r

f

REPORTS ON RESEARCH FROM THE WOODS HOLE OCEANOGRAPHIC INSTITUTION

Vol. 41, No. 2 • 1998 • ISSN 0029-8182

*•

*••?

Dark, particle-rich Intltnl dermal plumes surge
Ironi a Mack smoker chimney photographed in
IWSul 17 Son the Cast Pacific Hise.

( )n-iii HI ^ is published seiiii-annu.illv In I hr \\oods I lole
Oceanographic Institution, Woods Hole. MA 02543.

-,(IS 1'XM- r.lh-hnp \\wuulimedu oie.mns

o«-ntiii\ jnd its logo are "Registered Trademarks ol tin-
\\ ..... I- ll"l< < it eanographic Institution. AH Rights Reserved

A calendar-yeai f Viv/m/s snhsi iiplum is ,i\,nl,ihlc Im sr> m

thel'S SIN in ( .uuida. I In- \VHOI Publication P.it kaije.

including Oceanus magazine and lUww/s //'j/c('///7v///s-(;i

quarterly puhln iiimii 1m \\ IK H Associates and liu-mlsi is

a\aikihli' lur;i $25 calendar-year Ice in Ilk- 1 'S. s '.() in ( ,in.ul,i

( till Mile \nrlh Anii'i nil. Ihranniuil li.r foi t)ir(ittti\ in.iiFii/inc

iinl\ is sj'i ,nnl I lie l'nl)li( alinn Kit k;ii*i'oisK vl() In u-crixt-

the publications, please call (toll free) l-.suu-jm-hlis.nr

v\ i il «• \\'l H )l I 'nhlu at ion Sei \ lies. P( ). Hi i\ ",l 1 1 1 ",
XovlVdloul, MA (HT I.'. IMMl',

In purchase single and back-issue copies ol (hrnnn^. please
umtiifl |ane Hopewood, WH01-MS#5, Woods Hole, MA

:V Plume: 5«8-liW-T,l(. lax '.US 457 J I SL1

Checks should hi'dia\\n nn a I 'S h.mk in I 'S di ill. us and
inaile |>ay;tl>le In \\umis I lok'l )t earner aplnr Inslilntion.

\\'llen sending change ol addiess, plrasr inrlnde ni.iihnif

kihc-l. Claims tor missing numbers fiom the US

will be honored Within Ihree ninnlhsnl pnblu alum

overseas, six months.

1 upui^hl l IL)')S h\ \\oods I lolef lieaimiiiaphu lust ilntion.

Permission to photOCOpJ lor ink-i nal or peismial use or I he

internal or personal use of specific clients is granted hy

th i-iiini\ to hi ii.ii icsaiul ollu-r users resist eivd \\ilh the

( <>p\iinhi ( learance Center (CCC). provided ilia! ihrkisr

fee of $2 per copy of the article is paid dirct tly to (A (.'.
222 Rosewood Drive, Danvers, MA 01923. Special requests

should he aildii's.si'il lo I he ( >t rnini^ edilor

\\'OIR|S I Inlet Jceano^raphic I nsl it nl ion is an 1 <|iu! I inpln\
in MI ( )[ipoilnnil\ and Allinnali\e A( lion 1 ni|iln\iv

O Printed on HVVI led paper

REPORTS ON RESEARCH FROM THE WOODS HOLE OCEANOGRAPHIC
Vol. 41, No. 2 • 1998 • ISSN 0029-8182

The Mid-Ocean Ridge

Biologists' First Look at Vent Communities — Galapagos Rift, 1979 1

ByJ. Frederick Grassle

Deep-Sea Diaspora 6

The LARIT Project Explores How Species Migrate from Vent to Vent
By Lauren Mullineaux and Donal Manahan

Life on the Seafloor and Elsewhere in the Solar System 10

IfVolcanism Plus Oceans Can Support Life at the Vents. WliyNot on Other Planetary Bodies?
By John R. Delaney

ALISS in Wonderland 14

Imaging Ambient Light at Deep-Sea Hydrothermal Vents
By Sheri N. White and Alan D. Chave

The Cauldron Beneath the Seafloor 18

Percolating Through Volcanic Subsurface Rocks, Seawater Is Chemically Transformed
By Susan E. Humphris and Tom McCoIlom

How to Build a Black Smoker Chimney 22

The Formation of Mineral Deposits at Mid-Ocean Ridges
By Margaret Kingston Tivey

The Big MELT 27

The Mantle Electromagnetic and Tomography Experiment
By Donald Forsyth

Using Seismic Waves to "See" a Slice of the Oceanic Crust 30

ByJ. Pablo Canales

A Current Affair 32

Measuring Electrical Conductivity Hundreds of Kilometers within the Earth
By Robert Evans

Hitting the Hotspots 34

o r

Afeu' Studies Reveal Critical Interactions Between Hotspots and Mid-Ocean Ridges

Byjian Lin

Editor: Laurence Lippselt • Designer: Jim Canavan

© Woods Hole Oceanographic Institution

Robert B. Gagosian, Director -James E. Moltz, Chairman of the Board of Trustees

James M. Clark, President of the Corporation • Robert D. Harrington, Jr., President of the Associates

Jacqueline M. Hollister. Associate Director for Communications and Development

(_< ivi i; This M.M k snmkri. ainoni> (lie first ru-i photographed. u;is Inn IK I on (he I as| Pacific Hisc
dn i i ii» I he 1 979 expedition on \\hich black smokers were firs! discovered. Photo by Dudley Foster

Pfc '
lire

'Nothing Could Diminish the Excitement
Of Seeing the Animals for the First Time'

Biologists' First Look at Vent Communities — Galapagos Rift, 1979

J. Frederick Grassle

Director, Institute of Marine and Coastal Sciences, Rutgers University

he scientists who made the surprising discovery of teeming life around hydro! hermal vents of the
Galapagos Rift in 1977 were geologists and geochemisls. They had not expected lo find spectacular
colonies of previously unknown. large animals on the deep seafloor.

idn'l take long after the first amazing reports arrived for deep-sea biologists to mobilize a Galapagos
voyage. Holgerjannasc.il, Howard Sanders, and I from the Woods Hole Oceanographic
Institution. Ruth Turner and Carl Berg of Harvard Uni-
versity, and Dob I fessler of Scripps Insl it ill ion of Ocean-
ography (SIO) mel in Woods Hole and decided to send
the National Science foundation (\SI-")a proposal fora
return trip. I would coordinate (he proposal and be
chief scientist of I he expedition. After dis-
cussions with many colleagues, we de- , -—
vised a cruise on a ship with well-
equipped labs, combined with an Alviii
bottom-station expedition. We would
subject the newly _ —
discovered coinniu-
nities to the full
arsenal of tech-
niques available
lo modern biolouv.

Top (left to right):

Scienlisls 1 1 mil ;i wide
range ofdigcipliiirs
( here pausing lor an
Informal dinner on
deck) assembled for
llu> l'ir-,1 expedition lo
study hydmllicmuil
venl ecosystems in
1979. UIIOIs sub-
mersible , ill in »iis
extensively modified
l» accommodate new
equipment lor I he
biological dive*. . \lvin
is deployed between
the catamarans of it*
original lender. I ulu.

Kelow. a llonrislihi!;
colony ol' I lie red-
lipped Inbeworin

I'm ill 1^1-1 I i i'i I Grasslc

dives to the vents in

l9"*)abiMtrd.-l/r/M.

The original proposal to NSF. submitted in Janu-
ary of 1978, contained subsections from 19 princi-
pal investigators representing many institutions
(see box below). We posed three main questions
about the biological communities at the vents:
1) What is the basis for the origin and concentra-
tion of life at hydrothermal vents?

Early Vent Investigators

The collection of NSF proposals funded in 1978 for the first biologi-
cal expedition to the vents included:

• Physiological and morphological description, estimation of chemosyn-
thetic productivity and rates of microorganism activity Holger Jannasch
(WHOI), David Karl (University of Hawaii). Jon Turtle ( University of Texas),
and Carl Wirsen (WHOI)

• Small-scale distribution and community structure ofbenthic species Fred
Grassle (WHOI), Robert Hessler (Scripps Institution of Oceanography),
and Howard Sanders (WHOI)

• Reproduction, larval dispersal Ruth Turner and Carl Berg (Harvard)

• Genetic studies of population differentiation in molluscs Judy Grassle (Ma-
rine Biological Laboratory)

• Ecological energetics ofmytilids Ken Smith (SIO)

• Feeding relationships of species including field experiments Berg, F.
Grassle, Hessler, Sanders, and Turner

• Age structure and rates of mollusc growth Rich Lutz (Yale), Don Rhoads
(Yale), and Karl Turekian (Yale)

• Effects of temperature, oxygen, and pressure on metabolic rates of 'inver-
tebrates Jim Childress (University of California, Santa Barbara)

• Substrate and cofactor binding, catalytic efficiencies and structural sta-
bility of enzymes George Somero (SIO).

Joining our expeditions were Sandy Williams (WHOI), who collected
new physical measurements on currents, John Edmond (Massachusetts
Institute of Technology), who took chemical measurements of nutrients,
hydrogen sulfide, and oxygen, and scientists of WHOI analytical facili-
ties, who gathered chemical measurements of dissolved and particulate
organic carbon. Dan Cohen (NOAA Systematics Laboratory) studied the
vent fish, and Meredith Jones of the Smithsonian Institution continued
his earlier studies on the vestimentiferan tubeworms. John Baross (Or-
egon State University), who worked on microbiological material from the
1977 cruise, joined the geochemistry leg of the 1979 cruise.

Much to my chagrin, a proposal to study the food chain relationships
of microorganisms was not funded by the NSF — I believe because the
oceanographic application of immunological analyses had not been pub-
lished and was not familiar to reviewers. Perhaps as a consequence, we
are still uncertain about how various major microbiological taxa con-
tribute to primary productivity at the vents. — Fred Grassle

2) How have large organ-
isms adapted to elevated
temperatures, high
pressure, and hydrogen
sulfide?

3) Do the communities
that characterize each
vent represent a) stages
of succession, b) discrete
islands resulting from

; chance immigration, or
= c) organisms settling in
i response to different
| chemical gradients in
\ each area?
| Our three-leg NSF
! program was combined
' with a geological study
supported by the Office of Naval Research and
conducted by Jerry van Andel (Oregon State Univer-
sity), Bob Ballard (WHOI), and Kathy Crane
( WHOI ). The Angus ( Acoust ically NaviGated Under-
water System) team was led by Earl Young (WHOI).
Ballard had negotiated with both the National Geo-
graphic Society (NGS) and Disney Studios concern-
ing a television special. NGS won and, in return for
the privilege of documenting our efforts, provided
funds for five additional Alvin dives.

Total research funds added up to approximately
$600.000, nearly $200,000 of which went to equip
and support Alvin. In the year that elapsed between
submitting the proposal and sailing the following
January, Alvin was extensively modified to augment
its capabilities. NGS agreed to provide a pan-and-
tilt system for a new prototype RCA digital televi-
sion camera, which led to the addition ofAlvin's
much-needed second arm. During 1978, we tested
special lights for underwater filming. WHO Is Cliff
Winget designed a special basket to hold all the
necessary sampling equipment within reach of
Alvin's arms. We built an elaborate multi-valved
system, controlled by the flip of electronic switches,
to pump seawater from any of an arsenal of nozzles
into sample bottles— through either of two sizes of
filters to sample microorganisms, or through a large
hose, so that fragile animals could be gently slurped
into a large, segregated container. We devised an
insulated sampling container; a variety of samplers
for microorganisms and particulate organic matter;
a 35mm stereo camera with temperature probe:
apparatus for incubation experiments; in situ
respirometers to measure the animals' oxygen up-
take and hence their metabolism; fish, crab, and
larvae traps; and a device for deploying plates to
learn how larvae are dispersed and settle. Tetrahe-
drons of thick wire were alternately painted black
and white every 2 centimeters to provide three-
dimensional markers for photographs.

The 14 members of the biology team who sailed

'Vol.41, No. 2 • 1998

on the first \sg were distributed among WHOI's
R/V Lulu (Alviris tender), R/V Gillis, owned by the
Navy and operated by the University of Miami, and
Camelot, a sailing vessel chartered by NGS. When
we arrived in Panama on January 10. Lulu was
already there. Gillis arrived two days later, having
run aground in the Panama Canal. Gillis was so
stuffed with science equipment for subsequent legs
of the cruise that we could not enter the labs. In the
tropical heat, we unloaded several tons of this
equipment onto the dock and then loaded many
tons of our own equipment that we had shipped
separately — with only one case of heat exhaustion,
ably treated by Ruth Turner. Gillis was supposed to
arrive at the vent site two days before Lulu to locate
the vents within a patch ot seafloor about 50
meters in diameter and 2.500 meters deep, some
380 kilometers northwest of the Galapagos Islands.
However, an engine overhaul delayed the ship for
two days and Gillis arrived at the same time as Lulu.
As a consequence, Alviris first two dives had to be
made away from vents while Angus, the 2-ton
towed camera sled, searched for them. Ballard
estimated that it would take at least 30 hours to
locate the vents. The tense moments during a
meeting on Gillis to plan the field
experiments, filmed by NGS cinema-
tographer Jim Lipscomb and shown in
the 1979 Emmy- Award-winning NGS
special Dive to the Edge of Creation,
were genuine, as was the relief when a
few vent clams showed up on the last
few frames of a long series of photos
taken by Angus.

The work on Lulu did not go
smoothly. There was difficulty with
Alviris hydraulic system and dives
often ended abruptly as the alarm for a
hydraulic leak sounded— once before
we even reached the bottom. After
most dives, the hydraulic system was
taken apart and the Alvin crew worked
through the night and into the morn-
ing to fix the submersible in time for
the next day's scheduled dive. Despite
the long hours, the crew's morale remained high,
and the pilots would tease "the most concerned
scientist" of the day by leaving his — neither Ruth
nor Kathy were teased — equipment off the basket
until the last minute.

At a precruise meeting near the end of 1978,
each investigator had provided written notes speci-
fying minimum needs, and these were invaluable in
setting priorities during the first few dives. After six
dives out of a projected 10 on the vents, four at a
vent site dubbed "Mussel Bed" and two at another
called "Garden of Eden," the system that hoisted the
Alvin cradle to Lulu's deck level failed, and we sud-
denly had to start the long trip back to Panama (an

especially long haul for those who had to endure the
none-too-gracious accommodations aboard the
slow-going Lulu).

Even though each dive had been planned as if it
were our last, there was still much to do. No one was
very happy, even though we had accomplished a
significant part of the work. During each dive, we
collected many animal specimens and water and
microorganism samples. We made in situ metabolic
measurements of the animals, placed experiments,
and took many photographs with the stereo close-
up camera or the prototype RCA video camera. We
did have some failures, specifically our current
meter and thermistors, which produced no useful
measurements. The animals collected during each
dive were either sent to G/7//S for physiological
evaluation, preserved, or placed in culture. We were
concerned about the effects of decompression as
the animals were hauled to the surface and by the
long transfer time between ships. We made adjust-
ments: Jim Childress shifted from mussels to crabs
for his laboratory physiological studies after our
minnow-baited traps proved extraodinarily success-
ful at catching crabs. When a specimen of a delicate
orange benthic siphonophore (a relative of the Por-

tuguese man-of-war) fragmented during transit to
the surface, we fashioned on board a new sampling
container to capture one intact.

Each night, we held an intensive debriefing ses-
sion with each day's divers. By asking questions and
recording the conversation, we began to develop a
picture of the ecosystem and what was important.
NGS photographers Al Giddings and Emory Kristof
were full participants in the research and Al's un-
derwater video usually gave us better views of the
animals than were available by direct observation
through the A/i'/'n's small viewports. Even more
important, the video images gave everyone, not only
those in Alvin, an ability to see the animals.

. l/i7//> meter-long
temperature probe

r\l rinU tcmuril .1
community of
u.i l.if In-ill crabs
perched alop pillim
lux a and a dense field
ol mussels.

OCEANUS <

Minnow-baited

(raps proved quite

successful for

capturing crab

specimens lor
physiological studies.

I'.nh I', ill.ii il examines

a Inhcuorm several

meters long, brought

!<• the surface

I>V I/I7H.

Nothing could diminish the excitement of seeing
the animals for the first time. But aboard Alvin there
were few glamorous or even quotable exclamations,
as each diver struggled to record everything that
happened, operate cameras, and work with the pilot
to collect samples. Transcripts of Alvin tape record-
ings often consisted of dispassionate strings of time
marks, temperature readings, and names of animals.
NGS's Jim Lipscomb did provide a good record of
enthusiastic first impressions, such as Bob Hessler's
excited commentary as the first person to dissect
one of the 2-meter-long Riftia tubeworms: The
animals I usually work on are about this long, he
said, holding two fingers almost together. "To be
working with these things, where you could use a
knife and fork and spoon for dissection, is abso-
lutely remarkable!"

When we returned to the Galapagos vents for
Leg 2 on February 9, the research shifted to Ballard's
geology program. But Bob made room for some
biology on his five geology dives, and the five NGS-
funded dives emphasized biology. These 10 dives on
the second leg made a huge
difference to the biology
program's overall success.
When we left the ship on
February 26. NGS photo-
graphic specialist Pete
Petrone, who had discov-
ered how to process color
film at sea using seawater,
provided each scientist
with 20 of the best 35mm
photographs taken on the
cruise, and Emory Kristof
supplied a tape of underwa-
ter video highlights, for
scientific use only, until
after Dive to the Edge of
Creation aired. Eleven addi-
tional dives at the begin-
ning of December 1979

completed experiments begun in the first two legs
and added to the sample collections.

By the end of 1981, the 1979 Galapagos dives
had yielded 32 papers. My lab alone sent informa-
tion and materials to more than 40 additional
investigators. By 1986, the 1979 expeditions to the
Galapagos vents had contributed to 79 papers. By
then, most of the papers included information
from vent systems explored by 1982 expeditions to
the East Pacific Rise at 21°N and to Guaymas Ba-
sin. The discovery of 350°C+ hydrothermal fluid
pouring from "black smoker" chimneys in 1979
added a new dimension to hydrothermal vent
biology: a tremendous diversity of microorganisms
including Archaea growing above 100°C, and
Alvinella pompejana worms living at temperatures
up to 50°C. The Guaymas vents, spread over a large
area of soft sediments, provided yet another set of
habitats for new forms of hydrothermal vent life.
My greatest regret from these expeditions is that
we never had sufficient time to explore the full
diversity of life in each area. Basic exploration for
smaller organisms and new species took second
place to more focused research objectives.

