Previous Page Table of Contents Next Page

Poster papers (continue)


In situ observations of the scavenging fauna of the South Georgia slope

M.A. Collins1, I. Everson1, R. Patterson2, P.M. Bagley2, C. Yau3, M. Belchier1 and S. Hawkins1
1 British Antarctic Survey, NERC
High Cross, Madingley Road, Cambridge, CB3 OET, UK

2 Oceanlab, Zoology Department, Aberdeen University
Newburgh, Aberdeenshire, AB41 6AA

3 Department of Ecology & Biodiversity, The University of Hong Kong
Pokfulam Road, Hong Kong


The South Georgia region of the Scotia Sea is characterized by high biomass and productivity of both phytoplankton and zooplankton (Atkinson et al. 2001). This high surface productivity supports large populations of penguins, seals and fish, which are heavily reliant on Antarctic krill, which forms the basis of the pelagic food-web. This high productivity will also produce substantial deposition of material on the sea-floor on both the continental shelf, slope and the abyssal plain, which may stimulate a diverse scavenging community.

The fish fauna of the South Georgia shelf has been described in some detail (Gon and Heemstra 1990, Kock 1992), largely as a consequence of commercial interest in species such as mackerel icefish (Champsocephalus gunnari) and the marbled notothen (Notothenia rossii). But, the deeper fauna has been largely overlooked, and with the fishery for toothfish operating in deep-water it is important to establish the composition and abundance of the deep-sea fauna. However, investigating the demersal deep-sea fauna can be problematic, particularly as traditional commercial trawls do not extend much below 1000 m and alternate gears such as Agassiz and single warp trawls are considerably smaller sampling devices.

An alternative method of studying deep-sea fauna is through the use of baited cameras, which have been used to investigate the abundance and behaviour of scavengers in the Atlantic (Collins, Priede and Bagley 1999, Jones et al. 1998, Priede and Bagley2000) and Pacific Oceans (Priede, Bagley and Smith 1994) and the Mediterranean Sea (Jones et al. 2003). This method is highly selective exclusively attracting scavengers, but with knowledge of their behaviour, it does permit density estimates, either by using the first arrival time (Priede and Merrett 1996, 1998) or by estimating the area of the odour plume from which scavengers are attracted (Collins et al. 2002, Sainte-Marie and Hargrave 1987).

In 1997 the Government of South Georgia and the South Sandwich Islands (GSGSSI) funded a pilot study to use a baited camera system to investigate toothfish populations at South Georgia and subsequent studies have been undertaken in 2000 and 2003. The data from the 1997 and 2000 surveys have been used to describe the scavenging fauna at South Georgia (Yau et al. 2002), assess abundance of stone crabs (Collins et al. 2002), identify commensal relationships between fish and crabs (Yau, Collins and Everson 2000) and investigate the utility of using arrival time at bait to assess density of toothfish (Yau et al. 2001). Here we present the new data from 2003 and take the opportunity to review the data so far, focusing in particular on the distribution and behaviour of toothfish.


During three cruises at South Georgia (September 1997, F.V. Argos Galicia; January 2000, F.V. Argos Galicia; January 2003, F.P.V. Dorada) baited camera systems were deployed on the South Georgia slope from depths of 200–2500 m (Table 1, Figure 1). The baited camera systems were modifications of the Aberdeen University Deep Ocean Submersible (AUDOS) system (Bagley and Priede 1997, Priede and Bagley 2000) (Figure 2), which were designed to photograph and track scavenging fish and invertebrates on the sea floor.

Details of deployments with numbers of toothfish encounters and maximum numbers of crabs

Exp.Camera SystemDateAreaLatitudeLongitudeDepth (m)Temp. CToothfish EncountersMax. ParalomisCrabs Neolithodes
135 mm7-Sep-97SR53°19.1' S41°56.7' W10391.5130
235 mm8-Sep-97SR53°31.2' S40°20.5' W11491.52180
335 mm10-Sep-97SR53°46.0' S41°59.1' W10001.4341
435 mm11-Sep-97SR53°17.7' S42°12.0' W7471.89431
535 mm13-Sep-97SG53°35.2' S37°59.2' W10001.6127/40
635 mm17-Sep-97SG53°45.1' S35°59.8' W1100 4440
735 mm18-Sep-97SG54°14.9' S35°15.8' W775 033/30
835 mm19-Sep-97SG54°30.9' S35°06.1' W1487 1281
935 mm21-Sep-97SG55°25.4' S34°55.3' W625 6244
1035 mm22-Sep-97SG55°09.1' S36°21.2' W1143 4371
1135 mm23-Sep-97SG55°02.6' S36°59.6' W12751.35203
1235 mm24-Sep-97SG54°52.7' S37°58.7' W1178 1261
1335 mm27-Sep-97SG53°44.4' S39°22.6' W872 321/20
14DV17-Jan-00SR53°16.9' S42°22.1' W7192.27400
15DV17-Jan-00SR53°26.2' S41°35.6' W10851.88220
16DV21-Jan-00SG53°35.5' S36°20.9' W10351.624741
17DV22-Jan-00SG54°05.5' S35°20.9' W11141.638170
18DV22-Jan-00SG54°05.0' S35°20.1' W12941.644271
19DV23-Jan-00SG54°37.3' S34°47.6' W7801.914391
20DV23-Jan-00SG54°36.8' S34°42.8' W1005 10221
21DV24-Jan-00SG54°38.1' S34°37.6' W12501.432122
22DV24-Jan-00SG54°38.1' S34°32.6' W1518 21151
23DV26-Jan-00SG54°50.7' S38°31.5' W11201.67240
24DV26-Jan-00SG54°50.9' S38°31.5' W13351.61630
25DV27-Jan-00SG54°23.7' S39°28.7' W9461.731080
26DV27-Jan-00SG54°23.9' S39°23.6' W12021.718301
27DV29-Jan-00SR53°36.1' S40°44.9' W12831.718210
28DV29-Jan-00SR53°36.2' S40°45.7' W1140 18201
29DV11-Jan-03SG53°48.2' S36°7.0' W10562.0560
30DV13-Jan-03SG53°23.2' S36°53.0' W16111.457270
31DV13-Jan-03SG53°16.9' S36°54.4' W23350.9000
32DV15-Jan-03SG53°48.0' S39°28.4' W10182.057251
33DV18-Jan-03SR53°39.9' S40°37.7' W4712.4590
34DV20-Jan-03SR53°32.9' S41°7.8' W10052.255281
35DV25-Jan-03SG54°54.2' S38°9.6' W11601.92352
36DV26-Jan-03SG54°53.6' S38°24.7' W19841.2030
37DV27-Jan-03SG55°28.9' S36°13.3' W18961.3010
38DV27-Jan-03SG55°18.4' S36°17.1' W7692.253190
39DV29-Jan-03SG54°39.8' S35°35.2' W13561.559100
40DV30-Jan-03SG54°49.4' S34°26.0' W7902.153261
41DV31-Jan-03SG53°51.1' S35°40.4' W8902.054350

