Fine-scale segmentation of volcanic/hydrothermal systems along fast-spreading ridge crests

Fine-scale (3rd and 4th order) segments of the mid-ocean ridge (MOR) are morphologically defined units of crustal accretion with enigmatic origins. At fast-spreading centers, 3rd order segments are bounded by discontinuities in the structure of the axial high. Most 3rd order segments are 20F10 km long, and terminate at ridge axis discontinuities (RADs) exhibiting 0.2–3 km lateral offsets and/or z20 m axial depth increases. Nested within 3rd order segments are shorter 4th order segments bounded by smaller RADs. Along the fast-spreading East Pacific Rise (EPR) at 10802V– 9808VN and 17811V–42VS, spatial distributions of active high-temperature hydrothermal vents, biological communities, and lava flow morphology are known accurately from extensive near-bottom surveys and submersible dives along a total of seven 3rd order segments and twenty-four 4th order segments. The cumulative northern and southern EPR data show that 3rd order segments (but not 4th order segments) consistently exhibit apparently higher rates of lava effusion near segment middles in comparison to segment ends, supporting the hypothesis that each 3rd order segment is a discrete, centrally fed volcanic system, or bvolcanic segmentQ. Spatial analyses of the cumulative hydrothermal data reveal a striking increase in abundance of hydrothermal features within the middle portions of 3rd order segments. Approximately 60% of ongoing high-temperature focused flow is concentrated within the middle 40% of 3rd order segments, and twice as many biological communities are found at mid-segment in comparison to segment ends. The two independent data sets—volcanic and hydrothermal—both indicate that magmatic heat sources are focused beneath the central portions of 3rd order segments, and that magmatic/hydrothermal systems are spatially and temporally segmented at 3rd order length and time scales. We propose that 3rd order segmentation arises from processes controlling melt supply to the middle-tolower crust; and, that 4th order segmentation largely arises from more rapid/frequent upper crust processes of crack formation and dike intrusion. Future work is needed to test the applicability of our results along portions of the fast-