Full text of "Oceanus"
REPORTS ON RESEARCH FROM WOODS HOLE OCEANOGRAPHIC INSTITUTION
VOL.42, NO.2 • 2004 '^ISSN 0029-81 82
" -
-
is/ap
'•"•**.
.-
-*: j?r.
^11
Investigating Earth's dynamic processes
This may sound like heresy, but for
some of us at Woods Hole Ocean -
ographic Institution, the ocean is a
bit of a nuisance. All that lovely blue water
can get in the way.
The ocean is a barrier that impedes
our ability to understand how our planet
works. It conceals powerful and fascinat-
ing forces that are constantly shaping and
reshaping Earth's surface in ways that
really make our planet unique.
Earth is not cold, dead, and static. Its
surface is hot. It's moving. It's constantly
changing. But just a half-century ago,
we were largely unaware of our planet's
extraordinary dynamism because it pri-
marily occurred in a place where we
couldn't easily observe it.
The ongoing, fundamental forces that
forge our planet — that generate earth-
quakes and volcanoes; that perpetually
create and destroy Earth's crust; that rip
apart continents and smash them into one
another; that create mountains like the
Himalayas and island chains like Hawaii;
that open and close ocean basins; that
forge mineral deposits and generate oil
and gas; and that brew chemical cauldrons
that sustain rich communities ot lite in the
sunless depths — most of this action occurs
beneath the oceans.
Water is a blessing that supports life
on Earth, but it is a dreadful medium for
exploration. It is largely impenetrable to
light, so we can't see through it. We can't
view most of Earths surface with a tele-
scope, as we can with Mars. Flying vehi-
cles through the viscous medium ot water,
under conditions of crushing pressure and
complete darkness, poses daunting techni-
cal challenges.
Although we have fully mapped the
waterless surfaces of Venus, Mars, and
the moon in detail, we have mapped
only 5 percent of the entire seafloor at
the same resolution. Just 50 years ago,
we were as ignorant about our home
planet as we were about our solar system
nearly 500 years ago— before the astron-
omer Copernicus told us that the Earth
revolved around the sun, rather than
vice versa.
Since then we have learned that the
seafloor is not some vast, placid beach. In
the 1950s and 1960s, we discovered that
the globe is encircled by an active volcanic
mountain chain. It bisects the ocean floor
and stretches continuously for more than
75,000 kilometers (45,000 miles)— more
than five times the length of the Andes,
Rocky, and Himalayan mountains com-
bined. Over millions of years, this mid-
ocean ridge system continually spews
lava, creating new ocean crust that repaves
most of the planet's surface.
The seafloor is also rite with deep
trenches, where old, cold ocean crust sinks
back into Earth's interior and is recycled.
At both ridges and trenches, volcanism
and earthquakes are rampant. Indeed,
about 80 percent of volcanic and seismic
activity on Earth occurs under the sea.
Fueled by heat emanating from Earth's
core, the engine that drives much of this
activity is the mantle— the layer of our
2 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
•
planet between the crust and core. At the
high temperatures and pressures found
within Earth's mantle, solid rocks can
deform. (Think about how a blacksmith
heats iron to a temperature just below its
melting point to bend and shape a horse-
shoe.) Solid rocks within Earth's mantle
can flow, with hot buoyant material rising
and cold, dense material sinking.
This convection drives the motions of
our planet's thin, rigid outer layer, which
is broken into Earth's great tectonic plates.
The plates move apart and together, con-
tinually (albeit slowly) changing the face
of the planet. The con-
tinents atop the plates
are carried along as
passive riders.
In some cases, we
are learning that the
rending and collisions
of continents have led
to changes in the cir-
culation ot the oceans,
or the atmosphere,
or chemicals cycling
among the Earth,
can spawn changes in Earth's climate.
In the late 1970s, the surprising dis-
covery of life thriving at deep-sea hydro-
thermal vents revolutionized our con-
cepts of where and how life can exist. An
abundance of life flourishes in conditions
we had considered too extreme, sup-
ported by chemicals created by processes
occurring within the planet itself. More
recently, we have seen evidence that pre-
viously unimagined and potentially huge
communities of microbial life reside deep
within the Earth. These discoveries have
fundamentally changed our perspective
on the origins of life
on Earth and redi-
rected our approaches
to searching for extra-
terrestrial life.
Unforeseen discov-
< eries — such as plate
- tectonics and chemo-
4 synthetic deep-sea
^ life — have transformed
3 our understanding ot
Susan Humphris, the first Director of the Earth. But oceanog-
WHOI Deep Ocean Exploration Institute, raphy is a very young
ocean, and atmosphere, peers through an Alvin viewport before science. The oceans
All of these, in turn, the sub descends to the seafloor. remain a frontier.
Dramatic advances in deep-submer-
gence vehicles and technologies now
provide the potential for unprecedented
access to the oceans and seafloor — and
unprecedented discovery. New robotic
systems and oceanographic instruments
are being developed to remain in the
oceans for long periods — to go beyond
learning what's down there and begin to
make inroads into learning more about
what's going on down there.
That is the mission of the Deep Ocean
Exploration Institute at Woods Hole
Oceanographic Institution: to investigate
Earth's dynamic processes by exploring
the frontier where they are occurring. We
journey into uncharted waters— or more
precisely, under them— to reveal the his-
tory and natural engineering of the planet
we call home.
— Susan Humphris
Susan Humphris is Chair of the WHOI
Geology and Geophysics Department.
She was the first Director of the WHOI Deep
Ocean Exploration Institute, serving from
2000 until lime 2004, when she was succeeded
bv Dan Fornari.
Composite photo, above: IMAX film by William Reeve and
Stephen Low, Stephen Low Productions
l/oods Hole Oceanographic Institutio
Contents
Motion in the Mantle:
The Engine that Drives the Earth
6
7
13
Motion in the Mantle
New tools and techniques reveal the inner workings of
our planet — Robert Detrick
Conduits Into Earth's Inaccessible Interior
Hot plumes from deep within the planet bring up telltale
chemical clues about the mantle — Stan Hart
If Rocks Could Talk. ..
The ion microprobe extracts hidden clues about our
planet's history and evolution — Nobu Shimizu
I £ Listening Closely to 'See' into the Earth
A new national facility of cutting-edge seafloor seismo-
graphs probes Earth's interior — John Collins
Emerging Ocean Technology:
New Windows into the Oceans and Earth
TA Realizing the Dreams of da Vinci and Verne
A diverse fleet of innovative deep-submergence vehicles
heralds a new era of ocean exploration — Dan Fornari
~) C Unique Vehicles for a Unique Environment
New autonomous robots will pierce an ice-covered ocean
and explore the Arctic abyss—Robert Reves-Sohn
TO Seeding the Seafloor with Observatories
Scientists extend their reach into the deep with pioneer-
ing undersea cable networks — Alan Chave
0 ~\ A Sea Change in Ocean Drilling
Scientists launch a new drill ship and ambitions research
plans— Dennis Nonnilc and Richard A. Kerr
Creating the Seafloor:
The Formation of New Ocean Crust
0 £ Earth's Complex Complexion
Expeditions to remote oceans expose new variations in
ocean crust— Henry J.B. Dick
AC\ Unraveling the Tapestry of Ocean Crust
Scientists follow a trail of clues to reveal the magmatic
trickles and bursts that create seafloor — Peter Kelemen
A A Paving the Seafloor — Brick by Brick
New vehicles and magnetic techniques reveal details of
seafloor lava flows — Maurice Tivey
Earthquakes and Volcanoes:
Linking the Engine to Earth's Movements
^10 Earthshaking Events
Research on land and sea reinvigorates hopes of forecast-
ing where earthquakes are likely to occur — fian Lin
54
Ears in the Ocean
Hydrophones reveal a whole lot of previously undetected
seafloor shaking going on — Deborah Smith
C 7 Peering into the Crystal Fabric of Rocks
When you get right down to it, earthquakes and volca-
noes have atomic-scale causes — Greg Hirth
COVER: A striated tongue of basalt lava — extruded from the seafloor
like toothpaste from a tube — lies atop older lava at the Galapagos
Spreading Center 1,668 meters (5,463 feet) below the surface of the
Pacific Ocean. In 2002, scientists diving in Alvin discovered an exten-
sive community of giant clams up to 1 foot long (Calpytogena mag-
nifica) thriving on low-temperature hydrothermal fluids venting from
cracks in seafloor lava. (With grateful acknowledgement to David
Metz of Canon, Inc. -USA and George Moss for the loan of a Canon
EOS ID digital camera system used in Alvin to take these high-reso-
lution photographs) Photo credit: Tim Shank, and the Alvin Deep
Submergence Operations Group
4 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Earth-Ocean Interactions:
Where the Solid and Liquid Earths Meet
£T\ The Remarkable Diversity of Seafloor Vents
Continuing explorations reveal an increasing variety of
hydrothermal systems — Margaret Kingston Tivey
£ £. When Seafloor Meets Ocean, the Chemistry Is Amazing
In more places, scientists are finding large amounts of
natural gas on the ocean bottom — Jean Whelan
Life on Earth:
The Co-evolution of Life and the Planet
~1~) Is Life Thriving Deep Beneath the Seafloor?
Recent discoveries hint at a potentially huge and diverse
subsurface biosphere— Carl Wirsen
"70 The Evolutionary Puzzle of Seafloor Life
Scientists are assembling critical pieces to reconstruct the
history of life on the ocean floor — Timothy M. Shank
0£ Living Large in Microscopic Nooks
New-found deep-sea microbes rearrange thinking on the
evolution of the Earth, and life on it — Katrina Edwards
Shifting Continents and Climates:
How Earth Dynamics Change the Climate
QQ Shifting Continents and Climates — Mike Carlowic:
91
95
Moving Earth and Heaven
Colliding continents, the rise of the Himalayas, and the
birth of the monsoons — Peter Clift
How the Isthmus of Panama Put Ice in the Arctic
Drifting continents open and close oceanic gateways
— Gerald Hang, Ralf Tiedemann, and Lloyd Keigwm
1930
EDITOR: Laurence Lippsett
CONTRIBUTING EDITORS: Mike Carlovvicz, Vicky Cullen, & Kate Madin
DESIGNER: Jim Canavan, WHOI Graphic Services
WHOI PRESIDENT AND DIRECTOR: Robert B. Gagosian
CHAIRMAN OF THE BOARD OF TRUSTEES: James E. Moltz
CHAIRMAN OF THE CORPORATION: Thomas Wheeler
DIRECTOR OF COMMUNICATIONS: James M. Kent
Oceanus is printed by Woods Hole Oceanographic Institution, and
updated online at www.oceanusmag.whoi.edu
Shipping and handling tor two issues per year is $15 in the U.S., $20 in
Canada, and $25 U.S. elsewhere. To receive the print publication, email
whoi@cdmweb.com, or call (toll-free in North America) 1-800-291-
6458, outside North America call 508-966-2039, fax 508-992-4556, or
write: WHOI Publication Services, P.O. Box 50145, New Bedford, MA
02745-0005. For single back issues, visit the WHOI online store: http://
shop.whoi.edu
For bulk issues or bulk back issues, contact Jane Hopewood, WHOI,
Woods Hole, MA 02543. jhopewood@whoi.edu, phone: 508-289-3516;
fax:508-457-2182.
Checks should be drawn on a U.S. bank in U.S. dollars and made pay-
able to: Woods Hole Oceanographic Institution.
When sending change of address, please include mailing label. Claims
from the U.S. for missing issues will be honored within three months of
publication; from overseas, six months.
Permission to photocopy for internal or personal use or the internal
or personal use of specific clients is granted by Oceanus to libraries and
other users registered with the Copyright Clearance Center (CCC), pro-
vided that the base fee of $2 per copy of the article is paid directly to:
CCC, 222 Rosewood Drive, Danvers, MA 01923. Address letters to the
editor to oceanusmag@whoi.edu
Woods Hole Oceanographic Institution is an Equal Employment
Opportunity and Affirmative Action Employer.
Oceanus and its logo are Registered Trademarks of the Woods Hole
Oceanographic Institution.
Copyright ©2004 by Woods Hole Oceanographic Institution. All Rights
Reserved. Printed on recycled paper.
Woods Hole Oceanographic Institution
The Engine that
Motion in the Mantle
New tools and techniques reveal the inner workings of our planet
By Robert Detrick, Senior Scientist and Vice
President for Marine Facilities and Operations
Woods Hole Oceanographic Institution
Poets and philosophers have celebrated
the timelessness ot the land around us
for eons, but the solid Earth is actually a
very dynamic body. Great tectonic plates
are in constant motion at Earth's surface.
Earthquakes and volcanic eruptions
are manifestations of these movements on
human time scales. But over millions of
years, the movements ot Earth's tectonic
plates rearrange the face of the Earth.
They cause continents to rift and drift,
creating entirely new ocean basins. Col-
lisions between plates squeeze ancient
oceans until they disappear and produce
majestic mountain ranges such as the Alps
and Himalayas.
Like an auto mechanic who has to
look "under the hood" to see the engine
that powers your car, geologists need
to look deep within Earth's interior to
understand the tremendous underly-
ing forces that build and shape Earth's
surface. We are now beginning to com-
prehend and describe the role that the
Earth's mantle plays in driving changes
on the planet we live on.
Hot, flowing rocks
Earth's mantle is the solid, rocky inte-
rior ot our planet that extends from the
base of the crust all the way down to
Earth's core, about 2,900 kilometers (1,800
miles) below the surface. Although they
are solid, the rocks in Earth's mantle can
deform and flow by viscous creep over
long time periods. At first glance, this
might seem odd; after all, the rocks we
find on Earth's surface are cold and brit-
tle, and they fracture or break if they are
deformed. On a large scale, this same pro-
cess causes earthquakes.
But as any blacksmith knows, when a
hard, brittle material like iron is heated to
a temperature just below its melting point,
it becomes malleable. Similarly, given
enough time, at the high temperatures and
pressures found within Earth's interior,
mantle rocks can deform and flow like
Continued on page 11
Older, colder (and denser) oceanic plates
collide with other plates and plunge
back into the mantle at subduction
zones, forming deep ocean trenches
and volcanic island arcs.
Is the mantle one big p
At mid-ocean ridges, mantle rock rises, melts, and
erupts to form new oceanic crust and volcanic
The underlying "flow" of rocks in the mantle drives geological phenomena at Earth's
surface, ranging from earthquakes and volcanoes to the creation of mountains and
oceans. As any blacksmith knows, when a hard, brittle material like iron is heated to
temperatures just below its melting point, it becomes malleable. Similarly, the high
6 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.ed
Drives the Earth
Conduits Into Earth's Inaccessible Interior
Hot plumes from deep within the planet bring up telltale chemical clues about the mantle
By Stan Hart, Senior Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Jules Verne wrote about a way to jour-
ney to the center of the Earth, but
unfortunately, we haven't found it yet. So
we really don't know what happens deep
inside our planet.
Earth's interior has yielded its secrets
only gradually over the past few decades.
Yet it contains evidence about how our
planet originally formed some 4.6 bil-
or is it double-decked?
mountain chains. Newly created seafloor crust
spreads outward from the ridges.
In some seafloor regions, unusually hot
areas of the mantle form narrow, isolated
plumes that erupt through rigid oceanic
plates to form volcanic islands
and seamounts.
temperatures and pressure within Earth's mantle deform rocks so that they can flow
like a slowly moving liquid. Hot materials rise and cold materials sink in circular
convection cells. Scientists are pursuing evidence to determine if the entire mantle
convects (right side of diagram), or if the mantle is two-tiered (left side of diagram).
lion years ago. And it holds clues to the
processes that shape the surface of the
Earth — from the benign processes that
produce things we covet (ore and mineral
deposits) to the disruptive processes we
yearn to predict and avoid (great earth-
quakes and volcanic eruptions).
We get glimpses ot Earth's interior from
mountain-building processes that occa-
sionally thrust rocks to the surface from
depths as great as 95 kilometers (60 miles).
Volcanic eruptions sometimes bring up
rocks from depths of 200 kilometers (125
miles). But Earth's rocky mantle contin-
ues down for another inaccessible 2,690
kilometers to the core-mantle boundary
at 2,890 kilometers (1,800 miles). Nor will
we ever directly view Earth's metallic core,
which extends another 3,481 kilometers
(2,163 miles) down to Earth's center.
Plumes that form islands
One way to probe the deep Earth is
remote sensing, using seismic waves gen-
erated by earthquakes. (See "Listening
Closely to 'See' into the Earth," page 16.)
By studying the varying speeds at which
these waves travel through rocks, we can
infer a great deal about the rocks' chemical
composition and their varying tempera-
tures in different regions of the mantle.
Still, the technique is indirect, and unfor-
tunately, waves sometimes travel at similar
velocity through rocks with different
chemistry, providing ambiguous clues.
A more promising trail was blazed in
1972 when Jason Morgan of Princeton
University proposed the "mantle plume"
hypothesis. Mantle rocks are hot enough
to be in slow but constant motion. But
Woods Hole Oceanographic Institutio
A FOUNTAIN OF CLUES — Lava erupting on volcanic islands such as Hawaii is brought to the surface by narrow plumes of hot, buoyant rock
originating in the mantle. The volcanic rocks contain chemical "fingerprints" that reveal their age, source, and formation — giving geochemists
clues to the inner workings of the mantle.
occasionally, unusually hot areas can lorm
upwelling plumes. On the scale of the
entire mantle, these plumes are narrow
(with a width-to-height ratio similar to
a 6-inch strand of spaghetti). But these
plumes can be vast — plume "heads" can
reach an estimated 500 to 1,000 kilometers
(310 to 620 miles) in diameter.
Plumes start at the bottom of the
Earth's mantle and carry material near to
Earth's surface — a trip that takes tens of
millions ot years. As these deep mantle
rocks approach 50 miles from the surface,
decreasing pressure allows them to par-
tially melt. The melts, or magmas, may
leak through the overlying cold rigid sur-
face layer, or plate, to form volcanoes.
Most islands in the oceans are com-
posed of volcanic rock derived from such
plumes, as best we can tell. This seems like
a promising avenue to reveal what lies at
the bottom of the mantle. All we have to
do is travel to island paradises and bring
home volcanic rocks for analysis.
But, alas, it's not that easy. Mantle rocks
are chemically altered during the melt-
ing processes, and scientists have spent
the last 20 years trying to reliably unravel
these processes. Furthermore, only the
mantle rocks that actually melted end up
in labs — what's left behind? And finally,
we're not sure all plumes come from the
bottom ot the mantle.
Where do plumes come from?
Seismology provides our best way to
detect plumes and map their sources, but
the evidence is inconclusive. There are
two schools of thought on the origins of
plumes, the "layered mantle" school and
the "whole mantle" school.
"Layerists" claim that the mantle has
two layers. Below 660 kilometers (410
miles), pressure and temperature condi-
tions cause mineralogical changes in
rocks. The deeper rocks are denser, and
do not "flow" in the same way shallower
rocks do. As a result, rocks above and
below the 660-kilometer "boundary" con-
vect separately. In this model, plumes arise
solely from the bottom ot the top layer.
Virtually nothing from the deepest parts
ot the mantle ever comes near the surface.
"Wholists," on the other hand, main-
tain that plumes come from the bottom
of the mantle (1,800 miles deep), at the
boundary of Earths core. Heat from the
core disturbs the overlying mantle, lead-
ing to plume formation. If accurate, this
model gives hope that we will someday
understand the chemistry of the deep-
est mantle (and if we're very lucky, of the
outer core, provided some core material
occasionally mixes into rising plumes).
"Fingerprinting" volcanic rocks
To explore the origin ot plumes, geo-
chemists analyze rocks from ocean volca-
noes tormed by plumes. They use a form
of isotopic fingerprinting.
The rocks contain traces of long-lived
radioactive isotopes left over from the
initial building of our solar system. These
elements have known half-lives — the
time required tor one-half of a quantity
8 Oceanus Magazine -Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
ot "parent" isotopes to decay into more
stable "daughter" isotopes.
We can calculate the rock's age by mea-
suring the relative amounts of remaining
radioactive parent and its daughter prod-
ucts. And we can determine the "parent"
rocks from which the specimen originally
descended (or more accurately in this
case, "ascended"). In this way, we can trace
a rock specimens lineage.
If the mantle convected in a single unit,
it would produce volcanic rocks with a
single homogeneous chemistry. A two-
tiered mantle would produce, at most,
two types of rocks. But, once again, the
solution is not easy. Our analyses point to
more chemical diversity in the mantle.
Pedigree and mongrel rocks
\\'e have identified four "lineages" of
plumes, and mixtures thereof. Some oce-
anic islands appear to be virtually pure
species. Most are just mongrels.
The purebred ocean island rocks have
been given particular labels. Rocks from
Pitcairn Island, of HMS Bounty fame,
are labeled "EMI" (Enriched Mantle 1).
Rocks from Samoa, of "Coming of Age"
fame, are "EM2" (Enriched Mantle 2).
Rocks from Mangaia, in the Cook/ Aus-
tral chain, are called "HIMU" (High
"Mu," from the Greek symbol "u," which
represents a uranium/lead ratio). Rocks
from mid-ocean spreading ridges, which
are fed from the shallow upper mantle,
are "DMM" (Depleted Mid-ocean ridge
basalt Mantle).
Note that even these purebred island
rocks show some crossbreeding. For
example, specimens from Vailulu'u and
Malumalu, two newly discovered undersea
volcanoes in Samoa, show both the purest
EM2 yet sampled, as well as a variety of
less pure offspring. An unexplained curi-
osity is that all of the purebred islands are
in the Southern Hemisphere.
Some of the well-known "mongrel"
islands are Hawaii, Tahiti, and the Galapa-
gos, Azores, and Canary Islands. Hawaii
is notable for being both the largest of all
mantle plumes (as measured by total vol-
canic output over the millennia) and by far
the most intensively studied by geochemists.
Recycled crust
The history of these purebred mantle
rock species has been the subject of con-
troversy for decades. Were they born when
the Earth was born, somehow escaping
mixing in the constantly convecting man-
tle? Are they primordial, homogenous,
and once-pristine mantle rock species that
have been contaminated by injections of
rock from Earth's surface (as argued by
Albrecht Hofmann of Max-Planck Institut
tur Chemie and William M. White of Cor-
nell University in 1982)?
The rigid plates of rock that cover
Earth's surface ultimately plunge back
into the mantle— in the great subduction
zones that border the Pacific Ocean and
elsewhere. These plates carry ocean crust
and its veneer of accumulated sediment
into the mantle— sometimes deeply. This
crust and sediment reintroduce chemi-
cals into the mantle, which can confound
our analyses.
For example, oceanic crust, during its
millions of years of exposure to seawater,
sequesters naturally radioactive uranium
trom the oceans. When this crust is sub-
ducted into the deep mantle, the uranium
in it would decay, over a billion years, into
significant accumulations of uranium's
daughter isotope, lead-206. This ancient
oceanic crust material, heated up again in
the deep mantle, might once again become
buoyant — and thus a new plume is born!
Rocks from this plume would have a high
(and highly misleading) ratio of lead-
206 — the same fingerprint that typically
distinguishes HIMU rocks.
Over millions of years, ocean crust also
accumulates an overlying veneer of sedi-
ments containing rubidium, uranium, and
HOT OFF THE VOLCANO— Shielding his face with an asbestos glove, WHOI geochemist Stan
Hart uses a rock hammer to collect freshly erupted lava from Kilauea Volcano in Hawaii. The
lava, orginating from the mantle, provides chemical clues to the inner workings of the Earth.
Woods Hole Oceanographic Institute
lead from non-mantle sources. After sub-
duction, hibernation in the mantle, and
rejuvenation as a new plume, rocks con-
taining recycled sediment material could
contain "enriched" isotopic signatures
that can confound our ability to trace the
rocks' ultimate source in the mantle.
The "recycling" proposed by Hotmann
and White must occur, and it explains
many aspects of the chemical "zoo" we
observe in mantle rocks. Nevertheless,
there is no unanimity whatsoever in the
geochemical community as to whether
this model, or any of its amended versions,
is the "real" scenario.
Deep Earth odyssey
In 2004, here is the current state of
our knowledge about the mantle. We
know that Earth's mantle is convecting,
but we don't know whether it is whole-
mantle or layered convection. We are vir-
tually certain that plumes exist, but don't
know with certainty where they originate.
We know Earth's mantle is not chemically
homogeneous, but rather a chemical zoo
with at least four purebred species and a
bewildering pack of mongrels. We don't
know where any of these species come
from, nor can we trace the pedigrees of
the purebreds.
Given that it's taken four decades of
concerted effort to get this far, why are we
optimistic and excited? Because in recent
years, major links have formed among the
fields of geochemistry, seismology, rock
physics, and convection modeling. All
these fields have also profited from the
development ot incredible new technolo-
gies that are allowing us to circumvent
roadblocks that have stymied our scien-
tific research for years.
Scientists must converse across dis-
ciplines more than ever. We need, each
of us, to become broader— Renaissance
scholars, as it were. We need to maintain
access to cutting-edge technologies and
continue to hone new ones. Then, pos-
sibly, an article on Earth's mantle written a
decade from now will contain the answers
to the questions that linger in this one.
Volcanic islands, like dogs, can be pedigrees or mutts
By analyzing radioactive isotopes in rocks from various volcanic islands, geochemists have
determined that all islands are not alike. Most islands are made of mixtures of many rock
types. But some islands are primarily composed of one of four chemically distinct rock types:
EM 1 (Enriched Mantle 1), EM2 (Enriched Mantle 2), HIMU (High "Mu, "from the Greek symbol
p); and DMM (Depleted Mid-ocean ridge basalt Mantle). These chemical distinctions
provide clues to understand the underlying mantle plumes that create the islands.
Stan Hart was born and raised in Lynn, Mass., a liability mitigated by living on
the border of Lynn Woods, a many-square-mile tract afforest, granite ledges, and
a pristine lake (the town water supply). Though prohibited, fishing in this water
became an early passion; incessant roaming in the woods was permitted both by
~ parents and town law. Undergraduate years were spent commuting to MIT, rock
§ climbing the cliffs of New Hampshire, four-event skiing with the MIT Ski Team,
; and working summers in the Appalachian Mountain Club hut system. These out-
s door leanings led to a switch of major from chemistry to geology (the first course
• in organic chemistry helped). A wonderfully stimulating year at Cal Tech, a
^ master's degree in geochemistry, and daughter No. 1 cemented Stan's career path.
He was wooed back to the gentility of MIT for a Ph.D., followed by a postdoc (and 14 more years!) at
the Department of Terrestrial Magnetism, Carnegie Institution of Washington. Hart was called back to
MIT for the retirement of his Ph.D. advisor, and he professored therefor 14 years before opting for the
more rural life of Falmoiith for daughter No. 2 and son No. 1. He passed the 15-year mark at WHOI in
2004, and his climbing and skiing fervors have morphcd into an addiction for running. His predisposi-
tions for travel and research are being satisfied by intensive study of the many volcanoes of Samoa (the
youngest of which is still under water, and in active eruption).
1 0 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Continued from page 6
a slowly moving fluid, even though they
would appear solid to us. (See "Peering
into the Crystal Fabric of Rocks," page 57.)
These movements are gradual in human
terms (a few centimeters a year), but over
a hundred million years, mantle rocks can
move thousands of kilometers.
The energy that drives this movement
is heat within the Earth, which comes
from two main sources. One is the resid-
ual heat left over from the formation of
our planet 4.6 billion years ago. The radio-
active decay ot naturally occurring chemi-
cal elements in the Earth — most notably
uranium, thorium, and potassium— also
releases energy in the form of heat.
These two sources of heat warm
Earth's mantle and cause it to rise and
sink, much like soup in a pot on a stove.
Material heated from below gets hotter
and rises. It reaches the surface, releases
its heat, becomes colder and denser, and
sinks again.
Ocean trenches and ridges
The motion of tectonic plates, and all
ot their associated volcanic and earth-
quake activity, are believed to be the sur-
face manifestation of similar thermal con-
vection occurring in Earth's mantle. The
rigid outer layer of our planet, called the
lithosphere, is the cold, top boundary of
convection cells in the mantle. Two fea-
tures of the ocean floor are the results of
this active process.
At ocean trenches, old, cold, and
dense lithosphere sinks back into the hot-
ter mantle below, dragging surface plates
along with it. The continents are carried
along by these moving surface plates,
producing the continental drift that, for
instance, rifted and separated Africa and
South America.
Beneath ocean spreading centers, two
plates are moving apart. Hot, buoyant,
solid mantle flows upward. It begins to
melt as it reaches shallower areas where
pressures are lower. That leads to volcanic
activity that forms new ocean crust along
the great mid-ocean ridge system.
A 'CAT scan' of Earth's interior
Tonga Island Arc
Tonga Trench
Seismic tomography is the geophysical analog of a medical CAT scan.
Seismologists measure the speed of seismic waves generated by large earthquakes
(white circles) that propagate through the mantle. They can then map temperature
variations in the mantle, because seismic waves travel more slowly though hotter
regions (yellow and red) and faster through colder regions (green).
The green region here marks the slab of old oceanic plate plunging back into the
mantle at the Tonga Trench in the South Pacific Ocean.
In addition to this plate-scale flow,
plumes ot hot material can rise at various
places from within the convecting mantle
to form ocean island volcanoes or "hot-
spots" such as Hawaii. Indeed, convection
currents in the mantle are responsible for
most of the volcanic and tectonic pro-
cesses that we see at Earth's surface.
A double-decked mantle?
Mantle convection is the "engine" that
drives our dynamic Earth. But since scien-
tists cannot directly observe the workings
of this engine, they have debated the exact
mechanisms of this convection for more
than half a century.
For example, does convection occur
over the entire mantle at once, creating
a huge pot of essentially the same soup?
Or is it layered — with separate convec-
tion cells occurring in an "upper man-
tle" (extending from near the surface to
a depth of about 660 kilometers, or 410
miles) and in an underlying "lower mantle"
(with denser rocks under higher pressure)?
This second scenario would result in
two chemically distinct reservoirs in the
mantle that almost never mix, like oil
and water. The first model would seem to
require a more compositionally uniform,
well-mixed mantle. Two lines of evidence
give contrasting answers to the question.
Woods Hole Oceanographic Institution
Lavas erupted at mid-ocean ridges and
at oceanic islands have different compo-
sitions ot trace elements and noble gases.
(See "Conduits Into Earth's Inaccessible
Interior," page 7.) This strongly suggests
that distinct geochemical reservoirs exist
in the mantle, and seems to favor a lay-
ered convection model with different
mantle compositions in the upper and
lower mantle.
Geophysical 'CAT scans'
In the mid-1990s, however, seismolo-
gists studied how seismic waves gener-
ated by large earthquakes propagated in
the mantle and made a startling discovery.
Using a technique called seismic tomogra-
phy— the geophysical analog ot a medical
CAT scan — they were able to map varia-
tions in the speed of the seismic waves
traveling through the mantle. We can inter
temperature variations in the mantle from
this, because seismic waves travel more
slowly through hotter regions and taster
through colder regions.
The seismic tomographic images
showed that in some cases, cold tectonic-
plates sink to depths of only about 600
to 700 kilometers — near the base ot the
upper mantle. In other cases, they sink
more than halfway into the lower mantle.
These results clearly show that the den-
sity contrast (of lighter and denser rocks)
at the base of the upper mantle is not a
barrier to mixing between the upper and
lower mantle.
How can we then explain the evidence
tor distinct reservoirs in the mantle? In
other words, why isn't our mantle soup
well-mixed?
This vexing question is unresolved.
Some scientists have proposed that
there are "unstirred" blobs ot material
floating in the mantle that have per-
sisted throughout most of geologic his-
tory. Others have suggested that a dense,
poorly mixed region exists in the lower
1,000 kilometers of the mantle. This
region is chemically distinct and rich in
heat-producing elements, and it generates
flow into "hotspot" areas.
An ocean-bottom seismometer on R/V Ewing
awaits deployment to the seafloor during a
1 997 experiment in the Pacific Ocean to
elucidate how magma chambers feed
magma to overlying mid-ocean ridges.
Other unanswered questions
There are other puzzling questions
about how the mantle convection engine
works. For example, seismic tomogra-
phy studies of the shallow mantle beneath
oceanic spreading centers reveal a broad
region ot slow seismic velocities. This
suggests the presence ot partially melted
material across an area of several hundred
kilometers. Yet volcanic activity at mid-
ocean ridges occurs over remarkably nar-
row zones just a few kilometers wide. We
don't know how melting produced over
such a broad region in the mantle can be
focused into such a narrow zone of erup-
tive activity at the seafloor. (See "Unravel-
ing the Tapestry of Ocean Crust," page 40.)
Another major debate centers around
the origin of hotspot volcanic centers such
as Hawaii and Iceland. Some researchers
argue that hotspots are caused by rising
plumes of hot mantle material originating
in the "unstirred" reservoir just above the
Earth's core. Others think that processes
occurring in the uppermost mantle cause
some hotspots. Answering such questions
will require the integration of data from
seismology, geochemistry, and mineral
physics with fluid-dynamical modeling of
mantle convection.
New tools to look into the Earth
This is a particularly exciting time
for studies of the oceanic upper mantle.
Advances in ocean-bottom seismic instru-
mentation now make it possible to record
high-quality seismic data from earth-
quakes at sites over the two-thirds of the
planet covered by water. (See "Listening
Closely to 'See into the Earth," page 16.)
Advances in geochemical and petrolog-
ical studies over the past decade, including
new analytical and sampling techniques,
are defining important new limits on the
mantle conditions that produce melting.
And in recent years we've seen tre-
mendous increases in the affordability
and speed of computer systems, improved
numerical techniques, and a growing
understanding of the complex processes of
melting and viscous flow.
These advances are allowing the devel-
opment of the first global-scale, self-con-
sistent, three-dimensional models of man-
tle convection that can explain our surface
observations of tectonic plate motions,
gravity, and topography. These studies will
lead to much more complete understand-
ing of mantle convection.
Like the auto mechanic at your local
garage, we are finally "lifting the hood"
to view Earth's engine and the forces that
drive changes on the face ot the Earth.
Bob Detrick, a marine geophysicist, became Vice President for Marine Facili-
ties and Operations at WHO1 in 2004. As a young boy, Detrick's family vaca-
tions to Colorado from his hometown of Pittsburgh sparked a lifelong interest
= in geology and a love of mountains and world travel. In 1970 he came to
| Woods Hole as a Summer Student Fellow and was captivated by the adven-
5 lure and challenge of studying the geology of the seafloor. As first a Joint Pro-
gnun Student and later as a scientific staff member, Detrick has spent nearly
t-* •? 20 years at WHOI and participated in more than 30 different oceanographic
531 ^ research cruises. His research interests encompass marine seismology, ocean
crustal structure and tectonics, especially along mid-ocean ridges, and mantle dynamics. In his new
position, Detrick will be responsible for developing and implementing a strategy Jar integrating WHOI
ships, vehicle systems (manned submersible, remotely operated vehicles, and autonomous underwater
vehicles) and ocean observatories to provide researchers at WHOI with unmatched access to the sea.
1 2 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
If Rocks Could Talk. . .
The ion microprobe extracts hidden clues about our planet's history and evolution
By Nobu si, i in i / ii . Senior Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Every rock on Earth contains a clock, a
thermometer, and a barometer. Inside
all rocks are elements, or isotopes of ele-
ments, called "natural tracers." By exam-
ining the presence, proportion, and dis-
tribution of natural tracers within rocks,
we can reveal the conditions under which
the rocks formed. They can tell us when
the rocks formed (clocks), how fast they
cooled and how they crystallized (ther-
mometers), and the temperatures and
pressures they experienced at their cre-
ation (barometers).
Just as radioactive tracers are used
to understand the dynamics ot chemi-
cal reactions, natural tracers in rocks
can be used to help decipher the whens,
wheres, and hows of the complex geologi-
cal processes that create and maintain our
planet. With the right tools, we can extract
long-dormant, hidden information about
The tales that rocks can tell
The ion microprobe at WHOI precisely measures very small amounts of isotopes in rocks and other samples, revealing hidden clues about
how our planet works. The national facility, supported by the National Science Foundation, is housed in an 800-square-foot laboratory.
5 Lenses and deflectors located
throughout the ion microprobe
continue to direct the ion flight path.
magnet
6 A powerful magnet
sorts ions according
to subtle differences
in their mass,
separating lead into
206pb, 20?pb, and
208pb isotopes, for
example.
7 Ions from the
original sample are
separated by their
charge and mass,
similar to the way a
prism breaks up light
into constituent
parts. Scientists can
choose to measure
one specific ion mass,
using one mode. Or
they can measure up
to five different ion
types simultaneously.
8 Ion counters
measure the different
types of ions extracted
from the sample.
electrostatic analyzer
4 Two powerful,
oppositely charged plates
sort ions based on their
charge.
3 Lenses and deflectors produce carefully
regulated electrical fields within the ion
microprobe, steering the electrically
charged ions along a controlled flight path
1 Primary ion
beam accelerates
ions (charged
atoms) to high
speed and focuses
them into a narrow
beam.
sample
insertion tube
exit
slit
sample
2 Like a line of billiard
balls striking a racked set
of balls, the ion beam
impacts rock samples,
causing ions from the
sample to "sputter" off.
screen -
9 The ion microprobe can also
create an image showing where
mass-selected ions were
distributed in the original
sample.
Woods Hole Oceanographic Institution
91
C
Earth's inner workings from rocks.
The Ion Microprobe Facility at Woods
Hole Oceanographic Institution is just
such a tool. With it, we can peer far
back in time and deep into the Earth. To
understand processes that form new oce-
anic crust, for example, we have used the
WHOI ion microprobe to study ancient
rocks from the crust and underlying
mantle, which have been thrust up and
exposed on land (in formations known as
ophiolite massifs). And we have compared
those with rocks from active mid-ocean
ridges on the seafloor.
