[This article is in response to a feedback question I received some time ago. The reader brought to my attention several Geological Society of America (GSA) field trips led by a group of young-Earth creationists (YECs) last year. Although young-Earth (Flood) geology was not expressly taught on the field trips, the YEC leaders visited several sites, which they believe challenge the conventional geologic timescale. I spent some time researching the claimed examples of a young Earth, and have focused here on their presentation regarding sand injectites found along the Ute Pass Fault near Manitou Springs, Colorado. Thank you again for the feedback, and I look forward to hearing more of your questions!]
Unconsolidated Earth under pressure: sand injectites in the geologic record
What is a sand injectite? In short, it is a term applied to an irregular sand body—in the form of a pipe, dike, sill, or diapir—that formed when already deposited sand was remobilized in the subsurface. Imagine standing on a sealed tube of toothpaste, and then puncturing the container with a nail, except…while the tube is buried under a layer of mud. The toothpaste, representing unconsolidated sandy sediment, is then injected upward into the overlying sediments. The resulting intrusions have been called sand injectites, sand pipes, clastic dikes, and sand diapirs, depending on their form.
Occasionally, the remnants of injectites are visible at the surface, such as in Kodachrome Basin State Park in southern Utah, or the Panoche Hills in California (Hurst et al., 2011), and outcrop examples have been known for more than 100 years. But geologists have only recently investigated the processes behind their formation. One reason is that the kinematics behind sand injection are difficult to characterize without subsurface imaging and complex physical modeling (e.g. Huuse et al., 2010; Ross et al., 2011)—tools not available to the typical field sedimentologist. Another reason is found in the following excerpt from Schlumberger, a petroleum exploration and production group, who noted:
Under certain conditions, unconsolidated sand is remobilized and forced upward through overlying layers. Called injectites, these sands can have high porosity and permeability and play a huge role in planning and optimizing hydrocarbon recovery.
Hurst et al. (2005) echoed these descriptors and determined that sand injectites constitute an excellent, but relatively unexplored, play in petroleum exploration, where hydrocarbon preservation potential was high. For reference, a play in the oil industry refers to a type of deposit or structure (channel sands, dune fields, submarine canyons) that could potentially trap and preserve hydrocarbons (oil and gas). Some major sand injectites, such as in the North Sea, are comprised of well sorted, homogenous, highly porous and permeable sandstones. In the oil industry, these characteristics are of prime importance when it comes to recovering the maximum amount of oil from a reservoir. Thus, sand injectites make ideal reservoir rocks, and their irregular shape aids in trapping oil and gas.
The moral of the story is simply this: petroleum exploration companies have a lot of money, and are willing to spend that money researching aspects of geology that help them better recover oil and gas. Since Dixon et al. (1995) first explored the importance of diapiric sand in petroleum systems 16 years ago, our understanding of sand injectites has grown exponentially.
Modern understanding of sand injection: triggers and fluidization mechanisms
Sand injectites begin as relatively flat (tabular) bodies of sandy sediment, such as those deposited in coastal margins or eolian dunes (e.g. Mississippi River delta and Saharan desert, respectively). During periods of rising sea level, or high subsidence, the sand is overlain by fine-grained muds, or in some cases, evaporites, which may act as a low-permeability seal during burial. Normally, water in the pore spaces of both sediment types would escape as the rock pressure increases, allowing both the mud and sand to compact—the first step of lithification. If the geometry is just right, however, the surrounding mudstone can effectively prevent pore water from escaping the sand body during burial. Not only does this cause the sand layer to become overpressured (a condition that occurs when the pore-water pressure is higher than from the weight of overlying rock alone), but it prevents cementation—the next step of lithification.
Though sandstone lithification essentially halts in the scenario above, the surrounding mud continues to undergo diagenetic modification. First, the mudstone undergoes physical compaction, in which pore water is allowed to escape. At deeper burial (6,000–9,000 ft; 100–110°C), montmorillonite (a common clay mineral) converts to illite. The process involves loss of mineral-bound water to adjacent sedimentary units (fluid migration), as well as volume loss (since illite is smaller), causing clay-rich layers to fracture at depth (Selley, 1998).
