Soft-Sediment Deformation and Dune Collapse in the Navajo Sandstone Colby Ford

Loma Linda University TheScholarsRepository@LLU: Digital Archive of Research, Scholarship & Creative Works

Loma Linda University Electronic Theses, Dissertations & Projects

12-2015 Soft-sediment Deformation and Dune Collapse in the Navajo Sandstone Colby Ford

Follow this and additional works at: http://scholarsrepository.llu.edu/etd Part of the Geology Commons, Hydrology Commons, and the Sedimentology Commons

Recommended Citation Ford, Colby, "Soft-sediment Deformation and Dune Collapse in the Navajo Sandstone" (2015). Loma Linda University Electronic Theses, Dissertations & Projects. 334. http://scholarsrepository.llu.edu/etd/334

This Thesis is brought to you for free and open access by TheScholarsRepository@LLU: Digital Archive of Research, Scholarship & Creative Works. It has been accepted for inclusion in Loma Linda University Electronic Theses, Dissertations & Projects by an authorized administrator of TheScholarsRepository@LLU: Digital Archive of Research, Scholarship & Creative Works. For more information, please contact [email protected].

LOMA LINDA UNIVERSITY School of Medicine in conjunction with the Faculty of Graduate Studies

______

Soft-sediment Deformation and Dune Collapse in the Navajo Sandstone

Colby Ford

______

A thesis submitted in partial satisfaction of the requirements for the degree Master of Science in Geology

______

December 2015

, Chairperson Kevin E. Nick, Associate Professor of Geology

Gerald Bryant, Director, Colorado Plateau Field Institute, Dixie State University

H. Paul Buchheim, Professor of Geology

iii ACKNOWLEDGEMENTS

Funding and facilities for this project provided by the Department of Earth and

Biological Sciences, Loma Linda University, and by a grant from the Geological Society of America. Special thanks to Paul Buchheim (Loma Linda University) for input during the development of this project, and to Rosmarie Bisquera for field assistance.

iv CONTENT

Approval Page ...... iii

Acknowledgements ...... iv

List of Figures ...... vii

List of Tables ...... viii

Abstract ...... ix

Chapter

1. Expanded Introduction and Background ...... 1

Introduction ...... 1

Previous Research ...... 4

History of eolian research ...... 4 History of SSD research...... 6

References ...... 9

2. Architectural Evidence of Dune Collapse in the Navajo Sandstone, Zion National Park, Utah ...... 15

Introduction ...... 21

Purpose of this report ...... 16

Methods...... 19 Results ...... 22

Sedimentary and deformational structures...... 30

Discussion ...... 37

Initial conditions in the dune field ...... 38 Early liquefaction and incipient collapse ...... 40 Dune Collapse ...... 41 Production of uneven bounding surface ...... 42 Re-establishment of normal dune migration ...... 44 Implications for the Horowitz dune collapse model ...... 47

Conclusions & future research ...... 48 References ...... 49

3. Expanded discussion ...... 52

Addressing Horowitz’ predictions ...... 52 Relationship to other projects underway...... 56 Conclusions ...... 56 References ...... 58

vi FIGURES

Figures Page

1. Schematic of Horowitz dune collapse model...... 3

2. Schematic of Horowitz dune collapse model, based on outcrop ...... 17

3. Map showing study location ...... 20

4. Overhead view of study location ...... 22

5. Generalized stratigraphic column ...... 23

6. Digital elevation maps of bounding surfaces ...... 25

7. Cross-section of dune collapse complex ...... 27

8. Photo of “Keystone” outcrop ...... 28

9. Photo showing contorted chert nodules ...... 29

10. Thin section of the fractured end of a chert nodule ...... 30

11. Photo showing evidence of vertical movement of fluidized material...... 31

12. Photo of intense deformation in the North-central portion of the outcrop ...... 32

13. Photo of brecciated and displaced carbonate mudstone laminae ...... 33

14. Stereonet diagram of mapped faults ...... 35

15. Photo of a clastic injection pocket ...... 36

16. Photo of a small shear feature causing offset across the upper bounding surface ...... 37

17. Conceptual dune collapse model part A ...... 40

18. Conceptual dune collapse model part B...... 42

19. Conceptual dune collapse model part C...... 44

20. Conceptual dune collapse model part D ...... 46

vii TABLES

Tables Page

1. List and descriptions of named units in outcrop ...... 24

viii ABSTRACT OF THE THESIS

Soft-sediment Deformation and Dune Collapse in the Navajo Sandstone

Colby Ford

Master of Science Graduate Program in Geology Loma Linda University, December 2015 Dr. Kevin E Nick, Chairperson

The Canyon Overlook Trail of Zion National Park follows an outcrop of Navajo

Sandstone, which displays a uniquely well-exposed assemblage of features associated with failure of the lee face of a large eolian dune, and run-out over an expanse of interdune sediments downwind of that bedform. Exposed features include dramatic folds in the interdune succession and a stacked series of thrust sheets incorporating both interdune and overlying dune deposits. Thrust surfaces display consistent strikes, parallel to those of undeformed foresets, and incorporate zones of brittle failure and fluid deformation, including folds overturned in the direction of foreset dip. These features correspond to predictions made by the Horowitz (1982) model of dune collapse, formulated from less fortuitously exposed architectures in the Navajo Sandstone. Unlike the Horowitz (1982) model, however, this site preserves distinct indications that the bulk of deformed material accumulated above the level of the contemporary interdune surface, in an aggradational succession.

Paleotopographic reconstruction, based on preserved facies relationships at this site, indicates the presence of a large dune, partially encroaching upon a well-developed wet interdune succession, made up of two half-meter carbonate mud layers, separated by a meter of medium-grained sand. Trapping of pore water pressure between these mud

ix layers during liquefaction reduced shear strength in this interval, facilitating the collapse of the lee face of the upwind dune into the interdune area, and transmitted resultant shear forces to distal portions of the interdune expanse, in the shallow subsurface. Shear failure developed along bedding planes in the horizontally laminated carbonate muds, which provided both lubrication of the shear surfaces and structural support for the preservation of coherent thrust sheets during production of an imbricated succession of shear zones in the toe portion of the slump. Individual shear surfaces exposed in this outcrop extend for up to 50 m along strike and dip north up to 55°. Upturned mud layers in the toe of the slump resisted deflation, promoting preservation of an irregular interdune topography, over which the reorganized dune ultimately advanced.

Keywords: Dune collapse, soft-sediment deformation, Navajo Sandstone, interdune, paleohydrology.

x CHAPTER 1

EXPANDED INTRODUCTION AND BACKGROUND

Introduction

The Navajo Sandstone is a Formation of the Glenn Canyon Group, and represents an eolian dune system comparable in extent to the modern Sahara Desert. The Navajo

Sandstone records deposition from the beginning of the Pliensbachian through the mid-

Toarchian ages of the lower Jurassic. The Aztec Sandstone and Nugget Sandstone are equivalents in other parts of the Utah-Idaho trough, but the erg attained its maximum thickness in the central part of the basin, what is now Southern Utah. Seismic activity, which triggered abundant soft-sediment deformation (SSD) in the Navajo, was probably generated by ongoing continental flexure generating subsidence in the Utah Idaho trough

(P. A. Allen, Verlander, Burgess, & Marc Audet, 2000).

Soft-sediment deformation reveals a great deal about the internal structure and paleoenvironmental setting of eolianites. As eolian deposition naturally produces poorly- consolidated deposits, only two other factors must be supplied to drive liquefaction: pore fluids and trigger mechanisms. While gasses may play a role in relatively rare liquefaction features (Lohse, Rauhe, Bergmann, & Van Der Meer, 2004), liquefaction typically requires the sand to be fully saturated with water. In deposits with as much volume as the Navajo sandstone, saturation of sands near the surface requires a significant input of water into the basin. Triggering mechanisms vary, but for large-scale or widespread synchronous liquefaction as is common in the Navajo Sandstone, seismic activity is usually invoked.

1 Dune collapse, first suggested by Horowitz (1982), has been heavily cited in the literature as a model for the production of large-scale SSD in eolianites. In his model

(refer to Figure 1), high water table conditions in the dune field would manifest as a high liquefaction potential in the interdune spaces. As the lee slope of the upwind dune is always much steeper than the stoss slope of the downwind dune, an asymmetrical pore fluid pressure field would have been produced during liquefaction events, once the granular framework has been destabilized, with a high pressure gradient under the toe of the upwind dune. Therefore, liquefaction will tend to undermine and destabilize the upwind dune, potentially causing a slump. When the upwind dune slumps, material will be driven into the interdune, driving profound deformation of the interdune sediments, and producing a distinct “thrust ramp” morphology. In Horowitz’s model, the deposit would then deflate to the water table, producing a flat upper truncation for the deformation feature.

Figure 1: Re-drawn from Horowitz (1982). The original dune collapse model. A: Liquefied sand in the interdune area is acted upon by an asymmetric pore fluid pressure field, with material under the lee slope of the upwind dune experiencing greater overburden pressure due to its steeper slope. B: The liquefaction undermines the lee slope, causing it to collapse into the interdune area. A distinct slump morphology is formed. C: The deposit erodes to the water table, producing a flat upper truncation of the deformation feature.

3 A number of problems are, however, apparent in Horowitz’s original model. The first is that the sediments and deformation structures Horowitz studied were indistinct, lacked morphological diversity, and provided little outcrop exposure; leading him to propose a basic dune collapse model which would leave few distinct physical traces, other than profound deformation of primary sedimentary structures. The second, more subtle, and more serious problem arises from his assertion that dune collapse structures were truncated above by flat bounding surfaces, created by deflation to the water table.

