Fusion of multi-source precipitation records via coordinate-based generative models
Sencan Sun, Congyi Nai, Baoxiang Pan, Wentao Li, Lu Li, Xin Li, Efi Foufoula-Georgiou, Yanluan Lin · 2025
Read Fusion of multi-source precipitation records via coordinate-based generative models by Sencan Sun, Congyi Nai, Baoxiang Pan, Wentao Li, Lu Li, Xin Li, Efi Foufoula-Georgiou, Yanluan Lin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12865118.txt.
Description
Sencan Sun and seven co-authors blend rain records from many sources into one better map, in a 2025 paper. Rain is hard to measure well. Gauges are trusted but sparse. Satellites see almost everywhere but with errors, and computer models carry their own bias.
Their system, PRIMER, learns from all three sources at once. A diffusion model first learns broad rain patterns, then fine local detail from gauges. The trained system can fix biased datasets and zoom coarse maps into fine ones. It also worked on live forecasts and future climate runs without retraining. Weak sources, combined well, became a strength.
Catalog record
| CREATOR | Sencan Sun, Congyi Nai, Baoxiang Pan, Wentao Li, Lu Li, Xin Li, Efi Foufoula-Georgiou, Yanluan Lin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |