4.1. Photographs of clastic dikes. (A) Multiple crosscutting clastic dikes on Highway 240 near Pinnacles Overlook. (B) Clastic dike along the Old Northeast Road showing sheetlike morphology. (C), (D), (E) Close-up of clastic dikes showing crenulated texture. (F) Mud cracks from strata near Door and Window Overlook. Note size difference and fracture pattern compared to clastic dikes. Photos by the authors.
THE BADLANDS THAT WE SEE TODAY ARE THE RESULT OF both depositional and postdepositional processes. Our understanding of these processes and their role in the formation of the Badlands is only possible as a result of the significant erosion that characterizes this region. We discuss postdepositional features first, with the understanding that it is erosion that has allowed us to study these features.
The rock strata that comprise the Badlands were once loose sand, silt, and clay, with the occasional wind-deposited tephra and freshwater limestone. Upon burial, these sediments were turned to rock by compaction and cementation during the process of lithification. Compaction is due to the enormous overlying weight of additional sediments that are brought into the basin. As a result, the thicker the accumulation of sediments, the greater the reduction of porosity (void space) between the grains. Cementation is the process by which individual grains of sediment are bound together by the precipitation of secondary minerals, commonly calcite, quartz, or iron oxide, out of groundwaters that have moved through the sediment. Geologists refer to lithification as just one form of diagenesis, the sum of physical and chemical changes that can be induced in sediment upon burial.
Additional chemical diagenetic changes include the wholesale change of certain minerals into new phases, such as the dewatering of iron oxyhydroxides to hematite and the conversion of smectite (swelling) clays to illite clay by the addition of potassium. Along with cementation, entirely new minerals can grow in available pore spaces. In the White River Group, the common secondary minerals include calcite, gypsum, quartz, chalcedony, barite, various uranium-bearing minerals, and zeolites (Retallack, 1983b; Terry and Evans, 1994). Many of these diagenetic changes require microscopes to detect small-scale mineralogical changes and specialized instruments to measure the chemical makeup of rock samples. Diagenetic changes that can be seen with the naked eye include the overall reddish appearance (due to the recrystallization of various yellowish-brown iron oxyhydroxides into hematite) that are seen in many of the stripes that cut across the Badlands, and secondary mineralization within fractures caused by tectonic forces that compressed and stretched rock units of the Badlands.
Some of the more prominent postdepositional features of the Badlands are the numerous nearly vertical sheets of resistant material that crosscut the horizontal layering of the Badlands (Fig. 4.1). These features, referred to as either clastic dikes or chalcedony veins, depending on the type of material that is present, vary in size and in regional and stratigraphic distribution, and are formed by different processes.
Clastic dikes are composed of small particles of sediment that have been cemented to form ridgelike structures that cut across the Badlands. Their resistance to erosion is greater than that of typical Badlands rocks, which results in “fins” of clastic dike material that stand out in erosional relief and commonly help to support less resistant Badlands materials (Fig. 4.1). The appearance of these dikes can vary from light beige colors that are similar to typical Badlands strata to those that are stained with a greenish color. When viewed from the edge, these features range from paper-thin bodies to tens of centimeters thick that cut across tens of meters of solid rock. Some are relatively smooth along their surfaces, whereas others appear to be crenulated, similar to a wavy potato chip (Fig. 4.1). These dikes sound a glassy ring when pieces are struck together. When viewed along the surface of the ground or from the air, these sheets of clastic material are sometimes straight lines, but in other instances they form curvilinear features that seem to randomly run across the surface for tens to hundreds of meters. Models describing the formation of these dikes have included large-scale desiccation cracks that were filled with sediments (Retallack, 1983b), fractures that were formed (and later filled) in response to tectonic and structural strain (Smith, 1952), and injection of liquefied sediments from deeper in the Badlands strata (Whelan, Hamre, and Hardy 1996).
Modern-day desiccation cracks (mud cracks) form by the drying and shrinkage of clays and can be found in such settings as muddy stream banks or lakes that have dried. The resulting cracks decrease in width downward and tend to join other cracks to form polygonal patterns ranging in scale from several centimeters across to tens of meters. These cracks can be infilled with new sediment and preserved in the geologic record (Fig. 4.1). With respect to the clastic dikes in the Badlands, some of these features pinch out and narrow in a downward direction. The most common examples of this are found in association with green-colored clastic dikes.
