Lessons from land to sea
A good amount of our exploration of the seafloor and understanding of its hydrothermal processes and products stems from our deep knowledge, largely gained from industry, of these same processes that formed ores mined on land. The geological record is replete with massive accumulations of base and precious metal sulfides in submarine volcanic and volcano-sedimentary terrains, so-called volcanogenic massive sulfide (VMS) deposits (or volcanogenic hydrothermal massive sulfide, or VHMS deposits as the Australians prefer to call them). Typically, these terrains are found in extensional basins above subduction zones, behind or within island arcs where volcanoes produce lavas and other products ranging in composition from basalt to rhyolite.
Some of the ore deposits are huge. Windy Craggy in northwestern British Columbia contains about 300 million metric tons of ore-grade sulfides, and Kidd Creek in northeastern Ontario contains about 170 million metric tons. The Iberian Pyrite Belt of southern Spain and Portugal contains 85 known VMS deposits totaling 1,765 million metric tons.
These numbers are big, and studies of these deposits and others like them have provided extensive knowledge on how and where VMS deposits form although it took some time for some of this research to be accepted. In 1847, the French scientist Elie de Beaumont surmised that what we now call VMS deposits were formed when minerals precipitated from volcanic and metalliferous emanations; but his work went unrecognized. In the early 1900s, Japanese scientists understood correctly that their Miocene-age Kuroko VMS deposits were formed by hot springs on the seafloor; but nobody in the West read their papers written in Japanese.
This image of a black smoker on the East Pacific Rise first appeared in the November 1979 issue of National Geographic magazine and introduced many people to black smokers. Photo courtesy of Dudley B. Foster, Woods Hole Oceanographic Institution.
Finally, in the West in the late 1950s, a young Dick Stanton declared that massive sulfide ores formed in island arcs. Those of Bathurst, New Brunswick, he said, were simply sulphide metamorphic rocks formed by springs and fumaroles in the seaboard areas of the adjacent volcanic terrain. Some vilified Stanton for this, at the time, heretical opinion, which countered the prevailing view that such deposits were formed by massive replacement of their volcanic host rocks. The discovery of black smokers forming massive sulfides on the modern seafloor would seem to still such opposition for all time, but what goes around comes around. Modern research shows that replacement is in fact an important process.
From the geological record, we know that large VMS deposits have several features in common. For many, it is clear that they form in an extensional tectonic setting, although compression commonly follows. Thick continental crust typically underpins the volcanic rocks adjacent to the magma that are the immediate host rocks of the deposit; for example, 30 to 40 kilometers of continental crust sit beneath a veneer of arc volcanics that contain the famous and extensively studied Kuroko deposits.
A heat engine of either magma or very hot rock is required to generate the hydrothermal fluid. Metals and other elements are leached from rocks in a high-temperature reaction zone by heated seawater and the much altered, buoyant fluid rises rapidly. A metal-rich fluid that is generated by reaction with basal passes through and further reacts with sediments, as it does at Guaymas Basin, before discharging onto the seafloor.
On land, the heat engine, in the form of sill-like bodies, can be identified in some ancient VMS terrains such as Matagami (Bell River Complex) and Noranda (Flavrian Intrusion), both of Archean age in Quebec. Careful studies of ancient terrains also reveal local structures at the site of deposition that serve to focus the fluid flow onto or just beneath the seafloor. We also know something of the nature of the ore-forming fluid as it is preserved in fluid inclusions: it is usually acid, moderately saline and generally less than 350 degrees Celsius. And we know a lot about the alteration of the rocks beneath the ore deposits.
The optimum conditions for the formation of VMS deposits, as ascertained from land studies, are best met in an island-arc or back-arc setting. This fact is one reason why a significant amount of research focuses on exploring the seabed of modern arcs for hydrothermal activity. What other tectonic terrains should we explore?
We have good examples on the modern seafloor of the Kuroko- and Cyprus-type VMS deposits in volcanic rocks; but not for the Besshi- and Sedex-types, in which sediments, rather than volcanics, play a predominant role. Perhaps this lack of evidence is not surprising, especially for the Sedex-type. These deposits probably formed from relatively low-temperature, highly saline and therefore dense fluids that ponded in depressions. Such fluids would not generate a plume in the seawater column, a signal relatively easy to detect for deposits forming on sediment-starved mid-ocean and back-arc ridges. Another reason for their apparent absence is that we simply havent looked for them, at least not hard enough.
Seemingly, everywhere that marine scientists look for hydrothermal activity on mid-ocean ridges and back-arcs even in some unexpected places they find it. Hydrothermal sites are now known in all the worlds oceans except where no one has explored, such as the High Arctic. Some mid-ocean sites that are heavily sedimented such as Middle Valley on the Juan de Fuca Ridge, Escanaba Trough on the Gorda Ridge, Guaymas Basin, and various deeps of the Red Sea, such as the Atlantis II have some similarity to Besshi and Sedex settings but lack at least one key feature. Perhaps the time is ripe to extend our seafloor investigations to those settings that more closely mimic these missing ore types.
