Targeting regenerative farming practices to increase crop yields globally

Targeting regenerative farming practices to increase crop yields globally | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Targeting regenerative farming practices to increase crop yields globally Kpade O. L. Hounkpatin, Emanuela De Giorgi, Mika Jalava, Jeroen Poelert, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6409921/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Regenerative farming practices (RFP) such as no-tillage (NT), cover crops (CC), agroforestry (AF), and organic farming (OF) are increasingly being promoted to improve soil health and sustainably increase food production. However, how the suitability and impact of these practices varies across the landscape is unclear. Here, we evaluate the environmental suitability for each of these four practices across the world’s croplands and identify areas where these practices could increase crop yields. To achieve this purpose, a Random Forest model was used to estimate and map the relative yield change globally using field-scale experiments from multiple meta-analyses linked with global gridded climate, soil and environmental datasets, at 5 arc-min resolution. Areas with increasing yields varied across practices and regions, ranging from 0.86 to 60% of the potential areas of the cropland. When evaluating the area coverage for various RFP, whether individually or together with other practices, it appeared that AF would be more suitable for increasing yields with about 60% of the cropland area followed by cover crops (59%), no-tillage (32%) and organic farming (1.3%). For possibilities where more than two RFP might potentially be suitable, cover crop occurred more frequently alongside agroforestry (CC, AF), organic farming (OF, CC) and no-tillage (NT, CC, AF). These results highlight how regenerative framing practices’ impact on yield varies across places and can be used to target policies and actions to have a greater impact on both soil health and food production. Biological sciences/Ecology/Agri ecology Earth and environmental sciences/Environmental sciences/Environmental impact regenerative farming approaches suitability no-tillage cover crops agroforestry organic farming Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Food demand will increase globally over the coming decades as the global population is projected to surpass nine billion by 2050 1 . According to some estimates, the growing and increasingly affluent population will require approximately 50–100% more food in 2050 than is produced today 2 – 4 . Meanwhile, a recent evaluation of the world´s land by FAO revealed that one-third of soils in the world are degraded and fertile soil is being lost at the rate of 24 billion tons of topsoil every year 5 . This is typically a result of unsustainable land-use and management practices, such as natural vegetation removal, intensive agricultural operations, under (or over) fertilization, and erosion 6 , 7 . Meeting future food demand and lowering the environmental stressors would require a complete change in paradigm and the adoption of sustainable means of production at a global scale. In that regard, scientists widely agree that humanity cannot sustainably increase its use of land, water and other key resources for food production 8 – 10 . Rather, for many scholars and policy makers, the way forwards lies in sustainable intensification with the adoption of management practices that are more resilient to extreme weather events and that would result in yields being increased without any further conversion of agricultural land and without exacerbating soil degradation 2 , 6 , 11 , 12 . This acknowledges the need for integrated soil conservation and nutrient management strategies that will result in a productive agricultural system which restore degraded soils and ecosystems and improve soil quality, while at the same time reduce net anthropogenic emissions and enhances the natural resource base and environment. Multiple soil conservation and nutrient management strategies that are related to sustainable intensification practices are embodied in the concept of regenerative agriculture. Although the definition for regenerative agriculture is rather broad, soil conservation is considered to be its bottom line 13 with the aim at reducing the negative footprint of farming on the environment while at the same time improving and restoring the agri-environment to a better productive state 14 – 16 . Frequently reported regenerative farming practices (hereafter mentioned as RFP), are no tillage (NT), agroforestry (AF), cover crop (CC) and organic farming (OF) 17 – 20 due to their potential to improve soil health and increasing soil productivity in the context of sustainable agricultural practices. In recent years, several studies that analyzed and summarized the impacts of different RFP on soil properties 19 , 20 , crop yields 21 , soil microbial biomass 22 , greenhouse gas emissions 23 are available at global scale. The impacts of RFP on yields were reported to vary with crop types, agricultural management practices, climate zones and geographical areas 21 , 24 . Implementing NT resulted in lower, equal or higher yields compared to conventional tillage 25 , 26 . Similar trends were also reported for OF global analysis reporting 18–50% reduction in yield especially in North America and Europe 27 while about 16% increase is reported for tropical countries in Africa 28 . A global meta-analysis shows that AF resulted in average increase of 7–16% in crop yield especially in subtropical and tropical zones 29 , while in average 2.6% reduction occurred in European field experiments, slightly varying depending on the density and age of the trees 30 . Findings related to CC suggest up to 14% yield increase especially in coarse soil texture and dryland areas along with the use of leguminous cover crops 31 . On the other hand, up to 3% yield reductions were observed under CC especially for cash crops in temperate conditions 32 , 33 . Thus, RFP have led to either increase or decrease in crop productivity compared to conventional techniques, with the direction and magnitude of the treatment outcome being dependent on various factors. Consequently, there is no universal “rule” regarding the outcome of specific RFP, as many variables must be considered, including soil properties, climate, crop type, topography etc. Moreover, existing studies have focused mainly on the variation in productivity for specific RFP without investigating their comparative potential across various factors and determining which management practice(s) could be more beneficial towards yield increase for a specific location. Identifying potential geographical locations where the implementation of one or more RFP would lead to an increase in productivity could equip decision makers towards a sustainable informed farming and use of land-based resources. Here, we identify potential areas where RFP maintain or increase crop yields. We conducted a spatial suitability analysis using observations from multiple global meta-analyses and global spatial datasets for climate / bio climate, topography, soil properties, and vegetation productivity. Then, we used a Random Forest model as learning algorithm to perform a regression that links our set of global spatial predictor variables against observations of the RFP effect size. Maps resulting from the upscaling of the effect size (ES) to a global scale and its related uncertainties allowed us to identify where each management practice would potentially be suitable without decreasing the yields or, preferably, even increasing them. Our findings contribute to the global efforts towards sustainable farming which still require empirical information towards the different opportunities embedded into the RFP for increasing global food. Methods Data collection The data used in this study are derived from various meta-analyses across the world (Fig. 1 ). The main inclusion criteria were that studies ought to: (1) be global focusing on RFP, (2) have a collection of plot level experiments, (3) have records of x, y coordinates of the experiment locations, (4) have quantitative information about control and treatment yields. The search was restricted to meta-analyses focusing on the following RFP: no-tillage (NT), agroforestry (AF), cover-crop (CC) and organic farming (OF). Data were aggregated from: (1) studies used to create the FarmGeek ( https://www.farmgeek.xyz/homehttps://www.farmgeek.xyz/home ) platform which synthetized peer-reviewed literature on the outcomes of agricultural management practices and food system interventions, (2) Xu et al. 21 and (3) Jian et al. 6 . The main studies considered from FarmGeek are: Pittelkow et al. 26 (NT), Ding et al. 34 (OF) and Felix et al. 35 (AF). Preliminary quality check was carried out to identify and remove replicates in the individual studies reported in these meta-analyses. Initially, studies reported yields for 124 crops, before being grouped here into seven main groups. The most cultivated crops in the world such as maize, wheat, soybean and rice were considered separately while the remaining were classified into other cereals, cash-crops, vegetables & fruits and others (see Supplementary Table 1). The ES, i.e. the response ratios (RR) of crop yield to these management systems was calculated as follows: \(\:{ln}\left(RR\right)={ln}\left({X}_{T}/{X}_{C}\right)\) (1) where X T and X C are the yield value under treatment (NT, AG, CC, or OF) and control, respectively 36 . Environmental data and feature selection The gridded environmental data considered in this study can be divided into four main groups (Table 1 ): topography (11 covariates), soil properties (9 covariates), climate / bio-climatic (31 covariates), and land cover (3 covariates). Below each group is briefly introduced and justified. Topography : Many studies consider topographical variables as key drivers of spatial variability in crop yield as they interact with weather to influence soil temperature and moisture 37 , 38 . The following topographical variables from the global study of Amatulli, et al. 39 were considered: elevation, slope, aspect (cosine and sine), plan curvature, profile curvature, topographic position index, terrain ruggedness as well as the Shannon index of geomorphological landform which is an indicator of the diversity of geomorphological landforms. Additionally, the geomorphological landform grid data produced by Iwahashi, et al. 40 as well the Fraction of Absorbed Photosynthetically Active Radiation by Hackländer, et al. 41 were considered. Soil properties : Crop yields are closely dependent on soil properties which control nutrient movements, aeration, nutrient cycling and root growth 42 , 43 . Global soil properties such as soil texture (sand, silt, clay), bulk density (BD), soil organic carbon (SOC), world reference-based soil types, pH were downloaded from the SoilGrids platform 44 . The global stock of soil Olsen phosphorus came from the global study carried out by McDowell et al. 45 while the soil moisture for the 2001–2020 period dataset was gotten from the TerraClimate 46 and the average was computed for prediction. Climate / bio-climatic layers : Climatic variables have been documented to have a major impact on crop growth and food production worldwide 47 – 49 . This study compiled different CHELSA climate and bioclimatic variables covering the 1979–2013 period 50 (Bio1-Bio19, see Supplementary Table 2 for full definition). A multisource (SM2RAIN-ASCAT 2007–2021, CHELSA Climate and WorldClim) average for monthly precipitation (1 km) was considered in this study along with its standard deviation ( https://zenodo.org/records/6458580 ). The long-term averaged monthly mean (2000–2017) time series and standard deviation of the MODIS land surface temperature (daytime and nighttime) were also used as predictors ( https://doi.org/10.5281/zenodo.1420114 ). The global aridity index was obtained from the Consortium for Spatial Information 51 . In addition, the Growing degree days (GDD) for maize, wheat, rice and soybean used in this study were sourced from Ahvo, et al. 52 . Solar radiation was derived from digital elevation model using the SAGA software 53 . Land cover : The Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) are both vegetation indices which are frequently used for crop yield related studies 54 – 56 . The mean of the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) from the MODIS collections (2000–2023) were computed using google earth engine. The bare soil data consisting in the fractional area not covered by vegetation was accessed from the spatial data platform ( https://eo-data.csiro.au/remotesensing/rapp-help/ ) of the National Landcare Regional Partnerships Program 57 . Table 1 Environmental variables (in bracket are abbreviations). Covariates Unit Resolution References Soil properties Sand % 250 m Poggio, et al. 44 Silt % 250 m Poggio, et al. 44 Clay % 250 m Poggio, et al. 44 Ph - 250 m Poggio, et al. 44 Bulk density (bd) kg/dm³ 250 m Poggio, et al. 44 World reference-based soil types (wrb) - 250 m Poggio, et al. 44 Soil organic carbon (SOC) g/kg 250 m Poggio, et al. 44 Soil Olsen phosphorus concentrations (phosphorus) mg/kg 1000 m McDowell et al. 45 Soil moisture (sm) mm 200 m Abatzoglou, et al. 46 Topography Cosine of aspect (cosasp) ° 0.0083 ° Amatulli, et al. 39 Digital elevation model (dem) m 0.0083 ° Amatulli, et al. 39 Shannon index of geomorphological landform (indxshan) - 0.0083 ° Amatulli, et al. 39 Plan curvature (plc) ° m − 1 0.0083 ° Amatulli, et al. 39 Profile curvature (prc) ° m − 1 0.0083 ° Amatulli, et al. 39 Sine of aspect (sinasp) ° 0.0083 ° Amatulli, et al. 39 Slope geomorphological landform ° 0.0083 ° Amatulli, et al. 39 Topographic position index (tpi) - 0.0083 ° Amatulli, et al. 39 Terrain ruggedness(tr) - 0.0083 ° Amatulli, et al. 39 Fraction of Absorbed Photosynthetically Active Radiation (fapar) % 0.0020 ° Hackländer, et al. 41 Geormorphological landform - 0.0083 ° Iwahashi, et al. 40 Climatic / Bio-climatic Bioclimatic variables (Bio1-Bio19) - 0.0083 ° Karger, et al. 50 Growing degree days ° C 0.0083 ° Ahvo, et al. 52 . Temperature (temp) ° C 0.0083 ° Hengl, et al. 58 MODIS land surface temperature daytime (lstd) ° C 0.0083 ° Hengl, et al. 58 Standard deviation of the MODIS land surface temperature (sd lstd) ° C 0.0083 ° Hengl, et al. 58 MODIS land surface temperature night time (lstn) ° C 0.0083 ° Hengl, et al. 58 Annual Precipitation (pcp) mm 0.0083 ° Hengl, et al. 58 Standard deviation Annual Precipitation (sdpcp) mm 0.0083 ° Hengl, et al. 58 Solar radiation (srad) W/m 2 0.0083 ° Mujić and Karabegović 53 Aridity index (aridity) - 0.0083 ° Zomer, et al. 51 Vegetation cover Bare soil (bs) - 0.0020 ° Becker-Reshef, et al. 57 Normalized Difference Vegetation Index (ndvi) - 500 m Shammi and Meng 54 Enhanced Vegetation Index (evi) - 500 m Shammi and Meng 54 Data analysis with a machine learning model The gridded covariate data were overlayed with the ES data to create the regression matrix (Fig. 2 a) for further analysis with the Random Forest which was used as the reference modelling technique for prediction. The basic implementation of Random Forest falls short of considering the spatial context especially when validation strategies ignore spatial autocorrelation in the data. To overcome this, we conducted the RF modelling using the Leave-Location-Out cross-validation (LLOCV), that considers potential spatial autocorrelation. LLOCV involves training models repeatedly by leaving the data from one location or a group of locations out and using the remaining to validate model 59 . Modelling was carried out considering different management and crops. Firstly, all the management data were merged and modelling was conducted without accounting for any crops to get the overall trend in the data. Secondly, analysis was carried out considering each management and related crops. Following the principle of parsimony, a feature selection was carried out by: (1) removing highly correlated (> 0.70) covariates before modelling, (2) using the forward feature selection (FFS) method that functions in combination with target-oriented performance to detect and remove variables that lead to overfitting 59 . The data were split, with 80% of the samples to train the models while 20% were used as an independent validation set (Fig. 2 a). For cases where the training observation is less than 200 samples, only a cross validation was carried out to allow the model to learn on the full dataset. There is a general agreement for using cross-validation for training small size data 60 , 61 . Random Forest parameters such as ntree, mtry (number of variables at each split) and the minimum node size (minimum number of observations for each node) can be tuned to improve prediction performance. To enable faster computer processing, the default value was used for the ntree (ntree = 500) while the latter two were subjected to fine-tuning using the grid search method. We used the “caret” R Package 62 to carry out the different model calibrations based on a three-time repeated 10-fold cross-validation. The prediction uncertainties were derived through Quantile Regression Forests (QRF) 63 . QRF are a generalization of Random Forest (Fig. 2 b). In contrast to Random Forest which keeps only the mean of the observations that fall into each tree node, QRF retain the value of all observations in this node considering thereby the spread of the response variable 63 . Consequently, it has the advantage of producing the prediction interval by assessing the distribution of observed response variables at each leaf of the tree. Thus, a 90% uncertainty interval can be computed by considering 0.05 quantile as the lower bound, and the 0.95 quantile as the upper bound. Consequently, the 90% uncertainty maps related to the ES prediction maps were produced and used for assessing the accuracy of the predictions. Selected gridded covariates were resampled at a resolution of 5 arcmin for the prediction of the ES and related uncertainties. Areas that were not croplands were masked out using the cropland mask produced based on SPAM 2020 crop production data by Yu, et al. 64 . The RF performance (Fig. 2 b) was assessed based on the coefficient of determination (R 2 ), Root Mean Squared Error (RMSE) and the Lin’s concordance correlation coefficient (LCCC). The LCCC measures both the accuracy and precision of the relationship between the observations and the predictions 65 . Developing suitability maps The ES measures the magnitude of the treatment (NT, CC, AF, OF) impact. Values of ES near zero indicate small effects in terms of absolute magnitude while those far from zero indicate large effects. The assumption is that the RFP would perform better than the conventional agriculture resulting in positive and potentially large ES. It is therefore possible to spot out on a gridded map of ES the direction (locations where a given treatment has positive or negative ES) as well as the magnitude of its impact (ES values). The implementation of a given RFP is potentially suitable for a given location (1) if the predicted ES at that very location is different from zero and positive (> 0) and (2) if its corresponding prediction uncertainty is the lowest at that very location compared to the remaining RFP (see example in Supplementary Fig. 1). The following steps (Fig. 2 c) were followed in that regards: build the Random Forest model based on the environmental factors and the observations (ES for a given treatment and a particular crop/crop group) create ES map for each RFP map the uncertainty by predicting the prediction confidence interval create suitability map for each RFP choose most suitable RFP considering areas with positive ES along with the related uncertainty values (step 3). At a given location, consider the RFP which has positive ES and the lowest uncertainty compared to the remaining RFP. repeat step 5 to consider more than one RFP. Relative importance of the predictor variables We used the Shapley values (Fig. 2 b) as a metric for assessing variable importance. Shapley values were established in relation to the cooperative game theory proposed by Shapely 66 and recently revised by Lundberg et al 67 . In a cooperative game, “players” have the possibility to function as a team towards achieving a common goal with the consideration of a fair distribution of payoffs among the members of the coalition. In machine learning, the application of the Shapley value intends to fairly assign credit for a model’s output among its input features. Consequently, this theory allocates in the context of this study a contribution value to each feature involved in prediction of the