10.5 Verification of Nssl’s Multisensor Precipitation Estimation Techniques in Arizona

Accurate estimates of precipitation type and amount are important for water management at the Salt River Project (SRP) located in Arizona. The primary focus of water operations at the SRP is management of reservoirs on the Salt and Verde Rivers. Knowledge of precipitation accumulating in watersheds (WSs) has historically been provided by a network of rain gauges located throughout and adjacent to the WSs. While gauge data are considered ground truth, they do not provide temporally or spatially continuous coverage of precipitation falling into the WSs. Also, gauge measurements may not be representative of the character of the precipitation on a case-by-case basis. Finally, one or more gauges may have reporting errors. To overcome these deficiencies, the National Severe Storms Laboratory (NSSL), in partnership with the SRP, has developed and deployed a multisensor (MS) system designed to provide widespread quantitative precipitation estimates. The Quantitative Precipitation Estimation and Segregation Using Multiple Sensors (QPE SUMS) system (Gourley et al. 2001) combines satellite, numerical model, multiple radar, lightning, surface data, and rain gauge data to provide a number of precipitation products, including precipitation type and multi-hour accumulations. As an example of its uniqueness to specific problems encountered in Arizona, QPE SUMS employs a brightband identification (BBID) algorithm that identifies situations that result in gross overestimates in traditional radar-only products. Overestimation is prevented by employing a methodology of using the correlation between satellite infrared (IR) cloud top temperatures and calculated precipitation rates. In addition to improving the magnitude of precipitation estimates, the satellite technique also greatly increases the coverage area. While QPE SUMS has been employed in Arizona for several years, a systematic study of its performance has not been conducted (mainly due to lack of significant precipitation events). This paper provides a preliminary evaluation of QPE SUMS products for the 13-14 February 2003 event. An overview and description of key components and parameters of QPE SUMS are provided in section 2. Section 3 provides background on the SRP domain including the Salt and Verde WSs and associated gauges used for verification. Section 4 describes