Habitat associations and assemblage structure of demersal deep-sea fishes on the eastern Flemish Cap and Orphan Seamount

Journal Pre-proof Habitat associations and assemblage structure of demersal deep-sea fishes on the eastern Flemish Cap and Orphan Seamount Brynn M. Devine, Krista D. Baker, Evan N. Edinger, Jonathan A.D. Fisher PII:

S0967-0637(19)30356-5

DOI:

https://doi.org/10.1016/j.dsr.2019.103210

Reference:

DSRI 103210

To appear in:

Deep-Sea Research Part I

Received Date: 8 March 2019 Revised Date:

20 December 2019

Accepted Date: 28 December 2019

Please cite this article as: Devine, B.M., Baker, K.D., Edinger, E.N., Fisher, J.A.D., Habitat associations and assemblage structure of demersal deep-sea fishes on the eastern Flemish Cap and Orphan Seamount, Deep-Sea Research Part I, https://doi.org/10.1016/j.dsr.2019.103210. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Elsevier Ltd. All rights reserved.

1

Habitat associations and assemblage structure of demersal deep-sea

2

fishes on the eastern Flemish Cap and Orphan Seamount

3 4

Brynn M. Devine1*, Krista D. Baker2, Evan N. Edinger3, Jonathan A.D. Fisher1

5 6 7

1

Centre for Fisheries Ecosystems Research, Fisheries and Marine Institute of Memorial University of Newfoundland, 155 Ridge Road, St. John’s, NL A1C 5R3, Canada

8 9

2

Department of Fisheries and Oceans Canada, 80 East White Hills Road, St. John’s, NL, A1C 5X1, Canada

10 11

3

Geography Department and Biology Department, Memorial University of Newfoundland, St. John’s, NL, A1B 3X9, Canada

12

*Corresponding author. Email address: [email protected]

13 14 15

Abstract Advancements in remote-sampling and optical technologies have considerably improved

16

our understanding of fish-habitat relationships and assemblage structure in the deep ocean

17

through direct observations. The composition and complexity of seafloor habitats can strongly

18

influence species diversity and distributions, but the relative importance of different

19

microhabitats - both abiotic and biotic - is poorly understood. We examined differences in fish

20

species composition and relative abundance between different physical (sediment type and

21

boulder density) and biological (coral and sponge densities) habitat categories through in-situ

22

observations from remotely-operated vehicle surveys of unglaciated deep-sea features off the

23

coast of Newfoundland, Canada. Fish-habitat relationships were observed across 61 km of

24

seafloor and spanning depths of 875 –3003 m at five dive locations, with additional

25

quantification of fish behaviour and assemblage patterns. Distinct assemblages occurred among

26

depth zones (ANOSIM, Global R=0.47, p=0.001), although global tests of overall biological and

27

physical habitats were not significant. Significant pairwise differences in assemblages were only

28

observed in more complex physical habitats (e.g. boulder fields, and outcrops) and complex

29

biological habitats (e.g. dense corals) compared to less complex areas of fine-grain sediment or

30

locations with no or few corals and sponges present. Our results suggest overall structural

31

complexity of physical and biogenic habitat features may be particularly important to some deep-

32

sea fishes. Until further details of these relationships can be explored, conservation efforts should

33

strive to protect a wide-range of microhabitats to preserve valuable fish habitats in the deep-

34

ocean environment.

35 36

Keywords

37

Deep-sea fishes; Assemblages; Distributions; Remotely Operated Vehicle (ROV); Community

38

analyses; Habitat associations

39 40 41

1. Introduction Habitat composition, complexity, and heterogeneity can be strong predictors of species

42

diversity and abundance in a variety of marine communities (Menge & Sutherland, 1976; Tews

43

et al. 2004; Gratwicke & Speight, 2005). Benthic habitats provide structure and space for shelter

44

and can play an important role in feeding and reproduction of demersal fishes (Beck et al. 2001;

45

Teagle et al. 2017), and influence intra- and inter-species interactions and trophodynamics (Diehl

46

1992; Grabowski & Powers, 2004). In general, habitat complexity and heterogeneity decrease

47

with depth across continental margins, as variability in substrate types, grain size, food

48

availability, and average epifauna size decline (Levin et al. 2001; Carney, 2005). However,

49

transitional slope habitat between the shelf and margin edge, often punctuated by submarine

50

canyons, sediment flows, and subject to internal tides and eddies, can offer complex and variable

51

habitats beyond the shelf edge (Levin & Dayton, 2009; Buhl-Mortensen et al. 2010). Physical

52

structures, including geomorphological features such as outcrops or canyons and variability in

53

slope and substrate types, can add topographic complexity and heterogeneity across the seafloor.

54

Corals, sponges, and other benthic macrofauna add emergent biogenic structural complexity to

55

the seafloor, however, physical habitat factors such as sediment composition and current regimes

56

can in turn largely dictate their distributions (Guinan et al. 2009; Baker et al. 2012a). As habitats

57

become more uniform with increasing depth, the relative importance of structural habitat may

58

also change when contrasted against a landscape with fewer features.

59

Habitat associations for deep-sea fishes are not well-known. The small percentage of the

60

deep seafloor that has been directly observed (Clark et al. 2016) limits our understanding of

61

small-scale patterns in distribution, composition, and availability of different microhabitats in

62

deep-sea benthic ecosystems; as such the role and relative influence of different microhabitats on

63

the distribution and assemblage structure of deep-sea fishes is poorly understood. Advancements

64

in optical technologies and remotely operated vehicles (ROVs) offer the opportunity to observe

65

in-situ fine scale habitat associations of deep demersal fishes to explore habitat use, assemblage

66

composition, and fish behaviour at depth (Ross & Quattrini, 2007; Anderson et al. 2009; Baker et

67

al. 2012b; Milligan et al. 2016). These direct observations are beginning to shed light on fish-

68

habitat relationships in the deep ocean, although many questions regarding temporal variability

69

in habitat use, connectivity between broader habitat patches, and the relative importance of

70

abiotic and biotic habitat factors remain.

71

Several studies have examined fish-habitat associations using both abiotic and biotic

72

habitat variables, with results suggesting both aspects can influence species assemblages and

73

abundance. On the southern slope of the Grand Banks of Newfoundland, Baker et al. (2012b)

74

found geomorphic features such as bedrock outcrops harbour unique fish assemblages and more

75

frequent observations of relatively rare species compared to other habitat types. Substrate or

76

sediment type may also affect fish distributions; pairwise comparisons of various habitat types

77

by Ross et al. (2015) indicated fishes associated with sand substrates differed significantly from

78

fishes in other habitats. Other studies report the role of deep-water corals as important fish

79

habitat (D’Onghia, 2019), with higher fish abundance and species richness (Costello et al. 2005),

80

and higher densities of larger individuals in cold-water coral habitats (Husebo et al. 2002). Off

81

the coast of Norway, deep-water reefs of Lophelia pertusa coral support higher densities of

82

gravid redfish Sebastes viviparous (Fosså et al. 2002), suggesting coral habitats may play an

83

important functional role for some fish species.

84

However, not all studies observed changes in fish abundance and/or assemblage

85

composition in relation to corals (D’Onghia et al. 2010; Baker et al. 2012b; Biber et al. 2014).

86

Auster (2005) found fish communities in dense coral habitats were functionally similar to those

87

within less complex habitats. Although high densities of Sebastes spp. were observed in areas of

88

dense corals as reported in earlier studies (Mortensen et al. 2005; Fosså et al. 2002), Auster

89

(2005) found no significant difference in densities between regions of dense corals and outcrop-

90

boulder habitats with sparse corals, concluding overall habitat complexity may be a better

91

predictor. However, Krieger and Wing (2002) observed 85% of large Sebastes spp. in the Gulf of

92

Alaska associated with coral-covered boulders, despite < 1% of boulders containing corals.

93

Whether species-specific preferences or inter-annual variability in habitat associations explain

94

these discrepancies is unknown and warrants further investigation.

95

Recent technological advances in accessing deep environments have increased global

96

interest in exploitation of deep-sea resources, with rising prospects for mineral mining and oil

97

exploration (Wedding et al. 2015; Cordes et al. 2016) and commercial fisheries pushing into

98

deeper waters (Morato et al. 2006). Slow recovery rates and high vulnerability to physical

99

disturbances often characterize deep-sea communities (Clark et al. 2015). Bottom impacts from

100

human activities can cause significant redistribution and erosion of sediments (Martin et al.

101

2014) with potentially devastating effects on fragile habitat-forming epifauna (Koslow et al.

102

2001; Ramirez-Llodra et al. 2011), such as reduced diversity and changes in biomass,

103

abundance, and community composition in areas exposed to physical disturbances (Baker et al.

104

2009; Clark et al. 2015). For example, many cold-water corals in the Northwest Atlantic grow at

105

rates of 4 – 25 mm per year, with ages of some gorgonian colonies over 200 years (Sherwood &

106

Edinger, 2009) and up to 8,000 years for larger Northeast Atlantic reef systems (Hovland et al.

107

1998), corresponding to centuries-long recovery rates for damaged reefs (Friewald et al. 2004).

108

Likewise, slow growth, late maturation, and longevity in many deep-sea fishes leaves them

109

particularly vulnerable and slow to recover from disturbances such as habitat degradation and

110

overfishing (Koslow et al. 2000; Cailliet et al. 2001; Baker et al. 2009). Therefore, establishing

111

fish-habitat relationships and the value of particular habitats to fishes in the deep ocean can

112

highlight essential habitats and aid direction of marine protection planning and conservation

113

efforts.

114 115

This study opportunistically examined fish-habitat relationships from in-situ video data collected during habitat surveys at five locations along the Flemish Cap and Orphan Seamount in

116

the Northwest Atlantic Ocean. We examine patterns in fish occurrence and assemblages in

117

relation to depth, dive location, and habitat classification factors that separate physical

118

topographic complexity and coral and sponge density, and also examine fish behaviour at each

119

observation. These factors are used to explore the overall influence of complexity of habitat on

120

assemblages and test whether abiotic or biotic factors best predict fish distributions.

121 122

2. Materials and methods

123

2.1 Survey design

124

Benthic habitats and fauna were explored along slopes of the Flemish Cap and the

125

Orphan Seamount located east of the Grand Banks of Newfoundland, Canada (Figure 1) in July

126

2010 using the remotely operated vehicle (ROV) ROPOS (Remotely Operated Platform for

127

Ocean Science) (CSSF, 2010). The objective of this cruise was to survey geology and coral

128

distributions given the unique geomorphological history of these seafloor features (Edinger et al.

