These awards involve an investigation of the causes and consequences of intraplate volcanism in the Tertiary-Recent magmatic province in eastern Oregon. The project will focus on four main questions related to generating the intraplate magmatism: 1) Is a plume required? ; 2) What role does subduction play? ; 3) What role does lithospheric structure play in generating/controlling magmatism and how has magmatism affected lithospheric structure? ; and 4) What role does continental crust play in magmatism and how is continental crust modified by magmatism? To answer these questions the PI's will acquire passive (James and Fouch) and active (Keller) seismic data, incorporating data from the Bigfoot array (US Array), new geochronological and geochemical data (Carlson, Duncan, Grunder, Hart) and experimental petrological investigations (Grove) on targeted magmatic products and undertake numerical and laboratory dynamical modeling.
The theory of plate tectonics well explains why geologic activity (earthquakes, volcanos) is clustered near plate boundaries. What it does not well explain is geologically active areas removed from plate boundaries. One excellent example is the Pacific Northwest of the United States. Here, subduction of the Pacific Ocean plate beneath North America creates large earthquakes along the fault where the oceanic plate plunges into the mantle. Water carried into Earth's interior by the subducting oceanic plate lowers the melting point of the rocks in Earth's interior, instigating the volcanism that forms the Cascades. This is the expected plate boundary activity. Over the last 100 million years, however, the volcanism and earthquakes have not been confined just to the Cascades, but includes very large volume volcanism up to over 1000 km east of the plate boundary. In the last 20 million years, this activity includes the Columbia River flood basalts (about 250,000 km3 of lava), the volcanic trace of the Snake River Plain, ending in currently active Yellowstone, and the abundant, young and still active, volcanism in eastern Oregon that forms the High Lava Plains (HLP). This study combined seismic imaging and modeling of flow patterns in the rocks beneath this area with various geologic, chronologic, and geochemical studies of the volcanic rocks erupted onto the overlying crust. The seismic results imaged the subducting oceanic plate to about 500 km depth, but then found it to break up into a number of fragments that are scattered around the upper mantle beneath the Pacific Northwest. Numeric and tank modeling shows that the downgoing oceanic plate induces a number of flow patterns that bring surrounding material rapidly towards the subduction zone at the plate boundary, then down with the subducting plate. The flow pattern suggested by the modeling was confirmed by seismic techniques sensitive to the direction of movement in Earth's interior. This background induced flow brings deeper, hotter, material towards the surface where it can undergo melting to produce the lavas that erupt in eastern Oregon. Both the seismic imaging and geochemical methods to define the depth of origin of the magmas point to a very shallow (50-100 km deep) melt zone beneath the HLP consistent with the idea that this wide spread volcanism primarily results from shallow mantle flow induced by the subducting oceanic plate. In contrast, the melt zones beneath the Snake River Plain/Yellowstone (SRP/Y) trace continue to depths of at least 150 km and perhaps as deep as 200 km. At depths of 400-500 km, however, we imaged a stranded fragment of an older piece of the subducted oceanic plate, whose northern edge lies directly below the northern edge of the SRP. Beneath the fragment of the oceanic plate, and directly beneath Yellowstone, we image a conduit of hot mantle that extends to at least 1000 km depth at which point our techniques can follow it no deeper, though it may continue much deeper in the mantle. These results are consistent with the idea that the SRP/Y volcanism is caused by uprise of a conduit of very deep, hot, mantle that melts to form a hot "spot" of volcanism. The "spot" becomes a line, like the SRP, because North America has moved to the southwest over a fixed hotspot in the mantle over the last 12 million years. The stranded piece of oceanic plate, which because it is cold and stiff, interferes with the rise of the plume and causes it to spread out into a sheet and then flow around the oceanic plate on the north. This complication to the traditional hot-spot model explains many features of SRP/Y volcanism that previously had been used to argue against a deep mantle source. The results of the project show that while geologic activity is concentrated at plate boundaries, the consequences of a continent overriding an oceanic plate can extend will into the continental interior as a result of the flow patterns instigated by the sinking plate. In the western US, this activity is enhanced by the presence of a deep mantle plume rising to produce the volcanism of the SRP/Y trace.
