CCAMLR

Since 2002, the Working Group on Ecosystem Monitoring and Management has been undertaking a work program to develop a management procedure for krill fisheries. To date, this has primarily focussed on subdividing the region-wide krill catch limit in the southwest Atlantic into small-scale management units through a number of workshops. Computer simulations are being used by WG-EMM to evaluate the costs and benefits to achieving conservation and fishery objectives of the six different strategies for undertaking the subdivision. This paper further elaborates the Ecosystem Productivity, Ocean, Climate modelling framework. In particular, it outlines the general considerations for designing a simulation environment for evaluating the strategies for subdividing the krill catch limit and, on the basis of these considerations, provides an update of the conceptual structure of EPOC and outlines the templates developed for EPOC to facilitate this work.

A useful way of testing assessment methodology and related decisions is through the framework of management strategy evaluation and the simpler form of assessment strategy evaluation. Such a framework uses an operating model of the system to generate data against which the assessment model, or the entire management system can be evaluated against a number of suitably chosen performance measures. This paper defines the term Assessment Strategy Evaluation (ASE) and provides a practical framework for carrying out such an evaluation. The framework is implemented in the R programming language and the technical details and decisions made in the formulation of this framework are included in this paper. Initial testing of this framework with respect to the Patagonian Toothfish Dissostichus Eleginoides Fishery in Division 58.5.2 is described.

An updated version of the Spatial Multi-species Operating Model (SMOM) of krill-predator-fishery dynamics is described. This has been developed in response to requests for scientific advice regarding the subdivision of the precautionary catch limit for krill among 15 small-scale management units (SSMUs) in the Scotia Sea, to reduce the potential impact of fishing on land-based predators. SMOM has been revised from the original version presented in Plagányi and Butterworth (2006a) in three main ways: 1) Accounting for seasonality; 2) Explicitly modelling fish and whales in addition to penguin and seal predators; and 3) Addition of an alternative movement model based on the results from the OCCAM model.
This modelling framework provides an example of a method for bounding some of the uncertainty associated with multi-species models used for management. Results are presented as probability envelopes rather than in point estimate form, giving a truer reflection of the uncertainty inherent in outcomes predicted on the basis of multi-species models as well as highlighting how such probability envelopes could be narrowed given improved data on key parameters such as survival. Results are useful for evaluating different spatial allocations of krill catches. An example is given of how such a framework can be used to develop a management scheme which includes feedback through management control rules.

We modelled the ray population at South Georgia using a surplus production model implemented in a Bayesian framework. Catch and CPUE data were reconstructed for the time series 1985 – 2006 from a variety of sources, and included the consideration that since 2004 the practice of cutting rays off lines at the water surface should have increased survivorship of discarded rays, although this survivorship has been found to decrease as depth of fishing increases. Our analysis is concentrated on catches deeper than 800m, and we have assumed that the majority of catches were of the Amblyraja variant species previously described.
The assessment indicated that the data were informative with respect to carrying capacity but not with respect to r, the intrinsic rate of growth of the ray population. We chose a conservative, low value of r thought to be representative of long-lived rajids, but conducted sensitivity tests with higher and lower values. The assessment suggests that the level of mortality currently experienced by the population is sustainable, and that currently the population is well above Bmsy. Whilst current catches appear not to be significantly impacting the ray population, trends to deeper water fishing and increasing numbers of autoliners (whose catch rates of rays is greater than Spanish gear longliners) should continue to be observed closely. Future analyses should make use of the increasing accuracy and duration of CPUE data and the results of an ongoing mark-recapture experiment for rays at South Georgia.

Three Operating Models (OMs) reflecting an “Optimistic”, “Intermediate” and a “Pessimistic” current status for the toothfish resource in the Prince Edward Islands region are developed which take account of the different selectivities of past longline and pot fisheries. These models are used for trials of a candidate Management Procedure (MP) which could provide future TAC recommendations for this resource. The MP uses two data sources: the recent trend in longline CPUE and the mean length of the catches made. This MP provides encouraging performance over the wide range of scenarios considered, increasing catches substantially if the resource is above MSYL, while increasing more slowly if the resource is heavily depleted while nevertheless securing stock increase with high probability.

An updated Antarctic toothfish (in the Ross Sea) stock assessment was done by means of the TISVPA model (2006 data on catch-at-age and cpue and the data on tag recaptures were included). All the three sources of information taken separately (and together) indicate historical increase in stock biomass, perhaps, due to development (broadening) of the fishery. The results show that, according to the Constable & de la Mare (1996) decision rules, annual catches of 23 000 tonnes per year are suitable.

In 2006 the CCAMLR scientific committee noted several features of exploratory Dissostichus fisheries in the southern Indian Ocean (58.4) which gave cause for concern as to the status of the resource in this area, and the lack of a scientific basis for setting catch limits in these areas (SC-CAMLR XXV, paragraphs 4.184-4.192).
In its management advice for this and other exploratory fisheries, the Scientific Committee requested urgent consideration by Members on of methods for collecting of data and of assessing these stocks. We outline a methodology to assess the BANZARE fishery including:
• Identification of grounds through analysis of spatial pattern in catch and effort;
• Construction of standardised CPUE series for each ground;
• DeLury/Leslie analysis of standardised CPUE of Dissostichus and major bycatch groups to provide initial estimates of biomass and rates of depletion in each ground until such time as they can be improved; and
• Estimation of the proportion of the stock which can be harvested (?) to satisfy the CCAMLR decision rules.
• Analysis of relative catch rates of bycatch groups and target species
The input data for this assessment is the C2 catch and effort data held by CCAMLR for the fishery in this Division since it began in 2003/04.
The results presented in this paper indicate that this methodology will produce an improved understanding of the trends in this fishery and greatly assist in developing management advice for the stocks of Dissostichus and major bycatch species on BANZARE Bank.
We welcome recommendations from WG-SAM as to the most productive way forward to an assessment for this Division to be considered at CCAMLR XXVI.

