🔗 Share this article

As Tropical Storm Melissa swirled south of Haiti, weather expert Philippe Papin felt certain it was about to escalate to a major tropical system.

As the lead forecaster on duty, he forecasted that in just 24 hours the weather system would intensify into a category 4 hurricane and start shifting towards the Jamaican shoreline. Not a single expert had previously made such a bold prediction for quick intensification.

However, Papin had an ace up his sleeve: AI technology in the form of Google’s new DeepMind hurricane model – released for the initial occasion in June. And, as predicted, Melissa evolved into a storm of astonishing strength that ravaged Jamaica.

Growing Reliance on AI Predictions

Forecasters are heavily relying upon the AI system. On the morning of 25 October, Papin explained in his public discussion that the AI tool was a key factor for his certainty: “Approximately 40/50 Google DeepMind ensemble members indicate Melissa reaching a Category 5 storm. While I am not ready to predict that intensity yet given track uncertainty, that remains a possibility.

“There is a high probability that a period of quick strengthening will occur as the storm drifts over exceptionally hot ocean waters which is the most extreme marine thermal energy in the entire Atlantic basin.”

Surpassing Conventional Systems

Google DeepMind is the pioneer AI model focused on tropical cyclones, and currently the first to beat standard weather forecasters at their own game. Through all tropical systems this season, Google’s model is top-performing – surpassing human forecasters on path forecasts.

The hurricane eventually made landfall in Jamaica at maximum strength, among the most powerful coastal impacts recorded in almost 200 years of record-keeping across the region. The confident prediction likely gave residents additional preparation time to get ready for the catastrophe, potentially preserving lives and property.

The Way Google’s Model Functions

The AI system works by identifying trends that conventional time-intensive scientific weather models may overlook.

“They do it far faster than their physics-based cousins, and the computing power is more affordable and time consuming,” stated Michael Lowry, a ex meteorologist.

“What this hurricane season has demonstrated in short order is that the newcomer AI weather models are competitive with and, in some cases, more accurate than the slower traditional weather models we’ve relied upon,” he added.

Clarifying AI Technology

To be sure, the system is an example of AI training – a method that has been employed in research fields like meteorology for a long time – and is not creative artificial intelligence like ChatGPT.

AI training takes large datasets and extracts trends from them in a manner that its model only requires minutes to generate an answer, and can operate on a desktop computer – in sharp difference to the primary systems that authorities have used for decades that can take hours to run and require some of the biggest high-performance systems in the world.

Professional Reactions and Upcoming Developments

Still, the reality that Google’s model could outperform previous top-tier traditional systems so rapidly is truly remarkable to meteorologists who have spent their careers trying to predict the most intense storms.

“I’m impressed,” said James Franklin, a former forecaster. “The data is sufficient that it’s pretty clear this is not just chance.”

Franklin said that although the AI is beating all other models on forecasting the future path of hurricanes globally this year, similar to other systems it sometimes errs on high-end intensity forecasts inaccurate. It struggled with another storm previously, as it was also undergoing rapid intensification to category 5 above the Caribbean.

During the next break, Franklin said he plans to talk with the company about how it can make the DeepMind output even more helpful for forecasters by providing additional internal information they can use to evaluate the reasons it is coming up with its answers.

“A key concern that troubles me is that while these forecasts seem to be really, really good, the results of the system is kind of a black box,” said Franklin.

Broader Sector Trends

Historically, no a commercial entity that has produced a top-level weather model which grants experts a view of its methods – unlike most other models which are offered free to the public in their full form by the authorities that created and operate them.

The company is not the only one in starting to use artificial intelligence to address difficult weather forecasting problems. The US and European governments are developing their respective artificial intelligence systems in the development phase – which have demonstrated better performance over earlier non-AI versions.

The next steps in artificial intelligence predictions seem to be startup companies tackling formerly difficult problems such as sub-seasonal outlooks and improved early alerts of tornado outbreaks and flash flooding – and they are receiving federal support to pursue this. One company, WindBorne Systems, is even launching its proprietary weather balloons to fill the gaps in the US weather-observing network.