![]() |
| Weddell Sea Tides. Amplitude (color scale on right) and phase (black lines) for the sea surface elevation amplitude of the semidiurnal M2 tidal constituent. |
Complex interactions between the ocean, atmosphere, sea ice, and floating glacial ice sheets in the Weddell Sea lead to surface waters becoming cooler, saltier, and therefore denser. The dense water sinks, carrying surface-acquired properties such as O2 and CO2 into the deep ocean where they are distributed throughout the world ocean. This process is called "ventilation" and plays a major role in the coupled global ocean and atmosphere. Any modification of the sea ice distribution by oceanic or atmospheric forcing is also relevant to climate modeling because the reflectivity of ice is much greater than open water, thus more ice implies much greater reflection rather than absorption of incoming solar radiation.
The most vigorous cooling and salinization of surface water occur over the continental shelf and under the ice shelves around the perimeter of the Weddell Sea. Many processes help to modify the surface water, and determine the final property of the modified water as it sinks down the continental slope. Tides play a role in this system, both directly and through interactions with other components of the atmosphere-ocean-ice (AOI) environment. Some of the mechanisms that are of interest to us are as follows.
For these reasons, we have been funded to carry out several studies to improve our understanding of the distribution of tidal energy in the Weddell and Scotia Seas, and to investigate the effects of tides on other processes. Our studies to date fall into two categories: improving numerical models that predict tidal energy distribution; and using the models to understand the role of tides in the large-scale circulation of the Weddell Sea and its cover of sea ice. Future funded work includes a detailed study of the interactions of ocean tides with the glacial ice shelves. More information on specific projects can be found by following the links below.
Based on application of empirical models to the modeled barotropic tidal flow [Robertson et al., 1998], the pycnocline over the upper continental slope in this region may undergo significant mixing due to shear instabilities associated with baroclinic tides and other internal gravity waves. We also expect turbulence in the boundary layer to be significant. To investigate these processes, we start with a quasi-2-D version of POM, assuming no variability in the along-slope direction, and considering only the dominant semidiurnal M2 tide. The parameters of interest to us include:
Stress applied to the base of pack ice by ocean tidal currents is one component determining ice motion. An analysis of the motion of satellite-tracked drifters, and comparisons with an updated version of our ocean tidal model [Robertson et al., 1998], shows that tidal currents affect ice motion along the Ronne Ice Front, the northern Filchner Trough, and the outer shelf and upper continental slope of the southern and western Weddell Sea. Over the remainder of the Weddell Sea, near-inertial response to wind stress, and direct forcing by tidal-band energy in the wind stress spectrum, account for most tidal-band ice motion. Ice divergence also shows energy at tidal frequencies, suggesting that tides can contribute to the oceanic heat loss to the atmosphere by an increase in lead fraction in tidally active areas. Along the southern and western continental shelf break and the Ronne Ice Front, the RMS horizontal divergence calculated from our Weddell Sea ocean tides model is about 2x10-6 s-1, consistent with a mean tide-forced lead fraction of 3%. The diurnal and semidiurnal model divergence fields are shown in figure 2. Most of the horizontal tidal divergence along the shelf break is due to advection of the water column over sloping topography by the tides: continuity requires that the water column’s area change in inverse proportion to the water depth.
This work has been prepared as manuscript entitled "Tidal-band ice motion and divergence in the Weddell Sea", now submitted to J. Geophys. Res. An Abstract is shown below.
TIDES IN THE WEDDELL SEA
Robin Robertson, Laurie Padman*, and Gary D. Egbert
College of Oceanic and Atmospheric Sciences
Oregon State University
Corvallis, Oregon
*
Now at Earth and Space Research
Seattle, Washington
We use a high-resolution barotropic tidal model to predict tidal elevations and currents in the Weddell Sea. The ocean cavity under the Filchner-Ronne Ice Shelf is included in the model domain. Tidal elevations exceed 1 m at the back of the Filchner-Ronne and Larsen Ice Shelves. Tidal velocities are small over the deep basins but are generally greater than 10 cm s-1 over the continental shelves. Velocities occasionally reach 1 m s-1 in the shallow water near the General Belgrano Bank and under the Ronne Ice Shelf near the ice front. Model performance was evaluated through comparisons with TOPEX/Poseidon altimetry, bottom pressure gauge records, and current meter data. The largest discrepancies between the model results and measurements occur over the continental slope and under the ice shelves. The principal error sources are believed to be inaccurate bathymetry in our model, tidal analysis limitations associated with short data record lengths, and omission of baroclinic tides. Model results indicate that tides play a significant role in the circulation and heat flux in the Weddell Sea. We discuss the influence of tides on mean flow through the modified effective bottom drag, and the generation of baroclinic tides and other internal gravity waves through interactions of the tide with topography.