At the time that Horowitz was writing, this was a fairly widely-accepted mechanism for creation of the bounding surfaces in eolian deposits (Stokes, 1968). However, subsequent research has shown that the typical eolian bounding surfaces are time-transgressive surfaces created by deflation in the interdune space as dunes migrate, whereas deflation of the entire dune field to the water table would only take place during a major erosional episode (Blakey, Peterson, & Kocurek, 1987; Brookfield, 1977; Kocurek, 1981; Kocurek

& Havholm, 1993; Kocurek & Hunter, 1986). Therefore, for dune collapse to be generally followed by deflation to the water table would imply that the dune collapse process is intimately linked with the formation of major diastems. However, this is not appropriate for what is essentially a mass-wasting process, which is fairly mundane in every way except its paleoenvironmental context.

Previous Research

History of Eolian Research

The North American craton, and especially the region of the Western United

States known as the Colorado Plateau, includes the greatest volume of eolian

4 accumulation in the entirety of the rock record. Deposition of wind-blown sand came in two major pulses, one in the upper Paleozoic, and a second in the Jurassic.

Accumulations preserved from these episodes are divided between as many as three dozen named, well-exposed, and well-studied formations. In fact, the eolianites of the

Colorado Plateau, and especially the Navajo Sandstone, is probably the most thoroughly studied accumulation of rock in the world (Blakey et al., 1987).

The first serious description of the area’s geology is a fruit of the Whipple expedition, which was dispatched by the Secretary of War in 1853 to find a route for a railroad extending from the Mississippi river to Los Angeles (Marcou, 1856). These and other early pioneers, regarded the arid, canyon-riddled Colorado Plateau as an obstacle rather than a land of opportunity. It was a barrier to westward expansion. The nation finally began to recognize the region’s harsh beauty after the daring and well-publicized expeditions of John Wesley Powell (Powell, 1875). Serious study of the region was aided greatly by the conservationist movement of the late 19th and early 20th century, especially the efforts of Presidents Roosevelt and Taft who established National Monuments surrounding the Grand Canyon and what would later be called Zion Canyon.

Herbert Gregory’s survey (1917) provided many of the initial paleoenvironmental interpretations for the rock units of the Colorado Plateau, interpretations which would be considered authoritative for the first half of the 20th century (Gregory, 1917b; Van Loon

& Brodzikowski, 1987). The reinterpretation of the Navajo Sandstone as a subaqueous deposit by Freeman & Visher (1975) triggered a second wave of investigation into eolian processes, architecture, and structures. While their reinterpretation was not ultimately accepted by the scientific community, William Freeman and Glenn Visher’s paper

5 pointed out weaknesses in old interpretive criteria which new research worked to correct

(Blakey et al., 1987; Brookfield, 1977; Doe & Dott, 1980; Freeman & Visher, 1975;

Hunter, 1977; Kocurek, 1991; Kocurek & Hunter, 1986; Rubin & Hunter, 1982). The recognition of eolian units’ value as petroleum reservoirs, coupled with the development of the discipline of sequence stratigraphy, fueled continued eolian research and paleogeographic interpretation from the 1990s through the present (P. A. Allen et al.,

2000; Blakey & Ranney, 2008; Kocurek & Havholm, 1993; Parrish & Falcon-Lang,

2007; Rodriguez-Lopez, Clemmensen, Lancaster, Mountney, & Veiga, 2014; Romain &

Mountney, 2014; Rowe, Loope, Oglesby, Van der Voo, & Broadwater, 2008).

History of SSD Research

SSD features were first described by some of the pioneers of modern geology, including Darwin and Lyell (Darwin, 1851; Lyell, 1838). Their significance, however, was not appreciated at that early stage. While their presence was noted, they were interpreted as oddities, superficial features, or simply bad outcrop. Their presence was not noted specifically during early explorations of the Colorado Plateau deposits

(Marcou, 1856; Powell, 1875), but clastic pipe features were noted in the Entrada

Sandstone around the turn of the century, when these formations were being formalized

(Newsom, 1903). Early, fundamental work on the Navajo Sandstone included little interpretation of SSD features, although they did not go unnoticed (Gregory, 1917a,

1950). The first use of SSD as paleoenvironmental indicators in the Navajo Sandstone came from Kiersch (1950), who used extensive liquefaction features to propose climatically-driven episodes of high water table conditions.

6 Field and laboratory studies on SSD in sandstones, especially in the early 1970s, dramatically improved the interpretive power of deformation features, especially by establishing the effects of moisture on cohesion and shear resistance (J. R. L. Allen &

Banks, 1972; Kiersch, 1950; Lowe & LoPiccolo, 1974; McKee & Bigarella, 1972;

McKee, Douglass, & Rittenhouse, 1971; McKee, Reynolds, & Baker Jr., 1962; Rettger,

1935; Stokes, 1968). In 1975, the body of evidence for the influence of water on the

Navajo Sandstone, led to the high profile and controversial re-interpretation of the

Navajo as a marginal marine sand wave deposit by Freeman and Visher (Folk, 1977;

Freeman & Visher, 1975; Picard, 1977). While the immediate result of Freeman and

Visher’s paper was merely a number of largely philosophical defenses of previous interpretive criteria, it also spawned a flurry of research which lasted into the mid-1980s.

(J. R. L. Allen, 1986; Doe & Dott, 1980; Glennie & Buller, 1983; Horowitz, 1982;

Hunter & Kocurek, 1986; Kocurek & Dott, 1981; Kocurek & Hunter, 1986) Of particular importance for SSD were two papers, one by T. R. Doe & R. H. Dott Jr. (1980) and the other by Daniel Horowitz (1982). The former established both the terminology and mathematics of SSD in unconsolidated sands, supporting an eolian interpretation of the

Navajo Sandstone (Doe & Dott, 1980). The latter provided a process by which large- scale SSD features could be generated without major marine incursions (Horowitz,

1982). These articles have become keystones in the study of eolianite SSD.

The conceptual model advanced by Horowitz ultimately stems from the study of liquefaction-induced slumping of earth-fill dams. Horowitz himself references the work of H. Bolton Seed particularly heavily, especially Seed’s study of the collapse of the

Lower San Fernando Dam (Seed, 1979a, 1979b; Seed & Idriss, 1971). Although this dam

7 was an earth and rockfill dam, and its materials barely resemble those of an eolian sand dune, its collapse due to liquefaction produced features similar in their morphology to the

SSD features Horowitz was studying. Furthermore, the dam was comparable in size to the best estimates of typical Navajo dunes, and its downstream slope (which collapsed due to liquefaction) was an angle of repose surface, similar to the lee slope of a sand dune.

As important as this phase of research was for making a useful interpretive tool out of SSD features, the advent of Sequence Stratigraphy and new approaches in basin analysis led to a shift away from the use of SSD as a general paleoenvironmental indicator, and toward actualistic studies. Most recent work falls into two categories: geotechnically- and seismically-focused experimental analysis of liquefaction, (P. A.

Allen et al., 2000; Hilbert-Wolf, Simpson, Simpson, Tindall, & Wizevich, 2009; Holzer

& Youd, 2007; Hunter & Kocurek, 1986; Jones & Omoto, 2000; Kuhn, 2005; Loope,

Elder, Zlotnik, Kettler, & Pederson, 2013; Montenat, Barrier, Ott d′Estevou, & Hibsch,

2007; Moretti, Alfaro, Caselles, & Canas, 1999; Nichols, Sparks, & Wilson, 1994;

Obermeier, 1996; Owen, 1996; Owen, Moretti, & Alfaro, 2011) and, more commonly, actualistic studies of specific features or classes of features over limited aerial and stratigraphic extents (Blakey, 1996; Bromley, 1992; Bryant, 2011; Bryant & Miall, 2010;

Bryant, Monegato, & Miall, 2013; Chan & Bruhn, 2014; Eisenberg, 2003; Fernandes, de

Castro, & Basilici, 2007; Glennie & Buller, 1983; Hurst & Glennie, 2008; Huuse,

Shoulders, Netoff, & Cartwright, 2005; Kocurek & Hunter, 1986; Loope et al., 2013;

Loope & Rowe, 2003; Loope, Rowe, & Joeckel, 2001; Mahaney et al., 2004; Moretti,

8 2000; Mountney & Thompson, 2002; Netoff, 2002; Netoff & Shroba, 2001; Parrish &

Falcon-Lang, 2007; Plaziat, Aberkan, & Reyss, 2006; Seiler & Chan, 2008).

9 References

Allen, J. R. L. (1986). Earthquake magnitude-frequency, epicentral distance, and soft- sediment deformation in sedimentary basins. Sedimentary Geology, 46, 67-75.

Allen, J. R. L., & Banks, N. L. (1972). An interpretation and anlysis of recumbent-folded deformed cross-bedding. Sedimentology, 19, 257-283.

Allen, P. A., Verlander, J. E., Burgess, P. M., & Marc Audet, D. (2000). Jurassic giant erg deposits, flexure of the United States continental interior, and timing of the onset of Cordilleran shortening. Geology, 28(2), 159. doi:10.1130/0091- 7613(2000)28<159:jgedfo>2.0.co;2

Blakey, R. C. (1996). The Kayenta-Navajo transition: a critical environmental change in Lower Jurassic rocks of the Colorado Plateau. In M. Morales (Ed.), The Continental Jurassic (pp. 477-478). Flagstaff, Arizona: Museum of Northern Arizona.

Blakey, R. C., Peterson, F., & Kocurek, G. (1987). Synthesis of late Paleozoic and Mesozoic eolian deposits of the Western Interior of the United States. Sedimentary Geology(56), 3 - 125.

Blakey, R. C., & Ranney, W. (2008). Ancient landscapes of the Colorado Plateau: Grand Canyon Association.

Bromley, M. (1992). Topographic inversion of early interdune deposits, Navajo Sandstone (Lower Jurassic), Colorado Plateau, USA. Sedimentary Geology(80), 1 - 25.

Brookfield, M. E. (1977). The origin of bounding surfaces in ancient aeolian sandstones. Sedimentology(24), 303 - 332.

Bryant, G. (2011). Outcrop Studies of Soft-sediment Deformation Features in the Navajo Sandstone. University of Toronto.

