Saha et al. (2025) Development of daily downscaled, bias-corrected CMIP6 climate datasets for estimating reference evapotranspiration (ETo) in South Asia

Identification

• Journal: Scientific Data
• Year: 2025
• Date: 2025-11-28
• Authors: Aniruddha Saha, Manoj Jain, P. V. Joshi, Subhankar Das, Naimesh Singh Rawat
• DOI: 10.1038/s41597-025-06149-4

Research Groups

• Department of Hydrology, Indian Institute of Technology Roorkee, Roorkee, India
• Department of Soil and Water Systems, University of Idaho, Moscow, ID, USA
• Moscow Forestry Sciences Laboratory, USDA Forest Service-Rocky Mountain Research Station, Moscow, ID, USA

Short Summary

This study developed daily bias-corrected and downscaled CMIP6 climate datasets for South Asia, including temperature, solar radiation, wind speed, and relative humidity, which were then used to estimate reference evapotranspiration (ETo) at a 0.25° spatial resolution for historical and future climate scenarios. The resulting datasets significantly reduce biases compared to original CMIP6 outputs, providing a crucial resource for regional hydrological and climate impact assessments.

Objective

• To develop a high-resolution (0.25°), daily, bias-corrected reference evapotranspiration (ETo) dataset for South Asia using outputs from 12 CMIP6 General Circulation Models (GCMs) for historical and four Shared Socioeconomic Pathways (SSPs), employing the FAO-56 Penman-Monteith method.

Study Configuration

• Spatial Scale: South Asia (India, Bangladesh, Nepal, Bhutan, Pakistan, Afghanistan, Sri Lanka) at 0.25° × 0.25° spatial resolution.
• Temporal Scale: Daily data for the historical period (1960–2014) and future scenarios (2015–2100) under SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5.

Methodology and Data

• Models used:
  • 12 CMIP6 General Circulation Models (GCMs)
  • FAO-56 Penman-Monteith (PM) method for ETo estimation.
  • Detrended Quantile Mapping approach for bias correction and spatial downscaling.
• Data sources:
  • CMIP6 GCM outputs (daily temperature, solar radiation, wind speed, relative humidity).
  • ERA-5 reanalysis data (daily, 0.25° spatial resolution) as observational reference for bias correction.
  • Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) (30 m resolution, upscaled to 0.25°).
  • Gridded atmospheric carbon dioxide (CO2) concentrations (historical 1960-2013, future 2015-2100, 1° spatial resolution, downscaled to 0.25°).

Main Results

• Daily bias-corrected datasets of temperature (K), solar radiation (W/m²), wind speed (m/s), relative humidity (%), and ETo (mm) were developed at 0.25° spatial resolution for South Asia.
• Bias correction using Quantile Mapping successfully removed systematic biases in CMIP6 outputs, with the bias-corrected ETo climatology closely aligning with ERA5 observations for the historical period (1960–2014).
• Inter-variable dependencies among the four climatic sub-variables were preserved after univariate bias correction.
• Projected ETo shows an overall increase across South Asia under SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios, with the highest increases observed in Afghanistan, northern Pakistan, and northern India.
• Under the SSP3-7.0 scenario, a projected decrease in ETo is observed over most of the Indian mainland and Bangladesh.
• While mean temperatures are projected to rise across all SSPs, ETo trends exhibit more complex responses due to varying contributions from the individual climate variables.

Contributions

• Delivers the first regional-scale, bias-corrected, daily ETo dataset for South Asia, derived from 12 CMIP6 projections using the FAO-56 Penman-Monteith method.
• Provides high-resolution (0.25°) daily datasets of key climate variables and ETo for both historical (1960-2014) and future (2015-2100) periods under four SSPs.
• Employs a trend-preserving detrended Quantile Mapping approach, ensuring the integrity of climate change signals during bias correction.
• Incorporates the influence of atmospheric CO2 concentration on surface resistance in ETo calculations, enhancing the physical realism of projections.
• The publicly available datasets serve as a valuable foundation for improved regional water resource assessments, agricultural planning, and climate impact studies in a climate-vulnerable and densely populated region.

Funding

• Prime Minister’s Research Fellowship (PM-31-22-630-414) under the Government of India.
• National Supercomputing Mission (NSM) for computing resources of ‘PARAM Ganga’ at the Indian Institute of Technology Roorkee, supported by the Ministry of Electronics and Information Technology (MeitY) and Department of Science and Technology (DST), Government of India.

Citation

@article{Saha2025Development,
  author = {Saha, Aniruddha and Jain, Manoj and Joshi, P. V. and Das, Subhankar and Rawat, Naimesh Singh},
  title = {Development of daily downscaled, bias-corrected CMIP6 climate datasets for estimating reference evapotranspiration (ETo) in South Asia},
  journal = {Scientific Data},
  year = {2025},
  doi = {10.1038/s41597-025-06149-4},
  url = {https://doi.org/10.1038/s41597-025-06149-4}
}