Influence of Vertical Air Motion on the Radar Quantitative Precipitation Estimation

The radar quantitative precipitation estimation (QPE) is one of the main purpose of weather radar application. QPE products are applied very well due to wide space coverage, good precision and high spatial and temporal resolution of precipitation information. Main influencing factors cause differences between the QPE from radar and ground observation include the accuracy of the radar reflectivity, inconsistent of spatial and temporal between the radar and surface observation, and complex precipitation particle raindrop spectrum distribution. Air vertical motion effect in precipitation system and its temporal variation of random fluctuation is another important factor. Raindrop spectral distribution is considered with the development of radar QPE in recent years, and its falling speed can be achieved at the same time from PARSIVEL disdrometer. The air vertical motion acquired from data of PARSIVEL can be used to analyze its influence to the radar QPE. Using PARSIVEL data from the Southern China Monsoon Rainfall Experiment (SCMREX) during May and June 2014 at Yangjiang, Guangdong Province, several precipitation processes are analyzed, including 5 stratiform cloud (SC) precipitation events and 6 convective cloud (CC) precipitation events. The vertical air motion is retrieved and their influences on the QPE precision for both SC and CC are analyzed.The vertical air motion influencing value for 5 SC events are between-0.18 mm·h^-1 and-1.05 mm·h^-1, ranging from 13.61% to 13.99%. The vertical air motion influencing value for the six CC events are between 5.44 mm·h^-1 and 24.81 mm·h^-1, ranging from-38.59% to 25.92%. The influence on CC is greater than that on SC. PARSIVEL observation is applied to estimate the A and b coefficient in Z-R relation. The average deviation estimates SC under stationary atmospheric condition is 10.9% and 9.2% under non-stationary atmospheric condition. The vertical air motion effect partly offset by the uncertainty of the estimated precipitation Z-R relation. Average deviations of radar QPE are 25.5%, 51.2% under the stationary and non-stationary atmospheric conditions. After considering the raindrop spectrum distributions, the error of radar QPE is mainly from the vertical air motion. The deviation of QPE is related to data duration, shorter data usually lead to greater deviation. Simulation experiments are also carried out using PARSIVEL data to investigate this influence.