129

2011). The Flemish Cap, a circular plateau of continental crust ~58,000 km2 in area, rises to 140

130

m depth approximately 600 km east of Newfoundland; the Flemish Pass, which extends below

131

1000 m depth, separates the Flemish Cap from the Grand Banks shelf (King et al. 1985). Unlike

132

the continental margin of Newfoundland and Labrador, the Flemish Cap was not glaciated during

133

the late Quaternary, hence it hosts a different suite of surficial geological features from the

134

continental margin (Edinger et al. 2011). Located within a mixing zone of the cold northern

135

Labrador Current and warm southerly Gulf Stream, average water temperatures here typically

136

exceed those on the adjacent northern Grand Banks (Stein, 2007). The Orphan Seamount is a

137

volcanic submarine feature located 620 km north-east of St. John’s, Newfoundland, and

138

approximately 9 km east of the southern portion of the Orphan Knoll. The seamount is roughly

139

14 km wide at its base, with a depth of 1932 m at its peak (Pe-Piper et al. 2013).

140

A total of four dives (R1335 – 1337, R1339) were analysed along the south and eastern

141

slopes of the Flemish Cap, and a single dive (R1340) on the Orphan Seamount. As fish

142

distributions were not the survey focus, dive locations were selected based on geological and

143

biological (i.e. corals, sponge) habitats of interest. The location, time, depth, temperature, and

144

orientation of the ROV was recorded at 1-second intervals throughout each dive. Survey depths

145

ranged from 875 to 3003 m, with dive transect distances ranging from 7.7 to 15.6 km (Table 1).

146 147 148

Figure 1 Map indicating locations of each of five ROV dives conducted during a 2010 survey of the Flemish Cap and Orphan Seamount off the Grand Banks of Newfoundland, Canada.

149

150 151 152

2.2 Video processing For all analyses, video from the forward-facing colour camera positioned 0.8 m high on

153

ROPOS was used, with white LED lighting and scale reference lasers placed 10 cm apart. All

154

video footage was analyzed using the open-source VLC media player (Version 2.2.6). Location

155

and depth information for each fish observation were acquired from video timestamps on the

156

ROV navigation log. All fish were identified to the lowest possible taxonomic level based on

157

morphological and behavioural characteristics, using video clips for consultation with taxonomic

158

experts when necessary. Individuals were identified to species when possible, however several

159

groups were identified only to family level (i.e. Myctophidae and Gonostomatidae) as voucher

160

specimens are needed for further taxonomic resolution. Behaviour was recorded for each fish at

161

first observation, including reaction to the ROV (no reaction, avoidance, and attraction), in-situ

162

behaviour (hovering, resting, hiding, and swimming), along with estimated position relative to

163

the seafloor (on bottom <10 cm, off-bottom <100 cm, high-off bottom >100 cm). Positions for

164

fishes associated with vertical habitat (e.g. outcrop) were estimated using the same criteria but in

165

the horizontal plane. To reduce double-counting fish, individuals that approached the ROV from

166

behind were not counted.

167

Habitats within the field of view (approximately 24 m2) at each fish observation were

168

categorized in two ways using physical factors relating to the substrate composition and

169

topography of the seafloor, and using biological factors relating to coral and sponges densities.

170

Physical habitats included: 1) fine-grain sediments, 2) coarse sediments, 3) coarse sediments

171

with boulders, 4) boulder field, and 5) outcrop. ‘Fine-grain sediments’ and ‘coarse sediments’

172

were characterized by either finer sediments (sand, silt, clay, or finer gravel) or coarser

173

sediments (pebble, cobble) (Wentworth, 1922), respectively, with few <3 boulders (>25 cm)

174

present. ‘Coarse sediments with scattered boulders’ described coarser sediments with scattered

175

(n=3-10) boulders. Habitat with numerous (>10) boulders over fine or coarse sediments defined

176

‘boulder fields’, and regions comprised of vertical, cliff-like structures of exposed bedrock were

177

classified as ‘outcrop’. Biological habitats include 1) absent, 2) sparse corals, 3) sparse sponges,

178

4) sparse mixed, 5) dense corals, 6) dense sponges, and 7) dense mixed. ‘Absent’ habitats lacked

179

visible corals or sponges present. The term ‘sparse’ describes habitats with <10 colonies, and

180

‘dense’ habitats were comprised of >10 colonies. ‘Mixed’ refers to presence of both corals and

181

sponges, at the associated density descriptor (i.e. ‘sparse mixed’ contained both coral and

182

sponges at densities <10 colonies). Encrusting corals/sponges and other epibenthic fauna such as

183

crinoids or anemones were not included in biological habitat density estimates.

184

Fish observations in relation to species-specific coral or sponge densities were not

185

examined in the present study, although previous studies documented coral and sponge diversity

186

in these regions (Meredyk, 2017; Miles, 2018). These studies showed variable dominant coral

187

taxa and relative abundance among dives. Lowest coral abundance was observed in dives R1335

188

and R1336 on the southern Flemish Cap, dominated by Acanella spp. and Chrysogorgia

189

agassizii. Dive R1337 on the eastern Flemish Cap encountered the highest coral concentrations,

190

namely Anthomastus spp. and Nephthelidae soft corals, and Isididae corals and Anthomastus spp.

191

dominated Dive R1339 (Miles, 2018). The Orphan Seamount (R1340) supported the highest

192

coral diversity, comprised largely of Zoantharian corals, as well as Isididae and Pennatulacea sea

193

pens (Meredyk, 2017). Corals within all dives were predominately small in size (colonies <20

194

cm) and, although species vary morphologically, they presumably offer similar structure/shelter

195

at these small sizes. Sponges are exceedingly difficult to identify from video data, and the

196

various species have yet to be classified or scored for relative abundance. A vast majority of

197

sponges observed here were small (<10 cm) unidentified, yellow sponges, possibly belonging to

198

the genus Geodia, and a limited number of vase-like and glass sponges (Class Hexactinellida).

199 200

2.3 Data analysis

201

Multivariate analyses were used to explore differences in fish assemblages within and

202

between depths and both physical and biological habitat types. Fish observations were sorted

203

based on the following depth classes: 1) middle slope <1500 m, 2) lower slope 1500 – 2500 m,

204

and 3) margin edge >2500 m. These groups were selected based on structure of depth zones

205

along continental margins (see Buhl-Mortensen et al. 2010) and to ensure adequate

206

representation of physical and biological habitat combinations within each depth class for

207

comparison. Patterns among location and depth groups were explored by calculating taxa

208

richness and Shannon diversity indices (H’) for each dive and depth class, and accumulation

209

curves of taxa richness generated for each dive.

210

To explore whether physical or biological habitat features better predict fish assemblage

211

composition, two separate assemblage analyses were conducted in PRIMER 7 (v7.0.10, Primer-

212

E, Plymouth, UK; Clarke & Gorley, 2015) using abundance values of each taxa per sample unit,

213

where each sample unit uniquely combined dive, depth class, and physical (or biological) habitat

214

types (see methods in Ross & Quattrini, 2007; Baker et al. 2012b). Abundance values of each

215

unique taxa per sampling unit were standardized based on the relative survey distance within

216

each sample unit (i.e. total surveyed distance of each habitat type within each depth class and

217

dive). Fish observations were relatively evenly dispersed along each dive transect (Figures S1-

218

S5), with most observations <100 m apart. Given the uniformity of fish observations, the total

219

distance of each habitat was measured in ArcGIS (ArcMap v10.3.1, ESRI) based on fish

220

observations and associated habitats along each transect, measuring total distance between

221

observation points within each habitat type.

222

A non-metric multidimensional scaling (nMDS, Kruskal and Wish, 1978) ordination plot

223

and associated group-average linked hierarchical cluster dendrogram were generated based on a

224

Bray-Curtis similarity matrix constructed from fourth-root transformed abundance data that

225

down-weighted dominant species and increased relative influence of rare species. Two-way

226

analyses of similarities (ANOSIM) based on Spearman rank correlations tested for differences in

227

fish assemblages based on depth class and both physical and biological habitat types to explore

228

which habitat component influenced fish communities most, and pairwise tests used to explore

229

differences between groups. SIMPER (similarity percentages) subsequently determined which

230

species contributed most to similarities and dissimilarities between groups.

231 232 233

3. Results A total of 66.4 hrs of bottom video footage was captured, covering a total distance of

234

55.3 km and spanning depths from 875 – 3003 m (Table 1) across the five dives. Fine-grain

235

sediment dominated the sampled physical habitat (31.3 km), followed by outcrop (7.8 km) and

236

coarse sediment (7.1 km) habitats (Table 2). Sponges were the dominant biological habitat

237

feature, with ~60% of the surveyed distance comprised of sparse (19.7 km) and dense (12.7 km)

238

sponge habitats (Figure A6). Survey depth coverage was similar between <1500 m and >2500 m

239

groups (15.2 km and 10.6 km, respectively), with highest coverage at depths between 1500 and

240

2500 m (29.4 km) (Table 2).

241 242

3.1 Species occurrence & habitat specificity

243

A total of 6,938 fish were observed, representing at least 45 species or unique taxa from

244

at least 19 families and 16 orders (Table 3; Figure 2, Figure S7). A majority of individuals were

245

identified to species (n=4,392), with the remaining observations to genus (n=173) or family

246

(n=1,298) level. Macrourids were the most abundant taxa for all dives (n=4,065), comprised of at

247

least 13 species, followed by Antimora rostrata (n=993) and unidentified mesopelagic fishes

248

(n=867). Rare species included Apristurus spp. (n=2), Histiobranchus bathybius (n=1),

249

Lipogenys gillii (n=2), and

250 251 252 253 254 255 256

Figure 2 Photos of classified habitat types and fishes observed during 2010 ROV surveys off Flemish Cap and Orphan Seamount: (a) Antimora rostrata swimming over fine grain sediments with no corals or sponges present, (b) sparse corals over coarse sediments with scattered boulders, (c) Chaunax spp. resting over coarse sediments, (d) dense sponges on outcrop wall, (e) Centroscyllium fabricii swimming over sparse sponges, (f) Coryphaenoides rupestris in sparse mixed habitat, (g) Neocyttus helgae among sparse mixed over coarse sediments, (h) Boulder field with dense mixed, (i) Macrourus berglax on bedrock outcrop wall with dense corals.

