🔗 Share this article

When Tropical Storm Melissa was churning south of Haiti, meteorologist Philippe Papin had confidence it would soon escalate to a major tropical system.

Serving as primary meteorologist on duty, he predicted that in a single day the storm would intensify into a severe hurricane and begin a turn in the direction of the coast of Jamaica. Not a single expert had ever issued such a bold forecast for rapid strengthening.

But, Papin possessed a secret advantage: artificial intelligence in the form of the tech giant’s recently introduced DeepMind cyclone prediction system – released for the first time in June. And, as predicted, Melissa evolved into a storm of remarkable power that ravaged Jamaica.

Growing Dependence on AI Predictions

Forecasters are increasingly leaning hard on Google DeepMind. On the morning of 25 October, Papin explained in his official briefing that the AI tool was a primary reason for his certainty: “Roughly 40/50 Google DeepMind ensemble members show Melissa reaching a most intense hurricane. While I am unprepared to forecast that strength at this time due to track uncertainty, that remains a possibility.

“There is a high probability that a period of rapid intensification will occur as the storm moves slowly over very warm sea temperatures which is the highest marine thermal energy in the entire Atlantic basin.”

Surpassing Traditional Models

The AI model is the first AI model dedicated to hurricanes, and currently the first to outperform traditional weather forecasters at their specialty. Through all tropical systems this season, the AI is the best – even beating experts on track predictions.

Melissa ultimately struck in Jamaica at maximum strength, one of the strongest landfalls ever documented in almost 200 years of data collection across the Atlantic basin. The confident prediction probably provided people in Jamaica additional preparation time to prepare for the catastrophe, possibly saving people and assets.

The Way The Model Functions

The AI system works by spotting patterns that traditional 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,” said Michael Lowry, a former forecaster.

“What this hurricane season has demonstrated in short order is that the recent artificial intelligence systems are on par with and, in certain instances, more accurate than the slower physics-based forecasting tools we’ve relied upon,” he added.

Clarifying Machine Learning

It’s important to note, Google DeepMind is an example of AI training – a method that has been employed in data-heavy sciences like meteorology for a long time – and is distinct from creative artificial intelligence like ChatGPT.

Machine learning processes mounds of data and extracts trends from them in a such a way that its system only requires minutes to come up with an result, and can operate on a standard PC – in sharp difference to the flagship models that governments have used for decades that can require many hours to process and require some of the biggest high-performance systems in the world.

Professional Reactions and Upcoming Advances

Still, the fact that the AI could outperform previous gold-standard legacy models so rapidly is nothing short of amazing to meteorologists who have spent their careers trying to predict the world’s strongest weather systems.

“I’m impressed,” said James Franklin, a retired expert. “The sample is now large enough that it’s evident this is not a case of chance.”

He said that while Google DeepMind is beating all competing systems on forecasting the trajectory of storms worldwide this year, similar to other systems it occasionally gets extreme strength predictions wrong. It had difficulty with Hurricane Erin earlier this year, as it was also undergoing rapid intensification to category 5 north of the Caribbean.

During the next break, Franklin said he intends to talk with Google about how it can make the DeepMind output more useful for forecasters by offering extra internal information they can use to evaluate exactly why it is producing its conclusions.

“The one thing that troubles me is that while these predictions seem to be highly accurate, the output of the model is essentially a opaque process,” said Franklin.

Broader Sector Trends

There has never been a commercial entity that has produced a high-performance forecasting system which allows researchers a view of its methods – unlike nearly all systems which are provided at no cost to the general audience in their full form by the authorities that created and operate them.

Google is not alone in adopting AI to address challenging meteorological problems. The authorities also have their respective artificial intelligence systems in the development phase – which have demonstrated improved skill over previous non-AI versions.

Future developments in artificial intelligence predictions seem to be startup companies tackling previously difficult problems such as sub-seasonal outlooks and better advance warnings of tornado outbreaks and flash flooding – and they are receiving federal support to pursue this. One company, WindBorne Systems, is even deploying its proprietary atmospheric sensors to address deficiencies in the US weather-observing network.