{"@context":{"@language":"en","Affiliation":"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool","AggregatedSourceRepository":"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider","Campus":"https:\/\/open.library.ubc.ca\/terms#degreeCampus","Creator":"http:\/\/purl.org\/dc\/terms\/creator","DateAvailable":"http:\/\/purl.org\/dc\/terms\/issued","DateIssued":"http:\/\/purl.org\/dc\/terms\/issued","Degree":"http:\/\/vivoweb.org\/ontology\/core#relatedDegree","DegreeGrantor":"https:\/\/open.library.ubc.ca\/terms#degreeGrantor","Description":"http:\/\/purl.org\/dc\/terms\/description","DigitalResourceOriginalRecord":"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO","FullText":"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note","Genre":"http:\/\/www.europeana.eu\/schemas\/edm\/hasType","GraduationDate":"http:\/\/vivoweb.org\/ontology\/core#dateIssued","IsShownAt":"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt","Language":"http:\/\/purl.org\/dc\/terms\/language","Program":"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline","Provider":"http:\/\/www.europeana.eu\/schemas\/edm\/provider","Publisher":"http:\/\/purl.org\/dc\/terms\/publisher","Rights":"http:\/\/purl.org\/dc\/terms\/rights","RightsURI":"https:\/\/open.library.ubc.ca\/terms#rightsURI","ScholarlyLevel":"https:\/\/open.library.ubc.ca\/terms#scholarLevel","Supervisor":"http:\/\/purl.org\/dc\/terms\/contributor","Title":"http:\/\/purl.org\/dc\/terms\/title","Type":"http:\/\/purl.org\/dc\/terms\/type","URI":"https:\/\/open.library.ubc.ca\/terms#identifierURI","SortDate":"http:\/\/purl.org\/dc\/terms\/date"},"Affiliation":[{"@value":"Science, Irving K. Barber Faculty of (Okanagan)","@language":"en"},{"@value":"Biology, Department of (Okanagan)","@language":"en"}],"AggregatedSourceRepository":[{"@value":"DSpace","@language":"en"}],"Campus":[{"@value":"UBCO","@language":"en"}],"Creator":[{"@value":"Emde, David","@language":"en"}],"DateAvailable":[{"@value":"2022-01-17T21:24:23Z","@language":"en"}],"DateIssued":[{"@value":"2021","@language":"en"}],"Degree":[{"@value":"Master of Science - MSc","@language":"en"}],"DegreeGrantor":[{"@value":"University of British Columbia","@language":"en"}],"Description":[{"@value":"Over the last 200 years, conversion of noncultivated land for agriculture has substantially reduced global soil organic carbon (SOC) stocks in upper soil layers. While many agricultural management practices have been well-studied, the specific effects on SOC of cropping systems that incorporate irrigation are poorly understood. Given the large, and expanding, agricultural landbase under irrigation across the globe, this is a critical knowledge gap for climate change mitigation. I undertook a systematic literature review and subsequent meta-analysis of data from studies that examined changes in SOC through time on irrigated agricultural sites across the globe. I followed this work with a regional assessment of mineral-associated organic carbon (MAOC) deficits in irrigated perennial cropping systems of the Okanagan Valley, British Columbia. In the meta-analysis study, I investigated changes in SOC with the following objectives: i) to examine the impact of irrigated agriculture on SOC storage, ii) and to identify the conditions under which irrigated agriculture is most likely to enhance SOC. In the regional assessment of MAOC deficits, I compared specific surface area (SSA) and fine fraction soil texture in combination with random forest and stepwise multiple regression with Akaike Information Criterion modelling in order to iii) determine the best model approach for estimation of MAOC, iv) determine if there is a MAOC deficit present in Okanagan soils, and v) identify the Okanagan soil types with the greatest potential to store additional MAOC. Overall, the meta-analysis indicated that irrigated agriculture has increased SOC stocks by 5.9% globally; however, changes in SOC varied by climate and soil depth. Models using random forest with either SSA or fine fraction soil texture produced the most accurate estimates of current MAOC concentration. By employing a 90th quantile method, we further determined that woody cropping systems across the Okanagan Valley have the potential to store an additional 28.5 million kg of MAOC in the top 30 cm of the soil profile. This work sheds light on the nuances of SOC change across irrigated agricultural systems, outlines the viability of MAOC deficit modelling, and will help guide future research on the impacts of irrigated agriculture on SOC.","@language":"en"}],"DigitalResourceOriginalRecord":[{"@value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/80664?expand=metadata","@language":"en"}],"FullText":[{"@value":"  UNDERSTANDING THE IMPACT OF IRRIGATED AGRICULTURAL SYSTEMS ON SOIL ORGANIC CARBON STORAGE  by David Emde  B.Sc., The University of British Columbia, 2019   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE COLLEGE OF GRADUATE STUDIES (Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan) December 2021    \u00a9 David Emde, 2021ii  The following individuals certify that they have read, and recommend to the College of Graduate Studies for acceptance, a thesis entitled: Understanding the impact of irrigated agricultural systems on soil organic carbon storage Submitted by David Emde in partial fulfilment of the requirements of the degree of Master of Science in Biology.   Dr. Melanie Jones, Faculty of Science, UBC Okanagan Co-Supervisor  Dr. Kirsten Hannam, Faculty of Science, UBC Okanagan Co-Supervisor  Dr. Lael Parrott, Faculty of Science, UBC Okanagan Supervisory Committee Member  Dr. Denise Neilsen, Faculty of Science, UBC Okanagan Supervisory Committee Member    iii  Abstract Over the last 200 years, conversion of noncultivated land for agriculture has substantially reduced global soil organic carbon (SOC) stocks in upper soil layers. While many agricultural management practices have been well-studied, the specific effects on SOC of cropping systems that incorporate irrigation are poorly understood. Given the large, and expanding, agricultural landbase under irrigation across the globe, this is a critical knowledge gap for climate change mitigation. I undertook a systematic literature review and subsequent meta-analysis of data from studies that examined changes in SOC through time on irrigated agricultural sites across the globe. I followed this work with a regional assessment of mineral-associated organic carbon (MAOC) deficits in irrigated perennial cropping systems of the Okanagan Valley, British Columbia. In the meta-analysis study, I investigated changes in SOC with the following objectives: i) to examine the impact of irrigated agriculture on SOC storage, ii) and to identify the conditions under which irrigated agriculture is most likely to enhance SOC. In the regional assessment of MAOC deficits, I compared specific surface area (SSA) and fine fraction soil texture in combination with random forest and stepwise multiple regression with Akaike Information Criterion modelling in order to iii) determine the best model approach for estimation of MAOC, iv) determine if there is a MAOC deficit present in Okanagan soils, and v) identify the Okanagan soil types with the greatest potential to store additional MAOC. Overall, the meta-analysis indicated that irrigated agriculture has increased SOC stocks by 5.9% globally; however, changes in SOC varied by climate and soil depth. Models using random forest with either SSA or fine fraction soil texture produced the most accurate estimates of current MAOC concentration. By employing a 90th quantile method, we further determined that woody cropping systems across the Okanagan Valley have the potential to store an additional 28.5 million kg of MAOC in the top 30 cm of the soil profile. This work sheds light on the nuances of SOC change across irrigated iv  agricultural systems, outlines the viability of MAOC deficit modelling, and will help guide future research on the impacts of irrigated agriculture on SOC.  v  Lay Summary Irrigated agriculture is expanding across the globe, but the effects of irrigation on soil organic carbon (SOC) stocks are poorly understood. We combined data from studies all over the world to examine changes in SOC at irrigated agricultural sites. In contrast to many agricultural systems, which tend to lose soil carbon with time, we found that SOC increased by 5.9% on irrigated sites, overall. SOC increases were greatest in surface soils, in finer-textured soils, in drier climates, and under sprinkler irrigation. However, not all soil carbon remains in the soil for long periods of time. We then took a deep dive into determining how much of the long-lasting forms of carbon are stored in Okanagan Valley orchards and vineyards using carbon modelling. We determined that Okanagan Valley orchard and vineyard soils have the potential to store an additional 28.5 million kg of this carbon in the upper 30cm.  vi  Preface Chapter 2: The contents of this chapter have been published: Emde, D., Hannam, K.D., Most, I., Nelson, L.M., and Jones, M.D. (2021), Soil organic carbon in irrigated agricultural systems: A meta-analysis. Glob Change Biol, 27: 3898-3910. https:\/\/doi.org\/10.1111\/gcb.15680. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use in non-commercial and no modification or adaptations are made. Permission to modify the manuscript for use in this thesis was obtained prior on November 25, 2021. The systematic review protocol modified for use in this chapter and the related publication was first outlined by Kirsten Hannam and Ilka Most. All figures and text were produced and written respectively by me and edited with contributions by my co-authors. Data collection for this meta-analysis was initiated by Kirsten Hannam, and taken over shortly thereafter by me. I collated the data, contacted authors for additional data, developed all gap-filling and data standardization methodology, ran all the statistical analyses, and wrote the first draft of the manuscript.  Chapter 3: The contents of this chapter have been submitted for review to Frontiers in Soil Science on November 9, 2021 as \u201cEstimating mineral associated organic carbon deficits in soils of the Okanagan Valley: A regional study with broader implications\u201d, by Emde, D., Hannam, K.D., Midwood, A.J., and Jones, M.D. All figures and text were produced and written by me and edited with contributions by my co-authors Kirsten Hannam, Andrew Midwood, and Melanie Jones. The data used in this study were generated by Andrew Midwood and his team at the University of British Columbia, after which I contacted authors of the papers that formed the basis for my initial modelling for vii  code and clarification. I carried out further data collation and calculations to generate the dataset used for analysis and producing models. Andy Midwood and Kirsten Hannam identified the initial specific surface area calculation via % moisture approach, which I adapted for use in this chapter. The overall research question and approach was conceived of, and developed by, Kirsten Hannam, Andrew Midwood, and me.   ii  Table of Contents Abstract ......................................................................................................................... iii Lay Summary ................................................................................................................. v Preface .......................................................................................................................... vi Table of Contents .......................................................................................................... ii List of Tables ............................................................................................................... vii List of Figures ............................................................................................................. viii Acknowledgements ....................................................................................................... x Dedication ..................................................................................................................... xi Chapter 1: Introduction ................................................................................................. 1 1.1 Soil organic carbon dynamics ............................................................................ 2 1.1.1 Types of soil organic carbon ....................................................................... 3 1.1.1.1       Particulate organic carbon fraction ...................................................... 3 1.1.1.2 Mineral-associated organic carbon fraction ......................................... 4 1.1.2 Carbon sequestration ................................................................................. 4 1.2 Effects of agricultural management practices on soil organic carbon ................. 6 1.2.1 Tillage practices ......................................................................................... 6 1.2.2 Crop rotation or multiple crops .................................................................... 7 1.2.3 Soil amendments ........................................................................................ 7 1.2.3.1       Nitrogen fertilizer ................................................................................. 8 1.2.3.2 Organic amendments .......................................................................... 8 1.2.3.3 Biochar ................................................................................................ 9 iii  1.2.4 Cover crops .............................................................................................. 10 1.2.5 Irrigation ................................................................................................... 10 1.3 Potential for additional SOC sequestration ...................................................... 12 1.3.1 Estimating mineral-associated organic carbon deficit ............................... 13 1.3.2 Expectations for mineral-associated organic carbon deficits ..................... 14 1.4 Research goals ............................................................................................... 15 References ................................................................................................................ 17 Chapter 2: Soil organic carbon in irrigated agricultural systems: A meta-analysis ...................................................................................................................................... 28 Abstract...................................................................................................................... 28 2.1 Introduction ..................................................................................................... 29 2.2 Methods .......................................................................................................... 33 2.2.1 Study selection ......................................................................................... 33 2.2.2 Data Collection ......................................................................................... 35 2.2.3 Publication bias ........................................................................................ 39 2.2.4 Meta-analysis ........................................................................................... 39 2.3 Results ............................................................................................................ 42 2.3.1 Overall changes by depth increment ........................................................ 42 2.3.2 Effects of climate, texture, and irrigation method ...................................... 42 2.3.2.1 Effects of aridity ................................................................................. 43 2.3.2.2 Effects of soil texture ......................................................................... 45 2.3.2.3 Effects of irrigation method ................................................................ 45 iv  2.3.3 Effects of study duration ........................................................................... 46 2.3.4 Effects of initial SOC stock ....................................................................... 48 2.4 Discussion ....................................................................................................... 49 2.4.1 Climate ..................................................................................................... 49 2.4.2 Soil texture ............................................................................................... 52 2.4.3 Irrigation method ...................................................................................... 53 2.4.4 Limitations and potential for future study direction .................................... 55 References ................................................................................................................ 60 Supplementary Materials ........................................................................................... 71 Chapter 3: Estimating mineral associated organic carbon deficits in soils of the Okanagan Valley: A regional study with broader implications .............................. 101 Abstract.................................................................................................................... 101 3.1 Introduction ................................................................................................... 104 3.2 Methods ........................................................................................................ 109 3.2.1 Data acquisition and variable selection ................................................... 110 3.2.1.1 Okanagan soils database ................................................................ 110 3.2.1.2 Methods of estimating specific surface area .................................... 113 3.2.1.3 Variable selection ............................................................................ 115 3.2.2 Model selection for MAOC estimation ..................................................... 116 3.2.2.1 Stepwise multiple regression with AIC ............................................. 117 3.2.2.2 Random Forest modelling ............................................................... 117 3.2.2.3 Model performance ......................................................................... 118 v  3.2.3 Mineral-associated organic carbon deficit estimation .............................. 118 3.2.4 Calculating MAOC stocks and stock deficits ........................................... 120 3.3 Results .......................................................................................................... 121 3.3.1 Carbon fractions and soil properties ....................................................... 121 3.3.1.1 Correlations between measured MAOC and other soil variables ..... 121 3.3.1.2 Correlations between measured POC and other soil variables ........ 121 3.3.1.3 Correlations between measured SOC and other soil variables ........ 122 3.3.2 Predicting current MAOC: model comparison ......................................... 122 3.3.3 Mineral-associated organic carbon formation capacity ........................... 124 3.3.4 Deficits in MAOC: concentrations ........................................................... 125 3.3.5 Deficits in MAOC: stocks ........................................................................ 127 3.4 Discussion ..................................................................................................... 128 3.4.1 Variable selection ................................................................................... 128 3.4.2 Prediction of current MAOC using stepwise and random forest models .. 131 3.4.3 Mineral-associated organic carbon formation capacity ........................... 132 3.4.4 Mineral-associated organic carbon deficits in Okanagan Valley soils ..... 133 3.4.5 Limitations of 90th quantile deficit modelling and specific surface area estimations ........................................................................................................... 135 3.5 Conclusion .................................................................................................... 136 Acknowledgements .................................................................................................. 136 References .............................................................................................................. 138 Supplementary Materials ......................................................................................... 146 vi  Chapter 4: Conclusion .............................................................................................. 152 4.1 Soil organic carbon in irrigated agricultural systems ...................................... 153 4.2 Estimating MAOC deficits .............................................................................. 154 4.3 Strengths, limitations, and future study directions .......................................... 155      vii  List of Tables Chapter 2 Table 1: Summary of variables collected.  ............................................................ 37 Chapter 2: Supplementary Materials S3: Crop types included in each crop category .................................................... 78 S6: Variable significance ...................................................................................... 81 S7: Agronomic history .......................................................................................... 84 Chapter 3 Table 1: Model performance metrics.  ................................................................ 123 Table 2: Pedotransfer functions.  ....................................................................... 124 Table 3: Difference in whole profile MAOC values between soil groups.  ........... 125 Chapter 3: Supplementary Materials Table S1: Mean MAOC concentration by cropping system and depth  ............... 149 Table S2: Mineral-associated organic carbon deficit by sol type and depth ........ 151    viii  List of Figures Chapter 2 Figure 1: Count of publications included in meta-analysis.  .................................. 34 Figure 2: Spatial distribution of study sites.  ......................................................... 35 Figure 3: Change in SOC by aridity, soil texture, and irrigation method.  .............. 44 Figure 4: Change in SOC by study duration. ........................................................ 47 Figure 5: Change in SOC stock by initial SOC stock. ........................................... 48 Chapter 2: Supplementary Materials S1: Systematic review protocol ............................................................................ 71 Figure S2: Bulk density model variable importance plot ....................................... 77 Figure S4: Publication bias funnel plots ............................................................... 79 S5: Fail-safe N output .......................................................................................... 80 Figure S8: SOC stocks by study duration and depth ............................................ 88 Figure S9: SOC stocks by study duration, depth, and climate .............................. 89 Figure S10: SOC stocks by study duration, depth, and soil texture ...................... 90 Figure S11: SOC stocks by study duration, depth, and irrigation method ............. 91 Figure S12: Rate of SOC change by climate category ......................................... 92 Figure S13: Rate of SOC change by soil texture .................................................. 93 Figure S14: Rate of SOC change by irrigation method ......................................... 94 S15: Studies included in meta-analysis ................................................................ 95 Chapter 3 Figure 1: Outline of MAOC deficit determination process. .................................. 110 Figure 2: Soil texture composition of study samples. .......................................... 111 ix  Figure 3: Comparison of specific surface area and soil texture variables. .......... 113 Figure 4: Correlation matrix of measured soil characteristics. ............................ 116 Figure 5: Predicted MAOC values vs. measured MAOC values. ........................ 119 Figure 6: MAOC concentration and MAOC formation capacity by soil type. ....... 126 Figure 7: Mineral associated organic carbon deficits by soil type. ...................... 127 Chapter 3: Supplementary Materials Figure S1: Full correlation matrix ....................................................................... 146 Figure S2: MAOC model random forest variable importance plot ....................... 147 Figure S3: Comparison of SSA pedotransfer functions and BET analysis .......... 148   x  Acknowledgements This thesis and the resulting publications form a part of the Agricultural Greenhouse Gases Program of Agriculture and Agri-food Canada (Project AGGP2-25). I would like to thank my co-supervisors Melanie Jones and Kirsten Hannam for their irreplaceable guidance throughout all stages of my research. Additionally, I would like to thank Andrew Midwood for the guidance and co-authorship opportunities that arose from collaborating with him throughout my research, and in particular the development of the dataset used in Chapter 3. Thank you also to all of the research technicians and research assistants whose hard work was responsible for the data used herein. Finally, thank you to my wife, Janine Wood, for her endless support and encouragement.    xi  Dedication This thesis work is dedicated to my wife, Janine Wood, without whom this research would never have been possible. Her support, encouragement, and kind patience while I prattle on about soil carbon and my growing fascination with soil science has been a beacon throughout my life as a graduate student.  This thesis is also dedicated to my parents, Karl and Lynn Emde, who always encouraged me to pursue my interest, wherever they led, and helped me with all things great and small throughout my life. Finally, this work is also dedicated to the pillars of scientific excellence, Melanie Jones, Kirsten Hannam, and Andrew Midwood, whose expertise and patience have grown in me a passion for all things soils.1  Chapter 1: Introduction Soil organic matter (SOM) is a continuum of decomposing plant and animal residues; soil microorganisms, including bacteria, fungi, and their exudates; and any other form of organic matter present in the soil (Paustian et al., 1997; Stockmann et al., 2013). As the foundation to healthy, productive soil, SOM is especially important to agriculture, where management practices shift SOM dynamics, increasing or decreasing the amount of SOM stored or released from the soil. Although SOM only accounts for between 1 and 6 % of soil mass in mineral soils, it plays an integral role in an array of essential soil functions (Weil & Brady Emeritus, 2016; Wiesmeier et al., 2019).  Soil organic matter provides a broad range of complex and interacting edaphic benefits including improvement of water holding capacity, soil aggregation, erosion prevention, and nutrient cycling (Hudson, 1994; B. Minasny & McBratney, 2018). Further serving as a reservoir for the nutrients responsible for soil fertility, SOM provides a large number of ion exchange sites capable of binding often limiting nutrients (Beare et al., 1994). Additionally, the increase in beneficial physicochemical characteristics associated with increased SOM directly and indirectly impact the microbial activity in soil, often increasing microbial diversity and breakdown of organic matter (Wiesmeier et al., 2019). This breakdown of organic matter is the mechanism by which soil organic carbon (SOC) turnover occurs and contributes up to 80% of the carbon in the stable fraction of organic matter (Gleixner et al., 2002; Liang & Balser, 2011). While the definition of \u2018stability\u2019 is evolving as we gain a better understanding of the mechanisms involved, there appear to be a suite of mechanisms involved in SOC stability, including physical protection of carbon in soil aggregates, organo-mineral complexing, and cation bridging (Angst et al., 2021; Rasmussen et al., 2018; Six et al., 2002). Ultimately, all SOC can be decomposed under the right conditions, however, carbon which is resistant to oxidation and 2  mineralization by microbes stays in the soil longer and is instrumental in atmospheric greenhouse gas mitigation efforts (Beare et al., 2014; McNally et al., 2017).   Conversion of land from native vegetation to agricultural crops over the last 200 years has reduced global SOC stocks in the top 2 m of soil by an estimated 133 Pg (Sanderman et al., 2017). Much of the carbon lost from soils due to land conversion is released into the atmosphere as CO2, along with an estimated 460 Pg over the same time period from industrial emissions and fossil fuel combustion (Le Quere et al., 2015). Carbon dioxide accumulates in the atmosphere and is the leading factor causing the recent, anthropogenic, acceleration of global warming (Anderson et al., 2016). Soil carbon sequestration, with a potential sequestration rate of 0.4 to 1.3 Pg C yr-1, has been identified as one of the most promising methods by which we may reduce atmospheric greenhouse gases (McNally et al., 2017) and support net-zero carbon commitments, like Canada\u2019s Strengthened Climate Plan, aimed at exceeding 2030 Paris agreement carbon reduction goals and achieving net-zero emissions by 2050 (Service Canada, 2020).  1.1 Soil organic carbon dynamics Agricultural lands cover approximately 11% of the globe\u2019s land surface at an estimated 1.5 billion hectares (Thomson, 2003). Research has shown that adoption of beneficial agricultural practices has the potential to affect SOM change on a large scale. In order to understand how beneficial agricultural practices affect SOM, it is important to understand the different forms of soil organic carbon and the mechanisms by which carbon is stored and released from agricultural soils.  3  1.1.1 Types of soil organic carbon Carbon in soil can take on many forms and is often described as a continuum beginning with undecomposed organic matter (e.g., plant and animal matter) and ending with soil organic carbon (Melillo et al., 1989). Soil organic carbon itself is divided into carbon fractions that vary in the availability of their constituent carbon for biological decay. Particulate organic carbon (POC) is often an indicator of soil quality and decomposes quickly, whereas mineral associated organic carbon (MAOC) is more stable and can take decades to centuries to decompose because it is not readily available to the microorganisms largely responsible for its breakdown (Ferraz de Almeida et al., 2019).  1.1.1.1 Particulate organic carbon fraction Particulate organic carbon (sometimes referred to as the \u201cactive\u201d carbon pool) is composed of organic matter in the early stages of the SOM continuum, and includes newly senesced plant debris as well as living and recently dead organisms (Hu et al., 2018; Midwood et al., 2021; Six et al., 2002). This fraction is largely unbound in the soil matrix, is actively decomposed, has a relatively quick turnover rate (days to years), and has no upper soil capacity threshold. Additionally, POC is the portion of the soil carbon pool most available to soil organisms for energy and its increase is thereby correlated with increases in nutrient cycling (Cleveland et al., 2007; Gulde et al., 2008). Because of the rapid pace of microbial decomposition and the many beneficial properties of POC, changes in POC content can be a sensitive indicator of management-caused changes in soil health (Gulde et al., 2008; Skjemstad et al., 2006).   4  1.1.1.2 Mineral-associated organic carbon fraction The terms \u2018stable carbon\u2019 and \u2018fine fraction carbon\u2019 are often used interchangeably with respect to SOC and generally refer to the highly processed carbon fraction that is bound to mineral surfaces in the soil matrix (Six et al., 2002). This MAOC fraction is primarily composed of microbially derived organic matter (metabolic byproducts and microbe necromass, including the proteins found in the cell walls of dead microbes), and plant-derived organic matter (primarily root exudates and microbially processed lignin; (Angst et al., 2021). Bound, as they are, via organo-mineral complexing and cation bridging with soil minerals, these low atomic weight carbon molecules are largely unavailable to the microorganisms responsible for further decomposition of carbon (Angst et al., 2021; Midwood et al., 2021; Rasmussen et al., 2018; Six et al., 2002). Additionally, physical protection in soil aggregates can also inhibit decomposition of SOC (Gulde et al., 2008; Mikutta et al., 2007; Rowley et al., 2021). As such, this slow-cycling carbon pool can take decades to centuries to decay (Romero-Olivares et al., 2017), and is therefore of particular importance to carbon sequestration and greenhouse gas mitigation efforts (Paustian et al., 2000).  