Vegetation Optical Depth (VOD) is a satellite-based measure of how much water is stored in plants and trees, which helps monitor drought, forest health, and climate change. Satellite missions like SMAP and SMOS provide global VOD data, but we need accurate ground-based measurements to validate these satellite readings. One method, GNSS-Transmissometry (GNSS-T), uses signals from navigation satellites to estimate how much vegetation blocks the signal. However, GNSS-T data is often limited and affected by complex signal behaviors like scattering and reflections, making it difficult to use directly. To address this, our project combines real-world GNSS-T data with synthetic data generated using a “digital twin” — a detailed 3D model of the forest built from laser (LiDAR) scans. This model simulates how signals travel through the forest with high physical accuracy, capturing effects that real measurements often lack. We are testing this approach in a forest in Georgia, where we collect both GNSS-T and LiDAR data to refine our simulations. Once validated, this system will generate large volumes of realistic synthetic data to train machine learning models. These models will help create more accurate and continuous VOD maps, improving the reliability of satellite-based environmental monitoring.