Scottish scallop stocks: results of 2023 stock assessments
This report presents the results of Scottish regional king scallop stock assessments carried out by the Marine Directorate including data up to 2022. It also provides background information on Scottish fisheries for scallops, and the current management and regulatory framework.
3. Results and discussion by area
3.1. Regional and temporal trends
Since 2016, total Scottish scallop landings have declined substantially, from over 12 thousand tonnes in 2016 to just over six thousand tonnes in 2022. The majority of the landings are dredge caught, and it is the reduction in this fishery that is responsible for the overall decline in landings (Tables A.2 and A.3). Dive caught scallops typically making up less than 10% of the total landings.
Temporal trends in total landings (by all vessels into all countries) vary considerably between assessment areas and are shown in Figure 3.1 (and Table A.4). In some areas, particularly to the west of Scotland, there appear to have been substantial declines in the fisheries over the last 10 years whilst in other areas such as Orkney and Shetland the landings show a generally increasing trend over the time series of data.
The spatial distribution of UK landings in 2022 (split by landing country: Scotland and Other UK) is shown in Figure 3.2. ICES rectangle 49E8 (to the west of Shetland) has been the most important rectangle for Scottish landings in recent years, with almost 900 tonnes of scallops landed from this location in 2022. Over the last five years, the Shetland assessment area has accounted for around 20% of the total Scottish landings. The statistical rectangles off the west coast and northeast coast of Scotland are also important, resulting in the Northwest and Northeast assessment areas being the next most important after Shetland. There are also substantial scallop fisheries further south in the Irish Sea, off the northeast English coast and in the English Channel which are also prosecuted by Scottish vessels, although with the majority of the landings reported into non-Scottish UK ports.
The main dive fisheries are located in the coastal waters of the west of Scotland and also around Orkney (Figure 3.3) where diving accounted for over 50% of total landings in 2022.
3.2. Clyde
Description of the fishery
Landings from this area have fluctuated markedly over time, from over 400 tonnes per year in the late 1970s to under 20 tonnes in 1990. There has been a general decline in the fishery over the last 10 years, with landings of 133 tonnes in 2022. (Figure 3.1). A substantial proportion of landings in this area are taken by the dive fishery (~35% in 2022), with around five vessels regularly operating out of Tarbert. The dredge fishery is prosecuted by both local and visiting vessels, with the majority of these landings going into Campbeltown, Carradale and Tarbert. Note that the partitioning of landings from statistical rectangle 40E4 into east (Clyde) and west (West of Kintyre) components relies on an estimation procedure (See Section 3.2) and hence there is likely to be greater uncertainty associated with the total landings values for Clyde than for other areas. The estimated landings from the Clyde area were updated for 2002 onwards in the current assessment report.
Data
Raised catch-at-age data are available for the Clyde, although in some years these are based on quite low levels of sampling (Table A.5). In 2019, MD began a dredge survey in the Clyde. So far, the survey has been relatively exploratory in nature and a set of stations with appropriate coverage of the scallop grounds is in the process of being established. It is envisaged that as the survey coverage becomes more standardised from year to year and the time series increases in length, there should be sufficient data to provide a stock assessment for scallops in this area.
3.3. East Coast
Description of the fishery
The scallop fishery in the East Coast assessment area developed in the 1990s. There has been marked variability in the landings throughout the time period, from 313 t in 2001 to a high of almost 3,000 t landed in 2016 (Figure 3.1). Since then landings have declined quite substantially and were 463 t in 2022. The landings from this area are almost entirely from the dredge fleet. A large number of vessels operate in the area, some based locally and others which are more itinerant, fishing on different grounds around the coast of the UK. Over 80% of landings from this area go into ports in northeast Scotland (Fraserburgh and Peterhead).
Catch-at-age data
Sampling of the landings has been carried out since the beginning of the fishery (Table A.5). In recent years, sampling levels have been particularly poor and this is likely to be due to a lack of sampling opportunities given the recent lower level of landings. The Covid-19 pandemic is also likely to have resulted in reduced sampling opportunities.
Catch-at-age data for the East Coast are shown in Figure 3.4 and Table A.6 for 1991 onwards. No specific age classes consistently dominate the landings and there are no apparent trends in age composition. The high landings in 1994-1995 consist mainly of young (ages 4-6) individuals from the 1989-1991 year classes, which dominate the landings in 1999-2000 at older ages (8-10+). The catch-at-age data show consistently low numbers of individuals at young ages (three and four year olds) indicating only partial recruitment to the fishery up to age five.
Biological data
The mean weights-at-age are shown in Figure 3.5 and Table A.7. There are no apparent systematic temporal trends although interannual fluctuations in mean weight-at-age are similar across age classes.
Exploratory analysis
Mean standardised catch-at-age data by proportion are shown in Figure 3.6 with dark bubbles illustrating above average values. The data provide some indications of relative year class strength, with the 1989 and 1999 year classes (recruiting at age three in 1992 and 2002 respectively) appearing well above average and those of the mid 1990s being particularly low. These strong and weak year classes are well tracked at subsequent ages. In recent years, the data appear to be more noisy and it is difficult to identify clear year class signals.
Details of the surveys which have been carried out in the East Coast assessment area are given in Table A.8. A partial North Sea scallop survey was conducted in 1993, with full coverage of the East Coast assessment area beginning in 1994. However, the survey was not conducted consistently by the same vessel (RV Clupea) until 1997 onwards with a change to RV Alba na Mara in 2008. No comparative tows were conducted to compare catch rates between vessels. Previous scallop stock assessments have suggested that despite standardisation of catch rates (to account for differences in the number of dredges towed and dredge width), survey vessel may have a significant impact on catchability. Therefore, the survey data are treated as two separate time series. The Clupea dredge survey runs from 1998 to 2007 and the Alba survey from 2008 onwards, but with a missing year in 2020 due to disruption associated with the Covid-19 pandemic. (Table A.8 and A.9).
Mean standardised survey catch rates-at-age by cohort are shown in Figure 3.7 for the two surveys separately (Clupea and Alba). Following a number of weak year classes during the mid-1990s, the Clupea survey suggests good recruitment in years 2001 and 2002 (1998 and 1999 cohorts) and tracks these cohorts with above average catch rates across a range of age classes. Furthermore, this survey identifies a strong 2003 cohort. The early part of the Alba survey also identifies the 2003 cohort as well as 2004-2005 to be strong cohorts, seen as above average indices at age 3-5 in 2008 (Figure A.2). Although the data from the Alba survey appear to be more noisy with less internal consistency, there is an indication that recruitment has been relatively lower in recent years (many of the log mean standardised values are below zero).
