Assessment Model

Data Sources

Commercial catch and effort data used in this assessment are based on those entered into Tasmanian Octopus Fishery (TOF) Commercial Catch, Effort & Disposal Record logbook returns. Octopus fishing records comprise individual demersal longline lifts, with catches per line reported as weight, and effort per line reported as the number of pots (i.e., ‘pot-lifts’). These data are used to calculate catch per unit effort (CPUE). Additional octopus catch data are reported from the Rock Lobster and Scalefish fisheries (tonnage of by-catch) and can be accessed in assessment reports of those fisheries; these data are not included in CPUE calculations.

Since November 2004, a 50-pot sampling programme has been conducted, where fishers are required to collect all octopus caught in 50 randomly selected pots from a single line, representing 10% of a standard commercial line. From these 50-pot samples, the numbers of males and females of each species and the percentage of pots with eggs are recorded. The total and gutted weight of the catch was also recorded from 2004 to 2010. Fishers are required to sample at least 50 pots per line from at least one line per fishing day, and at least one line per distinct area fished in each day. Areas are distinct when lines are located entirely on different substrates or are separated by more than 10 nautical miles. For the purposes of the 50-pot sampling programme, the TOF area has been separated into five geographic regions (see Fishing Regions).

Weight-at-age is highly variable in octopus due to a high individual variability and a rapid response to environmental factors (Leporati et al., 2008; André et al., 2009). This introduces stochasticity in catch weight so that it becomes difficult to use when interpreting trends in population size. The 50-pot samples provide numbers of octopus, which is more representative of the state of the stock. This practice aims to enhance the understanding of the stock status, particularly at a finer spatial scale (i.e., block level). New logbook requirements recently implemented will lead to improved data collection for the 50-pot samples.

In the 2021/22 season, commercial data also exist for the developmental permits on the east, south, and west coasts of Tasmania. This fishing is outside the normal TOF operations and has been summarised separately in the assessment report.

Data Analysis

Catch, Effort and CPUE

A fishing year from 1st March to the last day of February has been adopted for annual reporting, which reflects the licensing year. Catches are analysed fishery-wide and by fishing blocks. For the purpose of assessment, catch, effort and CPUE analyses are restricted to commercial catches of Pale Octopus for the period March 2000 to February 2022.

Data on logbook returns include gutted and non-gutted (i.e., whole) weights. All gutted weights were converted to whole weight as follows:

Whole weight = 1.233472 * Gutted weight

where Whole weight and Gutted weight are in kilograms. This relationship between Whole and Gutted weight was estimated from 8,510 individuals recorded in the 50-pot sampling dataset between December 2004 and April 2010.

The number of pots pulled (pot-lifts) is used as a measure of effort in this assessment. Catch returns for which effort information is incomplete are flagged and excluded when calculating effort or catch rates. However, in recent years the amount of incomplete logbook entries has been negligible to nil. All records are included for reporting catches.

The impact of soak time (the time during which the fishing gear is actively in the water) was determined by analysing CPUE trends (in catch number per pot) through time for the 50-pot sampling data. Exploration of this influence was discussed in detail in the 2015/16 stock assessment (Emery & Hartmann, 2016), where no relationship between soak time and CPUE was apparent. Therefore, soak time is not considered in the resultant catch standardisation process below.

CPUE of Pale Octopus have been standardised using a generalised linear model (GLM) to reduce the impact of obscuring effects such as fishing year or season on the underlying trends (Kimura, 1981, 1988). However, while standardised catch rates are preferred over the simple geometric mean, other factors may remain unaccounted for that obscure the relationship between standardised catch rates and stock size, such as increasing fisher efficiency or spatial shifts in fishing effort from areas of low to higher catch rates.

There is currently only one operator representing the TOF, which uses two different vessels. The vessels cooperate, with the vessel pulling the gear not necessarily being the same vessel that set it. Consequently, vessel and skipper were not included in the GLM. The depth at which the gear is set is variable; however, the inclusion of the term Depth in the GLM did not significantly explain variation in CPUE (p > 0.05), and this term was excluded from the final model. Catch and effort data for many fishing blocks across multiple years were absent or insufficient for testing the interaction between Licensing Year and Fishing Block (i.e., for testing whether CPUE distribution among fishing blocks varied among licensing years), and this interaction term was excluded from the final model. Factors considered in the final GLM were year, month, and block. A lack of spatial block data for multiple trips early in the time series led to 229 t of catch data (12.5% of total catch over the time series) being omitted from the subsequent catch standardisation process.

