Description: The economic and population growth in India accompanied by increasingly frequent and severe disasters, such as floods and forest fires, raises concerns about the future. In recent decades, satellite technology has become an important tool for monitoring the Earth's surface, including natural hazards and land surface changes. Our USAID (United States Agency for International Development) project, REmote Sensing for Forest Renewal, Ecosystem Services, and Sustainable Hydrological Management (REFRESH), aims to improve forest sustainability in India with the help of satellite geodetic and remote sensing data. Here we report the progress for the physical science component of the REFRESH Project. Our team has developed a deep-learning model for classifying 10 types of Land Use/Land Cover (LULC) to detect long-term changes in forest areas over India. With collaboration from Indian partners, we validate our classification results using their in-situ land cover data. We used PlanetScope daily images at 3–5 m spatial resolution to monitor the evolutions of floods and forest fires over India. Other satellite data products include in-land water level virtual stations over rivers, dams, and lakes using three decades of satellite radar altimetry, daily-sampled GRACE/GRACE-FO Level 1B gravimetry data to monitor monsoonal floods and seasonal droughts, spaceborne lidar and GEDI for mapping canopy changes, and deep-learning downscaled satellite gravimetry measured total water storage anomalies over entire India. Finally, we plan to generate spaceborne GNSS-Reflectometry and NASA’s CYGNSS for water/land classification data products over India. The data products can be retrieved, visualized, and analyzed on an elaborate observation portal.