The catch rates of scallops (age three and 4+ separately) at stations across the East Coast assessment area between 2017 and 2022 are shown in Figure 3.8. Despite the reduction in the number of survey stations, it is clear that there has been a reduction in survey catch rates in 2022 across the East Coast assessment area, particularly compared to 2019 and earlier. Catches of age three individuals (age at recruitment) are also less widespread in 2022 compared to other years.
Further plots exploring the quality and consistency of both the catch and survey data (catch curves and scatterplots) are included in the Appendix (Figures A.1-A.4) and these clearly show the low catchability of age three scallops in the survey.
Final assessment
The exploratory catch and survey data analysis indicates highly variable catch rates of age two individuals. In addition, the catch rates of the 10+ age group in the survey are very noisy. These data are, therefore, excluded from the final assessment.
A summary of the assessment input data is provided below:
| Data Type | Year Range | Age Range | Notes |
|---|---|---|---|
| Catch numbers-at-age | 1991 - 2022 | 3 - 10+ | - |
| Catch weights-at-age | 1991 - 2022 | 3 - 10+ | - |
| Clupea Survey numbers-at-age | 1998 - 2007 | 3 - 9 | - |
| Alba Survey numbers-at-age | 2008 - 2022 | 3 - 9 | Missing year in 2020 due to Covid-19 pandemic |
| Stock weights-at-age | 1991 - 2022 | 3 - 10+ | Assumed equal to catch weights-at-age |
| Maturity-at-age | 1991 - 2022 | 3+ | All ages assumed mature |
| Natural mortality-at-age | 1991 - 2022 | 3+ | 0.15 all ages (based on likely longevity) |
The approach to choosing a final model configuration was to begin with a largely default SAM configuration file and modify the model settings (i.e. increase complexity and number of parameters) with the aim of improving residual patterns and fit to the data (in terms of AIC). The main features of the final assessment model for East Coast scallops can be summarised as follows:
- Fishing mortality at ages 8 and above are assumed equal (coupled).
- Survey catchabilities are uncoupled across all ages for the Clupea survey and are coupled at ages 7 and above for the Alba survey.
- Catch observation variance parameters are coupled across ages 5 to 9 and uncoupled at other ages. To account for the variation in sampling levels over time, the number of catch samples per year is used to weight the data such that the catch data from those years with a greater number of samples is given a greater weight in the assessment.
- Survey observation error is coupled for ages 5 and above (both surveys).
- The Alba survey observations are modelled with AR(1) covariance structure (rather than independent). Correlation parameters for ages 3-5 were uncoupled from other ages.
- Recruitment at age 3 is modelled as a random walk.
- Fishing mortality-at-age is modelled with AR(1) and process variance parameters are coupled across all ages.
- Process variance in stock numbers-at-age were assumed coupled with the exception of age 3 (the age at recruitment).
The final SAM configuration file is given in Table A.10.
The stock summary from the SAM assessment is shown in Figure 3.9 and Table A.11. More detailed outputs, including estimates of model parameters, stock numbers-at-age and fishing mortality-at-age can be found in Tables A.12 – A.14.
The model predictions of catch and survey numbers-at-age track the observations relatively well across most ages (Figures A.5 – A.7). The exceptions to this are the younger ages in both catch and surveys, and the 10+ age group in the catch which are estimated to have greater observation uncertainty (logSdLogObs parameters 0, 1, 3, 4 and 7 in Table A.12).
There are no major patterns or trends in the one-step-ahead residuals or process residuals (Figures A.8 and A.9) and nearly all errors lie within ±3 standard deviations. (There is a single outlier in each case).
The full set of input data, model configuration and assessment results can be found on the stockassessment.org website (East Coast results).
Retrospective analysis
The retrospective plots shown in Figure 3.10 indicate that in general the assessment tends to overestimate the recruitment, and consequently the SSB, in the final year (i.e. that estimates are revised downwards with each additional year’s data). Mohn’s ρ (rho, average under-/overestimation) is often used as a measure of assessment performance. For SSB, this is calculated as 0.08 (averaged over the last five assessments) i.e. 8 % overestimation of SSB. With the exception of the peel ending in 2017, Fbar is relatively well estimated and Mohn’s ρ is calculated as 12 %. Typically assessments with Mohn’s ρ for SSB estimated within -15% and 20% are considered sufficiently robust for the provision of catch advice by ICES (ICES, 2020).
Stock summary
Following a number of very strong year classes between 2000 and 2015, recruitment is estimated to have been below average for the years since then. This has resulted in a decline in biomass to just below the long term average. Despite the decline in stock biomass, Fbar(4-8) has declined to one of the lowest values of the time series due to the substantial reduction in catch since 2017. There appears to be no clear relationship between stock size (SSB) and recruitment to the fishery at age three (Figure A.10).
Provisional reference points are defined below.
| Reference point | Value | Notes |
|---|---|---|
| F0.1 | 0.19 | Derived from 2023 assessment results. Used as a conservative proxy for FMSY (See Figure A.11). |
| Bpa | 6 359 tonnes | Bloss × exp(1.645 × σ); Bloss = 4 576 tonnes (lowest SSB (1991) from 2023 assessment), σ = 0.2 (CV on estimate of SSB 2022). |
The final estimates of current Fbar and SSB for 2022 are 0.065 and 10,095 t respectively. Comparing these values to the provisional reference points indicates that the stock is being fished below the FMSY proxy and SSB is well above the precautionary biomass reference point (Bpa).
Comparison with previous assessment
A comparison between the latest assessment and that given in the 2017 report is presented in Figure 3.11. Despite the differences in approach, the historical trends show good agreement. In the previous TSA assessment, Fbar was defined as average fishing mortality across ages four to six whereas in this assessment the age range is four to eight. Therefore it is unsurprising that there are differences between the estimates of Fbar in the two assessments. The latest assessment also estimates a slightly higher biomass in the later years of the assessment which may be associated with the retrospective pattern in the previous assessment (which showed a tendency to revise estimates of biomass upwards as additional years of data were included).
3.4. North East
Description of the fishery
The North East scallop fishery developed in the 1980s and landings peaked in the mid-1990s at over 3,000 t. Since then, there has been considerable fluctuation in the size of the fishery with landings increasing from 507 t in 2019 to 1664 t in 2021 and then subsequently declining by approximately 50% to 891 t in 2022 (Figure 3.1). The current fishery is principally a dredge fishery prosecuted by a large fleet of both local and visiting vessels. Almost 50% of landings go into Fraserburgh and a further 35% into Buckie and Peterhead.