257

258 259

a

b

c

d

e

f

g

h

i

260 261

Coelorinchus caelorhincus (n=1). Six taxa occurred at all five dive locations: Antimora rostrata,

262

Bathysaurus ferox, Coryphaenoides carapinus, Gaidropsarus spp., Hydrolagus affinis, and

263

Synaphobranchus kaupi. Several taxa occurred only in a single dive, including observations of

264

Cottunculus sp. (R1335 only) and Histiobranchus bathybius (R1336 only) along the southern

265

Flemish Cap, Apristurus spp. and Neocyttus helgae on the eastern Flemish Cap (R1337 only),

266

Coelorinchus caelorhincus on the northeastern slope of the Flemish Cap (R1339 only), and

267

Simenchelys parasitica observed exclusively on the Orphan Seamount (R1340 only). Taxonomic

268

richness was highest in the two shallowest dives (R1335, S= 28; R1337, S=28), followed by

269

R1340 (S=24) and lowest in the deeper waters along the Flemish Cap (R1336, S=21; R1339,

270

S=20), with approximately 50% of local richness observed within the first 5 hours of each dive

271

(mean = 4.8 hrs ± 2.0 S.D.; Figure S8) Calculations of Shannon-Weiner diversity indices for dive

272

locations indicate highest diversity at the Orphan Seamount (R1340, H’=2.34), followed by the

273

shallowest dive R1335 (H’=2.18). Diversity indices for the remaining three dives were quite

274

similar, with an average H’ index of 1.94 (range=1.93-1.95) although lack of replication

275

precluded tests of statistical significance among dives. Taxa richness and diversity were highest

276

(S=42, H’=2.21) at depths along the lower slope (1500-2500m) compared to depths of the

277

middle slope (<1500 m; S=32, H’=2.03) and deep seafloor (>2500m; S=20, H’=2.10).

278

Many taxa spanned a wide range of sampled depths (Figure 3), with Antimora rostrata

279

(876 – 2969 m), Macrourid sp. 2 (1215 – 2969 m), Alepocephalus spp. (1200 – 2890m),

280

Coryphaenoides carapinus (1302 – 2969 m), and Aldrovandia spp. (1319 – 2924 m) spanning

281

the greatest depth range. Several taxa were limited to upper slope locations, including

282

Centroscyllium fabricii (<1243 m), Cottunculus thompsonii (<1215 m), Caelorinchus

283

caelorhincus (<1357 m), and Neocyttus helgae (<1457 m). Simenchelys parasitica (>2750 m),

284

Chaunax spp. (>2334 m), Histiobranchus bathybius (>2323 m), Lipogenys gillii (>2171 m), and

285

Coryphaenoides armatus (>2090 m) were among the deepest-dwelling taxa, observed

286

exclusively below 2000 m. The deepest fishes Coryphaenoides armatus (3003 m),

287

Halosauropsis macrochir (3002 m), and Bathyraja spp. (3001 m) occurred at depths >3000 m.

288

Some taxa occurred almost exclusively in low complexity habitats (Table 4) – either fine

289

or coarse sediments with no or few boulders present – including Centroscyllium fabricii,

290

Cottunculus spp. and Cottunculus thompsonii, and all rajids with the exception of a single

291

Amblyraja jenseni observed along a bedrock outcrop. Ophidiidae and Neocyttus helgae also

292

associated with outcrops in high numbers. Sediment grain size was apparently important for

293

some taxa; Bathysaurus ferox occurred exclusively over fine-grained sediments whereas

294

Chaunax spp. and Lepidion eques were only observed over coarse or hard-bottom habitats. Taxa-

295

specific habitat associations were also observed in regard to biological habitat features (Table 4).

296

Several taxa primarily occurred in areas with no or sparse corals and/or sponges present,

297

including Polyacanthonotus spp., Bathysaurus ferox, Chaunax spp., and Aldrovandia spp. In

298

contrast, Neocyttus helgae, Lepidion eques, Apristurus spp., Coelorincus caelorhincus occurred

299

almost exclusively in dense coral/sponge habitat, whereas Notacanthus chemnitzii, Gaidropsarus

300

spp., and Macrourus berglax occurred in relatively higher numbers in areas of dense coral and/or

301

sponge cover.

302 303 304 305 306

307 308

309

Figure 3 Depth distribution of fish taxa observed during 2010 ROV surveys off Newfoundland, Canada.

310 311

3.2 Fish behaviour In response to the approach of the ROV, most fish (n=5,432) exhibited no obvious

312

reaction (Table 5). Relatively common taxa (those represented by more than two individuals)

313

were not observed to consistently avoid or be attracted to the vehicle. Centroscyllium fabricii

314

occasionally approached the ROV (~27% of encounters), with those individuals actively

315

swimming toward and often circling the ROV. Mesopelagic fishes (i.e. myctophids, unidentified

316

mesopelagic fishes) frequently avoided the ROV, rapidly swimming and colliding with the ROV

317

or the seafloor prior to quickly swimming away. Nearly half of Hydrolagus affinis individuals

318

displayed clear avoidance behaviour, as did a small proportion (~12 – 30% of encounters) of

319

some macrourid species. However, individuals of both H. affinis and macrourids were often

320

relatively high off the seafloor, and therefore potentially had a greater propensity for avoidance

321

behaviour to prevent collision with the moving ROV.

322

In-situ behaviour of fishes at first observation (Table 5) indicate a vast majority of

323

individuals either actively swimming (n=3,082) or hovering above the seafloor (n=3,689). Far

324

fewer individuals were observed resting on the seafloor (n=148) or hiding within habitats (n=19).

325

Most macrourids were hovering, with the exception of Coryphaenoides armatus, which was

326

almost as likely to be actively swimming (~42%). Several cryptic taxa, including Bathysaurus

327

ferox, Chaunax spp., and Cottunculus spp., were consistently observed resting on the seafloor.

328

Only Gaidropsarus spp. and Coryphaenoides carapinus individuals exhibited hiding behaviour

329

by more than 2 individuals within a taxon, which hid beneath overhanging boulders or outcrop

330

ledges, or sheltering within or underneath corals or sponges.

331

Most fish closely associated with the seafloor (n=3629), as opposed to off-bottom

332

(n=1304) or high off-bottom (n=962). Position relative to the seafloor varied between taxa, both 17

333

among and within familial groups (Table 5). Many individuals found predominately or

334

exclusively on-bottom belonged to taxa with morphologies suited to a benthic existence,

335

specifically members of the families Rajiidae, Chaunacidae, Psychrolutidae, and Bathysauridae.

336

A majority of individuals in several other taxa also occurred predominately on-bottom, including

337

Gaidropsarus spp., Lepidion eques, Ophidiidae, and Reinhardtius hippoglossoides. Taxa

338

observed more frequently not directly associated with the seafloor include all mesopelagic

339

groups (Myctophidae, Gonostomatidae, and unidentified mesopelagics), Simenchelys parasitica,

340

Lipogenys gillii, and Hydrolagus affinis. Macrourids differed notably in position, with

341

Macrourus berglax and Nezumia bairdi nearly exclusively found on-bottom (>95%

342

encountered). Coryphaenoides carapinus and Coryphaenoides rupestris were also encountered

343

more often on-bottom (~ 80% and 60%, respectively); however, Coryphaenoides armatus were

344

observed on-bottom and off-bottom in near equal frequency. Additional variations between

345

closely related taxa were also observed, namely between notacanthids (~62% Polyacanthonotus

346

spp. on-bottom versus ~63% Notacanthus chemnitzii off-bottom) and halosaurids (~77%

347

Halosaurus machochir on-bottom versus ~56% Aldrovandia spp. off- or high off-bottom).

348 349 350

3.3 Assemblage analyses For both physical and biological assemblage analyses, the global ANOSIM indicated

351

significant differences in assemblages between depth groups (physical: R=0.47, p=0.001;

352

biological: R=0.45, p=0.001), and nMDS plots for both analyses clearly illustrated these

353

differences (Figure 4). Assemblages also differed significantly between each depth category in

354

pairwise comparisons (R≥0.35, p<0.002). SIMPER results from both habitat assemblage

355

analyses (Table S1-S2) indicate the macrourids Coryphaenoides carapinus, Coryphaenoides 18

356

rupestris, and Macrourus berglax were the main contributors (cumulative ~35%) to dissimilarity

357

between the middle slope (<1500 m) and both lower slope (1500 – 2500 m; Avg. Dissimilarity:

358

64.6 – 67.8%) and margin edge (>2500 m; Avg. Dissimilarity: 87.8 – 90.3%) depth groups,

359

driven by higher abundance of Macrourus berglax in shallowest depth group and relatively few

360

Coryphaenoides carapinus at depths below 1500 m. Dissimilarity between lower slope and

361

margin edge (Avg. Dissimilarity: 63.6 – 67.2%) mainly resulted from higher abundance of

362

Coryphaenoides armatus and fewer Antimora rostata in the deepest depth class.

363 364 365 366

Figure 4 MDS plots of Bray-Curtis similarity matrices based on fish observations across (a) physical habitats, (b) depth groups within physical habitat matrix, (c) biological habitats, and (d) depth groups within biological habitat matrix. Dotted lines indicate 50% similarities based on dendrogram cluster analyses.

19

367 368

Global ANOSIM did not indicate significant differences in assemblages across all physical

369

habitats (R=0.02, p=0.40; Table S3). However, pairwise tests showed a significant difference in

370

assemblages between outcrop and fine-grain sediment habitats (R=0.29, p=0.032) and between

371

boulder field and fine-grain sediments (R=0.23, p=0.039), but no differences between other

372

physical habitats. The high relative abundance of Antimora rostrata and Coryphaenoides

373

carapinus across all physical habitats greatly influenced similarity within groups, with these two

374

species contributing a combined 38 – 71% similarity in each group. The absence of

375

Halosauropsis macrochir, myctophids, macrourid sp. 1, and relatively low abundance of

376

Synaphobranchus kaupi on boulder fields all contributed to dissimilarity between boulder field

377

and fine-grain sediment habitat. The absence of Bathysaurus ferox and relatively low abundance

378

of Synaphobranchus kaupi, myctophids, Halosauropsis macrochir and higher relative abundance

379

of Gaidropsarus spp. and ophidiids on outcrops all contributed to dissimilarity between outcrop

380

and fine-grain sediment habitats.

