INNOVATIVE APPROACHES TO GA U GE RESILIENCE OF MANAGED RAINFED AGRICULTURAL SYSTEMS By Eeswaran Rasu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Crop and Soil Sciences Doctor of Philosophy 20 20 ABSTRACT INNOVATIVE APPROACHES TO GA U GE RESILIENCE OF MANAGED RAINFED AGRICULTURAL SYSTEMS By Eeswaran Rasu Increasing climate extremes have devastated both crop yields and farm economies, especially in the rainfed agricultural systems of the Midwestern United States. Furthermore, these extreme events are projected to increase in the future due to climate change. As a result, alternative agricultural practices are becoming more common in eff orts to mitigate the impacts of extreme climatological events. Nevertheless, the level of resilienc y of these practices has not yet been adequately quantified due to the l ack of robust metrics available to address the complexity of agricultural systems whi le being simple enough to be measured at different scales. To this end, three studies were conducted and compiled into this dissertation. These studies were carried out in a long - term cropping system experiment at the W.K. Kellogg Biological Station and th e Kalamazoo River W atershed located in the Southwestern Michigan of the United States . The first study was designed to evaluate the applicability of soil moisture metrics to gauge resiliency of four differently managed rainfed agricultural treatments at th e field scale. The robustness of these metrics was assessed over a long - period (1993 - 2018) for a corn - soybean - wheat rotation by monitoring crop growth and yield in response to climate variability . Results demonstrated that the soil moisture metrics can be used as indicators of resilience at the field scale. The no - till treatment had the highest level of resilience as quantified by soil moisture retention, effectiveness of reducing drought severity , crop yields, and stability of yields. Although the organ ic treatment substantially improved resiliency in terms of soil moisture conservation and drought mitigation than the conventional treatment , the limitation of available nitrogen significantly reduced corn and wheat yields. Meanwhile, the reduced input tre atment was the least resilient as it was vulnerable during extreme climate conditions. The second study was performed to evaluate the climate resilience of four rainfed agricultural treatments in terms of profitability and farm risks for the same corn - soy bean - wheat rotation . Crop production and management data were used to conduct enterprise budgeting and risk analysis. The means and volatility of estimated net returns and risk preferences were used as the evaluation metrics . According to the results of th is study, the organic and the no - till treatments had higher resilience than the conventional and the reduced input treatments as they were projected to generate greater net revenues with higher stability. Furthermore, these treatments were promising to cat er to a large group of farmers with different risk preferences. Meantime, conventional and reduced inputs treatments were found to be adversely affected by climate extremes. The goal of the third study was to examine the impacts of large - scale adaptation o f conservation agricultural practice s (i.e., no - till treatment) on resilience in comparison to traditional practices (i.e., conventional treatment). Similar to the previous two studies, the corn - soybean - wheat rotation was applied on all agricultural land u se in the Kalamazoo River W atershed during the period of 1993 - 2019. Crop and groundwater models were integrated to derive the resilience metrics, namely r echarge, groundwater level , soil moisture, yield, and net return . Results showed clear improvement in all metrics under the no - till treatment. Therefore, the adoption of the no - till could improve the overall resilience of the corn - soybean - wheat rotation . Together, these studies present a set of robust metrics to quantify the resilience of diverse rainfed agricultural systems both at the field and watershed scale. Copyright by EESWARAN RASU 20 20 v This dissertation is dedicated to my loving parents, wife, and brothers for all the love and care they have given me. vi ACKNOWLEDGMENTS I am very grateful to owe my gratitude to a great many people and institutions who guided, helped, and supported me during my doctoral program in Crop and Soil Sciences at Michigan State University . At this moment of accomplishment, I would first offer my sincere gratitude to my major advisor Dr. Pouyan Nejadhashemi for his excellent supervision, constructive feedback , kind g uidance , and sparing valuable time throughout my program . Without which it would have never been a reality to complete this task. I am thankful for everything I have learn ed from him and for everything he has done for me. My deep sense of gratitude goes t o my advisory committee, Dr. Steven R. Miller, Dr. Alexandra Kravchenko, and Dr. Shu - Guang Li, for their support and guidance to plan, execute and evaluate my doctoral research work. I would also like to thank Dr. Karen Renner, Dr. James Kells, and Mrs. Ma ckenzie Graham for all their support from the Department of Plant, Soil and Microbial Sciences to complete this program. I would like to extend my gratitude to the Department of Community Sustainability for awarding me for the C.S. Mott predoctoral fellows hip, Kellogg Biological Station (KBS) for offering me KBS - Long Term Ecological Research (LTER) summer graduate fellowship, and the College of Agriculture and Natural Resources for providing me the dissertation completion fellowship and other financial supp ort during my program. The C.S. Mott fellowship has also provided me with an opportunity to follow a dual major in Ecological Food and Farming Systems. I would sincerely thank Dr. Amor Ines for providing me an opportunity to work on some of his projects du ring the early stages of the program. vii I gratefully acknowledge the principal investigators and technical staff of the LTER experiment at the KBS for their support throughout this research. Dr. Stephen K. Hamilton, Dr. Neville Millar, and the Information Manager Sven Bohm should be thanked for their prompt responses for any clarifications regarding Main Cropping System Experiment (MCSE) at KBS. I acknowledge Dr. Peter Grace from Queensland University of Technology, Australia, and Dr. Philip Robertson from Michigan State University for proposing best - suited cultivars for the DSSAT framework. Dr. Cheryl Porter from the University of Florida is thanked for her quick responses for the clarifications during the crop modeling in DSSAT. My sincere appreciation goe s to my colleagues ; Dr. Babak Saravi, Sebastian Hernandez - Suarez , Josué Kpodo , Anna Raschke , Ian Kro p p , Suyog Chaudhari, Dr. Prakash Jha, Abhijeet Abhishek, and Hannah Ferriby for their incredible assistance and encouragement during this dissertation research. In addition , I would also thank Dr. Zachary Curtis , Dr. Umesh Adhikari , and Dr. Huasheng Liao from the Hydrosimulatics team for assisting in regional groundwater modeling. Finally, I would like to e spec ially thank my family . To my mother, father, and lovely wife Hirishani for encouraging me to p ursu e my doctoral studies and always being a constant source of love and emotional support from miles apart. To my brothers, Rathan, Thuva, and Thusi , for always being there to make fun and enjoy life. Thank you for all the love and blessings you have given me especially during a global pandemic, COVID - 19 . viii TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ......................... xi LIST OF FIGURES ................................ ................................ ................................ ..................... xiii KEY TO ABBREVIATIONS ................................ ................................ ................................ ....... xv 1. INTRODUCTION ................................ ................................ ................................ ................... 1 2. REVIEW OF LITER ATURE ................................ ................................ ................................ .. 4 2.1 Overview ................................ ................................ ................................ ............................... 4 2.2 Rainfed Agriculture ................................ ................................ ................................ ............... 5 2.2.1 Global Challenges in Rainfed Agriculture ................................ ................................ ..... 5 2.2.2 Rainfed Agriculture in the Midwestern United States ................................ .................... 6 2.2.3 Challenges of Rainfed Agriculture in the US Midwest ................................ .................. 6 2.3 Resilience in Rainfed Agriculture ................................ ................................ ......................... 7 2.4 Interventions to Improve Resilience in Rainfed Agriculture ................................ ................ 8 2.4.1 Agronomic Interventions ................................ ................................ ................................ 8 2.4.2 Genetic Interventions ................................ ................................ ................................ ...... 9 2.4.3 Decision Support Interventions ................................ ................................ ...................... 9 2.5 Promising Agricultural Systems to Improve Resilience ................................ ..................... 10 2.5.1 Conservation Agriculture ................................ ................................ ............................. 10 2.5.2 Organic Agriculture ................................ ................................ ................................ ...... 12 2.5.3 Reduced Input Agriculture ................................ ................................ ........................... 13 2.6 Metrics of Resilience in Agricultural Systems ................................ ................................ .... 13 2.6.1 Soil Moisture ................................ ................................ ................................ ................ 15 2.6.2 Drought Indices ................................ ................................ ................................ ............ 16 2.6.3 Farm Economics ................................ ................................ ................................ ........... 22 2.6.4 Ecosystem Services ................................ ................................ ................................ ...... 24 2.7 Modeling Resilience at a Large Scale ................................ ................................ ................. 27 2.7.1 Crop Modeling ................................ ................................ ................................ .............. 28 2.7.2 Hydrological Modeling ................................ ................................ ................................ . 30 2.8 Adaptation of Resilient Practices ................................ ................................ ........................ 32 2.9 Summary ................................ ................................ ................................ ............................. 34 3. INTRODUCTION TO METHODOLOGY ................................ ................................ ........... 36 4. EVALUATING THE APPLICABILITY OF SOIL MOISTURE - BASED METRICS FOR GAUGING THE RESILIENCY OF RAINFED AGRICULTURAL SYSTEMS ....................... 39 4.1 Introd uction ................................ ................................ ................................ ......................... 39 4.2 Materials and Methods ................................ ................................ ................................ ........ 44 4.2.1 Study Location and Site Description ................................ ................................ ............ 45 4.2.2 Data Collection ................................ ................................ ................................ ............. 48 4.2.3 Calculating the Metrics of temporal Dynamics of Soil Moisture to Examine the Relative Resilien ce of Different Rainfed Agricultural Treatments ................................ ....... 51 ix 4.2.4 Evaluate the Sensitivity of the Mean Relative Difference of Soil Moisture to Climate Variability and its Reflections on Crop Growth and Yield in Different Treatments ............ 53 4.2.5 Compare the Effectiven ess of Different Treatments on Reducing Agricultural Drought Severity .. ................................ ................................ ................................ ................................ 55 4.3 Results and Discussion ................................ ................................ ................................ ........ 58 4.3.1 Ranking the Resilience of Soil Moisture in Different Treatments ............................... 58 4.3.2 Temporal Persistence of Soil Moisture in Different Treatments ................................ .. 63 4.3.3 Evaluating the Effects of Treatments and Climate Variability on Mean Relative Difference of Soil Moisture and its Reflections on Growth and Yield of Crops .................. 64 4.3.4 Comparing the Effectiveness of Different Treatments on Reducing Agricultural Drought Severity ................................ ................................ ................................ .................... 69 4.4 Conclusions ................................ ................................ ................................ ......................... 74 4.5 Acknowledgment ................................ ................................ ................................ ................ 77 5. EVALUATING THE CLIMATE RESILIENCE IN TERMS OF PROFITABILITY AND FARM RISK MANAGEMENT FOR A LONG - TERM CORN - SOYBEAN - WHEAT ROTATION UNDER DIFFERENT TREATMENT SYSTEMS ................................ ................. 78 5.1 Introduction ................................ ................................ ................................ ......................... 78 5.2 Materials and Methods ................................ ................................ ................................ ........ 83 5.2.1 Study Location and Details of the Experiment ................................ ............................. 84 5.2.2 Enterprise Budget Analysis to Estimate Net Return under Different Production and Treatment Systems ................................ ................................ ................................ ................ 87 5.2.3 Categorization of Climate Variability ................................ ................................ .......... 89 5.2.4 Evaluating the Effects of Different Production and Treatment Systems and Climate Variability on the Expected Net Return ................................ ................................ ................ 90 5.2.5 Defining Climate Resilience ................................ ................................ ......................... 91 5.2.6 Evaluating the Risk Level for the Adaptation of Different Production and Treatment Systems by Producers ................................ ................................ ................................ ............ 93 5.3 Results and Discussion ................................ ................................ ................................ ........ 94 5.3.1 Effects of Treatment Systems and Climate Variability on the Net Return of the Different Production Systems ................................ ................................ ................................ 94 5.3.2 Resilience of Production and Treatment Systems as Measured by the Net Return ..... 95 5.3.3 Producer Risk Levels for the Adaptation of Different Production and Treatments Systems .. ................................ ................................ ................................ .............................. 100 5.4 Conclusions ................................ ................................ ................................ ....................... 103 5.5 Acknowledgment ................................ ................................ ................................ .............. 104 6. QUANTIFICATION OF RESILIENCE METRICS AS AFFECTED BY A CONSERVATION AGRICULTURAL PRACTICE AT A WATERSHED SCALE ............... 106 6.1 Introduction ................................ ................................ ................................ ....................... 106 6.2 Materials and Methods ................................ ................................ ................................ ...... 109 6.2.1 Study Area ................................ ................................ ................................ .................. 111 6.2.2 Description of Long - Term Field Experiment ................................ ............................. 111 6.2.3 Description of the Study Watershed ................................ ................................ ........... 113 6.2.4 Crop Modeling ................................ ................................ ................................ ............ 115 6.2.5 Groundwater Modeling ................................ ................................ ............................... 119 6.2.6 Simulation of Crop Yields and Recharge at a Watershed Scale ................................ . 122 x 6.2.7 Assessment of Resilience as Af fected by the Adoption of No - Till Agriculture ........ 122 6.3 Results and Discussion ................................ ................................ ................................ ...... 125 6.3.1 Calibration of the Crop Model ................................ ................................ .................... 125 6.3.2 Calibration of the Groundwater Model ................................ ................................ ...... 126 6.3.3 Resilience as Affected by the Adoption of No - Till Agriculture ................................ 127 6.4 Conclusions ................................ ................................ ................................ ....................... 137 6.5 Acknowledgment ................................ ................................ ................................ .............. 138 7. CONCLUSIONS ................................ ................................ ................................ ................. 139 8. RECOMMENDATIONS FOR FUTURE RESEARCH ................................ ..................... 142 APPENDIX ................................ ................................ ................................ ................................ . 144 REFERENCES ................................ ................................ ................................ ........................... 165 xi LIST OF TABLES Table 2.1. Commonly used available agricultural drought indices. ................................ ............. 17 Table 4.1. Climate variability for each crop in the rotation. ................................ ......................... 57 Table 4.2. Relative positions and ranking based on MRD, and their re spective percentages for each agricultural system under each climate variability category. ................................ ............... 61 Table 4.3. The means of MRD, total bio mass and yield for different treatments and crop types .. ................................ ................................ ................................ ................................ ........ 65 Table 5.1. Conventional and organic commodity prices used to value the crops in different treatment systems. ................................ ................................ ................................ ......................... 89 Table 5.2. Probabilities for the effects evaluated in the statistical mixed model for th e net return ................................ ................................ ................................ ................................ ......... 95 Table 5.3. Means of net return under different production systems as affected by climate variability. ................................ ................................ ................................ ................................ ..... 97 Table 5.4. Payoff matrix of production systems under different treatment systems as affected by climate variability. ................................ ................................ ................................ ...................... 102 Table 6.1. The goodness of fit parameters of the calibrated crop model to simulate yield and soil moisture under the conventional and no - till treatments. ................................ ............................. 125 Table 6.2. The mean yield and net return for different crops under two treatments in the Kalamazoo River watershed. ................................ ................................ ................................ ...... 136 Table S4.1. Details of the four annual row crop treatments of MCSE investigated in this study ................................ ................................ ................................ ................................ ...... 145 Table S4.2. P - values for the effects evaluated in the statistical mixed model for MRD, total biomass and yield. ................................ ................................ ................................ ....................... 151 Table S4.3. Coefficient of variation of total crop biomass and yield of crops under different treatments as affected by climate extremes. ................................ ................................ ............... 152 Table S5.1. Details and timing of management operations for corn - soybean - wheat rotation under the conventional treatment (CON). ................................ ................................ ............................. 153 Table S5.2. Details and timing of management operations for corn - soybean - wheat rotation under the no - till treatment (NT). ................................ ................................ ................................ ........... 154 xii Table S5.3. Details and timing of management operations for corn - soybean - wheat rotation under the reduced input treatment (RI). ................................ ................................ ................................ 155 Table S5.4. Details and timing of management operations for corn - soybean - wheat rotation under the organic treatment (OR). ................................ ................................ ................................ ........ 156 Table S5.5. Means of yield under different production and treatment systems as affected by climate variability. ................................ ................................ ................................ ...................... 157 Table S6.1. Soil properties at the KBS Main Cropping System Experiment site used to develop the sequential DSSAT model. ................................ ................................ ................................ ..... 160 Table S6.2. Probability values for the significance of the effects evaluated in the statistical mixed model for crop yields and net return. ................................ ................................ .......................... 160 xiii LIST OF FIGURES Figure 4.1. The experimental fields of the Main Cropping System Experiment (MCSE) at the Kellogg Biological Station (KBS) in Michigan, United States. ................................ ................... 47 Figure 4.2. Cumulative probability d istribution of seasonal precipitation for the period of 30 years (1989 - 2018) at KBS weather station. ................................ ................................ .................. 56 Figure 4.3. Ranked MRD of vo lumetric soil moisture and ITS for each treatment during the dry years. ................................ ................................ ................................ ................................ ............. 59 coefficient of different treatments for the duration of the study. ................................ ................................ ................................ ................................ ............. 64 Figure 4.5. Percentage of agricultural drought severity events based on different treatments and climate variabilities during the experimental period (1993 - 2018). ................................ .............. 71 Figure 4.6. Probability dist ribution of drought experienced by each agricultural system as represented by SWDI during the experimental period (1993 - 2018). ................................ ........... 72 Fi gure 5.1. The location of the Main Cropping System Experiment (MCSE) in Southwest Michigan, United States. ................................ ................................ ................................ ............... 84 Figure 5.2. The probab ility distribution of total seasonal precipitation (1989 - 2018) at the KBS weather station. ................................ ................................ ................................ ............................. 90 Figure 5.3. Scatter plot between the expected net return and the relative standard deviation for different treatments as affected by climate variability in different production systems. .............. 99 Figure 5.4. Box plot of net return of corn production system under different treatments as affected by climate variability. ................................ ................................ ................................ ... 100 Figure 6.1. An overview of the modeling process. ................................ ................................ ..... 110 Figure 6.2. Location of the experimental site and th e Kalamazoo River Watershed in Michigan, USA. ................................ ................................ ................................ ................................ ............ 113 Figure 6.3. Comparison between simulated groundwater heads and observed groundwater heads ................................ ................................ ................................ ................................ ....... 12 7 Figure 6.4. The mean annual recharge from different crops under two treatments in the Kalamazoo River watershed. ................................ ................................ ................................ ...... 128 Figure 6.5. Monitoring well (MW) locations superimposed over the 2019 head distribution under conventional treatment. ................................ ................................ ................................ ............... 130 xiv Figure 6.6. Simulated water table heads under the conventional and no - till treatments for the six monitoring wells. ................................ ................................ ................................ ........................ 131 Figure 6.7. The probability distribution for the mean, maximum, and minimum of MRD across different soils in the Kalamazoo River watershed for the period of 1993 - 2019 as affected by the conventional (a) and the no - till (b) treatments. ................................ ................................ ........... 134 Figure 6.8. The probab ility distribution for the mean, maximum, and minimum of SWDI across different soils in the Kalamazoo River watershed for the period of 1993 - 2019 as affected by the conventional (a) and the no - till (b) treatments. ................................ ................................ ........... 134 Figure S4.1. Ranked MRD of volumetric soil moisture and ITS for each treatment during the normal years. ................................ ................................ ................................ ............................... 146 Figure S4.2. Ranked MRD of volumetric soil moisture and ITS for each treatment during the wet years. ................................ ................................ ................................ ................................ ........... 147 Figure S4.3. Means of soil organic carbon content in different treatments during the period of 1989 - 2001 in this experiment. ................................ ................................ ................................ .... 148 Figure S4.4. Association between mean relative difference (MRD) of volumetric soil moisture and soil organic carbon content in the treatments investigated in this study. ............................. 149 Figure S4.5. Total biomass (a - c) and yield (d - f) of crops as affected by the interaction between treatment and climate variability. ................................ ................................ ............................... 150 Figure S5.1. Box plot of net return of soybean production system under different treatments as affected by climate variability. ................................ ................................ ................................ ... 158 Figure S5.2. Box plot of net return of wheat production system under different treatments as affected by climate variability. ................................ ................................ ................................ ... 159 Figure S6.1. Results of the calibrated steady - state groundwater model including head contours, color map for head, and velocity vectors. ................................ ................................ ................... 161 Figure S6.2. Mean annual recharge from corn (a), soybean (b), and wheat (c) across different soils in the Kalamazoo River watershed for the period betwe en 1993 - 2019 as affected by the conventional and the no - till treatments. ................................ ................................ ...................... 162 Figure S6.3. Mean yield of corn (a), soybean (b), and wh eat (c) across different soils in the Kalamazoo River watershed for the period between 1993 - 2019 as affected by the conventional and the no - till treatments. ................................ ................................ ................................ ........... 163 Figure S6.4. Average net return of corn (a), soybean (b), and wheat (c) across different soils in the Kalamazoo River watershed for the period between 1993 - 2019 as affected by the conventional and the no - till treatments. ................................ ................................ ...................... 164 xv KEY TO ABBREVIATIONS ADI: Accumulated Drought Index AMSL: Above Mean Sea Level APSIM: Agricultural Production Systems Simulator ARMS: Agricultural Resource Management Survey BD: Bulk D ensity C O BRA : Community Based Resilience Assessment CON: Conventional COSA: C ommittee on S ustainability A ssessment CR I STAL : Community - based Risk Screening Tool A daptation and Livelihood CV: Coefficient of Variation DEM: D igital E levation M odel DSSAT: Decision Support System for Agrotechnology Transfer EMV: E xpected M onetary V alue ET: Evapotranspiration ET 0 : Reference Evapotranspiration ETDI: Evapotranspiration Deficit Index FAO: Food and Agriculture Organization FC: Field Capacity HBV: Hydrologiska Byråns Vattenbalansavdelning IFOAM: International Federation of Organic Agriculture Movements IGW: Interactive Groundwater IISD: International Institute for Sustainable Development ITS: Index of Temporal St ability KBS: Kellogg Biological Station xvi KRWC: Kalamazoo River Water Council LTER: Long - Term Ecological Research MAGNET : Multi - scale Adaptive Global Network MCSE: Main Cropping System Experiment MDEQ : Michigan Department of Environmental Quality MEA: Millennium E cosystem A ssessment MRD: Mean R elative D ifference MW: Monitoring Well NCEI: National Centers for Environmental Information NLCD: National Land Cover Database NOAA : National Oceanic and Atmospheric Administration NRMSE: N ormalized R oot M ean S qu are E rror NSE: Nash - Sutcliffe M odel E fficiency C oefficient NT: No - till OR: Organic PDSI: Palmer Drought Severity Index PET: P otential E vapotranspiration RATALF: Resilience, and Adaptation and Transformation Assessment Framework RDI: Reconnaissance Drought Index RHEAS: Regional Hydrological Extremes Assessment System RI: Reduced Input RIMA: Resilience Index Measurement and Analysis Model RSD: Relative S tandard D eviation RWD: Relative Water Deficit SALUS: System Approach to Land Use Sus tainability xvii SHARP: Self - evaluation and Holistic Assessment of climate Resilience of farmers and Pastoralists SMDI: Soil Moisture Deficit Index SWAT: Soil and Water Assessment Tool SWDI: Soil Water Deficit Index SWL: Static Water Level UNDP: United Nations Development Program US: United States USDA: United States Department of Agriculture USGS: United States Geological Survey VIC: Variable Infiltration Capacity VRD: Variance of Relative Difference WEAP: Water Evaluation and Planning System WMO: World Meteorological Organization WP: Wilting Point WS: Water Stress Ratio WSA: Water S tress A nomaly 1 1. INTRODUCTION Sustainably increasing the production of edible crops by 60 - 120 percent to feed the increasing global population by 2050 is a major challenge of the 21 st century (Berners - Lee et al., 2018; Pradhan et al., 2015; Tilman et al., 2011) . Food crops are produced both under rainfed and irrigated agricultural systems around the world. Remarkably, rainfed agriculture contributes to 60 percent of the total food production throughout the world (FAO, 2017) , while in the United States, the major (USDA, 2014) . The rainfed agricultural system is dominant in the Midwestern United States, which is highly productive, economically vital, technologically advanced , and predominantly produces row c rops of corn, soybean, and wheat (Franzluebbers et al., 2011) . Unfortunately, the regional climate of the Midwestern States is changing as a result of anthropogenic global climate change, and the increasing frequencies of climate extremes such as droughts, floods, and heatwaves have been documented (Andresen et al., 2012; Hatfield et al., 2018; Pryor et al., 2014) . These extreme events significantly impact crop yields and may offset the productivity increment achieved through genetics and management advance ments within the past few decades (Hatfield et al., 2018; Wang et al., 2016) . Soil moisture is the most limiting factor of productivity in rainfed agriculture. The availability of soil moisture is dictated by the seasonality of precipitation (i.e., onset, intensity, and frequency) and its interactions with the soil - plant - atmosphere continuum (Rost et al., 2009) . This dependency on precipitation makes the rainfed system very vulnerable to climate extremes. For example, the 2012 drought devastated the crops and economies of rainfed farmers in the US Midwest (Boyer et al., 2013; Rippey, 2015) . It is projected that climate extremes, such as these, will only increase in frequency and intensity in the future (Jin et al., 2018) . 2 There is an increasing consensus among stakeholders that the resilience of these production systems must be improved. Resilience can be defined as the ability of a system to maintain its structure and functions in the face of climate perturbations (Holling , 1973; Walker et al., 2004) . Resiliency allows the system to continue to provide its services, which in agriculture , means the growth and yield of crops, stress tolerance, profits, and other ecosystem services offered by agricultural systems . Nevertheless , increasing evidence demonstrates that conventional agriculture, which is currently widely practiced around the world, has become vulnerable to the stressors associated with climate change and variability (Adhikari et al., 2015; Lesk et al., 2016) . Moreov er, conventional agriculture has been shown to contribute to global environmental change and therefore be unsustainable (Foley et al., 2011) . As a result, alternative agricultural practices, such as conservation agriculture, have been proposed. However, th e resilience of these practices against climate extremes is not yet well understood and the risk levels and profitability concerns associated with adoption have prevented widespread implementation (Eeswaran, 2018; Mausch et al., 2017; Roesch - McNally et al. , 2018) . Resilience is applicable to a wide range of disciplines (e.g., biological systems, engineering, economics, social welfare, politics, etc.), and these disciplines determine how resilience should be quantified and what metrics should be used to gau ge the resiliency (Quinlan et al., 2016) . Although several metrics have been previously applied to measure the resilience of agricultural systems, most of them have been qualitative . Moreover, these metrics often fail to capture the temporal dynamics at both the field and regional scales . Therefore, this dissertation aims to advance the understanding of the following knowledge gaps and thereby contribute to the crop and soil sciences to better manage agricultural systems in a changing environment. 3 Knowledge Gap 1: Lack of a robust resilience metric to address the complexity in agricultural systems as related to temporal variabilities while being simple enough to be measur ed at the field scale. Knowledge Gap 2: R esilience matrices are needed to account for long - term profitability and risks at the farm scale for alternative management systems (treatments) . Knowledge Gap 3: Absence of knowledge on how the conservation agricu lture (i.e., no - till) affects the overall resiliency at large scales (e.g., a large watershed). To address the knowledge gap 1 , the following objectives were accomplished: (1) rank the relative resilience of different rainfed agricultural systems using the metrics of temporal dynamics of soil moisture; ( 2) evaluate the robustness of the soil moisture metrics of temporal dynamics on crops growth and yield under climate extremes; and ( 3) compare the effectiveness of different rainfed agricultural systems on r educing agricultural drought severity. To fulfill the knowledge gap 2 , the following objectives were studied: (1) evaluate the effects of climate variability on farm net returns under different production and treatment systems ; and (2) assess the risk lev el for the adaptation of different production systems under different treatments . To ascertain the knowledge gap 3 , the following objectives were examined: (1) a ssess recharge , groundwater table, and soil moisture variabilities for the long - term corn - soybean - wheat rotation under the conventional and the no - till practices at a watershed scale ; ( 2) e stimate yields and net returns under the conventional and the no - till practices within a large , diverse watershed; ( 3) e valuate the overall changes in resiliency as affected by the adaptation of no - till for row crop agriculture in the watershed . 4 2. REVIEW OF LITERATURE 2.1 Overview Global environmental changes pose far - reaching challenges to our food production systems ( Campbell et al., 2016; Myers et al., 2017; Wheeler and von Braun, 2013) . With the increasing frequency of climatic extremes, farmers face a daunting task to maintain their crop productivity. Impacts of these extreme climatic events disproportionately affec t the rainfed agricultural systems because of its total dependency on precipitation to rebuild the soil moisture required for crop growth (Kuwayama et al., 2019; Minoli et al., 2019; Rost et al., 2009; Sweet et al., 2017) . Meanwhile, the conventional pract ices that are commonly implemented to manage rainfed agricultural systems have become more unreliable in certain regions (e.g., Midwestern United States ) to climate change and variability. In fact, conventional practices are known to exacerbate extreme events via contributing to global environmental changes (Bennett et al., 2014; Lesk et al., 2016) . To overcome these challenges, alternative agricultural practices have been introduced as agronomic interventions to improve the resilience of crop production systems (Branca et al., 2013; Michler et al., 2019) . However, the resilience of these management practices for climate risk management in rainfed agricul ture should be quantified using appropriate set of metrics to assess their benefits. Therefore, the goal of this review is first to explore the challenges and opportunities in rainfed agricultural systems, then to understand the concept of resilience while learning how the resiliency can be measured, and finally to identify the knowledge gaps within the context of managed rainfed agricultural systems. 5 2.2 Rainfed Agriculture Agriculture can be classified into two types, namely, rainfed agriculture and irri gated agriculture, depending on the source of water used to grow crops. Rainfed agriculture entirely relies on infiltrated precipitation (predominantly rainfall) for crop production, while irrigated agriculture depends on full or supplementary irrigation. Globally, rainfed agriculture accounts for 80% of the cultivated lands and contributes to nearly 60% of total food production in a wide range of production systems (FAO, 2017) . Furthermore, the majority of the cereals are cultivated as rainfed crops. The p ercentages of corn, wheat, rice, and other coarse grains produced in rainfed systems are 82%, 66%, 40%, and 86%, respectively (Rosegrant et al., 2002) . Regionally, rainfed farmlands account for more than 95% in sub - Saharan Africa, almost 90% in Latin Am erica, 60 - 65% in Asia , and 75% in East and North Africa (Wani et al., 2009) . Meanwhile, in the United States, about 94% of farmland acres are under rainfed agriculture (USDA, 2014) . The productivity of rainfed agriculture in temperate regions such as in Eu rope and North America is much higher than the productivity in humid and dry tropical regions. (FAO, 2011) . This discrepancy in productivity is associated to the variations in soil fertility, climate, and biotic stressors (Rosenzweig and Liverman, 1992) . 2.2.1 Global Challenges in Rainfed Agriculture Water stress is the major challenge in rainfed agriculture systems. Meanwhile, since producers do not have a control over the timing and amount of the precipitation , the water use efficiency of rainfed systems is generally low (Rao et al., 2015; Wani et al., 2009) . Other constraints that substantially reduce the productivity of rainfed agriculture are soil degradation, soil organic matter depletion, soil erosion, nutrient deficiencies, environmental pol lution, low external inputs, low investment capacity, and poor market linkages (Rao et al., 2015; Wani et al., 6 2009) . Moreover, increase in rainfall variability (due to climate change) is becoming another important challenge in rainfed farming systems (Kas sie et al., 2014) . 2.2.2 Rainfed Agriculture in the Midwestern United States Rainfed agriculture systems in the Midwestern United States are highly productive than the rainfed systems in Africa, Asia, and Latin America (Rosegrant et al., 2002) . Moreover , they are ecologically diverse, practice a high level of farm technology, and receiving annual precipitation of greater than 500 mm (Franzluebbers et al., 2011) . In this region , the rainfed production systems are economically important, and most of the corn and soybean crops , are produced either as a monocrop or in rotation (Hatfield, 2012) . The productivity of rainfed crops steadily increased in the past as a result of genetic improvements (closer to 70%) and management intensification (closer to 30%) such as plant populations, row spacing, soil fertility, cover crops, rotation , farm mechanization and advancements in crop protection (Franzluebbers et al., 2011; Hatfield et al., 2018) . Moreover, with the development of herbicide - resistant crop varieties, adoption of conservation agriculture practices such as no - tillage has been substantially increased to counteract with the problems of soil erosion, nutrient leaching , runoff, and yield instability (Franzluebbers et al., 2011) . 2.2.3 Challenges of Rainfed Agriculture in the US Midwest Regardless of the overall increase in productivity, crop yields in the US Midwest have shown to be vulnerable to interannual variability in the climate (Hatfield et al., 2018; Hatfield, 2012; Ortiz - Bobea et al., 2018) . This phenomenon can be explained by the fact that the water availability for rainfed agriculture is determined by the seasonal characteristics of precipitation (i.e., intensity and frequency) and its interaction with the soil - plant - atmospher e continuum (Rost 7 et al., 2009) . Therefore, rainfed agriculture is much vulnerable to the effects of climate variability and extremes. Changes in the Midwest regional climate associated to the anthropogenic activities resulted in increased frequencies of extreme climatic events such as droughts, floods, and heatwaves (Andresen et al., 2012; Dai et al., 2016; Pryor et al., 2014) . For example, the 2012 Midwest drought devastated major crops and the economic base of rainfed farms in this region (Fuchs et al., 2015; Mallya et al., 2013; Otkin et al., 2016) . As a result of this drought, corn yield in the US has fallen for three consecutive years (2010 - 2012) for the first time since 1928 - 1930 (Rippey, 2015) . These climate extremes will continue to have substantia l impacts not only on the national economy but also on international trade (Boyer et al., 2013) , since the share of Midwestern region on U.S agricultural exports are relatively large, especially for corn and soybean. Nitrate leaching is identified as a maj or threat to environmental quality and human health (Bowles et al., 2018) . Since intensive agricultural management is practiced in this region, nitrate leaching has been found to be another important challenge in this system (Hussain et al., 2019; Martinez - Feria et al., 2019) . However, the adoption of conservation agriculture practices has been shown to minimize the leaching losses of nitrogen and increase the nitrogen use efficiency of crops (Palm et al., 2014; Syswerda et al., 2012) . 2.3 Resilience in Rainfed Agriculture Resilience was first defined by Holling (1973) in the context of ecological systems . Accordingly, resilience is the capacity of a system to absorb disturbance s and reorganize while changing. Thus, it can still maintain es sentially the same function, structure, identity, and feedbacks (Holling, 1973; Walker et al., 2004) . In rainfed agriculture, resilience can be defined 8 as the ability of a rainfed system to maintain its structures and patterns of behavior in the face of st ressors such as climate perturbations so it can continue to provide its services and desirable outcomes such as crop growth, yield and other ecosystem services (Tendall et al., 2015; Urruty et al., 2016) . Resilience also refers to the ability of an agricul tural system to develop capacities to cope and adapt to a new condition (e.g., climate shocks and extremes) via appropriate management practices (Bousquet et al., 2016) . Since soil moisture is the key determinant of productivity in rainfed agriculture, To w et al. ( 2011) argued that the ability of a system or its components to recover from water stress could also be considered as a measure of resilience. This could be achieved by implementing management interventions that can alleviate the exposure and seve rity of water stress in a way that an extreme climatic event shall not reduce the crop yields significantly. 2.4 Intervention s to Improve Resilience in Rainfed Agriculture I nterventions to enhance the resilience in rainfed agriculture can be broadly categ orized as agronomic interventions , genetic interventions, and decision support systems. As climate change is already hampering agriculture around the world, these interventions are also known to support the key objectives of climate - smart agriculture (Lipp er et al., 2014) . Below, each of these interventions is described in detail. 2.4.1 Agronomic Interventions Agronomic interventions may include crop, soil, and water management options aimed at improving resilience in agricultural systems. Agronomic interventions such as adjustment of planting and harvesting times, altering fertilizer and water application rates (Howden et al., 2007; Nouri et al., 2017; Rurinda et al., 2015) , mulching (Erenstein, 2003; Qin et al., 2015) , crop diversification and agroforestry (Altieri et al., 2015; Gan et al., 2015; Lin, 2011; Mbow et al., 9 2014) and adoption of conservation agricu lture (Delgado et al., 2013; Michler et al., 2019) have varying potentials to build resilience in rainfed agriculture. 2.4.2 Genetic Interventions Genetic interventions involve the develop ment of n ovel crop germplasm s with improved tolerance to environment al stresses such as drought and heat and/or genotypes with phenological adjustments so that they can escape and /or avoid overlapping with the occurrences of such stress events (Ainsworth and Ort, 2010; Ceccarelli et al., 2010; Davies et al., 2011) . The int roduction of these crop genotypes with improved stress tolerance has substantially increased the climate - resilience of major crops such as corn and soybean around the world (Cairns et al., 2012; Chapman et al., 2012; Sadok and Sinclair, 2011) . 