Introduction
Fossilized pollen grains can be used to understand the terrestrial past. It can be very useful when near complete records are preserved in bogs or lakes. Percentages of pollen can change between stratigraphic levels, allowing us to look into the past at the types of vegetation that were present and compare it with today.
Margret Davis, in seeking to expand information contained in pollen percentage diagrams, estimates rates of deposition of pollen in the sediment. Changes in deposition rates infer changes in species abundances of plants better than pollen percentages. She compares these rates as they change from level to level. Her results allow for interpretations of the past and have expanded previous interpretations based on pollen percentages; giving us good insights to the historic climate of Rogers Lake in Southern Connecticut.
Rogers Lake
The lake is 6 km east of the Connecticut River. It is a damned lake, surrounded by houses, and there are probably a lot more now. What used to be farm land is now deciduous forest, all second growth, made up by oaks, red maple, hickory, birch, ash, pine, beech, elm and sugar maple. ( in order of greatest to least in abundance)
Methods
Two cores of sediment were collected from Rogers Lake, one in the North basin and one in the South. Samples were taken from cores using the spatula technique where sediment was packed into a 1-ml porcelain spatula. Supposedly this technique is pretty accurate and sample error is of the same magnitude as error in pollen determination . The use of digital cameras would probably aid in narrowing down that error.
The spatula samples were treated with KOH and acetolysis. Assays were made by using the aliquot slide method. They used Eucalyptus pollen as the control-which apparently is not present in the lake. Counts of the aliquot slides provided estimation of pollen concentration and calculations for percentages. 1/3 of pollen grains could not be determined because they were torn or distorted.
Results/Discussion
Pollen percentage: percentage of any given type divided by total grains in sample. These are plotted against depth or age in pollen percent diagram.
Sediment matrix accumulation rate: the net thickness of sediment accumulated per unit of time or amount of time per unit of thickness of sediment, after compaction and diagenesis .(fig 2) slope equals rate of accumulation. The deepest sample was 14,300 years old, showing a rapid rate of sediment accumulation.
Problems: radiocarbon C14 occurs in atmosphere so testing samples older than 2,500 years are shown to be younger than the true ages. This can cause changes in the slope and alter results. Davis reports that the changes are small and are not significant.
The second problem stems from interpretation. Some older accumulation rates were used for comparison that were incomplete. The size of margins are not predictable and can fluctuate based on the site, however Davis assures us that the margin or error is small and should be considered normal.
The second problem stems from interpretation. Some older accumulation rates were used for comparison that were incomplete. The size of margins are not predictable and can fluctuate based on the site, however Davis assures us that the margin or error is small and should be considered normal.
Pollen concentration: the number of grains per unit volume of wet sediment. AKA absolute pollen frequency (APF) fig 3. Pollen concentration is plotted against radiocarbon age, not corrected for C14 in atmosphere.
Pollen disposition/accumulation rate or pollen influx: net number of grains accumulated per unit of area of sediment surface per unit of time. To get this Davis multiplies pollen concentrations by the sediment accumulation rate. (fig 3) The total influx of pollen increased between 12,000-9,000 years ago from 1,000 to 50,000 grains/cm2 per year, then dropping to 20,000 and eventually increasing again. Because there is little for Davis to compare her results to, she cannot effectively say why these changes are occurring.
Pollen stratigraphy and interpretation
Davis separates different zones based on stratigraphy and carbon dating. I will try to be as brief as possible for this section.
The first and oldest zone is the Herb pollen zone (14,300-12,150 years ago) where the oldest and siltiest sediments had the highest percentage of herbaceous plants and grasses with few pollen from trees.
The next zone is A-1: transition from herbs to spruce(11,700-12,150 years ago) maximum of birch pollen and rising percentages of spruce
Zone A-2-3: spruce oak zone (10,200- 11,700 years ago) high percentage of spruce pollen and by a late-glacial percent max for oak, hornbeam and ash.
Zone A-4: spruce fir zone (9,100-10,200 years ago) decrease in oak, and max for spruce, fir and larch frequencies.
Zone B: pine zone (7,900-8,100 years B.P.) high in white pine frequencies, over 50% of pollen is pine, a large change in the trees may indicate change in climate, but much of this area has not been sampled thoroughly.
Zones C 1,2,&3 oak zones (7,900- present)
Conclusions
Several things to be noted.
- The sudden increase in pollen deposition rates for trees between 14,000 and 9,000 years ago. This fits with the idea that as tundra receded, it was replaced by woodland, and then forests as trees increased in frequency.
- There was a large increase in Oak pollen 11,500 years ago. This could be due to an overall increase in oak abundance or it could be due to a change in the wind or weather.
- 8,500-9,500 years ago there was a large influx in pine pollen, as boreal pollen decreased. This signifies a large change in plant species as the forest is becoming more deciduous. This large change could be due to climate change.
- C-1a shows an increase of ragweed 7-8,000 years ago, possibly due to a prairie period that was hot and dry (xerothermic or zerothermic?). This could have been overlooked because of the large focus on tree pollen and not herbaceous plants.
- The changes in pollen frequency that were thought to represent the same zerothermic period in S. New England were found in younger sediment (2,000-4,500 years BP). Only over the last 2,000 years have the pollen in S. Connecticut been similar to modern assemblages. This is evidence that may suggest that these forests are of very recent origin.
Qs:
What were some previous studies that could have influenced this study?
Has this paper influenced any recent studies?
What does m Davis mean when she states that these forests might be of recent origin?
How glad are you that you do not study pollen?
