There’s a new paper out in Quaternary Research by Liu et al. entitled “Vegetation history in central Kentucky and Tennessee (USA) during the last glacial and deglacial periods.” This is an interesting paper for a number of reasons, the first is that it’s a (relatively) old record. I’m someone who has lived their whole life above the maximum extent of the last glacial, so it’s difficult for me to imagine landscapes without knowing that they’ve been glaciated at some point in their recent history.
Second, this is a paper that deals with no-analogue vegetation (based on pollen records) that shows the peak non-analogue signal predates northern signals by ~2000 years. It’s hard to know what happens after 13kyr, which is when no-analogues further north begin to peak. One potential mechanism for the development of no-analogue vegetation is the loss of megaherbivores that would have had a mega-impact on vegetation, particularly deciduous trees, through browsing. These results show that non-analogue communities are a time transgressive phenomenon, at least with relation to other sites in the region. If that’s the case, then it might imply that climate is a stronger forcing for non-analogue communities. Unfortunately there was very little Sporormiella reported in the cores, so it is difficult to know the extent to which megafauna occupied this landscape. The other issue is that the density of sites in this region is low enough that it is difficult to know whether this is really a time-transgressive signal, or whether we’re inappropriately interpolating local signals over a large space because there’s a lack of data in this region.
That last point speaks to the third reason this paper is great. We don’t really know much about this region during the last-glacial period. You can see in Figure 2 that if we exclude the two sites from Liu et al. there’s essentially nothing south of central Indiana on the western side of the Appalachians. In part this is a function of the availability of sites, but the history of palynology in the US likely plays a role as well. My advisor used to tell me that you could tell where people liked to vacation by looking at where pollen samples in BC came from (see Figure 1 in Goring et al., 2009 for example). I suppose in this case it’s more a matter of “you can tell where the palynologists come from”. People cored close to home.
The final reason I’m interested in this paper is entirely self-serving. Our sedimentation paper from last year defined a set of mean deposition rates (cm/yr) for eastern North America, following up on work by Webb and Webb in 1988, but with the added intent of setting reasonable and empirically estimated priors for Bacon. It’s a tricky paper in a sense. Maarten Blaauw has already re-set the default priors in Bacon based on our estimates, and I think there’s some interesting patterns in the sedimentation rates (and all the code & data are included as a supplement), but given that, there’s not a reason to explicitly cite the paper (but you should cite it!).
That’s why what Liu et al have done is pretty cool. They use the probability distributions of sedimentation rates we estimate in the Holocene to determine the likelihood that there are hiatuses in the record. When they find deposition rates that are far too slow (or sedimentation rates that a far too high) then they start to check for hiatuses. Neat.
Anyway, it’s a good paper, with some cool vegetation dynamics going on. I’d be interested in your thoughts, feel free to comment!