University Place

cc >>

Tom Zinnen

Welcome, everyone, to Wednesday Nite at the Lab. I'm Tom Zinnen. I work here at the UW Madison Biotechnology Center. I also work for UW Extension Cooperative Extension, and on behalf of those folks and our other sponsors, Wisconsin Public Television, Wisconsin Alumni Association, and the Science Alliance, thanks for coming to Wednesday Nite at the Lab. We do this every Wednesday night, 50 times a year. Tonight, I have the pleasure to introduce to you Eric Carson. He's a fluvial geomorphologist with the Wisconsin Geological and Natural History Survey, and he's an assistant professor in the UW Extension Department of Environmental Sciences. His current research interests include Quaternary landscape development and fluvial systems near forming glaciers, uh-oh, near former glacier margins in the upper Midwest and intermontane west. I take that to mean he studies the geology where the glaciers drew back. >> Exactly. >> The impacts of post-glacial climate change and historic land use on sediment mobility, flood behavior, and land from genesis. And applications of Quaternary geochronology to the glacial record of North American mid-continent, which is right here. He received his BS from West Virginia University in 1996, his MS and PhD from UW Madison here in 1998 and 2003, respectively, and was a lecturing professor at San Jacinto College in Houston, Texas, from 2003 to 2008. Please join me in welcoming Eric to Wednesday Nite at the Lab.

APPLAUSE

Tom Zinnen

>>

Eric Carson

Thank you, Tom. Thank you all for showing up. I appreciate the turn out and the opportunity to talk about some of the research that I'm doing here at the Wisconsin Geological Survey. I was hired at the survey four years ago now, started in the summer of 2008, and the Wisconsin Geological Survey has had a long record of Earth surface processes research relating to the glaciers, relating to the glacial record of Wisconsin, which certainly isn't surprising given the extensive record here, but haven't had much of a record with having people with an interest in streams working in the state. And so when I was hired the idea was that I would be turned loose in the southwestern corner of the state, the driftless area, which was never covered by glaciers and allowed to do my own thing and study the geologic processes there. So as a good geologist, I tend to think of Wisconsin looking something like this rather than the way we see it today.

LAUGHTER

Eric Carson

I'll wait and see if we can get a pointer instead of just an hour glass symbol there. Okay, well, regardless, the idea is if we go back in time, maybe 18,000-20,000 years ago, this is the image of what Wisconsin would have looked like. It was covered by the southern margin of the Laurentide Ice Sheet. So the ice sheet that covered all of modern Canada and extended down into northern United States came down, and as it came down into the upper Midwest and into Wisconsin, it divided into several sub-lobes and partitions. And across Wisconsin, you can see they're labeled here, the Superior, Chippewa, Wisconsin Valley lobes, and then down in our neck of the state, we have the Green Bay lobe and the Lake Michigan lobe. And the Green Bay lobe you can see extended right down here into the area of the Baraboo Hills and down to just south and west of Madison. You can also see that it dammed up Glacial Lake Wisconsin. But then you can also see this large area here in the southwestern part of the state, and this is the driftless area. Now, if we pull away the glaciers and look at things as it is today, the driftless area itself still stands out pretty clearly, pretty distinctly in the landforms that we have generated. If we look at it, you can still see the effects of the last glacial cycle on the landscape. So here's the margin and behind it the landscape is entirely derived from the glacial processes that have gone on. Whereas, down in the driftless area, it's primarily stream incision that's happened. Now, in the previous slide we saw what the last glacial cycle looked like with ice coming down primarily from the north and from the northeast, but during the Quaternary period, during the last 2.5 million years or so, there have been numerous cycles of glaciations. And so during this time this driftless area stands out in the Midwest as an area that has escaped glaciation entirely. We will restart later...

INAUDIBLE

Eric Carson

Okay, so we see a landscape that's very distinct in the upper Midwest. We have upland surfaces that are covered by a thick mantel of wind-blown silt that's blown in during the glacial cycles. We have the Lower Wisconsin River Valley that's incised several hundred feet. This is taken from Ferry Bluff just near Sauk City, and it's about 250 feet, I believe, from where this photo was taken down to the river. And then in much of the Lower Wisconsin River Valley, the valley itself was actually dug out another 200-300 feet down and has filled up with sediment. >> You can see the nude beach in the distance there. >> If you want to look that way, you can see that.

