Razor clam surveys - Firth of Clyde and Solway: report

Results from the 2023 Firth of Clyde survey

Water column – Firth of Clyde 2023

The water column was generally well mixed although slight to moderate haloclines and/or thermoclines were present at around 4 m depth on some days (Appendix I: Figure 9). Water temperatures were between 12.0 and 14.2°C and salinities between 29.4 and 33.1.

Video tows – Firth of Clyde 2023

Summaries of the video tows are given in Appendix II: Table 1. A total of 98 tows were completed along the Ayrshire coast within the area where electrofishing is permitted. For comparative purposes a further 31 tows were completed outside the scientific trial areas.

The Ayrshire coastline consists of a relatively narrow strip of sand interspersed with rocky outcrops and areas of coarser ground. There are also numerous obstacles, such as pipelines and cables, so the local knowledge of the skipper and crew was essential for safe operation of the survey. The razor clam diving operations in the Clyde tend to be restricted to shallow depths, at most 20 m but typically less than 15 m. A limited number of tows were conducted in water deeper than 15 m with the aim of seeing whether there are repositories of Ensis below the normal diving depths. However, the sediment on these deeper tows was coarse silt and mud and very few razors were observed. The video cameras produced clear images, even on deeper tows and towards the end of the day, which allowed easy identification of organisms on the sediment (Appendix I: Figure 10).

The mean tow duration was 13 mins (± 2.7 std dev) with only a few tows terminated early due to changing sea conditions or encountering obstacles such as rocks (Appendix I: Figure 11a). Most tows were longer than 50 m (Appendix I: Figure 11b) which was the minimum distance recommended in Fox et al. (2019). Based on simulations the precisions of mean density estimates are likely to be improved by increasing the number of video tows within an area, rather than substantially increasing the lengths of individual tows. However, towing speed (Appendix I: Figure 11c) is also an important consideration as it influences the length of time an area of seabed is exposed to the electrical field. Mean towing speed was 4.0 ± 1.1 m min^-1 (mean ± std dev) so that most of the estimated exposure times were longer than 30 s (Appendix I: Figure 11d). The average swept area of the video tows was 75 ± 21 m² (mean ± std dev). From the video observations, the sediments were mainly fine rippled sand mixed with ground shell fragments with coarser sands in some locations.

Razor clam sizes – Firth of Clyde 2023

Overall, it was possible to estimate live sizes for 3,634 E. siliqua, with 71% of these measurements coming from tows conducted within the permitted electrofishing grounds. Comparing mean lengths for E. siliqua from tows within the presently fished and outside the trial area suggests that razor clams outside the fished area were slightly larger (Appendix II: Table 2). The largest mean size by site was recorded from tows conducted in Saddell Bay, which is outside of the trial area although the largest individual E. siliqua was recorded in Ayr Bay, a site within the permitted fishing area.

Many of the length datasets were skewed and multimodal meaning that simple comparisons of the average lengths may be mis-leading and comparing the whole length distributions is a better approach. Pooling the E. siliqua class 1 objects length data from all the tows suggests there were two or three size modes, one at around 200 – 210 mm; a second less obvious around 140 – 150 mm and a third around 100 – 110 mm (Appendix I: Figure 12a). There were also a small number of clams at less than 50 mm shell length, which were assumed to be E. siliqua because of the overall dominance of that species in the surveys. The size distribution of E. siliqua from tows conducted within the scientific trial area was similar to the overall pattern (Appendix I: Figure 12b). In comparison, the size distribution of E. siliqua from tows outside of the trial area had a slightly higher proportion of medium sized razor clams (Appendix I: Figure 12c). Compared with the other sites, Carradale and Saddell Bays, which are both in the unfished area, were dominated by large sized razors (Appendix I: Figure 13).

Ensis magnus are not normally collected by this fishery and the vast majority (86%) of the records of this species came from Machrie Bay (outside of the permitted trial area).

Razor clam densities – Firth of Clyde 2023

Summary statistics for E. siliqua densities comparing tows undertaken within and outside of the present Scottish Government scientific trial area in the Firth of Clyde are given in Appendix II: Table 3. The maximum density across all three size categories was 2.60 m^-2 and was found outside of the trial area, whilst within the fishery area the maximum density was 1.41 m^-2. Mean densities across all three size categories were 0.53 ± 0.03 (mean ± SE) for the whole survey, and 0.47 ± 0.03 (mean ± SE) and 0.72 ± 0.09 m^-2 (mean ± SE) for the in-trial and out-trial areas respectively. However, density distributions were often skewed so that the medians may be more representative, these being 0.43 m^-2 for the whole survey, and 0.37 and 0.61 m^-2 for the in-trial and outside-trial area tows respectively.

