Members Can Post Anonymously On This Site
NASA Attorney Enjoys Serving Others and Contributing to the Team Mission
-
Similar Topics
-
By NASA
Explore This Section Earth Earth Observer Editor’s Corner Feature Articles Meeting Summaries News Science in the News Calendars In Memoriam Announcements More Archives Conference Schedules Style Guide 31 min read
Summary of the 2025 GEDI Science Team Meeting
Introduction
The 2025 Global Ecosystem Dynamics Investigation (GEDI) Science Team Meeting (STM) took place April 1–3, 2025 at the University of Maryland, College Park (UMD). Upwards of 60 participants attended in-person, while several others joined virtually by Zoom. The GEDI Mission and Competed Science Team members were in attendance along with the GEDI NASA program manager and various postdoctoral associates, graduate students, collaborators, and data users – see Photo. Participants shared updates on the GEDI instrument and data products post-hibernation with the GEDI community. They also shared progress reports on the second Competed Science Team cohort’s projects as well as applications of GEDI data.
This article provides a mission status update and summaries of the presentations given at the STM. Readers who would like to learn more about certain topics can submit specific questions through the GEDI website’s contact form.
Photo. Attendees, both in person and virtual, at the 2025 GEDI Science Team meeting. Photo credit: Talia Schwelling Mission Status Update: GEDI Up and Running After its Time in Hibernation
When the 2023 GEDI STM summary was published in June 2024 – see archived article, “Summary of the 2023 GEDI Science Team Meeting” [The Earth Observer, June 18, 2024] – GEDI had been placed in a temporary state of hibernation and moved from the International Space Station’s (ISS) Japanese Experiment Module–Exposed Facility (JEM–EF) Exposed Facility Unit (EFU)-6 to EFU-7 (storage).
Two years later, as the 2025 GEDI STM took place, the GEDI instrument was back in its original location on EFU-6 collecting high-resolution observations of Earth’s three-dimensional (3D) structure from space.
DAY ONE
GEDI Mission and Data Product Status I
Ralph Dubayah [UMD—GEDI Principal Investigator (PI)] opened the STM with updates on mission status (see previous section) and the development of current and pending GEDI data products.
Following its hibernation on the ISS from March 2023–April 2024, the GEDI mission entered its second extension period. Since re-installation, the instrument’s lasers have been operating nominally, steadily collecting data, increasing coverage, and filling gaps. As of November 27, 2024, GEDI had collected 33 billion Level-2A (L2A) land surface returns, with approximately 12.1 billion passing quality filters. Since the last STM, an additional 1422 new simulated GEDI footprints have been added to GEDI’s forest structure and biomass database (FSBD), which is a database of forest inventory and airborne laser scanning data (ALS) from around the globe that is used for cal/val of GEDI data. The FSBD now has 27,876 simulated footprints in total – see Figure 1. This data will support improved L4A biomass algorithm calibration.
Figure 1. Training samples, or simulated footprints, are derived from coincident forest inventory and ALS data. DBT = deciduous broadleaf, EBT = evergreen broadleaf, ENT = evergreen needleleaf, GSW = grass, shrub, woodland. Figure credit: David Minor Version 2.1 (V2.1) of GEDI L1B, L2A, L2B, and L4A data products are the latest product releases available for download. This version incorporates post-storage data through November 2024. In January 2025, the team also released the new L4C footprint-level Waveform Structural Complexity Index (WSCI) product using pre-storage data. The upcoming V3.0 release will incorporate pre- and post-storage data that will improve quality filtering, geolocation accuracy, and algorithm performance.
Although GEDI met its L1 mission science requirements before entering hibernation, orbital resonance on the ISS impacted GEDI’s coverage in the tropics. To help address these gaps, the team is exploring data fusion opportunities with other missions – e.g., NASA-Indian Space Research Organisation Synthetic Aperture Radar (NISAR), the Deutsches Zentrum für Luft- und Raumfahrt’s (DLR – German Aerospace Center) Terra Synthetic Aperture Radar–X (TerraSAR-X) and TerraSAR add-on for Digital Elevation Measurement (TanDEM-X) missions, and the European Space Agency’s upcoming forest mission – Biomass. [UPDATE: Biomass launched successfully on April 29, 2025 from Europe’s Spaceport in Korou, French Guiana, and NISAR launched July 30, 2025 from the Satish Dhawan Space Centre located on Sriharikota Island in India.] Additional ongoing mission team efforts include advancing waveform processing, developing gridded products tailored to end-user needs, understanding error and bias, and continuing expansion of the FSBD.
Dubayah concluded by highlighting the steady rise in GEDI-related publications and datasets appearing in high-impact journals, including PNAS, Nature, and Science families. Visit the GEDI website to gain access to a comprehensive list of GEDI-related publications.
After hearing general updates from the mission PI, attendees heard more in-depth reports on science data planning, mission operations, and instrument status.
Scott Luthcke [NASA’s Goddard Space Flight Center (GSFC)—GEDI Co-Investigator (Co-I)] reported on Science Operations Center activities, including geolocation performance and improvements. He shared that the Science Planning System, which is used to plan GEDI data acquisition locations, has been upgraded to improve targeting capabilities using high-resolution Reference Ground Tracks. The Science Data Processing System also underwent a technical refresh that increased computational and storage capability and has completed processing and delivery of all V2.1 data products, including post-storage data (April–November 2024), to the Land Processes and Oak Ridge National Laboratory (ORNL) Distributed Active Archive Centers (DAACs).
Luthcke explained that V2.1 improves on precision orbit determination, precision attitude determination, tracking point modeling, time tags, and oscillator calibration. Looking ahead, V3.0 will enhance range bias calibration, improved pointing bias calibration, and modifications to L1A, L1B, L2A, and L2B products. Luthcke also discussed updates to the L3 data product, which include corrected timing and range bias, improved positioning and elevation, and a wall-to-wall 1-km (0.62-mi) elevation map to be released alongside V3.0.
Tony Scaffardi [GSFC—GEDI Mission Director] provided an update on the Science and Mission Operations Center since its post-hibernation return to science operations on June 3, 2024. He addressed various on-orbit events that may have briefly disrupted data collection and reviewed upcoming ISS altitude plans. As of March 2025, each of the instrument’s three lasers logged over 22,000 hours in firing mode, collecting more than 20 billion shots each, with 72% of that time directly over land surfaces. As of April 2025, 95,346 hours of science data have been downlinked, averaging 51.21 GB of data per day.
Bryan Blair [GSFC—Deputy PI and Instrument Scientist] concluded this section of the meeting with a discussion of GEDI instrument status, reporting that all three lasers are operating nominally and that both the detectors and digitizers continue to perform well. He noted that the laser pulse shapes have remained stable since the mission began, indicating consistent system performance over time. Blair also addressed the inherent challenges of operating in space, such as radiation exposure, and emphasized the importance of designing systems for graceful degradation. A recent firmware update was successfully applied to all three digitizers, and no life-limiting concerns have been identified to date.
Competed Science Team Presentations – Session I
Jim Kellner [Brown University—GEDI Co-I] kicked off the Competed Science Team (CST) presentations with an overview of his work investigating the role of stratification and quality filtering to improve GEDI data products and the impact of stratification error on prediction. He explained how GEDI quality filtering and aboveground biomass density (AGBD) model selection and prediction rely heavily on stratification by plant functional type (PFT) and geographic world region. Thus, evaluation and improvement of stratification and quality filtering will help maximize the number of usable GEDI shots, some of which are potentially excluded unnecessarily. To support these improvements, Kellner is exploring replacement of the current 1-km (0.62-mi) stratification layer with a 30-m (98-ft) product derived from Landsat and similarly upgrading the 500-m (1640-ft) phenology stratification layer to a 30-m (98-ft) Landsat version. These changes aim to improve the L4A footprint-level AGBD estimates in particular, but flow through to the GEDI L4B data product.
Birgit Peterson [United States Geological Survey (USGS), Earth Resources Observation and Science (EROS) Center] presented her research on the decomposition of GEDI waveforms to derive vegetation structure information for 3D fuels and wildfire modeling, emphasizing the importance of consistent and comprehensive information on vegetation status for effective wildland fire management. Canopy structure data, like that provided by GEDI, can play a key role in developing physics-based fire behavior models, such as QUIC-Fire. With study sites in South Dakota, the Sierra Nevada, and dispersed around the southeastern United States, Peterson’s work aims to demonstrate how vegetation structure parameters needed to run the QUIC-Fire model can be derived from GEDI waveform data.
David Roy [Michigan State University] shared updates on his CST project leveraging GEDI data to improve understanding of species-specific tropical forest regrowth in central Africa. Focusing on the Mai Ndombe region of the Democratic Republic of the Congo (DRC), the project aims to quantify forest regrowth by integrating GEDI-derived structural data with satellite and airborne laser scanner (ALS) based maps of forest height. Roy emphasized the potential of secondary and recovering forest conservation as a low-cost mechanism for carbon sequestration and climate change mitigation. GEDI data combined with satellite maps provides new opportunities to quantify forest regrowth and carbon sequestration in secondary forests at finer detail, although high species diversity and varying regrowth rates can be complex to assess with remote sensing. Roy also presented a 2025 paper validating the GEDI relative height product in the DRC and at two US temperate forest sites with a simple method to improve the GEDI canopy height using digital terrain heights measured by airborne laser scanning (ALS).
Perspectives I
After the morning CST presentation session, meeting attendees heard the first perspectives presentation from Amanda Whitehurst [NASA Headquarters (HQ] GEDI Program Scientist and NASA Terrestrial Ecology Program Manager]. Whitehurst is new to the GEDI Program Scientist role; she used this opportunity to officially introduce herself to the ST and expressed her enthusiasm for the work ahead She commended the GEDI team on the impressive accomplishments of the mission to date, and spoke about the exciting potential for continued data collection and scientific discovery through the program.
Matteo Pardini [DLR] shared his perspective on the potential of combining synthetic aperture radar (SAR) with lidar data to improve four-dimensional (4D) forest structure mapping. He highlighted DLR’s TerraSAR-X and TanDEM-X missions, which have been acquiring interferometric data since 2007 and 2010, respectively. Both missions are expected to continue acquiring data through 2028. The TanDEM-X Global Digital Elevation Model, covering 150 million km2 (58 million mi2) with approximately 1-m (3-ft) accuracy, can be used to derive forest height and biomass. The fusion of TanDEM-X and GEDI data can improve biomass estimates – see Figure 2 – and help researchers parameterize the relationship between coherence and forest structure. Pardini also previewed the upcoming BIOMASS mission, which will operate at a lower frequency and be able to penetrate vegetation, providing complementary information to the TerraSAR-X and TanDEM-X missions.
