Winter Storms and Ocean Eddies Duel Over the Energy That Drives Deep Mixing

Deep in the subtropical Northwestern Pacific, roughly 1,500 kilometers southeast of Japan, a lone mooring has recorded a puzzle that has long haunted physical oceanographers: two winter storms of very different strength delivered wildly different amounts of energy to the sea surface, yet the swirling near-inertial waves they left churning in the ocean’s interior carried almost exactly the same amount of energy. A new study published in Ocean Science by Hongkai Wang, Zifei Chen, and colleagues at the Institute of Oceanology of the Chinese Academy of Sciences now offers a detailed explanation, and it centers on an unexpected player in the ocean’s energy budget: mesoscale eddies, the slow-spinning whirlpools hundreds of kilometers wide that pepper the world’s ocean like weather systems in the atmosphere.

Near-inertial waves, or NIWs, are internal ocean waves that oscillate at frequencies close to the local inertial frequency, the rotation rate imposed by the Earth at a given latitude. They are generated when winds change rapidly in time and direction, most dramatically during hurricanes and typhoons, but also during the broader, longer-lived winter storms that sweep across the mid-latitude ocean. Once generated in the wind-stirred surface mixed layer, these waves radiate downward into the ocean interior, where they steepen, shorten, and eventually break, injecting turbulent energy that mixes water masses across density surfaces. That diapycnal mixing is not a curiosity: it helps sustain the global meridional overturning circulation, the planet-scale conveyor belt that redistributes heat, carbon, and nutrients. Munk and Wunsch famously estimated in 1998 that roughly 2.1 terawatts of power are needed to keep that circulation and the abyssal stratification going, with wind forcing expected to supply about 1.2 terawatts of the total.

The trouble is that the numbers have never quite added up. Global estimates of the wind power pumped into near-inertial motions in the mixed layer range from about 0.3 to 1.5 terawatts, yet modeling work by Furuichi and colleagues suggested that only around 0.1 terawatts actually penetrates from the mixed layer into the ocean interior. Something must either boost or bleed away the energy along the way, and mesoscale eddies have emerged as prime suspects. Their strain fields can either hand energy to near-inertial waves or siphon energy out of them, and recent studies have suggested these exchanges can be remarkably efficient, in some cases rivaling the wind input itself.

To catch these exchanges in the act, the team deployed a subsurface mooring at 25 degrees North, 146 degrees East, near the northern flank of the North Pacific Subtropical Countercurrent, a region where a shallow eastward flow overlays the westward North Equatorial Current. This vertical shear favors baroclinic instability and seeds an energetic eddy field, making the site an ideal natural laboratory. From April 2017 to June 2018, the mooring carried two 75-kilohertz acoustic Doppler current profilers mounted at 400 meters depth, which jointly measured velocities in the upper 900 meters at hourly intervals with 8-meter vertical resolution, along with an array of conductivity-temperature-depth instruments and temperature loggers. The researchers focused on the winter window from late November 2017 to the end of January 2018, combining the mooring data with ERA5 wind reanalysis, satellite altimetry, and a coupled reanalysis product from the Met Office.

Two winter storms crossed the site during this period. The first, from 6 to 18 December 2017, packed mean winds of 8.5 meters per second and gusts reaching 13.8 meters per second. The second, from 6 to 17 January 2018, was noticeably weaker, with mean winds of 6.9 meters per second and a maximum of 10.2 meters per second. Using the classic slab model of Pollard and Millard, the team calculated that the first storm injected about 11.5 kilojoules per square meter of near-inertial energy into the mixed layer, roughly three times the 3.1 kilojoules per square meter delivered by the second. Slab-model near-inertial velocities peaked near 0.6 meters per second during the first storm but only about 0.2 meters per second during the second.

Yet when the researchers examined the near-inertial kinetic energy actually observed in the thermocline, the two events were nearly indistinguishable. The first event pushed near-inertial velocities up to 0.4 meters per second and energy densities to 20 joules per cubic meter at a depth of 120 meters; the second reached 0.21 meters per second and 21 joules per cubic meter at 220 meters. Both wave packets propagated down to roughly 300 meters, and rotary spectral analysis confirmed that the motions were strongly clockwise-polarized, as expected for near-inertial waves in the Northern Hemisphere, with near-inertial energy exceeding the diurnal and semidiurnal tidal energy in the upper ocean by a wide margin.

