GSA Annual Meeting in Phoenix, Arizona, USA - 2019

Paper No. 216-5
Presentation Time: 2:50 PM


FUSTON, Spencer, WU, Jonny and LIN, Yi-An, Department of Earth and Atmospheric Sciences, University of Houston, 3507 Cullen Blvd, Rm. 233, Houston, TX 77204-5008

Subduction of the now-vanished “Resurrection” plate, and two bounding mid-ocean ridges, has been proposed from two coeval belts of Paleocene to Miocene near-trench magmatism in Alaska and Oregon. We test the presence of this cryptic plate with a plate reconstruction since 50 Ma, constrained by unfolded (i.e. structurally restored) Farallon and Pacific-Kula slabs from four recent global and local mantle tomographies including MITP16 and UUP07. Our 3-D slab mapping reveals a steeply-dipping (>45°) Aleutian slab that is flattened across the uppermost lower mantle. East of Anchorage, this slab becomes shallowly-dipping (<45º). Below Vancouver Island, the Cascadia slab has a ~45° dip. This slab becomes shallowly-dipping to the south, where it extends to the longitude of Illinois and is flattened across the uppermost lower mantle. We also mapped localized high-velocity anomalies below western Yukon at ~500-600 km depth. Slab unfolding from six transects across Alaska and the Aleutian Islands reveals pre-subduction slab lengths of ~1,500 to 6,300 km (east to west). When reconstructed within a global plate model, the Aleutian slabs account for ~50 Ma of Pacific-Kula subduction west of Anchorage but only ~30 Ma to the east, forming an apparent gap in restorable oceanic lithosphere below southeast Alaska. Similarly, seven transects across the western United States reveal pre-subduction slab lengths of ~1,500 to 6,000 km (north to south) that account for only ~25 Ma of Farallon subduction below Vancouver Island (i.e. a second gap) and ~80 Ma near southern California. Intersection of the bounding edges of these apparent gaps with North America since ~50 Ma show close correspondence with the Coast Range Basalt Province (northwest USA), Sanak-Baranof belt (Alaska), and younger near-trench magmatism between the two. Thus, we suggest that our reconstructed gaps between the Kula and Farallon plates are slab windows that constrain the locations of the Kula-Resurrection and Resurrection-Farallon ridges since at least 50 Ma. The localized high-velocity tomographic anomalies below western Yukon fit within our plate reconstruction as a “Resurrection plate”. Our slab reconstruction provides new, independent spatiotemporal constraints on the lost Resurrection plate and NW Cordillera plate tectonics since early Eocene time.