1

1

2

The Magnetic Signature of Hydrothermal Systems in Slow Spreading Environments 3

Maurice A. Tivey1 and Jerome Dyment

2 4

5

1Dept. of Geology and Geophysics 6

Woods Hole Oceanographic Institution 7

Woods Hole, MA, 02543 8

9

2Equipe de Géosciences Marines, 10

Institut de Physique du Globe de Paris and CNRS 11

4 place Jussieu, 75005 Paris, France 12

13

AGU Monograph: Diversity of hydrothermal systems on slow spreading ocean ridges 14

Editors: Peter Rona, Colin Devey, Jerome Dyment and Bramley Murton 15

16

17

2

18

1. Abstract 19

Slow-spreading midocean ridges like the Mid-Atlantic Ridge host a remarkable diversity of 20

hydrothermal systems including vent systems located on the neovolcanic axis, large axial 21

volcanoes, in transform faults and non-transform offsets, and associated with low-angle detachment 22

faults, now recognized as a major tectonic feature of slow spreading environments. Hydrothermal 23

systems are hosted in various lithologies from basalt to serpentinized peridotite and exposed lower 24

oceanic crust. The substantial variations of hydrothermal processes active in these environments 25

have important implications for the magnetic structure of oceanic crust and upper mantle. 26

Hydrothermal processes can both destroy the magnetic minerals in basalt, diabase and gabbro and 27

create magnetic minerals by serpentinization of ultramafic rocks and deposition of magnetic 28

minerals. We report on the diversity of magnetic anomaly signatures over the vent systems at slow 29

spreading ridges and show that the lateral scale of hydrothermal alteration is fundamentally a local 30

phenomenon. This highly focused process leads to magnetic anomalies on the scale of individual 31

vent fields, typically a few 100 meters or less in size. To detect such features, high-resolution, 32

near-bottom magnetic surveys are required rather than sea surface surveys. High-resolution 33

surveys are now more tractable with deep-towed systems, dynamically-positioned ships and with 34

the recent development of autonomous underwater vehicles, which allow detailed mapping of the 35

seafloor on a scale relevant to hydrothermal activity. By understanding these present-day active 36

hydrothermal systems, we can explore for yet to be discovered buried deposits preserved off-axis, 37

both to determine past history of hydrothermal activity and for resource assessment. 38

39

40

3

2. Introduction 41

It has long been recognized that geothermal activity can produce distinctive magnetic 42

anomalies in terrestrial volcanic rocks (e.g. Hochstein and Soengkono, 1997). There are numerous 43

ground and aeromagnetic surveys documenting the occurrence of reduced or weakly magnetic 44

anomalies over geothermal areas in New Zealand and Iceland (e.g. Watson-Munro, 1938; Studt, 45

1959; Palmason, 1975). The source of the reduced magnetization has variously been attributed 46

either to the high temperatures found in these systems that exceed the Curie temperature of the 47

magnetic minerals in the host rock [Watson-Munro, 1938] or to hydrothermal alteration and 48

destruction of the magnetic minerals [Studt, 1959; Browne, 1978]. A review of the early literature 49

by Hochstein and Soengkono [1997] revealed that alteration of the magnetic mineral magnetite to 50

less magnetic minerals was by far the most important mechanism in generating the anomalies. 51

Interestingly, they also found that in liquid-dominated geothermal systems, magnetite was the first 52

mineral to be replaced by non-magnetic minerals such as pyrite, but in vapor-dominated systems 53

that were oxidizing, magnetite is stable and remains magnetic. It was further noted, however, that 54

geothermal systems often went through phases of liquid- and then vapor-dominated periods, so that 55

even a temporary stage of liquid-dominated activity could permanently erase the magnetic 56

properties of the host rock [Hochstein and Soengkono, 1997]. 57

58

In the marine realm, the recognition of magnetic anomalies associated with hydrothermal 59

systems has been slower to gain acceptance. While early studies of the rock magnetic properties of 60

basalt ranging from terrestrial lava outcrops [Ade-Hall et al., 1971], to ocean crustal drilling [Ade-61

Hall et al., 1973], dredged samples [Irving, 1970] and lab experiments [Johnson and Merrill, 1972; 62

1973] all demonstrated that magnetization was significantly affected by low-temperature alteration, 63

4

this process was considered relatively regional and broad-scale in effect [Johnson and Atwater, 64

1977]. Other than an overall decrease in magnetic anomaly amplitude, the effect of concentrated 65

high-temperature hydrothermal alteration was not considered widespread or common. Although it 66

was recognized early on that hydrothermal mineral deposits could form at mid-ocean ridge (MOR) 67

spreading centers [Bostrom and Peterson, 1966; Spooner and Fyfe, 1973; Hutchinson, 1973; 68

Sillitoe, 1973; and Bonatti, 1975], it was not until the discovery of the mineral-rich sediments in 69

deep brine pools of the Red Sea [Degens and Ross, 1969] and active hot springs on the seafloor in 70

the Galapagos Rift [Corliss et al., 1979] that the possibility of crustal alteration associated with 71

hydrothermal vent systems became evident. Clues to the possibility of magnetically disturbed areas 72

in upflow zones related to hydrothermal activity came from studies in ophiolites. A detailed 73

ground level magnetic survey over a massive sulfide body in the Troodos ophiolite complex in 74

Cyprus revealed a strong magnetic low over the stockwork zone, which is hosted within a basaltic 75

lava sequence at Agrokipia [Johnson et al., 1982]. Rock magnetic studies revealed that the 76

titanomagnetite in the basalt had been replaced with non-magnetic sulfide minerals and the basalt 77

itself had been altered to a clay, silica and chlorite assemblage within the stockwork upflow zone 78

[Johnson et al., 1982; Hall, 1992]. 79

80

Rona [1978] published on the possible magnetic signature of hydrothermal alteration and 81

volcanogenic mineral deposits in oceanic crust and proposed that magnetic anomalies might be an 82

effective method of detecting and characterizing such areas and deposits. Since that time, we have 83

seen the discovery of over 200 active vents on the seafloor in many different environments 84

[Hannington et al., 2005]. As we have come to learn, the detection of water column plumes from 85

active vents and seafloor photography and visual imagery of distinctive biological communities 86

5

have become a successful method of vent discovery. The geophysical signature of hydrothermal 87

systems based on crustal properties has been a lot more difficult to employ primarily due to the 88

small scale of the hydrothermal features (e.g. typically only a few 100s of meters across) compared 89

with the ocean depth (kilometers), which imposes a resolution limitation on sea surface 90

measurements. Progress has been made, however, with the advent of deep-towed geophysical 91

surveys [Tivey et al., 2003], near-bottom submersible and Remotely-Operated Vehicle (ROV) 92

surveys [Tivey et al., 1993; 1996; Dyment et al., 2005] and most recently autonomous underwater 93

vehicle (AUV) surveys [Tivey et al., 1997; 1998; 2006]. These high-resolution surveys now reveal 94

that hydrothermal systems do indeed have a detectable geophysical response and we can now 95

embark on the next challenge of documenting the detailed subsurface structure of these systems and 96

to apply these techniques to finding older and non-venting deposits both on and off the MOR. In 97

the following discussion, we have chosen to focus primarily on the magnetization response, which 98

removes the latitude skewing effect of the magnetic field and magnetization vector directions on 99

the magnetic anomaly signal. Magnetization is typically computed from magnetic field data by 100

assuming the direction of the magnetic field and magnetization vector and a source body geometry, 101

either constant layer thickness or some geometrical shape (e.g. pipe) [e.g. Parker and Huestis, 102

1974]. 103

104

Not all types of hydrothermal activity will result in reduced magnetization and magnetic 105

mineral destruction. In some environments hydrothermal activity can result in magnetite 106

formation. The process of serpentinization can lead to significant magnetite deposition and there is 107

now clear evidence that magnetite formation can occur not only through normal seafloor alteration 108

of peridotite [Dunlop and Prevot, 1982; Smith and Banerjee, 1985; Bina and Henry, 1990; 109

6

Nazarova, 1994; Oufi et al., 2002], but also can occur where there are vigorous hydrothermal 110

systems hosted in serpentinized peridotite bodies [Murton et al., 1994; German et al., 1996; 111

Gebruk et al., 1997; Charlou et al., 1998; Barriga et al., 1998; Kelley et al., 2001] that have 112

ongoing magnetite formation. These ultramafic-hosted vent systems appear to produce positive 113

magnetic anomalies either related to the formation of magnetite through serpentinization [Dyment 114

et al., 2005] or by the mineralization process. 115

116

Magnetite can also form as part of the mineralization process. An example is the sediment-117

hosted Bent Hill hydrothermal system in Middle Valley on the northern Juan de Fuca Ridge, which 118

produces a positive magnetic anomaly related to a magnetite zone at depth [Tivey, 1994], that was 119

subsequently confirmed by drilling [Zierenberg et al., 1998]. Other magnetic minerals can form as 120

a consequence of hydrothermal activity. The most common is pyrrhotite (Fe7S8), an iron sulfide 121

that has magnetic behavior that varies according to its crystal structure and composition. 122

Monoclinic pyrrhotite is weakly magnetic (ferromagnetic) with a Curie temperature of 325°C, but 123

the magnetic properties weaken with greater sulfur content as the iron sulfide becomes more pyritic 124

in composition (FeS2). Hexagonal pyrrhotite (Fe9S10) is nonmagnetic (antiferromagnetic) at room 125

temperature although it is weakly magnetic (ferromagnetic) above its lambda transition temperature 126

of ~200°C when thermally activated vacancy ordering occurs until its Curie temperature at 265°C 127

is reached. Some of the vent structures on the Juan de Fuca ridge contain pyrrhotite [Tivey and 128

Delaney, 1985; 1986]. 129

130

The slow-spreading Mid-Atlantic Ridge (MAR) shows a remarkable diversity in the types 131

of hydrothermal systems with significant differences in the tectonic and volcanic setting combined 132

7

with highly variable host rock lithologies (see Table-1 and Fig. 1). Basalt-hosted hydrothermal 133

vent systems range from what are now recognized as low-angle detachment fault controlled 134

systems such as TAG (26°N) and the recently discovered systems at 13-14°N [Murton et al., 2007; 135

Searle et al., 2007], to vents directly located at the neovolcanic axis such as Snake Pit at 23°N and 136

Turtle Pits at 4°48’S, to those associated with axial volcanoes such as Lucky Strike and Menez 137

Gwen [Langmuir et al., 1997; Ondreas et al., 1997]. Ultramafic hydrothermal vent systems range 138

from the peridotite-hosted Logatchev and Rainbow vent systems found at segment discontinuities 139

[German et al., 1996; Fouquet et al., 1997; Gebruk et al., 1997] to the carbonate dominated Lost 140

City located on the Atlantis transform at 30°N [Kelley et al., 2001]. Below we discuss the magnetic 141

properties and resultant magnetic anomalies found over this remarkable range of hydrothermal 142

systems present on the slow-spreading MAR. Magnetic anomalies are one of the few methods at 143

present that provide insight into the subsurface structure of active hydrothermal systems. 144

Understanding the geometry and distribution of this subsurface crustal structure of active 145

hydrothermal vent systems should allow us to more confidently search and characterize those 146

ancient deposits now preserved off-axis on the flanks of the MOR system. 147

148

3. Magnetism of basalt-hosted systems on the MAR 149

3.1 TAG Hydrothermal Field (26° 08’N 44°45’W) 150

The Trans-Atlantic Geotraverse (TAG) program in 1973 provided the first clues to the 151

existence of hydrothermal activity on the slow spreading Mid-Atlantic Ridge near 26°N [Rona, 152

1973]. A water temperature and camera survey of the eastern rift valley wall showed evidence of 153

elevated temperature near the seafloor and low temperature alteration and mineral deposition [Rona 154

et al., 1975; Scott et al., 1974]. A detailed sea surface magnetic survey of the region (Plate 1a,b) 155

8

revealed a distinctive zone of reduced magnetic intensity directly over the spreading axis 156

[McGregor and Rona, 1975; McGregor et al., 1977; Rona, 1978]. Detailed magnetic modeling of 157

this anomalous zone suggested that block rotation could not cause the reduction in magnetic 158

intensity and it was suggested instead that hydrothermal alteration was responsible for this 159

reduction in magnetic intensity [McGregor et al., 1977]. It wasn’t until many years later in 1985 160

that active black-smoker high temperature hydrothermal activity was discovered in the TAG area 161

