A Comparison of Radar Rainfall Estimates and Rain Gage Measurements during Two Denver Thunderstorms
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH
Pagination or Media Count:
It has long been hoped that rainfall measurements made with weather radars would be able to supplement or even replace rain gage networks. During the summer of 1991 Colorado State Universitys CSU Department of Atmospheric Science participated in an experiment which was intended, in part, to determine how useful radar measurements are in helping to estimate heavy rainfall from thunderstorms. A dense array of 41 automatic rain gages operated by the Denver Urban Drainage and Flood Control District UDFCD provided precipitation measurements to compare with the radar results. Two thunderstorms were observed the first occurred on June 6 and produced up to 43.7 mm 1.72 in of rain in 85 minutes, while the August 5 storm resulted in peak rainfall amounts of 37.8 mm 1.49 in in 60 minutes. Unfortunately, estimates of rainfall made by using a relation between the radar measured reflectivity Z and the rain rate R are frequently in error by large amounts. The radar estimate over gage 540 during the June 6 storm was 58.8 mm 2.31 in higher than the gage amount, an error of 201. During the August 5 event gage 1720 recorded 37.8 mm, while the radar estimate was 27.8 mm 1.09 in lower. Such large errors are unacceptable for most meteorological or hydrological purposes. The use of different Z-R relations provided overall improvement during both storms, while reflectivity thresholds resulted in better estimates during the June 6 storm, in which hail was present.