381

Among biological habitats, global ANOSIM was not statistically significant (R=0.09,

382

p=0.061; Table S4). Despite this, pairwise test results yielded significant comparisons, including

383

dense coral versus absent (R=0.46, p=0.008) and sparse sponge versus dense coral (R=0.37,

384

p=0.018) habitats. Again, the high relative abundance of Antimora rostrata and Coryphaenoides

385

carapinus greatly influenced similarity within biological habitat groups. These two species were

386

the main contributors within most groups, with the exception of Macrourus berglax contributing

387

>45% similarity in dense coral habitat. As a result, the relatively high abundance of Macrourus

388

berglax in these habitats, along with the lower relative abundance of Antimora rostrata and

389

Synaphobranchus kaupi contributed the greatest dissimilarity between dense coral habitat and 20

390

areas with no sponges or corals present. The absence of Coryphaenoides armatus and macrourid

391

sp.2 in dense coral habitat and higher relative abundance of Macrourus berglax and myctophids

392

and lower abundance of Antimora rostrata and Synaphobranchus kaupi contributed to

393

dissimilarity between dense corals and sparse sponges habitats.

394 395 396 397

4. Discussion We observed distinct fish assemblages between middle slope (<1500 m), lower slope

398

(1500 – 2500 m), and margin edge (>2500 m) depth zones. The influence of physical and

399

biological habitats on assemblages varied, with significant differences only for habitats of

400

contrasting topographical relief. This suggests that habitat complexity may be important for

401

some deep demersal fishes, with significant differences between habitats of lowest complexity

402

(i.e. absent, sparse sponge) and highest complexity (i.e. dense corals, outcrops, boulder fields).

403

Multiple studies report the influence of depth on fish assemblage structure throughout the

404

world’s oceans. Off southern Australia, Williams et al. (2018) observed differences in fish

405

assemblages between depth classes along the slope (200 – 3000 m), in addition to changes in

406

biomass, diversity, and density of fishes across depth gradients. Vertical changes in assemblage

407

composition with depth can often be associated with vertical changes in water masses and/or

408

variables associated with unique water masses (i.e. oxygen and temperature) (Menezes et al.

409

2009; Williams et al. 2018). Changes in water masses, along with changes in topography and

410

food availability, are identified as the main drivers behind depth-structured demersal fish

411

assemblages (Haedrich, 1997; Clark et al. 2010), with extrapolation to the deep pelagic

21

412

environment where hydrography and food availability similarly influence vertical distribution of

413

fish species with depth (Sutton, 2013).

414

Pairwise tests for both physical and biological habitat analyses suggest overall habitat

415

complexity as an important indicator of fish assemblages. The two most topographically

416

complex physical habitats – boulder fields and bedrock outcrops - hosted significantly different

417

assemblages compared to less complex substrates. Likewise, fish assemblages differed

418

significantly between biological habitats of dense corals versus both absent and sparse sponges

419

habitats. Dense corals were observed on all physical habitat types (Figure S6) and appeared

420

particularly important to some fish species.

421

Habitat complexity can strongly affect the structure and dynamics of fish communities,

422

with positive relationships between faunal richness and abundance with increased complexity in

423

a variety of marine ecosystems (Luckhurst & Luckhurst, 1978; Gratwicke & Speight, 2005;

424

Caddy, 2007; Linley et al. 2017; D’Onghia 2019). The presence of structures such as corals and

425

sponges can provide shelter from predators, and these emergent structures can modify near-bed

426

hydrodynamics, altering water flow patterns that may enhance particle entrainment or refuge

427

from currents (Zedel & Fowler, 2009; Buhl-Mortensen et al. 2010). By providing shelter and

428

enhanced food supply, as well as potential spawning and nursery habitats, these features may

429

enrich local productivity and improve overall fitness for some species. Complexity varies with

430

species-specific morphology and size (Buhl-Mortensen et al. 2010). Even smaller sponges may

431

provide adequate complexity in dense concentrations (Beazley et al. 2015), although small

432

sponges observed in the present study did not yield significant pairwise tests between dense

433

sponges and lower complexity habitats.

22

434

Variable relationships between non-reef forming cold-water coral species and fish

435

assemblages have been reported. Ross & Quattrini (2007) found unique fish assemblages

436

associated with cold-water corals along the US Atlantic slope. Others have shown species-

437

specific fish associations with gorgonian corals (Krieger & Wing, 2002; Mortensen et al. 2005)

438

and soft corals (Heifetz, 2002). However, Baker et al. (2012a) found no relationship between

439

corals and fish assemblages, regardless of coral density and colony size. Likewise, based on

440

analyses from over 100 submersible dives off southern California, Tissot et al. (2006) concluded

441

that despite the co-occurrence of fishes and structure-forming invertebrates in the same physical

442

habitats, the two groups were not necessarily functionally related. These contradictory

443

conclusions highlight the need for further sampling to determine the level of microhabitat

444

associations and the factors that influence their variability.

445

As in previous work along the Newfoundland slope waters (Baker et al. 2012a), we

446

observed two species relatively rare to the region, Neocyttus helgae and Lepidion eques, in

447

strong association with outcrop and/or boulder habitats. These two species also occurred almost

448

exclusively in areas of high coral density (dense corals or dense mixed), limiting observations to

449

the most physically and biologically complex habitats. Similar studies indicate close association

450

of Neocyttus helgae with cold-water corals in the NE Atlantic (Costello et al. 2005; Milligan et

451

al. 2016), potentially utilizing these biological and physical structures for foraging or flow refuge

452

behaviours (Auster et al. 2005). In the NE Atlantic Lepidion eques has been found to associate

453

with complex substrates and emergent structures such as solitary corals and gorgonians (Alves,

454

2003; Söffker et al. 2011). Fewer direct observations of Lepidion eques exist in the NW Atlantic

455

beyond our study, with limited other western Atlantic ROV surveys (Baker et al. 2012a; Biber et

23

456

al. 2014; Lepidion sp. from Quattrini et al. 2017) also reporting associations with complex

457

habitats (outcrops, dense corals).

458

The benefits of using ROVs for non-extractive survey of deep fish populations are far-

459

reaching, preserving habitats and providing in situ direct observations of fish-habitat

460

associations. Multiple studies have explored the behaviour of motile fishes in response to the

461

presence of an ROV, reporting that fishes may respond to the presence of light, noise, and water

462

and/or sediment displacement generated by the vehicle (Trenkel et al. 2004; Stoner et al. 2008).

463

In our survey, a majority of individuals did not noticeably react to the ROV, and we observed

464

strong avoidance or attraction behaviours in only in a few taxa. Species-specific attraction or

465

avoidance behaviour has been observed (Lorance and Trenkel, 2006; Baker et al. 2012a),

466

potentially biasing fish observations from ROV surveys. However, a recent study comparing

467

ROV and trawl survey methods reported 20 times more fish observations with ROVs (Ayma et

468

al. 2016). Although this finding does not exclude the possibility of the ‘false absence’ of a

469

species that actively avoids the ROV and remains out of the field of view, Ayma et al. (2016)

470

results in conjunction with issues of catchability and species avoidance also inherent to trawls

471

(Winger et al. 2010) suggest that ROVs may provide a better indication of relative species

472

abundance.

473

Noting limited knowledge about habitat relationships in the deep sea, variables such as

474

temperature, current regimes, food availability and potentially other unknown factors not

475

explored here could drive small-scale patterns in fish distributions. We considered biological

476

habitat exclusively in the context of corals and sponges, however, several other habitat-forming

477

epibenthic organisms (e.g. stalked crinoids, urchins, and anemones) were also present, although

478

fewer in number. Crinoid beds along the shelf-break off the coast of western Italy in the

24

479

Mediterranean Sea support high fish densities, particularly of juveniles and spawners (Colloca et

480

al. 2004). Most research on the role of biotic habitats for fishes in the deep ocean have focused

481

on corals and sponges; whether other epibenthic megafauna may also influence species

482

assemblages requires further study (Buhl-Mortensen et al. 2010).

483

Our study surveyed fish during a period of a few weeks within a single year, therefore we

484

cannot preclude the possibility that our observations represent seasonal or life stage-based

485

subsets of habitat associations for some species. Some fish rely on specific microhabitats for

486

spawning substrate, as evidenced by the attachment of skate and shark egg cases to gorgonian

487

corals (Ebert and Davis, 2007; Etnoyer and Warrenchuk, 2007) and numerous accounts of teleost

488

egg deposition within a variety of sponges (Barthel, 1997; Busby et al. 2012; Chernova, 2014).

489

Catsharks (Chondrichthyes: Scyliorhinidae) in particular use a wide variety of benthic

490

invertebrates for egg case attachment (Vazquez et al. 2018), potentially explaining the presence

491

of Apristurus spp. only in dense coral habitats. Some researchers have suggested that deep-sea

492

fishes could use these habitats and other structures facultatively, and that fishes associate with

493

habitat structures based on their complexity, regardless of the composition. Better understanding

494

of how microhabitats influence the demographics of deep-sea fishes requires surveys of fish-

495

habitat associations spanning a variety of locations and range of available habitat types, both

496

geographically and across seasons and life history stages. While microhabitats were examined as

497

categorical coral and sponge densities, future studies should examine fish relationships with

498

enumerated concentrations of specific coral, sponge, and other epibenthic habitat-forming

499

species.

500 501

Given our rudimentary understanding of the role of various habitats in the deep sea, habitat conservation should implement an approach to ensure preservation of a wide variety of

25

502

habitats. Habitat availability presumably affects habitat specialists more than generalists

503

(Swihart et al. 2003), therefore predicting how potential threats to deep-sea habitats might impact

504

fish communities hinges upon understanding how fish communities partition across various

505

microhabitats. Given increased cumulative impacts of human activities in much of the world’s

506

oceans (Halpern et al. 2015), and despite limited understanding of habitat extent in the deep

507

ocean, increased global interest in deep benthic resources could potentially alter habitat

508

landscapes in the deep sea (Ramirez-Llodra et al. 2011).

509

Despite the vulnerability of deep-sea fishes to habitat disturbance and overfishing, efforts

510

to conserve deep-sea habitats in the waters off Newfoundland, Canada have been modest and

511

controversial. As part of Canada’s commitment to designate 10% of its waters as marine-

512

protected and conservation areas by 2020, the Northeast Newfoundland Slope Closure Area was

513

announced in December 2017, adding 55,353 km2 toward this goal. This region spans a range of

514

depths and contains some areas with high concentrations of corals and sponges. While it is

515

closed to all bottom fishing activities, roughly 35% of the reserve remains open to offshore oil

516

and gas exploration (WWF, 2018). Although the impacts of infrastructure installation are

517

typically restricted to a localized area (~100 m) surrounding the drill site, impacts from

518

discharges can reach distances exceeding 2 km (Cordes et al. 2016), and impacts of a major spill

519

could extend much farther. Sedimentation and displacement by infrastructure could have severe

520

and persistent impacts on fragile habitats within the area, but whether these effects extend to fish

521

communities is unknown.

