On-going and past research

Note that throughout this section you can see abstracts of papers and talks by selecting hyperlinks. Some older research topics remain to provide some context for ongoing work. This only gets updated on rare occasions; see my cv for current publications.

Research Topics


Vertical Extent of Plate Boundaries
Both rigid plate and continuum deformation models have had success in explaining different aspects of global tectonics. Within deforming regions, though, it is unclear just exactly how surface deformation is accommodated at depth. Do distributed faults at the surface reflect a broad region of deformation, or is the upper crust detached from the mantle such that a plate boundary could be a very localized shear zone in the mantle? This question has led us to New Zealand. From December 2000 to mid 2002 we deployed 10 arrays of seismometers across the Marlborough Fault Zone in the north part of the South Island of New Zealand. Analysis of this data for anisotropy from S waves and Ps converted arrivals is being examined to detect shear zones reflecting these different styles of deformation. Initial results were discussed by Wilson et al., 2001 and Wilson et al., 2003; our interpretation is that the Moho is uncut and underlies a pervasively deformed lower crust that spreads the surface slip over a broader area (Wilson et al., 2004). Deeper levels of the evolving Pacific-Australian plate boundary were explored by Boyd et al. (2007).
 
Uplift and Geodynamics of the Sierra Nevada
The Sierra has been a life-long focus of my research; this led to me writing a trade book, The Mountains that Remade America, to be published by the University of California Press in 2017. My work in the Sierra Nevada started with a review integrating pertinent geological, geomorphological, and geophysical observations bearing on the late Cenozoic uplift of the range and tectonism in the Basin and Range immediately to the east (Jones, 1987). This was followed by a seismological field experiment of the southern Sierra that indicated that the range's uplift probably is supported within the mantle and not by a crustal root formed in the Mesozoic as had previously been asserted (Jones, Kanamori, and Roecker, 1994 ). This in turn has led to my participation in a multidisciplinary Continental Dynamics program (1993 Southern Sierra Nevada CD) transect of the range, principal results from the many experiments in this project support the inferred absence of a crustal root and likely presence of an anomaly in the upper mantle (Wernicke et al., 1996). Results from use of small-aperture, mixed short-period and broadband seismic arrays in this experiment (Jones and Phinney, 1998 ) confirm the absence of a thick crust and the importance of dynamics within the mantle in the history of the Sierra, though some other observations remain problematic (Savage et al., 1994). Additional work with these arrays indicated interesting opportunities for analysis or local and regional earthquakes (Phinney, Jones, and Parker, 1994). While this work has improved our understanding of the modern forces balancing the elevation of the Sierra, it has created problems for tectonic models for the history of the Sierra. This led to a second CD project (The Sierran Paradox Continental Dynamics Project) which included a new deployment of seismometers in the southern Sierra to conduct experiments on shear-wave tomography and anisotropy. Initial results have shown great variations in the anisotropy and shear-wave velocities in the upper mantle under the southern Sierra (Jones and Phinney, 1999). This work has reinforced the suggestion of other investigators that eclogitic material was removed from below the Sierran crust about 3.5 Ma; such a removal will drive uplift, extension, and likely force shortening across the Coast Ranges, effects which can be documented in California (Jones et al., 2002, 2003, 2004). It is even possible that this could influence rates of motion on the San Andreas fault system. I assisted Jeff Unruh and Egill Hauksson in evaluating the effect of the Isabella anomaly on upper crustal