If enough data is available, Flood Hub provides an animation of water depth compared to a human body - for example, showing water that’s ankle-deep at a particular location will be waist height by 11am tomorrow. It can also make longer-term flood frequency projections, providing information about the danger of a flood in the next two to five years.
A spokesman for Civil Defence’s National Emergency Management Agency said, “While we are aware that there are a number of private suppliers of weather information, it would not be appropriate for Nema to comment on the merits of any single provider. However, we note that MetService is the Government’s official provider of severe weather warnings, and Civil Defence Groups issue local flood warnings, and we encourage the public to stay informed through these agencies.”
Regardless, Google isn’t positioning Flood Hub as a replacement for any existing emergency alerts, or the likes of Niwa’s Flood Frequency tool.
A Google disclaimer says Flood Hub predictions “are still a work in progress and are for informational purposes only. You should not use them as a sole source of data in the event of an emergency, but as potentially one data point in addition to other sources of data - e.g. the Government, weather services etc”.
Flood Hub requires you to turn on Google Location Services.
A feature for real-time alerts to your Android mobile, or to an iPhone with Google’s Search or Maps app installed, won’t initially be available in NZ. Google says it’s aiming to make the alert data available to central and local government first, then to individuals down the track.
Google pitches Flood Hub as a tool to help countries deal with the immediate effects of climate change.
The system was first introduced for flood-prone areas of India and Bangladesh in 2018, then to 18 more countries last year. With this week’s expansion, Flood Hub is now in 80 countries.
Google said in a post: “In 2021, we sent 115 million flood alert notifications to 23 million people over Search and Maps, helping save countless lives.”
“During our rollout period in Myanmar, our forecasts were integrated by the WFP [UN World Food Programme] into their ADAM [Advanced Disaster Analysis and Mapping] alerting system. And During the Mocha Cyclone, which killed over 400 people [in India and Bangadesh], our Flood Hub forecasts were used in this report they issued, alongside other sources,” Google crisis response lead Yossi Matias said in comments emailed to the Herald.
The system now covers some 460m people worldwide.
NZ faces more frequent, more intense natural disasters
“Since early 2023, New Zealanders have faced extreme weather events unlike anything we’ve seen before. Historic flooding and cyclones have ripped through communities, and while natural disasters are common here - causing an average of $790m in damages each year - they’re unfortunately increasing in frequency and intensity due to climate change, threatening people’s safety and livelihoods,” Matias said.
Flooding affects more than 250m people globally each year, causing more than US$10 billion in damage, Matias said.
Flood Hub’s AI is based on diverse, publicly available data sources, such as weather forecasts and satellite imagery, he said.
The technology then combines two models: the Hydrologic Model, which forecasts the amount of water flowing in a river, and the Inundation Model, which predicts what areas are going to be affected and how deep the water will be, Matias said.
Separately, Google said it had given $200,000 for NZ cyclone relief, made up of funds from the company and donations from staff.
In 2021, the company launched a quake alert system in NZ, which collates data from smartphones running Google’s Android operating system - it utilises the accelerometer built into most modern phones, which is capable of detecting seismic waves. If waves are detected, an alert is to other Android users in nearby areas, giving them a warning that a quake could be on the way. The system caused a false alarm for a quake off Hawkes Bay in August 2021, which in the final event was too deep and far offshore to be felt, but the system was praised for its accuracy after a large quake struck near Taranaki in January last year.
