CCAMLR

This paper investigates the influence of mixing of fish, and the uneven distribution of tag placements and recapture effort, on bias in the Petersen estimate. It does so by constructing a linear model of the South Georgia toothfish fishery, simulating fish movements within this system and overlaying various combinations of tagging and recapture effort to investigate bias. The fishable grounds around South Georgia were divided into 77 very small scale boxes lying along the 1000m contour. The uneven distribution of animals was simulated by adjusting an average movement rate downwards when animals encountered a high CPUE box and upwards in a low CPUE box so that they were retained in high CPUE boxes. The model incorporates the facility for releases by box over a number of years.
The model performed as expected with test situations. It produced a near-perfect estimate of stock size when there was an ideal distribution of tags and/or fishing effort; by ideal we mean that either tagging or fishing effort was in direct proportion to CPUE. When both tagging and fishing effort were non-ideal, eg when effort was concentrated away from tag concentrations, or overly concentrated in them, the Petersen estimator either over-estimated or underestimated (respectively) the true population size. When run on the real tag release data, and using CPUE from 2002-2004 and recapture effort in 2003 and 2004, the model indicated that the Petersen equation produced an under-estimate of true population size. Although we do not advocate using the magnitude of the estimated bias to correct the tagging estimate made last year for 48.3, we do conclude that the particular distribution of tag releases and recapture effort at South Georgia is likely to lead to an under-estimate of the true population size rather than an over-estimate of it.

ASPM have been proposed and applied for Patagonian toothfish stock assessment at CCAMLR Subarea 48.3. Last results obtained from this model, discussed at WG-FSA (2004), do not show acceptable fit with standardized CPUE series and observed length proportions in the catches.
In this paper, we discuss some of the problems related with available CPUE data and estimate new vulnerability patterns to produce a good fit of the model both, to CPUE series and proportion-at-length data from CCAMLR dataset.

Quantifying the catch rates and biomass of by-catch species on CCAMLR’s fishing grounds is an essential component of the assessment advice prepared by WG-FSA. However, such analyses are problematic because the CCAMLR datasets are incomplete and have a high occurrence of ‘missing catch values’.
A method to treat ‘missing catch values’ using estimates derived from the mean weights of by-catch species by fishing gear, region and period is outlined. This method would improve the consistency of the CCAMLR datasets and allow quantitative analyses of by-catch to be based on the best, scientific data available.

Information on methods aimed at mitigating incidental mortality resulting from fisheries interactions have been released in a variety of local, national and international media. Recent published reviews in the field of bycatch mitigation have typically had a species or fishing method focus. This report presents the results of the seabird component of a global review of mitigation methods aimed at reducing mortalities of protected seabirds, marine mammals and reptiles and corals due to interactions with fishing gear in New Zealand fisheries and fisheries that operate using similar methodologies. The application of these mitigation methods to New Zealand fisheries were assessed, recommendations for the fisheries management made, and areas for further research in New Zealand identified. Factors influencing the appropriateness and effectiveness of a mitigation device include the fishery, vessel, location, seabird assemblage present and time of year (i.e. season). As such, there is no single magic solution to reduce or eliminate seabird bycatch across all fisheries. Realistically a combination of measures is required, and even within a fishery there is likely to be individual vessel refinement of mitigation techniques in order to maximise their effectiveness at reducing seabird bycatch. Retention or strategic management of offal and discards are recommended as the most effect measure to reducing seabird bycatch in longline and trawl fisheries. Other recommended methods for both demersal and pelagic longlining include paired bird-scaring lines, line-weighting and night-setting (in some fisheries). Along with offal and discard management, paired bird-scaring lines and reducing the time the net is on (or near) the surface are likely to be the most effective regime at this point to mitigate seabird interactions with the warp cables and net respectively. However urgent investigation is needed into more effective measures at reducing seabird interactions with the trawl nets.

A new method for estimating illegal fishing effort is put forward. The results from this new method are similar to those of the Agnew and Kirkwood (2005) method, and this suggests that the current method is adequate under circumstances of low evasion and when good knowledge exists that zero observations reflect zero illegal fishing. The new method performs better in the case of zero detections and can potentially better handle the evasion of detection by illegal activity.
Both the new and the current method suffer from the fact that the observation method used directly affects the system. This is the prevention/detection problem, in which the greater the number of detections for a given level of illegal fishing, the more often the illegal fishers will curtail their fishing trips. This leads to a negative correlation between the amount of fishing and the estimated amount of fishing for a given number of illegal cruises.
As the number of illegal cruises increases, both the estimate and the average amount of illegal fishing increase. This gives some confidence that the method can produce results that have a degree of legitimacy. However, the range of actual fishing (in the simulation datasets) for a given estimated level of fishing is very large. This range of uncertainty increases as the evasion rate increases.
This research suggests that it would be possible to calculate a precautionary assessment of illegal fishing such that the actual number of illegal fishing days is less than, or equal to, the precautionary assessment with some given level of confidence (for example 80%).