Geology and Geomorphology of Coral Pink Sand Dunes State ...Geology and Geomorphology of Coral Pink Sand Dunes State Park, Utah RichardL.Ford1,ShariL.Gillman1,DavidE.Wilkins2,WilliamP.Clement3,

Geology and Geomorphology ofCoral Pink Sand Dunes State Park, UtahRichard L. Ford1, Shari L. Gillman1, David E. Wilkins2, William P. Clement3,and Kathleen Nicoll4

Geology of Utah’s Parks and Monuments2010 Utah Geological Association Publication 28 (third edition)D.A. Sprinkel, T.C. Chidsey, Jr., and P.B. Anderson, editors

1Department of Geosciences, Weber State University, Ogden, UT 844082Department of Geosciences, Boise State University, Boise, ID 837253Center for Geophysical Investigation of the Shallow Subsurface,Boise State University, Boise, ID 837254Department of Geography, University of Utah, Salt Lake City, UT 84112

ABSTRACTCoral Pink Sand Dunes State Park, located in southwestern Kane County, Utah, contains a variety of geologic features

including one of the largest areas of freely migrating dunes in the Colorado Plateau. The semiarid climate, strong pre-vailing southerly winds, sparse vegetation, and abundant supply of sand-sized sediment make this area susceptible to eo-lian processes.

Picturesque exposures of Jurassic rocks are present within the park. The stratigraphic sequence ranges from the Tri-assic-Jurassic Moenave Formation to the Middle Jurassic Carmel Formation. The most widespread bedrock unit exposedwithin the park is the Navajo Sandstone (Lower Jurassic). The Navajo Sandstone is also widely exposed across the Moc-casin Terrace southwest of the park and is the most likely source for the sand that comprises the dune field. The “coralpink” color of the dune sand is the result of iron-oxide stains on the surface of the sand grains inherited from the sourcesandstones.

Migrating dunes, whose morphology is primarily a function of wind characteristics, include transverse ridges,barchanoid ridges, and a solitary star dune. Dunes influenced or impeded by topographic obstacles or vegetation includeclimbing dunes, echo dunes, parabolic dunes, vegetated linear dunes, and nebkhas.

We divide the dune field into major geomorphic units based on the dominant dune type. A largely stabilized (vege-tated) sand sheet and partially stabilized, poorly organized dunes are present at the southern (upwind) end of the dunefield. The active core of the dune field contains transverse ridges and barchanoid ridges. Barchanoid ridges at the north-ern (downwind) end of the active core grade into climbing dunes that ramp up the bedrock escarpment associated withthe Sevier fault. The climbing dunes in turn grade into large parabolic dunes that dominate the downwind end of the dunefield.

Coral Pink Sand Dunes lies within the structural transition zone between the Great Basin section of the Basin andRange province to the west, and the core of the Colorado Plateau to the east. The north-south-trending Sevier fault cutsthrough the length of the park. The fault trace is marked by a west-facing bedrock escarpment that divides the park intotwo topographic units (a forested plateau to the east and a relatively low-lying valley floor to the west) and acts as a majorcontrol over the accumulation of sand within the dune field.

Important events recorded in the geologic features of the park include the Triassic and Jurassic depositional history ofthe Glen Canyon Group, the Cretaceous to Cenozoic structural history of the Colorado Plateau, and the late Holocene his-tory of the active dunes. Optically stimulated luminescence dates from the active core of the dune field indicate thatHolocene eolian deposition began at least 4,000 years ago. Radiocarbon dating of organic materials from an exhumed soilsurface suggests a period of landscape stability approximately 500-200 years ago, coincident with the Little Ice Age. Den-drochronologic data from the ponderosa pines in the park, along with historic photographs, indicate the dune field has ex-perienced alternating wet periods and drought since the end of the Little Ice Age, which have influenced vegetation cov-erage and dune activity in the area

371

INTRODUCTIONGeographic Setting

Sand dunes occur in two distinct habitats; along thecoasts of seas and rivers, on the one hand, and on thebarren waterless floors of deserts, on the other....Here[in deserts] instead of finding chaos and disorder, theobserver never fails to be amazed at a simplicity of form,an exactitude of repetition and geometric order un-known in nature on a scale larger than that of crys-talline structure.

— Ralph A. Bagnold, 1942

The many geologic wonders of southern Utah includea picturesque area of active dunes in southwestern KaneCounty, located about 27 miles (43 km) northwest ofKanab near the Utah-Arizona border (figure 1). This is oneof the largest areas (approximately 5.4 square miles or 14km2) of freely migrating dunes within the ColoradoPlateau Province. This dune field has become known asthe Coral Pink Sand Dunes. The term “coral” in this caserefers to the deep orange pink color of the sand and hasnothing to do with the marine reef-building organisms ofthe same name. Using a soil-color chart, geologists or soilscientists would formally describe the sand as reddish yel-low (5YR 6/8); it is probably a good thing that a geologistdid not name these dunes. A wide variety of eolian (an ad-jective referring to wind activity, derived from the name ofthe Greek god of the wind, Aeolus; also commonly spelled

as aeolian) features and landforms are present within theCoral Pink Sand Dunes, making this an excellent field lab-oratory in which to learn about the role of wind in shapingthe surface of the Earth.

The northern half of the dune field is federal land ad-ministered by the Bureau of Land Management (BLM).The southern half of the dune field and adjacent land(3,730 acres or 15 km2) comprises Coral Pink Sand DunesState Park (CPSDSP), the subject of this paper. The Yel-lowjacket Canyon 7.5-minute topographic quadranglecovers the park area.

The park was established in 1963 and initially visitedprimarily by small numbers of off-highway vehicle (OHV)enthusiasts. Over the years visitation has greatly in-creased and today the majority of park visitors arrivewithout OHVs to enjoy the scenic beauty of this distinctivelandscape. Park facilities include a 22-unit campground,modern restrooms with hot showers, a boardwalk over-look, and a one-half-mile nature trail. In addition, the parkserves as an excellent base for exploring other areas of ge-ologic interest in southern Utah, including Zion and BryceCanyon National Parks, Cedar Breaks and Grand Stair-case–Escalante National Monuments, and KodachromeBasin State Park.

Sand Dune Road, a paved road that runs south fromU.S. Highway 89 through Yellowjacket Canyon, and Han-cock Road, a paved road that runs southwest from U.S.Highway 89, provide access to the park (figure 1). Togeth-er, Sand Dune Road and Hancock Road comprise the state-designated Ponderosa/Coral Pink Sand Dunes ScenicBackway.

R.L. Ford and S.L. Gillman Geology of Coral Pink Sand Dunes State Park, Utah

372

Pipe SpringsNat'l Monument

KANAB

FREDONIA

CPSDSP

Glendale

Mount Carmel Jct.

Mount Carmel

Orderville

San

dD

une

Rd

Hancock Rd

UTAH UTAH

ARIZONA ARIZONA

WA

SH

ING

TON

CO

UN

TY

KA

NE

CO

UN

TY

ZIONNATIONAL

PARK

LongValleyJct.

DIXIENATIONALFOREST

DIXIENATIONALFOREST

KAIBABINDIANRESERVATION

BRYCE CANYONNAT'L PARK

PINK CLIFFS

WHITE CLIFFS

VERMILLION CLIFFS

GRANDSTAIRCASE-ESCALANTE

NATIONALMONUMENT

389 89

8969

9

14

0

0

5

10

10

20

20

30 40

30 miles

km

UTAH

Figure 1. Location of Coral Pink Sand Dunes State Park (CPSDSP) in Kane County, Utah.

TopographyTopographically, the park can be divided into two dis-

tinct units. The eastern half is a forested upland, part ofthe Moquith Mountains, bounded by a west-facing es-carpment. This escarpment forms the eastern redrockbackdrop to the dunes, visible from the dune field andcampground to the west (figure 2). This escarpment isfault-controlled (see Structure section) and increases inheight from north to south, reaching a maximum of ap-proximately 700 feet (213 m) near the southern end of thepark. Although given the name “mountains,” the Mo-quiths are actually a deeply incised plateau that is tilted to-ward the northeast. The highest elevation within the park,6,977 feet (2,126 m), lies on this plateau. The fault-con-trolled escarpment exposes the same rock units that formthe Vermillion Cliffs of the Grand Staircase, which lie gen-erally east and south of the park. For a complete descrip-tion of the rock units and features that comprise the GrandStaircase of southwestern Utah (see Grand Staircase–Es-calante National Monument article in this volume).

The western half of the park is relatively low terrainlying between the Moquith Mountains and the mesa justnorthwest of the park. Though unnamed on modern topo-graphic maps, this mesa was called Esplin Point by Gre-gory (1950) in his early regional study, a name we will usein this paper. Esplin Point lies at the eastern end of a lineof cliffs and mesas that extend from Elephant Butte to thepark, which collectively form the western expression ofthe White Cliffs of the Grand Staircase. Esplin Point has amaximum elevation of approximately 6,850 feet (2,090 m).Most of the dune field within the park boundaries lies onthis lowland at the base of the scarp bounding the Mo-quith Mountains. Near the northern end of the park thedunes climb the escarpment and continue to the northeast,forming that part of the dune field on land administeredby the BLM.

The park area is drained by a number of ephemeralstreams whose channels extend from the upland areas, oneither side of the dune field, to the margin of the dune fielditself. Channels are largely absent within the dunefield,owing to the high infiltration rates within the dune sand.However, the steep ephemeral channels which drain theeastern escarpment act as tributaries to Sand CanyonWash, lying at the base of the escarpment and along theeastern margin of the dune field. Sand Canyon Washflows southward to Kanab Creek and hence to the Col-orado River. The lowest elevation within the park, ap-proximately 5,620 feet (1,710 m), is in the channel of SandCanyon Wash where it crosses the park’s southern bound-ary. Thus, the total topographic relief within the park isapproximately 1,357 feet (416 m). The drainage divide be-tween Sand Canyon Wash and Yellowjacket Canyon to thenorth is located near the intersection of Hancock and SandDune roads, approximately 0.9 miles (1.4 km) north of thepark. North of this divide, Sand Dune Road descends Yel-lowjacket Canyon, a north-flowing tributary of the EastFork of the Virgin River.

Climate and VegetationEolian processes are important in areas with strong

prevailing winds, sparse vegetation, and an abundant sup-ply of sand-sized sediment. This is why most of theworld’s major dune fields are found in association witharid or semiarid climates. The Coral Pink Sand Dunes areno exception, being located within the extensive semiaridclimatic zone of Utah known as steppe (Köppen climaticclassification: BS). In Utah, the steppe climate is transi-tional between the true arid/desert climates of the GreatBasin and the moister climates of the state’s mountainousregions (Murphy, 1981). The mean annual temperature inKanab, Utah, is 54.4° F (12.4° C); the mean annual precipi-tation is 13.3 inches (33.8 cm) (Utah Climate Center, as re-ported in Pope and Brough, 1996). The distribution of pre-cipitation in this area, as evidenced by the data fromKanab, is distinctly bimodal with winter-summer precipi-tation peaks and spring-autumn drought. The closestweather station to the park with long-term wind data isLas Vegas, Nevada, located approximately 140 miles (225km) to the southwest. The mean annual surface wind atthis location blows from the southwest at 9.3 mph (15.0kph). It is interesting to note that in Las Vegas, the twomonths of May and June have the greatest average windspeed, which coincides with the two driest months of theyear in the Kanab area. If this relationship is extrapolatedto the Coral Pink Sand Dunes, one would hypothesize thateolian activity would generally peak during this time.

Outside of the distinctive habitats of the dune field it-self (see Castle, 1954), the dominant vegetation of the west-ern half of the park is typical of pinyon-juniper woodlands(Pinus edulis, Juniperus spp.) and sagebrush scrub(Artemisia spp.) of the Colorado Plateau—with one no-table and scenic exception. Impressive stands of pon-

D.A. Sprinkel, T.C. Chidsey, Jr., and P.B. Anderson, editors 2010 Utah Geological Association Publication 28

373

Figure 2. View to the east from the boardwalk overview of the bedrockescarpment that forms the eastern boundary of the dune field. The es-carpment was created by movement along the Sevier fault (see Struc-ture section) and forms the western boundary of the forested MoquithMountains in the background. A solitary star dune (see Classic Geo-logical Sites section) is visible in the middle ground.


derosa pine (Pinus ponderosa) are present within the morestabilized areas of the dune field. Ponderosa pines alsodominate the forested upland of the eastern side of thepark.

Previous InvestigationsGeologic aspects of Coral Pink Sand Dunes have been

previously reported in more broad-based investigations,most notably Gregory (1950), Doelling and others (1989),Anderson and Christenson (1989), and Davis (1999). Cas-tle (1954) investigated the soil conditions and vegetationcommunities within the dune field. We are greatly in-debted to these researchers, as much of this paper is a syn-thesis of their work. However, this is the first work tofocus specifically on the geology of Coral Pink SandDunes. One of our major contribution to the geologicknowledge of this area lies in the complete characteriza-tion of the dune field, including its subdivision into dis-tinct geomorphic units that reflect the relationship be-tween surface forms and dominant surface processes. Inaddition to our literature review, we conducted an analy-sis of color aerial photography (scale 1:4,000), augmentedby reconnaissance-level field research during the springand fall of 1999.

Subsequent research by the authors has utilizedground-penetrating radar (GPR) to characterize the inter-nal structure of the dunes (Wilkins and others, 2002; 2003).Wilkins and others (2005) also developed a chronology ofHolocene dune activity based on radiocarbon and optical-ly stimulated luminescence (OSL) dating. Wilkins andFord (2007) used repeat photography and spatial statisticsto document recent changes in the spatial organization ofthe dune field.

STRATIGRAPHY AND GEOMORPHOLOGYThe generalized geologic map of the park area (figure

3) and diagrammatic cross section (figure 4) provide a ideaof the geology. The general geology of the park can becharacterized as an area of cliff-and-bench topography, de-veloped on Mesozoic sedimentary rocks, with a Quater-nary dune field at the surface. The oldest Mesozoic unitexposed at CPSDSP is the Moenave Formation (Triassic-Jurassic). If one were to drill below the present-day dunefield, the Chinle (Upper Triassic) and Moenkopi (LowerTriassic) Formations would be successively encounteredbelow the Moenave Formation (figure 4). The youngestbedrock unit exposed within the park boundary is thelowest member of the Carmel Formation (Middle Juras-sic); the Co-op Creek Limestone Member of Doelling andothers (1989). The various unconsolidated eolian depositsand active dunes, for which the park is named, are theyoungest of all the geologic units in the area.

Mesozoic RocksThe picturesque Moenave/Wingate, Kayenta, and

Navajo Formations make up the Glen Canyon Group.