Figure 2. Biomass estimates over the Amazon basin at 25-m (82-ft) resolution derived using a fusion of data from NASA’s GEDI and DLR’s TanDEM-X missions. Figure credit: Wenlu Qi CST Presentations – Session II
Chris Hakkenberg [University of California, Los Angeles (UCLA)] opened the second CST presentation session with a discussion on his research using GEDI to characterize fuel structure, burn severity, and post-fire response across the regions of California affected by wildfire. He began by highlighting significant land cover changes resulting from wildfires in recent years that are visible as enormous [greater than 100 km2 (38 mi2)] conversions from forest to grass/scrub in the National Land Cover Dataset. Hakkenberg’s project aims to examine the role of fuel structure in driving fire severity patterns, improve burn severity maps using GEDI for change detection, and characterize post-fire response using data from Landsat 5, 7, and 8 and GEDI. He noted that while fire behavior is heavily dependent on weather, topography, and fuels – only fuels can be actively managed. GEDI provides valuable insights into forest fuel structure by measuring canopy volume (total fuel quantities) and vertical continuity (how fire may spread through those volumes). Hakkenberg and his team found that vertical fuel continuity metrics were stronger predictors of severity than fuel volume, especially in extreme weather conditions, and are most closely related to the high-burn severities that can delay long-term recovery. Finally, Hakkenberg presented research that combines GEDI and Landsat to improve burn severity assessments, which will be the focus on the next phase of this research project.
Sean Healey [U.S. Forest Service (USFS)—GEDI Co-I] presented an overview of the Online Biomass Inference using Waveforms and iNventory (OBI-WAN) project. OBI-WAN provides globally consistent estimates of biomass and carbon, as well as changes in these estimates over time, for user-defined areas and periods of interest rather than fixed 1-km (0.62-mi) squares. The project leverages GEDI L4A models to predict biomass at the footprint level and uses this dense collection of footprints to create local-level biomass models with Landsat (assuming consistent calibration of Landsat through time). To quantify uncertainty in change estimates, OBI-WAN employs a statistical method called bootstrapping, which can be embedded into customized accounting systems through a powerful programming interface accessed through Google Earth Engine.
Data Product Status I
Michelle Hofton [UMD—GEDI Co-I] and Sarah Story [GSFC] returned to the topic of GEDI data product status. They presented an update on the GEDI L2A product, which includes ground topography and canopy height measurements. Encouragingly, preliminary testing shows that GEDI’s post-storage performance has remained consistent with pre-storage. Hofton explained GEDI observations are compared with high-quality intersections with the Land, Vegetation, and Ice Sensor (LVIS) (an airborne lidar) data in order to assess GEDI data quality and accuracy. She highlighted the use of bingos – pairs of GEDI waveforms believed to be spatially coincident in vegetated areas – as a valuable tool for assessing geolocation and waveform errors as well as algorithm performance. As of December 2024, more than one million bingos had been collected. Hofton and Story concluded with a preview of anticipated updates to the L2A product for V3.0, including new quality flags for data cleaning and a refined algorithm selection approach.
John Armston [UMD—GEDI Co-I] presented on GEDI L2B data, which provides gridded footprint-level [25-m (82-ft) resolution] metrics such as canopy cover (see Figure 3), plant area index (PAI), plant area volume density (PAVD), and foliage height diversity (FHD). Waveform analysis will remain largely unchanged from V2.0. He shared that the upcoming V3.0 release will differ from V2.0 in that it will use GEDI-derived canopy-ground reflectance ratios — rather than values derived from NASA LVIS — to estimate canopy cover, thus allowing for spatial variability. Waveform analysis will remain largely unchanged from V2.0. Armston also presented 1-km (0.62-mi) leaf-on and leaf-off gridded L2B canopy cover fraction maps using both pre- and post-storage data (April 2019–November 2024), explaining how post-storage data were used to fill gaps. Additionally, the mission team has mapped GEDI canopy cover distributions using a H3-indexing API developed by Tiago de Conto [UMD], which are being used to improve GEDI L2A algorithms for ground detection. V3.0 will offer a more direct measure of canopy structure to complement L2A relative height metrics by improving quality flags and including relative canopy height metrics. Finally, the team presented progress on the independent validation of GEDI L2B V3.0 algorithms and products using the GEDI FSBD and NASA LVIS campaign data from Costa Rica, Gabon, French Guiana and the United States.
Figure 3. GEDI L2B leaf-on canopy cover fraction map derived from data obtained April 2019–October 2024. Figure credit: John Armston Jamis Bruening [UMD] shared the final data product update of the day. He discussed GEDI’s L4B gridded aboveground biomass density (AGBD) product, which is a 1-km (0.62-mi) raster dataset representing area-level estimates of mean AGBD and associated uncertainty across the mission’s range of observation. GEDI’s L4B estimates are derived from the footprint-level L4A AGBD predictions through one of two statistical modes of inference. Currently, hybrid estimation is used to generate L4B. This approach uses GEDI data as the sole input and requires at least two GEDI tracks in a 1-km (0.62-mi) grid cell to produce a mean estimate. The hybrid estimator also provides a standard error, accounting for both model variance in the L4A predictions and GEDI’s sampling uncertainty. To address gaps in GEDI’s coverage where hybrid estimates cannot be produced, the team has begun implementing an alternative inference mode, called generalized hierarchical model-based (GHMB) estimation. GHMB incorporates auxiliary imagery, such as Landsat, SAR, and GEDI’s L4A predictions, to infer mean biomass and its standard error. Although the addition of post-storage data has increased GEDI’s coverage, GHMB remains essential for producing a complete, gap-free 1-km (0.62-mi) AGBD map. Both hybrid and GHMB approaches will soon be used together to generate a global, gap-free L4B product. Users can expect the release of V3.0 L4B estimates – featuring hybrid and GHMB models of biomass inference using both pre- and post-storage data – later in 2025.
What’s Next?
Day one concluded with John Armston, who presented on the potential new satellite laser altimetry mission called Earth Dynamics Geodetic Explorer (EDGE), which was competitively selected for a Phase A Concept Study under NASA’s Earth Systems Explorer Announcement of Opportunity. If selected, EDGE would launch in 2030 and operate for a two-year mission, providing a critical link between current and future satellite laser altimetry missions.
Armston explained that EDGE addresses two of the targeted observables identified in the 2017 Earth Science Decadal Survey – terrestrial ecosystem structure and ice elevation. It provides a dramatic improvement in coverage and resolution over current active missions by operating in a Sun-synchronous orbit that will enable the direct measurement of change in the three-dimensional (3D) structure of vegetation and the surface topography of ice at the spatial and temporal scales needed to observe the driving processes. EDGE will provide fine-scale detail of ecosystem structure in some of the world’s most critical and challenging-to-quantify regions, including the boreal, transforming the field’s understanding of global terrestrial ecosystem structure and its response to natural and anthropogenic change over all of Earth’s wooded ecosystems.
DAY TWO
Data Product Status II/Extended and Demonstrative Products I
Jim Kellner began day two with an L4A footprint-level AGBD product update. His presentation focused on current product status and planned evaluation of and improvements to the L4A algorithm. Since the last STM, L4A V2.1 was updated to include data through MW 311 (through November 2024) and is now available to end users. V3.0, along with an updated Algorithm Theoretical Basis Document (ATBD), is expected later in 2025. The revised ATBD will outline enhancements to the waveform simulator, quality filtering, stratification, and model selection thanks in part to the availability of on-orbit data. V3.0 will also benefit from the ingestion of approximately 35% more simulated waveforms that passed quality assurance and quality control in the FSBD, significantly expanding training data coverage, particularly in Africa and North America. Kellner noted that users should be aware of key differences between L2 and L4 quality flags; L4 flags account for factors such as sensitivity, water presence, urban conditions, and phenology. Additionally, updates to the selected models may lead to changes in AGBD estimates, which will be more clearly communicated in the V 3.0 release. Comparing pre- and post-storage data, Kellner and his team found that AGBD estimates remain stable across both periods. He encouraged users to review the updated ATBD upon release to fully understand the changes and their implications.
Tiago de Conto [UMD] presented the new GEDI L4C WSCI product, which was released in May 2024 and available through the ORNL DAAC – see Figure 4. This footprint-level metric captures the amount and variability of canopy structure in 3D space, reflecting the richness of structural information underlying any given GEDI observation. It synthesizes multiple structural attributes into a single metric and incorporates elements of both vertical and horizontal variability. WSCI models are trained at the PFT level (i.e., deciduous broadleaf trees, evergreen broadleaf trees, evergreen needleleaf trees, and the combination of grasslands, shrubs, and woodlands) using crossovers of GEDI and airborne lidar point clouds. While WSCI tends to scale with canopy height, the relationship varies across biomes. Looking ahead, de Conto previewed forthcoming WSCI–SAR fusion work designed to produce wall-to-wall maps that are suitable for applications, such as change detection. Early fusion results using data from the European Union’s Copernicus Sentinel-1 (a synthetic aperture radar mission) and the Japan Aerospace Exploration Agency’s (JAXA) Advanced Land Observing Satellite Phased Array L-band Synthetic Aperture Radar (ALOS-PALSAR) show stable prediction performance across different biomes and time periods as well as consistent performance against airborne lidar wall-to-wall reference data.
Figure 4. The global frequency distribution of GEDI L4C Waveform Structural Complexity Index. Figure credit: Tiago de Conto Paul May [South Dakota School of Mines and Technology] presented his work predicting interpolated waveforms, along with their associated uncertainties, over USDA Forest Inventory and Analysis (FIA) field plots across the contiguous United States (CONUS). This project aims to develop regression models that convert GEDI’s waveform data into measurements of key forest attributes and enhance monitoring capabilities for a variety of applications. The resulting data product – GEDI-FIA Fusion: Training Lidar Models to Estimate Forest Attributes – was released June 2025 and is publicly available through the ORNL DAAC.
Sean Healey presented ongoing work on the GEDI L4D Imputed Waveform product, led by postdoctoral researcher Eugene Seo [Oregon State University]. This product aims to generate a wall-to-wall 30-m (98-ft) resolution map of GEDI waveforms across the globe in 2023. To achieve this, Seo, Healey, and Zhiqiang Yang [USFS] are using a k-nearest neighbor (k-NN) imputation approach to address areas without GEDI observations. The model operates at a 10-km (6-mi) scale but draws neighbors from a surrounding 30×30 km (19×19 mi) window. The resulting 30-m (98-ft) resolution imputed waveform map is aligned with Landsat data from 2023. Users can expect the release of the L4D product later in 2025.
