ROGER P. HEWITT and DAVID A. DEMER, Antarctic Ecosystem Research Group, Southwest Fisheries Science Center, La Jolla,California 92038
The primary objectives of the bio-acoustic sampling program were to map the mesoscale (10s of kilometers) dispersion of krill ( Euphausia superba ) near the South Shetland Islands, to estimate their biomass, and to determine their association with predator foraging patterns, water-mass boundaries, spatial patterns of primary productivity, and bathymetry. Secondary objectives were to describe cross-sections of volume backscattering strength along transects through Admiralty Bay, across the shelf break to the north and south of the South Shetland Islands, and across the Antarctic Convergence in Drake's Passage. In addition, in situ target strength (TS) measurements of individual zooplankton were made; these data will be used to develop or enhance TS models for various macrozooplankton and nekton.
Acoustic data were collected using a multifrequency echo sounder (Simrad EK500) configured with downlooking 38-, 120-, and 200-kilohertz transducers mounted in the hull of the ship. Pulses were transmitted every 2 seconds at 1 kilowatt for 1.0 millisecond duration at 38 kilohertz, 1.0 millisecond duration at 120 kilohertz, and 0.6 millisecond at 200 kilohertz. Geographic positions were logged every 60 seconds. Ethernet communications were maintained between the EK500, a UNIX workstation, and a Windows-95 workstation. The UNIX workstation was used for system control, data logging, and data postprocessing, including interpretation of echograms, echo-integration, and TS analyses. Processed data were passed to the Windows-95 PC for gridding and contouring of integrated volume backscattering strength. System calibrations were conducted before and after the surveys using standard sphere techniques while the ship was at anchor in Martel Inlet, King George Island, and off Cape Shirreff, Livingston Island.
For the purposes of generating distribution maps, bottom return, surface turbulence, and system noise were eliminated from the echograms. The remaining volume backscatter was attributed to biological scatterers and was integrated over depth (from 15 to 500 meters for the 38 kilohertz data, from 15 to 225 meters for the 120 kilohertz data, and from 15 to 175 meters for the 200 kilohertz data) and averaged over 0.1-nautical-mile (185-meter) distance intervals. A 3015 cell grid was imposed on the survey area; adjustments were made for the scaling discrepancy between longitude and latitude; and integrated volume backscattering values were interpolated at grid nodes using Krigging methods and assuming a linear variogram model.
For the purpose of generating an estimate of krill biomass density, all volume backscattering at 120 kilohertz was assumed to be from krill. Integrated volume backscattering strength per unit sea surface area was converted to estimates of krill biomass density by applying a factor equal to the quotient of the weight of an individual krill and its backscattering cross-sectional area, summed over the sampled length frequency distribution for each survey (Hewitt and Demer 1993). Mean density for the Elephant Island area (as indicated by the box on figure 1) was estimated by treating the mean biomass density on each of the nine transects as an independent estimate of the mean density over the survey area (Hewitt and Demer 1993).
Survey A (27 January to 10 February, figure 1) was conducted to map the mesoscale dispersion of krill near the South Shetland Islands and to estimate the biomass of krill in a 12,600-square-nautical-mile area centered on Elephant Island. Each survey consisted of approximately 2,000 nautical miles of transects conducted between 106 stations. Station work included a conductivity-temperature-depth (CTD) cast and an Isaacs-Kidd Midwater Trawl (IKMT) plankton tow. Mean krill density was calculated from nine north-south transects with 15-nautical-mile spacing between lines. Survey D (17-24 March) was greatly curtailed because of severe time constraints resulting from an engine failure on the survey vessel.
A discontinuous band of high integrated volume backscattering strength was mapped on the north side of the South Shetland Islands (figure 1). This band tended to be north of the frontal zone running parallel to the archipelago, widening and spreading onto the shelf near Elephant Island. Other areas of high integrated volume backscattering strength were mapped on the shelf near Livingston Island and in deeper water in Bransfield Strait. IKMT samples indicated that the krill north of the South Shetland Islands were adults in advanced maturity stages, whereas the krill caught on the island shelf and in the Bransfield Strait were a mixture of 1-year-old juvenile and adult krill.
Krill biomass, estimated from the acoustic data for the area outlined in the box on figure 1, was the highest estimated since 1992 (table).
Figure 2 is a compressed echogram describing a cross-section of volume backscattering strength from Station A44 through Station A100 (bold transect line in figure 1). A notable feature is the krill layer between 25 and 100 meters depth extending from halfway between Stations A44 and A45 to Station A47; a large aggregation of surface feeding fin whales was observed in this area. Also notable is the approximately 100-meter-thick layer of krill rising toward the surface as sunset approached and abruptly ending at 200 meters water depth. The echogram also suggests a downward migration of krill prior to sunrise on the south side of the South Shetland shelf and scattered krill swarms across Bransfield Strait and onto the Antarctic Peninsula shelf.
Hewitt, R.P., and D.A. Demer. 1993. Dispersion and abundance of antarctic krill in the vicinity of Elephant Island in the 1992 austral summer. Marine Ecology Progress Series, 99(1-2), 29-39.