These units have historically been regarded to be Triassicand Jurassic in age; however subsequent stratigraphicwork has shown that the Glen Canyon Group is entirelyEarly Jurassic in age (Pipiringos and O’Sullivan, 1978;Imlay, 1980; and Doelling and others; 1989). The generalstratigraphy and depositional environments of the variousJurassic-age bedrock formations exposed in the park areaare discussed below. A generalized stratigraphic columnis presented in figure 5.Moenave Formation (Jmo)

In western Kane County, the Moenave Formation (Tri-assic-Jurassic) overlies the Chinle Formation and theboundary between them was thought to represent a peri-od of erosion called the J-0 unconformity (Pipiringos andO'Sullivan, 1978), but new evidence suggests the contact isconformable and the "J-0" unconformity is within lowerMoenave strata (Lucas and other, 2005) (figure 5). TheMoenave Formation is now divided into three members(in ascending order, the Dinosaur Canyon and WhitmorePoint), The Springdale Sandstone was formerly the uppermember but is now reassigned to the basal Kayenta For-mation (Lucas and Tanner, 2006). Only the upper part ofthe Moenave Formation is exposed in fault blocks of theSevier fault zone south of the park (figure 3).

Thin bedded sandstone and siltstone with beddingparallel lamina and low-angle cross-beds in the DinosaurCanyon Member and siltstone, fine-grained sandstone,thin limestone, and mudstone with freshwater-fish fossilsin the Whitmore Point Member suggest a terrestrial depo-sitional environment for the Moenave Formation. Specif-ic environments probably included lakes, mudflats, andfluvial channels as part of a broad floodplain (Doellingand others, 1989).Kayenta Formation (Jk)

The Kayenta Formation (Lower Jurassic) uncon-formably overlies the Moenave Formation (figure 5). TheKayenta Formation is exposed within the park in the Sevi-er fault zone along the base of the escarpment that formsthe western edge of the Moquith Mountains (figure 3).The Springdale Sandstone Member is also exposed in faultslices along the base of the Moquith Mountains. The unit,160 to 185 feet (49–56 m) thick, consists of very finegrained, pale-reddish-brown sandstone with subordinateamounts of cliff-forming conglomerate (Doelling and oth-ers, 1989). The Kayenta Formation above the Springdaleconsists primarily of deep red siltstones and mudstones.Smaller amounts of lenticular, medium-grained, trough-cross-bedded sandstone are also present. In Kane County,the thickness of the Kayenta Formation (not including theSpringdale Sandstone Member) varies from 190 to 340 feet(58–104 m) (Doelling and others, 1989).

Although the Kayenta Formation is mostly fluvial inorigin (formed by shifting braided streams), some eoliansandstone and lacustrine limestone beds, deposited in in-terfluve areas, are locally present (Doelling and others,

374


375

Figure 3. Geologic map of the Coral Pink Sand Dunes area in Kane County, Utah (after Sargent and Philpott, 1987). The map legend is includedwith the cross section shown in figure 4 and identifies the geologic units and symbols used. Note the trace of the Sevier fault running north-souththe length of the park.


1989). These streams originated in highlands to the east,probably the Uncompahgre uplift (Barnes, 1993).

A thin, though mappable, tongue of sandstone, knownas the Tenney Canyon Tongue, extends eastward from themain body of the Kayenta Formation into the lower por-tion of the overlying Navajo Sandstone (figure 5). Theportion of the Navajo Sandstone lying between the mainbody of the Kayenta Formation and the Tenney CanyonTongue is called the Lamb Point Tongue of the NavajoSandstone (Doelling and others, 1989). These two bedrockunits are also exposed within the park along the trace ofthe Sevier fault (figure 3).

Lamb Point Tongue of the Navajo Sandstone (Jnl)

The Lamb Point Tongue (Lower Jurassic) is named fora location on the Vermilion Cliffs south of the park. Theunit consists of white, tan, or gray fine-grained sandstonewith thick sets of eolian cross-beds, locally greater than 25feet (7.6 m). The lower and upper contacts with theKayenta Formation are sharp. This unit is 90 to 410 feet(27–125 m) thick and pinches out to the west of the park.The Lamb Point Tongue contains distinctive contorted

cross-beds in the upper 10 to 15 feet (3–4.6 m) of the unit.These contorted beds have been interpreted to be the re-sult of slumping on the steep downwind side of dunes orpossibly the result of earthquake-induced liquefaction(Doelling and others, 1989). The Lamb Point Point unit isan intertongue of the Navajo Sandstone resulting fromwidespread eolian deposition in Early Jurassic time.

Tenney Canyon Tongue of the Kayenta Formation (Jkt)

The Tenney Canyon Tongue (Lower Jurassic) of theKayenta Formation is pale-reddish brown in color andcomprised of siltstone, mudstone, and very fine thin-bed-ded to laminated sandstone, representing floodplain andchannel deposition. It is 90 to 170 feet (27–52 m) thick andtypically displays low-angle cross-bedding. This unit hassharp contacts with both the underlying Lamb PointTongue and the overlying main body of the Navajo Sand-stone (Doelling and others, 1989). The tongue thins fromwest to east and pinches out east of Kanab, Utah.

Navajo Sandstone (Jn)

The Navajo Sandstone (Lower Jurassic) stands out as

376

Figure 4. Cross section along line A-A’ shown in figure 3. Note the offset in the geologic units created by movement along the Sevier fault zone.

the premier formation of the remarkable scenery of theColorado Plateau. In the vicinity of CPSDSP, the NavajoSandstone is a thick unit, 1,800 to 2,300 feet (549–701 m), ofwell-sorted, fine-grained quartz sandstones of varying col-ors and degrees of cementation. In addition to the cross-bedded sandstone, the Navajo contains a few lenticularbeds of dense limestone or dolostone. These carbonaterocks are thought to have accumulated in interdune lakesor playas. Being more resistant than the sandstones, thelimestones and dolostones commonly form benches orshelves (Doelling and others, 1989).

The Navajo Sandstone is world-renowned for its elab-orate array of high-angle cross-beds and stark erosionalforms, including cliffs, domes, and monuments. The lightcolors of the Navajo have been described variously aswhite, tan, buff, salmon, pink, vermillion, brown, red, yel-low, cream, orange and gray. The sandstone is weakly ce-mented by calcite or dolomite and varying amounts of

iron oxides. Hematite and goethite cements, produce thered, orange, and yellow colors; ferrous-iron minerals con-tribute to the brown and occasional green colors (Doellingand others, 1989). The diagenetic coloration of the NavajoSandstone and related units tells an interesting story ofiron cycling, documented in the recent studies of Chan andothers (2000), Chan and Perry (2002), and Beitler and oth-ers (2003 and 2005).

The Navajo Sandstone is the most extensive bedrockunit within the park (figure 3). The lower and upper partsof the formation are more strongly cemented than the mid-dle third and thus are cliff-forming units. The lower cliff-forming unit is typically reddish in color and, along withthe underlying Kayenta Formation, form the VermillionCliffs of southwestern Utah. The upper cliff-formingsandstones typically lack the reddish hues characteristic oflower section and thus form the White Cliffs of the GrandStaircase. The lower reddish cliff-forming portion of the


377

Figure 5. Stratigraphic column of the Mesozoic bedrock units present within Coral Pink Sand Dunes State Park (modified from Doelling and oth-ers, 1989).


Navajo is exposed in the bedrock escarpment east of theactive dunefield, whereas the upper white cliff-formingportion of the Navajo is exposed in the face of Esplin Point(figure 6), the prominent mesa located due north of thepark’s campground. In active gullies and washes alongthe margin of the Coral Pink Sand Dunes, the NavajoSandstone—presumably the middle to upper third of theformation—directly underlies the modern dunes.

Deposition of the Navajo Sandstone occurred in alarge coastal to inland desert, possibly similar to the mod-ern Sahara. Orientation of the cross-bedding, essentiallythe lithified slip faces of the Jurassic dunes, indicates thatthe Jurassic winds blew primarily from the north andnorthwest (Stokes, 1986; Parrish and Peterson, 1988; Peter-son, 1988). The mineralogy of the Navajo Sandstone, 90 to98 percent clear quartz with minor amounts of feldspar,magnetite, tourmaline, staurolite, zircon, garnet, and mica,suggests the source area of the sand was dominated bymetamorphic rocks (Doelling and others, 1989). Recent re-search on the detrital zircons found within the NavajoSandstone suggests that the sediments were eroded fromthe Appalachian Mountains, Canadian Shield, and the An-cestral Rocky Mountains (Dickinson and Gehrels, 2003).The sediments were probably transported westward bylarge rivers and deposited on a coastal plain and shallow-water shelf that lay up-wind of the Navajo erg. Geologistsuse the Arabic term “erg” to refer to large (≥ 48 squaremiles or 125 km2) desert areas dominated by wind-de-posited sand and complex dune forms (smaller areas ofdunes, such as the Coral Pink Sand Dunes, are defined as“dune fields” [Pye and Tsoar, 1990]).

Some early workers (e.g. Stanley and others, 1971) fa-vored a marine origin for the Navajo Sandstone as largeoffshore sandbars cross-bedded by ocean currents. Thiscontroversy was essentially put to rest by the work ofHunter (1976, 1977), who established the criteria for dis-tinguishing eolian and water-laid cross-stratification. The

discovery of local occurrences of petrified trees within thesandstone (presumably spring-fed desert oases) and ani-mal tracks (from vertebrates, including dinosaurs, and in-vertebrates) preserved within the interdune limestoneadded further evidence of terrestrial deposition (Barnes,1993; Wilkens and Farmer, 2005; Parrish and Falcon-Long,2007). Recently BLM geologists confirmed the existence ofa substantial dinosaur trackway, known to OHV enthusi-asts for some time, located on BLM land just southwest ofCoral Pink Sand Dunes State Park (Mathews and others,2010). The exposure in Navajo Sandstone, known as theMoccasin Mountain Tracksite, contains tracks from at leastsix different types of dinosaurs.Temple Cap Sandstone (Jtc)

A major unconformity (J-1) marks the top of the Nava-jo Sandstone (figure 5). Above it is the Temple Cap Sand-stone of Middle Jurassic age, named for the East and WestTemple features of Zion National Park. This formationcontains, at its base, a prominent reddish, slope-formingsiltstone and sandy siltstone, that is 40 to 50 feet (12–15 m)thick (Sinawava Member). This member is overlain by upto 150 feet (46 m) of light-gray to tan, cross-bedded, cliff-forming sandstone (White Throne Member) (Doelling andothers, 1989). The sandstones and siltstones of the lowerSinawava Member are marine to marginal marine(Sprinkel and others, 2009); the cross-bedded sandstonesof the White Throne Member represent aeolian deposition.However, the two members are difficult to separate owingto the extensive interfingering between the two (Doellingand others, 1989).

The Temple Cap Sandstone is exposed in the centralportion of the face of Esplin Point north of the camp-ground (figure 6). This outcrop belt extends to the westernedge of the park (figure 3).Co-op Creek Limestone Member of theCarmel Formation (Jc)

The Carmel Formation (Middle Jurassic), named forMt. Carmel, Utah, is a lithologically complex unit of lime-stones and sandstones that is subdivided into numerousmembers (Hintze, 1988; Doelling and others, 1989). Thelowermost Co-op Creek Limestone Member caps EsplinPoint (figure 6). Co-op Creek beds also occur just west ofthe road near the northern boundary of the park. The Co-op Creek Member consists of a basal siltstone, 10 to 20 feet(3–6 m) thick; however, this unit is similar to the SinawavaMember of the Temple Cap and is considered the upperpart of the Temple Cap (Sprinkel and others, 2009). Theremaining Co-op Creek Members is more than 200 feet(61+ m) of thin- to medium-bedded light-gray limestoneand shaly limestone (Doelling and others, 1989).

The Co-op Creek Limestone Member of the CarmelFormation conformably overlies the Temple Cap Sand-stone (figure 5). The Carmel Formation in turn is locallytruncated by the unconformity at the base of the Creta-ceous System. Most importantly, the Carmel Formation

378

Figure 6. View to the northwest of Esplin Point, part of the WhiteCliffs. Navajo Sandstone (Jn), Temple Cap Sandstone (Jtc), and the Co-op Creek Limestone Member of the Carmel Formation (Jc) are exposedin the cliff face. Foreground area is covered by mixed eolian and allu-vial sediments (Qae).

marks the end of terrestrial deposition and the return of re-gional shallow marine conditions to this part of Utah dur-ing Middle Jurassic time. Subsequent geologic events up-lifted the sediments of the Carmel Formation from belowsea level and raised them to their present elevation of over6,600 feet (2,011 m) in the vicinity of the park.

Quaternary Deposits and Geomorphology ofthe Dune Field

Although the local Jurassic bedrock units are impres-sive, the foremost geologic features of this park are thevarious unconsolidated eolian deposits and landformsthat make up the dune field. Whereas the Jurassic forma-tions provide information about ancient depositional envi-ronments, the modern dunes are the result of recent andon-going geological processes during the Quaternary Peri-od (the last 2.6 million years of Earth’s history).

For a dune to form, a small patch of sand must first ac-cumulate where the wind speed has been reduced, gener-ally by an increase in surface roughness. Once formed thepatch of sand grows by trapping bouncing (saltating)grains that are unable to rebound upon impact as easily asthose that impact harder non-sandy surfaces (an exampleof a positive feedback). As a result of the separation anddeceleration of the airflow on the lee (downwind) side ofthe growing sand body, sand accumulates more rapidlythan it is removed by eolian erosion; thus the dune in-creases in size (Summerfield, 1991). A mature dune is thusshaped by the airflow over it and in turn modifies the air-flow. As a result, many dunes attain a characteristic equi-librium profile of three components: (1) facing upwind is agently inclined (typically 10–15 degrees) backslope, (2) fac-ing downwind is a steep slipface, standing at the angle ofrepose for sand (generally 30–34 degrees), and (3) separat-ing the two slopes, a dune crest (Ritter and others, 1995).

This asymmetric equilibrium profile can be seen in a vari-ety of dune types with differing plan forms (figure 7).

Wind direction and velocity, sand supply, and thepresence of topographic obstacles or stabilizing vegetationare the most important factors influencing dune morphol-ogy or type. Considering these factors, it is useful to clas-sify the various dune types as either free dunes, whosemorphology is primarily a function of wind direction andsediment supply, or impeded dunes whose form is greatlyinfluenced by the effects of vegetation, topographic obsta-cles, or highly localized sediment sources (Summerfield,1991). Many of the classic dune types are present withinCoral Pink Sand Dunes State Park (table 1).