The Galapagos investigators completed most of
their work and at least partially answered their
original questions: Life at vents was distributed
primarily according to the flow and composition of
hydrothermal fluid, so that competition and preda-
tion among the vent animals were less important
than among similarly dense assemblages of animals
living in shallow waters. Despite the high pressure
and low temperature of the deep, the respiration
rates of vent species were generally comparable to
those in shallow-water animals. High concentra-
tions of blood pigments, such as hemoglobin, were
found in crabs, mussels, clams, and Riftia pachyptila
tubeworms. compensating for the low partial pres-
sure of oxygen at depth, and, incidentally, giving the
animals with hemoglobin
their vivid red hues. Karl
Turekian's Yale team
showed that clams grew
large and fast— 22 centime-
ters in 10 years. Ken Smith
demonstrated that mussels
transplanted close to the
vents increased their respi-
ration and growth rates and
stopped growing when
moved away. All signs
pointed to a highly produc-
tive ecosystem sustained by
the ephemeral, high-energy
flow of hydrothermal fluid.

In his experiments,
Holgerjannasch filled
1 syringes containing radio-
? active carbon dioxide com-

' Vol. 41, No. 2« 1998

bined with vent fluids and incubated them over two
days, proving that deep-sea bacteria used hydrogen
sulfide from vent fluid to convert carbon dioxide
into organic carbon, which they incorporated into
their cells. We had not anticipated the presence of
these chemosynthetic bacteria in the tissues of
Riftia pdchyptila, clams, and mussels. Discovery of
these symbionts explained why these animals grew
rapidly to a large size. A surprising variety of free-
living chemosynthetic microorganisms was identi-
fied, and. per unit area, vents were found to be
among the most productive ecosystems known. A
central mystery— how larvae were
able to find new vents nearly every
generation— was partially answered
by the early genetic study of mus-
sels: Juvenile mussels were geneti-
cally quite different from older
classes, indicating larvae arrive in
pulses from distant vents.

Although the diversity of species
was low in comparison with other
deep-sea environments, a remark-
able number of new genera, fami-
lies, and subfamilies was described
from the samples collected. In
general, life in the deep sea remains
grossly undersampled and poorly
known, and the mid-ocean ridges
are not an exception. After many
more expeditions in the two de-
cades since 1979. biological studies
have been conducted at 31 vent
sites, but the distribution and habi-
tats of most species are still not
adequately described.

In a review of the literature through 1998 that
appeared in Advances in Marine Biology. V. 34,
Verena Tunnicliffe (University of Victoria), Andrew
McArthur (Marine Biological Laboratory) and
Damhnait McHugh (Harvard University) found that
75 percent of the 367 species and 40 percent of more
than 200 genera of animals known to occur only at
hydrothermal vents have been found at only one
hydrothermal site. This narrow biogeographic dis-
tribution of most of the vent fauna is intriguing,
even though the unevenness of sampling and the
great distances between sampled sites make any
conclusions about fauna! distribution uncertain.
Many species and genera remain to be discovered
from hydrothermal vents. Plankton sampling, fur-
ther application of molecular techniques to identify
and track cohorts of larvae settling at vents, and
better understanding of the physical processes
associated with vent plumes are making the pro-
cesses transporting larvae to newly formed vents
less mysterious (see article, page 6). and rates of
dispersal are beginning to be estimated. The study
of vent microorganisms — on surfaces, below the

seafloor, in the water column, and in association
with other organisms — remains an important fron-
tier. The diversity of microorganisms and their
functional roles in the ecosystem are not yet well-
described, despite more widespread sampling and
the application of molecular techniques.

We have subsequently learned that each seg-
ment of mid-ocean ridge has its own characteristic
spreading rate and history of volcanic and tectonic
activity, which lead to different patterns of hydro-
thermal flows over time and space, and to differ-
ences in chemical composition at various vent

sites. This information provides the template for
understanding processes controlling the evolution
of vent fauna.

The cycle of birth and death of individual vents
and vent fields and their spacing along the mid-
ocean ridge are not well-understood. Along with
increased exploration of new sites, the next few
years bring the hope for one or more seafloor obser-
vatories at hydrothermal vents. By continuously and
simultaneously observing the biological, geological,
chemical, and physical processes occurring at these
sites, we can learn how all these processes interact.
The greatest legacy of the first vent studies has been
the collaboration of scientists from the many disci-
plines of ocean science to learn how a previously
unknown ecosystem functions. This will be the
model for discovery as other worlds are explored.

A summer in a WHOI geology laboratory following his sophomore
year m college started Fred Grassle's career studying life on the
seafloor. After 20 years on the WHOI scientific staff, he started a
new Institute of Marine and Coastal Sciences at Rutgers Univer-
sity. Administrative duties brought an end to his hydrothermal
vent cruises, but he still manages time for studying life on the
shelf and deep-sea bottom off New Jersey.

\ respirometer
measures (lie o\v»en
uptake ol mussels
and hem r Ihi-ii
metabolism rales.
Scientists found thai
despite the high
pressure and lm\
temperature of the
deep. the respiration
rales of vent animals
were general!)

I I Mll| 1.1 I .ll llr tO

those in sliallou
water animals.

To learn In m

deep sea vcnl species

disperse through the

ocean and coloni/.e

III'W1 liullnlllrMli.il

sites, researchers in

the I.ARVK Project

are i m esl ii*al i 111; t he

complete lile cycles of

several species,

including three in the

photo above: the

Itihcunrm /.'//////

pachyptUa, the vcnl

crah B\ thugrum

thermydron,and

the innssel

lititlninoilitilii*

llu'rillll/lllilll'..

\ sMilaclic loam lloal
marks the location <>

an experimenli

Mm k amid a colon

ol (he I uliruiii i

RiftiapcuJiyptili

Deep-Sea Diaspora

The LARVE Project Explores How Species Migrate from Vent to Vent

Lauren Mullineaux

Associate Scientist, Biology Department

Donal Manahan

Associate Professor, Department of Biological Sciences, University of Southern California

When spectacular biological communities
were first discovered at hydrothermal
vents in 1977, biologists puzzled over
two main questions: How did these oases of large
and abundant animals persist in the deep sea,
where food is typically scarce? And how did these
unusual species, which occur only at vents, manage
to colonize new vents and avoid extinction when
old vents shut down?

Efforts to solve the first question have resulted

in fascinating insights into the chemosynthetic
microbes that form the base of the vent food chain,
as well as the physiological adaptations that allow
the tubeworms, bivalves, and crabs to thrive. The
second question, however, has largely remained an
enigma. We do know that most of the species prob-
ably disperse between vents in a larval stage that
drifts through the water, but we know very little
about how this process works. Over the past two
decades it has become clear that unraveling this
mystery will require an interdisciplinary and coop-
erative scientific approach. The LARVE (Larvae At
Ridge VEnts) Project was created to provide just
such a collaborative framework for investigators
from many different disciplines.

LARVE is part of a larger program called RIDGE
(Ridge Inter-Disciplinary Global Experiments),
which is funded by the National Science Foundation
to provide a coordinated, interdisciplinary research
program aimed at understanding the geology, phys-
ics, chemistry, and biology of processes occurring
along the global mid-ocean ridge system. At
present, 10 principal investigators from eight differ-
ent universities and research institutions are work-
ing in the LARVE team, and others are expected to
join over the next few years.

The goal of the LARVE Project is to investigate

•Vol.41, No. 2 • 1998

the processes by which larvae disperse through the
ocean from one vent system to another. The LARVE
Project aims to understand how these processes
determine how far and wide species can migrate,
how new vent biological communities are estab-
lished, and how their populations become increas-
ingly but differentially diverse. Answering these
questions requires expertise in a variety of
fields, including biology, chemistry, geol-
ogy, and physical oceanography. The
links among these fields are illustrated
by following an example species, the
vestimentiferan tubeworm Riftia
pachyptila, through different stages of
its life cycle.

Riftia pachyptila reproduce by releasing
eggs and sperm into the water. The fertilized egg
then develops into a larva called a trochophore. The
length of time a larva can survive in the water de-
pends on its physiology — that is, how much lipid it
has and how quickly it metabolizes this stored
energy. Although the larvae use their cilia for slow
swimming, their horizontal motion is determined
largely by hydrodynamics, including fluid flows near
the vents and larger-scale oceanic circulation pat-
terns. However, larval behavior, such as vertical
swimming, can position the larva in different flows
and cause them to disperse at very different speeds
and even in opposite directions.

Larvae that survive the perils of predation and
starvation in the water column must still locate a
suitable vent habitat for settlement— possibly by
responding to a chemical
cue at a particular vent
site. After a larva settles
into the vent environ-
ment and metamorpho-
ses into a juvenile stage,
its survival is controlled
by its physiological toler-
ances, nutritional require-
ments, and interactions
with other tubeworm
species, mussels, crabs,
and fish that may be
competitors or predators.

Successful colonists
contribute their genes to
the population, and if
larval exchange between
two populations is fre-
quent, this gene flow
keeps the populations
genetically similar. If,
however, larval dispersal
between two popula-
tions becomes inhibited,
the populations may
diverge genetically over

time, and speciate (evolve into separate species).

For their initial studies, LARVE Project research-
ers chose a site near 9°50'N, 104°17' W along an axial
summit valley of the East Pacific Rise. The site is
ideal because it consists of a chain of vents that
support diverse and abundant communities of vent
species. The bathymetry of this section of the ridge
is well-surveyed, and the geochemistry of the
vents has been monitored since 1991, when
a well-documented volcanic eruption at
the site wiped out the existing commu-
nity and created a blank slate for a new
community to form. Researchers are
concentrating on the chain of vents in this
area to study how species reproduce,
progress through larval stages, settle at vents,
and survive. Larger-scale studies of the gene flow
and physical oceanography are conducted along the
East Pacific Rise and elsewhere.

We presently know a little about stages in the life
cycle of a few different vent species, but we do not
have a comprehensive picture of how extensively, in
both distance and number, any single species can
disperse. LARVE researchers have targeted several
species for detailed investigations of the species'
entire life cycle: how they reproduce, how the larvae's
behavior and physiology interact with processes that
transport them physically, how larvae settle at new
vents in response to chemical or physical cues, how
larvae survive in the complex chemical and biologi-
cal environment of hydrothermal vents and how
these processes contribute to gene flow. Selected

A magnified image
slums .1 'J'J < l.n nlil
microscopic embryo
ol Ihc lnbo\vorm
Riftia pachyptila. Its
actual diameter is
ahmil 100 microns
(.1)001 motors).

Adam Marsh. University of

The- lift- cycle- ufltiftia
liinli\i>lilii illusl rales
the complexity of
factors— involving
biology, chemistry,
geology, and physical
oceanography — that
all play rules in ill"
migration, resettle-
ment, ami special inn
ol \onl species.

OCEANUS '

Newly designed
high-pressure

system** lor enduring

M-nl specie's' lanac

ullon scientists lo

examine theheliinior.

energy stores, and

metabolic rales ol

lanac. uhicli all

contribute to their

ability In siir\ivc

migration to ncu

vent sites.

A venl colon) ol the

scrnnlid poluliaclc

/ afninatttbus nlvini

thmcs near a

llMlllllllrMII.il M III

species include two
tubeworms (Riftia
pachyptila and the
smaller Jericho worm,
Tevniajerichonana), the
vent crab Bythograea
thermydron, the mussel
Bathymodiolus
thennophilus, and several
\ species of gastropods.

The major challenge
; of the project is adapting
techniques used to study
shallow-water communi-
| ties to the remote, deep-
| sea vent system, with its
I extreme pressures, high
I temperatures, and inhos-
: pitable chemistry. The
I field studies require
I delicate in situ manipu-
i lations by submersible,
t. as well as recovery of
: live, undamaged speci-
mens for shipboard experiments. The laboratory
studies require incubating and observing larvae in
chambers that must be maintained at ambient
deep-sea pressures with precise control over the
thermal and chemical environment.

One of the more novel applications for laboratory
studies of larvae is the design of two high-pressure
systems for culturing larvae (photos above left). The
system for behavioral observations consists of a
cylindrical chamber outfitted with a viewport and
illumination system. The chamber for physiological
experiments is a smaller flow-through system that
allows researchers
to introduce fluid
containing chemi-
cal tracers or to
extract fluid and
larvae for experi-
ments and other
metabolic measure-
ments. Using these
systems, investiga-
tors can explore
questions about the
energy stores and
metabolic rates of
larvae, and about the behaviors that may affect their
hydrodynamic transport.

Currents near mid-ocean ridges are affected by
both ridge topography and buoyant plumes of hydro-
thermal fluids flowing from the vents. Preliminary
studies using current meters moored along ridges
indicate that currents are "steered" to run parallel to
the ridge. Flow speeds, however, can be much faster
at some heights above the ridge than others, and

currents on the two ridge flanks may be oriented in
opposite directions. Therefore, small movements by
larvae, up or down, can have a substantial impact on
the direction and speed of their dispersal.

The hot, buoyant fluids flowing from the vents
further complicate the currents by forming a plume
that rises several hundred meters above the ridge.
The earth's rotation interacts with the buoyant
plume to create a circulation cell within the plume
that keeps some of the plume water near the source
vent. If a larva becomes trapped in this spinning
buoyant plume, it may recolonize the vent where it
was spawned. To characterize larval transport in the
specific flows near the 9°N East Pacific Rise study
site, current meters have been deployed along the
ridge axis, and future studies using neutrally buoy-
ant floats are being planned.

One type of larva, the late stage (or rnegalopa)
larva of the vent crab Bythograea, may not need
high pressure for study of some of its behavior. This
astonishing larva has been kept alive for more than
10 months at surface pressure by Chuck Epifanio,
Anna Dittel. and Craig Gary at the University of
Delaware. During that time these researchers have
been able to observe individuals metamorphosing
into juveniles, providing a potential system for
investigating how larvae may settle at potential
vent sites in response to physical or chemical cues.
Unfortunately, this pressure tolerance is uncom-
mon in most vent species, and settlement cues for
other species will have to be investigated under
pressure, using laboratory culture chambers or in
situ field experiments.

Field experiments to explore how larvae settle at
new vent sites and how juveniles survive are con-
ducted with hundreds of simple, inexpensive basalt

blocks deployed by
the submersible
Alvin, operated by
Woods Hole
Oceanographic
Institution
(WHOI).The
blocks are placed
within and away
from vigorously
venting sites for
various periods of
time to understand
the effects of vari-
ous environments and neighboring colonists on the
survival of settling larvae. Some blocks are pro-
tected from crabs and fish by cages, and others have
partial cages as controls, to examine the effects of
predators on colonists (see photos, opposite). Large
numbers of vent species' larvae settle onto the
blocks in a matter of months. The experimental
design has been adapted to the deep-sea vents from
classical studies of the rockv intertidal environment

8. Vol.41, No. 2- 1998

by investigators at WHOI (Lauren Mullineaux),
Pennsylvania State University (Chuck Fisher and
Steve Schaeffer), and the University of North Caro-
lina (Pete Peterson and Fiorenza Micheli).

By necessity, the experimental approach to
questions of larval settlement and survival is con-
ducted on small areas over short time periods. But
obviously it would be valuable to make observa-
tions over many years. Time-series observations of
undisturbed faunas are rare in the deep sea, but a
1 -kilometer-long vent "sanctuary" was established
after the 1991 East Pacific Rise eruption, allowing
Rich Lutz's lab at Rutgers University to monitor
changes in vent communities for almost a decade.
This monitoring is done with time-lapse cameras
left at individual vents and repeated camera sur-
veys conducted with Alvin and the remotely oper-
ated vehicle Jason.

Studies of genetic exchange between vents
started soon after vents were discovered and con-
tinue along the East Pacific Rise and elsewhere in
the world's oceans. Studies on the species targeted
by LARVE are currently under way as part of the
effort by Bob Vrijenhoek's group at Rutgers to un-
derstand genetic exchange on a regional and global
scale. The intent is to identify vent populations that
have significantly different genetic compositions
and to characterize how these populations are
separated geographically. Are they separated by a
few kilometers between neighboring vents? Or by
gaps between segments of the ridge? Or by an ocean
basin or a continent? This information will then be
compared to predictions, based on the larval stud-
ies, of which geographic features may pose barriers
to dispersal and, therefore, gene flow.

The first cruises specifically associated with the
LARVE project went to sea in 1997. and about 10
cruises are expected to visit the East Pacific Rise
through the year 2000. The field observations and
data are essential ingredients for subsequent theo-
retical studies that will lead to a cohesive under-
standing of the complex processes that allow such a
wealth of life to thrive on the bottom of the ocean.

MM* »

» '« i .'>

Funding from the National Science Foundation and support
from the RIDGE office helped the LARVE Project metamor-
phose from idea to reality.

Lauren Mullineaux first became interested in the ecology of
hydrothermal vent systems during the early 1 980s as a graduate
student at Scripps Institution of Oceanography in Robert
Hessler's lab, a hotbed of pioneering work on vent communities.
He In'isely) counseled her to choose a less risky project for a
thesis, but the fascination with vents was solidly in place. After
arriving at WHOI she jumped at the chance to study larvae of
vent species during a cruise along the Juan de Fuca Ridge with
Ed Baker (NOAA) and Peter Wiebe (WHOI). Now. many cruises
and AJvin dives later, she still gets a thrill from visiting vents in
person, but an even greater satisfaction from witnessing the
inspiration and excitement of another scientist on his or her first
sojourn to a vent.

Born and raised in Ireland, Donal Manahan became interested
in marine biology and the study of larvae as an undergraduate
working on oyster culture in Ireland. As a graduate student at the
University of Wales in the UK, he studied larvae metabolism,
specifically questions such as how much energy larvae need to
grow and how they get that energy from the ocean. He came to
America in 1 980 as a postdoctoral researcher at the University of
California. Irvine, He has expanded his interests to study how
animal life develops in extreme environments, such as Antarc-
tica. That naturally drew him to investigate the early stages of
animals that survive being cold and hungry at another similarly
extreme environment, hydrothermal vents. Although the vents
themselves are very hot. the seawater just a few feet away is very
cold, and like polar regions, has no obvious source of food.

wander amid a

thriving colony ol

Itflllll WW//W//.S

thfrini>i>liilii\

St irnl isl s h.i\ r

deployed hundreds ol
l>.i--.ill blocks (Iff!)
within and away from
xcnt .11 r.i - lor >arions
periods of time to
l< .11 n lum lurvuc

-rltli- .il these sites.

