return to SeafloorScott's HomePage

IN RIDGE Events Dec. 1999

Volcanoes of the southern East Pacific Rise:

A New View of Crustal Accretion and Ridge Segmentation at Super-fast Spreading Rates

Scott M. White , Ken C. Macdonald , and Rachel M. Haymon

Dept. Geological Sciences and Marine Science Institute, University of California, Santa Barbara

and the Sojourn-2 Science Party (S. Baron, B. Bezy, E. Birk, L. Crowder, G. Levai , L. Magde, J. O’Neill, D. Schierer, P. Sharfstein, S. Sudarikov, D. Wright)

 

Introduction

Meter-scale DSL-120 sonar mapping and co-registered Argo II photographic observations reveal changes in volcanic structures and inferred lava effusion rate that closely follow the 3^rd order structural segmentation of the ridge axis on the southern East Pacific Rise (SEPR) at 17°11’-18°37’S. Near 3^rd order segment ends, we observe abundant pillow lavas and lava domes that suggest an eruptive style dominated by low-effusion rate eruptions. Near 3^rd order segment centers, we find smooth lobate and sheet lava flows associated with collapsed lava lakes that suggest high-effusion rate eruptions.

Evidence from previous geochemical and seismic studies indicates a segmented magma supply to fast-spreading mid-ocean ridges [e.g. Langmuir et al., 1986; Sinton et al., 1991; Batiza, 1996; Carbotte et al., 1997; Dunn and Toomey, 1997; Singh et al., 1998; Toomey et al., 1998]. Our fine-scale volcanological observations merged with the insights provided by earlier geochemical and seismic work on mid-ocean ridges permits us to develop a model for the significance of 3rd order ridge axis discontinuities (RAD’s) as boundaries between individual volcanic systems along the ridge axis (Table 1).

Table 1. Characteristics of segmentation on fast-spreading ridges updated from [Macdonald et al., 1991]. Changes based on detailed observations of southern East Pacific Rise, shown in italics.

 

Data Collection

To achieve both the high level of resolution needed to image important volcanological features (eruptive fissures, volcanic vents, and lava flow fields), and the extensive areal coverage needed to survey entire ridge segments, we employed the "nested-survey" concept with SeaBeam 2000, DSL-120, and Argo II data. Initial mapping with the SeaBeam 2000 multibeam sonar system accurately located the ridge axis along the ridge segments selected for the DSL-120 survey. On this part of the SEPR, the ridge axis lies (1.) at the crest of a smoothly domed axial high at 17°11’-56’S, (2.) to the west of a 15-30m high west-facing scarp that apparently defines a half-graben at 17°56’-18°09’S, (3.) within a broad but shallow trough at 18°09’-22’S, and (4.) within a deeper but narrower trough at 18°22’-37’S. Using the DSL-120 sonar system, we then imaged a 900-1000 m wide sidescan reflectivity swath, and a 700-800 m wide bathymetric swath, continuously along 150 km of the ridge crest. The DSL-120 survey provided a base-map for the far more detailed observations of the axial neovolcanic zone with Argo II at 17°15’-17°40’S and 18°23’-18°29’S. During the 1996 Sojourn 2 expedition to the SEPR [Haymon et al., 1997], Argo II was configured with 3 video cameras, 2 still cameras, Imagenex 675 kHz profiling sonar, 200 kHz sidescan sonar, CTD, transmissometer, and magnetometer.

 

Lava lakes with overflow tubes and channels

At the crest of the smoothly domed portion of the axial high, two areas of extensive collapse in the volcanic carapace were found. Both areas were completely imaged at the meter-scale using the DSL-120 sidescan sonar system, and subsequently one area was densely surveyed with the Argo II photo-acoustic sled. Both areas of extensive collapse have a similar structure (Figure 1). The collapses are 10-15 m deep, elongate parallel to the spreading axis, and never more than 100 m wide. Both zones of collapse continue for approximately 7 km along-axis, but are composed of several elongate collapse pits arranged en echelon. These areas of extensive collapse are morphologically similar to the axial summit collapse trough at 9°26’-56’N on the EPR [Haymon et al., 1991; Fornari et al., 1998].

