Pink et al. (2025) Increased rainfall-runoff drives flood hazard intensification in Central Himalayan river systems

Identification

• Journal: Scientific Reports
• Year: 2025
• Date: 2025-12-08
• Authors: Ivo Pink, Sim Reaney, R. J. Hardy, C. Isabella Bovolo
• DOI: 10.1038/s41598-025-26815-2

Research Groups

• Department of Geography, Durham University, UK
• Institute of Hazard, Risk and Resilience, Durham University, UK

Short Summary

This study provides the first large-ensemble, regional climate-change impact assessment of design floods for the Central Himalayan Karnali River, projecting significant increases in 1% annual exceedance probability flood magnitudes (up to +79% by 2060–2099 under high emissions) primarily driven by increased rainfall-runoff.

Objective

• To provide a comprehensive regional climate change impact assessment of design flood hazards (specifically 1% Annual Exceedance Probability) in the Central Himalayan Karnali River, quantifying future flood magnitudes and their driving hydrological processes under different emission scenarios, while accounting for climatic, hydrological, and statistical uncertainties.

Study Configuration

• Spatial Scale: Karnali River catchment in Nepal and China, modeled at a 500 m × 500 m grid resolution.
• Temporal Scale: Baseline (1975–2014), Near-future (2020–2059), and Far-future (2060–2099) timeframes, with daily time steps for simulations.

Methodology and Data

• Models used:
  • Spatial Processes in Hydrology (SPHY) hydrological model (version 3.0)
  • Ensemble of 12 bias-corrected and downscaled CMIP6 Global Circulation Models (GCMs)
  • Generalised Likelihood Uncertainty Estimation (GLUE) framework for hydrological uncertainty
  • Flood Frequency Analysis (FFA) using the L-Moments approach with Wakeby distribution and a bootstrapping framework for sampling uncertainty.
• Data sources:
  • Satellite-based precipitation estimates (GPM IMERG Final Precipitation L3 1 Month V006) disaggregated with in-situ measurements.
  • Reanalysis temperature data (WATCH forcing Data methodology applied to ERA-Interim dataset - WFDEI).
  • Bias-corrected and downscaled CMIP6 climate projections for temperature and precipitation (SSP245 and SSP585 scenarios).
  • Daily discharge and stage-discharge observations from the Department of Hydrology and Meteorology (DHM), Nepal.
  • Satellite-based annual actual evapotranspiration (ETa) estimates (MODIS).
  • SRTM 90 m V4.167 for topography.

Main Results

• For the far-future (2060–2099), the 1% Annual Exceedance Probability (AEP) flood magnitudes are projected to increase by 40% (medium-emissions, SSP245) and 79% (high-emissions, SSP585) relative to the baseline (1975–2014).
• Rainfall-runoff is the dominant driver of flood intensification, contributing ≥ 90% of the additional flood water in future scenarios.
• The mean flood discharge is projected to increase by 1,380–4,581 cubic meters per second (m³/s) compared to the baseline, with rainfall-runoff accounting for 91–93% of this increase.
• Mean snowmelt contributions to flood discharge are projected to decrease from 3% (baseline) to 0-1% (far-future, SSP585), primarily due to an earlier onset of the melting season.
• The hydrological model ensemble is the largest source of uncertainty in the baseline (52% of overall variance), while climate model uncertainty becomes increasingly dominant in the far-future (up to 49% for SSP585).
• Flood Frequency Analysis (FFA) uncertainty is significantly affected by the timing and occurrence of rare extreme precipitation events, emphasizing the need for large climate model ensembles.

Contributions

• Provides the first comprehensive regional climate change impact assessment of design flood hazards (1% AEP) for the Central Himalayan Karnali River, utilizing high-resolution, large-ensemble modeling with the latest CMIP6 climate projections.
• Quantifies and decomposes climatic, hydrological, and statistical uncertainties in future design flood projections, including the novel incorporation of sampling uncertainty into the Flood Frequency Analysis (FFA).
• Identifies rainfall-runoff as the overwhelming dominant driver (≥90%) of projected flood intensification in the region, offering crucial information for flood risk management.
• Proposes a robust modeling framework for predicting Annual Exceedance Probability (AEP) flood magnitudes under climate change, emphasizing the need for large, diverse climate ensembles (Atmosphere Only General Circulation Models and Earth System Models) to capture extreme event uncertainty.

Funding

• Charles Wilson Doctoral Studentship

Citation

@article{Pink2025Increased,
  author = {Pink, Ivo and Reaney, Sim and Hardy, R. J. and Bovolo, C. Isabella},
  title = {Increased rainfall-runoff drives flood hazard intensification in Central Himalayan river systems},
  journal = {Scientific Reports},
  year = {2025},
  doi = {10.1038/s41598-025-26815-2},
  url = {https://doi.org/10.1038/s41598-025-26815-2}
}