Life in the World’s Oceans 07

Biological research on seamounts has been limited, and we have a poor understanding of global seamount biodi-versity.. To mark the end of the fi rst Census of Marine Life, this chapter a

Life in the World’s Oceans, edited by Alasdair D McIntyre

© 2010 by Blackwell Publishing Ltd.

123

Life on Seamounts

Mireille Consalvey 1 , Malcolm R Clark 1 , Ashley A Rowden 1 , Karen I Stocks 2

1 National Institute of Water and Atmospheric Research, Wellington, New Zealand

2 San Diego Supercomputer Center, University of California San Diego, La Jolla, California, USA

of Seamount Research

The rugged terrain and vast mountain ranges that rise from

our continents inspire a strong passion, perhaps epitomized

by the fi rst climbing of Mount Everest by Sir Edmund

Hillary and Sherpa Tenzing Norgay in 1953 In the same

decade marine scientists Bruce Heezen and Marie Tharp

were looking down, deep into our oceans, mapping the

Atlantic seafl oor Painstakingly assembling echo - sounded

data, they revealed for the fi rst time the extent of the mid

Atlantic ridge At 20,000 km, it easily surpasses the length

of the Himalayas, Andes, and Rockies combined, and is the

longest mountain range on Earth On this ridge and

else-where in the oceans stand undersea mountains, or seamounts

(Box 7.1 ), the largest of which rise many kilometers from

the sea fl oor

The bathymetry of our oceans is now resolved at a scale

and detail unimaginable by early pioneers Yet despite

advances in ocean mapping we are still unable to answer

seemingly simple questions such as how many seamounts

there are To even begin to estimate the global number of

seamounts requires advanced computational technologies

In 2007 Hillier & Watts took 40 million kilometers ’ worth

of echosounder depth measurements and predicted the

occurrence of around 40,000 seamounts over 1,000 m tall,

most not yet discovered Widening their scope to include

seamounts > 100 m high, the authors predicted about

200,000, and speculated there could be as many as 3 million seamounts (Fig 7.1 )

Biological research on seamounts has been limited, and

we have a poor understanding of global seamount biodi-versity So far, fewer than 300 seamounts have been studied

in suffi cient biological detail to describe adequately the assemblage composition of seabed organisms Furthermore, sampling has been biased toward larger fauna such as fi shes, crustaceans, and corals (SeamountsOnline; Stocks 2009 ) Carl L Hubbs (1959) was one of the fi rst biologists to work on seamounts, and the questions he posed in 1959 remain relevant half a century later What species inhabit seamounts and with what regularity and abundance? How did these species disperse to, and establish on, seamounts? What bearing may the determined constitution of these isolated faunas have on our ideas concerning past and present oceanic circulation and temperatures? Do banks and seamounts provide stepping stones for trans - oceanic dispersal? To what degree has isolation led to speciation? What factors are responsible for the abundance of life over seamounts?

However, one of Hubbs ’ questions was to be answered quickly: are demersal or pelagic fi shes suffi ciently abundant

on seamounts to provide profi table fi sheries? Seamounts host signifi cant commercial fi sheries in many parts of the world Traditional handline fi sheries were likely the fi rst

fi sheries associated with seamounts (Marques da Silva & Pinho 2007 ) as far back as the fourteenth century (Brewin

et al 2007 ), and continue to the present day In the 1970s

deep - sea trawling began in earnest, targeting large seamount associated fi sh aggregations (Clark et al 2007a ) with

nations sending hundreds of vessels around the world ’ s oceans So far, at least 77 commercially valuable fi sh species have been fi shed on seamounts (Rogers 1994 ) Since the

Seamounts are prominent features of the world ’ s

underwa-ter topography, found in every ocean basin (Fig 7.1 ) They

are generally volcanic in origin, and often conical in shape

Over geological time seamounts sink (through isostatic

adjustment) and erode to become less regular The

topog-raphy of seamounts can be complex and within any

seamount one may find terraces, canyons, pinnacles,

crev-ices, and craters

Seamounts are traditionally defined by geologists as

having an elevation greater than 1,000 m above the

seabed (Menard 1964 ) Biologists now widely include peaks less than 1,000 m in their definitions, for there is

no known ecological reason for this cutoff height Pitcher

et al (2007) defined a seamount as any topographically

distinct seafloor feature that is greater than 100 m but which does not break the sea surface to become an island This definition excludes large banks and shoals (as they differ in size) and topographic features on con-tinental shelves (because of their proximity to other shallow topography)

