Patterns of habitat utilization by deep-sea fish in the Bay of Biscay, NE Atlantic

Índice do artigo

A análise de vídeos obtidos através do submersível “Nautile”, na Baía de Biscaia, Oceano Atlântico NE, revelou que as espécies de peixes demersais estão associadas a tipos diferentes de habitats marinhos. Analisaram-se quatro transetos de mergulho com o objetivo de identificar as espécies de peixes, invertebrados bentónicos e as características ambientais a estas associadas. Analisaram-se 480 peixes, pertencentes a 19 taxa, através de métodos de análise estatística canónica. A epifauna marinha bentónica foi dominada por uma diversidade elevada de organismos, indicando gradientes diferentes de corrente vertical e horizontal. Ambos os fatores bióticos, abióticos e utilização de habitat influenciaram a seleção de habitat dos peixes demersais. A espécie mais abundante de peixe, olho de vidro laranja (Hoplostethus atlanticus), apresentou uma associação evidente a habitats complexos, incluindo colónias de corais. A espécie de peixe Neocyttus helgae, ocorreu em habitats associados a correntes fortes, inclinação elevada e presença de corais. Outras espécies de peixes, como por exemplo, o lagartixa-da-rocha (Coryphaenoides rupestres) e o moreão-de-natura (Synaphobranchus kaupii), ocorreram numa diversidade elevada de habitats e mostraram capacidade de adaptação em relação à variação dos diferentes fatores ambientais analisados. Os métodos destrutivos de pesca, como o arrasto, devem ser banidos nas zonas de corais de água fria e estabelecida uma rede de zonas protegidas como método de precaução de gestão das pescas.



Mendes Alves, Dário1 2 *,Uiblein, Franz3, Santos, Jorge1, Lorance, Pascal4

1Norwegian College of Fishery Science, University of Tromsø, 9037 Tromsø, Norway
2Department of Zoology, Institute of Zoology, University of Salzburg, Hellbrunnerstr. 34, 5020 Salzburg, Austria
3Institute of Marine Research, P.O. Box 1870, N-5817 Bergen, Norway
4Ifremer, Institut français de recherche pour l’exploitation de la mer, Centre Atlantique - Rue de l'Ile d'Yeu - BP 21105 - 44311 Nantes Cedex 03, France

*Corresponding author: Este endereço de email está protegido contra piratas. Necessita ativar o JavaScript para o visualizar.

Short title – Patterns of habitat utilization by deep-sea fish


Analysis of video recordings made during dives of the submersible “Nautile” in the Bay of Biscay, NE Atlantic Ocean, indicates that demersal fish species are associated with different marine habitat types. Four different dive transects were studied with respect to fishes, benthic invertebrates and associated environmental characteristics. A total of 480 fishes belonging to 19 taxa were ordinated by means of canonical correspondence analysis. Deep-sea epifauna was dominated by a diverse assemblage of suspension feeders indicating different gradients of vertical and horizontal current flow. Both biotic and abiotic factors influence fish habitat distribution. The most abundant fish the orange roughy (Hoplostethus atlanticus), showed a clear association with complex bottoms, including coral colonies. The false boarfish (Neocyttus helgae), occurred in habitats associated with strong currents, high slope inclination, and corals. Others such as roundnose grenadier (Coryphaenoides rupestris) and Kaup's arrowtooth eel (Synaphobranchus kaupii) occurred in a wider range of habitat types. Deep-sea destructive fishing methods, like trawling, should be banned in the areas of cold water corals and a network of no take zones should be introduced as a precautionary method of fisheries management.


Key words

Deep-sea conservation, Deep-sea ecology, Demersal fishes, Deep sea epifauna, Deep-sea image analysis, Submersibles

 1. Introduction

Little is known about the behaviour, spatial distribution and microhabitat utilization of deep sea fishes. The increasing coverage by photographic surveys and video recordings taken from submersibles and remotely operated vehicles (ROV) have clearly suggested the existence of species-specific habitat associations (Lorance et al. 2002; Milligan et al. 2016; Soffker et al. 2011; Uiblein et al. 2003). Demonstration of these associations is not easy, because many environmental variables, in addition to fish behavioural selectivity and plasticity may be involved (Baker et al. 2012; Demestre et al. 2000). The existence of strong fish-benthos associations may change our perception of the deep ocean, but it also raises some conservation concerns with regard to deep-sea fisheries because certain fishing practices may affect both the targeted species and the benthic fauna to which they are typically associated (Huvenne et al. 2016).