2.4.3 Decisi on Support Interventions Decision support interventions mainly refer to the utilization of weather and seasonal climate forecast information to make agricultural decisions in the face of uncertainties in the upcoming days/season (Hansen et al., 2011; Klemm and McPherson, 2017; Meinke et al., 2006) . Forecast products can potentially influence many steps throughout the agricultural production such as selection of crop s and varieties , changing planting dates, input management, changing land management practices, adjusting marketing practices, and determining index - based insurance s to farmers thereby improve the climate - resilience in rainfed agriculture (Crane et al., 2011; Klemm and M cPherson, 2017) . With increasing real - time or near - real - time earth and environmental observations and modeling tools, decision support systems such as Regional Hydrological Extremes Assessment System (RHEAS) offers agricultural practitioners and policymake rs a framework to estimate onset, severity, recovery, and duration of regional 10 droughts and their impacts on expected crop yield outlooks in the vulnerable regions (Andreadis et al., 2017) . 2.5 Promising Agricultural Systems to Improve Resilience To improv e resilience, agricultural systems should be persistent, adaptive, and transformative at appropriate times and places. Persistence may include absorbing shocks of agricultural risks like weather extremes, invasive species, pest/disease outbreaks, and price fluctuations. Agriculture also needs to be adaptive to the long - term changes in the environment, such as the effects of climate change, soil degradation, and environmental pollution. Finally, a resilient agriculture system must be able to transform in to n ew modes of operation without harming the human and natural environment (Bennett et al., 2014) . Conservation and organic agriculture systems have many promising characteristics that could improve resiliency (Michler et al., 2019; Milestad and Darnhofer, 20 03) . 2.5.1 Conservation Agriculture Conservation agriculture co nsists of three agronomic principles , namely, minimum soil disturbance/no - till, permanent soil cover by crop residues/cover crops , and crop rotations (Hobbs et al., 2008; Palm et al., 2014) . Conservation agriculture was originated in rosion by wind and water (Baveye et al., 2011) and significantly expanded during mid - to late - herbicides, herbicide - tolerant crop species, and improved farm mechanization (Derpsch et al., 2010) . Since then, it has been rapidly adopted in North America, South America, and Australia primarily on large - scale mechanized farms, with heavy applications of herbicides to control weeds that are otherwise normally controlled by tillage (Palm et al., 2014) . Although it was originally designed for high - input agricultural systems in temperate regions, it has gained 11 momentum with smallholder farmers in Asia and Africa (Brouder and Gomez - Macpherson, 2014) due to its numerous benefits beyond soil erosion control . Globally, conservation agriculture was practiced in 180 Million hectares of cropland (12.5% of total cropland) by 78 countries during the period of 2015 - 2016 (Kassam et al., 2019) . During the same period, the adoption of conservation agriculture in the U nited States was 43.2 Million hectares, which corresponds 35% of total cropland (Kassam et al., 2019) . However , in the US Midwest, farmers practice various intensities of conservation agriculture with different combinations of agronomic principles (Denny e t al., 2019) , making them varying degrees of resilience to climate shocks and extremes. In comparison to conventional tillage, the no - till systems showed the highest level of resilience as a result of improved water availability and soil quality, that hel ps to avoid substantial reductions in crop yields during extreme climatic events (Delgado et al., 2013; Harrington and Tow, 2011; Lal et al., 2012; Michler et al., 2019) . In contrast, few other studies (e.g., Pittelkow et al., 2015 a ; Powlson et al., 2014) have shown yield reductions with no - till and limited potential to improve resilience . Interestingly, a recent global meta - analysis highlighted that the no - till system , when it is combined with other two conservation agriculture principles (residue retentio n and crop rotation), significantly increases crop yields , especially in rainfed crops (Pittelkow et al., 2015 b ) . This could be due to the capture of snow, reduction of runoff and soil evaporation (with the retention of crop residues), and creation of bett er soil structure and rooting patterns through crop rotations to store more water in the root zone (Franzluebbers et al., . 12 2.5. 2 Organic Agriculture The movement for organic agricultur e began in the first half of the 20 th century and have of Organic Agriculture Movements (IFOAM) institutionalized four principles of organic agriculture, namely , health, ecology, fairness, and care, with the main focus of protecting mankind and environment in the process of food production (Luttikholt, 2007) . Organic agricultural systems are well known to improve soil organic matter, especially when practiced for a long - term (Pimentel et al., 2005) . I t has been estimated that for every 1% organic matter, soil can hold approximately 10,000 - 11,000 liters of plant - available water per ha down to about 30 cm of soil depth (Gomiero et al., 2011) . Therefore, the water holding capacity of soils under organic agriculture is greater than the soils under conventional agriculture (Lotter et al., 2003; Mäder et al., 2002) . Moreover, organic agriculture has shown number of promising features to build re silience and reduce environmental impacts in various food systems thus serves as climate - friendly farming (Milestad and Darnhofer, 2003; Scialabba and Müller - Lindenlauf, 2010; Tuomisto et al., 2012) . Even though the yield gap between organic and high input conventional agriculture is larger mainly due to nutrient limitation (De Ponti et al., 2012) , organic systems achieve comparable or even higher yields in many developing countries, as compared to the current conventional systems (Scialabba and Müller - Lind enlauf, 2010) . In contrast, the production cost of certified organic farms is generally higher as the producers spend more on labor, insurance, certification, and marketing costs than the conventional farmers (Uematsu and Mishra, 2012) . 13 2.5.3 Reduced Input Agriculture Reduced input agriculture or low external input agriculture is aimed at minimizing the environmental pollution of conventional agriculture practices (Odum, 1987) . Therefore, direct or indirect use of petrochemicals - based inputs (e.g., fertili zers, pesticides) are substantially reduced in this system in comparison to a conventional system (Buttel et al., 1986) . Reduced input agriculture can be considered as a step toward sustainable agriculture since it improves ecological processes within the agricultural systems (Stinner and House, 1987) . However, it has limited potential to improve the overall profit of agricultural productions (Kessler and Moolhuijzen, 1994) . To improve the productivity of reduced input agriculture, several management practi ces were introduced such as integrating leguminous cover crop and intercropping (Robertson et al., 2014) . 2.6 Metrics of Resilience in Agricultural Systems Appropriate metrics are required to quantify the resiliency of promising agricultural systems. The m easure of resilience can be used to maintain or shift a system towards a more desirable and sustainable state, track thresholds of potential concerns, and help with assessments on how the system is being managed and can be further improved (Quinlan et al., 2016) . Meantime, evaluation of resilience often involves a holistic approach that incorporates social, economic, and environmental dimensions of resilience (COSA, 2017) . Because of the complexity and interactions in these three dimensions, food system res ilience is generally assessed qualitatively (Toth et al., 2016) . However, qualitative assessments are case - specific, and their applicability across scales varies based on the assumptions. Various organizations and individuals have developed and employed se veral tools to measure the resilience of farming systems in many regions of the world (Douxchamps et al., 14 2017) . These tools have been generally applied to large socioeconomic units (e.g., households/ communities/ administrative units/ national scale). For example, the Food and Agriculture Organization of the United Nations (FAO, 2016) developed the Resilience Index Measurement and Analysis Model (RIMA), which has been applied in many African countries to measure the climate - resilience of agricultural commu nities (Serfilippi and Ramnath, 2018) . Few other examples are; Community Based Resilience Assessment (CoBRA) developed by United Nations Development Program (UNDP, 2013) , Self - evaluation and Holistic Assessment of climate Resilience of farmers and Pastora lists (SHARP) adopted by the Food and Agriculture Organization of the United Nations (Choptiany et al., 2017) , Community - based Risk Screening Tool - adaptation and Livelihood (CRiSTAL) developed by International Institute for Sustainable Development (IISD, 2 014) , Climate Vulnerability and Capacity Analysis (CVCA) established by Care International (Care, 2009) and Resilience, and Adaptation and Transformation Assessment Framework (RATALF) developed by The Commonwealth Scientific and Industrial Research Organiz ation . These tools consist of a combination of measurement indices to quantify the resilience of . Generally, these metrics include the means and variance of agricultural production (Di Falco and Chavas, 2008; Zampieri et al., 2020) , crop yields (Birthal et al., 2015; Li et al., 2019; Martin and Magne, 2015) , profit (Browne et al., 2013; Komarek et al ., 2015; Seo, 2010) , revenue (Kandulu et al., 2012; Rigolot et al., 2017; Tibesigwa and Visser, 2015) , labor productivity (Komarek et al., 2015) , crop failure (Jones and Thornton, 2009) , dietary diversity (Dillon et al., 2015) , farming risks (Komarek et al ., 2015) , agricultural gross domestic product (Hsiang and Jina, 2014) and expenditure for food consumption/food security (Alfani et al., 2015) . In the context of agricultural development, 15 climate - resilience was monitored and evaluated using income, food av ailability, land area, and labor force (Douxchamps et al., 2017) . Nevertheless, these indicators are often used at a large scale and show non - linear responses to climate variability depending upon various characteristics of farms and farmers (Di Falco and Chavas, 2008; Tittonell, 2014) . Although there is a growing interest in the concept of resilience in the face of global environmental changes, there is no pragmatic guidelines available to quantify the overall resilience of farming systems (COSA, 2017) . Mo reover, existing tools and frameworks often fail to capture the spatial and temporal dynamics of resilience at various scales (Dixon and Stringer, 2015; Douxchamps et al., 2017) . Therefore, new metrics are needed to address the complexity in agricultural s ystems while being simple enough to be measured at different scales. 2.6.1 Soil Moisture Soil moisture primarily controls the productivity of crops in rainfed agricultural systems (Jägermeyr et al., 2016) , and is considered among 27 key indicators of clima te - resilience proposed by the committee on sustainability assessment (COSA, 2017) . Precipitation is the source of soil moisture in rainfed farming; therefore, the seasonality of rainfall, such as onset, intensity, and frequency, determines the availability of water resources to crops (Robinson et al., 2019) . Moreover, soil moisture affects several hydrological processes such as evapotranspiration, runoff, and recharge to aquifers hence considered as one of the components of hydrological cycle (Robinson et a l., 2019) . Meanwhile, irregular rainfall associated to the climate variability has reduced the plant available water, and water stress has been identified as an important abiotic stress that significantly affects crop yields (Dinar et al., 2019; Ortiz - Bobe a et al., 2019; Rossato et al., 2017) . That is why Rockström et al. ( 2010) argued that any investment that focuses on the improvement of rainfed agriculture should primarily consider water management. 16 Spatiotemporal dynamics of soil moisture should be tak ing into account in order to evaluate soil moisture as a metric of resilience. Mean relative difference (MRD) of soil moisture, indices of the spatiotemporal var iabilities of soil moisture (Liu et al., 2018; Vachaud et al., 1985) . These indices were commonly used to identify temporally stable or representative locations to employ soil moisture monitoring equipment (Barker et al., 2017; Brocca et al., 2010; Starks et al., 2006; Zhou et al., 2007) and/or to validate remotely sensed soil moisture products (Cosh et al., 2008; Jacobs et al., 2004; Wagner et al., 2008) . Moreover, these indices were also used to study the spatiotemporal dynamics of soil moisture in hillslope (Gao et al., 2016; Liu et al., 2018) or under diverse land uses (Hu et al., 2010) or in different depths of the soil (Gao and Shao, 2012; He et al., 2019) . In contrast, the applicability of these indices to quantify the resilience of different a gricultural systems has not been studied before. 2.6.2 Drought Indices Shortage of soil moisture to the crops for a few weeks due to the absence of rainfall can lead to the development of agricultural droughts (Esfahanian et al., 2017) . Drought affects both quantity and quality of crop yields depending on the following factor s; timing of drought in relation to the crop's growth stage , reliability of water resources, vulnerability of cropping system, and socioeconomic conditions of farmers (Rey et al., 2017) . Hence, drought indicators can be used as metrics of resilience in agr icultural systems and building drought resilience is essential to sustain agricultural production in the United States (Brusberg and Shively, 2015) . The agricultural drought indices, which are commonly used are presented in Table 2.1 (Esfahanian et al., 20 17; Moorhead et al., 2015) . 17 Table 2. 1 . Commonly used available agricultural drought indices. Agricultural Drought Index Inputs Required Reference Soil Water Deficit Index Soil moisture, field capacity, wilting point Martínez - Fernández et al., 2015 Soil Moisture Deficit Index Soil moisture Narasimhan and Srinivasan, 2005 Evapotranspiration Deficit Index Actual Evapotranspiration (ET), potential evapotranspiration (PET) Narasimhan and Srinivasan, 2005 Palmer Drought Severity Index (PDSI)/ Palmer Z i ndex Precipitation, temperature, available soil water capacity, and latitude of the location Palmer, 1965; Alley, 1984 Reconnaissance Drought Index Actual ET, PET Tsakiris and Vangelis, 2005 Accumulated Drought Index Precipitation, reference ET (ET 0 ) Sivakumar et al., 2011 Relative Water Deficit Actual ET, PET Sivakumar et al., 2011 Crop Moisture Index Same as PDSI Palmer, 1968 Vegetation Condition Index Satellite images Kogan, 1995 Standardized Vegetation Index Remote sensing data Peters et al., 2002 Soil Water Deficit Index (SWDI): SWDI can be used to quantify the agricultural drought when temporal soil moisture data is available (Martínez - Fernández et al., 2015) . SWDI can be calculated using the following formula: (2.1) w here , is the percentages of the volumetric soil moisture (cm 3 /cm 3 ) , is the percentage of the field capacity (cm 3 /cm 3 ) of the soil and is the percentage of the permanent wilting point (cm 3 /cm 3 ) . When SWDI is positive, the soils have excess water; when it equals zero, the soil is at the fi eld capacity (i.e., no water deficit). Negative values indicate drought, and the soil reaches the wilting poi nt - 10. At this point, the soil water content is below the lower limit of water available for plant uptake (Savage et al., 1996). 18 According to calculated SWDI values , drought severity in agricultural systems can be c lassified - - 2 > SWDI > - 5, as - 5 > SWDI > - - 10 SWDI (Martínez - Fernández et al., 2015) . Although the annual and monthly time scales of evaluation seem adequate to monitor the effects of meteorological and hydrological drought s (Sharma, 1997) , a shorter period of analysis is required to evaluate the effects of agricultural drought. A weekly period is commonly recommended for cropping systems as their growth stages show different levels of sensitivity to drought, and farmers require a sufficient window of time for management interventions such as irrigation planning (Purcell et al., 2003) . For this purpose, SWDI is appropriate to evaluate the effects of droughts in agricultural systems. Moreover, the SWDI is more comprehensive compared to the Soil Moisture deficit Index (SMDI) in capturing soil moisture deficit since it uses several soil parameters . In contrast, the SMDI only uses one parameter soil moisture (Esfahanian et al., 2017) . Soil Moisture deficit Index (SMDI): SMDI only requires the soil moisture in the root zone as input (Narasimhan and Srinivasan, 2005) . The weekly percentage of soil moisture deficit or excess can be calculated as follows : (2.2) (2.3) where , is the soil water deficit (%), is the mean weekly soil water (mm) in the soil profile, is the long - term median of available soil water (mm) in the soi l profile, is the long - term maximum of available soil water (mm) in the soil profile, and 19 is the long - term minimum of available soil water (mm) in the soil profile. i denotes the year while j denotes the week. Using the above equation, the seasonality inherent in soil water is removed, and the deficit values can be compared across seasons. The soil water deficit (SD) values during a week could range from 100 to +100, indicating very dry to v ery wet conditions, respectively. As the SD values for all the subbasins were scaled between 100 and +100, they are also spatially comparable across different climatic zones (Narasimhan and Srinivasan, 2005) . Thereafter, SMDI for any given week can be ca lculated by: (2.4) where, t is the time in weeks. SMDI during any week will range from - 4 to +4, representing the wors t drought and no drought conditions, respectively. Evapotranspiration Deficit Index (ETDI) : ETDI is calculated using a procedure similar to the calculation of SMDI. However, a water stress ratio is initially computed rather than using the evapotranspirati on (ET) alone (Narasimhan and Srinivasan, 2005) . T he weekly water stress ratio ( WS ) is calculated as follows : (2.5) wher e, PET is the weekly potential evapotranspiration of the crop, and AET is the weekly actual evapotranspiration. Values of WS may range from 1 to 0, where 1 indicating zero actual evapotranspiration and 0 indicating the evapotranspiration occurring at poten tial rates. After calculating WS , the maximum, minimum , and median water stress es are used to calculate the weekly water stress anomaly (WSA). The WSA values will range from - 100 to +100 , with 20 negative values indicating dry conditions and positive values indicating wet conditions (Narasimhan and Srinivasan, 2005) . Under water - limited conditions, crop transpiration will not occur at the potential rates. Therefore, the ratio of actual ET to PET can indicate the stress on the plant , thus, ETDI may be more suitable to evaluate the impacts of agricultural drought than the indices that use precipitation. Meanwhile, the major limitation to this drought index is that the actual crop ET data may not often be measured in agricultural settings . Furthermore , th e PET must be available for the same crop as the actual ET. Although the r eference ET ( ET 0 ) can replace PET , ET 0 is only applicable for the reference crop , and d ifferent crops will have different relationships to ET 0 depending on their growth stage - specifi c crop coefficients (Moorhead et al., 2015) . Palmer Drought Severity Index (PDSI)/ Palmer Z index : As part of the original work done by Palmer in the early 1960s, the Palmer Moisture Anomaly Index ( Z Index ) is usually calculated on a monthly basis along wi th PDSI output as the moisture anomaly. The PDSI addresses two of the most elusive properties of droughts namely, intensity and their beginning and ending time (Alley, 1984) . However , this index is computationally intensive and less transparent compared to the SWDI, SMDI, and ETDI , because it requires a total of eight parameters such as precipitation, temperature, available soil water capacity , and latitude of the location of interest (Alley, 1984; Ficklin et al., 2015; Jacobi et al., 2013; Palmer, 1965) . Moreover, the Z - index has been used with a coarse spatial resolution of 7000 - 100,000 km 2 and monthly temporal resolution, while SMDI and ETDI were used with a finer spatial resolution (16 km 2 ) and weekly temporal resolution (Narasimhan and Srinivasan, 2005) . Reconnaissance Drought Index (RDI) : RDI was introduced by Tsakiris and Vangelis ( 2005) with the objective of developing an agricultural drought index accounting for both 21 precipita tion and ET. RDI first calculates the aridity index using precipitation and potential evapotranspiration. The aridity index is calculated as the ratio between the long - term annual mean of precipitation and the annual mean of ET 0 . RDI will then be calculated as a standardized or normalized ratio between the cumulative precipitation and cumulative ET 0 for a given period (Tsakiris and Vangelis, 2005) . RDI can be also calculated for various time scales; however, ET data may not be avai lable in all the locations, and an accurate estimation of ET requires numerous inputs (Moorhead et al., 2015) . Accumulated Drought Index (ADI) : ADI was formulated by the Integrated Center of Agrometeorological Information in Brazil (Sivakumar et al., 201 1) . ADI is calculated as follows: (2.6) where, DI is derived from the relationship between precipitation and ET as detailed in Sivakum ar et al. ( 2011) , n is the number of time periods, and N is the number of periods with less than 10 mm of precipitation. Even though this index uses both precipitation and ET to assess the drought conditions, it was not well - established and verified in the drought assessment literature (Moorhead et al., 2015) . Relative Water deficit (RWD): RWD is calculated as the ratio of actual ET to PET (Sivakumar et al., 2011) . (2.7) The RWD will be zero if actual ET is occurring at the rate of PET, that indicates no water stress to crops. RWD can be calculated for different time scales; no netheless, the actual ET data 22 is difficult to obtain, and it varies for different crops. This is the major drawback of this index to implement it across different landscapes. 2.6.3 Farm Economics Farm profit is the key concern of producers since the liv elihood of most of the farmers is primarily depends on their farming enterprises. Farm profits are generally measured as the earnings before interest taxes and amortization. Additionally, net farm revenue, operating profit margin ratio, rate of return on f arm assets, and rate of return on farm equity can be used to quantify the profitability of agricultural enterprises (Langemeier, 2016) . Farmers depend on net farm income for; living expenses, to pay their debts, and to purchase new assets to continue their farming operations. The United States had 6.8 million farms in 1935; however, according to a recent survey conducted in 2019, there were 2.02 million farms located in 897 million acres of land, with an average farm size of 444 acres (ERS, 2020) . This decline in the number of farms highlights the increasing agricultural productivity and opportunities for nonfarm employment. Therefore, agricultural production in the United States is currently concentrated to a smaller number of large, specialized farms, mostly in rural areas. Te chnological advancements in crop and animal genetics, agrochemicals, farm machinery, and management were the driving force for the growth in agricultural outputs (ERS, 2020) . The net farm income of the whole country is calculated by subtracting total farm expenses from gross farm income. Net farm income (inflation - adjusted) was forecasted to increase by 21.7% in 2020, to 102.7 billion US dollars by the United States Department of Agriculture. Total cash receipts from major row crops corn, soybean, and wheat were 50.1 billion, 34.2 billion, and 9.0 billion US dollars, respectively, in 2019. The median of the total 23 household income from all farming households was 72,481 US dollars in 2018. In contrast, the annual farm sales of around half of the US farms were under 10,000 US dollars, and these farm families mostly rely on off - farm employment for their family income. Meantime, large - scale farms had an annual farm income of 348,811 US dollars (ERS, 2020) . Farm economics is one of the three pillars of sustainable agriculture (Pham and Smith, 2014) . Crop productivity, profitability, and stability are the indicators used to gauge sustainability in farming systems (Pham and Smith, 2014) . For the households depending on farming for their livelihood, the gross farm inco me should be able to compensate both fixed and variable costs involved in their farming operation. Meanwhile, most of the developing countries have tropical climates and smallholdings, where farm families cannot meet their income needs alone from farming. Therefore, they need to seek off - farm income to maintain their farming operations (Li et al., 2012) . According to Tey and Brindal ( 2015) , the farm profit is determined mostly by the scale of operation, operational efficiency, and output prices. Improving t he operational efficiency through precision management of resources such as irrigation water and nutrients has substantially improved the profitability in farming systems (McConnell and Burger, 2011; Sapkota et al., 2014; Spencer et al., 2019) . Conservatio n management systems also have improved the profitability of farming systems (Jat et al., 2014; Ngwira et al., 2013) ; however, some other studies have shown mixed results (Mafongoya et al., 2016; Plastina et al., 2020) . he challenges of climate extremes, such as implementing any resilient practices, maybe costlier and riskier than the business - as - usual practices; therefore, the long - term profitability is the most important factor that determines adaptation of such practices (Kumar et al., 2016; Mausch et al., 2017; Sain et al., 2017) . 24 Moreover, farmers demonstrate different preferences toward risks, dep ending on the socioeconomic factors. (Brink and McCarl, 1978; Lu et al., 2003) . For example, a highly profitable production system may also be highly risky because of the variabilities on net revenues and/or market demands, thus may not be preferred by a r isk - averse farmer. Farm profitability also determines the costs and benefits of farm insurance policies, which are important tools for climate risk management (Annan and Schlenker, 2015; Tack and Ubilava, 2015) . Resilience in terms of farm economics can be measured using both the mean and volatility of the expected net revenues or profitability, where a system with higher mean and lower volatility can be taken as a relatively resilient system (Abson et al., 2013; Browne et al., 2013) . The volatility, as it measures the variations in farm income, can also capture the risk associated with the system (Browne et al., 2013) . It is essential that the producers make informed decisions considering the most appropriate risk management strategies tailored to their ada ptation systems and the probability of risks (Vigani and Kathage, 2019) . Conversely, an appropriate cost - benefit analysis or enterprise budgeting is required to reasonably estimate the net return in farming enterprises (Plastina et al., 2020) . To perform a n accurate enterprise budgeting, correct data on the quantities of inputs, outputs, and associated costs during the time of operation are essential for the period of analysis, which may not adequately available from farm records. 2.6.4 Ecosystem Services E cosystems are a vital part of the environment, and all the benefits that humans receive from the natural environment are called ecosystem services (Costanza et al., 1997; Mengist et al., 2020) . Agricultural ecosystems are managed to primarily produce food, fodder, fiber, and fuel; 25 however, they also offer various ecosystem services that are essential for the functioning of the environment (Dale and Polasky, 2007; Swinton et al., 2007; Tancoigne et al., 2014; Wood et al., 2015) . According to the Millennium E cosystem Assessment, ecosystem services can be broadly grouped into four categories, namely, provisioning services, regulating services, supporting services, and cultural services (Fisher et al., 2009; MEA, 2005) . Provisioning services include any goods t hat are directly consumed by people, such as food, water, energy, raw materials, timber, ornamentals, genetic and medicinal resources. Regulatory services comprise carbon sequestration, climate regulation, mitigation of floods, control of soil erosion, man agement of pest/disease outbreak, purification of water, air, and soil, decomposition, and detoxification. Supporting services involve natural processes such as photosynthesis, cycling of water and nutrients, soil formation, pollination, and biodiversity. Tangible benefits such as recreation, education, inspiration, ecotherapy, and other spiritual values of the ecosystems come under the cultural services (Garbach et al., 2014; MEA, 2005; Wei et al., 2021) . Although the food production systems have the poten tial to offer numerous ecosystem services, the current trend of conventional agriculture deliberately aimed on maximizing few provisioning services (e.g., food, fodder, fuel) with the simplification of agricultural landscapes rather than harnessing a range of ecosystem services (Bommarco et al., 2013; Gaba et al., 2015; Robertson and Swinton, 2005) which in turn affects the resilience and sustainability of the agricultural systems. Remarkably, ecosystem services and system resilience are strongly related to each other. Resilient ecosystems can withstand or quickly recover from environmental disturbances such as climate perturbations and continue to supply ecosystem services. On the other hand, the 26 ecosystem with lower resilience may be vulnerable to extreme events (Biggs et al., 2012; Fedele et al., 2017; Montoya and Raffaelli, 2010) . Therefore, the loss of ecosystem resilience could compromise ecosystem services that are indispensable for agricultural systems, especially in the face of increasing climatic ex tremes (DeClerck et al., 2016; El Chami et al., 2020; Swift et al., 2004) . For example, recharge is an important water - related ecosystem service in agricultural systems (Coates et al., 2013; Serna - Chavez et al., 2014) . Recharge is the deep percolation of water leaving the vadose zone from agricultural farms and contribute to replenishing the aquifers and the groundwater - dependent ecosystems such as wetlands and streams (Gordon et al., 2010; Sampath et al., 2015) . A significant portion of the land area is u tilized to grow agricultural crops; thus, large quantities of recharge to groundwater occurs from the cropland areas. In the meantime, groundwater resources are highly exploited to cater to the requirements of increasing human population. Extraction of gro undwater above the rate of local recharge has significantly reduced the groundwater storage in the aquifers that has affected the baseflow to streams, groundwater - fed wetlands, and other dependent habitats and species (Dalin et al., 2017; Scanlon et al., 2 012; Wada et al., 2010) . Recharge may behave differently in response to various tillage practices 2011; Owens, 1994) . Since reduced tillage or zero tillage is one of the important principles of conservation agriculture, it is important to understand how recharge responds to such tillage regimes when conservation practices are implemented at regional scales. If conservation tillage could increase the recharge and thereby improve groundwater levels, it will increase the resiliency in the face of declining groundwater resources. 27 2.7 Modeling Resilience at a Large Scale Resilience in agricultural systems not only confined to farm scales but also expand across regional scales (Bailey and Buck, 2016; Scherr et al., 2012) . To evaluate the resilien ce at larger scales, appropriate metrics should be measured. However, it is often impossible to measure the resilience metrics at regional scales due to the lack of resources to collect data. Meanwhile, modeling approaches can be implemented to solve this problem. Model is the representation of a conceptualized system, which consists of a set of equations to mimic the system's behavior. Models assist in understanding and explaining the performance of the systems at different scales. Depending on the struct ure, functions, and purposes, models can be classified as statistical, mechanistic, deterministic, stochastic, dynamic, static, process - based, descriptive, and explanatory models (Murthy, 2004) . Statistical models are based on historical observations, and the relationships between inputs and outputs are developed using statistical techniques (Jones et al., 2017a) . Mechanistic models explain not only the relationships between model parameters but also detail the mechanisms of the modeling process. Deterministic models simulate predictions on dependent variables without any associated probability distribution, varia nce, or random element, while a probability element is attached to each output in stochastic models. Dynamic models represent the behavior of a system; therefore, time is included as a variable; however, static models do not include time as a variable; hen ce dependent and independent variables will have constant values over a given period of time. Process - based models are often computer models that use several mathematical equations to detail the processes occurring at different levels in a system. Descript ive models are used to describe the system without much emphasis on the mechanisms, 28 while explanatory models describe the mechanisms and processes that cause the functioning of the systems (Murthy, 2004) . Most of the crop and hydrological models that are u sed to evaluate the performances of agricultural systems are process - based models (Jones et al., 2017a; Siad et al., 2019) . Therefore, these models can be implemented to produce resilience metrics such as crop growth, yield, revenues, and soil water balanc es at different scales ranging from field to region. 2.7.1 Crop Modeling Development of crop models to characterize the processes of crop growth began between 1950 - 1960 and currently achieved remarkable improvements in agricultural system models (Boote et al., 2013; Jones et al., 2017a) . Crop models can be broadly categorized into predictive and explanatory models (Di Paola et al., 2016) . Predictive models are used to predict the yield mostly based on empirical functions (i.e., statistical relationships) be tween environmental variables and crop yield using few easily derived parameters. Meanwhile, explanatory models are built by considering plant - environment interactions and ecophysiological processes (e.g., photosynthesis, leaf area expansion, biomass parti tioning) that govern the growth and yield formation of crops (Di Paola et al., 2016) . Explanatory models are also called process - based models and continue to be updated. Meanwhile, the number of parameters that are required to explain each process within t he model has been increased that ultimately elevate the complexity of these models compared to predictive models. Therefore, hybrid models (i.e., mixing both empirical and mechanistic approaches) are often used to evaluate the performance of different agri cultural systems (Di Paola et al., 2016) . Crop models require weather, soil, crop cultivar, and management information to simulate the processes such as photosynthesis, respiration, water and nitrogen balances, 29 phenology, partitioning, and senescence gener ally at a daily time scale (Boote et al., 2013; Jones et al., 2003) . Numerous crop models have been developed in various countries. Each of these models has its own strengths and weaknesses (Di Paola et al., 2016) . Some of the popular crop models are; Agri cultural Production Systems Simulator (APSIM) (Keating et al., 2003) , AquaCrop (Vanuytrecht et al., 2014) , CropSyst (Stöckle et al., 2003) , Daisy (Abrahamsen and Hansen, 2000) , Decision Support System for Agrotechnology Transfer (DSSAT) (Jones et al., 2003 ) , EPIC (Williams et al., 1984) , InfoCrop (Aggarwal et al., 2006) , MONICA (Nendel et al., 2011) , Oryza (Bouman et al., 2001) , System Approach to Land Use Sustainability (SALUS) (Basso et al., 2006) , STICS (Brisson et al., 1998) , and WOFOST (Diepen et al., 1989) . Among these crop models, DSSAT and APSIM operate as a platform for multiple crop modules for major food crops. For example, within the DSSAT framework, CERES - Rice, CERES - maize, CERES - wheat, and CROPGRO - soybean modules are available to model rice, maize, wh eat, and soybean crops, respectively (Jones et al., 2003) . DSSAT currently consists of process - based simulation modules for 42 major crops, and the APSIM has a unique capability of modeling intercropping systems (Hoogenboom et al., 2019; Jones et al., 2017 b; Keating et al., 2003) . DSSAT is the most widely used crop modeling system in the world. The applications of DSSAT include the evaluation of genetic improvement (Boote et al., 1996) , assessment of the impacts of climate change and variability (Fodor et al., 2017; Rosenzweig et al., 2014) , optimization of management practices such as tillage, water, and nutrients (Iocola et al., 2017; Joshi et al., 2017; Kropp et al., 2019; Liu et al., 2013; Malik and Dechmi, 2019; Roy et al., 2019) , evaluation of crop re sponses to environmental stresses (Liu et al., 2016; Saseendran et al., 2015) and yield gap analysis (Teixeira et al., 2019) . Meanwhile, DSSAT was applied for yield 30 forecasting, disease management, precision farming, decision support, and policy analysis in agriculture (Boote et al., 1996; Jha, 2019; Shelia et al., 2015; Thorp et al., 2008) . Meanwhile, the sequence modeling procedure in DSSAT enables us to simulate cr op rotations (Bowen et al., 1998; Liu et al., 2013; Salmerón et al., 2014) . Crop models have been successfully used to implement hypothetical experiments and/or to simulate field experiments to assess various genetic × environment × management (G×E×M) int eractions from field to regional scales (Adnan et al., 2019; Balboa et al., 2019; Eitzinger et al., 2017; Jin et al., 2019; Peng et al., 2020) . Increasing access to biophysical data from remote sensing and integration of data assimilation techniques have s ubstantially improved the scale of implementation of crop models (Jin et al., 2018) . Therefore, crop modeling can help to evaluate the resilience of different production systems and to quantify their responses to different adaptation measures (Challinor e t al., 2018; Rötter et al., 2018) . 2.7.2 Hydrological Modeling Hydrological models are used to predict and manage the distribution and fluxes of water as a function of various climate, soil, and physiographical characteristics that are used to describe the system within the model (Siad et al., 2019) . Hydrological models that describe the land surface processes are called rainfall - runoff models, and those that describe the processes in the saturated zone are referred to as groundwater models. A large number of hydrological models are currently available to model hydrological processes from the scale of small catchments to the globe (Siad et al., 2019) . Some of these models are; Variable Infiltration Capacity (VIC) model (Liang et al., 1994) , MIKE - SHE (Refsgaa rd and Storm, 1995) , Hydrologiska Byråns Vattenbalansavdelning (HBV) (Lindström et al., 1997) , Soil and Water Assessment Tool (SWAT) (Gassman et al., 2007) , Water Evaluation and Planning System (WEAP) (Sieber and 31 Purkey, 2015) , Soil Water Atmosphere Plant (SWAP) (Van Dam et al., 2008) and HYDRUS . MODFLOW (Harbaugh, 1995) and Interactive Groundwater (IGW) (Li and Liu, 2006) are the commonly available hydrological models to simulate groundwater systems. Each model has its unique feature s and applications, and their choice and implementation depend on data availability and modeling objectives. Furthermore, hydrological models can be coupled with crop models to better evaluate the water movement through the plant - soil - atmosphere continuum (Siad et al., 2019) . Groundwater models use the finite difference approximation of the governing partial differential equation to solve flow conditions in confined and unconfined aquifers as follows (Heath, 1983) : (2.8) where, S s is the specific storage coefficient, h is the hydraulic head, t is time, K is the saturated hydraulic conductivity, is the mathematical gradient opera tor, and q is the net flux (i.e., source (+) or sink ( - ) term). Regional groundwater modeling requires numerous data inputs, which include aquifer properties, lithology, hydrographic information, land use, climate data, topography, recharge, and static wa ter levels (Liao et al., 2020) . Calibrated groundwater models are very useful tools to evaluate the impacts of land use changes and pumping on the groundwater flow dynamics (Sampath et al., 2015) , groundwater quality (Curtis et al., 2018) , and contaminant transport in the saturated zone (Hadley and Newell, 2014) . 32 2.8 Adaptation of Resilient Practices The adaptation usually refers to the process, action, or outcome of farmers and farming systems to better cope with the changing condition, stress, hazard, or risk (Smit and Wandel, 2006) . Climate change adaptation in the context of agriculture could be identified as the adjustment of agronomic practices, agricultural processes, and capital investments in response to extreme events (Ainsworth and Ort, 2010) . Dep ending on the timing of implementation, adaptations can be proactive (anticipatory) or reactive. Based on their degree of spontaneity, they could be either autonomous or planned. Adaptation in farming communities is closely related to their respective adap tive capacity and vulnerability. The concepts of adaptation, adaptive capacity, vulnerability, resilience, exposure, and sensitivity are closely related and have wide applications to global environmental changes and its impacts on agricultural systems. Vul nerability is determined by both the differential exposure and sensitivity of farming communities to climate change and to the adaptive capacities of those communities to deal with the effects associated with exposures. Exposure and sensitivity signify the likelihood of experiencing detrimental effects, while the livelihood characteristics of agricultural systems influence its sensitivity to such exposure (Smit and Wandel, 2006) . Agriculture has evolved through centuries in different regions of the world; t herefore, it has an immense diversity of management practices to be used for adaptation to the impacts of climate variability. Such practices may involve adjustment of the timing of planting and harvesting, replacing crop species and cultivars with more ap propriate thermal duration or stress tolerance, developing new crop varieties with improved traits, precision management of agricultural inputs such as water and nutrients and improvement of climate forecasting to reduce the production risks (Delgado et al ., 2013; Howden et al., 2007; Minoli et al., 2019) . 33 Many agricultural systems can provide both adaptation and mitigation benefits synergistically if they are designed and managed appropriately in a large landscape (Harvey et al., 2014) . This is because mos t of the agricultural adaptation options for climate extremes have positive impacts on mitigation. These include measures that reduce soil erosion, reduce leaching of nitrogen and phosphorous, conserving soil moisture, increasing the diversity of crop rota tions, and modification of microclimate to reduce temperature extremes (Smith and Olesen, 2010) . Cooper et al. ( 2009) argued that, the development of temperature - adapted varieties, together with improved farm management practices, could result in practical ly complete mitigation of the negative impact of temperature rise and rainfall variability. According to Smit and Skinner ( 2002) , the adaptation strategies of farming communities in response to climate change can be categorized into four groups namely, technological developments (e.g., crop improvement, weather and climate information systems, and resource management innovation), government programs and insurance (e.g., agricultural subsidies and livelihood support programs, farm insurance, and r esource management programs), farm practices (e.g., farm production, irrigation, and timing of operation) and farm financial management (e.g., crop insurance, crop shares, income stabilization, and household income). The ability of farmers to adapt to clim ate change will depend on several socioeconomic factors. Land tenure, farm size, ownership of agricultural assets and livestock, gender, age, and availability of inputs decide the technology adoption under climate change (Asfaw et al., 2014) . In Africa, th e adaptation of conservation farming practices (minimum tillage, permanent soil cover, and crop rotations) determine by the availability of extension services and rainfall variability (Arslan et al., 2014) . This shows the role of agro - ecological and socioe conomic 34 constraints in explaining adoption, as well as the potential role and effectiveness of outreach programs to support it. Recent survey studies among the farmers in the Midwestern United States indicate that most of the producers recognized the chang ing climate (Arbuckle et al., 2013; Doll et al., 2017; Mase et al., 2017) and have, therefore, implemented different climate adaptation measures (Denny et al., 2019; Morton et al., 2015) . For example, farmers who better understand the causes and casualties of climate change are increasingly using climate information, conservation practices, crop insurance, and other alternative options for mitigating risks to climate change (Arbuckle et al., 2013; Haigh et al., 2015; Mase et al., 2017; Morton et al., 2017) . However, recognition of the threats due to climate change and the rate of uptake of climate - resilient practices still inadequate compared to the rate of changes in climate variability (Church et al., 2017; Lemos et al., 2014) . This is because there are ma ny barriers to adaptation and operating at different levels. Barriers are the obstacles that hinder the planning and implementation of adaptation measures. Generally, farmers are constrained by financial barriers, socio - cultural barriers, institutional bar riers, technological barriers, and a lack of information on the impacts of climate extremes (Antwi - Agyei et al., 2015) . To overcome these barriers, a comprehensive and dynamic policy approach is required considering all the aspects ranging from individual farmer awareness to the establishment of efficient markets (Howden et al., 2007) . Moreover, there is a need to engage stakeholders at different levels from different sectors with diverse and often contesting types of expertise, experience values, and inter ests (Vervoort et al., 2014) . 2.9 Summary Overall, increasing climate extremes have become the major challenge for the rainfed agricultural systems in the Midwestern United States. Although farmers have realized the causes 35 and casualties of these extreme climatic events, the uptake of adaptation measures was not adequate to face the challenges. Alternative agricultural practices such as conservation agriculture and organic agriculture have been introduced and endorsed for adaptation to counteract the effec ts of climate change and variability. However, the resilience of these promising management systems has not been adequately quantified due to the lack of metrics to capture the temporal variabilities both at field and regional scales. Therefore, carrying o ut research studies to explore the metrics that are practical and able to quantify the resilience of different rainfed systems would aid in identifying appropriate options for climate risk management in agriculture. 