Joshi et al. (2025) Assessment of precipitation and its extreme precipitation changes over the Himalayan Bhilangana River Basin, India

Identification

• Journal: Theoretical and Applied Climatology
• Year: 2025
• Date: 2025-11-18
• Authors: Bhupendra Joshi, Vishal Singh, V. K. Chandola, Atar Singh
• DOI: 10.1007/s00704-025-05881-6

Research Groups

• Department of Agricultural Engineering, Banaras Hindu University, Varanasi, India
• Centre for Cryosphere and Climate Change Studies, NIH, Roorkee, Uttarakhand, India

Short Summary

This study assessed historical and projected changes in extreme precipitation over the Himalayan Bhilangana River Basin, India, by first developing a high-resolution bias-corrected precipitation dataset and then analyzing CMIP6 model outputs. The findings indicate consistent declines in short-duration extreme precipitation events (Rx1day, Rx5day), fewer consecutive dry days, and a marked increase in consecutive wet days, particularly in high-altitude regions.

Objective

• To identify the most accurate gridded precipitation products for the Bhilangana Basin through comprehensive evaluation against available gauge data using continuous and categorical metrics.
• To reconstruct a high-resolution (0.10°) precipitation dataset via hybrid merging and bias correction of top-performing gridded precipitation products.
• To assess historical (1995–2023) and future (2025–2100) precipitation extremes using ETCCDI indices under SSP245 and SSP585 scenarios using a statistically downscaled CMIP6 multi-model ensemble.

Study Configuration

• Spatial Scale: Bhilangana River Basin, India, spanning approximately 1,482 km² (between 30°20′71″ to 30°50′79″ N and 78°30′22″ to 79°03′53″ E), with an elevation range of 769 to 6,614 meters above sea level. The reconstructed precipitation dataset has a spatial resolution of 0.10°.
• Temporal Scale:
  • Historical/Baseline: 1995–2023 (for ETCCDI indices), 2000–2023 (for reconstructed dataset), and 2012–2015 (for gauge comparison).
  • Future Projections: Overall period 2025–2100, categorized into Near Future (NF: 2025–2054) and Far Future (FF: 2061–2090).

Methodology and Data

• Models used:
  • Climate Models: Multi-model ensemble of five selected CMIP6 models (BCC-CSM2-MR, CMCC-CM2-SR5, INM-CM5-0, KIOST-ESM, NESM3) under SSP245 and SSP585 scenarios.
  • Bias Correction: Linear Scaling (LS) for CMIP6 outputs and Cumulative Distribution Function (CDF) based Quantile Mapping (QM) for gridded precipitation datasets.
  • Data Assimilation/Merging: Hybrid statistical merging approach (two-phase quantile regression) to reconstruct a high-resolution gridded precipitation product.
• Data sources:
  • Gridded Precipitation Datasets: APHRODITE (0.25°), CHIRPS (0.25°), ERA5-reanalysis (0.25°), ERA5-Land (0.10°), GPM (0.10°), IMD (0.25°), IMDAA-reanalysis (0.12°), and Princeton University (0.25°).
  • Observation Data: Daily precipitation data from the Koteshwar station (Central Water Commission - CWC) for 2012–2015.
  • Extreme Climate Indices: 10 precipitation indices recommended by the Expert Team on Climate Change Detection and Indices (ETCCDI).

Main Results

• Gridded Dataset Performance: APHRODITE demonstrated the strongest agreement with observed station data (correlation coefficient (r) of 0.95, Root Mean Square Error (RMSE) of 42.54 mm/month, lowest Mean Absolute Error (MAE) and Percentage Bias (PBias), and highest Critical Success Index (CSI)). IMD was ranked second best, while Princeton performed the weakest.
• Reconstructed Precipitation Dataset: A high-resolution (0.10°) bias-corrected precipitation dataset for 2000–2023 was successfully reconstructed by merging IMD, ERA5-Land, and GPM using quantile mapping. This dataset showed a 5% to 20% improvement in accuracy across the basin compared to APHRODITE.
• Future Precipitation Extremes (2025–2090):
  • Maximum 1-day (Rx1day) and 5-day (Rx5day) precipitation: Consistent declines are projected across both SSP245 and SSP585 scenarios, with Rx1day reductions ranging from -50% to -72% and Rx5day reductions from -35% to -55%. High-altitude regions show the most significant declines.
  • Consecutive Dry Days (CDD): Projected to decrease, with reductions ranging from -5% to -50% across scenarios and time horizons, indicating fewer dry spells.
  • Consecutive Wet Days (CWD): Projected to increase substantially, with changes ranging from 120% to 200%, indicating more frequent wet spells.
  • Heavy and Very Heavy Precipitation Days (R ≥ 10 mm, R ≥ 20 mm, R95p, R99p): Mixed trends for R ≥ 10 mm (some increases in lower elevations, significant declines in high elevations), but generally a decline in very heavy precipitation events (R ≥ 20 mm, R95p, R99p). R99p shows declines of -40% to -70%.
  • Simple Daily Intensity Index (SDII) and Yearly Total Precipitation on Wet Days (PRCPTOT): Both indices show decreasing trends, with SDII reductions of -25% to -55% and PRCPTOT reductions of -5% to -40%.
• Regional Sensitivity: High-altitude areas are identified as particularly susceptible to extreme precipitation changes, increasing the vulnerability of glaciers.

Contributions

• Developed a robust, high-resolution (0.10°) bias-corrected gridded precipitation dataset for the data-scarce Bhilangana River Basin in the Central Himalayas, integrating multiple satellite and reanalysis products with gauge observations.
• Provided a comprehensive evaluation and ranking of various gridded precipitation products against in-situ data using a combination of continuous and threshold-based statistical metrics, offering critical insights into their performance in complex mountainous terrain.
• Generated long-term projections (1995–2100) of extreme precipitation indices under CMIP6 SSP245 and SSP585 scenarios, revealing significant future changes in precipitation patterns, including declines in short-duration extremes and shifts in dry/wet spell frequencies.
• Highlighted the elevated sensitivity of high-altitude regions to future precipitation shifts, underscoring potential risks to glacier stability, flood regimes, and hydropower reliability in the Himalayas.
• Emphasized the importance of such evidence-based insights for developing climate-adaptive water management strategies, disaster preparedness, and risk reduction policies for vulnerable communities in the Himalayan region.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript. This research was exclusively conducted using in-house resources and support from Banaras Hindu University (BHU) and the National Institute of Hydrology (NIH).

Citation

@article{Joshi2025Assessment,
  author = {Joshi, Bhupendra and Singh, Vishal and Chandola, V. K. and Singh, Atar},
  title = {Assessment of precipitation and its extreme precipitation changes over the Himalayan Bhilangana River Basin, India},
  journal = {Theoretical and Applied Climatology},
  year = {2025},
  doi = {10.1007/s00704-025-05881-6},
  url = {https://doi.org/10.1007/s00704-025-05881-6}
}