LAUGHTER

Eric Carson

And in between these upland surfaces and the deeply incised Lower Wisconsin River Valley, we have a landscape that has a level of topography that's not seen anywhere else in the Midwest. And so if we sort of take a step back and look at the Midwest as a whole, we can see how this area came to be and why it came to be when we look at the record of Quaternary glaciations. So on here we have the three major episodes of glaciations that have occurred during the Quaternary period. So in the gray area here in Iowa down through Missouri we have what we refer to as the pre-Illinoian glaciation. This was a cycle of glaciations that occurred starting about 2.5 million years ago and extending to about 500,000 years ago. Primarily, that ice was centered in western Canada and flowed down from the north and the northwest into the upper Midwest. Then from a period of about 300,000 years ago to 130,000 years ago we had what was referred to as the Illinoian glaciation. So these deposits are shown in gold, and there's our pointer. We can see that they come down and clearly derive their name from being well expressed in the state of Illinois and bound to the driftless area just down here on the south and the southeast. And then most recently we have the Wisconsinan or the Wisconsin glaciation. So this is the entire area that's shaded in green. And this was an ice cap that was centered more in central Canada over towards the Hudson Bay more. So the ice was actually flowing down from the north and the northeast into the region. And so when we see this, we see that the driftless is this one spot that stands out that was never surrounded by ice all at one time but at the different time periods, it has been surrounded and it is the one area that has never been covered by ice. Okay, if we look back at the topographic map, what I'm going to do here is highlight the two areas where I've been primarily working since I've started at the Wisconsin Geological Survey, and these are the two areas that I'm going to be talking about. First, I'll be talking about some work here in Grant County. So the southwestern most county in the state. This is the area where the pre-Illinoian, or oldest, glaciations have budded against the driftless area. So I'll talk about some of the work that has been teasing out how we can learn about the distribution of those deposits. And then back here closer to home in Madison, right along this last glacial maximum margin, work that I've been doing pinning down the exact timing of the glacial record. Okay, so let's start down in Grant County. So up in the left corner here we have a DM of the county. Wisconsin River coming down here, the Mississippi River flowing along here, and the area that's covering the majority of the slide is highlighted here in gold. I guess it's probably about 23 years ago now my PhD adviser, Jim Knox, was doing some work down in this area, put in a core hole right here above the town of Glen Haven. That core location is up on the high bluffs, so right at the very top of the bluffs above the Mississippi River. Much of the landscape here is buried by wind-blown silt. So there's 15, 20, 25, as much as 30 feet of wind-blown silt covering the bedrock. When Jim put in this core, he found, I think my recollection is, right around 28 feet of this wind-blown silt on top the bedrock. And then right down at the base right before he hit bedrock, he hit about two feet worth very dense, dark gray clay that had flecks of minerals that aren't found in the bedrock around here. And what he interpreted this is as one of the pre-Illinoian glacial advances. That this was a remnant of till from one of these pre-Illinoian advances that was coming in from the west from Iowa and Minnesota and actually topping up onto the modern bluffs above the Mississippi River. Not surprisingly, he wasn't able to find it in any other locations. Partially this has to do with the fact that back in the '80s the coring equipment that he had didn't give him the ability to core many locations very effectively and partly because if we're thinking of a glacial deposit that may be a half million or a million years old, there's probably not a very extensive record of it left. Probably most of it has been eroded away. So it might have been just good fortune on his part that he found that deposit in that one location. Now, when I showed up and began working, I had access to a quicker, more portable coring mechanism, and so in the past two years I've put in a whole series or cores here trying to find more of this same deposit. And to give you an idea of how patchy this material can be, here is Jim's original core. He calls this his Adrian site because the Adrian family owns the land. The first core I put in was within a hundred feet of where his core was. And when I put the core in, it went down to bedrock and there wasn't any tilling in my coring location a hundred feet away. So sort of an idea of how patchy this old material could be. So I put in maybe two dozen, two and a half dozen, cores. What you can see here is a very limited area right along the bluff where I found several additional cores that had this old till right at the base. So it's a really good indication that we do have the presence of ice up on this surface. Like I said, the fact that there are the blue squares which are locations that there was no till, the fact that we're finding those intermixed, like I said, isn't at all surprising because the age of this till. One of the other interesting things that I found in the course of this coring are these two locations that are shown in gold, and rather than finding till, what I found in those two locations was thick accumulations of sand and gravel. And that's exactly the kind of thing that you typically see out right in front of a glacier. Glaciers that are in this landscape position, in this part of the world, in this time in the geologic history tend to have a lot flowing around at their margin flushing a lot of water out from underneath them and carrying a lot of sediment, and a lot of that is sand and gravel, that this outwashes, what we refer to it. And it was in some ways unsurprising but in some ways very gratifying to be able to find it in these two locations, further evidence of this glacial advance. So it gives us an idea that we had ice here on the landscape coming up. Now, the Mississippi River at this time probably was not incised as deeply as it is now, probably in somewhat a similar landscape position, but we still have the good evidence of ice coming across what is today the Mississippi and occupying just barely into Wisconsin with outwash coming out. Now, when we compare this to additional work in the driftless area, we see that it's not the only pre-Illinoian deposit that we can find in Grant County and in western Wisconsin. So this blue dashed line is going to be the margin of the glacial deposits that I identified. There are also additional deposits right here in the mouth of the Mississippi River. These are referred to as the Bridgeport Deposits. These were identified by Jim Knox in the geography department and John Attig in my office. It is also a pre-Illinoian ice advance. In the case of the coring that I've done, if I can get my pointer, the coring that I've done in this area, the fact that the deposits are pre-Illinoian are pretty much demonstrated by the fact that they're covered by this thick accumulation of wind-blown silt, 25 or 30 feet of silt that dates well back into the geologic record. So that's what demonstrates the age of these deposits. In the case of the Bridgeport deposits here in the mouth of the Mississippi, there is additional information that gives us a little bit better chronologic control. So this is a cross-section looking along the Wisconsin River. So looking from as far west as about Sauk City, Wisconsin, right near the Baraboo Hills all the way down to the mouth of the Wisconsin River at Prairie du Chien. So we see a number of different surfaces here. This the bedrock surface of the Wisconsin Valley that's buried by several hundred feet of sand and gravel. And we'll once again restart later.

