CCAMLR

An adaptive approach was used to assess the Champsocephalus. gunnari stock in Area 48.3. Values of catch/effort by age group and by year of fishing were used in conjunction with abundance indices from trawl surveys in tuning the VPA.
The C. gunnari stock at the start of the 1990/91 season is estimated at 184 thousand tonnes.
Calculation of TAC was carded out keeping in mind the change to the fishing regime brought about by a change in mesh size (90 mm).
A TAC of 59.4 thousand tonnes for C. gunnari is advised for an optimal regime of fishing operations at a level of F0.1 = 0.65.

An adaptive approach was used to assess the Notothenia gibberifrons stock in Area 48.3. It is demonstrated that when tuning the VPA using abundance indices from trawl surveys it is imperative to estimate the catchability coefficient. This parameter was equal to 0.326 to 0.635.
An adaptive approach in this case involved using a target function in the form of the sum of square deviations in logarithms of observed and calculated biomass indices and resulted in an estimate of from 21.5 to 42.1 thousand tonnes for N. gibberifrons biomass in the 1989/90 season. It is thus possible to suggest a TAC of 1.9 to 2.7 thousand tonnes.
A similar approach is recommended for assessing the stock of other commercial species in this area.

The method of refining the natural mortality coefficient is based on the notion that the free component of the regression equation between the mean-weighted coefficient of natural mortality as defined by the VPA and fishing effort must be equal to zero if the natural mortality coefficient has been correctly identified and something other than zero if it has not been correctly chosen.
As a result of this mathematical modelling study using C. gunnari in Area 48.3 as an example, it was shown that the proposed method facilitates such a refinement. When a mistake is made in the choice of natural mortality coefficient, the estimate of the free component in the equation will be other than zero. A bias in the estimate can be positive if the assumed natural mortality coefficient is less than the “true" value and close to zero or slightly greater than zero if the assumed value is greater than the "true" one

Shnyute’s generalised cohort analysis of size composition in catches was used to estimate the Dissostichus eleginoides stock in Area 48.3. This made it possible to ignore the hypothesis of equilibrium in the stock and to examine the dynamics of its abundance and biomass.
At the start of the 1990/91 season the D. eleginoides stock is estimated to be 91.5 thousand tonnes.
It has been determined that for optimal results, fishing operations should be conducted at a level of F0.1. In accordance with such a policy, a TAC of 8.8 thousand tonnes is proposed for the 1991/92 season.
A sensitivity analysis demonstrated that the stock size and TAC were determined with a high degree of robustness

The seventh cruise of the RV Atlantida was a continuation of regular AtlantNIRO research into the status of Chaenichthyidae and Nototheniidae fish in Statistical Subarea 48.3 (South Georgia, Shag Rocks and Black Rocks). The minimum biomass of commercial fish species was calculated from the results of a census trawl survey carded out in April-May 1991. This paper contains data on distribution, biology and size composition of commercial fish species.

Genetic population structure of the mackerel icefish, Champsocephalus gunnari, in waters from around South Georgia, South Orkneys, and Heard Island (Australian Antarctic) was examined using allozyme electrophoresis.
Icefish were collected from Heard Island (N = 47; November 1990), South Georgia (N = 267; January/February 1991) and the South Orkneys (N = 44; January/February 1991) and subjected to starch gel electrophoresis. sixteen putative enzyme-coding loci were examined, of which 7 were polymorphic at the 0.95 criterion: glutamate oxaloacetate transaminase, GOT-I 2.6.1.1; α - glycerophosphate dehydrogenase, α-GPDH 1.1.1.8; isocitrate dehydrogenase, ICD 1.1.1.42; malate dehydrogenase, MDH-III 1.1.1.37; mannose phosphate isomerase, MPI 5.3.1.8; phosphoglucose mutase, PGM-II 2.7.5.1; superoxide dismutase, SOD 1.15.1.1).
Routine scoring of 358 fish at the 16 loci revealed below average levels of genetic variability, with the proportion of polymorphic loci, P = 0.175, mean heterozygosities per locus, HL = 0.058, and the effective number of alleles = 1.04 - 1.88.
The majority of alleles conformed to Hardy-Weinberg expectations, with significant deviations due exclusively to heterozygote deficiencies. Genetic differentiation was detected among samples, both within (South Georgia) r and between (South Georgia vs. Shag Rocks; South Georgia vs. South Orkneys) sea areas. Nei’s (1973) gene diversity analysis indicated that over 85% of the total gene diversity was due to the within-population component, with progressively higher between population values with increasing geographic separation. Samples from Heard Island were most genetically distinct ( I = 0.900 - 0.930; D = 0.072 - 0.106).
The apparent erosion of genetic variability may be related to abrupt changes in population size. Enhanced mobility of fish between sea areas was suggested by an overall reduction in genetic differentiation within and between South Georgia and Shag Rocks (compared with 1990), and the greater conformity to Hardy-Weinberg equilibrium. Genetic data support the notion of at least partial migration of C. gunnari between South Georgia and the South Orkneys, and possible speciation on a geographic scale.

Recently a longline fishery has developed for the Patagonian Toothfish, Dissostichus eleginoides, in the vicinity of Shag Rocks, South Georgia. Assessment of this stock is not possible by conventional methods because the bulk of the adult population may be dispersed over a wide area in deep water whereas the Shag Rocks’ shelf may be a nursery ground for juveniles. An assessment of the stock has been made by extrapolating the population parameters from trawl surveys to derive estimates for the whole stock.

Two similar methods for preparing aggregate length frequency distributions from survey data are discussed. The Mackerel Icefish, Champsocephalus gunnari at South Georgia is used as an example.

Discrepancy was noted between the age length key and modes in the length distribution of Ch. gunnari from the Hill Cove survey around South Georgia in January 1990. A revised age length key for this survey is proposed, based on data from a previous survey in the area. The revised age length key corresponds well to modes in the length distribution from the survey and to another key derived from samples collected during a subsequent survey in the same season.