#1
Top-scoring, highly novel machine learning approach for weather and climate prediction.
Abstract
Machine learning models have shown great success in predicting weather up to two weeks ahead, outperforming process-based benchmarks 1 , 2 , 3 , 4 , 5 . However, most existing approaches do not fully integrate data assimilation into the prediction workflow 6 , 7 , and often suffer from error accumulation in autoregressive computation, limiting their applicability to predictions beyond the medium range 8 and climate projections 9 . We introduce Generative Assimilation and Prediction (GAP), a unified deep generative framework that connects data assimilation, weather forecasting, and climate simulation through a shared probabilistic representation of atmospheric states. GAP learns the climatological distribution from reanalysis data, and adapts it to different tasks by incorporating observational, predictive, and forcing constraints. Evaluated across a range of weather-climate tasks, GAP produces probabilistic weather forecasts competitive with leading forecasting systems, using its own assimilated initial states derived from 5% synthetic perfect observations, with particular skill in capturing extreme events such as atmospheric rivers, heatwaves and the Madden-Julian Oscillation. At seasonal timescales, GAP shows promising predictive skill when constrained by oceanic boundary forcing. Moreover, GAP supports stable millennial-scale climate simulations that reproduce atmospheric variability from daily weather fluctuations to decadal oscillations. These results suggest that generative modeling of the climatological distribution offers a viable foundation for seamless weather and climate prediction, thereby opening a path toward integrated, observation-to-climate workflows.
Machine learning models have shown great success in predicting weather up to two weeks ahead, outperforming process-based benchmarks 1 , 2 , 3 , 4 , 5 . However, most existing approaches do not fully integrate data assimilation into the prediction workflow 6 , 7 , and often suffer from error accumulation in autoregressive computation, limiting their applicability to predictions beyond the medium range 8 and climate projections 9 . We introduce Generative Assimilation and Prediction (GAP), a unified deep generative framework that connects data assimilation, weather forecasting, and climate simulation through a shared probabilistic representation of atmospheric states. GAP learns the climatological distribution from reanalysis data, and adapts it to different tasks by incorporating observational, predictive, and forcing constraints. Evaluated across a range of weather-climate tasks, GAP produces probabilistic weather forecasts competitive with leading forecasting systems, using its own assimilated initial states derived from 5% synthetic perfect observations, with particular skill in capturing extreme events such as atmospheric rivers, heatwaves and the Madden-Julian Oscillation. At seasonal timescales, GAP shows promising predictive skill when constrained by oceanic boundary forcing. Moreover, GAP supports stable millennial-scale climate simulations that reproduce atmospheric variability from daily weather fluctuations to decadal oscillations. These results suggest that generative modeling of the climatological distribution offers a viable foundation for seamless weather and climate prediction, thereby opening a path toward integrated, observation-to-climate workflows.
#2
Landmark paleoclimate study revising understanding of abrupt ocean circulation transitions.
Abstract
Dansgaard-Oeschger (DO) events during the last ice age are thought to reflect rapid shifts in the Atlantic Meridional Overturning Circulation (AMOC). Greenland ice cores suggest warming transitions within only a few decades—or even just over 10 years. Using precisely dated speleothem oxygen-isotope records from the Asian monsoon region, we have shown that the same DO warmings unfolded over more than a century. Isotope modeling revealed that Greenland records primarily capture a rapid atmospheric reorganization triggered by the initial sea-ice retreat in the northeastern North Atlantic, whereas the parallel AMOC recovery and hemispheric thermal adjustment proceeded much more slowly. These findings reveal that natural AMOC transitions evolved more slowly than the rapid North Greenland oxygen-isotope shifts would imply, thereby refining our understanding of “abruptness” during DO events.
