Land cover classification and change detection analysis using LandSat series and geospatial datasets in Nepal from 1980 to 2010
Authors:
- 1 July 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 414-418
Abstract
Nepal was ranked as one of the most climate vulnerable country in the world and a better understanding of vegetation patterns and their transformations is critically needed for environmental studies. Studies have shown that monitoring land use and land cover changes with conventional ground survey methods are time consuming and labor intensive. Remote sensing and GIS techniques have been widely employed by researchers due to its cost effective, time efficient and product consistency. This paper illustrates the method to develop land cover maps in Nepal in 1980s, 1990s, 2000s, and 2010s using LandSat series with the support of ancillary spatial datasets. The support vector machine (SVM) supervised classification techniques was used to develop land cover products of each decade. The spatial analysis of time series land cover products over time was employed to examine the changes. The accuracy of different land cover types ranged from 46% to 94% and the overall accuracy of land cover maps in 2010 is around 83% based on the ground truth data collected by the local collaborators. The results show a decrease trend of forest coverage although the rate of deforestation was declining over time. Meanwhile, the results also indicated an expansion of agriculture land in order to sustain the growing population. Similarly, the urban extent have been sprawling over time. The study provide accurate means of measuring the spatial distribution of major land cover types in Nepal over the last four decades that forms valuable resources for devise sustainable land use and environmental research.Keywords
This publication has 4 references indexed in Scilit:
- Croplands in West Africa: A Geographically Explicit Dataset for Use in ModelsEarth Interactions, 2004
- Land use change and biogeochemical controls of nitrogen oxide emissions from soils in eastern AmazoniaGlobal Biogeochemical Cycles, 1999
- Support-vector networksMachine Learning, 1995
- Relating Patterns of Land‐Use Change to Faunal Biodiversity in the Central AmazonConservation Biology, 1994