A topic of current interest is whether global climate change can be observed in the changing characteristics of precipitating weather systems. Climate research suggests likely changes in the intensity, frequency, and duration of precipitation events. Since precipitation occurs in association with systems that evolve and move in time, explicit analysis of these characteristics requires an entity-based or object-oriented approach. Such an approach involves the use of automated procedures to identify weather systems of interest and measure the appropriate characteristics of each system. This type of analysis has been typically performed for specific weather phenomena via visual inspection of weather data. Such visual analysis is very labor intensive, therefore studies have been limited to a relatively short time period (<5 years). Recently developed automated object-oriented analysis techniques allow for the analysis of a much larger data set, such as that obtained from the 10+ year archive of the NCEP 4km Stage II hourly radar-gage national precipitation analyses. Using data mining and image processing tools, a database of sub-daily timescale precipitation systems over the U.S. will be developed. Precipitating weather systems will be characterized by their size, intensity, duration, and spatial organization. These events will also be classified by their temporal/spatial scales, precipitation intensity, and spatial organization, following the automated classification procedures developed by Baldwin et al. (2005). The methodology and results from this ongoing work will be presented at the conference.