Members Can Post Anonymously On This Site
Comet Makes a Pit Stop Near Jupiter's Asteroids
-
Similar Topics
-
By NASA
6 min read
Smarter Searching: NASA AI Makes Science Data Easier to Find
Image snapshot taken from NASA Worldview of NASA’s Global Precipitation Measurement (GPM) mission on March 15, 2025 showing heavy rain across the southeastern U.S. with an overlay of the GCMD Keyword Recommender for Earth Science, Atmosphere, Precipitation, Droplet Size. NASA Worldview Imagine shopping for a new pair of running shoes online. If each seller described them differently—one calling them “sneakers,” another “trainers,” and someone else “footwear for exercise”—you’d quickly feel lost in a sea of mismatched terminology. Fortunately, most online stores use standardized categories and filters, so you can click through a simple path: Women’s > Shoes > Running Shoes—and quickly find what you need.
Now, scale that problem to scientific research. Instead of sneakers, think “aerosol optical depth” or “sea surface temperature.” Instead of a handful of retailers, it is thousands of researchers, instruments, and data providers. Without a common language for describing data, finding relevant Earth science datasets would be like trying to locate a needle in a haystack, blindfolded.
That’s why NASA created the Global Change Master Directory (GCMD), a standardized vocabulary that helps scientists tag their datasets in a consistent and searchable way. But as science evolves, so does the challenge of keeping metadata organized and discoverable.
To meet that challenge, NASA’s Office of Data Science and Informatics (ODSI) at the agency’s Marshall Space Flight Center (MSFC) in Huntsville, Alabama, developed the GCMD Keyword Recommender (GKR): a smart tool designed to help data providers and curators assign the right keywords, automatically.
Smarter Tagging, Accelerated Discovery
The upgraded GKR model isn’t just a technical improvement; it’s a leap forward in how we organize and access scientific knowledge. By automatically recommending precise, standardized keywords, the model reduces the burden on human curators while ensuring metadata quality remains high. This makes it easier for researchers, students, and the public to find exactly the datasets they need.
It also sets the stage for broader applications. The techniques used in GKR, like applying focal loss to rare-label classification problems and adapting pre-trained transformers to specialized domains, can benefit fields well beyond Earth science.
Metadata Matchmaker
The newly upgraded GKR model tackles a massive challenge in information science known as extreme multi-label classification. That’s a mouthful, but the concept is straightforward: Instead of predicting just one label, the model must choose many, sometimes dozens, from a set of thousands. Each dataset may need to be tagged with multiple, nuanced descriptors pulled from a controlled vocabulary.
Think of it like trying to identify all the animals in a photograph. If there’s just a dog, it’s easy. But if there’s a dog, a bird, a raccoon hiding behind a bush, and a unicorn that only shows up in 0.1% of your training photos, the task becomes far more difficult. That’s what GKR is up against: tagging complex datasets with precision, even when examples of some keywords are scarce.
And the problem is only growing. The new version of GKR now considers more than 3,200 keywords, up from about 430 in its earlier iteration. That’s a sevenfold increase in vocabulary complexity, and a major leap in what the model needs to learn and predict.
To handle this scale, the GKR team didn’t just add more data; they built a more capable model from the ground up. At the heart of the upgrade is INDUS, an advanced language model trained on a staggering 66 billion words drawn from scientific literature across disciplines—Earth science, biological sciences, astronomy, and more.
NASA ODSI’s GCMD Keyword Recommender AI model automatically tags scientific datasets with the help of INDUS, a large language model trained on NASA scientific publications across the disciplines of astrophysics, biological and physical sciences, Earth science, heliophysics, and planetary science. NASA “We’re at the frontier of cutting-edge artificial intelligence and machine learning for science,” said Sajil Awale, a member of the NASA ODSI AI team at MSFC. “This problem domain is interesting, and challenging, because it’s an extreme classification problem where the model needs to differentiate even very similar keywords/tags based on small variations of context. It’s exciting to see how we have leveraged INDUS to build this GKR model because it is designed and trained for scientific domains. There are opportunities to improve INDUS for future uses.”
This means that the new GKR isn’t just guessing based on word similarities; it understands the context in which keywords appear. It’s the difference between a model knowing that “precipitation” might relate to weather versus recognizing when it means a climate variable in satellite data.
And while the older model was trained on only 2,000 metadata records, the new version had access to a much richer dataset of more than 43,000 records from NASA’s Common Metadata Repository. That increased exposure helps the model make more accurate predictions.