Sargent and Philpott (1987) divide the dune field intotwo units: (1) a larger core of active sand dunes and (2) asmaller area of inactive eolian sand along the southernmargin of the dune field. Doelling and others (1989) mapthe dune field as a single unit. Neither report identifies thevarious types of dunes present in the area. Based on fieldobservations and an analysis of aerial photography, wesubdivide the active core of the dune field into five (5) mapunits that reflect the dominant dune type present (figure8). In addition, we retain the stabilized-sand unit (Qes) ofSargent and Philpott (1987). These units are describedbelow in order of their occurrence, starting at the southend of the dune field.Stabilized Eolian Sand Sheet (Qes)

In the southwestern corner of the park, and extendingbeyond the park boundaries, is an area of wind-blownsand largely stabilized by sagebrush scrub (figure 8). Al-though slipfaces are not preserved on the dune forms, thisunit does retain some original depositional relief in theform of rolling hills of sand. The western and southernboundary of this unit is gradational with the mixed eolian


379

TranverseRidges BarchansBarchanoid

Ridges

ParabolicDunesStar Dunes Linear Dunes

B

Figure 7. Morphology of the major types of dunes. Arrows indicate the formative prevailing wind direction (modified from McKee, 1979).


and alluvial sediments of the valley floor (Qae), whichsupport a pinyon-juniper woodland. The interior bound-ary with the partially stabilized transverse dunes (Qeps) issharper, being marked by a distinct change from sage-brush scrub to a more sparsely vegetated area. This low-relief accumulation of eolian sand is best termed a sandsheet. Sand sheets develop under conditions unfavorableto dune formation, including a high water table, periodicsurface flooding, and a vegetative cover (Lancaster, 1995).Each of these conditions, which act to limit the amount ofsand available for dune formation, may be significant inthis area, especially the cover of vegetation.Partially Stabilized Dunes (Qeps)

Located at the southern (upwind) end of the dunefield and interior of the stabilized sand sheet is an area ofpartially stabilized dunes (figure 8). Dune morphologywithin this map unit is poorly developed and somewhatchaotic, but includes small transverse and parabolic-likeforms (discussed in detail in following sections). Many ofthe low rolling dunes in this area lack obvious slipfaces.Topographically, this map unit is transitional, in terms ofdune activity, relief and vegetative cover, between the sta-bilized sand sheet (Qes) to the south and the core of activedunes (Qetd) to the north.

The most distinctive characteristic of this geomorphicunit is the presence of scattered ponderosa pines (Pinusponderosa) amongst the dunes (figure 9). Ponderosa-pineforests are best developed on the Colorado Plateau on

mesa tops and mountain slopes at elevations between6,900 and 8,200 feet (2,100–2,500 m) (Betancourt, 1990).Their presence at this relatively low elevation, approxi-mately 5,680 to 5,800 feet (1,731–1,768 m), on a basin floorotherwise dominated by a pinyon-juniper woodland, iscurious. Ponderosa pines can tolerate a variety of soil con-ditions, but trees on thin, dry soils are usually dwarfed(Harlow and Harrar, 1968). These trees are not dwarfedand thus indicate a moisture regime more typical of coolerhabitats at higher elevations. We speculate that the south-ward slope of the dune field, combined with the high hy-draulic conductivity of dune sand, produces a localizedsouthward flow of shallow groundwater. The water tableand ground surface may converge as one moves south to apoint that is within the rooting depth of ponderosa pines.Four-year-old trees may have taproots 4 to 5 feet (1.2–1.5m) long (Harlow and Harrar, 1968).

The ponderosa pines in Qeps are found in various sit-uations, ranging from being rooted in shallow (less thanone meter deep over bedrock) cover sheet sand to havingpartially buried trunks (up to an estimated 4 meters) in thedunes. The duff from the trees (e.g., needles, branches,and cones) appears to add stability to the dune and swalesurfaces under and around the trees, and as a result verylittle surface dune activity is observed in the stands.

In the summers of 2002 and 2003, 15 ponderosa in thissection of CPSDSP were cored. Using standard den-drochronological methods, the cores were processed,dated, and the rings visually cross-dated (Stokes and Smi-

380

Table 1. Classification of dune types present at Coral Pink Sand Dunes State Park (modified from Summerfield, 1991). See figure 7 for diagrams ofthe major dune types.

DUNE TYPE MORPHOLOGY WIND REGIME / MODE OF DEVELOPMENT

A. FREE DUNESTransverse ridge asymmetric ridge, straight crest unidirectional windsBarchanoid ridge asymmetric ridge, sinuous crest unidirectional windsStar dune central peak with 3 or more arms multidirectional windsB. IMPEDED DUNESBlowout elliptical depression localized deflationParabolic dune u-shaped in plan, arms point upwind locally deflated sand partially anchored by

vegetationNebkha elliptical mound accumulation around a clump of vegetation

Climbing dune sandy ramp on the windward side of accumulation in zone of disrupted airflowlarge topographic obstruction

Echo dune asymmetric ridge parallel to and separated accumulation in zone of disrupted airflowfrom the windward side of a largetopographic obstruction

Vegetated linear dune symmetric ridge parallel to prevailing remnant arm of an eroded parabolic dunewind

ley, 1996). The range of variability in the trees’ environ-ments, that is, either partially buried or rooted in a swale,results in different growth-ring responses to the same an-nual rainfall variability. Rings in cores collected from treespartially buried in the dunes show wider rings than treesin the swales show in the same years; it is thought that thedunes allow greater retention of soil moisture duringdrought or dry years, and also enhance the growth in wet-ter years. These differences, though, prevent statisticalcross-dating so these cores are best used to provide mini-mum ages of tree and stand establishment (Wilkins andothers, 2005).

Because sampling was stratified towards the oldesttrees, the majority of those cored are greater than 150 yearsold. Minimum dates of establishment range from the oldestat 1562 CE to the youngest tree in the last decade of the 20thCentury, indicating the stand may have first become estab-lished around the beginning of the Little Ice Age, sometimearound the early 15th century (Bradley and Jones, 1993) buthas continued to survive through the present.

The ages of the trees in CPSDSP were compared to apublished ponderosa tree-ring chronology from NavajoMountain (Dean and Bowden, 1971), 150 miles to the eastat roughly the same elevation. Establishment dates of theCPSDSP trees correspond broadly with periods of widering widths in the Navajo Mountain trees, indicating theybecame established during periods of more effective mois-ture. A similar comparison against the reconstructedPalmer Drought Severity Index for the dunes (Cook andothers, 2004) also supports tree-establishment dates corre-sponding with higher levels of moisture, such as wouldhave existed during the Little Ice Age.

A model of stand establishment and dune activity inthis part of the dune field can be reconstructed using theseobservations. Assuming trees need moist conditions and astable surface to germinate, ages for the oldest trees of 440years suggests that activity in this portion of the dune fieldhas been climatically limited, at least intermittently, sincethe beginning of the Little Ice Age. Those same climaticconditions would result in sediment-limited conditions, as


381

0

0 21

1

toU.

S.89

Qes

to AZ

6850 ftx

NN

S o y

Cattle Guard Turnout

Overview Area

Star Dune

Roadsto U.S. 89

Esp

linPo

int

2 mi

3 km

Geomorphic Units:

Qelp - Large parabolic dunes

Qecd - Climbing dunes

Qemp - Marginal parabolic dunes

Qetd - Transverse dunes

Qeps - Partially stabilized dunes

Qes - Stabilized sand sheet

Moq

uith

Mou

ntai

nsSa

ndD

une

Roa

d

Hancock Road

Qelp

Qetd

Qemp

Qes

Sev

ier

fau

lt

Qecd

Qeps

Qes

EXPLANATION

State Park Boundary

Cattle Guard Turnout

Overview Area

Star Dune

Roads

Geomorphic Units:

Qelp - Large parabolic dunes

Qecd - Climbing dunes

Qemp - Marginal parabolic dunes

Qetd - Transverse dunes

Qeps - Partially stabilized dunes

Qes - Stabilized sand sheet

M

Figure 8. Geomorphic map of the Coral Pink Sand Dunes. Trace of theSevier fault from Sargent and Philpott (1987). See text for detailed de-criptions of the Quaternary-age map units.

0

0 21

1

NN

2 mi

3 km

G


sand does not travel well when conditions are damp (Ko-curek and Lancaster, 1999). Additional limitations on thesand available for transport are imposed by the needleduff observed on the surfaces of the dunes. Finally, astrees have increased in height, the resulting feedback is in-creased friction and reduced wind velocity at the surface,further limiting eolian transport. Thus, dune activity inthis portion of the dune field may have been waning sincethe establishment of the first trees over 450 years ago.Active Core of Transverse Dunes (Qetd)

The most active portion of the dune field is character-ized by the presence of large, well-developed transversedunes, amid very sparse vegetation. These are the signa-ture dunes of the park, and this area is the most popularwith the OHV recreationists. This portion of the dunefield lies on the valley floor immediately below the west-facing escarpment of the Moquith Mountains (figure 8).

Transverse dunes, generally associated with unidirec-tional winds, are asymmetric ridges of sand with a singleslipface orientation. The ridge crests are oriented perpen-dicular or transverse to the prevailing winds (Summer-field, 1991). The crest of the transverse dunes at CPSDSPhave a prominent northwest-southeast orientation, indi-cating a prevailing southwesterly wind, and rise 30 to 50feet (9–15 m) above their adjacent interdune swales.

Two of the three major types of transverse dunes arepresent in the Coral Pink Sand Dunes (isolated barchansare not present here). In the southern portion of the mapunit, the dunes have relatively straight crests and are thusclassified as transverse ridges (figure 7). The crest-to-crestspacing of the ridges is typically 600 to 700 feet (183–213m). Simple transverse ridges are relatively rare. Appar-ently, minor surface irregularities commonly createcorkscrew-like vortices in the airflow over a transverseridge. These vortices distort the ridge into a sinuous formcalled a barchanoid ridge (Summerfield, 1991). The trans-verse ridges at CPSDSP grade downwind (to the north)into classic barchanoid ridges (figure 10), with an increase

in the dune spacing (up to 1,000 feet or 305 m). Thesebarchanoid ridges display the characteristic alternatinglinguoid (protruding) and barchanoid (recessed) elements(figure 7). Owing to the complex airflow over these sinu-ous ridges, the barchanoid element of one ridge is com-monly aligned with a linguoid element in the adjacentridge, creating a complex crestal network referred to asaklé or fishscale patterns (Pye and Tsoar, 1990). A weakaklé pattern is present at the northernmost end of this mapunit, where the barchanoid ridges impinge upon thebedrock escarpment to the east.

The convergence between the barchanoid ridges andthe bedrock escarpment has also produced several echodunes (figure 11). Echo and climbing dunes (discussedbelow) are examples of topographically controlled dunes,dunes that owe their existence and morphology to interac-tions between sand-transporting winds and topographicobstacles (Lancaster, 1995). Echo dunes commonly form infront of cliffs (slope > 50 degrees). A vortex forms immedi-ately upwind of the cliff that acts to “sweep out” a corri-dor between the cliff and the slipface of the dune (Cookeand others, 1993). The result is the slipface of the dunetends to parallel, or echo, the shape of the cliff face. Echodunes are best developed where the prevailing wind isnear perpendicular to the trend of the cliff or escarpment.Since the sand-transporting winds within the park strikethe bedrock escarpment obliquely, only the eastern end ofthe barchanoid ridges display this echo morphology.Small Marginal Parabolic Dunes (Qemp)

Along the western margin of the dune field, approxi-mately a quarter mile south of the “cattle guard” area (seeClassic Geological Sites section) is an area dominated bysmall parabolic dunes (figure 8). Parabolic dunes are U- orV-shaped in plan with two trailing arms that point upwind(figure 7). Typically, there is a large mound of sand with a

382

Figure 10. View to the east of barchanoid ridges within the active coreof transverse dunes (Qetd). Note the steep slipface and sinuous crest ofthe dune.

Figure 9. View to the south of ponderosa pines (Pinus ponderosa)growing among partially stabilized dunes (Qeps) at the southern end ofthe dune field.

steep slipface at the downwind end of the dune. The out-side slopes of the trailing arms are almost always vegetat-ed (Pye and Tsoar, 1990). Parabolic dunes are common insand accumulations stabilized by vegetation. Localizeddisturbance of the vegetative cover and subsequent eolianerosion can give rise to circular or elliptical depressionscalled blowouts. The eroded sand is then depositeddownwind, near the edge of the disturbed area, as a para-bolic dune (Summerfield, 1991).

The orientation of these parabolic dunes suggest theyare controlled by the same wind regime responsible for thelarge area of transverse dunes. However, these are muchsmaller features, generally less than 12 feet (3.7 m) inheight.

Climbing Dunes (Qecd)

Wind encountering the windward face of a topo-graphic obstacle, such as a hill or scarp, with a slope of 30to 50 degrees will be decelerated such that a dune is de-posited upwind of the obstacle. In this situation, unlike anecho dune, the fixed eddy or vortex is small or absent.Thus, the dune can bank up against the scarp as a sandyramp, creating a so-called climbing dune. When the rampattains an equilibrium profile sand can then be carried upand over it (Cooke and others, 1993). This is the processthat has taken place at the north end of the park where theactive transverse dunes obliquely converge with the fault-line scarp of the Sevier fault (figure 8). Figure 12 is aground-penetrating-radar (GPR) transect from the areathat clearly shows a bedrock scarp buried by a ramp ofclimbing dunes. The climbing dunes have a slope of 5 to12 degrees and display well preserved internal slip faces.The original slope of the bedrock scarp is estimated tohave been 45 to 50 degrees, based on the profile of small

outcrops of Navajo Sandstone poking through the sandramp. The area of climbing dunes is more heavily vege-tated than the active core of transverse dunes. The dunemorphology is a mixture of poorly developed transverseand parabolic forms.Large Parabolic Dunes (Qelp)

Where the climbing dunes reach the top of the escarp-ment they grade into an extensive geomorphic unit domi-nated by parabolic dunes (figure 8). This geomorphic unit,essentially the northern half of the dune field, lies beyondthe park on land administered by the BLM. Slipface ori-entation in this part of the dune field indicates a morewesterly prevailing wind (S. 50˚ W.) compared to the areaof transverse dunes (S. 40˚ W.), probably owing to deflec-tion as the regional wind is funneled through the gap be-tween the eastern escarpment and Esplin Point. The top ofthe escarpment appears to act as a flow-expansion point,which reduces wind velocity and promotes deposition.

These parabolic dunes are much larger than thosealong the margin of the dune field farther south. The com-plex pattern of coalesced and nested dunes in this area,typical of most occurrences of vegetated parabolic dunes,is suggestive of alternating periods of dune migration andstabilization (Cooke and others, 1993). Here the rows ofcoalesced dunes have an average crest-to-crest spacing ofapproximately 1,300 feet (396 m). Large circular to ellipti-cal blowouts are present upwind of many of the dunes. Acharacteristic feature of this part of the dune field is thepresence of small groves of ponderosa pines in the inter-dune swales. Unlike the ponderosa pines present at theextreme southern end of the dune field, these occur atmore typical elevations of 6,100 to 6,500 feet (1,859 to 1,981m).

