A Deep Learning–Based Framework for ESM Climate Downscaling and Its Application to the U.S. Northeast

Abstract. Simulation of the carbon cycle in climate models is important due to its impact on climate change, but many weaknesses in its reproduction were found in previous models. Improvements in the representation of the land carbon cycle in Earth system models (ESMs) participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6) include the interactive treatment of both the carbon and nitrogen cycles, improved photosynthesis, and soil hydrology. To assess the impact of these model developments on aspects of the global carbon cycle, the Earth System Model Evaluation Tool (ESMValTool) is expanded to compare CO2-concentration- and CO2-emission-driven historical simulations from CMIP5 and CMIP6 to observational data sets. A particular focus is on the differences in models with and without an interactive terrestrial nitrogen cycle. Overestimations of photosynthesis (gross primary productivity (GPP)) in CMIP5 were largely resolved in CMIP6 for participating models with an interactive nitrogen cycle but remaining for models without one. This points to the importance of including nutrient limitation. Simulating the leaf area index (LAI) remains challenging, with a large model spread in both CMIP5 and CMIP6. In ESMs, the global mean land carbon uptake (net biome productivity (NBP)) is well reproduced in the CMIP5 and CMIP6 multi-model means. However, this is the result of an underestimation of NBP in the Northern Hemisphere, which is compensated by an overestimation in the Southern Hemisphere and the tropics. Carbon stocks remain a large uncertainty in the models. While vegetation carbon content is slightly better represented in CMIP6, the inter-model range of soil carbon content remains the same between CMIP5 and CMIP6. Overall, a slight improvement in the simulation of land carbon cycle parameters is found in CMIP6 compared to CMIP5, but with many biases remaining, further improvements of models in particular for LAI and NBP is required. Models from modeling groups participating in both CMIP phases generally perform similarly or better in their CMIP6 compared to their CMIP5 models. This improvement is not as significant in the multi-model means due to more new models in CMIP6, especially those using older versions of the Community Land Model (CLM). Emission-driven simulations perform just as well as the concentration-driven models, despite the added process realism. Due to this, we recommend that ESMs in future Coupled Model Intercomparison Project (CMIP) phases perform emission-driven simulations as the standard so that climate–carbon cycle feedbacks are fully active. The inclusion of the nitrogen limitation led to a large improvement in photosynthesis compared to models not including this process, suggesting the need to view the nitrogen cycle as a necessary part of all future carbon cycle models. Possible benefits when including further limiting nutrients such as phosphorus should also be considered.

Deep Learning (DL) has recently emerged as an innovative tool to downscale climate variables from large-scale atmospheric fields under the perfect prognosis (PP) approach. Different Convolutional Neural Networks (CNN) have been applied under present-day conditions with promising results, but little is known about their suitability for extrapolating future climate change conditions. Here, we analyze this problem from a multi-model perspective, developing and evaluating an ensemble of CNN-based downscaled projections (DeepESD) for temperature and precipitation over the European EUR-44i (0.5º) domain, based on eight GCMs from the Coupled Model Intercomparison Project Phase 5 (CMIP5). To our knowledge, this is the first time that CNNs have been used to produce multi-model ensembles of downscaled projections, allowing to quantify inter-model uncertainty in climate change signals. The results are compared with those corresponding to an EUR-44 ensemble of regional climate models (RCMs) showing that DeepESD reduces distributional biases in the historical period. Moreover, the resulting climate change signals are broadly comparable to those obtained with the RCMs, with similar spatial structures. As for the uncertainty of the climate change signal (measured on the basis of inter-model spread), DeepESD yields a smaller uncertainty for precipitation, but a similar uncertainty for temperature. To facilitate further studies of this downscaling approach we follow FAIR principles and make publicly available the code (a Jupyter notebook) and the DeepESD dataset. In particular, DeepESD is published at the Earth System Grid Federation (ESGF), as the first continental-wide PP dataset contributing to CORDEX (EUR-44).

