ASSESSMENT OF URBAN HEAT ISLAND (UHI) USING GEOMATICS TOOLS

Abstract

URBAN HEAT ISLAND (UHI) refers to the observed temperature difference between urban environment and the surrounding rural regions. Local-climate changes due to urbanization are epitomized by the urban heat island (UHI). In this study, main emphasis is given on the methodology which is used to find Land Surface Temperature (LST). Mono window algorithm is used for LST computations which was proposed by Qin et al. in 2001.This algorithm is specifically designed to convert Landsat TM/ETM+/OLI digital numbers to Land Surface Temperature(LST). The mono window algorithm for LST requires only three parameters: transmittance, emissivity, and effective mean atmospheric temperature, to retrieve land surface temperature from Landsat TM/ETM+/OLI data. Spectral range used for thermal band is 10.4 to 12.5 μm which records the radiation of earth surface. There are various parameters which are considered while computing LST. The parameters like Spectral Radiance and At Satellite Temperature plays an important role in LST computations. In this study, Dehradun city is selected for analysis of urban heat island (UHI) effect. The analysis is carried out using Landsat TM, ETM+ and OLI data from 1998 to 2015 over Dehradun city region. Relation between normalized difference vegetation index (NDVI) and LST is studied and it shows that there exists negative correlation between NDVI and LST. Moreover, there exists a negative correlation between emissivity and LST which implies NDVI and emissivity are positively correlated. Land use classification is also studied in order to showcase the % change in each land cover class from 1998 to 2015 and how LST is increasing in various land cover classes over the time frame. It was observed that barren land and range land were significantly reduced during the time period 1998 – 2015 whereas crop land and urban areas were significantly increased since 1998 to 2015. LST varies mainly in the three temperature slabs i.e. (35-400C), (40 - 450C) and (>450C). Over the time span, the areas with the temperature 40 - 450C were increased most significantly whereas the areas in the core of the city i.e. the areas having temperature > 450C were also increased. After assessment in ArcMap 10.2.2, it is observed that the dense forest areas above the Dehra Dun valley i.e. Sahastradhara Forest area is not affected by urban heat island. However the areas near the Tons river are affected to some extent from 1998 to 2015. The reason for this is not primarily urbanization but various activities which lead to deforestation and thus increase of barren land exposing land to heat island and hence increase in surface temperature.