3-5 Evaggelikis Scholis, 14231 Nea Ionia, Greece

NASA’s new AI map tool can forecast exactly how climate change will transform your street

Originally posted on sciencefocus.

A new tool could help you make important life decisions in the face of climate change.

Did the increase in extreme weather events across the globe in 2023 spike your eco-anxiety? If not, a new artificial intelligence (AI) tool that can predict how climate change could impact your house will.

Revealed at COP28, the AI tool, developed by NASA and IBM, will help us monitor Earth from space – measuring environmental changes that have already happened while also making more accurate predictions about the future. While not designed specifically to scare you, the system could help you escape deadly weather events in future.

It’s a bit like Google Earth – except that you can toggle different algorithms to reveal overlays such as tree cover, carbon emissions, and flooding and wildfire risk. Plus, the developers are making it available to everyone when it becomes available in 2024. This includes nations, corporations, charities – and you.

In theory, you could even use it to plan where you should and shouldn’t travel – or where you should buy a house, according to Dr Juan Bernabe-Moreno, director of IBM Research Europe for Ireland and UK.

“There are many ideas about what you can do – the use of the application is really up to the people,” Bernabe-Moreno told BBC Science Focus. “But instead of having to be a big tech [corporation] to create this application, making it open-source means putting it in the hands of the community.”

So how does it work? The tool is an AI-powered foundational model: a model that can map out complex systems using raw data. IBM built it using NASA’s expertise and, crucially, their massive datasets, such as information from satellites.

While you won’t need any fancy tech like, say, having your own data centre to use it, the AI tool may not be able to run on your trusty old laptop.

The foundational model doesn’t require much but, according to Bernabe-Moreno, is likely to need a “few” GPUs: the part of your computer that enables graphics and visuals. (High-performance gaming laptops often have two GPUs.)

In the future, applying this kind of generative AI to the weather could lead to more accurate forecasts – as well as predictions about hurricanes, droughts, and other extreme events. This could help us understand exactly how changes in the climate – such as ice loss at the poles – could impact our lives.

Crucially, these real projections may help authorities to put emergency plans in place.

Being open-source, the tool will promote accountability as communities can hold governments to the promises they make at events like COP28. In fact, the Government of Kenya has been using an existing version of the model to track the progress of its reforestation program.

This is part of what Bernabe-Moreno calls the “democratisation of weather and climate”. He said: “We’ve put the ability to model climate and weather in the hands of the community, and that’s brilliant.”

Source: sciencefocus

Related Posts