Tea is a manufacturing beverage that is popular around the world. In value chain analysis to increase efficiency, remote sensing technology can be developed to monitor the phenomenon of land use land cover (LULC) change and vegetation health conditions. This study aims to identify LULC in tea plantations, identify the health condition of tea plantations, then analyze spatial trends of changes in tea productivity in Gunungmas Afdeling-1 to changes in tea area or land use conditions. Identification of changes in LULC in tea plantations can be carried out using remote sensing technology and machine learning, in this study using GEE with Sentinel-2A. LULC identification was generated using a supervised classification with the random forest method on the GEE. Tea Productivity trends from 2019 to 2020 decreased and from 2020 to 2021 increased. The results of this study show that the trend of changes in the area of tea plantations classification class is decreasing. Based on the NDVI result, most of the reduced area of tea plantations is in areas with healthy vegetation. The trends in tea productivity changes are not in line with changes in the land cover area of tea plantation classification class and tea vegetation health condition.