Deep-Sea Research II 45 (1998) 465—515
Temporal and spatial patterns of biological
community development at nascent deep-sea
hydrothermal vents (9°50@N, East Pacific Rise)
Timothy M. Shank!,*, Daniel J. Fornari", Karen L. Von Damm#,
Marvin D. Lilley$, Rachel M. Haymon%, Richard A. Lutz!
! Institute of Marine and Coastal Sciences, Rutgers University, New Brunswick, NJ 08903-0231, USA
" Department of Geology & Geophysics, Woods Hole Oceanographic Institution, Woods Hole,
MA 02543, USA
# Department of Earth Sciences, University of New Hamphsire, Durham, NH 03824-3589, USA
$ School of Oceanography, University of Washington, Seattle, WA 98195, USA
% Department of Geological Sciences, University of California at Santa Barbara, Santa Barbara,
CA 93106, USA
Received 10 March 1997; received in revised form 19 August 1997
Abstract
The April 1991 discovery of newly formed hydrothermal vents in areas of recent volcanic
eruption between 9°45@N and 9°52@N on the East Pacific Rise provided a unique opportunity to
follow temporal changes in biological community structure from the ‘‘birth’’ of numerous
deep-sea hydrothermal vents. In March l992, DS» Alvin was used to deploy an on-bottom
observatory, the Biologic—Geologic Transect, to monitor faunal succession along a 1.37 km
segment of the axial summit caldera between 9°49.61@N and 9°50.36@N (depth &2520 m).
Photo- and videographic documentation of megafaunal colonization and chemical analyses of
diffuse hydrothermal fluids associated with many of these developing communities within the
Transect were performed in March 1992, December 1993, October 1994, and November 1995.
Photographic and chemical time-series analyses revealed the following sequence of events in
low-temperature venting areas. (1) Immediately following the 1991 eruption, hydrogen sulfide
and iron concentrations in diffuse fluids were extremely high ('1 mmol kg~1) and microbially
derived material blanketed active areas of venting in the form of thick microbial mats.
(2) Mobile vent fauna (e.g. amphipods, copepods, octopods, and galatheid and brachyuran
crabs) and non-vent fauna (e.g. nematocarcinid shrimp) proliferated in response to this increased biological production. (3) Within 1 yr of the eruption, areal coverage of microbial mats
* Corresponding author. Tel.: 001 908 932 8959; fax: 001 908 932 6557; e-mail: shank@imcs.rutgers.edu.
0967-0645/98/$19.00 ( 1998 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 6 7 - 0 6 4 5 ( 9 7 ) 0 0 0 8 9 - I
466
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
was reduced by &60% and individuals of the vestimentiferan tube worm ¹evnia jerichonana
settled gregariously in areas where diffuse flow was most intense. (4) Two years after the
eruption, maximum levels of H S decreased by almost half (from 1.90 to 0.97 mmol kg~1) and
2
dense thickets of the vestimentiferan Riftia pachyptila dominated vent openings previously
inhabited by ¹evnia jerichonana. (5) Three years after the eruption, maximum hydrogen sulfide
levels declined further to 0.88 mmol kg~1 and mussels (Bathymodiolus thermophilus) were
observed on basaltic substrates. (6) Four years after the eruption, galatheid crabs and serpulid
polychaetes increased in abundance and were observed close to active vent openings as
maximum hydrogen levels decreased to 0.72 mmol kg~1. Also by this time mussels had
colonized on to tubes of Riftia pachyptila. (7) Between 3 and 5 yr after the eruption, there was
a 2- to 3-fold increase in the number of species in the faunal assemblages. In the absence of
additional volcanic/tectonic disturbance, we predict that mytilid and vesicomyid bivalves will
gradually replace vestimentiferans as the dominant megafauna 5—10 yr following the eruption.
We also anticipate that the abundance of suspension feeders will decline during this period
while the abundance of carnivores will increase. We hypothesize that the above series of events
(1—7) represents a general sequence of biological successional changes that will occur at newly
formed low-temperature deep-sea hydrothermal vents along the northern East Pacific Rise and
contiguous ridge axes.
Megafaunal colonization at deep-sea hydrothermal vents is considered to be the consequence
of an intimate interaction of the life-history strategies of individual species, physical oceanographic processes, and the dynamic hydrothermal environment. Our observations indicate that
the successful sequential colonization of dominant megafaunal vent species, from ¹evnia
jerichonana to Riftia pachyptila to Bathymodiolus thermophilus, also may be strongly influenced
by temporal changes in geochemical conditions. Additional evidence demonstrating the close
link between diffuse vent flux, fluid geochemistry, and faunal succession included the rapid
death of several newly formed biological assemblages coincident with abrupt changes in the
geochemical composition of the venting fluid and the local refocusing or cessation of vent flow.
These correlations suggest that future models of faunal succession at hydrothermal vents along
intermediate to fast-spreading mid-ocean ridges should consider not only the interplay of
species-specific life-history strategies, community productivity, and physical oceanographic
processes, but also the influence of changing geochemical conditions on the sequential colonization of megafaunal species. ( 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
In many space-limited marine communities, abiotic disturbance events are key
structuring forces (Connell, 1979; Sousa, 1980; Keough, 1984; Pickett and White,
1985). Studies investigating the re-establishment and development of communities
after natural, large-scale catastrophic disturbances have been limited largely to
terrestrial ecosystems following volcanic eruptions [e.g. Anak Krakatau Island (Bush
and Whittaker, 1991; Thornton and Walsh, 1992), Mount St. Helens (Dale, 1991),
Pitcairn Island (Diamond, 1994), and Surtsey Island (Fridriksson and Magnusson,
1989)]. Hydrothermal vent habitats in the deep ocean also experience large-scale
catastrophic events. Frequent volcanic eruptions and tectonic disturbance (Haymon
et al., 1993; Embley et al., 1995), as well as inherently unsteady vent fluid convection
through young oceanic crust (Watremez and Kervevan, 1990), create highly variable,
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
467
transient habitats. Understanding the processes by which deep-sea hydrothermal
vent-endemic species successfully establish and maintain communities in patchily
distributed, ephemeral habitats has been a fundamental goal of hydrothermal vent
ecology (Tunnicliffe, 1991). The April 1991 discovery of newly formed hydrothermal
vents in volcanically active areas of the fast-spreading (11—12 cm yr~1; Carbotte and
MacDonald, 1992) East Pacific Rise (EPR) between 9°45@N and 9°52@N (Haymon
et al., 1993) presented a rare opportunity to document the colonization of newly
formed vents, examine the chemical, geological, and oceanographic processes that
impact community establishment, and define time-scales over which communities
develop, undergo structural/faunal changes, and eventually die. Additionally, this
event afforded the opportunity to assess existing hypotheses of faunal succession at
deep-sea hydrothermal vents (Hessler et al., 1988; Fustec et al., 1987; Jollivet, 1993)
and develop new descriptive models.
Most previous studies of temporal and spatial change in mid-ocean ridge (MOR)
vent communities have been conducted without knowing the age of the existing
community relative to the inception of hydrothermal activity caused by volcanic and
tectonic activity (e.g. Fustec et al., 1987; Hessler et al., 1988; Johnson and Tunnicliffe,
1988; Tunnicliffe and Juniper, 1990; Tunnicliffe et al., 1990). Attempts to relate the
relative age of lava flows to faunal composition have been conducted by Milligan and
Tunnicliffe (1994) along the Cleft segment, Juan de Fuca Ridge, and Geistdoerfer et al.
(1995) and Auzende et al. (1994) along the southern EPR. The relative age of biological
communities and the initiation of venting at certain vent sites at 13°N EPR also have
been inferred from records of seismic activity by Jollivet (1993). During the past
7 years, three submarine eruptions and the accompanying onset of hydrothermal
activity have been documented on the MOR: (1) along 9°—10°N EPR in 1991
(Haymon et al., 1993; Gregg et al., 1996); (2) at CoAxial Seamount of the Juan de Fuca
Ridge in 1993 (Delaney and Embley, 1993; Embley et al., 1995; Tunnicliffe et al., 1998);
and (3) along the northern Gorda Ridge near 42°40@N in 1996 (Embley et al., 1996).
Anthropogenically induced venting brought about by deep-drilling at Middle Valley
(Holden, 1996) and the recent eruption of Loihi Seamount (Duennebier et al., 1997)
represent other fortuitous events that provide opportunities for further study of the
natural development of vent-endemic faunal communities.
Over the past 15 years, observations during repeat visits to eastern Pacific vent sites
indicate that well-established biological vent communities undergo substantial changes in their composition and distribution (Fustec et al., 1987; Hessler and Smithey,
1983; Hessler et al., 1985; Hessler et al., 1988; Tunnicliffe et al., 1990; Van Dover and
Hessler, 1990; Jollivet, 1993). Wholly or partially dead beds of clams (Calyptogena
magnifica) observed at 21°N EPR between 1982 and 1990 attest to the effect discontinuous vent emission can have on biological communities (Lutz, 1991; Fisher et al.,
1988). The abundance of Riftia at 21°N was virtually eliminated by intensive research sampling in 1982, but dramatically increased by 1990 (Lutz, 1991). Vent sites
near 13°N EPR experienced periods of marked instability in venting activity between
1982 and 1990 (Jollivet, 1993). Inactive periods of venting were characterized by
a sharp decrease in Riftia populations, an increase in scavengers, and a persistence of
mytilid bivalves (Bathymodiolus thermophilus). Reactivation of extinct vent areas was
468
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
correlated with increased numbers of ¹evnia jerichonana and Bythograea thermydron
(Jollivet, 1993). Under seemingly constant venting conditions at Rose Garden vents
(Galápagos Rift) from 1979 to 1990 (Hessler and Desbruyères, 1991), the abundance of
Riftia declined sharply and was replaced with the mytilid B. thermophilus and the
vesicomyid C. magnifica. At the Mushroom Vent site on Axial Seamount along the
Juan de Fuca Ridge, the natural development of vestimentiferan communities was
significantly altered by organism sampling and the excavation and removal of sulfides
(Tunnicliffe and Juniper, 1990). It thus appears that the spatial readjustment of vent
emissions, altered intensity of fluid flux, and the physical effects of tectonic perturbations and anthropogenic disturbance can have dramatic effects on the temporal and
spatial variation in vent community structure. These studies also demonstrate that
dramatic faunal changes can be observed over periods of several years, although
distinguishing between naturally occurring faunal changes and the effects of anthropogenic disturbance (e.g. extensive sampling, manipulative experiments, and submersible maneuvering) can be extremely difficult (Tunnicliffe, 1990).
In this study our main goals are to: (1) document the temporal sequence of species
colonization and dominance within spatially separated low-temperature faunal communities; (2) assess the rate at which various species colonize new vents; (3) correlate
changes in community composition with changes in diffuse vent fluid flux and
geochemistry; and (4) develop a descriptive model of vent faunal ‘‘succession’’ along
certain intermediate- to fast-spreading MORs in the eastern Pacific. In order to
accomplish these goals, we established a long-term study area for monitoring seafloor
processes between 9°49.61@N and 9°50.36@N on the EPR crest (Fig. 1). This area,
termed the Biologic—Geologic Transect (Transect), extends along a 1.37 km long
section of the floor of the axial summit caldera (ASC) and traversed seven hightemperature vents and at least 17 areas of low temperature, diffuse flow that formed
during the 1991 eruption. Photographic and videographic images were obtained
15 months prior to the eruption, and 2 weeks, 11 months, 32 months, 42 months, and
55 months after the eruption in order to document temporal changes in biological
community structure. In addition to the extensive photographic and videographic
images, concomitant time-series measurements and analyses of low-temperature
diffuse vent fluid chemistry and temperature permitted correlation of abiotic factors
with observed changes in community structure.
2. Methods
2.1. Study area
The following considerations were used in selecting the region between 9°49.61@N
and 9°50.36@N as the Transect area: (1) a November 1989 photosurvey (conducted
using the deep-towed vehicle Argo) identified this area of the EPR crest as one of
the most hydrothermally active segments between 9°09@N and 9°54@N, (Haymon
et al., 1991) with well-developed vent communities between 9°45@N and 9°51@N; (2) in
April 1991, venting in the area between 9°49.61@N and 9°50.36@N ranged from
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
469
Fig. 1. Bathymetric map of the East Pacific Rise ridge crest between 9°49@N and 9°51@N indicating the
location of the 1.37 km long Biologic—Geologic Transect (yellow boxed area) and principal high temperature vents (shown in white and vertically exaggerated) in this region. Grid (interval"20 m) constructed
from Alvin altitude and depth, and corrected Sea Beam depth data. Inset (top left) shows the location of the
9°—10° North field area between the Clipperton (C) and Siquieros (S) Transform Faults, and other EPR vent
fields located at 21°N, 13°N, and 11°N, with the Rivera (R) and Orozco (O) Transform Faults separating
these sites. Map (at right) depicts the positions of the 210 Biomarkers of the Biologic— Geologic Transect
deployed along nascent and pre-eruption vent areas within the axial summit caldera (ASC) between
9°49.61@N and 9°50.36@N in March 1992. Red squares and black diamonds indicate the relative positions of
the Biomarkers and the walls of the axial summit caldera, respectively. The ‘‘Hole-to-Hell’’ (black box) area
is bounded by two areas of high-temperature venting (‘‘Bio 9’’ and ‘‘P vent’’). Arrows indicate locations of
biological colonies, as well as two pre-eruption communities (Bio182 Musselbed and the Tube Worm Pillar)
whose development has been documented since the April 1991 eruption.
470
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Fig. 2. Polyethylene ‘‘Biomarkers’’ deployed along the axial summit caldera floor. Each Biomarker
consisted of a vertical polyethylene panel (45 cm]30 cm]1 cm) and an attached, horizontal overhanging
polyethylene ‘‘flap’’ (15 cm]30 cm]1 cm). Analog numbers (1—210), an optically readable eight-bit binary
code system (consisting of 2.5 cm holes, recognizable in the event of fouling), and reflective and phosphorescent tapes were employed to facilitate Biomarker recognition by a variety of submersible systems. In order
to stabilize the on-bottom position, each Biomarker was weighted with a 0.5 kg stainless steel anchor bar
with a swivel that permitted rotation of the marker in the presence of currents and submersible propwash.
Inset shows Biomarker d62 on the deck of the R/» Atlantis II.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
471
high-temperature (392°C) fluids emanating from orifices in bare basalt (with little or
no associated sulfide deposits) to clouds of diffuse flow issuing from fissures, cracks,
and deep pits of collapsed lava. The fresh lava flows and numerous nascent hydrothermal vents were covered with thick microbial material with a notable absence of
megafauna; (3) in March 1992, newly established vent-endemic invertebrates were
found inhabiting these same areas of diffuse venting.
2.2. Transect installation
In March l992, 210 polyethylene markers, referred to as Biomarkers (Fig. 2), were
deployed from Alvin at intervals ranging from &1 to 15 m along the Transect.
Attempts to space the Biomarkers farther apart in less hydrothermally active areas
and closer together in areas of intense venting activity were made along the length of
the Transect (see Fig. 3). The Biomarkers provide benchmarks that unequivocally
locate meter-scale areas in which temporal change can be characterized quantitatively. In order to assess the extent to which Biomarkers moved, we installed a fixed
marker with a 3—5 kg anchor immediately adjacent to five Biomarkers at various
intervals along the Transect. Since the establishment of the Transect, these Biomarkers have not moved with respect to their paired fixed marker. A few Biomarkers
located on the margins of fissures, or on the rims of collapse pits, have shifted position
due to the collapse of lobate lava surfaces.
Biomarkers were deployed and surveyed during Alvin dives 2496, 2499, and 2501.
The location of each Biomarker was determined to within 2—4 m in geodetic coordinates by acoustic ranging from Alvin to a permanently-moored, long-baseline acoustic
navigation system. Transponders were surveyed acoustically using global positioning
system (GPS) data; root mean square errors of transponder positions were 1.1 m.
