SARF098:PAMP Refreshment Study – final report: A statistical analysis of sea - lice medicine use and benthic monitoring at Scottish marine salmon farms (2002 – 2014).

Executive summary
The louse Lepeophtheirus salmonis is a crustacean ectoparasite that infests salmon. The control of lice is the major issue currently facing the Scottish salmon farming industry. The infeed anti-louse chemicals emamectin benzoate (EMB) and teflubenzuron (TBZ) are consented for use in Scotland. Following ingestion EMB or TBZ enter the fishes’ tissue which becomes toxic to feeding lice. EMB and TBZ are both excreted by the fish over an extended period (~200 days) and enter the environment via faeces. EMB and TBZ are both particle affinitive, accumulate in the sediments around the fish-farm and may pose a hazard to non-target benthic crustacea.
SEPA are the consenting authority in relation to EMB/TBZ use in Scotland (Note: TBZ is no longer used in Scottish salmon farms). Under the farm-operating consent, SEPA requires farmers to report the results from grab-based sampling of sediment around the farm site (at cage edge (CE), intermediate (AZE) and distant Reference (Ref) stations) in relation to macrobenthos, particle size and organic content. SEPA also collates current flow-data from around farms, the monthly use of EMB/TBZ and the results from EMB/ TBZ seabed residue monitoring.
The objectives of the present work were to assess the relationship between EMB/TBZ use and benthic assemblages, particularly crustacea, around fish-farms based on SEPA-collated data. There were insufficient data to assess the relationship between sensitive species and TBZ. In relation to EMB, the analysis was split between ‘per-production-cycle’ (PPC, ~20 month salmon production cycle (PC)) and per site total (PST) analyses with consideration of changes occurring at the CE, AZE and, particularly, the Ref stations. Within the data extracted from the databases (N=1259 samples from 99 sites) there were maxima of 3 and 10 kg of EMB used PPC and PST respectively (the PST consisted of the cumulative EMB used across several PCs). Analyses were based on generalised linear mixed models, with Bayesian inference, to predict the independent mean EMB effect (with 95% credible intervals) based on an average site, under average conditions were the maximum EMB to be used (PPC or PST). At Reference stations, on a PPC basis, there was strong evidence of a substantial decline, in crustacean richness and abundance, of 40% (4 - 63%) and 66% (29 – 87 %) respectively associated with 3 kg of EMB use. At Reference stations, on a PST basis, the EMB –crustacean association was more significant with expected reductions in richness and abundance being 64% (21 - 82%) and 96% (74 - 100%) indicating wide-scale, cumulative impacts and incomplete recovery between successive EMB treatments. Whilst it is possible that the modelled reductions in crustacea were attributable to factor(s) that were associated with EMB use, rather than directly caused by EMB, this is unlikely given the range and nature of the co-variables included in the models.
The ecosystem consequences of the observed reduction in crustacea are not known but crustacea include important fishery species such as crabs and lobsters. Understanding the consequences of EMB/TBZ use on benthic systems is essential to ensure the ongoing sustainable development of the Scottish fish-farming sector. The evidence suggests that benthic crustacea may not be adequately protected by the current regulation of EMB use in Scottish salmon farms.