Assessing the role of urban green infrastructure in mitigating summertime Urban Heat Island (UHI) effect in metropolitan Shanghai, China

The importance of urban green infrastructure (UGI) in ameliorating the urban thermal environment and reinforcing urban climatic adaptation has attracted growing concerns. In this study, 36 representative urban thermal hotspots (UTHSs) in Metropolitan Shanghai, which vary in urban setting features and exposure to surface urban heat island (SUHI) effect, were used to quantitatively assess the UGI's role in mitigating the SUHI effect measured with the satellite-retrieved land surface temperatures (LSTs). We observed significantly contrasting LST differences between UTHSs ranging from highly to slightly urbanized areas. Moreover, we found the contrasting performances of forward stepwise regression (FSWR), partial least squares regressions (PLSR), and random forest regressions (RFR) models for quantitatively assessing the UGI's role in mitigating the SUHI effect. The FSWR models showed high r2 (averaged 0.816) but presented the worst performance for interpreting the results; PLSR models showed the highest r2 (averaged 0.820), presenting the best performance for interpreting the results; RFR models showed the lowest r2 (averaged 0.708), presenting an acceptably moderate performance for interpreting the results. Our findings provided an applicable methodology with repeatability for assessing UGI's role in mitigating the SUHI effect elsewhere, regarding the complexity of urban setting features.