GEDI Applications and Perspectives II
Neha Hunka [European Space Agency] shared her work using GEDI to fill gaps in the Republic of Sudan’s National Forest Inventory (NFI) in support of their Forest Reference Level (FRL) report to the United Nations Framework Convention on Climate Change (UNFCCC). Using existing NFI data for calibration, Hunka and colleagues developed a geostatistical model-based approach that interpolates between NFI sample units, allowing predictions of AGBD to be made in areas of interest – see Figure 5. (Hunka was lead author on a 2025 paper in Remote Sensing of Environment that describes a similar approach to what she described in this presentation.) UNFCCC called for the modeling approach to be transparent and replicable. Hunka emphasized the importance of access to and preparation of covariate data and called for greater capacity-building and knowledge-transfer support to help other countries adopt GEDI in their reporting. Sudan’s submission marks the first time GEDI data has been used in an FRL report.
Figure 5. A geostatistical model-based approach uses data from the Republic of Sudan’s National Forest Inventory (NFI) for calibration and interpolates between NFI sample units, allowing predictions of aboveground biomass density where desired. Figure credit: Neha Hunka Forests cover about 30% of Earth’s land area, store over 80% of terrestrial biomass and carbon, and absorb around 30% of anthropogenic carbon dioxide (CO₂) emissions. While storing carbon in forests can help mitigate carbon emissions, deforestation, disturbances, shifting global economy, and low confidence in forest carbon credits add risk and uncertainty to this strategy. By monitoring forest biomass, some of these risks can be alleviated. Stuart Davies [Smithsonian Institution] joined the STM to present his work on GEO-TREES, a global forest biomass reference system aiming to provide high-quality, publicly available ground data from a network of long-term forest inventory sites to improve biomass mapping on the global scale. Despite many Earth observing (EO) missions focused on forest biomass, a lack of standardized ground reference data has hindered accurate validation. GEO-TREES addresses this need, by fostering collaboration between carbon monitoring, biodiversity research, and EO communities. The project includes 100 core sites and 200 supplementary sites across tropical and temperate regions, selected to represent environmental and human-use gradients, with greater emphasis on sampling in the tropics. Each core site follows Committee on Earth Observation Satellites (CEOS) protocol and includes three types of measurements: forest plot inventory plot, terrestrial laser scanning, and ALS.
CST Presentations – Session III
Atticus Stovall [GSFC] shared first-year findings from his research on post-fire disturbance forest recovery in Mediterranean ecosystems – specifically Spain and Portugal – where the frequency and intensity of wildfires have significantly increased in the 21st century. Using Iberian Forest Inventory ALS data and GEDI footprint data, Stovall and his team showed that GEDI can be used to assess post-fire change as well as evaluate degradation patterns from increasing fire recurrence and intensity. Stovall shared examples demonstrating the use of GEDI to detect both immediate fire effects as well as recovery after disturbance, including stand-replacement and understory clearing. By overlaying disturbance maps with GEDI data, the team observed that recovery rates differ across height class. Looking forward, they plan to investigate how recovery rates vary across environmental gradients and incorporate field plot data to validate their findings.
KC Cushman [ORNL] presented on biomass calibration and validation (cal/val) activities for the NISAR mission, which launched in July 2025. She outlined the general approach to the NISAR biomass algorithm, which uses multiple observations from NISAR every year to produce annual biomass estimates at 1-ha (0.004 mi2) resolution. Cal/val efforts will use ALS to link sparse field data to larger landscapes with estimates at two or more sites in 15 different ecoregions. NISAR has supported cal/val field plot data collection in Spain, South Africa, and various National Ecological Observatory Network (NEON) sites in addition to ALS campaigns at Agua Salud, Panama, near the Los Amigos Biological Station, Peru, in the Chaco ecosystem, Argentina, and near Madrid, Spain.
Chi Chen [Rutgers University] presented his work exploring vertical acclimation of vegetation canopy structure and photosynthetic activities using GEDI data. Chen’s research aims to generate gap-free, high-temporal-frequency canopy profile data and to develop a novel framework that integrates GEDI observations into a multi-layer canopy process model. By training a random forest model with spatially discontinuous GEDI PAVD profiles and multiple features, e.g., multiband spectral reflectance, tree height, and forest type, Chen and his team successfully estimated spatially continuous PAI profiles across different canopy heights. The team cross-validated their predicted PAI with GEDI PAI, NEON PAI, and LAI measurements from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument on NASA’s Terra and Aqua platforms. These data could be used to study seasonal variation in different canopy heights. Using a Global Multilayer Canopy OPTimization (GMC-OPT) model, they also found that GEDI-informed data has the potential to identify the vertical position of “net” seasonal leaf turnover – ultimately improving the accuracy of estimates of carbon and water fluxes.
CST Presentations – Session IV
Marcos Longo [Lawrence Berkeley National Laboratory (LBNL) in transition to Brazilian National Institute for Space Research (INPE)] and his team presented a proposed project that integrates GEDI data with process-based models to assess the impact of wildfires on forest structure, recovery, and ecosystem function. As western US forests face increasing wildfire risk due to drier climates, more human ignitions, and a legacy of fire suppression, changes in forest structure, composition, and function are likely to become more detectable over time. This project, under the leadership of Robinson Negron Juarez [University of California, Irvine and LBNL—PI] aims to quantify biomass changes in mixed conifer forests across California, Oregon, and Washington using both ALS and GEDI data. The team plans to use GEDI L2A and L2B data to assess immediate fire impacts on forest structure, investigate post-fire forest recovery, and establish relationships between forest structure and fire intensity/severity. This information will inform process-based models – e.g., the U.S. Department of Energy’s Functionally Assembled Terrestrial Ecosystem Simulator (FATES) – and support a better understanding of forest resilience under fire disturbance regime changes.
Ovidiu Csillik [Wake Forest University] presented work using GEDI and ALS to investigate biomass and structural changes in tropical forests. The work, conducted with Michael Keller [NASA/ Jet Propulsion Laboratory (JPL), USFS—PI], aims to use models to quantify changes in tropical forest biomass and evaluate understanding of topical forest productivity drivers. The project will use ALS data from over the Brazilian Amazon and other sites in Brazil, Gabon, French Guiana, Costa Rica, Mexico, and Borneo alongside GEDI data over pantropical regions around the globe. Csillik and Keller are currently conducting ALS–ALS and GEDI–GEDI comparisons, and are planning to estimate aboveground biomass change from ALS–ALS, ALS–GEDI, and GEDI–GEDI comparisons at both regional and pantropical scales from 2008–2026.
Zhenpeng Zuo [Boston University (BU)] presented his team’s research, led by Ranga Myneni [BU—PI], using mechanistic model-GEDI integration to map potential canopy top height (pCTH) and inform forest restoration planning. Predicting restoration potential is challenging, as empirical, deep-learning, and mechanistic methods vary in their accuracy, interpretability, and spatial detail. This project uses a mechanistic model based on water use and supply equilibrium, calibrated using GEDI canopy height metrics, to predict pCTH. The team found this approach produced robust pCTH predictions and shows that the Eastern US has vast restorable areas. Future work will expand to dynamic modeling to incorporate disturbance risks and effects under different climate scenarios.
DAAC Reports
Rupesh Shrestha [ORNL DAAC] presented the status of GEDI L3 and L4 datasets at the ORNL DAAC – see Figure 6. Since the 2023 STM, three new datasets have been released: L4B (country-level summaries of aboveground biomass), L4C (footprint level waveform structural complexity index), and a GEDI-FIA (fusion dataset for training lidar models to forest attributes). In total, almost 34,000 unique users have downloaded GEDI L3 and L4A-C data 13,770,648 times, with L4A being the most popular at 13.1 million downloads. As of April 30, 2025, all GEDI footprint-level datasets from L1–L4 are available with data through mission week 311 (November 2024), besides L4C. Users can look forward to a GEDI L4D Imputation Dataset later in 2025 along with the much-anticipated V3.0 GEDI data product release. All levels of GEDI data can now be accessed in one place through the NASA Earthdata Search and Data Catalog. In addition to the data products themselves, data tools and services, publications citing GEDI data and GEDI data tutorials and workshops can be found at the ORNL DAAC website. The ORNL DAAC provides data user support through the Earthdata Forum, or via their email uso@daac.ornl.gov.
Figure 6. GEDI L3 and L4B data projected on NASA WorldView/GIBS API. Explore the program here. Figure credit: Rupesh Shrestha Jared Beck [Land Processes Distributed Active Archive Center (LP DAAC)] presented on GEDI data products at the LP DAAC, a USGS–NASA partnership that archives and distributes lower-level GEDI products (GEDI L1B, L2A, and L2B). The LP DAAC has distributed over seven petabytes of GEDI data so far, and is now exclusively distributing GEDI data through NASA’s Earth Data Cloud. GEDI L2A is the most popular of the products in terms of terabytes of distribution. Like the ORNL DAAC, data user support also flows through the NASA Earthdata Forum. Tutorials can be found on GitHub. All levels of GEDI data can now be accessed in one place through the NASA Earthdata search and data catalog options.
GEDI Extended and Demonstrative Products II
Scott Goetz [NAU] discussed his team’s research leveraging GEDI data for biodiversity applications, emphasizing its potential to help improve species distribution models and the high value of understanding forest structure for conservation assessments. He highlighted a 2022 Nature Ecology & Evolution article showing that forests with higher structural integrity and cover reduced the extinction risk for over 16,000 threatened or declining species. Another 2023 study in Nature demonstrated how biodiversity indicators, such as habitat cover, canopy structure, and human pressures, can influence the effectiveness of protected areas. In order to have a wider variety of gridded products to work with for species distribution models, Pat Burns [NAU], Chris Hakkenberg, and Goetz developed the Gridded GEDI Vegetation Structure Metrics and Biomass Density at Multiple Resolutions product that has been released through the ORNL DAAC and Google Earth Engine along with a data descriptor paper published in a Nature Scientific Data paper – see Figure 7. Burns elaborated that, relative to fusion products, gridded GEDI products performed better when measuring structure, especially in the understory. The team is now comparing species distribution models in mainland Southeast Asia using fusion versus solely GEDI data.
