Projects

EOLab's research activities revolve around harnessing the power of geospatial data and remote sensing for the greater good, with a focus on land surface-atmosphere interactions and, to a lesser extent, coastal monitoring. To this end, we lead several competitive research projects to develop, validate, and employ multiple products, with a strong emphasis on the use of AI, while maintaining a rigorous understanding of the underlying processes. Our research spans the acquisition and processing of diverse geospatial and other complementary datasets, including satellite imagery from multiple agencies and providers (primarily Copernicus, NASA, and USGS). In our research endeavors, we leverage the power of cloud computing, with a particular emphasis on platforms like Google Earth Engine, to harness the vast potential of geospatial data for innovative insights and solutions. Our commitment to FAIR and Open Science principles underpins all our research efforts, ensuring transparency and accessibility for all.

Project highlights:

Project FOCUS (Horizon 2020):

Project FOCUS was lead by the eoLab (UC) and was designed to develop AI-driven innovative ways to detect tree decline caused by the pinewood nematode. A new, and successful method was created using a combination of satellite data, hyperspectral sensors onboard planes and drones, and laboratory analysis. A broad network of stakeholders was involved, participating in the co-design of the solution to amplify impact.

Project eANDES (CONCYTEC):

Project eAndes is co-lead by the eoLab in partnership with local partners in Peru. It seeks to develop new, high-quality and locally relevant AI-driven land cover products from Sentinel-1/-2 data and incorporate GPM precipitation data to study change dynamics in Lake Junín and enhance stewardship. A strong capacity-building component is included in the project.

Project MDV (La Caixa Foundation):

The scientific lead of this project by the eoLab enabled the development of accurate and frequently updated land cover maps for Portugal, using Deep Learning methods and Sentinel-2 data.

Check back soon, for more information and updated list of projects.