Late Pleistocene Dunes along the Dempsey Divide

Extensive sand sheets and associated dune fields are common along the major braided streams crossing the Great Plains and cover much of the northern half of Nebraska. These dunes are mostly of Middle to Late Holocene age (e.g. Ahlbrandt et al. 1983; Arbogast 1995; Ferring 1995; Holliday 1985, 1989; Madole 1995; May et al. 1994; Muhs and Holliday 1995; Muhs et al. 1997; Swinehart et al. 1988; Wright et al. 1985). Since 1994, we have been studying a dune field in far western Oklahoma that has proven to be considerably more ancient.

The Dempsey Divide is the high interfluve between the North Fork of the Red and the Washita rivers in Beckham and Roger Mills counties, Oklahoma. A narrow east-west peninsula of basal Tertiary Ogallala Formation, averaging 8-10 km in width, occupies the crest of the divide in southern Roger Mills County. The Ogallala Formation outcrop is underlain unconformably by Permian redbeds. The sandy Ogallala clastics have been extensively reworked into an aeolian landscape of scattered dunes, which vary widely in size and shape, and internally drained deflation basins. Tributaries of the North Fork and Washita are actively eroding the edges of this Ogallala outlier, particularly the steep-gradient Washita tributaries to the north.

A remarkable concentration of Late Paleoindian, Late Archaic and Woodland archaeological sites occurs along the divide, particularly along the outcrop boundary between the Ogallala and the Permian redbeds (Bement and Buehler 1994; Buehler 1997; Thurmond 1990, 1991a, 1991b, 1991c, 1997; Wyckoff 1992). The contrasting lithology of the Tertiary and Permian units produces two distinct landscapes, with different soils, plants and animals. The deep, sandy, mollic soils and gentle relief of the Ogallala outcrop support a mosaic of forest, brush and tall grass, with a diverse assemblage of small game and wild plant foods. The Permian outcrops are characterized by thin, fine-textured soils under a much lower biomass of short to mid-height grasses, with narrow riparian belts of limited species diversity. Shortgrass attractive to bison dominates the Permian outcrops. Nearly every stream valley transecting the Tertiary/Permian outcrop boundary is spring-fed, as groundwater flows from the base of the Ogallala aquifer. Archaeological sites are concentrated along the ecotonal edge associated with this outcrop boundary, in proximity to the ubiquitous spring-fed streams. Knappable stone is available in gravels lagged by the westward erosional recession of the Ogallala Formation. Finally, the divide afforded pedestrian hunter-gatherers a ready ‘gangplank’ for travel between the Llano Estacado to the west and the Rolling Plains to the east.

In hopes of obtaining data pertinent to a reconstruction of Quaternary geomorphic and climatic change on the Dempsey Divide, since 1994 we have been documenting and dating paleosols in dunes, upland valley fills and pond deposits from the crest of the divide northward to the Washita floodplain. We have dated Late Holocene valley fill paleosols in eight locations which superficially replicate the bimodal Copan/Delaware Canyon paleosol sequence (Ferring 1982; Hall 1982, 1990; Hall and Lintz 1984). However, we have also documented multiple paleosols at five exposures dating 50 BC to AD 1650 which manifest a roughly 400 year pluvial/interpluvial cycle, with the pluvials lasting 185 years on average, and the interpluvials 205 years. Our ongoing analysis of marsh, stream and pond deposits dating 9-28 ka BP, which are now 30 m above the modern streams and as much as 2 km beyond the modern Ogallala Formation erosion front, indicate the Ogallala boundary on the north side of the divide has receded at a remarkable average rate of some 8 m per century over the past twenty-eight millennia, and that the landscape immediately beyond the erosion front has lowered an average of 15 cm per century.

