Area of velocity anomalies

This figure shows velocity perturbations relative to a 1-D starting model on a pseudo-plane parallel to the seafloor at a depth of 3 km. The four plots below correspond to the regions boxed in blue. For each plot, the "footprint" of the velocity anomaly was measured at a range of depths. These values can be thought of as the area contained within contours exceeding the stated threshold. The size of each anomaly is plotted as a function of depths.

postscript version

The three plots above span (left) a 10km x 10km box around the caldera anomaly, (middle) a similar box around the east flank low velocity zone, and (right) a larger box which essentially sums the two regions together.

The very low velocity anomalies (< -1 km/s) peak at about 2.8 km below seafloor. This is the most likely center of the magma chamber. Below 3 kilometers, the area of these anomalies drops off. At depth, there is little of this material left. However, the region of "mild" low velocities (< -0.5 km/s) is quite large in the lower crust. There is no appearent bottom on this feature. One explanation for this is that transient or diffuse magma is sourced through the lower crust. It is never in great enough quantities to depress seismic velocities much more than 0.5 km/s however. In the mid-crustal magma chamber centered on 2.7 km, this melt collects. Pooling magma increases the partial melt fraction enough to create a siginificant seismic signature exceeding -1-2 km/s. The small amount of low velocity material in the upper crust may be remnant melt from recent eruptions or weakened rock from hydrothermal cracking. The latter interpretation is also supported by the lack of low velocities immediately above the magma chamber at a depth of 2.0 km. Fluid may not be able to penetrate this region due to its depth or perhaps the extreme heat.

postscript version The region spanned in this calculation is 12 km SSW of the caldera. The velocity contrast is not as great as in the caldera magma chamber. The size of the anomaly peaks at ~2.5 km below seafloor. Note the conspicuous lack of any low velocity signature below the LVZ. This feature may be an abandoned reservoir that is no longer being sourced from below. That is corroborated by the lack of large velocity perturbations (< -1.0 km/s) anywhere in the feature. It is also possible that this feature is not sourced from below but rather from the caldera magma body. In a rift situation it is easy to support the lateral transport of magma. Since this would likely occur via diking, as opposed to diffuse flow, one would not necessarily expect to observe a low velocity region connecting the two. Such a feeder tube is not observed here. The seafloor over this feature is about 500 meters deeper than at the caldera. So the 2.5 km depth of this anomaly is actually slightly deeper than the caldera magma chamber in absolute terms. When the 2 km of overlying water is considered however, hydrostatic pressure near this feature is essentially the same as in the central magma chamber. Ridge parallel diking resulting from crustal extension provides a plausible, though not proven, mechanism for lateral magma injection.

http://www.ldeo.columbia.edu/~mwest/AX/F/areas