Seamounts

Box 7.1

Fig 7.1

Location of 63,000 seamounts collated from verified regional datasets, or estimated from satellite altimetry or vessel track sounding data (CenSeam 2009)

1960s the total international catch of demersal fi shes on

seamounts by distant - water fi shing fl eets is estimated to be

over 2.25 million tonnes (Clark et al 2007a ), although the

true extent of trawling on seamounts may never be known

through a combination of catches not being reported, or

catches coming from wider areas than just seamounts

(Watson et al 2007 )

Historically seamount ecosystems have not been well

protected (Probert et al 2007 ) and have been affected by

fi shing activities that can cause declines in fi sh stocks and

visible damage to benthic habitat (Davies et al 2007 ) A

great deal of fi shing effort has, and continues to, occur

on the high seas and many fi sheries proceeded largely

unregulated, falling outside of any nation ’ s jurisdiction

Although the United Nations and regional fi sheries

management organizations are becoming more effective,

enforcement of regulations on the high seas remains a

challenge

Emergent threats such as deep - sea mineral extraction

and indirect threats to all deep - sea habitats are also

increas-ingly being considered, such as rising CO 2 (Guinotte et al

2006 ) High - profi le governmental and non - governmental

initiatives have elevated the position of seamounts in the

public eye

Recognizing it is not feasible to sample all of the

world ’ s seamounts, research efforts needed to be

coordi-nated to assess the current state of knowledge, fi ll critical

knowledge gaps, and target understudied regions and

seamount types The Global Census of Marine Life on

Seamounts (CenSeam) has provided a focal point for

coor-dinating global research and for communicating research

results to stakeholders seeking scientifi c advice and

guid-ance To mark the end of the fi rst Census of Marine Life,

this chapter addresses some of the core research questions

that have faced seamount researchers, including those of

the CenSeam project, over the past fi ve years It also

indi-cates where seamount research is likely to be directed in

the future

Marine Life on Seamounts (CenSeam)

The fi eld of seamount biology has grown in recent decades,

as shown by the increasing number of scientifi c

publica-tions each year (Brewin et al 2007 ) The Census fi eld

project CenSeam started in 2005, and has served to bring together more than 500 seamount researchers, policy makers, environmental managers, and conservationists from every continent At the outset of CenSeam, our understanding of seamount ecosystems was hampered by signifi cant gaps in global sampling, a variety of approaches and sampling methods, and a lack of large - scale synthesis; scientifi c attention was not yet consistent with their potential biological and ecological value (Stocks et al

2004 ) CenSeam has aimed to do the following: (1) syn-thesize and analyze existing data (Box 7.2 ); (2) coordinate and expand existing and planned research (Box 7.3 ); (3) communicate the fi ndings through public education and outreach; and (4) identify priority areas for research and foster scientifi c expeditions to these regions CenSeam researchers have augmented sampling efforts and analyses

in the well studied Southwest Pacifi c and Northern Atlan-tic CenSeam has also identifi ed three key undersampled regions: the Indian Ocean, the South Atlantic, the Western

Since 2005, SeamountsOnline (Stocks 2009 ) has been

col-lecting data on species that have been recorded from

seamounts globally, and making them available through a

free online data portal (Fig B7.2 ) By bringing together

global seamount data into a standardized, searchable,

electronic format, SeamountsOnline facilitates research

and management objectives looking at patterns across

dif-ferent seamounts and regions

Through a map interface, users can select seamounts

of interest, or see the distribution of taxa globally Users can

search for information by management boundaries, such

as Exclusive Economic Zones, and biogeographic region-alizations, such as Longhurst Provinces Seamounts can also be searched by summit depth Taxonomic searching has options for searching by phylum, class, order, or family,

in addition to genus or species All species observations in SeamountsOnline are also contributed to the Ocean Bio-geographic Information System ( www.iobis.org ), which integrates data from across all the Census of Marine Life projects