During dive 2504, Alvin continuously surveyed the Transect by traversing at an
altitude of 1—4 m over the Biomarkers in a north to south direction at a constant
speed of &12 m/min.
2.3. Photographic documentation
Photographic surveys. Photographic surveys were conducted near or within the
Transect area during six cruises between 1989 and 1995 (Fig. 1). Image data collected
between 1989 and 1992 consisted of sporadic 35 mm photographs, color video
recorded on VHS, 8 mm or Hi-8 tape, and black and white video from a wide-angled
silicon-intensified target (SIT) camera. From 1993 to 1995, imaging capabilities
improved with the use of metal-halide lighting (developed by M. Olsson and Deep-Sea
Power and Light) and a high-resolution color 3-chip charged-couple device (CCD)
video system (developed by W. Lange and Woods Hole Oceanographic Institution)
recorded to a broadcast-quality Betacam SP format. The 3-chip CCD camera
(capable of 750 lines of television resolution) was equipped with a zoom lens with
a focal length of 5—47 mm providing abundant, high-quality images that could be
digitally scanned, thereby facilitating accurate reproduction of precise azimuths and
positions of observations made at many sites along the Transect. Scaling of animals
472
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Fig. 3. Map of the northern Transect region between 9°50.285@N and 9°50.345@N (a 120 m distance)
featuring the location of the six vestimentiferan colonies (as they appeared in March 1992) in close
proximity to five high-temperature vents along the central primary eruptive fissure within the Hole-to-Hell
area. Unlabeled lava flow types on the ASC floor are areas where flows are a jumbled mixture of adjacent
flows. Axis tick"1 m.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
473
and features within images was essential to estimating organism size, abundance, and
area occupied on the seafloor. The 3-chip camera was fitted with four, orthogonally
mounted 635 nm red-orange micro-lasers to provide quantitative scale within images
(Tusting and Davis, 1992). Three of the lasers were parallel to the optical axis of the
camera, while the fourth laser was pointed at an angle such that its beam converged
with its opposing laser at exactly 2 m distance in water. Calibration and confirmation
of this laser configuration was performed using a 50 cm2 grid placed 2 m from the
camera lens by Alvin’s manipulator claw at a depth of &200 m as the submersible
descended through the water column. This multi-beam laser scaling array permitted
the determination of object size, as well as the distance from the camera to the object
on the seafloor (Tusting and Davis, 1992; 1993). Additionally, Biomarkers imaged
adjacent to animal communities were used as a check on the accuracy of quantitative
laser measurements.
Enumeration of ‘‘Colonies’’. We refer to discrete vent areas containing developing vestimentiferan populations (and subsequent colonizers) as ‘‘colonies’’ which
are identified by the nearest Biomarker number (e.g. Bio9 and Bio141; Fig. 3). The
3-chip camera was mounted on the Alvin’s basket or on the starboard manipulator
arm and positioned parallel to the submersible’s long axis. At each colony, video
images were collected at a minimum of three pre-determined headings at a variety
of focal lengths. Far-field and vertical-incidence (down-looking) imaging (4—8 m
from the subject plane) also was conducted to determine the area occupied by the
colony, and close-up macro-scale imaging (1—2 m from subject, full zoom) allowed size
and abundance determination of individual species. Three to five video frames
covering at least three different headings at each developing vent colony were
captured digitally using Media Cybernetics Image Pro-Plus software and a Coreco
Oculus-TCX/MX image capture board. From the positions of the laser dots within
each digital image, perspective and vertical grids were constructed and overlaid on the
image. Direct counts of individual organisms and determination of area within the
field of view were obtained using methods previously described by Tusting and Davis
(1993).
The pre-determined submersible headings from which video images were recorded
were selected to minimize topographic relief and reduce the error inherent with the
assumption of a flat seafloor in perspective grid analysis. Additional error in quantitative estimates of seafloor area and vestimentiferan abundance was introduced by the
variation in thickness of individual tube worm assemblages, which increased over
time. Average variability of estimates obtained from images of vestimentiferan colonies ranged from $15 individuals (in colonies under &150 ind.) to $95 individuals
(in colonies exceeding &800 individuals) or 10—12%. For individual size determination, accuracy varied widely, depending on the taxa imaged. Among tube worms, the
frequent inability to adequately observe the point at which the posterior end of the
tube was in contact with substrate prevented accurate determination of tube length in
&90% of the individuals within a given colony. For tube worms for which the entire
tube was visible, estimates varied $3 cm, depending on the oblique angle of the
camera and the angular position of the posterior end. The variability in the diameter
of the laser spot on the seafloor had little effect on quantitative estimates, as spot
474
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
diameter ranged from 0.5 to 1.5 cm, depending upon the focal distance and the
intensity of shimmering vent fluids.
Strip transect surveys. Strip transect surveys (McCormick and Choat, 1987;
Buckland et al., 1993) were conducted using Alvin by traversing over Biomarkers from
north to south at 12 m min~1. The 3-chip camera was positioned at an oblique (&40°)
angle 3—6 m above the seafloor during transect surveys to determine gradients in the
distribution of mobile and sessile megafauna over time. Transect runs were conducted
at the beginning and end of each dive program in December 1993, October 1994, and
November 1995. Scoring of identifiable mobile organisms was performed every 3—7 s
of continuous video. Seafloor area was determined using horizontal and vertical laser
distance every 12—17 s of continuous video data (equating to &3—15 m of linear
seafloor distance), depending on laser visibility. Running averages of seafloor area
were computed and the number of organisms per square meter determined.
Computer-generated maps (Fig. 3) delineating the spatial distribution of biological
and geological features along the length of the Transect were constructed by integrating Biomarker position, submersible navigation, and prominent topographic features
in the ASC during each field program.
»ent fluid chemistry. Diffuse vent fluids were sampled from low-temperature
faunal communities using the NOAA manifold water sampler (Massoth et al., 1988)
equipped with both ‘‘major’’ and ‘‘gas tight’’ titanium water samplers in April
1991, December 1993, October 1994, and November 1995 (Von Damm et al., 1994;
Von Damm et al., 1996; Lilley et al., 1996). Sampling locations within developing
communities were within &1.5 m in successive years. During field programs in 1993,
1994, and 1995, two to ten discrete water samples were obtained from within these
communities and subsequently analyzed using methods described in Von Damm et al.
(1997) to determine the chemical composition and concentration of numerous chemical species including hydrogen sulfide, iron, carbon dioxide, methane, magnesium,
chloride, and silica. In each case, the manifold intake wand was carefully positioned at
the base of vestimentiferan tubes. Extreme care was taken during water sampling to
ensure minimal disturbance to the surrounding biological communities. Temperatures of diffuse fluids were monitored during each field program using the Alvin
temperature probe and manifold intake wand, and between field programs since
December 1993 using self-recording, time-lapse temperature (HOBOTM) probes
(Fornari et al., 1994; Shank et al., 1995).
3. Results
3.1. Geologic and hydrothermal setting
The physical characteristics of the ASC between 9°09@N and 9°54@N and relationships to hydrothermal vent distribution, faults, and fissures are described by Fornari
et al. (1990), Haymon et al. (1991, 1993), Fornari and Embley (1995), Wright et al.,
(1995), and Fornari et al. (1998). In plan-view the ASC outline is sinuous (Figs. 1 and 3),
and its width in the 9°49@—51@N region varies between 40 m, near 9°50.8@N, and 110 m,
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
475
just north of 9°49.0@N. The height of the ASC walls within the Transect area varies
from &5—8 m. In places, the wall consists of irregular stair-steps with one or two
narrow (1—3 m wide) benches formed by apparent still-stands of lava that represent
eruptions which partially filled the ASC locally. Lava pillars (2—8 m high) commonly
support archways of uncollapsed roof sections of ponded lava flows along the margins
and the floor of the ASC (Fig. 4a and b).
Between 9°49@ and 9°51@N, the primary eruptive fissure inferred to have fed the 1991
eruption (Haymon et al., 1993; Gregg et al., 1996) can be followed nearly continuously
over distances of hundreds of meters. This fissure often comprises an anastomosing
single crack &1—3 m wide and &2—5 m deep, which cuts across hackly, sheet and
ponded flows in the ASC floor (Fig. 4c). The location of the fissure within the ASC is
nearly coincident with the line of Biomarkers (Fig. 1). Based on field relationships
observed between April 1991 and November 1995, considerable post-volcanic modification has influenced the current plan-view geometry of the fissure. This modification
is principally manifested in the variable collapse of lava crusts along the fissure
margin, short (&5—20 m) offsets of the primary fissure (e.g. at Bio9 vent, Fig. 3), and
occasional short parallel cracks or elongate collapse depressions. A region of extensively collapsed lobate lavas was also present along the length of the Transect in
a zone &50—200 m wide on the ASC margin (Fig. 4b). This large-scale porosity and
permeability offers great potential for circulating hydrothermal fluids and harboring
subsurface microbial communities.
The Transect was divided into eight regions in order to characterize the distribution
of biological communities, hydrothermal setting, and lava morphology (Table 1).
Nascent hydrothermal venting and incipient biological colonization occurred
between Biomarkers d1 and d160. This &1.02 km region within the Transect
was chosen as the primary area for studying vent community development and has
been termed the ‘‘Phoenix’’ vent site. Between Biomarkers d160 and d210, there
were two areas in which extensive communities existed prior to the 1991 eruption:
a basalt/ sulfide edifice known as ‘‘Tube Worm Pillar’’ (an 11 m high structure covered
by an extremely large population of Riftia pachyptila), and a &20]50 m bed of
mytilid bivalves (Bathymodilus thermophilus) and serpulid polychaetes (¸aminatubus
alvini and Protis hydrothermica). The proximity of both nascent and previously
established biological communities in the Transect provided the ability to monitor
changes in early and mature stages of community development in spatially proximal
areas.
Within the Transect there were four areas (each less than 40 m2) of high-temperature venting ('350°C). High-temperature vents activated as a result of the 1991
eruption are near Biomarker d9 (Bio9 Prime vent and Bio9 vent) and Biomarker
d35 (P vent, P Middle vent, and Bucket-lid 8 vent) (Fig. 3). Two areas of focused
flow, present prior to the eruption, that continue to be active are: (1) the Tube
Worm Pillar (near Biomarker d188); and (2) a spindle-shaped chimney called
‘‘Damocles Sword’’ present at the east margin of the ASC near Biomarker d72. A few
individuals of Riftia and Bathymodiolus thermophilus were present at the base of
Damocles Sword in October 1991 (Chevaldonné et al., 1995). Time-series data of
vent fluid temperature for each high-temperature vent in the 9°49@N—51@N area are
476
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
presented and discussed by Oosting and Von Damm (1996). Faunal colonization
patterns and development of high-temperature chimneys within the Transect are
discussed in Shank et al. (in prep.).
3.2. Low-temperature vent colonization (1989—1991)
Images of the ASC floor taken using Argo in 1989 between 9°49@N and
9°51@N revealed the presence of well-developed animal communities similar to those
encountered along the Galápagos Rift in 1985. Extensive beds of bathymodiolid
mussels, serpulid worms, galatheid crabs, vesicomyid clams, and swarms of amphipods were observed on the margins of the ASC. The presence of only sparse
populations of Riftia pachyptila, in a predominantly recumbent posture, and the
biogeographic dominance of bathymodiolid mussels between 9°46@ and 9°52@N in
1989 suggests that these vent assemblages were in the later stages of vent community
development.
During the 1991 eruption, fresh lava flows over-ran several existing communities
(e.g. Tube Worm Barbecue; Fig. 1) and draped small, recently toppled chimneys
(Haymon et al., 1993). At that time, high-temperature venting through bare basaltic
lava and widespread unorganized diffuse flow emanating through a network of cracks,
holes, deep pits, and 1—5 m wide fissures in the seafloor between 9°49.6@N and 9°50.4@N
(Figs. 5a and 6a) also were observed (Haymon et al., 1993). White filamentous
microbial mats, &1—10 cm-thick, blanketed up to 50 m2 areas of fresh lava flows
surrounding nascent venting areas. These mats were composed of a single type of
bacterial ‘‘sheath-like’’ filament covered with particles of silica and arranged into
‘‘brittle flakes and flexible sheets’’ (Nelson, 1991; Haymon et al., 1993). Biogenic
particles from these highly structured stratums were swept upwards from the seafloor
by strong hydrothermal flow creating ‘‘snowstorms’’ that reached 50 m above the
ASC floor (Haymon et al., 1993). In April 1991, diffuse fluid temperatures within the
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&c
Fig. 4. Video images of the principal geological features on the axial summit caldera (ASC) floor in
December 1993. (a) The surface of a 1991 lava flow that ponded near the east wall (near Biomarker d110) of
the ASC. The lava pillars which support the lobate crust are &2 m high. The extensive plates of talus in the
foreground of the image are collapsed pieces of lobate crust that were once part of the flow surface. The
4—5 cm thickness of the lobate crust indicates that the 1991 flow remained ponded for &4 h prior to the
waning of the eruption (Gregg et al., 1996). Distance across the bottom of the image is &4 m, and the view
is to the southeast (167°). (b) The section of the axial summit caldera floor near Biomarker d175 (middle of
image). Large feature in upper left is a &4 m tall by &5 m wide lava remnant created by two large lava
pillars that nearly coalesced during the 1991 eruption. The ‘‘hole’’ in the middle served as a principal surface
conduit for lateral lava movement within the axial summit caldera. The foreground is covered by platey
collapse talus on a low-relief (&1 m) pressure ridge in the ASC floor. Distance across the bottom of the
image is &2.5 m, and the view is to the south (173°). (c) The western margin of the primary eruptive fissure
(from lower left to upper-middle of image) that fed the 1991 eruption. Clouds of ‘‘smoky’’ diffuse flow issue
from the fissure (at left). The terrain in this video image is just south of the Bio141 and Bio142
vestimentiferan colonies (Biomarkers d143 and d144 are visible). Brachyuran and galatheid (at base of
lava pillar) crabs are visible among narrow basaltic cracks lined with bacteria. The lava pillar in middle left
of image is &4 m tall, and the view is to the south-southeast (&158°).
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
477
478
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
479
Transect ranged from 22°C (Hole-to-Hell area) to 29°C (Biomarker d141 area) to
55°C (Y Vent at the base of the Tube Worm Pillar, Fig. 1) in comparison to 2.1°C for
ambient bottom water. Corresponding iron (relative to chloride) and hydrogen sulfide
concentrations were unusually high (0.15 mmol kg~1 and 0.15—8.5 mmol kg~1, respectively) in these areas. Sessile vent-endemic megafauna were not observed in any
newly-formed vent openings.