The present work investigates the distribution of deep-sea fishes using a multi-factor approach. The micro-distribution of fishes was tentatively associated to geological, hydrographic and benthic variables in a semi-quantitative method.

Observations along four transects on the continental slope of the Bay of Biscay (Figure 1) were used to assess if latitudinal or depth gradients affected these small-scale associations (Latrouite et al. 1999).


art fig1

Figure 1. Map showing the four dive stations in the Bay of Biscay, NE Atlantic (Dives 22, 34, 35 and 37). Transects were performed approximately perpendicular to the continental slope.

2. Material and Methods

2.1. Field and image-based observations

The four dives analysed in this study took place in 1998 during a cruise in the Bay of Biscay performed by the submersible “Nautile” of IFREMER, where the main objective was to make direct underwater observations on fishes and invertebrates of commercial interest and the impact of fishing gears. The details of the sampling were given by Latrouite et al. (1999) (Figure 1, Table 1). The submersible cruised 1.5 m above the sea bottom at variable speed, normally within the range of 0.5-0.7 knots. The estimated visual field was about 10 meters wide (5 m on each side) and 10-15 m long (ahead), depending on the turbidity of the water, plankton density, and seabed topography. Both a fixed camera and a mobile camera were operated to record video sequences of the area covered, and navigation and physical records were kept automatically. The methodology differed from normal transect sampling because the submersible made frequent stops to study in detail the habitat surrounding individual fish.


Table 1. Dive transects main characteristics.

Dive 22

Dive 34

Dive 35

Dive 37






GPS Position

47°54'N 08°11W

44°43'N 02°09W

46°15'N 04°34'W

47°28'N 06°41'W

Distance crossed (meters)





Time hours





Duration (minutes)





Depth (meters)





Dive path inclination











Bottom type



Soft sediment



Fine sediment









Sampling units in the present study were individual fish. Other macro-nekton (e.g. cephalopods and crustaceans) were only sporadically observed and were not considered in the analysis.

In the laboratory, videotapes were analysed using a Super VHS video player and a large screen, or processed by means of computer and video with software Microsoft Media Player©®. Recording of habitat factors started immediately after an individual fish appeared in a video sequence. Traits such as size, shape and colour were used for identification of the fish and bottom epifauna. Even under good conditions of light and turbidity only individuals or colonies larger than about 5 cm could be suitably recognized. These were identified to the lowest possible taxonomic level. Microhabitat of each individual was characterized according to depth, temperature, slope inclination, bottom type and complexity, current velocity, benthos type and cover (Tables 1, 2, 3). In this study, microhabitat refers to the ensemble of biotic and abiotic conditions of the fish surroundings.


Table 2. Number of individuals per species per dive transects and codes in brackets.

Fishes Dive 22 Dive 34 Dive 35 Dive 37
Alephocephalidae (Ale) 0 2 5 0
Anguiliformes (Ang) 12 5 18 1
Beryx decadactylus (Ber) 0 0 0 10
Chimaerids (Chi) 7 4 4 36
Coryphaenoides rupestris (Cor) 61 7 103 0
Galeus melastomus (Gal) 0 0 0 0
Helicolenus sp. (Hel) 0 0 0 41
Hoplostethus atlanticus (Hoa) 4 1 334 0
Lepidion eques (Lep) 52 16 12 0
Mesopelagic fish (M) 7 3 0 0
Macrouridae - others (Mac) 15 30 10 5
Mora moro (Mm) 18 4 0 0
Molva molva (Mol) 0 0 0 40
Moridae - others (Mor) 6 2 10 0
Neocyttus helgae (Neo) 0 3 14 0
Notacanthus sp. (Not) 1 3 10 23
Sharks (others) (Sha) 3 3 29 1
Synaphobranchus kaupii (Syn) 57 88 26 0
Trachyscorpia sp. (Tra) 8 0 0 0
Total 251 171 575 157

Table 3. Categories used to define the fish habitats.


