Abstract:
The importance of forests is hard to underestimate. Seeing as forests are of so
much value, several studies have been done to assess forest cover, both on small and
large scale.
Natural and human-caused characters that can be seen on the surface of Earth is
Land cover. Land use, on the other hand, points out events that take place there and its
present use. Land Use and Land Cover caused by human actions such as deforestation,
are one of the crucial elements of global forest cover change. Remote sensing is a tool
that is much essential for producing maps depicting Land use and Land cover by means
of image classification.
Supervised Image Classification was applied while focusing on forest area in
the AOI, over the past three decades (1990-2020). This was done for classifying and
mapping changes in land use and cover. This research also includes carbon stock
assessment over the study period.
The major LULC classified were Agricultural land, Water bodies, Forest area,
Settlement and Others. For an effective Image Classification, various aspects had to be
considered.
Although forest area reduced from 1990 to 2009, a sharp increase was seen in
the next decade (between 2009 and 2020) by 56%. This can be linked to new plantations
under BTTAP in the region. Almost 836 km2 of land was found to be covered with
forests. The net change was recorded as a 32% increase in forest land in the study area
over three decades. Same trend was seen in Carbon stocks. While carbon stocks
decreased from 1990 to 2009, they increased by 408 tons per km2 in the last decade.
Overall, classification accuracy of the study fell between 76.8% and 83.5% and
hence the data and classified images can be used for additional research. The study
offers important pieces of information which environment managers and decisionvii
makers can utilize to encourage plantation of trees and save existing forests in the
country to combat climate change.
