CR 87:99-116 (2022)  -  DOI: https://doi.org/10.3354/cr01686

ABSTRACT: Land cover and topography influence land surface temperature (LST). The purpose of this study was to determine whether there is a correlation between Moderate Resolution Imaging Spectroradiometer (MODIS) LST and air temperature (AT) under various physiographic conditions in Iran. A novel method was developed to determine the match-up time between the LST of the Terra/Aqua overpass and the AT observation at an appropriate distance. Numerous statistical metrics were used, including the correlation coefficient (R), root mean square differences (RMSD), and relative root mean square differences (RRMSD). The findings indicate that LST and AT have a strong correlation coefficient (0.7-0.9) in Iran. Intra-annual changes in R, on the other hand, show higher values during the cold months than during the warmer months. The differences between LST and AT were lowest at the coasts and highest in the central and northwestern deserts of Iran according to the spatial distribution of the RRMSD. The RRMSD for MODIS Terra night LST (LSTTN) was lower than for MODIS Aqua day LST (LSTAD). Seasonal differences between LST and AT were observed, particularly at higher elevations and slopes. The most important factors that result in a higher correlation and a smaller difference between LST and AT are surface moisture, vegetation density, and land cover dynamics (LCD). The LST/AT correlation and difference can also be affected by the daily and intra-annual temperature cycle. The findings can be applied to further studies, e.g. when estimating AT from LST in deserts and at high elevations.

KEY WORDS: LST · Air temperature · MODIS · Iran · Statistical comparison