Overview

GEDI’s gridded biomass and forest structure products have a core resolution of 1 km, based on aggregating footprints within these cells. However, even after four years in orbit, some cells still lack GEDI tracks. Many stakeholders need gap-free maps at finer spatial resolutions. To address this, we combine GEDI’s sparse observations with data from other space-based sensors, particularly Synthetic Aperture Radar (SAR), which offers complete coverage, weather independence, and sensitivity to canopy structure. Since 2019, under NASA’s Carbon Monitoring System (CMS), we have been combining interferometric SAR (InSAR) data from TanDEM-X (TDX) with GEDI data to produce continuous, high-resolution maps of canopy height and biomass. These maps, which include uncertainty estimates, cover extensive areas of the pantropics and allow for the study of biomass changes.

Motivation for GEDI-TanDEM-X Data Fusion

1. Enhanced Spatial Resolution: GEDI’s lidar data, while providing detailed observations of forest structures, is limited to a minimum resolution of 1 km. Fusion with TDX allows for much finer resolution, as small as 1 ha, which is crucial for detailed ecosystem monitoring.
2. Comprehensive Coverage: TDX offers high-resolution interferometric coherence data and can derive topography and canopy structure globally, with complete coverage every three years. This ensures extensive and consistent data for biomass and structural analysis.
3. Improved Data Accuracy: While TDX faces limitations in penetrating vegetation due to its X-band radar and single-polarization at the global scale, combining its data with GEDI’s detailed forest structure information improves the accuracy of height and biomass estimates.
4. Long-Term Monitoring Capability: With TDX operating from 2010 to 2025, this fusion supports long-term studies of forest structure and biomass changes, providing critical data for ecosystem monitoring and climate change studies.

Background on TanDEM-X Mission and Collaboration with DLR

Launched in 2010, the TanDEM-X mission uses a bistatic Synthetic Aperture Radar (SAR) system with two X-band satellites flying in close formation. Its primary goal is to create a highly accurate global digital elevation model (DEM). Operating until 2025, the mission supports a range of SAR applications, notably for monitoring vegetation and forests.

From 2010 to 2015, TanDEM-X surveyed the Earth’s entire land surface twice, achieving its main objective. It then shifted to scientific tasks like forest mapping, snow and ice detection, and monitoring of volcanoes, urban areas, and ocean currents. Between 2017 and 2020, TanDEM-X conducted another global survey to create the Change DEM, documenting surface changes like glacier melting and deforestation. From 2023 to 2025, the mission is conducting a third global survey to update data on forests, permafrost, and ice regions. UMD’s collaboration with German Aerospace Center (DLR) under the CMS project allows UMD team members to access the global archive for generating forest height and biomass maps.

GEDI-TDX Fusion Framework

The GEDI-TDX fusion framework leverages GEDI waveform data to derive mean reflectivity profiles for inverting TDX heights, calibrated with GEDI canopy data using regional models. These height data are then combined with GEDI biomass estimates to establish biomass-height equations, enabling biomass estimation at resolutions as fine as 25 m using Generalized Hierarchical Modeling Based inference (GHMB) for comprehensive uncertainty assessment. Our approach of change involves two methods: 1) using Landsat disturbance records and GTDX biomass to gauge biomass accumulation rates post-disturbance and assess carbon losses, and 2) directly employing GTDX biomass maps across different epochs to quantify biomass changes in areas, accompanied by robust uncertainty.

GEDI-TDX Fusion Data Products

1. Data Processing: Over 30,000 TDX scenes processed, covering 10,000,000 km2.
2. Outputs: Detailed canopy height and biomass maps at various resolutions (25 m, 100 m, 1 km), with uncertainty layers included; Change layers will be included in the next delivery.
3. Coverage: Mexico, Amazon Basin, French Guiana, Gabon, and U.S. forests (Hawaii, Maryland, Washington State, etc.); Data product is expanding to other regions of the pantropics.

GEDI-TDX fusion biomass estimates over the Amazon basin at 25 m resolution.

GEDI-TDX fusion biomass and biomass uncertainty estimates at 25 m.

Comparison of GEDI-TDX fusion biomass products with GEDI L4B in the Amazon Basin.