We can also peer into rocks from the
surface to incredible depths — almost
450 kilometers down— by probing min-
eral inclusions in some diamonds formed
under pressures at great depth. (Inclusions
are minute foreign bodies enclosed within
the mass ot another mineral.)
Extracting information from rocks
The ion microprobe offers great advan-
tages over previous methods to glean
natural tracer information out of rocks.
Before, scientists had to break apart sam-
ple rocks and extract minerals contain-
ing specific tracers. The purified minerals
were then chemically processed, and the
amounts or types of tracers were deter-
mined using various instruments.
It is a painstaking and time-consum-
ing process, and something important is
destroyed in the process ot mineral extrac-
tion and purification: the textural rela-
tionships in which mineral crystals occur
in the rock. Rock texture is significant
because it reflects the dynamic conditions
under which minerals crystallized, and
it presents a geologic framework within
which to interpret the tracer information.
For example, if a rock forms while
conditions around it are changing, the
minerals in the rock will show different
textures or grain sizes depending on the
conditions. This information is lost in
traditional processing, but retained with
ion microprobe analysis because the rock
is not broken up. With the ion micro-
probe, we can look at the composition
of very small samples and identify com-
ponents in situ, even over distances only
micrometers apart.
From electron beams to ion beams
The first tools that allowed analysis of a
What the ion microprobe can tell us
How do rocks melt and migrate beneath the seafloor?
The WHOI ion microprobe is help-
ing researchers shed light on a funda-
mental, but still largely unknown pro-
cess that shapes our planet — how rock
deep beneath mid-ocean ridges melts
to form magmas, and how the melted
rock then migrates toward Earths sur-
face at the ridges. The instrument has
provided the first unequivocal evidence
that a type of melting called "near-
fractional melting" is occurring deep
beneath ridges.
In near- fractional melting, solid and
melted rock do not stay together. The
melt finds pathways and immediately
escapes the not-yet-melted rock.
We deduced the occurrence of this
type of melting by examining composi-
tions of the solid rocks found at mid-
ocean ridges. The work is based on our
knowledge that when rocks melt, ele-
ments are distributed between solid
and liquid phases in a particular way.
Evidence for near-fractional melting
was first found by determining abun-
Tiny glass inclusions, about 50 micro-
meters in diameter, are enclosed in olivine
crystals from mid-ocean ridge basalt. The
amounts and distribution of isotopes of
lead within the inclusions yield informa-
tion about the mantle source of the basalt.
dance patterns of particular trace ele-
ments (the rare earths, or lanthanides)
found in a mineral (diopside), which is
itself present in very small amounts in
rocks (peridotites) dredged from ocean
ridge fracture zones. Researchers could
not have analyzed such small quantities
with conventional geochemical tech-
niques, but the ion microprobe enabled
researchers to determine lanthanide
abundance patterns in samples and
demonstrate this type of melting.
The same melting has now also
been identified in mid-ocean ridge
basalt, in tiny glass particles, or glass
(melt) inclusions, incorporated into
crystals of olivine, a mineral that
forms in basalts as they cool and
solidify. By determining the isoto-
pic composition of an element, lead,
in melt inclusions, we are changing
our views about the mantle source
for mid-ocean ridge basalt, and about
how magma migrates from the mantle
to the ocean floor.
We know that melt inclusions are
entrapped in olivine at shallower, less-
pressurized levels beneath the seafloor.
But the variability of lead isotopes that
we are observing in melt inclusions in
seafloor rocks samples indicates that
melts from different mantle sources
are rising separately and are not mix-
ing before they reach the level at which
olivine forms.
1 4 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
sample's composition without chemically
processing it were earlier electron-beam
microprobes. These machines generated
electron beams and tocused and directed
them at a rock sample. The electrons hit-
ting a sample caused the production of
X-rays, and measuring the X-ray spec-
tra allowed us to determine the chemical
composition of the samples.
In contrast, ion microprobes use
focused beams of ions (charged atoms) to
bombard a sample. Ions are much heavier
than electrons, and the ion beam causes
the sample to eject atoms and ions, rather
than just emit X-rays. When the ion beam
strikes the sample, atoms and ions are
"sputtered" (sprayed out) from the sample.
The ion microprobe has two basic
parts: the ion-beam source, which
focuses and directs microbeams of ions
onto the sample; and the mass spectrom-
eter, which measures the signal intensi-
ties (abundances) of ions ejected from
the sample. Sputtered ions are accelerated
into the mass spectrometer, where the
ions are detected and distinguished based
on their different masses and charges.
From the intensities of ions of different
elements, we can determine the chemi-
cal composition of the sample. From the
intensities of isotopes of specific ele-
ments, we can identify the isotopic com-
position of the elements.
The ion microprobe adds new dimen-
sions to our analyses for two reasons. It
can detect tracer elements with much
greater sensitivity. It also gives us the
ability to determine both the chemi-
cal composition (the elements present)
and the isotopic composition (the pro-
portions of different isotopes of the ele-
ments) of minute amounts of tracers, in
small samples, without breaking the tex-
tural context in the rock.
After a 17-year global migration from his na-
tive Tokyo to Washington, D.C., then Paris,
France, and Cambridge, Mass., Nolni Shimizii
joined the WHOI staff in 1988. He is the "guru
of ion probology" in Earth sciences, and a
practicing therapist in various relationships
(mostly geochemical).
Extracting records of past hurricanes and ocean temperatures from corals
We can study more than rocks with the ion microprobe.
WHOI researchers Anne Cohen and Graham Layne have
used the facility to examine the hard calcium carbonate
skeleton in corals, which can reveal a record of hurricanes
and past sea surface temperatures. They sample areas small
enough to see layers of skeleton produced by the coral in a
day or less.
For the hurricane studies, they use the ion microprobe to
determine the ratio of two stable isotopes of oxygen in the
carbonate. A higher percentage of the lighter oxygen isotope
is incorporated into coral skeletons during periods of very
heavy rainfall, which occur during hurricanes. Ancient or
fossil corals thereby can reveal a record of hurricane events
and patterns stretching into the past.
To estimate past seawater temperatures, researchers often
use strontium/calcium ratios in coral skeletons (the higher
the strontium, the lower the temperatures). But Cohen used
the ion microprobe to reveal a subtle wrinkle in the relation-
ship between the element ratios and water temperatures.
Cohen studied corals living in Woods Hole harbor
(Astrangia poculata), some of which have algal cells liv-
ing symbiotically in their tissue, and some of which lack the
algae. The photosynthesizing algae directly affects the coral's
ability to deposit hard skeleton material and build large
coral reefs.
Cohen showed for the first time that strontium/calcium
ratios were systematically different in corals with and with-
out algae. Thus,
the strontium/
calcium ratio
can be used to
determine past
water tempera-
tures only in
skeletal mate-
rial produced
by algae-free
coral, or in the
absence of
sunlight.
Dotted line in left photo indicates
where a thin, 5-millimeter cross
section of coral was removed and
placed in the WHOI ion microprobe.
Black line of spots in the photo
above shows the path where the
ion beam harvested ion samples
from the coral for analysis.
Woods Hole Oceanographic Institution
Listening Closely to "See7 into the Earth
A new national facility of cutting-edge seaf loor seismographs probes Earth's interior
By John Collins, Associate Scientist
Geology & Geophysics Department
Woods Hole Oceanographic Institution
Chemists can monitor reactions in test
tubes in their labs. Ecologists can go
into the field to make observations. But
Earth scientists interested in the struc-
ture of Earth's deep interior don't have
the luxury ot seeing their subject close at
hand. Without a way to travel through the
Earth, we have had to rely on ways to "see"
Earth's structure from a great distance.
To accomplish that, our method
employs sound, rather than sight. When-
ever an earthquake occurs, scientists can
tune in and "listen" to it. We use seis-
mometers and seismographs that measure
and record earthquake-generated seismic
waves that travel along Earth's surface and
through its interior. By analyzing these
waves, we can infer a great deal about the
characteristics of the materials the waves
are traveling through. (See "Earthquakes
and Seismic Waves," page 18.)
Within the last decade, two factors
have helped make seismology the pre-
eminent tool for determining Earth's hid-
den interior structure. First, more per-
manent and temporary seismographs to
record seismic waves have been distrib-
uted around the globe. Second, interna-
tional policies have made archives of seis-
mic data freely and widely available on the
Internet to any investigator.
Today, excitement and anticipation
are growing because of new generations
Revealing the plume beneath Hawaii
iUiil lini^njiui^ far li»d fi»J iltua,
pint '.•->
THe OBSs will recorff
rthquake-generated
. Jsmic waves traveling
trffough the plumsT"
revealing new insights on
howjnantle plumefworl
1 6 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
of seismographs designed for use in the
oceans. These new instruments are being
built at Woods Hole and two other ocean-
ographic institutions, and they will com-
prise a new national pool of instruments
for use by the scientific community. This
new national facility will let us monitor
more of the planet with the precision we've
long wished for, and thus enhance our
ability to answer fundamental questions
about our planet.
Mapping Hawaii's plume
Until now, almost all seismic observa-
tories have been located on land, which
accounts for only 30 percent of Earths sur-
face. The lack of evenly distributed seis-
mograph coverage in the oceans has lim-
ited our understanding of Earth's structure
at both regional and global scales.
On a regional scale, consider the task of
understanding the structure of the upper
mantle beneath the small geographic
region ot Hawaii. Hawaii is thought to be
the surface expression of a buoyant plume
ot hot rock in the upper mantle, possibly
originating at the core-mantle boundary
at a depth of about 2,900 kilometers (1,800
miles). We don't know the width, tempera-
ture, or depth of the plume.
To find out, we will need to record seis-
mic waves that travel from a known source,
through the plume, to receiving seismo-
graphs. But the waves must travel through
as little as possible of the rest of the Earth,
so that we can attribute any anomalies or
alterations in the waves entirely to their
passing through the plume.
Hawaii's landmass is so small, however,
that land seismic stations cannot be posi-
tioned in places to intercept waves that
travel through the plume at depth. So we
haven't been able to measure how deep the
plume is. Nor are the islands close enough
together to let us accurately measure the
plumes diameter. The only way to deter-
mine the plume's structure is by placing
seismographs in appropriate locations to
intercept waves that have traveled only
through the plume — and those locations
are on the seafloor.
SHORE TO SHIP — A vanload of new WHOI "D2" ocean-bottom seismographs is readied for
field testing. The D2s are small and light, for easy deployment and recovery. With a six-month
battery capacity, they are designed for relatively short-term experiments. The D2s are available
to scientists through the newly created National Ocean-Bottom Seismograph Instrumentation
Pool. WHOI Senior Engineer Ken Peal, who helped design the D2, is in the background.
Seismic waves through the Earth
To probe Earth's interior on a more
global scale, the most useful earthquakes
to study are large and deep. But such
earthquakes do not occur uniformly over
the planet. They happen only in specific
geographic areas called subduction zones,
where tectonic plates plunge back into
the mantle.
Seismic waves from these earthquakes
radiate through Earth's deep interior to
the other side of the globe. But because
existing land-based seismic stations aren't
distributed uniformly on Earth's surface,
we don't receive any information from
many areas. That leaves us with large
gaps in our knowledge about some parts
The D2's electronic data logger (above),
which records seismic wave measurements,
resides in a glass ball in one half of the pea-
nut-shaped 02. Batteries are in the other half.
of Earth's deep interior.
Very large earthquakes can actually
set the whole planet vibrating in a series
of oscillations, commonly likened to the
ringing of a bell, which affect the entire
surface of the Earth. But we can't mea-
sure them over large areas where there
are no seismographs, once again leaving
us with an incomplete picture of Earth's
interior. In fact, the optimum location to
observe and measure seismic waves that
pass through Earth's core lies in the central
Atlantic Ocean — where there are tew seis-
mographic stations!
Seismic data gaps in the ocean
The oceanographic community has
long recognized the need for seafloor
seismic stations. Since the 1970s, several
oceanographic institutions — including
Woods Hole, Scripps Institution of Ocean-
ography, and the University of Texas Insti-
tute of Geophysics — have designed, built,
and operated ocean floor seismic-moni-
toring instruments, known as ocean-bot-
tom seismographs (OBSs) and ocean-bot-
tom hydrophones (OBHs). The research
was funded primarily by the Office of
Naval Research.
For several reasons, these first-gen-
eration OBS and OBH instruments didn't
give us a clear picture of Earth's structure.
Woods Hole Oceanographic Institution
Earthquakes and seismic waves
When an earthquake occurs, rocks
at a fault line slip or rupture, and a por-
tion of Earth's crust physically moves.
That releases energy, and two types of
seismic waves radiate outward from the
earthquake through Earths interior and
along its surface. Compression waves
alternately compress and release rocks
in the direction the waves are moving
(similar to the air compression we hear
as sound). Shear waves move rocks per-
pendicularly to the direction the waves
are moving.
Seismographs (seismometers and
associated recording systems) detect
and measure these waves. Compres-
sion and shear waves travel through the
Earth at different speeds. By measur-
ing the speed of the waves as they travel
through different positions within the
Earth, we can draw inferences about
the temperature, composition, and
degree of deformation of the material
that the waves travel through. In turn,
better estimates of these characteristics
improve our understanding of hidden
processes that occur in the Earth's core,
mantle, and crust.
With seismic measurements we can
also locate an earthquake's source. The
characteristics of an earthquake itself,
such as its location, magnitude, fault
orientation, and fault slip, are impor-
tant for understanding tectonic pro-
cesses at global and regional scales, and
seismology is essential for understand-
ing the physics of earthquake initiation
and rupture.
7 -Shear waves (S-waves) move rocks
perpendicularly to the direction the
waves are moving (similar to waves at the
beach).
2-Compression waves (P-waves)
alternately compress and release rocks
in the direction the waves are moving
(similar to the air compression that we
hear as sound).
They required a lot ot power, had limited
storage capacity, and used inexact clocks
that couldn't determine (with enough
accuracy) the times when waves from an
earthquake arrived.
But today's seafloor seismic instru-
ments are being built with low-power
digital electronics, very large storage
capacities, and high-precision clocks that
drift only a millisecond a day or less. The
older instruments measured Earth move-
ments with periods of a second or so, but
new seafloor seismometers are sensitive
to motions with periods of hundreds to
tenths of seconds, which can tell us much
more about the Earth's deep structure.
And new high-dynamic-range analog-
to-digital converters improve upon older
converters, which tended to distort both
small signals (such as from a distant mod-
A D2 is field-tested on the seafloor near the
TAG hydrothermal vent site in the North
Atlantic. A weight anchors it to the bottom.
The seismometer is housed in the silver
canister at right.
erate earthquake) and large signals (such
as from a large local earthquake).
A new national seismic facility
Another obstacle in the early days
was a limited availability ot OBSs. Most
early instruments were available only
to investigators at institutions that built
the instruments, and these investigators
didn't necessarily pursue scientific ques-
tions of interest to a broader Earth science
community. In addition, early seismic-
data were stored in scattered sites and in
a variety of formats, so many of the data
remained inaccessible because they were
hard to find and use.
Now, with National Science Founda-
tion funding, three institutions— Woods
Hole Oceanographic Institution (WHOI),
the Institute of Geophysics and Planetary
Physics at Scripps Institution of Ocean-
ography (SIO), and the Lamont-Doherty
Earth Observatory of Columbia Univer-
sity—will form the National OBS Instru-
mentation Pool.
Each institution will design, build, and
operate instruments, and provide engi-
neering and technical support to deploy
the instruments on the seafloor. Investi-
gators can request to use OBSs from the
national facility as a part of the standard
NSF proposal process, and other private
and public organizations can use them as
18 Oceanus Magazine • Vol. 42, No. 2 • 2004
availability permits. All data collected with
these new instruments will be centrally
archived at the same national data reposi-
tory used by the U.S. land seismological
community. After two years' proprietary
access for investigators, the data will be
available to all.
Two new kinds of OBSs
\VHOI and SIO each will build and
operate two distinct types of OBSs. One
will be used for short deployments to
record predominantly short-period (i.e.
high-frequency) Earth motions. The
other can be deployed for a year or more
to record Earth motions over a much
broader range of periods.
Short-period OBSs are usually used in
closely spaced arrays for durations of days.
They typically record seismic waves from
artificial sources, such as those used by the
oil-exploration industry, which use com-
pressed air to produce a brief wave pulse.
These instruments are used to determine
the structure of the outer tens of kilome-
ters of the seafloor.
The WHOI short-period OBSs, called
"D2s," have a six-month battery capac-
ity and are equipped with a seismometer
and a hydrophone. They can detect Earth
motions in both vertical and horizon-
tal directions, and can even record small,
short-period earthquakes, such as those
associated with hydrothermal vent activity.
In typical experiments using short-
period OBSs, large numbers of instru-
ments must be deployed repeatedly during
30-day cruises. So the D2s were designed
to be small and light, for easy recov-
ery. They don't need to be opened at sea;
instead recorded data can be offloaded
easily onto a shipboard data- storage sys-
tem over an Internet connection.
A long-awaited experiment
The first short-deployment experi-
ments using D2s were completed in the
summer of 2002. In April of 2004, we
undertook our first really challenging
experiment, deploying and recovering
some 150 D2s during a 35-day cruise.
All three institutions are building
long-deployment OBSs. These carry
long-period pressure sensors and seis-
mometers capable of recording Earth
motions with periods ranging from about
40 seconds to 0.1 seconds. The WHOI
long-deployment OBS will have the same
data logger used in the D2 OBS, but will
be equipped with a WHOI-designed
acoustic modem that will let investiga-
tors on research vessels remotely retrieve
data from the OBS sitting on the seafloor.
This capability will allow investigators to
check that the OBS is performing ade-
quately before the ship leaves the area
for a year.
The first long-term deployment is
scheduled to take place in December
2004 ott Hawaii in the Plume Lithosphere
Undersea Melt Experiment (PLUME). The
goal of this two-year long experiment — a
collaboration between WHOI, SIO, the
University ot Hawaii, and the Carnegie
Institution ot Washington — is to image the
Hawaiian plume. This one experiment has
the potential to revolutionize our under-
standing of how Earth's mantle convects.
For Earth scientists, the future looks
bright, and will sound even better.
Born and raised in Ireland,
John Collins came to the
,~, United States in 19S1 and en-
1 tered the MIT/WHOI Joint
1 Program. After two years of
= postdoctoral work in Austra-
t lin, lohn returned to WHOI
12 where he has been ever since.
Collins' scientific interests focus on investigating
the structure of the oceanic crust and lithosphere,
and he has played a major role in developing a
wide variety of ocean-bottom seismic instrumen-
tation. He has participated in about 20 research
cruises tliat have taken him from the Pacific to
the Atlantic. Since 2003, he has served as the
manager of WHOI Ocean Bottom Seismometer
Instrument Pool.
Broadband
Seismic Sensor
Differential
Pressure Gauge
Batteries & Electronics
Corrodible Line
A NEW GENERATION OF SEISMOGRAPHS— WHOI is designing and building ocean-bottom seis-
mographs (OBSs) for long-term deployments.The four orange fiberglass "hardhats," mounted
on a plastic grillwork, contain batteries and electronics. A differential pressure gauge measures
earthquake-generated waves in the water. The seismometer is housed in a metal sphere at-
tached to an swivel at right. When the OBS is deployed, a line corrodes, positioning the seismom-
eter on the seafloor.
Woods Hole Oceanographic Institution
en
O
Realizing the Dreams of da Vinci and Verne
A diverse fleet of innovative deep-submergence vehicles heralds a new era of ocean exploration
01
By Daniel Fornari, Director
Deep Ocean Exploration Institute
and Senior Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Leonardo da Vinci made the first
drawings of a submarine more than
500 years ago, and Jules Verne published
20,000 Leagues Under tlie Sea in 1875.
But only in the past few decades has the
dizzying pace of technological advances
allowed us to realize their dreams of
exploring the ocean depths and taking
humans to the seafloor.
For the past 40 years, submersible
human-occupied vehicles (HOVs) such
as Alvin have given scientists direct access
to the seafloor and the ability to explore
it from a firsthand and up-close perspec-
tive— one they could only fantasize about
from the decks of ships. But even more
recently, humans have explored the abyss
with vehicles that even da Vinci and Verne
never conceived: remotely operated vehi-
cles (ROVs) and autonomous underwa-
ter vehicles (AUVs). The latest generation
ot these innovative deep-submergence
vehicles has enhanced human access to
extreme abyssal environments and has
greatly expanded the capabilities of Earth
and ocean scientists to investigate the far
reaches and depths of the global ocean.
ROVs, AUVs, and HOVs
ROVs are unoccupied underwater
vehicles controlled by a pilot aboard a
support ship and tethered to a fiber-optic
cable. The cable offers unlimited power to
the vehicle, so ROVs can stay on the bot-
tom longer than HOVs. The cable also
transmits real-time images and data to sci-
entists aboard ship. The ROVs pilot uses
the dexterous, force-feedback manipulator
arms to collect samples, perform experi-
ments, and deploy, service, and download
sensors in the deep.
AUVs are unoccupied, untethered
vehicles that are dispatched on pre-pro-
grammed missions in the ocean. Like
ROVs, they can operate submerged for
longer periods than HOVs. But their free-
swimming abilities, combined with more
precise control over their movements,
also allow them to explore and map much
more of the seafloor and the water above
it per dive.
AUVs and ROVs, in concert with
HOVs, will play indispensable roles in
establishing and servicing long-term sea-
floor observatories. (See "Seeding the
Seafloor with Observatories," page 28.)
Together, the complementary capacities
ot all three types of deep-submergence
vehicles provide synergies that have revo-
lutionized how scientists conduct research
in the ocean.
Revolutionary discoveries
The era of modern oceanography was
launched by the HMS Challenger expe-
dition (1872-76), and until recently has
relied on surface ships that go on expe-
ditions lasting from weeks to months to
collect data. Technical advances in instru-
ments, especially after World War II, let
scientists collect more and better data,
which fueled great leaps in knowledge
about the Earth and ocean.
Geophysicists mapped striking sea-
floor features, ranging from deep
trenches to the mid-ocean ridge system —
the globe-encircling underwater volcanic
mountain chain where the ocean crust is
born. (See "Unraveling the Tapestry of
Ocean Crust," page 40.) These discoveries
led to the plate tectonics revolution in the
early 1970s, which created a fundamental
new framework for understanding how
the Earth works.
Physical oceanographers pieced
together a general understanding ot the
physics that control the ocean's circulation
and water masses. With climatologists,
they realized the importance of interac-
tions between the oceans and atmosphere
in controlling Earth's climate.
It was only in 1977 that biologists,
geologists, and geochemists found lush
biological communities living oft chemi-
cals issuing from deep-sea hydrothermal
vents on the crest of the volcanic mid-
ocean ridge. This unexpected discovery
transformed conceptual thinking about
how and where life could exist on this and
other planets and has stimulated new lines
of inquiry into the origins of life itself.
(See "Is Life Thriving Deep Beneath the
Seafloor?" page 72, and "The Evolutionary
Puzzle of Seafloor Life," page 78.)
The fourth dimension: time
The discovery of hydrothermal vents
also catalyzed a realization that now domi-
nates the thinking of marine scientists—
the idea that myriad geological, chemical,
biological, and physical processes in the
deep ocean and on the seafloor are inter-
connected. (See "Living Large in Micro-
scopic Nooks," page 86.)
To observe and understand interrelated
20 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
The National Deep Submergence Facility
Woods Hole Oceanographic Institution operates this "fleet" of deep-submergence
research vehicles, which are used by scientists throughout the oceanographic com-
munity. It is part of the University-National Oceanographic Laboratory System,
an organization of 62 academic institutions and laboratories engaged in oceano-
graphic research. The National Deep Submergence Facility is funded by the Nation-
al Science Foundation, the National Atmospheric and Oceanic Administration, and
the Office of Naval Research.
Alvin
A 23-foot (7-meter)-
long human-occu-
pied vehicle (HOV)
takes two scientists
and a pilot as deep
as 4,500 meters
(14,764 feet). /
Argo II
A towed imaging and mapping
vehicle, with video and still
cameras and several different
acoustic sensors. It can operate
around the clock to depths of
20,000 feet (6,000 meters).
Medea
DSL-120A Depressor
DSL-120A
A towed 11 -foot (3. 3
vehicle equipped vJ
frequency (120 kilohertz) sonar to
map the seafloor at about 2-meter
resolution. It operates to depths
of 20,000 feet (6,000 meters). The
depressor isolates DSL-120A from
sea surface motion.
Jason 2 and Medea
A remotely operated vehicle (ROV) controlled
via a fiber-optic tether by a pilot aboard ship.
It operates to depths of 6,500 meters '"" "c
feet). Medea isolates Jason 2 fro ~"
surface motion.
Woods Hole Oceanographic Institut;
_o
o
Ol
5
processes that change over time, scien-
tists need to collect a variety of data — over
spatial scales ranging from centimeters to
kilometers and time spans ranging from
seconds to days, years, and decades. They
need to establish a continuous, compre-
hensive, long-term presence in the sea and
on the seafloor — instead of trying to piece
together processes by taking intermittent
snapshots of a relatively few places and
events. The difference in approach is like
seeking to understand family dynamics by
looking at a photo album versus spending
a few weeks with a family.
Here is where new ROVs and AUVs
will excel. Equipped with new suites ot
sensors, an expanding fleet ot autonomous
and remote deep-submergence vehicles
will give scientists more time to explore,
with expanded capabilities to map, sample,
and measure, over more territory — includ-
ing remote and inhospitable portions of
the oceans that have defied comprehensive
exploration by surface vessels.
Gliders, drifters, and REMUS
At the Woods Hole Oceanographic
Institution (WHOI), the synergy and col-
laboration among engineers
and scientists have consis-
tently pushed the envelope on
robotic oceanographic
technology. As a
result a diverse range
of vehicles has evolved
from drawing board to prototypes
and now into second generations of
vehicles working routinely on the ocean
frontier. Individual types of vehicles are
adapted and equipped to accomplish spe-
cific missions.
In the coming decades, for example,
oceanographers will be eager to measure
physical and chemical processes that drive
the world's ocean circulation and influence
Earth's climate. Many of these interactions
occur between the atmosphere and ocean
over vast regions, between and across
oceans. To this day, many oceans — includ-
ing the South Atlantic, Arctic, Indian, and
Southern Oceans— have not been well-
studied because ot their great size, remote
locations, or severe conditions (ranging
from sea ice to stormy seas).
Though satellites provide global cov-
erage, they cannot
provide data much
beyond the sea
surface. AUVs are
probably the only
way that we will fill in
these large gaps in our (Remote Environmental Sensing UnitS
REMUS
knowledge and gain a full understanding
of the short- and long-term oceanographic
processes within nearly half of Earth's
ocean basins.
For this mission, autonomous gliders
and drifters are being developed that can
travel across open oceans over hundreds of
miles and several weeks, taking measure-
ments along the way. Drifters such as Argo,
RAFOS, and Spray (now being developed
by Brechner Owens at WHOI and Russ
Davis at Scripps Institution of Oceanogra-
phy) are pre-programmed to deflate and
inflate a bladder, which causes them to
sink as much as 2,000 meters (6,500 feet)
in the ocean and then rise again to the
surface as they are carried along by cur-
rents. Gliders are essentially drifters with
wings that provide lift and allow them to
move horizontally. At WHOI, Dave
Fratantoni and colleagues are
leading efforts to use and develop
new glider systems.
Equipped with diverse
oceanographic sensors,
gliders and drifters can
make fine-scale mea-
surements ot tempera-
ture, salinity, current speed,
phytoplankton abundances,
and chemical changes,
and then surface periodi-
cally to transmit the data
via satellite to scientists on
shore. Fleets of these vehicles,
numbering in the hun-
dreds and eventually
thousands, will be able
Glider
to make comprehen-
sive studies of vast oceanic
ABE
(Autonomous Benthic Explorer)
regions. The portability of these vehicles
also makes them useful to study ephem-
eral or localized phenomena, such as phy-
toplankton blooms or upwelling events.
For surveys, mapping, and data col-
lection in shallow depths (330
feet, or 100 meters)
and coastal ocean
regions, WHOI
scientists and engi-
neers led by Chris-
topher von Alt built REMUS
(Remote Environmental Sensing Units).
This class of AUVs has already logged
thousands ot research missions. Specially
modified REMUS-based vehicles have been
used to search for mines in Iraqi harbors
and for cracks in tunnels supplying water to
New York City from upstate reservoirs.
A pioneer in the ocean frontier
In the deep ocean, the Autonomous
Benthic Explorer (ABE), developed by
WHOI researchers Dana Yoerger, Albert
Bradley, and Barrie Walden, has been a
pioneer. It has provided a testbed for inno-
vative robotics and electronics that have
demonstrated the viability and value of
deep-submergence AUV technology for a
wide range of oceanographic research. It
can dive to depths of 5,000 meters (16,500
teet) tor 16 to 34 hours, equipped with
an assortment of sensor packages (such
as high-resolution sonar, salinity, tem-
perature, and chemical recorders, current
meters, and magnetometers) to accom-
plish a variety of scientific missions, often
during the same dive.
AUVs add value to oceanographic
expeditions by col-
lecting data
autono-
mously
while
ships
simulta-
neously
acquire
data using
more traditional
means. AUVs can also
o
1
22 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Sentry
maximize the ettectiveness of
other vehicles. During a
2002 expedition to
the Galapa-
gos Rift led by
WHOI biolo-
gist Tim Shank
and his NOAA
colleague Steve
Hammond, ABE
demonstrated that it
could survey the seafloor by night, surface
at dawn, and deliver high-precision maps
that scientists in Alvin used to guide their
explorations that day.
ABE's capability to adapt its naviga-
tion to maintain a precise course over
rugged seafloor topography gives it the
ability to make high-precision seafloor
maps. In a typical dive, ABE's sonars can
image features less than a meter in length
and a few tens of centimeters tall on a
square kilometer of terrain. That is the
equivalent of being able to see footprints
on a football field from the bleachers.
The scale and resolution of these maps
alone are giving scientists the ability to
correlate seafloor features and biologi-
cal and geological processes in ways that
were previously impossible.
Sentry, Puma, and SeaBED
AUVs' high-resolution mapping abili-
ties will also play a key role in the develop-
ment of long-term, deep-sea observatories
by identifying optimal locales to deploy
sensors measuring a wide range of chemi-
cal, biological, and geological processes
over time. In the future, deep-sea obser-
vatories will include docking stations for
AUVs. These AUVs will be programmed
to be dispatched from their docks to rap-
idly respond to fast-breaking or ephemeral
events in the oceans that ships could never
reach in time to observe — an earthquake,
for example, or a temperature or chemical
change — and conduct timely sampling or
deploy experiments.
As good as ABE is, WHOI engineers
are striving to make it and its progeny bet-
ter. For instance, ABE was designed to be
*
SeaBED
able to move
in any direc-
tion or turn
in place
so it could
maneuver
close to the
bottom. It does this well, but
at the cost of efficiency in trav-
eling straight or up and down. It is
ideal for close-to-the-bottom
r\
surveying and photogra-
phy. The immense value
of these maps spurred
WHOI engineers to I
design a second-genera-
tion vehicle called Sentry,
optimized for sonar surveys
in rugged terrain. Sentry will
give up the ability to move
directly sideways or to hold posi-
tion and heading, but it will be much more
efficient in forward travel, steep climbs
and dives, and vertical motion.
WHOI scientists are also
designing other AUVs
with specialized capa-
bilities for specific
missions. One example
is SeaBED, developed
by WHOI scien-
tist Hanumant
Singh and col-
leagues. It is an AUV
that can fly slowly or
hover over the sea-
floor to depths of 6,000
feet (2,000 meters), making it particularly
suited to collect highly detailed sonar and
optical images of the seafloor. With Singh
and other colleagues, WHOI scientist Rob
Reves-Sohn is developing Puma and Jaguar,
two AUVs designed to search for and inves-
tigate hydrothermal vents under the ice-
covered Arctic Ocean. (See "Unique Vehi-
cles for a Unique Environment," page 25.)
ROVs and HROVs
AUVs' advantages are complemented
by ROVs, such as the pioneering Jason
ROV, first developed in the late 1980s by
Jaguar
Andrew Bowen, Dana Yoerger, and col-
leagues in the WHOI Deep Submergence
Laboratory (DSL), under the leadership of
Robert Ballard. The lab designed and built
an improved second-generation Jason
ROV, which was launched in 2002 and
is now in service as part ot the National
Deep Submergence Facility at WHOI.
One disadvantage of ROVs, however,
is that they can't cover as much ground as
AUVs in the same amount of
time. The ROV tether, which
can be thousands of meters
long and an inch thick, pro-
duces drag on the vehicle, and
makes it less maneuverable and
vulnerable to entanglement, espe-
cially in difficult terrains. To
combine the strengths ot both
types of vehicles, Bowen and
Yoerger of the WHOI Deep
Submergence Laboratory, in col-
laboration with Louis Whitcomb of Johns
Hopkins University, have begun to design
a Hybrid Remotely Operated Vehicle
(HROV), which will be able to
switch back and forth to
operate as either an AUV
or an ROV on the same
cruise. It will use a light-
weight fiber-optic cable,
only 1/32 of an inch in
diameter, which will allow
the HROV to operate and
maneuver at unprecedented
depths without the high-drag and
expensive cables and winches typically
used with ROV systems. Once the HROV
reaches the bottom, it
will conduct mis-
sions while
paying out
as much as
20 kilo-
meters
(about 1 1
miles) ot
microcable.
Pilots on
surface vessels will
remotely control
HROV
(Hybrid Remotely
Operated Vehicle)
Woods Hole Oceanographic Institut -
Conceptual Design for Alvin Replacement
I
o
I
01
c
'§>
01
Thrusters(6total)
Improved Science
Capabilities
• Dive to 6,500 meters allow!
access to 99% of seafloor
• Improved fields of view fo
pilot and observers
• Larger interior space and
increased science payload
• Variable ballast for mid-water studies
• High-speed data transmission to surf;
via microfiber cable
Improved Operational and
Maintenance Features
• Descends and ascends faster; longer time on bottom
• Improved navigation and communication
• Reduced physical and chemical disturbance to science study areas
• Improved battery access; fewer personnel sphere penetrations
• Improved safety systems
Observer viewports (2 of 4)
Personnel sphere
Manipulator (1 of 2)
Lights & Strobes
Still & Motion cameras
Pilots viewport
Retractable sample basket
Design Specifications
• Present Alvin capabilities
not compromised
• Size no larger than current DSV Alvin
Ascent/descent to 6,500 meters-2.5 hours
• Improved viewport size, arrangement, and number
• Increased sphere interior volume (27 cubic foot increase)
• Battery capacity increased by 30%
• Automated position keeping in all axes
• Operating costs comparable to current DSV Alvin
• Launch and recovery using R/V Atlantis
the HROV via the microcable, which
will be jettisoned upon completion of
the mission. Untethered, the HROV will
guide itself to the surface for recovery by
a ship, and the microcable is then recov-
ered for reuse.
The HROV will bring ROV capabili-
ties to places where it could not be used
before, such as the ice-covered Arctic. If
the ROV cable is severed during opera-
tions, the AUV capabilities will automati-
cally take over to continue the mission
autonomously or to return the vehicle
to the surface. The HROV will also be
capable of diving to 1 1,000 meters (36,000
feet) — deep enough to explore the deepest
parts of the world's oceans in the trenches
of the western Pacific.
A new replacement for Alvin
The remote capabilities of AUVs and
ROVs have proved enormously valuable,
but there is still no substitute for being
there. In particular, two-dimensional
images from ROVs still cannot provide
the direct, three-dimensional, full-con-
textual vision of the human eye, com-
bined with the ingenuity of the human
mind on the scene.
Forty years after Alvin was delivered
in 1964, WHOI scientists have embarked
on designing a new replacement HOV,
funded by the National Science Founda-
tion, with many improvements, including
an increased diving depth of 6,500 meters
(21,325 feet) that will allow it to reach
99 percent of the seafloor. (At its current
depth limit of 4,500 meters or 14,765 feet,
Alvin can reach 63 percent of the seafloor.)
Together, all these new deep-submer-
gence vehicles will be at the vanguard of a
new era of ocean exploration, leading us
deeper into the ocean frontier and augur-
ing a new era of discovery.
— Oceanus Editor Laurence Lippsett
contributed to this article.
Dan Fornari (left) became Director of the Deep Ocean Exploration Institute
at WHOI in 2004. He is internationally recognized for his research on volca-
nic and hydrothermal processes and mapping at mid-ocean ridges, and on the
, structure and magmatic processes at oceanic transforms and oceanic islands,
:- such as Hawaii and the Galapagos. Over 35 years, he has participated in 55
~- research cruises in the Pacific, Atlantic, and Indian Oceans, and has coin-
's pleted more than 100 dives in Alvin and other Navy subtnersibles. In 1993,
~ he became the first Chief Scientist for Deep Submergence at WHOI, and
£ helped usher in a new era in deep-submergence technology using Alvin with
the RO\' lason 2 and high-resolution sidescan sonar systems. Fornari has served on numerous scientific
panels, and national and international committees, including the President's Commission on Ocean
Exploration in 2003. Together, he and Susan Humphris developed Dive and Discover, an education and
outreach Web site (www.divediscover.whoi.edu) that brings the excitement of oceanographic science to
thousands of students each day. He is ably assisted most days by his personal assistant, Riley (right).
