Shallow Geophysical Survey Across the Pahrump Valley Fault Zone,
California-Nevada Border

Gordon Shields, Kip Allander, Russell Brigham, Ryan Crosbie, Lorenzo Trimble,
Mike Sleeman, Richard Tucker, Hongbin Zhan, and John N Louie
Seismological Laboratory (174), Mackay School of Mines, University of Nevada, Reno

Published in BSSA, v. 88 (April 1997), p. 270-275.

Abstract

We employ seismic reflection; magnetics; and electromagnetics to precisely locate the southern extension of the Pahrump Valley Fault Zone (PVFZ) and examine its subsurface geometry. The southern extension of the PVFZ is active and represents a potential seismic hazard for Las Vegas. We observe pronounced magnetic and conductivity anomalies, and truncations of reflectors in the seismic profiles coincident with one of three Holocene scarps in southern Pahrump Valley. These geophysical techniques in combination can locate faults more precisely than the presence of eroded scarps.

Introduction

The Pahrump Valley Fault Zone (PVFZ) lies within Pahrump Valley, between the Spring Mountains of Nevada and the Kingston and Nopah Ranges of California, and follows the California-Nevada state line (Figure 1). It forms part of the broad zone of northwest-trending right-lateral shear known as the Walker Lane belt (Stewart, 1988). One of the first studies to mention the potential for right-lateral displacement on this fault zone is Stewart et al. (1968), in which an estimate of the total offset is given as 16 to 19 km on a northern segment of this fault in Stewart Valley. In the vicinity of our study area, evidence was first reported for the trace of the PVFZ in satellite images, but no clear indications of the amount of offset were found (Burchfiel et al., 1983; Stewart and Crowell, 1992). In a recent Master's thesis, Hoffard (1991) mapped in detail the faults of Pahrump Valley, including the PVFZ. Although Hoffard's study area ends just north of ours, she reports the age of the youngest geomorphic surface displaced by the fault is mid to late holocene, and that it therefore qualifies as potentially active. In the absence of detailed mapping in our study area, this segment of the fault is inferred to be a southern extension of the PVFZ that connects with the Stateline fault at the southern end of Mesquite Valley (Hoffard, 1991; MIT Field Geophysics Course, 1985), nearly 50 km to the southeast. In our study, completed in the spring of 1994, we use the results of seismic reflection, magnetic, and electromagnetic (EM) surveys to examine the geometry and activity of this southern extension of the PVFZ.

Figure 1: Location map showing the study area in Pahrump Valley and environs: PV, Pahrump Valley; PVFZ, Pahrump Valley Fault Zone; SV, Stewart Valley; MV, Mesquite Valley; AM, Ash Meadows; SM, Spring Mountains; NR, Nopah Range; KR, Kingston Range; OSTH, Old Spanish Trail Highway. Our surveys follow the Old Spanish Trail Highway, which runs between Tecopa, California and Nevada highway 160. The PVFZ trends NW essentially coincident with the CA-NV state line.
(click on figure for a full-size image)

Our geophysical surveys follow the Old Spanish Trail Highway, which runs between Tecopa, California and Nevada highway 160. A set of west facing, northwest trending scarps cross Southern Pahrump Valley nearly perpendicular to the highway, in the vicinity of the California-Nevada border. The map of Wright et al. (1981) indicates the fault is concealed where it crosses the Old Spanish Trail Highway nearly 3 km east of the state line, somewhat at odds with our reconnaissance. Elevation changes on the order of 10 m over a horizontal distance of 200 m accompany each of the three lineaments along this segment of the PVFZ. The total elevation gain over the 5 km long stretch of road that crosses all three scarps is about 70 m. The first goal of our study is to definitively identify these scarps as tectonic, and therefore part of the PVFZ. Upon finding clear geophysical signatures of a fault, an additional goal became discerning the strike and sense of offset of these features, or at least constraining these parameters within certain bounds. Such information will be important for a future examination of the seismic potential of the PVFZ, which holds importance for the seismic hazard to Las Vegas (Wyman et al. 1993).

Methods

Our magnetic and gravity surveys completely crossed Southern Pahrump Valley; however, we report here the results of high resolution seismic, magnetic and EM data collected at the PVFZ scarps in the center of the valley. Measurement points along the roadside were surveyed with an EDM theodelite, providing precise locations for all the geophysical surveys.

