Applying geographic object-based image analysis (GEOBIA) and data mining to identify secondary forests succession on Santarem Region, Para, Brazil

The secondary forest has a low diversity of species, however, they have major importance for the reestablishment of ecosystem functions and nutrient stocks lost from the original forests, as well as higher carbon absorption rate of the mature forests. In this study, we developed an approach using geographic object-based image analysis (GEOBIA) to classify the forest succession stages in a study area with 11,124 km 2 around Santarem (Para State, Brazil). Among the results, we did 19 different classifications covering 1984 to 2016, having identified a variable pattern in the extension of two secondary succession classes (5 to 15 years, >30 years). Another relevant result was the modeling of a decision tree applicable to surface reflectance images collected by the LANDSAT satellites, processing these classifications attributes in a data mining software.