Satellite sensors can now provide an amazing level of detail of the Earth surface, yet with sparse and imperfect surface sensors to validate them, and due to their relatively short record (a few decades) their usefulness when used on their own is somewhat limited. To make new leaps in understanding environmental change and to improve prediction we must find intelligent ways to combine satellite data with surface sensors and the output from physics-based environmental simulators (e.g., climate models). To bridge these spatial scales and various modalities we are creating a team of scientists and engineers to build and deploy open-source toolkits and examples driven by real-world case studies.
This project aims to lay the foundations for ambitious research programmes to tackle our greatest environmental changes. We will develop and deploy reproducible and interpretable methods to increase scientific understanding, build tools to help environmental measurement planning, and provide the underpinning tools for intelligent real-time monitoring. The project aims to strengthen collaborations across the environmental and machine learning/AI communities by enhancing open-source and user-friendly computing platforms, including Pangeo.
The project will help researchers around the globe to access and make more efficient use out of the vast volumes of environmental data collected by national research facilities, often in somewhat ad hoc and sometimes unplanned ways. This work will support broader scientific communities where sensors are periodically upgraded (e.g., earth observation satellites), where there are sparse sensor networks (e.g., Antarctica, high mountainous regions), or where placement is not by design (e.g., ship-based sensors for ocean monitoring). The research will inform future planning of sensor development, placement and implementation.