constructional features that either are precipitated
from the hydrothermal fluids themselves when they
encounter cold, alkaline sea water or are formed by
precipitation due to the mixing of hydrothermal fluids
with sea water, or, and more usually, both of these
processes. Although these chimneys comprise metal
sulfides and sulfates, other minerals are also present,
usually in lesser quantities, although on sedimentcovered ridges carbonates, for example, may also be
abundant in these chimney or mound structures. Lowtemperature fluids, generally p351C, but sometimes
approaching 1001C, have usually deposited (if they
ever contained) much of their metal load below the
seafloor and hence usually neither smoke nor have
specific constructional features associated with their
fluid flow. Because of this lack of specific structure, the
fluid flow is often less organized, and hence is referred
to as ‘diffuse.’ Some authors use this term to refer to
fluids exiting directly from the basaltic substrate.
These sites are usually well-colonized by various types
of vent megafauna, making it difficult, if not impossible, to identify a specific orifice. Other authors
use this same term to refer to fluids that are oozing out
of large sulfide structures. At some sites, only hightemperature ‘black smoker’ vents are found, while at
other sites only the lower-temperature diffuse venting
is found (as at the Galapagos Spreading Center where
venting was first discovered). At other sites, both focused and diffuse flow may occur right next to one
another, and, at least at one site, we have now observed low-temperature diffuse flow evolving into
high-temperature focused flow over several years. For
all types of ‘low temperature and/or diffuse flow,’ all of
the fluids that have been sampled to date appear to
contain some fluid that has reacted at significantly
greater temperatures than are measured directly in
these fluids. Hence one could argue that, at least at the
ridge axes, all fluids are ultimately high-temperature
fluids and that some of them have undergone significant mixing with sea water at some depth, likely
relatively shallow, within the oceanic crust. With the
continued inability to drill these systems, our knowledge of their subsurface hydrology is rudimentary at
best. One might note that fluids with temperatures in
the B100–2001C range have been left out of the above
classification. This is because there are few fluids that
have been sampled in this temperature range, which
may reflect a true lack of abundance of fluids at such
temperatures, or simply a sampling gap.
The Influence of Volcanic Events
In the late 1980s there was a real dichotomy in our
understanding of the controls on seafloor hydrothermal systems. We knew of individual vents that
were chemically stable for years at a time, yet each
individual vent had a distinctive chemistry, and rapid
changes in the hydrothermal plumes above these sites
had been noted. In 1991 some of this puzzle was
resolved with the first opportunity to sample a midocean ridge hydrothermal site immediately following
a volcanic eruption and to study the evolution of this
site on timescales now approaching decadal. The
discovery of an eruption on the East Pacific Rise at
9145–52
0 N was fortuitous, but it could be sampled
and documented with the DSV Alvin within weeks of
the volcanic event. Since that time using the US
Navy’s SOSUS system, it has been possible to determine real-time when some of these events are occurring on the Juan de Fuca and Gorda Ridges, but
immediate response to these events have been limited
to surface ship observations due to logistical constraints. The events studied so far have been characterized by a volcanic eruption at the seafloor, the
intrusion of a dike, or both. The knowledge gained
from responding to these events has revolutionized
our understanding of these systems, especially their
pronounced temporal variability on very short
timescales (much less than days to weeks). Presumably we will be able to study more of these events
and to gain a sense of their frequency, perhaps as a
function of spreading rate, as we begin to instrument
more of the ridge crest with hydrophones to detect
the T-phase signals associated with these events.
The Influence of Tectonic/Cracking Events
Magmatic events have provided new insights into the
processes that drive hydrothermal systems and their
fluid compositions on intermediate- to fast-spreading
ridges, but presumably tectonic events are more important (or at least more frequent than magmatic
events) on slow-spreading ridges. As we have not yet
been able to observe one of these events, we cannot
assess their importance. In a relatively small cracking
event on a fast-spreading ridge observed with a
seismic array, changes in fluid temperatures were
marked, and changes in fluid compositions were
profound, leading to major changes in the biological
communities existing at this site. Presumably the
observation of one or more tectonic events on slowspreading ridges will also provide critical new insights into how hydrothermal systems on these types
of ridges function and evolve.
On-axis versus Off-axis
The discussion above has focused on the axial component of midocean ridge hydrothermal systems.
