Forecasting Better the Tropical Cyclone Intensity in Western North Pacific is Now Possible

Considered one of the most powerful forces of nature, tropical cyclones (TCs) can bring extreme havoc on people, animals, the environment, and human populations. 

So far, experts have strived to improve the TC forecasting method, aiming to save lives. As a result, and only in the last few decades, TC tracking over the western North Pacific (WNP) has significantly improved. 

Here is what you need to know.

A Better Forecast System to Save Lives

A bold team of researchers from the Department of Atmospheric and Ocean Sciences/ Insitute of Atmospheric Sciences at Fudan University joined forces with the Shanghai Typhoon Institute of China Meteorological Administration. As a result, they created a brand-new forecasting technique that offers more accurate TC intensity forecasts.

How they did it

Researchers based the novel method on the logistic growth equation. They also needed to mix something known as SWR (stepwise regression) and machine learning (LightGBM) techniques. 

The results indicate that the new method offers much less important TC intensity forecast error than previous schemes.

Furthermore, the team compared novel LightGBM-data with results from the SWR and found something genuinely intriguing.

Apparently, the LightGBM technique consistently exceeded conventional SWR-based schemes.

Prof. Ruifen Zhan, who led the new research, explains:

“The new scheme also shows the potential for forecasting TC rapid intensification and rapid weakening, and for extending the current 5-day forecast time limit to 7 days.”

Such a thing is more than very useful. It’s revolutionary, a truly novel method of forecast that could save our lives.

Of course, more research will soon follow to overcome the issue of insufficient samples by mixing the transfer learning ways based on the recent research. That is extremely necessary because we can use it in operational forecasts.

As previously mentioned, TCs can cause extreme havoc upon our lives, communities, environment, and more. Therefore, finding the most precise ways of forecasting them should always be mandatory.