Forecasting flood hazards in real-time: A surrogate model for hydrometeorological events in an Andean watershed

Growing urban development, combined with the influence of El Niño and climate change, have increased the threat of large unprecedented floods induced by extreme precipitation in populated areas near mountain regions of South America. High-fidelity numerical models with physically-based formulations can now predict inundations with a substantial level of detail for these regions, incorporating the complex morphology, and copying with insufficient data and the uncertainty posed by the variability of sediment concentrations. These simulations, however, might have large computational costs, especially if many scenarios need to be evaluated to develop early-warning systems and trigger preemptive evacuations. In this investigation we develop a surrogate model or meta-model to provide a rapid response flood prediction to extreme hydrometeorological events. We characterize the storms with a small set of parameters and use a high-fidelity model to create a database of flood propagation under different conditions. We perform an interpolation and regression procedure by using kriging on the space of parameters that characterize the events, approximating efficiently the flow depths in the urban area. This is the first application of its kind in the Andes region, which can be used to improve the prediction of flood hazards in real conditions, employing low computational resources. It also constitutes a new framework to develop early warning systems to help decision makers, managers, and city planners in mountain regions.