Since both modes of compaction cause the mud to shrink, they can potentially undermine the seal that had kept the sand body overpressured. Alternatively, rising hydrostatic pressure in the underlying sandstone will inevitably fracture the mudstone when the upward normal stress overcomes the strength of the cap rock. In either case, high-pressure streams of water are forced upward into the overlying sediments, along with unconsolidated sand. At this point, the extent and geometry of sand injection is only a matter of physics, obviously dependent on the parameters of each scenario (overlying lithology, burial depth, initial hydrostatic pressure, etc.).
To add some perspective, Vigorito and Hurst (2010) reported fluid pressures of ~25 MPa, or 3,625 psi, after mobilization had occurred, and estimated that 27 MPa (~4,000 psi) was necessary to cause fracturing of the mudstone seal. Compare these pressures to the average 30–35 psi in your tires! Scott et al. (2009) estimated subsurface sandstone velocities up to 9.43 m/s, or some 21 mph. Sand injection is no gradual process.
Hurst et al. (2011) summarized a number of proposed triggers for sand mobilization: seismic events, fluid migration, igneous intrusion, and even meteor impacts. These mechanisms are not mutually exclusive to a scenario involving overpressure, however, and are more likely complementary (i.e. the straw that broke the camel’s back; see Huuse et al., 2010). For example, soft-sediment deformation is common in tectonically active regions, like the Late Cretaceous Sevier Foreland Basin of southern Utah, exposed near Cedar City (Parowan Canyon) and Gunlock. If a fluid-saturated sand body is already at high pressure and unconsolidated, even a modest earthquake could set the catastrophic dewatering process into motion.
And for the record: yes, catastrophic processes are perfectly consistent with uniformitarianism!
Sand injectites are dominantly fine to medium-grained, showing graded sedimentary structures that depend on the flow characteristics (banding in lower flow regimes; absence of structure in highest flow regimes; Hurst et al., 2011). Erosion of the surrounding bedrock may also occur. Cylindrical pipes commonly contain fine-grained sand at the core, surrounded by brecciated fragments toward the edge (Hurst et al., 2011).
Young-Earth arguments based on sand injectites
Young-Earth geologists have long argued that sand injectites are problematic for the ‘uniformitarian’ timeline, because they find it inconceivable that buried sand could remain unlithified for thousands to millions of years. Rather, they will argue that sand injectites (and other examples of soft-sediment deformation) warrant a significant rescaling of the geologic timescale—in this case, from hundreds of millions of years to less than 5,000 years. But is the argument premature, given our current understanding of post-depositional sand injection? I will examine two major cases in point here, and conclude that sand injectites are not problematic for the conventional geologic timeline.
Kodachrome Basin State Park, UT
Columnar sand pipes were cited early on as evidence against the conventional geological time scale by Roth (1992), who posited that the Jurassic sandstones of Kodachrome Basin State Park should have lithified (cemented) before the supposed remobilization. He argued that sandy sediments would have to remain unlithified for some 150 million years, based on field relationships. William Hoesch of ICR restated the case here, expressing his doubt with “quotation marks” that sediments remained unconsolidated for more than even 10 million years.
Missing overburden in the Young-Earth timeline
Roth (1992) argued erroneously, however, that movement of the sand occurred as late as Pleistocene, not realizing this would require the process to take place under only a few hundred feet of overburden (i.e. very low pressure). More likely, the sand injected later in the Jurassic (~140–150 Ma; see Netoff, 2002), long before the erosional unconformity at the base of the Upper Cretaceous Dakota Sandstone was formed (~90 Ma). The injectites did not pierce Pleistocene-age sediments, but rather those sediments were deposited on top of weather-resistant quartz arenites of the columnar sand bodies.
