VOLCANIC EVENTS AT MID-OCEAN RIDGES

Motivating questions:

How does volcanism at and near mid-ocean ridges vary as a function of spreading rate?
What do these variations tell us about the overall characteristics of fast vs. slow spreading ridges?
What do the volcanic features in the Oman ophiolite sequence say about the evolution of the ridge?

Volcanic events at active MORs:

fundamental differences between fast and slow MORs
on-axis volcanism
of-axis volcanism and seamounts
lava flow morphology
magmatism vs. tectonism
summary and expectations for Oman observations

Volcanic stratigraphy in the Oman ophiolite:

flow stratigraphy
model of evolution

References

VOLCANIC EVENTS AT ACTIVE MID-OCEAN RIDGES

I. Differences between fast and slow MORs

	fast	intermediate	slow
full spreading rate, mm/yr	80-160	40-80	10-40
ridge examples	southern EPR	JdFR, Gorda, Galapagos, SE Indian	MAR
morphology	axial highs	...	axial valleys
relief across ridge	low	...	high
melt lens?	steady-state, continuous	...	none detected
vent morphology	fissures more common	...	point sources more common
flow morphology	sheet flows dominate	...	pillow lavas dominate
eruption rate	higher	...	lower

II. Observations and/or detection of volcanic events at MORs

hydrothermal plumes (chemical and T anomalies) above vents
repeated surveying (e.g., SeaBeam bathymetry, sidescan sonar)
MORB geochemistry (e.g., may be consistent with dike propagation and/or evolution of a magma body)
bottom pressure recorders on calderas of axial volcanoes (e.g., Axial Volcano, Juan de Fuca) to measure inflation/deflation of summit
US Navy Sound Surveillence System hydrophones (SOSUS, operational since 1993)--pick up acoustic noise (T-waves) generated by submarine earthquakes of magnitude <3 and allow for real-time observations

Several historical eruptions have been documented. For example:
- Juan de Fuca ridge: Cleft (1980's), CoAxial (1993), Blanco FZ (1994)
- northern East Pacific Rise
- Gorda Ridge (1996)

III. On-axis volcanism

The neovolcanic zone is the region along a ridge where magmatism is focussed, and its width and morphology appears to vary with the ridge spreading rate (see Figure 1):

At fast ridges, the neovolcanic zone is <250 m wide, with a continuous linear axial summit trough (AST) on the ridge crest up to 40 m deep and 250 m wide. The AST indicates focussed magmatism and a greater frequency of events; its continuity indicates along-axis magma supply. (MAGMATISM dominates.)

Slow ridges have a large, deep axial valley up to 20 km wide with a central neovolcanic zone up to 12 km wide. There is often a discontinuous axial valley ridge (AVR) up to 5 km wide in the valley as well as evidence for eruptions along axis-parallel faults and along ridge-perpendicular zones. The axial valleys are extensively faulted, and large earthquakes are common. (TECTONISM dominates.)
Lava flow morphology also varies with ridge spreading rate (see Figure 2): Lavas at the neovolcanic zone of fast ridges are dominantly fissure-fed sheet flows, ranging from flat, thin flows to jumbled sheets up to 1 m thick. At slow ridges, pillow lavas dominate. These lavas form hummocks up to 500m in diameter, or hummocky ridges up to 2 km long.
Seamounts are small (<<1 km diameter), point-source piles of lava. These are not found within the neovolcanic zones of fast ridges, but are common features within in the neovolcanic zones of slow ridges, as part of the AVR.

IV. Off-axis volcanism

Eruptions can occur outside the neovolcanic zone. Off-axis flows may be younger than predicted by their distance from the ridge axis and the known ridge spreading rate. Thin sediment cover, a fresh glassy appearance, and young U-series disequilibrium ages also suggest recent eruption. Off-axis flows (up to 4 km off-axis) are seen along the EPR (see Figure 3) and may also be occurring along the JdFR. Off-axis flows are syntectonic with faulting (i.e., are often associated with ridge-parallel fissures), and petrologically appear to be fed by spatially distinct and restricted sources (either a different plumbing system or the edge of an axial magma body).
Off-axis seamounts are common along slow-spreading ridges either as single cones or chains. Seamounts are not found at the ridge crest of fast-spreading ridges, but do occur within 15 km of the axis.
Off-axis seamounts at slow-spreading ridges are generally chemically depleted and more mafic than axial valley lavas, so they probably do not originate in crustal magma chambers.
Seamounts on fast-spreading ridges are always off-axis, and often their craters are offset from the center towards the ridge axis.