seismicity, reaching the counterintuitive conclusion that modern day forces are directed upwards, deflecting principal strain rate axes in south-central California (Unruh et al., 2014). The material that was removed is probably in two large seismic velocity anomalies seen at either end of California's Great Valley; we have imaged the southern body containing high-wavespeed, low attenuation material, finding it plunges to the east or southeast under the southern Sierra (Boyd et al., 2004, Jones et al., 2014, Bernardino et al., 2015, 2016). We have found that a Rayleigh-Taylor instability is capable of explaining how this material was removed despite the cold initial temperatures; the key is recognizing the breakdown of power-law creep at high stresses (Molnar and Jones, 2004). Testing these ideas seismologically (Sierra Nevada EarthScope Project) and through a collaborative, multidisciplinary attack (Sierra Nevada Drips Continental Dynamics Project) has been a major focus since 2005. Results of this work indicate considerable similarity along the Sierra; a bright and fairly shallow Moho exists along the crest from the southern Sierra into the Lake Tahoe region and a deep and vague Moho in the foothills (Frassetto et al., 2011). P-wave tomography (Reeg et al., 2008; Jones et al., 2014) also shows a lot of continuity along strike of the Sierra, with a high wavespeed body in the crust to uppermost mantle along the Sierran foothills that ends near deeper high wavespeed anomalies. Ongoing work (2016) is to try and separate intrinsic and scattering-caused attenuation of teleseismic signals and to use Steve Roecker's full waveform software to understand the "Pn shadow zone" that was the original basis for inferring a thick crust under the Sierra Nevada.
Laramide Orogeny/Southern Rockies
Understanding the creation of mountains in the Southern Rocky Mountains between about 70 and 45 million years ago has been a major tectonic challenge for more than a century and the associated topography of the region, which might be from a younger event, remains an enigma. Created more than 1000 km from the subduction zone on the west side of North America, most workers prefer an explanation requiring the subducting Farallon slab to skim along the base of the North American lithosphere. However, a detailed examination of the conflicting requirements of different datasets makes untenable the few unified explanations for the geometry of and stresses induced by this slab (e.g., Jones 2014). Lang Farmer, Shijie Zhong (Physics), Brad Sageman (Northwestern) and I developed an alternative framework for explaining the Laramide (Jones et al., 2011; Jones et al, 2007) that relies more on the characteristics of the overriding North American plate interacting with a shallowing slab. Some additional work on this process was described by Harig et al. 2010, which attempts to reconcile older analytical solutions for the controls on slab dip with more recent numerical simulations of this process. An outgrowth of this too has been the development of a hypothesis that the high elevations of much of the Laramide orogen as well as less deformed areas surrounding it was caused by hydration of the lower crust destroying garnet (Jones et al., 2011, Jones et al., 2015).
Structure Associated with the Coso Geothermal Area
An outgrowth of the Sierran work was a major deployment of seismometers in the Coso Geothermal area from 1998 to 2000; short period seismometers in arrays of 5-11 sensors were placed at about 20 different locations in the area. Receiver function analysis both located the magma in the subsurface and found no evidence of a deeper magma chamber (Wilson et al., JGR, 2003). Also absent was any indication of the regional west-dipping negative conversion inferred to represent a master detachment fault to the north. Instead we posit that the magma body has reorganized the strain system such that subhorizontal shear zones extend out from the base of the magma body, separating little deformed upper crust from highly sheared lower crust. This interpretation is supported by numerical simulations of extension of a magma body (C. K. Wilson PhD thesis, unpublished work). This data had a second life as the material used in an ambient noise tomographic inversion for crustal structure in the region (Yang et al., G^3, 2011) that found an interesting correlation of low mid-upper crustal wavespeeds and areas of late Tertiary and Quaternary magmatism.
