Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques

Applications of Nuclear Techniques for GHGs

Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques

Mohammad Zaman • Lee Heng • Christoph Müller Editors

# Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques

Applications of Nuclear Techniques for GHGs

Editors Mohammad Zaman Soil and Water Management and Crop Nutrition Section International Atomic Energy Agency (IAEA) Vienna, Austria

Christoph Müller Institute of Plant Ecology Justus Liebig University Giessen Giessen, Germany

Lee Heng Soil and Water Management and Crop Nutrition Section International Atomic Energy Agency (IAEA) Vienna, Austria

#### ISBN 978-3-030-55395-1 ISBN 978-3-030-55396-8 (eBook) https://doi.org/10.1007/978-3-030-55396-8

© International Atomic Energy Agency, Vienna 2021. This book is an open access publication.

The opinions expressed in this publication are those of the authors/editors and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

Open Access This book is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

## Foreword

Delivering on the 2030 Agenda and the Sustainable Development Goals requires tackling global challenges that threaten food security, livelihoods, well-being and lives of billions of people. One of the greatest challenges and most pressing issues of our time is climate change, which exacerbates crop failures, fisheries depletion, erosion of livelihoods, environmental degradation, spread of infectious diseases, desperate competition for scarce natural resources, etc. While having disproportionate effects on regions, sectors and communities, the social, economic and environmental consequences of climate change are global in scope and unprecedented in scale. Adapting to them in the future will be costly, more difficult and complex.

It is therefore imperative and urgent to tackle the root causes of climate change while fostering adaptation to its impacts. The key action is to limit the cumulative level of greenhouse gas (GHGs) emissions that is increasing the average global temperature, which in turn is shifting weather patterns and rising sea levels. As we can only manage what we can measure, mitigation can only be successful when based on accurate measurements of GHGs emissions, along with the understanding of the processes underpinning these emissions. For this, nuclear and isotopic techniques have a comparative and competitive advantage over other methods as they offer a unique ability to precisely measure GHGs and identify their sources of production. Furthermore, they provide an insight into the movement and fate of nutrients, such as Nitrogen and Carbon, in our agroecosystems, from storage in soils, transfer to water bodies, to emission into atmosphere.

In this regard, the Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, through its Soil and Water Management and Crop Nutrition Section, has been leading, over the last four decades, laboratory and field coordinated research activities in partnership with national and international research institutes to develop advanced methodologies and tools using stable isotopes of nitrogen ( 15N) and carbon (13C) to devise novel mitigation techniques and management strategies for GHGs emissions from our agroecosystems. Key to success in this endeavour is the strategic and transformational partnerships with world-leading research centres such as the German Science Foundation research unit DASIM.

I am pleased to write the preface for this book, a culmination of a highly productive collaboration with a distinct partner (DASIM), a consortium of eminent German scientists and associated international leading experts, jointly working on trace gas emissions and nuclear techniques to measure and identify GHG processes, including at the agroecosystem level. It comprises eight chapters covering GHG emission from soil fauna and plants, methane production in ruminant animals, non-isotopic and micrometeorological methods, laboratory and field techniques, isotopic techniques to measure GHGs and identify their sources, and climate-smart agriculture practices for GHGs mitigation. This is a very important contribution to fostering and attaining the climate-smart agriculture "triple wins'' of reduced emissions, enhanced adaptation and increased agricultural productivity.

In summary, this book provides protocols, methodologies and standard operating procedures stemming from the most recent research and technology development on GHGs emissions in a variety of agroecosystems. I highly recommend it to researchers, scientists, engineers and practitioners who deal with issues pertaining to GHG emissions and climate change. I would like to conclude by thanking the authors, and their collaborators, for their efforts and contributions.

> Qu Liang Director, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture International Atomic Energy Agency (IAEA) Vienna, Austria

## Preface

The rapid change of global climate due to increased emission of anthropogenic greenhouse gases (GHGs) has led to increased extreme weather events such as droughts, floods and heatwaves. The emissions of major GHGs including carbon dioxide (CO2), methane (CH2) and nitrous oxide (N2O) have had a profound impact on the global climate including global warming and on the sustainability of agricultural production systems. Agricultural activities and land use changes contribute approximately 25% of the total GHG emissions, mainly due to poor farming practices including the inefficient use of chemical fertilisers, improper use of farm effluent and manure, overgrazing, and deforestation. While agriculture is contributing appreciably to climate change through GHG emissions, it is also a victim of climate change due to the negative impact climate change has on soil quality and crop productivity. Thus, it is paramount to find an integrated solution to reduce GHG emissions while at the same time making soils more resilient against climate change.

Precise measurement of GHG emissions across different agroecosystems and identifying their microbial sources within soils poses an immense challenge to researchers. Thus, the use of stable isotopic techniques offers the best choice for precise measurements of GHGs and identifying their sources of production. The methods outlined in this book apply, in principle, to all soil-derived greenhouse gases (i.e. CO2, CH4 and N2O) since most methods can simultaneously determine all three gases. However, there is a focus on emissions of gaseous N (N2O, N2) and methane (CH4). To understand effects of ecosystem management, and climate change on the emission of N2O and its mitigation, it is essential to determine the associated effects on the ratio of N2O/N2. Accordingly, current state-of-the-art methods and proposed future methodologies for determining N2 are also presented. Pathway-specific GHGs will be presented in the context of the soil processes (e.g. N transformation processes for the evaluation of N2O emission processes) to provide a toolbox for the understanding of GHGs in relationship to the overall elemental cycles. Thus, this book aims to provide a comprehensive overview of current and possible future technologies used to determine GHGs (especially N2O and CH4). Recent developments and discoveries of GHG sources such as the emissions from plants, fungi and fauna, as well as from ruminants and aquatic ecosystems are also presented.

The objective of this book is to present protocols, methodologies and standard operating procedures (SOPs) for measuring GHGs from different agroecosystems using isotopic and related techniques, which can also be used to validate climate-smart agricultural practices to mitigate GHGs. The material presented in this book should be useful for both, beginners in the field, to obtain an overview of the current methodology, and experienced researchers who need a hands-on description of current methodologies. We hope that the methods described in this book are easy-to-understand and applicable to a range of users with different expertise and backgrounds. This book is an outcome of the collaboration between the Soil and Water Management and Crop Nutrition Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, Department of Nuclear Sciences and Applications, International Atomic Energy Agency (IAEA), Vienna, Austria and the German Science Foundation research unit DASIM (Denitrification in Agricultural Soils: Integrated control and Modelling at various scales) among other institutes. DASIM is a consortium of scientists in Germany with associated partners worldwide which studies, in detail, the processes of GHG emissions using nuclear and stable isotopic techniques of 15N, 18O and 13C to understand and quantify gaseous N fluxes. This information has been compiled from the latest published literature and from authors' own publications specific to the subject matter (especially with expertise coming from the DASIM network).

Vienna, Austria Dr. Mohammad Zaman Vienna, Austria Dr. Lee Heng Giessen, Germany Prof. Christoph Müller

## Acknowledgements

We thank Mr. Liang Qu, Director NAFA, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, and Ms. Mokhtar Najat, Deputy Director General, Department of Nuclear Sciences and Applications, International Atomic Energy Agency (IAEA), Vienna, Austria, for their encouragement, support and timely help to write this publication. We are thankful to the late Prof. R. P. Pharis, Department of Biological Sciences, University of Calgary, Canada for his critical review and feedback on an earlier version of this book. We are grateful to the Hessian Agency for Nature Conservation, Environment and Geology (HLNUG) for their long-term support of the climate change programme that greatly supported this work. This book project was realised as part of the research unit DFG FOR2337: "Denitrification in Agricultural Soils: Integrated Control and Modelling at Various Scales (DASIM)" of the Deutsche Forschungsgemeinschaft (DFG).

## How to Cite the Book

Zaman M, Kleineidam K, Bakken L, Berendt J, Bracken C, Butterbach-Bahl K, Cai Z, Chang SX, Clough T, Dawar K, Ding WX, Dörsch P, dos Reis Martins M, Eckhardt C, Fiedler S, Frosch T, Goopy J, Görres C-M, Gupta A, Henjes S, Hofmann MEG, Horn MA, Jahangir MMR, Jansen-Willems A, Lenhart K, Heng L, Lewicka-Szczebak D, Lucic G, Merbold L, Mohn J, Molstad L, Moser G, Murphy P, Sanz-Cobena A, Šimek M, Urquiaga S, Well R, Wrage-Mönnig N, Zaman S, Zhang J, Müller C (2021) Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options Using Nuclear and Related Techniques Springer ISBN 978-3-030-55395-1, https://doi.org/10.1007/978-3-030- 55396-8.

## Contents


G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller 4 Micrometeorological Methods for Greenhouse Gas Measurement .......................................... 141 M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. A. Horn, M. E. G. Hofmann, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller 5 Direct and Indirect Effects of Soil Fauna, Fungi and Plants on Greenhouse Gas Fluxes ................................ 151 M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller 6 Methane Production in Ruminant Animals.................... 177 M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller 7 Isotopic Techniques to Measure N2O, N2 and Their Sources ............................................... 213 M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes,

M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir,


## Editors and Contributors

## About the Editors

Mohammad Zaman Soil and Water Management and Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria.

He is working as a Soil Scientist/Plant Nutritionist covering countries in Africa, Asia, Middle East, Europe, Central and Latin America on mitigating greenhouse gas (GHG) emissions and developing climate-smart agricultural practices for enhancing food production, minimise land degradation and environmental sustainability using nuclear and related techniques. Prior to joining the SWMCN Section, he worked for 22 years in GHG mitigations, enhancing soil fertility and integrated plant nutrient management at different research, academic, commercial and international organisations in both developing and developed countries. He is author and coauthor of one book, several book chapters and over 65 research publications in refereed journals. He has won several awards for his professional achievements including the IAEA Superior Achievement Award in 2018, and the IAEA Best Technical Cooperation Award in 2018 and Lincoln University, Canterbury New Zealand Doctorate Scholarship in 1995. e-mail: m.zaman@iaea.org

Lee Heng Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria.

She is working as a Section Head of Soil and Water Management and Crop Nutrition (SWMCN) Subprogram in the Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture. She has more than 27 years' experience in soil– plant–water interactions, agricultural water management and water use efficiency, integrated nutrient water interactions, and diffuse pollution control for sustainable agricultural production systems, at both national and international levels. Her work is covering countries in Africa, Asia, Europe, and Central and Latin America on sustainable land and water management for climate-smart agriculture and the efficient use and conservation of agricultural resources for enhancing food production and environmental sustainability. e-mail: l.heng@iaea.org

Christoph Müller Institute of Plant Ecology, Justus Liebig University Giessen and School of Biology and Environmental Science and Earth Institute, University College Dublin.

He is Professor of Experimental Plant Ecology (Giessen) and Adjunct Professor of Soil Science (Dublin). He is director of the Institute of Plant Ecology, and he is also the spokesperson of the German Science Foundation research unit DASIM ("Denitrification in Agricultural Soils: Integrated control and Modelling at various scales"). His main research interest is in feedback mechanisms of climate change on ecosystem processes. His group is managing one of the worldwide longest running Free Air Carbon Dioxide Enrichment (FACE) studies on grassland. He has developed stable isotope tracing methods to quantify the gross elemental transformation rates and processes of trace gas emissions from terrestrial ecosystems. e-mail: cmueller@uni-giessen.de; christoph.mueller@ucd.ie

## Contributors

Lars Bakken Faculty of Chemistry, Biotechnology and Food Science, Norwegian University of Life Sciences (NMBU), Aas, Norway (Chap. 3).

He is a professor in microbiology. His main interest is microbial nitrogen transformations, anaerobic respiration and regulatory biology of respiration in microbes, and implications for the emission of N2O. e-mail: lars.bakken@nmbu.no

Jacqueline Berendt Grassland and Fodder Sciences, Faculty of Agriculture and the Environment, University of Rostock, Germany (Chap. 7).

She is a Ph.D. student at the University of Rostock in the WETSCAPES project. She is measuring N2O emissions from peatlands and studying their microbial production pathways. During a research stay at the Lincoln University in New Zealand, she performed field experiments regarding tracer application methods with 15N. e-mail: jacqueline.berendt@uni-rostock.de

Conor Bracken School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland (Chap. 7).

He is a Ph.D. candidate based in University College Dublin. His primary research interests are investigating N2O emissions from grassland agriculture and applying cavity ring down spectroscopy measurements of N2O isotopomers to understand underlying soil N cycling processes. His research also examines the effectiveness of multispecies swards as a N2O mitigation strategy. e-mail: conor. bracken.1@ucdconnect.ie

Klaus Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Germany; Institute of Atmospheric Physics (IAP-CAS), Chinese Academy of Sciences, Beijing, China; International Livestock Research Institute (ILRI), Nairobi, Kenya (Chap. 3).

He is Head of Division of BioGeoChemical Processes at IMK-IFU, Guest Research Professor at IAP-CAS, and Principal Scientist at ILRI. His main research focus is on biogeochemical cycles of C and N in terrestrial ecosystems, and more specifically on processes governing the spatio-temporal variation of ecosystem-hydrosphere-exchange processes of environmental important substances under current and future climatic conditions. His work encompasses measurement and modelling aspects from site to regional scales. e-mail: klaus.butterbachbahl@kit.edu

Zucong Cai School of Geography Sciences, Nanjing Normal University, Jiangsu, China (Chap. 7).

He is Professor of Soil Sciences in the School of Geography Sciences, Nanjing Normal University. His main research interest is in the CH4 and N2O emissions from agroecosystems, particularly from rice paddy soils, organic carbon and nitrogen transformations in soils and cycles in terrestrial ecosystems. His group is exploring the smart nutrient managements based on the match of dominant inorganic nitrogen species in soil with the nitrogen species preferentially taken up by crop and the methodologies for quantifying the relative contributions of N2O producing processes to N2O emissions from terrestrial ecosystems by using isotopic method. email: zccai@njnu.edu.cn

Scott X. Chang Department of Renewable Resources, University of Alberta, Edmonton, Alberta, Canada T6G 2E3 (Chap. 8).

He is a Professor in forest soils and nutrient dynamics. His main research interests are in soil biogeochemistry, and the application of soil science in land reclamation, agriculture, forestry, and global change. Scott is currently a regional editor for Biology and Fertility of Soils, and an editorial board member for Pedosphere, Biochar and Forests. He served as Chair of the Soil Fertility and Plant Nutrition Commission of the International Union of Soil Science, Chair for the Forest, Range and Wildland Soils Division of the Soil Science Society of America, and Chair of the Alberta Soil Science Workshop. He is a Fellow of the Soil Science Society of America, American Society of Agronomy, and the Canadian Society of Soil Science email: sxchang@ualberta.ca

Tim Clough Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, New Zealand (Chap. 2).

He is a Professor of Environmental Biogeochemistry with research focusing on nitrogen and carbon cycling in agricultural ecosystems and natural environments. Central to this research is the use of stable isotopes to understand carbon and nitrogen dynamics in soils and the associated greenhouse gas emissions and their potential mitigation. e-mail: timothy.clough@lincoln.ac.nz

Khadim Dawar Department of Soil and environmental Sciences, University of Agriculture, Peshawar-Pakistan (Chap. 2).

He is an Assistant Professor of Soil Sciences. His main interest is to study the emissions of greenhouse gases from soil, enhancing nutrient and fertiliser use efficiency using nuclear and related techniques. e-mail: khadimdawar@yahoo.com

Weixin Ding Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China (Chap. 2).

He is professor of Soil Science, leading a working group on C and N cycling in soils, greenhouse gas emissions and their mitigations. He has published more than 110 papers in this area and has been supported by the fund for outstanding young scholar from National Natural Science Foundation of China. e-mail: wxding@mail. issas.ac.cn

Peter Dörsch Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway (Chap. 3).

He is a research professor in biogeochemistry with focus on GHG research. His work spans from controlled process to GHG emission studies on various scales. e-mail: peter.doersch@nmbu.no

Christian Eckhardt Institute of Plant Ecology, Justus Liebig University Giessen (Chap. 3).

Christian is a Ph.D. student at the Justus Liebig University working on mitigation options for GHGs in temperate grasslands. He has developed automated systems that combine various analyzers for GHG measurements under laboratory and field conditions, in particular connections between Cavity Ring Down Spectroscopy (CRDS) and automated chamber systems. e-mail: christian.eckhardt@bio.uni-giessen.de

Sebastian Fiedler Grassland and Fodder Sciences, Faculty of Agriculture and the Environment, University of Rostock, Germany (Chap. 7).

He studied agroecology at the University of Rostock and wrote his Ph.D. thesis on greenhouse gases from agricultural soils. To better understand biogeochemical cycles, he focused afterwards on stable isotope methods with 15N to trace sources of nitrous oxide. His second interest lies on climate change adaption of agriculture. e-mail: sebastian.fiedler@uni-rostock.de

Torsten Frosch Leibniz Institute of Photonic Technology, Technical University Darmstadt (Chap. 2).

He is the head of the spectroscopic sensing group at Leibniz-IPHT. One of his main research interests focuses on the development of enhanced Raman spectroscopy for simultaneous multi-gas analysis, including N2, N2O, O2, H2, CH4, CO2, and stable isotopes (13C, 15N, and 18O). This high selectivity enables the study of complex gas exchange processes, including pathways of the nitrogen cycle. Research regarding novel photonic enhancement techniques provide high sensitivities for trace gas analysis. His research was awarded with the Bunsen Kirchhoff Award for analytical spectroscopy (GDCh). e-mail: torsten.frosch@gmx.de

John Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya (Chap. 6).

He holds a Ph.D. (Ruminant Nutrition) from the University of New England (Australia) and bachelor's degree (Hons) in Animal Production from the University of Queensland. His main focus in the last two decades has been the measurement and mitigation of ruminant GHGs, variously in Australia, sub-Saharan Africa and south-east Asia. Recently his work has been pivotal in producing improved emission estimates for East African livestock, expanding the dairy farmers' training programme in Kenya. He holds Senior Research Fellow positions (adjunct) at the universities of Melbourne and Pretoria (South Africa) and currently lives in rural Australia. e-mail: j.goopy@cgiar.org

Carolyn-Monika Görres Department of Soil Science and Plant Nutrition/ Department of Applied Ecology, Hochschule Geisenheim University, Germany (Chaps. 2 and 5).

She is a landscape ecologist with a Ph.D. in agroecology. Her field of expertise is the exchange of greenhouse gases between soils and the atmosphere in terrestrial ecosystems with a focus on chamber-based flux measurements. In her current research, she investigates the effects of soil fauna on soil CH4 fluxes using stable carbon isotopes and soil acoustics. e-mail: carolyn.goerres@hs-gm.de

Abhishri Gupta Independent Consultant India (Chap. 8).

She completed her Master in Global Change: Ecosystem Science and Policy from University College Dublin, Ireland and Justus Liebig University, Giessen, Germany. Her research focus has been on Climate Smart Agriculture and its use in reducing food insecurity and agricultural greenhouse gas emissions. e-mail: abhishrig@gmail.com

Sina Henjes Institute of Microbiology, Leibniz University Hannover (Chap. 3).

She is a Ph.D. student in the group of Marcus A. Horn and member of the DASIM research unit. She is investigating the regulation, ecophysiology and kinetic parameters of N-gas flux associated anaerobic microbial communities in agricultural soils. Part of these analyses are carried out via automated gas chromatographic measurements of small-scale soil slurry incubations. e-mail: sina. henjes@ifmb.uni-hannover.de

Magdalena E. G. Hofmann Picarro B.V., 's-Hertogenbosch, The Netherlands (Chaps. 2 and 5).

She is working as Application Scientist for Picarro B.V., and has a passion for science and technology as evidenced by over a decade of research experience focused on stable isotope and trace gas analysis. A geochemist by training, Magdalena obtained a Ph.D. from the University of Göttingen, Germany, and a PostDoc at the University of Utrecht in the Netherlands. e-mail: mhofmann@picarro.com

Marcus A. Horn Institute of Microbiology, Leibniz University Hannover (Chap. 3).

He is a microbial ecologist and Professor of Soil Microbiology at the Leibniz University Hannover, (co-)editor of two microbiological journals, and PI of the research unit DASIM. N- and C-cycle associated microbes, including those related to greenhouse gas metabolism in soils and sediments (methanogens, intermediary ecosystem metabolism related microbes), and their ecology have been studied for more than a decade. Expertise includes regulation of process-associated microbial community structures and their activity patterns spanning scales from pure cultures to ecosystem manipulations. Approaches include isolation of target organisms, genomics, transcriptomics, gene marker-based diversity and expression analyses, kinetic analyses, stable isotope probing and tracing approaches, and diverse analytical techniques for the analyses of microbial transformations (i.e., LC/IC-MS, GC-PDHID). Research has been conducted at sites in Germany, Finland, Russia, and Brazil, including permafrost-affected areas. e-mail: horn@ifmb.uni-hannover.de

Mohammad M.R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Bangladesh (Chap. 2).

He is a Professor of Soil Science at Bangladesh Agricultural University. His main research interests are greenhouse gas measurement and mitigation in rice paddy and other agroecosystems; and nutrient transformations, pools, dynamics and balance using isotopic approach. e-mail: mmrjahangir@bau.edu.bd

Anne Jansen-Willems Institute of Plant Ecology, Justus Liebig University Giessen (Chap. 7).

She is a postdoctoral researcher at the Institute of Plant Ecology, Giessen, Germany, and a member of the research unit DASIM. Her main research interest is in modelling nitrogen transformations in the soil, with a current focus on the development and application of the 15N tracing tool Ntrace. e-mail: anne.willems@bot2. bio.uni-giessen.de

Kristina Kleineidam Institute of Plant Ecology, Justus Liebig University Giessen (Chaps. 2 and 7).

She is research associate at the Institute of Plant Ecology. Kristina is the coordinator of the German Science Foundation research unit DASIM (Denitrification in Agricultural Soils: Integrated control and Modelling at various scales) that tries to disentangle, with an interdisciplinary team, the complex interactions of physical, biochemical and microbial processes in soils in response to various boundary conditions to determine and model denitrification activities including dinitrogen (N2) emissions. e-mail: kristina.kleineidam@bot2.bio.uni-giessen.de

Katharina Lenhart University of Applied Sciences, Berlinstraße 109, 55411 Bingen am Rhein (Büdesheim), Germany (Chap. 5).

She is Professor for Botany, Limnology and Ecotoxicology at University of Applied Sciences in Bingen, Germany. Her field of expertise is the production and consumption of trace gases in terrestrial and aquatic ecosystems, with focus on plants and fungi. With her research projects she focuses on metabolic pathways of N2O and CH4 production and the quantification of emission rates by using stable isotopes. e-mail: k.lenhart@th-bingen.de

Dominika Lewicka-Szczebak Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Poland (Chap. 7).

She is postdoctoral researcher investigating soil nitrogen transformations with particular interest on gas emissions (N2O and N2) from agricultural soils. She is developing methods for applying natural abundance stable isotopes of N2O to identify N2O production pathways and quantify N2O reduction to N2 (in her own DFG project collaborating with DASIM).

e-mail: dominika.lewicka-szczebak@uwr.edu.pl

Gregor Lucic Picarro Inc., Santa Clara, California, United States (Chaps. 2 and 5). Gregor is a product manager at Picarro Inc., and is responsible for their greenhouse and trace gas family of analyzers. In his research career he studied active and diffuse degassing from small and large volcanic centers around the world. His passion for novel measurements and instrumentation brought him to Picarro, where he continues to support a broad spectrum of research projects. Gregor obtained his Ph.D. from McGill University, Canada. e-mail: glucic@picarro.com

Lutz Merbold Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, (Chap. 4).

Lutz research focuses on the detailed understanding of greenhouse gas (GHG) exchange in ecosystems worldwide, including wetlands, woodlands, forests, savannas, grasslands and agroecosystems. By applying micrometeorological methods such as the eddy covariance technique he defines major meteorological as well as biological factors influencing ecosystem GHG exchange and links the biogeochemical cycles of carbon, nitrogen and water.Besides studying the ecosystem scale, he is further interested in the exchange of GHGs at the process level (leafs, soils and animals using GHG chambers) and larger scales (regional to global) using modeling (empirical, semi-empirical and process-based biogeochemical models) and remote sensing approaches, while linking environmental aspects to productivity.Beyond the quantification of GHG exchange he is driven by a simple question: "How do ecosystems function now and how will they function in the future?" Lutz is motivated to educate undergraduate and graduate students in environmental sciences but also to apply recent research findings in the real world e.g. by doing environmental assessments of agricultural production systems in developing countries as currently done as head of Mazingira Centre at ILRI. e-mail: l.merbold@cgiar.org

Joachim Mohn Laboratory for Air Pollution and Environmental Technology, Empa Dübendorf (Chaps. 6 and 7).

He is leading the group Emissions & Isotopes at the research institute Empa. Joachim's work is focusing on anthropogenic and natural emissions of air pollutants, in particular diffuse emissions of trace gases and their isotopes. His main research interest is on the development and application of on-line analytical techniques and referencing strategies for the analysis of greenhouse gas isotopologues (e.g. N2O, CH4) to quantify emissions, identify precursors and trace transformation pathways. e-mail: joachim.mohn@empa.ch

Lars Molstad Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway (Chap. 3).

He is a senior engineer specialized in designing and programming robotized analytical solutions. He is currently doing a PhD on mathematical modelling of water filtration processes at the Faculty of Science of Technology, NMBU.e-mail: lars.molstad@nmbu.no

Gerald Moser Institute of Plant Ecology, Justus Liebig University Giessen (Chaps. 2, 3 and 7).

He is senior scientist at the Institute of Plant Ecology (Giessen). Gerald is the director of the Environmental Monitoring and Climate Change Impact Research Station. Since 1998 one of the world-wide longest running Free Air Carbon Dioxide Enrichment (FACE) Experiment is running there and the first free air combination of CO2 Enrichment and air Temperature warming (T-FACE), where the air is heated directly, started there in 2018. His main research interest is the impact of climate change on ecosystem processes and feedback mechanisms using stable isotope technique to disentangle the connections of ecosystem carbon and nitrogen cycles. e-mail: gerald.moser@bio.uni-giessen.de

Paul Murphy Environment and Sustainable Resource Management Section, School of Agriculture and Food Science, and UCD Earth Institute, University College Dublin, (Chap. 7).

He is Assistant Professor of Soil Science in the UCD School of Agriculture and Food Science and a fellow of the UCD Earth Institute. Pauls's research interests are in soil nutrient cycling and management, particularly relating to soil management practices, water quality and greenhouse gas emissions. Paul has a particular interest in N2O emissions from soils and the application of CRDS techniques and has played a leading role in establishing the UCD Earth Institute Environmental Isotopes Lab. e-mail: paul.murphy@ucd.ie

Marcio dos Reis Martins EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica/RJ Brazil (Chap. 2).

Marcio is a Soil Scientist. His main research interest is nitrogen dynamic in the soil-plant system in general, and measurements of ammonia, N2O and CO2. He has a long-term experience in determination of ammonia and N2O fluxes from agricultural soils. e-mail: marcio.dosreismartins@agroscope.admin.ch

Alberto Sanz-Cobena Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain (Chap. 8).

He is Associate Professor at the Technical University of Madrid (UPM) and researcher at the Centre for the Management of Agricultural and Environmental Risks (ETSIAAB, UPM). Beyond quantifying and assessing the impact of crop management on N dynamics and GHG emissions at plot scale, he also works on a more integrated view of complex problems such as GHG emissions and other nitrogen compounds and their mitigation on the provincial, watershed, national and large (Mediterranean) areas. e-mail: a.sanz@upm.es

Miloslav Šimek Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, České Budějovice and Faculty of Science, University of South Bohemia, České Budějovice, Czech Republic (Chap. 2).

He is a professor of Agricultural Microbiology. His main research interest is microbial nitrogen and carbon transformations in soil in general, and formation of trace gases including nitrous oxide, methane and carbon dioxide in particular. He has a long-term experience in determination of GHG fluxes from agricultural and forest soils. e-mail: miloslav.simek@bc.cas.cz

Segundo Urquiaga EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica/ RJ Brazil (Chap. 2).

He is a Soil Scientist. His main research interest is nitrogen and carbon transformations in soil in general, and measurements of ammonia and N2O, CH4 and CO2. He has a long-term experience in determination of GHG fluxes from agricultural and forest soils. e-mail: segundo.urquiaga@embrapa.br

Reinhard Well Thünen Institute of Climate-Smart Agriculture, Braunschweig (Chaps. 2 and 7).

He is a soil scientist leading a working group on N cycling in soils and its implication for greenhouse gas emissions. A major focus of his work consists of developing analytical as well as experimental techniques to investigate pathways of gaseous N fluxes at field and lab scales. His actual tasks focus on climate and land use effects on the emission of greenhouse gases from soils. He teaches Nitrogen Isotope Techniques in Terrestrial Ecosystems at the Universities of Göttingen and Braunschweig. e-mail: reinhard.well@thuenen.de

Nicole Wrage-Mönnig Grassland and Fodder Sciences, Faculty of Agriculture and the Environment, University of Rostock, Germany (Chap. 7).

As Professor of Grassland and Fodder Sciences, Nicole is searching for sustainable solutions for livestock fodder production, including improved nutrient use. For more than 20 years, she has been using and developing methods for distinguishing among sources of nitrous oxide from soils using stable isotopes. Within the DASIM consortium, she is responsible for further improving these methods and using them in-field to provide data for improved modelling of denitrification and other sources of N2O from distinct soil volumes. e-mail: nicole.wragemoennig@uni-rostock.de

#### Shazia Zaman (Chapter 2)

She holds a Ph.D. from University of Canterbury, New Zealand. After completing her Ph.D., Shazia continued post-doctoral research at Canterbury University and Adelaide University, Australia in the areas of the synthesis of nitrogenous heterocycles of biological importance. Shazia is author and coauthor of one book chapter and several research publications in refereed journals.

Jinbo Zhang School of Geography, Nanjing Normal University (Chap. 7).

He is Professor of Geography. His main research interest is in soil nitrogen transformations and feedbacks between plants and soil nitrogen transformations using stable isotope tracing methods. He is author and coauthor of one book, several book chapters and over 100 research publications in refereed journals. e-mail: zhangjinbo@njnu.edu.cn

## Acronyms and Abbreviations




## List of Figures







#### List of Figures xxxvii


## List of Plates





#### List of Plates xliii



## List of Tables



## **Chapter 1 Greenhouse Gases from Agriculture**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** The rapidly changing global climate due to increased emission of anthropogenic greenhouse gases (GHGs) is leading to an increased occurrence of extreme weather events such as droughts, floods, and heatwaves. The three major GHGs are carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). The major natural sources of CO2 include ocean–atmosphere exchange, respiration of animals, soils

M. Zaman (B) · L. Heng

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang

Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

T. Clough Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

© The Author(s) 2021

M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_1

(microbial respiration) and plants, and volcanic eruption; while the anthropogenic sources include burning of fossil fuel (coal, natural gas, and oil), deforestation, and the cultivation of land that increases the decomposition of soil organic matter and crop and animal residues. Natural sources of CH4 emission include wetlands, termite activities, and oceans. Paddy fields used for rice production, livestock production systems (enteric emission from ruminants), landfills, and the production and use of fossil fuels are the main anthropogenic sources of CH4. Nitrous oxide, in addition to being a major GHG, is also an ozone-depleting gas. N2O is emitted by natural processes from oceans and terrestrial ecosystems. Anthropogenic N2O emissions occur mostly through agricultural and other land-use activities and are associated with the intensification of agricultural and other human activities such as increased use of synthetic fertiliser (119.4 million tonnes of N worldwide in 2019), inefficient use of irrigation water, deposition of animal excreta (urine and dung) from grazing animals, excessive and inefficient application of farm effluents and animal manure to croplands and pastures, and management practices that enhance soil organic N mineralisation and C decomposition. Agriculture could act as a source and a sink of GHGs. Besides direct sources, GHGs also come from various indirect sources, including upstream and downstream emissions in agricultural systems and ammonia (NH3) deposition from fertiliser and animal manure.

K. Dawar

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

W. X. Ding Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch

Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

C.-M. Görres

Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

M. E. G. Hofmann Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh **Keywords** Climate change · GHG · N2O · CO2 · NH3 · Animals

## **1.1 Introduction**

The global climate is changing rapidly. This leads to the increasing occurrence of extreme weather events such as droughts and floods. The major cause of these events is the rising temperature in the Earth's atmosphere, which is driven by increasing emissions of climate-relevant greenhouse gases (GHGs) that trap heat in the atmosphere. Major GHGs include carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) (Fig. 1.1).

Carbon dioxide is the major GHG responsible for the increasing greenhouse effect of the atmosphere. Key natural sources of CO2 include ocean–atmosphere exchange, respiration of animals, soils (microbial respiration) and plants, and volcanic eruption. Major anthropogenic sources of CO2 include burning of fossil fuel (coal, natural gas, and oil), deforestation, and the cultivation of land that increases the decomposition of soil organic matter and crop and animal residues (Xu and Shang 2016).

Aside from CO2, CH4 is a major GHG, which is emitted by natural and anthropogenic processes. Natural sources of CH4 emission include wetlands, termite activities, and occan. Paddy fields used for rice production, livestock production systems

J. Zhang

School of Geography, Nanjing Normal University, Nanjing, China

K. Lenhart

Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

D. Lewicka-Szczebak Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold

Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

M. Šimek

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

(enteric emission from ruminants), landfills, and production and use of fossil fuels are the main anthropogenic sources of CH4. Furthermore, CH4 can be produced by anaerobic mineralization by methanogenic archaea in both natural and man-made systems. Also, plants have been shown to emit CH4.

The third major GHG is N2O (Zaman et al. 2012). Besides being a major GHG, N2O is a major ozone-depleting gas (Ravishankara et al. 2009). Oceans and soils under natural vegetation are non-anthropogenic sources of N2O. However, at a global scale, the emission of N2O is mostly caused by, or related to, anthropogenic agricultural and other land-use activities. The atmospheric concentration of N2O has increased by more than 20% from ~271 ppb to 331 ppb since the industrial era (ca. 1750) to 2018 (WMO 2019). Over the last decade, the rate of N2O increase was equal to 0.95 ppb yr−<sup>1</sup> (IPCC 2013b; WMO 2019) with an increasing trend (Makowski 2019; Thompson et al. 2019). In 2006, the total anthropogenic source of N2O was 6.9 Tg N2O-N. Out of these direct emissions, agricultural sources dominated (4.1 Tg N2O-N), while indirect emissions accounted for 0.6 (with a range of 0.1–2.9) Tg N2O-N (IPCC 2013a). Such a large N2O emission is attributed to various factors including the intensification of agricultural and other human activities, increased use of synthetic fertiliser (119.4 million tonnes of N worldwide in 2019), inefficient

M. dos Reis Martins · S. Urquiaga

EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

R. Well

Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany S. Zaman

University of Canterbury, Christchurch, New Zealand

use of irrigation water, deposition of animal excreta (urine and dung) from grazing animals, excessive and inefficient application of farm effluents and animal manure to croplands and pastures, and management practices that enhance soil organic N mineralisation and C decomposition. These activities affect the N cycle. The N cycle is rather complex, and it has even been disrupted due to increased N inputs and intensification of agriculture. Sources of increased N inputs, in particular reactive N, into the N cycle stem, for instance, from the Haber–Bosch process (Erisman et al. 2011), thereby transforming the N cycle into the so-called N cascade, which is characterised by the release of reactive N forms into the environment with various consequences (Sutton et al. 2011). There are still many uncertainties concerning the N cycle. For example, the role of individual factors controlling the occurrence and rate of the key N transformation processes, such as denitrification and nitrification, is uncertain (Butterbach-Bahl et al. 2013; Müller and Clough 2014; Smith 2017). Nitrification results in ammonium (NH4 +) being converted to nitrate (NO3 <sup>−</sup>) under aerobic conditions, while denitrification is the reduction of NO3 <sup>−</sup> to N2 under anaerobic conditions. Nitrifier denitrification occupies the niche between nitrification and denitrification and occurs as oxygen concentrations approach an anaerobic status. Under these conditions, nitrifiers actually convert nitrite into N2O and N2 instead of nitrate (Wrage-Mönnig et al. 2018).

The three GHGs (CO2, CH4, and N2O), derived from various sectors, play a major role in regulating Earth's temperature (Fig. 1.1). Without GHGs in the atmosphere, the average global soil surface temperature would be ~19 °C, compared to the present values of 14 °C (Hossain 2019).

Recent data from the UN Intergovernmental Panel on Climate Change (IPCC) clearly show that anthropogenic emissions of GHGs are at the highest in history (IPCC 2014a). Since 1990, Earth's average surface air temperature has increased by about 0.8 °C, with much of the emission increases taking place since the mid-1970s (Fig. 1.2).

The global warming potential (GWP) of a GHG relates to the amount of heat trapped by a certain mass of a gas to the amount of heat trapped by a similar mass of CO2 calculated over a 100-year time horizon (IPCC 2016). For example, the GWP of N2O is 265–298, which means if the same masses of N2O and CO2 were emitted into the atmosphere, N2O would trap 265–298 times more heat than CO2 over a 100-year time period (Table 1.1) (IPCC 2016).

The Kyoto Protocol was negotiated by parties to the United Nations Framework Convention on Climate Change (UNFCCC) in an effort to stabilise the continued increase in atmospheric GHG concentrations. The Kyoto protocol outlines GHG reduction targets for participating countries. The signatories to the protocol must develop and report on their annual national inventory of anthropogenic GHG emissions. Guidelines on how to construct inventories were prescribed by the IPCC (2014b). Country-specific emission data can be considered but that requires accurate inventory data based on precise GHG measurements.

**Fig. 1.2** Recent anthropogenic emissions of GHGs (IPCC 2014a; WMO 2019)


**Table 1.1** Global warming potential (GWP) and atmospheric lifespan of various GHGs (IPCC 2016)

## **1.2 Impact of Ammonia on GHG Emissions**

Besides direct sources, GHGs also come from different indirect sources such as upstream and downstream in agricultural systems (Plate 1.1).

Ammonia (NH3) itself has no direct greenhouse effect. It is a gas with a relatively short residence time in the atmosphere (2–10 days) compared to some GHGs, such as CO2 (3–4 years), CH4 (12 years), and N2O (114 years). However, after NH3 is emitted to the atmosphere and reacts with acids, it forms salts. These salts then return to The Earth's surface and act as a N source source for N2O emissions, similar to a fertiliser-N application.When the soil is submitted to conditions near the optimum for

**Plate 1.1** Schematic representation of direct (from the cropping system) and indirect (upstream and downstream) GHG emissions from crop production. As an example, ammonia emitted from a cropping system will be deposited and potentially oxidised to nitrate, which can further be denitrified, thus enhancing the risk for N2O emission (Aguilera Fernández 2016)

urease activity (e.g. pH close to neutrality, soil moisture near field capacity, temperature >30 °C), the N losses through NH3 volatilisation from urea-based fertilisers applied on soil surface can be as high as 50% (Martins et al. 2017; Rochette et al. 2013). Therefore, the measurement of NH3 emission is important to estimate indirect N2O emissions derived from soil amendments, such as urea-based fertilisers, green manures, animal excreta, or ammonium-based fertilisers in alkaline soils. A default emission factor defined by IPCC, known as EF4, can be applied for the estimation of indirect N2O emissions derived from volatilisation of NH3 and other nitrogen oxides (NOx) (IPCC 2006). The mean value of EF4, considering the N volatilisation and consequent re-deposition, is 0.01 kg N2O-N per kg N volatilised as NH3 with an uncertainty ranging from 0.002 to 0.05 (IPCC 2006). Therefore, management options that reduce NH3 volatilisation from soils are considered mitigation practices because they reduce indirect N2O emissions (IPCC 2014a; Lam et al. 2017).

## **1.3 Aim of the Book**

In 1992, the International Atomic Energy Agency (IAEA) published a "Manual on measurement of CH4 and N2O emissions from agriculture" (IAEA 1992). Since the publication of the manual, progress has been made in analytical techniques. The progress includes advances in automation technologies as well as the theoretical understanding of how soil microbial processes affect CH4 and N2O emissions and the factors influencing those microbial processes.

Hence, the aim of this book is to provide an updated account of the state-of-theart techniques to measure direct GHG emissions (Plate 1.1), as a necessary step to propose and assess any mitigation strategy. The focus is on CH4 and N2O emissions. Additionally, information on techniques to measure indirect GHG sources is provided. Indirect GHG sources in this book include volatilised ammonia (NH3). NH3 is a reactive N gas highly affecting the environment through eutrophication and acidification of natural ecosystems as well as human health due to the promotion of particulate matter formation (Sanz Cobena et al. 2014). Moreover, in line with the 1992 IAEA Tecdoc document, the hands-on approach is also followed here so that researchers, who want to use the techniques described in this book, can easily apply them to their own work.

## **References**


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 2 Methodology for Measuring Greenhouse Gas Emissions from Agricultural Soils Using Non-isotopic Techniques**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** Several approaches exist for measuring greenhouse gases (GHGs), mainly CO2, N2O, and CH4, from soil surfaces. The principle methods that are used to measure GHG from agricultural sites are chamber-based techniques. Both open and closed chamber techniques are in use; however, the majority of field applications use closed chambers. The advantages and disadvantages of different chamber techniques

M. Zaman (B) · L. Heng

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

© The Author(s) 2021 M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_2

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

and the principal steps of operation are described. An important part of determining the quality of the flux measurements is the storage and the transportation of the gas samples from the field to the laboratory where the analyses are carried out. Traditionally, analyses of GHGs are carried out via gas chromatographs (GCs). In recent years, optical analysers are becoming increasingly available; these are user-friendly machines and they provide a cost-effective alternative to GCs. Another technique which is still under development, but provides a potentially superior method, is Raman spectroscopy. Not only the GHGs, but also N2, can potentially be analysed if the precision of these techniques is increased in future development. An important part of this chapter deals with the analyses of the gas concentrations, the calculation of fluxes, and the required safety measures. Since non-upland agricultural lands (i.e. flooded paddy soils) are steadily increasing, a section is devoted to the specificities of GHG measurements in these ecosystems. Specialised techniques are also required for GHG measurements in aquatic systems (i.e. rivers), which are often affected by the transfer of nutrients from agricultural fields and therefore are an important indirect source of emission of GHGs. A simple, robust, and more precise method of ammonia (NH3) emission measurement is also described.

**Keywords** GHG · Chamber-based technique · Raman spectroscopy · Ammonia emission · Optical spectroscopy

T. Clough

K. Dawar

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

W. X. Ding

Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad

Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch

Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

C.-M. Görres Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

## **2.1 Introduction**

Given the complexity of emissions, process-based models are not able to accurately estimate daily fluxes or the variations in fluxes due to variations in management practices. This limits our understanding of the factors affecting greenhouse gas (GHG) emissions and eventually restricts the development of agricultural management options that minimise GHG emissions. The Intergovernmental Panel on Climate Change (IPCC) requires local data based on field studies. Estimation of emission factors (EF) along with quantification of EF-associated physical, chemical, and biological processes that produce CH4 and N2O is required for field-scale GHG measurements. Field measurement of GHG is the basis of GHG flux estimates and a means of evaluating potential countermeasures for reducing emissions (Minamikawa et al. 2015). Several approaches exist for measuring GHG fluxes from soil surfaces. The two most important approaches are chamber-based methods and micrometeorological techniques (Denmead 2008; Oertel et al. 2012) (Chap. 4); more sophisticated approaches include space and airborne measurements.

Micrometeorological techniques usually integrate much larger surface areas in comparison to chamber-based techniques, thereby substantially reducing spatial variability problems that are inherent to chamber-based methods (Mosier 1990). Micrometeorological techniques are often more expensive, require special analytical instruments, and need knowledge and expertise that are largely not available in most developing countries. There is also a third technique that could be used for gas flux estimation–the measurement of gas concentration in different layers of soil.

M. E. G. Hofmann

Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh

D. Lewicka-Szczebak

Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

Information about gas concentration in a soil profile can also be used for gas flux prediction (Chirinda et al. 2014; Kammann et al. 2001). However, this technique requires additional information on soil physical and chemical properties, including hydraulic characteristics, to calculate GHG fluxes based on gas diffusion in the soil matrix (Diel et al. 2019). Therefore, in most instances, chamber-based methods have been used to study GHG fluxes from agricultural soils. Nonetheless, in combination with chamber-based techniques, soil profile techniques can provide valuable additional information to explain and analyse GHG emissions from the soil surface (Müller et al. 2004).

## **2.2 Chamber-Based Methods**

The great majority of GHG emission studies published in the past three decades have used chamber-based techniques–in particular, non-flow-through, non-steady-state chambers (Rochette 2011). These methods have been described in detail in several excellent review papers (De Klein and Harvey 2012; Hutchinson and Livingston 1993; Hutchinson and Rochette 2003; Mosier 1989). The following text mainly comes from these reviews that precisely address the topic and represent a comprehensive overview of the method. In addition to information from the literature, experimental data and our own experience with field gas flux determination will also be presented.

Mosier (1990) characterised three basic chamber-based techniques: open soil chambers (open dynamic chambers) that use flow-through air circulation, closed soil chambers with closed-loop air circulation (closed dynamic chambers), or no air circulation (static closed chambers).

M. Šimek

M. dos Reis Martins · S. Urquiaga

R. Well

J. Zhang

K. Lenhart

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany

S. Zaman

University of Canterbury, Christchurch, New Zealand

School of Geography, Nanjing Normal University, Nanjing, China

Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

## *2.2.1 Advantages and Disadvantages of Closed Chamber-Based Methods*

The closed chamber technique has several advantages (Mosier 1990; Oertel et al. 2016), including the following:


The closed chamber technique, however, has some limitations:


In summary, closed-chamber methods represent an inexpensive and easy to use technique, suitable for the determination of GHG fluxes between the soil and the atmosphere from a wide range of agroecosystems. However, several aspects must be considered when using the closed chamber-based methods, including the following:


Some of the problems with field GHG flux measurements are the large spatial variability of gas fluxes, and the high and often unpredictable temporal changes of fluxes. Other problems associated with field GHG flux measurements include the following:


## *2.2.2 Principles and Applications of Chamber-Based Techniques for Gas Flux Measurement*

There is no best technique for GHG flux determination; each technique has its advantages and disadvantages, and no single approach is applicable for all conditions or purposes. An excellent overview of the principles and applications of chamber-based techniques for gas flux measurements was provided by Livingston and Hutchinson (1995). The following text is based on that publication and provides selected information, for practical reasons not every publication mentioned by the authors in their text is cited. More details and references can be found by Livingston and Hutchinson (1995). Some data and experiments taken from Šimek et al. (2014) are also included in this chapter. More recent developments in chamber-based techniques for N2O flux measurements are discussed and summarised in De Klein and Harvey (2012) and Oertel et al. (2016). The factors causing high temporal and spatial variabilities in GHG emissions are outlined in the following sections.

## *2.2.3 Gas Exchange Processes*

Rates of gas exchange between the soil and the atmosphere are usually extremely variable in time and space, making the determination of gas fluxes very complicated and challenging. Movement of gas molecules is due to either mass flow (advective transport) or molecular diffusion. *Diffusive transport* is described by Fick's law and affected by gas permeability, i.e. the ease with which gases move through soil, varies over several orders of magnitude in relation to the shape, size, orientation of the soil pores and soil water content. *Advective transport* of gas occurs in response to a difference in total pressure between the soil air and the atmosphere, which is described by Darcy's law. *Water* substantially affects the movement of gases in soil. Diffusivity of gases is about 104 times smaller in water than in air (although the rates differ for different gases). *Plants* influence the exchange of gases; typically, the presence of plants results in increases in gas fluxes from and to the soil. Plants (mainly vascular plants) function as a direct pathway for the flow of trace gases through (often specialised) plant tissues (aerenchyma system); plants also alter conditions in the rhizosphere and therefore directly or indirectly influence gas formation and transport. Moreover, plants also consume and produce several gases, including CH4 (Liu et al. 2015) or N2O (Müller 2003). From a practical point of view, the presence of plants usually makes the gas flux determination even more challenging in comparison with bare soil.

## *2.2.4 Chamber Types*

There are many types of closed/static chambers, typically developed by researchers for specific purposes (De Klein and Harvey 2012; Oertel et al. 2016; Saggar et al. 2007). The chambers may be made from various materials, including metals, plastics, and glass, and can have different designs, sizes, shapes, and volumes; chambers as small as 50 cm3 and as large as ca 1 m3 have been used for field flux determination from the soil surface. Obviously, chamber materials should be chemically inert, and thus, neither react with the gases being measured nor emit any contaminants. Recommended materials therefore include stainless steel, aluminium, and glass, while the use of polycarbonate, polyethylene, methyl methacrylate, and polyvinyl chloride should be checked for their suitability before use. The schematic diagram of the most common metal chamber is shown in Plate 2.1. This is similar to the design proposed by De Klein and Harvey (2012).

Metal chambers, as shown in Plate 2.1, represent the best choice for many reasons. Metals are not permeable to gases and are inert if materials such as stainless steel are used and can be manufactured in local workshops. The type of material used is important because galvanised steel or normal steel may alter the soil conditions by releasing zinc and iron ions that have the potential to affect microbial activity; thus, it is recommended to use stainless steel. However, when compared to plastic

**Plate 2.1** A schematic diagram **a** of three parts base frame, extension or enlargement and top lid with GHG sampling ports, a complete metal closed chamber in the paddy field (**b**), and specially designed chamber for maize plants (**c** and **d**)

2 Methodology for Measuring Greenhouse Gas Emissions … 19

**Plate 2.1** (continued)

chambers, they can be more expensive, heavier, and less available. Also, insulation is required to minimise temperature fluctuations inside the chamber which in turn would affect fluxes of GHG's, and microbial processes that drive their production. Metal chambers typically consist of two or three parts: the bottom part (also called the base, frame, or soil collar/gutter), the top part (i.e. the chamber), and perhaps a suitable extension (Plate 2.1). The bottom part (frame) should be inserted into the soil at least 2 weeks before the first sampling and permanently installed to minimise soil disturbances effects. Chambers shall be insulated (e.g. using foam or polystyrene with reflective foils) to avoid unnatural heating during chamber closure. This allows repeated gas flux measurements in the same place, e.g. during the whole season. The two/three-part chamber design is strongly recommended, as the disturbances of soils prior to the measurements are eliminated. This, however, means that the frame should ideally be placed into the soil a few weeks before the measurements commence (De Klein and Harvey 2012; Oertel et al. 2016). Precautions should be taken in grazed sites so that neither the chamber nor the animals are endangered.

Plastic chambers have often been used for GHG flux determination (Plate 2.2a, b) (Zaman et al. 2009). The most critical issue for this type of chamber is the nature of the plastic used for making the chamber. Most plastic materials show permeability for gases, as well as the ability to emit some gases or react with them, e.g. hydrocarbons. The advantages of using plastic are (among others) the general availability, easiness to work with, low weight, and able to easily glue different parts together. Plastic chambers consist of two parts (a plastic vessel without a base/collar/gutter) and a lid. The plastic vessel is usually inserted into the soil at least 2–3 days prior to the

**Plate 2.2** Plastic chamber made up of two parts: vessel and lid for measuring greenhouse gas emission from **a** pasture soil and **b** arable and vegetable croplands

2 Methodology for Measuring Greenhouse Gas Emissions … 21

**Plate 2.3** Glass chamber consists of only one part (Šimek, personal communication) for measuring GHG emission from pasture soils

flux measurement. For gas sampling, a lid containing a gas sampling port (rubber septum) connected to a three-way valve is carefully placed on top of the vessel using a gas-tight seal (Plate 2.2a, b). After gas samples are collected, the lid is removed. Plastic chambers can be either transparent or opaque.

Glass chambers have been used less frequently, although the material (glass) is probably the best material to use, considering the inertness and the low gas permeability. However, glass is fragile which makes it very problematic to work with. Therefore, there are more disadvantages than advantages to use glass chambers. Still, one type of glass chambers has been tested for gas flux measurements (Plate 2.3). The major disadvantage of this chamber type is the limited size of the bottles available. Bottle volume is usually between 100 and 2000 cm3, and surface area covered by the chamber is less than 100 cm2, which is too small for most uses. Glass chambers consist of a single part (without a base/collar/gutter) and are transparent.

## *2.2.5 Chamber Design*

Critical aspects of the chamber-based methodology include several construction considerations, especially materials, dimensions, gas tightness, and insulation (Table 2.1).

All details for the construction of closed chambers (Table 2.1) shall be taken into consideration in order to maximise the accuracy and precision of the measurements


**Table 2.1** Static chamber design requirements and design recommendations (adapted from De Klein and Harvey (2012))

of soil GHG fluxes. A first factor that shall be considered in terms of dimension is the height of the chamber (De Klein and Harvey 2012). The homogeneity of the air in the headspace can be compromised in higher chambers (e.g. >40 cm). To minimise this effect, small fans connected to rechargeable batteries can be used inside the static chambers or even pumping air from the syringe inside the chamber before gas sampling. However, for the latter care has to be taken to avoid the creation of pressure artifacts in the absence of fans (Christiansen et al. 2011). In addition, higher chambers also lead to poor detection of low gas flux due to the dilution of gas derived from soil with the air of chamber headspace. On the other hand, excessive reduction of the chamber height increases the influence of the chamber deployment on the gas diffusion from soil to chamber causing a bias. In some cases, the use of high chambers or chamber extensions is necessary. For example, when growing plants should be incased, in this case, researchers shall ensure that chambers do not physically injure the plants. If taller chambers (e.g. >60 cm) are used, longer deployment time is necessary to improve the detectability of soil gas fluxes. As a rule of thumb, when the chamber height is doubled, the deployment time should also be doubled.

The chamber base must be inserted into the soil deep enough to prevent gas leakage from chamber headspace. There is not a clear consensus in the literature regarding a minimum depth for the insertion of chamber bases (or frames) into soil. Depths found in the literature range from 5 to 20 cm (Hutchinson and Livingston 2001; Martins et al. 2015; Martins et al. 2017; Zaman et al. 2009). Special care shall be taken regarding the depth of base insertion when GHG emissions are determined from sandy soils because of the higher risk of gas leakage by lateral diffusion. Another very important aspect of the prevention of gas leakage from chamber headspace is base-chamber sealing. The use of a trough soldered on the top of the base and filled with water immediately prior to the base-chamber coupling has been shown to be an efficient method of sealing (De Klein and Harvey 2012). Another option for sealing is the use of gaskets that are compressed by fasteners at the time of base-chamber placement. The advantage of using gaskets is the ability to seal chambers used in areas that are not flat which may exist in natural areas (i.e. forests, hilly pastures) or smallholder cropping systems.

Stainless steel and PVC materials are the most commonly used materials to construct static chambers for field deployment. When soil gas flux is being measured in pasture systems in the presence of animals, cages will need to be used to prevent chambers from being damaged. To avoid unnatural heating in the chamber during gas sample collection, both insulating materials such as foam or polystyrene and reflective foils should cover them. The material for insulation is usually non-expensive and can be easily found in the market. The chamber insulation minimises the changes in air temperature in the chamber headspace, reducing biases due to temperature effects on gas diffusion from soil. The use of small vent tubes is recommended to avoid the effects of pressure difference inside and outside of the chamber on gas diffusion from soil (Xu et al. 2006). Detailed information on how to determine the best diameter and length of vent tubes has been previously presented (Hutchinson and Livingston 2001; Hutchinson and Mosier 1981; Parkin and Venterea 2010). For further reading the paper of the Global Research Alliance by Clough et al. (2020) on chamber design considerations is recommended.

## *2.2.6 Chamber Operation, Accessories, Evacuation of Exetainers, and Gas Flux Measurement*

Any chamber, plastic or metal, must be rigid enough to be repeatedly used in the field. The procedure of using static chambers for gas flux measurements from pasture, crop, and vegetable lands is similar. This includes (i) chamber base/collar insertion into the soil, and deployment of the top part if the chamber consists of the two or three parts as shown in Plates 2.1 and 2.2a, b, (ii) closing the chamber (placing either lid/upper part with stopper and septum), and (iii) repeated collection of GHG samples from each chamber using a syringe at specific timings such as 0, 30, and 60 min. Time intervals of gas sampling always depend on factors such as specific conditions, the purpose of the study, and on the gas(es) to be determined.

In addition to chamber design and chamber deployment into soil, having proper gas sampling skills is important to achieve the best quality data for GHG emission (Table 2.2).

Chamber bases shall be installed long enough before measurements commence to allow for conditions to approximate the ambient (Plate 2.4). This might take as little as one hour on coarse-textured soils, while a few days may be needed on clayey soils, provided unvegetated area is investigated. In some cases, even weeks may be required to allow root regrowth. This will avoid any potential impacts of root death, which disrupts C and N cycling with potential effects on CO2 and N2O production and consumption in the soil profile. Among annual crops, chamber bases should be installed shortly after seeding, to allow roots to grow into the inner area. Soil water content can impact chamber performance in several ways. Researchers walking around the chambers, especially in very wet conditions, can compact the soil. Chamber bases may also affect lateral surface water flow, and they should be relocated when soil water content inside the chamber differs from the surrounding area. Finally, under very dry conditions, clayey soils may shrink away from the edge of the chamber base. In such circumstances, researchers should carefully loosen and tamp down the soil at the outer edge of the chamber base prior to measurement, to fill the gap and improve the seal between the soil and the chamber base (De Klein and Harvey 2012).

To collect gas samples from each chamber, researchers shall have accessories including syringes (60 ml), three-way taps (Luer-Lock), 12 ml pre-evacuated exetainers, and needle of 0.45 mm × 13 mm (Plate 2.5).

The exetainers are usually pre-evacuated; however, if reused, they must be reevacuated. However, if an evacuation manifold is not available, also a method is available to use unevacuated vials (for more details, see below).

Various glass vials or vessels have been used for temporary gas sample storage and transportation from field to laboratory before gas analysis. For short-term storage of a large number of samples, inexpensive polypropylene syringes have been used; their use is, however, rather limited because of the possibility of gas sample loss during prolonged storage, for example, when the storage time exceeds 2 days (Rochette


**Table 2.2** Recommendations for static chamber deployment (adopted from De Klein and Harvey (2012))

and Bertrand 2003). If these types of syringes are used, the gas transfer to suitable vials should be done immediately after the gas sampling. Septum-sealed containers of different materials, volumes, and overall qualities have been used for gas storage. The best solution is arguably a glass vial of several ml in volume (8–20 ml), evacuated and sealed with a special gas-tight stopper. In this context, two possible sources of errors exist. First, the quality of the evacuation, even if the vials were bought as evacuated, an additional evacuation prior to use is required. If the vial contains N2, argon (Ar), or another inert gas such as helium (He) or has been purged with such a gas, it does

**Plate 2.4** Metal chamber **a** frame, **b** body and lid, **c** complete chamber for gas flux measurement in the field (material: stainless steel)

**Plate 2.5** Accessories for gas sampling

not create a problem for the analysis of GHGs. However, the vial shall not contain any traces of the gas which is to be determined in the gas sample. There is another related problem: if the amount of inert gas in the vial is too high, the gas sample added to the vial for storage purpose is substantially "diluted", and this affects the gas concentration in the subsample which is taken later for gas analysis. To overcome this potential source of error, vials shall be evacuated before gas sampling using a high vacuum pump (Plate 2.6). This process of evacuation takes about 3–5 min per sampling batch. After ca 5–10 times of repeated use of the vials, the septum should be replaced with a new one. Usually, septa can be separately purchased from the supplier.

It is also a good practice to fill the gas vials first using inert gas (e.g. Ar, He, N2, depending on the purpose), and then to evacuate them–and to do so (filling, evacuation) repeatedly (3–5 times), or to flush the vials with the inert gas and then use the evacuation system below (Plate 2.6).

**Plate 2.6** A high-efficiency vacuum pump connected to a system for simultaneous evacuation of multiple gas vials

In any case, when using vials for temporary gas storage, it is strongly recommended to check the quality of vacuum and overall quality (see below) of vials and stoppers before using them regularly. If gas analyses are not performed right away, then researchers shall store the gas vials with headspace samples in an insulated box (to avoid large temperature changes), and transport the vials to the laboratory as soon as possible, or pack and send them, preferentially in an insulated box for gas concentration determination. For prolonged storage periods, it is recommended to store standard (calibration) gas mixture in the same way as the gas samples. Comparison of direct standard analysis and analysis of several samples containing standard gas mixture, instead of unknown samples, yields the correction factor necessary for sample dilution and gas leakage calculations. For longer storage, it is advisable to store samples with an overpressure in the vial which is often done anyway, e.g. if a sample loop of the analytical instrument has to be filled which is often the case if sample analyses occur via an autosampler.

The second major risk in using glass vials is that the seals or septa are made from plastic material. The stopper must ensure "gas-tightness" in two aspects: it must be nonpermeable to the gas to be analysed (no gas is diffusing to and from the vial) and it must be inert (no gas is generated by the material of the stopper during gas storage). For example, common silicone stoppers are very gas permeable (!) whilst other materials sometimes create a large amount of gases, typically light hydrocarbons including methane (CH4). Inconsistencies related to gas sampling and storage and possible errors are often ignored which may lead to large errors.

Glass vials (e.g. Exetainer®, Labco Limited, High Wycombe, UK) are now commonly used as air sample containers, and procedures have been developed for their use (Plate 2.7). While different sizes are available, 6- and 12-ml septum-capped glass vials are most commonly deployed with gas sample volumes as small as 1 ml being removed for analysis. Such glass vials have screw-on plastic caps with rubber septa. Experience shows that gas tightness is achieved when the cap is screwed on "finger tight", followed by another quarter-turn. Different septa are available (De Klein and Harvey 2012); as the materials differ in their composition and properties, proper septa must be selected with respect to the gas(es) to be stored in vials and then analysed.

Chamber deployment duration should be long enough to allow flux calculation. This is governed by the accuracy of the analytical instrument used for determining the concentration increment over time (i.e. to determine a flux, the change in gas concentration needs to be higher than the standard deviation of typically 10 repeatedly sampled ambient air standards). However, problems may arise that are associated with changes to the chamber physical environment, and the risk of leaks, which increases with deployment time. Therefore, short chamber deployment periods are recommended (De Klein and Harvey 2012), with each deployment not to exceed 2 h in general. Sampling is often carried out at 0, 30, and 60 min (Zaman et al. 2009). The chamber deployment duration also depends on practical considerations including the number of headspace samples to be taken during the enclosure period, the number of simultaneously deployed chambers, and the number of field operators (De Klein and Harvey 2012).

Sequencing and grouping of chamber measurements vary depending on deployment duration, experimental design, and availability of human resources. The number of chambers that can be handled by one operator increases with deployment duration but decreases with the number of headspace samples to be collected and the distance between chamber installations. Chamber size and height, or stacking requirement (tall crops), may also have an impact on the number of chambers an operator can handle. The time interval between sampling two chambers varies, depending on their location, but it is usually ≥60 s (De Klein and Harvey 2012). In the case of a measurement design with repeated treatments, groups of chambers handled together should represent an entire repetition of treatments. Because GHG flux measurement often takes a long time, it is important to sample different treatments within a replicate block in as short a period as possible when there are many chambers to be sampled.

**Amount of headspace air to be sampled**: The greater the headspace volume to be taken, the better the characterisation of accumulation of trace gases and the less biased each individual flux estimate will be. Generally, 3–4 or more gas samples

**Plate 2.7** Gas collecting exetainer for gas storage

are recommended to be taken during deployment, to adequately assess the quality of the calculated flux (detection of outliers and technical problems during handling and analysis of samples), and to account for the increase in non-linear rates of gas concentration with deployment time (De Klein and Harvey 2012; Rochette 2011). Less intensive chamber headspace sampling may be acceptable for certain situations. Any consideration around reducing headspace sampling intensity should be based on minimising the overall uncertainty of the flux estimate. For example, when the spatial variability of fluxes is exceptionally high, it may be preferable to deploy a greater number of less-intensively sampled chambers (two or three samples) to improve plot-level flux estimates, even if this comes at the cost of increased uncertainty in individual chamber estimates. However, if the priority is to generate a representative flux–through the sampling of multiple chambers per plot and assumption of a linear increase in headspace gas concentration, rather than multiple sampling from the headspace of fewer chambers–it is essential to qualify any potential bias introduced by only taking two or three headspace samples per deployment. To reduce the number of samples but still cover the spatial variability, gas pooling techniques are available (Arias-Navarro 2013). Each dataset of four or more headspace samples should be statistically analysed to see if there is non-linearity. At the end of the experimental period, researchers shall summarise this information, provide a percentage of cases when linearity was observed, and then cite this alongside their calculated fluxes. This will provide an indication of the bias–hence confidence–in the results that may have been introduced by assuming linearity in the flux calculation (De Klein and Harvey 2012).

Headspace air sampling usually begins as soon as the chamber is deployed (at time 0), and then in selected intervals: as outlined above (Plate 2.8).

In the case of a measurement design with repeated treatments, groups of chambers handled together should represent the entire repetition of treatments. This avoids temperature-induced biases and enables valid statistical comparisons of fluxes. However, the sampling sequence shall vary between sampling dates, to avoid any potential bias from always sampling in a particular order (De Klein and Harvey 2012). Modern CRDS (Cavity Ring-Down Spectroscopy) techniques for GHG concentration measurements overcome this problem by analysing the gas concentration every few seconds if the analyser is directly connected to the chamber (for more information on the CRDS technique see below, Sect. 2.9).

Daily GHG emissions are often estimated from a single measurement made at the time of the day when the flux is believed to equal its daily mean. For example, in the absence of transient fluxes following a disturbance of soil N2O producing processes (N application, tillage, and rainfall), diurnal dynamics of fluxes are mostly governed by soil temperature where the main production of the gas occurs (De Klein and Harvey 2012). Research has generally indicated that sampling fluxes when temperature in the plough layer is close to its daily mean are often indicative of the average daily flux. However, data by Šimek et al. (2014) show that diurnal variation in flux rates can be very high and difficult to predict. Periodic measurements of the diurnal pattern in soil N2O emission during an experiment are the best way to determine when a single sampling time is representative of mean daily fluxes. However, such

**Plate 2.8** Gas sampling through a syringe using a plastic/metal chamber

measurements require resources that few projects can afford, and temperature in the plough layer remains the most frequently used index for determining the best single time of flux measurement in a day. Moreover, most experimental designs and measurement protocols assume that diurnal emission patterns are the same in all treatments and throughout the year. However, this may not always be the case. For example, if treatments affect the amount of crop residue retention at the soil surface, the time of daily minimum and maximum soil temperature at a given depth will likely differ among treatments. Similarly, placing N fertilisers at different depths can also produce different temporal patterns in surface N2O fluxes (De Klein and Harvey 2012).

Seasonal/annual variations of gas fluxes should also be taken into account. As discussed in detail by De Klein and Harvey (2012), the major problem is related to a short period of flux determination (ca. one hour) done with relatively long intervals (from 1 to 14 days), and the need for integrating the data over a much longer period (season or year). Consequently, it is crucial to select an adequate number and time of sampling events when linear interpolation is used between sampling points for temporal integration of emissions (De Klein and Harvey 2012). Theoretically, the GHG flux measurements shall be done every day but for practical reasons, much longer intervals are often selected. If the GHG peaks of the fluxes can roughly be predicted in advance, then sampling at least twice per week, or ideally daily, is recommended. However, after heavy rainfall events or with other rapidly changing conditions when high emission rates are expected (e.g. freezing–thawing or wetting– drying cycles) measurements should be performed immediately after the event and closely followed for the next day or two because peaks might appear once the diffusional constraint due to surface water has subsided. During periods when fluxes are low (e.g. prolonged drought), measurements should be performed at least once a week. When fluxes have returned to background levels, the sampling interval can be increased further.

Spatial integration of fluxes is extremely important due to the enormous spatial heterogeneity of the soils. Together with the temporal variability, the spatial heterogeneity of fluxes represents one of the most difficult features related to integrated gas flux determination from plots and ecosystems. As suggested by De Klein and Harvey (2012), in experiments that determine emissions from a particular practice, selecting small and uniform areas consistent with the measurements being made will help to minimise interference from spatial heterogeneity in background emissions. The location of these relatively homogeneous areas within a landscape–such as a grazed paddock or cropland–can be determined before the experiment, using preliminary flux sampling. However, while this approach usually helps to reduce uncertainty in estimates of the influence of management effects, it does not account for interactions with other soil factors influencing gas dynamics across a given landscape. The number of replicate measurements can often be reduced if preliminary observations have identified the homogeneity of the experimental site.

To deal with large spatial heterogeneity, 2–3 replicated chambers per plot (treatment) have often been used. However, this depends on the available human and financial resources to collect gas samples and their analysis. For further reading the papers of the Global Research Alliance by De Klein et al. (2020a) on Health and safety consideration, by Harvey et al. (2020) on sample collection, storage and analysis and by Charteris et al. (2020) on deployment and source variability are recommended.

## *2.2.7 Gas Pooling to Address the Spatial Variability of Soil GHG Fluxes*

Soil–atmosphere exchange of GHGs is notoriously variable at spatial scales. Overcoming this variability is a major issue if fertiliser treatments or other management options need to be compared. The spatial variability of soil GHG fluxes is due to small-scale variability of soil properties, soil environmental conditions, and processes of N and C ecosystem turnover driven by microbes and plants (e.g. Butterbach-Bahl et al. 2002). Spatial variability can be addressed best by increasing the number of replicates and by using larger chambers. But as increasing the number of chambers is not always feasible, one may also use the gas pooling technique as outlined in Plate 2.9.

High-tech equipment such as Cavity Ring-down Spectroscopy (CRDS), Gas Chromatograph (GC), and mass spectrometers and 15N labelled fertiliser are used for measuring GHGs, and their isotopic signatures are expensive and require special

**Plate 2.9** Schematic of the gas pooling technique as described in Arias-Navarro et al. (2013). **a** taking of gas samples from five different chambers and mixing of samples within one syringe (**b**). Injecting the mixed gas samples in a vial (**c**) for different sampling times (**d**). Finally, analysis of the gas sample by gas chromatography

technical skills to operate them; therefore, limited field studies have been carried out to quantify GHGs emissions from agriculture worldwide, especially in developing countries. Therefore, it is necessary to identify appropriate methodology and provide suitable guidelines and protocols to help researchers to measure GHGs with greater accuracy and precision.