Most of the eastern and western boundaries of this ge-omorphic unit are marked by what is best termed a vege-tated linear dune. The western feature is quite evident asone travels on Hancock Road. Linear dunes are orientedroughly parallel to the prevailing wind (figure 7). Thereare many subtypes and their origin is generally poorly un-derstood (Cooke and others, 1993). However, these fea-tures are clearly the remnant arms of vegetated parabolicdunes, the noses and other arms having been blown away.Fluvial Deposits (Qa, Qp, Qae)

Besides the eolian deposits and landforms, a variety ofunconsolidated water-laid deposits occur in the park area(figures 3 and 4). The sand and gravel associated with ac-tive stream courses are mapped as alluvium (Qa). Thelargest accumulation of alluvium within the park coin-cides with the channel of Sand Wash, located along theeastern margin of the dune field. The most widespreadgroup of unconsolidated deposits in the park area is bestclassified as mixed eolian and alluvial deposits (Qae) (fig-ure 3). These deposits form when accumulations of wind-blown sand are later reworked by sheetwash or channelflooding during torrential rains (Doelling and others,


383

Figure 11. View to the east of a barchanoid ridge (modern dune sand)impinging on an outcrop of Navajo Sandstone (Jurassic dune sand).Local airflow has modified the barchanoid ridge into an echo dune.Young field assistant in the lower left corner is approximately 4.6 feet(1.5 m) tall.


1989). These mixed deposits cover much of the area im-mediately west of the dune field, including the park camp-ground. Unconsolidated gravel deposits lying on bedrockat elevations significantly above the active channels arepediment or terrace deposits (Qp). These remnant de-posits indicate that the valley floor once stood at a higherelevation and has since been lowered by erosion. In thepark area, pediment gravels are located along both thebase of the fault-controlled escarpment to the east and thebase of Esplin Point to the west (figure 3). The pedimentgravels (Qp) are the oldest of the Quaternary deposits inthe area, probably deposited during the Pleistocene Epoch(approximately 2.6 million to 10,000 years ago). All of theother Quaternary units, both aeolian and fluvial, weremost likely deposited during the Holocene Epoch, or thelast 10,000 years.Succession of Dune Forms at Coral Pink Sand Dunes

The succession of dune types within this dune fieldcan be generalized as a downwind sequence (south tonorth) of stabilized and partially stabilized sand sheetsand dunes, transverse ridges, barchanoid ridges, climbingdunes, and parabolic dunes. This is a complex sequencethat reflects the interplay between topography, sedimentsupply, vegetation, and possibly age. The progression offorms at Coral Pink Sand Dunes is best understood by con-sidering it to be two subsequences with different primarycontrols.

The southern half of the sequence represents adjust-ments in form and process to a topographically controlledincrease in sand accumulation with increasing distancefrom the source area. This subsequence culminates in thedeposition of the climbing dunes against the fault-linescarp of the Sevier fault. Other dune fields of the western

United States with similar sequences and a pronouncedtopographic control on dune successions include the Lyn-ndyl Dunes of Utah and Great Sand Dunes National Mon-ument, Colorado (Sack, 1987).

The northern subsequence, climbing dunes to para-bolic dunes, reflects a decrease in sand accumulationdownwind of the source area, owing to a progressive de-crease in the sediment supply and an accompanying in-crease in vegetation. Similar controls affect the sequenceof dunes at the Killpecker Dunes of Wyoming and WhiteSands National Monument, New Mexico (Sack, 1987).

STRUCTURAL GEOLOGYRegional Setting

Coral Pink Sand Dunes State Park (CPSDSP) lies with-in the zone of structural transition between two major ge-ologic or geomorphic provinces of western North Ameri-ca; the Great Basin section of the Basin and RangeProvince, and the Colorado Plateau province (figure 13).These two provinces are distinguished from one anotherbased on differences in their depositional history and styleof structural deformation and related differences in theirtopography. West of Cedar City and the Hurricane Cliffs(figure 13) lies the Great Basin, an area characterized bynorth-south-trending fault-block mountain ranges sepa-rated by broad intervening basins with internal drainageand filled with Tertiary to Quaternary alluvium and lakedeposits. The boundary between the transition zone andthe Great Basin coincides with the trend of the Paragonah,Gunlock, and Grand Wash faults (figure 13). The bound-ary between the transition zone and the more stable Col-orado Plateau interior to the east is not as sharply demar-

384

0 m 100 m 200 m 300 m 400 m 500 m

5 m

10 m

15 m

20 m

25 m

climbing dune foresets

dune-bedrock interface Scar

p-lik

efe

atur

e

vertical exaggeration 10x

SW NE

20 m

25 m

10 m

15 m

20 m

0 m

5 m

100 m 200 m 300 m 400 m 500 mm

Figure 12. 500-m GPR transect within the area of climbing dunes (Qecd). The transect crossed bedrock outcrops whose locations were used to in-terpret the subsurface contact between the active dunes and the underlying bedrock (dashed line). Note the internal radar stratigraphy within theclimbing dune.

cated, but generally coincides with the Paunsaugunt fault(Anderson and Christenson, 1989).

The transition zone, designated the High Plateaus sec-tion (of the Colorado Plateau) by some workers (for exam-ple Davis, 1999), lies between the core areas of the Basinand Range and Colorado Plateau and thus displays char-acteristics of both major provinces. In simplest terms, thetransition zone is that part of the Colorado Plateau thatwas affected by Basin and Range extensional faulting,more characteristic of the style of deformation further tothe west. Basin and Range extension was initiated duringMiocene time (approximately 15 million years ago) and isstill active today (Davis, 1999). This deformation createdeast-west-tilted structural blocks bounded by widelyspaced, north-south-trending, high-angle normal faults;namely the Hurricane, Sevier, and Paunsaugunt faults(Anderson and Christenson, 1989; Davis, 1999). Coral

Pink Sand Dunes State Park lies astride the Sevier fault(figures 3 and 13), the central of the three major Basin-and-Range faults of southern Utah.

Sevier FaultThe Sevier fault in Utah is the northern portion of a

300-mile-long (500 km) fault zone that includes theToroweap and Aubray faults in Arizona (Anderson andChristenson, 1989; Davis, 1999; Lund and others, 2008).The fault extends from the Grand Canyon to central Utah,where it loses its discrete character within Miocene vol-canic rocks. The north-south trace of the Sevier fault cutsCPSDSP into two roughly equal halves (figure 3).

In the vicinity of the park, the Sevier fault strikes ap-proximately N. 30˚ E. and dips about 75˚ W. It is not a sin-gle fault plane but rather a complex zone of braided fault


385

NN

• • • • • • • • • • • • • • • • • • • • • • • • • • • •• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

P

Pa

Basin & RangeProvince

0

0

50

50

100 150

100 mi

UTAH

ARIZONA

NE

VAD

A

LAS VEGAS

CPSDSPKanab

ColoradoPlateau

Province

Richfield

Sevi

erFa

ult

St.George

G

RR

GW

WH

T

CedarCity

< Basin and Range /Colorado Plateau

structural transition >

••• • Province boundariesState lines

CPSDSP = Coral Pink SandDunes State Park

Normal fault, bar & ball ondown-thrown side

G - Gunlock faultP- Paragonah faultT - Toroweap, W- WashingtonH - Hurricane faultGW -Grand Wash faultPa - Paunsaugunt faultRR - Reef Reservoir fault

EXPLANATION

Figure 13. Basin and Range/Colorado Plateau structural transition zone in southwest Utah (modified from Anderson and Christenson, 1989). Notethe location of Coral Pink Sand Dunes State Park along the trace of the Sevier fault.


segments (figures 3 and 4). As a normal fault, the Sevierfault was created by extensional stresses that produced rel-ative motion along the fault zone such that rocks lyingabove the fault plane and west of the fault’s surface tracemoved down relative to the up-thrown block of rocks be-neath the dipping fault zone and east of the surface trace(figure 4). Strata within the western down-thrown blockdip very gently (< 5˚) to the north, as evidenced by the ex-posures in the face of Esplin Point. The blocks between thefault braids, within the up-thrown eastern block, are local-ly tilted up to 60˚ (Doelling and others, 1989). Farther east,away from the fault zone, the strata again display a verygentle north to northeast dip.

Total normal displacement along the Sevier fault zoneincreases from south to north, ranging from approximate-ly 1,560 feet (475 m) at the Utah-Arizona border to 4,920feet (1,500 m) in southern Sevier County. The stratigraph-ic displacement within the park is obvious because beds ofthe upper portion of the Navajo Sandstone (Lower Juras-sic), exposed on the western down-thrown side of thefault, are abutted against rocks of the up-thrown blockranging from Moenave Formation (Lower Jurassic) tolower portions of the Navajo Sandstone (figures 3 and 4).Anderson and Christenson (1989) estimated the total dis-placement in the vicinity of CPSDSP at 1,560 feet (475 m).This fault offset has produced the prominent west-facingbedrock escarpment, a fault-line scarp, east of the camp-ground and main area of active dunes (figure 2).

Movement along the Sevier fault probably began 15 to12 million years ago, corresponding to the beginning ofBasin and Range deformation in this area (Davis, 1999).Assessing the recent history of movement along this faulthas proven to be difficult (Lund and others, 2008).Throughout most of its length in southern Utah, the Sevi-er fault cuts through Mesozoic-age bedrock. This limitsthe opportunity to document the recent history of move-ment, which is typically accomplished through the studyof fault scarps and stratigraphic offset within unconsoli-dated surficial deposits (Quaternary age). However, stud-ies of the Sevier fault north and south of the park area sug-gest movements resulting in surface faulting during thelate Tertiary to Pleistocene (approximately 4 to 0.56 millionyears ago) (Anderson and Christenson, 1989; Lund andothers, 2008). Two geomorphic aspects of the fault in thegeneral vicinity of the park provide indirect support forPleistocene deformation (Anderson and Christenson,1989). First, the west-facing bedrock scarp displays a dis-tinctive profile, with a moderately sloping rounded upperpart and steeper cliff-forming lower part. In several placeswithin the park these two slope elements are separated bya small bench or step in the scarp profile. Anderson andChristenson (1989) suggest that recent movement, ratherthan differential erosion, is responsible for these profilecharacteristics. The second feature suggestive of recentmovement along the fault is the presence of a small butdistinctive closed basin at Clay Flat, approximately 1 mile(1.6 km) north of the park. This area of deposition (de-

pocenter) is located at a major left step in the trace of theSevier fault. The local stress field has apparently created asmall pull-apart basin that interrupts the northward trans-port of sediment by ephemeral streams in Yellow Jacketand Sethys Canyons. In order to maintain such a local de-pocenter, active fault movement and related subsidenceduring the late Quaternary is probably required (Ander-son and Christenson, 1989).

Coral Pink Sand Dunes State Park lies within thesouthern end of the Intermountain Seismic Belt (ISB), thenorth-south-trending zone of seismicity extending fromnorthwestern Montana to the Las Vegas, Nevada, area.The Wasatch fault of northern Utah is the most infamousfeature within this zone of active earthquake activity.Earthquakes in the southern portion of the ISB are gener-ally small to moderate in size, typically less than Richtermagnitude (ML) 4.0, and in many cases are not associatedwith the mapped Quaternary faults of the area (Andersonand Christenson, 1989). The largest historic earthquake inthe vicinity of CPSDSP was the magnitude 6.3 (ML) PineValley earthquake in 1902, located 21 miles (33 km) northof St. George. Estimated magnitude-5.7 and 5.5+ (ML)earthquakes occurred east of the park near Kanab in 1887and 1959, respectively. Compared to the Hurricane fault40 miles (64 km) to the west, which generated the 1992 St.George earthquake (Mw = 5.8), the Sevier fault has had amuch lower level of seismicity during historic time. Onlytwo historic (1900-1985) earthquakes have been recordedalong the main trace of the Sevier fault in Utah (Davis,1999), neither of them within the park. Anderson andChristenson (1989) raise the possibility that the Sevier faultmay be locally deforming by continuous aseismic faultcreep, as opposed to large ground-rupturing earthquakes.Either way, the low-level earthquake activity, coupled withthe geomorphic character of the fault-line scarp notedabove, suggest that the Sevier fault is still active today.

Deformation-Band Shear ZonesThe Navajo Sandstone of southwestern Utah displays

a variety of outcrop-scale structures that resemble veins ordikes weathering out in positive relief from their hostsandstone. These features, called deformation bands, havebeen recently studied and analyzed in careful detail byDavis (1999) and Fossen and others (2007). Deformationbands are the result of porosity reduction, grain crushing,grain fracturing, and grain flow that occur during the de-formation of porous granular materials. Davis recognizestwo varieties of deformation bands, cataclastic and non-cataclastic, both of which are present in the Navajo Sand-stone within the vicinity of CPSDSP. Cataclastic deforma-tion bands are closely associated with the regional faultsand folds of the area, including the Sevier fault. Slip with-in these features is indicated by slickensides, and a dis-tinctive ladder structure is common. The most distinctivemanifestation of cataclastic deformation bands are the for-mation of “fault fins,” triangular blades of sandstone up

386

to 26 feet (8 m) tall (see cover photograph of Davis, 1999).Fault fins are produced by differential weathering and ero-sion; the reduced porosity within the deformation bandsincreases their resistance. Tilted deformation bands thenserve to protect underlying sandstone from erosion(Davis, 1999).

The second or noncataclastic type of deformationband is present in almost every outcrop of the uppermost33 to 50 feet (10–15 m) of the Navajo Sandstone (Davis,1999). The noncataclastic deformation bands are vein-likestructures that weather in positive relief but show minimalevidence of grain-scale fracturing and no slickensides.Davis (1999) concludes that noncataclastic deformationbands formed in Jurassic time by the loading of under-compacted dune sand of the Navajo Sandstone by sedi-ments that now make up the Carmel Formation. Aided bygroundwater infiltration, the Navajo sands failed by verti-cal compaction. The deformation bands mark the zoneswhere this volume reduction took place. The uppermostpart of the Navajo Sandstone outcrops near the northwestcorner of the park. Large blocks of sandstone, apparentlyquarried from that area, have been placed at camp sitesand along the loop road. “These blocks are museumpieces of . . . noncataclastic deformation bands” (Davis,1999, p. 66) (figure 14).

GEOLOGIC HISTORYA complete geologic history of the area around Coral

Pink Sand Dunes State Park (CPSDSP) is beyond the scopeof this paper. Interested readers should consult Petersonand Turner-Peterson’s (1989) overview of the ColoradoPlateau, as well as other papers in this volume. We willfocus on those geologic events that have specifically lefttheir mark on the landscape of the park. The most impor-tant events are the Triassic-Jurassic deposition of the GlenCanyon Group (Moenave, Kayenta, and Navajo Forma-tions), the Cretaceous to Cenozoic structural deformationof the Colorado Plateau, and the Quaternary formationand evolution of the active dunes. This selective history isprimarily abstracted from the work of Baars (1983), Stokes(1986), Harris and Tuttle (1990), Patton and others (1991),Barnes (1993), Morris and Stueben (1994), Elias (1997),Davis (1999), and Blakey and Ranney, 2008).