This was a cool study with a lot of parallels to VonPost’s work. It seems like the author was fairly conservative with the radiocarbon dating ages and in making assumptions based on the changes observed in pollen counts. It’s really cool that she is able to make inferences about the climate of the recent past based on this technique, but as stated above she does a good job of finding other factors that could have also led to changes in the pollen record besides changes in vegetation associated with climate. The ragweed case was illustrative of this where she seemed to feel the pollen could have blown in from the Midwest, or it could indicate a transition period in the Rodgers lake area where this weedy species was present. I also liked how she made an effort to confirm or question her findings with other local and regional pollen deposition studies. I thought the two samples taken were also useful by increasing the sample size even though I can imagine counting pollen from just one sample was a real nightmare.
ReplyDeleteThe third pollen paper I’ve read that investigates the plant community assemblages of this time period in this same region. They all basically describe the same thing – the successional change with the retreat of the ice sheets followed by climate driven reconfigurations. I guess the new thing this paper offers is the distinction between using deposition rate from pollen concentrations rather than percentages which offers a more reliable inference of species abundance. I just noticed that in our table of contents is states that Margaret did the translation for the von Post paper. I’m not quite sure why this was included in the case studies section. Perhaps Lennart von Post was the first to analyze a pollen record which is why he was included in the methodological advances section but since his pollen record study had already been added why include another in this volume?
ReplyDeleteI'm with Martina on this one concerning the inclusion of two pollen study papers. However, I wondered why the von Post paper, that did not have any figures or tables was included, especially with the plethora of methodological advances omitted from the section. While the von Post paper was one of the first to analyze pollen, I feel that this paper itself was better able to express the important implications of pollen studies.
DeleteThese case studies seem awfully thorough if nothing else. I know the introduction listed this paper as changing the way people approached how to quantify pollen assemblages and analyze those assemblages, so that is pretty cool I suppose. Pollen records are something used in other fields besides ecology to give an idea of historical flora distributions, so that is another cool thing. I have a hard time putting these case studies into a larger context and appreciate what they have contributed to the field of ecology.
ReplyDeleteThis seems to be one of those studies that really needs some background and context to truly appreciate. Peet does a good job of this in the introduction, mentioning (1) the methodological advance of using pollen influx as an estimate of absolute rather than just relative abundance, and (2) recognition that community assemblages existed in the past that no longer exist today. Sometimes modern analogues can be found in other geographic locations (e.g. the Midwest), but not necessarily. I was able to appreciated this paper more than von Post, as it is more concise, has better visuals, and quantifies things more precisely, but I guess von Post made the more significant methodological advance by more-or-less inventing palynology in 1916. It's easy to think of papers like Davis' as isolated case studies that only give us fuzzy, anecdotal views of the past, but in fact there have been a number of efforts to synthesize and aggregate results from many different lakes across a region. For example, the PaleoEcological Observation Network is group that is reconstructing forest structure and composition across the Northeast over the past 2000yrs based on a synthesis of a bunch of paleo and historical data and some fancy stats. See https://www3.nd.edu/~paleolab/paleonproject/
ReplyDeletePeet mentions that, along with Whittaker and Curtis & McIntosh, Davis' study helped put the nail in the coffin of the Clementsian paradigm. I can see how this study is relevant to that debate in the individualistc rise and fall of taxa and the non-analogue communities, but Davis doesn't discuss her results in these terms. I do wonder if Peet is overstating the role of this paper in the individualistic v. organismic conversation, maybe because he really wanted to talk about Whittaker and found this to be the a logical connection. I do have to say that I'm very surprised and sad there's no Whittaker in this book. I guess he was too long-winded, but his Sante Catalina Mountains paper is only 24pgs.
In reply to Martina's comment, maybe two pollen papers were included in the volume because pollen studies in general have been so important to establishing maps for the history of vegetation since the last ice age, which have supported the Gleasonian rather than Clementsian view. Von Post's was the first, and Davis' represented innovative new methods. It is striking that in 1916, Von Post commented, "And additional more comprehensive studies of the present day pollen rain must be carried out before the pollen diagrams can be interpreted in detail." 53 years later, Davis followed through on this.
ReplyDeleteThis study makes me think about resurrection studies where people are looking at seed banks embedded in ice core or sediments to understand the influence of the past climate in the genetic variation of that particular organism further back in time. Comparing changes in adaptive and neutral variation in certain genes people can have different insight of climate change effects upon plants an animals.
ReplyDeleteAnother pollen study, while I am glad I don’t work with pollen, I find these studies interesting. The idea of using pollen sedimentation to track vegetation succession is pretty cool. Pollen is used for other purposes as well, such as dating objects, determining where an object (such as one of forensic interest) has been it is even used by a lab-mate of mine to determine what plants her bees (Melissodes) are utilizing.
ReplyDeleteIt is always amazing that we can see these kinds of changes purely based on what kinds of pollen. I think this study still falls into similar traps as we talked about before, i.e. wind pollinators being more heavily represented, but that doesn't detract from it offering a cool way of looking at stuff and it is a good case study.
ReplyDeleteI don't know that the overrepresentation of wind pollinated plats is bad or, for that matter should be viewed as a trap. If we can identify the numbers and species (or more likely genera) of the wind pollinated plants in an area, without too much more effort, we could infer the other members of the historical plant community based on modern communities. This hypothesis regarding the historical plant assemblages could be used to search for evidence of the other, expected plants or used in historical biogeography or climate studies. Its interesting to think about what we would or wouldn't know about prehistoric ecosystems without pollen data.
Delete