LAUGHTER

Eric Carson

We have the modern river surface shown here in the solid line and then a series of terraces, a series of outwash terraces. So surfaces built up of this sand and gravel that gets washed out of glaciers. So this would have been from the ice that was at the Baraboo Hills carrying sand and gravel down the Wisconsin River Valley and producing these terraces that are mirror images to the modern river topography, just higher surfaces. But then here in the blue we also have these Bridgeport deposits. In the mouth of the river, there is an actual moraine. There's a topographic landform which is exceedingly uncommon for glacial deposits of this age. There is a terrace associated with it, and there is outwash deposits associated with it, and they actually, you notice that these ones that are associated with the modern river grade downward to the west. So they're grading downward in the direction that the water is flowing. This Bridgeport deposit grades in the opposite direction. It grades downward to the east. So right here is the Bridgeport terrace, grades down to the east, and what this tells us is that when ice was at this Bridgeport moraine location it was coming from the west, coming out of Iowa and Minnesota, and was completely blocking the mouth of the Wisconsin River. And, in fact, it was completely reversing the flow of the river. And during this time, the river was flowing back to the east back here towards the Madison/Sauk City area and in some unknown direction flowing off by another root into the Mississippi or perhaps flowing into the Great Lakes basin. It's demonstrated by both the slope of the terrace and also the structure of the sediments within the terrace tell us that they were deposited by water flowing to the east. As far as age control on this, Jim Knox and John Attig took paleomagnetic measurements from the sediment. So they measured the magnetism of the Earth's magnetic field that was preserved in the sediments, and they have what we refer to as a reverse polarity. So these were deposited, these sediments were deposited prior to the last time that the Earth's magnetic field reversed. So that occurred 790,000 years ago. So these deposits of the Bridgeport moraine and Bridgeport terrace are something greater than 790,000 years ago. In order to have a soft sediment landform like this Bridgeport moraine still expressed on the Earth's surface is actually pretty intriguing to us. What this tells us is we had multiple advances of pre-Illinoian ice up against the western margin of modern day Wisconsin. We have some pretty good evidence that those two deposits, the deposits from my coring up on the bluffs and the deposits from the Bridgeport moraine, almost certainly were not the same age. They were probably two completely different pre-Illinoian ice advances. But what we're seeing is at multiple times we had ice coming up and delineating this western margin of the driftless area. Now, if we move to the opposite end of things, if we move to the east edge of the driftless area right here in the Madison/Sauk City/Baraboo Hills area, we're up against the most recent ice margin. We're up against the late Wisconsin ice margin. And this is an area that's intriguing to us, first because the landforms are much more present on the landscape so it should be a much easier proposition to tackle. It's also intriguing to us because despite the fact that the Wisconsin deposits are the most recent, we have very poor age control on the exact timing of events. I told you at the beginning that the Wisconsin deposits are, in general, something less than 30,000 years old, and we as geologists have a ballpark understanding of the timing of the late Wisconsin glaciation, but we don't have a very refined understanding of it. And that's kind of interesting considering that the history of studying glacial deposits here in Wisconsin goes well back into the 19th century with TC Chamberlin and Charles Van Hise here at the university. And there's some very good reasons for it. This quote up here is from a recent survey publication of the glacial deposits of Wisconsin. This is by Kent Syverson and Pat Colgan, and what they know is that the late Wisconsin maximum, the timing of exactly when the ice was fluctuating isn't constrained in Wisconsin because there isn't organic material during that time period. And organic material, of course, is important because radiocarbon dating for a half century now has been our bread and butter tool for producing chronologic control, for developing chronologic control. And so in Wisconsin there simply hasn't been found any significant amount of organic material during this huge window that spans the peak of the Wisconsin glaciation. Now, we can look at this in the larger picture if we look at the radiocarbon data that are available. So this is from a publication that Lee Clayton from our office put together back in 2001. What he's gone through here, he's gone through the published literature and he's pulled out all the radiocarbon ages from large wood. That's important because having large organic material gives us some context to where it was deposited. And he's plotted it up with age on the vertical axis. So going from 12,000 to 15,000 years ago back to over 35,000 years ago. And then on the horizontal axis he's plotted it in its position from north to south. So here's the Illinois/Wisconsin state line, Illinois extending off to the right, Wisconsin extending up to the left. And as we move forward in time from, say, 35,000 years ago, what we see is that the northern limit of organic material keeps receding. It recedes farther and farther south until a peak somewhere in the mid '20s. And then this front of organic material begins advancing north again. Now, the obvious interpretation of this is that this is a reflection of the climate changing and the ice advancing out onto the landscape, and as the landscape begins to get covered with ice, clearly you're not going to have organic material, certainly in the areas that are covered by ice but also there's going to be an additional fringe of land that isn't going to be supporting much vegetation growth. We have quite a bit of geologic evidence that during the late Wisconsin maximum here in the upper Midwest there were permafrost conditions. So the ground was permanently frozen year-round. So this is a kind of condition where you wouldn't expect organic material to be growing. So this makes good intuitive sense to us but it's problematic