Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1 Introduction The debate over the existence and/or the type of crustal root underlying the Sierra Nevada Mountains began during the 1950s. Some guy proposed that there ought to be a deep root based on elementary gravity and compensation calculations. These calculations are plausible considering given know data today about other mountain ranges throughout the world, but for the Sierra Nevadas, the story is much more complicated. In 1973, Carder proposed that in fact there is no root based on teleseimic arrivals over the seimic network. His paper is still cited by some authors today. However, the truth is with the unrealistic velocity model used in his calculations and that the data was unreversed, his model cannot be verified. In 1980, Pakiser and Brune also thought this, and wrote a analytical paper on the travel times through the Sierras and caluclating several arrival times and only then choosing the best model. Their results, a 50 km root in beneath the Sierra Nevadas in the northern and southern Sierras. Since the initial bodies of work, the debate continues. More extensive work has been completed in the southern Sierras, but little has been done to connect the evolution of the Sierra Nevada root as whole for one primary reason, the data that exists north or Fresno, CA is sparse. This paper aims at expanding on the current and well thought-out model for the southern Sierras with seismic refraction data as a control, and what the Sierra’s geophysical structure may appear toward the north of most previous work. The root beneath the Sierra Nevada Mountains should be around 75 km thick given the estimated Mesozoic elevation, the size of the batholith, and the petrology of the xenoliths entrained during volcanic eruptions (Fliedner et al., 2000, Jones et al., 2004). Crustal roots formed at an active continental margin are typically composed of calc- 2 alkaline granite facies. Because of the crustal root’s depth (temperature-pressure gradient and gravity), it often differentiates into more of less intermediate and mafic parts. Either way, it is agreed upon the magmatism is a result of decompression melting from the delaminated root and that there is a small amount of melting at the base of root by asthenosphere (Manley et al., 2000). It is has been shown through several types of geophysical work that the root beneath the southern Sierra Nevada Mountains has delaminated, creating a west centered drip into the mantle (Fliedner et al., 2000, Zandt et al., 2004, Wernicke et al., 1996). It is debated as to how much of the root foundered and whether the delamination occurred throughout the entire Sierra Nevada, or whether it’s progressing northward from the already foundered southern root. There is also debate over the actual process and cause of the foundering of the root. Some argue that negative buoyancy from the eclogized root caused the separation (Zandt et al., 2004), while others argue that an outside force such as extension between Death Valley and Sierras caused stress forming factures within the crust and allowing the heavy root to founder (Manley et al., 2000). Extensive geochemical and petrologic work as been completed within the southern Sierra Nevada and its eastern vicinity. The volcanics of this region provide key insight to the timing and possibly the extent of the delamination of the crustal root. Based on work from Ducea and Saleeby (1996) the composition of xenoliths entrained in Miocene and Pliocene volcanics, they suggest that delamination occurred between 10-3 Ma. Petrologically, the Sierras, at one point in time must have had a complete crustal root in order to have produced the volcanics from the Mesozoic batholith, in addition to the required support for the paleo-elevation of the Sierras (Manley et al., 2000). Although the 3 Sierra have gained over a kilometer of elevation during the Cenozoic, the Sierras had an elevation of at least X (based on paleo-botany of seeds within the Sierras, as well as the amount of erosion and sediments that were deposited into the Great Valley). This bracket of volcanism is coincident with the beginning of the Cenozoic uplift of the Sierras (Jones et al., 2004). This over one kilometer gain in elevation also corresponds to extension in Death Valley (ca. 7 or 6 Ma started (McKenna and Hodges, 1990 cited in Jones et al., 2004, and Wernicke and Snow, 1998). Other concurrent events include the valleys east of Sierras filling with sediment during the late Miocene and early Pliocene (Bacon et al., 1982, Burchfiel et al., 1987, Lueddecke et al., 1998), folding of the California coast ranges and the spreading of the Gulf of California during the Pliocene (Axen and Flether, 1998). Cenozoic volcanism within the Sierra Nevada are marked by two major pulses. The first, occurring in the Miocene, 12 to 8 Ma, produced typical subalkalic to alkalic trending basalts- typical of contential eruptions. After a period of no volcanism, a quick spurt of magmatism between 4 to 3 Ma produced ultra potassic rocks, occurring in a limited area of the southern/central Sierras, specifically in Kings and San Joaquin volcanic fields. Followed by quiet, lava was again erupted during the Quaternary within the Big Pine and Long Valley volcanic fields, however, this time once again producing more typical subalkalic volcanics (Manely et al., 2000). From petrologic analysis alone, many cite the start of the delamination at ca. 3.5 Ma, due to the unqiue Pliocene lavas. They were just a pulse, short-lived, no progression north or south, as moving of plates, only concentrated in southern Sierras and Owens Valley (Manley, et al., 2000). Low Nubd. K-metasomatized, mantle not involved in 4 Sierra magmatism during Cenozoic (Farmer et al., 2002). Xenoliths also indicated that it was shallow (40 km). Newley studied rocks with mafic to intermediate compositions indicated that they have K and P, but also the lowest SiO2 and highest MgO for any known Sierran rocks (Farmer et al., 2002). Highly potassic mafic to intermediate rocks are constrained within the San Joaquin and Kings volcanic fields during the Pliocene (Farmer et al., 2002). Not parent from lower crustal rocks, must have come from the mantle. Miocene volcanics in the central Sierras derived from LILE, Pb-metasomatized, High Nb peridotite, upwelling asthenosphere, or post-Mesozoic lithospheric mantle. Melting from extension in Death Valley and Slab window (Farmer et al., 2002). (From Manley et al. (2000) based on their compilation of over 500 samples, they created a simple timeline for volcanism within the Sierra and surrounding region. Before 13 Ma, also magmatism south of Garlock fault (Mojave desert). Ca. 13 Ma, Sierra Nevada, White mtns. And some in Mojave. Ceased in ca. 9 Ma. Abrubtly began and then ceased between 4-3 Ma. Only within Sierra Neavda, Owens Valley, Coso Range (all high potassic). Activity in ca. 1.5 Ma Coso, Big Pine Long Valley, Golden Trout, which are volcanic fields we see today, all east of Sierra frontal fault.) Crust began delamination process at ca. 3.5 Ma determined by the plus mafic potassic volcanism within and east of the Sierras preceded and followed by times of no volcanism (Manely et al., 2000). Farmer et al.