WG-SAM-2021 and WG-EMM-2021 reviewed available information for krill assessment simulations in Subarea 48.1 using the Grym. Whilst no agreement on parameters was achieved at WG-SAM-2021, the Working Group agreed that an ensemble approach using multiple parameter value combinations could be used (WG-SAM-2021, paras 3.21-3.22.)

WG-EMM-2021 provided initial parameters for the assessment simulations noting that alternate parameters could be tested to compare (WG-EMM-2021, paras 2.32-2.33 and Table 1).

Here we present the results of model ensembles for values either provided directly to the CCAMLR e-group on ‘GYM/Grym assessment model development’, or calculated based upon data submitted to the e-group.  Code and outputs of the models are available on github (https://github.com/ccamlr/Grym_Base_Case /tree/Simulations).

A collaborative research program has been undertaken by Japan and South Africa since 2013 to enhance data collection and analysis in the subarea 48.6 under CM 21-02. Spain joined the proposal starting from 2018/19 fishing season in order to contribute to the data acquisition and to speed up the integrated assessments of the Antarctic toothfish (Dissostichus mawsoni) stock in this subarea (WG-FSA-18/34).

The continuation of the three-member research proposal for 2021/22 season is presented to ensure continuity of previous research activities. Data and investigations about the population structure and various demographic parameters of D. mawsoni using trotline (JPN and ZAF) and Spanish longline (ESP) gears, established tagging techniques, pop-up tags and genetic analysis will provide the basis for the development of spatial population models and assessments in support of management advice. An Integrated Stock Assessment (ISA) which takes into account the tag time series from southern research areas of Subarea 48.6 is going to be continually developed until the end of the 2023/24 season.

Based on suggestion from WG-SAM-2021 (report of WG-SAM-2021, para 8.4), the research plan have been revised as follows:

• Explaining the importance of understanding stock connectivity between research blocks in the area (seamounts versus continental shelf),
• indicating further details about how the stock structure will be represented in the planned CASAL assessment for the region,
• increasing the otolith sampling rate from “10 otoliths per 5 cm length bin” to “20 otolith per 5 cm length bin), and
• indicating minimum sampling requirements for by-catch species and designed to meet the research objectives.

The WG-SAM-2021 recalled that a structured fishing design was necessary to optimise tagging performance evaluation (report of WG-SAM-2021, para 8.4). However, as already described in 3(a) “Research survey/fishing design”, the area is not suitable to set a stratified sampling design as the fishing grounds with broad environmental characteristics such as a complex of seamounts, hills and ridges is expected to be small relative to the size of the research block. Therefore, no depth stratification is proposed in the current research plan.

The updated CCAMLR Research Plan – Research Proponent Self-Assessment can be found in Appendix 1.

Here, we provide diagnostic plots for the 2021 assessment model for Kerguelen Island EEZ Patagonian toothfish (Dissostichus eleginoides) presented in Massiot-Granier et al. (2021a), following the recommendations of WG-SAM-2015 (SC-CAMLR-XXXIV 2015 Annex 5). The stock assessment model is described in Massiot-Granier et al. (2021a), and a detailed description of the stock area, stock assessment methods and the stock assessment parameters are given in MassiotGranier et al. (2021b).  

This paper presents an updated stock assessment for the Patagonian toothfish (Dissostichus eleginoides) fishery of the Kerguelen Islands EEZ, CCAMLR Division 58.5.1. It includes commercial fishery data up to the end of the season 2019/2020 (end of August 2020). The updated assessment model is based on the best available data and includes abundance estimates from random stratified trawl surveys (POKER, 2006, 2010, 2013 and 2017), longline tag-release and tag-recapture data from 2007 to 2020 and priors and estimates on model parameters such as life traits (e.g. growth, maturity and natural mortality). It also uses commercial data that account for fish removals (fishing mortality and depredation) and support the estimation of year class strength (YCS) and selectivity functions.

Compared to the previous assessment that was presented to WG-FSA 19 (MassiotGranier et al., 2019), this assessment accounts for (1) updated catch and biological data (2019/2020), (2) updated age data from a new 4-year reading program, (3) updated priors on YCS and period of estimation (4) and updated depredation rate. All model runs were conducted using CASAL 2.30 2012-03-21 rev. 4648, that was agreed on by WG-SAM-14.

The updated model leads to a higher estimate of the virgin spawning stock biomass SSB₀ than in 2019 with an estimate of 233 130 tons (95% CI: 207 030 – 265 460). The estimate of the current SSB status of the stock is 68,8% (95% CI: 64,5% - 73,1%) and the current catch limit meets the CCAMLR decision rules.  

This paper presents an updated stock assessment for the Patagonian toothfish (Dissostichus eleginoides) fishery of the Crozet Islands, CCAMLR sub-area 58.6, with data up to the end of the season 2019/2020 (End of August 2020). This updated assessment model is based on the best available data and includes longline tag-release and tag-recapture data from 2007 to 2020 and priors and estimates on life traits model parameters (e.g. growth, maturity and natural mortality). It also uses commercial data that account for fish removals (fishing mortality and depredation) and support the estimation of their associated selectivity functions.

Compared to the previous assessment presented at WG-FSA 19 (Massiot-Granier et al., 2019, WG-FSA-2019/57 Rev. 1), this assessment accounts for (1) updated data (2019/2020) and (2) updated depredation rates. All model runs were conducted with the CASAL 2.30 2012- 03-21 rev. 4648, that was agreed on by WG-SAM-14.