The basic system consists of an aluminium frame on which a camera system is mounted, a current meter (Sensortec), acoustic releases (Mors AR and RT) and a battery. Buoyancy was provided by glass spheres (Benthos Inc., each giving 24 kg positive buoyancy) attached to a 100 m mooring line. In 1997 a conventional, high capacity still camera (Ocean Instruments) was used, while in 2000 and 2003 a video camera (JVC Colour Video Camera, TK-C1380 in housing with controller) was used.

Map showing the locations of baited camera deployments at South Georgia in 1997 (red), 2000 (green) and 2003 (blue)


Diagrammatic view of the baited camera system used in 1997 (A) and photograph of the system used in 2000 and 2003 (B)


The camera systems descended by free-fall. It had 100 kg of ballast, which held the rig in position on the seafloor. The ballast, with a graduated cross and baits attached, remained on the seafloor and was connected to the AUDOS vehicle by a 2 m length of wire. The cross therefore rested 30–50 cm above the seafloor on top of the ballast; positive buoyancy of the mooring line held the AUDOS 2 m above the cross. Each deployment was baited with four squid (Illex argentinus) hung from the cross with sardines inserted in the mantle cavity and attached to the ballast (total 800 g).

Each experiment lasted between 6 and 10 hours. The conventional time-lapse camera system (1997) took photographs every minute. The camera was loaded with Ektachrome 200 ASA film with a capacity of approximately 750 full frames. Small strips of film were developed on board to test that the camera was working, but the bulk of the film was developed after the cruise.

The video camera systems (2000, 2003), recorded a total of one hour of video and were typically programmed to record 45 consecutive seconds in each 2½ minutes for the first two hours; 45 seconds in each five minutes for the next two hours and 45 seconds every 15 minutes during hours five and six. The camera recorded onto digital videotape and after recovery a copy was made on the SVHS. The camera viewed an area of sea-floor of approximately 4.9 m2. The current meter was programmed to record depth, temperature, current direction and current speed at one-minute intervals throughout the deployment.

On completion of the experiment the camera systems were released from the ballast by an acoustic signal from a Mors deck unit (TT301) using a transponder lowered into the water. The vehicles surfaced at a rate of approximately 0.8 ms-1 under their own buoyancy. A marker buoy was attached at the surface to the end of the mooring, which incorporated a VHF radio beacon (Novatech) together with a large pink flag. This aided in its location and recovery.


3.1 Scavenging fauna

The scavenging fauna was dominated by lithodid crabs (Paralomis formosa and P.spinosissima), toothfish, grenadiers and morids (Table 2, Figure 3). Lithodid crabs were present at all but the deepest deployment (2 235 m). Toothfish were not seen at the three deepest deployments and were also absent from Deployment 7 (1997, 775 m). Grenadiers (Macrourus sp.) were frequently attracted to the bait, but the species could not be identified with any certainty from the videos or still photographs. Three species (M. holotrachys, M. carinatus and M. whitsoni) were taken in trawls (up to 900 m) on the South Georgia slope during the 2003 survey. Two species of morid were also seen at the bait: Antimora rostrata, which is easily recognizable from the black colour and a second species that was either Lepidion ensiferus or Halargyreus johnsoni. Two species of skate were seen, Bathyraja meridionalis and Raja georgianus. See Yau et al. (2002) for full details of the scavenging fauna from the 1997 and 2000 surveys.

3.2 Toothfish abundance and distribution

The 1997 and 2000 cruises were standard South Georgia groundfish surveys. Camera deployments were made each evening and recovered in the morning but the depth choice was constrained by the requirements of the trawl survey. In 2003 the survey was designed to investigate deep-sea fauna in a series of down-slope transects, with the camera deployments extending to depths in excess of 2000 m. The number of toothfish photographed at the bait was low during the 1997 survey, but greater numbers were attracted to the baits in 2000. However, numbers were again low in 2003. The low number of encounters with toothfish in the 1997 survey was attributed to the effects of the powerful flash associated with the stills camera. During the 2003 survey, three deployments were made at depths greater than 1800 m and toothfish were not seen in these deployments (Table 1). These deep deployments corresponded to the lowest temperatures (< 1°C). Even excluding these, there were less toothfish encounters in 2003. The maximum encounter rate for toothfish was at depths of 1200–1400 m (Figure 4A), corresponding to temperatures of around 1.5°C (Figure 4B).