This research evaluated the performance of 19 Coupled Model Intercomparison Project phase 6 (CMIP6) models in simulating the Arctic sea ice from the perspective of annual cycle, spatial pattern and temporal variation. Based on the evaluation, five models with better comprehensive performance capacity were optimized as the ensemble to project the response of the East Asian winter climate to the sea ice-free Arctic occurring under the SSP2-4.5 and SSP5-8.5 scenarios, respectively. The ensemble projections indicate that the sea ice-free Arctic is followed by a weakening of the East Asian winter monsoon, which is characterized with the shallower East Asian trough and weaker East Asian jet stream. Concurrently, the winter surface net radiation flux is projected to increase in the East Asian-western North Pacific region. These changes favour large-scale warming in the East Asian-western North Pacific region. Moreover, the warming is more pronounced under the sea ice-free Arctic of SSP2-4.5 than under that of SSP5-8.5. The winter precipitation tends to increase along the East Asian coast from South China to the Sea of Okhotsk. Such an increase is closely associated with the enhancement of low-level moisture. Due to larger enhancement of moisture, there appears greater increase of precipitation in the monsoon region from South China to Japan under the sea ice-free Arctic of SSP5-8.5 compare to that under SSP2-4.5. Substantial changes are also projected for the temperature and precipitation extremes, such as a general increase in warm days and warm nights and an overall intensification of extreme precipitation in the East Asian-western North Pacific region.

In the era of global warning, the insight of future climate and their changing extremes is critical for climate-vulnerable regions of the world. In this study, we have conducted a robust assessment of Regional Climate Model (RCM) results in a monsoon-dominated region within the new Coupled Model Intercomparison Project Phase 5 (CMIP5) and the latest Representative Concentration Pathways (RCP) scenarios. We have applied an advanced bias correction approach to five RCM simulations in order to project future climate and associated extremes over Bangladesh, a critically climate-vulnerable country with a complex monsoon system. We have also generated a new gridded product that performed better in capturing observed climatic extremes than existing products. The bias-correction approach provided a notable improvement in capturing the precipitation extremes as well as mean climate. The majority of projected multi-model RCMs indicate an increase of rainfall, where one model shows contrary results during the 2080s (2071–2100) era. The multi-model mean shows that nighttime temperatures will increase much faster than daytime temperatures and the average annual temperatures are projected to be as hot as present-day summer temperatures. The expected increase of precipitation and temperature over the hilly areas are higher compared to other parts of the country. Overall, the projected extremities of future rainfall are more variable than temperature. According to the majority of the models, the number of the heavy rainy days will increase in future years. The severity of summer-day temperatures will be alarming, especially over hilly regions, where winters are relatively warm. The projected rise of both precipitation and temperature extremes over the intense rainfall-prone northeastern region of the country creates a possibility of devastating flash floods with harmful impacts on agriculture. Moreover, the effect of bias-correction, as presented in probable changes of both bias-corrected and uncorrected extremes, can be considered in future policy making.