## *2.2.8 GHG Measurements in Paddy Rice System*

Unlike other field crops, rice is usually grown in flooded conditions. Paddy rice is a large anthropogenic source of CH4. In recent years, it has become evident that there has been a major increase in the use of N fertilisers in rice agriculture, making rice fields a significant source of N2O as well.

The closed chamber method, as described earlier, is commonly used to measure GHGs from rice paddy. In comparison to micrometeorological methods, closed chamber techniques are virtually the only available option because of its ease of implementation, low cost, and high logistical feasibility. The United Nations Framework Convention on Climate Change's (UNFCCC) clean development mechanisms (CDM) also recommend carrying out GHG measurements using the closed chamber method (UNFCCC 2008). However, the design of closed chambers for measuring GHG under rice paddy is different from those used for grassland and cropping systems. The chamber should be equipped with a small fan (battery-operated fan to homogenise the air inside the chamber headspace), a thermometer inside the chamber (to monitor temperature changes during the gas sampling period), a vent stopper, a gas sampling port (preferably a tube connected to a valve) (Plate 2.10), and an air buffer bag (1 l Tedlar bag). This air buffer bag compensates for both higher air pressure caused by increased gas production and lower gas pressure caused by gas sampling (Minamikawa et al. 2015). A rectangular chamber (transparent or opaque), with double deck and adjustable height, covering multiple plants of the area occupied by a single rice hill or two hills, is recommended (Plate 2.11). Chamber height should be higher than the rice plant. For the double-deck chamber, a water seal or a suitable gasket between the base and the chamber is required to ensure the gas-tight connection. The belowground depth of the base should be 10–15 cm.

As discussed above (2.2.4), a double-deck chamber should have three components, i.e. chamber base (made of stainless steel and 15 cm deep for rice) that has a trough shouldered in the top of the base filled with water immediately prior to the basechamber coupling; chamber top facilitating with gas sampling point and a fan; and

**Plate 2.10** Closed chamber used for collection of GHG samples from rice field adopted from Pathak et al. (2013)

**Plate 2.11** A metal chamber of 40 × 40 cm is required to cover at least one hill

an extension that connects the chamber lid and the base. The extension and the lid can be made of polycarbonate or stainless steel (Plate 2.10). Handling of polycarbonate is easier than the stainless-steel chambers. Wrapping the polycarbonate chambers with insulating papers may reduce heat increment inside the chambers. Further, when the rice plants are smaller, only the base and the lid can be used without connecting the extension part. It is critical for rice fields, to insert the base to a depth of about 15 cm to restrict lateral flow of nutrients, particularly N, from outside the chamber and vice versa. After chamber installation, the protocol for collection of GHG samples, and sample storage is similar as described above.

## *2.2.9 Analysis of GHG Samples on a Gas Chromatograph (GC)*

To avoid changes in concentration during storage, the GHG samples collected from field/lab trials and stored in vials are transported to the laboratory and analysed for trace gas concentrations. An over-pressurisation of the sample gas in the vial ensures that no gas from outside can dilute the sample gas. Crucial recommendations for gas sample collection, storage, and analysis are listed in Table 2.3 (Kelliher et al. 2012). Gas chromatography (GC) is mostly used for analysis of trace gases, including N2O, CH4, and CO2 (Plate 2.12).


**Table 2.3** Basic recommendations for GHG sample collection, storage, and analysis (adapted from Kelliher et al. 2012)

**Plate 2.12** A gas chromatograph (**a**) and schematic diagram system for GHG analysis (**b**)

A GC with a sample loop allows the analysis of gas mixtures, and the configuration ensures that the same gas volumes are always analysed under the same condition (pressure and temperature). Besides, gas samples and standard gases are treated always in the same way. Separation of the gas mixture into single gases (CO2, CH4, and N2O) is achieved by passing the sample gas via a carrier gas through a packed column (e.g. a 1/8 analytical column packed with Haysep Q and/or Molsieve). A carrier gas, usually N2, He, or Ar, is used, which passes continuously through the system at a constant flow rate. Standard gas chromatographic procedures allow the quantification of CH4, CO2, and N2O in the same sample. To ensure the same conditions for all samples (samples and standards), gas samples are usually injected into a sample loop at constant temperature and pressure (the loop typically has a volume of 0.5–5 ml). After the separation, the gases are analysed with different detectors.

Methane, like all other hydrocarbons, can be burnt, and this feature is used in a specific detector: a flame ionisation detector (FID). After the gas sample enters the FID, it is burnt creating a proportional number of free electrons that generate a current at the collector electrode, which is passed on as an electric signal to the integration unit. When a GC with FID has an attached system with Ni catalysts for conversion of CO2 to CH4, it can also be used for the analysis of CO2 concentration. Otherwise, a GC equipped with a thermal conductivity detector (TCD) is often used to measure CO2 concentrations. Concentrations of N2O are analysed with a 63Ni-electron capture detector (ECD) operating at column, injector, and detector temperature of 65, 100, and 280 °C, respectively. An anode is inserted into a small, well-isolated, foil-lined box. The carrier gas (recommended Ar + 5% CH4) with the gas sample can pass through the detector. The radioactive 63Ni-foil (ß emitter) delivers electrons in the anode interior. The electrons are drawn by the anode in the middle and are "caught"; the number of caught electrons is determined by the electric pulse frequency at the anode. If an electrophile and electron-catching substance (such as N2O) streams through the space around the anode, it takes up electrons according to its concentration and "electrophilicity". To collect the same number of electrons as before, the electric pulse frequency of the anode must be raised, and this change in pulse frequency is a measure of the amount of the electrophilic substance.

Since the different gases pass through the analytical column at different speeds (e.g. in the order: CH4, CO2, and N2O), it is possible to analyse all three gases in one sample. First, the elution of the column is passed through the FID, and CH4 is successfully captured by the FID. A switching valve (usually a pneumatical switch) will switch the gas stream from the column from the FID to the ECD detector. Depending on the flow rate of the carrier gas as well as the oven temperature of the GC where the analytical column is located, the analysis time of one sample is typically 3–6 min. In addition, a pre-column is often installed in line with the analytical column to capture all slower moving substances. Once all gases of interest have entered the analytical column, usually slow-moving substances still remain in the pre-column. These substances will then be cleaned from the pre-column via a back-flushing mode with the carrier gas. If that is not done, there is a danger that these substances would appear at some later stage and disrupt the analysis of later samples. To perform the switching, usually a second pneumatically operated 10-port valve is used. For further reading the paper of the Global Research Alliance by Harvey et al. (2020) on gas analysis is recommended.

## **2.3 Methods to Quantify GHG Production in the Soil Profile**

So far we have presented methods to quantify GHG fluxes at the soil–atmosphere interface. However, the various gases are produced in the soil profile and there in sites which are suitable for the activity of microorganisms. Thus, when we are talking about gaseous emissions, we are dealing with two processes that go hand-in-hand: (1) the production of GHGs in suitable soil microsites, and (2) the transport of GHGs from the production site to the soil surface. The transport of GHGs is a diffusion process which is governed by a range of variables such as temperature, soil moisture, soil texture, and the properties of the gas in question. With the help of gas diffusion, based on Fick's law, it is possible to calculate the movement from the production site to the soil surface. The production site is often assumed to be close to the soil surface where most of the management takes place but the main production site can also be deeper in the soil profile (Müller et al. 2004). The gas dynamics in the soil profile can be determined by soil air samplers. Various soil air sampling devices have been developed over the years including (a) stainless steel tubes which are blocked at the end but have close to the tip a radial arrangement of holes for soil air intake (Dörr and Münnich 1987), (b) flexible plastic tubes that allow gas diffusion but are impermeable for water. These can be inserted horizontally at a certain depth (Jacinthe and Dick 1996). The advantage of the second system is that the gas production can be assigned to a specific depth, while gas taken in with the first system could have potentially drawn into the sampler from other depths. For the tube samplers principally two different materials, differing in their diffusive properties, are used: silicone or air permeable, hydrophobic, polypropylene ( Accurel®). Both materials can easily be shaped into a coil and inserted at a specific soil depth. However, gas diffusion through silicone is much slower than through Accurel®. Thus, silicone cannot be used for continuous sampling but requires a roughly 24 h equilibration time between samplings (Kammann et al. 2001). This allows discrete gas samplings at a minimum time resolution of approximately one day. The gas diffusion through Accurel® is so quick that continuous sampling is possible (Neftel et al. 2000) which allows for in-field online measurements (Jochheim et al. 2018). The analysis of gas samples is similar to the gas sample analysis from chamber samples.

Plate 2.13 shows an automated setup where Accurel® tubings (Plate 2.13a) are inserted into a soil profile at various depths (Plate 2.13b). The samplers are connected

**Plate 2.13** Air sampler setup using Accurel® tubing with **a** soil air sampler with in- and outlet to allow continuous analysis, note the chicken wire around the sampler is there to protect the material from rodent bites, **b** soil profile setup with soil air samplers (right) and soil moisture/temperature sensors (on left) which are connected to a datalogger, **c** manifold system with quick connector gas sampling ports for different depths, **d** discrete sampling with a syringe and an exetainer vial, first the sample will be taken by the syringe and then the three-way-tap will be turned towards the evacuated exetainer and the gas in the syringe will be transferred to the vial, **e** the manifold can also directly be connected to an autoanalyzer arrangement for automated in situ measurements (see also Sect. 3.2.2 for further information)

**Fig. 2.1** Raw soil gas concentrations of CO2, CH4, and N2O at different depths, analysed with the automated system described in Plate 2.13. (1.11.2019, FACE—Research station, Institute of Plant Ecology, Justus Liebig University Giessen). Highlighted data are used to show the further analysis steps presented in Fig. 2.2

via a teflon tubing to a manifold system at the top of the soil which is fitted with quickconnectors (Plate 2.13c) to allow for discrete sampling using a syringe arrangement (Plate 2.13d) or for connection to an automated arrangement consisting of an LI-COR 8100/8150 multiplexer connected to a CRDS analyzer (Picarro G2508) (Plate 2.13e). For more details on the automated system, see Chap. 3, Sect. 3.2.2.

Figure 2.1 shows a typical output of an in-field measurement campaign. The advantage is that both gas fluxes at the soil surface (if automated chambers are used) together with the soil gas concentrations can be monitored in situ at the same time (see Sect. 3.2.2).

Each air sampler is analysed for a certain period of time (typically 5 min) in a closed-loop system. A decline of the concentration (CO2, N2O), or increase under subambient conditions (CH4) indicates a contamination with ambient air which will be corrected via the following regression analysis. First of all, the results during the time when equilibrium is reached, i.e. between start of sample analysis (t0) and toffset, are discarded. This period is determined by a moving regression analysis from t0 till the end of the sample analysis. The dynamics of the intercept of this moving regression indicates the time (toffset) when the adjustment period is completed (see Fig. 2.2b). The most robust measurement is usually CO2 which will also be used to determine toffset (Fig. 2.2). The concentrations of CO2, CH4, and N2O in the samplers are then determined by a linear regression between toffset and the end of the sample analysis. In the example presented in Fig. 2.2, the resulting concentrations (i.e. the intercept of the regression at t0) were 2700 ppm for CO2, 0.571 ppm for N2O, and 1.494 ppm for CH4.

**Fig. 2.2** Determination of the soil air sampler concentrations (**a**) is based on a moving regression analysis (**b**) (data were taken from the results indicated by the shaded area in Fig. 2.1)

## **2.4 Standard Operating Procedure (SOP) for Gas Flux Measurement**

## *2.4.1 Field Gears and Equipment Needed for GHG Sampling*


## *2.4.2 Step-Wise Procedure (SOP) for GHG Measurements*


**Plate 2.14** A schematic diagram illustrating the collection of gas samples through a large chamber in a pineapple field

the experimental area to protect the field site from animals. In the case of an open grazing system, fence the experimental site 2–3 months prior to treatment application to minimise the effect of animal excreta (urine + dung). If the site is fenced so long in advance, the grazing impact on plants has to be simulated. If possible carry out representative screening of the area including soil analyses, plant analysis, and gas measurements to determine blocks to decide where to place the experimental plots.


## *2.4.3 Gas and Soil Sampling*


**Plate 2.15** Gas collection through a syringe in the field


## *2.4.4 Safety Measures for GHG Sampling*


For further reading the paper of the Global Research Alliance by de Klien et al. (2020b) on safety measures is recommended.

## **2.5 Calculation of GHG Fluxes**

## *2.5.1 Overview*

1. Analyse reference gases (i.e. gases that are available from commercial companies with a known concentration) using a GC to make a calibration curve (Fig. 2.5).

#### 2 Methodology for Measuring Greenhouse Gas Emissions … 45

**Fig. 2.3** Concentration of CO2, N2O and CH4 in the headspace during the incubation

**Fig. 2.4** Calculated emissions of CO2, N2O and CH4 during a 35 day measuring campaign


## *2.5.2 Calibration*

A **calibration** is a procedure to convert the GC output into a concentration unit, typically parts per million (ppm) or parts per billion (ppb). To develop a calibration curve (Fig. 2.5), normally three to four gas standards of known concentration are injected into a GC and analysed. Standard gas containing gas mixtures at increasing concentrations, contained in gas cylinders, can be ordered from commercial gas companies. With increasing gas concentrations, the GC output also increases. Either a linear increase (e.g. for CH4) or a non-linear (CO2 and N2O) increase is observed which can be described by the following equations (Eqs. 2.1 and 2.2). Note, if the increase is linear, the term "a" in Eq. 2.2 is zero and the entire equation is reduced

**Fig. 2.5** Calibration with a linear (left) or quadratic (right) regression line

to a linear regression.

$$\text{Linear regression: } y = a\mathbf{x} + b \tag{2.1}$$

$$\text{Quadratic regression:}\,\mathbf{y} = a\mathbf{x}^2 + b\mathbf{x} + c\tag{2.2}$$

where

*x* = area values (area of the standards, output from the GC), *y* = % or ppm values (from the standards), and *a*, *b*, *c* = regression parameters. Steps of the regression analysis:


## *2.5.3 Calculation of the Gas Concentration and Fluxes*

After all regression parameters are identified, the calculation of the concentration is, depending on the gas, carried out according to Eqs. 2.1 or 2.2.

From concentration to flux

The gas fluxes under a closed chamber are calculated for the duration of the gas sample collection. To do this, the concentrations are determined at several points in time (Fig. 2.6).

Based on the changes in concentration over time, the slope of the regression line at time = 0 is calculated (this corresponds to parameter "*a*" in Eq. 2.1 and parameter "*b*" in Eq. 2.2). Therefore, the slope of the regression line provides the flux rate as concentration/time. The unit of the flux rate is ppm h−1.

Note, typically for N2O and CO2, we observe a positive slope, i.e. emission from soil to the atmosphere and for CH4 under aerobic conditions the slope is negative, i.e. uptake of CH4 by soil.

## *2.5.4 Conversion from Concentration to Mole*

Transformation of concentration (ppm) in mole using the ideal gas law (Eq. 2.3):

$$m = \frac{P \cdot V}{R \cdot T} \tag{2.3}$$

where

*n* = Number of moles of the examined gas *P* = Atmospheric pressure (Pa) [~100,000 = 1000 hPa] (to be measured) *R* = Gas constant (J mol−<sup>1</sup> K−1) [8.314] *T* = Temperature (K) [273.15 + *t* °C] (*t* is the temperature to be measured) *V* = Volume of gas (i.e. N2O, CO2 or CH4) in the chamber (m3). This is calculated by the multiplication of gas concentration with total chamber volume (*V*tot) (ppm × 10−<sup>6</sup> × *V*tot).

#### **Why do we convert gas concentration to its mass?**

Gas concentration does not provide information about the total amount of gas measured or emitted. The smaller the chamber volume, the higher the concentration increases. Imagine a chamber volume of 1 m3 where gas concentration increases at 100 ppm h−1. If the chamber volume is only 0.5 m3, this concentration increase would double up to 200 ppm h−1.

*Think: what would be the concentration change if the chamber volume would be 2 m3?*

#### *The answer is 50 ppm h*−*<sup>1</sup>*

Hence, to know the exact amount of a gas, in addition to the concentration (ppm or ppb), the volume and area of the chamber (*V*tot, *A*), the atmospheric pressure (*P*, the higher the pressure the more gas molecules in the chamber), and the temperature (*T*, higher temperatures decrease the number of gas molecules per volume) must all be considered (see Eq. 2.3).

#### **Example**:

Temperature (*T*): 20 °C Temperature: unit transformation °C to K: 20 + 273.15 = **293.15** K Air pressure (*P*): 100,000 Pa Chamber volume (*V*tot): 0.02 m3 Chamber area (*A*): 0.1 m2 Concentration increase of the gas (CO2) at *t* = 0 (-*C*): 1000 ppm CO2 h−<sup>1</sup>

Molecular weight of CO2: 44.009 g mol−<sup>1</sup>

Note, in the ideal gas law the Volume, *V*, refers to the gas we are interested in, i.e. CO2, CH4, N2O. So, first of all the volume of this gas within the chamber volume, *V*tot, is calculated:

$$V = V\_{\text{tot}} \, \text{\* } \Delta C \, \text{\*} \, 10^{-6} = 0.02 \, \text{\*} \, 1000 \, \text{\*} \, 10^{-6} = 0.00002 \, \text{m}^3 \, \text{h}^{-1}$$

$$\frac{\Delta n}{\Delta t} = \frac{100,000 \times 0.00002}{8.314 \times 293.15} = 0.00082 \,\text{mol} \,\text{CO}\_2 \,\text{h}^{-1} \tag{2.4}$$

Multiplied with molar mass of CO2 (44.009 g mol−1), this corresponds to 0.03607 g CO2 h−<sup>1</sup> or 0.00986 g CO2–C (if only the active C component is applied with a molwt of 12.011 g mol−1). This is now the emission rate from the plot the chamber has covered. To standardise the emission rate, we express it per m−2:

$$F = \frac{0.00986}{0.1} = 0.0986 \,\text{g } \text{CO}\_2-\text{C } \text{m}^{-2} \,\text{h}^{-1} \tag{2.5}$$

The following information is required for flux calculation:


#### **Units and factors**:

ppm: 10−<sup>6</sup> or 1 μl L−<sup>1</sup> ppb: 10−<sup>9</sup> ppt: 10−<sup>12</sup>

#### **Molar masses**:

CO2: 44.009 g mol−<sup>1</sup> C: 12.011 g mol−<sup>1</sup> N2O: 44.013 g mol−<sup>1</sup> N: 14.007 g mol−<sup>1</sup> CH4: 16.043 g mol−1.

For further reading the paper of the Global Research Alliance by Venterea et al. (2020) on flux calculation is recommended.

For further reading the papers of the Global Research Alliance by De Klein et al. (2020a) on data reporting and further calculations, by Dorich et al. (2020) on gap-filling techniques and by Giltrap et al. (2020) on modelling approaches are recommended.

## *2.5.5 Data Analysis*

Data analysis is extremely important to produce realistic emission data. The most appropriate flux calculation method must be selected, and best interpolation of noncontinuous measurements adopted to obtain best estimates of emissions and emission factors (EF) (Venterea et al. 2012) (Table 2.4). It is suggested that both gas analyses and data analysis, including appropriate statistical analysis, are performed in a laboratory equipped with staff familiar with all the required skills.


**Table 2.4** Basic recommendations for GHG data analysis (adapted from Venterea et al. 2012)

**Plate 2.16** A schematic representation of gas sampling, analysis, and interpretation (© FAO/IAEA Mohammad Zaman)

A summary of GHG measurements, analysis, and data interpretation is presented in a schematic diagram below in Plate 2.16.

## **2.6 Analysis of GHG Samples with Optical Gas Analysers**

Gas chromatography is still the most widely used analytical technique for measuring GHGs from chambers (Plate 2.12a, b). It is a well-established, reliable, and robust method; the GC can also be linked to isotope-ratio mass spectrometers (IRMS) for analysis of the abundance of isotopes. In recent years, GCs have become more portable and automated, which makes it possible to run them unmanned in the field in connection with automated chambers. However, the greatest disadvantage of gas chromatography is that one can only measure discrete samples, and it takes several minutes to run a sample, which limits the number of gas samples that can be run (Rapson and Dacres 2014). These disadvantages can be overcome by employing optical gas analysers, which can conduct continuous gas measurements at a high temporal resolution (seconds to Hertz). The measurement principle of optical gas analysers utilises the ability of small molecules (e.g. H2O, CO2, CH4, N2O, and NH3) to absorb infrared (IR) and near-infrared (NIR) light at unique wavelengths. Each molecule has a unique combination of atoms and, as a result, produces a unique IR spectrum when illuminated with IR light, which allows its identification. The light absorption at a specific wavelength, which is measured as light attenuation by a detector, is proportional to the concentration of a gaseous compound (Hensen et al. 2013; Rapson and Dacres 2014).

The general setup of an optical gas analyser consists of a light source from which light travels through the gas sample to an appropriate detector. The path that the light takes between the light source and the detector is called the optical path (Werle 2004). When the optical path lies directly in the outside air, it is called an open path system, whereas when the optical pass is enclosed inside a measurement cell (or cavity) where sample gas has to be pumped into, it is referred to as a closed path system (Hensen et al. 2013; Peltola et al. 2014). Depending on the specific optical technique utilised in an analyser, several mirrors and/or optical filters are added to the optical path to focus the light beam to increase light intensity and the length of the optical path. The main optical techniques employed for quantifying GHG are non-dispersive infrared spectroscopy (NDIR), Fourier-transform infrared spectroscopy (FTIR), photoacoustic spectroscopy (PAS), tunable laser absorption spectroscopy (TLAS), cavity ring-down spectroscopy (CRDS), and off-axis integrated cavity-output spectroscopy (OA-ICOS) (Werle 2004; Hensen et al. 2013; Peltola et al. 2014; Rapson and Dacres 2014).

NDIR and FTIR analysers use light sources that emit broadband IR radiation, e.g. IR lamps or black body light sources. In an NDIR analyser, broadband IR radiation passes unfiltered through the air sample. An optical filter in front of the detector determines which wavelengths are detected and thus which molecules are quantified. NDIR analysers are quite cheap and robust and often are used for quantifying CO2 and H2O in the air (Tohjima et al. 2009; Keronen et al. 2014). In FTIR analysers, the source radiation is constantly modulated by a set of mirrors called a Michelson interferometer containing different combinations of frequencies. For each combination, the amount of IR absorbed by the gas sample is measured. A Fourier transform is then applied to the raw data to calculate the absorption at each wavelength for the complete IR spectrum. This method allows the simultaneous quantification of many different gas species in air. Depending on the resolution of the FTIR analyser, it determines only gas concentrations (low-resolution FTIR) or isotopomers simultaneously (high-resolution FTIR) (Griffith et al. 2012; Rapson and Dacres 2014). In PAS, the light source is often a heated black body, but it can also be a laser. In contrast to NDIR and FTIR, PAS takes only indirect measurements of light absorption. The light passes through an optical filter to pre-select a specific wavelength, and a light chopper periodically "switches" the modulated light on and off before it is directed into the cavity via a mirror. Molecules heat up and expand when they absorb the modulated light, resulting in a pressure rise in the cavity. The light chopping creates pressure variations, which in turn generates acoustic signals that can be detected by microphones. The acoustic signals are proportional to the gas concentration of the target gas species (Iqbal et al. 2013; Rapson and Dacres 2014).

Analysers based on NDIR, FTIR, or PAS can operate with the measurement cell at ambient atmospheric pressure. This fact and the chosen light source reduce material costs and lead to lower prices in comparison to laser-based analysers utilising TLAS, CRDS, or OA-ICOS. Laser-based analysers do not operate with broadband IR radiation, but instead are tuned to the unique absorption line of a specific trace gas (Hensen et al. 2013). The cavity in such an analyser is kept at sub-ambient pressure, which results in narrower IR absorption lines and thus a higher gas species selectivity (Rapson and Dacres 2014). However, it also requires the analysers to be equipped with vacuum pumps and vacuum-proof tubing, tube fittings, and valves. The great advantage is that laser-based analysers are capable of fast and the most sensitive measurements of trace gas concentrations, as well as stable isotope compositions (including isotopomers) in the air (Hensen et al. 2013; Rapson and Dacres 2014; Brannon et al. 2016).

Nowadays, TLAS is the most common laser-based absorption technique for quantifying GHG concentrations in air. Commercially available analysers often employ either tunable diode laser absorption spectroscopy (TDLAS) or tunable infrared laser differential absorption spectroscopy (TILDAS) with quantum cascade lasers (QCL). Detailed information about these techniques is available in Li et al. (2013, 2014). The main disadvantage of TLAS is its susceptibility to instrument noise because the analysers rely on the measurement of a small change in light intensity against a large background light signal. To drastically improve sensitivity, TDLAS and TILDAS analysers are commonly equipped with multipass cells. In multipass cells, the light beam travels several times between aspherical mirrors before exiting the cavity in the direction of the detector, resulting in optical path lengths of several up to hundreds of metres. However, the mirrors are extremely sensitive to changes in alignment and, therefore, more sensitive to vibrations. This has led to the development of highfinesse optical cavities, which allow the build-up of large amounts of light energy in the cavity, boosting analyser sensitivity and robustness, but also decreasing cavity and analyser size. High-finesse optical cavities are the basis of CRDS and OA-ICOS (Rapson and Dacres 2014).

In OA-ICOS analysers, the laser beam enters the cavity at a non-zero angle so that the photons make thousands of passes between high-reflectivity mirrors before leaving the cavity again. This increases the effective optical path length to several thousand metres, enhancing the measured light absorption in comparison to multipass cells. The optical path length depends only on optical losses in the cavity, and therefore a precise laser beam alignment is not necessary, and the analysers are also less susceptible to slight changes in mirror alignments (Bakhirkin et al. 2004; Peltola et al. 2014; Rapson and Dacres 2014; Lebegue et al. 2016). In contrast, CRDS does not consider the absolute absorption intensity by the target gas species, but rather the rate of light intensity decay in the cavity. When the laser is turned on, the cavity is quickly filled with NIR light. As soon as a light intensity threshold is reached, the laser is abruptly turned off, and the light leaks out of the cavity with an exponential decay rate through the cavity mirrors. Additional light absorption in the cavity by the target gas species accelerates the light intensity decay rate, also known as ring-down time. The switching on and off of the laser takes place within microseconds, and the ring-down times are used to determine the concentrations of the target gas species in the cavity. Since the actual measurements take place when the laser is turned off, CRDS is not susceptible to laser intensity fluctuations or absolute laser power. The disadvantages, though, are that the mirrors have to stay very clean, and high-speed detection electronics are required (Rapson and Dacres 2014; Brannon et al. 2016). A laser-based optical technique, which has just been recently commercialised for GHG measurement, is laser dispersion spectroscopy (LDS). It tries to overcome the main limitation of TLAS, the detection of a small signal change against a large background, by measuring molecular dispersion instead of molecular absorption. The measurement signal is encoded in the light phase making the analyser more resilient to dirt and water vapour in the cavity because these compounds only affect light intensity (Nikodem and Wysocki 2012).

The past two decades have seen a rapid development of optical techniques and optical analysers for GHG measurements, and this trend is projected to continue over the coming years. One optical technique that might become more important is enhanced Raman spectroscopy (see Sects. 2.8.2 and 2.8.3). The range of commercially available optical gas analysers for different applications can be quite confusing; thus, Table 2.5 summarises the key points one should consider before purchasing a specific optical analyser.

For a comparison of laser spectroscopic analyses of N2O isotopomers, see Chap. 7, Sects. 7.3.4 and 7.3.5.

## **2.7 Hands-On Approaches Using a CRDS Analyser**

## *2.7.1 Overview of the CRDS Techniques for Determining GHG Concentrations and Soil Fluxes*

As outlined in Sect. 2.6 ("Analysis of GHG samples with optical gas analysers"), the advent of laser-based techniques has enabled real-time measurement of soil gas fluxes in the field. In recent years, CRDS has become of particular interest to researchers due to its ability to measure GHG concentrations with a very high precision in field settings. For example, the Picarro G2508 multi-species gas analyser allows CO2, CH4, N2O, NH3, and H2O to be measured simultaneously. CRDS analysers have three key advantages for soil gas flux measurements in comparison to traditional GC measurements: (1) it enables the detection and measurement of low levels of GHG fluxes. Areas with low emission rates, while not significant in the short-term, may be of great interest to calculate the global GHG budget when extrapolated over longer timescales and large surface areas; (2) multi-species measurements allow for direct correlations between gas species. The processes and sources associated with the emission (or uptake) of GHGs may require more than one axis for proper characterisation; and (3) the high sampling frequency of real-time measurements leads to a better characterisation of soil flux profiles and shorter closure times for soil flux chambers. In addition, the real-time nature of this system provides instant feedback on site selection and omits sampling and storage effects.


**Table 2.5** Key features of optical analysers



#### **Table 2.5** (continued)

#### **Table 2.5** (continued)



#### **Table 2.5** (continued)

The sub-section below describes the fundamentals of CRDS and discusses how a CRDS analyser may be paired with commercial and home-made closed chambers to create a field and/or laboratory deployable flux system.

## *2.7.2 Theory: Near-Infrared Absorption Spectroscopy Fundamentals*

Nearly every small gas-phase molecule (e.g. H2O, CO2, CH4, N2O, etc.) has a unique near-infrared absorption spectrum, and this is generally also true for different isotopologues (12CO2 vs. 13CO2, 12CH4 vs. 13CH4, etc.) (for details, see Chap. 7). At atmospheric pressure, these spectra are very broad and continuous, but at sub-atmospheric pressure the contributions from individual vibrational modes are well resolved so that each spectrum is made up of distinct absorption lines; each line is centred at a characteristic wavelength and possesses an extremely predictable line shape. Under these conditions, it is generally possible to find one or more narrow wavelength frequency ranges where the target molecule has strong absorbance features, and the absorbance contributions from other molecules are minimal. Therefore, the spectrum for each molecular species can be thought of as a unique and very information-rich "fingerprint".

After choosing a specific wavelength range, the concentration of the target gas can be measured by determining light absorption. In principle, the concentration could be inferred from the light loss at the absorbance maximum of the target gas; however, in practice, it is far more useful to scan over a specific absorption line by using a tunable, narrow-band laser to measure absorption in small steps over the relevant frequency range. The main advantage of determining the full spectrum of an absorption line is that one can measure multiple gases at the same time and that it also allows to measure the background level of light absorbance in one step. A measured absorption spectrum *s* can then be described as the weighted sum of contributions from individual gases and from the background light loss (Fig. 2.7) and (Eq. 2.6).

$$ds = b + a\_1 c\_1 g\_1 + a\_2 c\_n g\_n \tag{2.6}$$

where *b* is the background light loss, *g* is the spectrum specific for each gas, *a* is a scaling factor describing the fundamental absorbance strength of the gas, and *c* is the scaling factor the concentration of each individual gas molecule. If the values of *b*, *a,* and *g* are known, then the gas concentrations *c* can be determined from the observed spectrum *s*. Since *a* and *g* are fundamental properties of the respective gas, they can be established in advance. The background light loss *b* is specific for each instrument but is generally highly stable over time.

## *2.7.3 Operational Principle of Cavity Ring-Down Spectroscopy*

Cavity ring-down spectroscopy allows the concentration and isotopic composition of gases to be measured with very high precision. A key feature of CRDS is that

**Fig. 2.7** An example spectrum "*s*" (top left) is the sum of background light loss "*b*", and scaled gas-specific absorbance spectra "*g*1" and "*g*2" (bottom). In this fictional example, *s* = *b* + 1000 × *g*<sup>1</sup> + 500 × *g*<sup>2</sup>

it measures the decay of light in the measurement cell instead of measuring absolute absorption intensities, where the term "ring down" in CRDS is referring to this principle of measuring the light loss in the cavity over time. The cavity itself consists of three high reflectivity mirrors, and the light from a continuous-wave, single-frequency tunable laser diode is reflected between the mirrors (Fig. 2.8). When the laser is turned on, the cavity quickly fills with the circulating laser light. Because the mirrors have slightly less than 100% reflectivity (99.999%), a small amount of light continuously leaks out and is measured with the photodetector. As soon as the light intensity reaches a threshold (within a few tens of microseconds), the laser is abruptly turned off, and the light already within the cavity bounces between the mirrors (about 100,000 times). As the light circulates between the mirrors, light is absorbed by the target gas in the cavity, and the photodetector measures the exponential decay curve, also known as ring downs. In the absence of the absorbing gas, the decay rate is determined solely by the reflectivity of the mirrors.

The ring downs are measured in real time by the photodetector, and the exponential decay curve consists of several thousand individual light measurements. A dedicated signal processor fits this curve to determine the decay rate. To obtain a spectrum, the laser frequency is tuned over a sequence of narrow frequencies, and the ring-down rate is measured for each step (Fig. 2.9). Once a spectrum is obtained, algorithms are used to recover the spectrum contributions from each gas and from background light

**Fig. 2.8** High finesse optical cavity with a three-mirror configuration. Top: no absorbing species in the cavity = long ring-down time. Bottom: absorbing species present in the cavity = short ring-down time

**Fig. 2.9** The laser frequency is stepped, and a ring-down measurement is made at each step, generating a series of exponential decay curves (left). The software fits each curve to discover its decay coefficient. Representing these coefficients as absorbance, we obtain an absorbance spectrum (right). Numbers indicate corresponding decay curves and absorbance spectrum data

loss processes, as previously described. Critically, the decay rates are independent of the initial laser light intensity, and as such, the measurement is not affected by laser intensity fluctuations or absolute laser power.

## *2.7.4 Minimum Detectable Flux (MDF)*

CRDS analysers allow GHG concentrations to be measured with a higher frequency and precision than generally possible with GC techniques. These key advantages allow either to determine GHG fluxes with a higher precision or to achieve the same level of precision within a shorter amount of time. The latter is of importance because a reduction in enclosure time minimises negative effects resulting from the use of a closed static chamber.

Christiansen et al. (2015) investigated the relationship between enclosure time and instrument precision and introduced the concept of minimum detectable flux (MDF) to determine the lower limit of flux rates that can be achieved with a given measurement precision (Eq. 2.7):

$$\text{MDF} = \left(\frac{A\_{\text{A}}}{t\_{\text{c}}}\right) \left(\frac{VP}{ART}\right) \tag{2.7}$$

where *A*<sup>A</sup> is the analytical precision in ppm, *t*<sup>c</sup> is the closure time of the chamber in hours, *V* is the chamber volume in L, *P* is the atmospheric pressure in Pa, *A* is the chamber area in m2, *R* is the gas constant in L K−<sup>1</sup> mol−1, and *T* is the ambient temperature in K. The equation describes that the MDF can be reduced by either increasing the enclosure time or by increasing the analytical precision (reducing the analytical uncertainty).

Nickerson (2016) highlighted the practical importance of the increased measurement frequency of CRDS analysers since an increase in measurement number *n* will reduce the standard error *A*SE compared to the raw precision *A*<sup>A</sup> of the CRDS analyser (as *A*SE = *A*a/ √*n*). Therefore, he suggested that the effective MDF should consider the measurement/sampling frequency (Eq. 2.8):

$$\text{MDF} = \left(\frac{A\_{\text{A}}}{t\_{\text{c}}\sqrt{\frac{t\_{\text{c}}}{p\_{\text{s}}}}}\right)\left(\frac{VP}{ART}\right) \tag{2.8}$$

where *p*<sup>s</sup> is the sampling periodicity in hours. For a CRDS analyser, the sampling periodicity is about every 8 s. The GC measurements are based on discrete samples, and the sampling frequency can vary significantly based on the research focus, but typically samples are taken every 5–15 min. Consequently, real-time CRDS analysers achieve a higher precision for a flux estimate than GC measurements, given the same chamber system, deployment time, and assuming a similar raw precision for the two techniques (Fig. 2.10). Or in other words, a prescribed MDF can be achieved within a shorter amount of time with CRDS compared to GC measurements.

**Fig. 2.10** Effect of sampling/measurement frequency on the time needed to reach a certain CH4 minimum detectable flux (MDF). Real-time CRDS analysers report a concentration measurement about every 8 s, while discrete sampling for GC measurements commonly occurs every 5–15 min (i.e. 300–900 s). Due to the higher integration time, the CRDS analyser achieves a prescribed MDF within a shorter amount of time (an MDF of 0.1 μmol CH4 m−<sup>2</sup> h−<sup>1</sup> is reached within 700 s with CRDS and within about 2300 s for 5 min interval GC sampling). The illustration assumes that the CRDS and GC measurements show the same raw precision. MDF was calculated on the following parameters: *<sup>A</sup>*<sup>a</sup> <sup>=</sup> 0.01 ppm, *<sup>V</sup>* <sup>=</sup> 0.5 m3, *<sup>P</sup>* <sup>=</sup> 101325 Pa, *<sup>A</sup>* <sup>=</sup> 1 m2, *<sup>R</sup>* <sup>=</sup> 8.31448 J K−<sup>1</sup> mol−1, and *T* = 295 K (figure adapted from Nickerson 2016)

## *2.7.5 Selecting the Appropriate Flow Path*

CRDS analysers have a continuous flow-through design that pulls gas through the analyser at a fixed flow rate (typical flow rate for a concentration CRDS analyser is ~250 ml min−1). Automated pressure control loops within the analyser can be adjusted to gradual changes in input pressure and flow, but under normal operation, gas enters and leaves the analyser at the same flow rate. This design characteristic makes it suitable for use in recirculation experiments. Depending on the chamber design, the CRDS analyser can act as the primary source of flow or as a secondary sampling source, referred to as in-line or parallel, respectively.

**Fig. 2.11** Flow path of an in-line arrangement with a CRDS analyser in combination with a recirculation pump and a soil chamber. Additional communication or control cables are not shown

## *2.7.6 In-Line Flow Path*

In this setup, the CRDS analyser is typically the primary source and control of flow in the recirculation setup (Fig. 2.11). It is commonly used with custom-made chambers or commercial chambers that do not employ their own flow control. When designing such a setup, users need to be mindful of any restrictions or rapid changes of flow that may occur during measurement. Inlet tubing length should not exceed 15 m and should have appropriate filtration in place to protect the analyser from dust and liquid water (see the next section). The CRDS analyser acts as the primary data repository for concentration and external sensor data (chamber temperature and pressure) and can control and coordinate chamber operation (if available).

## *2.7.7 Parallel Flow Path*

The parallel flow arrangement is typically used when the analyser acts as a complementary or additional instrument in an existing recirculation setup. The user may have an existing flow setup that is controlled by a primary pump, or they may be using an existing chamber-analyser system (e.g. LI-COR LI-8100A, for details, see Chap. 3, Sect. 3.2.2). Since the CRDS analyser is no longer the primary source of flow control, it is essential that the flow rate of the main recirculation setup is higher than that of the CRDS analyser. The greater the difference between the two, the smaller the influence of the CRDS analyser will be.

## *2.7.8 Multiple Chambers*

It should be noted that many commercial and homemade chambers often come with multiplexers–the ability to deploy and connect multiple chambers to a single analyser. A CRDS analyser can be used with a multiplexer in both the in-line or parallel arrangements.

The connection location of the CRDS analyser should follow the same approach as shown in Figs. 2.12 and 2.13. In either case, the multiplexer is located in the place of the chamber.

**Fig. 2.12** Flow path of a parallel arrangement with a CRDS analyser, a recirculation pump, and a mobile soil gas survey system (LI-COR LI-8100A). Additional communication and control cables are not shown. Note the higher flow rate of the main recirculation setup at 1000 ml min−1, and the parallel connection of the CRDS inlet and return on the return flow of the LI-COR analyser

**Fig. 2.13** Flow path for a multiple chamber setup. Deployments with multiple chambers are popular when a large surface area is to be monitored over a long period of time

## *2.7.9 Calibration*

The concentration readings of a CRDS analyser behave very stable over time, and therefore, it is not necessary to calibrate the concentrations daily or weekly. However, it is advisable to check the accuracy and performance of a CRDS analyser on a monthly to yearly basis. This can be done by measuring standards from compressed gas cylinders. The concentrations measured by the CRDS analyser are then compared to the accepted value of the gas cylinder. The quality/accuracy of the final analyser calibration reflects (1) the number of standards, (2) the compositional range of the standards, (3) the method of standard measurement, and (4) the accuracy and precision of the standard itself:


## *2.7.10 Advanced Application Considerations: Filtration of Gas Samples*

The precision of an absorption measurement within a CRDS analyser can degrade under the presence of foreign particles that scatter or absorb light. To overcome this potential issue, the inlet of a CRDS analyser is generally equipped with two 4 nm particulate filters. One is user-replaceable, and the other can only be replaced in a cleanroom environment. To prevent long-term damage to the sample handling of the analyser and maintain a healthy cavity, external filters are recommended.

Of primary concern are particles that can potentially be pulled into the sample handling of the analyser via the inlet port. In order to select and build a proper filtration system, the user must first assess which, if any, particles are potentially present in the field. Arid environments with high wind speeds or other mechanical means of increasing particle load (walking, driving, animals, etc.) pose a much greater danger to the analyser than vegetated surfaces. In most cases, a cheap, user-replaceable Acrodisc®, 0.2–1 μm is enough to remove most particles. The frequency of replacement should be dictated by the particle load of the gas sample and the flow rate through the filter.

## *2.7.11 Liquid Water and Water Vapour*

One of the biggest risks to any field-based system is exposure to liquid water. In this section, we focus on the introduction of liquid water through the inlet of the analyser as a result of (1) condensation or (2) actual sampling of liquid water (precipitation, ground saturation, etc.).

Long-term deployments of CRDS analysers can subject them to environments with temperature variations that exceed the operating range of the system (10– 35 °C). Custom or commercially available enclosures are used to provide a measure of temperature control and/or protection from the elements (rain, snow, wind, etc.). The temperature difference between the exterior and interior of the enclosure can lead to condensation in sampling lines. This problem is particularly evident in scenarios where hot and humid gas is brought into a cold enclosure. To reduce the likelihood of condensation, desiccants such as Drierite or magnesium perchlorate can be used to reduce the water vapour level in the sample gas. Such desiccants are also practical if water vapors need to be reduced below 4 vol%, the operational limit for a CRDS analyser.

If condensation, or the presence of liquid water through direct sampling, cannot be avoided, then a water trap must be used before the inlet of the analyser. The volume of a water trap should be appropriate to handle any input of liquid water. It is also possible to install water traps with automatic draining capability so that the system may operate in a user-free fashion. It is important to note that a water trap will add additional volume to the recirculation system, something that needs to be considered when flux calculations are performed.

## *2.7.12 CRDS-Specific Considerations*

A CRDS analyser is a laser absorption spectrometer. Under specific conditions, the composition of the sample gas may have unintended effects on the spectra. If levels of ethane, ethylene, acetylene, or hydrogen sulphide exceed atmospheric levels, the user should consult manufacturer documentation to determine the effects of these

**Fig. 2.14** Example setup of a CRDS analyser for field deployment in a forest environment. The copper scrubber is in place as trace amounts of H2S are expected, the water trap is used to minimise the change of liquid water entering the system due to ground saturation during long rainfall events, and the particulate filter is there to prevent particles that may enter the air stream. Not shown on this image is an enclosure for the system and a stable supply of power

species on their measurements of greenhouse gases. In many cases, the installation of an additional scrubber may be used to remove an interfering species. One such example is the removal of hydrogen sulphide using laboratory-grade copper filings (Malowany et al. 2015). Figure 2.14 illustrates how a recirculation setup can be adjusted for specific field conditions by adding a particulate filter, a water trap, and a copper scrubber.

The composition of the carrier gas, specifically any significant changes to the N2–O2 ratio of the carrier gas, will have pressure broadening effects on the spectra (Nara et al. 2012). This can result in a degradation of accuracy. The installation of an additional oxygen sensor inside a CRDS analyser, or any external oxygen sensor, can help the user detect changes in the N2–O2 ratio and apply a correction. The manufacturer also offers a pre-configured mode for pure N2 carrier gases. It should be noted that major changes to Argon (Ar) or Helium (He) will have the same effects. More importantly, a pure He carrier can permanently damage a CRDS analyser.

## *2.7.13 Datalogging and Flux Processing*

All concentration data obtained using a CRDS analyser is automatically stored on the hard drive of the analyser. This raw data can then be processed using the CRDS software or other, third-party compatible software that is usually supplied by the chamber manufacturer to obtain flux values. In order to convert concentrations into fluxes, the software will ask the user to enter the chamber and tubing volume, chamber surface area, chamber and soil temperature, and chamber pressure. Depending on the software package used, it may be possible to monitor the accumulation of the gases in real time or apply post-processing using three or more fitting algorithms: linear (Eq. 2.9), quadratic (Eq. 2.10) (Wagner et al. 1997), or Hutchinson and Mosier (1981) (Eq. 2.11).

2 Methodology for Measuring Greenhouse Gas Emissions … 69

**Linear**:

$$y = \text{slope} \times X + \text{background} \tag{2.9}$$

**Quadratic**:

$$\mathbf{y} = \mathbf{a} \times \mathbf{X}^2 + \text{slope} \times \mathbf{X} + \text{background} \tag{2.10}$$

#### **Hutchinson and Mosier**:

$$\mathbf{y} = \text{background} + \text{step}(1 - \mathbf{e}^{-\text{slope} \times X/\text{step}}) \tag{2.11}$$

Custom fitting options are available in certain software packages (Fig. 2.15).

CRDS analysers are compatible with the streaming of data to external datalogger using RS232, Ethernet, or analogue communication protocols. In addition, it is also possible to incorporate external sensor data such as pressure and temperature probes within the soil flux chamber. Detailed support and integration documentation are available from the manufacturer.

**Fig. 2.15** Example of a Hutchinson and Mosier fitting algorithm applied to a measurement of H2O on a CRDS analyser

## **2.8 Enhanced Raman Spectroscopy of Greenhouse Gases**

Enhanced Raman spectroscopy emerged in recent years as a powerful analytical tool for highly selective identification and simultaneous quantification of multiple components in complex gas mixtures (Hanf et al. 2014; Frosch et al. 2013a; Jochum et al. 2015a; Hippler 2015; Kiefer 2015; Knebl et al. 2017). Conventional Raman spectroscopy was already discovered in the 1920th (Long 2002; Smekal 1923; Raman and Krishnan 1928; Landsberg and Mandelstam 1928; Kramers and Heisenberg 1925), but its application in real-world applications was very limited until recently, due to the inherently weak signal intensities and technical limitations.

## *2.8.1 Raman Spectroscopy of Gases*

Raman spectroscopy is based on inelastic scattering of light. When monochromatic laser light is guided through a gas sample, most photons pass the analyte volume without any interaction and thus provide no gas specific chemical information. The scattering cross section σ is only around 10−<sup>27</sup> cm2 for typical gases. Thus, only a small fraction of the photons is scattered by the gas molecules which also have a very low number density. The major part of the scattered radiation results from elastic scattering (Rayleigh scattering) and has the same frequency as the excitation light. Inelastic scattering occurs with approximately three orders of magnitude smaller intensity. The frequency shift of the inelastically scattered light is caused by the energy transfer in the rotational, vibrational, or rotational–vibrational states of the molecules and is thus the basis of the unique chemical selectivity of Raman spectroscopy. Raman scattering and the more established IR absorption are complementary effects that probe molecular vibrations and rotations. Raman scattering depends on the changes in the polarisability of the molecule during the rotation/vibration, while IR absorption relies on changes in the (permanent) dipole moment. Thus, IR absorption spectroscopy is not sensitive to diatomic homonuclear molecules such as oxygen (O2), hydrogen (H2), or nitrogen (N2) which provide no permanent dipole moment for light interaction. In contrast, inelastic Raman scattering does not require a permanent dipole moment of the molecules, and the specific molecular transitions must not be matched with the excitation wavelength. Thus, Raman spectroscopy is a powerful method for the analysis of all but noble gases with only one monochromatic laser source. A whole suite of biogenic gases, including CO2, N2O, O2, CH4, H2, and N2, can be identified and quantified simultaneously (Fig. 2.16).

**Fig. 2.16** Raman spectrum of the rotational and rovibrational bands of a complex mixture of biogenic gases, including CO2, N2O, O2, CH4, H2, and N2. Reprinted with permission from Hanf et al. (2015a). Copyright 2015 American Chemical Society

## *2.8.2 Enhanced Raman Gas Spectroscopy*

Conventional Raman spectroscopy is an extremely weak process. Recent advances in laser as well as detector technology, new highly efficient optical filters, and optical components develop Raman spectroscopy into a comprehensive technique for multigas analysis. The Raman Stokes intensity *I*<sup>R</sup> depends on the laser intensity *I*0, the angular frequencies of the laser ω<sup>L</sup> and the scattered light ω<sup>R</sup> as well as the polarisability α of the molecule, and the number *N* of molecules of the measured gas (Eq. 2.12).

$$I\_{\mathbb{R}} = \text{const} \ast N \ast I\_0 \ast (\omega\_{\mathbb{L}} - \omega\_{\mathbb{R}})^4 \ast |\alpha|^2 \tag{2.12}$$

**Fig. 2.17** Comparison of conventional Raman spectroscopy (**a**) and fibre-enhanced Raman spectroscopy (**b**). Conventional Raman setups suffer from the limited interaction volume of excitation light and analyte gas (**a**). In FERS, the excitation light is guided over an extended range within the hollow core of the optical fibre, which also functions as minimised analyte container (**b**). An excellent light–analyte interaction is achieved in fibre-enhanced Raman spectroscopy and a high number of molecules contribute to the Raman signal (Frosch et al. 2013b). Reprinted with permission from Frosch et al. (2013b). Copyright 2013 American Chemical Society

The originally small inelastic scattering signals of the gas molecules must be increased with the development of enhancement techniques based on the above-given parameters in order to achieve high sensitivities. Raman gas sensors are nowadays based on specific multi-pass cavities, high-pressure cells, or the application of highpower lasers (Li et al. 2008; Kiefer et al. 2008; Schiel and Richter 1987). Two very promising techniques, which were recently developed, are fibre-enhanced Raman spectroscopy (FERS) (Hanf et al.2014; Knebl et al. 2018) and cavity-enhanced Raman spectroscopy (CERS) (Frosch et al 2013a).

In FERS, the number *N* of gas molecules that contribute to the Raman signal (Eq. 2.10) is extremely increased. In conventional Raman spectroscopy, the signal is excited and collected from a small scattering volume (Fig. 2.17a) (Frosch et al. 2013b). This limitation can be overcome with the help of hollow-core optical sensor fibres. In elaborated micro-structured hollow-core optical fibres, the light can be guided with extremely low attenuation within a certain spectral range (Russell 2003; Knight 2003; Hartung et al. 2015; Yan et al. 2017a, b). By filling the gas in the same inner hollow core of the fibre, light and analyte are both guided in the same volume and thus the analyte is efficiently excited over an extended path length (Fig. 2.17b) and the number of scattering molecules is strongly increased (Frosch et al. 2013b). In other words, the Raman fibre sensor can be seen as optimised cuvette and provides strongly enhanced analytical sensitivity (Hanf et al. 2014; Frosch et al. 2013b; Knebl et al. 2019; Jochum et al. 2016; Hanf et al. 2015a; Boegoezi et al. 2015; Sieburg et al. 2019). A highly efficient FERS setup was developed for gas analysis (Figs. 2.17 and 2.18).

In CERS, the weak inelastic scattering signals of the gas molecules can be increased by up to six orders of magnitude with the help of a high finesse optical cavity. One of the developed gas sensors consists of a miniaturised laser diode with λexc. = 650 nm (Frosch et al. 2013a; Keiner et al. 2013, 2014) which is passively frequency locked and feedback coupled to a high finesse cavity, enabling a power

#### 2 Methodology for Measuring Greenhouse Gas Emissions … 73

**Fig. 2.18** Design of the optical setup for fibre-enhanced Raman spectroscopy (FERS) (Hanf et al. 2014) consisting of Laser, telescope (T), long-pass dichroic beam splitter (DC), objective lens (OL), fibre adapter assembly (A1 and A2), hollow-core photonic crystal fibre (PCF), power metre (PM), pinhole (P), edge filter (E), aspheric lens (L), spectrometer (SPEC), and CCD detector. Reprinted with permission from Hanf et al. (2014). Copyright 2014 American Chemical Society

build-up to 100 W (Frosch et al. 2013a). The cavity components are aligned for spatial mode matching of the input beam and the Gaussian beam supported by the power build-up cavity (PBC), while the facet of the laser diode helps in stabilising mode matching by spatial filtering (Frosch et al. 2013a). Such arrangement of the PCB is extremely stable to mechanical vibrations (Frosch et al. 2013a), and it has been shown that concentration fluctuations of about 50–100 ppm can be monitored within measurement times of one second (Frosch et al. 2013a).

Main advantages of enhanced Raman spectroscopic gas sensing include the following:


**Fig. 2.19** Combined Raman spectra containing the rovibrational bands of 14N2, 14N15N, 16O2, 18O16O, 18O2, 16O2, 12C16O2, 13C16O2, and 12C18O16O. Adapted with permission from Knebl et al. (2019). Copyright 2019 American Chemical Society


## *2.8.3 Enhanced Raman Spectroscopic Analysis of Greenhouse Gases*

Research regarding enhanced Raman spectroscopy of biogenic gases can be divided into three areas: (i) The capability for unambiguous identification and quantification of various gas mixture components to follow the formation, spread, and exchange as well as storage of gases (Frosch et al. 2013a; Jochum et al. 2015a, b; Keiner et al. 2013, 2014, 2015a, b; Sieburg et al. 2017, 2019; Hanf et al. 2015b). One focus is the analysis of the respiratory quotient (RQ, CO2 released per O2 consumed during respiration) as an indicator of changes in substrate use and metabolism (Jochum et al. 2015a; Keiner et al. 2013; Sieburg et al. 2017; Hanf et al. 2015b); (ii) the monitoring of stable isotopes (12/13CO2, 14/15N2, 14/15N2O, and 16/18O2) alongside unlabeled gases as tracers for specific pathways (Knebl et al. 2019; Keiner et al. 2014, 2015a, b); and (iii) the unique ability for direct quantification of nitrogen at natural background to study pathways of the nitrogen cycle (e.g. denitrification, N2 fixation) (Keiner et al. 2015a; Kumar et al. 2018; Jochum et al. 2017).

The RQ value was analysed as an indicator of changes in plant metabolism under drought stress (Fig. 2.20) (Hanf et al. 2015b). It was discovered that pine (*Pinus sylvestris*) can switch from carbohydrate-dominated respiration to a mixture of substrates during several days of drought stress, but spruce (*Picea abies*) cannot (Hanf et al. 2015b). The onsite analysis of depth profiles of soil gases in the Hainich critical zone exploratory showed that the concentrations of O2 and CO2 were largely decoupled, and complex processes in previously uncharacterised environments can be studied (Sieburg et al. 2017). The ability to monitor the inert tracer sulphur hexafluoride (SF6) alongside biogenic gases under consideration allows for thorough online gas leakage correction to avoid under- or overestimation of biological activity such as respiration or photosynthesis (Fig. 2.21) (Jochum et al. 2015b).

The discriminatory power of Raman spectroscopy was used to monitor several stable gas isotopes simultaneously to investigate the labelling of young poplar trees

**Fig. 2.20** Example for monitoring dark respiration rates of a branch of untreated pine. The individual respiration rates (both O2 and CO2) are used to calculate subsequent RQ values (Hanf et al. 2015b). Adapted with permission from (Hanf et al. 2015b). Copyright 2015 Royal Society of Chemistry

under aphid infestation with 13CO2 to analyse the possible incorporation of 13C in defense compounds (Keiner et al. 2014, 2015b). A combination of 13C-labelling and RQ analysis was applied to investigate the microbial degradation of 13C-labelled benzene in soil against the background of the heterotrophic soil respiration (Jochum et al. 2015a). By combining 13CO2 and 12CO2 as well as 18O2 and 16O2 measurements in one setup, it was proposed to use carbon dioxide and oxygen isotopologues to track and disentangle different overlaying processes and to help elucidating the contributions of photosynthesis, photorespiration, and respiration to the net gas exchange of plants (Knebl et al. 2019).

The nitrogen evolution was continuously monitored over the stepwise enzymatic denitrification of labelled and unlabeled nitrate by *Pseudomonas stutzeri* (Fig. 2.22) (Keiner et al. 2015a). The simultaneous quantification of the whole gas phase also enabled the contactless and sterile online acquisition of the pH changes in the *P. stutzeri* culture by the stoichiometry of the redox reactions during denitrification and the CO2-bicarbonate equilibrium. Continuous pH-monitoring–without the need to insert an electrode into a sterile solution–elucidated an increase in the slope of the pH value coinciding with an accumulation of nitrite, which in turn led to a temporary accumulation of N2O, due to an inhibition of N2O reductase (Keiner et al. 2015a). The gas quantification was complemented with the analysis of nitrate and nitrite concentrations for the online monitoring of the total nitrogen element budget (Fig. 2.23) (Keiner et al. 2015a). In an investigation of the thiosulfate- and hydrogen-driven autotrophic denitrification by a microbial consortium enriched from groundwater of an oligotrophic limestone aquifer, the turnover reactions of electron donors (thiosulfate and H2) were traced, as well as electron acceptor (nitrate), gaseous intermediates, and end products (15N2, 15N2O, CO2, H2, 14N2, and O2) in the headspace, using Raman gas spectroscopy (Kumar et al. 2018). N2 production and H2 consumption rates were calculated under denitrifying conditions and followed the electron donor usage of the bacterial consortium. Recently, the biological nitrogen fixation of a *Medicago sativa*–*Rhizobium* consortium was, for the first time, directly investigated at natural background and without a proxy or isotopic labelling, by continuously

#### 2 Methodology for Measuring Greenhouse Gas Emissions … 77

**Fig. 2.23** Concentration courses of nitrate, nitrite, nitrous oxide, and dinitrogen during the successive reduction NO3 <sup>−</sup> → NO2 <sup>−</sup> → NO → N2O → N2 (Keiner et al. 2015a). Continuous Raman gas concentrations of 15N2, 15N2O, and CO2 are represented by solid lines. The total nitrogen balance from all nitrogen components was calculated and is given as grey line (*N*tot) (Keiner et al. 2015a). Adapted with permission from Keiner et al. (2015a). Copyright 2015 Elsevier

analysing the amount of atmospheric N2 in static environmental chambers (Jochum et al. 2017).

Enhanced Raman gas spectroscopy combines the unmatched analytical prowess of Raman spectroscopy with the enhancement of small signals through the sophisticated use of optical cavities (CERS) and hollow-core optical fibres (FERS). Enhanced Raman spectroscopy is a powerful technique for simultaneous multigas analysis, including N2, N2O, O2, H2, CH4, CO2, and stable isotopes (13C, 15N, and 18O). This high selectivity enables the study of complex gas exchange processes, including pathways of the nitrogen cycle.

## **2.9 GHG Fluxes from Aquatic Systems**

Following the first set of IPCC guidelines being produced (IPCC1995), further developments were needed to improve national inventory calculations. These included a quantifiable way to determine N2O production after the leaching and runoff of N from agricultural soils, in order to improve the methodology for calculating annual N2O emissions from agricultural soils at a national level (Mosier et al. 1998). The second phase approach assumes that all N cycles within 1 year and makes no allowance for potential sequestration within the soil, which may be subsequently released on longer timescales (Mosier et al. 1998).

The approach taken in the second phase of inventory development for determining N2O emissions from agricultural soils was to include direct emissions from agricultural fields, direct emissions from animal production systems, and indirect emissions. Indirect emissions result from volatilisation and subsequent deposition of NH3 and NOx, leaching and runoff of N, and human consumption of crops followed by sewage treatment. Nitrogen leaching and runoff from agricultural systems (NLEACH) include N applied as fertiliser (NFERT; kg N year−1), and N derived from animal excretion (NEX; kg N year−1) so that manure produced during grazing is captured. The fraction of fertiliser and excreta N that moves through the leaching and runoff pathways (FRACLEACH) was estimated by Mosier et al. (1998) to range from 0.1 to 0.8. Some countries have developed country-specific values for FRACLEACH (Thomas et al. 2005). Thus, the flux of N leached (NLEACH, kg N year−1) (Eq. 2.13) is

$$\text{NLEACH} = \text{[NFERT} + \text{NEX]} \times \text{FRACLEACH} \tag{2.13}$$

Production of N2O occurs when the leached N moves into the groundwater and surface drainage, rivers, and estuaries with emission factors for these three zones designated as EF5-g, EF5-r, and EF5-e, respectively (Mosier et al. 1998). The sum of these three components provides the N2O emission factor (EF5) for N2O originating from NLEACH, where N2O (l) is the N2O emissions associated with agricultural N lost via leaching and runoff (kg N year−1) (Eq. 2.14):

$$\mathrm{N\_2O(l)} = \mathrm{NLEACH} \times \mathrm{EF\_{\mathcal{S}}} \tag{2.14}$$

Supersaturation of N2O in ground and surface waters occurs due to either N2O leaching from the soil or because microbial processes utilise NLEACH to form N2O in the ground and surface waters.

The approach taken to determine EF5 seeks to compare the emission based on the N loading, which is comparable with calculating direct emission factors from agricultural soils. For soils, a direct emission factor is determined by applying a quantity of N to plots and measuring the resulting N2O emissions relative to control plots. A different approach has been taken to determine EF5. Briefly, the N2O concentration in a water sample is measured, and the degree of N2O saturation is determined. The EF5 value is then determined based on the ratio of N2O-N: NO3 <sup>−</sup>-N or on the basis of assumed N2O emissions using the degree of saturation to predict a flux. To use the IPCC methodology to determine EF5 values requires that the concentration of dissolved N2O is first established so that the flux of N2O can be determined.

## *2.9.1 Determining Dissolved N2O Concentrations*

There are several methods in the literature that are available to measure dissolved gas depending on the source of the water sample. Water samples may be taken from groundwater, drains, and open water bodies such as rivers, lakes, and estuaries. When sampling water bodies for dissolved gases, it is important to also record the water temperature in situ, the atmospheric pressure, and to take some gas samples of the atmosphere at the sampling site. Atmospheric pressure will vary with altitude and weather conditions. A further requirement is to know the salinity of the water being sampled since this also affects the solubility of gases.

Nitrous oxide produced at depth within a soil can be readily dissolved in soil pore water or water draining from the soil in question. When water samples are collected using automated samplers, there remains the need to prevent degassing of the sample. In an ideal situation, drainage water samples would be collected as drainage occurred. However, this is not always feasible. Roper et al. (2013) developed a technique to facilitate dissolved N2O sampling from automated water sampling bottles. A 10 ml volumetric pipette is placed, inverted inside the sample bottle, within the autosampler. The inversion of the pipette allows for simple modifications for length and enables faster filling while simultaneously minimising turbulence during filling (Roper et al. 2013). The inverted pipette means that the water contained in the pipette has a reduced surface area compared to the larger sample bottle. A water sample for dissolved gas determination is taken manually by withdrawing 5 ml of the 9 ml of water from inside each pipette using a syringe equipped with a 20-gauge 30.5 cm needle (Popper®, Fisher Scientific) and placing the sample in a 12 ml Exetainer (Labco International, UK). The Exetainers were previously flushed, using ultrapure helium, and brought to atmospheric pressure with helium in advance of water sampling. Exetainers also had a microbial inhibitor, placed inside prior to evacuation, to prevent microbial activity altering dissolved N2O concentrations. Generally, the microbial inhibitor of choice is mercuric chloride (HgCl2). The inhibitor must not physically or chemically alter the water sample in a manner that could alter the N2O concentration. Once sampling is complete, the dissolved N2O embodied in the water equilibrates with the gaseous phase in the headspace of the Exetainer. It is thus important to record the laboratory temperature when this equilibrium is attained prior to sampling the gas headspace.

A headspace sample equilibration technique can also be used to sample dissolved N2O in groundwater. Clough et al. (2007a) used this technique to study dissolved N2O in water samples obtained from piezometers. Using a Masterflex L/S portable peristaltic pump (Cole Parmer, Vernon Hills, Il, USA) with the pumping rate adjusted to minimum flow, to avoid out-gassing, water was pumped from the piezometers (Plate 2.17). The water was pumped to waste, until a volume of water equal to several internal piezometer volumes had been pumped, and then diverted into a 60 ml plastic syringe, by way of a two-way stopcock, until 20 ml had been collected. The resulting 20 ml sample was then injected into a pre-evacuated (−0.93 atm) 160 ml serum bottle fitted with a rubber suba seal. Then using a helium-flushed syringe, fitted with a stopcock and hypodermic needle, connected to a helium-filled Tedlar bag, helium was injected into the serum bottle until atmospheric pressure was reached. This was visually determined by inverting the serum bottle so that the water sample covered the helium needle inlet: when bubbling ceased, the serum bottle headspace was at atmospheric pressure. The sample is thus ready for equilibration (Plate 2.18).

A third example of where water samples are commonly taken for headspace equilibration is that of open waterways. For example, Beaulieu et al. (2012) measured dissolved N2O in river water by taking 140 ml polypropylene syringes equipped with

**Plate 2.17** Groundwater being sampled from a piezometer tube using a peristaltic pump. Note the syringe is kept under water to prevent any contamination by air while it is being filled

stopcocks, and slowly filling these with water at 5 cm depth. The water sampling was performed slowly to avoid degassing the sample water. These samples were stored under water during transit to the laboratory whereupon a 20 ml headspace was created by transferring 20 ml of high purity helium to the sample syringe.

**Plate 2.18** A water sample has been previously injected into the serum bottle. Here, the He is being injected, with the bubbles indicating He gas is still entering the serum bottle, once bubbling ceases the headspace will be at atmospheric pressure, the needle will be removed, and equilibration of the water and gas phases can occur

In all these previous examples, headspace equilibration was achieved by gently shaking the sample for >5 min so that the N2O dissolved in the water sample equilibrates with the gas phase in the headspace. The addition of a microbial preservative prevents N2O consumption and/or production of N2O. This equilibration, as we will see below, is temperature-dependent. So, it is important to also measure the laboratory temperature at which the equilibration is performed. This means that the water in the sample needs to be at the same temperature as the laboratory, so if the sample has been in a fridge it must come to room temperature.

After shaking and equilibration, the headspace gas can then be sampled and analysed on a gas chromatograph (GC) to determine the N2O concentration. Typically, this involves using a syringe equipped with a stopcock and a hypodermic needle to transfer a suitable volume of gas to a GC sample vial, for example, an Exetainer. In the first example above, the water was equilibrated in an Exetainer, and further gas transfer was not required. Even with a small volume of water, the GC analyses were able to be performed. Often, however, a gas sample is taken from the serum bottle or syringe headspace to transfer into a GC sample vial.

The actual GC analysis of a gas sample is discussed elsewhere in this publication. The output from the analysis will provide the concentration of N2O in the equilibration vessel's headspace (*C*g), typically with units of μl l−1.

The next step is to convert this value to one with units of μmol l–1. This is performed using the ideal gas equation (Eq. 2.15):

$$PV = nRT\tag{2.15}$$

where *P* is the pressure (atm.), *V* is the volume (*l*), *n* is moles, *R* is a constant (0.0821 l. atm. mol−<sup>1</sup> K−1), and *T* is the temperature (*K*). Rearranging equation (2.15) as follows gives moles: if the volume of gas is input as μl, then the output is μmol (Eq. 2.16):

$$n = \frac{PV}{RT} \tag{2.16}$$

What we wish to determine is how "saturated" in N2O the water sample is, with respect to river water in equilibrium with the atmosphere. So first we must calculate the N2O concentration in the water that would be in equilibrium with the atmosphere (*C*eq; μmol l−1): we must know the partial pressure of the N2O in the atmosphere at the sampling site, the water temperature at the sampling site, and the atmospheric pressure at the sampling site. Then, *C*eq is determined using (Eq. 2.17):

$$C\_{\rm eq} = \beta \times P\_{\rm gas} \times P\_{\rm barometric} \tag{2.17}$$

where*P*gas is the partial pressure of the gas (atm atm−1) and β is the Bunsen coefficient (mol l−<sup>1</sup> atm−1). Values for β are found in the publication of Weiss and Price (1980), who measured the solubility of N2O in pure water and seawater over a temperature range of 0–40 °C and a salinity range of 0–40‰. *P*barometric is the atmospheric pressure during field sampling (atm).

For example, assume a river temperature of 12 °C and a salinity of zero, then β equals 3.741 × 10−<sup>2</sup> mol l−<sup>1</sup> atm−1, atmospheric partial N2O pressure of 0.32 × 10–6 atm atm−1, and a barometric atmospheric pressure of 1 atm, and then *C*eq equals 0.012 μmol l−1.

The next step is to determine how saturated in N2O the water sample (*C*water) is with respect to the atmospheric equilibrium concentration (*C*eq) just established. To do this, we must know the N2O concentration in the headspace of the equilibration vessel (*C*g). This is provided from the GC results, with units of μl l−1. We also need to know the water temperature during equilibration so that the correct value of β can be used. The unit of β in Weiss and Price (1980) is moles l−<sup>1</sup> atm−1, so the value of β is first converted to units of moles l−<sup>1</sup> atm−<sup>1</sup> in order to be compatible with the units of *C*g. Finally, we need to know the atmospheric pressure inside the equilibration vessel. This would be set at 1 atmosphere if equilibrating the headspace with atmospheric pressure (e.g. the Tedlar bag method discussed above), but if the atmospheric pressure when measured differs, then the measured value should be used. Then the following calculation is used (Eq. 2.18):

$$C\_{\text{water}} = \beta \times C\_{\text{g}} \times P\_{\text{barometric}} \tag{2.18}$$

For example, if we have a laboratory equilibration temperature of 20 °C, and salinity remains at zero, then β equals 2.875 × 10−<sup>2</sup> mol l−<sup>1</sup> atm−<sup>1</sup> (Weiss and Price 1980) which, using the ideal gas law equals 0.644 l l−<sup>1</sup> atm−1. If the concentration of N2O in the equilibration headspace is 0.35 × 10−<sup>6</sup> μmol l−1, and the barometric atmospheric pressure is 1 atm, and then *C*water equals 0.2255 μl l−1, which, using the Ideal Gas Law, equates to 0.0101 μmol l−1.

