Time series datasets prior to 1981 are modeled using climatologically-aided interpolation (CAI), which uses the long-term average pattern (i.e., the 30-year normals) as first-guess of the spatial pattern of climatic conditions for a given month or day. CAI is robust to wide variations in station data density, which is necessary when modeling long time series. Data is based on monthly modeling.
These datasets use whatever station networks and data sources are available for the relevant period. See PRISM datasets for more information; the data available from this page is “AN81m” (monthly). Note that individual grids contain non-climatic variations due to station equipment and location changes, stations openings and closings, and varying observation times. These datasets should be considered "stable" (unlikely to change until a major version change).