Lights brighten the night sky in this image of Europe, including Poland, taken from the International Space Station. NASA
As extreme weather events increase
around the world due to climate change, the need for further research into our
warming planet has increased as well. For NASA, climate research involves not
only conducting studies of these events, but also empowering outside
researchers to do the same. The artificial intelligence (AI) efforts
spearheaded by the agency offer a powerful tool to accomplish these goals.
In 2023, NASA teamed up with IBM
Research to create an AI geospatial foundation model. Trained on vast amounts
of NASA’s widely used Harmonized Landsat and
Sentinel-2 (HLS) data, the model provides a base for a variety of AI-powered studies
to tackle environmental challenges. In keeping with open science
principles, the model is freely available for anyone to access.
Foundation models serve as a
baseline from which scientists can develop a diverse set of applications,
enabling powerful and efficient solutions. “Foundation models only know what
things are represented in the data,” explained Manil Maskey, the data science
lead at NASA’s Office of the Chief Science Data Officer (OCSDO). “It’s like a
Swiss Army Knife—it can be used for multiple different things.”
Once a foundation model is created,
it can be trained on a small amount of data to perform a specific task. To
date, the Interagency Implementation and Advanced Concept Team (IMPACT) along
with collaborators have demonstrated the geospatial foundation model’s
capabilities by fine-tuning it to detect burn scars, to delineate flood water,
and to classify crop and other land use categories.
Rectangular ponds for shrimp farming line the coast of northern Peru in this image captured on March 14, 2024 by the OLI-2 (Operational Land Imager-2) on Landsat 9. NASA Earth Observatory / Lauren Dauphin
Because of the computational
resources required to create the initial foundation model, a partnership was
necessary for success. In this case, NASA brought the data and scientific
knowledge, while IBM brought the computing power and AI algorithm optimization
expertise. The team’s shared commitment to making their research accessible
through open science principles ensures that their model can be useful to as
many researchers as possible.
“To build a foundation model at
scale, we realized early on that it's not feasible for one institution to build
it,” Maskey said. “Everything we have done on our foundation models has been
open to the public, all the way from pre-training data, code, best practices,
model weights, fine-tuning training data, and publications. There’s
transparency, so researchers can trace why certain things were used in terms of
data or model architecture.”
Following on from the success of
their geospatial foundation model, NASA and IBM Research are continuing their
partnership to create a new, similar model for weather and climate studies.
They are collaborating with Oak Ridge National Laboratory (ORNL), NVIDIA, and
several universities to bring this model to life.
This time, the main dataset will be
the Modern-Era
Retrospective analysis for Research and Applications, Version 2 (MERRA-2), a huge collection of atmospheric reanalysis data
that spans from 1980 to the present day. Like the geospatial foundation model,
the weather and climate model is being developed with an open science approach,
and will be available to the public in the near future.
Covering all aspects of Earth
science would take several foundation models trained on different types of
datasets. However, Maskey believes those future models might someday be
combined into one comprehensive model, leading to a “digital twin” of the Earth
that would provide unparalleled analysis and predictions for all kinds of
climate and environmental events.
Whatever innovations the future
holds, NASA and IBM’s geospatial and climate foundation models will enable
leaps in Earth science like never before. Though powerful AI tools will enhance
researchers’ work, the team’s dedication to open science supercharges the
possibilities for discovery by allowing anyone to put those tools into practice
and pave the way for groundbreaking research to help better care for the
planet.
For more information about open
science at NASA, visit:
https://science.nasa.gov/open-science/
By Lauren Leese
Web Content Strategist for the Office of the Chief Science Data Officer
No comments:
Post a Comment