Then following a mass balance equation (Hamilton and Ostrom 2007; Beaulieu et al. 2012), the total amount of N2O present in the equilibration vessel is calculated as follows (Eq. 2.19):

$$(C\_{\rm water}^{o})(V\_{\rm water}) = (C\_{\rm water})(V\_{\rm water}) + \left(C\_{\rm gas}\right)(V\_{\rm gas})\tag{2.19}$$

where *V* water and *V*gas are the volumes of water (*l*) and gas in the equilibration vessel, respectively, and *C*water and *C*gas are the respective concentrations (μmol l−1) in the water and gas phases following equilibration. Then, since all the N2O came from the original water sample, the concentration of N2O in this water sample *Co* water , with units of μmol l−1, is calculated as (Eq. 2.20)

$$\left(C\_{\text{water}}^{o}\right) = \frac{(C\_{\text{water}})(V\_{\text{water}}) + \left(C\_{\text{gas}}\right)(V\_{\text{gas}})}{(V\_{\text{water}})} \tag{2.20}$$

Thus, using the value for *C*water determined above, with equilibration volumes for water and gas of 0.12 l and 0.02 l, respectively, the value of *C<sup>o</sup>* water equals 6.93 × 10−<sup>2</sup> μmol l−1. The degree of saturation, or saturation ratio, in this example, is thus expressed as a ratio of *C<sup>o</sup>* water to *C*eq. For the example above, this equates to 5.70.

## *2.9.2 Determining N2O Fluxes from a Water Body*

Gas fluxes from a water body may be calculated using the following equation (Beaulieu et al. 2012), where *k* is the gas transfer velocity with units of distance per unit time, e.g. cm h−<sup>1</sup> (Eq. 2.21):

$$F = k \left( C\_{\text{water}}^{\circ} - C\_{\text{eq}} \right) \tag{2.21}$$

If the value of *C*eq has not been determined, but atmospheric N2O concentration and water temperature in situ are known, then the flux of N2O off the water surface may also be calculated as follows (Reay et al. 2003) (Eq. 2.22):

$$F = k \left( C\_{\text{water}}^{o} - \frac{C\_{\text{a}}}{K\_{\text{h}}^{\prime}} \right) \tag{2.22}$$

where *K* <sup>h</sup> is Henry's law constant (Weiss and Price 1980) for N2O, which is calculated as the equilibrium concentration in the gas phase (moles per unit volume air) divided by the equilibrium concentration in the water phase (moles per unit volume water), and *C*<sup>a</sup> is the concentration of N2O in air (moles per unit volume air). Theoretical background and commonly used quantities and units for *K* <sup>h</sup> can also be found in Sander (2015). To derive sensible flux units, the units in Eqs. (2.21) and (2.22) must be consistent with those used for the value of *k*. So, if *k* has units of cm h−1, the units of concentration must be in mol cm−<sup>3</sup> with the resulting flux given in mol cm<sup>2</sup> h−1. Similarly, if *k* and concentrations are in units of m d−<sup>1</sup> and mol m−3, respectively, then the flux of N2O has units of mol m2 d−1.

## *2.9.3 Determining Gas Transfer Velocity (K)*

The value of *k* varies due to the turbulent mixing of the water body at the water–atmosphere interface. Turbulence within the water body may result from shear stresses on the bed of streams or rivers, or tidal currents and is thus dependent on the depth of the water body and the speed at which it flows. It may also result from the wind. Thus, it can be expected that small, shallow, and sheltered water bodies, such as agricultural drains, will have values of *k* more strongly influenced by water depth and/or speed (current), while deeper large open expanses of water such as estuaries, large rivers, lakes, and oceans will be more strongly influenced by wind, although tidal currents in estuaries and winds that oppose the direction of the water body current can all influence turbulence.

Methods to derive values for *k* include (i) measuring the N2O flux using floating chambers (or fixed chamber) and back calculating *k*, (ii) using published relationships between *k* and wind speed to derive the value of *k*, (iii) tracer gas methods, and (iv) modelling of water body and micrometeorological parameters, or a combination of these approaches.

Measuring the N2O flux with chambers that float on the water body surface is commonly performed. Besides the requirements of chamber design that optimise sensitivity of the measurement, discussed elsewhere for land-based gas measurement (see Sect. 2.2), the chamber must obviously float, but with minimal intrusion into the water body when floating, have a vent to allow equilibration of the headspace pressure during the placement of the chamber on the water surface, be insulated to prevent temperature changes inside the chamber, and have a low profile to reduce the influence of wind moving the chamber across the water surface at directions opposed to the current. Such an effect creates additional turbulence and can lead to an overestimation of N2O fluxes (Clough et al. 2007b). Similarly, tethering the chamber so that it cannot drift in the current can also increase turbulence inside the chamber and lead to erroneous fluxes being derived (Hartman and Hammond 1984). If the water body is large, the chamber can be tethered to a boat, with a slackline, while both drift in the current (Beaulieu et al. 2012). Procedures for gas sampling, analysis, and calculation of fluxes are consistent with those for land-based chambers (Sect. 2.2): best practice involves taking several chamber headspace samples over time to enable the fitting of models to determine if the change in headspace N2O concentration fits a linear or non-linear trend, with fluxes calculated accordingly (Beaulieu et al. 2012; Hutchinson and Mosier 1981) (Plate 2.19).

For large water bodies, the value of *k* can be estimated using its relationship to wind speed (de Wilde and de Bie 2000; Barnes and Owens 1998) using equations that derive *k* from wind speed (Liss and Merlivat 1986; Wanninkhof 1992; Clark et al. 1994; Wanninkhof 2014). This methodology is best suited to deeper large open expanses of water, such as estuaries, lakes, and oceans. However, caution is required, as water currents can still affect turbulence in large open water bodies as demonstrated by Beaulieu et al. (2012): the value of *k* in the Ohio River was attributed to not just wind but both current and wind speeds, with 46% of *k* at low wind speeds (0.5 m s−1) resulting from water currents which were reduced to 11% at higher wind speed (>2.0 m s−1).

The principle of the **tracer gas method** is to release a soluble gas at a given point in the water body under study and to follow the dissipation of the injected gas over distance and time. Trace gas studies have been performed in lakes (Cole and Caraco 1998), oceans (Wanninkhof 1992), estuaries (Clark et al. 1994), and streams and drainage ditches (Harrison and Matson 2003; Premaratne et al. 2017). Trace gases should be absent from the study environment, and sulphur hexafluoride (SF6), methyl chloride (CH3Cl), propane (C3H8), krypton (Kr), and Freon-12 have been used. Studies by Jin et al. (2012) and Premaratne et al. (2017) provide examples of the technique using CH3H8, which is briefly described below. The tracer gas is first bubbled through a large container of water to saturate the water body with C3H8. Typically, this is a large carboy with the exact volume-dependent on the size of the water body under study and the ensuing release rate of the gas saturated water into the water body, assumed here to be a stream. Also dissolved in the water is an inert chemical tracer that can be used to estimate stream flow rate and dilution of the 2 Methodology for Measuring Greenhouse Gas Emissions … 85

**Plate 2.19** Floating chambers being used to determine the N2O flux from the LII river in Canterbury, New Zealand. Note the high number of replicates. Floating chambers are drifting freely in the current, but loosely tethered to a person walking alongside on the river bank who can retrieve the chambers for sampling (**a**), and fixed chamber being used to determine the GHGs flux from the river in China that can be raised or lowered with the fluctuation of water level (**b**)

added water. For example, this inert tracer might be bromide (Br−) or a chloride salt (Cl−). If the background conductivity of the water body is high, then a dye such as the rhodamine dye may be used. The water is injected at a known constant rate into the stream, using a peristaltic pump. Combining gas and chemical tracers in the one vessel prevents uneven mixing of the tracers in the stream following injection, which might occur if they are added from separate containers. However, for larger water bodies, this will be impractical, and the gas tracer will be delivered by gently releasing gas directly into the water body. For example, in the Iroquois River, Laursen and Seitzinger (2004) introduced tracer gas by staking a perforated garden hose to the stream bed perpendicular to the channel flow direction. Sampling stations are set up at intervals downstream of the tracer injection site, dependent on-stream speed and flow, with the first site several metres downstream to ensure adequate mixing of the injected water and to allow for any ebullition of tracer gas bubbles to be completed. Again, with larger rivers, these sampling stations will be at greater distances (e.g. kilometres) from the tracer injection site. At the sampling stations, water samples are taken to determine trace gas concentrations using the equilibration technique, and floating chambers may also be deployed to measure trace gas fluxes. In addition, water chemistry, micrometeorological parameters (e.g. wind speed, air temperature, and solar irradiance) and hydrological parameters are measured: depth, *h* (m); width, *w*(m); and velocity, *U* (m s−1) enable the volumetric flow (*Q*, m3 s−1) to be calculated.

If floating chambers are deployed, an estimate of *k* can be made using the equilibrium concentrations of the trace gas and the measured flux, thus, for C3H8 (Eq. 2.23):

$$k\_{\rm C\beta\mathcal{H}\_8} = F\_{\rm C\beta\mathcal{H}\_8} / \left(C\_{\rm water}^o - C\_{\rm eq}\right) \tag{2.23}$$

Using this CH3H8 transfer velocity, the N2O transfer velocity *k*N2O can be estimated as follows (Eq. 2.24):

$$k\_{\rm N\_2O} = \left(\mathrm{Sc\_{N\_2O}/Sc\_{C\_3H\_8}}\right)^{-n} \times k\_{\rm C\_3H\_8} \tag{2.24}$$

where *Sc* is the temperature-dependent Schmidt number for CH3H8 and N2O, and *n* is the Schmidt exponent (equal to 0.5 for surfaces with waves and 0.67 for surfaces without waves (Jähne et al. 1987). A Schmidt number represents the dimensionless ratio of the kinematic viscosity to the diffusion coefficient of the gas of interest, which is temperature-sensitive. Thus, it is a common practice to normalise the value of *k*N2O to a value of *Sc* = 600, which is the Schmidt number for carbon dioxide at 20 °C in freshwater (Wanninkhof 1992) (Eq. 2.25):

$$k\_{600-\text{N}\_2\text{O}} = \left(600/\text{Sc}\_{\text{N}\_2\text{O}(T)}\right)^{-n} \tag{2.25}$$

where *Sc*N2O(*<sup>T</sup>* ) is the Schmidt number for N2O at temperature *T*, and *k*<sup>600</sup>−N2O is the standardised gas transfer velocity for N2O at *Sc* = 600. The Schmidt number can be calculated from published equations for C3H8 and N2O according to Wanninkhof

#### 2 Methodology for Measuring Greenhouse Gas Emissions … 87

(1992), where

$$\text{Sc}\_{\text{CylH}} = 1911.1 - 118.11 \times T + 3.4527 \times T 2 - 0.04132 \times T 3 \tag{2.26}$$

$$\text{Sc}\_{\text{N}\_2\text{O}} = 2055.6 - 137.11 \times T + 4.3173 \times T2 - 0.05435 \times T3 \tag{2.27}$$

Raymond et al. (2012) revisit these Schmidt numbers following a meta-analysis of 563 experiments.

The dissolved tracer gas (C3H8) concentration data for each sampling station can also be used to estimate *k*. As the tracer gas moves downstream some of this gas is lost across the air–water interface and this can be defined in terms of a first-order transfer rate (Clark et al. 1994; Laursen and Seitzinger 2004) as follows (Eq. 2.28):

$$F = hK \left( C\_{\rm m} - C\_{\rm eq} \right) \tag{2.28}$$

where *Cm* is the mean concentration of the tracer gas in the water, *h* is the mean water depth, and *K* is the gas exchange coefficient (Clark et al. 1994). By combining Eqs. (2.21) and (2.28), gas transfer velocity may be defined as (Eq. 2.29)

$$k = h \frac{\left(C\_{\rm m} - C\_{\rm eq}\right)}{\left(C\_{\rm water}^{\circ} - C\_{\rm eq}\right)} K \tag{2.29}$$

The slope of the ratio of tracer gas (C3H8): chemical tracer (Br−) ratios versus time allows the gas exchange coefficient (*K*) to be determined (Wilcock 1988; Chapra and Wilcock 2000). Following corrections for background C3H8 or Br−, *K*C3H8 can be determined by plotting the ln(C3H8 concentration/Br concentration) versus water travel time (Jin et al. 2012). The assumption is made that the ratio of C3H8: Br during transit in the stream is altered solely as a result of C3H8 loss to the atmosphere. Then, the C3H8 exchange coefficient (*K*C3H8 ) is used to derive the N2O exchange coefficient (*K*N2O) as follows (Eq. 2.30):

$$K\_{\rm N\_2O} = \left( S c\_{\rm N\_2O} / S c\_{\rm C\_3H\_8} \right)^{-n} \times K\_{\rm C\_3H\_8} \tag{2.30}$$

The N2O exchange coefficient (*K*N2O) is then multiplied by *h* to determine the air–water N2O transfer velocity (*K*N2O) that can be used to calculate the N2O flux (Eq. 2.28), where it is assumed that the water column is well mixed. The mean value of *h* may be determined as follows (Eq. 2.31):

$$h = \frac{\mathcal{Q}}{\omega U} \tag{2.31}$$

## *2.9.4 Models for Determining N2O Fluxes from Water Bodies*

Numerous models exist in the literature for estimating *k,* and it is beyond the scope of this section to cover all of these. However, N2O emissions from agricultural drains and first-order streams are recognised hotspots for N2O emissions (Reay et al. 2003; Beaulieu et al. 2008; Outram and Hiscock 2012; Turner et al. 2015; Davis and David 2018), and these sites readily allow the collection of variables required for modelling. One model which has been used extensively is that of O'Connor and Dobbins (1958) who introduced the model to predict *k* based on water turbulence (Eq. 2.32):

$$k = \left(\frac{DU}{h}\right)^{0.5} \tag{2.32}$$

where *U* is the stream velocity (m s−1), *h* is the average river depth (m), and *D* (m2 s−1) is the diffusion coefficient of the respective gas in water at the water's given temperature. This model has been widely used. More recently, in a meta-analysis of 563 direct gas tracer release experiments, Raymond et al. (2012) found that the gas transfer velocity scaled with the product of stream slope and velocity, and they reported further models for estimating *k* and Schmidt numbers in streams. Other models (e.g. Schwarzenbach et al. 1993) have also been used to determine *k* by partitioning the individual contributions of water and wind-driven turbulence effects (Beaulieu et al. 2012; Hamma-Aziz et al. 2016).

## *2.9.5 Other Factors to Consider*

Besides measurements of dissolved gases in the water body, consideration should also be given to reporting other parameters or features at the site that will enable interpretation of the results, extrapolation of the results, modelling, and comparison with other studies. Water bodies such as drains and streams are unlikely to have constant chemistry throughout the day, and diel variation is commonly observed in temperature, dissolved oxygen (DO), and pH (Harrison et al. 2005; Clough et al. 2007b). Similarly, seasonal changes also occur. Such changes have implications for N cycling, and thus the amount of N2O dissolved in the water. A key driver of DO, besides respiration, is the rate of photosynthesis. This is affected by how clear the water is; so measurements of turbidity and sunshine hours should also be reported. Similarly, wind speed and direction are other critical parameters to gather. Dissolved organic C and inorganic-N species, both known to affect N2O production or consumption transformations, should also be reported.

## *2.9.6 Determining EF5*

Once the N2O flux has been established, the EF5 value can be established. First, the N2O flux is converted to an N2O-N flux. This will have units of mass of N per unit area per unit time. To establish the EF**<sup>5</sup>** value, the amount of NO3 <sup>−</sup> moving through a given area per unit time is then calculated, which then enables the EF**<sup>5</sup>** ratio to be determined. For example, if a 1-m-wide stream is 1 m deep and has a speed of 1 m s−1, then the stream flows through one square metre every second. If the NO3 <sup>−</sup>-N concentration in the water is 0.2 mg l−1, then 0.2 g of NO3 <sup>−</sup>-N flows through underneath the 1 m2 surface every second. If an N2O-N flux is 0.5 μg m−<sup>2</sup> s−1, then the ratio is 0.5 × 10−6: 0.2, or 2.5 × 10−6. The EF5 value is then determined based on the ratio of N2O-N: NO3 <sup>−</sup>-N or on the basis of assumed N2O emissions using the degree of saturation to predict a flux.

Units of N2O-N flux and NO3-N can be adjusted for the time period concerned. Obviously, stream chemistry and N2O fluxes are dynamic, and so it is expected that EF5 could also vary. Thus, modelling of these parameters at varying scales is useful for predicting EF5 over different periods. For example, Reay et al. (2003) successfully modelled the N2O flux along a drain with N2O emissions calculated on a "per metre of stream" basis, with predicted N2O losses from the preceding 1 m stretch being subtracted from the dissolved N2O concentration for each 1 m section of the drain.

## **2.10 Indirect GHG Emissions–Ammonia Emissions**

Ammonia (NH3) itself has no direct greenhouse effect. It is a gas with relatively low residence time in the atmosphere (2–10 days) compared to some GHG, such as CO2 (3–4 years), CH4 (9 years), and N2O (150 years) (Hobbs 2000). However, after NH3 enters the atmosphere and reacts with acids forming salts, it returns to the earth's surface and causes N2O emissions similar to a fertiliser-N application. When the soil is subjected to conditions near to optimal for urease activity (e.g. pH close to neutrality, soil moisture near the field capacity, temperature > 30 °C), the N losses through NH3 volatilisation from urea-based fertilisers applied on soil surface can be as high as 50% (Rochette et al. 2013; Martins et al. 2017). Therefore, the measurement of NH3 emission is important to estimate indirect N2O emissions derived from soil amendments, such as urea-based fertilisers, green manures, animal excreta, or ammonium-based fertilisers in different soils. A default emission factor defined by IPCC, known as EF4, can be applied for the estimation of indirect N2O emissions derived from volatilisation of NH3 and other nitrogen oxides (NOx) (de Klein et al. 2007). The mean value of EF4, considering the N volatilisation and consequent re-deposition, is 0.01 kg N2O-N per kg N volatilised as NH3 + NOx with an uncertainty ranging from 0.002 to 0.05 (de Klein et al. 2007). Management options that reduce NH3 volatilisation from soils are considered mitigating practices because they reduce indirect N2O emissions (IPCC 2014; Lam et al. 2017).

**Plate 2.20** Schematic representation of circular plots for measurements of NH3 volatilisation using mass balance integrated horizontal flux (IHF) method. A mast with NH3 samplers installed at different heights is placed in the centre the plot. A mast placed in an unfertilised area is used to quantify the background soil NH3-N emissions (zero N control)

There is a considerable range of non-isotopic methods that can be used to quantify the NH3 volatilisation from soils. Chamber-based methods are the most common. They are generally inexpensive compared to other more sophisticated methods, such as micrometeorological, and are easier to install and handle. On the other hand, the micrometeorological method has the advantage of being more representative for larger areas and therefore are more representative of the field scale. The classical micrometeorological method to measure NH3-N emissions is the mass balance integrated horizontal flux (IHF) method (Leuning et al. 1985). This method is based on the assumption that "vertical NH3-N flux from a fertilised area of limited upwind extent" is equal to "the integrated horizontal NH3-N flux at known downwind distance" (Wilson et al. 1983). This method uses NH3 samplers attached to masts placed in the centre of a circular field plot (Plate 2.20).

The NH3 samplers with mounting pivots and fins are installed in different heights of the mast above the soil surface (Leuning et al. 1985; Misselbrook et al. 2005; MSU 2019). Therefore, the samplers are pointed into the wind quantifying the horizontal NH3 fluxes at different levels (Plate 2.21), which are used in the final calculation of the vertical NH3-N flux (VF), in μgNm-2 s-1, as shown in the following equation (Eq. 2.33):

$$VF = \frac{1}{r} \left[ \int\_0^{\frac{\pi}{4}} \left( \frac{MN}{EAS \times t} \right)\_{\text{treated}} \times \text{d}z - \int\_0^{\frac{\pi}{4}} \left( \frac{MN}{EAS \times t} \right)\_{\text{background}} \times \text{d}z \right] \tag{2.33}$$

where *r* is the plot radius (fetch length), *MN* is the mass of N captured in the sampler, *EAS* is the effective cross-sectional area of the sampler, *t* is the sampling period, and *z* is the height of the sampler (MSU 2019).

The use of relatively sophisticated techniques, such as the mass balance IHF method described above, is often not possible due to limitations of financial resources or lack of skilled technicians. Thus, chamber-based methods are often employed instead, because of their viability and ease to work with. Part of the procedures

described above for the chambers used for measurement of N2O fluxes is also valid for the chambers used to quantify NH3 volatisation (Tables 2.6 and 2.7). A difference between these two types of chambers is the way how total N losses over the period of measurements are estimated. The usual continuous NH3 captured by acid traps during the entire period of measurement provides direct results of time-integrated N losses, i.e. there is no collection of temporally separated samples for a final calculation of fluxes as usually performed for N2O measurements.

Chamber-based methods have been used since the mid-twentieth century to test the effectiveness of N management practices aiming at a reduction of NH3-N losses (Volk 1959). An example of a chamber design used to collect volatilised NH3 is the semiopen collector described by Nômmik (1973). A concise schematic representation of this chamber with the foam discs with an acid solution used for NH3 trapping is shown in Plate 2.22.

Another common type of chamber is based on the use of a low vacuum system to collect NH3, which consists of tubes connecting the chamber to a vacuum pump and a small flask containing an acid solution used as NH3 trap (Kissel et al. 1977).

To warrant the precision and accuracy of measurements, i.e. reducing the uncertainty for extrapolation of NH3-N losses to larger areas, some procedures of installation and handling the chambers under field conditions should be warranted. For instance, the installation and replacement of the acid traps should be performed in a short period to avoid a time-dependent bias. Methods employing soil collars should be preinstalled at least 24 h before the beginning of measurements to avoid undesirable effects of soil physical disturbance on the NH3 volatilisation process. Moreover, the efficiency of NH3 captured by some type of chambers is usually not 100%, demanding the use of correction factors for the final calculation of NH3-N losses. These correction factors are usually predefined in previous studies in which calibration is performed using reference methods, such as the 15N-balance method (Araujo et al. 2009; Jantalia et al. 2012; Martins et al. 2021b).

The advantages and disadvantages of usual methods for NH3 volatilisation using non-isotopic techniques are shown in Table 2.6.


**2.6**SomemethodsusedtoquantifytheNH3volatilisationfromsoils

92 M. Zaman et al.


2 Methodology for Measuring Greenhouse Gas Emissions … 93

**Table 2.7**

Example of a spreadsheet

 for the calculation

 of NH3-N losses from the data of N

concentration

 in the solution extracted from the acid traps (foams)

## *2.10.1 A Simple Low-Cost Chamber to Quantify NH3 Volatilisation*

Among all methods for NH3 volatilisation measurements, a low-cost chamber has been used successfully under field conditions (Araújo et al. 2009; Nichols et al. 2018). This chamber costs less than one US dollar and can be manufactured in any region of the world, including those regions with limited financial resources for scientific research. Studies conducted under field conditions have been showing that this chamber can be used with sufficient accuracy and precision in experiments aiming at testing management practices for NH3-N loss abatement. Details of the design of this chamber were presented by Araújo et al. (2009), Jantalia et al. (2012), Nichols et al. (2018), and Martins et al. (2021a,b). Briefly, this chamber is made by cutting the bottom of a plastic soda bottle. After cutting, the bottom is attached to the top of the bottle finish, serving as protection against rainwater. A vertically hung foam strip (250 mm long, 25 mm wide, and 3 mm thick) pre-soaked in a 10 ml solution of sulphuric acid (1 mol/L) plus glycerol (2%, v/v) is placed inside the chamber to capture NH3. A small plastic pot is used inside the chamber to retain the remaining acid solution not absorbed by the foam. A basket made of wire and hung from the bottle finish is used as a support for the plastic pot and the foam inside the chamber. The steps to manufacture and anchor the chamber in the soil are shown in Plates 2.23 and 2.24.

Metallic supports can be used to anchor the chamber (Plate 2.24). The amounts of "outside NH3", i.e. NH3 not directly derived from soil, are usually not significant 2 Methodology for Measuring Greenhouse Gas Emissions … 95

**Plate 2.23** Steps to manufacture a low-cost chamber to measure ammonia volatilisation under field conditions (Araujo et al. 2009; Martins et al. 2021b)

**Plate 2.24** A simple design of an open chamber after the assemblage of the parts. A correction factor of 1.7 is used to convert the amount of NH3 trapped in the total volatilised NH3 (Araujo et al. 2009; Martins et al. 2021b)

in this type of chamber, even though the non-significance of the amount of "outside NH3" can be checked by installing some blank chambers in the area of NH3 measurements (e.g. one "blank" chamber per experimental block). Some eventual NH3 detected in these "blank" chambers can be subtracted from the total NH3 trapped in the chambers installed in treated and control plots. This "blank" chamber is the same described above, but the soil surface is covered using an impermeable barrier (e.g. an inert and impermeable plastic sheet).

Considering that the area circumscribed by chambers (<0.1 m2) is usually much smaller than the final area used for extrapolation of the NH3-N losses (e.g. hectare), the minimisation of the inherent uncertainty depends on the use of some strategies. For instance, one way of reducing uncertainty is the employment of multiple chambers per field plot. This strategy can increase the capacity of the detection of significant NH3-N loss abatement by adequate N management practices. In a field plot experiment, Martins et al. (2021a) estimated that, by using five chambers per plot (31 m2), the margin of error was less than 15% of the real NH3-N loss. To optimise the fieldwork and the laboratory analyses, the use of multiple chambers per plot can be employed only in the critical periods of emissions, i.e. focusing sampling in the period of a high probability of occurrence of significant NH3 fluxes and associated errors. For example, after easy hydrolysable N sources are deposited on the soil surface (e.g. fertilisers, excreta), the bulk of NH3-N losses usually occurs in the first 2 weeks after N deposition. Therefore, using multiple chambers in this period has a high probability of significantly reducing errors associated with the use of chambers. It is worth noting that the conventional use of multiple plots per treatment (replicates) in field experiments intrinsically improves the precision of NH3-N loss estimates, even when using a single chamber per plot.

In addition, a significant part of the errors in measurements of NH3 volatilisation can be reduced if it is ensured that the amounts of fertiliser applied in the relatively small area circumscribed by the chamber are equivalent to the N rate tested in the area of the plot, in kg ha-1. The use of microplots inside the main field plots reduces this kind of uncertainty. An example calculation to define the amount of fertiliser applied in a microplot with a chamber is presented below.

Example calculation:


The amount of urea-N required for the entire plot is (Eq. 2.34)

$$\frac{10,000\,\mathrm{m}^2}{25\,\mathrm{m}^2} = \frac{150,000\,\mathrm{g}\,\mathrm{N}}{\mathrm{Amount of N in the plot}}$$

$$\begin{aligned} \mathrm{Amount of N in the plot} &= 15\,\mathrm{g}\,\mathrm{N} \times 25\\ \mathrm{Amount of N in the plot} &= 375.0\,\mathrm{g}\,\mathrm{N} \\ \frac{46\,\mathrm{g}\,\mathrm{N}}{100\,\mathrm{g}\,\mathrm{urea}} &= \frac{375.0\,\mathrm{g}\,\mathrm{N}}{\mathrm{Amount of urea}\,\mathrm{in}\,\mathrm{the}\,\mathrm{plot}} \\ \mathrm{Amount of urea}\,\mathrm{in}\,\mathrm{the}\,\mathrm{plot} &= \frac{375.0 \times 100\,\mathrm{g}\,\mathrm{urea}}{46} \end{aligned}$$

$$\begin{aligned} \mathrm{Amount of urea}\,\mathrm{in}\,\mathrm{the}\,\mathrm{plot} &= \,\mathrm{815.2\,\mathrm{g}\,\mathrm{urea}} \end{aligned} \tag{2.34}$$

The amount of urea-N required for the microplot is

10,000 m2 <sup>0</sup>.9 m2 <sup>=</sup> <sup>150</sup>,000 g N Amount of Ninthe microplot Amount of Ninthe microplot = 15 g N × 0.9 Amount of Ninthe microplot = 13.50 g N 46 g N 100 g urea <sup>=</sup> <sup>13</sup>.50 g N Amount of ureainthe microplot Amount of ureainthe microplot <sup>=</sup> <sup>13</sup>.<sup>50</sup> <sup>×</sup> 100 g urea 46 Amount of ureainthe microplot = 29.35 g urea Amount of ureainthe remaining area ofthe plot = 815.20 − 29.35 Amount of ureainthe remaining area ofthe plot = 785.85 g urea

Another way to avoid a bias due to a slight unevenness in the fertiliser distribution in the plot is the application of a precise amount of fertiliser in the area circumscribed by the chamber. This procedure ensures a proper extrapolation of the NH3–N loss captured by the chamber to the total NH3–N loss per hectare. An example calculation of the amount of urea using the data of the calculations above, considering the "plastic bottle" chamber method, is presented as follows (Eq. 2.35):

• Diameter of the chamber: 100 mm

Area circumscribed bythe chamber <sup>=</sup> <sup>π</sup> <sup>×</sup> diameter ofthe chamber<sup>2</sup> 4 Area circumscribed bythe chamber = 3.14159 × 100 × 10−<sup>3</sup> 2 4 Area circumscribed bythe chamber <sup>=</sup> <sup>0</sup>.007854 m2 10, 000 *m*<sup>2</sup> <sup>0</sup>.007854 m2 <sup>=</sup> <sup>150</sup>, <sup>000</sup> *g N* Amount of Ninthe area ofthe chamber Amount of N appliedinthe area ofthe chamber = 15 g N × 0.007854 Amount of N appliedinthe area ofthe chamber = 0.11781 g N = 117.81 mg N 46 mg N 100 mg urea <sup>=</sup> <sup>117</sup>.81 mg N Amount of ureainthe area ofthe chamber Amount of ureainthe area ofthe chamber <sup>=</sup> <sup>117</sup>.<sup>81</sup> <sup>×</sup> 100 mg urea 46 Amount of ureainthe area ofthe chamber = 256.11 mg urea (2.35)

The foams capturing NH3 in the "plastic bottle" chamber should be periodically replaced in the course of the measurement period. The frequency of replacement depends on the purpose of the NH3-N loss quantification. When the measurements are performed to understand the temporal dynamics of NH3-N losses, more frequent replacements of the foams should be adopted (e.g. every 2 days or on a daily basis). By contrast, when the purpose is to assess only the total amount of NH3-N loss, the intervals of replacements of the foams can be increased, but not exceeding 5 days (Araújo et al. 2009). Considering the amount of sulphuric acid in the solution, each foam has the NH3 trapping capacity equivalent of more than 300 kg N ha-1. Considering also that the foams are replaced several times in the course of the measurement period, the saturation or reduction of trapping efficiency is very unlikely for the usual NH3 fluxes measured in soils.

When an easily hydrolysable N source is applied, such as urea-based fertilisers in agricultural soils or urine in pasture areas, the significant fluxes of NH3 are usually very intense in the first days and usually return to the background NH3 emissions in the first two weeks (Martins et al. 2021a). On the other hand, the application of more recalcitrant N sources causes prolonged NH3 fluxes above the background emissions, demanding measurements for longer periods. For instance, the measurements of NH3-N losses derived from plant residues with high C/N ratio usually should be performed for several weeks or months after the amendment. If promptly performed after collection in the field, the analysis of N in the foams can indicate the need for further NH3-N measurements.

**Plate 2.25** Orbital shaker with the Erlenmeyer flasks used for the extraction of NH4 <sup>+</sup> from the foams coming from the chambers

The plastic pots with the collected foam strips should be transported to the laboratory for the extraction of N trapped in the foams (NH4 <sup>+</sup> form). To extract the NH4 + from the foams, the content of the plastic pots (foam + solution) is transferred to an Erlenmeyer flask, and then deionised water (40 ml) is added for the extraction of the NH4 +. Alternatively, KCl solutions can also be used (Jantalia et al. 2012), but some tests indicate that pure distilled water is able to extract the total quantity of N trapped in the foams. After that, the Erlenmeyer flasks are shaken on an orbital shaker, usually for less than 30 min (Plate 2.25).

After the NH4 <sup>+</sup> is extracted from the foams, the N concentration of the extraction solution can be analysed using one of the many methods available (Bremner 1965; Keeney and Nelson 1982). Some examples of usual methods for the determination of NH4 +-N in solution are continuous flow analysis, steam distillation, and colorimetric techniques. Continuous flow system, such as flow injection analysis (FIA), is quite adequate because they are usually rapid and very sensitive, which are significant advantages when working with a high number of chambers and with many replacements of foams during the period of measurement of NH3 volatilisation. Standard solutions of ammonium salt (e.g. sulphate) should be analysed along with the extraction solutions. An example spreadsheet for illustrating the calculation for the conversion of the N concentration in the extraction solutions to the total NH3-N loss per hectare is detailed in Table 2.7. In the example shown below, considering a total NH3-N loss of 54.90 kg N ha−<sup>1</sup> in the N-fertilised treatment and subtracting 0.79 kg N ha−<sup>1</sup> (Zero N treatment, i.e. "*background*" emission), we estimated that 36% of the applied N was lost as volatilised NH3.

## **2.11 Gas Production Processes in Terrestrial Ecosystems**

The total production of a certain gas from the soil is often the sum of several soil fluxes that can occur in diverse microsites of the soil (e.g. aerobic and anaerobic microsites), for instance, soil microbial processes for N2O production, including nitrification, denitrification, co-denitrification, and organic N oxidation to mineral N (Zaman et al. 2012; Butterbach-Bahl et al. 2013; Müller et al. 2014; Smith 2017). The microbial processes of N2O and non-greenhouse gas N2 production in soils, sediments, and groundwater across the landscape may occur simultaneously and depend on the physical (moisture contents or O2 level and permeability) and chemical conditions [N form (i.e. organic N, NH4 +-N, NO2 <sup>−</sup>-N, and NO3 <sup>−</sup>-N), pH and organic C contents] (Zaman et al. 2012). Such complexity of gas production processes in soil makes the precise measurements of GHG flux, especially across different agroecosystems, very difficult and challenging. The identification of process and pathway-specific GHG fluxes can be tackled by suitable stable isotope-tracing techniques which are described in detail in Chap. 7.

## **References**


Klein CAM, Harvey MJ (eds) Nitrous oxide chamber methodology guidelines. Global Research Alliance on agricultural greenhouse gases. Ministry for Primary Industries, New Zealand, pp 56–72


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 3 Automated Laboratory and Field Techniques to Determine Greenhouse Gas Emissions**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** Methods and techniques are described for automated measurements of greenhouse gases (GHGs) in both the laboratory and the field. Robotic systems are currently available to measure the entire range of gases evolved from soils including dinitrogen (N2). These systems usually work on an exchange of the atmospheric N2 with helium (He) so that N2 fluxes can be determined. Laboratory systems are often

M. Zaman (B) · L. Heng

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl

Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang

Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

© The Author(s) 2021 M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_3

109

used in microbiology to determine kinetic response reactions via the dynamics of all gaseous N species such as nitric oxide (NO), nitrous oxide (N2O), and N2. Latest He incubation techniques also take plants into account, in order to study the effect of plant–soil interactions on GHGs and N2 production. The advantage of automated in-field techniques is that GHG emission rates can be determined at a high temporal resolution. This allows, for instance, to determine diurnal response reactions (e.g. with temperature) and GHG dynamics over longer time periods.

**Keywords** GHG · N2O · N2 · Helium

## **3.1 Automated Laboratory Techniques**

This section focuses on automated high-resolution determinations of anaerobic atmospheric trace gas (H2, NO, N2O, CO2, CH4) production kinetics, and nitrate (NO3 <sup>−</sup>) nitrite (NO2 <sup>−</sup>) analysis systems designed for ex situ incubations. Typically, such systems combine a temperature-controlled sampling unit holding several septum vials (with or without stirring or shaking) under a fully programmable robotic arm of an autosampler, which is used to repeatedly pierce the septa of the incubation vessels to subsample the headspace for concentration measurements by gas chromatography and/or other dedicated gas analysers. Processes are studied in different

T. Clough

K. Dawar

```
Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan
```
W. X. Ding

Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad

Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

C.-M. Görres Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

matrices, ranging from repacked or suspended soils (Horn et al. 2006; Palmer et al. 2012, 2015, 2016; Zhu et al. 2013; Schmidt et al. 2016), plant material (Keppler et al. 2006), and water samples (McCrackin and Elser 2010) to extracted bacterial cells (Dörsch et al. 2012; Brenzinger et al. 2015) and cultures of isolated (Lycus et al. 2017) or type strain organisms (Bergaust et al. 2012; Horn et al. 2005) including knockout mutants. Automated incubation setups are a refinement of classical incubation techniques still used to determine process rates of "respiration" (e.g. by soda-lime trapping of evolved CO2) or "denitrification enzyme activity" (DEA; N2O accumulation in the presence of acetylene). While automated batch incubations allow for estimating process potentials and apparent Michaelis–Menten parameters (*V*max, *Km*) under quasi-steady-state conditions in short-term assays with ample substrates, high-resolution incubations can also be used to study "gas kinetics" of selected processes with finite amounts of substrates (Palmer et al. 2012). Gaseous intermediates (e.g. in denitrification) evolve transiently before gaseous end products accumulate to amounts given by the added substrate. In this way, high-resolution gas kinetics can provide insights into the metabolic functioning under changing substrate conditions. Typically, incubations are initiated under oxic or anoxic conditions by flushing the headspace of the incubation bottles with various gas mixtures. The transition from oxic to anoxic metabolism can be studied by letting the sample consume a finite amount of O2 while monitoring the evolution of gaseous intermediates (NO, N2O, H2), and final products (N2, CH4, H2S) involved in anoxic metabolism (Schlüter et al. 2019). Notably, replacing the bottle atmosphere with He or a mixture of He and O2 allows for quantifying N2, the final product of denitrification which cannot be detected at ambient N2 background levels (but see section on Raman spectroscopy

M. E. G. Hofmann

Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh

D. Lewicka-Szczebak

Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

#### J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

#### P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

Chap. 2, Sect. 2.10). Using appropriate inhibitors (Yoshinari and Knowles 1976; Bedard and Knowles 1989; Taylor et al. 2015), selected processes can be studied.

High-resolution gas kinetics are also the method of choice to study the induction of enzymatic processes involved in GHG turnover, which can be studied by sequential analysis of gene expression levels or other genetic markers (Vollack and Zumft 2001; Bergaust et al. 2010; Liu et al. 2014; Brenzinger et al. 2015). In this way, the ecophysiology of specific microbial groups or organisms can be studied in great detail and advance our understanding of regulatory aspects involved in GHG turnover (Bakken et al. 2012; Braker and Conrad 2011; Hink et al. 2017).

## *3.1.1 Technical Challenges*


M. Šimek

M. dos Reis Martins · S. Urquiaga

R. Well

Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany

#### S. Zaman

University of Canterbury, Christchurch, New Zealand

#### J. Zhang

School of Geography, Nanjing Normal University, Nanjing, China

K. Lenhart

Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil


typically eluate close to each other, and integration routines have to be designed to account for these dynamics. Likewise, it may be difficult to integrate small O2 peaks next to large N2 peaks.


In the following sections, two automatic incubation systems will be described.

## *3.1.2 System 1*

The system for determining anaerobic atmospheric trace gases, namely, H2, NO, N2O, CH4, and CO2 comprises a gas chromatograph (7890B GC, Agilent Technologies Inc., Santa Clara, CA, USA) equipped with a split/splitless injector as well as a micro-packed column (Shin Carbon ST 80/100, 2 m, 0,5 mm ID, Part#: 19043, Restek Corporation, Bellefonte, PA, USA) and a capillary column (RT-Molecular Sieve 5A, 30 m, 0,53 mm ID, 50 μm, Cat#: 19723, Restek Corporation, Bellefonte, PA, USA; Plate 3.1). A pulsed discharge helium ionization detector (PDHID, V1D-3-I-HP-220, Valco Instruments Company Inc. VICI AG International, Houston, TX, USA) is coupled to the outlet of the ShinCarbon column via a restriction capillary, and a thermal conductivity detector (TCD, G3440B, Agilent Technologies Inc., Santa Clara, CA, USA) is coupled to the outlet of the molecular sieve 5A column. The ShinCarbon column acts as a water-insensitive molecular sieve where CO2 and N2O elute (which is essentially impossible for "regular" molecular sieve columns that trap CO2). Baseline separation of O2 and N2 in the % range is achieved on the molecular sieve 5A after "heart cutting" of the combined peak eluting from the ShinCarbon column.

The TCD is used for analytes in the % range. The PDHID is a universal detector which offers high sensitivity in the higher ppb to lower % range without the need to fulfil the increasing requirements for operating a radionuclide-dependent electron

**Plate 3.1** Setup of the gas chromatographic system. Switch represents a "Microfluidics Deans Switch" enabling "heart cutting" of peaks eluting from the ShinCarbon column

capture detector (ECD) (Forsyth 2004) or to purchase a rather cost-intensive mass sensitive detector (MSD). The gas chromatograph is operated with helium 6.0 as carrier gas that is purified by two sequential helium purifiers (V1HPM-220, Valco Instruments Company Inc. VICI AG International, Houston, TX, USA). One of the outlets of either the PDHID or the TCD can be connected to an NO analyser equipped with a chemiluminescence detector (CLD; Sievers NOA 280i, Zysense LLC, Weddington, NC, USA) to increase sensitivity for NO. The digital signal of the nitric oxide analyser is sent to an analogue-to-digital converter (A/D converter ADC 39500E, Agilent Technologies). The signals of all three detectors are recorded by the software Chemstation (Chemstation Rev. C.01.07.SR2 [255], Agilent Technologies Inc., Santa Clara, CA, USA) on a Windows computer, which is likewise used to control the gas chromatograph.

A robotized incubation system (Combi PAL-xt, CTC Analytics, Zwingen, Switzerland) is connected to the gas chromatograph to complete the setup (Plate 3.2). The Combi PAL-xt is equipped with two trays for 32 vials (10 or 20 ml) each. One of the trays can be temperature controlled in the range of 4 °C–70 °C. Furthermore, it comprises a temperature-controlled agitator (temperature range: 30 °C–200 °C; agitation speed: 250–750 rpm) shaking up to six samples at one time. A Y-cable (custom-made; Autosampler Guys, Alexandria, VA, USA) is used to enable the gas chromatograph to control both the incubation system and the A/D converter of the nitric oxide analyser. This has the advantage that all subordinate devices will only start a run when the gas chromatograph is "ready". Chronos xt 4.9 Master-Software for PAL samplers (Axel-Semrau, Sprockhövel, Germany) controls the robotized incubation system.

Altogether, this system allows for a cost-efficient online quantification of gas production and consumption during incubation experiments as well as the analysis of stored gas samples. A simultaneous quantification of H2, O2, N2, NO, CO2, CH4, and N2O is achieved. Although not in focus here, C2H2, C2H4, and C2H6 are likewise easily detectable.

**Plate 3.2** Photograph of the robotized incubation system (**a**) and configuration of the robotized incubation system, including the gas chromatograph (GC) and the nitric oxide analyser (NOA) with a chemiluminescence detector (CLD) connected via an A/D converter (ADC). "Switch" indicates an ethernet switch (**b**)

#### **3.1.2.1 Measurements of Gas Samples**

Analysis parameters are adjusted according to the gases to be measured as well as their expected concentrations. The most important consideration is whether the separation of a given analyte will be sufficient on the ShinCarbon column or whether "heart cutting" and further separation on the molecular sieve column is required. Specifically, baseline separation of O2 and N2 in the % range requires passage through the molecular sieve column. However, CO2, N2O, and NO are strongly retained or trapped, preventing detection. The choice of detector is generally dependent on the concentration of the gases with the PDHID being significantly more sensitive than the TCD. However, the TCD has a broader range of linear response than the PDHID. Multiple "heart cuts" (i.e. switches) between the molecular sieve—TCD path and the restrictor—PDHID path are possible during one run, enabling the quantification of trace concentrations of certain analytes in the presence of huge concentrations of others.

The following protocol provides good separation of the gases H2, NO, CO2, CH4, N2O, O2, and N2 (Table 3.1 and Fig. 3.1). The injection volume of gas samples is 250μl. The system operates in constant pressure mode with a pressure of 379.21 kPa. The inlet is set to 160 °C, the total flow is 40.782 ml min−1, and the split flow is 12.977 ml min−1. The septum purge flow is 3 ml min−1. Carrier gas is directed to the PDHID. A "heart cut" is performed via the Deans switch at 0.6 min to direct the flow to the TCD via the molecular sieve column, and it switches back at 0.76 min to redirect the flow to the PDHID. The outlet of the PDHID is connected to the CLD inside the nitric oxide analyser. The temperature protocol starts at 90 °C. This temperature is kept for 2 min before it is ramped at 50 °C min−<sup>1</sup> to 120 °C and held for 2.1 min. Finally, the oven is further heated to 200 °C (50 °C min−1), and the column is baked out for 2 min at this temperature. After 10 min, the gas chromatograph is

**Table 3.1** Retention times, limits of detection (LOD), limits of quantification (LOQ), coefficient of variation (CV at 500 ppm apart from N2), and linear range for different gases. Injection volume was 250 μl


an.d., not determined

bDetermined at 2% v/v N2

**Fig. 3.1** Chromatograms of the three detectors connected to the GC system: CLD (blue), TCD (green), and PDHID (red)

ready to start a new run. H2, CH4, CO2, and N2O are detected on the PDHID, O2 and N2 on the TCD, and NO on the CLD (Fig. 3.1). The dynamic range for most of the tested gases includes four orders of magnitude with a single detector (Table 3.1).

The minimum change in concentration which can be regarded significant is 6.3 ppm for H2, 5.0 ppm for CH4, 2.9 ppm for CO2, and 8.7 ppm for N2O at an initial concentration of 500 ppm. For N2, the minimum significant change in concentration is 0.2 vol.% (2000 ppm) at an initial concentration of 2%.

**Fig. 3.2** Contamination of He-filled sample vials (500 mbar overpressure) with N2 from air after multiple punctures

#### **3.1.2.2 Possible Sample Containers**

The system can be adapted to a variety of vial types. The sample seats in the provided autosampler trays accommodate 20 ml glass vials (75.5 mm × 22.5 mm with rounded bottom) or 10 ml vials with specific inserts (LEAP PAL Parts and Consumables LLC, Raleigh, NC, USA). Magnetic caps can be used to enable the autosampler to move the vials from the trays to the agitator if required. Acrylic glass blocks with milled sample seats (custom-made) can accommodate up to 100 exetainers (3 ml or 12 ml; Labco, Lampeter, UK) and other vial types offering the possibility for simple measurement of stored gas samples. Butyl septa (T.H. Geyer GmbH and Co. KG, Renningen, Germany) are sufficiently gastight after multiple punctures during incubation experiments (Fig. 3.2). However, the tightness of the septa should be carefully checked to ensure the reliability of the data obtained.

#### **3.1.2.3 Semi-automated Determination of Nitrate and Nitrite**

An analysis of the dissolved N-oxides NO3 <sup>−</sup> and NO2 <sup>−</sup> is required for the calculation of N and electron recoveries along with the determination of N-gases. We use a semiautomated, highly sensitive system based on reduction of NO3 <sup>−</sup> and/or NO2 <sup>−</sup> and analysis of the resulting nitric oxide. The lower detection limit for NO3 <sup>−</sup> or NO2 − in liquid samples is 100 nM.

The heart of the system is a nitric oxide analyser (Sievers 280i, Zysense LLC, Weddington, NC, USA) with a chemical reactor reducing N-oxides to NO ("purge vessel"). The purge vessel contains a reducing agent and is heated by a circulating water bath (Grant Instruments, Cambridge, UK). Up to 96 samples can be placed in a Combi PAL-xt autosampler (CTC Analytics, Zwingen, Switzerland) equipped with a large wash module and standard tray holders for 2 ml HPLC vials. A small volume of liquid sample (10–100 μl) is injected into the purge vessel. The evolving nitric oxide is purged out of the vessel by a stream of nitrogen and reaches the CLD inside the NOA. 50 mg of potassium iodide dissolved in 0.5 ml of double-distilled water and added to 4.5 ml of glacial acetic acid are used as reducing agent for NO2 − in the purge vessel. Nitrate and NO2 <sup>−</sup> are both reduced to NO when using 40 mg of vanadium chloride in 5 ml hydrochloric acid as reducing agent. The purge vessel is heated to 45 and 90 °C by a circulating water bath for NO2 <sup>−</sup> only and combined NO3 <sup>−</sup>/NO2 <sup>−</sup> analyses, respectively. Water vapour is condensed before escaping the vessel by a second circulating water bath, set at 4 °C to cool the upper part of the vessel. In addition, a propylene filter is installed in the tubing which connects the purge vessel outlet to the nitric oxide analyser to prevent any liquids from reaching the CLD inside the NOA. A sodium hydroxide trap filled with 1 M NaOH is installed in line with the purge vessel in case of the combined NO3 <sup>−</sup>/NO2 <sup>−</sup> analysis to neutralize any HCl before it reaches the CLD (Fig. 3.2). The data is transferred to a Windows computer which also controls the autosampler via the software Cycle Composer Pal1 (CTC Analytics, Zwingen, Switzerland).

Before the analysis a calibration is performed using NO2 <sup>−</sup> or NO3 <sup>−</sup> standards in the range of 100 nM to 100 mM. The standards as well as the samples are placed in 2 ml HPLC glass vials (neoLab Migge GmbH, Heidelberg, Germany), fitting into the sample tray and are sealed with septa and caps. The autosampler can perform up to 50 injections before the reducing agent needs to be replaced manually. Each measurement takes 60 s, and the needle is flushed with double-distilled water twice in the large wash modules of the Combi PAL autosampler before taking the sample and after injecting.

The concentrations of NO2 <sup>−</sup> can be calculated directly from the peak areas using the calibration curve. In the case of NO3 <sup>−</sup>, the actual concentration is obtained by subtracting the NO2 <sup>−</sup> concentration from the combined NO3 <sup>−</sup>/NO2 <sup>−</sup> concentrations obtained (Plate 3.3).

#### **3.1.2.4 Application of the Automated Incubation System to Determine Denitrification Kinetics**

The following protocol can be applied for the estimation of denitrification potentials by use of soil slurry incubations and simultaneous analysis of denitrification associated trace gases. The determination of N2O production and N2O consumption rates are used to derive NO3 <sup>−</sup> dependent apparent Michaelis Menten parameters for soil denitrification.

**Plate 3.3** Schematic diagram of the semi-automated NO3 <sup>−</sup> and NO2 − analysis system

#### *General Experimental Setup*

Soil slurries are set up and pre-incubated under anoxic conditions to reduce the background of soil endogenous N-oxides. Pre-incubated soil slurries are supplemented with substrate and spiked with different concentrations of NO3 <sup>−</sup>. Acetylene (10% v/v) is added to the gas phase of one of two subsets of samples to inhibit N2O reduction (acetylene inhibition technique; Yoshinari and Knowles 1976). Subsequently, the concentration of N2O in the headspace is monitored via gas chromatographic analyses while shaking the samples on a regular basis. The amount of N2 which is produced in the non-inhibited samples can be estimated from the difference of the N2O concentrations of the two subsets for low concentrations of supplemental NO3 <sup>−</sup> or measured directly when high concentrations of NO3 <sup>−</sup> in the mM range are supplemented.

#### *Materials*


#### *Experimental Procedure*

For the pre-incubation, the soil is diluted 1:5 with distilled water in a Duran glass bottle. The bottle is closed with a butyl stopper, and the slurry is made anoxic by applying alternating cycles of vacuum and helium for five times.

A minor overpressure of helium (200 mbar) is applied to avoid mass flow of air into the bottle. The soil slurry is pre-incubated on a roller mixer until endogenous N-oxides are depleted. Incubation times are soil-dependent. 1 ml samples of the soil slurries are taken every 24 h using a helium-flushed syringe and needle to monitor combined NO3 <sup>−</sup>/NO2 <sup>−</sup> concentrations with the semi-automated NO3 <sup>−</sup>/NO2 <sup>−</sup> analysis system. The sample is centrifuged for 5 min at 16,000 × g and 4 °C before the supernatant is filtered through a 0.2 μm pore size membrane to remove microorganisms as well as any remaining particles. This filtrate is then used for NO3 <sup>−</sup>/NO2 <sup>−</sup> analysis (Plate 3.4). Alternatively, colorimetric methods can be used. The pre-incubation is stopped when NO3 <sup>−</sup> and NO2 <sup>−</sup> are depleted.

The pre-incubated soil slurry is aliquoted into 20 ml glass vials. Each vial receives 15 ml of the soil slurry via small funnels while shaking continuously by hand to avoid sedimentation. Acetate is added as C-source for denitrification as it is one of the most important fermentation products found in soils potentially fuelling denitrification. The vials are crimp-sealed with butyl septa and magnetic caps. To restore anoxic

**Plate 3.4** Flushing of incubation vials containing soil slurry

conditions in the samples, the septa are pierced by a thin needle and the headspace is evacuated and helium-flushed five times leaving an overpressure of 620 or 800 mbar in the subsets that will or will not receive acetylene, respectively. Those vials with 620 mbar overpressure receive 0.9 ml acetylene via a helium flushed gastight syringe with a thin needle, resulting in 800 mbar overpressure and 10% (v/v) acetylene in the headspace.

Finally, all vials receive the respective concentration of NO3 <sup>−</sup> ranging from 20– 1000μM from anoxic stock solutions via a 100μl micro-volume liquid syringe. After placing the vials in the temperature-controlled tray holders of the incubation system, the gas chromatographic analysis starts. Samples are transferred from the trays to the temperature-controlled agitator and vigorously shaken for 2 min before each analysis to allow for equilibration between slurry and headspace. The autosampler syringe is flushed with helium for 60 s before piercing the septum. It draws a 250 μl sample of the headspace and directly injects it into the gas chromatograph. Each vial is sampled 4 to 8 times during the incubation period depending on the initial NO3 − concentration and the denitrification potential.

In addition to the gas analysis, NO3 <sup>−</sup>, NO2 <sup>−</sup> and ammonium can be analysed from t0 as well as from tend samples to check for N-recovery. For NO3 <sup>−</sup> and NO2 − analysis, the samples are centrifuged as stated above and analysed with the semiautomated NO3 <sup>−</sup> and NO2 <sup>−</sup> analyser (Plate 3.4). Ammonium binds to the cation exchange sites of soil particles. Thus, extraction with 2 M potassium chloride is performed to extract ammonium prior to quantification. 1 ml samples are mixed with 1 ml of 2 M potassium chloride in an Eppendorf tube before being centrifuged for 15 min at 16740 × g and 4 °C. The supernatant is then used for a colorimetric assay (Berthelot reaction). In brief, 100 μl of supernatant is mixed with 50 μl of 2% sodium phenoxide, 25 μl of 0.005% sodium nitroprusside, and 25 μl of sodium hypochlorite. The absorption of the formed indophenol is measured at 630 nm in the spectrophotometer after 30 min of incubation at 30 °C in the dark.

#### *Data Evaluation*

The gas chromatographic data is evaluated by firstly transforming peak areas into ppm values with the help of calibration curves. These ppm values are then corrected for the pressure in the vial headspace, which decreases with the number of samplings. It is important to account for the soluble fraction of CO2 and N2O in water. The temperature-dependent Bunsen coefficients of these gases can be used to estimate the physically dissolved fraction. For nitrous oxide, the Bunsen coefficient is 0.85 at 10 °C, 0.71 at 15 °C, 0.61 at 20 °C, and 0.52 at 25 °C. For carbon dioxide, it is 0.99 at 10 °C, 0.85 at 15 °C, 0.74 at 20 °C, and 0.65 at 25 °C (Blachnik 1998). Physically dissolved CO2 reacts in a pH-dependent manner with water to H2CO3 that further reacts to bicarbonate and carbonate. This pH-dependent chemically dissolved fraction also needs to be considered.

#### 3 Automated Laboratory and Field Techniques … 123

Calculation of decreasing headspace pressure after sampling:

$$m\_0 = \frac{p\_0 \times V\_h}{R \times T} \tag{3.1}$$

$$p\_s = \frac{p\_0 \times V\_h}{V\_h + V\_s} \tag{3.2}$$

$$m\_s = \frac{p\_s \times V\_s}{R \times T} \tag{3.3}$$

$$p\_1 = \frac{(\mathbf{n\_0} - \mathbf{n\_s}) \times \mathbf{R} \times T}{V\_h} \tag{3.4}$$

in the likely case that *p*<sup>1</sup> = *p*s, this equation simplifies to:

$$p\_1 = \frac{p\_0 \times V\_h}{V\_s + V\_h}$$


Calculation of amount of substance from ppm values:

$$
\mathfrak{n}\_{\text{total}} = \mathfrak{n}\_{\text{gas}} + \mathfrak{n}\_{\text{diss}} + \mathfrak{n}\_{\text{chem}} \text{ (last term only if needed)} \tag{3.5}
$$

$$\mathfrak{m}\_{\rm gas} = \mathfrak{y}\_i \times \frac{\mathfrak{p}\_h \times V\_{\rm gas}}{\mathbf{R} \times T} \tag{3.6}$$

$$
\mathfrak{m}\_{\rm diss} = \mathfrak{a} \times \frac{\mathfrak{n}\_{\rm gas}}{\mathcal{V}\_{\rm gas}} \times \mathcal{V}\_{\rm liq} \tag{3.7}
$$

$$
\mathfrak{n}\_{\text{chem}} = \mathfrak{n}\_{\text{diss}} \times \mathbf{10}^{pH - pKa} \tag{3.8}
$$


## *3.1.3 System 2*

The described robotized gas analysis system is an improved version of the system described by Molstad et al. (2007). The current system has been published by Molstad et al. (2016). The system has replaced the microGC with a traditional GC (Agilent 7890A), equipped with various valves, columns, and detectors (TCD, FID, and ECD) to analyse all relevant gases (except NO) by a single injection. The setup has two PLOT columns: a Molesieve column for separation of N2 and O2, and a PLOTQ column to separate CH4, N2O, and CO2. The system has three loop injectors–one for each PLOT column and one for injecting samples into the chemiluminescence NO detector. Short, packed Hayesep columns are used to retain and backflush water coming with the sample.

The use of ECD ensures sensitive measurements of N2O at low concentrations, and the FID detects CH4 in the sub-ppm range, thus allowing quantification of these gases at below ambient levels in the atmosphere (which are 0.35 ppmv for N2O, 1.84 ppmv for CH4). At high concentrations, N2O can be quantified via the TCD signal, which is linear, ensuring accurate determination of N2O over a wide concentration range.

The Gilson 222 (XL) autosampler of the old system has been replaced with the larger CTC GC-PAL (rebranded by Agilent), including a custom-made holder for a hypodermic needle. Up to 30 stirred and 16 non-stirred 120 ml serum flasks, incubated in a water bath, can now be monitored in one experiment (Plate 3.5).

To transfer a sample from the headspace to the three sampling loops (Plate 3.6), a peristaltic pump (Gilson Minipuls 3) is used. We now use Iso-Versinic Fluoroelastomer (IF) tubing (17 cm, ID = 2 mm; product nr 770130, Saint-Gobain or product nr F117744, John Morris) instead of Marprene tubing. This is because the diffusion of

**Plate 3.5** Water bath with stirrer plates (blue) and rack (green) that can hold up to 30 stirred and 16 non-stirred bottles. The lid fits onto the tops of the bottles. Other types of bottles may be supported by making other versions of the two plates that serve as rack and lid. The autosampler (not shown) is attached behind the bath. The bath, rack, and lid can be produced from CAD drawings with a CNC router, using seven PVC plates of 2 cm thickness. The CAD drawings are available upon request

**Plate 3.6** Architecture of the GC system. The GC exists in slightly different versions, equipped with one or two ovens (one for the PLOTQ column and one for the Molesieve column) which are held at constant temperature

N2 and O2 is slower through IF than Marprene, and because this replacement reduced the carryover of N2O and CH4 (see leaks, carryover, and dilution tests below).

In all other aspects, the new system is equal to the original (Molstad et al. 2007) regarding the repeated sampling by peristaltic pumping, replacement of the sampled gas by reversed pumping (injecting He), and transfer of gas to a chemiluminescence NO analyser. In short, samples are taken from 10–120 ml serum flasks using an autosampler connected to a peristaltic pump. The autosampler operates a needle which pierces the septa on the flasks (never twice at the same spot), and the pump takes approximately 1 ml headspace gas from the flasks into the sample loops of the GC and NO analyser. After injection (the turning of the valves leading to the gas in the loops into the GC columns and the NO analyser), the pump is reversed, pumping helium and the portion of the up-pumped gas in the pipelines back into the flasks, thus minimizing dilution of the gas in the flasks while maintaining atmospheric pressure.

The system monitors O2, N2, N2O, NO, CO2, and CH4 (and various other gases if needed) by repeated sampling of the headspace, while sustaining the gas pressure near 1 atmosphere in the flasks due to replacement of sampled gas with He.

#### **3.1.3.1 The GC**

The flow scheme of the GC system is shown in Plate 3.6. Loop injection to the PLOT columns (Molesieve and PLOTQ) is done by Valve 3 and 4 via Hayesep backflush columns, which prevent water from entering the PLOT columns. For the Molesieve column, also CO2 is backflushed. The third loop injection valve (Valve 2) sends samples to the NO analyser. Like in Sitaula et al. (1992), Valve 1 directs the outlet from the TCD to the FID most of the time to prevent oxidation of the detector, except for a brief period when N2O is eluted from the PLOTQ column.

The flowchart for the valve positions during a run is shown in Plate 3.6, and a chromatogram of air is illustrated in Fig. 3.3. Figure 3.4 shows the valve positions, column back pressures, and substance elution times during a run. Further information about the GC is found in Table 3.2.

Slightly different versions of the system have been constructed with our help elsewhere. Currently, there are five systems operating in China and three in Norway. These systems are identical in principle, the main difference being that some of the GC's have the PLOTQ column in a separate oven. Our experience is now that this is not necessary for the compounds we are analysing. In one version, an H2 analyser is attached in parallel to the FID. A split in front of the FID directs a minor amount of the column flow into a Peak Performer 1 RCP H2 analyser.

#### **3.1.3.2 The NO Analyser**

For the analysis of nitric oxide, we use either a modified Teledyne 200 NO/NOx analyser or a Sievers NO 280i. The modification of the Teledyne analyser is held a trade secret by the Norwegian distributor of Teledyne (Industriell Måleteknikk, Oslo,

**Fig. 3.3** Chromatogram of an air sample (5.5 min). The peaks before the 3rd minute come from the PLOTQ column, the others come from the Molesieve column. CO2, N2, and O2 (at 2.1, 3.5, and 4 min) are detected on the TCD. Methane is detected, at 1.5 min from the PLOTQ column and at 5 min from the Molesieve column, on the FID, and also, for high concentrations, on the TCD. For a 1.2 min period, the gas from the TCD is switched from FID to ECD, allowing the quantification of low N2O concentrations. At high concentrations, N2O is also detected by the TCD, thus extending the linear range much beyond that of the ECD


**Fig. 3.4** Valve positions, column pressures, and eluted peaks during a single run (5 min)


**Table 3.2** Characteristics of the GC

Norway) but involves an improvement of the response time. Without this modification, the Teledyne analyser has proven to be unsuitable for the incubation system, which injects a short pulse of 0.5 ml gas into the analyser. (An unmodified analyser gives a one-minute response to a 0.5 ml injection; depending on the NO concentration, this response can be either a broad flat-topped peak or a sharp peak followed by one minute of noise.) Based on this, it is recommended to either buy the Teledyne instrument from the Norwegian distributor or acquire a different NO analyser, like the (much more sensitive) Sievers NOA 280i, which is also supported by the software, when building the incubation system.

#### **3.1.3.3 Leaks, Carryover, and Dilution of Headspace by Sampling**

The new system has another type of injection (loop injection versus the solenoidvalve system of the microGC), and this leads to less dilution of the headspace per sampling and lower leaks. We have also replaced the Marprene tubing with Iso-Versinic Fluorelastomer tubing, which is more gastight, though less durable.

By sampling repeatedly from 120 ml flasks containing air (no liquid), we have found that the dilution of the headspace per sampling is 0.71%, which corresponds to replacing 0.85 ml of gas with helium. This test was conducted with active sampling to the NO analyser; without this (i.e. without turning Valve 2), the dilution per sampling was 0.36% or 0.43 ml. These volumes are lower than the sum of the volumes of the sample loops because they are heated to 100 °C. The theoretical loss for a 1 ml sample loop should be approximately 1.0 ml\*293 K/373 K = 0.785 ml if the temperature in the bottle is 293 K. The dilution is slightly higher than this due to the mixing of sample gas with helium in the pump tubing and pipes.

We have also tested the leak of O2 and N2 into the flasks per sampling by repeated sampling from flasks with He (1 atm) at 20 °C. This was done with two types of tubing, Marprene (ID = 1.6 mm), and Iso-Versinic Fluorelastomer (ID = 2 mm).

**Table 3.3** Carryover for N2O and CH4 in five consecutive helium samples after a standard containing 151 ppmv N2O and 1% CH4. Standard errors are, for N2O: < 2ppbv; for CH4: < 0.015ppmv (Two replicates for Marprene, three for Iso-Versinic)


The results for Marprene are

N2: 10 ppmv/sampling = 50 nmol N2/sampling O2: 5 ppmv/sampling = 25 nmol O2/sampling

These leak estimates were obtained by running the pump at a speed resulting in a flow of 9 ml min−1, sample pumping for 20 s = 3 ml total volume taken into the loops. If using slower pumping speed and longer times (3 ml min−1, 60 s pumping), the leaks were approximately doubled.

The results for Iso-Versinic Fluorelastomer, pumping 12 ml min−<sup>1</sup> (due to larger ID) for 15 s:

N2: 2.3 ppmv/sampling = 11 nmol N2/sampling

O2: 1.2 ppmv/sampling = 6 nmol O2/sampling

The leaks can be further reduced by enclosing the pump in a helium chamber. A simple plastic cover over the pump head, flushed with He (10 ml min−1), reduced the leaks of N2 and O2 to 1.6 and 0.8 ppmv/sampling.

The carryover of N2O and CH4 from one sample to the next was measured by sampling from a standard containing 151 ppmv N2O and 1 vol.% CH4 before sampling in sequence five bottles containing only helium. The results (Table 3.3 and Fig. 3.5) show that, compared to Marprene, the Iso-Versinic tubing has less than half the carryover for N2O, and less than one-fifth for CH4. With Iso-Versinic, the carryover to the first helium sample yielded 0.055 ppmv N2O (0.037% of the previous sample) and 0.15 ppmv CH4 (0.0015% of the previous sample). The carryover of O2 and N2 was measured in a similar way by sampling from a helium bottle after sampling from a bottle with 20% O2 and 79% N2. The carryover of O2 and N2 was not detectable with Iso-Versinic; with Marprene, it yielded approximately 7 ppmv O2 and 16 ppmv N2. For CO2, a bottle with 8.5% CO2 yielded a carryover of 13.3 ppmv (0.016%) for Marprene and 7.3 ppmv (0.0086%) for Iso-Versinic

#### **3.1.3.4 Software**

The system is controlled by Python software which checks the readiness state of the chromatography software, controls the autosampler and the peristaltic pump, triggers the GC, records the output of the NO analyser, and sorts and tabulates the

**Fig. 3.5** Leak test: Results of repeated sampling from two 120 ml bottles containing helium. The O2 (blue dots) and N2 (red dots) concentrations increase on average by 1.2 ppmv and 2.3 ppmv per sampling, respectively. The insert shows the leaks, the difference between consecutive measured ppmv values. The leaks do not appear to increase over the 60 samplings (These measurements have been adjusted for the dilution caused by replacing the sampled gas with helium; unadjusted measurements are shown in grey)

results. The bottles may be given individual rates of sampling, and the sampling rate of each bottle may be easily changed during an experiment (The problem of collisions of sampling time is solved by putting the bottles in a priority queue and linking together bottles that must be sampled in sequence). It is also possible to let the sampling rates automatically depend on the measured concentrations during a run (for example, measuring more often when the oxygen concentration runs low), but this requires some programming from the experimenter. For the inspection of results during an experiment, a graphical user interface is provided which lets the experimenter quickly plot time series of raw data or estimated concentrations for chosen substances and bottles.

#### **3.1.3.5 Data Handling and Analysis**

The system produces large amounts of data, requiring adequate handling. We have developed software that compiles the data as a result matrix for each gas (vial nr \* time), and calculation routines for estimation of gas transport between headspace and liquid, concentrations of the gases in the liquid, and the kinetics of transformations (Molstad et al. 2007). The software has been used in many studies of N2O/N2 gas kinetics, NO scavenging kinetics, analyses of oxic and anoxic respiration, etc. (see publication list of NMBU Nitrogen Group https://www.nmbu.no/en/research/groups/ nitrogen/publications). The software is provided free of charge on request.

#### **3.1.3.6 Documentation**

This description of the robotic incubation system is modified from a technical report (Molstad et al. 2016) which is available on Researchgate. The report should be referred to by its DOI, together with Molstad et al. (2007), when publishing results obtained with this incubation system.

## **3.2 Automated Chamber Systems for Field Measurements**

## *3.2.1 Field Techniques Using GC Systems*

Accurate measurement of GHG flux from soils is challenging because of the high temporal and spatial variabilities in the processes contributing to GHG emissions (Butterbach-Bahl et al. 2016), which result in a variability of soil surface fluxes often spanning 2–3 magnitudes. Addressing these two components of variability remains a challenge.

