Shallow plumbing systems for small-volume basaltic volcanoes

Eruptive dynamics in basaltic volcanoes are controlled, in part, by the conduit geometry. However, uncertainties in conduit shape and dike-to-conduit transition geometry have limited our predictive capability for hazards assessments. We characterize the subvolcanic geometry of small-volume basaltic volcanoes (magmatic volatile-driven eruptions, 0.1 to 0.5 km(3)) based on a synthesis of field studies of five basaltic volcanoes exposed to varying degrees by erosion and exhibiting feeder dikes, conduits, and vent areas <= 250 m depth. Study areas include East Grants Ridge (New Mexico, USA), Basalt Ridge, East Basalt Ridge, Paiute Ridge, and Southeast Crater Flat (Nevada, USA). Basaltic feeder dikes 250 to 100 m deep have typical widths of 4-12 m, with smooth host-rock contacts (rhyolite tuff). At depths less than 100 m, heterogeneities in the host rock form preferential pathways for small dike splays and sills, resulting in a 30-m effective width at 50 m depth. The development of a complex conduit at depths less than 70 m is reflected in bifurcating dikes and brecciation and incorporation of the country rock. The overall zone of effect at depths less than 50 m is <= 110 m wide (220 m elongated along the feeder dike). Based on comparisons with theoretical conduit flow models, the width of the feeder dike at depths from 250 to 500 m is expected to range from 1 to 10 m and is expected to decrease to about 1-2 m at depths greater than 500 m. The flaring shape of the observed feeder systems is similar to results of theoretical modeling using lithostatic pressure-balanced flow conditions. Sizes of observed conduits differ from modeled dimensions by up to a factor of 10 in the shallow subsurface (< 50 m depth), but at depths greater than 100 m the difference is a factor of 2 to 4. This difference is primarily due to the fact that observed eroded conduits record the superimposed effects of multiple eruptive events, while theoretical model results define dimensions necessary for a single, steady eruption phase. The complex details of magma-host rock interactions observed at the study areas (contact welding, brecciation, bifurcating dikes and sills, and stoping) represent the mechanisms by which the lithostatic pressure-balanced geometry is attained. The similarity in the normalized shapes of theoretical and observed conduits demonstrates the appropriateness of the pressure-balanced modeling approach, consistent with the conclusions of Wilson and Head (J Geophys Res 86:2971-3001, 1981) for this type of volcano.