Hoesch argued for a Cretaceous-aged injection, based on soft-sediment deformation in the Dakota Sandstone, but the two are not necessarily related. While common in Cretaceous formations in southern Utah, soft-sediment deformation (a typical sign of seismic activity) also occurs in Jurassic units (Netoff, 2002). Both records of seismic disturbance are consistent with the Mesozoic tectonic setting of southern Utah, during which time the Sevier orogenic (fold-thrust) belt was developing to the west.
Despite the uncertain timing of sand injection, it appears to have occurred at least several million years after deposition, based on the biostratigraphic constraints of overlying Jurrasic units. Deposition of the Carmel Formation, for example, is estimated at ~170–164 Ma. Sandstone injectites sourced from the Carmel Formation cut the overlying Entrada Formation, which was in place by 161 Ma. Thus a minimum of ~3 million years passed between deposition and injection. So how did the Carmel sandstones remain unconsolidated for such a period of time?
Salt: geological Tupperware
Evaporite layers, which are impermeable, cap the Carmel Formation locally and could have served as an extremely effective seal during burial. They would also prevent circulation of meteoric water to the buried sandstone. Not only would the Carmel sandstones become overpressured, but pore waters would lack the ions and oxidation state necessary for cementation to proceed. In passive margin sequences, the geothermal gradient is also typically low, so sediments must be buried more deeply than normal to reach a given temperature. Thus quartz cementation would not have occurred before the evaporite seal was broken during burial.
Geophysicist Glenn Morton has similarly commented on the arguments of Roth (1992). He correctly points out that cementation is not simply a function of age, and cites examples from personal experience where deeply buried sediments are still unconsolidated—some below well cemented strata! I will expand on his reasoning later on, with a closer look at cementation processes.
I do not mean to suggest that sand injectites at Kodachrome Basin are not mysterious formations—even counterintuitive on some level. These incredible statues defy tangible experience, and even challenge some very old geological dogmas. But they are not, after all facts are considered, inconsistent with the accepted timing and origin of geological strata. On the contrary, a greater challenge remains to those that believe these injectites formed during or after the Flood, while still unlithified, and yet cemented well enough in the time since the Flood to be exposed as weather-resistant landforms today.
Ute Pass Fault and associated sand dikes near Manitou Springs, CO
Every summer, the picturesque, mountain town of Manitou Springs—located just west of Colorado Springs, CO—hosts a massive tourist population. In addition to the unbeatable scenery, unique shopping experience, and local dining outlets like the Wine Cellar (my personal ‘shout-out’), nearby geological attractions such as Garden of the Gods and Cave of the Winds attract visitors from across the country—myself included (in fact, I spent part of my honeymoon there)!
The structural history of Manitou Springs region is equally enticing. Over the past ~60 million years, the Ute Pass Fault (a high-angle reverse fault) has exposed the Mesoproterozoic Pike’s Peak Granite to the south of the town. Paleozoic and Mesozoic sedimentary rocks were upwarped during the Laramide Orogeny, and are now exposed along the Front Range (e.g. Garden of the Gods). Numerous sand dikes are also found within extensional fractures of the Pike’s Peak Granite. Austin and Morris (1986) note that most dikes are found in the hanging wall of reverse faults along the Front Range, and strike parallel to Laramide faults.
Sand dikes of the Front Range in Colorado are fundamentally different from examples I cited above. Rather than piercing upward into sedimentary strata, these dikes formed when unconsolidated sand moved downward to fill extensional fractures. Nonetheless, sand dikes associated with the Ute Pass Fault are incredible examples of soft-sediment deformation (i.e. remobilization of unconsolidated sand), and are worth exploring further.