Automated chamber systems are increasingly used to address temporal variability. These systems consist of a set of automated chambers, mostly the closed chamber type if fluxes of rather non-reactive trace gases such as CH4, CO2, or N2O are measured, or of the dynamic chamber type, if soil–atmosphere fluxes of reactive gas species such as NO, NO2, or NH3 are measured (e.g. Butterbach-Bahl et al. 1997).

One of the first times an automated closed chamber system was used was for the quantification of CH4 fluxes from rice paddy fields in Italy (Schütz et al. 1989). These authors showed that CH4 fluxes vary strongly on diurnal to seasonal scales and, by using automated chamber systems with sub-daily measurements of fluxes, provided for the first time an accurate quantification of cumulative seasonal CH4 fluxes. The development of automated measuring chambers continued over years, and Papen and Butterbach-Bahl (1999), Gasche and Papen (1999) and Butterbach-Bahl et al. (2002) provided examples of how such systems can be used for quantifying the soil–atmosphere exchange of N2O, NO, NO2, and CH4 continuously over entire years (Plate 3.7). These measurements revealed that even during freeze–thaw periods forest soils are extremely strong sources for the atmospheric GHG N2O (Papen and Butterbach-Bahl 1999) and that in years during which such freeze–thaw N2O fluxes occur, the annual flux magnitude might be a factor of 2–10 times higher than in years without freeze–thaw events (Plate 3.7).

Only by using automated chamber systems, Wolf et al. (2010) were able to accurately quantify N2O fluxes during freeze–thaw events in steppe ecosystems of Inner

**Plate 3.7** Schematic drawing of an automated chamber system for measurements of soil GHG fluxes with five chambers and a gas chromatograph equipped with an Electron Capture detector (for N2O analysis) and Flame Ionization Detector (for analysis of CH4 plus CO2 [the latter requires use of a methanizer]) (right panel). The right panel show a picture of an automated chamber with pistons to close/open chambers by compressed air (top), a gas chromatograph fixed to a small box (middle), and a dataset of mean daily N2O and NO fluxes as obtained by the use of an automated closed chamber system (N2O) and an automated dynamic chamber system. Further details describing these measurements are provided in Papen and Butterbach-Bahl (1999) and Gasche and Papen (1999)

Mongolia. In this study, not only high seasonal, but also diurnal flux variations were observed. Meanwhile, to the knowledge of the authors, approx. 20–30 groups worldwide are using automated chamber systems, i.e. still in a minority of GHG studies are such systems used for quantifying GHG emissions from upland and lowland agricultural, forest, or grassland ecosystems, while installations are spanning various climate zones, i.e. from the tropics to alpine systems. Most recently, those automated closed chamber systems were combined with automated lysimeter systems, so that environmental N and C losses can be quantified at the same time (Fu et al. 2017, Plate 3.8). At the same time, the use of laser spectroscopy instead of gas chromatography allowed to lower the flux detection limit significantly, so that closing times for chambers could be reduced (Savage et al. 2014).

But how important are high-frequency measurements of GHG emissions from soils? Does this indeed significantly improve the accuracy of annual flux estimates? Or do weekly or monthly measurements suffice to conclude on annual exchange rates

**Plate 3.8** Automated GHG flux measurements at a grassland site in Bavaria by the Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Garmisch-Partenkirchen, Germany, using a closed chamber robot system in combination with lysimeters. For this installation, undisturbed soil monoliths (1 m diameter, 1.8 m depth) were taken at various locations along a climate gradient and re-installed at a common site. This setup allows to monitor shortand long-term effects of climate change (changes in rainfall and temperature) on ecosystem CO2, N2O, and CH4 fluxes as well as on groundwater recharge and leaching of nitrogen compounds (DON, NO3 <sup>−</sup>, NH4 +). Photo: courtesy of Dr. Ralf Kiese, who is also leading this research programme at IMK-IFU, which is part of the German Helmholtz TERENO initiative

of GHG between soils and the atmosphere? These questions were addressed in a study by Barton et al. (2015) using 28 published datasets of sub-daily N2O fluxes from a variety of different terrestrial ecosystems across the globe. The aim of the study was (a) to provide guidelines for sampling frequency to estimate annual N2O fluxes using manual chambers and (b) to quantify the uncertainty if the sampling frequency is lowered to 2–3 times per week, weekly, or monthly or if the sampling frequency is varied, with intensive measurements during episodes with expected high fluxes (e.g. following fertilization events or first rainfalls following a dry period), but lower frequency measurements during periods with expected low fluxes (e.g. dry periods). The analyses by Barton et al. (2015) show that "Nitrous oxide emissions need to be measured daily to accurately determine annual N2O flux in environments where data has not previously been collated", and that daily sampling was required to achieve annual N2O fluxes within 10% accuracy compared to sub-daily measurements with automated chambers. Lowering the measuring frequency to weekly, two weekly or monthly measurements result in an under- or over-estimation of annual N2O fluxes by up to 93.5%. Overall, the authors suggested that the current uncertainty in global terrestrial N2O budgets associated with the upscaling of field-based datasets can be decreased significantly using adequate sampling frequencies (daily, every second day, or by using automated measuring systems). For further reading the paper of the Global Research Alliance by Grace et al. (2020) on automated techniques is recommended.

## *3.2.2 Combination of Automatic Chamber System and CRDS Analyser for Field GHG Flux Measurements*

The LI-8100A Automated Soil CO2 Flux System (Licor, Inc.) is designed to measure CO2 efflux with an IR gas analyser (IRGA) from soils and low vegetation using up to 16 automated chambers with the closed non-steady-state transient approach controlled by a multiplexer unit LI-8150. This system allows the combination with a CRDS analyser (see Chap. 2, Sect. 2.7), capable of measuring other trace gases, including CH4, N2O, and NH3. They can be interfaced with the LI-8100A System and the SoilFluxPro data processing software (freely downloadable from the LI-COR website) to compute fluxes for the additional gases. There is an application note available that describes how to integrate a third-party analyser like the Picarro G2508, which can measure CO2, CH4, N2O, NH3, and H2O simultaneously, into the system and the procedure used to compute fluxes of additional gases in SoilFluxPro software (Licor 2016).

There are three options of combining the two systems (see Plates 3.9 and 3.10); the first two options (Plate 3.9a, b) connecting the CRDS analyser to the primary sampling loop or as a subsampling loop are described in the application note and need adjustments of the flow and a calibration of the IRGA (see application note, Licor 2016). The third mode of gas flow path does not need these adjustments, and the LI-8100A might also be used without the CRDS analyser without the need for readjustments.

If the system is used in humid and cold regions, a water trap or nafion tube to dry the measurement gas flow is recommended to protect the CRDS analyser from condensed water. In the LI-COR control software, LI-8100A the total gas volume of the system (chamber, tubes, analyser internal tubing, water trap) has to be defined for correct flux calculations. With the LI-8100A software, the valve system of the Multiplexer and the measurement sequence of the 16 automated chambers can be controlled.

It is important to synchronize the system clocks of the IRGA and the Picarro frequently (daily–weekly); otherwise, the SoilFluxPro software cannot match the LI-COR and the Picarro dataset for flux calculations properly. Matching of the LI-COR and the Picarro dataset in the SoilFluxPro software is also explained in the application note (Licor 2016).

For long-term field measurement campaigns, the Picarro Analyser should be placed in a water proof box. If this system is used in a hot environment, the housing box for the Picarro Analyser has to be cooled so that temperature in the box stays

**Plate 3.9** Schematic modes of combining the LI-8100A automated chamber system including up to 16 clear or opaque chambers with a CRDS Picarro five-gas analyser G2508 for simultaneous flux measurement of CO2, CH4, N2O, NH3, and H2O: **a** entire measurement gas flow enters the IRGA first, then the CRDS analyser, **b** only part of the measurement gas flow from the IRGA Outlet enters the CRDS analyser, and **c** the IRGA unit is only used to control the pump and valve system of the multiplexer but not for CO2 flux measurement. The multiplexer has 16 ports and can be connected to clear and opaque chambers, to soil air probes or aboveground sampling ports

**Plate 3.10** The combined automatic chamber system with CRDS analyser for field GHG-flux measurement in the Giessen climate change experiment combining air Temperature warming of + 2 °C and free-air CO2 Enrichment of +20% (T-FACE). In the left-hand box, the Picarro analyser is installed including its monitor and pump. The LI-8150 Multiplexer (big yellow box) is connected to seven automated chambers and eight soil air probes and switches every 5 to 30 min from one port to the next controlled by the LI-8100A Analyzer Control Unit (small yellow box). The screen (lower left side) shows stable N2O concentrations (zero flux, 1st panel above), CH4 uptake (2nd panel), stable H2O concentrations (3rd panel), and ecosystem CO2 respiration (4th panel on the screen)

below 35 °C; otherwise, the analyser will go into error. A cooling solution with a small air condition unit is possible and is described in the Picarro Application Note AN041 (Anthony and Silver 2020).

This combination of an automatic chamber system and CRDS Analyser allows to set up a high-frequency measurement sequence including up to 16 ports. A combination of clear and opaque chambers, soil air probes (e.g. Accurel© tubes installed in different soil depths), and aboveground air sampling ports is possible to create detailed data on GHG concentrations in different soil depths and aboveground and the resulting gas fluxes. The LI-8100A system allows to install and measure soil temperature and moisture probes at each automated chamber to record the top soil conditions during flux measurement. The result of such data sets allows a better understanding of hot moments of GHG emission peaks during long-term measurement campaigns and the hotspots of GHG production within the soil profile.

3 Automated Laboratory and Field Techniques … 137

## **References**


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 4 Micrometeorological Methods for Greenhouse Gas Measurement**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** Micrometeorological techniques are useful if greenhouse gas (GHG) emissions from larger areas (i.e. entire fields) should be integrated. The theory and the various techniques such as flux-gradient, aerodynamic, and Bowen ratio as well as Eddy correlation methods are described and discussed. Alternative methods also used

M. Zaman (B) · L. Heng

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang

Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

T. Clough Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

© The Author(s) 2021

M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_4

141

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

are Eddy correlation, mass balance techniques, and tracer-based methods. The analytical techniques with current state-of-the-art approaches as well as the calculation procedures are presented.

**Keywords** Eddy correlation · Flux gradient · Energy balance · Mass balance

## **4.1 Introduction**

Gas transport takes place through the eddying motion of the atmosphere, which transfers sections of air from one level to another; this is used as the basic concept of the micrometeorological approaches used to measure GHG fluxes to or from the soil surface (Denmead 1983). Viscous forces reduce turbulence in distances less than 1 mm, and molecular diffusion is crucial for transport within these very short distances. The basic transportation of gas throughout the free atmosphere, up to within approximately 1 mm of the emitting or absorbing surface, is caused by the turbulent diffusion, where individual eddies are relocated. Therefore, in practically measured distances, the overriding approach is turbulent transport. In the easiest micrometeorological methods, it is possible to measure the GHG flux through sensing the

K. Dawar

W. X. Ding Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China


J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

C.-M. Görres

Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

M. E. G. Hofmann Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

velocities and concentrations of components of the turbulence (Fowler and Duyzer 1989).

With the exception of the mass balance method (see below), micrometeorological methods are based on the assumption that the GHG flux to or from the surface is identical to the vertical flux measured at the reference level some distance above the surface (Eugster and Merbold 2015). These two quantities may differ as a consequence of three processes: (i) chemical reaction within the air column between the measurement level and the surface; (ii) changes in concentration with time and therefore changes in the storage of the trace gas within the air column; and (iii) horizontal gradients in air concentration leading to advection (Fowler and Duyzer 1989).

Additionally, prerequisites of these methods include large uniform surface areas, uniform flux to the surface in the upwind area that effects the sample point (fetch), and the atmospheric conditions must be unremitting throughout each measurement period. Measuring the flux at the selected sampling points, on flat homogeneous terrain, provides the average vertical flux over the upwind fetch. The constant flux layer is at a range extending vertically to approximately 0.5% of the upwind uniform fetch. Therefore, the constant flux layer, on a uniform 200 m field, is located around 1 m deep at the downwind edge. The height of the sampling points should be placed within the height range where this vertical flux is constant (Monteith 1973; Monteith

D. Lewicka-Szczebak

Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold

Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

M. Šimek

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

M. dos Reis Martins · S. Urquiaga EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

R. Well Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany and Unsworth 1990). It is evident that experimental areas need to be quite large and uniform before these techniques can be used in accordance with theory.

Two general micrometeorological techniques are used to measure trace gas flux density: eddy correlation and flux gradient (gradient diffusion). The description of these methods is derived primarily from Baldocchi et al. (1988), Denmead (1983), and Fowler and Duyzer (1989). While the overall principle of micrometeorological measurements has not changed, there have been advancements in analysers capable of measuring other greenhouse gases such as CH4 and N2O besides CO2 and H2O in relevant temporal resolution to allow continuous field-scale observations of these gases. Similarly, standardization approaches have been implemented during the past decade, allowing comparison of measurements across sites, ecosystem types, and climatic regions (Baldocchi 2014; Eugster and Merbold 2015).

## **4.2 Flux-Gradient Method**

Though there are differences, the flux-gradient theory assumes that molecular diffusion is the same process as the turbulent transfer of gas. However, molecular diffusion occurs due to the random motion of molecules, while turbulent transfer, or eddy diffusion, occurs when parcels of air move from one level to another. This results in eddy diffusion generally being several times greater than molecular diffusion. Wind speed, height above surface (i.e. soil, plant canopy, or water), the vertical temperature gradient, and the aerodynamic roughness of the surface are what determine the actual magnitudes. The gradient is positive by convention if the concentration increases towards the surface, and vice versa. Additionally, the turbulent flux is proportionate to the product of an eddy diffusion coefficient and the average vertical concentration gradient of the gas.

The basic data for numerous methods of measuring vertical fluxes over large surfaces are stipulated by measurements within the constant flux layer of gradients with height of wind velocity, concentration of trace gases, and air temperature. The Bowen ratio (energy balance) and the aerodynamic techniques are two of the more popular methods of calculating vertical fluxes.

J. Zhang School of Geography, Nanjing Normal University, Nanjing, China

S. Zaman

University of Canterbury, Christchurch, New Zealand

K. Lenhart Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

**Plate 4.1** Typical instrumentation for flux-gradient measurement in the field

## **4.3 Aerodynamic Method**

The aerodynamic method is based on the relationship between the momentum flux equation and the wind speed gradient. This technique requires the measurement of wind speed at two or more heights and the concentration gradient of the gas at these heights. The trace gas flux is influenced by vertical air density gradients due to water vapour and heat fluxes, and stability corrections must be made (Fowler and Duyzer 1989).

Flux-gradient measurements can be used to estimate emissions from field-size areas of land: 1–100 ha. Plate 4.1 shows the system employed for gases which can be analysed directly in a flowing stream. Gas streams are pumped through sampling tubes from different heights on a gas sampling mast, and wind speed and temperature at different heights are also measured.

## **4.4 Bowen Ratio (Energy Balance Method)**

This procedure is based on the energy balance at the surface, and measurements of the wind velocity profile are not necessary. The inward net radiation is divided among the soil heat flux, sensible heat flux, and the latent heat flux. The Bowen ratio consists of the ratio of sensible to latent heat fluxes. The measurements required for this method are the vertical gradients of gas concentration, humidity, and temperature. Without introducing the uncertainty and difficulty that occurs with stability corrections, which are necessary for the aerodynamic method, these measurements specify approximations of the fluxes of sensible heat, the trace gas, and water vapour. Furthermore, this method allows for the calculation of the rate that water evaporates (Denmead 1983).

The largest and most important drawback for trace gas fluxes is the substantial net radiation fluxes required by energy balance methods. In cloudy, night, or winter conditions, the available net radiation is frequently too small to permit satisfactory flux estimates (Fowler and Duyzer 1989). A further night-time problem is that condensation of dew on radiation instruments leads to erroneous measurements. The employment of both aerodynamic and energy balance methods would seem advisable at any time, but particularly so when the diurnal pattern of trace gas loss is being investigated (Denmead 1983).

## **4.5 Eddy Correlation Approach**

Using the eddy correlation technique, a trace gas flux density (vertical transport of gas past a point in the atmosphere) is obtained by correlating the instantaneous vertical wind speed at a point with the instantaneous concentration of that gas. In the natural environment, the eddies which are important in the transport process occur with frequencies extending up to 5 or 10 Hz. Therefore, a rapid response detector is required (Denmead 1983; Aubinet et al. 2012). With a fast response detector, eddy correlation methods offer many advantages. In particular, they require a minimum number of assumptions about the nature of the transport process. The method requires real time, continuous measurement of wind speed, and gas concentration at only one height above the surface. The technique can also be used inside plant canopies, above soils, and at night (Eugster and Merbold 2015). The method does require fast responding computing facilities to cope with the rapid data acquisition rates and rather large storage capacity. Moreover, the correction of the trace gas density is required with respect to water vapour and heat transfer. These corrections become quite large for some gases (Denmead 1987; Burba et al. 2008; Webb et al. 1980). Applications of both eddy correlation and flux-gradient approaches are limited to situations in which the air analysed has passed over a homogeneous exchange surface for a long distance so that profiles of gas concentration in the air are in equilibrium with the local rates of exchange. The methods assume that horizontal concentration gradients are negligible as a standard requirement to apply the eddy correlation technique (Baldocchi et al. 1988, 2001). With the development of easy-to-deploy quantum cascade laser absorption spectrometers QCLAS (i.e. Mohn et al. 2008, 2013; McManus et al. 2010), field applications of these instruments to continuously measure CH4 and N2O exchange above ecosystems are possible. Previous tunable diode laser (TDL) absorption techniques (Thurtell et al. 1991) were often difficult to maintain in the field over a long period. This was particularly the case of the TDL technique which is based on infrared (IR) absorption spectrometry, whereby the extent of absorption depends upon path length, line strength, and absorber concentration. Liquid nitrogen temperature diodes are commercially available to cover the IR spectrum from about 2 to 10 pm, the region where most trace gases have absorption spectra. The diode laser is mounted in a liquid nitrogen-cooled dewar. A heater mounted inside the dewar controls precisely the laser temperature in the 78–110 K region. The centre frequency of the laser emission is controlled by the cold head temperature. Each diode is temperature tunable over about 100 cm−1. With the newly developed and by now commercially available instruments, difficult setups including liquid nitrogen supply in the field are no longer required. Nevertheless, the instruments capable of measuring CH4 and N2O at high temporal resolution as needed for the application in eddy correlation setups still need to be handled with more care than instruments capable of measuring CO2 and H2O concentration. A recent overview paper highlights the necessary steps to set up reliable and long-lasting measurements of CH4 and N2O with the eddy covariance technique (see below).

Standardization of GHG flux measurements was already initiated two decades ago (Aubinet et al. 2012). With the establishment of environmental research infrastructures in Europe, the US and Australia (i.e. Integrated Carbon Observation Systems–ICOS, National Ecological Observatory Network–NEON, and the Terrestrial Ecosystem Research Network–TERN), further steps towards standardized GHG flux measurements were made. These go beyond the standardization of eddy flux measurements and further include GHG chamber measurements as well as standardized collection of ancillary data to interpret GHG flux data. Most of the best practices concerning eddy covariance (EC) flux measurements, including the actual setup of EC towers, the data collections as well as calculation procedures have been summarized in a recent Special Issue in International Agrophysics (Franz et al. 2018). While the approaches explained in this special issue are largely focusing on the highest quality of data, modified and often simpler approaches including instruments that may be used in more remote environments where grid power may not be available are also covered and referred to for more details.

## **4.6 Alternative Micrometeorological Methods**

## *4.6.1 Eddy Accumulation*

The eddy accumulation technique has been proposed as a possible means of measuring the fluxes of constituents, e.g. N2O. The technique involves the collection of upward and downward transported material in two separate containers at rates proportional to the vertical wind speed. The flux is evaluated as the difference in concentration accumulated over the sampling period. Unfortunately, there are substantial practical difficulties in measuring small concentration differences in the accumulators and sampling the air according to vertical wind velocity (Baldocchi et al. 1988).

## *4.6.2 Mass Balance Technique*

The mass balance method has been used to measure fluxes of ammonia from small fields (Denmead et al. 1977; Wilson et al. 1983). Gas flux density is related to the horizontal distance from the upwind edge of the field and the top of the air layer influenced by the emission of the gas. The method assumes that the mean horizontal turbulent flux is much smaller than the mean horizontal advective flux. The top of the air layer influenced by the emission of the gas is a function of stability and surface roughness which can be estimated (Denmead 1983). It is recommended that concentrations of the gas and wind speeds are measured at five or more heights above the surface. The horizontal distance from the upwind edge must be known. Thus, to minimize the effect of changing wind direction on this horizontal distance, it is recommended that experiments be conducted in a circular plot with the instrument array in the centre (Baldocchi et al. 1988).

## **4.7 Non-isotopic Tracer Release and Measurement Methods**

Methods that have been used to consider an assortment of atmospheric transport and diffusion problems include non-isotopic tracer techniques (e.g. using halogenated gases such as sulphur hexafluoride (SF6)) (Benner and Lamb 1985). Procedures exercising tracer release and measurement techniques have been developed to approximate flux rates from area and point sources. These tracer release and measurement techniques can be utilized when a definite plume of the target gas, methane, for example, can be readily detected in the ambient environment. When these conditions are present, based on analysis of downwind and upwind samples, the plume of dispersed emissions can be located. Next, the tracer gas (e.g. SF6) should be released, at a known rate, in a configuration similar to that of the target gas. For example, if release of the target gas is at several distinct points, the release of the tracer gas should be at those same points. To ensure that the locations and shapes of the target and tracer plumes match one another, downwind and upwind samples are analysed. The known release rate of the tracer and the ratio of the tracer and target gas concentration are used to calculate the flux rate of the target gas. The ideal conditions for this technique are with steady meteorological surroundings, which allow the target and tracer plumes to comprehensively mix. Further, the flux rates need to be sizeable enough to permit the detection of the target and tracer gases at distances that are great enough to enable thorough mixing. To determine the desired flux rate if there is insufficient knowledge about the pattern of the target gas release, it is possible that single-point release of the tracer can be used and an analysis of the ratio of the crosswind integral of the tracer and target gas plumes.

It is essential to conduct immediate, on-site, analysis to achieve the most success. Chapter 2 describes the preferred equipment for continuous mobile gas analysis.

## **References**


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 5 Direct and Indirect Effects of Soil Fauna, Fungi and Plants on Greenhouse Gas Fluxes**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** Soils harbour diverse soil fauna and a wide range of soil microorganisms. These fauna and microorganisms directly contribute to soil greenhouse gas (GHG) fluxes via their respiratory and metabolic activities and indirectly by changing the physical, chemical and biological properties of soils through bioturbation, fragmentation and redistribution of plant residues, defecation, soil aggregate formation,

M. Zaman (B) · L. Heng

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl

Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang

Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

© The Author(s) 2021 M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_5

151

herbivory, and grazing on microorganisms and fungi. Based on recent results, the methods and results found in relation to fauna as well as from fungi and plants are presented. The approaches are outlined, and the significance of these hitherto ignored fluxes is discussed.

**Keywords** Soil fauna · Fungi · Microorganisms · GHG

## **5.1 Greenhouse Gases from Soil Fauna**

## *5.1.1 Introduction*

Soils harbour a diverse group of fauna. Based on their size and the resulting occupied soil space, soil animals can be grouped into microfauna (<200 μm: protists, some nematodes), mesofauna (0.2–2 mm, e.g. nematodes, microarthropods, enchytraeids, molluscs), and macrofauna (>2 mm, e.g. earthworms and other worms, ants, beetles, termites, spiders, molluscs) (Lavelle et al. 2006). These animals directly contribute to soil GHG fluxes via their respiratory and metabolic activities and indirectly by changing the physical, chemical and biological properties of soils through bioturbation, fragmentation and redistribution of plant residues, defecation, soil aggregate

T. Clough

K. Dawar

W. X. Ding

Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad

Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

C.-M. Görres Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

formation, herbivory, and grazing on microorganisms and fungi. Additionally, microhabitats are created that can support greater microbial activity than bulk soil does (Brown et al. 2000; Lubbers et al. 2013). Thus, soil fauna can substantially influence the spatial and temporal variability of GHG fluxes in the field. Additionally, climate, abiotic soil conditions, land management and interactions within the soil food web modify the abundance, activity and vertical distribution of fauna in soils. The magnitude of the effect of soil fauna on soil GHG fluxes remains poorly quantified. Most of our current knowledge comes from laboratory experiments, while field data are scarce, which are often controversial and have been limited to only a few regions and species. Filser et al. (2016) provided the latest review and an extensive literature list on soil fauna and its effects on soil organic matter (SOM) turnover and nutrient cycling. The following sections provide a broad overview of the current knowledge of direct and indirect effects of soil animals on soil GHG fluxes and summarize available field methods for quantifying these effects.

## *5.1.2 Overview of Fauna on GHG Emissions*

#### **5.1.2.1 Carbon Dioxide (CO2)**

Spatial clustering of soil fauna can result in an increased spatial variability of soil CO2 emissions due to the creation of CO2 point sources (Ohashi et al. 2007). Positive correlations between soil CO2 emissions and faunal biomass have been observed (Binet et al. 1998), and faunal respiration can contribute between 2 and 40% of total

M. E. G. Hofmann

Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh

D. Lewicka-Szczebak Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

soil respiration (Briones et al. 2004; Lubbers et al. 2013). Ignoring these soil CO2 flux hotspots might result in substantial errors in ecosystem carbon (C) balances (Ohashi et al. 2007). However, indirect effects of the presence of soil fauna on total soil respiration often seem to be more important than direct respiratory contributions.

The most studied soil fauna groups with respect to soil CO2 fluxes are earthworms, potworms (*Enchytraeids*), termites and beetles. Earthworms are divided into three functional groups based on their feeding and movement patterns in soils. Anecic earthworms live in deeper soil zones where they ingest moderate amounts of mineral soil as well as litter which they drag down from the soil surface into their burrows. Earthworms living and feeding mainly in the rhizosphere are called endogeic. These earthworms ingest substantial amounts of mineral soil. Epigeic earthworms live and feed preferentially in the litter zone above the mineral soil (Horn et al. 2006). Soil properties are altered by its passage through the earthworm gut, and earthworm casts, burrows and middens provide microhabitats for smaller soil animals and microorganisms (Brown et al. 2000; Lubbers et al. 2013). Interactions of functional earthworm groups seem to lead to greater mean soil CO2 emissions compared to soils where only one functional earthworm group is present (Speratti and Whalen 2008). Increased soil CO2 emissions through earthworm activity seem to be transient, though in the long-term earthworms might increase C storage in soils through stabilization of SOM in stable micro-aggregates (Lubbers et al. 2013; Six and Paustian 2014).

Earthworm abundances increase and species compositions change when moving from conventional tillage systems to systems with reduced or no tillage (Lubbers et al. 2013). In conventionally tilled fields, enchytraeids seem to be more important for organic matter (OM) mineralization than earthworms (van Vliet et al. 2004). Like earthworms, enchytraeids are burrowing animals that can ingest large amounts of soil, including fungi, bacteria, algae and even dead bodies of larger invertebrates (Briones et al. 2004; van Vliet et al. 2004).

M. Šimek

M. dos Reis Martins · S. Urquiaga

R. Well

J. Zhang

K. Lenhart

Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany

S. Zaman

University of Canterbury, Christchurch, New Zealand

School of Geography, Nanjing Normal University, Nanjing, China

Termites are primarily concentrated in tropical grasslands and forests, and based on their nesting habits are classified as mound-building, wood-nesting or subterranean termites (Jamali et al. 2011a). Termite mounds are point sources of CO2 emissions (Brümmer et al. 2009) which can show large seasonal patterns due to seasonal population dynamics of termites inhabiting the mounds, as well as changes in gas diffusivity of the mound walls (Jamali et al. 2011b, 2013). These patterns are controlled by changes in temperature, moisture, food quantity and quality, as well as the termites' life cycle, and can be highly species-specific (Jamali et al. 2011b). Although total mound CO2 emissions are mainly comprised of termite respiration, microbial respiration within the mound walls can also be a significant contributor. Suitability of mound walls as habitats for microorganisms depends on properties such as mound bulk density and wall thickness that again can vary widely between termite species (Jamali et al. 2013). Termite CO2 emissions might be negligible on an ecosystem scale, but uncertainties in CO2 flux estimates are high (Brümmer et al. 2009; Jamali et al. 2013). These uncertainties are due to a lack of field studies for termites that do not construct mounds, uncertainties in the global estimates of total termite biomass and number of nests, and a lack of process understanding of gaseous exchange between termites and the atmosphere (Jamali et al. 2011a).

In recent years, another group of arthropods has received increasing attention regarding its effects on soil GHG fluxes–dung beetles. On farmland with grazing animals, dung pats are GHG flux hotspots. Dung beetles are important contributors to dung composition and can significantly alter the temporal pattern of CO2 emissions from dung pats. Dung beetles can either enhance or suppress CO2 emissions from dung pats based on the species and their feeding behaviours (Penttilä et al. 2013; Piccini et al. 2017).

#### **5.1.2.2 Methane (CH4)**

Soil fauna groups known to emit CH4 are termites, scarab beetles, millipedes and cockroaches. Emissions have been observed from tropical as well as temperate species, but not from all tested species within a group, and within-group variability is high (Egert et al. 2005; Sustr and Šimek 2009; Brune 2010; Kammann et al. 2017). Methane is produced by methanogenic microorganisms in the insect gut which either live on intestinal surfaces or as endosymbionts inside gut-inhabiting protozoa (Brune 2010). Like termite respiration, termite CH4 fluxes scale with termite biomass (Jamali et al. 2011b). Methane emitted by termites can significantly affect the CH4 balance of an ecosystem, offsetting part of the CH4 sink of the surrounding soil (Jamali et al. 2011a). However, not all CH4 produced by termites reaches the atmosphere because part of it is oxidized to CO2 by methanotrophic microorganisms during its diffusive passage through mound walls and the soil surrounding termite nests (Sugimoto et al. 1998; Jamali et al. 2011a, 2013). Termites can also directly enhance CH4 oxidation as their burrowing activities increase soil diffusivity, and the increasing soil CH4 concentration supports a larger and more active methanotrophic community (Bender and Conrad 1995). Methane emissions from other soil-dwelling insects have thus far been regarded as negligible at the ecosystem scale (Egert et al. 2005; Sustr and Šimek 2009). However, recent studies suggest that CH4 emitted by Scarab beetle larvae have the potential to significantly enhance soil CH4 oxidation and thus affect the soil CH4 balance in well-aerated upland soils (Kammann et al. 2017). However, the observations on dung beetles are controversial. Methane emissions from dung pats, especially if they are from grazing dairy cows, can be extremely high, being able to switch a field from a net CH4 sink to a net CH4 source. Depending on species, dung beetles have been shown to either drastically reduce or increase dung pat CH4 fluxes (Penttilä et al. 2013; Piccini et al. 2017). Adult scarab beetles can also be CH4 sources (Bijnen et al. 1996). Earthworms are generally not considered as CH4 emitters (Drake et al 2006), but recently, CH4-emitting earthworms were discovered in Brazil (Schulz et al. 2015).

#### **5.1.2.3 Nitrous Oxide (N2O)**

Earthworms are the only faunal group for which a considerable amount of literature on their effects on soil N2O fluxes is available (Kuiper et al. 2013). Earthworms themselves as well as their casts and burrow walls can be N2O emission hotspots, able to increase soil N2O emissions by more than 40% (Brown et al. 2000; Lubbers et al. 2013). In contrast to CH4-emitting soil fauna, earthworms do not have a quantitatively significant indigenous microbial biome in their guts. Instead, N2O emissions are due to the activation of nitrate- and nitrite-reducing bacteria in the ingested material during its gut passage. Conditions favourable for denitrifiers inside the earthworm gut include anoxia, an ample supply of C as well as nitrate and nitrite, a suitable pH, and a high moisture content (Drake and Horn 2006; Horn et al. 2006). Gut passage does not only affect soil denitrifiers, but the overall soil bacterial community composition. This is due to selective digestion of bacteria by the earthworms and the mixing of ingested material with mucus, an aqueous secretion rich in organic molecules (Drake et al. 2006). However, the impact of earthworms on functional bacterial soil communities remains unclear (Chapuis-Lardy et al. 2010).

The impact of earthworms on soil N2O emissions depends on soil type, environmental conditions, earthworm functional groups and species compositions, but the emissions themselves do not seem to scale with earthworm biomass (Chapuis-Lardy et al. 2010). The impact of earthworms on soil structure, gas diffusion, and N and C availability seems to have a much larger effect on soil N2O fluxes than direct N2O emissions from the earthworms (Bertora et al. 2007; Kuiper et al. 2013). Increase in N2O emissions in the presence of earthworms has only been observed for soils to which crop residues or fertilizers were applied (Chapuis-Lardy et al. 2010; Speratti and Whalen 2008). Earthworms initially speed up residue mineralization and thus increase inorganic N availability. However, as already seen for CO2 emissions, this seems to be a transient phenomenon and in the long-term earthworm presence may lead to lower N2O emissions, compared to fields without an abundant earthworm population (Bertora et al. 2007; Chapuis-Lardy et al. 2010). The earthworm effect on N2O fluxes due to residue incorporation can be partly replaced by ploughing in agricultural systems (Bertora et al. 2007). Other soil invertebrate fauna can also significantly modify soil N2O fluxes, both positive and negative, with the effect being dependent on the fauna's ability to modify soil structure and its relative biomass in the soil fauna food web (Kuiper et al. 2013; Penttilä et al. 2013; Piccini et al. 2017). However, for soil fauna groups other than earthworms, studies have mainly focused on N mineralisation, rather than on direct gaseous N fluxes (van Vliet et al. 2004; Kuiper et al. 2013).

## *5.1.3 Field Methodology*

Soil fauna can substantially influence the spatial and temporal variability of GHG fluxes in the field through direct and indirect effects, but it is often logistically or methodologically not possible to quantify them in situ. Nevertheless, a basic knowledge of the local soil fauna can significantly aid the interpretation and upscaling of GHG flux data. Thus, in the planning phase of a GHG flux study, one should consider the following questions:


There are no standard field methods and measurement protocols available specifically for quantifying soil fauna GHG fluxes since the vast majority of studies have been in the laboratory or in field mesocosms under controlled and simplified conditions. However, all of the measurement methods described in this book can potentially be adapted to quantify soil fauna GHG fluxes in situ, and examples of such adaptations are listed in the following paragraphs.

Measurements of net GHG fluxes with the chamber or eddy covariance method (Chap. 4) include all soil fauna effects at the plot scale regardless of the available knowledge on these animals at a particular field site. Abundances and total biomass of key species and functional groups can be quantified in accompanying soil sampling campaigns. Abundance, weight and vertical distribution in the soil are the easiest parameters to obtain for soil fauna characterization. Greenhouse gas emissions of excavated animals can be directly estimated in the field by incubating them in hermetically sealed vessels, preferably glass vessels. Vessel size (e.g. Exetainers, test tubes, glass jars) and incubation time depend on animal size and expected emission strength (Sustr and Šimek 2009). Figure 5.1 shows a freshly excavated Scarab beetle larva (2.6 g) which was incubated in a 110 ml test tube sealed with a rubber stopper for an hour.

**Fig. 5.1** Freshly excavated Scarab beetle incubated in a 110 mL test tube

At the end of the incubation period, 25 ml of gas were extracted from the test tube with a syringe for gas concentration analysis. The GHG emission rates are then calculated from the increase of gas concentrations in the headspace air of the incubation vessel over time. Vessels should be kept in the shade or a cooler to prevent a large temperature increase relative to the soil layer from which the animals were excavated. For incubation times over an hour, wet pieces of paper towel should be added to the vessels to keep the animals from drying out. Each incubation series should include blank measurements (=empty sealed vessels). In some studies, individuals were rinsed with water before incubating them to remove attached soil particles (e.g. Horn et al. 2006). Each vessel should only be used once in the field as earthworms and scarab beetle larvae can defecate during the incubation. Any emissions from the faeces can be determined by sealing the vessel again for at least an hour after the animal has been removed. These incubations can also be performed on soil fauna which is kept in the laboratory under controlled conditions. However, one has to keep in mind that GHG emissions may strongly depend on the available food source and thus, laboratory and field quantifications might not be directly comparable.

Chambers can also be used to measure directly net GHG fluxes by soil fauna clusters like dung pats or termite mounds (Penttilä et al. 2013; Jamali et al. 2011a, b). The flexibility of the chamber and collar design permits it to fit as precisely as possible to the size of the clusters, thus reducing the relative contribution of soil adjacent to the clusters to the measured GHG flux. However, when targeting soil faunal effects on GHG fluxes with the chamber method, there are two additional considerations which have to be taken into account during construction and installation of permanent collars. First, permanent collar installation can constrain the horizontal movement of soil fauna and, as a result, change their behaviour. Depending on the targeted soil fauna groups, it might not be possible to find a collar insertion depth that adheres to the guidelines for airtight chamber measurements and permits unconstrained soil fauna movement at the same time. Second, the chosen collar material has to be durable enough to withstand any fragmentation attempts by the soil fauna, especially the mandibles of macro-arthropods (e.g. scarab beetle larvae).

Net GHG fluxes at soil fauna clusters may also be quantified using soil gas sampling probes (see Chap. 2, Sect. 2.3.) and the flux gradient technique. Jamali et al. (2013) installed nylon tubes in termite mounds for measuring CO2 and CH4 concentrations. Net mound CO2 and CH4 fluxes measured with chambers were significantly correlated to the internal mound CO2 and CH4 concentrations, respectively. There was also a significant relationship between net mound CH4 fluxes and internal mound CO2 concentrations. However, all these relationships were highly speciesspecific, and using internal mound gas concentrations as a proxy for determining net mound fluxes results in higher uncertainties in the flux estimates compared with direct chamber measurements. Nevertheless, in areas where chamber measurements on termite mounds are not possible, this method can be used with great caution to derive at least rough estimates of net mound CH4 and CO2 emissions (Jamali et al. 2013).

Stable isotope methods provide possibilities to measure GHG fluxes in situ without disturbing the soil system and are especially helpful to estimate gross fluxes, such as the contribution of soil fauna. Sugimoto et al. (1998) estimated the proportion of CH4 oxidized by diffusion through the walls of termite mounds from the difference in δ13C-CH4 between the CH4 produced inside the mounds and the CH4 captured outside during chamber measurements (natural abundance approach). An isotope pool dilution technique developed by von Fischer and Hedin (2002) allows the simultaneous estimation of gross CH4 production and gross CH4 oxidation rates in soils. Originally developed for incubating intact soil cores in hermetically sealed vessels, this technique can also be used in combination with chambers (Yang and Silver 2016). Immediately after a chamber is placed airtight onto a collar to start a flux measurement, 13C-CH4 is injected into the chamber headspace to reach an isotopic enrichment of 2–10 atom % 13C-CH4. However, headspace CH4 concentration should remain at the ambient level and not increase by more than 0.1 ppm. The chamber headspace can additionally be labelled with trace amounts of SF6 (~10 ppb) to quantify diffusional losses of the 13C-CH4 label from the chamber and then correct the measurement results accordingly. After labelling, the chamber measurement proceeds as usual, taking gas samples either manually with a syringe or automatically with an attached gas analyser (Yang and Silver 2016). Section 5.3 provides a practical example of the analysis of manually collected discrete gas samples with a stable carbon isotope analyser. Gross CH4 production and gross CH4 oxidation rates are estimated by fitting equations for the change of the amount of labelled CH4, the total amount of CH4, and the isotope ratio over time (von Fischer and Hedin 2002). This technique has the potential to be used to quantify in situ CH4 production of soil-dwelling Scarab beetle larvae and other CH4-emitting macro-arthropods in well-aerated upland soils, although it is yet to be field-tested for this purpose. The various CH4 production and oxidation processes can also be analysed via a stable isotope tracing model similar to *Ntrace*Gas described in Chap. 7, Sect. 7.5.7.

Another way to use stable isotopes in soil fauna GHG flux studies is to add 13C and/or 15N labelled plant material to soils.Methane production by scarab beetle larvae in well-aerated upland soils feeding on the residues could be estimated by quantifying the emission of 13C-CH4 from the soil by using chambers. Measuring 15N-N2O soil emissions using chambers allows estimation of the plant material contribution to the overall N2O emissions. Excavated earthworms can be freeze-dried, ball-milled and dried at 60 °C for 15N determination. This information can untangle which of the earthworm species is most active in digesting plant material residues (Giannopoulos et al. 2010).

## **5.2 Greenhouse Gases from Fungi and Plants**

Many experiments focus on the greenhouse gas balance of an ecosystem. Therefore, CH4, N2O and CO2 fluxes from the surface to the atmosphere are measured, either with chamber techniques on a number of plots or–on larger scale–via eddy covariance techniques. With these techniques, the net fluxes of GHG between the ecosystem and the atmosphere can be quantified and the sink or source strength can be determined. Nevertheless, this approach neglects simultaneously occurring production and consumption processes of CH4 and N2O plants that are part of the soil–plant–atmosphere continuum (SPAC).

Currently, when investigating greenhouse gas fluxes and emission pathways, the focus lies on microbial-derived greenhouse gas emissions from soils. Recently, it has been shown for a broad range of species that eukaryotic organisms–namely, algae (Scranton and Brewer 1977), fungi (Lenhart et al. 2012), cryptogams (Lenhart et al. 2015), animal cells (Ghyczy et al. 2008) and higher plants (Keppler et al. 2006) release considerable amounts of CH4 and N2O to the atmosphere. From those organisms, only fungi are known to produce N2O from soil. In addition, fungi can also emit CH4 (Lenhart et al. 2012). Plants can either be a source of CH4 (Keppler et al. 2006) and N2O (Lenhart et al. 2019) themselves, or they can have a "chimney" function, transporting dissolved CH4 and N2O via the transpiration stream. Depending on ecosystem and environmental conditions, this will lead to an over- or underestimation of GHG fluxes (Machacova et al. 2016). Lichens and mosses will be of minor importance in most agricultural ecosystems due to their low abundance in managed ecosystems. Nevertheless, depending on cryptogamic biomass and metabolic activity, cryptogam-derived emissions are of relevance in natural ecosystems.

**Fig. 5.2** Conceptual scheme of sources and sinks of soil air CH4 in an aerated soil ecosystem (Jugold et al. 2012). Methane uptake from the atmosphere is the dominant process, biotic CH4 formation (red arrows) under oxic conditions in plants and fungi, biotic CH4 formation under anoxic conditions in free-living Archaea and larvae containing methanogenic Archaea in their gut and chemical formation of CH4 (blue arrows) from soil organic matter are sources of soil air CH4

## *5.2.1 Methane (CH4)*

Investigating CH4 production in aerated soil in the field is difficult because it occurs simultaneously to CH4 consumption. For aerated ecosystems (e.g. upland soils), CH4 consumption usually exceeds CH4 production by several orders of magnitude; thus, it is extremely difficult to detect and quantify small CH4 production rates. Methane consumption by methanotrophic bacteria is the only biological CH4 sink of atmospheric CH4 and causes–due to the arising concentration gradient–flux of CH4 from the atmosphere into the soil. Besides atmospheric CH4, which is usually the predominant CH4 source in aerated soils, several biotic or abiotic sources are known in the plant–soil system (Wang et al. 2013). An overview is given in Fig. 5.2.

Abiotic CH4 formation from soil organic matter is triggered by solar radiation, temperature and wetting–drying cycles. For a detailed description of abiotic greenhouse gas sources, we refer to Wang et al. (2017). Biotic formation of CH4 occurs in anoxic microsites and in the gut of soil macrofauna by methanogenic archaea. Plants release CH4 from roots to soil air and from aboveground biomass directly to the atmosphere. Due to their strong connection via mycorrhiza, it is virtually impossible to distinguish between root- and fungi-derived CH4 in nature.

Methane formation of methanogenic archaea is restricted to anoxic conditions. Thus, free-living archaea occur in deeper soil layers, water-saturated soils, anoxic microsites or within the gut of soil animals (Hackstein and Stumm 1994). In contrast to *Archaea,* plants and fungi are not restricted to anoxic conditions. Thus, plants and fungi emit CH4 in the presence of oxygen (i.e. in upland soils where methaneconsuming bacteria occur), although their emission rates are in general much lower than the rates that can be achieved by methanogenic*Archaea* under anoxic conditions. Moreover, the fluxes of CH4 from these sources (Fig. 5.2) are characterized by a high temporal and spatial variability, depending on the environmental conditions such as temperature, pH, water content, O2 concentration and substrate availability.

Our knowledge about CH4 formation from plants and fungi is still rudimentary. Physical damage (Wang et al. 2009; Lenhart et al. 2015), UV-radiation (Bruhn et al. 2014) and inhibition of the cytochrome C oxidase (Wishkerman et al. 2011) leads to an increased formation of CH4 by plants. The biochemical pathways leading to CH4 formation in plants and fungi are unknown, although various organic compounds like pectins (Keppler et al. 2008), lignins (Vigano et al. 2008), hydrocarbons (Etiope and Klusman 2002), ascorbic acid (Althoff et al. 2010) and methionine (Lenhart et al. 2015) were identified as a precursor for CH4 production.

By using isotopic labelling techniques (13C), it is possible to identify the precursors of CH4. When 13C labelled compounds are added that are metabolized to CH4, also the 13C label in headspace-CH4 will be labelled. Moreover, using position-specific components (i.e. compounds where only one atom is 13C labelled), it is even possible to determine a specific atom or functional group of a molecule that is converted to CH4. One example of position-specific labelling is the amino acid methionine, where only the sulphur bound methyl group of methionine is 13C labelled (S-13CH3). In a laboratory approach, it was shown that sterile cultivated plants are able to convert this methyl group to CH4 (Lenhart et al. 2015). In a similar experiment where fungi were supplemented with acetate, and where either the carboxy group (HOO13C-CH3) or the methyl group (HOOC-13CH3) was 13C labelled (Fig. 5.3), it was shown that fungi can convert the methyl group of acetate to CH4.

## *5.2.2 A Laboratory Approach to Study CH4 Production from Plants and Fungi*

Traditional approaches that measure greenhouse gas fluxes from the plant–soil system often measure net fluxes; determination of gross rates is rare. Only one-field study of plant-derived CH4 exists (Wang et al. 2008). However, to get insights into the complex system of soil air CH4 sources and sinks occurring in the plant–soil system, we suggest investigating CH4 production processes under controlled laboratory conditions. We are aware that transfer of laboratory results to the field site is very complex and not always reasonable. On the other hand, with our current knowledge, it is not possible to conduct the following experiments in the field.

To quantify CH4 formation, a *closed chamber system* (Fig. 5.4) must be used to achieve a sufficient CH4 enrichment. For open flux chamber systems, the CH4 (and N2O) production rates of plants and fungi are probably too low to be detected. This might cause artificial effects, e.g. by affecting gas diffusion (including O2 availability) in the soil. When measuring CH4 production in the plant–soil system, methane oxidation by methanotrophic bacteria must be inhibited. This can be achieved with the gaseous inhibitors Difluormethane (DFM, 1 ml l−1) or Acetylene (1 ml l−1). Both substances are reversible inhibitors of the enzyme methane monooxygenase, resulting in a nearly complete inhibition of CH4 consumption by methanotrophic bacteria.

**Fig. 5.4** Example of a closed chamber system that can be used to measure trace gas emissions from organisms–also under sterile conditions. If the lid is equipped with two Festo tube connectors, the flask can be connected to a gas analyser for automated analysis (see Sect. 3.2.2)

## *5.2.3 Measuring Procedure*

Field-fresh soil samples should be sieved (2–6 mm) to increase the diffusivity of the inhibitors in the soil. When collecting soil cores with vegetation (e.g. grassland soil), the efficiency of inhibitors is limited due to the lower diffusivity compared to sieved soil.

Insert the field-fresh sample into an airtight flask or chamber equipped with a septum to collect gas samples or connect it to an online system for trace gas measurements (e.g. Picarro system, Plate 3.10). Add either DFM or Acetylene. To accelerate the diffusion of the inhibitor into the soil, we recommend pressure fluctuations, e.g. by "pumping" air in and out with a syringe (=volume change).

**Attention**: Those inhibitors often cause interferences with a laser-based system → check before the start of the incubation. It is therefore recommended to collect gas samples and measure the concentration with a GC system.

The time to collect gas samples depends on the concentration change, which in turn depends on the amount of sample, the volume of the flask/chamber and the production rate (mostly temperature dependent). For a temperate grassland soil, we obtained good results with 2.7 l headspace, 300 g samples (soil + vegetation), and 25 °C and 15 h incubation. A pre-experiment is recommended to determine the emission rates of CH4, N2O and CO2 under "normal" conditions before working with inhibitors. Depending on the outcome of this pre-experiment, the system must be adjusted (volume, temperature, amount of sample, sampling time). For all incubations, we add three control flasks where only water was added and no changes in CH4 concentration over time can be expected. In those flasks, gas concentration should remain constant.

## **5.3 Measuring Discrete Gas Samples with a Cavity Ring-Down Spectrometer for CO2 and CH4 Concentration and Carbon Isotope Analysis**

In greenhouse gas flux studies, it is common practice to perform the gas concentration and isotope analysis with a gas chromatograph and mass spectrometer, respectively. Some commercially available optical gas analysers are capable to combine both analyses, thus simplifying sample collection and sample processing for the user. The gas analyses to quantify the concentrations of GHG emissions and the stable isotope compositions as described in Sects. 5.1 and 5.2. were predominantly carried out with an optical analyser. One such analyser is the Picarro G2201-*i* cavity ringdown spectrometer. The Picarro G2201-*i* is a field-deployable analyser capable of simultaneous concentration and δ13C measurements for both CO2 and CH4. It can be directly connected to a chamber for continuous closed-loop measurements or employed for the analysis of discrete gas samples, which is the focus here. The following sections address sample injection modes, carrier gas stream and sample volume choice, as well as sample processing times specifically for the Picarro G2201 *i*, but large parts of the information are either directly or in a slightly modified version transferable to other commercially available optical gas analysers.

#### *Sample injection mode*

An important gas analyser property to consider when designing an experiment is its sample injection mode since this has implications on the maximum number of processible samples and the method of sample storage prior to analysis. The Picarro G2201-*i* has three possible sample injection modes: (a) injecting each sample manually with a syringe directly into the carrier gas stream, (b) injecting each sample manually into a Picarro Small Sample Introduction Module (SSIM) which automatically passes the samples on into the carrier gas stream, or (c) equipping the SSIM with a manifold to completely automate the injection of multiple samples (see Fig. 5.5). For all three discrete injection modes, the basic requirement is the supply of the Picarro G2201-*i* with a continuous carrier gas stream of either dry zero air (i.e. air with <1 ppm CO2 and CH4 and <10 ppm H2O) or dry standard air with ambient trace gas concentrations (Dickinson et al. 2017) at a pressure of ≥3 psi (0.2 bar) and <8 psi (~0.5 bar). The slight overpressure at the sample inlet ensures that the analyser can draw in the required ~25 ml min−<sup>1</sup> into its optical cavity.

An example of a custom-made direct manual sample injection setup has been described by Dickinson et al. (2017). It was tested with different Picarro gas analysers, but the principle of operation can also be transferred to other optical gas analysers. In general, the preferred tubing material for the carrier gas stream line is stainless steel due to its gas tightness and chemical inertness, but PTFE and FEP tubing can also be used, with FEP being the least expensive material. In this case, FEP tubing was equipped with two Luer lock three-way valves. The valve closest to the analyser is the designated sample injection port. Regardless of the sample injection mode, the tubing between the sample injection port and the analyser's sample inlet should be minimized to decrease potential dead volumes, mixing and lag times between sample injection and the actual measurement inside the analyser. The other valve is used for controlling the carrier gas stream. Samples were delivered to the carrier

**Fig. 5.5** Comparison of different discrete sample injection modes: *left* direct manual injection (figure modified after Dickinson et al. 2017), *center* injection via Small Sample Introduction Module (SSIM) and *right* automated injection of multiple samples via a 16-port-manifold. The manifold allows to attach up to 8 sample bags (every second inlet is capped off for purging cycles)

gas stream with a gastight syringe equipped with a push-button valve and Luer lock fitting. Once the syringe is connected to the sample injection port, the three-way valve is closed manually to stop the carrier gas stream. Then the syringe valve is opened, and the gas sample drawn steadily into the analyser without any further assistance since the analyser operates with a vacuum inside the optical cavity. After complete evacuation of the syringe, the carrier gas stream is manually switched back on. A sample volume of 50 ml was estimated to be the minimum volume for reliable operation of this injection mode. Larger volumes can be administered as well but would require some adjustments in the data processing routine which are described in Dickinson et al. (2017). This custom-made injection setup requires a well-cleaned and silicone-lubricated glass syringe to ensure smooth sample evacuation. Any friction between the plunger and the syringe wall can cause the plunger to jump during sample evacuation which can create small pressure fluctuations inside the analyser's optical cavity and thus increase measurement noise.

Injecting discrete gas samples via the SSIM avoids the problem of possible pressure variations. The SSIM is composed of a 20 ml sample chamber with a sample injection port, a solenoid valve system, an internal pressure sensor and an external vacuum pump. It is inserted into the carrier gas stream line as close as possible to the analyser's sample inlet. The solenoid valve system can shut off the sample cell from the carrier gas stream, but the carrier gas stream into the analyser is never interrupted. At the beginning of a discrete sample measurement, the SSIM is purged with carrier gas and subsequently evacuated several times to remove residues of previous samples. The last pre-measurement step is always a sample chamber evacuation. Injections of discrete sample volumes between 20 and 25 ml can be performed actively with a syringe, or by attaching larger sample containers (e.g. Teflon bags) to the SSIM sample injection port from which samples can be drawn in by the sample chamber vacuum. Once sample injection into the SSIM sample chamber is complete, the sample injection port is closed and the solenoid valve connecting the sample chamber with the carrier gas stream is opened to release the sample into the carrier gas stream. As the pressure in the sample chamber decreases, the outlet valve of the analyser slowly closes to maintain a constant pressure in the cavity. In this way, the flow rate through the cavity is reduced to ensure that the residence time of the sample in the cavity is maximized for a longer peak integration. It is also possible to inject less than 20 ml of gas into the SSIM sample chamber. In that case, the remaining sample chamber volume is filled up with carrier gas prior to releasing the sample to the analyser, resulting in sample dilution. This process is later in this case study referred to as automatic zero air dilution. When working in the dilution mode, one has to make sure that the CO2 and/or CH4 concentrations are within the dynamic range of the analyser. The standard specification range for the G2201-*i* is 380–2000 ppm for CO2 and 1.8–1000 ppm for CH4 (unless the analyser was upgraded for low and/or high CO2 concentration measurements).

To automate discrete sample injection, the SSIM can be equipped with a 16-port manifold. The 16-port manifold is a rotary valve that allows to attach up to 8 discrete samples, while the other 8 ports must be closed off so that they can be used for purging between measurements to reduce memory effects due to sample carry-overs. Also, when attaching the manifold to the SSIM, it is important to keep the tubing as short as possible to reduce dead volume effects which might degrade the accuracy for concentration measurements. The sequence of sample and standard measurements can be defined in the SSIM control software (referred to as 'Coordinator' software) also allowing for repeated sample injection. After defining the measurement sequence and other measurement parameters, the SSIM performs a purge and pump step and then asks the user to attach all sample containers to the manifold keeping the valves to the container closed. The Coordinator software then starts another purge and pump cycle for all selected input ports, and once completed, the user has to manually open all sample container valves before the actual unattended sample measurement starts. The duration of a single measurement takes 12 or 8 min in the standard or fast measurement mode, respectively (Fig. 5.5).

#### *Effect of carrier gas and sample volume on the measurement result*

The Picarro G2201-*i* continuously measures gas concentrations and δ13C in its optical cavity at a rate of approximately three to five seconds depending on whether one focuses only on the stable carbon isotopes of one gas species or the simultaneous quantification of both δ13C-CO2 and δ13C-CH4. The carrier gas stream provides a baseline measurement on which discrete sample measurements are superimposed as peaks. Choosing dry standard air instead of zero air as carrier gas is foremost an economical choice, but it has important implications for the data processing. Adding zero air to a discrete gas sample dilutes it resulting in lower CO2 and CH4 concentrations. However, the isotopic signature of the sample remains unchanged since no additional 12C and 13C are added to the sample. There are three possible reasons for diluting discrete gas samples: (a) it was not possible to collect sufficiently large gas sample volumes during an experiment, (b) CO2 and/or CH4 concentrations in the samples are too high and exceed the analyser's operational range, or (c) the samples contain contaminants (e.g. ammonia, hydrogen sulphide, ethylene) in high enough concentrations to negatively affect the spectroscopic measurements. When the Picarro G2201-*i* is coupled with the SSIM, the analyser's software automatically calculates the isotopic signature for each discrete gas sample regardless of whether it has been diluted with zero air in the SSIM or not, and no further data processing by the user is required. However, there is a limit to sample dilution. If the absolute concentrations of 12C and 13C in the sample approach the lower end of the analyser's operational range, instrument noise can increase to a degree that a reliable estimate of the isotopic signature becomes impossible.

Under certain conditions, the Picarro G2201-*i* is also able to provide reliable CO2 and CH4 concentration measurements (Picarro Application Note AN038). As previously mentioned, zero air addition to a discrete gas sample dilutes its CO2 and CH4 concentrations and ultimately biases the measurement result towards the carrier gas. Biases can also be introduced by sample carry-overs. Sample carry-overs are significantly reduced in the SSIM by the purging and evacuation cycles taking place immediately prior to sample injection. To physically limit zero air dilution of the sample, it is important to minimize the path length between the SSIM injection port and the sample, e.g. by using a septum port, and to inject at least 20 ml of gas into the SSIM sample chamber. If less sample is injected, a bias in the concentration data is unavoidable and has to be corrected by post-processing the data.

Dickinson et al. (2017) have developed data post-processing routines to correct for biases in the analyser's output data, both for the concentration and the δ13C values. Post-processing of the δ13C-CO2 and δ13C-CH4 is required when dry standard air is used as carrier gas since it already contains CO2 and CH4 at ambient levels with a specific stable carbon isotopic signature. Another important point to consider when opting for dry air as carrier gas is that its CO2 and CH4 concentrations and their isotopic signatures have to be significantly different from the expected values in the discrete gas samples because otherwise no unambiguous peak identification in the analyser's continuous data output can be achieved.

#### *Sample processing time*

The processing time for a single discrete gas sample depends on the injection mode, the sample volume and required analyser precision. For the direct injection of 50 ml samples into the carrier gas stream, the sample-to-sample time is ~5 min. Half of this time is needed for complete sample evacuation from the syringe. If one likes to use this injection mode with sample volumes >50 ml, sample-to-sample time will increase accordingly (Dickinson et al. 2017). For the combination of the Picarro G2201-*i* with the SSIM, the user has the possibility to choose between a standard and a fast measurement mode, which have a sample-to-sample time of 15 and 10 min, respectively. This includes 3 min for the purge and evacuation cycle to clean the SSIM and the analyser between samples. The actual measurement times for the δ13C and concentration values are 4 min and 9 min for the fast and standard mode, respectively. The analyser's software reports the discrete sample measurement results as averages with their respective standard deviations for these time intervals. Longer measurement time per sample increases measurement precision, but significantly reduces the amount of processible samples per day. The sample-to-sample time is independent of the gas sample volume injected into the SSIM sample cell since the SSIM adds carrier gas to the sample cell if less than 20 ml have been injected. As a result, the SSIM always releases the same gas sample volume into the carrier gas stream. However, if a user is not accustomed yet to manually injecting discrete samples into the SSIM sample cell or if there is a problem during the injection, significant time can be lost during this step and sample-to-sample times can increase up to 20 min. Daily sample throughput can be maximized by automating the entire sample injection process, i.e. equipping the SSIM with a programmable manifold. This does not decrease sample-to-sample times, but samples can also be processed during laboratory off-hours since no personnel has to be present for the sample injection process. For the calculation of the total number of processible samples per day, one does not only have to take into account the number of discrete gas samples obtained from an experiment, but also allocate time for instrument warm-up periods and instrument maintenance as well as the regular analysis of gas standards.

#### *A practical example of discrete gas sample measurements with a Picarro G2201***-***i*

This example demonstrates step-by-step workflow for discrete gas sample measurements with a Picarro G2201-*i* equipped with an SSIM. The focus of this example is solely on the physical handling of the gas samples and the analyser setup. Post-processing of the measurement data to account for biases will not be discussed but we refer to the considerations described by Dickinson et al. (2017). The gas samples were from a field study using chambers to measure soil CO2 and CH4 fluxes. 45 ml gas samples were taken from the chambers and stored in 20 ml screw-capped glass vials which were sealed with pierceable grey chlorobutyl rubber septa. The overpressure serves as protection against sample contamination during transport in case of minor leakages, and to ensure the extraction of ~20 ml of gas from the vials for injection into the SSIM to minimize sample dilution.

Figure 5.6 shows an example of how a sample measurement plan for the analysis of discrete gas samples with a Picarro G2201-*i* can look like. The first row identifies the experiment and the day on which the gas samples were obtained. The second row lists the names of the files in which the measurement data is stored on the analyser. As soon as the analyser is operational after being switched on, it automatically starts recording all data it collects regardless of whether the user will later use the data for analysis or not. These continuous data are stored in dat files at a sampling rate of approximately three to five seconds depending on the chosen measurement mode. In this example, the dat file name includes the date and the instrument time at which it was generated. To operate the SSIM, the CRDS coordinator software has to be installed on the analyser and launched, as well as a software monitoring the pressure inside the SSIM sample cell. The coordinator software automatically detects the peaks and integrates the signals to provide average δ13C and gas concentration values and their uncertainties for each discrete sample injected via the SSIM. The discrete data are stored in csv files, and the file names also include the date and the instrument time at which it was generated as well as the abbreviation SSIM. When using the concentration data, e.g. for the calculation of soil chamber fluxes, one has to make sure to use the dry concentration values, not the uncorrected wet concentration values as this could lead to flux underestimation.


**Fig. 5.6** Example of a measurement plan for a Picarro G2201-*i* equipped with an SSIM

The measurement plan in Fig. 5.6 contains columns for sample descriptions (vial and sample id), a running number which is taken from the csv file (SSIM Run Num), the injected sample volume (ml injected), the pressure inside the SSIM sample cell after sample injection (SSIM pressure injec.) and after automatic zero air dilution (SSIM pressure dilut.), the time at which each sample was injected into the SSIM (injection time), and comments. The measurement schedule consists of three different phases–a start-up phase, a leak testing and calibration phase, and the actual measurement phase. The start-up phase includes all the necessary steps to get the analyser ready for discrete sample injection. If the analyser is not running yet, but still has to be switched on, it needs at least 45 min to warm up and achieve the necessary vacuum in its optical cavity, after which it should run for about two more hours just measuring room air for laser stabilization. This is to ensure the best possible analyser performance. The next step is to connect the carrier gas stream to the SSIM and the SSIM to the analyser (Fig. 5.7). Before conducting any discrete sample measurements, the analyser should be purged with the carrier gas for about an hour to ensure the absence of memory effects. A well-established carrier gas reading is also essential for any required data post-processing due to memory effects and data biases (Dickinson et al. 2017). These preparatory steps can also be conducted the day before the actual discrete measurements. However, if the carrier gas is zero air, it

**Fig. 5.7** A Picarro G2201-*i* (large box) equipped with an SSIM (small box to the right) during measurements. A syringe containing a discrete sample can be seen attached to the SSIM sample injection port

is not recommended to have the analyser measure the gas stream for several hours as the uninterrupted absence of CO2 and CH4 can cause the analyser to drift and offset the analyser's calibration.

Once the analyser is ready for the injection of discrete gas samples, it is very important to perform leak tests and any calibration checks deemed necessary before measuring unknown samples. The most common location for leakages is the sample injection port. In this example, the SSIM injection port was equipped with a so-called septum injector nut with ¼" GC septum (VICI International, Schenkon, Switzerland). The septum used was a 6-mm-thick EC grade, high-temperature silicone septum (Trajan Scientific Europe Ltd, Milton Keynes, United Kingdom). A septum injector nut provides the shortest possible injection pathway from the sample to the SSIM sample cell. It is screwed onto the SSIM injection port, but it is very important not to overtighten the nut. Nut overtightening can significantly compress the septum, thus reducing the septum's flexibility and resealing capability, ultimately resulting in leakage. Additionally, a tightly compressed septum is more difficult to pierce with a needle.

To check the overall system's gas tightness, one should monitor the water concentration and the SSIM pressure. When the dry carrier gas stream is connected to the analyser, the water concentration should drastically drop below 10 ppm and then very slowly continue to decrease throughout the measurement day. The slow continuous decrease is due to the fact that it takes a significant amount of time to purge an analyser completely of all moisture as water vapour attached to the tubing walls will only slowly detach and enter the carrier gas stream. To specifically check the gas tightness of the SSIM, one should first perform a discrete sample injection without actually injecting a sample (Fig. 5.6, SSIM Run Num 1). The normal injection routine for the SSIM consists of six steps: (1) The SSIM coordinator prompts the user to attach a syringe with a vacuum-proof valve to the sample injection port with the syringe valve closed. (2) The sample cell including the tubing to the injection port (and thus the syringe needle) is purged with carrier gas and subsequently evacuated to remove any gas residues. (3) The user is prompted to open the syringe valve, and the sample is sucked into the sample chamber by the vacuum as indicated by a slowly inward moving plunger. As the vacuum in the sample cell is diminished by the inflowing sample, the plunger movement will slow down, and the user has to press in the plunger completely to finish the sample injection. Once the sample has been injected, the syringe can be removed. (4) The sample cell is filled up with carrier gas if necessary (=automatic zero air dilution). (5) The gas in the sample cell is released into the carrier gas stream.( 6) The SSIM is purged with carrier gas and evacuated several times to prepare it for the next sample. When no sample is injected during step 3, the pressure in the sample cell should remain stable at ~0 Torr (Fig. 5.6, SSIM pressure injec.). If the injection port is not gastight, the pressure in the sample cell will slowly increase. This test should be regularly repeated since the septum wears down with time and then has to be changed. If no sample is injected, the sample cell is filled with ~20 ml of carrier gas in step 4. The total amount depends on the actual pressure of the carrier gas stream which is depicted by "SSIM pressure dilute" in Fig. 5.5. The user's manual reading of this value from the analyser screen will vary a bit because this is only a short pressure spike in the sample cell before the sample is released to the analyser. However, the exact value can be retrieved later from the dat file if necessary. Apart from the stable vacuum during step 3, the discrete carrier gas sample measurement should have a water vapour concentration only slightly above the reading of the continuous carrier gas stream if the injection port is gastight.

Apart from leakage checks, the SSIM pressure can be used to estimate the volume for a gas sample if the user does not know it. The relationship between the gas volume and the pressure in the SSIM sample cell is theoretically perfectly linear. The first six measurements in Fig. 5.5 were all injections of known gas sample volumes which were used to construct a simple linear regression model. The resulting equation was

## SSIM pressure injec. [*T orr*] = 45*.*576*T orr/m L* ∗ *V olumeinjected* [*m L*] + 15*.*130*T orr*

with R2 = 1.000. If the user is not able to establish this linear relationship with an R<sup>2</sup> value of 1, this is also indicative of leakages in the system. This equation was used to later calculate the gas sample volumes for the unknown samples (Fig. 5.6, SSIM Run Num 7–16). These samples were extracted from the 20-ml screw-capped glass vials. The syringe used was a 25-ml 1025 SL Hamilton SampleLock Syringe for 22 gauge needles. To retrieve a discrete sample from a glass vial, the syringe needle was inserted into the vial and the syringe plunger pulled out all the way to the 25 ml mark. The plunger was held in this position for about five seconds to assure that a pressure equilibrium was reached between the vial and the syringe as the air needed some time to flow through the needle. Then the syringe valve was closed, and the plunger could be released. When transferring the sample from the vial to the syringe, the sample gas will equilibrate between these two volumes and only a part of the actual sample gas can be used for analysis, the remainder stays behind in the vial. Thus, as mentioned before, it is mandatory to have an overpressure in the vial of at least the volume which one wishes to inject into the analyser. After the sample was secured in the syringe, the plunger was gently pressed into the syringe until it met a slight resistance and then released again. The mark at which the plunger finally stopped was an indicator of the sample volume retrieved from the vial, albeit a very rough one. This value was noted down in the "ml injected" column; however, the precise SSIM pressure reading during sample injection was later used to derive a better sample volume estimate as described above.

At the end, a small remark regarding the syringe needle. We recommend the usage of side port needles to prevent septum coring. Septum coring means that the needle detaches a piece of silicone from the septum while piercing it. Usually, this silicone piece gets stuck inside the needle completely blocking it. A sure sign for coring is that the syringe plunger does not start to move when the syringe is subjected to the SSIM sample cell vacuum. However, needle blockage also means that no sample can be lost when coring occurs. One can simply close the syringe valve, remove the syringe, clean the needle, and repeat the injection. Needles which are open at the tip are much more prone to coring than side port needles. Side port needles are more expensive than other needle types; however, in the long run, frequent septum coring requires the septum in the injector nut to be changed more often and one also loses valuable measurement time.

## **References**


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 6 Methane Production in Ruminant Animals**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** Agriculture is a significant source of GHGs globally and ruminant livestock animals are one of the largest contributors to these emissions, responsible for an estimated 14% of GHGs (CH4 and N2O combined) worldwide. A large portion of GHG fluxes from agricultural activities is related to CH4 emissions from ruminants.

M. Zaman (B) · L. Heng

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang

Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

T. Clough Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

© The Author(s) 2021

M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_6

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

Both direct and indirect methods are available. Direct methods include enclosure techniques, artificial (e.g. SF6) or natural (e.g. CO2) tracer techniques, and micrometeorological methods using open-path lasers. Under the indirect methods, emission mechanisms are understood, where the CH4 emission potential is estimated based on the substrate characteristics and the digestibility (i.e. from volatile fatty acids). These approximate methods are useful if no direct measurement is possible. The different systems used to quantify these emission potentials are presented in this chapter. Also, CH4 from animal waste (slurry, urine, dung) is an important source: methods pertaining to measuring GHG potential from these sources are included.