Although some clastic dikes can be seen to narrow downward (which would support a desiccation model), when viewed from the air, these dikes do not create large-scale polygonal networks similar to modern mud cracks. In addition, shrinkage via desiccation only occurs in moist clay-size sediments, whereas these dikes crosscut tens of meters of solid rock of varying grain size. It is possible that these cracks were at one time open to an ancient surface sometime in the geologic past, but no such relationship has yet been documented. If these fractures were indeed open to an ancient land surface, such features as fossil bones from animals that fell into these fissures or evidence of younger sediments filling dikes that cut across older strata should be present. To date no such evidence has been documented, although no such study has yet been carried out in the Badlands. On one occasion, one of the authors (Terry) observed a fossil leaf impression in clastic dike material from strata equivalent to the Badlands at the Flagstaff Rim locality southwest of Casper, Wyoming. On another occasion the same author observed a clastic dike composed of pea-size gravel cutting through the Chadron Formation in the Big Corral Draw area in the South Unit of Badlands National Park (Terry and Spence, 1997). Both of these observations would suggest that at least some of these dikes were indeed open to the surface in the past, but they do not support the model of large-scale desiccation cracks.
Fractures in rock are commonly caused by geological forces of stress and strain acting over broad regions. In order to form large, open fractures, rock strata must be stretched. This induces tensional forces that can eventually fracture the rock to create open fissures. These fissures tend to be vertical and parallel to each other. Such forces were created by the uplift of the Black Hills, but this occurred before the majority of the Badlands were deposited. It is possible that some fracturing is related to the formation of the asymmetric basin that contains most of the Badlands strata in this region (Fig. 4.2). This basin is hinged on the northeast edge and drops downward to the south and west. This type of basin requires crustal thinning (tensional forces) and could be a consequence of aftereffects of the uplift of the Black Hills as the surrounding strata on the Great Plains could again relax after uplift.
If this model is correct, the majority of clastic dikes should show a roughly southeast-to-northwest orientation, parallel to the major structural axes of the Badlands. Although many dikes do follow this orientation, many more run southwest to northeast, suggesting that either a secondary stress field of extension was present or that fracturing was due to compressional forces, which creates sets of fractures that cut across each other. Tensional fracturing also does not explain the exquisite curvilinear nature of many of these fractures when viewed from the air.
Our final hypothesis to explain these features is injection of clastic materials from below. Under high pressures, sediments can become suspended in a fluid and injected into overlying strata. Such conditions are common during earthquakes, during which the intense shaking of the ground liquefies unconsolidated sediment as groundwater is forcefully ejected to the land surface (liquefaction). This process can result in localized sand volcanoes or small-scale dikes that are injected into surface soils. These dikes can be traced downward over several meters and can be seen to connect with laterally extensive beds of sandy material that were originally deposited by normal geologic processes, such as a former river.
In deeper geologic settings the fluidization of sediments and subsequent injection is more complicated. Certain minerals are more susceptible to breakdown after burial. In the case of the Badlands, the large volume of volcanic ash that was brought into this region by wind and rivers would have been easy to modify upon burial. This modification, referred to as desilicification and dewatering, creates a situation in which masses of sediment at depth become liquefied as the ash breaks down. The overlying layers of sediment exert significant downward pressure on these liquefied beds, which forces the liquid up and along any fracture that can be exploited.
Evidence of injection is prominent along the Old Northeast Road just north of Cedar Pass (Fig. P.1). These features were described by Whelan, Hamre, and Hardy (1996) and include upward-radiating fingers of dike material that branch off thicker dikes, and dikes that divide and rejoin. Upon closer examination, the dike material appears to be laminated. In some instances the dike fill appears to be cross-bedded in an upward direction. Both of these features are indicators of flow. Clastic dikes in this area also manifest a crenulated appearance, some of which may be the result of sculpting and scouring during fluid flow (Fig. 4.1). None of these features would be present if the dike fill was the result of passive infilling of a fracture open at the land surface. In addition, some of these dikes show multiple generations of dike formation, which appear as variations in the color of the dike material or grain size.
4.2. Regional geologic map (A) and cross section (B) showing the geographic and structural relationships between the deposits of the White River Group, the Black Hills, and the bounding faults that created this basin.
One other possible explanation is the conversion of opaline silica to more stable forms by recrystallization (Davies et al., 2006). This process produces excess water and a reduction in sediment volume, creating overpressurized conditions and eventual fluidization of sandy materials and upward injection into overlying strata. This phenomenon has been documented at shallow depths (less than 50 m) in biosiliceous-enriched marine sediments. Although biosiliceous deposits are not known within White River Group, the underlying Cretaceous Pierre Shale, which would have a greater abundance of biosiliceous materials, could possibly be responding to such a recrystallization mechanism. If this is indeed the case, clastic dikes should be more widespread throughout the region because the Pierre Shale underlies the entirety of South Dakota.