Young-Earth Creationists lead a GSA field trip to the Front Range
William Hoesch and other young-Earth geologists led a field trip at the Geological Society of America annual meeting held in Denver last year (Ross et al., 2010; abstract available here). They argued that the Cambrian Sawatch Sandstone injected into Pike’s Peak granite, which was fractured during the Laramide Orogeny and thrust on top of the Cambrian sandstone, some 430 million years after deposition. How did it turn out? One sympathetic spectator noted:
…a bunch of PhD creationist geologists led a field trip for the premier, annual secular geology meeting. I was there on that trip…and it was like music to my ears to have 16 PhD geologists stumped.
Austin and Morris (1986) originally advanced the argument that the timing and distribution of the sandstone dikes challenged the conventional geologic timescale. Following Kost (1984), they determined the Cambrian Sawatch Formation (~500 Ma) to be the sediment source based on similarities in textural and compositional maturity (although grains within the sand dikes were better sorted and cemented by hematite, rather than dolomite).
Most peculiar about the sand dikes is that they intrude older igneous and metamorphic rocks (Harms, 1965). Thus extensional faulting (pulling apart) of the crystalline rock was necessary for injection to take place, rather than failure of an overlying seal or cap rock. If the timing of fault formation can be constrained, however, so can the timing of sand injection.
Most injectites are found within proximity to the Ute Pass Fault, a dominantly Laramide structure, and so the timing of injection has been argued to be Cretaceous or later (less than 65 Ma) by Austin and Morris (1986). But if injection occurred as a result of Laramide movement on the Ute Pass Fault, one must explain how sandstone could remobilize after more than 430 million years of burial.
An unrealistic timeline: burial history of the Sawatch Formation
Although no geologist would suggest that lithification is simply a function of time (e.g. Selley, 1998), the proposed 430 million-year time gap of Austin and Morris (1986) would constitute a daunting challenge to the conventional age assignments. The Cambrian Sawatch Formation is not simply old, but it has since been buried by more than 2 miles of sediment. Moreover, there is no impermeable cap rock that would cause overpressuring or prevent circulation of diagenetic fluids.
Austin and Morris (1986) are correct about one thing: the Sawatch Formation could not have remained unlithified until the Laramide Orogeny, unless we are hopelessly mistaken about the age of either event. But the assertion that deposition and injection all took place during or shortly after the Flood is not the only alternative hypothesis. In fact, that scenario is falsified rather easily.
Genetic link between the Sawatch and Fountain formations
The Fountain Formation, also exposed near Manitou Springs, was deposited between the Late Pennsylvanian and Early Permian (Sweet and Sorreghan, 2010). Though dominantly sandstone, the unit is stratigraphically complex, characterized by numerous shallowing-upward cycles. Lithologies range from fine-grained mud, silt and sand to coarse, pebble conglomerates. Sweet and Sorreghan (2010) interpreted both marine and terrestrial depositional environments, and concluded that deposition took place in a fan-delta system, in which uplift to the west drove progradation of sediments toward the marine basin that covered the modern Great Plains.
Based on the geometry of the Fountain Formation, along with clast-size distribution, Sweet and Sorreghan (2010) also concluded that cyclic deposition of the Fountain Formation was driven by movement along the ancestral Ute Pass Fault, during uplift of the ancestral Rocky Mountains. While the Ute Pass Fault exposed near Manitou Springs today is a Laramide feature, the region has been tectonically active since the Cambrian (Myrow et al., 2003).
Conglomerate facies of the Fountain Formation provide further evidence for this depositional model. Sweet and Sorreghan (2010) used petrography to identify earlier Paleozoic clasts within the Fountain Formation, including weathered pebbles from the Sawatch Formation. In other words, the ancestral Ute Pass Fault, also a reverse fault, exposed older Paleozoic rocks as the Fountain Formation was being deposited to the northeast.