The density data can be further summarised by site (Appendix I: Figure 14 and Appendix II: Table 4). Median densities of large size (> 150 mm shell length) E. siliqua were between 0.20 and 0.24 m^-2 in Irvine, Ayr and Culzean Bays but higher in Turnberry Bay. Higher densities of large size E. siliqua were also found in Carradale and Saddell Bays which are outside of the trial fishery area. For medium sized (≥ 100 and < 150 mm shell length) E. siliqua, median densities were quite variable being less than 0.1 m^-2 in Irvine, Carradale and Saddell Bays but up to 0.22 m^-2 in Culzean Bay. Median densities of small sized (< 100 mm shell length) E. siliqua were lower only reaching marginally above 0.1 m^-2 in Turnberry and Carradale Bays (Appendix I: Figure 14d).

The spatial distributions of E. siliqua densities in the different size categories for sites within the Scottish Government scientific trial are shown in Appendix I: Figures 15 to 18. Data from sites outside the trial area are shown in Appendix I: Figures 21 and 22. Densities are shown as pink filled circles with the diameters scaled in proportion to the average density of razor clams recorded on that tow. For each size category, the scaling goes up to the maximum density recorded across all the tows in the Firth of Clyde survey i.e. it is possible to compare the charts between sites for each size category, but care should be taken comparing charts within a site for the different size categories.

Although large razors (≥ 150 mm shell length) were found throughout the area (Appendix I: Figures 15 and 21), the highest densities were in Turnberry Bay. The densities of medium sized razors (≥ 100 and < 150 mm shell length) seemed to increase moving south from Irvine Bay (Appendix I: Figures 17 and 18). Patches of small size razors (< 100 mm shell length) were found scattered across Ayr, Culzean and Turnberry Bays (Appendix I: Figures 18 and 20). In Irvine Bay, small size razors were largely confined to a patch off Ardrossan harbour while densities in Ayr, Culzean and Turnberry Bays were slightly higher.

Considering sites outside of the present trial, large razors were found at all three locations (Figure 21a) and at densities generally higher than within the fished area (Appendix I: Figure 14). In contrast, medium sized razors seemed mainly confined to Machrie Bay (Figure 21b) where maximum densities exceeded those within the fished area. Small size razors (< 100 mm shell length) were found distributed patchily in Machrie, Saddell and Carradale Bays (Appendix I: Figure 22).

In summary, within the trial fishing area, the highest densities of large size E. siliqua were observed in Turnberry Bay although even here the maximum density was less than 1 m^-2. Compared with Irvine, Ayr and Culzean Bays, densities of large razors in Machrie, Carradale and Saddell Bays, which are outside the permitted fishing area, tended to be higher. These differences likely reflect the recent fishing history as these latter sites have not been fished for 5 – 6 years. However, densities of medium and small sized razors were quite variable across all the sites and there was no obvious link with whether a site was within or outside of the permitted trial area. Small sized razors were present across all surveyed sites although higher densities were confined to smaller patches, especially in Irvine and Machrie Bays.

Razor clam depth distribution – Firth of Clyde 2023

There did not appear to be any obvious relationships between the densities of E. siliqua in each size category and the water depths at the time of sampling (Figure 23a – d).

Rates of partial emergence – Firth of Clyde 2023

The mean percentage of partially emerged E. siliqua was 19 ± 1 % (mean ± SE), although rates as high as 60 and 67% occurred on individual tows in Irvine Bay and Machrie Bay (Appendix I: Figure 24a). There was no obvious relationship between percentage partial emergence and mean exposure time (Appendix I: Figure 24b). However, for water depth there did seem to be a relationship with partial emergence which tended to be higher on deeper tows (Appendix I: Figure 24c). There did not appear to be any relationship with E. siliqua densities (Appendix I: Figure 24d).

In a previous survey of the Ayrshire coast, it was noted that partial emergence rates averaged around 20% but were higher at certain sites e.g. Culzean Bay being particularly high at 41% (Fox, 2018). This pattern was also seen in the present surveys where partial emergence rates averaged 20% for Irvine and Ayr Bays but reached 30% in Culzean Bay. In addition to a possible depth effect, razor clam emergence rates may be at least partially related to sediment type which will vary somewhat between sites.

Incidental observations – Firth of Clyde 2023

The total numbers of incidental organisms recorded on the videos are summarised in Appendix II: Table 5. Crabs (probably the common shore crab, Carcinus maenas) were the most common incidental organism overall. The next most common incidental organisms were sandeels (Ammodytidae), followed by hermit crabs and various starfish (mostly Asterias rubens but also occasional cushion stars). Small, stunned flatfish were also observed, although some were seen to recommence active swimming as the video cameras passed overhead. Two species of diving birds, the cormorant (Phalacrocorax carbo) and eider duck (Somateria mollissima) were observed adjacent to the survey vessel, and occasionally on the underwater cameras, although neither species was seen in large numbers.