SeamountsOnline: Providing Researchers and Managers with Tools for

Finding and Accessing Information on the Biological Communities that live on

Seamounts

Box 7.2

On seamount voyages researchers will typically conduct a

bathymetric survey (usually using multibeam sonar) of the

target seamount The resulting baythmetric map provides

the basis for more detailed planning of the sampling

program: plans that will take into account factors such as

seamount size, shape, and depth Echosounder

informa-tion can also be used to identify substrate type and can

guide sampling to target soft and hard bottoms

The sampling gear and methodology used will depend

on the nature of the research and the in situ conditions, for

example weather and substrate For biodiversity surveys,

camera transects (undertaken using towed camera plat-forms, remotely operated vehicles, or submersibles) should

be performed where possible Remote methodologies have the advantage of being non - destructive and enabling researchers to view intact community composition, and to potentially gain valuable information on animal behavior However, to quantify biodiversity fully, “ ground - truthing ” is required, and physical collection is vital for the completion

of a full taxonomic inventory A combination of sampling gears (for example grabs, corers, dredges, sleds, trawls) may be deployed on seamount surveys, but the hard and

Sampling Seamounts

Box 7.3

Fig B7.2

SeamountsOnline is a free online data portal that makes available global data on species that have been recorded from seamounts (seamounts.sdsc.edu, see also its sister database SeamountCatalog at earthref.org/cgi - bin/er.cgi?s=sc - s0 - main.cgi)

(A) (B)

Fig B7.3

Researchers will typically conduct (A) a bathymetric survey followed by (B) camera transects e.g., Deep Towed Imaging System (pictured) and then

collect physical specimens using a (C) beam trawl and/or (D) epibenthic sled (National Institute of Water and Atmospheric Research)

rough ground that frequently prevails on seamounts may

limit researchers to the use of towed dredges or sleds

(Fig B7.3 )

The sample from each gear type deployed is sorted

on board and separated out as close to species or

puta-tive species (that is, apparently morphologically distinct

organisms, sometimes called operational taxonomic units)

as possible The samples are then chemically fixed or

frozen (following taxa - specific recommendations, as well

as taking into account genetic sampling requirements)

At the end of the voyage the samples will be delivered

to taxonomists who will complete the final faunal inventory This assessment of biodiversity can take many years, based on the high numbers of samples and low numbers

of taxonomists

Two working groups have helped drive the CenSeam research effort; the Data Analysis (DAWG) and Standardi-zation Working Group (SWG) Members of each group have convened several workshops to tackle specific research questions and challenges, for example standardizing survey design, sampling, and analysis techniques (where possible) to facilitate geographic comparisons

and Southern Central Pacifi c, and researchers have worked

toward securing funding to sample these regions So far,

CenSeam - linked scientists have participated in over 20

voyages

To help focus global research efforts, the CenSeam community identifi ed two overarching priority themes: (1) What factors drive community composition and diver-sity on seamounts, including any differences between

Seamount ecosystem

Fig 7.2

The CenSeam field program aims to

investigate what factors drive

community composition and

diversity on seamounts, and to

understand better the impacts of

human activities such as fishing on

seamounts (Erika Mackay, National

Institute of Water and Atmospheric

Research)

seamounts and other habitat types? (2) What are the

impacts of human activities on seamount community

struc-ture and function? (Fig 7.2 ) Within these themes key

questions were developed to address where more science

was needed to improve our understanding of the structure

and functioning of seamount ecosystems and to inform

management and conservation objectives These questions

will be used below to present what is known about

seamounts so far, and how CenSeam has contributed to

this knowledge The CenSeam research effort has focused

on seamount mega - and macroinvertebrates

Community Composition and

Diversity on Seamounts?