24 Oceanus Magazine -Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
ehicles for a Unique Environment
New autonomous robots will pierce an ice-covered ocean and explore the Arctic abyss
By Robert Reves-Sohn, Associate Scientist, Geology and Geophysics
Department, Woods Hole Oceanographic Institution
Imagine you have inherited a magnificent medieval cas-
tle. You wander its corridors, climbing spiral staircases
to hidden towers, delving purposefully into subterranean
caverns, and delighting in the details of .
its architecture, history, and artistic treasures. Over
time you come to realize there is a great North Wing
that has long been sealed off from the rest of the castle.
You've found old documents in the library describ-
ing construction of the North Wing, and it appears as
though it was built using rare materials that are not
found anywhere else in the castle. As best as you can
tell the castle's main thermostat is inside the North
Wing, which adds some urgency because lately
the castle seems to be getting inexplicably
warmer. And, perhaps most intriguing,
recent evidence suggests that some- ^f
thing— perhaps even something jr
unusual— might actually be
living in there.
Even so, it might be
more accurate \S) con-
fess that you've just got
to gain entrance to the
North Wing— because not
being able to enter rooms in
your own house is unbearable.
Puma ana Jaguar are autonomous
underwater vehicles (AUVs) designed
' to overcome the technical challenges
that now preclude under-ice operations
in the Arctic Ocean. They will home in
to an acoustic beacon and latch onto a
wire suspended from a hole in the ice.
Puma has sonars and sensors to search
wide areas and detect temperature,
chemical, or turbidity signals from
hydrothermal vent plumes (the green
lasers detect particulates in the water).
Puma can track the plume back to its
seafloor source, where Jaguar then will
be deployed to hover with camera and
lighting systems, high-resolution sonar,
and a manipulator arm for close-up
imaging, mapping, and sampling.
Jaguar.
t
Woods Hole Oceanographic Institutio
CTI
_o
o
01
01
Finding a way in
You come to realize, however, that
you're not the first to try. Numerous
intrepid individuals have dedicated
themselves to the pursuit over the years,
their stories comprising a veritable tome
of frustration and failure. And why?
Because the North Wing is hidden under
a moat ot water more than two miles
deep, which, in turn, is covered by a per-
manent layer of ice.
Moreover, it is so far north that the
compasses and gyroscopes typically used
for navigation are essentially useless. The
tools required to get through ice and into
the abyss to explore the North Wing can-
not be bought at any price. You will have
to make them yourself.
You have probably guessed by now that
the castle in this mental exercise is Earth,
and the North Wing is the vast, ice-cov-
ered Arctic Ocean Basin. If the old adage
is true that we know more about the sur-
face of our neighboring planets than we
do about Earth's ocean basins (and it is),
then nowhere is it more true than the
Arctic Basin.
A blank spot on the map of Earth
Deep-sea research is hard enough as
it is. But cover the ocean you're trying to
explore with a permanent ice cap, limit
your available field season to a few months
that are not too cold and dark, and factor
in a generally inaccessible location at the
very top of the world, and you can begin
to appreciate why we know so little about
the Arctic Basin.
In fact, were it not for a few Rus-
sian and American scientists whiling the
months away in camps on drifting ice
floes, we would know almost nothing at all
about the Earth's great North Wing. (Actu-
ally, U.S. and Soviet navies also gathered
data about the Arctic during the heyday
of Cold War submarine warfare, but this
information is generally classified.)
The human spirit cannot abide a puzzle
with a missing piece. But this is especially
so when the missing piece could fill in
crucial details about the origin of Earth's
oceans, the evolution of life, and our plan-
ets susceptibility to climate change.
An unexplored frontier
For climatologists and physical ocean-
ographers, it is often said that the Arctic is a
canary in the environmental coal mine. In
a warming world, the Arctic's delicately bal-
anced ocean circulation and sea ice appears
vulnerable to disruptions that could have
dramatic impacts on Earth's oceans and cli-
mate. Thus, climate change drives a large
percentage of Arctic research.
But the Arctic Basin is so unknown
and unique, it probably holds more undis-
covered scientific treasures than any other
ocean basin on Earth. Perhaps the hardest
challenge is deciding which fundamental
scientific questions to attack first.
For marine biologists, for example, the
Arctic represents a potential gold mine.
About 65 million years ago, the Arc-
tic Ocean basin became enclosed, with
no deepwater connections to any other
ocean basin on Earth. Species and bio-
logical communities in the Arctic Ocean
may have developed and evolved in isola-
tion, possibly making the Arctic a sort of
marine equivalent of Australia. (See "The
Evolutionary Puzzle of Seafloor Life,"
page 78).
A 2 1 st-century voyage of discovery
In 2001, Woods Hole Oceanographic
Institution (WHOI) scientist Henry Dick
was part of a team that conducted the
most detailed exploration to date of the
Gakkel Ridge, which transects the eastern
Arctic Basin and is perhaps the most enig-
matic tectonic plate boundary on Earth.
Like all mid-ocean ridges, the Gakkel
Ridge is an undersea volcanic mountain
chain where magma erupts to create new
ocean crust that spreads out on both sides
of the ridge. It was thought to be spreading
so slowly, however, that it would have little
volcanic and hydrothermal activity on it.
But Dick and colleagues found tantaliz-
ing clues ot active volcanism and ubiqui-
tous hydrothermal venting on the Gakkel
Ridge. What's more, they gathered evi-
An expedition in 2007 is planned to search
for, map, and sample hydrothermal vents for
the first time beneath the ice-covered Arctic
Ocean, along the Cakkel Ridge.
dence that seafloor spreading on the Gak-
kel Ridge occurs in a fundamentally dif-
ferent way compared to other previously
explored ridges. (See "Earth's Complex
Complexion," page 36.)
Dick and colleagues recovered rocks
from the Gakkel Ridge composed ot mate-
rials that normally reside in the mantle
deep within Earth's interior, and that are
rarely found on Earth's surface. These
rocks are perhaps the closest modern ana-
logues to the kind of volcanic rocks that
erupted billions of years ago in the early
stages ot Earth's history. They have a dis-
tinct chemistry that affects their interac-
tion with seawater circulating through
hydrothermal vent systems. (See "The
Remarkable Diversity of Seafloor Vents,"
page 60.) These chemical reactions release
exceptionally large amounts ot chemi-
cal "food" for the kinds of evolutionarily
ancient microbes that reside at the roots
of the Tree of Lite. (See "Is Life Thriving
Deep Beneath the Seafloor?" page 72.)
The hydrothermal vent fields on the Gak-
kel Ridge could therefore provide a means
to study hydrothermal activity on an early
26 Oceanus Magazine • Vol.42, No. 2 • 2004 • oceanusmag.whoi.edu
Earth, and possibly even provide clues to
the origin of life on this planet.
Many questions, little data
Scientists only began to get their first
detailed look at the Arctic seafloor between
1995 and 1999, when the U.S. Navy and the
National Science Foundation (NSF) teamed
up to use Na\T nuclear submarines tor
unclassified scientific investigations.
We still have almost no data from
most of the mountain chains in the Arctic
Basin, so we're still guessing about their
composition, age, and origin. Yet, the
sparse evidence we do have suggests that
the Arctic Basin did not form the same
way other ocean basins did, and that it
may have the oldest extant ocean crust in
the world.
That's why we dream about retrieving
data from beneath the Arctic ice cap, and
why we have begun to harness 21st-century
technology to make those dreams come
true. Nuclear subs are far too expensive to
build and operate, and besides, they are
crushed like tin cans at full Arctic seafloor
depths. Instead, in our quaint New England
village of Woods Hole, we are developing
autonomous underwater vehicles (AUVs)
and other advanced deep-submergence
technologies that can open up the Arctic
Basin to scientific investigation.
Overcoming the ice barrier
Though many scientists dream of get-
ting beneath the ice to study the Arctic
Basin, the long-standing commitment to
ocean technology and instrumentation at
WHOI gives scientists here the opportu-
nity to turn dreams into reality. We for-
mally began a program to develop AUVs
specifically for under-ice operations in
the Arctic in January 2000 when the NSF
Office of Polar Programs awarded WHOI
a grant to design, fabricate, and test a
prototype vehicle called APOGEE, the
Autonomous Polar Geophysical Explorer.
The objective was to develop a swimming
robot that can carry a variety of scien-
tific sensors to explore the most inacces-
sible regions of the Arctic Basin, and that
can be deployed and recovered through a
small hole in the ice.
Life is easier if AUVs can simply pop
up on the ocean surface for recovery, but
Arctic pack ice adds complications. APO-
GEE was designed with critical acoustic
navigation and control systems that allow
it to navigate to a homing beacon, latch
itself to a wire suspended from a hole
in the ice, and ultimately be recovered
by scientists on an icebreaker or an ice
camp. Without this essential capability,
an AUV in the Arctic would almost cer-
tainly be lost.
To the Arctic and beyond?
We conducted our last set of sea tri-
als with APOGEE in 2003. Now, we have
begun to build the next generation of Arc-
tic vehicles — under a joint grant (with the
University ot Maryland's Space Systems
Lab) trom the Astrobiology Science and
Technology tor Exploring Planets program
of the National Aeronautics and Space
Administration.
Our mission will be to use AUVs to
find, map, and sample hydrothermal
vent fields on the Gakkel Ridge. We will
develop instrumentation that will guide
future efforts to search for life on Europa,
a Galilean moon of Jupiter, which may
have two necessary ingredients for life:
active volcanism and an ocean — albeit an
ice-covered one.
The mission will also allow us to study
Arctic vent fields for the first time. We
will use a small group of purpose-built
AUVs, each with different characteristics
and equipped with state-of-the-art sen-
sor systems. They will work in concert
to study Arctic vent fields. For example,
a "bloodhound" AUV (named Puma, for
Plume Mapper) will be equipped with
sensors that can detect tiny telltale tem-
perature, chemical, or turbidity signals
in the water. It will survey a wide area to
"snift out" one of the hydrothermal vent
plumes that Henry Dick got whiffs of in
2001 and follow it back to its vent field
source on the seafloor.
Once a vent is found, "hummingbird"
AUVs (named Jaguar), will be deployed.
These will be able to hover in place, and
equipped with camera and lighting sys-
tems, high-resolution sonar, and a manip-
ulator arm with storage canisters, they will
be used for mapping, imaging, and sam-
pling at vent sites.
Fully autonomous methods have never
been used to find and image, not to men-
tion obtain samples from, vent fields in
any ocean, not to mention an ice-covered
one. The technical challenges are serious
and legion, but they are worthy of a cut-
ting-edge oceanographic institution such
as WHOI. We are of necessity drawing
on the expertise, inspiration, and creativ-
ity of dozens of experts hailing from every
department within the institution.
There is no denying that we are
attempting an ambitious project that faces
stiff technical challenges, but this ener-
gizes and motivates us. Ultimately, we will
succeed or fail based on the talent and
dedication of the scientists, engineers, and
technicians who are conceiving, designing,
and fabricating the new instrumentation
we will take to the Arctic. Anyone who
knows the people on our team would be
reluctant to bet against us.
Robert Reves-Sohn is
an Associate Scientist
in the Geology and
g Geophysics Depart-
£ ment at Woods Hole
5 Oceanographic In-
-^ stitution. He decided
™ to pursue a career in
geophysics when, as a grad student at Scripps
Institution of Oceanography, he read the in-
struction sheet on "How to Prepare a Perfectly
Putrid Poster" Jor his first American Geophysi-
cal Union meeting back in 1992. The official
instructions suggested that the presenter eschew
logical arrangement of the material, use illeg-
ible fonts, and, most importantly, "consume
copious quantities of beer" before presenting the
poster. He immediately realized he had found
the peer group he had been looking for. His
interest in AUVs is a natural outgrowth of his
innate tendencies as a teehnophile, although the
notion of using AUVs to get under the Arctic ice
pack came to him while playing with a Sesame
Street submarine and giving his (then) newborn
(now S-year-old) daughter a bath.
Woods Hole Oceanographic Institution 27
jf
o
Seeding the Seafloor with Observatories
Scientists extend their reach into the deep with pioneering undersea cable networks
<v
By Alan Chave, Senior Scientist
Applied Ocean Physics & Engineering Dept.
Woods Hole Oceanographic Institution
It would be a lot easier to explore the
deep ocean, it we only had some electri-
cal outlets and phone jacks on the seafloor.
With 21st-century technology, we are
starting to install some.
On land, Earth scientists can plug their
instruments into electric power lines or
rig them with solar panels to make long-
term measurements of earthquakes, the
planet's magnetic field, and other episodic
or ongoing geophysical processes. But
many of Earths most fundamental, planet-
shaping processes occur only beneath the
ocean, where deploying similar instru-
ments has presented unique challenges.
Expeditions to remote ocean regions are
typically more expensive and time-consum-
H20(Hawaii-2 Observatory)
In 1998, scientists used the remotely operated vehicles Jason and Medea to create the
pioneering long-term seafloor observatory called H20 (Hawaii-2 Observatory). They
spliced an abandoned submarine telephone cable into a termination frame. The frame
relays power and communications to a junction box, which serves as an electrical outlet
for scientific instruments.
ing than land-based expeditions. Marine
scientists are also limited by the availability
of ships and must contend with corrosion
problems peculiar to ocean environments.
And without sunlight or a continuous elec-
tricity supply, scientists have had a limited
capacity to supply power to instruments
and data recorders in the ocean.
As a result, the record of land-based
measurements contrasts starkly with the
near-total absence of long-term geophysi-
cal data from the seafloor. Instead, ocean
scientists tend to have a lot of snapshots
of what is happening in the ocean. Since
three-quarters of the Earth is covered by
ocean, that's like trying to monitor the
dynamics of a household by observing
events for just a few hours a month in only
one bedroom and a closet.
Long-term presence on the seafloor
To comprehend Earth's dynamic behav-
ior, ocean and Earth scientists must do
more than observe small regions for short
periods. Advances in communications,
robotics, computers, and sensor technol-
ogy now make it possible to get wider and
longer views of the oceans.
Remotely operated vehicles (ROVs)
and autonomous underwater vehicles
(AUVs) can work longer and deeper, and
with ever-growing abilities. (See "Real-
izing the Dreams of da Vinci and Verne,"
page 20.) Fiber-optic cables and other
communications technologies allow more
data to be transferred at greater speeds.
And with new materials, we can develop
hardy instruments that can withstand
harsh seafloor conditions.
Establishing a long-term presence in
28 OceanusMagazine-Vol.42, No. 2- 2004 • oceanusmag.whoi.edu
In 2003, H2O got its first renovation. ROV Jason 's manipulator arm
plugs an instrument into the deep-sea observatory's junction box.
Jason's arm pours glass beads into a container used to bury a
seismometer beneath seafloor sediments.
the oceans is no longer a dream. At the
H2O "cabled" seafloor observatory (mid-
way between Hawaii and California), the
worlds second deep-sea science station
has been setting a new scientific precedent
since 1998 (the first was established off
Japan in 1993).
In 2005, the Monterey Accelerated
Research System (MARS) project is
scheduled to begin full-time observation
of phenomena in Monterey Canyon off
California. And by the end of the decade,
ocean scientists hope to wire and network
an entire tectonic plate for observation
through the North East Pacific Time-inte-
grated Undersea Networked Experiments
(NEPTUNE) program.
These initial outposts on the ocean
frontier will allow us to examine in detail
the interactions that shape the seafloor,
generate earthquakes, fuel volcanoes, form
ore and oil deposits, transport sediments,
circulate currents, and support lite in
deep-ocean environments.
Bringing the power of the Internet to
the seafloor, these cabled observatories will
connect scientists in their labs directly to
submarine experiments. Scientists will be
able to monitor and adjust instruments,
or to dispatch AUVs to observe episodic
events that previously went undetected.
These observatories will also offer students
and the public an unprecedented, intimate
view ot ocean exploration in action.
H20 — a pioneering observatory
In 1998, Woods Hole Oceanographic
Institution (WHOI) and University ot
Hawaii researchers seized an opportu-
nity to take a significant first step toward
opening the relatively unexplored sub-
merged regions of Earth to more thor-
ough examination.
Beneath 5,000 meters of water (16,400
feet), a submarine telephone cable called
Hawaii-2 (HAW-2) stretched across the
seafloor from Hawaii to California. AT&T
laid the cable in 1964, but when it broke
in 1989, the company looked to its fiber-
optic future and decided not to fix it.
Instead, it donated the cable to the scien-
tific community.
The HAW-2 cable is like a long exten-
sion cord delivering electrical power to
the seafloor and providing a means for
two-way, shore-to-seafloor communica-
tions. With funding from the National Sci-
ence Foundation (NSF), a WHOI/Hawaii
science team developed and built a junc-
tion box to be spliced into one end of the
HAW-2 cable. The junction box is like an
eight-socket power strip for the seafloor.
No longer constrained by battery
power and limited data recorders, we
could install instruments to take con-
tinuous measurements ot slowly evolv-
ing Earth processes and rapidly occurring
events. The dream of America's first long-
term, deep-ocean observatory became
a reality with the establishment of the
Hawaii-2 Observatory, or H2O.
The first two instruments plugged into
H2O were a seismometer and a deep-
water pressure gauge. The seismometer
records seismic waves generated by earth-
quakes, allowing scientists to locate and
study the sources. The pressure gauge aids
researchers in detecting tsunamis, large
waves generated by earthquakes in the
open ocean.
In the summer of 2003, H2O got its
first renovation. The junction box was
raised, upgraded with the latest communi-
cations and power interlaces, and lowered
again to the seafloor.
In 2005, several new instruments are
scheduled to be installed. Magnetometers
and other instruments will be set up as
part of a seafloor geomagnetic observa-
tory. A new benthic biology experiment,
which includes digital cameras and sedi-
ment traps, will be able to observe what
is living on and falling to the deep ocean
floor. A 1.5-kilometer (1-mile) cable will
be laid from the junction box to a bore-
hole where another more sensitive seis-
mometer will be deployed.
MARS and VENUS
The next step for deep-ocean observa-
tories is to incorporate fiber-optic com-
munications, allowing high-speed, high-
bandwidth transfer of science data and
Woods Hole Oceanographic Institution 29
01
_o
o
<v
MARS (Monterey Accelerated Research System)
The next step for deep-sea observatories will be MARS, a test bed using a fiber-optic cable
that allows high-speed, high-bandwidth communications and data transfer. MARS will
have a 40-mile cable along the north side of Monterey Canyon, connecting a shore station
at Monterey Bay Aquarium Research Institute (MBARI) to an undersea node that serves as a
power and data-transfer station for instruments.
computer commands. In the fall of 2002,
NSF provided the funding to build a test
bed for fiber-optic cabled observatories — a
proving ground for the next generation of
seafloor observatories.
The Monterey Accelerated Research
System (MARS) will consist of 70 kilome-
ters (44 miles) of submarine cable laid out
from a shore station along the northern
side of the Monterey Canyon to a single
science node located 1,200 meters (almost
4,000 feet) below the ocean surface. The
science node will serve as a power and
data-transfer station for instruments.
Ocean scientists from around the
world will be able to design new instru-
ments and then test them by plugging into
one of MARS's four standardized ports.
Each port will support data transfers of
up to 100 megabits per second — compa-
rable with some of the fastest land-based
commercial data networks. The cable also
will supply up to 10 kilowatts of power —
enough to supply a few terrestrial houses,
and several orders of magnitude more
power than can be supplied with batteries.
Partners in the MARS program include
the Monterey Bay Aquarium Research
Institute (MBARI), Woods Hole Ocean-
ographic Institution, the University of
Washington (UW), and the National
Aeronautics and Space Administrations
Jet Propulsion Laboratory (JPL).
The University of Victoria (UV) is
simultaneously developing a complemen-
tary shallow-water test bed — known as the
Victoria Experimental Network Under the
Sea (VENUS)— in the Strait of Georgia
and Saanich Inlet between Vancouver and
Victoria, British Columbia.
On to NEPTUNE
Just off the coast of the Pacific North-
west of Canada and the United States lies
an ocean scientist's dream. Nearly all of
the major Earth-shaping features and pro-
cesses— seafloor volcanism, hydrothermal
vent systems, earthquakes, seafloor spread-
ing, and subduction zones— converge in
a reasonably small geographic area. At its
coastal edge, sediments from the North
American continent pour into the deep
sea, while the waters teem with life in the
Pacific Northwest's great fisheries.
Quite simply, the Juan de Fuca tectonic
plate is a comprehensive natural labora-
tory for ocean science. The plate's proxim-
ity to shore and relatively small size make
it a cost-effective candidate for incremen-
tal but eventually extensive cabling.
The North East Pacific Time-inte-
grated Undersea Networked Experiments
(NEPTUNE) project aims to establish
an extensive Earth/ocean observatory
across and above the Juan de Fuca Plate
off the West Coast of the United States
and Canada. Researchers are propos-
ing to lay down 3,000 kilometers (1,865
miles) of high-speed fiber-optic subma-
rine cables that will link a series of 30
seafloor nodes, each about 100 kilome-
ters (62 miles) apart. Those nodes would
support thousands of assorted measur-
ing instruments, video equipment, and
robotic vehicles that could upload power
30 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
and download data at undersea docks.
Unlike conventional telephone cables,
which supply power from shore in a
straight line, end to end, NEPTUNE
would operate like a power grid, distribut-
ing power simultaneously and as needed
throughout the network. More than 100
kilowatts ot power would flow through
the system (roughly the amount of power
needed to supply 50 to 100 homes at any
given moment).
NEPTUNE would work much like a
campus data network, with nodes analo-
gous to buildings and each instrument like
a computer workstation. It would provide
real-time transmission of data and two-
way communications for up to 30 years.
Developers of the NEPTUNE network
include many of the same institutions
involved in previously mentioned observa-
tory initiatives (WHOI, UW, JPL, MBARI,
and UV). We hope NEPTUNE is opera-
tional by 2008. It will cost approximately
$200 million to develop, install, and operate
through the first five years. An entire net-
work of seafloor instruments, distributed
over an area of 500 by 1,000 kilometers
(310 to 620 miles), would cost less than one
satellite or one Coast Guard icebreaker.
Alan Chave is a Senior Scientist in the WHOI
Applied Ocean Physics and Engineering De-
partment, where he specializes in going to meet-
ings and writing reports. When time allows,
he builds "toys" to throw in the ocean (hoping
they will come back) and has been quite suc-
cessful at this over the years. His major research
interest is the design, construction, installation,
and servicing of ocean observatories. His group
provided the technical lead for the Hawaii-2
Observatory, and he acts as project scientist for
the ambitious NEPTUNE project planned for
installation in 2008-2009. Chave lives in Fal-
mouth with his wife, daughter, and Portuguese
water dog.
VENUS (Victoria Experimental Network Under the Sea)
The University of Vancouver is developing a shallow-water undersea observatory called
VENUS in the Strait of Georgia between Victoria and Vancouver, Canada.
NEPTUNE (North East Pacific Time-integrated
Undersea Networked Experiments)
The NEPTUNE project aims to establish an Earth/ocean observatory across the Juan de
Fuca Plate off the U.S. West Coast. Researchers propose to lay 1,865 miles of fiber-optic
submarine cables linking 30 seafloor nodes that support assorted instruments and
robotic vehicles.
Explorer
Plate
.' . ;
Victoria'
' Seattle-
\
S-r— •«r.
Juande \
Fuca Plate v\
Portland
North
American
^ ' Junction Boxes — Plate Boundaries
o —Fiber Optic Cable Borders
•* • Moorings Rivers
Plate
Woods Hole Oceanographic Institution 31
1
o
I
o
A Sea Change in Ocean Drilling
Scientists launch a new drill ship and ambitious research plans
By Dennis Normile and Richard A. Kerr
Science Magazine
In the early 1960s, geologists took
their first shot at drilling all the way
through Earth's crust and into its man-
tle with the Mohole Project. It turned
out to be a disaster. Named for the
Mohorovicic discontinuity, the bound-
ary between the crust and mantle, the
ambitious attempt to penetrate 6 kilo-
meters of crustal rock was sunk by cost
overruns and management problems
and scrapped after a few test holes.
But out of that debacle came a
highly successful international scien-
tific endeavor. The decision to drill
Mohole from a barge — to take advan-
tage of the fact that the oceanic crust
Reprinted and adapted with permission from
"A Sea Change in Ocean Drilling," Science
300:410-412 © 2003, American Association for
the Advancement of Science.
is much thinner than the continental
crust — laid the foundation for mod-
ern-day scientific ocean drilling. And
researchers have exploited the world it
opened up to make seminal
discoveries about the planet.
Now, those efforts are about
to enter a new era.
Over the past 40 years,
researchers have drilled
more than 2,900 holes in the
ocean floor, retrieved 319
kilometers of mud and rock
core, and studied 35,000
samples. The legacy of
ocean drilling includes val-
idating the theory ot plate
tectonics and tracing
Earths changing climate back 100 mil-
lion years, as well as inventing the field
of paleoceanography.
Since 1984, that work has been car-
ried out under the 22-country Ocean
Drilling Program (ODP), a unique
effort that ended in September 2003.
But it will be replaced by something
even more ambitious: In October 2003,
Japan and the United States inked
an agreement formally creating the
Integrated Ocean Drilling Program
(IODP). It will eventually include 20
or so other countries, cost twice as
much to operate as its fore-
runner, and use two, and
at times three, ships rather
than one.
Initially, IODP will rely
on an upgraded U.S. drill
ship, either a revamped ver-
Since 1 985, the research vessel JOIDES Resolution has been the workhorse for scientific ocean drilling. Through 2003, the 470-foot-long ship with
its 202-foot derrick has drilled more than 1,800 holes in the ocean crust and retrieved samples at some 670 sites.
32 Oceanus Magazine • Vol.42, No. 2 • 2004 • oceanusmag.whoi.edu
sion of OOP's workhorse, the IOIDES Res-
olution, or a new vessel with similar capa-
bilities. By late 2006, it will be joined by a
brand-new ocean drilling vessel, Japan's
Chikyu, equipped with technology that
will allow it to literally break new ground.
Together, the two ships will enable
Earth scientists to bore more and much
deeper holes than is currently possible
and in locations that are now inaccessible.
There are even going to be "mission-spe-
cific platforms" that will drill niche loca-
tions such as the icy Arctic Ocean and
shallow coastal waters.
A new drill ship for a new era
The biggest change in operational
capabilities will come when the 210-meter,
57,500-ton, $475 million Chikyu starts
drilling. For all its achievements, the Reso-
lution has serious limitations. It can't drill
in shallow water or farther down than 2
kilometers. Nor can it tolerate the icy con-
ditions of the Arctic Ocean. What's more,
sedimentary basins have been largely off-
limits because oil and gas deposits have
posed safety and environmental hazards.
The Chikyu will overcome some of
these constraints. It will have a second
pipe, called a riser, that will enclose the
drill pipe and allow circulation of a heavy
but fluid drilling mud that will flush
debris from deep holes and shore up
unstable sediments. The arrangement will
also protect against blowouts when the bit
penetrates pressurized oil or gas deposits.
Attempts at drilling very deep holes using
the Resolution were frustrated by the fric-
tion and by debris piled up in the hole.
"Because of the capabilities ot the riser
vessel, [all sorts of drilling] projects will
be more viable," said Hisatake Okada, a
paleoceanographer at Hokkaido Univer-
sity in Sapporo.
But all of this comes at a steep price.
The annual budget of OOP ran about $80
million, with 60 percent of that sum put
up by the U.S. National Science Founda-
tion (NSF) and the rest split among the
other member countries. Countries spent
additional funds to support scientists ana-
A NEW DRILL SHIP— The Integrated Ocean Drilling Program's drill ship Chikyu ("Earth" in
Japanese) is launched in Japan, in 2002. The 57,500-ton, 2 1 0-meter (689-foot)-long ship will be
capable of drilling 7 kilometers (4.35 miles) below the seafloor — sufficient to reach the mantle. Its
derrick was installed in September 2003 and after sea trials, Chikyu should be ready by late 2006.
lyzing drilling samples and data.
In comparison, lODP's annual oper-
ating budget is expected to start at
$160 million and rise depending on the
amount and nature of drilling carried
out. Japan and the United States will split
at least two-thirds of the operating costs
equally, with other countries providing
the rest — and also funding mission-spe-
cific platforms.
Researchers are arguing that the sci-
entific advances will be worth the price,
from a better understanding ot earthquake
mechanisms and the history of global
climate change to the discovery of new
energy sources and unusual microbes for
use in biotechnology. And governments so
far seem convinced.
Exploring large igneous provinces
Ocean drilling's first significant achieve-
ment came in geophysics. "Past successes
have changed our understanding of how
Earth works," said oceanographer Larry A.
Mayer of the University of New Hampshire,
Durham. By dating rock recovered from
numerous seafloor locations, researchers in
the early 1970s confirmed the basic cycle
of plate tectonics: New ocean crust forms
at mid-ocean ridges and spreads outward
toward deep-sea trench subduction zones.
Crustal drilling also showed how great
upwellings of hot rock, called plumes, could
create chains of islands and seamounts such
as Hawaii. (See "Motion in the Mantle,"
page 6.)
These discoveries have raised new
questions about solid Earth cycles and
geodynamics, one of three broad themes
in lODP's initial science plan. Earlier
drilling showed that large parts of the
crust were formed by anomalous volcanic
events separate from plate tectonics.
Oceanic plateaus, so-called large igne-
ous provinces, formed mostly during the
mid-Cretaceous period 100 million to 140
million years ago when massive amounts
of material burst through tectonic plates,
venting heat and magmatic gases from
Earth's interior. These features have as
yet been barely sampled by drilling.
Researchers hope that data from a combi-
nation of numerous shallow holes drilled
by a riserless ship and deep holes drilled
later by Cliikyu may relate these events to
Earth's evolution and reveal whether or
not they triggered climatic changes that
led to mass extinctions.
Another major geophysical target will
be subduction zones, where the clash of
Woods Hole Oceanographic Institution
Drilling the seafloor
Derrick
The moon pool is a 7-
meter (23-foot)-wide
hole below the drill
ship's derrick through
which the drill pipe is
lowered.
The process of lower-
ing the drill bit, which
is affixed to the end
of the drill pipe, takes
about 72 hours in
5,500 meters (18,045
feet) of water. To core
through the seafloor,
the entire drill pipe is
rotated.
Thrusters mounted
beneath the ship keep
the massive vessel from
drifting off the drill site.
Drill crews thread pipe
sections — each about
28.5 meters (93.5 feet)
long and weighing
about 874 kilograms
(1,925 pounds).
Acoustic beacons
guide crews to re-entry
cones, which are used
to resample previously
drilled holes.
sinking and overriding plates generates 90
percent of the worlds earthquakes. Chikyu's
first target, reached by consensus, will be
the Nankai Trough subduction zone off-
shore ot Honshu, Japan's main island.
Chikyu's riser will allow boring through
the deep sedimentary deposits atop over-
riding plates. Those deposits were off-lim-
its to the Resolution because of the danger
ot a blowout caused by inadvertently tap-
ping into oil and gas deposits and by the
depth of the fault target.
Gaku Kimura, a geologist at Univer-
sity of Tokyo, says Chikyu will also be
able to install a new generation of instru-
ments in the bore hole to monitor fault
zone temperatures, stresses, deformation,
and fluid pressures.
"This is a completely different scientific
approach" to studying rock samples, said
Kimura. "It's like the difference between
studying a live human being and dissect-
ing a corpse." An improved understand-
ing of earthquake mechanisms could help
Japan and other onshore communities
assess the risk of future earthquakes.
IODP may even take another shot at
penetrating the Mohorovicic discontinu-
ity. With the lubricating drilling mud cir-
culating through its riser, Chikyu should
be able to reach 6 kilometers and into the
upper mantle. Such a hole would help
refine knowledge ot the structure, com-
position, and physical properties of the
oceanic crust.
Probing climate changes on Earth
Although geophysics was the prime
motivation for the first ocean drilling
cruises, scientists in other disciplines soon
capitalized on the data obtained from the
cores. "Paleoceanography is one of the
strong successes of ocean drilling," said
Jerry McManus, a paleoceanographer at
the Woods Hole Oceanographic Institu-
tion in Massachusetts.
Paleoceanographers recognized that
cores recovered from layered sediments
provided clues to a variety of climatic phe-
nomena, sometimes going back 120 mil-
lion years. McManus credits ocean drilling
34 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
with clinching the orbital theory of cli-
mate change over millions of years, when
Earth's wobbling drove climate oscilla-
tions. It also documented extreme climates
such as the thermal maximum ot the late
Eocene (55 million years ago) and rapid
climate change.
But the use of drilling for paleoceano-
graphic studies has been held back by the
limitations ot the Resolution. It cannot
drill in water much shallower than 100
meters, ruling out the inner continen-
tal shelves and coral reefs that hold long
records of climate and sea-level change.
And it can't handle more than the passing
bit of sea ice, which has kept it entirely out
of the Arctic Ocean.
"You could lay out all the existing Arc-
tic cores in my office," said oceanogra-
pher Theodore Moore of the University of
Michigan, Ann Arbor. A successful mis-
sion to the deep Arctic, he said, would
provide "an entire history from 50 million
years ago to the present."
Under lODP's initial science vision,
mission-specific platforms capable ot
drilling in niche locations would play
a major role in studying environmen-
tal change, processes, and effects. Euro-
pean scientists, for example, obtained
sufficient funding to send a drill ship to
retrieve long sediment cores from the
Arctic Ocean in the summer of 2004.
Seafloor fuel, sub seafloor life
The third leg of the IODP scientific
tripod, studying the deep biosphere and
the sub-seafloor ocean, is also the newest.
The original ocean drillers never imagined
there could be life within the extreme tem-
peratures, pressures, and chemical envi-
ronments ot the ocean floor. But reports
of microbial colonies at seafloor vents
and volcanic rifts demonstrated other-
wise. Now some experts in extremophiles,
as these microbes are called, believe that
as much as two-thirds of Earth's micro-
bial population may be buried in oceanic
sediment and crust. (See "Is Life Thriving
Deep Beneath the Seafloor?" page 70.)
One major challenge will be to define
CORE CURRICULUM — A team of scientists on the drill ship Resolution examines seafloor core
samples to reconstruct events and phenomena occuring over millions of years of Earth history.
the range of temperatures, pressures,
chemistry, and other conditions under
which these seafloor communities thrive
and to map their geographical distribu-
tion. Researchers would also like to clarify
whether these microbes get their nutrients
from material that filters down from the
surface or from updrafts of fluids flowing
through the interface between sediments
and hard rock.
The findings "could revolutionize
ideas about the origins of life," said Asa-
hiko Taira, director-general of the Center
tor Deep Earth Exploration at the Japan
Marine Science and Technology Center.
Researchers also hope to add to the hand-
ful of industrially useful microbes already
isolated from deep-sea regions.
Another underresearched area of
inquiry is gas hydrates, deposits of ice-
encapsulated methane. Although a poten-
tial new source of clean energy, they also
could release a significant amount of
greenhouse gases into the atmosphere if
thawed as a result of global warming. Sci-
entists want to learn how microbes gen-
erate the methane, how hydrates form,
and whether methane can be produced
at prices competitive with those of other
fuels. (See "When Seafloor Meets Ocean,
the Chemistry Is Amazing," page 66.)
These intriguing questions weren't even
on the radar screen of those working on
the ill-fated Mohole Project. But if IODP
comes up with some answers, its scientists
will owe a debt of gratitude to those who
conceived and carried out the first scien-
tific attempt to probe what lies beneath
Earths oceanic crust.
Dennis Normile of Science
magazine studied civil
engineering at Villanova
University and briefly
considered going on for
graduate work in geology.
Instead he ended up doing
his best to counter the ef-
fects of earthquakes on buildings as a structural
engineer working in Anchorage, Alaska. He has
been the Japan correspondent for Science since
1995 and has followed the evolution of the IODP
ever since it was first proposed at a meeting in
Hayama. Japan, in 1996.
Richard A. Kerr of Science
magazine has covered
the Earth and planetary
sciences (and a bit of pa-
leontology) since 1977 at
Science. He went there
from the University of
Rhode Island a week af-
ter successfully defending his dissertation on the
hutnics in seawater. The frustrations of trying to
understand such gunk were enough to drive linn
into journalism, where the breadth of an educa-
tion in oceanography has served him well.
Woods Hole Oceanographic Institutio
m
01
ui
01
01
c
Earth's Complex Complexion
Expeditions to remote oceans expose new variations in ocean crust
By Henry J.B. Dick, Senior Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Even as you read this, Earth's crust is
continually being reborn and recycled
in a dynamic process that fundamentally
shapes our planet. We're not generally
aware of all this action because most of
it occurs at the seafloor, under a formi-
dable watery shroud, and often in remote
regions of the oceans.
The creation and cooling of oceanic
crust is the primary means by which
heat escapes from Earth's interior. This
dynamic planet-scale crucible transports
heat, chemicals, and minerals from Earth's
interior to its surface.
Over the long haul, the process ulti-
mately determines Earth's chemical
makeup. It affects the amount of carbon
dioxide and water in the crust, oceans, and
atmosphere, and it produces zinc, copper,
and other mineral deposits, including some
gold and silver. So how Earth's crust forms
is more than just an academic question.
The seafloor layer cake
Several generations of scientists have
dredged rock samples from the seafloor,
employed submarines and robots to study
it, and even drilled into it to learn a con-
siderable amount about the shallow oce-
anic crust. (See "A Sea Change in Ocean
Drilling," page 32.) We've analyzed seismic
waves that penetrate and reflect off rock
MAIDEN VOYAGE — Scientists and crew on the first cruise of the U.S. icebreaker Healy in 200 1 successfully performed 200 dredge operations in the
ice-covered Arctic Ocean, collecting some 4,000 seafloor rock samples.
