Geochemical, structural, and morphological variability at intermediate-spreading ridges are manifestations of mantle-melting and magmatic differentiation processes and their complex interplay with tectonism. Because melt production, transport, accumulation, and eruption processes can vary on the segment-scale over decades (Rubin et al., 2005), MOR study requires the ability to identify, sample, and date individual flow units (Rubin et al., 2001) in order to capture a representative range of volcanological and geochemical variability through time. Even when detailed seafloor observations or mapping exists (e.g., the Juan de Fuca Ridge) understanding the fine-scale petrologic, volcanologic, and magmatic evolution of mid-ocean ridges requires evaluating age relationships among lava flow sequences (Christie, 1994, Rubin & MacDougall, 1990). Here we detail recent results at Axial Seamount, which is part of a larger collaborative study to investigate timescales and magnitudes of geological, volcanological, and chemical variability on the southern JdFR (including CoAxial and northern Cleft).

Axial Seamount hosts the shallowest (~1.5 km bsf; Carbotte et al., 2008) and likely most robust axial magma chamber of the JdFR. Migration of the JdFR due to northwesterly absolute plate motion has led to intersection with the Cobb hotspot and magma oversupply at the present site of Axial Seamount since <0.5 Ma (Karsten & Delaney, 1989). Multibeam mapping and coregistered sidescan data collected with MBARI's AUV D. Allan B. at Axial Seamount since 2006 total ~73 km² and include the horseshoe-shaped summit caldera (3x 8x 0.15 km) and rim. The only known eruption of Axial Seamount occurred in 1998. Lateral dike injection and eruption were associated with an earthquake swarm (Dziak & Fox, 1999) and produced vigorous sheet lava flows and abundant pyroclasts (high effusion rate) from at least two collinear fissures at the southern end of the caldera (Chadwick et al., 2006, Embley et al., 1999, Fox et al., 2001). West et al. (2001) suggest that melts reside in the crust beneath Axial Seamount for hundreds to a few thousand years, but inflation since the 1998 eruption suggest that the magma chamber will be replenished in just ~22 years (Chadwick et al., 2006, Nooner & Chadwick, 2009).

Petrochemical and morphological variability, recorded in historical and older lava sequences, offer insight into timescales and pathways of melt generation and differentiation and replenishment in crustal reservoirs. A chronological framework for lava sequences is constructed using uranium-series model ages of the lavas and radiocarbon ages of coexisting benthic and planktic foraminifera at the sediment-lava contact. (We have requested NSF funding to supplement these efforts with geomagnetic paleointensity methods.) Our observation of >10 distinct, large aphyric, very lightly sedimented (and therefore <500a) lava flows within the caldera suggests an eruption recurrence of <50 a. Further, our seafloor observations and preliminary data suggest a significant petrogenetic change occurred beneath Axial Seamount within the last millennium. Eight dated lava flows from the summit caldera and flanks have eruption ages of 600-1200a, MgO ≥7.5 wt%, and are plagioclase-phyric. In contrast, younger lavas (undated but with thin sediment cover) from the same areas are aphyric with lower MgO, possibly reflecting enhanced crystal separation or longer magma residence times since ~600 a.

A higher degree of magma mixing (homogenization) in the magma reservoir beneath Axial Seamount is inferred from a limited range in trace element ratios (e.g. La/Yb; Chadwick et al. 2005 and our unpublished data) compared to lavas from adjacent segments, but modern radiogenic isotope data are too sparse to evaluate. Our  data confirm that Axial Seamount has the lowest (²³⁰Th)/(²³²Th) on the JdFR axis but (²³⁰Th)-excesses are similar to adjacent segments. These observations suggest that source compositions are more variable than melting processes on the southern JdFR. Like Axial Seamount under the influence of the Cobb hotspot, inflated crust at Endeavour segment is associated with the source of the nearby Heckle Seamount. However, there are fundamental unresolved differences between these two sections highlighted by (²³⁰Th)-excesses and geochemical variances at Endeavour that are much higher than elsewhere else on the JdFR.