V. Lava flow morphology

Experiments with hot wax extruded into cold water quantitatively show flow morphology as a function of extrusion rate, which affects the cooling rate, and slope. Lava morphology is quantified by a dimensionless parameter y, the ratio of the time required for a solid crust to form on the flow surface to a characteristic time scale for horizontal advection. Pillow lavas form at high cooling rates and low extrusion rate or low y; lavas become more sheet-like as the cooling rate decreases relative to a higher extrusion rate, at high y (see Figure 4). To summarize, the following is taken from Gregg and Fink (1995):

flow morphology	psi	cooling rate	slope	extrusion rate
pillows	...	high	low	low
lobate sheets	...	...	...	...
lineated sheets	3	...	...	...
ropy sheets	10	...	...	...
jumbled sheets	30	low	high	high

Gregg et al. (1996) used the slope, morphology and volumes of submarine flows along with such experimental results to estimate eruption durations:
- 9°50'N, EPR, 1991 (3-5x106 m3, ponded and overtopped the ASC, see Figure 6b): <2hrs
- CoAxial segment, JdFR, 1993 (~5x106 m3, mostly pillows): ~10 days
The correlation of the ratio of sheet to pillowed flows with ridge spreading rate is reasonable given the presence of a continuous magma supply under fast ridges and the lack of one under slow ridges. Across an intermediate-spreading-rate ridge, flow morphology, like ridge morphology, can be highly variable (see Figure 5).
Other features of submarine lava flows include lava pillars (Figure 6) and tumuli or inflated flows (Figure 7).

VI. Magmatism vs. tectonism

At fast ridges, magmatism dominates. There is a large magma supply beneath the ridge, so dike intrusions are more frequent. This keeps tensional stresses perpendicular to the axis low, so that to erupt lavas onto the surface, dikes must have relatively high internal pressure. This results in higher effusion rates and more sheet-like flows.
At slow ridges, tectonism dominates. There is little magma under the ridge, so dikes intrude at lower pressures and intrusions are infrequent. This means the tensionals stress perpendicular to the ridge increases until faulting (or intrusion) occurs.
Fast ridges tend to have narrow grabens (corresponding to the AST) along their axis, and slow ridges have very wide axial valleys bounded by faults. Mechanical modeling has shown how the width of grabens above intrusive dikes corresponds to the ridge spreading rate (see Figure 8): as a dike intrudes to shallow levels in the crust the distance between the zones of maximum tensional strain decreases. The magnitude of the dike-induced strain also increases as the dike moves upward. A graben forms on the surface when the dike-induced + ambient stress reaches a critical point. (The ambient stress is a function of the time since the last rifting event.)
It is speculated that ridges may evolve from a more magmatically robust phase to a more tectonically-controlled one (see Figure 9). The difference between fast and slow ridges would be the timescale over which this evolution (or cyclicity) takes place.
MORB chemistry is dictated by the balance between magmatism and tectonism.

VII. Summary and expectations for Oman observations

Volcanic events vary in size and temporal and spatial frequency along mid-ocean ridges corresponding to spreading rate (see Figure 10). In general, fast ridges are characterized by frequent small events, and slow ridges by infrequent large events. The morphology of ridge axes is also related to spreading rate (see Figures 1 and 11).
The Oman ophioliate is thought to represent a fast-spreading-rate ridge axis, so we should expect the volcanic sequence to show evidence for frequent events which mostly produced sheet-like flows.

VOLCANIC STRATIGRAPHY IN THE OMAN OPHIOLITE

I. Flow stratigraphy

The following descriptions are taken from Alabaster et al. (1982); the distribution and stratigraphy of the units is shown in Figure 12:

The Geotimes Unit (which was shown on the cover of Geotimes in 1975) is a unit 1-1.5 km thick, directly atop the sheeted dike complex. Because of its stratigraphic location and the similarity of its trace element compositions to the sheeted dikes, the unit is thought to represent "spreading axis volcanism." The unit is composed mostly of pillows interbedded with some massive columnar jointed flows. Pillow are large (up to 2 m diameter) and elongate.
The Lasail Unit is named for the Lasail mine, which is a massive sulfide deposit found between the Geotimes and Lasail lavas. In the field, this unit is 0-750 m thick; where it does not outcrop there is a layer up to 2 m thick of sediments. The unit's occurrence in patches indicates that it probably represents seamounts that were built on top of the Geotimes lavas. The Lasail unit is basaltic but is also composedof swarms of thin intrusive sheets of intermediate composition (basalts to andesites). The orientations of these sheets suggest they emanate from late gabbro-diorite-tonalite-trondhjemite complexes. The unit is composed of pillows (which are smaller than the Geotimes pillows) with microphenocrysts of olivine and clinopyroxene. The interseamounts areas have interbedded pillows and sheet flows that are thin and "transitional."
The Alley Unit has its type locality in a major graben structure named the Bowling Alley. This unit is thin, at most 750 m thick. The lavas are mostly mafic, but the unit also has a series of rhyolites and obsidian flows which are located along major NW-SE faults. The Alley unit is comprised of large vesicular pillows interbedded with massive flows.
The Cpx-phyric Unit is volumetrically the smallest volcanic unit in the Oman complex. These lavas are basaltic, with distinct phenocrysts of clinopyroxene. These lavas are thin pillowed flows found along major faults.
The Salahi Unit is restricted to the Wadi Salahi area, and is fed by 20 m thick dikes crosscutting the Alley unit. These dikes are oriented NW-SE. Unlike the other units, The Salahi has a chemical signature of intraplate basalts. This unit is separated from the Alley lavas by a layer of sediments, and is composed of a layer up to 20 m thick of small pillows overlain by 60 m of massive columnar flows.

II. Model of evolution
Figure 13 shows the interpretation of Alabaster et al. (1982) for the evolution of the Oman lavas described above, based on geochemistry and petrogenetic modeling. Additionally:

Petrologenetic modeling shows that the Geotimes and Lasail units can be linked by partial melting and fractional crystallization.
The Alley and Cpx-phyric units can also be linked to the same source, but this source mantle is more depleted than the Geotimes-Lasail source.
The Salahi unit require about 15% partial melting of its source, while the other units need about 30% partial melting. Global correlations of geochemical data suggest that the extent of mantle melting beneath a ridge increases with increasing spreading rate. This suggests the Salahi was erupted at a slower spreading rate than the other units, such as when the ridge was being obducted.

This topic will be presented in detail by Renee and Mark.

REFERENCES

Observations at active ridges

Appelgate, B and Embley, RW, 1992, Submarine tumuli and inflated tube-fed lava flows on Axial Volcano, Juan de Fuca Ridge, Bulletin of Volcanology, 54:447-458.
Ballard, RD, Holcomb, RT and van Andel, TH, 1979, The Galapagos rift at 86šW: 3. Sheet flows, collapse pits, and lava lakes of the rift valley, Journal of Geophysical Research, 84(B10):5407-5422.
Bonatti, E and Harrison, CGA, 1988, Eruption styles of basalt in oceanic spreading ridges and seamounts: effect of magma temperature and viscosity, Journal of Geophysical Research, 93(B4):2967-2980.
Chadwick, WW and Embley, RW, Graben formation associated with recent dike intrusions and volcanic eruptions on the mid-ocean ridge, Journal of Geophysical Research, in press.
Chadwick, WW and Embley, RW, 1994, Lava flows from a mid-1980s submarine eruption on the Cleft segment, Juan de Fuca Ridge, Journal of Geophysical Research, 99(B3):4761-4776.
Embley, RW and Chadwick, WW, 1994, Volcanic and hydrothermal processes associated with a recent phase of seafloor spreading at the northern Cleft segment: Juan de Fuca Ridge, Journal of Geophysical Research, 99(B3):4741-4760.
Embley, RW, Chadwick, W, Perfit, MR and Baker, ET, 1991, Geology of the northern Cleft segment, Juan de Fuca Ridge: Recent lava flows, sea-floor spreading, and the formation of megaplumes, Geology, 19:771-775.
Embley, RW, Chadwick, WW, Jonasson, IR, Butterfield, DA and Baker, ET, 1995, Initial results of the rapid response to the 1993 CoAxial event: Relationships between hydrothermal and volcanic processes, Geophysical Research Letters, 22(2):143-146.
Perfit, MR and Chadwick, WW, Magmatism at mid-ocean ridges: constraints from volcanological and geochemical investigations, submitted.
Perfit, MR, Fornari, DJ, Smih, MC, Bender, JF, Langmuir, CH and Haymon, RM, 1994, Small-scale spatial and temporal variations in mid-ocean ridge crest magmatic processes, Geology, 22:375-379.