Scavenging species identified from baited camera studies on the South Georgia slope

 SpeciesDepth RangeEncounters/ Max
FishDissostichus eleginoides471–16114027847
 Macrourus sp.764–2 328255555
 Antimora rostrata466–232810172122
 Lepidion ensiferus/ Halargyreus johnsoni769–1896010129
 Raja georgianna719–15190111
 Raja sp A.21018–18960126
InvertebratesParalomis formosa/ Paralomis spinossissima471–18964410835
 Neolithodes diomedea764–1519422
 Thymops birsteini471–1896373

For fish species numbers refer to the number of encounters and for invertebrates to the maximum number at any time.

1 All morids from 1997 and 2000 recorded as Antimora rostrata, but probably included Lepidion ensiferus and, or, Halargyreus johnsoni.

2 Recorded as Bathyraja meridionalis in Yau et al. (2002).

Scavengers from the South Georgia slope

A. Video still (Jan 2000) of aggregation of crabs (P.formosa) - B. Photograph of Parolomis formosa and P. spinosissima - C. Toothfish (Dissostichus eleginoides) - D. Video still of skate, Raja spA.


Relationship between toothfish encounters and deployment depth (upper panel) and temperature (lower panel) from the baited video camera deployments in 2000 and 2003


3.3 Toothfish behaviour

Swimming: Toothfish generally approached the bait from down-current, swimming close to the sea-floor with gentle sculling motions of the pectoral fins (i.e. labriform swimming). Occasionally toothfish were startled by knocking each other, touching crab spines or the cross and swam away rapidly. On one occasion a toothfish was startled outside the field of view, and swam rapidly through the field of view, skirting around the ballast. The fish was approximately 80 cm long and swam with a velocity of 3.1 body lengths sec-1.

Bait attendance: Toothfish numbers did not accumulate at the bait, fish usually arrived individually, investigated the area around the bait and if they could not easily obtain the squid bait would depart. Fish occasionally rested on the sea-floor close to the bait, but usually only for a minute or two.

Colour change: In the video camera deployments the lights were programmed to turn on before the camera started. There were many incidents when a toothfish was in the field of view when the video turned on and the fish were frequently very pale. The toothfish were subsequently seen to change colour, changing from light, back to dark, in less than 30 secs. On other occasions a cloud of sediment was visible when the camera came on, indicating that a fish (probably toothfish) had been startled by the lights, and swam away rapidly.

Taking the bait: Toothfish were videoed taking the bait on numerous occasions, usually the fish would grab the squid, pull it from the cross and swallow it whole. On one occasion a toothfish rotated three times in order to remove the bait from the cross. Toothfish also tried to eat the current ribbons that were attached to the ends of the cross and were probably covered in bait.

Interactions with crabs: When large numbers of crabs (Paralomis sp.) were present at the bait toothfish did not stay around long. There was no evidence that toothfish prey on crabs. Toothfish frequently came into contact with the spines of the crabs and rapidly swam away.

3.4 Crab abundance

The principal crab species attracted to the bait was Paralomis formosa, although Paralomis spinnossissima was also seen during shallower deployments. The much larger Neolithodes diomedea was seen occasionally. However, when large numbers of crabs were present at the bait, it was difficult to distinguish between the Paralomis species. Data from the higher resolution 35 mm camera system in 1997 suggested that P.spinossissima was limited to depths of less than 600 m.

Crab numbers increased rapidly at the bait, and crabs remained in the area while bait was accessible. The crabs formed large clumps around the squid bait and were occasionally seen fighting for parts of the bit.

From the current meter data, the size of the odour plume (assumed to be an elipse) can be estimated at a given time, and with knowledge of crab walking speed, and numbers at the bait, an estimate of density can be obtained. Density estimates were made from the 2000 survey (see Collins et al. 2002) but the analysis has not yet been undertaken on the 2003 data.


The South Georgia slope supports a diverse community of scavengers that respond rapidly to the arrival of bait on the sea-floor. The main scavenging species are crabs of the genus Paralomis and Patagonian toothfish (Dissostichus eleginoides). Smaller scavengers such as amphipods could not be quantified. The data suggest that the abundance of scavengers is lower at depths greater than 1800 m, which may be a consequence of reduced food availability at this depth or the physiological intolerance of the scavengers to greater depths and lower temperatures.

The data suggest that D. eleginoides are absent from depths greater than 1800 m and, or, temperatures of less than 1.3°C. Unlike Dissostichus mawsoni, D. eleginoides does not have anti-freeze glycopeptides in its blood (Eastman 1990) and it is therefore likely that low temperatures limit its distribution.

The 35 mm camera system used in the 1997 survey produced high resolution images from which the fauna could be better identified, but it was thought that the powerful flash may have discouraged toothfish from attending the bait. On the other hand, the video camera system, with lower light levels, gave poorer resolution for species identification but, for the common species provided valuable behavioural data. Even with the comparatively low-light levels on the video camera, there was still evidence that toothfish reacted to the lights, either by rapidly leaving the illuminated area (evidence by clouds of sediment when the camera started) or by rapid colour change. The ability to change colour appears unusual for a deep-sea fish species, where light is usually low or absent, but the toothfish clearly attempt to camouflage themselves against the background, indicating that the fish have the ability to detect and respond to light of varying intensity. Juvenile, and occasionally adult, toothfish inhabit shallow water, where the ability to detect and respond to light will be advantageous, so it appears likely that adults have retained this ability.

