The half space model of lithospheric cooling with age predicts that the heat flux through the ocean floor on ridge flanks will vary in proportion to the inverse square root of its age, but across older ocean floor measured heat flow values vary more slowly than this, again favoring a plate model. The GDH1 model of Stein & Stein (1992) predicts the following values for heat flow, q (mWm-2) as a function of age, t (Ma):


50 100


Fig. 6.9 Observed depth and heat flow data for oceanic ridges plotted as a function of lithospheric age, and compared to the predictions of three thermal models: HS, half space model; PSM, model of Parsons, Sclater and McKenzie; GDHI, global depth and heat flow model of Stein and Stein (redrawn from Stein & Stein, 1996, by permission of American Geophysical Union. Copyright © 1996 American Geophysical Union).

q = 510t_1/2 for t < 55 Ma and q = 48 + 96exp(-t/36) for t > 55 Ma.

The variation of heat flow with age predicted by all three thermal models is illustrated in Fig. 6.9b and compared to observed heat flow values. It will be noted that observed values for younger lithosphere have not been plotted. This is because there are large variations in the heat flux measured in young oceanic crust (Fig. 6.4). The values obtained are typically less than those predicted by the models and there is now thought to be good reason for this. In particular, there is a large scatter in heat flow magnitude near the crests of ocean ridges. Thermal lows tend to occur in flat-floored valleys and highs within areas of rugged topography (Lister, 1980).

Blanketing by sediment does not appear to be the cause of the low heat flow because the troughs are within the least sedimented areas of the ridge and also the youngest and therefore hottest. To explain these phenomena it was proposed that the pattern of heat flow is controlled by the circulation of seawater through the rocks of the oceanic crust.

Although the penetration of water through the hard rock of the sea floor at first seems unlikely, it has been shown that thermal contraction can induce sufficient permeability for efficient convective flow to exist. The cracks are predicted to advance rapidly and cool a large volume of rock in a relatively short time, so that intense localized sources of heat are produced at the surface. Active geothermal systems that are driven by water coming into contact with near-molten material are expected to be short-lived, but the relatively gentle circulation of cool water, driven by heat conducted from below, should persist for some time. However, as the oceanic crust moves away from the ridge crest, and subsides, it is blanketed by impermeable sediments, and the pores and cracks within it become clogged with minerals deposited from the circulating water. Ultimately heat flux through it is by conduction alone and hence normal heat flow measurements are obtained. This "sealing age" of oceanic crust would appear to be approximately 60 Ma.

Detailed heat flow surveys on the Galapagos Rift revealed that the pattern of large-scale zoning and the wide range of individual values are consistent with hydrothermal circulation (Williams et al., 1974). Small-scale variations are believed to arise from variations in the near-surface permeability, while larger-scale variations are due to major convection patterns which exist in a permeable layer several kilometers thick which is influenced by topography, local venting, and recharge at basement outcrops. The penetration of this convection is not known, but it is possible that it is crust-wide. It is thought that hydrothermal circulation of seawater in the crust beneath ocean ridges transports about 25% of the global heat loss, and is clearly a major factor in the Earth's thermal budget (Section 2.13).

The prediction of hydrothermal circulation on mid-ocean ridges, to explain the heat flow values observed, was dramatically confirmed by detailed investigations at and near the sea floor at ridge crests, most notably by submersibles. Numerous hydrothermal vent fields have been discovered on both the East Pacific Rise and the Mid-Atlantic Ridge, many of them revealed by the associated exotic and previously unknown forms of life that survive without oxygen or light. The physical and chemical properties of the venting fluids and the remarkable microbial and macrofaunal communities associated with these vents, have been reviewed by Kelly et al. (2002). The temperature of the venting fluids can, exceptionally, be as high as 400°C. The chemistry of the hydrothermal springs on the East Pacific Rise and Mid-Atlantic Ridge is remarkably similar, in spite of the great difference in spreading rates, and suggests that they have equilibrated with a greenschist assemblage of minerals (Campbell et al., 1988). Surprisingly perhaps, because of the cooler environment at the ridge crest, there are high levels of hydrothermal activity at certain locations on the very slow- and ultraslow-spreading Gakkel Ridge. This appears to result from the focusing of magmatic activity at these points, producing higher temperatures at shallow depths (Michael et al., 2003).

