Research and application of X-band hydrological radar rainfall retrieval model in Hunan Province
HE Bingshun
ZHAO Yanwei
Abstract：To address the limitations of traditional rain gauge networks,such as extensive monitoring blind spots and insufficient spatiotemporal resolution,in mountain flood prevention,this study develops a high-precision rainfall retrieval model based on X-band dual-polarization hydrological radar using observational data from four radars in Hunan Province.Combining traditional empirical modeling with machine learning techniques,a multi-parameter joint retrieval approach was established.In the empirical models,relationships between reflectivity(ZH),specific differential phase(KDP),and differential reflectivity(ZDR)and rainfall intensity were constructed to estimate precipitation.The machine learning models applied algorithms including k-nearest neighbor(KNN),support vector regression(SVR),random forest(RF),artificial neural network(ANN),and recurrent neural network(RNN)for rainfall fitting.Results show that among the empirical models,the ZH-R model performs best when rainfall intensity is below 10 mm/h,whereas the KDP-R model performs better when it exceeds 10 mm/h.Among the machine learning models,the RNN achieves the best performance,with a root mean square error(RMSE)of 3.125,a coefficient of determination(R2)of 0.981,and a Pearson correlation coefficient of 0.992,outperforming both other machine learning and empirical models.The integration of X-band radar with multi-parameter correction and machine learning techniques significantly enhances rainfall retrieval accuracy,providing more reliable data support for the"three defense lines"system in mountain flood disaster prevention.
Keywords：three defense linesrainfall radarrainfall retrievalrecurrent neural networkmountain flood disasterempirical modelmachine learning model
Publication Date：2025-10-30
Online Publishing Date：2025-11-14(First online date of this platform, not the publication date of the document)
Pages：10( 37-46 )