Many debates have focused on the importance of
these axial systems compared to hydrothermal
HYDROTHERMAL VENT FLUIDS, CHEMISTRY OF 85
from the hydrothermal fluids themselves when they
encounter cold, alkaline sea water or are formed by
precipitation due to the mixing of hydrothermal fluids
with sea water, or, and more usually, both of these
processes. Although these chimneys comprise metal
sulfides and sulfates, other minerals are also present,
usually in lesser quantities, although on sedimentcovered ridges carbonates, for example, may also be
abundant in these chimney or mound structures. Lowtemperature fluids, generally p351C, but sometimes
approaching 1001C, have usually deposited (if they
ever contained) much of their metal load below the
seafloor and hence usually neither smoke nor have
specific constructional features associated with their
fluid flow. Because of this lack of specific structure, the
fluid flow is often less organized, and hence is referred
to as ‘diffuse.’ Some authors use this term to refer to
fluids exiting directly from the basaltic substrate.
These sites are usually well-colonized by various types
of vent megafauna, making it difficult, if not impossible, to identify a specific orifice. Other authors
use this same term to refer to fluids that are oozing out
of large sulfide structures. At some sites, only hightemperature ‘black smoker’ vents are found, while at
other sites only the lower-temperature diffuse venting
is found (as at the Galapagos Spreading Center where
venting was first discovered). At other sites, both focused and diffuse flow may occur right next to one
another, and, at least at one site, we have now observed low-temperature diffuse flow evolving into
high-temperature focused flow over several years. For
all types of ‘low temperature and/or diffuse flow,’ all of
the fluids that have been sampled to date appear to
contain some fluid that has reacted at significantly
greater temperatures than are measured directly in
these fluids. Hence one could argue that, at least at the
ridge axes, all fluids are ultimately high-temperature
fluids and that some of them have undergone significant mixing with sea water at some depth, likely
relatively shallow, within the oceanic crust. With the
continued inability to drill these systems, our knowledge of their subsurface hydrology is rudimentary at
best. One might note that fluids with temperatures in
the B100–2001C range have been left out of the above
classification. This is because there are few fluids that
have been sampled in this temperature range, which
may reflect a true lack of abundance of fluids at such
temperatures, or simply a sampling gap.
The Influence of Volcanic Events
In the late 1980s there was a real dichotomy in our
understanding of the controls on seafloor hydrothermal systems. We knew of individual vents that
were chemically stable for years at a time, yet each
individual vent had a distinctive chemistry, and rapid
changes in the hydrothermal plumes above these sites
had been noted. In 1991 some of this puzzle was
resolved with the first opportunity to sample a midocean ridge hydrothermal site immediately following
a volcanic eruption and to study the evolution of this
site on timescales now approaching decadal. The
discovery of an eruption on the East Pacific Rise at
9145–52
0 N was fortuitous, but it could be sampled
and documented with the DSV Alvin within weeks of
the volcanic event. Since that time using the US
Navy’s SOSUS system, it has been possible to determine real-time when some of these events are occurring on the Juan de Fuca and Gorda Ridges, but
immediate response to these events have been limited
to surface ship observations due to logistical constraints. The events studied so far have been characterized by a volcanic eruption at the seafloor, the
intrusion of a dike, or both. The knowledge gained
from responding to these events has revolutionized
our understanding of these systems, especially their
pronounced temporal variability on very short
timescales (much less than days to weeks). Presumably we will be able to study more of these events
and to gain a sense of their frequency, perhaps as a
function of spreading rate, as we begin to instrument
more of the ridge crest with hydrophones to detect
the T-phase signals associated with these events.
The Influence of Tectonic/Cracking Events
Magmatic events have provided new insights into the
processes that drive hydrothermal systems and their
fluid compositions on intermediate- to fast-spreading
ridges, but presumably tectonic events are more important (or at least more frequent than magmatic
events) on slow-spreading ridges. As we have not yet
been able to observe one of these events, we cannot
assess their importance. In a relatively small cracking
event on a fast-spreading ridge observed with a
seismic array, changes in fluid temperatures were
marked, and changes in fluid compositions were
profound, leading to major changes in the biological
communities existing at this site. Presumably the
observation of one or more tectonic events on slowspreading ridges will also provide critical new insights into how hydrothermal systems on these types
of ridges function and evolve.
On-axis versus Off-axis
The discussion above has focused on the axial component of midocean ridge hydrothermal systems.
Many debates have focused on the importance of
these axial systems compared to hydrothermal
HYDROTHERMAL VENT FLUIDS, CHEMISTRY OF 85