Abstract. The uptake of anthropogenic carbon (Cant) by the ocean leads to ocean acidification, causing the reduction of pH and the saturation states of aragonite (Ωarag) and calcite (Ωcalc). The Arctic Ocean is particularly vulnerable to ocean acidification due to its naturally low pH and saturation states and due to ongoing freshening and the concurrent reduction in total alkalinity in this region. Here, we analyse ocean acidification in the Arctic Ocean over the 21st century across 14 Earth system models (ESMs) from the latest Coupled Model Intercomparison Project Phase 6 (CMIP6). Compared to the previous model generation (CMIP5), models generally better simulate maximum sea surface densities in the Arctic Ocean and consequently the transport of Cant into the Arctic Ocean interior, with simulated historical increases in Cant in improved agreement with observational products. Moreover, in CMIP6 the inter-model uncertainty of projected changes over the 21st century in Arctic Ocean Ωarag and Ωcalc averaged over the upper 1000 m is reduced by 44–64 %. The strong reduction in projection uncertainties of Ωarag and Ωcalc can be attributed to compensation between Cant uptake and total alkalinity reduction in the latest models. Specifically, ESMs with a large increase in Arctic Ocean Cant over the 21st century tend to simulate a relatively weak concurrent freshening and alkalinity reduction, while ESMs with a small increase in Cant simulate a relatively strong freshening and concurrent total alkalinity reduction. Although both mechanisms contribute to Arctic Ocean acidification over the 21st century, the increase in Cant remains the dominant driver. Even under the low-emissions Shared Socioeconomic Pathway 1-2.6 (SSP1-2.6), basin-wide averaged Ωarag undersaturation in the upper 1000 m occurs before the end of the century. While under the high-emissions pathway SSP5-8.5, the Arctic Ocean mesopelagic is projected to even become undersaturated with respect to calcite. An emergent constraint identified in CMIP5 which relates present-day maximum sea surface densities in the Arctic Ocean to the projected end-of-century Arctic Ocean Cant inventory is found to generally hold in CMIP6. However, a coincident constraint on Arctic declines in Ωarag and Ωcalc is not apparent in the new generation of models. This is due to both the reduction in Ωarag and Ωcalc projection uncertainty and the weaker direct relationship between projected changes in Arctic Ocean Cant and changes in Ωarag and Ωcalc.

Considering plateau climate and complex terrain of the upper Yellow River Basin, understanding changes in climate extremes has become increasingly urgent. This study has highlighted the historical changes in climate extremes from 1960 to 2022 based on 20 extreme climate indices, and future changes in climate extremes until 2100 under two Shared Socioeconomic Pathways (SSP126 and SSP585) based on the Coupled Model Intercomparison Project phase 6 (CMIP6) models. We found that historical spatial and temporal evolutions of precipitation extremes (PEs) and temperature extremes (TEs) primarily exhibit increasing trends. The frequency and intensity of PEs primarily show an increasing trend, while the duration of PEs primarily shows a decreasing trend. Both the frequency and duration of cold extremes primarily show a decreasing trend, while the intensity of cold extremes, as well as the intensity, frequency, and duration of warm extremes, primarily show an increasing trend. Future PEs and TEs are expected to continue to intensify even under the most ideal scenario (i.e., SSP126), and these are anticipated to further intensify with increasing radiative forcing levels and greenhouse gas concentrations. Results could provide scientific references for better coping with extreme climate changes in regions of complex terrain and scarce observation station.

In this study, a machine learning model was used to investigate the potential consequences of climate change on vegetation growth. The methodology involved analyzing the historical Normalized Difference Vegetation Index (NDVI) data and future climate projections under four Shared Socioeconomic Pathways (SSPs). Data from the Global Inventory Monitoring and Modeling System (GIMMS) dataset for the period 1981–2000 were used to train the machine learning model, while CMIP6 (Coupled Model Intercomparison Project Phase 6) global climate projections from 2021–2100 were employed to predict future NDVI values under different SSPs. The study results revealed that the global mean NDVI is projected to experience a significant increase from the period 1981–2000 to the period 2021–2040. Following this, the mean NDVI slightly increases under SSP126 and SSP245 while decreasing substantially under SSP370 and SSP585. In the near-term span of 2021–2040, the average NDVI value of SSP585 slightly exceeds that of SSP245 and SSP370, suggesting a positive vegetation development in response to a more pronounced temperature increase in the near term. However, if the trajectory of SSP585 persists, the mean NDVI will commence a decline over the subsequent three periods (2041–2060, 2061–2080, and 2080–2100) with a faster speed than that of SSP370. This decline is attributed to the adverse effects of a rapid temperature rise on vegetation. Based on the examination of individual continents, it is projected that the NDVI values in Africa, South America, and Oceania will decline over time, except under the scenario SSP126 during 2081–2100. On the other hand, the NDVI values in North America and Europe are anticipated to increase, with the exception of the scenario SSP585 during 2081–2100. Additionally, Asia is expected to follow an increasing trend, except under the scenario SSP126 during 2081–2100. In the larger scope, our research findings carry substantial implications for biodiversity preservation, greenhouse gas emission reduction, and efficient environmental management. The utilization of machine learning technology holds the potential to accurately predict future changes in vegetation growth and pinpoint areas where intervention is imperative.

