Lindenlaub et al. (2026) Characteristics of agricultural droughts in CMIP6 historical simulations and future projections

Identification

• Journal: Earth System Dynamics
• Year: 2026
• Date: 2026-01-14
• Authors: Lukas Lindenlaub, Katja Weigel, Birgit Haßler, Colin Jones, V. Eyring
• DOI: 10.5194/esd-17-81-2026

Research Groups

• University of Bremen, Institute of Environmental Physics (IUP), Bremen, Germany
• Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR), Institut für Physik der Atmosphäre, Oberpfaffenhofen, Germany
• National Centre for Atmospheric Science and School of Earth and Environment, University of Leeds, Leeds, United Kingdom

Short Summary

This study investigates changes in agricultural drought characteristics using 18 CMIP6 Earth System Models under various future scenarios (SSP1-2.6, SSP2-4.5, SSP5-8.5). It projects that the global harvest area experiencing extreme drought conditions by the end of the century could range from 10% to 40%, driven primarily by increasing reference evapotranspiration.

Objective

• To characterize changes in agricultural drought event characteristics (intensity, distribution, spatial extent) in CMIP6 historical simulations and future projections under different Shared Socioeconomic Pathways (SSPs).
• To evaluate the performance of Earth System Models in simulating drought-related variables by comparing historical simulations with reanalysis datasets.

Study Configuration

• Spatial Scale: Global, with a focus on 42 IPCC AR6 WG1 reference regions, specifically 11 harvest-relevant regions. Data regridded to 1° × 1° resolution.
• Temporal Scale:
  • Historical simulations: 1950–2014 (65-year reference period for SPEI calibration), evaluated for 1980–2014 (35 years).
  • Future projections: 2015–2100 (86 years) for SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios.
  • Drought index: 6-month accumulation (SPEI6).
  • Analysis periods: 2015–2034 and 2081–2100 for decadal change rates; 2070–2100 for end-of-century distributions.

Methodology and Data

• Models used: 18 CMIP6 Earth System Models (ACCESS-CM2, AWI-CM-1-1-MR, BCC-CSM2-MR, CanESM5, CMCC-ESM2, CNRM-CM6-1, EC-Earth3-Veg-LR, FIO-ESM-2-0, GFDL-ESM4, GISS-E2.1G, INM-CM5-0, IPSL-CM6A-LR, KACE-1-0-G, MIROC6, MPI-ESM1-2-LR, MRI-ESM2.0, NorESM2-MM, UKESM1-0-LL).
• Data sources:
  • Climate Model Data: CMIP6 ScenarioMIP simulations for SSP1-2.6, SSP2-4.5, SSP5-8.5.
  • Reanalysis Data: ERA5 (monthly averaged and hourly data for temperature, wind speed, surface pressure, shortwave radiation, precipitation, soil moisture), CRU TS v. 4.07 (temperature, precipitation, approximated ET0).
  • Satellite Data: CDS's combined satellite soil moisture dataset (CDS-SM).
  • Land Use Data: GFSAD1KCM dataset (1 km crop mask) for identifying harvest regions.
  • Tools/Indices: Standardized Precipitation Evapotranspiration Index (SPEI6) calculated using the ASCE standardized reference crop evapotranspiration equation (Penman-Monteith variant) and generalized log-logistic distribution. ESMValTool (v2.11) for data loading, pre-processing, and diagnostics. SPEI R-package for SPEI calculation.

Main Results

• CMIP6 models show high pattern correlation (>0.8) with ERA5 for surface downwelling shortwave radiation, reference evapotranspiration (ET0), surface pressure, and daily minimum/maximum temperature (1980-2014). Precipitation correlations are lower (0.6-0.8), and wind speed and soil moisture patterns have the lowest agreement (<0.6).
• All 18 models project a decrease in global mean water budget (P-ET0) and a general shift towards drier conditions (negative SPEI) for 2015-2100, especially in higher emission scenarios (SSP2-4.5, SSP5-8.5). The increase in ET0 is significantly higher than the increase in precipitation.
• Significant regional differences in SPEI changes are projected, with most parts of North and South America, Africa, and Southwest Asia showing decreasing water budgets, while Arctic, Subarctic, and some mountain regions show increasing water budgets.
• Under SSP5-8.5, conditions considered the driest 2.3% of months during 1950-2014 are projected to become "normal" or "moderate" in arid regions (e.g., Mediterranean) by 2100.
• For SSP5-8.5, 40% of the harvest regions' surface is projected to be under extreme drought conditions during Northern Hemisphere autumn by the end of the century. Under SSP1-2.6, this area is projected to be less than 10%.
• Increasing reference evapotranspiration (ET0) is identified as the dominant driver for drier conditions in arid and semi-arid regions.

Contributions

• Extends IPCC AR6 model benchmarking by 15 years and includes additional drought-related variables.
• Provides a comprehensive analysis of agricultural drought characteristics (intensity, distribution shifts, spatial extent) across 18 CMIP6 models under three SSP scenarios, with a focus on 11 key harvest regions.
• Quantifies the projected shift of SPEI distributions, showing that historical extreme drought conditions could become the new normal in arid regions under high emission scenarios.
• Highlights the increasing seasonal dependency and spatial extent of extreme droughts, particularly in Northern Hemisphere autumn.
• Emphasizes the role of increasing reference evapotranspiration as the dominant driver for future drying trends.
• Offers a scientific foundation for further impact and mitigation studies by providing detailed regional and scenario-specific drought projections.
• Integrates methods into the open-source ESMValTool framework, ensuring reproducibility and reusability for future CMIP analyses.

Funding

• Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) (grant no. EY 22/2-1)
• European Research Council (ERC) Synergy Grant "Understanding and modeling the Earth System with Machine Learning (USMILE)" (grant agreement no. 855187)
• European Union's Horizon 2020 research and innovation program (grant agreement 101003536, ESM2025 – Earth System Models for the Future)
• Central Research Development Fund at the University of Bremen (funding no. ZF05/2020/FB1/Causal inference for Earth System Models)
• Deutsches Klimarechenzentrum (DKRZ) (projects no. BD0854 and BD1083)
• University of Bremen (for article processing charges)

Citation

@article{Lindenlaub2026Characteristics,
  author = {Lindenlaub, Lukas and Weigel, Katja and Haßler, Birgit and Jones, Colin and Eyring, V.},
  title = {Characteristics of agricultural droughts in CMIP6 historical simulations and future projections},
  journal = {Earth System Dynamics},
  year = {2026},
  doi = {10.5194/esd-17-81-2026},
  url = {https://doi.org/10.5194/esd-17-81-2026}
}