Mapping Major Land Use And Land Cover Dynamics In Growing Urban Areas Based On Landsat Imagery Using Google Earth Engine

In the present context of rapid change which is occurring either due to the natural or artificial phenomenon, tracking environmental changes with the aid of remote sensing applications has been an important tool for multidisciplinary sectors. One of these applications is the construction of land use and land cover maps through the image classification process. The Google Earth Engine is becoming a crucial tool in remote sensing data analysis as it provides cloud-based platform with huge amount of multiple source satellite data and excellent computation services. The present study aims to know the extent of land cover in Tokha Municipality of Kathmandu of the year 2001, 2009 & 2019 along with the dynamics between the year and the rate of change. In this study, Landsat image of Tokha Municipality of the year 2001, 2009 &2019 were used for LULC supervised classification in Google Earth Engine platform. Five LULC classes were decided and classified using random forest classification, and the output map was obtained with an overall accuracy of 94.8%, 88.40% & 86.95% for the year 2000, 2009 & 2109 respectively. It was found that most of the landcover was dominated by agricultural land and very few were covered by waterbody. From the analysis section, it was clear that major LULC change between 2000 and 2009 was seen in otherland with the total decrease of 354.84 ha. whereas settlement area of 61.5 ha. significantly increased between 2009 and 2019. For the whole study period annual rate of change was found to be in increasing for the categories settlement (1.558%) and forest (1.072%) and decreasing for Agriculture (0.846, waterbody (0.001%) and otherland (1.786%).