1.1.2 Carbon sequestration Carbon sequestration begins with the conversion of atmospheric CO2 to plant tissue via photosynthesis. This plant-bound carbon then enters the soil directly as root exudates, root turnover, or from the decomposition of leaves, fruit, bark, roots, and branches when the plant or plant tissue dies, or by transfer to the hyphae of mycorrhizal fungi. In many agricultural systems, above-ground plant tissue is removed during harvest; therefore root systems are responsible for a majority of the plant-based carbon sequestration in many managed soils (Paustian et al., 2000).  Soil microorganisms thereafter play a primary 5  role in the conversion of plant matter into both POC, and the more recalcitrant MAOC (Kleber, 2010; Poeplau et al., 2018).  Soil microorganisms are responsible for the decay of organic material across the SOC continuum in response to their own energy and nutrient requirements, a process that accounts for up to 80% of the stable carbon fraction (Simpson et al. 2007; Wiesmeier et al. 2019). Because SOC decay is microbially mediated, factors that affect the rate of microbial respiration similarly affect carbon sequestration rates. For this reason, climate can have a great effect on SOC content in soils on a global scale (e.g., both plant productivity and microbial activity are generally higher in warm, moist areas, than in cool or dry ones; (Wiesmeier et al., 2019)). Once plant and animal biomass has entered the soil ecosystem as SOM, the majority of it is considered POC, and is the raw material for in carbon mineralization and plant nutrient uptake ((Das et al., 2017). A smaller amount will transition into the stable form of carbon that is often found in the finer-textured soil fraction, namely MAOC. Enhancement of this smaller, mineral-bound, pool is generally the goal of long term carbon sequestration efforts because of its presumed longer residence time in soils (Budiman Minasny et al., 2017; Wagner et al., 2007). While many studies have established the fine soil fraction (i.e. silt and clay-sized particles) as the most accurate predictor of MAOC in soil, recent studies have shown a more complex interaction of soil pH with aluminum-oxyhydroxides or exchangeable calcium cations as the driving force behind stable carbon contents, rather than the clay or silt content, per se (Rasmussen et al., 2018). In acidic soils, aluminum- and iron-oxides stabilize soil carbon by forming organo-metal complexes and promoting soil aggregation, whereas in alkaline soils, calcium forms cation bridges between fine soil carbon and mineral surfaces (Kaiser & Guggenberger, 2000; Matus et al., 2006; Mikutta et al., 2007; Rowley et al., 2021). 6   1.2 Effects of agricultural management practices on soil organic carbon The conversion of land from non-cultivated to managed agricultural systems can reduce the SOC content in affected soils by as much as 30 - 80% (Poeplau et al., 2018; Wei et al., 2014; Wiesmeier et al., 2019). This loss has been attributed to a range of factors including greater exposure of the soil surface, altered water dynamics, plant residue removal, and loss of aggregation. By contrast, recovery of crop land to a natural state can increase SOC (Post & Kwon, 2000). Most agricultural management practices have well-investigated effects on the SOC contents of converted land.  1.2.1 Tillage practices The effects of tillage practices on SOC have been widely studied:  reduced or no-tillage has generally been shown to increase SOC compared with more mechanically disruptive full tillage methods (Z. Bai et al., 2018; Wander & Bidart, 2000; Wiesmeier et al., 2019). However, recent studies have shown that reduced or no-tillage treatments tend to increase SOC only up to plough depth; in deeper soils, there is no overall increase in SOC with reduced tillage, but rather a redistribution of SOC within the soil profile (Z. Bai et al., 2018; Wiesmeier et al., 2019). Angers and Eriksen-Hamel (2008) even go as far as to suggest recalcitrant forms of SOC (such as MAOC) are increased under full inversion tillage (a tillage practice by which the surface 20 - 30cm is fully inverted in situ) compared with no-tillage.  7  1.2.2 Crop rotation or multiple crops Increasing crop diversity, including through crop rotations, generally results in higher amounts of SOC than monoculture cropping systems under conventional tillage (Blanco-Canqui, 2021). However, while root carbon input, soil microbial diversity, and aggregate stability are all improved with crop diversity, conversion from continuous monoculture to some form of crop rotation or fallow treatment does not consistently increase SOC content (Wiesmeier et al., 2019). In fact, in cases where an agricultural system already employs no-tillage practices, changing the cropping complexity has no significant effect on SOC (West & Post, 2002).  Silvopastoral and agroforestry systems have varying effects on the SOC content depending on the nature of the converted agricultural land. For example, planting trees in pasture systems tends to decrease SOC at least in the short term (Shi et al. 2013; Upson et al. 2016), while conversion to agroforestry from cropland stimulated significant increases in SOC at all measured depths (up to 60 cm; (Shi et al., 2013)). Since fine root production at depth is one of the primary carbon sources for deep, stable carbon stocks, it is likely that stand age plays a large role in the carbon dynamics of silvopastoral and agroforestry systems (Jackson et al., 2000; Norby & Jackson, 2000).  1.2.3 Soil amendments Application of soil amendments, including nitrogen fertilizer, manure, compost, and biochar, are arguably the most direct way to influence the SOC content of the soil (Chenu et al., 2019). However, the individual characteristics of the cropping system and the confounding effects of other management practices on the same parcel of land play 8  an important role in determining the efficacy of soil amendments for increasing SOC (Chenu et al., 2019; Trost et al., 2013).   1.2.3.1 Nitrogen fertilizer Nitrogen fertilizer, despite co-varying with SOC in most soils, does not intrinsically increase SOC. Tautges et al. (2019) found that the addition of nitrogen fertilizer, did not change the SOC content in wheat-fallow systems in long-term studies in California. Similar results were found in both greenhouse (Ren et al., 2014) and intensive maize cropping systems (Brown et al., 2014), which showed that SOM only increased when fertilizer was applied in tandem with manure. Further, this was the case regardless of whether prior nitrogen levels were determined to be insufficient for the crop or in excess (Brown et al., 2014; Ren et al., 2014).   1.2.3.2 Organic amendments Organic amendments typically have a positive effect on SOC; however, while organic amendments are a relatively direct method by which carbon may be added to soil, their application is often inefficient (Aguilera et al., 2013; Chenu et al., 2019; Wiesmeier et al., 2019; Wuest & Gollany, 2013).  For example, Tautges et al. (2019) reported that maize-tomato rotations saw a 12.6% increase in SOC across a 2 m profile when winter cover crops and composted poultry manure were applied. That being said, the degree to which SOC is retained in a given soil varies greatly depending not only on the physico-chemical properties and of the soil, and the management practices of the agricultural plot as discussed herein, but on the type of organic amendment applied (Hu et al., 2018; Zhao et al., 2013). Ghosh et al. (2012) showed that while SOC increased overall in a 45-9  year trial of rice-wheat systems in the indo-Gangetic plains of eastern India, only 6% of the C applied as green manure was retained in the soil, while 26% of the C applied as farm yard manure was retained.  1.2.3.3 Biochar Biochar as a soil amendment is part of a system of climate-smart agricultural management practices, often combined with practices such as conservation tillage and cover crops, shown to enhance the soil properties responsible for SOC sequestration (X. Bai et al., 2019). While the physicochemical effects of biochar on soil are still being evaluated, its beneficial effects on soils are quite promising, particularly in acidic soils (Chintala et al., 2014; Yuan et al., 2011). Perhaps linked to its liming effect on acidic soils, biochar has been shown to increase SOC in agricultural soils by as much as 39% (X. Bai et al., 2019). This contrasts with the relatively limited increases from other established climate smart agricultural practices shown in a meta-analysis study by X. Bai et al. (2019) in which cover cropping resulted in a 6% increase, and conservation tillage caused a 5% increase. As a soil amendment, biochar holds promise for increasing SOC while counteracting the intrinsic acidification of intensely managed agricultural soils (X. Bai et al., 2019). It should be noted, however, that the overall effects of biochar application on C capture at a global scale are less clear; biochar must be collected from another ecosystem and manufactured, a process which mimics rapid OM decomposition and the related efflux of CO2, thereby potentially counteracting SOC gains in the field to which it is applied. 10   1.2.4 Cover crops The use of cover crops has been shown to produce substantially higher SOC content than conventionally-managed crops (Aguilera et al., 2013; Kuo et al., 1997; Paustian et al., 2000; Tribouillois et al., 2018). However, as with the adoption of no- or reduced-tillage, the enhancement of SOC with the use of cover crops may be restricted to the soil surface. For example, in a long-term maize-tomato trial in California, Tautges et al. (2019) showed that while there was a 3.5% increase in SOC at plough depth, there was a 10.8% decrease at the 30 - 200 cm depth increment. Thus, the addition of winter cover crops to a conventionally managed agricultural system actually decreased overall SOC by 7.3% when sampling was conducted to greater soil depths (Tautges et al., 2019). Because soil hydrology is altered by cover crops, decreases of SOC at depth may be partially explained by reduction of water percolation and the consequent changes to soil processes related to SOC occlusion (i.e., soil aggregation) and transport of SOC (Tautges et al., 2019).  1.2.5 Irrigation None of the management practices discussed here occur independently, and the practice most commonly associated with them is irrigation. The use of irrigation in otherwise water-limited ecosystems increases productivity of cropland and, thus, has the potential to increase SOC in agricultural soils. However, the degree to which SOC is increased, and the relative influence of irrigation compared with other confounding agricultural practices on the same site is not well documented (Condron et al., 2014; Wiesmeier et al., 2019).  11  The few studies that have attempted to quantify the effects of irrigation on SOC typically highlight the difficulty in isolating the effects of irrigation on SOC from the effects of other management practices, and instead focus on simple agricultural systems with few other inputs, such as pastures, or compare the effects of different water sources on soil properties, rather than the effects of irrigation per se (Andrews et al., 2016; Condron et al., 2014; H\u00e4ring et al., 2017; Ramirez-Fuentes et al., 2002). That being said, in a heavily-cited review, Trost et al. (2013) estimated that irrigation in arid or semi-arid climate regions increased SOC by 11 % to 35 %. This effect is believed to be heavily influenced by climatic factors and initial SOC levels and to taper out with depth. Further, recent studies have suggested that irrigation actually decreases SOC at depth, negating any irrigation-caused increases in SOC near the soil surface (Condron et al., 2014). In managed pasture systems in New Zealand, irrigation even decreased SOC at all depths (Mudge et al., 2017). Irrigation is known to interact with other management practices that increase plant productivity and alter microbial activity (Trost et al., 2016). For example, adopting reduced- or no-tillage practices is believed to improve water use efficiency and biomass production and generally enhances the positive effects of irrigation on SOC in water-limited environments (Kochsiek et al., 2009; Trost et al., 2016). Conversely, increased soil moisture as a result of irrigation can also increase microbial activity, thereby accelerating decomposition of SOC (Liu et al., 2008; Trost et al., 2016). Whether irrigation-caused increases in plant productivity, and subsequent additions of plant-derived C to the soil, outweigh the irrigation caused enhanced microbial decay of SOC is expected to depend on climate and varies by agricultural system. The effects of irrigation on SOC may also be confounded by alterations in soil hydrological characteristics caused by the physical effects of applying water to the soil 12  surface (Trost et al., 2013). For example, the repeated wetting and drying of some clay-rich soils can increase large aggregate formation, thereby positively influencing SOC storage as well as the water holding capacity and water infiltration characteristics of the soil (Trost et al., 2016). However, regular infiltration of irrigation water may also function to break up larger, less stable, aggregates into smaller, water-stable microaggregates by removing water soluble compounds and translocating clay particles responsible for macro-aggregate formation (Gulde et al., 2008). Additionaly, irrigation may also promote the formation of soil crusts, which could reduce the efficacy of irrigation for improving plant productivity over the long-term (Fattah & Upadhyaya, 1996). Therefore, the effects of irrigation on SOC are likely to differ among irrigated agricultural systems with divergent physico-chemical characteristics, and as such, there is no consensus effect of irrigation on SOC. Where data are available, there is often great variability in the observed trends. As the irrigated land-base expands to keep up with global food requirements, and the agricultural community seeks to adopt more climate-smart management practices, it is important to understand the specific impact that irrigated agricultural systems have on SOC.  1.3 Potential for additional SOC sequestration In many cases, intensive agricultural practices can disrupt the soil aggregates and organo-mineral complexes responsible for storage of carbon in the soil and result in losses of SOC as CO2 via the acceleration of organic matter decay (Z. Bai et al., 2018; Trost et al., 2013). In order for environmental efforts regarding increased sequestration of soil carbon to be successful we must better understand the potential for individual soils to store carbon. Current global estimates of potential carbon storage are based on broadly defined \u201cland use\u201d metrics for a given geographical area and do not account for 13  the differing capacity of soils with different physicochemical characteristics to store carbon (McNally et al., 2017; Smith et al., 2005).  Because of the need to store carbon over the long term, previous efforts to quantify this upper threshold for a given soil type relied on consideration of the fine soil fraction and its capacity to stabilize SOC as MAOC (Beare et al., 2014; McNally et al., 2017). This MAOC stabilization capacity represents the maximum level of low molecular weight carbon that a soil can stabilize in the mineral fraction under the current soil, environmental, and agronomic management conditions. The MAOC stabilization capacity for a given soil type is used alongside current MAOC concentration to determine the MAOC deficit.  1.3.1 Estimating mineral-associated organic carbon deficit The MAOC deficit is defined as the difference between the MAOC stabilization capacity of a given soil and the soil\u2019s current MAOC concentration (Beare et al., 2014; McNally et al., 2017). In order to focus efforts to increase MAOC sequestration where there is the greatest chance of success, it is important to identify the soils that are not at maximum capacity and that may realistically benefit from adoption of management practices that increase carbon in soils. The difficulty arises in determining the theoretical upper limit of carbon stabilization for a given soil. Most previous studies used a calculation similar to those proposed by Hassink (1997) or Six et al. (2002) in which MAOC stabilization capacity was predicted using a least squares regression model of the relationship between the mass proportion of the fine soil particles (clay and fine silt content) and previously measured MAOC. However, it has since been shown that this approach likely underestimates the sequestration potential for soils because it assumes that most soils are already at MAOC stabilization capacity. Further, these techniques cannot account for 14  soils with more complex clays (Feng et al., 2013; McNally et al., 2017). In light of this, Beare et al. (2014) and McNally et al. (2017) adopted a method by which a best fit quantile regression model was used on a set of improved predictors, drawn from a national database of over 1500 soil profiles, to estimate the MAOC stabilization capacity of soils with a range of clay types (Beare et al., 2014; McNally et al., 2017). The 90th quantile of MAOC values in the dataset were interpreted as representing the MAOC stabilization capacity for those soils. This improved model utilized a specific surface area proxy calculation (rather than the simpler, but less accurate, clay and silt concentration used by Six et al. (2002) and others) to estimate the available surface area for adsorption of fine fraction carbon. Regardless of the method of calculation, exploring the potential for increased MAOC storage, and identifying soils most likely to sequester carbon in a more stable form is integral to greenhouse gas mitigation efforts and to meeting Canada\u2019s Strengthened Climate Plan, which is aimed at exceeding its 2030 Paris Climate Accord goals.  1.3.2 Expectations for mineral-associated organic carbon deficits Given the degree to which SOC is lost from soils upon land use changes, it follows that the vast majority of agricultural soils are likely at a SOC deficit, particularly those under intensive management practices. In studies done on French agricultural sites, the soils determined to be at their MAOC stabilization capacity were those naturally high in SOC; that is, grasslands, meadows, and farming systems in which large amounts of manure were applied (Angers et al., 2011; Poeplau et al., 2011).  The same soil physicochemical characteristics that are most commonly associated with higher SOC levels are likewise implicated with increased MAOC levels and smaller SOC and MAOC deficits. Thus, soils with a larger fine soil fraction are 15  considered to be less likely to have a significant SOC deficit, although this is may not be the case in an intensive agricultural setting. Conversely, sandy soils, with a smaller fine fraction are likely to have a higher SOC deficit, although their capacity to sequester C, in general, will be lower. Soils at higher altitudes and cooler climates, such as carbon rich meadows, have the potential for a small SOC deficit due to lower rates of soil C decay (Angers et al., 2011; Wiesmeier et al., 2019).  1.4 Research goals This thesis and the resulting publications form a part of the Agricultural Greenhouse Gases Program of Agriculture and Agri-Food Canada (Project AGGP2-25), developed with the overall aim of identifying orchard and vineyard management practices that can be used by orchardists and viticulturalists to mitigate greenhouse gas emissions by promoting soil carbon storage. In line with these objectives, this thesis is comprised of two main studies that aim to investigate the SOC storage in irrigated agricultural systems and the potential for further SOC storage in irrigated agricultural soils: i) a global meta-analysis outlining patterns of SOC change in irrigated agricultural systems, and ii) an analysis of MAOC deficits in the orchards and vineyards of the Okanagan Valley, BC, Canada. The first study investigated changes in SOC across the globe by climate (aridity), soil texture classification, and the method of irrigation with the following objectives: i) to examine the impact of irrigated agriculture on SOC storage, ii) and to identify the conditions under which irrigated agriculture is most likely to enhance SOC. The second study, involved a regional assessment of MAOC deficits in which I compared specific surface area (SSA) and fine fraction soil texture in combination with random forest and stepwise multiple regression with Akaike Information Criterion modelling in order to iii) 16  determine the best model approach for estimation of stable carbon, iv) determine if there is a MAOC deficit present in Okanagan agricultural soils, and v) identify the Okanagan soil types with the greatest potential to store additional MAOC.   17  References Aguilera, E., Lassaletta, L., Gattinger, A., & Gimeno, B. S. (2013). Managing soil carbon for climate change mitigation and adaptation in Mediterranean cropping systems: A meta-analysis. Agriculture, Ecosystems & Environment, 168, 25\u201336. Anderson, T. R., Hawkins, E., & Jones, P. D. (2016). CO2, the greenhouse effect and global warming: from the pioneering work of Arrhenius and Callendar to today\u2019s Earth System Models. Endeavour, 40(3), 178\u2013187. Andrews, D. M., Robb, T., Elliott, H., & Watson, J. E. (2016). Impact of long-term wastewater irrigation on the physicochemical properties of humid region soils: \u201cThe Living Filter\u201d site case study. Agricultural Water Management, 178, 239\u2013247. Angers, D. A., Arrouays, D., Saby, N. P. A., & Walter, C. (2011). Estimating and mapping the carbon saturation deficit of French agricultural topsoils: Carbon saturation of French soils. Soil Use and Management, 27(4), 448\u2013452. Angers, D. A., & Eriksen-Hamel, N. S. (2008). Full-inversion tillage and organic carbon distribution in soil profiles: A meta-analysis. Soil Science Society of America Journal. Soil Science Society of America, 72, 1370\u20131374. Angst, G., Mueller, K. E., Nierop, K. G. J., & Simpson, M. J. (2021). Plant- or microbial-derived? A review on the molecular composition of stabilized soil organic matter. Soil Biology & Biochemistry, 156, 108189. Bai, X., Huang, Y., Ren, W., Coyne, M., Jacinthe, P.-A., Tao, B., Hui, D., Yang, J., & Matocha, C. (2019). Responses of soil carbon sequestration to climate smart agriculture practices: A meta-analysis. Global Change Biology. https:\/\/doi.org\/10.1111\/gcb.14658 18  Bai, Z., Caspari, T., Gonzalez, M. R., Batjes, N. H., M\u00e4der, P., B\u00fcnemann, E. K., de Goede, R., Brussaard, L., Xu, M., Ferreira, C. S. S., Reintam, E., Fan, H., Miheli\u010d, R., Glavan, M., & T\u00f3th, Z. (2018). Effects of agricultural management practices on soil quality: A review of long-term experiments for Europe and China. Agriculture, Ecosystems & Environment, 265, 1\u20137. Beare, M. H., Hendrix, P. F., Cabrera, M. L., & Coleman, D. C. (1994). Aggregate-protected and unprotected organic matter pools in conventional- and no-tillage soils. Soil Science Society of America Journal, 58(3), 787-795.  Beare, M. H., McNeill, S. J., Curtin, D., Parfitt, R. L., Jones, H. S., Dodd, M. B., & Sharp, J. (2014). Estimating the organic carbon stabilisation capacity and saturation deficit of soils: a New Zealand case study. Biogeochemistry, 120(1), 71\u201387. Blanco-Canqui, H. (2021). No-till technology has limited potential to store carbon: How can we enhance such potential? Agriculture, Ecosystems & Environment, 313, 107352. Brown, K. H., Bach, E. M., Drijber, R. A., Hofmockel, K. S., Jeske, E. S., Sawyer, J. E., & Castellano, M. J. (2014). A long-term nitrogen fertilizer gradient has little effect on soil organic matter in a high-intensity maize production system. Global Change Biology, 20(4), 1339\u20131350. Chenu, C., Angers, D. A., Barre, P., Derrien, D., Arrouays, D., & Balesdent, J. (2019). Increasing organic stocks in agricultural soils: Knowledge gaps and potential innovations. Soil and Tillage Research, 188, 41\u201352. Chintala, R., Mollinedo, J., Schumacher, T. E., Malo, D. D., & Julson, J. L. (2014). Effect of biochar on chemical properties of acidic soil. Archives of Agronomy and Soil Science, 60(3), 393\u2013404. 19  Cleveland, C. C., Nemergut, D. R., Schmidt, S. K., & Townsend, A. R. (2007). Increases in soil respiration following labile carbon additions linked to rapid shifts in soil microbial community composition. Biogeochemistry, 82(3), 229\u2013240. Condron, L. M., Hopkins, D. W., Gregorich, E. G., Black, A., & Wakelin, S. A. (2014). Long-term irrigation effects on soil organic matter under temperate grazed pasture: Pasture production and soil organic matter. European Journal of Soil Science, 65(5), 741\u2013750. Das, D., Dwivedi, B. S., Singh, V. K., Datta, S. P., Meena, M. C., Chakraborty, D., Bandyopadhyay, K. K., Kumar, R., & Mishra, R. P. (2017). Long-term effects of fertilisers and organic sources on soil organic carbon fractions under a rice\u2013wheat system in the Indo-Gangetic Plains of north-west India. Soil Research, 55(3), 296\u2013308. Emde, D., Hannam, K. D., Most, I., Nelson, L. M., & Jones, M. D. (2021). Soil organic carbon in irrigated agricultural systems: A meta-analysis. Global Change Biology. https:\/\/doi.org\/10.1111\/gcb.15680 Fattah, H. A., & Upadhyaya, S. K. (1996). Effect of soil crust and soil compaction on infiltration in a yolo loam soil. Transactions of the ASAE,  39(1), 79\u201384. Feng, W., Plante, A. F., & Six, J. (2013). Improving estimates of maximal organic carbon stabilization by fine soil particles. Biogeochemistry, 112(1), 81\u201393. Ferraz de Almeida, R., Rodrigues Mikhael, J. E., Oliveira Franco, F., Fonseca Santana, L. M., & Wendling, B. (2019). Measuring the labile and recalcitrant pools of carbon and nitrogen in forested and agricultural soils: A study under tropical conditions. Forests, Trees and Livelihoods, 10(7), 544. Ghosh, S., Wilson, B., Ghoshal, S., Senapati, N., & Mandal, B. (2012). Organic 20  amendments influence soil quality and carbon sequestration in the Indo-Gangetic plains of India. Agriculture, Ecosystems & Environment, 156, 134\u2013141. Gleixner, G., Poirier, N., Bol, R., & Balesdent, J. (2002). Molecular dynamics of organic matter in a cultivated soil. Organic Geochemistry, 33(3), 357\u2013366. Gulde, S., Chung, H., Amelung, W., Chang, C., & Six, J. (2008). Soil carbon saturation controls labile and stable carbon pool dynamics. Soil Science Society of America Journal. Soil Science Society of America, 72, 605\u2013612. H\u00e4ring, V., Manka\u2019abusi, D., Akoto-Danso, E. K., Werner, S., Atiah, K., Steiner, C., Lompo, D. J. P., Adiku, S., Buerkert, A., & Marschner, B. (2017). Effects of biochar, waste water irrigation and fertilization on soil properties in West African urban agriculture. Scientific Reports, 7(1), 10738. Hassink, J. (1997). The capacity of soils to preserve organic C and N by their association with clay and silt particles. Plant and Soil, 191, 77\u201387. Hudson, B. D. (1994). Soil organic matter and available water capacity. Journal of Soil and Water Conservation, 49(2), 189\u2013194. Hu, J., Wu, J., & Qu, X. (2018). Decomposition characteristics of organic materials and their effects on labile and recalcitrant organic carbon fractions in a semi-arid soil under plastic mulch and drip irrigation. Journal of Arid Land, 10(1), 115\u2013128. Jackson, R. B., Schenk, H. J., Jobbagy, E. G., Canadell, J., Colello, G. D., Dickinson, R. E., Field, C. B., Friedlingstein, P., Heimann, M., Hibbard, K., Kicklighter, D. W., Kleidon, A., Neilson, R. P., Parton, W. J., Sala, O. E., & Sykes, M. T. (2000). Belowground consequences of vegetation change and their treatment in models. Ecological Applications: A Publication of the Ecological Society of America, 10(2), 470\u2013483. 21  Kaiser, K., & Guggenberger, G. (2000). The role of DOM sorption to mineral surfaces in the preservation of organic matter in soils. Organic Geochemistry, 31(7), 711\u2013725. Kleber, M. (2010). What is recalcitrant soil organic matter? Environmental Chemistry , 7(4), 320\u2013332. Kochsiek, A. E., Knops, J. M. H., Walters, D. T., & Arkebauer, T. J. (2009). Impacts of management on decomposition and the litter-carbon balance in irrigated and rainfed no-till agricultural systems. Agricultural and Forest Meteorology, 149(11), 1983\u20131993. Kuo, S., Sainju, U. M., & Jellum, E. J. (1997). Winter cover crop effects on soil organic carbon and carbohydrate in soil. In Soil Science Society of America Journal, 61(1), 145-152.  Liang, C., & Balser, T. C. (2011). Microbial production of recalcitrant organic matter in global soils: implications for productivity and climate policy [Review of Microbial production of recalcitrant organic matter in global soils: implications for productivity and climate policy]. Nature Reviews. Microbiology, 9(1), 75; author reply 75. Liu, C., Holst, J., Br\u00fcggemann, N., Butterbach-Bahl, K., Yao, Z., Han, S., Han, X., & Zheng, X. (2008). Effects of irrigation on nitrous oxide, methane and carbon dioxide fluxes in an Inner Mongolian steppe. Advances in Atmospheric Sciences, 25(5), 748\u2013756. Matus, F., Amigo, X., & Kristiansen, S. M. (2006). Aluminium stabilization controls organic carbon levels in Chilean volcanic soils. Geoderma, 132(1), 158\u2013168. McNally, S. R., Beare, M. H., Curtin, D., Meenken, E. D., Kelliher, F. M., Calvelo Pereira, R., Shen, Q., & Baldock, J. (2017). Soil carbon sequestration potential of permanent pasture and continuous cropping soils in New Zealand. Global Change 22  Biology, 23(11), 4544\u20134555. Melillo, J. M., Aber, J. D., Linkins, A. E., Ricca, A., Fry, B., & Nadelhoffer, K. J. (1989). Carbon and nitrogen dynamics along the decay continuum: Plant litter to soil organic matter. Plant and Soil, 115(2), 189\u2013198. Midwood, A. J., Hannam, K. D., Forge, T. A., Neilsen, D., Emde, D., & Jones, M. D. (2020). Importance of drive-row vegetation for soil carbon storage in woody perennial crops: A regional study. Geoderma, 377, 114591. Midwood, A. J., Hannam, K. D., Gebretsadikan, T., Emde, D., & Jones, M. D. (2021). Storage of soil carbon as particulate and mineral associated organic matter in irrigated woody perennial crops. Geoderma, 403, 115185. Mikutta, R., Mikutta, C., Kalbitz, K., Scheel, T., Kaiser, K., & Jahn, R. (2007). Biodegradation of forest floor organic matter bound to minerals via different binding mechanisms. Geochimica et Cosmochimica Acta, 71(10), 2569\u20132590. Minasny, B., Malone, B. P., McBratney, A. B., Angers, D. A., Arrouays, D., Chambers, A., Chaplot, V., Chen, Z.-S., Cheng, K., Das, B. S., Field, D. J., Gimona, A., Hedley, C. B., Hong, S. Y., Mandal, B., Marchant, B. P., Martin, M., McConkey, B. G., Mulder, V. L., \u2026 Winowiecki, L. (2017). Soil carbon 4 per mille. Geoderma, 292, 59\u201386. Minasny, B., & McBratney, A. B. (2018). Limited effect of organic matter on soil available water capacity: Limited effect of organic matter on soil water retention. European Journal of Soil Science, 69(1), 39\u201347. Mudge, P. L., Kelliher, F. M., Knight, T. L., O\u2019Connell, D., Fraser, S., & Schipper, L. A. (2017). Irrigating grazed pasture decreases soil carbon and nitrogen stocks. Global Change Biology, 23(2), 945\u2013954. 