The updated model leads to equivalent estimate of the virgin spawning stock biomass SSB0 than the one obtained in 2019, with an estimate of 55 740 tons (49 220 - 60 500 tons). The estimate of the current SSB status of the stock is 65,2% (61.4 % - 68.5 %) and the current catch limit satisfies the CCAMLR decision rules.

A research plan for Subarea 88.3 was commenced by Korea in 2016 under CM 24-01, New Zealand joined the proposal in 2017. The joint research proposal was for 3 years covering the period 2017/18, 2018/19 and 2019/20. It was designed to build on Korea’s previous research by continuing to focus on research blocks where tagged fish have previously been released on the slope, while also prospecting two of the northern seamount complexes and two boxes on the continental shelf, where no research has occurred. Following a proposal from the Ukraine to join the research plan in 2018, the Scientific Committee requested an integrated proposal be developed for all three Members (SC XXXVII paragraph 3.191). This proposal updates the final year (19/20 fishing season) of the research plan with the recommendation from the Scientific Committee to integrate the Ukrainian research with the Korean and New Zealand research plan. We propose to continue the research on Dissostichus spp. in Subarea 88.3 from 2021/22 to 2023/24. The main objective of the proposal is to determine the abundance and distribution of Antarctic toothfish in Subarea 88.3. Secondary objectives are to improve understanding of stock and population structures of toothfish in Statistical Area 88, to carry out calibration trials among the vessels, to collect data on the spatial and depth distributions of bycatch species, and to trial scientific electronic monitoring technologies.

A 2-year programme was implemented in the 2019/20 and 2020/21 fishing seasons in the Ross Sea region to tag and release skates for population size estimation and to validate the thorn ageing method for Antarctic starry skate (Amblyraja georgiana). The program has been largely successful to date, with a total of 8,506 skates tagged and released over the past two seasons in the Ross Sea region, with a further 484 individual skates voluntarily tagged in the Amundsen Sea Region. Recaptures from these releases will be used to monitor trends in population size through time with additional tagging occurring periodically. In addition, 2,117 skates that were tagged and released in the 2019/20 season on Australian, Spanish, New Zealand, and United Kingdom vessels were also injected with either oxytetracycline hydrochloride or strontium chloride in order to incorporate chemical marks into the caudal thorns for age validation. A total of 44 skates tagged during the program have been recaptured to date. Results from the age validation experiments, as well as those of biological and movement analysis, will be provided to future WG-FSA meetings. Tagged skates recaptured on all vessels in the Ross Sea region in future fishing seasons should be biologically sampled and a selection of caudal thorns removed and frozen, and returned to NIWA, Wellington, for subsequent laboratory analysis.

The paper provides a preliminary examination of catches and data holdings for bycatch species in the Ross Sea toothfish fishery. Bycatch species composition varied between management areas; however landings of most species groups were generally highest in SSRUs 88.1H and 88.1I in the S70 management area. As found in other areas of the Convention, macrourids were the most commonly observed bycatch group by both weight and numbers. Macrourids, skates, icefish, eel cods and morid cods comprised almost 99.5% of the total bycatch by weight. A large amount of length, weight, sex, and gonad stage data, as well as material for ageing (i.e., otoliths for finfish, whole individuals or caudal thorns for skates), has been collected for key species since 2014, in particular the starry skate Amblyraja georgiana, violet cod Antimora rostrata, the macrourids Macrourus caml and M. whitsoni, and for the most common eel cod species in the Ross Sea, Muraenolepis evseenkoi.

Based on the findings of this paper, we recommend:

1. A revised data collection plan be developed to direct continued collection and support analysis of data for bycatch species. Such a plan should balance data collection with other observer commitments, and be aligned with both a revised fishery-based research plan as well as the broader objectives of the Ross Sea region Marine Protected Area Research and Monitoring Plan.
2. Characterisations of inter alia recent trends in catch, effort, catch rates, and biological information should be compiled for commonly observed species so that quantitative risk assessments can be developed assess the risk of fishing on these species.
3. To assist with the cataloguing and analyses of biological data:

• i)     The "Revision of CCAMLR catch reporting forms" e-group explore ways to indicate whether non-otolith tissue samples where retained on the biological form; and
• ii)     The e-group also explore ways to indicate whether individuals recorded in the biological form were tagged.

4. That the Secretariat periodically summarise data available for bycatch species from toothfish fisheries in other areas and incorporate this information into fishery reports.

The Casal2 stock assessment software package is being developed to implement an updated generalised age-structured population model that allows more flexibility in specifying the population dynamics, parameter estimation, and model outputs than CASAL.

The package is approaching a development point where it can be considered for use by CCAMLR for tag-based toothfish assessments. Casal2 models can be built with user-defined categories such as area, sex, and maturity, and has been constructed to allow for alternative parameterisations of processes and observations to be compared between models. The structural elements of a model in Casal2 are generic and user defined. Casal2 models can be used for a single population with a single anthropogenic event (i.e., a single fishery), or for multiple species and populations, areas, and/or anthropogenic or exploitation methods.

Casal2 is open-source and is available at https://github.com/NIWAFisheriesModelling/Casal2. Comparisons of CASAL and Casal2 in a few stock assessments for New Zealand species show that they provide the same results for estimation of key productivity parameter, but more validation is required.

A Casal2 workshop will be held later in 2021 for scientists who wish to engage in the development and testing of Casal2, and we invite Members to participate in this workshop and in an e-group to develop test cases for presentation at WG-SAM in 2022.