522 523 524

5. Conclusions 26

525

Ensuring protection of essential fish habitats and adhering to ecosystem-based

526

management initiatives requires better knowledge of fish-habitat relationships. We found distinct

527

fish assemblages based on depth but evidence of physical and biological habitat factors was

528

equivocal, with significant comparisons observed only between the least and most complex

529

habitats (i.e. outcrops, boulder fields, dense corals) . Although this study elucidates broad

530

patterns in assemblage structure, more in situ research is needed to further explore these

531

relationships, the relative importance of specific biological habitats, and if this importance varies

532

seasonally or throughout ontogeny. Until these relationships are identified, management and

533

conservation of deep demersal fish and fisheries requires greater attention to habitat conservation

534

and should strive to preserve a wide range of both physical and biological benthic habitats.

535 536

Acknowledgements

537

We would like to thank the crew of CCGS Hudson and the staff of the Canadian Scientific

538

Submersible Facility for ROPOS operation and support. We also thank taxonomic experts Sheila

539

Atchison, David Smith, and Kenneth Tighe for their help with fish identifications.

540 541

Funding: This work was supported by Fisheries and Oceans Canada International Governance

542

Strategy, NSERC Discovery Grants awarded to E.Edinger and J.Fisher, NSERC PGS-D

543

fellowship to K.Baker, OSIRA grant from the NL Research and Development Corporation

544

awarded to B.Devine, and a NSERC ship-time grant to E. Edinger and colleagues.

545

Declarations of Interest: None

27

546 547

References

548

Alves D.M. (2003) Behaviour and patterns of habitat utilisation by deep-sea fish. Analysis of

549

observations recorded by the submersible Nautilus in" 98" in the Bay of Biscay, NE

550

Atlantic. Master's thesis, Universitetet i Tromsø.

551

Anderson T.J., Syms C., Roberts D.A., Howard D. F. (2009) Multi-scale fish–habitat

552

associations and the use of habitat surrogates to predict the organisation and abundance

553

of deep-water fish assemblages. Journal of Experimental Marine Biology and Ecology

554

379: 34-42.

555 556 557

Auster P.J. (2005) Are deep-water corals important habitats for fishes? In: Cold-water Corals and Ecosystems. Springer, Berlin, Heidelberg: pp. 747-760. Auster P.J., Moore J., Heinonen K.B., Watling, L. (2005) A habitat classification scheme for

558

seamount landscapes: assessing the functional role of deep-water corals as fish habitat.

559

In: Cold-water corals and ecosystems. Freiwald A., Roberts J.M. (eds). Springer, Berlin,

560

Heidelberg, pp 761-769.

561

Ayma A., Aguzzi J., Canals M., Lastras G., Bahamon N., Mechó A., Company J.B. (2016)

562

Comparison between ROV video and Agassiz trawl methods for sampling deep water

563

fauna of submarine canyons in the Northwestern Mediterranean Sea with observations on

564

behavioural reactions of target species. Deep Sea Research Part I: Oceanographic

565

Research Papers 114: 149-159.

566

Baker K.D., Devine J.A., Haedrich R.L. (2009) Deep-sea fishes in Canada’s Atlantic: population

567

declines and predicted recovery times. Environmental Biology of Fishes 85: 79.

568

Baker K.D., Wareham V.E., Snelgrove P.V., Haedrich R.L., Fifield D.A., Edinger E.N.,

569

Gilkinson K.D. (2012a) Distributional patterns of deep-sea coral assemblages in three

570

submarine canyons off Newfoundland, Canada. Marine Ecology Progress Series 445:

571

235-249.

28

572

Baker K.D., Haedrich R.L., Snelgrove P.V., Wareham V.E., Edinger E.N., Gilkinson K.D.

573

(2012b) Small-scale patterns of deep-sea fish distributions and assemblages of the Grand

574

Banks, Newfoundland continental slope. Deep Sea Research Part I: Oceanographic

575

Research Papers 65: 171-188.

576

Barthel D. (1997) Fish eggs and pentacrinoids in Weddell Sea hexactinellids: further examples

577

for the structuring role of sponges in Antarctic benthic ecosystems. Polar Biology 17: 91-

578

94.

579

Beazley, L., Kenchington, E., Yashayaev, I., & Murillo, F. J. (2015). Drivers of epibenthic

580

megafaunal composition in the sponge grounds of the Sackville Spur, northwest

581

Atlantic. Deep Sea Research Part I: Oceanographic Research Papers 98: 102-114.

582

Beck M.W., Heck K.L., Able K.W., Childers D.L., Eggleston D.B., Gillanders B.M., Halpern B.,

583

Hays C.G., Hoshino K., Minello T.J., Orth R.J. (2001) The identification, conservation,

584

and management of estuarine and marine nurseries for fish and invertebrates: a better

585

understanding of the habitats that serve as nurseries for marine species and the factors

586

that create site-specific variability in nursery quality will improve conservation and

587

management of these areas. Bioscience 51: 633-641.

588

Biber M.F., Duineveld G.C., Lavaleye M.S., Davies A.J., Bergman M.J., van den Beld I.M.

589

(2014) Investigating the association of fish abundance and biomass with cold-water

590

corals in the deep Northeast Atlantic Ocean using a generalised linear modelling

591

approach. Deep Sea Research Part II: Topical Studies in Oceanography 99: 134-145.

592

Buhl‐Mortensen L., Vanreusel A., Gooday A.J., Levin L.A., Priede I.G., Buhl‐Mortensen P.,

593

Gheerardyn H., King N.J., Raes M. (2010) Biological structures as a source of habitat

594

heterogeneity and biodiversity on the deep ocean margins. Marine Ecology 31: 21-50.

595

Busby M.S., Blood D.M., Fleischer A.J., Nichol D.G. (2012) Egg deposition and development of

596

eggs and larvae of bigmouth sculpin (Hemitripterus bolini). Northwestern Naturalist 93:

597

1-16.

598 599

Caddy J.F. (2007). Marine habitat and cover: their importance for productive coastal fishery resources. UNESCO, Paris, France.

29

600

Cailliet G.M., Andrews A.H., Burton E.J., Watters D.L., Kline D.E., Ferry-Graham L.A. (2001)

601

Age determination and validation studies of marine fishes: do deep-dwellers live

602

longer? Experimental Gerontology 36: 739-764.

603 604 605

Carney R.S. (2005) Zonation of deep biota on continental margins. In: Oceanography and Marine Biology. CRC Press: pp. 221-288. Chernova N.V. (2014) New species of the genus Careproctus (Liparidae) from the Kara Sea with

606

notes on spongiophilia, reproductive commensalism between fishes and sponges

607

(Rossellidae). Journal of Ichthyology 54: 501-512.

608

Clark M.R., Althaus F., Williams A., Niklitschek E., Menezes G.M., Hareide N.R., Sutton P.,

609

O’Donnell C. (2010) Are deep‐sea demersal fish assemblages globally homogenous?

610

Insights from seamounts. Marine Ecology 31: 39-51.

611

Clark M.R., Althaus F., Schlacher T.A., Williams A., Bowden D.A., Rowden A.A. (2015) The

612

impacts of deep-sea fisheries on benthic communities: a review. ICES Journal of Marine

613

Science 73: i51-i69.

614 615 616 617 618

Clark M.R., Consalvey M., Rowden A.A. (2016) Biological sampling in the deep sea. John Wiley and Sons. Clarke K.R., Gorley R.N. (2015) Getting started with PRIMER v7. PRIMER-E: Plymouth, Plymouth Marine Laboratory. Colloca F., Carpentieri P., Balestri E., Ardizzone G.D. (2004) A critical habitat for

619

Mediterranean fish resources: shelf-break areas with Leptometra phalangium

620

(Echinodermata: Crinoidea). Marine Biology 145: 1129-1142.

621

Cordes, E.E., Jones, D.O., Schlacher, T.A., Amon, D.J., Bernardino, A.F., Brooke, S., Carney,

622

R., DeLeo, D.M., Dunlop, K.M., Escobar-Briones, E.G. and Gates, A.R., 2016.

623

Environmental impacts of the deep-water oil and gas industry: a review to guide

624

management strategies. Frontiers in Environmental Science, 4, p.58.

625 626

Costello M.J., McCrea M., Freiwald A., Lundälv T., Jonsson L., Bett B.J., van Weering T.C., de Haas H., Roberts J.M., Allen D. (2005) Role of cold-water Lophelia pertusa coral reefs

30

627

as fish habitat in the NE Atlantic. In: Cold-water Corals and Ecosystems. Springer Berlin

628

Heidelberg: pp. 771-805.

629

CSSF (2010). Canadian Scientific Submersible Facility. http://www.ropos.com

630

Diehl S. (1992) Fish predation and benthic community structure: the role of omnivory and

631 632

habitat complexity. Ecology 73: 1646-1661. D’Onghia G., Maiorano P., Sion L., Giove A., Capezzuto F., Carlucci R., Tursi A. (2010) Effects

633

of deep-water coral banks on the abundance and size structure of the megafauna in the

634

Mediterranean Sea. Deep Sea Research Part II: Topical Studies in Oceanography 57:

635

397-411.

636

D’Onghia G. (2019) 30 Cold-Water Corals as Shelter, Feeding and Life-History Critical Habitats

637

for Fish Species: Ecological Interactions and Fishing Impact. In: Orejas C., Jiménez C.

638

(eds) Mediterranean Cold-Water Corals: Past, Present and Future. Coral Reefs of the

639

World, vol 9. Springer, Cham.

640 641

Ebert D.A., Davis C.D. (2007) Descriptions of skate egg cases (Chondrichthyes: Rajiformes: Rajoidei) from the eastern North Pacific. Zootaxa 1393: 1-18.

642

Edinger E.N., Sherwood O.A., Piper D.J., Wareham V.E., Baker K.D., Gilkinson K.D., Scott

643

D.B. (2011). Geological features supporting deep-sea coral habitat in Atlantic

644

Canada. Continental Shelf Research 31(2): S69-S84.

