Members Can Post Anonymously On This Site
AI-powered satellite data reveals clouds in 3D
-
Similar Topics
-
By NASA
At Goddard Space Flight Center, the GSFC Data Science Group has completed the testing for their SatVision Top-of-Atmosphere (TOA) Foundation Model, a geospatial foundation model for coarse-resolution all-sky remote sensing imagery. The team, comprised of Mark Carroll, Caleb Spradlin, Jordan Caraballo-Vega, Jian Li, Jie Gong, and Paul Montesano, has now released their model for wide application in science investigations.
Foundation models can transform the landscape of remote sensing (RS) data analysis by enabling the pre-training of large computer-vision models on vast amounts of remote sensing data. These models can be fine-tuned with small amounts of labeled training and applied to various mapping and monitoring applications. Because most existing foundation models are trained solely on cloud-free satellite imagery, they are limited to applications of land surface or require atmospheric corrections. SatVision-TOA is trained on all-sky conditions which enables applications involving atmospheric variables (e.g., cloud or aerosol).
SatVision TOA is a 3 billion parameter model trained on 100 million images from Moderate Resolution Imaging Spectroradiometer (MODIS). This is, to our knowledge, the largest foundation model trained solely on satellite remote sensing imagery. By including “all-sky” conditions during pre-training, the team incorporated a range of cloud conditions often excluded in traditional modeling. This enables 3D cloud reconstruction and cloud modeling in support of Earth and climate science, offering significant enhancement for large-scale earth observation workflows.
With an adaptable and scalable model design, SatVision-TOA can unify diverse Earth observation datasets and reduce dependency on task-specific models. SatVision-TOA leverages one of the largest public datasets to capture global contexts and robust features. The model could have broad applications for investigating spectrometer data, including MODIS, VIIRS, and GOES-ABI. The team believes this will enable transformative advancements in atmospheric science, cloud structure analysis, and Earth system modeling.
The model architecture and model weights are available on GitHub and Hugging Face, respectively. For more information, including a detailed user guide, see the associated white paper: SatVision-TOA: A Geospatial Foundation Model for Coarse-Resolution All-Sky Remote Sensing Imagery.
Examples of image reconstruction by SatVision-TOA. Left: MOD021KM v6.1 cropped image chip using MODIS bands [1, 3, 2]. Middle: The same images with randomly applied 8×8 mask patches, masking 60% of the original image. Right: The reconstructed images produced by the model, along with their respective Structural Similarity Index Measure (SSIM) scores. These examples illustrate the model’s ability to preserve structural detail and reconstruct heterogeneous features, such as cloud textures and land-cover transitions, with high fidelity.NASAView the full article
-
By NASA
NASA/Joel Kowsky On Dec. 4, 2024, NASA astronauts Loral O’Hara, left, and Jasmin Moghbeli spent a moment in part of the Earth Information Center, an immersive experience combining live NASA data sets with innovative data visualization and storytelling at NASA Headquarters in Washington.
O’Hara and Moghbeli spent six months in space as part of Expedition 70 aboard the International Space Station. On Nov. 1, 2023, they performed a spacewalk together that lasted 6 hours and 42 minutes.
Image credit: NASA/Joel Kowsky
View the full article
-
By European Space Agency
On 1 December 2024, BepiColombo flew past Mercury for the fifth time. During this flyby, BepiColombo became the first spacecraft ever to observe Mercury in mid-infrared light. The new images reveal variations in temperature and composition across the planet's cratered surface.
View the full article
-
By NASA
iss071e650763 (Sept. 14, 2024) — The long exposure photograph taken by NASA astronaut Matthew Dominick shows star trails, streaks of city lights, and two Roscosmos crew ships, the Soyuz MS-26 docked to the Rassvet module (foreground) and the Soyuz MS-25 (background) docked to the Prichal docking module, as the International Space Station orbited 265 miles above central China.NASA Space Station trajectory data is now available to the public!
This data, called an ephemeris, is generated by the ISS Trajectory Operations and Planning Officer (TOPO) flight controllers in the Mission Control Center at NASA’s Johnson Space Center. TOPO keeps track of where the ISS is, where it is going to be, and most importantly makes sure it isn’t at risk of colliding with other objects in space. At ISS’s altitude, a very thin atmosphere is still present. This thin atmosphere creates drag and over time can cause TOPO’s predicted ISS trajectory to accumulate error. Because of this, TOPO updates the predicted trajectory approximately three times a week, so the ISS Flight Control Team has the best trajectory estimate possible. An accurate trajectory is essential for maintaining communications links, planning visiting vehicle rendezvous, and ensuring ISS’s path is clear of any potential collisions.
The links above and below are to the most current posted ephemeris. The ephemeris is in the CCSDS Orbital Ephemeris Message (OEM) standard and is available in .txt and .xml file formats. Each file contains header lines with the ISS mass in kg, drag area in m2, and drag coefficient used in generating the ephemeris. The header also contains lines with details for the first and last ascending nodes within the ephemeris span. Following this is a listing of upcoming ISS translation maneuvers, called “reboosts,” and visiting vehicle launches, arrivals, and departures.
After the header, ISS state vectors in the Mean of J2000 (J2K) reference frame are listed at four-minute intervals spanning a total length of 15 days. During reboosts (translation maneuvers), the state vectors are reported in two-second intervals. Each state vector lists the time in UTC; position X, Y, and Z in km; and velocity X, Y, and Z in km/s.
Orbit Ephemeris Message (OEM)
https://nasa-public-data.s3.amazonaws.com/iss-coords/current/ISS_OEM/ISS.OEM_J2K_EPH.txt https://nasa-public-data.s3.amazonaws.com/iss-coords/current/ISS_OEM/ISS.OEM_J2K_EPH.xml Users of this data should monitor this page for information regarding any future changes to the file format. Past data postings can be found archived on data.nasa.gov by searching “ISS COORDS.”
NOTE: NASA is providing this information for use by the general public. The OEM data format is supported natively by many commercial spaceflight software applications. Please consult your application’s support documentation for specific details on how to deploy this data.
View the full article
-
By European Space Agency
A pair of spacecraft were launched together today from India with the potential to change the nature of future space missions. ESA’s twin Proba-3 platforms will perform precise formation flying down to a single millimetre, as if they were one single giant spacecraft. To demonstrate their degree of control, the pair will produce artificial solar eclipses in orbit, giving prolonged views of the Sun’s ghostly surrounding atmosphere, the corona.
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.