Buildings and Floods Project

Our built environment is at risk of flooding and erosion due to climate change and the continued expansion of our urban footprint. To manage this risk, decision makers currently rely on models and databases that were assembled using coarse data about the built infrastructure and natural landscape. New technologies and analysis techniques increase the resolution of data but introduce a new set of problems related to data management and the extraction of useful information from massive volumes of raw data that must be solved before models can be made more robust. The focus of the proposed partnership is the micro- or property-scale characterization of building attributes and drainage details to support a novel flood modeling strategy and risk assessment. We will use innovative data gathering technologies and analysis techniques to vastly improve the quality and utility of available information, and then link the information to databases and risk assessment tools that are already being developed by our partners. Data gathering technologies will include street level photography and aerial lidar scans to obtain complete surveys of urban environments. Analysis techniques will include image analysis to extract information about buildings and urban drainage systems and digital elevation model analysis to assess micro-scale flood risk. The results will feed into the Linkable Open Data Environment (LODE) developed by Statistics Canada, the Risk and Return on Investment Tool (RROIT) developed by Credit Valley Conservation, and the CanFlood risk modelling toolbox developed by Natural Resources Canada. Long-term benefits to Canada will include new robust methods to enable high-resolution open database of our cities, more precise modelling of urban floods, and better management and mitigation of flood risk.

Four objectives
Four objectives for the project include: 1) image and lidar data aquisition using unmanned aerial vehicles and street view; 2) analysis with computer vision and machine learning to extract relevant parameters; 3) automation of property-scale flood model setup; and 4) micro-scale flood risk analysis.




Chul Min Yeum, Assistant Professor, Department of Civil and Environmental Engineering, University of Waterloo (Computer Vision for Smart Structures Lab)

Jennifer Drake, Associate Professor, Department of Civil and Environmental Engineering, Carleton University (Smart, Healthy, and Sustainable Communities and Environments Lab)

Derek Robinson

Derek Robinson, Associate Professor, Department of Geography, University of Waterloo (Modelling and Spatial Analysis Lab)


Sriram Narasimhan, Professor, Department of Civil Engineering, University of California in Los Angeles (Sensing & Robotics for Infrastructure Lab)

Buildings and Floods Student Advertisement274 KB