Factors Affecting Deforestation in Paunglaung Watershed, Myanmar using Remote Sensing and GIS

Abstract

The objective of this study is to predict the factors affecting deforestation based on the relationship between detected forest cover changes and biophysical factors of the study area by using spatial and statistical analysis. The study area is Paunglaung watershed, one of the important watersheds in Myanmar. This study used maximum likelihood classification (MLC) to compare Landsat TM images of 1989 and Landsat ETM + images of 2000 in order to assess deforestation. The study revealed that 32.61km^2 or 0.7% of the forest cover was annually lost in the watershed area. An empirical spatial model of deforestation was developed by relating a set of variables measuring landscape attributes to the spatial occurrence of deforestation. Variables derived from the GIS based analysis were used to verify spatially the close statistical relationship between dependent and independent variables by logistic regression model. This study provides both statistical and spatial confirmation of the importance of access in the location of deforestation.