Forces producing uplift in the western U.S.
A major outgrowth of the Sierran work has been an effort to separate the relative roles of the crust and upper mantle in supporting topography in the western U.S. By integrating seismic information with an assumption of isostasy, I have inferred that some puzzling variations in the geology in the western U.S. (e.g., the difference between northern and southern Basin and Range geology and topography) are a result of density or compositional variations in the mantle (Jones et al., 1992). This in part motivated a 3-D inversion of local and teleseismic travel times in southern Nevada for structure of the crust and upper mantle across the boundary from the higher Great Basin to the lower central Basin and Range (Chen, Roecker, Jones, and Gomberg, 1993, ms. in permanent limbo). Some of this work helped inform a more recent attempt to isolate the sources of buoyancy in the central Basin and Range (Schulte-Pelkum et al., 2011). I applied similar techniques to results from the Rocky Mountain Front experiment that helped to demonstrate the role of the mantle in the elevation of the Rockies (Sheehan et al., 1995) and will be exploiting this further in analyzing results of the 1994-5 Colorado Plateau- Great Basin experiment (Jones, 1996; Jones et al., 1995). This work led to wider efforts by myself, Jeff Unruh, and Leslie Sonder to quantify the potential energy budget of the lithosphere with the goal of explaining the forces both today and through the Cenozoic that have deformed the U.S. Cordillera (e.g., Jones, Sonder, and Unruh, 1996); our first paper documents that such internally-derived forces are capable of producing the deformation observed eastward of the San Andreas- Eastern California Fault Zone systems (Jones, Unruh, and Sonder, 1996). With the passage of Earthscope's Transportable Array across the western U.S. and the development of ambient-noise tomography capable of producing good crustal wavespeed structures, student Will Levandowski and I have collaborated with Weisen Shen and Mike Ritzwoller to reconsider the variations in buoyancy supporting topography and their implications for body force balance in the western U.S. (Levandowski et al., 2011). A rather distantly related project initiated by Karl Mueller is a flexural analysis of the uplifted shorelines in southernmost California and northern Baja California that reveals a westward migration of a mantle source of buoyancy in the past ~1 M.y. (Mueller et al., 2009). My interest has been attracted to understanding the origins of the elevations of areas not profoundly affected by Cenozoic tectonism such as the High Plains and Colorado Plateau. We advanced the hypothesis that hydration of the lower crust has been responsible for a considerable part of this uplift (Jones et al., 2015). Efforts to isolate and prove or disprove this hypothesis (Jones et al. 2016) have included an attempt to strip off the known effects of surface faulting and sedimentation (Bogolub and Jones, 2016) and applying an expansion of the Levandowski et al. formulation relating seismic wavespeeds and densities to the Plains and Colorado Plateau (Levandowski et al., in review, 2016). A more distantly related piece of research questioning the magnitude and definition of dynamic topography argues that sub-lithospheric density anomalies are not responsible for large magnitude continental topography (Molnar et al., 2015)
Vertical Axis Rotations in Contractional Folds