Effective management of any seamount ecosystem must be

based on a solid understanding of the seamount

commu-nity, and associated physical and biological processes

Fur-thermore, it is important to determine the interactions of

seamount communities with those in the wider deep - sea

realm

7.3.1 Seamount c ommunity

c omposition and d iversity

The dominant large fauna of hard substrate on many deep

sea seamounts are attached, sessile organisms that feed on

particles of food suspended in the water (Fig 7.3 ) These

suspension feeders are predominantly from the phylum

Cnidaria, which includes stony corals, gorgonian corals,

black corals, sea anemones, sea pens, and hydroids Deep sea (or cold water) corals are one of the most studied groups, and CenSeam has promoted research on their

global distributions (Clark et al 2006 ; Rogers et al 2007 ; Tittensor et al 2009 ) Corals can grow as individual

colo-nies, or can coalesce as reefs; potentially providing complex three - dimensional habitat for a wide range of other animals, providing more refuge, an enhanced supply of food, surface area for settlement, and microhabitat variability to support

a greater faunal diversity than less complex habitat However, the role of biogenic habitat in the deep sea has only recently emerged as an area of both academic and conservation interest, and only a few quantitative studies have been made of the relationship between biogenic habi-tats and the composition of seamount fauna (see, for

example, O ’ Hara et al 2008 )

In the literature there are many studies that describe the

fi sh that live on seamounts The most recent review by

Morato et al (2004) identifi ed a total of 798 species of

seamount fi sh, though exactly how to defi ne a “ seamount

fi sh ” is not straightforward Commonly cited examples

include the orange roughy ( Hoplostethus atlanticus ), alfon-sino ( Beryx splendens ), Patagonian toothfi sh ( Dissostichus eleginoides ), oreos ( Pseudocyttus maculatus , Allocyttus niger ), and pelagic armourhead ( Pseudopentaceros wheeleri )

(Table 7.1 ) Sharks and tuna are also reported as occurring

on seamounts, and in the waters above some shallow seamounts serranids (including sea basses and the groupers) and jacks are observed to spawn (Morato & Clark 2007 ) Seamounts are popularly referred to as hot spots of high species richness in the deep sea However, many research-ers are failing to fi nd support for this premise Stocks & Hart (2007) report variability but no overall trend of ele-vated species richness across approximately 18 studies

(A) (B)

Fig 7.3

Corals on seamounts can grow as reefs or as individual colonies (A) Solenosmilia variabilis on Ghoul Seamount (approximately 1,000 m; New Zealand;

National Institute of Water and Atmospheric Research) (B) Paragorgia arborea and a dense population of basket stars Gorgonocephalus sp on San Juan

Seamount (USA; courtesy of the Monterey Bay Aquarium Research Institute) (C) Viminella on the summit of Condor Seamount (200 m; Azores, North

Atlantic; © Greenpeace/Gavin Newman) (D) Paragorgiid, acanthogorgiid, and chrysogorgiid corals on Pioneer Seamount (approximately 1,700 – 1,800 m;

Northwestern Hawaiian Islands; 2003 NWHI exploration team: Amy Baco - Taylor, Chris Kelley, John Smith, and pilot Terry Kerby, NOAA Office of Ocean

Exploration and Hawaii Undersea Research Laboratory)

comparing seamounts to either surrounding deep sea or

nearby continental margins However, sampling - related

issues complicate such comparisons, an issue that has been

addressed within the CenSeam Data Analysis Working

Group (DAWG) Taking sampling factors into account,

DAWG member O ’ Hara (2007) compared levels of

ophiuroid species richness between seamount and non

seamount areas (for the latter by randomly generating

pop-ulations from areas and depth ranges that refl ected the

typical sampling profi le of seamounts) and concluded that

seamounts do not show elevated levels of species richness

At macro - ecological scales, the fauna of individual

seamounts have been found to broadly refl ect the species

pools present on neighboring seamounts and continental

margins (see, for example, Samadi et al 2006 ; Stocks &

Hart 2007 ; McClain et al 2009 ; Clark et al 2010 ; Brewin

et al 2009 ) Although the main body of evidence suggests

that broad assemblage composition may be similar to sur-rounding deep - sea environments, community structure

may differ between habitats For example McClain et al

(2009) present preliminary evidence that the faunal com-munities of Davidson Seamount (off the west coast of the USA), although similar in composition to adjacent canyon habitat, are structurally different, particularly in the frequency of occurrence of particular species