Jurassic Environmental ChangeFor much of the Paleozoic Era (about 54 to 250 million

years ago), Utah was situated on the western edge of theNorth American continent and dominated by shallow-ma-rine deposition. During Middle Triassic time (245 to 230million years ago) western Utah was cut off from the oceanto the west by a highland barrier, termed the Meso-cordilleran High (Stokes, 1986), which was located in whatis now eastern Nevada. To the east, in what is now west-ern Colorado, was another highland area called the Un-compahgre uplift, an eroded remnant of the AncestralRocky Mountains. Thus, the subsequent Triassic-Jurassic

formations of the Glen Canyon Group were deposited in asedimentary basin that formed between these two high-lands, dominated by terrestrial depositional environ-ments. The Mesocordilleran High blocked moisture-lad-den air masses and served to enhance arid conditions inUtah by creating a rainshadow. The Mesocordilleran Highwas also a source area for rivers flowing eastward and car-rying sediment into the basin.

Fluvial and lacustrine deposition during theLate-Tri-assic and Early Jurassic resulted in the sandstones, silt-stones, and freshwater limestone of the Moenave Forma-tion. During this time (approximately 200 to 196 millionyears ago) the North American continent was driftingnorthward through equatorial latitudes. The rock typesand sedimentary structures preserved within the MoenaveFormationsuggest a landscape of low-gradient streamchannels with broad floodplains containing lakes andmudflats. Fossil remains of large freshwater fish, similarto sturgeon, as well as numerous dinosaur footprints andtracks (see Milner and others, 2006), have been found inthe Whitmore Point Member.

As North America drifted north away from the equa-tor, the climate in what is now southern Utah became cool-er with wet summers and dry winters (Barnes, 1993).Streams deposited silt and sand in channels and on flood-plains; dinosaurs left their footprints in the damp sedi-ments that would become the Kayenta Formation (196 to190 million years ago).

Continued northward drift brought a large portion ofNorth America under the influence of subtropical highpressure, resulting in increasing aridity and the establish-ment of a huge Jurassic sand sea or erg. One estimate in-dicates that the erg may have covered 256 million squaremiles (6.63 x 105 km2) (Marzolf, 1988). However, the inter-tonguing of the Kayenta and Navajo Formations repre-sented by the Lamb Point and Tenney Canyon indicates an


387

Figure 14. Noncataclastic deformation bands in a boulder of NavajoSandstone along the campground loop, Coral Pink Sand Dunes StatePark.


initial period of time during which fluvial and eolian sys-tems vied for dominance in southern Utah. A Jurassicdesert, represented by the Lamb Point Tongue, appears tohave advanced from the east, burying and replacing thefluvial channels and floodplains of the Kayenta Forma-tion. The margins of the sandy desert temporarily reced-ed back to the east, replaced by fluvial environments of theTenney Canyon Tongue. The desert once again advancedand overwhelmed the fluvial landscape. Eolian deposi-tion, represented by the main body of the Navajo Sand-stone, then dominated this area for approximately 15 mil-lion years (190 to 175 million years ago).

Paleogeographic reconstructions suggest that the ergwas situated near the western edge of the continent be-tween approximately 18˚ and 25˚ north latitude, under theinfluence of onshore north-northwesterly winds associat-ed with the eastern edge of a subtropical high (Marzolf,1988; Peterson, 1988; Blakey, 2008; Blakey and Ranney,2008). A shallow marine shelf, estuary, or large lake bor-dered the lowland desert on the west. Peterson (1988) sug-gests that the source area for the sand was an uplifted areain central Montana. Rivers carried sediment westward tothe coast from this upland, longshore currents moved thesediment southward, and then waves and northwesterlywinds delivered the sediment to Utah. Marzolf (1988) sug-gests a different source for the sand. He suggests thatrivers draining northwestward from upland areas in Ari-zona delivered sediment to the head of the marine embay-ment to the west, where the silt and clay was winnowedand the sand was returned inland by the onshore winds.

Navajo Sandstone deposition ended, possibly the re-sult of climate change, in the vicinity of CPSDSP whenstreams, loaded with red mud, flooded the dunes and par-tially truncated them. For wind-blown sediments to bepreserved as eolian sandstones, they must be deposited inan actively subsiding basin. In the case of a coastal erg, ifdeposition does not keep pace with subsidence, the seawill transgress or flood the lowland. This is what occurredalong the western edge of the Navajo erg during MiddleJurassic time, resulting in a major unconformity (J-1) (fig-ure 6). A shallow seaway extended into southwesternUtah from the north.

The mostly marine sediments flooded the region andbecame the Sinawava Member of the Temple Cap Sand-stone (Middle Jurassic). Sea level fluctuated during Tem-ple Cap deposition, which allowed eolian deposition toquickly resume producing the White Thrown Member ofthe Temple Cap Sandstone. Sea level rose again, but thistime the sea flooded farther inland completely inundatingthe erg field resulting in deposition of limestone beds ofthe Co-op Creek Member of the Carmel Formation (Mid-dle Jurassic).

Upper Jurassic through Lower Cretaceous beds arenot present in the vicinity of the park and were probablyremoved by a long period of uplift and erosion associatedwith the Laramide orogeny. The nearest outcrops of Cre-taceous rocks are poor exposures of Upper Cretaceous

Tropic Shale and Dakota Sandstone located near Mt.Carmel. Thus, the modern sand dunes of the park lie di-rectly on Lower Jurassic formations, most notably theNavajo Sandstone. The boundary between the two aeolianunits, one lithified and one still moving, represents a hia-tus or gap in the stratigraphic record of approximately 180million years.

Cretaceous to Cenozoic TectonicsTowards the end of Mesozoic time, and continuing

into the Cenozoic, western North America was subjectedto a mountain-building episode known as the Laramideorogeny (approximately 90 to 50 million years ago). Thisorogeny was the result of compressional stresses generat-ed by plate-tectonic convergence between the NorthAmerican plate and the oceanic Farallon plate to the west.Subduction of the Farallon plate, and related folding andthrust faulting, gave rise to the Rocky Mountains. TheColorado Plateau was also greatly affected by this defor-mational event (see Davis and Bump, 2009). Compressionof the Colorado Plateau region was accommodated bybasement-cored reverse faulting and associated foldingwhich produced the Kaibab uplift, the San Rafael Swell,and the numerous broad anticlines and synclines of theKaiparowits Plateau, among other major structures. Sincethe beginning of the Laramide orogeny the nearly circularColorado Plateau has acted as a coherent structural blockand has been uplifted more than 6,600 feet (2,012 m) (Mor-ris and Stubben, 1994).

One hypothesis (Beghoul and Barazangi, 1989) sug-gests the Farallon plate may have been key to this uplift,becoming attached during subduction to the bottom of theNorth American plate in the vicinity of the ColoradoPlateau. During mid-Cenozoic time the Farallon plate sep-arated from the North American plate and sank into themantle. As the Farallon plate remnant sank, hot materialfrom the asthenosphere rose to fill the void. This materialthermally expanded as it rose and some of it intruded thelower continental crust, thereby increasing its thicknessand temperature. The Colorado Plateau is thus seen tohave been uplifted by isostatic compensation resultingfrom a combination of heating and increased crustal thick-ness.

A second deformational event, Basin and Range ex-tension, also had a pronounced effect on the geology of thepark. This episode of normal faulting and crustal exten-sion began approximately 15 million years ago, in mid-Miocene time, and is still active today (Davis, 1999). Thereis no consensus among geoscientists as to the plate-tecton-ic explanation for this extension. Some (for example Stew-art, 1971) suggest that a slowing of convergence betweenthe Farallon and North American plates coupled with arollback of the subducting Farallon plate essentially creat-ed back-arc spreading within the continental crust. Others(for example Dickinson, 1979) have suggested that theNorth American plate completely overrode the mid-oceanridge along the western boundary of the Farallon plate,

388

thus creating a window in the subducting slab throughwhich heat and magma from the asthenosphere could riseto lift and thin the overlying continental crust. Either way,the beginning of this extensional deformation was coinci-dent with the cessation of subduction along the Californiacoast and the initiation of the transform boundary be-tween the North American and Pacific plates marked bythe San Andreas fault.

Whatever its cause, this extensional deformation cre-ated the distinctive topography of the Basin and RangeProvince, including the Great Basin section in Nevada andwestern Utah, as well as the three major high-angle nor-mal faults located within the southwestern margin of theColorado Plateau—namely the Hurricane, Sevier, andPaunsaugunt faults (figure 13). Movement along the Sevi-er fault has had a major influence on the geologic historyof the park, creating the fault-line scarp that dominates thepresent topography. Without this fault scarp, the variedMesozoic strata would not be exposed within the park. Inaddition, this scarp is in part responsible for the concen-tration and accumulation of wind-blown sand at this loca-tion.

The uplift of the Colorado Plateau caused a majorchange in the nature of the dominant geologic processes.Whereas the Mesozoic Era was dominated by deposition,the subsequent Cenozoic geologic history of the ColoradoPlateau has been dominated by erosion, producing theworld-renowned canyonlands and cliff-and-bench topog-raphy. This erosion is very evident in the landscape ofCPSDSP, exposing the picturesque formations of the GlenCanyon Group. However, localized eolian deposition dur-ing the Holocene Epoch has created the features for whichthe park is named.

Late Holocene History of the Dune FieldThe Quaternary Period is subdivided into the Pleis-

tocene and Holocene Epochs, with the boundary betweenthe two occurring approximately 10,000 years ago. ThePleistocene is noted for its alternating glacial and inter-gla-cial conditions. The Holocene is essentially the most re-cent and on-going inter-glacial time period that beganwhen the last ice age (the Wisconsin glaciation) ended.Late Pleistocene full-glacial (22,000 to 18,000 years ago)conditions on the Colorado Plateau probably inhibitedmajor eolian transport and deposition. Paleobotanical ev-idence suggests that late Wisconsin winters were wetterand summers were cooler and drier than present (Betan-court, 1990). Drier summers probably reflect a decrease inthe availability of subtropical moisture via the southwestmonsoon, whereas wetter winters were caused by a south-ern shift in the polar jet and an increased occurrence of Pa-cific storms and associated frontal precipitation over thePlateau. The result was coniferous woodlands, includinglimber pine, juniper, Colorado blue spruce, Douglas-fir,and sagebrush, were present in areas that now supportpinyon-juniper woodlands. The area of the Coral PinkSand Dunes may have supported a coniferous woodland

during the late Pleistocene. Notably, ponderosa pine wasapparently absent from the Colorado Plateau during thistime, due to a lack of summer moisture and/or a reductionof thunderstorm activity and lightning strikes; ponderosapine distribution is in part linked to fire (Betancourt, 1990).

The Pleistocene-Holocene transition on the ColoradoPlateau is generally marked by gradual changes in vegeta-tion that indicate warming and a progressive decrease ineffective moisture. Macrobotanical remains from the Es-calante River basin indicate decreased abundances of mes-ophytic plants and an upslope retreat of Douglas fir andspruce (Withers and Mead, 1993). However, the early tomid-Holocene (10,000 to 6,000 years ago) on the ColoradoPlateau saw a significant increase in summer precipitation(and thunderstorm activity?) (Betancourt, 1990). Pon-derosa pine expanded during this period beyond even itspresent distribution. Even during the latter part of this pe-riod, commonly associated with hot-dry conditions inother parts of the southwest (the altithermal), the Col-orado Plateau appears to have been wetter than today.The most likely explanation is more abundant subtropicalmoisture from an intensified Bermuda High and south-west monsoon (Betancourt, 1990). Pollen evidence fromPosy Lake on the Aquarius Plateau, north of the park, in-dicates that the period of enhanced summer precipitationpersisted until 5,500 years before present (B.P.) (Shafer,1989). Lake levels were lowest at this site from about 5,000to 3,500 years B.P., indicating greater aridity than today. .The oldest dates that we have obtained from the depositsin the dune field, approximately 4000 yr B.P. (discussedbelow) coincide with this period of regional aridity.Late Holocene Geochronology and Dune Activity

Using a combination of dating methods, including ra-diocarbon dating, optically stimulated luminescence(OSL) dating, and dendrochronology, we have been able toreconstruct the changes in the environment and resultingchanges in dune activity over the past several millenniapreserved within the Holocene stratigraphy of the dunefield (Wilkins and others, 2005; 2007).

Dune migration and erosional lowering of the upwindsurfaces of some of the tallest transverse dunes (Qetd) hasuncovered the tops of several large ponderosa pine snags,in situ and upright, presumably killed through deep bur-ial by passing dunes. Radiocarbon dating (note: all agesare reported at the 2σ level and calibrated using int-cal09.14c; Stuiver and Reimer, 1993) of the outer rings of abranch near the top of a ten meter-tall snag gives an age ofdeath of 190±50 14C yr BP (Cal. 1643–1950 CE; Beta 181947);ring counting revealed 188 rings, indicating that the standof trees was established for several hundred years before itwas buried.

In a swale (approximately 500 meters north of thesnags) between active transverse dunes, a fallen and heav-ily weathered ponderosa pine rooted in bedrock has beenuncovered by dune migration. Radiocarbon dating of thepith, or center rings, at its base yields an age germination


389


of 250±40 14C yr BP (Cal. 1485–1950; Beta 181943); countingthe rings from the pith to the outer edge of the stem indi-cated a minimum age of 75 years before its death

One hundred meters west of the downed ponderosa,recent dune movement has also exhumed a paleosurfaceconsisting of a muddy soil capping older eolian sediments.The older sediments are two to three meters of friablecross-bedded eolian sand comprised of three units, eachone to two meters thick, all with very similar grain miner-alogy, color, and sorting. The uppermost eolian unit is un-conformably overlain by a more resistant silty mud layer.The mud and soil layer, 0.25 meters thick, exhibits flowfeatures at the base, including rip-up clasts, in the silty ma-trix characteristic of a sheetflood deposit. The mud layeris topped by an organic layer consisting of what appears tobe a duff layer from woodland forest floor, including char-coal and plant macrofossils of juniper berries, cones (Pinusedulis), stems and roots in situ, indicating a pinyon-ju-niper woodland had once occupied the site.