Shallow Conductivity Survey
Magnetic Survey
Our magnetic survey traversed the entire valley with 100 m station spacing, but at the westernmost PVFZ scarp spacing was decreased to $\approx$30 m. Across the second and third scarps station spacing was $\approx$60 m. In all 70+ total-field measurements were taken in a 3.5 km span across the PVFZ scarps using a Scintrex proton-precession magnetometer, corrected for drift by subtracting measurements from an automated base station. Repeat measurements were made at the western-most scarp to confirm the reproducibility of the anomaly.

Shallow Seismic Reflection Survey
The seismic reflection profiles cover only the two western-most scarps. Using an off-end spread geometry with a 5 m source interval, the two profiles extend 450 m and 300 m. A 5 kg sledgehammer provided the source, stacked 12 times for each source location. Each record contains 12 channels, for a maximum offset of 60 m with 5 m geophone spacing. 500 ms of data were recorded with a Bison 9000 digital recorder from single vertical component 100 Hz geophones, though sufficient signal for interpretation occurred in only the first 250 ms. This corresponds to a maximum reflection depth of 300 m; however, high frequency airwave noise obscures some of the later arriving energy. To process the data, we used the Common Depth Point (CDP) velocity analysis and stacking procedure, followed by migration in 100 m wide data sections (Telford et al., 1990)

Results

The data from all three surveys clearly contain anomalies in the vicinity of the scarps. However, because they each image different aspects of the structure, in combination the three types of data are much more convincing than individually. At scarp 1 (the western-most scarp), both the EM and magnetic data contain significant, essentially coincident anomalies relative to the rest of the data we collected (Figure 2). The EM31 data do not lend themselves to modeling; however, we know they represent the average apparent conductivity between the instrument and the maximum sensitivity depth. In this case we can say that in the vicinity of scarp 1 the conductivity increases 2 to 3 times over the background conductivity to a depth of at least 3 m. The total width of the main data anomaly here is close to 100 m, and considering the maximum penetration depth is only 3 m, this indicates the near-surface conductivity anomaly is also this wide. It should also be noted that the anomaly peaks in three other offset profiles (not shown) across scarp 1 have values of 120, 110 and 85 mS/m, significantly larger than the 60 mS/m anomaly along the continuous profile. At the second and third scarps, the conductivity data also show anomalies, though not as significant and convincing as at scarp 1.

Figure 2: Plot of magnetic and conductivity profiles with scarp topography. The scarps are numbered for reference in the text. Large anomalies exist at scarp 1 for both data types. The magnetic data show a regional increase to the east toward the Spring Mountains. The entire valley-wide magnetic profile is plotted below the scarp data to show the absence of magnetic anomalies anywhere other than at the PVFZ.
(click on figure for a full-size image)

The magnetic data contain a distinct anomaly at scarp 1 (Figure 2); however, no clear anomaly appears at either of the other two scarps. What makes the magnetic anomaly at scarp 1 quite convincing is the lack of any other such anomaly in the entire valley-wide profile (Figure 2) except for slight perturbations over the other two scarps. Also evident is the trend of increasing magnetic field to the northeast towards the Spring Mountains, which is a regional magnetic high on Saltus and Ponce's (1988) magnetic map. Unlike the EM data, the magnetic data potentially represent features much deeper than a few meters.

Figure 3 shows attempts at forward modeling the magnetic data from scarp 1, using a magnetic susceptibility of 0.001 e.m.u., which is at the low end of the range of values for common magnetic minerals. Of the two most likely sources of the anomaly, the steeply dipping prism (representing a mineralized fault plane) fits the the overall shape of the data anomaly better than the offset stratigraphic layer. The general level of misfit in both models is due to the regional magnetic trend, which was not removed from the data. The depths of these features are not precisely constrained, but the anomaly width gives a reasonable approximation. Based on the straight-slope rule of thumb (Milsom, 1989, p.58), the depth to the top of the magnetized body is approximately 100 m, which is on the order of 50 times deeper than the measured conductivity anomalies.

