Scientists at the National Center for Atmospheric Research (NCAR) have developed a breakthrough method using unsupervised machine learning to accurately predict summer monsoon rainfall across Arizona and New Mexico months in advance. This method analyzes atmospheric variables such as temperature, humidity, winds, and pressure from long-range weather model forecasts, particularly leveraging the European Centre for Medium-Range Weather Forecasts (ECMWF) model known for reliable atmospheric predictions. The improved forecasts enable water managers to make informed decisions on reservoir storage and release, helping to better manage dwindling water resources amid the ongoing megadrought in the American Southwest.

Join for free to read