The occurrence of Sawatch-sourced pebbles in the Fountain Formation has significant implications for the timing of sand injection, since we may conclude that emplacement of the sand injectites occurred after the deposition of the Sawatch Formation (496 Ma), but prior to deposition of the Fountain Formation (~305 Ma). Moreover, the Sawatch Formation had to be lithified—at least on the upthrown block—before it could erode into pebbles and be deposited in conglomerates of the Fountain Formation. Sand injection did not occur during the Laramide Orogeny, because the Sawatch Formation was already lithified by the late Middle Paleozoic, more than 200 million years earlier.
Syntectonic deposits in a Flood model?
This sedimentological constraint constitutes a major challenge to Austin and Morris’ interpretation of the geologic history, since they must regard both the Sawatch and Fountain formations as Flood deposits. How did the Sawatch Formation lithify within less than a year? And if it did, then how was it injected into the Pikes Peak Granite later in the Flood, during ‘Laramide’ movement along the Ute Pass Fault? Austin and Morris thus face the same challenge they raise, and their interpretation of the sand dikes is simply not tenable in light of all geological data.
But the question still remains: when did sand injection occur? And how did it happen? Not considering the paleogeography and seismic history of the region, Austin and Morris (1986) glanced over the answer in their original paper:
Some workers…recognize the fundamental impossibility of keeping the Sawatch Sandstone…unlithiﬁed and deeply buried for 430 million years until the Laramide Orogeny…These workers tend to negate the important ﬁeld relationships and suggest that the dikes were actually intruded in the Cambrian while the Sawatch Sandstone was unconsolidated. Evidence of Cambrian or Ordovician tectonics of a magnitude able to open up extension fractures hundreds of feet wide, however, has not been found on the Ute Pass Fault. (emphasis added)
Austin and Morris (1986) thus ruled out the possibility that sand injection occurred in the early Paleozoic (Cambrian/Ordovician) because 1) sand dikes are found along the Ute Pass Fault—a Cenozoic structure; and 2) they believe that only the Laramide Orogeny was powerful enough to form the wide extensional fractures now hosting the Cambrian sand. But there are a few fatal flaws in this line of reasoning.
Tectonic blunders in the arguments of Austin and Morris
Sand injection could not have taken place during the Cenozoic, because uplift of the modern Rocky Mountains was driven by contractional deformation—namely, the Laramide Orogeny. The Ute Pass Fault is a reverse fault, which forms when rocks are compressed together, but sand dikes occur within extensional faults. In the latter case, rocks are pulled apart, so the tectonic features are mutually exclusive. Austin and Morris‘ suggestion that Laramide tectonism was “of sufﬁcient magnitude to open up the large extension fractures” is blatantly contradicted by the field evidence they had already cited. A more parsimonious conclusion is that extensional faulting occurred early in the Paleozoic (Cambrian–Ordovician), allowing for sand injection. Sand dikes were then exposed by uplift and erosion, driven by tectonic contraction, during the Cenozoic.
Austin and Morris (1986) argue that “the coincidence of the dikes along the Ute Pass Fault, a proven Laramide structure, cannot be accidental…”—and they are right. So why should sand dikes be found in proximity to and strike along Laramide faults if they were not formed at the same time? One could answer this question by a simple experiment. All you need to do is take a hammer to a brick, so that it cracks from top to bottom. Then, use a vice to squeeze the fractured brick together until the pieces break and move past each other. As you might expect, the brick will break along already formed fractures (i.e. where it is already weak).
In geological systems, this phenomenon is known as reverse-reactivation of normal faults (e.g. Kelly et al., 1999). During periods of tectonic extension, normal faults and extensional fractures form. Later, when the same rocks undergo compression, reverse faults form preferentially along older fault planes. This process not only explains the association of early Paleozoic sand dikes with Cenozoic reverse faults (namely why sand dikes run parallel to the Laramide Ute Pass Fault) and the high angle of the Ute Pass Fault (in contrast to a low-angle thrust fault), but also solves the apparent time gap of Austin and Morris (1986).
