The land-use/land-cover pattern of a region is an outcome of natural and anthropogenic process. Land-use /land-cover change has become a central component of current strategies in managing natural resources and monitoring environmental changes. The present study was carried out with an integrated approach using Remote Sensing and GIS techniques together with socio-economic data for land cover change detection. Landsat TM of 2002 and ETM+ of 2010 imagery were used to evaluate forest cover dynamics during 2002-2010 in the Laljhadi Forest (Corridor), Kanchanpur district of Nepal. The aims of the study were to quantify and map the spatio - temporal pattern of forest cover change in term of land use changes between 1996- 2010 and to explore the proximate cause behind those processes. Supervised classification was used to prepare land use maps using the maximum likelihood algorithm. Image classification was carried out by emphasizing six main categories. Ground verification was done in November, 2010 along with questionnaire survey.
The study revealed that there is a net decrease in forest cover from 1996 to 2010. In 1996 forest cover age area was 63.73%, where in 2002 it was 47.73% and in 2010 it was 35.9 %. This shows that there is a rapid decrease in forest cover from 1996 to 2002 with the rate of changes 4.90% and 3.57% from 2002 to 2010. This compensates to increase in bush area where in 1996 its coverage was 1.37% of the total study area, which increased to 25% and in 2010 to 29%. The overall accuracy was 78% for both year maps.
The underlying causes of forest cover change are multifarious including encroachment by the flood victims (93%) (Dodha River), open grazing of cattle including the critical area of forest (95%) and people’s desire to expand forest land into cultivation land within the forest area and moreover rapidly increasing population with the growth rate of 3.9%, which is higher than of Nation. The broad pattern of major land use /land cover changes are known with some confidence, and the literature is rich in contending explanation for them. To advance, we need much more precise and spatially congruent datas, so that there could be harmony between resources and the local users.