Statistical analysis of slab-window related plutons: The Sanak-Baranof belt of southern Alaska

The Sanak-Baranof belt of southern Alaska is a chain of near-trench plutons that formed during the subduction of a spreading-ridge and the subsequent formation of a slab-window. Intrusions within this belt have a time-transgressive distribution that ranges from 61-50 Ma over its 2100 km length, and they are thought to track the movement of the ridge-trench-trench triple junction and the slab-window. We analyzed the spatial distribution, the orientation and the frequency-size statistics of 64 plutons within the Sanak-Baranof belt. These data indicate that intrusions within the Sanak-Baranof belt have a power-law frequency-size distribution. The power-law has the form N~A^{ -0.76} +/-0.15 which is quite similar to that calculated for other magmatic systems (Pelletier, 1999). We hypothesize that such a distribution is related to how magma propagates through the crust, which is similar for both the Sanak-Baranof belt and `normal' arc environments. A statistical analysis of the spatial relationships of the Sanak-Baranof belt plutons indicate that magmatism has a uniform nearest-neighbor distribution at the scale of individual plutons, but is clustered at scales larger than 100-200 km. One interpretation is that the spacing of magmatic centers is related to segmentation of the subducting ridge. However, between the western and eastern portions of the Sanak-Baranof belt there is a change in the spacing of magmatic centers. In the western portion, magmatic centers have a spacing of approximately 165 km, whereas in the east (within the Chugach metamorphic complex), magmatism is continuous at the scale of our analysis. We hypothesize that in the western portion of the belt the subducting spreading-ridge was oriented at an oblique angle (30°) to the trench, and the triple junction was migrating rapidly to the east at 19 cm/yr. In the eastern portion of the Sanak-Baranof belt, the trailing edge of the plate was parallel to the subduction zone causing the rate of spreading-ridge subduction to slow considerably. The decrease in subduction rate led to more pervasive magmatism and a higher regional metamorphic grade. This provides support for the existence and complete subduction of the Resurrection Plate by 50 Ma as hypothesized by Haeussler et al, 2002.