The labriform swimming of the toothfish appears designed to conserve energy, typically sculling close to the sea-floor with gentle beats of pectoral fins. However, when startled they are capable of more rapid, sub-carangiform swimming and the speed of one fish of 3.1 BL sec-1 is quicker than cod are reported to swim at similar temperatures (Videler and Wardle 1991). In general, the toothfish did not remain long at the bait, which may be a consequence of the lights, but may also be indicative of plentiful food. Priede et al. (1991) and Priede and Merrett (1998) have demonstrated that the staying time of grenadiers at bait is related to the overlying productivity and thus, probably, to food availability on the sea-floor.

Providing that the bait was accessible, crab numbers accumulated at the bait over time, which permited estimates of local density to be made from the arrival times (Collins et al. 2002). However, such an approach is not possible for toothfish, which stayed only briefly at the bait. Local crab density was estimated to be as high as 25000 km-2 with a mean density of 8318 km-2 (depth range: 719–1 518 m).

The relationship between Paralomis formosa and the liparid Careproctus sp. was first described (Yau et al. 2000) from the 1997 survey with the high resolution camera, but subsequent work with the video system has shown that the fish are able to move between crab hosts and aggregations at bait may provide opportunities for fish to switch host and, perhaps, feed on scavenging amphipods.


Atkinson, A., M.J. Whitehouse, J. Priddle, G.C. Cripps, P. Ward & M.A. Brandon 2001. South Georgia, Antarctica: a productive, cold water, pelagic ecosystem. Marine Ecology-Progress Series, 216, 279–308.

Bagley, P.M. & I.G. Priede 1997. An autonomous free-fall acoustic tracking system for investigation of fish behaviour at abyssal depths. Aquatic Living Resources, 10, 67–74.

Collins, M.A., I.G. Priede & P.M. Bagley 1999. In situ comparison of activity in two deep-sea scavenging fishes occupying different depth zones. Proceedings of the Royal Society B, 266, 2011–2016.

Collins, M.A., C. Yau, F. Guilfoyle, P. M. Bagley, I. Everson, I.G. Priede & D. Agnew 2002.Assessment of stone crab (Lithodidae) density on the South Georgia slope using baited video cameras. ICES Journal of Marine Science, 59, 370–379.

Eastman, J.T. 1990. The biology and physiological ecology of Nototheniod fishes p.34–51. In Fishes of the Southern Ocean (ed. O. Gon and P.C. Heemstra), JLB Smith Institute of Ichthyology.

Gon, O. & P.C. Heemstra 1990. Fishes of the Southern Ocean. Grahamstown: J.L.B.Smith Institute of Ichthyology.

Jones, E.G., M.A. Collins, P.M. Bagley, S. Addison & I.G. Priede 1998. The fate of cetacean carcasses in the deep-sea: observations on consumption rates and succession of scavenging species in the abyssal north-east Atlantic. Proceedings of the Royal Society London B, 265, 1119–1127.

Jones, E.G., A. Tselepides, P.M. Bagley, M.A. Collins & I.G. Priede 2003. Bathymetric distribution of some benthic and benthopelagic species attracted to baited cameras and traps in the Eastern Mediteranean. Marine Ecology Progress Series, 251, 75–86.

Kock, K.-H. 1992. Antarctic Fish and Fisheries. Cambridge: Cambridge University Press.

Priede, I.G. & P.M. Bagley 2000. In situ studies on deep-sea demersal fishes using autonomous unmanned lander platforms. Oceanography and marine biology: An annual review, 38, 357–392.

Priede, I.G., P.M. Bagley, J.D. Armstrong, K.L. Smith &N.R. Merrett 1991. Direct measurement of active dispersal of food-falls by deep-sea demersal fishes. Nature, 351,647–649.

Priede, I.G., P.M. Bagley, P.M. &K.L. Smith 1994. Seasonal change in activity of abyssal demersal scavenging grenadiers Coryphaenoides (Nematonurus) armatus in the eastern Pacific Ocean. Limnology and Oceanography, 39, 279–285.

Priede, I.G. & N.R. Merrett 1996. Estimation of abundance of abyssal demersal fishes; a comparison of data from trawls and baited cameras. Journal of Fish Biology, 49,207–216.

Priede, I.G. & N.R. Merrett 1998. The relationship between numbers of fish attracted to baited cameras and population density; Studies on demersal grenadiers Coryphaenoides (Nematonurus) armatus in the abyssal NE Atlantic Ocean. Fisheries Research, 36,153–157.

Sainte-Marie, B. & B.T. Hargrave 1987. Estimation of scavenger abundance and distance of attraction to bait. Marine Biology, 94, 431–443.

Videler, J.J. & C.S. Wardle 1991. Fish swimming stride by stride: speed limits and endurance. Reviews in Fish Biology and Fisheries, 1, 23–40.

Yau, C., M.A. Collins, P.M. Bagley, I. Everson, C. Nolan & I.G. Priede 2001. Estimating the abundance of Patagonian toothfish Dissostichus eleginoides using baited cameras: a preliminary study. Fisheries Research, 51, 403–412.

Yau, C., M.A. Collins, P.M. Bagley, I. Everson & I.G. Priede 2002. Scavenging by megabenthos and demersal fish on the South Georgia slope. Antarctic Science, 14,16–24.

Yau, C., M.A. Collins & I. Everson 2000. Commensalism between a liparid fish (Careproctus sp.) and stone crabs (Lithodidae) photographed in situ using a baited camera. Journal of the Marine Biological Association of the United Kingdom, 80, 379–380.

