Pasture Condidition map (10m) of the Brazilian Cerrado

Pasture Condition map of 2020 for the Brazilian Cerrado made by Remote Sensing and GIS Lab (LAPIG/UFG). The production of this map used the cloud computing power of the Google Earth Engine cloud platform coupled with data from ESA's Sentinel 2 A/B satellite and the Random Forest algorithm to produce a pasture map and a pasture conditions map. The training samples used were based on ~60,000 samples with spectral behavior correlated with Embrapa field data using the Time-Weighted Dynamic Time Warping (TWDTW) algorithm developed by the Instituto Nacional de Pesquisas Espaciais (INPE).

The classes presented in the map are related to:

1 - Signs of Severe Degradation process

2 - Signs of Moderate Degradation process

3 - Signs of Absent Degradation