(2002) have mapped these potassic rich volcanics and they do extend north to about 38 N. And in the INC study area alone there are volcanics in the Long Valley area. The volcanic eruption of these xenoliths, high potassium occurred south of Long Valley between 3-4 Myr ago, aiding a time bracket of when the initial foundering, or detachment of the ultra mafic rocks began (Zandt et al., 2004). In 5 the study area of Zandt et al. (2004), they show the presence dense garnet pyroxnenite xenoliths in mid-Miocene volcanics within the Sierra Nevada. These ultramafic rocks are much more dense than typical mantle peridotites, and Zandt et al. (2004) states that these are prone to separation from their granitoid counterparts. Jones et al. (2004) requires more than just the weight of the elclogized crust, but also a shear. Zandt et al. (2004) does suggest shear interms of mantle anisorty in negative polarities where the root has already separated. Cenozoic volcanism within the Sierra Nevadas are marked by two major pulses. Before 13 Ma, magmatism was located south of the Garlock fault within the Mojave desert. Around ca. 13 Ma magmatism began within the Sierra Nevada and White Mountains. The first, occurring in the Miocene, 12 to 8 Ma, produced typical subalkalic to alkalic trending basalts- typical of contential eruptions. Adburbty ceased. After a quiet period of volcanism, a quick spurt of magmatism between 4 to 3 Ma within Sierra Nevada, Owens Valley, and the Coso Range. produced ultra potassic rocks, occurring in a limited area of the southern/central Sierras, specifically in Kings and San Joaquin volcanic fields. Followed by quiet, lava was again erupted during the Quaternary within the Big Pine, Golden Trout and Long Valley volcanic fields, all easr of the Sierra frontal fault.However, this time once again producing more typical subalkalic volcanics. (Manely et al., 2000). However, potassic volcanism itself, doesn’t indicated delamination, however they are similar to Andes and Tibet though, says more about crustal root composition (Manley et al., 2000). The question is raised as what is happening in the southern and northern Sierras. It is clear from geophysical data that there is no southern root (Zandt et al., 2004, 6 Fliedner et al., 2000). However, the Kern Volcanic field erupted Pliocene rocks that are similar to Miocene rocks (Farmer et al., 2002), not from delamination in opinion and results. Also, there is no strong evidence for these Pliocene, K-rocks north of 38 N, but the root was most likely root removed along the entire the Sierra Nevada (Jones et al., 2004) because all Sierras have been uplifted during the Cenozoic. Maybe no volcanics north b/c old craton, Sr line or area that the entire mantle lithosphere was removed. (Jones et al., 2004). OR just not enough evidence in north for the rocks in Lake Tahoe, west 2-4 Ma (D.S. Harwood reported in Saucedo and Wagner, 1992 in Jones et al., 2004). Subsidense area determined by flow of drainage into the Great Valley. North of Fresno flow northward, while drainage flows into the Tulare basin. This subsidense possibly began 3-4 Myr and moved southwestward (Zandt et al., 2004). However the INC experiment does not directly follow this lodgic since the drainage pattern into the Tulare basin is south of the INC crustal welt. A possible explanation is that the delamination and drip is more than just the separation of the root, but as Jones et al. (2004) suggests also due to a horizontal shear, aiding the delamination by fracturing the crust, or some other process. If the deepest part of the delamination is still within the Tulare basin region, also associate with the highest SIerran topography, then welt moving into the mantle would have a greater pull on the crust to the north, having a zipper effect. However drainages did incise, because of uplift everywhere From the most recent inversion model, we observe a crustal root of only 44 km +/- 2 km centered 40 km west of the topographic crest at the latitude of approximately 37 N. The root is shallower than what the depth the root ought to be of at least 75 km by Fliedner et al. (2000). This observed crustal root as described as a “crustal welt” (Zandt et 7 al., 2004) or “small crustal root” (Fliedner et al., 1996) has steep sides and a velocity of 7.6 km/s. Previous results in the Sierras, just south of the INC transect (36.5 N), are in agreement with the location and depth of the crustal root (Ruppert et al., 1998, Zandt et al., 2004, Fliedner et al., 1996, Fliedner et al., 2000). Ruppert et al., (1998) show a 40-42 km root centered 80 km west of topographic crest at 36.5 N and based on the same data, but employing different modeling techniques, Fliedener et al., show a 43 km root east of Fresno that centered 40 km west of topographic crest. The INC transect only extends to Fresno, CA, but based on previous work, the crust thins to 28-33 km west of the root beneath the Great Valley (well data from Wentworth, 1990 in Fliender et al., 2000, Ruppert et al., 1998, and Holbrook and Mooney, 1992). INC Refraction Experiment In August 2004, the Nevada Seismological Laboratory with the University of Nevada, Reno completed a 600 km-long refraction transect extending from the IdahoNevada border, through Battle Mountain, Nevada and western Nevada, across the Long Valley caldera and the central Sierra Nevada, and into Fresno, California (see Figure 1). Spaced approximately 1.5 km apart along the 600 km-long transect, 411, 24-bit single channel portable seismographs, known as RefTek “Texans,” connected to 4.5 Hz geophones recorded mining blasts and small earthquakes. The instruments were programmed to record for a total of 96 hours (four 24-hour periods) on the dates of August 16, 17, 19, and 20, 2004 (Julian Days 229, 230, 232, and 233). Although mine blasts are triggered only during daylight hours, in order to insure the instruments would gather earthquakes occurring at night. In the previous 2002 Northern Walker Lane 8 (NWL) refraction experiment, instruments were only programmed to record during working hours. Valuable earthquakes that would have served as refraction reversals were missed (Louie et al., 2004). The recorder array consisted of Texan instruments with 64 or 32 megabytes of memory. The 32 Mb instruments were retrieved halfway through the experiment for a 24-hour period to download data and