Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

W.M. Laird


The geomorphology and glacial geology of the north half of Barnes County in east central North Dakota is described and interpreted. The area comprised a portion of the Western Lake section of the Central Lowland province and was informally designated the Middle Sheyenne River area without defining its boundaries. Four physiographic subdivisions in the Middle Sheyenne River area were assigned: 1. The Kensal Glacial Plain subdivision that embraces land between the Sheyenne River and the west county boundary. 2. The Sheyenne River Valley subdivision. 3. The Glacial-Lake Plains subdivision that consists of three separate but genetically related areas of lacustrine silts. 4. The Luverne Ridge and slope subdivision that includes the long straight upland area that trends north-south through E. 57 W., and the eastward sloping surface to the east county boundary. Glacial and glaciofucial landforms of subdivisions are described and their mode of origin discussed.

Pleistocene stratigraphy is discussed under two headings: Indeterminant stages, and the Wisconsinan stage. Two deposits of indeterminant age are represented. One consists of inclusions of till of orange color in a buff till matrix. The boundaries of the inclusions are sharp, the calcite and dolomite content is three to four times that of the enclosing till, and the lithology is distinctive. The inclusions are interpreted at a till older than the enclosing till matrix. No source of the orange is known in eastern North Dakota.

The other till of indeterminant age is gray, dense, stony clay that is mapped as a lithostratigraphic unit and named the Sheyenne Valley Formation. It is characterized by joints and pebble molds that are lined with stains of iron and manganese. Where exposed, it is capped by boulder pavement that seperates it from the overlying buff till. It forms several topographic promontories in the Glacial-Lakes Plains and Luverne Ridge and Slope subdivisions.

In the Wisconsinan stage, four drifts occur in the area of this report. The oldest is interpreted as Buchanan (?) drift that occurs as a series of drumlinoid hills in the northwestern part of the county. It is believed these hills are an overridden and moraine of Buchanan (?) drift reshaped by Kensal ice. Sharp lithologic variation from surrounding Kensal till, remnants of outwash southeast of the lines of hills, and that parallel to the strike of the Buchanan end moraine in Stutsman County, all prompt this interpretation.

Kensal drift, the oldest continuous surface drift in Barnes County, covers the entire area west of the Sheyenne and Baldhill Creek Valley. Cooperstown end moraine bounds the Bald Creek Valley, crosses the Sheyenne Valley near the southeast corner of T. 143 N., R. 58 W,. apparently formed a re-entrant, then continuous south along the east edge of the Sheyenne Valley. The uppermost Wisconsinan drift in Barnes County is herein named the Luverne drift. This unit consists of a long straight and moraine, its attendant outwash on the west, and sloping ground moraine to the east edge of the county.

Historical interpretation places all Barnes County surface glacial deposits in Late Wisconsinan subsequent to the emplacement of ice that formed the Streeter Moraine and the dead-ice features of the Burnstad drift on the Coteau du Missouri. Three terraces in the Sheyenne Valley post-date the Kensal and Cooperstown moraines, but LuTerne meltwater may have helped form the third of lowest. Data is summarized showing the relationship of these terraces to diversion of Lake Souris water and deposition of the Sheyenne Delta in Lake Agassiz I.

Effects on Barnes County glacial chronology of the contrasting views of Leighton and Elson concerning existence of Valders ice ln North Dakota are summarized. If Leighton was correct, Valders ice formed the Kensal moraine which correlates with the Bigstone moraine in Minnesota. In this view Kensal, Cooperstown, and Luverne moraines are all post-Two Creekan in age. Elson’s view that Valders ice did not reach North Dakota would place Barnes County deposits in the Woodfordian or pre-Two Creeks interval. If this view is accepted, available radiocarbon dates of 11,650 years from wood in the Burnstad drift cannot be associated with active Burnstad ice. Rather, an earlier date for Burnstad advanve and subsequent stagnation must be assigned to allow subsequent active ice to build moraines in Barnes County and to disappear from North Dakota by Two Creeks time.

Block (554041 kB)

Included in

Geology Commons