The observation of large vertical-axis folds adjacent to large strike-slip faults (below) brings up the question whether such rotations might be present in other environments. One of the most promising places to look is probably the Colorado Plateau, where monoclines overlie diversely oriented basement faults unlikely to have reactivated in purely dip-slip mode. As discussed at GSA in 2000 (Jones, 2000), some observations of folding in Washington state can be interpreted as a product of something like oroflexural bending approaching the fold axis. Similar processes provide an alternate explanation to vertical-axis rotations seen in the California Coast Ranges, consistent with some obliquity on the deeper thrusts noted by Jones and Wesnousky, (1992). In the Plateau, we might expect monoclines to also have such rotations simply because the diversity of fold orientation seems unlikely to be matched by an equally diverse driving stress. If this mechanism proves to be of importance, then many assumptions about the relationship of fold axis to paleostress directions are incorrect. An initial foray into the Plateau (Tetreault and Jones, 2001) found no rotation in one structure; additional work has turned up rotations in the Grand Hogback (Tetreault et al, 2003). Field work in the folds on the west side of the San Joaquin Valley explored the possible existence of large rotations from shear along the folds' axes (T├ętreault and Jones, 2007). Detailed work on the Grayback Monocline (Tétreault et al., 2008) indicates that rotations are limited to the forelimb of trishear folds, which seems to also be the case in the somewhat less asymmetric folds in the Coalinga area (Tétreault thesis).
Oroflexural bending of the crust and tectonics of Lake Mead region.
A bending origin for large (1-1000 km) "oroflexural" bends of mountain ranges adjacent to some strike-slip faults has only been verified within the past ten years through the use of paleomagnetic observations (e.g., Nelson and Jones, 1987). My co-PIs, Leslie Sonder and Steve Salyards, and I have found a strikingly systematic variation of paleomagnetic rotation with position relative to the Las Vegas Valley Shear Zone in southern Nevada; this variation is far more consistent with continuum deformation of the Earth than with rotations of rigid blocks larger than 1-3 km (Sonder, Jones, Salyards, and Murphy 1994). Areas previously mapped as rigid blocks have internal variations in paleomagnetic rotation, indicating that they are not truly rigid but must deform internally. Mapping conducted by a University of Colorado M.S. student (Colin Shaw) revealed little macroscopic deformation (fault offsets > ~1 m) between sites with very different paleomagnetic declinations. In addition to the general questions of such "oroflexural" deformation, our observations in this area indicate that existing palinspastic reconstructions crossing the area (which assumed no vertical-axis rotations) should be revised. We have constructed firm stratigraphic ties across the Las Vegas Valley Shear Zone using magnetostratigraphy (Salyards, Jones, and Sonder, 1994; also unpublished work). As this area lies within the only strip of the Basin and Range with a firm geologically-based estimate for total extension, results of this work could influence most estimates of overall Basin and Range extension.
Slip-Partitioning in the Western U.S.
The existence of active, parallel strike-slip and dip-slip (or oblique-slip) faults has often required exotic explanation. Several investigators have linked such faults into a single system where oblique slip has been partitioned between the two (or more) faults. Steve Wesnousky (Univ. Nevada-Reno) and I have taken this idea and applied it to historical faulting along the San Andreas fault and found simple explanations for differing behaviors of faults adjoining the San Andreas (Jones and Wesnousky, 1992). Others have explained the close association of dip-slip faults with the San Andreas as indicating that the San Andreas is substantially weaker than adjoining faults; within the context of slip-partitioning, we have found that the San Andreas and adjacent faults can be equally weak, and variations in slip obliquity determined by the geometry of faulting. Other simple predictions of this model include variations in slip rate along the San Andreas and estimates of recurrence times of earthquakes on dip-slip or oblique-slip structures adjacent to the San Andreas fault.