At the other end of the spectrum are the numerous veins of chalcedony that can be seen in certain parts of the park. The chalcedony veins are commonly various shades of blue, olive, yellow-gray, and occasionally red, and they can vary in width from knife-edge to several centimeters. Although the veins are most commonly composed of chalcedony, calcite and gypsum are sometimes present, as are pockets of typical Badlands mudstone (Fig. 4.3).
The chalcedony veins appear to be restricted both stratigraphically and regionally throughout the park. The most prominent zone of chalcedony veins is primarily restricted to the upper part of the Chadron Formation and ends just below, or slightly within, the overlying Scenic Member of the Brule Formation. They can be easily seen along Highway 44 headed toward Scenic, South Dakota, in the narrow neck of NPS property that crosses the road (Fig. P.1). The veins stand out in erosional relief as sharp, low-standing ridges of material approximately 5 to 10 cm above the bedrock. Over time, these sheets of material break off and cover the surrounding slopes, which give the Badlands a slightly blue tint. On a regional scale, the chalcedony veins appear to be concentrated toward the southwest part of the park, especially in and around Chamberlain Pass, Sheep Mountain Table, and the White River Visitor Center (Fig. P.1). Veins identical in mineralogy but with a much deeper blue color, referred to as blue agate, can be found in Toadstool Geologic Park north of Crawford, Nebraska.
The genesis of the chalcedony veins is enigmatic. Unlike the clastic dikes, which are primarily vertical features, chalcedony veins are both vertical and horizontal, although vertical orientations are most prominent. At scales of 10 m2 the chalcedony veins appear as rough polygonal patterns reminiscent of large desiccation cracks (Fig. 4.3). Upon closer examination, the veins show evidence of multiple episodes of mineral precipitation, sometimes as alternating colors of chalcedony and other times as alternating bands of gypsum and chalcedony. Some veins also show evidence of fracture and subsequent healing. In some instances, such as near the White River Visitor Center, chalcedony veins experienced ptygmatic folding (Fig. 4.3), suggesting compaction of strata surrounding the vein and subsequent deformation. On the very small scale, chalcedony occasionally manifests as three-dimensional boxworklike structures that pinch out over tens of centimeters in all directions. In other instances, isolated bodies of chalcedony can be found that display bladed forms, similar to desert roses (Fig. 4.3), which suggests replacement of an original calcite or barite precursor by silica. Amorphous, vesicular masses of chalcedony are sometimes found in association with bladed morphologies (Fig. 4.3).
The underlying cause behind the genesis of these veins is still a mystery. Any proposed hypothesis to explain the formation of these veins must be able to account for the production and availability of a large volume of free silica. The White River Group is enriched in volcanic ash, the hydrolysis of which produces smectite clays and free silica. The late Eocene Chadron Formation is abundant in smectite, which gives this unit the thick popcorn-textured weathered surface and rounded, haystacklike hills compared to the overlying Oligocene Brule Formation. The smectite enrichment in the Chadron Formation is thought to represent ancient weathering and soil formation under humid, forested conditions that gave way to drier conditions (and hence less hydrolysis) in the Oligocene Brule Formation (Retallack, 1983b; Terry, 1998, 2001). It is unlikely that a greater amount of hydrolytic weathering during the late Eocene is responsible for contributing silica for these veins because the veins crosscut lithified sediments of the Chadron Formation and extend up into the lowermost Brule Formation.
4.3. Photographs of chalcedony veins and other siliceous accumulations. (A) Top view of vein showing multiple generations of vein fill and stress-related shearing (white lines and arrows). (B) Polished section of chalcedony vein showing clay masses (cl) inside of chalcedony (ch). (C, D) Patch of chalcedony veins just south of Sheep Mountain Table. Note curvilinear nature and rough polygonal outline. (E) Vertical vein exposure with ptygmatic folding (P). (F) Boxwork structure of chalcedony ranging from knife-edge to centimeters in thickness. (G) Bladed chalcedony pseudomorph from north end of Sheep Mountain Table. Original mineralogy is unknown. Coin for scale is 1.8 mm wide. (H) Frothy mass of chalcedony from north end of Sheep Mountain Table. Void spaces originally filled with mudstone. Coin for scale is 1.8 mm wide. Photos by the authors.
4.4. Dillon Pass fault and the structural features in the north central area of Badlands National Park. The structural contours are on the elevations of the Hay Butte marker. The fine lines are ridge lines along the Badland Wall and below the Sage Creek Rim.