Bryant, G., & Miall, A. (2010). Diverse products of near-surface sediment mobilization in an ancient eolianite: outcrop features of the early Jurassic Navajo Sandstone. Basin Research, 22(4), 578-590. doi:10.1111/j.1365-2117.2010.00483.x

Bryant, G., Monegato, G., & Miall, A. (2013). An example of liquefaction-induced interdune sedimentation from the early Jurassic Navajo Sandstone, USA. Sedimentary Geology, 297, 50-62. doi:10.1016/j.sedgeo.2013.09.001

Chan, M. A., & Bruhn, R. L. (2014). Dynamic liquefaction of Jurassic sand dunes: processes, origins, and implications. Earth Surface Processes and Landforms. doi:10.1002/esp.3539

10 Darwin, C. (1851). Geological observations on coral reefs, volcanic islands, and on South America: being the geology of the voyage of the Beagle, under the command of Captain Fitzroy R. N., during the years 1832 to 1836. London,: Smith, Elder & co.

Doe, T. W., & Dott, R. H. (1980). Genetic significance of deformed cross bedding--with examples from the Navajo and Weber Sandstones of Utah. Journal of Sedimentary Petrology, 50(3), 793-812.

Eisenberg, L. (2003). Giant stromatolites and a supersurface in the Navajo Sandstone, Capitol Reef National Park, Utah. Geology, 31(2), 111. doi:10.1130/0091- 7613(2003)031<0111:gsaasi>2.0.co;2

Fernandes, L. A., de Castro, A. B., & Basilici, G. (2007). Seismites in continental sand sea deposits of the Late Cretaceous Caiuá Desert, Bauru Basin, Brazil. Sedimentary Geology, 199(1-2), 51-64. doi:10.1016/j.sedgeo.2005.12.030

Folk, R. L. (1977). Stratigraphic Analysis of the Navajo Sandstone: A Discussion. Journal of Sedimentary Petrology, 47(1), 475 - 483.

Freeman, W. E., & Visher, G. S. (1975). Stratigraphic Analysis of the Navajo Sandstone. Journal of Sedimentary Petrology, 45(3), 651 - 668.

Glennie, K. W., & Buller, A. T. (1983). The Permian Weissliegend of NW Europe: the partial deformation of aeolian dune sands caused by the Zechstein transgression. Sedimentary Geology, 35, 43-81.

Gregory, H. E. (1917a). Geology of the Navajo country, a reconnaissance of parts of Arizona, New Mexico, and Utah. Washington,: Govt. print. off.

Gregory, H. E. (1917b). Geology of the Navajo Country: A Reconnaissance of Parts of Arizona, New Mexico, and Utah. Washington D.C.: Government Printing Office.

Gregory, H. E. (1950). Geology and geography of the Zion Park region, Utah and Arizona. USGS.

Hilbert-Wolf, H. L., Simpson, E. L., Simpson, W. S., Tindall, S. E., & Wizevich, M. C. (2009). Insights into syndepositional fault movement in a foreland basin; trends in seismites of the Upper Cretaceous, Wahweap Formation, Kaiparowits Basin, Utah, USA. Basin Research, 21(6), 856-871. doi:10.1111/j.1365- 2117.2009.00398.x

Holzer, T. L., & Youd, L. T. (2007). Liquefaction, Ground Oscillation, and Soil Deformation at the Wildlife Array, California. Bulletin of the Seismological Society of America, 97(3), 961-976.

Horowitz, D. H. (1982). Geometry and origin of large-scale deformation structures in some ancient wind-blown sand deposits. Sedimentology(29), 155-180.

11 Hunter, R. E. (1977). Basic types of stratification in small eolian dunes. Sedimentology(24), 361-387.

Hunter, R. E., & Kocurek, G. (1986). An experimental study of subaqueous slipface deposition. Journal of Sedimentary Petrology, 56(3), 387-394.

Hurst, A., & Glennie, K. W. (2008). Mass-wasting of ancient aeolian dunes and sand fluidization during a period of global warming and inferred brief high precipitation: the Hopeman Sandstone (late Permian), Scotland. Terra Nova, 20(4), 274-279. doi:10.1111/j.1365-3121.2008.00817.x

Huuse, M., Shoulders, S. J., Netoff, D. I., & Cartwright, J. (2005). Giant sandstone pipes record basin-scale liquefaction of buried dune sands in the Middle Jurassic of SE Utah. Terra Nova, 17(1), 80-85. doi:10.1111/j.1365-3121.2004.00587.x

Jones, A. P., & Omoto, K. (2000). Towards establishing criteria for identifying trigger mechanisms for soft-sediment deformation: a case study of Late Pleistocene lacustrine sands and clays, Onikobe and Nakayamadaira Basins, northeastern Japan. Sedimentology, 47, 1211-1226.

Kiersch, G. A. (1950). Small-scale structures and other features of Navajo sandstone, northern part of San Rafael Swell, Utah. AAPG Bulletin, 34(5), 923-942.

Kocurek, G. (1981). Significance of interdune deposits and bounding surfaces in aeolian dune sands. Sedimentology(28), 753-780.

Kocurek, G. (1991). Interpretation of ancient eolian sand dunes. Annual Review of Earth and Planetary Sciences(19), 43-75.

Kocurek, G., & Dott, R. H. (1981). Distinctions and uses of stratification types in the interpretation of eolian sand. Journal of Sedimentary Petrology, 51, 579-595.

Kocurek, G., & Havholm, K. G. (1993). Eolian sequence stratigraphy-a conceptual framework MEMOIRS-AMERICAN ASSOCIATION OF PETROLEUM GEOLOGISTS (pp. 393-409).

Kocurek, G., & Hunter, R. E. (1986). Origin of Polygonal Fractures in Sand, Uppermost Navajo and Page Sandstones, Page, Arizona. Journal of Sedimentary Petrology, 56(6), 895 - 904.

Kuhn, G. G. (2005). Paleoseismic features as indicators of earthquake hazards in North Coastal, San Diego County, California, USA. Engineering Geology, 80(1-2), 115- 150. doi:10.1016/j.enggeo.2005.04.006

Lohse, D., Rauhe, R., Bergmann, R., & Van Der Meer, D. (2004). Granular physics: creating a dry variety of quicksand. Nature, 432, 689-690.

12 Loope, D. B., Elder, J. F., Zlotnik, V. A., Kettler, R. M., & Pederson, D. T. (2013). Jurassic earthquake sequence recorded by multiple generations of sand blows, Zion National Park, Utah. Geology, 41(10), 1131-1134. doi:10.1130/g34619.1

Loope, D. B., & Rowe, C. M. (2003). Long-Lived Pluvial Episodes during Deposition of the Navajo Sandstone. Papers in the Earth and Atmospheric Sciences. Retrieved from http://digitalcommons.unl.edu/geosciencefacpub/211

Loope, D. B., Rowe, C. M., & Joeckel, R. M. (2001). Annual monsoon rains recorded by Jurassic dunes. Nature, 412, 64 - 66.

Lowe, D. R., & LoPiccolo, D. (1974). The characteristics and origins of dish and pillar structures. Journal of Sedimentary Petrology, 44(2), 484-501.

Lyell, C. (1838). Elements of geology. London: John Murray.

Mahaney, W. C., Milner, M. W., Netoff, D. I., Malloch, D., Dohm, J. M., Baker, V. R., . . . Komatsu, G. (2004). Ancient wet aeolian environments on Earth: clues to presence of fossil/live microorganisms on Mars. Icarus, 171(1), 39-53. doi:10.1016/j.icarus.2004.04.014

Marcou, J. (1856). Resume and field notes. Washington D.C.: US Government.

McKee, E. D., & Bigarella, J. J. (1972). Deformational structures in Brazilian coastal dunes. Journal of Sedimentary Petrology, 42(3), 670-681.

McKee, E. D., Douglass, J. R., & Rittenhouse, S. (1971). Deformation of Lee-Side Laminae in Eolian Dunes. Geological Society of America Bulletin, 82, 359-378.

McKee, E. D., Reynolds, M. A., & Baker Jr., C. H. (1962). Experiments on intraformational recumbent folds in crossbedded sand. (165).

Montenat, C., Barrier, P., Ott d′Estevou, P., & Hibsch, C. (2007). Seismites: An attempt at critical analysis and classification. Sedimentary Geology, 196(1-4), 5-30. doi:10.1016/j.sedgeo.2006.08.004

Moretti, M. (2000). Soft-sediment deformation structures interpreted as seismites in middle-late Pleistocene aeolian deposits (Apulian foreland, southern Italy). Sedimentary Geology, 135, 167-179.

Moretti, M., Alfaro, P., Caselles, O., & Canas, J. A. (1999). Modelling seismites with a digital shaking table. Tectonophysics(304), 369-383.

Mountney, N. P., & Thompson, D. B. (2002). Stratigraphic evolution and preservation of aeolian dune and damp/wet interdune strata: an example from the Triassic Helsby Sandstone Formation, Cheshire Basin, UK. Sedimentology(49), 805-833.

13 Netoff, D. I. (2002). Seismogenically induced fluidization of Jurassic erg sands, south- central Utah. Sedimentology(49), 65-80.

Netoff, D. I., & Shroba, R. R. (2001). Conical sandstone landforms cored with clastic pipes in Glen Canyon National Recreation Area, southeastern Utah. Geomorphology(39), 99-110.

Newsom, J. F. (1903). Clastic dikes. Geological Society of America Bulletin, 14(1), 227- 268.

Nichols, R. J., Sparks, R. S. J., & Wilson, C. J. N. (1994). Experimental studies of the fluidization of layered sediments and the formation of fluid escape structures. Sedimentology(41), 233-253.

Obermeier, S. F. (1996). Use of liquefaction-induced features for paleoseismic analysis - An overview of how seismic liquefaction features can be distinguished from other features and how their regional distribution and properties of source sediment can be used to infer the location and strength of Holocene paleo-earthquakes. Engineering Geology(44), 1-76.