Understanding changes in precipitation extremes is essential for assessing climate variability in vulnerable regions. This study evaluates historical trends (1963–2023), historical simulations (1850–2014), and future projections (2015–2100) of extreme precipitation indices in Rio Grande do Norte (RN), a semi‐arid state in northeast Brazil characterised by high climatic variability. Daily rainfall data from 115 meteorological stations and multi‐model simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6) were analysed under three Shared Socioeconomic Pathway (SSP) scenarios: SSP1‐2.6, SSP2‐4.5, and SSP5‐8.5. Eleven standardised precipitation indices defined by the Expert Team on Climate Change Detection and Indices (ETCCDI) were computed. Model performance was assessed using Taylor diagrams, and temporal trends were analysed using the Mann–Kendall test. As an innovation, a Hydroclimatic Extremes Index (HCEI) was developed to integrate the effects of both prolonged dry periods and intense rainfall events by combining Consecutive Dry Days (CDD) and R95pTOT through Z ‐score normalisation. This metric revealed a significant upward trend in HCEI values across all mesoregions of RN, particularly in the West and East, indicating an increased occurrence of compound hydroclimatic extremes. Peaks in HCEI coincided with major El Niño episodes, suggesting strong links between large‐scale climate drivers and local hydroclimatic dynamics. The results point to a consistent intensification of dry spells and extreme precipitation, especially under the high‐emission SSP5‐8.5 scenario. Despite limitations in spatial resolution, CMIP6 models captured large‐scale spatial patterns and project escalating hydroclimatic extremes throughout the 21st century. These findings offer critical insights for regional adaptation planning and water resource management under a changing climate.

The rapid expansion of Earth system model (ESM) data available from the Coupled Model Intercomparison Project Phase 6 (CMIP6) necessitates new methods to evaluate the performance and suitability of ESMs used for hydroclimate applications as these extremely large data volumes complicate stakeholder efforts to use new ESM outputs in updated climate vulnerability and impact assessments. We develop an analysis framework to inform ESM sub‐selection based on process‐oriented considerations and demonstrate its performance for a regional application in the US Pacific Northwest. First, a suite of global and regional metrics is calculated, using multiple historical observation datasets to assess ESM performance. These metrics are then used to rank CMIP6 models, and a culled ensemble of models is selected using a trend‐related diagnostics approach. This culling strategy does not dramatically change climate scenario trend projections in this region, despite retaining only 20% of the CMIP6 ESMs in the final model ensemble. The reliability of the culled trend projection envelope and model response similarity is also assessed using a perfect model framework. The absolute difference in temperature trend projections is reduced relative to the full ensemble compared to the model for each SSP scenario, while precipitation trend errors are largely unaffected. In addition, we find that the spread of the culled ensemble temperature and precipitation trends includes the trend of the “truth” model ∼83%‐92% of the time. This analysis demonstrates a reliable method to reduce ESM ensemble size that can ease use of ESMs for creating and understanding climate vulnerability and impact assessments.