3.3. Low-temperature vent colonization (1992—1995)
The sequential pattern of colonization exhibited by the dominant megafaunal
species was remarkably similar within all colonies (including those spatially separated) throughout the Transect (Table 2, and Figs. 5 and 6). In every area of persistent
venting, bacteria-dominated vent openings activated during the eruption were rapidly
colonized and dominated by vestimentiferans; first by ¹evnia jerichonana (within 11
months) and then by Riftia pachyptila (within 32 months). For convenience, these
species will be referred to as ¹evnia and Riftia, respectively (both are members of
monospecific genera). The vent-endemic mussel Bathymodiolus thermophilus was
initially observed adjacent (0.5—6 m distant) to these vestimentiferan colonies 10 months
after the establishment of Riftia, and affixed to tubes of Riftia 13 months later. During
the course of our studies we did not observe successful colonization by vesicomyid
clams within the Transect area. Mullineaux et al. (1998) reports finding clams (2 ind.)
identified as Calyptogena magnifica (of undetermined size) on basaltic settlement
panels deployed near the base of P vent from December 1991 to November 1994. We
observed no small specimens of clams despite intensive searching in November 1995
utilizing a macro-video camera capable of resolving 1 cm of seafloor at a camera
distance of 2 m. It is conceivable, however, that vesicomyid clams occupied microhabitats not readily discernible by macro-video (e.g. the undersurface of lobate lava
crust). The observed sequence of faunal colonization (as illustrated by the Biomarker
b&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Fig. 5. Temporal sequence of faunal community development at Biomarker d9 (Bio9) located along
a short offset of the primary eruptive fissure in the northern Hole-to-Hell area. Arrow in c, d, and e indicates
comparable corresponding location on the seafloor in successive years. (a) No vent megafaunal organisms
were present when vigorous hydrothermal activity was initiated, and profuse microbial material expelled
from a fissure within 15 m of the subsequent location of Biomarker d9, view is to the southeast. (b) By
March 1992, the fissure was colonized by an extensive population of ¹evnia jerichonana next to Biomarker
d9, with Riftia pachyptila noticeably absent (view is to the northeast). (c) By December 1993, numerous
Riftia pachyptila had settled and rapidly grown to form a dense thicket, engulfing the existing ¹. jerichonana
(note Biomarker d16 at left; view is to the east, 087°). (d) Between December 1993 and October 1994,
continued rapid colonization and growth by large numbers of Riftia pachyptila dramatically increased the
density, lateral extent, and vertical growth within this colony (view is to the east, 088°). (e) By November
1995, the density and lateral extent of the Bio9 colony continued to increase. The colony was now occupied
by over &2000 R. pachyptila. About half of the ¹. jerichonana assemblage documented in 1992 was alive in
1995 at the base of the R. pachyptila tubes (¹. jerichonana not visible in this image; view is to the east, 091°).
The staining of worm tubes with a rust-colored, ferrous-oxide precipitate was coincident with increased
concentrations of iron in the diffuse vent fluids (maximum temperature recorded was 33°C) supporting this
colony.
480
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
d141 site, Fig. 7) and the coincident variability in vent emission and fluid chemistry
are described below.
3.3.1. March 1992
»ent emissions. The reduction of widespread vent emissions between April 1991
and March 1992 formed discrete venting areas from &5 to 100 m2. Thick clouds of
diffuse vent flow and shimmering water exited from the primary eruptive fissure and
collapsed lava pits. A marked decline in the thickness and areal extent of microbial
mats reflected the spatial reduction and organization of venting over this same time
period. Filamentous mats were restricted to circular patches (&1 to 10 m2) around 17
areas of diffuse flow. Samples from several vent-associated microbial mats 300 m
north of the Transect area contained extremely high abundances of siphonostomatoid
copepods and pardaliscid amphipods. This high abundance prevented the accurate
collection of microbial counts necessary to assess levels of microbial production
(D. Nelson, pers. comm., 1993).
Megafaunal colonization. Within 11 months of the eruption, ¹evnia colonized 17
diffuse venting areas distributed over the 900 m distance separating Biomarkers d8
and d142. These vestimentiferan-dominated areas were in discrete patches (&1 to
4 m2) separated from each other by distances of 2 to 340 m. ¹evnia abundance within
colonies ranged from &30 to over 200 individuals in March 1992 (Table 2). All
Transect colonies were situated on the sides of the eruptive fissures or adjacent to
these fissures along numerous small cracks (coated with white microbial mats) in
the ASC floor. Associated with these ¹evnia colonies were several species of
limpets (e.g. ¸epetodrilus elevatus and ¸. pustulosus), which densely covered
(100—500 ind. m~2) the neighboring seafloor. These limpets were prominent within
circular patches conspicuously devoid of microbial mat suggesting that these limpets
were actively grazing within these mats. Brachyuran crabs (Bythograea thermydron)
were also present on the adjacent seafloor (&10 ind. per colony), and were infrequently observed crawling among ¹evnia tubes. Present on the dorsal surface of several B.
thermydron were numerous actively ‘‘hopping’’ siphonostomatoid copepods (Aphotopontius acanthinus; Humes and Lutz, 1994). Large numbers of amphipods (Halice
hesmonectes, Martin et al., 1993; Kaartvedt et al., 1994; Van Dover et al., 1992)
swarmed &1 to 3 m above &35% (6 of 17) of the ¹evnia colonies (Table 2).
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&c
Fig. 6. Temporal sequence of faunal community development at Biomarker d119 (Bio119). (a) Microbial
material and cloudy shimmering water dominated the area adjacent to the central fissure in April 1991
(view to the north, &007°). (b) By March 1992, a small number of ¹evnia jerichonana were observed within
this fissure (Biomarker d119 at top right; view to the south, &176°). (c) By December 1993, increasing
numbers ¹. jerichonana, as well as several newly established Riftia pachyptila, were present. Zoarcid fish
were also abundant (view to the north, &352°). (d) Between December 1993 and October 1994,
¹. jerichonana and R. pachyptila abundance and coverage continued to increase along the fissure,
enveloping almost half of Biomarker d119 (view is to the north, 359°). (e) By November 1995, the
R. pachyptila colony increased by &50 individuals, while small patches containing dead ¹. jerichonana
were evident (lower right). Video image (e) was frame-captured during a spawning event by R. pachyptila.
A spawn ‘‘cloud’’ is visible mid-way above the Biomarker d119 number and the top edge of the image.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
481
482
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
483
Galatheid crabs (Munidopsis subsquamosa) and nematocarcinid shrimp (Nematocarcinus ensifer?) dominated peripheral areas (&4—40 m from vent openings). The highest
densities of nematocarcinids were invariably 5—20 m south of each venting area in
densities (up to 0.5 m~2) greater than those of galatheids.
Fluid chemistry. No low-temperature vents were sampled in March 1992, because
the manifold sampler was not available for diffuse fluid sampling.
3.3.2. December 1993
»ent emissions. Spatial variability in local-scale, diffuse vent flow was evident in
1993 as venting continued to be more spatially discrete (actually ceasing in some
areas) along the primary eruptive fissure. Specifically, flow was: (1) waning around the
southern base of P Vent (Fig. 3); (2) completely absent between Biomarkers d40 and
d81; and (3) severely reduced between Biomarkers d84 and d112. ¹evnia and Riftia
subjected to the cessation of vent flow were either dead (e.g. at Bio63) or dying (e.g. at
Bio36 and Bio89, Table 2). Vent flow from a pit just south of P vent decreased
markedly and the population of live ¹evnia within this pit (Bio36 colony) diminished
by &50%. This colony also contained the fewest number of Riftia (&20 ind.) within
active sites along the Transect in December 1993. Unlike the slowly dying Bio36 and
Bio89 colonies, rapid termination of vent flow in the Bio63 region by December 1993
coincided with the death of the Bio63 colony, and presumably preempted the successful establishment of Riftia. Large numbers of colonial siphonophores (¹hermopalia
taraxaca; up to 7 ind. m~2) moved via their numerous long filamentous tentacles
among worm tubes in these dead and dying areas (Fig. 8d). Increased densities of
galatheid crabs also were associated with these moribund colonies.
Megafaunal colonization. By December 1993, 32 months after the 1991 eruption,
extensive microbial mats were no longer present within the Transect. Small dense
patches of microbial mat were limited to the surfaces of high-temperature chimneys.
Surviving vestimentiferan colonies within the Transect region were dominated by
Riftia (20 to over 600 ind. per colony). Linear growth rates exhibited by 1.5 m tall
individuals of Riftia were &85 cm/yr (increase in tube length between March 1992
and December 1993) (e.g. Fig. 5c). These rates are the fastest reported to date for any
species of vestimentiferan tube worm (Lutz et al., 1994). Extensive colonies of Riftia
were present in each of the pre-existing ¹evnia colonies (e.g. Bio8, Bio119 and Bio141,
b&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Fig. 7. Maps depicting the temporal and spatial sequence of faunal change within the Biomarker d141
area, including the spatial distribution of Biomarkers, the aerial coverage of growing faunal assemblages,
and the distribution of brachyuran and galatheid crabs in March 1992, December 1993, October 1994 and
November 1995. The Biomarker d141 area features a broadening, well-defined, bacterial-stained fissure
that was dominated by ¹evnia jerichonana in 1992 and by Riftia pachyptila in 1993, 1994, and 1995. Mussels
were first observed colonizing cracks and crevices surrounding the Bio141 colony in October 1994, and
increased in abundance between October 1994 and November 1995. By November 1995, a few individual
mussels were directly attached to R. pachyptila worm tubes. The areal coverage of serpulid polychaetes
expanded over time as they encroached toward vent openings, particularly between October 1994 and
November 1995. Large swarms of amphipods (Halice hesmonectes) were associated with vesitmentiferan
colonies in each year. Axis ticks"1 m.
Biomarker
Distance
(m)
Geological
Regional characteristics
hydrothermal
Biological
Northern
Terminus
1—7
40
2—7 m wide central fissure;
glassy, lobate, ropy lava to
sheet flow surfaces
Vigorous diffuse venting from base of
fissure from Biomarker d5 south into
Hole-to-Hell
Brachyuran crabs and nematocarcinid
shrimp dominated fissure margin from
1992 to 1995
‘‘Hole-to-Hell’’
8—39
70
5—10 m diameter collapse pits
on the margin of a 1—5 m wide
fissure; ropy, hackly and sheet
flow lava surfaces
High-temperature vents [Bio9 Prime
and Bio9 (368°C venting from orifices
in bare basalt in 1991) located in the
north and P vent (368°C in 1991), P
Middle, and Bucket-lid 8 vent in the
south]; diffuse flow throughout the
area in 1991 and 1992; by 1995, diffuse
flow had increased concentrations of
iron relative to hydrogen sulfide in the
northern end, and diffuse vent flow
had ceased in the southern end
Thick microbiol mats in 1991; in 1992,
¹evnia jerichonana colonies near
Biomarkers d8, d9, d12, d13, d29,
and d36 developed along the central
fissure; by 1995, vent fluids had very
low H S and high iron concentrations
2
within dying vestimentiferan colonies
near Biomarkers d12-d13 and flow
had ceased within dead colonies near
Biomarkers d29-d36
South of
‘‘Hole-to-Hell’’
40—99
400
2—10 m wide continuous fissure
flanked by broad intermittent
collapse with fresh glassy plates
and broken sheet lava
Vigorous diffuse flow was associated
with tube worm colonies in 1992 (29°C
in 1991); by 1993, cessation of vent
flow occurred north and south of
Biomarker d82:; high-temperature
(339°C in 1991) vent (Damaclese
Sword) near Biomarker d72
¹evnia jerichonana colonies (Bio 61,
Bio62, Bio63, Bio65, Bio66, Bio77,
Bio82, and Bio89) developed by 1992.
By 1993, Bio61, Bio63, Bio77, and
Bio89 colonies were dead with little or
no Riftia pachyptila colonization, while
diversity and biomass increased within
the Bio82 colony from 1992 to 1995
Central
Transect
100—112 150
Low relief hackly and ropy lava
flows with abundant collapse
and lava pillar remnants
No diffuse or high-temperature
venting was observed in this region
No vent-endemic megafauna were
observed in this region; fauna present
were nematocarcinid shrimp,
synaphobranchid and macrouid fish
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Transect
region
484
Table 1
Summary of the regional geological, hydrothermal, and biological characteristics of the Biologic—Geologic Transect. Marked changes in hydrothermal emissions
and coincident changes in vent community structure, including the death of faunal assemblages, are featured. The &17 nascent vent areas, comprising the
1.02 km-long Phoenix area, are in the northern area between Biomarkers d1 and d160. Pre-eruption communities occupy a 40 m long section on the ASC floor
just north of the Southern Terminus. The Tube Worm Pillar marks the southern limit of observations made after March 1992
113—160 360
2—4 m wide central fissure
bordered by sheet, hackly, and
ropy lavas; fissure broadens
near Biomarker d141
Intense diffuse bacterial ‘‘snowblower’’type venting (23°C) from the fissure
adjacent to Biomarker d141 in 1991;
fissure from Biomarker d116 to d120
and Biomarker d137 to d143 still
vigorously venting diffuse flow in 1995
(27°C)
Thick microbial mats in 1991 were
reduced to 1—2 m wide patches by 1992;
the region between Biomarkers d117 to
d121 was covered by serpulids and
colonized by patches of mussels between
1994 and 1995; Bio119, Bio141, and
Bio142 vestimentiferan colonies thrived
from 1992 to 1995
Southern
Transect
161—180 250
Highly chaotic collapse region;
fissure not well-defined;
numerous areas of ‘‘contact’’
where older sediment-dusted
lobates are partially buried by
younger glassy lava flows
No diffuse or high-temperature flow
was observed in this region
No sessile megafauna observed in this
region; area dominated by
nematocarcinid shrimp and relatively
few galatheid crabs
Tube Worm
Pillar
181—192
40
Lobate lava fields with extensive
collapse; no definable fissure in
the area; mussel bed on top of
lobate lava remnants on western
ASC rim; an 11 m tall lava/
sulfide pillar (‘‘Tube Worm
Pillar’’.) is situated on a circular
pedestal in a 30—40 m wide and
8—10 m deep collapse pit
No visible flow ever observed in the
mussel bed; low-temperature fluid
(55°C) exiting from a broken lava
pillar (Y vent) at the base of the Tube
Worm Pillar exhibited a high H S
2
concentration, !9 mmole/kg, in 1991
which decreased to &0.55 mmole/kg
in 24°C fluids by 1995
Two communities existed prior to the
1991 eruption; the Biomarker d182
mussel-serpulid bed and the ‘‘Tube
Worm Pillar’’; mussels (no clams
observed) appeared alive in 1992, but by
1995 all appeared dead; numerous
galatheid crabs in the mussel bed in
1992, gastropods abundant in 1993; the
Tube Worm Pillar densely inhabited by
Riftia pachyptila since at least 1989,
mobile megafauna increasing since
1991, brachyuran crabs and zoarcid fish
commonly observed from 1992 to 1995
Southern
Terminus
193—210
60
Series of chaotic collapse pits
and lava remnants
Intense, smokey diffuse flow emanating
from deep pits in 1992; spatial
reduction of diffuse vent flow evident
between 1992 and 1995
Brachyuran crabs were the only fauna
visible due to intense water turbidity in
1992; several vestimentiferan colonies
developed in this area between 1993 and
1995, but documentation is limited
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Biomarker
d141
485
Species/Feature
Bio8
91
Bacterial mats
Bythograea thermydron
¹hermarces andersoni
¸epetodrilus spp.
¹evnia jerichonana
Riftia pachyptia
Amphipod spp.
Alvinocaris lusca
Paralvinella grasslei
¹hermopalia taraxaca
Munidopsis subsquamosa
Halice hesmonectes swarm
Bathymodiolus thermophilus
Serpulid polychaetes (per m2)
Cyanathea hydrothermala
R. pachyptila spawning event
Temperature (max. °C)
Methane
Carbon dioxide
Iron
Hydrogen sulfide
'20 m2
10
1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Bio9
92
93
94
(5 m2
10
1
100
200
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
10
(5
150
100
600
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
20
(5
200
50
1100
30
—
—
—
—
—
(5
—
—
—
—
—
—
—
—
95
91
—
30
(10
'300
50
2000
50
—
—
—
1
—
(15
—
2
—
—
—
—
—
—
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
22—
—
—
—
92
93
(5 m2
10
(5
100
150
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
10
(5
200
100
400
30
(5
—
—
—
—
—
—
—
—
—
—
—
—
—
94
95
—
10
(5
350
50
800
30
(5
—
—
—
—
—
—
—
—
32
0.004
8.4
0.064
0.80
—
20
(5
'400
50
2000
90
(5
—
—
—
—
—
—
—
—
33
0.005
8.8
0.240
0.194
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Transect Colony
486
Table 2
Appearance and average abundance of species, maximum diffuse fluid chemical concentrations (in mmoles kg~1), and other features observed at select nascent
vent areas in which vestimentiferan-dominated colonies formed within the Biologic—Geologic Transect. Colonies similar in abundance to their nearest
neighboring colony are not shown here, but include Bio28, Bio59, Bio61, Bio66, Bio77, and Bio99. Observations were made within 2 weeks (April 1991), 11
months (March 1992), 32 months (December 1993), 42 months (October 1994) and 55 months (November 1995) from the initiation of venting activity. Abundance
estimates include &1 m2 of seafloor surrounding each colony (*"vestimentiferan colony on fissure margin only; -"sample taken in the vicinity; ‡"limit of
detection; —"not observed (despite detection efforts) or sampled; ( )"empty tubes, animals considered dead; ND " temperature anomaly not detected)
Table 2. Continued
Transect Colony
Species/Feature
Bio12
Bacterial mats
Bythograea thermydron
¹hermarces andersoni
¸epetodrilus spp.