A method of dating, called optically stimulated lumi-nescence (OSL) dating, is a useful tool in desert environ-ments to date the time since sand was last exposed to sun-light and buried. Samples of sand for OSL dating were col-lected at the upper contact of each buried eolian unitbelow the exhumed soil. Dating of the samples record anupper limiting age for the termination of eolian activityand deposition related to that unit. The OSL ages for dep-osition and burial are in the proper stratigraphic order,with the upper unit terminating 0.51±0.06 ka (1435–1555CE) at a depth of 0.7 m below the surface. The two lowerunits had ages of 2.8±0.22 ka (2.0 m, 1015–525 BCE) and4.1±0.19 ka (2.8 m, 2205–1905 BCE), respectively (Wilkinsand others, 2007).

Radiocarbon dates were also obtained from this site.A sample of the silty mud was collected from the base ofthe layer and was dated using the bulk organic carboncomponent, yielding an age of 470±50 14C yr B.P. (Cal.1400–1490 CE; Beta 190240). A pinecone embedded in thesurface soil and duff layer yielded an age of 200±40 14C yrB.P. (Cal. 1761–1803 CE; Beta 181944; Wilkins and others,2005).

The OSL dating of the eolian sand and radiocarbondating of the exhumed soil, plant matter, and tree snagsprovide the basis for reconstructing the latest Holocene en-vironments in the Coral Pink Sad Dunes. The lumines-cence dates indicate that eolian activity has been present inthe park area for at least the last 4000 years, though ourunderstanding is still limited as to how episodic or per-sistent the earlier periods of eolian activity were. Theyoungest eolian unit, initiating some time prior to its 500OSL yr BP age, is likely related to one or more of themegadroughts in the western US in the 13th, 14th, andearly 15th centuries (Oviatt, 1988; Stahle and others, 2007;Mensing and others, 2008) and also coincides with the re-gional abandonment of Anasazi farming settlements (Ben-son and others, 2007). Eolian activity in the park area dur-ing that time is broadly similar to activity reported for the

Great Sand Dunes (Colorado; Forman and others, 2006)and the Tusayan Dunes (Arizona; Stokes and Breed, 2003).

The termination of eolian activity and apparentlyabrupt transition to wetter conditions—as recorded in thewaterlain muds over the eolian sediments, expansion ofthe pinyon-juniper woodland into the dunes, and estab-lishment of the ponderosa in the dune field—is dated tothe start of the Little Ice Age (LIA). The onset of dune sta-bility corresponds to a period of valley deposition in thesurrounding region, as reported by Hereford (2002), aswell as an absence of prolonged, multidecadal drought(Cook and others, 2004). Hereford (2002) proposes that theaccumulation of valley fill resulted from a general absenceof large floods. In the area of the Coral Pink Sand Dunes,the Little Ice Age appears to have been a time of landscapestability and diminished eolian transport, interspersedwith periods of drought and resurgence of eolian activitysuch as that which led to the burial and death of pon-derosa pines in the active core of the dune field. The agesof the snags in the dunes (250±50 and 190±50 14C yr BP)correspond to a period of eolian activity at the Great SandDunes in Colorado (Forman and others, 2006), centeredaround 1800 CE that is attributed to a period of reducedprecipitation in the region.

The last half of the 19th century marked the end of theLittle Ice Age in the Colorado Plateau region (Hereford,2002). Data from Cook and others (2004) and Stahle andothers (2007) both report a number of short-term severedroughts during this period; tree rings in cores collected inCPSDSP also record these droughts as very narrow rings,most notably a severe drought in the 1890s that lasted sev-eral years.

The end of the Little Ice Age is also marked by a returnof frequent large floods in the region (Hereford, 2002; Ely,1997)—the early 20th century is marked by a prolongedperiod of increased moisture termed a “pluvial” by Stahleand others (2007), and is recorded in CPSDSP tree cores asvery wide rings. The pluvial was followed immediately bythe extreme continental-scale drought of the 1930s knownas “the Dust Bowl” (Stahle and others, 2007). H.E. Grego-ry (1950), the first geologist to record a visit to the CoralPink Sand Dunes, reported a highly active dune field in1937; this activity was likely a response to the drought con-ditions still present from the early part of that decade. Aer-ial photos of the dunes acquired in 1957 and 1960 alsoshow very active dunes, likely a response to the extremedrought in the southwestern U.S. in the 1950s (Stahle andothers, 2007).

In the four decades since those photos, 1961–2000,local climate records record a slight increase in mean an-nual precipitation over the previous three decades,1931–1960. In response, the vegetation at the margins ofthe dune field expanded and become more broadly estab-lished in the dunes—aerial photos acquired in 1997 showa contraction of active dune areas since 1960 (Wilkins andFord, 2007).

390

GEOLOGIC UNIQUENESS OF THE PARKNaturalist Freeman Tilden (1970) concluded that sand

dunes are nature’s nudes. Indeed, there is something verypleasing, even sensual, about the curves and bare contoursof active dunes. Dunes bring to mind romantic adventuresin Old World deserts, as well as frontier tales from theAmerican West. It is no wonder then that the Coral PinksSand Dunes have been used as locations for several Holly-wood films: Arabian Nights (1942)—“corny escapist stuff”with an enslaved Sheherazade and a retired Sinbad theSailor; Mackenna’s Gold (1969)—“overblown adventuresaga about search for lost canyon of gold,” and One LittleIndian (1973)—“AWOL cavalry corporal escaping throughthe desert with young Indian boy and a camel” (Maltin,1997). Geologic history, OHV recreation, film making, andwildlife habitat (to be discussed below)—all make this isan interesting, important, and valuable landscape.

Source of the SandAlthough the commercial use of the Coral Pink Sand

Dunes dates to 1942, the earliest scientific treatment of thedune field is Herbert Gregory’s (1950) description of thedunes in his general survey of the geology and geographyof the Zion National Park area:

Likewise the disintegrated products from the bare sand-stones of South Mountain are carried eastward acrossMoccasin Terrace to form the dunes along the cliffs atSand Canyon and the great sand flats at the head ofThree Lakes Canyon. With local accretions en routethese deposits are redistributed over Wygaret Terraceand eastward to the base of the Kaiparowits plateau.During sandstorms great quantities of the finest dustrise high into the air and move eastward to distantplaces or after darkening the skies with whirling cloudsreturn to the place from which they came. . . . Anotherlarge area of dunes extends from the head of Cotton-wood and Water Canyons northwestward across Sandand Yellow Jacket Valleys to the base of the Block Mesasnear Esplin Point. For a distance of about 8 milesnorthwest of Riggs Spring dunes as much as 200 feethigh are nearly stationary; they support growths ofpiñon and juniper and in wet seasons hold lakes andgive rise to springs. At the head of Sand Canyon theyare in motion and in their migration are burying anduncovering trees and converting the original escarp-ment of the Sevier fault into a slope (Gregory, 1950, p.188).

Botanist Elias Castle (1954) may have been the first touse the term “coral pink” to describe the color of thedunes. His report, probably the first to focus specificallyon the dunes in the park area, summarizes the soil condi-tions and their influence on the various vegetation com-munities within the dune field.

Kane County has many aeolian deposits though mostare quite small. In addition to the Coral Pink Sand Dunes,the Sand Hills, which extend across U.S. 89 between theSevier fault and Kanab Creek, are large enough to be por-trayed on regional geologic maps (Sargent and Philpott,1987; Doelling and others, 1989). Of the many Jurassicsandstone units exposed in the vicinity, the Navajo, Page,and Entrada Sandstones are thought to be especially pro-ductive in supplying sand for dunes. Doelling and others(1989) suggest that the friable and poorly cemented mid-dle portion of the Navajo Sandstone provided the bulk ofthe reddish sand for the Coral Pink Sand Dunes. The red-dish orange (coral pink) color of the quartz sand is likelyinherited from iron oxide grain coatings and cement of theNavajo Sandstone (Chan, 2010, personal communication).

Although Gregory (1950) indicates a western source(South Mountain) for the sand, the prominent southwest-northeast orientation of the dune field and prevailingsouthwesterly winds suggests to us a source to the south-west. The middle pink portion of the Navajo Sandstone isexposed extensively across the Moccasin Terrace, south-west of the park. In addition, extensive surficial deposits,primarily deposits of fine-grained sand deposited bymixed alluvial and eolian processes (Qae), cover the Moc-casin Terrace (Doelling and others, 1989). These unconsol-idated deposits are a likely source for the sand that com-prises the Coral Pink Sand Dunes.

This suggested source indicates a very interesting paththrough the rock cycle for these sand grains—ancient eo-lian sandstones weathering to produce modern aeoliansediments. Recall that recent studies (Dickinson andGehrels, 2003) suggest that rocks as far away as the Ap-palachian Mountains were weathered and eroded to pro-duce sediment that was transported westward by rivers toa marine embayment during Jurassic time. This sedimentwas transported southward by ocean currents, progres-sively winnowed by wave and tidal action, and eventual-ly transported back to land by onshore winds. These sandgrains were then deposited in large dunes of a lowlanderg. The dunes were buried by younger sediments andeventually lithified to sandstone, what we now call theNavajo Sandstone. These eolian sandstones remainedburied for perhaps 125 million years or more. Uplift of theColorado Plateau during the Tertiary Period set the stagefor the recent and on going weathering and erosion, re-leasing the sand grains from the ancient rocks. Onceagain, perhaps 170 million years latter, the sand grains areeroded and again transported by wind. The modern windregime has concentrated many of these sand grains in theCoral Pink Sand Dunes and continues to slowly transportthem to the northeast, to encroach on and bury JurassicNavajo Sandstone outcrops (figure 11).

Formation of the Dune FieldAn obvious question is why have these modern eolian

sediments been concentrated here? Although no localwind data are available, a park brochure (Utah State Parks


391


and Recreation, no date) and interpretative signage de-scribe the existence of several regional winds that con-verge in the vicinity of the park. A break in the VermillionCliffs south of the park, due to offset on the Sevier fault,likely allows winds to accelerate through the gap betweenthe Moquith and Moccasin Mountains and to entrain loosesediment on the Moccasin Terrace. As the gap widens inthe vicinity of the park and the winds impinge upon thebedrock scarp of the Sevier fault, velocity may decrease toallow deposition. Named for an Italian physicist, the ac-celeration of air or liquid as it is forced through a narrowconstriction is called the Venturi effect. The southern por-tion of the dune field has a distinct north-south orienta-tion, possibly reflecting the influence of this southerlywind.

About 2 miles (3.2 km) north of the southern end ofthe dune field, in the generally vicinity of the park camp-ground, the orientation of the dune field changes to south-west-northeast. This may reflect the influence of prevail-ing southwesterly winds blowing up from Rosy Canyon,another wind gap in the Vermillion Cliffs to the southwest.These southwesterly winds may be funneled between theMoccasin Mountains to the south and the Block Mesas, in-cluding Esplin Point, to the north. The fault-related scarpon the eastern edge of the dune field may act to slow thesewinds as well, thus initiating deposition at the base of thescarp. The sand deposited at the base of the scarp is theobvious source for the dunes that climb the fault scarp andextend approximately 4 miles (6.4 km) northeast onto BLMland. This secondary sand movement may be facilitatedby an acceleration of the wind as it flows through the nar-row, 1-mile (1.6 km) gap between Esplin Point and thescarp of the Sevier fault. Thus, the Coral Pink Sand Dunesappear to owe their existence to an abundant supply offine-grained sand from the Navajo Sandstone and otherunits, strong prevailing southerly and southwesterlywinds, and the decelerating influence of the Sevier fault-line scarp. Specific data about the local wind regime areneeded to confirm the highly speculative hypothesis pre-sented above.

LOCATION AND DESCRIPTIONOF CLASSIC GEOLOGICAL SITES

WITHIN THE PARKOverview Area and Nature Trail

A parking lot and overview area (figure 8) is locatedon the east side of the campground road between theranger station and the campground. Most non-OHV visi-tors to the park utilize the boardwalk here to access thedune field. The overview is built atop what is best termeda vegetated linear dune (Tsoar and Moller, 1986) thatmarks the eastern boundary of this portion of the dunefield. Unlike the vegetated linear dunes along the marginsof the dune field to the north, the origin of this feature isnot obvious. The best interpretation we can provide for

this example is that it is the result of complex airflow alongthe boundary between the dunefield and the vegetatedvalley floor.

From this vantage point one can look to the south andsee the gap between the Moquith Mountains (east, left)and Moccasin Mountain (west, right) through whichsoutherly winds may be accelerated by the Venturi effect.One can also look to the southwest across much of theMoccasin Terrace, the probable source area for much of thesand in the dunes. To the north-northeast is the narrowgap between Esplin Point and the west-facing escarpmentof the Sevier fault, which may facilitate acceleration of thewind and the transport of sand over the escarpment toform the large parabolic dunes at the northern end of thedune field (Qelp). Directly east of the overview are the ac-tive core of the dune field (Qetd) and the prominentbedrock escarpment of the Sevier fault (figure 2).Nebkhas

A short nature trail begins beyond the boardwalkoverview and traverses into the dune field. Interpretativesigns along the trail focus on the relationships betweenvegetation and certain dune forms. Here one can see ex-cellent examples of nebkhas (also called shrub-coppicedunes), mounds or hummocks of sand trapped by clumpsof vegetation (figure 15). Although present in each geo-morphic unit of the dune field, nebkhas are particularyabundant here along the soutwestern margin of the activecore of transverse dunes (Qetd).

On flat surfaces wind velocities may decrease as muchas 50 percent on the downwind side of a bush or shrub(Francis, 1994). Both the decrease in velocity and the phys-ical obstruction of the stems or branches cause depositionof wind-blown particles within the plant, including sand,silt, organic debris, and nutrients attached to dust nuclei.Such accumulations may locally improve the soil fertility.

392

Figure 15. View to the west of a nebkha (shrub-coppice dune), anchoredby Wyethia scabra (known commonly as mule ears), within the activecore of transverse dunes (Qetd). Young field assistant is approximate-ly 4.6 feet (1.5 m) tall.

A nebkha will develop where the plants grow fast enoughto remain on top of the accumulating mound. Nebkhasmay eventually reach 3 to 15 feet (1–3 m) in height and 6to 30 feet (2–5 m) in diameter. Typically they are oval or el-liptical in plan form. The nebkhas in the Coral Pink SandDunes are generally less than 6 feet (2 m) in height. Here,several different plant species at are adapted to the stressof progressive burial by wind-blown sand, but Wyethiascabra (mule ears) is by far the most common. Wyethiascabra nebkhas are particularly abundant along the mar-gins of the dune field. In some areas the relief of thenebkha has been increased by deflation or eolian erosionof the ground adjacent to the nebkha, locally exposing thematted and twisted stems and roots of the mound-formingplant for a considerable depth below the top of the dune(Castle, 1954).Star Dune

The large dune directly east of the boardwalkoverview is an excellent example of a star dune (figures 2and 8). Star dunes commonly have a high central peakand three or four curving arms, with multiple slip faces,that radiate outward from the central mass. Star dunesform where winds from more than two directions havecontributed to dune growth. They tend to grow vertically,rather than migrating horizontally, and contain a greatervolume of sand than any other dune type (Lancaster,1995). McKee (1982) noted that most of the star dunes ofthe Namib Desert, southwestern Africa, developed fromlinear dunes, although some are associated with trans-

verse dunes, as is the case in the Coral Pink Sand Dunes.Star dunes typically achieve heights of 45 to 450 feet

(14-137 m), though some modern star dunes in the NamibDesert are reported to rise as much as 900 feet (274 m)above their interdune base (McKee, 1982). The large stardune at CPSDSP rises approximately 100 feet (30 m) aboveits base and has three sinuous arms and several smallersecondary arms.

