Comparison of ground-based radar and geosynchronous satellite climatologies of warm season rainfall over the United States

Warm season continental rainfall exhibits organization over a wide range of scales, from 1 to >1000 km. Over the continental U.S., individual rainfall event statistics; diurnal and semi-diurnal cycles; and sub-seasonal rainfall variations are captured rather well by the meteorological radar (WSR-88D) network. Though local uncertainties in instantaneous rainfall rate may approach a factor of two, this event-resolving dataset is genuinely unique with respect to continental climatology applications. Dynamically-based climatology findings, based on such data, have been published by Carbone et al. 2002 [JAS]; Tuttle and Carbone 2004 [MWR]; and Ahijevych et al. 2004 [MWR].

Continental regions outside the United States have insufficient radar coverage to produce event-resolving climatologies. However, geostationary satellite data at infrared frequencies are almost universally available. Properly conditioned, these data can serve a similar purpose, especially when augmented by radar and passive microwave datasets.

This paper compares warm season radar-rainfall statistics over the U.S. with those derived from GOES infrared brightness temperature data. We examine the principal regimes of inconsistency and uncertainty. A transfer function between the two databases is explored as a means to minimize relative bias in the statistics of event phase, phase speed, spatial span and duration. Such statistics infer both local and global auto-covariance properties of warm season rainfall within the diurnal cycle and with respect to topographical forcings caused by the Rockies, Appalachians, and sea/land breeze modulations.

The principal motivation for this study is to reduce the relative bias and uncertainty inherent to similar climatologies, which are currently being developed for other continents It is anticipated that intercomparisons over the U.S. will lend insight to the relative (radar-satellite) biases and therefore reduce uncertainty in the climatological statistics of other continents.