**Keywords** GHGs · Animals · Direct and indirect emission · SF6 · CH4

## **6.1 Introduction**

Agriculture contributes more than 25% to the total greenhouse gases (GHGs) globally, and ruminant livestock animals are one of the largest contributors to these emissions, responsible for an estimated 14% of GHGs worldwide (Tubiello et al. 2014). In non-industrialised countries, emissions from livestock (methane and nitrous oxide

K. Dawar

W. X. Ding Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

C.-M. Görres Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

M. E. G. Hofmann Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

combined) may be the most important source of GHGs. However, measurement– though critical for meeting international obligations and for assessing mitigation options–is not commonly undertaken in most countries.

Many methods are available for estimating enteric methane (CH4) production in ruminant animals. They can be broadly classified into two groups–direct and indirect. The important distinction is that direct methods measure CH4 produced by the ruminant animals in some manner, whereas indirect methods infer CH4 production from parameters that are associated with CH4 production in the ruminant animals. In all cases, the methods have strengths and weaknesses and need to be selected with care for the particular objective in mind. The choice of method will depend on available financial and technical resources and the purpose of the measurement, including whether interactions between ruminant animals and environment are important to the research question. If it is impractical to use any of the direct methods, then less precise indirect methods can be used. This chapter will include examples of both direct and indirect measurement methods. Direct methods include enclosure techniques, artificial (e.g. SF6) or tracer techniques based on herd-level experiments in natural ventilated cattle housings (e.g. using CO2 as internal tracer), and micro-meteorological methods using open-path lasers. Under the indirect methods, emission mechanisms are understood, where the CH4 emission potential is estimated based on the substrate characteristics and the digestibility (i.e. from volatile fatty acids). These approximate

D. Lewicka-Szczebak

Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold

Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

M. Šimek

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

M. dos Reis Martins · S. Urquiaga EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

R. Well

Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany

methods are useful if no direct measurement is possible. This chapter provides an overview of the different systems used to quantify these emission potentials are presented. Also animal wastes are an important source of CH4 from animal waste (slurry, urine, dung) is an important source; thus, methods pertaining to measuring GHG potential from these sources are also included.

## **6.2 Direct Measurements**

## *6.2.1 Enclosure Techniques*

All enclosure methods rely on the principle of measuring either continuously or intermittently, the concentration of CH4 in and the total flow of air from around the animal. Methods vary in technical complexity, ease of operation, and precision.

#### **6.2.1.1 Total Enclosure of Animal**

The technique of open-circuit indirect respiration calorimetry has been routinely used with many species of ruminant animals to determine partition of dietary energy. This involves the measurement of oxygen (O2) consumption and carbon dioxide (CO2) production. For ruminants, the emphasis is quite different–determination of total CH4 production, arising largely from rumen fermentation to provide more precise estimates of relationships between dietary energy intake and CH4 production.

Models to estimate national and global CH4 emission from sheep and cattle at the farm level are mostly based on data of indirect calorimetric measurements, which most precisely measure the relationship between enteric CH4 and feed intake (Johnson and Ward 1996). Respiration chambers are used to measure CH4 at an individual animal level–their use is technically demanding, and the number of animal measurements possible will be determined by the availability of physical infrastructure (number of chambers). However, these systems provide the most precise measurements on enteric CH4 production.

Although design of the chambers varies, the basic principle remains the same. Chambers are constructed to house the subject animals, which are then sealed off and their environment is controlled (Plate 6.1). All open-circuit chambers are characterised by an inlet and exhaust, so animals breathe in a one-way stream of air passing

S. Zaman

J. Zhang

K. Lenhart

University of Canterbury, Christchurch, New Zealand

School of Geography, Nanjing Normal University, Nanjing, China

Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

through the chamber space. Air can be pulled through each chamber by exhaust fans, generating a negative pressure within the chamber, which minimises loss of air from the chamber. However, in practice, CH4 can still be lost from chambers (down the concentration gradient) that are imperfectly sealed, so, Gas recovery is not a maintenance task. Thresholds for chamber temperature (<27 °C), relative humidity (<90%), CO2 concentration (<0.5%) and ventilation rate (250–260 l min−1) have been described (Pinares-Patiño et al. 2011). Although these parameters may be varied, it is very important to ensure that subject animals remain in their thermo-neutral zone while being measured, or voluntary intake is likely to be compromised. In practice, chambers in tropical and sub-tropical regions will need to be fitted with air conditioning units, which will also provide a degree of dehumidification and a ventilation system. This ensures that chambers can be maintained at a constant temperature or at near ambient temperatures so as to capture the normal diurnal pattern of CH4 production (Tomkins et al. 2011). Feed bins and automatic water systems may also be fitted with electronic scales and meters, respectively, to monitor feed and water intake during CH4 measurement periods.

CO2 and CH4 concentrations are measured by sampling incoming and outgoing air, using infrared laser gas analysers, infrared photoacoustic monitors or gas chromatography systems. The other essential measurement is airflow over the measurement period, which is generally either a 24 or 48 h period. The accuracy and thus the validity of measurements are dependent on the sensitivity of the gas analysers used and routine calibration of these devices. Direct calibration of chambers is performed by releasing a known volume and concentration of standard gas to estimate recovery values (Klein and Wright 2006). Alternatively, a gas is released at a known rate until the concentration inside the chamber has equilibrated. Measurements are also influenced by the environmental conditions including temperature, humidity, atmospheric pressure and incoming air composition, and these parameters need to be measured and recorded as part of the measurement process. The calibration of the gas analysers must be accurate and reproducible for long-term usage.

Apart from technical complexity, an important limitation of the technique is that normal animal behaviour and movement is restricted when animals are kept in the respiration chambers (Table 6.1). Animals to be measured will benefit from acclimation in chambers prior to confinement and measurement to minimise alterations in their behaviour: such as decreased feed intake. Even so, there is clear evidence that this will happen in a small proportion of animals regardless of training, which should be borne in mind when interpreting data (Robinson et al. 2014). Using transparent construction material in chamber design allows animals to have visual contact with the other housed animals. The chamber should be sufficiently rigid to tolerate normal animal behaviour and if possible, a metabolism stall should be provided within the chamber to restrain the animal. Provisions should of course be made for feeding and watering the animal.

Substantial costs will be incurred with construction and maintenance of the opencircuit respiration chambers; the requirement for high-performance and sensitive gas analysers and flow meters need consideration in design and construction. Only a few

**Plate 6.1** Schematic diagram of open-circuit calorimeter to measure CH4 production in animals


**Table 6.1** Recommended dimensions (m) of chambers for different livestock classes

animals can be used for measurements within chambers at any one time. Nevertheless, respiration chambers are well suited to study the differences between treatments in response to mitigation strategies and continue to be regarded as the "gold standard" for measuring individual animal emissions.

Air conditioners comprising of a cooling coil and associated refrigerator are essential in hot regions; air filter and coil heater are optional.


#### **Measurement of Air Volume**

Several methods are available to determine the volume of air passing through the chamber. The simplest device (which requires no electrical supply) is a commercial dry gas meter, which gives a direct measure of total gas flow. Other alternatives include electronic turbine flowmeters and venturi apparatus. These give instantaneous rates of gas flow and are highly reliable. However, the need to integrate the individual airflow rates and the relative sophistication of such apparatus precludes their use in laboratories without adequate and reliable electrical supply. It is necessary to correct the total flow of gas to Standard Temperature and Pressure (STP). This requires a determination of atmospheric pressure and temperature.

#### **Measurement of CH**<sup>4</sup> **from Chamber Exhaust**

The exhaust air needs to be analysed for CH4 concentration. Several possibilities exist for CH4 measurement. Most commonly, gas samples will be taken at regular (4–12 min) intervals from each chamber in turn, for ~2 min. (to allow for adequate flushing of the residual from the previous sample) using an automated switching device. The actual measurement time required will depend on the measurement apparatus itself (as outlined in the previous section). A possible alternative and simpler method is to acquire a small, representative subsample of the chamber gas over the entire run using a low flow rate pump (such as a peristaltic pump) and stored in gasimpervious (e.g. polyvinyl fluoride (PVF); Tedlar®) bags. It is important that the flow rate of the aliquot collection is constant, thereby providing an integrated representation of the entire collection. After the run has been completed, CH4 concentration may be measured by the methods outlined in Sect. 6.8.

#### **6.2.1.2 Head Box**

Without the use of a tracer that moves with CH4, it is essential to collect all the expired and eructated (belching) gas. Drawing a stream of air past the muzzle of the animal by enclosing the animal's head in a hood can do this. The method requires considerable training of the animals. The hood can be made sufficiently large so that it is much less restrictive–thus more easily accepted. Additionally, the animal can be fed and have access to water during the collection of gas samples. In both methods, the principle concern is to have a sufficient outflow of gas to ensure there is lower gas pressure in the hood and gas lines and that all the "leaks" are inward. However, there is still considerable scope for part of the sample to leak, up the concentration gradient and be "lost". Hence, performance of gas recoveries is essential in properly calibrating the method. With the accuracy of available analytical equipment, the dilution of CH4 by air drawn past the animal's head does not present a problem.

The ventilated hood system is a simplification of a whole animal respiration chamber, as it measures the gas exchange only from the head rather than the whole body. Modern ventilated hood systems, applied for CH4 measurements, have been used in Japan, Thailand (Suzuki et al. 2007, 2008), the USA (Place et al. 2011), Canada (Odongo et al. 2007), and Australia (Takahashi et al. 1999). Recently, Fernández et al. (2012) described a mobile open-circuit respiration system. The ventilated hood system used by Suzuki et al. (2007, 2008) consists of a head cage, the digestion trial pen, gas sampling and analysis, behaviour monitoring, and data acquisition system. Like the whole animal chambers, it is equipped with a digestion pen for feed intake and excreta output measurements. An airtight head cage is located in front of the digestion pen and is provided with a loose-fitting sleeve to position the animals' head. Head boxes are provided with blowers, to move the main air stream from the inlet to the exhaust. Flow meters correct the air volume for temperature, pressure and humidity. Air filters are set up to remove moisture and particles out of the gas samples, which are sent to the gas analysers (Suzuki et al. 2007). The mobile system of Fernández et al. (2012) contains a mask or a head hood connected to an open-circuit respiration system, which is placed on a mobile cart.

The ventilated hood system is a suitable method under some circumstances: especially where open-circuit chambers are not viable. A critical limitation of the hood system is that extensive training to allow the test animals to become accustomed to the hood apparatus is absolutely essential–thus, while it can be used to assess potential of feeds, it is not suitable for screening large numbers of animals, primarily because of the high degree of training required. A further consideration is that hoods capture only measurements of enteric methanogenesis, excluding the proportion in flatus.

#### **Design of Hoods**

The hood should be designed to provide sufficient feeding space and enough room for the animal to move its head in an unrestricted way. A wide variety of materials may be used to build a box that is reasonably airtight. The most common materials used are wood and metal. While they can be custom-built, it is also possible to use plastic or metal drums or pre-constructed packing crates. It is a major advantage to have a clear removable panel to provide access for feeding and for checking the animal. This clear panel helps to maintain normal animal behaviour, particularly if other animals are visible to the experimental animal during the period it is in the hood. The animal should be restrained while its head is in the hood and the design of the hood depends on the facility available for restraint. For example, if animals are held in metabolism cages where they cannot turn around, a canvas sleeve can be fitted around the neck and connected to the hood as shown in Plate 6.2. This allows the animal to stand, eat and lie down during the measurement period. It may also be necessary to restrain the animal within the hood by means of a halter or collar. Hoods may also be built around yokes or even head bales at the end of a working race. It is desirable to minimise the amount of air leakage around the neck and head: a sleeve is an effective means of achieving this. This can be tied around the neck using a drawstring. The sleeve can be constructed from any material, but canvas or heavy cotton is most suitable. The length of the sleeve should be enough to allow the animal to stand up, lie down and have unrestricted access to feed and water.

An example for sheep (arrangement similar to that shown in Plate 6.2) uses a box made of 9 mm plywood that has solid sides 0.9 m × 0.4 m top and bottom (0.4 m × 0.6 m). The front and back panels (0.9 m × 0.6 m) have windows 0.25 m × 0.25 m for the animal's head and 0.5 m × 0.5 m for feeding and observation. A removable clear plastic or Perspex panel 0.6 m × 0.6 m is required for the feeding/observation window. The dimensions of the box can be varied to accommodate standard feeders, a water trough and the layout of the animal housing. The length of the sleeve should be around 0.35 m and will taper from a diameter of 0.25 m at the point where it is attached to the hood to 0.15 m so as to fit over the animal's head and be secured around the neck. The dimensions of the hood should be increased for cattle (approximately 3 times larger than for sheep). This will vary considerably according to the size

**Plate 6.2** Gas sampling from hood: subsampling over 24 h for subsequent analysis

of the cattle and the type of feed (diet). The basic principle for the hood is that it should be sufficiently large for the animal's comfort while maintaining the ability to feed concentrate and roughage diets. Attention should also be given to minimise the places where gas can leak from the system

*Gas sampling*: The methods for collecting the gas sample, including gas lines, pumps and meters, are similar irrespective of whether hoods or chambers are used.

*Sampling and airflow*: Airflow of 50–70 l min−<sup>1</sup> is suggested for measuring CH4 production in sheep. This gives concentrations of between 100 and 500 ppm (v/v) of methane in the airflow from the hood, but this may need to be modified, depending on the range over which the measuring device exhibits a linear response. For example, for cattle a flow rate of anywhere from 9 to 21 l sec−<sup>1</sup> the actual flow rate will be determined by the concentration range deemed most suitable for the gas measurement apparatus used, which may be indicated, depending on the measurement device, live weight of the animal and level of feeding. These levels of gas flow will also be sufficient to provide the animal with fresh air and maintain CO2 levels below 1%. Gas flow rates are easily controlled by altering the speed of the pump or even by an in-line flow restrictor but must not be altered during a run. A single pump can be used to sample a number of animals providing each sampling line has independent control of flow rate and a separate airflow meter. A schematic diagram is shown in Plate 6.2.

*Sampling lines*: The gas sampling lines can be secured overhead. The diameter of the tubing should be between 1.5 and 2.0 mm ID for sheep and 2–5 mm ID for cattle and can be constructed of materials such as copper, PVC or flexible rubber or plastic hoses, although CH4 will leak from plastic or PVC tubing, and the use of these materials in a sampling line is better to avoid if possible.

**Plate 6.3** Gas sampling and continuous analysis of a subsampled air stream

*Filter*: It is advisable to have a simple filter system in the main sampling line in order to remove insects and feed particles that may enter the line. A jar containing plastic scourers or glass wool is adequate.

*Gas flow measurement*: The total volume of gas drawn past the animal must be measured accurately using a commercial dry gas meter.

#### **Analysis and Subsampling Systems**

*Continuous gas analysis*: It is possible to analyse the CH4 content of the gas from the animal by incorporating an infrared gas analyser in the main line, or in a stream of gas taken from the main line (see Plate 6.3). With this system, the output from the meter needs to be recorded continuously and integrated over time.

Calibration of this equipment requires relatively large quantities of gas, and this can present a potential source of error and an on-going cost. In addition, it requires relatively sophisticated recording and integration equipment or a data logger. Placing the infrared gas analyser in the main line is not recommended unless there is a specific need to measure changes over time.

*Time scale for measuring CH4 production*: The extrapolation of short-term measurements of CH4 production by subsampling over only part of the day can be very misleading. There is considerable variation in the rate of CH4 production during the day, and it can change considerably after feeding. Unless the measurement is for specific screening purposes, and the time of sampling is standardised with respect to feeding and other animal husbandry practices, measuring for periods of less than 24 h is not recommended.

*Calculation*: The production of CH4 is calculated by multiplying the concentration of CH4 in the subsample by the total volume of gas passing through the hood. For example, a daily airflow of 86,000 l past the head of a sheep with a CH4 concentration of 200 ppm (v/v) gives a daily production of 17.2 l day−<sup>1</sup> (0.77 mol day−1). In cattle, corresponding values might be a daily airflow past the animal of 720 000 l and a concentration of 250 ppm (v/v). In this instance, CH4 production would be 180 l day−<sup>1</sup> (8.04 mol day−1).

#### **Greenfeed**® **Emission Monitoring Apparatus**

Greenfeed® is a patented device (Zimmerman and Zimmerman, 2012) that measures and records short-term (3–6 min duration) CH4 emissions from individual cattle repeatedly over 24 h by attracting animals to the unit using a "bait" of pelleted concentrate. By being available 24 h/day, potential sampling bias is reduced, and the technique has been shown to provide comparable estimates to that produced by both respiratory chamber and SF6 techniques (Hammond et al. 2013). However, a significant limitation of the technique is the requirement to supply an "attractant" to lure the animal to use the facility, consisting of up to 1 kg of concentrate pellets per day. This will certainly affect DMP and may also alter VFA profiles or the overall digestibility of the diet consumed. Attempts to use energy-neutral attractants, such as water, have proven equivocal (J Velazco pers. comm.).

#### **6.2.1.3 Polytunnel**

The polytunnel is an alternative to respiration chambers although operation and measurements are somewhat simpler. Methane emissions from individuals or small groups of animals can be acquired under some degree of grazing. This allows test animals to express normal grazing behaviour, including diet selection over the forages within the polytunnel space. It has been used in the UK to measure CH4 emissions from ruminants under semi-normal grazing conditions. Murray et al. (2001) report CH4 emissions from sheep grazing of two ryegrass pastures and a clover/perennial ryegrass mix pasture using this methodology. Essentially, polytunnels consist of one large inflatable or tent-type tunnel made of heavy-duty polyethylene fitted with end walls and large diameter ports. Air is drawn through the internal space up to 1.0 m3s−<sup>1</sup> (Lockyer and Jarvis 1995). In general, polytunnels may be used where emissions from fresh forages are of interest because animals can be allowed to graze a confined area of known quality and quantity. When the amount of available forages is depleted, the tunnel is moved to a new patch.

Airflow rate can be measured at the same interval as the CH4 levels are assessed or can be continuously sampled at the exhaust port (Lockyer 1997). Micropumps may be used to pass the exhausted air to a dedicated gas analyser or a GC (Murray et al. 2001). Data from all sensors can be sent to a data logger, which records flow rate, humidity and temperature within the tunnel, as well as gas production from the livestock. Samples of the incoming and exhaust air can be taken as frequently as necessary, depending on the required accuracy. The samples can be either taken manually or by an automatic sampling and injection system.

The polytunnel system requires frequent calibration to assure a good recovery rate, which is performed by the same principle as the chamber technique. Methane measurements can be collected over extended periods of time. Fluctuations can be expected to occur due to changes in animal behaviour, position relative to the exhaust port, internal temperature and relative humidity and grazing pattern of the animal: eating, ruminating or resting (Lockyer and Jarvis 1995; Lockyer and Champion 2001). The polytunnel is suitable for measuring CH4 emissions under semi-normal grazing conditions. It has been reported that the tunnel method gives lower readings of CH4 concentration (15%), compared to the respiration chamber method, suggesting that animals actually consume less–this requires further investigation. Recovery rate is high in both systems, 95.5–97.9% versus 89.2–96.7% for tunnel and chambers, respectively (Murray et al. 1999). With an automated system, measurements can be performed with high repeatability. The polytunnel system is portable and can be used on a number of available pastures or to also browse shrubs. Its utility is limited by the inability to capture and record feed intake.

#### **6.2.1.4 Portable Accumulation Chambers (PAC)**

PACs consist of a clear polycarbonate box of approximately 0.8 m3, open at the bottom and sealed by achieving close contact with flexible rubber matting. Methane production is measured by the increase in concentration, which occurs over approximately a 1 h period. PACs were designed to screen large numbers of sheep to identify potentially low and high methane-emitting individuals and thus develop estimates of the genetic parameter in sheep populations. This technique initially showed good agreement with respiratory chamber measurements (Goopy et al. 2009, 2011) and subsequent investigations demonstrated such measurements to be moderately repeatable in the field and to have potential for genetic screening of the animals (Goopy et al. 2014). Longer term comparisons of PAC measurements and respiratory chamber data, however, suggest that these two methods may be measuring quite different traits and further investigations are required before committing significant resources to pursue measurements using PACs (Robinson et al. 2015a, b).

## **6.3 Tracer Techniques**

## *6.3.1 Use of SF6 Bolus*

The sulphur hexafluoride (SF6) technique is a direct measurement of the CH4 emission of individual animals. This technique can be performed on an individual animal under normal grazing conditions and can also be employed under more controlled conditions where intake is measured and/or regulated.

The SF6 principle relies on the insertion of a permeation tube with a predetermined release ratio of SF6 into the rumen, via the mouth (per os). Air from around the animal's muzzle and mouth is drawn continuously into an evacuated canister connected to a halter fitted with capillary tube around the neck. Johnson and Ward (1996) provide a detailed description of the methodology.

The regulation of the duration of collection of each sample is achieved by altering the length and/or diameter of the capillary tube (Johnson and Ward 1996). Several modifications have since been reported with specific applications (Goopy and Hegarty (2004); Grainger et al. 2010; Ramirez-Restrepo et al. 2010). Most recently, Deighton et al. (2014) have described the use of an orifice plate flow restrictor which considerably reduces the error associated with sample collection and thus should be considered in preference to the traditional capillary tube flow restrictors. At completion of sample collection, the canisters are pressurised with N2 prior to compositional analysis by gas chromatography. Enteric CH4 production is estimated by multiplying the CH4/SF6 ratio by the known permeation tube release rate corrected for actual duration of sample collection and background CH4 concentrations (Williams et al. 2011), which are determined by sampling upwind ambient air concentrations. Williams et al. (2011) emphasised the importance of a correct measurement and the reporting of the background concentrations, especially when the method is applied indoors. Methane is lighter (16 g mol−1) compared to SF6 (146 g mol−1); therefore, it will disperse and accumulate differently depending on ventilation, location of the animals and other building characteristics.

This method enables sampling of gas concentrations in exhaled air of individual animals, and it also takes into account the dilution factor related to air or head movement. A significant limitation of this method, however, is the high within- and between-animal variation. Grainger et al. (2007) reported a variation within animals between days of 6.1% and a variation among animals of 19.7%. Pinares-Patiño et al. (2011), monitoring sheep in respiration chambers simultaneously with the SF6 technique, reported a higher within (×2.5) and between (×2.9) animal variance compared to the chamber technique combined with a lower recovery rate (0.8 ± 0.15 vs. 0.9 ± 0.10). These sources of variation need to be taken into account in determining the number of repeated measures needed to obtain accurate results. Moate et al. (2015) have recently described the use of Michaelis–Menten kinetics to better predict the discharge rate of the SF6 capsules, which should potentially reduce the error associated with estimating discharge rates and also prolong the useful life of experimental subjects through the improved predictability of discharge rates over much longer intervals.

The SF6 technique gives animals the ability to move and graze normally on test pastures. This makes the method suitable for examining grazing management effects on CH4 emission (Pinares-Patiño et al. 2007), but it does so at a cost. The SF6 method is less precise, less physically robust (high equipment failures) and more labour-intensive than respiration chamber method (Plate 6.4).

The permeation tubes consist of a closed stainless-steel tube capped at one end with a Teflon disc held in place with a standard SwagelokTm nut. The tube is filled with SF6 while being cooled with liquid N2 causing it to reach solid state on contact. It is then allowed to equilibrate to a fixed temperature in a water bath. Permeation rates are determined gravimetrically through repeated weighing on a precision balance over a period of several weeks. Typical permeation rates are of the order of 0.5– 2.5 mg day−1, although they may be considerably higher if circumstances dictate. Permeation tubes should be inserted into the rumen a minimum of 3 days prior to the first scheduled collection to allow steady-state conditions to be reached. Insertion can be accomplished with a balling/drenching gun or other similar methods for inserting boluses into the rumen.

A leather pad attached to the noseband of a halter serves as an anchor point for the sample line near the animal's nose and mouth. A small piece of plastic tubing is attached to a filter and oriented such that it is placed over one of the nostrils (see Plate 6.5). A filter (10 µm) connected to the upstream end of the sample line protects the flow restrictor from becoming plugged. Fastening the tube to the sides of the halter helps to protect the capillary tubing and reduces animal irritation. The capillary tube-collection flask connection should be via a quick-connect fitting to simplify flask exchange. A soft rope fastened around the neck with a clasp that can be attached to the collection flask helps stabilise the flask and takes pressure off the capillary tube.

The collection vessel should be large enough to accommodate the size of the desired sample (i.e. ~5 l for cattle), should be able to withstand a vacuum and should have a valve for sealing the flask. Immediately prior to use, all air should be removed

**Plate 6.5** Position of the evacuated vessels on the animal

from the flask and the valve closed. After fastening the flask to the supporting neck rope and attaching the capillary tube, the time of day should be noted, and the valve opened. When the sampling time is complete, the flask is removed for analysis. If repeated collections are desired, another flask should be added after the first one is filled. It is recommended that many measurements are made on each animal, and that total 24 h emissions are also reported. The diameter and length of the capillary tubes needed depend on the rate at which sampling is desired. The size of the capillary tube bore should be such that the evacuated sample flask fills to about 1/2 atmospheric pressure over the desired sampling time. The flask pressure should be measured, and the flask then filled with N2 to bring it to positive pressure (around 1.5 atmospheres). Both pressures need to be recorded to enable the extent of dilution by N2 to be calculated.

The vessel samples can be quickly and accurately analysed for the tracer using an electron capture gas chromatograph. Methane can be analysed using a gas chromatograph with an FID detector (see Sect. 6.8).

The emission rate of CH4 (L hr−1) is calculated from

$$\text{QCH}\_4 = \text{QSF}\_6 \frac{[\text{CH}\_4]}{[\text{SF}\_6]} \tag{6.1}$$

where QSF6 is the release rate of SF6 in litre hr−1, [CH4] is the concentration of CH4 in the gas sample and [SF6] is the concentration of SF6 in ppm (v/v).

## *6.3.2 Tracer Ratio Method for Emission Measurements in Naturally Ventilated Housing*

An established approach to determine emissions at herd level in naturally ventilated housings is the use of artificial tracers with known source strength (Mohn et al. 2018; Ogink et al. 2013; Schrade et al. 2012). The most frequently used tracer gas for emission measurements in livestock husbandry is SF6, as it is chemically extremely stable and has a concentration in ambient air of only a few ppt (10−<sup>12</sup> mol mol−1). The most widely used dosing principle is constant SF6 dosing, which can be implemented for continuous or semi-continuous measurements. The quantification of emissions with the tracer ratio method is based on the assumption that the tracer gas (i.e. SF6) release sufficiently mimics the release of the target substances from surfaces (e.g. NH3) or animals (CH4, CO2). The mass flow of the target gas (mtarget, e.g. mCH4) is calculated from the ratio of the background-corrected target (ctarget) and tracer gas concentration (ctracer) and the mass flow of the tracer gas (mtracer) according to mtarget = mtracer x ctarget/ctracer. In contrast to CO2 and heat balance methods, tracer ratio methods are also applicable to naturally ventilated housings with large openings and areal sources like outdoor exercise areas and open slurry storages (Schrade et al. 2012), because the mass flow of the tracer can be adapted to the dilution ratio.

The uncertainty level of tracer gas methods is highly dependent on the accuracy and distribution of the tracer gas dosing and the number and positioning of the sampling points (Calvet et al. 2013). Homogenous dispersion can best be achieved by dosing a diluted tracer gas, with similar density to ambient air, in a dosing grid next to the emitting areas or objects. Representative, preferably homogenously mixed, air samples can be collected using an air sampling grid with critical orifices. Validation experiments of the tracer ratio method have demonstrated that the technique is suitable for both areal and point emission sources and can achieve an uncertainty of less than 10% for target gas mass emissions, which is superior to other currently available methods (Mohn et al. 2018; Schrade et al. 2018). The sensitivity of the technique is mainly given by the capability for background correction, i.e. to separate and subtract emissions in the close vicinity of the housing. This is especially important for calm nights where, depending on the topographic conditions, the target gas accumulates in the nocturnal boundary layer.

A specific variant of the tracer ratio method, using two (or more) different tracer gases (e.g. SF6, SF5CF3), enables the independent assessment of housing areas (Mohn et al. 2018; Schrade et al. 2018). In an experimental housing with identical, but spatially separated housing areas, this approach can be applied to quantify the reduction potential of abatement measures (e.g. flooring, feeding) under comparative measurement conditions (e.g. climate). Additionally, using two different tracers, cross-contaminations between experimental areas are readily apparent by enhanced concentrations of the tracer gas in the respective other section.

## *6.3.3 Application of CH4: CO2 Ratio*

Madsen et al. (2010) proposed using the ratio of CH4: CO2 in exhaled breath to assess enteric CH4 production in ruminants. The principle relies on knowledge of intake, energy content and heat increment of the feed ration consumed. Haque et al. (2014) applied this method using a fixed heat increment factor. Hellwing et al. (2013) has regressed open-circuit chamber measurements of Daily Methane Production (DMP) in cattle against estimates calculated using CH4: CO2 ratios and found them to be only moderately correlated (R2 = 0.4), suggesting that this method is unsuitable for precision measurements.

## **6.4 Micrometeorological Estimates**

## *6.4.1 Open-Path Lasers*

The use of open-path lasers combined with a micrometeorological dispersion method can now be used to measure enteric CH4 emissions from herds of animals and facilitates whole-farm CH4 measurements across a number of pastures.

The open-path laser method for whole-farm CH4 measurements is already in use in Canada (McGinn 2006; Flesch et al. 2005, 2007), Australia (Loh et al. 2008; McGinn et al. 2008; Denmead 2008; Tomkins et al. 2011), New Zealand (Laubach and Kelliher 2005) and China (Gao et al. 2010). Methane concentration measurements are performed using one or more tuneable infrared diode lasers mounted on a programmable and motorised scanning unit (Tomkins et al. 2011). The tuneable infrared laser diode beams to a retroreflector along a direct path, which reflects the beam back to a detector. The intensity of the received light is an indicator of the CH4 concentration (ppm) along the path. In an optimal situation, there should be at least one path for each predominant wind direction: one path upwind (background CH4) and multiple paths downwind (CH4 emission) of the herd. This method assumes that the herd acts as a surface source or, when individual animals can be fitted with GPS collars, individual animals are treated as point sources. Regardless of application, the CH4 concentration is calculated as the absorption ratio of the external absorption to internal reference-cell absorption, of the infrared laser beam as it travels along the path (Flesch et al. 2004, 2005). Continuous measurements are required for the CH4 concentration and the environmental measurements, the latter being recorded by a weather station, i.e. atmosphere temperature, pressure, and wind direction and speed (Loh et al. 2008, 2009). Data management can be achieved statistically by merging all data, including GPS coordinates of the field (paddock) or individual animals, from a number of averaging time periods using statistical software. After integrating, WindTrax software (Thunder Beach Scientific, Nanaimo, Canada) uses a backward Lagrangian Stochastic (bLS) model to simulate CH4 emission (grams day−<sup>1</sup> animal−1), by computing the line average CH4 concentrations with atmospheric dispersion conditions.

The data integrity of the open-path laser method will be highly dependent on environmental factors and the location of test animals. Flesch et al. (2007) described several criteria to determine data integrity using an open-path laser method. These criteria are based on wind turbulence statistics, laser light intensity, R2 of a linear regression between received and reference waveforms, surface roughness, atmospheric stability, and the source location (surface or point source). Invalid data can be generated by misalignment of the laser, unfavourable wind directions, surface roughness or periods in which the atmospheric conditions (rain, fog, heat waves, etc.) are unsuitable for applying the model (Freibauer 2000; Laubach and Kelliher 2005; Loh et al. 2008). To optimise the positioning of the equipment, all of these meteorological and physical aspects of the experimental site must be taken into account (Flesch et al. 2007; Loh et al. 2008, 2009). Moreover, the measurement area is restricted by the length of laser paths when using a surface source approach. It is especially important to define the herd location, as uneven distribution of the herd results in miscalculations of the CH4 concentration. Tomkins et al. (2011), comparing open-circuit respiration chambers with the open-path laser technique, reported estimated CH4 emissions using the bLS dispersion model of 29.7 ± 3.70 g kg−<sup>1</sup> dry matter intake (DMI) relative to 30.1 ± 2.19 g kg−<sup>1</sup> DMI measured using open-circuit respiration chambers.

The open-path laser method does not interfere with the normal grazing behaviour of the cattle and is non-invasive. Spatial variability is taken into account in these measurements, as the method can simulate gas fluxes over a large grazing area. Moreover, the tuneable diode laser is highly sensitive and has a fast response to changes in CH4 concentration with detection limits at a scale of parts per trillion (ppt) (McGinn 2006). The labour intensity is low, although the equipment requires continuous monitoring. This method is expensive, which reflects not only the requirement of sensitive and rapid-response instruments to analyse CH4 concentration, but also the requirement to capture micrometeorology data. Diurnal variations due to grazing and rumination pattern, pasture composition, and individual variation need to be considered in planning experimental protocols and are dependent on the purpose of the study to prevent over- or under-estimation of the total CH4 emission. Furthermore, DMI determination is not very accurate as these are based on predictive models using the relationship between live weight (LW) and LW gain, following assumptions of the ARC (1980).

## **6.5 Short-Term Measurements**

While most assessments of enteric methane emissions are focused on Daily Methane Production (DMP), or the derivative: Methane Yield (MY), there is an increasing impetus to estimate the emissions of large numbers of animals in their productive environment. This is driven both by the demand for data to establish genetic parameters for DMP and/or to verify mitigation strategies or GHG inventories. Here, this area will be discussed only briefly, as there is at present limited scope for the application of these technologies in the developing countries. For the interested reader, the area has been ably reviewed by Hegarty (2013).

## *6.5.1 Spot Sampling Using Lasers*

Spot measurements of CH4 in the air around cattle's mouths have been made using laser devices to provide short-term estimates of enteric CH4 flux (Chagunda et al. 2009; Garnsworthy et al. 2012). These estimates are then extrapolated to represent DMP. Chagunda and Yan (2011) have claimed correlations of 0.7 between laser and respiratory chamber measurements, but this claim is based on the laser apparatus measuring CH4 concentrations in the outflow of respiratory chambers, rather than from the animals themselves; thus, this "method" remains unproven under realistic field conditions.

## **6.6 Indirect Methods**

The procedures described below are "indirect" methods for approximating CH4 production and do not measure levels of CH4 per se that are produced by ruminant animals. Therefore, they should be used only if more precise measurements are logistically unachievable (e.g. for regional or country-wide surveys), or there is a need to fill gaps in a much larger survey, or obtain preliminary data before embarking on more extensive studies using more accurate direct techniques.

## *6.6.1 Methane Emissions from Feed and Feed Characteristics*

Enteric methane production (EMP) can be estimated from intake and diet quality (digestibility), and there are a number of extant algorithms that attempt to quantify this, although it has been demonstrated that estimates of emissions can vary by 35% or more for a particular diet (Tomkins et al. 2011). Diet quality (i.e. digestibility) can be inferred from the analysis of representative samples of the rations or pasture consumed, but where intake is not measured, estimation of EMP faces considerable challenges. Models which estimate intake based on diet quality or particular feed fractions assume ad libitum access, and in situations where animals are corralled without access to feed overnight (as is frequently the case in developing countries), the validity of this assumption is likely violated (Hendricksen and Minson 1980; Jamieson and Hodgson 1979). In such cases, intake can be inferred from energy requirements **(**Energy for LW flux; maintenance+lactation and pregnancy+locomotion) using published estimates such as the Nutrient Requirements of domestic ruminants (CSIRO 2007) to convert physical values to energy values and so infer intake of the estimated diet. However, the inherent variability in feed composition, intake patterns of fermentation and feed partitioning, makes precise prediction of CH4 production from feed characteristics problematic, though there are many examples in the literature of attempting to do so.

Blaxter and Clapperton's (1965) landmark review of experimental results from measuring CH4 generation in cattle and sheep indicates that both intake and digestibility determine the amounts of methane generated and do so in an interrelated manner. However, based on several hundred animal measurements over a range of diets, they proposed the following equation:

$$\text{CH}\_4\text{(kcal/100 kcal GE)} = 1.30 + 0.112D + L(2.37 - 0.05D) \qquad (6.2)$$

where

D = digestibility of gross energy, and L = level of intake relative to maintenance.

Moe and Tyrell (1979) and Holter and Young (1992) both developed more complex equations using carbohydrate fractions to predict CH4 generation in dairy cattle:

$$\mathrm{CH}\_{4}\mathrm{(MJ\ d^{-1})}=3.406+0.510\,\mathrm{s}\\
\text{double residue (kg fed)}+1.736\,\mathrm{hemiccell}\\
\mathrm{ulos}\,\left(\mathrm{kg fed}\right)$$

$$+2.648\,\mathrm{cellulose}\,\left(\mathrm{kg fed}\right)\tag{6.3}$$

The ability of each of these equations to accurately predict methane generation has been assessed by Wilkerson et al. (1995). Using data from 602 cattle, they concluded that all equations showed moderate to large errors of prediction, with Blaxter and Clapperton (1965) and Moe and Tyrell's (1979) equations having the least error. Contrary to the conclusions of Wilkerson et al. (1995), Benchaar et al. (1998), in reanalysing data from published studies, concluded that the equations of both Blaxter and Clapperton (1965) and Moe and Tyrell (1979) had coefficients of determination of less than 0.6, with high errors of prediction.

Recently, an alternative algorithm for Methane Production Rate (MPR-CH4 g d−1) for cattle consuming tropical forages has been proposed by Charmley et al. (2016) to address the lack of data for estimating EMP for ruminants grazing in tropical systems.

Estimations of CH4 production from feed characteristics alone should at best be considered as being preliminary calculations, to only be used prior to the start of proper measurements; however, a methodology based on IPCC Tier II enteric CH4 estimation promises more accurate estimates of DMP of ruminants under field conditions. The general approach, which was developed specifically for smallholder farming systems in Africa (Goopy et al. 2018), uses total metabolic energy requirements (MERtotal) of individual cattle on a seasonal basis calculated by summing the estimated MER for maintenance (MERM), LW gain or loss (MERG/L), lactation (MERL) and locomotion/traction(MERT). Intake can be inferred as a function of MERtotal and the weighted mean DM digestibility (DMD) of the seasonal feed baskets in the study area. DMI was used as the basis to estimate daily CH4 production rate (MPR).

## *6.6.2 Emissions from Volatile Fatty Acids (VFAs)*

Increasing interest in reducing emission of the important GHG methane has created a need for reliable indicators of daily methane production by ruminants without resorting to lengthy and intensive calorimetry studies. Feed intake explains the majority of differences seen in DMP. However, difficulty in assessing feed intake by grazing animals limits the application of this predictor and thus, a predictor of daily CH4 production, which could be measured in the field, was sought. Enteric methane production is a direct function of the net amount of H2 liberated through fermentation of feed, and there is an extensive literature which relates in vitro total fermentation to total gas production (e.g. Menke et al. 1979), and to CH4 production in particular (i.e. Demeyer and Van Nevel 1975). Stoichiometric relationships developed in vitro typically explain more than 95% of observed H2 present in VFAs. Thus, it has been considered that measurement of VFA levels in one or more rumen fluid samples may give a useful prediction of CH4 production on a given day. Isotopic studies (Leng 1970; Sharp et al. 1982; Sutton et al. 2003) relating VFA concentration to VFA production indicated that a moderately strong relationship exists, suggesting that spot measures of VFA concentration (as a proxy for daily VFAs production) may provide a useful way of estimating daily CH4 production in the field.

However, one study comparing multiple samples of VFA taken over the day, with simultaneous enteric methane production measurements, found that the average concentration of individual or total VFAs explained less than 25% of the variance in daily CH4 production (Robinson et al. 2010) and is, thus, a poor predictor of methane production of sheep, which have typical morning and afternoon feeding periods. Sharp et al. (1982) also found poor-to-moderate correlations between CH4 production rate and concentrations of acetate (r = − 0.1), propionate (r = 0.63), and butyrate (r = 0.92). Sutton et al. (2003) found similar relationships to Sharp et al. (1982) between VFA production rate and VFA concentration for cattle fed a mixed diet twice daily.

The disparate response of methane production and VFA concentration to feeding may result from changes in VFA absorbance and rumen volume with feeding. Variable absorption of VFAs due to differences in factors such as pH and osmolarity that change with feeding is known (Dijkstra et al. 1993). Additionally, rumen volume would itself directly affect VFA absorption by means of changing absorptive surface area available. Increases in rumen volume with feeding level are also known (Purser and Moir 1966) and could also explain how fermentation and methane production could increase without causing a proportional increase in VFA concentration, by simply increasing the volume of rumen water in which the VFAs were dissolved. Rumen volume has been observed to increase by 20–34% in the first hour after feeding (Stewart et al. 1958). Such an increase would serve, not only to proportionally dilute VFA concentrations, but also to increase rumen surface area by an estimated 13–21%, thereby considerably enhancing the opportunity for diffusion of VFA across the rumen wall. Thus, using spot sampled concentrations of VFA in the rumen cannot be recommended as a method for estimating DMP in ruminants.

## *6.6.3 In Vitro Incubations*

The amount of gas released from the fermentation process and from the buffering of VFAs is related to the kinetics of the fermentation of a known amount of feedstuff (Dijkstra et al. 2005). Several systems for measuring in vitro gas production have been developed which vary considerably in complexity and sophistication. Menke et al. (1979) described a manual method using gastight syringes, which involves constant registering of the gas volume produced. More recently others have described a system using pressure transducers (Pell and Schofield 1993; Theodorou et al. 1994; Cone et al. 1996). Variants now available as proprietary systems (RF, ANKOM Technology) use radio frequency pressure sensor modules that communicate with a computer interface and dedicated software which records gas pressure values.

The basic principle of the in vitro technique relies on the incubation of rumen inoculum with a feed substrate under an anaerobic environment in gastight culture bottles. Gas accumulates throughout the fermentation process, and a cumulative volume is recorded with gas volume curves being generated over time. To estimate kinetic parameters of total gas production, gas production values are corrected for the amount of gas produced in a blank incubation, and these values can be fitted with time using a non-linear curve fitting procedure in GenStat (Payne et al. 2011) or other suitable software. Headspace gas samples are taken to analyse the gas compositions and determine actual CH4 concentrations, typically by gas chromatography, although a "quick and dirty" alternative has been developed–specifically a strongly basic solution such as NaOH is introduced into the vessel, which will subsequently cause the CO2 to enter solution. The remaining gas is assumed to be CH4.

Gas production is only one of the outputs of microbial fermentation, and the quality of the information derived can be improved by also considering substrate disappearance with concomitant production of VFAs (Blümmel et al. 2005).

There are two main forms of artificial rumen. In one form, rumen contents, freshly removed from a donor animal, are incubated in vitro in batch culture, while in the other form continuous culture is established. The former type of culture is valid for a period of hours, whereas the latter culture, which represents an anaerobic system not identical to the rumen, can be sustained for days or weeks. A procedure for batchtype culture is described below, while the rumen simulation technique (RUSITEC) (Czerkawski and Breckenridge 1977) is a proven system used widely throughout the world.

Neither type of in vitro system is very good for predicting methane production in vivo. The batch culture system generally lacks pH control, and the stoichiometry of product formation cannot be guaranteed to be the same as that occurring in the donor animal. For example, lactic acid is often detected in short-term in vitro incubations but not in the animal. In the RUSITEC system, the pattern of VFAs produced is usually similar, though not identical, to the in vivo situation.

Nevertheless, in vitro systems have some value for comparative purposes, e.g. for measuring the effects of additives on factors that control fermentation. The continuous culture is more valuable than the batch system, because it can take into account any adaptation the microbial population makes in response to the additive. This is particularly important for additives that affect CH4 production. Some types of material, such as chlorinated hydrocarbons, are highly effective in the short term, but the rumen microbial population adapts eventually to become insensitive to the inhibitor. Microbial additives, on the other hand, require adaptation of the population to detect an effect, a situation that does not occur in the batch system.

## *6.6.4 Batch Systems*

A variety of types of apparatus can be used. These differ in their cost and ease of use and range from a simple conical flask to a pH and Eh-controlled fermenter. The following criteria are essential to all of the systems:

Rumen contents must be removed from the rumen just before use; anaerobic conditions must be maintained during both transfer of the sample to the incubation vessel and during the incubation. The fermentation liquid must be incubated at 39 °C and agitated sufficiently to maintain some of the ciliate protozoa in suspension. The donor animals should be fed the feed to be used during the incubations.

#### **6.6.4.1 Incubation of Rumen Contents in Glass Syringes: A Simple Artificial Rumen**

The simplest type of short-term artificial rumen consists of test tubes or flasks with rumen contents incubated at 39 °C. Anaerobic conditions are maintained by bubbling CO2 during incubation or by providing the vessels with Bunsen valves. The latter arrangement is difficult to manipulate (addition of substances, withdrawal of samples); an important component of fermentation products (gas) cannot be quantified, and the free venting of gas can result in contamination when radioisotopes are used. The procedure described below, which was developed by Czerkawski and Breckenridge (1977), is simple, and the apparatus is inexpensive; it does not suffer from the disadvantages discussed above, and it can be readily adapted to different requirements.

**Plate 6.6** Simple apparatus for incubating small samples of rumen contents in a water bath

#### *Procedure*


useful since, with care, a curve relating gas production with time can be drawn. The shape of this curve can provide useful information about the extent and magnitude of the reaction in the incubated sample.


#### **6.6.4.2 Continuous Fermenters**

Three main types of continuous systems are used. The first is actually a sequential culture, where a small amount of solid feed is added to a sample of rumen fluid which is then sub-cultured daily (Merry et al. 1987). Whether methane production is sustained and mimics accurately, the real situation is unclear. The second type involves the continuous flow of both solid and liquid phases (Hoover et al. 1976). The third type is the RUSITEC scheme where solids are received once daily, but in the liquid phase by continuous flow. Use of the RUSITEC scheme requires a considerable input of time and expertise and is not recommended unless the specific objective is particularly suited to the apparatus, i.e. only where small quantities of the potential modifying agents are available, and the small-scale and between-vessel reproducibility offers a major advantage over animal experiments.

## **6.7 Methane from Animal Wastes**

Animal waste is a significant source of CH4. The amount produced depends on the diet of the animal. If the waste is stored anaerobically in lagoons or in liquid slurry tanks, CH4 production can be closely related to the disappearance of organic matter. Temperature and relative humidity affect the rate at which the faecal matter dries out under grazing and feed conditions and determines how long fermentation continues. Estimates of CH4 production from these systems can be obtained by adapting techniques developed for field use. The closed chamber method described for groups of animals (Chap. 2) and in rice paddies is likely to be the best method to use in measuring CH4 production from lagoons and tanks. The emission rate of CH4 that is likely to arise from animal waste can be very large. It may require a shorter sampling time and may also require mechanical mixing of the vented gas to prevent concentration gradients from forming in the chamber or space.

For waste lagoons, floating chambers of the types used in water studies can be employed (Chap. 2). Alternatively, modifications of the hood and facemask technique may be used, whereby air is drawn over the surface of the animal waste and subsampled for analysis. Non-isotopic tracers such as SF6 can be used for measuring CH4 production in barns and other enclosed areas where animal waste may accumulate.

## **6.8 Storage and Analysis of Samples**

## *6.8.1 Storage of Samples*

Unlike solid or liquid biological samples, which can often be frozen, dried, inactivated or preserved chemically, gas samples require very specialised handling during all stages of their analysis. Furthermore, the gases dealt with here are colourless and they have no odour. Thus, whatever the procedure, whether during sampling, storage and analysis, exceptional care should be taken to avoid losses and to eliminate any chance of such losses going undetected.

Methane passes through plastic materials, and care is required when storing gas samples prior to analysis. Glass and metal provide the most reliable materials for storing gas-containing CH4. Methane is also relatively insoluble in water, and storage under water can provide a useful means of containing the gas.

Metal and glass syringes greased with Vaseline provide safe storage over periods of 2–3 days. These should be sealed with metal taps or hypodermic needles inserted into a rubber stopper. An important rule is that gas samples should be analysed as soon as possible after sampling; if it is necessary to store gas samples, this storage should be reduced to a minimum. Metal taps (not plastic) should be used to seal syringes. Alternatively, steel needles can be used and pushed into a rubber stopper to seal them. Bags made from PVF are satisfactory for temporary storage of gas prior to analysis, as are metal-coated liners for wine and fruit juice casks. If the aim is to store the samples for long time before analysis is carried out, then Exetainers shall be used (Plate 6.7).

## *6.8.2 Analysis of Samples*

The methods shown in the above section require concentration measurements of CH4, SF6, volatile fatty acids, O2, N2, H2 or CO2. For concentration analysis, the appropriate method must be chosen under consideration of prevailing conditions, such as available apparatus, required sensitivity and the need for information about other associated gases (e.g. CO2 or H2). Essentially three methods are available. Whichever method is chosen, all the precautions listed above must be taken.

**Plate 6.7** Exetainers for gas storage

#### **6.8.2.1 Flammable Gases (CH4 and VFA)**

Flammable gases can be measured with a GC equipped with a flame ionisation detector (FID). A detailed description of this method is given in Chap. 2.

The main principle of a GC is that the gas sample (with is a mixture of several gases, e.g. O2, CO2, CH4 and others like VFA or SF6) is split into its components in a separation column. The separation columns differ in their composition, length and diameter. Make sure the column of your analytical system is appropriate for the respective gases to be measured. The manufacturer is the best source of information on appropriate columns for your measurements.

When a sample of CH4-containing gas is injected into the GC, the gas mix is separated into its components. Depending on the retention time in the column, one compound after the next reaches the FID (CH4 will pass through very fast). Flammable gases will trigger a signal in the detector, which is registered as a sharp peak. The area of this peak will be proportional to the concentration of CH4.

To determine the gas concentration in your sample, it is necessary to use reference gases (i.e. gases with a known gas concentration) during all GC measurements. A detailed description of the subsequent calculations and interpretation of your samples is given in Chap. 2.

#### **6.8.2.2 Thermal Conductivity Detector (TCD)**

Gas chromatographs fitted with this type of detector come with comprehensive descriptions for use supplied by the maker, who will also supply suitable columns and other auxiliary equipment. The use of the standard gas samples and calibration of the instrument are also dealt with by the manufacturer.

The value of this technique is that it is possible to separate all the permanent gases (O2, N2, H2, CO2, and, of course, CH4) and to determine their relative concentrations. It is customary to use two columns, a Porapak Q column which will separate CO2 from a composite peak containing all the other gases, and a Molecular Sieve 5A column which separates H2, O2, N2 and CH4, with CO2 becoming part of the baseline. This means that samples have to be analysed twice. However, it is also possible to put the two columns in series on the two sides of the detector (Czerkawski and Clapperton 1968). In this method, the gas (argon) passing through one side of the detector acts as reference, and then the polarity is reversed and the other side of the detector acts as reference. It is then only necessary to inject one sample and simply to reverse the polarity of the detector during the analysis.

#### **6.8.2.3 Infrared Detector (IRD)**

An infrared gas analyser is capable of measuring CH4 within the required range of concentrations using,


#### **6.8.2.4 Laser Techniques**

As described in Sect. 2.7, open-path lasers and also so-called cavity ring-down spectroscopy (CRDS) are nowadays available to determine the concentration of all relevant gases. Details on the use of these techniques can be found in Chap. 2, Sect. 2.7.

*Choice of analyser*: In choosing the most appropriate CH4 analyser, it is necessary to have a suitable measurement range. In most situations, 0–500 ppm CH4 is quite satisfactory, but by variations in analytical tube length, this range can easily be attenuated or extended. Most manufacturers can supply such options.

Of equal importance is the choice of analyser, i.e. whether it is a single channel or dual channel. In the former type, the single analytical tube is used to analyse both zero and CH4-containing samples, but it is not possible to accommodate changes in background CH4 concentrations or atmospheric pressure changes. In a dual-channel analyser, the use of two optically balanced analytical tubes permits the sample (unknown) gas to be measured against background air at all times in a different mode. By this procedure, changes in background CH4 concentrations and atmospheric pressure can be fully considered, and the reliability of the resultant data increases.

*Analysis*: Infrared gas analysers measure CH4 concentration in a steady stream of sample gas. Calibration procedures generally use a reference gas, i.e. N2 or outside air if N2 is unavailable, and a higher standard gas of known concentration. The reference and standard gases should be on either side of the range of the expected concentration of the sample gas to ensure accurate measurement of sample gas concentration. Care must be taken to ensure that the flow of sample gas to the analyser is within the recommended guidelines for the analyser and is constant throughout the analysis period. The sample gas should be dried before use. This may be accomplished by sending the gas stream through a short piece of pipe or tubing that has been loosely filled with a drying compound such as AquasorbTm or DrieriteTm or even anhydrous Na2SO4.

After the system is equilibrated and the reading is steady, the analyser reading is recorded and inserted into the calculation equation specific for the analyser in use.

#### **6.8.2.5 Non-flammable Gases (CO2 and SF6)**

Concentration measurements of CO2 and SF6 require a GC equipped with an electron capture detector (ECD) and TCD. Note that some GCs comprise both detectors in a row so that the gas sample passes both detectors. In this case, you get gas concentration of flammable gases and non-flammable gases.

Standards should be analysed daily. A 30 ml sample of gas from a collection flask should be removed with a syringe (plastic or glass), and part of this gas is passed through the loop and the loop full of gas is injected onto the column. SF6 will elute at approximately 20 s, and a sharp peak will be produced. O2 has a retention time of 50 s and will be the next peak produced. It will be well resolved from the SF6. Samples may be analysed at approximately four-minute intervals. The detection limit is approximately one part per trillion (ppt).

#### **6.8.2.6 Standards and Calibration**

In open-circuit calorimetry for measurements of CH4 concentration and ultimately its production, it is necessary to conduct regular calibration of the CH4 analyser and of the whole system.

(i) *Analyser calibration*

With respect to the analyser, where an infrared apparatus is available, this should be calibrated before and after use in order to establish both zero and set point deflection and to accommodate any analyser drift. The zero-point measurement can be achieved by appropriate use of a CH4-free air sample obtained from a fresh air source at a suitable distance from any livestock or livestock waste source. It is advisable that the gas sample is filtered and dried (e.g. through a calcium chloride tower) before injection into the analyser. A primary standard containing CH4 can be acquired in the first instance through a suitable commercial company or an international organisation. This standard ought to contain between 250 and 500 ppm CH4 and should be used initially to calibrate the analyser. Once this has been achieved, gas from a larger cylinder containing a similar but unspecified concentration of CH4 should be introduced into the analyser, and through repeated measurement its concentration should be established (secondary standard). Thereafter, this standard sample should be used regularly to standardise the analyser, and the primary standard should only be used to recalibrate new cylinders of the CH4 standard as required.

Experience shows that often successful research work involves some improvisation. This is particularly true when one develops new methods. An example of the type of problems that can be encountered relates to efforts made by scientists in a particular institute to develop a simple method for measuring the concentrations of CH4 in rumen gases. The apparatus required frequent calibration with standard gas mixtures. These gases were contained in two small aerosol cans and were considered to be so valuable that they were kept in a safe. During the development work, all the gas in one can was used up and it became imperative to do something before the remaining standard gas in the other can was used up since the delivery of another can would take 6–8 months. This was done by partly evacuating an old N2 cylinder in the laboratory and carrying it to a nearby anaerobic digester on the institute farm where it was filled with CH4-containing gas. The cylinder was then taken back to the laboratory and "topped up" with nitrogen to about 10 atmospheres, resulting in some 7% CH4 in nitrogen. The accurate concentration of CH4 in this secondary standard was determined using the precious primary standard. The secondary standard was then used during the remainder of the study and for many months thereafter.

Local conditions are very important. For instance, anaerobic digesters are very common in some countries, and it is easy to obtain CH4 for the secondary standard. If there are no digesters, some other source of CH4 has to be found; even a simple chemical method (i.e. methane derived from aluminium carbide) may be suitable.

#### (ii) *Whole systems*

With respect to calibration of whole systems such as respiration chambers, etc., to estimate the recovery of CH4 and check both the analyser and non-analyser components of the system, the preferred procedure is to introduce a gas of known CH4 concentration into the chamber over a minimum period of 6–8 h, and to determine its recovery downstream. Methane emission from the cylinder can be determined by a recording of cylinder weight loss over the test period and the CH4 concentration of the cylinder gas, whilst quantitative recovery of CH4 has to cover the period of time required for the concentration to return to baseline. Ideally, the recovery of CH4 should be within ±3% of added methane. If the results are outside this range, the test must be repeated, and if still unsatisfactory, individual components of the system should be isolated and checked for satisfactory function.

A direct method of calibration involves weighing the CH4 source and adding CH4 to the chamber. CH4 emission and analysis of outflow gas are continued until a measurable quantity of CH4 has been removed from the source. The source is then closed off, and analysis of outflow gas for CH4 content continued until outflow concentration returns to original background level.

An alternative to the burning of alcohol to produce CO2 can be simulated quite easily by acidifying a BaCO3 solution with dilute hydrochloric acid within the chamber. The generation of CO2 should be controlled by drip addition of the acid or by slow pumping of the acid to avoid massive fluctuations in CO2 content of the outflowing air.

## **References**


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 7 Isotopic Techniques to Measure N2O, N2 and Their Sources**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** GHG emissions are usually the result of several simultaneous processes. Furthermore, some gases such as N2 are very difficult to quantify and require special techniques. Therefore, in this chapter, the focus is on stable isotope methods. Both natural abundance techniques and enrichment techniques are used. Especially in

M. Zaman (B) · L. Heng

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai

School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang

Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

T. Clough Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

© The Author(s) 2021

M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_7

the last decade, a number of methodological advances have been made. Thus, this chapter provides an overview and description of a number of current state-of-theart techniques, especially techniques using the stable isotope 15N. Basic principles and recent advances of the 15N gas flux method are presented to quantify N2 fluxes, but also the latest isotopologue and isotopomer methods to identify pathways for N2O production. The second part of the chapter is devoted to 15N tracing techniques, the theoretical background and recent methodological advances. A range of different methods is presented from analytical to numerical tools to identify and quantify pathway-specific N2O emissions. While this chapter is chiefly concerned with gaseous N emissions, a lot of the techniques can also be applied to other gases such as methane (CH4), as outlined in Sect. 5.3.

**Keywords** 15N2O · 15N2 · 15N tracer technique

## **7.1 Introduction**

In this chapter, we are presenting techniques utilising the stable isotope 15N to better understand the N cycle but more importantly to determine GHG gas fluxes that cannot be quantified or are difficult to quantify with any non-isotopic technique. The

K. Dawar

W. X. Ding Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

C.-M. Görres Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

M. E. G. Hofmann Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

stable isotope 15N was discovered in the 1920s (Naudé 1929a, b) and the advantage of using this isotope in agriculture, for the determination of the N use efficiency has been recognised and applied since 1943 (Norman and Werkman 1943). Also, microbiologists have utilised the new possibilities that 15N can offer, to quantify the turnover rates of individual processes in the N cycle (Hiltbold et al. 1951) based on dilution principles (Kirkham and Bartholomew 1954). Moreover, 15N allowed for the first time the development of techniques to quantify the loss of N2 against a huge atmospheric N2 background (Hauck et al. 1958). Also, the identification which of the processes contributing to total N2O emissions (Butterbach-Bahl et al. 2013) is unthinkable without the use of advanced 15N tracing techniques (Müller et al. 2014). With the development of new and advanced analytical techniques, it is now possible to also use information on the position of the 15N (i.e. central, alpha and terminal, beta position) in N2O, i.e. the isotopomers (of one isotopologue), providing information

D. Lewicka-Szczebak

G. Lucic Picarro Inc., Santa Clara, CA, USA

#### L. Merbold

Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

M. Šimek

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

M. dos Reis Martins · S. Urquiaga EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

R. Well Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany

S. Zaman University of Canterbury, Christchurch, New Zealand

J. Zhang School of Geography, Nanjing Normal University, Nanjing, China

K. Lenhart Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

on the origin without the addition of 15N labelled fertiliser. Note, isotopologues are molecules that differ in their isotopic composition, isotopomers are molecules with the same isotopic atoms but differing in their position, and isotopocules is the generic term for both isotopologues and isotopomers. There is a wealth of information that we can obtain from using diverse isotopic approaches based on 15N or 18O labelling but also on natural abundance techniques that take advantage of the different metabolism with which for instance N2O is produced. Thus, 15N provides us with a toolbox to identify emission pathways and in turn provides information on effective mitigation techniques.

## **7.2 15N Gas Flux Method (15N GFM) to Identify N2O and N2 Fluxes from Denitrification**

## *7.2.1 Background*

N2O reduction to N2 is the last step of microbial denitrification, i.e. anoxic reduction of nitrate (NO3 <sup>−</sup>) to dinitrogen (N2) with the intermediates NO2 <sup>−</sup>, NO and N2O (Firestone and Davidson 1989; Knowles 1982). Commonly applied non-isotopic techniques enable us to quantitatively analyse only the intermediate product of this process including NO and N2O, but not the final product, N2, a non-greenhouse gas. The challenge to quantify denitrification rates is largely related to the difficulty in measuring N2 production due to its spatial and temporal heterogeneity and the high N2-background of the atmosphere (Groffman et al. 2006). There are three principal ways to overcome this problem: (i) adding NO3 <sup>−</sup> with high 15N enrichment and monitoring 15N labelled denitrification products (15N gas flux method, 15N GFM) (e.g. (Siegel et al. 1982)); (ii) adding acetylene to block N2O reductase quantitatively and estimating total denitrification from N2O production (acetylene inhibition technique, AIT) (Felber et al. 2012); (iii) measuring denitrification gases during incubation of soils in absence of atmospheric N2 using gas-tight containers and an artificial helium/oxygen atmosphere (HeO2 method; (Butterbach-Bahl et al. 2002; Scholefield et al. 1997; Senbayram et al. 2018)). Each of the methods to quantify denitrification rates in soils has various limitations with respect to potential analytical bias, applicability at different experimental scales and the necessity of expensive instrumentation that is not available for routine studies. Today the AIT is considered unsuitable to quantify N2 fluxes under natural atmosphere, since its main limitation among several others is the catalytic decomposition of NO in presence of O2 (Bollmann and Conrad 1997), resulting in unpredictable underestimation of gross N2O production (Nadeem et al. 2012). The 15N gas flux method requires homogenous 15N-labelling of the soil (Mulvaney and Vandenheuvel 1988) and under natural atmosphere, it is not sensitive enough to detect small N2 fluxes (Lewicka-Szczebak et al. 2013). Direct measurement of N2 fluxes using the HeO2 method is not subject to the problems associated with 15N-based methods (Butterbach-Bahl et al. 2013) but the need for sophisticated gas-tight incubation systems limits its use. When applying 15N GFM in the laboratory, sensitivity can be augmented by incubation under an N2-depleted atmosphere (Lewicka-Szczebak et al. 2017; Meyer et al. 2010; Spott et al. 2006). In the following, the basic principle, limitations, bias and application examples are presented and discussed.

## *7.2.2 Principles of the 15N Gas Flux Method*

The 15N gas flux method consists of quantifying N2 and or N2O emitted from 15Nlabelled NO3 <sup>−</sup> applied to soil in order to quantify fluxes from canonical denitrification (Mulvaney and Vandenheuvel 1988; Stevens et al. 1993), where N2 and N2O are formed from the combination of two NO precursor molecules. Under certain preconditions, it is also possible to identify the production of hybrid N2 or N2O (i.e. molecules formed from the combination of N atoms from one source of oxidised N, e.g. NO2 <sup>−</sup>), and another source of reduced N (e.g. NH3 or NH2OH) via anaerobic ammonia oxidation (annamox) or co-denitrification (Laughlin and Stevens 2002; Spott and Stange 2007; Spott et al. 2011). To quantify canonical denitrification, experimental soil is amended with NO3 <sup>−</sup> highly enriched with 15N. The 15N gases evolved are collected in closed chambers and 15N emission is calculated from the abundance of N2 and N2O isotopologues in the chamber gas. 15N enrichment of N2 in the gas samples are typically close to natural abundance because the amount of N emitted from the 15N-labelled soil is small compared to the atmospheric background. Precise techniques of isotope analysis are, therefore, necessary.

#### **7.2.2.1 The Non-random Distribution of Atoms in the N2 Molecule**

The 15N gas flux method is based on the assumption that within N2 or N2O from a single source of a given 15N abundance, the N2O isotopologues of a distinct number of 15N substitutions follow a random (binomial) distribution, as given by the terms in (Eq. 7.1):

$$\left((p+q)^2 = p^2 + 2pq + q^2\right) \tag{7.1}$$

where p is the atom fraction of 14N, q the atom fraction of 15N and p + q is equal to unity (Hauck et al. 1958).

If N2O or N2 from two different N pools, one background pool of natural 15N abundance (0.3663 atom %) and the second enriched in 15N are mixed, the distribution deviates from the binomial pattern. Given the distribution of N2 or N2O isotopologues emitted from the first (background) N pool (*abg*) including non-labelled N2 and N2O (derived from the atmosphere and possibly non-labelled N2O from non-labelled N sources in soil) and the resulting mixture (*a*m), the 15N abundance in the 15N-labelled second pool (*ap*) and the fraction of N2O or N2 originating from that labelled pool (*f <sup>p</sup>*) can be determined (e.g. Bergsma et al. 2001; Spott et al. 2006). To calculate *f* <sup>p</sup> values, the nitrogen isotope ratios 29R(29N2/ 28N2) and 30R(30N2/ 28N2) are used. In case of N2, the three isotopologues 14N14N and 14N15N and 15N15N are detected. For N2O, one option is to directly analyse intact N2O molecules, consisting of N and oxygen (O) and analysing molecular masses 44, 45 and 46. It has to be taken into account that these molecular masses include not only N- but also O-substituted isotopocules and thus the following 6 species: 14N14N16O with mass-to-charge (m/z) 44, 14N14N18O (m/z 46), the isotopomers 14N15N16O and 15N14N16O (both m/z 45), 14N14N17O (m/z 45) and 15N15N16O) (m/z 46). To calculate 15N pool-derived N2O, 29R and 30R of the N2O–N is calculated taking into account the natural abundance of 17O- or 18O-substituted isotopocules (14N14N18O and 14N14N17O) due to their mass overlap with the 15N-substituted isotopocules (Bergsma et al. 2001). Alternatively, N2O can be reduced to N2 prior to IRMS analysis (Lewicka-Szczebak et al. 2013), thereby allowing direct determination of 29R and 30R of N2O–N.

There are various calculation procedures that have evolved over time (Hauck et al. 1958; Mulvaney 1984; Arah 1992; Nielsen 1992, Well et al. 1998; Spott et al. 2006). In Eqs. 7.2 and 7.3 we show one example (Spott et al. 2006), where the fraction of N2 or N2O evolved from the 15N-labelled NO3 <sup>−</sup> pool (*f <sup>p</sup>*) is calculated:

$$f\_p = \frac{a\_m - a\_{bg}}{a\_p - a\_{bgd}}\tag{7.2}$$

where *am* is the 15N abundance of the total gas mixture

$$a\_m = \frac{\,^{29}R + 2 \ast ^{30}R}{2\left(1 + \,^{29}R + \,^{30}R\right)}\tag{7.3}$$

and *abg* is the 15N abundance of atmospheric background N2. The 15N abundance of the 15N-labelled nitrate pool undergoing denitrification is

$$a\_p = \frac{{}^{30}x\_m - a\_{b\emptyset d} \* a\_m}{a\_m - a\_{b\emptyset d}} \tag{7.4}$$

where <sup>30</sup>*xm* is the measured fraction of m/z 30 in the total gas mixture:

$$\mathbf{x}^{\;30}\mathbf{x}\_m = \frac{\mathbf{x}^{\;00}\mathbf{R}}{1 + \,^2\mathbf{R}^2 + \,^3\mathbf{R}}\tag{7.5}$$

The same calculations can be used for N2 and N2O, resulting in respective values for fractions of pool-derived N (*f* p\_N2; *f* p\_N2O) and for the respective 15N abundances of the active N pools *(a*p\_N2; *a*p\_N2O).

If only m/z = 28 and m/z = 29 are determined during isotope analysis of N2, then emission of 15N2 is underestimated (Hauck et al. 1958). The extent of underestimation is related to the 15N atom fraction of the NO3 <sup>−</sup> pool from which N2 is emitted (Well et al. 1998) and *f* <sup>p</sup> can thus be calculated if the 15N enrichment of the denitrified N pool is known (Mulvaney 1984):

$$f\_p = \left(^{29}R\_{sa} - ^{29}R\_{bg}\right) / \left(2 \, a\_p \left(1 - a\_p\right)\right) \tag{7.6}$$

where lower case sa and bg denote sample and background (typically ambient air), respectively. An alternative equation yielding *f* <sup>p</sup> from 29R that is more complex, but also more precise, is given by Spott et al. (2006).

In many studies, a 15N atom fraction of 0.99 was selected for the 15N enrichment of applied NO3 <sup>−</sup> ( 15aNO3) in order to maximise 30R (see Fig. 7.1), thus yielding better 30R signals. However, there are also reasons to keep *15aNO3* between about 0.6 and 0.4, since 30R is only detectable with high fluxes due to a typical high IRMS background signal at m/z 30 (see next section), so that *f <sup>p</sup>* has to be calculated from 29R only using Eq. 7.6. But *f <sup>p</sup>* calculated from Eq. 7.6 with a given 29R is relatively insensitive to changes in *ap* between 0.4 and 0.6 since the nominator yields, e.g. for *ap* between 0.4 and 0.6, values between 0.48 and 0.5. Hence, uncertainty in the estimation of *ap* within that range causes minor uncertainty in calculated *f <sup>p</sup>* (Well and Myrold 1999).