645 646 647

Etnoyer P., Warrenchuk J. (2007) A catshark nursery in a deep gorgonian field in the Mississippi Canyon, Gulf of Mexico. Bulletin of Marine Science 81: 553-559. Ewing G.P., Kilpatrick R. (2014) Estimating the gear footprint of demersal trawl and longline

648

fishing gears used in the Heard Island and McDonald Islands fisheries. In: Welsford

649

D.C., Ewing G.P., Constable A.J., Hibberd T., Kilpatrick R. (eds.). Demersal fishing

650

interactions with marine benthos in the Australian EEZ of the Southern Ocean: An

651

assessment of the vulnerability of benthic habitats to impact by demersal gears. Final

652

Report FRDC Project 2006/042, Department of the Environment, Australian Antarctic

653

Division and the Fisheries Research and Development Corporation, Hobart, Tasmania,

654

Australia: pp 265. 31

655

Fosså J.H., Mortensen P.B., Furevik D.M. (2002) The deep-water coral Lophelia pertusa in

656

Norwegian waters: distribution and fishery impacts. Hydrobiologia 471: 1-12.

657

Grabowski J.H., Powers S.P. (2004) Habitat complexity mitigates trophic transfer on oyster

658 659

reefs. Marine Ecology Progress Series 277: 291-295. Gratwicke B., Speight M.R. (2005) The relationship between fish species richness, abundance

660

and habitat complexity in a range of shallow tropical marine habitats. Journal of Fish

661

Biology 66: 650-667.

662

Guinan J., Grehan A.J., Dolan M.F., Brown C. (2009) Quantifying relationships between video

663

observations of cold-water coral cover and seafloor features in Rockall Trough, west of

664

Ireland. Marine Ecology Progress Series 375: 125-138.

665 666 667

Haedrich, R. L. (1997). 3 Distribution and Population Ecology. In Fish Physiology (Vol. 16, pp. 79-114). Academic Press. Halpern B.S., Frazier M., Potapenko J., Casey K.S., Koenig K., Longo C., Lowndes J.S.,

668

Rockwood R.C., Selig E.R., Selkoe K.A., Walbridge S. (2015) Spatial and temporal

669

changes in cumulative human impacts on the world’s ocean. Nature Communications 6:

670

7615.

671 672 673

Heifetz J. (2002) Coral in Alaska: distribution, abundance, and species associations. Hydrobiologia 471: 19-28. Hovland M., Mortensen P.B., Brattegard T., Strass P., Rokoengen K. (1998) Ahermatypic coral

674

banks off mid-Norway: evidence for a link with seepage of light hydrocarbons. Palaios

675

13: 189–200.

676 677 678

Husebø Å., Nøttestad L., Fosså J.H., Furevik D.M., Jørgensen S.B. (2002) Distribution and abundance of fish in deep-sea coral habitats. Hydrobiologia 471: 91-99. King L.H., Fader G.B., Poole W.H., Wanless R.K. (1985) Geological setting and age of the

679

Flemish Cap granodiorite, east of the Grand Banks of Newfoundland. Canadian Journal

680

of Earth Sciences 22: 1286–1298.

32

681

Koslow J., Boehlert G.W., Gordon J.D.M., Haedrich R.L., Lorance P., Parin N. (2000)

682

Continental slope and deep-sea fisheries: implications for a fragile ecosystem. ICES

683

Journal of Marine Science 57: 548-557.

684

Koslow J.A., Gowlett-Holmes K., Lowry J.K., O’hara T., Poore G.C.B., Williams A. (2001)

685

Seamount benthic macrofauna off southern Tasmania: community structure and impacts

686

of trawling. Marine Ecology Progress Series 213: 111-125.

687 688

Krieger K.J., Wing B.L. (2002) Megafauna associations with deepwater corals (Primnoa spp.) in the Gulf of Alaska. Hydrobiologia 471: 83-90.

689

Kruskal, J. B., & Wish, M. (1978). Multidimensional scaling. 1978. Beverly Hills, CA.

690

Levin L.A., Etter R.J., Rex M.A., Gooday A.J., Smith C.R., Pineda J., Stuart C.T., Hessler R.R.,

691

Pawson D. (2001) Environmental influences on regional deep-sea species

692

diversity. Annual Review of Ecology and Systematics 32: 51-93.

693 694 695

Levin L.A., Dayton P.K. (2009) Ecological theory and continental margins: where shallow meets deep. Trends in Ecology and Evolution 24: 606-617. Linley T.D., Lavaleye M., Maiorano P., Bergman M., Capezzuto F., Cousins N.J., D’Onghia G.,

696

Duineveld G., Shields M.A., Sion L., Tursi, A., Priede I.G. (2017) Effects of cold‐water

697

corals on fish diversity and density (European continental margin: Arctic, NE Atlantic

698

and Mediterranean Sea): Data from three baited lander systems. Deep Sea Research Part

699

II: Topical Studies in Oceanography 145: 8-21.

700

Lorance P., Trenkel V.M. (2006) Variability in natural behaviour, and observed reactions to an

701

ROV, by mid-slope fish species. Journal of Experimental Marine Biology and Ecology

702

332: 106-119.

703 704 705

Luckhurst B.E., Luckhurst K. (1978) Analysis of the influence of substrate variables on coral reef fish communities. Marine Biology 49: 317-323. Martin J., Puig P., Masque P., Palanques A., Sanchez-Gomez A. (2014) Impact of bottom

706

trawling on deep-sea sediment properties along the flanks of a submarine canyon. PLoS

707

One 9: p.e104536.

33

708

Menezes, G. M., Rosa, A., Melo, O., & Pinho, M. R. (2009). Demersal fish assemblages off the

709

Seine and Sedlo seamounts (northeast Atlantic). Deep Sea Research Part II: Topical

710

Studies in Oceanography, 56(25), 2683-2704.

711

Menge B.A., Sutherland J.P. (1976) Species diversity gradients: synthesis of the roles of

712

predation, competition, and temporal heterogeneity. American Naturalist 110: 351-369.

713

Meredyk S.P. (2017) Physical characterization and benthic megafauna distribution and species

714

composition on Orphan Knoll and Orphan Seamount, NW Atlantic. M.Sc. thesis,

715

Memorial University of Newfoundland. https://research.library.mun.ca/.

716

Miles L.L. (2018) Cold-water coral distributions and surficial geology on the Flemish Cap,

717

Northwest Atlantic. M.Sc. thesis, Memorial University of Newfoundland.

718

https://research.library.mun.ca/.

719

Milligan R.J., Spence G., Roberts J.M., Bailey D.M. (2016) Fish communities associated with

720

cold-water corals vary with depth and substratum type. Deep Sea Research Part I:

721

Oceanographic Research Papers 114: 43-54.

722 723 724

Morato T., Watson R., Pitcher T.J., Pauly D. (2006) Fishing down the deep. Fish and Fisheries 7: 24-34. Mortensen P.B., Buhl-Mortensen L., Gordon D.C., Fader G.B.J., McKeown D.L., Fenton D.G.

725

(2005) Effects of fisheries on deepwater gorgonian corals in the Northeast Channel, Nova

726

Scotia. American Fisheries Society Symposium, Vol. 41, pp. 369.

727

Pe-Piper G., Meredyk S., Zhang Y., Piper D.J., Edinger E. (2013) Petrology and tectonic

728

significance of seamounts within transitional crust east of Orphan Knoll, offshore eastern

729

Canada. Geo-Marine Letters 33: 433-447.

730

Quattrini A.M., Demopoulos A.W., Singer R., Roa-Varon A., Chaytor, J.D. (2017) Demersal

731

fish assemblages on seamounts and other rugged features in the northeastern

732

Caribbean. Deep Sea Research Part I: Oceanographic Research Papers 123: 90-104.

733

Ramirez-Llodra E., Tyler P.A., Baker M.C., Bergstad O.A., Clark M.R., Escobar E., Levin L.A.,

734

Menot L., Rowden A.A., Smith C.R., Van Dover C.L. (2011) Man and the last great

735

wilderness: human impact on the deep sea. PLoS One 6: p.e22588. 34

736

Ross S.W., Quattrini A.M. (2007) The fish fauna associated with deep coral banks off the

737

southeastern United States. Deep Sea Research Part I: Oceanographic Research Papers

738

54: 975-1007.

739

Ross S.W., Rhode M., Quattrini A.M. (2015) Demersal fish distribution and habitat use within

740

and near Baltimore and Norfolk Canyons, US middle Atlantic slope. Deep Sea Research

741

Part I: Oceanographic Research Papers 103: 137-154.

742

Sherwood O.A., Edinger E.N. (2009). Ages and growth rates of some deep-sea gorgonian and

743

antipatharian corals of Newfoundland and Labrador. Canadian Journal of Fisheries and

744

Aquatic Sciences 66(1): 142-152.

745

Söffker M., Sloman K.A., Hall-Spencer J.M. (2011) In situ observations of fish associated with

746

coral reefs off Ireland. Deep Sea Research Part I: Oceanographic Research Papers 58(8):

747

818-825.

748 749

Stein M. (2007) Oceanography of the Flemish Cap and Adjacent Waters. Journal of Northwest Atlantic Fisheries Science 37: 135–146.

750

Stoner A.W., Ryer C.H., Parker S.J., Auster P.J., Wakefield W.W. (2008) Evaluating the role of

751

fish behavior in surveys conducted with underwater vehicles. Canadian Journal of

752

Fisheries and Aquatic Sciences 65: 1230-1243.

753 754 755

Sutton T.T. (2013) Vertical ecology of the pelagic ocean: classical patterns and new perspectives. Journal of Fish Biology 83: 1508-1527. Swihart R.K., Gehring T.M., Kolozsvary M.B., Nupp T.E. (2003) Responses of

756

‘resistant’vertebrates to habitat loss and fragmentation: the importance of niche breadth

757

and range boundaries. Diversity and Distributions 9: 1-18.

758

Teagle H., Hawkins S.J., Moore P.J., Smale D.A. (2017) The role of kelp species as biogenic

759

habitat formers in coastal marine ecosystems. Journal of Experimental Marine Biology

760

and Ecology 492: 81-98.

761

Tews J., Brose U., Grimm V., Tielbörger K., Wichmann M.C., Schwager M., Jeltsch F. (2004)

762

Animal species diversity driven by habitat heterogeneity/diversity: the importance of

763

keystone structures. Journal of Biogeography 31: 79-92. 35

764

Tissot B.N., Yoklavich M.M., Love M.S., York K., Amend M. (2006) Benthic invertebrates that

765

form habitat on deep banks off southern California, with special reference to deep-sea

766

coral. Fishery Bulletin 104(2): 167-181.

767 768 769

Trenkel V.M., Lorance P., Mahévas S. (2004) Do visual transects provide true population density estimates for deepwater fish?.ICES Journal of Marine Science 61: 1050-1056. Vazquez D.M., Belleggia M., Schejter L., Mabragaña E. (2018) Avoiding being dragged away:

770

finding egg cases of Schroederichthys bivius (Chondrichthyes: Scyliorhinidae) associated

771

with benthic invertebrates. Journal of Fish Biology 92: 248-253.