The Common Metadata Repository is the backend behind the following data search and discovery services:
Earthdata Search International Data Network Learning to Love Rare Words
One of the biggest hurdles in a task like this is class imbalance. Some keywords appear frequently; others might show up just a handful of times. Traditional machine learning approaches, like cross-entropy loss, which was used initially to train the model, tend to favor the easy, common labels, and neglect the rare ones.
To solve this, NASA’s team turned to focal loss, a strategy that reduces the model’s attention to obvious examples and shifts focus toward the harder, underrepresented cases.
The result? A model that performs better across the board, especially on the keywords that matter most to specialists searching for niche datasets.
From Metadata to Mission
Ultimately, science depends not only on collecting data, but on making that data usable and discoverable. The updated GKR tool is a quiet but critical part of that mission. By bringing powerful AI to the task of metadata tagging, it helps ensure that the flood of Earth observation data pouring in from satellites and instruments around the globe doesn’t get lost in translation.
In a world awash with data, tools like GKR help researchers find the signal in the noise and turn information into insight.
Beyond powering GKR, the INDUS large language model is also enabling innovation across other NASA SMD projects. For example, INDUS supports the Science Discovery Engine by helping automate metadata curation and improving the relevancy ranking of search results.The diverse applications reflect INDUS’s growing role as a foundational AI capability for SMD.
The INDUS large language model is funded by the Office of the Chief Science Data Officer within NASA’s Science Mission Directorate at NASA Headquarters in Washington. The Office of the Chief Science Data Officer advances scientific discovery through innovative applications and partnerships in data science, advanced analytics, and artificial intelligence.
Share
Details
Last Updated Jul 09, 2025 Related Terms
Science & Research Artificial Intelligence (AI) Explore More
2 min read Polar Tourists Give Positive Reviews to NASA Citizen Science in Antarctica
Article
6 hours ago
2 min read Hubble Observations Give “Missing” Globular Cluster Time to Shine
Article
6 days ago
5 min read How NASA’s SPHEREx Mission Will Share Its All-Sky Map With the World
Article
7 days ago
Keep Exploring Discover Related Topics
Missions
Humans in Space
Climate Change
Solar System
View the full article
-
By European Space Agency
Astronomers have confirmed the discovery of a rare celestial visitor: a comet from beyond our Solar System.
Officially named 3I/ATLAS, this newly identified interstellar object is only the third of its kind ever observed, following the famous 1I/ʻOumuamua in 2017 and 2I/Borisov in 2019.
View the full article
-
By NASA
5 Min Read What Are Asteroids? (Ages 14-18)
What are asteroids?
Asteroids are rocky objects that orbit the Sun just like planets do. In fact, sometimes asteroids are called “minor planets.” These space rocks were left behind after our solar system formed about 4.6 billion years ago.
Asteroids are found in a wide range of sizes. For example, one small asteroid, 2015 TC25, has a diameter of about 6 feet – about the size of a small car – while the asteroid Vesta is nearly 330 miles in diameter, almost as wide as the U.S. state of Arizona. Some asteroids even have enough gravity to have one or two small moons of their own.
There are more than a million known asteroids. Many asteroids are given names. An organization called the International Astronomical Union is responsible for assigning names to objects like asteroids and comets.
This illustration depicts NASA’s Psyche spacecraft as it approaches the asteroid Psyche. Once it arrives in 2029, the spacecraft will orbit the metal-rich asteroid for 26 months while it conducts its science investigation.NASA/JPL-Caltech/ASU What’s the difference between asteroids, meteors, and comets?
Although all of these celestial bodies orbit the Sun, they are not the same. Unlike asteroids, which are rocky, comets are a mix of dust and ice. Meteors are small space rocks that get pulled close enough to enter Earth’s atmosphere, where they either burn up as a shooting star or land on the ground as a meteorite.
What are asteroids made of?
Different types of asteroids are composed of different mixes of materials. Most of them are made of chondrites, which are combinations of materials such as rocks and clay. These are called “C-type” asteroids. Some, called “S-type,” are made of stony materials, while “M-type” asteroids are composed of metallic elements.
NASA’s Dawn spacecraft captured this image of Vesta as it left the giant asteroid’s orbit in 2012. The framing camera was looking down at the north pole, which is in the middle of the image.NASA/JPL-Caltech/UCLA/MPS/DLR/IDA How did the asteroids form?
Asteroids formed around the same time and in the same way as the planets in our solar system. A massive, dense cloud of gas and dust collapsed into a spinning disk, and the gravity in the disk’s center pulled more and more material toward it. Over time, these pieces repeatedly collided with each other, sometimes resulting in smaller fragments and other times clumping together, resulting in much bigger objects.