Owen, G. (1996). Experimental soft-sediment deformation: structures formed by the liquefaction of unconsolidated sands and some ancient examples. Sedimentology, 43, 279-293.

Owen, G., Moretti, M., & Alfaro, P. (2011). Recognising triggers for soft-sediment deformation: Current understanding and future directions. Sedimentary Geology, 235(3-4), 133-140. doi:10.1016/j.sedgeo.2010.12.010

Parrish, J. T., & Falcon-Lang, H. J. (2007). Coniferous trees associated with interdune deposits in the Jurassic Navajo Sandstone formation, Utah, USA. Palaeontology, 50(4), 829 - 843.

Picard, M. D. (1977). Stratigraphic Analysis of the Navajo Sandstone: A Discussion. Journal of Sedimentary Petrology, 47(1), 475 - 483.

Plaziat, J.-C., Aberkan, M. h., & Reyss, J.-L. (2006). New late Pleistocene seismites in a shoreline series including eolianites, north of Rabat (Morocco). Bulletin de la Société Géologique de France, 177(6), 323-332.

Powell, J. W. (1875). Exploration of the Colorado River and Its Canyons. New York: Cover Press.

Rettger, R. E. (1935). Experiments on soft-rock deformation. AAPG Bulletin, 19(2), 271- 292.

Rodriguez-Lopez, J. P., Clemmensen, L. B., Lancaster, N., Mountney, N. P., & Veiga, G. D. (2014). Archean to Recent aeolian sand systems and their sedimentary record: Current understanding and future prospects. Sedimentology, 61, 1487-1534.

14 Romain, H. G., & Mountney, N. P. (2014). Reconstruction of three-dimensional eolian dune architecture from one-dimensional core data through adoption of analog data from outcrop. AAPG Bulletin, 98(1), 1-22. doi:10.1306/05201312109

Rowe, C. M., Loope, D. B., Oglesby, R. J., Van der Voo, R., & Broadwater, C. E. (2008). Inconsistencies Between Pangean Reconstructions and Basic Climate Controls. Science, 318, 1284-1286.

Rubin, D. M., & Hunter, R. E. (1982). Bedform climbing in theory and nature. Sedimentology(29), 121-138.

Seed, H. B. (1979a). Considerations in earthquake-resistant design of earth and rockfill dams. Geotechnique, 29(3), 215-263.

Seed, H. B. (1979b). Considerations in the earthquake-resistant design of earth and rockfill dams. Geotechnique, 29(3), 215-263.

Seed, H. B., & Idriss, I. M. (1971). Simplified procedure for evaluation of soil liquefaction potential. Journal of the Soil Mechanics and Foundations Division, 97(9), 1249-1273.

Seiler, W. M., & Chan, M. A. (2008). A Wet Interdune Dinosaur Trampled Surface in the Jurassic Navajo Sandstone, Coyote Buttes, Arizona: Rare Preservation of Multiple Track Types and Tail Traces. Palaios, 23(10), 700-710. doi:10.2110/palo.2007.p07-082r

Stokes, W. L. (1968). Multiple parallel-truncation bedding planes--a feature of wind- deposited sandstone formation. Journal of Sedimentary Petrology, 38(2), 510- 515.

Van Loon, A. J., & Brodzikowski, K. (1987). Problems and progress in the research on soft-sediment deformations. Sedimentary Geology, 50, 167-193.

15 CHAPTER 2

ARCHITECTURAL EVIDENCE OF DUNE COLLAPSE IN THE NAVAJO

SANDSTONE, ZION NATIONAL PARK, UTAH

Introduction

The outcrop of Navajo Sandstone along the Canyon Overlook Trail, in Zion

National Park, records the signature of an extraordinary soft-sediment deformation (SSD) event during the deposit’s early history, enabling an unusually detailed reconstruction of dynamics in the original depositional environment. Architectural relationships are clearly defined and well-displayed, providing an excellent testing ground for various hypotheses regarding deformation controls and dynamics. This report focuses on the applicability of a model proposed by Horowitz (1982) that is commonly invoked to explain Navajo SSD.

The Navajo Sandstone was deposited in the Utah-Idaho Trough (Imlay, 1980;

Peterson, 1994), a foreland (Bjerrum and Dorsey, 1995) or back-arc (Lawton, 1994) basin which stretched from what is now Northern Arizona to Southern Idaho, in the Early

Jurassic, prior to establishment of the Western Interior Basin. In its southern extent, the

Navajo erg interacted with fluvial systems of the Kayenta Fm, which maintained high water table conditions in that region throughout much of the Navajo depositional history

(Blakey, 1996; Loope & Rowe, 2003). The Navajo erg hosted dunes hundreds of meters high, with wavelengths of kilometers. Where high water table conditions prevailed, interdune spaces were often occupied by ephemeral or perennial ponds, which trapped fine sand and mud in lenticular accumulations decimeters thick, across km-scale expanses

(Kocurek & Havholm, 1993). In many cases, high evaporation rates also led to the deposition of carbonates in the more persistent interdune ponds (Bryant & Miall, 2010;

16 Parrish & Falcon-Lang, 2007). The resultant lenses of horizontally stratified sediments are frequently associated with major SSD features, especially in the southern and eastern reaches of the Navajo Sandstone, where they have been studied extensively in the last several decades (Allen, Verlander, Burgess, & Marc Audet, 2000; Blakey, Peterson, &

Kocurek, 1987; Bryant, 2011; Chan & Bruhn, 2014; Doe & Dott, 1980; Eisenberg, 2003;

Horowitz, 1982; Kocurek & Hunter, 1986; Loope, Elder, Zlotnik, Kettler, & Pederson,

2013; Parrish & Falcon-Lang, 2007).

Purpose of This Report

Horowitz (1982) proposed earthquake-induced dune collapse as a general mechanism for the generation of large-scale deformation features in eolian dune environments (Figure 2).

Figure 2: Horowitz’ dune collapse model, redrawn from Horowitz (1982).

18 His model reinforced Doe and Dott’s (1980) comprehensive discussion of Navajo SSD in refuting the claims of Freeman and Visher (1975) and other critics of the eolian interpretation of Navajo deposition (Jordan, 1965; Marzolf, 1969; Stanley et al., 1971;

Visher, 1971) that widespread SSD in the Navajo constitutes evidence for a subaqueous origin.

Horowitz’s dune collapse model was bolstered by geotechnical particulars of modern liquefaction-induced slumps and slides, especially the 1971 collapse of the

Lower San Fernando Dam (Horowitz, 1982; Seed, 1979). This engineering analysis provides a compellingly pragmatic and quantitative dimension to the dune collapse model. Perhaps because of this, Horowitz’ (1982) paper has been referenced liberally in most subsequent literature on eolian SSD (Chan & Bruhn, 2014; Eisenberg, 2003; Hurst

& Glennie, 2008; Plaziat, Aberkan, & Reyss, 2006). However, this model, as originally proposed, poses several problems.

First, it has been cited as a general mechanism for the production of large-scale

SSD features in eolian dune accumulations, even though the particulars of many documented cases show no linkage to topography (see Bryant et. al. in this volume). The occurrence of deformation features whose extents surpass the likely wavelength of eolian bedforms, whose fold orientations are inconsistent with horizontal translation of material, or whose intensity attenuates upward into undeformed foresets, precludes surface processes as the primary driver of deformation morphology (Bryant & Miall, 2010). A second, related problem, is that the original dune collapse model provides few physical criteria by which to positively identify the dynamics of dune collapse and distinguish them from other SSD processes. Finally, Horowitz implied that dune collapse should be

19 followed by deflation of the deposit to the water table. This was in line with a hypothesis for the generation of eolian bounding surfaces (Stokes, 1968) which was widely held at the time. However, as shown in Figure 2, this would not result in aggradation of the succession (Bryant et al., this volume) and a typical (first-order) boundary between successive deposits (Kocurek & Havholm, 1993), but would produce a more extensive diastem, sometimes referred to as a “super-surface” (Kocurek, 1988).

This report documents a more distinctive example of dune collapse than has previously been available, provides more robust physical criteria for the recognition of dune collapse features, and demonstrates that this process operated during aggradational phases of the accumulation history.

Methods

The Canyon Overlook Trail provides ready access to most of the outcrop details featured in this report. A cross-section of the interval in question is visible at a distance from the switchbacks on Highway 9 below the tunnel. Access to the outcrop itself is easiest from the Canyon Overlook trailhead at the Eastern end of the tunnel (Figure 3).

Figure 3: Map showing the location of the study site

While the complex erosional pattern of this part of the canyon makes it difficult to place isolated outcrops in panoramic context, it does provide a great deal of surface area.

High-precision GPS mapping of the upper and lower bounding surfaces throughout the outcrop facilitated the creation of elevation maps in ArcGIS, using the Kriging interpolation method. These elevation maps and fault planes were in turn transformed

21 into a hypothetical cross-section illustrating the internal architecture of the deformation complex.

Wherever possible, we measured the attitude of visible faults, noting the textures present along each fault surface. Faults in this deposit fall into two general categories: deformation bands (Fossen, 2007) and synsedimentary shear surfaces. Deformation bands are typically vertical or sub-vertical (parallel with the main regional jointing pattern), show very limited offset, and develop a cataclastic texture (green planes on Figure 14).

Two of these deformation bands which exhibited clear grain fracturing were recorded for descriptive purposes (see Figure 14), but all others were disregarded. Synsedimentary faults, on the other hand are not associated with grain fracturing due to shearing, and as such do not have weathering properties much different from the cross-bedded Navajo

Sandstone.

The rocks themselves were characterized using petrographic microscopy and x- ray diffractometry, demonstrating the texture and mineralogy at various parts of the deposit. For the most part, these details do not contribute to a study of the dune collapse process, and will be omitted from this report. However, in one particular case, petrography addresses the timing of cement formation in the interdune deposit.