23  Norby, R. J., & Jackson, R. B. (2000). Root dynamics and global change: seeking an ecosystem perspective: REVIEW Root dynamics and global change. The New Phytologist, 147(1), 3\u201312. Paustian, K., Andr\u00e9n, O., Janzen, H. H., Lal, R., Smith, P., Tian, G., Tiessen, H., Noordwijk, M., & Woomer, P. L. (1997). Agricultural soils as a sink to mitigate CO 2 emissions. Soil Use and Management, 13(s4), 230\u2013244. Paustian, K., Six, J., Elliott, E. T., & Hunt, H. W. (2000). Management options for reducing CO2 emissions from agricultural soils. Biogeochemistry, 48(1), 147\u2013163. Poeplau, C., Don, A., Six, J., Kaiser, M., Benbi, D., Chenu, C., Cotrufo, M. F., Derrien, D., Gioacchini, P., Grand, S., Gregorich, E., Griepentrog, M., Gunina, A., Haddix, M., Kuzyakov, Y., K\u00fchnel, A., Macdonald, L. M., Soong, J., Trigalet, S., \u2026 Nieder, R. (2018). Isolating organic carbon fractions with varying turnover rates in temperate agricultural soils \u2013 A comprehensive method comparison. Soil Biology & Biochemistry, 125, 10\u201326. Poeplau, C., Don, A., Vesterdal, L., Leifeld, J., Van Wesemael, B., Schumacher, J., & Gensior, A. (2011). Temporal dynamics of soil organic carbon after land-use change in the temperate zone - carbon response functions as a model approach. Global Change Biology, 17(7), 2415\u20132427. Post, W. M., & Kwon, K. C. (2000). Soil carbon sequestration and land-use change: processes and potential. Global Change Biology, 6(3), 317\u2013327. Ramirez-Fuentes, E., Lucho-constantino, C., Escamilla-Silva, E., & Dendooven, L. (2002). Characteristics, and carbon and nitrogen dynamics in soil irrigated with wastewater for different lengths of time. Bioresource Technology, 85(2), 179\u2013187. Rasmussen, C., Heckman, K., Wieder, W. R., Keiluweit, M., Lawrence, C. R., Berhe, A. 24  A., Blankinship, J. C., Crow, S. E., Druhan, J. L., Hicks Pries, C. E., Marin-Spiotta, E., Plante, A. F., Sch\u00e4del, C., Schimel, J. P., Sierra, C. A., Thompson, A., & Wagai, R. (2018). Beyond clay: towards an improved set of variables for predicting soil organic matter content. Biogeochemistry, 137(3), 297\u2013306. Ren, T., Wang, J., Chen, Q., Zhang, F., & Lu, S. (2014). The effects of manure and nitrogen fertilizer applications on soil organic carbon and nitrogen in a high-input cropping system. PloS One, 9(5), e97732. Romero-Olivares, A. L., Allison, S. D., & Treseder, K. K. (2017). Decomposition of recalcitrant carbon under experimental warming in boreal forest. PloS One, 12(6), e0179674. Rowley, M. C., Grand, S., Spangenberg, J. E., & Verrecchia, E. P. (2021). Evidence linking calcium to increased organo-mineral association in soils. Biogeochemistry. https:\/\/doi.org\/10.1007\/s10533-021-00779-7 Sanderman, J., Hengl, T., & Fiske, G. J. (2017). Soil carbon debt of 12,000 years of human land use. Proceedings of the National Academy of Sciences of the United States of America, 114(36), 9575\u20139580. Service Canada. (2020, November 19). Net-zero emissions by 2050. https:\/\/www.canada.ca\/en\/services\/environment\/weather\/climatechange\/climate-plan\/net-zero-emissions-2050.html Shi, S., Zhang, W., Zhang, P., Yu, Y., & Ding, F. (2013). A synthesis of change in deep soil organic carbon stores with afforestation of agricultural soils. Forest Ecology and Management, 296, 53\u201363. Six, J., Conant, R. T., Paul, E. A., & Paustian, K. (2002). Stabilization mechanisms of soil organic matter: Implications for C-saturation of soils. Plant and Soil, 241, 155\u2013176. 25  Skjemstad, J. O., Swift, R. S., & McGowan, J. A. (2006). Comparison of the particulate organic carbon and permanganate oxidation methods for estimating labile soil organic carbon. Soil Research, 44(3), 255\u2013263. Smith, P., Andren, O., Karlsson, T., Perala, P., Regina, K., Rounsevell, M., & Wesemael, B. (2005). Carbon sequestration potential in European croplands has been overestimated. Global Change Biology, 11(12), 2153\u20132163. Stockmann, U., Adams, M. A., Crawford, J. W., Field, D. J., Henakaarchchi, N., Jenkins, M., Minasny, B., McBratney, A. B., Courcelles, V. de R. de, Singh, K., Wheeler, I., Abbott, L., Angers, D. A., Baldock, J., Bird, M., Brookes, P. C., Chenu, C., Jastrow, J. D., Lal, R., \u2026 Zimmermann, M. (2013). The knowns, known unknowns and unknowns of sequestration of soil organic carbon. Agriculture, Ecosystems & Environment, 164, 80\u201399. Tautges, N. E., Chiartas, J., Gaudin, A. C. M., O\u2019Geen, A. T., Herrera, I., & M Scow, K. (2019). Deep soil inventories reveal that impacts of cover crops and compost on soil carbon sequestration differ in surface and sub-surface soils. Global Change Biology. https:\/\/doi.org\/10.1111\/gcb.14762 Thomson, K. J. (2003). World agriculture: towards 2015\/2030: an FAO perspective. Land Use Policy, 20(4), 375. Tribouillois, H., Constantin, J., & Justes, E. (2018). Cover crops mitigate direct greenhouse gases balance but reduce drainage under climate change scenarios in temperate climate with dry summers. Global Change Biology, 24(6), 2513\u20132529. Trost, B., Prochnow, A., Drastig, K., Meyer-Aurich, A., Ellmer, F., & Baumecker, M. (2013). Irrigation, soil organic carbon and N2O emissions. A review. Agronomy for Sustainable Development, 33(4), 733\u2013749. 26  Trost, B., Prochnow, A., Meyer-Aurich, A., Drastig, K., Baumecker, M., & Ellmer, F. (2016). Effects of irrigation and nitrogen fertilization on the greenhouse gas emissions of a cropping system on a sandy soil in northeast Germany. European Journal of Agronomy: The Journal of the European Society for Agronomy, 81, 117\u2013128. Wagner, S., Cattle, S. R., & Scholten, T. (2007). Soil-aggregate formation as influenced by clay content and organic-matter amendment. Journal of Plant Nutrition and Soil Science, 170(1), 173\u2013180. Wander, M. M., & Bidart, M. G. (2000). Tillage practice influences on the physical protection, bioavailability and composition of particulate organic matter. Biology and Fertility of Soils, 32(5), 360\u2013367. Weil, R. R., & Brady Emeritus, N. (2016). The Nature and Properties of Soils. Pearson Education. Wei, X., Shao, M., Gale, W., & Li, L. (2014). Global pattern of soil carbon losses due to the conversion of forests to agricultural land. Scientific Reports, 4, 4062. West, T. O., & Post, W. M. (2002). Soil organic carbon sequestration rates by tillage and crop rotation: A global data analysis. Soil Science Society of America Journal. Soil Science Society of America, 66(6), 1930\u20131946. Wiesmeier, M., Urbanski, L., Hobley, E., Lang, B., von L\u00fctzow, M., Marin-Spiotta, E., van Wesemael, B., Rabot, E., Lie\u00df, M., Garcia-Franco, N., Wollschl\u00e4ger, U., Vogel, H.-J., & K\u00f6gel-Knabner, I. (2019). Soil organic carbon storage as a key function of soils - A review of drivers and indicators at various scales. Geoderma, 333, 149\u2013162. Wuest, S. B., & Gollany, H. T. (2013). Soil organic carbon and nitrogen after application 27  of nine organic amendments. Soil Science Society of America Journal. Soil Science Society of America, 77, 237\u2013245. Yuan, J.-H., Xu, R.-K., Wang, N., & Li, J.-Y. (2011). Amendment of acid soils with crop residues and biochars. Pedosphere, 21(3), 302\u2013308. Zhao, X., Hu, K., & Stahr, K. (2013). Simulation of SOC content and storage under different irrigation, fertilization and tillage conditions using EPIC model in the North China Plain. Soil and Tillage Research, 130, 128\u2013135.  28  Chapter 2: Soil organic carbon in irrigated agricultural systems: A meta-analysis  Authors: David Emde1, Kirsten Hannam2, Ilka Most1, Louise Nelson1, and Melanie Jones1 1Biology Department, University British Columbia, Okanagan Campus, Kelowna, BC, Canada 2Agriculture and Agri-Food Canada, Summerland Research and Development Centre, BC, Canada  Corresponding Author: David Emde (dsemde@gmail.com, +1 250 241 0241)  Keywords: Irrigation, soil organic carbon, agriculture, aridity, climate, soil texture, soil depth  Abstract Over the last 200 years, conversion of noncultivated land for agriculture has substantially reduced global soil organic carbon (SOC) stocks in upper soil layers. Nevertheless, practices such as no- or reduced tillage, application of organic soil amendments, and maintenance of continuous cover can increase SOC in agricultural fields. While these management practices have been well-studied, the effects on SOC of cropping systems that incorporate irrigation are poorly understood.  Given the large, and expanding,  29  agricultural landbase under irrigation across the globe, this is a critical knowledge gap for climate change mitigation. We undertook a systematic literature review and subsequent meta-analysis of data from studies that examined changes in SOC on irrigated agricultural sites through time. We investigated changes in SOC by climate (aridity), soil texture, and irrigation method with the following objectives: i) to examine the impact of irrigated agriculture on SOC storage, and ii) to identify the conditions under which irrigated agriculture is most likely to enhance SOC. Overall, irrigated agriculture increased SOC stocks by 5.9%, with little effect of study length (2 \u2013 47 years). However, changes in SOC varied by climate and soil depth, with the greatest increase in SOC observed on irrigated semi-arid sites at the 0 - 10 cm depth (14.8%). Additionally, SOC increased in irrigated fine- and medium-textured soils but not coarse-textured soils. Furthermore, while there was no overall change to SOC in flood\/furrow irrigated sites, SOC tended to increase in sprinkler irrigated sites, and decrease in drip irrigated sites, especially at depths below 10 cm. This work sheds light on the nuances of SOC change across irrigated agricultural systems, highlights the importance of studying SOC storage in deeper soils, and will help guide future research on the impacts of irrigated agriculture on SOC.  2.1 Introduction Irrigated agricultural land amounts to 275 million hectares worldwide and accounts for 40% of global food production (The United Nations World Water Development Report 2014, 2014). As consumption per capita increases and greater demand is put on agricultural resources, irrigated agriculture is projected to increase by an average of 1.65 million hectares per year until 2030 (The United Nations World Water Development Report 2014, 2014). By enhancing agricultural productivity per unit area of land,  30  particularly in arid and semi-arid climates, irrigation may also help mitigate greenhouse gas emissions by reducing the rate at which unmanaged systems under native vegetation are converted to agronomic systems, thereby preventing the massive losses of soil organic carbon (SOC) often associated with land conversion (McGill et al., 2018; Poeplau et al., 2011; Wei et al., 2014; Wiesmeier et al., 2019). Soil organic carbon, and the organic matter in which it is bound, are integral to maintaining key soil functions in agricultural soils, e.g., water and nutrient retention, soil microbial activity, and maintenance of a healthy physicochemical balance (Trost et al. 2013). These functions, in turn, help maintain agricultural productivity, which is crucial for keeping up with the growing demand for food (Wiesmeier et al., 2016). Irrigation has the potential to affect SOC dynamics in agricultural soils not only by changing soil moisture dynamics and crop productivity, but also by translocating soluble material through the soil profile, by transporting material across the soil surface, and by modifying the metabolic behaviour of microbial communities (Trost et al., 2013; Minasny et al., 2017; Tautges et al., 2019; Xu et al., 2016).  There are three broad types of study that can be used to examine changes in SOC storage under irrigated agriculture: i) studies that compare SOC in irrigated and rainfed plots; ii) studies that compare SOC in unmanaged \u2018natural\u2019 and irrigated plots; and iii) studies that compare SOC in irrigated plots at the beginning and end of an experiment. Studies that compare SOC in irrigated and rainfed field plots present the most obvious opportunity for studying the effects of irrigation on SOC, but such experiments can only be conducted in sufficiently humid regions, where crop production is possible (albeit less successful) without supplementary irrigation. Given that irrigation is needed most in arid\/semi-arid regions that do not receive sufficient natural precipitation to support commercial agriculture, comparisons of SOC in irrigated and rainfed plots cannot capture the effects of irrigated agriculture in the regions where  31  irrigation is most widely applied. Further, comparisons of SOC in irrigated and rainfed plot are often confounded by differences in crop type or planting density.  Similarly, studies that compare SOC under \u2018natural\u2019, or \u2018unmanaged\u2019 vegetation and adjacent irrigated fields are also confounded by differences in plant species (i.e., agronomic versus native species), and by the effect of land-use conversion, which often causes dramatic losses in SOC (Don et al., 2011; Guo & Gifford, 2002; Ogle et al., 2005; Rusinamhodzi et al., 2011).  Consequently, we chose to focus our analysis on studies that compared SOC in the same irrigated plots or fields at the beginning and end of an experiment. The requirement for irrigation is highest in climatic regions where low natural rainfall limits crop productivity. In a recent literature review, Trost et al. (2013) calculated that irrigation in regions with a semi-arid climate increased SOC by 11% to 35% relative to that under native vegetation, and suggested that this effect was heavily influenced by initial SOC levels and was strongest in surface soils. However, a recent soil survey of the western Mediterranean basin near the Spanish-Portuguese border revealed that SOC storage tends to be greater in irrigated Fluvisols, Luvisols and Calcisols than in equivalent rainfed soils but similar in irrigated and rainfed Cambisols (Telo da Gama et al., 2019).   The effect of irrigation on SOC accumulation in more mesic climatic regions is even less well defined. In a long-term study at the Kellogg Biological Station in Michigan, USA McGill et al. (2018) found that SOC increased in irrigated corn fields relative to rainfed controls at a rate of 1% yr-1 for 12 years. By contrast, studies in Brandenburg Country, Germany by Ellmer and Baumecher (2002) and Vienna, Austria by Dersch and B\u00f6hm (2001) found no changes in SOC with irrigation of cereal crops in studies ranging from 21 to 65 years. Rotenberg et al. (2005) reported that irrigation of vegetable crops at  32  the University of Wisconsin\u2019s Hancock Agricultural Research Station caused SOC stocks to decline by over 18%. Moreover, irrigation of temperate pastureland in New Zealand caused SOC to decrease at depth, with losses substantial enough to negate any increases in SOC near the soil surface (Condron et al., 2014).  Thus, there is no current consensus on whether the overall effect of irrigation on SOC stocks is positive, neutral, or negative; and how it varies with climate zone. As global efforts to reduce greenhouse gas emissions intensify, it is more important than ever to investigate the potential for agriculture to promote soil carbon sequestration. The degree to which SOC is increased, and the relative influence of irrigation compared with other confounding agricultural practices on the same site is not well documented (Condron et al., 2014; Wiesmeier et al., 2019). The few studies that have attempted to quantify the effects of irrigation on SOC have highlighted the difficulty in isolating the effects of irrigation per se on SOC from the effects of other agricultural management practices; as such, many studies have focussed on simple agricultural systems with few other inputs, such as pastures. Alternatively, many studies have compared the effects of irrigation with freshwater and wastewater on soil properties (Andrews et al., 2016; H\u00e4ring et al., 2017; Ramirez-Fuentes et al., 2002). In order to improve our understanding of the impact on SOC of irrigation with freshwater, we undertook a meta-analysis of the global literature to i) summarise and characterise changes in SOC in irrigated agriculture and ii) identify the conditions under which irrigated agriculture is most likely to enhance SOC.    33  2.2 Methods 2.2.1 Study selection Each of the studies included in our analysis was an agricultural experiment that included irrigation with freshwater as part of plot management, held management practices consistent throughout the trial, and measured SOC at least twice during the study. The use of irrigation was the unifying theme for selecting studies to include in our analysis.  However, a variety of other management practices were also applied in each study, including tillage, crop rotation, manure and fertilizer application, etc.; selected study sites were also planted to a range of crop types and were irrigated using diverse methods. Thus, we determined that it was not possible to isolate the effect of irrigation per se on SOC stock using our approach. Instead, our aim was to characterize trends in SOC under \u2018irrigated agriculture\u2019, which includes a suite of diverse management practices. To that end, peer reviewed research papers were selected from Web of Science and Google Scholar using the following search string: agric* AND irrigat* AND ((\"soil carbon\" OR (soil AND \"inorganic carbon\")) OR \"soil nitrogen\" OR \"soil pH\") NOT (rice OR tropic* OR forest*) OR (agricultur* OR farm*) AND (irrigat*) AND (soil NEAR\/2 (carbon OR nitrogen))   34  Research papers were then systematically assessed to determine their suitability for inclusion in the meta-analysis using a set of predefined criteria (Supplementary Materials S1). Only data from field studies were included (i.e. no greenhouse or laboratory studies). Also excluded were i) data from experiments conducted in rice paddies or forests, ii) papers that provided insufficient detail about study design, iii) papers that included only one growing season, and iv) papers that used irrigation water sources other than fresh water. The final literature search was completed on January 14, 2019, and included studies published from January 1985 to July 2018, with 68.6% published after 2010 (Figure 1). Although our initial intention was to characterise changes in SOC, soil inorganic C (SIC) and total soil N associated with irrigated agriculture (Supplementary Materials S1), we found few relevant SIC and total soil N data; as a consequence, we chose to focus our meta-analysis on SOC only.  Manual screening of papers that passed these filters resulted in 35 eligible studies, covering 42 study  Figure 1: Count of publications included in meta-analysis dataset by year of publication.  35  sites, and including 297 observations; 38.5% of the study sites were located in North America, 20.5% in Asia, 25.6% in Europe, 2.6% each in the Middle East and Africa, and 5.1% each in South America, and Oceania (Figure 2).  2.2.2 Data Collection We compiled soil organic carbon (SOC) data from all papers that met the selection criteria described above. Soil organic C data collected at the time of treatment establishment (i.e., \u2018time 1\u2019) were considered data from the control treatment and SOC data collected at least one year after the study was established (i.e., \u2018time 2\u2019) were considered data from the \u2018experimental\u2019 treatment. If multiple years of post-treatment SOC data were reported, data were taken from the final year of the study only, to avoid pseudoreplication.  Most papers reported SOC data as either C concentration (i.e., SOC kg-1 dry soil) or C stock (Mg SOC ha-1). Given that changes in SOC stock most accurately reflect Figure 2: (a) Spatial distribution of study sites included in the meta-analysis, including 297 observations from 42 study sites. (b) Breakdown of study sites included in meta-analysis dataset, by geographical area.   36  changes in soil carbon storage, SOC concentration data were converted to SOC stocks, when necessary, using the following equation: \ud835\udc46\ud835\udc42\ud835\udc36 \u0be6\u0be7\u0be2\u0bd6\u0bde (\ud835\udc40\ud835\udc54 \ud835\udc36 \u210e\ud835\udc4e\u0b3f\u0b35)  = \ud835\udc46\ud835\udc42\ud835\udc36\u0bd6\u0be2\u0be1\u0bd6.  \u00d7 \ud835\udc35\ud835\udc37 \u00d7 \ud835\udc61 \u00d7 0.1  (1) where \ud835\udc46\ud835\udc42\ud835\udc36\u0bd6\u0be2\u0be1\u0bd6. is soil organic carbon concentration in g kg-1, \ud835\udc35\ud835\udc37 is bulk density in g cm-3, \ud835\udc61 is the thickness of the depth increment (cm), and 0.1 is the conversion factor for Mg ha-1. Despite its importance for determination of soil properties and SOC stock, bulk density (\ud835\udc35\ud835\udc37) was reported in only 43% of the included studies. We used a random forest algorithm to estimate the missing bulk density values for each soil depth category, using all available predictor variables, in a manner analogous to a pedotransfer function (Supplementary Materials S2) (Chen et al., 2018; Akpa et al. 2016; Sequeira et al., 2014). Random forest (RF) works by combining a large number of regression trees, trained using bootstrap aggregation, to build a robust predictive model that is resistant to noise in the data (Breiman 2001). The R code for this RF model is available at https:\/\/github.com\/dsemde\/Emde-et-al.2021-public. When data were presented in figures, rather than tables, values were estimated using WebPlotDigitizer (https:\/\/apps.automeris.io\/wpd\/). Standard deviation (\ud835\udc46\ud835\udc37), and number of replicates (\ud835\udc5b) were also recorded. Only 30% of the papers reported \ud835\udc46\ud835\udc37 or standard error (\ud835\udc46\ud835\udc38). Where \ud835\udc46\ud835\udc38 was reported, the standard deviation was calculated as  \ud835\udc46\ud835\udc37 =  \ud835\udc46\ud835\udc38 \u00d7 \u0da5 \ud835\udc5b (2)  37  where SE is the standard error and n is the number of observations. If no SD or SE was reported, the SD was estimated using the mean coefficient of variation (Jerabkova et al., 2011).  Categorical and continuous meta-data that could be used as possible predictors of irrigation-related changes in SOC were also collected from each study (Table 1). When key meta-data were not provided, they were estimated, where possible. Missing elevation data were filled in using longitude and latitude values reported in the paper and the rasterized ETOPO1 Global Relief Model (Fick and Hijmans, 2017). Average temperature and precipitation data were typically reported for the specific study period; however, there were cases where a standard 30-year average was reported instead. In light of this, longitude and latitude data were again used to extract 30-year averages from global rasters from worldclim.org for all sites, in order to both standardize the existing values and fill in those that were missing. Table 1. Summary of variables collected from included studies. Data in parentheses indicate the percentage of study sites that reported each variable. Site details Agricultural details Irrigation details Sample collection Geographical location (100) Crop type (100) *Years since irrigation (100) Study scale (100) Elevation (59.2) Multiple crops\/rotation (100) Irrigation method (92.1) *Sample depth (100) Avg. annual precip. (76.3) Tillage frequency (68.4) Irrigation water source (98.7) *Soil bulk density (43.0) Avg. annual temp. (39.5) Tillage type (78.9) Irrigation water pH (9.2) *Soil texture (% sand\/clay) (97.4) Climate (80.3) Crop residue removal (77.6) Irrigation water HCO3 (2.6) *Soil pH (52.6) **Aridity index Grazing status (88.2) Irrigation water calcium (3.9) *Number of samples (n) (100)  Use of cover crops (96.0) Irrigation water Organic C (0) *Organic C (incl. SD) (30.0)  Inorganic N application (90.8) Irrigation water N (2.6)   Herbicide use (65.8)    Organic matter application (92.1)    Inorganic C application (90.8)   * Data were collected separately for the control and experimental treatments. ** Aridity Index was collected from a rasterized GeoTIFF using reported longitude and latitude values.  38   Soil depth measurements varied between studies. In order to standardize comparisons among studies, data were placed in soil depth categories based on the most common sample depths across all studies: 0-10 cm, 10-20 cm, 20-30 cm, 30+ cm. Data from studies that sampled soils outside of these ranges were standardized, as described in Angers and Eriksen-Hamel (2008). That is, values were fitted to a specific depth category by first finding the median of the reported depth increment, and then determining the depth category into which the median value fell. Where studies reported more than one SOC measurement in one depth category, a single value was calculated, using a weighted average.  Given that climate is a strong determinant of irrigation requirements and the impact of irrigation on SOC (Trost et al., 2013), we believed it was important to identify climate zones both accurately and consistently across sites. Therefore, a rasterized GeoTIFF containing global aridity index values was used along with climate delineations outlined in Trabucco & Zomer (2018) to determine the aridity index and a more fine-grained aridity category for each study location. Aridity index was calculated as the ratio of precipitation to potential evapotranspiration according to the following equation: \ud835\udc34\ud835\udc5f\ud835\udc56\ud835\udc51\ud835\udc56\ud835\udc61\ud835\udc66 \ud835\udc3c\ud835\udc5b\ud835\udc51\ud835\udc52\ud835\udc65 =  \ud835\udc40\ud835\udc52\ud835\udc4e\ud835\udc5b \ud835\udc34\ud835\udc5b\ud835\udc5b\ud835\udc62\ud835\udc4e\ud835\udc59 \ud835\udc43\ud835\udc5f\ud835\udc52\ud835\udc50\ud835\udc56\ud835\udc5d\ud835\udc56\ud835\udc61\ud835\udc4e\ud835\udc61\ud835\udc56\ud835\udc5c\ud835\udc5b\ud835\udc40\ud835\udc52\ud835\udc4e\ud835\udc5b \ud835\udc34\ud835\udc5b\ud835\udc5b\ud835\udc62\ud835\udc4e\ud835\udc59 \ud835\udc45\ud835\udc52\ud835\udc53\ud835\udc52\ud835\udc5f\ud835\udc52\ud835\udc5b\ud835\udc50\ud835\udc52 \ud835\udc38\ud835\udc63\ud835\udc4e\ud835\udc5d\ud835\udc5c \u2212 \ud835\udc47\ud835\udc5f\ud835\udc4e\ud835\udc5b\ud835\udc60\ud835\udc5d\ud835\udc56\ud835\udc5f\ud835\udc4e\ud835\udc61\ud835\udc56\ud835\udc5c\ud835\udc5b (3) Aridity index categories: arid (0.03 - 0.2), semi-arid (0.2 - 0.5), dry sub-humid (0.5 - 0.65) and humid (> 0.65) (Trabucco & Zomer, 2018). Other categorical variables were designated using similar approaches to those in previous meta-analyses. For example, crop information was converted from the originally reported crop species to one of three categories indicating the dominant crop type for  39  the study period, as outlined by Aguilera et al. (2013). These categories were: cereals, including crop rotations in which cereals were the dominant crop; horticulture, including crop rotations in which vegetables were the dominant crop; and woody perennial crops, including orchards and vineyards (Supplementary Materials S3). Crop types were considered dominant if they accounted for the greatest portion of the crops in rotation, and present in the study plot for the greatest portion of the study period. Soil texture categories were grouped according to Jian et al. (2020) as either coarse-, medium-, or fine-textured, based on USDA soil texture categories. Coarse-textured soils included sand, loamy sand, and sandy loam; medium-textured soils included sandy clay loam, loam, silt loam, and silt; and fine-textured soils included clay, sandy clay, clay loam, silty clay, and silty clay loam. Finally, study duration categories were determined as in Xu et al. (2019): short term (\u22645 years), medium term (6-15 years), and long term (>15 years).  2.2.3 Publication bias Publication bias was analyzed using both funnel plots (including trim\/fill methods; (Halupka & Halupka, 2017; Viechtbauer, 2010)) and Rosenberg\u2019s failsafe N (Rosenberg, 2005). Using these measurements, the likelihood of publication bias was assessed as non-problematic (Supplementary Materials S4 and S5) and is not discussed further.  2.2.4 Meta-analysis In meta-analysis, either a fixed effect or random effect model can be generated. If the data set is sufficiently large and there is very small inter-study heterogeneity, a fixed effect model may be used (Bashir & Conlon, 2018; Field & Gillett, 2010). However, our meta-analysis consisted of studies from around the world, including sites with a broad  40  range of SOC contents and soil characteristics; we therefore adopted a random-effect approach. We used natural logarithmic response ratios to calculate the relative effect sizes of various management practices, environmental factors, and physicochemical characteristics over time in \u2018experimental\u2019 (time 2, as described above) plots relative to \u2018controls\u2019 (time 1, as described above). This metric allowed us to compare the proportional change in SOC across studies in our global dataset (Gurevitch et al., 2018; Hedges et al., 1999). Response ratios were calculated as: \ud835\udc45\ud835\udc45\u0bcd\u0bc2\u0bc6\u0bbe \u0b35 \u0be9\u0be6 \u0bcd\u0bc2\u0bc6\u0bbe \u0b36  =  \u0d6c\ud835\udc46\ud835\udc42\ud835\udc36\u0bcd\u0bc2\u0bc6\u0bbe \u0b36\ud835\udc46\ud835\udc42\ud835\udc36\u0bcd\u0bc2\u0bc6\u0bbe \u0b35\u0d70 (4) where \ud835\udc46\ud835\udc42\ud835\udc36\u0bcd\u0bc2\u0bc6\u0bbe \u0b35is the soil organic carbon stock at \u2018time 1\u2019 and \ud835\udc46\ud835\udc42\ud835\udc36\u0bcd\u0bc2\u0bc6\u0bbe \u0b36is the soil organic carbon stock at \u2018time 2\u2019. In order to interpret effect size more easily, response ratios were further transformed into percent SOC stock change: \ud835\udee5\ud835\udc46\ud835\udc42\ud835\udc36\u0be6\u0be7\u0be2\u0bd6\u0bde  (%)  = ( \ud835\udc52\u0bcb\u0bcb\u0cc5\u0cba\u0cbe\u0cb6 \u0c2d \u0ce1 \u0cc5\u0cba\u0cbe\u0cb6 \u0c2e \u2212 1) \u00d7 100  (5) where \ud835\udc45\ud835\udc45\u0bcd\u0bc2\u0bc6\u0bbe \u0b35 \u0be9 \u0bcd\u0bc2\u0bc6\u0bbe \u0b36 is the log-transformed response ratio used to compare the change in SOC stock between \u2018time 1\u2019 and \u2018time 2\u2019 as described in equation 4. According to this equation, negative \ud835\udee5\ud835\udc46\ud835\udc42\ud835\udc36\u0be6\u0be7\u0be2\u0bd6\u0bde values indicate a loss of SOC stock over time, positive values indicate a gain in SOC stock over time, and values of \u20180\u2019 indicate no change between \u2018time 1\u2019 and at \u2018time 2\u2019. That is to say, these analyses were used to assess the effect of irrigation on total carbon storage. Soil organic carbon change rate was calculated using the values from Equation 5 in combination with each individual study duration as follows:  41  \ud835\udc46\ud835\udc42\ud835\udc36\u0be6\u0be7\u0be2\u0bd6\u0bde  \ud835\udc50\u210e\ud835\udc4e\ud835\udc5b\ud835\udc54\ud835\udc52 \ud835\udc5f\ud835\udc4e\ud835\udc61\ud835\udc52 (% \ud835\udc66\ud835\udc52\ud835\udc4e\ud835\udc5f\u0b3f\u0b35) =\ud835\udee5SOC\u0b71\u0b72\u0b6d\u0b61\u0b69\ud835\udc60\ud835\udc61\ud835\udc62\ud835\udc51\ud835\udc66 \ud835\udc51\ud835\udc62\ud835\udc5f\ud835\udc4e\ud835\udc61\ud835\udc56\ud835\udc5c\ud835\udc5b  (6) where \ud835\udee5SOC\u0b71\u0b72\u0b6d\u0b61\u0b69 and \ud835\udc60\ud835\udc61\ud835\udc62\ud835\udc51\ud835\udc66 \ud835\udc51\ud835\udc62\ud835\udc5f\ud835\udc4e\ud835\udc61\ud835\udc56\ud835\udc5c\ud835\udc5b (in years) differed on a per-study basis. To further examine the importance of study duration on changes in SOC storage in irrigated agriculture, we additionally placed each data point into one of three study duration categories: short (\u2264 5 years between initial (t1) SOC measurement and final (t2) SOC measurement), medium (6 - 15 years) and long (> 15 years) (Supplementary Materials S8). Data analyses were conducted by soil sample depth category as well as for the full soil profile, because previous meta-analyses have shown that sample depth is a strong predictor of management-caused changes in soil carbon storage (Bai et al., 2019; Du et al., 2017). To explore the importance of soil properties, climatic factors, and management practices on changes in SOC stocks and SOC change rate, we used a combination of simple linear regression (for continuous explanatory variables) and ANOVA (for categorical explanatory variables) as appropriate. From these analyses we selected a sub-set of variables that best explained changes in SOC (i.e., aridity class, and irrigation method), or that warranted further exploration due to their well-documented importance in determining SOC storage (i.e., soil texture). The weighted means and SDs of this sub-set of explanatory variables were determined using the R package \u2018Weighted.Desc.Stats\u2019 (Parchami, 2016). These estimators and their 95% confidence intervals are reported in Figure 3. Treatment effects were considered significant if their 95% confidence interval did not cross 0.  All analyses were carried out in R version 3.6.3 and 4.0.0 (R Core Team, 2020). The meta-analyses were conducted using the \u2018metafor\u2019 package (Viechtbauer, 2010).  42  Simple linear regressions were carried out using the linear model function (lm), and ANOVA was carried out using the ANOVA function (aov) with Tukey\u2019s HSD (TukeyHSD) used as an a posteriori test.  2.3 Results 2.3.1 Overall changes by depth increment Overall, irrigated agriculture increased SOC stocks by 5.9% (black datapoint in the Full Profile panel of Figure 3). Analyzed by depth increment, irrigated agriculture increased SOC by 10.9% at the 0 - 10 cm depth, but did not cause significant changes in SOC at the 10 - 20 cm, 20 - 30 cm, or 30+ cm depths.  2.3.2 Effects of climate, texture, and irrigation method Based on our understanding of the drivers of SOC storage, several explanatory variables were initially considered for their possible importance in modifying the strength and direction of changes in SOC stocks caused by irrigated agriculture (Table 1). According to the criteria used to select variables included in the final analysis (Supplementary Materials S6), we found that aridity class, irrigation method, and soil texture were strong predictors of change in SOC in irrigated agricultural systems. The utility of study duration as a predictor of changes in SOC is discussed separately below. Average annual precipitation, average annual temperature, elevation, crop type, and other management practices had no or only minor importance and are not discussed further (Supplementary Materials S6).   