search

PalEON-Project / stepps-calibration

STEPPS pollen-veg calibration model code and paper.
0 stars 1 forks source link

Representativeness of sites #4

Closed andydawson closed 9 years ago

andydawson commented 9 years ago

Lake Ocheda (high maple) and Tamarack Creek (high tamarack). Review pubs associated with these sites to determine if they are representative. Potentially use adjacent samples if spike appears only in pre-set sample?

andydawson commented 9 years ago
Lake Ocheda

Can not find any pubs associated with this data, emailed Eric and Simon.

Eric wrote that he suspects this may be a typo since Acer rubrum does not occur in SW MN where the site is. He emailed Ed and Barb to see if they still have the original data sheets.

Update: Barb did have the original data sheets and the count is correct (according to the data sheets). Both Barb and Eric suspect that what was id'ed as Acer rubrum is really Acer negundo, which grows in swampy areas in SW MN. They note that this count was early in her career, and that there is a strong possibility than an Acer anther got into the sample resulting in an inflated count.

Note that this elevated Acer count only occurs in the sample we identified as pre-settlement. Eric suggested setting Acer count to 0. This is what we've done for now, but we'll need to discuss the appropriateness of this data fix :) .

Tamarack Creek

At Tamarack Creek, the relative proportion of Tamarack began to rise at a sample depth of about 80 cm, and this rising trend continued until the most recent sample (core top). This is the only site we have that has a high proportion of Tamarack.

This site is in what is know as the Driftless Area in WI, which was unglaciated. The Driftless Area is part of the Upper Mississippi River Watershed, and consists of rugged and steep topography. In Davis et al. (1979), they mention the difficulties associated with interpreting pollen spectra from areas of moderate to high relief, in part because it is difficult to determine the source area. I think this site is representative of the Tamarack Bogs of the Driftless Area, but I don't think Tamarack Bogs are the dominant ecosystem type in this region. In Knoot et al. (2015), they look at the relative change in composition for the Driftless Area using PLS and FIA data. They show that there is a change from Oak-dominated forests to Other Species, but they don't mention Tamarack at all in the article, and I think when they talk about Other species they are talking about Hardwoods (a bit unclear).

Also, we do not have any other pollen samples near this site - in fact, there is a large radius around this site for which there are no pollen samples.

Let's talk about this at our next group discussion.

andydawson commented 9 years ago
Lake Ocheda

During the June 2015 meeting, we discussed two options:

  1. setting Acer to 0 in the pre-settlement sample or
  2. using the sample immediately below the selected pre-settlement sample in place of the pre-settlement sample. The current pre-set sample is at 36 cm, while the sample immediately below is at 40 cm, so I think this would be fine.

For now, I am setting Acer to 0 in the current pre-sett sample.

Tamarack Creek

Again, during the June 2015 meeting we discussed setting Tamarack to 0 at this site, and this is what I have done. We also noted it may be interesting to assess the effects of setting this to zero, and/or removing the site altogether at a later time. Will revisit this idea at some time in the future.

Note that both of these decisions will be explained in the manuscript.

paciorek commented 9 years ago

what was the proportion of tamarack in that sample? Just curious how much other pollen is in there.

On Thu, Jun 11, 2015 at 8:17 AM, Andria Dawson notifications@github.com wrote:

Lake Ocheda

During the June 2015 meeting, we discussed two options:

  1. setting Acer to 0 in the pre-settlement sample or
  2. using the sample immediately below the selected pre-settlement sample in place of the pre-settlement sample. The current pre-set sample is at 36 cm, while the sample immediately below is at 40 cm, so I think this would be fine.

For now, I am setting Acer to 0 in the current pre-sett sample.

Tamarack Creek

Again, during the June 2015 meeting we discussed setting Tamarack to 0 at this site, and this is what I have done. We also noted it may be interesting to assess the effects of setting this to zero, and/or removing the site altogether at a later time. Will revisit this idea at some time in the future.

Note that both of these decisions will be explained in the manuscript.

— Reply to this email directly or view it on GitHub https://github.com/PalEON-Project/stepps-calibration/issues/4#issuecomment-111169733 .

andydawson commented 9 years ago

For Tamarack Creek, there 40 % of the arboreal pollen is tamarack.

andydawson commented 9 years ago

Have modified the data as indicated above, and explained these modifications in the manuscript.