Field observations of an internal solitary wavetrain impacting a shoaling bottom are presented. Measurements of the spatio‐temporal characteristics of the shoaling waves are given, as well as estimations of the mixing they may have caused upon impact. The observations are discussed in the context of numerical simulations, laboratory experiments, and hypotheses recently raised on the origin and evolution of internal solitary waves in coastal environments.

[1] A parameterization for the on-shelf mass flux induced by upwelling through a shelf break submarine canyon is estimated by laboratory spin-up experiments. We determine the effects of a submarine canyon on flow evolution implicitly by measuring the topographic drag force in the context of a heuristic model. Trials were performed across a range of values for the shelf break velocity, Coriolis frequency, and buoyancy frequency. Assuming the drag force within the canyon is balanced locally by rotation, we propose a parameterization for upwelling through a canyon provided that the Coriolis frequency, buoyancy frequency, shelf break velocity, and canyon dimensions at the shelf break depth are known. We use our results to compare wind-forced and canyon-forced upwelling over Astoria Canyon off the coast of Washington State. The analysis suggests that canyonforced upwelling through Astoria Canyon is of equal importance to wind-forced upwelling directly above it on seasonal scales.

Internal waves heave the background flow through which they propagate. If the background flow is vertically sheared, the high-pass-filtered velocity field will thus contain signals of both the wave velocity and the heaved flow. Under conditions of large wave amplitude and large background shear-a common situation for nonlinear internal waves in coastal waters-the velocity fluctuations caused by wave heaving of the background flow can be comparable to the wave velocity itself. This complicates the inference of wave properties such as energy flux and propagation direction. The present study deals with methods to infer propagation direction in such situations. Attention is given to three methods that may be applied to acoustic Doppler current profiler measurements: a ''filtering'' method that estimates wave signals from high-pass-filtered time series, a ''beamwise'' method that infers wave direction from lagged correlations of echo intensity between the spatially separated acoustic beams of the profiler, and a ''modal'' method that separates background and wave signals by regressing the high-pass-filtered velocity field onto a normal-mode wave model. The methods are tested using synthetic datasets. The results suggest that the filtering method is biased by wave heaving of the background shear, while the beamwise and modal methods are resistant to heaving. The beamwise method provides accurate predictions of wave propagation angle for cases in which the measurements have high temporal resolution and the environment exhibits no depth-averaged background flow. The limitation on depth-averaged flow is relaxed for the modal method, but it requires the measurement of stratification. These issues are illustrated, and the applicability of these methods is explored with a series of sensitivity tests, and it is found that the different methods perform well under different circumstances.

[1] The development of seismic oceanography requires direct comparison of seismic data to high-resolution oceanographic measurements over long horizontal scales. Here, we compare multichannel seismic (MCS) reflection images to 300 km of spatially-coincident, high-resolution (<1 km) expendable bathythermograph (XBT) surveys that were collected near a frontal region of the Gulf Stream. Fronts, eddies, tendrils, and interleaving were among the features identified in both data sets. In some cases, identification of features would be difficult if only hydrographic data were collected at conventional spatial scales. Comparing MCS reflection images with others derived purely from hydrographic data reveal many similarities and show that interleaving can be clearly identified with seismic methods. Varied time lags between MCS and hydrographic data collection identified the need for the separation between collecting both data sets to be short (i.e. hours to days), with advective processes and decorerlation time scales of desired features affecting acceptable sampling strategies. Citation: Mirshak, R., M. R.