MS Science (MSS) are responsible for the science aspects of the razor clam trial. Live samples, provided by trial participants, are being used to derive length/weight relationships and information on size at maturity and spawning period of razor clams in the different trial areas. Participants are also providing measurements of the razor clams landed from different trial areas.
In addition, MSS is analysing REM data to study where catches are taken and combining fishing effort and reported landings data (where date of reported landings is consistent with detected fishing activity) to monitor landings per unit effort. MSS have also been involved in the design and commissioning of surveys. Progress and findings over the reporting period are summarised below.
1) Biological data
Samples of live razor clams were routinley sent to MSS by trial participants on an approximately monthly basis since August 2018 (Table 1). Sample numbers were low in 2018 as vessels gradually entered the fishery and improved after engagement with the fishers to remind them of the importance and protocols for submitting live samples. The number of samples is highest from the West Coast SW trial zone area and this is reflective of the higher number of vessels fishing in that area. As of January 2021, 51 samples had been received and 48 fully processed with over 5,100 razor clams weighed, measured and dissected. Three samples were not processed because of the poor condition of the razors due to delays in the delivery chain. Samples have been obtained from all four zones and eight of the trial areas. The samples consisted of, almost exclusively, the pod razor Ensis siliqua. A few specimens of the bendie razor, E. magnus, (previously known as E. arcuatus) were identified in samples from sites in the West coast SW zone. These were excluded from analyses. Due to the Covid-19 pandemic and working from home regulations, there were no live razor samples received from April 2020 to January 2021.
|Firth of Forth||Outer Herbrides||West Coast NW||West Coast SW|
Length / weight relationships
Data has been analysed to derive length / weight relationships for the four trial zones (Figure 4). Length / weight relationships are required to convert survey estimates (sizes of razor clams and numeric density) into biomass density (weights) in the surveyed areas. They are also used in the process of checking the raised sampling length frequency distributions to landed weights.
Further analyses will include examining differences between trial zones, areas, months and years but these are more difficult to interpret because of the paucity of samples from some of the trial areas. Increased communication with stakeholders and the introduction of a monthly schedule has improved the coordination of samples but unfortunately sampling was stopped in April 2020 because of the Covid-19 pandemic.
Razor shells have been sent to the Scottish Association for Marine Science (SAMS) for age determination. This is being done with a view to deriving estimates of area specific growth parameters for use in length based stock assessments.
The biological analyses have been limited over the last year because sampling of live razors was stopped due to the Covid-19 pandemic. No further work was conducted on sexing of razors or the size of maturity but we provide below an update on histology samples analysed (up to February 2020) to investigate time of spawning.
Sections of razor tissue were examined microscopically to determine the stage of gonadal development and staged according to the six stages of the gametogenic cycle of E. siliqua published by Darriba et al., 2005. The observed seasonality of stages, for the sexes combined, is shown Figure 5. We were unable to identify all of the stages of the reproductive cycle and appear to have missed animals at the critical stage IIIA, the condition when the gonads are ripe. Samples collected in May 2019 were starting to spawn (IIIB) or in the stage of exhaustion (IV). In July, the majority of samples were exhausted (IV) or at rest (Stage 0). The early stage of gametogenesis, stages I and II, were evident in samples taken in August and October 2019 respectively, with more animals in stage II from October through to February 2020.
On the basis of the work done to date, it seems likely that spawning occurs in March and or April. However, further sampling is required to confirm this. This would also be the best time of year to obtain samples which include better representation of smaller size classes to derive estimates of size at maturity.
Razor clam landings length measurements
Trial participants and associated processors, have provided measurements of the lengths of razor clams (Figure 6) landed from each of the four trial zones and eight trial areas. As of the end January 2021, a total of 189 length samples were submitted to MSS (Table 2) from 23 different vessels. Collectively, a total of 43,011 razors have been measured by trial participants. Sample numbers were low in 2018 as vessels gradually entered the fishery, and improved in 2019 after engagement with the fishers to remind them of the requirements and protocols for submitting measurement samples of landings. No samples were received in the months of April, May and June in 2020 because of the Covid-19 pandemic. Sampling resumed in July 2020 and the importance of this sampling was reemphasised. As a result, 102 length samples (from 18 different vessels) were received from July 2020 to January 2021.
|Firth of Forth||Outer Herbrides||West Coast NW||West Coast SW|
Landings from the West coast NW and West coast SW trial zones have been sampled on a regular basis, i.e. in most months of the year for 2019 and 2020 (Table 2). The number of samples is highest from the West Coast SW trial zone area and this is reflective of the higher number of vessels fishing in that area. Coverage of other zones, where there are fewer vessels authorised to participate in the trial, was sporadic and less satisfactory but there have been improvements in the number of samples from the Firth of Forth since July 2020.
The length frequency distributions (the representation of different sizes classes) for the four trial zones are shown in Figure 7. In the short term, any marked changes in the length frequency distribution of razors landed from particular grounds could be indicative of changes in the health of the stocks. In the longer term, this type of data combined with information on quantities landed and growth parameters can be used in stock assessments to indicate whether fishing mortality F is above or below an estimate of FMSY.
