Robert H. Bourke

Sea ice thickness distribution in the Arctic Ocean

Abstract Data from the unclassified literature were reviewed to determine the regional and seasonal distributions of sea ice thickness, pressure ridging statistics, frequency of occurrence of polynyas, and keel/sail height ratios. Seasonal and regional maps and histograms of these properties were constructed. The majority of the data were obtained from submarines equipped with a narrow-beam, upward-looking sonar. As determined from an analysis of 17 submarine cruises, the overall mean thickness of Arctic sea ice above 65° N, including both deformed and undeformed ice, is 2.9 m with a standard deviation of 1.8 m. The overall seasonal mean ranges from approximately 2.4 m in spring to 3.3 m in summer. Local mean ice drafts ranged from less than 1 m near the marginal ice zone to greater than 7 m to the north of the Canadian Archipelago. Histograms of sea ice draft reflect a bimodal distribution in winter and spring, an effect of the presence of thin first year ice. Due to ice melt in summer and autumn only a single mode of much thicker multiyear ice is observed.

Precipitation over the Pacific Ocean, 30°S to 60°N

Abstract By using present weather observations taken by ships and relating them to a given amount of precipitation, new estimates of oceanic rainfall for the Pacific Ocean between 30°S and 60°N have been derived. Satellite microwave measurements and Taylors (1973) island analysis support our findings. Annual and quarterly rainfall maps, drawn from our estimates, agree with other modem, land-derived values, but provide greater detail. Between the equator and 60°N, the annual depth and volume rainfall totals are 1282 mm and 1.16×105 km3, respectively. Maps of amplitude and phase show that most of the rainfall north of 28°N occurs in winter, while maximum rainfall occurs in July and August in the tropics. Diurnal rainfall, studied at selected locations, is at a minimum at noon in all but the western pan of the North Pacific. Here there is no distinct minimum.

Contour mapping of Arctic basin ice draft and roughness parameters

A data base of ice draft and roughness parameters has been constructed for selected portions of the Arctic Ocean based upon analysis of under-ice draft distribution data acquired by inverted echo sounder systems on submarines. From the voyages of 12 submarines which traversed the Alaskan, Canadian, and central Arctic regions of the Arctic Ocean during the summer and winter seasons, a series of mean ice draft and deep-draft keel statistics was calculated for 50-km segments along each submarine track. Contour maps of the mean ice draft, its standard deviation, the mean keel draft, and the spatial frequency of ice keels were constructed. They show that the greatest ice drafts, the roughest ice, and the greatest number of deep-draft keels are found off the north coasts of the Canadian Archipelago and Greenland due to ice convergence on these land barriers.

Acoustic travel‐time perturbations due to shallow‐water internal waves and internal tides in the Barents Sea Polar Front: Theory and experiment

During August 1992, a combined acoustics/physical oceanography experiment was performed to study both the acoustical properties and the ocean dynamics of the Barents Sea Polar Front in the region near Bear Island. Oceanographic observations from shipboard hydrography and moored sensors allowed the construction of the internal wave frequency spectrum for the area. A rapidly sampled tomographic section from a 224‐Hz, 16‐Hz‐bandwidth acoustic source to a 16‐element vertical receiving array enabled the monitoring of travel‐time fluctuations over the internal wave frequency band. To describe the measured acoustic fluctuations, theoretical expressions have been developed for the travel‐time variances which are functions of the internal wave oceanographic field, the local acoustic propagation characteristics, and the acoustical system’s properties. Both ray and mode theory expressions are generated, as the experiment was performed in shallow water and both ray and mode arrivals were resolvable. Comparison of the...

Precipitation over the Atlantic Ocean, 30°S to 70°N

Abstract New estimates of rainfall over the Atlantic Ocean between 30°S to 70°N have been constructed based an a technique that uses the present weather observations taken by ships. Annual and quarterly rainfall maps are presented. Between the equator and 60°N, the average annual rainfall depth is 1034 mm and the annual volume is 3.93 × 104km3. Compared to the Pacific, the Atlantic is significantly drier and has less extreme values. Maps of amplitude and phase show that most of the North Atlantic cast of 60°W experiences a inter peak rainfall. The South Atlantic experiences its peak rainfall in the Southern Hemisphere summer.

The Barents Sea Polar Front in summer

In August 1992 a combined physical oceanography and acoustic tomography experiment was conducted to describe the Barents Sea Polar Front (BSPF) and investigate its impact on the regional oceanography. The study area was an 80 × 70 km grid east of Bear Island where the front exhibits topographic trapping along the northern slope of the Bear Island Trough. Conductivity-temperature-depth, current meter, and acoustic Doppler current profiler (ADCP) data, combined with tomographic cross sections, presented a highly resolved picture of the front in August. All hydrographic measurements were dominated by tidal signals, with the strongest signatures associated with the M2 and S2 semidiurnal species. Mean currents in the warm saline water to the south of the front, derived from a current meter mooring and ADCP data, were directed to the southwest and may be associated with a barotropic recirculation of Norwegian Atlantic Water (NAW) within the Bear Island Trough. The geostrophic component of the velocity was well correlated with the measured southwestward mean surface layer flow north of the front. The frontal structure was retrograde, as the frontal isopleths sloped opposite to the bathymetry. The surface signature of the front was dominated by salinity gradients associated with the confluence of Atlantic and Arctic water masses, both warmed by insolation to a depth of about 20 m. The surface manifestation of the front varied laterally on the order of 10 km associated with tidal oscillations. Below the mixed layer, temperature and salinity variations were compensating, defining a nearly barotropic front. The horizontal scale of the front in this region was ∼3 km or less. At middepth beneath the frontal interface, tomographic cross sections indicated a high-frequency (∼16 cpd) upslope motion of filaments of NAW origin. The summertime BSPF was confirmed to have many of the general characteristics of a shelf-slope frontal system [Mooers et al., 1978] as well as a topographic-circulatory front [Federov, 1983].