Some Mocks arc
protected In cages
(center and right I lo
delerniine I he effects
ol predators on
colonists. At right,
a block has been
coloni/ed In
\esliinenlileran
I ul K-U 01 in-- alter
i > months on
I he sea floor.

OCEANUS

Life on the Seafloor
and Elsewhere in the Solar System

If Volcanoes Plus Oceans Can Support Life at the Vents,
Why Not on Other Planetary Bodies?

John R. Delaney

Professor of Oceanography. University of \Vashington

A mi i>. i u of digital

images shows the

-nUnlr chimney

"Roane" before it \\a>

retrieu'd from the

seafloor in 1998.

I In- llmil- \i nli il
h\ I'.i .,1 1 ii -i 1 1 >| n ii I r, I

a \arirty of

inn liiinu, ini-in-

The RIDGE program (Ridge Inter-Disciplin-
ary Globe Experiments) was sharply focused
on the global spreading center system, but
the program's goals were broadly defined. RIDGE
was designed to explore the causes, consequences,
and linkages associated with the physical, chemical,
and biological processes that transfer mass and
energy from the interior to the surface of the planet
along the mid-ocean ridges.

That broad mission statement left a lot of room
for the "I" in RIDGE. But one lesson of RIDGE is
that truly interdisciplinary research is difficult to
do. It is difficult to get funding for interdisciplinary
research. It is difficult to overcome the language
barriers across a spectrum of fields ranging from
molecular biology to seismology. Nevertheless, the
fields are related, and that is another lesson of
RIDGE: Scientists must keep an eye on adjacent
fields. Increasingly, a number of researchers believe
that communication across
fields will spawn some of the
major scientific discoveries in
the coming decades.

With that perspective, the
RIDGE program might be
considered a work-in-progress.
Perhaps it is best viewed in the
context of events that hap-
pened before it was initiated
and activities that may take
place as RIDGE evolves. \Ve
often look back in order to
look forward, and looking
back to the late 1970s, two
major voyages of discovery
occurred within months of
one another. One voyage was
to the bottom of the ocean
and the other to the far
reaches of the solar system.
These discoveries set the stage
for RIDGE.

In 1977, scientists using
Alvin dove to the ocean bot-
tom near the Galapagos Is-
lands. They found evidence of
\ volcanism. but. unexpectedly.
I thev also discovered a lush

from Iheseafl

ffleochemical

,h

Mill II

idechimnej "Roane" is retrieved
u pedition in Julj I'Wsledhj
sit

id

th

of Washington. Studio of
licrohial communities \\ilhin
• I the chimneys m.n shed li^ht
nil the possibility ol lite on

thehol.siiHul •
on the ori>;in ( I
other planetary hodies.

animal community. An abundance of crabs, clams,
mussels, and worms tipped with brilliant red
plumes was found thriving near zones of active
seafloor volcanism. Until then, the seafloor had
been regarded as relatively barren terrain, where the
only nutrient sources were scant dregs that perco-
lated down from the sunlit surface. Distilled to its
essence, this first discovery was that volcanic activ-
ity in the presence of liquid water can support life
\vithout sunlight.

Shortly after that Alvin dive, the I 'oyager I space-
craft set off to explore the outer solar system. Eigh-
teen months later, it had reached Jupiter and, like
Ah'in. found something unexpected. As Voyager I
flew past lo, the innermost moon of Jupiter, it photo-
graphed nine ongoing, simultaneous volcanic erup-
tions ejecting material hundreds of kilometers above
the moon's surface. Quite unlike the earth's long-

10 -Vol. 41, No. 2- 1998

dead moon, lo turned out to be the
most volcanically active body in the
solar system. The second fundamental
discovery, simply put. was that there
were more volcanoes in the solar
system than we had ever dreamed.

Two ideas, nearly 20 years old:
Undersea volcanoes can support life:
there are many volcanoes in the solar
system. Only now are these two ideas
beginning to interact with each other
to guide exploration for life in outer
space. But these ideas remained
worlds apart when the RIDGE pro-
gram was launched in the 1980s.

A primary goal of the RIDGE pro-
gram was to look at events taking place on ridge
crests in real time, rather than simply to map the
products of those processes. From 1986 to 1990,
Robert Embley and Edward Baker, from the Na-
tional Oceanic and Atmospheric Administration
(NOAA) Pacific Marine Environmental Laboratory
(PMEL) in Seattle, Washington, and Newport,
Oregon, identified likely volcanic activity associ-
ated with large water-column plumes along the
southern end of the Juan de Fuca Ridge about 180
miles off the Pacific Northwest coast. But the link
between the large plumes and an active eruption
was mostly an insightful inference that could not
be confirmed by direct observation. At the time,
scientists on cruises involved in the area did note
the presence of flocculated bacterial mats in the
vicinity of fresh seafloor lavas.

In April 1991, researchers exploring the East
Pacific Rise off the coast of Mexico found them-
selves in the right place at the right time. The cruise
was led by Rachel Haymon of the University of
California, Santa Barbara, and Daniel Fornari of
Woods Hole Oceanographic Institution, who lis-
tened raptly as divers aboard Alvin reported near-
whiteout conditions at the dive site. Clumps and
clusters of what seemed to be bacterial products
billowed up in huge plumes from beneath the sea-
floor. Through the bacterial haze, they saw
tubeworms newly barbecued and folded in fresh
lava. The researchers quickly drew the conclusion
that the gods of serendipity had delivered them into
the midst of either an active or a very recent erup-
tion. These experiences substantially enhanced the
resolve of RIDGE and NOAA vent scientists to pur-
sue active events on ridge crests in real time.

With the end of the Cold War. oceanographers
gained an invaluable tool: limited access to selected
data from the Navy's classified SOund Surveillance
System. SOSUS was an array of hydrophones origi-
nally designed to detect enemy submarines. On
June 26, 1993, just four days after gaining access to
some SOSUS data, Christopher Fox and colleagues
at PMEL in Newport detected a series of earth-

quakes along the Juan de Fuca Ridge. In two days,
the locus of earthquake activity had migrated 50
kilometers northward along the ridge axis.

The opportunity to catch a ridge eruption in
flagrante finally had presented itself, but taking full
advantage of the windfall was not automatic. Some
in the scientific community still had doubts about
this type of activity and viewed the effort as "fire-
engine chasing." Unlike the pursuit of fire engines,
however, it is much harder to find ships (and funds)
to pursue a possible eruption in the middle of an
ocean. To arrive on site within days or weeks of an
eruption was not easy. It meant persuading scien-
tists to yield some of their hard-won ship time,
usually scheduled more than a year before, to chase
a possible chimera and either curtail or ignore their
own funded research. Such behavior in the competi-
tive, peer-reviewed world of oceanographic research
could jeopardize longer-term funding for more
predictable, if less exciting, science.

Nevertheless, within days, following a shore-to-
ship call from Fox, Richard Thomson of the Institute
of Ocean Sciences in Sidney, British Columbia, a
member of the RIDGE Steering Committee, briefly
redirected an ongoing cruise operating nearly 120
miles to the north. Thomson and colleagues discov-
ered a major hot-water plume directly above the site
where the migrating earthquakes had stalled. Embley
and his NOAA colleagues, Baker and Bill Chadwick,

lo, .1 moon ol Inpilcr
(above), where sur-
prisingly active
volcanism was de-
led cil hy NASA's
Voyager I probe
(lell tin l')7<).

In the late l')7(ls. h\o
unique research
vehicles. \\IIOIs

l/i m I lirlim i .ind

NASA's Im/ii'rr /
i.ilimn. made sepa-
rate lull complemen-
tary discoveries ill
inner and outer
spare, llv discovering
thriving biological
eoniniiinilies near
seafloor volcanic
vents. , \lvin helped
slum that u.ih i plus
volcanism can sup-
port lilcMillionl
sunlight, \in-agrrl
discovered active
Milcanism on Jupiter's
moon lo. sliouini;
that volcanism was
more common in the
solar system than
previously believed.
These two ideas have
only recently com-
bined to "iiidc
scientists' search
lor life on other
planetary bodies.

OCEANUS- 11

Ill I'I'll l/imJmr

onto a recently

erupted \nli .HIM

zone along the Hast

r.n il i» Kisc in-. n

I0°\, where scientists

found In sh I. n. i .in<l

huge- billowing mats

.nil I I Illslri -. 4>|'

bacterial products.

A sini|ilihrd \crsion

ol the |»h\ lo^rnrt n

Ircc created by C arl
\Voose (I nivorsilvol

Illinois) sIlOUS lilt'

three domains of life:

I'.. n in i.i. I in .11 i.i

and Archaca. In !')'):{.

scientists discovered

massive outpourings

of hyperthennophilic

Archaca mi erupting

mill in c. Ill ri<li>es.

also reacted within 10 days and redirected a long-
planned cruise to what became known as the
CoAxial Segment of the ridge. Using the Canadian
vehicle ROPOS (Remotely Operated Platform for
Ocean Science), Embley and colleagues found a fresh
lava flow that already had been colonized with bright
yellow mats of bacteria. Farther south, they discov-
ered massive amounts of bacterial products surging
from beneath the ocean floor, in a fashion strongly
reminiscent of the East Pacific Rise event in 1991.

As these reports were relayed on shore, two
University of Washington (UW) colleagues, Paul
Johnson and Russell McDuff, and I worked with
dispatch to request support from the National
Science Foundation for RIDGE funds to explore
the new eruptions with Alvin at the earli-
est possible time. Our argument was
that for the first time ever we could Animals plants

approach a seafloor event with the full \ Fungi

knowledge of what was taking place,
when it began, what its extent was.
and how it was unfolding.
The submersible is
nearly always Bacteria

booked, but there

o
was an opening in ^f^

October, four long

months after Fox

first detected the

telltale seismic

activity. By the

time we put to sea

with a scientific crew

comprising both academic and

NOAA researchers, we were already

building on the work of many people: Embley. Baker,

Haymon, Fornari, Fox, and Thomson. Then two

others, Fred Spiess and John Hildebrand of the

Scripps Institution of Oceanography, University of

California, San Diego, contributed, on very short

Aquifex

Thermotoga

Unknown common
ancestor

notice, an eleventh-hour sonar mapping program of
the eruption area that helped guide our Alvin dives.

We found vast volumes of microbial material
issuing from the seafloor, as had been observed
before. But because of experience on previous expe-
ditions, microbiologists from John Baross' labora-
tory at UW were on board to culture bacterial
samples that we collected. Then-graduate-student
Jim Holden established that the volcanic microbes
belonged to an ancient group of organisms that are
so genetically distinct that they represent a separate
and fundamental branch on the tree of life distinct
from bacteria, plants, and animals. They are known
as Archaea, or ancient ones, and they are
hyperthermophilic, meaning they survive, in fact
thrive, at temperatures greater than 90°C. We don't
know yet whether the eruptions flush pre-existing
microorganisms inhabiting the subseafloor or
whether the eruptions release nutrients that trigger
a volcanic microbial bloom that simply overflows
the available space in seafloor rocks. But it was clear
that, as hypothesized earlier by Baross and his col-
league Jody Deming of UW, the rocks below the
seafloor are populated with microbial communities
that we need to explore.

Discovering the outflow of Archaea along erupt-
ing spreading centers has been one of RIDGE's
major successes. It required long-term commitment
and cooperation of many scientists in the face of
considerable difficulty. Since the Coaxial Event, a
number of similar rapid responses have been
mounted to document these types of events, and in
each case, high-temperature microbes have been
cultured from the effluent associated with
the volcanic-tectonic activity.

These observations are giving rise
to a potentially controversial hy-
pothesis: The brittle outer shell of
any volcanically active, water-satu-
rated planet may harbor a poten-
tially vast subsurface
microbial biosphere.
Indeed, this deep,
hot, chemically rich
environment within
the earth's volcanic
shell appears to
have the necessary
ingredients to foster
the critical reactions
e*-vv that could have created

the building blocks of life. In
a 1983 paper, Baross, Sarah
Hoffman, and Jack Corliss, all then at
Oregon State University, first suggested that life
may have originated on the earth not in warm,
shallow pools struck by lightning, but rather in
sunless, deep-sea volcanic vents. And if such an
evolution can occur on the earth, perhaps some-

Desulfurococcus

Pyrodictium

Methanopyrus

Methanococcus

Sulfolobus

-—
Thermoproteus

Pyrobaculum
Pyrococcus

Methanothermus
Archaeoglobus

12 «Vol. 41, No. 2. 1998

Rocky Interior

Rocky Interior

thing similar has occured elsewhere in
the solar system where active volcan-
ism and liquid water are juxtaposed.

One of the most intriguing solar
bodies likely to support such life forms
may be ID'S nearest neighbor, Europa —
lupiter's second moon and one of the
most beautiful planetary bodies in the
solar system. It has a smooth, highly
reflecting, almost pearl-like appear-
ance from a distance because its sur-
face is completely covered with ice and
nearly devoid of large craters. Recent
close-up images of Europa's surface
taken from the Galileo spacecraft
display a more chaotic texture, which
has been interpreted by the Galileo
imaging team as a cluster of blocklike
icebergs. The tops of these striated
blocks appear to be "floating" about
200 to 250 meters above a surrounding
slurry of much smaller ice fragments,
much the way a large ice cube would
float in a slush of crushed ice. The
height of the ice block suggests that
only one-ninth of its bulk projects
above the surface. That implies that
the blocks, which clearly have floated away from a
solid ice "shoreline," are about 3 to 5 kilometers
thick. With a calculated bulk density of nearly 3.0
grams per centimeter, Europa must have a rocky
interior, and the best estimates indicate that the
water layer (whether liquid or solid) above the more
dense rocky material is about 100 kilometers thick.
If that is true, a relatively thin outer layer of ice may
float atop a much thicker layer of liquid water,
which directly overlies Europa's higher-density
interior. In other words, there may well be another
ocean in the solar system, and it may be main-
tained by volcanism within the rocky interior.

Some scientists speculate that to maintain a
liquid water body on a frigid satellite that is slightly
smaller in diameter than the earth's moon, Europa,
like its neighbor lo. may harbor volcanic activity
within. In Roman mythology. Jupiter, king of the
gods, and the maiden Europa conducted a torrid
affair. In a modern scientific parallel, the tidal rela-
tionship between the huge planet and its diminutive
moon creates significant heat. Jupiter's enormous
gravitational embrace, combined with the resonant
interplay of nearby satellites, alternately squeezes
and stretches Europa, generating internal friction
and possible volcanism, as it clearly does on lo. The
possibility of a planetary body hosting both an
ocean and submarine volcanoes makes similar
systems on the earth useful analogs for searching
for life elsewhere in the universe. By designing inno-
vative strategies to learn more about the relation-
ships between volcanoes and life here on Earth, we

Metallic Core

Cold, Brittle Surface Ice

Water Layer

Metallic Core

Warmer, Convecting Ice

Ice Covering

Liquid Ocean Under Ice

Water Layer

not only learn a great deal about how our own
planet functions. We also gain valuable new insights
into how to approach similar systems in our solar
system and beyond.

John Delaney completed a degree in geology at Lehigh Univer-
sity in Bethlehem. Pennsylvania, before going to the University of
Virginia for a masters degree. As an ore deposit geologist in
Maine, he became fascinated with processes that concentrate
metals. Gravitating to the heart of the copper mining industry, he
searched for base and precious metals in Colorado. Utah. Ne-
vada, and Arizona, while studying economic geology at the
University of Arizona in Tucson. After six months living in and
working on active volcanoes of the western Galapagos
Islands, he decided to study active volcanism for the rest of
his life, and completed a dissertation on submarine
volcanic gases.

His research and teaching have focused on active
submarine volcano-hydrothermal systems along the
global spreading-center network. In 1 980 a unique set of
rocks from the Mid-Atlantic Ridge recovered aboard
Alvin provided clear evidence that seafloor fracturing and
mineral deposition was identical to quartz veins beneath
massive copper-iron sulfide deposits on land — bringing
Delaney full circle to his original geological interests. The
recognition that submarine volcanic gases provide an essential
nutrient source for the microbial communities that are the base of
the chemosynthetic food chain at ridge crests took another of his
original research pursuits in an exciting, unanticipated direction.

Delaney enjoys the poems of Imiku poet Matsuo Basho ( 1 644-
1694) — a master at capturing the essence of an experience in
very few words:

Breaking the silence of an ancient pond,

Afrogjumped into water.

Deep resonance!

Wliether the pond is only a pond, or the pond is a mind and the
frog is an idea, is left to the reader. In many ways, the simplicity
and elegance of such a distillation is akin to what scientists strive
to extract from their observations.

These drawings
depict two proposed
models ol'lhe subsur-
face si rui-lii re of
Jupiter's moon
I in op.i \o conclu-
sive proof hus yet
been found ili.il an
ocean exists on
Kuropa. hill geologic
features on its sur-
face, imaged by
NASA's (inlileo space-
craft, might he ex-
plained either hy the
existence of a warmer,
convecling ice layer,
located several
kilometers helow a
cold, hrillle surface
ice crust ( lop model ).
in In . i layer of liquid
water with a possible
depth of more I han
100 kilometers
(bottom model). If an
ocean I (H) kilometers
(or 60 miles) deep
existed below a
I iiiii|i.in ice crust 15
kilometers ( III miles)
thick, it would be II)
limes deeper than
any ocean on the
earl hand would
contain twice as
much water as I he
earth's oceans and
rivers combined.

The icy exterior of
I in o|i.i. a moon of
Jupiter, may be hiding
a deep ocean and
interior volcanism —
two necessary ingre-
dients for life. The
gravity Held on
Buropa is about
one-seventh thai
of the earth.

NASA Voyager Image

OCEANUS-13

SS in Wonderland

v:

•* /

^' 4

•Kf*K

ALISS (Ambienl

I i^lll Ini.i^iii^.iiiil

Spectral Syslcin)

Collects <l.ll .1 ill

"I," vent at 9°\ on
the East Pacific liisc.
icus the camera

\u lasers
i ALISS
distance
I under-.

