ML, Computer Visualization and Physics-Based Modeling for Urban Climate Digital Twin Framework

Machine Learning, Computer Visualization and Physics-Based Modeling for Urban Climate Digital Twin Framework
by Prof. Dev Niyogi,
Jackson School of Geosciences the University of Texas, Austin
April 12, 2023
Sponsored by IEEE GRSS
Cities are impacted by extreme weather and climate events- and cities, in turn, also create their own local climate through urban heating and rainfall changes. Representing cities and the urban interactions within dynamical weather prediction and climate models is challenging. This challenge is not just because of a city’s relatively small spatial footprint relative to the regional landcover, but also the need for integrating the within-city processes within the urban dynamics. So coupling of cities and larger climate is an important societal issue and a fascinating dynamical modeling problem as we are often at the limits of how different equations can be applied without violating the inherent assumptions. This is where the recent growth in computer visualization and machine learning (ML) techniques has been fascinating to help bridge the scales and physics with behavioral or unsensed aspects of the urban systems. This talk will present some ongoing work and highlight upcoming challenges and opportunities under the urban climate digital twin framework that is being continually constructed to study and model weather / climate processes for complex systems and settings like cities, surrounding areas and neighborhoods.
More details at: www.grss-ieee....