36 3. INTRODUCTION TO METHODOLOGY This dissertation comprised of three research studies that are critically important to quantify the level of resiliency in agricultural settings. The first paper, Evaluating the Applicability of Soil Moisture - based Metrics for Gauging the Resiliency of Rainfed Agricultural Systems systems at the field scale. Soil moist ure is the major limiting factor of crop productivity in rainfed agriculture. Hence, the spatiotemporal dynamics of soil moisture is one of the key indicators of resilience. We hypothesized that above - average soil moisture during the growing season could i mprove the resiliency of rainfed agricultural systems. T hus, the elevated soil moister content can positively affect the growth and yields of crops , especially during climate extremes such as droughts . This hypothesis was tested in a long - term crop rotatio n (corn - soybean - wheat) setting . The experimental fields were managed under four treatments, namely, conventional, no - till, reduced input , and organic treatments. This long - term cropping system experiment was located at the W.K. Kellogg Biological Station i n the temperate humid climate of Southwest Michigan, United States. The data on precipitation, soil moisture, total crop biomass, crop yield, and soil organic carbon were collected for the duration of this study. Four soil moisture - based metrics, namely, t he mean relative difference , coefficient , the i ndex of temporal stability , and the soil water deficit index , were computed for each treatment. Next, the robustness of these metrics under long term (1993 - 2018) climate variabiliti es was evaluated for reducing drought severity and maintaining/improving growth and yield of crops using different statistical techniques . Evaluating the C limate R esilience in terms of P rofitability and F arm R isk M anagement for a L ong - T erm C orn - S oybean - W heat R otation under D ifferent 37 T reatment S ystems terms of profitability and risks that are essential concerns of producers. Three alternative treatments include the no - till treatment , the reduced input treatment, and the organic treatment, were compared against the conventional treatment for the same crop rotation (i.e., corn - soybean - wheat). The annual crop management and production data for ea ch treatment were obtained for the period of study (1993 - 2018) from the W.K. Kellogg Biological Station . Means and volatility of expected net returns and the risk preferences were used as metrics to evaluate the resilience. The net returns for each crop pr oduction and treatment systems were estimated using static enterprise budgeting, while the volatility of net returns was calculated using the relative standard deviation . The payoff matrix tool was utilized to assess the risk preferences for each productio n and treatment systems. A statistical mixed model was used to quantify the effects of long - term climate variability on the estimated net returns. Finally, resilient treatments were defined as those with high er expected net revenues with lower volatility t o the conventional treatment and also have a wide range of risk preferences for the adaptation. Quantification of Resilience Metrics as Affected by a Conservation Agricultural Practice at a Watershed Scale quantification of resilience from the field scale to a larger scale (e.g., a watershed). This study was implemented in the Kalamazoo River watershed located in the Southwestern Michigan of the United States . The long - term da ta collected from the W.K. Kellogg Biological Station for the conventional and the no - till treatments were used to develop a sequence crop model to represent the corn - soybean - wheat rotation, and this model was calibrated using the yield and soil moisture data for the study period (1993 - 2019). The soil, clima te, and land use data were collected for the watershed scale simulation . Meanwhile, yield, soil moisture, and recharge values were obtained 38 from the calibrated crop model for different soils and climate within the study area. The net return was calculated using an enterprise budget for each crop. Yield, net return, and recharge data were analyzed using a mixed model to exploit the statistical significance between two treatments. The daily soil moisture data over the growing season was used to compute soil m oisture metrics, namely, the mean relative difference and the soil water deficit index . The annual recharge data comes from the crop model for the agricultural areas, while a calibrated groundwater model was used to calculate a long - term mean recharge valu es from the non - agricultural areas. This information was integrated to assess the long - term changes in the water table under both treatments. Finally, r echarge, groundwater table , soil moisture, yield , and net return were used to evaluate the overall chang es in resiliency as affected by the adaptation of no - till as a conservation agriculture treatment in the watershed . 39 4. EVALUATING THE APPLICABILITY OF SOIL MOISTURE - BASED METRICS FOR GAUGING THE RESILIENCY OF RAINFED AGRICULTURAL SYSTEMS 4. 1 Intr oduction Rainfed agriculture systems account for 80 percent of the croplands in the world and contribute to nearly 60 percent of total food production (FAO, 2017) . Meanwhile, in the United States, about 94 percent of farmland acres are considered rainfed a griculture (USDA, 2014) . Rainfed agricultural systems in the United States are economically important, ecologically diverse, technologically advanced, and are most common in the eastern half of the mainland where annual precipitation is greater than 500mm. In this region, the majority of the corn and soybean crops are produced from these systems, either as a monocrop or in rotation. The productivity of these crops steadily increased in the past as a result of genetic improvements (about 70%) and management interventions (about 30%). Moreover, with the development of glyphosate - resistant crop varieties, adoption of conservation agriculture practices such as no - tillage has been substantially increased to counteract the problems of soil erosion, nutrient leachi ng and runoff, and yield instability (Franzluebbers et al., 2011) . Regardless of this overall increase in productivity, crop yields have been shown to be vulnerable to interannual variability in the climate (Hatfield et al., 2018; Hatfield, 2012; Lesk et al., 2016; Thornton et al., 2014) . Like in other areas around the world, changes in the regional climate in the Midwestern states result in increasing the frequencies of extreme events such as droughts, floods, and heatwaves (Andresen et al., 2012; Hatfiel d et al., 2018; Pryor et al., 2014) . The water availability for rainfed agriculture is primarily controlled by the seasonal pattern of precipitation (intensity and frequency) and its interactions with the soil - plant - atmosphere continuum (Rost et al., 2009) . This makes rainfed crop production much vulnerable to the effects 40 of climate variability and extremes. For example, the flash drought in 2012 devastated major crops and economic base of rainfed farms in the Midwestern region (Fuchs et al., 2012; Mal lya et al., 2013; Otkin et al., 2018, 2016) . Flash droughts are associated with decreased precipitation and humidity, increased solar radiation, and elevated temperatures, leading to reduced availability of soil water to crops (Ford and Labosier, 2017) . As highlighted by Rippey, (2015) , due to this drought development, corn yield in the US has fallen for three consecutive years from 2010 - 2012 for the first time since 1928 - 1930. These climate extremes will have substantial impacts not only on the national ec onomy but also on international markets (Boyer et al., 2013) . Resilience can be generally defined as a capability of a system to recover from stressors (Holling, 1973) . Therefore , climate - resilience can be considered as the ability of a system to maintain its structures and patterns of behavior in the face of climate perturbations. This allows the system to continue to provide its services, which in the case of agriculture, is the growth and yield of crops (Tendall et al., 2015; Urruty et al., 2016; Walker et al., 2004) . It also refers to the ability of a system to develop capacities to cope with, adapt, and potentially transform the appropriate management practices to face the challenges of the climate shocks and extremes (Bousquet et al., 2016) . Specific to rainfed agriculture, climate - resilience can be used to describe the ability of the components of the system to recover from water stress (Tow et al., 2011) . This could be achieved by implementing management interventions that will keep the soil moisture at relevant levels in a way that extreme climatic events shall not reduce the crop yields significantly. Such actions to enhance the climate - resilience in rainfed agriculture can be broadly categorized as genetic interventions, informed decisions, and agr onomic interventions. Genetic interventions involve developing new germplasm with improved tolerance to environmental stresses such as drought and heat stress and/or crop genotypes with phenological 41 adjustments to avoid such stresses (Ainsworth and Ort, 2 010; Ceccarelli et al., 2010; Davies et al., 2011) . Adoption of these genotypes with improved stress tolerance has increased the climate - resilience of corn and soybean production systems around the world (Cairns et al., 2012; Chapman et al., 2012; Sadok an d Sinclair, 2011) . Informed decisions refer to the utilization of seasonal climate forecast information (Hansen et al., 2011; Meinke et al., 2006) . Forecast products such as - month outlooks) can potenti ally influence most of the agronomic and genetic interventions. Agronomic interventions may include, adjustment of planting and harvesting times, altering fertilization rates and irrigation practices (Howden et al., 2007; Nouri et al., 2017; Rurinda et al. , 2015) , mulching ( Erenstein, 2003; Qin et al., 2015) , crop diversification and agroforestry (Altieri et al., 2015; Gan et al., 2015; Lin, 2011; Mbow et al., 2014) and adoption of conservation agriculture (Delgado et al., 2013; Michler et al., 2019) . There is also evidence for the improvement of climate - resilience under organically managed agricultural systems (Scialabba and Mller - Lindenlauf, 2010; Tuomisto et al., 2012) due to the enhancement in soil quality and reduction in environmental impacts. In this study, we are mainly focusing on agronomic interventions, especially conservation agriculture as climate - resilience practices. Conservation agriculture comprises of three management principles: minimum soil disturbance/no - till, permanent soil cover by crop residues/cover crops, and crop rotation (Hobbs et al., 2013; Pittelkow et al., 2015a) . These conservation practices are adopted at various intensities and combinations in the Midwestern states to provide varying degrees of resilience to climate shocks and extremes (Denny et al., 2019) . In comparison to conventional tillage, the no - till/zero - tillage systems showed the highest level of climate - resilience as a result of improved water availability and soil quality, that helps to avoid substantial reductions i n crop yields during extreme climatic events (Delgado et al., 2013; 42 Harrington and Tow, 2011; Michler et al., 2019) . In contrast, some other studies (e.g., Pittelkow et al., 2015a; Powlson et al., 2014) have shown that no - till reduces crop yields compared to conventional tillage and their potential for climate - resilience is limited. Interestingly, Pittelkow et al., 2015b , in their comprehensive review, highlighted that no - till, when combined with the other two principles of conservation agriculture (residue retention and crop rotation), significantly increases the crop yields in rainfed systems. This could be due to the ability of the system to capture snow, reduction of runoff and soil evaporation (with the retention of crop residues), and creation of bette r soil structure and rooting patterns through crop rotations to store more water in the root zone . To evaluate responses of varying agronomic interventions of climate - resilience practices, there is a need for quantification metrics for system resiliency. Resilience metrics can be used to shift a system towards more desirable and sustainable states, track thresholds of potential concerns, and help with assessments on how the system is being managed (Quinlan et al., 2016) . The Committee on Sustainability Assessment (COSA) stated that gauging resilience generally involves a holistic approach that incorporates social, economic, and environmental dimensions of resilience (COSA, 2017) . Because of the complexity a nd interactions in these three dimensions, food system resilience is often assessed qualitatively (Toth et al., 2016) . However, qualitative assessments are region - specific and subject to variations in assumptions. Therefore, multiple tools have been develo ped to evaluate the climate resilience of food production systems in many parts of the world (Douxchamps et al., 2017) . These tools have often been applied to large socioeconomic units (e.g., households/communities/administrative regions/national scale). F or example, the Resilience Index Measurement and Analysis Model (RIMA) developed by the Food and Agriculture Organization of the United Nations (FAO, 2016) , is increasingly used to measure the climate - resilience of 43 agricultural communities in many African countries (Serfilippi and Ramnath, 2018) . Other such tools are Community Based Resilience Assessment (CoBRA) developed by United Nations Development Program (UNDP, 2013) , Self - evaluation and Holistic Assessment of c limate Resilience of farmers and Pastoralists (SHARP) used by FAO (Choptiany et al., 2017) , Community - based Risk Screening Tool - adaptation and Livelihood (CRiSTAL) developed by International Institute for Sustainable Development (IISD, 2014) , Climate Vulne rability and Capacity Analysis (CVCA) developed by Care International (Care, 2009) and Resilience, Adaptation and Transformation Assessment Framework (RATALF) developed by The Commonwealth Scientific and Industrial Research Organization 2015) . In general, these tools use indices known as resilience metrics to evaluate the flexibility of a system . Means and variance of agricultural production (Di Falco and Chavas, 2008) , crop yields (Birthal et al., 2015; Martin and Magne, 2015) , profit (B rowne et al., 2013; Komarek et al., 2015; Seo, 2010) , revenue (Kandulu et al., 2012; Rigolot et al., 2017; Tibesigwa and Visser, 2015) , labor productivity (Komarek et al., 2015) , crop failure (Jones and Thornton, 2009) , dietary diversity (Dillon et al., 20 15) , farming risks (Komarek et al., 2015) , agricultural gross domestic product (Hsiang and Jina, 2014) , and expenditure for food consumption/food security (Alfani et al., 2015) have been used as resilience metrics. These metrics are often used in combinati on and have shown non - linear responses to climate variability depending upon various characteristics of farms and farmers (Di Falco and Chavas, 2008; Tittonell, 2014) . Despite the growing knowledge in this area, there is still no consensus on how resilienc e should be measured and no universal tool available to quantify resilience at different scales. Moreover, existing tools and frameworks often fail to capture the spatial and temporal dynamics of resilience (Dixon and Stringer, 2015; Douxchamps et al., 201 7) . 44 To address these challenges, there is a need for new metrics to address the complexity in agricultural systems while being simple enough to be measured and adopted by individual farmers. Therefore, we propose a new measure to overcome these challenges and constraints by providing a case - specific definition of resilience and confining our focus to long - term agronomic performance (soil moisture , growth, and yield of crops) under different rainfed agricultural systems in an experimental farm scale. Howev er, it is important to note that this study is only focusing on farm - scale resiliency and therefore is not considering other sociological/institutional characteristics that are important beyond farm - scale, which can be evaluated through existing resilience metrics, which were discussed earlier. In order to achieve this goal, three objectives need to be satisfied: 1) rank the relative resilience of different rainfed agricultural systems using the metrics of temporal dynamics of soil moisture; 2) evaluate the robustness of the soil moisture metrics of temporal dynamics to growth and yield under climate extremes; and 3) compare the effectiveness of different rainfed agricultural systems on reducing agricultural drought severity. 4. 2 Materials and Methods Root z one soil moisture is the key determinant of productivity in rainfed agricultural systems (Jägermeyr et al., 2016) . The ability of a system to store a substantial amount of soil moisture is one of the key indicators of resilience, as recommended by COSA, (2017) . In this study we applied a combination of soil moisture - based metrics to gauge the resiliency of rainfed agr icultural systems. Long - term (1993 - 2018) agricultural experiment data on soil moisture, total crop biomass, crop yield, and soil organic carbon were collected, which provides an excellent opportunity to conduct an exhaustive evaluation of resiliency in agr icultural systems. We hypothesize that a rainfed agricultural system has a higher degree of resilience if it can maintain above - average soil moisture; thus, this relative resilience will beneficially affect the yields, 45 especially during periods of climate variabilities (e.g., drought). We test this hypothesis using soil moisture - based metrics in four differently managed long - term rainfed row crop treatments, namely, conventional treatment (CON), no - till treatment (NT), reduced input treatment (RI), and orga nically managed treatment (OR). These soil moisture - based metrics include three metrics of the temporal stability of soil moisture and an agricultural drought index. The soil moisture temporal dynamics metrics were selected to evaluate the relative resilie nce of different rainfed agricultural treatments (Objective 1), then we examined the robustness of these soil moisture metrics for different crops under climate extremes (i.e., dry and wet years) during the growing season. The results of this section can p rove whether the soil moisture temporal dynamics metrics can be used to evaluate the resiliency of different agricultural systems as measures of growth and yield (Objective 2). Finally, an agricultural drought index was selected to evaluate the effectivene ss of different rainfed agricultural treatments on reducing the agricultural drought severity (Objective 3). 4. 2 . 1 Study Location and Site Description This study was conducted at the Kellogg Biological Station (KBS), where Long - Term Ecological Research (LTER) experiment is implemented to evaluate the performance of different annual and perennial crops under varying management intensity gradients (Roberts on and Hamilton, 2015) . KBS is located in Southwest Michigan at 288 AMSL, within the northern - 2010, the mean 1,005 mm, with 511 mm of the total precipitation falling as rain during the summer growing season (May - September) (NCEI, 2019) . The evapotranspiration in this region is water - limited during the warmer part of the year and energy - limited during the colder months (Mcvicar et al., 2012) . The soil of this experimental 46 station is classified as fine - loamy, mixed, mesic Typic Hapludalf (Kalamazoo loam series) developed on glacial till and outwash (Syswerda and Robertson, 2014) . The texture of the Ap horizon (0 - 30 cm) of this soil is loam or sandy loam (43% sand, 38% silt, and 19% clay). The pH, bulk density, and total soil carbon are 5.5, 1.6 g cm - 3 , and 12.85 g kg - 1 , respectively (Crum and Collins, 1995) . The Main Cropping System Experiment (MCSE), established in 1988 (Figure 4. 1), is comprised of seven model ecosystems namely, four annual row crop systems with different management intensity (treatments), Poplar ( Populus deltoides × P. nigra ), continuous Alfalfa ( Medicago sativa ), and an early successional vegetation community (Robertson and Hamilton, 2015) . For this study, we selected the first four annual row crop treatments that were designed in a management intensity gradient. 47 Figure 4. 1 . The experimental fields of the Main Cropping System Experiment (MCSE) at the Kellogg Biological Station (KBS) in Michigan, United States. 48 Crops were established and measurements began in 1989. All agricultural systems within the experiment have been managed as rainfed and each of them is as signed to six replicates (blocks) of one - ha, plots (87 × 105 m) in a randomized complete block design. The four annual row crop treatments consisted of maize ( Zea mays ) - soybean ( Glycine max ) - winter wheat ( Triticum aestivum ) rotations managed as (i) convent ional treatment (CON), (ii) no - till treatment (NT), (iii) reduced input treatment (RI), and (iv) organically managed (USDA certified organic) treatment (OR). The conventional treatment is considered as the control. This allows us to measure the resilience of selected agricultural systems in comparison to the conventional system. The reasons behind the selection of the conventional treatment as control are; 1) majority of the croplands (>85% globally and >65% in the United States) are under conventional agri culture (Kassam et al., 2019) and 2) conventional system has been identified as the significant contributor of pollutions to the environment (Foley et al., 2011; Meier et al., 2015) . All row crop treatments were planted and harvested at the same time. A r outine experimental design was followed only from 1993; therefore, we confined this study for the period of 1993 - 2018. The detailed description of the four annual row crop treatments is presented in the Supplementary Materials section (Table S 4. 1). 4. 2. 2 Data Collection To achieve the objectives of this research, gravimetric soil moisture, soil organic carbon, total crop biomass, and yield were measured for the period (1993 - 2018) from the experimental plots at KBS. Gravimetric soil moisture data were colle cted, and then respective bulk density data were used to convert it into volumetric soil moisture. This volumetric soil moisture data was used to calculate the metrics of soil moisture temporal dynamics to evaluate the relative resilience of different rain fed agricultural treatments (Objective 1). In addition, soil organic carbon data were 49 collected to explore its associations with observed soil moisture dynamics. Total crop biomass and yield data were collected to evaluate the robustness of the soil moistu re metrics to gauge resiliency to climate extremes in terms of growth and yield of crops (Objective 2). Finally, the volumetric soil moisture data was used to calculate the drought index to evaluate the effectiveness of different rainfed agricultural treat ments on reducing the agricultural drought severity (Objective 3). Soil moisture: Gravimetric soil moisture measurements began in 1989 in all MSCE agricultural treatments. These measurements were employed each year, typically biweekly throughout the growin g season (April - October). Soils were collected from five permanent sampling stations established in each replicate (plot). Two soil cores were taken from each sampling station using soil augers to represent 0 - 25 cm depth. These ten samples then composited by the physically mixing of individual soil cores taken within a replicate into one homogenous sample. Composite soil samples were sieved through a 4 mm screen to remove debris and homogenize the sample. About 40 - 50 g of sieved composite samples were taken into soil moisture cans, and then oven - moisture ( g ) was calculated as follows (Reynolds, 1970) : ( 4. 1) Bulk density: Bulk density is used to convert gravimetric soil moisture into volumetric soil moisture. It was measured by taking core samples at different soil horizons (Blake and Hartge, 1986) . In this study, we used the gravel - free bulk density values measured at 0 - 2 5 cm depth. Bulk density ( BD ) was calculated as follows (Carter and Gregorich, 2008) : ( 4. 2) 50 Bulk density measurements were availab le for MCSE during the years of 1996, 2001, 2010 and 2013. Therefore, bulk density measurements from the years 1996, 2001, 2010 and 2013 were used to convert gravimetric soil moisture measurements for the periods of 1993 - 2000, 2001 - 2009, 2010 - 2012, and 201 3 - 2018, respectively. E quation 4. 3 (Evett, 2008) was used to covert gravimetric soil moisture into volumetric soil moisture ( v ), assuming the density of water is 1 gcm - 3 . ( 4. 3) Soil organic carbon : Soil organic carbon from the surface soils for each treatment and respective replicates was available for the period of 1989 - 2001. Even though the measurements have been made for total soil carbon (organic and inorganic forms), extensive testing of KBS sur face soils has shown soil inorganic carbon to be non - detectable; thus, total soil carbon is identical to total organic carbon. To measure the total soil carbon, subsamples were oven - dried at for at least 48 hours until no further mass loss occurs. Dr ied and finely grounded soil samples are weighed into small foil capsules which are combusted in an automated CHN (carbon, hydrogen and nitrogen) analyzer that measures the amount of released CO 2 using gas chromatography. Soil carbon values are expressed a s a percentage of carbon in dry soil. We utilized these soil organic carbon values to explore its associations with soil moisture dynamics. Crop biomass and yield: Total crop biomass was measured from 1993 at peak growth of biomass. Aboveground biomass hand - clipped within 1 m 2 quadrat at each of the five sampling stations per replicate, once in a growing season. Collected biomass was oven - hours and weighed. Seed yield of crops was measured during crop harvest determined by m achine harvesters appropriate to each crop. The standardized moisture content for yield measurement was 15.5% for corn and 13% for wheat and soybean. 51 4. 2 . 3 Calculating the Metrics of temporal Dynamics of Soil Moisture to Examine the Relative Resilience o f Different Rainfed Agricultural Treatments Metrics of the temporal dynamics of soil moisture are the mean relative difference (MRD), index of temporal stability (ITS), and non - coefficient ( r s ). MRD is used to present the relative ranks of different rainfed agricultural treatments in terms of root zone soil moisture content, ITS is used to show the variability of the MRD ranking over the growing season and r s was used to indicate the persistenc e of the relative MRD ranks over the period of study (Jacobs et al., 2004; Joshi et al., 2011; Vachaud et al., 1985) . In the past, temporal stability analysis of soil moisture was used to identify time - stable points or representative locations to employ mo nitoring networks/sensors (Barker et al., 2017; Brocca et al., 2010; Starks et al., 2006; Zhou et al., 2007) and/or to validate remote sensing soil moisture products (Cosh et al., 2008; Jacobs et al., 2004; Joshi et al., 2011; Wagner et al., 2008) . In othe r studies, the metrics were used to study the spatiotemporal dynamics of soil moisture in hillslope (Gao et al., 2016; Liu et al., 2018) or under different land use (Hu et al., 2010) or in diverse soil layers (Gao and Shao, 2012; He et al., 2019) . In this study, we will use these metrics to gauge the relative resilience of different rainfed agricultural treatments. Mean Relative Difference: MRD was introduced by Vachaud et al. (1985) to study the temporal stability of spatially measured soil moisture. The mean relative difference (equation 4 .4 ) together with the index of temporal stability (equation 4. 7) have been used in the spatiotemporal analysis of soil moisture (Gao and Shao, 201 2; He et al., 2019; Joshi et al., 2011; Liu et al., 2018) . Here, we use these metrics to evaluate the spatiotemporal dynamics of soil moisture across different agricultural treatments. The MRD (% cm 3 /cm 3 ) for treatment in a particular growing season (year) is defined as: 52 ] ( 4. 4) ( 4. 5) where, is the volumetric soil moisture (% cm 3 /cm 3 ) measured in a treatment T at time t . is the mean volumetric soil moisture of all treatments. N T is the number of treatments. Nt is the number of soil moisture measurements (sampling days) during the particular growing season. A negative MRD value indicates that the treatment is drier than t he field - averaged soil moisture, whereas a positive MRD value signifies that the treatment is wetter than the field mean soil moisture (Joshi et al., 2011) . Variance of Relative Difference (VRD) : The variance of the relative difference for each treatment is calculated as: ( 4. 6) The MRD quantifies the deviation of the soil moisture in a particular treatment, and the VRD quantifies the accuracy of that measurement (Joshi et al., 2011) . Here we calculated VRD to derive ITS in the next step (equation 4. 7). Index of Temporal Stability: ITS can be derived by considering both MRD and VRD (He et al., 2019; Jacobs et al., 2004; Liu et al., 2018) ; therefore, ITS represents both bias and accuracy metrics.: ( 4. 7) In a rank ordered MRD and ITS plot, treatments with MRD values close to zero and with smaller ITS values can be considered temporally more stable (He et al., 2019; Joshi et al., 2011; 53 Liu et al., 2018) . However, our intention in this work is to find out relatively wetter treatment (i.e., positive MRD). Because we hypothesized that if an agricultural system can hold more soil moisture due to conservation pr actices (e.g., no - till) during the growing period than a conventional system, it would positively be affecting the growth and yield of crops in rainfed agricultural systems. Non - The non - parametric Spearm (Vachaud et al., 1985) was used to examine the persistence of MRD ranks over the 26 - year study period for each treatment. The r s is expressed as: r s = ( 4. 8) where, R ij is the rank of MRD in treatment i on the year j , R ik is the rank of MRD in treatment i on the following year k , and n is the number of years. The r s was calculated for each agricultural system where r s of 1, for any treatment, represent the MRD having the same rank between the years j and k . Therefore, higher values of r s (values closer to 1) represent higher temporal persistence of relative ranking over the study period (Liu et al., 2018) . 4. 2 . 4 Evaluate the Sensitivity of the M ean R elative Difference of Soil Moisture to Climate Variability and its Reflections on Crop Growth and Yield in Different Treatments Long - term experimental data on soil moisture, total crop biomass, crop yield, and soil organic carbon were collected from KBS - LTER data catalog and processed using a python script in Wing Pro Version 7.1.2 (Wingware, Cambridge, Massachusetts, USA) and Microsoft Excel Version 2016 (Microsoft Corporation, Redmond, Washington, USA). In this study, a mixed model (Milliken and Johnson, 200 9) was used to explore the statistical significance of random and fixed effects of independent variables on selected response variables. 54 The statistical model for evaluating the effects of treatment, climate variability and year on selected response varia bles (i.e., MRD, total crop biomass and yield) was specified as : y ijk = µ + clim + trt i + yr k + trt i × clim + blk j ( yr k ) + trt i × yr k + e ijk ( 4. 9) where, y ijk : is the vector of observation (response variable) collected for the i th treatment, within j th block on the k th year. µ : is the overall mean, trt : is the fixed effect of treatments, which represents different agricultural systems, blk : is the random effect of the replications (blocks) nested within years, clim : is the fixed effect of the climate variability (dry, normal and wet), yr : is the random effect of the year, trt × clim : is the interaction between the effects of treatment and the fixed effect of climate variability, trt × yr : is the interaction between the effects of treatment, the random effect of the year, and e is the residual. The effect of seasonal mean temperature was also initially included in the statistical model as a fixed effect; however, the temperature effect and all the interaction terms wi th temperature were non - significant at p response variables. Thus, we decided not to include the effect of seasonal mean temperature in the final statistical model. The non - significant effect of seasonal mean temperature was most likel y a result of a little coefficient of variation (CV = 4.8%) observed for this continuous variable. The normality of the residuals and homogeneity of variances were checked for each response variable (Milliken and Johnson, 2001) . The data for all response v ariables were normal, and their variances were homogenous. Analyses were performed by crop (separately for each crop), using PROC GLIMMIX procedure (Milliken and Johnson, 2009) in SAS software version 9.4 (SAS Institute Inc. Cary, North Carolina, USA). Fi xed effects were tested using F - test (Steel et al., 1980) , and random effects were tested using Wald test for covariance (Wald, 1943) . Tukey Test for mean separation was performed when significant differences were detected for the fixed effects (Lee and Le e, 55 2018) , and significant differences were evaluated at the 0.05 probability level (Steel et al., 1980) . (Weaver et al., 2017) was performed using MINITAB 15 (Minitab, LLC. State College, PA, USA) to determine the relationshi p between MRD vs. soil organic carbon. 4. 2 . 5 Compare the Effectiveness of Different Treatments on Reducing Agricultural Drought Severity Agricultural drought is defined as a period of soil moisture deficiency resulted in the shortage of precipitation that occurs for a few weeks of duration (Esfahanian et al., 2017). Drought is the major growth - and yield - limiting abiotic stress in rainfed agricultural systems. A U.S. based study estimated yield reductions in row crops as high as 8% in the rainfed agriculture counties of Midwest due to recent drought events (Kuwayama et al., 2019). Moreover, increasing climate variability and extremes make rainfed agriculture more vulnerable to droughts (Hatfield et al., 2018), which increases the adoption of conservation practices among farmers (Ding et al., 2009). Therefore, we decided to evaluate the impacts of these treatments on agricultural drought severity and occurrence using the Soil Water Deficit Index (SWDI). SWDI was considered for this study sin ce it requires fewer inputs and is more widely used than other agricultural drought indices such as Palmer Moisture Anomaly Index (Alley, 1984; Palmer, 1965), Soil Moisture Deficit Index (Narasimhan and Srinivasan, 2005), and Evapotranspiration Deficit Ind ex (Narasimhan and Srinivasan, 2005). In addition, when time - series soil moisture measurements are available from the root zone, SWDI can be successfully implemented at field scale to assess agricultural drought (Martínez - Fernández et al., 2015) with the a dditional information on field capacity and permanent wilting point of that particular soil. Ultimately, the evaluation of the agricultural 56 drought severity will enable us to gauge the relative resilience of different rainfed agricultural systems. Catego rization of climate variability: Total seasonal precipitation (April - October) for the period of 30 years (1989 - 2018) was collected from the KBS weather station located within the experimental site, and the probability distribution was calculated to categor ize the climate variability, namely dry years (cumulative probability less than 33.3%), normal years (cumulative probability in between 33.3% and 66.6%) and wet years (cumulative probability greater than 66.6%) as shown in Figure 4. 2. Figure 4. 2 . Cumulative probability distribution of seasonal precipitation for the period of 30 years (1989 - 2018) at KBS weather station. 57 Based on the frequency analysis explained above, dry years were categorized as the years wet years were categorized as the years that the precipitation is in between. Classification of the entire experimental period (1993 - 2018) into the above climate v ariability categories for each crop in the rotation is given in Table 4. 1. Table 4. 1 . Climate variability for each crop in the rotation . Crop Categories of climate variabilities Years Corn Dry year s 1996, 1999, 2002, 2005 Normal year s 1993, 2014, 2017 Wet year s 2008, 2011 Soybean Dry year s 1997, 2012 Normal year s 2003, 2009 Wet year s 1994, 2000, 2006, 2015, 2018 Wheat Dry year s 1995, 1998, 2007 Normal year s 2004, 2010, 2016 Wet year s 2001, 2013 Soil Water Deficit Index (SWDI): Soil water deficit index is an agricultural drought indicator developed by Martínez - Fernández et al. (2015) based on water deficit accumulation or soil water storage. SWDI is the fraction between the differences of i) volumetric soil moisture and field ca pacity, and ii) plant available water content, which is the difference between field capacity and permanent wilting point. This fraction is then multiplied by 10 to obtain SWDI (equation 4. 10), and the respective drought severity levels: ( 4. 10) FC is the field capacity (% WP is the permanent wilting point (% cm3/cm3). The SWDI is calculated for the root zone on all the da tes of soil moisture measurement for each treatment 58 during the growing season. The median volumetric soil moisture of six replicates was used to calculate SWDI. Since the soil texture of this experimental station is fine - loamy, we selected the field capaci ty value of 27% and the wilting point of 12% as proposed by Ratliff et al. (1983) and Hanson et al. (2000). When SWDI is positive, the soils have excess water; when it equals zero, the soil is at the field capacity (i.e., no water deficit). A negative MRD indicates the drought, and - 10. After this point, the soil water content falls below the lower limit of plant - available water (Savage et al., 1996). Based on calculated SWDI, drought sev - - 2 > SWDI > - - 5 > SWDI > - 10, and - 10 > SWDI (Martínez - Fernández et al., 2015). The number of droughts in each severity cat egory was added for each treatment during the study period 1993 - 2018, and the percentage of drought events was calculated and compared under the categories of climate variability (dry, normal, and wet years). Moreover, SWDI for the entire study period (199 3 - 2018) was organized in descending order for each treatment to perform probability analysis as described by Alizadeh (2013). This would allow analyzing the behavior of drought severity under different treatments in response to climate variability. 4. 3 R esults and Discussion 4.3.1 Ranking the Resilience of Soil Moisture in Different Treatments As discussed earlier, we hypothesize that if soil moisture in an agricultural system can remain wetter than the conventional (control) system over the growing seas on due to treatment, it would beneficially affect the growth and yield of crops in rainfed agriculture. Therefore, treatments were ranked on the ascending order of MRD for the study period (1993 - 2018) and 59 presented together with ITS for each climate variab ility category as dry years (Figure 4. 3), normal years (Figure S 4. 1), and wet years (Figure S 4. 2). Figure 4. 3 . Ranked MRD of volumetric soil moisture and ITS for each treatment during the dry years. Note: Crop grown is given next to the respective year for each plot. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. According to the relative positions of MRD during dry years (Figure 4. 3), a no - till treatment maintained higher soil moisture (and higher MRD) than a conventional treatment. Similarly, the organic treatment always performed better than the reduced input tr eatment, while the conventional treatment performed better than the reduced input treatment in the majority of years. During normal and wet years, no - till and organic treatments were better than the 60 conventional treatment, while reduced input treatment per formed equally to the conventional treatment in most of the normal and wet years (Figures S 4. 1 and S 4. 2). Table 4. 2 presents the number of occurrences in each rank, based on MRD and the relative positions of MRD (either negative or positive) for different categories of climate variability. The rankings are based on the ascending order of MRD in which a wetter system is represented by the highest - ranking (rank=4). During dry and normal years, the majority of the times, the conventional treatment represents the second rank, the no - till treatment represents the fourth rank, reduced input treatment represents the first rank, and the organic treatment represents the third rank. This indicates that the no - till treatment is the wettest and organic treatment is wet ter than the conventional treatment; however, the reduced input treatment is drier than the conventional treatment during dry and normal years. During wet years, the majority of the times conventional treatment represents the lowest rank (i.e., rank 1) thu s it was the driest of the treatments, while no - till treatment represents the highest rank (i.e., rank 4) thus wettest of the treatments. Organic and reduced input treatments fall in between; however, the organic treatment is still wetter than the reduced input treatment. Concerning the relative positions of MRD, the no - till treatment is wetter (100% of years having positive MRD) than all other treatments during dry, normal, and wet years. The organic treatment performs similar to the conventional treatment, while the reduced input treatment is drier than conventional treatment in dry years (100% of years having negative MRD). During convent ional treatment. Meanwhile, the reduced input treatment performs equally to the conventional treatment during normal and wet years. 61 Table 4. 2 . Relative positions and ranking based on MRD, and their respective percentages for each agricultural system under each climate variability category . Category of climate variability Treatment * Relative positions of MRD (Percent of the total years) Ranking based on MRD Number of occurrences in each rank (Years) MRD< 0 MRD>0 1 2 3 4 Dry years CON 77.8 22.2 2 4 3 0 NT 0 100 0 0 0 9 RI 100 0 7 2 0 0 OR 77.8 22.2 0 3 6 0 Normal years CON 87.5 12.5 2 5 1 0 NT 0 100 0 0 1 7 RI 87.5 12.5 6 2 0 0 OR 50 50 0 1 6 1 Wet years CON 100 0 5 4 0 0 NT 0 100 0 0 0 9 RI 100 0 3 5 1 0 OR 44.4 55.6 1 0 8 0 * CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. The highest level of soil moisture resilience shown by the no - till treatment irrespective of climate variability. This can be due to improved soil water dynamics. In the no - till treatment, a greater amount of crop residues is retained in the soil compared to the conventional treatment (Palm et al., 2014; Pittelkow et al., 2015b) . As a result, organic matter and biotic activity significantly increase in the topsoil, which leads to greater wet aggregate stability and macropore connectivity while reducing soil compaction (Blanco - Canqui and Ruis, 2018; Palm et al., 2014; Perego et al., 2019) . Therefore, water infiltration increases, surface runoff, and soil evaporation decreases, which results in an increased amount of plant - available water (Liu et al., 2013; Pa lm et al., 2014; Thierfelder et al., 2013; Thierfelder and Wall, 2009) . The impact of the no - till treatment on higher soil moisture and water use efficiency is also evident under irrigation (Gathala et al., 2013; Grassini et al., 2011) . The average measure d soil organic carbon in our 62 experiment for the 13 - year period (1989 - 2001) supports these arguments where the no - till treatment had significantly higher soil organic carbon than the conventional treatment (Figure S 4. 3). Additionally, the organic treatment and the reduced input treatment also had significantly higher soil organic carbon compared to the conventional treatment. However, accumulation of soil organic carbon in tilled reduced input and organic treatments are unlikely and may be a result of legumi nous winter cover crop established in these treatments (Robertson et al., 2014) . We also found a significant positive correlation between the mean relative difference (MRD) of volumetric soil moisture and the soil organic carbon (Figure S 4. 4). This highlig hts the increase in soil moisture retention with increasing soil organic carbon, which was observed in the no - till and the organic treatments. This is one of the main reasons for the increasing yield of crops with increasing soil organic carbon (Oldfield e t al., 2019) unless other resources are limited. The next best treatment in terms of soil moisture resilience was the organic treatment. This can be due to increased biological activity such as an abundance of earthworms that could beneficially affect the soil water dynamics; however, activity of earthworms will be limited in other systems as a result of the application of herbicides (Bai et al., 2018) . Moreover, leguminous winter cover cropping may also have beneficially affected the soil water dynamics in the organic system (Basche et al., 2016) . Even though the reduced input system had a leguminous winter cover crop, it had the lowest soil moisture resilience may be because of the impacts of herbicides on soil structure by reducing soil biological activit ies (Basche and DeLonge, 2019) . As we look for an agricultural system with higher MRD, most of the time, the system with greater MRD also has higher ITS (Figures 4. 3, S 4. 1, and S 4. 2) . However, under a few instances, lower MRD can obtain a higher ITS. This can be explained by Equation 4. 7 in which the ITS is always a positive value regardless of whether MRD is a positive or negative number. 63 Therefore, ITS is not a suitable metric for gauging the resilience of an agricultural system as related to soil moisture temporal dynamics in a growing season. 4. 3. 2 Temporal Persistence of Soil Moisture in Different Treatments coefficient ( r s ) of different treatments for the duration of the study (1993 - 2018) is shown in Figure 4. 4. Accordingly, the no - till treatment showed the highest temporal persistence ( r s oisture among other treatments for the long - term (26 - years). The conventional treatment showed the lowest temporal persistence. This is due to its ranking of the MRD, which behaves differently in different years (Figures 4. 3, S 4. 1, and S 4. 2). Thus, the abi lity of the conventional treatment to maintain the resilience of soil moisture for an extended period is limited. The temporal persistence of the organic treatment and the reduced input treatment fall in between conventional and no - till treatments; however , the organic treatment performs a little better than the reduced input treatment. Therefore, the ability of the organic treatment to maintain the resilience of soil moisture for a long period is greater than the reduced input treatment. This observation i s in support of the long - term effects of the no - till treatment in soil moisture conservation (Bai et al., 2018; Lampurlanés et al., 2016) . Moreover, Castellini et al. (2019) detected a significantly higher number of micropores under the long - term no - till t reatment compared to the conventional treatment. This can be another reason for the highest temporal persistence of soil moisture in the no - till treatment. Meanwhile, the moderate level of temporal persistence of soil moisture observed for organic and redu ced input treatments can be attributed to increased soil moisture conservation with the winter cover crop applied to these treatments (Basche et al., 2016) . Cover crops are also beneficial in reducing annual deep drainage and soil evaporation (Yang et al., 2020) , thereby increasing soil water availability in the root zone. 64 Figure 4. 4 . study. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. 4. 3 . 3 Evaluating the Effects of Treatments and Climate Variability on Mean Relative Difference of Soil Moisture and its Reflections on Growth and Yield of Crops The mixed model (equation 4. 9) explained in Section 2.5 was used to analyze the MRD, total biomass (at peak growth), and yield, considering them as response variables for individual crop types. The probability of the effects on the above response variables on each crop is presented in Table S 4. 