Catch-at-age data
Sampling of the landings has been carried out since the beginning of the fishery (Table A.5). In recent years, sampling levels have been particularly poor and this is likely to be due to a lack of sampling opportunities given the recent lower level of landings. The Covid-19 pandemic is also likely to have resulted in reduced sampling opportunities.
Catch-at-age data for the North East are available from 1984 onwards. The data are shown in Figure 3.12 and Table A.15. In the early part of the time series, catches were more dominated by individuals in the 10+ age category, whereas more recently, the catches consist largely of age four to seven year olds (with the exception of 2019 where there are a high proportion of 8-10+ in the landings). The catch-at-age data show consistently lower numbers of individuals at younger ages likely indicating only partial recruitment to the fishery up to age five.
Biological data
The historical mean weights-at-age show variability, but no systematic trend until the mid-2000s when mean weights of older individuals increased until 2008 and then declined to more ‘normal’ values (Figure 3.13 and Table A.16). This coincides with the period when sampling levels became more variable and when ages nine and ten plus, in particular, were less apparent in the sampled landings. The increase in mean size at older ages in this period may therefore be an artefact associated with sampling variability. Similarly, in recent years sampling levels have reduced which may explain the large interannual variability in mean weights at older ages (particularly the change from 2019 to 2020).
Exploratory analyses
Mean standardised commercial catch-at-age data by proportion are shown in Figure 3.14 with dark bubbles illustrating above average values. Following a period of apparently poor recruitment in the late 1980s, the commercial catch-at-age data consistently suggest above average catches for the 1988 to 1991 cohorts across most age classes. Data from the more recent period also suggests some years with stronger recruitment, but these signals are less clear. In the 2018-2020 period, there is an indication of above average catch numbers at the older ages of cohorts estimated as below average at time of recruitment, suggesting a period of increased survival/lower fishing mortality or potentially a change in fishery selectivity.
Details of the surveys which have been carried out in the North East assessment area are given in Table A.8 and discussed in Section 3.3. Previous scallop stock assessments have suggested that despite standardisation of catch rates (to account for differences in the number of dredges worked and dredge width), survey vessel may have a significant impact on catchability. Therefore, in this assessment, the survey data are treated as two separate series. The Clupea dredge survey runs from 1997 to 2007 and the Alba survey from 2008 onwards, but with a missing year in 2020 due to disruption associated with the Covid-19 pandemic. (Tables A.8 and A.17).
Mean standardised survey catch rates-at-age by cohort are shown in Figure 3.15 for the two surveys separately (Clupea and Alba). Following a number of weak year classes during the mid-1990s (recruiting in the late 1990s at age 3), the Clupea survey suggests good recruitment in the years 2001 and 2002 and tracks these cohorts (1998 and 1999) with above average catch rates across a range of age classes. The early part of the Alba survey also identifies the 2003 and 2004 year classes (recruitment in 2006 and 2007) as strong across a range of older age classes. Thereafter there appear to be some periodic fluctuations in year class strength although the index at age 3 does not appear to be a good indicator of abundance at older ages.
The catch rates of scallops (age three and 4+ separately) at stations across the North East assessment area between 2017 and 2022 are shown in Figure 3.8. Coverage of the grounds (in terms of number of stations) was consistent between 2021 and 2022 but somewhat reduced in the central Moray Firth compared to earlier years. The majority of tows contain both recruits (age 3) and older individuals in 2022 suggesting relatively widespread recruitment in 2022. Following a reduction in catch rates off the northeast Aberdeenshire coast in 2019, these have increased again in 2021 and 2022.
Further plots exploring the quality and consistency of both the catch and survey data (catch curves and scatterplots) are included in the Appendix (Figures A.12 to A.15).
Final assessment
The exploratory catch and survey data analysis indicates highly variable catch rates of age two individuals. In addition, the catch rates of the 10+ age group in the survey are very noisy. These data are, therefore, excluded from the final assessment.
A summary of the assessment input data is provided below:
| Data Type | Year Range | Age Range | Notes |
|---|---|---|---|
| Catch numbers-at-age | 1984 - 2022 | 3 - 10+ | - |
| Catch weights-at-age | 1984 - 2022 | 3 - 10+ | - |
| Clupea Survey numbers-at-age | 1997 - 2007 | 3 - 9 | - |
| Alba Survey numbers-at-age | 2008 - 2022 | 3 - 9 | Missing year in 2020 due to Covid-19 pandemic |
| Stock weights-at-age | 1984 - 2022 | 3 - 10+ | Assumed equal to catch weights-at-age |
| Maturity-at-age | 1984 - 2022 | 3+ | All ages assumed mature |
| Natural mortality-at-age | 1984 - 2022 | 3+ | 0.15 all ages (based on likely longevity) |
The main features of the final model can be summarised as follows and are broadly similar to those for the East Coast assessment:
- Fishing mortality at ages 9 and above are assumed equal (coupled).
- Survey catchabilities are uncoupled across all ages for the Clupea survey and are coupled at ages 7 and above for the Alba survey.
- Catch observation variance parameters are coupled across ages 5 to 9 and uncoupled at other ages. To account for the variation in sampling levels over time, the number of catch samples per year is used to weight the data such that the catch data from those years with a greater number of samples is given a greater weight in the assessment.
- Survey observation error is coupled for ages 4 and above (both surveys).
- Recruitment at age 3 is modelled as a random walk
- Fishing mortality-at-age is modelled with AR(1) and process variance parameters are coupled across all ages.
- Process variance in stock numbers-at-age were assumed coupled with the exception of age 3 (the age at recruitment). Model convergence issues were encountered when estimating the process error for ages 4 and above, therefore this parameter was fixed at an arbitrary small value. (-5 on log scale and small in comparison to the estimated recruitment process error).
The final SAM configuration file is given in Table A.18.
The stock summary from the SAM assessment is shown in Figure 3.16 and Table A.19. More detailed outputs, including estimates of model parameters, stock numbers-at-age and fishing mortality-at-age can be found in Tables A.20-A.22.
The model predictions of catch and survey numbers-at-age track the observations relatively well across most ages (Figures A.16-A.18). The exceptions to this are the younger ages in both catch and surveys, and the 10+ age group in the catch which are estimated to have greater observation uncertainty (Table A.20).
There are no major patterns or trends in the one-step-ahead residuals or process residuals (Figures A.19 and A.20) and nearly all errors lie within ±3 standard deviations. There is some evidence in the catch-at-age residuals of increased uncertainty in recent years (one-step-ahead residuals since 2017) which potentially suggests that the use of the number of samples as an annual weighting factor is insufficient to account for the actual variability in the data (Figure A.19).