There is currently only one family business representing the TOF, which operates from two different vessels. The vessels cooperate, with the vessel pulling the gear not necessarily being the same vessel that set it. Consequently, vessel and skipper were not included in the GLM. The depth at which the gear is set is variable and the inclusion of depth in the GLM was not useful for explaining the data. Factors considered in the GLM were year, month, and block. A lack of spatial block data for multiple trips early in the time series led to 228 t of catch data (13.3% of total catch over the time series) being omitted from the subsequent catch standardisation process. The GLM was applied to weight per pot for the whole commercial dataset and number per pot for the 50-pot sampling dataset. This process removes the effect of season and location so that trends in CPUE are more accurately reflective of change in octopus abundance.

Stock Status Definitions

To assess the status of Pale Octopus in the TOF in a manner consistent with the national approach (and other jurisdictions), we have adopted the national stock status categories used in the 2020 Status of Australian Fish Stock (SAFS) report (Piddocke et al., 2021). These categories define the assessed state of the stock in terms of recruitment overfishing, which is often treated as a limit reference point. If a stock falls below this limit reference point, it is deemed that recruitment is impaired and its productivity reduced. Fisheries are ideally also managed towards targets that maximise benefits from harvesting, such as economic yield or provision of food. However, the scheme used here does not attempt to assess the fishery against any target outcomes. Determination of stock status into the below categories is based on temporal and spatial trends in commercial catch, effort and standardised CPUE data from the TOF.

Reference Points

During the decade from 2004/05 to 2013/14 (the reference period for the TOF), fishing activity was concentrated in the eastern Bass Strait around Flinders Island and in fishing blocks close to Stanley. Total catch for the fishery fluctuated around approximately 75 – 100 t per year and effort fluctuated around approximately 250,000 to 350,000 pot-lifts per year. Stock assessment methods were developed based on the assumption of ‘negligible shifts in the distribution of fishing effort’.

In 2019/20, catch and effort data from TOF no longer met this assumption and were deemed inappropriate for assessing against fishery-wide performance metrics. This is because, in 2019/20, substantial shifts in fishing effort and catch occurred within the fishery, leading to the introduction of additional analyses and regional assessment approaches.

Following the 2019/20 assessment of the TOF Pale Octopus stock as Depleting, fishery stakeholders initiated a meeting among representatives of the Institute for Marine and Antarctic Studies (IMAS), the Department of Natural Resources and Environment Tasmania (NRE Tas), and the Tasmanian Seafood Industry Council (TSIC). During the meeting, IMAS presented stock assessment outcomes for the 2019/20 season for extended subsequent discussion. The meeting concluded with the aim to establish potential reference points for future management of effort in all different regions currently monitored in the context of the 50-pot sampling programme (see Fishing Regions). Reference points for sustainable effort, rather than catch, were considered desirable in terms of both practical implementation as well as fishery impact. A draft report was developed by IMAS researchers and forwarded to NRE Tas for feedback, with principle outcomes based on catch-only simulations similar to those presented in the 2020/21 Tasmanian Octopus Assessment (Fraser et al. 2022). Summary statistics of annual effort in the different fishery regions from the reference period (2004/05 to 2013/14) and over recent years (2018/19 to 2020/21) highlight the above-mentioned shift in the distribution of fishing activity (Table 1, Table 2).

Table 1 Summary statistics of annual effort (number of pot-lifts) by region for the reference period 2004/05 to 2013/14.

Region	Mean	Standard deviation	Minimum	Maximum	Median
King Island East	1,900	3,414	0	11,000	500
Stanley	176,100	86,323	51,500	304,500	171,250
Flinders Island West	68,855	62,480	2,000	215,000	44,500
Flinders Island East	34,900	52,537	0	174,500	20,500
All regions combined	281,755	113,675	93,550	447,500	281,250

Table 2 Summary statistics of annual effort (number of pot-lifts) by region for the period 2018/19 to 2020/21.

Region	Mean	Standard deviation	Minimum	Maximum	Median
King Island East	168,833	54,903	110,500	219,500	176,500
Stanley	152,333	68,776	97,500	229,500	130,000
Flinders Island West	5,667	2,754	2,500	7,500	7,000
Flinders Island East	833	1,443	0	2,500	0
All regions combined	327,667	17,961	311,500	347,000	324,500