2. Accordingly, the effect of treatments ( trt ) on MRD was strongly significant ( p <0.0001) for all crops. However, the effect of climate variability ( clim ), year ( yr ), and interaction terms ( trt × clim and trt × yr ) were not significant. This means that MRD can be used to differentiate the impacts of different treatments regardless of climate variabilities and the random yearly effects . For total biomass and yield the effect of treatment ( trt ) was strongly significa nt ( p <0.0001) in corn and wheat while it was significant ( p< 0.05) in soybean. Therefore, 65 we first compared the treatments against the means of MRD, means of total biomass, and means of yield for each crop type in Table 4. 3: Table 4. 3 . The means of MRD, tot al biomass and yield for different treatments and crop types*. Treatment Corn Soybean Wheat MRD Total biomass (Mgha - 1 ) Yield (Mgha - 1 ) MRD Total biomass (Mgha - 1 ) Yield (Mgha - 1 ) MRD Total biomass (Mgha - 1) Yield (Mgha - 1) CON - 0.0549 c 14.26 a 6.78 b - 0.0551 c 5.33 a 2.40 b - 0.0202 b 8.07 a 3.74 a NT 0.1121 a 15.08 a 7.80 a 0.1055 a 5.46 a 2.83 a 0.1293 a 8.45 a 3.90 a RI - 0.0602 c 14.45 a 6.81 b - 0.0492 c 5.21 ab 2.62 ab - 0.0742 b 6.86 b 3.38 b OR 0.0136 b 10.14 b 4.30 c 0.0072 b 4.81 b 2.33 b - 0.0204 b 4.59 c 2.08 c * Means with the same letter in each column are not significantly different at p<0.05. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. The comparison of treatments versus means of MRD (Table 4. 3) showed that the no - till treatment had significantly higher MRD than the conventional treatment, which was ultimately reflected on the yield where it was significantly greater under the no - till treatment in corn and soybean. However, significantly higher MRD was not r eflected in the growth where the biomass under the no - till treatment was not significantly different from the biomass under the conventional treatment. MRD in the reduced input treatment was not significantly different from the conventional treatment, whic h was reflected in the growth and yield of corn and soybean, where their biomass and yield were not significantly different (Table 4. 3). Meanwhile, the MRD in the organic treatment was significantly greater than that in the conventional system for corn and soybean (Table 4. 3). This was not reflected in the growth as it was significantly lower in the organic treatment for both crops. Fur thermore, the yield of corn was significantly lower in the organic treatment than the conventional treatment. In contrast, the yield of soybean in the organic treatment was not significantly different from the yield in the conventional treatment. In wheat, MRD was significantly higher in the no - till treatment than the conventional treatment; however, growth and yield were not significantly different. Moreover, MRD in the reduced input and 66 organic treatments was not significantly different from the conventio nal treatment in wheat; however, growth and yield were significantly lower in these two treatments than the conventional treatment. In addition to significant treatment effects ( trt ) , the interaction between treatment and climate variability ( trt × clim ) w as also significant for the total biomass and yield of crops (Table S 4. 2). The occurrence of treatment by climate interaction ( trt × clim ) highlights that the performances of treatments change in different categories of climate variability; thus, worthy of an investigation. As shown in Figure S 4. 5, the cereals (corn and wheat) presented the highest growth and largest grain yield in the normal years, while the wet years produce higher growth and yield for soybean. Furthermore, we are particularly interested in the total biomass and yield performances of treatments under dry extreme of the climate variability. Total biomass of corn and soybean were not different among the treatments (Figure S 4. 5; a and b) during dry years. However, the no - till treatment had eq uivalent total biomass for wheat to that in the conventional treatment, while wheat growth was smaller in the reduced input and organic treatments (Figure S 4. 5; c). The no - till treatment produces higher yields than the conventional treatment for soybean an d wheat during dry years (Figure S 4. 5; e and f) while corn yield in the no - till treatment was similar to that in the conventional treatment (Figure S 4. 5; d). Dry year yields of corn and soybean under the reduced input treatment and the organic treatment we re comparable to the yields under the conventional treatment (Figure S 4. 5; d and e); however, these treatments did not perform well on the yield of wheat compared to the conventional treatment (Figure S 4. 5; f). Furthermore, the interaction between treatmen t and year ( trt × yr ) was also significant for the total biomass and yield of corn (Table S 4. 2). The occurrence of treatment by years interaction ( trt × yr ) is expected due to the number of years evaluated (26 - years). In addition, this can be 67 originated fr om interannual variations on other above - and below - ground environmental variables (e.g., solar radiation, vapor pressure deficit, and nutrient dynamics) which could also affect the growth and yield but were not included in the model (equation 4. 9) used fo r this study. In summary, significantly greater soil moisture retention, as quantified by MRD under the no - till treatment, was reflected on the significantly higher yields in corn and soybean than the conventional treatment when averaged across all the years. Moreover, the no - till soybean and wheat produced substantially higher yields than the conventional soybean and wheat during dry years. Similar to our study, significantly higher yields under the no - till treatment were also observed in several previous studies when c ombined with crop residue retention in long - term crop rotations (Corbeels et al., 2014; Deines et al., 2019; Pittelkow et al., 2015b; Rusinamhodzi et al., 2011) . Moreover, Syswerda et al. (2012) found significantly lower nitrate leaching loss under the no - till treatment than the conventional treatment in this same experiment at KBS - LTER. Hence, the no - till treatment has the ability to use a greater amount of nitrogen in the process of yield formation. Although the organic treatment had significantly greate r MRD in corn and soybean than the conventional treatment, it did not produce significantly higher biomass or yield. However, organic soybean yield was equivalent to the conventional treatment even during dry years. Significantly higher MRD for no - till and organic treatments in corn and soybean than the conventional treatment signifies a greater soil moisture retention over the growing season under no - till and organic treatments. As previously discussed, this can be attributed to the improvement in soil phy sical properties and soil organic carbon with these conservation systems (Hobbs et al., 2008; Valkama et al., 2020; Verhulst et al., 2010; Williams et al., 2017) . Additionally, the decrease in evaporation, increase in infiltration, and the ability to store more soil moisture under 68 no - tillage produces greater soil water storage (Blevins et al., 1971; Lal et al., 2012; Lampurlanés et al., 2016) . Meanwhile, t he major r eason behind the significant growth and yield reduction, especially in cereal crops (i.e., corn and wheat) with the organic treatment versus the other treatments, was nitrogen deficiency as organic treatment lacks exogenous nitrogen fertilizer application (Robertson et al., 2014) . Growth and yield reduction were evident even if though higher soil water retention was shown to be higher for organic treatment than the conventional and reduced input treatments in our study. Interestingly, the yield of soybean i n the organic treatment was comparable to that in the conventional and reduced input treatments even during dry years. This is because soybean has a synergistic relationship with nitrogen fixing bacteria, which reside on its roots and fix atmospheric nitro gen (Hungria and Mendes, 2015) . The equal performance of MRD in the reduced input treatment to the conventional treatment was reflected as the equivalent growth and yield of corn and soybean but not in wheat. In wheat, growth and yield were significantly lower in the reduced input treatment than the conventional treatment. Although reduced input treatment had the potential for soil quality improvement as a result of leguminous winter cover crop, this potential was limited because of herbicide and inorganic fertilizer application that substantially reduces the activities of soil biota (Bai et al., 2018; Tsiafouli et al., 2015) . Therefore, the reduced input treatment performs inferior to the organic treatment on MRD in corn and soybean even though its perform ance is comparable to the conventional treatment. Equal performance of the reduced input treatment to the conventional treatment on crop growth and yield of corn and soybean is due to its equal performance on soil moisture retention, as it was shown in thi s study. Nonetheless, the growth and yield of wheat in the reduced input treatment were significantly lower than the conventional treatment. This is associated to the lack of nitrogen to wheat as it is planted in the fall 69 immediately following the soybean harvest, which leaves a relatively low amount of nitrogen - rich crop residue for the wheat crop in the reduced input treatment, whereas corn and soybean follow nitrogen - fixing red clover winter cover crop (Robertson et al., 2014) that can supplement a reduc ed rate of nitrogen fertilizer application. This proves the ability of the winter red clover cover crop (Gentry et al., 2013; Vyn et al., 2000) to supply sufficient nitrogen to the reduced input system. Interestingly, the coefficient of variation (CV) of total crop biomass and yield for different treatments under two extremes of climate variability (i.e., dry and wet) showed that the no - till treatment had the lowest CV values among other treatments except for wheat during wet years (Table S 4. 3). The lowest CV value indicates fewer variations, hence greater stability of growth and yield of these crops, which also highlights the resilience of the no - till treatment. This finding was in support of Verhulst et al. (2011) , where they have shown that higher soil m oisture retention by the zero - tillage practice resulted in a more stable agronomic system than the conventional system, as it utilizes rainfall more efficiently. In contrast, Rusinamhodzi et al. (2011) argued that yield stability was not substantially impr oved by the no - till system. 4. 3 . 4 Compar ing the Effectiveness of Different Treatments on Reducing Agricultural Drought Severity According to the classification of drought severity (Martínez - Fernández et al., 2015 ), the drought categories that are crucial for crop production are moderate, severe, and extreme. This is because the loss of soil moisture as a percentage of total plant available water is 20 - 50%, 50 - 100%, and 100% for moderate, severe, and extreme droughts, respectively (Martínez - Fernández et al. , 2015) . These losses of plant available soil moisture are translated to some crop damages under moderate drought, likely crop damages under severe drought, and major crop damages 70 under extreme drought (Svoboda et al., 2002) . Mild drought is not associated with significant crop damages; therefore, the effectiveness of different treatments on reducing moderate, severe, and extreme droughts are only assessed here. In general, the numbers of severe and extreme drought events we re decreased in all treatments when climate variability shifted from dry to wet (Figure 4. 5). This is because increasing precipitation improves the availability of soil moisture to crops, thus reduces severe and extreme categories of agricultural drought, while increasing the drought events with lower severity (i.e., moderate/mild) or no drought. In terms of having drought - free events/no drought, no - till treatment is superior and organic treatment is better than the conventional treatment, while reduced inp ut treatment is the worst. For example, the no - till treatment and the organic treatment had 467% and 67% higher drought - free events than the conventional treatment, respectively, during dry years, while the reduced input treatment had no drought - free event s during dry years. The above observations were also noticed during normal and wet years where the no - till treatment and the organic treatment had substantially higher drought - free events than the conventional treatment, while the reduced input treatment h ad substantially lower drought - free events than the conventional treatment. Moreover, the effectiveness of the no - till system, especially on reducing moderate, severe, and extreme drought events is much higher than any other treatment irrespective of clima te variability. During dry years where the drought can be prominent, the no - till treatment had 4%, 23%, and 57% lower severity than the conventional treatment, respectively, for the moderate, severe, and extreme droughts. The effectiveness of the organi c system on reducing moderate, severe, and extreme drought events is still better than the conventional system. The percentages of moderate and severe drought events in the organic treatment were 20% and 17% lower than the conventional 71 treatment during dry years, respectively. Nonetheless, the effectiveness of the reduced input system is broadly limited. The reduced input treatment had 12% and 50% higher moderate and extreme drought events than the conventional treatment during dry years, respectively. Howe ver, it had 10% lower severe drought events than the conventional treatment during the dry years (Figure 4. 5). Figure 4. 5 . Percentage of agricultural drought severity events based on different treatments and climate variabilities during the experimenta l period (1993 - 2018). Note: The numbers of soil moisture measurements (events) available to calculate the percentage of different drought severities in the dry, normal, and wet years were 92, 80, and 98, respectively. CON: Conventional treatment; NT: No - t ill treatment; RI: Reduced input treatment; OR: Organic treatment. Probability analysis of SWDI for the entire experimental period showed that no - till treatment has the highest probability (22%) to have drought - free events (Figure 4. 6). Meanwhile, the org anic, the conventional, and the reduced input treatments are ranked next with a probability 72 of 12.5%, 7.8%, and 4%, respectively. The no - till treatment and the organic treatment had substantially lower probabilities under the impactful drought events (i.e. , moderate, severe, and extreme) than the conventional treatment, while these probabilities were larger in the reduced input treatment (Figure 4. 6). Figure 4. 6 . Probability distribution of drought experienced by each agricultural system as represented by SWDI during the experimental period (1993 - 2018). Note: Red dashed lines represent the boundary of different drought severity levels. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. Bo th of the above analyses have shown that the effectiveness of the no - till treatment to reduce moderate, severe, and extreme drought events is much greater than all other treatments. In agreement with these findings, Lal et al. (2012) showed that compared t o the conventional system, the no - till system had 86 mm more available soil water at planting during drought - hit 73 (2011 - 2012) in Akron, Colorado. This is due to greater soil water conservation under the no - till system as resulted by reduced runoff, evaporat ion rate, and increased capture of snow (Al - Kaisi et al., 2013; Lal et al., 2012) . Furthermore, Thierfelder and Wall (2010) found three to five times higher infiltration for no - till plots compared to conventional plots in Africa. They argued that the no - ti ll would increase the soil moisture and enable crops to mitigate the effects of droughts. Therefore, the effectiveness of the no - till treatment to reduce agricultural drought severity is much higher than the conventional treatment. The organic treatment is inferior to the no - till treatment but superior to the conventional treatment on reducing moderate, severe, and extreme droughts. The primary reason for the higher effectiveness of the organic treatment to reduce agricultural drough t compared to the conventional treatment could be due to higher water holding capacity of soils under the organic management (Lotter et al., 2003) . In the temperate climate of Switzerland, water holding capacity was reported 20 - 40% greater in organically m anaged soils in comparison to those managed conventionally (Mäder et al., 2002) . Pimentel et al. (2005) quantified 15 - 20% higher soil water availability in organic systems than the conventional systems in the long - term Rodale Institute Farming Systems tria l in Pennsylvania. This is because of the presence of higher soil organic matter in organically managed soils, and it has been estimated that for every 1% soil organic matter, soil can hold 10,000 - 11,000 liters of plant - available water per ha of soil down to about 30 cm soil depth (Gomiero et al., 2011) . In our organic treatment study, the soil organic matter was 0.3% higher than that in the conventional treatment, when soil organic carbon (Figure S 4. 3) converted to organic matter. Cover crops are grown und er organic treatment also may have contributed to soil moisture conservation (Basche et al., 2016; Yang et al., 2020) . 74 Finally, the effectiveness of the reduced input treatment to mitigate drought occurrences (moderate, severe, and extreme) is limited comp ared to other treatments (Figures 4. 5 and 4. 6). This may be due to its poor soil biology resulted from herbicide applications when compared to the organic treatment (Basche and DeLonge, 2019) and tillage operation compared to the no - till treatment. However , it was not clear why the reduced input treatment had lower effectiveness to reduce drought severity than the conventional treatment. In summary, these findings highlight that the no - till treatment is the best, the organic treatment is better, and the red uced input treatment is worst when compared to the conventional treatment in terms of the effectiveness of lowering agricultural drought. 4. 4 Conclusions Understanding the factors that affect agricultural system resiliency are critical in designing adaptation measures in the face of increasing climate variability and change. Soil moisture primarily controls the productivity of rainfed agricultural systems; soil moisture dynamics can be an indicator of resilience. Building on this, we evaluated the applicability of soil moisture - based metrics to gauge resiliency of three promising rainfed agricultural treatments, namely the no - till, the organic, and the reduc ed input treatments against the conventional treatment. The results of this study showed that among the metrics of temporal dynamics of soil moisture, MRD, and r s were suitable to gauge the resiliency of different agricultural systems. In contrast, ITS was not applicable since it cannot capture relative soil moisture dynamics. In addition, SWDI was able to gauge the resiliency of different treatments in terms of the effectiveness of reducing agricultural drought severity. Based on MRD and r s , the no - till t reatment had the highest resilience to climate variability among the studied treatments due to maintaining the highest level of soil moisture 75 during the growing season. Moreover, the effectiveness of the no - till system on agricultural drought mitigation wa s superior among the other treatments. This improvement in resilience due to the no - till treatment was reflected not only in the significantly higher yields of crops even during dry years, but also on stable crop growth and yields during years with climate extremes. Concerning the improvement in resiliency from promising treatments, organic treatment was the second - best treatment after the no - till. Significantly greater MRD in corn and soybean and higher r s were observed with the organic treatment compared to the conventional treatment. Furthermore, the organic treatment was inferior to the no - till treatment but was superior to the conventional treatment on reducing drought severity. However, this improvement was not evident on significantly higher biomass or yield due to nitrogen limitation to the crops in the organic treatment (Robertson et al., 2014) . However, organic soybean yield was equivalent to the conventional treatment even during dry years because of the ability of soybean to fix nitrogen (Hungria and Mendes, 2015) . The greater soil moisture resilience in the no - till and the organic treatments can be attributed to the improvement in soil physical properties and soil organic carbon that improves soil water storage in these management systems (Gomier o et al., 2011; Lal et al., 2012) . This was observed in our study as well where we found significant positive correlation between MRD and soil organic carbon. Equal performance of MRD in the reduced input treatment to the conventional treatment was refle cted as the equivalent growth and yield of corn and soybean but not in wheat. The reason for the significant reduction on growth and yield of wheat in the reduced input treatment, compare to the conventional treatment, may be associated to the lack of nitr ogen for wheat as it is planted in the fall immediately following the soybean harvest, which leaves the relatively 76 lower amount of nitrogen - rich crop residue for the wheat crop in the reduced input treatment. In contrast, corn and soybean follow nitrogen - f ixing red clover winter cover crop (Robertson et al., 2014) . Moreover, the effectiveness of the reduced input treatment to mitigate drought occurrences was also limited compared to other treatments. In summary, this study showed that MRD, r s , and SWDI are applicable in combination to evaluate resilience in different rainfed agricultural systems as related to the soil moisture content, growth, and yield. The no - till system had the highest resiliency than the conventional treatment in terms of higher soil moisture retention, higher effectiveness for drought mitigation, larger crop yields, and increased stability of yields. The yields in the no - till treatment were 15%, 18% and 4.3% greater than the conventional treatment for corn, soybean and whe at, respectively. Although the organic treatment had substantially higher resiliency in terms of grater soil moisture retention and drought mitigation than the conventional treatment, yields were significantly lower, especially for cereals (i.e., corn and wheat) as a result of nitrogen limitation. Even though the yields of corn and soybean in the reduced input treatment were comparable to those in the conventional treatment, the reduced input treatment had the limited capacity to recover from extreme condit ions and improve resiliency in terms of soil moisture retention and drought mitigation. Finally, the proposed approach here can be improved in future studies by increasing the frequency of soil moisture measurements over the growing season at different dep ths of the root zone. In addition, we are recommending the expansion of the study to larger spatial scales to better capture the robustness of these metrics under a variety of rainfed agriculture systems in the USA Midwestern region and around the world. 77 4. 5 Acknowledgment Financial support for this research was provided by KBS - LTER Summer Graduate Fellowship 2018 and C.S Mott Predoctoral Fellowship in Sustainable Agriculture. LTER data was funded by the National Science Foundation Long - term Ecological Res earch (NSF - LTER) Program (Grant DEB 1637653) at the Kellogg Biological Station and by Michigan State University AgBioResearch. This work was also supported by the USDA National Institute of Food and Agriculture, Hatch project MICL02653. 78 5. EVALUATING TH E CLIMATE RESILIENCE IN TERMS OF PROFITABILITY AND FARM RISK MANAGEMENT FOR A LONG - TERM CORN - SOYBEAN - WHEAT ROTATION UNDER DIFFERENT TREATMENT SYSTEMS 5. 1 Introduction Growing anthropogenic activities that generate greenhouse gas emissions have caused glob al warming and lead to significant changes in the climate system. These changes include increasing ambient temperature, precipitation variability, and the frequency of extreme events such as droughts, floods, and heatwaves (IPCC, 2018) . In many regions of the world, climate change and extremes have reduced productivity of major food and feed crops (Adhikari et al., 2015; Challinor et al., 2014; Lesk et al., 2016; Rojas - Downing et al., 2017) , leading many commentators to proclaim that climate change, along w ith the need to feed a human population of about 9.8 billion by the end of 2050 , poses a significant threat to regional and global food security (Branca et al., 2013; Pradhan et al., 2015; Rojas - Downing et al., 2017) . US agriculture is not immune to climate change. Despite improved plant genetics and management practices that have brought about long - term improvements in yields, extreme weather events tied to climate change may offset these yield gains (Hatfield et al., 2018; Ortiz - Bobea et al., 2018 ; Schlenker et al., 2006). Furthermore, climate variability is a major cause for annual variations in crop yields in the Midwestern United States (Hatfield, 2012; Wang et al., 2016) and accounts for more than 60% of the yield variability of major food crops (Ray et al., 2015). The Midwestern US is one of the most productive and economically important agricultural regions in the world but is increasingly experiencing extreme weather events tied to climate change, including drou ghts and floods (Andresen et al., 2012; Hatfield et al., 2018) . Projecting into the future, an ensemble of eight climate models showed that the frequency of drought in the 79 US Midwest is to increase from the present rate of once every five years to once eve ry other year by 2050 (Jin et al., 2018) . Such projected drought events are expected to drive yield losses of corn, soybean, and wheat outpacing the rate of productivity gains through improved CO 2 fertilization, cultivars and agronomic practices (Jin et al., 2018; Lobell et al., 2014; Troy et al., 2015; Zipper et al., 2016) . Such effects may result in amplified economic swings for consumers and producers. For example, the 2012 North American drought - impacted Southern states the most, where corn prices res ponded with a 53% increase in 2012 - 2013 relative to the previous 5 - year average price and by 146% compared to the decade of 2000 - 2009 (Boyer et al., 2013) . The substantial extent of the row crop operations in the Midwest is under rain - fed production. Because precipitation is the primary source of moisture in rain - fed agriculture, the impacts of droughts are much more significant in such systems where farmers have limited options for irrigation (Kuwayama et al., 2019; Sweet et al., 2017) . Be sides impacting plant available moisture, Midwest growers must contend with the timing of extreme moisture events, as extreme rain events tied to climate change are most apt to occur during planting and harvesting times (Tomasek et al., 2017) , times wher e timely access to fields is paramount to maintaining productivity and profitability of their farms. Recent studies of Midwest farmers show that growers recognize the changing climate (Arbuckle et al., 2013; Doll et al., 2017; Mase et al., 2017) and have, therefore, implemented different adaptation measures (Denny et al., 2019; Morton et al., 2015) . For example, farmers who better understand the impacts of climate change and are able to attribute it to the anthropogenic activities are increasingly using cli mate information, conservation practices, crop insurance and other alternative techniques for mitigating risks to climate change (Arbuckle et al., 2013; Haigh et al., 2015; Mase et al., 2017; Morton et al., 2017) . However, recognition of the threats due to climate 80 change and the uptake of climate - resilient practices still wanes (Church et al., 2017; Lemos et al., 2014) . Therefore, the path to greater food stability and security may be a greater concerted effort to promote climate change - resilient practices. Climate resilience can be defined as the ability of an agricultural system to keep up with its structures and to ensure provisioning of its functions in the face of climate variability and extremes. For agriculture, this can be done by improving resilien ce capacities, namely robustness, adaptability, and transformability (Meuwissen et al., 2019; Tendall et al., 2015; Urruty et al., 2016) of agricultural practices . Robustness is the capacity of the agricultural system to withstand climate extremes. Adaptab ility is the capacity to change the agricultural practices, marketing, and risk management to reduce the impacts of climate extremes without altering the structures and feedback mechanism of the system. Transformability is the capacity to substantially cha nge the structures, feedback mechanisms, and functions in response to climate extremes (Meuwissen et al., 2019) . Therefore, the goals of improving climate resilience include not only increasing the productivity of crops in the face of climate extremes but also improving the ecosystem services provided in nature while minimizing the environmental degradation from farm - related activities (Peterson et al., 2018) . Alternative agricultural systems have been endorsed to promote climate resilience compared to the conventional agriculture (Branca et al., 2013; Michler et al., 2019; Scialabba and Müller - Lindenlauf, 2010; Tuomisto et al., 2012) . These alternative systems may include no - till/reduced tillage practices, crop rotations, cover crops, reduced applications of inorganic inputs, and certified organic crop production. According to some researchers, such production systems have the potential to mitigate environmental pollution and greenhouse gas emissions, while substantially improving the soil and water quality in the agricultural ecosystems (Behnke et al., 81 2018; Martens et al., 2015; Syswerda and Robertson, 2014) . However, these treatments have not gained sufficient popularity among farmers due to higher risks and profitability concerns (Mausch et al., 2017; Roesch - McNally et al., 2018) and general resistance to changes in farm practices (Fleming and Vanclay, 2010 ; Takahashi et al., 2016) . Some even assert that old practices are strongly reinforced by the markets, legislation, and agribusiness companies that greatly benefit from this intensive system (Roesch - McNally et al., 2018) . Hence, conventional agriculture is still widely used regardless of being vulnerable to climate extremes, the cause of environmental pollution, and a source of carbon emissions (Bennett et al., 2014; Foley et al., 2011) . Farm profits are an important source of family and regional incomes. For some, farm income may be the sole source of family earnings. Farm earnings are measured in farm profits or the differences between farm income and farm expenses. Maintaining a consistent and predictable flow of annual earnings is desirable (Martens et al., 2015) . Therefore, climate - induced variations implement climate - resilient practices. However, there are costs and risks to adopting new practices, and growe rs should target long - term, economic resilience when considering what practices to adopt (Kumar et al., 2016; Mausch et al., 2017; Sain et al., 2017) . Economic resilience can be quantified using both the mean and volatility (Abson et al., 2013; Browne et al., 2013) of the expected net returns or profitability where a system with higher mean and lower volatility can be taken as a relatively resilient system. In general, there are tradeoffs between profitability and farm risk management as different farmers reflect different behaviors towards risk (Brink and McCarl, 1978; Lu et al., 2003) . For example, a system may be solidly profitable but risky due to the interannual variations on net returns and/or market demands. Such a venture with high expected returns but a high degree of variation in expected outcomes 82 may not be pursued, depending on risk tolerances (preferences) of the farmer. Profitability also affects the costs and benefits of crop insurance policies, which is considered to be an important tool for climate risk management (Annan and Schlenker, 2015; Tack and Ubilava, 2015) . Therefore, both profitability and risks should be evaluated to identify climate - resilient systems that can be successfully transformed into farm - level adaptation. Data from long - running agricultural experiments are befitting an evaluation of the economic resilience and risks associated with alternative agricultural systems, as they can capture historical climate information and its impacts on farm net returns. The Main Cropping Sy stem - Term Ecological Research program provides an effective backdrop to look at the long - term effect of climate on crop resilience under alternative agricultural treatment systems. Three alte rnative treatment systems (i.e., no - till, reduced input, and organic) for a rotational row crop production system (corn - soybean - wheat) were compared to conventional treatment. In a similar experiment, Swinton et al. ( 2015) compared the profitability of the se treatments for the period of 1993 - 2007. That study compared the economic values of ecosystem services in the context of climate change. The current study augments Swinton et al. ( 2015) by broadening the time - ec onomic incentives for adopting alternative agricultural practices as it relates to economic returns and risks to those returns. Accordingly, this study was designed to evaluate the climate resilience of three alternative treatment systems in terms of lon g - term profitability and risks compared to the conventional system. To accomplish this task, the following objectives and hypotheses were formulated and tested. 1) to evaluate the effects of climate variability on farm net returns under different productio n and treatment systems here we hypothesize that c limate variability has a significant 83 impact on expected net returns and a treatment system with the higher net return and lower volatility will have greater climate resilience for a production system; 2) to evaluate the risk level for the adaptation of different production systems under different treatments here we hypothesize that one treatment system dominates the other s in terms of both risk and net return and would offer different adaptation alternatives for farmers depending on their risk preferences. 5. 2 Materials and Methods Profitability and associated risks are the outcomes of climate resiliency and determine farm - level adaptation in agricultural systems (Mausch et a l., 2017) . In this study, we evaluate the climate resilience of three alternative treatment systems implemented in a long - term corn - soybean - wheat rotation in comparison to the conventional treatment. Expected profits and annual variation in expected profit s, as a measure of economic risk, are used as evaluation metrics. Annual crop management and production data during the period of 1993 - 2018 (26 years) were collected, and the net returns for each production and treatment systems were estimated via static e nterprise budgeting. First, we quantify the effects of climate variability on the means of net returns using a statistical mixed model and volatility of net returns using relative standard deviation (Objective 1). We judge resiliency based on high expected net revenues and low expected annual variation in net revenues by defining the resilient system as the one which has the highest expected net return and lowest volatility. Finally, we assess the risk level of different production and treatment systems as affected by climate variability using the payoff matrix tool (Objective 2). This will enable us to identify if a climate - resilient system can be welcomed among the farmers with a wide range of risk preferences for adaptation. 84 5. 2.1 Study Location and Det ails of the Experiment This study was conducted using long - term data generated at KBS located in Southwest Michigan, United States. This region represents a humid continental climate and the Köppen c limate c lassification subtype of "Dfa" (Peel et al., 200 7) . Soils at KBS mainly consist of Kalamazoo (fine loamy) and Oshtemo (coarse loamy), both mixed, mesic Typic Hapludalfs that mainly differ in the thickness of the Bt horizon (Syswerda and Robertson, 2014) . A series of Long - Term Ecological Research experiments have been established at KBS. One such experiment is the Main Cropping System Experiment (MCSE), which is in the geographic coordinates of 42.41 N, 85.37 W, and the altitude of 288 m AMSL (Figure 5. 1). Figure 5. 1 . The location of the Main Cropping System Experiment (MCSE) in Southwest Michigan, United States. The MCSE began in 1989 and c onsisted of seven agricultural treatment systems. First four are annual row crop production systems (corn - soybean - wheat rotation), which are under differen t intensities of management and the remaining three are the perennial systems namely , p oplar ( Populus deltoides × Populus nigra ), continuous a lfalfa ( Medicago sativa ) and an early successional vegetation community (Robertson and Hamilton, 2015) . Although the MCSE 85 started in 1989, an appropriate experimental design was followed only from 1993; therefore, this study involves the first four annual row crop production systems of the MCSE for the period of 1993 - 2018. The four annual row crop product ion systems comprised of corn ( Zea mays ) - soybean ( Glycine max ) - winter wheat ( Triticum aestivum ) in rotation and managed as (i) conventional (CON) , (ii) no - till (NT), (iii) reduced input (RI) , and (iv) US Department of Agriculture (USDA) certified organic ( OR) treatments . Hence, here we consider different crops as production systems while different management as treatment systems. All these production and treatment systems have been managed as rainfed , and each of them is assigned to six replicates (blocks) of one - ha, in a randomized complete block design. Moreover, crops have been planted and harvested at the same time for each treatment and replicate during their usual growing seasons. During the cropping rotations of corn and soybean, CON, NT, RI treatment s have been planted with Roundup - ready seeds while the OR treatment has been planted with conventional and untreated seeds. There was no difference between the treatments on seed types during the winter wheat cycle. The timing of management operations of t his corn - soybean - wheat rotation under each treatment is presented in supplementary materials (Table S 5 . 1 - S5.4 ). The detail agronomic management of the different treatment systems is as follows: Conventional treatment : Crops were planted following the primary tillage and soil finishing. Primary tillage was applied using moldboard plo ugh until 1998 and after that using chisel plough. Secondary tillage was applied by disking before planting during the years when wheat cr op is planted. Inter - row cultivation was performed for corn and soybean. Fertilizer application rates were based on the soil - test recommendations for each crop. Weeds were 86 controlled by a broadcast application of appropriate herbicides, depending on the we ed intensity in each crop. No type of manure or compost or insecticides was applied. No - till treatment : Crops were planted under zero tillage operations using a no - till drill. Fertilizer application rates were based on the soil - test recommendations for each crop. Weeds were controlled by the broadcast application of appropriate herbicides, depending on the weed intensity in each crop. No type of manure or compost or insecticide was applied. Reduced input treatment : Crops were planted following the primary tillage and soil finishing. Primary tillage was applied using moldboard plo ugh until 1998 and after th at using chisel plough. Secondary tillage was applied by disking before planting during the years when wheat crop is planted. Inter - row cultivation was performed for corn and soybean. Nitrogen fertilizer and herbicide rates were applied as one - third of nit rogen and herbicides applied to the conventional treatment . H erbicides were not broadcasted but banded within rows . Phosphorus and potassium fertilizer application rates were based on the soil - test recommendations for each crop. A w inter cover crop was pla nted following the corn and wheat crops of the rotation to supply nitrogen to the following crop. Commonly, cereal rye ( Secale cereal ) was planted following corn, while red clover ( Trifolium pratense ) was planted following wheat. No type of manure or compost or insecticide was not applied. Organic treatment : Crops were planted following the primary tillage and soil finishing. Primary tillage was applied using moldboard plo ugh until 1998 and after tha t using chisel plough. Secondary tillage was applied by disking before planting during the years when wheat crop is planted. Inter - row cultivation was performed for corn and soybean. A w inter cover crop was planted following the corn and wheat crops of the rotation to offer nitrogen to the following crop. Commonly, cereal rye ( Secale cereal ) was planted following corn, while red clover 87 ( Trifolium pratense ) was planted following wheat. No manure - based fertilizers were applied. This is a USDA certified organi c treatment; therefore, no chemical fertilizers/herbicides/insecticides were applied. 5. 2.2 Enterprise Budget Analysis to Estimate Net Return under Different Production and Treatment Systems Calculation of cost component : Data on crop management, quantities of inputs, and the rates of application for all four treatments for the period of 1993 - 2018 were obtained from the agronomic log of the KBS. Variations in operating costs across treatments were estimated using 2018 ent erprise budgets from Clemson University Cooperative Extension for corn, soybean, and wheat (Clemson, 2020) . Annual operating costs were static using 2018 prices and reflect the primary operating costs for each treatment . While the use of static prices simp lifies the analysis, it also holds constant commodity prices highlighting the role of variations in cropping systems inputs and practices in determining economic outcomes as affected by climate variability. Additionally, though other well - established cro p enterprise budgets exist for modeling economic practices across all crops in this analysis and provided sufficient detail required to model the economics of diff erent treatments modeled here. Budgets with this level of detail were not available for Michigan - specific farm operations. Clemson University enterprise budgets were modified based on each production and treatment system. To this end, farm machinery opera tions, and crop scouting were valued using the 2018 custom machine and work rate estimates of Michigan State University Extension (Battel, 2018) . Costs captured for some of the farming activities in the previous year for the following crop were included in the budget of the main cropping year in which the harvest 88 occurs. For example, costs of land preparation and planting needed for wheat and for establishing cover crops to benefit corn and soybean during the previous year were considered in the calculation of the main cropping year budget of the respective crops. Michigan crop insurance payments for each commodity were used to account for the annual crop insurance amounts. The fixed cost, which generally involves land rent, farm machinery, farm insurance, a nd any overhead costs were excluded in the budget analysis due to complexity in making reasonable estimates for this research experiment - based management system. Therefore, the cost component consists only of the variable cost. Calculation of return s : Crop yield data was collected from the MCSE for the individual replicates of each production system. Crops from the conventional, no - till, and reduced input systems were priced using conventional commodity prices, while organic crops were priced using the orga nic commodity prices (Table 5. 1). Prices for the conventional and organic commodities were obtained from the Agricultural Resource Management Survey (ARMS) of USDA (USDA, 2020) . Recent ARMS commodity price data was available for the corn, soybean, and whea t during the years of 2010, 2006 , and 2009, respectively. Hence, ARMS price data was used as static commodity price in this study because of its sufficient statistical reliability and the availability of organic commodity prices. Expected annual profits we re estimated as the calculated net returns over the 26 years of data. Expected gross revenues were estimated from projected gross revenues, comprised of yield and crop prices (Table 5. 1). Expected costs were fixed based on fixed input costs specific to eac h treatment. The expected net return was then estimated as the average difference between projected gross revenues and costs. 89 Table 5. 1 . Conventional and organic commodity prices used to value the crops in different treatment systems*. Crop Conventional price (US dollar /bu shel ) Organic p rice (US dollar /bu shel ) Corn 4.33 7.13 Soybean 5.52 14.20 Wheat 4.65 8.02 *Note: Conventional and organic prices of corn, soybean, and wheat were obtained from ARMS conducted during the years of 2010, 2006, and 2009, respectively (USDA, 2020) . 5. 2.3 Categorization of Climate Variability The study period (1993 - 2018) was classi fied into dry, normal, and wet categories of climate variability based on seasonal precipitation that covers primary growing season (April - October). To do this, thirty years (1989 - 2018) of total seasonal precipitation data collected from the KBS weather st ation located within the experimental site . This is because climate variability should be analyzed for a relatively long period comprising at least three consecutive ten - year periods (WMO, 2017). Annual seasons were categorized as follows: i) dry years when the cumulative probability of aggregate seasonal precipitation is less than 33.3%; ii) normal years when the cumulative probability of aggregate seasonal precipitation is in between 33.3% and 66.6% and iii) wet years when the cumulative probability of aggregat e seasonal precipitation is greater than 66.6% . These breakouts are as shown graphically in Figure 5. 2. Accordingly, dry years were the years that the seasonal 90 years that seasonal precipitation is in between. Classification of the study period into the categories of climate variability for each crop in the rotation is also shown in Figure 5. 2. Fi gure 5. 2 . The probability distribution of total seasonal precipitation (1989 - 2018) at the KBS weather station. Note: The dark solid blue line indicates the cumulative probability, while the bar graph shows the total seasonal precipitation. Years and culti vated crops are given inside the bars following the categorization of climate variability. 5. 2.4 Evaluating the Effects of Different Production and Treatment Systems and Climate Variability on the Expected Net Return A statistical mixed model, which incor porates both fixed and random effects of relevant independent variables (Milliken and Johnson, 2009) , was used t o evaluate the effects of the treatment systems and the climate variability on the net return. The statistical model of our estimates (i.e., net return) was specified as: y ijk = µ + clim + trt i + yr k + trt i × clim + blk j ( yr k ) + trt i × yr k + e ijk , ( 5. 1) where , y ijk is the vector of expected net returns, or annual profit, calculated for the i th treatment, within j th block on the k th year , µ - is the mean, trt - is the fixed effect of treatment , blk - is the random effect of the replicates/ blocks nested within years ( yr ) , clim - is the fixed effect of 91 the climate variability (dry, normal and wet) , yr - is the random effect of the year, trt × clim - is the interaction term between the effects of treatment and the fixed effect of climate variability, trt × yr - is the interaction term between the effects of treatment and the random effect of the year , and e - is the residual. The estimated net return data passed t he normality test for the residuals and temporal homogeneity of variances (Milliken and Johnson, 2001) . Statistical analysis w as performed by the production syste m (separately for each crop), using PROC GLIMMIX procedure (Milliken and Johnson, 2009) in the SAS software (V ersion 9.4 SAS Institute Inc. Cary, North Carolina, US). Fixed effects were tested by F - test (Steel et al., 1980) , and random effects were tested by Wald test for covariance (Wald, 1943) . Tukey Test for mean separation was performed when significant differences were detected for the fixed effects (Tukey, 1977) , and significant differences were evaluated at the 0.05 probability level (Steel et al., 1 980) . 