The full set of input data, model configuration and assessment results can be found on the stockassessment.org website (North East results).
Retrospective analysis
The retrospective plots shown in Figure 3.17 indicate that in general the assessment tends to overestimate the recruitment, and consequently the SSB, in the final year (i.e. that estimates are revised downwards with each additional year’s data). This is most apparent in the SSB peels ending in 2020 and 2021 which show substantial over-estimation while the earlier peels are more consistent with the final assessment. Mohn’s ρ (average under-/overestimation) for SSB is calculated as 14% and for Fbar is -15%. Typically assessments with Mohn’s ρ for SSB estimated within -15% and 20% are considered sufficiently robust for the provision of catch advice by ICES (ICES, 2020).
Stock summary
A number of very strong year classes in the early 1990s resulted in a rapid increase in stock size and catches in that period. After low recruitment in the years 2015-2016 followed by low SSB (2016-2019), both recruitment and SSB are currently around the long-term average. The fluctuations in catch (and fishing mortality) closely track those of recruitment (and SSB) suggesting that the fishery targets the strong year classes. Fishing mortality in 2022 is estimated to be just below the time series average. There appears to be no clear relationship between stock size (SSB) and recruitment to the fishery at age three (Figure A.21). The largest variability in recruitment is apparent at low SSB.
Provisional reference points have been estimated for this stock and are given below.
| Reference point | Value | Notes |
|---|---|---|
| F0.1 | 0.16 | Derived from 2023 assessment results. Used as a conservative proxy for FMSY. |
| Bpa | 9 706 tonnes | Bloss × exp(1.645 × σ); Bloss = 6 985 tonnes (lowest SSB (1988) from 2023 assessment), σ = 0.2 (max(0.2, CV on estimate of SSB 2022)) |
The final estimates of current Fbar and SSB for 2022 are 0.164 and 10,617 t respectively. Comparing these values to the provisional reference points indicates that the stock is being fished just above the FMSY proxy and the SSB is above the precautionary biomass reference point.
Comparison with previous assessment
The last Scottish scallop assessment report was published in 2017 (Dobby et al., 2017). A comparison between the latest assessment and that given in the 2017 report is presented in Figure 3.18. The two assessments are very consistent, with the main difference being in the estimates of fishing mortality. In the current assessment, Fbar was defined as the average fishing mortality over ages four to eight whereas in the previous TSA assessment the age ranged from four to six years, and therefore it is unsurprising that there are differences between the estimates.
3.5. North West
Description of the fishery
The North West assessment area covers much of the west coast of Scotland and the waters around the Hebrides. There is a long history of scallop fishing in this area (Figure 3.1). The main fishing grounds are to the north of Skye and further south around the Inner Hebrides. Landings have declined over the past six years and were 1 216 t in 2022, 50% of the 2016 landings (and approximately one quarter of peak landings in the early 2000s). There is also a small but significant dive fishery in this area which largely operates in the sheltered inshore waters around Kyle, Mallaig, Oban and Ullapool. In 2022, the dive fishery accounted for just over 10% of total scallop landings in the North West area.
Catch-at-age data
The North West area has generally been well sampled since the late 1980s. As in other assessment areas, sampling levels have declined in recent years (Table A.5), likely due to a combination of fewer sampling opportunities due to the Covid-19 pandemic and a smaller fishery.
Catch-at-age data for the North West are available from 1982 onwards. The data are shown in Figure 3.19 and Table A.23. In the early part of the time series, a substantial proportion of the catch was from the 10+ age category, whereas more recently, the catches consist largely of age four to seven year olds. The catch-at-age data show consistently lower numbers of individuals at younger ages likely indicating only partial recruitment to the fishery up to age five.
Biological data
The mean weights-at-age are shown in Figure 3.20 and Table A.24. The mean weights of individuals aged five to ten shows a gradual increase from the late 1980s to mid-1990s, and also in more recent years (2016 onwards) with similar interannual variations across age classes. Mean weights for those age categories which are less important in the catch show greater fluctuations. From 2007 to 2008, mean weights show a sudden increase across a number of ages coupled with an unusual age structure (low numbers at age 8+ in 2008). This could be due to either unrepresentative landings being sampled (low number of samples in this period, particularly 2007, Table A.5) or potentially to errors during the sampling process.
Exploratory analyses
Mean standardised catch-at-age data by proportion are shown in Figure 3.21 with dark bubbles illustrating above average values. There is some evidence that data track relative year class strength during the 1990s and early 2000s. However, the more recent data appear very noisy and indicate generally below average proportions of ages three and four followed by mostly above average proportions at older ages (with the exception of scallops aged 9 and older).
Details of the surveys which have been carried out in the North West assessment area are given in Table A.25. A number of partial dredge surveys of the west of Scotland were carried out in the late 1980s: however, the surveys were not conducted consistently by the same vessel. From 1993 the survey was conducted consistently by RV Aora I with a change to RV Aora II in 2003 and a further vessel change to RV Alba na Mara in 2008. No comparative tows were conducted to compare catch rates between vessels. Previous scallop stock assessments have suggested that despite standardisation of catch rates (to account for differences in the number of dredges towed and dredge width), survey vessel may have a significant impact on catchability. Therefore, the survey data are treated as separate time series for each vessel (Table A. 26). No survey was conducted in 2020 due to disruption associated with the Covid-19 pandemic. In 2021, there was further disruption to the survey resulting in only 28 stations being carried out across a small proportion of the assessment area (Figure 3.8). For that reason the survey data for 2021 are considered unlikely to be representative of the stock and are excluded from the assessment.
Log mean standardised survey catch rates-at-age by cohort are shown in Figure 3.22 for the three surveys separately (Aora I, Aora II and Alba). There is reasonable tracking of cohort strength in both Aora surveys. The Aora I survey identifies high recruitment in the early 1990s and then increasing recruitment in the late 1990s, which is estimated relatively consistently over a wide range of age classes also in Aora II. The Alba survey is much noisier, although there appears to be reasonable consistency in cohort trends for ages four to six. The missing years of survey data (2020 and 2021) make it difficult to draw conclusions about year class strength from these data for the more recent time period.
Further plots exploring the quality and consistency of both the catch and survey data (catch curves and scatterplots) are included in the Appendix (Figures A.23 to A.26).