Further evidence that hydrothermal circulation occurs comes from the presence of metalliferous deposits at ridge crests. The metals are those known to be hydrothermally mobile, and must have been leached from the oceanic crust by the ingress of seawater which permitted their extraction in a hot, acidic, sulfide-rich solution (Rona, 1984). On coming into contact with cold seawater on or just below the sea floor the solutions precipitate base metal sulfide deposits. The presence of such deposits is corroborated by studies of ophiolites (Section 13.2.2).


Models for the formation of oceanic lithosphere normally require a magma chamber beneath the ridge axis from which magma erupts and intrudes to form the lava flows and dikes of layer 2. Solidification of magma within the chamber is thought to lead to the formation of most of oceanic layer 3 (Section 6.10). Evidence for the presence of such a magma chamber has been sought from detailed seismic surveys at ridge crests employing refraction, reflection, and tomographic techniques.

On the fast-spreading East Pacific Rise many of the surveys have been carried out in the area north of the Siquieros Fracture Zone between 8° and 13°N. The area centered on the ridge crest at 9°30'N has been particularly intensively studied (e.g. Herron et al., 1980; Detrick et al., 1987; Vera et al., 1990). More recently additional experiments have been carried out at 14°15'S, on one of the fastest spreading sections of the ridge (Detrick et al., 1993a; Kent et al., 1994). All of these studies have revealed a region of low seismic velocities in the lower crust, 4-8 km wide, and evidence for the top of a magma chamber at varying depths, but typically 1-2 km below the sea floor. There is some indication that the depth to the magma chamber is systematically less at 14°S compared to 9°N on the East



Fig. 6.10 The variation of P wave velocity in the oceanic crust, at the crest of the East Pacific Rise at 9°30 N, deduced from expanded spread (ESP) and common depth point seismic profiling. Shaded area indicates a region with a high percentage of melt. An interpretation of the velocities in terms of rock units, and an indication of the extent of the zone of anomalously low seismic velocities (LVZ), are also shown (redrawn from Vera et al., 1990, by permission of the American Geophysical Union. Copyright © 1990 American Geophysical Union).

10 E

Fig. 6.10 The variation of P wave velocity in the oceanic crust, at the crest of the East Pacific Rise at 9°30 N, deduced from expanded spread (ESP) and common depth point seismic profiling. Shaded area indicates a region with a high percentage of melt. An interpretation of the velocities in terms of rock units, and an indication of the extent of the zone of anomalously low seismic velocities (LVZ), are also shown (redrawn from Vera et al., 1990, by permission of the American Geophysical Union. Copyright © 1990 American Geophysical Union).

Pacific Rise, suggesting an inverse correlation between magma chamber depth and spreading rate (Detrick et al., 1993b). The interpretation of Vera et al. (1990) of results obtained at 9°30'N, using multi-channel, expanded spread reflection profiling, is shown in Fig. 6.10. They considered that only the volume in which the P-wave velocity is less than 3 km s-1 can be regarded as a melt lens, and that the region in which the P-wave velocity is greater than 5 km s-1, which includes much of the low velocity zone, behaves as a solid. Detrick et al. (1987) demonstrated that a strong reflector, thought to be associated with the top of the magma chamber, can be traced as a nearly continuous feature for tens of kilometers along the ridge axis. Much of the more recent work, typically employing tomographic techniques (Section 2.1.8), suggests that the region in which there is a high melt fraction, probably no more than 30% crystals so that the shear wave velocity is zero, is remarkably small, perhaps no more than a few tens of meters thick, and less than 1 km wide (Kent et al., 1990, 1994; Caress et al., 1992; Detrick et al., 1993a; Collier & Singh, 1997). Thus most of the low velocity zone beneath the ridge crest behaves as a solid and is interpreted as a region of anomalously hot rock.

In contrast to the picture that has emerged for the East Pacific Rise, most seismic studies of the slowly spreading Mid-Atlantic Ridge recognize a low velocity zone in the lower crust beneath the ridge crest but have not yielded any convincing evidence for a magma chamber or melt lens (Whitmarsh, 1975; Fowler, 1976; Purdy & Detrick, 1986; Detrick et al., 1990). However, Calvert (1995), in reanalyzing the data of Detrick et al. (1990) acquired at 23°17'N, isolated reflections from a presumed magma chamber at a depth of 1.2 km and with a width of 4 km.