Emplacement of these veins would have required water capable of carrying large amounts of dissolved silicon, calcium, and sulfur in order to produce chalcedony and gypsum. Because the majority of the veins end within the lower part of the Brule Formation, it is most likely that the veins (and fluids) originated at depth and migrated upward. This is supported by the presence of veins within outcrops of the Cretaceous Pierre Shale along Sage Creek near Sage Creek Campground (Plate 1). Several towns in and around the Badlands make use of geothermal wells that produce water heated at depth. Most notable is the town of Hot Springs, South Dakota, on the southern tip of the Black Hills; Wall, South Dakota, just north of the Badlands; and Philip, South Dakota, to the northeast of the Badlands. Geothermal waters are commonly enriched in dissolved minerals that precipitate upon exposure to air (e.g., the mineral precipitates around geysers at Yellowstone National Park), but precipitation at depth is also possible if fluids become oversaturated with dissolved solids. The repetitive and alternating episodes of chalcedony/gypsum mineralization within the same vein structure, healing of fractured veins, and limited areal extent of chalcedony veins support a geothermal model. According to Evans and Terry (1994) and Evans (1999), localized occurrences of tufas and tufa pipes can be found throughout the Badlands and are interpreted to be the result of regional groundwater flow and recharge off the eastern flank of the Black Hills. The distribution of these tufas and pipes may be influenced by structure, such as the numerous normal faults that radiate off the Black Hills.
According to Hoff et al. (2007), chalcedony veins are organized and consistent in their geometry and orientation on the small scale (100 m2), but over larger regions of southwest South Dakota and northwest Nebraska, the orientation of the veins becomes increasingly random. This suggests a lack of tectonic influence and supports a syneresis model of vein genesis. Maher and Shuster (2012) also describe localized areas of uniform vein orientations, especially in association with faults, with increasingly random orientations at larger scales. They noted that chalcedony veins are concentrated at multiple stratigraphic levels within the Badlands, and they can be seen to transition into clastic dike materials upsection. Many of the veins appear to have been subjected to compaction, as assessed by their striated outer textures, suggesting that sedimentary volume loss associated with compaction and diagenetic phase changes in silica, smectite clays, or zeolite minerals, dewatering, and fluid migration are likely responsible for the genesis of these features.
4.5. Faulting along Highway 240. Arrows show relative direction of displacement; X–X' denote offset beds. Photo by the authors.
Postdepositional tectonic forces modified the strata of the Badlands with the formation of folds and faults. Folding is best exemplified by the Sage Arch, an area of uplift and warping in the Dillon Pass area of the park that is oriented northwest–southeast (Plate 8). Associated with this feature, especially in Dillon Pass and along the Sage Creek Rim Road, are numerous faults that were created by the tensional thinning of the rock strata as a result of the uplift of the Sage Arch (Fig. 4.4). These faults, referred to by geologists as normal faults, are characterized by the downward slipping of overlying bodies of rock (headwall block) along a fault plane relative to the underlying footwall block (Plate 8). The movement along these faults can range from only a meter or so to tens of meters and are easily recognized by the juxtaposition of distinctly different rock strata or striping within the buttes. Normal faulting is common along the Badlands Wall from the Sage Creek Rim Road area to the southeast along the Highway 240 loop road, and through the Cedar Pass area of the park (Fig. 4.5). On a larger scale, these same tensional forces created the asymmetric basin that protects the Badlands strata from erosion (Fig. 4.2).
The Badlands that we see today are the result of a shift from deposition to erosion that began about 660,000 years ago (Stamm et al., 2013). The last remnants of deposition in the park are preserved as patches of large gravels and cobbles, referred to as the Medicine Root Gravels, which overlie the lithified Badlands bedrock (Fig. 4.6). During this time, eastern South Dakota was repeatedly subjected to glacial advances and retreats of the Laurentide ice sheet, which generated the relatively flat terrain by deposition of glacial till, a mixture of grain sizes from clay to boulders carried by the glacial ice. The last of these glacial advances, the Wisconsinan glacial episode, pushed the Missouri River to its current position. East of the river, glacial deposits covered older bedrock.
West of the river, bedrock remained exposed at the surface, including deposits of the former Western Interior Seaway and the Badlands, and was subjected to extensive erosion primarily by rivers such as the Bad, White, and Cheyenne, but some erosion was due to wind that carried the lighter particles farther to the east. Erosion and downcutting by rivers is common, although the exact reason for this period of incision in western South Dakota is unknown. Possible triggers include a pulse of regional tectonic uplift, climate change, or a reduction in regional base level. The change from deposition to erosion was also enhanced by the actions of the Cheyenne River, which eroded headward and eventually captured the original drainages flowing east from the Black Hills and starved the Badlands of river-borne sediment. Eolian sedimentation was still possible, as documented by the accumulations of loess on the tops of tables across the region (Rawling, Fredlund, and Mahan, 2003).
