Surface temperature variability analysis of an urban area using Landsat ETM+ thermal images

ABSTRACT The issues relating to global warming and urban heat island have praised the global inclination toward the study of surface temperature (ST) variations. ST is undeniably considered as one of the most significant parameters to assess the impact of temperature variability in any urban environment. The present work tries to extend the work beyond the thermal image processing for variability analysis with a multiscale cell division approach. Subsequently, a radical approach in conjunction with formal statistical techniques is attempted to characterize the spatial variability for the directional attributes. These methods may be helpful in detecting and quantifying the spatial variability at major and minor scales. The spatial variability is investigated to illustrate the spatial distribution of temperature over urban–rural (Rurban) areas contributing to the heat island. The results seemingly revealed that there is an aggregation of the spatial thermal gradient along Rurban areas and vice versa, whereas urban fringe exhibited very complex results of temperature variability for the various land use. The study also validated the approaches for integration of spatial variability techniques with the formal statistical approach as a reliable mechanism to monitor the thermal dynamics in the urban environment.