The first dune we dated (Trammell Dune #1) is a large linear one (12 m high, covering 4 ha, oriented NW-SE) on the Washita floodplain at Cheyenne, adjacent to the north bank of the modern river channel. We documented a mollic paleosol at the base of the dune, at the same elevation as the modern floodplain surface, and three melanized horizons within the dune. We expected the subdune soil to be Early Holocene, and the soils within the dune to be Late Holocene. The subdune soil dated 22,850 +/- 290 yr BP (NZA-4069). The soils within the dune, representing mesic intervals of dune vegetation and stability, dated 18,870 +/- 230 yr BP (NZA-4081, 1.5 m above dune base), 14,120 +/- 190 yr BP (NZA-4082, 2 m above dune base) and 4,670 +/- 150 yr BP (NZA-4083, 2.4 m above dune base).

Surprised by these dates, we shifted our focus some 15 km to the southwest to the crest of the Dempsey Divide to see how the dunes atop the divide compared in age. We had also assumed these dunes to be largely of Altithermal origin. Olson Dune #1, a 6 m high, 1.8 ha, SSW-NNE oriented linear dune, yielded a date of 25,970 +/- 270 yr BP (NZA-5739) from a melanized horizon 3 m above the dune base. A date of 9,370 +/- 97 yr BP (NZA-5738) was obtained from a second paleosol 4.6 m above the dune base. Next we cored Olson Dune #2 (a 4.5 m, 0.4 ha, round dune), 2 km southwest of Olson #1, and sampled three melanized epipedons within the dune. The lowest paleosol, .6 m above the base, dated 20,070 +/- 340 yr BP (NZA-6183). The next one up, 2.8 m above the dune base, dated 17,520 +/- 180 yr BP (NZA-6182). The uppermost soil, 3.7 m above dune base, dated 10,750 +/- 120 yr BP (NZA-6181). We cored an interdune basin 100 m to the southeast of Olson Dune #2, and assayed a date of 21,970 +/- 210 yr BP (NZA-5799) on a melanized silt loam 4 m below the modern surface. Finally, at Olson Dune #4, a 3.5 m high, 0.4 ha, SSW-NNE oriented linear dune 1.6km northwest of Olson #1, we dated a paleosol .7 m above the dune base at 17,930 +/- 180 yr BP (NZA-6198).

These are not ordinary Great Plains dunes by any measure. Again, the dunes atop the divide are quite variable in size, shape and orientation. They tend to be widely scattered. Most interesting to us is the fact that the dunes we have dated appear to have been fixed in place for up to twenty-six millennia, accumulating during interpluvials and stabilizing under vegetation during the pluvials, but not appreciably moving. There is considerable cementation of the sand by carbonates in these dunes, which is not surprising given the Ogallala Formation origin of the sand. Carbonate consolidation during the pluvials is probably responsible for the locational stability of the dunes. Analogous carbonate cementation of the soils in the broad areas between the dunes, and the patchy distribution of sandy parent material characteristic of the Ogallala, are presumably responsible for the wide separation of the dunes.

The dunes we have radiocarbon dated on and adjacent to the Dempsey Divide are far more ancient than prior work on the Great Plains had led us to expect. The dune dates are remarkably consistent within our study area, and correlate well with the dates for black mat formation in the Trans-Pecos Glacial Lake King, east of El Paso (Wilkins and Currey 1997), with highstands at Glacial Lake San Augustin, west of Socorro (Phillips et al. 1992) and with stream discharge peaks into Glacial Lake Estancia southeast of Albuquerque (Allen 1993; Allen and Anderson 1993, 1995). We believe we are seeing western Oklahoma expressions of the glacial climatic oscillation well known in the north Atlantic (Bond et al. 1997; Dansgaard et al. 1993; Oppo et al. 1998), expressed in the southwestern United States as pluvial maxima in the El Nino-Southern Oscillation (Benson et al. 1996; Heusser and Sirocko 1997; Oviatt 1997) within an overall pattern of a southward displacement of the southern branch of the jet stream (Kutzbach and Guetter 1986; Hostetler at al. 1994, Benson et al. 1998). A comparison of the Lake King, Lake Estancia, and Dempsey Dune dates (Table 1) appears to support the inference of a 400-year pluvial cycle for the region during the Last Glacial Maximum proposed by Phillips et al. (1992). The fact that we see the same 400 year in the Late Holocene paleosols of the last two millennia within our study area is intriguing, and suggests a persistent periodicity in the regional climate that manifests itself in both glacial and interglacial times.