Lessons from sea to land
Marine studies of the past two decades have confirmed much of what had been hypothesized about VMS ores. They have also contributed to our fund of knowledge about such deposits and may even provide some new guides for their exploration on land.
Bob Ballard, now head of the Institute for Exploration in Mystic, Conn., and Jean Francheteau, now at the University of Western Brittany in Brest, France, pointed out a long time ago before the complexities of seafloor spreading ridges were properly known that hydrothermalism on mid-ocean ridges was more prevalent in the ridge axis shallowest region, the so-called bathymetric minimum. This trend is understandable, given our current knowledge of axial magmatism, in which seismic studies show that magma (the heat engine) is closer to the seafloor and more prevalent along a ridge axis. The trend is also consistent with the changing compositions of mid-ocean ridge basalts approaching the bathymetric minimum and the main hydrothermal center. This change is recorded, for example, in the basalts large ion lithophile elements, potassium, barium and strontium. The bathymetric minimum is obvious on the modern seafloor, but how might it be detected in ancient terrains on land? The progressive change in basalt composition provides a clue that needs to be tested.
Without a doubt, the most important information to come from seafloor studies lies in vent fluids. Hydrothermal geochemists such as the late John Edmond of the Massachusetts Institute of Technology, his former student Karynne von Damm, who is now at the University of New Hampshire, Jean-Luc Charlou of IFREMER (the French Institute of Research and Exploitation of the Sea), Toshitaki Gamo of Tokyo University and David Butterfield of NOAA and the University of Washington perfected the technique of sampling vent fluids from submersibles. As a result, we now have a huge collection of such data from many different sites in different geological settings.
Armed with these data, we can much better understand the nature of fluid-rock interactions and the causes of variability in the compositions of sulfide ores. An earlier supposition was that, because hydrothermal solutions were produced at mid-ocean ridges when heated seawater reacted with basalt of more or less the same composition everywhere, all vented end-member fluids (i.e., uncontaminated by entrained fresh seawater) should have the same composition. We now know otherwise: end-member fluids have a wide range of compositions. Much of this variability arises because initial heating of a hydrothermal system brings unique physical and chemical conditions, because high-density and low-density saline phases separate and each carry different dissolved constituents; and because a hydrothermal system stabilizes as it matures.
A further complexity is that magmatic fluid rich in metals and volatiles might also contribute to the hydrothermal circulation system, and thus to ore formation. This metalliferous fluid was first identified and analyzed in a modern submarine setting by Kaihui Yang and the author in a paper published in the Oct. 3, 1996, Nature, although we were not able to demonstrate that this magmatic fluid actually made it into the hydrothermal system. We described precipitates of ore metals within the magmatic vapor bubbles of melt inclusions contained within phenocrysts and in small vesicles of an andesite from the Manus back-arc basin. In the August 2002 Economic Geology, we showed that the metals in these precipitates change progressively from more copper-rich to more zinc-rich with fractionation of the lavas from basalt to rhyolite. Interestingly, ancient ores in basalts are rich in copper and those in rhyolite are rich in zinc.
Other researchers, such as Yves Fouquet of IFREMER, as well as Cornel de Ronde of the Institute of Geological and Nuclear Sciences in New Zealand, Peter Herzig of Freiberg University in Germany and their colleagues, have provided isotopic and chemical evidence from gases for the presence of magmatic fluids in marine hydrothermal systems.
Giant ore deposits of more than 100 million metric tons require a superabundant supply of metals. Simple leaching of volcanic rocks or sediments by heated seawater cannot easily provide this supply within reasonable times. And observed volumes of leached rock beneath ancient VMS deposits would not yield enough metals. Perhaps a clue to the origin of giant VMS deposits and evidence for their presence in a particular locality lies in the unequivocal but elusive evidence for the release of magmatic fluid from shallow intrusions.
Many studies of hydrothermal plumes show that free venting of hydrothermal fluid into seawater results in the loss of as much as 90 percent of the metals in the form of particulates that are widely dispersed. Tim McConachy of the Commonwealth Scientific and Industrial Research Organisation in Australia points out that a typical plume from a black smoker discharging at 2 meters per second can produce 1 million metric tons of particulates in 10,000 years, while a plume discharging at 200 meters per second can produce the same amount in 100 years. Also, we know from isotopic dating by Claude Lalou of CNRS in Gif-sur-Yvette, France, that seafloor hydrothermal systems can last for at least 100,000 years. However, according to the model by Richard Feely of NOAAs Pacific Marine Environmental Laboratory in Seattle, 75 percent of the particulates representing the coarsest fraction that are most prone to settle near their source are too dispersed over several square kilometers of the seafloor to form a sulfide ore. Fallout of these particulates contaminates seafloor sediments, although the amount of metalliferous sediment actually observed in volcanic terrains of the modern seafloor is surprisingly small.
