Introduction

The Consortium for Continental Reflection Profiling (COCORP) collected data across the Death Valley domain in 1982 (Figure 1). Using conventional seismic processing Serpa et al. [1988] located upper- and mid-crustal reflectors. Geist and Brocher [1987] modeled a subset of first arrivals from lines 9 and 11 with iterative ray tracing to obtain shallow velocities, and provide a detailed lithologic interpretation. We extend their analyses using all the recorded data along lines 9, 10, and 11 (Figure 1), with a method that does not involve ray tracing and requires less a priori information. We also obtain a regional velocity model to the west of the highly extended domain (Figure 1) using surface-wave analysis, to constrain deeper velocities and complement Gibbs and Roller's [1966] velocity model from within the domain. We compare our results with regional models for the highly extended region as well as adjacent geologic provinces, and constrain the nature of extension beneath Death Valley.

Methods

An advantage of the simulated-annealing optimization is that it provides us with a suite of ``final'' models that can fit the data equally well. We can choose from this set one model that best fits our a priori knowledge about the region. The suite of models allows us to estimate the uncertainties associated with the optimized velocities. We trace rays through our velocity models and use the standard deviation over the suite to find out which parts of each model are best resolved, as in Pullammanappallil and Louie [1994].

The second part of our study uses surface-wave dispersion analysis of regional earthquake phases to constrain crustal velocities from 3 km to the Moho. Surface waves of aftershocks of the June 28, 1992, Landers earthquake at Deep Springs (DSP on Figure 1) traverse the Mojave block and west of Death Valley near the extended domain. We construct Rayleigh and Love wave group-velocity dispersion curves using the single-station ``peak and trough'' method [Brune et al., 1960]. An interactive modeling program we developed using the method of Takeuchi and Saito [1972] fits an observed dispersion curve to calculated curves by trial and error. Our method gives average velocity variations, but the depths of layers are not unique. In general, the fundamental-mode Rayleigh waves are more sensitive to the S-wave velocity variations at shallow depths, while the Love waves are affected more by velocities in the deeper horizons.

Results

The profile of line 10 extends from the southern part of the Black Mountains in the west to the southern end of the Nopah Range in the east (Figure 1). The Greenwater, Chicago, and Amargosa Valleys appear as near-surface 2.5 km/s low-velocity zones in the optimization result (Figure 2b). Higher velocities (~4 km/s) are associated with the pre-Neogene basement rocks beneath the Black Mountains and Nopah Range, gradually increasing to 5.8 km/s at 3.4 km depth. Ray tracing finds that resolution is good everywhere except under the Nopah Range.

Line 11 is an axial profile from southern Death Valley to central Death Valley in the north, entirely within the basin region (Figure 1). It indicates thickening of the 2 km/s low-velocity sediments toward the central Death Valley basin (Figure 2c). Velocities stay below 5.8 km/s until we reach depths of 3 to 4 km.

Velocities are similar where lines 9 and 11 intersect, but basin velocities are 0.5 km/s lower on line 11 (Figures 2a and 2c). The disagreement may result from the lateral smearing that occurs as we project crooked lines into straight velocity modeling sections. Line 9 smears in higher velocities from the basin walls, while line 11 smears in lower velocities from central Death Valley. Evidently first-arrival picks can constrain basin velocities here to only 0.5 km/s accuracy, although they strongly constrain basin-floor refraction velocities [Pullammanappallil and Louie, 1994].

The highest velocities we observe in the three profiles are about 5.8 km/s at 3 to 4 km depth. Ray tracing experiments show that any existing higher-velocity bodies between 2 and 4 km depth below the ranges would have been detected, as they would strongly refract energy to the surface. These experiments also indicate that velocity maxima obtained below the basins are well constrained. The COCORP lines do not transect any observed core complexes, and line 9 is well south of the Death Valley Turtlebacks (Figure 1; Holm et al., 1994).

Figure 3 summarizes the results of our surface-wave dispersion study. The computed dispersion curves and derived velocities reflect the cumulative effect of propagation in and near the highly extended Death Valley region, and through the less-extended Mojave region. Crustal velocities for the Mojave obtained by workers such as Hadley and Kanamori [1977] and Hearn and Clayton [1986] do not show any anomalously shallow high velocities there despite a complex tectonic history. Hence, any anomalies in the Death Valley region, if present, should appear in our results.

