BAYDAR et al. (2026) Evaluating soybean yield responses to future climate change and irrigation regimes: a DSSAT multi-model assessment

Identification

• Journal: Frontiers in Plant Science
• Year: 2026
• Date: 2026-03-27
• Authors: Alper BAYDAR, Yeşim Bozkurt Çolak, Mete ÖZFİDANER, Hüdaverdi Gürkan, Engin Gönen
• DOI: 10.3389/fpls.2026.1811299

Research Groups

• Department of Biosystems Engineering, Faculty of Agriculture, Siirt University, Siirt, Türkiye
• Department of Biosystems Engineering, Faculty of Agriculture, Malatya Turgut Özal University, Malatya, Türkiye
• Alata Horticultural Research Institute, Ministry of Agriculture and Forestry, Mersin, Türkiye
• Turkish State Meteorological Service, Ankara, Türkiye
• Oil Seeds Research Institute, Ministry of Agriculture and Forestry, Osmaniye, Türkiye

Short Summary

This study evaluated soybean yield responses to future climate change and irrigation regimes in the Mediterranean region of Türkiye using the DSSAT-CROPGRO-Soybean model with multi-GCM projections. It found that while elevated CO2 could increase yields, especially under the high emission scenario (RCP 8.5), seasonal soil water availability remained the primary constraint, highlighting the critical role of irrigation for stabilizing production.

Objective

• To evaluate the impact of future climate change (temperature, precipitation, and atmospheric CO2 concentrations) under different irrigation regimes (full, 70%, 50% replenishment, and rainfed) on soybean phenology, growth, and yield in the Mediterranean region of Türkiye. This was achieved using a calibrated DSSAT-CROPGRO-Soybean model driven by bias-corrected projections from an ensemble of three Global Climate Models (GCMs) across three future periods and two emission pathways (RCP 4.5 and RCP 8.5).

Study Configuration

• Spatial Scale: Field experiments were conducted at the Tarsus location of the Alata Horticultural Research Institute in Mersin, Türkiye (36°53′ N, 34°57′ E; elevation 12 m), characterized by a semi-arid Mediterranean climate. Climate projections were downscaled to a 20 km spatial resolution for the region.
• Temporal Scale:
  • Historical baseline for climate projections: 1970–2000
  • Model calibration: 2024 growing season
  • Model validation: 2025 growing season
  • Future projection periods: Near-future (2016–2040), Mid-century (2041–2070), Late-century (2071–2098)

Methodology and Data

• Models used:
  • Crop Simulation Model: DSSAT-CROPGRO-Soybean (Version 4.8.5.0)
  • Global Climate Models (GCMs) from CMIP5 ensemble: HadGEM2-ES, MPI-ESM-MR, GFDL-ESM2M
  • Regional Climate Model for dynamic downscaling: RegCM4.3.4
• Data sources:
  • Field observations (2024-2025 growing seasons) for calibration and validation: phenological stages (emergence, first flowering, first pod, physiological maturity), canopy height, maximum leaf area index (LAI), aboveground biomass, and grain yield under three irrigation treatments (I100, I70, I50).
  • Soil physical and chemical properties (particle-size distribution, water retention characteristics at -33 kPa and -1500 kPa, bulk density, pH, electrical conductivity).
  • Daily meteorological observations from the study area (1970-2000 baseline) for bias correction of climate projections.
  • Future climate projections (daily maximum temperature, daily minimum temperature, daily precipitation) from bias-corrected GCM outputs under RCP 4.5 and RCP 8.5 emission pathways.
  • Representative atmospheric CO2 concentrations for each RCP and future period (e.g., RCP 4.5: 436-534 ppm; RCP 8.5: 450-846 ppm; historical baseline: 347 ppm).

Main Results

• The calibrated DSSAT-CROPGRO-Soybean model accurately reproduced soybean phenology, canopy development, biomass accumulation, and grain yield across all irrigation treatments during both calibration and validation. Taylor diagram analysis showed high correlation coefficients (generally > 0.95), R² values ranging from 0.74 to 0.99, and acceptable Root Mean Square Deviation (RMSD) values.
• Bias-corrected GCM projections indicated a consistent and progressive warming trend across all models and RCPs, with RCP 8.5 showing more pronounced temperature increases (e.g., daily maximum temperature increases from approximately 4.4-7.1 °C in the near-future to 5.2-9.7 °C in the late-century relative to the baseline). Precipitation projections showed higher inter-model variability, with a general drying trend more pronounced under RCP 8.5.
• Under irrigated conditions, soybean yields were consistently higher under RCP 8.5 than RCP 4.5 across all future periods and GCMs, with the most significant separation observed in the late-century (2071–2098). Yield increases under RCP 8.5 ranged from approximately 4–17% compared to RCP 4.5 in the late-century under irrigation.
• Under rainfed conditions, RCP 8.5 generally produced higher yields than RCP 4.5 (increases of approximately 6–20% in the late-century), but absolute yield levels remained substantially lower and more variable than under irrigated systems.
• Elevated atmospheric CO2 concentrations associated with RCP 8.5 partly mitigated the adverse effects of warming, leading to higher yields, especially when water supply was maintained at adequate levels through irrigation.
• Seasonal soil water availability remained the primary limiting factor for soybean yield, particularly under rainfed conditions, where the beneficial CO2 fertilization effect was insufficient to overcome moisture limitations.

Contributions

• This study provides the first calibration and validation of the DSSAT-CROPGRO-Soybean model under three distinct irrigation regimes (full, 70%, and 50% replenishment) for the specific semi-arid Mediterranean agro-climatic conditions of Türkiye, offering a robust representation of local crop water use and yield responses.
• It integrates bias-corrected climate projections from an ensemble of three Global Climate Models (HadGEM2-ES, MPI-ESM-MR, GFDL-ESM2M) under two emission pathways (RCP 4.5 and RCP 8.5) across multiple future time horizons (near, mid, and late-century) for soybean in the Eastern Mediterranean region, addressing a significant regional research gap.
• The research offers a comprehensive understanding of how interacting drivers—warming, CO2-induced physiological responses, and irrigation management—shape future soybean yield trajectories and associated water requirements through a joint evaluation.
• The findings provide a reliable basis for quantifying climate-related risks and supporting adaptation-focused irrigation planning and water management strategies for sustainable soybean production in semi-arid Mediterranean environments.

Funding

This research was funded by the General Directorate of Agricultural Research and Policy under Project No: TAGEM/TSKAD/T1/23/A11/P3/6417.

Citation

@article{BAYDAR2026Evaluating,
  author = {BAYDAR, Alper and Çolak, Yeşim Bozkurt and ÖZFİDANER, Mete and Gürkan, Hüdaverdi and Gönen, Engin},
  title = {Evaluating soybean yield responses to future climate change and irrigation regimes: a DSSAT multi-model assessment},
  journal = {Frontiers in Plant Science},
  year = {2026},
  doi = {10.3389/fpls.2026.1811299},
  url = {https://doi.org/10.3389/fpls.2026.1811299}
}