change batteries. The seismographs were programmed to sample at 50 Hz, slower than commonly used in refraction recording. The recovery rate for instrument hardware was 99.5%. The data recovery rate was over 99%. The deployment consisted of three download/home stations located in Battle Mountain, NV, Austin, NV, and Bishop, CA. Eight teams of two students were responsible for the deployment and pick-up of approximately 50 Texans instruments. The teams buried each instrument at least one foot (0.3 m) beneath the ground surface to help diminish noise and insulate instrument clocks from changes in temperature. Noise such as traffic and industrial operations played a minimal role, as the transect only crossed three highways (US 395, US 50, I-80) in this undeveloped region. In addition, two teams hiked fifteen of the 411 instruments into the John Muir Wilderness and over the Sierra Nevada crest, recording at 11,500 ft (3500 m) elevation. This is now the first continuous refraction experiment to cross the Sierra Nevada. The success rate for deploying instruments at assigned sites was 100%. However, the pick-up rate was 99.56% due to the loss of two instruments during a desert flash flood. Modeling our experiment after Harder and Keller (2000), we utilized blasts from Nevada gold mines: Barrick GoldStrike, Round Mountain, and Cortez, as seismic sources for this experiment. However, our experiment tested the limitations of the original 9 experiment; Harder and Keller (2000) used one mine blast to record a high-density profile out to a distance of 150 km. This experiment takes the original idea a step further by looking at first arrivals out to a distance of at least 400 km, and by recording earthquakes. In disseminated gold deposits, high revenues and levels of production depend on crushing large quantities of rock. Our experiment reaped the benefits with the recording of 200,000 lb (90,000 kg) ANFO blasts. Explosives were detonated within arrays of 40-ft (12 m) holes, providing a nicely controlled source. The shot holes were ripple-fired, but video recording shows all firing is within a 0.5 second interval. We also employed earthquakes located by the California Integrated Seismic Network and the Western Great Basin Seismic Network. These earthquakes provided reversals from the Barrick GoldStrike blasts, as well as crustal refraction values along the interior of the transect. Mine blasts from Barrick GoldStrike, located near Battle Mountain, Nevada, served as the primary shots for the 2004 seismic refraction experiment. Following the mine’s regular blasting schedule, the instruments captured at least one blast on each day of the experiment. Processing The record has minimal, but standard processing. The processing of the records included: a center function, subtracting each vector's arithmetic mean from the amplitude of elements in the vector; a bandpass filter of 2-16 Hz with a 20 percent taper; a trace equalization gain that gains each vector by the L1 norm of the whole trace- it is equivalent to an AGC (automatic gain control) with the window length set to the trace length; and, the majority of the picks were made with a clip of 2.5 times the RMS (root- 10 mean-squrared) velocity. A Dip filter was also used to verify picks and to view data, but not used for actually picking the data. The Dip filter is usually used in reflection processing to correct for the overcorrection of primary reflection and undercorrection of secondary arrivals. The filer rejects dips that curve down with offset, positive dip Inversion Tomography models from first arrival picks were produced using the optimized, finite differencing scheme of Pullammanappallil and Louie (1994). The program continues to produce models until a global minimum in terms of error is reached. This optimized method does not limit the model product as a typical refraction inversion might. A common problem addressed by Fliedner et al. (2000) is the smoothing over of a laminated Moho, however the optimized inversion is one step in limiting this type of error. The synthetics produced by Louie et al. (2004) for this type of refraction experiment show that this inversion method is reliable and accurate. Grid spacing, 2.5x5 km Monte Carlo-based optimization “general simulated annealing, invert first arrivals to get velocity. Fast finite-difference solution of the eikonal equation. Convergence of of model is independent of the initial model. Not a linear methods. Produces many final models with least-square error. nonlinear Results Barrick GoldStrike blast records 11 Mine blasts from Barrick GoldStrike, located near Battle Mountain, Nevada, served as the primary shots for the 2004 seismic refraction experiment. Following the mine’s regular blasting schedule, the instruments captured at least one blast on each day of the experiment. The Barrick GoldStrike mine is an open pit with approximately 2 by 3 km dimensions and a maximum depth of 5 km from the local elevation. Although blasts were recorded on each day of the experiment, this paper only discusses and utilizes the Thursday and Friday blasts for the model inversion. On August 19, 2004, Julian day 232, Barrick GoldStrike ripple fired two blasts. First arrivals from the blasts, the first blast, 14007 kg (30880 lb or 10036 kg TNT equivalent), and the second blast, considerably larger, 62541 kg (137880 lb or 45787 kg TNT equivalent), can be seen across the entire 600 km record (Figure 2a). Figure 2b, displays the Barrick GoldStrike 232 blast in reduced time (time-offset/velocity). The time is reduced by 7.8 km/s, the continental average velocity of the Moho. Displaying arrivals in reduced time allows the slopes of the refracted arrivals to be more easily distinguished. In actual time, slopes of the refracted arrivals often look too similar. In the reduced time records shown, if the velocity is 7.8 km/s the slope refraction is flat. If the refracted arrivals are slower than 7.8 km/s they will have a negative slope, and if faster, a positive slope. The crossover occurs at approximately 127 km from the blast location, suggesting a 30 km-thick crust just southwest of Barrick GoldStrike. Typical cross-over distances range from 150200 km for average 35 km thick, continental crust. This also concurs with the 1986 PASSCAL refraction experiment that found 30 km thick crust within the vicinity of the INC transect (see Figure 9b). Although battery life was tested before the deployment, by the last day of the 12 experiment many of the Texan instrument batteries died, resulting in missing traces within records from the Friday (233) recording. Because of the missing traces, the start of the Barrick GoldStrike blast 233 record was not picked. However, the first arrivals to the southwest of Barrick GoldStrike were picked to self-check the picks of the