Towards an international census of marine life research program on seamounts

K.I. Stocks1 and G.W. Boehlert2
1 San Diego Supercomputer Center
University of California
9500 Gilman Drive
San Diego MC 0505, La Jolla, CA 92093-0505, USA

2 Hatfield Marine Science Center
Oregon State University
2030 SE Marine Science Dr. Newport , OR 97365-5296, USA


The Census of Marine Life (CoML) is a worldwide science initiative promoting research to assess and explain the diversity, distribution, and abundance of species throughout the world's oceans (<>; Decker and O'Dor 2003). As one of several activities, the CoML fosters international field programmes to facilitate research into under-explored marine ecosystems. Recognizing the growing scientific interest in seamounts, the Census of Marine Life hosted an international workshop on seamounts on 22–24 August 2003 at the Hatfield Marine Science Center in Newport, Oregon, USA. The goals of the workshop were to (a) evaluate the existing state of knowledge of seamounts, (b) determine the priorities for future seamount research, (c) outline the next steps required to address these research priorities and (d), evaluate the potential role of the CoML in fostering research progress through an international field programme.


The workshop concluded that seamounts represent important ecosystems for studies that have not, to date, received scientific attention consistent with their biological and ecological value (Stocks, Boehlert and Dower 2004). A comprehensive understanding of ocean biodiversity and biogeography will require directed study of seamounts to learn of their unique features. Specifically, it was noted that seamounts

Important policy and management decisions regarding seamounts will be made in the next five to seven years (e.g. UN General Assembly considerations of marine protected areas) and good science will be essential for guiding management and conservation efforts. Further, the workshop participants determined that a CoML field programme on seamounts could have a valuable role in stimulating and coordinating seamount research. Workshop presentations on past and planned seamount work highlighted that seamounts are an area of active research. However, important science questions, detailed below, will not be addressed by simply continuing with a “business as usual” approach. In particular, there is a compelling need for an international effort to promote and coordinate future field efforts and synthesize existing knowledge in order to extend our results beyond individual seamount ecosystems.

Seamount locations

Squares indicate seamounts that have been sampled biologically (from SeamountsOnline; Small black dots indicate predicted locations of ~14 000 seamounts (from Kitchingman and Lai 2004).



To provide a scientific framework for the envisioned CoML program on seamounts, the following priority science question was articulated: “What roles do seamounts play in the biogeography, biodiversity, productivity, and evolution of marine organisms, and what is their effect on the global, oceanic ecosystem?”. This primary theme was further sub-divided into three specific research themes.

Theme 1

Given the large number of seamounts globally, can we categorize seamount community structures and, or, develop proxy variables in order to (a) use our knowledge from a limited number of well-known seamounts to make predictions about unknown ones, (b) efficiently guide future research programmes and (c), understand the key processes regulating the structure and maintenance of seamount communities?

One topic considered was whether some minimal set of physical factors might be formulated in order to provide a biologically meaningful description and categorization scheme for seamounts. Although by no means exhaustive, the following list of factors were identified as being important to consider in any such scheme:

Theme 2

How do seamount communities, both within and between seamounts, differ in ecological structure and function? This theme explicitly recognized that there can be substantial patchiness within a given seamount as well as between seamounts, and that variability at both scales is important for understanding seamount ecosystems. Particular questions of interest include: How do the physical characteristics of a given seamount influence the composition of communities that occupy its various habitats? What are the roles of biological interactions (e.g. trophic structure and food web function) both within seamount communities and with surrounding ocean communities? How vulnerable are seamount ecosystems to disturbance, and how might the structure of these ecosystems change in response to natural (e.g. seasonal variability, inter-annual cycles, climate change) and anthropogenic (e.g. overfishing) influences? What roles do larval dispersal and recruitment dynamics play in the long-term persistence of seamount populations? How do the surrounding deep-sea communities interact with seamounts? On what scales do seamounts influence the biological and physical structure of adjacent oceanic habitats (i.e. what is the spatial and ecological “footprint” of a seamount or seamount chain in the surrounding ocean)?

Theme 3

On a broader scale, what roles do seamounts play in global oceanic ecosystems with respect to (a) biogeography, (b) biodiversity, (c) evolution and (d), productivity? This theme will involve the synthesis of seamount studies from around the world. Specifically, workshop participants proposed investigating issues such as whether seamounts act as centres of speciation or as refugia for relict populations, to what extent do seamounts serve as stepping-stones for trans-oceanic dispersal and whether they represent regional hotspots of biological production, which may be important e.g. for migratory species.


To address the priorities stated above, workshop participants identified several key planning and research activities. The primary focus will be on developing an international scientific programme that concurrently catalyzes and coordinates field sampling while continuing databasing and data analysis efforts.

Activity 1: Promoting future field sampling

Given that fewer than 150 of the tens of thousands of seamounts have been explored in any detail (Figure 1), new field research is obviously critical to improving our understanding of seamount biogeography. Therefore, supporting and coordinating existing field efforts and developing new collaborative projects have been identified as high priorities for a future CoML seamount program. While funding is always a limiting factor for expedition work, the growing concerns, both within many countries and internationally, over human impacts on seamounts may open new sources of funding. In addition to leveraging funds for new initiatives, future seamount research should also be linked to existing sampling programmes and, where appropriate, national agendas (e.g. several nations, including Australia, New Zealand and Canada, already have programmes underway to protect certain seamounts within their Exclusive Economic Zones).

Workshop participants recommended that a planning stage for this activity address:

Activity 2: Networking and coordination

The scope of the science recommended above will require substantial coordination within the international scientific community. Geographically, it is desirable (indeed essential) that many countries participate, given how widely distributed seamounts are and how many lie in international waters. Scientifically, expertise from a wide range of fields (e.g. genetics, population biology, fisheries biology, physical oceanography, geology, taxonomy, ecosystem ecology, etc) will be needed. Further, a variety of existing programmes (e.g. MAR-ECO <>; OASIS <>; NOAA Ocean Explorations <>) are undertaking seamount research. These programmes typically have objectives that are consistent with the science objectives identified by this workshop, and can provide established networks of experts. Facilitating collaboration among these projects, while not competing with them, will be critical.