[Rona 1985; Rona et al., 1986]. A large circular mound ~100 m in diameter and 50 m high was 162

documented at 26°08’N and 44°45’W at the foot of the eastern rift valley scarp with a central black 163

smoker complex discharging 350°C fluids [Campbell et al., 1988]. 164

165

To investigate the relationship between reduced magnetic anomaly intensity and crustal 166

magnetization, Wooldridge et al., [1990] studied the rock magnetic properties of dredged and 167

submersible rock samples from TAG and other known or suspected hydrothermal sites on the MAR 168

and Gorda Ridge in the Pacific. Contrary to what was expected, the basalt samples showed no 169

evidence of pervasive hydrothermal alteration and showed instead high natural remanent 170

magnetization (NRM), high Koenigsberger ratio values (the ratio of remanent to induced 171

magnetization) and strong stability, in other words, the typical magnetic parameters of young, 172

relatively fresh, MOR basalts. These results suggest that if hydrothermal activity were indeed 173

affecting oceanic crustal magnetization, then the zones of alteration are discrete and localized 174

perhaps along faults or the alteration is confined to the crust at depth beneath more recent lava 175

flows. 176

177

9

The original magnetic field data from the 1973 survey were reprocessed by Wooldridge et 178

al. [1992] and analysed using the inversion techniques of Parker and Huestis [1974] modified for 179

three-dimensions by Macdonald et al. [1980]. In addition to TAG, Wooldridge et al., [1992] also 180

investigated sea surface magnetic anomalies over several other known and suspected sites of 181

hydrothermal activity on the MAR at 15°N and 17°N and also the Sea Cliff vent field on the Gorda 182

Ridge at 42°45’N. They suggested that magnetic anomaly lows over these regions were primarily 183

caused by regional-scale high-temperature hydrothermal alteration of the crust with subsidiary 184

factors at each site varying from captured reversely magnetized crust, to thinner extrusive crust 185

arising from magmatic cyclicity or due to the raised temperatures in the crust above an axial 186

magma chamber. In the latter case, an increase in crustal temperature in a region above a magma 187

chamber would cause the magnetic minerals to exceed their Curie temperature i.e. the temperature 188

above which magnetism ceases to occur, which for the dominant magnetic mineral in basalt, 189

titanomagnetite, is approx. 160-300°C [Wooldridge et al., 1992]. 190

191

Sea surface magnetic surveys are necessarily limited in their resolution by the water depth 192

(3-4 km), the distance from the source region. However, most of the discovered hydrothermal vent 193

fields to date have spatial extents on the scale of 100s of meters [Hannington et al. 2005], which 194

means that sea surface data could detect regional kilometer scale alteration but lacks the spatial 195

resolution needed to resolve individual vent systems. Near-bottom surveys provide magnetic 196

measurements on the same spatial scale as individual vent deposits and thus can tie the anomalous 197

field more directly to the deposits. The first unambiguous documentation of the magnetic effects of 198

a seafloor hydrothermal field in detail was obtained by the submersible ALVIN in a 1990 survey 199

over the active TAG mound [Tivey et al., 1993]. A distinct zone of reduced magnetization was 200

10

computed from magnetic field data collected along submersible tracks at an altitude of 20 meters 201

above the mound (Plate 2). Magnetic modeling of this zone of reduced magnetization suggested a 202

pipe-like source body with a radius of 100 m, a depth extent of 500 m and a magnetization contrast 203

of 12 A/m. Inversion results suggested a narrower zone some 80 m across (Plate 2). This narrow 204

pipe-like body is directly comparable in size to the upflow zones and stockwork pipes typically 205

found in terrestrial volcanogenic massive sulfide ore deposits such as the Cyprus ophiolite 206

[Johnson et al. 1982; Richards et al., 1989]. To answer the question of whether the zone of reduced 207

magnetization beneath the active TAG mound could be responsible for the more regional anomaly 208

measured at the sea surface, a computation of the total magnetic moment suggested that this 209

discrete zone of demagnetization could not produce the anomaly [Tivey et al., 1993]. The 210

estimated total magnetic moment for the TAG active mound is 108 Am

2; which is calculated from 211

the computed magnetization contrast, the area of the low magnetization zone and the thickness of 212

the source body. A similar calculation for the magnetic moment required to generate the sea 213

surface anomaly is 5 x 1010

Am2, (see Tivey et al., 1993), which means that several hundreds of 214

sources (500) similar to the TAG active mound would be needed to create the surface anomaly. 215

Such sources clearly do not exist. Tivey et al., [1993] concluded that the regional zone of reduced 216

magnetization measured at the sea surface was a reflection of crustal scale processes rather than an 217

ensemble of individual vent deposit anomalies. 218

219

In 1994, new insight into the source of the regional magnetic anomaly at TAG came from 220

the analysis of a deeptow sidescan and magnetic survey of the area [Kleinrock and Humphris, 221

1996] as part of a site survey effort prior to drilling by the Ocean Drilling Program (ODP) 222

[Humphris et al. 1995; Herzig et al., 1998]. The deeptow sidescan and magnetic data were 223

11

collected at an altitude of 100 m, with line spacing of 700 m, and encompassed an area of 200 sq. 224

km extending across the TAG segment spreading axis and 10 km from north to south. The 225

magnetic data analysis revealed that the zone of reduced magnetization is clearly located over the 226

eastern wall of the rift valley and is not directly associated with the active and inactive 227

hydrothermal mounds (Plate 1c,d). Combined with the earlier discovery of gabbros on the eastern 228

valley wall [Zonenshain et al., 1989], further analysis suggested that the regional magnetic low is 229

the result of crustal thinning and the loss of the extrusive crustal layer caused by 4 km of horizontal 230

extension on a low-angle normal fault [Tivey et al., 2003]. It was suggested that the location of the 231

active TAG mounds and adjacent relict mounds on the hanging wall of this detachment fault was 232

one reason why these systems had been active for such a long period of time [Lalou et al., 1993; 233

1998; You and Bickle, 1998]. Further confirmation of the geometry of the detachment faulting was 234

provided by a seismic refraction survey and microearthquake study, which imaged the crustal 235

structure and revealed the steep dip of the detachment fault at its toe [Canales et al., 2007; 236

deMartin et al., 2007]. The recognition that low-angle detachment fault systems are much more 237

prevalent than first thought in slow spreading environments [Smith et al., 2006; 2008] coupled with 238

recent discoveries of additional active and inactive hydrothermal activity [Searle et al., 2007; 239

Murton et al., 2007; Cannat et al., 2007; Fouquet et al. 2007] associated with these types of fault 240

systems suggests that these environments are a major new addition to the inventory of 241

hydrothermal activity at slow spreading ridges [Escartin et al., 2008]. 242

243

ODP drilling at TAG (Leg 158) was intended to sample the subsurface structure of an active 244

hydrothermal system and to investigate the mineral and host rock zonation and fluid-rock 245

interactions [Herzig et al. 1998; Humphris et al., 1996]. Drilling conditions were difficult due to 246

12

the harsh conditions and core recovery was generally poor (av. 12%). However, a penetration of 247

125 m was accomplished and the samples helped reveal the processes active within the mound. The 248

mound was found to consist of 4 different types of breccias: a massive pyrite breccia, an anhydrite-249

rich breccia, a quartz-sulfide breccia stockwork at depths > 40 mbsf and a quartz-chlorite breccia 250

stockwork at depths >100 mbsf. The drill-hole results confirmed that the zone of upflow is 251

narrowly constrained beneath the active mound and is likely to be on the order of 80 m in diameter, 252

consistent with the submersible magnetic inversion results (Plate 2). 253

254

Rock magnetic measurements of the breccias reveal very weak magnetization with NRMs 255

generally >> 1 A/m and low magnetic susceptibility [Zhao et al., 1998]. Magnetic tests of the 256

sulfide samples suggest that pyrrhotite and trace amounts of magnetite are present and contribute to 257

the magnetic properties of these samples. Holes TAG-2 and TAG-4 drilled the margins of the 258

active mound and did intercept basalt and silicified basalt breccia at depth. Magnetic 259

measurements of these samples show NRM values more typical for basalt (~10 A/m) and higher 260

magnetic susceptibility. Magnetic tests suggest a pseudo-single domain grain size and high Curie 261

temperatures of 500-540°C indicative of low-Ti titanomagnetite. Unoxidized titanomagnetite has a 262

Curie temperature of 160°C, while completely oxidized titanomaghemite has a Curie temperature 263

of 350°C [Ade-Hall et al., 1971]. Zhao et al. [1998] suggests the possibility of high-T deuteric 264

oxidation of these basalt samples. Perhaps the most surprising result, however, was the fact that the 265

basalts also showed much shallower inclinations (20.8°) than predicted (55°) based on the site 266

latitude [Zhao et al., 1998]. The explanation for this shallow inclination remains to be resolved. 267

The inclination could be the result of tilting/rotation due to faulting, although no obvious signs of 268

this are seen in other data. The magnetization could simply be remagnetized through chemical 269

13

alteration, although such remagnetization is unlikely to be in a direction different than the ambient 270

field. Another speculation is that the basalts may have captured a geomagnetic excursion for which 271

there are several candidates in the Brunhes recent past (e.g. Laschamps, Lake Mungo, etc) [Zhao et 272

al., 1998]. Age dating and magnetic paleointensity measurements could help to test this possibility. 273

274

In summary, hydrothermal activity at the TAG active mound and nearby inactive mounds 275

does result in localized demagnetized zones but these are not responsible for the observed sea 276

surface anomaly. This regional sea surface anomaly is the result of crustal-scale processes related 277

to a loss in magnetic source layer (extrusive lavas) by low-angle detachment faulting which appears 278

to act as a mechanism to promote hydrothermal activity on the hanging wall of the fault. 279

280

3.2 Snake Pit Hydrothermal zone (23°22.08’N 44°57.00’W) 281

The Snake Pit hydrothermal area was discovered by deeptow camera in May 1985 during a 282

site survey for bare-rock drilling sites on the MAR just south of the Kane fracture zone at 23°22N 283

[Kong et al., 1985]. The site is located on the neovolcanic ridge within the axial rift valley, 25 km 284

south of Kane fracture zone. The deeptow camera imaged pockets of greenish-white mottled 285

sediment populated by crabs and spindly worms subsequently found to be shrimp and eels (after 286

which the vent site was named). The vent area was explored and drilled as part of ODP Leg 106, 287

site 649 [Detrick et al., 1986]. The pre-drill camera survey found an 11 m tall black smoker 288

chimney and an extensive area of mineralization and hydrothermal chimney debris. A transect of 289

holes were drilled with very limited core recovery; only one hole, site 649B recovered 6 m of ore 290

minerals before intercepting hard substrate. Only a small amount of basaltic material (chips) was 291

recovered from the drilling. This basalt was mostly glassy and aphyric and apparently unaffected 292

14

by hydrothermal alteration. The basalt also had a relatively high MgO content, which indicates 293

slightly more primitive lava than the basalt recovered to the north at site 648 (Serocki volcano) 294

[Detrick et al., 1986]. The area was subsequently visited by submersibles Alvin [Karson and 295

Brown, 1988], Nautile [Mevel et al., 1989; Dubois et al., 1994] and Mir [Lisitsyn et al., 1989]. The 296

pre-ODP drilling surveys included sea surface magnetic surveys [Schulz et al., 1988], which 297

showed no pronounced magnetic anomaly low as was noted over the TAG area [Wooldridge et al., 298

1992; Tivey et al., 1993]. Inspection of the magnetic inversion for crustal magnetization [Schulz et 299

al., 1988, Fig.9] shows that ODP site 649 at the Snake Pit vent site is located between two along-300

axis magnetization highs, but as discussed earlier, it is unlikely that the magnetic effects of 301

hydrothermal alteration related to individual vent fields can be resolved by the sea surface magnetic 302

data. To date, no near-bottom magnetic data have been collected over the Snake Pit area. 303

304

3.3 Lucky Strike (37°17’N 32°17’W) 305

Following a series of research cruises to survey the MAR axis in the early 1990’s under the 306

FARA program (French-American cooperative agreement, Needham et al., 1992), a detailed 307

sampling program (FAZAR) collecting rock samples every 5 km along the neovolcanic zone 308

between 34° and 41°N recovered one dredge at 37°17’N with sulfides and hydrothermal vent 309

organisms [Langmuir et al., 1997]. This serendipitous event thus coined the name of this area as 310

“Lucky Strike”. The Lucky Strike ridge segment located between 37° and 37°35’N is the third 311

segment south of the Azores platform and represents a hot-spot influenced segment that provides an 312

opportunity to document the potential impact of hot-spot magmatism on a hydrothermal system. 313

The ridge segment is 65 km long and marked by a rift valley that has remarkably uniform width of 314

11 km wide and a prominent axial volcano 430 m high reaching 1550 m at its summit located at the 315

15

center of the segment (Plate 3). The summit of the axial volcano can be divided into two halves, a 316

linear narrow ridge to the west and a more circular edifice to the east that is marked by 3 small 317

cones which enclose a lava lake at the summit [Fouquet et al., 1995]. Hydrothermal activity is 318

located on the periphery of this lava lake. Several submersible dive programs documented the 319

presence of high temperature black smoker chimneys, extensive areas of diffuse flow and sulfide 320

deposits distributed around the lava lake margins [Fouquet et al., 1994a, b; Langmuir et al., 1997; 321

