WSR-88D quantitative precipitation estimates along the northern California coast during the 1998 El Niño

The winter season storms associated with the 1998 El Niño had a substantial impact on California, especially its coastal regions. A series of storms in January and February produced heavy rainfall and extensive flooding, which led to the loss of 33 lives and more than $300 million in property damage. The network of 10 cm Doppler radars operated by the NWS (WSR-88D) was designed in part to monitor these types of events and provide quantitative precipitation estimates (QPE). However, the mountainous terrain in coastal California leads to difficulties in obtaining accurate estimates. This is due to factors such as the siting of radars at relatively high altitudes (300-1000 m above sea level) to limit topographic obscuration. In combination with the relatively shallow (at times < 4 km) nature of coastal orographic precipitation, this scenario can lead to underestimates of precipitation rate because most of the precipitating cloud is below the lowest scan of the WSR-88D. However, overestimates of precipitation rate can occur if the lowest scan of the WSR-88D intercepts melting ice particles in the radar bright band.

A better understanding of the relationship between precipitation rate at the levels scanned by the WSR-88D and precipitation rate at or near the surface is needed to improve these estimates, an effort that requires knowledge of the vertical profile of precipitation rate throughout the precipitating cloud. To address this issue, our study uses WSR-88D data from San Francisco (KMUX) in combination with observations collected during the 1998 California Land-falling Jets (CALJET) experiment. The key CALJET observations are from a 10 cm vertically pointing Doppler radar located in the coastal mountains north of San Francisco. This instrument provides the vertical structure of precipitation from the surface up to cloud top, including those regions below the scanning level of KMUX.

Analysis of the vertically pointing Doppler radar data has revealed distinctly different vertical precipitation structures in the cool, warm, and cold sectors of land-falling extratropical cyclone systems. In addition to variations in the height of the bright band, the rate of reflectivity decrease with height above the bright band differs between each of these regimes. These results have relevance to the QPE problem since this type of information can be used to make corrections to flawed WSR-88D precipitation estimates. In this presentation, we will show how the corrections are calculated and applied, the subsequent improvements in QPE, and possible limitations in the technique.