Recent eruptions on the CoAxial segment of the Juan de Fuca Ridge: Implications for mid-ocean ridge accretion processes

R. W. Embley,¹ W. W. Chadwick,² M. R. Perfit,³ M. C. Smith,³ and J. R. Delaney⁴

¹Pacific Marine Environmental Laboratory, National Oceanic and Atmospheric Administration, Hatfield Marine Science Center, Newport, Oregon 97365
²Cooperative Institute for Marine Resources Studies, Oregon State University, Hatfield Marine Science Center, Newport, Oregon 97365
³Department of Geology, University of Florida, Gainesville, Florida, 32611
⁴School of Oceanography, University of Washington, Seattle, WA 98195

Journal of Geophysical Research, 105(B7), 16,501–16,525 (2000).
Copyright ©2000 by the American Geophysical Union. Further electronic distribution is not allowed.

1. Introduction

The realization in the 1960s that the world-encircling mid-ocean ridge (MOR) generates most of the Earth's volcanic activity has led to an increasing focus on this system with progressively more powerful mapping systems [Detrick, 1986]. Although the scale of the feature (~60,000 km) and its relative remoteness have been impediments to a comprehensive study, the combination of digital swath mapping bathymetric and side-scan sonar systems, detailed MOR basalt (MORB) geochemical analyses, common depth point seismic profiling, and the use of deep-towed cameras and submersibles have resulted in significant progress during the last 2 decades in defining the major volcanic and tectonic "units" of the system [Detrick et al., 1987; Lonsdale, 1977; Macdonald and Fox, 1983; Sempere et al., 1993] and in understanding some of the fundamental accretionary processes which occur at the ridge crest [Ballard et al., 1979; Kappel and Ryan, 1986; Macdonald and Fox, 1983; Tucholke and Lin, 1994]. Although these studies have provided good definition of long-term (~10–10 years) patterns of crustal accretion and the tectonic framework of the oceanic crust, a detailed understanding of the temporal and spatial dynamics of MOR magmatism has been lacking. Unlike terrestrial volcanic areas, many of which are closely monitored with local seismic arrays capable of detecting and locating volcanogenic seismic events (M < 4), most of the MOR is monitored solely by the worldwide seismic net, which is limited to a detection threshold of M  4 [Fox, 1993/1994].

The first significant data about processes associated with discrete accretion events along the fast and intermediate spreading rate portions of the MOR became available from studies of sites on the Juan de Fuca Ridge (JdFR) and northern East Pacific Rise through the serendipitous discovery of seafloor eruptions [Chadwick et al., 1991; Embley et al., 1991; Haymon et al., 1993]. Detection of volcanic events on the Juan de Fuca and Gorda Ridges became possible in 1991 when the U.S. Navy allowed the National Oceanic and Atmospheric Administration (NOAA) access to the Sound Surveillance System (SOSUS) submarine hydrophone arrays in the northeast Pacific. The high efficiency of sound propagation in seawater in the Sound Fixing and Ranging (SOFAR) channel of the upper ocean lowers the detection threshold by about 2 orders of magnitude and increases the location precision by about 1 order of magnitude, making possible remote detection of, and subsequent response to, volcanic events in this area [Fox, 1993/1994]. As of late 1999, three eruptions on northeast Pacific spreading centers have been detected with this method and verified in the field [Dziak and Fox, 1999; Dziak et al., 1995; Fox and Dziak, 1998]. The first one was at the CoAxial segment in 1993.

In June 1993, within a week after the real-time acoustic monitoring system became operational, an unusual seismic swarm began on the northern part of JdFR just north of Axial Volcano (Figure 1) [Fox et al., 1995]. The swarm initially migrated north along the ridge axis for at least 25 km [Dziak et al., 1995] to the northern (deeper) end of the segment, where it subsequently became concentrated at approximately the center of the axial valley at 46°31N [Embley et al., 1995] (Figure 1). The only known eruption associated with the 1993 event occurs within this cluster of seismicity [Embley et al., 1995] (Figure 1). Several rapid response cruises took place between July and October 1993 and included water column plume surveys from the research vessels Tully, Discoverer, and Atlantis II, dives with the Remotely Operated Platform for Ocean Science (ROPOS) and manned submersible Alvin, and near-bottom surveys with the Scripps Deep-Tow and a U.S. Geological Survey (USGS) towed camera system. The details of the 1993 seismic swarm and the initial documentation of the water column and seafloor manifestation of the diking/eruption episode have previously been described in a series of papers [e.g., Fox, 1995].

Figure 1. Location map of central Juan de Fuca Ridge with place names used in text. Inset at bottom right corner shows location of area relative to northwest United States and Canada. AV is abbreviation for Axial Valley, V is for Volcano, and FZ is for fracture zone. Location of epicenters from June–July 1993 seismic swarms. Dashed lines show trends of Axial Volcano North Rift Zone (AVNRZ) and CoAxial neovolcanic zone. Diameters of circles are approximate length along axis of venting as observed in 1993. Contour interval is 200 m. Boxes outline areas covered by Figures 2, 5, 8, and 11.

The initial data sets generated several intriguing questions that have major implications not only for this particular event but also for the nature of MOR crustal accretion. An important first-order question was: Where did the dike come from? The tertiary water-borne seismic wave (T wave) epicenters were initially located west of CoAxial's axial valley, and the activity was originally thought to have begun on the north rift zone of Axial Volcano, which overlaps with the southern part of the CoAxial segment (Figure 1). However, the fresh lava and the active hydrothermal sites discovered during the 1993 cruises were aligned along the neovolcanic zone of CoAxial, with no indication that Axial had been involved. Water column and bottom surveys during the 1993 cruises localized three discrete venting sites (Flow site, Floc site, Source site) which were separated by 15–20 km from one another (Figure 1). So, a second question was: What was the geologic and hydrothermal manifestation of the event along the path of dike injection? Detailed seafloor mapping of the segment and time series measurements at these vent sites over several years subsequent to the diking event have helped us interpret what parts of the segment were perturbed by this event and which were unaffected. A third question was: Was the dike an isolated event or part of a longer-lasting episode of crustal accretion for this section of ridge? On the central volcanos of Iceland (our closest subaerial analogy to the MOR), stress builds up over long periods of time (hundreds of years) and is released during multiple events that occur on a decadal time scale [Björnsson, 1985], but until now, there has not been an analogous record of crustal accretion observations anywhere on the submerged portion of the MOR.

To answer these questions, the initial event response expeditions were followed by several other surface ship and submersible cruises in 1993–1996. Submersible dives in 1994–1995 continued the chemical and biological time series studies at the vent sites discovered in 1993 and conducted detailed mapping of the sites. A segment-scale mapping program in 1994–1996 used deep-towed side-scan sonar surveys, camera tows, and rock coring along the CoAxial neovolcanic zone and the north rift zone of Axial Volcano in order to place the detailed work at the active vent sites into a regional context.

This paper synthesizes the available SeaBeam, side-scan, towed camera, and submersible data to relate the 1993 CoAxial accretion event to the recent structural and magmatic history of the area and to the broader issue of MOR accretion processes. We believe our data demonstrate that (1) the 1993 dike intruded from a magma source beneath the CoAxial segment (rather than Axial Volcano) with a significant component of lateral propagation along the neovolcanic zone of the segment, and (2) the 1993 eruption was not an isolated event but was one of at least three volcanic events in the 12-year period between 1981 and 1993 which must have relieved a significant level of stress along the segment.

Return to Abstract or go to next section

PMEL Outstanding Papers

PMEL Publications Search

PMEL Homepage