Deep-sea fish

Four offshore marine regions (OMR): Bailey, Faroe Shetland Channel (FSC), Hatton and Rockall are considered here (Figure 1). These OMRs occur in two distinct biogeographical areas separated by the Wyville Thompson Ridge (WTR), which lies at around 60 ˚N and acts as a barrier between two water masses. To the north of the WTR, the FSC is exposed to cold Arctic waters, while the remaining OMRs to the south experience warmer Atlantic waters. As a result, the composition of the fish assemblages in these two biogeographical regions are quite distinct.

Rockall Trough, lying to the west of Scotland (making up large parts of the Bailey and Rockall OMRs), is considered the ‘cradle of deep-sea biological oceanography’ due to the pioneering work undertaken here by Victorian scientists in the mid-19^th century (Gage, 2001). These early expeditions dispelled the previously held wisdom that these depths were devoid of life. It is now established that fish biodiversity increases between depths of 400 – 1000 m (Clarke et al., 2015).

There is a long history of exploitation of fish stocks in parts of these offshore areas. Rockall Bank, for example, has supported a fishery for more than 200 years (Newton et al., 2008). Deep-water trawl fisheries began operating in the area during the 1970’s, but the rapid increase in commercial exploitation did not begin until 1989 and remained unregulated till 2003 (Gordon, 2003; Clarke et al., 2015).

Examples of species that have been commercially exploited in the area include: monkfish (Lophius piscatorius), black scabbardfish (Aphanopus carbo), torsk (Brosme brosme), ling (Molva molva), argentines (Argentina spp.), blue ling (Molva dipterygia), roundnose grenadier (Coryphaenoides rupestris), orange roughy (Hoplostethus atlanticus) and some of the deep-water sharks (mainly leafscale gulper (Centrophorus squamosus) and Portugese dogfish (Centroscymnus coelolepis)) (Campbell et al., 2011, see Priede 2019 for a review of some important deep-water fish species in the area). Several decades of exploitation has resulted in significant declines in the abundance of some of these commercially important species, and even commercial extinction in the case of the orange roughy (Dransfeld et al., 2013).

Management efforts in the area were first introduced through a total-allowable-catch (TAC) management system in 2003 and over subsequent years through gear bans (e.g. gillnets), with the fishery for deep-water sharks phased out in 2010 (Neat et al., 2015). More recently, in an effort to protect vulnerable deep-sea marine ecosystems, the EU introduced a ban on all bottom trawling gears operating below 800 m within its waters (EC, 2016).

The EC’s Marine Strategy Framework Directive (EC, 2008), is the policy framework for the protection of the marine environment and the environmental pillar of the Integrated Maritime Policy. It aims to achieve Good Environmental Status of the EU’s maritime waters by 2020, and to protect the marine resource base upon which economic and social activities depend. It does this by setting out 11 high-level descriptors of environmental status, the first of which aims to: “maintain biodiversity in line with a natural state appropriate for a particular area”. Assessing diversity across time can give an indication of how a system is responding to the pressures placed on it. The diversity of organisms in living systems effects how they respond to changes in the environment, underpins ecosystem function, and provides ecosystem goods and services that benefit humanity, such as the provision of food (Hiddink et al., 2008; Campbell et al., 2011).

The measures of diversity used in this assessment are, briefly:

Species richness

The number of different species present, with no consideration of abundance.

Shannon’s diversity index

This index takes account of both richness and evenness (proportional abundance of each species) and represents the uncertainty surrounding an unknown individual drawn from an assemblage. If an assemblage has high evenness, i.e. where all species are represented by similar numbers of individuals, then identifying an unknown individual has a high associated uncertainty (it may equally be one of a number of species). If, however, the assemblage is uneven, with one or a few species dominating, then the unknown individual may be predicted with less uncertainty as it is more likely to come from one of the dominating species. It is calculated using: -∑Pi ln(Pi), where Pi is the proportion of individuals belonging to species i.

Simpson’s diversity index

This index takes account of both richness and evenness and represents the probability that two randomly chosen individuals belong to different species. It is calculated using: 1 - ∑Pi2, where Pi is the proportion of individuals belonging to species i.

See Magurran (2004) and Morris et al. (2014) for more information.