Freight Footprints and Urban Heat Islands: Interactions between Freight Facilities, Built Environment and Thermal Consequences

Previous studies on urban heat island (UHI) and its influencing factors have focused on built environment and urban functional areas. However, freight facilities, large-scale facilities with impervious surfaces and high albedo characteristics widely used in the era of e-commerce, have not been fully examined in terms of their potential impacts on land surface temperature (LST). In this study, we applied satellite image datasets and machine learning techniques to identify freight facilities, and employed K-means clustering, multiple linear regression (MLR), and eXtreme Gradient Boosting (XGBoost) to investigate how different aggregation patterns of freight facilities, in conjunction with built environmental factors, differentially affect LST. The results show that this method can accurately capture the freight building facilities compared to other data forms. Second, freight facilities have a significant positive impact on LST; the existence of freight facilities will significantly increase the annual average LST by approximately 0.25°C, an effect that amplifies substantially to 0.393°C during summer. Subsequently, we categorized grids into three groups based on the scale, quantity, agglomeration level, and location of freight facilities, and validated the heterogeneous effects of their combination with built environment on LST. The warming effect of impervious surfaces is substantially amplified in freight facilities area, while ecological functions are suppressed or even nullified in highly concentrated area. Additionally, complex and non-linear relationships between the built environment and LST reflect across different freight clusters. This study provides actionable insights for planners and policymakers to develop freight facility planning strategies that prioritize ecological sustainability and long-term development.