CCAMLR

The diet of the white-chinned petrel at Bird Island, South Georgia was studied during the chick-rearing period in 1986 by quantitative analysis (by weight, frequency of occurrence and number of individuals) of regurgitated or lavaged adult stomach contents. Antarctic krill was the most important constituent of the diet, comprising over 90% of prey items and forming 47% of the diet by weight; fish and squid occurred In 67% and 35% of samples and formed 33% and 19% of the diet by weight, respectively. Decapods, amphipods and salps occurred in a few samples. The fish were mainly lanternfish (Myctophidae) of 8 species (chiefly in the genera Electrona and Gymnoscopelus) , forming 80% by numbers and 52% by mass of fish prey and the notothenioid Patagonotothen guntheri (14% by number and 35% by mass). Of the squid species taken, the ommastrephid Martialia hyadesi comprised 57% by number and 52% by mass and the gonatid Gonatus antarcticus 14% by number and 42% by mass. These dietary data confirm white chinned petrel as the most important avian consumer of fish and squid at South Georgia (and the third most important consumer of krill). In 1986 the notothenioid fish were probably obtained via commercial fishing operations but the myctophids and squid were probably live-caught, most likely at night. Meal size increased rapidly until chicks were three weeks old and then remained constant until the chicks were within 10 days of fledging, when it decreased. Meal delivery rate was high (one per day) for young chicks (1-10 days old) and thereafter fluctuated between 0.56 and 0.88 meals per day until close to fledging, when it was halved. These provisioning rates, and the proportion of krill in the diet, are higher than those recorded previously at South Georgia and Indian Ocean sites probably reflecting high local availability of krill at South Georgia in 1986. In many respects white-chinned petrels at South Georgia are intermediate ecologically between prions and albatrosses, although specialised in their extensive consumption of myctophids. Because krill and all their main fish prey are currently the targets of substantial commercial fishing and the main squid prey (Martialia) is a potential target, the role and status of white-chinned petrels is of additional importance.

Published in J. exp. Biol. 000: 000-000 (1994)

A method is described for calculating indices of sea ice distribution from digital data distributed by the National Snow and Ice Data Centre, Colorado, USA. Data consist of an image of the Antarctic, made up of 312x332 pixels, each having a dimension of 25x25km, for every day in the year. Primary data of the distance of sea ice from CEMP sites, the northerly position of the sea ice edge, and the percentage ice cover in subareas, are entered into the CCAMLR database. Indices of ice cover are calculated from these primary data. The techniques used are validated against alternative known data.

The potential of multi-frequency acoustic technology for quantifying fish and plankton is widely recognized. At a U.S. GLOBEC workshop in 1991, one of the considered problems was measuring macrozooplankton and micronekton. It was concluded that in order to survey animals in the size range 0.5-5 cm, it would be expedient to use at least three and perhaps as many as 8-10 frequencies in the range 38-420 kHz. Here, a specific design is evaluated: ten more or less standard, approximately logarithmically spaced frequencies are chosen. Each frequency represents the resonant condition of a circular piston transducer with 10-deg beamwidth. The performance of this system is assessed through the maximum detection range of single targets, assuming transmission near the cavitation limit but consistent with dynamic strength and heat generation in ceramic elements, isotropic ambient noise, and detection threshold of 20 dB. Target strength is treated as a parameter, with investigated values from -130 to -50 dB. Performance assessment for a volumetric distribution of scatterers is similarly treated.

Design of sondes for in situ measurement of zooplankton or other scatterers requires choosing among alternative transducer geometries. This contribution addresses the problem of choosing between cylindrical and circular piston transducers by comparing the performance of the two according to the principle that the acoustically active areas be equal. Computations are performed with the actual dimensions of six fabricated cylindrical transducers, whose beam patterns have been measured by the manufacturer at a total of 11 frequencies spanning the range 27-710 kHz. Nominal power levels assigned to the cylindrical transducers are also used for both transducer types. Comparison of theoretically computed beam patterns with measurement gives confidence in the radiation model, which is used to compute the directivity index and on-axis sensitivity loss due to curvature of the cylindrical transducers, referred to as the curvature loss. Under identical conditions of excitation, isotropic ambient noise, and detection threshold of 20 dB, the active sonar equation is exercised to estimate the maximum detection range of both single targets and multiple targets distributed throughout the sampling volume. In every single case, the performance of the equal-area circular piston is superior to that of the corresponding cylindrical transducer. This is directly attributable to differences in directivity index and curvature loss. Other, pragmatic considerations argue for the choice of the circular piston transducer over the cylindrical transducer. Three problems requiring future treatment are identified.