Sea level rise in the Arctic Ocean

About 60 tide-gauge stations in the Kara, Laptev, East-Siberian and Chukchi Seas have recorded the sea level change from the 1950s through 1990s. Over this 40-year period, most of these stations show a significant sea level rise (SLR). In light of global change, this SLR could be a manifestation of warming in the Arctic coupled with a decrease of sea ice extent, warming of Atlantic waters, changes in the Arctic Ocean circulation, and an increase in coastal erosion and thawing of permafrost. We have analyzed monthly mean sea level data and assessed the role that different factors may play in influencing the process of sea level change in the Arctic Ocean. Analysis of the observational data and model results shows that changes in the patterns of wind-driven and thermohaline circulation may account for most of the increase of sea level in the Arctic Ocean and their cumulative action can explain more than 80% of the sea level variability during 1950–1990.

The Jan Mayen Current of the Greenland Sea

Oceanographic measurements acquired by the USNS Bartlett during September 1989 in and north of the area of the Jan Mayen Current (JMC) show that, in terms of upper layer baroclinic flow, about half of the JMC is a wide meander in the East Greenland Current (EGC) and about half continues eastward to close the Greenland Gyre (GG) on the south. This phenomenon has been charted in the past but has elicited no comment. Surface drifters and a numerical model corroborate this behavior. At deeper depths (>100 m) the meander dissipates with the flow becoming more easterly. A drifting float at 500 m depth confirms that the flow is relatively undeviated toward the east ultimately merging with the northward flowing Norwegian Atlantic Current. Near the GG the fresh water content of the water column is low in the upper layers but high in the middepths. Where there is a substantial content of EGC water (e.g., in the JMC), the converse is usually true. The FWC of the GG computed for the period 1953–1966 was smaller by a factor of 2 in both layers than in 1989, indicating that 1989 is in another period of low salinity comparable to that of 1968, the period of the Great Salinity Anomaly described by Dickson et al. (1988).

Noise source level density due to surf. I. Monterey Bay, CA

Ambient noise measurements made in Monterey Bay, CA, in 1981 were reduced by estimations of wave-breaking noise and the residual noise was combined with modeled transmission loss (TL) to estimate the spectral source level of surf-generated noise. A Hamilton geoacoustic model of the coastal environment was derived and used in a finite-element parabolic equation propagation-loss model to obtain TL values. Estimates of both the continuous, or local, and discrete components of wave-breaking noise intensity were subtracted from the total measured noise field to determine the contribution due to surf only. Surf breaking on a uniform 12.5-km linear section of beach near Ft. Ord was found to be the dominant source of surf-generated noise. Estimated noise source level densities for heavy surf at Ft. Ord beach varied from 138 dB ref. 1 /spl mu/Pa Hz/sup -1/2/ m at 1 m from the source at 50 Hz to 107 dB at 1 kHz, with a slope of about -5 dB per octave. Although these results must be considered as preliminary, since they are based on a small number of measurements, they may he useful for prediction of ambient noise in other littoral regions.

Estimation of bottom scattering strength from measured and modeled mid-frequency sonar reverberation levels

Hamilton-type geoacoustic models were developed for Area Foxtrot, a shallow water test bed south of Long Island, for emerging active sonar systems where the surface sediment type is highly spatially variable. Reverberation levels (RL) were modeled using the finite-element parabolic equation (FEPE) propagation model to augment the generic sonar model (GSM) propagation model because the bottom loss model in GSM did not estimate transmission loss (TL) accurately in shallow water. FEPE estimates reveal that there is a greater than 15 dB difference between TL for sand and that for silt-day sediments in Area Foxtrot. The comparison between modeled RL and measured RL (from a 1991 active sonar exercise) enabled bottom scattering strength kernels to be developed for Area Foxtrot. Bottom scattering strength was found to be a function of sediment type. Hard sand sediment has a bottom scattering strength which obeys Lamberts law (sin/sup 2/ /spl theta/) while that of silt-clay sediment is consistent with sub-bottom volume scattering (sine). The RLs in Area Foxtrot are azimuth-dependent and are a function of TL and bottom scattering strength (and hence bottom sediment type). Sonar beams steered towards the hard sand show higher RLs than for silt-clay, and knowledge of the sediment type and its spatial variation must be known to model RL accurately. A method to determine sediment type using measured RLs and RL slopes is given.