772 773

VideoLan Organization. (2016) VLC media player (Version 2.2.6). http://www.videolan.org/vlc/

774

Wedding L.M., Reiter S.M., Smith C.R., Gjerde K.M., Kittinger J.N., Friedlander A.M., Gaines

775

S.D., Clark M.R., Thurnherr A.M., Hardy S.M., Crowder L.B. (2015) Managing mining

776

of the deep seabed. Science 349(6244): 144-145.

777 778 779

Wentworth C.K. (1922) A scale of grade and class terms for clastic sediments. Journal of Geology 30: 377-392. Winger P.D., Eayrs S., Glass C.W. (2010) Fish behavior near bottom trawls. In: Pingguo H.,

780

Winger P. (eds.) Behavior of marine fishes: capture processes and conservation

781

challenges Wiley-Blackwell, Ames, Iowa, pp. 65-103.

782

Williams, A., Althaus F., Pogonoski J., Osterhage D., Gomon M., Graham K., Appleyard S.A.,

783

Gledhill D., Bray D., McMillan P., Green M. (2018) Composition, diversity and

784

biogeographic affinities of the deep-sea (200–3000 m) fish assemblage in the Great

785

Australian Bight, Australia. Deep Sea Research Part II: Topical Studies in

786

Oceanography 157: 92-105.

787 788 789

WWF (2018) World Wildlife Fund http://www.wwf.ca/?27841/Oil-and-gas-dont-mix-withconservation-WWF-Canada-says. Accessed on 22 June 2018. Zedel L., Fowler W.A. (2009). Comparison of boundary layer current profiles in locations with

790

and without corals in Haddock Channel, southwest Grand Banks. In: The ecology of

791

deep-sea corals of Newfoundland and Labrador waters: biogeography, life history, 36

792

biogeochemistry, and relation to fishes. Canadian Techical Report of Fisheries and

793

Aquatic Sciences 2830: 97-104.

794 795 796 797 798 799

Table 1 Summary of dive details for 2010 ROV cruise, indicating depth range (m), distance sampled (km), and duration (h) for each dive. Depth range (m)

Time on bottom (h)

Distance (km)

Dive

Date

Region

R1335

12 July

875 – 1840

13.78

10.0

South Flemish Cap

R1336

13 July

2224 – 2900

11.1

7.7

South Flemish Cap

R1337

14 July

1020 – 2195

16.68

15.6

East Flemish Cap

R1339

17-18 July

1363 – 2463

10.05

10.2

Northeast Flemish Cap

R1340

19-20 July

1870 – 3003

14.8

11.9

Orphan Seamount

800 801 802 803 804 805 806 807

Table 2 Number of fish observations in each physical and biological habitat type, separated by depth groupings. Values in parentheses represent total survey distance in meters of each habitat type. Habitat type

<1500 m

1500 – 2500 m

>2500 m

Total

2184 (9957)

1550 (16785)

219 (4549)

3953 (31291)

448 (2487)

551 (3975)

53 (668)

1052 (7130)

251 (1253)

321 (3816)

28 (441)

600 (5510)

54 (248)

211 (2160)

25 (1173)

290 (3581)

296 (1259)

612 (2676)

135 (3855)

1043 (7790)

143 (660)

585 (4848)

164 (2901)

892 (8409)

17 (87)

135 (1675)

13 (321)

165 (2083)

Physical Fine-grain sediments, no or few boulders Course sediments, no or few boulders Course sediments, scattered boulders Boulder field Outcrop Biological No corals or sponges Sparse corals

37

Sparse sponges

1361 (5059)

1286 (10775)

174 (3879)

2821 (19713)

Sparse mixed

21 (129)

227 (2108)

46 (450)

294 (2687)

Dense corals

176 (748)

112 (790)

-

288 (1538)

Dense sponges

889 (5242)

461 (4573)

58 (2859)

1408 (12674)

Dense mixed

626 (3279)

439 (4643)

5 (276)

1070 (8198)

808 809 810

Table 3 Number of individuals of each fish taxon observed during the five analysed dives. Taxa Chimaeridae Scyliorhinidae Etmopteridae Rajidae

Synaphobranchidae

Notacanthidae

Halosauridae

Alepocephalidae Gonostomatidae Bathysauridae Myctophidae Chaunacidae Macrouridae

Moridae Lotidae

R1335 Hydrolagus affinis Apristurus spp. Centroscyllium fabricii Amblyraja spp. Amblyraja jenseni Bathyraja spp. Bathyraja spinicauda Rajella spp. Rajidae (unknown) Histiobranchus bathybius Simenchelys parasitica Synaphobranchus kaupi Synaphobranchidae (unknown) Lipogenys gillii Polyacanthonotus spp. Notacanthus chemnitzii Notacanthidae (unknown) Aldrovandia spp. Halosauropsis macrochir Halosauridae (unknown) Alepocephalus spp. Gonostomatidae Bathysaurus ferox Myctophidae (unknown) Chaunax spp.

R1336 2

R1337 1

17 3 1

R1339 2 2 9 2

R1340 4

1

1 1

1 1 1

2 1 1

9

1 1

334 9

1

57

26

1

1 1 9 3

1 4

24 2

1

1 1

4 49

16 1 1 14

2

2

3 9 5 11

13 18 1

3

8 1 2 25

1 54

5

Coelorinchus caelorhincus Coryphaenoides armatus Coryphaenoides carapinus Coryphaenoides rupestris Macrourus berglax Nezumia bairdi Macrourid sp. 1 Macrourid sp. 2 Macrourid sp. 3 Macrourid sp. 4 Macrourid sp. 5 Macrourid sp. 6 Macrourid sp. 7 Macrouridae (unknown)

161 402 180 93 21 10 2 3 1

Antimora rostrata Lepidion eques Gaidropsarus spp.

38

7 146

15 59 1 3

1

4

1

2 313

62

152

1 43 843 925 832 123 42 74 5 6 3 7 2 1159

454 5 9

200 1 38

53 1 2

993 7 63

2 400 517 514 28 3 3

1

456

176

242

44

13

1

5

1 4 13 6 136 2 3

Total 10 2 26 6 1 1 1 6 12 1 3 427 14 2 37 16 3 18 68 1 26 2 12 93 6

30 123 4

2 4 1

Ophidiidae Oreosomatidae Psychrolutidae Pleuronectidae Unknown

Ophidiidae (unknown) Neocyttus helgae Cottunculus sp. Cottunculus thompsonii Reinhardtius hippoglossoides

2

2

2

8

14 8 1 3 52

1

1 1 1 1 27 867 37 6938

8 1 2 25

Fish unknown (sp. 1) Fish unknown (sp. 2) Fish unknown (sp. 3) Fish unknown (sp. 4) Fish unknown (Anguilliform) Fish unknown (mesopelagic) Fish (unknown)

1 27

1 1 13 628 17 2701

Grand total

39

3 3 6 538

4 160 7 2584

2 51 1 619

1 5 25 6 496

811 812

Table 4 Relative abundances of fish taxa observed on each physical and biological habitat type, standardized based on the total number of fish observed on each habitat ((Number of Sp.A on Habitat X/Total number of fish observed on Habitat X) * 100). Biological habitats

Physical habitats

Taxa

FGS

CS

CSB

BF

O

Absent

SS

SC

SM

DS

DC

DM

Chimaeridae Scyliorhinidae Etmopteridae

Hydrolagus affinis Apristurus spp. Centroscyllium fabricii

0.13 0.00 0.48

0.10 0.10 0.38

0.33 0.00 0.33

0.34 0.34 0.00

0.10 0.00 0.10

0.22 0.00 0.22

0.07 0.00 0.46

0.61 0.00 0.00

0.34 0.00 0.00

0.21 0.07 0.43

0.00 0.35 0.35

0.09 0.00 0.37

Rajidae

Amblyraja spp. Amblyraja jenseni Bathyraja spp. Bathyraja spinicauda Rajella spp. Rajiidae

0.13 0.00 0.00 0.03 0.13 0.20

0.10 0.00 0.10 0.00 0.00 0.19

0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.34 0.69

0.00 0.10 0.00 0.00 0.00 0.00

0.11 0.00 0.00 0.00 0.45 0.00

0.14 0.04 0.04 0.00 0.00 0.21

0.00 0.00 0.00 0.00 0.00 1.21

0.00 0.00 0.00 0.00 0.00 0.00

0.07 0.00 0.00 0.00 0.07 0.21

0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.09 0.09 0.09

Synaphobranchidae

Histiobranchus bathybius Simenchelys parasitica Synaphobranchus kaupi Synaphobranchidae

0.03 0.03 8.17 0.20

0.00 0.00 3.52 0.19

0.00 0.00 2.50 0.17

0.00 0.00 1.03 0.34

0.00 0.19 4.70 0.19

0.11 0.11 9.75 0.11

0.00 0.00 8.44 0.32

0.00 0.00 3.03 0.00

0.00 0.00 0.68 0.34

0.00 0.14 4.40 0.14

0.00 0.00 0.69 0.00

0.00 0.00 2.90 0.09

Notacanthidae

Lipogenys gillii Polyacanthonotus spp. Notacanthus chemnitzii Notacanthidae

0.03 0.68 0.13 0.03

0.00 0.10 0.48 0.10

0.00 0.50 0.17 0.17

0.00 0.00 0.00 0.00

0.10 0.58 0.48 0.00

0.11 0.78 0.11 0.00

0.00 0.78 0.14 0.04

0.00 1.82 0.61 0.00

0.00 1.02 0.00 0.00

0.00 0.00 0.43 0.07

0.00 0.35 0.69 0.00

0.09 0.09 0.19 0.09

Halosauridae

Alepocephalidae

Aldrovandia spp. Halosauropsis macrochir Halosauridae Alepocephalus spp.

0.30 1.24 0.03 0.33

0.10 0.67 0.00 0.10

0.50 0.50 0.00 0.00

0.34 0.00 0.00 0.69

0.10 0.86 0.00 0.96

0.45 3.48 0.00 1.01

0.21 0.82 0.00 0.50

1.21 2.42 0.00 0.00

1.36 2.72 0.00 0.34

0.07 0.07 0.00 0.00

0.00 0.00 0.35 0.00

0.09 0.09 0.00 0.19

Gonostomatidae Bathysauridae Myctophidae Chaunacidae

Gonostomatidae Bathysaurus ferox Myctophidae Chaunax spp.