Dansgaard-Oeschger (DO) events during the last ice age are thought to reflect rapid shifts in the Atlantic Meridional Overturning Circulation (AMOC). Greenland ice cores suggest warming transitions within only a few decades—or even just over 10 years. Using precisely dated speleothem oxygen-isotope records from the Asian monsoon region, we have shown that the same DO warmings unfolded over more than a century. Isotope modeling revealed that Greenland records primarily capture a rapid atmospheric reorganization triggered by the initial sea-ice retreat in the northeastern North Atlantic, whereas the parallel AMOC recovery and hemispheric thermal adjustment proceeded much more slowly. These findings reveal that natural AMOC transitions evolved more slowly than the rapid North Greenland oxygen-isotope shifts would imply, thereby refining our understanding of “abruptness” during DO events.
#3
Reveals critical feedbacks amplifying extreme heat waves, with implications for climate extremes.
Abstract
The most extreme heat waves are intensifying faster than moderate events across many mid-latitude regions. Climate models underestimate these trends, suggesting an incomplete representation and understanding of the underlying physical processes. Here, we use an explainable machine-learning framework to investigate interactions among atmospheric circulation, surface soil moisture, and temperature, and their role in recent heat extremes. We show that a three-way interaction among these components becomes active once a critical temperature threshold is exceeded, contributing, on average, ∼10% to peak temperature anomalies during the most extreme heat events across ∼28% of humid mid-latitude regions. We further reveal that this interaction is systematically underestimated in CMIP6 and CESM2 Large Ensemble simulations. Soils in climate models are too dry at the initiation of heat waves, therefore suppressing subsequent soil-moisture depletion and inhibiting feedback activation. These findings help explain regional model biases in upper-tail warming and highlight avenues for improving future projections of extreme heat events.
The most extreme heat waves are intensifying faster than moderate events across many mid-latitude regions. Climate models underestimate these trends, suggesting an incomplete representation and understanding of the underlying physical processes. Here, we use an explainable machine-learning framework to investigate interactions among atmospheric circulation, surface soil moisture, and temperature, and their role in recent heat extremes. We show that a three-way interaction among these components becomes active once a critical temperature threshold is exceeded, contributing, on average, ∼10% to peak temperature anomalies during the most extreme heat events across ∼28% of humid mid-latitude regions. We further reveal that this interaction is systematically underestimated in CMIP6 and CESM2 Large Ensemble simulations. Soils in climate models are too dry at the initiation of heat waves, therefore suppressing subsequent soil-moisture depletion and inhibiting feedback activation. These findings help explain regional model biases in upper-tail warming and highlight avenues for improving future projections of extreme heat events.
#4
Innovative analytics for wildfire smoke and air quality, relevant for environmental monitoring.
Abstract
Abstract Wildfire smoke is drastically altering the air quality landscape across the United States (US), forcing environmental agencies to confront frequent exceedances of the National Ambient Air Quality Standards (NAAQS). Currently, the operational workflow for developing an Exceptional Event Demonstration (EED) is a resource-intensive process requiring disparate data portals and specialized software. This makes it challenging for state and federal agencies to demonstrate the relationship between smoke and air pollution. Additionally, few tools exist for health scientists to rigorously assess the PM 2.5 and O 3 impacts of wildfire smoke. To meet this need, we developed “Smokelyze”, an open-access, web-based app. Based on peer-reviewed research, it integrates surface observations, satellite data, meteorological data combined with machine learning (Generalized Additive Models or GAMs) to quantify smoke contributions to PM 2.5 and O 3 at nearly every regulatory monitor in all states. The app features a user-friendly workspace for visualizing the results, with fusion of our GAMs with multiple surface, satellite and transport parameters, including EPA’s AirNow monitoring data, NOAA HMS smoke polygons and fire locations, NASA’s TEMPO and TROPOMI for NO 2 and HCHO retrievals, and on-demand cloud-based HYSPLIT model results. Smokelyze provides an optimized environment for environmental agencies to understand and quantify the degree to which wildfire smoke influences PM 2.5 and O 3 . This understanding can be used to support state efforts to build EEDs. Smokelyze also provides a consistent tool to quantify surface smoke for health researchers, and democratizes complex atmospheric analyses and transforms static academic methodologies into an accessible, rapid verification product.