36 Oceanus Magazine -Vol. 42, No. 2- 2004 • oceanusmag.whoi.edu
layers deep in the crust in an effort to
decipher its characteristics— similar to
the way physicians use an MRI to peer
below the skin. (See "Listening Closely to
'See' into the Earth," page 16.) We've also
studied ophiolites— isolated portions of
the seafloor that tectonic forces have
thrust up and exposed on continental
margins. (See "Unraveling the Tapestry of
Ocean Crust," page 40.)
From early studies, a simple picture
emerged: It seemed the ocean crust was
relatively homogenous in composition,
structure, and thickness — sort of a geo-
logical three-layer cake about 6 to 7 kilo-
meters (3.7 to 4.3 miles) thick. On top
was lava that spilled out and cooled rap-
idly into a glassy substance called basalt,
which carpeted the ocean bottom. Below
were great, vertical sheets of molten rock
called dikes — the pathways by which
magma was injected to the surface from
deeper layers. Finally, lying atop the man-
tle itself, was a lower layer composed of
magma that rose directly from the man-
tle, cooled more slowly, and crystallized
into a rock known as gabbro.
This was a neat picture and a great first
step, but nature, like life, usually turns out
not to be so simple. And so it is with sea-
floor crust.
Forays to remote mid-ocean ridges
The layer cake model began to crumble
in the 1980s when scientists began dredg-
ing, drilling, and diving around the great
transform faults that offset volcanic mid-
ocean ridge segments in the Atlantic and
Indian Oceans. They gathered new evi-
dence suggesting that the gabbro layer of
ocean crust may be missing entirely in
some locations and thicker than expected
in others. It turned out that ocean crust,
forged under different circumstances in
different places, could be very different.
Surprising discoveries further reshaped
our concept of ocean crust after several
recent research voyages to the world's
slowest-spreading ocean ridges, which
are located in remote, largely unexplored
regions. In particular, our work on the
BREAKING THE ICE— The U.S. icebreaker Healy performed well on its 2001 Arctic Ocean
expedition to explore the Gakkel Ridge — the deepest and slowest-spreading ridge on Earth.
Southwest Indian Ridge in December
2000 and January 2001 revealed an area
several thousands of miles square where
there appeared to be almost no crust at all;
rather, the Earth's mantle rises up between
the diverging Antarctic and African plates
to spread directly onto the seafloor. There
is no layer cake here — just the mantle plate
beneath the seafloor.
Despite the ultraslow spreading and
apparent lack of volcanism, these same
cruises discovered areas with massive sul-
fide deposits of potential economic impor-
tance. These deposits are formed by miner-
als precipitating out of hydrothermal fluids
rising out ot the crust. (See "The Remark-
able Diversity of Seafloor Vents," page 60.)
This was a complete surprise to those who
believed such deposits would be found only
at faster-spreading ridges, where magma
rose more actively from the mantle.
In luly 2001, we ventured to the Arc-
tic aboard the new U.S. Coast Guard ice-
breaker/research vessel Healy. We made
the first detailed maps ot large portions
of the Gakkel Ridge, which extends from
north of Greenland almost to Siberia. It
is both the deepest ocean ridge, ranging
from 3 to 5 kilometers (1.8 to 3 miles)
deep, and the slowest-spreading, ranging
from one inch per year near Greenland
to half an inch per year at its eastern end
off Siberia.
Theory predicted that as seafloor
spreading slowed along the ridge, volca-
nism would wither and the ridge would
become essentially a crack in the planet
where solid mantle rock would be pulled
up by the spreading plates to form new
seafloor. We did, indeed, find mantle rock
rising in great solid slabs to form new sea-
floor; but we also found that isolated vol-
canoes persisted as far as we could survey
to the east. The generation of magmas in
the Earth proved tar more complicated
than anyone imagined!
Woods Hole Oceanographic Institutic
Gakkel Ridge Bathymetry
AMORE Cruise(Aug - Oct 2001) f
USCGC Healy 8 PFS Polarslern
Seabeam and Hydrosweep Data
Depth ikiiompiersi
o
o
<=
ID
01
in
01
at
c
Along the Gakkel Ridge, we not only
sampled more hydrothermal deposits, we
also detected abundant active hydrother-
mal venting in a region where current the-
ory predicted their absence. The discovery
offers the potential to find vent sites with
unique fauna that have evolved in isola-
tion from those in other oceans.
These discoveries have now led to the
realization that instead of two great classes
of ocean ridges — slow and fast — there is
a third category, ultraslow, which may
make up as much as one-third ot the
global ocean ridge system. These ultraslow
Icebreaker
ridges — so unlike the more explored and
better-known Atlantic and Pacific Ocean
ridges — represent a new frontier.
New territories and technologies
From all of this, it is clear that, despite
many decades ot seafloor study, we have
A NEWLY DISCOVERED TYPE OF MID-OCEAN RIDGE— Volcanic activity at mid-ocean ridges creates new seafloor crust that spreads outward
to cover 70 percent of Earth's surface. Recent expeditions have shown that instead of just two classes of ridges — fast-spreading and slow-
spreading — there is now a third, ultraslow. Ultraslow-spreading ridges, which may make up one-third of the global ocean ridge system, have
distinctive characteristics. Like other mid-ocean ridges, ultraslow ridges have areas where magma rises from the mantle and erupts at the seafloor
to create new ocean crust. But in between, there are also amagmatic zones, where solid slabs of mantle rock rise directly to the seafloor.
38 Ocean us Magazine- Vol. 42, No. 2- 2004 • oceanusmag.whoi.edu
The 200 7 AMORE (Arctic Mid-Ocean Ridge Expedition)
collected SeaBEAM sonar data to create this detailed 1,000-
kilometer (620-mile) bathymetric map of the Cakkel Ridge.
just begun to delineate the varied charac-
ter of ocean crust. Our recent discoveries
resulted from pushing current technolo-
gies to their limits, creating new ones, and
exploring uncharted territories.
The Ocean Drilling Program, for
example, recently drilled the first deep
hole into the lower ocean crust, in an area
where faulting has exposed a deep section
of the lower ocean crust and mantle. Cores
from this hole, nearly a mile deep, confirm
that the composition of the lower ocean
crust differs as one goes farther down.
But how significant is a single hole? The
answer lies in detailed mapping of large
areas of the seafloor around this drill site,
and at other tectonic windows into the
lower ocean crust.
Development of new tools and tech-
niques will bring further progress in
understanding the modified (or crumbled
or perhaps crazy) seafloor layer cake.
We are envisioning new, broader
approaches to map large areas of sea-
floor without an expensive ship in con-
stant attendance— easily deployable sea-
floor rock drills for collecting samples, for
example, and the marriage of new instru-
ments with autonomous vehicles. (See
"Realizing the Dreams of da Vinci and
Verne," page 20.)
With these approaches, marine geolo-
gists will create a series of touchstones at
specific sites across the ocean basins. By
extrapolating sections in between these
touchstone sites with remote geophysi-
cal sensing, we will be able to paint a fully
detailed picture of the skin that gives
Earth its unique complexion.
ULTRASLOW BUT VOLCANICALLY ACTIVE— Scientists had predicted that the Gakkel Ridge was
spreading far too slowly to promote volcanism, but an expedition in 200 1 found surprising
evidence for active volcanoes and hydrothermal vents.
Henry Dick /!<is been a Senior Scientist at Woods Hole since 1990. He
first became interested in geology as a boy when he found a rock collec-
tion in an outbuilding at his grandparents' home in Vancouver, Wash.
His great-grandfather was a geologist sent out west in the 1890s by his
uncle. Spencer Fullerton Baird, who also founded the U.S. Department
of Fisheries. For his Ph.D. at Yale, Dick backpacked more than 60 square
miles of the rugged Kalmiopsis Wilderness in southwestern Oregon,
mapping ancient ocean crust and mantle formed at a former island arc.
He came to Woods Hole to find out if rocks, at mid-ocean ridges were really different from those
he'd seen in the Oregon coast ranges and worked with Wilfred Bryan in the Geology and Geophys-
ics Department. The answer proved to be yes, and he had so much fun finding out, that he's stayed
around. Dick works almost exclusively on regions of the Earth with an average population density
of less than one person per thousand square miles. He is currently involved in the discovery of a new
class of ocean ridges found in the Arctic Ocean and around Antarctica with colleagues at WHOL the
University of Tulsa, and the Max Planck Institute in Germany. A lot of his time is spent fishing for
broken bits of the Earth's mantle found in great faults along the ocean ridges. He has a remarkably
dedicated wife named Winifred and three children, Helene, Spencer, and Lydia, who think their dad
goes to sea too much. He's promised them he won't even look at the ocean for at least a year, mud:
less get his feet wet — after his next cruise this fall!
Woods Hole Oceanographic Institution 39
Unraveling the Tapestry of Ocean Crust
Scientists follow a trail of clues to reveal the magmatic trickles and bursts that create the seaf loor
o
o
01
in
o>
re
2!
u
By Peter Kelemen, Senior Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Most people know that oceans cover
about 70 percent of Earth's surface.
Fewer people realize that the crust beneath
oceans and continents is fundamentally
different. Why this is so remains a mystery
that scientists are still trying to solve.
Oceanic crust is generally composed of
dark-colored rocks called basalt and gab-
bro. It is thinner and denser than continen-
tal crust, which is made ot light-colored
rocks called andesite and granite. The low
density of continental crust causes it to
"float" high atop the viscous mantle, form-
ing dry land. Conversely, dense oceanic-
crust does not "float" as high — forming
lower-lying ocean basins. As oceanic crust
cools, it becomes denser and ultimately
sinks back into the mantle under its own
weight after about 200 million years.
Earth's continental crust, on the other
hand, is up to 4 billion years old, and it
is thought to be the product of geologic
recycling processes far more complicated
than those that create ocean crust. If we
can decode and read the relatively sim-
ple story of how oceanic crust is formed,
we may someday be able to decipher the
more complex record of how the conti-
nents developed.
Sounding out seafloor structure
Because most oceanic crust is hid-
den from view beneath many kilome-
THE SEAFLOOR, ON LAND— In a few places on Earth, blocks of oceanic crust (called ophiolites)
have been thrust onto the continents, giving scientists the unusual chance to get a firsthand
look at rock formations that were once beneath the seafloor. The largest ophiolite is in Oman.
ters of water, our research must be con-
ducted "remotely," often using acoustic
techniques. Sound — emanating from an
earthquake, an explosion, or a relatively
benign source known as an airgun — trav-
els through different rocks at different
speeds. Geophysicists infer the basic geo-
logic structure of underlying rocks by
measuring the time it takes tor sound to
travel from one source to many different
receivers, or from many sources to a single
receiver. (See "Listening Closely to 'See'
into the Earth," page 16.)
In the oceans, this technique has
yielded a simple picture of a basaltic, lay-
ered crust about 7 kilometers (4.3 miles)
thick, underlain by the mantle. Rock sam-
ples obtained via dredging, submersible
operations, and drilling confirm that the
top of the oceanic crust, where it is not
obscured by sediments, is composed of
basaltic lava that originates in the mantle.
At the dawn of the modern theory of
plate tectonics in the 1960s, geologists
and geophysicists realized that the entire
oceanic crust was created from basaltic
lava along linear chains of seafloor vol-
canoes known as mid-ocean ridges, or
spreading ridges. Seafloor spreading car-
ries older oceanic crust away from the
ridges over tens of millions of years, until
it cools, becomes denser, and "tails" back
into the mantle in areas known as sub-
duction zones.
Seafloor clues in the desert
In a few places on Earth, blocks ot oce-
anic crust, called "ophiolites," have been
thrust, relatively intact, onto the conti-
nents during collisions between tectonic
40 Oceanus Magazine -Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
plates. Tilting and subsequent erosion
allow scientists to walk through a sec-
tion that once extended 25 kilometers (15
miles) into Earth's interior. The largest and
best exposed of these, the Oman ophiol-
ite near the Persian Gulf, comprises about
ten blocks that together cover roughly the
same area as Massachusetts.
The great extent of these ophiolites,
once deep beneath the seafloor but now
exposed, provides a comprehensive view
of the internal geometry of oceanic plates
that is unmatched by any sampling or
imaging technique at sea. Like pot shards
covered with hieroglyphics, ophiolites
open a window onto an ancient, largely
vanished world, and provide a rare avenue
for systematic investigation.
In the late 1960s and early 1970s, geol-
ogists and geophysicists observed simi-
larities between the layered structure of
oceanic crust, as interpreted from sound
velocities, and the layering in ophiol-
ites. A thin, upper layer in oceanic crust
(with low sound velocities) corresponds
to a layer of sediments and lava flows in
ophiolites. A deeper layer (with faster
sound velocities) corresponds to an ophio-
lite layer of "gabbro," which formed when
molten basalt solidified beneath Earth's
surface. In both oceanic crust and ophio-
lites, the gabbro layer is underlain by the
mantle, which extends thousands of kilo-
meters down to Earth's core.
A striking feature of well-exposed
ophiolites is a continuous layer of
"sheeted dikes" that lies between the lava
and the gabbro. These dikes are tabular
rock formations, about a meter wide, cre-
ated by periodic bursts of molten rock.
They stand side by side, like soldiers in
formation, each dike adjacent to neigh-
boring dikes, or sometimes leaning or
intruding into them.
This recurring structural pattern
occurs because all oceanic crust is newly
created at spreading mid-ocean ridges on
a kind of continuous conveyor belt: Each
dike, in a simple view, forms directly at the
center of a ridge. It then spreads out from
the ridge center, as another dike forms
WALKING ON THE OCEAN FLOOR— WHO! scientists Peter Kelemen (top arrow) and Greg Hirth
(50 meters directly below) walk on rocks that once were in the upper mantle below the seafloor.
In this photomosaic of a mountainside in Oman (and in photo on preceding page), light-colored
rocks (dunite) are ancient channels through which melt once flowed through the mantle.
behind it, in an ongoing process that cre-
ates the continuous layer observed in
ophiolites. Nothing like that happens in
continental crust, where new dikes more
randomly intrude older rock.
Going with the flow
During the 1970s and 1980s, geo-
physicists and geologists strove to under-
stand how basaltic lava forms beneath
spreading ridges. They theorized that
because the oceanic plates pull apart at
the surface, new material must rise to fill
the gap. As the material rises, the pres-
sure that helps keep it solid decreases.
This allows hot mantle rocks to partially
melt and produce basaltic liquid. This so-
called "melt" is less dense than surround-
ing solids, and so it buoyantly rises to the
surface to form the crust.
However, this theory raises as many
questions as it answers. From lava com-
positions, we know that from an enor-
mous volume of mantle rock, only small
amounts of rock partially melt to create
oceanic crust. Melt forms in micron-size
pores along the boundaries of innumer-
able crystal grains across a mantle region
that is 100 to 200 kilometers wide and
100 kilometers (61 miles) deep. From
this vast region, however, the melt some-
how is focused into only a 5-kilometer
(3 mile)-wide zone at a spreading ridge.
How is lava channeled from tiny pores
in a broad area ot melting into a narrow
area where it forms new oceanic crust
topped by massive lava flows?
My colleagues in exploring this mys-
tery, working in various combinations,
have included Greg Hirth, Nobu Shi-
mizu, and Jack Whitehead at Woods Hole
Oceanographic Institution (WHOI),
Marc Spiegelman of Lamont-Doherty
Earth Observatory, French geologists
Adolphe Nicolas and Francoise Boud-
ier, Massachusetts Institute of Technol-
ogy graduate student Vincent Sailers,
and MIT/WHOI Joint Program students
Einat Aharonov, Mike Braun, Ken Koga,
and Jun Kornaga. Our research has been
funded by the U.S. National Science
Foundation, the WHOI Interdisciplinary
and Independent Study Award program,
and the Adams Chair at WHOI.
Woods Hole Oceanographic Institution 41
o
o
in
0)
We have shown that melt travels
through the mantle in porous channels,
similar to channels filled with gravel that
provide permeable pathways through
clay-rich soil. Melt rising through the
hot mantle can partially dissolve miner-
als around them and gradually enlarge the
pores along the boundaries between indi-
vidual crystal grains. This, in turn, creates
a favorable pathway through which more
melt can flow— in a positive feedback loop
that spontaneously creates channels that
focus the flow.
Small channels formed in this fashion
coalesce to form larger channels, in a net-
work analogous to a river drainage system.
The number and size of melt flow chan-
nels we observe in the mantle section of
ophiolites support these theories.
Melt lenses and periodic bursts
New questions arose. If melt flows
through the mantle in micron-scale pores
along the boundaries of crystal grains,
where does it accumulate to form mas-
sive lava flows at spreading ridges? And, it
porous flow is a continuous, gradual pro-
cess, what causes the periodic bursts of
molten rock that create new dikes?
Once again, the Oman ophiolite pro-
vided clues. Nicolas and Boudier found
small formations of gabbro, called sills,
embedded in the shallowest mantle rocks.
Chemical analyses ot these sills indicated
that they crystallized from the same melt
that formed gabbro, sheeted dikes, and
lava flows in the crust. In addition, the
gabbro, dikes, and lava flows all had an
identical, distinctive pattern of alternating
bands of dark and light minerals.
It seemed to us that the entire gabbro
layer in the Oman ophiolite crust, from
uppermost mantle to the surface, could
have formed when melt material periodi-
cally collected in relatively small pools that
subsequently crystallized into solid "melt
lenses." Over time, a myriad of these melt
lenses accumulates — embedded within
each other and stacked atop each other or
side by side— to produce gabbro's rocky,
banded fabric.
Clogged pores build up pressure
Why would melt lenses first appear
in the uppermost mantle, immediately
beneath the base of the crust? We pro-
pose that such lenses form where melt,
approaching the seafloor, begins to cool.
Melt rising through the hot mantle can
dissolve minerals surrounding it to create
pore spaces, but cooling melt will begin to
crystallize and clog pores.
Two scenarios are possible: When the
supply of melt from below is low, con-
duits become narrower. The melt is forced
outward around impermeable barriers,
migrating via diffuse porous flow along
crystal grain boundaries throughout sur-
rounding rock.
But when melt supply is large, as
it is immediately beneath a spreading
ridge, buoyant melt accumulates beneath
impermeable barriers and creates excess
pressure. Eventually, the melt bursts
through the barriers and creates a melt-
tilled tracture that intrudes the over-
lying crust. It the tracture propagated
high enough in the crust, it would form
a sheeted dike, and if it reached even
higher, it would spill out onto the sea-
floor and feed a lava flow.
In this cycle of buildup and release,
minerals alternately crystallize and melt
under conditions of higher and lower
pressure. At relatively high pressure, much
less ot the light-colored mineral (plagio-
clase) is termed, compared with darker-
colored minerals. At lower pressure, the
proportion of plagioclase is larger. Thus,
periodic pressure changes result in the
light-and-dark banding observed in ophi-
olite gabbros.
Paths of most resistance
Working trom geological evidence
in ophiolites, together with physical and
chemical theory, we hypothesize that there
are two distinct ways to transport melt
that forms oceanic crust. Within the melt-
ing region in the mantle, melt can dissolve
minerals and create additional pore space.
As a result, continuous, high-porosity con-
duits form a coalescing drainage network
that focuses melt transport to the spread-
ing ridge.
At shallow levels beneath the ridge,
cooling melt begins to crystallize, clogging
pore space along crystal grain boundar-
ies. As a result, flow becomes diffuse, and
melt accumulates beneath impermeable
barriers. Pressure builds up until the melt
periodically bursts through overlying bar-
riers, and melt-filled fractures are injected
into overlying rocks to feed dikes and lava
flows. Together, these processes form a
highly organized system that consistently
produces new oceanic crust with a regular
structure along spreading ridges.
In our ongoing research, we are more
rigorously testing theories about how
porous conduits form in the mantle. We
seek to understand in more detail how
melt lenses form beneath spreading ridges.
And we want to figure out the factors
that determine why and when diking and
eruption events occur.
Peter Kelemen was at WHO/, us a postdoc and scientist, for 15 years.
i During this time he and his wife, Rachel Cox, acquired a house, two kids,
- and Rachel's Ph.D. in physiology from Boston University Medical School.
• During the first Jour of his six years as an undergraduate, Kelemen was
= an English and philosophy major at Dartmouth College. He then real-
5 i:ed he would need to get a job when he graduated. In the meantime,
he learned technical climbing techniques. He reasoned that it would be best to work outside in the
mountains, and so switched to a major in Earth sciences. In 1980, Peter and friends founded Dihe-
dral Exploration, a consulting company specializing in "extreme terrain mineral exploration." Until
1991, he split his time between geological research and mineral exploration, in the process obtain-
ing a Ph.D. from the University of Washington. As a mineral exploration consultant and research
scientist, Kelemen has been fortunate to work in the mountains of California, Oregon, Washington,
British Columbia, Peru, the Yukon, Alaska, the Indian Himalaya, and Karakorum Ranges, East
Greenland, along the Mid-Atlantic Ridge, and in the Oman ophiolite, where oceanic crust has been
thrust on land. In 2004. he became the Arthur D. Storke Professor at Columbia University.
42 Oceanus Magazine • Vol.42, No. 2 • 2004 • oceanusmag.whoi.edu
How is ocean crust made?
The crust beneath oceans and continents is fundamentally different.
Continental crust is made of light-colored rocks called andesite and
granite. Ocean crust is composed of dark-colored rocks called basalt
and gabbro. Ocean crust originates as a "melt" that forms in submicro-
scopic pores in rocks in Earth's hot mantle and rises to the surface.
Scientists have pieced together clues to discover: 1) how melt that
forms over hundreds of kilometers in the mantle is focused into a
five-kilometer volcanic zone beneath mid-ocean ridges, and 2) how
oceanic crust is formed with a relatively uniform, three-tiered structure
consisting of gabbro, sheeted dikes, and lava flows.
-« Scenario 1
When the supply of rising melt is low, it is
forced outward and around impermeable
barriers and trickles along tiny pore spaces
throughout surrounding rock.
Scenario 2 »•
When the supply of rising melt is large, it
accumulates beneath impermeable barriers.
Pressure builds until the melt bursts through
the barriers and creates a melt-filled fracture
that intrudes the overlying crust. If the
fracture propagates high enough in the crust,
it forms a sheeted dike. If it reaches even
higher, the melt spills over on the seafloor and
feeds a lava flow that solidifies into basalt.
~
V-
r levels beneath the ridge, the
egins to crystallize, clogging
ow channels and creating solid, impermeable
— = — TWo scenarios ensue (above).
3 Small channels coalesce to form larger
channels, in a network analogous to a river
drainage system, focusing melt toward a
mid-ocean ridge.
2 Rising melt partially dissolves minerals
around it, enlarging micron-scale channels
between mineral crystals and creating wider
pathways for additional flow.
.000005 meters
1 Hot mantle rocks partially liquefy. This "melt" is less
dense than surrounding solids and buoyantly rises.
Dunites
Segments of melt
channels break off,
solidify, and move
outward as the
seafloor spreads.
They create rock
formations called
dunites, often seen
in ophiolites .
Woods Hole Oceanographic Institution 43
o
o
01
I/I
01
Paving the Seaf loor — Brick by Brick
New vehicles and magnetic techniques reveal details of seaf loor lava flows
By Maurice A. Tivey, Associate Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Most ot Earth's crust is manufac-
tured at the bottom of the sea.
Deep beneath the waves and beyond our
view, magma erupts along a 40,000-mile
volcanic mountain chain that bisects
the ocean floors and encircles the globe.
The lava flowing from these mid-ocean
ridges solidifies into new ocean crust
that spreads out and paves the surface of
our planet.
That's the "big picture." But our abil-
ity to understand all the subtle and com-
plex details of this fundamental, planet-
shaping process is blocked by the oceans
themselves. Miles of water prevent us from
seeing the seafloor directly, and we can't
survive the darkness and high pressures at
the seafloor to explore it firsthand.
Exploring from afar
Imagine explorers from another planet
hovering in a spaceship high above a large
American city. From this lotty perspective,
the explorers would deduce that the urban
landscape was not natural, but constructed
somehow in particular ways.
To understand this landscape, they
would want to know: When were individ-
ual structures made? Of what materials? In
what sequence? How are the roads, build-
ings, and sidewalks erected? Why are they
located where they are? Which came first?
The extraterrestrials would have to zoom
down to get a closer look and perhaps
take some close-up photos and samples to
untangle the myriad factors that cumula-
tively result in cities.
For ocean scientists to zoom down and
take a closer look at the seafloor requires
specialized equipment. In the past decade,
we have taken big technological leaps.
New undersea vehicles have given us
unprecedented access to the seafloor and
new abilities to collect previously unat-
tainable data.
Our goal is to understand the seafloor
processes that turn molten lava into fresh
crust. In particular, we are developing an
44 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
innovative technique: measuring the mag-
netic properties of very young seafloor
rocks. It is revealing new details of how,
where, and when lava erupts, and how it
flows and accumulates on the ocean floor.
Telltale magnetic clues
As hot lava erupts, magnetic iron
oxide crystals (magnetite) within it orient
themselves to align with Earth's magnetic
field. When the lava cools and solidifies,
its magnetic crystals are "flash frozen" -
pointing north. The rocks' magnetic direc-
tion, or "polarity," is preserved.
But the magnetic north pole has not
always been where it is now. Throughout
Earth's 4.6 billion-year history, its mag-
netic field has flip-flopped several times —
with the magnetic north sometimes facing
south, or vice versa, as it is today.
The time periods when many of these
magnetic reversals occurred are well-doc-
umented. So seafloor lavas provide a built-
in chronometer or calendar that we can
use to determine when they were created.
Thus, the continually forming ocean crust
is kind of a tape recorder of Earth's mag-
netic field history.
Most recent lava flow
N Mid-
ocean ridge axis
TRACKING SEAFLOOR LAVA FLOWS — By superimposing magnetic measurements on detailed
seafloor topography maps like this one, scientists can distinguish how, when, and where
individual lava flows occurred on mid-ocean ridges. Younger lava has the highest magnetic
intensities (red and yellows). Above, the most recent lava flow erupted from the ridge axis,
overlaid older lava flows, and pooled to the left of the axis.
A magnetic mirror image
In the late 1960s, this phenomenon
provided crucial evidence confirming that
the seafloor was indeed spreading apart.
This concept is at the core of the revolu-
tionary theory of "plate tectonics."
In the early 1960s, scientists analyzed
the magnetic properties of rocks in ter-
restrial outcrops and developed a time-
scale that chronicled the reversals of
Earth's magnetic field. At sea, scientists
collected magnetic data from sensors
towed from ships and found remarkable
magnetic patterns in seafloor rocks that
were subsequently correlated with the
magnetic reversal "clock" established in
terrestrial rocks.
On both sides ot mid-ocean ridges,
scientists found a pattern ot alternating
magnetic "zebra stripes." "Black" stripes
represented rocks imprinted when Earth's
magnetic field was in a normal direc-
tion, as it is today, and "white" represented
rocks imprinted when the field was in a
Woods Hole Oceanographic Institution 45
o
o
Ot
c
Magnetic 'zebra stripes'
Seafloor lavas have built-in magnetic "clocks" that reveal their age. When seafloor lava
solidifies at the seafloor, its magnetic crystals are quenched in alignment with Earth's
magnetic field, and the rocks' magnetic "polarity" is preserved. But Earth's magnetic field
has reversed many times over the planet's history — with the magnetic north sometimes
facing south, or vice versa, as it is today.
Spreading ridge
I Normal
' magnetic
polarity
4 million
years ago
2 million
years ago
Present day
Million years ago
New seafloor is created at mid-ocean ridges (with the prevailing magnetic signature)
and spreads out in both directions, creating a symmetrical "zebra-stripe" pattern of
alternating rocks with either "reversed" or "normal" polarity.
reversed direction.
The stripes ran parallel to the ridges,
and the pattern was astonishingly symmet-
rical on either side of the ridges. This mir-
ror image could form only if new seafloor
was created at the crests of ridges (with the
prevailing magnetic signature) and then
spread outward in both directions.
A new way to measure 'young' rock
Historically, magnetic data were col-
lected with magnetometers towed at the
ocean surface by research ships. They pro-
vided a large-scale picture (like the view of
a city from a high-flying spaceship) of how
ocean crust forms over many millions of
years and hundreds ot miles.
To get a more detailed view, we began
to look at rocks less than 100,000 years old
and within just a few miles of the ridge
crest. This was problematic, because the
most recent magnetic reversal occurred
780,000 years ago. We had no way ot find-
ing the age of lava younger than this.
To solve this dilemma, we began to
examine not only the polarity of Earth's
magnetic field, but also the strength of the
field, or intensity. The intensity of Earth's
magnetic field has also varied dramatically
through time, and ocean floor sediments
have preserved a record ot Earth's mag-
netic field intensity over the past few
hundred thousand years.
So while magnetic polarity measure-
ments give us ages within millions to hun-
dreds of thousands of years, intensity mea-
surements give us ages within thousands
to tens ot thousands of years — a minute
hand on the magnetic clock. Thus, within
a patch of young lava with the same mag-
netic orientation, we can distinguish older
and younger rocks, and we can begin to
unravel the sequences in which they were
deposited on the seafloor.
Unprecedented seafloor access
To detect magnetic intensities, we
must take measurements within meters
of the rocks — something we just could
not do until recently. In 1993, we received
an opportunity to use a magnetometer
aboard the submersible Alvin, and we
confirmed for the first time that we could
detect strong magnetic intensity signals
in freshly erupted lava. The invention and
development at WHOI of the Autono-
mous Benthic Explorer (ABE) in the
mid-1990s made it possible to measure
magnetic intensities over wide swaths of
seafloor lava.
ABE can maintain a stable ride and a
constant altitude over changing seafloor
contours— features that make it well-
suited for collecting high-resolution mag-
netic measurements. Over several hours,
it can survey 20 to 25 kilometers ( 12 to
15 miles) of seafloor. At the same time,
its sonar can take measurements to create
fine-scaled topographical maps.
The maps give us detailed, three-
dimensional pictures of contorted amalga-
mations of lava flows. By superimposing
magnetic measurements on these maps,
we can distinguish the extent and volume
of individual lava flows in the upper crust
and tell when they erupted. We can begin
to unravel how flows are buried by sub-
sequent flows like a deck of cards. Alvin
plays a crucial complementary role, giving
us an essential visual picture of the sea-
scape and the ability to sample fresh lava
for precision dating.
46 Oceanus Magazine -Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
THE AUTONOMOUS BENTHIC EXPLORER fABEJ— Developed by Al Bradley, Dana Yoerger, and colleagues at WHO!, ABE can maintain a stable
ride and constant altitude over changing seafloor topography. It can make detailed seafloor maps and collect high-resolution magnetic data.
Many bursts or a 'crack of doom'?
My colleagues in this National Science
Foundation-sponsored research— Hans
Schouten, Dan Fornari, and Ken Sims at
WHOI, and Jeffrey Gee at Scripps Insti-
tution of Oceanography— focus on the
ocean bottom because seafloor lava flows
are young and not yet buried by thick
muds. They also have not lost their mag-
netic intensity due to chemical alteration
with seawater, so they accurately record
Earths recent magnetic field history.
The lavas found at mid-ocean ridges
erupt more often, are better-preserved,
and are less disturbed than those found
on continents. These attributes give us an
opportunity to get a more "brick-by-brick"
understanding of how the surface of Earth
is paved.
For example, does lava erupt at a cen-
tral "crack of doom" atop a ridge, and then
spill over and cascade downhill to bury
older lava flows (as portrayed in the right
side of the illustration on pages 44-45)?
Does it burst from several outlying magma
chambers through narrow channels to cre-
ate discrete patches of seafloor (as in the
left side of the illustration)? Or are both
processes at work?
Answers to these questions will delin-
eate the width of the zone along mid-
ocean ridges in which new seafloor crust
is formed. It will also reveal how fre-
quently and periodically ridge eruptions
take place— which, in turn, will tell us
about the heartbeat of magma movements
deeper in the Earth.
Maurice Tivey followed
his grandfather and fa-
ther to sea, as they both
5 served in the British
g Royal Navy. His marine
1 science career began with
5 geological studies at Dal-
^^ ^ housie University in Nova
6 Scotia and rock magne-
tism work at the University of Washington. He
came to WHOI in 1988 as a Postdoctoral Scholar.
Maurice has been involved in 32 research voyages
and made 27 dives in deep-sea submersibles in-
cluding Alvin, the French sub Nautile, and Japan's
Shinkai 6500. His research interests encompass all
of magnetism but especially focus on high-resolu-
tion magnetic measurements of the seafloor, and
what they can tell us about how the ocean crust
is formed and how the field has changed through
time. When at home, he plays soccer with friends
and coaches in the local youth league. He recently
bought a telescope to peer into the heavens, as an
excuse to get more gadgets and as c,n alternative
to continually looking under water.
Woods Hole Oceanographic Institution 47
£
re
t:
re
Earthshaking Events
New research on land and sea reinvigorates hopes of forecasting where earthquakes are likely to occur
By Jian Lin, Associate Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
When I was still a schoolboy in
China, two major earthquakes
occurred, about a year apart. They had a
profound impact on my lite and on the
Chinese people.
The first quake, with a magnitude of
7.3, struck the northern city of Haicheng
on Feb. 4, 1975. For six months before the
quake, a series of much smaller quakes
had rumbled in the region. They accel-
erated on Feb. 3, and in the early morn-
ing of Feb. 4, the government began to
issue warnings, shutting down factories
and urging people to remain outdoors.
Despite the frosty weather, people moved
into open fields, where children watched
movies instead of going to school. The
shock arrived at
7:36 p.m. While
more than one
million dwellings
were badly dam-
aged, the quake's Tjentsin0 ^
Gulf of Chihli
Magnitude 7.8 quake chincho,
July 28, 1976
Beijingo
^ Tangshan
1
death toll was relatively low at 1,328. The
foreshocks had offered a warning signal,
and this amazingly successful earthquake
prediction probably saved tens of thou-
sands of lives.
Seventeen months later, the people ot
Tangshan, a city 280 miles southwest of
Haicheng, were not
so lucky. On July
28, 1976, a mag-
nitude 7.8 earth-
quake released the
energy equivalent of
Magnitude 7.3 quake
February 4, 1975
A CITY FLATTENED — Without warning, a magnitude 7.8 earthquake struck the city of Tangshan in China, killing 242,769 people and destroying
90 percent of the city's buildings. A similar earthquake struck 1 7 months earlier about 280 miles northeast near Haicheng. But a series of smaller
foreshocks provoked Chinese officials to issue warnings urging people to remain outdoors, and only 1,328 people died.
48 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
ABOVE AND BELOW THE OCEAN— Faults on land, like the San Andreas Fault near Palmdale, Calif, (left), are similar to those in the ocean, like the
Atlantis Transform Fault in the Atlantic Ocean (right, three times vertical exaggeration). But in many ways, oceanic faults are easier to study and
offer the potential to discover fundamental aspects about earthquakes that are applicable to land.
400 Hiroshima atomic bombs combined.
There were no foreshocks, and therefore
no warning.
The quake struck at the worst time,
3:42 a.m., when most people were sleeping
in their beds. It lasted only about 90 sec-
onds, but about 90 percent of the houses
and buildings in Tangshan collapsed. A
total of 242,769 people died, and 169,851
were severely injured, according to an offi-
cial tally.
Bamboo and body bags
The Tangshan earthquake occurred
during a political era when China
shrouded itself from outside eyes. So the
devastation it wrought did not make head-
lines in the West. But it left an imprint on
me and other Chinese that is perhaps as
profound and indelible as September 1 1th
on the current generation.
My home province in southern China
harbors bamboo forests, and much of our
harvest was sent to Tangshan to construct
emergency shelters. I also recall vividly
that factories in my home city made a
huge quantity of large plastic bags to be
sent to Tangshan. So many people died,
those plastic bags were needed to bury
the dead.
Undoubtedly, these events encouraged
me to become a geophysicist and earth-
quake researcher— to seek to understand
the fundamental physics of earthquakes
and learn, for example, why foreshocks
preceded the quake in Haicheng, but not
in Tangshan. Underlying all the scientific
efforts of earthquake researchers are the
goals of forecasting earthquakes and sav-
ing lives.
In many ways, earthquake research is
like cancer research: They are scientific
challenges that offer huge potential ben-
efits to society, yet both have turned out to
be far more complex and intractable than
we thought and hoped they would be. But
scientists have made important advances
in cancer research, and I believe that new
seismological research on land and in the
oceans has us on the path to make similar
advances in seismology.
Location, location, location
We know, in general, where earthquakes
typically occur. They happen near faults,
or fractures in the Earths crust where rock
formations— driven by the inexorable
movements of Earths tectonic plates-
grind slowly against each other and build
RUPTURED LANDSCAPE — WHOI geophysicist Jian Lin stands near the plainly visible surface of
the Pleasant Valley Fault in Nevada, which caused a magnitude 7. 1 earthquake in 1 954.
Woods Hole Oceanographic Institution 49
I
-
•••'• > •-.,. • ••".•-;•'..•' '"' -^ "^. "*--v. ' '
5& im^?*
'
WHERE THE FAULTS LIE — A satellite image shows a portion of the San Andreas Fault and neighboring faults (white lines) in Southern California.
up stress. At some point, stress surmounts
friction, and the rocks slip suddenly, releas-
ing earthshaking seismic energy.
We have a general idea where major
faults are near Earth's tectonic plate bound-
aries. Many faults in California are named,
including the famous and highly visible
San Andreas Fault, which stretches from
north of San Francisco to northern Mexico.
But there are many much smaller and less
well-known faults that might prove even
more dangerous because they are closer to,
and even directly underneath, urban cen-
ters such as Los Angeles. In many places,
including right beneath Los Angeles, faults
don't break the surface. We have only vague
ideas about the locations ot these so-called
"blind faults," until surprising earthquakes
illuminate their existence.
