🔗 Share this article

As Tropical Storm Melissa was churning off the coast of Haiti, weather expert Philippe Papin felt certain it would soon grow into a monster hurricane.

As the primary meteorologist on duty, he forecasted that in a single day the weather system would become a severe hurricane and start shifting towards the Jamaican shoreline. No forecaster had previously made this confident prediction for rapid strengthening.

But, Papin possessed a secret advantage: artificial intelligence in the guise of the tech giant’s new 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 Forecasting

Meteorologists are heavily relying upon Google DeepMind. On the morning of 25 October, Papin clarified in his public discussion that Google’s model was a key factor for his confidence: “Approximately 40/50 AI simulation runs indicate Melissa becoming a most intense storm. While I am unprepared to forecast that strength at this time given track uncertainty, that is still plausible.

“It appears likely that a phase of quick strengthening is expected as the system drifts over very warm sea temperatures which is the most extreme marine thermal energy in the entire Atlantic basin.”

Surpassing Traditional Systems

Google DeepMind is the first AI model focused on tropical cyclones, and now the first to beat standard weather forecasters at their specialty. Through all tropical systems so far this year, the AI is top-performing – even beating human forecasters on track predictions.

Melissa eventually made landfall in Jamaica at maximum intensity, one of the strongest landfalls recorded in nearly two centuries of data collection across the Atlantic basin. The confident prediction likely gave residents extra time to get ready for the disaster, possibly saving people and assets.

How The System Functions

Google’s model operates through identifying trends that conventional time-intensive physics-based prediction systems may overlook.

“The AI performs much more quickly than their physics-based cousins, and the computing power is less expensive and time consuming,” stated Michael Lowry, a former meteorologist.

“What this hurricane season has proven in quick time is that the newcomer AI weather models are competitive with and, in some cases, superior than the slower physics-based forecasting tools we’ve traditionally leaned on,” he added.

Clarifying Machine Learning

To be sure, Google DeepMind is an instance of machine learning – a method that has been used in research fields like meteorology for a long time – and is not creative artificial intelligence like ChatGPT.

AI training processes large datasets and pulls out patterns from them in a manner that its system only takes a few minutes to come up with an result, and can do so on a standard PC – in strong contrast to the flagship models that authorities have utilized for years that can require many hours to run and need some of the biggest supercomputers in the world.

Expert Responses and Future Developments

Nevertheless, the reality that the AI could exceed previous gold-standard legacy models so rapidly is truly remarkable to meteorologists who have dedicated their lives trying to forecast the most intense weather systems.

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

He said that while the AI is beating all other models on predicting the trajectory of storms globally this year, like many AI models it sometimes errs on extreme strength predictions wrong. It had difficulty with another storm previously, as it was also undergoing quick strengthening to maximum intensity north of the Caribbean.

During the next break, he stated he plans to talk with Google about how it can make the DeepMind output even more helpful for forecasters by offering additional internal information they can use to evaluate the reasons it is producing its answers.

“The one thing that troubles me is that although these predictions appear really, really good, the output of the model is kind of a black box,” said Franklin.

Wider Industry Developments

There has never been a commercial entity that has produced a top-level weather model which grants experts a peek into its techniques – in contrast to nearly all systems which are offered at no cost to the general audience in their full form by the governments that designed and maintain them.

The company is not alone in adopting AI to solve difficult weather forecasting problems. The US and European governments also have their respective AI weather models in the works – which have demonstrated improved skill over previous non-AI versions.

Future developments in AI weather forecasts seem to be startup companies tackling formerly tough-to-solve problems such as sub-seasonal outlooks and improved advance warnings of severe weather and sudden deluges – and they are receiving federal support to pursue this. A particular firm, WindBorne Systems, is also deploying its proprietary weather balloons to address deficiencies in the US weather-observing network.