Abstract Wildfire smoke is drastically altering the air quality landscape across the United States (US), forcing environmental agencies to confront frequent exceedances of the National Ambient Air Quality Standards (NAAQS). Currently, the operational workflow for developing an Exceptional Event Demonstration (EED) is a resource-intensive process requiring disparate data portals and specialized software. This makes it challenging for state and federal agencies to demonstrate the relationship between smoke and air pollution. Additionally, few tools exist for health scientists to rigorously assess the PM 2.5 and O 3 impacts of wildfire smoke. To meet this need, we developed “Smokelyze”, an open-access, web-based app. Based on peer-reviewed research, it integrates surface observations, satellite data, meteorological data combined with machine learning (Generalized Additive Models or GAMs) to quantify smoke contributions to PM 2.5 and O 3 at nearly every regulatory monitor in all states. The app features a user-friendly workspace for visualizing the results, with fusion of our GAMs with multiple surface, satellite and transport parameters, including EPA’s AirNow monitoring data, NOAA HMS smoke polygons and fire locations, NASA’s TEMPO and TROPOMI for NO 2 and HCHO retrievals, and on-demand cloud-based HYSPLIT model results. Smokelyze provides an optimized environment for environmental agencies to understand and quantify the degree to which wildfire smoke influences PM 2.5 and O 3 . This understanding can be used to support state efforts to build EEDs. Smokelyze also provides a consistent tool to quantify surface smoke for health researchers, and democratizes complex atmospheric analyses and transforms static academic methodologies into an accessible, rapid verification product.
#5
Global-scale analysis of drought propagation from meteorological to groundwater, key for water security.
Abstract
Abstract. Groundwater stores a third of all global freshwater and supports water supply, irrigation and ecosystems across the world. As such, groundwater drought can have wide-reaching financial, social and environmental impacts, particularly when drought events are prolonged or multi-year. Although recent work has made significant progress in understanding the drivers and patterns of multi-year meteorological droughts, we do not know how this signal translates into multi-year groundwater drought, where subsurface processes, anthropogenic influences and abstractions can alter the meteorological signal. This is particularly true at the global-scale, where a major barrier to understanding large-scale groundwater drought dynamics is the difficulty of obtaining consistent and comprehensive groundwater data. In this research, we use a new global hyper-resolution (∼1 km) groundwater dataset to investigate the global patterns and drivers of groundwater drought from 1960–2019, with a specific focus on multi-year events. We start by characterizing the propagation of meteorological (represented by SPEI-12) to groundwater drought, evaluating how and to what extent the sub-surface plays a role in modulating the meteorological drought signal. Subsequently, we define three global groundwater response types that provide a framework for understanding the processes and geo-physical drivers of normal versus multi-year groundwater droughts. We find that 35 % of the world has an average groundwater drought duration which is multi-year. In 83 % of these locations, the subsurface extends the meteorological drought signal, whereas in the remaining 17 %, groundwater drought duration appears to be primarily driven by meteorological anomalies. Our analysis offers new insights into global-scale drought exposure by identifying regions which have been most vulnerable to multi-year groundwater drought in the past, as well as those which might be more vulnerable in the future. Importantly, our typology also highlights areas where multi-year groundwater droughts can be anticipated based on meteorological drought anomalies and can therefore inform strategies for managing and mitigating future water scarcity risks.
