IBM and NASA’s Marshall Space Flight Center today announced a collaboration to use IBM artificial intelligence (AI) technology to discover new insights from NASA’s vast wealth of Earth and geospatial science data. The joint work will apply foundational AI modeling technology to data from NASA’s Earth-observing satellites for the first time.
The foundational models They are types of AI models that are trained on a large set of unlabeled data, which can be used for different tasks and allow information about one situation to be applied to another. These models have advanced rapidly in the field of natural language processing (NLP) technology over the past five years, and IBM is pioneering applications of foundational models beyond language.
Earth observations that allow scientists to study and monitor our planet are being collected at an unprecedented rate and volume.. New and innovative approaches are required to extract insights from these vast data resources. The aim of this work is to provide researchers an easier way to analyze and gain insights from these large data sets. IBM’s foundational modeling technology has the potential to accelerate the discovery and analysis of this data for rapidly advance scientific understanding of Earth and the response to climate-related problems.
IBM and NASA plan to develop several new technologies to extract information from Earth observations. One of the projects will train a foundational IBM geospatial intelligence model with NASA’s Harmonized Landsat-Sentinel-2 (HLS) data set, a record of Earth’s surface and land-use changes captured by Earth-orbiting satellites. By analyzing petabytes of satellite data to identify changes in the geographic footprint of phenomena such as natural disasters, cyclical crop yields, and wildlife habitats, this foundational modeling technology will help researchers provide critical analyzes of our planet’s environmental systems.
Another result of this collaboration is expected to be an easily searchable corpus of Earth science literature. IBM developed a natural language processing model trained with nearly 300,000 Earth Science journal articles to organize the literature and facilitate the discovery of new knowledge. With one of the largest AI workloads trained on Red Hat OpenShift software to date, the model uses PrimeQA, an open source multilingual question-and-answer system from IBM. Beyond providing a resource for researchers, the new language model for Earth Sciences could be incorporated into NASA science data management and administration processes.
“The beauty of foundational models is that they can potentially be used for many downstream applications.”explained Rahul Ramachandrana senior research scientist at NASA’s Marshall Space Flight Center in Huntsville, Alabama, United States. “The development of these foundational models cannot be tackled by small teamshe added.Teams from different organizations are needed to bring their different perspectives, resources and skills to bear“.
“The foundational models have proven their worth in natural language processing, and now is the time to expand them into new domains and modalities important to business and society.“, it states Raghu Ganti, IBM Principal Investigator. “The application of foundational models to geospatial data, sequences of events, time series, and other non-linguistic factors within Earth Science data will make valuable knowledge and information available to a much broader group of researchers. companies and citizens. Ultimately, it could make it easier for more people to work on some of our most pressing climate issues.“.
5695_.jpg)
Under the agreement, other possible projects between IBM and NASA include the creation of a foundational model for weather and climate prediction using MERRA2, a dataset of atmospheric observations. This collaboration is part of the NASA’s Open Source Science Initiative, a commitment to build an open scientific community, inclusive, transparent and collaborative over the next decade.