Ondreas et al., 2009]. The presence of a lava lake at the summit also suggests recent magmatic 322

activity and the potential for an active magma chamber directly beneath the edifice. Confirmation 323

of the presence of an active axial crustal magma chamber at a depth of 3 km was obtained by a 324

seismic reflection survey over the Lucky Strike volcano [Singh et al., 2006]. Ongoing studies 325

including a recent on-bottom gravity study have been completed across the area to detect the 326

presence of this magma chamber as part of the MOMAR observatory effort. 327

328

Publically available sea surface magnetic data (National Geophysical Data Center (NGDC): 329

www.ngdc.noaa.gov) were compiled over the Lucky Strike area. Inversion for crustal 330

magnetization was performed using the Parker and Huestis [1974] approach assuming a source 331

layer thickness of 1 km, a geocentric dipole direction for the magnetization and a pass band with 332

wavelength cutoffs of 237 and 2 km respectively. The magnetization inversion shows a slight 333

weakening of the axial magnetization signal over the center of the segment compared with the 334

segment ends (Plate 3b), which is typical of MAR segments in general [Grindlay et al., 1992; 335

Ravilly et al., 1998; Tivey and Tucholke 1998;] and has been explained in various ways including 336

changes in along-axis crustal thickness and composition, in the depth of the Curie isotherm, in the 337

basalt petrology, and in the degree of alteration. In the case of the Lucky Strike segment, the 338

16

presence of a central volcano argues against thinned crust but might support thermal 339

demagnetization due to an elevated Curie isotherm. The inversion and analysis of a dense sea 340

surface magnetic anomaly compilation has revealed a strong, ~7 nT magnetization low over the 341

Lucky Strike lava lake and hydrothermal area [Miranda et al., 2005]., suggesting that the 342

hydrothermal systems located on the summit of the volcano may have sufficient volume to be 343

imaged magnetically from the sea surface. A detailed near-bottom sidescan sonar, phase 344

bathymetry and magnetic survey was completed over the axial volcano in 1996 at an altitude of 100 345

m [Fornari et al., 1996; Scheirer et al., 2000; Humphris et al., 2002]. The near-bottom magnetic 346

data were upward continued to a level plane, bordered with the regional data and inverted for 347

crustal magnetization assuming a source layer thickness of 1 km. A geocentric dipole direction for 348

the magnetization was assumed and the inversion solution was filtered with a wavelength passband 349

of 2 and 237 km to ensure convergence. The magnetization inversion results, shown in Plate 3d, 350

shows two distinct zones of reduced magnetization. A 1.5 x 3 km zone of low magnetization is 351

located over the easternmost cone of the summit and a longer, linear magnetization low is found to 352

the west of the summit (Plate 3d). These zones appear to align with the crustal fault systems 353

imaged by the reflection seismic data (Plate 3) that bound the summit area [Singh et al., 2006]. 354

These bounding faults are interpreted by Singh et al. [2006] to be pathways of hydrothermal 355

recharge for the axial hydrothermal system beneath the summit and thus may have suffered 356

enhanced alteration resulting in the linear zones of reduced magnetization on a crustal scale (Plate 357

3d). Confirmation of this crustal alteration within these fault zones remains to be confirmed by in 358

situ observations and rock sample recovery and analysis. More detailed surveys are ongoing at 359

Lucky Strike as part of the MOMAR observatory effort and high resolution camera-tow magnetic 360

data were collected as part of this effort but remain to be analyzed [Escartin et al., 2006]. 361

17

362

3.4 Menez Gwen (37°50’N 31°30’W) 363

The Menez Gwen hydrothermal field was discovered during submersible dives on the ridge 364

segment between 37°35’N and 38°N just north of the Lucky Strike segment [Fouquet et al., 365

1994a,b]. This segment also has a well-developed circular volcano similar to Lucky Strike, which 366

is centrally located within the rift valley (Plate 4). The 800 m high volcano is 16 km in diameter 367

and has a 2 km wide by 6 km long axial graben at its summit, which is ~300 m deep, open to the 368

north and south and floored by very fresh volcanics [Ondreas et al., 1997]. A younger 120 m high, 369

700 m diameter volcanic cone is located towards the northern end of the graben. The vent field 370

was discovered on the east and southeast slopes of this volcanic cone at depths of 840 to 865 m and 371

hosted in young fresh pillow lava [Desbruyeres et al., 2001]. In terms of magnetism, the area has 372

not been mapped in detail by near-bottom sensors but sea surface magnetic data do exist. An 373

inversion of sea surface magnetic field data compiled from NGDC (using the Parker and Huestis 374

inversion approach [1974] and assuming a 1 km thick source layer and geocentric dipole direction 375

for the magnetization) shows that the central portion of the Menez Gwen ridge segment dominated 376

by the large axial volcano does have a reduced magnetization zone (Plate 4). This reduced zone of 377

magnetization is similar to other ridge segments of the MAR, which show similar reduced 378

magnetization at segment centers and enhanced magnetization at segment ends [Grindlay et al., 379

1992; Ravilly et al., 1998; Tivey and Tucholke, 1998]. Nevertheless, the juxtaposition of reduced 380

magnetization at the magmatically robust crustal center certainly suggests that processes other than 381

crustal thickness are acting to reduce the magnetic effect of the crust (Plate 4). 382

383

3.5 South Atlantic Vent Sites 384

18

Over the past few years, MOR exploration has also begun to include the MAR south of the 385

large-offset Romanche fracture zone in the south Atlantic to test ideas of biogeographic barriers to 386

vent faunal populations. Plume mapping [German et al., 2002, 2005; Devey et al., 2005] and in 387

situ mapping and sampling [Koschinsky et al., 2006; Haase et al., 2007; German et al., 2008] has 388

now revealed the presence of several active hydrothermal sites in the South Atlantic between 2° 389

and 14°S. Many of these sites appear to be hosted on basaltic substrate. At 4°48’S, there are 390

several active black smoker fields (Comfortless Cove, Turtle Pits and Red Lion) at depths ranging 391

from 3050 m at Red Lion field to 2990 m at Turtle Pits. The Turtle Pits sites are boiling and the 392

measured vent exit temperatures reach 407°C [Haase et al.,2007; Koschinsky et al., 2007]. These 393

are located within the axial rift on very fresh volcanic lava, which may have recently erupted. The 394

vent fluid chemistry has high hydrogen content relative to methane along with significantly reduced 395

chloride concentrations indicative of phase separation presumably related to a very recent intrusion 396

and lava eruption [Koschinsky et al., 2005; 2008]. Interestingly, the inactive chimneys at Turtle 397

Pits have abundant hematite in association with pyrite and magnetite [Haase et al., 2007] and the 398

presence of such oxidizing conditions may suggest that positive magnetic anomalies might be 399

associated with these deposits. Only sparse sea surface magnetic coverage is available on this 400

portion of the MAR [Bruguier et al., 2003] and the low latitude is not favorable for collecting high 401

quality magnetic data. Near-bottom ABE magnetic data were collected over the 4°48S area in 2005 402

(Plate 5), flown at an altitude of approx. 50 m with a nominal 40-50 m line spacing [German et al, 403

2008; Haase et al., 2007]. The primary magnetic sensor of ABE was not functioning for this 404

survey and so we have processed the back-up ABE sensor (Crossbow) which results in a noisier but 405

satisfactory result. Inversion analysis of these ABE magnetic data reveal zones of discrete reduced 406

magnetization associated with the Red Lion and Comfortless Cove vents areas (Plate 5d). The 407

19

Turtle Pits site also lies on the edge of a large ~200 m diameter zone of reduced magnetization 408

(Plate 5d) perhaps suggesting a wider area of alteration buried under the recently erupted lava flows 409

[Haase et al., 2007]. 410

411

Further south, the Nibelungen vent field (8°17’S) was discovered at 2905 m depth 412

[Koschinsky et al., 2005; 2006; German et al., 2008; Yoerger et al., 2007]. This vent field is 413

located on the face of a steep fault facing outward from the rift axis. Although the rocks adjacent to 414

the vent site appear to be dominantly volcanic pillow lava, the vent fluid chemistry suggests a 415

serpentinite influence. Finally, a small low temperature vent field (max. temperature 17°C), the 416

Lilliput Field, was discovered at 9°33’S at 1500 m on the summit of an axial volcanic ridge [Haase 417

et al., 2005; Koschinsky et al., 2006; Yoerger et al., 2007]. Near-bottom magnetic data over these 418

areas remain to be analyzed. 419

420

3.6 Ultramafic hosted vent deposits 421

Rocks from the mantle and the deep crust are observed on transform fault and fracture zone 422

scarps in different spreading environments, from fast (e.g. Garrett Fracture Zone or FZ) to slow 423

spreading centers (e.g. Marie-Celeste FZ; Kane FZ; Saint Paul FZ). However, only slow and 424

ultraslow spreading centers expose such deep seated rocks at the spreading axis, both as a result of 425

low magma supply [e.g. Cannat, 1993; Dick et al., 2003; Cannat et al., 2006] and tectonic 426

denudation [e.g. Tucholke et al., 1998; Smith et al., 2008]. Although outcropping mantle rocks 427

likely reflect a colder thermal regime, several active hydrothermal sites have nevertheless been 428

discovered in ultramafic environments on the Mid-Atlantic Ridge. These sites exhibit a rather large 429

diversity, including low temperature (7-9°C) fluids and diffuse flow at Saldanha (see Table-1); 430

20

focused low temperature (40-75°C) fluids and carbonate chimneys at Lost City and focused high 431

temperature fluids and sulfide precipitates at Rainbow (365°C) and Logatchev (350°C). In addition 432

to the northern MAR, evidence for ultramafic-hosted hydrothermal vents have also been found in at 433

least three locations on the Southwest Indian Ridge [Tao et al., 2007]. 434

435

To date only the Rainbow area has been surveyed for magnetism in any detail, i.e. both sea-436

surface and deep-sea magnetic surveys. Rainbow is located at 36°14’N, 33°54’W, on the western 437

side of a N-S hill, 13 km long and 8 km wide, which separates two adjacent segments (Plate 6a). 438

This hill is mostly comprised of serpentinized peridotite and may represent an oceanic core 439

complex at the footwall of a detachment fault, although no clear striations or mullions have been 440

identified so far. The hydrothermal site extends over 300 m in a E-W direction and 200 m in a N-S 441

direction, at a depth of about 2300 m (Plate 7a). It was first explored in 1997 [Fouquet et al., 1997] 442

shortly after the associated hydrothermal plume was detected [German et al., 1996; Charlou et al., 443

1996]. This plume is one of the strongest in the Atlantic Ocean as far as CH4, H2 and Fe contents 444

are concerned [Charlou et al., 2002], suggesting fluid-mantle rock interaction and serpentinization. 445

Rainbow vent fluids appear to be very reducing, highly acidic (pH 2.8) and very high in Fe 446

[Douville et al., 2002]. 447

448

Plate 6b shows the sea-surface scalar magnetic anomaly compiled from the available 449

shipborne data, and Plate 6c shows the equivalent magnetization derived from this anomaly 450

assuming a 500 m thick magnetic layer whose top is bounded by the topography. The magnetic 451

lineations show a local N20°E direction and a general N40°E direction, in agreement with, 452

respectively, the local structural direction of each ridge segment and the overall direction of the 453

21

Mid-Atlantic Ridge in the area, as mirrored in the topography (Plate 6a). The axial magnetic 454

anomaly (the normal Brunhes period, 0-0.78 Ma) is marked by the highest amplitudes, up to 40 455

A/m, crossing the map on the NE-SW diagonal. The reversed Matuyama period corresponds to the 456

negative equivalent magnetization of -20 A/m that dominates the SE and NW corners and only 457

interrupted by Chron 2 (the normal polarity Olduvai interval) marked by a positive magnetization 458

up to 20 A/m. The axial magnetic anomaly displays significant along-axis variations. The AMAR 459

segment (north of Rainbow) shows a strong equivalent magnetization (35 A/m) at the segment 460

southern end near 35°17’N and a much weaker one (5 to 10 A/m) at the segment centre near 461

35°22’N, in agreement with the general observation of Ravilly et al. [1998]. Conversely, the 462

segment south of Rainbow shows a strong equivalent magnetization all along the segment, maybe 463

because this segment is too short to exhibit the characteristic signature of a typical MAR segment. 464

The Rainbow vent site is located between the two segments within the inside corner of a non-465

transform offset in an area which exhibits a lower equivalent magnetization of 15 A/m (Plate 6c). 466

Such a value may however not be meaningful, as the computation of an equivalent magnetization 467

assumes a constant thickness of the magnetic layer – an assumption which is probably violated at 468

segment ends. 469

470

A deep-sea magnetic survey was carried out at Rainbow in 2001 by ROV Victor of 471

IFREMER using the deep-sea vector magnetometer developed by the Ocean Research Institute of 472

the University of Tokyo. The vent site was systematically surveyed at an altitude of about 10 m, 473

with profiles spaced by about 60 m. No high-resolution bathymetric data was collected, as this 474

capability was yet not available on ROV Victor. The magnetic data were corrected for the 475

magnetic effect of the ROV and inverted to magnetization using the topography and the altitude of 476