Abstract. Groundwater stores a third of all global freshwater and supports water supply, irrigation and ecosystems across the world. As such, groundwater drought can have wide-reaching financial, social and environmental impacts, particularly when drought events are prolonged or multi-year. Although recent work has made significant progress in understanding the drivers and patterns of multi-year meteorological droughts, we do not know how this signal translates into multi-year groundwater drought, where subsurface processes, anthropogenic influences and abstractions can alter the meteorological signal. This is particularly true at the global-scale, where a major barrier to understanding large-scale groundwater drought dynamics is the difficulty of obtaining consistent and comprehensive groundwater data. In this research, we use a new global hyper-resolution (∼1 km) groundwater dataset to investigate the global patterns and drivers of groundwater drought from 1960–2019, with a specific focus on multi-year events. We start by characterizing the propagation of meteorological (represented by SPEI-12) to groundwater drought, evaluating how and to what extent the sub-surface plays a role in modulating the meteorological drought signal. Subsequently, we define three global groundwater response types that provide a framework for understanding the processes and geo-physical drivers of normal versus multi-year groundwater droughts. We find that 35 % of the world has an average groundwater drought duration which is multi-year. In 83 % of these locations, the subsurface extends the meteorological drought signal, whereas in the remaining 17 %, groundwater drought duration appears to be primarily driven by meteorological anomalies. Our analysis offers new insights into global-scale drought exposure by identifying regions which have been most vulnerable to multi-year groundwater drought in the past, as well as those which might be more vulnerable in the future. Importantly, our typology also highlights areas where multi-year groundwater droughts can be anticipated based on meteorological drought anomalies and can therefore inform strategies for managing and mitigating future water scarcity risks.
#6
New global dataset of sea surface currents using physics-informed deep learning, valuable for oceanography.
Abstract
Abstract. Surface ocean circulation shapes climate, air-sea exchange, and the transport of heat, carbon, and other tracers, yet most long-term satellite-based global datasets rely primarily on geostrophic balance and therefore miss important ageostrophic motions. Here, we reconstruct global daily sea surface circulation at 0.25° resolution from 1993 to 2023 using a physics-informed deep learning framework that integrates satellite altimetry with momentum constraints from geostrophic balance, wind stress, and nonlinear advection. The resulting dataset is dynamically consistent, with a mean momentum-balance error below 1.5 %, and reduces the median velocity error against independent mooring observations to 0.07 m s⁻¹, compared with 0.20 m s⁻¹ for geostrophic currents and 0.10 m s⁻¹ for Ekman-corrected currents. We show that geostrophic flow sets the large-scale circulation, whereas Ekman and nonlinear contributions are smaller but comparable in magnitude to each other. Although nonlinear advection contributes little to relative vorticity, it strongly shapes surface divergence and the fine-scale structure of eddy kinetic energy. Our results show that nonlinear ageostrophic flow is an essential component of global surface circulation and that neglecting it limits our ability to resolve surface transport and variability on climatically relevant scales.
Abstract. Surface ocean circulation shapes climate, air-sea exchange, and the transport of heat, carbon, and other tracers, yet most long-term satellite-based global datasets rely primarily on geostrophic balance and therefore miss important ageostrophic motions. Here, we reconstruct global daily sea surface circulation at 0.25° resolution from 1993 to 2023 using a physics-informed deep learning framework that integrates satellite altimetry with momentum constraints from geostrophic balance, wind stress, and nonlinear advection. The resulting dataset is dynamically consistent, with a mean momentum-balance error below 1.5 %, and reduces the median velocity error against independent mooring observations to 0.07 m s⁻¹, compared with 0.20 m s⁻¹ for geostrophic currents and 0.10 m s⁻¹ for Ekman-corrected currents. We show that geostrophic flow sets the large-scale circulation, whereas Ekman and nonlinear contributions are smaller but comparable in magnitude to each other. Although nonlinear advection contributes little to relative vorticity, it strongly shapes surface divergence and the fine-scale structure of eddy kinetic energy. Our results show that nonlinear ageostrophic flow is an essential component of global surface circulation and that neglecting it limits our ability to resolve surface transport and variability on climatically relevant scales.
#7
Links ancient CO2 levels to Southern Ocean biological pump efficiency, advancing paleoclimate and carbon cycle knowledge.