We were surprised to learn, through the study of aeri-al photographs, that the star dune did not exist in 1960, butis clearly evident on photographs from 1997 (Wilkins andothers, 2003). The photographs suggest that several trans-verse dunes merged, sometime between 1960 and 1997, toform the lone star dune. We have collected data from sev-eral ground-penetrating radar (GPR) transects in the areaof the star dune since 2000 (Wilkins and others, 2002;2003). Data quality is excellent with strong reflectors up to60 feet (20 m) below the surface being recorded. The pres-ence of unidirectional, high-angle cross-bedding in a GPRtransect that crosses an arm of the star dune (figure 16)supports the transverse-dune parentage of this feature.Most star dunes are thought to be composed of low- tomoderate-angle ripple-laminated sand (Nielson and Ko-curek, 1987). Thus the internal structure and recent histo-ry of this dune suggests a distinctive mode of star duneformation.Barchan (Current Crescent)

Also visible from the boardwalk overview is a largebarchan-like dune, approximately 0.2 miles (0.3 km)


393

10 90807020 30 40 50 60 100

5

10

30

20

25

15

Dep

th(m

)Transect Distance (m)

OHV disturbance?

Dune Trough

Dune-Bedrock Interface

NE dipping foresets

Truncationsurfaces

10 20 30 40tansecrT

50 60e (m)t Distanc

8070 90 100

5

10

15

(m)

20

25

epth

(D

30

Figure 16. 100-m GPR transect from the base to crest of the star dune, oriented SW-NE. The dune-bedrock contact was confirmed with a hand auger.The internal radar stratigraphy shows high-angle slip faces and truncation surfaces that record multiple generations of migrating dunes. The re-flectors below the dune-bedrock contact are suggestive of southwest dipping cross-beds within the Navajo Sandstone.


south-southwest of the star dune. Barchans are character-ized by a distinctive crescentic shape in plan view, a steepconcave slipface, and “horns” extending downwind fromthe central mass of the dune (figure 8). Barchans are thestable dune type in areas where the directional variabilityof the wind is less than 15˚. Isolated barchans generallyoccur in areas with a limited sand supply; as the sand sup-ply increases barchans coalesce to form barchanoid ridges(Lancaster, 1995). However, it is puzzling that this dunetype so indicative of a unidirectional wind regime is insuch close proximity to a star dune, indicative of multidi-rectional winds. A mitigating factor may be the fact thatthe isolated barchan at CPSDSP is not a freely migratingdune, but instead has formed largely upwind of a smallknob of Navajo Sandstone, which can be observed in thedune’s slip face and GPR transects. We believe this duneis not a true barchan, but instead is best termed a currentcrescent. Current crescents are horseshoe-shaped dunesthat accumulate where winds are deflected around andover topographic obstacles such as boulders, escarpments,or hills (Pye and Tsoar, 1990). Sand is deposited both infront of the obstacle and on either side in two taperingwings that resemble the horns of a true barchan. The 1960and 1997 aerial photography support this interpretation,as the current crescent formed when a migrating trans-verse dune impinged onto and wrapped around thebedrock knob. We speculate that the formation of the cur-rent crescent may have modified the local wind regime,concentrating flow immediately downwind and leading tothe formation of the isolated star dune (Wilkins and oth-ers, 2003).

Tiger Beetle Conservation Area(“Cattle Guard Turnout”)

Approximately 1.6 miles (2.6 km) north of the rangerstation, on the east side of the road, is a turnout and smallparking area locally referred to as the “cattle guard” area(figure 8). This site provides easy access to the active coreof transverse dunes (Qetd) and the area of small marginalparabolic dunes (Qemp). From this area one can also seethe barchanoid ridges (figure 10) impinge upon thebedrock escarpment, transforming into echo dunes andclimbing dunes. This is also an excellent location to ob-serve the ecological differences between the largely un-vegetated dune ridges and the vegetated interduneswales. Some of the interdune swales are incised, expos-ing their shallow stratigraphy. This stratigraphy, which in-cludes mudcracks and darker, organic-rich (?) layers, indi-cates cycles of water ponding and desiccation and periodsof weak soil development within these interdune areas.

This general location is also the habitat of the CoralPink Sand Dunes (CPSD) tiger beetle (Cicindela limbata al-bissima), which is only known to inhabit this dune field.The biology and ecology of the CPSD tiger beetle has beenlargely revealed through the research of entomologistsBarry Knisley (Randolph-Macon College, Ashland, Vir-

ginia) and James Hill. The following account is summa-rized from their work, primarily Knisley and Hill (2001).

Tiger beetles are predatory insects that prefer open,sparsely vegetated habitats, such as sand dunes. Manyspecies of the genus Cicindela inhabit dune fields of thewestern United States and several are endemic to specificdune fields, including the CPSD tiger beetle. Surprisingly,the CPSD tiger beetle appears to regularly inhabit only asmall portion (approximately 200 acres or 81 ha) of thetotal dune field, making its geographical range one of thesmallest of any organism known to science.

In 1994 the Southern Utah Wilderness Alliance peti-tioned the U.S. Fish and Wildlife Service (USFWS) to listthe CPSD tiger beetle as an endangered species. Current-ly, the CPSD tiger beetle is recognized as a candidate forlisting as an endangered or threatened species. In 1998the USFWS, BLM, Utah Department of Natural Resources,and Kane County signed an agreement that established a200-acre (81 ha) conservation area within CPSD State Park.Off-highway vehicles are not permitted in the westernportion of the conservation area where most of the tigerbeetles are found. The eastern portion of the conservationarea serves as a travel corridor between the southern andnorthern portions of the dune field. An additional 370 acre(121 ha) conservation area has been established on BLM-administered land at the extreme northern end of the dunefield where a small number of dispersing tiger beetleshave been observed. The CPSD tiger beetle was once con-sidered a subspecies of the Sandy Tiger Beetle (Cicindelalimbata). DNA studies published in 2000 (Morgan and oth-ers, 2000) show it to be a distinct and separate species andhighlight the importance of the conservation measurestaken within the park.

Although the range of the CPSD Tiger beetle is quitesmall, as is its total population, there is no evidence of aprogressive decline in CPSD tiger beetles during the timethat Knisley and Hill (2001) have monitored their popula-tion (1992-present). They believe that the implementationof the conservation plan in 1998, which protects the bee-tles’ primary habitat, will promote the long-term existenceof this species within the Coral Pink Sand Dunes.

A puzzling question, for which there is presently nodefinitive answer, is why is the distribution of the CPSDtiger beetle so restricted within the overall dune field?Knisley and Hill (2001) point out that there is significantvariation in the geomorphology, vegetation, prey insects,and OHV activity throughout the dune field. They notethat the percent vegetation cover in the interdune swalesof the primary habitat/conservation area is significantlygreater than in the swales of other areas of the dune field.Probably related, the numbers and types of prey speciesare also significantly greater in the conservation area com-pared to other areas of the dune field. Are these differencesrelated to purely natural factors, such as the geomorphictransition from more active, less vegetated barchanoiddunes in the southern portion of the dune field to the lessactive, more vegetated climbing and parabolic dunes to

394

the north? Or are the habitat differences related to OHVactivity? Knisley and Hill (2001) suggest that high OHVuse contributes to the highly dynamic nature of the south-ern portion of the dune field, decreasing the cover of veg-etation in interdune swales. These questions point out thedistinctly interdisciplinary (biology and geology in thiscase) nature of many of the resource and land-use issues orconcerns now faced by human societies.

Final CommentsThis paper has attempted to summarize our current

understanding of the particular geomorphic systemknown as the Coral Pink Sand Dunes. We know of noother dune field in Utah, with the possible exception of theLynndyl Dunes (BLM Little Sahara Recreation Area) (Sack,1987), that contains the diversity of eolian landforms pres-ent here. Hopefully we have shown that much can belearned at Coral Pink Sand Dunes State Park, both infor-mally and individually by observant visitor-naturalistsand formally and cooperatively by professional scientists.Coral Pink Sand Dunes will continue to provide a naturallaboratory in which to observe and discern the complex in-teractions among atmospheric, geomorphic, and ecologi-cal processes. The authors plan to continue research herein the hopes of developing a more complete chronology ofdune formation and evolution during the Holocene. Ofcourse one doesn’t have to be a scientist to appreciate theimportance and beauty of this landscape.

To stand in the midst of a sea of great dunes is to havethe sensation of being utterly alone, and vulnerable, in aworld of scorching elements and soul-shaking beauty.

— Jan DeBlieu, 1999

ACKNOWLEDGMENTSWe would like to thank Margie Chan (University of

Utah), Greg Schlenker (Kleinfelder), and Sarah George(Utah Museum of Natural History) for their careful andthoughtful review of our manuscript. In addition, thecomments and suggestions of editors Doug Sprinkel andPaul Anderson greatly improved the final paper. BarryKnisley (Randolph-Macon College) generously shared hisknowledge of tiger beetle ecology and dune environmentsand materials related to the park. Stephanie Rodgers do-nated her time and expertise in preparing some of the fig-ures. Thanks are also due to the staff of the Utah StateFilm Commission for researching the use of the park areaas a film location. Stuart Ford provided assistance andcomic relief during several of the early field sessions.Weber State undergraduates Spencer Riley and ScottWheeler provided field assistance during the initial GPRsurveys. We thank the following entities for their fundingof our post-2000 studies in the park: Weber State Universi-ty; Boise State University; Gladys W. Cole Award (Geolog-

ical Society of America); NASA; and The Royal SocietyLastly, we thank the park staff, especially Rob Quist (for-mer Park Manager) and Carl Davis (retired ranger), for fa-cilitating our field work and, more importantly, for theirprofessional management of this important site. Their jobis not easy as the constituencies who use and value thispark are diverse. We all owe them our gratitude.

REFERENCESAnderson, R.E., and Christenson, G.E., 1989, Quaternary

faults, folds, and selected volcanic features in theCedar City 1°x 2° quadrangle, Utah: Utah Geologicaland Mineral Survey Miscellaneous Publication 89-6,29 p.

Baars, D.L., 1983, The Colorado Plateau - a geologic histo-ry: Albuquerque, University of New Mexico Press, 279p.

Bagnold, R.A., 1942, The physics of blown sand and desertdunes: New York, William Morrow & Company, 265 p.

Barnes, F.A., 1993, Geology of the Moab area: Moab,Canyon Country Publications, 264 p.

Beghoul, N., and Barazangi, 1989, Mapping high Pn veloc-ity beneath the Colorado Plateau constrains upliftmodels: Journal of Geophysical Research, v. 94, p.7083-7104.

Beitler, B., Chan, M.A., and Parry, W.T., 2003, Bleaching ofJurassic Navajo Sandstone on Colorado PlateauLaramide highs: evidence of exhumed hydrocarbonsupergiants?: Geology, v. 31, p. 1041-1044.

Beitler, B., Parry, W.T., and Chan, M.A., 2005, Fingerprintsof fluid flow: chemical diagenetic history of the Juras-sic Navajo Sandstone, southern Utah: Journal of Sedi-mentary Research, v. 75, p. 545-559.

Benson, L., Peterson, K., and Stein, J., 2007, Anasazi (pre-Columbian native-American) migrations during themiddle-12th and late-13th centuries—were theydrought induced? Climatic Change, v. 83, p. 187-213.

Betancourt, J.L., 1990, Late Quaternary biogeography ofthe Colorado Plateau, in Betancourt, J.L., Van Deven-der, T.R., and Martin, P.S., editors, Packrat middens -the last 40,000 years of biotic change: Tucson, The Uni-versity of Arizona Press, p. 259-292.

Blakey, R.C., 2008, Pennsylvanian-Jurassic sedimentarybasins of the Colorado Plateau and Southern RockyMountains, in Miall, A.D., editor, Sedimentary basinsof the United States and Canada: Amsterdam, Elsevi-er, p. 245-296.

Blakey, R., and Ranney, W., 2008, Ancient landscapes of theColorado Plateau: Grand Canyon Association, 176 p.

Bradley, R.S., and Jones, P.D., 1993, ‘Little Ice Age’ summertemperature variations: their nature and relevance torecent global warming trends: The Holocene, v. 3, p.367-376.

Castle, E.S., 1954, The succession of vegetation on a south-ern Utah sand dune: Provo, Brigham Young Universi-ty, M.S. thesis, 98 p.


395

Chan, M.A., and Parry, W.T., 2002, Rainbow of rocks, mys-teries of sandstone colors and concretions in ColoradoPlateau canyon country: Salt Lake City, Utah Geologi-cal Survey, 17 p.

Chan, M.A., Parry, W.T., and Bowman, J.R., 2000, Diage-netic hematits and manganese oxides and fault-relatedfluid flow in Jurassic sandstones, southeastern Utah:American Association of petroleum Geologists Bul-letin, v. 84, p. 1281-1310.

Cook, E.R., Woodhouse, C.A., Eakin, C.M., Meko, D.M.,and Stahle, D.W., 2004, Long-term aridity changes inthe western United States: Science, v. 306, p. 1015-1018.

Cooke, R., Warren, A., and Goudie, A., 1993, Desert geo-morphology: London, CRC Press, 526 p.

Davis, G.H., 1999, Structural geology of the ColoradoPlateau region of southern Utah with special emphasison deformation bands: Geological Society of AmericaSpecial Paper 342, 157 p.

Davis, G.H., and Bump, A.P., 2009, Structural geologicevolution of the Colorado Plateau, in Kay, S.M.,Ramos, V.A., and Dickinson, W.R., editors, Backboneof the Americas: shallow subduction, plateau uplift,and ridge and terrane collision: Geological Society ofAmerica Memoir 204, p. 99-124.

Dean, J.S., and Bowdwn, D.O., 1971, Navajo Mountaintree-ring data: Boulder, CO, International Tree-RingData Bank, IGBP PAGES/World Data Center for Pale-oclimatology, NOAA/NCDC Paleoclimatology Pro-gram.

DeBlieu, J., 1999, Wind: how the flow of air has shaped life,myth, and the land: Boston, Mariner Books, 294 p.

Dickinson, W.R., 1979, Cenozoic plate tectonic setting ofthe cordilleran region in the United States, in Armen-trout, J.M., Cole, M.R., and Terbest, H., editors, Ceno-zoic paleogeography of the western United States: So-ciety of Economic Mineralogists and Paleontologists,Pacific Coast Paleogeography Symposium 3, p. 1-13.