Forecasting the timing of earthquakes
is more difficult. Our best estimates are
currently based on the general theory
that faults accumulate strain and slip on a
somewhat regular schedule. This isn't very
useful for very large faults and great earth-
quakes, which require centuries of strain
buildup before they rupture.
From Joshua Tree to Hector Mine
In our quest to forecast earthquakes,
scientists have searched tor any signs ot
precursors that might provide warning.
Even a few minutes warning would be use-
ful, providing enough time to stop trains,
shut down nuclear power plants, or turn
off the gas in your house, tor example.
Scientists have investigated changes
in groundwater levels and electromag-
netic fields and unusual animal behavior
before major quakes, but the evidence so
far, though intriguing, is otten complex
and not conclusive. Seismologists are also
examining changes in seismic patterns
that may foretell large quakes — an increase
in smaller earthquakes, as happened
before Haicheng, or perhaps a dying out
of smaller earthquakes immediately before
the Big One.
With colleagues Ross Stein of the U.S.
Geological Survey (USGS), Geoffrey King
of Institut de Physique du Globe in France,
and Andrew Freed of Purdue University,
I have explored an intriguing seismic pat-
tern in Southern California that we believe
can improve our capability to identify
areas that are more susceptible to future
earthquakes. Our research has been sup-
ported by the National Science Founda-
tion (NSF), the USGS, and the Southern
California Earthquake Center.
Since 1992, a sequence of four moder-
ate to major earthquakes has occurred in
the Mojave Desert — near Los Angeles, but
fortunately in sparsely populated, outly-
ing regions. In April 1992, a magnitude
6.1 quake struck near the town of Joshua
50 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Tree, on a small fault about 20 kilometers
east of the San Andreas Fault. Two months
later, on another fault about 30 kilometers
north, a magnitude 7.3 quake occurred in
Landers— followed just three hours later
by a magnitude 6.3 quake near the town of
Big Bear.
Do quakes "talk' to each other?
In the wake of these earthquakes,
we calculated how rocks shifted and
deformed. We constructed a three-dimen-
sional computer model that simulated
how stress was transferred throughout the
rocks in the regions, from the brittle upper
crust to the more fluid-like lower crust.
The model pointed to an increase in pres-
sure of 1 to 2 bars in the direction of the
town of Hector Mine, about 20 kilometers
northeast of Landers.
One bar is equivalent to the pressure
exerted by pushing your hands together.
Does strain 'creep' through the crust?
Barstow O
1999
Hector Mine quake
magnitude 7.1
San Andreas Fault
* O Palmdale
^^^
^^^L^
Mojave
Segment
(Last ruptured in1 857)
Los
Angeles
1992
Landers quake
magnitude 7.3
Big Bear quake \ yv
San Bernardino
Mountain Segment
(Last ruptured inl 81 2)
Coachella
Valley Segment
(Last ruptured inl 680)
1992
Joshua Tree quake
magnitude 6.1
Since 1 992, a sequence of four moderate to major earthquakes has occurred in sparsely
populated places in the Mojave Desert, near the towns of Joshua Tree, Landers, Big Bear,
and Hector Mine.
Computer models (right) simulated how
the Joshua Tree, Landers, Big Bear, and
Hector Mine quakes alleviated stress
in some places (blue areas) and
transferred and increased stress
(red areas) to adjacent regions.
In a domino-like effect, strain can
"creep" through the crust, interact
with neighboring faults, and trigger
another earthquake elsewhere. Scien-
tists believe that the Landers and Hector
Mine quake sequence has shifted stress toward the San Bernadino Mountain segment of
the San Andreas Fault and increased the likelihood of a quake there.
That's not much, but applied over a wide
area that is on the verge ot a quake, it adds
up. In 1999, a magnitude 7.1 quake hit Hec-
tor Mine. We believe the Landers and Big
Bear quakes transferred stress, through the
crust and the slowly creeping upper mantle,
toward the Hector Mine fault and hastened
its eventual failure seven years later.
Our research indicates that as an earth-
quake alleviates stress on one fault, it can
shift that stress to adjacent areas. In a dom-
ino-like effect, strain can "creep" through
the crust, interact with neighboring faults,
and trigger another earthquake elsewhere.
In this way, earthquakes carry on a
conversation with each other. If we listen
in, we can begin to understand and moni-
tor how one quake can influence the tim-
ing and location of earthquakes on faults
"down the line."
Subsequent research on fault systems
in Turkey and Japan has shown that these
concepts of "creeping" and "stress trig-
gering" may give us a valuable new tool
Woods Hole Oceanographic Institution 51
re
3
o-
C
a
Eavesdropping on oceanic quakes
In early May of 1 998, hydrophones began to record a flurry of small earthquakes near
the Siqueiros Transform Fault, an 80-mile fault sandwiched between two volcanic mid-
ocean ridges in the Pacific Ocean. The flurry culminated in a magnitude 5.8 quake on
May 7 0. This was followed by another flurry
of smaller quakes in an outlying area about HH£T ~5-Wtm£fUfr>-
20 kilometers away and another magnitude
5.8 quake only 18 hours after the first one.
New findings like this reinvigorate hope that
further understanding offoreshocks can
provide warning of larger quakes, at least in
some places on Earth.
to assess where earthquakes are more
likely to happen next. We believe, for
example, that the Landers and Hector
Mine quake sequence has combined to
shift stress toward the San Bernardino
Mountain segment of the San Andreas
Fault and increases the likelihood ot a
quake there.
Faults under the oceans
Simple explanations for earthquakes on
land have been difficult to unravel because
continental crust is old and usually has a
long, complex, and tortured history. It is
a conglomeration of many types of rocks
that have been distorted by erosion and
chemical "weathering" over tens of mil-
lions to hundreds of millions of years;
overwritten by many periodic bursts of
volcanism; and repeatedly fractured and
refractured by earthquakes.
These confounding complications
have encouraged seismologists to study
a simpler system that naturally cre-
ates earthquakes — on the bottom of the
ocean, where the majority ot Earth's tec-
tonic plates collide and where about
80 percent of seismic activity on Earth
takes place. Many of the plate borders
are marked by volcanic mountain chains
called mid-ocean ridges, where new
ocean crust is created. The mid-ocean
ridges are offset, in a zigzag pattern, by
great transform faults, which resemble
tanks on land.
The seafloor offers many advantages:
ocean crust that is pristine and com-
posed of only a few types of rocks; fault
systems that are uncomplicated and
undistorted; and far more seismic activ-
ity to observe. By studying seafloor sys-
tems, we have the potential to discover
fundamental aspects about earthquakes
that are applicable to land.
Searching for warning signals
Studying earthquakes in the oceans
presented its own difficulties in the past.
Beyond the expenses and logistics of sea-
going research, seismometers previously
could not stay down long enough to cap-
ture seismic events and could not record
a broad range ot seismic waves. But the
growing potential of long-term ocean
observatories, new generations ot seismic
instruments, and exponentially increased
computer power are dramatically increas-
ing our chances of making seismological
breakthroughs. (See "Seeding the Seafloor
with Observatories," page 28, and "Listening
Closely to 'See' into the Earth," page 16.)
Previous experiments by research-
ers have shown tantalizing evidence
of precursory "slow" quakes before a
major quake on oceanic faults. Instru-
: ments installed on land recorded a very
J slow creeping event 100 seconds before
| a magnitude 7.0 quake in 1994 on the
: Romanche Transform Fault in the equa-
: torial Atlantic Ocean, a fault very similar
5 to San Andreas. A similar experiment in
the Prince Edward Transform Fault in
; the Indian Ocean showed a slow creep-
ing event 15 seconds before a magnitude
6.8 earthquake.
In 2004, funded by NSF and the
WHOI Deep Ocean Exploration Insti-
tute, we analyzed seismic data collected
by the National Oceanic and Atmospheric
Administration (NOAA) since 1995 along
the Siqueiros Transform Fault, an 80-mile
fault sandwiched between two volcanic
mid-ocean ridges in the Pacific Ocean.
NOAA's seismic instruments are sus-
pended in the SOFAR (sound fixing and
range) channel, an exceptionally good
zone of sound transmission that is 700
to 1,500 meters beneath the sea surface.
Shaking from even a small quake in the
ocean crust can vibrate into the SOFAR
channel and be recorded by NOAA instru-
ments hundreds of kilometers away. (See
"Ears in the Ocean," page 54.)
The NOAA instruments recorded
many small quakes on the Siqueiros
Transform Fault. Trish Gregg, a graduate
52 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Quakes shift stress from place to place
Direction of stress
Fault ruptures
violently in an
earthquake
• Fault line
Direction of stress
Decreasing stress Increasing stress
A computer model simulated how stress on rocks is transferred in a region. At left, rocks on either side of a fault push against each other,
building up stress. At right, the stress overcomes friction, and the fault ruptures violently in an earthquake. In the aftermath, stress is
alleviated in some areas (blue) and transferred and increased to other neighboring areas (red).
student in the MIT/WHOI Joint Program,
Deborah Smith, a geophysicist at WHOI,
and I read the records carefully, and we
were surprised! Beginning in early May
of 1998, several small earthquakes began,
culminating in a magnitude 5.8 quake on
May 10.
As we did in California, we calculated
how this main shock changed the stress
field in the region. In the areas where we
calculated that the stress was transferred,
smaller quakes continued to occur. Then,
another flurry of smaller quakes began to
occur in an outlying area about 20 kilome-
ters away, followed by another magnitude
5.8 quake in that area only 18 hours after
the first one.
Flurries of foreshocks also appear to
have occurred prior to a number of other
moderate-size earthquakes in our study
region in the Pacific. These new findings
give us confidence that we can figure out
how foreshocks are created and that we
can calculate how and where one earth-
quake can transfer stress and trigger
another earthquake nearby. The research
reinvigorates hopes that at least some
earthquakes on Earth have relatively
high predictability.
A rising phoenix
Almost three decades after its devas-
tating earthquake, Tangshan is a flourish-
ing new city. Its population has increased
by more than 50 percent since 1976.
Tangshan was one of the first Chinese cit-
ies honored by the United Nations for its
hospitable residential environments. A
new city landmark — the Phoenix Hotel —
rises 112 meters into the sky and is spe-
cially designed for resisting shakes from
major earthquakes.
However, the people of Tangshan are
fully aware of the constant danger posed
by a very complex network ot faults in
northern China. Recently, I began to col-
laborate with colleagues to evaluate how
the Haicheng, Tangshan, and other mod-
erate-size quakes in the region may have
"talked" to each other. We hope to assess
more accurately seismic risks for the Bei-
jing-Tianjing-Tangshan triangle, the
true cultural and political heartland of
China — so that the future will bring us
fewer earthquakes without any warning
like Tanshan.
Jian Lin was born and raised in China's southern coastal city ofFuzhou,
not far from the quake-prone island of Taiwan. When he was small, his fa-
ther explained to him that the dangling light in their house went into a wild
f. swing because of a major quake in Taiwan. The mid-1970s was an unusu-
allv apprehensive era when the whole nation of China felt as if quakes could
occur anywhere at any time. Lin became a voluntary "earthquake watcher"
? in school. He kept a diary of water level clianges in an abandoned well and
?. the gentle tilting of the ground, phoning in readings to a local seismological
center. He studied physics and seismology at the University of Science and
-: Technology of China, conducted Ph.D. research at Brown University in Prov-
= idence, R.I., and pursued earthquake geophysics at the U.S. Geological Sur-
< vey in Menlo Park, Calif. At WHOI, he has investigated mid-ocean ridges,
hotspots, and underwater volcanoes, while continuing his work on quakes. The Institute for Scientific
Infcirmalicin recently recognized Lin as among the most cited authors on earthquake research in the
past decade. His paper with Geoffrey King (Institut de Physique du Globe) and Ross Stein (USGS) was
the most cited earthquake research article in the past decade.
Woods Hole Oceanographic Institution 53
Ears in the Ocean
Hydrophones reveal a whole lot of previously undetected seafloor shaking going on
c
IB
re
3
1
By Deborah Smith, Senior Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
It you sought to delve into the forces that
drive and shape the face of the Earth
and that distinguish it from all other plan-
ets in our solar system, you would shine a
spotlight on the mid-ocean ridges.
This 75,000-kilometer (45,000-mile)-
long volcanic mountain chain bisects
the seafloor and wraps around the entire
globe. It is the site where magma con-
tinuously erupts to create new crust. As
the crust spreads out on both sides of the
ridges, it paves the surface of the planet
and sets in motion the tectonic forces that
cause continents to rip apart and collide,
and oceans to open and close.
This planetary extravaganza, full ot
fury and sound, is accompanied by a con-
stant drumbeat of earthquakes ^^^^
and volcanic eruptions. But the
oceans act like a great blue cur-
tain, completely shrouding our
view and muting the sound.
About 80 percent of volca-
nic and earthquake activity on
Earth occurs on the seafloor, but
it is like those proverbial trees
that fall in forests when nobody's
there. If we could eavesdrop on
all that seismic activity, we could
glean a great deal of information
about the workings of our planet.
To do that, Earth scientists
have employed instruments that
record seismic waves generated
by earthquakes and volcanoes.
(See "Listening Closely to 'See'
into the Earth," page 16.) But
these studies have suffered from two pri-
mary limitations. The instruments have
been installed in land-based networks,
which cover broad areas over long time
periods but can detect only large-magni-
tude earthquakes in the oceans. Experi-
ments using ocean-bottom seismometers
can detect seafloor earthquakes with pre-
cision, but they are best-suited to monitor
small areas.
But now, the ending of the Cold War
has given Earth scientists an unprec-
edented opportunity to take advantage of
a tool created to wage that war — a project
launched by the U.S. Navy in the 1950s
that went by the code name Jezebel.
SOFAR, so good
The roots of Jezebel actually began at
Woods Hole Oceanographic Institution
SOFAR channel .
I Hydrophone
-1,000 meters
Mid-Ocean
Ridge
New hydrophone arrays have been developed to detect and monitor
previously "unheard" volcanic and earthquake activity in the oceans.
They are deployed in the SOFAR channel, a layer of water in the ocean
that channels sound waves generated by seismic events and allows
them to be transmitted and detected thousands of miles away.
during World War II when two scien-
tists, Maurice Ewing and J. Lamar Worzel,
began to conduct basic research on acous-
tics in the ocean— seeking any advantages
that would help the Navy detect enemy
submarines or mines, or help U.S. subs
avoid detection. In a critical experiment,
they detonated 1 pound of TNT under
water near the Bahamas and detected the
sound 2,000 miles away near West Africa.
The test confirmed Ewing's theory that
low-frequency sound waves were less eas-
ily scattered or absorbed by water and
could travel far without losing energy.
The key, however, was the discovery of a
layer of water in the ocean that acted like a
pipeline to channel low-frequency sound
and transmit it over vast distances. This
sound pipeline, called the SOFAR (Sound
Fixing and Ranging) channel, was inde-
^^^^ pendently discovered by Russian
scientist Leonid Brekhovshkikh,
working simultaneously with
underwater explosions in the
Sea of Japan.
The key to the SOFAR channel
is that sound energy, traveling in
waves, speeds up in waters where
..- temperatures are warmer (near
t the surface) or where pressure is
t higher (at depths). But in between
;' lies the SOFAR channel, which is
s bounded by water layers where
: sound velocities increase. The
i boundaries act like a ceiling and
floor: When sound energy enters
the channel from below, it slows
down. When it hits the ceiling, it
does not keep going, but rather,
it is refracted back downward.
54 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Hydrophone location
• Earthquake recorded by hydrophones
Earthquake recorded by land-based seismographs
SOUND AND FURY ON THE SEAFLOOR — An array of six hydrophones (yellow stars) recorded 7,785 seismic events (red dots) between 7 999 and
2002, mostly along the Mid-Atlantic Ridge and the Puerto Rico and Caribbean Trenches. A close-up of one portion of the ridge reveals that land-
based seismographs recorded far fewer sea floor seismic events (white triangles) over a much longer period (26 years).
Then it hits the floor, and it is refracted back
upward. In this way, sound is efficiently
channeled horizontally with minimal loss of
signal over thousands of kilometers.
The Navy immediately saw the value
of the SOFAR channel, launching feze-
bel, which later became the Sound Sur-
veillance System, or SOSUS. It deployed
a network of underwater microphones,
called hydrophones, connected by under-
sea cables to onshore facilities. The hydro-
phones were installed in locations to
optimally exploit the SOFAR channel.
The Navy could detect radiated acous-
tic energy of less than a watt at ranges of
several hundred kilometers— a sensitivity
that often could distinguish the number of
propellers a submarine had.
The tall of the Berlin Wall in 1989
marked the end of the Cold War, and in
1991, the Navy declassified SOSUS and
entertained the idea of allowing scientists
to use it for basic research. Scientists at the
National Oceanic and Atmospheric Admin-
istration's Pacific Marine Environmental
Laboratory (PMEL) in Seattle had the clever
idea of using the Northeast Pacific Ocean
SOSUS array to detect low-level seismic
waves on the Juan de Fuca Ridge off the
Pacific Northwest. The waves were gener-
ated by magmatic eruptions or rocks slip-
ping along faults to cause earthquakes.
But since SOSUS covered only a por-
tion of the world ocean, PMEL launched
a strategy to develop moored, autono-
mous hydrophones that could be deployed
for long durations in the SOFAR chan-
nel to monitor seismicity in more remote
ocean regions. In 1996, PMEL deployed
long-term hydrophones in the eastern
equatorial Pacific between 20°N and 20°S
and began monitoring them successfully.
That led me to ask if NOAA could deploy
similar hydrophones in my neck of the
woods — the North Atlantic Ocean.
With support from the National Sci-
ence Foundation, our research team —
myself, Christopher Fox and Haru Mat-
sumoto of NOAA PMEL, and Maya
Tolstoy of Lamont-Doherty Earth Obser-
vatory, and later Robert Dziak of NOAA
PMEL and Oregon State University —
deployed six autonomous hydrophones,
designed and built by NOAA's PMEL, in
the North Atlantic in 1999. In a sweet
coincidence, we used Lament's ship, R/V
Maurice Ewing, named after the SOFAR
channel's co-discoverer. We placed the six
hydrophones on the flanks of the Mid-
Atlantic Ridge between 15° and 35°N—
Woods Hole Oceanographic Institution 55
§
I
o-
n
three on each side, each spaced about 550
nautical miles apart.
Eavesdropping on the Earth
We selected this section of the Mid-
Atlantic Ridge because it has been studied
intensively over many years, beginning
with the famous FAMOUS (French-Amer-
ican Mid-Ocean Undersea Study) Proj-
ect in 1974 — the first expedition using
human-occupied vehicles to explore a
mid-ocean ridge — and more recently by
the French-American Atlantic Ridge proj-
ect, which extended from 1 5°N to 40°N.
The entire length of the ridge in this
region, as well as a few areas extending
onto the ridge flanks, has been surveyed
using multibeam sonar. This provides
detailed bathymetric maps that give sci-
entists the seafloor equivalent of the lay
ot the land. In addition, the area includes
several sites, such as the TAG hydrother-
mal vent field, where detailed multidisci-
plinary investigations are taking place.
All of these studies provide a frame-
work—a wide view and select close-up
images of the region — on which we now
can superimpose a soundtrack of seismic
events. By combining these data, we can
correlate the occurrences, locations, and
timing of various seismic events with the
seascape on which they are occurring.
We can begin to unravel, for example,
a whole series of questions and cause-and-
eftect relationships: How frequently does
magma erupt on mid-ocean ridges, and
in which locations? And how do these
temporal and spatial patterns influence
the geological landscape, and vice versa?
Where and when do different magnitudes
of earthquakes occur? Can they give us
clues to help us predict where and when
earthquakes will likely occur on land?
An unheard seismic symphony
The ability to attack all these questions
hinges on getting good hydrophone data.
In short, the array worked beautifully.
Between 1999 and 2002 the Mid-Atlan-
tic hydrophones recorded 7,785 seismic
events in the studv area — more than five
times as many as were recorded by seis-
mometers on land.
The hydrophones also detected seismic
events that release energy at levels 1,000
times less than events detected by land-
based seismometers (down to magnitude
2.5). These events, mostly small faults
moving, previously fell on the deaf ears of
land-based networks, which do not detect
seafloor seismic events smaller than about
magnitude 4.5.
This data set provides us with an
unparalleled view of the seismicity of
the ridge over a broad region and over a
broad range of event magnitudes. Joined
by new collaborators — DelWayne Boh-
nenstiehl of Lamont-Doherty and Javier
Escartin, Mathilde Cannat, and Sara
Bazin of the Institut de Physique du
Globe — we have begun to piece together
temporal and spatial patterns of seismic
events, which, in turn, can shed light on
fundamental planet-shaping processes
that occur at the ridges.
Stripes and gaps
Initial results show areas on the ridge
with intense and persistent seismic activ-
ity, which we call "stripes," that stand in
sharp contrast with areas that lack seis-
micity, which we call "gaps." Our hypoth-
esis is that the "stripes" correspond to
areas where the crust is thicker, colder,
and more brittle. In these places, stress
builds in rocks until a threshold is reached
and the rocks slip suddenly, causing earth-
quakes. The gaps represent areas where
the crust is hotter and thinner, and stress
is more easily accommodated by gradual
deformation in the rocks, perhaps produc-
ing earthquakes too small to be detected
by hydrophones. (See "Peering into the
Crystal Fabric ot Rocks," page 57.)
The gaps might also turn out to corre-
late with the presence in the crust of a type
of rock called serpentinite, which is more
slippery than other crustal rocks. It slides
rather than breaks, and therefore reduces
the probability of earthquakes. Clues like
these, discovered on the seafloor, could
apply to land and aid in the quest to fore-
cast where earthquakes are more likely to
occur. (See "Earthshaking Events," page 48.)
On a small scale, the distribution of
seismicity on mid-ocean ridges has impor-
tant implications for hydrothermal vents
and their biological communities because
magmatic eruptions initiate, maintain,
and destroy them. (See "The Evolutionary
Puzzle of Seafloor Life," page 78, and "Is
Life Thriving Deep Beneath the Seafloor?"
page 72.)
On a larger scale, we are analyzing seis-
mic patterns for clues that will more pre-
cisely reveal the directions in which three
of Earth's tectonic plates are moving. A
cruise in 2005 will investigate a telltale area
ot high seismicity that we discovered 70
kilometers west of the Mid- Atlantic Ridge
between 12.5°N and 13.5°N, which we hope
will reveal the location of the triple junction
where the South American, North Ameri-
can, and African Plates all meet.
Long-term hydrophone arrays have
given us our first glimpse of the robust
seismic activity rumbling on the seafloor.
As seismic monitoring continues and
expands into other oceans, we will accu-
mulate a treasure trove of new informa-
tion to explore the Earth.
Deborah Smith grew up in the cornfields of Illinois. She now
travels to the far reaches of land and sea to investigate how
volcanoes are built. She has hiked through tropical forests
in Hawaii and Tahiti and over the cold, barren terrain of
Iceland. Slic has mapped underwater volcanoes along a mid-
ocean ridge at the location where one can't be farther away
from land and still be on the planet: halfway between New
Zealand and South America. In an effort to share the experi-
ences of scientists with. the general public, she has developed
a Web site that profiles the careers of women oceanographers
(http://www.womenoceanographers.org).
56 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.ed
Peering into the Crystal Fabric of Rocks
When you get right down to it, earthquakes and volcanoes have atomic-scale causes
By Greg Hirth, Associate Scientist
Geology & Geophysics Department
Woods Hole Oceanographic Institution
ock solid" is an oxymoron, to my
L\vay of thinking. Oh, the expression
does have some truth in that minuscule,
superficial portion of our planet where
humans dwell. But the majority of rocks
nearly everywhere else in the Earth are
continually changing their physical char-
acteristics, lust below Earth's surface, rocks
are constantly subjected to heat that causes
them to flow like syrup, and to intense
stress that causes them to crack like glass.
On a planetary scale, the flowing and
breaking of rocks causes volcanism and
earthquakes, which pose hazards to people.
But on an even more fundamental scale,
the face of the Earth is continually shaped
and reshaped because rocks aren't rock
solid. They slide, crack, flow, and melt.
These abilities underlie larger processes by
which mountains and volcanoes are made,
continents are rifted, new seafloor is cre-
ated, and the great tectonic plates compris-
ing Earth's surface are set in motion to col-
lide or slide against each other.
It turns out that the same forces that
affect rock formations across continents
and ocean basins — heat and pressure-
also affect the deformation of rocks on the
microscopic scale. Rocks are composed
of a fabric of crystals. Just the way a solid
metal paper clip — when heat or force is
applied— can bend, break, or stretch, so,
too, can the crystal fabric within rocks.
The core of my research is to inves-
tigate the microstructures and micro-
mechanics of crystals within rocks. My
field of science, called rheology (from the
MAJOR CHANGES BEGIN MICROSCOPICALLY— Using optical and electron microscopes,
scientists can detect how crystals within rocks change their sizes, shapes, and orientations
when the rocks are subjected to heat and stress. These atomic-scale changes can ultimately
lead to earthquakes and volcanoes. Large crystals are evident in a photo of "undeformed"rock
from the Indian Ocean seafloor (top). But (bottom), the crystals are broken and ground up in
similar rock subjected to stress.
Woods Hole Oceanographic Institution 57
3
n
3
1
08
In this rock sample, heat has caused crystals to change shape and "flow" within the rock, which
accommodates strain without causing the crystals to break.
Greek word rheos, meaning current), is
the study of the flow of matter. Amazingly,
understanding what happens on an atomic
scale provides insights into what happens
on a planetary scale.
From crystals to earthquakes
My work is similar to that of mate-
rial scientists, who might examine the
microscopic textures of bricks or steel to
determine loads in house walls or fatigue
thresholds in airplane wings. My friends
jokingly call me a "rock psychologist,"
because I study the properties of rocks
under stress and strain.
Heat and stress cause chemical and
physical alterations and defects in the
crystal structure of rocks. Under stress,
atoms are pushed against other atoms,
sending further ripples of changes down
the line, like people crowding in subway
cars. Chemical bonds break. The rocks'
well-ordered lattice of atoms breaks down
and rearranges itself.
In short, crystals change their size,
shape, and orientation within rocks.
Using optical and electron microscopes,
we can detect these changes in rocks'
crystalline structure. The changes we
identify provide telltale information on
threshold temperatures or stress levels at
which the rocks' interior structure be^an
to crack, flow, or melt. Thus, rheology
holds the key to answering a fundamen-
tal question like "How deep into the crust
will earthquakes occur?"
Flowing rocks
Near the surface, rocks are unconsoli-
dated and porous. There's sufficient room
to accommodate any force applied. It's like
pushing a pile of sand. As a consequence,
large earthquakes are infrequent.
But deeper down in the Earth, there's
less excess room to accommodate strain.
The stress in rocks builds up. At a thresh-
old point, the stress surmounts the
strength of the rocks' crystal structure.
Bonds between atoms break suddenly. The
rocks crack and slide, releasing pent-up
energy. The result is an earthquake.
Deeper in the Earth, however, tempera-
tures rise. The heat encourages some— but
not all — of the atomic bonds within the
rocks' crystal lattice to break, freeing atoms
from their rigid atomic scaffolding to move
momentarily. The atoms reorient them-
selves in relation to neighboring atoms,
bonding with them in new alignments that
actually change the shape of the crystal.
This movement of atoms within rock
crystals (and the crystal shape-changing
it causes) accommodates the strain on the
rock, so that the stress never builds high
enough to cause the rock to bieak. At lower
temperatures, the same stress would shatter
all the atomic bonds at once. At higher tem-
peratures, the rocks "flow." At still higher
temperatures the rocks liquefy, or melt.
This phenomenon of flowing rocks
explains why earthquakes diminish deeper
in the Earth. Along plate boundaries,
such as the San Andreas Fault in Califor-
nia, earthquakes are restricted to depths
between 3 and 15 kilometers (2 to 9 miles).
In the lab and in the ocean
To test these theories, we conduct
experiments that subject various rock
samples to temperatures and pressures
they encounter within the Earth. By study-
ing the microstructure of the rock sam-
ples— before and after — we can map the
changes in crystal structure that accom-
modate different experimental strains, and
we can reveal much about the processes
that determine the rocks' strength and vul-
nerabilities.
But these experiments use samples on
the scale of one's pinkie finger and take
place over the course of a day. Do similar
microstructural changes occur in the "real"
Earth, over distances of tens of kilometers
and time spans of millions of years?
To confirm our laboratory experi-
ments, we investigated rocks collected
by the Ocean Drilling Program from the
Indian Ocean. With support from the
National Science Foundation (NSF), we
examined both undetormed rocks and the
same type of rock from deeper beneath
the surface, where it had been deformed
by heat and pressure.
These oceanic rocks are good models
for validating experimental analyses of
rocks because they are freshly made, with
no previous history of deformation, and
then they have been rapidly exposed to
cold seawater, which effectively quenches
them with their microstructure preserved.
Continental rocks, in contrast, are
old and have been subjected to repeated
shifts and traumas and exposed to high
temperatures over their long histories. All
these factors moditv their microstruc-
58 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
ture, often blurring and overprinting
records of their deformation.
We see similar microstructural
changes in oceanic rocks and in our exper-
imental samples. That gives us confidence
that the microscopic rock behavior and
mechanics observed in laboratory samples
can sharpen our understanding of rock
dynamics on a planetary scale.
Going against the crystal grain
With this confidence (and NSF sup-
port), we have applied a rheological
approach to elucidate details of many
other intriguing questions about the
Earth. For example, how is new ocean
crust created at mid-ocean ridges? This
process, which continually paves and
repaves Earth's surface and creates new
oceans and continents, starts on an atomic
level when crystals within rocks in the
mantle begin to melt.
How and when do rocks with differ-
ent types of crystals melt? How does this
melt dissolve the edges of other rock crys-
tals to create pore space in which it flows?
How do these microscopic rivers of mol-
ten rock coalesce, grow larger, and rise to
the surface to form new oceanic crust? All
these processes are ultimately controlled
by basic, atomic-scale physical and chemi-
cal properties of rocks, which I have been
investigating in collaboration with WHOI
colleagues Wenlu Zhu, Peter Kelemen, and
others. (See "Unraveling the Tapestry of
Ocean Crust," page 40.)
I have also been examining another
intriguing and important microscopic
phenomenon ot flowing rock. When
rocks' symmetrical crystal lattice breaks
down, crystals begin to deform and flow
in a particular direction. The rocks' crystal
lattice eventually realigns— not symmetri-
cally, but oriented in the particular direc-
tion that the crystals flowed in.
The phenomenon has important
implications for scientists trying to learn
more about Earth's interior using a pri-
mary tool: seismic waves. Earth scien-
tists deduce a great deal about the char-
acteristics of rocks within the Earth by
analyzing the speed of seismic waves
traveling through them. But seismic
waves will travel more slowly when they
go against the crystal alignment, and
faster when they travel in alignment with
the crystal structure.
A new generation of ocean-bottom
seismometers promises to lead to new dis-
coveries about the shrouded inner work-
ings ot the inaccessible mantle. (See "Lis-
tening Closely to 'See' into the Earth," page
16.) But a more detailed understanding of
mantle rheology will be critical to inter-
pret the new seismic data most precisely.
Life thrives between the cracks
More recently, we have also been apply-
ing rheological techniques to illuminate
how hydrothermal vent systems evolve.
These systems occur on mid-ocean ridges,
where magma from Earth's mantle rises
toward the surface and heats rocks below
the seafloor. (See "The Remarkable Diver-
sity of Seafloor Vents," page 60.)
Cold seawater percolates downward
through cracks in the ocean crust and is
heated up. The seawater reacts chemi-
cally with the rocks to create hot, mineral-
rich hydrothermal fluids that rise and are
vented at the seafloor, where they provide
nutrients that sustain thriving communi-
ties of deep-sea life.
Once again, this entire biogeochemical
cascade of events has a rheological foun-
dation—the cracks and conduits within
ocean crust that set the stage for the pro-
cess. We have been investigating the
micromechanics of cracking of peridotite
in the face of the hot and cold temperature
contrasts it experiences.
When peridotite cracks, it creates per-
meable pathways tor seawater to percolate
downward. The water chemically reacts
with peridotite to create another type of
rock, called serpentinite.
Serpentinite is a more slippery sub-
stance than peridotite, which may be an
important factor in causing fewer earth-
quakes along faults in the ocean. With
my WHOI colleagues Jian Lin and Jeff
McGuire, we are exploring how we can
use our knowledge ot earthquakes in the
oceans to increase our understanding ot
earthquakes on land, too. (See "Earthshak-
ing Events," page 48.)
In addition, the chemical reactions that
form the serpentinite from seawater may
provide critical conditions for the develop-
ment of some hydrothermal vents. Thus,
the micromechanics of fractures in rocks
and the evolution of permeability in oce-
anic crust will also play a key role in explor-
ing a new frontier: the potential for a deep
biosphere of microbial life beneath the sea-
floor. Microcracking continually creates
new mineral surfaces that are exposed to
fluids and microbes — creating the sites and
conditions for the biogeochemical reac-
tions that may sustain large communities
of microbes. (See "Is Life Thriving Deep
Beneath the Seafloor?" page 72.)
In the Earth sciences, microscopic cracks
can open entirely new vistas, and how mat-
ter flows turns out to matter quite a lot.
GregHirth spent much of his youth enthusiastically run-
ning around the woods of Ohio and the mountains of
Colorado. He went to Indiana University to study politi-
cal science with an idealistic goal of changing the role of
e politics in environmental affairs. But he quickly changed
_- his major to geology, motivated partly by opportunities
•e to work in the field. In his geology textbooks, Hirth read
£ about the research of his father, John, a material scientist.
\ That spurred his pursuit of a Ph.D. in rock deformation
at Brown University, where he met his wife, Ann Mulli-
gan (now an Assistant Scientist in the Marine Policy Center at WHOI). Hirth came to WHOI in 1993,
where he has pursued a myriad of research opportunities in the ocean, field, and lab (particularly with
colleagues at MIT). He is particularly grateful for his colleagues' encouragement to go back to his roots
and study geology in the field. Ann and Greg now enjoy gardening, following the fortunes of the Pa-
triots and Red Sox, and the wilds of Cape Cod, through which Hirth also leads trips for the Cape Cod
Bird Club.
Woods Hole Oceanographic Institution 59
The Remarkable Diversity of Seaf loor Vents
Continuing explorations reveal an increasing variety of hydrothermal systems
By Margaret Kingston Tivey, Associate Scientist
Marine Chemistry & Geochemistry Department
Woods Hole Oceanographic Institution
In late summer of 1984, 1 anxiously
awaited my first trip to the seafloor in
the submersible Alvin. There was a delay
in launching the sub, but I resisted the
urge to have a drink, anticipating one final
trip to the bathroom before crawling into
Alvin's three-person, 6-foot sphere for
eight hours. I was excited not only about
my first chance to dive, but about visiting
the home ot the seafloor rocks I had long
been studying for my master's thesis.
Since 1982, 1 had spent most of my
waking hours examining pieces of seafloor
vent deposits that had been recovered dur-
ing a routine dredging operation along
the Juan de Fuca Ridge off the Pacific
Northwest coast. Expecting to find com-
mon seafloor rocks called basalts, scien-
tists were surprised to pull up fragments
and boulders of massive sulfide covered
with small tubeworms. They had discov-
ered the fifth and, at the time, newest site
ot hydrothermal venting on the seafloor, a
place now known as the Main Endeavour
Field. These rocks helped launch my sci-
entific career.
In the years leading up to my 1984
dive, I had learned that hydrothermal vent
systems played a significant role in trans-
ferring heat and mass from the solid Earth
to the ocean, and that the vent sites host
unusual biological communities, including
tubeworms, bivalves, crabs, and fish that
thrive in the absence of sunlight. It was
also becoming clear that the relatively con-
stant chemistry of the ocean was in part
sustained by hydrothermal activity.
What my colleagues and I were only
beginning to realize then was that hydro-
thermal vent systems are like snowflakes—
no two are ever exactly alike.
Long-standing mysteries
When scientists in Alvin discovered the
first active hydrothermal system in 1977 at
the Galapagos Rift, they found warm fluids,
later determined to be a blend of cold sea-
water and hot vent fluids, seeping from sea-
>' •'
HEAT AND 'SMOKE' — When hot hydrothermal fluid jets from the seafloor and mixes with cold seawater, fine particles of dark metal sulfides
precipitate out of solution, creating the appearance of black "smoke."
60 Oceanus Magazine- Vol. 42, No. 2- 2004 • oceanusmag.whoi.edu
A seat loor observatory
In 2000-200 7, scientists collaborated on the most comprehensive study of a hydrothermal vent system to date, making
complementary and continuous observations at the Main Endeavour Field off the Pacific Northwest coast. This schematic depiction
shows the various instruments deployed and vehicles used (though the vehicles were not used simultaneously). They measured
currents, pressure, vent fluid temperatures and flow rates, chemical properties, and seafloor magnetic properties. The project, funded
by the National Science Foundation, is a model for future long-term seafloor observatories.
floor crust. Never-before-seen organisms
were present at the vents, including large
clams and tall, red-tipped tubeworms.
In 1979, a second active hydrother-
mal system was discovered along the East
Pacific Rise. At that site, much hotter flu-
ids (350°C) jetted from tall rock forma-
tions composed ot calcium sulfate (anhy-
drite) and metal sulfides. When the clear
hot fluid jetting from these chimney-like
structures mixed with cold seawater, fine
particles of dark metal sulfides precipi-
tated out of solution, creating the appear-
ance of black "smoke" (hence the name
"black smoker chimneys.")
The discovery of these vent systems
immediately answered a question long
posed by geophysicists: How is heat trans-
ferred from Earth's interior to the oceans?
