Keynote: Artificial intelligence for digital twins in natural and agricultural sciences

Summary

Geo-information has become a social commodity that moves fast into the big data era: Increasingly more data sources are becoming available, from affordable miniature devices, to smart phones, IoT sensors, high-resolution, multi-spectral remote sensing images, and social media. Detecting meaningful signals through vast volumes of environmental data to further progress our understanding of natural processes and socio-environmental system interactions, remains a challenge. This talk offers an outlook of the technical challenges ahead, and some solutions from ongoing work with my team, towards data-driven intelligence with applications in ecosystem services, environmental policy, food security, and crop yield prediction at field and regional scales.