(r

lulled

KSS al •

center)

1mm Al ISSs

pressure window.

rhevent

shrimp liintn m i\

exoculata suarm

near a lnilrullicilll.il

\enl site on llle

Mill \ll.inl i< I'.ulyc.

Imaging Ambient Light at Deep-Sea Hydrolhermal Vents

Sheri N. White

WHOI/MIT Joint Program Student

Alan D. Chave

Senior Scientist. Geology & Geophysics Department

In 1985. Cindy Van Dover, then a graduate stu-
dent in biology in the MIT/WHOI Joint Pro-
gram, discovered a novel light-sensing organ on
a unique species of shrimp that lives at high-tem-
perature, black smoker chimneys on the Mid-Atlan-
tic Ridge. If this photoreceptor were indeed some
sort of primitive "eye," the question instantly arose:
At depths of some 3,600
meters, where sunlight
cannot penetrate, what
are these shrimp looking
at? The search for a
source of light in deep-
sea hydrothermal envi-
ronments began.

The shrimp that Van
Dover investigated,
Rimicaris exoculata, lack
normal eyes and eye-

stalks; hence the name exoculata, which means
"without eyes" in Latin. This fact did not seem
surprising considering the pitch-black depths where
these shrimp dwell. But ensuing anatomical investi-
gations revealed the existence of a large, reflective
organ on the shrimp's dorsal side beneath the
thoracic shell. Subsequent studies of its structure
and pigment supported
Van Dover's hypothesis
that it was some type of
eye. The organ does not
create images, but it is a
very sensitive photore-
ceptor. It is uniquely
adapted to maximize the
absorption of light at
| about 500 nanometers,
| in the blue-green part of
g the spectrum. But what

14 • Vol. 41, No. 2 • 1998

kind of light were these shrimp "looking" at and
where was it coming from?

The most plausible light source is thermal radia-
tion from the extremely high-temperature (above
350°C) fluids discharging from the vents. Thermal
radiation is the light emitted from a hot, dense
object— the same phenomenon that causes the
heating element, or burner, on an electric stove to
glow. In 1988, on a cruise to the Juan de Fuca Ridge
in the northeastern Pacific, Van Dover persuaded
John Delaney of the University of Washington to
take a picture of a black smoker vent with a charge-
coupled device (CCD) camera while all of the sub-
mersible Alvin's external lights were turned off and
the portholes blocked. Indeed, the image revealed a
glow emanating from the vent orifice (photo be-
low). Since that time, a group of marine geophysi-
cists, chemists, biologists, and physicists has been
studying the phenomenon to learn what causes the
light, what the light can tell us about vent proper-
ties, and how biological communities at the vents
respond to the light.

The first confirming image of "vent glow" offered
no data to characterize the light. In an effort to
obtain a rough spectrum of the light. OPUS (Optical
Properties Underwater Sensor) was built at Woods
Hole Oceanographic Institution. OPUS is a simple
photometer that measures light intensity at various
wavelengths. OPUS consisted of four to eight photo-
diodes covered with individual optical bandpass
filters, which together cover the spectrum from 400
nanometers (in the blue part of the visible) to 1.000
nanometers (well into the
near infrared). The visible
region of the electromag-
netic spectrum (that is,
what humans are capable of
seeing) ranges from 400
nanometers (violet) to 750
nanometers (red), with the
infrared region extending
on to about a million na-
nometers. The search for
light is limited to the 400- to
1,000-nanometer region
because below and above
that range, light intensity
greatly decreases as it
passes through water. This phenomenon is called
attenuation; light is attenuated the least at approxi-
mately 500 nanometers, allowing blue-green light to
travel the farthest and giving the ocean its charac-
teristic color.

OPUS was deployed using Alvin at black smoker
chimneys on the Mid-Atlantic Ridge and the East
Pacific Rise during four cruises between 1993 and
1997. The OPUS data revealed that most of the vent
glow is in the infrared region of the spectrum
(greater than 750 nanometers), which suggests that

the light is primarily due to thermal radiation. The
OPUS data were consistent with the spectrum of a
theoretical "black body" — an ideal body that ab-
sorbs and emits all the radiation falling on it. Light
emitted from a black body is dependent only on the
temperature of the body. For a 350°C black body, a
graph of light emission would peak at longwave-
lengths (about 4,600 nanometers), with a tail ex-
tending down into the 400- to 1,000-nanometer
region (see figure on page 16). OPUS recorded light
emitted in this tail region.

OPUS successfully provided a rough spectrum of
the light emitted at vents. That is important in
determining possible light sources, since certain
mechanisms can cause light emission at certain
wavelengths. However, OPUS lacked an imaging
capability, and knowing
where the light is emitted at
a black smoker (at the
orifice or higher in the
plume, for example) may
also hold important clues
to which sources are more
likely. To obtain both im-
j ages and spectral informa-
* tion. a special charge-
5 coupled device (CCD) cam-
1 era was commissioned by
! WHOI. It is called ALISS
1 (Ambient Light Imaging
\ and Spectral System).

The development of CCD
technology in the 1970s was of great interest to
astronomers who measure very faint sources of
light— stars billions of light-years away. A CCD is
basically a silicon chip that is divided into a number
of pixels (picture elements), each of which acts like a
bucket holding incoming photons. The advantage of
a CCD is that it can digitally image very faint objects
by using long exposures to increase the number of
photons captured in each pixel. Long exposure
times are necessary for ALISS to capture vent glow,
which is too faint to be seen by the human eye.

Sheri \\hilechecks
Ihc alignment of the
ALISS tiller array
during a fall
1997 cruise.

I he llr-,1 image of
\cni glou taken on
the Juan de I lica
Kidgein 19SS.

OCEANUS-15

10"

750 1.000 2,000

Wavelength (nanometers)

Light emission from
-i "black hotly" (an
ideal hotly thai
ahsorhs and emits all
radiation falling
on it ) peaks at 4,600
nanometers with the
tail extending down-
ward in the region ol
the light spertrnni
visible to the human
eye (400 lo 750
nanometers).

Plot of actual data

Ihlne dots I collected

at "I1" \eill on the

Mill Atlantic liidge

and Iliedala

expected from a

llimiTlical lil.K k

hotly at : :«MI°C'

(red curve).

The ALISS cam-
era, built by
Princeton Instru-
ments Inc. of Tren-
ton, New Jersey,
contains a CCD chip
with a 1,024-by-
1,024-pixel grid. It
has a set of optics,
designed by Truax
Associates of
Southington, Con-
necticut, consisting
of nine individual
lenses that divide the CCD plane into nine smaller
tiles, each roughly 340 by 340 pixels. The nine lenses
in front of the chip are each covered by individual
optical bandpass filters with wavelength band-
widths of roughly 50 nanometers and 100 nanom-
eters. Thus, ALISS can obtain nine separate and
simultaneous pictures — each in a different wave-
length band. Equipped with two separate nine-filter
arrays, ALISS can obtain images in 18 wavelength
bands per dive.

ALISS is controlled by a scientist inside Alvin
using a personal computer, which also stores the
ALISS data. Two lasers mounted on top of ALISS
cross at a distance of 50 centimeters in front of the
camera to aid the Alvin pilot in positioning the
camera within its focusing range of 50±25 centime-
ters. Because the light emanating from vents is so
faint, ALISS must take a number of long exposure
images (5 minutes each), which can then be added
together and/or averaged electronically. This re-
quires the submarine to remain stationary and in

the dark (all exter-
nal lights and
lasers turned off
and all viewports
blacked out) for 5
minutes at a
time — not a com-
pletely comfort-
able prospect
when you are a
mere half a meter
from water hot
enough to melt
Alvin's viewports.
A number of corrections are made to the ALISS
images during processing to account for variations
in the response from one pixel to the next, as well
as for surplus charge buildup in the pixels that is
unrelated to the incoming light. The camera was
calibrated between cruises at the California Insti-
tute of Technology with the assistance of Palomar
Observatory scientists, who have extensive experi-
ence with CCD devices used, in their case, to ex-
plore outer space.

600 700 800

Wavelength (nanometers)

ALISS has been deployed at deep-sea hydrother-
mal vents in the 9°N area of the East Pacific Rise
(November-December 1997) and in the main En-
deavour vent field of the Juan de Fuca Ridge off the
Pacific Northwest coast (June-July 1998). Images
have been collected from black smokers and flange
pools whose venting fluids range in temperature
from 250° to 375°C. Black smokers and flange pools
are very distinct types of hydrothermal venting. At
some large chimney structures (particularly on the
Juan de Fuca Ridge), flanges are formed on the sides
of the chimneys, much like the eaves on a house.
Exiting hydrothemal fluid collects underneath these
flanges, forming high-temperature, relatively sta-
tionary, clear pools.

ALISS's images tell us a lot about the light emit-
ted at hydrothermal vents. The light's intensity is
highest at the vent orifice and decreases farther
from it. Because of rapid mixing with seawater, the
temperature of hydrothermal fluids drops from
about 350°C at the orifice to about 10°C at a dis-
tance of some 20 centimeters above the orifice.
Therefore, the higher light intensities appear to
correspond to the higher temperatures. An example
of the spectra derived from the ALISS images is
shown in the figure below left. The greatest amount
of light is seen in the far red and near infrared re-
gion of the spectrum (greater than 700 nanom-
eters). This matches the spectrum of an ideal black
body at similar temperatures. Thus, both the spatial
and spectral observations suggest a large compo-
nent of thermal radiation.

The data collected from black smokers and
flange pools on both the East Pacific Rise and the
Juan de Fuca Ridge show very similar results at the
longer wavelengths. In all cases, thermal radiation
appears to dominate, with hotter vents emitting a
greater amount of light. However, while flange pools
show no light in the visible range, some of the black
smoker data do show a very low level of visible light.
Given the temperatures of the vents, the existence
of visible light cannot be explained by thermal
radiation. Thus, it is still possible that other mecha-
nisms are also contributing to vent light. These
mechanisms include, but are not limited to, such
phenomena as sonoluminescence (the emission of
light due to the implosion of small bubbles) and
crystallo- and triboluminescence (the emission of
li^ht due to the crystallization and fracturing of
crystals, respectively).

At some hydrothermal vents, the extremely high
temperatures and relatively low pressures (with
respect to those at the ocean floor) can allow hydro-
thermal fluids to boil. Under these circumstances,
bubbles will form and sonoluminescence remains a
possibility. At all vents, minerals crystallize and
precipitate rapidly as the hot metal-rich hydrother-
mal fluids mix with colder seawater. As these miner-
als form, thermal shock might cause them to frac-

16 • Vol. 41, No. 2 • 1998

ture and emit light. Many of the minerals present at
hydrothermal vents emit light in this way under
laboratory conditions. Further examination of exist-
ing ALISS data and new data to be collected on the
Mid-Atlantic Ridge should help to verify what
sources contribute to vent light.

In calculating the contribution of thermal radia-
tion to vent light, three factors must be known: the
temperature of the fluids, the intrinsic attenuation
of light by water in the hydrothermal vent environ-
ment, and the emissivity of vent fluids. Alvin tem-
perature probes can measure the temperature of the
vents, so these are well-documented. Attenuation
experiments, using both OPUS and ALISS with an
external light source, have helped to determine how
much light is attenuated as it passes through the
water between the vents and the ALISS instrument.
Emissivity is a measurement of how much light is
actually emitted by an object. The emissivity of vent
fluids is related to the reflection, scattering, and
transmission of light through them. Thus, measur-
ing emissivity is not an easy task. An experiment
planned for a year 2000 cruise to hydrothermal

vents on the Mid-Atlantic Ridge will help ascertain
reasonable values of emissivity. An external light
source will be directed at the vents at angles of 0° to
180° from ALISS. The reflection, scattering, and
transmission of light through the vent fluid can be
measured by the camera and then used to calculate
the emissivity. In situ measurements of tempera-
ture, attenuation, and emissivity will greatly en-
hance our analysis of the ALISS images and will
help to elucidate just how big a role thermal radia-
tion plays in the phenomenon of vent glow.

Understanding what causes vent glow and char-
acterizing the light is just the first step. At present,
it appears that long-wavelength light at vents is
dependent on temperature, which offers some in-
triguing potential applications. Images of vent light
could be used to create thermal maps of vent
plumes, which may help elucidate processes by

which vent fluids and
seawater mix at vents.
If light in a certain
wavelength band can
be found to relate to
the chemistry of the
vent, imaging vents
may provide an in
situ way of analyzing
vent chemistry.

Then there is the
question of biological
response to light.
While we know that
the vent shrimp
Rimicaris exoculata
can detect vent light, it
has not yet been
proven whether they
actually do or how
they use that informa-
tion. Do they use their
photoreceptor to "see" the vents and thus locate
food? Or do they "see" the vents only to avoid the
high-temperature fluids that could cook them?

The shrimp are not the only fauna living at
vents, and perhaps we will find that other organ-
isms can also detect vent light. While some scien-
tists speculate that life may have begun in hydro-
thermal vent environments, others take it a step
further and wonder if photosynthesis could have
originated at vents, as well. The discovery of a non-
solar source of light at the ocean floor adds to the
list of amazing vent phenomena that were once
thought impossible.

Development of the ALISS camera system was funded by
the National Science Foundation Oceanographic Instrumenta-
tion Program.

Sheri White is a Ph,D. candidate in the Department of Geology &
Geophysics. Wfiile she once dreamed of being an astronaut, Sheri
now sets her sights down instead of up — some 2,500 meters down
to deep-sea hydrothermal vents. When she's not at sea or in front
of her computer, she is a force to contend with on the rugby pitch.

Alan Chave is a Senior Scientist in WHOI's Department of Geol-
ogy and Geophysics, where he says he specializes in going to
meetings and writing reports. When time allows, he builds "toys"
to throw in the ocean (hoping they will come back) and has been
quite successful at this.

The ALISS camera,
mounted on l/r///s
science basket, looks
at "P" venial1)1 A on
ilir I .isi Piirillc HIM-.

At It'll. AMSS
i run di-il nine ini.im-.
of"P"vi'nl iii niiif
different wavelengths
nl h^lii I In- iin.i^i- in
llu1 shortest wave-
li-nulli i.ilniul Till
nanometers) is in llu1
upper righthand
roriu'r. I he intake in
llu- loupes! wave-
length I some <).)()
nanometers I
is in the lower
Icllli.uiil corner.

These two photos

slum "P" vent illinni-
i i,il ril In . l/r/'//'s lights
(lell ) and imaged
with ambient lijjht in
the 870-nanometer
band (riylil).

OCEANUS-17

A nick sample,
recovered by drilling
1 16 mclcrs belim Ilio

active scafloor

Imlrolhcrinal vonl silo

,il L!d \.inllii- Mill

Atlantic Itidgr. .shims

hint I he ruck has been

altered by read ions

'.', il li M'.m.ilri al

temperatures of about

300°C. Pieces of

hii:hl\ altered rork

("ray) an1 cemented

l«!>rllii'i uilh minerals

mil as iron Mlllides

(gold-colored) anil

quartz (white).

The Cauldron Beneath the Seafloor

Percolating Through Volcanic Subsurface Rocks,
Seawater Is Chemically Transformed into Hydrothermal Fluid

Susan E. Humphris

Senior Scientist, Geology & Geophysics Department

Thomas McCollom

Postdoctoral Fellow, Marine Chemistry £ Geochemistry Department

Hydrothermal

Fluid Seawater

Temperature ("
Acidity (at 25 C)
Dissolved Oxygen
Hydrogen Sulfide (mM) 2.3-3

Sodium (mM)
Potassium (mM)

Calcium (nM) 30.8
Magnesium (mM) ,

' 4

Silica (nifty 20.75
Chloride (mM) '636

Manganese (uM)

Iron(uM) 5590 0.0015

Copper (vM) 98-120 0.007

Zinc(nM) 47-53 0.01

A comparison of characteristics and chemical com-
position slums the distinct tlillerences between
seawater and hydrothermal vonl llniil. in this case
llnid from lhi> I VG livdriilherinal site on I he Mid-

\ll.mln Ki.lu.

Just over 20 years ago, scientists exploring the
mid-ocean ridge system first made the spec-
tacular discovery of black smokers— hydrother-
mal chimneys made of metal sulfide minerals that
vigorously discharge hot, dark,
particulate-laden fluids into the
ocean. The ultimate source of
the fluid venting from these
smokers is seawater, but a com-
parison of chemical composi-
tions shows that seawater and
hydrothermal fluid are dis-
tinctly different. The vent fluids
are not only far hotter than sur-
rounding seawater, they are also
more acidic and enriched with
metals, and have much higher
concentrations of dissolved gases,
such as hydrogen, methane, and
hydrogen sulfide (see table below). The
metals trans-
ported by the fluids fre-
quently form ore deposits
at the seafloor, and the
dissolved gases support a
prolific biological commu-
nity that derives its energy
from chemical reactions
rather than sunlight. By
what processes is seawater
turned into this remarkable
i fluid that emanates from
I black smokers?

The answer lies beneath
\ the seafloor, within the
L oceanic crust. The mid-
1 ocean ridge system, which
I forms where the earth's
I tectonic plates are spread-
1 ing apart, is volcanically
| active and the site of nu-
; merous heat sources, which

c

5 induce seawater to circu-
late through the permeable
oceanic crust. It is esti-
mated that the equivalent
of an entire ocean's worth

of water circulates through the mid-ocean ridge
hydrothermal systems every 10 million years or so.
As the seawater percolates through subseafloor
rocks, a complex series of physical and chemical
reactions between seawater and volcanic rocks
drastically changes the chemical

composition of both the seawa-
ter and the rocks. These chemi-
cal reactions not only influence
the composition of the oceanic
crust, they also play a role in
regulating the chemistry of the
oceans.

