Using SVM to Provide Precipitation Nowcasting Based on Radar Data
In recent years, SVM (Support Vector Machine) has been widely used in the field of weather forecasting, especially in the medium and long-term weather forecasting, but it is seldom used in the precipitation nowcasting. Without considering other meteorological factors, this paper uses SVM method in precipitation nowcasting based on the radar images. The statistical results of four difference thunderstorm events shown that the method based on SVM has good performance in the precipitation nowcastings in 0-2 h lead-time forecasting.
KeywordsPrecipitation nowcasting Radar data SVM
This research was financially supported by the Natural Science Foundation of Guangxi (NO. 2018JJA150144,2018GXNSFAA294079) and the National Science Foundation of China (NO. 61562008, 41401524, 4166010274).
- 1.Yu, C., Yongyi, X.: A new method for dealing with non-linear classification and regression problems (1) - brief introduction of support vector machine method. J. Appl. Meteorol. 15(3), 345–354 (2004)Google Scholar
- 2.Feng, H., Chen, Y.: Application of Support Vector Machine (SVM) in weather forecasting, a new method for dealing with non-linear classification and regression problems. J. Appl. Meteorol. 15(3), 335–365 (2004)Google Scholar
- 5.He, J., Chen, J., et al.: A multi-time scale SVM method for local short-term rainfall prediction. Meteorology 43(4), 402–412 (2017)Google Scholar
- 7.Li, H.: Statistical Learning Method, pp. 95–130. Tsinghua University Press (2012)Google Scholar
- 8.Zhou, Z.: Machine Learning. Tsinghua University Press, Beijing (2016)Google Scholar