LAUSR

Abstract A harmful effect of anthropogenic activities in urban environments is the increases of thermal discomfort and subsequently, a negative effect on humans’ mental and physical performance. Therefore, it is of high importance to detect, monitor, and predict thermal discomfort, especially its temporal and spatial patterns in cities. The objective of this study is to propose a new method for modeling outdoor thermal comfort based on remote sensing and climatic datasets. To do so, several datasets were utilized, including those from Landsat, Moderate Resolution Imaging Spectroradiometer (MODIS), Digital Elevation Model (DEM) from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), and climatic datasets from local meteorological stations. The method was experimented in the city of Tehran, Iran. For modeling outdoor thermal comfort, the Least Squares Adjustment (LSA) model was presented based on the Principle Component Analysis (PCA). In this model, the Principle Components (PCs) of the environmental and surface biophysical parameters were considered as independent variables and Discomfort Index (DI) as dependent variable. Finally, by determining the optimal values of the adjustment coefficients for each independent variable, maps of outdoor thermal comfort at different timestamps were produced and analyzed. The results of the modeling showed that correlation coefficient and Root Mean Square Error (RMSE) between the modeled and observed outdoor thermal comfort values at the meteorological stations for the training data sets were 0.86 and 1.80, for the testing data set were 0.89 and 2.04, respectively, while it was 0.85 and 1.15 for the self-deployed devices. The average values of DI in warm season of year was 8.5 °C higher than the cold season of the year. Further, in both warm and cold seasons of year the mean value of DI for bare land was found higher than other land covers, whereas that of water bodies lower than others. Our findings suggest that efficiency can be achieved for modeling outdoor thermal comfort using LSA with remote sensing and climatic datasets.