An investigation on seasonal variability between LST and NDWI in an urban environment using Landsat satellite data

Subhanil Guha,Himanshu Govil,Monika Besoya

DOI: https://doi.org/10.1080/19475705.2020.1789762

2020-01-01

Geomatics, Natural Hazards and Risk

Abstract:<span>The urban landscape is considered the most complex and heterogeneous landscape among the different land surface features. It rises the land surface temperature (LST) to a large extent compared to the surrounding rural body. This investigation deals with the seasonal variability between LST and normalized difference water index (NDWI) on the different land surfaces in Raipur, India by using sixty-four Landsat images from 1991–92 to 2018–19. The results show that the post-monsoon season indicates the best correlation (0.42) between LST and NDWI, followed by the monsoon (0.34), pre-monsoon (0.25) and winter (0.04). The water bodies reflect a moderate negative correlation of LST-NDWI in all the four seasons (−0.49 in pre-monsoon, −0.33 in monsoon, −0.31 in post-monsoon and −0.45 in winter). On green vegetation, this LST-NDWI correlation is strongly positive in pre-monsoon (0.67) season, moderate positive in monsoon (0.43) and post-monsoon (0.50) seasons, and weak negative in winter (0.25) season. The built-up area and bare lands build a weak positive correlation of LST-NDWI in all the four seasons (0.24 in pre-monsoon, 0.21 in monsoon, 0.27 in post-monsoon and 0.15 in winter). This study can be beneficial for land use planning and management of any city under a similar physical environment.</span>