Ocean internal tides suppress tropical cyclones in the South China Sea

Tropical cyclones (TCs) are devastating natural disasters that significantly impact coastal regions. Accurate forecasting of both TC trajectory and intensity is crucial for effective disaster mitigation. While TC track forecasting has improved considerably in recent decades, intensity forecasting remains a challenge. TC intensity is governed by complex interactions between internal dynamics and external oceanic and atmospheric conditions. Key factors influencing intensification include vertical wind shear (which inhibits intensification), mid-atmosphere relative humidity, sea surface temperature (SST), and potential intensity (the theoretical maximum intensity attainable under given SST and atmospheric conditions). SST is particularly important, as it determines the enthalpy flux from the ocean to the TC. Strong winds associated with TCs entrain subsurface cold water, leading to SST cooling that reduces enthalpy supply and thus hinders intensification. Recent research has highlighted the role of pre-existing oceanic processes, such as mesoscale eddies and barrier layers, in modulating this SST cooling effect. The South China Sea (SCS) is a region of high TC frequency, yet it exhibits surprisingly weak TC intensification compared to other global TC-active basins. Previous studies have focused on TC genesis and interannual to decadal variability in the SCS, linking these to the East Asian monsoon, ENSO, and the Pacific Decadal Oscillation. However, few studies have investigated the specific intensification characteristics and the underlying atmospheric and oceanic mechanisms. Recent observational studies suggest that the SCS's powerful internal tides (ITs), interacting with TC-generated near-inertial waves, could amplify upper-ocean turbulent mixing and significantly impact SST cooling, thereby influencing TC intensification. This study aims to investigate whether and how these ITs influence TC intensification in the SCS.

The literature on tropical cyclone intensification highlights the importance of various atmospheric and oceanic factors. Emanuel (1999) emphasized the thermodynamic control of hurricane intensity, while Frank and Ritchie (2001) detailed the effects of vertical wind shear on simulated hurricanes. Studies such as those by Knaff et al. (2005, 2018) developed operational statistical schemes for typhoon intensity prediction. The role of SST in TC intensification has been widely explored (Schade, 2000), with studies showing the negative impact of TC-induced SST cooling (Chang & Anthes, 1979; Price, 1981; Bender & Ginis, 2000; D’Asaro et al., 2007). The influence of pre-existing oceanic conditions, such as mesoscale eddies and barrier layers, on SST cooling and TC intensification has also been demonstrated (Balaguru et al., 2012; Shay et al., 2000; Lin et al., 2005). In the SCS, previous research focused on TC genesis and interannual/interdecadal variability (Wang et al., 2007; Zuki & Lupo, 2008; Goh & Chan, 2010; Lee et al., 2012). However, the unique intensification characteristics of SCS TCs and the role of internal tides were under-explored until recent observational studies suggesting a potential link between internal tides, near-inertial waves, and amplified turbulent mixing (Guan et al., 2014; Liu et al., 2018).

This study utilized a comprehensive approach combining analysis of long-term TC datasets with numerical modeling. Four decades (1979-2019) of global TC data were analyzed to compare TC frequency, intensification characteristics, and related atmospheric and oceanic environmental factors between the SCS and other global TC-active basins. TC intensification rate was calculated for each TC track point as the 24-hour intensity change. The probability of rapid intensification (RI) was calculated using two methods: a TC-case based method and a TC track-point based method. The percentage of intense TCs (categories 3-5) was also calculated using both methods. Atmospheric and oceanic environmental factors were computed using 41 years of global datasets and composited for each TC-active basin. These factors included SST, potential intensity (calculated using a Fortran program provided by K. Emanuel), vertical wind shear (calculated using ECMWF interim database data), and relative humidity (also from ECMWF data). To address potential confounding effects, the researchers excluded TCs within 100 km of land or with landfall within 24 hours. A fixed-window analysis was also conducted to assess the influence of basin size on TC intensification. The study analyzed the TC self-induced oceanic cooling effect using daily satellite microwave SST data. Pre-TC SST was obtained by averaging SSTs 1-7 days before TC occurrence. TC-induced SST cooling was calculated as the difference between the mean SST during and after the TC and the pre-TC SST. Surface enthalpy flux was estimated using bulk aerodynamic formulae. Two complementary numerical models were used to investigate the impact of ITs on TC cooling. The first was the Price model, which is computationally efficient and used to simulate a large number of TC cases in the SCS, both with and without ITs incorporated into the model. The second was a coupled ROMS (Regional Ocean Modeling System) and WRF (Weather Research and Forecasting) model, providing a high-resolution case study of Typhoon Megi (2010), with and without ITs, for more detailed examination of the processes involved. For the ROMS model, the impact of ITs was included by imposing four primary tidal constituents on the lateral open boundary. In situ ocean current observations from global mooring and buoy databases were also used to compare ocean responses to TCs under similar conditions with and without ITs. This involved careful selection of comparable TC cases captured by moorings in the SCS (with ITs) and in the North Atlantic (without ITs).

