HydroNetNZ: An AI-driven mobile app for impact-based flood forecasts and warnings at household level in New Zealand

Flood warnings today tell people what the weather will be. What households actually need to know is whether water will reach their property, and how deep it will get.

​

Answering that takes high-resolution maps of flood extent and depth, derived from river flow, sea level, and local topography. Conventional inundation models can produce those maps, but they are too computationally heavy to run inside the window where a warning still changes anyone's decision.

​

HydroNetNZ closes that gap in three steps. First, AI models generate high-resolution flood extent and depth maps in a fraction of the time conventional modelling requires, fast enough to run during an event. Second, an Impact-Based Forecasting and Warning (IBFW) platform and mobile app translate those maps into localised warnings, delivered at household level rather than catchment or region. Third, residents submit crowdsourced flood reports through the app, giving ground-truth observations where instrumentation doesn't reach.

​

Those reports feed back into the models, improving accuracy with every event and strengthening community resilience.