Barrick GoldStrike 232 blast on the day before. Often when picking past 400 km from the source, it can be tricky to distinguish between Pn refracted arrivals and noise. By having more than one record from similar events, the first arrivals are more easily distinguished. While there are some discrepancies between the blast picks, this assures that the inversion model isn’t over-fitting the data. On Friday, August 20, 2004, Julian day 233, Barrick GoldStrike ripple fired 3 blasts. The first blast was only 1361 kg (3000 lb, with a 975 kg TNT equivalent. The second and third blasts were 24405 kg (53805 lb, 17487 kg TNT equivalent) and 48906 kg (107820 lb, 36567 kg TNT equivalent). Because the first blast was small (in comparison) first arrival picks were taken from the start of the second blast. The crossover occurred at 125 km from the blast, and again arrivals could be seen out to a distance of at least 450 km, only because so many Texan records were missing. For a typical, northern Basin and Range crust of 30 km with a Moho velocity of 7.8 km/s, there was no delay in any of the picks. The minimum delay of –0.29 seconds occurs about 20 km after the cross-over distance, and is within in error. A +/- 0.5 second error in a pick only results in +/- 2 km in depth for a Moho with a 7.8 km/s velocity. For a slower or faster Moho or crustal velocity, the difference is negligible for the scale of this discussion. The maximum delay of -1.49 seconds, occurred past 450 km from the blast location. These early arrivals are not reflected in the inversion model, but the inversion 13 program best fit the earlier arrivals (see plot). Only Pg picks were made north of Barrick. It is possible that there are one Pn arrivals, however they weren’t clearly distinguishable. The picks have a maximum delay of 1.298 seconds near Barrick GoldStrike. This is significantly more than the other Barrick blast. Possibly resulting crust more like 32 km just southwest of Barrick whereas the previous inversion model had more like 30 km thick crust. The minimum delay of –1.298 seconds out toward the end of the record, agreeing more with Barrick 232 record. Paso Robles, CA earthquake record During the 2002, Northern Walker Lane transect, the Texan instruments recorded a quarry blast in Watsonville, California. The quarry blast was reported on the USGS seismic network. This led us to program for 24 hour increments, as previously mentioned, so the instruments could capture earthquakes occurring anytime. There was also a greater possibility for capturing smaller earthquakes and more numerous quakes as the transect ran through the Long Valley area and into Fresno, California. The experiment was fortunate and recorded two Lake Nacimiento, CA earthquakes, in the vicinity of Paso Robles, CA, directly on strike with transect, providing a reversal to the Barrick GoldStrike blasts. The two earthquakes, M 2.8 Paso Robles, CA and M 3.8 Lake Nacimiento, have first arrivals visible on our records out 750 km from their epicenters. The Lake Nacimiento earthquake, M 3.8, although visible, the larger earthquake provided more difficult first-arrival Pn picks. The complicated nature of the earthquake provided more emergent Pn arrivals. The Paso Robles, CA 233 earthquake record is pickable across the 14 entire record, and remarkably continuous. Although the earthquake was far enough away to allow visible PmP arrivals, because of the size of the earthquake, the arrivals are not discernable among the later, high frequency energy. Because this earthquake occurred during the last day of the experiment, like the Friday Barrick 233 record, many of the instruments suffered battery failure, making the picking through the Sierra Nevadas slightly more difficult. All of the picks on the Paso Robles 233 record are Pn first arrivals because the earthquake had an epicenter of 7800 km depth and an offset of 151 km from the first Texan instrument. Back calculating from the first picks, the Pn cross-over distance should be approximately 150-250 km from the earthquake’s epicenter. This calculated cross-over distance is in agreement with a refraction study just south of the INC transect, who recorded a cross-over distance of 150-175 km from their surface shot point (Ruppert et al., 1998). Tomography models The initial tomography inversion, Model 1, used one blast from Barrick GoldStrike and the 2.8 M Paso Robles, CA earthquake. The inversion model yielded a crustal root beneath the Sierra Nevada Mountains with an approximate depth of 50 km, assuming a Moho velocity of 7.6 beneath the Sierra Nevada (Fliedner et al., 1996). To the northeast of the root, the inversion reveals a typical 30-32 km crust in northern Nevada, deepening to the north toward the Nevada-Idaho state border. The model agrees with the 1986 PASSCAL refraction and the 40N COCORP reflection experiment that both revealed 30 km thick crust in northern Nevada. A recent refraction experiment extending west from the NV-ID state border, shows deeper, 35-40 km thick crust in this area, thus 15 the reason for the deepening of the Moho on this paper’s initial model interpretation (Klemperer et al., 2007). The mid-crustal velocities also seem to support this deepening trend to the north. In order to confirm and expand on the experiment’s basic model, a second tomography inversion was completed. The second model, Model 2, utilized a re-picked Paso Robles, CA earthquake and the Barrick GoldStrike 232 blast, as well as an additional Barrick GoldStrike blast on Julian day 233. The second inversion confirms our original picks and adds clarification to features in the initial version. From Model 2, we observe a crustal root of only 44 km +/- 2 km with steep sides, centered 40 km west of the topographic crest at the latitude of approximately 37 N. The depth of the root partly depends on whether one considers the Moho to have a velocity of 7.6 km/s below the Sierra Nevada or a more traditional 7.8 km/s Moho velocity. In this experiment’s model, and in the models by other workers (Fliedner et al.,1996, Ruppert et al., 1998, Fliedner et al., 2000), we see a laminated Moho beneath the Sierra Nevada. This paper uses convention from more recent papers (Fliedner et al.,1996, Ruppert et al., 1998, Fliedner et al., 2000), and employs a velocity of 7.6 km/s, the top of the laminated Moho as the velocity for the Sierra Nevada root. Within northern Nevada, we observe 30-32 +/- 2 kmthick crust. In agreement with the Model 1 as well as the 1986 PASSCAL and the COCORP 40 N transect, this gives confidence in the picks and inversion. There is little transition between the steep-sided root on it’s eastern side and the Basin and Range. At the latitude of Long Valley, we immediately observe 30 km-thick crust. Assuming a root velocity of 7.6 km/s and a crustal velocity in