The study revealed that the SCS has the weakest TC intensification characteristics globally, despite favorable atmospheric and oceanic conditions. The TC intensification rate in the SCS was only about half the global average, and the probability of rapid intensification was only one-third of the global average. The SCS also had the lowest percentage of intense TCs. Analysis of atmospheric and oceanic factors showed that none of the typical factors (SST, potential intensity, vertical wind shear, relative humidity) could explain the unusually weak intensification in the SCS. The small basin size and surrounding landmasses were also ruled out as explanations. Instead, the study found that the SCS exhibited the strongest TC-induced SST cooling and the lowest enthalpy flux among all global TC-active oceans. Despite higher pre-TC SST, the SCS experienced the most rapid SST drop during TC passages, significantly reducing enthalpy supply to TCs. The exceptionally strong cooling effect in the SCS could not be attributed to differences in TC attributes (intensity, size, translation speed) or subsurface thermal stratification alone. The key finding is that the SCS's exceptionally powerful internal tides (ITs) play a critical role in the strong cooling effect. The ITs, generated by the interaction between astronomic tides and sharp ocean ridges, interact with TC-generated near-inertial waves. This interaction dramatically amplifies turbulent mixing in the upper ocean, leading to stronger subsurface entrainment of cold water and hence greater SST cooling. In situ ocean current observations from a mooring in the SCS (with ITs) and a buoy in the North Atlantic (without ITs) under similar TC conditions corroborated these findings, showing significantly larger amplitudes and higher energy levels in the SCS, with evidence of secondary waves generated by nonlinear interactions between near-inertial waves and background ITs. Numerical experiments using both the Price model and the coupled ROMS-WRF model confirmed the role of ITs in enhancing TC cooling. The Price model simulations underestimated SST cooling by 34% in the absence of ITs, demonstrating the crucial contribution of ITs to the cooling process. The coupled ROMS-WRF simulations for TC Megi showed significantly stronger subsurface turbulent mixing and greater SST cooling in the presence of ITs, resulting in weakened TC intensity and rainfall. The spatial inhomogeneity of ITs strength in the SCS, along with seasonal variations, could also further modulate the suppression of TC intensification. This suggests that the interaction between ITs and TC-generated near-inertial waves is a critical mechanism driving the unusually strong cooling in the SCS, ultimately suppressing TC intensification.

This study provides compelling evidence that ocean internal tides (ITs) play a previously unrecognized, significant role in suppressing tropical cyclone intensification in the South China Sea (SCS). The unusually strong TC-induced cooling effect, directly linked to the interaction between ITs and TC-generated near-inertial waves, contradicts expectations given the generally favorable atmospheric and oceanic conditions for intensification in the SCS. The findings highlight the crucial need to incorporate the complex IT-TC interaction into operational weather prediction systems to improve TC intensity forecasts in the SCS and other regions with strong ITs. This research sheds light on a previously unknown interaction between a large-scale oceanic process and a high-impact weather phenomenon. The implications extend beyond the SCS, suggesting that ITs might influence weather systems in other regions where strong ITs are present, warranting further investigation.

This study demonstrates that the exceptionally strong internal tides in the South China Sea significantly suppress tropical cyclone intensification via a unique mechanism involving enhanced turbulent mixing and SST cooling. The findings highlight the importance of incorporating IT-TC interactions in weather prediction models to improve forecast accuracy. Future research should investigate the extent to which this mechanism operates in other regions with strong ITs and explore its potential influence on other weather phenomena.

The study acknowledges limitations due to the limited availability of in situ ocean current observations under TCs in the SCS and other ocean basins. The comparison of TC-ocean interactions with and without ITs was limited by the availability of suitable observational data, requiring careful selection of comparable TC cases. While the numerical models used were sophisticated, they rely on parameterizations and simplifications that might not fully capture the complexity of IT-TC interactions. The study's focus on the SCS might limit the generalizability of findings to other regions with varying IT characteristics and TC behaviors.