22

the submersible [Honsho, 1999, Nakase, 2003; Kitazawa, 2006; Honsho et al., 2009]. The 477

resulting map (Plate 7b) shows a very strong positive magnetization (up to 35 A/m) associated to 478

the western side of site Rainbow, whereas it does not exceed 10 A/m in other areas. Seafloor 479

submersible observations suggest that the western part of Rainbow consists of well-developed and 480

relatively mature vents edifices, whereas the three easternmost vents (Plate 7a) may have appeared 481

more recently and may still be immature [Y. Fouquet, pers. comm., 2004]. 482

483

The host rocks of the Rainbow region are comprised of a non-mineralized, lizardite-484

dominated serpentinite resulting from a static, relatively low-temperature, retrograde 485

serpentinization [Marques et al., 2006]. The sulfide-bearing serpentinites, on the other hand, are 486

characterized by a stockwork zone of semi-massive sulfides (Cu-Zn-(Co) type) that are found 487

restricted to the vent area and is not thought to be genetically linked to the earlier serpentinization 488

history [Marques et al., 2007]. This overprinting of sulfide mineralization results in zonation and 489

replacement textures more typically found in mafic systems [Doyle and Allen, 2003], but with some 490

ultramafic tendencies including Ni-rich stockwork sulfides and methane anomalies [Marques et al., 491

2007]. In terms of the source of the strong magnetization found over the main Rainbow vent area, 492

it would appear that the main serpentinite host rock is only weakly magnetic so that the formation 493

of magnetite by the serpentinization process may not be the main source of magnetization. At the 494

vent site, the serpentinized peridotite would therefore be overprinted by the stockwork sulphide 495

mineralization inferring that the source of the magnetization anomaly could be due to the presence 496

of magnetic iron sulfide such as pyrrhotite or even nickel, if present in sufficient quantity. Rock 497

magnetic measurements of the sulfide phases are needed to test this assertion. 498

499

23

In summary, the Rainbow hydrothermal site and the serpentinized peridotite hill upon which it 500

stands is associated with relatively low equivalent magnetization inverted from the sea-surface 501

magnetic anomalies, between two magnetic highs associated with the nearby ridge segment ends 502

(Plate 6d). Conversely, the vent field itself corresponds to a very strong magnetic signal at its 503

western edge as inverted from deep sea vector magnetic anomalies. These observations can be 504

reconciled by recognizing that both data sets do not reflect the same scale of features of the 505

seafloor. Deep-sea magnetic anomalies recorded at 1 – 10 m above seafloor are much more 506

sensitive to shallow sources (i.e. in the uppermost tens to hundreds of meters) which dominate the 507

signal, whereas sea surface magnetic anomalies, recorded 2500 m above the seafloor, integrate the 508

effect of the magnetization carried by all layers of the oceanic crust. The “crust”, in this case 509

serpentinized peridotite, around the Rainbow vent site has a relatively low magnetization, upon 510

which is superimposed a smaller but more strongly magnetized shallow body. A likely magnetic 511

source for this superficial body is magnetic sulfide mineralization possibly a stockwork zone of 512

pyrrhotite or perhaps more exotic minerals of Ni composition. 513

514

The high-temperature ultramafic-hosted Logatchev and Achadze vent sites have also 515

recently been the target of deep-sea high resolution bathymetric and magnetic measurements 516

[Fouquet et al., 2007; Ondreas et al., 2007]. These results are currently being analyzed but are 517

required to generalize the observations gathered at Rainbow to the focused, high-temperature 518

hydrothermal vents in an ultramafic environment. Furthermore, a very high resolution 519

photographic, bathymetric and magnetic study carried out at Rainbow in mid-2008 will help to 520

better constrain the source of the strong Rainbow magnetic anomaly. The magnetic signature of 521

other low temperature ultramafic sites such as Saldanha (diffuse, very low temperature flow) and 522

24

Lost City (focused, low temperature vents) remains relatively unknown at present. Rock magnetic 523

measurements have been made on basaltic samples from the Saldanha area showing typical low-524

temperature alteration but no measurement of the magnetic properties was carried out on ultramafic 525

rocks from the hydrothermal area [Miranda et al., 2002]. A deeptow magnetic anomaly profile 526

across the Lost City vent field showed a pronounced low suggesting that serpentinization is not 527

contributing to the magnetic signal [Gee and Blackman, 2004]. Finally, at least three hydrothermal 528

plumes have recently been detected by Chinese investigators along the ultraslow Southwest Indian 529

Ridge [Tao et al., 2007], an area which displays many serpentinite outcrops [Cannat et al., 2006]. 530

Given these recent results, it is likely that the number of hydrothermal sites located on serpentinite 531

outcrops are more numerous than first thought. Indeed, it is also likely that slow spreading ridges 532

in general show a more diverse range of hydrothermal systems than might have been predicted only 533

5 years ago (e.g. Escartin et al., 2008). 534

535

Conclusions 536

We can make the following generalizations and conclusions concerning the magnetic response of 537

hydrothermal systems in the slow spreading midocean ridge environment. 538

539

1) In basaltic-hosted hydrothermal vent systems, crustal magnetization is strongly impacted by 540

hydrothermal fluid circulation, particularly in the upflow zones where buoyant high-temperature 541

fluid rises to the seafloor through the crust. The lateral scale of these upflow zones appears to be 542

on the order of 100 m or less compatible with observations of ancient systems found in ophiolites 543

[e.g. Richards et al., 1989]. The depth of these upflow zones remains to be quantified but they are 544

likely to extend through the extrusive crust into the intrusive dike section. Further, also by analogy 545

25

to ancient deposits, the lateral alteration gradients in the upflow zone are likely to be quite sharp 546

resulting in well-defined magnetic contrasts [e.g. Tivey and Johnson 2002]. 547

548

2) From the few high-temperature, focused-flow ultramafic-hosted systems surveyed to date 549

(Rainbow), we find strong positive magnetic anomalies possibly associated with magnetic sulfide 550

deposition or perhaps other metallic mineral compounds that are magnetic such as Ni (or Cr-Fe). If 551

confirmed, the role of magnetite formation as a result of serpentinization of the host peridotite may 552

be only of secondary effect in producing magnetic anomalies. More work is needed to fully 553

characterize this conclusion. 554

555

3) Little to no magnetic work has been done over low-temperature ultramafic hosted vent systems 556

like Saldanha or Lost City. Preliminary indications from a near-bottom magnetic profile over Lost 557

City are that while serpentinization is occurring at depth, it is not producing sufficient magnetite to 558

create a positive anomaly at the vent site [Gee and Blackman, 2004]. More work is also needed 559

here to fully quantify the magnetic effects in these low temperature, low flow ultramafic hosted 560

systems. 561

562

4) The slow spreading environment as exemplified by the Mid-Atlantic Ridge shows a remarkable 563

diversity of hydrothermal systems that was perhaps unanticipated a few years ago. The diversity of 564

host rocks found in slow spreading environments also allows for a greater range in the types of 565

mineralization found at slow spreading ridges. Also, the role of low-angle faulting and detachment 566

tectonics provides a newly recognized pathway for fluid flow that will allow for hydrothermal 567

systems to maintain activity away from the conventional spreading axis environment. It was 568

26

already known that the slow spreading environment hosts longer lived hydrothermal systems than 569

fast to intermediate spreading centers where volcanic activity may bury vent systems on a regular 570

basis. This proclivity to longer lived activity would potentially lead to larger hydrothermal deposits 571

and perhaps a greater chance of preservation in the off-axis realm. 572

573

5) Magnetic imaging provides a viable method to search for these hydrothermal deposits and with 574

advances in gradiometer and vector measurement technology, which would allow for source depth 575

estimation and improved lateral interpolation of spatial anomaly character, we foresee better 576

quantification of these anomalous magnetic zones in the future. 577

578

Acknowledgments 579

JD acknowledge discussions with and support from Yves Fouquet (IFREMER) and Kensaku 580

Tamaki (University of Tokyo) as well as many individuals from CNRS-INSU, IFREMER, and 581

ORI-University of Tokyo who made cruise IRIS of R/V L'Atalante on site Rainbow a success. This 582

is IPGP contribution 2518. MAT acknowledges support for this work by the Woods Hole 583

Oceanographic Institution (WHOI) Deep Ocean Exploration Institute Fellowship funds. MAT 584

would also like to thank the many members of the WHOI Deep Submergence Lab and the crews of 585

research vessels that have collected magnetic data for him on various vehicles over the past few 586

years. MAT thanks Chris German for permission to use the South Atlantic ABE magnetic data. 587

Helpful reviews from an anonymous reviewer, Joaquim Luis, and volume editor, Peter Rona, 588

greatly improved a previous draft of this manuscript.589

27

References 590

Ade-Hall, J.M., H.C. Palmer, and T.P. Hubbard, The magnetic and opaque petrological response of 591

basalt to regional hydrothermal alteration, Geophys. J. R. astr. Soc., 24, 137-174, 1971. 592

Ade-Hall, J.M., F. Aumento, P.J.C. Ryall, R.E. Gerstein, J. Brooke, and D.L. McKeown, The mid-593

Atlantic ridge near 45°N, 21, Magnetic results from basalt drill cores from the median 594

valley, Can. J. Earth Sci., 10, 679-696, 1973. 595

Barriga, F.J.A.S., Y. Fouquet, A. Almeida, M. Biscoito, J.-L. Charlou, R.L.P. Costa, A. Dias, 596

A.M.S.F. Marques, J.M.A. Miranda, K. Olu, F. Porteiro, and M.G. P.S. Queiroz, Discovery 597

of the Saldanha Hydrothermal Field on the FAMOUS Segment of the MAR (36°30 N). 598

AGU Fall Meeting, Eos Transactions, 79 (45), F67, 1998. 599

Beltenev, V., A. Nescheretov, V. Shilov, A. Shagin, T. Stepanova, G. Cherkashev, B. Batuev, M. 600

Samovarov, I. Rozhdestvenskaya, I. Andreeva, I. Fedorov, M. Davydov, L. Romanova, A. 601

Rumyantsev, V. Zaharov, N. Lunaeva and O. Artem'eva, New discoveries at 12°58'N, 602

44°52'W, MAR: Professor Logatchev-22 cruise, initial results, InterRidge News, 12 (1), 13-603

14, 2003. 604

Beltenev, V., A. Shagin, V. Markov, I. Rozhdestvenskaya, T. Stepanova, G. Cherkashev, I. 605

Fedorov and I. Poroshina, A new hydrothermal field at 16°38.4'N, 46°28.5'W on the Mid-606

Atlantic Ridge, InterRidge News, 13, 5-6, 2004. 607

Bina, M.M., and B. Henry, Magnetic properties, opaque mineralogy and magnetic anisotropies of 608

serpentinized peridotites from ODP Hole 649, Phys. Earth Planet. Int., 65, 88-103, 1986. 609

Bonatti, E., Metallogenesis at oceanic spreading centers, Ann. Rev. Earth Planet. Sci., 3, 401-431, 610

1975. 611

28

Bostrom, K., and M.N.A. Peterson, Precipitates from hydrothermal exhalations of the East Pacific 612

Rise, Econ. Geol., 61, 1258-1265, 1966. 613

Browne, P.R.L., Hydrothermal alteration in active geothermal fields, Ann. Rev. Earth Planet. Sci., 614

6, 229-250, 1978. 615

Bruguier, N.J., T.A. Minshull, and J.M. Brozena, Morphology and Tectonics of the Mid-Atlantic 616

Ridge, 7–12°, J. Geophys. Res., 108, 2093, doi:10.1029/2001JB001172, 2003. 617

Campbell, A.C., M. R. Palmer, G. Klinkhammer, T. S. Bowers, J. Edmond, J. Lawrence, J. Casey, 618

G. Thompson, S. Humphries, P. Rona, and J. Karson, Chemistry of hot springs on the Mid-619

Atlantic Ridge: TAG and MARK sites, Nature, 335, 514-519, 1988. 620

Canales, J.P., R. Sohn, and B.J. DeMartin, Crustal structure of the Trans-Atlantic Geotraverse 621

(TAG) segment (Mid-Atlantic Ridge, 26°10'N): Implications for the nature of hydrothermal 622

circulation and detachment faulting at slow spreading ridges, Geochem. Geophys. Geosys., 623

8 (Q08004), doi:10.1029/2007GC001629, 2007. 624

Cannat, M., Emplacement of mantle rocks in the seafloor at mid-ocean ridges, J. Geophys. Res., 98, 625

4163-4172, 1993. 626

Cannat, M., A. Briais, C. Deplus, J. Escartín, J. Georgen, J. Lin, S. Mercouriev, C. Meyzen, M. 627

Muller, G. Pouliquen, A. Rabain, and P. da Silva, Mid-Atlantic ridge - Azores hotspot 628

interactions: Along-axis migration of a hotspot-derived magmatic pulse 10 to 4 myrs ago, 629

Earth Planet. Sci. Lett., 173, 257-269, 1999. 630

Cannat, M., D. Sauter, V. Mendel, E. Ruellan, K. Okino, J. Escartin , V. Combier, and M. Baala, 631

Modes of seafloor generation at a melt-poor ultraslow spreading ridge, Geology, 34 (7), 632

605, 2006. 633

29

Cannat, M., H. Ondreas, Y. Fouquet, S. Silantiev, E. Hoise, F. Fontaine, and Scientific Party, 634

Geological context of ultramafic-hosted hydrothermal vent fields in the 13-15N region of 635

the Mid-Atlantic Ridge: Preliminary results of the Serpentine cruise, Eos Trans. AGU, 88 636