Dickinson, W.R., and Gehrels, G.E., 2003, U-Pb ages of de-trital zircons from Permian and Jurassic eolian sand-stones of the Colorado Plateau, USA: paleogeographicimplications: Sedimentary Geology, v. 163, p. 29-66.

Doelling, H.H., Davis F.D., and Brandt, C.J., 1989, The ge-ology of Kane County, Utah: Utah Geological andMineral Survey Bulletin 124, 192 p.

Ely, LE., 1997, Response of extreme floods in the south-western United States to climatic variations in the lateHolocene: Geomorphology, v. 19, p. 175-201.

Forman, S.L., Spaeth, M., Marin, L., Pierson, J., Gomez, J.,Bunch, F., and Valdez, A., 2006, Episodic late Holocenedune movements on the sand-sheet area, Great SandDunes National Park and Preserve, San Luis Valley,Colorado, USA: Quaternary Research, v. 66, p. 97-108.

Fossen, H., Schultz, R.A., Shipton, Z.K., and Mair, K., 2007,Deformation bands in sandstone: a review: Journal ofthe Geological Society, London:, v. 164, p. 1-15.

Francis, C.F., 1994, Plants on desert hillslopes, in Abra-hams, A.D., and Parsons, A.J., editors, Geomorpholo-

gy of desert environments: London, Chapman andHall, p. 243-254.

Gregory, H.E., 1950, Geology and geography of the ZionPark region Utah and Arizona: U.S. Geological SurveyProfessional Paper 220, 200 p.

Harlow, W.M., and Harrar, E.S., 1968, Textbook of den-drology - covering the important forest trees of theUnited States and Canada (5th edition): New York,McGraw-Hill Book Company, 512 p.

Harris, A.G., and Tuttle, E., 1990, Geology of nationalparks (4th edition): Dubuque, Kendall/Hunt Publish-ing, 652 p.

Hereford, R., 2002, Valley-fill alluviation during the LittleIce Age (ca A.D. 1400–1880), Paria River basin andsouthern Colorado Plateau, United States: GeologicalSociety of America Bulletin, v. 114, p. 1550-1563.

Hintze, L.F., 1988, Geologic History of Utah: BrighamYoung University Geology Studies, Special Publica-tion 7, 202 p.

Hunter, R.E., 1976, Comparison of eolian and subaqueoussand-flow cross strata [abs.]: American Association ofPetroleum Geologists Bulletin, v. 60, p. 683-684.

Hunter, R.E., 1977, Basic types of stratification in small eo-lian dunes: Sedimentology, v. 24, p. 361-387.

Imlay, R.W., 1980, Jurassic paleogeography of the conter-minous United States in its continental setting: U.S.Geological Survey Professional Paper 1062, 134 p.

Knisley, C.B., and Hill, J.M., 2001, Biology and conserva-tion of the Coral Pink Sand Dunes tiger beetle, Cicin-dela limbata albissima: Western North American Natu-ralist, v. 60, no. 4, p. 381-394.

Kocurek, G., and Landcaster, N., 1999, Aeolian system sed-iment state: theory and Mojave Desert Kelso dunefield example: Sedimentology, v. 46, p. 505-515.

Lancaster, N., 1995, Geomorphology of desert dunes: Lon-don, Routledge, 290 p.

Lucas, S.G., andMilner, A.R.C., 2006, Cochostraca from theLower Jurassic Whitmore Point Member of the Moe-nave Formation, Johnson farm, southwestern Utah, inHarris, J.D., Lucas, S.G., Spielmann, J.A., Lockley,M.G., Milner, A.R.C., and Kirkland, J.I., editors, TheTriassic-Jurassic terrestrial transition: New MexicoMuseum of Natural History and Science Bulletin 37, p.421-423.

Lucas, S.G., and Tanner, L.H., 2006, The Springdale Sand-stone of the Kayenta Formation, Lower Jurassic ofUtah-Arizona, in Harris, J.D., Lucas, S.G., Spielmann,J.A., Lockley, M.G., Milner, A.R.C., and Kirkland, J.I.,editors, The Triassic-Jurassic terrestrial transition:New Mexico Museum of Natural History and ScienceBulletin 37, p. 71-76.

Lund, W.R., Knudsen, T.R., and Vice, G.S., 2008, Paleoseis-mic reconnaissance of the Sevier fault, Kane andGarfield Counties, Utah: Utah Geological Survey Spe-cial Publication 122, 31 p.

Maltin, L., editor, 1997, Movie and video guide (1998 edi-tion): New York, Signet Reference, 1620 p.

396


Matthews, N.A., Christensen, T., Titus, A.L., Noble, T.A.,and Breithaupt, B.A., 2010, Using GIS to track di-nosaur footprints: Take It Outside at the Jurassic Moc-casin Mountain Tracksite, Utah: Geological Society ofAmerica Abstracts with Programs, v. 42 (3), p. 5.

Marzolf, J.E., 1988, Controls on late Paleozoic and earlyMesozoic eolian deposition of the western UnitedStates: Sedimentary Geology, v. 56, p. 167-191.

McKee, E.D., 1979, Introduction to a study of global sandseas, in McKee, E.D., editor, A study of global sandseas: U.S. Geological Survey Professional Paper 1052,p. 1-19.

McKee, E.D., 1982, Sedimentary structures in dunes of theNamib Desert, South West Africa: Geological Societyof America Special Paper 188, 64 p.

Mensing, S., Smith, J., Norman, K.B., and Allan, M., 2008,Extended drought in the Great Basin of western NorthAmerica in the last two millennia reconstructed frompollen records: Quaternary International, v. 188, p. 79-89.

Milner, A.R.C., Lockley, M.G., and Johnson, S.B., 2006, Thestory of the St. George dinosaur discovery site at John-son Farm: an important new Lower Jurassic dinosaurtracksite from the Moenave Formation of southwest-ern Utah, in Harris, J.D., Lucas, S.G., Spielmann, J.A.,Lockley, M.G., Milner, A.R.C., and Kirkland, J.I., 2006,The Triassic-Jurassic terrestrial transition: NewMexicoMuseum of Natural History and Science Bulletin 37, p.329-345.

Morgan, M., Knisley, C.B., and Vogler, A.P., 2000, New tax-onomic status of the endangered tiger beetle Cicindelalimbata albissima (Coleoptera: Cicindelidae): evidencefrom mtDNA: Annals of the Entomological Society ofAmerica, v. 93, p. 1108-1115.

Morris, T.H., and Stubben, M.A., 1994, Geologic contrastsof the Great Basin and Colorado Plateau, in Harper,K.T., St. Clair, L.L., Thorne, K.H., and Hess, W.M., ed-itors, Natural history of the Colorado Plateau andGreat Basin: Niwot, University of Colorado Press, p.98-25.

Murphy, D.R., 1981, Climatic zones, in Atlas of Utah:Provo, Brigham Young University Press, p. 55.

Nielson, J., and Kocurek, G., 1987, Surface processes, de-posits, and development of star dunes: Dumont dunefield, California: Geological Society of America Bul-letin, v. 99, p. 177-186.

Oviat, C.G., 1988, Late Pleistocene and Holocene lake fluc-tuations in the Sevier Lake Basin, Utah, USA: Journalof Paleolimnology, v. 1, p. 9-21.

Parrish, J.T., and Falcon-Lang, H.J., 2007, Coniferous treesassociated with interdune deposits in the JurassicNavajo Dandstone formation, Utah, USA: Paleontol-ogy, v. 50, p. 829-843.

Parrish, J.T., and Peterson, F., 1988, Wind directions pre-dicted from global circulation models and wind direc-tions determined from eolian sandstones of the west-ern United States—a comparison: Sedimentary Geolo-

gy, v. 56, p. 261-282.Patton, P,C., Biggar, N., Condit, C.D., Gilliam, M.L., Love,D.W., Machette, M.N., Mayer, L, Morrison, R.B., andRosholt, J.N., 1991, Quaternary geology of the Col-orado Plateau, in Morrison, R.B., editor, Quaternarynonglacial geology—Coterminous U.S.: Boulder, Col-orado, Geological Society of America, The Geology ofNorth America, v. K-2, p. 373-406.

Peterson, Fred, 1988, Pennsylvanian to Jurassic eoliantransportation systems in the western United States:Sedimentary Geology, v. 56, p. 207-260.

Peterson, Fred, and Turner-Peterson, C.E., 1989, Geologyof the Colorado Plateau, inHanshaw, P.M., editor, Sed-imentation ands basin analysis in siliciclastic rock se-quences, Volume 1, Field trips for the 28th Interna-tional Geological Congress: Washington, D.C., Ameri-can Geophysical Union, 65 p.

Pipiringos, G.N., and O’Sullivan, R.B., 1978, Principle un-conformities in Triassic and Jurassic rocks, western in-terior United States - a preliminary survey: U.S. Geo-logical Survey Professional Paper 1035-A, 29 p.

Pope, D., and Brough, C., 1996, Utah’s weather and cli-mate: Salt Lake City, Publishers Press, 245 p.

Pye, K., and Tsoar, H., 1990, Aeolian sand and sand dunes:London, Unwin Hyman, 396 p.

Ritter, D.F., Kochel, R.C., and Miller, J.R., 1995, Process ge-omorphology (3rd edition): Boston, WCB/McGraw-Hill, 546 p.

Sack, D., 1987, Geomorphology of the Lynndyl Dunes,west-central Utah, in Kopp, R.S., and Cohenour, R.E.,editors, Cenozoic geology of western Utah - sites forprecious metal and hydrocarbon accumulations: UtahGeological Association Publication 16, p. 291-299.

Sargent, K.A., and Philpott, B.C., 1987, Geologic map ofthe Kanab quadrangle, Kane County, Utah, and Co-conino County, Arizona: U.S. Geological Survey Geo-logic Quadrangle Map GQ-1603, scale 1:62,500.

Shafer, D.S., 1989, Holocene climatic patterns inferredfrom pollen and macrofossil records of Holocene lakesof the northern Colorado Plateau, Utah and Arizona[abs.]: Geological Society of America Abstracts withPrograms, v. 21, no. 5, p. 142.

Sprinkel, D.A., Kowallis, B.J., Doelling, H.H., and Kuehne,P.A., 2009, The Middle Jurassic Temple Cap Forma-tion, southern Utah-radiometric age, palynology, andcorrelation with the Gypsum Spring member of theTwin Creek Limestone and the Harris Wash Memberof the Page Sandstone.: Geological Society of AmericaAbstracts with Programs, v. 41, no. 7, p. 690.

Stanhle, D.W., Fye, F.K., Cook, E.R., and Griffin, R.D., 2007,Tree-ring reconstructed megadroughts over NorthAmerica since A.D. 1300: Climatic Change, v. 83, p.133-149.

Stanley, K.O., Jordan, W.M., and Dott, R.H., 1971, New hy-pothesis of Early Jurassic paleogeography and sedi-ment dispersal for western United States: AmericanAssociation of Petroleum Geologists Bulletin, v. 55, p.


397

10-19.Stewart, J.H., 1971, Basin and Range structure - a system ofhorsts and grabens produced by deep seated exten-sion: Geological Society of America Bulletin, v. 82, p.1019-1050.

Stokes, M.A., and Smiley, T.L., 1996, An introduction totree-ring dating: Tucson, University of Arizona Press.73 p.

Stokes, S., 1994, Latest Holocene mobilisation (sic) eventsin the southwestern United States [abs.]: GeologicalSociety of America Abstracts with Programs, v. 26, no.7, p. A-89.

Stokes, S., and Breed, C.S., 1993, A chronostratigraphic re-evaluation of the Tusayan Dunes, Moenkopi Plateauand Ward Terrace, northeastern Arizona, in Pye, K.,editor, The dynamics and environmental context ofAeolian sedimentary systems: Geological Society(London) Special Publication 72, p. 75-90.

Stokes, W.L., 1986, Geology of Utah: Utah Museum of Nat-ural History Occasional Paper 6, 309 p.

Stuiver, M., and Reimer, P.J., 1993, Extended 14C databaseand revised CALIB radiocarbon calibration program:Radiocarbon, v. 35, p. 215-230.

Summerfield, M.A., 1991, Global geomorphology: NewYork, Longman Scientific & Technical, 537 p.

Tilden, F., 1970, The national parks (revised and enlargededition): New York, Alfred A. Knoff, 608 p.

Tsoar, H., and Moller, J.T., 1986, The role of vegetation inthe formation of linear dunes, in Nickling, W.G., edi-tor, Aeolian geomorphology (Proceedings of the 17thAnnual Binghamton Geomorphology Symposium):Boston, Allen & Unwin, p. 75-95.

Utah State Parks and Recreation, no date, Coral Pink SandDunes State Park - geology of the park and surround-ing area: Utah State Parks and Recreation brochure, 4p.

Wilkens, N.D., and Farmer, J., 2005, Paleoecology of theNavajo Sandstone interdune deposits: Canyon Legacy(Dan O’Laurie Museum of Moab), v. 54, p. 39-42.

Wilkins, D.E., and Ford, R.L., 2007, Nearest neighbormethods applied to dune field organization: the CoralPink Sand Dunes, Kane County, Utah, USA: Geomor-phology, v. 83, p. 48-57.

Wilkins, D.E., Ford, R.L., and Clement, W.P., 2002, Internalstructure of a star dune imaged using ground-pene-trating radar (abstract): EOS, Transactions of theAmerican Geophysical Union, v. 83, p. F587.

Wilkins, D.E., Ford, R.L., and Clement, W.P., 2003, GPRand GIS use in documenting a distinctive mode of stardune formation: XVI INQUACongress Programs withAbstracts: Reno, NV, The Desert Research Institute, p.85.

Wilkins, D., Ford, R.L., Clement, W., and Nicoll, K., 2005,Little Ice Age behavior of the Coral Pink Sand Dunes,Kane County, Utah: Geological Society of America Ab-stracts with Programs, v. 37, p. 426.

Wilkins, D., Ford, R.L., Clement, W., and Nicoll, K., 2007,

Multiproxy evidence for late Holocene climaticchange, Coral Pink Sand Dunes, Utah (abstract): Asso-ciation of American Geographers 2007 Annual Meet-ing (San Francisco, CA) Online Program.

Withers, K., and Mead, J.L., 1993, Late Quaternary vegeta-tion and climate in the Escalante River basin on thecentral Colorado Plateau: Great Basin Naturalist, v. 53,p. 145-161.

398


Documents

Geology and Geomorphology of Coral Pink Sand Dunes State ...Geology and Geomorphology of Coral Pink Sand Dunes State Park, Utah RichardL.Ford1,ShariL.Gillman1,DavidE.Wilkins2,WilliamP.Clement3,