Descriptive data, such as facies associations, internal architecture, and stratigraphic relationships are most readily communicated using annotated photos. The outcrop has a great deal of three-dimensional complexity, making it a prime candidate for new three-dimensional mapping techniques. A computer model of one particularly relevant outcrop was produced using a free web-based photogrammetry application,

22 123D Catch by Autodesk. This model will be included in the supplementary content of this article.

Figure 4: Map of the research area. A – A’ is the line of the cross-section (Figure 7), and the red outline approximates the current lateral boundaries of the interdune succession, with the majority of the deformation complex visible in the central and Western portions. However, intense deformation, including major shear zones, is visible along the walls of the small canyon to the southeast of the trail, which will be discussed in conjunction with Figure 18.

Results

The outcrop area is oblong with its longer dimension perpendicular to the local paleocurrent direction (local measurements average 181°). The western edge is eroded through the deformed interval, and the upwind extent is buried in the canyon wall.

Therefore, its original extent can only be estimated. However, good outcrop constraints on the maximum extents, downwind and perpendicular to the paleocurrent, suggest that

23 the affected area approached 0.5 km2. Deformation is most intense in the northwestern portion of the outcrop, nearer to the projected detachment faults, and centers on the interdune deposits. Figure 4 shows an overhead view of the research area, with the approximate boundaries of deformation within the study section.

Figure 5: Generalized stratigraphic column of the deformation complex, provided as a generalized cross section and key to the relative positions of named units within the deformation complex. There is significant vertical exaggeration for clarity and abbreviations on the right are explained in Table 1.

24 Figure 5 gives a generalized stratigraphic section of the deposit. Key stratigraphic units and boundaries within the deformation architecture are identified in Table 1.

Table 1: List and descriptions of identified units and relevant surfaces in this outcrop.

SDD Subsequent Dune Deposits. Fine- to medium-grained cross-bedded sandstone, no visible deformation. A single major cross-bed set overlies all other outcrop features in this location. XSS2 Fine- to medium-grained cross-bedded sandstone, no visible deformation. A 1-2 m cross-bed set overlying a small portion of the deformation complex, onlapping the deformation complex’s upper bounding surface (UBS). Only visible in a small portion of the outcrop. XSS1 Fine- to medium-grained cross-bedded sandstone, with rare examples of small (cm-scale) deformation features. A 1-2 m cross-bed set overlying a small portion of the deformation complex, onlapping the deformation complex’s upper bounding surface (UBS). Only visible in a small portion of the outcrop. UBS The irregular surface truncating the major deformation feature. DSS Fine- to medium-grained sandstone. Sedimentary structures vary, with some areas entirely massive, and other areas containing relict cross-bedding, often rotated from typical cross-bed dips, and crossed by many small shear surfaces. CM2 Sandy carbonate mudstone, pink to burgundy in color, and in some areas containing chert laminae or nodules. In its undeformed state, it has cm-scale laminations. IDSS Intermediate Sandstone. Massive fine- to medium-grained sandstone, separating the two carbonate mudstone layers. CM1 Sandy carbonate mudstone, pink to burgundy in color, cm-scale horizontal lamination. Unlike CM2, no chert nodules have been located in this unit. LBS The surface below CM1, which truncates the previous dune deposit (PDD). Generally flat, although some minor local topography is apparent in the GPS data. PDD Previous Dune Deposits. Fine- to medium-grained cross-bedded sandstone, limited deformation. This is a single set of crossbeds that underlies the deformed interval at this location.

25 The deposit’s lower boundary (LBS) is a planar, crossbed set boundary (first- order surface) at the base of the interdune succession, typically dipping at 2-4°, representing some combination of the dune climb angle and local structural dip. The interdune succession is composed of two ~50cm carbonate mudstone layers (CM1, CM2) separated by approximately one meter of massive sandstone (IDSS). Where there is material between the top of the interdune succession and the upper truncation of the deposit (DSS), this material is the fine- to medium-grained sandstone typical of the

Navajo, sometimes with preserved cross-bedding, sometimes entirely massive. Where cross-bedding is preserved (as in the extreme western edge of the deposit, at the overlook), it is crossed by major shear zones. In several places, DSS is visibly rotated, with a final orientation of cross-bedding varying from shallowly north-dipping (against paleocurrent) to oversteepened south-dipping. Where material is rotated, it is always associated with shear zones. The upper truncation surface (UBS) is highly variable, with its most dramatic topography associated with major shear zones.

Figure 6: Digital elevation maps of the lower (left) and upper (right) bounding surfaces, the points from which they were derived, and the line along which the cross section (Figure 7) was drawn. The cross-section line was drawn through the area with the highest point density, in the direction of the paleocurrent.

Interpolation of the GPS data points yielded the DEM overlay in Figure 6.

Because of access restrictions and outcrop along which the relevant surfaces are not exposed, data coverage is not optimal for all areas of the outcrop. The resultant digital elevation maps shown are fairly accurate in areas with a high GPS point density, although they may be highly erroneous in the extrapolated regions (such as the Southeast corner of the LBS map, Figure 6 left). The Kriging interpolation method tends not to yield smooth

27 surfaces, so sharp breaks or corners in the maps above and the cross section below are likely to be artifacts. The general trend, however, matches field observations of the trend of the UBS and LBS, including the locations and attitudes of outcropping faults.

In general, the LBS climbs fairly steadily from north to south, the slight structure

(evident to the right of Figure 7) appears to be real, as it coincides with high GPS point density, although it is very gradual and not noticeable in the field. The gradual offset may be due to shear along the recent regional jointing pattern discussed in the methods section of this report. The shape of the UBS is very apparent in both this cross-section and field observations, with the UBS climbing steeply (relative to the LBS), reaching a high point near the middle of the cross-section, and then falling gradually.

Figure 7: Cross-section of the slump feature created from interpolated upper and lower bounding surfaces. 2:1 vertical exaggeration. Green lines indicate the upper and lower bounding surfaces created from the interpolation of GPS data. The carbonate mudstone layers (burgundy lines) are interpreted where the lines are dotted, due to vegetation cover and limited outcrop exposure. Where lines are solid, the presence of carbonate mudstones derived from CM1 and/or CM2 are confirmed, and their dip is accurate (corrected for vertical exaggeration).

28 Material just above the upper bounding surface (XSS1, XSS2, or SDD, depending on location) is, for the most part, made up of normal cross-bedded sandstone with no evidence of shearing or other disruption (Figure 8). Cm-scale deformation features, however, can be found in some places, which will be described later in this section. The upper bounding surface, wherever visible, is flat or irregular and truncates deformation structures.

Figure 8: The "Keystone" outcrop. Green line traces the UBS, and the red line traces the shear surface responsible for this particular morphology. Note that the material above the upper bounding surface is undeformed.

29 The deformation complex exhibits two general lithologies. First is well-sorted, well-rounded fine- to medium-grained sandstone, which is carbonate cemented and often hematite stained. This is the typical lithology of cross-bedded Navajo sandstone. A variation of this same fine- to medium-grained sandstone is visible in IDSS and the clastic injection feature discussed later in this section. This IDSS material is distinguishable in hand sample by its massive texture, and is petrographically distinct due to significantly greater porosity, and near total dissolution of feldspar grains. The second major lithology is a fine-grained, dark red to pink carbonate mudstone, with significant

(10-50%) silt and mud-sized siliciclastic contribution, associated with interdune deposits.

Chert, as lensoidal nodules or mm-scale laminae, is common is some samples. In one location, chert laminae are fractured and contorted in association with deformation of

CM2 (Figure 9). Thin sections prepared from this area indicate that subsequent diagenetic processes affected the fracture surfaces of the chert laminae (Figure 10).

Figure 9: Typical interdune material, in this case CM2. Fresh surfaces are darker red in color. Note the fractured and contorted chert nodules (white streaks).

Figure 10: Thin section showing the fractured end of a chert lamina. Note ingrowth of carbonate minerals along the bottom (bottom of figure) and the fractured end (left of figure) of the lamina.

Sedimentary and Deformational Structures

Primary sedimentary structures found in the study area are those typical for the

Navajo: horizontal stratification in the interdune deposits and large-scale cross-bedding otherwise. These primary structures, however, are often highly modified by deformation structures. Dish and pillar structures are present in the extreme southwest of this outcrop area (Figure 11), but such simple deformation structures are rare otherwise. More often, deformation is profound and results in a combination of partial or total effacement of

32 primary sedimentary structures (see DSS in Figure 12), major shear surface development

(see Figure 12), and brecciation and displacement of resistant materials (see CM1 in

Figure 13).

Figure 11: Evidence of fluidization and vertical movement of material just below the interdune succession, near the eastern boundary of the dune collapse feature. No evidence of the main slump is visible here, and deformation seems to have little or no component of horizontal translation.

Figure 12: Intense deformation in the North-central portion of the outcrop, photo taken facing east. Blue lines trace shear surfaces with a normal sense of slip, red lines trace those with reverse sense of slip. Fine black lines added to emphasize drag folding. Green lines trace the upper and lower truncation surface of the deformation complex. Black boxes obscure hikers on the trail.

It is worth noting that the greatest intensity of deformation coincides with the greatest thickness of the interdune succession, and therefore likely the center of the interdune deposit. Deformation intensity attenuates toward the lateral margins of the deposit (Figure 13) which displays cm-scale offsets of material and little development of shear zones.

Figure 13: Brecciation and displacement of the upper layers of CM1 into DSS. Major inclusions derived from CM1 outlined in black for clarity. Note parting of material along bedding planes in CM1, another example of deformation taking advantage of existing bedding.

Synsedimentary shear surfaces typically originate in or near the interdune mud layers below, are truncated above by the upper bounding surface, and show no change in texture on the shear surface due to grain fracturing (orange planes on Figure 14). The most important criterion for deciding which faults are synsedimentary is this truncation.