During the planning phase of a CoML seamount project, international planning workshops will bring together the varied scientific expertise required and help to engage scientists who were not represented at the initial CoML seamount workshop. By these means we will involve a broader group of scientists and expertise to improve the design of the scientific programme. Presentations at established scientific meetings will be used to raise awareness of the project.

Activity 3: Databasing and retrospective analyses

A substantial body of work exists on seamount biogeography and ecology (Figure 1). To date, however, these data remain fragmented and, in many cases, are all but inaccessible to the scientific community. Several information systems relevant to seamount biology were demonstrated at the workshop: SeamountsOnline <>, the Seamount Catalog of EarthRef <> and the MBARI Video Annotation and Reference System <>. The workshop participants recommended that the planning phase for a global CoML seamount project must continue the development of an online seamounts database and help create an analysis and synthesis effort on existing data. This is not to say that future field studies should not be undertaken until such a synthesis is complete, but rather that full advantage must be taken of existing data to assist in the planning and refining of future field efforts.

After the planning phase is complete, efforts to further develop databases should continue for the repository and archiving of new data collected during the field programme and as the programmes' contribution to the Ocean Biogeographic Information System (Zhang and Grassle 2003,

Activity 4: Outreach

As photogenic and exciting habitats, seamounts have a rich potential for public outreach and for pre-collegiate education efforts. Because of the emerging interest in conservation issues of seamounts -such as national and international efforts to site marine protected areas on seamounts and manage seamount fisheries -communicating science results to managers and decision makers was highlighted as a special emphasis for a future CoML seamounts programme. While it is beyond the scope of the CoML to make policy recommendations, scientific outcomes from seamount research are likely to be critical to decision making. Because the traditional lines of communication between science and management (i.e. publication in the primary research literature) are often not effective, new mechanisms should be considered.


The next actions towards realizing a Census of Marine Life International Field Program on seamounts will be further outreach efforts to scientists not represented at the workshop, further refinement of the science plan and the development of a programme secretariat for coordination. Anyone interested in participating in this effort or learning more can contact the authors or visit and follow the links to the CoML Seamount Program.


The authors would like to thank the Census of Marine Life and the Alfred P. Sloan Foundation for providing the funding necessary to hold this workshop. We also thank the workshop participants for their valuable contributions before, during, and after the workshop: Nancy Baron, Jacqui Burgess, Gregor Caillet, Bernd Christiansen, Sabine Christiansen, Judith Connor, Cynthia Decker, Heino Fock, Karen Garrison, Linda Glover, Raymond Highsmith, Baban Ingole, Eva Ramirez Llodra, Jon Moore, Bhavani E Narayanaswamy, Ron O'Dor, Alex Rogers, Evgeny Romanov, Ashley Rowden, Thomas Schlacher, Tim Shank, Ross Shotton, Igor Smirnov, Paul Tyler, Franz Uiblein, Mike Vecchione, Eric Vetter, Waldo Wakefield, Alan Williams, Mary Yoklavich.


Decker, C.J. & R. O'Dor 2003. A Census of marine life: unknowable, or just unknown? Oceanologica Acta, 25: 179–186.

Kitchingman, A. & S. Lai 2004. Inferences on potential seamount locations from mid-resolution bathymetric data. In T. Morato and D. Pauly (eds) Seamounts: Biodiversity and Fisheries. Fisheries Center Research Reports 12(5): 7–12.

Koslow, J.A., K. Gowlett-Holmes, J.K. Lowry, G.C.B. Poore & A. Willams 2001. Seamount benthic macrofauna off southern Tasmania: community structure and impacts of trawling. Mar. Ecol. Progr. Ser. 213: 111–125.

Parin, N.V., A.N. Mironov & K.N. Nesis 1997. Biology of the Nazca and Sala y Gomez Submarine Ridges, an Outpost of the Indo-West Pacific Fauna in the Eastern Pacific Ocean: Composition and Distribution of the Fauna, its Communities and History. Adv. Mar. Biol. 32: 145–242.

Probert, P.K., D.G. McKnight & S.L. Grove 1997. Benthic invertebrate bycatch from a deep-water trawl fishery, Chatham Rise, New Zealand. Aquatic Conservation: Mar. Freshw. Ecosyst. 7: 27–40.

Richer de Forges, B., J.A. Koslow & G.C.B. Poore 2000. Diversity and endemism of the benthic seamount macrofauna in the southwest Pacific. Nature 405: 944–947.

Rowden, A.A., S. O'Shea & M.R. Clark 2002. Benthic biodiversity of seamounts on the northwest Chatham Rise. Marine Biodiversity Biosecurity Report No. 2. Wellington: New Zealand Ministry of Fisheries, 21pp.

Smith, D.K. & T.H. Jordan 1988. Seamount statistics in the Pacific Ocean. J. Geophys. Res. 93: 2899–2919.

Stocks, K.I., G.W. Boehlert & J.F. Dower 2004 in press. Towards and international field program on seamounts within the Census of Marine Life. Archive of Fisheries and Marine Research 51.

Zhang, Y. & J.F. Grassle 2003. A portal for the Ocean Biogeographic Information System. Oceanologica Acta 25: 193–197.

Environmental assessment of fisheries -working towards an ecologically sustainable future

Victoria Wilkinson
Sustainable Fisheries, Department of Environment & Heritage
GPO Box 787, Canberra, Australia

The depletion of fish stocks and the ecological sustainability of global fisheries are issues of international concern. The Australian Government has responded to these concerns by incorporating ecological sustainability requirements into Commonwealth environment and fisheries legislation.

