An intercomparison has been conducted between two gauge-based precipitation climatologies, the PRISM of Daly et al. (1994) and the PREC/L of Chen et al. (2002), to examine how orographic enhancements in precipitation may impact the quantitative accuracy in the interpolated precipitation fields over mountainous areas. In the PRISM, the monthly precipitation climatology is calculated via a rainfall - elevation relationship that has been established empirically for each calendar month by local comparisons, while in the PREC/L it is defined by an inverse-distance interpolation of gauge observations with no orographic considerations.

First, the PRISM and PREC/L mean precipitation fields are compared on a 0.5o lat/lon grid over the contiguous United States. While good agreement in both quantitative magnitude and spatial distribution patterns are observed between the two climatologies over areas with relatively flat terrain, significant underestimates are reported in the PREC/L compared to the PRISM over the mountainous areas in the western United States. Quantitative examinations of the two climatologies over the mountainous areas reveal a clear relationship between the departures in precipitation climatology and the grid box elevation, implying that the differences between the PREC/L and the PRISM are related to differences in the manner in which orographic effects are handled.

To understand how precipitation is influenced by topography, the relationship between the monthly station precipitation climatology and station elevation is examined statistically using the GHCN Version 2 data set. Orographic enhancements in precipitation are observed over all mountainous areas examined in this study. In a selected region, the monthly precipitation climatology at a station increases with its elevation at a rate that differs for different seasons and for different locations relative to the wind direction. This rate of orographic enhancement in precipitation, however, can be expressed very well as a linear function of the mean precipitation over the region regardless of season and relative location. The existence of this linear relationship is confirmed for all of the regions examined in this study, but the coefficients in the relationship differ regionally.

The preliminary results of this study indicate that the underestimates of precipitation in the PREC/L are caused primarily by the lack of a procedure to account for orographic enhancements and that improvements are possible by making use of the quantitative relationship between the precipitation and elevation. Further work is underway to investigate how the linear relationship between precipitation enhancement and elevation changes over various regions and if or not it is possible to approximate the changes as a function of large-scale parameters so that a PRISM-like technique can be applied easily to improve the quality of the gauge-based precipitation analyses over various parts of the global land areas.