0.05 0.28 1.77 0.00

0.00 0.00 1.33 0.10

0.00 0.00 1.00 0.50

0.00 0.34 0.00 0.00

0.00 0.00 0.29 0.19

0.00 0.45 0.45 0.22

0.04 0.25 0.57 0.11

0.00 0.00 2.42 0.00

0.00 0.34 0.00 0.34

0.00 0.00 2.70 0.00

0.00 0.00 1.39 0.00

0.09 0.00 2.52 0.00

Macrouridae

Coelorinchus caelorhincus Coryphaenoides armatus Coryphaenoides carapinus Coryphaenoides rupestris Macrourus berglax Nezumia bairdi

0.03 0.53 10.88 10.37 8.10 2.28

0.00 0.57 10.08 25.86 11.12 1.33

0.00 0.83 16.00 20.50 14.17 1.50

0.00 1.72 29.31 6.21 13.10 0.00

0.00 0.58 12.08 9.78 26.27 0.96

0.00 1.35 21.19 4.26 1.23 0.67

0.00 0.64 14.36 12.05 5.92 1.63

0.00 1.21 12.12 13.94 17.58 0.61

0.00 1.02 32.65 5.10 7.14 0.00

0.00 0.36 4.76 15.06 16.97 3.98

0.00 0.00 5.21 24.65 35.42 0.69

0.09 0.28 4.77 21.12 24.77 1.12

40

Moridae

Macrourid sp. 1 Macrourid sp. 2 Macrourid sp. 3 Macrourid sp. 4 Macrourid sp. 5 Macrourid sp. 6 Macrourid sp. 7 Macrouridae Antimora rostrata Lepidion eques

0.78 1.11 0.08 0.10 0.08 0.15 0.00 14.39 15.08 0.00

0.29 0.76 0.00 0.10 0.00 0.00 0.19 14.07 18.92 0.00

0.17 1.00 0.33 0.17 0.00 0.00 0.00 15.33 15.67 0.67

0.00 0.69 0.00 0.00 0.00 0.34 0.00 24.83 13.79 0.00

0.67 1.34 0.00 0.00 0.00 0.00 0.00 26.65 6.14 0.29

1.57 3.81 0.11 0.45 0.11 0.00 0.00 21.97 12.56 0.00

0.71 1.17 0.11 0.04 0.07 0.11 0.00 19.78 12.41 0.04

1.21 1.21 0.00 0.00 0.00 0.00 0.00 18.79 12.73 0.00

1.02 1.36 0.00 0.34 0.00 0.00 0.00 25.85 13.95 0.00

0.21 0.00 0.00 0.00 0.00 0.14 0.07 11.08 16.69 0.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 13.89 10.07 0.00

0.00 0.09 0.09 0.00 0.00 0.19 0.09 9.53 19.16 0.56

Lotidae Ophidiidae Oreosomatidae Psychrolutidae

Gaidropsarus spp. Ophidiidae Neocyttus helgae Cottunculus spp. Cottunculus thompsonii

0.73 0.08 0.00 0.03 0.05

0.86 0.10 0.00 0.00 0.10

1.83 0.00 0.33 0.00 0.00

1.72 0.34 0.00 0.00 0.00

0.86 0.86 0.58 0.00 0.00

0.56 0.11 0.00 0.00 0.00

0.53 0.25 0.00 0.00 0.07

0.61 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00

2.13 0.36 0.00 0.07 0.07

0.69 0.35 1.74 0.00 0.00

0.93 0.00 0.28 0.00 0.00

Pleuronectidae

Reinhardtius hippoglossoides

0.76

1.43

0.17

0.34

0.48

0.67

0.74

1.21

0.68

0.71

0.00

1.03

0.03 0.00 0.03 0.03

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.34 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.00 0.04 0.00

0.61 0.00 0.00 0.61

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00

0.00 0.09 0.00 0.00

0.56

0.29

0.33

0.00

0.00

0.56

0.50

0.00

0.34

0.21

0.00

0.37

18.69 0.71

6.27 0.10

4.00 0.33

2.76 0.00

2.88 0.67

9.42 1.23

15.03 0.64

3.64 0.61

2.38 0.68

17.76 0.28

3.13 0.00

8.13 0.09

Unknown

Fish unknown (sp. 1) Fish unknown (sp. 2) Fish unknown (sp. 3) Fish unknown (sp. 4) Fish unknown (Anguilliform) Fish unknown (mesopelagic) Fish (unknown)

813 814 815 816 817

41

818 819

Table 5 Reaction to the ROV, behaviour, and relative position of individuals for each fish taxa observed during 2010 surveys off Newfoundland, Canada. Taxa

Behaviour

Reaction to ROV

Hiding

Hovering

Restring

Swimming

No reaction

Avoidance

Position

Attraction

On bottom

Off bottom

High off bottom

Chimaeridae

Hydrolagus affinis

-

30.0

-

70.0

50.0

40.0

10.0

30.0

30.0

40.0

Scyliorhinidae

Apristurus spp.

-

-

-

100.0

50.0

50.0

-

-

-

100.0

Etmopteridae

Centroscyllium fabricii

-

-

-

100.0

65.4

7.7

26.9

50.0

34.6

15.4

Rajidae

Amblyraja spp. Amblyraja jenseni Bathyraja spp. Bathyraja spinicauda Rajella spp. Rajiidae

-

-

66.7 100.0 41.7

33.3 100.0 100.0 100.0 58.3

50.0 100.0 100.0 66.7 50.0

50.0 100.0 33.3 50.0

-

100.0 100.0 100.0 100.0 66.7 100.0

33.3 -

-

Synaphobranchidae

Histiobranchus bathybius Simenchelys parasitica Synaphobranchus kaupi Synaphobranchidae

-

0.7 21.4

-

100.0 100.0 99.3 78.6

100.0 100.0 88.1 92.9

9.1 7.1

2.8 -

37.2 28.6

66.7 31.4 57.1

100.0 33.3 31.4 14.3

Notacanthidae

Lipogenys gillii Polyacanthonotus spp. Notacanthus chemnitzii Notacanthidae

-

50.0 51.4 25.0 -

-

50.0 48.6 75.0 100.0

50.0 94.6 93.8 100.0

50.0 5.4 -

6.3 -

62.2 25.0 100.0

29.7 62.5 -

100.0 8.1 12.5 -

Halosauridae

Aldrovandia spp. Halosauropsis macrochir Halosauridae

-

38.9 89.7 100.0

33.3 1.5 -

27.8 8.8 -

66.7 94.1 100.0

33.3 4.4 -

1.5 -

44.4 76.5 -

44.4 16.2 100.0

11.1 7.4 -

Alepocephalidae

Alepocephalus spp.

-

61.5

-

38.5

92.3

7.7

-

25.0

31.3

43.8

Gonostomatidae

Gonostomatidae

-

-

-

100.0

-

-

100.0

-

50.0

50.0

Bathysauridae

Bathysaurus ferox

-

-

100.0

-

100.0

-

-

100.0

-

-

Myctophidae

Myctophidae

-

6.5

-

93.5

33.3

57.0

9.7

10.8

39.8

49.5

Chaunacidae

Chaunax spp.

-

-

100.00

-

50.0

50.0

-

100.0

-

-

Macrouridae

Coelorinchus

-

100.0

-

-

100.0

-

-

100.0

-

-

42

caelorhincus Coryphaenoides armatus Coryphaenoides carapinus Coryphaenoides rupestris Macrourus berglax Nezumia bairdi Macrourid sp. 1 Macrourid sp. 2 Macrourid sp. 3 Macrourid sp. 4 Macrourid sp. 5 Macrourid sp. 6 Macrourid sp. 7 Macrouridae

0.6

58.1 89.9

0.2

41.9 9.3

86.0 86.8

11.6 12.7

2.3 0.5

41.9 79.8

44.2 10.6

14.0 9.6

0.1 0.1 0.8 0.2

77.0 83.3 85.4 85.7 75.7 100.0 83.3 100.0 100.0 50.0 86.1

4.3 0.8 0.3

22.9 12.2 13.0 14.3 24.3 16.7 50.0 13.5

75.4 96.8 96.7 88.1 71.6 100.0 83.3 100.0 100.0 100.0 90.6

23.9 2.8 3.3 11.9 28.4 16.7 9.1

0.8 0.5 0.3

60.2 94.7 96.7 83.3 59.5 20.0 66.7 100.0 100.0 100.0 66.7

23.8 3.4 3.3 9.5 29.7 40.0 33.3 13.5

16.0 1.9 7.1 10.8 40.0 19.8

Moridae

Antimora rostrata Lepidion eques

-

8.8 71.4

0.1 -

91.1 28.6

83.1 100.0

11.2 -

5.7 -

65.9 100.0

27.8 -

6.3 -

Lotidae

Gaidropsarus spp.

14.3

14.3

49.2

22.2

95.2

1.6

3.2

96.8

3.2

-

Ophidiidae

Ophidiidae

-

57.1

-

42.9

100.0

-

-

78.6

21.4

-

Oreosomatidae

Neocyttus helgae

-

87.5

-

12.5

87.5

12.5

-

25.0

37.5

37.5

Psychrolutidae

Cottunculus sp. Cottunculus thompsonii

-

-

100.0 100.0

-

100.0 100.0

-

-

100.0 100.0

-

-

Pleuronectidae

Reinhardtius hippoglossoides

-

-

59.6

40.4

67.3

30.8

1.9

98.1

-

1.9

Unknown

Fish unknown (sp. 1) Fish unknown (sp. 2) Fish unknown (sp. 3) Fish unknown (sp. 4) Fish unknown (Anguilliform) Fish unknown (mesopelagic) Fish (unknown)

-

100.0 18.5

100.0 3.7

100.0

11.1

100.0 11.1

100.0 100.0 44.4

100.0 29.6

100.0

77.8

100.0 100.0 100.0 77.8

25.9

-

1.8

0.2

98.0

27.2

64.4

8.4

18.5

44.5

37.0

-

50.0

2.6

47.4

86.8

10.5

2.6

35.1

40.5

27.0

100.0

820

43

• • • • •

Fish-habitat relationships are poorly understood in the deep sea Remotely-operated vehicles provide direct observations of fish assemblages Deep-sea fish assemblages differed significantly across depth groups Overall tests for habitat composition (physical & biogenic) were not significant Pairwise tests suggest habitat complexity most important for many species

Manuscript Number: DSR1_2019_57R1   Habitat associations and assemblage structure of demersal deep-sea fishes on the eastern Flemish Cap and Orphan Seamount

Declaration of interest: None