¹evnia jerichonana
Riftia pachyptila
Amphipod spp.
Alvinocaris lusca
Paralvinella grasslei
¹hermopalia taraxaca
Munidopsis subsquamosa
Halice hesmonectes swarm
Bathymodiolus thermophilus
Serpulid polychaetes (per m2)
Cyanathea hydrothermala
R. pachyptila spawning event
Temperature (max. °C)
Methane
Carbon dioxide
Iron
Hydrogen sulfide
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
(5 m2
10
(5
100
30
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
93
—
10
(5
200
30
200
20
—
—
—
—
1
—
—
—
—
—
—
—
—
—
94
—
10
(5
200
30
250
20
—
—
—
1
1
(5
—
—
—
—
—
—
—
—
95
—
5
(5
(40
(30)
10(250)
10
—
—
—
3
—
—
—
—
—
29
0.0003
5.2
0.28
)0.015‡
91
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
(5 m2
10
5
200
50
—
—
—
—
—
—
1
—
—
—
—
—
—
—
—
—
93
94
95
—
10
5
200
50
200
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
10
5
200
50
200
—
—
—
—
—
—
(5
—
—
—
—
—
—
—
—
—
10
5
—
(50)
25(200)
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
91
Bio13
487
488
Table 2. Continued
Transect Colony
Species/Feature
Bio36
Bacterial mats
Bythograea thermydron
¹hermarces andersoni
¸epetodrilus spp.
¹evnia jerichonana
Riftia pachyptila
Amphipod spp.
Alvinocaris lusca
Paralvinella grasslei
¹hermopalia taraxaca
Munidopsis subsquamosa
Halice hesmonectes swarm
Bathymodiolus thermophilus
Serpulid polychaetes (per m2)
Cyanathea hydrothermala
R. pachyptila spawning event
Temperature (max. °C)
Methane
Carbon dioxide
Iron
Hydrogen sulfide
'20 m2
—
1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
'5 m2
10
—
—
150
—
—
—
—
—
—
1
—
—
—
—
—
—
—
—
—
93
94
95
—
10
—
—
100(100)
20
—
(5
—
—
—
—
—
—
—
—
—
—
—
—
—
—
10
5
—
(150)
(20)
—
(5
—
—
(5
—
—
—
—
—
—
—
—
—
—
—
10
5
—
(150)
(20)
—
(5
—
—
(5
—
—
—
—
—
—
—
—
—
—
91
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
'5 m2
10
—
'500
150
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
93
—
—
—
—
(200)
—
—
—
—
(20
(5
—
—
—
—
—
—
—
—
—
—
94
95
—
—
—
—
(200)
—
—
—
—
—
(5
—
—
—
—
—
ND
—
—
—
—
—
—
—
—
(200)
—
—
—
—
—
—
—
—
—
—
—
9
—
—
—
—
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
91
Bio63
Table 2. Continued
Transect Colony
Species/Feature
Bio82
Bacterial mats
Bythograea thermydron
¹hermarces andersoni
¸epetodrilus spp.
¹evnia jerichonana
Riftia pachyptila
Amphipod spp.
Alvinocaris lusca
Paralvinella grasslei
¹hermopalia taraxaca
Munidopsis subsquamosa
Halice hesmonectes swarm
Bathymodiolus thermophilus
Serpulid polychaetes (per m2)
Cyanathea hydrothermala
R. pachyptila spawning event
Temperature (max. °C)
Methane
Carbon dioxide
Iron
Hydrogen sulfide
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
29—
—
—
1.01s
92
93
'5 m2
—
10
10
—
5
'200
'400
200
200
—
600
—
30
—
—
—
—
—
—
—
—
1
1
—
—
—
—
—
—
—
1
—
16
—
0.002
—
5.6
—
0.024
—
—
94
—
20
10
'600
200(50)
800
30
—
—
—
—
1
10
—
—
1
20
0.003
6.2
0.003
0.4
95
—
30
15
'2000
250(50)
'2000
50
(10
—
—
—
1
50
—
—
1
18
0.002
5.8
0.011
0.263
91
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
'5 m2
10
—
'500
150
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
93
94
95
—
20
5
200
500(500)
200
—
—
—
(5
—
—
—
—
—
—
—
—
—
—
—
—
10
5
—
(1000)
(200)
—
—
—
—
(5
—
—
—
—
—
ND
—
—
—
—
—
10
5
—
(1000)
(200)
—
—
—
—
(5
—
—
—
—
—
ND
—
—
—
—
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
91
Bio89
489
490
Table 2. Continued
Transect Colony
Species/Feature
Bio119
Bacterial mats
Bythograea thermydron
¹hermarces andersoni
¸epetodrilus spp.
¹evnia jerichonana
Riftia pachyptila
Amphipod spp.
Alvinocaris lusca
Paralvinella grasslei
¹hermopalia taraxaca
Munidopsis subsquamosa
Halice hesmonectes swarm
Bathymodiolus thermophilus
Serpulid polychaetes (per m2)
Cyanathea hydrothermala
R. pachyptila spawning event
Temperature (max. °C)
Methane
Carbon dioxide
Iron
Hydrogen sulfide
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
93
'5 m2
—
10
10
5
15
'200
'2000
50
150
—
50
—
30
—
—
—
—
—
—
—
—
1
1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
94
—
50
10
'2000
500
150
30
—
—
—
—
1
5
—
—
—
—
—
—
—
—
95
—
150
15
'2000
1000(250)
200(25)
50
(5
—
—
—
1
(30
(50
—
1
—
—
—
—
—
91
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
23 —
—
0.004 1.9 -
92
93
'5 m2
—
10
10
—
10
'200
'400
100
200
—
400
—
30
—
—
—
'20
—
—
—
—
1
1
—
—
—
—
—
—
—
2
—
26
—
0.002
—
9.4
—
0.021
—
0.944
94
95
—
20
30
'600
250
800
35
—
'20
—
—
1
(10
—
—
1
29
0.009
6.7
0.012
0.88
—
30
35
'2000
250
'2000
60
—
'20
—
—
1#
(15
(50
—
1
27
0.008
8.7
0.012
0.725
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
91
Bio141*
Table 2. Continued
Transect Colony
Species/Feature
Bio142
Bacterial mats
Bythograea thermydron
¹hermarces andersoni
¸epetodrilus spp.
¹evnia jerichonana
Riftia pachyptila
Amphipod spp.
Alvinocaris lusca
Paralvinella grasslei
¹hermopalia taraxaca
Munidopsis subsquamosa
Halice hesmonectes swarm
Bathymodiolus thermophilus
Serpulid polychaetes per m2)
Cyanathea hydrothermala
R. pachyptila spawning event
Temperature (max. °C)
Methane
Carbon dioxide
Iron
Hydrogen sulfide
'20 m2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
92
'5 m2
10
—
—
50
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
93
94
95
—
10
10
'200
50
400
—
—
'20
—
—
—
—
—
—
1
35
0.030
11.6
0.028
0.979
—
10
35
'400
50
600
—
—
'20
—
—
1
(15
—
—
—
28
0.004
2.7
0.005
0.68
—
10
25
'400
50
800
—
—
'20
—
1
1#
(75
(10
—
1
18
0.002
4.8
0.012
0.301
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
91
491
492
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Figs. 5 and 6). Individuals of Riftia reached maturity in less than 20 months as
evidenced by the documentation of four separate spawning events (a rapid ejection of
a cloudy material released from a single point below the obturaculum) within three
distinct colonies during 12 dives in December 1993 (Table 2).
The characteristic ‘‘haystack’’ morphology where Riftia tubes centrally located
within a colony were at least twice as tall as those inhabiting the periphery [described
by Hessler and Smithey (1983) and Hessler et al. (1985, 1988)], was displayed by Riftia
assemblages containing more than nine individuals. The length of these Riftia tubes
was proportional to the visually estimated intensity of vent emissions, suggesting that
Riftia growth was optimized for proximal venting conditions. Individuals of Riftia did
not appear to be limited to areas previously colonized by ¹evnia populations.
Seemingly sporadic, opportunistic colonization by Riftia was apparent as tufts comprised of 2—35 individuals, frequently colonized lava pillars, cracks, and areas between
lobate lava surfaces from which shimmering vent fluids emanated.
Faunal diversity, as determined from video analyses, increased from at least
5 species [bythograeid crabs, lepetodrilid limpets (perhaps 3 species), siphonostomatoid copepods, zoarcid fish, tevniid tube worms] in March 1992, to at least 11
species (additionally, bresiliid shrimp, alvinellid polychaetes, nebaliid leptostracans,
pardaliscid and lysianassid amphipods, and riftiid tube worms) in December 1993
(Table 2). In December 1993, the bresiliid shrimp Alvinocaris lusca was observed
among tubes of Riftia only at the extreme northern (Bio9, 5 ind.) and southern ends
(Tube Worm Pillar, 2 ind.) of the Transect. The polychaetous annelid Paralvinella
grasslei (present at the base of Riftia tubes, Fig. 8a) was seen in only the Bio141 and
Bio142 colonies, and never observed in any other Riftia colonies in the Transect area.
Fluid chemistry. Maximum diffuse fluid temperatures within the central and southern colonies varied from 16°C to 35°C in December 1993. Maximum hydrogen sulfide
concentrations declined from 1.9 mmol kg~1 in April 1991 to 0.979 mmol kg~1
in December 1993 (Table 2). Maximum diffuse fluid iron concentrations within
vestimentiferan colonies declined more sharply from 0.151 mmol kg~1 in April 1991
to 0.024 mmol kg~1 in December 1993. Within these same colonies, methane concentrations varied from 0.002 to 0.030 mmol kg~1 and carbon dioxide concentrations
ranged from 5.6 to 11.6 mmol kg~1 (Table 2).
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&c
Fig. 8. Continued
individuals in this image are less than &20 cm in length). (d) By December 1993, venting emissions had
completely ceased in the Bio63 colony and no living vestimentiferans (less than 5 Riftia pachyptila had
colonized this area) were observed (view is to the southwest, 192°; distance across the bottom edge of image
is &3 m). The only living megafuanal organisms among the worm tubes were numerous rhodalid
siphonophores (¹hermopalia taraxaca). ¹. taraxaca moved amongst the worm tubes and over the adjacent
basalt via their long filamentous tentacles. (e) In December 1993 and October 1994 the Bio12 colony was
a Riftia pachyptila assemblage (with associated dense amphipods) typical of those within the Transect at
these times. However by November 1995 (shown here), the entire colony and much of the surrounding
seafloor was covered with a rust-colored iron-oxide precipitate, and over 95% of the R. pachyptila were
dead, despite the extensive schlieren of 29°C fluids throughout the moribund colony. (f ) Amphipod swarms
reached up to 18 m3 above R. pachyptila colonies as shown here above the Bio141 colony in November
1995.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
493
Fig. 8. Video images illustrating characteristics of community structure within developing and dying
colonies and their associated abiotic habitat. (a) Paralvinella grasslei was observed between the basal
portion of Riftia pachyptila tubes within the Bio141 colony (four individuals shown here; view is toward the
east in 1993; distance across the lower edge of the image is &25 cm; note small R. pachyptila at lower left
and the presence of highly variable lepetodrilid limpet abundance on certain worm tubes). This was the only
vestimentiferan colony in which P. grasslei was observed. (b) Within several vestientiferan colonies, ¹evnia
jerichonana displayed gregarious settlement and colonization as indicated by numerous small individuals
observed here (arrow) in December 1993 near Biomarker d141 (view is to the northeast). (c) The first
documented occurrence of a glalatheid crab directly on the surface of a Riftia pachyptila tube (at right) took
place in November 1995, 55 months after the April 1991 eruption. Mussel (Bathymodiolus thermophilus)
densities observed in this image were typical of the densities observed adjacent to the Bio142 colony in
November 1995 (note byssal threads on yellow-brown mussel shells; view is to the southeast; ¹. jerichonana
494
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
495
3.3.3. October 1994
»enting emissions. The only evident variation in vent flux between December 1993
and October 1994 was the cessation of diffuse flow in areas already waning in
December 1993 (e.g. near P vent, Biomarker d84 and d112). All ¹evnia and Riftia
living proximal to these areas were dead by October 1994 (e.g. Bio89, Table 2).
Microbial mats were no longer observed within the Transect area. However, hyperthermophilic methanotrophic bacteria were isolated from diffuse fluids within the
Bio9 Riftia colony as well as from the base of the Tube Worm Pillar.
Megafaunal colonization. Between December 1993 and October 1994, individuals
of Riftia within thriving colonies grew an additional 0.2—1.0 m. The number of Riftia
continued to increase, doubling their densities in certain colonies (e.g. Bio8 and Bio9).
The amount of seafloor area occupied by Riftia increased little ((1 m2) in the
northern colonies, while central (Bio82 and Bio119) and southern colonies (Bio141
and Bio142) expanded along the primary fissure by an additional &2—10 m2, suggesting that the carrying capacity of Riftia in these areas had not been reached in
December 1993. Between December 1993 and October 1994, ¹evnia (with their tightly
aggregated tubes forming small tufts; Fig. 8b) continued to colonize the Riftiadominated colonies. Individuals of ¹evnia identified as alive in December 1993 were
still alive in October 1994. By October 1994, small bathymodiolid mussels (&2—11 cm
long) were nestled under pieces of basaltic talus and in narrow cracks in the seafloor
within 0.5—6 m of the tube worm assemblages (Fig. 8c, Table 2). Within adjacent
(1 m2) areas of seafloor surrounding the Transect colonies, mussel densities averaged
1—3 m~2, with an average of 9 individuals around a given Riftia community (Table 2).
The greatest mussel density (5 m~2) was observed near the Bio141 and Bio142
colonies.
Fluid chemistry. Concomitant with the observed faunal changes between December
1993 and October 1994 was a continued decline of hydrogen sulfide and iron
concentrations in 20°—32°C diffuse fluids (Table 2). In October 1994, hydrogen sulfide
concentrations ranged from 0.40 mmol kg~1 in the central (Bio82) Transect region to
0.68 mmol kg~1 (&0.3 mmol kg~1 lower than in December 1993) in the southern
(Bio141 and Bio142) region and 0.80 mmol kg~1 in the northern Hole-to-Hell
area (Bio9). Colonies that exhibited the largest increase in tube worm abundance
between December 1993 and October 1994 (e.g. Bio9 and Bio141) also contained
the highest H S concentrations. Maximum methane concentrations decreased from
2
0.03 mmol kg~1 in December 1993 to 0.009 mmol kg~1 in October 1994 (varied from
0.003 to 0.009). Carbon dioxide concentrations decreased dramatically from 9.4 and
11.6 mmol kg~1 to 6.7 and 2.7 mmol kg~1 within the southern Bio141 and Bio142
colonies, respectively, while carbon dioxide concentrations increased slightly from 5.6
to 6.2 mmol kg~1 within the central Transect’s Bio82 colony (Table 2).