because it crimps our ability to understand the exact timing of when the glacier fluctuations were occurring. Beyond that, there are a couple issues with using radiocarbon dating in glacial environments. Not only in the upper Midwest with the southern Laurentide Ice Sheet margin but in many of the former ice sheet margins, the permafrost conditions limit the amount of material you're going to have. So there very often isn't a whole lot of radiocarbon material, datable material, that extends back to the maximum ice conditions. In the cases that we do find organic material close to ice margins in glacial settings, there are further complications with interpreting the ages. The first of them is that a glacial environment, a glacial landscape, that's having ice on the landscape, having sediments deposited by the ice and by water in front of the landscape, this is an inherently messy kind of setting. There's a lot of water, a lot of very saturated sediments, a lot of mobility of sediments and mobility of the landscapes in whole. There are a couple cases where geologists have been able to document how long glacial landscapes tend to be mobile after the glaciers recede, and in the published literature there is cases where it's been documented that the landscape has been mobile for hundreds to as much as 2,000-3,000 years after the ice retreated away. Now, this is important to us because what we as geologists are interested in is the time of landform deposition because that's when the glacial action is occurring. What the organic material and what the radiocarbon dates are able to tell us is the timing of landform stabilization. And so if we have a lag of hundreds to thousands of years between the two, it's going to crimp our ability to use radiocarbon. The final one is because we have this lag in stabilization of landscapes, because we have a spatial lag, often, between where we find radiocarbon material and where the ice margin was, it's very rare that we have cases that we find organic material that we can correlate to exact ice margin positions. So we can tell from the geomorphology, from the landscape and the landforms that are posited, we can tell where the ice was we just don't have the ability to pin down an exact time for when the ice was there. What this leads to is sort of some circuitous routes, I guess is what you'd say for coming up with how and when the ice margins were fluctuating. One of the very typical ways of dealing with these issues of radiocarbon dates in glacial environments is try to accumulate them and use them in mass. So if radiocarbon dates can't tell us exactly where an ice margin was, at least they can tell us when a specific location was covered or uncovered by ice. And a great example of that is the Two Creeks forest bed that's up in east central Wisconsin, up just north of Sheboygan. So this is within the geology community, within the geomorphology and certainly radiocarbon community, this is a very significant site. This is actually one of the very first sites that was dated using radiocarbon dating. Two Creek forest bed is a location up in Sheboygan. It was well behind the ice margin when ice was at its maximum, so the location was buried by ice, and following the glacial maximum, ice was melting back and eventually melted back far enough to expose the Sheboygan area and uncover it from ice. And after it was exposed, a spruce forest grew in the location. And once this spruce forest had grown to maturation and was a fully grown forest, the ice re-advanced and physically ran over the forest. Up in Two Creeks you can find places where there are mature tree stumps still in place with the trunk snapped off. You can find entire tree trunks mixed into the till. And so the radiocarbon dates there give us very good control on when that location was uncovered by ice as it melted back and when it was buried again as the ice advanced forward. What it doesn't tell us, though, is how far north the ice retreated to expose it or how far south the ice advanced again. Now, if you take numerous locations like that, numerous locations where you have radiocarbon dating that tells you a specific location was exposed and uncovered by ice, you can start to marshal them and get an idea in space and time what areas were covered and what areas were uncovered by ice. So we express this in the geomorphology world in a diagram like this which is called a time distance diagram. So this has age on the vertical axis, going from old up to young, and then distance along the ice margin. So in this case, we're looking from, say, Green Bay down to Madison. And then the curve is the generalized location of where the ice margin was. Advancing we know that it had to back off to expose Waterloo and Cactus Rock here. We know it had to back off farther to expose Valders and the Sherwood sites. Then there was an advance, there was a retreat to expose the Two Creeks forest. So we can get the general pattern of ice advances and retreats, it still doesn't give us exact ice margin locations and times and that's actually really crucial. It's crucial for us if we're going to try to model the dynamics of the Laurentide Ice Sheet, if we're going to understand how fast and what mechanisms lead to its melting. It's crucial if we're going to model climate, if we want to have firm starting points and firm midpoints for climate change as indicated by exact ice margin positions in space and time. It's important for, say, biogeography. If you're interested in the migration of plant species in the landscape, it certainly helps if you can know when advances and retreats were occurring and in a very detailed sense. So refining beyond this has long been an interest within the geomorphology community. If radiocarbon isn't a possibility, if radiocarbon dating is simply compromised in the upper Midwest during the last glacial maximum, what we need to be interested in is other methods in dating. And about the time that I showed up at the Wisconsin Geological Survey, a few of my colleagues and now collaborators had began exploring the utility of what's referred to as optically stimulated luminescence dating, or OSL dating. And OSL dating is certainly very attractive in this kind of environment because OSL dating, rather than dating organic material, the life and death of organic material like radiocarbon does, OSL dating dates the amount of time that a sand grain has been buried away from the sun. And so certainly and environment like this an any