In addition to the loss of new sediment from the streams that were captured by the Cheyenne River, the strata of the Badlands themselves contributed to a positive feedback of erosion and exposure that created more surface area to be eroded. The reason for this enhanced feedback lies in the composition of the strata in the Badlands. In particular, the secret lies in the smallest of sedimentary particles, a particular clay known as smectite. Smectite is formed by the weathering of volcanic ash via hydrolysis. Once converted to smectite, the clay is susceptible to swelling in the presence of water, similar to a sponge. Some forms of smectite can expand up to 50 percent of their original volume. Upon drying, the smectite will shrink back to its original size, until the next wetting event. The entire sequence of strata exposed in the Badlands, from the late Cretaceous Pierre Shale up through the Sharps Formation of the Arikaree Group, contains smectite, but some units contain more than others, which is why the stratigraphic units in the Badlands have such different appearances, such as the haystack mound shape of the Chadron Formation versus the cliff and spire morphology of the Brule Formation (Plates 4, 8). The reason for this difference in smectite content can be related to two different processes: original weathering of the volcanic ash grains when the sediments were first deposited and exposed to ancient soil-forming processes, and the amount of volcanic ash that was delivered to the Badlands region.
Within the Pierre Shale, discrete episodes of volcanic ash deposition from late Cretaceous eruptions in the west are preserved as thin whitish-yellow layers of bentonite, a particular type of smectite named for Benton, Wyoming, where it was first described. Windblown ash was also introduced as part of the background sedimentation into the Western Interior Seaway. Smectite is also present in the Fox Hills Formation, but the greater amounts of clastic grains brought in by the rivers that constructed the delta diluted the overall concentration of the smectite clay. This is seen by the change in erosional styles of the rolling hills of the Pierre Shale versus the slightly more cliff-forming strata of the Fox Hills Formation (Plate 1).
With the switch to nonmarine deposition of the White River Group, volcanic ash was incorporated into river and lake deposits and was weathered in the soils that were active at that time. In the late Eocene, climates were humid and tropical, which promoted hydrolysis of the ash into smectite. With the continued lateral migration of meandering rivers, smectites and partially weathered volcanic ash were recycled and redeposited onto new floodplains to continue hydrolysis. The end result was a series of stratigraphic units, the Chamberlain Pass and Chadron formations, which became enriched in smectite compared to the overlying Brule Formation, which contains a greater proportion of unweathered volcanic ash grains. The reason for this greater concentration in ash within the Brule Formation may lie in the hypothesis of climate change from warmer and wetter conditions during the late Eocene to cooler and drier conditions of the Oligocene. It is also possible that the rate of new ash introduced into the Badlands region was too great for hydrolysis to keep pace. Calculations of sedimentation rates within equivalent deposits from Wyoming, Nebraska, and the Badlands suggest that sedimentation rates doubled from the Eocene into the Oligocene (Zanazzi et al., 2007; Terry, 2010).
Regardless of the cause, the result of these differing proportions of smectite in the bedrock of the park explains why the Chadron Formation tends to weather into mounds whereas the Brule Formation is more cliff forming. Geologists refer to this variable resistance to weathering and erosion as differential weathering, which can manifest as differences between geologic units or between individual small-scale beds in a particular unit. Along with the general geomorphology of these two formations, the surface texture of each of these units is the result of their smectite content. The Chadron Formation weathers into a popcorn texture as a result of the repeated shrink and swell of smectite (Fig. 4.6). When wet, the smectite is smooth, sticky, and slippery (a feature referred to locally as gumbo) and is almost impossible to walk or drive across. Upon drying, the popcorn texture becomes hard and abrasive, like sandpaper. This outer veneer of popcorn is gradually washed away by successive storm events and replaced by new popcorn that forms by the infiltration of water down to unaltered bedrock. This outer layer of popcorn can be quite thick – up to several meters in places near the base of some slopes. Beneath the popcorn surface of the Chadron Formation, the unaltered rock of the Badlands is hard, like any other rock, but a fragment of rock placed in a glass of water will break down in a matter of minutes, literally melting in place as the smectite absorbs the water, expands, and sloughs off the side of the rock fragment. The Brule Formation weathers by this same process, but the lesser amount of smectite translates to a thinner popcorn surface and a greater resistance to erosion, which is why the Brule Formation weathers into high spires and cliffs.