3.3.4. November 1995
»ent emissions. In November 1995, no spatial displacement of vent emissions was
detected and no additional cessation of flow was observed since the previous visit in
October 1994. However, vent openings that supported the Bio63 colony in March
1992 and ceased venting by December 1993 had been reactivated between October
496
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
1994 and November 1995 with 9°C vent fluids. No faunal response to this reactivation
was evident. Worm tubes of ¹evnia (dead since December 1993) in this colony showed
little decay or weathering after 23 months. Vent emissions in the central transect area
not only increased in intensity, but the areal extent of the Bio82 and Bio119 colonies
increased up to 20 m2 more than the northern and southern colonies, where the
average areal increase was 3 m2.
Megafaunal colonization. ¹evnia continued to colonize in tufts (&20 ind. per
colony) within Riftia communities. A few groups of ¹evnia observed alive in December
1993 were also identified as being alive in November 1995. The number of dead ¹evnia
observed within Riftia colonies also increased between October 1994 and November
1995 (e.g. at Bio119, Table 2), despite persistent vent flow. In November 1995, dead
¹evnia (tufts of empty tubes) were present in all active colonies except Bio141 and
Bio142. Riftia colonization dramatically increased at all active vents within the
Transect area (e.g. Bio9, Bio82, Bio141, and Y vent at the base of Tube Worm Pillar).
The increasingly dense matrix of vestimentiferan tubes (since 1993) provided microhabitats for expanding numbers of amphipods, leptostracans, and limpets (Table 2).
By November 1995 (55 months after the initiation of venting), Bathymodiolus
thermophilus colonized the seafloor surrounding every surviving vestimentiferan
colony in the Transect except the one at Bio9 (Table 2). Mussels (less than 12 ind.,
3—10 cm in length) were observed to be attached to tubes of Riftia (Fig. 8d), but only in
the southernmost colonies (Bio141 and Bio142). Mussel colonization surrounding the
central Bio119 colony was the most extensive along the Transect, as mussels formed
tight clumps (4—12 ind.) and linear rows within cracks that were spread over a 120 m2
area. The largest mussel (12 cm) within the Transect was observed in this area. The
abundance of serpulid polychaetes (¸aminatubus alvini and Protis hydrothermica,
indistinguishable in video) was extremely high (up to 125 ind. m~2) from the location
of the Bio119 colony in the central fissure across the low-relief ASC floor up to and on
top of the adjacent ASC walls. Serpulids also encroached in greater numbers from
peripheral areas around the Bio141 and Bio142 colonies to within centimeters of the
base of Riftia tubes. Abundances of both mussels and serpulids were much lower in the
northern end (Hole-to-Hell area) of the Transect. The first appearance of anemones
(3 ind. of Cyanathea hydrothermala, based on descriptions in Doumenc and VanPraët, 1988) within the Transect area was observed in November 1995 on slightly
iron-stained basaltic talus adjacent to the Bio8 colony.
The biological and geochemical changes that occurred within the Bio12 colony in the
northern Hole-to-Hell area were in marked dissimilarity to the range of biological and
geochemical changes that took place within the rest of the Transect colonies between
October 1994 and November 1995. The 29°C vent fluids bathing the Bio12 colony in
November 1995 contained extremely low concentrations of H S (at the limit of detection:
2
0.015 mmol kg~1) and relatively high concentrations of iron (&0.28 mmol kg~1, Table 2).
Methane concentrations of 0.0003 mmol kg~1 in the Bio12 fluids were an order of
magnitude lower than previously observed within the Transect, while concentrations
of carbon dioxide (5.2 mmol kg~1) were similar to those in other Transect colonies.
Rust-colored ferrous-oxide precipitation coated the entire vestimentiferan colony,
including brachyuran crabs, limpets, as well as much of the surrounding seafloor
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
497
(partially staining the Bio9 colony 4 m to the north, Fig. 5e). Only 10 of the &250
individuals of Riftia comprising this colony were considered to be alive (plumes
observed). No ¹hermopalia taraxaca (‘‘dandelions’’), commonly observed among dead
colonies lacking vent flow in December 1993, were observed within this dying assemblage.
Fluid chemistry. Between October 1994 and November 1995, H S concentrations
2
within diffuse fluids continued to decline in the Bio9, Bio82, and Bio142 colonies (from
0.40—0.80 mmol kg~1 to 0.194—0.301 mmol kg~1) with a slight increase only in the
Bio141 colony (from 0.68 to 0.725 mmol kg~1) (Table 2). Maximum diffuse fluid
temperatures in the Transect colonies declined by as much as 10°C. Methane concentrations showed only a slight general decrease between October 1994 and November
1995 (Table 2). Maximum measured iron concentrations remained low in the central
and southern colonies (&0.011 mmole kg~1 in Bio82 and 0.012 mmole kg~1 in
Bio141 and Bio142).
3.4. Patterns of mobile fauna distribution (**Transect Runs++)
In April 1991, only six brachyuran crabs (Bythograea thermydron) and one zoarcid
fish (¹hermarces cerberus) were documented along the length of the Transect. During
the next 11 months, the abundance of mobile vent organisms [e.g. Aphotopontius
acanthinus, Halice hesmonectes, Bythograea thermydron, Cyanograea predaetor,
Munidopsis subsquamosa, ¹hermarces cerberus, and an octopus (n. gen. n. sp., E.
Hochberg, pers. comm., 1996)] and mobile ‘‘non-vent’’ organisms (principally
nematocarcinid shrimp) dramatically increased within the Transect. Over 1207
brachyuran crabs, 127 galatheid crabs, 91 nematocarcinid shrimp, 17 zoarcid fish, 12
amphipod swarms, and 6 octopods were observed during a north to south transect
survey in March 1992. During transect runs between March 1992 and November
1995, the abundance of brachyuran and galatheid crabs increased by &12% and
137%, respectively; nematocarcinids increased by &8%, and zoarcids by a marked
892%. Dense ‘‘swarms’’ of a new species of pardaliscid amphipod, Halice hesmonectes
(Martin et al., 1993), were observed maintaining their position within meters above
low-temperature diffuse venting and shimmering water (Kaartvedt et al., 1994; Van
Dover et al., 1992). These ‘‘swarms’’ are believed to represent the greatest density of
invertebrates known in the deep sea, as numbers have been reported in excess of 1000
individuals per liter (Van Dover et al., 1992). Previous measurements of volume
occupied by these swarms was (1 m3 (Kaartvedt et al., 1994), while our observations
indicate that these swarms can occupy an area as small as &1 m3 above small groups
of mussels and up to &18 m3 overlying Riftia assemblages (Fig. 8f ). The distribution
and abundance of amphipod swarms and octopods (1 ind. sighted within &10 m of
same location each year suggesting territoriality) remained remarkably constant in
successive years. Amphipod swarms were observed above the same active vestimentiferan colonies (e.g. Bio12, Bio82, Bio119, and Bio141) from year to year (Table 2).
Changes in the distribution of brachyuran and galatheid crabs between March 1992
and November 1995 reflected the redistribution and cessation of venting emissions
over this time period (Fig. 9). Both species increased in density closer to the vent
openings, and galatheids (and nematocarcinid shrimp) tended to increase in density
498
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
just south of active vestimentiferan colonies. Both species of crabs were evenly
distributed along the Transect in March 1992, when diffuse flow was vigorous,
unorganized, and not yet as spatially discrete as in December 1993. As flow became
focused over time, the density of both crab species increased near diffuse and
high-temperature venting areas within the Transect (Fig. 9). The relative changes in
galatheid distribution paralleled those of brachyurans until November 1995, when the
number of galatheids increased from &50 to over 264 individuals between Biomarkers d114 and d120 resulting in higher densities than brachyurans (Fig. 9).
This shift in galatheid distribution was coincident with the most pronounced and
widespread colonization of mussels and serpulids (near the Bio119 colony) in the
Transect. At other sites in the eastern Pacific (e.g. 13°N EPR and Galápagos Rift),
galatheid crabs are commonly observed walking among vestimentiferan tubes.
However, galatheid crabs were never observed among or within a meter of any of
the newly-established vestimentiferan colonies within the Transect until 55 months
after the eruption (and then only one individual within the Bio8 and Bio142 colonies)
(Fig. 8c).
4. Discussion
4.1. Rates of biological processes on the fast-spreading East Pacific Rise
Since the discovery of deep-sea hydrothermal vents in 1976, there has been a growing body of evidence that rates of certain biological processes proceed relatively
rapidly in these unusual environments (Grassle, 1985; Lutz et al., 1988; Roux et al.,
1989; Tunnicliffe, 1991; Childress and Fisher, 1992). Documented temporal and
spatial changes within the Transect area provide dramatic, unambiguous evidence
that rates of colonization and growth of vestimentiferan tube worms, as well as diverse
faunal assemblages, are extremely rapid at deep-sea hydrothermal vents. The fastest
rates of growth reported for both ¹evnia (&30 cm/yr) and Riftia (&85 cm/yr) were
exhibited in the northern Hole-to-Hell area (Lutz et al., 1994). The rapid establishment of 17 ¹evnia colonies in virtually every locale where diffuse sulfide-rich hydrothermal discharge was most intense in April 1991 suggests that ¹evnia is a more
effective colonizer than other megafaunal species associated with these fast-spreading
MOR systems. Abundance estimates obtained 11 months after the eruption, from
colonies spanning 1 km of seafloor, suggest that founding population size of ¹evnia on
fast-spreading MORs may average between &30—150 ind. per colony. The presence
of ¹evnia with highly variable tube lengths (up to 30 cm) within each of the colonies
suggests the possibility of either continuous recruitment or a single recruitment event
containing individuals with highly variable, environmentally driven, growth rates.
Observations within the Transect also indicate that adult ¹evnia are more tolerant of
higher temperatures and sulfide concentrations than Riftia, as individuals preferentially colonized the most vigorous H S-rich diffuse venting areas, and active chimney
2
surfaces. The presence of numerous dead ¹evnia (that were documented alive in
previous years) coexisting with live Riftia and ¹evnia suggests that, under conditions
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
499
Fig. 9. The temporal distribution and density of brachyuran and galatheid crabs along the Transect
between March 1992 and November 1995. Nematocarcinid shrimp densities (not shown here) closely
mirrored those of galatheid crabs and often exceeded galatheid density and abundance in 1992. The major
high and low temperature venting areas are indicated in the uppermost graph.
500
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
of persistent venting (e.g. at Bio82 and Bio119), individuals of ¹evnia may have
a life-span of 4—5 yr.
The maximum estimated size of Bathymodiolus thermophilus individuals surrounding Riftia colonies was &12 cm in November 1995 (4.5 yr after the eruption).
According to estimated growth rates for B. thermophilus (Rhoads et. al., 1981; Fisher
et. al., 1988), this size corresponds to approximately 6.5 yr of age, and thus would have
had to colonize these vents late in 1989, 16 months prior to the eruption. Assuming
that these mussels were not present in the Transect colonies prior to the eruption,
rates of growth exhibited by B. thermophilus would be between 2.6 and 11.1 cm/yr
— rates considerably higher than those previously estimated for vent-endemic
mussels.
Within 2.5 yr of the inception of venting activity, 12 vent-associated species were
in the vicinity of newly-formed vents within the Transect area and, after one
additional year, 29 species were documented within this same area. In less than
5 yr, the accumulation of 32 species within the Transect accounted for 73% of the
total number of species sampled or observed along the length of the 9°—10°N ridge
crest (between 9°17@N and 9°54@N) (Table 3). Therefore the establishment of a community typical of those encountered between 9° and 10°N on the EPR required only
3.5 yr. Based on the faunal inventory of Tunnicliffe (1991), approximately 88% of
the species inhabiting this region are shared with 13°N EPR, 60% with 21°N EPR,
and 42% with vent fields along the Galápagos Rift. At least five taxa encountered
between 9°17@N and 9°54@N (12% of taxa found within this region) represent new
species (Table 3).
4.2. Patterns of mobile fauna distribution
Comparisons of various mobile vent and non-vent fauna over the entire Transect,
mapped from surveys in successive years, revealed that at any given point in time, the
distribution of crabs most reflected the spatial distribution of venting activity, with
scavenging galatheids favoring dying communities with little H S, and predatory
2
brachyurans preferring active areas and vestimentiferan colonies. Similar abundances
of nematocarcinid shrimp (Nematocarcinus ensifer?) and galatheid crabs along the
Transect in March 1992 and December 1993 suggest that these scavengers play
a more significant role in the trophic structure of northern EPR hydrothermal vent
sites than previously considered. Their tendency to aggregate 5—20 m south of active
diffuse venting areas may indicate a general southerly movement of organic material
within the ASC. At 13°N EPR, galatheids are believed to preferentially occupy areas
close to the Riftia thickets (Fustec et al., 1987; Jollivet, 1993). However, our results
suggest that these crabs have a lower tolerance to H S and/or other reduced chemical
2
species than brachyuran crabs (which are also frequently present on high-temperature
vents). It has been shown that B. thermydron has a greater resistance to sulfide (no
heart rate response up to 1400 lM of sulfide) than both M. subsquamosa and
Alvinocaris lusca (Vetter et al., 1987). We hypothesize that high levels of sulfide in
diffuse fluids during the 4 yr following the eruption (Table 3) prevented galatheids
from inhabiting their commonly observed habitats within ¹evnia and Riftia
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
501
Table 3
Taxonomic list of the 45 species sampled or observed between 9°17@N and 9°54@N along the EPR during
cruises associated with this study
Class
Order
Family
Genus
Hydrozoa
Anthozoa
Eocanthocephela
Obturata
Siphonophora
Actiniaria
Rhodalidae
Actinostolidae
Hypoechinorhynchidae
Riftiidae
Tevniidae
Polychaeta
Phyllodocida
¹hermopalia
taraxaca!
Cyanathea
hydrothermala!
Hypoechinorhynchus thermaceri!
Riftia
pachyptila!
¹evnia
jerichonana!
Oasisia
alvinae!
¸epidonotopodium fimbriatum?!
Branchipolynoe
symmytilida
Hesiolyra
bergi!
Nereis
sp.!
Amphisamytha
galapagensis!
Alvinella
pompejana!
Alvinella
caudata!
Paralvinella
grasslei!
¸aminatubus
alvini!
Protis
hydrothermica!
Calyptogena
magnifica
Bathymodiolus
thermophilus!
n. gen.
n. sp.!~#
Neolepetopsis
densata
Eulepetopsis
vitrea
¸epetodrilus
pustulosus!
¸epetodrilu
elevatus!
¸epetodrilu
cristatus
¸epetodrilu
ovalis!
Sutilizona
theca
Peltospira
operculata
Peltospira
delicata
Melanodrymia
n. sp.#
Cyathermia
naticoides
Provanna
sp.
Phymorrynchus
sp.!
Neolepas
sp.
unidentified gen.
sp.
Dahlella
caldariensis!
Aphotopontius
acanthinus!,#
Halice
hesmonectes!,#
»entiella
sulfuris!
n. gen.
n. sp.!,#
Bythograea
thermydron!
Cyanagraea
praedator!
Munidopsis
subsquamosa!
Alvinocaris
lusca!
¹hermarces
andersoni!
Bythites
hollisi!
Basibranchia
Terebellida
Bivalvia
Cephalopoda
Gastropoda
Crustacea
Pisces
Polynoidae
Hesionidae
Nereididae
Ampharetidae
Alvinellidae
Sabellida
Serpulidae
Eulamellibranchia
Filibranchia
Octopoda
Patellogastropoda
Vesicomyidae
Mytilidae
Octopodidae
Neolepetopsidae
Vetigastropoda
Lepetodrilidae
Neomphalina
Scissurellidae
Peltospiridae
Neotaenioglossa
Caenogastropoda
Pedunculata
Neomphalidae
Provannidae
Turridae
Scalpellidae
Leptostraca
Copepoda
Amphipoda
Nebaliidae
Siphonostomatoidae
Pardaliscidae
Lysianassidae
Decapoda
Bythograeidae
Perciforma
Galatheidae
Bresiliidae
Zoarcidae
Bythitidae
! Observed within the Transect.
" New species also present at 13°N EPR.