Figure 3: Two attempts at forward modeling the magnetic data at scarp 1 for the most likely fault-related structures: a vertical prism representing a mineralized fault plane (top); and an offset stratigraphic layer (bottom). A magnetic susceptibility contrast of 0.001 e.m.u. was used for the magnetic structures. Though both models are offset from the data due to the regional magnetic high centered under the Spring Mountains, the vertical prism more closely matches the shape of the anomaly and is the preferred model, representing magnetic mineralization of a steeply-dipping fault plane.
(click on figure for a full-size image)

Figure 4 shows the seismic sections across scarps 1 and 2. The record for scarp 1 contains a strong reflector at 40 ms (two-way travel time, corresponding to $\approx$26 m depth), continuous across the whole stack except for a slump-like feature near the middle of the profile. The section for scarp 2 shows a strong reflector at 75 ms ($\approx$50 m) in the middle of the profile, truncated at both ends. Air waves cause the high frequency reverberating noise in center of both sections that obscure possible reflectors there. The velocities used to estimate depths from the time sections come from our velocity analysis during stacking. Since the velocities are not known exactly, the estimated depths of these features may be in error up to $\pm$1 m for every 10 ms in the time section. Also the resolution is limited by the velocity and the dominant frequency; at 50 ms the horizontal resolution is about 13 m, at 150 ms the resolution is only about 23 m.

Figure 4: Stacked and migrated seismic sections including elevation static corrections for scarps 1 and 2, showing suggestions of truncated layers, though no absolute indications of vertical offset. Plotted above each section is the topography across the scarp, whose vertical scale does not apply to the section. For an approximate depth scale in the seismic sections: 40 ms $\approx$ 25 m; 75 ms $\approx$ 50 m; this corresponds to a vertical exageration of about 1.3. We believe the strong reflector that appears truncated or slumped below scarp 1 may be an ash bed that outcrops near scarp 2. The width of the disrupted zone at scarp 1 of $\approx$ 100 m matches the width implied in the conductivity data, and suggests significant lateral offset.
(click on figure for a full-size image)

Discussion

As noted in the introduction, several studies have categorized the PVFZ as right-lateral strike-slip based on satellite air-photo reconnaissance, and mapping of the fault zone north of our study area towards and in Stewart Valley (Stewart et al., 1968; Wright et al., 1981; Burchfiel et al., 1983; Hoffard, 1991; Stewart and Crowell, 1992). Additionally, the trace of the PVFZ can be followed southward to the State Line fault and northward through Stewart Valley to active faults in Ash Meadows and the southern Amargosa Desert (Hoffard, 1991), for a total length of over 100 km. More than 15 km of displacement has been proposed for this right-lateral fault system (R.L. Christiansen, written comm., in Sterwart et al., 1968; R.A. Schweickert, personal comm., in Hoffard, 1991), presumably continuous through Pahrump Valley. Because our geophysical anomalies align so closely with scarp 1, and given the suggestions of anomalies at the other scarps, we feel that we have recorded the geophysical signature of the right-lateral PVFZ. The conductivity and magnetic anomalies at the scarps further suggest that the scarps have not significantly eroded back from the fault traces and exist presently at the strands of the PVFZ. An east to west younging of the scarps based on their topographic morphology also seems consistent with our anomaly data. The western most scarp is the most recently active of these scarps as evidenced by the sharpest surface expression, and is the site of the only convincing conductivity and magnetic anomalies. We have therefore found a correlation between fault location, recency of activity, and anomaly location.

Several studies have documented conductivity and magnetic susceptibility increases across faults (Eberhart-Philips et al., 1995; Jones-Cecil, 1995; Wang, 1984). In particular, Wang (1984) presents a comprehensive study of the geophysical and geochemical properties of the central San Andreas and discusses the wide variety of chemical processes that can occur in saturated fault zones. However, very few minerals have the high magnetic susceptibilities suggested by the magnetic modeling at scarp 1; the three with the highest susceptibilities are magnetite, pyrrotite and illmenite (Telford et al., 1990; Milsom, 1989). These magnetic minerals could also cause the observed conductivity anomalies, if present in the upper few meters. For the source of such minerals, widespread hot spring mineralization has been documented on a section of the PVFZ beginning about 15 km north of our study area (Hoffard 1991), though no presently active hot springs exist along the PVFZ due to increasing groundwater withdrawals in the last decade. We believe that sufficient magnetic and conductive minerals remain in the fault zone at our study area from past hot spring activity to cause the observed anomalies.

