Figure 5.3 Observations and predictions at Ocean Station November (in the North Pacific Ocean) during June 1961. Upper panel: observed wind speed. Middle panel: observed (solid line) and predicted (dashed line) sea-surface temperatures. Bottom panel: observed (solid line) and predicted (dashed line) mixed layer depths (Clancy et al., 1981).
The diurnal ocean surface layer (DOSL) model, which predicts diurnal patterns of the sea-surface temperature (SST) field, has been developed to support high-frequency sonar operations against shallow targets (Clancy et al, 1991b; Hawkins et al, 1993). Diurnal variability in the mixed layer dominates short-term changes in the acoustic behavior of the surface duct. The well-known 'afternoon effect' discovered during Second World War denotes the loss of a surface duct due to the creation of a shallow, transient thermocline by local heating. Diurnal SST changes also affect the use and interpretation of satellite data. Specifically, satellites measure the skin temperature of the ocean, and these surface measurements may or may not be characteristic of the bulk mixed-layer temperature (the desired parameter for naval operations) depending on the amplitude and phase of the SST cycle at the time of measurement. Since the diurnal SST response is a strongly nonlinear function of the wind speed, smooth variations in the synoptic wind field can produce sharp horizontal SST gradients that might be misinterpreted as the thermal signature of ocean frontal features.
158 Propagation II: mathematical models (Part Two) 5.3 Shallow-water duct models
Acoustic propagation in shallow water is dominated by repeated interactions with the sea floor. Generally, shallow water is restricted to consideration of the continental shelves with depths less than 200 m. Detection ranges in shallow water are severely limited both by the high attenuation that results from interaction with the bottom and by the limited water depth, which will not support the long-range propagation paths available in deep water. In a recent book, Katsnelson and Petnikov (2002) discussed results from acoustical measurements made over the continental shelves of the Barents Sea and the Black Sea.
Determination of source location (bearing, range and depth) can be affected by the horizontal refraction caused by repeated boundary reflections over a sloping bottom. Doolittle etal. (1988) experimentally confirmed the horizontal refraction of CW acoustic radiation from a point source in a wedge-shaped ocean environment. A striking graphical presentation of a 3D ray trace in a complicated wedge-shaped ocean environment is illustrated in Figure 5.4 (Bucker, 1994). This ray trace vividly displays the effects of horizontal refraction caused by a sloping bottom boundary. The source (denoted by an asterisk) appears in the background. Such horizontal refractive effects complicate the determination of bearing angles between sources and receivers. Consequently, sonar detections made against targets in shallow water may need to be corrected for horizontal refraction.
It is convenient to categorize sound-speed profiles into generic groupings to facilitate subsequent discussions of shallow-water propagation. Assuming
linearly segmented profiles, three groupings of profiles can be distinguished by the degree of segmentation: Category I - linear, Category II - bilinear and Category III - multiply segmented. Subgroupings (labeled A, B,...) can be formed to further distinguish these profiles according to the sound-speed gradient.
Linear profiles consist of single segments that can be further distinguished according to their gradient as: I-A - positive-gradient (dc/dz > 0), I-B -negative-gradient (dc/dz < 0) or I-C - isovelocity (dc/dz = 0), where c is the speed of sound and z is the depth (measured positive downward).
Bilinear profiles consist of two segments and can be formed in two ways in the ocean. If a positive-gradient overlies a negative-gradient (II-A), then a surface duct is formed in the upper layer above the sound-speed maximum. If a negative-gradient overlies a positive-gradient (II-B), then a sound channel is formed at the juncture of the two segments (i.e. at the sound-speed minimum).
Multiply segmented profiles consist of three or more segments and can assume a variety of forms. However, the most common manifestation of this type occurs when a surface duct overlies a channel (III-A). Other manifestations typically involve multiple channels (III-B).
This classification system provides a convenient method for describing the general distribution of sound-speed profiles in shallow-water environments (with depths < 200 m). For example, Reise and Etter (1997) used this classification system in a sonar trade study that examined representative shallow-water profiles from the Pacific and Atlantic oceans, and the Mediterranean and Arabian seas (Table 5.1). Based on this small but representative sampling, the most common occurrence (42 percent of all profiles examined) was the bilinear profile with a surface duct (II-A). This form was almost twice as likely to occur in summer as in winter (64 versus 36 percent, respectively). The next most common occurrence (23 percent of all profiles examined) was the linear positive-gradient profile (I-A). This form occurred exclusively in winter (100 percent). No isovelocity cases (I-C) were encountered in the study. Multiply segmented forms (III-A and III-B) represented 15 percent of the profiles examined and were three times more frequent in summer than in winter (75 versus 25 percent). The linear negative-gradient profile (I-B) represented 12 percent of the profiles examined and occurred exclusively in summer (100 percent). Finally, the bilinear sound channel (II-B) represented 8 percent of the profiles examined and occurred as frequently in summer as in winter (50 percent each). It should be noted that these results might not be representative of every shallow-water region in every season.
Understanding and predicting acoustic sensor performance in shallow water is complicated by the relatively high temporal and spatial variability of the
Table 5.1 Categorization of shallow-water sound-speed profiles. Also shown are the probabilities of occurrence as determined by Reise and Etter (1997)
Group Linear Bilinear Multiply segmented designation/
subgroup positive I-B, negative I-C, isovelocity II-A, surface II-B, sound III-A, duct over III-B, multiple gradient (%) gradient (%) duct (%) channel (%) channel (%) channels (%)
Relative occurrence Summer Winter
Was this article helpful?