WSR-88D radar-derived rainfall estimates are generated by the Precipitation Processing Subsystem (PPS) within the NEXRAD algorithm set. Mean field and range dependent biases have been observed to be a significant source of error in WSR-88D precipitation estimates. A mean field bias-adjustment algorithm which is part of the PPS has never been used operationally due to unresolved communication issues between raingage data acquisition computers and radar sites. An improved bias adjustment algorithm has been developed and tested which should provide an improved estimate of the mean field bias. The new algorithm has been tested and applied in Stage II multisensor precipitation processing which occurs external to the RPG. A preliminary evaluation of the algorithm in use in real time at River Forecast Centers has been favorable, but, has also pointed out possible areas of improvement. In this paper, an extension of the algorithm in which a long term and short term bias are computed in parallel is discussed and evaluated.
In addition to the mean field bias adjustment, a technique has also been developed to account for the range dependent biases which are due solely to the non-uniformity of the vertical reflectivity profile (VRP). The methodology of applying a correction based on the VRP is discussed and examples are shown. Preliminary results indicate that the technique may work better for tropical events than for shallow stratiform events where a bright band may exist. One advantage of this technique is that it can be applied to NEXRAD data even if the mean field bias has not been computed due to lack of gage observations.