PrecipCorrDiff: A Deep Learning Framework for Generating 0.0625∘ High-Resolution IMD Gridded Precipitation Data
Abstract
High-resolution precipitation datasets are essential for weather forecasting, hydrological modeling, and climate risk assessment, yet observation-based gridded products often lack the spatial detail required for regional applications. Here we present PrecipCorrDiff, a diffusion-based downscaling framework built by modifying the original corrector diffusion based archetecture that incorporates a customized loss function and a multi-scale residual generation strategy to improve the precipitation especially the representation of extremes. PrecipCorrDiff generates 0.0625° resolution India Meteorological Department (IMD) precipitation fields from 0.25° gridded data while preserving precipitation statistics and reconstructing fine-scale spatial variability. Validation against independent IMD station observations, comparing the proposed framework with the SRGAN-based model and the original CorrDiff model, indicates that the proposed framework substantially improves the representation of spatial precipitation patterns, rainfall intensity distributions, and extreme rainfall events, while eliminating the systematic negative bias present in the previously developed SRGAN-based model. Across four climatologically distinct homogeneous Indian monsoon regions, PrecipCorrDiff reproduces observed precipitation characteristics with near-zero spatial bias, demonstrating the potential of diffusion-based downscaling for generating physically consistent, high-resolution precipitation datasets for hydrological, meteorological, and climate applications.
Related
- Zero‐Shot Transferability of Deep Learning Models for Precipitation Super‐Resolution Across Diverse Climate Regimes
- A Sensor-Adaptive Incremental Learning Framework for Artifact Detection in Satellite Precipitation Data
- Spatio-temporal precipitation downscaling at hourly resolution based on km-scale regional simulations
- High-resolution Calibrated Probabilistic Hourly Precipitation from a Deterministic Forecast
- Dependence of the precipitation intensity distribution on spatial and temporal resolution in observations
BibTeX
@misc{kumar2026precipcorrdiff,
title = {{PrecipCorrDiff: A Deep Learning Framework for Generating 0.0625$^\circ$ High-Resolution IMD Gridded Precipitation Data}},
author = {Bipin Kumar and Rajib Chattopadhyay and Harsh P. Bachhav and Jannu and Bhupendra Bahadur Singh and Ramesh Ragala and Ananda Kumar Das and Manish Modani and Thara Prabhakaran and Mrutyunjay Mohapatra},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.19994},
eprint = {2610.19994},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.19994},
}
RIS
TY - UNPB
TI - PrecipCorrDiff: A Deep Learning Framework for Generating 0.0625^∘ High-Resolution IMD Gridded Precipitation Data
AU - Bipin Kumar
AU - Rajib Chattopadhyay
AU - Harsh P. Bachhav
AU - Jannu
AU - Bhupendra Bahadur Singh
AU - Ramesh Ragala
AU - Ananda Kumar Das
AU - Manish Modani
AU - Thara Prabhakaran
AU - Mrutyunjay Mohapatra
PY - 2026
DA - 2026/10/14
PB - arXiv
DO - 10.48550/arxiv.2610.19994
UR - https://arxiv.org/abs/2610.19994
AB - High-resolution precipitation datasets are essential for weather forecasting, hydrological modeling, and climate risk assessment, yet observation-based gridded products often lack the spatial detail required for regional applications. Here we present PrecipCorrDiff, a diffusion-based downscaling framework built by modifying the original corrector diffusion based archetecture that incorporates a customized loss function and a multi-scale residual generation strategy to improve the precipitation especially the representation of extremes. PrecipCorrDiff generates 0.0625° resolution India Meteorological Department (IMD) precipitation fields from 0.25° gridded data while preserving precipitation statistics and reconstructing fine-scale spatial variability. Validation against independent IMD station observations, comparing the proposed framework with the SRGAN-based model and the original CorrDiff model, indicates that the proposed framework substantially improves the representation of spatial precipitation patterns, rainfall intensity distributions, and extreme rainfall events, while eliminating the systematic negative bias present in the previously developed SRGAN-based model. Across four climatologically distinct homogeneous Indian monsoon regions, PrecipCorrDiff reproduces observed precipitation characteristics with near-zero spatial bias, demonstrating the potential of diffusion-based downscaling for generating physically consistent, high-resolution precipitation datasets for hydrological, meteorological, and climate applications.
ER -