Abstract
Abstract The Early Eocene Climatic Optimum (EECO) was an interval in Earth’s history when reconstructed temperatures and atmospheric CO2 concentrations were among the highest of the Cenozoic Era (i.e., the last 66 million years). The causes of the high CO2 levels remain unclear. The Southern Ocean, a major “leak” in the global ocean’s biological carbon pump that has been implicated in the CO2 changes over recent ice age cycles, may also have played a role earlier in the Cenozoic. Here, we report foraminifera-bound δ15N measurements that show higher surface nitrate concentrations in the Southern Ocean during peak EECO. The associated venting of CO2 from the deep ocean to the atmosphere would have been amplified by the low seawater pH buffering capacity of the Eocene. Previously proposed low silicate rock weatherability during the Eocene may have weakened the “weathering thermostat”, allowing the reconstructed change in biological pump efficiency in the Southern Ocean to explain at least half of the atmospheric CO2 increase observed during the EECO.
Abstract The Early Eocene Climatic Optimum (EECO) was an interval in Earth’s history when reconstructed temperatures and atmospheric CO2 concentrations were among the highest of the Cenozoic Era (i.e., the last 66 million years). The causes of the high CO2 levels remain unclear. The Southern Ocean, a major “leak” in the global ocean’s biological carbon pump that has been implicated in the CO2 changes over recent ice age cycles, may also have played a role earlier in the Cenozoic. Here, we report foraminifera-bound δ15N measurements that show higher surface nitrate concentrations in the Southern Ocean during peak EECO. The associated venting of CO2 from the deep ocean to the atmosphere would have been amplified by the low seawater pH buffering capacity of the Eocene. Previously proposed low silicate rock weatherability during the Eocene may have weakened the “weathering thermostat”, allowing the reconstructed change in biological pump efficiency in the Southern Ocean to explain at least half of the atmospheric CO2 increase observed during the EECO.
#8
Integrates aerosol-cloud interactions with land-atmosphere coupling, improving cloud microphysics modeling.
Abstract
Abstract Clouds strongly impact net surface radiation, creating differential heating patterns, and trigger land breezes that in turn organize cloud development. These dial processes determine the location of persistent ecohydrological regimes at local scales. Aerosol–cloud interactions (ACI) directly impact the diurnal cycle of cloud optical depth, and therefore the surface energy budget; yet they are generally not explicitly represented in studies of land‐atmosphere interactions. To address this gap, a 3D Large‐Eddy Simulation model was modified to describe ACI in the gray zone. The goal is to simulate realistic cumulus cloud dynamics consistent with the spatial dynamics of land surface moisture and temperature, and associated boundary layer processes. The model is tested and evaluated using simulations at hectometer resolution with prescribed diurnal surface heating corresponding to idealized wet (low Bowen ratio [LBR]) and dry (high Bowen ratio [HBR]) surface boundary conditions. Expected cloud sensitivity is well captured: compared to LBR, HBR clouds have a significantly higher cloud base (about 1 km), up to 50% reduction in total cloud fraction, and cloudiness ends earlier by about 2 hr due to enhanced turbulence and lower moisture availability. Analyses of the evolution of 3D cloud drop size distributions show that the model captures the cloud life cycle, including aerosol activation, condensational growth, enhanced collision‐coalescence, and aerosol regeneration. The results show that resolved eddy‐scale (sub‐km) within‐cloud circulations modulate cloud microphysical structure by locally altering supersaturation at minute timescales. Successive aerosol regeneration yields coarser cloud condensation nuclei, producing a coarse‐aerosol mode, and impacting mid‐day cloud fraction and cloudiness persistence.