______________________________________________________________________________________

Table 1. Comparison of Lake King Black Mat Dates,

Dempsey Dune LGM Paleosol Dates, and

Estancia Basin Stream Discharge Peaks

______________________________________________________________________________________

We would like to thank William C. and Elaine Olson and Bill and Joyce Trammell for graciously granting us access to their property during the conduct of this research. Our thanks also to Dee Ann Story and an anonymous reviewer for their careful and constructive critiques of the first draft of this article.

References Cited

1983 The Dynamic Holocene Dune Fields of the Great Plains and Rocky Mountain Basins, U.S.A. In Eolian Sediments and Processes, pp. 379-406. M.E. Brookfield and T.S. Ahlbrandt, editors. Elsevier, New York.

1993 Late Quaternary Lacustrine Record of Paleoclimate from Estancia Basin, Central New Mexico, U.S.A. Ph.D. Dissertation, University of New Mexico, Albuquerque.

1993 Evidence from Western North America for Rapid Shifts in Climate during the Last Glacial Maximum. Science 260: 1920-1923.

1995 Field Guide to Estancia Basin, New Mexico. Compiled for PAGES, PEP-1 (Americas) Transect Science Meeting. Albuquerque.

1995 Paleoenvironments and Desertification on the Great Bend Sand Prairie in Kansas. Kansas Geological Survey Open File Report 95-33. Lawrence.

1994 Preliminary Results from the Certain Site: A Late Archaic Bison Kill in Western Oklahoma. Plains Anthropologist 39(148):173-183.

Benson, Larry V. , James W. Burdett, Michaele Kashgarian, Steve P. Lund, Fred M. Phillips and Robert O. Rye

1996 Climatic and Hydrologic Oscillations in the Owens Lake Basin and Adjacent Sierra Nevada, California. Science 274:746-749.

Benson, Larry V., Steve P. Lund, James W. Burdett, Michaele Kashgarian, Timothy P. Rose, Joseph P. Smoot and Martha Schwartz

1998 Correlation of Late Pleistocene Lake-Level Oscillations in Mono Lake, California, with North Atlantic Climate Events. Quaternary Research 49(1):1-10.

Bond, Gerard , William Showers, Maziet Cheseby, Rusty Lotti, Peter Almasi, Peter deMenocal, Paul Priore, Heidi Cullen, Irka Hajdas and Georges Bonani

1997 A Pervasive Millennial-Scale Cycle in North Atlantic Holocene and Glacial Climates. Science 278:1257-1266.

1997 Where’s the Cliff? Late Archaic Bison Kills in the Southern Plains. Plains Anthropologist Memoir 29:135-144.

Dansgaard, W. , S.J. Johnsen, H.B. Clausen, D. Dahl-Jensen, N.S. Gundestrop, C.U. Hammer, C.S. Hvidberg, J.P. Steffensen, A.E. Sveinbjornsdottir, J. Jouzel and G. Bond

1993 Evidence for General Instability of Past Climate from a 250-kyr Ice Core Record. Nature 364: 218-220.

1982 The Late Holocene Prehistory of Delaware Canyon, Oklahoma. North Texas State University Contributions in Archaeology 1.

1995 Middle Holocene Environments, Geology, and Archaeology in the Southern Plains. Geological Society of America Special Paper 297:21-66.

1982 Late Holocene Paleoecology of the Southern Plains. Quaternary Research 17:391-407.

1990 Channel Trenching and Climatic Change in the Southern U.S. Great Plains. Geology 18:342-345.

1984 Buried Trees, Water Table Fluctuations, and 3,000 Years of Changing Climate in West-Central Oklahoma. Quaternary Research 22:129-133.

1997 Millennial Pulsing of Environmental Change in Southern California from the Past 24 ky: A Record of Indo-Pacific ENSO Events? Geology 25(3): 243-246.