Nevada of 7.8 km/s, this forces the Moho interpretation to cut across the model’s velocity boundaries. Where the change from 7.6 16 to 7.8 km/s actually takes place is undeterminable from this study alone. We assume that the velocity changes at the corner of the root given that there is most likely upwelling asthenosphere against the recently delaminated mantle lithosphere (**Side note-if there is upwelling asthenosphere wouldn’t it be hotter, therefore slower?) (Zandt et al., 2004, Wernicke et al., 1996). On the north side of Barrick GoldStrike, where we have no Pn first arrival picks, we have extrapolated the Moho across to a depth of 35-38 km based on recent refraction experiment to the north of the INC transect (Klemperer et al., 2007). On the western side of the Sierra Nevada, within the Great Valley, we have again extrapolated the Moho, this time to a depth of 28-30 km based on refraction experiments within the Great Valley and coast ranges (Fliedner et al., 2000, Mooney and Weaver, 1991). Differences and Problems within the models On the whole, the picks were similar except for a few major differences. These changes led to the differences in the results, Models 1 and 2, from the inversion. The first obvious difference is that the first model has a complete Sierran root, and the second model doesn’t provide a clear maximum depth, and the two models roots varying in depth by 5-7 km. Despite the lacking a bottom to the Sierra Nevada root, Model 2 is an improvement for several reasons. Although the depth of the root is not defined, we assume the depth is approximately 42-45 km, whereas Model 1 had a root of 50 km. The main reason for this discrepancy is that the Paso Robles 233 earthquake picks used for the inversion of Model 1 were 2.5 seconds later. After picking the first arrivals from the Paso Robles earthquake for Model 1, there seemed to be a consistent 2.5 second delay 17 across the entire record. This delay was interpreted as a network error in the time that the earthquake was reported, and so 2.5 seconds was added to each pick time. After reevaluating the Paso Robles earthquake record, as well as re-picking the record, this standard error was not present. The picks used for the inversion producing Model 2 did not include any added time. This difference in pick times clearly accommodates the shallowing in depth of the crustal root in Model 2 (0.5 seconds=2 km, 2 seconds=8 km). However, the 2.5 second decrease in pick times doesn’t account for in the large increase in the velocity of the crustal root. Theoretically, it ought to be a trade-off: deeper root with low velocity, or fast shallow crust. Yet, the earlier pick times alone account for the change in depth. This lends us to the second major difference in Model 2. The Barrick GoldStrike 232 blast picks used for the inversion of Model 2 were consistent, except that the record was picked across the entire 500 km southwest of Barrick GoldStrike, whereas previously the record was only picked out to a distance of 350 km. It is possible that these picks with large offsets are too early, forcing the velocity within the root to be to fast. The INC inversion model shows a root with crustal velocities of 7.2-7.7 km/s, much faster than granulite facies (crustal root rocks), and much faster than the 6.0-6.2 km/s velocity of Fliedner et al., (2000) who show that these slower than typical midcrustal velocities extend down to the Moho. The high velocities are most likely an artifact from the inversion due to a lack of Pg arrivals within the Sierras. Fliedner et al., 2000 was able to determine these lower crustal velocities only by modeling the long-offset Pg arrivals. However, Ruppert et al. (1998) and Fliedner et al. (1996), modeled 6.4-6.6 km/s velocity to the base of the root from the same experiment, also in more agreement with Model 1. However, Model 2 and Model 1 have comparable velocities and depths just east 18 of the crustal root and extending the distance to Barrick GoldStrike, suggesting that the entire pick-set is not the source of the problem, but rather a few early picks forcing the entire velocity within the Sierras to be faster. Another argument for the faster arrivals within the crustal root are the extensive ophiolites beneath the Great Valley. Although west of the INC transect, the ophiolites extend to a depth of at least 25 km. Most of the rays from the Paso Robles earthquake, with an epicenter at 7800 km depth should bypass the fast ophiolites, however, it is possible that first arrivals in Fresno, CA are affected as they travel, up to the receivers, causing the entire inversion to adopt these fast arrivals. Also, it is under some debate as to whether or not the ophiolites extend to the base of the crust. Ruppert et al. (1998) argues that the accreted terrain extends down to the Moho, whereas more extensive work from Fliedner et al. (2000) argues that there is a low velocity layer of metasedimentary rocks, like the Franciscan, between the ophiolites and Moho. On a similar note, because the crustal root in the INC model is partly controlled by Pn arrivals from the Barrick GoldStrike blast, which we mainly attribute to the cause of the high velocity in the root, it is possible that rays traveling through the volcanic laden Long Valley caldera and vicinity had faster arrivals, and again possibly forcing the inversion to interpolate based on these fast arrivals. As mentioned, the first inversion for Model 1 did not include picks as far west as the inversion for Model 2. Both Ruppert et al. (1998) and Fliedner et al. (2000) agree that the Panamint Valley to the east of the Sierras has high lower crustal velocities of 6.8-7.0 at a depth 19-22 km, and with a Moho near 30 km. Ruppert et al. (1998) attributes this to Basin and Range underplating. This underplating most likely extends further north beneath the INC transect. Future work of 19 producing synthetics, or purposely picking later and inverting the data may prove useful in determining this error. For several reasons, despite the high velocity within the root, we assert that Model 1 is still a better model for interpreting the depth of the Moho. One, the depth of the Sierran root and the crustal thickness within Northern Nevada is consistent with other data in the study area, as will be discussed (Fliedner et al., 1996, Ruppert et al., 1998, Zandt et al., 2000, PASSCAL 1986). Second, we attribute the high velocity within the crustal root to an interpolation artifact of the inversion and not to any first arrival picks from the data, as the Hitplot (Figure 9) confirms. Third, the lack definition at the bottom of the Sierra Nevada crustal root could be caused by the tunneling of rays through the root, or more likely, the diffraction of seismic rays off of the steep sided walls (during the actual experiment, not