Seamounts are hypothesized to serve as biogeographical “ islands ” that could also function as shallow stepping stones across the abyssal plains The isolated nature of many sea-mounts has fueled the hypothesis that seasea-mounts can sup port high levels of endemicity, and numerous studies have

sup-ported this hypothesis (Richer de Forges et al 2000 ; see

review by Stocks & Hart 2007 ) However, so far, it is unlikely that we have identifi ed enough of the regional or global deep sea fauna to use the term endemic with any confi dence, and

Table 7.1

Distribution of main commercial fi sh species on seamounts (North Atlantic ( NA ); South Atlantic ( SA ); North Pacifi c ( NP ); South Pacifi c ( SP ); Indian Ocean ( IO ); Southern Ocean ( SO )) and the depth range commonly fi shed

Southern boarfish Pseudopentaceros richardsoni SA, SP, IO 600 – 900

Oreos Pseudocyttus maculatus, Allocyttus niger SA, SP, IO, SO 600 – 1,200

Redfish Sebastes spp ( S marinus , S mentella , S proriger ) NA, NP 400 – 800

apparently high levels of endemism may be an artifact of

sam-pling species - rich communities or uneven samsam-pling effort

Furthermore, many seamount fauna are recorded to have

global or near - global distributions including reef - building

scleractinian corals ( Lophelia pertusa , Solenosmilia

variabi-lis , and Madrepora oculata ) (Roberts et al 2006 ) and fi sh

species such as orange roughy (Francis & Clark 2005 )

In summary, seamounts can host abundant and diverse

benthic communities However, in many instances

com-munity composition is similar to those of adjacent habitats

including continental slope Today the concept of seamounts

being islands in the sea has little support, but more

sampling is required to be able to address this idea fully

7.3.2 Connectivity of s eamount

p opulations

Differences in the connectivity of faunal populations among

seamounts is almost certainly an important determinant of

community composition on seamounts, a potential driver of

endemicity, and a major consideration for the management

of seamount ecosystems The dispersal capabilities of deep sea fauna depend primarily on whether species can disperse

as adults, or only as eggs, larvae, and/or post - larvae; however, one cannot fully predict the distribution of a species based on larval life history alone (see Johannesson ’ s (1988) paradox

of Rockall) Relatively little is known about the life - history traits of deep - sea organisms, including those found on seamounts Studies so far have been restricted to a single seamount (Parker & Tunnicliffe 1994 ) or limited taxa (see, for example, Calder (2000) on hydroids) Distance from shore, or degree of isolation from other seamounts, is widely proposed as an important factor determining community composition and richness Leal & Bouchet (1991) report a signifi cant decline in species richness of prosobranch gastro-pods moving offshore along the Vit ó ria - Trindade seamount chain, but could not attribute this to any differences in larval life histories and posed that species may be passively dis-persed along the chain through “ island hopping ” In fact a suite of physical and biological factors also infl uence

dispersal, and hence connectivity, over space and time In a

major CenSeam review paper, Clark et al (2010) break these

factors down: (1) physical ocean structure (for example

hydrographic retention, large - scale and local currents), (2)

factors infl uencing larval development time (for example

temperature, food availability, predation), (3) habitat

avail-ability for larval settlement, and (4) post - settlement survival;

with interactions thereof driving variations in the dispersal

capabilities of fauna among seamounts

Genetic studies are essential to our understanding of

con-nectivity, but so far have been limited to few fauna and by the

sensitivity of the current techniques Historically, seamount

genetic connectivity studies focused on commercially fi shed

fauna but in recent years efforts have expanded to non

commercial fauna No consistent pattern has emerged, with

mixed results indicating both genetic differentiation and

genetic homogeneity between some commercially fi shed

fauna on seamounts, and those on oceanic islands and the

continental margins at both oceanic and regional scales

(Aboim et al 2005 ; Stockley et al 2005 )

