Abstract

We aim to develop a forecasting model for urban floods, by tightly integrating physical modelling with data-driven deep learning. Our primary application scenario are urban pluvial floods that occur when precipitation cannot be fully absorbed by the drainage system, therefore causing flooding and substantial damages, as well as disruption to socio-economic activities. Structural engineering solutions to cope with urban flooding are large and expensive to build and maintain, and it is important to also advance non-structural solutions, and in particular real-time flood prediction. The fast occurrence and relatively short duration of urban floods (“flash floods”) means that physically-based models are of limited use, due to their long run times. In this project, we will develop new Machine Learning (ML) methods to generate flood predictions with sufficient lead time, such that they can be used to alert the population and to plan mitigation and rescue actions. Our core research question in this project is: how can deep learning and physically-based hydraulic models be combined? Our hypothesis is that by exploiting hydraulic modelling knowledge, deep learning need not “ statistically learn physics from scratch”. By tightly integrating the two modelling approaches, we aim to get the best of both worlds: interpretability and adherence to physical constraints; outputs with well-calibrated uncertainty estimates; and the predictive power and speed of neural networks. The input data for the ML flood forecasting model will be rainfall forecasts provided by meteorological services (e.g. from weather radar), images and digital surface models. The ML-based flood model will produce spatially explicit, two-dimensional flood hazard maps with water depth, flood extent and flow velocity information.