In the citation above, Austin and Morris state that “evidence of Cambrian or Ordovician tectonics…has not been found on the Ute Pass Fault.” I would argue, however, that the sand dikes are themselves evidence of Cambro-Ordovician tectonics! Although most offset on the Ute Pass Fault occurred during the Laramide Orogeny, the fault zone is primarily a Paleozoic structure that also produced thick, syntectonic deposits during the Pennsylvanian (Sweet and Soreghan, 2010) and was simply reactivated in the latest Mesozoic to early Cenozoic.
Austin and Morris (1986) accused earlier workers (e.g. Kost, 1984) of ignoring vital field evidence to save the old-Earth paradigm, but after closer examination, it appears Austin and Morris are guilty the same to support their own claims. In fact, Kost (1984) also used paleomagnetic data from the sand dikes to argue for an early Paleozoic sand injection, but these data were conveniently overlooked.
Overall, sand injectites near Manitou Springs are not evidence for a faulty geologic timescale, as suggested by Austin and Morris (1986). Reinterpretation of the depositional and structural history of the Front Range on a ~5,000 year timeline would create countless geological problems in an effort to solve one or two problems that do not actually exist. In fact, it does not even solve this one or two!
Appendix: Cementation of sandstone
All clastic sedimentary rocks lose both porosity and permeability with depth. Understanding this phenomenon is crucial to the oil industry, since these characteristics may determine whether or not an oil reserve is recoverable. Selley (1998) notes that 1) the geothermal gradient, and 2) the pressure regime are the primary factors controlling cementation during burial.
The sandstone layers that sourced the Kodachrome Basin sand pipes and the clastic dikes near Manitou Springs were deposited in a passive margin and intracratonic setting, respectively. In both cases, the geothermal gradient and sedimentation rate are relatively low, implying that sediments could remain unconsolidated for a very long time.
Cementation also depends on the ion composition and oxidative state of pore waters. Since silica is relatively insoluble at low temperature and neutral pH, sandstone cementation does not occur until deep burial unless ample groundwater is allowed to circulate through the sediments. In some cases, particularly near faults, oxygen-poor waters with high amounts of dissolved iron are introduced to porous sandstones that are already saturated with oxygen-rich, meteoric water. Iron is insoluble in oxidative environments, so the result is a hematite-cemented sandstone, such as in Kodachrome Basin State Park.
In other cases, carbonate-rich waters may circulate down into porous sandstone. Since acidity is lost in the process, the carbonate ions precipitate within pore spaces of the sand, forming carbonate cements. The Cambrian Sawatch Formation was cemented by dolomite, apparently sourced from the overlying Ordovician limestone/dolomite. Sand dikes along the Ute Pass Fault, however, are cemented with hematite—evidence of hydrothermal fluid interaction.
The importance of oil in cementation
If hydrocarbons migrate through unconsolidated or poorly-cemented sandstone, they may prevent further cementation, or even dissolve certain cements already in place (namely hematite). In the American southwest, white, bleached horizons in otherwise red sandstone cliffs reflect this very process. Thus the prevalence of sand injectites as hydrocarbon reservoirs is not entirely coincidental.
Early charges of hydrocarbons are sometimes responsible for exceptionally high porosity and permeability in sandstones. The Coalinga Oil Field of California, for example, yielded far more oil that its counterpart field at Kettleman Dome, because cementation was prevented by an early hydrocarbon charge in the former. If sand injectites were to lack a sufficient seal to preserve hydrocarbons, however, microbially mediated degradation of the oil could lead to rapid carbonate cementation in oil-bearing injectites as they are exhumed (Jonk et al., 2005). Thus many sand injectites and sand pipes are exposed today as weather-resistant structures.
Austin, S.A., and Morris, J.D., 1986, Tight Fold and Clastic Dikes as Evidence for Rapid Deposition and Deformation of Two Very Thick Stratigraphic Sequences, in Walsh, R.E., Brooks, C.L., Crowell, R.S. (editors), Proceedings of the First International Conference on Creationism, Pittsburgh, p. 3–13.
