Using the Gradient Boosting Decision Tree to Improve the Delineation of Hourly Rain Areas during the Summer from Advanced Himawari Imager Data

Abstract

AbstractA new classification scheme based on the gradient boosting decision tree (GBDT) algorithm is developed to improve the accuracy of rain area delineation for daytime, twilight, and nighttime modules using Advanced Himawari Imager on board Himawari-8 (AHI-8) geostationary satellite data and the U.S. Geological Survey digital elevation model data. The GBDT algorithm is able to efficiently manage the nonlinear relationships among high-dimensional data without being affected by overfitting problems. The new delineation module utilizes several features related to the physical variables, including cloud-top heights, cloud-top temperatures, cloud water paths, cloud phases, water vapor, temporal changes, and orographic variations. The scheme procedure is as follows. First, we perform extensive experiments to optimize the module parameters such that the equitable threat score (ETS) reaches its maximum value. Then, the GBDT-based modules are trained and classified with the optimum parameters. Finally, validation datasets are applied to test the true performance of the GBDT-based modules. The agreement between the estimations and observations of the ground-based rain gauges is verified. Results show that the ETS values of the GBDT-based modules are 0.42 for the daytime, 0.30 for the twilight period, and 0.32 for the nighttime. The cloud water path and cloud phase features make the most significant contributions to the modules. Comparisons drawn with the two probability-related methods show that our new scheme presents great advantages in terms of statistical scores on the overall performance.