The history of these chemical
reactions is recorded in the miner-
als and chemical composition of
the rocks. By investigating samples
of rocks that have been altered, we
can learn about the sequence of wa-
ter-rock interactions taking place in
the subsurface. We can then begin to
understand the processes responsible for the chemi-
cal composition of vent fluids, the formation of
sulfide-rich mineral deposits, and the existence of
biological communities at hydrothermal vents.
Gaining access to investigate the subsurface
portion of a hydrothermal system is, of course, a
difficult problem, and scientists must employ sev-
eral different strategies. The most direct approach
is to find techniques to collect and analyze altered
rocks. One way is to drill a borehole through a
seafloor hydrothermal mineral deposit and recover
samples from the oceanic crust beneath. Over the
past few years, the international Ocean Drilling
Program (ODP) has conducted drilling operations
in two hydrothermal areas — one on the Juan de
Fuca Ridge off the northwestern US coast, and one
on the Mid-Atlantic Ridge about halfway between
Florida and West Africa. The drill cores recovered
from these sites allow scientists to study the vari-
ability in rock-water reactions that occur under the
different physical and chemical conditions found at
different depths within the earths crust. Ocean
drilling operations, however, are extremely expen-
sive and consequently have been carried out at only
a few locations.

Scientists can also collect seafloor rock samples

18- Vol.41, No. 2« 1998

by using dredges and small submarines
("submersibles") in areas where faults and fractures
have exposed rocks on the seafloor that were once
in the deep subsurface. The disadvantage of this
method is that the same processes that expose the
rocks may also muddle the spatial and temporal
relationships among individual samples. Neverthe-
less, much has been learned about the chemical
effects of water-rock reactions from dredge and
submersible samples. In many samples, the outer
rim, which has been altered by exposure to circulat-
ing hydrothermal fluids, can be compared to the
fresh, unaltered interior of the rock in order to
learn how the rock has been changed by the fluid
(see page 20).

While drilling, dredging, and submersibles can
be used to collect rocks to study the shallower por-
tion of the ocean crust, scientists have had to turn
to rocks on land to investigate deeper sections of
the hydrothermal system. In a few locations, includ-
ing sites in the western US, Oman, Cyprus, and the
west coast of Newfoundland, sequences of rocks
exposed on land resemble what scien-
tists believe to be the structure of the
oceanic crust. Many geologists think
that these rocks represent sections of
oceanic crust that have been thrust
onto the continents by tectonic move-
ments. Within these so-called
"ophiolite" sequences are ancient
analogs of seafloor hydrothermal
mineral deposits, and these sites pro-
vide another source of hydrothermally
altered rocks for study. But this
method, too, has pitfalls: In some
cases, ophiolite rocks have been al-
tered during the tectonic processes
that uplifted and thrust the oceanic
crust onto land. This subsequent
alteration often obscures the original
alteration that took place on the sea-
floor, making it difficult to use the
rocks to study submarine hydrother-
mal processes.

Scientists also employ experimen-
tal strategies in laboratories to under-
stand fluid-rock interactions, setting
up reactions between rocks and sea-
water under conditions simulating
those in a seafloor hydrothermal sys-
tem. The earliest of these experiments
actually pre-dated the discovery of
seafloor hydrothermal systems. In the
experiments, crushed rock samples
and seawater in varying proportions
are placed in a sealed reaction vessel
(commonly referred to as a "bomb"!),
which is then subjected to high tem-
peratures and pressures. These "cook-

and-look" experiments provide a way to explore
how reactions change as physical and chemical
conditions are varied, and they help scientists de-
termine how the chemistry of the fluid and the rock
evolves as the reactions proceed. Over the years,
experimentation has progressed to include "flow-
through" models that examine the changes in chem-
istry and physical state as fluids migrate through a
system. While laboratory simulations often result in
end-products that are somewhat different from
those observed in rock samples from actual altered
ocean crust, scientists have gained insights that
have been critical in deciphering the complex set of
water-rock reactions taking place in natural hydro-
thermal systems.

A third approach to understanding the chemis-
try of hydrothermal systems is geochemical model-
ing. Scientists have used models to investigate the
sequence of minerals that dissolve and precipitate
during fluid-rock reactions, as well as to examine
how the fluid changes its composition as it circu-
lates through the crust. These efforts depend on the

111 a Imllnlllri ni.il

i 111 ill. ii KIII system,
cold sea\*ater seeps

through I In- |ii-i UK-
.ll lie- MM I I. I. II .111.1

deeper subsurface
dikes. II undergoes
a series of chemical
reactions with sub-
surface rocks al
various temperatures

In i ir. ill- I nil In ill i .

thermal lluid thai
eventual!) \euis
at the sealloor.

OCEANUS • 19

\\lllll Srmni S, „.„

h-i Silvan llnmphris

( ll H I'^l I M 1 1 II I ) | II I •[ 1. 1 I . ' v

In dcplti> I In near

li.ii him IrgD// optical

and imaying system

operated hy \\1IOI,

Deep Submergence

Operations Group.

l"uril hchind ships at

• In I '• meters oil lllc

srufloor. \riii>ll

Collected \ I, In i .mil

si ill images of sca-
lldor hydrolhcrnial

U'lll s\slcills.

\ rock sample
dredged from the

Mi, I \ll.nili. Ki.l-.-

slumshou s< .m.ilri

Hci\. Mi'J lulv, ,-,-M

subsurface rocks

alters them antl

ccmenls them lo-

gclhcr. Hie rocks'

outer rims (gray) have

lii'i-n i In inn .ill\
changed liy interac-
tion uil h hoi seaualer
and can he easily
distinguished from

Ilif lrl.ll |\ r!\ llli.il

teied interior (brown).

l>\ comparing the

ge<irhrmislr\ ol the

rim and I he interior.

ii'searrheiscan

delenni;]: • . ays in

which cli'l:

e\ch;ii(L> ! irtneen
SCSIWut:-: •! lock.

availability of good ther-
modynamic data at the
temperatures and pres-
sures that occur in hy-
drothermal vent systems,
much of which has been
generated only in the
past few years. The mod-
els provide a framework
for integrating the obser-

O o

vations made from rock
samples and experimen-
tal studies, and they have
proven to be a powerful
tool to relate the changes
in fluid chemistry to the
alteration mineralogy of
the rocks.

By integrating results
from these different
investigative strategies, a
model is beginning to
' emerge of how seawater
chemistry changes in an active seafloor hydrother-
mal system, from the time it enters the oceanic
crust until it is discharged as a vent fluid. Conceptu-
ally, the circulation of seawater through the oceanic
crust can be divided into three parts (see page 19):

• A recharge zone, where seawater enters the crust
and percolates downward:

• A reaction zone at the maximum
depth of fluid penetration, the site
of the high-temperature reac-
tions that are thought to deter-
mine the final chemical char-
acteristics of the hydrother-

mal fluid; and

• An up/low zone, where the
buoyant hydrothermal fluids
rise and are discharged at
the seafloor.

The Recharge Zone

Seawater percolates readily into
the upper layer of the oceanic crust, which is con-
structed of highly porous and permeable volcanic
rocks that are broken apart in many places by cool-
ing cracks and tectonic fractures. As a consequence,
reactions between seawater and the exposed rocks
at relatively low temperatures up to about 60°C are
pervasive. Although reactions are relatively slow at
these low temperatures, they nevertheless begin to
change the composition of the seawater through two
processes. First, the seawater partially oxidizes the
crust, resulting in the removal of oxygen from the
seawater. Minerals containing iron in the rocks are
replaced by iron oxides and hydroxides (a process
analogous to the formation of rust), which also fill
veins and pore spaces in the upper crust. Second,

the reactions with seawater break down the original
rock minerals, replacing them with alteration miner-
als such as mica and clay. In the process, potassium
and other alkali elements, such as rubidium and
cesium, are transferred from seawater into the rocks.
Beyond about 300 meters into the oceanic crust,
penetration of seawater becomes more and more
restricted as the rocks' permeability decreases.
Larger fractures and fissures are more likely to
become the main conduits for fluid flow. As the
fluid (already oxygen- and alkali-depleted relative to
seawater) continues to penetrate downward toward
the heat sources, it becomes heated further, and
other reactions occur. At temperatures above about
150°C, clay minerals and chlorite precipitate out of
the fluid, essentially removing all of the magnesium
originally present in the fluid. The formation of clay
minerals and chlorite also removes hydroxyl ions
from the fluid, resulting in an increase in acidity
(that is, a lower pH). This increase in acidity, in
conjunction with the breakdown of the original
minerals in the rocks, causes calcium, sodium,
potassium, and other elements to be leached from
the rock into the fluid. Hence, the removal of potas-
sium (and the other alkalis) from the fluid at lower
temperatures is partially reversed at higher tem-
peratures at greater depths!

Another important reaction results in the forma-
tion of the mineral anhydrite (calcium sulfate). This
mineral possesses something called "retrograde
solubility," which means that instead
of becoming more soluble with
increasing temperature as most
minerals do, it becomes less
soluble. At the pressures
found at the bottom of the

ocean, this results in
', anhydrite precipitating
from seawater when
temperatures rise
above about 1 SOT. This
process removes about two-
thirds of the sulfate initially
present in the seawater and also limits the
calcium concentration of the fluid. At temperatures
greater than 250°C, the remaining sulfate in the
fluid reacts with iron in the crust to form metal
sulfide minerals.

The Rent-lion '/.one

The "reaction zone" designates the region where
high-temperature, water-rock reactions occur. This
zone is near the heat source that drives the circula-
tion system. The depth of the reaction zone de-
pends on the depth of the heat source and varies
from one mid-ocean ridge to another. On the fast-
spreading East Pacific Rise, the presence of a
magma lens at a depth between 1.5 and 2.4 kilome-
ters defines the lower limit of circulation, but sea-

20- Vol. 41, lo. 2. 1998

water may penetrate deeper on slower-spreading
ridges where no melt lens has been detected. Scien-
tists think reactions in this zone determine the final
chemical characteristics of the hydrothermal fluid.
Reactions at such high temperatures (up to 350° to
400°C) produce a characteristic suite of alteration
minerals (chlorite, sodium-rich feldspar, amphibole.
epidote. and quartz), which, in turn, controls the
fluid composition. Metals, such as copper, iron, and
zinc, as well as sulfur, are leached from the rock by
the acidic fluid. This provides the source of metals
for the massive sulfide deposits observed at the
seafloor, as well as the hydrogen sulfide to support
the chemosynthetically based hydrothermal biologi-
cal community.

I lie I 'pflowZone

Buoyancy forces cause the hot fluids to rise
rapidly toward the seafloor. much as hot air causes a
balloon to rise in the atmosphere. Initially, the
upflow is focused along a conduit of high perme-
ability, such as a fault surface. As it reaches shallow
depths, the flow may continue to be focused and
may discharge through a chimney, or it may follow
more tortuous pathways and be discharged as a
more diffuse flow (like water flowing through a
sponge). Continued high-temperature reactions
between the rock and the upward-flowing, metal-
rich, magnesium-depleted hydrothermal fluid pro-
duce an "alteration pipe" of highly altered rocks with
an interconnected network of veins filled with sul-
fides, silica, and chlorites. As focused high-tempera-
ture (350° to 400°C) fluids discharge at the seafloor
as black smokers, mixing with the surrounding
seawater causes metal sulfides to precipitate and
form massive sulfide deposits rich in iron, copper,
and zinc (see article, page 22). However, at locations
where the volcanic pile is especially permeable, the
upflowing hydrothermal fluid will mix with colder
seawater in the shallow subsurface, resulting in the
metal sulfides being precipitated beneath, rather
than at, the seafloor. The resulting lower-tempera-
ture fluids, depleted of metal sulfides, vent as "white
smokers," rather than particulate-laden black smok-
ers. Shallow subsurface mixing may also heat sea-
water to form anhydrite and cool hydrothermal
fluids to precipitate silica, both of which cement the
metal sulfides or seal fluid conduits.

Together, all the hydrothermal water-rock reac-
tions that occur — from the time seawater enters the
system to the time hydrothermal fluid leaves it-
play a role in regulating the chemistry of seawater.
But the relative importance of hydrothermal reac-
tions must be balanced with other factors that influ-
ence ocean chemistry — particularly, rivers, which
are the principal conduits by which most (but not
all) chemical elements enter the ocean. River input
provides a good measuring stick by which to com-
pare the relative contribution of hydrothermal activ-

ity to the fluxes of elements in and out of the ocean.
Hydrothermal vents are a source to the ocean of
alkali elements that leach from the crust during
hydrothermal alteration (although this process may
be tempered somewhat by lower-temperature
weathering of the shallow crust away from the
ridges, which removes alkali elements from
seawaterj.Vents also represent a significant source of
manganese input to the ocean. Most of the metals
present in hydrothermal fluids (iron, copper, zinc,
etc.) are removed rapidly by precipitation, either at
the seafloor or by mixing with seawater in the sub-
surface, so most of the metals do not enter the
oceans. On the other hand, hydrothermal circulation
removes magnesium and sulfate from seawater, so
the crust acts as a sink for these elements. The mag-
nesium loss is perhaps the most significant, and
hydrothermal activity may be a major mechanism of
balancing the magnesium budget in the ocean.

Susan Humphris's research is supported by the National
Science Foundation. Thomas McCollom is an NSF Earth
Sciences Postdoctoral Fellow.

Susan Humphris first came to Woods Hole from England in 1972
to enter the MIT/WHOI Joint Program. For her PhD. thesis, she
studied some rocks dredged from the ocean floor that had reacted
with seawater and determined the reactions that must have
occurred. Six months after she completed this work in 1976, the
first hydrothermal vents were discovered. She has spent more than
three years of her life on research vessels of various kinds, ranging
from traditional sailing ships, when she worked at the Sea Educa-
tion Association teaching oceanography to undergraduates, to
drilling vessels as a participant in the Ocean Drilling Program.
She has completed about 30 dives in submersibles and has used
ROV]ason to study new hydrothermal sites. In her spare time,
Humphris and her husband tend a large vegetable garden and
raise chickens and the occasional pig.

Tom McCollom's interests are in the organobiogeochernistry of
seafloor hydrothermal systems. He manages to squeeze in a little
research now and then between running around on the soccer
find ultimate fields, pedaling his bicycle, climbing up (or skiing
down) hills, dancing to his favorite bands, and birdwatching with
his wife. Ifer.

Chemical reactions
in hydrothermal vent
systems are a source
of elements ( positive
values) Icachiii!!, from
the ocean crusl to the
ocean, and a sink
(negative values) for
elements renio\ed
from scaualer and
incorporated into the
crust. In I lie cli. n I
above*, flnxe's of
elements into and onl
of seawater caused In
hydrolliermal activity
are compared to
element fluxes
caused by rivers.
Green bars indicate
minimum estimate's
of eleme'iit fluxes:
purple bars represent
maximum estimated
fluxes. ( Adapted
from Global ImpcuA of
Submarine //i tlro-
lliennnl I'rocexses.
The riniil Report.
RIDGE/VENTS
Workshop. 1994.)

OCEANUS • 21

3*

mf

How to

Build a

Black

Smoker

Chimney

, l/i /// •. manipulator
reaches louarda

1)1, H k Mllllkrl I llllM
MCy. SCCM III! iin-li I hi'

sul>'s\icu|>nrl.al

l7uSi>n I he |{asl

I'aciru I'.ise. llol

lnclrcilhriMi.il fluids

siirj;clhrc>ii!sli I In-

chimney al velocities

• il I In '» mclcrs per

second. The "black

smoke" consisls ul .in

abundance oldark.
fine-grained, sus-
pended particles lli.il

|ii «•( ipll.llr \\ IM-II tl)C

liol lluid mixes uilh
cold seaualer.

The Formation of Mineral Deposits
At Mid-Ocean Ridges

Margaret Kingston Tivey

Associate Scientist, Marine Chemistry & Geochemistry Department

Diving along the mid-ocean ridge at 21°N on
the East Pacific Rise, scientists within the
deep submersible Alvin peered through
their tiny portholes two decades ago to see an aston-
ishing sight: Clouds of billowing black "smoke" rising
rapidly from the tops of tall rocky "chimneys." The
"smoke" consisted of dark, fine-grained particles
suspended in plumes of hot fluid, and the "chim-
neys" were made of minerals that were rich in met-
als. Using specially designed fluid bottles and tem-
perature probes, Alvin took samples of these black
smoker chimneys, as well as the 350°C fluids venting

from them. Since then, scientists have observed and
sampled numerous active vent sites along portions
of the mid-ocean ridge in the Atlantic and Pacific
Oceans, and in back arc basins in the Pacific Ocean.
It has become abundantly clear that these high-
temperature seafloor hydrothermal systems are the
analogs to systems that created some of the world's
economically valuable mineral deposits, including
some that have been mined on land. In Cyprus and
Oman, for example, ore deposits of millions of tons
are found in ophiolites. portions of ancient seafloor
thrust onto land by tectonic forces.

lo. 2 • 1998

Scientists can gain much insight into hydrother-
mal processes through detailed studies of these
exposed areas of fossil systems, but only by investi-
gating active systems can they simultaneously
examine hydrothermal fluids and the corresponding
mineral deposits created by them. By analyzing
these fluids and deposits, we have been able to
formulate models to explain how submarine min-
eral deposits, from seafloor chimneys to great
subseafloor depths, are initiated and how they grow
in their early stages.

One of the most fascinating aspects of black
smoker chimneys is how rapidly they form. They have
been measured to grow (after upper parts of the
chimneys are razed by sampling) as fast as 30 centi-
meters per day. Examination of young chimney
samples, under the microscope
and by X-ray diffraction, re-
vealed that the earliest stage in
the creation of a black smoker
chimney wall involves precipita-
tion of a ring of a mineral called
anhydrite. The ring forms
around a jet of 350°C fluid,
which exits the seafloor at
velocities of between 1 and 5
meters per second. Anhydrite,
or calcium sulfate (CaSOJ, is an
unusual mineral because it is
more soluble in seawater at low
temperatures than at high
temperatures. Seawater con-
tains both dissolved Ca2* and
S04" ions, and when it is heated
to 150°C or greater, the ions
combine and anhydrite precipi-
tates. Hydrothermal fluids
contain little or no sulfate, so
the origin of the sulfate in the
precipitated anhydrite is seawa-
ter. Calcium, however, is present
in both seawater and hydrother-
mal fluid. That made it more
difficult at first to determine
whether the initial anhydrite
chimney wall formed solely
from seawater that was heated
by hydrothermal fluids, or from
the mixing of cold, sulfate-rich
seawater with hot, calcium-rich
hydrothermal fluid.

