Ocean tides can drag the atmosphere and cause tidal winds over broad continental shelves

The study investigates whether and how oceanic tidal currents can exert a frictional drag on the overlying atmosphere, generating measurable tidal winds within the atmospheric boundary layer and transferring tidal energy from the ocean to the atmosphere. Ocean tides, primarily driven by lunar and solar gravitational forces (with M2 at ~12.42 h as the dominant constituent in many locations), represent a major oceanic energy source and dissipate largely on continental shelves via bottom friction and conversion to internal tides. Recent work on current–wind interactions (current feedback, CFB) shows that surface ocean currents can modify surface stress, reduce atmospheric energy input to the ocean, and damp mesoscale and submesoscale activity. Because CFB effectively acts as a top drag across spatial and temporal scales, the authors hypothesize that tidal currents also impart momentum to the atmosphere, creating a high-frequency wind signal at tidal frequencies and constituting a previously unrecognized sink of tidal energy. The purpose is to quantify this mechanism, its energetic significance, and its expression as tidal winds over broad continental shelves, using the English Channel as a natural laboratory.

Background studies indicate that ocean tidal energetics are substantial (e.g., M2 work rate ~2.4 TW), with dominant dissipation on shelves by bottom friction and through internal tide generation. Work on air–sea current feedback establishes that currents can significantly modulate wind stress, reduce wind work on the ocean, and damp mesoscale variability, altering views of wind-driven circulation. Classical atmospheric tides are categorized as gravitational and thermal; while influential in the mesosphere and lower thermosphere, their tropospheric signatures are weak, contributing only small pressure variations that minimally affect sea level (e.g., ~1 cm via inverse barometer). Prior to this study, frictional atmospheric tides directly induced by ocean tidal currents had not been considered. The literature therefore motivates examining whether tidal currents can generate near-surface atmospheric tidal winds via CFB and how much tidal energy is lost to the atmosphere through this pathway.

The authors perform eddy-rich, fully coupled ocean–atmosphere simulations for the English Channel during 2010, complemented by in situ sea surface height (SSH) and wind observations. Two twin experiments are run: CTRL includes tides and current feedback (CFB); NOCFB includes tides but omits CFB.

• Ocean model: CROCO (built on ROMS), free-surface, terrain-following coordinates, split-explicit time stepping, Boussinesq and hydrostatic approximations. Domain spans 20°W–8°E and 43.3°N–61°N, grid 586×586 with 2.7–4.0 km resolution. Bathymetry from SRTM30_plus with Gaussian smoothing (width 4× grid spacing) and local smoothing (r=0.2) to reduce pressure gradient errors. Vertical grid: 50 levels with enhanced resolution at surface, bottom, and thermocline via stretching (parameters 6, 7, 8, 2 and 300 m). Initialization from daily GLORYS2V4; 1-year ocean-only spin-up with CFSR forcing, then coupled integration for 2010 with interannual GLORYS boundary forcing. Tidal forcing: TPXO tidal elevations and barotropic currents applied at open boundaries (Flather-type), plus internal tidal potential (astronomical, solid Earth body tide, self-attraction and load). Bottom drag via logarithmic law with roughness length z0 = 10−2 m. Vertical mixing via KPP.
• Atmospheric model: WRF v4.1 on a slightly larger domain than CROCO (to encompass sponge layers), 6 km resolution. Initial/boundary conditions from CFSR (~40 km), run for 2010. Surface fluxes (heat, freshwater, momentum) computed via bulk formula and passed to CROCO. The vertical turbulent diffusion scheme is modified to properly include surface current impacts on fluxes.
• Coupling: Hourly exchange via OASIS3. In CTRL, surface stress uses relative wind, τ = ρa CD (U − Uo)(U − Uo), where U is 10 m wind and Uo the surface ocean current; in NOCFB, τ = ρa CD U U (no current feedback). An additional experiment (CFB without tides) is mentioned to support conclusions on friction-induced tidal winds.
• Analyses: Comparison of CTRL vs NOCFB isolates the effect of CFB on tidal timescales. Spectral and co-spectral analyses are performed at mid-Channel and across the region: SSH spectra (model vs observations at the Bournemouth gauge), current and stress spectra (focus on M2), co-spectrum of wind work FK = τ^T U (with Fourier transforms to identify frequency-dependent contributions), and wind spectra at 10 m. Spatial maps of zonal tidal currents, stress anomalies (CTRL − NOCFB), wind anomalies (CTRL − NOCFB), and wind work are used to relate tidal currents to atmospheric response. Vertical profiles assess the penetration of tidal wind anomalies through the atmospheric boundary layer. Statistical coupling is quantified by bin-averaged scatter of zonal current vs 10 m wind anomalies across the Channel to estimate coupling coefficient sw (slope of linear fit). A global first-order estimate of M2 tidal winds is produced by scaling TPXO9-v5 M2 tidal currents with sw.

