This paper makes an attempt to map the changes in forest cover from 1999 to 2010 in Chamoli District of India using remote sensing techniques. Landsat ETM+ imageries of year 1999, 2006 and 2010, having a spatial resolution of 30m have been used to map the forest cover. The data was processed to obtain the forest cover for all images and then classified and analysed to derive forest cover change. The method used for obtaining forest cover was by deriving forest canopy density using NDVI. These results were further processed to obtain forest cover change during 1999-2006 and 2006-2010. The forest cover was then classified into 5 distinct forest density classes using the NDVI images. Change in forest cover was determined individually for each class. Class 5 represents very dense vegetation with canopy density more than 70%. The change observed in these class values had utmost importance as they indicate the loss or gain of dense forests in Chamoli which are endangered the most to degradation. After comparing and analysing the results of the change detection process, it was observed that during the period of 1999-2006, very dense vegetation exhibited a net increase of 111.67 sq. km and a net decrease of 98.19 sq. km in very dense vegetation was seen in the period of 2006-2010. Accuracy assessment was carried out using forest cover maps given in State Forest Reports published by FSI and results for 1999 and 2006 classified images showed and overall accuracy of 86% and 76% respectively.