The Driftless Area: Fewer glaciers but more topography than the rest of Minnesota

A post by Anne JeffersonTucked into the corner where Minnesota, Wisconsin, and Iowa meet there’s a special area with a Quaternary history that sets it apart from the rest of the northern United States.

At the Last Glacial Maximum, the Des Moines lobe lay to the west of this area and the Green Bay lobe lay to the east. But in this area, the land surface was not covered with ice. For this reason, extreme southeastern Minnesota, northeastern Iowa and western Wisconsin together are known as the Driftless Area, because drift is an old name for till, and where there were no glaciers, no till could be deposited.

Even before the last glacial period, the Driftless Area seems to have uniquely escaped the terrain smoothing, till depositing influences of the ice sheets. (Play with this animation to watch southeastern Minnesota avoid glacial advance after glacial advance.) The map below shows the maximum extent of glaciers at (a) 1 million years ago, (b) ~600,000 years ago, (c) ~250,000 years ago (the Illinoian glaciation) and (d) ~22,000 years ago (Wisconsinan glaciation). In all of those reconstructions, there’s a stippled doughnut hole that defines the Driftless Area.

Glacial advances relative to the formation of the Driftless Area (Reinertsen, 1992*)

Glacial advances relative to the formation of the Driftless Area (Reinertsen, 1992*)

While in most parts of the Upper Midwest, the Paleozoic stratigraphy is buried under glacial deposits, millions of years of uninterrupted erosion have spectacularly dissected the landscape of the Driftless Area, creating 150+ m bluffs and narrow valleys. This dissected landscape stands out in sharp contrast to the flatter glaciated areas which surround it, as shown in the image below.

Topography of the Driftless Area and surrounding regions

Topography of the Driftless Area and surrounding regions

On the ground, the topography is even more dramatic, particularly along the Mississippi River valley. In the image below, two hillsides part of Great River Bluffs State Park in Minnesota show the steepness of relief that can be found in the area.

Hillsides of the Driftless Area (image by McGheiver on Wikimedia)

Hillsides of the Driftless Area (image by McGheiver on Wikimedia)

Growing up in this region, I always wondered why people said the Midwest was flat!

*If anyone can help me track down the full citation of this article, I’d be most appreciative. I borrowed the image from here.

Categories: by Anne, geomorphology, outcrops, Pleistocene

Comments (21)

  1. Lockwood says:

    I think there’s an area east of Cleveland- much smaller in extent- that also was never glaciated. Most of the region is low, rolling hills, but the area south of Mentor is quite rugged. I’ll see if I can find a reference/link.

  2. Silver Fox says:

    Very cool! (Can’t help with a reference, though.)

  3. Eamon Knight says:

    I wish I’d known this five years ago when we drove from Quad Cities up the valley to Spillville (you’ll have to guess or Google why anyone might want to see Spillville, but it had nothing to do with geology). Then we crossed over to Wisconsin at Prairie du Chien, and toured a cave southeast of there that I think would still be within the Driftless Area. Beautiful scenery.

  4. Matt Kuchta says:

    Anne, I found this:
    Reinertsen, D.L., (1992). Guide to the Geology of the Galena Area. Illinois:
    Illinois State Geological Survey.

    Although that map isn’t as accurate as could be (during the mid Pleistocene, the DA probably wasn’t quite the complete “island” – aka nunatak – that the map implies).

    Where about in the DA were you from?

  5. Ann Willis says:

    Thanks Anne for letting me know about this. I did include it in Accretionary Wedge 29, since it was an area that hadn’t been talked about.

  6. RBH says:

    I also grew up in that general neck of the woods and appreciate this overview of the geology.

  7. troutbum says:

    thanks for the story i am fortunate to live near the driftless in austin. i could never leave the area.

  8. Pete Knapik says:

    Most of the DA lies in Wisconsin. Three local tourism organizations recently sponsored a very informative website,, that offers much information about Crawford and Vernon Counties and about the Kickapoo Valley. The “Driftless” brand is starting to develop. It implies organic, low impact, sustainability, silent sports, great scenery,…. The underlying rock is all limestone and sandstone, very porous. Thousands of springs combine to create hundreds of trout filled cold water streams. Overall the DA has very low population density and makes a great place for an escape from city life.

  9. Jakewobegon says:

    Lived in Vail & now live here. LOVE IT! May not be “the mountains,” but it’s got water, trees, 4 full seasons, cheaper cost of living & few tourists. 🙂

  10. Andy Wickert says:

    Hi Anne, I just bumped across this post. I hope that this comment is not so late that it will be lost in blog-history!

