Overview of the analytical framework and key results. (IMAGE)

Chinese Society for Environmental Sciences

Overview of the analytical framework and key results.

Caption

Overview of the analytical framework and key results. The study constructed a heat‑attributable mortality fraction as the dependent variable using time‑series epidemiological modelling and meta‑regression, and compiled explanatory variables including temperature, air pollution (PM2.5 and O3), built‑environment, demographic, and socioeconomic factors. Eight models were compared, with geographically neural network weighted regression (GNNWR) showing superior predictive performance and generalisation. The framework enables high‑resolution (100 m grid) risk mapping and identifies PM2.5 as the second strongest risk factor for heat‑related mortality, after temperature. GWR, geographically weighted regression; MGWR, multi‑scale GWR; GWRF, geographically weighted random forest; GWRBoost, geographically weighted gradient boosting; RF, random forest; OLS, ordinary least squares; MM, multiplicative model.

Credit

Environmental Science and Ecotechnology

Usage Restrictions

Credit must be given to the creator.

License

CC BY

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.