Meteorological Prediction Model of Aedes Density in Sanshui District, Foshan City Based on Random Forest
Abstract：To understand the complex relationship between meteorological factors and Aedes density, this study explores a meteorological prediction model for Aedes density in Sanshui District, Foshan City, to provide technical support for local governments in establishing dengue fever risk early warning mechanisms. Based on Aedes density monitoring data and corresponding meteorological data from March 2021 to October 2024 in Sanshui District, the impact mechanism of meteorological factors on Aedes density was systematically analyzed. Through Pearson correlation analysis, multiple linear regression, and random forest regression models, it was found that Aedes density in Sanshui District is significantly positively correlated with temperature (with the highest correlation coefficient of 0.508) and rainfall factors (with a correlation coefficient of 0.639 for the number of rainy days in a month), and negatively correlated with atmospheric pressure (correlation coefficient of -0.383). A random forest prediction model for Aedes density in Sanshui District was established for the first time, and further identified that the monthly average temperature, number of rainy days, and cumulative monthly rainfall were key driving factors, with their node purity contribution values (IncNodePurity) significantly higher than other variables. Model comparison showed that the random forest regression (R2=0.836) was more suitable for fitting complex nonlinear relationships compared to multiple linear regression (adjusted R2=0.446), with a significant improvement in prediction accuracy.
Keywords：Meteorological MonitoringDengue Fever Prevention and ControlAedes Density Prediction ModelLarval Mosquito Trap IndexRandom Forest AlgorithmSanshui District, Foshan City
Publication Date：2025-12-20
Online Publishing Date：2026-01-07(First online date of this platform, not the publication date of the document)
Pages：4( 67-70 )