Singh et al. (2026) Projected intensification of precipitation extremes in the Kosi Basin using CMIP6 models

Identification

• Journal: Scientific Reports
• Year: 2026
• Date: 2026-03-08
• Authors: Aditya Kumar Singh, Thendiyath Roshni, Vivekanand Singh
• DOI: 10.1038/s41598-026-43723-1

Research Groups

• Department of Civil Engineering, National Institute of Technology Patna, Bihar, India

Short Summary

This study evaluates and ranks thirteen statistically downscaled CMIP6 models for reproducing eight ETCCDI precipitation indices over the Kosi River Basin, identifying an optimal eight-member ensemble (AMME8) that projects a significant intensification of precipitation extremes under future warming scenarios.

Objective

• To evaluate and rank downscaled CMIP6 Global Climate Models (GCMs) over the Kosi Basin using a Multi Criteria Decision Making (MCDM)-Criteria Importance Through Inter-criteria Correlation (CRITIC) framework integrated with performance statistics.
• To identify an optimal ensemble subset that minimizes uncertainty and improves the simulation of precipitation extremes.
• To project future changes in precipitation extremes under Shared Socioeconomic Pathways (SSPs) SSP245 and SSP585 scenarios using the optimal ensemble and evaluate their hydrological implications for the Kosi Basin.

Study Configuration

• Spatial Scale: Kosi River Basin, India, at a 0.25° × 0.25° spatial resolution.
• Temporal Scale:
  • Historical period: 1985-2014
  • Near Future (EF): 2031-2060
  • Far Future (FF): 2061-2100

Methodology and Data

• Models used:
  • Thirteen statistically downscaled and bias-corrected CMIP6 Global Climate Models (GCMs): ACCESS-CM2, ACCESS-ESM1-5, BCC-CSM2-MR, CanESM5, EC-Earth3, EC-Earth3-Veg, INM-CM4-8, INM-CM5-0, MPI-ESM1-2-HR, MPI-ESM1-2-LR, MRI-ESM2-0, NorESM2-LM, and NorESM2-MM.
  • Arithmetic Multi-Model Ensembles (AMME): AMME3 (top 3 models), AMME5 (top 5 models), AMME8 (top 8 models), AMME13 (all 13 models).
  • Evaluation Framework: Eight ETCCDI precipitation indices, eight statistical indicators (Normalized Root Mean Square Error (NRMSE), Correlation Coefficient (CC), Mean Absolute Error (MAE), Mean Deviation (MD), Variance Explained (VE), Percent Bias (PBIAS), Mean Percentage Error (MPE), Sum of Squared Error (SSE)), Symmetric Uncertainty (SU) for post-assessment.
  • Weighting Method: Criteria Importance Through Inter-criteria Correlation (CRITIC) method.
  • Ranking Methods: Four Multi Criteria Decision Making (MCDM) techniques: TOPSIS, VIKOR, EDAS, and PROMETHEE-II.
• Data sources:
  • Observational/Reference: ERA5 reanalysis dataset (European Centre for Medium-Range Weather Forecasts - ECMWF).
  • Climate Models: Coupled Model Intercomparison Project Phase 6 (CMIP6) archive, statistically downscaled and bias-corrected to 0.25° × 0.25° spatial resolution.
  • Future Scenarios: Shared Socioeconomic Pathways (SSPs) SSP245 (intermediate stabilization) and SSP585 (high emission pathway).

Main Results

• MPI-ESM1-2-HR, INM-CM5-0, and BCC-CSM2-MR consistently outperformed other CMIP6 models in reproducing historical precipitation extremes over the Kosi Basin, while ACCESS-CM2 and NorESM2 variants showed weaker agreement.
• Error-based metrics (MD, MPE, SSE, NRMSE) received higher CRITIC weights, indicating their dominant role in evaluating model performance for precipitation extremes.
• The eight-member ensemble (AMME8) provided the best balance of accuracy and uncertainty reduction, most closely replicating the observed inter-relationships among precipitation extremes and achieving optimal symmetric uncertainty.
• Future projections using the AMME8 ensemble indicate a marked intensification of precipitation extremes under both SSP245 and SSP585 scenarios.
• The far future (2061-2100), particularly under SSP585, shows the strongest amplification with projected increases of up to 47% in annual precipitation (PRCPTOT), 60% in heavy rainfall days (R20mm), and nearly 79% in extremely wet days (R99p_total), suggesting heightened flood risk.

Contributions

• Developed a rigorous, multi-stage evaluation framework integrating objective CRITIC weights with four complementary MCDM techniques for robust and transparent identification of reliable CMIP6 models and ensemble configurations for regional hydro-climatic applications.
• Explicitly evaluated model performance based on individual ETCCDI extreme climatic indices, demonstrating its superiority over aggregated annual precipitation metrics for capturing model-specific strengths and weaknesses.
• Identified AMME8 as the optimal ensemble for the Kosi Basin, demonstrating that a balanced ensemble size (eight models) best preserves the physical coherence and inter-dependencies among precipitation extreme indices while effectively reducing uncertainty.
• Provided reliable, index-specific regional climate projections for precipitation extremes in the Kosi Basin under SSP245 and SSP585, highlighting significant future intensification and its implications for flood management and climate adaptation.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Citation

@article{Singh2026Projected,
  author = {Singh, Aditya Kumar and Roshni, Thendiyath and Singh, Vivekanand},
  title = {Projected intensification of precipitation extremes in the Kosi Basin using CMIP6 models},
  journal = {Scientific Reports},
  year = {2026},
  doi = {10.1038/s41598-026-43723-1},
  url = {https://doi.org/10.1038/s41598-026-43723-1}
}