ES. The calculation of the Shapley values was conducted with the “fastshap” R package 68 . In addition, partial dependence plot (PDP) which allows the assessment of whether the relationship between the target and a predictor variable is linear, monotonic or more complex was considered 69 . Results The focus in this result section is on presenting: (1) the outcome of the overlay of the yield changes with RFA using the effect size (ES) predictions and related uncertainty maps (Fig. 3 ), (2) the variable importance accounting for factors that influenced the ES predictions (Fig. 4 ), (3) the result of the suitability coverage of the different RFP based on the ES and related prediction uncertainties (Fig. 5 , 6 ). Spatial distribution of effect size (ES) and variable importance AF and CC seem suitable over most of the cropland areas for increasing yield at a global scale (Fig. 3 a,b). However, this potential for increasing yield was associated with a high level of uncertainty for about 96% (see Supplementary Table 3) of the cropland area AF. Similarly, about 49% and 38% of the cropland area were associated with high and medium level of uncertainty respectively under CC. On the other hand, NT and especially OF are associated with high occurrences of negative yield change respectively under 64% and 97% of the cropland areas (Fig. 3 c,d; see Supplementary Table 3). However, potential areas for increasing yield under NT are scattered all over the globe with much more prevalence in the upper part of North America, Latin America and Caribbean, West and East Africa, South and East Asia. Variable importance was evaluated to detect the most important predictors affecting the model predictions. They were identified based on the Shapley values which measure the magnitude of their contribution to the final model predictions. This was complemented with the PDP (i.e. partial dependence plot) to visualize how the model predictions change as the value of a particular predictor changes while holding the remaining constant. The analysis of the distribution of the Shapley values revealed that none of the environmental covariates contributed steadily to the increase of the ES over the whole range of their values with occurrences of specific ranges over which their impact on the model output was either low or high as further confirmed by the PDP (Fig. 4 , Supplementary Fig. 6–7). The detailed findings in relation to the variable importance are further presented in the following sections. Overall, the Precipitation of Driest Quarter (bio17) and the aridity index drive the spatial distribution of the ES when considering all management (Fig. 4 a). However, the order of importance varies based on the respective RFP as well as the crop group being considered under specific management (Fig, 4, Supplementary Fig. 6–7). AF was primarily affected by phosphorus and base soil (bs) at a global level (Fig. 4 d) while cropwise (Supplementary Fig. 6) climate variables either ranked first or second respectively for maize (Supplementary Fig. 6a, bio2: Mean Diurnal temperature Range), other cereal (Supplementary Fig. 6b, bio17) and vf & o (Supplementary Fig. 6c, bio7: Temperature Annual Range). The key variables influencing the ES under CC are topographical variables such as landform and Shannon index (Fig. 4 g). Crops under CC are mainly influenced by a combination of soil properties and climate (maize, wheat). For maize (Supplementary Fig. 6d) and vegetable, fruits and others (Supplementary Fig. 6f) crop groups, the soil moisture and Maximum Temperature of Warmest Month (bio5) as well as the Precipitation of Driest Month (bio14) were respectively identified as the most important variable affecting ES. The soil type (wrb) and silt content were ranked as the key variables for the wheat (Supplementary Fig. 6e) under CC. The results show that spatial distribution of the ES under NT was mostly associated with climate (Precipitation of Coldest Quarter (bio 19)) and topographical (sinus of aspect (sinasp)) variables (Fig. 4 j). This trend is also reflected at crop level (Supplementary Fig. 6) with most of the crops being primarily influenced by Climate variables especially bio17 for maize (Supplementary Fig. 6g), Mean Temperature of Driest Quarter (bio9) for rice (Supplementary Fig. 6h), aridity for soybean (Supplementary Fig. 6i), standard deviation of annual precipitation (sdpcp) for wheat and other cereal (Supplementary Fig. 6j,l), bio7 for cash crop (Supplementary Fig. 6k). OF was primarily affected (Fig. 4 m) by vegetation cover (Enhance vegetation index), climate (sdpcp) and soil properties (soil type (wrb)) with similar trend observed for maize (wrb, slope, Supplementary Fig. 6n), wheat (landform, wrb, Supplementary Fig. 6o), other cereal (Topographic position index (tpi), bare soil (bs), Supplementary Fig. 6p) and vf & o (sinasp, cosasp, soil moisture(sm), Supplementary Fig. 6q). Potential suitability areas for regenerative management practice Considering the predicted ES maps of the RFP and related uncertainties resulted in the identification (Fig. 5 ) of location with the highest productivity potential under the implementation of the related RFP. Any of such identified management is thereafter considered as the most suitable for that given location in comparison with the remaining RFP. Other potential configuration is labelled into the class “other_RFP” (Fig. 5 , 6 ). Mainly NT (i.e. no-tillage, Fig. 5 a,d) stood out among the RFP class with a global area coverage (0.90 billion ha, (Table 2 ) of 30% compared to 7% for AF (i.e. agroforestry, Fig. 5 a,b), 5.6% for CC (i.e. Cover Crop, Fig. 5 a,c), 1.3% for OF (i.e. Organic Farming, Fig. 5 a,e). At a regional scale (Table 2 ), the highest area coverage for NT is found in North America and South Asia while the lowest are in Latin America & Caribbean and Sub-Saharan Africa. The potential coverage of AF is restricted in regions such as Sub-Saharan Africa and South Asia followed by East Asia & Pacific. CC was found to be potentially more productive in Sub-Saharan Africa, East Asia & Pacific and Latin America & Caribbean. Regionally, the suitability coverage of the RFP for the different crops varied across crops and regions (Fig. 6 ; see Supplementary Table 4). NT was found to be more suitable for maize and wheat in Middle East & North Africa to Latin America & Caribbean (Fig. 6 a,b). For other cereal crop group, OF seems to be suitable at global scale in all regions (Fig. 6 c) while for the vegetable, fruits and others crop group NT and AF are more distinct (Fig. 6 d). For the latter crop group, NT is more prevalent at a regional scale in Latin America & Caribbean, North America, Europe & Central Asia. Table 2 Coverage areas of the regenerative farming practices (RFP). AF: Agroforestry, CC: Cover crop, NT: No-tillage, OF: Organic farming. Multiple RFP (AF,CC; CC,AF; NT,CC,AF) mean that each has positive yield at a given pixel location with ranking based on increasing level of uncertainty (see Supplementary Fig. 1). AF CC NT OF AF,CC CC,AF NT,CC,AF other_RFP All regions % 7.00 5.60 30.00 1.28 4.93 45.87 2.49 2.84 ha (x 10 9 ) 0.21 0.17 0.90 0.04 0.15 1.37 0.07 0.09 East Asia & Pacific % 6.81 6.69 28.92 0.26 5.79 44.90 3.39 3.24 ha (x 10 9 ) 0.04 0.04 0.18 0.00 0.04 0.28 0.02 0.02 Europe & Central Asia % 6.01 0.37 32.89 0.00 13.61 44.62 1.39 1.11 ha (x 10 9 ) 0.03 0.00 0.15 0.00 0.06 0.20 0.01 0.00 Latin America & Caribbean % 5.84 5.85 27.92 0.03 0.39 55.52 2.79 1.66 ha (x 10 9 ) 0.03 0.03 0.13 0.00 0.00 0.26 0.01 0.01 Middle East & North Africa % 5.69 3.90 29.67 1.55 5.08 46.25 4.11 3.75 ha (x 10 9 ) 0.01 0.00 0.04 0.00 0.01 0.06 0.01 0.00 North America % 5.53 0.73 42.79 0.17 9.41 38.87 0.93 1.56 ha (x 10 9 ) 0.02 0.00 0.15 0.00 0.03 0.14 0.00 0.01 South Asia % 8.93 4.64 42.24 0.62 1.01 36.78 2.74 3.05 ha (x 10 9 ) 0.03 0.01 0.12 0.00 0.00 0.11 0.01 0.01 Sub-Saharan Africa % 8.80 11.14 18.58 4.73 0.89 48.48 2.57 4.83 ha (x 109) 0.06 0.08 0.13 0.03 0.01 0.33 0.02 0.03 The study also considered the potential of having more than one management (Fig. 5 , 6 ) being suitable for a given location with all RFP having positive ES at that location with the order of ranking based on the magnitude of their respective uncertainty. RFP with the lowest uncertainty would rank first followed by the second which in turn would have lower uncertainty compared to the third for cases with three RFP (see Supplementary Fig. 1). Overall, cover crop and agroforestry (CC, AF; Fig. 5 g) had the highest suitability coverage with about 46% of the total share of the global cropland area (Table 2 ) while regionally the highest records were in Latin America & Caribbean (55%), Sub-Saharan Africa (48%) and Middle East & North Africa (46%) and the lowest in North America (39%) and South Asia (37%). Next are classes where AF took preeminence over CC (AF, CC; Fig. 5 .f) and where NT ranked first over CC and AF (NT, CC, AF; Fig. 5 h). Considering the occurrences of different of management at the same location for the crop groups shows different pattern for each crop. For maize (Fig. 6 a), three classed emerged in descending area coverage order OF, CC, AF; OF, AF, CC and CC, AF. For wheat (Fig. 6 b), mostly AF took preeminence over CC (AF,CC) while the remaining showed CC ranking first before AF (CC,AF) and OF (CC,OF). For the other cereal crop group (Fig. 6 c), the co-occurence of different RFP did not find expression while for the last crop group, vegetable, fruits and others (Fig. 6 d), OF, CC have better distinct coverage followed by CC, OF and NT, OF, CC. In total, cumulating the total area coverage either individually or together with other practices, it appeared that AF would be more suitable for increasing yields with about 60% of the cropland area followed by cover crops (59%), no-tillage (32%) and organic farming (1.3%). Discussion Suitability coverage of Regenerative Farming practices Our study examines the potential area suitability for different RFP as a strategy for defining geographical positions within agricultural land where crises of soil health, biodiversity, and food security could be addressed as a consequence of the implementation of more sustainable farming approaches. This was investigated by integrating via modelling different environment factors - soil properties, climate, vegetation and topography – and overlaying the spatial distribution of RFP related relative yield and corresponding uncertainties to establish potential locations for their feasibility. This analysis can serve as a crucial foundation for identifying hotspots where the implementation of one or more RFP would lead to an increase in productivity and thereby provide insight towards informed decision for policymakers and farmers. At a global scale, previous studies report NT and OF to be practiced on only about 9% 20 and 1.6% 70 of cultivated land respectively. About 15% of the cropland area is reported to be used for Agroforestry 71 . Poeplau and Don 72 reported about 25% of cropland to be suitable for CC. Our results predict a higher coverage potential for NT (30%) while reducing outcome of AF, CC and OF as single class to 7%, 5.6% and 1.3% (Fig. 5 ). The ES of the four RFP considered in this study were primarily moderated (Fig. 4 ) by climate variables while specifically, AF, CC and OF were mostly influenced by soil properties and topographical and vegetation variables. It seems therefore that local environment (soil properties, topography, vegetation cover) seem to prevail over global variable (climate) in the response of crops to the RFP. On the one hand, NT took preeminence over CC and AF when considering potential simultaneous occurrences of the RFP while CC came first in most instances as compared to AF (Fig. 5 ). This might have to do with the total number and spatial data distribution of each RFP with NT having the highest followed by CC. Consequently, spatial modelling uncertainties were relatively lower for areas where these managements have positive ES compared to the remaining. However, the distribution of the RFP in the current study were region and crop specific and are generally in line with previous studies 26 , 35 regarding the areas where these RFP are prevalent though with a higher magnitude for potential implementation. Previous studies found about 45% of the total global area are under NT in South America followed by the US with 32% 20 with the lowest in Europe (1%) and Africa (1%) 73 . Most studies in these regions report higher yield with NT compared to conventional farming in semi-arid and arid regions or drier conditions 74 , 75 . Such regions are captured in the present study with some locations in South Asia, North America, Middle East and North Africa especially for maize and wheat (Fig. 5 a,b). Generally, such a trend in water limited environment for NT is explained by improved infiltration and reduced evaporation resulting in an enhanced soil moisture availability 26 , 76 . The highest ES was observed under AF (Fig. 3 a) globally as well as for different crops (Supplementary Fig. 2–4). This observation aligns with the finding of Ren, et al. 36 who also found the yield increase under AF surpassing that of NT, CC and OF. This result can be attributed to various factors such as improvement in soil properties, increased capacity for erosion control with trunks, roots and litter reducing run-off, mitigation of crop pests and disease as well as better carbon storage 30 , 36 . For example, a meta-analysis for implementing AF over sub-Saharan Africa found the following: average crop yield increased almost by a factor of 2, soil fertility by a factor of 1.2, control of runoff and soil loss by a factor of 5 to 9, and infiltration by a factor of 3 as compared to the control 77 . However, the global ES distribution as found in Fig. 3 a was overoptimistic as few studies also reported negative ES under AF 29 , 30 . This was better controlled by considering the uncertainties which also appeared to be higher for AF compared to the remaining RFP. Consequently, mostly CC, NT or OF took preeminence when considering the potential co-occurences (Fig. 5 ,7). This resulted in a significant restriction of the suitability coverage for AF mostly into Sub-Saharan Africa, South Asia and East Asia & Pacific (Fig. 5 a,b). However, there might actually be higher potential for AF as Sprenkle-Hyppolite, et al. 78 evaluating the opportunities to increase tree cover without reducing yields in cropland found maximum tree cover increase potential of 22% for North America, 21% for Europe & Central Asia, 20% for East Asia & Pacific, 19% to Sub-Saharan Africa, 15% for Latin America & Caribbean and South Asia and 13% for Middle East & North Africa. The resulting tree cover increase map follows closely areas covered by the ES map for AF (Fig. 3 a) although high uncertainties were associated with these predictions. On the other hand, considering the cumulative occurrence of AF either alone or together with other RFP showed that AF would be more suitable for increasing yields with about 60% of the cropland area. Current results also fall within the major ecological regions identified by the FAO for implementing AF such as temperate, mediterranean, arid and semiarid, subhumid tropical (lowland), humid tropical (lowland) and highland 79 . Additionally, it was interesting to note AF prevalence also next to cover crop (CC, AF; Fig. 5 g; Fig. 6 a,b), no-tillage (NT, CC, AF; Fig. 5 h) or OF (OF, AF, CC; Fig. 6 a) in remaining areas such as North America, Middle East & North Africa, Latin America & Caribbean. CC was mostly represented when considering different combinations of RFP with about 46% of global coverage in combination with AF with about 36–55% share at regional scale in Latin America & Caribbean, Sub-Saharan Africa and Middle East & North Africa, North America and South Asia (Fig. 5 g). Both occupied areas where NT and OF have negative ES. Reduced productivity levels of NT are reported in the context of waterlogging and poor crop establishment, restricted root growth due to compaction, nutrient deficiencies 80 , decreased soil temperature especially for humid environment 81 . On the other hand, poor performance occurs often in OF system as a result of challenges related to weed and pests control and nutrients availability especially in nitrogen and phosphoru 27 , 82 , 83 . Though CC implementation involves also risks such as competition with the cover crops for water 84 , it is reported in different context in improving weed suppression 85 , 86 as well as enriching the soil via biological N fixation 87 . Most production on OF land (1.6% of global cropland 70 ) is reported to be taking place in Oceania (47%), Europe (23%), Latin America (13%), Asia (8.5%), Northern America (4.6%) and Africa (3.5%) 88 with the highest share of arable cropland being devoted to cereals (37%). We also found a high potential of OF for other cereals as single management (Fig. 6 c) as well as with maize crop when forming combination with CC and AF (OF, CC, AF; OF, AF, CC; Fig. 6 a). With the crop group related to vegetable, fruits and others its potential increase yield was linked with CC (OF, CC; Fig. 6 d). Different RFP have potential to lead to positive yield increase at the same locations. Our finding showed that two RFP (All crops: CC,AF; Wheat: AF,CC; other cereal: OF,AF; vegetables, fruits and others: OF, CC) and even three RFP (All crops: NT, CC, AF; Maize: OF, CC, AF; vegetables, fruits and others: NT,OF,CC) showed positive impact on yield in the same location (Fig. 5 , 6 ). This possibility might be further explored for promoting the implementation of more than one RFP at the same time during the growing season. A key criterion for implementing such a combination of practices might be the potential of one practice to cancel the limitations of another and thereby possibly further reduce the yield gap. For example, some findings report the inclusion of CC into organic systems as OF suffers often yield reduction because of poor weed control while CC on the other hand has potential to contribute more effectively to weed suppression 89 , 90 . Also, CC is introduced into NT (NT, CC) or AF (AF, CC) systems to provide nitrogen, add organic matter, further reduce soil erosion, improve soil structure thereby increasing the intrinsic fertility of the soil 91 , 92 . There is evidence that in AF, CC systems nutrient cycling processes involving litterfall from both shrubs or deep rooted trees and cover crops can mobilize nutrients from topsoil or the subsoil and this in turn result in the improvement of crop productivity 79 . Similarly, the benefits of integrating AF and NT are acknowledged 93 , 94 while combining OF, AF, NT does not receive the same attention in research compared to OF, NT, CC. Findings related to organic no-till system cover crops (OF, NT, CC) revealed 46% and 42% reduction in labour and fuel along with 27% yield reduction in a soybean cropping system 95 while about 14% reduction in reduced variable costs are recorded with additional 19% greater net revenue for the same crop 96 . The successes or failures in implementing more than one or two RFP are dependent on different variables such as farmers´ knowledge and resources, type and amount of CC in the field along with CC termination strategies 95 – 97 . Consequently, while acknowledging that combining several RFP can have greater benefits to soil health and potentially increase yield than either alone, there is a need to balance costs, practical feasibility and potential local outcomes for additional combination of these practices. The main criteria for deciding upon suitability in this study is the ES (i.e. Effect Size). Other focus of RFP such as the reduction of greenhouse gas emissions and the buildup of soil organic matter as well as the protection of biodiversity and reduction of soil erosion were not considered. These non-monetary advantages of RFP are mostly not tangible in the short term and several years of management might be required before any consistent outcome 27 . Although improved yield is commonly reported as key determinant for the adoption of many RFP 98 , 99 , these non-monetary advantages might actually be alternative motivations to implement these practices for farmers and/or decision makers. Further studies might focus on comparative analysis of the four RFP considered in this study regarding their feedback on greenhouse gas emission, biodiversity, soil organic carbon (SOC) and erosion. On the other hand, any farm operation resulting in reduced profits compared to current management practices might face reluctancy towards adoption 24 . Consequently, although looking beyond suitability into RFP adoption patterns and mechanisms was out of the scope of this study, it will be beneficial that more