Final assessment
The exploratory catch and survey data analysis indicates highly variable catch rates of age two individuals. In addition, the catch rates of the 10+ age group in the survey are very noisy. These data are, therefore, excluded from the final assessment.
A summary of the assessment input data is provided below:
| Data Type | Year Range | Age Range | Notes |
|---|---|---|---|
| Catch numbers-at-age | 1982 - 2022 | 3 - 10+ | - |
| Catch weights-at-age | 1982 - 2022 | 3 - 10+ | - |
| Aora I survey numbers-at-age | 1993 - 2002 | 3 - 9 | - |
| Aora II survey numbers-at-age | 2003 - 2007 | 3 - 9 | - |
| Alba survey numbers-at-age | 2008 - 2022 | 3 - 9 | No survey data for 2020 & 2021 due to disruption associated with Covid-19 |
| Stock weights-at-age | 1982 - 2022 | 3 - 10+ | Assumed equal to catch weights-at-age |
| Maturity-at-age | 1982 - 2022 | 3+ | All ages assumed mature |
| Natural mortality-at-age | 1982 - 2022 | 3+ | 0.15 all ages (based on likely longevity) |
The main features of the final model can be summarised as follows and are broadly similar to those for other assessment areas:
- Fishing mortality at ages 8 and above are assumed equal (coupled).
- Survey catchabilities are coupled at ages 8 and 9 for all surveys.
- Catch observation variance parameters are coupled across ages 5 to 10+ and uncoupled at other ages. To account for the variation in sampling levels over time, the number of catch samples per year is used to weight the data such that the catch data from those years with a greater number of samples is given a greater weight in the assessment.
- Survey observation error is coupled for ages 5 and above (all surveys). The Alba survey is allowed to have AR(1) covariance structure (rather than independent) with the age 3 to 4 correlation parameter uncoupled from other ages.
- Recruitment at age 3 is modelled as a random walk
- Fishing mortality-at-age is modelled with AR(1) and process variance parameters are coupled across all ages.
- Process variance in stock numbers-at-age were assumed coupled with the exception of age 3 (the age at recruitment). Model convergence issues were encountered when estimating the process error for ages 4 and above, therefore this parameter was fixed at an arbitrary small value. (-5 on log scale and small in comparison to the estimated recruitment process error).
The final SAM configuration file is given in Table A.27.
The stock summary from the SAM assessment is shown in Figure 3.23 and Table A.28. More detailed outputs, including estimates of model parameters, stock numbers-at-age and fishing mortality-at-age can be found in Tables A.29-A.31.
The model predictions of catch and survey numbers-at-age track the observations relatively well across most ages (Figures A.27-A.30). The exceptions to this are the younger ages in both catch and survey. In particular, the observation uncertainty estimated for age three in the catch is more than three times that at age four and more than four times that for ages five and above (Sd LogObs_0 compared to Sd_LogObs_1 in Table A.29).
There are no major patterns or trends in the one-step-ahead residuals or process residuals (Figures A.31 and A.32) and most errors lie within ±3 standard deviations. Two of the largest (negative) errors occur in the 2008 catch data at ages 9 and 10+. The rather strange age composition in this year was noted in the previous TSA assessment and was excluded from that assessment as it appeared to be highly influential in the fitting procedure (Dobby et al., 2017). In the SAM assessment presented in this report, those data appear to have relatively little influence, potentially due to being given lower weighting as a result of the comparatively lower sampling levels in that period.
The full set of input data, model configuration and assessment results can be found on the stockassessment.org website (North West results).
Retrospective analysis
The retrospective plots shown in Figure 3.24 indicate that in general the assessment tends to underestimate fishing mortality, and overestimate the recruitment and consequently the SSB, in the final year (i.e. that SSB estimates are revised downwards with each additional year’s data). Mohn’s ρ (average over-/underestimation) for SSB is calculated as 31% and for Fbar is -26%. In these circumstances (values outwith -15% to +20%), ICES would consider the assessment not to be sufficiently robust to be used for the provision of catch advice without additional precautionary considerations or rescaling of assessment results to account for the retrospective revisions.
Stock summary
A period of high recruitment from the early 1990s to 2000s resulted in an increase in stock size and catches in that period. Since then, recruitment and stock size have been at a relatively lower level and are estimated currently to be below the long-term average (although recently showing an increasing trend). Fishing mortality was high in the early 2000s and late 2010s but has since declined rapidly. There appears to be no clear relationship between stock size (SSB) and recruitment to the fishery at age three, with both low and high recruitment occurring across the range of SSB (Figure A.33).
Provisional reference points have been estimated for this stock and are defined below.
| Reference point | Value | Notes |
|---|---|---|
| F0.1 | 0.17 | Derived from 2023 assessment results. Used as a conservative proxy for FMSY |
| Bpa | 12 991 tonnes | Bloss × exp(1.645 × σ); Bloss = 8 768 tonnes (lowest SSB (2019) from 2023 assessment) σ = 0.239 (max(0.2, CV on estimate of SSB 2022)); |
The final estimates of current Fbar and SSB for 2022 are 0.155 and 11,958 t respectively. Comparing these values to the provisional reference points indicates that the stock is being fished below the FMSY proxy and the SSB is below the precautionary biomass reference point. However, given the closeness of estimated fishing mortality to the FMSY proxy reference point and that the retrospective analysis indicates that fishing mortality is likely to be underestimated, the conclusion on fishing mortality stock status should be treated with some caution. (The conclusions about SSB stock status are not affected by the retrospective bias due to the direction of the bias and that the stock is already considered to be below the reference point).
Comparison with previous assessment
The last Scottish scallop assessment report was published in 2017 (Dobby et al., 2017). There is reasonable consistency between the estimates of recruitment and SSB from the 2017 and most recent assessments, particularly over the early part of the time series (Figure 3.25). Since 2010, the estimates show a diverging pattern with the assessments from the latest assessment lower than those from the TSA. Some of the difference is likely to be due to the retrospective revisions in the assessment (downscaling of SSB with inclusion of additional data, Figure 3.24), but in addition there may be differences in the way the two methods are interpreting the data which may occur when data are particularly noisy. Estimates of Fbar are scaled up in the most recent assessment due to the use of an Fbar range of ages four to eight, compared to four to six in the previous assessment.
3.6. Orkney
Description of the fishery
The Orkney scallop fishery began in the 1970s but has remained relatively small in comparison to fisheries in other assessment areas. The majority of landings in this area are taken in the dive fishery. In recent years, this has consisted of 5 or 6 local vessels with the majority of landings going into Kirkwall. Most of the dredge landings are taken by visiting vessels that operate around the coast and are landed into ports on the mainland.