Figure 7. GEDI mean foliage height diversity (FHD) map using shots from April 2019 to March 2023 at 6-km (4.7-mi) spatial resolution. Red boxes indicate the approximate location of airborne lidar used for intercomparison. Three insets show GEDI mean FHD at finer spatial resolution [1 km (0.62 mi)] as well as more detailed airborne lidar coverage (red polygons). From left to right the insets show: Sonoma County, California, Coconino National Forest, Arizona, and Sumatra/Borneo. Figure credit: From Patrick Burns et al (2024) Nature Scientific Data Perspectives III
STM attendees concluded day two with a perspective talk from Marc Simard [JPL], who showcased a range of studies demonstrating diverse applications of GEDI data, opportunities for its improvement, and potential for informing future scientific research. Drawing on his own work, Simard shared examples of using GEDI data for cal/val of global Digital Elevation Measurement (DEM) and Digital Terrain Model (DTM), mapping global mangrove heights, monitoring forest growth, and analyzing hydrological processes. In more detail, he explained how he led the development of a 12-m (39-ft) spatial resolution global mangrove height product using GEDI and TanDEM-X data. Additionally, he discussed a study evaluating tree growth rates in the Laurentides Wildlife Reserve in Quebec, Canada using both GEDI and ALS. The analysis revealed an average growth rate of approximately 32 ±23 cm (12 ±9 in) per year. Finally, he presented a paper under review examining water level detection and hydrological conditions in coastal regions using GEDI alongside Ice, Cloud, and land Elevation Satellite 2 (ICESat-2) data. In closing, Simard emphasized that GEDI datasets can help identify critical data and knowledge gaps, guiding the development of new missions – e.g., the Surface Topography and Vegetation (STV) mission concept called for in the 2017 Earth Science Decadal Survey report. As described in the STV Study Team Report, the mission would focus on elevation and vertical structure to study the solid Earth, cryosphere, vegetation structure, hydrology, and coastal geomorphology.
DAY THREE
CST Presentations – Session V
Jody Vogeler [Colorado State University] opened the final CST presentation session with an overview of her research using GEDI data fusions to characterize post-fire landscapes and understand habitat refugia for the threatened Canada lynx (Lynx canadensis). This project builds on her team’s phase-I work, which produced 30-m (98-ft) resolution gridded GEDI fusion maps across six Western U.S. states to support habitat and diversity applications related to cavity-nesting birds, small mammals, and carnivore–prey relationships. The team is now focusing on validating and improving their GEDI fusion products within post-disturbance landscapes, specifically post-fire. Using this data, Vogeler and her team aim to better understand how post-fire structural information from GEDI improves their ability to understand lynx behavior–habitat relationships across early post-fire landscapes. This information can help evaluate what structural attributes determine post-fire refugia patch use by lynx. Next steps for this work include integrating GEDI V3.0 data upon its release, identifying new GEDI metrics and derived products, and incorporating lynx radio-collar data into their analyses. Vogeler also presented her work as co-PI on a NASA Ecological Conservation Project in Greater Kruger National Park, South Africa, where she and her team are developing spatial monitoring tools to support management and conservation planning.
Jingfeng Xiao [University of New Hampshire] provided updates on his team’s research using GEDI data to understand how structural diversity influences productivity and carbon uptake of forests in the United States. The project aims to assess GEDI’s ability to quantify structural diversity, investigate how that diversity regulates forest productivity and carbon uptake, and understand its role in resilience of forest productivity to drought. When analyzing the relationships of gross primary production (GPP) and evapotranspiration (ET) with canopy structure metrics, the team found that increased canopy structure complexity positively affected GPP and ET and reduced their seasonal variability. They also found that greater canopy complexity improved ecosystem resistance to drought. As part of the project, the team also produced 1-km (0.62-mi) resolution gridded maps of GPP and ET.
Lei Ma [UMD] delivered the final talk of the STM, presenting his project that integrates GEDI observations with mechanistic ecosystem modeling to quantify forest regrowth in a changing climate. Ma used GEDI data and the Ecosystem Demography (ED) model in his research and found that height and aboveground biomass (AGB) regrowth rates can be derived by combining GEDI and land-use and land-cover change data. Ma found that regrowth rates derived from different inputs are generally consistent at large scales but variable at fine scales. Notably, regrowth rates showed temporal dependence, decreasing by roughly 50% every decade. Lastly, Ma and his team found that spatial variation in height and AGB regrowth rates can be partially explained by environmental conditions and disturbance frequency.
Conclusion
The 2025 GEDI STM was especially exciting, as it came on the cusp of post-storage data being processed and released as V2.1. Additionally, it marked the first time the new CST cohort presented on their research and joined breakout sessions with the wider GEDI team. The meeting highlighted the mission’s ongoing success and scientific value following hibernation on the ISS. Looking ahead, data users can anticipate the V3.0 product release later in 2025.
Talia Schwelling
University of Maryland College Park
tschwell@umd.edu
Share
Details
Last Updated Aug 18, 2025 Related Terms
Uncategorized Earth Science View the full article
-
By NASA
Explore This Section Earth Earth Observer Editor’s Corner Feature Articles Meeting Summaries News Science in the News Calendars In Memoriam Announcements More Archives Conference Schedules Style Guide 21 min read
A Decade of Global Water Cycle Monitoring: NASA Soil Moisture Active Passive Mission
Introduction
The NASA Soil Moisture Active Passive (SMAP) mission, launched in 2015, has over 10 years of global L-band radiometry observations. The low frequency [1.4 GHz frequency or 21 cm (8 in) wavelength] measurements provide information on the state of land surfaces in all weather conditions – regardless of solar illumination. A principal objective of the SMAP mission is to provide estimates of surface soil moisture and its frozen or thawed status. Over the land surface, soil moisture links the water, energy, and carbon cycles. These three cycles are the main drivers of regional climate and regulate the functioning of ecosystems.
The achievement of 10 years in orbit is a fitting time to reflect on what SMAP has accomplished. After briefly discussing the innovative measurement approach and the instrument payload (e.g., a radiometer and a regrettably short-lived L-band radar), a significant section of this article is devoted to describing the mission’s major scientific achievements and how the data from SMAP have been used to serve society (e.g., applied sciences) – including SMAP’s pathfinding role as Early Adopters. This content is followed by a discussion of how SMAP has dealt with issues related to radio frequency interference in the L-Band region, a discussion of the SMAP data products suite, future plans for the SMAP active–passive algorithm, and a possible follow-on L-band global radiometry mission being developed by the European Union’s Copernicus Programme that would allow for data continuity beyond SMAP. This summary for The Earth Observer is excerpted from a longer and more comprehensive paper that, as of this article’s posting, is being prepared for publication in the Proceedings of the Institute of Electrical and Electronics Engineers (IEEE).
SMAP Measurement Approach and Instruments
The SMAP primary and operating instrument is the L-band radiometer, which collects precise surface brightness temperature data. The radiometer includes advanced radio frequency interference (RFI) detection and mitigation hardware and software. The radiometer measures vertical and horizontal polarization observations along with the third and fourth Stokes parameters (T3 and T4) of the microwave radiation upwelling from the Earth. The reflector boom and assembly, which includes a 6 m (20 ft) deployable light mesh reflector, is spun at 14.6 revolutions-per-minute, which creates a 1000 km (621 mi) swath as the SMAP satellite makes its Sun-synchronous orbit of the Earth – see Figure 1. This approach allows coverage of the entire globe in two to three days with an eight-day exact repeat. The radiometer instrument is calibrated monthly by pointing it to the deep sky.
Figure 1. An artist’s rendering of the SMAP Observatory showing both the radiometer and radar. Figure credit: NASA/Jet Propulsion Laboratory/California Institute of Technology The original SMAP instrument design included a companion L-band radar, which operated from April through early July 2015, acquiring observations of co- and cross-polarized radar backscatter at a spatial resolution of about 1 km (0.6 mi) with a temporal revisit of about three days over land. This data collection revealed the dependence of L-band radar signals on soil moisture, vegetation water content, and freeze thaw state. The radar transmitter failed on July 7, 2015. Shortly thereafter, the radar receiver channels were repurposed to record the reflected signals from the Global Navigation Satellite System (GNSS) constellation in August 2015, making SMAP the first full-polarimetric GNSS reflectometer in space for the investigation of land surface and cryosphere.
Scientific Achievements from a Decade of SMAP Data
A decade of SMAP soil moisture observations have led to a plethora of scientific achievements. The data have been used to quantify the linkages of the three main metabolic cycles (e.g., carbon, water, and energy) on land. They have also been used to improve drought assessments and flood prediction as well as the accuracy of numerical weather prediction (NWP) models. They are also used to measure liquid water and thickness of ice sheets, and sea surface salinity. The subsections that follow describe how SMAP data are being put to use in myriad ways that benefit society.
Quantifying Processes that Link the Terrestrial Water, Energy, and Carbon Cycles
The primary SMAP science goal is to develop observational benchmarks of how the water, energy, and carbon cycles link together over land. Soil moisture is the variable state of the land branch of the water cycle. It links the water cycle to the energy cycle through limiting latent heat flux – the change in energy as heat exchanges when water undergoes a phase change, such as evapotranspiration at the land–atmosphere interface. Soil moisture also links the water and carbon cycles, which is evident through plant photosynthesis. SMAP global observations of soil moisture fields, in conjunction with remote sensing of elements of the energy and carbon cycles, can reveal how these three cycles are linked in the real world as a benchmark for weather and Earth system models.
Photosynthesis is down-regulated by both the deficit in water availability and the lack of an adequate amount of photosynthetically active radiation. Global maps reveal how soil moisture and light regulate photosynthesis – see Figure 2. These benchmark observational results can be used to assess how Earth system models link to the three main metabolic cycles of the climate system.
Figure 2. Observed regulation of photosynthesis by water availability [left] and light availability [right]. Blue denotes greater limitation. Photosynthesis rates for both maps determined using solar-induced fluorescence (SIF) measurements (mW/m2 nm sr) from the Tropospheric Ozone Monitoring Instrument (TROPOMI) on the European Union’s Copernicus Sentinel-5P mission. Water availability was determined using soil moisture (SM) measurements from the Soil Moisture Active Passive (SMAP) mission. Light availability was determined using measurements of photosynthetically active radiation (PAR) from the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Terra and Aqua platforms. The resulting maps show the model slope (mW/m2/nm/sr) of the estimated SIF-SM relationship in the water-limited regime [left] and the model slope (10-3/nm/sr) of estimated SIF-PAR relationship in the light-limited regime [right]. Figure credit: Jonard et al (2022) in Biogeosciences Development of Improved Flood Prediction and Drought Monitoring Capability
SMAP products have also been widely used in applied sciences and natural hazard decision-support systems. SMAP’s observation-based soil moisture estimates offer transformative information for managing water-related natural hazards, such as monitoring agricultural drought – defined as a persistent deficit in soil moisture – and flood volumes – defined as the landscape’s water absorption capacity during precipitation events. The SMAP project produces a parallel, near-real-time data stream that is accessed by a number of federal and state agencies in decision-support systems related to drought monitoring, food security, and landscape inundation and trafficability.
