A field study of reflectivity and Z-R relations using vertically pointing radars and disdrometers

Observations from a 16-month field study using two vertically pointing radars and a disdrometer at Wallops Island are analyzed to examine the consistency of the multi-instrument observations with respect to reflectivity and Z-R relations. The vertically pointing radars were operated at S and K bands and had a very good agreement in reflectivity at a gate centered on 175 and 177 m above ground level over a variety of storms. This agreement occurred even though the sampling volumes were of different size and even though the S band measured the reflectivity factor directly, whereas the K-band radar deduced it from attenuated K-band measurements. Indeed, the radar agreement in reflectivity at the collocated range gates was superior to that between the disdrometer and either radar. This is attributed in large part to the spatial separation of the disdrometer and radar sample volumes, although the lesser agreement observed in a prior collocated disdrometer-disdrometer comparison suggests the larger size of the radar sample volumes as well as the better overlap also play a role. Vertical variations in the observations were examined with the aid of the two radar profilers. As expected, the agreement between the disdrometer reflectivity and the reflectivity seen in the vertically pointing radars decreased with height. The effect of these vertical variations on determinations of Z-R relation coefficients was then examined, using a number of different methods for finding the bestfitting coefficients. The coefficient of the Z-R relation derived from paired disdrometer rain rate and radar reflectivity decreased with height, while the exponent of the Z-R relation increased with height. The coefficient and exponent of the Z-R relations also showed sensitivity to the choice of derivation method [linear and nonlinear least squares, fixed exponent, minimizing the root-mean-square difference (RMSD), and probability matching]. The influence of the time lag between the radar and disdrometer measurements was explored by examining the RMSD in reflectivity for paired measurements between 0- and 4-min lag. The nolag conditions had the lowest RMSD up to 400 m, while 1-min lag gave the lowest RMSD at higher heights. The coefficient and exponent of the Z-R relations, on the other hand, did not have a significant change between no-lag- and 1-min-lag-based pairs. © 2009 American Meteorological Society.