Baco & Shank (2005) discovered relatively high levels

of genetic diversity, as well as low yet signifi cant levels of

population differentiation, for the precious coral

Coral-lium lauuense among several Hawaiian seamounts and

islands They suggested that C lauuense are primarily self

recruiting with occasional long - distance dispersal events

maintaining genetic connectivity between sites In contrast,

Smith et al (2004) provided evidence of widespread

dis-tribution of bamboo coral species in the Pacifi c which

were not endemic to seamounts However, the authors

could not rule out that the mitochondrial markers they

used in their analysis were insensitive to recent speciation

events Samadi et al (2006) determined that populations

of a gastropod with dispersive larvae were more similar

than populations of non - planktotrophic gastropod species

Samadi et al (2006) also determined that dispersive squat

lobster species were genetically similar among populations

on seamounts and the adjacent island slope

The potential ability of certain seamount fauna to

dis-perse widely is perhaps not surprising when compared with

the fi nding of dispersive fauna at other isolated deep - sea

habitats, such as hydrothermal vents or cold seeps (Samadi

et al 2007 ; see Chapter 9 )

As well as understanding the dispersal characteristics of

seamount fauna, it is vital to set these in the context of both

large - scale oceanic circulation and localized hydrological

phenomena For example, Taylor cones have been cited as

a possible trapping mechanism that may drive endemism

Mullineaux & Mills (1997) recorded larval concentrations

above and around Fieberling Guyot to be consistent with

modeled tidally rectifi ed recirculation over the seamount

Parker & Tunnicliffe (1994) proposed the presence of a

modifi ed Taylor cap on Cobb Seamount was important

for trapping short - lived larvae, but because water mass is

replaced approximately every 17 days, concluded that

medium and long - lived larvae would be dispersed Recent research by DAWG members has concluded, for some faunal groups, that seamount - scaled oceanographic reten-tion is weak compared with other ecological drivers of

community diversity on seamounts (Brewin et al 2009 )

To conclude, current understanding of the dispersal capabilities of seamount fauna, and the role of large and smaller - scale oceanographic processes, is limited and as such

we cannot assess the role that dispersal may play in produc-ing spatial differences between communities The premise that seamounts may be dominated by short lived or non planktonic larval phased fauna has not been widely tested

7.3.3 Environmental f actors

d riving d ifferences in d iversity and s pecies c omposition of

s eamount f auna

Seamount communities, as with slope and abyssal faunas, may exhibit latitudinal turnover in species composition For example, O ’ Hara (2007) reports a clear biogeographi-cal gradient for both seamount and non - seamount ophiuroid fauna from the tropics to the sub - Antarctic Though incom-plete, research so far has demonstrated that environmental factors can vary at large spatial scales, hence, can have the potential to infl uence community composition of deep - sea fauna

Seamounts differ in their location, depth, elevation, and

geological history (Rowden et al 2005 ), all factors that may

alter environmental conditions on large and small spatial scales and, in turn, infl uence seamount biodiversity and

species composition Clark et al (2010) list as among the

main factors that may determine the character of seamount benthic assemblages: seamount geomorphology, geological origin and age; local hydrodynamic regime (all preceding infl uence substrate type); light levels; water chemistry (for example oxygen); productivity of the overlying water (which relates to food availability); as well as the presence

of volcanic/hydrothermal activity (see Chapter 9 ), tempera-ture, and pressure All these factors may operate in tandem and can create a unique set of conditions for a region, for any given seamount, and within a seamount

Most marine animals are restricted to a limited

bathy-metric range (see, for example, Rex et al 1999 ), and recent

work by CenSeam - linked researchers has demonstrated that seamount assemblages can be depth stratifi ed (O ’ Hara

2007 ; Lundsten et al 2009 ) Work on the very deep slopes

of seamounts has been limited but new research indicates that these can support distinct assemblages (see, for example, Baco 2007 ) However, the depth - related patterns, and the drivers thereof, remain largely unexplored for seamounts, but environmental gradients that correlate with depth such as temperature, oxygen concentration, food

availability, and pressure are likely to be as important as

they are for other deep - sea habitats (Clark et al 2010 )

Large seamount chains can divert major currents, and

individual seamounts can affect localized hydrographic

events including the formation of a rotating body of water

retained over the summit of a seamount (Taylor cone,

which may fl atten to a cap) and the generation or

interac-tion with internal waves (White et al 2007 ) These may

infl uence the faunal composition through larval transport

(section 7.3.2 ) Additionally currents can be amplifi ed

around seamounts creating favorable conditions for

sus-pension feeders as the waters bring an increased particle

supply, as well as removing sediments (Genin 2004 )