4.6. Photographs of various features that postdate deposition of the Badlands strata. (A) Coarse unconsolidated gravel and cobble deposit resting unconformably on Badlands strata. This deposit is at the top of the hill directly east of the Pinnacles Overlook parking lot. Largest cobbles are 15 cm or more in length. (B) Surficial popcorn texture of the Chadron Formation. (C) Geodetic survey marker from the 1950s, now exposed approximately 30 cm above the land surface as a result of erosion of the surrounding bedrock. Photograph by Tyler Teuscher, courtesy of the National Park Service. (D) Dissection of the Badlands by water to create isolated tables (T). Over time, this process leads to large isolated tables, such as Sheep Mountain Table, southwest of Scenic, South Dakota. (E) Hoodoo at the base of Norbeck Pass along Highway 240. (F) Armored mud balls. Photo scale is 10 cm (4 inches). Photos by the authors.
On average, Badlands National Park loses almost 2 cm (1 inch) per year to erosion. This rate of erosion was determined by calculating the amount of sediment that was removed from underneath U.S. Geodetic Survey markers that were originally installed flush with the ground surface in the 1950s. Many of these markers are now lying exposed and on their side (Fig. 4.6). This extremely high rate of erosion also explains why plants that would normally slow the pace of erosion find it hard to gain a foothold. It is this high rate of erosion and the lack of vegetation that give the Badlands its name. Many areas have badlands, but these exposures in Badlands National Park serve as the archetypical example of this geomorphic feature. The overall high rate of erosion gives the White River its name. The whitish color is created by the suspended silts and clays that have been eroded from the surrounding Badlands, which prompted an early settler in this region to describe the White River as “too thick to drink and too thin to plow.”
The dissection of the Badlands has created particular geomorphic features. Some of these features are large in scale, whereas others are localized. The largest of these features are the numerous tabletop buttes throughout the park and the surrounding region, such as Sheep Mountain Table just southwest of the town of Scenic, South Dakota (Fig. P.1). From a distance, the tops of these tables, which in some cases are several hundred feet (50 to 100 m) above the surrounding prairie, can be visually reconnected from one to another (Fig. 4.6). The tops of these tables represent the last and highest stable geomorphic surface in this region before river incision became the dominant geologic process. The rivers did not incise at a steady rate. Along numerous rivers throughout this region are multiple terraces that represent periods of relative stability that allowed the river to laterally migrate back and forth within its valley for a period of time to form a small-scale floodplain before renewed downcutting lowered the level of the active river. Some rivers in and around the Badlands have three to four terrace levels above the active channel.
At a smaller scale within the strata of the Badlands, differential erosion exerts a powerful control on geomorphology. Channel sandstone bodies tend to be more resistant to erosion and can protect smectite-rich rocks below them. In special circumstances, blocks of sandstone can become isolated from the main body of the outcrop to form a hoodoo, a column of softer rock protected by a sandstone cap. Over time, these weaker columns of rock will erode, eventually toppling the sandstone cap. Hoodoos are found throughout the park, but they tend to be small-scale features (Fig. 4.6). At Toadstool Geologic Park in northwest Nebraska, large blocks of channel sandstone protect soft, smectite-rich mudstones to produce toadstools, which gave this park its name. The majority of the largest toadstools have since fallen, but new ones are presently forming.
The clastic dikes and chalcedony veins throughout the Badlands are more resistant to erosion than the surrounding rock. This increased resistance to erosion acts as an internal support to the surrounding strata and helps to hold up the ridges of the Badlands. When viewed from above, these dikes and veins have a preferred northwest–southeast and northeast–southwest orientation, which by default creates a preferred orientation in the buttes (Fig. 4.7). On a larger scale, this same orientation is expressed in river and stream drainages that dissect the Badlands. This symmetrical arrangement of rivers and streams is due to the exploitation of fractures in the bedrock that are less resistant to weathering and erosion.
Within the bedrock areas where dikes and veins are absent, the Badlands erode into small-scale gullies and rills with steep knife-edge ridge tops and slopes. Water is able to exploit smaller-scale fractures within the rock and, over time, develop networks of tubes and pipes under the popcorn weathered surface, similar to karst terrains (caves) that form in limestone. This piping erosion can, over time, wash away large areas of bedrock under the surface of the Badlands, which will eventually collapse in on itself and form a sinkhole.