Figure 1. DSL-120 backscatter image of a portion of Aldo Lake trough (leftmost panel), and geologic interpretation (second to left panel). The sidescan image shows higher backscatter reflectivity as darker areas. The line drawing interpretation outlines the major collapses that represent individual lava lakes with solid black line. The lava lakes are arranged en echelon, implying they formed over en echelon eruptive fissures. Argo II observations show that these collapses are mostly in hollow lobate lava, and had extensive areas of hydrothermal venting in 1996. Sheet lava flows and collapsed lava tubes are shown with light gray fill.

We observe in the DSL-120 sonar images that collapse areas are the source of lava tubes and sheet flows (Figure 1), and the loci of hydrothermal flow located with Argo II [Haymon, et al, 1997]. We hypothesize that these collapse features form directly over eruptive fissures. Fissure eruptions in Iceland and Hawaii [e.g. Thorarinsson, 1969; Lockwood et al., 1987] have produced en echelon segments similar in length and offset to the SEPR collapsed lava lakes. The Argo II visual imagery reveals that the margins of the collapses have flattened lobate and sheet-flow morphology with abundant relict lava channels or collapsed lava tubes, similar to the features produced by subaerial fissure eruptions. Both the local lava morphology of primarily smooth lobated and sheeted flows, and the relatively thin roof over large areas of void space adjacent to the collapses suggest a steady, high lava effusion rate [Gregg and Fink, 1995].

No such large collapses, or associated lava distribution systems, were observed in the DSL-120 records south of 17°56’S where the spreading axis lies along the floor of a deep fault-bounded axial trough. Yet, evidence of lava ponding and subsequent collapse is documented visually in the areas near 18°16’S and 18°25’S by the Argo II survey, as well as by submersible dives in this area [Auzende et al., 1996; Sinton et al., 1999]. It is likely that the high vertical scarps in the neovolcanic zone south of 17°56’S control the geometry of lava lakes by damming the lava flows, and acoustically obscure the collapses in the DSL-120 sonar images of the trough floor.

 

Axial Pillow Lava Domes

Low lava domes form on the ridge axis, and are found mainly in areas that are free of collapsed lava lake complexes, and lava channels. Imaged in 1996 using the DSL-120 sonar, the axial domes are subtle topographic features which average 200 meters in diameter and 20 meters in height (Figure 2). They are too large to notice in visual observations of the seafloor, yet too small to be recognized in previous lower-frequency sonar mapping. Some coalesce into rows aligned with the ridge axis, as if erupting from an axis-parallel fissure, although the outlines of discrete quasi-circular structures can still be traced in plan view (Figure 2). Axial domes can also be found as isolated edifices or in small clusters. Visual observations made with Argo II help to establish the primary vent origin of the axial lava domes imaged by the DSL-120 sonar [White et al., 2000]. Photos from Argo II reveal that the domes are made mainly of pillow lava, with minor lobate lava flows occurring primarily near the summits of some of the larger domes. A clear correlation that emerged from the survey data was the large number of axial domes near the ends of 3^rd order segments.

Figure 2. An example of axial pillow lava domes found at the end of a 3rd order ridge segment. Left panel shows the DSL-120 sidescan reflectance image from data gridded at 2m. Darker areas in the sidescan have higher backscatter reflectivity. On the right panel, the DSL-120 bathymetry, gridded at 5 m, has been contoured at 5 m intervals. Black outlines show the digitized boundaries of the domes. Large fissures are continuous through several domes.

In this study, the 3rd order structural ridge segments include all of the segments bounded by either non-overlapping offsets of the spreading axis, overlapping spreading centers (OSC’s), or saddle points that can be detected at SeaBeam2000 resolution (Table 2). When all 3rd order segments in the survey area are used, the frequency distribution of axial lava domes shows a clear pattern of decrease from segment end to segment center for normalized distance along 3rd order segments (Figure 3). No consistent pattern emerges when the distribution of axial lava domes is normalized to either higher or lower order ridge segments; only the 3^rd order segments display this pattern of volcanism.