Our data require the presence of a low-velocity (V_P = 3 km/s) 3 km thick layer at the top (Figure 3e), probably representing the numerous alluvial basins in the paths of the seismic waves. This is underlain by a ~15 km thick layer having a velocity of about 6.2 km/s. Below this velocity increases to 6.7-6.9 km/s down to the Moho.

For further constraints, we compute group velocity curves from a refraction velocity model [Gibbs and Roller, 1966]. Their NTS-Ludlow profile directly crosses the region that has undergone extreme extension (Figure 1). Then, we compare all the models against a hypothetical model having high mid-crustal velocities of 6.8 km/s at shallow depths of 5 km. This discriminates the response of a typical crustal model from one that has anomalously shallow mid-crustal velocities and a radically thinned upper crust. Figures 3d and 3e show this difference for periods between 7 and 20 seconds. Our deeper surface-wave study does not suggest the presence of any abnormally high-velocity zones in the shallow crust.

Discussion

Collectively, these results seem to rule out extensional mechanisms that require a change in the regionally ``normal'' upper- and mid-crustal velocities. Our data provide a test of extensional hypotheses involving large vertical movements or velocity alterations of the shallow crust. Major uplift of the middle crust, or radical thinning of the upper crust may not be consistent with the agreement between our observations and velocities in much less-extended regions.

Our results constrain the conditions under which fluid layer based extension might operate in this region [Wernicke, 1992; Jones et al., 1992; Wernicke et al., 1996]. Under this hypothesis, the normal velocities we observe in any of these areas would require the compensating medium to be very close in density to the upper crust. This density constraint would imply that this ``layer'' is rich in quartz. We would, however, expect a ``fluid layer'' derived from the middle crust during the Cenozoic to have a significantly higher velocity than similarly quartz-rich upper crust, as it would contain a significant fraction of very high-velocity metamorphic minerals at Greenschist grade or higher.

Figure 2 shows that such high-velocity materials cannot be within 4 km of the surface near Death Valley, and the mismatch of the ``Hypothetical'' velocity model to the data in Figure 3 shows that any ``fluid layer'' having mid-crustal velocities is likely to be at least 12 km deep, whether inside or near Wernicke's [1992] zone of extreme extension. With these constraints on the minimum thickness of the upper crust, the ``fluid layer'' hypothesis may be incapable of explaining extreme amounts of extension.

From the pure shear point of view, our results also require only moderate extension within the upper crust, perhaps by normal faulting [Wright and Troxel, 1973]. If pure shear were the only extensional mechanism operating in the region, the amount of extension should be less than 100%. To achieve the extreme extension observed by some workers [Wernicke et al., 1988; Wernicke, 1992], pure shear would require additional processes to operate in the lower crust below 15 km. But the absence of major Moho deflection [Wernicke et al., 1996], or more extreme topographic depression in areas where the crust has been strongly denuded [Wernicke, 1985; Gans, 1987] compared with the adjacent areas not denuded, seem to preclude any significant thinning of the crust as a whole. These facts rule out extreme extension by a conventional pure shear mechanism, perhaps in favor of the more moderate extensions of Wright and Troxel [1973].

Another hypothesis generally falling within the pure shear category that might explain the observed extension is the addition of magmatic material to the lower crust [Okaya and Thompson, 1986; Gans, 1987; Serpa, 1990]. Our results, which do not indicate any velocity alterations in the upper 15 km, can be consistent with such a model. This constraint also holds if magmatic additions are made to a simple-shear model (e.g., Figure 6 of Jones et al., 1992). Our observations thus constrain magmatic additions to amounts again not consistent with extreme extension, as the upper crust has not been significantly thinned.

Acknowledgments

References

Brune, J. N., Nafe, J., and Oliver, J., A simplified method for the analysis and synthesis of dispersed wave trains, J. Geophys. Res., 65, 287-304, 1960.

Gans, P. B., An open-system, two-layer crustal stretching model for the eastern Great Basin, Tectonics, 6, 1-12, 1987.

Geist, E. L., and Brocher, T. M., Geometry and subsurface lithology of southern Death Valley basin, California, based on refraction analysis of multichannel seismic data, Geology, 15, 1159-1162, 1987.

Gibbs, J. F., and Roller, J. C., Crustal structure by seismic-refraction measurements between the NTS-Ludlow, U. S. Geol. Survey Prof. Paper 550D, D125-D131, 1966.

Hadley, D., and Kanamori, H., Seismic structure of the Transverse Ranges, California, Geol. Soc. Amer. Bull., 88, 1469-1478, 1977.