glacial environment, certainly in Wisconsin we have plenty of sand around. And so if you can measure the time that these sand grains have been buried, then that's going to get around the problem of not having organic material. Now, the fundamentals of OSL dating are up here. I'll run through it really quickly. The idea is that mineral grains, particularly quartz grains, tend to accumulate electrons in their crystal lattice that are released by the decay of surrounding radioactive elements. Now, in any sedimentary package, you have a sand bar on a river, there's going to be quartz grains, there's going to be grains of other mineralogies that are going to have radioactive elements. Thorium, potassium, and uranium are the most common, and so as a sand grain that's buried in a sand bar, buried in a dune or wherever, it's going to be surrounded by these radioactive elements and the longer it's buried, the more it's going to accumulate these electrons. And so if we measure the amount of electrons that have been accumulated in a sand grain during the time since it was last buried, it's going to give us an indication of how long it has been buried. The measurement of that accumulation of energy is actually the optically stimulated part of it. Optically stimulating is what they do to the sand grain in the laboratory setting. We need to know a couple things. We need to know the dose rate, so that's basically how much uranium, potassium, and thorium are in the surrounding sediment. So that can be relatively easily measured with a salinometer. We also have to have to have some confidence, and this is an important one for our research, we have to have some confidence that this sand grain has been what we called zeroed out, and what that means is that all the electrons that accumulated in it were flushed out of it prior to its most recent burial. What we know is that short exposure to direct sunlight, something on the order of 15 to 20 seconds of exposure of a sand grain to direct sunlight, will flush its crystal lattice clean. So this is interesting if you think of the life cycle of a sand grain that it can be buried and exhumed and transported and buried and exhumed and transported multiple times over. And as long as the last time it was transported it was exposed to enough sunlight to flush out all the previously accumulated electrons, then it's going to give us a good date for how long it was buried that last time. Now, within the OSL dating community, OSL is a relatively new methodology. It was originally conceived and developed 35 or 40 years ago. It's been about the last two decades that it's been in common use, and the way to deal with this has been traditionally to work in geologic environments where you can intuitively demonstrate that the sand grains were transported far enough to be flushed clean. The way that's been done has primarily been working in areas where the sand was transported in the air column, transported wind-blown. So, intuitively, you can see if a sand grain was carried up in the air column it was going to be exposed to enough sunlight. What we're doing is starting to press the boundaries of this, starting to work in different geologic environments and address whether or not OSL dating can be used and address whether we can verify whether the sediments that we're going to be interested in have been flushed clean and are going to be viable dating methods. As a general rule, OSL dating is applicable and time ranges from about 300 years as a minimum, that's how long it needs to be buried to begin accumulating a measurable amount of electrons, up to about 100,000 years. So that's when the sand grain becomes saturated with electrons and can't accumulate any more. So that's the dating range. That fits perfectly into the late Wisconsin time frame. So it's the ideal tool in the respect for using. The methodology, the location and the setting that we're applying this, is in sand that was deposited in lakes right on the ice margin. So this is a different setting from the wind-blown environments that have typically been used, but it's got some great benefits for us with dealing with the Wisconsin geologic record. The first application of it, this is work that was started right when I showed up, we had this published last year in the journal Geomorphology, was using OSL dating in two small lake basins right here in the Baraboo Hills. And so these are both up high on the Baraboo Hills. They're not down in Devil's Lake. They're actually up high on top the quartzite. Let's zoom in on that. So here we have the Baraboo Hills. The red line on here is the ice margin. So everything to the east of that was covered with ice, and then we zoom in and the ice is sort of shaded in white in the larger image. And you can see that very convoluted ice margin in the Devil's Lake area. So ice coming in, blocking the Devil's Lake gorge to the north, blocking it to the south and looping around, and there was this area in here up on top the Baraboo Hills and this area over here on top the Baraboo Hills that were ice free. And in these two locations, Feltz Basin right here with this large lake shown and South Bluff Basin here with this smaller lake shown, there were two lakes that were formed precisely because the ice was there. These were small valleys up on the quartzite that normally would drain down into the lowlands down to the Wisconsin River, but during last glacial maximum time, those drainages were blocked by the presence of ice. And so lakes formed and those lakes only existed while ice was at its exact maximum. As soon as the ice started thinning and retreating back to the east, those lakes drained and no more sediment accumulated in it. So the idea for us is that this is a perfect environment where if we can derive reliable OSL dates from these, these dates can be unambiguously and entirely solidly correlated to the glacial event that happened to the ice margin position. And so with numerous dates from the Feltz Basin area, numerous dates from the South Bluff area, all of them converged to an average age of 18,500 years ago. Certainly this was the first case published of OSL dating in a glacial environment. Certainly it was one of the very first numeric controls on the timing of ice fluctuations in Wisconsin. So it was certainly significant work in both those regards. >> What does KA stand for? >> KA is thousands of years ago. So 18.6 KA is 18,600 years ago. Sorry. When we work with dating techniques in the geologic realm,

