An approach for mining the causation of heat island effect based on decision tree

Urban thermal environment is a complicated nonlinear system, which is affected by the integration of various influencing factors inside the city. To explore the mechanism of urban heat island (UHI) effect formation in the city of Shanghai, China, decision tree is applied to establishing the quantitative relationships between urban thermal environment and its influencing factors, which reflect ecological condition, built-up density and intensity, and human heat. The results indicate that it is feasible to simulate the intensity and distribution of land surface temperature (LST) field and predict the tendency of UHI effect in the future by decision tree. This research will help to take effective measures to alleviate heat island effect, and consequently reduce environmental pollution and improve urban environment.