Detection and Prediction of Peatland Cover Changes Using Support Vector Machine and Markov Chain Model

Detection and prediction of peatland cover changes should be conducted due to high rate deforestation in Indonesia. In this work we applied Support Vector Machine (SVM) and Markov Chain Model on multitemporal satellite data to generate the correspondings detection and prediction. The study area is located in the Rokan Hilir district, Riau Province. SVM classification technique used to extract information from satellite data for the years 2000, 2004, 2006, 2009 and 2013. The Markov Chain Model was used to predict future peatland cover. The SVM classification result showed that the mean Kappa coefficient of peatland cover classification is 0.97. Between years 2000 and 2013, the wide of non vegetation areas and sparse vegetation areas have increased up to 307% and 22%, respectively. While the wide of dense vegetation areas have decreased up to 61%. We found that a 3 years interval used in the Markov Chain Model leads to more accurate results for predicting peatland cover changes.