Gis Integration of Land Cover With Night-Time Lights for Spatiotemporal Evaluation of Urban Expansion

Abstract. Monitoring the urbanisation process requires accurate knowledge of the present and historical spatial extents of built-up area. Understanding the quantification of urban growth is a crucial component of urban and environmental planning. The current research is focussed on evaluating the spatiotemporal change evaluation of built-up area by using a combination Landsat 8 and NPP-VIIRS datasets with GIS integration. Firstly, spatial change analysis of the built-up area was done by employing the Land Use and Land Cover (LULC) classification methodology for the years 2014 and 2022. This was done using the Random Forest (RF) classifier within the Google Earth Engine (GEE) platform. Secondly, the change in built-up area was evaluated with respect to the NTL data. The night-time light (NTL) images detect artificial lights being radiated from cities at night. The NTL data was categorised into three classes and later used to categorise the built-up area into ‘core urban’, ‘peri-urban’ and ‘rural’. The proportion of 'core urban' pixels experienced substantial growth, increasing from 9.8% to 15.5% of the total pixels between 2014 and 2022. This represents a notable increase of 58.75%. Conversely, the 'peri-urban' pixels saw a smaller increase of 4.6%, while the 'rural' pixels witnessed a decline of 8.7%. The average built-up density of the 'core urban' area exhibited an increase from 0.385 in 2014 to 0.398 in 2022. These changes can be attributed to the progressive conversion of 'rural' pixels into 'peri-urban' areas, and subsequently, the transformation of 'peri-urban' areas into 'core urban' regions.