43  2.3.2.1 Effects of aridity In general, irrigated agriculture increased SOC stocks in drier climates (Figure 3). In arid regions, for example, mean SOC increased by 5.9% at the 0 - 10 cm depth; 17.2% at the 10 - 20 cm depth; and 14.8% at the 30+ cm depth (Figure 3). Changes in SOC at the 20 - 30 cm depth were not statistically significant. In semi-arid regions, SOC increased by 14.8% at the 0 - 10 cm depth and 4.1% at the 20 - 30 cm depth, with no statistically significant changes at other depths. In dry sub-humid regions, SOC increased by 4.6% at the 0 - 10 cm depth, and showed no statistically significant changes at the 20 - 30 cm and 30+ cm depths (there were no data available for the 10 - 20 cm depth). Irrigated agriculture in humid regions appeared to reduce SOC stocks at the 0 - 10 cm and 10 - 20 cm depths, but this is based on only two data points. 44  Figure 3: Change in soil organic carbon (SOC) stocks in irrigated agricultural systems by depth increment and climatic category, soil texture category, and irrigation method. Depth increments are indicated at the top of each panel. Filled circles with error bars represent the mean change in SOC and their respective 95% CI. Where 95% CI does not overlap 0% (indicated by a vertical red line), SOC has significantly changed over time. Datapoints may contain studies of varying duration, however a majority of study durations were <5 years. The numbers to the right of each datapoint indicate the total number of observations that were used to calculate the mean.   45  2.3.2.2 Effects of soil texture Over the whole profile, irrigated agriculture increased SOC stocks in medium- and fine-textured soils but not in coarse-textured soils (Figure 3). However, results varied by depth (Figure 3). In coarse-textured soils, irrigated agriculture reduced SOC at the 20 - 30 cm depth (-11.5%) but had no significant effect at other soil depths. In medium-textured soils, irrigated agriculture increased SOC at the 0 - 10 cm (16.8%), 10 - 20 cm (5.2%) and 30+ cm (7.0%) depths but had no significant effect at the 20 - 30 cm depth. In fine-textured soils, irrigated agriculture increased SOC at the 0 - 10 cm depth (12.9%) and 20 - 30 cm (4.7%) depths, and reduced SOC at the 10 - 20 cm (-5.4%) and 30+ cm (-8.0%) depths.  2.3.2.3 Effects of irrigation method Overall, SOC stocks increased significantly under sprinkler irrigation (9.5%) and showed no significant change in drip and flood\/furrow irrigated systems. The strength and direction of changes in SOC stocks also varied by irrigation method and soil depth (Figure 3). Drip irrigation increased SOC at the 0 - 10 cm depth (5.5%) but reduced SOC at the 10 - 20 cm (-5.9%), 20 - 30 cm (-14.8%), and 30+ cm (-5.3%) depths. By contrast, sprinkler irrigation increased SOC at the 0 - 10 cm depth (19.4%), reduced SOC at the 10 - 20 cm depth (-3.6%), and caused no change in deeper soils. Flood\/furrow irrigation increased SOC at the 20 - 30 cm depth (8.9%), reduced SOC at the 30+ cm depth (-8.5%), and caused no change in shallower soils.   46  2.3.3 Effects of study duration Although the t1 data used in our study was collected at the beginning of the published experiments, there is always the possibility that previous agricultural practices, including irrigation, carried out on the plots prior to these experiments may have had carry-over effects. However, we investigated the past history of each experimental site to ensure that differences in SOC between the beginning and end of each experiment were not confounded by recent land use change, e.g., conversion from \u2018non-agricultural\u2019 or \u2018unmanaged\u2019 land (Supplementary Materials S7).  Although site management history was not consistently well-documented, most of the selected studies were carried out on well-established agricultural research sites (52.8%); several of the remaining studies were carried out on long-term commercially managed sites or on sites with otherwise less well-documented management histories.  Only two studies (Undersander and Ger 1985; Sainju et al. 2014) appear to have been conducted on land that had recently (2 and 3 years prior, respectively) been converted from a \u2018natural\u2019 system to an agricultural system. Therefore, we are confident that most studies included in this analysis had not recently undergone land-use conversion from \u2018natural\u2019 systems.  If the start of a given experiment had coincided with the initiation of irrigation, however, then the change in SOC we observed might also be confounded by the effects of changes in management practices. However, our assessment of the study site descriptions from each publication suggests that most experimental plots had already been irrigated as part of previous management practices, i.e., prior to \u2018t1\u2019. Therefore, overall, the changes in SOC we observed appear to be caused by the suite of practices associated with irrigated agriculture, rather than by recent changes in management or land use. The treatment effects described above do not account for differences in study duration, i.e., the fact that treatment effects may increase, decrease or change direction  47  with time. In many cases, there were too few studies to draw conclusions about the effect of study duration on treatment effects when separated by study and soil depth categories (Supplementary Materials S9 through S11). Nevertheless, there were sufficient data to examine the effects of short- and medium-term study duration on changes in SOC at the 0-10 cm depth in i) semi-arid climates and ii) under sprinkler irrigation (Figure 4). In semi-arid climates, across irrigation methods and soil texture categories, irrigated agriculture increased SOC storage at the 0-10 cm depth in short term studies, but not in medium term studies; only one study reported changes in SOC over the longer-term. In sprinkler-irrigated systems, across climate types and soil texture categories, irrigated agriculture caused similar increases in SOC storage at the 0-10 cm depth over both the short and medium terms; again, only one study reported changes in SOC over the longer term. Clearly, longer-term monitoring of changes in SOC storage in irrigated systems is needed, although this analysis shows that even a few years of irrigation can cause changes in SOC. We also examined the importance of study duration on changes in SOC storage by calculating the rate of change in SOC (i.e., % change in SOC divided by study Figure 4: Change in soil organic carbon (SOC) stocks by study duration at the 0 - 10 cm depth increment for semi-arid, and sprinkler irrigated sites. Filled circles with error bars represent the mean change in SOC and their respective 95% CI. Where 95% CI does not overlap 0% (indicated by a vertical red line), SOC has significantly changed over time. The numbers to the right of each datapoint indicate the total number of observations used to calculate the mean.  48  duration; Supplementary Materials S12 through S14), by depth increment.  We found few discernible patterns in the data.  At the 20-30 cm depth, however, SOC decreased significantly at a rate of -0.38 %C year-1 under drip irrigation (Supplementary Materials S14).   2.3.4 Effects of initial SOC stock There was a slight negative relationship between initial SOC stock and the % change of SOC in irrigated agriculture at the 0 - 10 cm depth (p = 0.04), suggesting that irrigated agriculture is more likely to increase SOC in soils with lower initial SOC contents than in soils with higher initial SOC contents (Figure 5). Soils in arid climates tended to have lower initial SOC contents, while soils in semi-arid and dry sub-humid climates showed a wide range in initial SOC contents. It should be noted, however, that initial SOC stock Figure 5: Change in soil organic carbon (SOC) stocks by initial SOC stock at the 0 - 10 cm depth increment for arid, semi-arid, dry sub-humid, and humid climate zones (R2 = 0.0202; p value = 0.0423).   49  explained only a very small portion of the variation in change in SOC stock (R2: 0.02). No significant, discernible pattern was evident at greater soil depths (data not shown).  2.4 Discussion This study assessed overall trends in SOC storage in irrigated agricultural systems across the globe by compiling and analysing data from 35 published studies (Supplementary Materials S1). In most cases, the studies used in this analysis (list of included studies can be found in Supplementary Materials S15) did not aim to examine the effects of irrigation on SOC per se, but were included in our meta-analysis because they reported data that could be used to assess the impact that irrigated agriculture has on SOC stocks across our study categories. The use of irrigation was the single unifying theme for selecting these studies to include in our analysis. We found that irrigated agriculture tends to increase SOC stocks (by 5.9% overall), and that the effects are strongest in surface soils. Of the 32 explanatory variables that we considered (Table 1), aridity and irrigation method had the strongest effect on the scale and direction of change in SOC under irrigated agriculture. Average annual precipitation, average annual temperature, elevation, crop type, and other management practices (e.g., tillage), had no or only minimal importance as explanatory variables. We also assessed the role of soil texture in mediating changes in SOC under irrigated agriculture, due to its well-documented importance in controlling SOC storage.  2.4.1 Climate Irrigation in arid and semi-arid regions was associated with larger increases in SOC than irrigation in wetter climates (although there were fewer data points for dry sub-humid and  50  humid regions). Warmer soil temperatures promote evapotranspiration, which drives irrigation water demand upward. Thus, more irrigation water was likely applied in studies conducted in arid\/semi-arid regions than in more humid regions (Dong et al., 2015; Sch\u00fctt et al., 2014). Irrigation in water-limited environments increases plant productivity, which can result in greater carbon inputs to the soil; however, wetting soils may also stimulate microbial activity, which can result in the loss of soil carbon due to increased mineralization of SOM (David et al., 2018; Dong et al., 2015). These two factors (increased plant productivity and accelerated microbial decay of SOM) thus act in opposite directions with respect to the accumulation of SOC. Our results indicate that the effect of irrigation on plant growth outweighs the effect on SOM decay by microbes (carbon mineralization), particularly in surface soils of arid and semi-arid regions.  Assuming that irrigated agriculture improves plant growth (via increased photosynthetic carbon fixation) and, consequently, surface litter and below-ground (root) carbon inputs (Denef et al., 2008; Gillabel et al., 2007), the greatest improvements in SOC can be expected to extend from the soil surface to the maximum rooting depth, with smaller changes below the rooting zone (Trost et al., 2013). Data from semi-arid plots generally support this reasoning, with the greatest improvements in SOC at the 0 - 10 cm depth, smaller increases at the 20 - 30 cm depth, and no significant change at 10 \u2013 20 cm and 30+ cm depths. By contrast, data from arid plots showed the greatest improvements in SOC at the 10 - 20 cm and the 30+ cm depths. Although available data for irrigated arid sites are clearly limited, this pattern might be expected for crops that root more deeply to scavenge for available water lower in the soil profile (Guswa, 2008). This finding is particularly relevant because SOC that accumulates at depth is considered relatively resistant to decomposition (Das et al., 2017; Minasny et al., 2017).  51  The contrasting changes in SOC storage at depth in arid and semi-arid sites could also reflect alterations in soil hydrology caused by the downward percolation of applied water.  Again, such an effect would likely be strongest in the driest agricultural regions because of the requirement for more frequent and\/or greater irrigation. Repeated wetting and drying, such as that which occurs over irrigation cycles, can promote the formation of water-stable and micro-aggregates by altering cohesion and fragmentation processes in the soil; soil aggregation enhances both water holding capacity and water infiltration (Trost et al., 2013). Improved infiltration of water can positively influence SOC storage at depth by translocating soluble carbon downward in the soil profile, where it can be readily sorbed onto unsaturated soil particles and, thus, protected from mineralization (Minasny et al., 2017; Tautges et al., 2019; Xu et al., 2016). Few data have been published on the effects of irrigated agriculture on SOC storage in wetter climates, no doubt owing to the reduced need for irrigation in such regions. Our analysis detected a small, but significant, increase in SOC in irrigated dry sub-humid regions at the 0 - 10 cm depth but not at deeper depths, and a decrease in SOC at all depths in humid regions (although data were only available from one humid site). This is more or less in line with long-term (>10 years) studies conducted in sub-humid, humid, and tropical sites in Ethiopia and Brazil, which found no significant change in the SOC content with irrigation (De Bona et al., 2008; Getaneh et al., 2007). These studies were not included in our meta-analysis because they either did not report required study details (De Bona et al., 2008), or did not meet our criteria regarding consistent management practices (Getaneh et al., 2007), but their findings provide useful insights into the response of SOC to irrigation in wetter climates. In their review of a number of long-term agricultural studies, Trost et al. (2013) reported that increases in SOC due to irrigation depended not only on climate, but also  52  on initial SOC levels: humid and semi-arid sites with higher initial SOC tended to show low or no increase in SOC storage, while arid and semi-arid sites with lower initial SOC tended to show greater increases in SOC storage. While our analysis broadly supports a negative relationship between initial SOC storage and the scale and direction of changes in SOC storage in response to agricultural irrigation, this relationship was nuanced. While the greatest increases in SOC were indeed found in sites with lower initial SOC levels, so too were the greatest losses (Figure 5). The higher variability in changes in SOC storage observed for irrigated sites with initially lower SOC levels may simply be due to the fact that a greater number of studies have been conducted on sites with lower initial SOC stocks.  2.4.2 Soil texture Soils with a larger fine fraction tend to have a greater SOC storage capacity and, therefore, are expected to show greater increases in SOC due to irrigation (Wiesmeier et al., 2019; Zhong et al., 2018). With increases in clay content, irrigation is expected to favour formation of micro-aggregates (Trost et al., 2013; Wagner et al., 2007). As microaggregate formation increases, average pore size is decreased (Hassink et al., 1993). Since pore size determines accessibility of organic matter to microbes, a higher proportion of micro pores has the potential to decrease SOC mineralization (Xu et al., 2016). Overall, there was a trend toward larger gains in SOC in soils with fine or medium textures than in soils with coarse textures, as expected. However, SOC storage increased more consistently in medium-textured soils than in fine-textured soils, where SOC actually declined at the 10 - 20 cm and 30+ cm depths. Differences in the response of SOC by soil depth in fine-textured soils may be associated, at least in part, with the downward translocation of soil particles during the  53  percolation of irrigation water. In a study examining changes in soil properties of historically flood-irrigated fields that have been converted to drip irrigation, Puy et al. (2017) showed that soils directly under drippers had a higher ratio of coarse\/fine particles than adjacent, unirrigated soils in the same field.  This suggests that irrigation has the potential to shift SOC dynamics by translocating clay particles downward, thereby decreasing the proportion of the fine fraction in irrigated surface soils and altering soil hydrological properties at all depths (Warrington et al., 2007). Similarly, Drewry et al. (2020) found that soil bulk density increased, and macroporosity declined, in irrigated pastures and cropland in New Zealand. Further work examining changes in soil texture over time due to irrigation and at depth is necessary to better understand the role of soil texture in mediating irrigated agriculture-related changes in SOC, particularly on arid and semi-arid sites, where irrigation is employed most intensely.  2.4.3 Irrigation method The effect of irrigated agriculture on SOC storage also varied strongly by irrigation method and soil depth. In general, drip irrigation caused an increase in SOC storage in surface soils and a decrease in SOC storage below 10 cm, while sprinkler and flood furrow irrigation showed no consistent pattern with soil depth. Irrigation method likely plays an important role in determining the effects of irrigation on SOC content in agricultural soils by mediating changes in hydrological and physicochemical properties that vary with depth. In flood-irrigated agricultural plots established in the 10th - 13th centuries current era (CE) that were converted to drip irrigation, areas directly under drippers showed increases in SOC and marked textural changes, as discussed in section 4.2, whereas those adjacent to the drip zone lost SOC, likely due to increased SOC oxidation and reduced inputs of fresh plant biomass (Puy et al., 2017). Given that  54  inputs of both water and fertilizer (e.g., via fertigation) are much more localized under drip irrigation, the positive effects of irrigation on SOC are limited to those areas directly under the drippers (Kallenbach et al., 2010; Puy et al., 2017; S\u00e1nchez-Mart\u00edn et al., 2008). Differences in the distribution of water among irrigation methods (i.e., highly localized under drip irrigation versus more uniform under flood\/furrow and sprinkler irrigation) may account for the observed differences in SOC contents among irrigation methods reported here. This raises the possibility that soil sampling strategies influenced our results. For example, the drip irrigation studies included in our meta-analysis largely employed composite sampling, which included randomized sample locations across a plot, without defining whether those samples came from the \u2018wetted bulb\u2019 under the drippers; only one study specified that sampling was conducted directly within the drip zone. In order to properly capture the effects of drip irrigation on SOC from the perspective of atmospheric greenhouse gas mitigation and large-scale carbon stocks, sampling must be designed to represent the entire gradient of soil moisture contents across each plot. Soil organic carbon stocks under drip-irrigated agriculture decreased with depth in our analysis; however, it is difficult to know whether this reflects the sampling issues outlined above. Nevertheless, given the results from Puy et al. (2017), it seems likely that the pattern shown in Figure 3 is a reasonable representation of field-scale effects of drip irrigation on SOC. In addition to the highly localized placement of water caused by drip irrigation, careful control of irrigation volumes to prevent deep percolation of water under drip irrigation restricts drainage of water beyond the rooting zone (Sanchez-Mart\u00edn et al., 2010). Therefore, the observed decrease in SOC at depth is likely due to decreased microbial activity and\/or reduced root inputs outside the \u2018wetted bulb\u2019 beneath drip emitters (Liu et al., 2008; Wiesmeier et al., 2019).  55  Less spatially discrete irrigation methods, such as flood\/furrow and sprinkler irrigation, can increase the availability of soil water to both crop- and non-crop plant species. Indeed, the largest gains in SOC storage were observed in surface soils under sprinkler irrigation, which generally applies water across the entire surface of an agricultural field (i.e., to both crop and non-crop plants). While the growth of weeds and other non-crop plants may be counter to the objectives of conventionally-managed\/precision agriculture, they can also contribute to increased SOC stocks (Moonen & B\u00e0rberi, 2008; Petit et al., 2011). For example, inter-row spaces in orchards and vineyards are increasingly recognized as valuable targets for enhancing SOC stocks in agricultural soils (Midwood et al., 2020; Puy et al., 2017; Trost et al., 2013). Shifting the SOC dynamic for a whole agricultural field via irrigation may have notable benefits for long-term SOC change in agricultural soils, but the inherent inefficiency of applying irrigation water to support the growth of both crop- and non-crop species may not be sustainable as global demand for irrigation water increases (Levidow et al., 2014; Puy et al., 2017; United Nations Educational, Scientific and Cultural Organization, n.d.). As such, historically flood-irrigated agricultural fields are increasingly transitioning to more efficient drip irrigation systems (Puy et al., 2017). While this shift to drip irrigation may increase water use efficiency and decrease infrastructure related CO2 emissions, our results demonstrate that there are potentially far-reaching environmental trade-offs (i.e., reductions in SOC storage in deeper horizons) involved in adopting precision irrigation practices on a global scale.  2.4.4 Limitations and potential for future study direction Our data show that, overall, irrigated agriculture can increase SOC at all depths, but effects vary widely among climate categories, soil textures, and irrigation methods. Our  56  results show that in semi-arid sites, SOC increased across the full soil profile under irrigated agriculture (Figure 3), but it remains unclear whether increases in SOC due to irrigated agriculture are sufficient to reverse the frequently-reported losses of SOC caused by the conversion of natural, unmanaged ecosystems to agricultural systems. Further, increased SOC was associated with fine- and medium-textured soils at the 0 - 10 cm depth but patterns were less clear in deeper soils, suggesting that downward percolation of finer-textured particles may play an important role (Puy et al., 2017; Warrington et al., 2007). Observed differences in the pattern of change in SOC storage among irrigation methods, by contrast, offers direct insights into the effects of agricultural management practices on soil C dynamics but also highlights the difficulty in balancing management to maximize SOC storage and other environmental considerations, such as protecting water availability (Puy et al., 2017; Velasco-Munoz et al., 2019). Consequently, any prescriptive changes in irrigation management practices aimed at increasing SOC stocks must consider resource availability and the interactive effects of other management practices that are not directly discussed here. In their review, Trost et al.(2013) estimated that irrigation of semi-arid sites can increase SOC storage by 11 - 35%. Trost et al.(2013) used different inclusion criteria for their calculations: they included comparisons of rainfed and irrigated fields and comparisons of un-cultivated and irrigated fields, which we systematically excluded from our analysis because of low study numbers and confounding effects such as differences in plant species (i.e., agronomic versus native species), and the effect of land-use conversion.  Further, it is unclear to what soil depths the estimates reported in Trost et al. (2013) are referring.  However, our calculation of a 14.8% increase in SOC stocks at the 0-10 cm depth on semi-arid sites falls within their estimated range. That being said, we calculated much smaller gains in the full profile (8.0%) and deeper soil increments on  57  semi-arid sites (4.1% increase at the 20 - 30 cm depth and no significant change at the 10 - 20 cm and 30+ cm depths). This analysis brought to light critical gaps in the available data regarding irrigated agriculture across the globe. Despite its importance in soil dynamics, bulk density (BD) data were notably absent in a large number of studies. As such, considerable effort was required to accurately estimate bulk density values when they were not provided. Similarly, standard error or standard deviation data were often missing and required post hoc estimates from the assembled dataset. Any sort of gap-filling introduces uncertainty into the dataset and, as such, we recommend that this sort of data is reported in all future studies.  Additionally, a majority of studies did not report SOC values beyond near-surface depths. While we recognize that the greatest changes in SOC related to agricultural management practices are likely to occur near the soil surface, studies are increasingly showing that the overall effects of agricultural management and, in particular, irrigation, on SOC stocks are vastly misrepresented when only surface depths are considered (McNally et al., 2017; Mudge et al., 2017; Schipper et al., 2017; Trost et al., 2013). We therefore believe it is important that future studies include deeper sampling where possible. Finally, information regarding irrigated agricultural plots over longer duration (> 15 years) was notably limited, with 50% of the assembled studies being 5 years or less in duration, 39% being 15 years or less, and only 11% being longer than 15 years. Long-term studies meeting our study criteria were particularly absent for arid, dry sub-humid, and humid climate categories as well as for drip-irrigated plots. Further, while the literature often refers to the role of soil texture in SOC storage (Saiz et al., 2012; Trost et al., 2013; Wiesmeier et al., 2019; Zhong et al., 2018), few discuss the effect of irrigation  58  on changes in soil texture and the subsequent translocation of clay through the soil profile over the long term (Puy et al., 2017; Xu et al., 2016). Understanding how SOC dynamics are affected by translocation of clay (and nutrients) in response to irrigation practices is important to enhance the efficacy of global efforts aimed at offsetting greenhouse gas emissions via SOC sequestration.  In summary, we compiled and analysed data from 35 published studies that reported SOC stocks at the beginning and end of experiments conducted in irrigated agricultural systems across the globe. We found that irrigated agriculture tends to increase SOC stocks, particularly in surface soils, in fine- to medium-textured soils, in arid to semi-arid climates and under sprinkler irrigation.  Although numerous other variables, e.g., crop type, crop residue removal, and tillage practices, have been shown to have important effects on SOC storage, particularly in regional-scale studies, they were not significant contributors to the patterns of change in SOC storage observed under irrigated agriculture in this global-scale analysis. Annual precipitation, annual temperature and elevation also had no or only minimal value as explanatory variables. These findings demonstrate the value of considering aridity, irrigation method, and soil texture for future assessments of global-scale changes in SOC storage under irrigated agriculture.   Acknowledgements This work was funded by the Agricultural Greenhouse Gases Program of Agriculture and Agri-food Canada (Project AAGP2-25). Dr. Songchao Chen and Dr. Stephen I.C. Akpa provided R code snippets and advice in estimating bulk density values. Skye Wills similarly provided the R code snippets used in Sequeira et al (2014) for estimating bulk  59  density values. We are grateful to Ryan Stewart for putting us in touch with Jinshin Jian who provided us with R code snippets for generating multi-figure forest plots.  60  References Aguilera, E., Lassaletta, L., Gattinger, A., & Gimeno, B. S. (2013). Managing soil carbon for climate change mitigation and adaptation in Mediterranean cropping systems: A meta-analysis. Agriculture, Ecosystems & Environment, 168, 25\u201336. Akpa, S. I. C., Ugbaje, S. U., Bishop, T. F. A., & Odeh, I. O. A. (2016). Enhancing pedotransfer functions with environmental data for estimating bulk density and effective cation exchange capacity in a data-sparse situation. Soil Use and Management, 32(4), 644\u2013658. Andrews, D. M., Robb, T., Elliott, H., & Watson, J. E. (2016). Impact of long-term wastewater irrigation on the physicochemical properties of humid region soils: \u201cThe Living Filter\u201d site case study. Agricultural Water Management, 178, 239\u2013247. Angers, D. A., & Eriksen-Hamel, N. S. (2008). Full-inversion tillage and organic carbon distribution in soil profiles: A meta-analysis. Soil Science Society of America Journal. Soil Science Society of America, 72, 1370\u20131374. Bai, X., Huang, Y., Ren, W., Coyne, M., Jacinthe, P.-A., Tao, B., Hui, D., Yang, J., & Matocha, C. (2019). Responses of soil carbon sequestration to climate smart agriculture practices: A meta-analysis. Global Change Biology. https:\/\/doi.org\/10.1111\/gcb.14658 Bashir, Y., & Conlon, K. C. (2018). Step by step guide to do a systematic review and meta-analysis for medical professionals. Irish Journal of Medical Science, 187(2), 447\u2013452. Chen, S., Richer-de-Forges, A. C., Saby, N. P. A., Martin, M. P., Walter, C., & Arrouays, D. (2018). Building a pedotransfer function for soil bulk density on regional dataset  61  and testing its validity over a larger area. Geoderma, 312, 52\u201363. Condron, L. M., Hopkins, D. W., Gregorich, E. G., Black, A., & Wakelin, S. A. (2014). Long-term irrigation effects on soil organic matter under temperate grazed pasture: Pasture production and soil organic matter. European Journal of Soil Science, 65(5), 741\u2013750. Das, D., Dwivedi, B. S., Singh, V. K., Datta, S. P., Meena, M. C., Chakraborty, D., Bandyopadhyay, K. K., Kumar, R., & Mishra, R. P. (2017). Long-term effects of fertilisers and organic sources on soil organic carbon fractions under a rice\u2013wheat system in the Indo-Gangetic Plains of north-west India. Soil Research, 55(3), 296\u2013308. David, C., Lemke, R., Helgason, W., & Farrell, R. E. (2018). Current inventory approach overestimates the effect of irrigated crop management on soil-derived greenhouse gas emissions in the semi-arid Canadian Prairies. Agricultural Water Management, 208, 19\u201332. De Bona, F. D., Bayer, C., Dieckow, J., & Bergamaschi, H. (2008). Soil quality assessed by carbon management index in a subtropical Acrisol subjected to tillage systems and irrigation. Soil Research, 46(5), 469\u2013475. Denef, K., Stewart, C. E., Brenner, J., & Paustian, K. (2008). Does long-term center-pivot irrigation increase soil carbon stocks in semi-arid agro-ecosystems? Geoderma, 145(1), 121\u2013129. Dersch, G., & B\u00f6hm, K. (2001). Effects of agronomic practices on the soil carbon storage potential in arable farming in Austria. Nutrient Cycling in Agroecosystems, 60, 49\u201355.  62  Don, A., Schumacher, J., & Freibauer, A. (2011). Impact of tropical land-use change on soil organic carbon stocks - a meta-analysis. Global Change Biology, 17(4), 1658\u20131670. Dong, L., Yu, D., Zhang, H., Zhang, M., Jin, W., Liu, Y., & Shi, X. (2015). Long-term effect of sediment laden Yellow River irrigation water on soil organic carbon stocks in Ningxia, China. Soil and Tillage Research, 145, 148\u2013156. Drewry, J. J., Carrick, S., Penny, V., Houlbrooke, D. J., Laurenson, S., & Mesman, N. L. (2020). Effects of irrigation on soil physical properties in predominantly pastoral farming systems: a review. New Zealand Journal of Agricultural Research, 1\u201325. Du, Z., Angers, D. A., Ren, T., Zhang, Q., & Li, G. (2017). The effect of no-till on organic C storage in Chinese soils should not be overemphasized: A meta-analysis. Agriculture, Ecosystems & Environment, 236, 1\u201311. Ellmer, F., & Baumecker, M. (2002). 