Search

Improvement of quantitative precipitation estimation (QPE) is one of the primary benefits providedbypolarization radars. Ministryof Land, Infrastructure and Transportation Japan (MLIT) put Shakadakeoperational Cband Radar to practical use in1992. However, large improvement hadnotbeenrealizedbecause of its initial technical stage.Under these circumstances, first, this paper compares the performance of the conventional R (ZHH) relation and three polarimetric relations, which are R (ZHH, ZDR), R (KDP), R (KDP, ZDR) . At lower rainrates, the combinationof KDP and ZDR is less efficient because KDP becomes noisy. Next in order to improveQPE, a few combinations of estimation equations for operational purpose is proposed, considering thecharacteristic ofthose relations. Also validations with some other algorithms are conducted. As a result, the proposed algorithms improve the estimationof hourly rainfall and accumulated total rainfall.

Dual polarimetric radar has been developed and partly utilizedinmeteorology and hydrology in these twenty years because of its potential ability in estimating drop size distribution (DSD) and in discriminating precipitation type. Ministryof Infrastructure, Land and Transportation (MILT) in Japan put Shakadake operational C band Radar to practical use in 1992. However, large improvement had not been realized because of its initial technical stage. Under these circumstances, this study investigates observation precision of the latest C band polarimetric radar for promoting replacement of Japanese operational radars into operational polarimetric radars. As results, this study shows ZHH ZDR and KDP observed bythe polarization radar well fit to non-parametricallycomputed ZHH, ZDR and KDP using observed DSD, respectively. Also, this study shows “Radar observed ZDR-Dm of observed DSD” relation is consistent with observed DSD. Especially, KDP, which is free from attenuation problem, is suitable to estimate the strong rainfall. Parameter from the polarimetric radar has high possibility to improve real-time rainfall estimation.

The vertical profile of raindrop size distribution (DSD) is important for understanding rain processes. To improve the accuracy of rainfall intensity observed by precipitation radar and microwave radiometers carried on satellites, the information on the vertical profile of DSD is also important. The purpose of this study is to develop a database of DSD for the satellites precipitation retrieval by using a 400MHz-band wind profiler and COBRA (CRL Okinawa Bistatic polarimetric Radar) for various kinds of hydrometer (rain, snow, and mix phase). A new algorithm is proposed above the melting layer. As a result, in the case that the spectrum peak of air motion is difficult to be found from the observed Doppler spectrum above the melting layer, it is possible that DSD can be estimated by the proposed algorithm.

To improve the accuracy of the radar-estimated rainfall, we observe and analyze the vertical profile of the rain drop size distribution (DSD), then create a new formulation of vertical profile of DSD related to rainfall type and build up an algorithm for estimating ground rainfall using radar. Observation of the vertical profile of DSD is carried out by a vertical pointing VHF Doppler radar named the MU (Middle and Upper) Radar, which can detect vertical profiles of Doppler spectrum that composed of bothrain drop itself and air movements.

The rainfall intensity on the ground is usually not equivalent to the rainfall intensity estimated from the conventional radar. To improve the accuracy of the radar-estimated rainfall, we observe and analyze the vertical profile of the rain drop size distribution (DSD), then create a new formulation of the vertical profile of DSD. Based on the obtained information, first, a new formulation of DSD which considers rainfall type, rainfall intensity and observed DSD from the Disdrometer is developed. Then, the analysis of the vertical profile of DSD in the selected rainfall type is carried out to investigate the possibility of determing relationships among parameters for the DSD.