(52), Fall Meeting Suppl. Abstract T51F-02, 2007. 637

Charlou, J.-L., H. Bougault, Y. Fouquet, J.P. Donval, E. Douville, J. Radford-Knoery, M. Aballea, 638

H.D. Needham, P. Jean-Baptiste, P.A. Rona, C. Langmuir, and C.R. German, Methane 639

degassing, hydrothermal activity and serpentinization between the Fifteen Twenty fracture 640

zone area and the Azores Triple Junction area (Mid-Atlantic Ridge), J. Conf. Abstr., 1 (2), 641

771-772, 1996. 642

Charlou, J.-L., Fouquet, Y., Bougault H., Donval J.P., Etoubleau J., Jean-Baptiste P., Dapoigny A., 643

Appriou P. and P.A. Rona, Intense CH4 plumes generated by serpentinization of ultramafic 644

rocks at the intersection of the 15°20'N fracture zone and the Mid-Atlantic Ridge, Geochim. 645

Cosmo. Acta, 62 (13), 2323-2333, 1998. 646

Charlou, J.-L., J.P. Donval, E. Douville, P. Jean Baptiste, J. Radford-Knoery, Y. Fouquet, A. 647

Dapoigny, and M. Stievenard, Compared geochemical signatures and the evolution of 648

Menez Gwen (37°50'N) and Lucky Strike (37°17'N ) hydrothermal fluids, south of the 649

Azores Triple Junction on the Mid-Atlantic Ridge, Chemical Geology, 171, 49-75, 2000. 650

Charlou, J.-L., J.P. Donval, Y. Fouquet, P. Jean Baptiste, and N. Holm, Geochemistry of high H2 651

and CH4 vent fluids issuing from ultramafic rocks at the Rainbow hydrothermal field 652

(36°14'N, MAR), Chemical Geology, 191, 345-359, 2002. 653

Cherkashov, G.A., A.M. Ashadze, and A.V. Gebruk, New fields with manifestations of 654

hydrothermal activity in the Logatchev area (14N, Mid-Atlantic Ridge), InterRidge News, 9 655

(2), 26-27, 2000. 656

30

Corliss, J.B., J. Dymond, L.I. Gordon, J.M. Edmond, R.P. von Herzen, R.D. Ballard, K.E. Green, 657

D. Williams, A. Bainbridge, K. Crane, and T.H. van Andel, Submarine thermal springs on 658

the Galapagos rift, Science, 203, 1073, 1979. 659

Desbruyères, D., M. Biscoito, J.-C. Caprais, A. Colaço, T. Comtet, P. Crassous, Y. Fouquet, A. 660

Khripounoff, N. Le Bris, K. Olu, R. Riso, P.-M. Sarradin, M. Segonzac, and A. 661

Vangriesheim, Variations in deep-sea hydrothermal vent communities on the Mid-Atlantic 662

Ridge near the Azores plateau, Deep-Sea Research, 48, 1325-1346, 2001. 663

Degens, E.T., and D.A. Ross, Hot Brines and Recent Heavy Metal Deposits in the Red Sea, 600 664

pp., Springer-Verlag, New York, N.Y., 1969. 665

deMartin, B.J., R.A. Sohn, J.P. Canales, and S.E. Humphris, Kinematics and geometry of active 666

detachment faulting beneath the Trans-Atlantic Geotraverse (TAG) hydrothermal field on 667

the Mid-Atlantic Ridge, Geology, 35 (8), 711-714, 2007. 668

Detrick, R.S., J. Honnorez, and ODP Leg 106 Scientific Party, Drilling the Snake Pit hydrothermal 669

sulfide deposit on the Mid-Atlantic Ridge, Geology, 14, 1004-1007, 1986. 670

Devey, C.W., K.S. Lackschewitz, and E.T. Baker, Hydrothermal and volcanic activity found on the 671

southern Mid-Atlantic Ridge, EOS, 86, 209-212, 2005. 672

Dias, A.S., and F.J.A.S. Barriga, Mineralogy and geochemistry of hydrothermal sediments from the 673

serpentinite-hosted Saldanha hydrothermal field (36°34'; 33°26'W) at MAR., Mar. Geol., 674

225, 157-175, 2006. 675

Dick, H.J.B., J. Lin, and H. Schouten, An ultraslow-spreading class of ocean ridge, Nature, 426, 676

405-412, 2003. 677

Douville, E., J.-L. Charlou, E.H. Oelkers, P. Bienvenu, C.C.F. Jove, J.P. Donval, Y. Fouquet, D. 678

Prieur, and P. Appriou, The Rainbow Vent fluids (36 degrees 14'N, MAR); the influence of 679

31

ultramafic rocks and phase separation on trace metal content n Mid-Atlantic Ridge 680

hydrothermal fluids, Chemical Geology,184, 37-48, 2002. 681

Doyle, M., and R. Allen, Subsea floor replacement in volcanic-hosted massive sulfide deposits, Ore 682

Geol. Rev., 23, 183-222, 2003. 683

Dubois, J., The Gravinaut Cruise: Seafloor gravity and electromagnetism at the MARK-Snake Pit 684

area, InterRidge News, 3 (1), 3-6, 1994. 685

Dunlop, D.J., and M. Prevot, Magnetic properties and opaque mineralogy of drilled submarine 686

intrusive rocks, Geophys. J., 16, 763-802, 1982. 687

Dyment, J., K. Tamaki, H. Horen, Y. Fouquet, K. Nakase, M. Yamamoto, M., Ravilly, and M. 688

Kitazawa, A positive magnetic anomaly at Rainbow hydrothermal site in ultramafic 689

environment, Eos Trans. AGU, 86(52), Fall Meet. Suppl., Abstract OS21C-08, 2005. 690

Escartin , J., M. Cannat, G. Pouliquen, A. Rabain, and J. Lin, Constraints on the interaction 691

between the Mid-Atlantic Ridge and the Azores hotspot from bathymetry and gravity (36-692

39N), J. Geophys. Res., 106, 21719-21736, 2001. 693

Escartin, J., S. Soule, A. Bezos, M. Cannat, D. Fornari, V. Ballu, S. Humphris, and G.S. Party, 694

Patterns of volcanism and tectonism at a slow-spreading segment of the Mid-Atlantic Ridge 695

(Lucky Strike, 37N): preliminary results from near-bottom geological and geophysical 696

surveys, EOS Trans. AGU, 87 (52), Fall Meet. Suppl., Abstract B33D-05, 2006. 697

Escartin , J., D.K. Smith, J. Cann, H. Schouten, C.L. Langmuir, and S. Escrig, Central role of 698

detachment faults in accretion of slow spreading oceanic lithosphere, Nature (in press), 699

2008. 700

Fornari, D.J., S.E. Humphris, L.M. Parson, P. Blondel, and C.R. German, Near-bottom 701

investigations in the Lucky Strike segment-MAR 37°N using the fiber optic tethered 702

32

vehicles of the UNOLS national deep submergence facility: ROV Jason, ARGO-II and the 703

DSL-120 kHz sonar., Eos Trans. AGU, 77, F699, 1996. 704

705

Fouquet, Y., A. Wafik, P. Cambon, C. Mevel, G. Meyer, and P. Gente, Tectonic setting and 706

mineralogical and geochemical zonation in the Snake Pit sulfide deposit (Mid-Atlantic 707

Ridge at 23°N), Econ. Geol., 88 (8), 2018-2036, 1993. 708

Fouquet, Y., J.-L. Charlou, J.-P. Donval, J. Radford-Knoery, H. Pelle, H. Ondreas, N. Lourenco, M. 709

Segonzac, and M.K. Tivey, A Detailed Study of the Lucky Strike Hydrothermal Site and 710

Discovery of a New Hydrothermal Site: Menez Gwen; preliminary Results of the DIVA1 711

Cruse (5-29 May, 1994), InterRidge News, 3, 14-19, 1994a. 712

Fouquet, Y., J.-L. Charlou, J.-P. Donval, J. Radford-Knoery, H. Pellé, H. Ondréas, M. Ségonzac, I. 713

Costa, N. Lourenco, and M. K. Tivey, Geological setting and comparison of the Menez 714

Gwen and Lucky Strike vent fields at 37°17 N and 37°50 N on the Mid-Atlantic Ridge. 715

Preliminary results of the DIVA1 diving cruise with Nautile, Eos Trans. AGU, 75 (44), Fall 716

Meet. Suppl., 313, 1994b. 717

Fouquet, Y., H. Ondreas, J.-L. Charlou, J.-P. Donva, J. Radford-Knoery, I. Costa, N. Lourenco, and 718

M.K. Tivey, Atlantic lava lakes and hot vents, Nature, 377, 201, 1995. 719

Fouquet, Y., J.-L. Charlou, H. Ondreas, J. Radford-Knoery, E. Donval, E. Douville, R. Apprioual, 720

P. Cambon, H. Pelle, J.Y. Landure, A. Normand, E. Poncevera, C. German, L. Parson, F. 721

Barriga, I. Costa, J. Relvas, and A. Ribeiro, Discovery and first submersible investigations 722

on the Rainbow hydrothermal field on the MAR (36°14'N). Eos Trans. AGU, 78, F832, 723

1997. 724

33

Fouquet, Y., G. Cherkashov, J.-L. Charlou, H. Ondreas, M. Cannat, N. Bortnikov, S. Silantiev, J. 725

Etoubleau, and Scientific Party, Diversity of ultramafic hosted hydrothermal deposits on the 726

Mid-Atlantic Ridge: First submersible studies of Ashadze, Logatchev 2 and Krasnov vent 727

fields during the Serpentine cruise, Eos Trans. AGU, 88 (52), Fall Meeting Suppl. Abstract 728

T51F-03, 2007. 729

Gebruk, A.V., L.I. Moskalev, P. Chevaldonne, S.M. Sudarikov and E.S. Chernyaev, Hydrothermal 730

vent fauna of the Logatchev area (14 45, MAR): preliminary results from the first Mir and 731

Nautile dives in 1995, InterRidge News, 6 (2), 10-14, 1997. 732

Gee, J., and D.K. Blackman, Lineated near-bottom magnetic anomalies over an oceanic core 733

complex, Atlantis Massif (Mid-Atlantic Ridge at 30N), EOS Trans. AGU, 85 (17), Fall 734

Meeting Abstract: GP31A-15, 2004. 735

Gente, P., G. Ceuleneer, O. Dauteuil, J. Dyment, C. Honsho, C. Laverne, C. Le Turdu, N.C. 736

Mitchell, M. Ravilly, and R. Thibaud, On- and off-axis submersible investigations on an 737

highly magmatic segment of the Mid-Atlantic Ridge (21°40'N): the TAMMAR cruise, 738

InterRidge News, 5 (2), 27-31, 1996. 739

German, C.R., J. Briem, C. Chin, M. Danielsen, S. Holland, R. James, A. Jonsdottir, E. Ludford, C. 740

Moser, J. Olafsson, M.R. Palmer, and M.D. Rudnicki, Hydrothermal activity on the 741

Reykjanes Ridge: the Steinaholl vent-field at 63°06'N, Earth and Planet. Sci. Lett., 121, 742

647-654, 1994.German, C.R., G.P. Klinkhammer, and M.D. Rudnicki, The Rainbow 743

hydrothermal plume, 36°15'N, MAR, Geophys. Res. Lett., 23 (21), 2979-2982, 1996. 744

German, C.R., K. Richards, M.M. Lam, A. Thurnherr, J.L. Charlou, J. Radford-Knoery, J.P. 745

Donval, P. Jean-Baptiste, A.M. Le Clerc, A. Dapoigny, M. Cooper, M. Rudnicki, H. 746

Elderfield, H. Edmonds, D. Green, D. O'Brien, J. Patching, A. Khripounoff, P. Crassous, A. 747

34

Vangrisheim, D. Dixon, and P. Herring, A segment scale study of fluxes through the 748

Rainbow hydrothermal plume, 36N Mid-Atlantic Ridge, Eos Trans. AGU, 80 (46), 957-958, 749

1999. 750

German, C.R., D.P. Connelly, A.J. Evans, and L.M. Parson, Hydrothermal activity on the Southern 751

Mid-Atlantic Ridge, EOS Trans. AGU, 83 (47), Fall Meet. Suppl., Abstract V61B-1361, 752

2002. 753

German, C.R., and Scientific party, Hydrothermal activity on the southern Mid-Atlantic Ridge: 754

tectonically- and volcanically-hosted high temperature venting at 2-7°S, EOS Trans. AGU, 755

86, Fall Meet. Suppl., Abstract OS21C-04, 2005. 756

German, C.R., S.A. Bennett, D.P. Connelly, A.J. Evans, B.J. Murton, L.M. Parson, R.D. Prien, E. 757

Ramirez-Llodra, M. Jakuba, T.M. Shank, D.R. Yoerger, E.T. Baker, S.L. Walker, and K. 758

Nakamura, Hydrothermal activity on the southern Mid-Atlantic Ridge: Tectonically and 759

volcanically-controlled venting, 4-5°S, Earth Planet. Sci. Lett., 273, 332-344, 2008. 760

