Change Detection of Land Surface Temperature (LST) and some Related Parameters Using Landsat Image: a Case Study of the Ebinur Lake Watershed, Xinjiang, China

Abstract

This study assesses and detects land use/cover (LUC) and land surface temperature (LST) change using multi-temporal Landsat TM satellite data. NDVI, albedo and MNDWI were used to analyze the LST qualitatively. The results revealed that the accuracy of LST measurements in watershed is within 1.5 °C. Then, temperature changes between 1998 and 2011 were analyzed. The classifications of land surface temperatures lie in five categories as follows: lower (1.9–8.9 °C), low (8.9–15.9 °C), middle (15.9–22.9 °C), high (22.9–29.9 °C), higher (29.9–36.9 °C), and highest (36.9–43.9 °C). Second, east-west profiles of the characteristics of the distribution of LUC types were made based on 1998 and 2011 images. By comparing LSTs in these two years, one can conclude woodland-grassland has a very strong influence on temperature. Third, LST increased with the increases in the density of salinized and desert lands, but decreased with the increase in vegetation cover. The relationship between MNDWI and LST was significantly negatively correlated. Multiple regression analyses between LST and each index as well as elevation were created to evaluate the watershed thermal environment. This regression showed that NDVI, albedo, MNDWI and a digital elevation model were effective indicators for quantifying the effects of land use/cover change (LUCC) on LST, and the correlation coefficient R was 0.806. Finally, natural and human factors were important factors affecting temperature change. Generally, the temperature of the oasis was lower than the surroundings, which results in a 'cold island effect'.

from # All Medicine by Alexandros G. Sfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/2idgrf1
via IFTTT