Geospatial data is complex and often requires some level of technical expertise to navigate. As a result, these data tend to be accessible only to a limited number of researchers and scientists. Tailor-made that NASA As more data is collected from new satellites, these complexities only grow and may further limit the potential pool of people capable of extracting information and developing applications that could benefit society. If we add to that that In total we are talking about about 100 petabytes of information (equivalent to 50 billion pages of text, one on top of the other), the challenge is enormous.
Thus, navigating the more than 100 petabytes of data collected requires a strategy. This is what has led to NASA and Microsoft to explore the use of a personalized copilot using Azure OpenAI Service to develop Earth Copilot, an AI that could transform the way people interact with data on Earth.
NASA’s Earth Science Data Systems Program is responsible for collecting an incredible variety of data from space sensors and instruments. This data ranges from atmospheric conditions to land cover changes, ocean temperatures, and more. However, The scale and complexity of this information can be overwhelming. AI could streamline this process, reducing the time needed to obtain insights from Earth data to a matter of seconds.
Earth Copilot will be able to answer questions about our planet by condensing NASA’s wide range of geospatial information into easy-to-digest answers. For example, “What was the impact of DANA in Valencia?” or “How did the COVID-19 pandemic affect air quality in Spain?” With this, NASA aims to “democratize” access to scientific data.
“For many, finding and extracting information requires navigating technical interfaces, understanding data formats, and mastering the complexities of geospatial analysis: specialized skills that very few users possess – explains Tyler Bryson, one of those responsible for the development of Earth Copilot -. AI could speed up this process, reducing the time to obtain information from Earth data to a matter of seconds”.
But this AI will also have a huge impact on science. For example, scientists who need to analyze historical hurricane data to improve predictive models, or policy makers who want study deforestation patterns to implement environmental regulationsthey may find themselves unable to easily access the data they need.
Furthermore, as they continue to be released new satellites with new instruments and collecting more dataNASA is constantly faced with the challenge of building new tools to manage and make sense of this growing “library.”
The key to NASA’s Earth Copilot lies in the integration of cloud-based technologies such as Azure OpenAI Service, which provides access to powerful AI models and natural language processing capabilities that allow developers to integrate intelligent and conversational AI into their applications.
By combining these technologies, Earth Copilot allows users to interact with NASA archives through queries using simple language. For now, Earth Copilot It is only available to NASA scientists and researcherswho will evaluate the capabilities of the tool. They will then explore their integration into the data visualization, exploration and analysis platform (VEDA) from NASA, which already offers access to some of the agency’s data.