Grindlay, N.R., P.J. Fox, and P. Vogt, Morphology and tectonics of the Mid-Atlantic Ridge (25°-761

27°30'S) from Sea Beam and Magnetic data, J. Geophys. Res., 97, 6983-7010, 1992. 762

Hall, J.M., Interaction of submarine volcanic and high-temperature hydrothermal activity proposed 763

for the formation of the Agrokipia volcanic massive sulfide deposits of Cyprus, Can. J. 764

Earth Sci., 29, 1928-1936, 1992. 765

Haase, K.M., and Scientific Party, Young volcanism and related hydrothermal activity at 5°S on the 766

slow-spreading southern Mid-Atlantic Ridge, Geochem. Geophys. Geosys., 8 (11), Q11002, 767

doi:10.1029/2006GC001509, 2007. 768

Haase, K.M., and Scientific Party, Hydrothermal activity and volcanism on the southern Mid-769

Atlantic Ridge, EOS Trans. AGU, 86, Fall Meet. Suppl., Abstract OS21C-05, 2005. 770

35

Hannington, M., P. Herzig, P. Stoffers, J. Scholten, R. Botz, D. Garbe-Schönberg, I.R. Jonasson, 771

W. Roest, and Shipboard Scientific Party, First observations of high-temperature submarine 772

hydrothermal vents and massive anhydrite deposits off the north coast of Iceland, Marine 773

Geology, 177, 199-220, 2001. 774

Hannington, M.D., C.E.J. De Ronde, and S. Petersen, Sea-floor tectonics and submarine 775

hydrothermal systems, Econ. Geol, 100, 111-141, 2005. 776

Herzig, P.M., S.E. Humphris, D.J. Miller, and R.A. Zierenberg, Proc. ODP Scient. Results, Ocean 777

Drilling Program, College Station, TX, 1998. 778

Hochstein, M.P., and S. Soengkono, Magnetic anomalies associated with high temperature 779

reservoirs in the Taupo volcanic zone (New Zealand), Geothermics, 26 (1), 1-24, 1997. 780

Honsho, C., Magnetic structure of the Mid-Atlantic Ridge 21°40’N segment: an approach from 781

near-bottom magnetic measurements onboard a submersible, Ph.D thesis, University of 782

Tokyo, Tokyo, 1999. 783

Honsho, C., J. Dyment, K. Tamaki, M. Ravilly, H. Horen, and P. Gente (2009), Magnetic structure 784

of a slow spreading ridge segment: Insights from near-bottom magnetic measurements on 785

board a submersible, J. Geophys. Res., 114, B05101, doi:10.1029/2008JB005915 786

Humphris, S.E., P.M. Herzig, D.J. Miller, J.C. Alt, K. Becker, D. Brown, G. Brugmann, H. Chiba, 787

Y. Fouquet, J.B. Gemmell, G. Guerin, M.D. Hannington, N.G. Holm, J.J. Honnorez, G.J. 788

Iturrino, R. Knott, R. Ludwig, K. Nakamura, S. Petersen, A.-L. Reysenbach, P.A. Rona, S. 789

Smith, A.A. Sturz, M.K. Tivey, and X. Zhao, The internal structure of an active sea-floor 790

massive sulphide deposit, Nature, 377, 713-716, 1995. 791

Humphris, S.E., P.M. Herzig, D.J. Miller, and et al., Proc. ODP, Init. Repts., Ocean Drilling 792

Program, College Station, TX, 1996. 793

36

Humphris, S.E., D.J. Fornari, D.S. Scheirer, C.R. German, and L.M. Parson, Geotectonic setting of 794

hydrothermal activity on the summit of Lucky Strike seamount (37°17'N, Mid-Atlantic 795

Ridge), Geochemistry, Geophysics, Geosystems, 3 (8), 10.1029/2001GC000284, 2002. 796

Hutchinson, R.W., Volcanogenic sulfide deposits and their metallogenic significance, Econ. Geol, 797

68, 1223-1246, 1973. 798

Irving, E., The mid-Atlantic ridge at 45°N, XIV. Oxidation and magnetic properties of basalt; 799

review and discussion, Can. J. Earth Sci., 7, 1528-1538, 1970. 800

Johnson, H.P., and R.T. Merrill, Magnetic and mineralogical changes associated with the low-801

temperature oxidation of magnetite, J. Geophys. Res., 77, 334-341, 1972. 802

Johnson, H.P., and R.T. Merrill, Low-temperature oxidation of a titanomagnetite and implications 803

for paleomagnetism, J. Geophys. Res., 78, 4938-4949, 1973. 804

Johnson, H.P., and T. Atwater, Magnetic study of basalts from the Mid-Atlantic ridge, lat. 37oN, 805

Geol. Soc. Am. Bull., 88, 637-647, 1977. 806

Johnson, H.P., J.L. Karsten, F.J. Vine, G.C. Smith, and G. Schonharting, A low-level magnetic 807

survey over a massive sulfide ore body in the Troodos ophiolite complex, Cyprus, Mar. 808

Technol. Soc. J., 16, 76, 1982. 809

Karson, J., and J. Brown, Geologic setting of the Snake Pit Hydrothermal Site: An active vent field 810

in the Mid-Atlantic Ridge, Mar. Geophys. Res., 10, 91-107, 1988. 811

Kelley, D.S., J.A. Karson, D.K. Blackman, G.L. Früh-Green, D.A. Butterfield, D.A. Lilley, E.J. 812

Olson, M.O. Schrenk, K.K. Roe, G.T. Lebon, P. Rivizzigno, and A.-S. Party, An off-axis 813

hydrothermal vent field near the Mid-Atlantic Ridge at 30°N, Nature, 411, 145-149, 2001. 814

Kitazawa, M., Datation du fond oceanique: Apport des anomalies magnetiques de haute resolution 815

acquises par submersible, Ph.D thesis, Institut de Physique du Globe de Paris, Paris, 2006. 816

37

Kleinrock, M.C., and S.E. Humphris, Structural asymmetry of the TAG rift valley: evidence from a 817

near-bottom survey for episodic spreading, Geophys. Res Lett., 23 (23), 3439-3442, 1996. 818

Kong, L., W.B.F. Ryan, L.A. Mayer, R.S. Detrick, and K. Manchester, Bare-rock drill sites ODP 819

Legs 106 and 109: Evidence for hydrothermal activity at 23°N on the Mid-Atlantic Ridge, 820

EOS Trans. AGU, 66 (46), 936, 1985. 821

Kong, L.S., S.C. Solomon, and G.M. Purdy, Microearthquake characteristics of a mid-ocean ridge 822

along-axis high, J. Geophys. Res., 97, 1659-1685, 1992. 823

Koschinsky, A., D. Garbe-Schönberg, R. Seifert, H. Strauss, S. Weber, and H. Marbler, 400°C hot 824

boiling fluids from a hydrothermal field at 5°S on the Mid-Atlantic Ridge: Results of 825

Meteor Cruise M64/1, EOS Trans. AGU, 86, Fall Meet. Suppl., Abstract OS33A-1456, 826

2005. 827

Koschinsky, A., and Scientific Party, 4Discovery of new hydrothermal vent on the southern Mid-828

Atlantic Ridge (4°S - 10°S) during Cruise M68/1, InterRidge News, 15, 9-15, 2006. 829

Koschinsky, A., D. Garbe-Schönberg, S. Sander, K.L. Schmidt, H.-H. Gennerich, and H. Strauss, 830

Hydrothermal venting at pressure-temperaure conditions above the critical point of 831

seawater, 5°S on the Mid-Atlantic Ridge, Geology, 36 (8), 615-618, 2008. 832

Lalou, C., J.L. Reyss, E. Brichet, M. Arnold, G. Thompson, Y. Fouquet, and P.A. Rona, New age 833

data for Mid-Atlantic Ridge hydrothermal sites: TAG and Snakepit geochronology 834

revisited, J. Geophys. Res., 98, 9705-9713, 1993. 835

Lalou, C., J.L. Reyss, and E. Brichet, Age of sub-bottom suphide samples at the TAG active 836

mound, in Proc. ODP Scient. Results, edited by P.M. Herzig, S.E. Humphris, D.J. Miller, 837

and R.A. Zierenberg, pp. 111-118, Ocean Drilling Program, College Station, TX, 1998. 838

38

Langmuir, C., S. Humphris, D. Fornari, C. Van Dover, K. Von Damm, M.K. Tivey, D. Colodner, 839

J.-l. Charlou, D. Desonie, C. Wilson, Y. Fouquet, G. Klinkhammer, and H. Bougault, 840

Hydrothermal vents near a mantle hot spot: the Lucky Strike vent field at 37°N on the Mid-841

Atlantic Ridge, Earth Planet. Sci. Lett., 148, 69-91, 1997. 842

Lisitsyn, A.P., Y.A. Bogdanov, L.P. Zonenshain, M.I. Kuz'min, and A.M. Sagalevich, 843

Hydrothermal phenomena in the Mid-Atlantic Ridge at Lat. 26°N (TAG hydrothermal 844

field), Int. Geol. Rev., 31, 1183-1198, 1989. 845

Lourenço, N., J.M. Miranda, J.F. Luis, A. Ribeiro, L.A. Mendes Victor, J. Madeira, and H.D. 846

Needham, Morpho-tectonic analysis of the Azores volcanic plateau from a new bathymetric 847

compilation of the area, Mar. Geophys. Res., 20 (141-156), 1998. 848

Macdonald, K.C., S.P. Miller, S.P. Huestis, and F.N. Spiess, Three-dimensional modeling of a 849

magnetic reversal boundary from inversion of deep-tow measurements, J. Geophys. Res., 850

85, 3670-3680, 1980. 851

Marques, A.F.A., F. Barriga, V. Chavangnac, and Y. Fouquet, Mineralogy, geochemistry and Nd 852

isotope composition of the Rainbow hydrothermal field, Mid-Atlantic Ridge, Mineralium 853

Deposita, 41 (2), 52-67, 2006. 854

Marques, A.F.A., F.J.A.S. Barriga, and S.D. Scott, Sulfide mineralization in an ultramafic-rock 855

hosted seafloor hydrothermal system: From serpentinization to the formation of Cu-Zn-856

(Co)-rich massive sulfides, Marine Geology, 245, 20-39, 2007. 857

McGregor, B.A., and P.A. Rona, Crest of Mid-Atlantic Ridge at 26°N, J. Geophys. Res., 80, 3307-858

3314, 1975. 859

McGregor, B.A., C.G.A. Harrison, J.W. Lavelle, and P.A. Rona, Magnetic anomaly pattern on the 860

Mid-Atlantic Ridge crest at 26°N., J. Geophys. Res., 82, 231-238, 1977. 861

39

Melchert, B., C.W. Devey, C.R. German, K.S. Lackschewitz, R. Seifert, M. Walter, C. Mertens, 862

D.R. Yoerger, E.T. Baker, H. Paulick, and K. Nakamura, First evidence for high-863

temperature off-axis venting of deep crustal/mantle heat: The Nibelungen hydrothermal 864

field, southern Mid-Atlantic Ridge, Earth and Planet. Sci. Lett., 275, 61-69, 2008. 865

Mevel, C. ; J.M. Auzende; M. Cannat; J.P. Donval; J. Dubois; Y. Fouquet; P. Gente; D. Grimaud; 866

J.A. Karson; M. Segonzac; M. Stievenard, The Snake Pit ridge, M.A.R., 23°22'N: first 867

results of the HYDROSNAKE cruise, C. R. Acad. Sci. Paris, Ser. 2, 308, 545-552, 1989. 868

Miranda, J.M., P.F. Silva, N. Lourenço, B. Henry, R. Costa, and Saldanha Team, Study of the 869

Saldanha Massif (MAR): constraints from rock magnetic and geophysical data, Mar. 870

Geophys. Res., 23, 299-318, 2002. 871

Miranda, J.M., J.F. Luis, N. Lourenço and F.M. Santos, Identification of the magnetization low of 872

the Lucky Strike hydrothermal vent using surface magnetic data, J. Geophys. Res., 110, 873

B04103, doi:10.1029/2004JB003085, 2005. 874

Murton, B.J., G. Klinkhammer, K. Becker, A. Briais, D. Edge, N. Hayward, N. Millard, I. Mitchell, 875

I. Rouse, M. Rudnicki, K. Sayanagi, H. Sloan and L. Parson, Direct evidence for the 876

distribution and occurrence of hydrothermal activity between 27°N–30°N on the Mid-877

Atlantic Ridge, Earth Planet. Sci. Lett., 125, 119-128, 1994. 878

Murton, B., M. Unsworth, M. Harris, C. MacLeod, R. Searle, J. Casey, K. Achenbach, and C. 879

Mallows, Fluid flow at active oceanic core complexes, 13N Mid-Atlantic Ridge, Eos Trans. 880

AGU, 88 (52), Fall Meeting Suppl. Abstract T53B-1309, 2007. 881

Nakase, K., Magnetic structure of the Rainbow hydrothermal field at the Mid-Atlantic Ridge 882

36°14'N obtained from near-bottom geomagnetic measurements, M.Sc. thesis, University of 883