# New species.
Species
502
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
assemblages. As sulfide levels diminished within the Transect colonies, galatheids
moved closer to the vent openings. The increasingly discrete patterns of mobile faunal
distribution in time and space stress the importance of investigating temporal changes
in vent communities on large continuous spatial scales.
4.3. Influence of vent fluid chemistry
The April 1991 diffuse fluids (22°C to 55°C) contained H S concentrations which to
2
date, have only been found in fluids with temperatures exceeding &200°C (Von
Damm, 1995). Fluid compositions measured in April 1991 were unusual compared to
most diffuse flow because of their low chlorinities, high H S concentrations, and high
2
iron and manganese concentrations. For example, a maximum of 0.33 mmol l~1 of
H S was documented (close to the plumes Riftia) at the Galápagos Spreading Center
2
(Johnson et al., 1988a), while values of 1—2 mmol kg~1 were common at diffuse flow
vents associated with the 1991 eruption, and concentrations up to 8.5 mmol kg~1
(&27 times greater than at Galápagos) were measured at the base of the Tube Worm
Pillar (55°C at ‘‘Y’’ vent). Although the absolute H S concentrations in diffuse flow
2
vents along the Transect decreased in subsequent years, the absolute levels in November 1995 were still at least twice the maximum of those encountered at the Galápagos
Spreading Center (Johnson et al., 1988a). Temperature maxima within the Transect
(&30°C) also remained twice the maximum of Galápagos vents (e.g. Edmond et al.,
1979; Johnson et al., 1986; Johnson et al., 1988a). These unusually high H S concen2
trations in diffuse, low-temperature fluids may provide an explanation for the rapid
growth observed within the pioneering colonies of ¹evnia and Riftia. Symbiotic
bacteria within the vestimentiferan trophosome may have been influenced by the
‘‘abnormally’’ high levels of H S in the diffuse flow, with increased rates of producti2
vity and supported rapid linear worm tube growth (70% faster than any previously
reported for ¹evnia; Roux et al., 1989). The increased surface area of an enlarging
trophosome also may have served to support accelerated rates of growth.
Coincidentally, high concentrations of H S also may partially explain the lack of
2
colonization by Riftia to ¹evnia colonies in March 1992. While we realize that
colonization is a synthesis of many processes, we suggest that Riftia larvae may be less
tolerant to high levels of H S and iron than ¹evnia larvae, and thus incapable of
2
recruiting to post-eruptive vent areas until the high concentrations of reduced chemicals decrease. A sharp decline of iron was observed over the year following the
eruption (Von Damm et al., 1996). Time-series observations within the Transect and
at 13° N EPR (Fustec et al., 1987) suggest that the adult form of ¹evnia may be more
tolerant of high hydrothermal flux than adult Riftia (perhaps due to differences in
plume size and structure). Once Riftia was established within the Transect, colonies
exhibiting the largest increase in tube worm abundance (e.g., Bio9 and Bio141) also
consistently displayed the highest H S concentrations.
2
While there was no detectable migration of hydrothermal activity along the
length of transect over 55 months, death of newly formed biological assemblages
was either coincident with the cessation of flow or abrupt geochemical changes where
low-temperature fluids (29°C) were virtually depleted of H S. To the best of our
2
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
503
knowledge, there are only two instances where H S was considered to be absent from
2
hydrothermal emissions: at one high-temperature vent (291°C) and at several lowtemperature vents (&27°C) on the Juan de Fuca Ridge (Tunnicliffe et al., 1986).
However, no vent fauna were directly associated with these venting fluids. The death
of the Bio12 colony marks the first time that abrupt changes in vent-fluid geochemistry (loss of H S and a large increase of iron) have been correlated with the death of
2
a thriving vent community.
4.4. Patterns of faunal colonization
Temporal and spatial patterns of megafaunal colonization represent an intimate
relationship between the life-history characteristics of individual species, physical
oceanographic processes, and dynamic hydrothermal conditions. Colonization at
nascent hydrothermal vents within the Transect suggests a temporal sequence of
faunal change that also may be closely connected to changing geochemical conditions.
As all vent-endemic fauna ultimately depend on chemosynthesis for their survival, lifehistory strategies to ensure the persistence of their populations may be directly
influenced by changing geochemical conditions. Several other factors may profoundly
influence the successional patterns of megafaunal colonization observed at nascent
vents within the Transect.
In addition to the influences of fluid chemistry, priority effects, where the conditioning of substrates by microbial alteration or a preceding faunal species, may be
a relatively important mechanism for megafaunal colonization. For example, the
presence of high microbial production and byproducts may facilitate the early
recruitment of some species (e.g. ¹evnia) and/or preclude the recruitment of other
vent-endemic species (e.g. Riftia and Bathymodiolus thermophilus). If the role of
priority effects were a controlling force on the observed pattern of megafaunal
colonization, Riftia would require some additional biotically mediated cues to settle
that are not provided until ¹evnia sufficiently alters (physically or chemically) the vent
microhabitat. However, while the large majority of Riftia colonized to thriving ¹evnia
populations, ‘‘sporadic’’ colonization by Riftia in many areas where pre-existing
¹evnia were not present suggests that microhabitat pre-conditioning by ¹evnia is not
mandatory (and perhaps not even relevant). The colonization by ¹evnia to every area
of vigorous venting along the Transect (and not outside these areas) suggests that
there may be a causal relationship between ¹evnia larval recruitment and the earlier
presence of microbial material, the particular environmental conditions existing at
that time, and larval/symbiont interactions that place aggregates of ¹evnia close to
vent openings.
Species-specific larval recognition and acquisition of the appropriate endosymbiont
also may profoundly effect the timing of successful colonization. ¹evnia and Riftia
depend on intracellular, sulfur-oxidizing symbiotic bacteria for their energy source.
Both ¹evnia and Riftia harbor genetically identical endosymbionts (Edwards and
Nelson, 1991) that must be encountered in and ingested from the water column during
the worm’s larval stage (Jones and Gardiner, 1988; Southward, 1988; Cary et al., 1993).
Thus, the procurement of symbionts is prerequisite to the colonization of newly
504
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
formed vents. If both ¹evnia and Riftia larvae were present in the water column above
the Transect area in April 1991 and March 1992, and colonization of nascent vent
areas by ¹evnia preceded that of Riftia in vigorously venting areas, then ¹evnia larvae
may be more efficient with respect to symbiont recognition and/or acquisition than
are the larvae of Riftia. ¹evnia larvae also may be more sensitive to, or tolerant of, high
levels of hydrogen sulfide and iron or other physio-chemical constituents than Riftia.
In contrast, planktotrophic larvae of B. thermophilus are thought to acquire their
endosymbionts transovarially, from parent to offspring (Cary and Giovannoni, 1993).
Thus, one might expect mussel larvae to be one of the earliest colonizers of newly
formed vents. The fact that mussels were not observed within the Transect area prior
to October 1994 may be due to a combination of our inability to discern extremely
small individuals in the video images and the likelihood that they may be relegated to
living in cracks or on the underside of the basaltic crust (perhaps in order to avoid
potential predators; the underside of basaltic rock samples recovered in October 1994
and November 1995 within the Transect were colonized by numerous mussels less
than a 1 cm in length). While symbiont acquisition may impose important constraints
that functionally partition recruitment and colonization success of certain vent species
over time, these constraints alone do not adequately explain the observed patterns of
colonization within the Phoenix vent area.
The sequential appearance of megafaunal species at nascent vents also may be
influenced by the distance from larval sources. The distance of larval sources of the
dominant megafauna along the 9°—10°N segment to nascent vent areas can be
considered to be close (Riftia— less than &30 m east, &450 m south, and &700 m
to the north; ¹evnia— &700 m north, &1450 m to the south; and Bathymodiolus
thermophilus— less than &30 m east, and 430 m to the south). It should be noted that
another potential source of larvae were small beds of large vesicomyid clams present
700 m north of the Transect in 1992. In situ observations of two spawning episodes by
Riftia inhabiting the Tube Worm Pillar in November 1992 suggest that these individuals were supplying gametes to the water column (Van Dover, 1994). Given the
high levels of inferred long-range dispersal ability for Riftia (Black et al., 1994), ¹evnia
(Trivedi et al., 1994), and B. thermophilus (Craddock et al., 1995; Jollivet, 1996) along
the EPR and the close proximity of these potential larval sources, one might have
expected colonization by Riftia and B. thermophilus to have been concomitant with
¹evnia at uncolonized newly formed vents.
An alternative source of ¹evnia, Riftia, and B. thermophilus larvae was the water
column overlying the Transect vents between April 1991 and March 1992, where
physical oceanographic processes may have temporally partitioned the recruitment
and/or colonization success of these three species. Little is known about the dynamic
variability of bottom currents and hydrothermal plumes emanating from the ridge
crest in this area. From a series of CTD tows near the EPR crest between 9°—10°N,
Wijffels et al. (1996) showed that 3He-rich plumes spread westward from the ridge for
hundreds of kilometers. Stable periods of moderate bottom flow to the north along
the ridge axis followed by abrupt reversals to the south for periods of up to 2 months
have been recorded at 13°N EPR in 1991 (Chevaldonné et al., 1997). In contrast,
Cannon et al. (1991) showed, in a series of long-term current meter deployments on
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
505
the Juan de Fuca Ridge, that the topography of the ridge crest greatly affects the
variability in the local current structure. Our observations during hydrocasts conducted to track the hydrothermal plume above the Transect area also suggest large
temporal differences in current direction. Buoyant plume surveys conducted between
9°40@N to 9°50@N in December 1991 revealed a notable lack of identifiable vestimentiferan larvae, few mussel and clam larvae (Kim et al., 1994), and large numbers of
gastropod larvae (potentially 14 species) (Mullineaux et al., 1996). Hypotheses to
explain the temporal partitioning of larvae from correlations of current patterns and
relative distance from larval sources are difficult to formulate. However, given that
most megafaunal sources of larvae existed within 700 m of the Transect in April 1991,
we consider it likely that a pool of ¹evnia, Riftia, and B. thermophilus larvae coexisted
above the Transect vents immediately following the eruption.
The role of competition and predation as structuring forces in vent community
succession remains virtually unexplored. At the Galápagos Rift vents, Bathymodiolus
thermophilus is thought to outcompete Riftia via the removal of hydrogen sulfide from
vent fluids prior to the fluid reaching the worm plumes (Johnson et al., 1988b) and
through physical displacement of Riftia tubes and plumes (Hessler et al., 1988).
A recent study suggests that individuals of Riftia may be capable of producing an
additional connection via tube growth at its posterior end and dissolving its previous
attachment to the substrate (Gaill et al., 1997). Thus, Riftia may be able to modify its
relative position continually to maintain access to vent fluids. This adaptation would
yield a competitive advantage over ¹evnia, which apparently does not possess this
capability. Predation by crabs and fish on adult populations of organisms (e.g.
amphipods, Riftia plumes, and mussels) appeared to have little effect on community
composition. However, it is envisioned that predation during the initial stages
of community development, through the direct consumption of larvae and juveniles and/or consumption and removal of bacteria or microbial byproducts from the
basalt, may have a substantial impact on the subsequent patterns of successful
colonization.
In an effort to understand the mechanisms by which the above complex processes
interact to produce the sequence of colonization from microbial mats to ¹evnia, Riftia,
and B. thermophilus, patterns of faunal colonization and the temporal evolution of
vent fluid chemistry can be correlated to generate a general descriptive model of
low-temperature vent community succession along the fast-spreading EPR between
9°—10°N and contiguous ridge axes. Such models can provide the ecological context
for a broad spectrum of manipulative experiments (e.g. Mullineaux et al., 1998), as well
as the basis for assessing the impact of a variety of anthropogenic perturbations (e.g.
intensive sampling of constituent fauna, drilling of active vent sites, and extraction of
polymetallic minerals from accreting hydrothermal deposits) on biological communities at deep-ocean hydrothermal vents.
4.5. Descriptive models of low-temperature hydrothermal vent community succession
The relatively few hypotheses that attempt to explain patterns of succession at
deep-sea hydrothermal vents along the northern EPR and Galápagos Rift propose
506
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
that as hydrothermal flux and hydrogen sulfide concentrations diminish, bivalves gain
a competitive dominance and gradually replace tube worms as the dominant vent
megafauna (Hessler et al., 1988). Suspension feeders (e.g. serpulid polychaetes,
pectinids, siphonophores, and anemones) decrease with diminishing vent activity,
while carnivores (e.g. crabs, shrimp, and whelks) increase in abundance. Mussels are
hypothesized to be the last megafaunal survivors as hydrothermal activity ceases at
a given vent field.
The model of vent community succession we propose is based on the principle that
faunal succession is independent of species availability and assumes availability of
a suitable site. Pickett’s law of dynamic tolerance (Pickett and Mcdonnell, 1989) states
that an assemblage of species having differing tolerances to the abiotic environment
and differing capacities for interaction through resource use will sort through time in
order of their tolerances. Based on these assumptions, we hypothesize that the
following general sequence of biological successional changes occur at deep-sea
hydrothermal vents that form after volcanic eruptions along the fast-spreading EPR
and contiguous ridge axes.
Shallow dike intrusions, the extrusion of magma onto the seafloor, and the tectonic
perturbations associated with submarine eruptions on the MOR crest form the
shallow crustal permeability structure through which circulating hydrothermal fluids
mix with seawater before exiting the seafloor. This newly formed subsurface network
facilitates the potential release of either microbes or microbial byproducts from
a subsurface biosphere (a microbial source perhaps latent before the eruption/disturbance) (Deming and Baross, 1993), and creates new habitat space for vent-endemic
species. Microbially derived material quickly covers extremely vigorous diffuse vent
flow areas of the seafloor in the form of thick white microbial mats. Mobile vent and
non-vent fauna (e.g. amphipods, copepods, octopods, brachyuran crabs, galatheid
crabs and nematocarcinid shrimp) quickly move in to take advantage of the increase
in organic material. Larvae of the pioneering vestimentiferan species, ¹evnia
jerichonana, whether cued by physio-chemical stimuli such as high levels of iron,
hydrogen, and/or H S, or through efficient acquisition of their endosymbiont from the
2
water column, gregariously settle in the most intense areas of diffuse vent flow,
predominantly within primary fissures that served as sites of magma extrusion.
Within one year of an eruption, a marked reduction in the areal coverage of the
microbial mats will reflect the spatial reduction of hydrothermal flux and the geomorphology of each specific diffuse vent area. The characteristic bulls-eye distribution
pattern of vent fauna (Van Dover and Hessler, 1990) will become easily recognizable
at each of these individual vent areas by this time, and the gregarious nature of the
¹evnia settlement will place these founders of discrete colonies at the center of the
bulls-eye pattern [a pattern of colonization also observed by Fustec (1987) at 13°N
EPR].
Once levels of H S and other reduced chemical species have decreased appreciably
2
(within &2 yr) and ¹evnia colonies are established (&30 to 200 ind. per colony),
Riftia, after acquiring its symbiont from the water column or perhaps near newly
established individuals of ¹evnia, will rapidly grow and dominate all areas containing
¹evnia, as well as places far removed from intense diffuse flow. The increasing matrix
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
507
of Riftia tubes will provide microhabitats or refugia for numerous species including
amphipods, copepods, shrimp, Paralvinella grasslei, and zoarcid fish. ¹evnia colonization within vestimentiferan colonies will continue, as aggregates of tubes in small tufts.
Diffuse vent emissions will become spatially focused over time and any sessile
organisms that have colonized these outlying areas will die. As concentrations of H S
2
and other reduced chemical species continue to decline (within &3 yr), mussels will
begin to colonize the underside of lobate lava crust (perhaps due to their vulnerability
to predation as juveniles) and narrow cracks peripheral to diffuse vent openings
(meters away). Riftia also will colonize peripheral areas inhabited by mussels. Changes
in the spatial extent of fluid emissions within a given region will create a patchwork of
dead and thriving biological assemblages over time. Mobile fauna will alter their
distribution in response to both the spacing of actively venting areas and the location
of dead assemblages. Death by cessation of vent flow may occur during any successional stage (and may often occur in less than a year after the initiation of venting).
