IBM-built AI geospatial model earns NASA award for tackling critical climate challenges

Rahul Ramachandran of NASA IMPACT, left, Elizabeth Fancher of NASA IMPACT, Ankur Kumar of the University of Alabama in Huntsville (UAH), Sujit Roy of UAH, Raghu Ganti of IBM Research, David McKenzie of NASA, Muthukumaran Ramasubramanian of UAH, Iksha Gurung of UAH, and Manil Maskey of NASA IMPACT, right, accept the NASA Marshall Space Flight Center Group Achievement Award on Thursday, August 15, 2024 at NASA Marshall. (Image credit: NASA)

The team behind the AI geospatial foundation model, which includes members from IBM, NASA, and several other esteemed institutions, has been honoured with the NASA Marshall Space Flight Center Group Achievement Award.

They won the award for their contribution towards addressing critical climate challenges and providing a framework for environmental research, covering areas such as flood and wildfire detection and urban heat island analysis.

Launched in 2023 and released as an open-source application this year, the IBM-developed AI geospatial foundation model, based on NASA's Earth science satellite data, has proven its efficacy by reducing heat island effects by more than 3°C (5.4°F) in the United Arab Emirates and aiding reforestation efforts in Kenya. Additionally, it assisted in identifying urban heat stress and improving climate resilience strategies in the UK.

What is AI geospatial foundation model?

The AI geospatial foundation model combines extensive data from satellite imagery and weather information provided by NASA to produce actionable environmental insights. This foundational model is designed to address specific tasks such as flood detection, burn scar identification, and land use classification. It serves as a base for AI-driven environmental research, enhancing the precision and efficiency of weather forecasting and climate analysis.