Tokyo, Tokyo, 2002. 884

40

Nazarova, K., Serpentinized peridotites as a possible source for oceanic magnetic anomalies, Mar. 885

Geophys. Res., 16, 455-462, 1994. 886

Needham, H.D., M. Voisset, V. Renard, H. Bougault, O. Dauteuil, R. Detrick, and C. Langmuir, 887

Structural and volcanic features of the Mid-Atlantic rift zone between 40°N and 33°N, EOS 888

Trans. AGU, 73, 552, 1992. 889

Ondreas, H., Y. Fouquet, M. Voisset, and J. Radford-Knoery, Detailed study of three contiguous 890

segments of the Mid-Atlantic Ridge, South of the Azores (37°N to 38°30'N), using acoustic 891

imaging coupled with submersible observations, Mar. Geophys. Res., 19, 231-255, 1997. 892

Ondreas, H., M. Cannat, G. Cherkashov, Y. Fouquet, A. Normand, and Serpentine Scientific Party, 893

High-resolution mapping of the Ashadze and Logatchev hydrothermal fields, Mid-Atlantic 894

Ridge 13-15°N, Eos Trans. AGU, 88 (52), Fall Meeting Suppl. Abstract T51F-03, 2007. 895

Ondreas, H., M. Cannat, Y. Fouquet, A. Normand, P. M. Sarradin, and J. Sarrazin (2009), Recent 896

volcanic events and the distribution of hydrothermal venting at the Lucky Strike 897

hydrothermal field, Mid-Atlantic Ridge, Geochem. Geophys. Geosyst., 10, Q02006, 898

doi:10.1029/2008GC002171. 899

Oufi, O., M. Cannat, and H. Horen, Magnetic properties of variably serpentinized abyssal 900

peridotites, J. Geophys. Res., 107 (B5), 10.1029/2001JB000549, 2002. 901

Palmason, G., Geophysical methods in geothermal exploration, in Proceedings 2nd UN Symposium 902

on the Development and use of Geothermal Resources, pp. 1175-1184, U.S. Government 903

Printing Office, Washington, D.C., 1975. 904

Parker, R.L., and S.P. Huestis, The inversion of magnetic anomalies in the presence of topography, 905

J. Geophys. Res., 79, 1587-1594, 1974. 906

41

Pedersen, R.B., I.H. Thorseth, B. Hellevang, A. Schultz, P. Taylor, H.P. Knudsen, and B.O. 907

Steinsbu, Two vent fields discovered at the ultraslow spreading arctic ridge system, EOS 908

Trans. AGU, 86 (52), OS21C-01, 2005. 909

Pedersen, R.B., Discovery of the Loki Castle vent field at the Arctic Mid-Ocean Ridge., ESF Euro-910

MARC Conference, La Colle-sur-Loup, France, 15-17.09.2008, 2008. 911

Ravilly, M., J. Dyment, P. Gente, and R. Thibaud, Axial magnetic anomaly amplitude along the 912

Mid-Atlantic Ridge between 20°N and 40°N, Journal of Geophysical Research, 103 (B10), 913

24201-24221, 1998. 914

Richards, H.G., J.R. Cann, and J. Jensensius, Mineralogical zonation and metasomatism of the 915

alteration pipes of Cyprus sulfide deposits, Econ. Geol., 84, 91-115, 1989. 916

Roman, C., and H. Singh, A self-consistent bathymetric mapping algorithm, Jour. Field Robotics, 2 917

((1-2)), 26-51, 2007. 918

Rona, P.A., Marine Geology: The Trans-Atlantic Geotraverse, 252-256 pp., McGraw-Hill, New 919

York, 1973. 920

Rona, P.A., B.A. McGregor, P.R. Betzer, G.W. Bolger, and D.C. Krause, Anomalous water 921

temperatures over Mid-Atlantic Ridge crest at 26°N latitude, Deep-Sea Research, 22, 611-922

618, 1975. 923

Rona, P.A., Magnetic signatures of hydrothermal alteration and volcanogenic mineral deposits in 924

oceanic crust, J. Volc. Geoth. Res., 3, 219-225, 1978. 925

Rona, P.A., Black smokers and massive sulfides at the TAG hydrothermal Field, Mid-Atlantic 926

Ridge 26°N, EOS Trans. AGU, 66, 936, 1985. 927

Rona, P.A., G. Klinkhammer, T.A. Nelsen, J.H. Trefry, and H. Elderfield, Black smokers, massive 928

sulphides and vent biota at the Mid-Atlantic Ridge, Nature, 321, 33-37, 1986. 929

42

Scheirer, D.S., D.J. Fornari, S.E. Humphris, and S. Lerner, High-resolution seafloor mapping using 930

the DSL-120 sonar system: Quantitative assessment of sidescan and phase-bathymetry data 931

from the Lucky Strike segment of the Mid-Atlantic Ridge, Mar. Geophys. Res., 21, 121-932

142, 2000. 933

Scholten, J., F. Theilen, P. Herzig, M. Schmidt and Shipboard Scientific Party (1999). 934

"Hydrothermal activity along the Tjoernes Fracture Zone, north of Iceland: Initial results of 935

R/V Poseidon cruises 252 and 253." InterRidge News 8(2): 28-31. 936

Schulz, N.J., R.S. Detrick, and S. Miller, Two and three dimensional inversions of magnetic 937

anomalies in the MARK area (Mid-Atlantic Ridge 23°N), Mar. Geophys. Res., 10, 41-57, 938

1988. 939

Scott, R.B., P.A. Rona, B.A. McGregor, and M.R. Scott, The TAG hydrothermal field, Nature, 251, 940

301, 1974. 941

Searle, R., B. Murton, C. Mallows, J. Casey, K. Achenbach, M. Unsworth, and M. Harris, 942

Development of oceanic core complexes on the Mid-Atlantic Ridge at 13-14N: Deep-towed 943

geophysical measurements and detailed seafloor sampling, Eos Trans. AGU, 88 (52), Fall 944

Meeting Suppl. Abstract T51F-01, 2007. 945

Sillitoe, R.H., Environments of formation volcanogenic massive sulfide deposits, Econ. Geol, 68, 946

1321-1336, 1973. 947

Singh S., W. Crawford, H. Carton, T. Seher, V. Combier, M. Cannat, J.P. Canales, D. Dusunur, J. 948

Escartín, J.M. Miranda, Discovery of a magma chamber and faults beneath a Mid-Atlantic 949

Ridge hydrothermal field. Nature, 442, 1029-1032, 2006. 950

Smith, G.M., and S.K. Banerjee, Magnetic properties of plutonic rocks from the central north 951

Atlantic Ocean, Initial Rep. Deep Sea Drill. Proj., 82, 377-383, 1985. 952

43

Smith, D.K., J.R. Cann, and J. Escartín, Widespread active detachment faulting and core complex 953

formation near 13°N on the Mid-Atlantic Ridge, Nature, 442, 440-443, 2006. 954

Smith, D.K., J. Escartin , H. Schouten, and J. Cann, Fault rotation and core complex formation: 955

Significant processes in seafloor formation at slow spreading mid-oceanic ridges (Mid-956

Atlantic Ridge, 13-15°N), Geochem. Geophys. Geosys., 9 (3), Q03003, 2008. 957

Spooner, E.T.C., and W.S. Fyfe, Sub-sea-floor metamorphism, heat and mass transfer, Contrib. 958

Mineral. Petrol., 42, 287-304, 1973. 959

Studt, F.E., Magnetic survey of the Wairakei hydrothermal field, NZ. Jour. Geol. Geophys., 2, 746-960

754, 1959. 961

Sudarikov, S.M., A.M. Ashadze, T.V. Stepanova, A.B. Nesheretov, P.V. Vorobiov, R.V. Kasabov, 962

and A.S. Gutnikov, Hydrothermal activity and ore-forming in the rift zone of the Mid-963

Atlantic Ridge (new data). Dokl. Akad. Nauk. (in Russian), 311 (2), 440-445, 1990. 964

Tao, C., J. Lin, S. Guo, Y.J. Chen, G. Wu, X. Han, C.R. German, D. Yoerger, J. Zhu, N. Zhou, X. 965

Su, E.T. Baker, and S. Party, First Discovery and Investigation of a High-Temperature 966

Hydrothermal Vent Field on the Ultra- Slow Spreading Southwest Indian Ridge, EOS 967

Trans. AGU, 88 (52), Fall Meet. Suppl., Abstract T52B-07, 2007. 968

Thibaud R, Gente P, Maia M. A systematic analysis of the Mid-Atlantic Ridge morphology and 969

gravity between 15 degrees N and 40 degrees N: Constraints of the thermal structure. J 970

Geophys. Res. 103, 24223-24243 OCT 10 1998 971

Thompson, G., S.E. Humphris, B. Schroeder, M. Sulanowska, and P.A. Rona, Active vents and 972

massive sulfides at 26° N (TAG) and 23° N (Snake Pit) on the Mid-Atlantic Ridge, 973

Canadian Mineralogist, 26, 697-711, 1988. 974

44

Tivey, M., P.A. Rona, and H. Schouten, Reduced crustal magnetization beneath the active sulfide 975

mound, TAG hydrothermal field, Mid-Atlantic Ridge, at 26°N, Earth Planet. Sci. Lett., 115, 976

101-115, 1993. 977

Tivey, M.A., High-resolution magnetic survey over the Middle Valley mounds, northern Juan de 978

Fuca Ridge, in Proc. Ocean Drilling Program, Scientific Results, edited by M.J. Mottl, E.E. 979

Davis, A.T. Fisher, and J.F. Slack, pp. 29-34, Ocean Drilling Program, College Station, 980

TX, 1994. 981

Tivey, M.A., P.A. Rona, and M.C. Kleinrock, Reduced crustal magnetization beneath relict 982

hydrothermal mounds of the TAG hydrothermal field, Mid-Atlantic Ridge, 26°N, Geophys. 983

Res. Lett., 23, 3511-3514, 1996. 984

Tivey, M.A., A. Bradley, D. Yoerger, R. Catanach, A. Duester, S. Liberatore, and H. Singh, 985

Autonomous underwater vehicle maps seafloor, EOS, 78, 229-230, 1997. 986

Tivey, M.A., H.P. Johnson, A. Bradley, and D. Yoerger, Thickness of a submarine lava flow 987

determined from near-bottom magnetic field mapping by autonomous underwater vehicle, 988

Geophys.Res. Lett., 25 (6), 805-805, 1998. 989

Tivey, M., and B.E. Tucholke, Magnetization of 0-29 Ma ocean crust on the Mid-Atlantic Ridge, 990

25°30' to 27°10'N, J. Geophys. Res., 103 (B8), 17807-17826, 1998. 991

Tivey, M.A., and H.P. Johnson, Crustal magnetization reveals subsurface structure of Juan de Fuca 992

Ridge hydrothermal vent fields, Geology, 30, 979-982, 2002 993

Tivey, M.A., H. Schouten, and M.C. Kleinrock, A near-bottom magnetic survey of the Mid-994

Atlantic Ridge axis at 26°N: Implications for the tectonic evolution of the TAG segment, J. 995

Geophys. Res., 108, 2277, 10.1029/2002JB001967, 2003. 996

45

Tivey, M.A., W. Bach, M.K. Tivey, J. Seewald, P. Craddock, O. Rouxel, D. Yoerger, C. Yeats, T. 997

McConachy, M. Quigley, D. Vanko and the Scientific Party of MGLN06, Investigating the 998

influence of magmatic volatile input and seawater entrainment on vent deposit morphology 999

and composition in Manus Basin (back-arc) hydrothermal systems, EOS Trans. AGU, 87 1000

(52), Fall Meet. Suppl., Abstract B34A-01, 2006., 2006. 1001

Tivey, M.K., and J.R. Delaney, Sulfide Deposits from the Endeavour Segment of the Juan de Fuca 1002

Ridge, Marine Mining, 5, 165-179, 1985. 1003

Tivey, M.K., and J.R. Delaney, Growth of large sulfide structures on the Endeavour segment of the 1004

Juan de Fuca ridge, Earth Planet. Sci. Lett., 77, 303-317, 1986. 1005

Tucholke, B.E., J. Lin, and M.C. Kleinrock, Megamullions and mullion structure defining oceanic 1006

metamorphic core complexes on the Mid-Atlantic Ridge, J. Geophys. Res., 103, 9857-9866, 1007

1998. 1008

Watson-Munro, C.N., Reconnaissance survey of the variation of magnetic force in the New 1009

Zealand thermal regions, NZ. Jour. Sci. Tech., B20, 99-115, 1938. 1010

Wooldridge, A.L., S.E. Haggerty, P.A. Rona, and C.G.A. Harrison, Magnetic Properties and 1011

Opaque Mineralogy of Rocks from Selected Seafloor Hydrothermal Sites at Oceanic 1012

Ridges, J. Geophys. Res., 95, 12,351-12,374, 1990. 1013

Wooldridge, A.L., C.G.A. Harrison, M.A. Tivey, P.A. Rona, and H. Schouten, Magnetic modeling 1014

near selected areas of hydrothermal activity on the Mid-Atlantic and Gorda Ridges, J. 1015