The Australian Oceans Policy, released in 1998, announced the Australian Government's intention to require environmental assessment of Commonwealth managed fisheries and to remove the general exemption for fisheries from the export permit requirements of environmental legislation. The purpose is to independently audit the environmental performance of Australian fisheries to ensure that they are managed in an ecologically sustainable way.

This means that decision makers must ensure that the impacts of fishing on target, bycatch and by-product protected species and the wider marine environment are sustainably managed. Achievement of a balanced triple bottom line depends on an integrated, comprehensive assessment of impacts and not an exclusive focus on target stocks. The long-term economic viability of fisheries, and their continuing role in supporting communities, depends on governments, industry and communities working together to ensure that fishing practices are ecologically sustainable.

The assessment process is facilitating a change in management practices across Australia's commercial fisheries. There has been a positive shift away from largely target-species focused management to a more ecosystem-based approach.

The paper provided an overview of Australia's Commonwealth environmental assessment requirements and a brief analysis of progress with the assessments. It also provided examples of results from a range of Australian-managed fisheries, including a number of deep-sea fisheries.

Gulf Stream and formation of intraspecific structure in fishes of northeast Atlantic

G.G. Novikov1, A.N. Stroganov1, V.N. Shibanov2 and M.S. Shevelev2
1 Biological Faculty, Moscow State University
Leninskie gori, Moscow, 119992 Russia

2 Knipovich Polar Research Institute of Marine Fisheries and Oceanography (PINRO)
6, Knipovich Street, Murmansk, 183763, Russia

The essence of ontogeny is the transformation of hereditary information into a system of living connections of the phenotype with its environment. The natural selection is based on the evaluation of phenotypes, and is the mechanism of the information feedback conversion in the biogeocenosis.

I.I. Schmalhauzen (1968)


The geographic distribution and intraspecific diversity of marine fish is, to a considerable extent, determined by the variable environmental conditions created by the complicated system of oceanic currents. The Gulf Stream is no exception in this respect.

Migrations are among most important features of specie, whose life cycles are connected with currents. Migrations lead to an expansion in the range of the species and development of migration cycles in fish. The principal peculiarity of migrations is the integrity of passive dispersion of the young and active migration of subadult and mature individuals (Marti 1980). In the same publication Marti also stresses that “migrations do not favor speciation, and a species frequently preserves its uniformity in this case”. At the same time, “the wider the reproductive range and the more variable its environment, the more probably homing is directed not to the ‘home itself' but to ‘domestic conditions’; pelagic migrants spawning in deep water may be such an example. Therefore migrations always lead to isolated existence of separate biological groups in fish. In this connection investigation of migrations poses the problem of investigation of population structure of the species, which may be solved only by proper understanding of migration cycles of small taxonomic units and their reproductive ranges (i.e. different intraspecific groups such as ecological forms, races, populations, etc.).

A population may be understood as an reproductively isolated group, representing an elementary unit for the process of evolution and possessing two major qualities (a) genetic uniformity (specificity) and (b), genetic self-dependence (isolation). These are most important when analyzing intraspecific structure. But, it is precisely these parameters that are most controversial in the determination of population status for intraspecific groups (Novikov and Karpov 1989).


Genetic uniformity (specificity) of the population depends on mechanisms that maintain mixing of population (panmixia). In natural fish populations panmixia is sustained simultaneously both among spawning individuals and in sequences of generations.

There are several factors favoring a high level of panmixia in fish:

On the other hand, such parameters as numerous spawning grounds, the presence of circulating locking currents, creation of retention areas, local ecological niches, homing, favoring reproductive isolation, etc., intensify species differentiation (Novikov and Karpov 1989).

The intraspecific structure is more pronounced when the following features are observed:

Changes in any of these features will result in changes of the existing population boundaries with subsequent alteration of intraspecific relations among different groups of individuals. Appearance of gene flow among them will be accompanied by the formation of a cline or mosaic variability of characters. The presence of prolonged pelagic developmental stages coupled with directed drifting may result in a total absence of differentiation and give rise to genetic uniformity.

A similar pattern is observed in the intraspecific structure of the main commercial fish species in the North Atlantic. Here, the presence of principal “oceanic”populations with numerous smaller local groups is connected with certain regions and currents. The fact that such groups are evolutionary related is beyond question, and comparison of their distribution with the system of main ocean currents offers a clearer view on their origin and the degree of their genetic similarity.

One may easily imagine that both modification (functional) and genetic (structural) adaptations, determining later differences in their gene pools should develop within the area of any local group living in its specific environment. In this connection some authors believe that every species could be divided into a multitude of populations, adapted to their specific conditions (Altukhov 1990).

But this situation is far from being the case in all natural situation. The following variations are possible:

The combined influence of factors such as those limiting panmixia should be considered when analyzing the intraspecific structure of a species, because the presence of only one factor favoring panmixia condition, may be sufficient to neutralize all limiting factors. Unlike other groups of animals, fishes as a whole have significant factors contributing to the gene flow between different groups in space and time, i.e. across the generations.


In the North-East Atlantic different selective factors are in effect. The adaptations are developed in coping with conditions of a new local ecological niche, which are connected primarily with changes in intensity and direction of metabolic processes, such as alterations in growth rate, age of maturation, reproductive parameters, life extent, etc. (Table 1). They are induced by differences of environmental conditions in high seas and in coastal waters, including fiords and contribute to forming biological variability and development of different ecological forms. At the same time, the spatial isolation of habitats, including breeding areas, is responsible for establishing reproductive and genetic isolation, and, consequently, for complications in the intraspecific structure.