65 Years Long-term Experiments at Thyrow. Results for Sustainable Crop Production at Sandy Soils. Archives of Agronomy and Soil Science, 48(5), 521\u2013531. Field, A. P., & Gillett, R. (2010). How to do a meta-analysis. The British Journal of Mathematical and Statistical Psychology, 63(Pt 3), 665\u2013694. Getaneh, F., Deressa, A., & Negassa, W. (2007). Influence of small scale irrigation on selected soil chemical properties. Utilization of Diversity in Land Use Systems: Sustainable and Organic Approaches to Meet Human Needs. Tropentag, Witzenhausen, Germany. Gillabel, J., Denef, K., Brenner, J., Merckx, R., & Paustian, K. (2007). Carbon sequestration and soil aggregation in center-pivot irrigated and dryland cultivated  63  farming systems. Soil Sci. Soc. Am. J., 71, 1020\u20131028. Guo, L. B., & Gifford, R. M. (2002). Soil carbon stocks and land use change: A meta analysis. Global Change Biology, 8(4), 345\u2013360. Gurevitch, J., Koricheva, J., Nakagawa, S., & Stewart, G. (2018). Meta-analysis and the science of research synthesis. Nature, 555(7695), 175\u2013182. Guswa, A. J. (2008). The influence of climate on root depth: A carbon cost-benefit analysis. Water Resources Research, 44. W02427. https:\/\/doi.org\/10.1029\/2007WR006384 Halupka, L., & Halupka, K. (2017). The effect of climate change on the duration of avian breeding seasons: a meta-analysis. Proceedings. Biological Sciences \/ The Royal Society, 284(1867). https:\/\/doi.org\/10.1098\/rspb.2017.1710 H\u00e4ring, V., Manka\u2019abusi, D., Akoto-Danso, E. K., Werner, S., Atiah, K., Steiner, C., Lompo, D. J. P., Adiku, S., Buerkert, A., & Marschner, B. (2017). Effects of biochar, waste water irrigation and fertilization on soil properties in West African urban agriculture. Scientific Reports, 7(1), 10738. Hassink, J., Bouwman, L. A., Zwart, K. B., & Brussaard, L. (1993). Relationships between habitable pore space, soil biota and mineralization rates in grassland soils. Soil Biology & Biochemistry, 25(1), 47\u201355. Hedges, L. V., Gurevitch, J., & Curtis, P. S. (1999). The meta-analysis of response ratios in experimental ecology. Ecology, 80(4), 1150\u20131156. Hentati, O., Chaker, S., Wali, A., Ayoub, T., & Ksibi, M. (2014). Effects of long-term irrigation with treated wastewater on soil quality, soil-borne pathogens, and living organisms: case study of the vicinity of El Hajeb (Tunisia). Environmental Monitoring  64  and Assessment, 186(5), 2671\u20132683. Jerabkova, L., Prescott, C. E., Titus, B. D., Hope, G. D., & Walters, M. B. (2011). A meta-analysis of the effects of clearcut and variable-retention harvesting on soil nitrogen fluxes in boreal and temperate forests. Canadian Journal of Forest Research. Journal Canadien de La Recherche Forestiere, 41(9), 1852\u20131870. Jian, J., Du, X., Reiter, M. S., & Stewart, R. D. (2020). A meta-analysis of global cropland soil carbon changes due to cover cropping. Soil Biology & Biochemistry, 143, 107735. Kallenbach, C. M., Rolston, D. E., & Horwath, W. R. (2010). Cover cropping affects soil N2O and CO2 emissions differently depending on type of irrigation. Agriculture, Ecosystems & Environment, 137(3), 251\u2013260. Levidow, L., Zaccaria, D., Maia, R., Vivas, E., Todorovic, M., & Scardigno, A. (2014). Improving water-efficient irrigation: Prospects and difficulties of innovative practices. Agricultural Water Management, 146, 84\u201394. Le Quere, C., Moriarty, R., Andrew, R. M., Canadell, J. G., Sitch, S., Boden, T. A., & al., E. (2015). Global carbon project: The global carbon budget 2015 (V.1.0., issued Nov. 2015 and V.1.1, issued Dec. 2015) [Data set]. Environmental System Science Data Infrastructure for a Virtual Ecosystem; Carbon Dioxide Information Analysis Center (CDIAC), Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). https:\/\/doi.org\/10.3334\/CDIAC\/GCP_2015 Libohova, Z., Seybold, C., Wysocki, D., Wills, S., Schoeneberger, P., Williams, C., Lindbo, D., Stott, D., & Owens, P. R. (2018). Reevaluating the effects of soil organic matter and other properties on available water-holding capacity using the National Cooperative Soil Survey Characterization Database. Journal of Soil and Water  65  Conservation, 73(4), 411\u2013421. Liu, C., Holst, J., Br\u00fcggemann, N., Butterbach-Bahl, K., Yao, Z., Han, S., Han, X., & Zheng, X. (2008). Effects of irrigation on nitrous oxide, methane and carbon dioxide fluxes in an Inner Mongolian steppe. Advances in Atmospheric Sciences, 25(5), 748\u2013756. McGill, B. M., Hamilton, S. K., Millar, N., & Robertson, G. P. (2018). The greenhouse gas cost of agricultural intensification with groundwater irrigation in a Midwest US row cropping system. Global Change Biology, 24(12), 5948\u20135960. McNally, S. R., Beare, M. H., Curtin, D., Meenken, E. D., Kelliher, F. M., Calvelo Pereira, R., Shen, Q., & Baldock, J. (2017). Soil carbon sequestration potential of permanent pasture and continuous cropping soils in New Zealand. Global Change Biology, 23(11), 4544\u20134555. Midwood, A. J., Hannam, K. D., Forge, T. A., Neilsen, D., Emde, D., & Jones, M. D. (2020). Importance of drive-row vegetation for soil carbon storage in woody perennial crops: A regional study. Geoderma, 377, 114591. Minasny, B., Malone, B. P., McBratney, A. B., Angers, D. A., Arrouays, D., Chambers, A., Chaplot, V., Chen, Z.-S., Cheng, K., Das, B. S., Field, D. J., Gimona, A., Hedley, C. B., Hong, S. Y., Mandal, B., Marchant, B. P., Martin, M., McConkey, B. G., Mulder, V. L., \u2026 Winowiecki, L. (2017). Soil carbon 4 per mille. Geoderma, 292, 59\u201386. Moonen, A.-C., & B\u00e0rberi, P. (2008). Functional biodiversity: An agroecosystem approach. Agriculture, Ecosystems & Environment, 127(1), 7\u201321. Mudge, P. L., Kelliher, F. M., Knight, T. L., O\u2019Connell, D., Fraser, S., & Schipper, L. A.  66  (2017). Irrigating grazed pasture decreases soil carbon and nitrogen stocks. Global Change Biology, 23(2), 945\u2013954. Ogle, S. M., Breidt, F. J., & Paustian, K. (2005). Agricultural management impacts on soil organic carbon storage under moist and dry climatic conditions of temperate and tropical regions. Biogeochemistry, 72(1), 87\u2013121. Parchami, A. (2016). Weighted.Desc.Stat: Weighted Descriptive Statistics. https:\/\/CRAN.R-project.org\/package=Weighted.Desc.Stat Petit, S., Boursault, A., Le Guilloux, M., Munier-Jolain, N., & Reboud, X. (2011). Weeds in agricultural landscapes. A review. Agronomy for Sustainable Development, 31(2), 309\u2013317. Poeplau, C., Don, A., Vesterdal, L., Leifeld, J., Van Wesemael, B., Schumacher, J., & Gensior, A. (2011). Temporal dynamics of soil organic carbon after land-use change in the temperate zone - carbon response functions as a model approach. Global Change Biology, 17(7), 2415\u20132427. Puy, A., Garcia Aviles, J. M., Balbo, A. L., Keller, M., Riedesel, S., Blum, D., & Bubenzer, O. (2017). Drip irrigation uptake in traditional irrigated fields: The edaphological impact. Journal of Environmental Management, 202, 550\u2013561. Qian, B., Gregorich, E. G., Gameda, S., Hopkins, D. W., & Wang, X. L. (2011). Observed soil temperature trends associated with climate change in Canada. Journal of Geophysical Research, 116(D2), 291. Ramirez-Fuentes, E., Lucho-constantino, C., Escamilla-Silva, E., & Dendooven, L. (2002). Characteristics, and carbon and nitrogen dynamics in soil irrigated with wastewater for different lengths of time. Bioresource Technology, 85(2), 179\u2013187.  67  R Core Team. (2020). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. https:\/\/www.R-project.org\/ Rosenberg, M. S. (2005). The file-drawer problem revisited: a general weighted method for calculating fail-safe numbers in meta-analysis. Evolution; International Journal of Organic Evolution, 59(2), 464\u2013468. Rotenberg, D., Cooperband, L., & Stone, A. (2005). Dynamic relationships between soil properties and foliar disease as affected by annual additions of organic amendment to a sandy-soil vegetable production system. Soil Biology & Biochemistry, 37(7), 1343\u20131357. Rusinamhodzi, L., Corbeels, M., van Wijk, M. T., Rufino, M. C., Nyamangara, J., & Giller, K. E. (2011). A meta-analysis of long-term effects of conservation agriculture on maize grain yield under rain-fed conditions. Agronomy for Sustainable Development, 31(4), 657. Saiz, G., Bird, M. I., Domingues, T., Schrodt, F., Schwarz, M., Feldpausch, T. R., Veenendaal, E., Djagbletey, G., Hien, F., Compaore, H., Diallo, A., & Lloyd, J. (2012). Variation in soil carbon stocks and their determinants across a precipitation gradient in West Africa. Global Change Biology, 18(5), 1670\u20131683. S\u00e1nchez-Mart\u00edn, L., Arce, A., Benito, A., Garcia-Torres, L., & Vallejo, A. (2008). Influence of drip and furrow irrigation systems on nitrogen oxide emissions from a horticultural crop. Soil Biology & Biochemistry, 40(7), 1698\u20131706. Sanchez-Mart\u00edn, L., Meijide, A., Garcia-Torres, L., & Vallejo, A. (2010). Combination of drip irrigation and organic fertilizer for mitigating emissions of nitrogen oxides in semiarid climate. Agriculture, Ecosystems & Environment, 137(1), 99\u2013107.  68  Schipper, L. A., Dodd, M. B., Pronger, J., Mudge, P. L., Upsdell, M., & Moss, R. A. (2013). Decadal changes in soil carbon and nitrogen under a range of irrigation and phosphorus fertilizer treatments. Soil Science Society of America Journal. Soil Science Society of America, 77, 246\u2013256. Schipper, L. A., Mudge, P. L., Kirschbaum, M. U. F., Hedley, C. B., Golubiewski, N. E., Smaill, S. J., & Kelliher, F. M. (2017). A review of soil carbon change in New Zealand\u2019s grazed grasslands. New Zealand Journal of Agricultural Research, 60(2), 93\u2013118. Sch\u00fctt, M., Borken, W., Spott, O., Stange, C. F., & Matzner, E. (2014). Temperature sensitivity of C and N mineralization in temperate forest soils at low temperatures. Soil Biology & Biochemistry, 69, 320\u2013327. Sequeira, C. H., Wills, S. A., Seybold, C. A., & West, L. T. (2014). Predicting soil bulk density for incomplete databases. Geoderma, 213, 64\u201373. Tautges, N. E., Chiartas, J., Gaudin, A. C. M., O\u2019Geen, A. T., Herrera, I., & M Scow, K. (2019). Deep soil inventories reveal that impacts of cover crops and compost on soil carbon sequestration differ in surface and sub-surface soils. Global Change Biology. https:\/\/doi.org\/10.1111\/gcb.14762 Telo da Gama, J., Rato Nunes, J., Loures, L., Lopez Pi\u00f1eiro, A., & Vivas, P. (2019). Assessing spatial and temporal variability for some edaphic characteristics of Mediterranean rainfed and irrigated soils. Agronomy, 9(3), 132. Trabucco, A., & Zomer, R. J. (2018). Global Aridity Index and Potential Evapo-Transpiration (ET0) Climate Database v2. CGIAR Consortium for Spatial Information (CGIAR-CSI). Available Online: Https:\/\/figshare. com\/articles\/Global_Aridity_Index_and_Potential_Evapotranspiration_ET0_Climate 69  _Database_v2\/7504448\/1 (accessed on 25 June 2019). Trost, B., Prochnow, A., Drastig, K., Meyer-Aurich, A., Ellmer, F., & Baumecker, M. (2013). Irrigation, soil organic carbon and N2O emissions. A review. Agronomy for Sustainable Development, 33(4), 733\u2013749. The United Nations World Water Development Report 2014. (2014). http:\/\/www.unesco.org\/new\/en\/natural-sciences\/environment\/water\/wwap\/facts-and-figures\/all-facts-wwdr3\/fact-24-irrigated-land\/ Velasco-Munoz, J. F., Aznar-Sanchez, J. A., Batlles-delaFuente, A., & Fidelibus. (09\/2019). Sustainable irrigation in agriculture: An analysis of global research. WATER, 11(9), 1758. Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3). Wagner, S., Cattle, S. R., & Scholten, T. (2007). Soil-aggregate formation as influenced by clay content and organic-matter amendment. Journal of Plant Nutrition and Soil Science, 170(1), 173\u2013180. Warrington, D. N., Goldstein, D., & Levy, G. J. (2007). Clay translocation within the soil profile as affected by intensive irrigation with treated wastewater. Soil Science, 172(9), 692. Wei, X., Shao, M., Gale, W., & Li, L. (2014). Global pattern of soil carbon losses due to the conversion of forests to agricultural land. Scientific Reports, 4, 4062. Wiesmeier, M., Urbanski, L., Hobley, E., Lang, B., von L\u00fctzow, M., Marin-Spiotta, E., van Wesemael, B., Rabot, E., Lie\u00df, M., Garcia-Franco, N., Wollschl\u00e4ger, U., Vogel, H.-J., & K\u00f6gel-Knabner, I. (2019). Soil organic carbon storage as a key function of soils  70  - A review of drivers and indicators at various scales. Geoderma, 333, 149\u2013162. Fick, S.E. and R.J. Hijmans, 2017. WorldClim 2: new 1km spatial resolution climate surfaces for global land areas. International Journal of Climatology 37 (12): 4302-4315. Xu, H., Sieverding, H., Kwon, H., Clay, D., Stewart, C., Johnson, J. M. F., Qin, Z., Karlen, D. L., & Wang, M. (2019). A global meta\u2010analysis of soil organic carbon response to corn stover removal. GCB Bioenergy, 11(10), 1215\u20131233. Xu, X., Shi, Z., Li, D., Rey, A., Ruan, H., Craine, J. M., Liang, J., Zhou, J., & Luo, Y. (2016). Soil properties control decomposition of soil organic carbon: Results from data-assimilation analysis. Geoderma, 262, 235\u2013242. Zhong, Z., Chen, Z., Xu, Y., Ren, C., Yang, G., Han, X., Ren, G., & Feng, Y. (2018). Relationship between soil organic carbon stocks and clay content under different climatic conditions in central China. Forests, Trees and Livelihoods, 9(10), 598. 71  Supplementary Materials S1. Note: The following Systematic Review Protocol was created at the outset of this research project, to guide and document the data collection process. Once the literature search and subsequent data collection had been completed, it was determined that there were insufficient data to investigate the effects of irrigated agriculture on soil inorganic carbon or total soil nitrogen. We therefore focussed our analysis on soil organic carbon. Although data were gathered from three broad types of study: i) studies that compare SOC in irrigated and rainfed plots; ii) studies that compare SOC in unmanaged \u2018natural\u2019 and irrigated plots; and iii) studies that compare SOC in irrigated plots at the beginning and end of an experiment, we determined that there were only sufficient data to perform a robust meta-analysis on data from the third type of study. We therefore narrowed the scope of the meta-analysis, to examine the effects of irrigated agriculture on SOC dynamics over time on the same plots. Additionally, \u201ccontrol\u201d as described below is no longer relevant. Control plots are described in the Methods section of the accompanying manuscript. Systematic review protocol AGGP 2  Administrative Information Title:  Soil organic carbon in irrigated agricultural systems: A meta-analysis. Authors: David Emde, Kirsten Hannam, Ilka Most, Louise Nelson, and Melanie Jones  Department of Biology, University of British Columbia  72  Science Building, SCI372, 1177 Research Road, Kelowna, BC V1V 1V7 CANADA Support:  Agricultural Greenhouse Gases Program (AGGP2-025)  Introduction Rationale: Combustion of fossil fuels and land use change have contributed to a massive rise in atmospheric CO2 concentrations that is altering our climate at an alarming rate.  Soil management practices that optimize soil carbon storage, such as efficient, strategically-managed irrigation, have been identified as key tools for mitigating rising CO2 levels.  Numerous studies have investigated the effects of irrigation on soil C and N pools.  However, there is an urgent need for a more integrated understanding of the impact of irrigation on soil C and N storage in order to identify the conditions under which irrigation can best be used to mitigate increased atmospheric carbon dioxide concentrations and preserve\/improve soil quality. Objectives: The objectives of our study are: i. to examine the impact of irrigation on soil organic and inorganic carbon and nitrogen pools, and ii. to identify the site characteristics and irrigation methods that best facilitate soil C and N retention.      73  Methods Eligibility criteria: Studies will be selected according to the following criteria (as per the PICO process): \uf0b7 Population:  The population used for this analysis will include published data from field experiments documenting the effect of irrigation on soil C, soil N and\/or soil pH. Because irrigation is not expected to cause immediate measurable changes in soil total, organic and inorganic carbon stocks, only data from studies in which irrigation treatments were applied for at least one growing season will be included.  In order to ensure that all co-authors of this paper can read every source article, only papers that are available in English will be included. Only studies that are directly relevant to the agricultural sector in Canada will be included; thus, data from field studies conducted in tropical regions, on rice fields and on forested sites will be excluded.  Given their short-term nature, data from greenhouse and laboratory studies will also be excluded. \uf0b7 Intervention:  In order to examine the effect of irrigation method on soil C and N stocks, data from field studies using irrigation, in any form (e.g., furrow and flood irrigation; irrigation with high-pressure sprinklers; and irrigation with low-pressure sprinklers, sprayers, micro-jets and drip emitters) will be included. \uf0b7 Control: Control treatments will consist of sites or plots that have never been irrigated or have remained un-irrigated for a known period of time. Control sites or plots may or may not have been cultivated. \uf0b7 Outcome: Measurement data of interest will include: soil total C, organic C, inorganic C and total N content or concentration.   74  Information sources: The literature search will be restricted to two electronic databases: i. \u2018Web of Science\u2019 and ii. \u2018Google Scholar\u2019.   As well, reference sections from relevant literature reviews will be scanned to identify studies that did not turn up during searches of the electronic databases. Search strategy: Search terms will include: agric* AND irrigat* AND ((\"soil carbon\" OR (soil AND \"inorganic carbon\")) OR \"soil nitrogen\" OR \"soil pH\") NOT (rice OR tropic* OR forest*) Study selection: Literature search results will be uploaded to Open Science Framework (OSF), an internet based software program that facilitates collaboration among authors (David Emde, Ilka Most, Kirsten Hannam, and Melanie Jones) and reviewers. OSF will be used to store and manage gathered literature; Excel spreadsheets with extracted data; and relevant documents, including this search protocol. David Emde, Ilka Most and Kirsten Hannam will independently screen the titles, abstracts and results yielded by the literature search against inclusion criteria. Full reports will be given for all titles that appear to meet the inclusion criteria or where there is any uncertainty. Where necessary, review authors will discuss papers to determine whether they meet the inclusion criteria and seek additional information from study authors to resolve questions about eligibility. Rationale for the exclusion of literature will be recorded. Neither of the review authors will be blind to the journal titles or the study authors or institutions.  75  Data collection process: Relevant data will be extracted from papers and entered into a formalized, shared Excel spreadsheet. Where necessary, datapoints (i.e., means and error bars) will be estimated using the WebPlotDigitizer tool. Undefined error bars will be assumed to be standard errors. Study designs will be carefully scrutinized to validate claims of replication; where samples were collected less than 100 m apart within a single treatment area, they will be re-assigned an \u2018n\u2019 of 1, and reported values of standard deviation or standard error will be ignored. Data items: When available, the following response variables will be extracted from each selected paper. Units will be converted, where possible and appropriate:  \uf0b7 soil total carbon content and\/or concentration (plus bulk density) \uf0b7 soil organic carbon content and\/or concentration (plus bulk density) \uf0b7 soil inorganic carbon content and\/or concentration (plus bulk density) \uf0b7 soil nitrogen content and\/or concentration (plus bulk density) \uf0b7 soil pH \uf0b7 note that bulk density data will be used to convert soil C and N concentrations into soil C and N content When available, the following explanatory variables will be extracted from each selected paper:   76  \uf0b7 irrigation method and irrigation frequency \uf0b7 natural precipitation \uf0b7 soil texture and pH \uf0b7 crop type (i.e., forage, corn, fruit trees, field crops, etc.) \uf0b7 fertilizer and organic amendment (type and frequency)   77   Figure S2: Variable importance plot for the bulk density (BD) predictive algorithm used to determine missing BD values for C stock calculations. 78  Figure S3: Crop types included in each crop category for the meta-analysis. Cereals Horticulture Wood Crops Wheat Alfalfa Maize Barley Sorghum Legume Sugar beet Potato Snap pea Cucumber Cauliflower Red cabbage Wolfberry Grape Cotton Kiwi Apricot Olives  79   Figure S4: Publication bias for whole dataset shown by funnel plot. Funnel plot on the left shows the whole dataset; funnel plot on the right shows the results of the \u201ctrim and fill\u201d method. A lack of hollow dots on the funnel plot on the right indicates that no problematic publication bias is present.  80   S5: Fail-safe N calculation using the Rosenberg Approach. Fail-safe N Calculation Using the Rosenberg Approach  Average Effect Size:        0.1331 Observed Significance Level: 0.0002 Target Significance Level:   0.05  Fail-safe N: 192  81  S6: Explanatory variables that were evaluated for their role in determining the strength and direction of changes in SOC stocks in irrigated agricultural systems. Variables with \u2018Insufficient reported data\u2019 were unavailable for > 20% of the study sites and could not be reasonably estimated (e.g., using open-source raster data). Variable Inclusion details Significance Geographical Location Relevant aspects of geographical location described in other variables (i.e. temperature and precipitation). NA Elevation Insufficient reported data, filled in using rasters. Relevant aspects reflected in the Aridity Index. Not significant. Avg. annual precip. Insufficient reported data, filled in using rasters. Included in the Aridity Index calculation. Avg. annual temp. Insufficient reported data, filled in using rasters. Included in the Aridity Index calculation. Climate Reported often but inconsistent climate index. Aridity Index used as proxy. Aridity Index Calculated from rasters. Combines relevant aspects of precip. temp. and climate. R2: 0.1336, p-value: 0.0002507 Crop type Information not always complete. Not significant. Multiple crops\/rotation Information not always complete. Not significant. Tillage frequency Insufficient reported data. NA Tillage type Insufficient reported data. NA Crop residue removal Insufficient reported data. NA Grazing status Not discussed. Not significant. Use of cover crops Not discussed. Not significant. Inorganic N application Not discussed. Not significant.  82  Herbicide use Insufficient reported data. NA Organic matter application Not discussed. Not significant. Inorganic C application Not discussed. Not significant. Years since irrigation Not discussed. Not significant. Irrigation method Included in analyses. R2: 0.8383, p-value: 0.02318 Irrigation water source Only fresh water included in the analysis. NA Irrigation water pH Insufficient reported data. NA Irrigation water HCO3 Insufficient reported data. NA Irrigation water Ca Insufficient reported data. NA Irrigation water organic C Insufficient reported data. NA Irrigation water N Insufficient reported data. NA Study scale Not discussed. NA Sample depth Included in analyses. Significant when separated by study categories (aridity category, irrigation method, soil texture). Soil bulk density Insufficient reported data, filled in using pedotransfer analog.  Included in SOC stock calculations. Soil texture (% sand\/clay) Included in analyses for discussion purposes. Many papers have pointed out the importance of texture to SOC storage. Not significant. Soil pH Insufficient reported data. NA  83  Number of samples (n) Included in calculations. NA Organic C (incl. SD) Insufficient reported data, SD calculated using coefficient of variance. Included in effect size CI. NA  84  S7: Previous agronomic history of study sites, as reported in the literature.  Publication Climate Category Reported Agronomic History Alidoust et al. 2018 Arid Gradually deforested oak forest, converted to agriculture with conventional tillage. Bedbabis et al. 2015 Arid Drip irrigated olive orchard established in 1987, study ran from 2003 to 2012. Benjamin et al. 2010 Semi-Arid Conducted on USDA-ARS Central Great Plains Research Station. Irrigation regime in place for four years prior to study. Eleftheriadis and Turri\u00f3n 2014 Semi-Arid Deforested Quercus forests. Established agricultural land managed consistently with unchanging practices prior to the study. Faria et al. 2004 Semi-Arid Established intercropped, irrigated vineyard. Follet et al. 2005 Semi-Arid Conducted on Bajio Research Station, National Institute of Agricultural, Forestry, and Livestock Research - Mexico. For the year prior to establishment of experimental treatments the plots were conventionally farmed with winter wheat followed by sorghum, then disc plowed. Ghimire et al. 2017 Semi-Arid Conducted at the University of Wyoming Sustainable Agriculture Research and Extension Center. Planted with Continuous corn for two years prior to experiment. Halvorson and Jantalia 2011 Semi-Arid Continuously cropped with corn prior to study with center-pivot irrigation. Halvorson et al. 2003 Semi-Arid Continuously cropped with corn for 6 years prior to the study using a moldboard plow production system.  85  Hao et al. 2001  Semi-Arid For at least eight years prior to the study, the site was cropped with barley for silage. Little crop residue was returned to the soil. Hegde 1996 Semi-Arid Permanent plots under the All India Coordinated Research Project on Cropping Systems at Akola, Parbhani, and Rahuri in Maharashtra State, India. Hou et al. 2012 Semi-Arid Conducted at Yucheng Comprehensive Experiment Station of China Academy of Science. Before establishment of treatment plots, the study field was double-cropped with winter-wheat and summer maize under conventional tillage for five years. Residues from this crop were incorporated into the soil. Hulugalle et al. 2013 Semi-Arid Conducted in permanent plots at the Australian Cotton Research Institute, New South Wales, Australia. Iqbal et al. 2012 Arid Conducted on the Research Farm of the Institute of Soil and Environmental Sciences, University of Agriculture, Faisalabad, Pakistan. Existing farming systems in the region (not specified if this site is specifically included) include irrigated corn, wheat, cotton, sugarcane, rice, fodder, and pulses. Larney et al. 2017 Semi-Arid Conducted at the Vauxhall Sub-station of Agriculture and Agri-food Canada, in the Bow River Irrigation District of southern Alberta. Planted to barley one year prior to establishment of treatments. Leavitt et al. 2001 Arid Conducted at the University of Arizona Maricopa Agricultural Center. Lenka et al. 2014 Dry sub-humid Conducted on the research farm of the Indian Agricultural Research Institute, New Delhi. Libutti et al. 2018 Semi-Arid Conducted within fields of the Fiordelisi agricultural and food manufacturing company in the Apulia Region of Southern Italy.  86  Ma et al. 2018 Arid Rainfed agriculture established some time prior to 1969. Plots in the area may have been abandoned for some time prior to study due to saline groundwater, but it isn\u2019t clear whether the study plot falls under this category.  Montanaro et al. 2010 Semi-Arid Conducted at an established orchard in South Italy. Morug\u00e1n-Coronado et al. 2011 Semi-Arid Conducted on a cultivated terrace in Biar, Spain. Mu\u00f1oz and Ram\u00edrez 2007 Semi-Arid For one growing season prior to the initialization of treatment plots, the site was cropped with maize under irrigation using standard tillage practices including deep ploughing prior to planting and conventional tillage. Olson and Papworth 2006 Semi-Arid Study site established on 2 yr-old alfalfa. P\u00e9rez-Berm\u00fadez et al. 2016 Semi-Arid Conducted in a commercial winery in the D.D. Utiel-Requena area of Spain. The winery was planted 11 years prior to the study. Rochester 2011 Semi-Arid Conducted at the Australian Cotton Research Institute, Narrabri, New South Wales. Australia. Cropped with irrigated cotton and wheat for at least 25 years prior to the study. Rotenberg et al. 2005 Humid Conducted at the University of Wisconsin Hancock Agricultural Research Station. Sainju et al. 2014 Semi-Arid Study site converted from Conservation Reserve Program grassland to cropland three years prior to the first experimental measurement.  87  S\u00e1nchez-Garc\u00eda et al. 2016 Semi-Arid Conducted on an established commercial organic olive orchard within the \u201cSAT Casa Pareja\u201d farm. The farm has been organically managed for the 15 years prior to the study. Schmer et al. 2014 Dry sub-humid Conducted at the University of Nebraska Agricultural Research and Development Center in an established irrigated site under disk tillage. Silva et al. 2016 Semi-Arid Conducted at the Bebedouro experimental station of Embrapa Semi-Arid (Brazil).  Stewart et al. 2017 Semi-Arid Conducted at the Agricultural Research Development and Education Center near Fort Collins, CO. Field was previously continuously cropped to corn for 6 years under a conventional tillage management regime. Trost et al. 2016 Dry sub-humid Conducted at the Field Study and Research Station of the Humboldt-University of Berlin. Undersander and Ger 1985  Semi-Arid Conducted on plots converted from native buffalograss two years prior to the first organic matter measurement. Wang et al. 2014 Dry sub-humid Conducted at the Experimental Farm of the Northwest Agriculture and Forestry University of Yangling, Shaanxi province, China.  Xing et al. 2012 Humid Conducted on the Research Farm near the Potato Research Centre, Agriculture and Agri-food Canada in Fredericton, New Brunswick, Canada.    88    Figure S8: Change in SOC stocks by study duration (short (\u22645 years); medium (6-15 years); long (>15 years)) and soil depth.  Study duration is defined as the number of years between initial SOC measurement (t1) and final SOC measurement (t2)). Shaded grey area indicates 95% confidence interval. 89    Figure S9: Change in SOC stocks by study duration (short (\u22645 years); medium (6-15 years); long (>15 years)), soil depth and climatic category (noted at the top of each panel). Study duration is defined as the number of years between initial SOC measurement (t1) and final SOC measurement (t2)). Filled circles with error bars represent the mean change in SOC and their respective 95% CI. Where 95% CI does not overlap 0% (indicated by a vertical red line), SOC has significantly changed over time. The numbers to the right of each datapoint indicate the total number of observations used to calculate each datapoint.  90    Figure S10: Change in SOC stocks by study duration (short (\u22645 years); medium (6-15 years); long (>15 years)), soil depth  and soil texture category (noted at the top of each panel). Study duration is defined as the number of years between initial SOC measurement (t1) and final SOC measurement (t2)). Filled circles with error bars represent the mean change in SOC and their respective 95% CI. Where 95% CI does not overlap 0% (indicated by a vertical red line), SOC has significantly changed over time. The numbers to the right of each datapoint indicate the total number of observations used to calculate each datapoint.  91    Figure S11: Change in SOC stocks by study duration (short (\u22645 years); medium (6-15 years); long (>15 years)), soil depth and irrigation method (noted at the top of each panel). Study duration is defined as the number of years between initial SOC measurement (t1) and final SOC measurement (t2)). Filled circles with error bars represent the mean change in SOC and their respective 95% CI. Where 95% CI does not overlap 0% (indicated by a vertical red line), SOC has significantly changed over time. The numbers to the right of each datapoint indicate the total number of observations used to calculate each datapoint.  92     S12:  The rate of change in SOC (%SOC yr-1), by aridity category and soil depth. Calculated as %SOC divided by the duration of the study. Central bars represent the median data point. ANOVA with Tukey\u2019s HSD a posteriori revealed no statistically significant effects of aridity category, soil depth, or interactions between aridity category and soil depth.  93    S13: The rate of change in SOC (%SOC yr-1), by soil texture category and soil depth. Calculated as %SOC divided by the duration of the study. Central bars represent the median data point. ANOVA with Tukey\u2019s HSD a posteriori revealed no statistically significant effects of texture category, soil depth, or interactions between texture category and soil depth. \u201cNA\u201d texture category denotes sites for which soil texture data was not provided.  94    S14:  The rate of change in SOC (%SOC yr-1), by irrigation method and soil depth. Calculated as %SOC divided by the duration of the study. Central bars represent the median data point.  ANOVA with Tukey\u2019s HSD a posteriori revealed that at the 20-30 cm depth, SOC decreased under drip irrigation and increased under sprinkler and flood\/furrow irrigation. \u201cNA\u201d irrigation method denotes sites for which the irrigation method was not provided.   