studies report not only on the ecological and agronomic indicators of the RFP but also on their economic incidences especially in terms of profitability via cost benefit analysis. Additionally, the dataset considered in this study does not include the interaction between these approaches to quantitatively establish the yield gain from their association. On the other hand, practical issues might arise regarding the combination of some of these practices. For example, NT still allows pesticides for weed management to improve yields, while OF (very often) includes more tillage but no pesticides. A combination means that no pesticides can be used in no-tillage, which is obviously good for soil health, but most likely decreases possibilities for improved yields. There is therefore a need to capture the variabilities of the impact resulting from the combination of these practices on yield and other soil health indicators as compared to using either one alone. Although some local studies 91 , 92 – 94 investigated the outcome of combining different RFP at local scale, there is still a need for synthesis at a global scale. The magnitude and direction of the estimates of the ES related to the different RFP were most likely influenced by the specific locations and numbers of the respective studies for a given management. Overall, the locations of the data used in the present study (Fig. 1 ) are mostly from Sub-Saharan Africa for AF (99%), North America and Sub-Saharan Africa for CC (85%), Europe and North America for NT (67%), North America and Europe for OF (84%). In addition, regional and onsite properties especially local climates and soil properties were derived from external global data and might not necessarily correspond to local realities. This translated into low correlation with the predictor variables (Supplementary Table 5) resulting in a Lin’s concordance correlation coefficient between 0.17 and 0.65 (Supplementary Table 6). However, the use of the gridded dataset allowed us to scale-up the ES and have insight into the potential locations that would allow yield increase although local validations might be further required. This study considered the different RFP as a whole without investigating further into their specific variants. For example, different forms of AF range from alley cropping, forest farming, silvopastoralism, riparian forest buffers to windbreaks with some nuances and different degree of tree coverage in temperate and tropical areas 71 , 78 , 79 . On the other hands, NT 21 , 100 , CC 86 , 101 and OF 27 , 89 might occur with or without rotation, soil cover, or include different fertilizer input, amendment types, cover crop types, weed and pest management. While future studies might want to account for such details and differentiate between their unique contribution in yield change, we assumed that many of their influences might already be captured in the final yield recorded for the various managements. The dataset considered in the present study originate from experiments carried out in controlled settings under the directions of researchers. It is often reported that yield gains reported under such settings might actually be upper bounds estimates of the impact of the given RFP 24 . Farmers´ practices might not measure up to the optimal managements implemented in experimental plots resulting thereby in lower yields. Consequently, impacts of RFP measured under real world farm conditions are necessary to fully grasp their potential towards yield gains and thereby further improve the suitability coverage at both local and global scale. Our findings have also shown different areas which have potential for yield growth under implementation of single RFP and further suggested the potential for integrating many RFP for reaching the same purpose. We are however aware that the most suitable combination of RFP or preeminence of one of the RFP over the others is highly context-specific and can depend on many different variables, although local environments seem to be more important than climatic variables. The suitability maps provide great guidance for decision makers and policy programs, to find priority areas or hotspot areas of farming combinations that would potentially increase yields while improving sustainable land management. However, because of the data limitations mentioned earlier, local adaptation and validation of current results is critical to prevent implementing certain RFP that might be incompatible with local agricultural conditions and challenges. Conclusion This study allowed us to evaluate the suitability of different regenerative farming practices at global scale by considering areas where yield increase occurred along with the lowest uncertainty. Our findings confirm a high suitability coverage for no-tillage especially for maize and wheat when considering single management while organic farming was more prominent for other cereal crop groups. Overall, considering the potential of different practices for increasing yield at the same location revealed that cover crop has potential alongside agroforestry, no-tillage and organic farming. Likewise, organic farming as well as no-tillage project high potential for yield increase especially for the crop group related to vegetable, fruits and others crops. The global and crop specific suitability maps of single and combined regenerative farming practices presented in our study offer new perspectives on potential regions that could be subject for local validation towards implementation. They could also serve as potential baseline information source for situation where policy programs require hotspot areas where regenerative farming practices could enhance yields to encourage adoption. In that regard, future research and solutions should aim at investigating further beyond RFP related yield outcome by focusing also on the balance between costs and potential economic impact across different soil, climate, geography, weather, ecology and cultures. Declarations Author Contribution K.O.L.H and M.K. conceptualized and designed the research idea. K.O.L.H analyzed the data and wrote the main manuscript text. E.D.G conducted some preliminary analysis on some part of the data. E.D.G., M.J., J.P., P.W. and M.K. brought major input to the manuscript. All authors reviewed and commented on the manuscript. Acknowledgement This study was funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No. 819202), the Research Council of Finland’s Flagship Programme under project Digital Waters (grant no. 359248), Strategic Research Council (SRC) through project ‘Water & Food’ (grant no. 365512), and Maa- ja vesitekniikan tuki ry. References Gerland, P. et al. World population stabilization unlikely this century. Science 346 , 234-237 (2014). Tilman, D., Balzer, C., Hill, J. & Befort, B. L. Global food demand and the sustainable intensification of agriculture. Proceedings of the National Academy of Sciences 108 , 20260-20264 (2011). FAO. The state of the world's land and water resources for food and agriculture: Managing systems at risk . (Earthscan, 2011). Alexandratos, N. & Bruinsma, J. World agriculture towards 2030/2050: the 2012 revision. (2012). Bakker, H. The world food crisis: Food security in comparative perspective . (Canadian Scholars Press, 1990). Jian, J., Du, X. & Stewart, R. D. A database for global soil health assessment. Scientific Data 7 , 16 (2020). Eugenio, N. R., Naidu, R. & Colombo, C. M. Global approaches to assessing, monitoring, mapping, and remedying soil pollution. Environmental Monitoring and Assessment 192 , 1-2 (2020). Steffen, W. et al. Planetary boundaries: Guiding human development on a changing planet. Science 347 , 1259855 (2015). Foley, J. A. et al. Solutions for a cultivated planet. Nature 478 , 337-342 (2011). Fader, M., Gerten, D., Krause, M., Lucht, W. & Cramer, W. Spatial decoupling of agricultural production and consumption: quantifying dependences of countries on food imports due to domestic land and water constraints. Environmental Research Letters 8 , 014046 (2013). Garnett, T. et al. Sustainable intensification in agriculture: premises and policies. Science 341 , 33-34 (2013). Jules, P., Camilla, T. & Stella, W. Sustainable intensification in African agriculture. International Journal of Agricultural Sustainability 9 , 5-24 (2011). Schreefel, L., Schulte, R. P., De Boer, I., Schrijver, A. P. & Van Zanten, H. Regenerative agriculture–the soil is the base. Global Food Security 26 , 100404 (2020). O’Donoghue, T., Minasny, B. & McBratney, A. Digital Regenerative Agriculture. npj Sustainable Agriculture 2 , 5 (2024). Rehberger, E., West, P. C., Spillane, C. & McKeown, P. C. What climate and environmental benefits of regenerative agriculture practices? an evidence review. Environmental Research Communications 5 , 052001 (2023). Schulte, R. P. & Ditzler, L. in Future Food Systems 17-30 (Elsevier, 2024). Van Kessel, C. et al. Climate, duration, and N placement determine N2O emissions in reduced tillage systems: a meta‐analysis. Global Change Biology 19 , 33-44 (2013). Cárceles Rodríguez, B. et al. Conservation Agriculture as a Sustainable System for Soil Health: A Review. Soil Systems 6 , 87 (2022). Li, Y., Li, Z., Cui, S., Jagadamma, S. & Zhang, Q. Residue retention and minimum tillage improve physical environment of the soil in croplands: A global meta-analysis. Soil and Tillage Research 194 , 104292 (2019). Blanco-Canqui, H. & Ruis, S. J. No-tillage and soil physical environment. Geoderma 326 , 164-200 (2018). Su, Y., Gabrielle, B. & Makowski, D. The impact of climate change on the productivity of conservation agriculture. Nature Climate Change 11 , 628-633 (2021). Li, Y., Chang, S. X., Tian, L. & Zhang, Q. Conservation agriculture practices increase soil microbial biomass carbon and nitrogen in agricultural soils: A global meta-analysis. Soil Biology and Biochemistry 121 , 50-58 (2018). Abdalla, M. et al. Conservation tillage systems: a review of its consequences for greenhouse gas emissions. Soil Use and Management 29 , 199-209 (2013). Jain, M. et al. How much can sustainable intensification increase yields across South Asia? A systematic review of the evidence. Environmental Research Letters 15 , 083004 (2020). Pittelkow, C. M. et al. Productivity limits and potentials of the principles of conservation agriculture. Nature 517 , 365-368 (2015). Pittelkow, C. M. et al. When does no-till yield more? A global meta-analysis. Field Crops Research 183 , 156-168 (2015). Choden, T. & Ghaley, B. B. A portfolio of effective water and soil conservation practices for arable production systems in Europe and North Africa. Sustainability 13 , 2726 (2021). Hine, R. & Pretty, J. Promoting Production and Trading Opportunities for Organic Agricultural Products in East Africa Capacity Building Study 3. Organic Agriculture and Food Security in East Africa , 43-55 (2007). Baier, C., Gross, A., Thevs, N. & Glaser, B. Effects of agroforestry on grain yield of maize (Zea mays L.)—A global meta-analysis. Frontiers in Sustainable Food Systems 7 , 1167686 (2023). Ivezić, V., Yu, Y. & Werf, W. v. d. Crop yields in European agroforestry systems: a meta-analysis. Frontiers in Sustainable Food Systems 5 , 606631 (2021). Peng, Y., Wang, L., Jacinthe, P.-A. & Ren, W. Global synthesis of cover crop impacts on main crop yield. Field Crops Research 310 , 109343 (2024). Clark, A. J., Decker, A. M., Meisinger, J. J. & McIntosh, M. S. Kill date of vetch, rye, and a vetch—rye mixture: II. Soil moisture and corn yield. Agronomy Journal 89 , 434-441 (1997). Thorup-Kristensen, K., Magid, J. & Jensen, L. S. Catch crops and green manures as biological tools in nitrogen management in temperate zones. Advances in Agronomy 79 , 227-302 (2003). Ding, W. et al. Improving yield and nitrogen use efficiency through alternative fertilization options for rice in China: A meta-analysis. Field Crops Research 227 , 11-18 (2018). Félix, G. F., Scholberg, J. M., Clermont-Dauphin, C., Cournac, L. & Tittonell, P. Enhancing agroecosystem productivity with woody perennials in semi-arid West Africa. A meta-analysis. Agronomy for Sustainable Development 38 , 57 (2018). Ren, X., Zou, W., Jiao, J., Stewart, R. & Jian, J. Soil properties affect crop yield changes under conservation agriculture: A systematic analysis. European Journal of Soil Science 74 , e13413 (2023). Leuthold, S. J., Wendroth, O., Salmerón, M. & Poffenbarger, H. Weather-dependent relationships between topographic variables and yield of maize and soybean. Field Crops Research 276 , 108368 (2022). Kumhálová, J., Kumhála, F., Matějková, Š. & Kroulík, M. The relationship between topography and yield in different weather conditions. Precision Agriculture , 606-616 (2011). Amatulli, G. et al. A suite of global, cross-scale topographic variables for environmental and biodiversity modeling. Scientific data 5 , 1-15 (2018). Iwahashi, J., Kamiya, I., Matsuoka, M. & Yamazaki, D. Global terrain classification using 280 m DEMs: segmentation, clustering, and reclassification. Progress in Earth and Planetary Science 5 , 1-31 (2018). Hackländer, J. et al. Land potential assessment and trend-analysis using 2000–2021 FAPAR monthly time-series at 250 m spatial resolution. PeerJ 12 , e16972 (2024). Sainju, U. M. & Alasinrin, S. Y. Changes in soil chemical properties and crop yields with long‐term cropping system and nitrogen fertilization. Agrosystems, Geosciences & Environment 3 , e20019 (2020). Sainju, U. M., Liptzin, D. & Jabro, J. D. Relating soil physical properties to other soil properties and crop yields. Scientific Reports 12 , 22025 (2022). Poggio, L. et al. SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty. Soil 7 , 217-240 (2021). McDowell, R., Noble, A., Pletnyakov, P. & Haygarth, P. A Global Database of Soil Plant Available Phosphorus. Scientific Data 10 , 125 (2023). Abatzoglou, J. T., Dobrowski, S. Z., Parks, S. A. & Hegewisch, K. C. TerraClimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958–2015. Scientific Data 5 , 1-12 (2018). Kuradusenge, M. et al. Crop yield prediction using machine learning models: Case of Irish potato and maize. Agriculture 13 , 225 (2023). Wang, Q., Huang, K., Liu, H. & Yu, Y. Factors affecting crop production water footprint: A review and meta-analysis. Sustainable Production and Consumption 36 , 207-216 (2023). Grigorieva, E., Livenets, A. & Stelmakh, E. Adaptation of agriculture to climate change: A scoping review. Climate 11 , 202 (2023). Karger, D. N. et al. Climatologies at high resolution for the earth’s land surface areas. Scientific Data 4 , 1-20 (2017). Zomer, R. J., Xu, J. & Trabucco, A. Version 3 of the global aridity index and potential evapotranspiration database. Scientific Data 9 , 409 (2022). Ahvo, A. et al. Agricultural input shocks affect crop yields more in the high-yielding areas of the world. Nature Food 4 , 1037-1046 (2023). Mujić, N. & Karabegović, A. in 2023 18th Conference on Computer Science and Intelligence Systems (FedCSIS). 285-290 (IEEE). Shammi, S. A. & Meng, Q. Use time series NDVI and EVI to develop dynamic crop growth metrics for yield modeling. Ecological Indicators 121 , 107124 (2021). Liu, J. et al. Crop yield estimation in the Canadian Prairies using Terra/MODIS-derived crop metrics. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 13 , 2685-2697 (2020). Jhajharia, K. & Mathur, P. Prediction of crop yield using satellite vegetation indices combined with machine learning approaches. Advances in Space Research 72 , 3998-4007 (2023). Becker-Reshef, I., Justice, C., Whitcraft, A. K. & Jarvis, I. in IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium. 8155-8157 (IEEE). Hengl, T. et al. African soil properties and nutrients mapped at 30 m spatial resolution using two-scale ensemble machine learning. Scientific Reports 11 , 6130 (2021). Meyer, H., Reudenbach, C., Hengl, T., Katurji, M. & Nauss, T. Improving performance of spatio-temporal machine learning models using forward feature selection and target-oriented validation. Environmental Modelling & Software 101 , 1-9 (2018). Yadav, S. & Shukla, S. in 2016 IEEE 6th International conference on advanced computing (IACC). 78-83 (IEEE). Zhu, M. et al. The role of topography in shaping the spatial patterns of soil organic carbon. Catena 176 , 296-305 (2019). Kuhn, M. Package “caret”: Classification and Regression Training. R package version 6.0-91. (2022). Meinshausen, N. & Ridgeway, G. Quantile regression forests. Journal of Machine Learning Research 7 (2006). Yu, Q. et al. A cultivated planet in 2010–Part 2: The global gridded agricultural-production maps. Earth System Science Data 12 , 3545-3572 (2020). Lawrence, I. & Lin, K. A concordance correlation coefficient to evaluate reproducibility. 255-268 (1989). Shapley, L. in Contributions to the Theory of Games of Annals of Mathematics Studies Vol. 28 (eds K. Harold William & Albert William) 31-40 (Princeton University Press, Princeton, 1953). Lundberg, S. M. & Lee, S.-I. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems 30 (2017). Greenwell, B. Package “fastshap: fast approximate Shapley values.”, R package version 0.0. 5. (2021). Molnar, C. Interpretable machine learning . (Lulu. com, 2020). de la Cruz, V. Y. V., Cheng, W. & Tawaraya, K. Yield gap between organic and conventional farming systems across climate types and sub-types: A meta-analysis. Agricultural Systems 211 , 103732 (2023). Feliciano, D., Ledo, A., Hillier, J. & Nayak, D. R. Which agroforestry options give the greatest soil and above ground carbon benefits in different world regions? Agriculture, Ecosystems & Environment 254 , 117-129 (2018). Poeplau, C. & Don, A. Carbon sequestration in agricultural soils via cultivation of cover crops–A meta-analysis. Agriculture, Ecosystems & Environment 200 , 33-41 (2015). Friedrich, T., Derpsch, R. & Kassam, A. Overview of the global spread of conservation agriculture. Field actions science reports. The journal of Field Actions (2012). Ogle, S. M. et al. Climate and soil characteristics determine where no-till management can store carbon in soils and mitigate greenhouse gas emissions. Scientific Reports 9 , 11665 (2019). Díaz‐Zorita, M., Grove, J. H., Murdock, L., Herbeck, J. & Perfect, E. Soil structural disturbance effects on crop yields and soil properties in a no‐till production system. Agronomy Journal 96 , 1651-1659 (2004). Hansen, N. C., Allen, B. L., Baumhardt, R. L. & Lyon, D. J. Research achievements and adoption of no-till, dryland cropping in the semi-arid US Great Plains. Field Crops Research 132 , 196-203 (2012). Kuyah, S. et al. Agroforestry delivers a win-win solution for ecosystem services in sub-Saharan Africa. A meta-analysis. Agronomy for Sustainable Development 39 , 1-18 (2019). Sprenkle-Hyppolite, S., Griscom, B., Griffey, V., Munshi, E. & Chapman, M. Maximizing tree carbon in croplands and grazing lands while sustaining yields. Carbon Balance and Management 19 , 23 (2024). Nair, P. R. et al. Global distribution of agroforestry systems. An introduction to agroforestry: four decades of scientific developments , 45-58 (2021). Rusinamhodzi, L. et al. A meta-analysis of long-term effects of conservation agriculture on maize grain yield under rain-fed conditions. Agronomy for Sustainable Development 31 , 657-673 (2011). Li, R. et al. Potential mechanisms of maize yield reduction under short-term no-tillage combined with residue coverage in the semi-humid region of Northeast China. Soil and Tillage Research 217 , 105289 (2022). Niggli, U. Sustainability of organic food production: challenges and innovations. Proceedings of the Nutrition Society 74 , 83-88 (2015). Czarnecka, D., Czubacka, A., Agacka-Mołdoch, M., Trojak-Goluch, A. & Księżak, J. The occurrence of fungal diseases in maize in organic farming versus an integrated management system. Agronomy 12 , 558 (2022). Van Eerd, L. L., Chahal, I., Peng, Y. & Awrey, J. C. Influence of cover crops at the four spheres: A review of ecosystem services, potential barriers, and future directions for North America. Science of the Total Environment 858 , 159990 (2023). Qiu, T. et al. Optimizing cover crop practices as a sustainable solution for global agroecosystem services. Nature Communications 15 , 1-14 (2024). Osipitan, O. A. et al. Impact of cover crop management on level of weed suppression: a meta‐analysis. Crop Science 59 , 833-842 (2019). Barbieri, P., Starck, T., Voisin, A.