Enhancing Weather and Climate Forecasting Skill
SMAP’s enhancement of numerical weather prediction, model skill, and reduction of climate model projection uncertainties is based on the premise of the contribution of solar energy to weather and climate dynamics. Soil moisture has a strong influence on how available solar energy is partitioned into components (e.g., sensible heat flux versus latent heat flux) over land. The influence propagates through the atmospheric boundary layer and ultimately influences the evolution of weather.
To give an example, land surface processes can affect the evolution of the U.S. Great Plains low-level jets (GPLLJs). These jets drive mesoscale convective weather systems. Previous studies have shown that GPLLJs are sensitive to regional soil moisture gradients. Assimilation of SMAP soil moisture data improves forecasts of weakly synoptically forced or uncoupled GPLLJs compared to forecasts of cyclone-induced coupled GPLLJs. For example, the NASA Unified Weather Research and Forecasting Model, with 75 GPLLJs at 9 km (5.6 mi) resolution both with and without SMAP soil moisture data assimilation [SMAP data assimilation (DA) and no-DA respectively], shows how the windspeed mean absolute difference between SMAP DA and no-DA increase approximately linearly over the course of the simulation with maximum differences at 850 hPa (or mb) for the jet entrance and core – see Figure 3.
Figure 3. The impact of adding soil moisture data [SMAP data assimilation (DA) minus no-DA] to a model simulation from theNASA Unified Weather Research and Forecasting Model (NU-WRF)) of the Great Plains Low Level Jet (GPLLJ). The results show the mean over 75 independent GPLLJ events. The plots correspond to wind speed difference with height (y-axis) and time (hours on x-axis). The panels are for jet entrance [left], jet core [middle] and jet exit [right]. Soil moisture data assimilation enhances the intensity of the simulated GPLLJ. The stippling corresponds to 99% statistical confidence. Figure credit: Ferguson (2020) in Monthly Weather Review Measuring Liquid Water Content and Thickness of Ice Sheets
The mass loss of Greenland and Antarctica ice sheets contributes to sea-level rise – which is one of the most impactful and immediate damaging consequences of climate change. The melt rates over the last few years have raised alarm across the globe and impact countries with coastal communities. The cryosphere community has raised a call-to-action to use every observing system and model available to monitor the patterns and rates of land ice melt.
Surface melt affects the ice cap mass loss in many ways: the direct melt outflow from the ablation zone of the Greenland ice sheet, the structural change of the percolation zone of the Greenland ice sheet, changes in the melt water retention and outflow boundaries, changes in the structure of the Antarctic ice shelves, and destabilization of the buttressing of the glacier outflow through various processes (e.g., hydrofracturing and calving). The long-term climate and mass balance models rely on accurate representation of snow, firn, and ice processes to project the future sea level.
The SMAP L-band radiometer has relatively long wavelength [21 cm (8 in)] observations compared to other Earth-observing instruments. It enables the measurement of liquid water content (LWC) in the ice sheets and shelves as it receives the radiation from the deep layers of the snow/firn/ice column. Relatively high LWC values absorb the emission only partially, making the measurement sensitive to different liquid water amounts (LWA) in the entire column. Figure 4 shows the cumulative LWA for 2015–2023 based on SMAP measurements.
Figure 4. Total annual sum of SMAP daily liquid water amount (LWA) for 2015–2023. The black solid line on each map represents grid edges, and the grey color mask inside the ice sheet indicates melt detections by decreasing brightness temperature. Figure Credit: Andreas Colliander [Finnish Meteorological Institute]. The SMAP L-band radiometer has also been used to derive the thickness of thin sea ice [Soil Moisture and Ocean Salinity (SMOS) mission have been recalibrated to SMAP, using the same fixed incidence angle. The data show strong agreement and demonstrate clear benefits of a combined dataset. The L-band thin ice thickness retrievals provide a useful complement to higher-resolution profiles of thicker ice obtained from satellite altimeters (e.g. ESA’s CryoSat-2 and NASA’s Ice, Clouds and land Elevation Satellite–2 missions).
Extending and Expanding the Aquarius Sea Surface Salinity Record
The joint NASA/Argentinian Aquarius/Satélite de Aplicaciones Científicas (SAC)-D (Aquarius), which operated from 2011–2015, used an L-band radiometer and an L-band scatterometer to make unprecedented monthly maps of global sea surface salinity at 150-km (93-mi) resolution. The SMAP L-band radiometer has not only extended the sea surface salinity record in the post-Aquarius period, it has also increased the spatial resolution and temporal frequency of these measurements because of its larger reflector and wider swath. The increased resolution and revisit allow new and unprecedented perspectives into mixing and freshwater events, coastal plume tracking, and other more local oceanic features.
Providing New Perspectives on Global Ecology and Plant Water Stress
The L-band vegetation optical depth (VOD) – which is related to water content in vegetation – has been retrieved simultaneously with soil moisture using SMAP’s dual-polarized brightness temperatures and is being used to better understand global ecology. Water in above-ground vegetative tissue attenuates and thus depolarizes surface microwave emission, and VOD quantifies this effect. SMAP can provide global observations of VOD in all weather conditions with a two to three day temporal frequency. Changes in VOD indicate either plant rehydration or growth. Ecologists benefit from this new ecosystem observational data, which augments optical and near-infrared vegetation indices [e.g., leaf area index (LAI)] and has a higher temporal frequency that is not affected by clouds and does not saturate as rapidly for dense vegetation.
Examples of how the data have been used include deciphering the conditions when vegetation uptakes soil water only for rehydration (i.e., VOD increase with no LAI change) compared to plant growth (i.e., increase in both VOD and LAI). The applications of VOD are increasing and the ecology community views this product as a valuable additional perspective on soil–plant water relations.
At the moment, this measurement has no ground-based equivalent. Therefore, field experiments with airborne instruments and ground sampling teams are needed to firmly establish the product as a new observational capability for global ecology.
Applied Science Collaboration: SMAP Observations Serving Society
The SMAP project has worked with the NASA Earth Science Division Applied Sciences Program (now known as Earth Science to Action) and the natural hazards monitoring and forecasting communities for pre- and post-launch implementation of SMAP products in their operations. In some operational applications, for which long-term data continuity is a requirement, the SMAP data are still used for assessment of current conditions, as well as research and development.
The Original Early Adopters
Prior to its launch, the SMAP mission established a program to explore and facilitate applied and operational uses of SMAP mission data products in decision-making activities for societal benefit. To help accomplish these objectives, SMAP was the first NASA mission to create a formal Applications Program and an Early Adopter (EA) program, which eventually became a requirement for all future NASA Earth Science directed satellite missions. SMAP’s EA program increases the awareness of mission products, broadens the user community, increases collaboration with potential users, improves knowledge of SMAP data product capabilities, and expedites the distribution and uses of mission products after launch.
SMAP Data in Action
Several project accomplishments have been achieved primarily through an active continuous engagement with EAs and operational agencies working towards national interests. SMAP soil moisture data have been used by the U.S. Department of Agriculture (USDA) for domestic and international crop yield applications. For example the USDA’s National Agricultural Statistics Service (NASS) conducts a weekly survey of crop progress, crop condition, and soil moisture condition for U.S. cropland. NASS surveys and publishes state-level soil moisture conditions in the NASS Crop Progress Report.
The traditional field soil moisture survey is a large-scale, labor-intensive data collection effort that relies heavily on responses from farmers, agricultural extension agents and/or other domain experts for field observations. One weakness of these observations is that they are based on subjective assessments rather than quantitative measures and can lead to spatial inconsistency based on the human responses from the respective counties. Moreover, the NASS Crop Progress Reports do not provide specific geolocation information for the assessed soil moisture conditions – which are extremely useful metadata to provide to data users. NASS implemented the use of SMAP observations in their weekly reports during the growing period (March–November). SMAP maps estimated root-zone soil moisture for the week of November 14–20, 2022, over NASS Pacific (California and Nevada) and Delta (Arkansas, Mississippi and Louisiana) regional domains—see Figure 5.
Figure 5. SMAP-based soil moisture estimates for California, Nevada, Arkansas, Mississippi, and Louisiana, used by the U.S. Department of Agriculture’s (USDA) National Agricultural Statistics Service (NASS) in their weekly report covering November 14–20, 2022. These data are available for selected states at the NASS website linked in the text. Figure Credit: NASS SMAP Radio Frequency Interference Detection and Mitigation
Although SMAP operates within the protected frequency allocation of 1400–1427 MHz, the radiometer has been impacted by radio frequency interference over the mission lifetime. Unauthorized in-band transmitters as well as out-of-band emissions from transmitters operating adjacent to the allocated spectrum have been observed in SMAP measurements since its launch. The previously launched SMOS and Aquarius radiometers provide evidence of global RFI at L-band. Consequently, SMAP was designed to incorporate a novel onboard digital detector on the back end to enable detection and filtering of RFI. The radiometer produces science data in time and frequency, enabling the use of multiple RFI detection methods in the ground processing software.
On-orbit data demonstrate that the RFI detection and filtering performs well and improves the quality of SMAP brightness temperature measurements. The algorithms are most effective at filtering RFI that is sparse in time and frequency, with minimal impact on the noise equivalent delta temperature (NEDT) – a measure of the radiometer sensitivity. Some areas of the globe remain problematic as RFI that is very high level and persistent results in high percentages of data loss due to removal of contaminated data. A global map of RFI detection rate for January 2025 shows a large contrast between Eastern and Western Hemispheres and between Northern and Southern Hemispheres – see Figure 6. Regions of isolated RFI and severe RFI correspond to populated areas. A detection rate of 100% means all pixels are flagged and removed, resulting in data loss. Analysis of spectral information reveal many sources are likely terrestrial radar systems; however, many wideband, high-level sources and low-level, non-radar sources also persist. Over areas of geopolitical conflict, the time-frequency data show interference covering the entire radiometer receiver bandwidth.
Figure 6. Percentage of pixels on a 0.25° grid for January 2025 that have been flagged for removal by the Soil Moisture Active Passive radio frequency interference detection algorithms. Figure Credit: Priscilla N. Mohammed [GSFC] The RFI challenge is further addressed through official spectrum management channels and formal reports that include the geolocated coordinates of sources, interference levels, frequency of occurrence during the observed period, and spectral information – all of which aid field agents as they work to identify potential offenders. Reports are submitted to the NASA Spectrum office and then forwarded to the country of interest through the Satellite Interference Reporting and Resolution System.