Sus-pension feeders (for example corals, sponges, hydroids,

crinoids, anemones, sea pens, feather stars, and brittlestars)

have been found to dominate some seamounts (particularly

their peaks) (Genin et al 1986 ; Wilson & Kaufmann

1987 ), and large sessile fauna can, in turn, form structural

habitat for a diverse range of smaller, mobile fauna (section

7.3.1 )

Exposed rock surfaces are limited in the deep sea and

seamounts represent a signifi cant source of this substrate

(Gage & Tyler 1991 ) Soft sediments can also dominate

some seamounts, particularly fl at - topped seamounts, called

guyots, and in these circumstances community composition

can switch from suspension to deposit feeders, similar to

neighboring continental slopes (see, for example, Á vila &

Malaquias 2003 ; Lundsten et al 2009 ) The composition

of infaunal communities on seamounts have not been well

studied but include a wide diversity of polychaetes,

crus-taceans, mollusks, ribbon worms, peanut worms, and

oligochaetes, as well as meiofaunal organisms such as

nema-tode worms, loriciferans, and kinorhynchs (Samadi et al

2007 )

Although depth - related factors and substrate type

(including biogenic habitat) are important drivers of

com-munity composition on seamounts, more research is

required to describe and explore large - scale biogeographic

patterns on seamounts

7.3.4 Productivity on s eamounts

Enhanced productivity at seamounts is a widely cited

phenomenon, and seamounts appear to support relatively

large planktonic and higher consumer (fi sh) biomass when

compared with surrounding ocean waters, particularly

so in oligotrophic oceans (Genin & Dower 2007 )

Seamounts can each have their own local oceanographic

regimes (section 7.3.3 ), which could infl uence seamount

productivity

Elevated phytoplankton concentrations have been

observed on some seamounts (see, for example, Genin &

Boehlert 1985 ; Dower et al 1992 ; Mouri ñ o et al 2001 ),

and it has been theorized that nutrient - rich upwelled waters

and eddies around a seamount enhance surface primary productivity which leads to an energy transfer to higher trophic levels However, the persistence of upwelling does not generally seem suffi cient for such a transfer, making this an unlikely explanation for the elevated zooplankton and fi sh biomasses found over seamounts (Genin & Dower

2007 )

It is now proposed that the high biomass on some seamounts may be fueled not by enhancement of primary production but instead by a trophic subsidy to carnivores (Genin & Dower 2007 ) These authors proposed food inputs by the following routes: (1) bottom trapping of vertically migrating zooplankton, (2) greatly enhanced horizontal fl uxes of suspended food through current acceleration, and (3) amplifi cation of internal waves increasing horizontal fl uxes of planktonic prey Porteiro & Sutton (2007) proposed that fi sh behavior may have evolved to capitalize on the regular planktonic food supply passing over a seamount, enabling them to convert mid - trophic level biomass effi ciently to higher trophic levels

The importance of biogenic structure in supporting higher fi sh densities has been widely cited but work so far

has yielded mixed results (see, for example, Huseb ø et al

2002 ; Krieger & Wing 2002 ; Ross & Quattrini 2007 ) Hydrographic factors around deep - sea coral reefs may increase zooplankton density (Dower & Mackas 1996 ;

Huseb ø et al 2002 ) in turn benefi ting planktivorous fi sh,

but there is no consistent explanation for enhanced fi sh productivity over seamounts Fish aggregations may occur independently of biogenic fauna; for example, orange roughy are observed to spawn over seamounts but do not feed during spawning (Morato & Clark 2007 ) Clearly, more information on the life histories of many seamount

fi sh (for example larval stages, juvenile fi sh grounds) is needed

In summary, enhanced secondary productivity over seamounts is most likely attributable to a food supply exported from elsewhere, and not locally enhanced primary productivity The infl uence of bio - physical coupling, and the potentially complicated and varied interactions of forcing mechanisms and seamount topographies, is uncer-tain Vital to future research on seamount - related produc-tivity will be the establishment of long - term monitoring programs, in concert with the development of physical and trophic models

7.3.5 Trophic a rchitecture of

s eamount c ommunities

Large suspension feeders such as corals, sponges, and cri-noids can dominate the biomass of the seamount mega-benthos on hard substrates A dominance of suspension