Oceans make up two-thirds of Earth’s surface, but remain largely unexplored due to harsh conditions like strong currents, deep pressure, and limited visibility. To tackle this, IBM (IBM) Research, in partnership with the Plymouth Marine Lab (PML), UK STFC Hartree Centre, and the University of Exeter, has developed a breakthrough AI model called Granite-Geospatial-Ocean. Interestingly, this tool uses satellite imagery and real-world ocean data to map out marine ecosystems. The goal is to get a better understanding of ocean health, carbon cycling, and how these processes impact global climate.
Elevate Your Investing Strategy:
Take advantage of TipRanks Premium at 50% off! Unlock powerful investing tools, advanced data, and expert analyst insights to help you invest with confidence.
It’s worth noting that the model was trained on around 500,000 color-coded images from ESA’s Copernicus Sentinel-3 satellite and fine-tuned using limited but high-quality field data. Despite using just 100–200 observations in some tests, Granite-Geospatial-Ocean outperformed classical machine learning models in accurately predicting chlorophyll-a levels. These tiny microorganisms, called phytoplankton, not only support nearly all marine life but also absorb billions of tons of carbon from the atmosphere.
With climate change affecting ocean temperatures and reducing nutrient circulation, researchers are unsure how long phytoplankton can keep absorbing carbon at current levels. Granite-Geospatial-Ocean could help clarify this by providing more accurate global estimates and refining climate models. In addition to estimating phytoplankton levels, the model could be adapted to monitor harmful algae blooms, sediment runoff, and water quality. Researchers say this is a major leap forward in using remote sensing and AI to study the oceans.
Is IBM a Buy, Sell, or Hold?
Turning to Wall Street, analysts have a Moderate Buy consensus rating on IBM stock based on six Buys, six Holds, and one Sell assigned in the past three months, as indicated by the graphic below. Furthermore, the average IBM price target of $287.25 per share implies that shares are trading near fair value.
See more IBM analyst ratings
Disclaimer & DisclosureReport an Issue