Recently dead assemblages in these areas will be dominated by potentially scavenging
colonial siphonophores and galatheid crabs. Galatheid crabs and serpulids will
increase in abundance and approach active vent openings as hydrogen sulfide decreases. We hypothesize that galatheids have a limited tolerance to H S concentra2
tions that impedes their presence within newly-established faunal colonies. A gradual
decrease in H S over time will eventually permit them to occupy microhabitats
2
directly on worm tubes at vent openings. Within 4 years of the eruption, galatheids
and serpulids will thrive in areas occupied by increasingly dense mussel assemblages
that have developed along cracks adjacent to the vestimentiferan colonies. By this
time, mussels will be attached to the tubes of Riftia.
Based on times-series observations at 13°N EPR (Jollivet, 1993) and the Galápagos
Rift (Hessler et al., 1988), we hypothesize that (in the absence of volcanic, tectonic, or
anthropogenic disturbance) future observations within the Transect region will reveal
that: (1) hydrogen sulfide concentrations will continue to decrease; (2) Riftia pachyptila
will outcompete and replace ¹evnia jerichonana; (3) mytilids and vesicomyids will
gradually replace the vestimentiferans as the dominant megafauna; and (4) suspension
feeders will decrease, and carnivores will increase in abundance. Colonizing mussels
will not only outcompete vestimentiferans by displacing their plumes out of the flow,
but also by filtering the nutrients from the vent fluids before they can reach the worms’
plumes (Johnson et al., 1988b). The continued decrease in vent flux will permit clams
to settle in crevices unoccupied by mussels, and in the final waning stages of venting
within a vent field, mixotrophic mussels will stand the best chance for survival.
5. Summary and conclusions
Following the 1991 volcanic eruption between 9°45@N and 9°52@N on the East
Pacific Rise, time-series observations of community development and associated
geochemical conditions in many areas of nascent low-temperature hydrothermal
venting were conducted in March 1992, December 1993, October 1994, and November 1995. Much of the erupted volume of the 1991 flow was reimplaced into the
508
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
magma plumbing system at shallow crustal levels (manifested as chaotic collapse
morphology that dominates the ASC floor and margins from 9°45@—52@N). This
process, when coupled with subsequent dike intrusion events within the region, has
important implications for the continued heat potential to drive and sustain
hydrothermal circulation following an eruption. This geologic setting, and the
creation of a circulatory hydrothermal system that develops from it, can evoke
a potentially rapid biological response in the form of free-living microbial (and
presumably endosymbiotic bacterial) production, which, in turn, supports the structural development and maintenance of dense megafaunal assemblages.
Newly formed low-temperature vent areas were quickly enveloped by large blooms
of free-living, sulfur-oxidizing bacteria, potentially derived from a subsurface
biosphere. Numerous grazers, scavengers, and predators (e.g. brachyuran and
galatheid crabs, zoarcid fish, siphonostomatoid copepods, pardaliscid amphipods,
and several species of limpets) rapidly proliferated in response to the large microbial
production. Colonization by two symbiont-dependent species of vestimentiferan tube
worms was separated in time; pioneering colonies of ¹evnia jerichonana were established in the most intensely venting areas within eleven months, and populations of
Riftia pachyptila dominated these same areas 21 months later. Within 3.5 yr, individual Riftia pachyptila achieved tube lengths in excess of 2 m, and the increasing
density of worm tubes within developing colonies created microhabitats supporting
species diversity typical of northern EPR low-temperature vents. At this time, mussels
inhabited small basaltic cracks and numerous serpulid polychaetes colonized the
periphery of and immediately adjacent to the Riftia colonies; one year later small
numbers of mussels were fixed directly to the tubes of Riftia. Despite the presence of
well-established clam communities &700 m north of the Transect area, vesicomyid
clams were not observed within any of the newly formed vent areas.
Similar patterns of faunal succession were observed in all areas of persistent
low-temperature venting regardless of the distance separating the developing communities, suggesting that these patterns may be predictable. The close proximity of adult
megafaunal populations suggests that ¹evnia, Riftia, and B. thermophilus larvae were
equally available to colonize the Transect vents following the eruption. High levels of
iron and H S (1.0—2.0 mmol kg~1) were detected in diffuse vent fluid immediately
2
following the 1991 EPR eruption. The steady decline of H S and iron observed from
2
April 1991 to November 1995 was coincident with the sequence of vestimentiferan
colonization, subsequent mussel colonization, and the encroachment of serpulid
polychaetes and galatheid crabs toward vent openings over time.
The observed correlation between patterns of faunal succession and changing
geochemical conditions suggests that future models of faunal succession should
consider not only the interplay of species-specific life-history strategies, trophic
interactions (incorporating estimates of production and biomass), and physical
oceanographic processes, but also the effect changing geochemical conditions may
have on the sequential colonization of megafaunal species at hydrothermal vents
along intermediate to fast-spreading mid-ocean ridges. In addition, the potential for
high microbial production and associated abiotic conditions to facilitate the aggregated recruitment of ¹evnia larvae and/or preclude the recruitment of other sessile
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
509
organisms also should be considered in evaluating the underlying controls of community development.
While priority effects, species-specific symbiont acquisition, distances from larval
sources, physical oceanographic processes, and competitive/predatory interactions
may play important roles in the early assemblage of vent communities, species-specific
tolerances to geochemical conditions may also exert community structuring forces
responsible for the early recruitment success of ¹evnia over Riftia and the temporal
pattern of faunal colonization observed at numerous nascent vents within the Transect. The dynamic interaction of species’ life-history strategies, the transient nature of
vent habitats, and the variability in geochemical constituents in diffuse hydrothermal
fluids (whether serving as larval settlement cues or as barriers to colonization and
population growth) play important roles in the adaptation and distribution of
vent-endemic species.
Unlike previous attempts to model temporal and spatial changes associated
with deep-sea hydrothermal vent communities, our detailed biological and geochemical observations permitted the generation of a robust descriptive model of
community succession that takes advantage of (1) the unique opportunity to follow
temporal changes in biological community structure and vent fluid geochemistry
from the ‘‘birth’’ of numerous deep-sea hydrothermal vents; (2) the increased
frequency with which regular repeated observations were documented, and (3) the
non-invasive sampling techniques employed to ensure, as much as possible, that
the natural development of vent-endemic communities was documented. Colonization exhibited by fauna associated with high-temperature habitats may actively
influence the patterns of colonization displayed at low-temperature vents, as these
two habitats often share certain species, and may provide refugia or areas of settlement for the other. The incorporation of these patterns into our model in the near
future will allow us to develop a more comprehensive understanding of deep-sea
hydrothermal vent community development at intermediate to fast-spreading midocean ridges.
Acknowledgements
We thank Michael Black, Pierre Chevaldonné, Daniel Davis, Daniel Desbruyères,
Gyöngyvér Lévai, Mark Olsson, Anna-Louise Reysenbach, Cindy Van Dover, Verena
Tunnicliffe, and Waldo Wakefield for invaluable assistance and many constructive
comments; the officers and crew of both the R/» Atlantis II and DS» Alvin for their
invaluable technical expertise, assistance, and patience throughout the course of
numerous dive programs to the 9°50@N Biotransect area; and W. Lange for his
technical expertise in providing camera and recording systems critical to the success of
our studies; TMS thanks Robert Hessler for his support and encouragement. This
paper is Contribution number 97-06 of the Institute of Marine and Coastal Sciences,
Rutgers University and New Jersey Agricultural Experiment Station Publication No.
D-32402-4-97, supported by state funds, National Science Foundation (NSF) grants
OCE-89-17311 (RCV"Robert Vrijenhoek and RAL), OCE-92-17026 (TMS and
510
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
RAL), OCE-93-02205 (RCV"Robert Vrijenhoek and RAL), OCE-95-29819 (TMS
and RAL), OCE-96-33131 (RCV"Robert Vrijenhoek and RAL), OCE-91-00503
(DJF), OCE-90-20111 (RMH), OCE-91-01440, OCE-92-96158, OCE-93-03678
(KVD), OCE-91-00804 (MDL), and the National Institutes of Health (NIH) grant
PHSTW00735-01 (RAL).
References
Auzende, J.M., Ballu, V., Batiza, R., Bideau, D., Cormier, M., Fouquet, Y., Geistdoerfer, P.,
Lagabrielle, Y., Sinton, J., Spadea, P., 1994. Activité magmatique, tectonique et hydrothermale actuelle sur la Dorsale Est Pacifique entre 17° et 19° (campagne NAUDUR). Comptes
Rendus Academic Sciences Paris, 319, Série II, 811—818.
Black, M.B., Lutz, R.A., Vrijenhoek R.C., 1994. Gene flow among vestimentiferan tube worm
(Riftia pachyptila) populations from hydrothermal vents of the eastern Pacific. Marine
Biology 120, 33—39.
Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L., 1993. Distance Sampling: Estimating Abundance of Biological Populations. Chapman & Hall, London.
Bush, M.B., Whittaker, R.J., 1991. Krakatau: colonization patterns and hierarchies. Journal of
Biogeography 18, 341—356.
Cannon, G.A., Pashinski, D.J., Lemon, M.R., 1991. Middepth flow near hydrothermal
venting sites on the southern Juan de Fuca Ridge. Journal of Geophysical Research 96,
1 2815—12 831.
Carbotte, S.M., MacDonald, K.C., 1992. East Pacific Rise 8°—10°30@N: evolution of ridge
segments and discontinuities from SeaMARC II and three-dimensional magnetic studies.
Journal of Geophysical Research 97 (3569), 6959—6982.
Cary, S.C., Giovannoni, S.J., 1993. Transovarial inheritance of endosymbiotic bacteria in clams
inhabiting deep-sea hydrothermal vents and cold seeps. Proceedings National Academy of
Sciences 90, 5695—5699.
Cary, S.C., Warren, W., Anderson, E., Giovannoni, S.J., 1993. Identification and localization of
bacterial endosymbionts in hydrothermal vent taxa with symbiont-specific polymerase
chain reaction amplification and in situ hybridization techniques. Molecular Marine Biology and Biotechnology 2, 51—62.
Chevaldonné, P., Godfroy, A., Guezennec, J., Lesongeur, F., Fouquet, Y., Barbier,
G., Desbruyères, D., 1995. InterRidge, 4, 15—16.
Chevaldonné, P. Jollivet, D., Vangriesheim, A., Desbruyères, D., 1997. Hydrothermal-vent
alvinellid polychaete dispersal in the eastern Pacific: I. Influence of vent site distribution,
bottom currents, and biological patterns. Limnology and Oceanography 42, 67—80.
Childress, J.J., Fisher, C.R., 1992. The biology of hydrothermal vent animals: physiology,
biochemistry, and autotrophic symbioses. Oceanographical Marine Biology Annual
Review 30, 337—441.
Connell, J.H., 1979. Tropical rain forests and coral reefs as open non-equilibrium systems. In:
Population Dynamics. Anderson, R.M., Turner, B.D., Taylor, L.R., (Ed.). Blackwell, Oxford
pp. 141—163.
Craddock, C., Hoeh, W.R., Gusfatson, R.G., Lutz, R.A., Hashimoto, J., Vrijenhoek, R.C.,
1995. Evolutionary relationships among deep-sea mytilids (Bivalvia: Mytilidae)
from hydrothermal vents and cold-water methane/sulfide seeps. Marine Biology 124,
137—146.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
511
Dale, V.H., 1991. Mount St. Helens; revegetation of Mount St. Helens debris avalanche 10 years
post eruption. National Geographic Research and Exploration 7, 328—341.
Delaney, J.R., Embley, R.W., 1993. An October response cruise to the June-July intrusive/
eruptive events at the CoAxial segment of the Juan de Fuca Ridge. EOS, Transactions of
American Geophysical Union 75, 617.
Deming, J.W., Baross, J.A., 1993. Deep-sea smokers: windows to a subsurface biosphere?
Geochimica et Cosmochimica Acta N°Sp 57, 3219—3230.
Diamond, J.M., 1994. Pitcairn before the Bounty. Nature 369, 608.
Doumenc, D., Van-Praët, M., 1988. Actinies abyssales d’un site hydrothermal du Pacifique
oriental. Oceanological Acta, N°Sp 8, 61—68.
Duennebier, F.K., Becker, N.C., Caplan-Auerbach, J., Clague, D.A., Cowan, J., Cremer, M.,
Garcia, M., Goff, F., Malahoff, A., McMutry, G.M., Midson, B.P., Moyer, C.L.,
Norman, M., Okubo, P., Resing, J.A., Rhodes, J.M., Rubin, K., Sansone, F.J., Smith,
J.R., Spencer, K., Wen, X., Wheat, C.G., 1997. Researchers rapidly respond to submarine
activity at Loihi Volcano, Hawaii. EOS, Transactions of American Geophysical Union, 78,
229—233.
Edmond, J.M., Measures, C., Mangum, B., Grant, B., Sclater, F.R., Collier, R., Hudson, A.,
Gordon, L.L., Corliss, J.B., 1979. Ridge crest hydrothermal activity and the balances of the
major and minor elements in the ocean: the Galápagos data. Earth and Planetary Science
Letters 46, 19—30.
Edwards, D.B., Nelson, D.C., 1991. DNA-DNA solution hybridization studies of the bacterial
symbionts of hydrothermal vent tube worms (Riftia pachyptila and Tevnia jerichonana).
Applied and Environmental Microbiology 57, 1082—1088.
Embley, R.W., Chadwick, W.W., Jonasson, I.R., Butterfield, D.A., Baker, E.T., 1995. Initial
results of the rapid response to the 1993 CoAxial event: relationships between hydrothermal
and volcanic processes. Geophysical Research Letters 22, 143—146.
Embley, R.W., Chadwick, W.W., Shank, T., Christie, D., 1996. Geology of the 1996 Gorda
Ridge eruption from analysis of multibeam, towed camera, sidescan, and ROV data. EOS,
Transactions of American Geophysical Union 77(46), F1.
Fisher, C.R., Childress, J.J., Arp, A.J., Brooks, J.M., Distel, D., Favuzzi, J.A., Felbeck, H.,
Hessler, R.R., Johnson, K.S., Kennicutt, M.C., II., Macko, S.A., Newton, A., Powell, M.A.,
Somero, G.M., Soto, T., 1988. Microhabitat variation in the hydrothermal vent mussel,
Bathymodiolus thermophilus, at the Rose Garden vent on the Galápagos Rift. Deep-Sea
Research 35, 1769—1791.
Fornari, D.J., Haymon, R.M., Edwards, M.H., Macdonald. K.C., 1990. Volcanic and tectonic
characteristics of the East Pacific Rise crest 9°09@N to 9°54@N: Implications for fine-scale
segmentation of the plate boundary. EOS, Transactions of American Geophysical Union 71,
625.
Fornari, D.J., Van Dover, C.L., Shank, T.M., Lutz, R.A., Olsson, M., 1994. A versatile, low-cost
temperature sensing device for time-series measurements at deep-sea hydrothermal vents.
Bridge Newletter 6, 37—40.
Fornari, D.J., Embley, R.W., 1995. Tectonic and volcanic controls on hydrothermal processes
at the Mid-Ocean Ridge: an overview based on near-bottom and submersible studies. In:
Physical, Chemical, Biological, and Geological Interactions within Sea-floor Hydrothermal
Systems, Geophysical Monograph 91. Humphris, S.E., Zierenberg, R.A., Mullineaux, L.,
Thomson, R., (Eds.), American Geophysical Union, Washington, DC., pp. 1—46.
