Thursday, 7 August 2003: 12:00 PM
Correcting radar derived rainfall rates at the surface using Doppler velocities
It is ideal for hydrologists, especially urban hydrologists, to incorporate precipitation information from radar observation in their models rather than using scattered point measurements from rain gauges. However, precipitation estimates derived from radar need to be accurately represented on the surface for these applications. This study is a simplified examination of the error in surface precipitation rate caused by the wind drift of precipitation. One of the assumptions made in estimating rainfall rates from radar reflectivity is that rain falls vertically; applying this assumption leads to spatial errors as rain falls from cloud base to the surface. These errors can affect hydrological model results, which depend on the precision of rainfall estimation. Using Doppler wind fields near cloud base to estimate a horizontal wind component, combined with a vertical velocity of rainfall, can lead to better placement of precipitation quantities on the surface. This theoretical advection of precipitation from cloud base to ground level takes into account a horizontal wind from cloud base and applies it to gridded rainfall rates with varying average fall speed. The purpose of this theory is to examine the errors on the surface associated with the current assumption of vertical rainfall having no horizontal advection. Furthermore, different grid resolutions ranging from 2 km to 0.25 km will be used to examine the errors in reducing grid resolutions that are on the scale of urban hydrological models. Case studies of actual events and more complicated wind drift models will be used in the future to incorporate the effects of evaporation rates, drop sorting, terrain slopes and complex wind fields.
Supplementary URL: