AI and geospatial data, a duo to serve the planet
A quarter of a million terabytes of satellite data, or 25,000 times all the printed documents in the US Library of Congress – that’s what NASA’s various Earth science missions are expected to generate in 2024. Last August, the US agency announced the creation of a tool, developed in a public-private partnership with IBM Research, combining Artificial Intelligence (AI) and geospatial data. Based on an open-source AI model, the system is designed to facilitate the exploitation of the unprecedented volumes of data being collected. In particular, it will make it possible to track the impact of climate change or land use, monitor natural disasters and predict crop yields.
With the new generation of low-Earth orbit digital satellites, geospatial intelligence is building on an ever-increasing amount of reliable, real-time, high-resolution data. This market is growing rapidly, driven by technological advances. The global Earth observation market is expected to grow from $67.4 billion in 2022 to $119.9 billion in 2027, representing an average annual growth rate of 12.2%.[1] Long the preserve of the defence industry, geospatial data is now accessible to all sectors, enabling us to respond to a wide range of challenges. Using AI to process geospatial data opens up amazing possibilities. American giant Microsoft has designed a solution to increase the Earth’s visibility – a challenge in itself, given that 67% of the planet is covered by opaque clouds. AI analysis of geospatial data could also prove a valuable ally for marine biodiversity and the oceans, which produce over 50% of the planet’s oxygen. Developed by the European Earth observation programme Copernicus, and powered by ESA data, MyOcean Pro maps the oceans to help manage water, coastal areas and marine resources. Using proprietary algorithms, this visualisation tool measures water temperature, levels, oxygen content and acidity.
AI is also being used to process geospatial data for climate purposes. More than half of the 50 indicators identified by the UN for monitoring climate change are currently provided by satellite imagery. Convinced that AI and satellite imagery can transform the management of global warming, Kayrros, a leader in environmental intelligence, and climate and energy analysis using satellite imagery, aims to guide industry and public policy in the ecological transition. It does this by collecting geospatial data in real time and processing it using proprietary algorithms to produce precise measurements. Geospatial data can be used to reliably assess, model and better anticipate risk in all sectors, from finance and insurance to agriculture and firefighting. As firefighters can use geospatial data to identify areas at risk, the French specialist in geo-analysis technology has developed a forest fire risk prevention service for AXA. The Kayrros model also enables real-time monitoring of CO2 emissions and the volume of carbon sequestered by vegetation, and the detection of methane leaks on a global scale. The stakes are high: the warming power of methane is 84 times greater than that of CO2 over 20 years.
The massive adoption of AI by the space sector reflects two major trends that will define the coming decade. The first is the rise of AI, which is spilling over into all sectors and providing practical solutions. Processing the ever-increasing amount of geospatial data would be impossible without automation. This reflects the second trend, the strong innovation dynamic in the space sector, which is adopting the most advanced technologies and opening up to new commercial markets. At La Financière de l’Echiquier (LFDE), we see this synergy as fertile investment ground for innovation and a source of long-term opportunities. In the face of the uncertainty that characterises the anticipation of the future, one thing seems certain: the dizzying progress of these two themes will continue to exceed our most ambitious expectations. Lift-off is imminent!