-S. & Nesme, T. Biological nitrogen fixation of legumes crops under organic farming as driven by cropping management: A review. Agricultural Systems 205 , 103579 (2023). Willer, H., Schlatter, B. & Trávníček, J. The world of organic agriculture. Statistics and emerging trends 2023. Research Institute of Organic Agriculture (FIBL) (2023). Crystal-Ornelas, R., Thapa, R. & Tully, K. L. Soil organic carbon is affected by organic amendments, conservation tillage, and cover cropping in organic farming systems: A meta-analysis. Agriculture, Ecosystems & Environment 312 , 107356 (2021). Lynch, D. H. Soil health and biodiversity is driven by intensity of organic farming in Canada. Frontiers in Sustainable Food Systems 6 , 826486 (2022). Sánchez de Cima, D., Luik, A. & Reintam, E. Organic farming and cover crops as an alternative to mineral fertilizers to improve soil physical properties. International Agrophysics 29 (2015). Vincent-Caboud, L., Peigné, J., Casagrande, M. & Silva, E. M. Overview of organic cover crop-based no-tillage technique in Europe: Farmers’ practices and research challenges. Agriculture 7 , 42 (2017). Kisaka, M. O. et al. Integrating no-tillage with agroforestry augments soil quality indicators in Kenya’s dry-land agroecosystems. Soil and Tillage Research 227 , 105586 (2023). Octavia, D. et al. Smart agroforestry for sustaining soil fertility and community livelihood. Forest Science and Technology 19 , 315-328 (2023). Beach, H. M., Laing, K. W., Walle, M. V. D. & Martin, R. C. The current state and future directions of organic no-till farming with cover crops in Canada, with case study support. Sustainability 10 , 373 (2018). Dhakal, M. et al. Cover crop management strategies affect weeds and profitability of organic no-till soybean. Renewable Agriculture and Food Systems 39 , e3 (2024). Silva, E. M. Screening five fall-sown cover crops for use in organic no-till crop production in the Upper Midwest. Agroecology and Sustainable Food Systems 38 , 748-763 (2014). Ogieriakhi, M. O. & Woodward, R. T. Understanding why farmers adopt soil conservation tillage: A systematic review. Soil Security 9 , 100077 (2022). Agyekum, T. P., Antwi‐Agyei, P., Dougill, A. J. & Stringer, L. C. Benefits and barriers to the adoption of climate‐smart agriculture practices in West Africa: A systematic review. Climate Resilience and Sustainability 3 , e279 (2024). Su, Y., Gabrielle, B., Beillouin, D. & Makowski, D. High probability of yield gain through conservation agriculture in dry regions for major staple crops. Scientific Reports 11 , 3344 (2021). Porwollik, V., Rolinski, S., Heinke, J. & Müller, C. Generating a rule-based global gridded tillage dataset. Earth System Science Data 11 , 823-843 (2019). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 27 Jul, 2025 Reviews received at journal 27 Jul, 2025 Reviewers agreed at journal 08 Jul, 2025 Reviewers agreed at journal 06 Jul, 2025 Reviews received at journal 24 Jun, 2025 Reviewers agreed at journal 02 Jun, 2025 Reviewers invited by journal 23 Apr, 2025 Editor assigned by journal 09 Apr, 2025 Submission checks completed at journal 09 Apr, 2025 First submitted to journal 09 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6409921","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":446686282,"identity":"d214c55a-37c9-4ca0-92ac-35e02d6e4f07","order_by":0,"name":"Kpade O. L. Hounkpatin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEklEQVRIie2SsWrDMBCGz0smE60nXNJXUDGEQvIw9iIvpRS6eAoKAU0Br32MhEJnGYG9+AFcmsGl0C4Z0s2BDpWTEOgg07FQfSDBL+67k0AADsefxBNmU2YNuoRAzFEXyeB3SoRATVYmU9k/6qwAMHVShKWWED3/bGEzYi+SN/v2+jYsdaN3KeDQolCRLwIf3kO2KcqrZYT344ozpSpA21tYPhcBgI5XdSLRjzB+UhFTuYSZVdHeYt+eFPpllMfsY9cp9imFZ5ofFF4E3ZQV3rBehS49OfGZDmnNk/CCY/xQb+9UVaFVIaR8e25TPRrWfPy6nc7iLEvWTZpO8VJYnOP1DvuPtthXf8b+QxwOh+N/8w0wTGHXaiqExgAAAABJRU5ErkJggg==","orcid":"","institution":"Aalto University","correspondingAuthor":true,"prefix":"","firstName":"Kpade","middleName":"O. L.","lastName":"Hounkpatin","suffix":""},{"id":446686283,"identity":"7384f166-e460-44ea-8688-d2a4e64b50cb","order_by":1,"name":"Emanuela De Giorgi","email":"","orcid":"","institution":"Politecnico di Torino","correspondingAuthor":false,"prefix":"","firstName":"Emanuela","middleName":"","lastName":"De Giorgi","suffix":""},{"id":446686284,"identity":"14a5df31-1c05-4690-9dd1-29eceb846072","order_by":2,"name":"Mika Jalava","email":"","orcid":"","institution":"Aalto University","correspondingAuthor":false,"prefix":"","firstName":"Mika","middleName":"","lastName":"Jalava","suffix":""},{"id":446686285,"identity":"03d1bb32-4803-4e28-a08e-d08989bc54b0","order_by":3,"name":"Jeroen Poelert","email":"","orcid":"","institution":"Aalto University","correspondingAuthor":false,"prefix":"","firstName":"Jeroen","middleName":"","lastName":"Poelert","suffix":""},{"id":446686286,"identity":"48160786-1b9a-4d76-9f26-bf69a4e4b7da","order_by":4,"name":"Paul C. West","email":"","orcid":"","institution":"University of Minnesota","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"C.","lastName":"West","suffix":""},{"id":446686287,"identity":"01403ce2-632f-43ef-9501-9300c40ca9e4","order_by":5,"name":"Matti Kummu","email":"","orcid":"","institution":"Aalto University","correspondingAuthor":false,"prefix":"","firstName":"Matti","middleName":"","lastName":"Kummu","suffix":""}],"badges":[],"createdAt":"2025-04-09 08:53:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6409921/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6409921/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82166228,"identity":"b14bae94-45ce-4385-bf3b-2265e4ddc5a7","added_by":"auto","created_at":"2025-05-07 09:12:18","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":49684,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal distribution of the study sites. (Sources of the studies: Xu et al.\u003csup\u003e21\u003c/sup\u003e (NT), Jian et al.\u003csup\u003e6\u003c/sup\u003e (AF, CC, NT, OF), Pittelkow et al.\u003csup\u003e26\u003c/sup\u003e (NT), Ding et al.\u003csup\u003e34\u003c/sup\u003e (OF), Felix et al.\u003csup\u003e35\u003c/sup\u003e (AF)).\u0026nbsp;\u0026nbsp;\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/d92f1eae3e0ed82603167271.jpg"},{"id":82166245,"identity":"09db49a1-ae6a-4145-9e47-ae4735faec6d","added_by":"auto","created_at":"2025-05-07 09:12:18","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":90347,"visible":true,"origin":"","legend":"\u003cp\u003eGeneral data processing workflow for modelling and predicting the management types and related most important features\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/00c179809dee4877444a3aa8.jpg"},{"id":82166268,"identity":"74bbec7a-00ae-4006-9ad0-6741d9a92ec2","added_by":"auto","created_at":"2025-05-07 09:12:19","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":81270,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial distribution of the effect size and related uncertainty for the regenerative agriculture practices.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/69a79f3d9d176ab5d9160ff2.jpg"},{"id":82167080,"identity":"ba879239-0904-48c2-a1d8-138154028b29","added_by":"auto","created_at":"2025-05-07 09:20:18","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":305606,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal variable importance for the different regenerative agriculture practices. bio17: Precipitation of Driest Quarter, bio19: Precipitation of Coldest Quarter, bs: bare soil, sinasp: sine of aspect, evi: Enhanced vegetation index., sdpcp: Standard deviation Annual Precipitation, indxShan: Shannon index of geomorphological landform, wrb: world reference based soil types.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/19971ac6e974f5592f57c7e0.jpg"},{"id":82166262,"identity":"959d8524-c34f-4d40-b05a-a15107d1c247","added_by":"auto","created_at":"2025-05-07 09:12:19","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":255271,"visible":true,"origin":"","legend":"\u003cp\u003eMost suitable regenerative farming practices (RFP) for the global dataset without discriminating between crops. AF: Agroforestry, CC: Cover crop, NT: No-tillage, OF: Organic farming. Multiple RFP (AF,CC; CC,AF; NT,CC,AF) mean that each has positive yield at a given pixel location with ranking based on increasing level of uncertainty (see Supplementary Figure 1).\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/7722c31e43049b3b21ac1cb0.jpg"},{"id":82166265,"identity":"76100e8b-ab00-408b-a679-1995cf8c478f","added_by":"auto","created_at":"2025-05-07 09:12:19","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":317482,"visible":true,"origin":"","legend":"\u003cp\u003ePotential suitability areas for Regenerative Agriculture Practices (RFP) across different crop groups. AF: Agroforestry, CC: Cover crop, NT: No-tillage, OF: Organic farming. Multiple practices (AF,CC; CC,AF; CC,OF; NT,AF; OF,AF; OF,CC; OF,AF,CC; NT,OF,CC) mean that each of the RFP has positive yield at a given pixel location with ranking based on increasing level of uncertainty (see Supplementary Figure 1).\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/5182674cf533d9e705478a26.jpg"},{"id":82168787,"identity":"46e3bf9c-48fd-46de-8318-39a3c7181613","added_by":"auto","created_at":"2025-05-07 09:36:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2268527,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6409921/v1/636c4263-030d-4142-9bf8-7399438f1e5a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Targeting regenerative farming practices to increase crop yields globally","fulltext":[{"header":"Introduction","content":"\u003cp\u003eFood demand will increase globally over the coming decades as the global population is projected to surpass nine billion by 2050\u003csup\u003e1\u003c/sup\u003e. According to some estimates, the growing and increasingly affluent population will require approximately 50\u0026ndash;100% more food in 2050 than is produced today\u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Meanwhile, a recent evaluation of the world\u0026acute;s land by FAO revealed that one-third of soils in the world are degraded and fertile soil is being lost at the rate of 24\u0026nbsp;billion tons of topsoil every year\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. This is typically a result of unsustainable land-use and management practices, such as natural vegetation removal, intensive agricultural operations, under (or over) fertilization, and erosion\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMeeting future food demand and lowering the environmental stressors would require a complete change in paradigm and the adoption of sustainable means of production at a global scale. In that regard, scientists widely agree that humanity cannot sustainably increase its use of land, water and other key resources for food production\u003csup\u003e\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Rather, for many scholars and policy makers, the way forwards lies in sustainable intensification with the adoption of management practices that are more resilient to extreme weather events and that would result in yields being increased without any further conversion of agricultural land and without exacerbating soil degradation \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. This acknowledges the need for integrated soil conservation and nutrient management strategies that will result in a productive agricultural system which restore degraded soils and ecosystems and improve soil quality, while at the same time reduce net anthropogenic emissions and enhances the natural resource base and environment.\u003c/p\u003e \u003cp\u003eMultiple soil conservation and nutrient management strategies that are related to sustainable intensification practices are embodied in the concept of regenerative agriculture. Although the definition for regenerative agriculture is rather broad, soil conservation is considered to be its bottom line\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e with the aim at reducing the negative footprint of farming on the environment while at the same time improving and restoring the agri-environment to a better productive state \u003csup\u003e\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Frequently reported regenerative farming practices (hereafter mentioned as RFP), are no tillage (NT), agroforestry (AF), cover crop (CC) and organic farming (OF)\u003csup\u003e\u003cspan additionalcitationids=\"CR18 CR19\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e due to their potential to improve soil health and increasing soil productivity in the context of sustainable agricultural practices.\u003c/p\u003e \u003cp\u003eIn recent years, several studies that analyzed and summarized the impacts of different RFP on soil properties\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, crop yields\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, soil microbial biomass\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, greenhouse gas emissions\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e are available at global scale. The impacts of RFP on yields were reported to vary with crop types, agricultural management practices, climate zones and geographical areas\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Implementing NT resulted in lower, equal or higher yields compared to conventional tillage\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Similar trends were also reported for OF global analysis reporting 18\u0026ndash;50% reduction in yield especially in North America and Europe\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e while about 16% increase is reported for tropical countries in Africa\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. A global meta-analysis shows that AF resulted in average increase of 7\u0026ndash;16% in crop yield especially in subtropical and tropical zones\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, while in average 2.6% reduction occurred in European field experiments, slightly varying depending on the density and age of the trees\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Findings related to CC suggest up to 14% yield increase especially in coarse soil texture and dryland areas along with the use of leguminous cover crops\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. On the other hand, up to 3% yield reductions were observed under CC especially for cash crops in temperate conditions\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThus, RFP have led to either increase or decrease in crop productivity compared to conventional techniques, with the direction and magnitude of the treatment outcome being dependent on various factors. Consequently, there is no universal \u0026ldquo;rule\u0026rdquo; regarding the outcome of specific RFP, as many variables must be considered, including soil properties, climate, crop type, topography etc. Moreover, existing studies have focused mainly on the variation in productivity for specific RFP without investigating their comparative potential across various factors and determining which management practice(s) could be more beneficial towards yield increase for a specific location.\u003c/p\u003e \u003cp\u003eIdentifying potential geographical locations where the implementation of one or more RFP would lead to an increase in productivity could equip decision makers towards a sustainable informed farming and use of land-based resources. Here, we identify potential areas where RFP maintain or increase crop yields. We conducted a spatial suitability analysis using observations from multiple global meta-analyses and global spatial datasets for climate / bio climate, topography, soil properties, and vegetation productivity. Then, we used a Random Forest model as learning algorithm to perform a regression that links our set of global spatial predictor variables against observations of the RFP effect size. Maps resulting from the upscaling of the effect size (ES) to a global scale and its related uncertainties allowed us to identify where each management practice would potentially be suitable without decreasing the yields or, preferably, even increasing them. Our findings contribute to the global efforts towards sustainable farming which still require empirical information towards the different opportunities embedded into the RFP for increasing global food.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData collection\u003c/h2\u003e \u003cp\u003eThe data used in this study are derived from various meta-analyses across the world (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The main inclusion criteria were that studies ought to: (1) be global focusing on RFP, (2) have a collection of plot level experiments, (3) have records of x, y coordinates of the experiment locations, (4) have quantitative information about control and treatment yields. The search was restricted to meta-analyses focusing on the following RFP: no-tillage (NT), agroforestry (AF), cover-crop (CC) and organic farming (OF). Data were aggregated from: (1) studies used to create the FarmGeek (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.farmgeek.xyz/homehttps://www.farmgeek.xyz/home\u003c/span\u003e\u003cspan address=\"https://www.farmgeek.xyz/homehttps://www.farmgeek.xyz/home\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) platform which synthetized peer-reviewed literature on the outcomes of agricultural management practices and food system interventions, (2) Xu et al.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e and (3) Jian et al.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. The main studies considered from FarmGeek are: Pittelkow et al.\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e (NT), Ding et al.\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e (OF) and Felix et al.\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e (AF). Preliminary quality check was carried out to identify and remove replicates in the individual studies reported in these meta-analyses.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eInitially, studies reported yields for 124 crops, before being grouped here into seven main groups. The most cultivated crops in the world such as maize, wheat, soybean and rice were considered separately while the remaining were classified into other cereals, cash-crops, vegetables \u0026amp; fruits and others (see Supplementary Table\u0026nbsp;1). The ES, i.e. the response ratios (RR) of crop yield to these management systems was calculated as follows: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{ln}\\left(RR\\right)={ln}\\left({X}_{T}/{X}_{C}\\right)\\)\u003c/span\u003e\u003c/span\u003e (1) where X\u003csub\u003eT\u003c/sub\u003e and X\u003csub\u003eC\u003c/sub\u003e are the yield value under treatment (NT, AG, CC, or OF) and control, respectively \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEnvironmental data and feature selection\u003c/h3\u003e\n\u003cp\u003eThe gridded environmental data considered in this study can be divided into four main groups (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e): topography (11 covariates), soil properties (9 covariates), climate / bio-climatic (31 covariates), and land cover (3 covariates). Below each group is briefly introduced and justified.\u003c/p\u003e \u003cp\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eTopography\u003c/span\u003e: Many studies consider topographical variables as key drivers of spatial variability in crop yield as they interact with weather to influence soil temperature and moisture\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. The following topographical variables from the global study of Amatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e were considered: elevation, slope, aspect (cosine and sine), plan curvature, profile curvature, topographic position index, terrain ruggedness as well as the Shannon index of geomorphological landform which is an indicator of the diversity of geomorphological landforms. Additionally, the geomorphological landform grid data produced by Iwahashi, et al. \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e as well the Fraction of Absorbed Photosynthetically Active Radiation by Hackl\u0026auml;nder, et al. \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e were considered.\u003c/p\u003e \u003cp\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eSoil properties\u003c/span\u003e: Crop yields are closely dependent on soil properties which control nutrient movements, aeration, nutrient cycling and root growth\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Global soil properties such as soil texture (sand, silt, clay), bulk density (BD), soil organic carbon (SOC), world reference-based soil types, pH were downloaded from the SoilGrids platform\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. The global stock of soil Olsen phosphorus came from the global study carried out by McDowell et al.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e while the soil moisture for the 2001\u0026ndash;2020 period dataset was gotten from the TerraClimate\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e and the average was computed for prediction.