95  S15: Studies from which data were extracted for this meta-analysis. Alidoust, E., Afyuni, M., Hajabbasi, M. A., & Mosaddeghi, M. R. (2018). Soil carbon sequestration potential as affected by soil physical and climatic factors under different land uses in a semiarid region. Catena, 171, 62\u201371. Bedbabis, S., Trigui, D., Ben Ahmed, C., Clodoveo, M. L., Camposeo, S., Vivaldi, G. A., & Ben Rouina, B. (2015). Long-terms effects of irrigation with treated municipal wastewater on soil, yield and olive oil quality. Agricultural Water Management, 160, 14\u201321. Benjamin, J. G., Halvorson, A. D., Nielsen, D. C., & Mikha, M. M. (2010). Crop management effects on crop residue production and changes in soil organic carbon in the Central Great Plains. Agronomy Journal, 102, 990\u2013997. Eleftheriadis, A., & Turri\u00f3n, M.-B. (2014). Soil microbiological properties affected by land use, management, and time since deforestations and crop establishment. European Journal of Soil Biology, 62, 138\u2013144. Faria, C. M. B., Soares, J. M., & Le\u00e3o, P. C. S. (2004). Aduba\u00e7\u00e3o verde com leguminosas em videira no subm\u00e9dio S\u00e3o Francisco. Revista Brasileira de Ciencia Do Solo, 28(4), 641\u2013648. Follett, R. F., Castellanos, J. Z., & Buenger, E. D. (2005). Carbon dynamics and sequestration in an irrigated Vertisol in Central Mexico. Soil and Tillage Research, 83(1), 148\u2013158. Ghimire, R., Norton, U., Bista, P., Obour, A. K., & Norton, J. B. (2017). Soil organic matter, greenhouse gases and net global warming potential of irrigated  96  conventional, reduced-tillage and organic cropping systems. Nutrient Cycling in Agroecosystems, 107(1), 49\u201362. Halvorson, A. D., & Jantalia, C. P. (2011). Nitrogen fertilization effects on irrigated no-till corn production and soil carbon and nitrogen. Agronomy Journal, 103, 1423\u20131431. Halvorson, A. D., Mosier, A. R., Reule USDA, C. A., Building, D., Collins, S. 100 F., & Email:, C. O. 80526. (2003). Irrigated Crop management effects on productivity, soil nitrogen, and soil carbon. Fertilizer In Dustry R O Und Table, O c Tober, 28, 30. Hao, X., Chang, C., & Lindwall, C. W. (2001). Tillage and crop sequence effects on organic carbon and total nitrogen content in an irrigated Alberta soil. Soil and Tillage Research, 62(3), 167\u2013169. Hegde, D. M. (1996). Long-term sustainability of productivity in an irrigated sorghum-wheat system through integrated nutrient supply. Field Crops Research, 48(2), 167\u2013175. Hou, R., Ouyang, Z., Li, Y., Tyler, D. D., Li, F., & Wilson, G. V. (2012). Effects of tillage and residue management on soil organic carbon and total nitrogen in the North China Plain. Soil Science Society of America Journal. Soil Science Society of America, 76, 230\u2013240. Hulugalle, N. R., Weaver, T. B., Finlay, L. A., & Heimoana, V. (2013). Soil organic carbon concentrations and storage in irrigated cotton cropping systems sown on permanent beds in a Vertosol with restricted subsoil drainage. Crop & Pasture Science, 64(8), 799\u2013805.  97  Iqbal, M., Khan, A. G., Hassan, A. U., & Amjad, M. (2012). Soil physical health indices, soil organic carbon, nitrate contents and wheat growth as influenced by irrigation and nitrogen rates. Int. J. Agric. Biol., 14(1), 1\u201310. Larney, F. J., Pearson, D. C., Blackshaw, R. E., & Lupwayi, N. Z. (2017). Soil changes over 12 yr of conventional vs. conservation management on irrigated rotations in southern Alberta. Canadian Journal of Soil Science, 97(2), 249\u2013265. Leavitt, S. W., Pendall, E., Paul, E. A., Brooks, T., Kimball, B. A., Pinter, P. J., Jr., Johnson, H. B., Matthias, A., Wall, G. W., & La Morte, R. L. (2001). Stable-carbon isotopes and soil organic carbon in wheat under CO enrichment. The New Phytologist, 15, 0\u2013305. Lenka, S., Singh, A. K., & Lenka, N. K. (2014). Soil aggregation and organic carbon as affected by different irrigation and nitrogen level sin the maize-wheat cropping system. Experimental Agriculture, 50(2), 216\u2013228.  Libutti, A., Gatta, G., Gagliardi, A., Vergine, P., Pollice, A., Beneduce, L., Disciglio, G., & Tarantino, E. (2018). Agro-industrial wastewater reuse for irrigation of a vegetable crop succession under Mediterranean conditions. Agricultural Water Management, 196, 1\u201314. Ma, Q. L., Wang, Y. L., Li, Y. K., Sun, T., & Milne, E. (2018). Carbon storage in a wolfberry plantation chronosequence established on a secondary saline land in an arid irrigated area of Gansu Province, China. Journal of Arid Land, 10(2), 202\u2013216. Montanaro, G., Celano, G., Dichio, B., & Xiloyannis, C. (2010). Effects of soil-protecting agricultural practices on soil organic carbon and productivity in fruit tree orchards. Land Degradation & Development, 21(2), 132\u2013138.  98  Morug\u00e1n-Coronado, A., Garc\u00eda-Orenes, F., Mataix-Solera, J., Arcenegui, V., & Mataix-Beneyto, J. (2011). Short-term effects of treated wastewater irrigation on Mediterranean calcareous soil. Soil and Tillage Research, 112(1), 18\u201326. Mu\u00f1oz, A., L\u00f3pez-Pi\u00f1eiro, A., & Ram\u00edrez, M. (2007). Soil quality attributes of conservation management regimes in a semi-arid region of south western Spain. Soil and Tillage Research, 95(1), 255\u2013265. Olson, B. M., & Papworth, L. W. (2006). Soil chemical changes following manure application on irrigated alfalfa and rainfed timothy in southern Alberta. Canadian Journal of Soil Science, 86(1), 119\u2013132. P\u00e9rez-Berm\u00fadez, P., Olmo, M., Gil, J., Garc\u00eda-F\u00e9rriz, L., Olmo, C., Boluda, R., & Gavidia, I. (2016). Cover crops and pruning in Bobal and Tempranillo vineyards have little influence on grapevine nutrition. Scientia Agricola, 73(3), 260\u2013265. Rochester, I. J. (2011). Sequestering carbon in minimum-tilled clay soils used for irrigated cotton and grain production. Soil and Tillage Research, 112(1), 1\u20137. Rotenberg, D., Cooperband, L., & Stone, A. (2005). Dynamic relationships between soil properties and foliar disease as affected by annual additions of organic amendment to a sandy-soil vegetable production system. Soil Biology & Biochemistry, 37(7), 1343\u20131357. Sainju, U. M., Stevens, W. B., Caesar-TonThat, T., Liebig, M. A., & Wang, J. (2014). Net global warming potential and greenhouse gas intensity influenced by irrigation, tillage, crop rotation, and nitrogen fertilization. Journal of Environmental Quality, 43(3), 777\u2013788.  99  S\u00e1nchez-Garc\u00eda, M., S\u00e1nchez-Monedero, M. A., Roig, A., L\u00f3pez-Cano, I., Moreno, B., Benitez, E., & Cayuela, M. L. (2016). Compost vs biochar amendment: a two-year field study evaluating soil C build-up and N dynamics in an organically managed olive crop. Plant and Soil, 408(1), 1\u201314. Schmer, M. R., Jin, V. L., Wienhold, B. J., Varvel, G. E., & Follett, R. F. (2014). Tillage and residue management effects on soil carbon and nitrogen under irrigated continuous corn. Soil Science Society of America Journal. Soil Science Society of America, 78, 1987\u20131996. Silva, D. J., Bassoi, L. H., Rocha, M. G., Silva, A. O., & Deon. (2016). Organic and nitrogen fertilization of soil under \u201cSyrah\u201d grapevine: effects on soil chemical properties and nitrate concentration. REv. Bras. Clienc. Solo, 40(e0150073), 1\u201311. Stewart, C. E., Halvorson, A. D., & Delgado, J. A. (2017). Long-term N fertilization and conservation tillage practices conserve surface but not profile SOC stocks under semi-arid irrigated corn. Soil and Tillage Research, 171, 9\u201318. Trost, B., Prochnow, A., Meyer-Aurich, A., Drastig, K., Baumecker, M., & Ellmer, F. (2016). Effects of irrigation and nitrogen fertilization on the greenhouse gas emissions of a cropping system on a sandy soil in northeast Germany. European Journal of Agronomy: The Journal of the European Society for Agronomy, 81, 117\u2013128. Undersander, D. J., & Ger, C. R. (1985). Management on continuous production of irrigated winter wheat. Published in Agron. J., 77, 508\u2013511. Wang, S., Tian, X., Liu, T., Lu, X., You, D., & Li, S. (2014). Irrigation, straw, and nitrogen management benefits wheat yield and soil properties in a dryland agro-ecosystem. Agronomy Journal, 106, 2193\u20132201.  100  Xing, Z., Toner, P., Chow, L., Rees, H. W., Li, S., & Meng, F. (2012). Effects of hay mulch on soil properties and potato tuber yield under irrigation and nonirrigation in New Brunswick, Canada. Journal of Irrigation and Drainage Engineering, 138(8), 703\u2013714.  101  Chapter 3: Estimating mineral associated organic carbon deficits in soils of the Okanagan Valley: A regional study with broader implications  Authors: David Emde1, Kirsten D. Hannam2, Andrew J. Midwood1, Melanie D. Jones1  1Biology Department, University British Columbia, Okanagan Campus, Kelowna, BC, Canada 2Agriculture and Agri-Food Canada, Summerland Research and Development Centre, BC, Canada  Corresponding author: David Emde (dsemde@gmail.com) Keywords: mineral associated organic carbon, carbon deficit, soil carbon, carbon sequestration, carbon stabilization, saturation  Abstract In order to successfully reduce atmospheric CO2 by sequestering additional soil carbon, it is essential to understand the potential of a given soil to store carbon in a stable form. Carbon that has formed organo-mineral complexes with silt and clay particles is believed to be less susceptible to decay than non-complexed, or particulate, organic carbon. Although intensive agricultural practices often disrupt this stable mineral-associated organic carbon (MAOC), resulting in losses of MAOC and a consequent release of CO2, a previous study found that MAOC in the top 30 cm of the soil profile in orchard and  102  vineyards of the Okanagan Valley, Canada, had more than doubled over several decades of cropping. It is unclear, however, if these coarse-textured, geologically young soils have the capacity to store additional carbon or if they have reached \u2018saturation\u2019. Using direct measurements of MAOC on a subset of our samples, and an approach developed previously for primarily allophanic soils, we took a modelling approach to estimate MAOC for 537 samples of much coarser and younger soils from 99 non-cultivated and agricultural sites in the Okanagan Valley. We used specific surface area (SSA) as an indication of the mineral surface area available for sorption of organic matter, in addition to standard soil physicochemical data. We used both Random Forest and Stepwise Multiple Regression with Akaike Information Criterion (AIC) to determine a best fit model for predicting MAOC concentrations in these soils and to determine the most efficient method for estimating SSA for use in MAOC modelling. Random Forest modelling using SSA in addition to total SOC, exchangeable calcium, exchangeable potassium and soil pH was the most accurate predictive approach for determining MAOC in these soils (78.2% variance explained), although using soil texture instead of SSA produced almost equivalent results. In order to determine if a MAOC deficit existed for these soils, we then applied a quantile regression approach using the predicted 90th quantile of MAOC from our samples to represent the MAOC formation capacity of these soils. We determined that while MAOC deficits were present in all soils, and increased with depth, more clay rich soils had greater MAOC deficits ( 1.62 g kg-1 for 0-15 cm, 4.01 g kg-1 for 15-30 cm, and 5.80 g kg-1 for 30-60 cm), whereas sandier soils showed smaller deficits (1.01  g kg-1 for 0-15 cm, 2.72 g kg-1 for 15-30 cm, and 3.69 g kg-1 for 30-60 cm). Furthermore, the upper 30 cm of these soils have a current MAOC stock of 168 million kg over 8501 ha and have the potential to increase MAOC stock by 17% (28.5 million kg of MAOC) before they reach formation capacity. This study highlights the variability in  103  the MAOC formation capacity of soils with different physicochemical properties, and provides a useful framework for estimating MAOC concentrations and predicting MAOC deficits that could be applied to soils with a wide range of physico-chemical properties.  104  3.1 Introduction Land-use change and land management practices have altered the earth\u2019s largest terrestrial carbon pool: soil carbon (Schmidt et al., 2011). For example, since the advent of agriculture, conversion of non-cultivated land has decreased global C stocks by approximately 116 Pg (Sanderman et al., 2017), amounting to a loss of approximately 5% of the current global terrestrial soil carbon stock (Zomer et al., 2017). There is a growing understanding that agricultural management practices that promote soil carbon sequestration provide benefits not only to crop productivity and increased nutrient cycling, but also long term sequestration of CO2 from the atmosphere (X. Bai et al., 2019; Z. Bai et al., 2018; Emde et al., 2021; Midwood et al., 2021). Agricultural soils have thereby become a rallying point for carbon farming initiatives (Cabuzel, 2021) as well as broader-reaching net-zero carbon commitments, such as Canada\u2019s Strengthened Climate Plan, which is aimed at exceeding Canada\u2019s 2030 Paris agreement reduction goals and achieving net-zero emissions by 2050 (Service Canada, 2020). Soil organic matter (SOM) can be separated into two broadly defined pools: particulate organic matter, and mineral-associated organic matter. Particulate organic matter more closely resembles the plant, animal, and fungal material it originated from and tends to be unbound in the soil matrix (Midwood et al., 2021; Six et al., 2002). Mineral-associated organic matter, by contrast, is typically comprised of small molecular weight, microbially processed compounds, and is sorbed to the surfaces of mineral particles in the soil (Cotrufo et al., 2019; Lavallee et al., 2020; Poeplau et al., 2018). Here, we refer to the carbon component of POM and MAOM as particulate organic carbon (POC) and mineral-associated organic carbon (MAOC), respectively. Given that it forms part of the unbound organic matter in the soil, POC is more vulnerable to agricultural management practices that cause disturbance to the soil (i.e. tillage), and is  105  generally believed to have a relatively quick turnover time (Midwood et al., 2020; Poeplau et al., 2018; Wander et al., 1998; Wander and Bidart, 2000), whereas MAOC, which is physically protected from microbial decay, can remain in soils for decades to centuries (Lavallee et al., 2020; Mikutta et al., 2019; Trumbore, 1997). Furthermore, MAOC is far more resistant to changes in decomposition rate as temperatures increase (Benbi et al., 2014). Therefore, from a climate change mitigation perspective, the benefits of increasing MAOC are twofold: first, residence times of MAOC are greater than POC, and second, the vulnerability of MAOC to rising global temperatures is lower than POC (Trumbore, 1997). A weakness of current global estimates of soil carbon sequestration capacity is that they are limited to total soil carbon, rather than the more stable MAOC fraction of the total soil carbon pool. Further, these estimates are often not at a sufficiently fine resolution because they estimate soil carbon stocks by land use type, rather than by soil type and, therefore, do not accommodate differences in soil characteristics that are believed to influence MAOC formation, such as texture, pH, or exchangeable cations (Beare et al., 2014; Cotrufo et al., 2013; McNally et al., 2017). In order to focus carbon sequestration efforts on the adoption of management practices that increase MAOC in soils, it is important to identify soils that have not yet reached maximum MAOC capacity.  While carbon stabilization methods have been discussed more frequently in recent years, there are very few studies that attempt to quantify the potential for additional carbon storage in specific soils.  To quantify any potential for additional MAOC storage, two measures must be determined for a given soil: the current concentration of MAOC in the soil, and the maximum achievable concentration of MAOC for the soil (i.e, its \u2018stabilization capacity\u2019).  106  The difference between these two values is then considered to be the \u2018stable carbon deficit\u2019 (Beare et al., 2014; McNally et al., 2017). There is a general consensus that MAOC formation is strongly influenced by the fine fraction of the soil (Hassink, 1997; McNally et al., 2017). Clay and silt particles account for the majority of the surface area available for sorption of MAOC and, therefore, clay and silt content have often been used to estimate current MAOC concentrations (Feng et al., 2013; McNally et al., 2017; Skjemstad et al., 2006). More recently, however, McNally et al. (2017) found that current MAOC was estimated more accurately by using the specific surface area (SSA) of the soil coupled with extractable aluminum concentrations rather than soil texture per se. The New Zealand-wide soils database used in McNally et al. (2017) included primarily acidic soils of volcanic origin containing fine-textured allophanes. By contrast, many agricultural soils in the Northern Hemisphere are of glacial origin and are often coarse-textured; furthermore, those in semi-arid regions tend to be more alkaline (Rasmussen et al., 2018). In acidic soils, aluminum- and iron-oxides stabilize soil carbon by forming organo-mineral complexes, whereas in alkaline soils, calcium forms divalent cation bridges between fine soil carbon and mineral surfaces (Kaiser and Guggenberger, 2000; Matus et al., 2006; Mikutta et al., 2007; Rowley et al., 2021). Differences in how MAOC is formed in soils with varying pH, texture, and mineralogy highlights the importance of taking a more mechanistic approach to MAOC estimation - one that can accommodate individual differences in soil parent material (Matus, 2021; McNally et al., 2017). In order to do so, development of any model to estimate MAOC must use predictor variables appropriate for the soil type in the region. Current concepts of carbon stabilization capacity revolve around the notion of mineral saturation by fine fraction carbon; that is, the point at which carbon is bound to  107  all available mineral surfaces such that no further binding of carbon is possible (Angers et al., 2011; Vogel et al., 2014). Statistical methods offer insights into the theoretical MAOC saturation threshold values, but the mechanisms behind these values provide conceptual challenges. For example, any values generated by MAOC saturation threshold models are necessarily affected by current MAOC values, which are, in turn, influenced by current soil conditions. As such, the resulting MAOC saturation threshold determined by these models does not necessarily represent \u201csaturation\u201d, but is instead a reflection of the challenges imposed by the current climatic, agronomic, and soil physicochemical conditions. If some combination of those conditions change, the theoretical MAOC saturation capacity is also likely to change over time. Further, it has been shown that OM preferentially binds to rough mineral surfaces and existing organo-mineral clusters, to the degree that < 19% of the visible mineral surface area of Luvisols during a 42 day incubation experiment was occupied by OM (Vogel et al., 2014).  It follows that the MAOC saturation threshold value generated by model estimates is therefore a MAOC formation capacity under the current soil conditions, rather than a saturation threshold, per se. Therefore, instead of the term \u201csaturation\u201d, we recommend adopting the nomenclature \u201cMAOC formation capacity\u201d, as we have here. This term makes no assumptions regarding the limits of mineral saturation, and more accurately represents the values estimated using carbon stabilization models. The capacity for a soil to form MAOC is difficult to determine with confidence. Several studies have used the mass proportion of fine soil particles (clay and fine silt) to predict MAOC formation capacity directly using a least squares regression model (Hassink, 1997; Six et al., 2002). However, Six et al. (2002) and Beare et al. (2014), have shown that this approach likely underestimates the MAOC formation capacity due to limitations in the application of least squares regression (Feng et al., 2013). Two such  108  limitations stand out in particular. First, there is an underlying assumption in these regression models that soil mineralogy has no effect on C stabilization (Feng et al., 2013; Six et al., 2002). Second, the maximum capacity regression line generated by such models relies on the assumption that all samples used in the model are near their maximum C stabilization capacity (Feng et al., 2013). In light of these issues, McNally et al. (2017) endeavoured to improve the statistical method by which the MAOC formation capacity of New Zealand\u2019s soils could be estimated. To account for differences in mineralogy, current MAOC was estimated using SSA and concentration of exchangeable aluminum in the soils, and, further, MAOC formation capacity was considered to be the 90th quantile of MAOC in the samples, as calculated using a quantile regression model (McNally et al., 2017). This method thereby uses the upper range of measured MAOC values to estimate the theoretically obtainable, upper threshold of MAOC formation for a soil.  The objective of the current study was to adapt the method developed by McNally et al. (2017) for estimating the MAOC deficit for fine textured, acidic soils of volcanic origin in New Zealand so that it is appropriate for estimating the MAOC deficit for predominantly coarse-textured, alkaline soils of glacial origin in the Okanagan Valley, British Columbia, Canada. To do this, we had to determine the best predictor variables to include in the model to estimate current MAOC, and then select an appropriate model algorithm. One of the most important questions we sought to answer was whether soil texture or SSA was a better predictor variable for these soils. We used previously published chemical and texture data from a recent soil survey conducted in orchards and vineyards in this region (Midwood et al., 2021, 2020), as well as new data we collected on the specific surface area of the same soils. The five soil groups included in this study represent a range of textures, meaning the results may be used widely. Recent work  109  conducted on a subset of these soil samples showed that the concentration of MAOC in the top 30 cm of soil has more than doubled over several decades of perennial cropping relative to adjacent non-cultivated soils, and has a current MAOC stock of 168.0 million kg across 8501 ha. (Midwood et al., 2021). However, such MAOC accrual cannot be expected to continue indefinitely. Here, we wanted to: i) determine if there is still a MAOC deficit present in these soils, given the large, relatively recent accumulation of MAOC, and ii) identify soil types with the greatest potential to store additional MAOC. Because direct measurement of MAOC is difficult and time consuming, we first needed to: (iii) determine the best modelling approach for estimating current MAOC concentrations using commonly measured soil characteristics, which required (iv) determining the most efficient method for estimating SSA. Hence, we generated candidate models for estimation of MAOC and then, using the best performing model, estimated missing MAOC values for our dataset. Next, we used the 90th quantile approach established by McNally et al. (2017) to estimate MAOC formation capacity, by depth increment, for each of the five soil groups.  3.2 Methods The multi-stage process by which MAOC deficits in the soils Okanagan Valley perennial cropping systems were determined involved the careful selection and assessment of both soil physicochemical variables and modelling approaches. This process can be roughly grouped into three overall stages: i) data acquisition and variable selection; ii) model selection for MAOC estimation; and iii) MAOC deficit estimation (Figure 1).   110  3.2.1 Data acquisition and variable selection 3.2.1.1 Okanagan soils database The Okanagan Valley resides along a 200 km-long temperature and precipitation gradient, with 30-year annual daily temperature and precipitation averages of 18.0 \u00b0C and 279.5 mm in the southern part of the Valley, and 15.0 \u00b0C and 383.5 mm in the north (Environment and Canada, 2019). The region is known for its cultivation of woody perennial fruit crops, including wine grapes, apples, and sweet cherries. Five groups of common soil types were selected, based on the surficial parent material from which they developed: i) clayey glaciolacustrine sediments, ii) sandy or silty glaciolacustrine sediments, iii) eolian veneer over morainal deposits, iv) gravelly fluvioglacial deposits, and v) sandy or silty fluvioglacial deposits (Figure 2). These soil groups represent approximately 40% (8501 ha) of the agricultural landbase in the Okanagan Valley and were deliberately selected to capture a range of soil textures from across the Valley, although most had sandy loam to loamy textures.  Figure 1: Outline of the MAOC deficit determination process. Light gray sections denote the data acquisition and variable selection stage, middle gray denotes the model selection for MAOC estimation process, and dark gray denotes the MAOC deficit calculation process.  111  Composite soil samples were taken at three depths (0-15 cm, 15-30 cm, and 30-60 cm) from the drive rows and crop rows of four woody perennial cropping systems: drip-irrigated apples and grapes, and micro-spray-irrigated apples and cherries in each of the five soil groups. Non-cultivated, adjacent areas under native vegetation (typically grassland and shrubland, eg. Ponderosa pine (Pinus ponderosa), sagebrush (Artemisia spp.), yellow rabbitbrush (Chrysothamnus viscidiflorus), bluebunch wheatgrass (Elymus spicatus)) with similar soil classification, elevation, climatic conditions, and aspect were also sampled (Midwood et al., 2021). Data from 537 soil samples collected from 99 sites were used for the current study (18 uncultivated sites x 3 depths; 81 cultivated sites x 3 depths for drive- and crop-row; three individual samples were inaccessible at the 30 \u2013 60 cm depth); the soil sampling and analysis protocol was outlined in detail in Midwood et al. (2020). Briefly, soil samples were collected from between 10 and 20 randomly chosen Figure 2: Soil texture composition of all 537 samples separated by the five soil groups used in this study. Soil groups are defined by their surficial deposit classifications. Modified from Midwood et al. (2020).  112  points over an approximate area of 0.25 to 0.5 ha and composited by depth increment. In agricultural sites, separate composite samples were collected, by depth, from the crop rows and the drive rows (strips of herbaceous vegetation between the crop rows). Composite samples were then sieved to 8 mm to remove large stones and roots, air-dried, and then further sieved to 2 mm prior to further analysis.  Each sample was analysed for soil texture (% sand, % silt, and % clay); exchangeable Ca (meq 100g-1 dry soil), Mg (meq 100g-1 dry soil), K (meq 100g-1 dry soil), and Na (meq 100g-1 dry soil); total N (%); soil organic and inorganic carbon (SOC, %; SIC, %); mean \u03b413C composition (\u2030); and soil pH. Details on the specific methods applied can be found in Midwood et al. (2021, 2020). Mineral-associated organic carbon (MAOC) was measured on a subset of 216 samples, chosen to represent a cross-section of all soil types, cropping systems, and soil depths sampled, such that the subset matched the soil property representation of the larger dataset as closely as possible. The soil fractionation process is outlined in greater detail in Midwood et al. (2021) and follows the method described in Poeplau et al. (2018) as \u2018Par+Den5\u2019. Briefly, this method uses a combination of flotation and repeated wet sieving to produce a series of mechanically distinct fractions. The resulting soil fractions were simplified, following the approach used by Cotrufo et al. (2019), into the two soil carbon fractions used in this analysis: MAOM (<50 \u03bcm fraction = silt + clay) and POM (>50 \u03bcm fraction). Here, MAOC refers to the C content measured from the <50 \u03bcm fraction. This fractionated subset (216 soil samples x 2 fractions = 432 data points) was used as training data to develop a model that estimated current MAOC concentrations using other soil variables, with the best performing model being used to fill in missing MAOC values for the remaining 321 soil samples for which MAOC was not directly measured.    113  3.2.1.2 Methods of estimating specific surface area Specific surface area (SSA) represents the surface area of minerals (Figure 3) available for binding small molecular weight, microbially processed carbon (Beare et al., 2014; McNally et al., 2017); consequently, we wanted to include SSA in the models to estimate MAOC, but these data had not previously been measured on the Okanagan soil samples. Specific surface area analyses of soils can be difficult and time consuming; therefore, we used three approaches for measuring SSA: i) direct measurement of SSA on a small number of samples using specialized equipment based on the Brunauer, Emmett, and Teller theory (BET); ii) interpolation of the remainder of the samples based on the BET results obtained in i); and iii) measurements of water mass loss after drying soils at 120 \u00b0C as per Beare et al. (2014), described below. The BET theory of gas adsorption to solid particles (Brunauer et al., 1938) is used to directly measure SSA. Because this technique requires specialized equipment, and is time consuming and expensive to conduct, we selected only 10 samples for direct Figure 3: Comparison of SSAproxy (measured using moisture loss as in Equation 2) and soil texture variables for 537 samples collected from orchards, vineyards, and under native vegetation. Each soil sample has three points on the figure: one for Clay % vs. SSAproxy (p < 0.001; R2: 0.74), a second for Silt % vs. SSAproxy (p < 0.001; R2: 0.14), and a third for Sand % vs. SSAproxy (p < 0.001; R2: 0.53).  114  SSA measurement. The samples were selected across the full textural range of our entire 537 sample dataset, and were analysed using a TriStar II Plus, surface area and porosity analyser (Micrometrics Instrument Corp, Norcross, GA, USA). We then used a simple trend line of BET (i.e., actual SSA) to interpolate values between measured points to determine SSAinterpolated for the remaining 527 samples when ranked by moisture loss after oven drying (see below). As a less expensive alternative to the BET method for measuring SSA, the mass of water lost when air-dried soils are oven dried can be used as a reasonable approximation of BET (Beare et al., 2014). We assessed the utility of this method using sieved (2 mm), air-dried soil samples held at 30 \u00b0C and 30% humidity for at least one week in a controlled environment chamber (Conviron PGCFLEX model; Controlled, Environments Ltd., Winnipeg, Canada), then oven dried at 120 \u00b0C to a constant mass. We used the difference between the air-dried and oven-dried mass of the soil samples to estimate the mineral surface area available for binding carbon in soils, as was done for Allophanic, Brown, and Recent soils in New Zealand, using the following equation (Parfitt et al., 2001): \ud835\udc46\ud835\udc46\ud835\udc34\u0be3\u0be5\u0be2\u0beb\u0bec (\ud835\udc5a \u0b36 \ud835\udc54\u0b3f\u0b35)  =  2 \ud835\udc65 \ud835\udc4e\ud835\udc56\ud835\udc5f \u2212 \ud835\udc51\ud835\udc5f\ud835\udc66 \ud835\udc64\ud835\udc4e\ud835\udc61\ud835\udc52\ud835\udc5f \ud835\udc50\ud835\udc5c\ud835\udc5b\ud835\udc61\ud835\udc52\ud835\udc5b\ud835\udc61 (\ud835\udc54 \ud835\udc58\ud835\udc54\u0b3f\u0b35) (1) We then plotted the SSAproxy values against our measured and interpolated values of SSA (which relied on the BET method) and found that equation 1 greatly overestimated the SSA of the 432 soil fractions in the training set. We therefore calculated a new equation to estimate SSA using the \u2018moisture loss\u2019 method for soils in the Okanagan Valley as follows:  115  \ud835\udc46\ud835\udc46\ud835\udc34\u0be3\u0be5\u0be2\u0beb\u0bec (\ud835\udc5a \u0b36 \ud835\udc54\u0b3f\u0b35)  =  0.61 \ud835\udc65 \ud835\udc4e\ud835\udc56\ud835\udc5f \u2212 \ud835\udc51\ud835\udc5f\ud835\udc66 \ud835\udc64\ud835\udc4e\ud835\udc61\ud835\udc52\ud835\udc5f \ud835\udc50\ud835\udc5c\ud835\udc5b\ud835\udc61\ud835\udc52\ud835\udc5b\ud835\udc61 (\ud835\udc54 \ud835\udc58\ud835\udc54\u0b3f\u0b35) (2)  3.2.1.3 Variable selection In order to determine the strongest predictors of MAOC concentrations across soil types in the Okanagan Valley soils, we used Pearson correlation analyses to examine the relationships among measured soil properties (MAOC; SOC; SSAproxy; POC; % sand, % silt, and % clay; exchangeable Ca, Mg, K, and Na; total N; SIC; mean \u03b413C composition; C:N ratio; and soil pH) in order to determine the strongest predictors of MAOC, POC, and SOC concentrations (Supplementary Materials, Figure S1). While POC and SOC are not discussed at length in this study, correlation analyses were carried out to compare differences in drivers of soil C storage among each carbon fraction. Based on the results of correlation analysis, we parsed the variables down to SOC concentration, SSAproxy, exchangeable Ca, exchangeable K, % clay, % silt, and % sand, %N, C:N ratio, and soil pH (Figure 4). Specific surface area proxy and soil texture variables (% clay, % silt, and % sand) were included as indicators of the mineral area available for soil C binding, while exchangeable Ca was selected due to its strong correlation with MAOC in the more neutral soils of the Okanagan Valley (Rasmussen et al., 2018). While soil pH did not correlate well with current MAOC in our samples, it has previously been identified as an important factor in MAOC storage dynamics in the literature that formed the basis for our modelling approach (Beare et al., 2014) and variable selection process (Rasmussen et al., 2018). Exchangeable K, soil C, soil N, and C:N ratio were similarly  116  identified as strong predictors of MAOC stocks in European soils using Random Forest models analogous to those used in this study (Cotrufo et al., 2019).  