Earlier studies had shown that, contrary to
model predictions, not as much heat was
being transferred by conduction (particle-
to-particle transfer) near the ridge crests.
Scientists hypothesized that heat was
also transferred by convection, as fluid cir-
culated within the crust near mid-ocean
ridges. Sure enough, cold seawater is
entering cracks and conduits within sea-
floor crust. It is being heated by underly-
ing rocks and rising and venting at the
seafloor, carrying significant amounts of
heat from Earth to ocean.
The chemistry of the ocean
The discovery of vents allowed sci-
entists to begin to answer another major
question: How does the ocean maintain its
relatively constant chemical composition?
Over time, rivers drain materials into the
oceans, and winds blow in particles, some
of which dissolve to add chemical ele-
ments to the oceans. Some of these ele-
ments, in turn, exit ocean waters, settling
in ocean sediments, for example.
Many components of seawater—
including lithium, potassium, rubidium,
cesium, manganese, iron, zinc, and cop-
per— enter the oceans via vents. They are
leached from seafloor crust by subterra-
nean chemical reactions with hot hydro-
thermal fluids. Other hydrothermal vent
reactions draw elements out of seawater
and place them back into the Earth.
For example, magnesium eroded from
Woods Hole Oceanographic Institution 61
The birth of a black smoker
The key to the creation of black smoker chimneys is an unusual chemical property of the
mineral anhydrite, or calcium sulfate (CaSO4). Unlike most minerals, anhydrite dissolves
at low temperatures and precipitates at high temperatures greater than ~150°C (302 °F).
1 350°C hydrothermal
fluid (containing
dissolved Ca2+ ions)
exits the seafloor and
mixes with coldseawater
(containing dissolved
Ca2+ and 5O42' ions).
A ring of anhydrite
precipitates around a
vent opening and the jet
of hot fluid.
2 Hydrothermal fluid
and seawater mix
through the nascent
anhydrite wall around
the vent opening. Sulfide
and sulfate minerals
precipitate within pore
spaces of the wall, which
gradually becomes
less permeable. The
anhydrite wall also
provides a substrate, or
foothold, on which other
minerals can precipitate.
3 Different minerals
precipitate at different
temperatures.
Chalcopyrite is
deposited against
the inner wall, which
is adjacent to 3SO°C
hydrothermal fluid.
Anhydrite, iron, copper-
iron, and zinc-sulfide
minerals (such as pyrite,
marcasite, bornite,
sphalerite, wurtzite)
precipitate at lower
temperatures within
pore spaces of the
chimney wall.
land is carried to the ocean by rivers.
Yet magnesium concentrations have not
increased in the oceans. Scientists puzzled
for decades over where all the magnesium
could be going.
Scrutinizing hydrothermal vents,
researchers found that seawater entering
seafloor crust is rich in magnesium, but
fluids exiting the vent are free of it. Ocean-
ographers surmised that magnesium is left
behind in the crust, deposited in clay min-
erals as seawater reacts chemically with
hot rock.
Dive to Main Endeavour Field
In those early years when observations
were few and samples fewer, my thesis
rocks provoked considerable interest. In
some ways, the rocks were similar to those
recovered from the East Pacific Rise, but
in other ways they were quite different.
The Main Endeavour Field samples
were rich with amorphous silica, which
should precipitate only if the hydrother-
mal fluid had cooled without mixing with
seawater. And they did not contain anhy-
drite, a common vent chimney mineral
that dissolves in seawater at temperatures
less than about 150°C (302°F).
We theorized that the dredged pieces
must have come from the low-lying mounds
that lay beneath and around black smoker
chimneys. Our dives to the Endeavour Field
in 1984 would tell us if we were correct.
The trip to the seafloor took 90 min-
utes. As we approached the seafloor, the
pilot asked me to look out my viewport and
let him know when I saw bottom. A/vin s
lights were turned on. It was like the cur-
tain going up in a dark theater and the stage
lights going on. We were hovering over the
same basaltic rocks that I had spent count-
less hours studying in photographs.
Then to my right, I could see a rock
wall rising from the seafloor. It was obvi-
ous from the hedges of pencil-diameter
tubeworms sticking out of the tops and
sides of the cliffs that I was looking at large
hydrothermal vent structures. The view
was nothing like what had been described
at the East Pacific Rise. Instead of low-
62 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Hydrothermal Circulation
Seawater percolates through permeable seafloor crust, beginning a complex circulation process.The seawater
undergoes a series of chemical reactions with subsurface rocks at various temperatures and eventually changes
into hydrothermal fluids that vent at the seafloor.
1 0 Vent plumes disperse heat
and chemicals from hydrothermal
systems and may also disperse
vent fauna larvae. The plumes can
be used to locate unknown
vent sites.
1 Seawater percolates into
permeable ocean crust.
9 Chemical exchanges between
seawater and plume particles
further regulate ocean chemistry.
Phosphate, vanadium, and arsenic
from seawater are "scavenged"
by particles and sink to the seafloor. I
ral Particles
Black Smoker-
Metal Sulf ide Deposit \
8 Seawater is entrained into buoy-
ant plumes. The resulting mix (one
part hydrothermal fluid to 10,000
parts seawater) rises.
/ Fluids venting rapidly through
black smoker chimneys confront
cold seawater, causing metal
particles to precipitate and forming
dark, particle-laden plumes.
White
£. Circulating sea water reacts
chemically with volcanic ocean
crust rock. The reactions greatly
alter the fluids and the rocks.
3 Clay and sulfate minerals
precipitate from seawater as it
is initially heated. The resulting
fluid loses most or all of its
magnesium and sulfate.
D Fluids venting at the seafloor
mix with seawater, forming metal-
rich deposits similar to ore deposits
found on land. The chemical-rich
fluids support complex ecosystems
on and beneath the seafloor.
5 Hot, buoyant hydrothermal
fluids rise toward the surface.
4 Higher temperatures at
deeper depths trigger chemical
reactions that transfer metals,
silica, and sulfide from rocks to
fluids — resulting in hot, acidic
fluids containing abundant silica,!
metals, hydrogen sulfide,
hydrogen, and methane.
Source
Woods Hole Oceanographic Institution 63
Cross section of a black smoker chimney
As hydrothermal fluid vents from the seafloor and interacts with seawater, different
minerals precipitate at different temperatures, creating a multi-layered chimney.
Outer layers contain a mixture of
minerals that precipitate at lower
temperatures, including anhydrite
(CaSOj), which precipitates at
150°C or higher temperatures and
contains sulfate from seawater),
pyrite (FeS2), and
other fine-
grained
metal
sulfides.
Iron sulfides on the outer wall are exposed
to oxygen in seawater and are oxidized to
form iron oxides.
The initial chimney structure restricts mixing
with cold seawater. Chalcopyrite (CuFeS2),
which precipitates at >300°C, is deposited on
the inner walls.
A final inner coating of
sphalerite (ZnS), which
precipitates at ~300°C,
can form as hydrothermal
fluids cool.
lying mounds of sulfide debris topped by
active smokers, we saw steep-sided struc-
tures standing 15 meters (50 feet) high.
Why was this site so very different? How
could these chimneys stand there like multi-
story buildings without collapsing?
A fluid environment
Answers to these questions came from
studying new samples. We learned that the
tall chimneys structures were essentially
"cemented" — silica had filled pores in the
vent structure walls as the emerging flu-
ids cooled, making the structures sturdy.
But why was so much silica precipitating
at this site?
The fluid chemistry provided answers:
This vent site's fluids were rich in ammo-
nia (NH3). As the ammonia-rich fluids
cool, NH3 takes up excess H+ ions to form
NH4, which raises the pH of the fluids.
The higher pH likely allows silica to pre-
cipitate within Main Endeavour Field
structures; at sites with no NH3, low pH
probably inhibits the formation of amor-
phous silica.
As with almost every visit to a new
vent site, our survey of the Main Endea-
vour Field raised as many questions as it
answered. The differences among known
hydrothermal systems and the revelations
that accompanied each new discovery pro-
voked oceanographers to hunt for new
sites. We were explorers trying to learn
as much as we could. Hypotheses were
advanced, only to be proven wrong by yet
another discovery.
More visits to these seafloor hot springs
made it clear that all vent fluids are not the
same. Rather, the chemical composition
changes from ridge to ridge — and from
time to time.
Researchers returning to some vents
found that the chemistry of vent fluids was
not constant, changing on scales ranging
from days to years. These vent sites were
all associated with sites of recent mag-
matic activity, with recorded earthquakes
and evidence that dikes had been intruded
into the ocean crust and, at some sites,
that lava had been extruded onto the sea-
floor. The vent fluid compositions were
changing as these dikes cooled, and as flu-
ids penetrated deeper into the crust.
Searching for new vents
In the early 1980s, after a number of
vent sites had been found in the Pacific,
scientists began to wonder if hydrothermal
activity and active black smokers might
exist on the more slowly spreading Mid-
Atlantic Ridge. The hydrothermal vent
systems transfer large amounts of heat
from magma or newly solidified hot rock,
but on slow-spreading ridges the spread-
ing rate, and magma delivery rate, are
much less (about one-third) of those on
the northern East Pacific Rise and Juan de
Fuca Ridge.
To our surprise, exploration from the
mid-1980s through the early 1990s gradu-
ally made it clear that hydrothermal sys-
tems may be spaced farther apart on the
Mid-Atlantic Ridge, but they tend to gen-
erate much larger mineral deposits. Expe-
ditions to the Southwest Indian Ridge
in 2000 and the Gakkel Ridge (under
the Arctic Ocean) in 2001 revealed that
hydrothermal venting occurs on even the
slowest-spreading portions of the mid-
ocean ridge system.
New ways to find vents
Two decades of study have taught us
that there is no single type of seafloor
hydrothermal vent system. The plumb-
ing systems beneath the seafloor are both
diverse and incredibly complex.
Ocean scientists today are posing ques-
64 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
tions about the dimensions and evolution
of the hydrologic systems beneath vent
sites. We puzzle over how hot these fluids
get, how deep into the crust they descend,
and how tar they travel before venting at
the seafloor. And where does seawater
enter these systems?
To answer these questions, we will need
to continue exploring, not only over geo-
graphic space, but also over time. In the
early years, most vents were discovered
serendipitously; but as we've explored and
learned more about these systems, we've
been able to develop systematic methods
tor pinpointing sites.
For example, a technique ot" "tow-
yowing" has been developed, in which a
conductivity-temperature-depth (CTD)
sensor is raised and lowered through
the water column in a sawtooth pattern
above the ridge to map the locations of
plumes, and then to home in on and map
the buoyant portion of plumes coming
directly from active vent sites. This tech-
nique was used successfully to find vent
sites in the Pacific and Atlantic, and most
recently in the Indian Ocean.
Seafloor observatories
The need to explore the dynamics
of hydrothermal systems over time has
led to new technologies and the devel-
opment of seafloor observatories. New,
more precise and durable instruments
allow us to monitor temperature and
fluid chemistry at vent sites for hours,
days, or months— as opposed to observ-
ing those properties for brief moments
and grabbing one-time samples.
The future of hydrothermal studies
was displayed in a recent series of coordi-
nated experiments. With support from the
National Science Foundation's Ridge Inter-
disciplinary Global Experiments (RIDGE)
program, a team of researchers built a sea-
floor observatory on the Endeavour seg-
ment of the Juan de Fuca Ridge.
During the summers of 2000 and 2001,
scientists made complementary and con-
tinuous observations centered around
the Main Endeavour Field (the same site
I first visited in 1984) and at vent sites to
the north and south. The program goals
included making more accurate measure-
ments of the heat and mass flowing from
the system, and observing how the hydro-
thermal plumbing is influenced by tides
and by high-temperature reactions that
separate elements into saltier liquids and
more vapor-rich fluids (a process called
"phase separation").
Instruments were deployed to con-
tinuously monitor vent fluid tempera-
tures, flow rates, and chemical proper-
ties. Scientists also used newly developed
samplers to collect fluids at regular time
intervals. While these instruments were
in place, other researchers made acous-
tic images of vent structures and venting
fluids. Still others used the Autonomous
Benthic Explorer (ABE) to measure sea-
floor depths, magnetic signatures, and
water column properties above
vent fields.
Later in the program, the team
deployed a systematic array of current
meters, thermistor strings, magnetom-
eters, and tilt meters. Scientists even tested
techniques to "eavesdrop" on the data
being collected and download it without
removing the instrument from the vent.
The result of this collective effort was the
most comprehensive study ot a hydrother-
mal system to date, and a model for future
seafloor observatories.
A continually unfolding story
As we develop these new techniques
and instruments, our ability to explore
ongoing seafloor processes will grow.
More than a quarter-century into our
studies, we still find ourselves constantly
revising and refining our ideas about
hydrothermal systems.
At the same time, as we home in on the
fine details of how these systems work, we
continue to find new sites that completely
break the mold. As recently as Decem-
ber 2000, researchers diving in the Mid-
Atlantic discovered "Lost City," a vent site
located far away from the ridge axis, on
old rather than nascent seafloor crust,
and with 15-story-high white minaret-like
structures made of carbonate — a mineral
that is not found at most other known
vent sites.
So after 32 dives to the seafloor to
study vents, I am often still surprised, and
I am always awed. Even when I return to a
vent site that I've visited before, I still find
it an unbelievably beautiful sight to watch
jets of hot fluid mixing with seawater, and
unusual organisms that make their homes
near these vents. Like my colleagues, I
look for ways to make our studies more
precise, more methodical, and more con-
tinuous. But 20 years after my first dive, I
still enjoy seeing it all live. It's the differ-
ence between watching a movie of a water-
fall and standing next to one.
Margaret Kingston (Meg) Tivey, an aqueous geochemist, combines labora-
tory, theoretical and field studies to examine the formation of seafloor vent
0 deposits and the transfer of energy and mass through active hydrothermal
1 systems. Born and raised in Lexington, Mass., she majored in geology
at Stanford University (after taking a course with five field trips to local
i beaches and fault zones). As a graduate student at the University of Wash-
'3 ington, she was first introduced to seafloor hydrothermal systems when
1 massive sulfide samples with live tubeworms were recovered while dredg-
- ing along the Juan de Fuca Ridge— only the fifth discovery of an active
mid-ocean ridge hydrothermal system. Working on these samples, she received an M.S. in geological
oceanography in 1983. In 1984 she had her first opportunity to dive in Alvin, and shifted her disser-
tation to develop a theoretical model to link measured vent fluid compositions to observed mineral
assemblages in vent deposits. After receiving her Ph.D. in 19SS, she came to WHO/, where she is
now an Associate Scientist, studying active vent sites throughout the world using both human-occu-
pied ••ubmersibles and remotely operated vehicles. Her studies are aided by X-ray computed tomogra-
phy (CAT scans) to examine samples of mineral deposits in three dimensions; development and use
of numerical models to link vent fluid compositions to vent deposit mineralogy; and collaboration
with engineers to develop, build, and use instruments capable of making measurements in the high-
temperature and low-pH conditions present at vent sites.
Woods Hole Oceanographic Institution 65
When Seaf loor Meets Ocean, the Chemistry Is Amazing
In more and more places, scientists are finding vast amounts of natural gas on the ocean bottom
By Jean K. Whelan, Senior Research Specialist
Marine Chemistry and Geochemistry Dept.
Woods Hole Oceanographic Institution
Far more natural gas is sequestered on
the seafloor — or leaking from it — than
can be drilled from all the existing wells
on Earth. The ocean floor is teeming
with methane, the same gas that fuels our
homes and our economy.
In more and more locations through-
out the world's oceans, scientists are find-
ing methane percolating through the sea-
floor, bubbling into the water column,
collecting in pockets beneath seafloor sed-
iments, or solidifying in a peculiar icelike
substance, called methane hydrate, in the
cold, pressurized depths of the ocean.
Massive deposits of methane hydrates
could prove to be abundant reservoirs of
fuel. But in the past, these massive store-
houses of methane also may have "thawed"
suddenly and catastrophically, releasing
great quantities of climate-altering green-
house gas back into the atmosphere.
In some places, seeping methane sus-
tains thriving communities of exotic
organisms that harness the gas as an
energy source in their sunless environ-
ment. Below the seafloor, an unknown
but potentially vast biosphere of microbes
A BUBBLING, LIFE-SUSTAINING BREW — Evidence is steadily growing that methane seeping and bubbling from the seafloor is a widespread, but
previously overlooked, natural phenomenon. It can sustain communities of seafloor life, like these mussels in the Gulf of Mexico.
66 Oceanus Magazine- Vol. 42, No. 2 • 2004 • oceanusmag.whoi.ed
^ **^^<*^*3
FUEi. FROM THE DEPTHS? — Methane, the same natural gas that we use as fuel, solidifies in the cold, pressurized depths. It is encapsulated by
frozen water to form an icelike substance called methane hydrate, which could prove to be an abundant source of energy in the future.
may be making the methane that perco-
lates upward. (See "Is Life Thriving Deep
Beneath the Seafloor?' page 72.)
Other places on the seafloor show evi-
dence that pockets of gas trapped beneath
sediments have exploded to form "mud
volcanoes," or may have triggered seafloor
avalanches and tsunami waves.
An underestimated phenomenon
Until recently, scientists have largely
overlooked seafloor methane and its
potentially dramatic impacts. The prob-
Photos in this article: Copyright Woods Hole
Oceanographic Institution and the BBC Nat-
ural History Unit, courtesy of the WHOI Ad-
vanced Imaging and Visualization Laboratory
and Johnson-Sea-Link submersible, Harbor
Branch Oceanographic Institution.
lem is that methane commonly vents out
of isolated cracks in the seafloor — some so
small that they are easily missed by oce-
anic surveillance systems. Once out into
the ocean, the methane usually is diluted
rapidly by seawater, or it dissolves in sea-
water and is consumed by microorganisms
that convert it metabolically into carbon
dioxide. Unless you happen to be looking
in the right place at the right time, you'll
miss the show.
But evidence has steadily accumulated
that natural seepage of methane from the
seafloor is a large, continuous, and ubiqui-
tous phenomenon. When oceanographers
happen upon these vents (often called "cold
seeps"), the scene is often spectacular.
Several researchers have documented
large craters or pockmarks on the seafloor,
while others have described huge carbon-
ate mounds (formed by organisms that
ingest methane and produce carbonate).
Both are often relics of past seafloor gas
venting. Sometimes gas simply seeps from
the ocean floor and sustains communities
of unusual tubeworms, mussels, and other
creatures like those found at hydrothermal
vents. (See "The Evolutionary Puzzle of
Seafloor Lite," page 78.)
Gas frozen solid at the seafloor
The deep ocean floor around gas seep
sites is often covered by methane hydrates.
These are solid crystals of methane encap-
sulated in ice, which form under the low
temperatures and high pressures typical of
ocean depths greater than about 1,500 feet.
These hydrates look like seafloor car-
Woods Hole Oceanographic Institution 67
The seafloor is teeming with methane
Methane gas
Methane bubbles usually burst and cfeso/v
and are biodegraded by microorganisms.
Dissolved
methane
Deposits of solid methane
encapsulated in ice, called
methane hydrate, often
form at low temperatures
and high pressures at or
below the seafloor.
._ __, , s of methane bubble _.. ._
ch the surface. They can vent methane, a greenhouse gas,
o the atmosphere. They also create natural oil slicks, which
are now being used to locate new oil deposits.
n water
j|yt~*u~_ „„:„,- *„ *l J/ .._j._: ^.l__:..:_
communities of exotic animals in the sunless dep
Bacterial
mat and
gas hydrates
Gas
rdrate
Gas ,,k ,
hydrate
leakaoe
' Oil & gas
reservoir ..«•* \ ?
S
Methane gas flows upward through faults \
2 and cracks in sediments, sometimes leaking to
the seafloor or forming trapped pockets
of gas reservoirs.
reservoir
Faults x
& fractures
Oil and gas
Methane gas is created naturally under deep-sea sediments and in Earth's crust
from organic matter subjected to heat and pressure, or by bacteria producing it
as a metabolic end-product.
Scientists are discovering that abundant quantities of methane gas are continually seeping from the seafloor throughout the oceans. This wide-
spread but overlooked natural phenomenon has potentially dramatic implications on world energy supplies, life in the oceans, and Earth's climate.
68 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Methane gas
o *•
• o
trapped beneath
sediments can
build up pressure
until they explode
) to form "mud-
volcanoes. "
On continental
slopes, these may
trigger seafloor
avalanches
and tsunamis.
FED BY METHANE, LIFE FLOURISHES— Methane seeping from the seafloor sustains microbes
that serve as the base of the food chain for communities of animals, like these tubeworms,
which thrive in the sunless depths in the Gulf of Mexico.
bonate, but when chunks are broken off,
methane hydrates float upward (carbon-
ates sink). As those hydrates rise into
higher temperatures and lower pressures,
they decompose, releasing methane gas
into the ocean — a process akin to releasing
the pressure on a bottle of soda.
Energy companies have been eye-
ing methane hydrates as a potentially
tremendous new source of natural gas.
Since the 1930s, the use of natural gas has
increased fivefold to account for more
than 25 percent of the world's energy
consumption. With existing technol-
ogy, the world gas supply is estimated
to be 5,300 trillion cubic feet (tcf), Rob-
ert Kleinberg of Schlumberger and Peter
Brewer of Monterey Bay Aquarium
Research Institute reported in American
Scientist. At the current rate of global
consumption (about 85 tcf per year), a
60-year supply remains.
But the amount of gas at various loca-
tions around the world varies widely.
Russia and the Persian Gulf each have
about 1,700 tcf, while the total for North
America is about 260 tcf. Japan and
Europe import nearly all of their natu-
ral gas, while India and China have very
small domestic reserves.
A potential new energy resource
The untapped well of methane
hydrates holds the promise of energy
independence tor nations close to oceans
or permatrost regions (where condi-
tions and consistently cold temperatures
also create methane hydrates). Offshore
methane hydrates would provide the
U.S. alone an estimated potential natural
gas reserve of 300,000 tcf. Projections of
hydrate gas reserves in the ocean south
of Japan are 2,000 times that country's
very small existing natural gas reserves,
according to Kleinberg and Brewer.
Most of the world's gas hydrates are
sequestered in the deep ocean, presenting
great challenges for potential commer-
cial production. Hydrates dissolve quickly
when removed from the unique condi-
tions on the ocean bottom, so research-
ers must figure out how to either stabilize
them or produce and transfer fuel directly
from the seafloor.
Many known deep-water deposits, such
as the Blake- Bahamas Plateau off the Caro-
linas, are very diluted or spread across rela-
tively thin layers over wide areas, making
them very difficult to "mine" economically.
And deep-sea hydrates are often associ-
ated with complex biological communities
Woods Hole Oceanographic Institution 69
STANDING TALL — Dense colonies oftubeworms (Lamellibrachia luymesij aggregate around a cold seep site in the Gulf of Mexico. They form the
seafloor equivalent of hedges that provide habitat for many other invertebrates, such as mussels, crabs, shrimp, and snails.
<u
\j
o
that would be disrupted or destroyed by gas
extraction and production.
Recharged oil wells
Recent work by a number of labora-
tories suggests that free gas streaming
through the seafloor or seafloor hydrate
deposits may constitute yet another large
oceanic methane source. On the northern
continental slope of the Gulf of Mexico, for
instance, a process known as "gas washing"
fills subsurface petroleum reservoirs with
natural gas that flows upward from even
deeper reservoirs in the Earth's crust.
It has been estimated that less than
2 percent of generated oil and gas ever
makes its way into commercial reservoirs.
Of the residual oil, about half remains dis-
persed in the source rock and sediments.
The residual oil and organic mat-
ter in deeper sediments is subjected to
more heating and natural processing and
is broken down into natural gas. The gas
streams upward, washing out clogged pore
spaces and recharging many fuel reser-
voirs. Evidence comes from oil wells in the
northern Gulf of Mexico, where we have
observed significant changes in oil com-
positions over time scales as short as 10
years. The wells continue to produce long
after their expected lifetimes.
The other half of the residual oil leaks
upward and out of the sediments into ocean
bottom waters. Remarkable satellite pho-
tographs of the Gulf of Mexico and other
regions reveal slicks extending for miles in
areas where no oil production is occurring.
Similar photographs are now being used to
locate new oil and gas accumulations.
Methane-making microbes
Beyond the geological "cooking and
squeezing" processes that produce petro-
leum and gas, large quantities of gas also
are being produced biologically. Many
gas hydrate accumulations and ocean-
floor gas seeps consist of methane largely
derived from microorganisms.
Bacteria living in oxygen-poor areas
beneath deep-sea sediments on the sea-
floor produce methane as a major product
ot their metabolism. Some models suggest
that bacteria in sediments may account
for 10 percent of the living biomass on
Earth. In addition, microbial communities
beneath the seafloor, whose numbers are
entirely unknown, may also be producing
vast amounts of methane.
Global warming and tsunamis
The pervasive and ongoing movement
of methane gas— from seeps, decompos-
ing hydrates, gas washing, and microbial
sources — leads to some fascinating phe-
nomena and important questions.
Methane is a greenhouse gas that traps
heat about 20 times more effectively than
carbon dioxide. If methane deposits and
seeps prove to be ubiquitous in the oceans,
they are a potentially significant contribu-
tor to global warming.
Relatively modest changes in global
70 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.
PLUMES UNFURLED — Like flower petals atop stems, feather-like red plumes poke out of the tops of the tubeworms'thin, white tubes, which reach
about 50 centimeters (20 inches) high. The plumes act like gills, absorbing nutrients from seawater.
ocean temperatures or sea level could
trigger a massive release of oceanic meth-
ane. If a change in ocean bottom pressure
or a rise in water temperatures passes
a certain threshold, sizable methane
hydrate deposits could decompose rap-
idly and release a large quantity of heat-
trapping gas back into the atmosphere.
This scenario has been proposed as a
possible cause for some past episodes of
rapid global warming.
Evidence from the past suggests that
upward-seeping methane may pose
another threat: underwater avalanches.
Landslides at the edge of the continental
slope just off the East Coast of the United
States may have been triggered by pockets
of methane gas that had built up pressure
under a lid of overlying sediments and
exploded. Similar landslides today might
generate tsunamis that would hit the U.S.
coast. An offshore oil-drilling platform
that accidentally hit such a gas pocket
would also be endangered.
A wide-open new field
Many challenges remain ahead for
researchers. Methane seeps are widely dis-
tributed around the worlds oceans, yet
their discovery remains mainly serendipi-
tous. The volume of oil and gas seeping to
the floor throughout the worlds oceans
is also unknown, as most of the seafloor
remains unexplored.
Even in the cases of known seeps—
especially those found in and around
known oil and gas fields— data on the
rates of seepage are scarce. Yet evidence
suggests that gas seeps and methane
hydrate deposits may be even more per-
vasive than their known extent today and
may play a fundamental role in regulating
ocean chemistry, sustaining marine life,
and shaping seafloor geology.
Jean Whelan earned her bachelor's degree in chemistry
at the University of California, Davis, and her doctorate
*" in organic chemistry from the Massachusetts Institute of
Technology. Before coining to WHOL she carried out post-
doctoral work at Brandeis University and taught chemistry
9 3 at Fairleigh Dickinson University in Madison, N.J. She
studies how to use organic compounds to deduce geological
I 1 processes. Among her research focuses are the formation
1 and migration of petroleum, and she and colleagues have
I t; shown that large quantities of gas flowing through some of
' ^ the world':' oil and gas fields may be continuously altered
and sometimes refreshed by pools of hydrocarbons that lie deep within the Earth. Current research
focuses on how this gas seeping also affects the ocean. When she is not in her lab (or sometimes when
she is), she loves to sing. A contralto, she has sung both as a choir member and a soloist with many
Cape Cod choruses and chamber groups, as well as with her church choir.
Woods Hole Oceanographic Institution 71
Is Life Thriving Deep Beneath the Seaf loor?
Recent discoveries hint at a potentially huge and diverse subsurface biosphere
By Carl Wirsen, Oceanographer Emeritus
Biology Department
Woods Hole Oceanographic Institution
In 1991, scientists aboard the submers-
ible Alvin were in the right spot at the
right time to witness something extraor-
dinary. They had sailed into the after-
math of a very recent volcanic eruption
on the seafloor and found themselves in
a virtual blizzard.
They were densely surrounded by
floes ot white debris, composed of sul-
fur and microbes, that drifted more than
30 meters above the ocean bottom. The
seafloor was coated with a 10-centimeter-
thick layer of the same white material.
This vast volume of microbes did not
come from the ocean. The eruption had
flushed it out from beneath the seafloor.
The discovery was transforming.
It strongly suggested that previously
unimagined and potentially huge com-
munities of microbial life were thriving
in the dark, increasingly hot, oxygen-
depleted rocky cracks and crannies below
the ocean bottom. An abundance of life
apparently flourished in conditions we
had considered too extreme. It shattered
our narrow preconceptions and stretched
our view of the places and circumstances
that can harbor life.
'Everything is everywhere'
With our horizons expanded, we have
launched new initiatives in the past decade
to search for life deep within the Earth — to
A MICROBIAL BLIZZARD — In 1991, scientists aboard the submersible Alvin witnessed the aftermath of a very recent volcanic seafloor eruption
and found themselves in a torrent of white debris. The eruption flushed huge floes of microbes (and white sulfur filaments created by the
microbes) out of subsurface crevices and discharged them from the seafloor. The discovery pointed to previously unsuspected and potentially
huge communities of microbes living beneath the seafloor.
72 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
explore the so-called subsurface biosphere.
In recent years, scientists have discovered
many new subsurface biosphere habitats —
reaffirming the principle of the pioneering
microbiologist Martinus Willem Beijer-
inck (1851-1931), who said, "Everything is
everywhere, the environment selects." Bei-
jerinck's approach — to study "the relation
between environmental conditions and
the special forms of life corresponding to
them" — certainly applies to the subsurface
realm, where biology and microbiology
interact with geology and hydrology.
What organisms inhabit this deep bio-
sphere? How deep are they living? How
long can they survive under these con-
ditions? How have they adapted to take
advantage of energy supplied by the
planet, rather than by the sun?
What impact, in turn, does this bio-
sphere have on the oceans and the planet?
What can these hardy entrepreneurial
organisms teach us about the origin and
evolution of life on Earth? How can they
guide our search for life on other plan-
etary bodies?
We are at the frontiers of answering
these questions.
Better living through chemistry
The amazing discovery ot life at
seafloor hydrothermal vents in 1977
reminded us that solar energy, oxygen,
organic matter, and photosynthesis are not
the only fundamental building blocks and
chemical processes that foster life.
In place of energy from the sun, cer-
tain organisms use chemosynthetic reac-
tions to live and grow. They use inorganic
chemicals, such as hydrogen and hydro-
gen sulfide, rather than organic matter tor
their energy and carbon dioxide as their
source of carbon. Geothermal, rather than
solar, energy catalyzes chemical reactions
that generate life-sustaining chemicals
from rocks and seawater. Water is the
only absolutely essential ingredient.
But it was reasonable to assume that
conditions below the seafloor and deeper
into the subsurface would become more
extreme and life would become sparser
Examining the white floes discharged from
the 1991 seafloor eruption, WHOI scientists
discovered a new genus of bacteria called
Arcobacter. It lives in low-oxygen conditions
and metabolizes hydrogen su/fide to obtain
energy. An end-product of this metabolism
is a unique form of sulfur, which the bacteria
excrete in the form of solid, white filaments.
or nonexistent. Yet in the past decade we
have found an extraordinary diversity of
subsurface microbes living in a wide range
of conditions — buried deep within ocean
sediments, in hot ocean crust crevices, in
frozen polar soils, and in the subterranean
bowels of deep mines.
In all these places, individual species
have adapted to extreme conditions that
include high pressure, high and low tem-
peratures, unusual or toxic chemicals and
minerals, or low availability of essential
nutrients. Often they take advantage of
specific extreme conditions to carve out a
niche where they can thrive and other spe-
cies cannot.
Life finds a way — often cleverly
Take, for example, the mats of white
microbial sulfur debris witnessed by sci-
entists aboard Alvin in 1991. WHOI
Associate Scientist Craig Taylor and I
subsequently found that such mats are
produced by a genus of bacteria called
Arcobacter. It lives in low-oxygen condi-
tions and metabolizes hydrogen sulfide
(HiS) to obtain energy. An end-product of
this metabolism is a unique form of sultur,
which the bacteria ingeniously excrete in
the form of solid, white filaments.
Together, large populations of these
bacteria produce crosshatched mats of
these filaments. In the face of flowing sub-
surface hydrothermal fluids, these mats
help keep the bacteria anchored to rocky
surfaces where Arcobacter are perfectly
„<•" /
'
A titanium ring deployed at a Pacific hydro-
thermal vent site indicates the presence of
bacteria thriving beneath the seafloor. Within
days, Arcobacter bacteria, discharging from
the subsurface, rapidly colonize the ring, pro-
ducing a white sulfur filament mat up to
3 centimeters thick as they grow.
suited. They are bathed in hydrothermal
fluids percolating from the ocean crust,
which are low in oxygen and high in
hydrogen sulfide. In this niche, Arcobacter
feasts on ample HiS-rich fluids and out-
competes other oxygen-respiring bacteria.
It turns out that these discharged bac-
terial mats may also provide an important
carpeting around hydrothermal vents that
attracts other animals, such as Alvinella
tubeworms, and encourages them to set-
tle and grow. And when we looked closer
Woods Hole Oceanographic Institution 73
to home, we found Arcobacter bacteria
in sediments in the shallow depths of Eel
Pond in Woods Hole that grow and pro-
duce the same sulfur filaments as those at
the deep-sea vents.
The world's largest bacterium
Remarkable microbial adaptations like
this seem to be common nearly every-
where we look. In 1999, far from any
undersea volcanic areas, the worlds larg-
est bacteria were identified by an interna-
tional scientific team that included former
WHOI microbiologist Andreas Teske, who
is now at the University ot North Carolina.
They were found in the surface layers of
ocean sediments oft the coast ot Namibia,
where they find what they need: hydrogen
sulfide tor energy and nitrate to respire.
This bacterium, Thiomargarita nami-
bienus ("Sulfur pearl of Namibia")
reached sizes up to 750 microns (normal
bacteria are only 1 to 2 microns). It was
so large it could be seen with the naked
eye, and it shattered our conventional
wisdom that inherent bacterial physiol-
ogy prevented them from ever getting so
big. Their size is due to a large vacuole in
their cells, in which they store nitrate, as
do some hydrothermal vent microbes, to
survive periods when oxygen is lacking-
much the way we might store oxygen in
external SCUBA tanks to remain alive
under water.
Deep, dark, old, and cold
Arcobacter and Thiomargarita are
examples of well-adapted bacteria found
in the shallow subsurface. But deeper sub-
surface explorations in the past decade
have revealed unique, heretofore unknown
microbial habitats.
Some of the first investigations of the
deep subsurface were motivated by con-
cerns about pathogens and toxic chemi-
cals in groundwater supplies. The Wit-
waterstrand Deep Microbiology Project,
for example, a multinational effort led by
Swedish scientists, sampled groundwater
in fractured rock from 3-kilometer-deep
gold mines in South Africa and found a
wealth of microbial diversity in the deep
continental crust.
In 2000, researchers from West Ches-
ter (Penn.) University claimed to have
discovered the oldest-known living
microorganism in an ancient salt deposit
in New Mexico, buried 610 meters (2,000
feet) below ground. It was trapped in a
tiny brine-tilled pocket that formed in a
salt crystal 250 million years ago. Long
after the dinosaurs became extinct 65
million years ago, it lay in a dormant
state, waiting for the right conditions
to "awaken" its genetic machinery and
resume growing and reproducing, the
researchers said.
In the Arctic and Antarctic, scientists
have found metabolically active microbes
in subsurface permafrost frozen at tem-
peratures of -10°C ( 14°F) or colder for 2
million to 3 million years. High popula-
tions of viable microbes have been found
in oceanic sediment cores deeper than a
halt-kilometer, which would make them
older than 10 million years.
Some like it hot
Ultimately, a combination of physical
and chemical factors will set the limits at
which lite can exist. In general, increas-
ing pressure will not limit the depth at
which subsurface life is found. Increasing
heat is the primary limiting factor, and it is
doubtful that we have discovered the max-
imum temperature at which life can exist.
At hydrothermal vents, volcanic heat
has created an environment in which
hyperthermophilic (super-heat-loving)
microbes thrive. The maximum growth
temperature for a microorganism so far
was discovered in 2003 by scientists at the
University of Massachusetts. They called
it Strain 121, because it grows at a 121°C
(250°F). But scientists generally agree that
life could exist at temperatures as high as
140° to 145°C (284° to 295°F).
In the mid-1990s, scientists found
novel hyperthermophilic microbes in
hot oil reservoirs 3 kilometers below the
North Sea and the North Slope of Alaska.
Oil producers had thought that microor-
ganisms, which "sour" or contaminate oil,
were introduced into wells, but, in fact,
they are naturally occurring and live on
organic compounds in oil.
Such discoveries push our understand-
ing of the limits of life and the limits of
where to look for it. The largest known
biosphere— fully 80 percent of Earth's
available living space — is in the deep
ocean, yet this may be eclipsed by the sub-
surface biosphere as research into this
realm proceeds.
Drilling down to search for life
Deep-sea drilling remains the best
way to sample the subsurface, though it
has limitations. It is costly, and poten-
tially results in contamination of the sam-
ples retrieved.
The deep biosphere has been targeted
as a major research initiative of the new
multinational Integrated Ocean Drilling
Program, which operates deep-sea drill
ships for the oceanographic community.
(See "A Sea Change in Ocean Drilling,"
page 32.) A new permanent microbiology
laboratory was outfitted aboard the
JOIDES Resolution drill ship.