5. 2.5 Defining Climate Resilience As growers will take into account both the expected returns to a production system and the expected risks, in this section we define climate resilience based on both the means and volatility of the expected net returns. 5.2.5.1 Defining Climate Resilience Based on the Means of Expected Net Returns Of the four treatments modeled, the conventional tillage is considered as the baseline, or control case, for which alternative treatments are compared. The conventional system was taken as the control because of its widespread practice in present agricultu re throughout the world (Kassam et al., 2019) besides being vulnerable to climate change and extremes. We estimate expected economic returns as a mean of net return for each treatment system. Statistical significance was tested for each climate category - c ommodity combination (dry, normal and wet corn, soybean, and wheat), resulting in nine cases for each of the four 92 treatments. A treatment is deemed resilient to a climatic condition if the treatment provides six or more returns that are statistically hig her than the conventional system under all climatic conditions. A treatment system was identified as non - resilience if it performs worse than the conventional treatment in four or fewer cases as a measure of the mean of net return. A treatment system with five cases of net returns that significantly higher to the conventional treatment is considered as moderate resilience. 5.2.5.2 D efining Climate Resilience Based on the Volatility of Expected Net Returns The volatility of expected net return is another in dicator of climate resilience where an agricultural system with stable expected revenues can be more resilient while an agricultural system with volatile expected revenues will be less resilient to climate perturbations (Abson et al., 2013; Gil et al., 201 7; Urruty et al., 2016) . Relative standard deviation (RSD) can be used to quantify the volatility of agricultural outcomes (Rigolot et al., 2017) . RSD is the normalized measure of the dispersion of a probability distribution, which is defined as the ratio between the standard deviation and the absolute mean, presented in percentage (Abson et al., 2013) . The absolute mean is used in calculating RSD instead of the coefficient of variation to avoid negative measures of variation. In this way, RSD estimates are scale - invariant and comparable across all climate category - commodity combinations. In this study, we calculated the RSD for each treatment under each category of climate variability for all three production systems. Moreover, the distribution of expected profits for each treatment, as affected by climate variability, was also presented. Here, we define the resilient treatment as the one that holds a higher expected return with lower RSD and variability. 93 5.2.6 Evaluating the Risk Level for the Adaptation of Different Production and Treatment Systems by Producers Risk is defined as the chance of adverse outcomes associated with an action (Nelson, 1997) . While making decisions for the adaptation of different production and treatment systems, producers would pr efer to avoid risks. To understand the risks associated with each potential adaptation actions, producers require decision - making tools that permit them to incorporate uncertainty and risk into their adaptation planning. To this end, a payoff matrix can be considered as a suitable decision - making tool to analyze adaptation decisions in terms of alternative actions, possible events, and payoffs (Hoag and Parsons, 2010; Nelson, 1997) . We used the payoff matrix to evaluate the risk level for the adaptation of different production and treatment systems. The producer (decision - maker) can choose among alternative actions with differing predicted reward/risk structures that refer to different treatment systems for each production system. However, the expected outc omes (i.e., net return) from each action may depend on uncertain events, which represents climate variability impacts on grower net returns, for each of the commodity - treatment combinations. This uncertainty information can be incorporated into the payoff matrix as probabilities (Nelson et al., 1978; Senapati, 2020) . According to 30 - year (1989 - 2018) seasonal precipitation data from the KBS weather station (section 2.4.1), and the probability of having dry, normal, and wet years for a three - decadal time fram e was 0.3, 0.3, and 0.4, respectively. Producers can make decisions, either by incorporating the probability of uncertain events or not incorporating these events. There are two types of risk preferences among the producers who do not incorporate the prob abilities of uncertain events into their decision - making process. i) Risk - averse decision - makers: producers who select the best among the worst outcomes for 94 each adaptation action; ii) Risk - loving decision - makers: producers who select the best among the be st outcomes for each adaptation action. Producers who do incorporate the probabilities of uncertain events into their decision - making process can also be classified into two groups. i) Risk - neutral decision - makers: producers who maximize the expected monet ary value (EMV), which was calculated based on the probabilities of climate variability; ii) Risk - averse decision - makers: producers who maximize the EMV subject to some minimum level of safety. Here we decided the safety level as a minimum net return of 15 0 USD for at least 95% of the time. The above decision - rules were based on Nelson et al. ( 1978) and used to identify the treatment systems which can be selected by the producers with different risk preferences. 5. 3 Results and Discussion 5. 3.1 Effects of Treatment Systems and Climate Variability on the Net Return of the Different Production Systems The statistical mixed model (equation 1 ) was used to analyze the net returns taking it as a response variable for the individual production system (i.e., crop type ) . The probability of the fixed and random effects on the net return on each production sy stem is presented in Table 5. 2. Accordingly, the effect of treatments ( trt ) on the net return was significant ( p <0.01) for all crops. T he effect of climate variability ( clim ) was strongly significant ( p <0.0001) in soybean. However, factor variables entered the equation significantly, either directly or through mediated effects, as components of interaction terms. The interaction between treatment and climate variability ( trt × clim ) was significant ( p <0.01) for all crops , which indicates that the performanc es of treatments change in different climate conditions. Therefore, the effects of treatments on each production system were further investigated under the categories of climate variability, and this differential performance was used to evaluate the resili ence under section 3.2. Furthermore, the interaction 95 between treatment and year ( trt × yr ) was significant ( p <0.01) on the net return for corn and soybean. The interaction between treatment and year can be due to the interannual variations of crop yield, types, and rates of the applications of herbicides and fertilizers and that could lead to the interannual variations in the net return. Table 5. 2 . Probabilities for the effects evaluated in the statistical mixed model for the net return . Production system Effects in the statistical model Probability ( p - value) * Corn Treatment ( trt ) 0.007 Climate variability ( clim ) 0.413 Year ( yr ) 0.051 Interaction between treatment and climate variability ( trt × clim ) 0.0001 Interaction between treatment and year ( trt × yr ) 0.006 Soybean Treatment ( trt ) <0.0001 Climate variability ( clim ) <0.0001 Year ( yr ) 0.118 Interaction between treatment and climate variability ( trt × clim ) 0.002 Interaction between treatment and year ( trt × yr ) 0.003 Wheat Treatment ( trt ) 0.006 Climate variability ( clim ) 0.152 Year ( yr ) 0.294 Interaction between treatment and climate variability ( trt × clim ) 0.005 Interaction between treatment and year ( trt × yr ) 0.278 * Bold values denote statistical significance at the p<0.05 level. 5. 3.2 Resilience of Production and Treatment Systems as Measured by the Net Return The means of net returns for different production and treatment systems, as affected by climate variability, are presented in Table 5. 3. For a total of nine cases for each treatment, the organic treatment has significantly highe r net returns than the conventional system (eight cases). The no - till treatment had seven cases where the net returns were significantly higher than that of the conventional treatment. However, the reduced input treatment had only three cases of 96 significan tly higher net returns in comparison to the conventional treatment. Therefore, the findings assert that organic and the no - till treatments are resilient systems, while the reduced input system is not. Although the no - till treatment performs superior to al l other production systems on yield, in the majority of the crop and climate variability combinations (Table S 5.5 ), the organic treatment overrides the economic performance because of the relative revenue - cost differential, driven by higher per - unit sellin g prices and marginally lower production costs of organics (USDA, 2020) . When averaged across crops and the categories of climate variability, the mean net returns for the conventional, no - till, reduced input, and organic treatment s were 22 USD ha - 1 , 187 U SD ha - 1 , 13 USD ha - 1 , and 454 USD ha - 1 , respectively. This reinforces the findings of Swinton et al. ( 2015) , who showed that organic treatment dominates on profitability when it has been valued at the organic prices. When the net returns are averaged across treatments and categories of climate variability, the corn production system had the highest net return of 367 USD ha - 1 followed by wheat (1 32 USD ha - 1 ) and soybean (2 USD ha - 1 ). 97 Table 5. 3 . Means of net return under different production systems as affected by climate variability*. Production system Treatment system Expected Net return (US Dollars per hectare) Dry years Normal years Wet years Corn CON (50.30) c 468.40 b 463 .00 b NT 80.10 b 746.40 a 692.00 ab RI 66.30 b 310.70 c 478.00 b OR 586.60 a 340.60 bc 742.00 a Soybean CON ( 500 . 8 0) c ( 318.10 ) c (66.30) c NT ( 324.60 ) b ( 112.60 ) b 74.40 b RI ( 608.10 ) d ( 405.90 ) d ( 112.70 ) c OR 99.60 a 187.40 a 904.70 a Wheat CON (4.40) c 55.50 d 130.10 b NT 173.10 a 193.60 b 58.20 c RI 41.20 b 152.90 c 91.70 c OR 158.80 a 302.30 a 239.40 a * Means with the same letter in a single column for each crop are not significantly different at p<0.05. Means of net return with negative values are presented in parenthesis. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; and OR: Organic treatment. Means of net return, which are significantly higher in c omparison to the conventional treatment, are presented in bold letters. The stability of farm profits is directly related to resilience (Cabell and Oelofse, 2012) . The volatility, as measured by the RSD was presented against the net return in a scatter plo t in Figure 5. 3. Accordingly, the organic treatment can be identified as climate - resilient during dry and wet years, while the no - till treatment shows resiliency during the normal years in the corn production system. In the soybean production system, the o rganic treatment can be found as climate resilience across dry, normal, and wet years. Meanwhile, in the wheat production system, the organic treatment shows climate resiliency during wet and normal years, while the no - till system is climate - resilient duri ng dry years (Figure 5. 3). 98 The pattern of net returns for the study period as affected by climate variability in corn, soybean, and wheat production system is presented as boxplot in Figures 5. 4, S 5. 1, and S 5. 2, respectively. Overall, variations in net re turns are largest in corn while smallest in soybean. This may be due to higher interannual yield variability in corn compared to other crops, as observed by Kravchenko et al. ( 2005) and Porter et al. ( 1998) . Net returns were positive (gain) for corn and wh eat for all scenarios except for the conventional treatment during dry years. In contrast, the net returns were negative (loss) in all scenarios in soybean except for the organic and the no - till treatments during wet years. Despite lower yields in th e organic treatment (Table S 5. 5), organic offered higher net returns mainly because of premium prices and lower production costs compared to other treatment s, as highlighted by Nemes ( 2009) . Our findings also support those of Toliver (2010), who demonstrat es no - till treatment is profitable than the conventional system. This is mainly due to the higher yields in the no - till treatment in comparison to the conventional treatment (Table S 5. 5). No - till also has lower production cost from reduced tillage operations, which for chisel/moldboard plowing, can be cost intensive. 99 Figure 5. 3 . Scatter plot between the expected net return and the relative standard deviation for different treatments as affected by climate variability in different production syste ms. 100 Figure 5. 4 . Box plot of n et return of corn production system under different treatments as affected by climate variability . Note: The thick black line represents the median of the distribution; the red dot is the mean, the edges of the box are 25% and 75% quantiles, whiskers denote the range of the data, and the black dots are the outliers. 5. 3. 3 Producer Risk Levels for the Adaptation of Different Production and Treatments Systems The payoff matrix for the corn, soybean, and wheat production sy stems are presented in Tables 5. 4. According to the decision, rules explained in section 2.5, a risk - averse farmer who does not consider the uncertainty of climate variability, would prefer to select the organic treatment. In contrast, a risk - loving farmer would prefer to select the no - t ill treatment to produce corn. However, for a corn producer who considers the uncertainty of climate variability, risk - neutral action would be the selection of the organic treatment, while the risk - averse action could be the selection of either the no - till or the organic treatment (Table 5. 4). Regardless of the consideration of the uncertainty in climate variability, the organic treatment is shown to be the 101 selective action for the soybean and wheat producers with risk - averse, risk - loving, and risk - neutral preferences (Table 5. 4). No - till and organic treatments generally posit higher returns over most climatic conditions, though organic returns outperform others in adverse climate years, while no - till exhibits the highest returns for normal climate years. Ho wever, expected profits is only one consideration when selecting production practices. Risks, which generally relate to the minimum lower bound of outcomes, is also a factor. Those seeking strictly to minimize risk would select the same combination as thos e solely considering expected profits. That is, no - till provides the best downside, or risk, during normal years, while organic provides the best downside risk during adverse climatic years. Profitability is often positively related to risk (Tobin, 1958) ; however, this study shows that the organic treatment achieved higher profits with lower volatility, and showed appropriateness to cater to a range of farmers with different risk preferences. This is followed by the no - till treatment. According to the find ings of this study, the conventional and the reduced input treatments are poor substitutes to organic and no - till agriculture in terms of both expected net earnings and annual variations of those net earnings. 102 Table 5. 4 . Payoff matrix of productio n systems under different treatment systems as affected by climate variability . Production system Climate variability Probability Net Return (USD ha - 1 ) CON * NT RI OR Corn Dry year 0.3 (50.30) 80.10 66.30 586.60 Normal year 0.3 468.40 746.40 310.70 340.60 Wet year 0.4 463.00 692.00 478.00 742.00 Expected Monetary Value (EMV) * EMV CON = (0.3)(50.3)+(0.3)( 468.4 )+(0.4)( 463 ) 310.60 524.70 304.30 575.00 Soybean Dry year 0.3 (500.80) (324.60) (608.10) 99.60 Normal year 0.3 (318.10) (112.60) (405.90) 187.40 Wet year 0.4 (66.30) 74.40 (112.70) 904.70 Expected Monetary Value (EMV) * EMV CON = (0.3)( - 500.8 )+(0.3)( - 318.1 )+(0.4)( - 66.3 ) (272.10) (101.4) (349.30) 448.00 Wheat Dry year 0.3 (4.40) 173.10 41.20 158.80 Normal year 0.3 55.50 193.60 152.90 302.30 Wet year 0.4 130.10 58.20 91.70 239.40 Expected Monetary Value (EMV) * EMV CON = (0.3)( - 4.4 )+(0.3)( 55.5 )+(0.4)( 130.1 ) 67.40 133.30 95.00 234.10 Note: The worst outcome from each production and treatment system is in bold format . The best outcome from each production and treatment system is in italic format. The maximum EMV for each production system is in bold - italic format. * shows how EMV is calculated under the conventional treatment , and th e equation was used to calculate EMV in other treatments. The organic treatment commends a higher expected net revenue as a result of the current market value proposition afforded to the organic crops. The organic crop production practices, which consist of the use of non - genetically modified crops and avoidance of any synthetic agrochemicals, have comparably lower costs and higher revenues. Thus, more farmers can enter into the organic agriculture and increase the supply of organic produces. Of course, th e law of supply dictates that as entrants increase the supply of organic produce, the price will fall until all excess profits that can be gleaned from organics are exhausted. However, there are barriers to entry that may protect profits to the organic sec tor. Such barriers include the initial transition period and associated costs of three years to transition to certified organic production, where 103 farmers face declining transitional yields and revenues (Bravo - Monroy et al., 2016; Bowman and Zilberman, 2013 ) . Other barriers include surmounting the learning curve for managing a profitable organic operation. Though not a barrier to entry to organic, farmers are often unwilling to change current practices despite the potential for gains (Fleming and Vanclay, 20 10; Takahashi et al., 2016) . In addition, as organic operations become more common, abiotic stressors such as pests and diseases will likely increase production costs (Röös et al., 2018) , thereby reducing the net earnings advantage afforded by organics. Fi nally, there is the threat that widespread organic production has on aggregate food supply. If widespread organic production reduces aggregate farm productivity, the conventional market will shrink, giving rise to the price of non - organic foods, and thereb y eroding the organic food price advantage. That is, in the long run, many factors will be at play, impacting the future direction of the agri - food economy. 5. 4 Conclusions Identifying alternative agricultural practices to improve climate resi lience in current crop production systems is an urgent requirement to face the challenges of increasing climate extremes. Profitability and associated risks are the key determinants of climate resilience as they affect the decision of farmers for transitio n and adaptation. In this study, we evaluated the climate resilience of three alternative treatments, namely, no - till, reduced input, and organic treatments in comparison to the conventional treatment in the long - term (26 - years) corn - soybean - wheat rotation al production system in Southwest Michigan, US. Historical seasonal precipitation data was used to categorize the climate variability during the study period, and an appropriate enterprise budget analysis was conducted to derive expected annual net returns using the crop management and production data collected from the experimental station. A statistical mixed model was used to evaluate the effects of different treatments and climate variability on 104 the expected net returns. Mean and volatility of the expec ted net returns were used to define climate - resilient treatments. Additionally, the payoff matrix approach was used to identify the suitability of alternative treatments for farmers with different risk preferences. According to the findings of this study, the organic treatment showed the highest resilience, followed by the no - till treatment. The conventional and the reduced input treatments showed lower levels of resilience to climate change. The findings are significant in showing that no - till practices do minate conventional and reduced input practices in both expected annual net revenues with relatively lower risk to those revenues in light of climate change. It also shows that while organic production revenues are largely expected to exceed net revenues o f conventional food crops, for many commodities, organic systems may exert greater annual stability in revenues. However, market conditions assert that such an advantage is likely to wane over time as growers migrate to this more profitable option. Related to this is the question of why migration to organic has not occurred faster than experienced. Part of the reason may be the high transition costs going to organic, while another component likely arises from market structures that favor conventional practi ces and psychological barriers to significant disruptions of existing production practices. Overcoming these constraints will require policy and industry buy - in to alternative agricultural practices, including an expansion of crop insurance offerings, favo rable Natural Resources Conservation Service (USDA - NRCS) Environmental Quality Incentives Programs, and industry support for sustainably - produced food crops. 5. 5 Acknowledgment This research was f inancial ly support ed by KBS Long - Term Ecological Research Summer Graduate Fellowship in 2018, C.S Mott Predoctoral Fellowship in Sustainable Agriculture, and the USDA National Institute of Food and Agriculture Hatch project 105 MICL02653. KBS Long - Term Ecological Research experiment was funded by the National Science Foundation Long - term Ecological Research Program (DEB 1832042) at the Kellogg Biological Station and by Michigan State University AgBioResearch . 106 6. QUANTIFICATION OF RESILIENCE METRICS AS AFFECTED BY A CONSERVATION AGRICULTURAL PRACTICE AT A WATERSHED SCALE 6 .1 Introduction In similarity to other regions in the world, the Midwestern United States has already be en adversely impacted by climate change and variability (Andresen et al., 2012; Fuchs et al., 2015; Hatfield et al., 2018) , and the increasing climate extremes, such as droughts, are projected to increase in the future (Jin et al., 2018) . These extreme eve nts have lead to substantial crop yield losses (Hatfield et al., 2018; Wang et al., 2016) , affecting both producers and consumers. To counteract these drought extremes, groundwater based irrigation systems are widely used in the U.S. (Siebert et al., 2010) . However, extraction of groundwater for irrigation above the rate of recharge has significantly reduced groundwater levels, affecting the baseflow to streams, groundwater - fed wetlands, and other groundwater dependent habitats and species (Dalin et al., 20 17; Scanlon et al., 2012; Wada et al., 2010) . Therefore, there is an increasing consensus among researchers that the resilience and ecosystem services provided by agricultural production systems should be improved. Ecosystem services denote all the benefit s humans obtain from different natural systems for their physical and socio - economic prosperity (Costanza et al., 1997; Mengist et al., 2020) . Agricultural practices are responsible for the primary production of food and fiber, while providing numerous eco system services at different scales (Dale and Polasky, 2007; Power, 2010; Swinton et al., 2007; Tancoigne et al., 2014; Wood et al., 2015) . Comprehensive documentation of ecosystem services has been conducted within the framework of the Millennium Ecosyste m Assessment (MEA); accordingly, ecosystem services can be broadly 107 categorized based on provisioning, regulating, supporting, and cultural roles of the ecosystem (Fisher et al., 2009; MEA, 2005) . Supporting services are fundamental in nature; without them , other types of services cannot occur. Nevertheless, the current trend of agricultural intensification deliberately focuses on a few provisioning services (e.g., food, water, energy), through agricultural landscape simplification, rather than harnessing a range of ecosystem services (Bommarco et al., 2013; Gaba et al., 2015; Robertson and Swinton, 2005) which in turn affects the resilience and sustainability of the agricultural systems. This phenomenon is very common in the Midwestern United States (Landis , 2017) , which is one of the industrialized large - scale agricultural regions in the world , and contribut es significantly to global food security and the economy as it produces the majority of the U.S. row crops and several other food, feed, and fuel crops (Hatfield, 2012; Oppedahl, 2018) . Ecosystem services and resilience are interconnected, where the ecosystems with lower resilience are vulnerable to disturbances (e.g., climate perturbations) and higher resilience ensures a stable supply and/or recovery o f ecosystem services (Biggs et al., 2012; Fedele et al., 2017; Montoya and Raffaelli, 2010) . In other words, the loss of ecosystem resilience could compromise ecosystem services that are indispensable for sustainable agricultural production systems (DeCler ck et al., 2016; El Chami et al., 2020; Swift et al., 2004) . Therefore, increased resilience and ecosystem services can be seen as an opportunity for climate change adaptation and disaster risk reduction (Munang et al., 2013) . Improving agroecosystem services and resilience is not only confined to the farm scale, but can be expanded across the landscape (Bailey and Buck, 2016; Scherr et al., 2012) . For example, agricultural recharge, which is the water leaving the vadose zone fr om agricultural 108 farms, may contribute to groundwater - dependent wetlands, streams, and dependent species (Gordon et al., 2010; Sampath et al., 2015) beyond those farms. These groundwater - dependent systems deliver services such as microclimate regulation, wa ter for irrigation, flood mitigation, and control of pests and diseases (Griebler and Avramov, 2015; McLaughlin and Cohen, 2013) , which in turn enhance the resilience of agro - ecosystems. Although groundwater recharge is broadly considered as a provisioning service (Prudencio and Null, 2018; Serna - Chavez et al., 2014) , it is also indirectly linked to regulatory and support services. Therefore, recharge can be considered as a major water - related ecosystem service and can be used as a metric to evaluate resili ence in agro - ecosystems (Coates et al., 2013; Serna - Chavez et al., 2014) . R esilience signifies the ability of an agricultural eco system to maintain its structure and function in the face of disturbances (Walker et al., 2004) . The i nitial step of improving resilience is the asses s ment of resilience at appropriate scales. Resilience metrics are used to quantify resilience and can be used individually or in combination (Douxchamps et al., 2017; Serfilippi and Ramnath, 2018) . Commonly used resilience metrics ar e m eans and variance of agricultural production /yields (Di Falco and Chavas, 2008; Eeswaran et al., 2021; Martin and Magne, 2015) , profit /revenue (Browne et al., 2013; Kandulu et al., 2012; Komarek et al., 2015; Rigolot et al., 2017) , soil moisture (Eeswar an et al., 2021) , crop failure (Jones and Thornton, 2009) , and farming risks (Komarek et al., 2015) . No - till has been endorsed for enhancing ecosystem services such as carbon sequestration, greenhouse gas mitigation, microclimate regulation, control of nutrient leaching, soil erosion control and improving species richness (Lal, 2013; Robertson and Swinton , 2005; Syswerda and Robertson, 2014; Zhang et al., 2016) , often at the field scale. Considering all of the aforementioned benefits, there is an increasing trend in the adoption of no - till agriculture around 109 the world (Kassam et al., 2019) . However, there is a dearth of knowledge on how no - till affects the overall resilience at a larger scale. To fill this gap, we present an approach that integrates long - term field experimental data and modeling to evaluate an ecosystem service (i.e., groundwater recharge a nd water table) and resiliency (i.e., soil moisture, drought mitigation, yield, and net return) of convetinal and no - till practices in a large, diverse watershed. The objectives of this study are: 1) assess recharge, groundwater table , and soil moisture va riabilities for the long - term corn - soybean - wheat rotation under conventional and no - till practices at a watershed scale; 2) estimate yields and net returns under conventional and no - till practices within a large , diverse watershed; and 3) evaluate the over all changes in resiliency as affected by the adaptation of no - till as conservation agriculture. 6. 2 M aterials and Methods The modeling framework of this study is presented in Figure 6. 1. Initially, observed data from a long - term (1993 - 2019) cor n - soybean - winter wheat rotation experiment of both conventional tillage and no - till treatments were used to parameterize a crop model (i.e., the Decision Support System for Agrotechnology Transfer - DSSAT) (Jones et al., 2003) . Next, the DSSAT model was cali brated using the measured volumetric soil moisture and crop yield from the long - term field experiment. The calibrated DSSAT model was applied to individual fields within a large and diverse watershed. The results from the large - scale crop model were used t o calculate the annual recharge and resilience measures for individual fields. The simulated drainage from the crop model, i.e., the deep percolation from the bottom of the soil profile, was assumed to reach the water table instantaneously and act as rech arge from the agricultural land use (Xiang et al., 2020) . This assumption can be supported by the existence of permeable soils and strong connection between the surface and groundwater within the study 110 watershed (Grannemann et al., 2008) . Groundwater flow in the watershed was modeled using a process - based groundwater model called Inte ra ctive Groundwater (IGW) (Li and Liu, 2006; Liao et al., 2015 a ) and calibrated using static water level data. Finally, changes in the water table as ecosystem service and metr ics of resilience were evaluated as affected by the adaptation of a no - till treatment and compared to the base scenario (a conventional tillage treatment). Figure 6. 1 . An overview of the modeling process . 111 6. 2.1 Study Area Our research project comprises of both field experiments and modeling efforts. The following sections describe the study area for each of these efforts. 6. 2. 2 Description of Long - Term Field Experiment The DSSAT cropping system model (Jones et al., 2003) for the watershed scale evaluation was developed using the long - term experimental data collected from the Main Cropping System Experiment (MCSE) of the Kellogg Biological Station (KBS). KBS is located within the Kalamazoo River watershed in Michigan, U.S. at the coordinates of W and the altitude of 288m AMSL (Figure 6. 2). The annual precipitation at the KBS is about 1,027 mm, while the annual mean temperature is 10.1 C, ranging from the lowest monthly mean of - 9.4 C to the highest of 28. 9 C in January and July, respectively (Cusser et al., 2020) . This experimental site has fine loamy, well - drained, mesic Typic Hapludalf (Kalamazoo loam series) soils formed from the glacial till and outwash (Syswerda and Robertson, 2014) . The MCSE, established in 1989, consists of several experimental treatments of annual and perennial cropping systems. To meet the objectives of this study, only conventional and no - till treatments were considered of a corn ( Zea mays ) , soybean ( Glycine max ) , and winte r wheat ( Triticum aestivum ) annual rotation. Both treatments have been under rainfed management. Further, each of these experimental treatments consisted of six replicants (blocks) in a randomized complete block design , and each block has a dimension of 87 × 105 m . In the conventional treatment, c rops were planted following the primary tillage using moldboard plough until 1998 and there after using chisel plough . Primary tillage was followed by soil finishing each year . Disking was practiced as s econdary til lage before planting a wheat crop in the rotation while i nter - row cultivation was performed for corn and soybean. Nitrogen fertilizer 112 was applied as per the soil - test recommendations for each crop. Appropriate herbicides were broadcasted to control weeds d epending on the weed intensity . Crops were not applied with any manure or insecticides. The same management was used for the no - till treatment, except crops were planted without tillage using a no - till drill (Robertson and Hamilton, 2015) . Even though the MCSE was established in the late 1980s, an appropriate experimental design was adopted from 1993. Therefore, our study was designed for the experimental period of 1993 - 2019. The crop rotation begins with corn in 1993 and ends with wheat harvest in 2019, co vering nine complete rotations (27 years). The following data were used to parameterize the crop model developed for this experiment. The daily weather data (precipitation, maximum temperature, minimum temperature, and solar radiation) for the experimenta l period were obtained from the automated weather station located within the MCSE site. The soil analysis data of bulk density, organic carbon, total nitrogen, soil pH, extractable phosphorous, and exchangeable potassium at different depths were collected from previously published data (Crum and Collins, 1995) . Crop management data such as cultivar, planting (date of planting, planting method, planting distribution, planting density, row spacing, row direction, and planting depth), nitrogen fertilizer appli cation (date of application, type of nitrogen fertilizer, method of application, depth of application and quantity of application), tillage (date of tillage, tillage implement and tillage depth), and harvesting date were collected from the MCSE agronomic l og. The gravimetric soil moisture was measured typically in biweekly intervals at a depth of 0 - 25 cm from each replicate of the treatment during the study period. Periodically, updated soil bulk density data for the same depth (0 - 25 cm) was used to transfo rm gravimetric soil moisture into volumetric soil moisture. The detailed procedure for sampling gravimetric soil moisture and the conversion into volumetric soil moisture can be 113 found in Eeswaran et al. ( 2021) . Crop yields were measured at harvest using co mbine harvesters for the entire block. The seed yield was calculated based on the standard seed moisture level of 15.5% for corn and 1 2.5 % for wheat and soybean . Figure 6. 2 . Location of the experimental site and the Kalamazoo River W atershed in Michigan, USA. 6. 2. 3 Description of the Study Watershed The study was conducted in the Kalamazoo River watershed, which is in the Southwest part of Michigan, USA (Figure 6. 2). The wat ershed drains an extent of 5,232 km 2 from the 114 counties of Allegan, Barry, Calhoun, Eaton, Hillsdale, Jackson, Kalamazoo, Kent, Ottawa, and Van Buren into Lake Michigan near the towns of Saugatuck and Douglas (KRWC, 2011) . The hydrogeology of this watershed is determined by thick glacial deposits of sand and gravel that contribute to permeable soils and stable groundwater inflows (Wesley, 2005) . Generally, there is a high degree of connection between surface and groundwater in the basin (Grannemann et al., 2 008) . Soil groups which make up the watershed are 40% of sandy loam, 30% of loamy sand, 25% of clay loam, and 5% of organic soils (Wesley, 2005). The watershed has a gentle to moderate slope, and the drainage class is moderate to well - drained (Schaetzl et al., 2009) . The Kalamazoo River Watershed is historically well known for its richness in biodiversity, ecosystem services, and recreational opportunities as it consists of several lakes, headwater streams, wetlands, and flood plains that are heavily contr ibuted by its groundwater system (Alexander et al., 2014; KRWC, 2011) . A stable baseflow to streams and other habitats is essential to attenuate temperature extremes and to sustain aquatic life (KRWC, 2011) . In contrast, growing pressures from development, urbanization, and agricultural operations have significantly altered the hydrology and water quality within the watershed (Wesley, 2005) . Moreover, groundwater is extracted for industries, public water supply, domestic wells, irrigation, livestock, mining , and other commercial purposes; thus, groundwater withdrawal in this watershed is rated highest in the State of Michigan (Wesley, 2005) . The high groundwater withdrawal within the Kalamazoo River Watershed warranted its use for this study. Agriculture is the primary land use within the watershed (47%) followed by forest cover and successional vegetation (30%), lakes, wetlands, and flood plains (15%), and urban areas (8%) (Figure 6. 2; KRWC, 2011) . Row crops such as corn, soybean, and wheat dominate agricul tural lands while pasture, alfalfa, fruit crops, and livestock are also produced in the region. 115 The climate varies across the watershed depending on location, distance from Lake Michigan (lake effect), the formation of air masses, and atmospheric disturban ces. The mean annual - 865 mm, of which about half is snowfall (Wesley, 2005) . Watershed scale crop modeling was performed for the period 1993 - 2019 and the following da ta were collected for this task. The daily weather data (precipitation, maximum temperature, and minimum temperature) for the study period were obtained from eight meteorological stations in the Kalamazoo River Watershed (Figure 6. Centers for Environmental Information. To fill in the missing weather data, t he Soil Water Assessment Tool (SWAT) weather generator, i.e., WXGEN, was used (Sharpley and Williams, 1990) . The soil data for the watershed were downloaded from a global soil profile database for crop modeling applications available at Harvard Dataverse (Han et al., 2015) . This soil data is available as compatible to the DSSAT crop model (.SOL format) at 10 km resolution and recommended for large scale crop modeling (Han et al., 2019) . A total of 85 grids were found in the Kalamazoo River Watershed. The land use data were collected from National Land Cover Database (NLCD) 2013 (Homer et al., 2020) and the agric ultural land use (legend 82: cultivated crops) in the watershed was extracted using ArcGIS 10.6 (Esri, Redlands, California, USA). Finally, the soil grids were assigned to respective weather stations using geoprocessing tools ( Thiessen method ) in ArcGIS (T hiessen, 1911) . Therefore, a total of 85 modeling domains were used for crop modeling in the watershed. 6.2. 4 Crop Modeling Crop modeling for conventional and no - till treatments of the long - term experiment was performed in DSSAT. DSSAT is one of the most highly cited crop modeling platforms in global 116 agricultural research and currently consists of process - based simulation models for more than 42 crops (Hoogenboom et al., 2019; Jones et al., 2003, 2017) . DSSAT has been successfully imple mented in the evaluation of interactions among genetics, environment, and management at scales ranging from field to landscape (Adnan et al., 2019; Eitzinger et al., 2017) . This includes the assessment of genetic improvement (Boote et al., 1996) , evaluatio n of the impacts of climate change (Fodor et al., 2017; Rosenzweig et al., 2014) , optimization of management practices such as tillage, water, and nutrients (Iocola et al., 2017; Kropp et al., 2019; Liu et al., 2013; Malik and Dechmi, 2019; Roy et al., 201 9) , and yield gap analysis (Teixeira et al., 2019) . Moreover, DSSAT was applied for yield forecasting, precision farming, decision support, and policy analysis in agriculture (Boote et al., 1996; Shelia et al., 2015; Thorp et al., 2008) . Crop modeling can also offer valuable opportunities to evaluate resilience against climate extremes when integrated with long - term research experiments (Rötter et al., 2018) . In this study, the SEQUENCE modeling procedure (Bowen et al., 1998; Liu et al., 2013; Salmerón et al., 2014) in DSSAT - CSM was used to simulate the corn - soybean - winter wheat rotation for the conventional and the no - till treatments. The DSSAT version 4.7.5 (Hoogenboom et al., 2019) was used to simulate corn, soybean, and winter wheat by apply ing crop models of CERES - maize, CROPGRO - soybean, and CERES - wheat for the respective crops (Jones et al., 2003) . Weatherman application within the DSSAT (Pickering et al., 1994) was used to create DSSAT format (.WTH) weather files for the experimental perio d (1993 - 2019) using collected daily precipitation, maximum temperature, minimum temperature, and solar radiation from the MCSE site. The soil information (Kalamazoo Loam soil - MSKB 890006) was obtained from the DSSAT soil database and the Web Soil Survey (N RCS, 2020) , and the relevant model parameters, such as the saturated hydraulic conductivity (SSKS), were updated accordingly. The 117 soil analysis data collected from Crum and Collins, ( 1995) were used as the initial soil analysis values. The best cultivar o ptions suggested by Grace and Robertson for MCSE at KBS were available in DSSAT sequence models ( MSKB8902.SQX ) and were used to initialize the simulation (Hoogenboom et al., 2019) . Accordingly, four crop cultivars (two corn cultivars and one cultivar each for soybean and winter wheat) were used for crop modeling. The identification codes of the corn cultivars used are IB0090 and IB0093, both belong to the ecotype IB0001. The identification code of the soybean cultivar is 990002 (ecotype: SB0201 ) while the i dentification code for the wheat cultivar is IB 0488 (ecotype: USWH01 ). Planting information, nitrogen fertilizer applications, and harvesting information were incorporated for both treatments. The period between crops in the rotation was modeled as fallows . Irrigation information was not required as both treatments were managed as rainfed. Tillage operation for the conventional treatment was set as moldboard plow until 1998 and then as chisel plo w. For the no - till treatment, tillage was set to be a no - till drill. Treatments were appropriately assigned in separate files (.SQX), and simulation was initiated using the following methods: The Priestly - Taylor/Ritchie method was used to estimate evapotranspiration (Priestley and Taylor, 1972) , Suleiman - Ritchie meth od (Suleiman and Ritchie, 2003) was used to estimate soil evaporation, infiltration rate was estimated using the Soil Conservation Service method (SCS, 1985) , Century method (Parton, 1996) was used to simulate soil organic matter, and soil layer distributi on was set to the modified soil profile. The soil water balance was simulated in DSSAT as a function of daily precipitation, irrigation (if any), transpiration, soil evaporation, runoff, and drainage on a daily time step (Ritchie, 1998) . 118 D aily volumetric soil moisture was simulated for the depths of 0 - 5 cm, 5 - 15 cm, 15 - 22 cm, and 22 - 31 cm using the DSSAT model. Then, weighted average soil moistures were calculated for the comparison with the observed soil moisture at 0 - 25 cm depth. The root growth factor (SRGF), lower limit/wilting point (SLLL), drained upper limit/field capacity (SDUL) were manually adjusted to match the simulated and observed soil moisture to calibrate the DSSAT soil water balance module (Calmon et al., 1999; Fang et al., 2008) . The fina l soil properties generated from soil data calibration is presented in Table S 6. 1. Performance of the soil moisture and yield calibration was evaluated using coefficient of determination ( R 2 ) (Equation 6. 1), normalized root mean square error ( NRMSE ) (Equat ion 6. 2), and index of agreement ( d ) (Equation 6. 3). NRMSE and d are commonly used to statistically evaluate the goodness of fit between observed and simulated soil moisture and yield (Araya et al., 2017; Dokoohaki et al., 2016; Liu et al., 2013; Yang et a l., 2014) . The model performance according to NRMSE goodness of fit can be classified as 0 - 15% (good), 15 - 30% (moderate), and >30% (poor). Goodness of fit based on d (Willmott, 1982) can be categorized as <0.7 (poor), 0.7 - 0.8 (moderate), 0.8 - 0.9 (good), an d 0.9 - 1.0 (excellent) as proposed by Liu et al. ( 2013) . ( 6. 1) ( 6. 2) ( 6. 3) where, is the simulated i th value, is the observed i th value, is the mean of the simulated values, is the mean of the observed values, and n is the number of values. 119 6.2.5 Groundwater Modeling Groundwater flow in the shallow unconsolidated glacial deposits was modeled using Interactive Groundwater (IGW) , a groundwater modeling software introduced by Li and Liu ( 2006) , which uses the finite difference approximation of the governing partial differential equation (Equation 6. 4) to solve confined and un confined flow conditions: ( 6. 