Type (unit) or categories


Continuous (m)


Continuous (°C)

Current (Cur)

1. Absent/very low; 2. Low/moderate; 3. High/very high

Slope inclination (Slo)

1. Flat (0-5°); 2. Sloping (5-30°); 3. Steeply sloping (30-45°); 4. Steep (>45°)

Substrate Complexity (Sub)

1. Sedimentary; 2. Structured; 3. Complex

Ripple marks (Rip)

0. None; 1. Weak; 2. Strong

Bottom type

1. Coarse sediment; 2. Fine sediment

Bottom Structure (Bot)

Coverage by clast, rock or hard bottom: 1. (0-5%); 2. (5-25%); 3. (25-50%); 4. >50%

Benthos cover

Actinians (Act), Asteroidea (Ast), Pennatularians (Pen), Crinoids (Cri), Echinoids (Ech), Bryozoans (Bry), Hydroids (Hyd), Sponges (Spo), Scleractinians (Scl), Gorgonians (Gor), Antipatharians (Ant), Sea Cucumbers (Sea), Without benthos (Des)


Following a description of microhabitat use in single dives, a global dive analysis was attempted using multivariate techniques. Owing to the mixed nature of the variables (discrete and continuous), Canonical Correspondence Analysis (CCA) as implemented in the CANOCO 4© software (Leps & Smilauer 2003; Ter Braak & Smilauer 1998) was considerate appropriate for the exploratory analyses. The environmental variables of greater influence in the model fit were chosen by means of automatic forward selection using Monte-Carlo permutation tests (F-test: significance level set at alpha=0.05). A Monte-Carlo permutation test is a test of statistical significance obtained by repeatedly shuffling (permuting) the samples (Leps & Smilauer 2003). 

2.2. Statistical analyses

Prior to the analyses, the species, physical and environmental variables were coded according to the nomenclature given in Tables 1, 2 and 3. Stratified re-sampling of the original observation matrix was required to balance the number of observations between dives and fish species.120 observations were then randomly selected for each transect, and a matrix containing 4 x 120 individuals used for the global analysis.

Similarly, only the 11 most common types of epifauna, as well as one variable coding for absence of epifauna, were used as environmental variables, in order to down weight the occurrence of rare associations. A more detailed description of the numerical analyses is available in Mendes Alves (2003).

3. Results

The study area was heterogeneous with respect to bathymetry and other physical and geological characteristics (Table 1).

A coarse measure of the similarity among fish communities is apparent from the sets of species observed in the three deeper dives (Table 2), the composition and distribution of fish taxa being relatively similar in dives 22 and 34 over soft sediments (r=0.69) (Table 1), negligibly related to the set obtained over hard sediments and strong currents (dive 35). The fish assemblage encountered in the shallow central region (dive 37) differed clearly from those in deeper areas (Table 2).

The most frequent species in dive 22 were Coryphanoides rupestris, Lepidion eques and Synaphobranchus kaupii. Dive 34 revealed a high density of S. kaupii, while an enormous concentration of Hoplostethus atlanticus was found in dive 35. Helicolenus sp. was common along the upper slope (dive 37).

A global analysis of the four dives by means of CCA showed that a large part of the variation in the species-environment relationship could be explained, with partial percentages of 40.5% and 30.3% for the 1st (horizontal) and 2nd axes (vertical) axes, respectively. The 1st canonical axis was statistically significant as well as the relation between species and the environmental variables (p=0.005). The 2nd axis was not significant. A number of physical and geological variables, such as depth, slope, bottom structure, type of substrate, current, water temperature, and occurrence of ripple marks, as well as the prevalence of certain types of epifauna, such as scleractinians, sponges, pennatularians, hydroids and antipatharians, contributed significantly (at the 5% significance level) to the model. The CCA bi-plot illustrates the centres of distribution of the fish species and benthic fauna, and the directions of influence of the main physical and geological variables (Figure 2). The 1st axis, the axis carrying most information in the analysis, contrasted the fish-benthos associations in deep waters (on the left side) to those observed in shallower and warmer areas i.e. dive 37 (on the right side). Shallow and intermediate depths were often characterized by relatively high currents and coarse sediments or hard bottoms (high percentage of clast or rock on the bottom). In shallower areas, the bottom was either devoided of visible benthos or populated by sponges, scleractinians or hydroids. Molva molva, Helicolenus sp., Galeus melastomus and chimaerids were often found in association with these types of benthos.