Data
Catch sampling levels for Orkney have remained at a reasonable level in recent years (Table A.5) and raised catch numbers-at-age data are available. There are no fishery independent survey data available for this area which has precluded the development of an analytical assessment.
3.7. Shetland
Description of the fishery
The Shetland scallop fishery developed in the late 1960s. Landings have shown an almost continuously increasing trend since the SSMO took on responsibility for management of the fishery in this area and were just under 1 500 t in 2022. The fishery is almost entirely a dredge fishery and is prosecuted by over 25 local vessels, most of which are less than 10 m in length. These vessels fish out of many different harbours around the coast of Shetland, although the majority of landings go into Collafirth, Lerwick, Sullom, and Whalsay. A very small number of larger vessels from other areas also fish in the waters around Shetland. Note that landings reported here for Shetland include all landings from the Shetland assessment area and therefore may differ to those reported to the SSMO (which is responsible for management out to six nautical miles only).
Catch-at-age data
The landings from the Shetland area have been consistently well sampled since the late 1980s. (Table A.5). Samples from the Shetland area are collected and provided by staff from University of Highland and Islands under a Memorandum of Understanding with the MD.
Catch-at-age data for the Shetland area are available from 1986 onwards (Table A.32 and Figure 3.26). The catch-at-age data show consistently lower numbers of individuals at younger ages likely indicating only partial recruitment to the fishery up to age five. In the early part of the time series, a substantial proportion of the catch was from the 10+ age category and additionally, during the periods 1998-2000, 2011-2016, the catch large consists of older age individuals (> 50% are age seven and above).
Biological data
The mean weights-at-age are shown in Figure 3.27 and Table A.33. The mean weights of individuals show periodic fluctuations with higher values across all ages between 2005 to 2010, and then an increasing trend from around 2015 onwards, particularly at older ages. There appears to be less interannual variability in mean weights-at-age at Shetland compared to other assessment areas, which is likely due to the higher sampling levels in this area.
Exploratory analyses
Mean standardised catch-at-age data by proportion are shown in Figure 3.28 with dark bubbles illustrating above average values. There is some evidence that data track relative year class strength during the 1990s and mid 2000s. However, the more recent data appear quite noisy with no obvious signals of year class strength.
Details of the surveys which have been carried out in the Shetland assessment area are given in Table A.34. The survey has been carried out in the first quarter of the year, with the exception of 2002 when the survey was carried out three months earlier than usual at the end of 2001. From 1998 to 2008, the survey was conducted by the RV Clupea and since then by the RV Alba na Mara. No comparative tows were conducted to compare catch rates between vessels and previous scallop stock assessments have suggested that despite standardisation of catch rates (to account for differences in the number of dredges worked and dredge width), survey vessel may have a significant impact on catchability. Therefore, in this assessment, the survey data are treated as two separate time series.
The number of valid survey stations varies considerably, with bad weather often disrupting the survey. Typically, the stations which are missed due to bad weather are those to the southwest of Shetland and in other exposed locations. It is not possible to determine whether lack of data from these areas has significantly biased the survey catch-at-age indices in these years. Note that the exceptionally poor weather in 2014 and 2015 meant that not even a partial survey of the area could be conducted in either of those years.
Log mean standardised survey catch rates-at-age by cohort are shown in Figure 3.29 for the two surveys separately (Clupea and Alba). The Clupea survey data show similar general trends across ages, indicating weaker cohorts in the mid-1990s (recruitment in the late 1990s) followed by a period of stronger cohorts. The Alba survey is much noisier, potentially due to the generally poorer survey coverage over the last 10 years (missing surveys and lower number of stations).
On the whole, total catch rates appear to have reduced slightly in 2021 and 2022 compared to 2020 (Figure 3.31). This is most obviously apparent in the stations in Yell Sound (north facing bay) and in the stations between Bressay and Whalsay (central east coast). Catches of age three individuals (recruitment) remained widespread in 2022 and were present at all stations.
Further plots exploring the quality and consistency of both the catch and survey data (catch curves and scatterplots) are included in the Appendix (Figures A.35 to A.38).
Final assessment
A summary of the assessment input data is provided below:
| Data Type | Year Range | Age Range | Notes |
|---|---|---|---|
| Catch numbers-at-age | 1986 - 2022 | 3 - 10+ | - |
| Catch weights-at-age | 1986 - 2022 | 3 - 10+ | - |
| Clupea survey numbers-at-age | 1998 - 2008 | 3 - 9 | - |
| Alba survey numbers-at-age | 2009 - 2022 | 3 - 9 | No survey data for 2014 & 20215 due to poor weather |
| Stock weights-at-age | 1982 - 2022 | 3 - 10+ | Assumed equal to catch weights-at-age |
| Maturity-at-age | 1982 - 2022 | 3+ | All ages assumed mature |
| Natural mortality-at-age | 1982 - 2022 | 3+ | 0.15 all ages (based on likely longevity) |
The main features of the final model can be summarised as follows and are broadly similar to those for other assessment areas:
- Fishing mortality at ages 9 and above are assumed equal (coupled).
- Survey catchabilities are coupled at ages 8 and above (Clupea survey), and ages 7 and above (Alba survey).
- Catch observation variance parameters are coupled across ages 5 to 10+ and uncoupled at other ages. To account for the variation in sampling levels over time, the number of catch samples per year is used to weight the data such that the catch data from those years with a greater number of samples is given a greater weight in the assessment.
- Survey observation error is coupled for ages 5 and above for the Clupea survey, and coupled both for younger (ages 3-6) and older ages (ages 7-9) on the Alba survey. The Alba survey is allowed to have AR(1) covariance structure (rather than independent) with the age 3 to 4 correlation parameter uncoupled from other ages.
- Recruitment at age 3 is modelled as a random walk.
- Fishing mortality-at-age is modelled with AR(1) and process variance parameters are coupled across all ages, except age 3 (recruitment).
- Process variance in stock numbers-at-age were assumed coupled with the exception of age 3 (the age at recruitment). Model convergence issues were encountered when estimating the process error for ages 4 and above, therefore this parameter was fixed at an arbitrary small value. (-5 on log scale and small in comparison to the estimated recruitment process error).
The final SAM configuration file is given in Table A.36.
The stock summary from the SAM assessment is shown in Figure 3.31 and Table A.37 and indicates substantial uncertainty in estimates of the latest stock size and fishing mortality. More detailed outputs, including estimates of model parameters, stock numbers-at-age and fishing mortality-at-age can be found in Tables A.38-A.40.
