The Fate of the Juan de Fuca Plate

Mei Xue and Richard M. Allen


The Juan de Fuca plate is subducting beneath the northwestern United States and southwestern Canada. While the slab has been imaged to depths of $\sim $300 km beneath southern Washington and of at least 200 km beneath southern Oregon, there is little evidence for a slab deeper than $\sim $100 km east of High Cascades beneath central Oregon (Bostock, et al., 2002; Harris, et al., 1991; Iyer and Rite, 1981; Michaelson and Weaver, 1986; Rasmussen and Humphreys, 1988; Rondenay, et al., 2001). The apparent absence of the slab east of High Cascades can be interpreted as (1) a low high-velocity contrast making the slab indistinct from the surrounding mantle (Iyer and Stewart, 1977; Michaelson and Weaver, 1986); (2) a more vertical geometry of the slab (Michaelson and Weaver, 1986), or (3) a loss of seismic resolution. To image the slab beneath Oregon, we apply tomography technique using a dataset consisting of our own OATS deployment and all other available data.

Teleseismic tomography results

Figure 27.1: Seismic stations used in this study with a total number of 61. Inset shows the distribution of the 95 events used in the inversion for S-wave velocity model.
\begin{figure}\epsfig{file=meixue06_1_1.eps, width=8cm}\end{figure}

We have collected data from our own deployment of the Oregon Array for Teleseismic Study (OATS), an array extending northwest-southeast across Oregon from the coast to the McDemitt Caldera (Figure 27.1). We have also collected data from permanent networks (BK, CC, US, UO, UW, PN, IU, TA, LI), and temporary deployments (XJ, YC, YS) (Figure 27.1). We inspect events with magnitude 6.0 and above from July 19th, 2003 to Nov. 11th, 2004 for a total of 61 stations and follow the same inversion procedure as in (Allen, et al., 2002). For the Vs inversion, a total of 95 events (Figure 27.1) with clear S and SKS phases were recorded at 45 stations, and a total number of 2148 rays were used. For the Vp inversion, a total of 74 events with clear direct P phase were recorded at 46 stations, and a total number of 2043 rays were used.

Figure 27.2: Tomographic results showing S- and P- wave velocity structures beneath Oregon and results of inverting synthetic data created from various test slabs. All vertical slices show the same cross section as in (f), along the OATS line, where we have high resolution: (a) the Vs model; (b) the Vp model; resolution tests of S-wave velocity inversion for a slab ending at (c) 400 km depth, (d) 300 km depth, and (e) 500 km depth, and the input anomaly is 3$\%$ and dips 50$^{\circ }$, outlined by red lines; (f) a map shows the location of the cross-section.
\begin{figure}\epsfig{file=meixue06_1_2.eps, width=8cm}\end{figure}

Our tomography results show that the subducted slab extends to a depth of $\sim $400 km with a dip of $\sim $44$^{\circ }$(Figure 27.2, a and b). After correcting the oblique trend of the cross section, the real dip angle is $\sim $50$^{\circ }$Resolution tests show that structures at shallower depths are better recovered in terms of amplitudes and smearing is less significant. Here, we show three resolution tests for S-wave velocity models (Figure 27.2, c, d, and e). To test whether the structure we observed between 300 km and 400 km depth is caused by smearing shallow structure or not, we used a synthetic velocity anomaly with a high velocity to 300 km depth only. Figure 2d shows that a slab ending at 300 km depth is not sufficient to produce the observed slab structure between 300 km and 400 km depth. Figure 2c shows a case that the recovered structure resembles what we observed best, where the input slab extends to 400 km and stops there. To test whether the slab stops at 400 km, we conducted the test as shown in Figure 2e where the input slab extends to 500 km depth. Considering the depth range between 400 km to 500 km, the recovered structure differs from the observed structure in three aspects: (1) it has a much stronger velocity anomaly, (2) the transition of velocity anomaly from higher amplitudes to lower amplitudes is smooth, and (3) it didn't lose its width (Figure 27.2e). Thus we speculate that the slab likely stops at $\sim $400 km depth and does not extend deeper. The P-wave velocity model also suggests that the slab stops at $\sim $400 km (Figure 27.2b). This interpretation can be better tested when USArray data becomes available further east.


Our tomographic images clearly show the Juan de Fuca plate diving into the mantle beneath Oregon and continues east of the High Cascades with a dip of $\sim $50$^{\circ }$reaching a depth of $\sim $400 km. The slab dips shallower compared with its counterparts to north and south, which have a dip of $\sim $65$^{\circ }$(Harris, et al., 1991; Rasmussen and Humphreys, 1988). Resolution tests suggest there is little or no velocity anomaly associated with a slab below $\sim $400 km.


This work was supported by the NSF (EAR-0539987). We thank Gene Humphreys for allowing us using data from their deployment at Wallowa Mountain. The IRIS DMC provided seismic data. The figures were produced with SAC and GMT (Wessel and Smith, 1995).


Allen, R. M., et al., Imaging the mantle beneath Iceland using integrated seismological techniques, in Journal of Geophysical Research-Solid Earth, 107, 2325, doi:2310.1029/2001JB000595, 2002.

Bostock, M. G., et al., An inverted continental Moho and serpentinization of the forearc mantle, in Nature, 417, 536-538, 2002.

Harris, R. A., et al., Imaging the Juan Defuca Plate beneath Southern Oregon Using Teleseismic P-Wave Residuals, in Journal of Geophysical Research-Solid Earth, 96, 19879-19889, 1991.

Iyer, H. M., and A. Rite, Teleseismic P-wave delays in Oregon Cascades, in EOS, Trans. Am. geophys. Un., 62, 1981.

Iyer, H. M., and R. M. Stewart, Teleseismic technique to locate magma in the crust and upper mantle, in Magma Genesis, in Proceedings of the AGU Chapman Conference on Partial Melting in the Upper Mantle, edited by H. J. B. Dick, pp. 281-299, Oreg. Dep. Geol. Min. Ind. Bull, 1977.

Michaelson, C. A., and C. S. Weaver, Upper Mantle Structure from Teleseismic P-Wave Arrivals in Washington and Northern Oregon, in Journal of Geophysical Research-Solid Earth and Planets, 91, 2077-2094,1986.

Rasmussen, J., and E. Humphreys, Tomographic Image of the Juan-Defuca-Plate beneath Washington and Western Oregon Using Teleseismic P-Wave Travel-Times, in Geophysical Research Letters, 15, 1417-1420, 1988.

Rondenay, S., et al., Multiparameter two-dimensional inversion of scattered teleseismic body waves 3. Application to the Cascadia 1993 data set, in Journal of Geophysical Research-Solid Earth, 106, 30795-30807, 2001.

Wessel, P., and W. H. F. Smith, New version of the Generic Mapping Tools released, EOS, in Trans. Am. geophys. Un., 76, 329, 1995.

Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail:
© 2006, The Regents of the University of California