art fig2

Figure 2.Canonical correspondence analysis (CCA) ordination diagram of all the dives, showing associations between fish species (dots) and environmental variables (arrows) in the Bay of Biscay. Clusters of fish species showing similar association patterns are indicated by circles. Species names and environmental variables were coded for simplification, and their full names are given in Tables 1, 2 and 3.


Fish observed at intermediate and deep waters could not be sorted into discrete clusters. Instead, they were mostly found along a complex ecological gradient parallel to the 2nd plot axis. On or above soft bottoms, Anguilliformes, Moridae, Synaphobranchus kaupii, Lepidion eques and Mora moro were positively associated with Actinians and poorly structured habitat (Figure 2). At the centre of this gradient Coryphaenoides rupestris and other macrourids, mesopelagic fishes, and sharks were associated to pennatularians, asteroids, echinoderms and crinoids. Finally, Hoplostethus atlanticus formed a mono-specific cluster defined by the association with gorgonians, antipatharians and ripple seabed (Figures 2 and 3).


art fig3

Figure 3. (A) Hoplostethus atlanticus and diverse deep sea coral fauna (dive 35); (B) Molva molva and deep-sea scleractinian corals (dive 37); (C) Chimaera and benthos (dive 22); (D) Coryphaenoides rupestris over desert bottom (dive 34).

4. Discussion

The different organisms provided very useful information with regard to the current system prevailing in their habitats. In the case of gorgonians on the edge of hard substrate and rocks (van den Beld et al. 2017) and crinoids above the viscous sub layer of the benthic boundary layer were clear indicators of current flow (Tyler P & Zibrowius H 1992).

The deep-sea habitats investigated in the Bay of Biscay were highly diverse, and fish occurrence showed species-specific patterns in relation to physical, geological and biological characteristics (Table 1, Figure 2).

At the micro-scale level, there were important biological indicators of habitat and formation, particularly the presence of deep-water coral colonies of gorgonians and scleractinians (Figure 3). The occurrence of sponges and pennatularians were valuable indicators of distinct hydrological conditions. The most important physical characteristics determining fish occurrence were depth, temperature and current. Geomorphological factors were likely important, such as the substrate complexity, bottom structure and ripple marks.

Emphasis of the present study focused, however, on environmental heterogeneity as a determinant of fish occurrence at the micro-habitat level. The aim of the survey was multi-purpose. One of the limitative constraints in the investigation of the deep ocean is the inevitable bias imposed by the sampling instruments. The lights and sounds produced by the submergible can attract or scare fish in a selective way. Other factors that could have influenced data collection were sea floor relief, profusion of suspended particles and consequent visibility conditions, size of organisms, and occurrence of mimetic species.

Fish species analysed in the present study showed different distributions according to depth and temperature but this also reflects the different sampling characteristics of the four dives. Dive 37 was particularly shallow and warm (Table 1). It is well known that species inhabiting the deep-sea are zoned with depth (Uiblein et al. 1998). Diversity patterns of demersal fish assemblages can generally be comprehended in terms of interrelationships of predation, competition, environmental heterogeneity, and trophic level (Haedrich et al. 1980).

In the present study water current was qualitatively identified in two different ways, ‘current velocity’ and ‘current temporal characteristics’. Occasionally organisms typical of fast currents, such as gorgonians and antipatharians, were found in habitats showing relatively weak currents. Thus, caution was warranted in the utilization of current as explanatory variable of fish or micro-habitat distributions over larger scales in time and space.


Species-specific patterns and habitat use

Hoplostethus atlanticus was found in dive 35 (Table 2). However, most of these specimens were sampled only in a specific zone of the diving transect. This was a zone of high hydrological activity, rich in corals, like gorgonians and antipatharians, characterised by hard and complex bottoms. As pointed out by Baker et al. (2012) and Husebø A et al. (2002), these fish are probably associated with areas of high turbulence and mixing, and adopt calm areas when recovering between foraging trips. Sea fans and corals, fan-shaped gorgonians (Tyler & Zibrowius 1992) and other suspension feeders are common at sites of flow acceleration (Genin et al. 1986). The association of orange roughy to these organisms and associated habitat features suggests feeding concentrations of this benthopelagic fish (Rosecchi et al. 1988).