\u003c/p\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eClimate / bio-climatic layers\u003c/span\u003e: Climatic variables have been documented to have a major impact on crop growth and food production worldwide\u003csup\u003e\u003cspan additionalcitationids=\"CR48\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. This study compiled different CHELSA climate and bioclimatic variables covering the 1979\u0026ndash;2013 period \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e (Bio1-Bio19, see Supplementary Table\u0026nbsp;2 for full definition). A multisource (SM2RAIN-ASCAT 2007\u0026ndash;2021, CHELSA Climate and WorldClim) average for monthly precipitation (1 km) was considered in this study along with its standard deviation (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://zenodo.org/records/6458580\u003c/span\u003e\u003cspan address=\"https://zenodo.org/records/6458580\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The long-term averaged monthly mean (2000\u0026ndash;2017) time series and standard deviation of the MODIS land surface temperature (daytime and nighttime) were also used as predictors (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.1420114\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.1420114\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The global aridity index was obtained from the Consortium for Spatial Information\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. In addition, the Growing degree days (GDD) for maize, wheat, rice and soybean used in this study were sourced from Ahvo, et al. \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. Solar radiation was derived from digital elevation model using the SAGA software\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eLand cover\u003c/span\u003e: The Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) are both vegetation indices which are frequently used for crop yield related studies\u003csup\u003e\u003cspan additionalcitationids=\"CR55\" citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. The mean of the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) from the MODIS collections (2000\u0026ndash;2023) were computed using google earth engine. The bare soil data consisting in the fractional area not covered by vegetation was accessed from the spatial data platform (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://eo-data.csiro.au/remotesensing/rapp-help/\u003c/span\u003e\u003cspan address=\"https://eo-data.csiro.au/remotesensing/rapp-help/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) of the National Landcare Regional Partnerships Program\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEnvironmental variables (in bracket are abbreviations).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCovariates\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUnit\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eResolution\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eReferences\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"8\" rowspan=\"9\"\u003e \u003cp\u003e\u003cb\u003eSoil properties\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSand\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSilt\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClay\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBulk density (bd)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ekg/dm\u0026sup3;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWorld reference-based soil types (wrb)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSoil organic carbon (SOC)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eg/kg\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePoggio, et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSoil Olsen phosphorus concentrations (phosphorus)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emg/kg\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1000 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMcDowell et al.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSoil moisture (sm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e200 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAbatzoglou, et al. \u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003e\u003cb\u003eTopography\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCosine of aspect (cosasp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDigital elevation model (dem)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003em\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShannon index of geomorphological landform (indxshan)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePlan curvature (plc)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; m\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProfile curvature (prc)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; m\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSine of aspect (sinasp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSlope geomorphological landform\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTopographic position index (tpi)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTerrain ruggedness(tr)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmatulli, et al. \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFraction of Absorbed Photosynthetically Active Radiation (fapar)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0020\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHackl\u0026auml;nder, et al. \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGeormorphological landform\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIwahashi, et al. \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"9\" rowspan=\"10\"\u003e \u003cp\u003e\u003cb\u003eClimatic /\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003eBio-climatic\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBioclimatic variables (Bio1-Bio19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eKarger, et al. \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGrowing degree days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAhvo, et al. \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTemperature (temp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHengl, et al. \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMODIS land surface temperature daytime (lstd)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHengl, et al. \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStandard deviation of the MODIS land surface temperature (sd lstd)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHengl, et al. \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMODIS land surface temperature night time (lstn)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026deg; C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHengl, et al. \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnnual Precipitation (pcp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHengl, et al. \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStandard deviation Annual Precipitation (sdpcp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHengl, et al. \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSolar radiation (srad)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eW/m\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMujić and Karabegović \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAridity index (aridity)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0083\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eZomer, et al. \u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eVegetation cover\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBare soil (bs)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0020\u003csup\u003e\u0026deg;\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBecker-Reshef, et al. \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNormalized Difference Vegetation Index (ndvi)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e500 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eShammi and Meng \u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnhanced Vegetation Index (evi)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e500 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eShammi and Meng \u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003eData analysis with a machine learning model\u003c/h3\u003e\n\u003cp\u003eThe gridded covariate data were overlayed with the ES data to create the regression matrix (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) for further analysis with the Random Forest which was used as the reference modelling technique for prediction. The basic implementation of Random Forest falls short of considering the spatial context especially when validation strategies ignore spatial autocorrelation in the data. To overcome this, we conducted the RF modelling using the Leave-Location-Out cross-validation (LLOCV), that considers potential spatial autocorrelation. LLOCV involves training models repeatedly by leaving the data from one location or a group of locations out and using the remaining to validate model\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. Modelling was carried out considering different management and crops. Firstly, all the management data were merged and modelling was conducted without accounting for any crops to get the overall trend in the data. Secondly, analysis was carried out considering each management and related crops.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFollowing the principle of parsimony, a feature selection was carried out by: (1) removing highly correlated (\u0026gt;\u0026thinsp;0.70) covariates before modelling, (2) using the forward feature selection (FFS) method that functions in combination with target-oriented performance to detect and remove variables that lead to overfitting\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. The data were split, with 80% of the samples to train the models while 20% were used as an independent validation set (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). For cases where the training observation is less than 200 samples, only a cross validation was carried out to allow the model to learn on the full dataset. There is a general agreement for using cross-validation for training small size data\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Random Forest parameters such as ntree, mtry (number of variables at each split) and the minimum node size (minimum number of observations for each node) can be tuned to improve prediction performance. To enable faster computer processing, the default value was used for the ntree (ntree\u0026thinsp;=\u0026thinsp;500) while the latter two were subjected to fine-tuning using the grid search method. We used the \u0026ldquo;caret\u0026rdquo; R Package \u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e to carry out the different model calibrations based on a three-time repeated 10-fold cross-validation.\u003c/p\u003e \u003cp\u003eThe prediction uncertainties were derived through Quantile Regression Forests (QRF)\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. QRF are a generalization of Random Forest (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). In contrast to Random Forest which keeps only the mean of the observations that fall into each tree node, QRF retain the value of all observations in this node considering thereby the spread of the response variable\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. Consequently, it has the advantage of producing the prediction interval by assessing the distribution of observed response variables at each leaf of the tree. Thus, a 90% uncertainty interval can be computed by considering 0.05 quantile as the lower bound, and the 0.95 quantile as the upper bound. Consequently, the 90% uncertainty maps related to the ES prediction maps were produced and used for assessing the accuracy of the predictions.\u003c/p\u003e \u003cp\u003eSelected gridded covariates were resampled at a resolution of 5 arcmin for the prediction of the ES and related uncertainties. Areas that were not croplands were masked out using the cropland mask produced based on SPAM 2020 crop production data by Yu, et al. \u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. The RF performance (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb) was assessed based on the coefficient of determination (R\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e), Root Mean Squared Error (RMSE) and the Lin\u0026rsquo;s concordance correlation coefficient (LCCC). The LCCC measures both the accuracy and precision of the relationship between the observations and the predictions\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eDeveloping suitability maps\u003c/h3\u003e\n\u003cp\u003eThe ES measures the magnitude of the treatment (NT, CC, AF, OF) impact. Values of ES near zero indicate small effects in terms of absolute magnitude while those far from zero indicate large effects. The assumption is that the RFP would perform better than the conventional agriculture resulting in positive and potentially large ES. It is therefore possible to spot out on a gridded map of ES the direction (locations where a given treatment has positive or negative ES) as well as the magnitude of its impact (ES values). The implementation of a given RFP is potentially suitable for a given location (1) if the predicted ES at that very location is different from zero and positive (\u0026gt;\u0026thinsp;0) and (2) if its corresponding prediction uncertainty is the lowest at that very location compared to the remaining RFP (see example in Supplementary Fig.\u0026nbsp;1). The following steps (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec) were followed in that regards:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ebuild the Random Forest model based on the environmental factors and the observations (ES for a given treatment and a particular crop/crop group)\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ecreate ES map for each RFP\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003emap the uncertainty by predicting the prediction confidence interval\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003ecreate suitability map for each RFP\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003echoose most suitable RFP considering areas with positive ES along with the related uncertainty values (step 3). At a given location, consider the RFP which has positive ES and the lowest uncertainty compared to the remaining RFP.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003erepeat step 5 to consider more than one RFP.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e\n\u003ch3\u003eRelative importance of the predictor variables\u003c/h3\u003e\n\u003cp\u003eWe used the Shapley values (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb) as a metric for assessing variable importance. Shapley values were established in relation to the cooperative game theory proposed by Shapely\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e and recently revised by Lundberg et al\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e. In a cooperative game, \u0026ldquo;players\u0026rdquo; have the possibility to function as a team towards achieving a common goal with the consideration of a fair distribution of payoffs among the members of the coalition. In machine learning, the application of the Shapley value intends to fairly assign credit for a model\u0026rsquo;s output among its input features. Consequently, this theory allocates in the context of this study a contribution value to each feature involved in prediction of the ES. The calculation of the Shapley values was conducted with the \u0026ldquo;fastshap\u0026rdquo; R package\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. In addition, partial dependence plot (PDP) which allows the assessment of whether the relationship between the target and a predictor variable is linear, monotonic or more complex was considered\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe focus in this result section is on presenting: (1) the outcome of the overlay of the yield changes with RFA using the effect size (ES) predictions and related uncertainty maps (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), (2) the variable importance accounting for factors that influenced the ES predictions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), (3) the result of the suitability coverage of the different RFP based on the ES and related prediction uncertainties (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e,\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eSpatial distribution of effect size (ES) and variable importance\u003c/h3\u003e\n\u003cp\u003eAF and CC seem suitable over most of the cropland areas for increasing yield at a global scale (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea,b). However, this potential for increasing yield was associated with a high level of uncertainty for about 96% (see Supplementary Table\u0026nbsp;3) of the cropland area AF. Similarly, about 49% and 38% of the cropland area were associated with high and medium level of uncertainty respectively under CC. On the other hand, NT and especially OF are associated with high occurrences of negative yield change respectively under 64% and 97% of the cropland areas (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec,d; see Supplementary Table\u0026nbsp;3). However, potential areas for increasing yield under NT are scattered all over the globe with much more prevalence in the upper part of North America, Latin America and Caribbean, West and East Africa, South and East Asia.\u003c/p\u003e \u003cp\u003eVariable importance was evaluated to detect the most important predictors affecting the model predictions. They were identified based on the Shapley values which measure the magnitude of their contribution to the final model predictions. This was complemented with the PDP (i.e. partial dependence plot) to visualize how the model predictions change as the value of a particular predictor changes while holding the remaining constant. The analysis of the distribution of the Shapley values revealed that none of the environmental covariates contributed steadily to the increase of the ES over the whole range of their values with occurrences of specific ranges over which their impact on the model output was either low or high as further confirmed by the PDP (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Supplementary Fig.\u0026nbsp;6\u0026ndash;7). The detailed findings in relation to the variable importance are further presented in the following sections.\u003c/p\u003e \u003cp\u003eOverall, the Precipitation of Driest Quarter (bio17) and the aridity index drive the spatial distribution of the ES when considering all management (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). However, the order of importance varies based on the respective RFP as well as the crop group being considered under specific management (Fig, 4, Supplementary Fig.\u0026nbsp;6\u0026ndash;7).\u003c/p\u003e \u003cp\u003eAF was primarily affected by phosphorus and base soil (bs) at a global level (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed) while cropwise (Supplementary Fig.\u0026nbsp;6) climate variables either ranked first or second respectively for maize (Supplementary Fig.\u0026nbsp;6a, bio2: Mean Diurnal temperature Range), other cereal (Supplementary Fig.\u0026nbsp;6b, bio17) and vf \u0026amp; o (Supplementary Fig.\u0026nbsp;6c, bio7: Temperature Annual Range). The key variables influencing the ES under CC are topographical variables such as landform and Shannon index (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eg). Crops under CC are mainly influenced by a combination of soil properties and climate (maize, wheat). For maize (Supplementary Fig.\u0026nbsp;6d) and vegetable, fruits and others (Supplementary Fig.\u0026nbsp;6f) crop groups, the soil moisture and Maximum Temperature of Warmest Month (bio5) as well as the Precipitation of Driest Month (bio14) were respectively identified as the most important variable affecting ES. The soil type (wrb) and silt content were ranked as the key variables for the wheat (Supplementary Fig.\u0026nbsp;6e) under CC.\u003c/p\u003e \u003cp\u003eThe results show that spatial distribution of the ES under NT was mostly associated with climate (Precipitation of Coldest Quarter (bio 19)) and topographical (sinus of aspect (sinasp)) variables (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ej). This trend is also reflected at crop level (Supplementary Fig.