The average length of razor clams in the landings sampled (all years and zones) was 176mm, above the minimum landing size of 100 mm, which is understood to reflect the size preference of the market. Initial examination of the length frequency data (Figure 7) suggests some variation in the sizes of razor clams landed from the different trial zones which may reflect differences in the size structure of stocks. Smaller razors were evident in 2019 for the West Coast NW but are not present in 2020 because one vessel fishing in a particular area is no longer targeting those grounds. That reduced the number of samples from that area and there is an observed change in the length frequency distribution from a normal curve to more bimodal shape. This is reflective of the different size categories landed by the different vessels. There are no obvious shifts for the other zones and the presence of very small razors in the West Coast SW was investigated and found to be a fisher who was measuring the catch and not landings. This was resolved and the protocol further clarified.
The length frequency data will be further analysed at the finer spatial resolution of trial area to assess for any obvious shifts to smaller size categories at a localised level. Analyses which involve raising measurements to landed weights, to determine the proportions of different size classes harvested are ongoing. A time series of four to five years of data is generally required for length-based stock assessments and work has started to progress assessments for trial areas where sufficient data are available.
2) Remote Electronic Monitoring (REM)
Summary of fishery
MSS is analysing the data from REM devices to build up a precise footprint of the fishery and to study fishing behaviour. Some vessels move between areas within their allocated zone and fish in a number of production areas (areas classified by FSS for shellfish harvesting) whereas others tend to fish more locally in only one or two production areas. Most fishing takes place at depths of less than 10 metres and over limited extents of the various production areas. The data are being analysed to map electro-fishing effort.
The number of vessels fishing throughout 2020 was fairly stable (Figure 8) excluding February where vessel numbers were lower, and April – June when the fishery was paused due to Covid-19. Fishing effort was highest during the summer months, July – September (Figure 8). It is possible that Covid-19 affected the temporal distribution of fishing effort throughout the year, although with the short time-series of the trial any consistent seasonal trend in effort is not yet discernible.
Fishermen record landings on a tow-by-tow basis (as opposed to a daily) to enable MSS to refine spatial mapping of landings per unit effort (LPUE) and monitor trends during the trial. MSS has developed data checking procedures which identify instances of apparent mismatches between reported landings and fishing activity identified from REM sensor data. These are not uncommon in REM trials: however, it can take time to put processes in place to minimise occurrences to an acceptable level. Given the inconsistency in the temporal resolution of reported landings (tow-by-tow or daily) and the challenges in matching effort and landings data, electro-fishing activity and Fish1 reported landings have been matched on a daily basis to allow investigations of LPUE (kg/hour).
The distribution of reported landings across trial areas indicates that Firth of Clyde and Firth of Forth are the most harvested trial areas and are harvested across all months along with Colonsay (Figure 9). The remaining trial areas were not fished in all months.
Evaluation of integrating REM derived effort and reported landings data
As part of the scientific reporting on the Scottish razor clam scientific trial, a data pipeline is in development as part of the Scottish Government Data Science Accelerator Programme. The project will develop a documented process to integrate fishers' weekly Fish1 landings declaration forms with fishing activity identified from remote electronic monitoring systems on board vessels as a proof of concept for future REM implementation. For the 2020 trial year, the fishery recorded 1435 days-at-sea, of these 19% were missing either reported landings or REM detected fishing activity. This means that scientific evidence and analysis of trends is based on approximately 81% of the likely fishing activity of the trial fleet.
For 8% of days-at-sea vessels reported landings but no electro-fishing activity was identified by their on-board REM systems (Appendix 1; Figure 1). This occurred for just over half of participating vessels (59%) but for most the number of days affected was <5%. The most common cause for lack of corresponding REM fishing activity was a non-operational sensor attached to the generator monitoring electrical activity (66% of cases) (Appendix 1; Figure 2). Days where there was no indication of any vessel use whatsoever from on-board REM systems, which could be indicative of landings being reported incorrectly to a date other than the date in which fishing activity actually occurred accounted for only 1% of days-at-sea. Following the evaluation of data integration, the vessels where reported landings were missing REM electro-fishing effort were flagged to MS Compliance who reported already having notified the affected vessels and attempted to remedy the issue with either the REM system or suspected non-compliant behaviour.
For 11% of days-at-sea, on-board REM systems detected electro-fishing activity but vessels had no corresponding reported landings. This does not necessarily infer non-compliance but may be indicative of the poor data quality of weekly submitted log books. Approximately 10% of these cases were instances where the detected electro-fishing activity was for less than 1 hour which could indicate the generator was used for a non-fishing activity or for some reason electro-fishing was halted for the day after only a very brief operation (Appendix 1; Figure 3).
Landings per unit effort (LPUE) trends for fleet
Landings per unit effort (LPUE) is typically variable, among vessels and over time, which makes it difficult to reliably discern trends which might be related to stock abundance. Although, there is some suggestion of declines in catch rates in some trial areas and production areas over the reporting period relatively high catch rates have been maintained in others. LPUE can also be influenced by individual vessel fishing patterns, vessels moving between grounds etc. MSS continues to monitor LPUE in trial and production areas and is considering whether LPUE data in combination with changes in the length distributions of landings could be used as indicators of changes in stocks.
The LPUE has been fairly stable for the length of the trial for the West coast SW and Outer Hebrides zones (Figure 10). For the West coast NW zone the LPUE has increased over the duration of the trial: however, this appears to be attributable to three production areas with markedly different LPUE values having been fished at different periods of the trial. For the Firth of Forth trial area there is an indication that the LPUE has reduced over the timespan of the trial. This trial area was further investigated with a camera survey completed in 2020 to better understand the razor stocks in this area (https://doi.org/10.7489/12381-1).