Synaphobranchus kaupii preferred areas associated to currents, low subtract complexity and soft bottoms populated by asteroids (Tables 1 and 2; Figure 2). This can probably be explained by its adaptive adjustment of habitat selection and foraging behaviour. The presence of currents in its habitat can probably improve the strategy of food search through the canalization of odour plumes from food sources (Uiblein et al. 2002). The diet of S. kaupii is broad and typical of scavenger species (Priede et al. 1994). However, there was no strong evidence of these types of prey in this study. It is also known that generalist predators show wider distribution ranges than restricted predators (Haedrich et al. 1980). In all the dives S. kaupii appeared associated to cooler and deeper habitats and this is in agreement with other studies for the depth range considered (Merrett & Domanski 1985; Uiblein et al. 2002).

Coryphaenoides rupestris and S. kaupii shared similar habitats, depth and temperature ranges. This species showed a general tendency to environments containing pennatularians, asteroidea, as well as, bottoms characterized by fine sediment and sponges (Table 1; Figure 2). There is, however, strong evidence related to its mode of life and foraging flexibility. C. rupestris feeds upon on zooplankton and small mesopelagic fishes (Mauchline & Gordon 1984). This species seems to be rather more influenced by hydrological and dietary factors than by bottom structure.

Other Macrouridae followed the same patterns of habitat use as Coryphaenoides rupestris, probably due their biological affinities. Macrourids are generally known to have a rhythmic and active feeding behaviour, affected by the tide and transport of food (Guennegan & Rannou 1979; Mauchline & Gordon 1980).

The scorpaenid species Helicolenus sp. (personal observation; Figure 2) were mainly found associated with sponges. Scorpaenid fishes were observed and documented associated to sponges by (Smith & Hamilton 1983) in the Santa Catalina Basin using the submersibles Alvin and Sea Cliff. On the continental shelf off south western Norway (Husebø et al. 2002) noted a similar association for Sebastes sp., fish being often observed close to large sponges, resting or hiding in their concavities, and among stones. This association seems to be explained by the sheltering provided by these organisms (Husebø et al. 2002). Hydroids were occasionally associated to the microhabitat of Helicolenus sp. (Husebø et al. 2002) (Figure 2), but due to their small sizes and forms, accurate measurement of coverage was particularly difficult.

Molva molva, Galeus melastomus and chimaerids dwelled in a similar habitat type of Helicolenus sp. (Figure 2). Within these fish groups, Chimaerids were less associated to complex bottoms, and this is probably explained by their feeding behaviour, benthic affinity and active behaviour (Figure 3) (Lorance et al. 2000).

Little is known about the ecology of Lepidion eques. It has a diet typical of a euryphagic predator, with a wide variety of prey as amphipods, decapods, copepods and mysids (Mauchline & Gordon 1980). Inhabiting a similar habitat (Fig. 2), Mora moro and L. eques as well as other Moridae are known to adopt station holding behaviours (Uiblein et al. 2003). These fish seemed to choose interstices of the substrate as shelter from predators, and at larger sizes tend to occur in different types of habitats. When compared with the other species analyzed in the same dive 37, Beryx decadactylus seemed to prefer areas of higher substrate complexity (Table 1), mainly formed by small reefs of scleractinians and small rocky formations. Similar reasons as the ones suggested above for H. atlanticus, could possibly explain its association to complex reefs (Baker et al. 2012; Lorance & Trenkel 2006). Neocyttus helgae was another species strongly associated with coral colonies (personal observation; Table 2) (Milligan et al. 2016).

5. Conclusion

The current study provides added evidence to foregoing behavioural investigations of demersal fish species that many deep-water habitats are characterized by complex interactions between biological and physical habitat features.