<strong class="journal-contentHeaderColor">Abstract.</strong> Root-zone soil moisture is a key variable representing water cycle dynamics that strongly interacts with ecohydrological, atmospheric, and biogeochemical processes. Recently, it was proposed as the control variable for the green water planetary boundary, suggesting that widespread and considerable deviations from baseline variability now predispose destabilization of Earth System functions critical to an agriculture-based civilisation. However, the global extent and severity of root-zone soil moisture changes under future scenarios remains to be scrutinized. Here, we analyzed root-zone soil moisture departures from the pre-industrial climate variability for a multi-model ensemble of 14 Earth System Models (ESMs) in the Coupled Model Intercomparison Project Phase 6 (CMIP6) in four climate scenarios as defined by the Shared Socioeconomic Pathways (SSP), SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5, between 2021 and 2100. The analyses were done for 43 ice-free climate reference regions used by the Intergovernmental Panel on Climate Change (IPCC).We defined &lsquo;permanent departures&rsquo; when a region&rsquo;s soil moisture exits the regional variability envelope of the pre-industrial climate and does not fall back into the range covered by the baseline envelope until 2100. Permanent dry departures (i.e. lower soil moisture than pre-industrial variability) were found to be most pronounced in Central America, southern Africa, the Mediterranean region, and most of South America, whereas permanent wet departures are most pronounced in southeastern South America, northern Africa, and southern Asia. In the Mediterranean region, permanent departure may have already happened according to some models. By 2100, there is permanent departure in the Mediterranean in 60 % of the ESMs in SSP1-2.6, the most mitigated situation, and 100 % in SSP3-7.0 and SSP5-8.5, the medium-high and worst-case scenarios. North-Eastern Africa is projected to experience permanent departures in 50 % of the ESMs under SSP1-2.6, and 86 % under SSP5-8.5. The percentage of ice-free land area with departures increases in all SSP scenarios as time goes by.Wet departure are more widespread than dry departures throughout the studied timeframe, except in SSP1-2.6. In most regions, the severity of the departures increases with the severity of global warming. In 2050, permanent departures (ensemble median) occur in 10 % of global ice-free land areas in SSP1-2.6, and in 25 % in SSP3-7.0. By the end of the 21st century, the occurrence of permanent departures in SSP1-2.6 increases to 34 %, and in SSP3-7.0, 45 %. Our findings underscore the importance of mitigation to avoid further degrading the Earth System functions upheld by soil moisture.

<strong class="journal-contentHeaderColor">Abstract.</strong> Root-zone soil moisture is a key variable representing water cycle dynamics that strongly interacts with ecohydrological, atmospheric, and biogeochemical processes. Recently, it was proposed as the control variable for the green water planetary boundary, suggesting that widespread and considerable deviations from baseline variability now predispose destabilization of Earth System functions critical to an agriculture-based civilisation. However, the global extent and severity of root-zone soil moisture changes under future scenarios remains to be scrutinized. Here, we analyzed root-zone soil moisture departures from the pre-industrial climate variability for a multi-model ensemble of 14 Earth System Models (ESMs) in the Coupled Model Intercomparison Project Phase 6 (CMIP6) in four climate scenarios as defined by the Shared Socioeconomic Pathways (SSP), SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5, between 2021 and 2100. The analyses were done for 43 ice-free climate reference regions used by the Intergovernmental Panel on Climate Change (IPCC).We defined &lsquo;permanent departures&rsquo; when a region&rsquo;s soil moisture exits the regional variability envelope of the pre-industrial climate and does not fall back into the range covered by the baseline envelope until 2100. Permanent dry departures (i.e. lower soil moisture than pre-industrial variability) were found to be most pronounced in Central America, southern Africa, the Mediterranean region, and most of South America, whereas permanent wet departures are most pronounced in southeastern South America, northern Africa, and southern Asia. In the Mediterranean region, permanent departure may have already happened according to some models. By 2100, there is permanent departure in the Mediterranean in 60 % of the ESMs in SSP1-2.6, the most mitigated situation, and 100 % in SSP3-7.0 and SSP5-8.5, the medium-high and worst-case scenarios. North-Eastern Africa is projected to experience permanent departures in 50 % of the ESMs under SSP1-2.6, and 86 % under SSP5-8.5. The percentage of ice-free land area with departures increases in all SSP scenarios as time goes by.Wet departure are more widespread than dry departures throughout the studied timeframe, except in SSP1-2.6. In most regions, the severity of the departures increases with the severity of global warming. In 2050, permanent departures (ensemble median) occur in 10 % of global ice-free land areas in SSP1-2.6, and in 25 % in SSP3-7.0. By the end of the 21st century, the occurrence of permanent departures in SSP1-2.6 increases to 34 %, and in SSP3-7.0, 45 %. Our findings underscore the importance of mitigation to avoid further degrading the Earth System functions upheld by soil moisture.