The incorporation of “effective sample size” (ESS) in integrated assessments is an approximate but simple way of modelling the distribution of catch-at-age or catch-at-length frequencies using a multinomial likelihood when there is extra-multinomial heterogeneity in age class or length class frequencies. The ESS applied within the definition of the negative log-likelihood contribution to the objective function in CASAL determines the implicit weight given to the commercial catch-at-age or catch-at-length frequency data relative to the other types of data used in integrated assessments of toothfish stocks. An appropriate and accurate estimate of the ESS for catch frequency data for each fishery and fishing year is therefore important for such assessments and this issue is studied using simulation. Between-haul heterogeneity within fishing year was simulated using samples from the Dirichlet-multinomial (D-M) distribution with marginal class probabilities generated using a simple age-structured model incorporating fishing selectivity. Either between-year “process” or “systematic” error in these probabilities was also generated by varying one of the selectivity function parameters across years randomly or linearly, respectively. Five alternative methods of estimation of effective sample size were compared using this simulation model. Two existing methods are based on the lack-of-fit of predictions of class probabilities using aggregate year-level frequencies. The other three estimators use the haul-level frequencies, including a method based on a conditional profile maximum likelihood estimate of the D-M dispersion parameter. This last method generally gave the best estimator of an ESS that is appropriate for haul-level heterogeneity with another of the haul-level methods giving similar estimates. The year-level methods gave very inaccurate estimates of this ESS when process error variance was set to zero with relative mean square error an order of magnitude worse than the best two haul-level methods. When process error was incorporated one of the year-level methods gave reduced estimates of ESS. An appropriate distributional model that incorporates process error in addition to haul-level heterogeneity while giving a marginal variance relationship which allows an ESS to be defined does not appear to be available so heuristic arguments and simulation results are used to discuss the issue of estimating ESS in the presence of process error. It is shown that care should be taken to avoid year-to-year model lack-of-fit due to systematic deviations in observed versus predicted class frequencies being mistaken for process error and used to reduce the ESS inappropriately.

In this paper, we address a number of aspects of the model inputs and parameters of the Antarctic toothfish stock assessment for the Ross Sea fishery. In particular we review catch history, length-weight relationships, catch-at-length and catch-at age. In addition, we report some preliminary model runs that investigate the sensitivity of the 2006 stock assessment to changes in these model inputs and parameters.
Tree-regression methods were used to investigate the areal structure of the length distribution of Antarctic toothfish. While tree-regressions suggested strong evidence of a high degree of small-scale areal complexity, we were unable to provide a stratification that resulted in improved or consistent patterns in length frequencies over the duration of the fishery. Including terms for nation, vessel, or vessel type did not provide any additional information as these tended to be highly correlated with the location variables.
The catch and CPUE indices for the Ross Sea Antarctic toothfish fishery were updated, as are some modelling parameters, and methods for calculating age- and length-frequencies. Most of these changes did not have a significant impact on the assessment results.
We also provide an update of the numbers of fish scanned at length by New Zealand vessels, and the numbers of tagged fish recaptured. Inclusion of observations of the 2006 fish recaptured in 2007 had the greatest impact on the assessment model results. Dunn et al. (2007) noted that the locations of the 2007 recaptures were highly aggregated and were mostly located on four key locations in the Ross Sea, and most had moved only short distances. This confirms the concern that the key uncertainty underlying the current model is the impact of movements and spatial structure in the Antarctic toothfish population. In particular, the level and nature of the bias from non-homogeneous mixing assumptions of tagged fish.

Descriptive analyses of the toothfish tagging programme carried out in Subareas 88.1 and 88.2 since 2001 are updated. The paper provides a preliminary update of the tag-release and tag-recapture data that were presented at the October 2006 meeting of WG-FSA by including data from New Zealand vessels and preliminary data for other vessels that fished in 2007.
Release and recapture data that previously were unavailable for about half of the non-New Zealand vessels for 2004 are now available and described in this paper for the first time. Overall, a reported total of 12 177 Antarctic toothfish have been released and 333 recaptured, and 859 Patagonian toothfish released and 29 recaptured since 2001.
The number of tags recaptured in the Ross Sea in 2007 by New Zealand vessels was the highest annual recapture to date and double the number caught in 2006, although the nature of these recaptures suggests that assumptions of homogeneous mixing may need to be investigated. For the first time, long distance movements of Antarctic toothfish were observed from fish tagged by fishing vessels. A total of four fish moved significant distances from the slope fisheries in SSRUs 88.1H, 88.1I, and 88.1K to Terra Nova Bay in SSRU 88.1J. There was also some evidence that more fish are recaptured after a longer time at liberty on the slope than in the North.
However, we note that data from the 2007 season for the non-New Zealand vessels were incomplete at the time of this analyses and will need to be updated in future analyses.