In terms of deep-sea habitat conservation in the Bay of Biscay, destructive fishing methods, like trawling (Latrouite D et al. 1999), targeting important commercial sensitive species, such as H. atlanticus (low growth rate and high longevity), which are to some extent also associated to deep water corals, may destroy and put at risk these ecosystems.

Other threats as marine litter may affect cold water habitats (van den Beld I et al. 2017).             

Similar to other parts of the world (Friedlander et al. 2014), the introduction of deep sea marine protected areas in this region (Huvenne et al. 2016; Sanchez et al. 2013; van den Beld et al. 2017) may help the habitat reestablishment and protection of the deep sea resources, where there is clear evidence of anthropogenic destruction (Latrouite D et al. 1999).


This study was partially funded by the Research Council of Norway through an international bi-lateral research grant to Dário Mendes Alves, a thesis submitted in partial fulfilment of the requirements for the degree of Master of Science in International Fisheries Management. Department of Aquatic Bioscience Norwegian College of Fishery Science University of Tromsø, Norway. We also acknowledge Instituto Camões, Lisboa, Portugal, Municipal Library of Setúbal, Portugal, Library of the University of Tromsø, Norway. Thanks to Dr. H. Zibrowius, University of Marseille, France, Jens Revold, Bjørn Hersoug and Arne Eide, University of Tromsø, Norway.


Baker KD, Haedrich RL, Snelgrove PV, Wareham VE, Edinger EN & Gilkinson KD 2012. Small-scale patterns of deep-sea fish distributions and assemblages of the Grand Banks, Newfoundland continental slope. Deep-Sea Research I 65: 171–188.

Demestre M, Sánchez P & Abelló P 2000. Demersal fish assemblages and habitat characteristics on the continental shelf and upper slope of the north-western Mediterranean. Journal of the Marine Biological Association of the United Kingdom 80: 981-988.

Friedlander AM, Kostantinos A, Stamoulis, JN, Jeffrey CD & Tissot BN 2014. Understanding the Scale of Marine Protection in Hawai‘i: From Community-Based Management to the Remote Northwestern Hawaiian Islands. In: Magnus L. Johnson and Jane Sandell, editors, Advances in Marine Biology 69, Oxford: Academic Press, 153-203 pp.

Huvenne, VA, Bett, BJ, Masson, DG, Le Bas TP & Wheeler, AJ 2016. Effectiveness of a deep-sea cold-water coral Marine Protected Area, following eight years of fisheries closure Biological Conservation 200: 60–69.

Genin A, Dayton L & Spiess F 1986. Corals on seamount peaks provide evidence of current acceleration over deep-sea topography. Nature 322: 59-61.

Guennegan Y & Rannou M 1979. Semi-diurnal, rhythmic activity in deep-sea benthic fishes in the Bay of Biscay. Sarcia 64: 113-116.

Haedrich RL, Rowe GT & Polloni PT 1980. The Megabenthic fauna in the deep sea south of New England, USA. Marine Biology 57: 165-179.

Husebø A, Nøttestad L, Fosså JH, Furevik DM & Jørgensen SB 2002. Distribution and abundance of fish in deep-sea coral habitats. Hydrobiologia 471: 91-99.

Latrouite D, Désaunay Y, De Pontual H, Troadec H, Lorance P, Galgani F, Machado P, Bavouzet G, Noel P, Véron G, Danel P & Dugornay O 1999. Compte-rendu de mission à la mer – OBSERVHAL98 – Observations à finalité halieutique. Research report, IFREMER, Brest, RST 99-01.

Leps J & Smilauer P 2003. Multivariate Analysis of Ecological Data using CANOCO, Cambridge University Press. 248 pp.

Lorance P, Latrouite D & Seret B 2000. Submersible observations of elasmobranch species in the Bay of Biscay. Séret, B., Sire, J-Y. (Eds.), 3rd European Elasmobranch Association Meeting, Bologne sur Mer, Societé France Ichthyologie & IRD, Paris, 26-45.

Lorance P, Uiblein F & Latrouite D 2002. Habitat, behaviour and colour patterns of orange roughy Hoplostethus atlanticus (Pisces: Trachichthyidae) in the Bay of Biscay. Journal of the Marine Biological Association of the United Kingdom 82: 1-11.

