Download Volume II Accomplisments (28 Mb pdf). - IRIS

Slab Fragmentation and Edge Flow: Implications for the Origin of
the Yellowstone Hotspot Track
D.E. James (Carnegie Institution/DTM), M.J. Fouch (School of Earth and Space Exploration, Arizona State University), J.B.
Roth (Exxon/Mobil), R.W. Carlson (Carnegie Institution/DTM)
The Snake River Plain/Yellowstone (SRP/Y) volcanic complex is widely considered a classic example of a plume generated
continental hotspot (e.g. Smith et al., 2009). By that model, the plume head appeared ca 17 Ma near the Oregon-Idaho-Nevada
border with the outpouring of the Steens and Columbia River flood basalts. The SRP/Y hotspot is taken to be the product of
plume tail upwelling over the past 12 Ma. Our recent S-wave and P-wave tomographic images, however, instead suggest a subduction-related
process by which volcanism along the SRP/Y hotspot track results from slab fragmentation, trench retreat, and
mantle upwelling at the leading edge of the descending plate and around its truncated edges (Faccenna et al., 2010; K. Druken
and C. Kincaid, pers. comm.). Seismic images of the deeper upper mantle show that the subducted oceanic plate extends locally
eastward well into stable North America. The break-up and along-strike fragmentation of the descending plate is related in both
time and space to the onset of flood volcanism and the formation of the SRP/Y hotspot. A sub-horizontal branch of subducting
oceanic plate, orphaned from the descending plate by the northward migration of the Mendocino triple junction, resides in
the mantle transition zone (400-600 km) directly beneath the SRP/Y track (Figure 1a). Its truncated northern edge is parallel
to the northwestern margin of the hotspot track itself and marks the southern edge of a slab gap (Figure 1b). Numerical and
physical tank modeling show that a rapidly retreating and severely fragmented downgoing plate drives mantle flow around both
the tip and the edges of the descending slab. Our seismic results show that the morphology of the subducting slab is appropriate
for generating large-scale poloidal flow (flood volcanism) in the upper mantle during the re-initiation phase of slab descent
ca 20 Ma and for generating smaller-scale toroidal and poloidal upwellings (hotspot volcanism) around both the leading tip and
northern edge of the slab as it descends into the deeper upper mantle. Plate reconstructions are consistent with the timing and
position of both flood and hotspot volcanism.
Acknowledgements: This work is part of the High Lava Plains (HLP) Project, supported by Continental Dynamics NSF grants EAR-0507248 (MJF)
and EAR-0506914 (DEJ and RWC), as well as EAR-0548288 (MJF EarthScope CAREER grant). We thank the USArray team for a remarkable
effort and the superb efforts of the PIC staff and the HLP seismic team in the deployment and operation of the HLP seismic array. Data
were provided through the IRIS Data Management Center. We thank ranchers throughout the HLP region who hosted seismic stations. The
Eastern Oregon Agricultural Research Center provided critical logistical assistance for the High Lava Plains seismic team during this project.
SW
0
100
200
300
400
500
600
Remnant Slab
SRP/Y
NE
500
600
300
400
0
100
200
50˚N
48˚N
46˚N
44˚N
42˚N
500 km
124˚W 120˚W 116˚W 112˚W 108˚W
NWUS08z_S
50˚N
Depth = 500 km
Slab Gap
SRP-Y
48˚N
46˚N
44˚N
42˚N
Depth (km)
700
700
40˚N
40˚N
800
800
900
900
38˚N
38˚N
1000
1000
-1.0 -0.5 0.0 0.5 1.0
P-velocity Perturbation (%)
(a)
36˚N
-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
S-Velocity Perturbation (%)
(b)
36˚N
Figure 1. (a) P-velocity vertical cross-section parallel to plate convergence on axis of SRP/Y showing a sub-horizontal remnant of the Farallon slab adrift in the mantle
transition zone beneath the SRP/Y hotspot track, its leading edge just beneath Yellowstone; (b) S-velocity perturbations at 500 km depth a “slab gap” just north of
and parallel to the SRP/Y track itself. The slab gap extends from north of Yellowstone into eastern Oregon, its southern margin coinciding with the northern boundary
of the remnant slab in (a).
2010 IRIS Core Programs Proposal | Volume II | Upper Mantle Structure and Dynamics | II-219