Welcome to the Colorado State University Machine-Learning Probabilities Prediction Webpage!

Our research specializes in the prediction of extreme weather hazards via statistical postprocessing techniques. Forecast products are generated via Random Forest machine learning models, which predict the occurrence of hazards associated with deep convection (e.g., flash flooding, tornadoes, hail, and wind). You can learn more about our products in these peer-reviewed publications: Herman and Schumacher (2018), Herman and Schumacher (2018b), Hill et al. (2020), Schumacher et al. (2021), Hill and Schumacher (2021), and Hill et al. (2023).

Use the links above to browse our severe weather and extreme precipitation random forest probabilistic forecasts, as well as verification plots of our forecasts. All questions about the products and forecasts can be directed to Dr. Aaron Hill (aaron.hill@colostate.edu)

Summaries of these projects in CSU's SOURCE:
From research to real world: CSU atmospheric scientists develop heavy rainfall forecast tool used nationwide
CSU machine learning model helps forecasters improve confidence in storm prediction

This research is supported by National Oceanic and Atmospheric Administration Joint Technology Transfer Initiative Grants NA20OAR4590350 and NA21OAR4590187.