Home Articles Forest Change Detection

Forest Change Detection

S. V. Panikkar
Manager, Business Development, Mapping/GIS Group
Rolta India Ltd.

The landuse / land cover data of Dehradun area for a period of 60 years were analysed to study the changes in the forest cover.

GIS Technology is being increasingly used for a host of applications. One of the key areas of application is monitoring the environmental degradation brought out by an increase in the anthropogenic activities resulting in the removal of the forest cover. This article is based on a study of the effect of human interference on the environment, particularly depletion of the forest cover. The landuse / land cover data for a period of 60 years were analysed to study the changes in the forest cover, and then correlated with the occurrence of landslides in the area around Dehradun and Mussoorie in Uttar Pradesh, India.

In today’s world, man has proved to be an important agent bringing about changes in the natural processes through intense ‘development’ activities. The effect of the rapid urbanization of the Himalayan terrain in the area around Dehradun and Mussoorie in Uttar Pradesh has been studied. These townships are frequented by thousands of tourists every year, which has resulted in increased anthropogenic activities in terms of road and building constructions. The intense quarrying over a period of time for the limestone has also been added to the problem of environmental degradation. The result of all these has been that the once-dense forest cover is now depleted drastically. The problem has been further compounded through improper landuse practices in these depleted forests.

Forest Change Detection
The changes in the landuse / land cover in the area over a period of 60 years were analysed through data collected from the topographical maps and remote sensing media (IRS Data). The topographical maps based on the surveys carried out in 1930 (1:63,360) and 1960 (1:50,000) were used to prepare the landuse maps for these two periods. Supervised classification was undertaken to delineate the forest cover of the present day.


In 1930, the forest cover accounted for about 45% of the total study area of 445 sq. km. Within a period of three decades, by 1960, it was reduced to 150 sq. km (34% of the area). Between 1960 and 1990, there was a drastic reduction in the forest cover to about 82 sq. km (18.7 % of the study area). The current landuse practices in these depleted forests were determined in order to ascertain the causes for deforestation. About 22% of the forest cover was removed for cultivation. The maximum forest cover was however removed for the economic value of sal, banj, pine and fir, which forms the dominant vegetation in the area. This is indicated by the fact that 36% and 38% of the depleted forests are presently barren and with sparse vegetation respectively.

The occurrences of landslides were mapped using aerial photographs and remote sensing data. A cross tabulation of the depleted forest map with the landslide occurrence map showed that 60.27% of the landslides are in non-forested areas that were forested in 1930. The forest cover that has remained unchanged since 1930 i.e. for 60 years, was seen to account only for 2.9% of the landslides. This goes to show the relevance of preserving the forest cover goes to show the relevance of preserving the forest cover.


The study has demonstrated the utility of a GIS / Image Processing System to monitor changes in the forest cover, and utilise the data effectively to identify the causative factors. An ideal GIS system for such an analysis should have built-in support for raster and vector analysis. It should also have the capability to integrate the GIS data from the various commercially available packages without having to go through a process of translation. This is particularly pertinent in our country, where data are scarce and exist as “islands” of information.