Geophys. Res., 97, 10,911-10926, 1992. 1016

You, C.-F., and M.J. Bickle, Evolution of an active sea-floor massive sulfide deposit, Nature, 394, 1017

668-672, 1998. 1018

46

Yoerger, D.R., A.M. Bradley, M. Jakuba, C.R. German, T. Shank, and M. Tivey, Autonomous and 1019

remotely operated vehicle technology for hydrothermal vent discovery, exploration, and 1020

sampling, Oceanography, 20 (1), 152-161, 2007. 1021

Zhao, X., B. Housen, P. Solheid, and W. Xu, Magnetic properties of Leg 158 cores: The origin of 1022

remanence and its relation to alteration and mineralization of the active TAG mound, in 1023

Proc. ODP Sci. Results, edited by P. Herzig, S.E. Humphris, D.J. Miller, and R.A. 1024

Zierenberg, pp. 337-351, Ocean Drilling Program, College Station,TX, 1998. 1025

Zierenberg, R.A., Y. Fouquet, D.J. Miller and Shipboard Party, The deep structure of a seafloor 1026

hydrothermal deposit, Nature, 392, 485-488, 1998. 1027

Zonenshain, L.P., M.I. Kuzmin, A.P. Lisitsin, Y.A. Bogdanov, and B.V. Baranov, Tectonics of the 1028

Mid-Atlantic rift valley between the TAG and MARK areas (26-24° N): evidence for 1029

vertical tectonism, Tectonophysics, 159, 1-23, 1989.1030

47

Figure Captions 1031

1032

Figure 1. Location map of known hydrothermal sites on the Mid-Atlantic Ridge between 20°S and 1033

60°N. 1034

1035

Plate 1. Magnetic data over the TAG hydrothermal field on the Mid-Atlantic Ridge (From Tivey et 1036

al., 2003). a) Multibeam bathymetry showing the location of the active TAG mound (star). Bold 1037

box shows the outline of the deeptow sidescan and magnetic survey [Kleinrock and Humphris, 1038

1996]. b) Crustal magnetization computed from sea surface magnetic field data showing a 1039

magnetization low located within the Brunhes normal chron. Star is the TAG active mound c) 1040

High-resolution phase bathymetry map from the deeptow survey of Kleinrock and Humphris, 1041

[1996]. White stars represent from left to right, TAG, Alvin zone and Mir mound. d) Crustal 1042

magnetization computed from near-bottom magnetic data [Tivey et al., 2003] showing the zone of 1043

reduced magnetization over the eastern wall of the spreading segment indicating exposed less 1044

magnetic lower crustal rocks. White stars as in c). 1045

1046

Plate 2. Detailed bathymetry and magnetic data over the TAG active hydrothermal mound. Top 1047

panel shows high resolution multibeam bathymetry of the active TAG mound obtained by ROV 1048

Jason, contour interval is 5 m [Roman et al., 2007]. Bottom panel shows high-resolution near 1049

bottom magnetic data collected by submersible Alvin over the TAG active mound showing reduced 1050

magnetization directly over the mound (cont. int. 1 A/m) [Tivey et al., 1993]. The north-south 1051

extension of the reduced zone suggests a possible fault trend influence in the upflow zone. Bold 1052

line is the 3645 m isobath delineating the mound extent. Figure modified from Tivey et al., [1993]. 1053

48

1054

Plate 3. Bathymetry and magnetic data over the Lucky Strike hydrothermal area. a) ship 1055

multibeam bathymetry of the Lucky Strike segment showing the axial volcano within the rift valley 1056

that hosts the hydrothermal activity. b) Crustal magnetization computed from the sea surface 1057

magnetic data (see text for details) showing the Brunhes anomaly and the weaker crustal 1058

magnetization over the summit of Lucky Strike volcano. Bold line is the 1900 m isobath 1059

delineating the volcano. c) High-resolution phase bathymetry of the Lucky Strike volcano 1060

collected by a deeptow DSL120 survey [Scheirer et al., 2000]. Red stars show the location of 1061

hydrothermal activity on the summit of the volcano. d) Crustal magnetization computed from the 1062

DSL120 near bottom magnetic data (see text for details) showing magnetization lows associated 1063

with the major fault systems (shown in heavy bold dash) imaged by multichannel seismic data 1064

[Singh et al., 2006]. Red star indicates the summit hydrothermal vent area. These major fault 1065

systems may be hosting recharge for the hydrothermal system [Singh et al., 2006] resulting in 1066

enhanced low temperature alteration on a regional crustal scale. 1067

1068

Plate 4. Magnetic and bathymetry data of the Menez Gwen ridge segment: a) regional multibeam 1069

bathymetry data from [Lourenço et al. 1998] showing the presence of the axial split volcano within 1070

the rift valley reaching 800 m depth. Bold black line marks the spreading axis. b) a compilation of 1071

sea surface magnetic field data over the Menez Gwen segment (data from National Geophysical 1072

Data Center). c) Multibeam bathymetry [Cannat et al., 1999; Escartin et al., 2001]. Red star shows 1073

the location of the hydrothermal vent site within the central rift of the split axial volcano. d) 1074

Crustal magnetization computed from the sea surface magnetic data and showing the overall 1075

49

reduced magnetization over the axial volcano. Bold line is the 1000 meter isobath outlining the 1076

axial volcano. Red star is the location of the active vent site. 1077

1078

Plate 5. Magnetic and bathymetry data from the 4°48’S region on the southern Mid-Atlantic Ridge: 1079

a) Regional ship-based multibeam bathymetry courtesy of German et al., [2008]. b) High 1080

resolution near bottom multibeam bathymetry data from 2005 autonomous vehicle ABE missions 1081

151/153. c) ABE multibeam data showing the location and names of active vent sites with red 1082

stars. d) ABE near-bottom magnetic data inverted for relative crustal magnetization showing zones 1083

of reduced magnetization associated with the active vent sites show by red stars. 1084

1085

Plate 6. Sea surface magnetic and bathymetry data over the Rainbow hydrothermal area of the 1086

Mid-Atlantic Ridge at 36°14’N. a) Multibeam bathymetry from [Thibaud et al., 1998] showing the 1087

location of Rainbow ridge at the non-transform offset between the AMAR segment to the north and 1088

the South AMAR segment to the south. b) Sea surface magnetic data compilation of the Rainbow 1089

area (data from National Geophysical Data Center and cruise Iris, chief scientist Y. Fouquet) with 1090

ship tracks shown by fine black lines. c) Regional crustal magnetization computed from the sea 1091

surface magnetic field data showing the central Brunhes anomaly and the weak magnetic field over 1092

the Rainbow area. 1093

1094

Plate 7. Detailed map of the Rainbow hydrothermal area: a) Bathymetry of the Rainbow field 1095

with the extent of mineralization (Y. Fouquet, pers. comm., 2004) shown by the red outline. 1096

Individual vents are shown by the triangles and circles. Near bottom magnetic survey tracks shown 1097

50

dotted. b) Crustal inversion of the near bottom magnetic data showing that a strong positive 1098

anomaly is located over the main active vent field to the west end of the field. 1099

1100

51

Table 1 Hydrothermal vent sites along the Mid-Atlantic Ridge 1101

Vent Field Location Active

(A)

Inact.

(I)

Type 1 Terrain

2 Depth

(m)

Areal Size 3

sq. km

(m)

Reference

Loki’s Castle ~73.5°N, 4°W A High-T

300°C

Volcanic 2352 0.01

(100 x 100)

Pederson, 2008

Gallionella

Garden

71.3°N, 5.78°W A High-T

270°C

Volcanic

550 (500 x ?) Pederson et al., 2005

Soria Moria 71.26°N, 5.81°W A High-T

unreported

Volcanic 700 (100 x 100) Pederson et al., 2005

Kolbeinsey 67.083°N,

18.72°W

A Low-T

131°C

Volcanic 100-110 unreported Scholten et al., 1999

Hannington et al., 2001

Grimsey 66.60°N 17.67°W A High-T

248-

Volcanic 400 0.03

(300 x 100)

Scholten et al., 1999

Hannington et al., 2001

52

251°C

Reykjanes

(Steinaholl)

63.10°N, 24.55°W A Low-T

unknown

Volcanic 220-350 unknown German et al., 1994

Menez Gwen 37.84°N, 31.53°W A High-T

265 285ºC

Volcanic 840-870 0.04

(200 x 200)

Ondreas et al., 1997

Charlou et al., 2000

Lucky Strike 37.29°N, 32.27°W A High-T

170-

324°C

Volcanic 1700 0.4

(1000 x

400)

Langmuir et al., 1997

Charlou et al., 2000

Saldanha 36.57°N, 33.42°W A Low-T

CH4

7-9°C

Ultramafic 2200-2220 0.0004

(400 sq m)

Barriga et al., 1998

Dias and Barriga, 2006

Rainbow 36.23°N, 33.23°W A High-T

360°C

Ultramafic 2270-2320 0.025

(250 x 100)

German et al., 1996

Lost City 30.12°N, 42.12°W A Low-T Ultramafic 700-800 0.02

(200 x 100)

Kelley et al., 2001

Broken Spur 29.17°N, 43.17°W A High-T

356-

Volcanic 3100 0.0078

(7850 sq. m)

Murton et al., 1994

German et al, 1999

53

364°C

TAG 26.14°N, 44.83°W A High-T

363°C

Volcanic 3620-3700 25

(5000 x

5000)

Rona et al., 1993

MAR 24°30’N 24.5°N, 46.15°W I n/a Volcanic 3900 Unknown Sudarikov et al., 1990

Snake Pit 23.37°N, 44.95°W A High-T

335-

350°C

Volcanic 3440 0.045

(150 x 300)

Thompson et al., 1988

Fouquet et al., 1993

Puy des Folles 20.50°N, 45.64°W I n/a Volcanic 1940-2000 0.858 Gente etal., 1996

Zenith Victory 20.13°N, 45.65°W I n/a Volcanic 2370-2390 0.495 G. Cherkashov, pers.

comm. 2009

Krasnov 16.63°N, 46.47°N I n/a Volcanic 3700-3750 0.15

(500 x 300)

Beltenev et al., 2004

MAR 15°20’N 15.08°N, 44.8°W A Low-T Ultramafic 2452-4668 Unknown Charlou et al., 1998

Logatchev-1 14.75°N, 44.98°W A High-T

359°C

Ultramafic 2900-3050 0.032

(400 x 80)

Gebruk et al., 1997

54

Logatchev-2 14.72°N, 44.94°W A High-T

320°C

Ultramafic 2650 0.007 Cherkashev et al., 2000

Logatchev-5 14.75°N, 44.97°W I n/a Ultramafic 2900 unreported Fouquet et al., 2008

Semyonov 13.52°N, 44.95°W I n/a Volcanic 2400-2600 0.361 G. Cherkashov, pers.

comm. 2009

MAR 13°30’N 13.50°N, 44.83°W I n/a unreported unreported unreported Searle et al., 2007

MAR 13°20’N 13.33°N, 44.83°W I n/a unreported unreported unreported Murton et al., 2007

Ashadze 1 12.97°N, 44.86°W A High-T

370°C

Ultramafic 4080

(150 x ?)

Beltenev et al., 2003

Fouquet et al., 2007

Ashadze-2,3 12.97°N, 44.86°W A High-T Ultramafic 3300 0.106 Fouquet et al., 2007

Ashadze-4 ~12.97°N,

~44.86°W

I n/a Volcanic 4530 unreported Fouquet et al., 2007

Comfortless

Cove

4.80°S, 12.37°W A High-T

400-

464°C

Volcanic 2996 0.0001-

0.0002

(10 x 10)

(10 x 20)

German et al., 2005

Haase et al., 2005

Koschinsky et al., 2008

Turtle Pits 4.81°S, 12.37°W A High-T Volcanic 2990 0.0004 German et al., 2005

55

1102

1103

1 High (T > 150°C) vs. low (T < 50°C) exit-fluid temperature. 1104

2 terrain refers to the primary substrate hosting the vent site Volcanic 1105

3 Reported areal extent of hydrothermal deposits in sq. kilometers and equivalent area in meters in parentheses. 1106

407°C (20 x20)

0.0001

(10 x 10)

Haase, et al., 2005

Koschinsky et al., 2008

Red Lion 4.80°S, 12.38°W A High-T

193-

349°C

Volcanic 3050 0.000625

(25 x 25)

German et al., 2005

Haase et al., 2005

Koschinsky et al., 2008

Wideawake 4.81°S, 12.37°W A Diffuse Volcanic <2990 unreported German et al., 2005

Haase et al., 2005

Nibelungen

(Drachenschlund)

8.30°S, 13.51°W A High-T

153°C

Volcanic? 2915 0.001

~100 x 10

Koschinsky et al., 2006

Melchert et al., 2008

Lilliput 9.55°S, 13.21°W A Low-T

Diffuse

Volcanic 1500 unreported Koschinsky et al., 2006

56

1107

1108

Figure 1.

Plate 1.

Plate 2.

Plate 3.

Plate 4.

Plate 5.

Plate 6.

Plate 7.