Prediction of the future urban heat island intensity and distribution based on landscape composition and configuration: A case study in Hangzhou

During rapid urbanization, global land use and land cover change drastically, leading to increasingly severe problems in the urban environment. Among them, the urban heat island effect has attracted much attention owing to its magnification effect on climate change and its negative role in urban livability and sustainable development. In this context, predicting the future surface urban heat island intensity (SUHII) and distribution is important in dealing with the urban thermal environment. In this study, we conducted a prediction of SUHII using a random forest model based on future landscape patterns. According to the results, the built-up area is expected to grow to 3740.67 km(2), and the non-forest green space (GS2) is estimated to decrease by 1130.96 km(2) by 2030. In the context of urban expansion, the area with SUHII of higher than 5 ?C will increase significantly, showing a more concentrated spatial distribution, while the region with SUHII of below -2.5 ?C will decrease in coverage. In addition, our results showed that the scattered built-up areas and GS1, as well as concentrated GS2, will contribute to reducing SUHII. The results are expected to help policymakers and urban planners to design reasonable measures for achieving sustainable urban development.