The Galapagos Rift at 86°W: 5. Variations in volcanism, structure, and hydrothermal activity along a 30‐kilometer segment of the Rift Valley

Abstract

A 30‐km segment of the Galapagos Rift near 86°W has been mapped in detail using the Angus towed camera system, the submersible Alvin, and multi‐narrowbeam sonar data. Recent volcanic activity and active hydrothermal circulation are evident along the entire length of the segment mapped. There are, however, clear along‐strike variations in these processes which render previous two‐dimensional models obsolete. Although alternate explanations are possible, eruptive sequences appear to begin with the outpouring of surface‐fed sheet flows and end with more channelized pillow flows. In the western portion of the rift studied, sheet flows dominate with the entire valley floor covered by recent flows associated with a broad shield volcano. The eastern portion, on the other hand, is narrower; consisting primarily of less voluminous pillow flows of apparently the same youthful age. Three possible models for the volcanic evolution of this rift segment are presented. According to the first model, the extrusive portion of the crust is formed by a distinct volcanic episode, followed by a long period of volcanic quiescence. The volcanic phase begins with voluminous sheet flows emerging from numerous eruptive fissures, which in time evolve into a narrow pillow ridge. Farther along‐strike, where the flows are smaller and the extrusive zone narrow, the marginal portions undergo continued fissuring and subsequent uplift to form marginal highs and lows. This deformational activity also affects the extrusive zone once volcanic activity ends, converting the distinctly lobate topography of the active period into highly lineated fault‐controlled terrain. According to the second model, extension and voleanism can be viewed as a continuous process without major periods of volcanic quiescence. The initial lava flows of a new eruptive sequence fill low areas, frequently spilling over local sills and flooding much of the rift valley. These sheet flows are then capped by pillow ridges, but from an areal view sheet flows dominate, since they can flow over much greater distances. In the third model, along‐strike differences in flow type arise from ephemeral differences in spreading rate along adjacent segments of the rift separated by transient transform faults or adjacent segments of en echelon spreading segments. In this third model, eruptions dominated by sheet flows are associated with a faster spreading element than eruptions dominated by pillow flows. Active hydrothermal vent fields were found throughout the study area as well as numerous yellow and ocher accumulations of what appear to be massive sulfide deposits. These latter deposits are not associated with active venting, apparently reflecting previous active hydrothermal phases when the exiting temperatures were sufficently high to permit surficial sulfide deposition. The active vent fields exhibit along‐strike variations in their exiting temperatures, either due to ephemeral differences in spreading rate or variations in the volcanic flow sequences.