Given the highly erosive nature of the rock in the Badlands, it is fortunate that this part of the United States only receives approximately 13 inches of rain per year. Otherwise the Badlands would have washed away long ago. The rain that does fall tends to come in storms and rapidly inundates the landscape. The smectites in the bedrock swell and act as a protective barrier to retard deep penetration of precipitation, instead forcing the water to flow over the saturated surface into the ephemeral streambeds and narrow gullies throughout the exposures. With rapid precipitation events, this can induce flash flooding as the water quickly collects and moves downslope. These flash floods can be quite powerful and are capable of moving large cobbles and chunks of more resistant rock. One of the more interesting features of these flash flood events is the creation of armored mud balls (Fig. 4.6). As the flooding proceeds, small pebbles are rolled along the bottom of the streambed and pick up bits and pieces of sticky clay. As this pebble moves along and grows in size, other, smaller pebbles can be picked up and incorporated, similar to rolling a snowball downhill. The farther the mud ball travels, the larger it will become, until the stream is no longer able to move it. Most mud balls are the size of a golf ball to the size of a softball, although some the size of basketballs are occasionally seen.
4.7. Portion of the U.S. Geological Survey Wall SW 7.5 Minute topographic map showing the preferred orientation of buttes and ephemeral streams.
Not all of the water that falls on the Badlands is incorporated into gullies and streams. Some simply washes down the sides of the buttes and onto the prairie. As it does so, this water carries clays and silts with it that are deposited as a broad sheet at the base of the slope (colluvium). Over thousands of years, these individual paper-thin colluvial deposits can accumulate into packages several meters thick. The prairie vegetation continues to grow upward as more sediment is added, which in turn protects the colluvium from erosion. Eventually these packages of individual sheet-wash storm events will be dissected by erosion to form sod tables, isolated patches of laminated sheet-wash deposits several meters high that are capped by prairie vegetation (Fig. 4.8). The prairie cap represents the former stable land surface before dissection and can commonly be traced from sod table to sod table.
Upon closer inspection, the laminae of colluvium that comprise the sod tables are sometimes disrupted by darker bands 10 to 20 cm thick. These bands represent periods of geomorphic stability during which dark, organic-rich surface horizons of prairie soils could form as a result of slower rates of sediment influx. With the return of increased sediment input from the slope wash, the topsoil could not keep pace with deposition. In addition to buried soil horizons, sod tables also frequently display burrows and dens of prairie animals that were excavated before dissection of the landscape. Some sod tables preserve former campsites of indigenous populations that inhabited the Badlands thousands of years ago. In other parts of the park, where older bedrock of the Pierre Shale is exposed at the surface, dark-colored colluvium from the shale interfingers with light-colored river deposits coming out of the Badlands (Fig. 4.8).
4.8. Photographs of various features that postdate deposition of the Badlands strata. (A) Sod tables (S) formed by the dissection of sheet-wash deposits. The tops of the tables represent the former land surface. (B) Dark-colored colluvial sediments from the Pierre Shale interfingering with light-colored alluvium from stream flooding. (C) Rockfall in Norbeck Pass along Highway 240. (D) Earth flow in the Pierre Shale near the Sage Creek Campground. Photos by the authors.
The most dramatic types of erosion in the Badlands manifests as catastrophic mass wasting events, such as rockfalls, and slower-moving landslides, and can occur with or without the presence of heavy rains. Rockfalls can occur almost anywhere in the park but are most commonly associated with fractures in the bedrock that eventually break away or by undermining of cliff faces by rivers and streams during periods of heavy rain (Fig. 4.8). Other rockfalls are associated with faults, such as along the high cliff face of the Cliff Shelf Nature Trail in Cedar Pass, and form aprons of debris at the base of the cliffs. Highway 240 from the northeast entrance of the park descends from the upper prairie geomorphic surface to the lower prairie through Cedar Pass, just northeast of the Ben Reifel Visitor Center (Fig. P.1). This steep, twisting road is actually developed on several large landslide blocks that have been moving gradually downslope for decades. Evidence of this movement can be seen in the twisted stripes of rock along the highway and the constantly degrading road bed at the lip of Cedar Pass, which is split and offset by this constant movement. Periods of heavy rain tend to speed the rate of downhill movement. These landslide blocks were stabilized several years ago by installing artificial drainage systems to divert the water. On occasion, periods of heavy rain saturate less consolidated soils, causing them to flow downhill (an earth flow) on top of more solid bedrock beneath. This is especially common in the Pierre Shale (Fig. 4.8).