we're very often concerned with two issues

accuracy and precision. So accuracy is how well does the date we have correlate to the geologic event? And the precision is how tightly does it constrain it? How much error is involved in it? And one of the failings that we have with OSL dating right now is that the, certainly with this work we weren't able to test the accuracy portion of it. It was unsupported by other dating techniques, which was a worry to us but something we weren't able to address. The second is that the precision is relatively poor for OSL dating. If we had, with this date here of 18,600 from the South Bluff Basin, if we had radiocarbon date of 18,600 year, the lab error on that would probably be, today with AMS dating, would probably be 18,600 plus or minus 60 years. That's how precise radiocarbon dating is. With OSL dating, currently the errors associated with it are plus or minus 10%. So 18,600 plus or minus 1800 or 1900 years. >> What is OSL dating? >> OSL is dating how long the sand grain has been buried, how long ago it was deposited. >> What is OSL stand for? >> Optically Stimulated Luminescence. >> Thank you. >> You're welcome. When we're looking at this in the big picture, we wanted to address both of these. We wanted to try to address the accuracy issue to determine whether OSL dating can be verified to have reasonable accuracy and hopefully through laboratory techniques deal with the precision as well. And since we found several lake basins right here in the lower Wisconsin River Valley, this area is actually quite fertile in having lake basins associated with the ice margin position. So from previous mapping that had been done by my colleagues at the Geological Survey, we're aware of a whole host of small lake basins that exist on the landscape. We have several of them up here that are part of Glacial Lake Wisconsin. We have a whole series of them that are out in front of the maximum ice margin position down along the Lower Wisconsin River. There's a few extra lake basins that are behind the maximum position. So we have lakes in a variety of settings with a variety of different mechanisms forming them that we can continue to test the OSL dating mechanism. That's what we've been doing the last year or two. And for the rest of this talk I'm going to focus in on these two locations. So this is Black Earth Creek flowing from the Middleton area down into the Lower Wisconsin River, and this northwestern site I'm going to refer to as Marsh Valley. It is in Marsh Valley. And then this southeastern site I'll refer to as the Swamp Lovers site or Swamp Lovers Valley. It's in an unnamed tributary that much of the land is an LLC by the landowners. And they love swamps so they named it the Swamp Lovers LLC, that owns their land. So Swamp Lovers site or Swamp Lovers Valley is what I'll be referring to. And what I'm going to do is you're going to be seeing that map area there quite a few times. You'll be seeing it with geologic maps, you'll be seeing it zoomed in with topographic and digital elevation models. So kind of keep that in mind of where we are in the bigger picture with the Wisconsin River here in the northwest corner and the late Wisconsin maximum here on the eastern boundaries of it with Black Earth Creek cutting across the sort of central portion of that area. So let's take a look at this from the geologic perspective. One of the primary tasks that our office works on is creation of geologic maps. So this is the surficial geology map that was created for Dane County. So this shows all the surficial deposits in the county, mapped by two of my colleagues. And this is the area that we are looking at. And just to give you sort of a big picture idea, basically everything in the greens are the glacial deposits and everything in the washed out sort of whitish buff color with the oranges and pinks, those are stream deposits and the deposits in the eastern most driftless area. So we're seeing right up against this margin of the glaciated train against the driftless area. So let's zoom in on here and talk about the Marsh Valley site and the Swamp Lovers site. Here's Marsh Valley right here. During their mapping they actually identified this as a location that would have had a lake in it. There's not a lake currently, but these are lake deposits that are shown in, oops, excuse me, that are shown in blue. Here in the Swamp Lovers Valley, they identified that there were similar deposits just probably not extensive enough for them to map. So they knew that there were lakes there, and the idea of why there were lakes in these two streams is because both of them drain into streams that carried outwash, carried that sand and gravel being flushed out from the glaciers. So Marsh Valley drains directly into the Wisconsin River, Swamp Lovers Valley drains down into Black Earth Creek, and you need to imagine that during glacier times this outwash was coming down Black Earth Creek, it was coming down the Wisconsin River, and it was actually filling up and choking those valleys. They were what we call aggrading. They were filling up with sand, and as they were filling up, it was damming up Marsh Valley, it was damming up Swamp Lovers Valley, it was damming up, this was Dunlap Creek up here. It did the same thing to it. We focused in on these two for a very specific reason, though, because not only were they dammed up like that by the outwash on the main stream, but they were also in a position where the ice at its very maximum position, at its very maximum extent during the late Wisconsin, got just barely into their headwaters. So we can see the green off on the eastern part of the map area. And here we have the headwaters of the Swamp Lover Valley. Here we have the headwaters of the Marsh Valley, and the ice extended no more than a mile or a mile and a half into those valleys. So it was only when the ice was right at its maximum extent did it get into the headwaters of those valleys. Why is that important? It's important if we think of the processes that were going on in those two valleys. And so this next set of maps that I'm going to show are a conceptualization of how sediment was accumulating in these two valleys. And these actually date back to the time when John Attig and Lee Clayton mapped this area. This is how they conceived that sediment would have been accumulated. So here we have that same map area I've been showing. Wisconsin River up here, Black Earth Creek flowing off to the west, and this is a time slice taken while ice was advancing to its maximum. So we can see the ice up here in the northeast corner of the map, and we had outwash coming down both Black Earth Creek and down the Wisconsin River damming up lakes in our two sites. So a lake here and a lake here. And then if we step forward in time, we get to when ice was at its maximum and it was in the headwaters of Marsh Valley and Swamp Lovers Valley. And because ice was in the headwaters of those valleys, water was coming down and outwash was coming down those valleys. And the outwash was actually being deposited on top of those lake sediments. And then as soon as ice began to retreat, as soon as it backed off no more than a mile or so, it got out of Marsh Valley, it got out of Swamp Lovers Valley, and the supply of outwash down these valleys was cut off. But there was still outwash coming down Black Earth Creek, there was still outwash coming down the Wisconsin River, so they continued to aggrade. They continued to fill up with outwash. And so once ice began to retreat, lakes were reestablished in those two valleys. Now, that's the conception that they had from doing their mapping, from understanding the geologic processes. They never had means or reason to go out and test it by coring. Step forward 20 years in time to now when I'm working with the Survey and we're interested in this OSL dating, all the sudden there is a reason to do it. And so out we went to go and core. This is us with what's called a geoprobe coring equipment. The track vehicle on the left. It's a direct push system from with hydraulic pressure it pushes a core barrel down into the Earth and retrieves about an inch and a half diameter core barrel. Really nice, really quick coring unit that we have access to. You can see here we're certainly in the dead of winter, and this was previous winter when we actually had a winter and had snow and that kind of thing.