Figure 3. Histogram of the cumulative number of axial pillow lava domes for 3^rd order segments in the survey area. Bin interval is normalized to represent 20% of the distance from segment end to center in order to adjust for the different segment lengths.

 

Table 2. Third Order Ridge Axis Discontinuities in the study area, locations and description.

Latitude	Longitude	Structure/Morphology	Length (km)
17° 11’ S	113°8.5’ W	Start of survey midsegment
17° 18.5’ S	113°10.6’ W	Saddle point	14.7
17° 30’ S	113°13.5’ W	Left-stepping non-overlapping offset	22
17° 35.75’S	113°15’ W	Left-stepping non-overlapping offset	11
17° 40’S	113°15.75’W	Saddle point	8.1
17° 56’ S	113°17.75’W	Left-stepping overlapping spreading center	29.6
18° 02.5’ S	113°19’ W	Left-stepping non-overlapping offset	12.3
18° 09’ S	113°20.75’W	Left-stepping overlapping offset at transition between half-graben and full-graben ridge axis	12.4
18° 22’ S	113°22’ W	Left-stepping overlapping spreading center	24.6
18° 30.75’S	113°24.7’ W	Left-stepping non-overlapping offset	18.5
18° 33.5’S	113°25.25’W	End of survey North of overlapping spreading center

 

Volcanic Segmentation

Although 3rd order RAD’s are the smallest offsets of the ridge axis that we can resolve solely on the basis of multibeam bathymetry, they correlate to the variations in the volcanic structures observed independently by our near-bottom survey. We observe high-effusion rate lava flow morphologies and collapse features near 3^rd order segment centers, and low-effusion rate lava morphologies and volcanic edifices near 3^rd order segment ends. Thus, we propose that 3^rd order ridge segments represent individual volcanic systems with decreasing lava effusion rates away from the center of the system.

Rock dredging and wax coring of the ridge crest at a spacing much smaller than the 3^rd order segment length has found that changes in the lava chemistry occur at small RAD’s such as "DeVAL’s" and OSC’s [Langmuir et al., 1986; Sinton et al., 1991; Reynolds et al., 1992]. Seismic investigations have found that the properties of the axial magma chamber also change on a smaller scale than that defined by large OSC’s and transform faults [Singh et al., 1998; Carbotte et al., 1997]. These fine-scale variations of geophysical and geochemical data fit very well with our hypothesis that 3rd order segments behave as individual volcanic systems (Figure 4).

Figure 4. Schematic along-axis cross-section showing how the distribution of volcanic features is related to segmentation on the SEPR. The top panel shows one complete 2nd order segment, encompassing two 3rd order segments. Eruptive style along the ridge varies systematically from the low-effusion rate axial pillow lava domes found at the ends of 3rd order segments, to high-effusion rate lava lakes and lobate flows found near mid-segment. Lateral melt migration within the axial magma chamber (AMC) or variable magma supply below the AMC might lead to the pattern of volcanism observed on 3rd order ridge segments. Slight pinching and swelling of the AMC, or variations in melt:crystal ratios, might correspond to the 3rd order ridge segmentation, but the AMC usually disappears at 2nd order RAD’s.

 