\u0026nbsp;6) with most of the crops being primarily influenced by Climate variables especially bio17 for maize (Supplementary Fig.\u0026nbsp;6g), Mean Temperature of Driest Quarter (bio9) for rice (Supplementary Fig.\u0026nbsp;6h), aridity for soybean (Supplementary Fig.\u0026nbsp;6i), standard deviation of annual precipitation (sdpcp) for wheat and other cereal (Supplementary Fig.\u0026nbsp;6j,l), bio7 for cash crop (Supplementary Fig.\u0026nbsp;6k). OF was primarily affected (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003em) by vegetation cover (Enhance vegetation index), climate (sdpcp) and soil properties (soil type (wrb)) with similar trend observed for maize (wrb, slope, Supplementary Fig.\u0026nbsp;6n), wheat (landform, wrb, Supplementary Fig.\u0026nbsp;6o), other cereal (Topographic position index (tpi), bare soil (bs), Supplementary Fig.\u0026nbsp;6p) and vf \u0026amp; o (sinasp, cosasp, soil moisture(sm), Supplementary Fig.\u0026nbsp;6q).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003ePotential suitability areas for regenerative management practice\u003c/h3\u003e\n\u003cp\u003eConsidering the predicted ES maps of the RFP and related uncertainties resulted in the identification (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) of location with the highest productivity potential under the implementation of the related RFP. Any of such identified management is thereafter considered as the most suitable for that given location in comparison with the remaining RFP. Other potential configuration is labelled into the class \u0026ldquo;other_RFP\u0026rdquo; (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e,\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMainly NT (i.e. no-tillage, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,d) stood out among the RFP class with a global area coverage (0.90\u0026nbsp;billion ha, (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) of 30% compared to 7% for AF (i.e. agroforestry, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,b), 5.6% for CC (i.e. Cover Crop, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,c), 1.3% for OF (i.e. Organic Farming, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,e). At a regional scale (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), the highest area coverage for NT is found in North America and South Asia while the lowest are in Latin America \u0026amp; Caribbean and Sub-Saharan Africa. The potential coverage of AF is restricted in regions such as Sub-Saharan Africa and South Asia followed by East Asia \u0026amp; Pacific. CC was found to be potentially more productive in Sub-Saharan Africa, East Asia \u0026amp; Pacific and Latin America \u0026amp; Caribbean.\u003c/p\u003e \u003cp\u003eRegionally, the suitability coverage of the RFP for the different crops varied across crops and regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e; see Supplementary Table\u0026nbsp;4). NT was found to be more suitable for maize and wheat in Middle East \u0026amp; North Africa to Latin America \u0026amp; Caribbean (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea,b). For other cereal crop group, OF seems to be suitable at global scale in all regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec) while for the vegetable, fruits and others crop group NT and AF are more distinct (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed). For the latter crop group, NT is more prevalent at a regional scale in Latin America \u0026amp; Caribbean, North America, Europe \u0026amp; Central Asia.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCoverage areas of the regenerative farming practices (RFP). AF: Agroforestry, CC: Cover crop, NT: No-tillage, OF: Organic farming. Multiple RFP (AF,CC; CC,AF; NT,CC,AF) mean that each has positive yield at a given pixel location with ranking based on increasing level of uncertainty (see Supplementary Fig.\u0026nbsp;1).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNT\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eOF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAF,CC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCC,AF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNT,CC,AF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eother_RFP\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAll regions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e30.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e4.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e45.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e2.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eEast Asia \u0026amp; Pacific\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e28.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e5.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e44.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3.24\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eEurope \u0026amp; Central Asia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e32.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e13.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e44.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e1.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eLatin America \u0026amp; Caribbean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e27.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e55.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.66\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMiddle East \u0026amp; North Africa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e29.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e5.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e46.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3.75\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eNorth America\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e42.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e9.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e38.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.56\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eSouth Asia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e42.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e36.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 10\u003csup\u003e9\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eSub-Saharan Africa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e18.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e48.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e4.83\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eha (x 109)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe study also considered the potential of having more than one management (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e,\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) being suitable for a given location with all RFP having positive ES at that location with the order of ranking based on the magnitude of their respective uncertainty. RFP with the lowest uncertainty would rank first followed by the second which in turn would have lower uncertainty compared to the third for cases with three RFP (see Supplementary Fig.\u0026nbsp;1). Overall, cover crop and agroforestry (CC, AF; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eg) had the highest suitability coverage with about 46% of the total share of the global cropland area (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) while regionally the highest records were in Latin America \u0026amp; Caribbean (55%), Sub-Saharan Africa (48%) and Middle East \u0026amp; North Africa (46%) and the lowest in North America (39%) and South Asia (37%). Next are classes where AF took preeminence over CC (AF, CC; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.f) and where NT ranked first over CC and AF (NT, CC, AF; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eh).\u003c/p\u003e \u003cp\u003eConsidering the occurrences of different of management at the same location for the crop groups shows different pattern for each crop. For maize (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea), three classed emerged in descending area coverage order OF, CC, AF; OF, AF, CC and CC, AF. For wheat (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb), mostly AF took preeminence over CC (AF,CC) while the remaining showed CC ranking first before AF (CC,AF) and OF (CC,OF). For the other cereal crop group (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec), the co-occurence of different RFP did not find expression while for the last crop group, vegetable, fruits and others (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed), OF, CC have better distinct coverage followed by CC, OF and NT, OF, CC. In total, cumulating the total area coverage either individually or together with other practices, it appeared that AF would be more suitable for increasing yields with about 60% of the cropland area followed by cover crops (59%), no-tillage (32%) and organic farming (1.3%).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eSuitability coverage of Regenerative Farming practices\u003c/h2\u003e \u003cp\u003eOur study examines the potential area suitability for different RFP as a strategy for defining geographical positions within agricultural land where crises of soil health, biodiversity, and food security could be addressed as a consequence of the implementation of more sustainable farming approaches. This was investigated by integrating via modelling different environment factors - soil properties, climate, vegetation and topography \u0026ndash; and overlaying the spatial distribution of RFP related relative yield and corresponding uncertainties to establish potential locations for their feasibility. This analysis can serve as a crucial foundation for identifying hotspots where the implementation of one or more RFP would lead to an increase in productivity and thereby provide insight towards informed decision for policymakers and farmers.\u003c/p\u003e \u003cp\u003eAt a global scale, previous studies report NT and OF to be practiced on only about 9%\u003csup\u003e20\u003c/sup\u003e and 1.6%\u003csup\u003e70\u003c/sup\u003e of cultivated land respectively. About 15% of the cropland area is reported to be used for Agroforestry \u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e. Poeplau and Don \u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e reported about 25% of cropland to be suitable for CC. Our results predict a higher coverage potential for NT (30%) while reducing outcome of AF, CC and OF as single class to 7%, 5.6% and 1.3% (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The ES of the four RFP considered in this study were primarily moderated (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) by climate variables while specifically, AF, CC and OF were mostly influenced by soil properties and topographical and vegetation variables. It seems therefore that local environment (soil properties, topography, vegetation cover) seem to prevail over global variable (climate) in the response of crops to the RFP.\u003c/p\u003e \u003cp\u003eOn the one hand, NT took preeminence over CC and AF when considering potential simultaneous occurrences of the RFP while CC came first in most instances as compared to AF (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). This might have to do with the total number and spatial data distribution of each RFP with NT having the highest followed by CC. Consequently, spatial modelling uncertainties were relatively lower for areas where these managements have positive ES compared to the remaining. However, the distribution of the RFP in the current study were region and crop specific and are generally in line with previous studies\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e regarding the areas where these RFP are prevalent though with a higher magnitude for potential implementation. Previous studies found about 45% of the total global area are under NT in South America followed by the US with 32%\u003csup\u003e20\u003c/sup\u003e with the lowest in Europe (1%) and Africa (1%)\u003csup\u003e73\u003c/sup\u003e. Most studies in these regions report higher yield with NT compared to conventional farming in semi-arid and arid regions or drier conditions\u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e,\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e. Such regions are captured in the present study with some locations in South Asia, North America, Middle East and North Africa especially for maize and wheat (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,b). Generally, such a trend in water limited environment for NT is explained by improved infiltration and reduced evaporation resulting in an enhanced soil moisture availability\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe highest ES was observed under AF (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea) globally as well as for different crops (Supplementary Fig.\u0026nbsp;2\u0026ndash;4). This observation aligns with the finding of Ren, et al. \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e who also found the yield increase under AF surpassing that of NT, CC and OF. This result can be attributed to various factors such as improvement in soil properties, increased capacity for erosion control with trunks, roots and litter reducing run-off, mitigation of crop pests and disease as well as better carbon storage\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. For example, a meta-analysis for implementing AF over sub-Saharan Africa found the following: average crop yield increased almost by a factor of 2, soil fertility by a factor of 1.2, control of runoff and soil loss by a factor of 5 to 9, and infiltration by a factor of 3 as compared to the control\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. However, the global ES distribution as found in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea was overoptimistic as few studies also reported negative ES under AF\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. This was better controlled by considering the uncertainties which also appeared to be higher for AF compared to the remaining RFP. Consequently, mostly CC, NT or OF took preeminence when considering the potential co-occurences (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e,7). This resulted in a significant restriction of the suitability coverage for AF mostly into Sub-Saharan Africa, South Asia and East Asia \u0026amp; Pacific (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,b). However, there might actually be higher potential for AF as Sprenkle-Hyppolite, et al. 78 evaluating the opportunities to increase tree cover without reducing yields in cropland found maximum tree cover increase potential of 22% for North America, 21% for Europe \u0026amp; Central Asia, 20% for East Asia \u0026amp; Pacific, 19% to Sub-Saharan Africa, 15% for Latin America \u0026amp; Caribbean and South Asia and 13% for Middle East \u0026amp; North Africa. The resulting tree cover increase map follows closely areas covered by the ES map for AF (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea) although high uncertainties were associated with these predictions. On the other hand, considering the cumulative occurrence of AF either alone or together with other RFP showed that AF would be more suitable for increasing yields with about 60% of the cropland area. Current results also fall within the major ecological regions identified by the FAO for implementing AF such as temperate, mediterranean, arid and semiarid, subhumid tropical (lowland), humid tropical (lowland) and highland\u003csup\u003e\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. Additionally, it was interesting to note AF prevalence also next to cover crop (CC, AF; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eg; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea,b), no-tillage (NT, CC, AF; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eh) or OF (OF, AF, CC; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea) in remaining areas such as North America, Middle East \u0026amp; North Africa, Latin America \u0026amp; Caribbean.\u003c/p\u003e \u003cp\u003eCC was mostly represented when considering different combinations of RFP with about 46% of global coverage in combination with AF with about 36\u0026ndash;55% share at regional scale in Latin America \u0026amp; Caribbean, Sub-Saharan Africa and Middle East \u0026amp; North Africa, North America and South Asia (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eg). Both occupied areas where NT and OF have negative ES. Reduced productivity levels of NT are reported in the context of waterlogging and poor crop establishment, restricted root growth due to compaction, nutrient deficiencies\u003csup\u003e\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e, decreased soil temperature especially for humid environment\u003csup\u003e\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e\u003c/sup\u003e. On the other hand, poor performance occurs often in OF system as a result of challenges related to weed and pests control and nutrients availability especially in nitrogen and phosphoru\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e,\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e. Though CC implementation involves also risks such as competition with the cover crops for water\u003csup\u003e\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e\u003c/sup\u003e, it is reported in different context in improving weed suppression\u003csup\u003e\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e,\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e\u003c/sup\u003e as well as enriching the soil via biological N fixation\u003csup\u003e\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMost production on OF land (1.6% of global cropland\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e) is reported to be taking place in Oceania (47%), Europe (23%), Latin America (13%), Asia (8.5%), Northern America (4.6%) and Africa (3.5%)\u003csup\u003e88\u003c/sup\u003e with the highest share of arable cropland being devoted to cereals (37%). We also found a high potential of OF for other cereals as single management (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec) as well as with maize crop when forming combination with CC and AF (OF, CC, AF; OF, AF, CC; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). With the crop group related to vegetable, fruits and others its potential increase yield was linked with CC (OF, CC; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003eDifferent RFP have potential to lead to positive yield increase at the same locations. Our finding showed that two RFP (All crops: CC,AF; Wheat: AF,CC; other cereal: OF,AF; vegetables, fruits and others: OF, CC) and even three RFP (All crops: NT, CC, AF; Maize: OF, CC, AF; vegetables, fruits and others: NT,OF,CC) showed positive impact on yield in the same location (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e,\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). This possibility might be further explored for promoting the implementation of more than one RFP at the same time during the growing season. A key criterion for implementing such a combination of practices might be the potential of one practice to cancel the limitations of another and thereby possibly further reduce the yield gap. For example, some findings report the inclusion of CC into organic systems as OF suffers often yield reduction because of poor weed control while CC on the other hand has potential to contribute more effectively to weed suppression\u003csup\u003e\u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e89\u003c/span\u003e,\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAlso, CC is introduced into NT (NT, CC) or AF (AF, CC) systems to provide nitrogen, add organic matter, further reduce soil erosion, improve soil structure thereby increasing the intrinsic fertility of the soil\u003csup\u003e\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e91\u003c/span\u003e,\u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e\u003c/sup\u003e. There is evidence that in AF, CC systems nutrient cycling processes involving litterfall from both shrubs or deep rooted trees and cover crops can mobilize nutrients from topsoil or the subsoil and this in turn result in the improvement of crop productivity\u003csup\u003e\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. Similarly, the benefits of integrating AF and NT are acknowledged\u003csup\u003e\u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e93\u003c/span\u003e,\u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e94\u003c/span\u003e\u003c/sup\u003e while combining OF, AF, NT does not receive the same attention in research compared to OF, NT, CC. Findings related to organic no-till system cover crops (OF, NT, CC) revealed 46% and 42% reduction in labour and fuel along with 27% yield reduction in a soybean cropping system\u003csup\u003e\u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e95\u003c/span\u003e\u003c/sup\u003e while about 14% reduction in reduced variable costs are recorded with additional 19% greater net revenue for the same crop\u003csup\u003e\u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e96\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe successes or failures in implementing more than one or two RFP are dependent on different variables such as farmers\u0026acute; knowledge and resources, type and amount of CC in the field along with CC termination strategies\u003csup\u003e\u003cspan additionalcitationids=\"CR96\" citationid=\"CR95\" class=\"CitationRef\"\u003e95\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e97\u003c/span\u003e\u003c/sup\u003e. Consequently, while acknowledging that combining several RFP can have greater benefits to soil health and potentially increase yield than either alone, there is a need to balance costs, practical feasibility and potential local outcomes for additional combination of these practices.