Flood risk assessment is currently reliant on models and databases that were assembled with relatively coarse data about buildings and the landscape. New data-acquisition technologies offer the potential for a huge leap forward in resolution, but also introduce a new set of problems related to data management and the extraction of information. The focus of this project is the "micro" or property-scale characterization of building attributes and drainage details to support a novel flood modeling and risk assessment strategy. Data gathering technologies will include street level photography and lidar scans to obtain complete surveys of urban environments. Machine learning and other advanced analysis techniques will be used to extract information and parameterize urban watershed models. Models will be calibrated with collected field data and used to run climate and system component failure models to assess flood hazard probability. We have funding through the NSERC Alliance Grant program to work on this program in collaboration with government partners over the next five years.