A remarkable phenomenon in urban climate is the urban heat island (UHI) effect. Based on satellite imagery and geographical and climatological data, this study analyzes the patterns of land surface temperature (LST) and land use/land cover (LULC) in Hong Kong and the seasonal variations of the relationships between them. Remote sensing, Geographical Information System (GIS) and statistical methods are used for analysis. The results indicate that LST is significantly affected by LULC types. The ordinary least squared (OLS) models indicate that LST is positively related to Normalized Difference Built-up Index (NDBI) and Normalized Difference Bareness Index (NDBaI) but negatively related to Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), elevation and wind speed. The Geographically Weighted Regression (GWR) model indicates that LST is negatively related to NDVI and elevation. The magnitudes of the influences of indices vary with the season. The GWR model demonstrates better fitness compared to the OLS models. Regression results on the GIS maps reveal the fitness of the models across the study area and the spatial variations of the regression coefficients. Such information may help decision makers develop policies at the regional scale and fine-tune planning practices at the local scale to improve the urban thermal environment.