Major synsedimentary faults never cross or offset the UBS, except for one case, which will be discussed later, in relation to Figure 16. Most often, smaller synsedimentary faults

35 are visible only because they offset cross-beds, and typically have no micro-scale textural difference from the rest of the Navajo Sandstone. These minor shear zones are not included in Figure 14, as they are typically too numerous to measure accurately, and their attitudes reflect cm-scale deformation due to settling or drag folding, rather than the overall stress field of the dune collapse event.

Major synsedimentary faults, since they typically originate in the carbonate mudstone beds, often smear fine material along their shear planes. This grain size variation can then lead to weathering characteristics different from typical Navajo

Sandstone, although they are usually less resistant, rather than more resistant like the regional joints. Their attitude is highly variable, but synsedimentary faults in the area typically strike at approximately 300°, with dips typically varying between 20° and 35°, although in one case a shear surface dips at over 55° (Figure 14).

Figure 14: Stereonet diagram of mapped shear surfaces on the Canyon Overlook Trail. Synsedimentary shear surfaces marked as orange planes and poles, with more recent jointing marked as green planes and poles. Note the distinctly bimodal distribution.

In the north-central portion of the outcrop, a small (about 2 m wide, 1 m tall) pocket of brecciated material intrudes into a massive sandstone matrix and is interpreted to be a clastic injection feature. Thin sections collected from this feature show that the brecciated blocks are very similar to material in the carbonate mudstone beds, although with signs of different diagenetic history. The matrix of this pocket matches the lithology of IDSS, as previously discussed.

Figure 15: A meter-scale clastic injection pocket. Clasts preserve internal lamination in some cases, and are sourced from interdune mud layers. The matrix is a massive medium sand with extremely high porosity. This photo is taken facing west from the North-central portion of the outcrop. Figure 12 was taken from the same spot, facing across a small ravine.

While the material above the deformation complex is generally not deformed in any way, material just above the upper bounding surface is subject to minor (cm scale) deformation. This deformation always takes the form of shearing crossing the UBS, with up to 10 cm of offset, and with the faults typically developing along cross-beds in the

DSS.

Figure 16: A small deformation feature crossing the upper bounding surface. Shearing seems to have taken place along bedding planes in the material below, and is only visible because it involves the horizontally bedded dune plinth material above the bounding surface.

In addition, in some marginal areas of the outcrop (in both cases, just before the deformation complex pinches out), dinosaur tracks are visible in plan view and cross- section.

39 Discussion

Dune collapse, as a primarily surficial process, must be understood in its paleoenvironmental context. Dune size, morphology, wavelength, and other paleotopographic factors, in conjunction with the type and strength of the triggering mechanism exercised primary control on the extent of deformation. However, paleohydrology was the crucial variable governing susceptibility and sediment response to deforming forces, due to variations in the cohesiveness of uncemented sand as a function of moisture content. Paleohydrology also controlled the type and distribution of interdune sediments. Appropriate consideration of these factors is essential to the proper application of dune collapse models to the occurrence of soft-sediment deformation features in the Navajo Sandstone (See Bryant et al., in this volume).

Initial Conditions in the Dune Field

Estimating the original dune size is difficult for two main reasons. The first is the simple fact that a direct quantitative relationship between cross-bed thickness and dune height remains elusive, although various authors have speculated about possible links

(Blakey et al., 1987; Romain & Mountney, 2014; Rubin & Hunter, 1982). An accepted rule of thumb is that, under consistent conditions of bedform climb, eolian dune fields produce cross-bed sets about 10% as thick as the original dune height. The second difficulty is particular to this outcrop. Due to the production of an irregular bounding surface, the intense deformation of this crossbed set throughout much of its exposed extent, and the peculiarities of stratification produced during re-equilibration of the altered bedform to regional controls on dune size and shape, the metrics that might

40 normally be used to estimate dune size cannot be readily applied, here. In the larger area where this feature occurs, cross-bed sets are about 10m thick, suggesting that dunes in the area would have been approximately 100m in height. The extent of the carbonate mudstones in this outcrop point to an interdune area approximately 500m along paleocurrent, which in turn suggests a dune wavelength of about 1 kilometer. Trough cross-stratification apparent within some portions of the local succession indicates the probability of along-crest variations in dune form.

Variation in water table height supported episodic standing water conditions in the interdune area, driving carbonate deposition in pulses. The origin of the sand between the two carbonate mudstone layers (IDSS) cannot be determined on the basis of primary structures, since these were obliterated during the deformation event; however, there is no evidence for wholesale injection of this unit during deformation. The progradational relationships between CM1, IDSS, and CM2 suggest that they should be considered as a single interdune succession, developed during an extended history of dune migration under high water table conditions. Based on analogy to common patterns of interdune sedimentation in the Navajo, IDSS most likely originated as a horizontally stratified interdune deposit, which accumulated under moist, rather than flooded conditions, during a temporary decline in the elevation of the water table, between episodes of carbonate sedimentation. During the liquefaction event, pore fluids trapped in IDSS, between the carbonate mud layers would have facilitated efficient transfer of energy to distal portions of the interdune succession, transmitting lateral forces produced by the collapsing dune.

Fluid pressure in this unit may already have been elevated, prior to liquefaction, due to

41 continuity with a water table mound in the upwind dune (See Bryant et al., in this volume).

As DSS, the deformed cross-bedded sandstone, is present over much of the outcrop area, it is possible that there was a full cross-bed set (up to 10 m thick) deposited atop the interdune succession before the dune collapse event. This is not required by the observed architectural relationships, however, which are equally consistent with the identification of DSS as the basal portion of the collapsed dune. Further discussion will model the simpler case where the interdune succession is at the surface.

Early Liquefaction and Incipient Collapse

As the upwind dune advanced over the interdune succession, the additional overburden pressure under the lee slope would have primed particularly energetic fluidization processes in that region. It may be that, once a seismic trigger was supplied, pore fluids punched through the upper carbonate mudstone layer into the toe of the dune, contributing to its destabilization. This initial pulse of fluid escape may have produced features very similar to those described by Loope (2013) from two locations very near the

Canyon Overlook Trail, but higher in the section. If the incipient stages of dune collapse produced a multitude of small fluid escape features, these were obliterated by higher energy processes derived from the forces generated by dune collapse and the higher pressures developed in the confined aquifer between carbonate mud layers.

Figure 17: Conceptual model of this dune collapse feature, part A. Incipient collapse of a large dune migrating over a stacked interdune succession. Note that fluid escape which did not reach the surface would likely have been preserved as clastic injection pockets (Figure 15).

It is unlikely that all fluid escape pathways formed near enough to the toe of the dune or extended far enough vertically, to break out to the surface. In these cases, rather than forming outflow deposits, fluids would have injected the relatively coarse grainflow deposits, smoothing the pressure gradient, and leaving mobilized sediment just above the interdune. This, perhaps, explains the clastic injection feature pictured in Figure 15.

Dune Collapse

Consequent to liquefaction of the interdune sediments, pore fluid would have been expelled to the surface and into the plinth of the upwind dune. This reduced the

43 shear strength of the slope-buttressing mass by reducing the number and stability of grain-to-grain contacts. This destabilizing effect may have been augmented by local changes in slope, produced by subsidence during reconsolidation, especially under the dune plinth. Upon failure at the toe, the lee slope of the dune detached and slumped downward, subparallel to crossbedding. The greatest intensity of deformation coincides with the maximum thickness of carbonate mudstones, which reveals the role played by muds in this instance of dune collapse. While not triggering the event, two aspects of the interdune succession provided a unique susceptibility to the collapse of this particular dune. First, shearing across the interdune surface was accommodated by slip along bedding planes in the mud layers, and their fine sediment content lubricated the movement, facilitating the generation of multiple major synsedimentary faults (Figure

18). These stacked sediment wedges are rooted in a common decollement along the interdune succession, that buckled under compressive forces to produce discrete wedges.

They have no apparent connection to the production of multiple slump blocks in the dune face. Whether or not such slump blocks were produced is an open question at this site, since this upper portion of the slump is not exposed and probably was not preserved.

The second important link between the presence of the carbonate mudstones and the collapse of the dune is build-up of pore pressure between the impermeable muds during liquefaction. This would have accentuated grain framework disruption as fluid was forced to migrate laterally, and then escape more forcefully across the permeability barrier.

Figure 18: Conceptual model of this dune collapse feature, part B. The dune fails and slumps into the interdune area. Slumping is accommodated by the development of shear surfaces originating in bedding planes of the interdune muds. The liquefied interdune succession transmits the slump force throughout the interdune, causing small subsidiary shear surfaces to form in distal areas of the interdune.

Production of Uneven Bounding Surface

Once slumping had run its course, eolian transport conditions resumed primary control of the topography. It is likely that the slump resulted in a lowering and upwind migration of the dune crest, and locally altered aerodynamic conditions in favor of greater wind velocity over the failed dune (Kocurek, 1991). This would have readily mobilized dry sand from the dune itself, while more cohesive material resisted erosion. This cohesion was likely due to a combination of trapped moisture in fine sands, initial consolidation of muds, and early cementation. The presence of chert lamina, fractured during dune collapse, reveals that at least some early cement formation took place before

45 that event; although the other cements in the carbonate mudstones appear to result from multiple diagenetic phases.

Thus, carbonate mud layers thrust to high angles facilitated development of an irregular topography (with over 15 m of relief, see Figure 7) in the interdune. However, it is difficult to estimate the maximum amount of relief that was produced in the dune collapse event. UBS, like all other eolian bounding surfaces, is time-transgressive. The pile of slump material resulting from the dune collapse would have been exposed to wind erosion for decades or centuries (based on 1m/yr dune advance) before being covered by the next upwind dune, and the present irregularity of UBS likely underestimates the topographic relief of the slump pile immediately after the dune collapse event. Despite this, an average accumulation of 10m of this slumped material was preserved, similar to the expected amount of accumulation due to migration of a large dune in this area of the

Navajo. This emphasizes the fact that sediment accumulation in dune fields is not driven primarily by features of individual dunes, but rather by the process of bedform climb, which is in turn controlled by the interplay of sediment supply and current activity (see

Rubin & Hunter, 1982).