Objects with a lot of mass – like planets – produced enough gravity to pull themselves into spheres, but many smaller objects didn’t. These ended up becoming comets, small moons, and, yes, asteroids. Although some asteroids have a spherical shape, most have irregular shapes – sometimes oblong, bumpy, or jagged.
The main asteroid belt lies between Mars and Jupiter, and Trojan asteroids both lead and follow Jupiter. Scientists now know that asteroids were the original “building blocks” of the inner planets. Those that remain are airless rocks that failed to adhere to one another to become larger bodies as the solar system was forming 4.6 billion years ago.Credits: NASA, ESA and J. Olmsted (STScI) Where are asteroids found?
Most of the asteroids we know about are located in an area called the main asteroid belt, which is found in the space between Mars and Jupiter. But asteroids are found in other parts of the solar system, too.
Trojan asteroids orbit the Sun on the same orbital path as a planet. They’re found at two specific points on the planetary orbit called Lagrange points. At these points, the gravitational pull of the planet and the Sun are in balance, making these points gravity-neutral and stable. Many planets have been found to have Trojan asteroids, including Earth.
An asteroid’s location can also be influenced by the gravity of planets it passes and end up pushed or pulled onto a path that brings it close to Earth. When asteroids or comets are on an orbital path that comes within 30 million miles of Earth’s orbit, we call them near-Earth objects.
Illustration of NASA’s DART spacecraft and the Italian Space Agency’s (ASI) LICIACube, with images of the asteroids Dimorphos and Didymos obtained by the DART spacecraft.Credit: NASA/Johns Hopkins APL/Joshua Diaz Could an asteroid come close enough to hit Earth?
Yes! Throughout history, asteroids or pieces of asteroids have collided with Earth, our Moon, and the other planets, too. The effects of some of these impacts are still visible. For example, Chicxulub Crater was created 65 million years ago when a massive asteroid struck Mexico’s Yucatan Peninsula. The resulting cloud of dust and gas released into Earth’s atmosphere blocked sunlight, leading to a mass extinction that included the dinosaurs. More recently, in 2013, people in Chelyabinsk, Russia, witnessed an asteroid almost as wide as a tennis court explode in the atmosphere above them. That event produced a powerful shockwave that caused injuries and damaged structures.
This is why NASA’s Planetary Defense Coordination Office keeps a watchful eye on near-Earth objects. The Planetary Defense team relies on telescopes and observatories on Earth and in space to detect and monitor objects like these that could stray too close to our planet.
The agency is working on planetary defense strategies to use if an asteroid is discovered to be heading our way. For example, NASA’s DART (Double Asteroid Redirection Test) mission in 2022 was a first-of-its-kind test: an uncrewed spacecraft with an autonomous targeting system intentionally flew into the asteroid Dimorphos, successfully changing its orbit.
Jason Dworkin, OSIRIS-REx mission project scientist, holds up a vial containing part of the sample from asteroid Bennu in 2023.Credit: NASA/James Tralie How does NASA study asteroids?
NASA detects and tracks asteroids using telescopes on the ground and in space, radar observations, and computer modeling. The agency also has launched several robotic explorers to learn more about asteroids. Some missions study asteroids from above, such as the Psyche mission, launched in 2023 to study the asteroid Psyche beginning in 2029. Other missions have actually made physical contact with asteroids. For example, the DART mission mentioned above impacted an asteroid to change its orbit, and the OSIRIS-REx (Origins, Spectral Interpretation, Resource Identification and Security – Regolith Explorer) spacecraft collected a sample of material from the surface of asteroid Bennu and delivered the sample to Earth in 2023 for scientists to study.
Career Corner
Want a career where you get to study asteroids? Here are some jobs at NASA that do just that:
Astronomer: These scientists observe and study planets, stars, and galaxies. Astronomers make discoveries that help us understand how the universe works and how it is changing. This job requires a strong educational background in science, math, and computer science. Geologist: Asteroids are made of different types of rock, clay, or metallic materials. Geologists study the properties and composition of these materials to learn about the processes that have shaped Earth and other celestial bodies, like planets, moons, and asteroids. More About Asteroids
Asteroid Facts
Gallery: What’s That Space Rock?
Center for Near Earth Object Studies
Planetary Defense at NASA
Asteroid Watch: Keeping an Eye on Near-Earth Objects
View the full article
-
By NASA
3 min read
Preparations for Next Moonwalk Simulations Underway (and Underwater)
The Jet Propulsion Laboratory perfected aerogel for the Stardust mission. Under Stardust, bricks of aerogel covered panels on a spacecraft that flew behind a comet, with the microporous material “soft catching” any particles that might strike it and preserving them for return to Earth.NASA Consisting of 99% air, aerogel is the world’s lightest solid. This unique material has found purpose in several forms — from NASA missions to high fashion.