3.2.2 Model selection for MAOC estimation Two model algorithms were chosen for estimating current MAOC concentrations using the selected variables: stepwise multiple regression with AIC, because of its predominance in generating pedotransfer functions (Beare et al., 2014); and random forest, a relatively new, non-parametric machine learning approach with the potential to produce more accurate MAOC estimates (Cotrufo et al., 2019). Given the close relationship between SSAproxy and soil texture found in the literature, we generated two model iterations using either SSAproxy or soil texture in addition to all of the selected variables mentioned above, for each of the two algorithms: i) stepwise multiple Figure 4: Correlation matrix of soil characteristics measured on the 216 measured samples included in the models and for discussion. Numbers in each cell are the Pearson correlation coefficient between the two variables. Shade intensity is representative of the strength of the correlation, and hashed boxes represent negative correlations. Relationships in white cells with italicized correlation coefficients were not statistically significant (p value > 0.05). SSA interpolated values are specific surface area values calculated using the Brunauer, Emmet and Teller (BET) theory and values interpolated using a best fit line. Mineral associated organic carbon (MAOC) is the C concentration in the < 50 \u00b5m fraction, while Particulate organic carbon (POC) is the C concentration in the > 50 \u00b5m fraction. Soil organic carbon (SOC) is the total organic carbon present in the soil sample. Specific surface area proxy (SSAp;roxy) is the mineral surface area as measured using the moisture loss method as in Equation 2.  117  regression with AIC using SSAproxy values but no % clay and % silt; ii) stepwise multiple regression with AIC using % clay and % silt but no SSAproxy; iii) random forest algorithm with SSAproxy but no % clay and % silt; and iv) random forest algorithm with % clay and % silt but no SSAproxy.   3.2.2.1 Stepwise multiple regression with AIC The data used in the analysis (SOC, exchangeable Ca and K, SSAproxy, % clay, % silt, % sand, % N, soil pH, and C:N ratio) were largely skewed to the right, suggesting that there were few extremely high values in the samples (Ngailo and Vieira, 2012). To normalize the data for use with stepwise multiple regression, values were log-transformed (natural log) prior to applying stepwise multiple regression. Akaike\u2019s Information Criterion was used to select the best-fit model (McNally et al., 2017; Munro et al., 2020). Stepwise multiple regression with AIC was carried out using the stepAIC package (Venables and Ripley, 2002) in R version 4.1.1 (R Core Team, 2020).  3.2.2.2 Random Forest modelling The same data (SOC, exchangeable Ca and K, SSAproxy, % clay, % silt, % sand, % N, soil pH, and C:N ratio) were used to estimate current MAOC via random forest modelling. Random forest models work by combining a large number of regression trees, trained using bootstrap aggregation, to build a robust predictive model that is resistant to noise in the data (Breiman, 2001). The randomForest base package was used alongside caret for hyperparameter tuning (Kuhn, 2020; R Core Team, 2020). Model fit was determined from the fitted models using the \u201c% var explained'' output parameter.  118   3.2.2.3 Model performance The performance of the models developed to estimate MAOC concentrations were evaluated based on the respective model performance metrics, carried out using k-fold cross validation, for each of the model types (e.g. R2 for stepwise multiple regression with AIC, and \u2018% var explained\u2019 for Random Forest). Linear models that compared the estimated MAOC values from each of the four estimation models with the MAOC values measured on the 216-sample subset highlighted the difference in relative performance across the range of MAOC values (Figure 5). The model with the highest model performance metric was selected to estimate the current MAOC concentrations for the 321 samples for which MAOC concentrations had not been directly measured.  3.2.3 Mineral-associated organic carbon deficit estimation We estimated MAOC deficits as the difference between current MAOC concentrations and the MAOC formation capacity of each soil sample. Current MAOC concentrations had either been measured directly (216 samples) or estimated as described in 2.2 (321 samples). We used the multivariate quantile regression method described by McNally et al. (2017) to estimate the MAOC formation capacity. According to this method, the 90th quantile of current MAOC concentrations was designated as the formation capacity for the entire population of soil samples; 90th quantiles of current MAOC concentration were calculated separately for each of the five soil groups included in the dataset. The carbon  119  deficits were then calculated as the difference between the current measured (or estimated) value of MAOC concentration for each individual sample and the mean 90th quantile of MAOC for all samples within each soil group. Confidence intervals (95%) were calculated using a root mean square approach (McNally et al., 2017).  MAOC deficits were subsequently compared among soil groups and depths using a two-factor ANOVA with means separation using Tukey\u2019s Honestly Significant Difference (Tukey\u2019s HSD) test.  Figure 5: Values of MAOC from predictive models using SSAproxy as the mineral surface area availability metric (plots A and B), or texture (%clay and %silt) as the mineral surface area availability metric (plots C and D) plotted against measured MAOC values (C concentration in the < 50 \u00b5m fraction) on the same 216 samples. Plot A and C use Random Forest predictive algorithms (Plot A \u2018% variance explained\u2019: 78.17; Plot C \u2018% variance explained\u2019: 77.8) while plots B and D are from Stepwise Multiple Regression with AIC algorithms (Plot B R2: 0.713; Plot D R2: 0.717). The black line shows a 1:1 relationship.  120  3.2.4 Calculating MAOC stocks and stock deficits MAOC stocks (Mg ha-1) were calculated from MAOC concentrations (g kg-1) using an adjusted soil bulk density value (BD). This adjusted BD value was determined by correcting the whole soil bulk density (Midwood et al., 2021) for the fine fraction portions of the soil in which the MAOC is found. Silt (%) and clay (%) accounted for an average of 27% of the bulk soil mass over all measured depths (Midwood et al. 2021). The adjusted BD and the resultant MAOC stock values were therefore calculated as follows: \ud835\udc35\ud835\udc37\u0bd4\u0bd7\u0bdd\u0be8\u0be6\u0be7\u0bd8\u0bd7 = \ud835\udc35\ud835\udc37\u0bea\u0bdb\u0be2\u0bdf\u0bd8 \u0be6\u0be2\u0bdc\u0bdf  \u00d7 0.27 (3) \ud835\udc40\ud835\udc34\ud835\udc42\ud835\udc36 \u0be6\u0be7\u0be2\u0bd6\u0bde (\ud835\udc40\ud835\udc54 \ud835\udc40\ud835\udc34\ud835\udc42\ud835\udc36 \u210e\ud835\udc4e\u0b3f\u0b35)  = \ud835\udc40\ud835\udc34\ud835\udc42\ud835\udc36\u0bd6\u0be2\u0be1\u0bd6.  \u00d7 \ud835\udc35\ud835\udc37\u0bd4\u0bd7\u0bdd\u0be8\u0be6\u0be7\u0bd8\u0bd7  \u00d7 \ud835\udc61 \u00d7 0.1 (4) where \ud835\udc40\ud835\udc34\ud835\udc42\ud835\udc36\u0bd6\u0be2\u0be1\u0bd6. is mineral-associated organic carbon concentration in g kg-1; \ud835\udc35\ud835\udc37\u0bea\u0bdb\u0be2\u0bdf\u0bd8 \u0be6\u0be2\u0bdc\u0bdf is the measured, whole soil bulk density, in g cm-3; \ud835\udc35\ud835\udc37\u0bd4\u0bd7\u0bdd\u0be8\u0be6\u0be7\u0bd8\u0bd7 is the bulk density adjusted for the silt and clay soil fraction, in g cm-3; \ud835\udc61 is the thickness of the depth increment (cm), and 0.1 is the conversion factor for Mg ha-1 .  As with MAOC concentrations, MAOC stock (Mg ha-1) formation capacities were calculated using the 90th quantile method, but applied to values of MAOC stock. The MAOC stock deficit was calculated on a per-sample basis as the difference between the MAOC stock upper limit for each soil type, and the current MAOM stock.   121  3.3 Results 3.3.1 Carbon fractions and soil properties The first step in developing a model to estimate current MAOC concentrations was to determine the most appropriate predictor variables for the soil samples in our dataset. We conducted correlation analyses among the measured MAOC concentrations (determined for 216 of the 537 soil samples collected for the aforementioned Okanagan soil survey) and other variables measured on the same soil samples (see Section 2.2). Correlation analyses were also conducted for POC and SOC using the same dataset to compare differences in the drivers of soil C storage among each carbon fraction.  3.3.1.1 Correlations between measured MAOC and other soil variables Mineral-associated organic carbon showed the strongest positive correlations with concentrations of SOC and exchangeable Ca in the whole (unfractionated) soil (Figure 4). Mineral-associated organic carbon also correlated positively with those factors most associated with increased SSA (i.e., SSAinterpolated, SSAproxy, % clay, and % silt) and with exchangeable K and % nitrogen. Mineral-associated organic carbon was negatively correlated with % sand, and was not correlated with soil pH nor C:N ratio.  3.3.1.2 Correlations between measured POC and other soil variables Storage of soil carbon as MAOC and POC were driven by different soil properties (Figure 4). Other than positive correlations with concentrations of SOC, % N, exchangeable Ca, and SSAproxy in the whole (unfractionated) soil, POC was not  122  positively correlated with the same variables as MAOC. Instead, POC was negatively correlated with soil pH and % clay and showed no relationship with exchangeable K, SSAinterpolated, % silt, % sand, or C:N ratio.  3.3.1.3 Correlations between measured SOC and other soil variables The concentration of SOC in the whole soil was most strongly positively correlated with exchangeable Ca, SSAproxy, and % N, with weaker (but still positive) correlations with SSAinterpolated and % silt. Soil organic carbon was negatively correlated with % sand and soil pH, and was not related to exchangeable K, % clay, and C:N ratio (Figure 4).  3.3.2 Predicting current MAOC: model comparison  Following the correlation analyses, the variables shown in Figure 4 were then used to generate the four model types outlined in section Section 2.2, that is: i) stepwise multiple regression with AIC using SSAproxy values but no % clay and % silt; ii) stepwise multiple regression with AIC using % clay and % silt but no SSAproxy; iii) random forest algorithm with SSAproxy but no % clay and % silt; and iv) random forest algorithm with % clay and % silt but no SSAproxy. Based on comparisons of values of MAOC concentration predicted by the model versus measured values of MAOC (C concentration in the < 50 \u00b5m fraction; Section 2.1), random forest (Figure 5 A, C) performed better than stepwise multiple regression with AIC (Figure 5 B, D) for estimating MAOC concentrations (Table 1). The random forest model containing SSAproxy as an input variable explained the most variance (78.2%), followed very closely by the random forest containing soil texture as an input variable (77.8%). Variable importance plots identified total SOC, exchangeable  123  Ca, and SSAproxy as the most impactful variables for estimating MAOC (Supplementary Materials S2). Like the random forest models, stepwise multiple regression with AIC containing either SSAproxy or soil texture as input variables produced similar results (Figure 5), with 71.7% of the variance explained using soil texture and 71.3% explained using SSAproxy data (Table 1). For the model generated using SSAproxy data, variables selected by the best-fit AIC were total SOC, exchangeable Ca, SSAproxy, and soil pH. For the model using soil texture data, variables selected by the best-fit AIC were total SOC, exchangeable Ca, % clay, and % silt. Notably, exchangeable K, total N, and C:N ratio were not selected for either model. While the stepwise multiple regression models showed a greater spread in predicted MAOC values than the random forest models, these models have the advantage of providing coefficients for a simple pedotransfer function. The pedotransfer functions for calculating MAOC, when SOC, exchangeable   Table 1: Model performance metrics for MAOC estimation. SSAproxy is specific surface area determined using the moisture loss method.  Plot Model Variation Model Performance p value A2 Random Forest - SSAproxy % var. explained: 78.2% NA1 B3 Stepwise Multiple Regression with AIC - SSAproxy R2: 0.713 < 2.2e-16 C2 Random Forest - texture % var. explained: 77.8% NA1 D3 Stepwise Multiple Regression with AIC - texture R2: 0.717 < 2.2e-16 1 Random Forest models do not produce a p value or equivalent metric. 2 Variables included in Random Forest model A were SSAproxy, SOC, exchangeable Ca, exchangeable K, % N, soil pH, and C:N ratio. Variables included in Random Forest model C soil texture (% clay and % silt), SOC, exchangeable Ca, exchangeable K, % N, soil pH, and C:N ratio. 3 Variables selected by AIC for model B were SSAproxy, total SOC, exchangeable Ca, and soil pH. Variables selected by AIC for model D were % clay, % silt, total SOC, and exchangeable Ca.  124  Ca, clay and silt concentrations, SSA, and soil pH are known, were produced from the best-fit stepwise multiple regression with AIC for both soil texture (Equation 5) variables and SSAproxy (Equation 6), and are shown in Table 2.   3.3.3 Mineral-associated organic carbon formation capacity Current concentrations of MAOC, as estimated using the random forest model with SSAproxy, were highest in soil samples taken from the top 15 cm of drive row soils in all cropping systems (Supplementary Materials Table S1). Soils with surficial deposits comprised of clayey glaciolacustrine sediments had the highest capacity for MAOC formation as calculated using the 90th quantile approach (9.9 g kg-1), while those of sandy fluvioglacial deposits had the lowest (5.3 g kg-1) (vertical lines, Figure 6). The remaining three soil groups had MAOC formation capacities of 6.3 g kg-1, 7.1 g kg-1, and 9.0 g kg-1 for soils with surficial deposits comprised of silty glaciolacustrine sediment, morainal deposit, and gravelly fluvioglacial deposit soils, respectively (Figure 6).  Table 2: Pedotransfer functions produced from best-fit stepwise multiple regression with AIC models using either soil texture variables (% clay and silt; Equation 5) or SSAproxy (Equation 6). MAOC and SOC are measured in g C kg-1, Ca is exchangeable calcium measured in meq 100 g-1 of dry soil, SSAproxy is specific surface area determined using the moisture loss method, pH is soil pH, and clay and silt are measured in percent soil texture composition.  Pedotransfer Function  \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc40\ud835\udc34\ud835\udc42\ud835\udc36 \u0bd6\u0be2\u0be1\u0bd6.)  =  0.719 +  0.627. \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc46\ud835\udc42\ud835\udc36 \u0bd6\u0be2\u0be1\u0bd6.)  +  0.134. \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc36\ud835\udc4e)  +  0.013. \ud835\udc50\ud835\udc59\ud835\udc4e\ud835\udc66 + 0.004. \ud835\udc60\ud835\udc56\ud835\udc59\ud835\udc61 (5) \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc40\ud835\udc34\ud835\udc42\ud835\udc36 \u0bd6\u0be2\u0be1\u0bd6.)  = \u22120.460 +  0.483. \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc46\ud835\udc42\ud835\udc36 \u0bd6\u0be2\u0be1\u0bd6.)  +  0.164. \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc36\ud835\udc4e)  + 0.470. \ud835\udc59\ud835\udc5c\ud835\udc54(\ud835\udc46\ud835\udc46\ud835\udc34\u0be3\u0be5\u0be2\u0beb\u0bec)  +  0.046. \ud835\udc5d\ud835\udc3b (6)  125  3.3.4 Deficits in MAOC: concentrations Deficits in MAOC concentration were present in all soil groups (p < 0.001)), and increased with depth (p < 0.001; Table S2, Figure 6, Figure 7). In general, more clay rich soils, e.g., soils derived from clayey glaciolacustrine sediment, showed greater MAOC deficits ( 1.62 g kg-1 for 0-15 cm, 4.01 g kg-1 for 15-30 cm, and 5.80 g kg-1 for 30-60 cm), while sandier soils, e.g., soils derived from sandy fluvioglacial deposits, showed smaller deficits (1.01  g kg-1 for 0-15 cm, 2.72 g kg-1 for 15-30 cm, and 3.69 g kg-1 for 30-60 cm (Table 3)).   Table 3: Difference in whole profile MAOC values between soil groups. Significant differences between soil group means, as determined using ANOVA with means separation using Tukey\u2019s Honestly Significant Difference (Tukey\u2019s HSD) test are noted next to each difference (\u2018.\u2019 : p <= 0.1; \u2018*\u2019 : p <= 0.05; \u2018**\u2019 : p <= 0.01; \u2018***\u2019 : p <= 0.001).  Soil type Comparison Difference (mg C g-1 soil) p value MD3 - CGS1 0.20158 0.9643 MD3 - SFD5 1.550131 *** 0.00000  MD3 - SGS2 1.309291 *** 2.00E-04 MD3 - GFD4 -0.37832 0.7482 CGS1 - SFD5 1.348551 *** 2.00E-04 CGS1 - SGS2 1.107711 ** 0.0033 CGS1 - GFD4 -0.5799 0.3599 SFD5 - SGS2 -0.24084 0.9423 SFD5 - GFD4 -1.92845 *** 0.00000 SGS2 - GFD4 -1.68761 *** 0.00000 1 Clayey Glaciolacustrine Sediments 2 Silty Glaciolacustrine Sediments 3 Morainal Deposits 4 Gravelly Fluvioglacial Deposits 5 Sandy Fluvioglacial Deposits   126    Figure 6:  Current MAOC concentration, as determined by MAOC modelling in combination with measured MAOC where available, and MAOC formation capacity as determined using a quantile regression (\u03c4 = 0.90) model, separated by soil group and soil depth. Black dots represent the mean MAOC for each soil group and depth, with the number of samples in each group shown above. Bars represent the standard error of the mean. The vertical grey dashed line is the MAOC formation capacity.  127   3.3.5 Deficits in MAOC: stocks Bulk density measurements were obtained as per Midwood et al. (2021) for only the 0-15 and 15-30 cm depths. Therefore, MAOC stocks (Mg ha-1) and their resulting deficits were calculated only for the upper 30 cm of the soil profiles. Across all soil groups, the current MAOC stock was 11.34 Mg ha-1 on average, for the 0-15 cm depth, and 8.43 Mg ha-1 on average, for the 15-30  cm depth. Across all soil groups, deficits in MAOC stocks were an estimated 1.13 Mg ha-1 for the 0-15 cm depth, and an estimated 2.22 Mg ha-1 for the 15-30 cm depth. When considered across the 8501 ha of agricultural land represented by these soil groups across the Okanagan Valley, there is currently a MAOC stock of 168 065 Mg C in the upper 30 cm of the soil profile, and a MAOC stock deficit of 28,478 Mg C. Figure 7: Mineral-associated organic carbon deficits by soil group and depth, calculated as the difference between the MAOC formation capacity estimated using a 90th quantile approach and the current MAOC. Boxes show the 25th to 75th interquartile range, with the median shown as a black bar. Whiskers show the range of data present for each box. Significant differences between samples from different depths within a soil type, determined using ANOVA with means separation using Tukey\u2019s Honestly Significant Difference (Tukey\u2019s HSD) test are shown above each soil depth combination (\u2018.\u2019 : p <= 0.1; \u2019*\u2019 : p <= 0.05; \u2018**\u2019 : p <= 0.01; \u2018***\u2019 : p <= 0.001)..  128  3.4 Discussion 3.4.1 Variable selection Soil carbon modelling of the whole soil (total SOC) may be less effective than modelling mechanistically distinct soil carbon fractions. For example, our results showed that not only are there often different drivers of total SOC, POM, and MAOC, but that % clay, a well-established factor affecting soil carbon storage, is correlated in the opposite direction for MAOC and POM (Figure 4). For example, % clay in the samples used for this study had a significant positive correlation with MAOC, and a significant negative correlation with POC, such that the overall effect of % clay on total SOC was nearly neutral and not significant. As such, it isn\u2019t possible to make any determinations regarding the mechanisms of SOC retention by looking at the correlations between total SOC and these such soil properties. While POC is an important indicator of soil health, from a climate change mitigation standpoint, MAOC offers greater benefits in the form of long-term carbon storage in soils. One of the aims of this study was to provide a framework for estimating MAOC stocks and predicting MAOC deficits that could be applied to soils with a broad range of physico-chemical properties. Selecting variables for inclusion in current MAOC models should be done on a per-soil basis, as soil properties often interact to form the conditions necessary for MAOC storage. For example, the Okanagan soils used in this study developed from glacial outwash and debris left behind by retreating glaciers approximately 10,000 years ago and contain few pedogenic clays, but rather clay-sized primary minerals (Fulton, 1969; Ross et al., 1985). While arable soils across the globe typically have > 50% of their SOC in the form of MAOC (Christensen, 1992; Curtin et al., 2019), soils in Okanagan Valley orchards and vineyards have relatively low MAOC,  129  averaging only 27% of SOC across all plots. This difference may be attributed to the semi-arid climate; low productivity of native plants; and relatively young, therefore lightly weathered, soils with less absorptive surface area in the clay-sized fraction (Midwood et al., 2021; Ross et al., 1985). The mineral weathering process is responsible for the development of reactive (clay) surfaces that bind and help protect MAOC from microbial decay, thereby vastly increasing the residence time of soil C (Trumbore, 2009). It follows that soils with different combinations of soil genesis, soil age, climate, and native vegetation will result in unique physicochemical conditions. While % clay and SSA are both related to the mineral surface area potentially available for adsorption of fine fraction OM, models utilizing either soil texture or SSA tend to have varying accuracy in different soil types. Nano-scale analyses of OM surfaces; iron- and aluminum-oxides, and exchangeable calcium; as well as mineral surfaces, have highlighted the availability of, and mechanisms behind, mineral adsorption sites (Boiteau et al., 2020; Gerin et al., 2003; Woche et al., 2017; Yuan et al., 1998). In acidic soils, aluminum- and iron-oxides play an important role in MAOC formation by forming organo-metal complexes with low molecular weight OM, while in alkaline soils, exchangeable calcium serves a similar function by forming divalent cation bridges (Kaiser and Guggenberger, 2000; Mikutta et al., 2007; Rowley et al., 2021). This results in uneven adsorption of OM to mineral surfaces, and may partially account for differences in the efficacy of using soil texture versus SSA in previous fine fraction C deficit modelling. Clay (%) alone cannot account for unevenness in the binding capacity of clays, while SSA can at least partially account for differences in clay mineralogy (Boiteau et al., 2020; Lehmann et al., 2007). In fact, while utilization of the SSAproxy (air-dry vs. oven-dry weight) method used here may, strictly speaking, give a less accurate measure of overall fine fraction surface area than direct measurements of SSA obtained  130  via the Brunauer\u2013Emmett\u2013Teller (BET) isotherm, the SSAproxy may more accurately account for the increased water holding capacity of the adsorbed OM (Woche et al., 2017). If this is the case, then MAOC prediction models for soils with predominantly 1:1 clays, low overall clay content, or low MAOC will likely find that soil texture variables like % clay and % silt serve as well as SSA in predicting MAOC, as we have here. Because Okanagan Valley soils tend to be more neutral or alkaline, exchangeable Ca plays an important role in MAOC formation; the relationship between MAOC and exchangeable Ca in these soils showed a correlation coefficient of 0.71, which is equivalent to that between MAOC and overall total SOC (Figure 4). Variable importance ranking also showed that exchangeable Ca was the second-most important variable for estimating MAOC in the random forest models (second only to total SOC (Supplemental Materials S2)). In addition to cation bridging (Kalinichev and Kirkpatrick, 2007; Pennock et al., 2011; Sutton et al., 2005), exchangeable calcium plays an important role in SOC accumulation in neutral and alkaline soils via entrapment of occluded SOC in Ca-promoted aggregates (Boiteau et al., 2020; Mart\u00ed-Roura et al., 2019; Rowley et al., 2021). Calcium and low molecular weight organic C molecules work with clay minerals to form organo-clay complexes, by which carbon is bound to the edges and surface of clay minerals (Pennock et al., 2011). In addition to the stabilization of carbon due to divalent cation bridging via Ca2+ (Rowley et al., 2018), the prevalence of micropores in clay-rich soils physically protects trapped carbon from microbial interaction (Pennock et al., 2011). The result is that as exchangeable Ca levels increase in soil, MAOC also increases, even if soil conditions favour oxidation of SOM (Rowley et al., 2018).   131  3.4.2 Prediction of current MAOC using stepwise and random forest models Measures of SSA can be employed in association with other soil physico-chemical properties (such as exchangeable Ca and soil pH) to accurately estimate current MAOC concentrations in situations where it is not practical or affordable to measure MAOC directly (Balesdent et al., 1998; Beare et al., 2014; Hassink, 1997; 2017). However, the coefficient used to estimate SSA in Beare et al. (2014) grossly overestimated the SSA of the younger, coarser-textured soils used here, as measured using BET. We therefore adopted a smaller coefficient, which more closely represented the SSAinterpolated values obtained for these soils (Equation 2). In any case, the patterns of change in SSAproxy and BET-derived SSAinterpolated values were very similar (Supplementary Materials Figure S3). It follows that in order to accurately approximate the SSA for a given soil using the SSAproxy approach, comparisons with measured SSA values may need to be considered (as we did using the BET method), and the coefficient may need to be individually assessed for soils with divergent physicochemical properties. That being said, when using the resulting calculated SSAproxy value for MAOC estimation, the pattern of change and relative SSAproxy value are what drove model predictions, particularly with non-parametric model types such as Random Forest. As such, the SSAproxy approach to SSA estimation remains appropriate for the purposes of modelling soil parameters impacted by changes in the specific surface area of soils. A series of four models were created to compare the efficacy of soil texture and SSAproxy to estimate MAOC in perennial cropping systems of the Okanagan Valley, British Columbia, Canada (Figure 5). The two random forest models (one using SSAproxy and one using soil texture) performed equally well (with approximately 78% variance explained); the two stepwise multiple regression models (again, one using SSAproxy and  132  one using soil texture) also performed similarly (SSA with R2 of 0.71 and soil texture with R2 of 0.72). These results are in line with previous studies that concluded that the mass proportion of the fine fraction is the most quantitatively important factor in controlling the storage of C in the MAOC fraction (Angers et al., 2011; Feng et al., 2013; Hassink, 1997). However, these results contrast with McNally et al. (2017), which concluded that SSAproxy was a much better predictor of MAOC than soil texture. This difference may be due, in part, to factors related to soil genesis and clay characteristics discussed in Section 4.1.  3.4.3 Mineral-associated organic carbon formation capacity In order to estimate MAOC deficits, both current MAOC and MAOC formation capacity must be known or estimated.  While it is possible to experimentally examine the mechanisms behind MAOC storage in soils and to obtain accurate estimates of current MAOC values, it is difficult to determine the mechanisms and values behind the absolute capacity of a soil to stabilize carbon.  Current \u2018saturation\u2019 concepts of MAOC storage revolve around the notion that maximum MAOC storage is achieved when there is no more surface area available to bind small molecular weight, microbially processed, carbon (Angers et al., 2011; Vogel et al., 2014). However, it has been shown that OM preferentially binds not only to the edges and rougher surfaces of mineral particles, but also to existing MAOC clusters. As such, only a limited portion of clay particles participate in MAOC formation, and a majority of the available mineral surface area is left bare (Vogel et al., 2014). While availability of mineral surface area is undoubtedly an important factor in the formation of MAOC, it is only one of a number of factors, such as current environmental conditions,  133  agronomic practices, and soil physicochemical conditions, which potentially limit the MAOC formation capacity.  3.4.4 Mineral-associated organic carbon deficits in Okanagan Valley soils All soil groups in our Okanagan Valley dataset showed mean concentrations of MAOC below the theoretical upper storage limit. Soils with greater % clay, and at greater depths, tended to have greater MAOC deficits. Our models show that MAOC storage capacity increases with increasing clay content (Figure 6). Further, clay content tends to increase with depth (Figure 2), likely due to downward translocation of clay particles as irrigation and rainwater move through the soil profile (Emde et al., 2021; Puy et al., 2017; Xu et al., 2016). There also tends to be more MAOC at the surface, due to overall higher SOM via deposition of above ground plant litter, rhizodeposition and root turnover, and organic amendments (Kaiser and Kalbitz, 2012; Midwood et al., 2020). In particular, the surface soils of the drive row spaces in cherry orchards irrigated with micro-sprinkler irrigation had the highest overall MAOC of all of our cropping systems. This is likely due to the increased productivity of incidentally irrigated drive row vegetation, and the large amount of detritus resulting from tree trimming, grass mowing, and leaf litter left on the undisturbed soils of the drive row spaces (Midwood et al., 2020). The resulting MAOC deficit, therefore, is likely underestimated at the soil surface because there is an underlying assumption that some soils have achieved the maximum possible concentration of MAOC, and even further underestimated at depth, where inputs of plant litter and roots are relatively low while concentrations of clay, due to the downward translocation of clay particles, is relatively high.  134   While perennial cropping systems in the Okanagan valley already have significantly elevated soil C when compared to adjacent, non-cultivated areas under natural vegetation (Midwood et al., 2021), this research shows that there is still room to store additional soil C as MAOC. Therefore, efforts made towards increasing MAOC in these soils will not go to waste. Hereafter, the challenge is in further optimising the agricultural management practices already in place to promote further accumulation of MAOC, while not impacting the productivity of the orchards and vineyards. To this end, the drive-row spaces (the unmanaged spaces between crop rows) offer a promising focus. Passive OM inputs from pruning litter, and increased groundcover productivity due to over-spray from crop irrigation sources has inadvertently created spaces where soil C, including MAOC, increases at a rate beyond that of the managed crop rows (Midwood et al., 2020). Actively managing these spaces to further enhance soil C input has the potential to not only increase site-wide MAOC stocks but also positively affect soil characteristics related to soil health and good soil hydrology. Further, by limiting herbicide use on the crop rows, and instead using soil amendments (i.e., organic mulches) or cover crops to limit competitive weed growth, there is the potential to further increase MAOC input, and enhance soil moisture retention in the crop row. Improved water-use efficiency is a critical concern with this semi-arid region, where irrigation is required to maintain crop productivity, and the availability of water is limited. Thus, such changes could simultaneously enhance irrigation water-use efficiency, increase MAOC stocks, and improve agricultural productivity (Midwood et al., 2021, 2020).   