• Model–observation fidelity: CROCO reproduces observed SSH tidal spectra in the English Channel, with M2 dominating and realistic amplitudes for K1, M2, and higher harmonics; modeled tidal surface currents reach >3 m s−1 near coasts and ~2 m s−1 mid-Channel.
• Stress response: Surface stress anomalies (CTRL − NOCFB) mirror tidal current patterns and are strongly anti-correlated with currents (temporal correlation of zonal components ~−0.8 over a year). The stress spectrum in CTRL has a pronounced M2 peak in regions of strong tides.
• Energy pathway (wind work FK): In NOCFB, FK is generally positive (atmosphere forcing ocean). In CTRL, over strong tidal current regions, FK shows an energy sink from ocean to atmosphere at M2, evidenced by a large negative co-spectral peak at M2, indicating tidal energy transfer to the atmosphere via top drag.
• Tidal winds: The atmospheric response yields measurable tidal-frequency winds. Zonal wind anomalies (CTRL − NOCFB) spatially track tidal currents and stress anomalies; correlations between zonal current and wind anomalies are ~0.7 over a year with no detectable time lag (<1 hour). Wind spectra at 10 m in CTRL show a clear M2 peak absent in NOCFB. Tidal wind amplitudes can exceed 1 m s−1 and extend through the atmospheric boundary layer (example stable ABL height ~800 m), attenuating with height.
• Coupling coefficient: A linear relationship between tidal current and wind anomalies gives sw ≈ 0.32, implying tidal wind amplitudes are ~32% of surface current amplitudes (up to a few m s−1 in extreme tidal currents). The sign is consistent with negative surface stress response and energy transfer to the atmosphere.
• Global estimate: Scaling TPXO9-v5 M2 currents by sw suggests friction-induced M2 tidal winds occur mainly over broad shelves, reaching up to ~1.5 m s−1 near Alaska, Argentina, Norway, and the Amazon.
• Energetics magnitude: The inferred ocean-to-atmosphere M2 energy sink is ~0.5% of bottom drag dissipation, approximately 0.012 TW globally.

The findings directly confirm the hypothesis that tidal currents can drag the atmosphere via current–wind feedback, generating a distinct class of atmospheric tidal winds at M2 frequency within the boundary layer and establishing a non-negligible ocean-to-atmosphere tidal energy sink. The mechanism mirrors mesoscale CFB physics but at tidal frequencies, with stress anomalies anti-correlated to currents and an immediate wind response that penetrates the ABL through vertical mixing. The estimated coupling coefficient (sw ~0.32) enables first-order extrapolation, pointing to widespread presence of tidal winds over global broad-shelf regions where tidal currents are strong. This mechanism has implications for air–sea momentum and energy budgets and can influence near-surface wind statistics relevant to climate modeling, weather forecasting, marine operations, and renewable energy. For example, adding a 1 m s−1 tidal wind to an average 6 m s−1 flow could increase wind power by ~10% (given cubic dependence on wind speed), comparable to projected climate change effects. The study also suggests that accounting for tidal winds could reduce wind biases in coastal shelf regions and should be considered in coupled models and potentially parameterized in ocean-only systems.

This work identifies and quantifies a new atmospheric tidal regime generated by frictional coupling to oceanic tidal currents. Using high-resolution coupled simulations and observations in the English Channel, the authors show that tidal currents imprint M2-frequency signals on surface stress and low-level winds, creating tidal winds with amplitudes ~30% of the underlying current speed and transferring a measurable fraction of tidal energy (~0.012 TW) to the atmosphere. A simple global scaling indicates that such winds likely occur across many broad-shelf regions and can reach ~1.5 m s−1. These results call for inclusion of tidal wind effects in coupled models and development/validation of parameterizations for ocean-only models (e.g., stress or wind corrections tuned for tidal frequencies). Future research should refine global estimates with higher-resolution coastal bathymetry, assess variability across different tidal regimes and atmospheric conditions, and evaluate impacts on forecasting, climate simulations, and wind energy resource assessments.

• The estimated global tidal wind map scales TPXO9-v5 M2 currents by a constant coupling coefficient and likely underestimates localized maxima near small-scale unresolved bathymetry.
• The quantified energy sink (~0.5% of bottom drag; ~0.012 TW) is based on regional modeling and extrapolation; while indicative, it requires broader verification across diverse regions and tidal regimes.
• The English Channel case features weak mesoscale activity; responses may differ where mesoscale variability is stronger or boundary layer conditions vary.
• Proposed parameterizations for ocean-only models need testing and tuning specifically at tidal frequencies before operational use.