Strontium, which is
present in seawater and hy-
drothermal fluid, was used to
investigate this problem.
Strontium has the same
charge as calcium and a num-
ber of different, easily measur-
able isotopes. (Isotopes are

elements having the same number of protons in
their nuclei, but different numbers of neutrons.
Thus they share chemical properties but have
slightly different physical properties.) Strontium
can readily take the place of calcium in the crystal-
line lattice that forms when anhydrite precipitates.
The concentration of strontium, as well as the
ratio of two of its isotopes, strontium 87 and stron-
tium 86, were measured in both vent fluid and in
seawater. Because the ratio of strontium 87 to
strontium 86 is higher in seawater than in hydro-
thermal fluid, it is possible to determine whether
the source of the strontium (substituting for cal-
cium in newly formed anhydrite grains) is seawa-
ter or vent fluid. The answer is both: Anhydrite
walls form from the turbulent mixing of seawater

During Stage 1 oi'hlacksuu
ml calcium-rich seaualer.
oxides carried in llu*lmi 11
ahmelhc \cnl. Durini; Sla

ke

es

cl
f'l

iir

rl

es

chimne) growth, hoi. cal
Iling in precipitation c»l

SO |>i r< i|Hl.llr 1 .l|Hill\ (1
il i llllllllr\ Ul.m III till il
Ir bcijills to |M (•< ]|Hl.llr

s across I he porous u .ill
i the interstices of the «

linn I u h \rn

ingof calciu

nu lilt' mi \ ii

ial chimney

il [il.llr Illr il
,i.l\r. ll.HI .11

1 u In, Ii «ra<li

fluid mixes liirhulciillv uilli cold, -iill.il.
n sulfate (anhydrite). \lelnl sulfidesand
4 process, lot mmi:,i plume oldark particles
all ol anhydrite forms a surface on which
MI-I chimiie\ u.ill. Mixing ol sca»alcr and
d diffusion results in the deposition ill /in.
i.ilK makes the chimne) less porous and

h\ « 1. < ii IH-I ni.il fluid compo
copper-iron, and iron suill
np. ii .• mclal-rich.

OCEANUS • 23

Massive Pyrite
Breccia

Pynte-Anhydrite Breccia
Anhydrite Breccia

Pynte-Silica Breccia
Silicified Wall-Rock Breccia
Chloritized Basalt Breccia

m Basalt

III I III-. M ill-Ill. III!

drawing of the TAG

.Kim- Induil lin m,il

mound, hydrothermal

I Inn I rises rapidly and
exits the mound al (he
r.l.K U Smoker Com-
plex. Cold calcium-

.mc I ~ull.li <• i n 1 1
seawater is entrained

into I In- n M nun I

\\hrir il IIIIM-V uilli

hydrothermal fluid.

llu' mixing causes

anhydrite, pyrite, and

i-ll.ili ii|i\nlr In
|M i-< ipil.itr uiMilr I In-

mound. I his precipi-

l.iliiiu mi H-.IM-S llu-

aridily ol the hydro*

lln-iin.il lliu.l. /in,

and other elements,
such as silver, gold,
and cadmium, dis-
solve in tills .irilllc

lluul. .illou iui; them
to he carried In tin1
white smoker fluid to
the edges of the
mound at tile "Krem-
lin ;'; i. il -IT the
cooler temperatures
uilliin unite smoker
chimney ualLs cause
(he elements in
precipilale.

50m

and hydrothermal fluid, not just from the
rapid heating of seawater.

During this mixing, other processes occur during
early stages of chimney growth. Metal sulfides and
oxides (zinc sulfide, iron sulfide, copper-iron sulfide,
manganese oxide, and iron oxide) precipitate from
the vent fluids as fine-grained particles, most of
which form a plume of "smoke." Because bottom
seawater is denser than the mix of seawater and
hydrothermal fluid in the plume, the plume rises
some 200 meters above the ridge to a level of neutral
buoyancy. Some particles form close to the chimney
and become trapped within and between grains of
anhydrite within the nascent chimney walls. These
particles give the anhydrite, which is white in its
pure form, a gray to black color. Copper-iron sulfide
(chalcopyrite, or CuFeS.,) begins to precipitate and
plate the inner surface of the chimney. The evolving
chimney walls are thin, ranging from less than a
quarter of an inch to a few inches, but on either side
of them are large gradients of pressure, temperature,
and concentrations of elements. Aqueous ions,
including copper, iron, hydrogen sulfide. zinc, so-
dium, chloride, and magnesium, are transported
from areas of high to low concentrations (by diffu-
sion). These elements also are carried by fluids flow-
ing back and forth across the wall from areas of high
to low pressure (by advection). As a result of these
processes, zinc sulfide, iron sulfide, and copper-iron
sulfide precipitate in the interstices of the chimney

walls. The chimney walls thus become less porous
and more metal-rich over time.

Early examinations of black smoker chimneys
resulted in a model of chimney growth that is still
accepted nearly 20 years later. But in terms of
size and ore grade, black smoker chim-
neys are not important mineral deposits.
Most of the metals are lost into the
plume that rises into the water column
above the vents and is dispersed. In
the last decade, the focus of study has
shifted to larger deposits present
along mid-ocean ridges.
., In 1985, scientists aboard Woods

Hole Oceanographic Institution's
| Atlantis II discovered hydrothermal
vents at the Trans-Atlantic Geotraverse
(TAG) active mound, the single largest
known active mineral deposit along the
mid-ocean ridge. Roughly circular in plan
view, the TAG site has a diameter of about
150 meters and rises some 50 meters above
the seafloor, with an estimated mass of 3 mil-
lion tons. TAG has been intensively sampled.
The Alvin, Mir, and Shinkai submersibles have
been used to recover vent fluids, chimneys, and
other hydrothermal precipitates from the mound
surface. And in 1994, 17 holes were drilled into the
mound during Leg 158 of the Ocean Drilling Pro-
gram (ODP). Drillcore was recovered from five
mound areas to a maximum depth of 125 meters
below the seafloor.

As in studies of black-smoker chimneys, the
combination of vent fluid data and examinations of
anhydrite played an important role in determining
the processes involved in growth of the large TAG
mineral deposit. In 1990, two distinct fluid composi-
tions were observed to be exiting the TAG mound.
From the so-called Black Smoker Complex in the
northwest area of the mound, 366°C black smoker
fluid billowed from an aggregation of chimneys in a
huge black plume that shrouded nearly the entire
complex. Approximately 70 meters southeast of this
complex, however, clear fluid with temperatures of
less than SOOT emanated from an area called the
Kremlin, named for its 1- to 2-meter-high chimneys
with their distinctive, onion-shaped Byzantine
cupolas. The mineral composition of the black
smoker chimneys was very similar to those of black
smokers at other mid-ocean ridge vent sites. The
chimneys forming from the cooler white smoker
fluid, however, were quite different, containing
significant amounts of zinc, as well as cadmium,
silver, and gold. Analyses of the white smoker fluids
indicated that they were more acidic and contained
less copper, iron, calcium, and hydrogen sulfide, but
10 times more zinc, than the hotter black smoker
fluids. However, concentrations of other elements,
such as potassium, were identical, suggesting that

24- Vol.41, No. 2' 1998

the two fluids were related to one another.

A hypothetical series of steps was soon developed
to explain these observations. The deficits of copper,
iron, and hydrogen sulfide in the cooler white smoker
fluid could best be explained by the precipitation of
copper-iron sulfide and iron sulfide (that is. chal-
copyrite and pyrite) within the mound. In addition, it
was theorized that precipitation of anhydrite within
the mound could explain the lower calcium concen-
trations in white smoker fluids. To trigger the deposi-
tion of sulfates and sulfides inside the mound, sul-
fate-rich seawater would have to percolate down into
the mound and mix \vith the hotter hydrothermal
fluid. The precipitation of sulfides would release
hydrogen ions, making white smoker fluid more
acidic. The increased acidity, in turn, would cause
metals in the mound, such as zinc, cadmium, silver,
and gold, to dissolve. Once dissolved in the fluid,
these so-called "remobilized" elements could be
transported toward the upper edge of the mound.
This would explain the excess zinc observed in white
smoker fluid. At the seafloor, the white smoker fluid,
rich in remobilized metals, confronts 2°C seawater
just outside the chimney wall. Crossing this thermal
gradient, the fluid cools, and some metal-rich miner-
als precipitate. This process, known as "zone refine-
ment," explains how some ore deposits are separated
into large-scale zones containing different metals,
with copper in the center of the deposits, for ex-
ample, and zinc at the edges.

To determine whether the scenario described
above was reasonable, we used geochemical model-
ing calculations, which take into account the ther-
modynamics of a range of different chemical reac-
tions at high temperatures. The theoretical reac-
tions had to reproduce the already-well-docu-
mented composition of the less-than-300°C fluid
from combinations of the 366°C fluid and seawater.
These calculations demonstrated that the composi-
tion of the cooler white smoker fluid we observed
could theoretically result from mixing 86 percent
black smoker fluid with 14 percent seawater, which
would result in the precipitation of 19 parts anhy-
drite, 8 parts pyrite, and 1 part chalcopyrite within
the mound, as well as the remobilization of zinc and
other metals by the resultant acidic fluid.

These predictions were put to the test when the
TAG mound was drilled in the fall of 1994. One of
OOP Leg 158's major findings was that significant
amounts of anhydrite are present throughout the
mound. Anhydrite has not been seen in analogous
ophiolite structures probably because it dissolves
at lower temperatures and essentially has disap-
peared from land-based deposits. But anhydrite
was recovered from three of the five sites drilled at
TAG. It was present at the base of the deepest hole
drilled at the TAG site, to a depth of 125 meters
below the seafloor. The drilling revealed that the
mound contained wide, complex, anhydrite-rich

veins, ranging in size from 1 millimeter to 1 meter
wide, which formed as anhydrite precipitated in
cracks within the mound.

The discovery of anhydrite confirmed the predic-
tion that seawater was entering and traveling
through the mound. It also let us determine the
proportions of seawater that were mixing with
hydrothermal fluid within the mound. Analyses of
strontium isotopes from anhydrite grains recovered
from various depths in the mound demonstrated
that anhydrite was forming from mixtures that

ranged from 99.5 percent seawater and 0.5 percent
hydrothermal fluid to 52 percent seawater and 48
percent hydrothermal fluid.

Samples of anhydrite grains recovered from drilled
cores were also used to determine the temperature
and salinity of fluids at various points within the
mound. When mineral grains form, small amounts of
the fluid from which they are forming can be trapped
and enclosed within the precipitating grain. As a
result, small cavities can form within the mineral,

Author Margaret
They examines Ihe
lop ol a spire oi a
black smoker chim-
ney retrieved In l/ri'/i
from the I .1-1 I'aciflc
Rise at 17°S.

OCEANUS • 25

Thin-section samples

.,!,l,m,,,r, u.lll

specimens, examined

under a microscope,

I r\ r.ll metal snlllde

part ides ( black)

cmhedded in and

around anhydrite

crystals (clear).

which may contain one or more phases (liquid, vapor,
or solid). These are called fluid inclusions. The cavi-
ties can range in size from about 1 micron to greater
than 1 millimeter, though 3- to 20-micron inclusions
are most common. The anhydrite grains from the
TAG mound all contained abundant fluid inclusions
with two phases: liquid and a vapor bubble.

We analyzed fluid inclusions within mineral
grains by using a heating-freezing stage attached to
a high-magnification microscope. The salinity is
determined by freezing the fluid within the inclu-
sion at temperatures of -^rm/ t^_ -m"
less than -SOT and then
slowly heating until the
last ice melts. The tem-
perature of final melting
is a function of the salt
content in the fluid. The
temperature of the fluid
when it was trapped
within the mineral also
can be determined by
heating the fluid inclu-
sion and measuring the temperature at which the
vapor bubbles disappear. With adjustments for
undersea pressures, measurements of fluid inclu-
sion trapping temperatures from a number of differ-
ent anhydrite grains recovered from within the TAG
mound indicated very high temperatures (greater
than 337T) throughout most of the mound. At
depths greater than 100 meters below the seafloor,
trapping temperatures were in excess of 380T.

The combination of strontium-isotope and fluid-
inclusion analyses of TAG anhydrite grains not only
demonstrated that large amounts of seawater are
being entrained into the mound, they also showed
that anhydrite (and chalcopyrite and pyrite) precipi-
tated in the mound from seawater-hydrothermal
fluid mixtures that are greater than 50 percent sea-
water. That led to a dilemma and to another discov-
ery. The dilemma was that our geochemical model-
ing calculations predicted that if the mixing propor-
tions were greater than 50 percent seawater, and if
mixing alone determined the temperature of the
fluids in the mound, the fluid temperatures in the
mound should be relatively low (5° to 250°C). Our
fluid inclusion measurements, however, indicated
that the anhydrite grains were nearly all precipitat-
ing at much higher temperatures, 187T to 388T.

The logical conclusion is that the seawater and
seawater-hydrothermal fluid mixtures that are
entering and travelling through the mound are
being heated by conduction. Hot black smoker
fluids are flowing rapidly along a direct, highly
focused route up through the mound to the Black
Smoker Complex. Drilling within the mound re-
vealed extensive accumulations of breccia — rocks
made of sulfide-rich fragments that conduct heat
well. So it is reasonable to conclude that cold seawa-

ter is being conductively heated as it flows through
channels bounded by breccias. If enough seawater is
being heated in this way, black smoker fluid may be
cooled slightly as it rises through the mound to the
Black Smoker Complex— from the 388°C tempera-
tures found in the anhydrite grains near the base of
the mound to the 366°C temperatures of the exiting
black smoker fluids.

Results of drilling revealed other important infor-
mation on the internal structure of the TAG active
mound. By dating mound materials, it is possible to
reconstruct a history of
the mound, showing that
it has undergone re-
peated episodes of high-
temperature fluid flow,
punctuated by quiescent
periods, over a roughly
20,000-year interval.
When the high-tempera-
ture fluid flow ceases, the
anhydrite dissolves, and
the chimneys that the
anhydrite supports collapse, scattering fragments of
rock. When high-temperature fluid flows resume
and percolate through these fragments, anhydrite
precipitates and serves as a matrix that cements
together fragmented chimney pieces and mound
materials into breccia deposits. Textures within the
TAG breccias indicate that there have been multiple
cycles of mound material reworking — a likely conse-
quence of repeated episodes of anhydrite deposition
and dissolution.

Information from the TAG mound shows how
this kind of intermittent activity can, over long
periods of time, result in the gradual formation of
a large hydrothermal mineral deposit similar to the
ore bodies preserved in Cyprus. Studying the large
TAG active mound has greatly increased our un-
derstanding of how large mineral deposits like
these can form.

Meg Tivey's research on hydrothermal deposits has been
funded through the National Science Foundation and the
Joint Oceanographic Institutions/US Science Advisory
Committee. Her work on the TAG active mound has benefited
from collaborations with many scientists, including Susan
Humphris and Geoffrey Thompson (WHOI). Rachel Mills
(University of Southampton), and Mark Hannmgton (Geologi-
cal Survey of Canada).

Meg Tivey chose to major in geology after taking a course with
five field trips to local beaches and fault zones. She then worked
as a physical science technician at the US Geological Survey
before deciding to pursue graduate studies in marine geology at
the University of Washington. She now specialises in studies of
active seafloor hydrothermal systems. Her current projects
include examining the formation ofpolymetallic sulfide deposits,
continuing work on linking measured vent fluid compositions to
observed mineralogy of vent deposits, using X-ray computed
tomography (CAT scans) to examine seafloor sulfide samples in
llii-ff-dimensions. and working with engineers to build insii n
ments capable of measuring temperatures and flow rates on tin-
seafloor in high-temperature and low-pH fluids.

26 -Vol. 41, No. 2- 1998

The Big MELT

The Mantle Electromagnetic and Tomography Experiment
Was the Largest Seafloor Geophysical Experiment Ever Mounted

Donald W. Forsyth

Professor of Geological Sciences, Brown University

More than 95 percent of the earth's volca-
nic magma is generated beneath the
seafloor at mid-ocean ridges. As the
oceanic plates move apart at spreading ridges, hot
mantle rock rises from deep in the earth's interior to
replace material dragged away by the plates. This
ascent releases pressure that had kept the hot
mantle rocks solid. The mantle rock begins to melt
and form basaltic magma, which then percolates up
toward the surface through cracks and pores in the
remaining solid, yet deforming, rock. At the ridge,
the magma pools and solidifies to form a 6-to-7-
kilometer-thick layer of new basaltic crust. This
crustal layer underlies the seafloor throughout the
ocean basins.

Most of what we know about this process of
magma or melt production beneath ridges is in-
ferred from the volume of melt indicated by the
thickness of the crust and from the composition of
basalts recovered at the seafloor by dredging, drill-
ing, and submersible sampling. Much uncertainty
remains about where and how the melt forms and is
extracted from the mantle, because we have not had
any means to directly probe into the melt produc-

tion region tens to hundreds of kilometers beneath
the earth's surface.

Geophysical measurements, however, can provide
means to probe deep into the mantle to detect the
presence of melt or hot rock. For example, studying
seismic waves that travel through the mantle can tell
us something about the structure of the earth along
their paths, because these waves travel slower in hot
rocks and slower still if there are melt-filled cracks
or pores. In addition, the forces of flow and deforma-
tion within the mantle often cause crystals of olivine
or pyroxene, the most common minerals in the
mantle, to align in particular directions. This phe-
nomenon, in turn, speeds up or slows down the
propagation of seismic waves through the crystals,
depending on the direction and vibration of the
waves. We can also extract information by analyzing
the propagation of electromagnetic waves through
the mantle, because basaltic melt is a better electri-
cal conductor than solid rock. If the melt pockets
are connected together to form a conducting net-
work, the wave propagation is strongly affected.

In previous studies, scientists examined veloci-
ties of seismic waves traveling from their earth-
quake sources through ridges and to land-based
receivers. They mapped regions of slow wave propa-
gation near ridges, but found the data lacked the

An ocean holtoni
seismometer on the
Pacific Ocean floor
measures the velocity
of seismic waves
(ravelin!' through the
mantle and oceanic
crust. These data
allowed scientists to
probe deep within the
portion of the earth
that generates new
ocean floor.

