Estimating rainfall for short-term, high-impact rainfall events is made difficult by instrument limitations. Infrared and visible data are available up to every 15 minutes from geostationary-orbiting sensors, but are sensitive to only the properties of the cloud tops and hence have limited utility for estimating rainfall. On the other hand, data from microwave sensors are much more sensitive to vertical cloud properties but are available only from polar-orbiting instruments.

A number of algorithms have been developed that attempt to combine the best features of both data sets for short-term rainfall estimation, where accuracy and latency are both critical. Among these is the Self-Calibrating Multivariate Precipitation Retrieval (SCaMPR), which has been running in real time at the NOAA/NESDIS Center for Satellite Applications and Research (STAR) since late 2004. A number of enhancements have been made to SCaMPR, including the use of lightning data and a rainfall regime classification scheme, and the impacts of these changes for short-term, high impact rainfall events are described in this presentation.