dc.description.abstract | The study uses Chung-Li 52MHz VHF radar and two dimensional video disdrometer (2DVD or disdrometer) to investigate precipitation phenomenon in the air. We can obtain precipitation terminal velocity, echo power, spectral width, air velocity, air echo power, air spectral width, rainfall rate and drop size distribution (DSD) from each apparatus. The precipitation particles will attenuate electromagnetic wave propagation in the air, so the study of their characteristic is very important, especially for DSD in the air.
In the past few years, many researches suggested that Gamma DSD was the closest to real DSD. The study through the deconvolution method to retrieve Gamma DSD parameters N_0 、μ、Λ and further important parameters such as mass-weighted averaged diameter (D_m), median volume diameter (D_0) and rainfall rate (R) etc. in the air. We simulate radar precipitation spectrums and corresponding DSDs through the theoretical formula in order to confirm the deconvolution method can efficiently be used in radar observed data, and then consider the errors between deconvolution results and theoretical values with different signal-to-noise ratio (SNR). The results show the errors reduce while SNR increase.
The data in this study is the precipitation events from June 22th 22:10LT to June 23th 18:00LT, 2019. We compare precipitation parameters D_m and R retrieved by traditional method and deconvolution process with ground-based 2DVD and find that deconvolution method is more closer to the 2DVD observation then tradition does. Next, according to precipitation characteristics, we separate the precipitation events into two categories, stratiform and convective precipitations. The statistical parameters, i.e., correlation coefficient, root-mean-square error and cross correlation function are calculated to compare radar retrieved parameters with 2DVD. Finally, both categories improve their results after the time shift. However, due to the different air velocity, the time delay will also be different. Last, we investigate the relationship between precipitation terminal velocity and its spectral width. The scatter diagrams show that the performances become closer to the pattern of disdrometer’s results when SNR increase, so the level of SNR be an important index when retrieving precipitation parameters in the air. | en_US |