Scientists have also developed new
instrument packages that plug into and
seal drilled seafloor holes, where they
remain tor months. These probes otter
potential windows into the interacting
chemical, hydrological, geological, and
biological processes that occur beneath
the seafloor. These long-term observato-
ries have been dubbed "CORKs," which is
both an eponym and an acronym (Circu-
lation Obviation Retrofit Kits).
The real challenge is to develop sen-
sors that can be placed in situ in a way
that doesn't disrupt the ecosystems they
are meant to record and that are sensitive
enough to provide continuous, real-time,
monitoring ot processes occurring on
even a molecular scale.
Drilling cruises are scheduled to search
for microbial life buried hundreds of
meters deep under thick ocean sediments
piled atop ocean crust in volcanically qui-
escent continental slope regions. In 2000, a
74 Ocean us Magazine • Vol.42, No. 2 • 2004 • oceanusmag.whoi.edu
WORLD'S LARGEST BACTERIUM — In 1 999 scientists discovered a previously unknown bacterium, which is large enough to be seen with the naked
eye. Found off the coast of Namibia, the bacteria grow in long lines of single cells, each stuffed with reflective white globules of sulfur. The bacteria
resembled a string of pearls to its discoverers, who named it Thiomargarita namibienus ("Sulfur pearl of Namibia"). The bacteria have evolved to
live on seafloor sediments, where they find hydrogen sulfide for energy and nitrate for respiration. Their size is due to a large vacuole that fills the
interior of their cells like inflated balloons. The vacuole stores nitrate, giving Thiomargarita the ability to survive periods when oxygen is lacking —
a built-in equivalent of an oxygen-storing SCUBA tank that allows humans to remain alive under water.
consortium of Japanese scientists launched
a several-year project using drill ships,
manned submersibles, remotely operated
vehicles, and long-term sensors to explore,
drill, and monitor the subsurface bio-
sphere beneath hydrothermal vents near
Suiyo Seamount, an active subsea volcano
in the western Pacific.
Going to extremes
A major research goal of the Deep
Ocean Exploration Institute at Woods
Hole Oceanographic Institution is to
extend our subsurface search into condi-
tions on Earth that are deeper, hotter, and
harsher than anything previously studied.
We want to learn more about the biologi-
cal and geochemical interactions that take
place within this biosphere.
Any residents we find in these frontiers
may well be biochemical pioneers. In their
genes, they will still have the original blue-
prints for a wide range of possible biologi-
cal processes. Some of these processes — like
Arcobacters sulfur filament machinery, or
Woods Hole Oceanographic Institution 75
Thiomargarita's large, nitrate-storing vacu-
ole, or the extreme heat tolerance of Strain
121 — we may never have seen before.
Life on Earth and other planets
We may never know with complete
certainty where and how lite originated
on Earth, but the hot subsurface around
hydrothermal vents is a likely candidate.
It is an environment that seems to have all
the necessary ingredients to spark critical
chemical reactions that could create the
precursor building blocks ot living organ-
isms— ultimately resulting in amino acids
for proteins, the genetic machinery DNA
and RNA, sugars for energy, and lipids to
make membranes.
In a hot subsurface melting pot, far
from solar ultraviolet radiation that can
break down complex molecules, these
chemicals could find sanctuary in tiny
rocky crevices where they could congre-
gate, interact, and perhaps combine even-
tually with a membrane around them.
Below the sea, they would certainly be
sheltered from meteor bombardment and
other life-threatening conditions that buf-
feted the early Earth's surface.
Further insights into life's ability to
survive harsh conditions will guide our
search for extraterrestrial life. New evi-
dence from Mars shows that it once had
water, and it may once have had seas
A microbial garden beneath the seafloor
Recent discoveries have raised the possibility of a huge and diverse subsurface biosphere of microbial life. Below the ocean floor, a variety of
chemical reactions between seawater and rocks, taking place over a range of temperatures, creates a chemical bouillabaisse. These chemicals
diffuse upward and become sources of energy and carbon that sustain a wide variety of microbes. The microbes have evolved to take
advantage of specific conditions in particular niches.
76 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.ee
that left salt deposits like those in New
Mexico. Europa, (upiter's moon, is prob-
ably volcanic, and beneath its ice-covered
surface may lie oceans with hydrother-
mal activity. The same tools and tech-
niques we devise to search for life within
and beneath Earth's volcanic oceans will
prove useful there.
Our journey into Earth's subsurface
biosphere is a quest to find the limits
of life.
Curl \Virscn, a Massachusetts native, came to WHOI in 1968 and retired
in 2003 as a Senior Research Specialist, having pursued research in marine
microbiology over 35 years. Being among the first scientists to dive at the
newly discovered hydrotlierniiil vents in the late 1970s and early 1980s, he
never /iii ked novel rcsciirch opportunities (including "tasting" tlie lunch
| preserved for 11 months in Alvin after it sank in 1968 and was recovered
; in 1969). Now, as an Oceanographer Emeritus, he can pursue some ques-
tions that remained unanswered over the years, as well as some favorite
- outside WHOI activities. For him and his wife Joye, who also retired from
WHOI, dogs have always been a big part of their lives, and training them for field and obedience is
almost as much fun as rearing a litter. Four grandchildren and all the activities that come with being
a grandparent (fishing, camping, science) make life busier now than it ever has been.
A SMORGASBORD OF CHEMICAL REACTIONS— Microbes living in the subsurface at or near deep-sea hydrothermal vent sites exploit a wide
range of conditions. Here is a list of known or possible chemical reactions microbes use to live and grow.
1 Aerobic metal and metal sulfide oxidation
(Fe+2 or Mn+2 + 02 ->Fe+3 or Mn+3 + H20)
Aerobic sulfide/sulfur oxidation
(H2S + 02 -*H2S04)
^Aerobic methane oxidation
(CH4 + 02 ->C02 + H20)
Aerobic hydrogen oxidation
3) Anaerobic sulfate reduction
(5H2 + S04-»H2S + 4H20)
Anaerobic iron reduction
(Fe+3 + H2->Fe+2)
Sulfur reduction via organic carbon utilization
(Organic C + elemental S->H2S + C02)
Anaerobic sulfur respiration
(Elemental 5 + H2->H2S)
£> Anaerobic methane production (methanogenesis)
(4H2 + C02->CH4 + H20)
Woods Hole Oceanographic Institution 77
The Evolutionary Puzzle of Seaf loor Life
Scientists are assembling critical pieces to reconstruct the history of life on the ocean floor
By Timothy M. Shank, Assistant Scientist
Biology Department
Woods Hole Oceanographic Institution
From the far reaches of their empire,
the Romans brought back all sorts of
beasts for their menageries and gladiator
spectacles— lions from Africa, bears from
northern Europe, and ibexes from the
deserts of the Middle East. If the empire
had reached Australia, the Romans surely
would have imported kangaroos, koalas,
and other marsupials found nowhere else
on Earth.
Scientists have long pondered why dif-
ferent species are distributed in various
places — a field called biogeography. The
line of inquiry stretches from Aristotle,
through Carolus Linnaeus, the 18th-cen-
tury father of taxonomy, to Charles Dar-
win, up to E.O. Wilson, who coined the
modern term "biodiversity." Until the lat-
ter half of the 20th century, however, bio-
geography was a strictly terrestrial pursuit.
In 1977, scientists diving on the Gala-
pagos Rift in Alvin made a discovery
that shook the foundations of biology.
They found oases of animals thriving in
the sunless depths around hydrothermal
vents. Instead of photosynthetic plants,
chemosynthetic microbes constitute the
base of the food chain at vents. They
obtain energy from chemical-rich flu-
ids generated by volcanic processes on
mid-ocean ridges, the 75,000-kilometer
(45,000-mile) undersea mountain chain
that encircles the globe and marks the
edges of Earths tectonic plates.
Since the discovery of vents, scientists
have explored hundreds of volcanically
active vents in the Pacific, Atlantic, and
Indian Oceans. And we have found that
on the seafloor, as on land, distinct ani-
mal populations have evolved in differ-
ent regions.
A biogeographic seafloor tour
In the eastern Pacific, tubeworms domi-
nate vent sites, but they are notably absent
GIANT CLAMS — This vent site on the Galapagos Rift, discovered in 2002, is called "Caly field" after Calyptogena magnifies, a clam speices that
grows up to 1 foot long. It thrives on chemical nutrients in hydrothermal fluids seeping between crevices around seafloor pillow lava.
78 Oceanus Magazine • Vol. 42, No.2 • 2004 • oceanusmag.whoi.edu
at vents in the Atlantic. Instead, billions
of shrimp swarm at vents along the Mid-
Atlantic Ridge, which bisects the Atlantic
Ocean floor. Both Pacific and Atlantic vents
have mussels, but not the same species.
Scientists today recognize six major
seafloor regions — called biogeographic
provinces — with distinct assemblages of
vent animal species. Beyond the tube-
worm-dominated eastern Pacific, there are
two provinces in the North Atlantic, where
different species of shrimp and mussels
predominate at deep vent sites to the south
and shallower ones to the north.
The fourth province is in the north-
east Pacific, off the U.S. Northwest coast,
which shares similar species (clams,
limpets, and tubeworms) with the east-
ern Pacific, but different species of each.
Across the ocean in the western Pacific,
vents are populated by barnacles, mussels,
and snails that are not seen in either the
eastern Pacific or the Atlantic.
Scientists got their first chance to search
for vents in the Central Indian Ocean in
2001 and found the sixth province. These
vents are dominated by Atlantic-type
shrimp, but also had snails and barnacles
resembling those in the western Pacific.
Evolutionary detectives
All these regions contain the same basic
ingredients to support life— chemical nutri-
ents generated by geothermal processes at
hydrothermal vent sites. So why do vent
fauna differ in the Atlantic and Pacific, or
in the eastern and western Pacific? How do
we assemble these puzzle pieces to explain
the diversity and evolution ot vent species
throughout the world's oceans?
Evolutionary biologists are detectives,
gathering clues and sorting through events
and phenomena to reconstruct the pro-
cesses that generated the patterns we see
today. Suspected factors include:
• topographical seafloor features that help
or hamper the dispersal of species;
• the movement of Earth's plates, which
disconnects underwater mountain
chains, and closes and opens gateways
between oceans;
A MIXED COMMUNITY OF VENT LIFE— A cloud of flea-like crustaceans called amphipods hovers
around tubeworms encrusted with limpets and mussels at the 9°/V vent site in the eastern Pacific.
• deep-sea currents that aid or hinder the
dispersal of vent larvae; and
• migrations of species (over evolutionary
history) between vents and other seafloor
habitats that foster chemosynthetic ecosys-
tems. These include whale carcasses (called
"whale falls") and "wood tails" from ship-
wrecks or trees cast into coastal regions.
Which combinations of these variables
limited or encouraged the dispersal of
animal populations along the widely scat-
tered, ephemeral patchwork of active vents
on mid-ocean ridges? Which sent some
populations down divergent evolution-
ary pathways, led others to extinction, and
created fertile niches for yet others?
Breakthroughs in biotechnology that
allow rapid gene sequencing now give sci-
entists powerful new abilities to compare
genomes of different species. We can see
how closely related they are and examine
how far back in time they may have diverged
on the evolutionary tree. Determining evolu-
tionary relationships among seafloor species
and communities, and their distribution and
biodiversity, will help unravel the evolution
of life on Earth. And it will guide our search
for life on other planetary bodies, where che-
mosynthesis may reign.
Breaking the mountain chain
Ecosystems on land and the seafloor
differ substantially, but terrestrial evolu-
tionary lessons may still apply to the deep
sea. For example, as Australia separated
from the ancient supercontinent of Pan-
gaea and became an island, its animal
population became divorced from other
populations (including predators) and
began to evolve separately. Did something
similar happen on the ocean floor?
About 40 million years ago, a continu-
ous mid-ocean ridge system existed in
the east Pacific, extending from below the
equator to the coast of what is now the
Northwest United States. In the continu-
ing reorganization of tectonic plates on
Earth's surface, however, the North Ameri-
can Plate pushed westward. It began to
override the Pacific Plate, forcing a por-
tion of it underneath the North American
Plate. Plate tectonics effectively discon-
nected northeast Pacific ridges from the
rest of the Pacific ridge system.
As a result, northeast Pacific vent com-
munities diverged from their equatorial
Pacific cousins. Both regions' vents share
clams, limpets, and tubeworms, but not
Continued on page 82
Woods Hole Oceanographic Institution 79
On the Seafloor, Different Species
Thrive in Different Regions
Soon after animal communities were discovered
around seafloor hydrothermal vents in 1977, sci-
entists found that vents in various regions are
populated by distinct animal species. Scien-
tists have been sorting clues to explain how
seafloor populations are related and how
they evolved and diverged over Earth's
history. Scientists today recognize dis-
tinct assemblages of animal species
in six major seafloor regions (colored
dots) along the system of volcanic
mountains and deep-sea trenches
that form the borders of Earth's
tectonic plates. But unexplored
ocean regions remain critical
missing pieces for assembling
the full evolutionary puzzle.
• Northeast Pacific vent
communities are domi-
nated by "bushes" of skinny tube-
worms called Ridgea piscesae.
Western Pacific vent com-
munities are dominated by
barnacles and limpets, as well as
hairy gastropods, shown above.
Challenger Deep
Unusual life forms may have
evolved under conditions of
extreme pressure in this 1 1,000-
meter-deep trench, the deepest
part of the world's oceans.
New Zealand
This region has a full spectrum of
habitats supporting seafloor life
(hydrothermal vents, cold seeps,
whale carcasses, and wood from
shipwrecks and trees) in close
proximity. How have species
evolved in these diverse settings?
Chile Rise
This region has a variety of
chemosynthetic habitats and
geological features in close prox-
imity. How do seafloor popula-
tions diverge or converge at this
triple junction on the "highway"
of mid-ocean ridges?
80 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
• Shallow Atlantic vents
(800- 1 700-meter depths)
support dense clusters of mussels
on black smoker chimneys.
Deep Atlantic vent com-
munities (2500-3650-me-
ter depths) are dominated by
swarms of shrimp called Rimica-
ris exoculata.
/ Central Indian vent com-
/ munities are populated by
Western Pacific-type fauna, but
also have North Atlantic-type
shrimp species.
Eastern Pacific vent com-
munities are dominated
by tall, fat tubeworms called
Riftia pachyptila.
Southern Ocean
The Drake Passage may act as a
key link or bottleneck for larval
dispersal between the Atlantic
and Pacific. Whale carcasses and
shipwrecks (such as Shackleton's
Endurancej may offer refuges or
stepping-stones between vents.
South Atlantic
Powerful currents and huge
seafloor chasms (fracture zones)
may act as barriers blocking the
dispersal of vent larvae and dis-
connecting vent populations in
the North and South Atlantic.
Caribbean
In this region, methane seeping
from the seafloor also supports
animal communities. Did
animals migrate between "cold
seeps" and nearby hot vents over
evolutionary history?
Arctic Ocean
The Arctic Ocean has never
had deep connections with
other major oceans. It may
harbor fundamentally
different vent animals that
evolved in isolation over the
past 25 million years.
Woods Hole Oceanographic Institution 81
SWARMING SHRIMP — Like bees around a hive, shrimp aggregate to feed at the Rainbow vent site in the North Atlantic.
Continued from page 79
the same species. Northeastern tubeworms
(Ridgea piscesae) have skinny tubes, while
eastern tubeworms (Riftia pachyptila) have
fatter tubes.
A ridge too far
The geological evolution of mid-ocean
ridges and ocean basins influences biolog-
ical evolution in other ways.
Hydrothermal vents are typically found
in rift valleys at the crests of mid-ocean
ridges, but there are striking differences in
ridges. On the fast-spreading East Pacific
Rise, rift valleys are typically 200 meters
(656 feet) wide and 10 meters (33 feet)
deep. But on the slower-spreading Mid-
Atlantic Ridge, rift valleys are often 1 kilo-
meter (0.6 miles) wide and 2 kilometers
(1.2 miles) deep. Such deep valleys may
have become "dead-ends" for vent animals
inside, whose larvae could not get up and
over the valley walls to disperse and colo-
nize new vent sites in other valleys.
Pacific and Atlantic ridges differ in
other ways, too. Mid-ocean ridges are
broken into segments by extensive faults,
called fracture zones, which intersect the
ridges at roughly perpendicular angles.
Ridge segments move apart along fracture
zones, breaking the otherwise straight line
of the ridge and creating a zigzag pattern
of alternating ridges and fracture zones.
In the Pacific, two ridge segments may
be separated by 10-kilometer (6 mile)-long
fracture zones. But in the Atlantic, frac-
ture zones typically span hundreds of kilo-
meters—perhaps an unbridgeable gap for
vent larvae trying to disperse into an adja-
cent ridge segment to the north or south.
Closing ocean gateways
Plate tectonics may have played an evo-
lutionary role in another way — by open-
ing and closing passages between oceans.
As Pangaea began to break up 200 million
years ago, and North America and Europe
separated from Africa, the ancient Tethys
Ocean formed between them. It was the
precursor of the Mediterranean Sea, and
it allowed a free flow of waters from the
proto-Atlantic Ocean to the Indian Ocean.
Perhaps this bygone oceanic route links
Atlantic and Indian Ocean shrimp popu-
lations, which are diminished at shallow
Atlantic vents (800-1,700 meters/2,625-
5,577 feet), more abundant at deeper
Indian vents (2,400 meters/7,874 feet), and
swarming by the hundreds of thousands
at even deeper Atlantic vents (2,500-3,650
meters/8,202-11,975 feet). The shrimps'
predilection for greater depths may have
allowed their larvae to disperse via deep
currents from the Atlantic to the Indian
Ocean — an avenue unavailable to other
species— rather than via a route along
the mid-ocean ridge, through the South
Atlantic and around the Horn of Africa.
Perhaps the explanation for distinct
North Atlantic fauna — with its swarm-
ing shrimp and no tubeworms — involves
other ocean gateways. The North Atlan-
tic Ocean basin began to form about 180
million years ago, but South America and
Africa remained connected until about
110 million years ago. That means that
North Atlantic populations evolved for
some 70 million years before the South
Atlantic existed. Further, the Drake Pas-
sage between Antarctica and the tip of
South America, a crucial oceanic gateway
connecting the Pacific and Atlantic, did
not open until about 21 million years ago.
Genetic comparisons of shrimp popu-
82 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
lations throughout these oceans will allow
us to reconstruct the pathways of shrimp
migration over evolutionary history and
answer these questions.
The rise of the Isthmus of Panama
A similar combination of geologi-
cal and oceanographic factors may also
explain the Logatchev vent site. Discov-
ered in 1994, it is the only Atlantic vent
site where clams are known to exist.
Logatchev, the southernmost known vent
site in the Atlantic, is located just east of
the Caribbean Sea.
Did clams originally migrate from the
Pacific via an ancient seaway that con-
nected the Pacific and Atlantic— before
the Isthmus of Panama rose about 5 mil-
lion years ago to block it? Or did clam
populations from western Pacific vents
migrate around South America until they
were impeded south of Logatchev by
ocean currents or seafloor topography?
Or perhaps clams thrive in other Atlantic
vents that we haven't discovered yet.
Chasms and currents
Complex processes create the pat-
terns ot vent populations we have seen.
But we can't truly begin to assemble this
evolutionary puzzle without having all
the pieces. Major regions of the seafloor
remain unexplored.
We have not yet located vent sites in
the South Atlantic, for example. The puz-
zle pieces we do have indicate a discon-
nection between most fauna in the North
Atlantic and on other mid-ocean ridges.
The most intriguing explanation
may involve the Romanche and Chain
Fracture Zones located near the equa-
tor, which are particularly large — 1,000
kilometers (620 miles) long and almost 8
kilometers (5 miles) deep. A strong, deep
current flows along and through these
fracture zones, almost straight across
the South Atlantic (pushing a volume of
1 million cubic meters of water per sec-
ond at an average speed of 10 centime-
ters per second). We think that the com-
bination of these currents and fracture
A WHALE OF A MEAL — Orange microbes coat the skeleton of a whale that fell to the seafloor off
California. Craig Smith (University of Hawaii) and colleagues found that microbes decompose
whale tissue and bones, producing hydrogen sulfide nutrients that sustain thriving animal popu-
lations. Whale-fall communities share many species with other chemosynthetic seafloor commu-
nities, such as hydrothermal vents and seeps, and may act as stepping-stones between them.
zones may act as physical barriers block-
ing the transport of vent fauna between
the North and South Atlantic — a subsea
equivalent of a Berlin Wall.
Islands in the seafloor stream
Other topographical factors may also
explain the existance of unique biological
communities found at seamounts— sub-
merged extinct or active volcanoes that
typically are isolated from mid-ocean
ridges. Vent communities at Loihi Sea-
mount, an active subsea volcano near
Hawaii, and at Edison Seamount, off the
coast of Papua, New Guinea, each have
their own set of endemic species.
Are vent larvae from these isolated sea-
mount communities too far away to reach
mid-ocean ridges or other hydrothermal
vent locations? Or do seamounts, towering
miles above the seafloor, cause deep-ocean
currents to swirl around them— creating
impenetrable vortices that dispersing lar-
vae cannot breach?
Despite their isolation, seamounts may
play a vital role in the evolutionary his-
tory ot seafloor life. Standing high above
catastrophic seafloor events, such as mass
extinctions, they might have offered criti-
cal refuges for vent fauna. Or they may
have provided fortuitous stepping-stones
between mid-ocean ridge vent sites.
Deep-sea seeps and sanctuaries
Into this complex equation, scientists
have added new variables by discovering
other deep-sea habitats that foster chemo-
synthetic life.
In shallower seafloor regions on conti-
nental margins, for example, naturally cre-
ated methane and hydrogen sulfide seep
from the seafloor. These so-called "cold
seeps" support chemosynthetic ecosys-
tems (including tubeworms, mussels, and
shrimp) that are different but analogous to
vents. (See "When Seafloor Meets Ocean,
the Chemistry Is Amazing," page 66.)
Genetic studies have shown that tube-
worms at seeps and vents are not the same
specie, but the two habitats do share 13
species. Over evolutionary history, did
species migrate back and forth between
seeps and vents?
Fossil records show that ancestors of
Woods Hole Oceanographic Institution 83
r
ra
A WALL OF WATER — The Mid-Atlantic Ridge near the equator is offset by long, deep faults,
called fracture zones. A strong deep current flows along and through these fracture zones,
almost straight across the South Atlantic. The combination of currents and fracture zones may
act as a physical barrier blocking the transport of vent fauna between the North and South
Atlantic Oceans — a subsea equivalent of a Berlin Wall.
vent tubeworms, barnacles, and limpets
are as old as the dinosaurs. Several mass
extinctions, like the one that annihilated
the dinosaurs, have occurred in the deep
sea. Could seeps or vents act as deep-
sea refuges, where colonies of animals
survived catastrophes and subsequently
reseeded populations?
Whale falls and shipwrecks
In 1987, Craig Smith of the University
of Hawaii and colleagues found a thriving,
diverse animal community on a whale car-
cass that had fallen to the ocean bottom.
Microbes decomposing the whales soft
tissue and bone lipids produced hydrogen
sultide nutrients similar to those that sus-
tain vent communities. To date, more than
20 whale-fall ecosystems have been found.
They share 10 species with vents and 19
with seeps.
Similarly, wood — from trees discharged
into the coastal ocean by rivers and mud-
slides, or from shipwrecks in coastal and
more far-flung areas— also provides chem-
ical nutrients from decaying organic mat-
ter to support chemosynthetic animals.
Like seamounts and cold seeps, whale
and wood tails likely play important evo-
lutionary roles. Whale falls along migra-
tory routes, and trees deposited near river
mouths, may serve as stepping-stones
to help disperse seafloor fauna. Wooden
shipwrecks may have introduced more
recent seafloor stepping-stones.
Recent evidence indicates that wooden
shipwrecks began earlier than suspected in
history, were not confined to coastal areas,
and occurred more frequently than pre-
viously believed. Between 1971 and 1990
alone, an estimated 3,000 wooden ships
were wrecked.
Finding the right niche
We can add other possibilities to this
expansive list of factors that may have
influenced the evolutionary process. For
starters, the proliferation ot certain ani-
mals in certain locations is probably
encouraged or discouraged by subtle dif-
ferences in the chemistry of vent fluids,
seafloor rocks, larval swimming behaviors,
or other unknown factors.
Tubeworms may have an advantage
over shrimp at fast-spreading Pacific mid-
ocean ridges where eruptions occur more
frequently than they do in the Atlantic.
Eruptions destroy vent communities. Ani-
mals stand a tar greater chance ot avoiding
extinction caused by frequent eruptions if
they have evolved with the abilities to col-
onize sites, mature, and reproduce quickly.
On the other hand, as these sites mature
or become more stable, other animals may
seize the advantage and overwhelm the
original colonists.
According to one theory, the newly
forming Atlantic Ocean — small, shallow,
and more susceptible to evaporation— was
more saline than it is now. Shrimp may
have been more salt-tolerant and may
have settled first at nascent Atlantic vent
sites. Once they took hold, species such
as tubeworms could never attain a foot-
hold. Shrimp assiduously scrape the sur-
faces of vent chimneys to harvest bacte-
ria and, the theory continues, they would
have consumed any tubeworm larvae that
happened to settle long before the worms
could mature.
Filling in the critical gaps
To assess the critical factors that redi-
rected evolutionary pathways, the first
step is to locate and sample what's liv-
ing in key places for making evolutionary
comparisons. Is South Atlantic vent fauna
similar to that in the North Atlantic or
Pacific? We will find out during an expedi-
tion in 2005, funded by the National Oce-
anic and Atmospheric Administration's
Ocean Exploration program, to search for
vents in the South Atlantic.
Proposed expeditions aim to add miss-
ing pieces to the puzzle from the largely
A NEW HYBRID ROBOT— Scientists hope to
use the Hybrid Remotely Operated Vehicle
(HROV), now being designed and built at
Woods Hole Oceanographic Institution, to
explore Challenger Deep — a 35,800-foot-
deep seafloor trench in the Pacific. They will
search for unusual life that may have evolved
under conditions of extreme pressure.
84 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edi
^•I^^^HH^lH^^^^^^IHII^^^^^^^^^Hi^H
unexplored Southern Hemisphere. In 2006,
we hope to explore the seafloor oft New
Zealand, which otters a natural laboratory
to investigate relationships among the full
spectrum of habitats supporting seafloor
life (vents, seeps, whale falls, and ship-
wrecks) that exist in close proximity there.
Expeditions proposed for 2006 to the
Chile Rise otter a similar opportunity to
examine — all in one region — a diversity of
chemosynthetic communities, along with
a wide range ot plate tectonic processes,
from mid-ocean ridge spreading to an
oceanic plate being subducted back into
the mantle. It also includes a triple junc-
ture of mid-ocean ridges, where vent pop-
ulations may converge or diverge.
In 2006, NOAA's Ocean Exploration
program will hind expeditions to search for
vents near Antarctica, on the East Scotia
Rise and in the Bransfteld Strait, which is
strategically located at a narrow but criti-
cal juncture linking the Pacific and Atlantic
Oceans. The powerful Antarctic Circum-
polar Current, rushing easterly through the
Drake Passage, may prove to be a. magic
carpet dispersing Pacific vent larvae to the
Atlantic, a wall ot water preventing the
westward flow ot Atlantic larvae, or a bot-
tleneck choking flow in both directions.
The region also has a large whale popu-
lation, giving us opportunities to explore
whale falls. And the tempestuous Drake
Passage and Southern Ocean ice have
populated the seafloor with wood falls
from shipwrecks that we will search for —
including Sir Ernest Shackleton's famous
ship Endurance.
Under the ice and in the trenches
There is another icy, unexplored ocean
that we are eager to explore— the Arctic
Ocean. In 2001, the spectacularly success-
ful maiden voyage of the U.S. icebreaker
Healy found evidence from water samples
and seafloor rocks of far more volcanism
and hydrothermal venting than was pre-
dicted on the ultra-slow-spreading Gakkel
Ridge in the Arctic Ocean. (See "Earth's
Complex Complexion," page 36.)
Are Arctic vent species similar to those
An Antarctic cruise in 2007 will search for
the wreck of Ernest Shackleton's ill-fated ship
Endurance and test the theory that wooden
wrecks play an important role in sustaining
and dispersing seafloor populations.
in the Atlantic, the Pacific, or to neither?
Our current knowledge suggests that
the Arctic Ocean has never maintained
deep-water connections with neighbor-
ing oceans. We theorize that Arctic fauna
has evolved in isolation — producing fun-
damentally different kinds of species. Per-
haps species dwelling on the remote Gakkel
Ridge are "living fossils"— relicts of ancient
species that continue to thrive today just as
they have for tens of millions of years.
In 2007 we will return to the Gak-
kel Ridge on an expedition funded by
the National Science Foundation (NSF)
and the National Aeronautics and Space
Administration's Astrobiology Science
and Technology for Exploring Planets
(ASTEP) program. To explore beneath
the ice, we will employ new autonomous
underwater vehicles built at Woods Hole
Oceanographic Institution to work in ice-
covered oceans. (See "Unique Vehicles
for a Unique Environment," page 25). We
also hope to use the cutting-edge Hybrid
Remotely Operated Vehicle (HROV), now
being designed and built at WHOI. The
HROV will be able to operate at depths
up to 1 1,000 meters (36,000 feet) in two
modes: as an autonomous, or free-swim-
ming, vehicle for wide area surveys, and as
a tethered, or cabled, vehicle for close-up
sampling and other tasks.
When the HROV is ready in 2006, we
plan to take it on a cruise funded by the
NSF to the Challenger Deep, a trench off
the Marianas Islands in the western Pacific
that is 10,923 meters (35,838 feet) deep-
deeper than Mount Everest is tall. In a
place where the pressure reaches 16,000
pounds per square inch, we will test our
hypothesis that these extreme conditions
have spawned uniquely adapted life forms
on the seafloor.
Tim Shank grew up on the North Carolina coast and has been fascinated
with marine life for as long as he can remember. In the mid-1980s, his
professors at the University of North Carolina discovered hydrocarbon
seeps, sparking his interest in the evolution of life and chemosynthetic eco-
systems. Shank heeded the advice of his marine geosciences professor and
mentor, Dr. Conrad Neumann, who recommended that he study other
; passions, such as genetics or chemistry, and then apply them to marine
3 science research. After graduation, he worked in the molecular genetic
| environmental toxicology lab at the Environmental Protection Agency in
•f Research Triangle Park, N.C., honing his molecular skills for three years
— bejore entering graduate school at Rutgers University. He came to WHOI
as a Postdoctoral Scholar in 1999. Shank's research focuses on understanding the ecological factors
that affect the structure of diverse populations of deep-sea chemosynthetic species. He combines mo-
lecular genetic approaches and ecological field studies to understand the conditions and adaptations
that allow various species to migrate, evolve, and thrive in deep-sea habitats throughout the world's
oceans, including, more recently, seamounts. He has participated in more than 20 research cruises,
using lason and ABE, and has had more than 50 dives in Alvin. Shank's thirst for evolutionary his-
tory extends into another passion: genealogical and American history.
Woods Hole Oceanographic Institution 85
Living Large in Microscopic Nooks
Newly discovered deep-sea microbes rearrange thinking on the evolution of the Earth — and life on it
By Katrina Edwards, Associate Scientist
Marine Chemistry & Geochemistry Dept.
Woods Hole Oceanographic Institution
Between a rock and a hard place is the
proverbial worst spot for people to
find themselves in. But for certain deep-
sea microbes, it's the place to be. In 2000,
to our surprise, we found that microscopic
nooks and pits within volcanic seafloor
rocks harbor abundant colonies of previ-
ouslv unidentified microbes.
These microbes are different from
other microorganisms living in the sun-
less depths. They do not obtain the energy
they need to grow and multiply by metab-
olizing chemicals dissolved in seawater or
in hydrothermal fluids venting from the
seafloor. Instead, these newly discovered
microbes are living directly off minerals in
solid seafloor rocks.
The microbes are oxidizing iron in the
rocks, chemically altering the rocks, and
harnessing the energy produced by this
chemical reaction to live. Their discovery
has raised a slew of intriguing questions:
• Does our planet sustain abundant and
ubiquitous populations ot these microbes?
• Do they play a pivotal role in chemically
altering Earth's crust?
• Were they pioneering life forms on an
early Earth, which was largely devoid of
oxygen but full of iron?
• Do they exist on other iron-rich, oxygen-
-1 ••
RUST IN DAVEY JONES' LOCKER — Reddish-orange iron oxide (the same chemical compound we commonly refer to as "rust") coats the seafloor on
Loihi Seamount, an active underwater volcano 25 miles off the island of Hawaii. The material is made by an abundance of microbes that live and
grow by oxidizing iron directly from solid seafloor rocks. To study these newly discovered microbes, scientists have established FeMO — the Iron
(Fe)-oxidizing Microbe Observatory — on Loihi.
86 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.ed
poor planetary
bodies such
as Mars?
These previ-
ously inconspicu-
ous microorgan-
isms may turn out
to have starring
roles in shaping the
evolution of life on
Earth and other
planets, and shap-
ing the evolution of
the planet itself.
So why didn't we
notice them before?
Beyond the inher-
ent difficulties and
expense of search-
ing tor microor-
ganisms at the bottom of the ocean, the
answer is that we hadn't really looked for
them before. But now these easy-to-over-
look microbes have become hard to ignore.
Pumping iron on the seafloor
More and more, we are learning how
life on the Earth and the Earth itself—
biology and geology— are intimately inter-
twined and evolve together. Microbes
are ubiquitous catalytic agents, sparking
chemical reactions that alter the physi-
cal and chemical properties of their sur-
roundings. Beyond our scope of vision,
their cumulative metabolic activities play
a fundamental role in shaping and regulat-
ing our environment. (Our world would
be completely different, tor example, if
microorganisms did not continuously
decompose organic matter and transform
it back into inorganic material.)
A new field of study has arisen called
geomicrobiology. Scientists are now taking
a closer look at many unexplored regions
of our planet, and other planets, searching
for populations of unknown microbes that
may play major roles in cycling chemicals
through planetary systems.
In geomicrobiology, the borders
between rocks and living things are not
so ironclad. Many rocks are, however.
TRANSFORMING ROCK — Rusty-orange iron oxide coats the left side of this sample of seafloor
rock, where microbes have oxidized iron in the rock. They harness the chemical energy from
this reaction to live and grow. The microbes did not progress to the right side of the rock, which
remains its normal gray color.
and the microbes we found steal electrons
from iron atoms in the rock, changing
them from ferrous (Fe+2) to ferric (Fe+3).
With the energy produced by this chemi-
cal reaction, they convert carbon dioxide
(from seawater) into organic matter-
much the way plants and plankton use
solar energy and photosynthesis to accom-
plish the same.
Microscopic, but mighty
Iron is one of the most abundant and
reactive elements in the environment near
Earth's surface, so the discovery of iron-
oxidizing microbes raises the potential
that massive communities of them may
exist on Earth. If so, they could continu-
ally extract huge amounts of carbon diox-
ide from seawater and microscopically
exert a huge influence on ocean chemistry
over geologic time.
Does this large-scale drawdown of car-
bon dioxide from seawater help the oceans
absorb carbon dioxide, a critical green-
house gas, from the atmosphere? If so, it
would revise our understanding of how car-
bon cycles through the planetary system —
perhaps giving iron-oxidizing microbes an
important, previously unknown role in the
evolution of Earth's climate.
In their own way, the rise of micro-
scopic photosyn-
thetic plants caused
one of the most
devastating, perma-
nent alterations in
all of Earth's history.
They changed the
chemical compo-
sition of the near-
surface environ-
ment that all life
r- depended on, by
"v simply pumping
2 oxygen into Earth's
5 atmosphere.
*
Before then, nei-
ther the atmosphere
nor the oceans con-
tained much oxy-
gen, but the oceans
were filled with iron-rich rocks and tons
of dissolved iron. In such an iron-rich,
oxygen-poor environment, iron-oxidizing
microbes may have been dominant, pio-
neering life forms — a concept that com-
pels us to reassess our thinking about the
evolution of life on early Earth.
The existence of iron-oxidizing
microbes also redirects our search for
life elsewhere in the universe. Similar
microbes could have thrived, or still thrive,
in other iron-rich, oxygen-poor locales —
such as Mars, with its red, iron-rich soil, or
on the volcanic seafloor below the ice-cov-
ered ocean of lupiter's moon, Europa.
A search for unknown life
These unexpected new lines of inquiry
began in 2000 when former WHOI Post-
doctoral Scholar Tom McCollom and I,
with funding from the Mellon Founda-
tion and the National Science Foundation,
joined a research cruise aboard
R/V Atlantis off the Oregon coast.
Since the late 1970s, when hydro-
thermal vents were discovered, scientists
have focused on deep-sea chemosyn-
thetic microbes that derive energy from
dissolved hydrogen, hydrogen sulfide,
and methane emitted from these sites.
Though it is easier for microbes to draw
Woods Hole Oceanographic Institution 87
An experimental sample of seafloor rock is put back on the seafloor ...
it
energy from chemicals dissolved in sea-
water, WHOI biologist Carl Wirsen and
others found evidence ot sulfur-oxidiz-
ing bacteria that used solid minerals as
their only source of energy. (See "Is Life
Thriving Deep Beneath the Seafloor?"
page 72.)
Enormous amounts of sulfur and sul-
tides are found in vent chimney rocks, in
broken chimney rubble on the seafloor,
and in fine-grained mineral particles that
precipitate and "rain" out of plumes of
hydrothermal fluids spewing out of chim-
neys. We speculated that this little-rec-
ognized but potentially large source of
chemical energy may sustain important
microbial communities, which, in turn,
could play pivotal roles in altering the
chemistry of seafloor rocks and the
ocean itself.
