In a recent paper published in Geophysical Research Express, the team used data collected by NASA satellites to model signs of hurricane growth. Predicting the intensity of a storm when it’s brewing is difficult, and there are many factors involved, but NASA’s Jet Propulsion Laboratory has identified several key parameters that use artificial intelligence algorithms to make more accurate predictions.
Researchers have determined that the surge in rainfall near the storm’s core is a good indicator that the storm is strengthening further. In addition, the intensity of the storm can be determined in part by the amount of water ice present in the hurricane’s strong clouds. The stronger the wind, the greater the chance that the water will be brought high into ice.
And these key parameters can be discovered by NASA satellites, which can better predict storms by combining them with artificial intelligence algorithms. In testing their new framework for predicting changes in storm intensity, the researchers found that it was 40 to 200 percent more accurate than the standard method of predicting rapid enhancement, or RI, and that the false positive rate was only 4 to 7 percent higher. This is undoubtedly a very obvious improvement and the negative impact of the risk is very small.