Examining the Relationship between Land Use/Land Cover (LULC) and Land Surface Temperature (LST) Using Explainable Artificial Intelligence (XAI) Models: A Case Study of Seoul, South Korea

Abstract

Understanding the relationship between land use/land cover (LULC) and land surface temperature (LST) has long been an area of interest in urban and environmental study fields. To examine this, existing studies have utilized both white-box and black-box approaches, including regression, decision tree, and artificial intelligence models. To overcome the limitations of previous models, this study adopted the explainable artificial intelligence (XAI) approach in examining the relationships between LULC and LST. By integrating the XGBoost and SHAP model, we developed the LST prediction model in Seoul and estimated the LST reduction effects after specific LULC changes. Results showed that the prediction accuracy of LST was maximized when landscape, topographic, and LULC features within a 150 m buffer radius were adopted as independent variables. Specifically, the existence of surrounding built-up and vegetation areas were found to be the most influencing factors in explaining LST. In this study, after the LULC changes from expressway to green areas, approximately 1.5 °C of decreasing LST was predicted. The findings of our study can be utilized for assessing and monitoring the thermal environmental impact of urban planning and projects. Also, this study can contribute to determining the priorities of different policy measures for improving the thermal environment.

Spatiotemporal Thermal Variations in Moroccan Cities: A Comparative Analysis

“…Interestingly, in many cases, XGBoost outperformed LightGBM. Our methodology parallels other studies that have successfully employed similar techniques in analyzing urban heat dynamics [47]. Furthermore, the use of the SHAP-based explanation for interpreting the ML models has lent additional transparency and interpretability to the study, an aspect that aligns with recent calls for explainable AI [60].…”

“…For instance, Zhou et al ( 2022) leveraged the XGBoost model and the SHAP method, among other techniques, to explore the relationship between urban landscape structure and LST [46]. Similarly, Kim et al (2021) applied XGBoost and SHAP models to develop an LST prediction model for Seoul, South Korea [47]. Both studies underscored the significant influence of certain environmental factors on LST, thereby validating the utility of these ML algorithms in deciphering complex environmental relationships.…”

Exploring the Effects of Land Use/Land Cover (LULC) Modifications and Land Surface Temperature (LST) in Pune, Maharashtra with Anticipated LULC for 2030

“…The relationship between land use/land cover (LULC) and land surface temperature (LST) in Seoul was investigated by employing the explainable artificial intelligence (XAI) approach. Integrating XGBoost and SHAP models, the study maximizes LST prediction accuracy, identifying surrounding built-up and vegetation areas as significant factors influencing LST.The results provide valuable insights for assessing and monitoring the thermal environmental impact of urban planning and projects, aiding in the determination of policy priorities for improving the urban thermal environment [16].…”

Monitoring wetland evolution in the Yellow River delta from 1990 to 2020 based on long time series landsat data

“…After the heat energy of the sun is radiated to reach the ground, part of it is reflected and part of it is absorbed by the ground, which heats the ground, and the temperature obtained from the measurement of the temperature of the ground is LST [14] . LST is becoming more and more important for assessing the conditions of the land surface for various studies.…”