The model predictions of catch numbers-at-age track the observations relatively well across most ages, although the model estimates substantially greater uncertainty at ages 3 and 4 (Figures A.39 and Table A.38). The model estimates appear to follow the survey numbers-at-age reasonably well at ages five and above in the Clupea survey (Figure A.39). The fit to the Alba survey data appears much poorer across all ages and while the general increasing trend towards the end of the time series is picked up, the model fails to replicate much of the interannual variability and the lack of fit is attributed to noise in the data (Figure A.41). The estimated observation error associated with the Alba survey is twice that estimated for most ages (age 5 and above) in the Clupea survey (logSdLogObs_6 and logSdLogObs_7 compared to logSdLogObs_5 in Table A.38). There are no major patterns or trends in the one-step-ahead residuals or process residuals (Figures A.42 and A.43) and most errors lie within ±3 standard deviations.
The full set of input data, model configuration and assessment results can be found on the stockassessment.org website (Shetland results).
Retrospective analysis
The retrospective plots shown in Figure 3.32 indicate that the assessment underestimates fishing mortality, overestimates the recruitment, and consequently the SSB, in the final year (i.e. that estimates are revised downwards with each additional year’s data). Mohn’s ρ (average under-/overestimation) for SSB is calculated as 30% and for Fbar is -19%. In these circumstances (values outwith -15% to +20%), ICES would not consider the assessment to be sufficiently robust to be used for the provision of catch advice without additional precautionary considerations or rescaling of assessment results to account for the retrospective revisions.
Stock summary
A period of high recruitment from the early 1990s and late 2000s resulted in an increase in stock size and catches. Since 2017, the assessment estimates a sharply increasing trend in recruitment and subsequent stock size, but with very wide confidence intervals and a substantial retrospective pattern. Current recruitment and SSB are estimated to be the highest in the timeseries. While catches show a generally increasing trend, fishing mortality is estimated to have declined in recent years due to the rapid increase in stock size. There appears to be no clear relationship between stock size (SSB) and recruitment to the fishery at age three although the recent very high recruitments have all occurred at high SSB (Figure A.43).
Provisional reference points have been estimated for this stock and are defined below.
| Reference point | Value | Notes |
|---|---|---|
| F0.1 | 0.13 | Derived from 2023 assessment results (Figure A.45). Used as a conservative proxy for FMSY |
| Bpa | 5 197 tonnes | Bloss × exp(1.645 × σ); Bloss = 3 116 tonnes (lowest SSB (1990) from 2023 assessment), σ = 0.311 (max(0.2, CV on estimate of SSB 2022)). |
The final estimates of current Fbar and SSB for 2022 are 0.136 and 19,818 t respectively although the retrospective patterns suggest that the fishing mortality is likely to be underestimated and the stock size overestimated. Comparing these values to the provisional reference points indicates that the stock is being fished just above the FMSY proxy and the SSB is above the precautionary biomass reference point. Given the current estimated magnitude of the SSB (very far above reference point), the retrospective bias in this assessment is unlikely to affect the conclusions on stock status.
Comparison with previous assessments
A comparison between the latest assessment and that presented in the 2017 report (Dobby et al., 2017) is shown in Figure 3.33. The two assessments show good consistency in estimates of both recruitment and SSB, and the trend in fishing mortality is also similar. In the current assessment, Fbar is defined as the average fishing mortality over ages four to eight whereas in the previous TSA assessment the age range was four to six which accounts for differences between the estimates.
UHI-Shetland also conduct stock assessments of the scallops around Shetland and provide advice to the SSMO. Their advice on stock status is based on a landings per unit effort (LPUE) series which shows a slight increasing trend in recent years (MSC, 2016).
3.8. West of Kintyre
Description of the fishery
The West of Kintyre assessment area has a long history of exploitation with periods of high and low landings (Figure 3.1). A substantial reduction in landings has occurred over the last 10 years with total landings in 2022 of 755 t (down from over 3,000 t in 2012). Around 5% of total landings come from the dive fishery. The majority of landings are taken in the summer months by a fleet of over 30 vessels, some of which are local and some from further afield, including the Isle of Man and Northern Ireland. The main fishing grounds are around the islands of Islay and Jura and the southern end of the Kintyre peninsula and the majority of landings (> 80%) go into Campbeltown, Oban, Port Ellen, Stranraer, Tayinloan and West Loch Tarbert.
Note that the partitioning of landings from statistical rectangle 40E4 into east (Clyde) and west (West of Kintyre) components relies on an estimation procedure (See Section 3.2). Therefore, uncertainty associated with the total landings (and subsequently the catch-at-age data) is likely to be greater for the West of Kintyre than for the other main scallop areas. The estimated landings from the West of Kintyre area were updated for 2002 onwards in the current assessment report based on the estimation process that has been adopted this year.
Catch-at-age data
The West of Kintyre area has generally been well sampled since the mid-1980s (Table A.5) and raised catch-at-age data are available from 1982 onwards (Figure 3.34 and Table A.41). In the early part of the time series, scallops of age eight years and older, and particularly 10+, were well represented in the catches, but have been less evident since the 1990s. In contrast, the number of scallops at ages four to six in the catch has increased considerably since this time. There also appears to be a decline in the contribution of age three individuals to the total catch, particularly since around 2010, which could be indicative of weaker recruitment.
Biological data
The historical mean weights-at-age show variability, but no major trends until the mid-2000s when mean weights declined and subsequently gradually increased to more above average values across all ages at the end of the time series (Figure 3.35 and Table A.42).
Exploratory analyses
Mean standardised catch-at-age data by proportion are shown in Figure 3.36 with dark bubbles illustrating above average values. Despite the good sampling levels, the commercial catch-at-age data appear very noisy for the West of Kintyre. The data identifies a period of weak year classes during the late 1980s which are picked up in the data as below average catches at older ages in the early 1990s. Following that there is some evidence of stronger recruitment in the early 1990s and again in the late 2000s. However, relative year class strength is generally not consistently tracked through cohorts in these data.
Details of west coast scallop surveys which cover the West of Kintyre assessment area are given in Table A.25 and discussed in Section 3.5. No comparative tows have been conducted to compare catch rates between vessels and previous scallop stock assessments have suggested that despite standardisation of catch rates (to account for differences in the number of dredges worked by each vessel), survey vessel may have a significant impact on catchability. Therefore, in this assessment, the survey data are treated as three separate indices. The Aora I dredge survey ran from 1993 to 2002, the Aora II from 2003 to 2007 and the Alba from 2008 onwards, but with both 2020 and 2021 missing due to disruption associated with the Covid-19 pandemic (Tables A.25 and A.43). Within each of the three survey indices the seasonal timing of the survey has been relatively consistent over time.