Root-zone soil moisture is a key variable representing water cycle dynamics that strongly interacts with ecohydrological, atmospheric, and biogeochemical processes. Recently, it was proposed as the control variable for the green water planetary boundary, suggesting that widespread and considerable deviations from baseline variability now predispose destabilization of Earth System functions critical to an agriculture-based civilisation. However, the global extent and severity of root-zone soil moisture changes under future scenarios remains to be scrutinized. Here, we analyzed root-zone soil moisture departures from the pre-industrial climate variability for a multi-model ensemble of 14 Earth System Models (ESMs) in the Coupled Model Intercomparison Project Phase 6 (CMIP6) in four climate scenarios as defined by the Shared Socioeconomic Pathways (SSP), SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5, between 2021 and 2100. The analyses were done for 43 ice-free climate reference regions used by the Intergovernmental Panel on Climate Change (IPCC).We defined 'permanent departures' when a region's soil moisture exits the regional variability envelope of the pre-industrial climate and does not fall back into the range covered by the baseline envelope until 2100. Permanent dry departures (i.e. lower soil moisture than pre-industrial variability) were found to be most pronounced in Central America, southern Africa, the Mediterranean region, and most of South America, whereas permanent wet departures are most pronounced in southeastern South America, northern Africa, and southern Asia. In the Mediterranean region, permanent departure may have already happened according to some models. By 2100, there is permanent departure in the Mediterranean in 60 % of the ESMs in SSP1-2.6, the most mitigated situation, and 100 % in SSP3-7.0 and SSP5-8.5, the medium-high and worst-case scenarios. North-Eastern Africa is projected to experience permanent departures in 50 % of the ESMs under SSP1-2.6, and 86 % under SSP5-8.5. The percentage of ice-free land area with departures increases in all SSP scenarios as time goes by.Wet departure are more widespread than dry departures throughout the studied timeframe, except in SSP1-2.6. In most regions, the severity of the departures increases with the severity of global warming. In 2050, permanent departures (ensemble median) occur in 10 % of global ice-free land areas in SSP1-2.6, and in 25 % in SSP3-7.0. By the end of the 21st century, the occurrence of permanent departures in SSP1-2.6 increases to 34 %, and in SSP3-7.0, 45 %. Our findings underscore the importance of mitigation to avoid further degrading the Earth System functions upheld by soil moisture.

One of the major globally relevant systematic biases in previous generations of climate models has been an equatorward bias in the latitude of the Southern Hemisphere (SH) mid‐latitude tropospheric eddy driven westerly jet. The far‐reaching implications of this for Southern Ocean heat and carbon uptake and Antarctic land and sea ice are key reasons why addressing this bias is a high priority. It is therefore of primary importance to evaluate the representation of the SH westerly jet in the latest generation of global climate and earth system models that comprise the Coupled Model Intercomparison Project Phase 6 (CMIP6). In this paper we assess the representation of major indices of SH extratropical atmospheric circulation in CMIP6 by comparison against both observations and the previous generation of CMIP5 models. Indices assessed are the latitude and speed of the westerly jet, variability of the Southern Annular Mode (SAM), and representation of the Amundsen Sea Low (ASL). These are calculated from the historical forcing simulations of both CMIP5 and CMIP6 for time periods matching available observational and reanalysis data sets. From the 39 CMIP6 models available at the time of writing there is an overall reduction in the equatorward bias of the annual mean westerly jet from 1.9° in CMIP5 to 0.4° in CMIP6 and from a seasonal perspective the reduction is clearest in austral spring and summer. This is accompanied by a halving of the bias of SAM decorrelation timescales compared to CMIP5. However, no such overall improvements are evident for the ASL.

