The worst-class tropical cyclone-generated high waves identified by seismic noises

Abstract The extreme winds of tropical cyclones generate high waves over the ocean, causing severe damage to offshore facilities and coastal communities1. Disaster mitigation requires accurate prediction and forecasting of the worst-class high waves. Furthermore, wave development, an entity of momentum loss from a tropical cyclone, determines the intensity of the tropical cyclone2. However, the physics of wave development is not fully understood, and the number of mid-ocean observations during extreme tropical cyclones is extremely insufficient3. Therefore, physically and statistically evaluating worst-class high waves is difficult. However, ocean waves excite seismic noise (microseisms)4,5,6. Here, we utilize long-term microseisms observed by seismic observation networks7 to identify historical worst-class tropical cyclone-generated high-wave events around Japan. The identification is facilitated by using microseism energy within an 8–10 s period, and we show that Typhoon Wipha in 2013, Typhoon Lan in 2017, and Typhoon Hagibis in 2019 were the worst-class events in the past 20 years. Here, we show that microseism observations contain rich information on the development and propagation of waves depending on wave frequency, leading to a better understanding of the physics of high waves and accurate predictions of worst-class events.