LAUGHTER

we're very often concerned with two issues

This is the Swamp Lovers site. We had to do it in the winter. We have a short window at the tail end of the summer to do this kind of coring in the sites because the ground is either frozen or that tiny window when it's dry. Otherwise, this is a wet swamp that we couldn't get this coring rig out on to. So we took advantage of the cold weather back in February of 2010, put in some cores to verify what the stratigraphy was like, to see if it was like what had been conceived there and if it was going to be a good setting for radiocarbon dating. Collected cores, I want to tell you a little bit about what we found in them before I actually show you the stratigraphy, show you the layers of sediment in them. We were pleasantly surprised at the volume of material, what we refer to as paleoenvironmental proxies, so indicators of what the conditions were like. We found abundant and diverse assemblages of pollen, of phytolith, which are plant remains, and also of diatoms, which are algae that lived in the water column. That's what's shown on the far side of the image. Those are freshwater diatoms. All of those are very nice for us because they give us indications about what the climate, the water conditions, the water temperature, the salinity. So they started filling in and flushing out the story of what the conditions were like while this lake was in existence. What we also found that really shocked us and put big smiles on our face for weeks after finding it was an incredibly rich record of organic material in this lake. What is truly and honest to god, I know we as scientists love to inflate the value of our work...

LAUGHTER

we're very often concerned with two issues

Truly and honest to god, an unparalleled record of organic material for radiocarbon dating in the upper Midwest. From top to bottom in this lake in these cores we found plant macrofossils, identifiable pieces of aquatic vegetation. We found both terrestrial and marine gastropods, snail shells that were living either in the freshwater column or living on land and then getting washed in. And we found them from the top of our core all the way down to within a few centimeters of the base of the core. So stunning find in and of itself to have all that. So let's look at the stratigraphy. So this is a column showing what the layering of the sediment looked like. Like I said, it is a marsh today in the Swamp Lovers site. So we found several meters of marsh sediment, organic, rich mats of vegetation and fine grained sediment mixed together. Beneath that we found right around four meters of laminated, layered, silty clay. Exactly what we'd expect from a lake. So lake sediment. And on top of that we found outwash, sand and gravel, beneath that we found yet more lake sediment. So we found exactly what we'd expect to find based on the understanding of the setting from the mapping. This throws us for a loop when we actually verify the things we think.