OCEANUS • 27

\ \\IIOIocraii
! mi hnn seismometer

i DBS ) is depiomi

from K \ MrMllr

(Scripps Institution ol

Oceanography ) at I he

start ol Hie Mill I

I \periinenl in

November I 'W.V

Seismometers within

theycllou spheres

record ground motion

Ifencralcd In seismic

\\a\cs traveling

Illl'lll-Jl I 111' III. Illl I,

ami die oceanic crust.

In Ma\ I "«<>. an

.mi In n tin the OKS

was released and I In

instrument floated to

the surface, where il

«asreco\cred.

Mure than '-n ocean

bottom seismometers

(while triangles) were

deployed across and

along the Kasl Pacific

Rise olTthe west cuast

of Mexico dm mi; the

Mill Experiment to

prohc I he deep

structure beneath the

mid-ocean ridi»e.

\rro\\ « .show the

motions ol the l';it ilic

and N.I/I.I plates.

which are spreading

in opposite direct ions

from the ridge creM.

shown In shallower

depths (red).

resolution to pinpoint the width or depth to which
the region of melt production extends. That left
room for two competing theories of how magma is
generated beneath mid-ocean ridges: Some theo-
retical models predict that most of the upwelling
and melting takes place in a narrow zone, perhaps
less than 10 kilometers wide, directly beneath the
ridge axis. In other models, melting extends over a
broad region and the migrating melt is somehow
forced back to the ridge axis to form new crust.

The Mantle ELectromagnetic
and Tomography (MELT) Experi-
ment was designed to distinguish
between these competing theoreti-
cal models by investigating in detail
the structure beneath a spreading
center. In the largest deployment of
geophysical instrumentation ever
attempted on the seafloor. a total of
nearly 100 ocean bottom seismom-
eters (OBSs), electrometers, and
magnetometers were placed in
arrays across the East Pacific Rise
(see map below). More than 50
scientists from 12 research institu-
tions participated. The OBSs were
supplied by groups from Woods
Hole Oceanographic Institution
and Scripps Institution of Oceanog-
raphy (SIO), and electrometers and
magnetometers by scientists from
WHOI, Australia, France, and Japan. Beginning in
November 1995. the seismometers recorded seismic
waves generated by earthquakes around the world
that propagated through the mantle beneath the
ridge. In May 1996, the OBSs' anchors were released
and the instruments were recovered along with their
valuable recordings. The same cruise that retrieved
the OBSs also deployed the electrometers and mag-
netometers to begin a one-year period of recording
electromagnetic waves generated by ionospheric

-109'

-108W

-18'

-19'

0 2700 2800 2900 3000 3100 3200 3300 3400 3500 3600 3700 3800 3900 4000 4100 4500

Depth (meters)

28- Vol. 41, No. 2« 1998

West
400

Pacific
101 m

currents that penetrate deep into the mantle (see
article, page 32).

The East Pacific Rise at 17°S was chosen for the
experiment because it is in the middle of one of the
longest, straightest sections of the mid-ocean ridge
system and is spreading at close to the fastest rate,
about 14.5 centimeters per year. In addition, the
subduction zones of the Pacific Rim provided an
excellent surrounding source of seismic waves, since
earthquakes frequently occur in these zones, where
the seafloor created at the East Pacific Rise eventu-
ally sinks back into the
mantle. We were lucky
that our six-month re-
cording period was
seismically active; there
were several earthquakes
of magnitude 7 or larger,
as well as some good,
smaller events on other
parts of the East Pacific
Rise. One of the best
sources was an earth-
quake 150 kilometers
deep in the Tonga sub-
duction zone. We also
happened to record the
last three underground
tests of nuclear devices in
French Polynesia, but the
signals from these explo-
sions were too small to be
useful for our purposes.
During the deployment
cruise, a group led by
Robert Detrick and Pablo Canales from WHOI and
John Orcutt and Sara Bazin from SIO employed an
array of airguns towed behind the ship as artificial
sound sources to acquire data on the structure of
the oceanic crust (see article on page 30). Differ-
ences in crustal thickness are caused by variations
in the supply of magma to the spreading center and
thus provide additional information about the
process of melt production in the mantle.

Seismic results from the MELT Experiment
appear to have settled the long-standing debate
about the form of the upwelling beneath the ridge.
We found low seismic velocities through a zone
several hundred kilometers wide, indicating the
presence of melt at depths of 20 to more than 70
kilometers. This wide zone suggests that the sepa-
rating plates provide the primary force that drives
upward flow of the mantle beneath the East Pacific
Rise. It contradicts the theory that melting occurs
primarily in a narrow zone directly beneath the
ridge and that upwelling of the solid mantle is
driven and focused by buoyancy forces.

The seafloor begins to subside as it cools and
moves farther from the ridge, but at the East Pacific

Rise, it subsides more slowly on the western flank
than on the eastern. This suggests that the mantle is
hotter to the west. Nevertheless, we were surprised
by the apparent asymmetry of the mantle structure,
as evidenced by the seismic data. The low-velocity
region extends as much as 250 kilometers west of
the axis, but only about 100 kilometers to the east,
and the lowest velocities may even be located west
of the axis. Since there are also many more sea-
mounts on the western, or Pacific Plate, side of the
axis, the asymmetry may be related to melting and

Distance From Mid-Ocean Ridge Axis (kilometers)
200 0 200

East
400

Plate

imeters per year

I I

Oceanic Crust

Pacific Ocean

Nazca Plate

45 millimeters per year-

300 -

upwelling involved in the off-axis volcanism that
builds seamounts. However, this volcanism is much
less voluminous than that caused by the seafloor
spreading process, producing only 1 or 2 percent of
the oceanic crust in this area.

Another intriguing asymmetry was found by
Cecily Wolfe (WHOI) and Sean Solomon (Carnegie
Institution). One type of seismic wave, the shear
wave, splits into two components that travel at
different speeds in the upper mantle; shear waves go
faster when vibrating in a direction parallel to the
alignment of the crystals in the mantle— that is,
when the waves' vibrations are aligned with, rather
than against, the crystalline grain. Wolfe and
Solomon found that shear-wave splitting is twice as
large beneath the Pacific Plate as beneath the Nazca
Plate on the eastern flank, indicating that crystals
may be better-aligned beneath the Pacific Plate.
This difference may be caused by the different rates
of plate motion; relative to the deep mantle, the
Pacific Plate is moving twice as fast to the west as
the Nazca Plate is moving to the east.

We have determined that the melting region, the
region of anomalously low-velocity seismic waves,

Initial results from
the MEI.T Experi-
ment are shown in
this schematic cross-
section of the upper
mantle beneath I he
East Pacific Rise. Hie
melting region below
the mid-ocean ridge
extends over a broad
area several hundred
kilometers wide. The
region is asymmetri-
cal, with a wider /one
west of the ridge than
east of it. The melting
region also extends
far deeper than many
scientists have previ-
ously I hem l/rd: to
depths of 150 to 200
kilometers beneath
the ridge, although
the greatest concen
(ration of melt occurs
above 1 00 kilometers.

OCEANUS • 29

extends to depths of 150 to 200 kilometers beneath
the ridge, although the velocity begins to increase
and melt concentration decreases below about 100
kilometers (see figure on page 29). This depth is
confirmed by modeling of the electromagnetic
signals by Alan Chave and Rob Evans (WHOI) and
Pascal Tarits (Universite de Bretagne Occidentale,
France), which indicates that the mantle may be
conductive at depths of about 180 kilometers. This
is important because petrologists and geochemists
have long debated the depth to which melting ex-
tends. Some argue that most of the melting occurs
at depths shallower than about 60 kilometers. Oth-

ers suggest that as much as 40 percent of the melt-
ing must take place at depths of 70 kilometers or
more, where pressures make the mineral, garnet a
common and stable crystal. Our results indicate
that a significant amount of melting does occur in
the garnet region, but it is difficult to say exactly
how much. Probably no more than 1 or 2 percent of
the mantle exists as melt anywhere beneath the
ridge, even though the maximum degree of melting
may approach 20 percent. The degree of melting can
be much larger than the amount of melt present
because melt migrates effectively through cracks
and tubes toward the surface, removing the melt

Using Seismic Waves to See9
A Slice of the Oceanic Crust

J. Pablo Canales

Postdoctoral Guest Investigator, Geology & Geophysics Department

The primary source of seismic data for the MELT
Experiment came from natural earthquakes. But a
secondary data set was obtained using a large array
of airguns aboard R/V Melville and 15 ocean bottom seis-
mometers (OBSs). The airguns release into the water a
bubble of air compressed to 2,000 pounds per square inch.
When these bubbles pop they create a sound pulse that
travels through the water, penetrates the solid earth several
kilometers down through the crust and upper mantle, and
eventually returns to the seafloor, where it is recorded by the
OBSs. The physics of this phenomenon is basically the same
as that which changes a light beam's trajectory when it en-
ters a different medium, such as a glass, or when it is re-
flected in a mirror. The velocity of the seismic waves through
the earth's multi-layered interior reveals a lot of information
about its structure and composition.

The MELT Experiment was designed to determine the
thickness of the ocean crust in the research area and its
velocity structure (that is, how fast the seismic waves propa-
gate through the crust at different depths). There were sev-
eral important reasons to focus on the crustal structure:

1) Seismic waves originating at remote earthquakes travel
through the earth and arrive at the OBSs after passing
through the crust immediately beneath the instruments.
Before seismologists can use seismic wave measurements to
deduce the distribution of melt in the mantle, they must be
sure that variations in oceanic crustal structure are not
affecting their data.

2) Since the distribution of melt in the mantle affects its
density, it also affects the local gravity field, so gravity mea-
surements provide additional information with which to
deduce how melt is distributed deep beneath the seafloor.

But the gravity field is also affected by variations in the den-
sity and thickness of the oceanic crust. Once again, we must
quantify and account for the crustal contribution to the
gravity field before using gravity data to interpret the mantle.
3) In the area of the MELT Experiment, the Pacific Plate,
west of the East Pacific Rise ridge, has far more abundant
seamounts than the Nazca Plate, east of the ridge. Seismic
measurements can help us to determine if the more abun-
dant volcanism in the Pacific Plate is, as expected, associ-
ated with the formation of a thicker crust.

In our study we analyzed the travel times of three types of
seismic phases. A single seismic wave generated from a
single airgun shot consists of several phases, depending on
the number of rock layers that it travels through. Pg rays are
refracted within the crust; PmP rays reflect off the Moho
(the mantle-crust transition); and Pn rays are refracted in
the upper mantle (see figure opposite). Respectively, these
waves provide information on the speed of waves traveling
through the crust, the thickness of the crust, and the speed
of waves traveling through the mantle. We first obtained the
crustal velocity structure using the Pg phase. Then, we ob-
tained the average crustal thickness in four regions (at the
center and end of a ridge segment in both the Pacific and
Nazca plates) by matching the observed PmP and Pn travel
times with travel times predicted by various models of
crustal thickness. We tested a variety of models, changing
the crustal thickness from 4 to 7 kilometers, and selected
those that best fit the data.

Overall we did not find a resolvable difference in crustal
thickness between the Pacific and Nazca plates. This implies
that the asymmetries in depth and gravity measurements
observed on each side of the ridge axis must be caused by

30 'Vol. 41, No. 2« 1998

almost as fast as it is generated and leaving only a
small concentration behind in the mantle matrix at
any one time.

The MELT Experiment created a huge archive of
geophysical data. It will probably take several years
to complete analyses of this unique experiment and
to create a new generation of models of mantle flow
and melt production that will satisfy the many
powerful constraints provided by these observa-
tions. Even the preliminary results described here,
however, have significantly altered our understand-
ing of seafloor spreading and the formation of the
oceanic crust.

The MELT Experiment was funded by the National Science
Foundation through the RIDGE Program.

Don Forsyth is a marine geophysicist and earthquake seis-
mologist who concentrates on problems of plate tectonics: How
thick are the plates? What drives plate motion? How do new
plates form at mid-ocean ridges? Don graduated from the WH01/
MIT Joint Program in 1974 and has maintained connections with
\VHOI scientists ever since. He currently is chair of the Depart-
ment of Geological Sciences at Brown University. His first re-
search cruise was on R/V Chain out of Woods Hole. His latest was
in the Indian Ocean studying interactions ofhotspots and mid-
ocean ridges. On shore, he loves playing squash and basketball

Pg Phase PmP Phase — Pn Phase T Ocean Bottom Seismometers

o

0)

01
I/I

E
o

a
m
a

West 1 50

100 50 0 50

Distance from Ridge Axis (kilometers)

100

1 50 East

To measure the thickness of oceanic cnist near the F.ast Pacific Kise, \VHOI scientists measured the velocity of airgun-generated seismic waves that travel
through the crust, refract off rock layers, and are recorded by ocean bottom seismometers. A single seisimic wave from a single airgun shot consists of
several phases, or rays, depending on the number of layers it penetrates. I'g rays (blue) are refracted within the crust and provide information on the speed
ol 'waves traveling through it. I'nil' rays (orange) reflect off the Molio (the mantle-crust transition boundary) and provide data on crustal thickness. Pn rays
(green) are refracted in the upper mantle and otfer data on the speed of waves traveling through the mantle. Only one of every 10 rays is plotted.

density variations in the mantle, not the crust. Evidence for
the asymmetry in mantle structure was observed in the
teleseismic (waves from distant sources) data recorded on
the same OBSs (see article on page 27). The crustal thick-
nesses measured along a primary array of OBSs were 4.8 to
5.6 kilometers on the Pacific Plate and 5.1 to 5.7 kilometers
on the Nazca Plate. Along a secondary array, crustal thick-
nesses measured 5.4 to 6.2 kilometers on the Pacific Plate
and 5.8 to 6.3 kilometers on the Nazca Plate. So the more
abundant volcanism in the Pacific Plate is not creating a
thicker ocean crust.

We did observe some smaller-scale, local differences in
crustal thicknesses. The most noticeable and surprising one
is in the Nazca Plate, where we found thinner crust along a
line that crossed an inflated section of the ridge over a "melt
lens reflector"— an image of the roof of a chamber, where
magma has accumulated, and an indication of a robust
magma supply. We found thicker crust along a line that
crossed near the end of a ridge segment, where there is no

magma lens or other indications of a large magma supply.
The crusts in both areas are of similar age (0.5 million to 1.5
million years), but their thicknesses differ. Along with subse-
quent tectonic studies, this suggests that the configuration
of the ridge has changed since those crusts were formed,
moving crust from a volcanically active to an inactive loca-
tion, and vice versa.

WHOI participants in the the MELT experiment were funded by the
National Science Foundation. J. Pablo Canales was also supported by the
Mimsterio de Educacion (Spain)/Fulbright Program.

Pablo Canales conducted his Ph.D. thesis at the Institute of Earth Sciences of
Barcelona (Spain), studying the structure of the oceanic lithosphere affected
by hotspots in "exotic" areas, such as the Canary Islands, Tahiti, and the
Galapagos Islands. He first came to WHOI in 1 996 as a guest graduate
student and since April 1997 has been a postdoctoral guest investigator,
funded by a grant from the Commission for Educational Exchange between
the US and Spain (Fulbright Program) and the Department of Education of
Spain. His research has focused on seismic studies of mid-ocean ridges,
including the East Pacific Rise and the Mid-Atlantic Ridge.

OCEANUS'31

I In1 conductivity ol
the mantle beneath

the liast Pacific Rise

I- drpn Iril ill I hi-

example of an nun

MOM rcMill from

magnetotelluric

technique (MT) data

collected during the

Ml-l.l i:\pcrimonl.

Warm colors (white,

u'llou orange, and

rod) represent in-

creased conduct ivity

(lower resisli\ity ).

Cold colors (green,

blue, black) represent

loun conductivity'

(higher resistivity1).

I lie upper .id

kilometers ol I he

maiille appear to be

more conductive

beiical hi be Pacific

I'l.ilr ur.| oflhe

i ulur ill. in on Ilie

\a/ca Plate, east of

the ridge. The region

ol ln.Ji conduct ivity.

extending about <so

kilometers uest ot

I be ridi>c crest and

I mi,, I HO kilometers

deep. siii^esls deep

melting processes

allecled by I lie

presence ol water,

or it may simply

relied I lit- filed ol'

ualcron the mantle

resistivity itself.

A Current Affair

A New Seafloor Technique Measures
Electrical Conductivity Deep Within the Earth

Robert Evans

Associate Scientist, Geology & Geophysics Department

The MELT Experiment was the largest
seafloor geophysical experiment ever
attempted, and one of its major compo-
nents was MT, the magnetotelluric technique. MT
offers a valuable tool toward the MELT Experi-
ment's goal of probing the earth's inaccessible deep
interior. But the technique remains something of a
mystery even to many marine scientists. It has been
used widely on land, particularly for regional-scale
surveys, but only a few full-scale MT surveys have
been carried out on the seafloor.

The primary data collected by marine MT experi-
ments are measurements of changes in the earth's
electrical and magnetic fields at the seafloor. These

Distance from Mid-Ocean Ridge Axis (kiloi
134 80 27 0 27

neters)

34 187

fields are affected by electromagnetic currents within
the earth, and here's where MT's apparent complexity
starts— because the source of these currents is not
within the earth, but rather in the ionosphere.

Charged particles, emitted from the sun as a solar
wind, become trapped in the ionosphere by the
earth's magnetic field. These moving charges essen-
tially create a variety of electric currents encircling
the earth. If the earth were a perfect insulator, like
space, that would be the end of the story. But the
earth can conduct electricity. As these ionospheric
currents flow around the earth, they generate a
response within the planet itself. More specifically,
the pattern of ionospheric currents induces almost a
mirror-image pattern of currents within the earth.

These so-called "induced image currents" cause
changes in the earth's electric and magnetic fields.
These changes depend on the conductivity of the

earth's interior, which, in turn, is determined by the
composition and structure of the materials that
constitute our planet's interior. Thus, by measuring
changes in Earth's electric and magnetic fields at
the surface, we can effectively deduce its electrical
conductivity and reveal its interior structure. As
CAT scans reveal images and frameworks that
enable us to learn about the workings of the human
body, MT experiments similarly provide essential
cutaway views that allow us to learn about pro-
cesses taking place within our planet.

