Morphological Characteristics of Two MAR Spreading Segments
Local and regional variations in segment-scale morphological characteristics most likely reflect changes in crustal thickness and mantle thermal structure. The small-scale (tens of meters to hundreds of meters) volcanic and tectonic morphology also varies, which reflects changes in the processes acting at shallower levels.
Figures 1 and 2 present towed ocean bottom instrument (TOBI) high-resolution side-scan imagery mapped on to multibeam bathymetry for portions of two segments at the MAR opening at the same rate (~25 mm/yr), but with highly contrasting segment-scale morphology. TOBI provides high-quality images; in general, the image pixel size is a few meters [see Murton et al., 1992] . The resolution of a surface-ship-mounted, multibeam echosounder depends on water depth (range). At MAR depths, each depth value represents a patch of seafloor 100 200 m in diameter. Combining the side-scan imagery and multibeam bathymetry into a single composite image is extremely useful for interpreting the sonar imagery. Furthermore, combining the TOBI image and the colorized surface enhances the information available from the TOBI image alone.
Fig. 1. Three-dimensional perspective view of TOBI side-scan sonar data mapped onto the color bathymetric surface for a portion of the inner valley floor of a spreading segment near 29° N at the MAR. The area of the image is 15 km east to west and 13 km north to south. Several TOBI swaths have been merged together across the image. The track of the TOBI vehicle is at the center of a swath; scalloping along the track is an artifact caused by a bottom tracking problem. No information is obtained directly below the vehicle. Bright areas indicate a reflection, and dark areas indicate a shadow or attenuated return from sediment-covered terrain. The cratered seamount marked on the image is ~220 m in relief with a summit crater ~600 m in diameter. Note how the hummocky-textured AVR winds its way along the valley floor. Also, note the volcanic highs extending east/west from the AVR into its flanking deeps.
Fig. 2. Three-dimensional view of TOBI side-scan sonar data mapped onto the color bathymetric surface for a portion of the inner valley floor of a spreading segment near 25° N at the MAR. The area of the image is 15 km east to west and 22 km north to south. The morphological center of this segment is a topographic low filled with smooth-textured flows that are extensively fissured and faulted. Isolated volcanic constructs sit stratigraphically on top of the smooth flows. The region within the white rectangle is shown in detail in Figure 3.
Figure 1 shows a spreading segment located near 29° N. It is ~60 km long; the average width of the inner valley floor is 6.5 km. A large negative mantle Bouguer gravity anomaly (~20 mGal variation along the segment) is located near the center of the segment, suggesting a substantial increase in crustal thickness (3.5 km) at the center compared to the ends [Lin et al., 1990] . An AVR runs most of the length of the inner valley floor. The AVR widens and narrows along its length, but is typically a few kilometers wide and about 150 m high. Seamounts are scattered along it.
At the TOBI scale, the volcanic morphology of the 29° N segment is dominated by hummocky ridges and hummocky flows (Figure 1). A seamount ~220 m in relief with a large summit crater ~600 m in diameter sits within the AVR, and rises well above the average AVR height of ~150 m. The volcanic morphology of the AVR is relatively unfaulted, suggesting that eruptions here keep pace with tectonic deformation caused by seafloor spreading. In places, topographic highs extend east and west of the AVR; overlapping side-scan swaths show that they are flat-topped, smooth-textured, and commonly cut by small faults. On the west, one large, more or less continuous, fault displaces the crust vertically to form the bounding wall; by contrast, several smaller faults displace the crust a similar vertical distance to form the eastern bounding wall.
Figure 2 shows a spreading segment located near 25° N. It is ~55 km long; the average width of its inner valley floor is ~7.5 km. This segment also has a large negative mantle Bouguer anomaly near the center of the segment (~22 mGal variation) indicating crustal thickening [Smith et al., 1997] . In sharp contrast to the segment near 29° N, however, this segment contains no AVR. Instead, a shallow (~150 m) central low extends along much of its length.
By contrast to the hummocky AVR that has been built in the segment shown in Figure 1, the center of the segment located near 25° N (Figure 2) is covered by extensive smooth flows; small volcanic constructs are scattered across these smooth flows (Figure 3). The central topographic low that runs along the center of the segment is highly fissured and faulted. In some places volcanic features mask the faults and fissures, suggesting recent construction. A scarp a few hundred meters in relief has formed within the valley floor; sediment has accumulated to the east of the fault, indicating a lack of recent volcanism in this region.
