By Evan

Last time I posted, I said the paper I had worked on for some time was accepted for publication. A month later, I got an email saying it is published! I’ll admit that this was far faster than I was expecting, as my past experience was such that an accepted paper sat “in press” for many months. It feels pretty good to see your hard work in the nice journal typsetting, along with all the information needed to cite it.

I suppose at this point, it would be good to explain what it is all about!

About 20,000 years ago, most of Canada and some of the northern United States were covered by a massive ice sheet known as the Laurentide ice sheet. In my paper, I am focussing on the western part of it, outlined in red in this image by Dyke et al (2003).

Laurentide Ice Sheet at the height of glaciation
Laurentide Ice Sheet at the height of glaciation

I focused on this region, because their is a lot of uncertainty in the timing of when the ice disappeared from this area (known as retreat). Between 20,000 and about 6000 years ago, the ice in the red outlined region went from a massive sheet of ice that was up to several thousand metres thick, to being completely gone. The ice sheet here contained a lot of water, and it is hypothesized that a lake that was blocked by ice here catastrophically drained into the Arctic Ocean 13,000 years ago, triggering the cold period known as the Younger Dryas. The only problem? Thus far there is limited amounts of information on when the ice went away. And this is the focus of my paper.

Chronlogical information
Chronlogical information

The above figure is a compilation I made of radiocarbon, luminescence and cosmogenic dates that are relevant to the retreat of the ice sheet. Radiocarbon dates measure the amount of radioactive carbon-14 in fossils (the abundance which tells you how old the fossil is). Luminescence dates essentially tell you how long sand grains have been buried. Cosmogenic dates tell how long a boulder has been exposed to bombardment from solar and cosmic radiation. In all, there are many hundreds of dates. But not all dates are equal. In my analysis, I avoided using cosmogenic dates due to high uncertainties on the results. They give a general idea of when the ice sheet retreat happened (the papers dealing with these dates pretty much disproved the so called “ice free corridor” hypothesis – which was the idea that there was a gap between the Laurentide and Cordilleran ice sheets that allowed humans to enter North America), but for the purposes of ice sheet reconstructions, they are not precise enough. Luminescence dates only exist in a few select areas. There are many more radiocarbon dates, but dates on material such as bulk sediment have the potential to be contaminated with sources of “old carbon” (i.e. carbon that has less or no radiocarbon relative to atmospheric concentrations at the time of deposition). To be on the safe side, I didn’t use these in my main analysis.

GMT v4.5.3 [64-bit] Document from psbasemapIn order to assess the timing retreat, I first created a map of the assumed direction of ice sheet retreat (shown as white arrows in the figure above). I used various glacial landforms to determine the pattern. Using the logic that ice sheet retreat was progressive, I extrapolated the chronological samples backwards to create a probability map. An example of this is shown above. This particular examples indicates that most of this glacial lobe (shown in red) had to have been ice free in order for the sample to exist where it is. The sample does not give any information on whether or not there was ice in the blue region.

I’ll now give a couple of examples of the results of my analysis.


The above analysis is for 12,900 years ago, about the start of the aforementioned Younger Dryas. The light blue line is an ice sheet margin reconstruction by Dyke (2004). A number of people have suggested that Lake Agassiz, located in the southern part of the study area, may have drained northward into the Arctic Ocean at this time. However, at present there are no chronological constraints (yellow dots) that date to this time between Lake Agassiz and the Arctic Ocean (see orange hash line). Of course, the absence of evidence does not mean evidence of absence, it does lead to the need for more study in this region. At present, it cannot be confirmed that Lake Agassiz drained into the Arctic until the late Younger Dryas, if not until after, something supported by geomorphic evidence.


Another problem is the lack of data in the area known as The Barrens. There is a famous Canadian book called Lost in the Barrens, a story of two teenagers and their struggle after getting lost in the cold, treeless tundra. The picture that Mowat painted is completely accurate, and this foreboding place makes for a difficult field work area. By 7400 years ago (see the above picture), it is assumed that most of the ice in the western Laurentide ice sheet is gone, but there are no data points in this region until about 1000-1500 years after that. Anyone have $200,000+ for a field work campaign? (I just pulled that number out of the air, but I wouldn’t be surprised if it cost at least that much.)

All in all, the uncertainty in timing of ice sheet retreat is large in many regions, which affects the uncertainty on estimates of ice volume, which is my next goal. I have six months left in my PHD, so it will be a lot of work!