4 ) where , S s is the specific storage coefficient, h is the hydraulic head [L], t is time [T], K is the saturated hydraulic conductivity, is the mathematical gradient operator, q is the net source (positive) or sink (negative) flux term, including recharge, and surface seepage [LT 1 ] . IGW is periodicall y updated (see, e.g., Liao et al., 2015a, 2015b, 2020) ; for this study, the IGW model was developed, calibrated, and visualized using the new web - based version of IGW called MAGNET Mul ti - scale Adaptive Global Network 4 Water , accessible on the magnet4w ater website: https://www.magnet4water.com/magnet. The IGW modeling software is live linked to a database comprising terabytes of raw and derived data useful for the groundwater modeling. A high - resolution (10 m) digital elevation model ( DEM ) (NED USGS 2006) was used to map topographic variations (i.e., the aquifer top) and to simulat e groundwater - surface seeps in the watershed (see more below). The bottom boundary is represented by a spatially variable surface based on the top of the bedrock undern eath the unconsolidated sediments . For this project, the bedrock wa s assumed to be impermeable, however it is known that the lower portion of th is watershed has a bedrock that is very transmissive . The reason for assuming an impermeable bedrock is to bette r compare the impacts of different scenarios of agricultural management on restoring the unconfined aquifer, which is impossible to quantify if the bedrock is transmissive. The bedrock top elevation raster 120 (500 m resolution) was interpolated from borehole records found in the statewide water well database called Wellogic (MDEQ, 2020) . Hydraulic conductivity (K) of the aquifer was represented by a spatially - var ia ble, two - dimensional (2D) raster of horizontal hydraulic conductivity . This was generated by inte rpolating estimated K values from records in the Wellogic database, public water supply , and U.S. Geological Society aquifer - tests, and aquifer properties reported in the literature (State of Michigan, 2006) . Given that the horizontal extent of the model w as much larger than the vertical extent, it was hypothesized that flow was predominantly two - dimensional (2D) and that a 2D model could capture the dominant flow processes. The model extent was divided into 418 cells in the x - (west - east) direction and 258 cells in the y - (north - south) direction. The model was executed for the time period between 1993 - 2019 using a one - year time step. The initial condition was generated by running the model in steady - state mode to represent long - term mean conditions , since no data was available to prescribe the initial head distribution. Annual recharge distributions from the calibrated DSSAT SEQUENCE model were included in the source/sink term at each time step . In non - cropland areas, the long - term mean recharge applied in the steady - state model was used. Natural, long - term mean recharge to the aquifer was input to the steady - state model and was created following empirical methods presented by Holtschlag ( 1997) i nvolving observed streamflow hydrographs and information related to land use, soil conditions, and watershed characteristics (State of Michigan, 2006) . - state model and the subsequent transient model, groundwater dis charge into lakes, streams, and wetlands/springs - the major control of the long - term prevailing groundwater flow patterns wa s captured through the critical use of high - resolution Digital Elevation Models (DEMs). Specifically, the entire land surface, mo deled using 121 the 10 m DEM from NED USGS (2006), was treated as a one - way head - dependent boundary condition (seepage) . This allow ed groundwater to discharge to the surface where the groundwater level intercept ed the land surface. The flux per unit area leavi ng the aquifer wa s the product of the leakance (hydraulic conductivity per unit thickness) of the land surface with the difference between the land surface elevation and the head in the aquifer. Leakance is a calibration parameter that is manually calibrat ed . For example , if the leakance was too low the flooded area w ould be too large and vice versa (note: a final calibrated value of 1 day - 1 was used for transient simulation). Surface seepage maps at different time - steps were compared to the surface water f eatures obtained from NHD USGS (2010) to ensure that this approach effectively capture d the spatial patterns of groundwater discharge to the surface water bodies. Groundwater pumping was not represented in the initial condition model nor the transient simu - of both steady state and transient models . In short, recharge in the watershed wa s balanced by surface seepage to surface water bodies in the simulations presented here. T he initial head distribution was obtained from the steady - state solution, and the simulation was advanced in time by solving Eq. (4) with annual time - steps. Annual recharge distributions from the calibrated DSSAT SEQUENCE model for the conventional and no - till treatments were included in the source/sink term at each time step in sperate runs . In non - cropland areas, the long - term mean recharge applied in the steady - state model was used (State of Michigan, 2006) . All other aquifer properties / attributes from the steady - state model were applied during the transient simulation. In addition, a specific yield of 0.1 was assigned based on the aforementioned distribution of soil types in the watershed (detailed specific yield data was n ot available). 122 6.2. 6 Simulation of Crop Yields and Recharge at a Watershed Scale Calibrated DSSAT SEQUENCE model for the conventional and no - till treatments were used to simulate crop yields and agricultural recharge for the period of 1993 - 2019. The water shed was clustered according to climate and soil types. It was assumed that the corn - soybean - wheat rotation was planted on all agricultural land within the watershed. The crop model was later run for each unique set of climate and soil type under the conve ntional and no - till treatment scenarios. 6.2. 7 Assessment of Resilience as Affected by the Adoption of No - Till Agriculture A rising groundwater table from increased recharge is beneficial since many natural habitats, such as wetlands, depend on year - roun d groundwater availability (McLaughlin and Cohen, 2013; Sampath et al., 2015, 2016) . In addition, increases in soil moisture within the root zone can improve the resilience of rainfed agricultural productions (Eeswaran et al., 2021). Yield, net return, and soil moisture metrics, namely mean relative difference (MRD) and soil water deficit index (SWDI), were used as metrics of resilience. MRD and SWDI were shown to be suitable metrics to evaluate resilience in agricultural systems (Eeswaran et al., 2021) . MR D was presented by (Vachaud et al., 1985) to evaluate the temporal stability of spatially distributed soil moisture measurements. Additionally, treatment with a higher MRD was considered resilient to climate extremes , such as droughts (Eeswaran et al., 202 1) . The MRD during a particular growing season wa s computed as follows : ( 6. 5) ( 6. 6) where, is the simulated daily volumetric soil moisture for i th treatment on j th day. This soil moisture was derived as a weighted avera ge for 0 - 25 cm depth from the simulation 123 outputs. The number of treatments denoted by n . is the average volumetric soil moisture of all treatments and N is the total number of days in the growing season. In this study, the growing season was considered to start on April 1 st and end October 31 st , since it covered the critical stages of each crop and the MRD values were calculated in percentages. Probabilit y analysis (Alizadeh, 2013) was conducted for the annual MRD values, and probability curves were compared between treatments to assess the resilience as affected by the adoption of the no - till treatment. SWDI is an agricultural drought index proposed by Ma rtínez - Fernández et al. (2015) and can be implemented to assess droughts when continuous soil moisture data is available. The SWDI is calculated using the following formula : ( 6. 7) where, is the simulated daily volumetric soil moisture during the growing season as above. is the field capacity/drain upper limit, and is the wilting point/lower limit. and val ues were obtained from each selected soil file (Han et al., 2015) as weighted averages for the 0 - 25 cm soil depth. A particular soil will have excess water when SWDI is positive, soil will be at the field capacity when SWDI equals zero, and be in a drought phase when SWDI is negative. Moreover, drought severity categories can be classified based on SWDI - - 2 > SWDI > - 5, as - 5 > SWDI > - - 10 > SWDI (M artínez - Fernández et al., 2015). Calculated SWDI for the entire growing season (April - October) for each year during the study period (1993 - 2019) was used to calculate the median, mean, maximum, and minimum across all soils, and these values were later arra nged in descending order to perform probability analysis 124 for each treatment (Alizadeh, 2013). Probability curves were compared between treatments to assess the resilience of the no - till agriculture to drought. The net return was estimated through cost - bene fit analysis using the annual crop yields and the price received for crops in November 2018 in Michigan (USDA, 2019) . In 2018, the price of corn, soybean, and winter wheat was 131.50, 307.50, and 180.76 US dollars per ton, respectively. The cost was calcul ated using the variable cost involved in all agricultural operations for both treatments during the year 2018 in the long - term research experiment. This cost was estimated based on a detailed 2018 enterprise budget from Clemson University Cooperative Exten sion for the respective crops (Clemson, 2020) . The pricing of cost and benefit components were considered as static over the years of simulation and the fixed cost was excluded due to lack of information for reliable estimates. The yield, net return, and annual recharge were statistically analyzed in a mixed model (Equation 6. 8) to evaluate the significance of fixed and random effects on these response variables for each evaluated crop ( i.e., corn, soybean and wheat ) . ( 6. 8 ) where, is the response (grain yield / net return /annual recharge ) simulated for the i th treatment, within j th soil type on the k th cropping year; µ is the intercept ; is the fixed effect of the cropping year k ; is the fixed effect of the treatment i; represents the random effects of the jth soil type, with denotes the fixed interaction between the ith treatment and kth cropping year; is the random effect of the interaction between jth soil type and kth cropping year, with ; is the random effect of the interaction betw een the ith treatment and jth soil type, ; and is the error associated with each observation, with . To ensure the normality of the residuals and the homogeneity of 125 variances, the grain yield and annual recharge dat a were log - transformed . Transformations were not needed for net return. There were varying extents of acreage of agricultural land use for each soil in the watershed. Hence, the area of each soil was used as a weighting factor in the model. The comparison between the means w as performed using the Tukey - Kramer t est , assuming (Herberich et al., 2010) . All analyses were performed using the GLIMMIX procedure (Milliken and Johnson, 2009) in the SAS software version 9.4 (SAS Institute Inc. Cary, North Ca rolina, USA). 6.3 Results and Discussion 6.3.1 Calibration of the Crop Model The sequential DSSAT crop model was calibrated and validated for yield and soil moisture during the period of 1993 - 2019, which included nine complete rotations of corn - soybean - wheat crops. The performance of the model to simulate crop yields under both tre atments was measured by the goodness of fit indicators shown in Table 6. 1. According to the R 2 and d - index , the model performance was considered excellent, whereas the NRMSE indicated moderate performance (Liu et al., 2013; Willmott, 1982) . However, relati vely large NRMSE values are expected when modeling long - term crop performance for multiple growing seasons as a result of interannual variations. It is also important to note that the performance of the no - till model was slightly better than the convention al model. A similar performance was observed for the simulation of soil moisture. However, performance indicators show that the crop model was reasonably calibrated for the corn - soybean - wheat rotation (Table 6. 1). Table 6. 1 . The goodness of fit parameters of the calibrated crop model to simulate yield and soil moisture under the conventional and no - till treatments. Treatment Crop yield Soil moisture R 2 NRMSE (%) d - index R 2 NRMSE (%) d - index Conventional 0.73 27.6 0.92 0.74 29.0 0.8 No - till 0.75 26.6 0.93 0.74 19.3 0.9 126 6.3.2 Calibration of the Groundwater Model The steady - state simulation results are shown in Figure S 6. 1. The comparison between the simulated results (heads) of the steady - state model and Static Water Level (SWL) measurements from water well records i n the Wellogic database can be seen in Figure 6. 3 . SWL observations from 23,757 glacial wells were used to calibrate the model. The solid 45 - degree line between s imulated and actual observations while the dashed lines represent confidence intervals of one standard deviation (Figure 6. 3) . Calibration results show that the model performance was good , as indicated by a strong Nash - Sutcliffe model efficiency coefficien t (NSE) of 0.90. Even though there wa s slightly large spread of the data points , all data wa s centered around the line of perfect agreement. Th e center - focused distribution demonstrates that the model was able to capture the dominant spatial structure of the groundwater system (i.e., the distribution of groundwater recharge and discharge areas) . Slightly large spread in the data , as indicated by the root - mean - square error of 7.91 m, primarily reflects the significant noise embedded in the SWL observations (Curtis et al., 2018) . 127 Figure 6. 3 . C omparison between simulated ground water heads and observed groundwater heads . The solid red line in the calibration indicates a 1:1 perfect agreement. The dashed lines represent a confidence interval of one standard deviation. 6.3.3 Resilience as Affected by the Adoption of No - Till Agriculture In this study, we quantif ied resilience in terms of recharge, groundwater ta ble, soil moisture metrics, crop yield, and net return f or both the conventional and the no - till treatments. Treatments with higher recharge, groundwater table, soil moisture retention, ability to mitigate drought, larger crop yields , and higher net revenues were considered as resilient over the long - term (1993 - 2019) evaluation. 6.3.3.1 Recharge and Groundwater Table as Affected by the Adoption of No - Till Agriculture The statistical analysis for the annual recharge showed that the effects of trea tment, year, and the interaction between treatment and year were strongly significant (see the supplementary 128 material Table S 6. 2). The means of the annual recharge across different soils and years from each crop can be seen in Figure 6. 4. Results showed th at the no - till treatment significantly increased the annual recharge from all crops in comparison to the conventional treatment. The annual recharge from the no - till treatment for corn, soybean, and wheat were 12.4%, 6.2%, and 13.2% greater than the annual recharge from the conventional treatment, respectively. The soybean had the highest recharge followed by wheat and corn. Because the interaction effect between treatment and year was also significant for the annual recharge (Table S 6. 2), the comparisons b etween treatments for each crop during the period of study is presented in Figure S 6. 2. In most years, the no - till treatment had significantly higher recharge than the conventional treatment. The changes in recharge across the years can be attributed to th e changes in precipitation and crop growth, which affect other water balance components (Figure S 6. 2). Figure 6. 4 . The mean annual recharge from different crops under two treatments in the Kalamazoo River watershed. ** indicates strongly signif icant means at p<0.0001 . 129 The results from the transient simulation for the conventional and no - till treatments are presented in Figures 6. 5 and 6. 6. Figure 6. 5 shows the 2019 head distribution under the conventional treatment, and the location of the six monitoring wells where transient head results were reported. It is important to n ote that the changes in the water table at the watershed scale over time were difficult to distinguish, therefore no comparison of plan - view model results under each a gricultural scenario was presented . Therefore, temporal changes of groundwater levels were presented at each monitoring wells (Figure 6. 6). The time - series comparisons show that the no - till treatment result ed in higher water tables compared to the conventi onal treatment. The differences were typically small, with about 0.3 - 0.5 m at Monitoring Well 1 , 0.1 - 0.3 m at Monitoring Well 4 , and 0.1 m or less at the other locations. However, even a relatively small improvement in the groundwater table can have benefi cial effect s on streams and aquatic ecosystems in the Kalamazoo River Watershed, due to the large contribution of groundwater to streamflow in this region (Cooper and Merritt, 2012; Sampath et al., 2016) . 130 Figure 6. 5 . Monitoring w ell (MW) l ocations superi mposed over the 2019 head distribution under conventional treatment . 131 Figure 6. 6 . Simulated water table heads under the conventional and no - till treatments for the six monitoring wells . 132 As observed in our study, higher recharge in the no - till treatment simultaneously increased the groundwater table; however, the magnitude of change was dependent upon the characteristics of the underlying groundwater system (Figure 6. 6). The higher recharg e observed under the no - till treatment in this study may have been caused by the greater infiltration of rainwater (Nunes et al., 2018) . According to Kravchenko et al. ( 2011) , the no - till system establishes large pores associated with the undisturbed root channels created by the previous crops. The macropores in a no - till system may contribute to greater infiltration and thus recharge. In agreement with the findings reported here, Syswerda and Robertson ( 2014) also found higher downward drainage under the n o - till treatment compared to the conventional treatment. In many regions of the world, groundwater is being tapped at rates greater than the local recharge, leading to the depletion of aquifers (Dalin et al., 2017; Reitz et al., 2017) . Furthermore, increa sing climate variability has already posed additional challenges to water resources and accelerated stresses to the water - energy - food nexus (Smidt et al., 2016) . Therefore, an improved recharge and water table under the no - till practice can increase the re silience of the food systems, while also supporting the sustainability of groundwater - dependent ecosystems. 6.3.3.2 Soil Moisture Metrics as Affected by the Adoption of No - Till Agriculture The probability distribution of the of mean, maximum, and minimum o f MRD for both treatments across 85 soils over the period of the study is presented in Figure 6. 7. MRD measures soil moisture deviations from the average soil moisture of agricultural treatments, and a positive MRD signifies a wetter treatment while a nega tive MRD signified a drier treatment (Eeswaran et al., 2021) . The mean of the MRD clearly shows that the conventional treatment mostly (>93% probability) generated a negative MRD while the no - till treatment generate d a positive MRD. Therefore, the no - till treatment consistently retained higher soil moisture than the conventional 133 treatment. Based on the maximum line for the conventional treatment (Figure 6. 7a), it also had a small probability (<14%) to be wetter than the no - till treatment. Similarly, the minimum line of the no - till treatment (Figure 6. 7b) shows that it also had the chance to be drier than the conventional treatment by the same magnitude of probability as above. The probability distribution of SWDI across all soils over the study period is shown in Figure 6. 8. As shown in Figure 6. 8, the probability of having different drought severity levels can be analyzed based on respective SWDI values (Martínez - Fernández et al., 2015) . Based on the mean SWDI, the no - ti ll treatment had a 43% probability of having no drought events, which was substantially higher than the conventional treatment (38%). Moreover, the no - till treatment had a lower probability of having mild, moderate, severe, and extreme droughts in comparis on to the conventional treatment. According to the maximum SWDI, the no - till treatment had 78% probability to have drought free days while the probability for the conventional treatment was 75%. The minimum SWDI also showed that the no - till treatment (13%) had higher drought free days than the conventional treatment (10%). Thus, the no - till treatment was superior in mitigating drought compared to the conventional treatment in this watershed. 134 Figure 6. 7 . The probability distribution for the mean, maximu m, and minimum of MRD across different soils in the Kalamazoo River watershed for the period of 1993 - 2019 as affected by the conventional (a) and the no - till (b) treatments. Note: Red dashed line at zero MRD indicates the demarcation, where the positive MR D values signify wetter treatment while the negative MRD values signify drier treatment. Figure 6. 8 . The probability distribution for the mean, maximum, and minimum of SWDI across different soils in the Kalamazoo River watershed for the period of 1993 - 2 019 as affected by the conventional (a) and the no - till (b) treatments. Note: Red dashed lines are to demarcate different drought severity levels . Consistently higher soil moisture retention by the no - till treatment was due to the beneficial improvement of soil physical properties, such as water holding capacity (Moebius - Clune et al., 2008) . Furthermore, the no - till treatment has been found to increase rainwater 135 infiltration, decrease runoff, and to reduce soil evaporation thereby increasing the proportio n of available water in the root zone (Lal et al., 2012; Lampurlanés et al., 2016; Verhulst et al., 2011) . The ability of the no - till treatment to store more soil moisture could help to mitigate the impacts of droughts on the crops, as evident in this study. This agreed with the findings of Thierfelder and Wall ( 2010) where the no - till system performed better for soil water dynamics in a drought - prone region of Africa. Based on the above findings, the no - till treatment was more resilient than the conventional treatment and adaptation of the no - till management in the Kalamazoo River Watershed would enhance its resilience to extreme drought events, which are detrimental to rainfed systems. 6.3.3.3 Crop Yield and Net Return as Affected by the Adoption of the No - Till Agriculture The probabilities for the statistical significance of the effects evaluated for crop yield and net return is presented in Table S 6. 2. To perform this statistical analysis, the extent of each soil in the agricultural land use was used as a weighting factor, since it is critical to consider production area when comparing management effe cts at larger scales (Leng et al., 2019) . As a result, we evaluated the effects of treatments in the watershed over the entire study period with high confidence. The statistical analysis showed that the effect of treatments was strongly significant on the yield of corn and soybean, but not in wheat. Nonetheless, treatment effect was strongly significant for the net return from all crops. Furthermore, the effect of year and interaction between the treatment and year were significant for both yield and net re turn of all crops (Table S 6. 2). The means of crop yield and net return as separated by treatments across the years and soils are presented in Table 6. 2. Accordingly, the yield increased under the no - till treatment by 1.23%, 0.61%, and 0.24% for corn, soyb ean, and wheat, respectively. Deines et al. ( 2019) 136 reported a 3.3% and 0.74% yield improvement, respectively, for corn and soybean as a result of conservation tillage adoption in the US corn belt region. However, it is important to note that conservation t illage is a mixture of different intensities of reduced tillage and not necessarily entirely no - tillage. In this study, the net return was 20%, 23.4%, and 48.3% higher under the no - till treatment for corn, soybean, and wheat, respectively (Table 6. 2). The higher margin of net revenues for the no - till treatment was mainly because of its lower production costs compared to the conventional treatment. The no - till treatment was cheaper due to absence of tillage operations, even though the herbicide application r ates were higher than the conventional tillage. The costs to produce corn, soybean, and wheat conventionally were 918.84, 705.03, 586.56 USD/ha, respectively. On the other hand, no - till treatment costs were 867.36, 632.12, and 508.58 USD/ha, for corn, soyb ean, and wheat productions, respectively. As the interaction effects between treatment and year were significant for both yield and net return in all the crops, the strength of significance may vary across different years. This differential performance, as affected by treatment and years, is shown in Figure S 6. 3 (yield) and Figure S 6. 4 (net return). In summary, the no - till outperformed the conventional treatment in the majority of the years. Table 6. 2 . The mean yield and net return for different crops un der two treatments in the Kalamazoo River watershed*. Treatment Corn Soybean Wheat Yield (Mg/ha) Net return (USD/ha) Yield (Mg/ha) Net return (USD/ha) Yield (Mg/ha) Net return (USD/ha) Conventional 8.91 b 315.31 b 3.27 b 345.87 b 4.09 a 165.77 b No - till 9.02 a 378.47 a 3.29 a 426.78 a 4.10 a 245.88 a *Means with the same letter in each column are not significantly different at p<0.05. The no - till treatment increased crop yields in most studies around the world (Corbeels et al., 2014; Pittelkow et al., 2015; Rusinamhodzi et al., 2011) . However, some studies have found no significant effects on yield under the no - till systems (e.g., Daigh et al., 2018) , while a few 137 other studies reported reductions in crop yield (e.g., Powlson et al., 2014) . In contrast, to see the consistently outperforming trends under the adoption of the no - till agriculture the evaluation must be longer than a decade (C usser et al., 2020) . This study was built on this need and successfully captured the long - term impacts of the no - till treatment. The results showed that the adoption of the no - till treatment could significantly improve the resilience of agricultural system s by increasing crop yields and net return. The increment in crop yields under the no - till management can be attributed to the enhancement of soil physical, chemical, and biological properties (Nunes et al., 2018) . 6.4 Conclusions In this long - term study, we found that the adoption of no - till treatment for a corn - soybean - wheat rotation has potential to increase the resilience in the Kalamazoo River Watershed. This improvement of resilience was demonstrated using the following metrics: recharge, water table, soil moisture, drought vulnerability, yield, and net return. The no - till treatment had significantly higher annual recharge, for corn, soybean, and wheat which were 12.4%, 6.2%, and 13.2% greater than the annual recharge from the conventional treatment, r espectively. The highest recharge was observed for soybean followed by wheat then corn. The rise in the water table resulting from the adoption of the no - till treatment in the watershed ranged between 0.1 - 0.5 m, which could substantially contribute to repl enishing the aquifers and groundwater - dependent ecosystems. MRD of soil moisture clearly showed that the no - till treatment consistently maintained higher soil moisture compared to the conventional treatment, thus remained as a relatively wetter treatment. Therefore, the no - till treatment had a higher resilience against drought compared to the conventional treatment as quantified by the drought index (SWDI). Yields and net returns were also improved under the no - till treatment for all crops 138 in the rotation. When averaged across the years and soils, the no - till treatment produced 1.23%, 0.61%, and 0.24% higher grain yields for corn, soybean, and wheat, respectively. Moreover, the no - till generated 20.0%, 23.4%, and 48.3% higher net returns for corn, soybean, a nd wheat, respectively. There were two major assumptions in this study. First, all agricultural land use in the Kalamazoo River Watershed was assumed to be planted with a corn - soybean - wheat rotation. However, farmers plant a variety of crops throughout th e watershed; therefore, the findings are mostly applicable to the row crop rotations in this region. Secondly, we assumed that the deep percolation simulated by the crop model instantly reached the water table. This assumption is only valid in regions wher e there is a greater connection between the surface and groundwater, similar to our study area. To expand our approach to different landscapes with varying climate, soil, groundwater, and cropping systems, we recommend modifying both the crop and groundwat er modeling procedures adhering to site - specific parameters and requirements. 6.5 Acknowledgment This research project was financially supported by the Dissertation Completion Fellowship from the College of Agricultural and Natural Resources at Michigan S tate University. The KBS Long - Term Ecological Research (KBS - LTER) experiment was funded by the National Science Foundation Long - term Ecological Research Program (DEB 1832042) at the Kellogg Biological Station and by Michigan State University AgBioResearch. 139 7 . CONCLUSIONS In this dissertation, several innovative approaches were introduced within three studies to quantify the resilience of rainfed agricultural systems at both the field and watershed scales. The robustness of these approaches was tested using data from a long - term cropping system experiment at KBS and then scaled up to the Kalamazoo River Watershed located in the Southwestern portion of Michigan, USA. In the first study, the applicability of soil moisture metrics to gauge resiliency of differently managed rai nfed agricultural treatments were evaluated at the field scale. In addition, the robustness of soil moisture metrics was examined over a long - term period (1993 - 2018) by monitoring the impacts of climate variabilities on crop growth and yield for a corn - soy bean - wheat rotation. The major takeaways from this study are as follows: Soil moisture - based metrics can be used to measure the resilience in rainfed agricultural systems. The no - till treatment significantly improved the resiliency of a corn - soybean - wheat rotation than the conventional treatment in terms of higher soil moisture retention, higher effectiveness for drought mitigation, greater crop yields, and increased stability of yields. The organic treatment had substantially higher resiliency in terms of gr e ater soil moisture retention and drought mitigation than the conventional treatment ; however, nitrogen limitation significantly reduced yields, especially for cereal crops (i.e., corn and wheat) . The reduced input treatment was the least resilient as it had limited capacity to recover from the impacts of climate extreme s (i.e., drought). 140 In the second study, climate resilience of alternative treatments for a long - term (1993 - 2018) corn - soybean - wheat rotation was evaluated in terms of profitability and far m risks for adaptation using enterprise budgeting and risk analysis. Means and volatility of expected net revenues and risk preferences were used as evaluation metrics. The key take - home messages from this study are as follows: The organic treatment was ex pected to exceed the conventional treatment's net revenues and stability making it more resilient. The no - till treatment was superior to the conventional and the reduced input treatments in expected annual net revenues with lower risk under climate extr emes. Due to higher resilience and lower risk levels, the organic and no - till treatments were suitable to cater to a large group of farmers with different risk preferences for adaptation . The conventional and reduced input treatments were vulnerable to climate extremes and should not be promoted in areas with a high - level of climate variabilities. In the third study, the overall resilience of a conservation agricultural practice (i.e., no - till) was compared to the conventional tillage for the corn - soybean - wheat rotation at the watershed scale using an integrated framework of crop and groundwater models. Recharge, groundwater head , soil moisture, yield , and net return were used as resilience metrics. The key conclusions of this study are as follows: 141 The adoption of the no - till treatment increased the annual recharge and groundwater heads in the watershed over the long - term study (1993 - 2019), and the highest recharge was observed for soybean followed by wheat and corn. The rise in the groundwa ter table under the no - till treatment ranged from 0.1 m to 0.5 m, depending on the underlying groundwater system in the watershed , and ha d the potential to beneficially impact the groundwater - dependent ecosystems. Shifting from the conventional tillage to t he no - till treatment significantly increase d the yield of corn and soybean and the expected net returns from all crops. T he no - till treatment consistently retained higher soil moisture than the conventional treatment and thereby helped to mitigate the imp acts of droughts. The overall resilience of the corn - soybean - wheat rotation was substantially improved by the adoption of the no - till treatment. 142 8 . RECOMMENDATIONS FOR FUTURE RESEARCH This dissertation presented innovative approaches to quantify the resilience of rainfed agricultural systems at the field and watershed scales . The possible areas to further strengthen these approaches are as follows: Soil moisture - based metrics can be fur ther improved with more frequently measured soil moisture at different depths within the root zone. This can be achieved through installing automated soil moisture sensors in the treatment plots at sufficient densities. Remote sensing - based soil moisture m easurements (e.g., Soil Moisture Active Passive - SMAP, Soil Moisture Ocean Salinity - SMOS) can be applied to develop s oil moisture - based resilience metrics if their spatial resolutions are reasonably improved at the field scale. The approaches presented here can also be extended to the irrigated cropping systems if the frequency and amount of irrigation are periodically recorded. The current version of the DSSAT model does not allow intercropping. Therefore, developing the intercropping module for DSSAT can assist in the evaluation of different cover crop species on the resilience of cropping systems. In addition, simulation of crop responses to nutrients can be improved by developing potassium and phosphorus (e.g., wheat) uptake models. Seamless integration of crop and groundwater models can give us better insights to study the impacts of different agricultural practices on recharge and groundwater restoration. 143 In this study the annual recharge and groundwater average head were reported to gauge the resilienc y as related to ecosystem services. However, it is recommended to report these metrics during the dry season to better represent the impacts of groundwater restoration on ecosystem services. Studying different pathways to motivate producers to adopt climate - resilient practices considering market and policy implications. 144 APPENDIX 145 APPENDIX Table S4. 1 . Details of the four annual row crop treatments of MCSE investigated in this study . Treatment Description of Management Conventional (CON) Crops were planted in corn - soybean - winter wheat rotation. Primary tillage was practiced using moldboard plow until 1998 and chisel plowing was used from 1999 onward. Disks were used for secondary tillage before wheat planting , while the field was conditioned with a soil finisher prior to soybean and maize planting. Moreover, inter - row cultivation was practiced for soybean and maize. Fertilizers were applied at the rates based on soil - test recommendations for each crop. Herbici des were broadcasted within and between rows to control weeds. There were no applications of manure/compost or insecticides. No - till (NT) Crops were planted in corn - soybean - winter wheat rotation. These crops were established under zero tillage. Fertilizers were applied at rates based on soil - test recommendations for each crop. Herbicides were broadcasted within and between rows to control weeds. There were no applications of manure/compost or insecticides. Reduced input (RI) Crops were planted in corn - soybean - winter wheat rotation. Primary tillage was practiced using moldboard plow until 1998 and chisel plowing was used from 1999 onward. Disks were used for secondary tillage before wheat planting, while the field was conditioned with a soil fin isher prior to soybean and maize planting. Moreover, inter - row cultivation was practiced for soybean and maize. Nitrogen (N) fertilizer and herbicide inputs were applied as one - third of N and herbicides applied to the conventional system (Reduced input). T hese herbicides were banded within rows. Winter cover crop was established following the corn and wheat crops within the rotation with the intention of supplementing nitrogen to the following crop. Generally, cereal rye ( Secale cereal ) was planted followi ng corn, while red clover ( Trifolium pratense ) was planted after wheat. There were no applications of manure/compost or insecticides. Organically managed (USDA certified organic) (OR) Crops were planted in corn - soybean - winter wheat rotation. Primary till age was practiced using moldboard plow until 1998 and chisel plowing was performed from 1999 onward. Disks were used for secondary tillage before wheat planting, while the field was conditioned with a soil finisher prior to soybean and maize planting. More over, inter - row cultivation was practiced for soybean and maize. Winter cover crop was established following the corn and wheat crops within the rotation with the intention of supplementing nitrogen to the following crop. Generally, cereal rye ( Secale cer eal ) was planted following corn, while red clover ( Trifolium pratense ) was planted after wheat. This certified organic treatment was not applied with any chemical fertilizers/herbicides/insecticides. 146 Figure S4. 1 . Ranked MRD of volumetric soil moisture and ITS for each treatment during the normal years. Note: Crop grown is given next to the respective year for each plot. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. 147 Figure S4. 2 . Ranked MRD of volumetric soil moisture and ITS for each treatment during the wet years. Note: Crop grown is given next to the respective year for each plot. CON: Conventional treatment; NT: No - till treatment; RI: Red uced input treatment; OR: Organic treatment. 148 Figure S4. 3 . Means of soil organic carbon content in different treatments during the period of 1989 - 2001 in this experiment. The error bars represent the standard error. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organic treatment. 149 Figure S4. 4 . Association between mean relative difference (MRD) of volumetric soil moisture and soil organic carbon content in the treatments investigated in this study. 150 Figure S4. 5 . Total biomass (a - c) and yield (d - f) of crops as affected by the interaction between treatment and climate variability. The error bars represent the standard error. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; OR: Organ ic treatment. 151 Table S4. 2 . P - values for the effects evaluated in the statistical mixed model for MRD, total biomass and yield. Crop Effects in the statistical model Probability ( p - value) MRD Total biomass Yield Corn Treatment ( trt ) <0.0001** <0.0001** <0.0001** Climate variability ( clim ) 0.9250 0.5154 0.2641 Year ( yr ) >0.05 0.0563 0.0948 Interaction between treatment and climate variability ( trt × clim ) 0.1733 0.0002** 0.0003** Interaction between treatment and year ( trt × yr ) 0.2497 0.0171* 0.0043* Soybean Treatment ( trt ) <0.0001** 0.0020* 0.0198* Climate variability ( clim ) 0.9998 0.0024* <0.0001** Year ( yr ) >0.05 0.0845 0.0867 Interaction between treatment and climate variability ( trt × clim ) 0.1442 0.0071* 0.0031* Interaction between treatment and year ( trt × yr ) 0.4661 0.6700 0.8306 Wheat Treatment ( trt ) <0.0001** <0.0001** <0.0001** Climate variability ( clim ) 0.9805 0.7511 0.2335 Year ( yr ) >0.05 0.0698 0.0729 Interaction between treatment and climate variability ( trt × clim ) 0.8305 0.3534 0.1965 Interaction between treatment and year ( trt × yr ) 0.8107 0.0700 0.0640 **Strongly significant at p<0.001; *significant at p<0.05. 152 Table S4. 3 . Coefficient of variation of total crop biomass and yield of crops under different treatments as affected by climate extremes . Crop parameter Category of climate extreme Treatment Coefficient of variation (%) Corn Soybean Wheat Total crop biomass Dry year CON 28.1 26.3 30.9 NT 19 .0* 22.6 * 21 .0* RI 25.1 24 .0 22.7 OR 31.9 28.1 35.4 Wet year CON 27.9 18.6 15.1 NT 20 .0* 11.6 * 16.6 RI 38.1 21.2 7.4 * OR 22.1 20.2 21 .0 Crop yield Dry year CON 38.4 19 .0 17.7 NT 27.7 * 7.8 * 10 .0* RI 39.3 16.9 16.2 OR 33.2 18.3 28.4 Wet year CON 38.6 13.6 8.9 NT 28.9 * 12.3 * 13.4 RI 42.2 15 .0 8.5 * OR 30.7 13 .0 23.5 *Lowest coefficient of variation for the total crop biomass and yield within each climate extreme. 153 Table S5. 1 . Details and timing of management operations for corn - soybean - wheat rotation under the conventional treatment (CON). Month Week of the month Management operation Year April 2 0 - 0 - 60 potassium fertilizer application 1 May 1 Chisel plowing (moldboard plow until 1998) 1 May 1 Soil finishing 1 May 2 Corn planting 1 May 2 19 - 17 - 0 liquid fertilizer application 1 May 3 Herbicide Application 1 June 2 28 - 0 - 0 liquid nitrogen fertilizer application 1 October 3 Corn harvesting 1 November 1 Mowing 1 April 3 0 - 0 - 60 potassium fertilizer application 2 April 4 11 - 52 - 0 monoammonium phosphate fertilizer application 2 May 1 Chisel plowing (moldboard plow until 1998) 2 May 2 Soil finishing 2 May 3 Soybean planting 2 May 4 Herbicide application 2 July 3 Herbicide application 2 August 1 Miticide application (if required) 2 October 1 - 2 Soybean harvesting 2 October 2 Chisel plowing (moldboard plow until 1998) 2 October 2 Soil finishing 2 October 4 Wheat planting 2 March 4 19 - 19 - 19 N - P - K fertilizer application 3 April 2 Herbicide application 3 April 3 28 - 0 - 0 liquid nitrogen fertilizer application 3 July 2 Wheat harvesting 3 August 4 Mowing 3 September 2 - 3 Glyphosate application to kill 3 154 Table S5. 2 . Details and timing of management operations for corn - soybean - wheat rotation under the no - till treatment (NT) . Month Week of the month Management operation Year April 2 0 - 0 - 60 potassium fertilizer application 1 May 2 Corn planting (no - till drill) 1 May 2 19 - 17 - 0 liquid fertilizer application 1 May 3 Herbicide Application 1 June 2 28 - 0 - 0 liquid nitrogen fertilizer application 1 October 3 Corn harvesting 1 November 1 Mowing 1 April 3 0 - 0 - 60 potassium fertilizer application 2 April 4 11 - 52 - 0 monoammonium phosphate fertilizer application 2 May 3 Soybean planting (no - till drill) 2 May 4 Herbicide application 2 July 3 Herbicide application 2 August 1 Miticide application (if required) 2 October 1 - 2 Soybean harvesting 2 October 4 Wheat planting (no - till drill) 2 March 4 19 - 19 - 19 N - P - K fertilizer application 3 April 2 Herbicide application 3 April 3 28 - 0 - 0 liquid nitrogen fertilizer application 3 July 2 Wheat harvesting 3 August 4 Mowing 3 September 2 - 3 Glyphosate application to kill 3 155 Table S5. 3 . Details and timing of management operations for corn - soybean - wheat rotation under the reduced input treatment (RI) . Month Week of the month Management operation Year April 2 0 - 0 - 60 potassium fertilizer application 1 May 1 Chisel plowing (moldboard plow until 1998) 1 May 1 Soil finishing 1 May 2 Corn planting 1 May 2 19 - 17 - 0 liquid fertilizer application 1 May 3 Herbicide Application 1 June 3 Row cultivation (inter row cultivator) 1 October 3 Corn harvesting 1 November 1 Mowing 1 November 2 Planting cereal rye cover crop 1 April 3 0 - 0 - 60 potassium fertilizer application 2 April 3 0 - 48 - 0 p hosphate fertilizer application 2 May 1 Chisel plowing (moldboard plow until 1998) 2 May 2 Soil finishing 2 May 3 Soybean planting 2 June 3 Row cultivation (inter row cultivator) 2 July 3 Herbicide application 2 August 1 Miticide application (if required) 2 October 1 - 2 Soybean harvesting 2 October 2 Chisel plowing (moldboard plow until 1998) 2 October 2 Soil finishing 2 October 4 Wheat planting 2 April 2 Herbicide application 3 April 3 28 - 0 - 0 liquid nitrogen fertilizer application 3 July 2 Wheat harvesting 3 August 1 - 2 Planting red clover cover crop 3 156 Table S5. 4 . Details and timing of management operations for corn - soybean - wheat rotation under the organic treatment (OR) . Month Week of the month Management operation Year May 1 Chisel plowing (moldboard plow until 1998) 1 May 1 Soil finishing 1 May 2 Corn planting 1 June 1 Row cultivation (rotary hoe) 1 June 2 Row cultivation (rotary hoe) 1 June 4 Row cultivation (inter row cultivator) 1 July 1 Row cultivation (inter row cultivator) 1 October 3 Corn harvesting 1 November 1 Mowing 1 November 2 Planting cereal rye cover crop 1 May 1 Chisel plowing (moldboard plow until 1998) 2 May 2 Soil finishing 2 May 3 Soybean planting 2 June 2 Row cultivation (rotary hoe) 2 June 3 Row cultivation (rotary hoe) 2 July 1 Row cultivation (inter row cultivator) 2 July 3 Row cultivation (inter row cultivator) 2 October 1 - 2 Soybean harvesting 2 October 2 Chisel plowing (moldboard plow until 1998) 2 October 2 Soil finishing 2 October 4 Wheat planting 2 July 2 Wheat harvesting 3 August 1 - 2 Planting red clover cover crop 3 157 Table S5. 5 . Means of yield under different production and treatment systems as affected by climate variability* . Production systems Treatment system s C rop Yield ( Mgha - 1 ) Dry years Normal years Wet years Corn CON 4.98 a 8.27 b 8.14 a NT 5.5 0 a 9.48 a 9.25 a RI 5.67 a 7.22 c 8.47 a OR 4.29 b 3.78 d 5.1 0 b Soybean CON 1.41 b 1.79 b 3.04 b NT 1.96 a 2.25 a 3.42 a RI 1.61 b 1.71 bc 3.39 a OR 1.43 b 1.59 c 2.98 b Wheat CON 3.14 b 4.18 ab 3.96 a NT 3.50 a 4.6 0 a 3.35 b RI 2.64 c 4.04 b 3.5 0 b OR 1.75 d 2.31 c 2.24 c * Means with the same letter in a single column for each crop are not significantly different at p<0.05. CON: Conventional treatment; NT: No - till treatment; RI: Reduced input treatment; and OR: Organic treatment. Means of yield, which are significantly hig her in comparison to the conventional treatment, are presented in bold letters. 158 Figure S5. 1 . Box plot of net return of soybean production system under different treatments as affected by climate variability. Note: The thick black line represents the median of the distribution, the red dot is the mean, the edges of the box are 25% and 75% quantiles, whiskers denote the range of the data and the black dots are the outliers. 159 Figure S5. 2 . Box plot of net return of wheat production system under different treatments as affected by climate variability. Note: The thick black line represents the median of the distribution, the red dot is the mean, the edges of the box are 25% and 75% quantiles, whiskers denote the range of the data and the black dots are the outliers. 160 Table S6. 1 . Soil properties at the KBS Main Cropping System Experiment site used to develop the sequential DSSAT model. Soil depth Bulk density Organic carbon Sand Silt Clay Root growth factor in soil* Saturated hydraulic conductivity Field capacity water content at 33kPa* Wilting point water content at 1,500kPa* (cm) (g / cm 3 ) --------------- (%) ---------------- unitless cm / h cm 3 / cm 3 cm 3 / cm 3 0 - 10 1.60 1.10 43 38 19 1.0 0.36 0.267 0.125 10 - 22 1.60 0.90 43 38 19 0.8 0.36 0.267 0.137 22 - 31 1.60 0.70 31 47 22 0.5 0.25 0.267 0.137 31 - 41 1.60 0.30 33 44 23 0.4 0.20 0.295 0.165 41 - 51 1.60 0.22 56 19 25 0.3 0.20 0.297 0.165 51 - 61 1.60 0.10 62 17 21 0.3 0.20 0.267 0.137 61 - 75 1.60 0.05 69 12 19 0.2 0.96 0.267 0.137 75 - 89 1.60 0.02 89 4 7 0.2 1.98 0.160 0.060 89 - 102 1.60 0.02 88 5 7 0.1 20.0 0.160 0.060 102 - 120 1.60 0.02 88 5 7 0.1 20.0 0.160 0.060 *parameters used to calibrate the soil water module of the DSSAT. Table S6. 2 . Probability values for the significance of the effects evaluated in the statistical mixed model for crop yields and net return. Crop Fixed e ffect Probability ( p - value) of the parameters Yield Net return Recharge Corn Treatment ( trt ) <0.0001 <0.0001 <0.0001 Year ( yr ) <0.0001 <0.0001 <0.0001 Interaction between treatment and year ( trt × yr ) <0.0001 <0.0001 <0.0001 Soybean Treatment ( trt ) <0.0001 <0.0001 <0.0001 Year ( yr ) <0.0001 <0.0001 <0.0001 Interaction between treatment and year ( trt × yr ) <0.0001 <0.0001 <0.0001 Wheat Treatment ( trt ) 0.0856 <0.0001 <0.0001 Year ( yr ) <0.0001 <0.0001 <0.0001 Interaction between treatment and year ( trt × yr ) <0.0001 <0.0001 <0.0001 161 Figure S6. 1 . Results of the c alibrated steady - state groundwater model including head contours, color map for head, and velocity vectors . 162 Figure S6. 2 . Mean annual recharge from corn (a), soybean (b), and wheat (c) across different soils in the Kalamazoo River watershed for the period between 1993 - 2019 as affected by the conventional and the no - till treatments. Note: Strongly significant means (p<0.0001 ) are indicated by **, and non - 163 Figure S6. 3 . Mean yield of corn (a), soybean (b), and wheat (c) across different soils in the Kalamazoo River watershed for the period between 1993 - 2019 as affected by the conventional and the no - till treatments. Note: Strongly significant means (p<0.0001) are indica ted by **, significant means (p<0.05) are indicated by *, and non - 164 Figure S6. 4 . Average net return of corn (a), soybean (b), and wheat (c) across different soils in the Kalamazoo River watershed for the period between 1993 - 2019 as affected by the conventional and the no - till treatments. Note: Strongly significant means (p<0.0001) are indicated by **, significant means (p<0.05) are indicated by *, and non - significance cases are 165 REFERENCES 166 REFERENCES Abrahamsen, P., Hansen, S., 2000. Daisy: an open soil - crop - atmosphere system model. Environ. Model. Softw. 