DRIVING FACTOR ANALYSIS OF FOREST COVER CHANGE USING RANDOM FOREST METHOD (CASE STUDY: WEST SUMATERA)
Forest development of human needs. Most of the villages in West Sumatera province are recorded to be located in and around forests. West Sumatra's forest cover continues to decrease due to land clearing for the benefit of surrounding communities, land clearing for mining, roads, plantations...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/76406 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Forest development
of human needs. Most of the villages in West Sumatera province are recorded to be
located in and around forests. West Sumatra's forest cover continues to decrease
due to land clearing for the benefit of surrounding communities, land clearing for
mining, roads, plantations, agriculture and occurrences of fires. In this case, there
has not been a specific study to conduct an in-depth analysis related to the driving
factors of forest cover change in Sumatera Barat.
This research study uses the random forest method with the aim of predicting
changes in forest cover and indentifying the most contributing driving factors. Then
an analysis was carried out related to this driving factor at the research location.
The accuracy of predicting changes in forest cover was the highest at 0,99, that is
primary mangrove forest and secondary swampforest. The lowest accuracy at 0,93
for predicting changes from combining data on all forest cover types in West
Sumatera. The results of permutation importance calculations indicate that the
most influencing driving factors are rainfall, distance to the city and distance to the
road. The varying rainfalls in West Sumatera has different impacts on secondary
and primary dry forest cover such as changes to other type of land cover. One of
the effects of low rainfall turning forest cover into dry land agriculture. Forests
cover that influenced by the distance to the city are mangrove forests and swamp
forests that transform into settlements and dry agriculture land, evidenced by the
predicted forest cover changes location near to settlements and economic activity.
Forest cover that influenced by the distance to the road is plantation forest, as
evidenced by the location of the predicted forest cover changes to agricultures and
plantations having adequate road access. This facilitates the transportation of
agricultural products from fields to markets or processing factories. The results of
the analysis are expected to be used as a guideline in taking preventive actions in
accordance with policies to preserve forests cover and human welfare.
|
|---|