The presence of the strike-slip Owens Valley Fault and adjacent dip-slip Independence Fault in the California Basin and Range has been attributed to time-varying stresses; within the context of slip-partitioning it is more simply explained as a good example of partitioning of slip within a system not varying with time (Wesnousky and Jones, 1994). Furthermore, with the limited observations at hand it appears that the physical framework of slip partitioning can provide insights into the physics of low-angle normal faults. The northern Panamint Valley fault zone (to the east of the Owens Valley fault) was found to be a shallow low-angle normal fault (M.I.T. 1985 Field Geophysics Course and Biehler, 1987) of recent vintage (last 5 m.y.); it also has oblique slip (Burchfiel et al., 1987), which is impossible unless the fault is exceptionally weak. This inference is stronger if slip is partitioned between the Panamint Valley and Owens Valley systems. Future research will focus on identifying instances of oblique-slip and partitioned systems elsewhere in the Basin and Range with the goal of evaluating variations in strength between faults, variations in stress directions and extension directions within the Basin and Range, and variations in style of oblique-slip.

Seismotectonics in the Western U.S.
The modern expression of extensional faulting can be found in the seismicity and neotectonics of the Basin and Range. I have attempted to address two questions by examining seismicity: what relationship does modern deformation have to the plate margin, and is there evidence indicating that seismogenic low-angle normal faults exist in the Basin and Range today. The first question was examined as part of a deployment of seismometers in the High Sierra in 1988; earthquakes recorded during this time were used to construct a simple velocity structure and were then relocated (as were many other events after 1980). Focal mechanisms were obtained for many events; the pattern that emerged suggested left-normal slip occurs across the southern Sierra Nevada, linking the Central Basin and Range with the San Andreas Fault (Qian et al, 1990). This work was continued to the north by Edwards and Jones, 1998. The question of active low-angle normal faulting was considered in a microearthquake survey in the Hansel Valley region of northern Utah in 1983. Relocated earthquakes in that area reveal scattered seismicity of all types, but mostly normal faulting outside of a "structural know" where thrusting and oblique-slip faulting were prevalent. The pattern of seismicity was interpreted as revealing a low-angle normal fault south of the knot at about 5 km depth and a lateral ramp in the LANF at the knot. This inferred LANF projects up to the west to the Raft River Mountains; the ramp projects to a major left-step in that detachment system along the north side of the Raft River Mountains (Jones et al, 1993). At a somewhat longer timescale, geophysical evidence in northern Panamint Valley appears to require the presence of a young (post 4-5 Ma) fault at very shallow depths (< 3 km depth) (M.I.T. 1985 Field Geophysics Course and Biehler, 1987). The presence of such a young, flat (<15° Burchfiel et al, 1987) fault at such a shallow depth reinforces the likelihood that LANFs slip at seismogenic depths in the crust. A rather more unusual approach to using seismicity is in Unruh et al. (2014), where the principal components of the strain rate tensors were found to vary in south-central California, possibly caused by rebounding crust as the Isabella anomaly decouples from the lithosphere.
Mapping in Death Valley
A key area in unraveling the large extensional strains in the Death Valley region is in the Alexander Hills and adjoining China Ranch basin just south of Tecopa, California. This region is between proximal landslide deposits and the large silicic Miocene intrusive (Kingston Peak pluton) inferred to be the source, now 25 km distant. I mapped in this area in 1984-6 (with some field checking in 1993) and have found an extensional duplex linked to a right-lateral strike-slip fault, later left-lateral faulting, and a polygenetic history for the China Ranch basin. Part of the mapping will be published as part of a USGS GQ map (Wright, Jones, and Scott, in preparation for publication as a USGS OFR then GQ map-but don't hold your breath, I still haven't seen this move in a long time, and with Lauren Wrights's death, not sure this will ever emerge). Some structures appear to reflect the uplift and denudation of the Kingston Peak pluton; in particular, exposure of the Tertiary pluton in the eastern part of the field area might be a klippe of the pluton left as the roof of the pluton was transported to the west. Future work, including detailed gravity and magnetics, might determine the depth extent of this exposure.
Velocity inversions of seismic arrival times
Several different studies mentioned above use variations and improvements on techniques originally developed by Steve Roecker. Results of the studies are described in abstracts. All the inversions are least-squares style iterative inversions based on the ACH inversion method of Aki et al (1977). Constant improvements to the raw codes have been made through these investigations. Expansion of the codes to use reparameterizations (meta-blocks) for teleseismic inversions was accomplished for Jones, Kanamori, and Roecker (1994), as was the addition of the use of PKP arrivals, some clarification of the use of weights for teleseismic arrival times, and a simplified recovery of off-diagonal results in resolution and covariance matrices. Minor improvements (mostly i/o related) were made for local earthquake studies of Qian, Jones, and Kanamori (1990) and Magistrale, Kanamori, and Jones (1992). Roecker and Chen reformulated the inversion to allow for a far greater number of degrees of freedom while retaining the full least-squares inversion; this inversion has been used in Chen, Roecker, Jones, and Gomberg (1993). Analysis of the 1997 Sierran Paradox Experiment data used a few unusual tricks to better constrain the petrology and physical characteristics of the sub-southern Sierra, including inverting for wavespeeds of S-waves of different polarizations (Boyd et al., 2004). Tomography from the SNEP dataset from 2005-7 uses the Roecker inversions again, taking advantage of the finite-difference 3-D raytracer now available. These codes have been expanded once again for both studies of attenuation (Bernardino et al., 2016b) and shear-wave tomography (Bernardino et al., 2015, 2016a) in the Sierra.

Please send mail to cjones@colorado.edu if you encounter any problems or have suggestions.

C. H. Jones | CIRES | Dept. of Geological Sciences | Univ. of Colorado at Boulder

Last modified at October 15, 2016 5:02 PM