135  3.4.5 Limitations of 90th quantile deficit modelling and specific surface area estimations Despite having coarser parent materials, soils derived from gravelly fluvioglacial deposits showed MAOC deficits comparable to the most clay heavy soils in our dataset. This runs counter to our expectations and highlights both the strength and limitations of determining the MAOC formation capacity using real soil values.  While using real soil measurements to model MAOC formation capacity provides confidence that the formation capacity estimates are within attainable limits, they also carry the potential for anomalies. For example, analyses were done on the fine fraction of the soil and so after the > 2mm soil particles were removed from the gravelly fluvioglacial deposit derived soils prior analysis, the fine soil fraction (% clay and silt) comprised a greater relative portion of the overall soil texture. The current MAOC concentrations of these soils were additionally quite variable, with a few samples that were very high in MAOC. Consequently, the estimated MAOC formation capacity for this soil group was unexpectedly high. Given the middling current MAOC concentrations for most samples of this gravelly soil, the resultant MAOC deficit is disproportionately large.  Although the MAOC formation capacity of soils derived from gravelly fluvioglacial deposits would likely remain high as a result of the application of the 90th quantile method on highly variable data, if these results were to be scaled up to MAOC stocks using bulk density values adjusted for the coarse fragments, it is likely that soils derived from gravelly fluvioglacial deposits would show current MAOC stocks, formation capacities, and deficits more in line with what is expected from coarse soils. .   136  3.5 Conclusion As the world\u2019s largest terrestrial carbon pool, and one we manage daily for our own food and other resource needs, soils are increasingly recognized as one of our best options for sequestering carbon to combat climate change. Estimating the capacity for soils to stabilize carbon from the atmosphere is the first step in ensuring that climate mitigation efforts are successful in the long term. This study has shown that MAOC for soils under perennial cropping systems, growing on a range of soil types, can be accurately determined by utilizing a random forest model with either SSA (estimated by measuring the mass loss following oven-drying of soil) or the % of clay+silt, alongside concentration of SOC and exchangeable Ca. Furthermore, the stabilization capacity of soils with different surficial deposit types can be estimated using data on current MAOC concentrations and a quantile regression model (\u03c4 = 0.90). Not only were there MAOC deficits in all representative soil groups, but soils with greater % clay tended to have both greater MAOC formation capacity and greater MAOC deficits, with the exception of soils derived from gravelly fluvioglacial deposits.  This study provides further insight into the applicability of an established method of determining the stabilization capacity of soils, and highlights the importance of continued research into the mechanisms, and capacity of different soil types to sequester soil carbon over the long term.   Acknowledgements This work was funded by the Agricultural Greenhouse Gases Program of Agriculture and Agri-food Canada (Project AGGP2-25). Thank you to Tirhas Gebretsadikan, Naomi Yamaoka, Sophia Russo, Ieva Zigg, IIka Most, Allyson Dyck, Maya Bandy, Paige Munro,  137  Jadyn Patton, Shawn Kuchta, Istvan Losso, Brayden Jones, Seanna Zintel, Alene Wong, Kat Chen and Nora Skuridina who helped with the extensive soil sampling and laboratory work, and to the vineyard managers and orchardists who allowed us to take samples from their properties. 138  References Angers, D. A., Arrouays, D., Saby, N. P. A., & Walter, C. (2011). Estimating and mapping the carbon saturation deficit of French agricultural topsoils: Carbon saturation of French soils. Soil Use and Management, 27(4), 448\u2013452. Bai, X., Huang, Y., Ren, W., Coyne, M., Jacinthe, P.-A., Tao, B., Hui, D., Yang, J., & Matocha, C. (2019). Responses of soil carbon sequestration to climate smart agriculture practices: A meta-analysis. Global Change Biology. https:\/\/doi.org\/10.1111\/gcb.14658 Bai, Z., Caspari, T., Gonzalez, M. R., Batjes, N. H., M\u00e4der, P., B\u00fcnemann, E. K., de Goede, R., Brussaard, L., Xu, M., Ferreira, C. S. S., Reintam, E., Fan, H., Miheli\u010d, R., Glavan, M., & T\u00f3th, Z. (2018). Effects of agricultural management practices on soil quality: A review of long-term experiments for Europe and China. Agriculture, Ecosystems & Environment, 265, 1\u20137. Balesdent, J., Besnard, E., Arrouays, D., & Chenu, C. (1998). The dynamics of carbon in particle-size fractions of soil in a forest-cultivation sequence. Plant and Soil, 201(1), 49\u201357. Beare, M. H., McNeill, S. J., Curtin, D., Parfitt, R. L., Jones, H. S., Dodd, M. B., & Sharp, J. (2014). Estimating the organic carbon stabilisation capacity and saturation deficit of soils: a New Zealand case study. Biogeochemistry, 120(1), 71\u201387. Benbi, D. K., Boparai, A. K., & Brar, K. (2014). Decomposition of particulate organic matter is more sensitive to temperature than the mineral associated organic matter. Soil Biology & Biochemistry, 70, 183\u2013192. Boiteau, R. M., Kukkadapu, R., Cliff, J. B., Smallwood, C. R., Kovarik, L., Wirth, M. G., Engelhard, M. H., Varga, T., Dohnalkova, A., Perea, D. E., Wietsma, T., Moran, J. J., & Hofmockel, K. S. (2020). Calcareous organic matter coatings sequester siderophores in alkaline soils. The Science of the Total Environment, 724, 138250.  139  Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5\u201332. Brunauer, S., Emmett, P. H., & Teller, E. (1938). Adsorption of Gases in Multimolecular Layers. Journal of the American Chemical Society, 60(2), 309\u2013319. Cabuzel, T. (2021, April 27). Carbon Farming. https:\/\/ec.europa.eu\/clima\/policies\/forests\/carbon-farming_en Christensen, B. T. (1992). Physical Fractionation of Soil and Organic Matter in Primary Particle Size and Density Separates. In B. A. Stewart (Ed.), Advances in Soil Science: Volume 20 (pp. 1\u201390). Springer New York. Cotrufo, M. F., Ranalli, M. G., Haddix, M. L., Six, J., & Lugato, E. (2019). Soil carbon storage informed by particulate and mineral-associated organic matter. Nature Geoscience, 12(12), 989\u2013994. Cotrufo, M. F., Wallenstein, M. D., Boot, C. M., Denef, K., & Paul, E. (2013). The Microbial Efficiency-Matrix Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabilization: do labile plant inputs form stable soil organic matter? Global Change Biology, 19(4), 988\u2013995. Curtin, D., Beare, M. H., Qiu, W., & Sharp, J. (2019). Does Particulate Organic Matter Fraction Meet the Criteria for a Model Soil Organic Matter Pool? Pedosphere, 29(2), 195\u2013203. Emde, D., Hannam, K. D., Most, I., Nelson, L. M., & Jones, M. D. (2021). Soil organic carbon in irrigated agricultural systems: A meta-analysis. Global Change Biology, 27(16), 3898\u20133910. Environment, & Canada, C. C. (2019). Canadian Climate Normals 1981-2010 Station Data. Environment Canada. https:\/\/climate.weather.gc.ca\/climate_normals\/ Feng, W., Plante, A. F., & Six, J. (2013). Improving estimates of maximal organic carbon stabilization by fine soil particles. Biogeochemistry, 112(1), 81\u201393.  140  Fulton, R. J. (1969). Glacial lake history, southern interior Plateau, British Columbia. Department of Energy, Mines and Resources. Gerin, P. A., Genet, M. J., Herbillon, A. J., & Delvaux, B. (2003). Surface analysis of soil material by X-ray photoelectron spectroscopy. European Journal of Soil Science, 54(3), 589\u2013604. Hassink, J. (1997). The capacity of soils to preserve organic C and N by their association with clay and silt particles. Plant and Soil, 191, 77\u201387. Kaiser, K., & Guggenberger, G. (2000). The role of DOM sorption to mineral surfaces in the preservation of organic matter in soils. Organic Geochemistry, 31(7), 711\u2013725. Kaiser, K., & Kalbitz, K. (2012). Cycling downwards \u2013 dissolved organic matter in soils. Soil Biology & Biochemistry, 52, 29\u201332. Kalinichev, A. G., & Kirkpatrick, R. J. (2007). Molecular dynamics simulation of cationic complexation with natural organic matter. European Journal of Soil Science, 58(4), 909\u2013917. Kuhn, M. (2020). caret: Classification and Regression Training. https:\/\/CRAN.R-project.org\/package=caret Lavallee, J. M., Soong, J. L., & Cotrufo, M. F. (2020). Conceptualizing soil organic matter into particulate and mineral-associated forms to address global change in the 21st century. Global Change Biology, 26(1), 261\u2013273. Lehmann, J., Kinyangi, J., & Solomon, D. (2007). Organic matter stabilization in soil microaggregates: implications from spatial heterogeneity of organic carbon contents and carbon forms. Biogeochemistry, 85(1), 45\u201357. Mart\u00ed-Roura, M., Hagedorn, F., Rovira, P., & Romany\u00e0, J. (2019). Effect of land use and carbonates on organic matter stabilization and microbial communities in Mediterranean soils. Geoderma, 351, 103\u2013115.  141  Matus, F., Amigo, X., & Kristiansen, S. M. (2006). Aluminium stabilization controls organic carbon levels in Chilean volcanic soils. Geoderma, 132(1), 158\u2013168. Matus, F. J. (2021). Fine silt and clay content is the main factor defining maximal C and N accumulations in soils: a meta-analysis. Scientific Reports, 11(1), 6438. McNally, S. R., Beare, M. H., Curtin, D., Meenken, E. D., Kelliher, F. M., Calvelo Pereira, R., Shen, Q., & Baldock, J. (2017). Soil carbon sequestration potential of permanent pasture and continuous cropping soils in New Zealand. Global Change Biology, 23(11), 4544\u20134555. Midwood, A. J., Hannam, K. D., Forge, T. A., Neilsen, D., Emde, D., & Jones, M. D. (2020). Importance of drive-row vegetation for soil carbon storage in woody perennial crops: A regional study. Geoderma, 377, 114591. Midwood, A. J., Hannam, K. D., Gebretsadikan, T., Emde, D., & Jones, M. D. (2021). Storage of soil carbon as particulate and mineral associated organic matter in irrigated woody perennial crops. Geoderma, 403, 115185. Mikutta, R., Mikutta, C., Kalbitz, K., Scheel, T., Kaiser, K., & Jahn, R. (2007). Biodegradation of forest floor organic matter bound to minerals via different binding mechanisms. Geochimica et Cosmochimica Acta, 71(10), 2569\u20132590. Mikutta, R., Turner, S., Schippers, A., Gentsch, N., Meyer-St\u00fcve, S., Condron, L. M., Peltzer, D. A., Richardson, S. J., Eger, A., Hempel, G., Kaiser, K., Klotzb\u00fccher, T., & Guggenberger, G. (2019). Microbial and abiotic controls on mineral-associated organic matter in soil profiles along an ecosystem gradient. Scientific Reports, 9(1), 10294. Munro, P., Forge, T. A., Jones, M. D., & Nelson, L. M. (2020). Soil biota from newly established orchards are more beneficial to early growth of cherry trees than biota from older orchards. Applied Soil Ecology: A Section of Agriculture, Ecosystems & Environment, 155, 103658.  142  Ngailo, J. A., & Vieira, S. R. (2012). Spatial patterns and correlation of soil properties of a lowland soil. Journal of Soil Science and Environmental Management, 3(12), 287\u2013296. Parfitt, R. L., Whitton, J. S., & Theng, B. K. G. (2001). Surface reactivity of A horizons towards polar compounds estimated from water adsorption and water content. Soil Research, 39(5), 1105\u20131110. Pennock, D., Bedard-Haughn, A., & Viaud, V. (2011). Chernozemic soils of Canada: Genesis, distribution, and classification. Canadian Journal of Soil Science, 91(5), 719\u2013747. Poeplau, C., Don, A., Six, J., Kaiser, M., Benbi, D., Chenu, C., Cotrufo, M. F., Derrien, D., Gioacchini, P., Grand, S., Gregorich, E., Griepentrog, M., Gunina, A., Haddix, M., Kuzyakov, Y., K\u00fchnel, A., Macdonald, L. M., Soong, J., Trigalet, S., \u2026 Nieder, R. (2018). Isolating organic carbon fractions with varying turnover rates in temperate agricultural soils \u2013 A comprehensive method comparison. Soil Biology & Biochemistry, 125, 10\u201326. Puy, A., Garcia Aviles, J. M., Balbo, A. L., Keller, M., Riedesel, S., Blum, D., & Bubenzer, O. (2017). Drip irrigation uptake in traditional irrigated fields: The edaphological impact. Journal of Environmental Management, 202, 550\u2013561. Rasmussen, C., Heckman, K., Wieder, W. R., Keiluweit, M., Lawrence, C. R., Berhe, A. A., Blankinship, J. C., Crow, S. E., Druhan, J. L., Hicks Pries, C. E., Marin-Spiotta, E., Plante, A. F., Sch\u00e4del, C., Schimel, J. P., Sierra, C. A., Thompson, A., & Wagai, R. (2018). Beyond clay: towards an improved set of variables for predicting soil organic matter content. Biogeochemistry, 137(3), 297\u2013306. R Core Team. (2020). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. https:\/\/www.R-project.org\/  143  Ross, G. J., Hoyt, P. B., & Neilsen, G. H. (1985). Soil chemical and mineralogical changes due to acidification in Okanagan apple orchards. In Canadian Journal of Soil Science (Vol. 65, Issue 2, pp. 347\u2013355). https:\/\/doi.org\/10.4141\/cjss85-038 Rowley, M. C., Grand, S., Spangenberg, J. E., & Verrecchia, E. P. (2021). Evidence linking calcium to increased organo-mineral association in soils. Biogeochemistry. https:\/\/doi.org\/10.1007\/s10533-021-00779-7 Rowley, M. C., Grand, S., & Verrecchia, \u00c9. P. (2018). Calcium-mediated stabilisation of soil organic carbon. Biogeochemistry, 137(1), 27\u201349. Sanderman, J., Hengl, T., & Fiske, G. J. (2017). Soil carbon debt of 12,000 years of human land use. Proceedings of the National Academy of Sciences of the United States of America, 114(36), 9575\u20139580. Schmidt, M. W. I., Torn, M. S., Abiven, S., Dittmar, T., Guggenberger, G., Janssens, I. A., Kleber, M., K\u00f6gel-Knabner, I., Lehmann, J., Manning, D. A. C., Nannipieri, P., Rasse, D. P., Weiner, S., & Trumbore, S. E. (2011). Persistence of soil organic matter as an ecosystem property. Nature, 478(7367), 49\u201356. Service Canada. (2020, November 19). Net-zero emissions by 2050. https:\/\/www.canada.ca\/en\/services\/environment\/weather\/climatechange\/climate-plan\/net-zero-emissions-2050.html Six, J., Conant, R. T., Paul, E. A., & Paustian, K. (2002). Stabilization mechanisms of soil organic matter: Implications for C-saturation of soils. Plant and Soil, 241, 155\u2013176. Skjemstad, J. O., Swift, R. S., & McGowan, J. A. (2006). Comparison of the particulate organic carbon and permanganate oxidation methods for estimating labile soil organic carbon. Soil Research, 44(3), 255\u2013263.  144  Sutton, R., Sposito, G., Diallo, M. S., & Schulten, H.-R. (2005). Molecular simulation of a model of dissolved organic matter. Environmental Toxicology and Chemistry \/ SETAC, 24(8), 1902\u20131911. Trumbore, S. (2009). Radiocarbon and Soil Carbon Dynamics. Annual Review of Earth and Planetary Sciences, 37(1), 47\u201366. Trumbore, S. E. (1997). Potential responses of soil organic carbon to global environmental change. Proceedings of the National Academy of Sciences of the United States of America, 94(16), 8284\u20138291. Venables, W. N., & Ripley, B. D. (2002). Modern Applied Statistics with S (Fourth). Springer. http:\/\/www.stats.ox.ac.uk\/pub\/MASS4\/ Vogel, C., Mueller, C. W., H\u00f6schen, C., Buegger, F., Heister, K., Schulz, S., Schloter, M., & K\u00f6gel-Knabner, I. (2014). Submicron structures provide preferential spots for carbon and nitrogen sequestration in soils. Nature Communications, 5, 2947. Wander, M. M., & Bidart, M. G. (2000). Tillage practice influences on the physical protection, bioavailability and composition of particulate organic matter. Biology and Fertility of Soils, 32(5), 360\u2013367. Wander, M. M., Bidart, M. G., & Aref, S. (1998). Tillage impacts on depth distribution of total and particulate organic matter in three Illinois soils. Soil Science Society of America Journal. Soil Science Society of America, 62(6), 1704\u20131711. Woche, S. K., Goebel, M.-O., Mikutta, R., Schurig, C., Kaestner, M., Guggenberger, G., & Bachmann, J. (2017). Soil wettability can be explained by the chemical composition of particle interfaces - An XPS study. Scientific Reports, 7, 42877. Xu, X., Shi, Z., Li, D., Rey, A., Ruan, H., Craine, J. M., Liang, J., Zhou, J., & Luo, Y. (2016). Soil properties control decomposition of soil organic carbon: Results from data-assimilation analysis. Geoderma, 262, 235\u2013242.  145  Yuan, G., Soma, M., Seyama, H., Theng, B. K. G., Lavkulich, L. M., & Takamatsu, T. (1998). Assessing the surface composition of soil particles from some Podzolic soils by X-ray photoelectron spectroscopy. Geoderma, 86(3), 169\u2013181. Zomer, R. J., Bossio, D. A., Sommer, R., & Verchot, L. V. (2017). Global Sequestration Potential of Increased Organic Carbon in Cropland Soils. Scientific Reports, 7(1), 15554.  146  Supplementary Materials  Figure S1: Correlation matrix of all available soil characteristics. Cell colours are representative of the strength and direction of the correlation; dark blue is the strongest positive correlation, while dark red is the strongest negative correlation.   147   Figure S2: Random Forest variable importance plot for MAOC prediction model used for remainder of study.    148   Figure S3: Comparison of SSA pedotransfer functions and BET analysis. \u201cEq. 1\u201d and \u201cEq. 2\u201d refer to the equations outlined in section 2.1, and \u201cMoisture Loss\u201d is the calculated air-dry vs. oven-dry weight difference without adjustment. The Beare et al. 2014 equation uses a coefficient of 2, whereas to more closely match our BET measured SSA values, we adopted a coefficient of 0.61.   149  Table S1: Mean MAOC concentration for all cropping systems by depth. Cropping System Sample Location Depth (cm) Mean MAOC (g kg-1) sd drip apple Crop Row 0-15 5.025815 2.772879 drip apple Crop Row 15-30 3.707056 2.545832 drip apple Crop Row 30-60 2.82561 1.413746 drip apple Drive Row 0-15 6.362813 3.011948 drip apple Drive Row 15-30 4.363855 2.935224 drip apple Drive Row 30-60 2.840761 1.625211 drip grape Crop Row 0-15 4.605776 2.431222 drip grape Crop Row 15-30 3.570481 2.32327 drip grape Crop Row 30-60 2.861431 1.639028 drip grape Drive Row 0-15 5.123202 2.534825 drip grape Drive Row 15-30 3.66704 1.839499 drip grape Drive Row 30-60 2.53961 1.421968 micro-sprinkler apple Crop Row 0-15 5.157111 2.718299 micro-sprinkler apple Crop Row 15-30 3.758131 1.753523 micro-sprinkler apple Crop Row 30-60 2.473434 1.282848 micro-sprinkler apple Drive Row 0-15 6.352369 3.23885 micro-sprinkler apple Drive Row 15-30 3.881418 2.113104 micro-sprinkler apple Drive Row 30-60 2.443462 1.375894 micro-sprinkler cherry Crop Row 0-15 6.40005 2.515249 micro-sprinkler cherry Crop Row 15-30 4.221228 2.528067 micro-sprinkler cherry Crop Row 30-60 2.865955 1.280604 micro-sprinkler cherry Drive Row 0-15 7.106712 2.907405 micro-sprinkler cherry Drive Row 15-30 4.823403 3.017318 micro-sprinkler cherry Drive Row 30-60 2.888205 1.716741 non-cultivated  0-15 3.969871 2.33786  150  non-cultivated  15-30 2.961792 1.84253 non-cultivated  30-60 2.36644 1.627146    151  Table S2: Mineral associated organic carbon content deficits by soil type and depth. Soil Group Depth (cm) MAOC Deficit (g kg-1) Clayey Glaciolacustrine Sediments 0-15 1.62  15-30 4.01  30-60 5.80 Silty Glaciolacustrine Sediments 0-15 1.42  15-30 2.98  30-60 3.74 Morainal Deposits 0-15 2.91  15-30 4.14  30-60 5.02 Gravelly Fluvioglacial Deposits 0-15 2.79  15-30 4.53  30-60 6.03 Sandy Fluvioglacial Deposits 0-15 1.01  15-30 2.72  30-60 3.69   152  Chapter 4: Conclusion This thesis included two main studies:  i) a meta-analysis outlining patterns of SOC change in irrigated agricultural systems, and ii) an analysis of MAOC deficits in the orchards and vineyards of the Okanagan Valley, BC, Canada. The first study aimed to investigate changes in SOC in irrigated agricultural systems on a global scale by outlining patterns in SOC change in different climates, with different soil texture classification, and under varying irrigation methods. The second study aimed to estimate the potential for additional MAOC storage in the soils of irrigated orchards and vineyards in the Okanagan Valley, British Columbia, Canada. In doing so, I further aimed to establish a framework for modelling MAOC concentrations and MAOC formation capacity, including variable selection and mineral surface area estimation.  The meta-analysis study showed that the greatest increases of SOC in irrigated agricultural systems occur in semi-arid regions, such as the Okanagan Valley. Indeed, a recent study conducted in the Okanagan Valley found that SOC stocks were greater in irrigated orchards and vineyards than in adjacent un-irrigated, noncultivated areas (Midwood et al., 2020, 2021). The second study determined that soils from irrigated orchards and vineyards in the Okanagan Valley have the potential to store additional carbon as MAOC, but the degree to which this is the case depends on soil physicochemical characteristics. Together, my thesis work aimed to determine if and where there is the greatest potential for additional carbon storage in irrigated agricultural systems in the Okanagan Valley and beyond.   153  4.1 Soil organic carbon in irrigated agricultural systems For the first study, I compiled a dataset from 35 published studies, including 297 observations from 42 agricultural sites for which SOC measurements were taken at the beginning and end of the study period. While, on a global scale, these data show that irrigated agriculture can alter SOC stocks at all depths, the strength and direction of change varied by climate, soil texture, and irrigation strategy (Chapter 2, Figure 3). For example, irrigated agricultural systems in semi-arid regions showed the greatest increase in SOC across the full soil profile, with arid and dry sub-humid regions showing increased SOC near the surface, but not at depth. Irrigated agricultural systems in humid regions, by contrast, showed decreased SOC at all soil depths. Similarly, irrigated agricultural systems on medium- and fine-textured soils showed increased SOC in surface soils, but the downward percolation of fine soil particles (clay and fine silt) appears to cause inconsistent changes in SOC at depth. Finally, although sprinkler irrigation is a less efficient use of irrigation water, it was associated with a significant increase in SOC over the full soil profile. By contrast, drip irrigation, the most efficient irrigation strategy studied, showed only small gains in SOC near the soil surface (0 - 10 cm depth), and decreases at all other depths. It is important to note, however, that even in cases where there are significant increases of SOC across the whole soil profile, it remains unclear whether the SOC gains under irrigated agriculture are always sufficient to counteract losses in SOC caused by the conversion of unmanaged ecosystems for agriculture.  Heretofore unexamined at this scale, this study provides insights into global trends in SOC storage in irrigated agricultural systems. These results demonstrate that factors important at the plot level do not necessarily drive SOC dynamics on the global scale. For example, while crop and tillage practices are often considered important  154  factors in SOC accumulation for individual agricultural systems, the meta-analysis study showed that they were not significant drivers of SOC change across the globe. Instead, factors which relate to plant available water drive SOC change at the global scale (i.e., aridity, soil texture, and irrigation method). This has implications regarding the effective application of climate-smart agricultural management practices for mitigation of human caused global greenhouse gas emissions, and highlights areas where further research may yield important results.  4.2 Estimating MAOC deficits Stable carbon storage in soils is increasingly recognized as a pillar supporting global efforts to offset anthropogenic greenhouse gas emissions on the road towards net-zero carbon emission commitments (Service Canada, 2020). In order to focus efforts where they will be most likely to have a lasting effect, I aimed to estimate the capacity of different soil types under irrigated woody perennial crops in the Okanagan Valley to store additional MAOC. To this end, I have shown that, despite already having SOC levels higher than adjacent unmanaged areas under native vegetation, orchard and vineyard soils in the Okanagan Valley still have the potential to store additional MAOC (Midwood et al., 2020). The degree to which this is the case differed by soil type, however, and soils with greater % clay tended to have both greater MAOC formation capacity, and greater MAOC deficits. Soils with greater % sand showed the opposite trend, with both lower MAOC stabilization capacities and MAOC deficits. I further established a process by which MAOC can be estimated for soils, without the need of expensive measurements and difficult or time-consuming laboratory procedures. Mineral-associated organic carbon estimation frameworks that allow for  155  consideration of a broad range of soil types have the potential to open the door for stable carbon modelling at a larger scale than would otherwise be possible, and offers cost effective methods by which efforts to increase carbon sequestration in soils over the long term may be focussed.   4.3 Strengths, limitations, and future study directions Both parts of this thesis make clear the complex nature of soil carbon dynamics in soils with different physicochemical properties. The effect of irrigation on SOC and MAOC formation varies not only by soil texture, but also by method of irrigation and climate zone. Further, because MAOC is intrinsically linked to SOC, factors that affect SOC storage likewise affect MAOC storage, but to different degrees based on soil physicochemical properties and soil mineralogy. All of this is to say that it is not possible to make prescriptive recommendations for increasing SOC and MAOC in agricultural soils on a broad scale. Site and soil specific factors including interactive effects of agricultural management practices must be considered when adapting irrigation practices to enhance SOC and MAOC. The meta-analysis presented in Chapter 2 of this thesis shows that the effects of irrigated agriculture on SOC are not simple and straightforward: they are mediated by a suite of factors, including climate, soil texture, and irrigation method. Because meta-analysis combines the data of multiple peer-reviewed research papers, the results are considered to be statistically strong, and are less biased by anomalous samples or sites. However, this statistical strength depends on the quality of data reported from diverse studies; required data was not always presented. For example, while it has been well established in the literature that SOC varies greatly by soil depth (McNally et al., 2017;  156  Mudge et al., 2017; Trost et al., 2013), most studies only reported SOC values near the soil surface. Consequently, conclusions about changes in SOC at depth relied on relatively few studies. Similarly, study sites that use irrigation in more mesic areas are less common than those in arid and semi-arid sites. As such, while the patterns shown for more arid regions are robust because they draw on the findings of numerous study sites, those for more humid regions relied on data from only a few sites. That being said, the areas where data are most scarce provide clear guidance on future study directions. Specifically, this meta-analysis highlights the need for studies that look at SOC storage in irrigated agricultural systems at depth, in more mesic climate zones, and over longer durations. Conversely, for the MAOC deficit study, detailed, consistent, and thorough data were available for many soil chemical and physical measurements at a range of depths. The difficulty was determining a method by which the MAOC formation capacity of soils could be reasonably estimated. The 90th quantile modelling methodology represents the current \u2018best\u2019 method but this approach has its limitations. For example, 90th quantile models rely on the assumption that at least some of the soil samples are at (or near) MAOC formation capacity. However, there is no way to know for certain whether there are samples in the dataset that represent this state, because there is currently no way to directly measure MAOC formation capacity. Further, while using real soil measurements to model MAOC formation capacity provides confidence that the formation capacity reported is reasonably obtainable, it is certain to underestimate the MAOC formation capacity of a given soil, especially at depth. Finally, while estimates of MAOC formation capacity can be obtained using modelling techniques on grouped samples, we cannot know the actual mechanisms limiting MAOC formation. This study sheds light on the importance of continued research into the mechanisms of MAOC formation capacity,  157  and the ability of different soil types to produce MAOC. While these two studies were carried out on vastly different scales, they both identify the need for future study in at least two distinct areas. Firstly, the meta-analysis showed that, in general, irrigated agricultural systems increase SOC; however, it isn\u2019t clear whether this is sufficient to offset the SOC lost by conversion of non-cultivated land for irrigated agriculture. Land-use assessments and chronosequence studies could be carried out in combination with soil physicochemical data collection over as long a time-period as possible to determine whether there remains a carbon debt in irrigated agricultural systems. Secondly, soil carbon dynamics at depth in response to the movement of clay particles requires further attention. Soil texture - clay content in particular - is an important factor in soil carbon dynamics at any scale. As water percolates through the soil profile, clay particles often become mobile and translocate deeper into the soil. This change in soil texture over time isn\u2019t well studied, and it isn\u2019t clear how the translocation of clay particles affects soil carbon dynamics.     158  References McNally, S. R., Beare, M. H., Curtin, D., Meenken, E. D., Kelliher, F. M., Calvelo Pereira, R., Shen, Q., & Baldock, J. (2017). Soil carbon sequestration potential of permanent pasture and continuous cropping soils in New Zealand. Global Change Biology, 23(11), 4544\u20134555. Midwood, A. J., Hannam, K. D., Forge, T. A., Neilsen, D., Emde, D., & Jones, M. D. (2020). Importance of drive-row vegetation for soil carbon storage in woody perennial crops: A regional study. Geoderma, 377, 114591. Mudge, P. L., Kelliher, F. M., Knight, T. L., O\u2019Connell, D., Fraser, S., & Schipper, L. A. (2017). Irrigating grazed pasture decreases soil carbon and nitrogen stocks. Global Change Biology, 23(2), 945\u2013954. Service Canada. (2020, November 19). Net-zero emissions by 2050. https:\/\/www.canada.ca\/en\/services\/environment\/weather\/climatechange\/climate-plan\/net-zero-emissions-2050.html Trost, B., Prochnow, A., Drastig, K., Meyer-Aurich, A., Ellmer, F., & Baumecker, M. (2013). Irrigation, soil organic carbon and N2O emissions. A review. Agronomy for Sustainable Development, 33(4), 733\u2013749. ","@language":"en"}],"Genre":[{"@value":"Thesis\/Dissertation","@language":"en"}],"GraduationDate":[{"@value":"2022-02","@language":"en"}],"IsShownAt":[{"@value":"10.14288\/1.0406288","@language":"en"}],"Language":[{"@value":"eng","@language":"en"}],"Program":[{"@value":"Biology","@language":"en"}],"Provider":[{"@value":"Vancouver : University of British Columbia Library","@language":"en"}],"Publisher":[{"@value":"University of British Columbia","@language":"en"}],"Rights":[{"@value":"Attribution-NonCommercial-NoDerivatives 4.0 International","@language":"*"}],"RightsURI":[{"@value":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/","@language":"*"}],"ScholarlyLevel":[{"@value":"Graduate","@language":"en"}],"Supervisor":[{"@value":"Jones, Melanie D.","@language":"en"},{"@value":"Hannam, Kirsten","@language":"en"}],"Title":[{"@value":"Understanding the impact of irrigated agricultural systems on soil organic carbon storage","@language":"en"}],"Type":[{"@value":"Text","@language":"en"}],"URI":[{"@value":"http:\/\/hdl.handle.net\/2429\/80664","@language":"en"}],"SortDate":[{"@value":"2021-12-31 AD","@language":"en"}],"@id":"doi:10.14288\/1.0406288"}