The resolution of the paradox came from calculating the energy transfer rate between the eddy field and the waves, following the framework of Polzin, in which the transfer depends on the strain of the background geostrophic flow acting on the near-inertial velocities. During the first storm, the transfer rate was persistently negative, meaning the eddies were draining energy from the waves: integrated over the upper 300 meters, the eddies extracted about 1.3 milliwatts per square meter, equivalent to roughly 46 percent of the wind-generated near-inertial energy. During the second storm the sign flipped. The eddies supplied about 0.36 milliwatts per square meter, an amount equal to roughly 43 percent of the wind input, effectively topping up the weaker storm’s contribution. Strong transfers coincided with periods when the Okubo-Weiss parameter was positive, indicating that strain, rather than vorticity, dominated the eddy flow, consistent with earlier findings from the Gulf of Mexico. After accounting for these transfers, the residual energy inputs for the two events, about 1.5 and 1.2 milliwatts per square meter respectively, became strikingly similar, neatly explaining the comparable observed wave energies.

The team also checked other possible energy pathways. Because no direct turbulence measurements were available, they applied the Gregg-Henyey-Polzin parameterization to estimate dissipation rates from the observed shear and stratification. The time-averaged dissipation rates during the two storms, 3.9 and 5.0 times ten to the minus ten square meters per cubic second, were nearly identical and about an order of magnitude smaller than the wind input and eddy-wave exchange, so turbulent dissipation was unlikely to be the dominant controller. Notably, the near-inertial shear variance was three to four times larger than that of the tidal bands, confirming that the storm-driven waves, not the tides, were the main shear agents. Radiation of energy out of the observed region could not be quantified from a single mooring, but the authors note it may also have contributed, particularly for the first event, whose low-mode structure favors long-range propagation.

That modal structure turned out to be the study’s second major finding. Projecting the observed near-inertial velocities onto the first twenty vertical baroclinic modes revealed two fundamentally different wave characters. The December event was dominated by low modes: the first four modes carried 48 percent of the total near-inertial kinetic energy, the wave packet had a large vertical wavelength of 487 meters, and its downward group velocity reached 31.2 meters per day, with a decay time of about nine days. The January event was the opposite: modes five through eight carried 41 percent of the energy, the vertical wavelength shortened to 372 meters, the group velocity slowed to 14.4 meters per day, and the decay time shrank to five days. Intriguingly, the mixed layer was actually deeper during the second storm, about 70 meters versus 51 meters, contradicting the conventional expectation that deeper mixed layers favor low modes and pointing to other controlling mechanisms.

Those mechanisms, the authors argue, are the interplay of eddy vorticity and the modal fingerprint of the wind itself. During the first storm, a westward-moving anticyclonic eddy passed over the mooring, and the relative vorticity shifted from positive to negative with depth and time, a configuration that stretches the vertical wavelength of the waves and favors low modes. During the second storm, predominantly negative vorticity compressed the vertical wavelengths and boosted high-mode content. Meanwhile, projecting the wind-generated energy flux onto the modal basis showed that the December winds deposited half their energy in the first four modes, while the January winds put 41 percent of their energy into modes five through eight. Because low-mode waves travel far from their generation sites while high-mode waves break locally and drive mixing, these differences matter for where and how strongly the winter ocean mixes. The findings suggest that accurately modeling the ocean’s mixing budget, and therefore its role in climate, requires resolving not just the winds but the eddies that quietly tax or subsidize every storm’s energetic legacy.

Subject of Research: Near-inertial wave generation, eddy-wave energy exchange, and modal structure during winter storms in the subtropical Northwestern Pacific Ocean

Article Title: Energetic near-inertial waves induced by winter storms and mesoscale eddies in the subtropical Northwestern Pacific Ocean

Article References: Wang, H., Chen, Z., Diao, X., Yu, F., Liu, X., Ren, Q., & Nan, F. (2026). Energetic near-inertial waves induced by winter storms and mesoscale eddies in the subtropical Northwestern Pacific Ocean. Ocean Science, 22(5), 3105-3120. https://doi.org/10.5194/os-22-3105-2026

Image Credits: AI Generated

DOI: 10.5194/os-22-3105-2026

Keywords: near-inertial waves, mesoscale eddies, winter storms, ocean mixing, Northwestern Pacific, mooring observations, baroclinic modes, wind energy input, diapycnal mixing, energy transfer, thermocline, physical oceanography