麻豆淫院

Rapid Intensification (RI) of tropical cyclones (TCs), defined as an intensity increase of at least 13 m/s within 24 hours, remains one of the most challenging phenomena to forecast because of its unpredictable and destructive nature. Although RI accounts for only 5% of all TCs, its sudden and severe development poses significant risks to affected regions.

Traditional forecasting methods, such as numerical weather prediction and statistical approaches, often fail to consider the complex environmental and structural factors driving RI. While artificial intelligence (AI) has been explored as a means to improve RI predictions, most AI techniques have struggled with high false alarm rates and limited reliability.

To address this issue, researchers from the Institute of Oceanology of the Chinese Academy of Sciences (IOCAS) have developed a new forecasting model for RI of TCs based on "contrastive learning." The study was in the Proceedings of the National Academy of Sciences on January 21.

The new model has two inputs: Input A, a known RI TC sample, and Input B, an unknown sample to be forecasted. It extracts features from both inputs and calculates their distance in a high-dimensional space. If the distance is small, Input B is forecasted as an RI TC; if large, it is classified as a non-RI TC. Each unknown sample is compared with 10 known RI TC samples, and if more than five of the comparisons classify it as an RI TC, it is then classified as such.

Additionally, this study uses satellite imagery alongside atmospheric and oceanic data to balance RI and non-RI TC data. The model learns to differentiate between RI and non-RI TCs by comparing the two inputs during training.

When tested on data from the Northwest Pacific between 2020 and 2021, the method achieved an impressive accuracy of 92.3% and reduced false alarms to 8.9%. Compared to existing techniques, it improved accuracy by 12% and reduced false alarms by a factor of three, representing a major advancement in forecasting.

Although the model was initially trained using reanalysis data, the researchers created an operational forecasting scenario by substituting the reanalysis data with ECMWF-IFS numerical model forecast data from 2020 to 2021 as input.

The results demonstrated comparable forecasting accuracy, further validating the reliability of this approach and confirming its suitability for real-time forecasting scenarios. This capability for real-time forecasting can significantly enhance early warning systems, thus enhancing global disaster preparedness.

"This study addresses the challenges of low accuracy and high false alarm rates in RI TC forecasting," said Prof. Li Xiaofeng, the corresponding author. "Our method enhances understanding of these extreme events and supports better defenses against their devastating impacts."