Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2411.16098

Global spatio-temporal downscaling of ERA5 precipitation through generative AI

Luca Glawion·Julius Polz·Harald Kunstmann·Benjamin Fersch·Christian Chwala
View original source ↗Read source document ↗Export metadata ↓

Abstract

The spatial and temporal distribution of precipitation has a significant impact on human lives by determining freshwater resources and agricultural yield, but also rainfall-driven hazards like flooding or landslides. While the ERA5 reanalysis dataset provides consistent long-term global precipitation information that allows investigations of these impacts, it lacks the resolution to capture the high spatio-temporal variability of precipitation. ERA5 misses intense local rainfall events that are crucial drivers of devastating flooding - a critical limitation since extreme weather events become increasingly frequent. Here, we introduce spateGAN-ERA5, the first deep learning based spatio-temporal downscaling of precipitation data on a global scale. SpateGAN-ERA5 uses a conditional generative adversarial neural network (cGAN) that enhances the resolution of ERA5 precipitation data from 24 km and 1 hour to 2 km and 10 minutes, delivering high-resolution rainfall fields with realistic spatio-temporal patterns and accurate rain rate distribution including extremes. Its computational efficiency enables the generation of a large ensemble of solutions, addressing uncertainties inherent to the challenges of downscaling. Trained solely on data from Germany and validated in the US and Australia considering diverse climate zones, spateGAN-ERA5 demonstrates strong generalization indicating a robust global applicability. SpateGAN-ERA5 fulfils a critical need for high-resolution precipitation data in hydrological and meteorological research, offering new capabilities for flood risk assessment, AI-enhanced weather forecasting, and impact modelling to address climate-driven challenges worldwide.

Explore related subjects

physics.ao-phcs.CVcs.LGphysics.geo-ph

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Luca Glawion, Julius Polz, Harald Kunstmann, Benjamin Fersch, Christian Chwala. 2024-11-22. Global spatio-temporal downscaling of ERA5 precipitation through generative AI. https://doi.org/10.1038/s41612-025-01103-y

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Surface Stokes drift from compact drifting wave buoys

Surface Stokes drift depends strongly on the energy and directions of short waves, which are incompletely resolved by routine wave observations. We derive surface Stokes drift vectors from wave measurements collected by compact drifting buoys during three deployments in the North-East Atlantic and the Alboran Sea. The calculation uses vertical-acceleration spectra and first directional Fourier moments, which describe the mean wave direction and directional concentration at each frequency; it accounts for the Doppler shift caused by buoy motion relative to the water and adds a calibrated high-frequency tail above an intrinsic frequency of 0.7 Hz. Across 13,139 records, the median estimated speed is 0.081 m/s at a median wind speed of 6.8 m/s. Over the measured band of 0.04-1 Hz, accounting for wave directions reduces the magnitude by a median 39% relative to the unidirectional assumption. The median ratio of the parameterised tail magnitude above 0.7 Hz to the total estimated magnitude is 0.37. Comparisons with WAVEWATCH III and Copernicus Marine MFWAM show strong covariation and similar wind-dependent differences from the buoy-derived estimates. On the station-matched sample from the two Atlantic deployments, WAVEWATCH III directional spectra indicate that these differences within the compared band arise mainly from spectral levels rather than from net directional reduction. The observations provide constraints for model evaluation; the contribution of the unresolved short waves remains sensitive to the assumed spectral tail and its directional spreading.

physics.ao-ph↗
arXiv2026

Diffusion-Based Super-Resolution of Adriatic Sea Oceanographic Fields

High-resolution oceanographic fields are critical for resolving mesoscale and sub-mesoscale coastal dynamics, yet their generation remains constrained by both computational cost and observational sparsity. We present OcDiffSR, a conditional denoising diffusion probabilistic model (DDPM) for oceanographic super-resolution that reconstructs high-resolution sea-surface fields from coarse-resolution reanalysis inputs. The model is trained on ten years (2011-2020) of paired low-resolution (GLORYS12V1, 1/12) and high-resolution (Mediterranean Sea Physics Reanalysis, Med MFC, 1/24) data, and evaluated on an independent test year (2009) over the Adriatic Sea. OcDiffSR employs a conditional U-Net augmented with multi-scale low-resolution encoders, cross-attention bottleneck layers, and sinusoidal seasonal embeddings via Feature-wise Linear Modulation (FiLM), enabling joint super-resolution of sea-surface temperature (SST), salinity (SSS), and horizontal velocity components with visually coherent circulation patterns. Benchmarked against bilinear interpolation and the state-of-the-art residual diffusion model CorrDiff, OcDiffSR achieves substantially lower reconstruction errors for scalar fields (RMSESST=0.477 C, RMSESSS=0.346 psu), near-unity Pearson correlation (PCC >= 0.999), and high structural similarity (SSIM >= 0.964). For dynamical vector fields, OcDiffSR outperforms both baselines in absolute error and spatial coherence, though moderate correlation (PCC = 0.64) reflects the intrinsic stochasticity of oceanic velocity fields. Daily and monthly evaluations confirm temporal robustness across all seasons. These results establish OcDiffSR as a reliable framework for high-fidelity oceanographic downscaling and reanalysis enhancement, producing fields that are visually consistent with known ocean dynamics.

physics.ao-ph↗
arXiv2026

Unreported large errors from two PAMGuard three-dimensional localizers of whale calls

Confidence intervals of location (CIL) of calling marine mammals, derived from time-differences-of-arrival (TDOA) between receivers, depend on errors of TDOAs, receiver location, clocks, sound speeds, and location method. When these errors are minuscule, simulations yield small errors of PAMGuard's 3D simplex localizer when click sounds of beaked and sperm whales originate in a 1000 x 1000 x 1000 $\mbox{m}^3$ region using five receivers having horizontal and vertical separations of 1000 m and 150 m respectively. Realistic uncertainties of sound speed up to $\pm 10$ m/s lead to errors up to $10^{14}$ m. With clocks maintained by atomic standards and common practice of correcting TDOA from synchronization measurements at the start and end of an experiment, errors of location are up to $10^{4}$ m. Errors up to $10^2$ and $10^3$ m are found when the receiver's locations are uncertain within 10 and 40 m respectively. Errors of PAMGuard's 3D hyperbolic localizer are almost independent of the above uncertainties, yielding errors of location up to about $10^4$ m even when simulated errors are minuscule. Causes of PAMGuard's 3D location errors are unknown. These algorithms are briefly compared to another method designed to yield a reliable CIL.

physics.ao-ph↗