Areas of the pantropics that have been processed, are in preparation to be processed, and are scheduled to be processed. Map as of November 2024.

Related Publications and Presentations

Armston, J., Fatoyinbo, T., Qi, W., Pardini, M., Papathanassiou, K., Cazcarra-Bes, V., & Dubayah, R. (2020). “Towards pantropical structure and biomass mapping from fusion of GEDI and TanDEM-X data”. AGU Fall Meeting. B108-0033.

Armston, J., Qi, W., Dubayah, R., Stovall, A., Fatoyinbo, T., Saarela, S., Choi, C., Pardini, M., & Papathanassiou, K. (2022). “Fusion of GEDI with satellite interferometric SAR for high resolution monitoring of woody vegetation structure and biomass”. Living Planet Symposium.

Carcarra-Bes, V., Pardini, M., Choi, C., Guliaev, R., & Papathanassiou, K. (2021). “Tandem-X and Gedi Data Fusion for a Continuous Forest Height Mapping at Large Scales”. 2021 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 796-799. doi: 10.1109/IGARSS47720.2021.9554655.

Choi, C., Pardini, M., Roman, G., & Papathanassiou, K. (2022). “Fusion of TanDEM-X and GEDI data for mapping forest height in the Brazilian Amazon”. 2022 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

Choi, C., Carcarra-Bes, V., Guliaev, R., Pardini, M., & Papathanassiou, K. (2023). “Large-Scale Forest Height Mapping by Combining TanDEM-X and GEDI Data”. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 16, pp. 2374-2385. doi: 10.1109/JSTARS.2023.3244866.

Dubayah, R.O., Qi, W., Armston, J., Fatoyinbo, T., Papathanassiou, K., Pardini, M., Stovall, A., Choi, C., & Cazcarra-Bes, V. (2023). “Pantropical Forest Height and Biomass from GEDI and TanDEM-X Data Fusion”. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/2298.

Guliaev, R., Cazcarra-Bes, V., Pardini, M., & Papathanassiou, K. (2021). “Forest Height Estimation by Means of TanDEM-X InSAR and Waveform Lidar Data”. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 3084-3094. doi: 10.1109/JSTARS.2021.3058837.

Qi, W., Lee, S.-K., Hancock, S., Luthcke, S., Tang, H., Armston, J., & Dubayah, R. (2019). “Improved forest height estimation by fusion of simulated GEDI Lidar data and TanDEM-X InSAR data”. Remote Sensing of Environment, 221, 621–634. https://doi.org/10.1016/j.rse.2018.11.035.

Qi, W., Saarela, S., Armston, J., Ståhl, G., & Dubayah, R. (2019). “Forest biomass estimation over three distinct forest types using TanDEM-X InSAR data and simulated GEDI lidar data”. Remote Sensing of Environment, 232, 111283. https://doi.org/10.1016/j.rse.2019.111283.

Qi, W., Armston, J., Cazcarra-Bes, V., Papathanassiou, K., Pardini, M., Fatoyinbo, L., & Dubayah, R. (2020). “Fusing GEDI waveforms with spaceborne SAR data for improved forest structure estimation”. AGU Fall Meeting. B029-08.

Qi, W., Armston, J., Dubayah, R., Fatoyinbo, T., Stovall, A., Papathanassiou, K., Pardini, M., & Cazcarra-Bes, V. (2021). “High-resolution mapping of pantropical canopy height by integration of GEDI and TanDEM-X Data”. AGU Fall Meeting. B25G-1557.

Qi, W., Armston, J., Dubayah, R., Stovall, A., Fatoyinbo, T., Choi, C., Pardini, M., & Papathanassiou, K. (2022). “Pantropical Forest Structure and Biomass Mapping by Integrating GEDI Lidar and TanDEM-X InSAR Data”. AGU Fall Meeting.

Qi, W., Armston, J., Choi, C., Stovall, A., Saarela, S., Pardini, M., & Fatoyinbo, L. (2024). “Mapping large-scale pantropical forest canopy height by integrating GEDI lidar and TanDEM-X InSAR data”. Remote Sensing of Environment (Under 3rd review). Research square preprint doi: https://doi.org/10.21203/rs.3.rs-3306982/v1.