1985 Holocene Soil-Geomorphological Relations in a Semi-Arid Environment: The Southern High Plains of Texas. In Soils and Quaternary Landscape Evolution, edited by J. Boardman, pp. 325-357. John Wiley and Sons, Ltd. New York.

1989 Middle Holocene Drought on the Southern High Plains. Quaternary Research 31(1):74-82.

1994 Lake-Atmosphere Feedbacks Associated with Paleolakes Bonneville and Lahontan. Science 263:665-668.

1986 The Influence of Changing Orbital Parameters and Surface Boundary Conditions on Climate Simulations for the Past 18,000 Years. Journal of Atmospheric Science 435:1726-1759.

1995 Spatial and Temporal Patterns of Late Quaternary Eolian Deposition, Eastern Colorado, U.S.A. Quaternary Science Reviews 14:155-177.

1994 Late Quaternary Fluvial and Eolian Sediments: Loup River Basin and the Nebraska Sand Hills. Guidebook, Friends of the Pleistocene Field Trip, 13-31.

1995 Evidence of Active Dune Sand on the Great Plains in the 19^th Century from Accounts of Early Explorers. Quaternary Research 43(2):198-208.

Muhs, Daniel R., Thomas W. Stafford, James B. Swinehart, Scott D. Cowherd, Shannon A. Mahan, Charles A. Bush, Richard F. Madole and Paula B. Maat

1997 Late Holocene Eolian Activity in the Mineralogically Mature Nebraska Sandhills. Quaternary Research 48(2):162-176.

1998 Abrupt Climate Events 500,000 to 340,000 Years Ago: Evidence from Subpolar North Atlantic Sediments. Science 279:1335-1338.

1997 Lake Bonneville Fluctuations and Global Climate Change. Geology 25(2):155-158.

Phillips, Fred M., Andrew R. Campbell, Cynthia Kruger, Peggy Johnson, Randall Roberts and Eric Keyes

1992 A Reconstruction of the Response of the Water Balance in Western United States Lake Basins to Climatic Change. New Mexico Water Resources Research Institute Report 269.

1988 Field Guide to Geology and Hydrology of the Nebraska Sand Hills. Colorado School of Mines, Professional Contributions 9.

1990 Late Paleoindian Utilization of the Dempsey Divide on the Southern Plains. Plains Anthropologist Memoir 25.

1991a A Late Archaic/Woodland Lunate Stone Burial from Far Western Oklahoma. Bulletin of the Texas Archeological Society 60:241-255.

1991b Archaeology of the Dempsey Divide, a Late Archaic/Woodland Hotspot on the Southern Plains. Bulletin of the Oklahoma Anthropological Society 39:103-157.

1991c: An Additional Late Paleoindian Component on the Dempsey Divide. Bulletin of the Oklahoma Anthropological Society 39:159-169.

1997 A Recently Recorded Late Woodland Campsite on the Dempsey Divide: Thurmond Ranch #112. Newsletter of the Oklahoma Anthropological Society 45(4):9-11.

1997 Timing and Extent of Late Quaternary Paleolakes in the Trans-Pecos Closed Basin, West Texas and South-Central New Mexico. Quaternary Research 47(3):306-315.

1985 Pollen Diagram from the Nebraska Sandhills, and the Age of the Dunes. Quaternary Research 24:115-120.

1992 Archaic Cultural Manifestations on the Southern Plains. Revista de Arqueologia Americana 5: 167-199.

[1] J. Peter Thurmond, Rt. 1, Box 62-B, Cheyenne, OK, 73628-9729 (dempseydiv@aol.com).

[2] Don G. Wyckoff, Oklahoma Museum of Natural History, University of Oklahoma, 2401 Chautauqua Avenue, Norman, OK, 73072-7209 (xtrambler@ou.edu).

Lake King	Dempsey Dunes	Estancia Basin
17.2	-	17.2
-	17.5	17.6
-	17.9	18.0
19.0	18.9	-
-	-	19.8
-	20.1	20.1
-	-	21.7
-	21.9	21.9
22.6	-	-
-	22.8	-
-	-	23.2
24.7	-	-
-	26.0	-