Twenty-year temperature and precipitation extremes and their projected future changes are evaluated in an ensemble of climate models participating in the Coupled Model Intercomparison Project Phase 5 (CMIP5), updating a similar study based on the CMIP3 ensemble. The projected changes are documented for three radiative forcing scenarios. The performance of the CMIP5 models in simulating 20-year temperature and precipitation extremes is comparable to that of the CMIP3 ensemble. The models simulate late 20th century warm extremes reasonably well, compared to estimates from reanalyses. The model discrepancies in simulating cold extremes are generally larger than those for warm extremes. Simulated late 20th century precipitation extremes are plausible in the extratropics but uncertainty in extreme precipitation in the tropics and subtropics remains very large, both in the models and the observationally-constrained datasets. Consistent with CMIP3 results, CMIP5 cold extremes generally warm faster than warm extremes, mainly in regions where snow and sea-ice retreat with global warming. There are tropical and subtropical regions where warming rates of warm extremes exceed those of cold extremes. Relative changes in the intensity of precipitation extremes generally exceed relative changes in annual mean precipitation. The corresponding waiting times for late 20th century extreme precipitation events are reduced almost everywhere, except for a few subtropical regions. The CMIP5 planetary sensitivity in extreme precipitation is about 6 %/°C, with generally lower values over extratropical land.

Climate change induced extreme precipitation poses a significant threat to agricultural production. Such impacts extend beyond local agricultural production regions, generating remote and cross-sector impacts that disrupt global supply chains (GSCs). While the direct impacts of extreme precipitation on agricultural production have been widely studied, how such local impacts cascade through supply chain networks to remote places remains elusive, partly because of the complex interdependencies within the global trade systems. To address this, we propose a Resilience Enhancement in Supply Chains Under Environmental Shocks (RESCUES) framework. RESCUES couples an agricultural production loss model with a dynamic recursive economic network model. It allows us to identify channels through which the impacts of climate change on agricultural production propagate along GSCs to interconnected sectors and regions. We design nine climate shock scenarios (i.e., dry, wet, and compound precipitation anomalies with extreme, severe, and moderate levels of severity) using the latest Coupled Model Intercomparison Project Phase 6 (CMIP6) under two Shared Socioeconomic Pathways (SSPs) scenarios (SSP126 and SSP585). We then use RESCUES to simulate the GSCs dynamics over 2016-2050 under these nine scenarios. We find that direct agricultural losses driven by local precipitation anomalies can spread through GSCs to a wider range of countries and regions across the globe, creating large spatial spillover effects with direct and indirect economic losses. We estimate that the averaged per event total value-added (VA) losses caused by compound extremes is around $20.4/22.6 billion under SSP126/585, followed by dry extremes ($16.4/15.0 billion) and wet extremes ($8.7/11.6 billion). Moreover, the global distribution of direct and indirect losses exhibits high spatial heterogeneity. Countries with large agricultural outputs tend to have both high direct and indirect VA losses, especially in China, India, the United States, Russia, and Brazil. In contrast, poorer countries, such as Tanzania, Sudan, Myanmar, Yemen, Afghanistan, and Nepal, experience relatively larger direct losses, while rich regions heavily dependent on agricultural imports, including Hong Kong, Qatar, and Singapore, suffer relatively larger indirect losses. Considering that nations frequently implement export restrictions to ensure food self-sufficiency, we further design a hypothetical scenario to assess the global trade and economic impacts of near-term (2025-2030) agricultural export restrictions in four key food production regions (China, India, the United States, and Indonesia) under extreme precipitation anomalies. Our study highlights the importance of an integrated and comprehensive assessment of the risk footprint of climate change-related shocks, encompassing both direct and indirect impacts on GSCs.&amp;#160;