just an artifact of the inversion). Analysis of receiver function stations and PmP arrivals show a “Moho hole” at the location of the thick crust, or small crustal root (Zandt et al., 2004, Fliedner et al., 1996). Zandt et al. (2004) attributes this missing Moho to possible scattering of reflections and Ps conversions along a boundary that may have some topography due to deformation caused by delamination. However, small-scale topography along the Moho, shouldn’t affect Pn arrivals other than a smoothing over the region. Also, refraction experiments in the region were able to clearly define a base to the root (Fliedner et al., 1996 and Ruppert et al., 1998). In addition to the better agreement with surrounding geophysical data, as will be discussed, the preferment of Model 2 to Model 1 stems from several basic observations. The picks from the Paso Robles, CA earthquake have no clear delay beneath the Sierras. Unlike the Northern Walker Lane (NWL) refraction experiment that saw 3-4 second 20 delays beneath the 55 km root (Louie et al., 2004). A deeper root within this paper’s study area would require at least a 2 second delay, and one that is not consistent across the entire record, but localized to the Sierras. Also, the cross-over distance for the Paso Robles earthquake is only approximately 150-200 km, and as previously mentioned in agreement with Ruppert et al. (1998) Who also saw a cross-over of 150 km, but from a surface shot. Ruppert et al. (1998) also found a lack of delay in their PmP arrivals. Crustal thickness maps We have complied a contoured crustal thickness map (Figure 10 and 11) based on the NWL (Louie et al., 2004) and INC seismic refraction results presented in this paper, as well as previous geophysical studies within the Great Basin and Sierra Nevada (literature cited in the comprehensive Braile et al., 1989. We extracted information concerning crustal thickness from the crust-mantle (Moho) refraction velocities and arrival times, crossover distances, and from models presented in the literature. A kriging algorithm was used to create 1 km thick contours. The colors on the map are contoured every 1 km, and the black outlines are for every 5 km, showing a more general trend. Presented in Figure 10a is the compilation of all the data, including any discrepancies. The map doesn’t contain any data sets that may have scientifically of methodology issues. In the compilation map, Mooney and Weaver 1989 and Fliedner et al., 2000, served as base maps. Mooney and Weaver (1989) is an interpretation of the crustal thickness within the Sierra Nevada root and surrounding region utilizing most geophysical work before 1989. Their map relies heavily on result from Eaton et al. (1963) and Pakiser and Brune (1982) for crustal control within the Sierra Nevada. Fliedner et al. 21 (2000) expanded on the original Mooney and Weaver (1989) map by inverting the SSCD results along with other data sets including Savage et al. (1990) for the south-central and southern Sierras. Fliedner et al. (2000) point out that many older data sets, specifically Mooney and Weaver (1989), over estimate the depth of the Sierra Nevada root simply by semantics. As previously mentioned in this paper, current convention is to use the top of a laminated Moho, which for the Sierras is about 7.6 km/s, as the depth to calculate crustal thickness. Previously, a velocity of 7.8 km/s was used to describe the Moho beneath the Sierras (Pakiser and Brune, 1982), which is most likely the bottom of any lamination. From this compilation map, a preferred model was generated (Figure 11). The preferred model eliminates most of these velocity discrepancies, except where no other data exists. By compiling these data sets it allows better visualize of key features, and allows us to better address discrepancies and how proposed models fit the actual data. Several key features to note: the crustal root that is centered west of the topographic crest; the deepening of the root north of Fresno, California; the steep sided walls of the crustal root; and the dramatic shift to normal, 30-km thick Basin and Range crust. Discussion It has been show that results from the INC refraction experiment are in agreement with results and modeling from the SCCD experiement (Ruppert et al., 1998, Fliedner et al., 1996, Fliedner et al., 2000), as well as other work (Eaton et al., 1963, Mooney and Weaver, 1989, Zandt et al., 2004). However, these experiments were slightly south of the INC transect. Because the model for the southern Sierras is well developed, using the 22 INC and NWL refraction experiments, we have tied in major geologic events and supporting evidence for the entire Sierra Nevda. Between 12-10 Ma, the ecologized section of the root delaminated across entire Sierra Nevada. This coincides and is probably related to extension within the Death Valley region (Jones et al., 2002, and Jones et al., 1987). The plus of Miocene volcanism also occurs at this time across the entire Sierras. We suggest that the root was removed acorss the entire Sierras for several explanations. First, the Cenozoic uplift of the Sierras began around 12 Ma, and has been linked to the upwelling asthenosphere (Jones et al., 2002, Wernicke et al., 1998, Lui and Shen, 1996). For both the southern and northern parts of the Sierras, calculations show that for the gravity anomolies present in the Sierras, a thick, Airy-type crustal root, at any reasonable depth, cannot accommodate the measurements. Some compensation from the mantle is required (Mavko and Thompson, 1988 and Fliedner et al., 1996). The Sierra Nevada root should have a root of approximatley 75 km depth (Fliedner et al., 2000), but as discussed in the beginning of this paper the root only has a depth of 35 km in the south, thickening to only 55 km in the north, as visable on the preferred crustal thickness map (Figure 11). With the incorporation of results from the INC and NWL transect, this interpration is made possible. After a period of quiet, the pulse of potassic volcanism erupted in a localized area within the Sierra Nevada and east of the range (see outline in Figure 1) (Manley et al., 2000 and Farmer et al., 2003). The composition of the xenoliths entrained in the volcanics imply the sinking of the eclogite facies, the mafic lower crust, as well as the underlying mantle lithosphere (Manley et al., 2000). Between the latitude of 35-38 N, the slightly thinner root, approx. 