Artificial intelligence comes to the aid of meteorologists

Weather forecasting, now done with the help of giant supercomputers and long-term simulations, will soon take a new turn. Google recently announced that it is using machine learning to predict the weather. This is one of the projects of the artificial intelligence research laboratory DeepMind of this company. Google further added that with the help of this artificial intelligence tool, it is going to provide a 10-day weather forecast program in less than a minute with unprecedented accuracy.

Artificial intelligence in competition with traditional models

Steven Eastbrook, professor of computer science and director of the University of Toronto’s School of the Environment, said in an interview: “In the past year, tremendous progress has been made.” “The AI ​​for weather forecasting works the same way it does in other apps like ChatGPT,” Eastbrook explained. He also added: “Google’s GraphCast program uses the weather data of the last four decades in high volume and by analyzing the weather trend in the past, it goes towards its prediction. This is almost similar to what ChatGPT does. This chatbot does not know exactly what it is saying; But it predicts which word is likely to be placed after the previous words according to the literature used. GraphCast emulates the same approach. In fact, it accurately imitates supercomputers and will provide the results to users more quickly. According to Eastbrook’s analysis, this project can be completed very quickly. He noted that the potential for artificial intelligence is so high that it can detect major weather systems earlier than supercomputers. For example, in a pilot study, GraphCast was able to predict Hurricane Lee in Nova Scotia nine days in advance, which was six days ahead of traditional models.

This artificial intelligence tool can learn all the patterns used in the traditional weather forecasting system and use them with high accuracy. According to Eastbrook’s predictions, soon traditional systems will also benefit from this artificial intelligence tool and this will create a huge change in the progress of this field.

Global News meteorologist Anthony Farnell responded by saying that he was excited about the potential of AI in weather forecasting and had heard that the model could beat conventional numerical models. However, he is not worried about replacing meteorologists with a robot; Because, according to him, someone is always needed to explain to the people what is going to happen.

Conclusion

Artificial intelligence has made great progress in the field of weather forecasting and soon it is going to help meteorologists. This artificial intelligence tool uses machine learning technique and by examining and analyzing meteorological data for the last four decades, it provides a detailed forecast of its condition, which is more accurate than traditional models. This tool also helps the crisis management systems with its high speed of analysis to prevent irreparable financial and life damages from natural phenomena such as storms by evacuating areas in danger.