Monitoring Land Cover Change in Kathmandu City Using Spatial Metrics And Remote Sensing Techniques

Abstract

This paper explores the urban land cover changes in Kathmandu city area in the past one and half decades. Multi-temporal satellite images and spatial metrics were used to quantify the land cover changes. The study area was setup to 10x10 km covering the Kathmandu metropolitan area. After pre-processing the satellite images, supervised classification with maximum likelihood classifier was applied to create thematic urban land cover maps. Random sampling method was used to create geographic points for each thematic map to assess the accuracy of the maps. Five land use classes: urban builtup area, cultivated land, orchard, water and natural vegetation were identified. For detecting land cover changes quantitatively, a land use land cover conversion matrix was computed between the year 1989- 1999 and 1999-2005. Spatial metrics (patch density, largest patch index, edge density, Euclidian nearest neighbor mean, contagion index and area weighted mean patch fractal dimension index) were computed to evaluate landscape structure of the metropolitan area.0The overall result shows rapid expansion of urban built environment and shrinkage of the agriculture land. The agriculture and orchard lands were mostly transformed into urban uses. The existence of water and vegetation were found very low as compared to other land covers. The urban expansion trend was confined in the peri-urban of the urbanized territory. The overall urban landscape seems to be very complex and fragmented in later years.