42 km thick, is centered west of the highest topography (ie 23 Mt. Whitney, see Figure 1). This suggests that more of the root must have foundered in the southern and central Sierras than in the north. From reciever function analysis, there are no PmP arrivals at the location of the thickened crust (Zandt et al., 2004). Zandt et al. (2004) attributed this to the fact the crustal root is dripping into the mantle, and so the Moho is actual “missing.” Because of the evidence from the thickness of the root, the lack of PmP arrivals, and the localized potassic volcanics are interpted that the entire mantle lithosphere was removed in this region (Jones et al., 2003). The mantle drip model is also supported by the drainage patterns in the Great Valley as shown by Zandt et al. (2004). The INC results fall just north of Zandt et al. (2004) proposed mantle drip, but the drip maybe pulling areas the north, south. As previously mentioned it is unclear as to how much of the root to the north was removed. There are no large volcanic fields like in the south-central Sierras, but as Manley et al. (2003) points out, there has been little analysis in the northern region. But because of the location of the proposed Sr line (see Figure 1), the compositions of the xenoliths may be different, and the same type of eruptions cannot be expected. However, from the COCORP transect across the northern Sierras, there are no PmP reflections (Nelson et al., 1987). Klemperer et al. (1989) attribute this to fact that typically in old continental crust PmP reflections aren’t clear, as opposed to recently extended crust that typically have clear, continuous reflections as in the Basin and Range. At this time, there is no evidence that the mantle lithoshopere has been removed in the northern Sierras, and the depth of the root as determined by the NWL transect, 55 km-thick, suggests that it hasn’t, but this cannot be confirmed without further work. Because the root along the 24 entire Sierra Nevada is centered west of the topographic, we assert that the delamination of the root is in conjuction with deformation in the Basin and Range Province. Future work For the INC experiment, there is future work that could add valuable data to the region. Currently there are deep earthquakes within Fresno, CA that are not recorded on the seismic networks. Because there is no recorded time, we need to look at all 96 hours or recording, and at this time to do not have an effective method for doing this. A Fresno, CA earthquake would give better depth and velocity control in the Sierra Nevada. The instrument’s recorded several small, M 1-2 earthquakes in Tom’s Place and Mammoth, NV that would provide crustal velocity and depth control in the Long Valley area. The earthquakes were processed and inverted, but the start time of the earthquakes were off, resulting in an unrealistic geologic and geophysical model. By having the opportunity of re-locating these earthquakes, a more complete model would be achieved. In addition, to the earthquakes and blasts in-line with the INC transect. The experiment recorded off-line earthquakes in Tokop, NV and at Round Mountain mine. The sources will provide important fan-type shots, as well as PmP reflection data for further analysis. Conclusions 1. Some root, not as deep as ought to be 2. No model is confirmed, but can show that models showing no root are wrong i) Negative tests 3. Adds data to community in the area where no data previously existed 25 This material is based upon work supported by the U.S.Department of Energy under instruments numbered DEFG07-02ID14311 and DE-FG36-02ID14311, managedthrough the DOE Golden Field Office. The instruments used in the field program were provided by the PASSCALfacility of the Incorporated Research Institutions for Seismology (IRIS) through the PASSCAL InstrumentCenter at New Mexico Tech. Data collected during this experiment will be available through the IRIS DataManagement Center. The facilities of the IRIS Consortium are supported by the National Science Foundation underCooperative Agreement EAR-0004370 and the Department of Energy National Nuclear Security Administration. TheCalifornia Integrated Seismic Network (CISN) and the Western Great Basin Seismic Network (USGS CooperativeAgreement 04HQAG0004) provided earthquake locations used in this experiment. We would like to thank BarrickGoldStrike Round Mountain and Cortez mines for their cooperation and kind help. References cited 1986 PASSCAL Basin and Range Lithospheric Seismic Experiment Working Group, 1988. “The 1986 PASSCAL Basin and Range lithospheric seismic experiment.” Eos, v. 69, no. 20. Benz, H.M., R.B. Smith, and W.D. Mooney, 1990. “Crustal structure of the northwestern Basin and Range province from the 1986 program for array seismic studies of the continental lithosphere seismic experiment.” JGR, vol. 95, B13 Braile, L.W., W.J. Hinze, R.B. von Frese, and G.R. Keller, 1989. “Seismic properties of the crust and uppermost mantle of the conterminous United States and adjacent Canada, in Pakiser, L.C., and W.D. Mooney: Geophysical Framework of the Continental United States.” GSA Memoir, v. 172, p. 655-680. Carder, D.S., 1973. “Trans-California seismic profile, Death Valley to Monterey Bay.” BSSA, v. 63, no. 2, p. 571-586. 26 Catchings, R.D., and W.D. Mooney, 1991. “Basin and Range crustal and upper mantle structure,northwest to central Nevada.” JGR, v. 96, p. 6247-6267. Diment, W.H., S.W. Stewart, and J.C. Roller, 1961. “Crustal structure from the Nevada Test Site to Kingman, Arizona, from seismic and gravity observations.” JGR, vol. 66, p. 20l-2l. Ducea, M.N. and J.B. Saleeby, 1996. “Buoyancy sources for a large, unrooted mountain range, the Sierra Nevada, California: evidence from xenolith thrmobarometry.” JGR, v. 101, p. 822908244. Eaton, J.P., 1963. “Crustal structure from San Francisco, California, to Eureka, Nevada, from seismic refraction measurements.” JGR, v. 68, p. 5789-5806. Farmer, G.L., A.F. Glazner, and C.R. Manley, 2002. “Did lithospheric delamination trigger late Cenozoic potassic volcanism in the southern Sierra Nevada, California?” GSA Bulletin, v. 114, no. 6, p. 754-768. Fliedner, M.M., S. Ruppert, and Southern Sierra Nevada Continental Dynamics Working Group, 1996. “Three-dimensional crustal structure of the southern Sierra Nevada from seismic fam profiles and gravity modeling.” Geology, v. 24, no. 4, p. 367-370. Fliedner, M.M., S.L. Klemperer, and N.I. Christensen, 2000. “Three-dimensional seismic model of the Sierra Nevada arc, California, and its implications for crustal and upper mantle composition.” JGR, vol. 105, no. B5, p.10. Gibbs, J.F., and J.C. Roller, 1966. “Crustal structure determined by seismic-refraction measurements between the Nevada Test Site and Ludlow, California.” U.S. Geol. Survey Prof. Paper 550-D, p. D125-D131. Harder, S., and G. R. Keller, 2000, Crustal structure determined from a new wide-angle seismic profile in southwestern New Mexico: New Mexico Geol. Soc. Guidebook, 51st Field Conf. Southwest Passage – a trip through the Phanerozoic, 75-78. Hauge, T.A., R.W. Allmendinger, C. Caruso, E.C. Hauser, S.L. Klemperer, S. Opdyke, C.J. Potter, W. Sanford, L.D. Brown, S. Kaufman, and J. Oliver, 1987. “Crustal structure of western Nevada from COCORP deep seismic-reflection data.” GSA Bulletin, v. 98, p. 320-329. Hauser, E., C. Potter, T. A. Hauge, S. Burgess, S. Burtch, J. Mutschler, R. W. Allmendinger, L. Brown, Sidney Kaufman, and Jack E. Oliver, 1987. “Crustal structure of eastern Nevada from COCORP deep seismic reflection data.” GSA Bulletin, vol. 99, p. 833 - 844. 