Developing advanced urban-resolving tools for urban climate research

Urbanization represents one of the most significant anthropogenic changes to the Earth’s surface, with profound climatic effects across scales through modifying the surface biophysical properties and hence perturbing the Earth’s surface energy balance. An additional 2.5 billion people projected to reside in urban areas by year 2050, nearly doubling the world’s current urban population in just three decades. This inevitable urbanization coupled with climate change will not only expose cities and their residents to substantial risks across the world, but also presents a historic and time-sensitive opportunity to mitigate and adapt to the negative impacts of future changes and to advance global sustainable and resilient growth. Addressing this grand challenge, however, requires advanced data and tools that better represent and/or resolve urban effects and their complex two-way interactions with climate across spatiotemporal scales, both for improved scientific understanding of cities and for planning effective resilient strategies. Recent advances in AI/ML, satellite remote sensing, high-resolution urban-resolving Earth system modeling, and advanced computing have enabled development of many of these advanced tools and opened up promising opportunities. In this talk, I will present some examples, using our recent work, on how AI/ML, hybrid modeling, and advance computing could help empower urban climate research as well as the associated challenges.