**Fig. 7.1** Abundance of 28N2, 29N2 und 30N2 in air, in soil-emitted N2 evolved from NO3 − with 50 atom% 15N, and in a 1:1000-mixture without and with randomisation of isotopologues by N2 dissociation, respectively

To illustrate how the combination of denitrification rates (i.e. *f <sup>p</sup>*) and homogenous or non-homogenous 15N enrichment of the soil NO3 <sup>−</sup> pool affect instrumental raw data as well as calculated *f <sup>p</sup>* and *ap* values, some theoretical data are shown (Table 7.1). Three cases are represented, (1) the soil is homogenously labelled with 15N, (2) non-labelled soil-derived NO3 <sup>−</sup> dilutes the labelled pool to a different extent in the 0 to 10 and 10 to 20 cm layers, but N2 and N2O production rates in both layers are equal and (3) like case (2) except that production rates of both layers differ. It can be seen that only case (1) calculated using Eq. 7.4 yields results identical to ideal *ap* and *f <sup>p</sup>*. Equation 7.6 gives deviating results when used with *15aNO*<sup>3</sup> as this value differs from *ap*. In the case of (2) and (3), all calculations lead to some deviation due to the non-homogeneity in label distribution. Moreover, isotope ratios show that even at the high denitrification rate assumed (case 2, 542 g N ha−<sup>1</sup> 20 cm−<sup>1</sup> d−1), the increase in 29R (29Rm–29Ra) and 30R (30Rm–30Ra) was 9.2 × 10−<sup>6</sup> and 3 × 10−6, respectively, and thus only about one order of magnitude above typical instrumental precision (see Table 7.2).

## *7.2.3 Identifying the Formation of Hybrid N2 and/or N2O*

When N2 and N2O are formed from denitrification, both N atoms are derived from the 15N labelled pool, and in hybrid N2 or N2O only one N atom comes from the labelled pool (N oxides, i.e. NO2 <sup>−</sup>) and the other one comes from non-labelled reduced N (e.g. NH3, NH2OH or organic N). Hence, the contribution of hybrid processes is reflected by an increase in 29R only, while denitrification increases both 29R and 30R (Clough et al. 2001). Laughlin and Stevens (2002) derived equations to calculate the fraction of hybrid and non-hybrid N2, assuming that the measured 15N atom fraction of NO3 − also reflected the enrichment of the NO2 <sup>−</sup> that contributed one N atom to the hybrid molecules, and that the 15N abundance of the non-labelled sources (atmospheric N and non-labelled reduced N) were identical. An extended approach was developed allowing to take into account different 15N enrichment for all contributing sources, i.e. different values for atmospheric and reduced N (Spott and Stange 2007; Spott et al. 2011). Spott et al. (2011) used those equations to calculate co-denitrification in a soil slurry but pointed out that the approach would be subject to possible bias due to difficulty and inaccuracy when determining the 15N enrichment of the nitrite (NO2 −) pool contributing to the hybrid formation. For N2O mixtures consisting of N2O from only two sources, i.e. hybrid and non-hybrid N2O, the authors, therefore, suggest to use the indicator value Rbinom to assess the contribution of hybrid N2O. Rbinom reflects the fact that N2 or N2O isotopocules of each non-hybrid source contributing to a gas mixture are following a random (binomial) distribution, whereas this is not the case for the hybrid N2O. Rbinom values >1 indicate a significant hybrid contribution. While fluxes excluding hybrid N2O would always yield Rbinom ≤ 1, respective Rbinom values would not exclude the possibility of some hybrid contribution. Hence, Rbinom can only prove the existence (but not the absence) of hybrid fluxes. The limitation of this approach is that it does not work in the presence of additional sources, e.g. if





**Table 7.2** Typical precision (standard deviation, SD) for the nitrogen isotope ratios 29R (29N2/ 28N2) and 30R (30N2/ 28N2) by IRMS

there is N2O from unlabelled sources including atmospheric N2O. Thus, Rbinom does not work for N2 because there is always a high background of atmospheric N2. To our knowledge, systematic and quantitative studies on hybrid fluxes from soils, including quantification of average pool enrichment and its homogeneity or non-homogeneity, and estimation of resulting uncertainties, have not yet been accomplished.

## *7.2.4 Analysis of N2 and N2O Isotopologues*

Precise quantification of N2 and N2O isotopocules requires isotope ratio mass spectrometry (IRMS) where 29R and 30R are obtained from ion current ratios detected at Faraday collectors tuned for m/z 28, 29 and 30 (e.g. Lewicka-Szczebak et al. 2013). A double collector IRMS was used before multi-collector IRMS became available. Double collector IRMS required two measurements with the IRMS so that either 29N2 or 30N2 is positioned on the first collector (Siegel et al. 1982). Emission spectroscopy has also been used in the past to detect 28N2, 29N2 and 30N2 (Kjeldby et al. 1987), but its relatively low precision enabled only detection of large N2 fluxes. While dual inlet IRMS had been used with manual measurement of samples in glass containers that were sealed (Well et al. 1993) or isolated by stopcocks (Siegel et al. 1982), continuous flow IRMS enables automated injection of samples from septum capped vials since the 1990s (Stevens et al. 1993). Recently, further progress was obtained by automated analysis of N2, N2+N2O and N2O in one run, including N2O reduction to N2 (Lewicka-Szczebak et al. 2013). The latter enables the analysis of N2O-N at m/z 28, 29 and 30, thus excluding the need to conduct 17O and 18O corrections, yielding better precision, since O corrections are biased to some extent by natural variation of 17O and 18O (Deppe et al. 2017).

While quantification of 29R is quite robust, 30R is affected by the mass overlap of 30N2 with the most abundant isotopocule of NO (14N16O), since NO+ is formed at the hot filament in the ion source of the IRMS (Brand et al. 2009, Siegel et al. 1982) due to the omnipresence of oxygen traces. NO+ formation can be quantified by the ratio between ideal and measured 30R of standard gases, giving values of 0.15 to 0.06 for atmospheric N2 analysed in the instrumentation proposed by Lewicka-Szczebak et al. (2013). NO+ formation can be minimised by removal of all O sources (O2, H2O) from the samples and also from the carrier and reference gases. In some types of IRMS the NO+ background is too high and associated with extreme tailing of the m/z 30 peak. This makes it impossible to quantify 30R (Well et al. 1993). To overcome this limitation, a procedure to quantify 30R indirectly from 29R was developed where 29R had to be analysed twice, (i) in samples where the non-random distribution of N2 isotopocules was randomised by the temporary splitting up of N2 molecules during a gas discharge (see change in 29R due to randomisation in Fig. 7.1). Discharge was actuated using a microwaves source, initially offline in sealed glass tubes, later with online continuous flow IRMS, where the discharge occurred in the gas circuit connecting and IMRS (Well and Meyer 1998). An overview of the IRMS precision for 29R and 30R in N2 standard gases is given in Table 7.2, showing that repeatability for 29R varied significantly between instruments, but 30R is comparable. However, it is also evident that during the last 35 years (Siegel et al. 1982) there has been no substantial improvement in the measurement precision.

## *7.2.5 Detection Limit for ap and fp*

Because *f <sup>p</sup>* is calculated from two quantities, 29R and 30R, and the relationship between them depends on the 15N enrichment of the active N pool (*ap*, see Fig. 7.1), the limit of detection (LOD) for *f <sup>p</sup>* at given repeatability of 29R and 30R is variable. LOD for *f <sup>p</sup>* was thus determined for varying conditions using equations from Spott et al. (2006) using Monte Carlo modelling assuming a normal distribution of 29R and 30R errors (Standard deviation of repeated analysis of standard gas samples). The MS-Excel function *norm.inv* was used to create the normal distribution of values but allowing only a maximum deviation of 3 standard deviations, otherwise unrealistic outlier of 29R or 30R yield unrealistically high uncertainty. Different scenarios were tested (*f <sup>p</sup>* = 1 to 100 ppm; *ap* = 0.055 to 0.75 using repeatability for 29R and 30R of the first IRMS listed in Table 7.1). LOD is obtained for two cases: 1. Both 29R and 30R are taken into account to calculate both *ap* and *f <sup>p</sup>*; 2. *f <sup>p</sup>* is calculated using only 29R (using Eq. 7.4 in (Spott et al. 2006)) and *ap* is estimated either from soil extract analysis or from *ap* of N2O (e.g. Stevens and Laughlin 2002). Note that *ap* of N2O is usually much more reliable than *ap* of N2 since *f <sup>p</sup>* of N2O is typically large (often between 0.1 and 1) due to the fact that, in contrast to N2, N2O is an atmospheric trace gas. Conversely, *f <sup>p</sup>* of N2 is typically very small (usually <10−<sup>5</sup> in ambient atmosphere).

The first calculation is preferable because *ap* of N2 and N2O can be different (see Fig. 7.3) and *ap* of N2O can only be obtained if N2O can be directly measured by IRMS, which is only the case if concentrations are high enough (about 0.3 to 3 ppm necessary, depending on 15N enrichment of N2O). Since incubation under N2-depleted atmosphere improves *f <sup>p</sup>* sensitivity, LOD is also given for an artificial gas mixture containing 2% N2.

LOD results are as follows (Table 7.3): with high *f <sup>p</sup>* (i.e. ≥10 ppm) and high *ap* (i.e. ≥0.5) and ideal IRMS performance (Table 7.2) both calculations yield precise results. Under N2-depleted atmosphere, LOD is excellent (2 to 7 ppb N2, last columns). With lowering of *ap*, LOD gets worse if *ap* has to be calculated using 30R. But without using

**Fig. 7.2** Abundance of 28N2, 29N2 and 30N2 in N2 evolved from the 15N–labelled NO3 – depending on *ap* (Siegel et al. 1982)

30R and assuming an ideal *ap* value or estimating *ap* of N2 from direct determination of *ap* of N2O, LOD of *f <sup>p</sup>* is still excellent. This is because with decreasing *ap*, abundance of 15N15N (30N2), and thus 30R, decreases exponentially whereas the decrease of 29R ( 29N2) is much slower (see Fig. 7.2).

## *7.2.6 Limitations of the 15N Gas Flux Method (15N GFM)*

The following factors limit the applicability of the 15N GFM

#### **7.2.6.1 Inaccurate Definition of the Soil Volume Represented by Denitrification Measurements and Incomplete Recovery of Denitrification Gases**

The denitrifying soil volume is clearly defined if soil cores are entirely labelled with 15N and are incubated in closed systems. However, in situ measurement of denitrification in surface soils or subsoils with approaches other than the core methods do not include complete enclosure of the investigated soil. It is not possible to control the application of 15NO3 <sup>−</sup> accurately. Consequently, the soil volume represented by the detected denitrification gases is not exactly defined, and calculated denitrification rates are associated with uncertainty. Partial enclosure of the investigated soil is typically achieved by driving cylinders into surface soils. This option reduces the problem to a certain extent. Moreover, measuring the spatial distribution of the 15N label at the end of experiments (Well and Myrold 2002) helps to constrain the soil volume contributing to soil N2 fluxes that can be "seen" by 15N analysis of headspace gases.

An additional problem of open systems is the difficulty to determine the direction and strength of diffusional gas transport. When chamber methods are used to determine denitrification of surface soils, a significant fraction of the denitrification gases produced in the 15N-labelled soil diffuses into the subsoil and is thus not recovered in the chambers. Principally, this can be solved by modelling diffusion of 15N labelled gaseous denitrification products (see Sect. 7.2.7).

#### **7.2.6.2 The Problem of Non-homogenous 15N Enrichment of the NO<sup>−</sup> 3 Pool**

An overview of techniques to supply 15N-labelled NO3 – to the soil is given in Table 7.4. The 15N GFM is based on the assumption of an isotopically homogenous NO3 – pool. Because this condition is rarely achieved in soils, underestimation of denitrification rates up to 30% can result (Arah 1992; Mulvaney and VandenHeuvel 1988). An initial homogeneity can be obtained by intensive mixing of the soil, but this is a massive disturbance with huge potential effects on N processes including denitrification dynamics and is only adequate to simulate soil tillage with similar disturbance. But even with initially ideal tracer distribution, non-homogeneity inevitably develops over time, since N transformations including nitrification, denitrification and immobilisation are never homogenous in structured soil where aerobic and anaerobic domains coexist and organic matter fractions of varying reactivity are unevenly distributed. Injection of 15N tracer solution (Wu et al. 2012) increases moisture and inevitably produces non-homogeneity with maximum label concentration at the injection spots. Saturation and drainage (Nõmmik 1956) or soil water displacement by irrigation of lysimeters (Well et al. 1993) leads to an interim increase in moisture and causes loss of DOC. Labelling with gaseous NO2 was not a suitable way to achieve high and homogenous enrichment of soil NO3 <sup>−</sup> (Stark and Hart 1996). Consequently, non-homogeneity of the label distribution is probably the main source of bias of the 15N GFM. Often 15N tracer has been applied to the surface similar to conventional fertilisation (Baily et al. 2012). However, in this case, only fertiliser derived fluxes are detected initially, while during ongoing diffusion and leaching of NO3 <sup>−</sup>, the 15N labelled NO3 <sup>−</sup> pool rapidly changes its dimensions and thus non-homogeneity complicates the interpretation of results.

Possible causes and consequences of non-homogenous distribution of the 15Nlabel and denitrification /nitrification dynamics is illustrated using two conceptual models (Well et al. 2015). The first model shows how *ap* of N2 and N2O can differ due to non-homogeneity in 15N enrichment and also non-homogeneity in N2 and

**Fig. 7.3** Model 1 to explain why N2 and N2O from denitrification can originate from different effective 15N pools: In the lower pool with a higher 15N enrichment, N2 fluxes dominate over N2O, whereas the opposite is the case for the shallow pool with lower enrichment. Hence, emitted N2 is more enriched compared to emitted N2O

N2O production rates (Fig. 7.3). Even if equal amounts of 15N tracer solution could be applied to each soil layer, 15N enrichment of NO3 <sup>−</sup> would be variable due to the different dilution of the label via soil-derived NO3 <sup>−</sup>. Additionally, production rates of N2 and N2O and their ratio are typically spatially variable, which results in differing *ap* values for N2 and N2O (Fig. 7.4). The development of spatial heterogeneity in 15N enrichment and the consequences arising from the fact that nitrification and denitrification typically occur in different soil niches is shown with the second conceptual model (Fig. 7.4) that had been used to explain observations (Deppe et al. 2017). In that study, the soil had been mixed with 15N labelled NO3 <sup>−</sup> and non-labelled NH4 + and isotopic values of initial NO3 <sup>−</sup> and final NO3 <sup>−</sup> and N2O had been compared. Results showed that *ap* of N2O was similar to initial enrichment of soil NO3 <sup>−</sup> (13 atom% 15N), but final NO3 <sup>−</sup> enrichment of the bulk soil was much lower (about 3 atom% 15N) whereas *ap* of N2O did not change significantly. This was postulated to result from the dilution of the label only in aerobic domains where nitrification occurred, whereas in anaerobic microsites there was no nitrification, and hence no dilution of the label. But the undiluted microsites produced all or most of the N2O whereas there was negligible N2O flux from aerobic domains. While this discrepancy between 15N enrichment of NO3 <sup>−</sup> in the bulk soil and *ap* of N2O was certainly extreme in that study, similar process dynamics can be expected in many cases. Such

**Fig. 7.4** Model 2 to explain possible non-homogeneity in 15N-labelling of NO3 <sup>−</sup> in NH4 +-fertilised soil (Deppe et al. 2017). Colours represent enrichment (blue <sup>=</sup> nat abundance, red <sup>=</sup> max. 15N enrichment). a. Initial enrichment of NO3 <sup>−</sup> results from mixing of soil NO3 <sup>−</sup> and added 15N-NO3 <sup>−</sup>. b. Initial homogenous distribution of labelled NO3 <sup>−</sup> and non-labelled NH4 <sup>+</sup> in the soil matrix. c. In anaerobic microsites, nitrification is inhibited and the NO3 <sup>−</sup> pool of initial 15N enrichment is denitrified and produces N2O of identical enrichment. In aerobic domains, nitrification of non-labelled NH4 <sup>+</sup> produces non-labelled NO3 <sup>−</sup>, thus diluting the initial labelled NO3 − pool and emitting unlabelled N2O. Note that the 15N enrichment of NO3 − undergoing denitrification is larger than the average 15N enrichment of extracted NO3 <sup>−</sup> and of emitted N2O

non-homogeneity in label distribution and its dilution as well as N2 and N2O production leads to uncertainties in calculation of *f <sup>p</sup>* (see Table 7.1). But these examples also show that comparison of *ap* of N2 and *ap* of N2O can be used to identify heterogeneity in labelling and thus stress the importance of using analytical methods including 29R and 30R of N2 and N2O-N (Lewicka-Szczebak et al. 2013). Moreover, it shows that calculating *f <sup>p</sup>* based on 15N enrichment of bulk NO3 <sup>−</sup> from soil extraction (Eq. 7.6) can lead to severe bias, since the 15N enrichment of the active pool can strongly deviate from the bulk pool. Moreover, an advantage of the non-random distribution approach with N2 and N2O is that non-homogeneity is indicated by discrepancies between *ap* of N2, *ap* of N2O and 15aNO3, which is quite useful (Lewicka-Szczebak and Well 2020). But it also shows that hybrid fluxes are difficult to identify if label distribution is non-homogenous.

Further limitations of the 15N GFM have been reviewed previously (Aulakh et al. 1991; Groffman et al. 2006; Sgouridis et al. 2016). They include enhancement of denitrification by NO3 <sup>−</sup> application in unfertilised systems, gas entrapment in very wet or fully water-saturated soils or sediment, and limited residence time of applied 15NO3 −-N due to plant uptake and leaching.

**Table 7.3** Detection limit of the 15N GFM determined by Monte-Carlo modelling. Detection limit for the fraction of pool derived N2 (*f <sup>p</sup>* of N2) is given as 1 standard deviation (SD) in dependence of 15N enrichment of active labelled NO3-pool (ap) and magnitude of *f <sup>p</sup>* in atmospheres with 100% or 2% N2 and assuming IRMS precision of 29R and 30R according to the first instrument in Table 2


#### **7.2.6.3 Combining the 15N GFM with Modelling of Gross N Transformation**

The current model to analyse data for the 15N GFM cannot be used to solve situations that include multiple labelled pools and heterogeneity of process activity and thus yield variable results in terms of flux quantification. Therefore, more complex models are needed to fill this gap. A 15N tracing model had been developed to analyse N2O dynamics in terrestrial ecosystems, which builds on previous tracing models for the quantification of the main mineral N transformations and soil nitrite (NO2 −) dynamics (Müller et al. 2014). This model is thus a first step in taking more complex dynamics into account. Extending this approach to model heterogeneity of processes and pools might be a promising way to solve current limitations of the 15N GFM. For more information on the tracing technique see Sect. 7.5 of this chapter.

## *7.2.7 Evaluation of the 15N GFM*

While quantification of N2 and N2O fluxes from distinct N pools remains a challenge after several decades of method development and improvement, this is even more the case for robust evaluation of methods, as this requires that the reference method is quantitative and is applied under the same conditions as the tested method. From that perspective, all previous tests included some uncertainties to our knowledge and were thus not fully able to evaluate the 15N GFM. There have been several comparisons between 15N GFM and AIT with controversial results, i.e. reporting general agreement (Aulakh et al. 1991) and severe underestimation by AIT (Arah et al. 1993; Sgouridis et al. 2016). Aulakh et al. (1991) compared 15N GFM and AIT in the field and found that 15N fertiliser derived N2 + N2O fluxes were comparable to total N2O fluxes in presence of acetylene (C2H2), suggesting that both methods were in general agreement. However, in all comparisons, 15N fertiliser was surface applied, so only the soil volume reached by the fertiliser contributed to the surface flux, unlike the AIT, were a larger soil volume was reached by the gaseous acetylene supplied by perforated pipes or buried calcium carbide. Hence comparisons did not reflect equal parts of the soil profile. Interestingly, in most comparisons denitrification was enhanced by soil compaction or glucose amendment, to achieve detectable 15N2 fluxes against the atmospheric N2 background. Sgouridis et al. (2016) compared closed chamber 15N GFM using needle injection to distribute K15NO3 <sup>−</sup> evenly with the AIT "soil core" variant finding 3 to 5 times higher rates with 15N GFM. Kulkarni et al. (2014) conducted an extensive comparison of the HeO2 method using small cores (5 cm diameter × 5 cm height, incubated under HeO2 in the lab) with in situ measurement using the 15N GFM where KNO3 <sup>−</sup> with 99 atom% was sprayed on the soil surface. Authors discussed difficulties to compare measurements in view of O2 manipulation in the lab and uneven label distribution in the field as well as variable moisture and temperature conditions in the field, and also that there are N2 fluxes from sources other than NO3 <sup>−</sup> (Butterbach-Bahl et al. 2013). What is still needed for a quantitative evaluation of the 15N GFM is to incubate 15N-labelled soil in a HeO2 setup to allow direct comparison of GC- and IRMS based N2 fluxes.

If 15N GFM is conducted under conditions maximising sensitivity and minimising bias, it can be used to evaluate other methods as for example the N2O isotopocule approach to determine N2O reduction (Lewicka-Szczebak et al. 2017, Buchen et al. 2018) (see Sect. 7.3).

## *7.2.8 Lab and Field Experiments*

Initial application of 15N GFM in lab incubations was carried out in closed vessels (Melin and Nõmmik 1983; Siegel et al. 1982). Recently, some studies used N2 depleted atmosphere to increase sensitivity in soil incubations (Lewicka-Szczebak et al. 2017; Schorpp et al. 2016) achieving sensitivities for pool-derived N2 of approximately 50 ppb which is thus comparable to GC sensitivity for N2O and two order of magnitude more sensitive compared to 15N GFM under ambient atmosphere. Important to note is that this also improves precision for quantifying *ap* and thus yields more precise estimates for the dilution of the denitrified pool by soil-derived NO3 −.

A key feature of 15N GFM is in situ measurement of denitrification and today it must be considered the only available field method, since AIT has been found unsuitable (Felber et al. 2012; Nadeem et al. 2012; Sgouridis et al. 2016). But 15N GFM has been used far less compared to the AIT probably due to its low sensitivity and high effort and expense to keep high 15N labelling in the field for extended periods, and also because of the multiple sources of bias. 15N GFM has thus been primarily used for soil types and/or conditions with high denitrification potential, e.g. due to abundant organic C (e.g. in organic soils or after soil compaction Arah et al. 1993). Typically, experiments covered only certain phases of the year. Maybe the most extensive study (including an extensive review of past in situ measurements) was by Sgouridis et al. (2016) who conducted 15N GFM in 4 sites monthly during about 18 months. But it has recently been found that during field application of the 15N GFM, denitrification is severely underestimated because a large fraction of the labelled N2 and N2O produced is not emitted from into the soil surface but diffuses to the subsoil or accumulates in pore space (Well et al. 2019a). This was confirmed experimentally and production–diffusion modelling showed that under typical experimental conditions, denitrification rates would be underestimated by more than 50%. It was concluded that field surface fluxes of 15N-labelled N2 and N2O have been severely underestimated in the past, but that diffusion modelling can be used to correct data. Moreover, to overcome the poor sensitivity of in situ 15N GFM, a new procedure was developed to conduct the 15N gas flux method using artificial N2-depleted atmosphere also for field application (Well et al. 2019b), giving a sensitivity for N2 + N2O fluxes up to 80-fold better compared to the conventional 15N GFM under ambient atmosphere. Consequently, recent methodical improvements are promising to yield good progress in the study of denitrification control at the field scale. 15N GFM has been used extensively with water saturated cores of aquatic sediments, e.g. Enrich-Prast et al. (2015), where sensitivity is less critical due to the possibility to measure 15N-labelled N2 dissolved in pore water where atmospheric N2 background is small.

#### **7.2.8.1 In Situ Measurement in Subsoil and Groundwater**

Some modifications of the 15N GFM for subsurface applications had been proposed and applied. For water saturated subsoil of hydromorphic soils or deeper groundwater, the "push–pull" type experimental setup (Istok et al. 1997) was combined with 15N tracing (Addy et al. 2002; Well et al. 2003; Well and Myrold 1999), where 15N tracer solution is injected in groundwater wells and groundwater samples are subsequently extracted over time and analysed for 15N labelled N2 and N2O. Similar to using 15N GFM in water saturated sediment in the lab (see above, Enrich-Prast et al. 2015), this approach is quite sensitive since produced N2 mixes with the small N2 background of N2 dissolved in groundwater. The 15N push–pull approach has been compared to slurry incubations of aquifer samples in the lab (Eschenbach et al. 2015; Well et al. 2005) finding good agreement between both approaches. It has also been successfully applied for deeper groundwater up to 90 m depth (Eschenbach et al. 2015).

In the unsaturated zone, subsoil denitrification has been quantified in situ from the steady-state 15N2 + 15N2O concentration within a defined 15N-labelled soil volume (Well and Myrold 2002). Diffusion-reaction modelling has been used to quantify rates by fitting measured and modelled *f <sup>p</sup>* values, but accuracy of this approach was limited by the difficulty to quantify the volume of 15N-labelled soil, its gas diffusivity and its distribution in 15N enrichment.

## *7.2.9 Conclusions and Outlook*

The 15N GFM is a powerful approach to quantify soil denitrification and its N2O/(N2 + N2O) mole ratio, to distinguish N2O fluxes derived from NO3 <sup>−</sup> and other N sources and, under certain conditions, also to identify the formation of hybrid N2 and N2O fluxes. It is applicable in the lab as well as in the field. But it is based on a variety of assumptions and prerequisites that are not always easy or possible to validate or to fulfil. Therefore, and because of its high expense for isotope tracers, IRMS analysis and demanding experimental setups, it has until now rarely been used routinely to study denitrification. Moreover, systematic evaluation using independent methods, e.g. using the HeO2 method, is still pending. Progress has been made in automated IRMS approaches that can be established using commercially available devices with some custom-made modifications. While sensitivity was clearly improved in the lab by incubation under N2 depleted atmosphere, this has not yet been fully realised for field conditions. These are good reasons to intensify the use of 15N GFM in future N cycle research, since despite large efforts during preceding decades, the magnitude of denitrification is still the big unknown of the N cycle (Butterbach-Bahl et al. 2013; Müller and Clough 2014).

## **7.3 Isotopocule Techniques to Identify Pathway-Specific N2O Emissions**

## *7.3.1 Introduction*

N2O isotopocules are the chemically identical N2O molecules but differing either in the atomic mass due to a substitution of one atom with heavy isotope 15N or 18O (isotopologues: 14N14N16O; 15N14N16O; 14N14N18O), or in the location of 15N substitution (isotopomers: 14N15N16O; 15N14N16O) (Toyoda et al. 2017). Thus, the asymmetric NNO molecule has in total twelve distinct isotopocules, representing all possible combinations of the N isotopes 14N and 15N and the oxygen isotopes 16O, 17O and 18O and providing a wealth of interpretation perspectives. Most commonly the three isotopic characteristics (δ18O, δ15N<sup>α</sup> and δ15N<sup>β</sup>) are measured, reporting the relative differences of isotope ratios of the four most abundant N2O isotopocules


**Table 7.4** Overview of current 15N- labelling techniques

14N14N18O/ 14N14N16O (δ18O), 14N15N16O/ 14N14N16O (δ15N<sup>α</sup>) and 15N14N16O/ 14N14N16O (δ15N<sup>β</sup>) in relation to a measurement standard defined on an international isotope ratio scale, Air-N2 for 15N/14N and Vienna Standard Mean Ocean Water (VSMOW) for 18O/16O. The average of δ15N<sup>α</sup> and δ15N<sup>β</sup> is referred to as δ15Nbulk and the difference between δ15N<sup>α</sup> and δ15N<sup>β</sup> (i.e. δ15N<sup>α</sup>–δ15N<sup>β</sup>) is called δ15N-site preference (δ15NSP), or commonly as SP (Toyoda and Yoshida 1999).

Natural abundance isotopic signatures can be used as an alternative approach to 15N tracing to constrain N2O transformations in the environment. Variations in stable isotope abundances are due to the fact that for many biotic and abiotic processes, the reaction rates differ between isotopic species, e.g. reduction of 15NO2 <sup>−</sup> versus 14NO2 <sup>−</sup>, leading to a so-called isotopic fractionation. As the isotopic fractionation is distinct for certain reaction pathways, isotopic signatures of particular production pathways and reduction fractionation factors determined in laboratory pure culture studies can be used to differentiate processes from each other. Distinct process information is provided by the difference in 15N substitution between the central and terminal position within the N2O molecule (SP), which is independent of the precursor's isotopic composition and characteristic of specific reaction mechanisms or enzymatic pathways. The most common interpretation strategy used to date is the dual isotope plot, also known as "mapping" approach, presenting the relationship between two isotopic parameters–commonly δ18O/ δ15Nbulk, δ15NSP/ δ15Nbulk or δ15NSP/ δ18O. From such figures, estimates can be made about trends, probable dominance of particular pathways, or reduction progress (Toyoda and Yoshida 1999; Lewicka et al. 2017; Koba et al. 2009; Ibraim et al. 2019) (Fig. 7.5).

N2O isotopocules at natural abundance levels can be analysed by isotope ratio mass spectrometry (IRMS) (Toyoda and Yoshida 1999) and more recently midinfrared (MIR) laser spectroscopic techniques.

With N2O isotopic analysis, the qualitative information can be added to the quantitative information gained from the concentration measurements. This is to naturally occurring differences between N2O from various origins as a result of isotopic fractionation, which causes enrichment or depletion of the reaction product in heavy isotope. Typically, for biochemical reactions we deal with the product depleted in heavy isotopes, but different biochemical pathways show characteristic isotope fractionation, which results in larger or smaller isotope effects (ε, difference between substrate and product (Eq. 7.7)), including also possible inverse isotope effects (product enriched in heavy isotopes, negative ε).

$$
\varepsilon \cong \delta\_{\text{product}} - \delta\_{\text{substrate}} = \Delta\_{\text{product}/\text{substrate}} \tag{7.7}
$$

Isotope effect is often expressed as Δ values, representing the difference between δ values of product and substrate. The values of ε should be used for a particular chemical reaction or physical transformation and describe the characteristic isotopic fractionation for this process (so-called intrinsic isotope effects), whereas Δ values may also be applied to describe an isotopic change between initial substrate and the final product, which may be due to a chain of following reactions and diffusion. This is the case e.g. for denitrification where we can mostly only determine the overall observed isotope effect between NO3 <sup>−</sup> and N2O (also called apparent or net isotope effect, Δ*15Nbulk N2O/ NO3*<sup>−</sup>) but without insight into intermediate products (NO2 <sup>−</sup>, NO) we cannot determine the ε values of individual reduction steps.

Due to distinct isotopic fractionation for various biochemical reactions, the N2O isotopic studies have been often used to distinguish between different N2O production pathways, e.g., nitrification and denitrification (Cardenas et al. 2017; Deppe et al. 2017; Köster et al. 2015; Toyoda et al. 2011; Wolf et al. 2015), or between different microorganisms involved in N2O production, e.g. fungal and bacterial denitrification (Kato et al. 2013; Schorpp et al. 2016; Zou et al. 2014). Moreover, also N2O reduction can be potentially monitored by N2O isotopic data. The possible reduction of N2O to N2 during denitrification is associated with isotopic fractionation, which changes the isotopic signature of the residual N2O. Therefore, isotopic analyses of residual N2O can be used to estimate the magnitude of its reduction and thereby the N2 production (Kato et al. 2013; Lewicka-Szczebak et al. 2017; Toyoda et al. 2011). Comprehensive reviews on the use of N2O isotopocules to estimate N2O dynamics are given by Ostrom and Ostrom (2011), Decock and Six (2013), Toyoda et al. (2017) and Yu et al. (2020). The main problem in the interpretation of isotopocule analysis of emitted N2O is the parallel production, possibly from various pathways, and consumption due to reduction to N2.

## *7.3.2 Principles*

For a proper interpretation of the analysed isotopic values of emitted N2O, both the possible production pathways and consumption due to N2O reduction to N2 must be taken into account.

To be able to identify potential production pathways, we need the basic data of the characteristic isotopic signatures for particular pathways, so called **endmember values**. These are obtained from the pure culture studies, where specific microorganisms are incubated separately and N2O is collected and analysed. Numerous pure culture studies are summarised in detail in the recent review papers (Denk et al. 2017; Toyoda et al. 2017). N2O isotopic signatures for specific pathways were also determined in controlled incubation of the whole soil by applying conditions favouring specific pathways. Such experiments were also summarised before (Denk et al. 2017; Toyoda et al. 2017). Here we present an overview of the most common pathways including results from pure culture studies and controlled soil incubations with some necessary critical selection explained below (after (Denk et al. 2017; Lewicka-Szczebak et al. 2017; Toyoda et al. 2017, Yu et al., 2020)). For each isotopic signature (δ*15Nsp,* δ*18O, and* δ*15Nbulk )* the rules how to properly use endmember values are explained and for each N2O production process the range of values (minimal and maximal literature reported value), the mean (of all literature reported values) and the median (of all literature reported values) is given.

*δ15Nbulk* of the produced N2O depends on the precursor isotopic signature, i.e. on soil NO3 <sup>−</sup> for denitrification and soil ammonium for nitrification. Therefore, to compare any results with literature endmember values we need to calculate the N isotopic signature of the N2O in relation to the precursor, i.e. Δ*15Nbulk N2O/ NO3*<sup>−</sup> for denitrification and Δ*15Nbulk NH4+* for nitrification. Some pure culture denitrification studies also reported the isotope effect between nitrite and N2O (Δ*15Nbulk N2O/ NO3*−), especially for fungal denitrification, but for field studies, we usually analyse soil NO3 <sup>−</sup>. By calculating isotope effects between N2O and N precursors one should be aware that the reaction progress changes the isotopic signature of the precursor: the more substrate is consumed, the more 15N enriched gets its residual pool. Therefore, the precursor N isotopic signature at the beginning and at the end of an experiment may differ depending on the reaction progress. Moreover, the δ*15N* of the measurable bulk N pools (by soil extraction) may deviate from the δ*15N* of the active N2O producing pools if the fractionating processes are heterogeneously distributed. This is especially the case in unsaturated soils where NO3 <sup>−</sup> in anoxic microsites is denitrified and thus progressively enriched in 15N, while in aerobic domains nitrification adds NO3 <sup>−</sup> at a lower <sup>15</sup>*Nbulk NO*<sup>3</sup><sup>−</sup> enrichment. Substantial deviation between bulk soil and active pool enrichment has been recently shown in tracer studies in the laboratory (Deppe et al. 2017) and in the field (Buchen et al. 2016). This indicates that the interpretation based on δ*15Nbulk* values is very complex and requires a good understanding of N transformation processes in the soil (see also Sect. 7.5).

The following endmember values can be considered:


*δ15Nsp* of the produced N2O is independent of the precursor isotopic signature. Hence, unlike δ*15Nbulk* , the endmember values are identical in δ*15Nsp* of the produced N2O. Therefore, the measured N2O δ*15Nsp* values can be directly compared with the following endmember values:


*δ18O* depends on the isotopic signature of several possible precursors: NO3 −, NO2 <sup>−</sup>, H2O and O2. For oxic processes like nitrification the incorporation of O2 is important (Snider et al. 2011) whereas for anoxic processes the O of the substrate or of soil water can be incorporated in N2O. Theoretically, during nitrification (hydroxylamine oxidation) O in N2O originates from O2 and during denitrification O from NO3 <sup>−</sup> should be transferred to N2O. But this is additionally complicated by the exchange of O atoms between soil water and denitrification intermediates (Kool et al. 2007). The extent of this exchange differs for various bacterial and fungal species (Rohe et al. 2017), but it has been shown recently that for soil incubations it is rather high (Lewicka-Szczebak et al. 2016). Therefore, soil water isotopic signatures show the largest impact on the final δ18O values of N2O, hence it was suggested to present the results as 18ON2O/H2O if dealt with denitrification (Lewicka-Szczebak et al. 2016). However, in pure culture studies, this rule works for fungal denitrification but not very well for bacterial denitrification where NO3 <sup>−</sup> plays an important role as a precursor for O atoms in N2O (Rohe et al. 2017). Because of different patterns for different processes, we present a summary of the measured, uncorrected δ18O values and additionally for denitrification we also show 18ON2O/H2O values.


in δ*18ON2O/O2* of −2.9‰. Hence, for this process, the δ*18ON2O* range of 23.5 ± 3‰ can be accepted (Frame and Casciotti 2010; Sutka et al. 2006). For the plots in Fig. 7.5, the δ*18ON2O* values are shown, which were determined in experiments utilising the air δ18OO2 of 23.5‰. For each case study where deviations from the typical O2 value are known (e.g. due to consumption in water column), these values should be expressed relative to the actually measured δ18OO2.

The most common way of identifying various N2O producing pathways is a graphical presentation of the measured values together with the literature endmember values. From the graphs, we can often identify the dominant pathway. To obtain more precise quantitative information, the contribution of a pathway (A) can be calculated based on the measured N2O isotopic signature (δN2O) using the **isotope mass balance**:

$$
\delta\_{N2O} = \delta\_{\text{pathway A}} \cdot a + \delta\_{\text{pathway B}} \cdot (1 - a) \tag{7.8}
$$

It must be noted that for this calculation (Eq. 7.8), the δN2O value may not be changed due to N2O reduction. This is only fulfilled if reduction is inhibited, measured to be negligible or included in calculations as described below. Using one isotope signature (δ*15Nbulk* , δ*15Nspor* δ*18O*), we are able to determine the mixing ratios of two pathways. Applying more isotopic signatures can theoretically enable quantification of more pathways. However, the results are not very exact due to the sometimes wide ranges of possible isotopic values for different pathways and overlapping of these ranges for more pathways. For both, δ*15Nsp* and δ*18O*, the ranges for heterotrophic bacterial denitrification and nitrifier denitrification. Additional interpretation of δ*15Nbulk* can further help but is often problematic due to lacking information on precursor isotope values (Lewicka-Szczebak and Well 2020). To increase precision of such calculations, controlled soil incubations with the soil under study may help to determine more narrow ranges of endmember values characteristic for the particular soil (Lewicka-Szczebak et al. 2017).

But besides the mixing processes also isotopic **fractionation during N2O reduction** changes the final isotopic value of the residual N2O. During N2O reduction to N2 (the last step of bacterial denitrification) preferentially the N-O bonds between light isotopes (14N and 16O) are broken and as a result the residual unreduced N2O is enriched in 15N<sup>α</sup> and 18O. In consequence, δ*15Nsp*, δ18O and δ*15Nbulk* values of residual N2O increase with progressing reduction. The magnitude of the shift towards higher values depends on the amount of reduced N2O and the isotopic fractionation factor associated with the N2O reduction. Hence, if we know the fractionation factor and the δ value of initially produced N2O before reduction (δ*0*), we can calculate the amount of reduced N2O and thereby determine the magnitude of N2 flux based on the measured δ value of the residual N2O after reduction (δ*r*). This is calculated according to the following isotopic fractionation Eqs. 7.9 to 7.11 by applying Rayleigh model that is valid for closed systems, either in its exact form (Mariotti et al. 1981):

7 Isotopic Techniques to Measure N2O, N2 and Their Sources 239

$$\frac{1+\delta\_r}{1+\delta\_0} = \left(r\_{N2O}\right)^{\varepsilon\_{N2-N2O}}\tag{7.9}$$

or in simplified, approximated form:

$$
\delta\_r \cong \delta\_0 + \varepsilon\_{N2-N2O} \cdot \ln(r\_{N2O}) \tag{7.10}
$$

where δ<sup>r</sup> is the residual N2O isotopic signature, after reduction, δ<sup>o</sup> is the initial N2O isotopic signature, before reduction, ε*N2-N2O* is the isotopic fractionation factor associated with N2O reduction and *r*N2O is the residual unreduced N2O fraction (*r*N2O = *y*N2O/(*y*N2 + *y*N2O); (*y*: mole fraction))

The application of the closed system model has been confirmed by several studies (Köster et al. 2015; Lewicka-Szczebak et al. 2017, 2014). However, it was also suggested that an isotopic fractionation model for open systems could be suitable (Decock and Six 2013), which is associated with smaller apparent isotope effects during N2O reduction:

$$
\delta\_r = \delta\_0 - \varepsilon\_{\rm red} \left( 1 - r\_{N\_2O} \right) \tag{7.11}
$$

To be able to determine *r*N2O from N2O isotopic values of individual samples according to the above equations, isotopic fractionation factor associated with N2O reduction to N2 (ε*N2*-*N2O*) must be known. They were determined in numerous studies in controlled soil incubations (Jinuntuya-Nortman et al. 2008; Lewicka-Szczebak et al. 2014; Menyailo and Hungate 2006; Ostrom et al. 2007; Well and Flessa 2009) and the following ranges were obtained:


In the summary, we disregarded one study which provided an inverse isotope effect for ε15Nbulk *N2*-*N2O* and ε18O*N2*-*N2O* (Lewicka-Szczebak et al. 2014). These values might have been a result of untypical reduction conditions in the experiment or an experimental artefact (Denk et al. 2017), therefore, they are neglected here. From the study of Lewicka-Szczebak et al. (2015) only the data of moderate reduction (from Pool1) were summarised here, because it was shown that by very intensive reduction the results can be strongly affected by N2O diffusion. This depends on the balance between diffusive and enzymatic fractionation during N2O reduction (Lewicka-Szczebak et al. 2014). By nearly complete N2O reduction, we observe a relatively large impact of diffusive N2O fractionation, resulting in residual N2O more depleted in heavy isotopes, hence the apparent isotope effects are significantly lower, i.e. −2.7‰, −1.5‰, and −2.0‰ for ε15Nbulk *N2*-*N2O*, ε15Nsp*N2*-*N2O*, and ε18O*N2*-*N2O*, respectively (Lewicka-Szczebak et al. 2015).

It is often problematic to separate the impact on the final N2O isotopic values by the mixing endmember for the produced N2O and by the isotopic fractionation due to N2O reduction. The interpretations and calculations based on N2O isotopic studies are difficult when we deal with the simultaneous variations in *r*N2O and δ<sup>0</sup> values. Usually, to calculate *r*N2O a stable δ<sup>0</sup> is assumed (Lewicka-Szczebak et al. 2015) and to precisely determine temporal changes in δ0, we need independent data on *r*N2O (Köster et al. 2015). In field studies, both *r*N2O and δ<sup>0</sup> cannot be determined precisely, but the possible ranges for each parameter can be given (Zou et al. 2014).

It is often attempted to distinguish between mixing and fractionation processes by using the changes in the isotopic signatures and their relations: δ*15Nsp*/δ*18O*, δ*15Nsp*/δ*15Nbulk* , δ*18O*/δ*15Nbulk* . These relations differ for the N2O reduction process and for mixing processes due to differences in the respective isotope effects. From literature data on N2O reduction fractionation factors (Jinuntuya-Nortman et al. 2008; Lewicka-Szczebak et al. 2014; Menyailo and Hungate 2006; Ostrom et al. 2007; Well and Flessa 2009) the following ratios are determined:


Although the range of possible ε*N2*-*N2O* variations is quite large, it has been shown recently that the mean values and typical ε15NspN2-N2O/ ε18ON2-N2O ratios are well applicable for oxic or anoxic conditions unless N2O reduction is almost complete, *i.e*. the ratio N2O/(N2 + N2O) < 0.1, meaning more than 90% of N2O was reduced (Lewicka-Szczebak et al. 2015).

For comparison, here are the relations between isotopic signatures of emitted N2O resulting from mixing processes calculated based on literature ranges for mixing endmembers given above. Because of the overlapping endmember ranges, we cannot distinguish between all individual pathways, and we determine the slopes of mixing lines between selected endmember values (Figs. 7.5 and 7.6) as follows:

	- *δ15NSP/ δ18O* from −10.5 to 4.8 with a mean of 6.1;
	- *δ15NSP/ δ15Nbulk* from −4.6 to −0.5 with a mean of −1.2;
	- *δ18O/ δ15Nbulk* from −1.0 to 0.1 with a mean of −0.1.
	- *δ15NSP/ δ18O* from 1.1 to 1.4 with a mean of 1.3;
	- *δ15NSP/ δ15Nbulk* from −3.9 to 7.9 with a mean of −2.7;
	- *δ18O/ δ15Nbulk* from −2.8 to 6.4 with a mean of −2.2.

Fungal denitrification cannot be distinguished from relations including δ15Nbulk because of the overlapping range with bacterial denitrification (see Fig. 7.5). Anyway, relations including δ15Nbulk are difficult to use due to dependence of this isotope value on the precursor, which differ for nitrification and denitrification. Here

the relationships must be determined with isotope effect for δ15Nbulk, i.e. using Δ*15Nbulk (N2O/NH<sup>+</sup> <sup>4</sup> )* for nitrification and Δ*15Nbulk* ( *N2O/NO*<sup>−</sup> *<sup>3</sup>* ) for denitrification (see x-axis in Fig. 7.5). Often the isotopic signatures of the precursors are not known, which make the interpretation of δ15Nbulk values rather ambiguous. Nevertheless, some studies apply the δ*15Nsp*/ δ*15Nbulk* isotope maps for distinction of mixing and fractionation processes, but for such isotope maps, systematic changes in δ*15Nbulk* induced by systematic changes in the N isotopic composition of one of the precursors NH4 <sup>+</sup> or NO3 <sup>−</sup> could be misinterpreted as reduction events (Well et al. 2012; Wolf et al. 2015). Hence, the careful monitoring of precursor isotopic signatures is needed (Zou et al. 2014).

A *δ15Nsp***/** *δ15Nbulk* **isotope mapping approach** allowing for assessment of minimal and maximal reduced N2O fraction and nitrification and denitrification mixing ratios was proposed by Zou et al. (2014) (Fig. 7.5). Such an approach is most often used for distinguishing between nitrification and bacterial denitrification only. However, other cases have also been analysed (Zou et al. 2014). The calculation method presented (Fig. 7.5) assumes first mixing of N2O from different endmembers and afterwards its partial reduction. Two mixing lines are defined–for the minimum and maximum values for both endmembers as well as two reduction lines–with maximal and minimal slope. From the intercept 1 the maximal denitrification contribution is determined whereas from the intercept 2 the minimal one. Based on the difference between the sample point and intercept 1 or 2 the reduction contribution, respectively, maximal and minimal, is also determined. However, it must be noted that in case of significant admixture of fungal denitrification or nitrifier denitrification the results may be biased.

The application of *δ***15Nsp/ δ18O isotope mapping approach** may be easier since δ15Nsp and δ18O values are more stable in time (Lewicka-Szczebak et al. 2017; Wu et al. 2019), and δ18O values show narrower endmember ranges when compared to δ15N values. The distinction of mixing and fractionation processes is based on the different slopes of the mixing lines and the reduction line (Fig. 7.6).

Isotopic values of the samples analysed are typically located between these two, reduction and mixing, lines. Here we defined only one mixing line for the median values of bacterial and fungal denitrification and one reduction line with a mean slope. From sample's, location, we can estimate the impact of fractionation associated with N2O reduction and admixture of N2O originating from fungal denitrification. We can deal with two scenarios:


While both scenarios yield identical results for the admixture of N2O from fungal denitrification, the resulting reduction shift, and hence the calculated *rN2O* value, is higher when using Scenario 2. It is still not clear which scenario is more realistic. The uncertainty analysis of this method has been recently presented by Wu et al. (2019) and this approach has been successfully applied in the field case studies (Buchen et al. 2018; Ibraim et al. 2019; Verhoeven et al. 2019). However, after the appearance of those publications, it has been found that other δ18O values should be applied for nitrification (Yu et al., 2020). This summary reports the most current choice of endmember ranges, which differ from those presented recently (Buchen et al. 2018; Ibraim et al. 2019; Lewicka-Szczebak et al. 2017; Verhoeven et al. 2019; Wu et al. 2019).

## *7.3.3 Analysis of N2O Isotopocules by IRMS*

The most common method for N2O isotopocule analysis is isotope ratio mass spectrometry (IRMS). In order to perform N2O isotopic analysis the gas samples need to be purified, and N2O must be separated and pre-concentrated. First, water and CO2 are removed by chemical traps, and then N2O is concentrated with liquid N traps. Afterwards, the gases are separated with gas chromatography and finally introduced in the isotope ratio mass spectrometer.

In the mass spectrometer, N2O isotopocule values are determined by measuring *m/ z* 44, 45 and 46 of the intact N2O+ ions as well as *m/ z* 30 and 31 of NO+ fragment ions. This allows the determination of average δ15N (δ15Nbulk), δ15N<sup>α</sup> (δ15N of the central N position of the N2O molecule), and δ18O (Toyoda and Yoshida 1999). δ15N<sup>β</sup> (δ15N of the peripheral N position of the N2O molecule) is calculated from δ15Nbulk = (δ15N<sup>α</sup> + δ15N<sup>β</sup>) / 2 and 15N site preference (δ15Nsp) from δ15Nsp = δ15N<sup>α</sup>–δ15N<sup>β</sup>. Since the IRMS approach was developed simultaneously by two groups (Brenninkmeijer and Röckmann 1999; Toyoda and Yoshida 1999), two different nomenclatures had been introduced for the two positions of N2O-N. Hence, in some studies, the peripheral (β) position is referred to as 1- and the central (α) as 2-position (Brenninkmeijer and Röckmann 1999). The scrambling factor resulting from the exchange of 15N atoms on the ion source must be taken into account. The magnitude of the scrambling factor should be determined individually for each mass spectrometer (Röckmann et al. 2003). Also, 17O-correction should be taken into account, because 17O substitution is indistinguishable from 15N, therefore typical terrestrial 17O content (0.528) is assumed (Kaiser and Röckmann 2008).

Up to now, there are still no internationally agreed gaseous N2O reference materials for N2O isotopocule analyses. Usually, the laboratories calibrated pure N2O gas for isotopocule analyses in the laboratory of the Tokyo Institute of Technology according to the method of Toyoda and Yoshida (1999). Recently, the first interlaboratory comparison has been performed and now the standards from this study (REF1, REF2) are available for the laboratories and allow the performing of two-point calibration for δ15Nsp values (Mohn et al. 2014). This intercalibration study has shown that the two-point calibration method is necessary to obtain accurate δ15Nsp values. Recently, two N2O standards had been tested in a further interlaboratory comparison (Ostrom et al. 2018) and is available from United States Geological Survey (USGS).

The sample volume needed for the N2O isotopocule depends on the concentration and is about 100 ml for ambient N2O concentration samples (about 300 ppb) and about 10 ml for N2O concentration of above two ppm.

## *7.3.4 Laser Spectroscopic Analysis of N2O Isotopomers to Differentiate Pathways*

The invention and availability of non-cryogenic light sources in the mid-infrared (MIR) spectral range (Brewer et al. 2019) coupled with different detection schemes such as direct absorption quantum cascade laser absorption spectroscopy (QCLAS) (Mohn et al. 2010, Mohn et al. 2012, Wächter et al. 2008), cavity ring down spectroscopy (CRDS) (Erler et al. 2015) and off-axis integrated-cavity-output spectroscopy (OA-ICOS) (Wassenaar et al. 2018) has provided sensitive and fielddeployable laser spectroscopic analysers for N2O isotopocule analysis. These instruments can analyse the N2O isotopic composition in gaseous mixtures (e.g. ambient air) in a flow-through mode, providing real-time data with minimal or no sample pre-treatment, which is highly attractive to better resolve the temporal complexity of N2O production and consumption processes. Most importantly, MIR laser spectroscopy is selective for 17O, 18O and position-specific 15N substitution due to the existence of characteristic rotational-vibrational spectra (Gordon et al. 2017).

Therefore, laser spectroscopy has the potential to open a new field of research in the N2O biogeochemical cycle, but, applications remain challenging and are still scarce for the following main reasons: (1) laser spectrometers as any analytical instrument are subject to drift effects, in particular under fluctuating environmental conditions, limiting their performance (Werle et al. 1993); (2) changes in N2O concentration affect N2O isotope results when using the δ-calibration approach (Griffith 2018); (3) laser spectroscopic results are affected by mole fraction changes of atmospheric background gases (N2, O2, and Ar), called gas matrix effects, due to the difference of pressure-broadening coefficients, and potentially by spectral interferences from other atmospheric constituents (H2O, CO2, CH4, and CO, etc.), called trace gas effects, depending on the wavelength region used in an instrument. Spectral interferences are particularly pronounced for N2O due to its low atmospheric abundance in comparison to other trace gases; (4) only since recently two pure N2O isotopocule reference materials (USGS51, USGS52) have been made available through the United States Geological Survey (USGS) (Ostrom et al. 2018), which was identified as a major reason limiting interlaboratory compatibility (Köster et al. 2013; Mohn et al. 2014, 2016).

In a recent study, the most common commercially available N2O isotope laser spectrometers were carefully characterised for their dependence on N2O concentration, gas matrix composition (O2, Ar) and spectral interferences caused by H2O, CO2, CH4 and CO to develop analyser-specific correction functions. In addition, the authors suggest a step-by-step workflow that should be followed (Fig. 7.7) by researchers to acquire trustworthy N2O isotopocule results using laser spectroscopy (Harris et al. 2020).

## *7.3.5 Hands-on Approach to Use a CRDS Isotopic N2O Analyser*

#### *Introduction*

As an example, the Picarro G5101-*i* analyser can be used to determine N2O concentration, 15Nbulk isotope ratios and isotopomer values (15N<sup>α</sup> and 15N<sup>β</sup>) by continuous or discrete sample measurement. Small volume discrete samples (≤20 ml) can be measured using the SSIM (small sample isotope module) (see also Sect. 5.3.) peripheral unit in conjunction with the Picarro G5101-*i* analyser. The G5101-*i* analyzer is the predecessor of the current G5131-*i* analyzer which also measures δ18O in addition to δ15Nbulk, δ15N<sup>α</sup> and δ15N<sup>β</sup>. The SSIM can also be used to dilute samples. Larger volume samples (e.g. Tedlar bags) can be measured by direct input into the G5101-*i* analyser or through the 16-port distribution manifold. The 16-port distribution manifold allows for partial automation of measurement and can be used in conjunction with the SSIM for smaller volume samples (see also Fig. 5.5 that illustrates the coupling of a 16-port manifold and a SSIM). The SSIM can also be used to dilute samples.

#### *Principle*

Samples are measured using mid-IR laser by CRDS (cavity ring down spectroscopy). Measurement precision increases with measurement time. Several options are available for delivery of N2O samples into the analyser and how long measurements take. Sample volume and the required precision of measurements should be considered to decide which operational set up is the most appropriate.

#### *Apparatus*



	-
	-


**Fig. 7.7** Workflow to acquire trustworthy N2O isotopocule results using laser spectroscopy (Harris et al. 2020)

## *Consumables*


## *Sampling*

For discrete gas samples follow a suitable sampling procedure as outlined by De Klein and Harvey (2012). Small volume samples (≤20 ml) should be stored in septum capped vials, ensuring to overpressure when filling to prevent inward contamination by ambient air. Vials can be stored in a cool dry place. Larger volume samples in Tedlar bags should be measured ASAP as storage reliability decreases greatly after 24–48 h.

## *Operational Procedure*


## *Discrete samples (*≤*20 ml)*

• To measure small volume discrete samples (≤20 ml) allow the laser to run continuously for 24 h to ensure that the laser has been given sufficient time to stabilise.

	- Click OK. Select G5101-*i* for reference standard.
	- SSIM pressure measurements should be available in the GUI data key dropdown tabs. Select this parameter to monitor SSIM pressure visually on the left side of the GUI.

**Note**: Under vacuum, the SSIM pressure should be ~8 Torr or below. When a sample is injected the max pressure is reached upon filling the cavity with sample/ ZA. The max pressure should read between 980 and 1000 Torr. If the pressure is too high down-regulate the second pressure regulator. If the pressure is too low up-regulate the second pressure regulator being very careful not to exceed 8 psi.


**Note**: The calibration curve should be almost perfectly linear with a R2 = 0.99 +. Deviations from the curve or lower R2 values may indicate a leak. Check the injector nut and septum, change septum if necessary. Check the ZA line connections from the SSIM unit to the analyser. Tighten loose connections if necessary by finger tightening + ¼ turn with an adjustable spanner. **Never over tighten as this can lead to leaks**.


**Note**: A system alarm will probably appear on the top left of the GUI and a message stating "Pressure unlocked". This results from the SSIM being disconnected. To resolve: click Shutdown and select "Stop Analyser Software Only". Wait a couple of minutes and relaunch the analyser software by double clicking on the Picarro Switcher Mode icon on the desktop and selecting G5101-*i* Isotopic N2O and click launch. Monitor the system as it relaunches and wait until it begins measuring N2O parameters.

• To turn off the instrument completely click shutdown on the GUI.

#### **NB: Never leave the analyser measuring ZA overnight, this will lead to drift.**

#### *Expression of Results*


#### *Quality Assurance*

• Prior to taking gas samples in the field (i.e. from static chamber) ensure vials are properly sealed and that they have been flushed and evacuated three times.


#### *Reporting of Results*

Raw data files are automatically generated by the analyser and are stored on the instrument's computer as a DataLog\_User file. These raw data files can be found by following the file path: C:\UserData\DataLog\_User\Year\Month\Day. An example of the file naming convention is JBDS5030-20170331-140739Z-DataLog\_User. JBDS5030 refers to the instrument serial number. 20170331 is the Year, Month and Date the file was started. 140739 is the Hour, Minute and Second of when the file was started. There are a number of values available for the N2O parameters measured. The dry corrected values are the appropriate values to select for analysis.

When measuring discrete samples using the SSIM there is sufficient time between samples to record the real-time values on a separate spreadsheet that has been premade with sample reference numbers included.

#### *Safety*


• Read and follow the information in the Risk Assessments for the Stable Isotope Analysis lab.

#### *Trouble Shooting*

## • **Start-up:**

If the chiller line contains large air bubbles this may stop the circulation of water in the line. This can lead to the baseplate temperature being exceeded which causes the analyser to enter safe mode (error message appears in GUI). This problem should be avoided by keeping the cooling agent LIQ-702 (propylene glycol) in the buffer tank (externally mounted on the chiller cover) topped up to 90% of its full volume with deionised water. To do this unscrew the black cover and use a wash bottle to add in fresh deionised water. This can be done while the analyser is running. If the error message does appear this may require the instrument to be shut down and for the chiller line be flushed following the instructions provided in the installation manual for the installation of the water buffer tank.

## *7.3.6 Accuracy, Precision and Bias*

The analytical precision for IRMS measurements determined as standard deviation (1σ) of the internal standards for measurements of δ*15Nbulk ,* δ*18O and* δ*15Nsp* is typically 0.1, 0.1 and 0.5‰, respectively. Commercially available laser spectrometers at ambient N2O concentrations offer a precision of 0.2 to 1 ‰ for δ15N<sup>α</sup>, δ15N<sup>β</sup> and δ18O, which can be reduced to 0.1 ‰ at higher concentrations, or by using a preconcentration device. However, from the inter-comparison study, we see that the bias may be much larger, up to: for δ*15Nbulk* 0.8 and 2.8‰, and for δ*15Nsp* 4.3 and 3.7‰ for mass spectrometry and for laser spectroscopy, respectively (Mohn et al. 2014). But these potentially large errors can be minimised by a proper data calibration using two points standardisation with the reference gases that bracket the measurement range. Care must be also given when samples with high concentrations are diluted as the dilution matrix (typically Helium or N2) may apparently have an impact on the final result. The rule of identical treatment of standards and samples should be held, including identical dilution matrix and similar concentration range (Mohn et al. 2014).

Possible bias is also associated with calculations applied for data interpretation. Due to large ranges of literature data, the N2O source partitioning cannot be done precisely, but rather the ranges of possible results can be given. To increase precision of such methods controlled soil incubation can be applied to determine the soil specific endmember isotopic values or fractionation factors (Lewicka-Szczebak et al. 2017).

## *7.3.7 Examples of Laboratory Applications*

#### Köster et al. (2015)

This experiment applied an N2O isotopocule approach combined with conventional N2O and N2 flux measurements to study microbial pathways after different organic fertiliser applications. The direct determination of emitted N2 was used to take isotope effects during N2O reduction to N2 into account. The measured isotope signatures were corrected for isotope effects during N2O reduction with Eq. 7.10 using previously determined fractionation factor ranges. Based on the corrected values the isotope mass balance equations (Eq. 7.8) for δ*15Nsp* and δ*18O* were applied. The ranges for different pathways contribution were given for δ*15Nsp*- and δ*18O*based results and the common area for both was accepted as most probable. Two mixing scenarios were considered: bacterial denitrification and nitrification or bacterial and fungal denitrification. Although the range of possible results for endmembers contribution varied up to 30%, a clear increase in nitrification contribution with the incubation time has been documented.

Schorpp et al. (2016)

In this experiment, incubations with soil fauna were applied to check the impact on N2O and N2 emission of anecic earthworms and euedaphic collembola. Isotopocule approach was applied together with 15N tracing. Interpretation of the isotopocule results based on the δ*18O*-δ*15Nsp* isotope map, similar as presented in Fig. 7.6, including three possible mixing endmembers: bacterial and fungal denitrification and nitrification (hydroxylamine oxidation) and taking N2O reduction into account. Isotope data allowed concluding that the presence of collembolans shifted the process pathways towards bacterial denitrification although no change in N2O concentration could be noted.

Deppe et al. (2017)

In this incubation experiment high NH4 <sup>+</sup> concentrations in soil were established to check the supposed inhibition of nitrification. An isotopocule approach, together with 15N tracing and acetylene inhibition approach, was applied to gain insight into N2O production processes. Interpretation of the isotopocule results based on the δ*18O*-δ*15Nsp* isotope map, similar as presented in Fig. 7.6, including two mixing endmembers: denitrification and nitrification (hydroxylamine oxidation) and N2O reduction. This assumption of the mixing conditions appeared incorrect, since some data were located outside of the mixing and reduction lines. This indicated a substantial contribution of nitrifier denitrification and/or coupled nitrification-denitrification (10–40%) to total N2O production.

Cardenas et al. (2017)

Laboratory incubation was carried out at different saturation levels for a grassland soil and emissions of N2O and N2 were measured as well as the N2O isotopocules. Thanks to direct measurements of N2 flux, the extent of N2O reduction was known. Hence, the measured δ values were mathematically corrected to obtain the δ values of the produced N2O before reduction applying Eq. 7.9. An endmember mixing model (Eq. 7.8) was then used to calculate the percentage of bacterial N2O in the total N2O flux based on δ*15Nsp* and δ*18O*. To assess the uncertainty of this approach the ranges of possible endmembers isotopic signatures and reduction fractionation factors were taken into account. The variations of the bacterial N2O contribution due to assumed ranges of input values reached up to 40%. But still it allowed to distinguish the dominant pathways for different water saturation levels and indicated that only when the micropores become partially dry, the more aerobic soil conditions allow a higher contribution of nitrification. The dryer conditions in soil macropores did not result in significant changes in bacterial denitrification contribution.

## *7.3.8 Examples of Field Applications*

#### Toyoda et al. (2011)

N2O emitted from agricultural soils planted with rice, wheat, soybean, and vegetables, and treated with synthetic (urea or ammonium) and organic (poultry manure) fertilisers was analysed. The observed isotopic values for Δ*15N* and δ*15Nsp* were compared with literature endmembers of nitrifying and denitrifying bacteria. A characteristic relationship between δ*15Nbulk* and δ*15Nsp* during N2O reduction by denitrifying bacteria was used to quantify N2O reduction. The relative fraction of N2O derived from nitrification and the approximate progress of N2O reduction were calculated by a Monte Carlo method. Different scenarios for pairs of mixing endmembers were tested (nitrification and denitrification; nitrification and nitrifier–denitrification; fungal denitrification and denitrification; fungal denitrification versus nitrifier–denitrification) but due to overlapping ranges for δ*15Nsp* values it was chosen to consider only the mixing between bacterial nitrification and denitrification. It was found that the contribution from nitrification was relatively high (40%–70%) in soils amended with synthetic ammonium fertiliser, while denitrification was dominant (50%–90%) in the same soils amended with poultry manure.

#### Kato et al. (2013)

In this study, field samples from static flux chambers located on alpine meadow, shrub and wetlands were collected and analysed. Interpretation of results based on the relationship between δ*15Nbulk* and δ*15Nsp* (similar as presented in Fig. 7.5). A mixing of two endmembers was assumed: bacterial and fungal denitrification and subsequent N2O reduction. Applying literature values for endmembers and fractionation during reduction the contribution of fungal denitrification (from 23 to 41%) and degree of reduced N2O (from 83 to 93%) was calculated. The calculations were performed with Monte Carlo simulations and the assessed uncertainty of the results ranged from 17 to 23% for contribution of mixing endmembers and from 10 to 19% for degree of reduced N2O.

#### Zou et al. (2014)

Soil gas was collected from a highly fertilised tea field at 10–50 cm depths using a silicone tube. δ*15Nsp* –Δ*15Nbulk* isotope maps (Fig. 7.5) were applied for interpretations. The precursor isotopic signatures were determined, and the endmember ranges have been recalculated according to the measured precursor values for bacterial and fungal denitrification, nitrification and nitrifier denitrification. For the N2O reduction two scenarios were taken into account: assuming reduction after mixing and applying closed system dynamics and assuming reduction preceding mixing and applying open system dynamics. Predictions of δ*15Nsp* values for different scenarios, reduction degrees and mixing ratios were presented and compared to the measured results. The study identified the bacterial denitrification as the dominant process and allowed for indication of the particular events when the contribution of nitrification or fungal denitrification increased pronouncedly.

Wolf et al. (2015).

N2O isotopic analyses were done directly from the atmospheric surface layer (at 2.2 m height) applying a laser spectrometer connected to an automated N2O preconcentration unit. The isotopic signatures of soil-emitted N2O were derived using the Keeling plot approach, where δ values measured in the atmosphere surface layer are plotted versus the inverse of N2O mole fractions (for a background on Keeling plot analysis see Pataki et al 2003). The intercept of the linear regression line is interpreted as the isotopic composition of soil-emitted N2O. The interpretation of the results is based on isotope maps of δ*15Nsp* vs. δ*15Nbulk* and δ*15Nsp* vs. δ*18O*. These isotope maps allowed concluding that N2O was predominately formed by bacterial denitrification and that variations in isotopic composition may have been caused predominately by N2O reduction to N2. The study did not attempt to quantify the mixing ratios or N2O reduction. The high-frequency isotope data was combined with a biogeochemical model Landscape DNDC with a stable isotope model for nutrient cycles (SIMONE) to identify and address weaknesses in N cycling of the model (Denk et al. 2019).

## *7.3.9 Outlook*

N2O isotopocule analyses provide a unique possibility to get insight into processes contributing to N2O production as well as to assess the magnitude of N2O reduction and thereby also N2 flux. However, the information is still rather indicative than strongly quantitative. The calculation methods presented allow estimates of ranges of possible mixing ratios and reduction contribution rather than precise numbers. However, such information is also quite precious hence often not attainable by any other methods. 15N tracing, which is often a more precise tool, is much more expensive and laborious, moreover applicable only on a very limited space and time scale, hence much more constrained in application potential.

A promising perspective is to apply the N2O isotopocule analyses in combination with other methods, like with 15N tracing (Deppe et al. 2017; Schorpp et al. 2016) (see also Sect. 7.5.) or with process modelling (Bai and Houlton 2009; Denk et al. 2017) which vastly increases the interpretation potential of such studies. Moreover, more quantitative estimates can be expected if the isotopocule approach is calibrated using controlled incubations where endmember values and isotopic fractionation factors are determined for specific conditions using independent estimates of contributing processes, e.g. by direct measurement of N2 production or 15N tracing (Lewicka-Szczebak et al. 2017; Wu et al. 2019). The most recent idea for interpretation of N2O isotope data is the application of a N2O isotopocule model which incorporates all three measured isotopic signatures (δ15Nbulk, δ15Nsp and δ18O) (Lewicka-Szczebak and Well 2020).

## **7.4 Dual Isotope Method for Distinguishing Among Sources of N2O**

Various microbial processes can produce N2O (for a simplified overview, see Fig. 7.8). These may occur simultaneously in distinct soil microhabitats or take

**Fig. 7.8** Overview of N2O producing processes carried out by nitrifiers and denitrifiers in soils. Shown is also the use of O2 versus H2O as source of oxygen. Please note that the distinction between nitrification-coupled denitrification and fertiliser denitrification is purely methodological, as the organisms and pathways are identical, but the source of NO3 − differs. Within brackets, abbreviations of the pathways as used in the text are shown. DNRA: Dissimilatory NO3 − reduction to NH3 (Wrage-Mönnig et al. 2018)


place temporally separated with fluctuating soil conditions. Often, only nitrification and denitrification are considered to be the main sources. However, the methods often applied cannot distinguish among all sources. For example, using 15N tracing with labelled ammonium or NO3 <sup>−</sup> does not allow a distinction among N2O produced by nitrifiers either via hydroxylamine (termed here nitrification, N) or via nitrite reduction (nitrifier denitrification, ND), or by denitrifiers using NO3 <sup>−</sup> produced by nitrifiers (nitrification-coupled denitrification, NcD). All N2O produced by these sources is summarised as 'nitrification' by authors using this method. No method based on 15N alone can so far separate the sources shown in Fig. 7.5. However, a distinction is important, as ND can under certain conditions produce all N2O derived from NH4 + and has been reported to cause up to 90% of total N2O emissions (Kool et al. 2010).

A distinction between ND and other sources of N2O is possible if 18O labelling is used in addition to 15N labelling (Kool et al. 2011). As seen in Fig. 7.8, nitrifiers use distinct sources of O2 for the oxidation of NH3 and the subsequent oxidations of NH2OH and NO2 <sup>−</sup>. This is used in the dual isotope method, where 18O-labelled H2O is applied on top of 15N tracers. However, care has to be taken to account for O-exchange, which can occur between H2O and N oxides in all reactions of N oxides shown in Fig. 7.8. This is accomplished using the enrichment ratio retention (ERR) approach, where the enrichment ratio of 18O:15N of N2O is compared to that of NO3 <sup>−</sup> in incubations with either 15N- or 18O-labelled NO3 <sup>−</sup>. Then, we can differentiate among N2O produced by N, ND, NcD and fertiliser denitrification (FD).

