Mapping the Cerrado vegetation classes using Rapid Eye imagery

Abstract:

The Cerrado biome is a world’s important biodiversity hotspot. Its natural vegetation is being severely converted by agriculture and pasture, with a large risk to be reduced into small fragments. The Cerrado is the second largest biome of South America and is considered the most floristically diverse savanna in the world. Its vegetation is divided into three major physiognomies: grassland, forest and savanna. The main objective of this study is to classify the major vegetation classes of the Cerrado and its converted areas. A very common issue faced by scientists when using remote sensing to map the Cerrado’s vegetation is finding some confusion between the vegetation and the converted areas. It is believed that satellites with a finer spatial resolution would help improve the classification of this biome. In order to test this hypothesis, we have selected the RapidEye 3A imagery for the year 2012 to use in this work. Decision Trees method were tested in this work to classify the vegetation classes with an effective result. Mapping the Cerrado classes and its converted areas is very important to understand and develop studies to monitor its land use change and prioritize areas for conservation.