Frequency of alleles and affinity to oxygen of cod HB-1 in the cod populations

PopulationMean many-year demersal temperatures °CFrequency of allele Hbl-1
Cod of the Baltic Sea+4+60.03
Cod of the Danish Straights+4+170.7
Cod of the Northern Sea+7+150.5
Coastal cod of the Norway Sea+5+100.4-0.3
Arcto-Norwegian cod+2+50.2-0.1
Coastal cod of the Barents Sea+2+70.3-0.2
White Sea cod-1.5+100.2
Cod of the West Atlantic+1+50.1-0.01

To better understand the complex processes of development of intraspecific structure, which take place in the North-East Atlantic, we will analyze relationships between the most distant local groups of fish inhabiting the eastern region (White Sea) and central (oceanic) areas of the North Atlantic.

The distribution and life cycles of many species of fish in the North-East Atlantic agree well with distribution of major currents of the Gulf Stream that reach the Novaya Zemlya Archipelago and the White Sea. It is obvious that the formation of the ichthyofauna of this region took place under the direct influence of the Atlantic fauna. But, by no means were all species able to adjust to the new specific ecological niches.

Such species as cod, Atlantic wolfish (one of three species inhabiting the Barents Sea), several species of flatfishes, Atlantic salmon, herrings, sculpin, and some others, colonized separate bays of the White Sea and established local population groups. But the degree of isolation of these groups, from oceanic populations to those living in the Barents Sea, differs among the various species. One may suggest a high degree of reproductive and genetic isolation of the White Sea groups of Atlantic wolfish and sculpin, which are characterized by low migratory ability of mature individuals, relatively low abundance in the North-East Atlantic and some other features.

Several other species, such as haddock and ocean perch (Sebastes marinus) did not encounter appropriate conditions for reproduction, and did not form separate stocks. The White Sea represents a part of the foraging area for adult individuals of these fishes. It also may be “a zone of sterile exportation”for the Atlantic populations of these species. To the contrary, the Atlantic cod in the White Sea forms a local population, which previously has been attributed to a separate subspecies. But studies conducted during the last decade point to the fact that no marked genetic difference exists between groups of oceanic cod from the North-East Atlantic and the White Sea population.

One of the mechanisms, supporting the genetic uniformity of populations is the combining of different fish groups within the bounds of a common reproductive zone during spawning. The pattern of distribution of alleles within the area is equalized due to such factors as a spacious spawning area, presence of differently aged individuals in the brood stock, which reflects multiple spawning during several successive years, fractional spawning with a large number of batches of eggs released, high fecundity of females and the ability to spawn with different partners including young adult males from fiords, passive drifting of eggs and larvae, and high migratory activity of adults. All these factors favor realization of panmixia.

Fish migrations in Gulf stream branches into the White Sea


Spawning time and temperature conditions of low-vertebrae herring



Relationships between the Atlantic and White Sea groups may be described as follows. When population abundance is high, the main body of the group during foraging and nesting migrations occupies all possible areas appropriate for feeding and spawning, but first of all in the fiords. A decrease of population abundance or changes in hydrologic conditions leads to restricted in-shore cod migration, and “temporary isolates”or local groups which may originate in some fiords. Their genetic composition may alter rather quickly depending on the intensity of selective pressure.

But when the contact with the main body of population recommences, genetic adaptations of temporary isolates are absorbed (eliminated) by the original population. A peculiar fluctuating mechanism is developed changing the level of genetic polymorphism [diversity] in the united population. It has been known that “the gene flow and its consequences exert an inhibitory effect on the process of evolution”(Mayr 1968) primarily during its early stages, i.e. in the course of intraspecific differentiation (Novikov and Karpov 1989).

In different parts of the area various degrees of genetic isolation of local groups is determined by the specific character of dominating currents -are there strong constant dispersing currents such as the main branches of the Gulf Stream, or circular locking currents in fiords or over some banks? The diversity of ecomorphological features of these groups is determined by the specific character of the abiotic and biotic factors.

Atlantic herrings of the genus Clupea display similar intraspecific structure. Their movement into the Baltic and White Seas is accompanied by significant differentiation according to places and season of spawning (in spring, summer or autumn), and considerable variations in average vertebral number and in some genetic indices may be evident. This suggests a different degree of divergence in the two forms with small and large vertebral number, which allows some scholars to consider them as separate species.

A high degree of ecomorphological plasticity (a wide norm of reaction to different temperature and salinity) enables herring to settle different ecological niches. Specific hydrological conditions, including currents in some bays of the White Sea, are beneficial for ecological variability according to the time of spawning, reproductive conditions, rate of maturation, etc. (Figure 2). But genetic divergence is insignificant owing to conditions favorable to mixing of individuals from different bays during the drifting of larvae from one bay to another and in regions of over-wintering by adult individuals. It is supported by the presence of a clearly defined cline in allele frequencies in different bays, which correlates well with the direction of the resulting current in the White Sea.

Thus, the intra-specific structure is the result of the realization of adaptive specific potentials (“norm of reaction”of a gene pool) under particular ecological conditions including physical barriers.


Altukhov, Yu.P. 1990. Population genetics: Diversity and Stability. London: Harwood Acad.Publ. 362 pp.

Marti, Yu.Yu. 1980. Migrations of marine fishes. Moscow: Picshepromisdat. 248 pp.

Mayr, E. 1968. Animal species and evolution (russian edition). Moscow: Mir. 597 pp.

Novikov, G.G. & A.K. Karpov 1989. Ecological aspects of the population organization of fishes. Vestn. Moscow St. University, Biology. 44/4:3–10.

Previous Page Top of Page Next Page