SMAP Science Data Products
The current suite of SMAP science data products is available in the Table. The principal data products are grouped in four levels designated as L1–4. The L1 products are instrument L-band brightness temperature in Kelvin and include all four Stokes parameters (i.e., horizonal and vertical polarization as well as third and fourth Stokes). Both 6:00 AM equatorial crossing (descending) and 6:00 PM equatorial crossing (ascending data) are contained in the products. The user has access to quality flags of the conditions under which measurements are available for each project. The L1B products are time-ordered and include fore and aft measurements. L1C products are on the Equal-Area Scalable Earth V2 (EASE2) grid with polar and global projections. L2 data products are geophysical retrievals (i.e., soil moisture, VOD, and binary freeze/thaw classification on a fixed Earth grid). The L2 half-orbit products are available to the public within a day of acquisition. L3 products are daily composites and include all half-orbits for that day.
The SMAP project also produces L4 data that are the result of data assimilation. The L4 products take advantage of other environmental observations, such as precipitation, air temperature and humidity, radiative fluxes at the land surface, and ancillary land use and soil texture information, to produce estimates of surface [nominally 0–5 cm (0–2 in)] and subsurface (e.g., root-zone up to a meter) soil moisture. The data assimilation system is a merger of model and measurements and hence resolves the diurnal cycle of land surface conditions. The data assimilation system also provides estimates of surface fluxes of carbon, energy, and water, such as evaporation, runoff, gross primary productivity (GPP), and respiration. The difference between GPP and respiration is the net ecosystem exchange, which is the net source/sink of the carbon cycle over land.
The SMAP suite of products also include near-real-time (NRT) brightness temperature and soil moisture products for use in operational weather forecast applications. The NRT product targets delivery to users within three hours of measurement acquisition. The NRT uses predicted SMAP antenna pointing (instead of telemetry) and model predicted ancillary data (soil temperature) in order to support operational centers that require more than three hours of data products for updating weather forecast models. To date SMAP has met its required and target (for NRT) latency requirements.
Two other data projects merge synergistically with other (colocated) satellite measurements. The SPL2SMAP_S merges SMAP L-band radio brightness measurements with C-band synthetic aperture radar (SAR) measurements from the ESA Copernicus Sentinel-1 mission. The SAR data have high resolution and allow the generation of 1 and 3 km (0.62 and 1.8 mi) merged surface soil moisture estimates. The high resolution soil moisture information, however, is only available when there is coincident SMAP and Sentinel-1 measurements. The refresh rate of this product is limited and can be as long as 12 days.
The merged SMOS–SMAP passive L-band radiometry data allows the generation of global, near daily surface soil moisture estimates, which are required to resolve fast hydrologic processes, such as gravity drainage and recharge flux. These parameters are only partially resolved with the SMAP, with a two to three day data refresh rate. This product interpolates the multi-angular SMOS data to the SMAP 40º incident angle and uses all SMAP algorithms, including correction of waterbody impact on SMAP brightness temperature, and ancillary data for geophysical inversions to soil moisture and VOD, ensuring consistency. The combined SMAP–SMOS data product may not be available daily across locations, such as Japan, parts of China, and the Middle East, where RFI affects data collection.
Table. Soil Moisture Active Passive suite of science products are available through the National Snow and Ice Data Center, one of NASA’s Distributed Active Archive Centers.
Product Type Product description Resolution (Gridding) Granule Extent SPL1BTB Geolocated, calibrated brightness temperature in time order 36 km Half Orbit SPL1CTB_E Backus-Gilbert interpolated, calibrated brightness temperature in time order (9 km) Half Orbit SPL1CTB Geolocated, calibrated brightness temperature on Equal-Area Scalable Earth V2 (EASE2) grid 36 km Half Orbit SPL1CTB_E Backus-Gilbert interpolated, calibrated brightness temperature on EASE2 grid (9 km) Half Orbit SPL2SMP Radiometer soil moisture and vegetation optical depth 36 km Half Orbit SPL2SMP_E Radiometer soil moisture and vegetation optical depth based on SPL1CTB (9 km) Half Orbit SPL2SMAP_S SMAP radiometer/Copernicus Sentinel-1 soil moisture 3 km Sentinel-1 SPL3SMP Daily global composite radiometer soil moisture and vegetation optical depth based on SPL1CTB 36 km Daily–Global SPL3SMP_E Daily global composite radiometer soil moisture and vegetation optical depth based on SPL1CTB_E (9 km) Daily–Global SPL3FTP Daily composite freeze/thaw state based on SPL1CTB 36 km Daily–Global SPL3FTP_E Daily composite freeze/thaw state based on SPL1CTB_E (9 km) Daily–Global SPL4SMAU Surface and Root Zone soil moisture 9 km 3 hours – Global SPL4CMDL Carbon Net Ecosystem Exchange 9 km Daily–Global SPL1BTB_NRT Near Real Time Geolocated, calibrated brightness temperature in time order 36 km Half Orbit SPL2SMP_NRT Near Real Time Radiometer soil moisture 36 km Half Orbit L2/L3 SMOS SM SMOS soil moisture and VOD based on SMAP algorithms (9 km) Half Orbit/Daily Global Future Directions for the SMAP Active–Passive Algorithm
Although the SMAP radar failed not long after launch, the data that were collected have been used to advance the development of the SMAP Active–Passive (AP) algorithm, which will be applied to the combined SMAP radiometer data and radar data from the NASA–Indian Space Research Organisation (ISRO) Synthetic Aperture Radar [NISAR] mission, a recently-launched L-Band Synthetic Aperture mission to produce global soil moisture at a spatial resolution of 1 km (0.62 mi) or better. The high resolution product can advance applications of SMAP data (e.g., agricultural productivity, wildfire, and landslide monitoring).
Data Continuity Beyond SMAP
A forthcoming mission meets some – but not all – of the SMAP measurement requirements and desired enhancements. The European Union’s Copernicus Program Copernicus Imaging Microwave Radiometer (CIMR) mission is a proposed multichannel microwave radiometry observatory that includes L-band and four other microwave channels sharing a large mesh reflector. The mesh reflector is similar to the one that is used on SMAP, but larger. The successful SMAP demonstration of rotating large deployable mesh antennas for Earth observations has been useful to the CIMR design.
In terms of RFI detection capability, CIMR will also use an approach that is similar to SMAP. With regard to instrument thermal noise (NEDT) and data latency, CIMR meets or comes close to the next-mission desired characteristics and equals or exceeds SMAP in most of the attributes. The native L-band resolution of CIMR is ~60 km (37 mi); however, the measurements are coincident and higher-resolution measurements in this configuration allow reconstruction of L-band radiometry at higher resolution than CIMR’s L-band. It may be possible to combine the L- and C-bands and achieve a reconstructed ~15 km (9 mi) L-band product based on the coincident and overlapping measurements. A refresh rate of one day is possible with the wide-swath characteristic of CIMR.
CIMR is currently in development; the first version, CIMR-1A, is expected to launch within this decade and the second version, CIMR-1B, in the mid 2030s. Since the Copernicus program supports operational activities (e.g., numerical weather prediction), the program includes plans for follow-on CIMR observatories so that the data record will be maintained without gaps in the future.
Conclusions
The SMAP mission was launched in 2015 and has produced over 10 years of science data. Because of its unique instrument and operating characteristics, the global low-frequency microwave radiometry with the SMAP observatory has resulted in surface soil moisture, vegetation optical depth, and freeze/thaw state estimates that outperform past and current products. The data have been widely used in the Earth system science community and also applied to natural hazards applications.
The Earth system science and application communities are actively using the decade-long, high-quality global L-band radiometry. The intensity and range of SMAP science data usage is evident in the number of peer-reviewed journal publications that contain SMAP or Soil Moisture Active Passive in their title or abstract and use SMAP data in the study (i.e., search: www.webofscience.com data-base). The authors acknowledge that many publications escape this particular query approach. Currently the bibliography includes over 1700 entries and over 20,000 citations spanning several elements of Earth system science, including hydrologic science and regional and global water cycle, oceanic and atmospheric sciences, cryosphere science, global ecology as well as microwave remote sensing technologies.
To Learn More About SMAP
A more comprehensive bibliography of studies published based on SMAP data products, a set of one-page SMAP science and applications highlights in standardized format, and SMAP project documents including assessment reports are all available online via the links provided.
Acknowledgements
The authors wish to acknowledge the contributions of the SMAP Science Team, the SMAP Algorithm Development Team, and the SMAP Project Office engineers and staff. All of these teams contribute to the ongoing SMAP science product generation and uses reported in this article.
Dara Entekhabi
Massachusetts Institute of Technology
darae@mit.edu
Simon Yueh
Jet Propulsion Laboratory/California Institute of Technology
simon.h.yueh@jpl.nasa.gov
Rajat Bindlish
NASA Goddard Space Flight Center
rajat.bindlish@nasa.gov
Mark Garcia
Jet Propulsion Laboratory/California Institute of Technology
mark.d.garcia@jpl.nasa.gov
Jared Entin
NASA Headquarters
jared.k.entin@nasa.gov
Craig Ferguson
NASA Headquarters
craig.r.ferguson@nasa.gov
Share
Details
Last Updated Aug 18, 2025 Related Terms
Earth Science View the full article
-
By NASA
Explore This Section Earth Earth Observer Editor’s Corner Feature Articles Meeting Summaries News Science in the News Calendars In Memoriam Announcements More Archives Conference Schedules Style Guide 12 min read
Summary of the 54th U.S.–Japan ASTER Science Team Meeting
Introduction
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Science Team (ST) organized a three-day workshop that took place June 9–11, 2025, at the Japan Space System’s (JSS) offices in Tokyo, Japan. About 25 people from Japan and the United States participated during the in-person meeting – see Photo 1.
U.S. participants included representatives of NASA/Jet Propulsion Laboratory (JPL); two universities – University of Arizona (UA) and University of Pittsburgh (UPitt); and Grace Consulting. Japanese participants represented JSS, the Geologic Survey of Japan (GSJ), National Institute of Advanced Industrial Science and Technology (AIST), National Institute for Environmental Studies (NIES), and the Remote Sensing Technology Center of Japan (RESTEC). Participants from Ibaraki University (IU), Nagoya University (NU), University of Tokyo (UT), and University of Tsukuba (Uts) also joined.
