Adaptive Estimation of Local Rainfall from Radar Intensity using Rule-based Approach on Temporal and Spatial Data

• Main Authors: Rachaneewan Talumassawatdi, Chidchanok Lursinsap, Yan Yin
• Language: English
• Published: Science Faculty of Chiang Mai University 2019
• Subjects: Z-R relation; rainfall estimation; rule-based classification; regression analysis; similarity measures
• Online Access: http://it.science.cmu.ac.th/ejournal/dl.php?journal_id=6823; http://cmuir.cmu.ac.th/jspui/handle/6653943832/66121
• Institution: Chiang Mai University

To achieve the highest accuracy of rainfall estimation using radar measurements, the parameters a and b in Z=aRb relation must be adaptively computed from the local relevant factors such as rain intensity, cloud types, duration of rain, etc. In this paper, a new and practical method to compute the values of a and b is introduced. The new method considered the effects of the following factors, i.e. cloud-rain type, ratio of gauge rain intensity(G) with radar rain intensity(R) for the computation of a and b. A rule-based classification concept was deployed to classify the relevant factors into seven cases and the technique of regression analysis was applied to derive the values of a and b. To evaluate the performance of the proposed method in terms of G/R ratio, the method was tested with data collected from S-band radar in the central areas of Thailand. Compared with the traditionally used formulas of Z=200R1.6, Z=300R1.4, and general probability matching method, the new Z-R relation achieved higher accuracy by approximately 10-30%. Furthermore, a new concept of similarity measure was introduced to select the appropriate rain gauge as the representative of any rain gauge with incomplete data.