Hurricane Milton: Forecast improves using satellites and computing – 10/10/2024 – Science

Over the decades, it has become possible, thanks to images taken from satellites and meteorological stations, to determine the formation and development of hurricanes. Likewise, it has long been possible to predict how powerful they will be and where they will go. But what we’ve seen over the last 15 years is a huge increase in accuracy.

“The structure of forecasting extreme systems such as hurricanes has remained similar over the years,” says Marco Aurelio Franco, professor at IAG-USP (Institute of Astronomy, Geophysics and Atmospheric Sciences of the University of São Paulo). “But we have achieved very significant improvements in two main areas: in the computational power to accommodate, model and simulate extreme conditions; and in the significant increase in the number of meteorological stations, as well as satellites, that allow good coverage of the entire globe.”

Remember the basics: The appearance of hurricanes is the product of the interaction between air currents and oceans, which form around an area of ​​low air pressure, with closed circulation of winds, resulting in large storms. They follow the patterns of air currents, and these so-called tropical cyclones form over the sea, feed on it, and carry warm, moist air to transport it to the upper atmosphere.

As it advances toward land, it loses its strength, because it no longer has the ocean to feed it. However, depending on the force with which it reaches the continent, it can cause significant damage, as is happening now with Hurricane Milton in Florida.

Meteorologists can monitor and predict the movement and intensity of the phenomenon through numerical computer simulations, which, based on real data obtained from satellites and surface stations and on basic principles of physics, extrapolate what the coming hours and days, even storms, will be like. dissipate.

This process is fundamentally chaotic – it involves so many variables in nonlinear phenomena (influencing each other) that it often needs to be simplified in simulations. However, with the advancement of computing, these exercises are becoming better and better.

“Computers have gotten faster and better, but not only that,” Franco says. “The models have also gotten more complete (and complex).” “Previously, many components of these models were based on relatively rudimentary hypotheses about the behavior of the atmosphere and the interaction between the atmosphere and the biosphere/ocean. Today, these components represent much better at physical, spatial and temporal resolution the complex interactions that occur in the atmosphere.”

The USP researcher also highlights the importance of significant investments (particularly by rich countries) in high-quality, accurate satellites. “It has become easier to monitor the Earth’s climate in real time,” he says, noting that despite this, there is still a lack of surface weather stations, which are more accurate in characterizing the atmosphere than satellites.

Changing phenomenon

Another important aspect of understanding and predicting hurricanes is their connection to climate change. It is clear at this point that global warming, caused primarily by the burning of fossil fuels, is not only increasing the frequency of these major storms, but also increasing their intensity.

“In this case of Milton, in particular, it is related to a very large anomaly in the ocean surface temperature [aliás, não só da superfície, mas em camadas mais profundas também]“This anomalous warming is definitely due to climate change,” he says. “See, the heating is tied to the energy that was available at that location, which was used by the hurricane, in this case Milton, which caused it to change its classification very quickly.”

The importance of alerts

The main use of predictive models, of course, is to allow preventive action to be taken, with the possible evacuation of affected residents. But it is always necessary to take uncertainties into account, even with improved simulations. “Chaotic systems are difficult to model. Today we have a good understanding of them and how they occur in the atmosphere, but prediction is always probabilistic, meaning there is always a chance it will happen or not.”

As a result, it is natural that there will be some reluctance on the part of the authorities to alert the population, “precisely because of all the logistics involved, which can lead even to panic and more complex problems.”

Franco highlights that in the case of Milton, major US agencies issued numerous warnings, as soon as the hurricane’s rapid and abnormal growth was observed. Within a few hours, it went from Category 2 to Category 5. “Record-breaking! It was really impressive but difficult to model. It is worth noting that changes in climate due to climate change are still a challenge for models.”

These uncertainties, and the very nature of predictions, based on simulations that explore all possible outcomes, averaged out, with the greatest probability of occurring, explain the differences between what is expected and what actually happens. In Milton’s case, the hurricane reached the coast around 8:40 p.m. (Florida local time), which was slightly different than forecast. Its meeting point with Earth has also diverged, being further south than initially expected.