However, as noted by Eberhart-Philips et al. (1995), electromagnetic methods cannot easily distinguish between conductive minerals in a fault from fluids or clays. Therefore, a second possible source of the shallow conductivity anomalies is differences in water content between the fault and the surrounding sediments. Even if increased fluid content is the cause of the shallow conductivity anomaly at the scarp, it still implies some subsurface disruption to admit additional water in the vicinity of the scarp. In either case, the conductivity anomaly suggests that significant fracturing of the near-surface layers occurs over a zone approximately 100 m wide.

One potentially meaningful observation is that the peak of the magnetic anomaly is offset about 70$\pm$40 m (40 m is the sum of the magnetic and conductivity survey spacing) to the west of the peak of the conductivity anomaly (Figure 2). The scarp is west facing, consistent with a down-to-the-west component of motion. Because the magnetic data correspond to deeper features than the conductivity data, the offset between the anomaly peaks may be due to a westward dip of the fault plane. Also, the offset conductivity profiles (not shown) give an indication of the strike of the conductive feature. The scarps appear to locally trend about N60W, more westerly than the N45W trend of the PVFZ averaged over its length. The offset conductivity profiles support this more westerly local trend over the 100 m along-fault distance of the four conductivity profiles.

Because the seismic data more directly represent the physical structure of the fault than either the magnetic of conductivity data, the seismic data are potentially much more convincing. The strong reflections in the stacked seismic sections may be from ash beds similar to those we see cropping out in the scarps. We might expect any vertical offsets of reflectors in the seismic data to be similar to the surface offsets, but growth faulting or lateral offset of even slightly dipping structures could cause significant discrepancies in either direction between surface expression and subsurface vertical offsets. At scarp 1, the strong reflector at 40 ms is continuous through the profile except at the base of the scarp where it is seemingly broken up or possibly slumped downward. This 100 m wide zone of missing reflector is a likely candidate for the location of the main trace of the fault at a depth of about 26 m.

The offset of the reflector in the section implies less than 10 m of vertical displacement, but offset could be greater at depth in the case of growth faulting, or large lateral offsets could be masking the true vertical offest. Interestingly, though the widths of the conductivity anomaly and the disrupted reflector at scarp 1 are both about 100 m, they are offest from one another. The deeper seismic data show the disrupted layer west of of the conductivity anomaly. As with the deeper magnetic data, perhaps this westward offset is due to a westward dip of the fault plane.

At scarp 2, the seismic section also lacks clear evidence of vertical displacement, but a fairly continuous 160 m long reflector is seemingly truncated at both ends. This seismic section at scarp 2 would likely have been more useful had it been extended westward further toward the base of this degraded scarp. However, the apparent truncations offer some evidence for disruption of subsurface layers.

We have verified that significant structural features exist at the PVFZ scarps. However, because motion on the PVFZ is thought to be dominantly right-lateral, the gross characteristics of the fault zone may be the only means of addressing the total offset. Of course there is no precise relation between fault zone width or localized fault morphology and cumulative lateral offset, but we may be able to decide if the proposed 20$\pm$5 km of offset is consistent with the geophysical data we have collected. At scarp 1 the conductivity data and the seismic data both imply a 100 m wide zone of fracturing. For scarp 2, both data types suggest an even wider zone of disruption, perhaps double the width at scarp 1. Shedlock et al. (1990) noted widths of disrupted zones on the central San Andreas on the order of 100 m. This fact suggests we cannot rule out the proposed 15 km of right-lateral offset.

Conclusion

We have verified that the scarps along the southern extension of the PVFZ originated through faulting, using a combination of seismic reflection, magnetic, and electromagnetic data. The three techniques in combination provide convincing evidence for a precise fault location. Our surveys located the zone of disruption underlying the westernmost scarp, possibly showing a graben in the hanging wall. Additionally, the 100+ m width of the zone of disruption present in both the conductivity and seismic anomalies can be used to infer a significant amount of lateral offset. Our results highlight the potential utility of these geophysical techniques for locating faults in areas where the scarps are poorly preserved. The 50+ km length of the PVFZ, which increases to $\approx$100 km with the inclusion of the State Line fault to the south, potentially presents a serious seismic hazard to the Las Vegas metropolitan area. Located just 50 km west of Las Vegas at its closest reach, the full length of the PVFZ could be capable of a M$_{W}$ 7+ event.