Our goal in 2000 was to try to identify
unknown microbes that live off solid min-
erals and that might be mediating large-
scale geochemical changes on Earth.
The perfect niche for microbes
To explore what might be down there,
we used the submersible Alvin to place a
variety ot microbe-tree samples of natu-
ral seafloor rock back on the seafloor. Our
aim was to see what might "grow" on these
"blank slates."
WHOI geochemist Meg Tivey
retrieved our experimental samples for
us during an Alvin dive two months later
(See "The Remarkable Diversity of Sea-
floor Vents," page 60). To our surprise, we
found that many of the samples had thick
burnt-orange coatings of oxidized iron
(or "rust").
Using a scanning electron microscope,
we saw that the surfaces of the samples
were scarred with abundant pits and pores
less than 20 microns (0.0004 inches) deep
and wide. In these tiny pits were large
accumulations of corkscrew-shaped stalks
made of iron oxide, which created the
thick rusty coating.
Here's what we believe is happening:
Iron-oxidizing microbes exploit a niche
where the chemistry is just right. At first,
oxygen-loving microbes move into the
pits. They consume the available oxygen,
which is not replenished because seawater
does not readily flow into the restricted
pit areas.
That creates an ideal situation for the
iron microbes, which need low-oxygen
conditions. The tiny sheltered coves within
seafloor rocks contain just enough oxy-
gen from seawater for the iron microbes to
respire, but not an overwhelming amount
that would oxidize all the iron — with-
out microbial intervention — before the
microbes could use it.
As a byproduct of their iron-oxidizing
process, the microbes produce bundles of
iron-oxide stalks that resemble a little girl's
ringlets. These stalk accumulations effec-
tively cap the pits, maintaining the iron
microbes' preferred low-oxygen environ-
ment and securing their turf.
FeMO — a microbial observatory
The rapid proliferation and sheer
abundance of these iron microbes and
A BUCKETFUL OF DATA — WHOI scientists prepared plastic buckets filled with thin, microbe-free
samples of natural seafloor rock and placed them back on the seafloor. The experiment sought
to find out what might "grow" on these "blank slates. "To their surprise, the scientists found that
the samples were quickly colonized by intriguing microbes. (See diagram above.)
88 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
the quick chemical transformation of
the rocks they lived on were eye-open-
ing. Now we have mobilized research that
combines biology, chemistry, and geol-
ogy to explore many intriguing aspects of
these iron microbes.
Among the initial questions are: What
kinds of iron-oxidizing microbes are out
there? How many are there? How are they
making a living?
These species have been notoriously
difficult to grow in the laboratory and
therefore difficult to learn about. But in
our lab Dan Rogers and I, along with
WHOI biologist Eric Webb and others,
have made strides recently to culture and
interrogate these elusive microbes, and
we have begun to identify various species
of microbes and reveal their biochemical
machinery and metabolic capabilities.
Toward this end, we have just estab-
lished "FeMO" — an Iron (Fe)-oxidizing
Microbe Observatory — to study these
microbes at a site where they are diverse
and prolific. It is located at Loihi, an
active, submerged volcano, relatively con-
veniently located only 25 miles southwest
ot the big island of Hawaii.
To investigate the potential abun-
dance of iron microbes, WHOI geochem-
ist Wolfgang Bach and I analyzed rock
samples retrieved from an assortment of
holes drilled by the Ocean Drilling Pro-
gram into the exposed volcanic rock that
spreads out on both sides of the mid-
ocean ridge mountain chain encircling
the globe. We found that older rocks were
depleted of Fe+2 and full of Fe+3 — exactly
what iron-oxidizing microbes use up and
leave behind. The finding suggests that
mid-ocean ridge flanks represent mil-
lions of square miles ot fertile habitat for
iron microbes.
Life on early Earth and elsewhere
We have also begun to sequence
genomes of these microbes, in a proj-
ect with Mitch Sogin and Ashita Dhil-
lion at the Marine Biological Laboratory
in Woods Hole, funded by the National
Aeronautics and Space Administration's
Astrobiology Institutes Program (NAI).
These microbes are pioneers that prob-
ably lived billions of years ago on Earth
and may exist on other planetary bod-
ies. Identifying their genes, the enzymes
they produce, and the metabolic path-
ways these enzymes catalyze will reveal
an evolutionary heritage that will help us
unravel the emergence and development
ot lite on Earth and guide our search for
life elsewhere in the universe.
A key to reconstructing the evolution
of life on Earth and other planetary bodies
lies in the ability of scientists to read the
records, or "biosignatures," that long-dead
microbes leave behind in ancient or extra-
terrestrial rocks. To do that reliably, scien-
tists must be able to distinguish changes
caused by microbial activity from those
caused by abiotic oxidizing processes such
as rusting.
With this goal, scientists in our group,
including Bach, Postdoctoral Scholar
Olivier Rouxel, and graduate student Cara
Santelli, are advancing a range of new
approaches to gain understanding of how
microorganisms affect the microtextures,
isotopic chemistry, and history of the
rocks they interact with.
If we can unravel their story, these
long-neglected microbes will reveal a pro-
found tale about the co-evolution of Earth
and life.
Kutrina Edwards grew up in central Ohio, where she pursued an initial
early career in the family business of running a small municipal airport just
north of Columbus. She spent several years assisting her father and siblings
in general airport operations (graduating to the role of chief flight instruc-
tor), which she continued as she pursued a bachelor's degree in geology at
Ohio State University. Edwards then "retired" to attend the University of
Wisconsin, Madison, where she earned a Ph.D. in geomicrobiology — the
lirst degree in this field ever awarded by the university. Edwards and her
% family moved to Massachusetts in 1999 to join WHOI, where she estab-
=° lished a geomicrobiology lab. It focuses on "the tooth decay of the solid
Earth," she says, or more specifically, the transformation and degradation of Earth materials (rocks,
minerals, organic matter) by microbes. Edwards now enjoys deep-sea exploration, as long as some-
one else "flies" the submarine and she can focus on geomicrobiological research.
Woods Hole Oceanographic Institution 89
Shifting Continents and Climates
Sixty-five millions years ago, dino-
saurs had just become extinct, and
mammals were starting to dominate the
planet. Tropical conditions extended
to Northern Spain and the heartland
of North America. Large trees grew in
Greenland and Antarctica, and alligators
and primates could be found on Elles-
mere Island in Arctic Canada. Global
temperatures were 6° to 10°C (11° to
18°F) warmer than today, and the polar
regions were free of ice.
Since then, Earth's history has been
marked by a sustained and nearly con-
tinuous cooling trend, punctuated by
abrupt shifts and transitions. Today,
Homo sapiens dominate the landscape,
the poles are blanketed in ice, and over
the past 3 million years, massive conti-
nental glaciers have waxed and waned
in an ongoing era of ice ages. Our mod-
ern climate is a brief, temperate respite
from an otherwise cold cycle in Earth's
geologic life.
So how did our hothouse planet turn
into an icehouse planet?
Tectonic causes, climatic effects
One explanation for the change is
the steadily and substantially decreas-
ing levels of carbon dioxide and other
greenhouse gases in the atmosphere
(at least until the anomalous and very
recent post-Industrial Revolution era).
Less greenhouse gas means that less
heat is trapped in Earth's atmosphere.
But changes in Earth's atmosphere
cannot explain the full extent of global
cooling or periods of acute change. Nor
can scientists fully explain the causes of
the atmospheric changes themselves.
So what other forces or processes
might have rearranged Earth's climate
so dramatically?
In recent years, scientists have been
building a persuasive but still controver-
sial case that changes in the solid Earth
(the crust and mantle) spurred changes
in the liquid Earth (the oceans and
atmosphere). In other words, so-called
tectonic forces— the drifting and col-
lisions of Earth's tectonic plates— may
lead to climate changes.
Rising mountains, closing gateways
The following articles outline two
theories that link tectonic and climatic
changes. One theory, outlined by Gerald
Haug of the Eidgenossiche Technische
Hochschule (ETH) in Zurich, Switzer-
land, and colleagues, proposes that the
opening and closing of oceanic gateways
between land masses — a result of con-
tinental drift— may have altered global
ocean circulation patterns, which, in
turn, led to climate changes. Accord-
ing to another theory, outlined by Peter
Clift of Woods Hole Oceanographic
Institution, the uplift of great moun-
tain belts — caused by continental colli-
sions—may have disrupted atmospheric
circulation and triggered a cascade of
other climate changes.
"Understanding the links between
solid and liquid Earth systems is a first-
order scientific problem for the 21st cen-
tury," says Clift, a marine geologist.
The best evidence to reveal those links,
he notes, is buried under the seafloor.
—Mike Carhwicz
90 Ocean us Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
Moving Earth and Heaven
Colliding continents, the rise of the Himalayas, and the birth of the monsoons
By Peter Clift, Associate Scientist
Geology and Geophysics Department
Woods Hole Oceanographic Institution
Therefore will not we fear, though the Earth
be removed, and though the mountains be
carried into the midst of the sea. —Psaim46
As a geologist, I do not fear the pro-
cesses that carry Earth and mountain
into the sea. I rejoice in them.
The mountains rise, are lashed by wind
and weather, and erode. The rivers carry
mud and debris trom the mountains into
the ocean, where they settle onto the rela-
tively tranquil seafloor and are preserved.
The sediments bear evidence about where
they came from, what happened to them,
and when. By analyzing and dating these
seafloor sediments, scientists can piece
together clues to reconstruct when and
how fast their mountain sources rose to
great heights millions of years ago, and
how the climate and other environmental
conditions may have changed in response.
Linking mountains and monsoons
Tens of millions of years ago, a geo-
logical process was set in motion that
changed the planet. It produced some of
the world's most dramatic and extensive
mountain ranges. It probably created one
of the planet's most intense and impor-
tant climate phenomena — the Asian mon-
soons—which today pace and undergird
the health and welfare of billions of people
in South and East Asia, two-thirds of the
total population on the planet. And it may
have provoked large-scale environmental
changes in the past that brought hominids
out of trees and upright onto two feet.
All of these developments in recent
Earth history ultimately may be attributed
to the land masses now known as India
and Arabia, which began moving north
some 100 million years ago, on a collision
course with what is now Eurasia. Accord-
ing to plate tectonic theory, Earths crust is
composed of interlocking, moving oceanic
and continental plates. Scientists consider
the collision of the Indian and Eurasian
Plates the classic example of how plate
tectonics can alter the circulation ot the
oceans and atmosphere. Here's the hypo-
thetical sequence of events:
The birth of the monsoons
Before the Indian and Eurasian Plates
collided, an ancient ocean called the
Tethys, lay between Eurasia and Africa.
By about 55 million years ago, the conti-
nents squeezed out the ocean, and some
research suggests that the resulting rear-
rangement of ocean currents may have
provoked the strong global warming that
came shortly after.
As India smashed into Asia, the world's
tallest mountain ranges were thrust up
like the hood of a car in a head-on colli-
sion. On the Indian Plate, the Himalaya
Mountains were formed, spanning Paki-
stan, India, Nepal, and Bhutan. The Indian
HIGH AND MIGHTY — A view ofPangong Lake in the Ladakh region of Northern India, taken at an altitude of 18,000 feet, shows the great flat,
2-mile-high expanse of the Tibetan Plateau, extending in the background as far as the eye can see.
Woods Hole Oceanographic Institution 91
Colliding continents
1 20 million years ago
The Indian subcontinent was part of a
supercontinent called Gondwana. The
ancient Tethys Ocean existed between
the South American/African and Eurasian
supercontinents.
60 million years ago
The Indian subcontinent, moving toward
Asia at a speed of 10 centimeters per year,
heads toward a collision about 50 million
years ago.
Today
The India-Asia collision has closed
the ancient Tethys Ocean, created the
Himalayan, Karakoram, and Hindu Kush
mountain ranges, and uplifted the great
Tibetan Plateau.
Plate was shoved under the Eurasian Plate,
uplifting the Karakoram and Hindu Kush
Mountains in Afghanistan and Pakistan,
as well as the great Tibetan Plateau — an
expanse about 4.5 kilometers high and halt
the size of the continental United States.
The creation of this dramatic continental
topography launched a cascade of plan-
etary changes.
The Tibetan Plateau acts like a gigan-
tic exposed brick, absorbing summer heat
and heating the atmosphere above it. Hot
air rises, and cool, moist air — drawn in
from over surrounding oceans — rushes in
to replace it. That moist air is the source of
monsoon rains.
New evidence suggests that between
22 million and 15 million years ago, the
Asian monsoons may have begun to
strengthen. The onset ot the monsoons
may have been triggered when the Tibetan
Plateau reached a threshold height of 2 to
3 kilometers (1.2 to 1.8 miles).
Removing CO2 from the atmosphere
As the mountains rose upward, the
land became more exposed to the forces
ot weather and gravity. Rainwater contains
acids that chemically react with rocks.
In the process, called chemical weather-
ing, carbon dioxide is drawn out ot the
atmosphere and converted into carbonate
in rocks. As the monsoons strengthened,
chemical weathering increased.
As the mountains rose and monsoon
rains increased, rivers also swelled and cut
more deeply into the mountains, increas-
ing erosion and carrying more sediments
into the oceans. To give a sense of scale,
the Indus River today deposits about 1,000
million tons of mud and sand each year
onto the Indus Submarine Fan in the Ara-
bian Sea. Relieved of such massive sedi-
mentary weight, the mountains could be
thrust up higher, in a reinforcing cycle
that continued to increase monsoons, ero-
sion, and uplift.
Evolving climates
Climatically, research suggests that the
increasing rates of weathering and erosion
of the mountains converted large volumes
of carbon dioxide from the atmosphere
into carbonate sediments that eventu-
ally were deposited on the ocean bottom.
As carbon dioxide was drawn out of the
atmosphere, the global greenhouse effect
The rise of the Himalayas and the Tibetan Plateau
i the India-Asia collision, the Eurasian Plate was compressed and thickened to uplift the Tibetan Plateau. The bulk of the Indian Plate
continues to be thrust under the Eurasian Plate, further uplifting Tibet. Slices of the Indian Plate were scraped off to form the Himalayas.
92 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
How to make a monsoon
3 Cool, moist air is drawn in from over sur-
rounding oceans to replace the rising hot air.
The moist air is the source of monsoon rains.
2 Hot air rises high
above the Tibetan
Plateau.
4 Monsoon rains en-
gorge rivers flowing
from high mountains,
which carry tons of
sediments to the
1 The Tibetan Plateau
acts like a gigantic
exposed brick,
absorbing summer
heat and heating
the atmosphere
above it.
was reduced, setting the stage for long-
term cooling of the planet that culminated
in the ice ages of the last 2.7 million years.
In addition, an influx of chemical nutri-
ents into the ocean may have sparked
blooms ot phytoplankton. Microscopic
marine plants also extract atmospheric car-
bon dioxide via photosynthesis and convert
it into carbonate organic matter that settles
to the seafloor when the plankton die.
As monsoon winds strengthened, they
blew waters laterally across the ocean sur-
face. To replace these waters, cooler, nutri-
ent-rich waters upwelled from the depths
to the sunlit surface, providing all the
ingredients plankton need to thrive. Evi-
dence from seafloor sediment cores shows
an abundance of preserved microscopic
shells of plankton (and, by inference, a
strengthening of the monsoons) beginning
about 8.5 million years ago in the Arabian
Sea, though the situation in other parts of
Asia is less clear.
Evolving humans
Curiously, that same time period marks
crucial events in human evolutionary his-
tory. The cumulative effects of decreasing
atmospheric carbon dioxide reduced the
global greenhouse effect, creating a much
colder and drier Earth.
About 8 million years ago, paleonto-
logical evidence shows that the great apes
became extinct in Europe and Asia — vic-
tims of a colder, drier climate. They main-
tained populations only in Southeast Asia
and Africa.
About 7 million to 8 million years ago,
humans began to diverge from great apes
in Africa, which, though still equatorially
warm, began to dry out. Jungles and for-
ests turned into grasslands and deserts. The
climate shift provided evolutionary advan-
tages for bipedal hominids with larger
brains to cope with environmental changes.
Continental clues are erased
All this is an intriguing, but specula-
tive, theory.
Unfortunately, our detailed theoreti-
cal understanding of Earths climatic evo-
lution is not matched by a sufficiently
detailed record of the evolution of Tibetan
and Himalayan uplift and erosion. Theo-
ries concerning the uplift of Tibet set the
start date anywhere from 65 million years
ago to as recently as 2 million years ago.
This lack of consensus principally
reflects the lack of a good continental
geological record to chart the growth
ot the mountains. Because of chemical
weathering, continental sediments are
difficult to date, and erosion often wipes
away large and critical portions of the
record, destroying any hope for a contin-
uous chronology.
Fortunately, marine sediments pre-
serve robust, continuous records that
can link tectonic and climatic evolution.
From deep-sea cores, marine geologists
k,
FLOWING TO THE SEA — An aerial view of the Ganges River Delta shows tons of sediments being
poured into the northern Bay of Bengal off India.
Woods Hole Oceanographic Institution 93
FROM THE MOUNTAINTOPS TO THE BOTTOM OF THE OCEAN— Great rivers (the Ganges,
Brahmaputra, and Indus) transport large volumes of sediments from great mountains (the
Himalayas, the Hindu Kush and the Karakoram) into the ocean. The Bengal Fan extends 2,500
kilometers ( 7,535 miles) south into the Bay of Bengal and is 22 kilometers (13.5 miles) thick.
The Indus Fan is 1 0 kilometers (6 miles) thick and extends 1,000 to 1,500 kilometers (610 to 91 5
miles) into the Arabian Sea.
have pieced together a detailed record of
environmental change in Asia and Africa.
Many of those cores have come from the
Arabian Sea, the South China Sea, and the
Bay of Bengal, which offer fertile territory
for examining the interacting histories of
the solid Earth and its climate.
The Tibet-Himalaya region is drained
by some of the most vigorous rivers on
Earth. The Ganges and Brahmaputra River
systems transport large volumes ot detri-
tus from the rapidly eroding Himalayas
and deposit them in the Bengal Fan in the
Indian Ocean. The Bengal Fan is the larg-
est sediment body on the planet; it runs
2,500 kilometers (1,535 miles) south into
the Bay of Bengal and is 22 kilometers
(13.5 miles) thick.
The modern Indus River system drains
sediments from the high peaks of the
Karakoram, Hindu Kush, and Western
Tibet. It has created the 10-kilometer-
thick Indus Fan, which extends 1,000 to
1,500 kilometers into the Arabian Sea.
Located between the land masses of
Arabia and the Indian subcontinent, the
Arabian Sea is ideally placed to record the
effects of India's collision with the Asian
mainland. New data now suggest that the
Indus River and Fan system was initiated
shortly after the India-Asia collision about
55 million years ago, probably in response
to the initial uplift ot Tibet. This long his-
tory makes it a natural storehouse ot infor-
mation on how the mountains developed
over long time periods.
An archive buried on the seafloor
Studying this rich store of deep-sea
sediments, I have been able to estimate
that erosion increased around 16 million
years ago, somewhat earlier than the start
of the monsoon pattern. That pulse of ero-
sion appears to have been the result of
mountain building in the Karakoram.
Conversely, the record shows a
decrease in sedimentation rates about
5 million to 7 million years ago. Some
researchers have proposed this may be
related to the strengthening ot the mon-
soons. Increased monsoon rainfall, they
speculate, might have promoted the
growth of vegetation, stabilizing the slopes
and reducing erosion. Other scientists,
including me, believe this period was
drier, with less rainfall, less erosion, and
less seafloor sedimentation.
The marine sediments ot the Arabian
Sea and the Bay of Bengal hold the prom-
ise of allowing ocean scientists to make
direct correlations among the evolution
of the monsoon, the uplift of the Tibetan
Plateau, and the erosion of the plateau
by heavy monsoonal winds and rains.
With further deep-sea scientific drill-
ing and marine seismic research to reveal
sub-seafloor sedimentary layers, we can
expose this record in detail and discover
how Earths tectonic, climate, and perhaps
human, evolutions are all linked.
Peter Clift grew up far from the Himalayas, just outside London, England.
After degrees at Oxford and Edinburgh Universities and some very pleasant
field research in Greece to escape the English rain, Peter spent tune explor-
ing in central Asia, traveling overland from Pakistan to Beijing. Smitten
- with the travel bug, he made his first expedition to the Indian sub-conti-
§ nent where he began to reconstruct the history of mountain growth from
~ the sediments he found there. Shortly after, an invitation to sail on a cruise
~= in the Southwest Pacific led to a new love of sea-going science and an inter-
t est in the interactions between the land and oceans. He worked with the
_^___ £ Ocean Drilling Program at Texas A&M University before coming to Woods
Hole in 1 995 as an Assistant Scientist. When he is not tending to his fuzzy family of cats and birds
and cycling around town, Clift continues to do fieldwork in remote parts of Asia more often associ-
ated these days with CNN than scientific journals.
94 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edu
How the Isthmus of Panama Put Ice in the Arctic
Drifting continents open and dose gateways between oceans and shift Earth's climate
By Gerald H. Haug, Geoforschungszentrum
Potsdam (GFZ), Germany; Ralt'Tiedemann,
Forschungszentrum fur Marine Geowissen-
schaften, Germany; and Lloyd D. Keigwin,
Woods Hole Oceanographic Institution
The long lag time has always puzzled
scientists: Why did Antarctica become
covered by massive ice sheets 34 million
years ago, while the Arctic Ocean acquired
its ice cap only about 3 million year ago?
Since the end ot the extremely warm,
dinosaur-dominated Cretaceous Era
65 million years ago, heat-trapping green-
house gases in the atmosphere have steadily
declined (with the anomalous exception ot
the last century), and the planet as a whole
has steadily cooled. So why didn't both
poles freeze at the same time?
The answer to the paradox lies in the
complex interplay among the continents,
oceans, and atmosphere. Like pieces of a
puzzle, Earth's moving tectonic plates have
rearranged themselves on the surface of
the globe — shifting the configurations of
intervening oceans, altering ocean circula-
tion, and causing changes in climate.
The development of ice sheets in the
Southern Hemisphere around 34 million
years ago seems fairly straightforward.
The supercontinent of Gondwana broke
apart, separating into subsections that
became Africa, India, Australia, South
America, and Antarctica. Passageways
opened between these new continents,
allowing oceans to flow between them.
When Antarctica was finally severed
from the southern tip of South America
to create the Drake Passage, Antarctica
became completely surrounded by the
Southern Ocean. The powerful Antarctic
Circumpolar Current began to sweep all
the way around the continent, effectively
isolating Antarctica from most of the
warmth from the global oceans and pro-
voking large-scale cooling.
The Northern Hemisphere is more
problematic. From sediment cores and
other data, we know that until about 5
million years ago, North and South Amer-
ica were not connected. A huge gap— the
Central American Seaway— allowed tropi-
cal water to flow between the Atlantic and
Pacific Oceans.
A growing body ot evidence suggests
that the formation of the Isthmus of Pan-
ama partitioned the Atlantic and Pacific
Oceans and fundamentally changed global
How Antarctica got its ice sheets
In the continual movement of Earth's tectonic plates, Antarctica was severed from the southern tip of South America about 34 million years
ago, creating the Drake Passage. Antarctica became completely surrounded by ocean. The powerful Antarctic Circumpolar Current began
to sweep around the continent, isolating Antarctica from the warmth of the global oceans and provoking large-scale cooling.
Woods Hole Oceanographic Institution 95
ocean circulation. The closing of the Cen-
tral American Seaway initially may have
warmed Earths climate, but then set the
stage for glaciation in the Northern Hemi-
sphere at about 2.7 million years ago.
The Ocean Conveyor
A fundamental element of todays cli-
mate system is a conveyor-like ocean cir-
culation pattern that distributes vast quan-
tities of heat and moisture around our
planet. This global circulation is propelled
by the sinking of cold, salty— and there-
fore dense — ocean waters.
In today's ocean, warm, salty surface
water from the Caribbean, the Gulf of
Mexico, and the equatorial Atlantic flows
northward in the Gulf Stream. As the
warm water reaches high North Atlan-
tic latitudes, it gives up heat and mois-
ture to the atmosphere, leaving cold, salty,
dense water that sinks to the ocean floor.
This water flows at depth, southward and
beneath the Gulf Stream, to the South-
ern Ocean, then through the Indian and
Pacific Oceans. Eventually, the water
mixes with warmer water and returns to
Today's climate system is influenced by the
ocean's conveyor-like global circulation. Cold,
salty waters sink to drive the conveyor, and
warm surface currents complete the loop.
the Atlantic to complete the circulation.
The principal engine of this global cir-
culation, often called the Ocean Conveyor,
is the difference in salt content between
the Atlantic and Pacific Oceans. Before the
Isthmus of Panama existed, Pacific sur-
face waters flowed into the Atlantic. Their
waters mixed, roughly balancing the two
oceans' salinity.
About 5 million years ago, the North
American, South American, and Carib-
bean Plates began to converge. The grad-
ual shoaling of the Central American
Seaway began to restrict the exchange of
water between the Pacific and Atlantic,
and their salinities diverged.
Evaporation in the tropical Atlantic
and Caribbean left ocean waters there
saltier and put fresh water vapor into the
atmosphere. The Trade Winds carried the
water vapor from east to west across the
low-lying Isthmus of Panama and depos-
ited fresh water in the Pacific through
rainfall. As a result, the Pacific became
relatively fresher, while salinity slowly and
steadily increased in the Atlantic.
As a result of the Seaway closure, the
Gulf Stream intensified. It transported
more warm, salty water masses to high
northern latitudes, where Arctic winds
cooled them until they became dense
enough to sink to the ocean floor. The
Ocean Conveyor was rolling, drawing
even more Gulf Stream waters northward.
How does this make ice in the north?
Peter Weyl of Oregon State Univer-
sity hypothesized in 1968 that the closure
of the Central American Seaway and the
intensification of the Gulf Stream would
The rise of the Isthmus of Panama and the closing of the Central American Seaway
Surface waters flowed from the Pacific
into the Atlantic 1 0 million years ago
via an ocean gateway called the Central
American Seaway, and both oceans had
the same salinity.
About 5 million years ago, the North
American, South American, and Caribbean
Plates converged. The rise of the Isthmus
of Panama restricted water exchange
between the Atlantic and Pacific, and their
salinities diverged. The isthmus diverted
waters that once flowed through the
Seaway. The Gulf Stream began to intensify.
Today, evaporation in the tropical Atlantic
and Caribbean leaves behind saltier ocean
waters and puts fresh water vapor into
the atmosphere. Trade Winds carry the
water vapor westward across the low-lying
isthmus, depositing fresh water into the
Pacific through rainfall. Asa result, the
Atlantic is saltier than the Pacific.
96 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edi
have brought a critical ingredient tor
ice sheet growth to the Northern Hemi-
sphere—moisture. Weyl's theory assumed
that the closure of the Central Ameri-
can Seaway and the buildup of salt in the
Atlantic coincided with the growth of
northern ice sheets between 3.1 and 2.7
million years ago.
But doubts about this hypothesis sur-
faced in 1982, when Lloyd Keigwin of
WHOI found evidence in ocean sediments
that the closing ot the Isthmus of Panama
had influenced ocean circulation more
than a million years earlier. He demon-
strated that the salinity contrast between
the Atlantic and Pacific had already
started to develop by 4.2 million years ago.
In 1998, Gerald Haug and Ralf Tiede-
mann confirmed Keigwin's research with
higher-resolution data from sediment
cores. If the salinity had already changed
by 4.2 million year ago, why didn't gla-
ciation start until 2.7 million years ago?
On the contrary, the Earth experienced a
warm spell between 4.5 million and 2.7
million years ago.
That global warm spell, called the Mid-
Pliocene Warm Period, may also have
been related to the closing of the Central
American Seaway and the consequent
rearrangement of global ocean circulation.
An invigorated Ocean Conveyor could
have driven a stronger flow of deep waters
from the Atlantic to the North Pacific
Ocean, which is the end ot the line tor
deep-ocean circulation.
On their journey to the North Pacific,
these deep waters became enriched in
nutrients and carbon dioxide. In the Sub-
arctic Pacific, these deep waters could have
upwelled, rising to the sunlit surface to
provide the ingredients to spark enormous
blooms ot phytoplankton. Great abun-
dances ot silica and opal (the preserved
material from the phytoplankton shells)
Short-circuiting the Ocean Conveyor
What factors caused the Northern Hemisphere to develop sea ice and glaciers about 3 million years ago? The closing of the Central Ameri-
can Seaway intensified the Gulf Stream, the upper limb of a conveyor-like global ocean circulation called the Ocean Conveyor. The Gulf
Stream transported salt to the North Atlantic, making waters there dense enough to sink and drive the lower limb of the Conveyor. Accord-
ing to one theory, the Gulf Stream also transported moisture to the North Atlantic region and encouraged ice formation. But the Gulf Stream
also conveys heat, which would deter glaciation. Neal Driscoll and Gerald Haug proposed a solution to this apparent contradiction (below).
I
Woods Hole Oceanographic Institution 97
in seafloor sediments are evidence of both
the blooms and the strong upwelling.
The upwelling may have been so
strong, however, that the phytoplankton
did not keep pace; that is, more carbon -
dioxide-rich water from deeper ocean
regions upwelled than the phytoplank-
ton could use. Consequently, the excess
carbon dioxide "leaked" back into the
atmosphere, adding a greenhouse gas that
warmed the planet.
Short-circuiting the Conveyor
What shut down the Mid-Pliocene
Warm Period about 2.7 million years
ago? And what finally caused Northern
Hemispheric glaciation at about the same
time — but nearly 2 million years after the
Isthmus ot Panama termed?
Weyl's original theory of a stron-
ger, moisture-laden Gulf Stream raised
another nettlesome question: How could
the Gulf Stream — which transports not
only moisture but also heat to the North
Atlantic — lead to major Northern Hemi-
sphere cooling and the formation of ice?
Gerald Haug and Neal Driscoll of
Scripps Institution of Oceanography
proposed one solution. They postulated
that moisture carried northward by the
Gulf Stream was transported by prevail-
ing westerly winds to Eurasia. It fell as
rain or snow, eventually depositing more
fresh water into the Arctic Ocean — either
directly, or via the great Siberian rivers
that empty into the Arctic Ocean.
The added fresh water would have
facilitated the formation of sea ice, which
would reflect more sunlight and heat back
into space. It would also act as a barrier
blocking heat stored in the ocean trom
escaping to the atmosphere above the Arc-
tic. Both these phenomena would further
cool the high latitudes. In addition, Arc-
tic waters flowing back into the North
Atlantic would have become less cold and
salty — short-circuiting the efficiency of
the Ocean Conveyor belt as a global heat
pump to North Atlantic regions.
The tilt toward glaciation
These preconditions— moisture plus
an Arctic nucleus for cooling— would
have made the climate system highly sus-
ceptible to ice sheet growth. Even modest
changes in the global environment would
have been sufficient to tip the scales and
lead to the onset of major Northern Hemi-
sphere glaciation.
Just such a change occurred between
3.1 million and 2.5 million years ago, as
Earth's axis fluctuated so that the planet's
I
Gerald Hang studied marine geology at Kiel University, Germany. He was
then a postdoctoral scholar at the University of British Columbia, in Van-
couver, Canada, and at WHOI in Lloyd Keigwin's lab. In 1998-99, lie was a
researcher at the University of Southern California in Los Angeles. Between
2000 and 2002, he did his 'Habitation at ETH Zurich, Switzerland, and
was an adjunct scientist at WHOI. Since 2003, he has been a scientist at
Geoforschungszentrum Potsdam with a joint appointment as professor at
Potsdam University, Germany. Gerald benefited tremendously from his tune
at WHOI, both scientifically and in his passion for old cars, which he shares
with co-author Keigwin
Won/ Keigwin studied geology at Brown University and oceanography at the
Univeristv of Rhode Island. At Brown he was in the ROTC and after gradu-
ation, served on active duty for two years as a line officer on a destroyer es-
cort in Newport, R.I. He maintained his affiliation with the Naval Reserve,
retiring after 30 years as Captain USNR. At Brown, his interest in geology
was stimulated by Leo Laporte, and his introduction to oceanography was
through 70/in Imbrie. Keigwin's research interests have evolved from study-
ing ocean and climate changes that occurred millions of years ago to those
that occurred in recent centuries. Noting this progression, a National Science
Foundation prognin: manager quipped, "Soon he will be studying the future, and we won't have to
fund him any longer! ' \ \ hen not busy with family or having fun doing science, Keigwin works in his
barn, trying to keep old sports cars running.
tilt toward the sun was less than today's
angle of 23.45 degrees. Less tilt to the
Earth would have reduced the amount
and intensity of solar radiation hitting
the Northern Hemisphere, leading to
colder summers and less melting of win-
ter snows.
The onset of Northern Hemisphere
glaciation also affected the Subarctic
Pacific. It led to the formation about 2.7
million years ago of a freshwater lid at the
surface of the ocean, called a halocline.
This Arctic halocline would have created
a barrier to upwelling, which blocked
deep carbon-dioxide-rich deep waters
from rising to the surface. The "leak" of
heat-trapping carbon dioxide into the
atmosphere was stemmed, further cool-
ing the planet.
Many other ocean-atmosphere feed-
back mechanisms, resulting from the
opening and closing of oceanic gateways,
remain imperfectly understood. And sci-
entists are also exploring the ramifications
of other oceanic gateways.
Mark Cane of Columbia University
and Peter Molnar of the University of
Colorado, for instance, have suggested
that the uplifting and movement of the
Indonesian Islands between 5 million and
3 million years ago would have funda-
mentally redirected a smaller amount of
warm South Pacific water and a greater
amount of cooler North Pacific water
through the Indonesian Seaway. The con-
sequence might have been that the Pacific
changed from more permanent El Nino-
like conditions (which move heat from
the tropics to high latitudes) to a more La
Nina-like state (which would have cur-
tailed the heat transfer and cooled the
Northern Hemisphere).
The lessons from these vast geologic
and geographic changes are both elegantly
simple and excruciatingly complex. The
opening and closing of seaways has a pro-
found influence on the distribution of
fresh water, nutrients, and energy in the
global ocean. The coupling of these chang-
ing oceans with a changing atmosphere
inevitably means a changing climate.
98 Oceanus Magazine • Vol. 42, No. 2 • 2004 • oceanusmag.whoi.edi
MB1. WHO! I.1BKAKY
UH 1S3X A
The Ocean Institutes
In 2000, Woods Hole Oceanographic
Institution established four Ocean
Institutes to accelerate advances in
knowledge about the oceans and to
convey discoveries expeditiously into
the public realm. The Ocean Institutes'
goals are to catalyze innovative thinking
that can open up new scientific vistas,
to spur collaboration among scientists
in different disciplines, and to stimulate
a rich and productive educational envi-
ronment that will engage future lead-
ers of oceanography. Concurrently, the
Institute's mission is to shorten the time
between acquiring knowledge and mak-
ing it accessible to decision-makers who
can use this information to benefit soci-
ety and steward the Earth.
'.
The Deep Ocean Exploration Institute investigates Earths
dynamic processes — beneath the oceans where more than 80 percent
of all earthquake and volcanic activity occurs and where the clues to
understanding the inner workings of our planet lie. The seafloor is our
window into the dynamic, fundamental processes that generate natu-
ral disasters, produce oil and mineral resources, create and destroy
oceans, rend continents, build mountains and islands, and foster life.
The Deep Ocean Exploration Institute:
• explores how our dynamic planet evolves and changes
• examines the basic forces that create earthquakes and volcanoes
• develops technology related to seafloor observatories and
deep-submergence vehicles
• investigates unusual chemosynthetic communities of lite on and
below the seafloor
• explores potential new energy and mineral resources in the oceans
The Ocean Life Institute explores the oceans extraordinary diver-
sity of life — from microbes or whales — to identify ways to sustain
healthy marine environments and to learn about the origin and evo-
lution of life on Earth. The more we look into the oceans, the more
we find remarkable life forms thriving in environments ranging from
Antarctic sea ice to the volcanic crust below the seafloor.
The Ocean Life Institute:
. explores biodiversity in the oceans
• finds ways to monitor and sustain the health of marine ecosystems
• studies marine life's physiological and ecological adaptations
• investigates the evolution of life in Earths oceans
• develops new techniques and instruments to explore ocean life
The Ocean and Climate Change Institute seeks to understand the
role ot the ocean in regulating Earths climate and to improve our abil-
ity to forecast future climate change. The ocean stores vast quantities
of heat, water, and carbon dioxide and works with the atmosphere in
regulating global and regional climates — on time scales ranging from
days (storms and hurricanes), seasons (monsoons), years (El Ninos),
to centuries and longer.
The Ocean and Climate Change Institute:
• identifies the climatic effects of ocean circulation patterns
• develops an ocean-monitoring network to forecast climate changes
• examines geological records to better understand ocean behavior
• studies ocean dynamics that may trigger large, abrupt climate shifts
• evaluates the ocean's response to the buildup of greenhouse gases
The Coastal Ocean Institute examines one of the most vital — and
vulnerable — regions on Earth: the coast. Our planets exploding pop-
ulation has begun to put stress on the fragile coastal ocean and has
exposed more people to coastal hazards such as storms, beach erosion,
and pollution. Understanding the complex, delicately balanced pro-
cesses at work in coastal areas is the key to ensuring that they remain
productive and attractive.
The Coastal Ocean Institute:
• reveals basic processes underlying the coastal oceans fertility
• provides sound science to guide coastal management policies
• examines uses of coastal resources, such as wind, oil, and fisheries
• identifies strategies to mitigate coastal hazards and natural disasters
• promotes awareness of the coastal zone's importance to society
Woods Hole Oceanographic Institution 99