Why the sudden change? It can't be moisture. Precipitation does increase going east but only gradually. And there's no significant change in elevation, nothing to disrupt South Dakota's gentle downward slope from the Rocky Mountains eastward. But there is a huge difference in substrate, as Nicollet noted:
As it flows, a glacier grinds and planes the land and carries off the fragments in its base, making it even more abrasive. When it melts, it leaves behind thick deposits of ground-up material—Nicollet's "erratic deposites", now called ground moraine. After multiple advances and melting during interglacials, eastern South Dakota was covered in ground moraine—the source of the fertile soils east of the Missouri River.
Many years later, USGS geologist Richard Foster Flint agreed with Warren. "The belief, advanced as early as 1869, that an ice sheet flowing southwestward blocked these valleys and detoured the drainage so as to form the Missouri River is confirmed." As evidence, Flint described three anomalous features of the Missouri (two can be seen in the map above). First, it doesn't follow the gradual decrease in elevation west to the east. In fact it flows mostly south, perpendicular to regional slope. Second, its valley generally is steeper-walled than those of its tributaries, indicating youthfulness. Finally, distribution of its tributaries is very lopsided—all major streams enter from the west (Flint 1955).
In summary, the glacial deposits covering eastern South Dakota end at the Missouri River because that was the limit of glacial advance. But glaciers didn't stop there because of the river. In fact there was no river until ice blocked east-flowing streams, sending them south along the margin of the ice sheet to become part of the longest river in North America—the great Missouri, 2546 miles in length from its headwaters in the Rocky Mountains to its confluence with the Mississippi (2).
Travel up the swift-flowing river was terribly slow and arduous. The men paddled, sailed and too often pushed or pulled their boats, including a heavy metal-framed keelboat. (It was sent back down the river from their winter camp, with collections, reports and a map.) It's thought that each boatman ate on the order of nine pounds of meat per day! Fortunately game was abundant (Johnson 2022).
The Corps of Discovery reached the Pacific Ocean in November of 1805, but only after leaving the headwaters of the Missouri, crossing the continental divide, and traveling months by foot and canoe to the Columbia River and on to the coast. Clearly the Missouri did not offer a water route to the Pacific. Even so, it would be an important transportation corridor ... in spite of its treachery.
Hazards included sand bars, snags, and wrecked boats. An estimated 400 steamboats were sunk or otherwise destroyed on the Missouri; the average lifespan of a steamboat was five to seven years (Johnson 2022).
Memories had been surfacing of a little girl and her younger brother playing Lewis and Clark long ago. Those legendary explorers had struggled up the Missouri, enduring summer heat and mosquitos, surviving winter cold and food shortages, all the while not knowing what lay ahead, what discoveries awaited! Lewis and Clark captured our imaginations, and for me they still do. We lived far from the Missouri then, but now I had a chance to travel where they had traveled, to see the great river flowing as they had.
I started from the parking lot of the Adams Homestead and Nature Preserve north of the Missouri River in the southeast corner of South Dakota. The route followed dirt roads and was well signed. You can tag along on this aerial photo (modified from Google Earth).
(1) For more about the raging flood vs. glaciers debate and Nicollet's involvement, see Mssr. Nicollet & I consider Glacial Theory and Glacial Beauty in South Dakota.
(2) At their confluence, the Missouri River is longer than the Mississippi River (upstream). By convention the longer river retains its name. Why did the Missouri lose out to the Mississippi? See discussion here.
(3) Nearby McCook Lake (in the upper part of the aerial photo) also is an abandoned meander but is older, being shown on a map dated 1895. Perhaps it's a natural one.
(4) Construction of dams and levees drastically reduced the amount of sediment carried by the Missouri River, to 1% of what it had been (Johnson 2022). Even so, it continues to deliver more than half the silt emptied into the Gulf of Mexico! (more here)
Sources (in addition to links in post)
Flint, RF. 1955. Pleistocene geology of eastern South Dakota. Geol. Surv. Prof. Paper 262.
Gries, JP. 1996. Roadside Geology of South Dakota. Mountain Press Publ. Co.
Johnson, WC. 2022. The Missouri River. Chapter 10 in Johnson & Knight 2022.
Johnson, WC, and Knight, DH. 2022. Ecology of Dakota Landscapes; past, present, and future. Yale University Press.
Nelson, Mike. 2014, Jan 29. Geology: eastern South Dakota in CSMS Geology Post. I also thank Mike for directing me to Flint's 1955 paper.
Nicolett, JN. 1843. Report intended to illustrate a map of the hydrographical basin of the upper Mississippi river. US Senate, 28th Congress, 2nd session, no. 237. BHL
Warren, GK. 1869. General considerations regarding the physical features of these rivers: US Army, Corps of Engineers, Rept. Chief of Engineers, 1868, p. 307-314. Full report online. (Conflicting years are puzzling but real.)