Figure 19: Conceptual model of this dune collapse feature, part C. Slump material which originated in the upper portions of the dune would have been relatively dry and unconsolidated, and would have eroded quickly. Interdune material, however, would have maintained greater competence in erosional conditions due to its different grain size, moisture content, and in some cases, early cementation. Thus, the erosional surface which formed was not a typical flat eolian bounding surface, but rather shows a distinct topography.

Re-establishment of Normal Dune Migration

The portion of the upwind dune which survived would have continued to advance during active seasons, although its decreased size, and the pile of slump deposits extending downwind onto the interdune, must have altered the local aerodynamic regime.

However this affected the rate of re-equilibration, it is clear that wind currents scavenged material for small dunes that migrated more rapidly in front of the collapsed dune. As shown in Figure 5, two smaller sets of cross-bedded material overlie the deformation complex upwind of the most prominent paleotopographic feature (Figure 7). This

47 indicates that the re-establishment of dune/interdune topography in equilibrium with the regional pattern proceeded through a more complex intermediate condition. This may be compared to the persistent defects in dune organization produced by episodic incursions of ephemeral streams into modern ergs, as in the Namib. The smaller crossbed sets also help to bracket the distance away from the lee slope of the collapsed dune represented at the study site. In order for downwind sediment transport to occur independent of the reactivation of the larger bedform, the location of these deposits must have been far enough downwind of that topography to allow for re-attachment of current flow to the substrate and organization of transported sand into dune-scale bedforms. This provides an additional indication that the locus of lee slope failure remains buried in the outcrop, north of the study site. At the same time, preservation of such dramatic paleotopography below the advancing dune suggests a limited time of exposure to erosive forces before burial by dune advance.

Sites such as this one provide rare windows into erg equilibration processes, which are poorly understood. These processes have received little attention, not only because the data required for longitudinal studies in modern environments are scant, but because suitable outcrops of the ancient record are scarce, and the necessary tools of architectural interpretation are inadequately developed.

Figure 20: Conceptual model of this dune collapse feature, part D. The final deposit, with undeformed material preserved above the irregular upper bounding surface.

Although the material above the upper bounding surface (SDD) is clearly not affected by the gross deformation of the dune collapse process, there are small (cm scale) deformation features in these deposits that do not generally appear in Navajo dune deposits. This suggests that the slumped material continued to consolidate as subsequent dunes migrated across its surface (See comparable indications in Bryant et al., 2013). In at least two places, minor subsidence has been accommodated through shearing along bedding planes in the otherwise undeformed dune (see Figure 16). These minor faults are not consistently aligned with the shear surfaces produced by the dune collapse event, but rather adopt the attitude of the cross-bed laminae along which they developed.

49 Implications for the Horowitz Dune Collapse Model

In order to accommodate the constraints of the Zion study site, the dune-collapse model of Horowitz (1982), requires overhaul in two specific ways. First, the idea that dune collapse will be followed by deflation to the water table must be replaced with a context of accumulation controlled by sediment budget, with unlimited accumulation space. Sediment moisture content provided subsidiary controls on stratification type and the preservation of interdune topography.

The second overhaul is a simple expansion of the diagnostic criteria used to recognize dune collapse. Five criteria are discussed below. First, as Horowitz (1982) wrote, there should be evidence of horizontal movement of material to suggest dune collapse as a model for any specific outcrop feature. Many SSD features in eolian deposits only provide indicators of liquefaction and vertical fluid escape, and should not be ascribed to dune collapse. Second, there should be clear evidence of a vertical truncation of the deformation feature. Dune collapse is, by definition, a surficial process, and should not be applied generically to features which attenuate upward into undeformed overlying deposits. Third, as the materials involved in liquefaction-induced dune collapse encompass the boundary between saturated and unsaturated conditions, there should be evidence of both fluid and brittle deformation styles. While it is possible for the material containing brittle deformation indicators to be lost to deflation or effaced by subsequent liquefaction, other explanations should be explored for outcrops devoid of indicators of brittle deformation and synsedimentary faulting. Fourth, in the particular case of slope failure, primary shear surfaces should be aligned with the dune crest. Since the lee face of a dune is its steepest component and foresets represent the most significant internal

50 discontinuity, dune collapse is not likely to take place in any other direction. Care should be taken in the application of this criterion, however, since deep-seated, extensive zones of liquefaction may more broadly compromise the substrate below entire bedforms, causing them to founder in response to less uniform controls (Bryant and Miall, 2010;

Bryant et al., in this volume). This highlights the importance of the final criterion for application of a dune-collapse model, which is that the scale of deformation should correspond to the scale of the dunes involved. In the case of the feature examined in this study, the dune collapse event affected the majority of the interdune space, stretching hundreds of meters. Smaller dunes with shorter wavelengths may be expected to produce smaller dune collapse features, but the size (and wavelength) of dunes, based on the best available parameters for topographic reconstruction (Rubin & Hunter, 1982), must be carefully incorporated into any model featuring dune collapse.

Conclusions & Future Research

The Canyon Overlook Trail outcrop provides less ambiguous details of dune collapse processes than have been available previously, allowing for more comprehensive analysis. This enables a more reliable interpretation of dune collapse features at other locations.

The Horowitz (1982) model suffers from narrow assumptions regarding paleohydrological controls on the development of the accumulation. It should not be used

51 as a general case to explain the occurrence of large-scale SSD features. However, the characterization of deformation mechanics it provides is confirmed by this study and should continue to provide a useful model for the production of a subset of SSD features in ancient eolianites.

52 References

Bjerrum, C.J. and R.J. Dorsey (1995) Tectonic controls on deposition of Middle Jurassic strata in a retroarc foreland basin, Utah-Idaho Trough, western interior, USA. Tectonics, vol. 14, p. 962-978.

Blakey, R. C., Peterson, F., & Kocurek, G. (1987). Synthesis of late Paleozoic and Mesozoic eolian deposits of the Western Interior of the United States. Sedimentary Geology(56), 3 - 125.

Bryant, G. (2011). Outcrop Studies of Soft-sediment Deformation Features in the Navajo Sandstone. University of Toronto.

Chan, M. A., & Bruhn, R. L. (2014). Dynamic liquefaction of Jurassic sand dunes: processes, origins, and implications. Earth Surface Processes and Landforms. doi:10.1002/esp.3539

Doe, T. W., & Dott, R. H. (1980). Genetic significance of deformed cross bedding--with examples from the Navajo and Weber Sandstones of Utah. Journal of Sedimentary Petrology, 50(3), 793-812.

Eisenberg, L. (2003). Giant stromatolites and a supersurface in the Navajo Sandstone, Capitol Reef National Park, Utah. Geology, 31(2), 111. doi:10.1130/0091- 7613(2003)031<0111:gsaasi>2.0.co;2

Fossen, Haakon, et al. "Deformation bands in sandstone: a review." Journal of the Geological Society 164.4 (2007): 755-769.

Horowitz, D. H. (1982). Geometry and origin of large-scale deformation structures in some ancient wind-blown sand deposits. Sedimentology(29), 155-180.

Hurst, A., & Glennie, K. W. (2008). Mass-wasting of ancient eolian dunes and sand fluidization during a period of global warming and inferred brief high

53 precipitation: the Hopeman Sandstone (late Permian), Scotland. Terra Nova, 20(4), 274-279. doi:10.1111/j.1365-3121.2008.00817.x

Imlay, R.W., 1980. Jurassic Paleobiogeography of the Conterminous United States in Its Continental

Setting. U.S. Geological Survey Professional Paper 1062, 134 p.

Jordan, W.M. (1965) Regional environmental study of the Early Mesozoic Nugget and Navajo Sandstones. Ph.D. dissertation, Wisconsin University

Kocurek, G. (1991). Interpretation of ancient eolian sand dunes. Annual Review of Earth and Planetary Sciences(19), 43-75.

Kocurek, G., & Havholm, K. G. (1993). Eolian sequence stratigraphy-a conceptual framework MEMOIRS-AMERICAN ASSOCIATION OF PETROLEUM GEOLOGISTS (pp. 393-409).

Kocurek, G., & Hunter, R. E. (1986). Origin of Polygonal Fractures in Sand, Uppermost Navajo and Page Sandstones, Page, Arizona. Journal of Sedimentary Petrology, 56(6), 895 - 904.

Kocurek, G. (1988) First-order and super bounding surfaces in eolian sequences; bounding surfaces revisited. In: Late Paleozoic and Mesozoic eolian deposits of the Western Interior of the United States. (Ed G. Kocurek), Sedimentary Geology, 56, 193-206.

Lawton, T.F., 1994, Tectonic setting of Mesozoic sedimentary basins, Rocky Mountain region, United States, in Caputo, M. V., Peterson, J. A., and Franczyk, K. J., editors, Mesozoic systems of the Rocky Mountain region, USA: Denver, Colorado, Rocky Mountain Section SEPM, p. 1–26.

Loope, D. B., Elder, J. F., Zlotnik, V. A., Kettler, R. M., & Pederson, D. T. (2013). Jurassic earthquake sequence recorded by multiple generations of sand blows, Zion National Park, Utah. Geology, 41(10), 1131-1134. doi:10.1130/g34619.1

Marzolf, J. (1969) Regional stratigraphic variations in primary features of the Navajo Sandstone, Utah. Abstracts with Programs - Geological Society of America, 25, 115.

Parrish, J. T., & Falcon-Lang, H. J. (2007). Coniferous trees associated with interdune deposits in the Jurassic Navajo Sandstone formation, Utah, USA. Palaeontology, 50(4), 829 - 843.