Photo 1. Several attendees sit for a photo at the 54th ASTER Science Team meeting at the Japan Space System’s offices in Tokyo, Japan. Photo credit: Osamau Kashimura The main objectives of the 54th ASTER STM were to:
discuss impacts of the proposed NASA budget reductions for Fiscal Year (FY) 2026; respond to plans for future impacts on ASTER from possible power reductions on the Terra platform; receive updates on data acquisition status, data calibration and validation (cal/val) activities, data distribution plans, and applications using ASTER observations; and discuss the end-of-mission plans for Terra and ASTER and archive documentation requirements. The remainder of this article summarizes the highlights from the meeting, which includes an overview of the opening plenary session and summaries of the four working group sessions. A brief review of the closing plenary, which included summary reports from the chairpersons of all working groups, rounds out the report, followed by some overall concluding thoughts.
Opening Plenary Session
Yasushi Yamaguchi [NU—Japan ASTER ST Lead] and Michael “Mike” Abrams [JPL—U.S. ASTER ST Lead] welcomed participants and reviewed the agenda for the opening plenary and the schedule for the working group sessions.
Abrams presented highlights of science results based on ASTER data. He also discussed some issues that Woody Turner [NASA Headquarters—ASTER Program Scientist] had raised, including NASA’s response to the President’s proposed fiscal year (FY) 26 budget and the status of FY25 funding. Abrams reported that Terra passivation is currently scheduled for February 2027 and described Terra’s power status. [UPDATE: If the President’s proposed FY26 budget goes into effect without modification by Congress, the three Flagship missions will enter accelerated Phase F (closeout); Terra passivation would start in November 2025 and be complete by March 2026.]
Abrams reviewed the status of the Terra spacecraft, showing slides provided by Jason Hendrickson [GSFC]. The Flight Operations Team staffing remains constant. Science data capture for ASTER remains above 99%. The impact of the shunt failure on November 28, 2024 required the safe halting of the instrument. Visible-near-infrared (VNIR) observations resumed in mid-January, and thermal infrared (TIR) observations resumed in mid-May. Collision avoidance events continue to be part of normal operations.
Hitomi Inada [JSS] provided a status report on the ASTER instrument. Many of the monitored components (i.e., VNIR pointing motor) are beyond their original useful life in orbit, but the aging hardware shows no signs of wearing out or a decrease in performance. She showed data that indicated that the temperature and current telemetry trends remain stable.
Abrams presented ASTER product distribution statistics provided by Cole Krehbiel [Land Processes Distributed Active Archiver Center (LP DAAC]). The ASTER Digital Elevation Model continues to be the most ordered product among all users of ASTER data. As defined by the ST at the last meeting, most ASTER data products [e.g., Version 4 (V4) products] are being created and placed in a searchable/orderable archive that can be accessed through NASA’s Earthdata tool. Abrams reported that the LP DAAC started producing these files in January 2025 and will be finished before August 2026.
Koki Iwao [GSJ] presented AIST’s product distribution statistics. Over 4.7 million scenes have been acquired and processed to Level 1A (L1A) since June 10, 2025. AIST continues to distribute ASTER’s pseudo-natural color scenes in keyhole mark-up language (KML – a file format used to display geographic data) and scene-based Digital Elevation Models. The largest number of users of Japanese products are from the United States.
Tetsushi Tachikawa [JSS] summarized the status of ASTER observations since the beginning of the mission. He reported that all of the global observation programs are functioning normally, acquiring data as planned. Updates to the observation programs will be considered by this week’s working groups. Tachikawa also added that the change of the orbit repeat – after Terra’s October 2022 exit from the Morning Constellation – has been accommodated in the ASTER scheduler.
Abrams presented a report on behalf of Simon Hook [JPL], who was unable to attend the meeting. Hook’s information provides a status update for the multispectral TIR instrument on NASA’s ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) mission. Abrams also spoke about NASA’s future Surface Biology and Geology (SBG) mission, which is part of the planned Earth System Observatory.
Applications Working Group
The applications session provided a sampling of how ASTER data are used. A few examples are highlighted below. The second half of the session was devoted to a discussion of end-of-mission documentation requirements. This included a review of the NASA guiding document and sharing of existing documents.
Michael Ramsey [UPitt] presented work on forecasting volcanic activity with the ASTER long-term archive. His team developed a statistical detection code to extract accurate temperature anomalies for five test volcanoes over 25 years. They used these results to train a deep learning approach for anomaly detection in future TIR data. The method showed 73% success for Piton del la Fournaise volcano, Réunion island, and near 100% success for Sheveluch volcano, Kamchatka Krai, Russia.
Miyuki Muto [IU] reported on waste volume changes in 15 open landfills in developing countries using more than 20 years of ASTER time-series digital surface models – see Figure 1. The method was found to be consistent with reports using synthetic aperture radar (SAR) data, which dates to 2016. Thus, ASTER can provide a longer time series for future optical or radar studies.
Figure 1. Time variation in the relative volume of waste for landfills, obtained from ASTER time-series digital surface model data for the four Indian sites – Ghazipur, Bhalswa, Okhla, and Deonar. Figure credit: Figure taken from Muto and Tonooka (2025), Sensors Mike Abrams presented the 25-year history of ASTER data applied to geologic mapping and mineral exploration. He explained how the first published papers appeared a few years after launch and validated the unique mineralogical information contained in the ASTER data. Over the following 20 years, several reports from mineral exploration companies announced the discovery of gold, chromite, and lithium deposits, which were found largely based on analysis of ASTER data.
Calibration/Validation Working Group
The Calibration/Validation (cal/val) working group is responsible for monitoring the radiometric and geometric performance of ASTER’s VNIR and TIR instruments. Three different cal/val techniques are used including: analysis of onboard calibration lamps, comparison with onboard blackbodies, and measurements of pseudo-invariant ground targets during field campaigns. The L2 software algorithms are being updated for the final, archival processing which is anticipated to be completed in May 2026.
Bjorn Eng [JPL] reported that the newest version of the L2 algorithm for ASTER VNIR and TIR cal/val was delivered to the LPDAAC for ingest and testing. Eng explained how the new software includes Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) data, which allows users to create atmospheric profiles for temperature, pressure, water vapor, and ozone. MERRA-2 is an improvement – both spatially and temporally – over the National Centers for Environmental Prediction’s Global Data Assimilation System that is used in the original MERRA. The new L2 production algorithms were validated, and the LP DAAC began incorporating the algorithms into the static archive in January 2025.
Mike Abrams presented on behalf of Cole Krehbiel [LP DAAC] and reported on the assessment of geometric performance of the L1 processing software, which was updated to the new Landsat ground control point library. He also presented an improved global digital elevation model. The ASTER final processing campaign uses the improved control point library.
Satoru Yamamoto [GSJ] presented updates to the calibration trends of the onboard VNIR lamps. Two onboard calibrations were performed on September 20, 2024 and November 8, 2024. Several analyses of the calibration lamps showed no significant change in the data trends – see Figure 2. The signal-to-noise ratios are still greater than the requirement of 140.
Figure 2. Onboard lamp calibration data for Bands 1, 2, and 3. The lamp data show no significant change in the three bands after updating the calibration. Figure credit: Satoru Yamamoto Soushi Kato [RESTEC] presented results from his September 2024 field campaign in Nevada and Utah. The campaign was marked by clear weather during ASTER’s day and night overpasses. Kato compared his in situ TIR measurements with the standard ASTER temperature products from the LP DAAC. The agreement for the five AESTER TIR bands was within ± 1.5 K.
Hideyuki Tonooka [IU] presented the results of his TIR field calibration campaigns at the same time and location as those conducted by Kato (described in previous presentation summary). Additionally, he reported that several calibration campaigns conducted at Lake Kasumigaura, Japan were cancelled due to cloudy weather, which led to the suspension of ASTER data acquisition. Tonooka compared his in situ TIR measurements with the standard ASTER temperature products from the LP DAAC. The agreement for the five ASTER TIR bands was within ± 1.3 K, except for band 10 at the Utah site where the discrepancy was -2.3 K.
Temperature–Emissivity Working Group
This group focuses on ASTER’s kinetic temperature and emissivity products, as well as application of these products and review of the nighttime TIR global mapping program status.
Mike Abrams presented his analyses of the ASTER Level-2 Surface Kinetic Temperature Product (AST_08) for a nighttime scene acquired over Lake Tahoe, CA. He compared the on-demand MERRA-2 product from NASA’s Global Modeling and Assimilation Office with the archive-produced product. The comparison showed that the two products were identical, pixel-by-pixel. Abrams conducted a second analysis to compare the archived MERRA_2 AST_08 product with the on-demand Moderate Resolution Imaging Spectroradiometer (MODIS) AST-08 product to assess the difference in temperature due to improved MERRA-2 atmospheric parameters. The MERRA-2 product had lower temperature values for higher elevations and higher values for lower elevations with more column water vapor – see Figure 3. This result is physically correct and validates the improvement using MERRA-2 atmospheric data.
Figure 3. Colorized difference by temperature, in Kelvin, between the product using MERRA-2 and MODIS atmospheric values: blue -1.0 to -0.6; green -0.5 to -0.1; red 0.0; and yellow 0.1 to 0.5. Figure credit: Michael Abrams Hideyuki Tonooka discussed the status of installation of the JPL radiometer at Lake Kasumigaura. The plan is to mount the radiometer on an existing observation in the middle of the lake. The radiometer will be operated jointly by JPL and IU. The installation is planned to start in the Summer 2025.
Tetsuchi Tachikawa reviewed the status of the current Thermal Global Mapping acquisition program to acquire cloud-free TIR nighttime images over most of the Earth’s land surface. He explained that the program is refreshed every year, with most recent refresh beginning May 2025.
Operations and Mission Planning Working Group
The Operations and Mission Planning Working Group oversees and reviews the acquisition programs executed by the ASTER scheduler. Because ASTER data acquisitions have to be scheduled every day to accommodate ASTER’s average 8% duty cycle, ST members developed an automatic program to select 600–700 daily scenes from the possible 3000 plus images uploaded in the request archive.
Tachikawa reviewed the status of acquisition scheduling. Urgent observations receive the highest priority and can be scheduled close to acquisition time. Approximately 70 scenes are programmed per month – with over 95% acquisition success. By contrast, global mapping data acquisitions receive the lowest priority and are used to fill in the scenes for the daily quota. He explained that the goal of the ASTER is to have the instrument acquire at least one cloud-free image for every place on Earth. Due to persistent cloud cover, success is typically ~85% after several years, at which time the program is restarted. Tachikawa next gave short updates on three other acquisition programs that focus on islands, volcanoes, glaciers, and cloudy areas, respectively. The global volcano image acquisition program will continue with no change to the observation parameters. Acquisition of images of islands and over cloudy areas will also continue in current form. The global glacier acquisition program will be modified to change the VNIR gain settings to optimize images over snow and ice.