    I don’t know where your figure comes from, but there is a similar one in the article found here:

    Although there was initial speculation that the deeply-incised Paleozoic plateau of the Driftless Area had experienced erosion since before glacial times (and perhaps even represented the standard pre-glacial landscape of the Upper Midwest), the consensus for nearly a century is that it formed during glacial times, likely via large amounts of meltwater discharge. The age of the start of incision is post-classic-Kansan stage (westerly-sourced material found on the east side of the now-incised area in NW Illinois) and before ~790 ka (Paleomag). The story hasn’t all been unraveled yet, but it looks like this landscape is as glacial as the rest, except in a spectacularly different way!

  11. Michael says:

    The evolution of the Driftless Area is now explained in a new paper published in a peer-reviewed journal by Michael Iannicelli. It can be viewed, downloaded and printed for free (at no cost). Go to The Open Geology Journal website and navigate over to volume 4 – year 2010 and scroll down to “Evolution of the Driftless Area and contiguous regions of midwestern USA through Pleistocene periglacial processes”.

    • Andy Wickert says:

      Thanks for the link. I think you’d be hard-pressed to convince those who have worked in the area that it was all significantly glaciated. Certainly some parts of the “driftless area” were covered by early ice, but fragmentary evidence of this has been found. As for the others, the erosional process you seem to be suggesting should result in fragementary interfluves of old till remaining (as is normal in erosional environments – erosion does not bevel anything off nicely). Also, older tills have, in fact, been mapped south of the Wisconsin limits in the Driftless Area, and their material is sourced from the West (rather than the North), not in support of deposition by glaciers crossing the Driftless Area. The orientations of small valleys due to snow drift orientation is interesting though; I have never run across it before.

  12. Michael says:

    You may be able to comprehend the complete picture of the paper if I word it the following way: The erosional history starts with the dissection of the surface such as the photograph in figure 8. The snow dunes or snowbanks that cause the original dissection then also gradually cryopedimentate the width of the valleys so that surface sediment is removed in a widespread manner. My paper & other published papers have documented that both cryopedimentation & cryoplanation can completely bevel widespread surfaces such as the Old Crow Plain in the Yukon of Canada. Since it can do this, it can also easily bevel the “interfluves of till” that you mention as this is evidenced by just looking at 1:24,000 scale topo maps of the area which show flattened tops of the old interfluves in the area. My paper also points out that the DA was positioned very close to & in the lee of continental ice sheets so that snow-bearing winds (generated from the anticyclonic system that hovered the ice sheet) that blew to the SW had repeatedly deposited tons of snow primarily in the DA while the deposition of snow lessened to the SW, so that subsequent snowmelt erosion caused lop-sided erosion in producing the till pattern of : zero till > negligible till > scattered patches of till > concentrated patches of till > widespread & thick deposits of till. This pattern of till is directed to the SW as we go from the DA to the Paleozoic Plateau to the Iowan Erosion Surface and even past that into the Southern Iowan Drift Plain. Till may occur on only those old interfluves that are on the side of the Paleozoic Plateau that’s close to the I.E.S. The beveling is accomplished through solifluction transport which even operates on surfaces with inclinations greater than 0 degrees (see my paper on page 39; see also Hugh French’s book “The Periglacial Environment” on page 156. ////////////////// You also mention “old till” that was sourced from the west, but this is actually only “till-like material” and not till as reported by Knox, Attig and Johnson (1982–“Pre-Wisconsin deposits in the Bridgeport Terrace of the Lower Wisconsin River Valley”). They characterize this material as outwash. I did have this particular information in my original manuscript but the reviewers felt that the incorporation of this information was beyond the scope of the paper.

  13. Michael says:

    I meant to add the year to the citation of Hugh French’s book “The Periglacial Environment” (on page 156) which is 1996.

  14. Michael says:

    Another postscript that i should have added to the above is that the Iowan paha were originally interpreted as remnant interfluves of till (Ruhe et al, 1968). Iannicelli (2000; 2010) slightly modified that interpretation as specifically remnant snowmelt interfluves of till.

  15. Andy Wickert says:

    OK – I’m willing to entertain the idea that the entire area could have been beveled. I would generally think that a transport agent would be necessary in addition to frost-related weathering (with these transport processes generally causing low ridges to be preserved, at least), but I don’t know as much about high-latitude weathering processes. In my experience, complete peneplanation is always an oversimplification.

    I do find it odd that pre-Illinoian tills are reported in the topographically similar portion of SE MN but not in SW WI; that inconsistency is what does make me think that SW WI could have escaped all glaciation. This seems consistent with topographic steering of ice flow via the basins of Lakes Michigan and Superior.

