The Benefits of Combining Coupled Wave-Current Models with SAR Observations for the Interpretation of Ocean-Surface Currents
Trevor Macklin(1) , Judith Wolf(2) , Sarah Wakelin(2) , Christine Gommenginger(3) , Graham Ferrier(4) , and Alan Elliott(5)
West Hanningfield Road,
Chelmsford CM2 8HN,
(2) Proudman Oceanographic Laboratory, 6 Brownlow Street, Liverpool L3 5DA, United Kingdom
(3) National Oceanographic Centre, European Way, Southampton SO14 3ZH, United Kingdom
(4) University of Hull, Department of Geography, Hull HU6 7RX, United Kingdom
(5) University of Wales, Menai Bridge, Anglesey LL59 5EY, United Kingdom
Mesoscale features, such as eddies and fronts, are important phenomena that influence the ocean circulation on regional and global scales. General circulation models, for example OCCAM, now include these features, and are beginning to find widespread use. There are also increasing demands to monitor and predict changes in the current field in coastal waters. Historically, many of the agencies responsible for predicting water quality have used two-dimensional, depth-averaged models of the currents. Such models are inadequate for many applications, because they neglect important hydrodynamical processes. More recent models represent the three-dimensional structure of currents and they include the coupling between the currents and the surface waves. However, these still require suitable input data and boundary conditions in order to produce reliable outputs.
Synthetic-aperture radar (SAR) observations are one source of relevant information which can be combined with these models. The SAR image modulations are related to spatially varying surface currents via surface wave-current interactions, among other features which modify surface roughness e.g. slicks. The novel mode of Along-Track Interferometry (ATI) provides information on surface currents via estimates of line-of sight velocities.
Here we review the results from both SAR images and ATI over a variety of sites, including estuaries, other coastal waters and shelf seas. The analyses are based on forward modelling of the radar signatures using surface wave spectra from a third-generation wave model (WAM), coupled to a model of the current field, together with a short-wave model. The modelled signatures are then compared with the radar data in order to assess the ability to reproduce the observed oceanographic signatures. We summarise the present status of these models and we highlight the critical issues associated with them. One of these issues is the procedure to set the boundary conditions from the wind field in the coupled wave-current models. This is usually achieved by nesting high-resolution local-area models within low-resolution large-scale models. The accuracy of the available information on wind fields is a key factor for this procedure.
The modelling of the surface currents for idealised cases of current flow shows the relative importance of bottom friction and continuity of flow. The assumption of local one-dimensional flow which has been used by many authors appears to be a reasonable approximation for many cases, but more complicated behaviour is expected around the edges of bathymetric features. The effects of two-dimensional flow can therefore be incorporated as a correction to existing schemes which relate the bathymetry to the local one-dimensional flow.
Further research is needed to reduce the impact of uncertain factors in both the wave-current model and the SAR forward model. In particular, we identify needs for better descriptions of the wind field and the directional distribution of the short-wave component of the wave height spectrum. A promising aspect for further study is the combination of ATI with wind information retrieved from the conventional SAR backscatter.