Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI
Abstract The spatial and temporal distribution of precipitation significantly impacts human lives. While reanalysis datasets provide consistent long-term global precipitation information that allows investigations of rainfall-driven hazards like larger-scale flooding, they lack the resolution to cap...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-06-01
|
| Series: | npj Climate and Atmospheric Science |
| Online Access: | https://doi.org/10.1038/s41612-025-01103-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849335829373124608 |
|---|---|
| author | Luca Glawion Julius Polz Harald Kunstmann Benjamin Fersch Christian Chwala |
| author_facet | Luca Glawion Julius Polz Harald Kunstmann Benjamin Fersch Christian Chwala |
| author_sort | Luca Glawion |
| collection | DOAJ |
| description | Abstract The spatial and temporal distribution of precipitation significantly impacts human lives. While reanalysis datasets provide consistent long-term global precipitation information that allows investigations of rainfall-driven hazards like larger-scale flooding, they lack the resolution to capture the high spatio-temporal variability of precipitation and miss intense local rainfall events. Here, we introduce spateGAN-ERA5, the first deep learning-based spatio-temporal downscaling of precipitation data on a global scale. SpateGAN-ERA5 enhances ERA5 precipitation data from 24 km and 1 h to 2 km and 10 min, 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 downscaling challenges and supports practical applicability for generating high-resolution precipitation data for arbitrary ERA5 time periods and regions on demand. Trained solely on data from Germany and validated in the US and Australia, considering diverse climates, including tropical rainfall regimes, spateGAN-ERA5 demonstrates strong generalization, indicating robust global applicability. It fulfills critical needs for high-resolution precipitation data in hydrological and meteorological research. |
| format | Article |
| id | doaj-art-34db5f449d934fbc9aade3725f669577 |
| institution | Kabale University |
| issn | 2397-3722 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Climate and Atmospheric Science |
| spelling | doaj-art-34db5f449d934fbc9aade3725f6695772025-08-20T03:45:10ZengNature Portfolionpj Climate and Atmospheric Science2397-37222025-06-018111310.1038/s41612-025-01103-yGlobal spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AILuca Glawion0Julius Polz1Harald Kunstmann2Benjamin Fersch3Christian Chwala4Institute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU), Campus Alpin, Karlsruhe Institute of TechnologyInstitute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU), Campus Alpin, Karlsruhe Institute of TechnologyInstitute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU), Campus Alpin, Karlsruhe Institute of TechnologyInstitute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU), Campus Alpin, Karlsruhe Institute of TechnologyInstitute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU), Campus Alpin, Karlsruhe Institute of TechnologyAbstract The spatial and temporal distribution of precipitation significantly impacts human lives. While reanalysis datasets provide consistent long-term global precipitation information that allows investigations of rainfall-driven hazards like larger-scale flooding, they lack the resolution to capture the high spatio-temporal variability of precipitation and miss intense local rainfall events. Here, we introduce spateGAN-ERA5, the first deep learning-based spatio-temporal downscaling of precipitation data on a global scale. SpateGAN-ERA5 enhances ERA5 precipitation data from 24 km and 1 h to 2 km and 10 min, 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 downscaling challenges and supports practical applicability for generating high-resolution precipitation data for arbitrary ERA5 time periods and regions on demand. Trained solely on data from Germany and validated in the US and Australia, considering diverse climates, including tropical rainfall regimes, spateGAN-ERA5 demonstrates strong generalization, indicating robust global applicability. It fulfills critical needs for high-resolution precipitation data in hydrological and meteorological research.https://doi.org/10.1038/s41612-025-01103-y |
| spellingShingle | Luca Glawion Julius Polz Harald Kunstmann Benjamin Fersch Christian Chwala Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI npj Climate and Atmospheric Science |
| title | Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI |
| title_full | Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI |
| title_fullStr | Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI |
| title_full_unstemmed | Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI |
| title_short | Global spatio-temporal ERA5 precipitation downscaling to km and sub-hourly scale using generative AI |
| title_sort | global spatio temporal era5 precipitation downscaling to km and sub hourly scale using generative ai |
| url | https://doi.org/10.1038/s41612-025-01103-y |
| work_keys_str_mv | AT lucaglawion globalspatiotemporalera5precipitationdownscalingtokmandsubhourlyscaleusinggenerativeai AT juliuspolz globalspatiotemporalera5precipitationdownscalingtokmandsubhourlyscaleusinggenerativeai AT haraldkunstmann globalspatiotemporalera5precipitationdownscalingtokmandsubhourlyscaleusinggenerativeai AT benjaminfersch globalspatiotemporalera5precipitationdownscalingtokmandsubhourlyscaleusinggenerativeai AT christianchwala globalspatiotemporalera5precipitationdownscalingtokmandsubhourlyscaleusinggenerativeai |