TIDAL-BAND ICE MOTION AND DIVERGENCE
Laurie Padman 1, Christoph Kottmeier 2, and Robin Robertson 3
1
Earth & Space Research2
Alfred Wegener Institute for Polar and Marine Research3
College of Oceanic & Atmospheric SciencesWe describe the spatial variability of tidal-band ice motion and divergence in the Weddell Sea using analyses of the motion of four groups of satellite-tracked drifters deployed between 1986 and 1995, and comparisons with the output from a depth-averaged, ocean-only tidal model. The strongest tidal velocities, which can exceed 20 cm s-1, are found over the southern and western continental slope and shelf. Spatial derivatives of ice motion (divergence, curl, and shear) also contain tidal-band energy that varies spatially and with time. Tidal-band root-mean-square (RMS) divergence values (s (Ñ × D), where D(x,y,t) is the ice drift velocity) can exceed 2x10-6 s-1. In the diurnal band, the largest values of s (Ñ × D) are found along the southern and western outer shelf and upper slope, consistent with diurnal tides along the shelf being dominated by topographically trapped vorticity waves that have large cross-slope currents and small length scales associated with them. In the semidiurnal band, the largest values of s (Ñ × D) are found in several areas over the deep water of the Weddell Basin, the northern end of the Filchner Trough, and along the Ronne Ice Front. We attribute semidiurnal divergence over deep water, where local ocean tidal currents are weak based on both modeling and current meter data, to the near-inertial response of the ice cover to wind stress. The divergence-convergence cycle at tidal frequencies generates a mean lead (open water) area of 2-5% in energetic regions. The increased lead fraction implies a significantly increased area-averaged winter ocean-to-atmosphere heat exchange rate and salt flux into the upper ocean during refreezing in the leads. When values of s (Ñ × D) are averaged over the entire Weddell Sea west of the Greenwich meridian and south of 63oS (the tip of the Antarctic Peninsula), we find that near-inertial resonant ice response to wind stress, and the small but significant higher-frequency (tidal-band) energy in the wind field, provide most (about 80%) of the RMS ice divergence. Tidal forcing on the shelf and upper slope provides the remaining 20%. However, ocean-atmosphere interaction over the shelf and slope plays a critical role in Weddell Sea thermohaline forcing. Therefore, the distribution of tide-forced RMS ice divergence, rather than simply its contribution to the domain-averaged value, is relevant to Weddell Sea general circulation modeling.
A depth-averaged, ocean-only tidal model is used to interpret the buoy observations. The model is first compared with moored current meter measurements on the southern continental shelf and slope, and near the Ronne Ice Front. Modeled semidiurnal currents agree well with all measurements, after correcting for the meters’ locations in the thick frictional boundary layer that occurs near the critical latitude for the M2 tidal constituent (w (M2)=f at l crit» 74o28’ S). Diurnal tidal currents are well modeled at the Ronne Ice Front, but are overestimated along the southern shelf break and Filchner Trough. The diurnal tides are very sensitive to bathymetry, which is only poorly known in the southwestern Weddell Sea, even after inclusion of data collected during the ROPEX cruise in early 1998.
In the model velocity field, Eulerian RMS divergence of depth-averaged horizontal velocities is insignificant, except in shallow water very close to the coast. However, tidal-band advection across sloping topography creates "Lagrangian" RMS divergence values that are qualitatively consistent with the buoy group analyses, when comparable spatial scales (O(100) km) are used.
Back to Tide-Forced Ice motion section