    While I believe that it may be possible that the snow drifts would cause small-scale features to be oriented with them, they are too short-wavelength to control the incision of the main topographic feature of the region, namely, the valley of the Mississippi River. The incision of the Mississippi River and its tributaries through the Paleozoic Plateau is a story of larger forcings.

    As to deposits, you are right: I was thinking of the outwash.

  16. Michael says:

    To address your question of a transport agent, the answer is once again the solifluction or “s.s.s.” meaning solifluction and snowmelt sheetwash and the fact that solifluction can work even on the gentlest slopes of inclinations greater than 0 degrees (French, 1996, pg. 156). //////////////////////////////// Your question of the peculiar situation whereas SE Minnesota retained some of its till while SW Wisconsin did not is answered by the model given in Figure 1 of my paper whereas snow-bearing winds generated by the anticyclonic system hovering over the continental ice sheets were strongest coming off the lip of the glacial front while depositing tons of snow in the lee of the ice front. This is analogous to other geomorphic realms of sand sheets being deposited in the lee of highlands. My paper also gave a modern-day example of this that occurs in Greenland. Anyway, the snow-bearing winds were directed to the SW but lost vigor with increasing distances which means correlative less & less depositions of snow with distance which means less and less erosion with distance in the same direction. Thus, that is why SW Wisconsin didn’t retain its till while SE Minnesota retained its till in increasing amounts as you travel further west within Minnesota. The old model of the Great Lakes channeling the continental ice sheets around the DA would than mean that the DA is a “mega-nunatak” , however, today’s nunataks around the world only exist as very, very small areas of land which makes the “mega-nunatak” hypothesis doubtful. ///////////////////////// My paper does not make the claim that the influence of the snow dunes controlled the incision of the Mississppi River. This long & mighty river has a life of its own in dictating its own destiny. The NW-SE tributaries however, were explained in the paper as being formed & greatly influenced through differential frost-heaving in combination with the snow dunes. Snow dunes have a thermal property which inhibits vertical frost-heaving (no raising) of the ground directly underneath it, while at the same time, the terrain surrounding the snow dune does experience vertical frost-heaving (raising) … from this, we can envision temporary low embankments forming which caused funneling of the snowmelt along the orientation of the snow dune and that orientation was in a NW-SE direction. This explanation was given in Iannicelli (2000) and in my Driftless Area paper .

  17. Andy Wickert says:

    OK – then to continue playing Devil’s advocate, if snow processes stripped seds off of SW Wisconsin, then why is the drainage there not oriented like those in Iowa? I am also skeptical of conclusions based on broad extrapolations of general circulation model results (i.e., anticyclones), because not all of the geological data from the last glacial cycle agree with this and because local winds can be very different than these large-scale patterns.

    I think it would be really interesting if you made a numerical simulation of the thermal evolution of the snow dune fields to test their relationship to weathering – have you or has anyone else thought about it?

    In any case, I don’t really know enough about the very early glacial history of the area to offer any more real thoughts, so thanks for sharing your hypothesis.

  18. Michael says:

    Many drainage networks are oriented NW-SE in SW Wisconsin as pictured in figure 6 of my paper. If you are referring also to the major trunk valleys of SW Wisconsin that weren’t oriented NW-SE, most of these were formed as outwash rivers emanating from the terminal front of the continental glacier, thus, these are glacially-derived and are called proglacial rivers. Also, it should be taken into consideration of any rivers forming during the “marginal” interglacial periods (following immediately after glacial retreat) that were kick-started by any leftover, waning pluvial periods of rainfall. The more typical, full-blown interglacial periods are mainly characterized as warmer & drier times which of course, would have a negative effect on river-creation …. so I believe that leftover, waning pluvial periods would consist of rainfall contrasted by glacial times when pluvial periods consisted of snowfall. //////////////////////////// The anticyclonic wind model is valid because there are a few published papers in peer-reviewed journals by different people that support it. Besides, that wind direction which blew to the SW from off the ancient continental ice sheet can still be seen today by the Polar Easterlies winds blowing to the SW from off today’s polar ice cap above North America. The adverse interglacial winds of the Midwest would have been dominated by dry-winds blowing from out of the northwest which account for barchan sand dunes in the region. ////////////////////////////// Further research such as numerical simulations of snow dunes could be experimented with by other people who feel like fussing with this. It would be an expensive endeavor while such people would have to go to places where weathering caused by permanent or semi-permanent snow dunes is active in today’s world.

  19. Andy Wickert says:

    Thanks for the thoughts!

  20. Michael says:

    You’re very welcome (smiles).

    Links (2)
  1. Pingback: The flat of the land | Highly Allochthonous

  2. Pingback: Geology Links for January 19th, 2011 | The Geology News Blog