Hearn, T. M., and Clayton, R. W., Lateral velocity variations in southern California. I. Results for the upper crust from Pg waves, Bull. Seismol. Soc. Amer., 76, 495-509, 1986.

Holm, D. K., Fleck, R. J., Lux, D. R., The Death Valley Turtlebacks reinterpreted as Miocene-Pliocene folds of a major detachment surface, Jour. Geology, 102, 718, 1994.

Jennings, C. W., Geologic map of California, scale 1:750,000, Geol. Data Map 2, Calif. Div. Mines and Geol., 1977.

Jones, C. H., Wernicke, B. P., Lang Farmer, C., Douglas Walker, J., Coleman, D. S., McKenna, L. W., and Perry, F. V., Variations across and along a major continental rift: an interdisciplinary study of the Basin and Range Province, western USA, Tectonophysics, 213, 57-96, 1992.

McKenzie, D. P., Some remarks on the development of sedimentary basins, Earth Planet. Sci. Lett., 40, 25-32, 1978.

Miller, E. L., Gans, P. B., and Garing, J., The Snake Range decollement: An exhumed mid-Tertiary ductile-brittle transition, Tectonics, 2, 239-263, 1983.

Okaya, D. A., and Thompson, G. A., Involvement of deep crust in extension of Basin and Range province, Geol. Soc. Amer. Special Paper 208, Boulder, Colo., 15-22, 1986.

Pullammanappallil, S. K., and Louie, J. N., A generalized simulated-annealing optimization for inversion of first-arrival times, Bull. Seismol. Soc. Amer., 84, 1397-1409, 1994.

Serpa, L., de Voogd, B., Wright, L., Willemin, J., Oliver, J., Hauser, E., and Troxel, B., Structure of the central Death Valley pull-apart basin and vicinity from COCORP profiles in the southern Great Basin, Geol. Soc. Amer. Bull., 100, 1437-1450, 1988.

Serpa, L., Structural styles across an extensional orogen; results from the COCORP Mojave and Death Valley seismic transects, in Basin and Range Extensional Tectonics Near the Latitude of Las Vegas, Nevada, edited by B. P. Wernicke, pp. 335-344, Geol. Soc. Amer. Memoir 176, Boulder, Colo., 1990.

Stewart, J. H., Basin and Range structure: a system of horsts and grabens produced by deep-seated extension, Geol. Soc. Amer. Bull., 82, 1019-1044, 1971.

Stewart, J. H., Extensional tectonics in the Death Valley area, California: transport of the Panamint Range structural block 80 km northwestward, Geology, 11, 153-157, 1983.

Takeuchi, H., and Saito, M., Seismic surface waves, in Methods in Computational Physics 11, edited by B. A. Bolt, pp. 217-295, Academic Press, New York, 1972.

Vidale, J. E., Finite difference calculation of travel times, Bull. Seismol. Soc. Amer., 78, 2062-2076, 1988.

Wernicke, B., Uniform-sense normal simple shear continental lithosphere, Can. J. Earth Sci., 22, 108-125, 1985.

Wernicke, B., Cenozoic extensional tectonics of the U. S. Cordillera, in The Cordilleran Orogen: Conterminous U. S., edited by B. C. Burchfiel, M. L. Zoback and P. W. Lipman, pp. 553-582, The Geology of North America, G3, Geol. Soc. Amer., Boulder, Colo., 1992.

Wernicke, B., Axen, G. J., and Snow, J. K., Basin and range extensional tectonics at the latitude of Las Vegas, Nevada, Geol. Soc. Amer. Bull., 100, 1738-1757, 1988.

Wernicke, B., Clayton, R., Ducea, M., Jones, C. H., Park, S., Ruppert, S., Saleeby, J., Snow, J. K., Squires, L., Fliedner, M., Jiracek, G., Keller, R., Klemperer, S., Luetgert, J., Malin, P., Miller, K., Mooney, W., Oliver, H., and Phinney, R., Origin of high mountains in the continents, the southern Sierra Nevada, Science, 271, 190-193, 1996.

Wright, L. A., Late Cenozoic fault patterns and stress fields in the Great Basin and westward displacement of the Sierra Nevada block, Geology, 4, 489-494, 1976.

Wright, L. A., and Troxel, B. W., Shallow-fault interpretation of basin and range structure, in Gravity and Tectonics, edited by K. A. DeJong and R. Scholten, pp. 397-407, John Wiley, New York, 1973.