Such an individual volcanic systems would be immediately recognized as a volcano in other settings where eruptions tend to form large volcanic edifices. Thus, while 3rd order RAD’s on fast-spreading ridges are small offsets (<1 km) of the ridge axis, or significant saddle points in axial topography, their analogues at slow-spreading rates are gaps between large volcanic edifices on the floor of the median valley [Macdonald et al., 1991; Smith et al., 1999]. However, the thin lithosphere at the SEPR has a low flexural rigidity that prevents building a large volcanic edifice on the axis [Buck et al., 1997]. Rather than being a constructional volcanic ridge, the axial high of a fast-spreading ridge is thought to be an isostatic or flexural feature [Madsen et al., 1984; Eberle and Forsyth, 1998]. Because the axial high of the SEPR, and other fast-spreading ridges, are isostatic rises instead of being constructional ridges, the highest points on the ridge do not mark the centers of individual volcanic constructions. If the rise were a volcanic construction, the volcanic extrusive layer would be thickest at the ridge axis; however, seismic layer 2A, interpreted as the volcanic layer, is consistently thinnest here. Thus, the elevated topography of the ridge axis on fast-spreading ridges cannot be compared directly to constructional volcanic ridges like those found in the median valley of slow-spreading ridges [Smith et al., 1999]. The significance of 3rd order segmentation is that it reveals the volcanic segmentation of the ridge in the absence of large volcanic constructions.

In this context, what are 4th order segments? The RAD’s defining 4th order segment boundaries are so small that near-bottom, high-resolution images are usually required to notice them. In most cases, they are probably the products of axial fissure eruptions. This is how they were originally defined by Macdonald et al. [1991]. While 3rd order segments may remain distinct for hundreds to thousands of years, they cannot persist for longer than 103-105 yrs., or else they would create an off-axis discordant zone. Changing with each new event of crack propagation and dike intrusion, 4th order segments are even more ephemeral, and may persist as distinct segments for just tens to hundreds of years (Table 1).

 

Conclusion

The eruptive style and types of volcanic structures formed on the SEPR vary systematically over 3rd order segments in a manner that suggests 3^rd order segments are the volcanoes that comprise the SEPR. Axial volcanic edifices in the form of small pillow lava domes are found on fast-spreading ridges, and increase in abundance toward 3rd order RAD’s. The general pattern of increasing abundance of axial lava domes toward 3rd order RAD’s reflects a decrease in overall effusion rate toward segment ends. Near mid-segment, collapsed lava lakes are associated with thin-crusted, hollow lobate lava flows and smooth sheet flows that suggest high effusion rates during eruption. Lava conduits are common within these lobate and sheet lava regions, and collapses in the lobate lava suggests that these flows have inflated several meters in areas proximal to their eruptive vent. Although not mapped in their entirety, the eruptions producing collapsed lava lakes on the SEPR are probably more voluminous than the pillow flows forming the axial lava domes.

 

References

Auzende, J.-M., V. Ballu, R. Batiza, D. Bideau, J.-L. Chalou, M.-H. Cormier, Y. Fouquet, P. Geistodoerfer, Y. Lagabrielle, J. Sinton, and P. Spadea, Recent tectonic, magmatic, and hydrothermal activity on the East Pacific Rise between 17S and 19S: Submersible observations, J. Geophys. Res., 101, 17995-18010, 1996.

Batiza, R., Magmatic segmentation of mid-ocean ridges: a review, in Tectonic, Magmatic, Hydrothermal, and Biological Segmentation of Mid-Ocean Ridges, Geological Society of America Special Publication, 1, edited by C. J. Macleod, P. A. Tyler and C. L. Walker, 103-130, Geological Society of America, 1996.

Buck, W. R., S. M. Carbotte, and C. Mutter, Controls on extrusion at mid-ocean ridges, Geology, 25, 935-938, 1997.

Carbotte, S., J. C. Mutter, and L. Xu, Contribution of tectonism and volcanism to axial and flank morphology of the southern EPR from a study of layer 2A geometry, J. Geophys. Res., 102, 10165-10185, 1997.

Dunn, R.A., and D.R. Toomey, Seismological evidence for three-dimensional melt migration beneath the East Pacific Rise, Nature, 388, 259-262, 1997.

Eberle, M.A., and D.W. Forsyth, An alternative dynamic model of the axial topographic high at fast spreading ridges J. Geophys. Res., 103, 12309-12320, 1998.