Driven by the desire to create a 3D cloud, Greek artist, Ioannis Michaloudis, learned to use aerogel as an artistic medium. His journey spanning more than 25 years took him to the Massachusetts Institute of Technology (MIT) in Cambridge; Shivaji University in Maharashtra, India, and NASA’s Jet Propulsion Laboratory in Southern California.
A researcher at MIT introduced Michaloudis to aerogel after hearing of his cloud-making ambition, and he was immediately intrigued. Aerogel is made by combining a polymer with a solvent to create a gel and flash-drying it under pressure, leaving a solid filled with microscopic pores.
Scientists at JPL chose aerogel in the mid-1990s to enable the Stardust mission, with the idea that a porous surface could capture particles while flying on a probe behind a comet. Aerogel worked in lab tests, but it was difficult to manufacture consistently and needed to be made space-worthy. NASA JPL hired materials scientist Steve Jones to develop a flight-ready aerogel, and he eventually got funding for an aerogel lab.
The aerogel AirSwipe bag Michaloudis created for Coperni’s 2024 fall collection debut appears almost luminous in its model’s hand. The bag immediately captured the world’s attention.Coperni
The Stardust mission succeeded, and when Michaloudis heard of it, he reached out to JPL, where Jones invited him to the lab. Now retired, Jones recalled, “I went through the primer on aerogel with him, the different kinds you could make and their different properties.” The size of Jones’ reactor, enabling it to make large objects, impressed Michaloudis. With tips on how to safely operate a large reactor, he outfitted his own lab with one.
In India, Michaloudis learned recipes for aerogels that can be molded into large objects and don’t crack or shrink during drying. His continued work with aerogels has created an extensive art portfolio.
Michaloudis has had more than a dozen solo exhibitions. All his artwork involves aerogel, drawing attention with its unusual qualities. An ethereal, translucent blue, it casts an orange shadow and can withstand molten metals.
In 2020, Michaloudis created a quartz-encapsulated aerogel pendant for the centerpiece of that year’s collection from French jewelry house Boucheron. Michaloudis also captured the fashion and design world’s attention with a handbag made of aerogel, unveiled at Coperni’s 2024 fall collection debut.
NASA was a crucial step along the way. “I am what I am, and we made what we made thanks to the Stardust project,” said Michaloudis.
Read More Share
Details
Last Updated Jun 09, 2025 Related Terms
Technology Transfer & Spinoffs Spinoffs Technology Transfer Explore More
2 min read NASA Tech Gives Treadmill Users a ‘Boost’
Creators of the original antigravity treadmill continue to advance technology with new company.
Article 2 weeks ago 3 min read Winners Announced in NASA’s 2025 Gateways to Blue Skies Competition
Article 3 weeks ago 3 min read Meet Four NASA Inventors Improving Life on Earth and Beyond
Article 1 month ago Keep Exploring Discover Related Topics
Missions
Technology Transfer & Spinoffs
Stardust
NASA’s Stardust was the first spacecraft to bring samples from a comet to Earth, and the first NASA mission to…
Solar System
View the full article
-
By NASA
NASA’s James Webb Space Telescope recently imaged the Sombrero Galaxy with its NIRCam (Near-Infrared Camera), which shows dust from the galaxy’s outer ring blocking stellar light from stars within the galaxy. In the central region of the galaxy, the roughly 2,000 globular clusters, or collections of hundreds of thousands of old stars held together by gravity, glow in the near-infrared. The Sombrero Galaxy is around 30 million light-years from Earth in the constellation Virgo. From Earth, we see this galaxy nearly “edge-on,” or from the side.NASA, ESA, CSA, STScI After capturing an image of the iconic Sombrero galaxy at mid-infrared wavelengths in late 2024, NASA’s James Webb Space Telescope has now followed up with an observation in the near-infrared. In the newest image, released on June 3, 2025, the Sombrero galaxy’s tightly packed group of stars at the galaxy’s center is illuminated while the dust in the outer edges of the disk blocks some stellar light. Studying galaxies like the Sombrero at different wavelengths, including the near-infrared and mid-infrared with Webb, as well as the visible with NASA’s Hubble Space Telescope, helps astronomers understand how this complex system of stars, dust, and gas formed and evolved, along with the interplay of that material.
Learn more about the Sombrero galaxy and what this new view can tell us.
Image credit: NASA, ESA, CSA, STScI
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.