Thesis Title:Quantitative Precipitation Estimation Using Polarimetric Doppler Weather Radar Products

Abstract:The proposed research work is to develop better algorithms to improve the accuracy of quantification and characterisation of hydrometeors based on Polarimetric weather radar signatures, especially in Indian terrain. Objectives realised for the study include Quality control algorithms using dual-pol products, generation of advanced data products, microphysical studies of different types of precipitation events, development of dual polarization rainfall estimation algorithms, and validation of Polarimetric Weather Radar Data.Radar data can be contaminated by non-meteorological targets, such as normal ground clutter, anomalous propagation echoes, etc., within the scanning volume. To have considerable accuracy in the radar-based rainfall estimation, these clutters should be removed. The sporadic nature of anomalous propagation echoes makes their removal more difficult than that of normal ground echoes. A new algorithm has been developed for the removal of anomalous propagation and normal ground clutter echoes from radar data. A fuzzy logic-based approach has been followed, which uses the standard deviation of differential phase (SDPHI), vertical gradient of reflectivity (VGZ), dual pol texture ratio (DPTR), and correlation coefficient (rhv) to discriminate precipitation echoes and non-precipitating echoes. The algorithm is tested on data from different independent events. The accuracy obtained from the new algorithm is found to be better than the filtering techniques so far used.The microphysical studies of different types of precipitation events have been carried out. The precipitation characteristics of the Ockhi cyclone from the C Band DWR TERLS data observations have been analyzed. Few studies mention initial DWR observations of tropical cyclone (TC) over this low-latitude region below 23.50 N. A statistical analysis of precipitating clouds in a cyclone from its depression stage to severe cyclonic stage has been carried out. This study tries to classify and quantify the contribution of convective and stratiform rain to the total TC rainfall. The estimation of the contribution of convective and stratiform rain to the total TC rainfall can be a valid input for accurate modelling and forecasting.A novel method for generating QPE from the observations of Polarimetric Doppler Weather radars is being proposed. Following derived products, Constant Altitude Plan Position Indicator (CAPPI), reflectivity (Z), differential reflectivity (ZDR), and specific differential phase (KDP) are being used independently or jointly for estimating surface rainfall intensity. Generally, these values are taken at an altitude of 1Km from the observation of PDWR. Here, we introduce a new method to generate equivalent Z, ZDR, and KDP profiles based on the information derived from the Vertically Integrated Liquid (VIL) product. First, VIL for different altitude layers (layers below 4 Km) is computed across the volumetric scan. Then, the layer at which the maximum VIL value occurs is considered as a seeding parameter for identifying the maximum reflectivity value in the layer, and that is used to generate a Pseudo Z profile. Once the Pseudo-Z profile obtained in this method is used for generating corresponding ZDR and KDP values. These newly generated Pseudo profiles are used to compute the rainfall rate. This algorithm is tested with C Band DWR TERLS and X Band DWR Chennai data. The rain rate estimates from Pseudo profiles are found to be highly correlated with in-situ observations than that is obtained from the CAPPI-based approach. Thus, the new approach will enhance the QPE of Rain events from Polarimetric Doppler Weather Radar observations. Often, radar may fail to capture significant precipitation echoes since it is not scanning all 360 degrees simultaneously. Thus, the CAPPI layer obtained may not be the best approximation of the actual precipitating cloud. Hence, we are trying to make use of all the information gathered by radar in a single volume scan to produce a replica of the CAPPI layer, which will eventually give a better QPE estimate.