Log mean standardised survey catch rates-at-age by cohort are shown in Figure 3.37 for the three surveys separately (Aora I, Aora II and Alba). On the whole, the survey data appear quite noisy. The Aora I survey identifies several year classes recruiting in the mid-1990s as above average, with the pre-1990 cohorts mostly below average. The weak 1998 year class (recruitment at age three in 2001) is consistently estimated below average across the Aora and Aora II survey series up to age seven. The Aora I and II surveys suggest good recruitment in 2002 and tracks this 1999 cohort with above average catch rates across a range of age classes.
The Alba survey data provide some evidence of strong recruitment in the late 2000s, but relative year class strength is not estimated consistently at older ages. Thereafter there appear to be some periodic fluctuations in year class strength although the index at age 3 does not appear to be a good indicator of abundance at older ages. Overall, correlations between ages appear to be low (Figure A.49).
Further plots exploring the quality and consistency of both the catch and survey data (catch curves and scatterplots) are included in the Appendix (Figures A.46 to A.49).
Final assessment
A summary of the assessment input data is provided below:
| Data Type | Year Range | Age Range | Notes |
|---|---|---|---|
| Catch numbers-at-age | 1982 - 2022 | 3 – 10+ | - |
| Catch weights-at-age | 1982 - 2022 | 3 - 10+ | - |
| Aora I survey numbers-at-age | 1993 - 2002 | 3– 9 | - |
| Aora II survey numbers-at-age | 2003 - 2007 | 3 - 9 | - |
| Alba survey numbers-at-age | 2008 - 2022 | 3 - 9 | No survey data for 2020 & 2021 due to disruption associated with Covid-19 |
| Stock weights-at-age | 1982 - 2022 | 3 – 10+ | Assumed equal to catch weights-at-age |
| Maturity-at-age | 1982 - 2022 | 3+ | All ages assumed mature |
| Natural mortality-at-age | 1982 - 2022 | 3+ | 0.15 all ages (based on likely longevity) |
The main features of the final model can be summarised as follows and are broadly similar to those for other assessment areas:
- Fishing mortality at ages 9 and above are assumed equal (coupled).
- Survey catchabilities are coupled at ages 8 and 9 for all surveys.
- Catch observation variance parameters are coupled across ages 4 to 9 and uncoupled for other ages. To account for the variation in sampling levels over time, the number of catch samples per year is used to weight the data such that the catch data from those years with a greater number of samples is given a greater weight in the assessment.
- Survey observation error is coupled for ages 4 to 8 and uncoupled at other ages (all surveys).
- Recruitment at age 3 is modelled as a random walk
- Fishing mortality-at-age is modelled with AR(1) and process variance parameters are coupled across all ages.
- Process variance in stock numbers-at-age were assumed coupled with the exception of age 3 (the age at recruitment). Model convergence issues were encountered when estimating the process error for ages 4 and above, therefore this parameter was fixed at an arbitrary small value. (-5 on log scale and small in comparison to the estimated recruitment process error).
The final SAM configuration file is given in Table A.44.
The stock summary from the SAM assessment is shown in Figure 3.38 and Table A.45. More detailed outputs, including estimates of model parameters, stock numbers-at-age and fishing mortality-at-age can be found in Tables A.46-A.48.
The model predictions of catch numbers-at-age track the observations relatively well with the exception of ages three and 10+ where the model estimates greater uncertainty in the data (Figures A.50 and Table A.46). The picture is similar for the fit to the survey data with the age three indices typically most uncertain. Although most one-step ahead and process errors lie within ±3 standard deviations, there appears to be a larger proportion of relatively large negative residuals in the last few years of data (Figures A.54 and A.55). This could be due to increased uncertainty in the data which is not accounted for by the use of the number of samples as a weighting factor.
The full set of input data, model configuration and assessment results can be found on the stockassessment.org website (West of Kintyre results).
Retrospective analysis
The retrospective plots are shown in Figure 3.39. There is some tendency to overestimate SSB (and slightly underestimate F) in the final year, as estimates are revised downwards with each additional year’s data. Mohn’s ρ (average under-/over-estimation) for SSB is calculated as 12 % and -12 % for Fbar. All peels fall within the confidence intervals of the full time-series assessment. Typically, assessments with Mohn’s ρ for SSB estimated within -15% and 20% are considered sufficiently robust for the provision of catch advice by ICES (ICES, 2020).
Stock summary
A period of increasing recruitment during the 1990s and 2000s resulted in an increase in stock size and catches. Recruitment has declined substantially since 2009 and is around the long term average in 2022. Catches have also declined rapidly in recent years and are currently around the lowest of the time series. As a result, despite the reduction in recruitment, fishing mortality has fallen in the 2020s to the lowest level observed in the time series and stock biomass remains at a high level (although estimated with substantial uncertainty). There appears to be no clear relationship between stock size (SSB) and recruitment although on average, recruitment appears to be higher at high SSB (Figure A.56).
Provisional reference points have been estimated for this stock and are defined below.
| Reference point | Value | Notes |
|---|---|---|
| F0.1 | 0.18 | Derived from 2023 assessment results (Figure A.57). Used as a conservative proxy for FMSY |
| Bpa | 6,559 tonnes | Bloss × exp(1.645 × σ); Bloss = 4,720 tonnes (lowest SSB (1988) from 2023 assessment), σ = 0.2 (max(0.2, CV on estimate of SSB 2022)). |
The final estimates of current Fbar and SSB for 2022 are 0.084 and 13 979 t. Comparing these values to the provisional reference points indicates that the stock is being fished below the FMSY proxy and the SSB is well above the precautionary biomass reference point.
Comparison with previous assessments
A comparison between the latest assessment and that given in the 2017 report (Dobby et al., 2017) is presented in Figure 3.40. The estimates are generally similar, but show some divergence, with higher estimates of recruitment and SSB in the SAM assessment since around 2000. The newly implemented estimation process for distributing landings between the Clyde and West of Kintyre components of statistical rectangle 40E4 (Section 3.2) has resulted in an upwards revision in the total catches from the West of Kintyre (compared to the 2017 assessment report). This results in the most recent assessment estimating higher recruitment and higher SSB compared to that presented in 2017. In the current assessment, Fbar is defined as the average fishing mortality over ages four to eight whereas in the previous TSA assessment the age range was four to six which accounts for some of the differences between the F estimates.