15, 313 330. https://doi.org/10.1016/S1364 - 8152(00)00003 - 7 Abson, D.J., Fraser, E.D.G., Benton, T.G., 2013. Landscap e diversity and the resilience of agricultural returns: a portfolio analysis of land - use patterns and economic returns from lowland agriculture. Agric. Food Secur. 2, 2. https://doi.org/10.1186/2048 - 7010 - 2 - 2 Adhikari, U., Nejadhashemi, A.P., Woznicki, S.A. , 2015. Climate change and eastern Africa: a review of impact on major crops. Food Energy Secur. 4, 110 132. https://doi.org/10.1002/fes3.61 Adnan, A.A., Diels, J., Jibrin, J.M., Kamara, A.Y., Craufurd, P., Shaibu, A.S., Mohammed, I.B., Tonnang, Z.E.H., 2019. Options for calibrating CERES - maize genotype specific parameters under data - scarce environments. PLoS One 14, e0200118. https://doi.org/10.1371/journal.pone.0200118 Aggarwal, P.K., Banerjee, B., Daryaei, M.G., Bhatia, A., Bala, A., Rani, S., Chander, S., Pathak, H., Kalra, N., 2006. InfoCrop: A dynamic simulation model for the assessment of crop yields, losses due to pests, and environmental impact of agro - ecosystems in tropical environments. II. Performance of the model. Agric. Syst. 89, 47 6 7. https://doi.org/10.1016/j.agsy.2005.08.003 Ainsworth, E.A., Ort, D.R., 2010. How do we improve crop production in a warming world? Plant Physiol. 154, 526 530. https://doi.org/10.1104/pp.110.161349 Alexander, K., Jackson, J., Kikuyama, F., Sasamoto, B., Stevens, A., 2014. Kalamazoo River Watershed Land Conservation Plan. University of Michigan, School of Natural Resources and Environment, Ann Arbor, MI, USA. Alfani, F., Dabalen, A., Fisker, P., Molini, V., 2015. Can We Measure Resilience? A Proposed Meth od and Evidence from Countries in the Sahel (No. 7170). The World Bank. https://doi.org/10.1596/1813 - 9450 - 7170 Alizadeh, A., 2013. The Principles of Applied Hydrology, 36th Editi. ed. University of Mashhad. Al - Kaisi, M.M., Elmore, R.W., Guzman, J.G., Hanna, H.M., Hart, C.E., Helmers, M.J., Hodgson, E.W., Lenssen, A.W., Mallarino, A.P., Robertson, A.E., Sawyer, J.E., 2013. Drought impact on crop production and the soil environment: 2012 experiences from Iowa. J. Soil Water Conserv. 68, 19 24. https://doi.org/10.2489/jswc.68.1.19A Alley, W.M., 1984. The Palmer Drought Severity Index: Limitations and Assumptions. J. Clim. Appl. Meteorol. 23, 1100 1109. https://doi.org/10.1175/1520 - 0450(1984)023<1100:TPDSIL>2.0.CO ;2 167 Altieri, M.A., Nicholls, C.I., Henao, A., Lana, M.A., 2015. Agroecology and the design of climate change - resilient farming systems. Agron. Sustain. Dev. 35, 869 890. https://doi.org/10.1007/s13593 - 015 - 0285 - 2 Andreadis, K.M., Das, N., Stampoulis, D., Ine s, A., Fisher, J.B., Granger, S., Kawata, J., Han, E., Behrangi, A., 2017. The Regional Hydrologic Extremes Assessment System: A software framework for hydrologic modeling and data assimilation. PLoS One 12, e0176506. https://doi.org/10.1371/journal.pone.0 176506 Andresen, J., Hilberg, S., Kunkel, K., 2012. Historical Climate and Climate Trends in the Midwestern USA, U.S. National Climate Assessment Midwest Technical Input Report. Annan, F., Schlenker, W., 2015. Federal Crop Insurance and the Disincentive to Adapt to Extreme Heat. Am. Econ. Rev. 105, 262 266. https://doi.org/10.1257/aer.p20151031 Antwi - Agyei, P., Dougill, A.J., Stringer, L.C., 2015. Barriers to climate change adaptation: evidence from northeast Ghana in the context of a systematic literature review. Clim. Dev. 7, 297 309. https://doi.org/10.1080/17565529.2014.951013 Araya, A., Kisekka, I., Gowda, P.H., Prasad, P.V.V., 2017. Evaluation of water - limited cropping systems in a semi - arid climate using DSSAT - CSM. Agric. Syst. 150, 86 98. https://doi .org/10.1016/j.agsy.2016.10.007 Arbuckle, J.G., Prokopy, L.S., Haigh, T., Hobbs, J., Knoot, T., Knutson, C., Loy, A., Mase, A.S., McGuire, J., Morton, L.W., Tyndall, J., Widhalm, M., 2013. Climate change beliefs, concerns, and attitudes toward adaptation a nd mitigation among farmers in the Midwestern United States. Clim. Change 117, 943 950. https://doi.org/10.1007/s10584 - 013 - 0707 - 6 Arslan, A., McCarthy, N., Lipper, L., Asfaw, S., Cattaneo, A., 2014. Adoption and intensity of adoption of conservation farmin g practices in Zambia. Agric. Ecosyst. Environ. 187, 72 86. https://doi.org/10.1016/j.agee.2013.08.017 Asfaw, S., Battista, D.F., Lipper, L., 2014. Food security impact of agricultural technology adoption under climate change: Micro - evidence from Niger (No . ESA Working Paper No. 14 - 12). FAO, Rome, Italy. Bai, Z., Caspari, T., Gonzalez, M.R., Batjes, N.H., Mäder, P., Bünemann, E.K., de Goede, R., Tóth, Z., 2018. Effects o f agricultural management practices on soil quality: A review of long - term experiments for Europe and China. Agric. Ecosyst. Environ. 265, 1 7. https://doi.org/10.1016/j.agee.2018.05.028 Bailey, I., Buck, L.E., 2016. Managing for resilience: a landscape fr amework for food and livelihood security and ecosystem services. Food Secur. 8, 477 490. https://doi.org/10.1007/s12571 - 016 - 0575 - 9 Balboa, G.R., Archontoulis, S.V., Salvagiotti, F., Garcia, F.O., Stewart, W.M., Francisco, E., 168 Prasad, P.V.V., Ciampitti, I.A ., 2019. A systems - level yield gap assessment of maize - soybean rotation under high - and low - management inputs in the Western US Corn Belt using APSIM. Agric. Syst. 174, 145 154. https://doi.org/10.1016/j.agsy.2019.04.008 Barker, J.B., Franz, T.E., Heeren, D.M., Neale, C.M.U., Luck, J.D., 2017. Soil water content monitoring for irrigation management: A geostatistical analysis. Agric. Water Manag. 188, 36 49. https://doi.org/10.1016/j.agwat.2017.03.024 Basche, A.D., DeLonge, M.S., 2019. Comparing infiltration rates in soils managed with conventional and alternative farming methods: A meta - analysis. PLoS One 14, 1 22. https://doi.org/10.1371/journal.pone.0215702 Basche, A.D., Kaspar, T.C., Archontoulis, S. V., Jaynes, D.B., Sauer, T.J., Parkin, T.B., Miguez, F. E., 2016. Soil water improvements with the long - term use of a winter rye cover crop. Agric. Water Manag. 172, 40 50. https://doi.org/10.1016/j.agwat.2016.04.006 Basso, B., Ritchie, J.T., Grace, P.R., Sartori, L., 2006. Simulation of Tillage Systems Impact on Soil Biophysical Properties Using the SALUS Model. Ital. J. Agron. 1, 677 688. https://doi.org/10.4081/ija.2006.677 Battel, B., 2018. 2018 Custom Machine and Work Rate Estimates. Michigan State University Extension. Baveye, P.C., Rangel, D., Jacobson, A.R., Laba, M., Darnault, C., Otten, W., Radulovich, R., Camargo, F.A.O., 2011. From Dust Bowl to Dust Bowl: Soils are Still Very Much a Frontier of Science. Soil Sci. Soc. Am. J. 75, 2037 2048. https://doi.org/10.2136/sssaj2011.0145 Behnke, G.D. , Zuber, S.M., Pittelkow, C.M., Nafziger, E.D., Villamil, M.B., 2018. Long - term crop rotation and tillage effects on soil greenhouse gas emissions and crop production in Illinois, USA. Agric. Ecosyst. Environ. 261, 62 70. https://doi.org/10.1016/j.agee.201 8.03.007 Bennett, E., Carpenter, S., Gordon, L., Ramankutty, N., Balvanera, P., Campbell, B., Cramer, W., Foley, J., Folke, C., Karlberg, L., Liu, J., 2014. Toward a More Resilient Agriculture. Solutions 5, 65 75. Berners - Lee, M., Kennelly, C., Watson, R., Hewitt, C.N., 2018. Current global food production is sufficient to meet human nutritional needs in 2050 provided there is radical societal adaptation. Elem. Sci. Anthr. 6, 52. https://doi.org/10.1525/elementa.310 Biggs, R., Schlüter, M., Biggs, D., Bohen sky, E.L., BurnSilver, S., Cundill, G., Dakos, V., Daw, T.M., Evans, L.S., Kotschy, K., Leitch, A.M., Meek, C., Quinlan, A., Raudsepp - Hearne, C., Robards, M.D., Schoon, M.L., Schultz, L., West, P.C., 2012. Toward Principles for Enhancing the Resilience of Ecosystem Services. Annu. Rev. Environ. Resour. 37, 421 448. https://doi.org/10.1146/annurev - environ - 051211 - 123836 Birthal, P.S., Negi, D.S., Khan, M.T., Agarwal, S., 2015. Is Indian agriculture becoming resilient 169 to droughts? Evidence from rice production systems. Food Policy 56, 1 12. https://doi.org/10.1016/j.foodpol.2015.07.005 Blake, G.., Hartge, K.., 1986. Bulk density., in: Methods of Soil Analysis: Part 1 Physical and Minerological Methods. pp. 363 375. Blanco - Canqui, H., Ruis, S.J., 2018. No - tillage and soil physical environment. Geoderma 326, 164 200. https://doi.org/10.1016/j.geoderma.2018.03.011 Blevins, R.L., Cook, D., Phillips, S.H., Phillips, R.E., 1971. Influence of No - tillage on Soil Moisture. Agron. J. 63, 593 596. https://doi.org/ 10.2134/agronj1971.00021962006300040024x Bommarco, R., Kleijn, D., Potts, S.G., 2013. Ecological intensification: harnessing ecosystem services for food security. Trends Ecol. Evol. 28, 230 238. https://doi.org/10.1016/j.tree.2012.10.012 Boote, K.J., Jones , J.W., Pickering, N.B., 1996. Potential Uses and Limitations of Crop Models I. Model Use as a Research Tool. Agron. J. 716, 704 716. Boote, K.J., Jones, J.W., White, J.W., Asseng, S., Liazaso, J.I., 2013. Putting mechanisms into crop production models. Pl ant. Cell Environ. 36, 1658 1672. https://doi.org/10.1111/pce.12119 Bouman, B.A.M., Kropff, M., Tuong, T., Wopereis, M., ten Berge, H., van Laar, H., 2001. ORYZA2000: modeling lowland rice. International Rice Research Institute, Los Baños, Philippines. Bou M., Declerck, F., Dessard, H., Kautsky, E.E., Fabricius, C., Folke, C., Fortmann, L., Hubert, B., Magda, D., Mathevet, R., Norgaard, R.B., Quinlan, A., Staver, C. , 2016. Resilience and development: Mobilizing for transformation. Ecol. Soc. 21, 40. https://doi.org/10.5751/ES - 08754 - 210340 Bowen, W.T., Thornton, P.K., Hoogenboom, G., 1998. The simulation of cropping sequences using DSSAT, in: Tsuji, G.Y., Hoogenboom, G., Thornton, P. (Eds.), Understanding Options for Agricultural Production. Kluwer Academic Publishers, Dordrecht, pp. 313 327. https://doi.org/10.1007/978 - 94 - 017 - 3624 - 4_15 Bowles, T.M., Atallah, S.S., Campbell, E.E., Gaudin, A.C.M., Wieder, W.R., Grandy, A.S., 2018. Addressing agricultural nitrogen losses in a changing climate. Nat. Sustain. 1, 399 408. https://doi.org/10.1038/s41893 - 018 - 0106 - 0 Bowman, M.S., Zilberman, D., 2013. Economic factors affecting diversified farming systems. Ecol. Soc. 18. https:/ /doi.org/10.5751/ES - 05574 - 180133 Boyer, J.S., Byrne, P., Cassman, K.G., Cooper, M., Delmer, D., Greene, T., Gruis, F., Habben, J., Hausmann, N., Kenny, N., Lafitte, R., Paszkiewicz, S., Porter, D., Schlegel, A., 170 Schussler, J., Setter, T., Shanahan, J., Sha rp, R.E., Vyn, T.J., Warner, D., Gaffney, J., 2013. The U.S. drought of 2012 in perspective: A call to action. Glob. Food Sec. 2, 139 143. https://doi.org/10.1016/j.gfs.2013.08.002 Branca, G., Lipper, L., McCarthy, N., Jolejole, M.C., 2013. Food security, climate change, and sustainable land management. A review. Agron. Sustain. Dev. 33, 635 650. https://doi.org/10.1007/s13593 - 013 - 0133 - 1 Bravo - Monroy, L., Potts, S.G., Tzanopoulos, J., 2016. Drivers influencing farmer decisions for adopting organic or conven tional coffee management practices. Food Policy 58, 49 61. https://doi.org/10.1016/j.foodpol.2015.11.003 Brink, L., McCarl, B., 1978. The Tradeoff between Expected Return and Risk Among Cornbelt Farmers. Am. J. Agric. Econ. 60, 259 263. https://doi.org/10. 2307/1240057 Brisson, N., Mary, B., Ripoche, D., Jeuffroy, M.H., Ruget, F., Nicoullaud, B., Gate, P., Devienne - Barret, F., Antonioletti, R., Durr, C., Richard, G., Beaudoin, N., Recous, S., Tayot, X., Plenet, D., Cellier, P., Machet, J. - M., Meynard, J.M., Delécolle, R., 1998. STICS: a generic model for the simulation of crops and their water and nitrogen balances. I. Theory and parameterization applied to wheat and corn. Agronomie 18, 311 346. https://doi.org/10.1051/agro:19980501 Brocca, L., Melone, F., Moramarco, T., Morbidelli, R., 2010. Spatial - temporal variability of soil moisture and its estimation across scales. Water Resour. Res. 46, 1 14. https://doi.org/10.1029/2009WR008016 Brouder, S.M., Gomez - Macpherson, H., 2014. The impact of conservation agr iculture on smallholder agricultural yields: A scoping review of the evidence. Agric. Ecosyst. Environ. 187, 11 32. https://doi.org/10.1016/j.agee.2013.08.010 Browne, N., Kingwell, R., Behrendt, R., Eckard, R., 2013. The relative profitability of dairy, sh eep, beef and grain farm enterprises in southeast Australia under selected rainfall and price scenarios. Agric. Syst. 117, 35 44. https://doi.org/10.1016/j.agsy.2013.01.002 Brusberg, M.D., Shively, R., 2015. Building drought resilience in agriculture: Part nerships and public outreach. Weather Clim. Extrem. 10, 40 49. https://doi.org/10.1016/j.wace.2015.10.003 Buttel, F.H., Gillespie, G.W., Janke, R., Caldwell, B., Sarrantonio, M., 1986. Reduced - input agricultural systems: Rationale and prospects. Am. J. Alt ern. Agric. 1, 58 64. https://doi.org/10.1017/S0889189300000898 Cabell, J.F., Oelofse, M., 2012. An Indicator Framework for Assessing Agroecosystem Resilience. Ecol. Soc. 17, 18. https://doi.org/10.5751/ES - 04666 - 170118 Cairns, J.E., Sonder, K., Zaidi, P.H. , Verhulst, N., Mahuku, G., Babu, R., Nair, S.K., Das, B., Govaerts, B., Vinayan, M.T., Rashid, Z., Noor, J.J., Devi, P., San Vicente, F., Prasanna, B.M., 2012. Maize production in a changing climate. impacts, adaptation, and mitigation 171 strategies, Advance s in Agronomy. https://doi.org/10.1016/B978 - 0 - 12 - 394275 - 3.00006 - 7 Calmon, M.A., Batchelor, W.D., Jones, J.W., Ritchie, J.T., Boote, K.J., Hammond, L.C., 1999. Simulating Soybean Root Growth and Soil Water Extraction Using a Functional Crop Model. Trans. AS AE 42, 1867 1878. https://doi.org/10.13031/2013.13352 Campbell, B.M., Vermeulen, S.J., Aggarwal, P.K., Corner - Dolloff, C., Girvetz, E., Loboguerrero, A.M., Ramirez - Villegas, J., Rosenstock, T., Sebastian, L., Thornton, P.K., Wollenberg, E., 2016. Reducing risks to food security from climate change. Glob. Food Sec. 11, 34 43. https://doi.org/10.1016/j.gfs.2016.06.002 Care, 2009. Climate Vulnerability and Capacity Analysis (CVCA). Care International, Geneva, Switzerland. Carter, M.., Gregorich, E.., 2008. Soi l Sampling and Methods of Analysis, Second Edi. ed. CRC Press. Castellini, M., Fornaro, F., Garofalo, P., Giglio, L., Rinaldi, M., Ventrella, D., Vitti, C., Vonella, A.V., 2019. Effects of no - tillage and conventional tillage on physical and hydraulic prope rties of fine textured soils under winter wheat. Water (Switzerland) 11. https://doi.org/10.3390/w11030484 Ceccarelli, S., Grando, S., Maatougui, M., Michael, M., Slash, M., Haghparast, R., Rahmanian, M., Taheri, A., Al - Yassin, A., Benbelkacem, A., Labdi, M., Mimoun, H., Nachit, M., 2010. Plant breeding and climate changes. J. Agric. Sci. 148, 627 637. https://doi.org/10.1017/S0021859610000651 Challinor, A.J., Müller, C., Asseng, S., Deva, C., Nicklin, K.J., Wallach, D., Vanuytrecht, E., Whitfield, S., Rami rez - Villegas, J., Koehler, A. - K., 2018. Improving the use of crop models for risk assessment and climate change adaptation. Agric. Syst. 159, 296 306. https://doi.org/10.1016/j.agsy.2017.07.010 Challinor, A.J., Watson, J., Lobell, D.B., Howden, S.M., Smith , D.R., Chhetri, N., 2014. A meta - analysis of crop yield under climate change and adaptation. Nat. Clim. Chang. 4, 287 291. https://doi.org/10.1038/nclimate2153 Chapman, S.C., Chakraborty, S., Dreccer, M.F., Howden, S.M., 2012. Plant adaptation to climate change opportunities and priorities in breeding. Crop Pasture Sci. 63, 251 268. https://doi.org/10.1071/CP11303 Choptiany, J.M.H., Phillips, S., Graeub, B.E., Colozza, D., Settle, W., Herren, B., Batello, C., 2017. SHARP: integrating a traditional survey with participatory self - evaluation and learning for climate change resilience assessment. Clim. Dev. 9, 505 517. https://doi.org/10.1080/17565529.2016.1174661 Church, S.P., Haigh, T., Widhalm, M., de Jalon, S.G., Babin, N., Carlton, J.S. , Dunn, M., Fagan, K., Knutson, C.L., Prokopy, L.S., 2017. Agricultural trade publications and the 2012 Midwestern U.S. drought: A missed opportunity for climate risk communication. Clim. 172 Risk Manag. 15, 45 60. https://doi.org/10.1016/j.crm.2016.10.006 Cle mson, 2020. Agronomic Crops Enterprise Budgets [WWW Document]. Enterp. Budgets. URL https://www.clemson.edu/extension/agribusiness/enterprise - budget/index.html (accessed 6.1.20). Coates, D., Pert, P.L., Barron, J., Muthuri, C., Nguyen - Khoa, S., Boelee, E., Jarvis, D.I., 2013. Water - related ecosystem services and food security., in: Managing Water and Agroecosystems for Food Security. CABI, Wallingford, pp. 29 41. https://doi.org/10.1079/9781780640884.0029 Cooper, D.J., Merritt, D.M., 2012. Assess ing the water needs of riparian and wetland vegetation in the western United States. USDA For. Serv. Gen. Tech. Rep. RMRS - GTR - 282 1 125. https://doi.org/10.2737/RMRS - GTR - 282 Cooper, P., Rao, K., Singh, P., Dimes, J., Traore, P., Rao, K., Dixit, P., Twomlow , S., 2009. rainfed agriculture in the semi - arid tropics. J. SAT Agric. Res. 7, 1 19. Corbeels, M., Sakyi, R.K., Kühne, R.F., Whitbread, A., Kühne, R.F., Whitbread, A., 2014 . Meta - analysis of crop responses to conservation agriculture in sub - Saharan Africa. Copenhagen: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). COSA, 2017. Simpler Resilience Measurement: Tools to Diagnose and Improve How Households Fare in Difficult Circumstances from Conflict to Climate Change. The Committee on Sustainability Assessment. Philadelphia, USA. Cosh, M.H., Jackson, T.J., Moran, S., Bindlish, R., 2008. Temporal persistence and stability of surface soil moisture in a semi - arid watershed. Remote Sens. Environ. 112, 304 313. https://doi.org/10.1016/j.rse.2007.07.001 P., van den Belt, M., 1997. The value 260. https://doi.org/10.1038/387253a0 Crane, T.A., Roncoli, C., Hoogenboom, G., 2011. Adaptation to climate change and climate variability: The imp ortance of understanding agriculture as performance. NJAS - Wageningen J. Life Sci. 57, 179 185. https://doi.org/10.1016/j.njas.2010.11.002 Crum, J.R., Collins, H.P., 1995. KBS Soils. KBS LTER Special Publication. Zenodo 1 2. Curtis, Z.K., Li, S. - G., Liao, H. - S., Lusch, D., 2018. Data - Driven Approach for Analyzing Hydrogeology and Groundwater Quality Across Multiple Scales. Groundwater 56, 377 398. https://doi.org/10.1111/gwat.12584 Cusser, S., Bahlai, C., Swinton, S.M., Robertson, G.P., Haddad, N.M., 2020. 173 Biol. 26, 3715 3725. https://doi.org/10.1111/gcb.15080 Dai, S., Shulski, M.D., Hubbard, K.G., Takle, E.S., 2016. A spatiotemporal analysis of Mi dwest US temperature and precipitation trends during the growing season from 1980 to 2013. Int. J. Climatol. 36, 517 525. https://doi.org/10.1002/joc.4354 Daigh, A.L.M., Dick, W.A., Helmers, M.J., Lal, R., Lauer, J.G., Nafziger, E., Pederson, C.H., Strock, J., Villamil, M., Mukherjee, A., Cruse, R., 2018. Yields and yield stability of no - till and chisel - plow fields in the Midwestern US Corn Belt. F. Crop. Res. 218, 243 253. https://doi.org/10.1016/j.fcr.2017.04.002 Dale, V.H., Polasky, S., 2007. Measures of the effects of agricultural practices on ecosystem services. Ecol. Econ. 64, 286 296. https://doi.org/10.1016/j.ecolecon.2007.05.009 Dalin, C., Wada, Y., Kastner, T., Puma, M.J., 2017. Groundwater depletion embedded in international food trade. Nature 543 , 700 704. https://doi.org/10.1038/nature21403 Davies, W.J., Zhang, J., Yang, J., Dodd, I.C., 2011. Novel crop science to improve yield and resource use efficiency in water - limited agriculture. J. Agric. Sci. 149, 123 131. https://doi.org/10.1017/S00218596 10001115 De Ponti, T., Rijk, B., van Ittersum, M.K., 2012. The crop yield gap between organic and conventional agriculture. Agric. Syst. 108, 1 9. https://doi.org/10.1016/j.agsy.2011.12.004 DeClerck, F., Jones, S., Attwood, S., Bossio, D., Girvetz, E., Cha plin - Kramer, B., Enfors, E., Fremier, A., Gordon, L., Kizito, F., Lopez Noriega, I., Matthews, N., McCartney, M., Meacham, M., Noble, A., Quintero, M., Remans, R., Soppe, R., Willemen, L., Wood, S., Zhang, W., 2016. Agricultural ecosystems and their servic es: the vanguard of sustainability? Curr. Opin. Environ. Sustain. 23, 92 99. https://doi.org/10.1016/j.cosust.2016.11.016 Deines, J.M., Wang, S., Lobell, D.B., 2019. Satellites reveal a small positive yield effect from conservation tillage across the US Co rn Belt. Environ. Res. Lett. 14, 124038. https://doi.org/10.1088/1748 - 9326/ab503b Delgado, J.A., Nearing, M.A., Rice, C.W., 2013. Conservation Practices for Climate Change Adaptation, Advances in Agronomy. 121, 47 115. https://doi.org/10.1016/B978 - 0 - 12 - 407 685 - 3.00002 - 5 Denny, R.C.H., Marquart - Pyatt, S.T., Houser, M., 2019. Understanding the Past and Present and the Upper Midwest. Soc. Nat. Resour. 32, 807 826. https://doi. org/10.1080/08941920.2019.1574045 Derpsch, R., Friedrich, T., Kassam, A., Hongwen, L., 2010. Current status of adoption of no - till farming in the world and some of its main benefits. Int. J. Agric. Biol. Eng. 3, 1 25. 174 https://doi.org/10.3965/j.issn.1934 - 63 44.2010.01.001 - 025 Di Falco, S., Chavas, J.P., 2008. Rainfall shocks, resilience, and the effects of crop biodiversity on agroecosystem productivity. Land Econ. 84, 83 96. https://doi.org/10.3368/le.84.1.83 Di Paola, A., Valentini, R., Santini, M., 2016. An overview of available crop growth and yield models for studies and assessments in agriculture. J. Sci. Food Agric. 96, 709 714. https://doi.org/10.1002/jsfa.7359 Diepen, C.A., Wolf, J., Keulen, H., Rappoldt, C., 1989. WOFOST: a simulation model of crop production. Soil Use Manag. 5, 16 24. https://doi.org/10.1111/j.1475 - 2743.1989.tb00755.x Dillon, A., Mcgee, K., Oseni, G., Dillon, A., Mcgee, K., Oseni, G., 2015. Agricultural Production , Dietary Diversity and Climate Variability Agricultural Production , Dietary Diversity and Climate Variability. J. Dev. Stud. 51, 976 995. https://doi.org/10.1080/00220388.2015.1018902 Dinar, A., Tieu, A., Huynh, H., 2019. Water scarcity impacts on global food production. Glob. Food Sec. 23, 212 22 6. https://doi.org/10.1016/j.gfs.2019.07.007 Ding, Y., Schoengold, K., Tadesse, T., 2009. The Impact of Weather Extremes on Agricultural Practices? J. Agric. Resour. Econ. 34, 395 411. Dixon, J.L., Stringer, L.C., 2015. Towards a theoretical grounding of climate resilience assessments for smallholder farming systems in Sub - Saharan Africa. Resources 4, 128 154. https://doi.org/10.3390/resources4010128 Dokoohaki, H., Gheysari, M., Mou savi, S. - F., Zand - Parsa, S., Miguez, F.E., Archontoulis, S. V., Hoogenboom, G., 2016. Coupling and testing a new soil water module in DSSAT CERES - Maize model for maize production under semi - arid condition. Agric. Water Manag. 163, 90 99. https://doi.org/10 .1016/j.agwat.2015.09.002 Doll, J.E., Petersen, B., Bode, C., 2017. Skeptical but Adapting: What Midwestern Farmers Say about Climate Change. Weather. Clim. Soc. 9, 739 751. https://doi.org/10.1175/WCAS - D - 16 - 0110.1 Douxchamps, S., Debevec, L., Giordano, M. , Barron, J., 2017. Monitoring and evaluation of climate resilience for agricultural development A review of currently available tools. World Dev. Perspect. 5, 10 23. https://doi.org/10.1016/j.wdp.2017.02.001 Eeswaran, R., 2018. Climate Change Impacts an d Adaptation in the Agriculture Sector of Sri Lanka: What We Learnt and Way Forward, in: Filho, W.L., Manolas, E., Azul, A.M., Azeiteiro, U.M., McGhie, H. (Eds.), Handbook of Climate Change Communication. Springer, Cham, pp. 97 110. https://doi.org/10.1007 /978 - 3 - 319 - 70066 - 3_8 Eeswaran, R., Nejadhashemi, A.P., Alves, F.C., Saravi, B., 2021. Evaluating the applicability of 175 soil moisture - based metrics for gauging the resiliency of rainfed agricultural systems in the midwestern United States. Soil Tillage Res. 205, 104818. https://doi.org/10.1016/j.still.2020.104818 Eitzinger, A., Läderach, P., Rodriguez, B., Fisher, M., Beebe, S., Sonder, K., Schmidt, A., 2017. Assessing high - impact spots of climate change: spatial yield simulations with Decision Support System for Agrotechnology Transfer (DSSAT) model. Mitig. Adapt. Strateg. Glob. Chang. 22, 743 760. https://doi.org/10.1007/s11027 - 015 - 9696 - 2 El Chami, D., Daccache, A., El Moujabber, M., 2020. How Can Sustainable Agriculture Increase Climate Resilience? A System atic Review. Sustainability 12, 3119. https://doi.org/10.3390/su12083119 Erenstein, O., 2003. Smallholder conservation farming in the tropics and sub - the development and dissemination of mulching with crop residues and cover crops. Agr ic. Ecosyst. Environ. 100, 17 37. https://doi.org/10.1016/S0167 - 8809(03)00150 - 6 ERS, 2020. Farming and Farm Income [WWW Document]. Econ. Res. Serv. United States Dep. Agric. Washington, DC. URL https://www.ers.usda.gov/data - products/ag - and - food - statistics - charting - the - essentials/farming - and - farm - income/ (accessed 10.10.20). Esfahanian, E., Nejadhashemi, A.P., Abouali, M., Adhikari, U., Zhang, Z., Daneshvar, F., Herman, M.R., 2017. Development and evaluation of a comprehensive drought index. J. Environ. Mana ge. 185, 31 43. https://doi.org/10.1016/j.jenvman.2016.10.050 Evett, S., 2008. Gravimetric and volumetric direct measurements of soil water content, in: Field Estimation of Soil Water Content. A Practical Guide to Methods, Instrumentation and Sensor Techno logy. International Atomic Energy Agency, Vienna, Austria. Fang, Q., Ma, L., Yu, Q., Malone, R.W., Saseendran, S.A., Ahuja, L.R., 2008. Modeling Nitrogen and Water Management Effects in a Wheat - Maize Double - Cropping System. J. Environ. Qual. 37, 2232 2242. https://doi.org/10.2134/jeq2007.0601 and Agriculture Organization of the United Nations, Rome, Italy. FAO, 2016. Analyzing resilience for better targeting and acti on. Resilience Index Measurement and Analysis II (RIMA - II). Food and Agriculture Organization of the United Nations, Rome, Italy. FAO, 2017. The future of food and agriculture. Trends and challenges. Food and Agriculture Organization of the United Nations, Rome, Italy. Fedele, G., Locatelli, B., Djoudi, H., 2017. Mechanisms mediating the contribution of ecosystem services to human well - being and resilience. Ecosyst. Serv. 28, 43 54. https://doi.org/10.1016/j.ecoser.2017.09.011 Ficklin, D.L., Letsinger, S.L., Gholizadeh, H., Maxwell, J.T., 2015. Incorporation of the 176 Penman Monteith potential evapotranspiration method into a Palmer Drought Severity Index Tool. Comput. Geosci. 85, 136 141. https://doi.org/10.1016/j.cageo.2015.09.013 Fisher, B., T urner, R.K., Morling, P., 2009. Defining and classifying ecosystem services for decision making. Ecol. Econ. 68, 643 653. https://doi.org/10.1016/j.ecolecon.2008.09.014 Fleming, A., Vanclay, F., 2010. Farmer responses to climate change and sustainable agri culture. A review. Agron. Sustain. Dev. 30, 11 19. https://doi.org/10.1051/agro/2009028 Fodor, N., Challinor, A., Droutsas, I., Ramirez - Villegas, J., Zabel, F., Koehler, A. - K., Foyer, C.H., 2017. Integrating Plant Science and Crop Modeling: Assessment of the Impact of Climate Change on Soybean and Maize Production. Plant Cell Physiol. 58, 1833 1847. https://doi.org/10.1093/pcp/pcx141 Foley, J.A., Ramankutty, N., Brauman, K.A., Cassidy, E.S., Gerber, J.S., Johnston, M., Mueller, l, C., Ray, D.K., West, P.C., Balzer, C., Bennett, E.M., Carpenter, S.R., Hill, J., Monfreda, C., Polasky, S., Rockström, J., Sheehan, J., Siebert, S., Tilman, D., Zaks, D.P.M., 2011. Solutions for a cultivated planet. Nature 478, 337 342. https://doi.org/ 10.1038/nature10452 Ford, T.W., Labosier, C.F., 2017. Agricultural and Forest Meteorology Meteorological conditions associated with the onset of fl ash drought in the Eastern United States. Agric. For. Meteorol. 247, 414 423. https://doi.org/10.1016/j.agrf ormet.2017.08.031 Franzluebbers, A., Steiner, J.L., Karlen, D., Griffin, T., Singer, J., Tanaka, D., 2011. Rainfed Farming Systems in the USA, in: Rainfed Farming Systems. Springer, Dordrecht, pp. 511 560. Fuchs, B.A., Wood, D.A., Ebbeka, D., 2015. From to o much to too little: How the central US drought of 2012 evolved out of one of the most devastating floods on record in 2011. National Drought Mitigation Center (NDMC), University of Nebraska Lincoln, USA. Gaba, S., Lescourret, F., Boudsocq, S., Enjalbert, J., Hinsinger, P., Journet, E. - P., Navas, M. - L., Wery, J., Louarn, G., Malézieux, E., Pelzer, E., Prudent, M., Ozier - Lafontaine, H., 2015. Multiple cropping systems as drivers for providing multiple ecosystem services: from concepts to design. Agron. Sust ain. Dev. 35, 607 623. https://doi.org/10.1007/s13593 - 014 - 0272 - z Poppy, L., 2015. Diversifying crop rotations with pulses enhances system productivity. Sci. Rep. 5, 14625. https://doi.org/10.1038/srep14625 Gao, L., Lv, Y., Wang, D., Muhammad, T., Biswas, A., Peng, X., 2016. Soil water storage prediction at high space time resolution along an agricultural hillslope. Agric. Water Manag. 165, 122 130. https://doi.org/10. 1016/j.agwat.2015.11.012 Gao, L., Shao, M., 2012. Temporal stability of soil water storage in diverse soil layers. Catena 177 95, 24 32. https://doi.org/10.1016/j.catena.2012.02.020 Garbach, K., Milder, J.C., Montenegro, M., Karp, D.S., DeClerck, F.A.J., 2014. Biodiversity and Ecosystem Services in Agroecosystems, in: Encyclopedia of Agriculture and Food Systems. Elsevier, pp. 21 40. https://doi.org/10.1016/B978 - 0 - 444 - 52512 - 3.00013 - 9 Gassman, P.W., Reyes, M.R., Green, C.H., Arnold, J.G., 2007. The Soil and Wate r Assessment Tool: Historical Development, Applications, and Future Research Directions. Trans. ASABE 50, 1211 1250. https://doi.org/10.13031/2013.23637 Gathala, M.K., Kumar, V., Sharma, P.C., Saharawat, Y.S., Jat, H.S., Singh, M., Kumar, A., Jat, M.L., Hu mphreys, E., Sharma, D.K., Sharma, S., Ladha, J.K., 2013. Optimizing intensive cereal - based cropping systems addressing current and future drivers of agricultural change in the northwestern Indo - Gangetic Plains of India. Agric. Ecosyst. Environ. 177, 85 97 . https://doi.org/10.1016/j.agee.2013.06.002 Gentry, L.E., Snapp, S.S., Price, R.F., Gentry, L.F., 2013. Apparent red clover Nitrogen credit to corn: Evaluating cover crop introduction. Agron. J. 105, 1658 1664. https://doi.org/10.2134/agronj2013.0089 Gil, J.D.B., Cohn, A.S., Duncan, J., Newton, P., Vermeulen, S., 2017. The resilience of integrated agricultural systems to climate change. Wiley Interdiscip. Rev. Clim. Chang. 8, e461. https://doi.org/10.1002/wcc.461 Gomiero, T., Pimentel, D., Paoletti, M.G., 2011. Environmental impact of different agricultural management practices: Conventional vs. Organic agriculture. CRC. Crit. Rev. Plant Sci. 30, 95 124. https://doi.org/10.1080/07352689.2011.554355 Gordon, L.J., Finlayson, C.M., Falkenmark, M., 2010. Managi ng water in agriculture for food production and other ecosystem services. Agric. Water Manag. 97, 512 519. https://doi.org/10.1016/j.agwat.2009.03.017 Grannemann, N.G., Hunt, R.J., Nicholas, J.., Reilly, T.E., Winter, T.., 2008. The importance of ground wa ter in the Great Lakes Region, Water - Resources Investigations Report 00 4008, U.S. Geological Survey. Grassini, P., Yang, H., Irmak, S., Thorburn, J., Burr, C., Cassman, K.G., 2011. High - yield irrigated maize in the Western U.S. Corn Belt: II. Irrigation m anagement and crop water productivity. F. Crop. Res. 120, 133 141. https://doi.org/10.1016/j.fcr.2010.09.013 Griebler, C., Avramov, M., 2015. Groundwater ecosystem services: a review. Freshw. Sci. 34, 355 367. https://doi.org/10.1086/679903 Hadley, P.W., N ewell, C., 2014. The New Potential for Understanding Groundwater Contaminant Transport. Groundwater 52, 174 186. https://doi.org/10.1111/gwat.12135 Haigh, T., Takle, E., Andresen, J., Widhalm, M., Carlton, J.S., Angel, J., 2015. Mapping the decision points and climate information use of agricultural producers across the U.S. 178 Corn Belt. Clim. Risk Manag. 7, 20 30. https://doi.org/10.1016/j.crm.2015.01.004 Han, E., Ines, A., Koo, J., 2015. Global high - resolution soil profile database for crop m odeling applications. Harvard Dataverse. https://doi.org/10.7910/DVN/1PEEY0 Han, E., Ines, A.V.M., Koo, J., 2019. Development of a 10 - km resolution global soil profile dataset for crop modeling applications. Environ. Model. Softw. 119, 70 83. https://doi.o rg/10.1016/j.envsoft.2019.05.012 Hansen, J.W., Mason, S.J., Sun, L., Tall, A., 2011. Review of seasonal climate forecasting for agriculture in sub - Saharan Africa. Exp. Agric. 47, 205 240. https://doi.org/10.1017/S0014479710000876 Hanson, B.R., Orloff, S., Peters, D., 2000. Monitoring soil moisture helbs refine irrigation management. Calif. Agric. 54, 38 42. Harbaugh, A.W., 1995. MODFLOW - 2005, the US Geological Survey modular ground - water model: the ground - water flow process. US Department of Int erior, US Geological Survey. Harrington, L., Tow, P., 2011. Types of Rainfed Farming Systems Around the World, in: Rainfed Farming Systems. Springer, Dordrecht, pp. 45 74. https://doi.org/10.1007/978 - 1 - 4020 - 9132 - 2 Harvey, C.A., Chacón, M., Donatti, C.I., G aren, E., Hannah, L., Andrade, A., Bede, L., Brown, D., Calle, A., Chará, J., Clement, C., Gray, E., Hoang, M.H., Minang, P., Rodríguez, Torquebiau, E., Etten, J., Wollenberg Opportunities and Challenges for Integrating Adaptation and Mitigation in Tropical Agriculture. Conserv. Lett. 7, 77 90. https://doi.org/10.1111/conl.12066 Hatfield, J.L., Antle, J., Garrett, K.A., Izaurralde, R.C., Ma der, T., Marshall, E., Nearing, M., Philip Robertson, G., Ziska, L., 2018. Indicators of climate change in agricultural systems. Clim. Change 1 14. https://doi.org/10.1007/s10584 - 018 - 2222 - 2 Hatfield, J .L. , 2012. Agriculture in the Midwest, in: Winkler, J., Andresen, J., Hatfield, J, Bidwell, D., Brown, D. (Eds.), U.S. National Climate Assessment Midwest Technical Input Report. pp. 1 8. He, Z., Zhao, M., Zhu, X., Du, J., Chen, L., Lin, P., Li, J., 2019. Temporal stability of soil water storage in multiple so il layers in high - elevation forests. J. Hydrol. 569, 532 545. https://doi.org/10.1016/j.jhydrol.2018.12.024 Heath, R.C., 1983. Basic ground - water hydrology. U.S. Geological Survey. https://doi.org/10.3133/wsp2220 Herberich, E., Sikorski, J., Hothorn, T., 2010. A Robust Procedure for Comparing Multiple Means under Heteroscedasticity in Unbalanced Designs. PLoS One 5, e9788. 179 https://doi.org/10.1371/journal.pone.0009788 Hoag, D.L.K., Parsons, J., 2010. Risk Navigator SRM: An Applied Risk Management Tool. J. P robab. Stat. 2010, 1 17. https://doi.org/10.1155/2010/214358 Hobbs, P.R., Sayre, K., Gupta, R., 2008. The role of conservation agriculture in sustainable agriculture. Philos. Trans. Biol. Sci. 363, 543 555. https://doi.org/10.1098/rstb.2007.2169 Holling, C ., 1973. Resilience and Stability of Ecological Systems. Annu.Rev.Ecol.Syst. 4, 1 23. Holtschlag, D., 1997. A generalized estimate of ground - water - recharge rates in the Lower Peninsula of Michigan. Department of the Interior, US Geological Survey, Washingt on, DC. Homer, C., Dewitz, J., Jin, S., Xian, G., Costello, C., Danielson, P., Gass, L., Funk, M., Wickham, J., Stehman, S., Auch, R., Riitters, K., 2020. Conterminous United States land cover change patterns 2001 2016 from the 2016 National Land Cover Dat abase. ISPRS J. Photogramm. Remote Sens. 162, 184 199. https://doi.org/10.1016/j.isprsjprs.2020.02.019 Hoogenboom, G., Porter, C.H., Shelia, V., Boote, K.J., Singh, U., White, J.W., Hunt, L.A., Ogoshi, R., Lizaso, J.I., Koo, J., Asseng, S., Singels, A., Moreno, L.P., Jones, J.W., 2019. Decision Support System for Agrotechnology Transfer (DSSAT) Version 4.7.5 (https://DSSAT.net), DSSAT Foundation, Gainesville, Florida, USA. Howden, S.M., Soussana, J.F., Tubiello, F.N., Chhetri, N., Dunlop, M., Meinke, H., 2007. Adapting agriculture to climate change. Proc. Natl. Acad. Sci. 104, 19691 19696. https://doi.org/10.1073/pnas.0701890104 Hsiang, S.., Jina, A.., 2014. The causal effect of environmental catastrophe on long - run economic growth: Evidence f rom 6,700 cyclones (No. 20352). National Bureau of Economic Research. Cambridge, Massachusetts, USA. Hu, W., Shao, M., Han, F., Reichardt, K., Tan, J., 2010. Watershed scale temporal stability of soil water content. Geoderma 158, 181 198. https://doi.org/10.1016/j.geoderma.2010.04.030 Hungria, M., Mendes, I.C., 2015. Nitrogen Fixation with Soybean: The Perfect Symbiosis?, in: De Bruijn, F.. (Ed.), Biological Nitrogen Fixation. John Wiley & Sons, pp. 1009 1024. https://doi.org/10.1002/978111 9053095.ch99 Hussain, M.Z., Bhardwaj, A.K., Basso, B., Robertson, G.P., Hamilton, S.K., 2019. Nitrate Leaching from Continuous Corn, Perennial Grasses, and Poplar in the US Midwest. J. Environ. Qual. 48, 1849 1855. https://doi.org/10.2134/jeq2019.04.0156 I Development, Winnipeg, Canada. 180 Iocola, I., Bassu, S., Farina, R., Antichi, D., Basso, B., Bindi, M., Dalla Marta, A., Danuso, F., Doro, L., Ferrise, R., Giglio, L., Ginaldi, F., Mazzoncini, M., Mula, L., Orsini, R., Corti, G., Pasqui, M., Seddaiu, G., Tomozeiu, R., Ventrella, D., Villani, G., Roggero, P.P., 2017. Can conservation tillage mitigate climate change impacts in Mediterranean cereal systems? A soil organic carbon assessment using long term experiments. Eur. J. Agron. 90, 96 107. https://doi.org/10.1016/j.eja.2017.07.011 IPCC, 2018. Summary for Policymakers, in: Masson - Delmotte, V., Zhai, P., Pörtner, H.O., Roberts, D., Skea, J., Shukla, P.R., Pirani, A., Mou fouma - Okia, W., Péan, C., Pidcock, R., Connors, S., Matthews, J.B.R., Chen, Y., Zhou, X., Gomis, M.I., Lonnoy, E., Maycock, T., Tignor, M., Waterfield, T. (Eds.), Global Warming of 1.5°C. An IPCC Special Report on the Impacts of Global Warming of 1.5°C abo ve Pre - Industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate Change,. Intergovernmental Panel on Climate Change, Geneva, Switzerland. Jacobi, J., Perrone, D., Dun can, L.L., Hornberger, G., 2013. A tool for calculating the Palmer drought indices. Water Resour. Res. 49, 6086 6089. https://doi.org/10.1002/wrcr.20342 Jacobs, J.M., Mohanty, B.P., Hsu, E.C., Miller, D., 2004. SMEX02: Field scale variability, time stabili ty and similarity of soil moisture. Remote Sens. Environ. 92, 436 446. https://doi.org/10.1016/j.rse.2004.02.017 Jägermeyr, J., Gerten, D., Schaphoff, S., Heinke, J., Lucht, W., Rockström, J., 2016. Integrated crop water management might sustainably halve the global food gap. Environ. Res. Lett. 11, 025002. https://doi.org/10.1088/1748 - 9326/11/2/025002 Jat, R.K., Sapkota, T.B., Singh, R.G., Jat, M.L., Kumar, M., Gupta, R.K., 2014. Seven years of conservation agriculture in a rice wheat rotation of Eastern G angetic Plains of South Asia: Yield trends and economic profitability. F. Crop. Res. 164, 199 210. https://doi.org/10.1016/j.fcr.2014.04.015 Jha, P.K., 2019. Agronomic Management of Corn Using Seasonal Climate Predictions, Remote Sensing and Crop Simulatio n Models. Ph.D. Dissertation, Michigan State University , MI, USA . Jin, X., Kumar, L., Li, Z., Feng, H., Xu, X., Yang, G., Wang, J., 2018 a . A review of data assimilation of remote sensing and crop models. Eur. J. Agron. 92, 141 152. https://doi.org/10.1016/ j.eja.2017.11.002 Jin, Z., Ainsworth, E.A., Leakey, A.D.B., Lobell, D.B., 2018 b . Increasing drought and diminishing benefits of elevated carbon dioxide for soybean yields across the US Midwest. Glob. Chang. Biol. 24, e522 e533. https://doi.org/10.1111/gcb. 13946 Jin, Z., Archontoulis, S. V., Lobell, D.B., 2019. How much will precision nitrogen management pay off? An evaluation based on simulating thousands of corn fields over the US Corn - Belt. F. Crop. Res. 240, 12 22. https://doi.org/10.1016/j.fcr.2019.04.0 13 181 Jones, J.W., Antle, J.M., Basso, B., Boote, K.J., Conant, R.T., Foster, I., Godfray, H.C.J., Herrero, M., Howitt, R.E., Janssen, S., Keating, B.A., Munoz - Carpena, R., Porter, C.H., Rosenzweig, C., Wheeler, T.R., 2017a. Brief history of agricultural syst ems modeling. Agric. Syst. 155, 240 254. https://doi.org/10.1016/j.agsy.2016.05.014 Jones, J.W., Antle, J.M., Basso, B., Boote, K.J., Conant, R.T., Foster, I., Godfray, H.C.J., Herrero, M., Howitt, R.E., Janssen, S., Keating, B.A., Munoz - Carpena, R., Porte r, C.H., Rosenzweig, C., Wheeler, T.R., 2017b. Toward a new generation of agricultural system data, models, and knowledge products: State of agricultural systems science. Agric. Syst. 155, 269 288. https://doi.org/10.1016/j.agsy.2016.09.021 Jones, J.W., Hoogenboom, G., Porter, C.H., Boote, K.J., Batchelor, W.D., Hunt, L.A., Wilkens, P.W., Singh, U., Gijsman, A.J., Ritchie, J.T., 2003. The DSSAT cropping system model. Eur. J. Agron. 18, 235 265. https://doi.org/10.1016/S1161 - 0301(02)00107 - 7 Jones, P. G., Thornton, P.K., 2009. Croppers to livestock keepers: livelihood transitions to 2050 in Africa due to climate change. Environ. Sci. Policy 12, 427 437. https://doi.org/10.1016/j.envsci.2008.08.006 Joshi, C., Mohanty, B.P., Jacobs, J.M., Ines, A.V.M., 20 11. Spatiotemporal analyses of soil moisture from point to footprint scale in two different hydroclimatic regions. Water Resour. Res. 47. https://doi.org/10.1029/2009WR009002 Joshi, N., Singh, A.K., Madramootoo, C.A., 2017. Application of DSSAT Model to Si mulate Corn Yield under Long - Term Tillage and Residue Practices. Trans. ASABE 60, 67 83. https://doi.org/10.13031/trans.11545 Kandulu, J.M., Bryan, B.A., King, D., Connor, J.D., 2012. Mitigating economic risk from climate variability in rain - fed agricultur e through enterprise mix diversification. Ecol. Econ. 79, 105 112. https://doi.org/10.1016/j.ecolecon.2012.04.025 Kassam, A., Friedrich, T., Derpsch, R., 2019. Global spread of Conservation Agriculture. Int. J. Environ. Stud. 76, 29 51. https://doi.org/10.1080/00207233.2018.1494927 Kassie, B.T., Rotter, R.P., Hengsdijk, H., Asseng, S., Van Ittersum, M.K., Kahiluoto, H., Van Keulen, H., 2014. Climate variability and change in the Central Rift Valley of Ethiopia: challenges for rainfed crop production. J. Agric. Sci. 152, 58 74. https://doi.org/10.1017/S0021859612000986 Keating, B., Carberry, P., Hammer, G., Probert, M., Robertson, M., Holzworth, D., Huth, N., Hargreaves, J.N., Meinke, H., Hochman, Z., McLean, G., Verburg, K., Snow, V., Dime s, J., Silburn, M., Wang, E., Brown, S., Bristow, K., Asseng, S., Chapman, S., McCown, R., Freebairn, D., Smith, C., 2003. An overview of APSIM, a model designed for farming systems simulation. Eur. J. Agron. 18, 267 288. https://doi.org/10.1016/S1161 - 0301 (02)00108 - 9 Kessler, J.J., Moolhuijzen, M., 1994. Low external input sustainable agriculture: expectations and realities. Netherlands J. Agric. Sci. 182 Klemm, T., McPherson, R.A., 2017. The development of seasonal climate forecasting for agricultural producer s. Agric. For. Meteorol. 232, 384 399. https://doi.org/10.1016/j.agrformet.2016.09.005 Kogan, F.N., 1995. Droughts of the Late 1980s in the United States as Derived from NOAA Polar - Orbiting Satellite Data. Bull. Am. Meteorol. Soc. 76, 655 668. https://doi. org/10.1175/1520 - 0477(1995)076<0655:DOTLIT>2.0.CO;2 Komarek, A.M., Bell, L.W., Whish, J.P.M., Robertson, M.J., Bellotti, W.D., 2015. Whole - farm economic, risk and resource - use trade - offs associated with integrating forages into crop - livestock systems in we stern China. Agric. Syst. 133, 63 72. https://doi.org/10.1016/j.agsy.2014.10.008 Kravchenko, A.N., Robertson, G.P., Thelen, K.D., Harwood, R.R., 2005. Management, Topographical, and Weather Effects on Spatial Variability of Crop Grain Yields. Agron. J. 97, 514 523. https://doi.org/10.2134/agronj2005.0514 Kravchenko, A.N., Wang, A.N.W., Smucker, A.J.M., Rivers, M.L., 2011. Long - term Differences in Tillage and Land Use Affect Intra - aggregate Pore Heterogeneity. Soil Sci. Soc. Am. J. 75, 1658 1666. https://doi .org/10.2136/sssaj2011.0096 Kropp, I., Nejadhashemi, A.P., Deb, K., Abouali, M., Roy, P.C., Adhikari, U., Hoogenboom, G., 2019. A multi - objective approach to water and nutrient efficiency for sustainable agricultural intensification. Agric. Syst. 173, 289 302. https://doi.org/10.1016/j.agsy.2019.03.014 KRWC, 2011. Kalamazoo River Watershed Management Plan. Kalamazoo River Watershed Council, Prepared for the Michigan Nonpoint Source Program (Michigan Department of Environmental Quality and the United States Environmental Protection Agency). Kumar, N.S., Anuja, Rashid, M., Bandyopadhyay, S.K., Padaria, R., Khanna, M., 2016. Adaptation of farming community to climatic risk: Does adaptation cost for sustaining agricultural profitability? Curr. Sci. 110, 1216 122 4. Kuwayama, Y., Thompson, A., Bernknopf, R., Zaitchik, B., Vail, P., 2019. Estimating the impact of drought on agriculture using the U.S. Drought Monitor. Am. J. Agric. Econ. 101, 193 210. https://doi.org/10.1093/ajae/aay037 Lal, R., 2013. Enhancing ecosy stem services with no - till. Renew. Agric. Food Syst. 28, 102 114. https://doi.org/10.1017/S1742170512000452 Lal, R., Delgado, J.A., Gulliford, J., Nielsen, D., Rice, C.W., Pelt, R.S. Van, 2012. Adapting agriculture to drought and extreme events. J. Soil Wa ter Conserv. 67, 162 166. https://doi.org/10.2489/jswc.67.6.162A Lampurlanés, J., Plaza - Bonilla, D., Álvaro - Fuentes, J., Cantero - Martínez, C., 2016. Long - term analysis of soil water conservation and crop yield under different tillage systems in Mediterrane an rainfed conditions. F. Crop. Res. 189, 59 67. 183 https://doi.org/10.1016/j.fcr.2016.02.010 Landis, D.A., 2017. Designing agricultural landscapes for biodiversity - based ecosystem services. Basic Appl. Ecol. 18, 1 12. https://doi.org/10.1016/j.baae.2016.07.0 05 Langemeier, M., 2016. Measuring farm profitability. farmdoc Dly. 6, 1 2. University of Illinois at Urbana - Champaign , IL, USA. Lee, S., Lee, D.K., 2018. What is the proper way to apply the multiple comparison test? Korean J. Anesthesiol. 71, 353 360. htt ps://doi.org/10.4097/kja.d.18.00242 Lemos, M.C., Lo, Y.J., Kirchhoff, C., Haigh, T., 2014. Crop advisors as climate information brokers: Building the capacity of US farmers to adapt to climate change. Clim. Risk Manag. 4 5, 32 42. https://doi.org/10.1016/j .crm.2014.08.001 Leng, G., Peng, J., Huang, S., 2019. Recent changes in county - level maize production in the United States: Spatial - temporal patterns, climatic drivers and the implications for crop modelling. Sci. Total Environ. 686, 819 827. https://doi.o rg/10.1016/j.scitotenv.2019.06.026 Lesk, C., Rowhani, P., Ramankutty, N., 2016. Influence of extreme weather disasters on global crop production. Nature 529, 84 87. https://doi.org/10.1038/nature16467 Li, F.J., Dong, S.C., Li, F., 2012. A system dynamics model for analyzing the eco - agriculture system with policy recommendations. Ecol. Modell. 227, 34 45. https://doi.org/10.1016/j.ecolmodel.2011.12.005 Li, M., Peterson, C.A., Tautges, N.E., Scow, K.M., Gaudin, A.C.M., 2019. Yields and resilience outcomes of organic, cover crop, and conventional practices in a Mediterranean climate. Sci. Rep. 9, 12283. https://doi.org/10.1038/s41598 - 019 - 48747 - 4 Li, S.G., Liu, Q., 2006. A Real - Time, Interactive Steering Environment for Integrated Ground W ater Modeling. Ground Water 44, 758 763. https://doi.org/10.1111/j.1745 - 6584.2006.00225.x Liang, X., Lettenmaier, D.P., Wood, E.F., Burges, S.J., 1994. A simple hydrologically based model of land surface water and energy fluxes for general circulation mode ls. J. Geophys. Res. 99, 14415. https://doi.org/10.1029/94JD00483 Liao, H.S., Curtis, Z.K., Sampath, P.V., Li, S.G., 2020. Simulation of Flow in a Complex 322. https://doi.org/1 0.1111/gwat.12918 Liao, H.S., Sampath, P. V., Curtis, Z.K., Li, S.G., 2015a. Hierarchical modeling of a groundwater remediation capture system. J. Hydrol. 527, 196 211. https://doi.org/10.1016/j.jhydrol.2015.04.057 Liao, H.S., Sampath, P. V., Curtis, Z.K., Li, S.G., 2015b. Hierarchical Modeling and Parameter 184 Estimation for a Coupled Groundwater Lake System. J. Hydrol. Eng. 20, 04015027. https://doi.org/10.1061/(ASCE)HE.1943 - 5584.0001219 Lin, B.B., 2011. Resilience in Agriculture through Crop Diversification : Adaptive Management for Environmental Change. Bioscience 61, 183 193. https://doi.org/10.1525/bio.2011.61.3.4 Lindström, G., Johansson, B., Persson, M., Gardelin, M., Bergström, S., 1997. Development and test of the distributed HBV - 96 hydrological model. J. Hydrol. 201, 272 288. https://doi.org/10.1016/S0022 - 1694(97)00041 - 3 Lipper, L., Thornton, P., Campbell, B.M., Baedeker, T., Braimoh, A., Bwalya, M., Caron, P., Cattaneo, A., Garrity, D., Henry, K., Hottle, R., Jackson, L., Jarvis, A., Kossam, F., Mann, W., McCarthy, N., Meybeck, A., Neufeldt, H., Remington, T., Sen, P.T., Sessa, R., Shula, R., Tibu, A., Torquebiau, E.F., 2014. Climate - smart agriculture for food security. Nat. Clim. Chang. 