54 Peterson, F. (1994) Sand dunes, sabkhas, streams and shallow seas: Jurassic paleogeography in the southern part of the Western Interior Basin. In: Mesozoic Systems of the Rocky Mountain Region, USA (Ed J.A.P. Mario V. Caputo, and Karen J. Francsyk), pp. 233-272. SEPM (Society for Sedimentary Geology), Rocky Mountain Section, Denver, Colorado.

Romain, H. G., & Mountney, N. P. (2014). Reconstruction of three-dimensional eolian dune architecture from one-dimensional core data through adoption of analog data from outcrop. AAPG Bulletin, 98(1), 1-22. doi:10.1306/05201312109

Rubin, D. M., & Hunter, R. E. (1982). Bedform climbing in theory and nature. Sedimentology(29), 121-138.

Seed, H. B. (1979). Considerations in earthquake-resistant design of earth and rockfill dams. Geotechnique, 29(3), 215-263.

Stanley, K.O.; Jordan, W.M. and Dott, R.H., Jr. (1971) New hypothesis of Early Jurassic paleogeography and sediment dispersal for western United States. American Association of Petroleum Geologists Bulletin, 55, 10-19.

Stokes, W. L. (1968). Multiple parallel-truncation bedding planes--a feature of wind- deposited sandstone formation. Journal of Sedimentary Petrology, 38(2), 510- 515.

Visher, G.S. (1971) Depositional processes and the Navajo sandstone. Geological Society of America Bulletin, 82, 1421-1423.

55 CHAPTER 3

EXPANDED DISCUSSION

Addressing Horowitz’ Predictions

Horowitz’s 1982 dune collapse model was detailed and well-conceived, suggesting a number of specific, testable predictions. I will attempt to address each below, in order. In each case, the numbered sentences are aspects of the Horowitz model.

1. Liquefaction potential will be highest in the interdune area

a. There may be evidence of high water table conditions and liquefaction

unconnected to the dune collapse event.

Point 1 is a well-established fact in research into liquefaction (Bryant, Monegato, &

Miall, 2013), and its corollary, 1a (the supposition that liquefaction features unconnected to the dune collapse event may be visible nearby), is borne out near the fringes of the dune collapse feature on the Canyon Overlook Trail. Dinosaur tracks (Figure [dinosaur track]) may be seen near the margins of the interdune deposit, and pockets dish and pillar structures (Figure [dish and pillar structures]), which are indicative of vertical movement of pore fluids; all of which are apparently unaffected by the dune collapse event speak to the high potential for liquefaction of these interdune sediments.

2. The steeper lee slope of the upwind dune will exert greater pressure on interdune

sediments, creating a net stress field in the downwind direction

Point 2, unfortunately, is still purely theoretical, as the dune collapse process has not been described or measured in modern sediments. This may be a profitable direction for future studies. However, the observed tendency, in ancient eolianites, for folding to occur

56 preferentially in the direction of foreset dip, as noted by Rubin and Hunter, (as well as

Horowitz and others) does provide observational support.

3. Undermining of the dune’s toe would cause it to slump in the downwind direction

a. The movement of the slump would be taken up by one or more

synsedimentary faults, rather than by purely plastic deformation

b. The detachment surface of the slump should strike parallel to the dune

crest, and dip downwind

Point 3a is the first statement which must be tested on the outcrop. While there clearly are many shear surfaces, how do we know which are due to the recent jointing pattern which encouraged the erosion of the canyon, and which are soft sediment shear surfaces related to the dune collapse event? Recent joints, although very common in the outcrop, are easily distinguished from synsedimentary faults in two ways. First, since recent faults developed in cemented sandstone, shear stress broke grains apart, rather than causing grains to slide past each other. This creates narrow bands of material with higher porosity and more surface area on the grains, which leads to greater residence time for pore fluids, and in turn greater cementation. This is known as a cataclastic texture, and it is readily visible on the trail as it weathers more slowly than the rest of the sandstone, leaving raised fins. Synsedimentary shear planes, however, do not lead to breakage of grains, and therefore have little textural difference from the Navajo Sandstone as a whole. In addition, recent joints are all, or nearly all, vertical. Plotting the dips and strikes of shear surfaces on the outcrop on a stereonet graph yields a clear bimodal distribution, signifying different genetic processes (see Figure [stereonet]). Point 3b, unfortunately,

57 cannot be evaluated at this outcrop, as the detachment surface of the slump does not appear to be exposed, remaining in the body of the rock.

4. The slump would create one or more “thrust ramps,” as Horowitz named them,

referred to in this paper as synsedimentary thrust faults or simply thrust faults.

a. Thrust faults should strike parallel to the dune crest and dip upwind

Points 4 and 4a are easily addressed by the same stereonet (Figure [stereonet]), which shows that the synsedimentary thrust faults do indeed strike at approximately 300°, which is perpendicular to the regional paleocurrent direction reconstructed from the attitude of cross-bed laminae. This means that the faults would have run parallel to the dune crest, and they do indeed dip in the upwind direction.

5. The disturbance should be laterally extensive, but vertically limited within the

overall deposit, suggesting a surficial process

Point 5 is easily observed and demonstrated on the trail itself. The material above and below the deformed interdune succession is clearly not affected by the dune collapse event, so the entire thickness of this deformation feature is at most 25 meters, and averages 10 meters. Its horizontal extent, however, is considerable, with traces of the dune collapse event visible nearly a kilometer to the South. Liquefaction features are typically not significantly more extensive horizontally than vertically, so this points to an unusual process acting primarily at the surface (Bryant & Miall, 2010).

6. The collapse of a major dune should represent a significant topographic

disturbance to the dune field

58 a. Much of the lee face of the upwind dune, probably including the dune

crest, should drop down significantly as the slump rotates (Figure

[conceptual model B])

b. The lateral movement of dune and interdune material should raise the

average elevation of the interdune (Figure [conceptual model B])

Point 6 is entirely a logical conclusion following Point 5. Since dune collapse is a surficial process, its immediate effects will be surficial. Point 6a, like 3b, is impossible to test with this outcrop. In fact, it is highly improbable that any material from the vicinity of the dune crest would have been preserved (see [conceptual model c]). Point 6b is readily visible in Figure [keystone], where material is thrust upward at a steep angle in what was presumably the interdune area. The stratigraphic column (Figure [strat column]) and cross-section (Figure [cross-section]) demonstrate the thickening of the interdune succession more quantitatively. While it is not unusual in the Navajo Sandstone to have cross-bed sets up to 30 meters thick, it is exceedingly rare to see interdune successions more than 2 meters thick.

7. The dune collapse process should be followed by deflation of the deposit to the

water table, producing a flat upper truncation surface

Point 7 is, of course, a prediction Horowitz made which this study seeks to contradict.

It is clear from the “keystone outcrop” (Figure [keystone]) that this dune collapse event did produce surface topography, which was not subsequently eroded to a flat plane. More importantly, it demonstrably resulted in a net addition to the accumulation, not a net reduction. The evidence for dune collapse is preserved within the set produced by the failed dune, not within affected sets in the underlying accumulation. This event fits into a

59 succession of continuously climbing dunes. The resistance of this topography is primarily due to the cohesion of the interdune carbonate muds, which were likely partially cemented by this point (see Figure [contorted chert laminae]). Erosion of aeolian deposits to the water table represents a much more significant event than the collapse of a single dune, representing the creation of super surfaces (G. Kocurek, 1988). Dune collapse features would certainly represent extensive surficial disturbances, and, where the specific arrangement of interdune mud lenses is lacking, a flat truncation of the top of the deposit is entirely within reason. However, dune collapse features must be studied in an actualistic sense, as local phenomena. They cannot be used as a hallmark of a major regional unconformity, as they have been in the past (Eisenberg, 2003).

Relationship to Other Projects Underway

The presence of abundant tracks and surficial SSD features has been invoked, legitimately (Gary Kocurek & Hunter, 1986) and otherwise (Eisenberg, 2003), as evidence for a halt in normal depositional conditions, and the formation of a sequence boundary or supersurface. In studying the Navajo Sandstone and other eolianites, it is vital for researchers to remember that, although deposition may continue on geologic timescales, some surfaces in an active dune field may be exposed for months or years.

Another Loma Linda University project by Ranjan Fernando, is studying the generation of heavily trampled surfaces on the lee face of an advancing dune in the Navajo erg. This project and my own should hopefully help to dispel the myth that any disruption in the normal cross-bedding of a sandstone constitutes catastrophic regional changes in the paleoenvironment.

60 Conclusions

As demonstrated in this study, dune collapse is not genetically bound to the formation of supersurfaces or isolated within periods of non-deposition. In fact, dune collapse features can result in net accumulation, and do not require any reduction of the substrate. Dune collapse features may be expected to contribute to deposition, assuming active depositional conditions in the paleoenvironment as a whole. The Canyon Overlook

Trail outcrop provides less ambiguous details of dune collapse processes than have been available previously, allowing for more comprehensive analysis. This enables a more reliable interpretation of dune collapse features at other locations.

The Horowitz (1982) model suffers from narrow assumptions regarding paleohydrological controls on the development of the accumulation. It should not be used as a general case to explain the occurrence of large-scale SSD features. However, the characterization of deformation mechanics it provides is confirmed by this study and should continue to provide a useful model for the production of a subset of SSD features in ancient eolianites.

61 References

Eisenberg, L. (2003). Giant stromatolites and a supersurface in the Navajo Sandstone, Capitol Reef National Park, Utah. Geology, 31(2), 111. doi:10.1130/0091- 7613(2003)031<0111:gsaasi>2.0.co;2

Kocurek, G. (1988). First-order and super bounding surfaces in eolian sequences; bounding surfaces revisited. In: Late Paleozoic and Mesozoic eolian deposits of the Western Interior of the United States. Sedimentary Geology, 56, 193-206.

Kocurek, G., & Hunter, R. E. (1986). Origin of Polygonal Fractures in Sand, Uppermost Navajo and Page Sandstones, Page, Arizona. Journal of Sedimentary Petrology, 56(6), 895 - 904.