Like standard alternating currents in most house-
holds, which have a frequency of 60 Hertz, or one
cycle per 1/60 of a second, induced image currents

also alternate— though they do
so over a wide range of fre-
quencies. The variations, or
frequencies, we use in seafloor
MT range from periods of
about 100 seconds to several
hours. These variations are
caused by the chaotic nature of
the events that entrap ions
from the solar wind, as well as
by more regular events, such as
the earth's daily orbit around
the sun. The important point is
that different frequencies
penetrate the earth to different
depths. If induced image cur-
rents came in only one flavor,
we would be able to image the earth's interior at only
one depth. As it is, higher-frequency currents (with
one cycle per 100 seconds, for example) don't pen-
etrate deeply and can tell us about structure 10 to 15
kilometers deep; the lowest-frequency currents (with
one cycle per several hours) can tell us about depths
of several hundred kilometers.

The goal of the MELT Experiment was to map
basaltic melt, from its source within the mantle to
the base of the oceanic crust at the mid-ocean ridge
crest. While the earth can conduct electrical cur-
rents, most rocks, including those comprising the
mantle, do not conduct particularly well. This situa-
tion changes considerably when melt is present:
Pure basaltic melt is several orders of magnitude
more conductive than olivine, a common mantle
mineral. In the mantle melting column, we do not
expect to see pure melt, nor anything like it, but

32- Vol. 41, No. 2 • 1998

rather some distribution of streams and pools of
liquid melt within a matrix of solid mantle rocks. In
this case, how the melt is distributed is important.
It is possible to think of the melt as a network of
wires that connect parts of the mantle. If the melt
forms a well-connected network through the rock,
electric currents can flow and the mantle will be
electrically conductive. Of course, reality is more
complicated and other factors, such as water dis-
solved in the mantle rock, can affect conductivity.
These other factors are also important for under-
standing the whole process of melt production.

The MT component of the MELT Experiment
was a truly multinational effort involving more
than a dozen scientists
from Woods Hole
Oceanographic Institu-
tion and Scripps Institu-
tion of Oceanography in
the US, and from France.
Japan, and Australia.
Each group contributed
instruments to the array
and played a role in the
data analysis. From June 1996 to June 1997. 47 in-
struments were deployed at 32 seafloor sites to
measure the time variations of the electric and
magnetic fields. Two lines were set out. The main
southern line had 19 sites and crossed a magma-
rich segment of the East Pacific Rise ridge crest,
extending 200 kilometers on either side of the crest.
The second line of 13 sites crossed the ridge to the
north on a magma-starved ridge segment, extend-
ing 100 kilometers on either side of the axis.

Each group's instruments essentially did the
same thing: measure changes in the electric and
magnetic fields at the seafloor. But each group
accomplished this in slightly different ways, deploy-
ing very different-looking instruments. As in all
marine experiments, the environment makes sea-
floor MT measurements more difficult to make, but
in one way nature helps us. The ocean is electrically
very conductive and acts as a screen against elec-
tromagnetic noise — extraneous signals from other
sources that would confuse interpretation of the
data. On land, power lines, for example, can be a
nuisance. The seafloor, however, is electrically quiet,
making it possible to measure very small electric
field variations. The other part of the MT signal is
the seafloor magnetic field— not the steady field
trapped in lavas and used to identify magnetic
reversals, but the magnetic field variations linked to
ionospheric currents.

To a first order, the ratio of the electric to the
magnetic field at the earth's surface is a direct mea-
sure of the earths electrical conductivity. We calcu-
late this ratio for a range of current frequencies using
modern processing techniques. To produce a model
of the earth, data from all instruments have to be

examined through a process of numerical inversion.
The interaction of induced currents in the earth with
the conductive bodies we hope to image (such as the
melt column) affects the electric and magnetic fields
over a wide region of seafloor. Generally, it is not
possible to look at data from a single instrument and
interpret the underlying structure. Instead, we have
to use computer modeling to predict the fields that
the mantle would create and compare these answers
to data from all the instruments. The model is up-
dated to improve the agreement and the process is
repeated until a satisfactory model is found. There
are many pitfalls involved in this process, as well as
different ways of carrying it out. The groups involved
in the MELT Experiment
have been using a variety of
methods over the past few
months, and we are in the
process of comparing re-
sults and discussing their
implications.

The MT analyses are
still in their early stages,
but some first-order re-
sults are beginning to come through. The MT data
show an asymmetrical distribution of melt between
the areas west and east of the ridge crest, with a
more extensive region to the west. The melt col-
umn also appears to be a broader feature, with a
low percentage of melt in it, rather than a narrow
vertical column of melt directly beneath the ridge.
This indicates a more passive flow of mantle to-
ward the ridge crest. Deeper, we see some evidence
for a conductive mantle at depths greater than 150
kilometers. If this proves to be true, it could be
evidence for deeper melting— deeper than the part
of the mantle generally believed to be responsible
for most melt generation. However, in the final
analysis, water dissolved in the mantle rock may
prove an important factor in mantle conductivity
at this depth.

Funding for the MELT Experiment was provided by the
National Science Foundation through the RIDGE Program. The
many people involved in the MT component of MELT include:
Alan Chave, Bob Petitt and John Bailey (WHOI). Jean Filloux
and Helmut Moeller (SIO). Pascal Tarits (Universite de Bretagne
Occidentale), Martyn Unsworth and John Booker (University of
Washington), Graham Heinson and Anthony White (Flinders
University. South Australia), and Hiroaki Toh. Nobukazu Seama
and Hissashi Utada (University of Tokyo).

Rob Evans was an undergraduate in the Physics Department at
Bristol University in the UK when he saw an advertisement for a
Ph.D. project that involved a cruise to the East Pacific Rise. Not
letting the fact stand in Ins way that he knew next to nothing
about what a mid-ocean ridge was. he applied for the studentship
at Cambridge University, and the cruise to a sunny location. Since
then. Rob has worked with most of the groups worldwide that
carry out seafloor electromagnetic work. He did a postdoc in
Toronto, Canada, before coming to WHOI as an Assistant Scien-
tist. His lack of hair comes from the stress of doing marine science
and has no relationship to his heavy use of EM fields.

Rob Evans (left) and

llrlllllll Miirlln ill

Scripps deploy a

\\ HOI m.ii;Mr1innrlrl

Iroin I'. \ rinttn/t\on
(I'nhvrsil) nj
Washington),

OCEANUS • 33

A llol -|iol i ir.llrtl I lie

1 4.i ml of Iceland and
its characteristic
volcanic landscape.
Hotspots are rela-
tively small regions
on I he earth \\licrc
unusually hoi rocks
rise from deep inside
I he mantle layer.

Hitting the Hotspots

New Studies Reveal Critical Interactions
Between Hotspots and Mid-Ocean Ridges

Jian Lin

Associate Scientist, Geology & Geophysics Department

The great volcanic mid-ocean ridge system
stretches continuously around the globe for
60,000 kilometers, nearly all of it hidden
beneath the world's oceans. In some places, how-
ever, mid-ocean ridge volcanoes are so massive that
they emerge above sea level to create some of the
most spectacular islands on our planet. Iceland, the
Azores, and the Galapagos are examples of these
"hotspot" islands— so named because they are
believed to form above small regions scattered
around the earth where unusually hot rocks rise
from deep inside the mantle layer.

But hotspots may not be such isolated phenom-
ena. Exciting advances in satellite oceanography,
seismology, geochemistry, and geodynamics, along
with a treasure trove of new declassified data, are
revealing that hotspots appear to have important
and far-reaching impacts on a surprisingly large
percentage of the global ridge system. Some 44
hotspots have been identified around the globe, and
a large number of them are integrally connected to
the ridge system (see map at right). Indeed, these

hotspots may play a critical role in shaping the
seafloor— acting in some cases as strategically
positioned supply stations that fuel the lengthy
mid-ocean ridges with magma.

Studies of ridge-hotspot interactions received a
major boost in 1995 when the US Navy declassified
gravity data from its Geosat satellite, which flew
from 1985 to 1990. The satellite recorded in unprec-
edented detail the height of the ocean surface. With
accuracy within 5 centimeters, it revealed small
bumps and dips created by the gravitational pull of
dense underwater mountains and valleys. Research-
ers often use precise gravity measurements to probe
unseen materials below the ocean floor. In places
where the seafloor contour has been well-charted by
ship surveys, we can employ modeling to remove
the gravitational effects of seawater and the sea-
floor. The leftover signal, called the Bouguer
anomaly, reveals information about the rocks be-
neath the ridges and hotspots. Using this tech-
nique, we have detected unusually thick, hot crust
and mantle rocks beneath virtually all major

34- Vol. 41, No. 2 • 1998

hotspots located near ridges, including Iceland and
the Azores in the northern Atlantic Ocean, Tristan
de Cunha in the southern Atlantic Ocean, the
Galapagos and Easter Islands in the Pacific Ocean,
and the Marion and Bouvet hotspots in tin
southwest Indian Ocean.

Beneath the earth's thin
outer skin (called the lithos-
phere). mantle rocks creep
plastically in a layer known as
the asthenosphere, where
temperatures stay near the
rocks' melting point. Below the
mid-ocean ridges, mantle rocks
rise to fill the gap between two
separating plates. As they as-
cend, some mantle rocks liquefy
to form basaltic magmas, or
melts, and the buoyant magmas
float to the top of the mantle to form
oceanic crust.

By studying the chemical composition of rocks
dredged from the ocean floor, researchers have
determined that most melts are probably produced
at depths of 20 to 80 kilometers and at tempera-
tures of 1,150° to 1,400°C. Below a hotspot, however,
geochemical evidence shows that mantle rocks may
start melting at greater depths and higher tempera-
tures, giving rise to voluminous lavas and melts that
solidify to form shallow ridges such as the
Reykjanes Ridge near Iceland, underwater volcanic
plateaus such as the vast Kerguelen Plateau in the
southern ocean, hotspot islands such as Hawaii, and
smaller submerged volcanoes called seamounts.

By measuring the precise travel time of seismic
waves passing through the mantle rocks under
Iceland, researchers have recently identified a sur-
prisingly narrow cylindrical "root" of anomalously
hot rocks extending to at least 400 kilometers
beneath Iceland (see figure at
left). Below 400 kilometers,
seismic data cannot be easily
gleaned, but more indirect
seismic evidence suggests that
this narrow root, which is
about the width of Iceland,
may be underlain by a region of
anomalously hot mantle as
deep as 660 kilometers. Theo-
retical geodynamic models show
that such a narrow hotspot root
would generate a vigorous
source of heat, providing a huge
volume of lava to build and feed
Iceland's volcanic landscape. Even
more dramatically, this heat source
would also cause mantle rocks to migrate laterally
in the asthenosphere hundreds of kilometers out
from Iceland (see figure on page 36).

The Geosat data show that in the cases of Ice-
land and the Azores, a bulge of unusually thick and
elevated crust extends from the two hotspots in
both directions along the Mid- Atlantic Ridge. This
region of thick, hot crust and mantle extends 1,300
kilometers north and south of the Iceland hotspot
and some 1,000 kilometers north and south of the
Azores hotspot. Together, the two hotspots appear
to be feeding a huge supply of magma to nearly the

K\ mr

precise ii.acl limes
of seismic waves
passing through

mantle rocks hciu'alh
Iceland, scientists
hii\e recently idcnli-
linl .1 surprisingly
M.II i im and deep
cyliiulrical"roiil"ol
anomalously hot
rocks, extending lo ;i

depth III' ill Ir.lsMIHI

kilometers, and
proliiilily farther.

From: Woll'i-. B|arnason.
VanDecar, and Solomon. 1996.

A map of the world's

major holspols slums
that many ollheniare
integrally connected
to lhc!>lol>al mid-
ocean ridge system
( retl lines}, dreen
lines indicate snh-
dnclion /ones, « here
plates pinnae hack
into the mantle. Not
slumn is the |.m
May I'll hot spot north
of Iceland.

OCEANUS • 35

licccnlh declassified

x. ilrlhl r yi.Mih il.il.i
reveal bulges ol
unusually thick and
elevated oceanic crust
( red. yellim and green
on the map) extend-
ing hundreds of
kilometers from the
Ireland and Azores
holspols. Thehvti
holspols may be
feeding huge supplies
..I i u.i^i i i.i lo nearly
the entire northern
.segment ol'the Mid-
Atlantic Ridge.

Mjp produced by Jennifer Georgen,
M1T/WHOI Joint Program, with
data from of David Sandwell.
Scnpps Institution of Oceanogra-
phy, and Walter Smith. NOA.A

A theoretical
geodynamic model

slums that a narnm
Imlspol "root." such
as the one recently
discovered beneath
Iceland, \vonkl gener-
ate a vigorous source
ol heal (light colors)
and produce a huge
\olnine ol magma
that uiuilil migrate
laterally along and
across the ridge axes
hundreds ol kilome-
ters a\\ay from Ihe
hotspol source.

70'N

60"N

50"N

40"N

30'N

20"N

10'N

10"5

70"W

60"W

I — I I I I I I — I — I — L— L

-100 -80 -60 -40 -20 0 2

FAA (mGal)

entire northern segment of the Mid- Atlantic Ridge.
These hotspots, like others, don't seem to provide a
steady supply of magma, but rather periodic surges.
What regulates this episodicity is an intriguing

Across

, (lolome

ters)

Hotspot

0

question currently under
investigation.

In the far southern
Atlantic, a string of
hotspots — Tristan,
Gough, Discovery and
Shona— align with the
southern Mid-Atlantic
Ridge, and may be con-
tributing magma sup-
plies that maintain the
ridge (see map opposite).
Just to the east of the
Shona hotspot lies the
Bouvet hotspot, whose
geological importance
may go far beyond cre-
ation of the tiny, remote
island of Bouvet. Re-
searchers speculate that
the hotspot may play a
critical role in maintain-
ing the triple junction,
where the mid-ocean
ridge branches that
define the boundaries of
the South American.
African, and Antarctic
plates all intersect. Far-
ther east, the promi-
nence of the Kerguelen,
Crozet, and Marion
hotspots suggests that
ridge-hotspot interac-
tions have helped shape
the seafloor of the
southern ocean.

In the western Pa-
cific, our studies show a
complex pattern of
interaction between the

Galapagos hotspot and the Cocos-Nazca Ridge, the
boundary between the Cocos and Nazca plates.
About 8 million years ago, the ridge and hotspot
collided and the hotspot-induced melt flux was
estimated to be very
robust. But comparisons
of older and younger
crust in the region reveal
that the intensity of the
ridge-hotspot interaction
has been diminishing as
the Cocos-Nazca Ridge
gradually moved away
from the Galapagos
hotspot. Today the two
are 200 kilometers apart, and we
theorize that the hotspot will lose its im-
pact on the ridge when the two reach a distance

60

80

Along Ridge Axis (kilometers)
800

36- Vol.41, No. 2- 1998

of about 500 kilometers apart.

Emerging geophysical and geochemical
data indicate significant dissimilarities
among; hotspots and tell us that the inter-
actions between individual hotspots and
ridges have their own peculiar dynamics.
While hotspots such as Iceland and Hawaii
may have their origin deep in the mantle.
ot hers may simply reflect large concentra-
tions of unusual chemical properties in the
earth's shallow mantle. While Iceland's
impact on the ridge extends up to 1,300
kilometers north and south, the Marion
hotspots effect on the Southwest Indian
Ridge diminishes after only 300 kilometers.
Magma eruption rates vary among differ-
ent hotspots, and each hotspot has its own distin-
guishable geochemical signature, which provides
clues to understand how melting occurs at each
hotspot. Iceland and the Azores, for example, both
show an elevated ratio of stronium 87/stronium 86
isotopes in the basaltic rocks they produce, com-
pared with basalts produced in Mid-Atlantic Ridge
regions farther south. However, basalts generated by
the Azones hotspot do not exhibit the same el-
evated helium 3/helium 4 isotopic ratios that the
Icelandic basalts do.

Rapidly progressing techniques in geochemistry,
as well as in broadband seismology, autonomous
underwater geophysics, and geodynamic model-
ing—combined with the new global geophysical

20'S

30'S

40'S

data coverage — should make the coining decade a
most exciting time to study the fascinating geologi-
cal phenomenon of ridge-hotspot interactions.

The National Science Foundation supported the research
described. The author is indebted to MIT/WHOI Joint Program
graduates Garrett Ito (now at University of Hawaii), Javier
Escartin (now at CNRS. Paris, France), and current student
Jennifer Georgen for sharing exciting time at sea and ashore in
studying mid-ocean ridges and oceanic hotspots.

lian Lin joined WHOI in 1 988 after graduate study at Brown
University and research on earthquakes at the US Geological
Survey in Menlo Park, California. He divides his travel time
between going to sea and investigating quakes in California. His
latest expedition was on board the French ship LAtalante to the
Mid-Atlantic Ridge south of the Azores hotspot, where spectacu-
lar new underwater volcanoes were discovered.

60'S

-100 -80 -60 -40 -20 0 20 40 60 80

FAA(mGal)

Author Jian Lin
raptured this image

ol ll I'l.llllU Mill ,11111

landscape during a
research expedition.

Satellite gravity map

nl ilir urxin ii Indian
and southern Ml. ml ii
Ocean basins, as
i I-M-. iln I by satellite
gravity data. A string
ol holspolsl Tristan,
dough. Discovery,
and Shona) aligned
mill the southern
Mid-Atlantic liidgc
(MAR (maybe con-
tributing magma
supplies that main-
tain the ridge. Hie
Itomcl hot spot may
play a critical role in
maintaining the
triple junction, ulirn
three mid ocean
ridge branches and
the three boundaries
(white lines) separat-
ing the South Ameri-
can. Antarctic, and
African plates all
intersect. Farther
east, the Kergiiclen.
C ro/el. and Marion
holspolsinay ha\e
extensively shaped
the sealloor ol the
southern ocean.
SEIR is the Southeast
Indian Kidge.
OR is the Central
Indian Ridge, and
SXVIK is the South-
west Indian Kidge.

MIT/WHOI Joint Program, with
data ftom David Sandwell. Scnpps
Institution of Oceanography, and
Walter Smith. NOAA

OCEANUS • 37

>ny of small mussels encrusts the surface of a black smoker on me summit oflne Luc
photo was taken by WIIOI's remotely operated vehicle /»,«»« at a depth of 1,700 meters below sea level.

mount at :i7 ' \ on the Mid-Atlantic liidj-e

'bods Hole Oceanographic Institution

Woods Hole, MA 02543 • 508-457-2000
www.whoi.edu