Evaluations on Radar QPE using raindrop size distribution in Southern Luzon, Philippines

  • Southern Luzon shows larger but fewer raindrops compared to neighboring countries.
  • Four R(Z) relations were able to reduce radar-retrieved rainfall bias by up to 61%
  • R(KDP) performed the best by a huge margin in terms of error statistics

The study analyzed the raindrop size distribution (DSD) measured by an optical Parsivel disdrometer in Southern Luzon, Philippines and utilized it to generate dual-pol relations for the nearby Tagaytay radar. The relations were generated using two methods (Method 1 - gamma-based and Method 2 - linear fitting), four timeintegration steps (1-, 2-, 5-, and 10-min) and datasets from two periods (wet season and single event). The resulting quantitative precipitation estimates (QPEs) calculated from the generated R(Z) relations were compared to rain gauge stations near the disdrometer and were evaluated for the Tropical Storm Yagi Monsoon event of 10 August (2200 UTC) to 11 August (0400 UTC) 2018 using six statistics: Pearson’s correlation; mean error, percent bias, Nash-Sutcliffe Efficiency, mean absolute error, and root-mean-square error. Results show that the area’s DSD demonstrates relatively larger average raindrop diameters than some of its Asian counterparts, albeit a smaller number in the total number of raindrops when compared with the same areas. In terms of QPE evaluation, results showed a consistent pattern observed wherein the R(Z) relations using finer time steps (1- and 2-min) generally performed better than the longer ones. Moreover, Method 1 dominated Method 2 in terms of error statistics. As expected, Method 2 outperformed Method 1 in terms of r (as Method 2 itself is derived through linear fit). The best derived R(Z) relations were able to outperform other relations in terms of r, NSE, and RMSE. On the other hand, R(KDP) was able to perform the best in terms of ME, MAE, and pBIAS, reducing the bias of current standard method by up to 74%.

Read 2094 times
© 1990-2033 Terrestrial, Atmospheric and Oceanic Sciences (TAO). All rights reserved.

Published by The Chinese Geoscience Union