SPREADING-RIDGE SUBDUCTION AND THE KINEMATIC AND MAGMATIC EVOLUTION OF THE SANAK-BARANOF BELT, SOUTHERN ALASKA

The forearc Sanak-Baranof plutonic belt of southern Alaska is thought to have resulted from the subduction of a spreading-ridge and subsequent slab-window. The initiation of magmatism in the belt decreases in age from 61-62 Ma near Sanak Island (west) to 51-50 Ma on Baranof Island (east) at a nearly constant rate of 19 cm / year. However, between the western and the eastern Sanak-Baranof belt there exist variations in pluton spacing, geochronology, geochemistry and metamorphism. The average spacing of magmatic centers changes from 165± 88 km in the west to 67±41 km in the east. The duration of magmatism (U/Pb zircon and monazite ages) increases from 1-2 Ma in the west to 4-5 Ma in the east, and the duration of Ar-Ar biotite-cooling ages increases to the east. In agreement with extended cooling times, large regions of greenschist and amphibolite facies metamorphism are present in the eastern segments, whereas outside of narrow pluton aureoles high temperature metamorphism is not present in the western Sanak-Baranof belt. Most Sanak-Baranof plutons are interpreted to have formed from a mixture of mantle-derived basalt and partial melts of accretionary prism sediments. However, intrusions with adakitic characteristics such as high Sr/Y ratios are present in the east, but not in the west. This indicates that some degree of slab melting occurred in the eastern Sanak-Baranof belt. It is suggested that the above east-west variations can be explained by the oblique subduction of a segmented spreading-ridge along a curved continental margin.

Geochronology, pluton spatial relationships and plate tectonic reconstructions suggest that the angle between the ridge and the trench decreased from 30-40° in the west near Kodiak Island to almost 0° at Baranof Island. These same data suggest eastward decreasing rates of outboard plate convergence from 10 cm / yr. near Kodiak, to eventually 0 cm / yr. in southeast Alaska indicating a cessation of spreading-ridge subduction. Such kinematic variations would produce slab-windows with different characteristics. In the western Sanak-Baranof belt they would produce a small rapidly migrating slab-window, whereas in southeast Alaska the above values would result in a large stationary slab-window caused by the subduction of the trailing edge of down going plate.