The preparation of soil samples proceeds in a similar way as for other stable isotope methods. However, one has to keep in mind that water needs to be added as tracer, so that the water content during conditioning needs to be a bit less than intended for the incubation. So far, conditioning has been done at 40% water-filled pore space (of samples dried at 40°C), and incubation at 80%, but this is adaptable as long as the requirements for tracer additions are kept. Soil samples of 75–100 g have been incubated in glass jars of about 300 ml for 24–28 h. These ratios and times may be adapted, but care must be taken to ensure linear N2O production over the incubation period, as well as stable concentrations of substrates (including O2 and H2O). Much longer incubations are difficult, as the 18O enrichment of the soil H2O might change locally due to evaporation and addition of H2O. The occurrence of NO3 <sup>−</sup> assimilation and DNRA needs to be checked (indicated by enrichment of NH4 <sup>+</sup> in incubations with 15N-NO3 <sup>−</sup>) and accounted for if necessary, as in other 15N methods.

The treatments (TR) are established (Table 7.5) with proper replication (at least five times) after conditioning of the soil as needed. So far, added label has been


**Table 7.6** Oxygen sources of N2O in the different processes and pathways distinguished using the dual isotope method

enriched at 1.0 atom% for 18O and 40 atom% for 15N, but higher enrichments may be desirable to reduce the amount of substrates added, especially concerning the N-substrates in natural systems. Usually, 100 mg N kg−<sup>1</sup> soil has been applied, half each as NO3 <sup>−</sup> and as NH4 +. When applying less NO3 <sup>−</sup>, one has to consider that if NO3 <sup>−</sup> becomes limiting, the underlying assumption of the method that only NO3 − and no NO2 <sup>−</sup> is used in NcD and FD becomes invalid. This would result in an underestimation of NcD and an overestimation of ND. Additional incubations with 18O-NO2 <sup>−</sup> (which is currently not commercially available, though) or analysis of the 18O enrichment of the NO2 <sup>−</sup> pool may help to overcome this.

Immediately after establishing the treatments, the jars are closed, and samples are taken for N2O content and isotopic signature as explained in Chap. 3 and above. At the end of the incubation, soil samples are taken for analysis of mineral N and its isotopic signature (the latter only in TR 3 and 4), as well as the soil moisture content to verify that this did not change during incubation. Consider that the label added with 18O-H2O is diluted when mixed with moist soil.

For quantifying the O-exchange between N oxides and 18O-H2O, the ERR approach is used. It is assumed that the O-exchange is similar for denitrifiers and nitrifiers. This need not be true, as O-exchange by nitrifiers has often been found to be less than in denitrifiers. Such a discrepancy would lead to an underestimation of the N2O produced by ND and NcD. No O-exchange is assumed to affect N2O derived from N. The ERR is calculated in Eqs. 7.12 to 7.17 as follows:

$$ERR[\%] = 100 \frac{^{18}\text{O}(\text{N}\_2\text{O}\_{TR2})}{^{15}\text{N}(\text{N}\_2\text{O}\_{TR3})} / \frac{^{18}\text{O}(\text{NO}\_{3\_{TR2}}^{-})}{^{15}\text{N}(\text{NO}\_{3\_{TR3}}^{-})} \tag{7.12}$$

where 18O(YTRx) and 15N(YTRx) denote the 18O or 15N enrichment, respectively, of substance Y from treatment x. Without O-exchange, ERR is 100%. O-exchange (Oex) is then quantified as

$$O\_{ex} = 100 - ERR \tag{7.13}$$

Next, the percentage of N2O derived from NO3 − N2ONO<sup>−</sup> 3 and NH4 + N2ONH<sup>+</sup> 4 is calculated. N2ONO<sup>−</sup> <sup>3</sup> is defined as N2O from FD, whereas N2ONH4+ comprises the other three sources.

$$\mathrm{N\_2O\_{NO\_3^-}}[\%] = FD = 100 \frac{^{15}\mathrm{N(N\_2O\_{TR3})}}{^{15}\mathrm{N(N\_2O\_{TR3})} + ^{15}\mathrm{N(N\_2O\_{TR4})}} \tag{7.14}$$

$$\mathrm{N\_2O\_{NH\_4^+}}[\%] = \mathrm{N} + \mathrm{ND} + \mathrm{NcD} = 100 \frac{^{15}\mathrm{N(N\_2O\_{TR4})}}{^{15}\mathrm{N(N\_2O\_{TR3})} + ^{15}\mathrm{N(N\_2O\_{TR4})}} \quad (7.15)$$

If the 15N enrichment of N2O in TR4 does not exceed the 15N enrichment of NO3 − in the same treatment, all N2ONH<sup>+</sup> <sup>4</sup> might have been derived from NcD (maximum contribution of NcD, NcDmax = N2ONH<sup>+</sup> <sup>4</sup> , implying that ND and N were equal to zero). If not, NcDmax is calculated as follows:

$$\mathrm{NcD}\_{\mathrm{max}}[\%]=\mathrm{N}\_{2}\mathrm{O}\_{\mathrm{NH}\_{4}^{+}}\times\frac{^{15}\mathrm{N(NO}\_{3\tau\_{\mathrm{TR}}}^{-})}{^{15}\mathrm{N(NO}\_{3\tau\_{\mathrm{Rx}}}^{-})+^{15}\mathrm{N(NH}\_{4\_{7\mathrm{R}}}^{+})},\ \mathrm{if}\ ^{15}\mathrm{N(N}\_{2}\mathrm{O}\_{\mathrm{TR}4})>^{15}\mathrm{N(NO}\_{3}^{-}\mathrm{T}\_{\mathrm{R}4}).\tag{7.16}$$

To distinguish among the other N2O producing pathways considered, the actual O incorporation from H2O into N2O (AOI) is determined from TR1:

$$AOI\{\%\} = 100 \frac{\,^{18}O(N\_2O\_{TR1})}{\,^{18}O(H\_2O\_{TR1})} \tag{7.17}$$

This AOI may come from Oex quantified as shown above and the reaction stoichiometry of the different pathways as shown in Table 7.6.

A large AOI may thus be caused either by a larger contribution of pathways with a larger incorporation of 18O-H2O (ND or NcD) or by a larger Oex. For further evaluation, Oex is maximised, i.e. assumed to take place in the NH4 +-derived pathways to the same extent as in FD (Scenario A) or minimised, i.e. assumed to be absent in nitrification pathways (Scenario B). Furthermore, in Scenario A, the contributions of N and NcD are maximised, while in Scenario B, ND is maximised. Under both scenarios, a theoretical O incorporation (TOI) is calculated and compared to the AOI.

Under Scenario A, the TOI (TOIA) is calculated (Eq. 7.18) as

$$TIOI\_A = N\_2O\_{NO\_3^-} \times O\_{ex} + NcD\_{\text{max}} \left(2/3 + 2/\Im O\_{ex} - 1/\Im (O\_{ex})^2\right) \tag{7.18}$$

This calculation comprises Oex occurring during D (N2ONO<sup>−</sup> <sup>3</sup> × Oex) as well as from NcD stoichiometry (2/3 O from H2O) and Oex occurring during N to NO3 − and NcD (for further explanation, see Appendix 1 in Kool et al. 2009). If TOIA ≥ AOI, no contribution by ND is necessary to explain the AOI. The minimal contribution of ND, NDmin, is then set to zero, and the maximum contribution of N, Nmax = N2ONH<sup>+</sup> <sup>4</sup> −NcDmax. If not, ND must have contributed to N2O production (NDmin > 0), which implies at the same time a maximum contribution of N, Nmax Nmax < N2ONH<sup>+</sup> <sup>4</sup> <sup>−</sup>NcDmax . In this case, we can calculate the contribution of NDmin (Eq. 7.19) as follows:

260 M. Zaman et al.

$$ND\_{\rm min} = \frac{AOI - (N\_2O\_{NO\_3^-} \times O\_{ex})}{2/3 + 2/3O\_{ex} - 1/3(O\_{ex})^2} - NcD\_{\rm max} \tag{7.19}$$

Nmax is then equal to N2ONH<sup>+</sup> <sup>4</sup> −NcDmax−NDmin.

Under Scenario B, ND is maximised by assigning N2ONH<sup>+</sup> <sup>4</sup> to ND and assuming no Oex during this pathway, and in Eq. 7.20 TOIB is calculated as

$$TIO\_B = N\_2O\_{NO\_3^-} \times O\_{ex} + N\_2O\_{NH\_4^+} \times 0.5\tag{7.20}$$

If TOIB > AOI, not all N2ONH<sup>+</sup> <sup>4</sup> can have been derived from ND NDmax < N2ONH<sup>+</sup> 4 . In that case, some N2O must have originated from N (i.e. the minimum contribution of N, Nmin > 0), which will lower the TOI. However, if TOIB ≤ AOI, all N2ONH<sup>+</sup> <sup>4</sup> may have come from ND NDmax = N2ONH<sup>+</sup> 4 and the contribution of Nmin was zero. A larger AOI (TOIB < AOI) may either come from a contribution of NcD or Oex during ND, which was assumed not to take place under this scenario. As both may equally well explain the numbers, NcDmin is set to zero in this case and Oex assumed to have occurred during ND.

If Nmin was found to be larger than zero, we can calculate NDmax from this scenario as follows (Eq. 7.21):

$$ND\_{\text{max}} = \frac{AOI - N\_2O\_{NO\_3^-} \times O\_{ex}}{0.5} \tag{7.21}$$

In that case, Nmin = N2ONH<sup>+</sup> <sup>4</sup> −NDmax.

Thus, in the dual isotope method, the contribution of NcD is always maximised, and consequently minimum and maximum contributions of N and ND are estimated based on Scenario A and B. Applying this method allows insight into these three potential sources of N2O plus fertiliser denitrification. However, in soils, further microbial processes can lead to N2O production. In the following, we will briefly discuss potential effects of nitrification by heterotrophs and archaea, fungal denitrification, as well as DNRA and co-denitrification.

If N2O of nitrification by heterotrophs and archaea is produced by the same sources and similar pathways as in autotrophic nitrifiers, this should not interfere with the calculations. The contribution of N would then comprise that of other nitrifiers. However, archaea have also been suggested to produce N2O in a pathway similar to ND (Jung et al. 2014). If so, this would be included in the contribution of ND. However, the pathway of N2O production by archaea is not clear yet and needs further study (Stieglmeier et al. 2014), the outcome of which will also affect the calculations presented here. In soils where fungal denitrification occurs, this is counted as FD using the dual isotope method, if the fungi use added NO3 <sup>−</sup> as a source in a reaction similar to denitrification. Fungal denitrification may be quantified using the isotopomer method (Sect. 7.3), calling for a combination with the dual isotope method.

The occurrence of DNRA should be tested for as explained above. Should it lead to N2O production (Stevens et al. 1998), this would lead to an overestimation of N2O from FD. As DNRA may be important in soils (Rütting et al. 2011), this pathway should always be considered by checking for enriched NH4 <sup>+</sup> in incubations with added 15N-NO3 −.

Co-denitrifiers combine NO3 <sup>−</sup> or NO2 <sup>−</sup> with other nitrogenous compounds to produce N2O or N2. The occurrence of such a process could be quantified using the triple labelling 15N tracing model (Müller et al. 2014) in combination with nonrandom 15N distribution (Laughlin and Stevens 2002). Incorporating this would be an improvement of the dual isotope method, as co-denitrification could interfere with the source estimations presented above.

The dual isotope method could be further developed by incorporating better rates of Oex for the pathways starting from NH4 +. Despite potential for improvements, however, this method allows an estimation of the contributions of N, ND, NcD and FD to N2O production and should be applied to a range of soils to further our understanding of these sources of N2O and potential mitigation strategies.

## **7.5 Quantification of Gross N Transformation Rates and Process Specific N2O Pathways via 15N Tracing**

## *7.5.1 Background*

The N cycle is a conceptual model that illustrates where and in which form N is present in the environment and how N is transformed and exchanged between organic, mineral and gaseous N forms. Since the N cycle is a dynamic system not only the sizes of the different N pools, e.g. NH4 +, NO3 <sup>−</sup> or organic N but also the rates between the pools provide an understanding of the dynamic nature of this important elemental cycle in soils and aquatic systems (Ryabenko 2013). The most common and easiest approach to understand the dynamic nature of the N pools is the determination of net process rates, such as net mineralisation rates by calculating the difference in the size of the mineral N pool between two time points. If this rate is positive, we refer to a net mineralisation, if it is negative then we call it net immobilisation. Thus, a net rate always refers to the difference between the production and consumption of the N pools in question. It can easily be shown that different pairs of production and consumption rates will lead to exactly the same net result. Thus, the analyses of net rates do not provide a measure of the individual rates that are contributing to the observed net rate. The individual rates associated with the N pool in question are called gross transformation rates. However, the quantification of these individual rates is not trivial because they cannot be measured directly. The most commonly used method to quantify the gross rates is the isotopic dilution technique (Stark 2000). The principle of this technique relies on the 15N labelling of a certain N pool so that

**Fig. 7.9** Illustration of the dilution technique by considering pool size and 15N abundance at two time points according to Kirkham and Bartholomew (1954)

the gross rate entering this pool can be quantified by taking into account the change in pool size and 15N enrichment of at least two times after label addition (Fig. 7.9).

The example in Fig. 7.9 shows that the pool size is decreasing which means that a net immobilisation of N occurred. However, the decline in 15N abundance of the pool N during the same period also shows that N at natural abundance or low 15N abundance must have entered the pool N. Thus, via visual inspection of the data we can say that N must have entered but also left the pool and that the rate leaving the pool must have been faster than the rate entering the pool. To quantify the individual rates requires a numerical analysis via a suitable N cycle model. Based on a simple twopool N model, Kirkham and Bartholomew (1954) were the first to derive analytical equations that allowed the calculations of the two rates between two-time points, i.e. the gross mineralisation and immobilisation rates. The underlying assumptions are (i) 15N is homogeneously labelled and no preferential usage of either 15N or 14N occurs in the soil, (ii) immobilised N will not re-mineralise and (iii) N transformation rates follow zero-order kinetics (constant rates). The conceptual model of the Kirkham and Bartholomew approach and the equations derived for their model are illustrated in Fig. 7.10.

Since Kirkham and Bartholomew's pioneering work in the 1950s, analyses techniques have been developed which are based on more realistic conceptual N models. These include the division of the mineral N pool into NH4 <sup>+</sup> and NO3 <sup>−</sup> pools with separate immobilisation rates, the consideration of more than one organic N pool and additional N loss rates such as ammonia volatilisation and denitrification (Myrold and Tiedje 1986). The dilution technique works well in simple systems where the inflow into a pool occurs via a single gross N transformation rate. However, in reality, often more than one pathway contributes to the buildup of a pool size. This can be illustrated by the NO3 <sup>−</sup> pool in soil. Production of NO3 <sup>−</sup> can occur via oxidation

**Fig. 7.10** The conceptual model, the differential equations of the various pools and the closed-form analytical solutions for the individual gross rates (m and i) according to Kirkham and Bartholomew (1954). Note, Norg (assumed to contain only 14N) depicts the organic N pool which mineralises into mineral N (M) which consist of H (15N) and N (14N), M <sup>=</sup> <sup>H</sup> <sup>+</sup>N. The subscript 0 refers to the concentrations of the pools at time zero

of NH4 <sup>+</sup> to NO3 <sup>−</sup> (usually termed autotrophic nitrification), and via oxidation of organic N to NO3 <sup>−</sup> (usually termed heterotrophic nitrification) (Fig. 7.11).

Following the principles of the dilution technique, the total gross rate of NO3 − production can be quantified by labelling the NO3 <sup>−</sup> pool and following the concentrations and 15N dilution of this pool over time. This total NO3 <sup>−</sup> production rate includes both, autotrophic (oxidation of NH4 +) and heterotrophic nitrification (oxidation of Norg). To separate the two processes, in addition to the 15NO3 <sup>−</sup> label also the NH4 + pool should be labelled in a separate 15N labelling treatment. To keep the conditions in the two 15N labels the same, it is important to also apply NH4 <sup>+</sup> in the soil that has received the 15NO3 <sup>−</sup> label while NO3 <sup>−</sup> should be applied in the 15NH4 + treatment. Now, the 15N enrichment in the NO3 <sup>−</sup> when only NH4 <sup>+</sup> has been labelled

**Fig. 7.11** Conceptual model for nitrification

will provide a measure of autotrophic nitrification while heterotrophic nitrification can be calculated by difference: *Nh* = *N*tot–*Na* where *N*tot, *Na* and *Nh* refer to total, autotrophic and heterotrophic nitrification, respectively. In practice, to quantify *N*tot the dilution of the 15N labelled NO3 <sup>−</sup> pool (Fig. 7.10) can be used while *Nh* and *Na* can only be estimated via a simulation model that takes into account both nitrification rates (Barraclough and Puri 1995). A parameter optimisation technique can also be used to estimate *Na* or *Nh* (Myrold and Tiedje 1986). Thus, in modern 15N tracing applications dilution-enrichment principles will be taken into account which we refer to as tracing.

In models with several simultaneous N transformations, it is impossible to derive analytical solutions. Therefore, the development of 15N tracing models which use numerical solutions has become the state-of-the-art approach (Mary et al. 1998). These models rely on a set of differential equations for example of a simple system that describes the N cycle. Transformation rates between the various pools can be constant (zero-order kinetics) or are dependent on the pool size where the rate is originating from (first-order kinetics) or follow enzyme kinetics (i.e. Michaelis–Menten kinetics). While zero and first-order kinetics are described by one parameter, rates calculated via Michaelis–Menten kinetics are dependent on two parameters, i.e. the maximum velocity of the reaction rate and the half-saturation constant (Müller 2000). The determination of the parameters in such equation systems rely on parameter optimisation tools. A whole range of parameter optimisation tools are available and different algorithms have been used in 15N tracing models (Mary et al. 1998; Myrold and Tiedje 1986). More recently, parameter optimisation tools based on Bayesian probability have become more common because they allow the simultaneous optimisation of a large number of parameters (for more details see Müller et al. 2007). It should be noted that the sole purpose of tracing models is to quantify gross transformation rates and are therefore data analysis tools and should not be confused with simulation models.

In the following sections, current 15N tracing techniques are illustrated. This includes the description of experimental requirements to obtain suitable data sets and the subsequent model analysis. A number of 15N tracing models have been developed (e.g. FLUAZ, Mary et al. 1998), and here, the data analysis will be illustrated by the *Ntrace* model. This model is based on the tool presented by Müller et al. (2007) and has since been developed further to analyse data sets from a range of differently complex setups, including dynamics of nitrite, gaseous N emissions, soil–plant interactions, biochar, etc. An advantage of *Ntrace* is its flexibility to adapt to various conditions and models because it is programmed in MatLab with code that can easily be changed and amended.

## *7.5.2 Stable Isotope Tracing Technique*

A stable isotope tracing study consists of two parts, an experimental study where one or more pools are isotopically labelled and a data analysis tool (e.g. *Ntrace*) to quantify individual gross transformation rates. The technique can be regarded as a calculation procedure to quantify gross rates which cannot be quantified via any other means. Thus, both the tracing experiment and the numerical tool are building an analysis unit and it is important that the experimental approach is taking into account the requirements of the numerical analysis and vice versa. What is also important is that the quality of the final results critically depends on the data quality and therefore on the careful execution of the experimental part of the tracing study. For instance, data with high uncertainties may also result in gross N rates that are associated with large errors.

## *7.5.3 Setup of Tracing Experiments*

To be able to analyse experimental data with the *Ntrace* model, the experiment needs to be set up in a certain way. Based on the research questions both field and laboratory experiments can be carried out. The research question usually requires the setup of several treatments (e.g. effect of various soil amendments). To quantify the individual gross N transformation rates in each treatment usually a set of at least two 15N labels should be employed per treatment (i.e. 15N-labelled NH4 <sup>+</sup> and 15N-labelled NO3 −, typically applied as NH4NO3 to ensure the application of equal quantities of each N species). However, often multiple labels are used (e.g. very common is a triple labelling approach with NH4NO3 where either NH4 +, NO3 <sup>−</sup> or both moieties are 15N labelled).

Ideally the 15N label should be applied without enhancing the concentration because this will also have an impact on the N transformations. Thus, in ecosystems that are not used to receive large N concentrations often a high 15N enrichment (e.g. 99 atom% 15N) is applied at a very low application rate. However, in agricultural soils which receive N in the form of fertiliser, this is less of a problem. The advantage of applying a reasonable, but not unrealistically high, N concentration is that it can more homogenously be applied to the soil. In most cases, a 15N enrichment of a few percent (e.g. 10 atom% 15N) is sufficient to determine gross rates. However, in situations where, for instance, the nitrite or gaseous N species such as N2O are analysed, the labelled N pool (e.g. NO3 <sup>−</sup>) should ideally be enriched by approximately 50 atom% 15N which allows most precise analysis based on the expected 29/28 iron current (Stevens et al. 1993). The 15N solutions are made up according to standard calculations which are, for instance, summarised by Cabrera and Kissel (1989). To homogeneously label the soil a variety of application techniques are described in the literature ranging from application via side port needles in different depth, multiple needle applicators and automated techniques (Buchen et al. 2016) (Table 7.4). In field tracing studies often an application via a watering can is preferred, simply, because under field conditions when large plots of several m2 have to be treated, it is critical that the solutions are applied within a short time window to ensure the same starting conditions (Plate 7.1). This is particularly important if dynamically changing N species such as N2O should be compared among treatments (Moser et al. 2018).

**Plate 7.1** Application of 15N solution (15NH4NO3, NH15 <sup>4</sup> NO3) on plots in the field with a small watering device (**a**), and soil incubation in a climate chamber in suitable incubation jars (**b**)

However, the application rate should be slow enough to avoid by-pass in cracks and fissures down the soil profile because this would cause uneven distribution of the 15N.

The time of labelling should be carefully noted because the difference between 15N application and soil analysis during the experiment provides the time after N supply which is required for the model analysis. If both, soil extractions and gaseous measurements are planned then in the field an area for the gas sampling and an adjacent soil sampling area should be setup (in Plate 7.1, gas measurement are in the forefront, and the area for soil sampling is at the top). In laboratory incubations usually one set of jars is reserved for gaseous measurements (which will be extracted at the end of the experiment) while for each analysis day, separate sets of jars are prepared for destructive sampling. The question arises for how long we need to carry out a typical incubation study. Since the application of N may cause a stimulation of microbial activity resulting in faster gross N rates shortly after N application, the duration of a typical tracing study should be continued until after this initial stimulation has subsided. A typical duration of such a study is approximately 14 days. To characterise the non-linear dynamics of the gross N rates over time it is necessary to determine the N pool sizes and their 15N enrichment at least 5 times throughout that period. Gas analysis should be carried out more frequently but at least at the times when soils are extracted.

Soil incubations have typically been carried out under controlled conditions at a pre-defined moisture content (set to a certain water filled pore space (WFPS) or water holding capacity (WHC)) and temperatures in a climate chamber (Plate 7.1).

#### **Case study**

To investigate the effect of a nitrification inhibitor in two soils on gross N transformations the following setup is realistic (using a triple 15N labelling approach, numbers in brackets refer to the number of entities):

Soils (2) × Inhibitors (2) × 15N labels (3) × Replicates (3) × Time of soil extraction (5) = 180 jars.

Thus, a total of 180 jars (i.e. 36 jars per extraction day) need to be prepared. The label needs to be applied with minimal disturbance while providing an equal distribution in the soil. This can be done using a long needle with side ports. If 150 g of dry soil equivalent should be used per jar, then approximately 14 kg of soil is required from each soil.

The extraction times should be timed in such a way that the first extraction happens as soon as possible after 15N labelling (typically after 2 h), then on day 1, 3, 7 and 15. Note, soils can react quite differently, therefore, the times and duration of the experiment should be adjusted accordingly.

## *7.5.4 Analyses of Experimental Data*

#### **7.5.4.1 Soil Extraction**

If nitrite concentrations should be investigated it is recommendable to carry out the blending procedure of Stevens and Laughlin (1995). They discovered that nitrite is chemically reduced to N2 in the KCl extract at pH below 5.5. They recommended a soil extraction at pH 7 and fast soil extraction. The blending procedure of Stevens and Laughlin (1995) is typically carried out at a soil: solution ratio of 1:1 in a blender for 90 s (Plate 7.2).

Immediately after the blending, the soil suspension needs to be centrifuged at 2000 × g for 5 min, and the supernatant filtered sequentially through a GF/D and GF/F (Plate 7.2).

The extracts have to be analysed for NO3 <sup>−</sup> and NH4 <sup>+</sup> and possibly for NO2. Based on the concentration a certain μmol of N of each N species will then be converted to 15N-N2O or via a diffusion approach.

## **7.5.4.2 Chemical Conversion of Mineral N to 15N-N2O**

A precise method to determine the 15N content of ammonium, NO3 <sup>−</sup> and nitrite is via a method that converts the N species to nitrous oxide (N2O). The reduction of NO3 <sup>−</sup> to N2O is described by Stevens and Laughlin (1994). For this, sulphamic acid (2.5 ml of 0.2 M solution) is added to 50 ml soil extract and shaken by hand for 5 s to ensure conversion of NO2 <sup>−</sup> to N2. Then 5 ml of 1 M sodium acetate- 1 M acetic

**Plate 7.2** Extraction procedure for quick soil extraction (**a**) and glass filter unit for glass fibre filter papers (**b**)

acid buffer has to be added to increase the pH to 4.7. Then a CD-Cu reductor has to be placed in the bottle (Plate 7.3).

**Plate 7.3** Soil extracts are transferred to medical flasks for conversion of NO2 <sup>−</sup> and NO3 <sup>−</sup> to N2O; gas samples are taken through the septa with syringes and then are transferred to pre-evacuated exetainers (from left to right)

The flask, capped, has to be laid flat in an orbital incubator at 20°C and shaken at 120 rpm with an orbit diameter of 50 mm for 2 h. A gas sample of the headspace is analysed with an IRMS for the 15N content of the N2O. The 15N content of the NO3 <sup>−</sup> is considered to be the same as that of the N2O.

The production of N2O from NH4 <sup>+</sup> is described in Laughlin et al. (1997). Firstly, the ammonium must be diffused into (NH4)2SO4. For this, 50 ml of the soil extract has to be pipetted in the diffusion unit (Plate 7.4). Above this liquid, a small flask containing 3 ml of H2SO4 has to be placed. Before the diffusion jar is closed, 0.2 g of heavy MgO must be added. The MgO has to be brought into suspension by gentle swirling for 30 s. After this, the diffusion jar has to be left for 4 days. After diffusion of the NH3, the (NH4)2SO4-H2SO4 has to be poured into a 12 ml glass exetainer and evaporated to dryness in a 150 °C oven, before cooling it in a desiccator and sealing it with a septum and cap. Then the vial has to be evacuated and filled with He to atmospheric pressure. One ml of NaOBr, with the molarity of NaOH adjusted to 10 M has to be injected through the septum. The vial has to be tilted and the solution gently swirled to ensure that the NaOBr reacts with as much of the (NH4)2SO4 as possible. The concentration and 15N content of the N2O in each vial has to be determined by an IRMS system.

**Plate 7.4** Glass equipment used for the conversion of NH4 <sup>+</sup> to NH3 which is trapped in the acid contained in the small hanging flask

For the *Ntrace*Nitrite model, data on NO2 <sup>−</sup> concentration and 15N content are also necessary. The NO2 <sup>−</sup> concentrations can be determined by a manual photometer method.

The 15N content of the NO2 <sup>−</sup> extracts can also be determined by a method based on conversion to N2O as described by Stevens and Laughlin (1994). For this 50 ml of the soil extract has to be pipetted into a bottle. One ml of 1 M HCl and 0.5 ml of 0.04 M NH2OH has to be added to the bottle. The bottle should then be capped and laid flat in an orbital incubator and shaken at 120 rpm with an orbit of 50 mm for 16 h. A 12 ml sample of the headspace has to be transferred to an evacuated septum-capped glass vial, and the 15N content of the N2O in each vial can be determined by IRMS. The atom% excess in 15N in NO2 <sup>−</sup> is calculated as two times the 15N atom% excess in N2O minus the 15N atom% excess in NH2OH.

The specific steps of the conversion method to N2O are summarised below.

1. NH4 +-N: NH4 +-N is first oxidised to NO2 <sup>−</sup>-N by BrO<sup>−</sup> in a vacuum with N2O being the by-product (Eq. 7.22). The production of N2O can be catalysed by Cu<sup>+</sup> at the appropriate pH (Laughlin et al. 1997).

$$\rm NH\_4^+ + \rm NaBrO \rightarrow \rm NaBr + H\_2O + N\_2 \uparrow + N\_2O \tag{7.22}$$

2. NO3 <sup>−</sup>–N: NO2 <sup>−</sup>–N must be removed by NH2SO3H before NO3 <sup>−</sup>-N is reduced. NO3 <sup>−</sup>–N is reduced to NO2 <sup>−</sup>–N and NH2OH by copper-plating cadmium grains at a pH of 4.7. Then NO2 <sup>−</sup>–N reacts with NH2OH to produce N2O, and the production of N2O is positively correlated to the production of NO3 <sup>−</sup>-N (Eq. 7.23). The ammonium and N from other sources has no effect on the determination of NO3 <sup>−</sup>-N; (Stevens and Laughlin 1994).

$$\rm NO\_3^- \rightarrow \rm NO\_2^- + \rm NH\_2OH \rightarrow \rm N\_2O + 2H\_2O \tag{7.23}$$

3. NO2 <sup>−</sup>-N: NO2 <sup>−</sup>-N reacts with NH2OH to produce N2O (Eq. 7.24) and the reaction is pH-dependent. When pH < 4, the reaction rate increases rapidly, and the reaction time should be at least 16 h. Because N2O is formatted via an asymmetric intermediate (N-nitroso-hydroxyl-amine) under acidic condition, the reaction requires at least 10 μmol of NH2OH (Stevens and Laughlin 1994).

$$\text{HNO}\_2 + \text{NH}\_2\text{OH} \rightarrow \text{N}\_2\text{O} + 2\text{H}\_2\text{O} \tag{7.24}$$

The amount of N2O produced is about half of the theoretical yield. According to the isotopic distribution, the two N atoms in N2O are formed from NO2 <sup>−</sup>-N and NH2OH, respectively. Hence, the atom% in NO2 −- 15N needs to be calculated with Eq. 7.25 (assume the atom% 15N in NH2OH is 0.365 atom%) (Laughlin et al. 1997):

$$^{15}\text{Natom}\% \text{(NO}\_2^- - \text{N}) = 2 \times ^{15}\text{Natom}\% \text{(N}\_2\text{O)} - 0.365\text{ atom\%} \tag{7.25}$$

#### **Apparatus**

PT-IRMS (purge and trap system coupled to isotope ratio mass spectrometry)

Vacuum pump

50 ml reaction vials

Glass vials with Al caps and septums

#### **Reagents**

*NH*<sup>+</sup> *<sup>4</sup>* -*N to N2O method*:

MgO: combusted at 450 °C for 4 h

0.01 M H2SO4 with 0.5 mM CuSO45H2O

Basic NaBrO (10 M NaOH)

NO3 <sup>−</sup>-N to N2O:

0.2 M NH2SO3H

1 M CH3COOH- CH3COONa (pH = 4.7)

Copper-plated cadmium granules

NO2 <sup>−</sup>-N to N2O:

1 M HCl

0.04 M NH2OH-HCl

#### **Procedures**

1. NH4 +-N:


```
2. NO3
       −-N:
```
(a) Pipet 20–25 ml (about 20 μg N) of the soil extract into a flask. Add 2.5 ml 0.2 M NH2SO3H, and shake the flask for 5 min to ensure conversion of NO2 −-N to N2;

	- (a) Pipet 10–15 ml (about 0.5–1.0 μg NO2 <sup>−</sup>-N) of the soil extract into a 50 ml reaction bottle. Keep the bottle capped tightly, evacuate and then fill with pure He;
	- (b) Inject 1 ml 1 M HCl and 0.5 ml 0.04 M NH2OH-HCl into the bottle;
	- (c) Place the bottle on a shaker running at 120 rpm for 16 h;
	- (d) Transfer a known amount of sample to an evacuated septum-capped glass vial. The 15N content of the N2O is then determined with mass spectrometry. Finally, calculate the NO2 −- 15N using Eq. 7.25.

## **7.5.4.3 Inorganic Nitrogen Isotopic Analysis in Soil Extracts via the Diffusion Method**

An alternative method for the determination of 15N in NO3 <sup>−</sup> and NH4 <sup>+</sup> is the diffusion method. The diffusion method is easier to apply and has the advantage that only a solid analysis on an IRMS is required rather than a gas measurement. These Mass spectrometers are more readily available. However, it should also be pointed out that chemical conversion method described above is quicker and is free from contamination by atmospheric N. It has very low detection limits, which are 20 μg N for NH4 +-N, 5 μg N for NO3 <sup>−</sup>-N and 0.5 μg N for NO2 <sup>−</sup>-N.

## **Principle**

During diffusion, ammonium in the soil samples is converted to ammonia by the use of MgO (Eq. 7.26). Then the ammonia is absorbed by using a filter paper containing a weakly acidic absorbent liquid during the volatilisation process. For determination of NO3 <sup>−</sup>-N, titrate some alkaline reagent to remove NH4 +-N in the sample then add some Devarda's alloy to reduce the NO3 <sup>−</sup>-N into NH4 +-N.

$$2\text{NH}\_4^+ + \text{MgO} \rightarrow \text{Mg}^+ + 2\text{NH}\_3\uparrow + \text{H}\_2\text{O} \tag{7.26}$$

*Apparatus need are*: EA-IRMS Shaker 250 ml airtight containers Perforated silicon films

Perforated filter papers

paper clips

Glass beads

*Reagents*:

MgO: Combusted at 450 °C for 4 h

Devarda's alloy: crushed to allow passage through a 300-mesh sieve

1MH2C2O4

#### **Procedures**


$$E\_s = \frac{E\_m M\_{s+b} - E\_b M\_b}{M\_{s+b} - M\_b} \tag{7.27}$$

where Es is the corrected abundance 15N enrichment of the sample, Em is the enrichment of the sample + blank measured by mass spectrometry, Ms+b is the mass of N (sample + blank) recovered in the acid trap, Mb is the mass of N recovered in the acid trap from the diffusion blank, and Eb is the enrichment in the blank (assumed to be 0.3663 atom%).

Note, for soil extracts < 2 mg l−<sup>1</sup> in inorganic N concentration, 50 ml of extract is needed to ensure accurate determination. When the abundances of NH4 +-N and NO3 <sup>−</sup>-N are very different, it is better to diffuse NH4 +-N and NO3 <sup>−</sup>-N separately (do not use the same extract).

#### **7.5.4.4 Inorganic Nitrogen Isotopic Analysis in Soil Extracts at Natural Abundance**

The diffusion method and chemical conversion method described above are both suitable for N at high abundance, but not for N at natural abundance. They both have a high demand for N and high levels of background N can interfere with the reaction. There are two modified chemical methods for N isotopic analysis at natural abundance. These simplify the preparation procedures, reduce the preparation time and do not require large amounts of N. The chemical method for ammonium requires 2.5μg N in a 4 ml sample volume for analysis, and its accuracy of δ15N measurements is less than 0.3‰. The method for NO3 <sup>−</sup> needs only 4.5 μg N, and its accuracy of δ15N and δ18O reaches 0.31‰ and 0.55‰, respectively.

#### **Conversion of ammonium at natural abundance**

#### **Principle**

The method is to oxidise NH4 +-N to NO2 <sup>−</sup>-N by BrO<sup>−</sup> instead of extraction of NH4 +- N in solution. Subsequently, the NO2 <sup>−</sup>-N is reduced to N2O by NH2OH-HCl, thus replacing HN3 (Liu et al. 2014; Stedman 1959) (Eq. 7.28).

$$\text{NH}\_4^+ + \text{NaBrO} \rightarrow \text{NO}\_2^- + \text{NH}\_2\text{OH} \rightarrow \text{N}\_2\text{O} + \text{H}\_2\text{O} \tag{7.28}$$

#### **Apparatus**

#### PT-IRMS

Shaker

20 ml headspace glass vials: Acid rinsed and combusted at 450 °C for 4 h

#### **Reagents**

10 M NaOH: Evaporate 100 ml of 5 M NaOH to 50 ml of 10 M NaOH

NaBrO:

(a) Bromate–bromide stock solution: Mix 0.6 g NaBrO3 and BrNa into 250 ml DIW (deionised water) (can be stored ≥ 6 months);


NaAsO2: Mix5.1 g NaAsO2 and 100 ml DIW

NH2OH, HCl:


```
6 M HCl
```
5 M NaOH

#### **Procedures**


δ15NNH<sup>+</sup> <sup>4</sup> sample <sup>=</sup> (δ15NN2Osample <sup>−</sup> intercept)/slope (7.29)

where the intercept and slope are obtained from the linear regression of the δ15N measured and the δ15N assigned from N2O produced by the standards.

#### **Conversion of NO<sup>−</sup> <sup>3</sup> at natural abundance**

#### **Principle**

The NO3 <sup>−</sup>-N is reduced to NO2 <sup>−</sup>-N by copper-plated cadmium granules in a weakly alkaline environment (Eq. 7.30):

$$\rm{NO\_3^- + Cd + H\_2O \leftrightarrow NO\_2^- + Cd^{2+} + 2OH^-} \tag{7.30}$$

NO2 <sup>−</sup>-N is converted into N2O by N3 <sup>−</sup> in a weakly acid buffer. When pH > 2, the reaction will be (Eqs. 7.31−7.33):

$$\mathrm{NO}\_{2}^{-} + \mathrm{H}^{+} \leftrightarrow \mathrm{HNO}\_{2} \tag{7.31}$$

$$\text{HNO}\_2 + \text{H}\_2\text{O} \leftrightarrow \text{H}\_2\text{NO}\_2^+ + \text{OH}^-\tag{7.32}$$

$$\rm H\_2NO\_2^+ + N\_3^- \rightarrow N\_3NO + H\_2O \rightarrow N\_2O + N\_2 \tag{7.33}$$

When there is a large number of halogen ions present the reaction will be accelerated (Stedman 1959) (Eqs. 7.34, 7.35):

$$\text{H}\_2\text{NO}\_2^+ + \text{Cl}^- \rightarrow \text{NOCl} + \text{H}\_2\text{O} \tag{7.34}$$

$$\rm{NOCl} + \rm{N}\_{3}^{-} \rightarrow \rm{N}\_{3}\rm{NO} + \rm{Cl}^{-} \rightarrow \rm{N}\_{2}\rm{O} + \rm{N}\_{2} \tag{7.35}$$

N2O is an asymmetric molecule with a molecular structure of N-N-O. The δ15NAir of N2O is the mean value of δ15NAir in two N atoms (Eq. 7.36):

$$\delta^{15} \mathbf{N}\_{\text{Air}}{}^{0}{\!\!/ \_{0}} \text{(N}\_{2}\text{O)} = \frac{\delta^{15} \mathbf{N}\_{\text{Air}}{}^{0}{\!\!/ \_{0}} (^{15}\mathbf{N} - \mathbf{N} - \mathbf{O}) + \delta^{15} \mathbf{N}\_{\text{Air}}{}^{0}{\!\!/ \_{0}} \text{(N} - ^{15}\mathbf{N} - \mathbf{O})}{2} \tag{7.36}$$

and the N2O produced is composed of a N atom and an oxygen atom provided by the NO2 <sup>−</sup>-N and a N atom provided by N3 <sup>−</sup>, a N source. The isotope ratio of N and oxygen of the NO2 <sup>−</sup> is identical with that of NO3 <sup>−</sup> in the original solution. So, the following relationships hold (Eqs. 7.37, 7.38):

$$
\delta^{15} \text{N}\_{\text{Air}}\%\_{\text{ol}}(\text{N}\_{2}\text{O}) = \frac{\delta^{15} \text{N}\_{\text{Air}}\%\_{\text{ol}}(\text{N}\_{3}^{-}) + \delta^{15} \text{N}\_{\text{Air}}\%\_{\text{ol}}(\text{NO}\_{2}^{-})}{2} = \frac{\delta^{15} \text{N}\_{\text{Air}}\%\_{\text{ol}}(\text{N}\_{3}^{-}) + \delta^{15} \text{N}\_{\text{Air}}\%\_{\text{ol}}(\text{NO}\_{3}^{-})}{2} \tag{7.37}
$$

$$
\delta^{18} \text{O}\_{\text{SMOW}}\%\_{\text{ol}}(\text{N}\_{2}\text{O}) = \delta^{18} \text{O}\_{\text{SMOW}}\%\_{\text{ol}}(\text{NO}\_{2}^{-}) + \delta^{18} \text{O}\_{\text{SMOW}}\%\_{\text{ol}}(\text{NO}\_{3}^{-}) \tag{7.38}
$$

Therefore, when the isotope ratio of N3 <sup>−</sup> is constant, the N and oxygen isotope ratios of N2O produced is linear with the N and oxygen isotope ratios of NO3 <sup>−</sup>, and the theoretical slopes of their correlation curves are 0.5 (N) and 1.0 (O), respectively.

#### **Apparatus**

PT-IRMS

Shaker

pH metre

Peristaltic pump: flow rate ≥ 5 ml min−<sup>1</sup>

Water-thermostat

Fume hood

Filter papers

Headspace glass vials

## **Reagents**

0.5 M NaCl

0.5 M HCl

1MC3H4N2

NaN3- CH3COOH:


## 6 M NaOH

Copper-plated cadmium granules

Cadmium reduction column

## **Procedures**


(b) Copper-plated cadmium granules can be added directly to the samples. Place the samples in a shaker running at 200 rpm at 30 °C for 3 h to reduce NO3 <sup>−</sup>-N. Filter the sample into another vial.

5. Take a 16 ml sample and place it into a 50 ml headspace glass vial with the cap sealed. Evacuate the vial and fill it with He gas.


$$\delta^{15} \mathbf{N}\_{AirNO\_3^-} = (\delta^{15} \mathbf{N}\_{AirN\_2O} - \text{intercept})/\text{slope} \tag{7.39}$$

$$(\delta^{15}\mathbf{N}\_{(180)WNO\_3^-} = (\delta^{15}\mathbf{N}\_{\text{SMOWN}\_2O} - \text{intercept})/\text{slope} \tag{7.40}$$

where the intercept and slope are obtained from the linear regression of the δ15N and δ18O measured from N2O produced by the standards and the δ15N and δ18O are assigned from the standards.

## *7.5.5 15N Tracing Model Analyses via* **Ntrace**

#### **7.5.5.1 Data Requirements for the** *Ntrace* **Model**

Data obtained through the 15N tracing experiment will be further analysed by the *Ntrace* model to quantify gross N transformation rates and pathway-specific N2O emissions. The various *Ntrace* model versions differ in their data requirements. The *Ntrace*Basic model has the fewest data requirements. The other models require more data on top of the data required for the *Ntrace*Basic model. The *Ntrace*Basic requires the fertiliser application rate (in μmol N g−1) and its 15N excess (in atom%). It also requires the average NO3 <sup>−</sup> and NH4 <sup>+</sup> concentration and 15N excess (in atom% excess) including their standard deviations for each data point in time. The NO3 − and NH4 <sup>+</sup> concentrations should be given in the same unit as the fertiliser application rate. Next to this, a one-time measurement of total organic N (in %) is required. This measurement can be taken from basic soil characteristics. The *Ntrace*Plant model also requires plant N and plant 15N data, and total plant biomass data, at each time when destructive sampling was carried out, i.e. preferably the same time points when NO3 <sup>−</sup> and NH4 <sup>+</sup> were determined. Ideally, above- and belowground biomass (roots) should be determined. The *Ntrace*Urea model requires the urea application rate and its excess, and if plants are included, it also has the additional requirements of the *Ntrace*Plant model. The *Ntrace*Nitrite model requires measurements at multiple time points (preferably the same as for NO3 <sup>−</sup> and NH4 +) of the average NO2 − concentration and its 15N excess (in atom%) including standard deviations.

Transformation can follow zero-order, first-order, or Michaelis–Menten kinetics. The type of kinetics used needs to be specified for each transformation separately. If the transformation uses N from a large pool, it is generally appropriate to use zeroorder kinetics. For N transformations coming from pools that change rapidly it is generally more realistic to use first-order or Michaelis–Menten kinetics. Especially the transformations associated with the NH4 <sup>+</sup> consumption (e.g. nitrification) may be most realistically represented by Michaelis–Menten kinetics. However, under conditions when microbial activities may be affected by conditions other than substrate, the N transformation rate may also follow first-order kinetics. For example, when the rate is governed by temperature or soil moisture.

#### **7.5.5.2 The** *Ntrace* **Model System**

The 15N tracing model *Ntrace* described by Müller et al. (2007) is a tool to quantify gross soil N transformations; this model considers five N pools and twelve simultaneously occurring N transformations (Fig. 7.12). The five soil N pools are ammonium (NH4 +) adsorbed NH4 <sup>+</sup> (NH4 + ads); labile soil organic N (Nlab); NO3 <sup>−</sup>, nitrate stored ( NO3 <sup>−</sup>sto) and recalcitrant soil organic N (Nrec). The model considered twelve gross N transformations from these five N pools: mineralisation of recalcitrant organic-N to NH4 <sup>+</sup> (*MNrec*); mineralisation of labile N to NH4 <sup>+</sup> (*MNlab*); immobilisation of NH4 + into recalcitrant organic-N (*INH4\_Nrec*); immobilisation of NH4 <sup>+</sup> into labile organic-N (*INH4\_Nlab*); oxidation of NH4 <sup>+</sup> to NO3 <sup>−</sup> (*ONH4*); oxidation of recalcitrant organic-N to NO3 <sup>−</sup> (*ONrec*); immobilisation of NO3 <sup>−</sup> to recalcitrant organic-N (*INO3*); dissimilatory NO3 <sup>−</sup> reduction to NH4 <sup>+</sup> (*DNO3*); adsorption of NH4 <sup>+</sup> on cation exchange sites (*ANH4*); and release of adsorbed ammonia (*RNH4a*), adsorption and release of NO3 − on/from stored NO3 <sup>−</sup>, i.e. *ANO3s* and *RNO3s*, respectively. *Ntrace* is a family of 15N tracing models to quantify gross N transformations in soils and sediments. The model consists of a N transformation model that is programmed in Simulink, a graphical programming language associated to Matlab, and a parameter optimisation routine based on a Markov Chain Monte Carlo routine in combination with the Metropolis algorithm (Müller et al. 2007).

Several extensions exist of the *Ntrace*Basic model, namely *Ntrace*Plant (Fig. 7.13a), *Ntrace*Urea (Fig. 7.13b), *Ntrace*Nitrite (Fig. 7.14) and *Ntrace*Gas (Fig. 7.15). The boxes represent the different N pools, and the transformations are represented by the arrows between the boxes. For each model all transformations are quantified simultaneously (Fig. 7.12).

The Matlab-Simulink files (m-files and mdl-files) alongside their description that are part of the *Ntrace* model are presented in Table 7.7.

Currently, a new optimisation routine for the *Ntrace* model is being implemented. This will further improve optimisation speed, and more importantly be quicker to find a global minimum as opposed to a local one. The method used for determining optimal parameters will be Matlab's GlobalSearch algorithm (Ugray et al., 2007).

**Fig. 7.13 a** *Ntrace* Plant model, based on Inselsbacher et al. (2013) **b** *Ntrace*Urea model

Standard deviations of the optimised parameters will be determined as described by Gavin (2019).

## *7.5.6 Parameter Optimisation with* **Ntrace**

#### **7.5.6.1 Setup**

After filling out an input file for the model (DataNtrace.xlsx) that contains all the required data specified in data requirements including kinetics for each transformation, initial parameters and minimum and maximum values for the parameters, the model can be run (Fig. 7.16).

#### **7.5.6.2 Procedure**

The first step of the optimisation is generally done by hand. The model is run with the initial parameter set, and graphs of modelled versus measured data are inspected. From this, parameters are adjusted until a visually reasonable fit is obtained. Thereafter, all parameters are optimised simultaneously using a Markov chain Monte Carlo (MCMC) method, and this is explained in more detail by Müller et al. (2007). For this, the parameters are slightly adjusted, and the model is run. At the end of the run, the misfit is calculated based on the difference between the modelled and measured values. If the misfit is lower compared to the previous run, i.e. a better fit between modelled and measured data, the new parameter set is accepted, and it will start again by adjusting the newly accepted parameter set and execute the next iteration. If the




**Fig. 7.16** Data *Ntrace* Excel file that is used to setup *Ntrace* model runs. The sheets show: aboveleft: input sheet of experimental data, below-left: setup of model run with input of basic parameters, e.g. Norg content, N application and the 15N enrichment, right: parameter sheet. Note: yellow areas are input areas

misfit is higher, so a worse fit, there is also a chance the parameter set is accepted via a likelihood function. By this, it is possible that the algorithm moves out of a local minimum and enters the global minimum. If a new parameter set is not accepted, the last accepted parameter set will be used for the parameter adjustment. This iterative procedure of parameter adjustment and running the model should go on till the probability density functions (PDFs) are well characterised for all parameters. If the initial parameter set is fairly close to the optimal set, PDFs are generally well characterised after 50,000 to 100,000 iterations. So, for the final run, the maximum number of iterations is generally set between 50,000 and 100,000. During the optimisation, generally three parallel sequences, each with different starting parameters, are calculated. The number of parallel sequences should be defined before running the model, but three is generally appropriate. From these parallel sequences, a reduction factor is determined, which determines the accuracy of the sampling (Gelman et al. 2003). If the reduction factor is below a pre-specified number (default 1.1) for all parameters, the optimisation will also be stopped, regardless of reaching the maximum number of iterations. The reduction factor near 1 indicates that all parallel sequences resulted in statistically the same parameter set.

Inspection of the PDFs can show that for certain parameters the peak is close to zero. This would indicate that this particular parameter can be neglected. The parameter can then be set to zero, and excluded from optimisation when the model is re-run.

#### **7.5.6.3** *Ntrace* **Model Output**

At the end of the optimisation, the model output will be exported to an Excel file. This output contains the initial parameter value, the optimised parameter value, the standard deviation of the optimised parameter, the average N flow for each transformation, the overall R2 and the AIC. The standard deviation (SD) of the transformation rate is based on the SD and average (AVG) parameter values as shown in Eqs. 7.41 to 7.43:

$$SD\_{\text{transformation rate}} = \frac{AVG\_{\text{transformation rate}} \cdot SD\_{\text{parameter}}}{AVG\_{\text{parameter}}} \tag{7.41}$$

Response ratios between transformation rates can be calculated by McGeough et al. (2016):

$$\mathbf{R} = \ln \left( \frac{\overline{\mathbf{X}}\_{\text{E}}}{\overline{\mathbf{X}}\_{\text{C}}} \right) \tag{7.42}$$

With the associated standard deviation:

$$sd\_R = \sqrt{\frac{sd\_E^2}{n\_E \overline{X}\_E^2} + \frac{sd\_C^2}{n\_C \overline{X}\_C^2}}\tag{7.43}$$

where *XE* and *XC* are the average N transformations of the elevated and control group, *sdE* and *sdC* the associated standard deviations and *nE* and *nC* the repetitions.

To compare the effect of different treatments on transformation rates, the individual treatments have to be run separately. After this, the rates can be compared using a one-way ANOVA based on the averages and standard deviations. Pairwise comparisons can be calculated with the Holm-Šídák test.

Another way to compare gross N transformation is via the determination of least significant difference (LSD) as described by Müller et al. (2011).

Output of the correlation matrix can be used to find parameters that tend to be strongly constrained together. A correlation value of above 0.8 indicates that it is constrained. There is also another output file that gives the pool sizes and transformation rates for each time step. This output can be used to create graphs.

## *7.5.7 Determination of N2O Pathways*

#### **7.5.7.1** *Ntrace* **Approach to Quantify N2O Pathways**

Nitrous oxide (N2O) can be emitted via a number of pathways including inorganic and also organic pathways, involving a range of microbes (e.g. bacteria, fungi) (Butterbach-Bahl et al. 2013). The *Ntrace*Gas model can be applied to quantify N2O pathways based on the underlying N transformations and especially based on the nitrite dynamics (*Ntrace*Nitrite). To accurately estimate N2O pathways via *Ntrace*Gas also the N2 production is calculated. The two predominant biological processes for N2O production in soil are traditionally considered to be autotrophic nitrification and heterotrophic denitrification (Ambus 1998; Wrage-Mönnig et al. 2018; Wrage et al. 2001). In both processes NO2 <sup>−</sup> is the key precursor to N2O production. In nitrifier nitrification, it is rather NH2OH or at least something before nitrite. In nitrifier denitrification, it is nitrite.

Assuming that the N2O production is derived from a single NO2 <sup>−</sup> pool, the 15N enrichments of the N2O and the NO2 <sup>−</sup> should be similar. However, experimental data show that the enrichment of the N2O is deviating from the theoretical 1:1 line (Fig. 7.17) leading to the conclusion that the N2O originated from various NO2 − sub-pools and also from sources which were at or close to natural abundance. Based on the experimental setup, the only common unlabeled N pool in all 15N treatments is organic N. Therefore, two possible processes were included in the *Ntrace*Gas model to account for such a dilution effect:

(a) reduction of NO2 <sup>−</sup> originating from oxidation of organic N derived (NO2 <sup>−</sup>org) and

(b) hybrid-reaction for N2O production whereby one atom of the N2O is derived from an enriched NO2 <sup>−</sup> pool and another from organic N at natural abundance.

**Fig. 7.17** Relationships between nitrite and nitrous oxide **a** as well as the 15N enrichment **b** (Müller et al. 2014)

Heterotrophic nitrifiers can also denitrify (Blagodatsky et al. 2006; Papen et al. 1989) and it is, therefore, possible that NO2 <sup>−</sup>org in the *Ntrace*Nitrite model (Rütting and Müller 2008) originating from Norg oxidation could be further reduced to gaseous N products. A hybrid-reaction between NO2 <sup>−</sup> and organic N is also possible which occurs, for instance, in the fungus *Fusarium oxysporum* (Kurakov et al. 2000; Tanimoto et al. 1992) and possibly in other heterotrophic organisms (Kumon et al. 2002).

Based on the above considerations the *Ntrace*Gas model analyses four N2O processes. The entire model includes all the previous 15N tracing models (see *Ntrace* family above). In *Ntrace*Gas NO2 <sup>−</sup> sub-pools are reduced to associated N2O pools which may further be reduced to N2 (Eq. 7.44).

$$\text{2NO}\_2^-\text{}\_x \rightarrow \text{N}\_2\text{O}\_x \rightarrow \text{N}\_2\tag{7.44}$$

(x = nit, den or org)

In addition, a hybrid-reaction between denitrification derived NO2 <sup>−</sup> (NO2 <sup>−</sup>den) and recalcitrant organic N (Nrec) was introduced (Eq. 7.45).

$$\rm NO\_2^- \rm\_{den} + N\_{\rm rec} \to N\_2O\_{\rm cod} \to N\_2 \tag{7.45}$$

Each soil N2O sub-pool can be further reduced to N2 via specific N2O reduction rates or emitted to the atmosphere, which is governed by gas diffusion parameters. For N2O emission a first-order notation has been implemented (Cho and Mills 1979; p. 97 in Müller 2000).

The total N2O emission is calculated (Eq. 7.46) by

$$
\sum E\_x = \sum k\_x \cdot \text{N}\_2\text{O}\_x \tag{7.46}
$$

where Ex is the emission rate (μmol N g−<sup>1</sup> h−1), kx is the emission rate constant (h−1) and N2Ox the soil N2O pool concentration (μmol N g−1). The symbol x stands for the process specific pools, i.e. nit, den, org and cod.

In the following section, two simplified approaches are presented that are based on the abundance of NH4 +, NO3 <sup>−</sup>, Norg and N2O. The methods are based on the assumption that N2O is derived from three uniformly labelled pools, i.e. NH4 +, NO3 <sup>−</sup> and Norg. and bases the analysis on the 15N enrichments of the different N species.


**Fig. 7.18** Three source model for determination of N2O pathways based on Rütting et al. (2010), see text for details

## *7.5.8 Source Partitioning to Quantify N2O Pathways*

#### **7.5.8.1 Three-Pool Model**

Based on the two-pool source-partitioning model by Stevens et al. (1997) a three-pool solver method (Rütting et al. 2010) was developed. The solver method (Microsoft Excel 2007) calculates the N2O fractions associated with NH4 <sup>+</sup> (n) and NO3 <sup>−</sup> (d) by minimisation of the absolute difference between observed and calculated 15N enrichments of N2O according to the equation:

$$a\_{N2O} = d \ast a\_d + n \ast a\_n + (1 - d - n) \ast a\_o \tag{7.47}$$

where *n* and *d* are the fractions related to the NH4 <sup>+</sup> and NO3 <sup>−</sup> pools, respectively, and *ad* , *an* and *ao* represent the 15N abundance of the NO3 <sup>−</sup>, NH4 <sup>+</sup> and Norg (assumed to be at natural abundance) respectively. The data are setup in an Excel spreadsheet and the Excel Solver is used to minimise the difference between measured and calculated 15N N2O enrichments (Rütting et al. 2010) (Fig. 7.18).

With this method, it is possible to subdivide total N2O emission into the three sources, autotrophic nitrification, denitrification and heterotrophic nitrification.

#### **7.5.8.2 Four-Pool Model**

The three-pool model has been developed further by Jansen-Willems et al. (2016) to analyse four simultaneous processes (nitrification, denitrification, co-denitrification and oxidation of organic matter). The assumption is that isotopic discrimination is negligible. The conceptual model for this approach is illustrated in Fig. 7.19.

**Fig. 7.19** Conceptual model to analyse N2O pathways according to Jansen-Willems et al. (2016)

#### **Background and development of the four-pool model**

Each N pool contains both 15N and 14N atoms. If one N atom would be randomly selected from a pool, the chance it would be a 15N atom is equal to the 15N atom fraction of that pool. So for the NH4 <sup>+</sup> pool this would be an, for the NO3 <sup>−</sup> pool ad, and for the organic-N pool ao. The chance it would be a 14N atom equals 1 minus the 15N atom fraction of that pool. So, for the NH4 <sup>+</sup> pool it would be 1-an, for the NO3 − pool 1-ad, and for the organic-N pool it would be 1-ao.

N2O consists of two N atoms. For nitrification, denitrification and oxidation of organic N, both N atoms come from the same pool. For these three processes:


In Eqs. 7.48 to 7.50, ax would be an for nitrification, ad for denitrification and ao for oxidation of organic N.

$$\text{Chance of zero}^{15}\text{N atoms} : (\mathbf{l} - \mathbf{a}\_{\mathbf{x}})(\mathbf{l} - \mathbf{a}\_{\mathbf{x}}) = \left(\mathbf{l} - \mathbf{a}\_{\mathbf{x}}\right)^{2} \tag{7.48}$$

$$\text{Chance of one}^{15}\text{N atom}: \mathbf{a}\_{\mathbf{x}}(1-\mathbf{a}\_{\mathbf{x}}) + (1-\mathbf{a}\_{\mathbf{x}})\mathbf{a}\_{\mathbf{x}} = 2(1-\mathbf{a}\_{\mathbf{x}})\mathbf{a}\_{\mathbf{x}}\tag{7.49}$$

$$\text{Chance of two}^{1\text{\textquotedblleft N atoms}} : \mathbf{a\_{x}a\_{x} = a\_{x}^{2}} \tag{7.50}$$

For co-denitrification, one atom comes from the NO3 <sup>−</sup>, and one comes from the organic N pool. So, for co-denitrification, the chance that N2O contains


$$\text{Chance of zero}^{15}\text{N atoms:}\,(1-\text{a}\_{\text{d}})(1-\text{a}\_{\text{o}})\tag{7.51}$$

$$\text{Chance of one}^{15}\text{N atom:}\,\mathbf{a}\_{\mathbf{d}}(1-\mathbf{a}\_{\mathbf{o}}) + (1-\mathbf{a}\_{\mathbf{d}})\mathbf{a}\_{\mathbf{0}}\tag{7.52}$$

$$\mathbf{\color{red}{Chance of two}^{15}\mathbf{N} \text{ atoms: } \mathbf{a\_{d}a\_{0}}}\tag{7.53}$$

The N2O in the gas sample is assumed to come from one of four processes. The fraction that comes from nitrification is written as *n*, the fraction that comes from denitrification is written as *d* and the fraction that comes from oxidation of organic matter is written as *o*. The fraction that comes from co-denitrification is written as *c*. As these are the only four processes considered, the four fractions should add up to one. Therefore, the following two equations apply:

$$\mathbf{a} + \mathbf{d} + \mathbf{o} + \mathbf{c} = \mathbf{l} \tag{7.54}$$

$$c = 1 - a - d - o \tag{7.55}$$

The fraction of N2O in the gas sample that is expected to contain zero 15N atoms can be calculated by multiplying the fraction of that sample from a specific process by the chance that the N2O from that process contains zero 15N atoms. So for nitrification, this would be *n*(1–an) 2, and for co-denitrification this would be (1–*n*–*d*−*o*) (1–ad) (1–a0). This should be done for all four processes, and then should be added together (Eq. 7.56). The fraction of the N2O in the gas sample that is expected to contain one 15N atom is calculated in the same way (Eq. 7.57), and the expected fraction containing two 15N atoms as well (Eq. 7.58)

Chance of zero 15N atoms:

$$m(1 - \mathbf{a\_n})^2 + d(1 - \mathbf{a\_d})^2 + o(1 - \mathbf{a\_o})^2 + (1 - n - d - o)(1 - \mathbf{a\_d})(1 - \mathbf{a\_0}) \tag{7.56}$$

Chance of one 15N atom:

$$2n(1-\mathbf{a\_{n}})\mathbf{a\_{n}}+2d(1-\mathbf{a\_{d}})\mathbf{a\_{d}}+2o(1-\mathbf{a\_{o}})\mathbf{a\_{o}}+(1-n-d-o)(\mathbf{a\_{d}}(1-\mathbf{a\_{0}})+\mathbf{a\_{0}}(1-\mathbf{a\_{d}})) \tag{7.57}$$

Chance of two 15N atoms:

$$
tau\mathrm{N}\_2 + d\mathrm{a}\_d^2 + o\mathrm{a}\mathrm{N}\_2 + (1 - n - d - o)\mathrm{a}\_d\mathrm{a}\_0\tag{7.58}
$$

#### **7.5.8.3 Mass Spectrometer Measurements and Calculation of Fractions**

To determine the fractions of the different processes 45R and 46R measurements are needed. These need to be corrected for the presence of 18O. Therefore, this means that 45R is the fraction of N2O molecules containing one 15N atom divided by the fraction of N2O molecules containing zero 15N atoms, and 46R is the fraction of N2O molecules containing two 15N atoms divided by the fraction of N2O molecules containing zero 15N atoms.

The expected fractions for N2O containing zero, one or two 15N atoms are given in Eqs. 7.56−7.58. In the study published by Jansen-Willems et al. (2016), a0 was set to 0.003663 (natural abundance), and an and ad were considered to be the 15N abundance of NH4 <sup>+</sup> and NO3 <sup>−</sup>. Using these values, *n*, *d* and *o* were quantified using the *fminsearchbnd* function in MatLab (The MathWorks Inc, Natick, MA). Thus, from this, *c* could be calculated according to Eq. 7.55.

## **References**


some observations and possible lessons, In: EGU (Ed.), EGU General Assembly. EGU EGU2015−11636


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 8 Climate-Smart Agriculture Practices for Mitigating Greenhouse Gas Emissions**

**M. Zaman, K. Kleineidam, L. Bakken, J. Berendt, C. Bracken, K. Butterbach-Bahl, Z. Cai, S. X. Chang, T. Clough, K. Dawar, W. X. Ding, P. Dörsch, M. dos Reis Martins, C. Eckhardt, S. Fiedler, T. Frosch, J. Goopy, C.-M. Görres, A. Gupta, S. Henjes, M. E. G. Hofmann, M. A. Horn, M. M. R. Jahangir, A. Jansen-Willems, K. Lenhart, L. Heng, D. Lewicka-Szczebak, G. Lucic, L. Merbold, J. Mohn, L. Molstad, G. Moser, P. Murphy, A. Sanz-Cobena, M. Šimek, S. Urquiaga, R. Well, N. Wrage-Mönnig, S. Zaman, J. Zhang, and C. Müller**

**Abstract** Agricultural lands make up approximately 37% of the global land surface, and agriculture is a significant source of greenhouse gas (GHG) emissions, including carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O). Those GHGs are

M. Zaman (B) · L. Heng

e-mail: m.zaman@iaea.org; zamanm\_99@yahoo.com

K. Kleineidam · C. Eckhardt · A. Jansen-Willems · G. Moser · C. Müller Institute of Plant Ecology, Justus Liebig University Giessen, Giessen, Germany

L. Bakken Norwegian University of Life Sciences (NMBU), Aas, Norway

J. Berendt · S. Fiedler · N. Wrage-Mönnig University of Rostock, Rostock, Germany

C. Bracken

School of Agriculture and Food Science and Earth Institute, University College Dublin, Dublin, Ireland

K. Butterbach-Bahl Institute of Meteorology and Climate Research, Atmospheric Environmental Research (IMK-IFU), Karlsruhe Institute of Technology, Karlsruhe, Germany

Z. Cai School of Geography Sciences, Nanjing Normal University, Jiangsu, China

S. X. Chang Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2E3, Canada

T. Clough Department of Soil & Physical Sciences, Faculty of Agriculture & Life Sciences, Lincoln University, Lincoln, New Zealand

© The Author(s) 2021 M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8\_8

303

Soil and Water Management & Crop Nutrition (SWMCN) Section, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency (IAEA), Vienna, Austria

responsible for the majority of the anthropogenic global warming effect. Agricultural GHG emissions are associated with agricultural soil management (e.g. tillage), use of both synthetic and organic fertilisers, livestock management, burning of fossil fuel for agricultural operations, and burning of agricultural residues and land use change. When natural ecosystems such as grasslands are converted to agricultural production, 20–40% of the soil organic carbon (SOC) is lost over time, following cultivation. We thus need to develop management practices that can maintain or even increase SOC storage in and reduce GHG emissions from agricultural ecosystems. We need to design systematic approaches and agricultural strategies that can ensure sustainable food production under predicted climate change scenarios, approaches that are being called climate-smart agriculture (CSA). Climate-smart agricultural management practices, including conservation tillage, use of cover crops and biochar application to agricultural fields, and strategic application of synthetic and organic fertilisers have been considered a way to reduce GHG emission from agriculture. Agricultural management practices can be improved to decreasing disturbance to the soil by decreasing the frequency and extent of cultivation as a way to minimise soil C loss and/or to increase soil C storage. Fertiliser nitrogen (N) use efficiency can be improved to reduce fertilizer N application and N loss. Management measures can also be taken to minimise agricultural biomass burning. This chapter reviews the current literature on CSA practices that are available to reduce GHG emissions and

K. Dawar

Department of Soil and environmental Sciences, University of Agriculture, Peshawar, Pakistan

W. X. Ding Institute of Soil Science, Chinese Academy of Sciences, Nanjing, China

P. Dörsch · L. Molstad Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Aas, Norway

T. Frosch

Leibniz Institute of Photonic Technology, Technical University Darmstadt, Darmstadt, Germany

J. Goopy International Livestock Research Institute (ILRI), Nairobi, Kenya

C.-M. Görres Department of Soil Science and Plant Nutrition/Department of Applied Ecology, Hochschule Geisenheim University, Geisenheim, Germany

A. Gupta Independent Consultant India, Mumbai, India

S. Henjes · M. A. Horn Institute of Microbiology, Leibniz University Hannover, Hannover, Germany

M. E. G. Hofmann Picarro B.V., 's-Hertogenbosch, The Netherlands

M. M. R. Jahangir Department of Soil Science, Bangladesh Agricultural University, Mymensingh, Bangladesh increase soil C sequestration and develops a guideline on best management practices to reduce GHG emissions, increase C sequestration, and enhance crop productivity in agricultural production systems.

**Keywords** Agriculture · Carbon dioxide · Climate-smart agriculture · <sup>C</sup> sequestration · GHG · Methane · Mitigation · Nitrous oxide

## **8.1 Introduction on Climate-Smart Agriculture Practices and Greenhouse Gas Emissions**

Agriculture is a major source of greenhouse gases (GHGs) that affect climate change and is also itself a victim of climate change. Agricultural lands make up 37.6% of the global land surface, and agriculture is a significant source of GHG emissions (IPCC 2014; Smith et al. 2008), where CO2, CH4 and N2O are the major forms of trace gases that are responsible for the majority of the global warming effect. Agricultural GHG emissions are associated with agricultural soil management (e.g. tillage), use of both synthetic and organic fertilisers, livestock management, burning of fossil fuel for agricultural operations and burning of agricultural residues. In particular, agriculture can be the source for 52% and 84%

D. Lewicka-Szczebak

G. Lucic Picarro Inc., Santa Clara, CA, USA

L. Merbold Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya

J. Mohn

Laboratory for Air Pollution & Environmental Technology, Empa Dübendorf, Dübendorf, Switzerland

P. Murphy

Environment & Sustainable Resource Management Section, School of Agriculture & Food Science, and UCD Earth Institute, University College, Dublin, Ireland

A. Sanz-Cobena

Research Center for the Management of Environmental and Agricultural Risks (CEIGRAM), ETSIAAB, Universidad Politécnica de Madrid, Madrid, Spain

M. Šimek

Institute of Soil Biology, Biology Centre of the Czech Academy of Sciences, and Faculty of Science, University of South Bohemia, Ceské Budˇ ˇ ejovice, Czech Republic

M. dos Reis Martins · S. Urquiaga

EMBRAPA Agrobiologia Seropédica, Brazilian Agricultural Research Corporation, Seropedica, RJ, Brazil

R. Well

Thünen Institute of Climate-Smart Agriculture, Braunschweig, Germany

Laboratory of Isotope Geology and Geoecology, Institute of Geological Sciences, University of Wrocław, Wrocław, Poland

of global anthropogenic emissions of CH4 and N2O, respectively (Smith et al. 2008). Since the global warming potentials of CH4 and N2O are much higher than that of CO2 based on per unit mass and a 100-year time frame (IPCC 2014), advanced concepts are required to reduce agricultural emissions of CH4 and N2O.

