Source: NIWA
NIWA is using machine learning to forecast flood inundation in a fraction of the time required to run physical models.
NIWA Climate, Atmosphere & Hazards platform manager Nava Fedaeff leads the project – she says effective flood preparation and response requires detail beyond river flows.
“What people really want to know is not just whether the river is running high, but what areas will be flooded, and what’s at risk from that potential flooding. We’re exploring how AI will help us to move from weather forecasts to inundation forecasts quickly enough so that useful information gets to those who need it,” said Fedaeff.
Predicting flood maps with physical models can take 24 hours but with machine learning it takes only 1-2 minutes.
Five days ahead of an event, scientists combine several elements such as weather forecasting, river flow predictions, inundation mapping and exposure assessments. This enables them to produce models that detail – down to street level – people, property or infrastructure at risk when storms strike.
NIWA data scientist Dr Deidre Cleland used Westport as a case study in the project.
She has produced a StoryMap detailing how the system works – with maps, animations and graphics – outlining how her team validated the AI flood model against the real-life 2021 Westport flooding.
“Our next step is operationalising this machine learning capability so that rapid flood map forecasting is available for a real incoming flood event in Westport. We are also working on extending the machine learning approach to other locations around New Zealand, starting with those at highest risk of flooding,” said Dr Cleland.
Floods are New Zealand’s most frequent and costly natural disaster, meaning that fast and accurate forecasting of flood impacts is crucial for reducing the risk to life, property and infrastructure.