Abstract Clouds strongly impact net surface radiation, creating differential heating patterns, and trigger land breezes that in turn organize cloud development. These dial processes determine the location of persistent ecohydrological regimes at local scales. Aerosol–cloud interactions (ACI) directly impact the diurnal cycle of cloud optical depth, and therefore the surface energy budget; yet they are generally not explicitly represented in studies of land‐atmosphere interactions. To address this gap, a 3D Large‐Eddy Simulation model was modified to describe ACI in the gray zone. The goal is to simulate realistic cumulus cloud dynamics consistent with the spatial dynamics of land surface moisture and temperature, and associated boundary layer processes. The model is tested and evaluated using simulations at hectometer resolution with prescribed diurnal surface heating corresponding to idealized wet (low Bowen ratio [LBR]) and dry (high Bowen ratio [HBR]) surface boundary conditions. Expected cloud sensitivity is well captured: compared to LBR, HBR clouds have a significantly higher cloud base (about 1 km), up to 50% reduction in total cloud fraction, and cloudiness ends earlier by about 2 hr due to enhanced turbulence and lower moisture availability. Analyses of the evolution of 3D cloud drop size distributions show that the model captures the cloud life cycle, including aerosol activation, condensational growth, enhanced collision‐coalescence, and aerosol regeneration. The results show that resolved eddy‐scale (sub‐km) within‐cloud circulations modulate cloud microphysical structure by locally altering supersaturation at minute timescales. Successive aerosol regeneration yields coarser cloud condensation nuclei, producing a coarse‐aerosol mode, and impacting mid‐day cloud fraction and cloudiness persistence.
#9
Explores nitrogen cycling and productivity-environment trade-offs in agriculture under climate change.
Abstract
Partial replacement of mineral nitrogen (N) with manure is widely promoted to improve nutrient circularity and reduce N losses, yet its agronomic and environmental outcomes vary widely among sites and seasons. Because manure N becomes plant-available only through microbial decomposition, its performance depends on how release kinetics align with crop N demand under prevailing soil and climatic conditions. Here, we calibrated APSIM Next Generation against a winter wheat–summer maize field experiment in eastern China, explicitly representing manure decomposition and N mineralisation. We then simulated 0–100% substitution of mineral N with sheep manure across contrasting initial soil N, growing-season rainfall, and historical (1991–2020) and future (2041–2060; 27 CMIP6 models) climates. The model reproduced crop phenology (RMSE = 6.45 d), grain yield (R² = 0.79), soil mineral N (R² = 0.70) and cumulative N₂O emissions (R² = 0.73). Under the baseline climate, 15–45% substitution maintained grain yield within 0.6% of mineral fertiliser while reducing greenhouse-gas intensity by 14–41% in wheat and 9–25% in maize. Complete substitution achieved the largest emission reductions but reduced yield by up to 7.3%, increased interannual yield variability and raised the energy footprint. Rainfall most clearly determined the direction of yield–GHG trade-offs and explained 39–51% of the variance in simulated yield, whereas initial soil N mainly shifted response magnitude; GHG variance was dominated by rainfall in wheat (52%) and by substitution rate and climate scenario in maize (40% and 31%). Under SSP5–8.5, net N mineralisation increased by 167% in wheat and 70% in maize, whereas crop N demand changed by −1% and −47%, respectively. This widened the seasonal gap between N release and demand through crop-specific pathways: off-peak mineralisation in wheat, and earlier, truncated demand in maize. By quantifying how N release and crop demand diverge under warming, these results explain why no single replacement ratio is likely to remain optimal across crops and climates. Substitution strategies should therefore be designed for release–demand synchrony, matching the timing of plant-available N supply to crop demand, rather than for a fixed replacement fraction.
Partial replacement of mineral nitrogen (N) with manure is widely promoted to improve nutrient circularity and reduce N losses, yet its agronomic and environmental outcomes vary widely among sites and seasons. Because manure N becomes plant-available only through microbial decomposition, its performance depends on how release kinetics align with crop N demand under prevailing soil and climatic conditions. Here, we calibrated APSIM Next Generation against a winter wheat–summer maize field experiment in eastern China, explicitly representing manure decomposition and N mineralisation. We then simulated 0–100% substitution of mineral N with sheep manure across contrasting initial soil N, growing-season rainfall, and historical (1991–2020) and future (2041–2060; 27 CMIP6 models) climates. The model reproduced crop phenology (RMSE = 6.45 d), grain yield (R² = 0.79), soil mineral N (R² = 0.70) and cumulative N₂O emissions (R² = 0.73). Under the baseline climate, 15–45% substitution maintained grain yield within 0.6% of mineral fertiliser while reducing greenhouse-gas intensity by 14–41% in wheat and 9–25% in maize. Complete substitution achieved the largest emission reductions but reduced yield by up to 7.3%, increased interannual yield variability and raised the energy footprint. Rainfall most clearly determined the direction of yield–GHG trade-offs and explained 39–51% of the variance in simulated yield, whereas initial soil N mainly shifted response magnitude; GHG variance was dominated by rainfall in wheat (52%) and by substitution rate and climate scenario in maize (40% and 31%). Under SSP5–8.5, net N mineralisation increased by 167% in wheat and 70% in maize, whereas crop N demand changed by −1% and −47%, respectively. This widened the seasonal gap between N release and demand through crop-specific pathways: off-peak mineralisation in wheat, and earlier, truncated demand in maize. By quantifying how N release and crop demand diverge under warming, these results explain why no single replacement ratio is likely to remain optimal across crops and climates. Substitution strategies should therefore be designed for release–demand synchrony, matching the timing of plant-available N supply to crop demand, rather than for a fixed replacement fraction.