In addition to causing increased levels of GHG emission, human settlement in previously unpopulated areas means that natural ecosystems are converted to agricultural production, with 20–40% of the SOC lost following cultivation and with most of that loss occurring in the first a few years (Davidson and Ackerman 1993). A recent estimate indicates that 133 billion tonnes of SOC, which is about 8% of the total global SOC stock, had been lost from the top 2 metres of soil on a global scale since agriculture started about 12,000 years ago, with the rate of loss dramatically increased since the industrial revolution (Sanderman et al. 2017). The Sanderman et al. (2017) study also indicated that the percentage of SOC loss was greater on cropland but the total amount of SOC loss was slightly higher on grazing land as more than twice as much land is grazed. This indicates that there is a greater potential to improve the % SOC gain in cropland but there is a greater potential to increase total SOC storage in grazing land. One of the key aspects of SOC is that the soil and vegetation stores about three times the organic C of the atmosphere (Plate 8.1; FAO 2004), and thus small changes in the organic C stock in the soil and vegetation can cause a large effect on atmospheric CO2 concentration; therefore, great efforts must be made to increase SOC storage in and to reduce GHG emissions from terrestrial ecosystems. In managed systems, SOC storage can be increased by management practices such as avoiding the burning of crop residues after harvest, and the application of composts and biochar and animal manure to increase organic C input to the soil.

There is significant potential for the agriculture sector to contribute to the reduction of anthropogenic sources of GHG emissions since agriculture is a large source of GHG and to increase soil C storage since much soil C has been lost through cultivation (Plate 8.2; top panel). For example, agricultural management practices can be improved to reduce disturbance to the soil by decreasing the frequency and extent of cultivation as a way to minimise soil C loss and/or to increase soil C storage; if permanent vegetation can be maintained, soil C storage can increase, benefiting from the C cycle becoming more closed in the system and the soil being able to trap more C (Plate 8.2; bottom panel). Fertiliser N use efficiency (NUE) can be improved through strategic application of fertiliser so as to reduce N loss, whether it is through

S. Zaman

J. Zhang

K. Lenhart

Bingen University of Applied Sciences, Berlinstr. 109, Bingen 55411, Germany

University of Canterbury, Christchurch, New Zealand

School of Geography, Nanjing Normal University, Nanjing, China

**Plate 8.1** An illustration of the distribution of carbon in terrestrial ecosystems, including the atmosphere. (*Source* Schwartz 2014)

leaching or gaseous form of N loss. Management measures can also be taken to minimise agricultural biomass burning. Climate-smart agriculture (CSA) management practices, including strategic use of synthetic and organic fertilisers and water, conservation tillage, use of cover crops, and the use of amendments such as biochar, nitrification inhibitors and lime to agricultural fields, have been considered a way to reduce GHG emissions from agriculture (Bai et al. 2019; Lipper et al. 2014; Zaman et al. 2008a and 2009; Zaman and Blennerhassett 2010). The FAO defines CSA as a systematic approach for developing agricultural strategies that can ensure sustainable food security under predicted climate change scenarios (FAO 2013). Based on this definition, a range of agricultural practices can be developed to help improve food security and environmental quality simultaneously in the context of global change.

**Plate 8.2** An illustration of a disturbance to the soil causes an increased release of CO2 and other greenhouse gases (and leading to the soil being leaky, open treasure chest) and in soils with undisturbed vegetation will cause the carbon cycle to be more closed (*Source* FAO 2013)

Since the soil can act as a sink or source for CO2 and affect climate change, if we can enhance the C sink strength and remove more CO2 from the atmosphere by adopting CSA, then we can be in a win–win situation by not only combating the negative effects of climate change but also improving soil quality and health, including the retention of nutrients and water, and increased agricultural productivity.

## **8.2 Climate-Smart Agricultural Technology to Reduce GHG Emissions**

Climate-smart agriculture emphasises on improving risk management, enhancing information flows and promoting local institutions to increase the adaptive capacity of communities to climate change (Campbell et al. 2014), as such CSA plays a pivotal role in maintainable development. Climate-smart agricultural practices such as the use of cover crops, amendments and tillage management play key roles in reducing agricultural GHG emissions. Take N2O, a potent GHG, as an example, the emission of N2O is affected by many factors, such as the use of N process inhibitors (urease and nitrification inhibitors), the strategic application of synthetic fertilisers (the right type, the right amount that is based on crop requirement and soil tests, at the right plant growth stage, and the right method (e.g. even spread)), avoiding the application of N fertilisers to very wet or very dry soils, maintaining soil pH above 6 by adding lime to lower the N2O:N2 ratio, co-application of animal manure and chemical fertilisers to provide a more balanced nutrient supply, optimising animal stocking rate to avoid over-grazing, keeping animals off the pasture especially in the wet season to minimise N input and avoid soil compaction, and minimising the excessive use of farm machinery. Many of those fall under CSA practices. Below, key CSA practices and their effect on GHG emissions from agriculture are discussed in detail.

## *8.2.1 Nitrogen Process Inhibitors and Greenhouse Gas Emissions*

Multiple microbial soil processes contribute to GHG emissions. One of the major processes contributing to GHG emissions is the mineralization of organic matter by microbial organisms, a process also called soil heterotrophic respiration, where organic C is converted to CO2 and released to the atmosphere. Nitrification and denitrification are key processes contributing to N2O emissions from the soil, where ammonia-oxidation and successive nitrifier denitrification or denitrification, among other processes, can be important pathways for N2O emissions (Fig. 7.8) (Guo et al. 2018). In the ammonia oxidation process, N2O can be produced by the chemical decomposition of hydroxylamine (NH2OH). One potential way to mitigate N2O emissions is to use nitrification and urease inhibitors to slow down the rate of nitrification and reduce the availability of the substrate (NH4 +) for nitrification.

#### **8.2.1.1 Synthetic Nitrification Inhibitors**

Nitrification is the process where ammonium or ammonia is converted to nitrate, via nitrite as an intermediate N species. Since the formation of nitrite is usually the rate-limiting step in the nitrification process, nitrite usually does not accumulate in the soil, unless the soil has a high pH where nitrite oxidisers are inhibited (Rodgers 1986). During the nitrification process, N2O can be formed and emitted from the soil. The final product of nitrification, i.e nitrate, is subject to leaching losses, and when anaerobic conditions develop, nitrate is denitrified thus leading to the formation of N2O. Both nitrate leaching and denitrification are major pathways for N losses from the soil. Reducing the rate of nitrification can both conserve N in the soil and reduce N2O emissions (Abalos et al. 2012). In this respect, the application of nitrification inhibitors can effectively reduce nitrification rates and the buildup of nitrate which can further be denitrified. Nitrification inhibitors are a group of chemical compounds that slow down the conversion rate of ammonium or ammonia to nitrate by inhibiting the ammonia monooxygenase activity, through disrupting the activity of *Nitrosomonas* and *Nitrobacter* bacteria (Abalos et al. 2014). Nitrification inhibitors are applied with ammonium-, ammonia- or urea-based fertilisers as the application of such fertilisers substantially increases the rate of nitrification. The need for nitrification inhibitors in non-fertilised soils is rare as nitrification rates in such soils are low. The ammonium or ammonia can come from urine, manures, composts or crop residues as they decompose, or fertilisers such as ammonium sulphate or urea (Rodgers 1986). Under suitable conditions (e.g. warm soil temperature and a moisture content near field capacity, when there is still ample O2 available), nitrification occurs within days or weeks of the application of ammonium-based fertilisers and thus if nitrification can be effectively reduced within that timeframe, N loss from the system can be substantially reduced (Sanz-Cobena et al. 2017).

There are at least eight compounds commercially available as nitrification inhibitors but the most commonly used ones are 2-chloro-6-(trichloromethyl) pyridine (nitrapyrin), dicyandiamide (DCD), 3,4-dimethylpyrazole phosphate (DMPP, e.g. ENTEC®) (IPNI undated) and pronitradine (Centuro™). Nitrapyrin (commercial product: N-Serve™ and Instinct™) can be applied to the soil in different ways: if anhydrous ammonia is the fertiliser being applied, the nitrification inhibitor can be injected with the fertiliser, if solid N fertilisers are being applied, the inhibitor can be coated onto the fertiliser, and if manure is used as an organic fertiliser, the inhibitor can be mixed with the manure before manure application. Nitrapyrin is usually effective for less than 30 days after being applied to the soil and is volatile, therefore, the best way to apply this inhibitor is to incorporate it into the soil; DMPP can be effective for reducing nitrification rates for 25–70 days and is usually preblended with fertilisers; DCD can last 25–55 days and can be coated on solid fertilisers (Sanz-Cobena et al. 2012), or surface applied to soils that have been applied with manure or on grazing land to reduce nitrification from urine patches (IPNI undated). While DMPP is somewhat immobile, DCD can be relatively easily leached from the soil; those behaviours of the nitrification inhibitors should be considered when they are applied in the field.

DCD and DMPP have been found to be equally effective in changing soil inorganic N content, leaching of dissolved inorganic N (DIN) and emissions of N2O in a recent meta-analysis of field trials reported in the literature (Yang et al. 2016). Their costbenefit analysis (CBA) showed that the economic benefit was about seven times greater with DCD than with DMPP when applied with ammonium-based fertilisers to reduce nitrification (Yang et al. 2016). Those two nitrification inhibitors are among the most commonly used. DCD is cheaper and less volatile, but the application rate of DMPP is typically one-tenth that of DCD, and DMPP has a lower eco-toxicological effect for plants as is summarised in Yang et al. (2016). In an Australian study, DMPP application (as ENTEC®) decreased N2O emissions by 15% in a subtropical pasture in Queensland (Lam et al. 2018).

The effectiveness of the nitrification inhibitors can be affected by soil properties such as soil water content (Barrena et al. 2017), and soil organic matter and clay content (Zhu et al. 2019). Zhu et al. (2019) reported that the efficiency of DMPP in reducing nitrification and N2O emissions was lower in soils with high organic matter and clay contents, likely due to the high rates of adsorption of DMPP by soil organic matter and clay. The effectiveness of DCD in reducing N2O emissions from urine patches in New Zealand is highly season-specific, with reductions of 52, 39 and 16% in autumn, spring and summer, respectively, but DCD application increased NH3 emissions by 56, 9 and 17% in the respective seasons (Zaman et al. 2009). Management practices can also affect the efficiency of nitrification inhibitors. For example, biochar application to the soil has been shown to decrease the efficiency of DMPP both at 40 and 80% of water-filled pore space (WFPS) in a laboratory incubation study (Fuertes-Mendizábal et al. 2019). The use of nitrification inhibitors increases crop yield and NUE, but the effectiveness was greatest when they are used in coarse-textured soils, irrigated systems and/or crops receiving high rates of N fertiliser input (Abalos et al. 2014).

#### **8.2.1.2 Synthetic Urease Inhibitors**

Urease inhibitors retard the activity of urease, which exists in the soil and plant residues. Urease is involved in the conversion of urea to ammonium in a process called hydrolysis. Unfortunately, the urea hydrolysis process increases the pH of the soil and causes a large proportion of the formed ammonium to be volatilized as ammonia. Urease inhibitors would slow down the rate of hydrolysis or the rate of release of ammonium, reduce the loss of N as ammonia through volatilization, and increase the NUE of urea fertilisers applied to the soil. One of the main reasons for the improved NUE is for more urea to be washed into the soil over time (when the rate of hydrolysis is suppressed), as urea is highly soluble in water. The better contact with the soil increases the chances of the released ammonium to be adsorbed by the cation exchange sites.

The N-(n-butyl) thiophosphoric triamide (NBPT) and N-(n-propyl) thiophosphoric triamide (NPPT) are two chemicals that have been shown to be effective in inhibiting urease activities (Rodriguez et al. 2019; Lam et al. 2018; Zaman et al. 2009; Sanz-Cobena et al. 2008). Products containing those urease inhibitors include Agrotain™ (that contains NBPT) and Limus™ (that contains both NBPT and NPPT). The application of NBPT prior to urine deposition was more effective in reducing ammonia volatilization loss (17.5–27.6% reduction) as compared with applying the NBPT after the urine deposition (0.6–2.9% reduction) in pastureland in New Zealand (Rodriguez et al. 2019). The effectiveness of NBPT is highly season-specific for reducing NH3 volatilization loss from urine patches in New Zealand, with reductions of 29, 93 and 31% in autumn, spring and summer, respectively (Zaman et al. 2009). An Australian study reached similar conclusions that urea applied with NBPT (as Green UreaNV®) was effective in decreasing NH3 volatilization (by 44%) in a subtropical pasture in Queensland (Lam et al. 2018). The effectiveness of NBPT was greater in alkaline soils (pH ≥ 8) (Abalos et al. 2014). It has also been observed, under laboratory conditions, that application of urease inhibitors in soils where nitrification is the main pathway in the production of N2O (i.e. WFPS < 50%), could be an effective way to mitigate these N losses (Sanz-Cobena et al. 2014).

The combined application of urease inhibitor + the nitrification inhibitor, DCD, (e.g. Agrotain Plus) to inhibit both the hydrolysis of urea and the nitrification processes to minimise the N loss showed best results in maximising NUE. When urea and ammonium nitrate (UAN) was applied at 150 mg N kg−<sup>1</sup> in a sandy loam soil in the United States with Agrotain Plus, N2O emissions were reduced by 78% compared to the control (Cai et al. 2018). The use of double inhibitors containing both NBPT and DCD (3:7) in a New Zealand study reduced NH3 volatilization by 14, 78 and 9% in autumn, spring and summer, respectively, and N2O emissions by 37, 67 and 28%, respectively, from urine patches (Zaman et al. 2009). However, another study showed that adding DCD with NBPT did not further reduce NH3 volatilization loss, but in fact enhanced the volatilization loss by maintaining a higher soil NH4 + concentration and pH for a longer period of time after urea application, indicating that DCD co-applied with NBPT and urea could offset the effect of NBPT in reducing volatilization losses (Soares et al. 2012).

#### **8.2.1.3 Biological Nitrification Inhibitors**

Some plant species can release secondary metabolites through root exudation and/or from leaf litter. Such metabolites have the ability to suppress microbial nitrification (Souri and Neumann 2018). In some earlier studies, detectable biological nitrification inhibition (BNI) was found in root exudates of sorghum (*Sorghum bicolor* (L.)), pearl millet (*Pennisetum glaucum* (L.) R. Br.) and groundnut (*Arachis hypogaea* (L.)) among tested cereal and legume crops, while *Brachiaria humidicola* (Rendle) Schweick and *B. decumbens* Stapf had the highest BNI capacity among pasture grass species tested (Subbarao et al. 2007). In addition, when BNI compounds from root exudates were applied to the soil, their inhibitory effects on NO3 <sup>−</sup> formation lasted for more than 50 days (Subbarao et al. 2007). Linoleic acid, α-linolenic acid, and methyl linoleate, a fatty acid methyl ester of linoleic acid, are some of the example compounds that are effective in biologically inhibiting nitrification (Subbarao et al. 2008). Subbarao et al. (2007) suggested that some level of BNI is likely a widespread phenomenon in tropical pasture grass species and those properties could be used to suppress nitrification in natural or managed systems. The production in and release from plants of BNIs are triggered by the presence of NH4 <sup>+</sup> in the rhizosphere of plants, which means that BNIs are released where the majority of the nitrifier populations reside (Subbarao et al. 2013a). The pH in the rhizosphere will affect the release of BNIs from roots; for example, sorghum plants release BNIs from their roots in the presence of NH4 <sup>+</sup> when the rhizosphere pH is between 5.0 and 6.0 (Subbarao et al. 2013b), indicating the usefulness of BNI to reduce nitrification in alkaline soils will be non-existent or very low. More research is needed to understand and take advantage of BNI in agricultural production systems to reduce N loss and improve NUE.

## *8.2.2 Soil Amendments and Greenhouse Gas Emissions*

#### **8.2.2.1 Mulch**

Addition of mulch to the soil will change the availability of carbon (C) and other nutrients to microbial populations and will thus affect soil GHG emissions. Existence of a litter layer (mulch) can induce microbial N immobilisation in the litter layer, and result in reduced available N and reduced plant growth rate; however, the litter layer may benefit plant growth by conserving soil moisture (Matsushima and Chang 2006) but may reduce soil temperature and N mineralization rates (Matsushima and Chang 2007). Addition of mulch can immobilise mineral N in the soil and reduce the availability of NH4 <sup>+</sup> for nitrification and NO3 <sup>−</sup> for denitrification, and thus reduce N2O emissions as compared to no mulch addition (Wu et al. 2013). Using wood bark mulch reduced the nitrate concentration in the soil and cut soil N2O emissions by up to 28% in a grape (*Vitus vinifera* L. cv. Merlot) yard on a sandy loam soil in British Columbia, Canada, when measured over a two-year period (Fentabil et al. 2016). However, overloading of straw to the soil surface can delay seed germination and result in the need for additional fertiliser supply to compensate for the N that may be immobilised in the critical period of the early growing season (Procházková et al. 2003). As far as CO2 emissions are concerned, mulching will usually result in increased CO2 emissions due to the addition of labile C in the mulch, with the rate of CO2 emissions increasing with the increased rate of mulch addition (Wu et al. 2013). Major anthropogenic sources of methane emissions include fossil fuel production, landfills and livestock farming, but some agricultural soils can be an anthropogenic source of methane emissions as well (IPCC 2007). In rice paddy systems, straw application has been shown to increase CH4 emissions (Bossio et al. 1999; Ma et al. 2008); however, straw addition significantly reduced CH4 emissions under an aerobic condition in a laboratory incubation experiment, indicating that under upland conditions, straw application increased the soil's ability to take up CH4 (Tate et al. 2007).

#### **8.2.2.2 Biochar**

Biochar has been widely studied for its effects on GHG emissions. Biochar addition to the soil can change a range of soil properties, including the cycling of C and N. Biochar application has been widely reported to reduce N2O emissions (Wu et al. 2013; Cayuela et al. 2013; Chang et al. 2016; Hüppi et al. 2016). The application of biochar to the soil has been shown to reduce denitrification and decrease N2O emissions by 10–90% when tested on 14 different agricultural soils, where a consistent reduction of the N2O/(N2 + N2O) ratio was observed, indicating that biochar reduces N2O emissions by facilitating the last step of the denitrification process and producing more N2 rather than N2O (Cayuela et al. 2013). However, in some soils, biochar application can stimulate nitrification and increase N2O emissions; therefore, the effect of biochar application on N2O emissions is related to the dominant N2O formation pathway that operates in a soil (Sánchez-García et al. 2014).

Biochar application has been reported to reduce CH4 emissions from paddy soils, one of the largest anthropogenic sources of CH4 emissions on a global scale (Feng et al. 2012). The reduction in paddy CH4 emissions by biochar application was not a result of the inhibition of methanogenic archaea, but resulted from increased methanotrophic proteobacterial abundances and decreased ratios of methanogenic to methanotrophic abundances (Feng et al. 2012).

In agricultural production systems, a large quantity of crop residues are produced and the return of crop residues in the raw form vs after the crop residue is converted to biochar can have dramatic effects on the emissions of all three trace gases (Wu et al. 2013), and most research suggests that there are substantial beneficial effects to be gained in mitigating climate change by converting crop residues to biochar and applying biochar to the soil instead. The effect of biochar on GHG emissions itself is highly complex as many factors alter the function of biochar on C and N transformation processes and thus GHG emissions. When the effect of the stable C input in the form of biochar is considered, the application of biochar is considered an effective technique to mitigate climate change due to its negative emissions potential (0.7 Gt Ceq. Yr−1) and its lower impact on land, water use, albedo, energy requirement and cost as compared to other negative emissions technologies such as direct air capture, increased weathering that takes up CO2 from the air, bioenergy projects with C capture and storage, and afforestation/deforestation (Smith 2016).

#### **8.2.2.3 Liming to Shift the Balance Between N2O and N2 Emissions**

Soil pH is one of the key regulators of microbiological processes that affect N2O and N2 production and their ratio. The soil pH threshold for nitrification is 5; however, nitrification can occur even below pH 5 as some nitrifier strains are adapted to acidic conditions (Bouwman 1990). Denitrification has been reported to occur over a wide range of soil pH (5–8) (Flessa et al. 1998); however, laboratory experiments with artificially adjusted soil pH suggest that, under optimised conditions (very low pO2, NO3 <sup>−</sup> and with glucose amendment), denitrification can proceed even at pHs below 4 or above 10 (Šimek et al. 2002). Numerous laboratory and field studies have shown that soil pH affects N2O and N2 emissions and thus the ratio of these gases (e.g. Stevens and Laughlin 1998). In experiments conducted under controlled environmental conditions, raising soil pH to 7 through lime application has been found to significantly increase N2 emissions from pasture and wetland soils treated with cow urine, urea and KNO3 at 200 kg N ha−<sup>1</sup> (Zaman et al. 2007, 2008b). Similar trends of enhanced N2 emissions after raising soil pH to 7 was observed in pasture soils treated with urea/urine in a field experiment (Zaman and Nguyen 2010). In another study, a site with the greatest animal impact, the ratio of N2 to N2O produced during denitrifying enzyme activity (DEA) measurements was fivefold higher, and the pH was 2 units higher, than a site with the least animal impact, indicating that soil conditions were favourable for production of N2 rather than N2O in the area with intense excretal returns and treading (Hynšt et al. 2007).

Most researchers attribute high N2O and low N2 emissions in acidic conditions to the suppression of N2O-reductase (inhibition starts at soil pH 4.5) (Daum and Schenk 1998; Flessa et al. 1998; Stevens and Laughlin 1998; Zaman et al. 2007). It is also possible that denitrifying enzymes are susceptible at low soil pH and produce N2O from intermediate products (Nägele and Conrad 1990). However, the lower rates of N2 emissions and higher N2O:N2 ratio at low soil pH could be due to lower amounts of soil organic C and mineral N available to the denitrifying population rather than a direct effect of low pH on denitrification enzymes (Šimek and Cooper 2002). Regardless of the biochemical mechanism for soil pH effects on N2 emissions, raising soil pH through the application of amendments such as lime appears a viable approach to mitigate N2O emissions (Šimek et al. 2002; Zaman and Nguyen 2010; Zaman et al. 2007, 2008b).

## *8.2.3 Fertiliser Type and Management and Greenhouse Gas Emissions*

The use of different fertiliser types and the management of fertiliser applications can have marked effects on nitrification, denitrification and GHG emission rates (Mosier et al. 2006; Wang et al. 2018). As discussed in the earlier section, the type of fertiliser (some fertilisers are acid-forming while others raise soil pH when applied to the soil) applied can affect the total amount of GHG emitted as well as the N2O:N ratio in the emissions. A number of agronomic practices have been widely tested to minimise N losses from agricultural production systems, for example, alteration of the rate or timing of fertiliser application, such as autumn vs. spring, basal vs. broadcast, deep vs. surface applications, point injection placement of solutions, foliar applications of urea (Subbarao et al. 2013b), and split application. Fertilisers such as polythenecoated urea (PCU) that releases N slowly in the soil have been demonstrated to reduce nitrification (Zvomuya et al. 2003); however, the use of such fertilisers can be limited by the high cost for purchasing such fertilisers (Subbarao et al. 2013b).

## *8.2.4 Cropping Systems and Greenhouse Gas Emissions*

#### **8.2.4.1 Agroecosystems**

The type of cropping system used has a significant effect on GHG emissions as cropping systems will differ in their fertilisation regime, crop productivity (and thus the amount of organic matter input to and retention in the soil), crop species being used (N-fixing vs. non-N-fixing crop species), water management practices, and tillage management, among others (Snyder et al. 2009). On an extreme case, GHG emission rates and timing will be dramatically different between rice cultivation in flooded fields vs. wheat production in well-drained upland sites in a rice-wheat rotation, where the rice-wheat belt makes up 24–27 million ha in South and East Asia (Wassmann et al. 2004).

Rice cultivation in paddy fields is a unique cropping system in Southeast Asia. In Vietnam, for example, rice production is the largest source of agricultural GHG, with 37.4 Tg CO2 equivalent of total emissions, that account for 58% of agricultural GHG emissions in that country (United Nations 2013). Many factors, such as the management of fertiliser applications, animal manure and crop residue management, water regime used during rice production, and use of urease and/or nitrification inhibitors, will affect the emissions of GHGs from paddy fields. One of the biggest concerns of GHG emissions from paddy fields is the emission of CH4 as paddy fields are mostly submerged in water during the growing season and anaerobic condition caused substantial CH4 emission to occur. The CH4 emissions are the balance of CH4 production and oxidation in the soil, and are affected by factors influencing the transportation of CH4 from the anoxic soil/free-standing water to the atmosphere (Aulakh et al. 2001). Up to 80% of the CH4 produced in paddy soils is oxidised in the rhizosphere or microsites that are less anaerobic before it is released to the atmosphere (Sass et al. 1991; Holzapfel-Pschorn et al. 1985) and this process helps to dramatically reduce the rate of CH4 emissions from paddy soils.

Proper water management can substantially reduce CH4 emissions from paddy fields; midseason drainage has been reported to reduce CH4 emissions by 44% and alternating wetting–drying cycles at ten-day intervals by 61% as compared to continuously flooded management in southeast China (Lu et al. 2000). Application of urea can increase CH4 emissions from paddy fields as ammonium can inhibit CH4 oxidation (Conrad and Rothfuss 1991). However, others found that application of ammonium-based fertilisers can reduce CH4 emissions as ammonium enhances methanotrophic bacteria activities in the rhizosphere of rice plants (Bodelier et al. 2000). The effect of ammonium on CH4 oxidation is dependent on the CH4 concentration: inhibition at low initial CH4 concentration (500 μl l−1) but stimulation at high initial CH4 concentration (2000 μl l−1) (Cai and Mosier 2000). However, ammonium sulphate has been found to be a promising fertiliser to use, as opposed to urea, to mitigate CH4 emissions as sulphate enhances sulphate-reducing bacteria activities, which decrease the availability of substrates for methanogens in submerged soils (Yagi et al. 1997). Reductions in CH4 emissions after ammonium sulphate application has been reported to range from 10 to 67% (Schütz et al. 1989; Wassmann et al. 2000). Ammonium is the preferred N form for rice, therefore, applying ammonium form of N that does not enhance CH4 emissions would be beneficial for the environment. Application of sulphate-containing phosphorus fertilisers (Achtnich et al. 1995) and gypsum (CaSO4) (Lindau und Bollich 1993), a common soil amendment for sodic and/or alkaline soil reclamation, have been shown to reduce CH4 emissions from paddy soils. Therefore, choice of fertiliser is important in minimising CH4 emissions from rice paddies, in this case, the use of sulphate-based N or phosphorus fertilisers would be preferred in flooded rice fields.

Applying organic fertilisers such as animal manure and returning crop residue to the soil are necessary to maintain and enhance the sustainability of rice production. However, addition of organic materials to rice fields would increase CH4 emissions, and organic materials application to the soil can lower the soil redox potential and supply C to methanogens that are responsible for CH4 production in paddy soils. The production of CH4 in paddy soils is markedly influenced by the quality and quantity of organic materials added to the soil (Minasny et al. 2017). Therefore, proper manure and crop residue management strategies need to be developed to achieve environment-friendly rice cultivation. One potential alternative to the application of crop residues to paddy soil is to convert crop residues to biochar that is slow to decompose after soil application (Ippolito et al. 2012); biochar application to soil can increase soil aeration and soil C content but mitigate CH4 emissions as compared with the conventional crop residue application (Feng et al. 2012; Karhu et al. 2011; Liu et al. 2011; Xie et al. 2013). However, other studies reported increased CH4 emissions from rice fields after biochar application, which may be related to increased substrate supply and enhanced environment for methanogenic activity, and increased rice growth (Knoblauch et al. 2011; Lehmann et al. 2011). The effects of biochar application on CH4 emissions is thus dependent on soil type, agricultural management practices used, and the type of biochar applied (Lehmann et al. 2011; Waters et al. 2011). Site-specific research should be conducted before any recommendation on agricultural management practices is made for farmers to adopt.

In rice paddies, the flooded condition is conducive for denitrification to occur. Even though less attention has been paid to N2O as compared to CH4 emissions from paddy fields, N loss in the form of NH3 volatilisation and N2O emissions and NO3 <sup>−</sup> leaching affect GHG emissions as well as NUE in rice production systems. Therefore, urease inhibitors can be used to slow down the rate of urea hydrolysis when urea is used as the main N fertiliser (Rogers et al. 2015), while the use of nitrification inhibitors can reduce nitrate leaching loss (Li et al. 2008) and N2O emissions (Majumdar et al. 2000; Kumar et al. 2000).

#### **8.2.4.2 Organic Farming**

Organic farming has long been considered a viable agricultural practice to improve soil health, reduce the resource use intensity, reduce the environmental impact of agriculture and improve food quality (Squalli and Adamkiewicz 2018). The comprehensive study of Squalli and Adamkiewicz (2018), based on longitudinal state-level data in the United States collected between 1997 and 2010, demonstrates that a 1% increase in organic farming acreage can result in a 0.049% reduction in GHG emissions; however, they showed that the net effect of organic farming on GHG emissions is dependent on the contribution of transportation (fuel burning) on methane and nitrous oxide emissions, even though their calculation indicates that the negative environmental effect of transportation output associated with organic food production is small relative to the environmental benefits of organic farming. A study in Switzerland showed a 40% reduction in N2O emissions for organic compared to conventional systems; however, yield-scaled N2O emissions under silage maize were not different between organic and conventional systems (Skinner et al. 2019), indicating that even if we consider the lower yield in organic systems, organic farming does not have a negative effect on GHG emissions. In contrast, Aguilera et al. (2015) and Cayuela et al. (2017) showed in two Meta-analyses that fertilisation with solid organic manures (the most used form of N fertilisers in organic agroecosystems) led to the lowest N2O emission factor in Mediterranean cropping systems worldwide. On the other hand, we must recognise that organic farming does have a lower crop yield (c. 25% on average) which may require a larger area of land to be cultivated to produce the same amount of food compared to conventional farming (Kniss et al. 2016).

#### **8.2.4.3 Row, Intercropping and Crop Rotation**

When dealing with a cropping system that involves rice production (e.g. in Southeast Asia), it is often difficult to strike a balance in the mitigation of different GHGs. Using a rice-wheat crop rotation system as an example, the two crops markedly differ in their nature and intensity of GHG fluxes, where CH4 emissions are a major contributor to GHG emissions from rice paddies. In rice production systems, water regimes, rice cultivars and soil properties all markedly affect CH4 emissions (Cai et al. 2003). In addition, N2O is also emitted in large quantities from rice production systems following aerobic-anaerobic cycles; on the other hand, N2O is emitted in short-term pulses after fertilisation, heavy rainfall or irrigation events and is the main GHG emissions of concern in upland wheat production systems (Wassmann et al. 2004). It is often difficult to balance emissions between CH4 and N2O when designing GHG mitigation strategies in a rice-wheat system, as measures to reduce CH4 emissions often intensify N2O emissions (Wassmann et al. 2004).

In evaluating the effect of intercropping on GHG emissions, Ricord (2018) studied GHG emissions from a sole maize crop, a sole soybean crop and a maize–soybean intercrop and found that the cereal–legume intercropping system effectively reduced N2O emissions. In a similar study on the North China Plain, N2O fluxes were lower from maize–soybean intercropping than a maize monoculture system in three growing seasons (2013–2015), when all cropping systems were applied with 240 kg N ha−<sup>1</sup> as urea in two split applications (Shen et al. 2018). Shen et al. (2018) showed that the fertiliser N loss as N2O was lower in the maize-soybean intercropping (1.6%) and soybean monoculture (1.7%) than in the maize monoculture (2.3%), concluding that maize–soybean intercropping should be recommended as a climate-smart cropping systems for use on the North China Plain. A maize–wheat intercropping system coupled with reduced tillage and stubble mulching can increase grain production and decrease C emissions in an arid area in northwest China (Hu et al. 2015; Yin et al. 2018). Therefore, choice of a cropping system to use and the associated management practices are important decisions to make to minimise GHG emissions.

## **8.3 Climate-Smart Agriculture (CSA) Practices and C Sequestration**

Climate-smart agriculture has three key objectives: (1) to increase agricultural production per unit land area so as to increase income, food security and community development, (2) to improve the adaptive capacity at multiple levels, i.e. from the farm to the national level and (3) to reduce GHG emissions and to enhance C sinks in ecosystems (Campbell et al. 2014). Climate-smart agricultural management practices, including conservation tillage, use of cover crops, and biochar application to agricultural fields, has been widely considered a way to reduce GHG emissions from agriculture.

Conventional tillage has been identified as one of the causes of widespread land degradation problems, such as deterioration of soil structure, soil erosion and decreased soil fertility, affecting the long-term sustainability of agricultural production (Barber et al. 1996). Many climate-smart agricultural technologies have been tested to improve SOC storage in the agricultural landscape, and many of those have been demonstrated to be effective. Climate-smart agricultural technologies such as the use of cover crops, use of perennial crops, application of manure and biochar, reduced/minimum tillage or zero tillage, and crop rotation have all been shown to increase SOC storage.

Field experiments in Australia on light-textured soils in southern Australia indicate that conservation tillage (3–19 years in duration) was effective in increasing SOC levels as compared with conventional tillage, but only in areas with >500 mm annual precipitation and in the top 2.5–10.0 cm of the soil; the lack of conservation tillage effects on SOC levels in other climatic condition or soil layers was mainly attributed to low crop yield related to low rainfall, partial removal of stubble due to grazing and the high decomposition rate in areas with high air temperature (Chan et al. 2004).

## **8.4 Life Cycle Assessment (LCA) for Estimating the C Footprint of Agro-Food Systems**

The proposal of effective GHG mitigation strategies in the agri-food sector needs to be based on a whole-system approach. This means that not only direct emissions but also indirect GHG losses (both upstream and downstream from the production systems) must be considered. For that purpose, the calculation and use of the "C footprint" (CF) or "C budget" of agri-food entities, from products to systems, is primordial (highly important) (Plate 1.1).

The calculation of the CF of the agri-food system requires the accounting for GHG emissions that occur in each of the phases of food and feed production, including not only those that take place in the agricultural sector itself, that is, in crop fields and farms, but also during the manufacture of agricultural inputs, or those derived from the distribution, marketing and consumption of food, using an LCA approach.

The LCA adopts a "bottom-up" approach, that is, it records in detail the emissions generated along the product supply chain, using information on production technologies of the different goods and services. To this end, a "product system" has to be defined, which includes both the different phases of the supply chain (i.e. the "life cycle" of the product) and the exchanges that occur with the environment (i.e. GHG emissions); and a "functional unit" for each food (e.g. 1 kg of product). As a result of the application of an LCA, the "emission coefficients" (i.e. the amount of GHG emissions in kg CO2 eq./kg of product) are obtained, which can be applied to both intermediate and final products, whether domestic or imported.

In estimating the CF of the Spanish Agri-food sector, Aguilera et al. (2015) used the following information as a source (Plate 8.3): (i) inventory analysis based on previous work of the Laboratory of History of Agroecosystems (UPO), based on official data (Yearbook of Agricultural Statistics, FAOSTAT, National Emissions Inventory, etc.); (ii) industrial input emission factors based on "embodied energy" (Aguilera et al. 2015); (iii) Mediterranean N2O emission factors–meta-analysis (Cayuela et al. 2017); (iv) C sequestration with HSOC model (Aguilera et al. 2018); and (v) calculation of emissions associated with irrigation (Aguilera et al. 2019).

National inventories of atmospheric emissions, prepared by the signatory countries to verify compliance with the Kyoto Protocol, is the IPCC Tier 1 approach that is based on global emission factors. However, there is growing evidence that the

**Plate 8.3** An example of processes considered, and the main sources used to estimate the carbon footprint of the Spanish agri-food sector. Reproduced courtesy of Aguilera (2015)

factors differ depending on the type of climate, soil conditions and management, so it is necessary to use more specific factors for a more accurate assessment of emissions. The balance of C in the soil is a crucial process in the CF of agricultural products, since as we have seen it can compensate a large part of the emissions, when the soil gains organic matter and behaves as a sink; otherwise, the soil would contribute more to GHG emissions, when the soil loses organic matter. Despite this, it has been shown that, particularly under arid and semi-arid conditions, C in the soil is very sensitive to changes in management regimes, with changes that can range from the loss of half a ton of C per year in unfertilised soils, to the gain of more than one ton of C in soils in which organic amendments are applied.

The emissions derived from the production of agricultural inputs have decreased significantly due to improvements in efficiency in the industry; therefore, it is necessary to use temporal dynamic factors for the evaluation of historical GHG emissions. Quantitative reviews have been published that analyse precisely these factors for conditions comparable to those in European countries (Aguilera et al. 2015).

The sources of GHG emissions include the construction and maintenance of the agricultural infrastructure, the direct and indirect emissions associated with the use of energy, including traction animals, power generation and fuel use, and CH4 emissions from water bodies (reservoirs, rafts, ditches and canals) (Aguilera et al. 2019). Emissions related to the use of energy could be estimated considering the changes in the country's electric mix, in the energy efficiency of electric generation, and in fossil fuel extraction techniques, including associated methane emissions. The GHG emissions associated with water bodies should include CH4, CO2 and N2O.

Given the challenges that we face in reducing GHG emissions at the agri-food system level, it is necessary to advance our knowledge about effective mitigation strategies that are adapted to the soil-climatic conditions in each region, for example, by synthesising the existing relevant information regarding the main agricultural management practices and their impact on the mitigation of GHG emissions, C sequestration, other polluting compounds, as well as potential barriers and opportunities for the implementation of these strategies.

## **8.5 Conclusions**

We conclude that CSA practices, with an emphasis on climate change adaptation and mitigation, can take many different forms. The CSA practices have many roles to play in agricultural sustainability and in reducing GHG emissions, as well as in increasing soil C sequestration. Practices such as the use of nitrification and urease inhibitors, mulching, application of biochar to the soil, fertilisation management and use of intercropping and crop rotations are all options available to landowners to effectively adapt to and mitigate regional to global climate change. The reader is, however, cautioned that the best CSA practice to be applied to a specific system or location depends on many different factors. Region- or site-specific research is often needed prior to their application to determine if any of the CSAs might produce a positive result on climate change adaptation and mitigation.

## **References**


The opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the International Atomic Energy Agency, its Board of Directors, or the countries they represent.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 3.0 IGO license (http://creativecommons.org/licenses/by/3.0/igo/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the International Atomic Energy Agency, provide a link to the Creative Commons license and indicate if changes were made.

Any dispute related to the use of the works of the International Atomic Energy Agency that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the International Atomic Energy Agency's name for any purpose other than for attribution, and the use of the International Atomic Energy Agency's logo, shall be subject to a separate written license agreement between the International Atomic Energy Agency and the user and is not authorized as part of this CC-IGO license. Note that the link provided above includes additional terms and conditions of the license.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Index**

#### **A**

Acclimation, 181 Accuracy, 21, 25, 28, 32, 49, 54, 56, 65–67, 93, 94, 132, 133, 167, 181, 183, 188, 192, 220, 232, 251, 253, 274, 284 Accurel ®, 37, 38, 134 Acetate, 121, 162, 197, 268 Acetylene (C2H2), 67, 111, 115, 120–122, 162, 163, 216, 230, 254 Acetylene Inhibition Technique (AIT), 120, 216, 230, 231 Actual O Incorporation (AOI), 259, 260 Adaptation, 157, 199 A/D Converter (ADC), 115, 116 Aerobic, 5, 47, 99, 227, 228, 236, 254, 311, 316 Aerodynamic method, 145 Agilent ®, 114, 115, 124, 128 Agricultural drains, 83, 87 Agricultural emissions, 304 Agriculture, 1, 2, 5, 7, 32, 177, 178, 215, 301–305, 307, 315, 317 Agroecosystem, 15, 99, 313, 316, 318 Agro-food system, 317 Airborne, 13 Air ducting, 182 Air filtering, 57 Akaike Information Criterion (AIT), 216, 230, 231 Ammonia (NH3), 2, 6–8, 12, 50, 53, 77, 89– 91, 93–96, 98, 131, 134, 135, 148, 167, 192, 217, 220, 238, 257, 263, 269, 270, 272–274, 280, 307–310, 315 Ammonia volatilization, 7, 89–91, 93–96, 98, 263, 309, 310, 315

Ammonium (NH4 +), 5, 7, 42, 44, 89, 98, 99, 122, 133, 227, 228, 233, 236, 241, 242, 254, 255, 257–265, 268– 274, 276, 279, 280, 287, 288, 291, 307–311, 314 Anaerobic, 4, 5, 99, 110, 114, 198, 199, 206, 217, 226–228, 307, 314, 316 Analysis software, 58 Analytical solution, 263, 264 Animal waste, 178, 180, 201, 202 Anthropogenic emissions, 5, 6, 303 Ap (15N abundance), 218 Aquatic, 12, 77, 231, 262 Argon (Ar), 25, 26, 36, 67, 112, 128, 204, 245 Artificial rumen, 198, 199 Ascorbic acid, 161 Atmospheric lifespan, 6 Automated GHG flux, 133 Automatic chamber system, 134, 136 Automation, automated, 7, 35, 37–39, 50, 57, 58, 62, 78, 109–111, 113, 114, 118, 119, 131–136, 164, 165, 183, 188, 223, 232, 247, 256, 266

#### **B**

Background, background emission, 25, 31, 44, 52, 53, 59, 60, 98 Bacteria, 154, 156, 161–163, 235, 255, 286, 308, 314 Bacteria community, 156 Batch systems, 199 Bayesian, 264 Beetle, 152, 154–156

© The Editor(s) (if applicable) and The Author(s) 2021 M. Zaman et al. (eds.), *Measuring Emission of Agricultural Greenhouse Gases and Developing Mitigation Options using Nuclear and Related Techniques*, https://doi.org/10.1007/978-3-030-55396-8

Biochar, 265, 283, 304, 305, 311, 312, 315, 317, 319 Biochar application, 302, 309, 311, 312, 315, 317 Biological Nitrification Inhibition (BNI), 310 Biomass burning, 302, 305 Bowen ratio, 141, 145 Braun ®, 120 Bromide, 85–87, 275, 310 Bunsen coefficient, 81, 122, 124

#### **C**

13C, 75, 77, 159–162, 165, 167–169 Calibration, 27, 35, 44–46, 65, 66, 73, 93, 114, 119, 122, 134, 170, 171, 181, 186, 188, 204–206, 244, 250, 251, 253, 276, 278 Calorimeter, 182 Carbon (C), 2, 5, 24, 31, 42, 48, 49, 88, 98, 114, 121, 132, 154, 156, 164, 168, 200, 201, 237, 302, 304–306, 311, 312, 315, 317–319 Carbon dioxide (CO2), 1, 3–6, 11, 16, 22, 24, 34, 36, 39, 45, 47–51, 53, 58, 70, 71, 73–77, 86, 89, 110, 111, 113–117, 122, 124, 126, 127, 129, 131–136, 144, 147, 153–156, 159, 160, 164– 169, 171, 178–181, 185, 192, 193, 198, 199, 202–205, 207, 244, 245, 275, 302–304, 306, 307, 311, 312, 314, 318, 319 Carbon Footprint (CF), 317–319 Carbon sequestration; C sequestration, 302, 303, 317–319 Carboxyl-group, 271 Carrier gas, 36, 56, 67, 115, 116, 128, 165–168, 170–172 Cavity-Enhanced Raman Spectroscopy (CERS), 70, 72, 77 Cavity Ring-Down Spectrometer (CRDS), 16, 29, 31, 39, 51–53, 60–68, 134– 136, 164, 169, 204, 245, 247 Chamber dimension, 21, 22 Chamber height, 22, 23, 28, 33, 48, 221 Chamber type, 17, 21, 131 CH4:CO2 ratio, CH4/CO2 ratio, 193 Chemical conversion, 268, 272, 274 Chemical method, 206, 274 Chemiluminescence, Chemiluminescence Detector (CLD), 115–117, 119, 124 Chemstation software, 115

2-chloro-6-(trichloromethyl)-pyridine (Nitrapyrin), 308 Chromatogram, 117, 126, 127 13C label, 161, 162 Clean Development Mechanisms (CDM), 33 Climate chamber, 266, 267 Climate change, 133, 136, 302, 303, 305, 306, 312, 319 Climate change adaptation, 319, 320 Climate-Smart Agriculture (CSA), 301– 303, 305–307, 317, 319, 320 Close cycle, 73 Closed chamber, 11, 14, 15, 18, 21, 32, 33, 40, 46, 53, 91, 131–133, 162, 164, 201, 217, 230 Closed path, 51, 55 Clover (*Trifolium*), 187 Co-denitrification, 99, 217, 220, 261, 288– 290 CO2 enrichment, 136 Combi PAL, CTC analytics, 115, 119 Conceptual model, 226, 227, 262–264, 288, 289 Conservation tillage, 302, 305, 317 Continuous fermenter, 201 Coordinator software, 167, 169 Correction factor, 27, 93, 96 Cost-Benefit Analysis (CBA), 308 Cover crop, 302, 305, 306, 317 Cropping system, 7, 23, 33, 313, 314, 316, 317 Crop production, 7 Crop rotation, 316, 317, 319 Cultivation, 2, 3, 302, 304, 314, 315

#### **D**

Daily Methane Production (DMP), 187, 193–198 Datalogging, datalogger, 38, 42, 68 Data storage, 57 Denitrification, 5, 74–76, 99, 111, 113, 119, 121, 122, 216–218, 220, 221, 225– 228, 230–233, 235–239, 241–243, 253–258, 261, 263, 286–290, 307, 311–313, 315 Denitrifying Enzyme Activity (DEA), 111, 312 Deposition, 2, 5, 77, 309 Detection limit, 113, 118, 132, 194, 205, 224, 229, 272 Devarda's alloy, 273, 274 Dicyandiamide (DCD), 308–310

Index 331

Differential equation, 263, 264 Diffusion method, 272, 274 Difluoromethane (DFM), 162, 163 Dilution, 23, 66, 85, 112, 113, 124, 126, 128, 130, 166, 168, 170, 183, 189, 191, 192, 215, 227, 230, 249, 250, 253, 264, 286 Dilution technique, 159, 262–264 3,4-dimethylpyrazole phosphate (DMPP), 308, 309 Dinitrogen (N2), 5, 12, 25, 26, 36, 67, 70, 71, 73–77, 99, 109–111, 113–118, 120, 124, 126–130, 189–191, 200, 202, 204–206, 213–235, 239, 245, 253, 254, 256, 261, 267, 268, 270, 272, 286, 287, 311–313 Disposable syringe, 120 Dissimilatory Nitrate Reduction to Ammonia (DNRA), 257, 258, 261 Dissolved gas, 16, 78, 88 Dissolved Inorganic N (DIN), 308 Dissolved N2O, 78, 79, 88 Dissolved Organic Carbon (DOC), 226 Dissolved Organic Nitrogen (DON), 133 Disturbance, 20, 25, 29, 93, 226, 233, 267, 302, 304, 306 Diurnal flux variation, diurnal variation, 29, 131, 194 DNDC, 256 Drain, 15, 78, 88, 248 Dry Matter Intake (DMI), 194, 197 Dual isotope method, 233, 257, 258, 261 Duran ®, 120, 121

#### **E**

Earthworm, 152, 154, 156–158, 160, 254 Ecosystem scale, 155, 156 Eddy accumulation, 147 Eddy Covariance (EC), 54, 55, 58, 147, 157, 160, 171 Electron Capture Detector (ECD), 36, 113, 115, 124, 127, 128, 132, 205 Elemental Analyser-Isotope Ratio Mass Spectrometer (EA-IRMS), 273, 274 Emission Factor (EF), 7, 13, 25, 49, 77, 78, 89, 316, 318 Emission Factor (EF5), 77, 78, 88 Endmember, end-member, 235, 236, 238– 243, 253–256 Energy balance, 144–146 Enrichment Ratio Retention (ERR), 258, 259

Enteric Methane Production (EMP), 195– 197 Eppendorf ®, 120, 122 Equilibration, equilibrium, 37, 39, 75, 79– 83, 85, 114, 122, 146, 172 Evacuation, 24–27, 35, 78, 166–168 Evaporation, 258 Excel ®, 46, 224, 283–285, 287, 288 Exetainer, 24, 28, 38, 41, 44, 78–80, 118, 157, 202, 203, 269 Experimental setup, 120, 231, 232, 286 Ex situ incubation, 110

#### **F**

Fermenter, 199 Fertiliser Denitrification (FD), 257–261 Fertiliser type, 96, 313 Fertilizer management, 313–315 Fibre-Enhanced Raman Spectroscopy (FERS), 70–72, 77 Field method, 153, 157, 231 First-order kinetics, 264, 279 Flame Ionization Detector (FID), 36, 113, 124, 126–128, 132, 191, 203 Flow Injection Analysis (FIA), 98 Flow path, 62–65, 134 FLUAZ, 265 Flux gradient, 145, 159 Flux processing, 68 Food and Agriculture Organization (FAO), 50, 304–306 Footprint, 317–319 Fourier-Transform Infrared Spectroscopy (FTIR), 51 Four-pool model, 288 Fp (labelled pool), 218–222, 224, 225, 227– 229, 232, 287 Fractionation model, 239 Fragmentation, 151, 152, 159 Free Air Carbon dioxide Enrichment (FACE), 39, 136 Fungal denitrification, 236–239, 241–243, 253–255, 261 Fungi, 151, 152, 154, 160–163, 261, 286, 287

#### **G**

Gap filling, 49 Gas Chromatograph (GC), 12, 16, 31, 34– 36, 42, 44–46, 50, 53, 54, 56, 61–63, 80, 81, 113–117, 122, 124–126, 128,

129, 131, 132, 163, 164, 171, 187, 191, 203, 205, 230 Gas diffusion, 14, 15, 23, 37, 43, 156, 162, 287 Gas Flux Method (GFM), 214, 216, 217, 225, 226, 228–232 Gas law, ideal gas law, 47, 48, 81 Gas matrix, 73, 245 Gas storage, 25, 27, 28, 203 Gas transfer velocity, 82, 83, 86, 87 Gilson ®, 124 Glass fibre, 268 Glass syringe, 166, 199, 200, 202 Global change, 306 Global Warming Potential (GWP), 5, 6, 303 Grassland, pasture, 2, 5, 15, 20, 21, 23, 24, 33, 98, 131, 133, 155, 163, 164, 187– 189, 193–195, 254, 302, 307–310, 312 Greenhouse Gases (GHGs), 1–8, 11–16, 18, 20–26, 28–35, 37, 40, 41, 44, 49–54, 58, 61, 67, 69, 73, 77, 84, 89, 99, 109, 110, 112, 131–134, 136, 141– 144, 147, 151–153, 155, 157–162, 164, 177–180, 195, 197, 213, 215, 216, 301, 303–307, 311–319 Gross N transformation, 229, 262, 263, 265, 267, 279, 280, 285 Gross rate, 162, 262, 263–266 Groundwater, 75, 77–79, 99, 133, 231 Growth chamber

#### **H**

Hamilton ®, 82, 172 Head box, 183 Headspace equilibration, 79, 80 Heart cutting, 114–116 He incubation, 110 Helium (He), 25, 26, 36, 67, 78–80, 109, 111–115, 118, 120–122, 126, 128– 130, 216, 253, 269, 272, 277, 278 He method Heraeus ®, 121 Herbivory, 152 Herd level, 179, 192 Heterotrophic nitrification, 263, 264, 288 High resolution, 111, 112 Hood system, 184 Hot moment, 136 Hot spot, 136 Hybrid, 217, 220, 223, 228, 232, 286, 287 Hydrocarbon, 20, 27, 36, 161, 199

Hydrogen (H2), 67, 70, 71, 73, 75–77, 110, 111, 114–117, 126, 128, 167, 197, 202, 204 Hydroxylamine (NH2OH), 217, 220, 237, 254, 257, 258, 269, 271, 272, 274, 275, 307

#### **I**

Immobilisation, 226, 262, 263, 279, 280, 311 Indirect emission, 4, 77, 319 Infographics, 7, 50, 158, 318 Infra-red (IR), 50, 51, 70, 134, 146, 181, 186, 193, 204, 205 Inhibitor, 78, 112, 162–164, 199, 201, 267, 306–308, 310 Injection, sample injection, 35, 44, 85, 113, 116, 117, 119, 124, 126, 128, 165– 173, 188, 205, 223, 226, 230, 233, 247, 250, 251, 313 Inorganic nitrogen, 272, 274 *In situ*, 38, 39, 78, 82, 157, 159, 160, 225, 230–232 Integrated Carbon Observation Systems (ICOS), 147, 245 Integrated Horizontal Flux (IHF), 89, 90 Intercropping, 316, 319 Intergovernmental Panel on Climate Change (IPCC), 4–7, 13, 77, 78, 89, 196, 303, 304, 311, 318 International Atomic Energy Agency (IAEA), 7, 8, 50, 276 *In vitro*, 197–199 Isotope effect, 234–240, 242, 253 Isotope mapping, isotope mapping approach, 242, 243 Isotope Mass Balance (IBM), 238, 253 Isotope Ratio Mass Spectrometry (IRMS), 50, 218, 219, 223, 224, 229, 230, 232, 234, 244, 269, 271, 272 Isotopic analysis, 234, 244, 272, 274 Isotopic enrichment, 159 Isotopic fractionation due to N2O reduction, 240 Isotopic fractionation factors associated with N2O reduction to N2 (εN2-N2O), 239–241 Isotopologue, 58, 75, 214, 216–219, 223, 232 Isotopomer, 51, 52, 214, 216, 218, 232, 245, 247, 261 Iso-Versinic, 129 Iso-Versinic Fluorelastomer (IF), 124, 128, 129

#### **K**

Kinetics, 110–113, 119, 130, 198, 264, 281 Krypton (Kr), 85

#### **L**

Labco®, 28, 78, 118 Laboratory approach, 161, 162 Landscape DNDC, 256 Laser Dispersion Spectroscopy (LDS), 52 Laser spectroscopy, 132, 245, 246, 253 Laser technique, 194, 204 Leaching, 77, 78, 133, 226, 228, 305, 307, 308, 315 Leakage, 23, 27, 75, 113, 114, 169, 171, 172, 184 Least Significant Difference (LSD), 285 Li-Cor ®, 39, 64, 65, 134 Life cycle analysis, Life Cycle Assessment (LCA), 317, 318 Lignin, 161 Liming, 312 Limits of Detection (LOD), 117, 224, 225 Linde ®, 120, 121 Linear regression, 39, 45, 172, 194, 256, 276, 278 Livestock, 2, 4, 177, 178, 182, 188, 192, 205, 302, 303, 311 Live Weight (LW), 185, 194, 196, 197 Long-term measurement, 136 Low-cost chamber, 94, 95

#### **M**

Manifold, 24, 37, 38, 165, 167, 168, 247, 249 Mapping approach, 241, 242 Marprene, 124, 128, 129 Mass balance technique, 142, 148 Mass spectrometer, 31, 164, 244, 272, 291 Mass-to-charge (m/z), 218, 219, 223, 224, 234, 244 Matlab ®, 265, 280, 282, 291 *Medicago*, 76 Medical flask, 269 Memory effect, 167, 171 Mesocosm, 157 Metabolic Energy Requirements (MER), 196, 197 Methane (CH4), 1–8, 11, 13, 16, 17, 22, 27, 32, 34, 36, 39, 45, 47–50, 53, 58, 63, 70, 71, 75, 77, 89, 110, 111, 113– 117, 124, 126–129, 131–136, 144, 146–148, 155, 156, 159–169, 171,

177–183, 185–189, 191–199, 201– 207, 215, 245, 302–304, 311, 312, 314–316, 319 Methane Yield (MY), 194 Method, 11–15, 23, 24, 32, 33, 37, 49–51, 54, 58, 65, 70, 73, 78, 81, 83, 89–91, 93, 94, 96–98, 109, 112, 121, 141–148, 152, 157, 159, 165, 178–180, 182– 185, 188–190, 192–195, 198, 201– 204, 206, 213–217, 225, 226, 228– 230, 232, 241–244, 253, 255–258, 261, 262, 268–272, 274, 281, 287, 288, 306 Methyl Chloride (CH3Cl), 85 Methyl group, 161, 162 Methyl methacrylate, 17 Metropolis algorithm, 280 Michaelis–Menten kinetics, 189, 264, 279 Microbial activity, 17, 78, 152, 267, 279 Microbial biomass, 153–157, 161 Microcosm, *see* incubation Micrometeorological technique, 13, 144 Microsite, 37, 99, 161, 227, 228, 314 Mid-Infrared (MIR), 234, 245, 247 Minimum Detectable Flux (MDF), 49, 61– 63 Mitigation, 7, 8, 179, 182, 195, 216, 261, 316, 317, 319, 320 Model, 13, 22, 55, 83, 87, 172, 180, 194, 195, 226–229, 239, 254, 256, 263–266, 269, 270, 279–288, 318 Molecular sieve, molesieve, 114, 116, 124– 128, 204 Mole fraction, 124, 221, 239, 245, 256 Mulch, 311 Multispecies gas analyzer, 53

#### **N**

15N, 262, 263 15N abundance, 217–220, 263, 288, 291 National Ecological Observatory Network (NEON), 147 Natural isotope abundance, natural abundance, 159, 213, 216–218, 233, 234, 263, 274, 276, 286, 288, 291 Naturally ventilated housing, 192 N balance, 76 δ15Nbulk of N2O, 234–236, 238–242, 244, 253, 255, 256 N cycle, 5, 74, 77, 215, 232, 262–264 N deposition, 94 Near Infrared (NIR), 50, 52, 58

Net rate, 262


*Nitrobacter*, 308


N2O isotopomer, 53, 245 N2O mixing endmembers, 243 Non-Dispersive Infrared Spectroscopy (NDIR), 51 15N:18O, 15N/18O, 77, 216 N2O:NO3 <sup>−</sup> ratio, N2O/NO3 − ratio, 235, 236, 242 NO3 - :NO2 - ratio, NO3 - /NO2 - ratio NO2 - :N2O ratio, NO2 - /N2O ratio, nitrite to nitrous oxide ratio, 236 Non-random distribution, 217, 224, 228 N2O:N2 ratio, N2O/N2 ratio, 130, 307, 313 *Ntrace*, 160, 265, 269, 270, 279–287 15N tracing, 214, 215, 231, 233, 254, 256, 257, 262, 264, 265, 279 15N tracing model, 229, 261, 264, 265, 279, 280, 287 Nuclear techniques N transformation, 5, 88, 226, 229, 236, 263, 264, 266, 279, 280, 285, 286, 312 N uptake, 107

#### **O**

17O, 218, 223, 232, 244, 245 18O, 74, 77, 218, 223, 232–234, 239, 245, 258–260, 291 O exchange, 283 Off-Axis Integrated Cavity-Output Spectroscopy (OA-ICOS), 51, 52 O incorporation, 260 18O label, 258 18O labelling, 216, 257 δ18O of N2O, 238, 278 Open chamber, 91, 96, 132 Open-circuit calorimeter, 182 Open cycle Open path, 51, 55 Open-path laser, 178, 179, 193 Organic C (Corg), 25, 99, 231, 304, 307, 313 Organic farming, 315, 316 Organic fertiliser, 253, 302, 303, 305, 308, 315 Organic N (Norg), 2, 5, 99, 220, 262–264, 279, 284, 286–290 Oxygen (O2), 5, 54, 67, 70, 71, 73–77, 88, 99, 111, 113–117, 124, 126–130, 161, 162, 180, 202–205, 216, 218, 223, 230, 232, 234, 237, 238, 242, 245, 257, 258, 276, 277, 308 Ozone (O3), 2, 4, 6

#### **P**

Paddy rice, 32 Parameter optimisation, 264, 280, 281 Pasture, grassland, 2, 5, 15, 20, 21, 23, 24, 33, 98, 131, 133, 155, 163, 164, 187–189, 193–195, 254, 302, 307–310, 312 Pectin, 161 Photoacoustic Spectroscopy (PAS), 51 pH, soil pH, 7, 25, 42, 75, 76, 88, 89, 99, 122, 156, 161, 197, 199, 267, 268, 270, 271, 275–278, 307, 309, 310, 312, 313 Picarro ®, 39, 53, 113, 134–136, 163–165, 167–170, 247–249, 251 *Picea*, 74 *Pinus*, 74 Plant residue, 98, 151, 152, 309 Plant uptake, 228 *Pleurotus*, 162 PLOTQ column, 124–128 Polycarbonate, 17, 34, 188 Polyethylene, 17, 187 Polythene-Coated Urea (PCU), 313 Polytunnel, 187, 188 Polyvinylchloride (PVC), 17, 22, 23, 91, 125, 185 Portable Accumulation Chamber (PAC), 188 Position-specific component, 161 Power Build-Up Cavity (PBC), 72 Precision, 12, 21, 32, 53, 54, 56, 60–63, 65, 66, 93, 94, 96, 113, 168, 180, 190, 193, 220, 223, 224, 230, 239, 247, 250, 251, 253 Propane (C3H8), 85–87 *Pseudomonas*, 75, 76 Pulsed Discharge Helium Ionization Detector (PDHID), 113–117 Purge and Trap System-Isotope Ratio Mass Spectrometry (PT-IRMS), 271, 275, 277, 278

#### **Q**

Quadratic regression, 46 Quality assurance, 251 Quantum Cascade Laser Absorption Spectrometers (QCLAS), 146, 245 Quantum Cascade Lasers (QCL), 52 Quick connectors, 38

#### **R**

Raman spectroscopic gas analysis, 71, 77

Raman spectroscopy, 12, 53, 69, 70, 72, 73, 75, 77, 111 Ratek instruments ®, 121 Real-time, 53, 60, 62, 63, 68, 146, 245, 248, 252 Recalcitrance Recalcitrant N, 98 Recalcitrant SOM, 279 Regression; regression analysis, 39, 40, 44– 47, 172, 194, 256, 276, 278 Remote control, 57 Respiratory Quotient (RQ), 73–75 *Rhizobium*, 76 Rice, *Oryza*, 2, 4, 13, 15, 25, 32–34, 42, 131, 201, 255, 311, 314–316 Risk assessment, 44, 252 River, 12, 77–79, 81, 83–85, 87 Rm, 220, 222 Robotized incubation, 115, 116 Row cropping Rumen, 180, 189, 190, 197–201, 206 Rumen simulation, 198 RUSITEC, 198, 199, 201 Ryegrass (*Lolium*), 187

#### **S**

Sample dilution, 27, 166, 167, 169, 249 Sample injection port, 165, 166, 168, 171, 172 Sample measurement plan, 169 Scarab beetle, 155–160 Screw-capped glass vials, 169, 172 Semi-automation, automated, 118, 120–122 Septum coring, 172, 173 Septum, septa, 21, 22, 24–28, 56, 110, 113, 116, 118–122, 126, 163, 168, 169, 171–173, 223, 247, 250, 251, 269, 271, 272 ShinCarbon, 114–116 Side port needle, 172, 173, 247 Simulink ®, 280, 282, 283 Site Preference (SP), 233, 234, 244 Slurry, soil slurry, 114, 119–122, 124, 178, 180, 192, 201, 220, 231 Small Sample Introduction Module (SSIM), 165–172, 247–252 Software, 56–58, 61, 68, 115, 119, 128–130, 134, 167–169, 193, 198, 248, 249, 251 Soil air sampler, 37, 38, 40 Soil amendment, 7, 89, 265, 311, 314 Soil bacterial community, 156

Soil CH4 balance, 156 Soil denitrifiers, 156 Soil extraction, soil extract, 224, 228, 236, 266–269, 272–274 Soil fauna, 151–160, 254 Soil fauna characterization, 157 Soil fauna clusters, 158, 159 Soil food web, 153 Soil gas sampling, 159 Soil management, 302, 303 Soil N2O fluxes, 156, 157 Soil Organic Carbon (SOC), 302, 304, 317 Soil Organic Matter (SOM) fractions, SOM, 2, 3, 153, 154, 161, 163, 308, 309 Soil-Plant-Atmosphere Continuum (SPAC), 160 Soil profile, 14, 24, 37, 38, 136, 230, 266 Soil respiration, 75, 154 Source partitioning, 253, 287 Spectral interferences, 245 Spectroscopy, 16, 29, 31, 58, 60, 77, 223, 247 SP, δ15N-site preference (δ15NSP) of N2O, 233, 234, 244 Stable carbon isotope, 159, 167 Stable isotope, 52, 54, 74, 77, 99, 159, 160, 164, 213–215, 233, 252, 256, 258, 265 Stable isotope model for nutrient cycles (SIMONE), 256 Stainless steel, 17, 22, 23, 26, 33, 34, 37, 165, 247, 249, 251 Standard, 15, 27, 28, 35, 36, 40, 41, 44–46, 49, 62, 65, 66, 98, 114, 119, 129, 146, 157, 165–168, 181, 182, 184, 190, 204–206, 223, 224, 229, 233, 244, 247–250, 253, 266, 276, 278, 279, 285 Standard Operating Procedure (SOP), 40, 41 Storage, 12, 15, 24, 26, 27, 34, 35, 41, 44, 53, 57, 73, 143, 146, 154, 165, 192, 202, 248, 302, 304, 312, 317 Stream, 16, 36, 68, 83, 85, 87, 88, 113, 119, 145, 160, 165–168, 170–172, 181, 183, 186, 205 Subsoil, 225, 226, 231, 232 Sulphur Hexafluoride (SF6), 54, 75, 85, 148, 159, 178, 179, 187–192, 202, 203, 205 Sustainable production, 302 Syringe, disposable syringe, 22, 24, 25, 30, 32, 37, 38, 41, 43, 44, 56, 78–80, 112, 113, 120–122, 158, 159, 163,

165, 166, 168, 170–173, 198, 200– 202, 205, 247, 249, 250, 252, 269, 275 Syringe needle, 171, 172

#### **T**

Tecan ®, 121 Teflon, 37, 114, 166, 190 Temperature regulation, temperature, 3, 5, 7, 15, 20, 22, 23, 25, 27, 29, 30, 33, 36– 38, 40, 42, 43, 47, 48, 55, 57, 62, 64, 67–69, 78–83, 85–89, 110, 114–116, 122–125, 128, 133, 134, 136, 144– 147, 155, 158, 161, 164, 171, 181– 183, 187, 188, 190, 193, 201, 230, 248, 252, 267, 279, 308, 311, 317 Temperature warming, 136 TERENO, 133 Termite, 2, 4, 152, 154, 155, 158, 159 Terrestrial ecosystem, 2, 99, 132, 147, 229, 304, 305 Terrestrial Ecosystem Research Network (TERN), 147 Thermal Conductivity Detector (TCD), 36, 113–117, 124, 126–128, 203 Three-pool model, 287 Trace gas, 17, 28, 34, 51, 52, 73, 85, 110, 114, 119, 131, 134, 143–146, 163–165, 224, 245, 303, 312 Tracer gas method, 85, 192 Tracer ratio method, 192 Tracer technique, 148, 178, 179, 188 Tracing model, 160, 229, 265 Trajan scientific ®, 120, 171 Transfer velocity, 86, 87 Tubing, 37, 38, 52, 55, 57, 63, 68, 119, 124, 128, 129, 134, 165, 167, 171, 185, 190, 204, 205, 247–249, 251 Tunable Diode Laser Absorption Spectroscopy (TDLAS), 52 Tunable Diode Laser (TDL), 113, 146 Tunable Infrared Laser Differential Absorption Spectroscopy (TILDAS), 52 Tunable Laser Absorption Spectroscopy (TLAS), 51–53 Two-pool model, 263, 287

#### **U**

United Nations Framework Convention on Climate Change (UNFCCC), 5, 33 United Nations (UN), 5, 314

Index 337

United States Geological Survey (USGS), 244, 245, 276, 278 Upwind, 89, 143, 148, 189, 193 Urea, 7, 89, 96–98, 255, 279, 282, 283, 308–310, 312–316 Urea and Ammonium Nitrate (UAN), 310 Urease inhibitor, 307, 309, 310, 315, 319

#### **V**

Vacuum pump, 26, 27, 52, 93, 120, 166, 182, 249, 251, 271 Ventilated hood system, 183, 184 VICI ®, 114, 115, 171 Vienna Standard Mean Ocean Water (VSMOW), 233, 234 Volatile Fatty Acids (VFA), 178, 179, 187, 197–199, 202, 203 Volatilization, 7, 77, 89, 273, 309, 310

#### **W**

Water, 2, 5, 17, 22–25, 31, 33, 37, 40, 42, 43, 47, 48, 57, 64, 66–68, 78–88, 94, 98, 111, 114, 119, 121, 122, 124–126, 134, 144, 145, 158, 161, 164, 171, 181, 183, 184, 187, 190, 198, 200, 202, 226, 228, 231, 233, 237, 238, 242, 244, 252–254, 258, 275, 277,

278, 305, 306, 308, 309, 312, 314, 316, 319 Water Filled Pore Space (WFPS), 267, 309, 310 Water Holding Capacity (WHC), 267 Water management, 314 Water vapour (H2O), 6, 50, 51, 53, 54, 58, 66, 67, 69, 134–136, 144–147, 165, 171, 172, 223, 233, 237, 245, 248, 252, 258–260, 271, 273, 275, 276 Wavelength, 50, 51, 59, 70, 245 Wet concentration value, 170 Workflow for N2O isotopocule analysis, 245, 246

#### **X**

Xa, 221 Xm, 221 Xp, 221

#### **Y**

Yield scaled emissions, 316

#### **Z**

Zero-order kinetics, 263, 279