Liberal Arts and Sciences
Wen Lab
Led by Yixin Wen · Atmospheric Sciences(unverified)
What we study
The Wen Lab uses satellite data and machine learning for weather and water research. A recent student project builds a symbolic regression model to predict rainfall from infrared bands; papers cover precipitation downscaling and hydrologic AI agents.
What working here is like
- Computational
- No organisms (data only)
- Coding required
- Project
- 1
- Papers
- 5
Current projects
Symbolic Deep Learning in Meteorology: Distilling Physical Laws from Massive Sensing and Simulation Data
The lab builds a symbolic regression model to predict rainfall from satellite infrared data. Students will select the best satellite bands, run the model, and compare predictions to radar‑based rainfall measurements.
From the Spring 2025 UF Undergraduate Research Symposium · abstract book, p. 581
Summaries are written by AI in plain language from public symposium abstracts and may be imperfect.
Recent papers, explained
- Efficient Kilometer‐Scale Precipitation Downscaling With Conditional Wavelet Diffusion
2026 · Journal of Geophysical Research Machine Learning and Computation
Creates a fast AI model to turn coarse satellite rain data into detailed, high-resolution precipitation maps.
- AI Agent for Hydrologic Modeling: Definition, Development, and Application
2026 · Geophysical Research Letters
Builds an AI agent that turns natural-language flood forecasts into data retrieval, model runs, and reports, tested on a Texas flash flood.
Show 3 more papersShow fewer
- Advancing Science and Applications by Enhancing Data Products, Services, and User Support
2026 · Digital Repository at the University of Maryland (University of Maryland College Park)
Highlights the need for open, accessible Earth science data to support climate research and disaster response.
- Recent oceanic performance of GPM multisatellite precipitation estimates benchmarked by passive aquatic listeners
2025 · Journal of Hydrology
- Accelerating Earth Science to Action
2025 · Bulletin of the American Meteorological Society
Proposes a community platform to speed up the testing and use of Earth science data products.
Plain-language summaries are AI-written and reviewed by us; check the paper for details.
How to reach out
Yixin Wen hasn't set up this page yet. You can also look them up in the UF directory (search “Yixin Wen”).