4, 1068 1072. https://doi.org/10.1038/nclimate2437 Liu, B., Assen g, S., Liu, L., Tang, L., Cao, W., Zhu, Y., 2016. Testing the responses of four wheat crop models to heat stress at anthesis and grain filling. Glob. Chang. Biol. 22, 1890 1903. https://doi.org/10.1111/gcb.13212 Liu, S., Yang, J.Y., Zhang, X.Y., Drury, C.F ., Reynolds, W.D., Hoogenboom, G., 2013 a . Modelling crop yield, soil water content and soil temperature for a soybean maize rotation under conventional and conservation tillage systems in Northeast China. Agric. Water Manag. 123, 32 44. https://doi.org/10. 1016/j.agwat.2013.03.001 Liu, Y., Gao, M., Wu, W., Tanveer, S.K., Wen, X., Liao, Y., 2013 b . The effects of conservation tillage practices on the soil water - holding capacity of a non - irrigated apple orchard in the Loess Plateau, China. Soil Tillage Res. 130 , 7 12. https://doi.org/10.1016/j.still.2013.01.012 Liu, Z., Wang, Yanhui, Yu, P., Tian, A., Wang, Yarui, Xiong, W., Xu, L., 2018. Spatial Pattern and Temporal Stability of Root - Zone Soil Moisture during Growing Season on a Larch Plantation Hillslope in No rthwest China. Forests 9, 68. https://doi.org/10.3390/f9020068 Lobell, D.B., Roberts, M.J., Schlenker, W., Braun, N., Little, B.B., Rejesus, R.M., Hammer, G.L., 2014. Greater Sensitivity to Drought Accompanies Maize Yield Increase in the U.S. Midwest. Scie nce. 344, 516 519. https://doi.org/10.1126/science.1251423 Lotter, D.W., Seidel, R., Liebhardt, W., 2003. The performance of organic and conventional cropping systems in an extreme climate year. Am. J. Altern. Agric. 18, 146 154. https://doi.org/10.1079/AJ AA200345 Lu, Y. C., Teasdale, J.R., Huang, W. - Y., 2003. An Economic and Environmental Tradeoff Analysis of Sustainable Agriculture Cropping Systems. J. Sustain. Agric. 22, 25 41. https://doi.org/10.1300/J064v22n03_05 Luttikholt, L.W.M., 2007. Principles of organic agriculture as formulated by the International 185 Federation of Organic Agriculture Movements. NJAS - Wageningen J. Life Sci. 54, 347 360. https://doi.org/10.1016/S1573 - 5214(07)80008 - X Mäder, P., Fließbach, A., Dubois, D., Gunst, L., Fried, P., Niggl i, U., 2002. Soil fertility and biodiversity in organic farming. Science . 296, 1694 1697. https://doi.org/10.1126/science.1071148 Mafongoya, P., Rusinamhodzi, L., Siziba, S., Thierfelder, C., Mvumi, B.M., Nhau, B., Hove, L., Chivenge, P., 2016. Maize produ ctivity and profitability in Conservation Agriculture systems across agro - ecological regions in Zimbabwe: A review of knowledge and practice. Agric. Ecosyst. Environ. 220, 211 225. https://doi.org/10.1016/j.agee.2016.01.017 Malik, W., Dechmi, F., 2019. DSSAT modelling for best irrigation management practices assessment under Mediterranean conditions. Agric. Water Manag. 216, 27 43. https://doi.org/10.1016/j.agwat.2019.01.017 Mallya, G., Zhao, L., Song, X., Niyogi, D., Govindaraju, R., 2013. 2012 Mi dwest Drought in the United States. J. Hydrol. Eng. 18, 737 745. https://doi.org/10.1061/(ASCE)HE.1943 - 5584.0000786 Martens, J.R.T., Entz, M.H., Wonneck, M.D., 2015. Review: Redesigning Canadian prairie cropping systems for profitability, sustainability, a nd resilience. Can. J. Plant Sci. 95, 1049 1072. https://doi.org/10.4141/cjps - 2014 - 173 Martin, G., Magne, M.A., 2015. Agricultural diversity to increase adaptive capacity and reduce vulnerability of livestock systems against weather variability - A farm - scale simulation study. Agric. Ecosyst. Environ. 199, 301 311. https://doi.org/10.1016/j.agee.2014.10.006 Martinez - Feria, R., Nichols, V., Basso, B., Archontoulis, S., 2019. Can multi - strategy management stabilize nitrate leaching under increasing rai nfall? Environ. Res. Lett. 14, 124079. https://doi.org/10.1088/1748 - 9326/ab5ca8 Martínez - Fernández, J., González - Zamora, A., Sánchez, N., Gumuzzio, A., 2015. A soil water based index as a suitable agricultural drought indicator. J. Hydrol. 522, 265 273. ht tps://doi.org/10.1016/j.jhydrol.2014.12.051 Mase, A.S., Gramig, B.M., Prokopy, L.S., 2017. Climate change beliefs, risk perceptions, and adaptation behavior among Midwestern U.S. crop farmers. Clim. Risk Manag. 15, 8 17. https://doi.org/10.1016/j.crm.2016. 11.004 Mausch, K., Orr, A., Miller, B.P., 2017. Targeting resilience and profitability in African smallholder agriculture: Insights from ICRISAT - led research programs. FACETS 2, 545 558. https://doi.org/10.1139/facets - 2017 - 0029 Mbow, C., Smith, P., Skole, D., Duguma, L., Bustamante, M., 2014. Achieving mitigation and adaptation to climate change through sustainable agroforestry practices in Africa. Curr. Opin. Environ. Sustain. 6, 8 14. https://doi.org/10.1016/j.cosust.2013.09.002 186 McConnell, M., Burger, L.W ., 2011. Precision conservation: A geospatial decision support tool for optimizing conservation and profitability in agricultural landscapes. J. Soil Water Conserv. 66, 347 354. https://doi.org/10.2489/jswc.66.6.347 McLaughlin, D.L., Cohen, M.J., 2013. Rea lizing ecosystem services: wetland hydrologic function along a gradient of ecosystem condition. Ecol. Appl. 23, 1619 1631. https://doi.org/10.1890/12 - 1489.1 Mcvicar, T.R., Roderick, M.L., Donohue, R.J., Van Niel, T.G., 2012. Less bluster ahead? ecohydrolog ical implications of global trends of terrestrial near - surface wind speeds. Ecohydrology 5, 381 388. https://doi.org/10.1002/eco.1298 MDEQ, 2020. Wellogic System (periodically updated) [WWW Document]. Michigan Dep. Environ. Qual. URL https://secure1.state. mi.us/wellogic/ (accessed 9.20.20). MEA, 2005. Millennium Ecosystem Assessment - Ecosystems and Human Well - being: Synthesis. Island Press, Washington, DC. https://doi.org/10.5822/978 - 1 - 61091 - 484 - 0_1 Meier, M.S., Stoessel, F., Jungbluth, N., Juraske, R., Scha der, C., Stolze, M., 2015. Environmental impacts of organic and conventional agricultural products - Are the differences captured by life cycle assessment? J. Environ. Manage. 149, 193 208. https://doi.org/10.1016/j.jenvman.2014.10.006 Meinke, H., Nelson, R., Kokic, P., Stone, R., Selvaraju, R., Baethgen, W., 2006. Actionable climate knowledge: From analysis to synthesis. Clim. Res. 33, 101 110. https://doi.org/10.3354/cr033101 Mengist, W., Soromessa, T., Feyisa, G.L., 2020. A global view of regulatory ecos ystem services: existed knowledge, trends, and research gaps. Ecol. Process. 9, 40. https://doi.org/10.1186/s13717 - 020 - 00241 - w Meuwissen, M.P.M., Feindt, P.H., Spiegel, A., Termeer, C.J.A.M., Mathijs, E., Mey, Y. de, Finger, R., Balmann, A., Wauters, E., U H., Nicholas - Davies, P., Hansson, H., Paas, W., Slijper, T., Coopmans, I., Vroege, W., Ciechomska, A., Accatino, F., Kopainsky, B., Poortvliet, P.M., Candel, J.J.L., Maye, D., Severini, S., Senni, S., Soria no, B., Lagerkvist, C. - J., Peneva, M., Gavrilescu, C., Reidsma, P., 2019. A framework to assess the resilience of farming systems. Agric. Syst. 176, 102656. https://doi.org/10.1016/j.agsy.2019.102656 Michler, J.D., Baylis, K., Arends - kuenning, M., Mazvimav i, K., 2019. Conservation agriculture and climate resilience. J. Environ. Econ. Manage. 93, 148 169. https://doi.org/10.1016/j.jeem.2018.11.008 Organic Farming. J . Sustain. Agric. 22, 81 97. Milliken, G., Johnson, D., 2001. Analysis of messy data, Volume III: Analysis of Covariance, 1st editio. ed. Chapman and Hall/CRC. 187 Milliken, G. , Johnson, D., 2009. Analysis of messy data, Volume I: Designed Experiments, 2nd edi tio. ed. Chapman and Hall/CRC. Minoli, S., Müller, C., Elliott, J., Ruane, A.C., Jägermeyr, J., Zabel, F., Dury, M., Folberth, C., François, L., Hank, T., Jacquemin, I., Liu, W., Olin, S., Pugh, T.A.M., 2019. Global Response Patterns of Major Rainfed Crops to Adaptation by Maintaining Current 1480. https://doi.org/10.1029/2018EF001130 Moebius - Clune, B.N., van Es, H.M., Idowu, O.J., Schindelbeck, R.R., Moebius - Clune, D.J., Wolfe, D.W., Abawi, G.S., Thies , J.E., Gugino, B.K., Lucey, R., 2008. Long - Term Effects of Harvesting Maize Stover and Tillage on Soil Quality. Soil Sci. Soc. Am. J. 72, 960 969. https://doi.org/10.2136/sssaj2007.0248 Montoya, J.M., Raffaelli, D., 2010. Climate change, biotic interactions and ecosystem services. Philos. Trans. R. Soc. B Biol. Sci. 365, 2013 2018. https://doi.org/10.1098/rstb.2010.0114 Moorhead, J.E., Gowda, P.H., Singh, V.P., Porter, D.O., Marek, T.H., Howell, T.A., Stewart, B.A., 2015. Identifying and Evaluati ng a Suitable Index for Agricultural Drought Monitoring in the Texas High Plains. J. Am. Water Resour. Assoc. 51, 807 820. https://doi.org/10.1111/jawr.12275 Morton, L.W., Hobbs, J., Arbuckle, J.G., Loy, A., 2015. Upper Midwest Climate Variations: Farmer R esponses to Excess Water Risks. J. Environ. Qual. 44, 810 822. https://doi.org/10.2134/jeq2014.08.0352 Morton, L.W., McGuire, J.M., Cast, A.D., 2017. A good farmer pays attention to the weather. Clim. Risk Manag. 15, 18 31. https://doi.org/10.1016/j.crm.20 16.09.002 Munang, R., Thiaw, I., Alverson, K., Liu, J., Han, Z., 2013. The role of ecosystem services in climate change adaptation and disaster risk reduction. Curr. Opin. Environ. Sustain. 5, 47 52. https://doi.org/10.1016/j.cosust.2013.02.002 Murthy, V.R .K., 2004. Crop Growth Modeling and Its Applications in Agricultural Meteorology. Satell. Remote Sens. GIS Appl. Agric. Meteorol. 235 261. Myers, S.S., Smith, M.R., Guth, S., Golden, C.D., Vaitla, B., Mueller, N.D., Dangour, A.D., Huybers, P., 2017. Climat e Change and Global Food Systems: Potential Impacts on Food Security and Undernutrition. Annu. Rev. Public Health 38, 259 277. https://doi.org/10.1146/annurev - publhealth - 031816 - 044356 Narasimhan, B., Srinivasan, R., 2005. Development and evaluation of Soil Moisture Deficit Index ( SMDI ) and Evapotranspiration Deficit Index ( ETDI ) for agricultural drought monitoring. Agric. For. Meteorol. 133, 69 88. https://doi.org/10.1016/j.agrformet.2005.07.012 NCEI, 2019. Summary of monthly normals 1981 2010. Gull Lake Biological Station (Network 188 ID: GHCND:USC00203504), MI, USA. National Centers for Environmental Information, Asheville, NC, USA. NED USGS , 2006. National Elevation Dataset. Reston, Virginia: U.S. Geological Survey. http://ned.usgs.gov/ Ned/about.asp NHD USGS , 2010. USGS: National Hydrography Dataset. Reston, Virginia: U.S. Geological Survey. http://nhd.usgs.gov/index.html Nelson, A., Casler, G., Walker, A., 1978. Making Farm Decisions in a Risky World: A Guidebook. Oregon State Universit y Extension Service, Oregon State University Extension Service. Nelson, A.G., 1997. Teaching Agricultural Producers to Consider Risk in Decision Making. West. Agric. Econ. Assoc. 1 16. Nemes, N., 2009. Comparative Analysis of Organic and Non - organic Farmin g Systems: A critical Assessment of Farm Profitability. Food Agric. Organ. 1 33. Nendel, C., Berg, M., Kersebaum, K.C., Mirschel, W., Specka, X., Wegehenkel, M., Wenkel, K.O., Wieland, R., 2011. The MONICA model: Testing predictability for crop growth, soi l moisture and nitrogen dynamics. Ecol. Modell. 222, 1614 1625. https://doi.org/10.1016/j.ecolmodel.2011.02.018 Ngwira, A.R., Thierfelder, C., Lambert, D.M., 2013. Conservation agriculture systems for Malawian smallholder farmers: long - term effects on crop productivity, profitability and soil quality. Renew. Agric. Food Syst. 28, 350 363. https://doi.org/10.1017/S1742170512000257 Nouri, M., Homaee, M., Bannayan, M., Hoogenboom, G., 2017. Towards shifting planting date as an adaptation practice for rainfed w heat response to climate change. Agric. Water Manag. 186, 108 119. https://doi.org/10.1016/j.agwat.2017.03.004 NRCS, 2020. Web Soil Survey [WWW Document]. United States Dep. Agric. URL http://websoilsurvey.sc.egov.usda.gov/ (accessed 8.28.20). Nunes, M.R., van Es, H.M., Schindelbeck, R., Ristow, A.J., Ryan, M., 2018. No - till and cropping system diversification improve soil health and crop yield. Geoderma 328, 30 43. https://doi.org/10.1016/j.geoderma.2018.04.031 ., 2015. The resilience, adaptation and transformation assessment framework: from theory to application. CSIRO, Australia. Farming Systems, in: Rainfed Farming Sys tems. Springer, Dordrecht, pp. 101 132. https://doi.org/10.1007/978 - 1 - 4020 - 9132 - 2 Odum, E.P., 1987. Reduced - input agriculture reduces nonpoint pollution. J. Soil Water Conserv. 189 42, 412 414. Oldfield, E.E., Bradford, M.A., Wood, S.A., 2019. Global meta - anal ysis of the relationship between soil organic matter and crop yields. SOIL 5, 15 32. https://doi.org/10.5194/soil - 5 - 15 - 2019 Ortiz - Bobea, A., Knippenberg, E., Chambers, R.G., 2018. Growing climatic sensitivity of U.S. agriculture linked to technological change and regional specialization. Sci. Adv. 4, eaat4343. https://doi.org/10.1126/sciadv.aat4343 Ortiz - Bobea, A., Wang, H., Carrillo, C.M., Ault, T.R., 2019. Un packing the climatic drivers of US agricultural yields. Environ. Res. Lett. 14, 064003. https://doi.org/10.1088/1748 - 9326/ab1e75 Otkin, J.A., Anderson, M.C., Hain, C., Svoboda, M., Johnson, D., Mueller, R., Tadesse, T., Wardlow, B., Brown, J., 2016. Assess ing the evolution of soil moisture and vegetation conditions during the 2012 United States flash drought. Agric. For. Meteorol. 218 219, 230 242. https://doi.org/10.1016/j.agrformet.2015.12.065 Otkin, J.A., Svoboda, M., Hunt, E.D., Ford, T.W., Anderson, M. C., Hain, C., Basara, J.B., 2018. Flash droughts: A review and assessment of the challenges imposed by rapid - onset droughts in the United States. Bull. Am. Meteorol. Soc. 99, 911 919. https://doi.org/10.1175/BAMS - D - 17 - 0149.1 Owens, L., 1994. Impacts of soil N management of the quality of surface and subsurface water, in: Lal, R., Stewart, B.. (Eds.), Soil Processes and Water Quality. Lewis Publishers, Boca Raton, FL, pp. 137 162. Palm, C., Blanco - Canqui, H., DeClerck, F., Gatere, L., Grace, P., 2014. Con servation agriculture and ecosystem services: An overview. Agric. Ecosyst. Environ. 187, 87 105. https://doi.org/10.1016/j.agee.2013.10.010 Palmer, W.C., 1965. Meteorlogical Drought. U.S. Weather Bur. Res. Pap. No. 45. Parton, W.J., 1996. The CENTURY model , in: Powlson, D.S., Smith, P., Smith, J.. (Eds.), Evaluation of Soil Organic Matter Models. Springer - Verlag Berlin Heidelherg, pp. 283 291. https://doi.org/10.1007/978 - 3 - 642 - 61094 - 3_23 Peel, M.C., Finlayson, B.L., McMahon, T.A., 2007. Updated world map of the Köppen - Geiger climate classification. Hydrol. Earth Syst. Sci. 11, 1633 1644. https://doi.org/10.5194/hess - 11 - 1633 - 2007 Peng, B., Guan, K., Tang, J., Ainsworth, E.A., Asseng, S., Bernacchi, C.J., Cooper, M., Delucia, E.H., Elliott, J.W., Ewert, F., Gr ant, R.F., Gustafson, D.I., Hammer, G.L., Jin, Z., Jones, J.W., Kimm, H., Lawrence, D.M., Li, Y., Lombardozzi, D.L., Marshall - Colon, A., Messina, C.D., Ort, D.R., Schnable, J.C., Vallejos, C.E., Wu, A., Yin, X., Zhou, W., 190 2020. Towards a multiscale crop mo delling framework for climate change adaptation assessment. Nat. Plants 6, 338 348. https://doi.org/10.1038/s41477 - 020 - 0625 - 3 Perego, A., Rocca, A., Cattivelli, V., Tabaglio, V., Fiorini, A., Barbieri, S., Schillaci, C., Chiodini, M.E., Brenna, S., Acutis, M., 2019. Agro - environmental aspects of - year experience on 20 farms in the Po valley (Northern Italy). Agric. Syst. 168, 73 87. https://doi.org/10.1016/j.agsy.2018.10.008 Peters, A.J., Walter - Shea, E.A., Ji, L., Viña, A., Hayes, M., Svoboda, M.D., 2002. Drought monitoring with NDVI - based Standardized Vegetation Index. Photogramm. Eng. Remote Sensing 68, 71 75. Peterson, C.A., Eviner, V.T., Gaudin, A.C.M., 2018. Ways forward for resilience rese arch in agroecosystems. Agric. Syst. 162, 19 27. https://doi.org/10.1016/j.agsy.2018.01.011 Pham, L. Van, Smith, C., 2014. Drivers of agricultural sustainability in developing countries: a review. Environ. Syst. Decis. 34, 326 341. https://doi.org/10.1007/ s10669 - 014 - 9494 - 5 Pickering, N.B., Hansen, J.W., Jones, J.W., Wells, C.M., Chan, V.K., Godwin, D.C., 1994. WeatherMan: A Utility for Managing and Generating Daily Weather Data. Agron. J. 86, 332 337. https://doi.org/10.2134/agronj1994.00021962008600020023x Pimentel, D., Hepperly, P., Hanson, J., Douds, D., Seidel, R., 2005. Environmental, energetic and economic comparison of organic and conventional farming systems. Bioscience 55, 573 582. https://doi.org/10.1007/978 - 94 - 007 - 7796 - 5_6 Pittelkow, C M., Liang, X., Linquist, B.A., Van Groenigen, K., Lee, J., Lundy, M.E., Van Gestel, N., Six, J., Venterea, R.T., Van Kessel, C., 2015 a . Productivity limits and potentials of the principles of conservation agriculture. Nature 517, 365 368. https://doi.org/10.1038/natu re13809 Pittelkow, C. M., Linquist, B.A., Lundy, M.E., Liang, X., Van Groenigen, K.., Lee, J., Van Gestel, N., Six, J., Venterea, R.T., Van Kessel, C., 2015 b . When does no - A global meta - analysis. F. Crop. Res. 183, 156 168. https://doi.or g/10.1016/j.fcr.2015.07.020 Plastina, A., Liu, F., Miguez, F., Carlson, S., 2020. Cover crops use in Midwestern US agriculture: perceived benefits and net returns. Renew. Agric. Food Syst. 35, 38 48. https://doi.org/10.1017/S1742170518000194 Porter, P.M., Lauer, J.G., Huggins, D.R., Oplinger, E.S., Crookston, R.K., 1998. Assessing Spatial and Temporal Variability of Corn and Soybean Yields. J. Prod. Agric. 11, 359 363. https://doi.org/10.2134/jpa1998.0359 Power, A.G., 2010. Ecosystem services and agricultur e: tradeoffs and synergies. Philos. Trans. R. Soc. B Biol. Sci. 365, 2959 2971. https://doi.org/10.1098/rstb.2010.0143 191 Powlson, D.S., Stirling, C.M., Jat, M.L., Gerard, B.G., Palm, C.A., Sanchez, P.A., Cassman, K.G., 2014. Limited potential of no - till agri culture for climate change mitigation. Nat. Clim. Chang. 4, 678 683. https://doi.org/10.1038/nclimate2292 Pradhan, P., Fischer, G., van Velthuizen, H., Reusser, D.E., Kropp, J.P., 2015. Closing Yield Gaps: How Sustainable Can We Be? PLoS One 10, e0129487. https://doi.org/10.1371/journal.pone.0129487 Priestley, C.H.B., Taylor, R.J., 1972. On the Assessment of Surface Heat Flux and Evaporation Using Large - Scale Parameters. Mon. Weather Rev. 100, 81 92. https://doi.org/10.1175/1520 - 0493(1972)100<0081:otaosh>2. 3.co;2 Prudencio, L., Null, S.E., 2018. Stormwater management and ecosystem services: a review. Environ. Res. Lett. 13, 033002. https://doi.org/10.1088/1748 - 9326/aaa81a Pryor, S.C., Scavia, D., Downer, C., Gaden, M., Iverson, L., Nordstrom, R., Patz, J., Roberson, G.P., 2014. Climate Change Impacts in the United States Ch. 18: Midwest, in: Melillo, J., Richmond, T., Yohe, G. (Eds.), Climate Change Impacts in the United States: The Third National Climate Assessment. pp. 418 440. https://doi.org/10.7930/ J0J1012N.On Purcell, L.C., Sinclair, T.R., McNew, R.W., 2003. Drought Avoidance Assessment for Summer Annual Crops Using Long - Term Weather Data. Agron. J. 95, 1566 1576. https://doi.org/10.2134/agronj2003.1566 Qin, W., Hu, C., Oenema, O., 2015. Soil mulchi ng significantly enhances yields and water and nitrogen use efficiencies of maize and wheat: A meta - analysis. Sci. Rep. 5, 1 13. https://doi.org/10.1038/srep16210 assessi perspectives. J. Appl. Ecol. 53, 677 687. https://doi.org/10.1111/1365 - 2664.12550 Rao, C.S., Lal, R., Prasad, J.V.N.S., Gopinath, K.A., Singh, R., Jakkula, V.S., Sahrawat, K.L., Venkate swarlu, B., Sikka, A.K., Virmani, S.M., 2015. Potential and Challenges of Rainfed Farming in India, in: Advances in Agronomy. pp. 113 181. https://doi.org/10.1016/bs.agron.2015.05.004 Ratliff, L.F., Ritchie, J.T., Cassel, D.K., 1983. Field - Measured Limits of Soil Water Availability as Related to Laboratory - Measured Properties. Soil Sci. Soc. Am. J. 47, 770 775. https://doi.org/10.2136/sssaj1983.03615995004700040032x Ray, D.K., Gerber, J.S., MacDonald, G.K., West, P.C., 2015. Climate variation explains a thi rd of global crop yield variability. Nat. Commun. 6, 5989. https://doi.org/10.1038/ncomms6989 Refsgaard, J.C., Storm, B., 1995. MIKE SHE, in: Miller, P.C. (Ed.), Computer Models of Catchment Hydrology. Water Resources Publications, Colarado, USA, pp. 809 846. 192 Reitz, M., Sanford, W.E., Senay, G.B., Cazenas, J., 2017. Annual Estimates of Recharge, Quick - Flow Runoff, and Evapotranspiration for the Contiguous U.S. Using Empirical Regression Equations. JAWRA J. Am. Water Resour. Assoc. 53, 961 983. https:// doi.org/10.1111/1752 - 1688.12546 Rey, D., Holman, I.P., Knox, J.W., 2017. Developing drought resilience in irrigated agriculture in the face of increasing water scarcity. Reg. Environ. Chang. 17, 1527 1540. https://doi.org/10.1007/s10113 - 017 - 1116 - 6 Reynolds , S.G., 1970. The gravimetric method of soil moisture determination Part I A study of equipment, and methodological problems. J. Hydrol. 11, 258 273. https://doi.org/10.1016/0022 - 1694(70)90066 - 1 Rigolot, C., de Voil, P., Douxchamps, S., Prestwidge, D., Van Wijk, M., Thornton, P.K., Rodriguez, D., Henderson, B., Medina, D., Herrero, M., 2017. Interactions between intervention packages, climatic risk, climate change and food security in mixed crop livestock systems in Burkina Faso. Agric. Syst. 151, 217 224. https://doi.org/10.1016/j.agsy.2015.12.017 Rippey, B.R., 2015. The U.S. drought of 2012. Weather Clim. Extrem. 10, 57 64. https://doi.org/10.1016/j.wace.2015.10.004 Ritchie, J.T., 1998. Soil water balance and plant water stress, in: Tsuji, G.Y., Hoogenboom , G., Thornton, P. (Eds.), Understanding Options for Agricultural Production. Kluwer Academic Publishers, Dordrecht, pp. 41 54. https://doi.org/10.1007/978 - 94 - 017 - 3624 - 4_3 Robertson, G., Hamilton, S., 2015. Long - Term Ecological Research at the Kellogg Biol ogical Station LTER Site, in: Hamilton, S., Doll, J., Robertson, G. (Eds.), The Ecology of Agricultural Landscapes: Long - Term Research on the Path to Sustainability. Oxford University Press, New York, USA, pp. 1 32. Robertson, G.P., Gross, K.L., Hamilton, S.K., Landis, D.A., Schmidt, T.M., Snapp, S.S., Swinton, S.M., 2014. Farming for ecosystem services: An ecological approach to production agriculture. Bioscience 64, 404 415. https://doi.org/10.1093/biosci/biu037 Robertson, G.P., Swinton, S.M., 2005. Recon ciling Agricultural Productivity and Environmental Integrity: A Grand Challenge for Agriculture. Front. Ecol. Environ. 3, 38. https://doi.org/10.2307/3868443 Robinson, D.A., Hopmans, J.W., Filipovic, V., van der Ploeg, M., Lebron, I., Jones, S.B., Reinsch, S., Jarvis, N., Tuller, M., 2019. Global environmental changes impact soil hydraulic functions through biophysical feedbacks. Glob. Chang. Biol. 25, 1895 1904. https://doi.org/10.1111/gcb.14626 Rockström, J., Karlberg, L., Wani, S.P., Barron, J., Hatibu, N., Oweis, T., Bruggeman, A., Farahani, J., Qiang, Z., 2010. Managing water in rainfed agriculture The need for a paradigm shift. Agric. Water Manag. 97, 543 550. 193 https://doi.org/10.1016/j.agwat.2009.09.009 Roesch - McNally, G.E., Arbuckle, J.G., Tyndall, J. C., 2018. Barriers to implementing climate resilient agricultural strategies: The case of crop diversification in the U.S. Corn Belt. Glob. Environ. Chang. 48, 206 215. https://doi.org/10.1016/j.gloenvcha.2017.12.002 Rojas - Downing, M.M., Nejadhashemi, A.P., Harrigan, T., Woznicki, S.A., 2017. Climate change and livestock: Impacts, adaptation, and mitigation. Clim. Risk Manag. 16, 145 163. https://doi.org/10.1016/j.crm.2017.02.001 Röös, E., Mie, A., Wivstad, M., Salomon, E., Johansson, B., Gunnarsson, S., Wallenbeck, A., Hoffmann, R., Nilsson, U., Sundberg, C., Watson, C.A., 2018. Risks and opportunities of increasing yields in organic farming. A review. Agron. Sustain. Dev. 38, 14. https://doi.org/10.1007/s13593 - 018 - 0489 - 3 Rosegrant, M., Ca i, X., Cline, S., Nakagawa, N., 2002. The Role of Rainfed Agriculture in the Future of Global Food Production (No. 90). International Food Policy Research Institute, Washington, D.C, USA. Rosenzweig, C., Elliott, J., Deryng, D., Ruane, A.C., Müller, C., Ar neth, A., Boote, K.J., Folberth, C., Glotter, M., Khabarov, N., Neumann, K., Piontek, F., Pugh, T.A.M., Schmid, E., Stehfest, E., Yang, H., Jones, J.W., 2014. Assessing agricultural risks of climate change in the 21st century in a global gridded crop model intercomparison. Proc. Natl. Acad. Sci. 111, 3268 3273. https://doi.org/10.1073/pnas.1222463110 Rosenzweig, C., Liverman, D., 1992. Predicted effects of climate change on agriculture: A comparison of temperate and tropical regions, in: Majumdar, S. (Ed.), Global Climate Change: Implications, Challenges, and Mitigation Measures. The Pennsylvania Academy of Sciences. Rossato, L., Alvalá, R.C. dos S., Marengo, J.A., Zeri, M., Cunha, A.P.M. do A., Pires, L.B.M., Barbosa, H.A., 2017. Impact of Soil Moisture on Crop Yields over Brazilian Semiarid. Front. Environ. Sci. 5, 1 16. https://doi.org/10.3389/fenvs.2017.00073 Rost, S., Gerten, D., Hoff, H., Lucht, W., Falkenmark, M., Rockström, J., 2009. Global potential to increase crop production through water managemen t in rainfed agriculture. Environ. Res. Lett. 4, 044002. https://doi.org/10.1088/1748 - 9326/4/4/044002 Rötter, R.P., Appiah, M., Fichtler, E., Kersebaum, K.C., Trnka, M., Hoffmann, M.., 2018. Linking modelling and experimentation to better capture crop impa cts of agroclimatic extremes A review. F. Crop. Res. 221, 142 156. https://doi.org/10.1016/j.fcr.2018.02.023 Roy, P.C., Guber, A., Abouali, M., Nejadhashemi, A.P., Deb, K., Smucker, A.J.M., 2019. Crop yield simulation optimization using precision irrigatio n and subsurface water retention technology. Environ. Model. Softw. 119, 433 444. https://doi.org/10.1016/j.envsoft.2019.07.006 194 Rurinda, J., Wijk, M.T.V.A.N., Mapfumo, P., Descheemaeker, K., Supit, I., Giller, K.E., 2015. Climate change and maize yield in Glob. Chang. Biol. 21, 4588 4601. https://doi.org/10.1111/gcb.13061 Rusinamhodzi, L., Corbeels, M., Van Wijk, M., Rufino, M., Nyamangara, J., Giller, K., 2011. A meta - analysis of long - term effects of conserva tion agriculture on maize grain yield under rain - fed conditions. Agron. Sustain. Dev. 31, 657 673. https://doi.org/10.1007/s13593 - 011 - 0040 - 2 Sadok, W., Sinclair, T.R., 2011. Crops Yield Increase Under Water - of Recent Physiologic al Advances for Soybean Genetic Improvement, Advances in Agronomy. Elsevier Inc. https://doi.org/10.1016/B978 - 0 - 12 - 386473 - 4.00007 - 5 Sain, G., Loboguerrero, A.M., Corner - Dolloff, C., Lizarazo, M., Nowak, A., Martínez - Barón, D., Andrieu, N., 2017. Costs and benefits of climate - smart agriculture: The case of the Dry Corridor in Guatemala. Agric. Syst. 151, 163 173. https://doi.org/10.1016/j.agsy.2016.05.004 Salmerón, M., Cavero, J., Isla, R., Porter, C.H., Jones, J.W., Boote, K.J., 2014. DSSAT Nitrogen Cycle S imulation of Cover Crop - Maize Rotations under Irrigated Mediterranean Conditions. Agron. J. 106, 1283 1296. https://doi.org/10.2134/agronj13.0560 Sampath, P.V., Liao, H.S., Curtis, Z.K., Doran, P.J., Herbert, M.E., May, C.A., Li, S.G., 2015. Understanding the Groundwater Hydrology of a Geographically - Isolated Prairie Fen: Implications for Conservation. PLoS One 10, e0140430. https://doi.org/10.1371/journal.pone.0140430 Sampath, P.V., Liao, H.S., Curtis, Z.K., Herbert, M.E., Doran, P.J., May, C.A., Landis, D .A., Li, S.G., 2016. Understanding fen hydrology across multiple scales. Hydrol. Process. 30, 3390 3407. https://doi.org/10.1002/hyp.10865 Sapkota, T.B., Majumdar, K., Jat, M.L., Kumar, A., Bishnoi, D.K., McDonald, A.J., Pampolino, M., 2014. Precision nutr ient management in conservation agriculture based wheat production of Northwest India: Profitability, nutrient use efficiency and environmental footprint. F. Crop. Res. 155, 233 244. https://doi.org/10.1016/j.fcr.2013.09.001 Saseendran, S.A., Trout, T.J., Ahuja, L.R., Ma, L., McMaster, G.S., Nielsen, D.C., Andales, A.A., Chávez, J.L., Ham, J., 2015. Quantifying crop water stress factors from soil water measurements in a limited irrigation experiment. Agric. Syst. 137, 191 205. https://doi.org/10.1016/j.agsy .2014.11.005 Savage, M.J., Ritchie, J.T., Bland, W.L., Dugas, W.A., 1996. Lower limit of soil water availability. Agron. J. 88, 644 651. https://doi.org/10.2134/agronj1996.00021962008800040024x Scanlon, B.R., Faunt, C.C., Longuevergne, L., Reedy, R.C., Alley, W.M., McGuire, V.L., McMahon, P.B., 2012. Groundwater depletion and sustainability of irrigation in the US High Plains and Central Valley. Proc. Natl. Acad. Sci. 109, 9320 9325. 195 https://doi.org/10.1073/pnas.1200311109 Schaetzl, R.J., Darden, J .T., Brandt, D, S., 2009. Michigan Geography and Geology. Pearson Custom Publishing, New York. Scherr, S.J., Shames, S., Friedman, R., 2012. From climate - smart agriculture to climate - smart landscapes. Agric. Food Secur. 1, 12. https://doi.org/10.1186/2048 - 7010 - 1 - 12 Schlenker, W., Hanemann, W.M., Fisher, A.C., 2006. The Impact of Global Warming on U.S. Agriculture: An Econometric Analysis of Optimal Growing Conditions. Rev. Econ. Stat. 88, 113 125. https://doi.org/10.1162/rest.2006.88.1.113 Scialabba, N.E.H. , Müller - Lindenlauf, M., 2010. Organic agriculture and climate change. Renew. Agric. Food Syst. 25, 158 169. https://doi.org/10.1017/S1742170510000116 SCS, 1985. National Engineering Handbook Section 4: Hydrology. Soil Conservation Service, United States Department of Agriculture, Washington, DC. Senapati, A.K., 2020. Evaluation of risk preferences and coping strategies to manage with various agricultural risks: evidence from India. Heliyon 6, e03503. https://doi.org/10.1016/j.heliyon.2020.e03503 Se o, S.N., 2010. Is an integrated farm more resilient against climate change? A micro - econometric analysis of portfolio diversification in African agriculture. Food Policy 35, 32 40. https://doi.org/10.1016/j.foodpol.2009.06.004 Serfilippi, E., Ramnath, G., 2018. Resilience Measurement and Conceptual Frameworks: a Review of the Literature. Ann. Public Coop. Econ. 89, 645 664. https://doi.org/10.1111/apce.12202 Serna - Chavez, H.M., Schulp, C.J.E., van Bodegom, P.M., Bouten, W., Verburg, P.H., Davidson, M.D., 20 14. A quantitative framework for assessing spatial flows of ecosystem services. Ecol. Indic. 39, 24 33. https://doi.org/10.1016/j.ecolind.2013.11.024 Sharma, T.C., 1997. A drought frequency formula. Hydrol. Sci. J. 42, 803 814. https://doi.org/10.1080/0262 6669709492080 Sharpley, A.N., Williams, J.R., 1990. EPIC - The erosion/productivity impact calculator: 1. Model Documentation. U.S. Dep. Agric. Tech. Bull. 235. Shelia, V., Sharda, V., Hansen, J., Porter, C., Zhang, M., Aggarwal, P., Hoogenboom, G., 2015. CC AFS Regional Agricultural Forecasting Toolbox (CRAFT): software for forecasting of crop production, risk analysis and climate change impact studies, in: 2015 ASABE International Meeting. American Society of Agricultural and Biological Engineers, pp. 1094 1 116. https://doi.org/10.13031/aim.20152182505 Siad, S.M., Iacobellis, V., Zdruli, P., Gioia, A., Stavi, I., Hoogenboom, G., 2019. A review of coupled hydrologic and crop growth models. Agric. Water Manag. 224, 105746. 196 https://doi.org/10.1016/j.agwat.2019.1 05746 Sieber, J., Purkey, D., 2015. WEAP User Guide, Water Evaluation And Planning System. Stockholm Environment Institute, Somerville, MA, USA. Siebert, S., Burke, J., Faures, J.M., Frenken, K., Hoogeveen, J., Döll, P., Portmann, F.T., 2010. Groundwater u se for irrigation a global inventory. Hydrol. Earth Syst. Sci. 14, 1863 1880. https://doi.org/10.5194/hess - 14 - 1863 - 2010 Validation. Trans. ASABE 55, 1261 1274. https:/ /doi.org/10.13031/2013.42239 Sivakumar, M.V., Motha, R., Wilhite, D., Wood, D., 2011. Agricultural Drought Indices, AGM - 11, WM. ed, Proceedings of the WMO/UNISDR Expert Group Meeting on Agricultural Drought Indices, 2 - 4 June 2010, Murcia, Spain: World Mete orological Organization, Geneva, Switzerland, World Meteorological Organization, Geneva, Switzerland. Smidt, S.J., Haacker, E.M.K., Kendall, A.D., Deines, J.M., Pei, L., Cotterman, K.A., Li, H., Liu, X., Basso, B., Hyndman, D.W., 2016. Complex water manage ment in modern agriculture: Trends in the water - energy - food nexus over the High Plains Aquifer. Sci. Total Environ. 566 567, 988 1001. https://doi.org/10.1016/j.scitotenv.2016.05.127 Smit, B., Skinner, M.W., 2002. Adaptation options in agriculture to clima te change: a typology. Mitig. Adapt. Strateg. Glob. Chang. 7, 85 114. https://doi.org/10.1023/A:1015862228270 Smit, B., Wandel, J., 2006. Adaptation, adaptive capacity and vulnerability. Glob. Environ. Chang. 16, 282 292. https://doi.org/10.1016/j.gloenvcha.2006.03.008 Smith, P., Olesen, J.E., 2010. Synergies between the mitigation of, and adaptation to, climate change in agriculture. J. Agric. Sci. 148, 543 552. https://doi.org/10.1017/S0021859610000341 Spencer, G.D., Krut z, L.J., Falconer, L.L., Henry, W.B., Henry, C.G., Larson, E.J., Pringle, H.C., Bryant, C.J., Atwill, R.L., 2019. Irrigation Water Management Technologies for Furrow - Irrigated Corn that Decrease Water Use and Improve Yield and On - Farm Profitability. Crop. Forage Turfgrass Manag. 5, 180100. https://doi.org/10.2134/cftm2018.12.0100 Starks, P.J., Heathman, G.C., Jackson, T.J., Cosh, M.H., 2006. Temporal stability of soil moisture profile. J. Hydrol. 324, 400 411. https://doi.org/10.1016/j.jhydrol.2005.09.024 S tate of Michigan, 2006. Public Act 148 Groundwater inventory and map project (GWIM): Executive Summary. Michigan State University, East Lansing, MI, USA. Steel, R.G.D., Torrie, J.H., Dickey, D., 1980. Principles and Procedures of Statistics: A Biometrical Approach., Principles and procedures of statistics: a biometrical approach. New York: MacGraw - Hill. https://doi.org/10.2307/2530180 197 Stinner, B.R., House, G.J., 1987. Role of ecology in lower - input, sustainable agriculture: An introduction. Am. J. Altern. A gric. 2, 146 147. https://doi.org/10.1017/S0889189300009243 Stöckle, C.O., Donatelli, M., Nelson, R., 2003. CropSyst, a cropping systems simulation model. Eur. J. Agron. 18, 289 307. https://doi.org/10.1016/S1161 - 0301(02)00109 - 0 Suleiman, A.A., Ritchie, J. T., 2003. Modeling Soil Water Redistribution during Second - Stage Evaporation. Soil Sci. Soc. Am. J. 67, 377 386. https://doi.org/10.2136/sssaj2003.3770 Svoboda, M., LeComte, D., Hayes, M., Heim, R., Gleason, K., Angel, J., Rippey, B., Tinker, R., Palecki, M., Stooksbury, D., Miskus, D., Stephens, S., 2002. THE DROUGHT MONITOR. Bull. Am. Meteorol. Soc. 83, 1181 1190. https://doi.org/10.1175/1520 - 0477 - 83.8.1181 Sweet, S.K., Wolfe, D.W., DeGaetano, A., Benner, R., 2017. Anatomy of the 2016 drought in the Northeastern United States: Implications for agriculture and water resources in humid climates. Agric. For. Meteorol. 247, 571 581. https://doi.org/10.1016/j.agrformet.2017.08.024 Swift, M.J., Izac, A.M.N., van Noordwijk, M., 2004. Biodiversity and eco system services in agricultural landscapes are we asking the right questions? Agric. Ecosyst. Environ. 104, 113 134. https://doi.org/10.1016/j.agee.2004.01.013 Swinton, S.M., Jolejole - Foreman, M.C., Lupi, F., Ma, S., Zhang, W., Chen, H., 2015. Economic val ue of ecosystem services from agriculture, in: Hamilton, S., Doll, J., Robertson, G. (Eds.), Economic Value of Ecosystem Services from Agriculture. Oxford University Press, New York, USA, pp. 54 76. Swinton, S.M., Lupi, F., Robertson, G.P., Hamilton, S.K., 2007. Ecosystem services and agriculture: Cultivating agricultural ecosystems for diverse benefits. Ecol. Econ. 64, 245 252. https://doi.org/10.1016/j.ecolecon.2007.09.020 Syswerda, S.P., Basso, B., Hamilton, S.K., Tausig, J.B., Robertson, G.P., 2012. Lon g - term nitrate loss along an agricultural intensity gradient in the Upper Midwest USA. Agric. Ecosyst. Environ. 149, 10 19. https://doi.org/10.1016/j.agee.2011.12.007 Syswerda, S.P., Robertson, G.P., 2014. Ecosystem services along a management gradient in Michigan (USA) cropping systems. Agric. Ecosyst. Environ. 189, 28 35. https://doi.org/10.1016/j.agee.2014.03.006 Tack, J.B., Ubilava, D., 2015. Climate and agricultural risk: measuring the effect of ENSO on U.S. crop insurance. Agric. Econ. 46, 245 257. ht tps://doi.org/10.1111/agec.12154 Takahashi, B., Burnham, M., Terracina - Hartman, C., Sopchak, A.R., Selfa, T., 2016. Climate Change Perceptions of NY State Farmers: The Role of Risk Perceptions and Adaptive Capacity. Environ. Manage. 58, 946 957. https://do i.org/10.1007/s00267 - 016 - 0742 - y 198 Tancoigne, E., Barbier, M., Cointet, J. - P., Richard, G., 2014. The place of agricultural sciences in the literature on ecosystem services. Ecosyst. Serv. 10, 35 48. https://doi.org/10.1016/j.ecoser.2014.07.004 Teixeira, W.W. R., Battisti, R., Sentelhas, P.C., Moraes, M.F., Oliveira Junior, A., 2019. Soybean calibrated by cultivar maturity groups. J. Agron. Crop Sci. 205, 533 544. https://doi.org/10.1111/jac .12343 Tendall, D.M., Joerin, J., Kopainsky, B., Edwards, P., Shreck, A., Le, Q.B., Kruetli, P., Grant, Glob. Food Sec. 6, 17 23. https://doi.org/10.1016/j.gfs.2015 .08.001 Tey, Y.S., Brindal, M., 2015. Factors Influencing Farm Profitability, in: Lichtfouse, E. (Ed.), Sustainable Agriculture Reviews . Springer International Publishing, Cham, pp. 235 255. https://doi.org/10.1007/978 - 3 - 319 - 09132 - 7 Thierfelder, C., Chisui , J.L., Gama, M., Cheesman, S., Jere, Z.D., Trent Bunderson, W., Eash, N.S., Rusinamhodzi, L., 2013. Maize - based conservation agriculture systems in Malawi: Long - term trends in productivity. F. Crop. Res. 142, 47 57. https://doi.org/10.1016/j.fcr.2012.11.0 10 Thierfelder, C., Wall, P.C., 2009. Effects of conservation agriculture techniques on infiltration and soil water content in Zambia and Zimbabwe. Soil Tillage Res. 105, 217 227. https://doi.org/10.1016/j.still.2009.07.007 Thiessen, A.H., 1911. Precipitat ion avarages for large areas. Mon. Weather Rev. 39, 1082 1084. https://doi.org/10.1175/1520 - 0493(1911)39<1082b:PAFLA>2.0.CO;2 Thornton, P., Ericksen, P., Herrero, M., Challinor, A., 2014. Climate variability and vulnerability Glob. Chang. Biol. 20, 3313 3328. https://doi.org/10.1111/gcb.12581 Thorp, K.R., DeJonge, K.C., Kaleita, A.L., Batchelor, W.D., Paz, J.O., 2008. Methodology for the use of DSSAT models for precision agriculture decision support. Comput. Electron. Agric. 64 , 276 285. https://doi.org/10.1016/j.compag.2008.05.022 Tibesigwa, B., Visser, M., 2015. Small - Scale Subsistence Farming, Food Security, Climate Change and Adaptation in South Africa: Male - Female Headed Households and Urban - Rural Nexus, Economic Research S outhern Africa. Cape Town, South Africa. Tilman, D., Balzer, C., Hill, J., Befort, B.L., 2011. Global food demand and the sustainable intensification of agriculture. Proc. Natl. Acad. Sci. 108, 20260 20264. https://doi.org/10.1073/pnas.1116437108 Tittonell, P., 2014. Livelihood strategies, resilience and transformability in African agroecosystems. Agric. Syst. 126, 3 14. https://doi.org/10.1016/j.agsy.2013.10.010 199 Tobin, J., 1958. Liquidity Preference as Behavior Towards Risk. Rev. Econ. Stud. 25, 6 5. https://doi.org/10.2307/2296205 Toliver, D., 2010. Effects of No - Tillage on Crop Yields and Net Returns Across the United States. University of Tennessee. https://doi.org/https://trace.tennessee.edu/utk_gradthes/754 Tomasek, B.J., Williams, M.M., Davis, A.S., 2017. Changes in field workability and drought risk from projected climate change drive spatially variable risks in Illinois cropping systems. PLoS One 12, e0172301. https://doi.org/10.1371/journal.pone.0172301 Toth, A., Rendall, S., Reitsma, F., 20 food resilience. Urban Ecosyst. 19, 19 43. https://doi.org/10.1007/s11252 - 015 - 0489 - x Tow, P., Cooper, I., Partridge, I., Birch, C., Harrington, L., 2011. Principles of a systems approach to agri culture, in: Rainfed Farming Systems. Springer, Dordrecht. Troy, T.J., Kipgen, C., Pal, I., 2015. The impact of climate extremes and irrigation on US crop yields. Environ. Res. Lett. 10, 054013. https://doi.org/10.1088/1748 - 9326/10/5/054013 Tsakiris, G., Vangelis, H., 2005. Establishing a drought index incorporating evapotranspiration. Eur. Water 9, 3 11. Tsiafouli, M.A., Thébault, E., Sgardelis, S.P., de Ruiter, P.C., van der Putten, W.H., Birkho fer, K., Hemerik, L., de Vries, F.T., Bardgett, R.D., Brady, M.V., Bjornlund, L., Jørgensen, H.B., Christensen, S., Hertefeldt, T.D., Hotes, S., Gera Hol, W.H., Frouz, J., Liiri, M., , J., Wolters, V., Hedlund, K., 2015. Intensive agriculture reduces soil biodiversity across Europe. Glob. Chang. Biol. 21, 973 985. https://doi.org/10.1111/gcb.12752 Tukey, J., 1977. Exploratory Data Analysis, Volume 2. ed. Addison - Wesley Publishing Compa ny. Tuomisto, H.L., Hodge, I.D., Riordan, P., Macdonald, D.W., 2012. Does organic farming reduce - A meta - analysis of European research. J. Environ. Manage. 112, 309 320. https://doi.org/10.1016/j.jenvman.2012.08.018 Uematsu, H., Mi Ecol. Econ. 78, 55 62. https://doi.org/10.1016/j.ecolecon.2012.03.013 UNDP, 2013. Community based resilience assessment (COBRA) conceptual framework and methodology. United Natio ns Development Program, New York, USA. Urruty, N., Tailliez - lefebvre, D., Huyghe, C., Tailliez - lefebvre, D., 2016. Stability , robustness , vulnerability and resilience of agricultural systems . A review. Agron. Sustain. Dev. 36, 1 15. https://doi.org/10.1 007/s13593 - 015 - 0347 - 5 USDA, 2014. Census of Agriculture Highlights. Farms and Farmland. United States Department 200 of Agriculture, Washington, D.C, USA. USDA, 2019. Agricultural Prices. National Agricultural Statistics Service (NASS), United States Departmen t of Agriculture, Washington, DC. USDA, 2020. Commodity Costs and Returns [WWW Document]. URL https://www.ers.usda.gov/data - products/commodity - costs - and - returns/organic - costs - and - returns/ (accessed 6.10.20). Vachaud, G., Passerat de Silans, A. Balabanis, P ., Vauclin, M., 1985. Temporal Stability of Spatially Measured Soil Water Probability Density Function. Soil Sci. Soc. Am. J. 49, 822 828. Valkama, E., Kunypiyaeva, G., Zhapayev, R., Karabayev, M., Zhusupbekov, E., Perego, A., Schillaci, C., Sacco, D., Mor etti, B., Grignani, C., Acutis, M., 2020. Geoderma Can Geoderma 369, 114298. https://doi.org/10.1016/j.geoderma.2020.114298 Van Dam, J.C., Groenendijk, P., Hendriks, R.F.A. , Kroes, J.G., 2008. Advances of Modeling Water Flow in Variably Saturated Soils with SWAP. Vadose Zo. J. 7, 640 653. https://doi.org/10.2136/vzj2007.0060 Vanuytrecht, E., Raes, D., Steduto, P., Hsiao, T.C., Fereres, E., Heng, L.K., Garcia Vila, M., Mejias model. Environ. Model. Softw. 62, 351 360. https://doi.org/10.1016/j.envsoft.2014.08.005 Verhulst, N., Govaerts, B., Verachtert, E., Castellanos - Navarrete, A., Mezzalama, M., Wal l, P.C., Chocobar, A., Deckers, J., Sayre, K.D., 2010. Conservation agriculture, improving soil quality for sustainable production systems?, in: Lal, R., Stewart, B.. (Eds.), Advances in Soil Science: Food Security and Soil Quality. CRC Press, Boca Raton,F L, USA, pp. 137 208. https://doi.org/10.1201/EBK1439800577 Verhulst, N., Nelissen, V., Jespers, N., Haven, H., Sayre, K.D., Raes, D., Deckers, J., Govaerts, B., 2011. Soil water content, maize yield and its stability as affected by tillage and crop residue management in rainfed semi - arid highlands. Plant Soil 344, 73 85. https://doi.org/10.1007/s11104 - 011 - 0728 - 8 Vervoort, J.M., Thornton, P.K., Kristjanson, P., Förch, W., Ericksen, P.J., Kok, K., Ingram, J.S.I., Herrero, M., Palazzo, A., Helfgott, A.E.S., Wi lkinson, A., Havlík, P., Mason - - guided adaptive action on food security under climate change. Glob. Environ. Chang. 28, 383 394. https://doi.org/10.1016/j.gloenvcha.2014.03.001 Vigani, M., Kathage, J., 2019 . To Risk or Not to Risk? Risk Management and Farm Productivity. Am. J. Agric. Econ. 101, 1432 1454. https://doi.org/10.1093/ajae/aaz020 Vyn, T.J., Faber, J.G., Janovicek, K.J., Beauchamp, E.G., 2000. Cover crop effects on nitrogen 201 availability to corn fol lowing wheat. Agron. J. 92, 915 924. https://doi.org/10.2134/agronj2000.925915x Wada, Y., van Beek, L.P.H., van Kempen, C.M., Reckman, J.W.T.M., Vasak, S., Bierkens, M.F.P., 2010. Global depletion of groundwater resources. Geophys. Res. Lett. 37, L20402 . https://doi.org/10.1029/2010GL044571 Wagner, W., Pathe, C., Doubkova, M., Sabel, D., Bartsch, A., Hasenauer, S., Blöschl, G., Scipal, K., Martínez - fernández, J., Löw, A., 2008. Temporal Stability of Soil Moisture and Radar Backscatter Observed by the Advanced Synthetic Aperture Radar (ASAR). Sensors 8, 1174 1197. Wald, A., 1943. Tests of Statistical Hypotheses Concerning Several Parameters When the Number of Observations is Large. Trans. Am. Math. Soc. 54, 426 482. Walker, B., Holling, C.S., Carpenter, S.R., Kinzig, A., 2004. Resilience , Adaptability and Transformability in Soci al ecological Systems. Ecol. Soc. 9, 5. Wang, R., Bowling, L.C., Cherkauer, K.A., 2016. Estimation of the effects of climate variability on crop yield in the Midwest USA. Agric. For. Meteorol. 216, 141 156. https://doi.org/10.1016/j.agrformet.2015.10.001 Wani, S.P., Sreedevi, T.K., Rockström, J., Ramakrishna, Y.., 2009. Rainfed agriculture past trends and future prospects, in: Wani, S.P., Rockström, J., Oweis, T. (Eds.), Rainfed Agriculture: Unlocking the Potential. CAB International, Oxfordshire, UK, pp. 1 35. https://doi.org/10.1177/0019466220090204 Weaver, K.F., Morales, V.C., Dunn, S.L., Godde, K., Weaver, P., 2017. An Introduction to Statistical Analysis in Research: with Applications in the Biological and Life Sciences. John Wiley & Sons. Wei, Y., Wu , S., Jiang, C., Feng, X., 2021. Managing supply and demand of ecosystem services in dryland catchments. Curr. Opin. Environ. Sustain. 48, 10 16. https://doi.org/10.1016/j.cosust.2020.08.007 Wesley, J., 2005. Kalamazoo River Assessment. Michigan Department of Natural Resources, Fisheries Division, Special Report 35, Ann Arbor, MI, USA. Wheeler, T., von Braun, J., 2013. Climate Change Impacts on Global Food Security. Science. 341, 508 513. https://doi.org/10.1126/science.1239402 Williams, D.M., Blanco - Canqui , H., Francis, C.A., Galusha, T.D., 2017. Organic farming and soil physical properties: An assessment after 40 years. Agron. J. 109, 600 609. https://doi.org/10.2134/agronj2016.06.0372 Williams, J., Jones, C., Dyke , P.T . , 1984. A Modeling Approach to Deter mining the Relationship Between Erosion and Soil Productivity. Trans. ASAE 27, 0129 0144. https://doi.org/10.13031/2013.32748 202 Willmott, C.J., 1982. Some Comments on the Evaluation of Model Performance. Bull. Am. Meteorol. Soc. 63, 1309 1313. https://doi.or g/10.1175/1520 - 0477(1982)063<1309:SCOTEO>2.0.CO;2 WMO, 2017. WMO Guidelines on the Calculation of Climate Normals, WMO Report No. 1203. World Meteorological Organization, Geneva, Switzerland. Wood, S.A., Karp, D.S., DeClerck, F., Kremen, C., Naeem, S., Pal m, C.A., 2015. Functional traits in agriculture: agrobiodiversity and ecosystem services. Trends Ecol. Evol. 30, 531 539. https://doi.org/10.1016/j.tree.2015.06.013 Xiang, Z., Bailey, R.T., Nozari, S., Husain, Z., Kisekka, I., Sharda, V., Gowda, P., 2020. DSSAT - MODFLOW: A new modeling framework for exploring groundwater conservation strategies in irrigated areas. Agric. Water Manag. 232, 106033. https://doi.org/10.1016/j.agwat.2020.106033 Yang, J.M., Yang, J.Y., Liu, S., Hoogenboom, G., 2014. An evaluation of the statistical methods for testing the performance of crop models with observed data. Agric. Syst. 127, 81 89. https://doi.org/10.1016/j.agsy.2014.01.008 Yang, W., Feng, G., Read, J.J ., Ouyang, Y., Han, J., Li, P., 2020. Impact of cover crop on corn soybean productivity and soil water dynamics under different seasonal rainfall patterns. Agron. J. 1 15. https://doi.org/10.1002/agj2.20110 Zampieri, M., Weissteiner, C.J., Grizzetti, B., Toreti, A., van den Berg, M., Dentener, F., 2020. Estimating resilience of crop production systems: From theory to practice. Sci. Total Environ. 735, 139378. https://doi.org/10.1016/j.scitotenv.2020.139378 Zhang, X.Q., Pu, C., Zhao, X., Xue, J. - F., Zhang, R., Nie, Z. - J., Chen, F., Lal, R., Zhang, H. - L., 2016. Tillage effects on carbon footprint and ecosystem services of climate regulation in a winter wheat summer maize cropping system of the North China Plain. Ecol. Indic. 67, 821 829. https://doi.org/10.1 016/j.ecolind.2016.03.046 Zhou, X., Lin, H., Zhu, Q., 2007. Temporal stability of soil moisture spatial variability at two scales and its implication for optimal field monitoring. Hydrol. Earth Syst. Sci. Discuss. 4, 1185 1214. Zipper, S.C., Qiu, J., Kucha rik, C.J., 2016. Drought effects on US maize and soybean production: spatiotemporal patterns and historical changes. Environ. Res. Lett. 11, 094021. https://doi.org/10.1088/1748 - 9326/11/9/094021