We trace the Yellowstone hotspot track back to an apparent inception centered near the Oregon-Nevada border. We and others have concluded this is the locus of a starting plume or plume head. Consideration of this plume-head model leads us to discuss the following three implications.

(1) The apparent center of the relic plume head is about 250 km west of the location where both the trends of the younger hotspot track and the inferred plate motions would place the hotspot at 16 Ma. A possible explanation for this discrepancy is the westward deflection of the plume up the bottom of the inclined Vancouver slab. Plate tectonic reconstructions and an intermediate dip for the Vancouver slab indicate a plume head would have intersected the Vancouver slab.

(2) The postulated arrival of the plume head at the base of the lithosphere is temporally associated with the eruption of the Columbia River and Oregon Plateau flood basalts at 14-17 Ma; however, these basalts were erupted several hundred kilometers north of the apparent plume center. The postulated plume center is symmetrically located near the midpoint of the 1,100-km-long Nevada- Oregon rift zone. Strontium isotopic variations reflect crustal and mantle lithosphere variations along the trend of this rift zone, with the basalt area of Oregon and Washington lying west of the 0.704 line in oceanic crust, the apparent center in northern Nevada between the 0.704 and 0.706 line in intermediate crust, and the area of central and southern Nevada east of the 0.706 line in Precambrian continental crust. Geophysical modeling is consistent with a dense crust north of the Nevada-Oregon border and an asthenospheric low-density body that extends several hundred kilometers south and north of the Nevada-Oregon boundary. A reconstruction of the initial contact of the plume head with the lithosphere suggests relatively thin lithosphere at 17 Ma beneath Oregon and Washington, which would favor the spreading of the plume northward in this direction, more decompression melting in this “thinspot” area, and the eruption of basalt through dense, oceanic lithosphere. Thus, preferential extrusion of flood basalts north of the plume center may be the result of differences in the pre-plume lithosphere, and not the location of the center of the plume head.

(3) A plume head rising into the base of the lithosphere is expected to produce uplift, which we estimate to be about 1 km with a north-south dimension of 1,000 km. This plume-head uplift, followed by subsidence, is consistent with Cenozoic paleobotanical altitude estimates. Other climatic indicators show major aridity about 15 Ma in areas in the inferred precipitation shadow east of the inferred uplift. Indicators of climate about 7 Ma are compatible with an eastward migration of uplift to a site between the plume-head area and the present Yellowstone crescent of high terrain. The warm Neogene “climate optimum” correlates with 14- to 17-Ma flood basalt and rhyolite volcanism. The continued effects of Yellowstone plume-head uplift and ensuing plume-tail uplift, if real, could provide regional uplift that is geophysically plausible. Climatic modeling has shown that uplift of the age and latitude of the postulated Yellowstone plume-head uplift, if allied with Himalayan and perhaps other uplifts, could result in the late Cenozoic cooling that lead to the Pliocene-Pleistocene ice ages (Kutzbach and others, 1989; Ruddiman and others, 1989, 1997).

Thus, the postulated Yellowstone plume head could have played an important role in the late Cenozoic geologic history of the northern, interior part of the U.S. Cordillera. Future studies of the kind briefly discussed here should provide a better evaluation of the Yellowstone plume head concept.

Key words: hotspot, Yellowstone, plume head, starting plume, Columbia River basalt, mantle, Basin and Range