LAUGHTER

we're very often concerned with two issues

But we're going to go with it. In the very small red dots up here in this upper lake sediment, these are four of the organic samples that I've had radiocarbon dated. So you can see that they go entirely through that lake section from right at the top where it transitioned into a marsh all the way down to the bottom. So this is the point right here where we transition from sand and pebbles back into a lake when ice was retreating. So all the sudden, literally for the first time in the upper Midwest, we have a location where we can easily identify what was going on as far as glacial processes and all the sudden we have radiocarbon material to give us a control. And this is over and above everything we were originally looking for which was whether we'd have good sediment for OSL dating, which, sure enough, we have. So we have a sample here for OSL Dating, two samples from here at the bottom in the outwash for OSL dating. This is ideal for us if we're interested in accuracy and precision. All the sudden we get the precision of the radiocarbon date, we also have the cross-verification of the radiocarbon and the OSL dating to tell us whether or not the accuracy is there. So this is perfectly thrilling for us. If we look in a little bit bigger setting here, this is a cross-section going up the length of the valley from northeast to southwest. This is what we found from multiple coring locations. So here's a location up at the head of the valley that went through a similar stratigraphy. Here's our core shown on the right that went down through the outwash and into lake sediment. And we actually weren't able to retrieve samples, but we penetrated all the way down to the bedrock at the bottom of the valley. And then down at the southwest end, there's outwash that was piling up in the Black Earth Creek. And that visually has a different characteristic to it. Much larger material, different mineralogy in it. So visually recognizable as being different outwash than this that was coming down the valley. And then although it's subtle on this diagram, we've got the flat surface of this marsh, this former lake basin, and right down at the mouth where it empties into Black Earth Creek there's a little rise here where the outwash at Black Earth Creek had piled up, and that's actually there. You can drive out on highway 14 and just on the east side of Cross Plains, you can actually see that. There's a house on it, and you can actually see that it drops down a couple feet into this Swamp Lovers Valley. So exactly identical to what we had expected. If we look at our dates, we have a series of dates. So I'll go through these from top to bottom. With radiocarbon dates we have an age of about 7700 calendar years BP when this lake converted over to a swamp. We have at the other end of the lake section we have dates that are right around 17,100 calendar years for when the lake was established. So this lake was established then. That gives us good radiocarbon control on when the ice was retreating. It also tells us that this lake was in place on the landscape for something on the order of 10,000 years before it rolled over to being a marsh. The OSL dates we have, and these dates you can see are converted to calendars. You can see how tight the error is. This has an error of, what is that? 170 plus 450, so about 600 years is the total error on this laboratorian conversion to calendar years. The OSL date, this is 21,400 years plus or minus 3300 years. So a much larger spread consistent with typically what we find with OSL dating and particularly what we're finding with OSL dating in lake settings. Down here we have an age centered on 23,000, 23,500 years BP and 26,500 years BP. Because of the errors, these three dates are actually statistically indistinguishable from one another. We can't separate them, but what we can say are a couple things. We can say with confidence that the ice was at its absolute maximum position probably for a relatively short amount of time centered around 26,000 to 21,000 years BP. And we can also say that all of the dates, both radiocarbon and OSL, show complete internal consistency. As you get lower, you keep getting older and older ages which is exactly what we want to see. This is wonderful for us for our confidence for using OSL dating. It's wonderful for our empirical knowledge of glacial processes to now have these radiocarbon ages. If we look at this setting, we now have an age for 17.1 KA, 17,100 years ago, when a lake had been established here. We can go to the Marsh Valley site. So this is a view of it. This is just an absolutely gorgeous little lake setting just east of Mazo. So nice, flat lake plain setting that is now being cultivated. If you're out there in this little lake basin, you can actually look around it and all the way around it you can see a nice little break-in slope that was the former beach surface. So you can still see the morphology of this lake. This lake was actually in existence for quite a bit longer. It was established by about 20,100 calendar years ago based on, again, abundant radiocarbon dating from within it. And it existed as a lake until about 1500 years ago. So this lake was in place on the landscape for something on the order of 18,500 years. Okay, if we take the combination of all these, if we take the Baraboo OSL ages, if we take the OSL ages from the Swamp Lovers site, and the radiocarbon ages from the Swamp Lovers site and the Marsh Valley site, we can plot them up on here and see what kind of progress we're making of getting into that window of no age control during the late Wisconsin maximum. And this is what we get. So here's our radiocarbon age from Swamp Lovers. The, oops, excuse me, radiocarbon dates from Marsh Valley are the two lower red dots. This is the average OSL date from Swamp Lovers, and this is the OSL age from Baraboo. Put them all together, they actually have, as a group of three different locations that are under similar conditions, they have very nice consistency with the ages. They're going into this wilderness that we've had existing for a century now of no age control. So this is something that is simply thrilling for us as geologists working in this state. Now, if we pull back up a step to wrap up here, and I showed all these different locations, this has us really excited, excited on a number of fronts. In retrospect, looking back, it's not all that surprising that we didn't find any radiocarbon datable material here in these two sites. These were up extremely high on the landscape, would have been incredible hostile environments during the Wisconsin maximum. They were also up on the Baraboo quartzite, which in general is a tough place for stuff to grow because that bedrock is so inert. So it's not surprising to not have found organic material here. To have found such an abundance of organic material from here in Swamp Lovers and here in Marsh Valley really gives us confidence as we start coring into all these other lake basins that are shown here that they're going to have a high probability of also having radiocarbon material. So we're going to build what is truly, it's not world class because we're only talking North America here, but truly a continental record or glacial events that has no parallel as far as its chronologic control. The nice thing is when we look at these, these are all tied into different glacial events as well. So not only can we date that last glacial maximum that I've been talking about, but there's a series of events that we can date as well. I mentioned these three lake basins up here are part of Glacial Lake Wisconsin. So mapping and previous research into Glacial Lake Wisconsin indicates to us that it probably existed multiple times going back prior to the last glacial maximum. So OSL dating in those basins could give us information about ice advances prior to the last glacial maximum. We have a number of different lake basins here in the Sauk City area, right near Badger Ammunition Plant. There are related to Johnstown position, they're also related to recessional Elderon position because these were dammed up by outwash deposited by the Johnstown ice, and then they drained when ice was receded to the Elderon position. So they formed because of the Johnstown position, they drained because of the Elderon position. We have this lake here in the East Baraboo Basin that was only dammed up when ice was at this recessional position that we call the Elderon phase. We have lake over here. This is Glacial Lake Middleton. It only existed when ice was at this recessional phase that we call the Middleton phase. So we have a whole range of different timing questions and chronology questions that we can address with this. Really high level of confidence, really high level of excitement on it. By the time we're done with this, particularly if we get the amount of radiocarbon material that we're expecting, we'll have fully confirmed the viability of OSL dating in these environments. And so moving forward from that, we can actually expand back out and start looking across the entire margin of the southern Laurentide Ice Sheet and begin dating large-scale events. And with that, I'd like to thank you, and I'd be happy to take any questions.

APPLAUSE