Author Name: Carribean Channel, Cuba TV.
This story was originally published on canalcaribe.ict.cu. [1]

Artificial Intelligence: How will it influence the weather of the future?
Date: None

The development of radars, satellites and numerical models has transformed – for the better – climate and weather forecasts. Remember that these two words have different meanings. When we talk about climate, we refer to the state of meteorological variables (atmospheric pressure, temperature, relative humidity, wind direction and speed, among others), analyzed over a long period, usually , thirty years. Meanwhile, the weather can change from one day to the next, with the arrival of a cold front, or if phenomena such as thunderstorms and tornadoes occur. In the case of the part that we follow daily through the media, we find valuable information to make daily decisions.

But, beyond the advances, what are the expectations of the meteorologists dedicated to the forecast of the weather about the future of this science? We asked Dr. José Rubiera, a prominent figure in meteorology in Cuba and the rest of the Caribbean: “The prospects we have for the future forecast meteorologists, of course, improve weather forecasts.

That means, both those of a very small scale, of a few kilometers, of a few minutes, as is the case of severe local storms, tornadoes, and those of longer term: hurricanes, for example, improve trajectory forecasts , even more, to improve those of fast change of intensity, that are a pending subject in the whole world ».

And on this path, one of the most fascinating areas of study in Computer Science offers new possibilities: Artificial Intelligence.Machine Learning: what is it and how does it work? Machines with capabilities of the human brain, such as reasoning, creativity and learning: that’s what Artificial Intelligence is all about.

This discipline articulates several branches, among which Machine Learning stands out, in English. We are talking about computers capable of learning thanks to the experience they obtain from the analysis of large amounts of data. Its performance improves as more information is processed, and today this field of study has applications in many scientific and everyday areas. “When we see Facebook and review our messages, that selection of messages is being carried out for us by a Machine Learning algorithm, using what our tastes are and our previous behavior,” explains the MSc. Suilán Estévez, professor of Artificial Intelligence at the Faculty of Mathematics and Computing at the University of Havana.And what have been some of the applications of Machine Learning in meteorology? “First, as a complement to the physical models, which allows creating faster but less precise approximations.

On the other hand, to estimate specific variables such as rainfall or soil temperature,” says Estévez.In fact, the prediction of rainfall, especially for very specific areas, still presents a margin of uncertainty that can be significant. For this reason, the topic focuses on several of the Artificial Intelligence research and projects related to weather forecasting.

In particular, Machine Learning shows enormous potential for the development of meteorology. However, there are still limitations. For example, the volume and quality of the data to be processed, as well as the computing capabilities determine how accurate a result can be.

Surpass human beings?

When talking about Machine Learning, one of the issues up for debate is whether machines will replace human beings.

For Estévez, «it is very important to highlight that Machine Learning is not here to replace experts, since they are necessary when interpreting data, proposing which ones are useful to solve a problem or make decisions. The important thing is that this is a very powerful tool that will be used more and more in the future. “But it is only the meteorologist, the one who with his knowledge and also with his experience, with what he has lived, which allows him to see the future of the forecast, within these models, what will happen over time.

For this reason, the forecasting meteorologist is the one who makes the decision, sometimes very important decisions and the man can never be replaced, because he is the one who decides, ”explains Dr. Rubiera.

Artificial Intelligence applied to meteorology and many other areas: a topical and highly controversial topic. What do you think?

(C) Cuba TV, Cuban state owned media.
[1] Url: https://www.canalcaribe.icrt.cu/en/artificial-intelligence-how-will-it-influence-the-weather-of-the-future/