\u003c/p\u003e \u003cp\u003eThe main criteria for deciding upon suitability in this study is the ES (i.e. Effect Size). Other focus of RFP such as the reduction of greenhouse gas emissions and the buildup of soil organic matter as well as the protection of biodiversity and reduction of soil erosion were not considered. These non-monetary advantages of RFP are mostly not tangible in the short term and several years of management might be required before any consistent outcome\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Although improved yield is commonly reported as key determinant for the adoption of many RFP\u003csup\u003e\u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e98\u003c/span\u003e,\u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e99\u003c/span\u003e\u003c/sup\u003e, these non-monetary advantages might actually be alternative motivations to implement these practices for farmers and/or decision makers. Further studies might focus on comparative analysis of the four RFP considered in this study regarding their feedback on greenhouse gas emission, biodiversity, soil organic carbon (SOC) and erosion. On the other hand, any farm operation resulting in reduced profits compared to current management practices might face reluctancy towards adoption\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Consequently, although looking beyond suitability into RFP adoption patterns and mechanisms was out of the scope of this study, it will be beneficial that more studies report not only on the ecological and agronomic indicators of the RFP but also on their economic incidences especially in terms of profitability via cost benefit analysis.\u003c/p\u003e \u003cp\u003eAdditionally, the dataset considered in this study does not include the interaction between these approaches to quantitatively establish the yield gain from their association. On the other hand, practical issues might arise regarding the combination of some of these practices. For example, NT still allows pesticides for weed management to improve yields, while OF (very often) includes more tillage but no pesticides. A combination means that no pesticides can be used in no-tillage, which is obviously good for soil health, but most likely decreases possibilities for improved yields. There is therefore a need to capture the variabilities of the impact resulting from the combination of these practices on yield and other soil health indicators as compared to using either one alone. Although some local studies\u003csup\u003e\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e91\u003c/span\u003e,\u003cspan additionalcitationids=\"CR93\" citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e94\u003c/span\u003e\u003c/sup\u003e investigated the outcome of combining different RFP at local scale, there is still a need for synthesis at a global scale.\u003c/p\u003e \u003cp\u003eThe magnitude and direction of the estimates of the ES related to the different RFP were most likely influenced by the specific locations and numbers of the respective studies for a given management. Overall, the locations of the data used in the present study (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) are mostly from Sub-Saharan Africa for AF (99%), North America and Sub-Saharan Africa for CC (85%), Europe and North America for NT (67%), North America and Europe for OF (84%). In addition, regional and onsite properties especially local climates and soil properties were derived from external global data and might not necessarily correspond to local realities. This translated into low correlation with the predictor variables (Supplementary Table\u0026nbsp;5) resulting in a Lin\u0026rsquo;s concordance correlation coefficient between 0.17 and 0.65 (Supplementary Table\u0026nbsp;6). However, the use of the gridded dataset allowed us to scale-up the ES and have insight into the potential locations that would allow yield increase although local validations might be further required.\u003c/p\u003e \u003cp\u003eThis study considered the different RFP as a whole without investigating further into their specific variants. For example, different forms of AF range from alley cropping, forest farming, silvopastoralism, riparian forest buffers to windbreaks with some nuances and different degree of tree coverage in temperate and tropical areas\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e,\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e,\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. On the other hands, NT\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e100\u003c/span\u003e\u003c/sup\u003e, CC\u003csup\u003e\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e,\u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e101\u003c/span\u003e\u003c/sup\u003e and OF\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e89\u003c/span\u003e\u003c/sup\u003e might occur with or without rotation, soil cover, or include different fertilizer input, amendment types, cover crop types, weed and pest management. While future studies might want to account for such details and differentiate between their unique contribution in yield change, we assumed that many of their influences might already be captured in the final yield recorded for the various managements.\u003c/p\u003e \u003cp\u003eThe dataset considered in the present study originate from experiments carried out in controlled settings under the directions of researchers. It is often reported that yield gains reported under such settings might actually be upper bounds estimates of the impact of the given RFP\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Farmers\u0026acute; practices might not measure up to the optimal managements implemented in experimental plots resulting thereby in lower yields. Consequently, impacts of RFP measured under real world farm conditions are necessary to fully grasp their potential towards yield gains and thereby further improve the suitability coverage at both local and global scale.\u003c/p\u003e \u003cp\u003eOur findings have also shown different areas which have potential for yield growth under implementation of single RFP and further suggested the potential for integrating many RFP for reaching the same purpose. We are however aware that the most suitable combination of RFP or preeminence of one of the RFP over the others is highly context-specific and can depend on many different variables, although local environments seem to be more important than climatic variables. The suitability maps provide great guidance for decision makers and policy programs, to find priority areas or hotspot areas of farming combinations that would potentially increase yields while improving sustainable land management. However, because of the data limitations mentioned earlier, local adaptation and validation of current results is critical to prevent implementing certain RFP that might be incompatible with local agricultural conditions and challenges.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study allowed us to evaluate the suitability of different regenerative farming practices at global scale by considering areas where yield increase occurred along with the lowest uncertainty. Our findings confirm a high suitability coverage for no-tillage especially for maize and wheat when considering single management while organic farming was more prominent for other cereal crop groups. Overall, considering the potential of different practices for increasing yield at the same location revealed that cover crop has potential alongside agroforestry, no-tillage and organic farming. Likewise, organic farming as well as no-tillage project high potential for yield increase especially for the crop group related to vegetable, fruits and others crops. The global and crop specific suitability maps of single and combined regenerative farming practices presented in our study offer new perspectives on potential regions that could be subject for local validation towards implementation. They could also serve as potential baseline information source for situation where policy programs require hotspot areas where regenerative farming practices could enhance yields to encourage adoption. In that regard, future research and solutions should aim at investigating further beyond RFP related yield outcome by focusing also on the balance between costs and potential economic impact across different soil, climate, geography, weather, ecology and cultures.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eK.O.L.H and M.K. conceptualized and designed the research idea. K.O.L.H analyzed the data and wrote the main manuscript text. E.D.G conducted some preliminary analysis on some part of the data. E.D.G., M.J., J.P., P.W. and M.K. brought major input to the manuscript. All authors reviewed and commented on the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThis study was funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No. 819202), the Research Council of Finland\u0026rsquo;s Flagship Programme under project Digital Waters (grant no. 359248), Strategic Research Council (SRC) through project \u0026lsquo;Water \u0026amp; Food\u0026rsquo; (grant no. 365512), and Maa- ja vesitekniikan tuki ry.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGerland, P.\u003cem\u003e et al.\u003c/em\u003e World population stabilization unlikely this century. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e346\u003c/strong\u003e, 234-237 (2014).\u003c/li\u003e\n\u003cli\u003eTilman, D., Balzer, C., Hill, J. \u0026amp; Befort, B. L. Global food demand and the sustainable intensification of agriculture. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e108\u003c/strong\u003e, 20260-20264 (2011).\u003c/li\u003e\n\u003cli\u003eFAO. \u003cem\u003eThe state of the world\u0026apos;s land and water resources for food and agriculture: Managing systems at risk\u003c/em\u003e. (Earthscan, 2011).\u003c/li\u003e\n\u003cli\u003eAlexandratos, N. \u0026amp; Bruinsma, J. World agriculture towards 2030/2050: the 2012 revision. (2012).\u003c/li\u003e\n\u003cli\u003eBakker, H. \u003cem\u003eThe world food crisis: Food security in comparative perspective\u003c/em\u003e. (Canadian Scholars Press, 1990).\u003c/li\u003e\n\u003cli\u003eJian, J., Du, X. \u0026amp; Stewart, R. D. A database for global soil health assessment. \u003cem\u003eScientific Data\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 16 (2020).\u003c/li\u003e\n\u003cli\u003eEugenio, N. R., Naidu, R. \u0026amp; Colombo, C. M. Global approaches to assessing, monitoring, mapping, and remedying soil pollution. \u003cem\u003eEnvironmental Monitoring and Assessment\u003c/em\u003e \u003cstrong\u003e192\u003c/strong\u003e, 1-2 (2020).\u003c/li\u003e\n\u003cli\u003eSteffen, W.\u003cem\u003e et al.\u003c/em\u003e Planetary boundaries: Guiding human development on a changing planet. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e347\u003c/strong\u003e, 1259855 (2015).\u003c/li\u003e\n\u003cli\u003eFoley, J. A.\u003cem\u003e et al.\u003c/em\u003e Solutions for a cultivated planet. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e478\u003c/strong\u003e, 337-342 (2011).\u003c/li\u003e\n\u003cli\u003eFader, M., Gerten, D., Krause, M., Lucht, W. \u0026amp; Cramer, W. Spatial decoupling of agricultural production and consumption: quantifying dependences of countries on food imports due to domestic land and water constraints. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 014046 (2013).\u003c/li\u003e\n\u003cli\u003eGarnett, T.\u003cem\u003e et al.\u003c/em\u003e Sustainable intensification in agriculture: premises and policies. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e341\u003c/strong\u003e, 33-34 (2013).\u003c/li\u003e\n\u003cli\u003eJules, P., Camilla, T. \u0026amp; Stella, W. Sustainable intensification in African agriculture. \u003cem\u003eInternational Journal of Agricultural Sustainability\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 5-24 (2011).\u003c/li\u003e\n\u003cli\u003eSchreefel, L., Schulte, R. P., De Boer, I., Schrijver, A. P. \u0026amp; Van Zanten, H. Regenerative agriculture\u0026ndash;the soil is the base. \u003cem\u003eGlobal Food Security\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 100404 (2020).\u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Donoghue, T., Minasny, B. \u0026amp; McBratney, A. Digital Regenerative Agriculture. \u003cem\u003enpj Sustainable Agriculture\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 5 (2024).\u003c/li\u003e\n\u003cli\u003eRehberger, E., West, P. C., Spillane, C. \u0026amp; McKeown, P. C. What climate and environmental benefits of regenerative agriculture practices? an evidence review. \u003cem\u003eEnvironmental Research Communications\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 052001 (2023).\u003c/li\u003e\n\u003cli\u003eSchulte, R. P. \u0026amp; Ditzler, L. in \u003cem\u003eFuture Food Systems\u003c/em\u003e 17-30 (Elsevier, 2024).\u003c/li\u003e\n\u003cli\u003eVan Kessel, C.\u003cem\u003e et al.\u003c/em\u003e Climate, duration, and N placement determine N2O emissions in reduced tillage systems: a meta‐analysis. \u003cem\u003eGlobal Change Biology\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 33-44 (2013).\u003c/li\u003e\n\u003cli\u003eC\u0026aacute;rceles Rodr\u0026iacute;guez, B.\u003cem\u003e et al.\u003c/em\u003e Conservation Agriculture as a Sustainable System for Soil Health: A Review. \u003cem\u003eSoil Systems\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 87 (2022).\u003c/li\u003e\n\u003cli\u003eLi, Y., Li, Z., Cui, S., Jagadamma, S. \u0026amp; Zhang, Q. Residue retention and minimum tillage improve physical environment of the soil in croplands: A global meta-analysis. \u003cem\u003eSoil and Tillage Research\u003c/em\u003e \u003cstrong\u003e194\u003c/strong\u003e, 104292 (2019).\u003c/li\u003e\n\u003cli\u003eBlanco-Canqui, H. \u0026amp; Ruis, S. J. No-tillage and soil physical environment. \u003cem\u003eGeoderma\u003c/em\u003e \u003cstrong\u003e326\u003c/strong\u003e, 164-200 (2018).\u003c/li\u003e\n\u003cli\u003eSu, Y., Gabrielle, B. \u0026amp; Makowski, D. The impact of climate change on the productivity of conservation agriculture. \u003cem\u003eNature Climate Change\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 628-633 (2021).\u003c/li\u003e\n\u003cli\u003eLi, Y., Chang, S. X., Tian, L. \u0026amp; Zhang, Q. Conservation agriculture practices increase soil microbial biomass carbon and nitrogen in agricultural soils: A global meta-analysis. \u003cem\u003eSoil Biology and Biochemistry\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 50-58 (2018).\u003c/li\u003e\n\u003cli\u003eAbdalla, M.\u003cem\u003e et al.\u003c/em\u003e Conservation tillage systems: a review of its consequences for greenhouse gas emissions. \u003cem\u003eSoil Use and Management\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 199-209 (2013).\u003c/li\u003e\n\u003cli\u003eJain, M.\u003cem\u003e et al.\u003c/em\u003e How much can sustainable intensification increase yields across South Asia? A systematic review of the evidence. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 083004 (2020).\u003c/li\u003e\n\u003cli\u003ePittelkow, C. M.\u003cem\u003e et al.\u003c/em\u003e Productivity limits and potentials of the principles of conservation agriculture. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e517\u003c/strong\u003e, 365-368 (2015).\u003c/li\u003e\n\u003cli\u003ePittelkow, C. M.\u003cem\u003e et al.\u003c/em\u003e When does no-till yield more? A global meta-analysis. \u003cem\u003eField Crops Research\u003c/em\u003e \u003cstrong\u003e183\u003c/strong\u003e, 156-168 (2015).\u003c/li\u003e\n\u003cli\u003eChoden, T. \u0026amp; Ghaley, B. B. A portfolio of effective water and soil conservation practices for arable production systems in Europe and North Africa. \u003cem\u003eSustainability\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2726 (2021).\u003c/li\u003e\n\u003cli\u003eHine, R. \u0026amp; Pretty, J. Promoting Production and Trading Opportunities for Organic Agricultural Products in East Africa Capacity Building Study 3. \u003cem\u003eOrganic Agriculture and Food Security in East Africa\u003c/em\u003e, 43-55 (2007).\u003c/li\u003e\n\u003cli\u003eBaier, C., Gross, A., Thevs, N. \u0026amp; Glaser, B. Effects of agroforestry on grain yield of maize (Zea mays L.)\u0026mdash;A global meta-analysis. \u003cem\u003eFrontiers in Sustainable Food Systems\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 1167686 (2023).\u003c/li\u003e\n\u003cli\u003eIvezić, V., Yu, Y. \u0026amp; Werf, W. v. d. Crop yields in European agroforestry systems: a meta-analysis. \u003cem\u003eFrontiers in Sustainable Food Systems\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 606631 (2021).\u003c/li\u003e\n\u003cli\u003ePeng, Y., Wang, L., Jacinthe, P.-A. \u0026amp; Ren, W. Global synthesis of cover crop impacts on main crop yield. \u003cem\u003eField Crops Research\u003c/em\u003e \u003cstrong\u003e310\u003c/strong\u003e, 109343 (2024).\u003c/li\u003e\n\u003cli\u003eClark, A. J., Decker, A. M., Meisinger, J. J. \u0026amp; McIntosh, M. S. Kill date of vetch, rye, and a vetch\u0026mdash;rye mixture: II. Soil moisture and corn yield. \u003cem\u003eAgronomy Journal\u003c/em\u003e \u003cstrong\u003e89\u003c/strong\u003e, 434-441 (1997).\u003c/li\u003e\n\u003cli\u003eThorup-Kristensen, K., Magid, J. \u0026amp; Jensen, L. S. Catch crops and green manures as biological tools in nitrogen management in temperate zones. \u003cem\u003eAdvances in Agronomy\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 227-302 (2003).\u003c/li\u003e\n\u003cli\u003eDing, W.\u003cem\u003e et al.\u003c/em\u003e Improving yield and nitrogen use efficiency through alternative fertilization options for rice in China: A meta-analysis. \u003cem\u003eField Crops Research\u003c/em\u003e \u003cstrong\u003e227\u003c/strong\u003e, 11-18 (2018).\u003c/li\u003e\n\u003cli\u003eF\u0026eacute;lix, G. F., Scholberg, J. M., Clermont-Dauphin, C., Cournac, L. \u0026amp; Tittonell, P. Enhancing agroecosystem productivity with woody perennials in semi-arid West Africa. A meta-analysis. \u003cem\u003eAgronomy for Sustainable Development\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 57 (2018).\u003c/li\u003e\n\u003cli\u003eRen, X., Zou, W., Jiao, J., Stewart, R. \u0026amp; Jian, J. Soil properties affect crop yield changes under conservation agriculture: A systematic analysis. \u003cem\u003eEuropean Journal of Soil Science\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, e13413 (2023).\u003c/li\u003e\n\u003cli\u003eLeuthold, S. J., Wendroth, O., Salmer\u0026oacute;n, M. \u0026amp; Poffenbarger, H. Weather-dependent relationships between topographic variables and yield of maize and soybean. \u003cem\u003eField Crops Research\u003c/em\u003e \u003cstrong\u003e276\u003c/strong\u003e, 108368 (2022).\u003c/li\u003e\n\u003cli\u003eKumh\u0026aacute;lov\u0026aacute;, J., Kumh\u0026aacute;la, F., Matějkov\u0026aacute;, \u0026Scaron;. \u0026amp; Kroul\u0026iacute;k, M. The relationship between topography and yield in different weather conditions. \u003cem\u003ePrecision Agriculture\u003c/em\u003e, 606-616 (2011).\u003c/li\u003e\n\u003cli\u003eAmatulli, G.\u003cem\u003e et al.\u003c/em\u003e A suite of global, cross-scale topographic variables for environmental and biodiversity modeling. \u003cem\u003eScientific data\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1-15 (2018).