Lava flow modeling

Fink, JH and Griffiths, RW, 1992, A laboratory analog study of the surface morphology of lava flows extruded from point and line sources, Journal of Volcanology and Geothermal Research, 54:19-32.
Gregg, TKP and Fink, JH, 1995, Quantification of submarine lava-flow morphology through analog experiments, Geology, 23(1):73-76.
Gregg, TKP, Fornari, DJ, Perfit, MR, Haymon, RM and Fink, JH, 1996, Rapid emplacement of a mid-ocean ridge lava flow on the East Pacific Rise at 9š46-51¹N, Earth and Planetary Sciences Letters, 144:E1-E7.
Griffiths, RW and Fink, JH, 1992, Solidification and morphology of submarine lavas: A dependence on extrusion rate, Journal of Geophysical Research, 97(B13):19729-19737.

Observations in Oman

Alabaster, T and Pearce, JA, 1985, The interrelationship between magmatic and ore-forming hydrothermal processes in the Oman ophiolite, Economic Geology, 80(1):1-16.
Alabaster, T, Pearce, JA and Malpas, J, 1982, The volcanic stratigraphy and petrogenesis of the Oman ophiolite complex, Contributions to Mineralogy and Petrology, 81:168-183.

FIGURE CAPTIONS

(These will be scanned in...eventually.)

Figure 1. a) Cross-sections of fast, intermediate, and slow ridges showing the difference in width and character of the neovolcanic zone. From XXXX b) Rift valley depths also appear to vary with ridge spreading rate. From XXXX
Figure 2. a) Flow morphology (as sheets or pillows) vs. ridge spreading rate. From Bonatti and Harrison (1988). b) Schematic of expected along-axis variations in lava flow morphology. This diagram assumes that the magma lens beneath the ridge tapers off near transforms or overlapping spreading centers, as geophysical evidence suggests. From Bonatti and Harrison (1988).
Figure 3. An interpretation of the distribution of on- and off-axis lavas along the northern EPR, resulting from geochemical data from a suite of lavas across the ridge axis. From Perfit et al. (1994).
Figure 4. Flow morphology. Left column shows plan views of experiments in which hot wax was extruded into cold water, the middle figures are photos of submarine flows, and the figures are sketches. From Gregg and Fink (1995).
Figure 5. Lava flow morphologies across the Galapagos Rift at 86ûW. a) Geologic map. Contours are in fathoms (1 fathom = 1.8 m). b) Profiles of the Galapagos Rift showing flow forms. Numbers refer to locations on geologic map. This large variation on flow morphologies is an example of the general transitional-yet-highly-variable nature of intermediate-spreading-rate ridges. From Ballard et al. (1979).
Figure 6. a) Artist's conception of lava pillars on the seafloor. From Ballard et al. (1979). b) Interpretation of the filling and drainback of lava during a pre-1991 event on the EPR (9û46-51'N). Lava pillars form where trapped water (or fluid) rises through a flow, increasing the permeability and the local rate of cooling, and remain when lava drains away.
Figure 7. Inflated lava flows. The upper two figures are photos of inflated flows from the summit caldera of Axial Volcano on the JdFR. The bottom figure is a photo of an inflated flow lobe at Diamond Craters, OR. Flow inflation occurs as a result of overpressurization of lava under a solid crust, usually at low but continuous rates of effusion. From Appelgate and Embley (1992).
Figure 8. The width of the rift valley along a mid-ocean ridge is a function of the depth of dike intrusion and the magnitude of the dike-induced stress. From Chadwick and Embley, in press.
Figure 9. Models for the general evolution of mid-ocean ridges from a magmatically-dominated system to one dominated by tectonics. The magmatic stage is characterized by voluminous eruptions, followed by rift failure and less voluminous eruptions of more evolved compositions due to a cooling magmatic system. During the tectonic phase, there is amagmatic extension at slow ridges, or small eruptions and the formation of axial summit grabens at intermediate- and fast-spreading ridges. From Perfit and Chadwick, in review.
Figure 10. Estimated flow volumes, eruption interval, and frequency per 10 ka of volcanic events vs. ridge spreading rate. Eruption intervals and the number of eruptions are predicted from assumptions about the spreading rate and thickness of crust formed over 10 ka. From Perfit and Chadwick, in review.
Figure 11. Rift valley depth vs. ridge spreading rate. From XXXX
Figure 12. Geologic map and stratigraphy of the volcanic section in Oman. From Alabaster et al. (1982).
Figure 13. An interpretation of the evolution of the Oman volcanic stratigraphy, beginning about 95 Ma. From Alabaster et al. (1982).

Last updated January 28, 1998

aho@oregon.uoregon.edu