Metalliferous sediments decimeters to meters thick are common in the vicinity of ancient VMS deposits, which in some cases extend several kilometers along the ore horizon good examples are the Japanese tetsusekiei (iron-quartz), Noranda Contact Tuffs and Matagami Key Tuffite. The presence of such rocks is a good exploration guide for nearby VMS deposits, but explorationists have tried in vain to find geochemical clues in these rocks pointing to ore. Intuitively, one would think that such vectors must exist and perhaps further research into the modern seafloor metalliferous sediments will unlock this secret.
Much of the dissolved material may never reach the deposits chimney. Drilling into sulfide mounds by the Ocean Drilling Program and observations of the interiors of deposits incised by faults is revealing processes within the mound and beneath the seafloor. The current seafloor work, because it catches this process in action, is allowing us to quantify the process. A somewhat startling result from these studies is that extensive replacement is occurring beneath the seafloor, particularly replacement of sediments such as at Middle Valley; but also, according to Fouquet, of ultramafics at the Rainbow site on the Mid-Atlantic Ridge. Mounds grow by replacement beneath the seafloor and by internal inflation above the seafloor. This process may be an answer to the enigma of how large VMS deposits can form when free venting into the water column is known to be so very inefficient.
Land-based studies have implied that most VMS ores formed in depressions. But the modern seafloor has shown otherwise, revealing that the ores form on high-standing structures, such as ridges. Perhaps this difference is more apparent than real, because deposits forming on mid-ocean ridges tend to be confined to the axial graben, or depression, of what is otherwise a high-standing volcanic structure. The same is true of deposits forming in the caldera of ridge-associated volcanoes such as Axial Volcano at 46 degrees north on the East Pacific Rise, an area that has been the focus of the NOAA-Canada NeMO Project. (NeMO, or New Millennium Observatory, is a long-term observation of the biology, chemistry and geology of an active volcanic portion of the mid-ocean ridge.) It is also true of the large sulfide deposit forming in a graben on the flank of an off-axis volcano to the east of the East Pacific Rise at 13 degrees north, described by Roger Hekinian, formerly of IFREMER, and by Fouquet.
Can we mine ocean sulfides?
The largest of the known seafloor deposits could be an economic resource in
their own right. I have made this case in several articles over the past 15
years, most recently in the June 2001 issue of Geoscience Canada. Others who
have made similar cases are Ray Binns and David Decker of CSIRO Australia, in
a 1998 article in Scientific American Presents The Oceans, as well as Herzig
The seafloor hosts deposits of apparent size and grade that, if they were on land, would definitely be targets for further evaluation. That they are under several hundred to a few thousand meters of water is of no particular concern. Deep marine technology is up to the challenge. The oil industry moved offshore in the mid-20th century, despite huge resources remaining onshore and despite the technology for marine recovery of oil not having been perfected. Today, about one-third of the worlds oil production is from the offshore, and that portion is growing. Oil exploration is extending beyond 3,000 meters water depth, about the maximum depth at which potentially viable mineral deposits have been found.
Will the mining industry attempt ocean recovery of polymetallic sulfides? The industry was terribly disillusioned by the $650 million failure in the 1970s and 1980s of manganese nodule mining (which interestingly is again in vogue in Korea, Japan and China). Most companies are reluctant to try again.
Two small entrepreneurial companies, Nautilus Minerals Corp., based in Australia, and Deep Sea Minerals, based in the United States, are betting that a market will be out there for marine polymetallic sulfides. Nautilus has an exploration license from the government of Papua New Guinea covering 2,458 square kilometers of the Manus Basin and an application pending for an additional 580 square kilometers further east around Lihir Island where Ladolam, one of the worlds richest gold deposits, is being exploited. Deep Sea and their partner, Phelps Dodge, have several applications pending.
More than 130 years ago, the character Captain Nemo in Jules Vernes Twenty Thousand Leagues Under the Sea claimed, In the ocean depths, there exist mines of zinc, iron, silver and gold which would be quite easy to exploit. Perhaps, as with so many of his insights, Verne was right.
from an ancient ocean
Photo courtesy of Chris Kennedy, University of Toronto.
The biomass of seafloor hydrothermal systems is immense.
In a 1998 paper in Advances in Marine Biology (vol. 34), Verena Tunnicliffe
of the University of Victoria identified 443 invertebrate species, most
of which were new to science. More have been discovered since. Biologists
will be occupied for a very long time to determine what role each creature
plays. Perhaps some, like Junipers worm, play an important role
in precipitating metal sulfides.