Fornari, D. J., R. M. Haymon, M. R. Perfit, T. K. P. Gregg, and M. H. Edwards, Axial Summit Trough of the East Pacific Rise 9°N to 10°N: Geological Characteristics and Evolution of the Axial Zone on Fast-Spreading Mid-Ocean Ridges, J. Geophys. Res., 103, 9827-9855, 1998.

Gregg, T. K. P., and J. H. Fink, Quantification of submarine lava-flow morphology through analog experiments, Geology, 23, 73-76, 1995.

Haymon, R. M., D. J. Fornari, M. H. Edwards, S. M. Carbotte, D. Wright, and K. C. Macdonald, Hydrothermal vent distribution along the East Pacific Rise crest (9 degrees 09'-54'N) and its relationship to magmatic and tectonic processes on fast-spreading mid-ocean ridges, EPSL, 104, 513-534, 1991.

Haymon, R. M., and 15 others, Distribution of fine-scale hydrothermal, volcanic, and tectonic features along the EPR crest, 17° 15’- 18° 30’S: results of near-bottom acoustic and optical surveys, EOS, Trans. Am. Geophys. U., 78, F705, 1997.

Langmuir, C. H., J. F. Bender, and R. Batiza, Petrological and tectonic segmentation of the East Pacific Rise, 5°30'N-14°30'N, Nature, 322, 422-429, 1986.

Lockwood, J. P., J. J. Dvorak, T. T. English, R. Y. Koyanagi, A. T. Okamura, M. L. Summers, and W. R. Tanigawa, Mauna Loa 1974-1984; a decade of intrusive and extrusive activity, in Volcanism in Hawaii, USGS Professional Paper 1350, edited by R. W. Decker, T. L. Wright and P. H. Stauffer, 537-570, USGS, Reston, VA, 1987.

Macdonald, K. C., D. S. Scheirer, and S. M. Carbotte, Mid-ocean ridges; discontinuties, segments and giant cracks, Science, 253, 986-994, 1991.

Madsen, J. A., D. W. Forsyth, and R. S. Detrick, A new isostatic model for the East Pacific Rise crest, J. Geophys. Res., 89, 9997-10015, 1984.

Reynolds, J. R., C. H. Langmuir, J. F. Bender, K. A. Kasten, and W. B. F. Ryan, Spatial and temporal variability in the geochemistry of basalts from the East Pacific Rise, Nature, 359, 493-499, 1992.

Singh, S. C., G. M. Kent, J. S. Collier, A. J. Harding, and J. A. Orcutt, Melt to mush variations in crustal magma properties along the ridge crest at the southern East Pacific Rise, Nature, 394, 874-878, 1998.

Simkin, T., and L. Siebert, Volcanoes of the World, 349 pp., Geoscience Press and Smithsonian Institution, Tucson, AZ, 1994.

Sinton, J. M., S. M. Smaglik, J. J. Mahoney, and K. C. Macdonald, Magmatic processes at superfast spreading mid-ocean ridges: Glass compositional variations along the East Pacific Rise 13°-23°S, J. Geophys. Res., 96, 6133-6155, 1991.

Sinton, J. M. and 15 others, Volcanological investigations at superfast spreading: results from R/V Atlantis Cruise 3-31, RIDGE Events, 10, 17-23, 1999.

Smith, D. K., M. A. Tivey, H. Schouten, and J. R. Cann, Locating the spreading axis along 80 km of the Mid-Atlantic Ridge south of the Atlantis Transform, J. Geophys. Res., 104, 7599-7612, 1999.

Thorarinsson, S., The Lakagigar eruption of 1783, Bull. Volcanol., 33, 910-927, 1969.

Toomey, D. R., W. S. D. Wilcock, S. C. Solomon, W. C. Hammond, and J. A. Orcutt, Mantle seismic structure beneath the MELT region of the East Pacific Rise from P and S wave tomography, Science, 280, 1224-1226, 1998.

White, S. M., K. C. Macdonald, and R. M. Haymon, Basaltic lava domes, collapsed lava lakes, and volcanic segmentation of the southern East Pacific Rise, J. Geophys. Res., 105, 23,519-23,536, 2000.