Parabolic distribution of circumeastern Snake River Plain seismicity and latest Quaternary faulting: Migratory pattern and association with the Yellowstone hotspot

Abstract

The Intermountain and Idaho seismic belts within Idaho, Wyoming, and Montana form an unusual parabolic pattern about the axis of the aseismic eastern Snake River Plain (SRP). This pattern is also reflected in the distribution of latest Quaternary normal faults. Several late Cenozoic normal faults that trend perpendicular to the axis of the eastern SRP extend from the aseismic region to the region of latest Quaternary faulting and seismicity. A study of the late Miocene to Holocene displacement history of one of these, the Grand Valley fault system in southeastern Idaho and western Wyoming, indicates that a locus of high displacement rates has migrated away from the eastern SRP to its present location in southern Star Valley in western Wyoming. In Swan Valley the studied area closest to the eastern SRP, isotopic ages, and paleomagnetic data for over 300 samples from 47 sites on well‐exposed late Cenozoic volcanic rocks (the tuff of Spring Creek, the tuff of Heise, the Huckleberry Ridge tuff, the Pine Creek Basalt, and an older tuff thought to be the tuff of Cosgrove Road) are used to demonstrate differences in the displacement rate on the Grand Valley fault over the last ∼10 m.y. Tectonic tilts for these volcanic rocks are estimated by comparing the results of paleomagnetic analyses in Swan Valley to similar analyses of samples from undeformed volcanic rocks outside of Swan Valley. Basin geometry and tilt axes are established using seismic reflection profiles and field mapping. Combining these data with the tilt data makes it possible to calculate displacement rates during discrete temporal intervals. An average displacement rate of ∼1.8 mm/yr is calculated for the Grand Valley fault in Swan Valley between 4.4 and 2.0 Ma. In the subsequent 2.0‐m.y. interval the rate dropped 2 orders of magnitude to ∼0.014 mm/yr; during the preceding 5.5‐m.y. interval the displacement rate is ∼0.15 mm/yr, or about 1 order of magnitude less than the rate between 4.4 and 2.0 Ma. Mapping of fault scarps and unfaulted deposits along the Grand Valley fault system shows that latest Quaternary fault scarps are restricted to the portion farthest from the eastern SRP, the southern part of the Star Valley fault. Surface displacements estimated from scarp profiles and deposit ages estimated from soil development suggest a latest Quaternary displacement rate of 0.6–1.2 mm/yr for the southern portion of the Star Valley fault. Morphologic evidence suggests that this displacement rate persisted on the Star Valley fault throughout most of the Quaternary. The latest Quaternary displacement rate calculated for the southern portion of the Star Valley fault is similar to the rate calculated for Swan Valley during the interval from 2.0 to 4.4 Ma. This similarity, together with evidence for a low Quaternary displacement rate on the fault system in Swan Valley, suggests that the location of the highest displacement rate has migrated away from the eastern SRP. Other normal faults in southeastern Idaho, northwestern Wyoming, and southwestern Montana, while less well described than the Grand Valley fault system, exhibit a similar outward migrating pattern of increased fault activity followed by quiescence. Furthermore, a temporal and spatial relationship between fault activity and the 3.5 cm/yr northeastward track of the Yellowstone hotspot is observable on the Grand Valley fault system and on other north‐northwest trending late Cenozoic faults that border the eastern SRP. The temporal and spatial relationship of Miocene to present high displacement rates for other circumeastern SRP faults and the observable outwardly migrating pattern of fault activity suggest that a similar parabolic distribution of seismicity and high displacement rates was symmetrically positioned about the former position of the hotspot. Moreover, the tandem migration of the hotspot and the parabolic distribution of increased fault activity and seismicity are closely followed by a parabolic‐shaped “collapse shadow,” or region of fault inactivity and aseismicity. We suggest that the outwardly migrating pattern of increased fault activity (active region) results from reduced integrated lithospheric strength caused by thermal effects of the hotspot. Conversely, the outwardly propagating quiescent region is the result of a reduction or “collapse” of crustal extension rates caused by increased integrated lithospheric strength. Lithospheric strength in this region is increased by addition of mafic materials at the base of the crust and at midcrustal levels. Although the strength of the mantle portion of the lithosphere is reduced, the increased strength of the crust results in a total integrated increase in lithospheric strength. Paradoxically, the surface heat flow data suggest that the region within the interior parabola has a higher heat flow (after accounting for the cooling effects of the eastern SRP aquifer) than the adjacent regions, yet the interior region exhibits significantly lower extension rates. It appears that in this region the surface heat flow is not a good predictor of rates of lithospheric extension.