Tachikawa also discussed the effect of the ASTER shutdown in November 2024 and cessation of all ASTER data acquisitions. VNIR-only acquisitions were resumed in January 2025, and TIR acquisitions resumed in May 2025, with full operations and acquisitions of data from both VNIR and TIR instruments.
Closing Plenary Session
Each chairperson summarized the presentations, discussions, and recommendations that occurred during their respective working group session. The overall consensus maintained that the ASTER instrument is operating normally again – with no indications of any component failures. The ST is preparing to absorb the impact of the 50% budget reduction on the Flight Operation Team at GSFC. At this time, the main impact has been a small increase in lost data (1–2%) as a result of the absence of operators to attempt immediate recovery. The ST also approved plans for ASTER’s contribution to the Terra power mitigation plan, and the recommendation has been forwarded to the Terra Project Scientist and the Flight Operations Team.
Conclusion
The 54th ASTER ST Meeting successfully covered all critical issues introduced during the Opening Plenary Session. The ST worked on formulating priorities for reduction of ASTER instrument operations in response to possible future Terra power reductions. During working group sessions, participants received updates on a variety of topics (e.g., instrument scheduling, instrument performance, archiving plans, and new applications). Although this may be the last Joint U.S./Japan ASTER ST Meeting, the 55th joint meeting was tentatively scheduled for May 2026.
Acknowledgments
The lead author’s work on this article was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with NASA.
Michael Abrams
NASA/Jet Propulsion Laboratory/California Institute of Technology
mjabrams@jpl.nasa.gov
Yasushi Yamaguchi
Nagoya University/Japan Science and Technology Agency
yasushi@nagoya-u.jp
Share
Details
Last Updated Aug 18, 2025 Related Terms
Earth Science View the full article
-
By NASA
The 33rd SpaceX commercial resupply services mission for NASA, scheduled to liftoff from the agency’s Kennedy Space Center in Florida in late August, is heading to the International Space Station with an important investigation for the future of bone health.
The experiment will test how microgravity affects bone-forming and bone-degrading cells and explore potential ways to prevent bone loss. This research could help protect astronauts on future long-duration missions to the Moon and Mars, while also advancing treatments for millions of people on Earth who suffer from osteoporosis.
Mesenchymal stem cells (MSCs) are derived from human bone marrow and stained with rapid red dye NASA Space’s Hidden Health Mystery
During long-duration missions, astronauts may experience a gradual reduction in bone density—typically around 1% to 2% per month—even with consistent exercise routines. While scientists understand how bones work on Earth, they aren’t sure exactly why bones weaken so quickly in microgravity.
Previous research aboard the space station revealed that microgravity changes how stem cells behave and what substances they release. Scientists now want to dig deeper into these cellular changes to better understand what causes bone loss in space and explore potential ways to prevent it.
Blocking a Potential Bone Thief
The Microgravity Associated Bone Loss-B (MABL-B) investigation focuses on special stem cells called mesenchymal stem cells, or MSCs. As these cells mature, they build new bone tissue in the body.
Scientists suspect that a protein called IL-6 might be the culprit behind bone problems in space. Data from the earlier MABL-A mission suggests that microgravity promotes the type of IL-6 signaling that enhances bone degradation. The MABL-B experiment will investigate this by testing ways to block this IL-6 signaling pathway.
The experiment will grow mesenchymal stem cells alongside other bone cells in special containers designed for space research. Cells will be cultured for 19 days aboard the space station, with crew members periodically collecting samples for analysis back on Earth.
How this benefits space exploration
The research could lead to targeted treatments that protect astronauts from bone loss during long-duration missions to the Moon, Mars, and beyond. As crews venture farther from Earth, bone health becomes increasingly critical since medical evacuation or emergency return to Earth won’t be possible during Mars missions.
How this benefits humanity
The findings could provide new insights into age-related bone loss that affects millions of people on Earth. Understanding how the IL-6 protein affects bone health may lead to new treatments for osteoporosis and other bone conditions that come with aging.
Related Resources
Microgravity Associated Bone Loss-B (MABL-B) Microgravity Associated Bone Loss-A (MABL-A) Microgravity Expanded Stem Cells About BPS
NASA’s Biological and Physical Sciences Division pioneers scientific discovery and enables exploration by using space environments to conduct investigations not possible on Earth. Studying biological and physical phenomenon under extreme conditions allows researchers to advance the fundamental scientific knowledge required to go farther and stay longer in space, while also benefitting life on Earth.
View the full article
-
By NASA
5 Min Read NASA, Army National Guard Partner on Flight Training for Moon Landing
By Corinne Beckinger
When Artemis astronauts land on the Moon’s South Pole in a commercial human landing system, they will encounter a landscape pockmarked with deep craters, sloped connecting ridges, and harsh lighting conditions. The Moon’s lack of contrast, combined with its rolling terrain, will also pose a challenge, making it difficult for astronauts to overcome visual illusions on the lunar surface.
NASA astronaut Bob Hines (left) and Colorado Army National Guard HAATS instructor Ethan Jacobs practice landing procedures in the Rocky Mountains of Colorado in April 2025. Depending on the season, the snowy or dusty conditions can cause visual obstruction. Lunar dust can cause similar visual impairment during future crewed missions. In the mountains of northern Colorado, NASA and the U.S. Army National Guard are using military helicopters to develop a foundational lunar landersimulated flight training course to help astronauts practice flight and landing procedures for the Moon.
For decades, military helicopter pilots have trained at the HAATS (High-Altitude Army National Guard Aviation Training Site) in Gypsum, Colorado. In 2021, NASA and the Colorado Army National Guard began working together to develop a course specifically for the next generation of lunar explorers.
That NASA-specific course is scheduled to be finalized in August 2025, marking an important milestone for Artemis crewed landings training efforts.
“NASA is using a three-pronged approach with motion-based simulation, in-flight lunar landing analog training, and in-flight lunar simulation to build out its foundational training for Artemis Moon landings,” said NASA astronaut Doug Wheelock, who helped coordinate the training program. “Helicopters at or above 10,000 feet are not really efficient in the thin air, forcing us into operating with very thin power margins similar to the Apollo astronauts having to manage energy and momentum to land safely. The operations along with the terrain at the HAATS course in Colorado’s Rocky Mountains provide a valuable, real-world opportunity for Artemis astronauts to practice flying and landing in conditions similar to maneuvering a lander in the lunar environment.”
NASA astronaut Raja Chari participates in the HAATS course in April 2025. Since 2021, 22 NASA astronauts and one ESA (European Space Agency) astronaut have participated and evaluated the course based on functionality and Artemis mission needs. NASA/Laura Kiker NASA astronaut Raja Chari participates in the HAATS course in April 2025. Since 2021, 22 NASA astronauts and one ESA (European Space Agency) astronaut have participated and evaluated the course based on functionality and Artemis mission needs. NASA/Corinne Beckinger NASA’s human landing systems that will safely transport astronauts to and from the Moon’s surface will be provided by SpaceX and Blue Origin.
NASA’s Artemis III mission will build on earlier test flights and add new capabilities, including SpaceX’s Starship Human Landing System and advanced spacesuits, to send the first astronauts to explore the lunar South Pole and prepare humanity to go to Mars.
While each industry provider is responsible for training Artemis astronauts on its specific lander, NASA is establishing foundational training to help prepare astronauts for crewed flights.
Flight training opportunities like this are vital to mission success and crew safety.”
Doug Wheelock
NASA Astronaut
“Over the last few years, NASA and the Army National Guard have worked closely to evaluate training procedures and landing zone areas, incorporating accounts from Apollo astronauts,” Wheelock said. “During training flights at HAATS, astronauts can experience the visual illusions, cross-cockpit communication, and degraded visibility they may experience navigating to their landing zone near the lunar south pole. Flight training opportunities like this are vital to mission success and crew safety.”
Paired with trained instructors from the Army National Guard, astronauts fly to mountaintops and valleys in a range of aircraft, including LUH-72 Lakotas, CH-47 Chinooks, and UH-60 Black Hawks.
While one astronaut pilots the aircraft, an astronaut in the back charts the landing area, marking key landmarks, identifying potential hazards, and helping to track the flight path. Throughout the week-long course, the landing zones and situations become more challenging, allowing astronauts to experience team dynamics and practice communication skills they will need to land on the Moon.
“Our full-time Colorado Army National Guard pilots have thousands of flight hours navigating the Rocky Mountains at altitudes ranging from 6,500 to 14,200 feet, and we are reaching new heights by providing realistic and relevant training with NASA for Artemis,” said first sergeant Joshua Smith of the HAATS program. “Our Colorado Army National Guard pilots may not fly around the Moon, but we wear our motto, de monitbus ad astra — from the mountains to the stars — with pride.”
Fast Facts
On the Moon’s South Pole, the Sun is never more than 1.5 degrees above or below the horizon. With the Sun at such a low angle and with only a thin exosphere, shadows are stark, and astronauts may find it difficult to determine distances and heights.
The Moon’s atmosphere is extremely thin, with few particles, and is called an exosphere. The Moon’s exosphere is thin enough to glow in sunlight, which has been observed by spacecraft and some of the Apollo astronauts. The Moon’s surface is challenging to land on. There are inactive volcanoes, bounders, large basins, craters, and cracks in the Moon’s crust, caused by the Earth’s gravity tugging on the Moon. Moon dust can also obscure the view from the windows of a commercial human landing system, and affect sensors that relay important information, such as altitude and velocity, to astronauts. Through the Artemis campaign, NASA will send astronauts to explore the Moon for scientific discovery, economic benefits, and to build the foundation for the first crewed missions to Mars – for the benefit of all.
For more information about Artemis visit:
https://www.nasa.gov/artemis
Share
Details
Last Updated Aug 18, 2025 EditorBeth RidgewayContactCorinne M. Beckingercorinne.m.beckinger@nasa.govLocationMarshall Space Flight Center Related Terms
Human Lander Challenge General Human Landing System Program Marshall Space Flight Center Explore More
3 min read Human Rating and NASA-STD-3001
Article 3 days ago 3 min read NASA Seeks Proposals for 2026 Human Exploration Rover Challenge
Article 3 days ago 4 min read NASA IXPE’s ‘Heartbeat Black Hole’ Measurements Challenge Current Theories
Article 6 days ago Keep Exploring Discover More Topics From NASA
Artemis
Human Landing System
Earth’s Moon
The Moon makes Earth more livable, sets the rhythm of ocean tides, and keeps a record of our solar system’s…
Artemis III
View the full article
-
-
Check out these Videos
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.