Christopher O. Tiemann, Peter F. Worcester, Bruce D. Cornuelle
Scripps Institution of Oceanography, University of California
at San Diego,
La Jolla, California 92093, USA
Institut fur Meereskunde, University of Kiel, 24105 Kiel, Germany
Abstract. The Strait of Gibraltar Acoustic Monitoring Experiment was conducted in April 1996 to determine the feasibility of using acoustic methods to make routine, rapidly repeated transport measurements in the Strait of Gibraltar, as well as to explore the acoustic scattering caused by the unique internal wave field in the Strait. The acoustic data from high frequency (2-kHz) reciprocal transmissions across the Strait are unique in that they clearly isolate the acoustic effects of passing internal bores without the added complexity of surface and bottom interactions. Although the acoustic scattering caused by each internal bore is different, some common characteristics can be identified.
The basic circulation in the Strait of Gibraltar is relatively simple. An upper layer of warm, fresh, Atlantic water about 100 m thick flows east into the Mediterranean Sea, and a lower layer of colder, salty Mediterranean water flows west into the Atlantic. This mean flow is modulated by large tidal flows and is also subject to hydraulic controls at the shallowest sill (Camarinal Sill on the west side of the Strait) and the narrowest constriction (Tarifa Narrows.) Furthermore, there are tidal fluctuations in the depth of the interface between the upper Atlantic and lower Mediterranean water layers.
Perhaps the most interesting feature, though, is the propagation of
internal bores which are released at the Camarinal Sill at the relaxation
of most high tides and propagate east down the Strait. These bores can
have amplitudes as high as 100 m peak to peak and wavelengths of 1 km.
They are found at the interface depth and travel with a phase speed of
1-2 m/s. The bore eventually disintegrates into a train of as many as twelve
internal solitary waves. [Armi and Farmer, 1988; Farmer and Armi, 1988;
Watson and Robinson, 1989]
Three 2-kHz transceivers (labeled T1, T2, and T3 in Figure 1) were installed just above the bottom of the Strait at 200 m depth at the endpoints of two different acoustic paths (shown in red) across the Strait. One path is perpendicular to the flow while the other is roughly parallel. Reciprocal transmissions along the two paths were sent and recorded every 2 minutes. [Worcester et al., 1997]
In addition to the acoustic data that was recorded, the tilts and orientations
of the three moorings were measured as well and used to correct ray travel
times for instrument motion. Examination of the motion data for the T1
instrument showed brief but violent kicks occurring roughly every 12 hours
during spring tides. These kicks are due to the passing of an internal
bore over the instrument and are now being used as a "clock" of internal
The corrected, absolute travel times for several acoustic rays are shown as a "dot plot" in Figure 2. For each transmission from T1 to T2 occurring over the two weeks shown along the x-axis, a dot is placed along the vertical axis at each time when a peak in the acoustic amplitude occurs. The size of the dot is proportional to the signal-to-noise ratio. The times from the internal bore clock are overlaid in red, indicating the passing of an internal bore over the T1 instrument.
The transmissions through the strong tidal flows, internal bores, and trains of interfacial internal waves have complex travel time fluctuations and path structure. The dot plots for the T1-T2 path show that the earliest arrival, from a deep-going ray which samples only the lower Mediterranean water layer, was stable over the duration of the experiment and had a strong tidal signal. This early arrival was successfully tracked over all days of the experiment and is shown in green in Figure 2. The later arrivals are from shallow rays which sample the interface between the two water layers. They also show tidal variability, but the path structure is more complicated; the shallow rays are smeared into a broad cluster of arrivals that are difficult to track. Dot plots for the T1-T3 path are generally similar to those for T1-T2, but the tidal signal is much smaller for the deep ray arrivals as the acoustic path is perpendicular to the tidal currents.
A closer look at the dot plot data from Figure 2 shows an interesting effect of the passing bores on the acoustics. The travel time of the earliest ray decreases with each passing of the bore, followed by a sudden increase in travel time shortly after the bore has passed. Because the effects are slightly different with each bore occurrence, nine examples of travel time data showing bore effects from the days of spring tides have been stacked on each other, as shown in Figure 3, in order to identify some effects common in all cases. Figure 3 shows dot plot data from T1-T2 transmissions for 5 hours before and after nine bore crossings, where each of the nine examples has been shifted to align all the green tracked paths at the time of a bore occurrence.
Figure 3 shows that the brief decrease, then increase, in lower ray path travel times is a common acoustic effect seen each time an internal bore passes over the T1 instrument. Differential travel times from reciprocal transmissions along the T1-T2 path show such travel time changes to be an effect of temperature, rather than current effects. Perhaps warm shallow water is being displaced deeper by the bore to the depths of the instruments, increasing the soundspeed near the endpoints of the deep going rays. Also note how both the shallow and deep ray arrivals show much more scattering in the hours following a bore crossing, with some shallow rays arriving at the same time as the deep rays.
Figure 4 also shows stacked dot plot data for nine bore crossings during spring tides, but now for transmissions along the other acoustic path from T1 to T3. Once again there is the common feature of a brief decrease, then increase, in lower ray travel times shown in green, as the bore passes the instrument. The cloud of shallow ray arrivals shows a very sharp decrease in travel times, and for hours after the bore has passed some shallow rays are arriving earlier than the deep rays! This may be because the interface layer is deepened behind the passing bore and is slow in restoring itself, allowing shallow rays to travel through warmer water during that time.
The challenge is to explain these acoustic observations in terms of the physical processes occurring. Acoustic propagation models through range- and time-dependent soundspeed fields representing the Strait of Gibraltar and perturbed by internal bore models are being used to understand these observations and answer several other questions: To what aspects of the bore are the acoustics most sensitive? How complicated of an ocean model is necessary to reproduce the observations, and are existing ray trace codes adequate? Can acoustics be used to observe a bore?s direction of propagation and phase speed?
Armi, L., and D. M. Farmer (1988). The flow of Mediterranean Water through the Strait of Gibraltar. Progress in Oceanography, 21, 1-105.
Farmer, D. M., and L. Armi (1988). The flow of Atlantic Water through the Strait of Gibraltar. Progress in Oceanography, 21, 1-105.
Watson, G., and I. S. Robinson (1989). A Study of Internal Wave Propagation in the Strait of Gibraltar using Shore-Based Marine Radar Images. J. Phys. Oceanogr., 20, 374-395.
Worcester, P., U. Send, B. Cornuelle, and C. Tiemann (1997). Acoustic monitoring of flow through the Strait of Gibraltar. Shallow-Water Acoustics, Proc. of the International Conference on Shallow Water Acoustics, Beijing, China, 21-25 April 1997. R. Zhang and J. Zhou, eds., 471-477, China Ocean Press.