4.9. Physiographic map showing the late Pleistocene features of South Dakota and Nebraska. Winds blowing from the northwest (white arrows) through the Badlands and paralleling the Laurentide ice sheet margin picked up sand and silt grains. This sediment blew over the Pine Ridge and dropped sand-size grains in the dunes of the Sand Hills of southern South Dakota and central Nebraska (areas surrounded by a heavy line). The silt-size dust particles were transported across the Sand Hills and were deposited on the lee side of the dunes in eastern and southern Nebraska, forming the thick Peoria Loess. The cross-sectional model is modified from Muhs et al. (2008:fig. 23), after Mason (2001). The base map for this diagram is from the U.S. National Atlas Web site (http://nationalatlas.gov/mapmaker).
4.10. Aerial photograph of some of the Pleistocene dunes in the Big Badlands. The area shown is on the table lands between Big Hollow Creek and Cain Creek. The white dotted lines show the crests of some of the parabolic dunes in the area. The Pleistocene winds blew from the northwest, as shown by the alignment of the dunes with the “arms” of the dunes extending toward the northwest. Also shown are some of the blowouts (b), which are depressions formed by local erosion by winds. The base photo is from the U.S. Department of Agriculture, taken on September 25, 2011, and is available on Google Earth.
Water is not the only agent of erosion in the Big Badlands. Windstorms can occur in the late spring, summer, and fall and pick up a great amount of fine-grained sediment, forming huge dust clouds. This is especially true during droughts, such as in the 1930s and as recently as the early 2000s, but a huge amount of wind erosion in the Badlands occurred during the last glacial maximum, between 14,000 and 25,000 years ago (Muhs et al., 2008). Glaciers did not occur in the Big Badlands during the Pleistocene, but during the greatest maximum extent of the Laurentide ice sheet, the margin of the continental glacier was 240 km to the east, where the Missouri River now flows (Fig. 4.9). Climatic conditions in the Badlands during the last glacial maximum were dry and exceptionally windy, with strong winds blowing from the northwest paralleling the margins of the continental glacier. The dry and cold conditions supported little vegetation, and silt grains eroded from the siltstones of the Brule and the Sharps formations were blown to the southeast across the Sand Hills of Nebraska, then deposited in eastern and southern Nebraska as a series of very thick loess deposits (Fig. 4.9). The thickest and most widespread of these dust deposits is the Peoria Loess, which reaches a maximum thickness of 40 m in central Nebraska. The Peoria Loess covers all of southern and eastern Nebraska, extending into Kansas, across Iowa, and into Illinois. Radiometric ages of detrital zircon grains and lead (Pb) isotopes within potassium feldspar grains in the Peoria Loess of Nebraska indicate a source from the White River Group and Sharps Formation in the Big Badlands (Aleinikoff et al., 2008). The ancient volcaniclastic loess deposits of the Brule and Sharps formations provided a source of silt grains for the Pleistocene loess of Nebraska, just like the ancient sand dunes of the Navajo and Entrada sandstones of Arches National Park are sources of modern sand in eolian dunes in eastern Utah. The thick Pleistocene loess deposits in Nebraska are some of the most productive agricultural lands in the world, and much of these sediments were derived from the Brule and Sharps formations in South Dakota.
Quaternary eolian deposits in the Big Badlands occur on the tops of the higher tables, such as on Sheep Mountain Table, Quinn Table, and Bouquet Table (Fig. P.1). These deposits include loess deposits and sand dunes that have been dated as late Pleistocene in age (Burkhart et al., 2008). The sand dunes are especially well preserved on the tables south of the Badlands Wall, such as on Imlay Table and to the west of Cain Creek (Fig. 4.10). These sand dunes are all parabolic dunes that have long ridges that point upwind and connect on the downwind side. These dunes are all stabilized by vegetation and did not move even during the Dust Bowl of the 1930s. Their orientation indicates winds blowing from the northwest. The presence of Pleistocene sand dunes in the Badlands region is significant, for silt either in the bedrock or reworked into stream deposits will not be picked up into the atmosphere as dust from only the direct action of winds. This is because of the weak cementation of the siltstone bedrock, and because the smoothness of any silt–sediment surfaces prevents the wind from lifting the silt into the air. It takes bouncing (saltating) sand grains during windstorms to hit the siltstone or silty sediments and launch the silt into the atmosphere as dust. Thus, saltating sand grains during late Pleistocene windstorms abraded the siltstone outcrops and silty stream sediments, sending up clouds of silt-size dust from the Badlands that eventually settled in Nebraska, forming thick loess deposits. The current limited outcrops of the Poleslide Member of the Brule Formation and the Sharps Formation in the Big Badlands is a result of not only stream erosion but also wind erosion.