#10
Assesses global potential of no-tillage for soil carbon enhancement and GHG reduction in agriculture.
Abstract
Abstract Pursuing the balance between environmental sustainability and agricultural productivity remains a critical challenge for achieving global sustainable development goals. No-tillage (NT), as a key measure of sustainable conservation and climate-smart agriculture, offers a promising approach to addressing this challenge. However, its comprehensive effects on soil carbon accumulation, greenhouse gas emissions, and crop yields remain unquantified at the global scale, limiting our ability to optimize both environmental and agronomic outcomes. Here, we conducted such a global synthesis by combining a meta-analysis of 8,303 paired observations from 1,371 studies with a random forest model to assess the influence of site conditions and management on NT effects. Our meta-analysis demonstrates integrated benefits of NT in enhancing soil organic carbon (SOC) stocks in 0–30 cm layer by 11.0% and mitigating greenhouse gas (the sum of CO2, CH4, and N2O in CO2-equivalent) emissions by 13.5%, without a significant impact on crop yields (−0.6% ± 1.5%). The integrated benefits are stronger in drier climates and coarser-textured soils, and they exhibit significant time-dependent patterns. Our prediction identifies 44% of global croplands as being optimal for NT’s integrated benefits, mainly in Africa, South Asia, and North America. Adopting NT in these areas would result in a 4.8% increase in major food-crop yields, a 15.2% increase in 0–30 cm SOC stocks, and a 6.4% reduction in global cropland emissions, respectively. These findings underscore the relevance of implementing NT in suitable regions by adopting it into Good Agricultural Practice Guidelines.
Abstract Pursuing the balance between environmental sustainability and agricultural productivity remains a critical challenge for achieving global sustainable development goals. No-tillage (NT), as a key measure of sustainable conservation and climate-smart agriculture, offers a promising approach to addressing this challenge. However, its comprehensive effects on soil carbon accumulation, greenhouse gas emissions, and crop yields remain unquantified at the global scale, limiting our ability to optimize both environmental and agronomic outcomes. Here, we conducted such a global synthesis by combining a meta-analysis of 8,303 paired observations from 1,371 studies with a random forest model to assess the influence of site conditions and management on NT effects. Our meta-analysis demonstrates integrated benefits of NT in enhancing soil organic carbon (SOC) stocks in 0–30 cm layer by 11.0% and mitigating greenhouse gas (the sum of CO2, CH4, and N2O in CO2-equivalent) emissions by 13.5%, without a significant impact on crop yields (−0.6% ± 1.5%). The integrated benefits are stronger in drier climates and coarser-textured soils, and they exhibit significant time-dependent patterns. Our prediction identifies 44% of global croplands as being optimal for NT’s integrated benefits, mainly in Africa, South Asia, and North America. Adopting NT in these areas would result in a 4.8% increase in major food-crop yields, a 15.2% increase in 0–30 cm SOC stocks, and a 6.4% reduction in global cropland emissions, respectively. These findings underscore the relevance of implementing NT in suitable regions by adopting it into Good Agricultural Practice Guidelines.