Artificial Intelligence could be the key to more accurate and more timely hurricane forecasts, China-based researchers report in the scientific journal Nature.
In two journal papers, the researchers, including scientists affiliated with Chinese tech company Huawei, said their AI approach is an improvement over conventional, numerical weather prediction systems, known as NWP. , which is now used by meteorologists around the world.
“Trained on 39 years of global data, our program, Pangu-Weather, obtains stronger deterministic forecast results on reanalysis data of all tested variables compared to the best systems in NWP in the world,” wrote Kaifeng Bi, Lingxi Xie and other researchers.
The NWP has been used for years and has improved, but is still considered slow and subject to overestimations and inaccuracies, the paper says. The AI ​​approach also needs to improve but has “huge potential.” AI will likely be integrated with traditional forecasting methods to produce hybrid models, Lingxi Xie told the Daily Beast news outlet.
The researchers used a three-dimensional model, along with four decades of weather data and “reanalysis” to train neural networks to predict storm behavior and weather patterns, the Nature paper explains.
If proven successful, it is possible that the AI ​​system could help predict the last-minute movements of some storms, which other studies have suggested are on the rise. Meteorologists and officials in Florida were criticized last fall after Hurricane Ian suddenly turned east after it was first predicted to hit the Tampa area.
One of the Nature papers can be found here. Two comments on the paper online questioned some of the study’s methods.
Topics
Disaster Natural Disasters InsurTech Hurricane Data Driven Artificial Intelligence
Is this article valuable?
Here are several articles you may enjoy.
Interested in Ouch?
Get automatic alerts for this topic.