Lorance P & Trenkel VM 2006. Variability in natural behaviour, and observed reactions to an ROV, by mid-slope fish species. Journal of Experimental Marine Biology and Ecology 332: 106–119.

Mauchline J & Gordon JD 1980. The food and feeding of the deep sea morid fish Lepidion eques (Gunther, 1887) in the Rockall Trough. Journal of the Marine Biological Association of the United Kingdom 60: 1053-1059.

Mauchline J & Gordon JD 1984. Diets and bathymetric distributions of the macrourid fish of the Rockall Trough, north-eastern Atlantic Ocean. Marine Biology 81: 107-121.

Mendes Alves D 2003. Behaviour and patterns of habitat utilisation by deep-sea fish Analysis of observations recorded by the submersible Nautilus in "98" in the Bay of Biscay, NE Atlantic. A thesis submitted in partial fulfilment of the requirements for the degree of Master Science in International Fisheries Management. Department of Aquatic Bioscience Norwegian College of Fishery Science University of Tromsø. 58pp.

Merrett NR & Domanski PA 1985. Observations on the ecology of deep-sea bottom-living fishes collected off northwest Africa: II. The Moroccan slope (27º-34ºN), with special reference to Synaphobranchus kaupii. Biological Oceanography 3: 349-399.

Milligan RJ, Spence GJ, Roberts M & Bailey DM. 2016. Fish communities associated with cold-water corals vary with depth and substratum type. Deep-Sea Research I 114: 43–54.

Priede IG, Bagley PM, Smith A, Creasey S & Merrett NR 1994. Scavenging deep demersal fishes of the Porcupine Seabight, North-east Atlantic: observations by baited camera, trap and trawl. Journal of the Marine Biological Association of the United Kingdom 74: 481-498.

Rosecchi E, Tracey DM & Webber WR 1988. Diet of orange roughy, Hoplostethus atlanticus (Pisces: Trachichthyidae) on the Challenger Plateau. New Zealand Marine Biology 99: 293-306.

Sánchez, F, Morandeau, G, Bru, N. & Lissardy, M. 2013. A restricted fishing area as a tool for fisheries management: Example of the Capbreton canyon, southern Bay of Biscay, Marine Policy 42, 180–189.

Smith CR & Hamilton SC 1983. Epibenthic megafauna of a bathyal basin off southern California: patterns of abundance, biomass, and dispersion. Deep-Sea Research 30: 907-928.

Soffker M, Sloman KA & Hall-Spencer JM. 2011. In situ observations of fish associated with coral reefs off Ireland. Deep-Sea Research I 58: 818-825.

Ter Braak CJF & Smilauer P 1998. CANOCO reference manual and user's guide to canoco for windows: software for canonical community ordination (version 4) Ithaca, New York, Microcomputer Power

Tyler P & Zibrowius H 1992. Submersible observations of the invertebrate fauna on the continental slope southwest of Ireland (NE Atlantic Ocean). Oceanologica Acta 15: 211-226.

Uiblein F, Bordes F, Castillo R & Ramos AG 1998. Spatial distribution of shelf- and slope-dwelling fishes collected by bottom long-line off Lanzarote and Fuertventura, Canary Islands. Marine Ecology 19: 53-66.

Uiblein F, Lorance P & Latrouite D 2002. Variation in locomotion behaviour in northern cutthroat eel (Synaphobranchus kaupii) on the Bay of Biscay continental slope. Deep Sea Research I 49: 1689-1703.

Uiblein F, Lorance P & Latrouite D. 2003. Behaviour and habitat utilization of seven demersal fish species on the Bay of Biscay continental slope, NE Atlantic. Marine Ecology Progress Series 257: 223-232.

van den Beld I, Bourillet J, Arnaud-Haond S, Chambure L, Davies J, Guillaumont B, Olu K & Menot L. 2017. Cold-water coral habitats in submarine canyons of the Bay of Biscay. Frontiers in Marine Science 4:118. doi: 10.3389/fmars.2017.00118

van den Beld I, Guillaumont B, Menot L, Bayle C, Arnaud-Haond S & Bourillet J. 2017. Marine litter in submarine canyons of the Bay of Biscay. Deep Sea Research Part II: Topical Studies in Oceanography 145: 142-152.