Merged CaPA and MPE
The "Merged" dataset is created using both the Canadian Precipitation Analysis (CaPA) and the Multi-sensor Precipitation Estimate (MPE) datasets. Outside of the contiguous 48 United States, the CaPA data is copied as it is. Within the contiguous 48 states, the MPE data is resampled to match CaPA’s cellsize, and then copied. Within 10 km of the boundary of the contiguous 48 states, and over the Great Lakes, the data from CaPA and MPE are combined and interpolated.
Canadian Precipitation Analysis (CaPA)
The Regional Deterministic Precipitation Analysis (RDPA) configuration of the Canadian Precipitation Analysis (CaPA) is a real-time gridded precipitation product provided by Environment and Climate Change Canada (ECCC). The grid has a resolution of approximately 10 km and the domain covers all of North America (Canada, USA and Mexico). It uses gauge data, radar reflectivity and GOES imagery to modify a trial field provided by the GEM numerical weather prediction model using a statistical interpolation technique.
Multi-sensor Precipitation Estimate (MPE)
Hydro-meteorologists at the National Weather Service's (NWS's) 13 River Forecast Centers (RFCs) utilize the Multi-sensor Precipitation Estimator (MPE) to produce rainfall estimates that cover the 48 contiguous United States as well as portions of Canada and Mexico. Data from many sources are manually analyzed at the RFCs to produce the day's best precipitation estimate on a 4 km grid on a 1 hour time step.
Departures are calculated from 15-year averages (2004-2018). For each dataset (CaPA, MPE, and Merged), daily averages for each calendar day were created.