Dixon, R.J., Schoﬁeld, K., Anderton, R., Reynolds, A.D., Alexander, R.W.S., Williams, M.C., Davies, K.G., 1995, Sandstone diapirism and clastic intrusion in the Tertiary
submarine fans of the Bruce-Beryl Embayment, Quadrant 9, UKCS, in Hartley, A.J.,
Prosser, D.J. (editors), Characterisation of deep-marine clastic systems: Geological Society of London Special Publication, v. 94., p. 77–94.
Harms, J.C., 1965, Sandstone Dikes in Relation to Laramide Faults and Stress Distribution in the Southern Front Range, Colorado: Geological Society of America Bulletin, v. 76, p. 981–1002.
Hurst, A., Cartwright, J.A., Duranti, D., Huuse, M., Nelson, M., 2005, Sand injectites: an emerging global play in deep-water clastic environments: Petroleum Geology Conference Series, v. 6, p. 133–144.
Hurst, A., Scott, A., Vigorito, M., 2011, Physical characteristics of sand injectites: Earth-Science Reviews, v. 106, p. 215–246.
Huuse, M., Jackson, C.A., Van Rensbergen, P., Davies, R.J., Flemings, P.B., Dixon, R.J., 2010, Subsurface sediment remobilization and fluid flow in sedimentary basins: an overview: Basin Research, v. 22, p. 342–360.
Jonk, R., Hurst, A., Duranti, D., Parnell, J., Mazzini, A., Fallick, A.E., 2005, Origin and timing of sand injection, petroleum migration, and diagenesis in Tertiary reservoirs, south Viking Graben, North Sea: American Association of Petroleum Geologists, v. 89, p. 329–357.
Kelly, P.G., Peacock, D.C.P., Sanderson, D.J., McGurk, A.C., 1999, Selective reverse-reactivation of normal faults, and deformation around reverse-reactivated faults in the Mesozoic of the Somerset coast: Journal of Structural Geology, v. 21, p. 493–509.
Kost, L. S., 1984, Paleomagnetic and petrographic study of sandstone dikes and the Cambrian Sawatch Sandstone, east ﬂank of the southern Front Range, Colorado: Master’s Thesis, University of Colorado, Colorado, 173 p.
Myrow, P.M., Taylor, J.F., Miller, J.F., Ethington, R.L., Ripperdan, R.L., Allen, J., 2003, Fallen arches: Dispelling myths concerning Cambrian and Ordovician paleogeography of the Rocky Mountain region: Geological Society of America Bulletin, v. 115, p. 695–713.
Netoff, D., 2002, Seismogenically induced fluidization of Jurassic erg sands, south-central Utah: Sedimentology, v. 49, p. 65–80.
Ross, J.A., Peakall, J., Keevil, G.M., 2011, An integrated model of extrusive sand injectites in cohesionless sediments: Sedimentology, v. 58.
Ross, M.R., Hoesch, W.A., Austin, S.A., Whitmore, J.H., Clarey, T.L., 2010, Garden of the Gods at Colorado Springs: Paleozoic and Mesozoic Sedimentation and Tectonics: Geological Society of America Field Guides, v. 18, p. 77–93.
Roth, A., 1992, Clastic Pipes in Dikes in Kodachrome Basin: Origins, v. 19, p. 44–48.
Scott, A., Vigorito, M., Hurst, A., 2009, The process of sand injection: internal structures and relationships with host strata (Yellowbank Creek Injectite Complex, California, U.S.A.): Journal of Sedimentary Research, v. 79, p. 568 – 583.
Selley, R.C., 1998, Elements of Petroleum Geology: Academic Press, San Diego, 470 p.
Vigorito, M., and Hurst, A., 2010, Regional sand injectite architecture as a record of pore-pressure evolution and sand redistribution in the shallow crust: insights from the Panoche Giant Injection Complex, California: Journal of the Geological Society of London, v. 167, p. 889–904.
Featured image photo credit.