Google says the new AI allows almost instant weather forecasts & # 39; that can help keep accurate forecasts amid rapid climate change
- A new weather model allows almost instant forecasts
- Google says & # 39; nowcasting & # 39; can cast a prediction in minutes instead of hours
- The model uses machine learning and AI and showed early success.
- Traditional models still outperform long-term forecasts
Google is unleashing the power of its AI and machine learning algorithms behind the development of faster and more accurate weather forecasts.
On a blog postGoogle describes a new model developed by the company called "nowcasting", which claims to have initial success in accurately predicting weather patterns with almost instantaneous results.
According to a new article, the method is capable of producing predictions up to six hours in advance in just five to 10 minutes – numbers that, he says, surpass traditional models, even in the early stages.
While some traditional forecasts generate large amounts of data, they can also take hours to complete.
"A significant advantage of machine learning is that inference is computationally cheap, given an already trained model, allowing for almost instantaneous, high-resolution predictions native to input data," writes Google.
& # 39; If it takes 6 hours to calculate a forecast, it will only allow 3-4 runs per day and will result in data-based forecasts over 6 hours, which limits our knowledge of what is currently happening. On the other hand, nowcasting is especially useful for immediate decisions, from traffic routing and logistics to evacuation planning. & # 39;
When Google compared their accuracy to three current traditional models using historical weather data between 2017 and 2019, the researchers found that performance was equally good, if not better.
Google says its model can help solve weather forecasting problems caused by climate change – more erratic patterns
The ability to make quick and fast forecasts can be especially helpful as the global climate changes, making climate patterns more erratic and sometimes more extreme.
While speed is among the key benefits of the Google model, the company says broadcasting is now still being surpassed by traditional forecasting methods when it comes to long-term forecasting.
As a result, Google's new system is unlikely to completely replace traditional forecasting models, but will be used to complement them and fill gaps in short-term coverage.
As noted by The Verge, Google is not alone in developing new and advanced ways to predict the weather. Both IBM and Monsanto are working on their own methods.
Share or comment on this article: