Harbour Seal Decline – vital rates and drivers. Report to Scottish Government HSD2.

Numbers of harbour seals (Phoca vitulina) have dramatically declined in several regions of the north and east of Scotland, while numbers have remained stable or have increased in regions on the west coast. For any management and mitigation plans to address this situation, the relative contribution of various factors in the decline of harbour seals in Scotland needs to be identified, understood and assessed. Potential drivers of the decline include changes in prey quality and/or availability, increasing grey seal population size which may be influencing harbour seal populations through direct predation or competition for prey resources, and the occurrence and exposure of seals to toxins from harmful algae.
Previous work by Matthiopoulos et al. (2014) and Caillat and Smout (2015) developed and fitted an age-structured population model to data from the well-studied subpopulation of harbour seals in Loch Fleet (Moray Firth), to evaluate the contributions of different potential proximate causes to the observed decline. Work has continued to build on the original Moray Firth study, re-coding a simplified version of the population model in JAGS language. A decision support tool (DST) has also been developed to include a biologically realistic simulation model and a model-fitting step that attempts to recover the parameters used in the simulation. A simple population model was successfully fitted to historical data for Scapa Flow (Orkney), with the Markov chain Monte Carlo (MCMC) converging and estimating reasonable-seeming parameter values. The DST was used to explore fitting limited data sets. The simulation/fitting approach showed that the fitting software was able to estimate parameters from the data even when the data set was ‘thinned’ (data not available for every year) and when no pup count data were available.
Live capture-release studies were conducted in Orkney in April and May 2016 under the SMRU Animal (Scientific Procedures) Act, 1986, (Home Office Licence No. 192CBD9F). Adult and juvenile harbour seals were captured, individual covariate data were collected from each seal and telemetry tags (GSM/GPS and LO tags) were deployed on adult seals, primarily on females, to direct the photo-identification effort prior to and during the pupping season. Pregnancy status was determined from progesterone concentrations in the plasma and in blubber, and from 17 beta-oestradiol concentration in plasma. Results show the blubber concentrations of progesterone may be a much more reliable indicator of pregnancy than levels in plasma. The proportion of the live-captured adult females that were pregnant was 61.5% (95% CI 35% – 88%), which is lower than would have been expected. However, given the small sample size further investigations must be carried out before any conclusions can be drawn. Domoic acid concentrations in the urine and faecal samples collected from the live capture-release animals were determined. Two animals had levels below the limit of detection, but the majority (88%) were above this level, indicating some low level exposure. Additionally, a further six scats collected at the capture haulout sites during May and June were also analysed. Of these, three were positive for DA but the remainder were below the limit of detection or samples were too small for analysis. Two fishing trips to collect prey samples were undertaken in the waters off Scapa Flow on the west coast of Orkney mainland. A total of 85 fish guts were sampled: 35 cod samples, 12 haddock, 36 ling and two torsk. All fish viscera were analysed for domoic acid content, using the same method as for the seal samples. All samples were positive for domoic acid at or above the limit of detection, although, in general, concentrations in all fish sampled were at low levels.
Moult aerial helicopter surveys were conducted in August 2016 in Orkney as part of the annual surveys conducted by SMRU. Breeding aerial surveys were also conducted in 2016 in Scapa Flow (Orkney), Kintyre and Isle of Arran, and Loch Dunvegan, using a fix-wing aircraft and digital photography. The difficulty of locating seals at haulouts from the aircraft and the impossibility of identifying age classes in the digital photographs led to the decision of excluding such data from the population model.
A summary of all seal carcasses reported to SMASS within and nearby the study sites between June 2016 and March 2017 is provided, with details on species, age class and proximate cause of death when available.