27 Heimgartner, M., J.B. Scott, W. Thelen, C.T. Lopez, and J.N. Louie, 2005. “Variable crustal thickness in the western Great Basin: a compilation of old and new refraction data.” Geothermal Resources Council, Transactions, v. 29, p. 239-242. Holbrook, W.S., 1990, “The crustal structure of the northwestern Basin and Range Province, Nevada, from wide-angle seismic data.” JGR, vol. 95, no. B13:21 Johnson, L.R., 1965. “Crustal structure between Lake Mead, Nevada, and Mono Lake, California.” JGR, vol. 70, no. 12. Jones, C.H., 1987. “Is the extension in Death Valley accommodated by thinning of the mantle lithosphere beneath the Sierra Nevada, California?” Tectonics, v. 6, no. 4, p. 449473. Jones, C.H., and R.A. Phinney, 1998. “Constraints on the seismic structure of the lithosphere from teleseismic converted arrivals observed at small arrays in the southern Sierra Nevada and vicinity, California.” Journal of Geophysical Research, v. 103, p. 10,065-10,090. Jones, C.H., G.L. Farmer, and J. Unruh, 2004. “Tectonics of Pliocene removal of lithosphere of the Sierra Nevada, California.” GSA Bulletin, v. 116, no. 11/12, p. 14081422. Klemperer, S.L., T.A. Hauge, E.C. Hauser, J.E. Oliver, and C.J. Potter, 1986. “The Moho in the northern Basin and Range province, Nevada, along the COCORP 40N seismicreflection transect.” GSA Bulletin, vol. 97. Lerch, D.W., S.L. Klemperer, J.M.G. Glen, D.A. Ponce, E.L. Miller, and J.P. Colgan, 2007. “Crustal structure of the northwestern Basin and Range Province and its transition to unextended volcanic plateaus.” ??????? (currently personal communication). Louie, J.N., W. Thelen, S.B. Smith, J.B. Scott, M. Clark, and S. Pullammanappallil, 2004. “The Northern Walker Lane refraction experiment: Pn arrivals and the northern Sierra Nevada root.” Tectonophysics, v. 388, no. 1-4, p. 253-269. Lui, M., and Y. Shen, 1998. “Sierra Nevada uplift: a ductile link to mantle upwelling under the Basin and Range province.” Geology, v. 26, no. 4, p 299-302. Mangio, S.G., G. Zandt, and C.J. Ammon, 1993. “The receiver structure beneath Mina, Nevada.” Bulletin of Seismological Society of America, v. 83, p. 542 - 560. Manley, C.R., A.F. Glazner, and G.L. Farmer, 2000. “Timing of volcanism in the Sierra Nevada of California: Evidence for Pliocene delamination of the batholithic root?” Geology, v. 28, no. 9, p. 811-814. 28 Mavko, B.B., and G.A. Thompson, 1983. “Crustal and upper mantle structure of the northern and central Sierra Nevada.” JGR, vol. 88, no. B7 Mooney, W.D., and C.S. Weaver, 1989. Regional crustal structure and tectonics of the Pacific Coastal States; California, Oregon, and Washington, in Pakiser, L.C., and Mooney, W.D., Geophysical framework of the continental United States: Boulder, Colorado, Geological Society of America Memoir 172. Nelson, K.D., T.F. Zhu, A. Gibbs, R. Haris, J.E. Oliver, S. Kaufman, L. Brown, and R.A. Schwickert, 1986. “COCORP deep seismic reflection profiling in the northern Sierra Nevada, California.” Tectonics, v. 5, no. 2, p. 321-333. Özalaybey, S., M.K. Savage, A.F. Sheehan, J.N. Louie, and J.N. Brune, 1997. “Shearwave velocity structure in the northern Basin and Range province from the combined analysis of receiver functions and surface waves.” Bulletin of Seismological Society of America, v. 87, p.183 - 199. Pakiser, L.C. and J.N. Brune, 1980. Seismic models of the root of the Sierra Nevada. Science, vol. 210, no. 4474, p. 1088-1094. Potter, C.J., C. Liu, J. Huang, L. Zheng, T.A. Hauge, E.C. Hauser, R.W. Allmendinger, J.E. Oliver, S. Kaufman, and L. Brown, 1987. “Crustal structure of north-central Nevada; results from COCORP deep seismic profiling.” Geological Society of America Bulletin, v. 98, p. 330-337. Priestley, K.F., A.S. Ryall, and G.S. Fezie, 1982. “Crust and upper mantle structure in the northwest Basin and Range province.” BSSA, vol. 72, no. 3, p. 911-923. Prodehl, C., 1979. “Crustal structure of the western United States.” U. S. Geol. Survey Prof. Paper, P 1034:74 Pullammanappallil, S.K., and J.N. Louie, 1994. “A generalized simulated-annealing optimization for inversion of first-arrival times.” Bulletin of Seismological Society of America, v. 84, p. 1397Roller, J.C., 1964. “Crustal structure in the vicinity of Las Vegas, Nevada, from seismic and gravity observations.” U.S. Geol. Survey Prof. Paper 475-D, p. D108-D111. Roller, J.C., and J.H. Healy, 1963. “Seismic-refraction measurements of crustal structure between Santa Monica Bay and Lake Mead.” JGR, vol. 68, no. 20. Ruppert, S., M.M. Fliedner, and G. Zandt, 1998. “Thin crust and active upper mantle beneath the Southern Sierra Nevada in the western United States.” Tectonophysics, vol. 286, p. 237-252. 29 Ryall, A., and D.J. Stuart, 1963. “Travel times and amplitudes from nuclear explosions, Nevada Test Site to Ordway, Colorado.” JGR, vol. 68, no. 20 Saleeby, J. and Z. Foster, 2004. “Topographic response to mantle lithosphere removal in the southern Sierra Nevada region, California.” Geology, vol. 32, no. 3, p. 245-248. Savage, M.K., Li Li, J.P. Eaton, C.H. Jones, and J.N. Brune, 1994. “Earthquake refraction profiles of the root of the Sierra Nevada.” Tectonics, vol. 13, no. 4, p. 803-817. Spieth, M.A., D.P. Hill, and R.J. Geller, 1981. “Crustal structure in the northwestern foothills of the Sierra Nevada from seismic refraction experiments.” BSSA, vol. 71, p. 1075 - 1087 Stauber, D.A., and D.M. Boore, 1978. “Crustal thickness in northern Nevada from seismic refraction profiles.” BSSA, vol. 68, no. 4, p. 1049-1058. Taylor, S.R., 1983. “Three-dimensional crust and upper mantle structure at the Nevada Test Site.” JGR, vol. 88, no. B3. Thompson, G.A., R. Catchings, E. Goodwin, S. Holbrook, C. Jarchow, C. Mann, J. McCarthy, and D. Okaya, 1989. “Geophysics of the western Basin and Range province, in Pakiser, L.C., and W.D. Mooney: Geophysical Framework of the Continental United States.” GSA Memoir, v. 172. Wernicke, B., R. Clayton, M. Ducea, C.H. Jones, S. Park, S. Ruppert, J. Saleeby, J.K. Snow, L. Squires, M. Fliedner, G. Jiracek, R. Keller, S. Klemperer, J. Luetgert, P. Malin, K. Miller, W. Mooney, H. Oliver, and R. Phinney, 1996. “Origin of high mountains in the continents: the southern Sierra Nevada.” Science, v. 271, p. 190-193. Zandt, G., H. Gilbert, T.J. Owens, M. Ducea, J. Saleeby, and C.H. Jones, 2004. “Actice foundering of a continental arc beneath the southern Sierra Nevada in California.” Nature, vol. 431.