It seems unlikely therefore that steady state magma chambers exist beneath the axes of slowly spreading ridges. Transient magma chambers, however, related to influxes of magma from the mantle, may exist for short periods. In order to test this hypothesis a very detailed combined seismic and electromagnetic experiment was carried out across the Reykjanes Ridge south of Iceland (Sinha et al., 1998). This study was deliberately centered on a magmatically active axial volcanic ridge (AVR) on the Reykjanes Ridge at 57°45'N, and did reveal a melt lens and crystal mush zone analogous to those imaged on the East Pacific Rise. In this instance the melt lens occurs at a depth 2.5 km beneath the sea floor. The results of this study provide strong support for the hypothesis that the process of crustal accretion on slow-spreading ridges is analogous to that at fast-spreading ridges but that the magma chambers involved are shortlived rather than steady state. Despite its proximity to the Iceland hot spot, the ridge crest south of 58°N on the Reykjanes Ridge has the characteristics of a typical slow-spreading ridge: a median valley, and normal crustal thickness and depth.

The logistically complicated seismic experiments required to test for the presence or absence of a melt lens have yet to be carried out on the very slow- and ultraslow-spreading Gakkel Ridge. It seems extremely unlikely that melt lenses exist beneath the amagmatic segments of this ridge, in that these consist of mantle peridotite with only a thin carapace of basalts, but possible that transient melt lenses occur beneath the mag-matic segments and volcanic centers (Section 6.9). However, in 1999 seismological and ship-borne sonar observations recorded a long-lived magmatic-spreading event on the Gakkel Ridge that had characteristics more consistent with the magma being derived directly from mantle depths than from a crustal magma chamber (Tolstoy et al., 2001).

Sinton & Detrick (1992), taking account of the seismic data available at that time and incorporating new ideas on magma chamber processes, proposed a model in which the magma chambers comprise narrow, hot, crystal-melt mush zones. In this model magma chambers are viewed as composite structures compris ing an outer transition zone made up of a hot, mostly solidified crust with small amounts of interstitial melts and an inner zone of crystal mush with sufficient melt for it to behave as a very viscous fluid. A melt lens only develops in fast-spreading ridges where there is a sufficiently high rate of magma supply for it to persist at the top of the mush zone (Fig. 6.11a). This lens may extend for tens of kilometers along the ridge crest, but is only 1-2 km wide and tens or hundreds of meters in thickness. Slow-spreading ridges are assumed to have an insufficient rate of magma supply for a melt lens to develop (Fig. 6.11b) and that eruptions only occur when there are periodic influxes of magma from the mantle. Such a model is consistent with the seismic data from ocean ridges and petrologic observations which require magma to have been modified by fractionation within the crust, which could not occur in a large, well-mixed chamber. It also explains why less fractionation occurs in the volcanic rocks of slow-spreading ridges. A problem with this model, however, is that it is not apparent how the layered gabbros of layer 3 might develop.

Subsequent work by Singh et al. (1998), involving further processing of the seismic reflection data obtained by Detrick et al. (1993a) near to 14°S on the East Pacific Rise, was specifically targeted at identifying any along-axis variations in the seismic properties and thickness of the melt lens. Their results suggest that only short, 2-4 km lengths of the melt lens contain pure melt capable of erupting to form the upper crust. The intervening sections of the melt lens, 15-20 km in length, are rich in crystal mush and are assumed to contribute to the formation of the lower crust. It seems probable that the pockets of pure melt are related to the most recent injections of magma from the mantle.

Was this article helpful?

0 0
Boating Secrets Uncovered

Boating Secrets Uncovered

If you're wanting to learn about boating. Then this may be the most important letter you'll ever read! You Are Going To Get An In-Depth Look At One Of The Most Remarkable Boating Guides There Is Available On The Market Today. It doesn't matter if you are just for the first time looking into going boating, this boating guide will get you on the right track to a fun filled experience.

Get My Free Ebook

Post a comment