Fornari, D.J., Haymon, R.M., Perfit, M.R., Gregg, T.K.P., Edwards, M.H, 1998. Axial Summit
Caldera of the East Pacific Rise 9°N to 10°N: Geologic characteristics and evolution of the
axial zone on fast-spreading mid-ocean ridges. Journal of Geophysical Research (in press).
512
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Fridriksson, S., Magnusson, B., 1989. Development of an ecosystem on Surtsey with references
to Anak Krakatau. GeoJournal 28, 287—291.
Fustec, A., Desbruyères, D., Juniper, S.K., 1987. Deep-sea hydrothermal vent communities at
13°N on the East Pacific Rise: Microdistribution and temporal variations. Biology and
Oceanography 4, 121—164.
Gaill, F., Shillito, B., Ménard, F., Goffinet, G., Childress, J.J., 1997. Rate and process of tube
worm production by the deep-sea hydrothermal vent tube worm Riftia pachyptila. Marine
Ecology Progress Series 148, 135—143.
Geistdoerfer, P., Auzende, J.M., Batiza, R., Bideau, D., Cormier, M., Fouquet, Y., Lagabrielle,
Y., Sinton, J., Spadea, P., 1995. Hydrothermalisme et communautés animales associées sur la
dorsale du Pacifique oriental entre 17°S et 19°S (campagne Naudur, décembre 1993).
Comptes Rendus Academic Sciences Paris 320, Série IIa, 47—54.
Grassle, J.F., 1985. Distribution and biology of hydrothermal vent animals. Science 229, 713—717.
Gregg, T.K.P., Fornari, D.J., Perfit, M.R., Haymon, R.M., Fink, J.H., 1996. Rapid emplacement
of a mid-ocean ridge lava flow on the East Pacific Rise at 9°46@—51@N. Earth and Planetary
Science Letters 144, E1—E7.
Haymon, R.M., Fornari, D.J., Edwards, M.H., Carbotte, S., Wright, D., Macdonald, K.C., 1991.
Hydrothermal vent distribution along the East Pacific Rise Crest (9°09@—54@N) and its
relationship to magmatic and tectonic processes on fast-spreading mid-ocean ridges. Earth
and Planetary Science Letters 104, 513—534.
Haymon, R.M., Fornari, D.J., Von Damm, K.L., Lilley, M.D., Perfit, M.R., Edmond, J.M.,
Shanks, W.C., III, Lutz, R.A., Grebmeier, J.M., Carbotte, S., Wright, D., McLaughlin, E.,
Smith, M., Beedle, N., Olson, E., 1993. Volcanic eruption of the mid-ocean ridge along the
East Pacific Rise crest at 9°45—52@N: Direct submersible observations of seafloor phenomena
associated with an eruption event in April, 1991. Earth and Planetary Science Letters 119,
85—101.
Hessler, R.R., Smithey, W.M., 1983. The distribution and community structure of megafauna at
the Galápagos Rift hydrothermal vents. In: Rona, P.A., Boström, K., Laubier, L., Smith,
J.K.L., (Eds.), Hydrothermal Processes at Seafloor Spreading Centers. Plenum Press,
New York, pp. 735—770.
Hessler, R.R., Smithey, W.M., Keller, C.H., 1985. Spatial and temporal variation of giant clams,
tubeworms and mussels at deep-sea hydrothermal vents. Biological Society of Washington
Bulletin 6, 465—474.
Hessler, R.R., Smithey, W.M., Boudrias, M.A., Keller, C.H., Lutz, R.A., Childress, J.J., 1988.
Temporal change in megafauna at the Rose Garden hydrothermal vent (Galápagos Rift;
eastern tropical Pacific). Deep-Sea Research 35, 1681—1709.
Hessler, R.R., Desbruyères, D., 1991. Contrasting patterns of temporal changes at hydrothermal
vents between the EPR at 13°N and Rose Garden on the Galapagos Rift, Sixth Deep-Sea
Biology Symposium, Copenhagen, p. 36.
Holden, C., 1996. Oceanographers make their own hot springs. Science 274, 349.
Humes, A.G. Lutz, R.A., 1994. Aphotopontius acanthinus, new species (copepoda: siphonostomatoida), from deep-sea hydrothermal vents on the East Pacific Rise. Journal of Crustacean Biology 142, 337—345.
Johnson, H.P., Tunnicliffe, V., 1988. Time lapse photography of a hydrothermal system:
A successful one-year deployment. EOS, Transactions of American Geophysical Union 69,
1024—1026.
Johnson, K.S., Beehler, C.L., Sakamoto-Arnold, C.M., Childress, J.J., 1986. In situ
measurements of chemical distributions in a deep-sea hydrothermal vent. Science 231,
1139—1141.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
513
Johnson, K.S., Childress, J.J., Beehler, C.L., 1988a. Short-term temperature variability in the
Rose Garden hydrothermal vent field: an unstable deep-sea environment. Deep-Sea
Research 35, 1711—1721.
Johnson, K.S., Childress, J.J., Hessler, R.R., Sakamoto-Arnold, C.M., Beehler, C.L., 1988b.
Chemical and biological interactions in the Rose Garden hydrothermal vent field,
Galapagos spreading center. Deep-Sea Research 35, 1723—1744.
Jollivet, D., 1993. Distribution et évolution de la faune associée aux sources hydrothermales
profondes à 13°N sur la dorsale du Pacifique oriental: le cas particulier des polychètes
Alvinellidae. The% se de doctorat nouveau re& gime, Université de Bretagne Occidentale, pp.
357.
Jollivet, D., 1996. Specific and genetic diversity at deep-sea hydrothermal vents: an overview.
Biodiversity and Conservation 5, 1619—1653.
Jones, M.L., Gardiner, S.L., 1988. Evidence for a transient digestive tract in Vestimentifera.
Proceedings of the Biological Society of Washington 101, 423—433.
Kaartvedt, S., Van Dover, C.L., Mullineaux, L.S., Wiebe, P.H., Bollens, S.M., 1994. Amphipods
on a deep-sea hydrothermal treadmill. Deep-Sea Research 41, 179—195.
Keough, M.J., 1984. Dynamics of the epifauna of the bivalve pinna-bicolor: interactions among
recruitment, predation, and competition. Ecology 65, 677—688.
Kim, S., Mullineaux, L.S., Helfrich, K.R., 1994. Larval dispersal via entrainment into hydrothermal vent plumes. Journal of Geophysical Research 99, 12 655—12 665.
Lilley, M.D., Olson, E.J., McLaughlin, E.A., Von Damm, K.L., Bray, A.M., Buttermore, L.G.,
1996. The geochemical relationship between high temperature hydrothermal vents and
directly adjacent diffuse flow sites. EOS, Transactions of American Geophysical Union
77(46), F404.
Lutz, R.A., Fritz, L.W., Cerrato, R.M., 1988. A comparison of bivalve (Calyptogena magnifica)
growth at two deep-sea hydrothermal vents in the eastern Pacific. Deep-Sea Research 35,
1793—1810.
Lutz, R.A., 1991. Sixth Deep-Sea Biology Symposium, Copenhagen, October 1991.
Lutz, R.A., Shank, T.M., Fornari, D.J., Haymon, R.M., Lilley, M.D., Von Damm, K.L.,
Desbruyères, D., 1994. Rapid growth at deep-sea vents. Nature 371, 663—664.
Martin, J.W., France, S.C., Van Dover, C.L., 1993. Halice hesmonectes, a new species of
pardaliscid amphipod (Crustacea, Peracarida) from hydrothermal vents in the eastern
Pacific. Canadian Journal Zoology 71, 1724—1732.
Massoth, G.J., Milburn, H.B., Hammond, S.R., Butterfield, D.A., McDuff, R.E., Lupton, J.E.,
1988. The geochemistry of submarine venting fluids at Axial Volcano, Juan de Fuca Ridge:
new sampling methods and a VENTS program rational. In: DeLuca, M.P., Babb, I., (Eds.),
Global Venting, Midwater, and Benthic Ecological Processes, NURP Research Report 88-4,
29—59.
McCormick, M.I., Choat, J.H., 1987. Estimating total abundance of a large temperate-reef fish
using visual strip-transects. Marine Biology 96, 469—478.
Milligan, B.N., Tunnicliffe, V., 1994. Vent and nonvent faunas of Cleft segment, Juan de Fuca
Ridge, and their relations to lava age. Journal of Geophysical Research 99, 4777—4786.
Mullineaux, L.S., Kim, S.L., Pooley, A., Lutz, R.A., 1996. Identification of archaeogastropod
larvae from a hydrothermal vent community. Marine Biology 124, 551—560.
Mullineaux, L.S., Mills, S.W., Goldman, E., 1998. Recruitment variation during a pilot study of
hydrothermal vents (9°50@N, East Pacific Rise). Deep-sea Research II 45(1—3), 441—464.
Nelson, D., 1991. Rapid growth of unusual hydrothermal bacteria observed at new vents during
ADVENTURE dive program to the EPR crest at 9°45@—52@N. EOS, Transactions of
American Geophysical Union 72, 481.
514
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
Oosting, S.E., Von Damm, K.L., 1996. Bromide/chloride fractionation in seafloor hydrothermal fluids from 9°—10°N East Pacific Rise. Earth and Planetary Science Letters 144,
133—145.
Pickett, S.T.A., White, P.S., (Eds.), 1985. The Ecology of Natural Disturbance and Patch
Dynamics. Academic Press, Orlando, FL.
Pickett, S.T.A., McDonnell, M.J., 1989. Changing perspectives in community dynamics: a
theory of successional forces. Trends in Evolution and Ecology 4, 11—14.
Rhoads, D.C., Lutz, R.A., Revelas, E.C., Cerrato, R.M., 1981. Growth of bivalves at deep-sea
hydrothermal vents along the Galapagos Rift. Science 214, 911—913.
Roux, M., Rio, M., Schein, E., Lutz, R.A., Fritz, L.W., Ragone, L.M., 1989. Mesures in situ de la
croissance des bivalves et des vestimentifères et de la corrosion des coquilles au site
hydrothermal de 13°N (dorsale du Pacifique oriental). Comptes Rendus Academic Science
Se& r. III, 308, 121—127.
Shank, T.M., Fornari, D.J., Von Damm, K.L., Lilley, M.D., Van Dover, C.L., Lévai, G., Lutz,
R.A., 1995. Dynamic fluctuations and cyclic periodicities of low-T hydrothermal venting;
time-lapse multi-probe temperature measurements and video imaging at 9°49.8@N on the
East Pacific Rise- results from December 1993 to October 1995. EOS, Transactions of
American Geophysical Union 76, 701.
Sousa, W.P., 1980. The responses of a community to disturbance: the importance of successional age and species’ life histories. Oecologia 45, 72—81.
Southward, E.C., 1988. Development of the gut and segmentation of newly settled stages of
Ridgeia (Vestimentifera): implications for relationship between Vestimentifera and Pogonophora. Journal of the Marine Biological Association of U.K. 68, 465—487.
Thornton, I.W.B., Walsh, D., 1992. Photographic evidence of rate of development of plant cover
on the emergent island Anak Krakatau from 1971 to 1991 and implications for the effect of
volcanism. Geojournal 28.2, 249—259.
Trivedi, A.K., Black, M.B., Vrijenhoek, R.C., 1994. Allozyme diversity within and among tube
worm (vestimentiferan) species from deep-sea hydrothermal vents of the east pacific rise.
Isozyme Bulletin 23, 47.
Tunnicliffe, V., 1990. Observations on the effects of sampling on hydrothermal vent habitat and
fauna of Axial Seamount, Juan de Fuca Ridge. Journal of Geophysical Research 95,
12 961—12 966.
Tunnicliffe, V., 1991. The biology of hydrothermal vents: ecology and evolution. Oceanography
and Marine Biology Annual Review 29, 319—407.
Tunnicliffe, V., Botros, M., DeBurgh, M.E., Dinet, A., Johnson, H.P., Juniper, S.K., McDuff,
R.E., 1986. Hydrothermal vents on Explorer Ridge, northeast Pacific. Deep-Sea Research
33, 401—412.
Tunnicliffe, V., Garrett, J., Johnson, H.P., 1990. Physical and biological factors affecting the
behavior and mortality of hydrothermal vent tubeworms (vestimentiferans). Deep-Sea
Research 37, 103—125.
Tunnicliffe, V., Juniper, S.K., 1990. Dynamic character of the hydrothermal vent habitat and the
nature of sulphide chimney fauna. Progress in Oceanography 24, 1—13.
Tunnicliffe, V., Embley, R.W., Holden, J.F., Butterfield, D.A., Massoth, G.J., Juniper, S.K., 1998.
Biological colonization of new hydrothermal vents following an eruption on Juan de Fuca
Ridge. Deep-Sea Research 44, 1627—1644.
Tusting, R., Davis, D., 1992. Laser systems and structured illumination for quantitative
undersea imaging. Marine Technology Society Journal 26 (4), 5—12.
Tusting, R., Davis, D., 1993. Improved methods for visual and photographic benthic surveys.
Proceedings of Diving for Science-93, AAUS, pp. 157—172.
T.M. Shank et al. / Deep-Sea Research II 45 (1998) 465—515
515
Van Dover, C.L., Hessler, R.R., 1990. Spatial variation in faunal composition of hydrothermal
vent communities on the East Pacific Rise and Galápagos spreading center. In: McMurray,
G.R., (Ed.), Gorda Ridge: A Seafloor Spreading Center in the United States’ Exclusive
Economic Zone, Springer, New York, pp. 253—264.
Van Dover, C.L., Kaartvedt, S., Bollens, S.M., Wiebe, P.H., Martin, J.W., France, S.C., 1992.
Deep-sea amphipod swarms. Nature 358, 25—26.
Van Dover, C.L., 1994. In situ spawning of hydrothermal vent tubeworms (Riftia pachyptila).
Biological Bulletin 186, 134—135.
Vetter, R.D., Wells, M.E., Kurtsman, A.L., Somero, G.N., 1987. Sulfide detoxification by the
hydrothermal vent crab Bythograea thermydron and other decapod crustaceans. Physiological Zoology 60(1), 121—137.
Von Damm, K.L., Oosting, S.E., Buttermore, L.G., 1994. Chemical evolution of diffuse hydrothermal fluids from 9°49@—50@N EPR: The BIOTRANSECT area, EOS, Transactions of
American Geophysical Union 75, 601.
Von Damm, K.L., 1995. Controls on the chemistry and temporal variability of seafloor
hydrothermal fluids, In: Humphris, S.E., Zierenberg, R.A., Mullineaux, L., Thomson, R.,
(Eds.), Physical, Chemical, Biological, and Geological Interactions within Seafloor Hydrothermal Systems, Geophysical Monograph 91, American Geophysical Union pp. 222—247.
Von Damm, K.L., Bray, A.M., Buttermore, L.G., Lilley, M.D., 1996. Chemical changes between
high temperature and diffuse flow fluids at 9° 50@N East Pacific Rise. EOS, Transactions of
American Geophysical Union 77, 403—404.
Von Damm, K.L., Buttermore, L.G., Oosting, S.E., Bray, A.M., Fornari, D.J., Lilley, M.D.,
Shanks III, W.C., 1997. Direct observation of a high-temperature vent from vapor to brine.
Earth and Planetary Science Letters 149, 101—111.
Watremez, P., Kervévan, C., 1990. Origine des variations de l’activité hydrothermale: premiers
éléments de réponse d’un modèle numérique simple. Comptes Rendus Academie Sciences
Paris, Se& r. II 311, 153—158.
Wijffels, S.E., Toole, J.M., Bryden, H.L., Fine, R.A., Jenkins, W.J., Bullister, J.L., 1996. The water
masses and circulation at 10°N in the Pacific. Deep-Sea Research 43, 501—544.
Wright, D.J., Haymon, R.M., Fornari, D.J., 1995. Crustal fissuring and its relationship to
magmatic and hydrothermal processes on the East Pacific Rise crest (9°12@N to 54@N).
Journal of Geophysical Research 100, 6097—6120.
Marvin Lilley