\u003c/li\u003e\n\u003cli\u003eIwahashi, J., Kamiya, I., Matsuoka, M. \u0026amp; Yamazaki, D. Global terrain classification using 280 m DEMs: segmentation, clustering, and reclassification. \u003cem\u003eProgress in Earth and Planetary Science\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1-31 (2018).\u003c/li\u003e\n\u003cli\u003eHackl\u0026auml;nder, J.\u003cem\u003e et al.\u003c/em\u003e Land potential assessment and trend-analysis using 2000\u0026ndash;2021 FAPAR monthly time-series at 250 m spatial resolution. \u003cem\u003ePeerJ\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e16972 (2024).\u003c/li\u003e\n\u003cli\u003eSainju, U. M. \u0026amp; Alasinrin, S. Y. Changes in soil chemical properties and crop yields with long‐term cropping system and nitrogen fertilization. \u003cem\u003eAgrosystems, Geosciences \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e20019 (2020).\u003c/li\u003e\n\u003cli\u003eSainju, U. M., Liptzin, D. \u0026amp; Jabro, J. D. Relating soil physical properties to other soil properties and crop yields. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 22025 (2022).\u003c/li\u003e\n\u003cli\u003ePoggio, L.\u003cem\u003e et al.\u003c/em\u003e SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty. \u003cem\u003eSoil\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 217-240 (2021).\u003c/li\u003e\n\u003cli\u003eMcDowell, R., Noble, A., Pletnyakov, P. \u0026amp; Haygarth, P. A Global Database of Soil Plant Available Phosphorus. \u003cem\u003eScientific Data\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 125 (2023).\u003c/li\u003e\n\u003cli\u003eAbatzoglou, J. T., Dobrowski, S. Z., Parks, S. A. \u0026amp; Hegewisch, K. C. TerraClimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958\u0026ndash;2015. \u003cem\u003eScientific Data\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1-12 (2018).\u003c/li\u003e\n\u003cli\u003eKuradusenge, M.\u003cem\u003e et al.\u003c/em\u003e Crop yield prediction using machine learning models: Case of Irish potato and maize. \u003cem\u003eAgriculture\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 225 (2023).\u003c/li\u003e\n\u003cli\u003eWang, Q., Huang, K., Liu, H. \u0026amp; Yu, Y. Factors affecting crop production water footprint: A review and meta-analysis. \u003cem\u003eSustainable Production and Consumption\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 207-216 (2023).\u003c/li\u003e\n\u003cli\u003eGrigorieva, E., Livenets, A. \u0026amp; Stelmakh, E. Adaptation of agriculture to climate change: A scoping review. \u003cem\u003eClimate\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 202 (2023).\u003c/li\u003e\n\u003cli\u003eKarger, D. N.\u003cem\u003e et al.\u003c/em\u003e Climatologies at high resolution for the earth\u0026rsquo;s land surface areas. \u003cem\u003eScientific Data\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1-20 (2017).\u003c/li\u003e\n\u003cli\u003eZomer, R. J., Xu, J. \u0026amp; Trabucco, A. Version 3 of the global aridity index and potential evapotranspiration database. \u003cem\u003eScientific Data\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 409 (2022).\u003c/li\u003e\n\u003cli\u003eAhvo, A.\u003cem\u003e et al.\u003c/em\u003e Agricultural input shocks affect crop yields more in the high-yielding areas of the world. \u003cem\u003eNature Food\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1037-1046 (2023).\u003c/li\u003e\n\u003cli\u003eMujić, N. \u0026amp; Karabegović, A. in \u003cem\u003e2023 18th Conference on Computer Science and Intelligence Systems (FedCSIS).\u003c/em\u003e 285-290 (IEEE).\u003c/li\u003e\n\u003cli\u003eShammi, S. A. \u0026amp; Meng, Q. Use time series NDVI and EVI to develop dynamic crop growth metrics for yield modeling. \u003cem\u003eEcological Indicators\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 107124 (2021).\u003c/li\u003e\n\u003cli\u003eLiu, J.\u003cem\u003e et al.\u003c/em\u003e Crop yield estimation in the Canadian Prairies using Terra/MODIS-derived crop metrics. \u003cem\u003eIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2685-2697 (2020).\u003c/li\u003e\n\u003cli\u003eJhajharia, K. \u0026amp; Mathur, P. Prediction of crop yield using satellite vegetation indices combined with machine learning approaches. \u003cem\u003eAdvances in Space Research\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 3998-4007 (2023).\u003c/li\u003e\n\u003cli\u003eBecker-Reshef, I., Justice, C., Whitcraft, A. K. \u0026amp; Jarvis, I. in \u003cem\u003eIGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium.\u003c/em\u003e 8155-8157 (IEEE).\u003c/li\u003e\n\u003cli\u003eHengl, T.\u003cem\u003e et al.\u003c/em\u003e African soil properties and nutrients mapped at 30 m spatial resolution using two-scale ensemble machine learning. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 6130 (2021).\u003c/li\u003e\n\u003cli\u003eMeyer, H., Reudenbach, C., Hengl, T., Katurji, M. \u0026amp; Nauss, T. Improving performance of spatio-temporal machine learning models using forward feature selection and target-oriented validation. \u003cem\u003eEnvironmental Modelling \u0026amp; Software\u003c/em\u003e \u003cstrong\u003e101\u003c/strong\u003e, 1-9 (2018).\u003c/li\u003e\n\u003cli\u003eYadav, S. \u0026amp; Shukla, S. in \u003cem\u003e2016 IEEE 6th International conference on advanced computing (IACC).\u003c/em\u003e 78-83 (IEEE).\u003c/li\u003e\n\u003cli\u003eZhu, M.\u003cem\u003e et al.\u003c/em\u003e The role of topography in shaping the spatial patterns of soil organic carbon. \u003cem\u003eCatena\u003c/em\u003e \u003cstrong\u003e176\u003c/strong\u003e, 296-305 (2019).\u003c/li\u003e\n\u003cli\u003eKuhn, M. Package \u0026ldquo;caret\u0026rdquo;: Classification and Regression Training. R package version 6.0-91. (2022).\u003c/li\u003e\n\u003cli\u003eMeinshausen, N. \u0026amp; Ridgeway, G. Quantile regression forests. \u003cem\u003eJournal of Machine Learning Research\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e (2006).\u003c/li\u003e\n\u003cli\u003eYu, Q.\u003cem\u003e et al.\u003c/em\u003e A cultivated planet in 2010\u0026ndash;Part 2: The global gridded agricultural-production maps. \u003cem\u003eEarth System Science Data\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 3545-3572 (2020).\u003c/li\u003e\n\u003cli\u003eLawrence, I. \u0026amp; Lin, K. A concordance correlation coefficient to evaluate reproducibility. 255-268 (1989).\u003c/li\u003e\n\u003cli\u003eShapley, L. in \u003cem\u003eContributions to the Theory of Games of Annals of Mathematics Studies\u003c/em\u003e Vol. 28 (eds K. Harold William \u0026amp; Albert William) 31-40 (Princeton University Press, Princeton, 1953).\u003c/li\u003e\n\u003cli\u003eLundberg, S. M. \u0026amp; Lee, S.-I. A unified approach to interpreting model predictions. \u003cem\u003eAdvances in Neural Information Processing Systems\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e (2017).\u003c/li\u003e\n\u003cli\u003eGreenwell, B. Package \u0026ldquo;fastshap: fast approximate Shapley values.\u0026rdquo;, R package version 0.0. 5. (2021).\u003c/li\u003e\n\u003cli\u003eMolnar, C. \u003cem\u003eInterpretable machine learning\u003c/em\u003e. (Lulu. com, 2020).\u003c/li\u003e\n\u003cli\u003ede la Cruz, V. Y. V., Cheng, W. \u0026amp; Tawaraya, K. Yield gap between organic and conventional farming systems across climate types and sub-types: A meta-analysis. \u003cem\u003eAgricultural Systems\u003c/em\u003e \u003cstrong\u003e211\u003c/strong\u003e, 103732 (2023).\u003c/li\u003e\n\u003cli\u003eFeliciano, D., Ledo, A., Hillier, J. \u0026amp; Nayak, D. R. Which agroforestry options give the greatest soil and above ground carbon benefits in different world regions? \u003cem\u003eAgriculture, Ecosystems \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e254\u003c/strong\u003e, 117-129 (2018).\u003c/li\u003e\n\u003cli\u003ePoeplau, C. \u0026amp; Don, A. Carbon sequestration in agricultural soils via cultivation of cover crops\u0026ndash;A meta-analysis. \u003cem\u003eAgriculture, Ecosystems \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e200\u003c/strong\u003e, 33-41 (2015).\u003c/li\u003e\n\u003cli\u003eFriedrich, T., Derpsch, R. \u0026amp; Kassam, A. Overview of the global spread of conservation agriculture. \u003cem\u003eField actions science reports. The journal of Field Actions\u003c/em\u003e (2012).\u003c/li\u003e\n\u003cli\u003eOgle, S. M.\u003cem\u003e et al.\u003c/em\u003e Climate and soil characteristics determine where no-till management can store carbon in soils and mitigate greenhouse gas emissions. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 11665 (2019).\u003c/li\u003e\n\u003cli\u003eD\u0026iacute;az‐Zorita, M., Grove, J. H., Murdock, L., Herbeck, J. \u0026amp; Perfect, E. Soil structural disturbance effects on crop yields and soil properties in a no‐till production system. \u003cem\u003eAgronomy Journal\u003c/em\u003e \u003cstrong\u003e96\u003c/strong\u003e, 1651-1659 (2004).\u003c/li\u003e\n\u003cli\u003eHansen, N. C., Allen, B. L., Baumhardt, R. L. \u0026amp; Lyon, D. J. Research achievements and adoption of no-till, dryland cropping in the semi-arid US Great Plains. \u003cem\u003eField Crops Research\u003c/em\u003e \u003cstrong\u003e132\u003c/strong\u003e, 196-203 (2012).\u003c/li\u003e\n\u003cli\u003eKuyah, S.\u003cem\u003e et al.\u003c/em\u003e Agroforestry delivers a win-win solution for ecosystem services in sub-Saharan Africa. A meta-analysis. \u003cem\u003eAgronomy for Sustainable Development\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 1-18 (2019).\u003c/li\u003e\n\u003cli\u003eSprenkle-Hyppolite, S., Griscom, B., Griffey, V., Munshi, E. \u0026amp; Chapman, M. Maximizing tree carbon in croplands and grazing lands while sustaining yields. \u003cem\u003eCarbon Balance and Management\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 23 (2024).\u003c/li\u003e\n\u003cli\u003eNair, P. R.\u003cem\u003e et al.\u003c/em\u003e Global distribution of agroforestry systems. \u003cem\u003eAn introduction to agroforestry: four decades of scientific developments\u003c/em\u003e, 45-58 (2021).\u003c/li\u003e\n\u003cli\u003eRusinamhodzi, L.\u003cem\u003e et al.\u003c/em\u003e A meta-analysis of long-term effects of conservation agriculture on maize grain yield under rain-fed conditions. \u003cem\u003eAgronomy for Sustainable Development\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 657-673 (2011).\u003c/li\u003e\n\u003cli\u003eLi, R.\u003cem\u003e et al.\u003c/em\u003e Potential mechanisms of maize yield reduction under short-term no-tillage combined with residue coverage in the semi-humid region of Northeast China. \u003cem\u003eSoil and Tillage Research\u003c/em\u003e \u003cstrong\u003e217\u003c/strong\u003e, 105289 (2022).\u003c/li\u003e\n\u003cli\u003eNiggli, U. Sustainability of organic food production: challenges and innovations. \u003cem\u003eProceedings of the Nutrition Society\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 83-88 (2015).\u003c/li\u003e\n\u003cli\u003eCzarnecka, D., Czubacka, A., Agacka-Mołdoch, M., Trojak-Goluch, A. \u0026amp; Księżak, J. The occurrence of fungal diseases in maize in organic farming versus an integrated management system. \u003cem\u003eAgronomy\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 558 (2022).\u003c/li\u003e\n\u003cli\u003eVan Eerd, L. L., Chahal, I., Peng, Y. \u0026amp; Awrey, J. C. Influence of cover crops at the four spheres: A review of ecosystem services, potential barriers, and future directions for North America. \u003cem\u003eScience of the Total Environment\u003c/em\u003e \u003cstrong\u003e858\u003c/strong\u003e, 159990 (2023).\u003c/li\u003e\n\u003cli\u003eQiu, T.\u003cem\u003e et al.\u003c/em\u003e Optimizing cover crop practices as a sustainable solution for global agroecosystem services. \u003cem\u003eNature Communications\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1-14 (2024).\u003c/li\u003e\n\u003cli\u003eOsipitan, O. A.\u003cem\u003e et al.\u003c/em\u003e Impact of cover crop management on level of weed suppression: a meta‐analysis. \u003cem\u003eCrop Science\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 833-842 (2019).\u003c/li\u003e\n\u003cli\u003eBarbieri, P., Starck, T., Voisin, A.-S. \u0026amp; Nesme, T. Biological nitrogen fixation of legumes crops under organic farming as driven by cropping management: A review. \u003cem\u003eAgricultural Systems\u003c/em\u003e \u003cstrong\u003e205\u003c/strong\u003e, 103579 (2023).\u003c/li\u003e\n\u003cli\u003eWiller, H., Schlatter, B. \u0026amp; Tr\u0026aacute;vn\u0026iacute;ček, J. The world of organic agriculture. Statistics and emerging trends 2023. \u003cem\u003eResearch Institute of Organic Agriculture (FIBL)\u003c/em\u003e (2023).\u003c/li\u003e\n\u003cli\u003eCrystal-Ornelas, R., Thapa, R. \u0026amp; Tully, K. L. Soil organic carbon is affected by organic amendments, conservation tillage, and cover cropping in organic farming systems: A meta-analysis. \u003cem\u003eAgriculture, Ecosystems \u0026amp; Environment\u003c/em\u003e \u003cstrong\u003e312\u003c/strong\u003e, 107356 (2021).\u003c/li\u003e\n\u003cli\u003eLynch, D. H. Soil health and biodiversity is driven by intensity of organic farming in Canada. \u003cem\u003eFrontiers in Sustainable Food Systems\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 826486 (2022).\u003c/li\u003e\n\u003cli\u003eS\u0026aacute;nchez de Cima, D., Luik, A. \u0026amp; Reintam, E. Organic farming and cover crops as an alternative to mineral fertilizers to improve soil physical properties. \u003cem\u003eInternational Agrophysics\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e (2015).\u003c/li\u003e\n\u003cli\u003eVincent-Caboud, L., Peign\u0026eacute;, J., Casagrande, M. \u0026amp; Silva, E. M. Overview of organic cover crop-based no-tillage technique in Europe: Farmers\u0026rsquo; practices and research challenges. \u003cem\u003eAgriculture\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 42 (2017).\u003c/li\u003e\n\u003cli\u003eKisaka, M. O.\u003cem\u003e et al.\u003c/em\u003e Integrating no-tillage with agroforestry augments soil quality indicators in Kenya\u0026rsquo;s dry-land agroecosystems. \u003cem\u003eSoil and Tillage Research\u003c/em\u003e \u003cstrong\u003e227\u003c/strong\u003e, 105586 (2023).\u003c/li\u003e\n\u003cli\u003eOctavia, D.\u003cem\u003e et al.\u003c/em\u003e Smart agroforestry for sustaining soil fertility and community livelihood. \u003cem\u003eForest Science and Technology\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 315-328 (2023).\u003c/li\u003e\n\u003cli\u003eBeach, H. M., Laing, K. W., Walle, M. V. D. \u0026amp; Martin, R. C. The current state and future directions of organic no-till farming with cover crops in Canada, with case study support. \u003cem\u003eSustainability\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 373 (2018).\u003c/li\u003e\n\u003cli\u003eDhakal, M.\u003cem\u003e et al.\u003c/em\u003e Cover crop management strategies affect weeds and profitability of organic no-till soybean. \u003cem\u003eRenewable Agriculture and Food Systems\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, e3 (2024).\u003c/li\u003e\n\u003cli\u003eSilva, E. M. Screening five fall-sown cover crops for use in organic no-till crop production in the Upper Midwest. \u003cem\u003eAgroecology and Sustainable Food Systems\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 748-763 (2014).\u003c/li\u003e\n\u003cli\u003eOgieriakhi, M. O. \u0026amp; Woodward, R. T. Understanding why farmers adopt soil conservation tillage: A systematic review. \u003cem\u003eSoil Security\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 100077 (2022).\u003c/li\u003e\n\u003cli\u003eAgyekum, T. P., Antwi‐Agyei, P., Dougill, A. J. \u0026amp; Stringer, L. C. Benefits and barriers to the adoption of climate‐smart agriculture practices in West Africa: A systematic review. \u003cem\u003eClimate Resilience and Sustainability\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e279 (2024).\u003c/li\u003e\n\u003cli\u003eSu, Y., Gabrielle, B., Beillouin, D. \u0026amp; Makowski, D. High probability of yield gain through conservation agriculture in dry regions for major staple crops. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 3344 (2021).\u003c/li\u003e\n\u003cli\u003ePorwollik, V., Rolinski, S., Heinke, J. \u0026amp; M\u0026uuml;ller, C. Generating a rule-based global gridded tillage dataset. \u003cem\u003eEarth System Science Data\u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 823-843 (2019). \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-sustainable-agriculture","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [npj Sustainable Agriculture](https://www.nature.com/npjsustainagric/)","snPcode":"44264","submissionUrl":"https://submission.springernature.com/new-submission/44264/3","title":"npj Sustainable Agriculture","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"regenerative farming approaches, suitability, no-tillage, cover crops, agroforestry, organic farming","lastPublishedDoi":"10.21203/rs.3.rs-6409921/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6409921/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRegenerative farming practices (RFP) such as no-tillage (NT), cover crops (CC), agroforestry (AF), and organic farming (OF) are increasingly being promoted to improve soil health and sustainably increase food production. However, how the suitability and impact of these practices varies across the landscape is unclear. Here, we evaluate the environmental suitability for each of these four practices across the world\u0026rsquo;s croplands and identify areas where these practices could increase crop yields. To achieve this purpose, a Random Forest model was used to estimate and map the relative yield change globally using field-scale experiments from multiple meta-analyses linked with global gridded climate, soil and environmental datasets, at 5 arc-min resolution. Areas with increasing yields varied across practices and regions, ranging from 0.86 to 60% of the potential areas of the cropland. When evaluating the area coverage for various RFP, whether individually or together with other practices, it appeared that AF would be more suitable for increasing yields with about 60% of the cropland area followed by cover crops (59%), no-tillage (32%) and organic farming (1.3%). For possibilities where more than two RFP might potentially be suitable, cover crop occurred more frequently alongside agroforestry (CC, AF), organic farming (OF, CC) and no-tillage (NT, CC, AF). These results highlight how regenerative framing practices\u0026rsquo; impact on yield varies across places and can be used to target policies and actions to have a greater impact on both soil health and food production.\u003c/p\u003e","manuscriptTitle":"Targeting regenerative farming practices to increase crop yields globally","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-07 09:12:12","doi":"10.21203/rs.3.rs-6409921/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-28T00:12:30+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-27T16:22:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"265456021162952747917105456219761062805","date":"2025-07-08T05:01:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"175084406367934557542854930835839876696","date":"2025-07-06T15:16:24+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-24T10:54:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"149991358937141086858682461992993547709","date":"2025-06-02T08:51:42+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-23T05:20:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-04-09T21:50:14+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-09T17:29:31+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Sustainable Agriculture","date":"2025-04-09T08:47:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-sustainable-agriculture","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [npj Sustainable Agriculture](https://www.nature.com/npjsustainagric/)","snPcode":"44264","submissionUrl":"https://submission.springernature.com/new-submission/44264/3","title":"npj Sustainable Agriculture","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"532d2ac5-d333-4faf-9c94-12d1a3a82491","owner":[],"postedDate":"May 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":47544103,"name":"Biological sciences/Ecology/Agri ecology"},{"id":47544104,"name":"Earth and environmental sciences/Environmental sciences/Environmental impact"}],"tags":[],"updatedAt":"2026-01-12T04:54:07+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-07 09:12:12","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6409921","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6409921","identity":"rs-6409921","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}