Quantitative Precipitation Estimates Using Machine Learning Approaches with Operational Dual-Polarization Radar Data

Abstract

Traditional radar-based rainfall estimation is typically done by known functional relationships between the rainfall intensity (R) and radar measurables, such as R–Zh, R–(Zh, ZDR), etc. One of the biggest advantages of machine learning algorithms is the applicability to a non-linear relationship between a dependent variable and independent variables without any predefined relationships. We explored the potential use of two supervised machine learning methods (regression tree and random forest) in rainfall estimation using dual-polarization radar variables. The regression tree does not require normalization and scaling of data; however, this method is quite unstable since each split depends on the parent split. Since the random forest is an ensemble method of regression trees, it has less variability in prediction compared with regression trees, but consumes more computer resources. We considered several different configurations for machine learning algorithms with different sets of dependent and independent variables. The random forest model was appropriately tuned. In the test of variable importance, the specific differential phase (differential reflectivity) was the most important variable to predict the rainfall rate (residual that is the difference between the true rainfall rate and the one estimated from the R–Z relationship). The models were evaluated by 10-fold cross-validation. The best model was the random forest model using a residual with the non-classified training set. The results indicated that the machine learning algorithms outperformed the traditional R–Z relationship. Then, we applied the best machine learning model to an S-band dual-polarization radar (Mt. Myeonbong) and validated the result with ground rain gauges. The results of the application to radar data showed that the estimates of the residuals had spatial variability. The stratiform and weak rain areas had positive residuals while convective areas had negative residuals, indicating that the spatial error structure driven by the R–Z relationship was well captured by the model. The rainfall rates of all pixels over the study area were adjusted with the estimated residuals. The rainfall rates adjusted by residual showed excellent agreement with the rain gauge, especially at high rainfall rates.

Polarimetric Retrieval of Raindrop Size Distribution: Double‐Moment Normalization Approach and Machine Learning Techniques

“…Two cameras were installed at different heights to calculate the fall velocity of rainfall drops. We discarded the drops with unrealistic fall velocity of rain, the outliers that occurred by splashing and mismatching, and the discontinuous N(D) where one or more zero concentration bins exist between non‐zero concentration bins (Bang et al., 2020; Shin et al., 2021).…”

“…As a result, our training data set consists of measurements with and without random noise. In addition, to account for the precipitation system advection and to reduce measurement noises, we also included 5‐min time averaged polarimetric radar variables ( Z h , 5min , Z dr , 5min , K DP, 5min , and A H , 5min ) (Shin et al., 2021). To mitigate the impact of hail particles, time periods were eliminated from the analysis if the Z H exceeded 55 dBZ (Bringi & Chandrasekar, 2001).…”

Observational Characteristics of Dual-Polarization Radar for Heavy Rainfall Events in Capital Area

“…기후변화 영향으로 국내의 강우 현상이 불규칙적으로 변하 고 있다. 최근 5년 변화를 살펴보면 2016년과 2017년에는 강우 부족으로 지역적인 가뭄이 심각하게 나타났으나, 이후 강우량이 점차 증가하여 2020년에는 수도권에 기록적인 폭우 현상이 발생하였다 (KMA, 2016;2017;2020a) (Matrosov, 2021;Misumi et al, 2021;Reinoso-Rondinel and Schleiss, 2021;Shin et al, 2021;Thurai et al, 2021;Wang et al, 2021) (Meischner, 2005). 비차등위상은 주어진 거리에서 수평파와 수직파의 위상차 를 나타낸는 것으로 차폐, 지형, 이상전파의 영향을 적게 받기 때문에 반사도와 함께 강수 분석에 이용된다 (Kumjian, 2013).…”

Artificial Intelligent For Rainfall Estimation In Tropical Region : A Survey

“…Shin et al [13] used regression tree and random forest which were supervised machine learning methods with dual-polarization radar variables for rainfall estimation. The authors proved that the machine learning algorithms exceed the traditional R-Z relationship in the result.…”