Statistical Downscaling of Daily Precipitation from ERA5 Atmospheric Fields to IMERG over West Africa
Abstract
Global reanalyses are too coarse to resolve the convective systems that produce most West African monsoon rainfall. In this work, we use U-Net based architectures to downscale daily ERA5 fields (0.25°) to IMERG V07 satellite precipitation (0.1°) over West Africa for the June to October seasons of 2000 to 2024, using geopotential, temperature, humidity and wind at three pressure levels together with elevation and land fraction. Adding ERA5 precipitation as input slightly improves performance for this particular challenging region. We find that all the deterministic networks investigated for this work are too dry and lose almost all observed variance below a few hundred kilometers, which motivated our use of a GAN-based alternative. We also investigated permutation importance for the U-Net that ranks 500 hPa specific humidity first, consistent with the sensitivity of deep convection to mid-tropospheric moisture.
Related
- Spatio-temporal precipitation downscaling at hourly resolution based on km-scale regional simulations
- Stress-Testing Dynamical and Generative Downscaling Using Subseasonal Extreme Precipitation Forecasts
- Evaluating Cross-region Generalization for Wavelet-Diffusion Precipitation Downscaling
- SwAIther-Precip: Lead-Time-Aware Bias Correction Enables Kilometer-Scale Downscaling of Global AI Precipitation Forecasts over Switzerland
- Climate models lack decadal variability in precipitation because of a weak atmospheric response to sea surface temperature anomalies
BibTeX
@misc{sagna2026statistical,
title = {{Statistical Downscaling of Daily Precipitation from ERA5 Atmospheric Fields to IMERG over West Africa}},
author = {David Sagna and Redouane Lguensat and Samo Diatta},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.16816},
eprint = {2610.16816},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.16816},
}
RIS
TY - UNPB
TI - Statistical Downscaling of Daily Precipitation from ERA5 Atmospheric Fields to IMERG over West Africa
AU - David Sagna
AU - Redouane Lguensat
AU - Samo Diatta
PY - 2026
DA - 2026/10/12
PB - arXiv
DO - 10.48550/arxiv.2610.16816
UR - https://arxiv.org/abs/2610.16816
AB - Global reanalyses are too coarse to resolve the convective systems that produce most West African monsoon rainfall. In this work, we use U-Net based architectures to downscale daily ERA5 fields (0.25°) to IMERG V07 satellite precipitation (0.1°) over West Africa for the June to October seasons of 2000 to 2024, using geopotential, temperature, humidity and wind at three pressure levels together with elevation and land fraction. Adding ERA5 precipitation as input slightly improves performance for this particular challenging region. We find that all the deterministic networks investigated for this work are too dry and lose almost all observed variance below a few hundred kilometers, which motivated our use of a GAN-based alternative. We also investigated permutation importance for the U-Net that ranks 500 hPa specific humidity first, consistent with the sensitivity of deep convection to mid-tropospheric moisture.
ER -