Fig. 3. TOBI side-scan sonar image of the small volcanic ridge within the central low of the spreading segment near 25° N; it is marked on Figure 2. Illumination is from the right. The ridge is ~3.3 km long, ~400 m wide, and ~30 m high. The along-axis orientation of the ridge mimics neighboring faults and fissures, and it is inferred to have erupted from similar fissures. The smooth, unfaulted flows surrounding the ridge are interpreted to be part of the eruption that built the ridge. These flows cover pre-existing faults and fissures. Information about the small-scale volcanic products and faults gained from images such as these provide insight into the processes of magmatism and tectonism at the axis of the MAR (see Figure 4).
The differences in eruptive styles between the two segments that is, hummocky flows and the construction of an AVR at 29° N compared with flows that spread away from their vents rather than building a major edifice at 25 occur despite the similarity in crustal thickness and mantle temperature variations inferred from the gravity data. This implies that other variables play a major role in controlling the small-scale morphology of these segments.
Models for Eruptive Processes at the MAR
The results of work with the high-resolution TOBI side-scan imagery at the MAR, observations of subaerial eruptive products in Iceland and Hawaii, and results from theoretical models [e.g., Head et al., 1996] provide the basis for a working model of eruptive processes at the MAR (Figure 4).
Fig. 4. a) Cartoon of fissure eruptions and their volcanic products. Where the dike breaches the surface, a fissure eruption occurs. Initially, surface flows will erupt along the length of the fissure. Eruption conditions rapidly evolve, however. Hummocks and hummocky ridges (hummocks aligned along the fissure) are produced when cooling of the dike leads to an eruption that emanates from discrete vents. Seamounts are produced when the eruption centralizes to a single vent. The seamount at the front edge of the cartoon is shown to indicate that fissures may evolve quickly to produce only a seamount, and fissures may be covered by their products. b) TOBI side-scan sonar data of part of the AVR from the 29° N segment (Figure 1) showing volcanic features similar to those illustrated in the left panel. Illumination is from the right. The area of the image is 3 km east to west, and 5 km north to south. c) Photograph of an ~10-km-long section of the Laki Fissure zone in Iceland. A string of hummock-sized cones has been built. Although many of these cones have been partially destroyed, individual edifices can be identified. The intact cone in the center of the image is ~200 m in diameter and ~30 m high, comparable to the size of hummocks at the MAR.
We envision that, similar to magma transport in subaerial volcanic rift zones, magma is transported from the regions of accumulation for example, in the shallow mantle, at the base of the crust, or within the crust to the seafloor through dikes by a combination of lateral and vertical motion. Where the dike intersects the seafloor, a fissure eruption begins. On land, fissure eruptions commonly start with surface flows and fire fountaining along the fissure, but rapidly become localized to a number of discrete vents. Again, like subaerial eruptions, the length of surface flows and their morphology (hummocky or smooth) are controlled by variables such as lava viscosity, effusion rate, cooling rate, and underlying slope. Hummocks and hummocky ridges are produced as the eruption becomes confined to several discrete vents, similar to subaerial spatter cones or spatter ramparts that form along fissures (Figure 4). A large seamount is constructed when the eruption becomes restricted to a single vent, much like Pu'u `O'o on the East Rift Zone of Kilauea Volcano. Surface flows, hummocky ridges, and seamounts are therefore the possible constructional products of a single eruption. At larger scales, axial volcanic ridges are built by multiple dike intrusions and their associated eruptive products, which are focused within a narrow region of the axial zone.
Based on this model, the hummocks and hummocky ridges that dominate the small-scale volcanic morphology of the AVR in the 29° N segment are formed by the focusing of fissure eruptions to several vents; the seamount is formed by the rapid evolution of an eruption to a single vent. The meandering of the AVR along the length of the axis is caused by the misalignment of fissure eruptions, the location of which may be controlled by a changing stress regime along the axis. The extensive low-relief flows observed on the valley floor of the 25° N segment would thus be generated by the eruption of low viscosity lavas with high effusion rates through fissures. Large volumes of lava may be erupted. The fact that volcanic edifices have been constructed on top of these smooth flows suggests that one or more eruption variables changed during these or subsequent eruptions.