Acknowledgements

This work was completed under grant EAR 9405534 from the National Science Foundation. Dr. Ken Taylor of the Desert Research Institute, Reno, generously provided the seismic and EM field instruments, and Dr. Robert Karlin of UNR provided the potential-field instruments. The S.F. Hunt fund of the University of Nevada, Reno, kindly supplied transportation expenses. Thanks to Mark Stirling and Craig Jones for helpful reviews.

References

Anderson, J. G., S. G. Wesnousky, and M. W. Stirling (1996). Earthquake size as a function of fault slip rate, Bull. Seism. Soc. Am. bf 86, 683-690.

Burchfiel, B. C., G. S. Hamill IV, and D. E. Wilhelms (1983) Structural geology of the Montgomery Mountains and the northern half of the Nopah and Resting Spring ranges, Nevada and California, Geol. Soc. Am. Bull. 94, 1359-1376.

Eberhart-Philips, D., W. D. Stanley, B. D. Rodriguez, and W. J. Lutter (1995). Surface seismic and electrical methods to detect fluids related to faulting, J. Geophys. Res. 100, 12,919-12,936.

Hoffard, J. L. (1991). Quaternary tectonics and basin history of Pahrump and Stewart Valleys, Nevada and California, M.S. thesis University of Nevada, Reno, 138 p.

Jones-Cecil, M (1995) Structural controls of holocene reactivation of the Meers fault, southwestern Oklahoma, from magnetic studies, Geol. Soc. Amer. Bull. 107, 98-112.

Milsom, J. (1989). Field Geophysics Halsted Press, New York.

MIT Field Geophysics Course (1985). A geophysical study of Mesquite Valley: Nevada-California border, J. Geopys. Res. 90, 8685-8689.

Saltus, R. W. and D. A. Ponce (1988). Aeromagnetic map of Nevada; Las Vegas Sheet, Nevada Bureau of Mines and Geology Map 95.

Shedlock, K. M., T. M. Broucher, and S. T. Harding (1990). Shallow structure and deformation along the San Andreas Fault in Cholame Valley, California, based on high resolution reflection profiling, J. Geophys Res. 95, 5003-5020.

Stewart, J. H. (1988). Tectonics of the Walker Lane Belt, Western Great Basin: Mesozoic and Cenezoic deformation in a zone of shear, in Metamorphism and Crustal Evolution of the Western U.S., Ruby volume VII W. G. Ernst (Editor), Prentice-Hall, New Jersey, 683-713.

Stewart. J. H., J. P. Albers, and F.G. Poole (1968). Summary of regional evidence for right-lateral displacement in the western Great Basin, Geol. Soc. Am. Bull. 79, 1407-1414.

Stewart, J. H., and J. C. Crowell (1992) Strike-slip tectonics in the Cordilleran region, western United States, in GSA Mono. The Cordilleran Orogen; conterminous U. S., v. G-3 B. C. Burchfiel, P. W. Lipman, and M. L. Zoback (Editors), Geological Society of America, Boulder, Colorado, 609-628.

Telford, W. M., L. P. Geldart, and R. E. Sheriff (1990). Applied Geophysics Cambridge U.P., New York.

Wang, C.-Y. (1984). On the Constitution of the San Andreas Fault Zone in Central California, J. Geophys. Res. 89, 5858-5866.

Wells, D. L., and K. J. Coppersmith (1994). New empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface displacement, Bull. Seism. Soc. Am. 84, 974-1002.

Wernicke, B. (1992). Cenezoic extensional tectonics of the U. S. Cordillera, in GSA Mono. The Cordilleran Orogen; conterminous U. S., v. G-3 B. C. Burchfiel, P. W. Lipman, and M. L. Zoback (Editors), Geological Society of America, Boulder, Colorado, 553-581.

Wright, L. A., B. W. Troxel, B. C. Burchfiel, R. H. Chapman, and T. C. Labotka (1981). Geologic cross section from the Sierra Nevada to the Las Vegas Valley, eastern California to Southern Nevada, Geol. Soc. Amer. Map Chart Ser., MC-28M.

Wyman, R. W., M. Karakouzian, V. Bax-Valentine, D. B. Slemmons, L. Peterson, and S. Palmer (1993). Geology of Las Vegas, Nevada, United States of America, Bull. Assoc. Eng. Geol. 30, 33-78.

Shallow Geophysical Survey Across the Pahrump Valley Fault Zone, California-Nevada Border