Research on Multi-model Intelligent Forecasting Methods for Areal Rainfall in the Red River Basin
LI Jinling
HE Chunjiang
HUANG Heng
OU Chunmiao
Abstract：To enhance the forecasting capability of areal rainfall in river basins,grid-based precipitation forecast products by numerical models from ECMWF,Japan,Germany,and GRAPES_MESO were utilized.The bilinear interpolation method was applied to interpolate grid data to station points,and the least squares method was used to calculate correlation coefficients between automatic station observations and numerical forecasts from each model.The weights of each model were determined through linear regression,and a mul-tivariate dynamic regression method was employed to dynamically adjust these weights,establishing forecas-ting equations for areal rainfall in each basin.Taking April-June 2021 as an example,the mean absolute er-ror method was applied to evaluate the performance of intelligent forecasting and forecasters'subjective pre-dictions in various catchment areas.The results indicate that intelligent forecasting outperforms subjective forecasting by forecasters during the same period,demonstrating promising application prospects.This ap-proach improves the precision and accuracy of areal rainfall forecasts,leveraging meteorological predictions for reservoir regulation,load adjustment,and water level control.
Keywords：weather forecastingareal rainfallmultivariate dynamic regressionmulti-model intelligent forecastingHongshui River Basin
Publication Date：2025-04-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 48-51,57 )