Also, the highs extending to the east and west of the AVR in Figure 2 are morphologically similar to the perched lava ponds of the East Rift Zone of Kilauea Volcano [Wilson and Parfitt, 1993] . Perched lava ponds are thought to form when a channelized flow encounters a radical decrease in slope (such as that at the base of the AVR), and spreads out radially. Enhanced cooling in this new geometry causes the flow to come to rest forming flat, circular benches with steep, leading edges [Wilson and Parfitt, 1993] . Alternatively, these features could be isolated edifices that were built abutting the AVR or were partially buried by subsequent eruptions at the AVR as suggested by Smith and Cann [1992] . Determining the origin of these features awaits more detailed imaging and sampling such as has been done on Serocki Volcano located in an MAR segment south of the Kane transform [Bryan et al., 1994] . There geochemical evidence suggests that the flat-topped volcano was fed by a lava tube from the adjacent AVR.
The majority of MAR flows imaged by near-bottom side-scan sonar instruments have a hummocky surface morphology at a scale not defined by the multibeam bathymetry. The hummocks may be equivalent to tumuli that are ubiquitous topographic features on pahoehoe flows of basaltic volcanoes. Tumuli form where overpressure within a flow leads to swelling and cracking of the lava surface; they are typically a few meters high but can reach heights of 10 m, and are elongate to near-circular in plan shape [Walker, 1991] . Fields of circular tumuli in some places overlapping and piling up to form larger swells are seen on recent flows of the current eruptive vent on the East Rift Zone of Kilauea Volcano. Outbreaks of lava commonly form volcanic carapaces on tumuli, giving some the appearance of small eruptive centers. Whether or not the hummocky surface textures observed in the data are due to the formation of tumuli remains to be answered.
Future Directions
The increasing amount of high-resolution side-scan sonar images of mid-ocean ridges makes it possible to start examining the patterns in small-scale volcanic and tectonic morphology, and how they might relate to the processes of dike emplacement, eruption, and faulting. In addition, the capabilities of most near-bottom side-scan sonar systems have been enhanced to obtain co-registered fine-scale phase bathymetry. As more of these data are collected, we will be able to constrain the sizes and shapes of the small-scale volcanic and tectonic features generated at the axis of the MAR, and make more rigorous comparisons to subaerial features. Within the context of these data, it will be possible to design future geophysical and geochemical experiments that are on the same scale as those currently conducted at subaerial volcanic rift zones. Studies such as these, in combination with existing data, will rapidly advance our understanding of the magmatic and tectonic processes associated with rifting processes at the MAR.
Acknowledgments
The TOBI deep-towed side-scan vehicle was developed at the Institute of Oceanographic Sciences (now part of the Southampton Oceanography Centre) in the United Kingdom with funds from the Natural Environment Research Council. The Council also supported RRS Charles Darwin Cruise 65, on which the images were obtained.
References
Ballard, R. D., and Tj. H. van Andel, Morphology and tectonics of the inner rift valley at lat 36°50'N on the Mid-Atlantic Ridge, Geol. Soc. Am. Bull., 88, 507–530, 1977.
Bryan, W. B., S. E. Humphris, B. Thompson, and J. Casey, Comparative volcanology of small axial eruptive centers in the MARK area, J. Geophys. Res., 99, 2973–2984, 1994.
Head, J. W., III, L. Wilson, and D. K. Smith, Mid-ocean ridge eruptive vents: Evidence for dike widths, eruption rates, and evolution of eruptions and axial volcanic ridges, J. Geophys. Res., 101, 28,265–28,280, 1996.
Lin, J., G. M. Purdy, H. Schouten, J.-C. Sempere, and C. Zervas, Evidence from gravity data for focused magmatic accretion along the Mid-Atlantic Ridge, Nature, 344, 627–632, 1990.
Murton, B. J., I. P. Rouse, N. Millard, and C. G. Flewellen, Deep-towed vehicle explores the ocean floor, Eos, Trans. AGU, 73, 1–4, 1992.
Smith, D. K., and J. R. Cann, The role of seamount volcanism in crustal construction along the Mid-Atlantic Ridge at 24°–30° N, J. Geophys. Res., 97, 1645–1658, 1992.
Smith, D.K., et al., Mid-Atlantic Ridge Volcanism from Deep-Towed Side-Scan Sonar Images, 25°–29° N, J. Vol. Geotherm. Res., in press, 1997.
Walker, G. P. L., Structure, and origin by injection of lava under surface crust, of tumuli, "lava-rise pits", and"lava-inflation clefts" in Hawaii, Bull. Volcanol., 53, 546–558, 1991
Wilson, L., and E. A. Parfitt, The formation of perched lava ponds on basaltic volcanoes: the influence of flow geometry on cooling-limited lava flow lengths, J. Vol. Geotherm. Res., 56, 113–123, 1993.
Back to Science for Everyone
