Author
Stella Uiterwaal
Citations
Calder, W.J., Shuman, B., 2019. Detecting past changes in vegetation resilience in the context of a changing climate. Biology Letters 15, 20180768. https://doi.org/10.1098/rsbl.2018.0768
Davis, M.B., 1969. Climatic Changes in Southern Connecticut Recorded by Pollen Deposition at Rogers Lake. Ecology 50, 409–422. https://doi.org/10.2307/1933891
Author Background
Margaret B. Davis – the first (and only) woman in our textbook. She used fossils to study how historic changes in climate influenced communities. She taught at several universities and served as the president of the Ecological Society of America from 1987 to 1988.
W. John Calder is quantitative paleoecologist who studies links between wildfires, climate, and microbes. He is a postdoc in Alex Buerkle’s lab in Wyoming (the same lab as one of last week’s authors!).
Classic Paper
Fossil pollen in sediment provides a detailed, continuous record of past vegetation. This paper used changes in fossil pollen concentration and deposition rates at Rogers Lake in Connecticut to study vegetation around the lake in the last 14,000 years. To collect pollen, the author and a partner took several cores from which samples were taken for pollen analysis. Ages were determined by radiocarbon dating. Pollen was identified by size and morphology. The author finds evidence for tundra-like vegetation and then a transition to woodlands of various compositions (spruce, oak, fir). She also finds evidence for fir, larch, birch, alder, and pine trees. Around 8,000 years ago, white pine becomes dominant and other plants (like ferns) increase. This is likely due to a shift in climate in the region. After this period, vegetation becomes characteristic of prairie. In the last few centuries, pollen records reflect vegetation changes due to European settlers. The author also describes finding evidence for the introduction of new tree species such as chestnut and hickory in the last few thousand years. She attributes these sudden onsets of new species to migration from glacial refuges. In addition to confirming what other studies have concluded, this paper provides evidence of vegetation transitions that were previously not well understood.
Modern Paper
The authors of this paper studied resilience in palaeoecological records by looking at changes in vegetation in response to wildfires. To do this, they collected pollen and oxygen data from Rogers Lake and Summit Lake, both in the US. They also used previously constructed vegetation and wildfire history information. Wildfire data was obtained by looking for macroscopic charcoal deposits, indicative of high-severity wildifires within 1 to 3 km of a lake. To determine resilience, they used a linear model relating oxygen to conifer pollen percent to generate predicted vegetation values to which they compared observed values. Resistance is then the difference between the predicted value post-disturbance and the actual value.
Within their period of interest, they found that pollen levels declined and remained low at the Rogers Lake site. About 50% of the variance in pollen levels at this site could be explained by oxygen. At Summit Lake, they also found a decline in pollen over the last 2500 years, also due to oxygen levels. A century of high-frequency wildfires resulted in a noticeable decline in pollen levels, and the conifers never fully recovered from this disturbance. The authors also note that there was a lower response of conifers to climate post-wildfires. The authors conclude that climate was ultimately responsible for the reduction of conifers, but wildfires altered the vegetation-climate equilibrium.
My thoughts
I am impressed and intrigued by the methods used in both papers. I know very little about paleoecology, and it was neat to learn that pollen can be used to reconstruct vegetation composition around lakes. I also like that Rogers Lake was being used for these kinds of studies in the 1960s and is still being used for the same type of research today. A clear limitation of this method is the age of the lake. I wonder if there are older lakes around the world where similar work could be done.
The Davis 1969 paper is more in the realm of my understanding so I enjoyed it quite a bit. I don't read a ton of papers regarding pollen and paleoecology, but I found this one relatively easy to comprehend. It was fascinating to see the author piece together the past environments based on the pollen abundances in the lake core samples. I was curious to see why only two core samples were collected (one from either side of the lake). I would have thought that more than two core samples would have been taken to gain a better picture of the pollen abundances throughout time, but there may have been a restriction because of time or some other factor.
ReplyDeleteAs for the second paper, the Calder and Schuman (2019), it was a great follow-up to the original 1969 paper. It was particularly interesting to me that they tracked fires that affected the vegetation (presumably to determine an equilibrium, I think?). They seemed to jump into the idea of looking at wildfires in the methods without really explaining why they were doing this until the discussion, which bothered me. I also wish they would have spoken a little bit more about what equilibrium is at the end of their paper.
In the introduction of the Davis paper, she makes the argument that in sediment deposits, the yearly influx of each pollen type is independent of the influx rate for other pollen types. I don’t know a lot about paleoecology, but this seemed somewhat counterintuitive to me. Competition is frequently occurring between plant species, so wouldn’t their success in seed production/dispersal (i.e., reproductive events) have an influence on what is ultimately recorded in the pollen record? I think that the density of plants at any given time surrounding where the pollen samples were taken would have a substantial effect on what species are present in that sample, so I partially understood her logic in assessing abundance of species over geologic time but it was unclear to me how you tease apart those differences. I was also curious to learn from people more familiar with carbon dating – has the accuracy of the method she described (related to estimating matrix accumulation rates) improved at all, or is this just very difficult with sediment cores?
ReplyDeleteI thought the Calder and Shuman paper was neat in that it brought full circle some of the discussions we had at the beginning of the course regarding the idea of changing, dynamic equilibria and the idea of recovering/return back to the original steady-state, but they put these ideas into a paleoecology context. I thought that the two approaches they used at the two locations were interesting, but I didn’t really understand why they couldn’t compare both the newer data (2500 BP to present) and older data (13000-11000 BP) from Summit Lake to the older time period examined at Rogers Lake for a more meaningful comparison. I know they were interested in assessing the effects of wildfire severity in more recent history, but I had a hard time comparing the data presented in Figure 1 given the spatial and temporal separation.
I found Davis 1969 fascinating and extremely impressive. Reading through the extensive list of methods, it's amazing that she was able to both innovate the data collection and analysis at the same time she was drawing novel scientific insights. I have worked on glacial lakes in New England, and pollen analysis is still a very important technique there. It was fascinating to read about how it was developed.
ReplyDeleteI agree that the methods in the Calder and Shuman paper were interesting. The definition of resilience they use is different than the classical definition, which more or less equates to the "the latency of a system to move from one stable state to another." The classical definition, in seeking to be comprehensive, is also extremely vague and difficult to operationalize. I therefore found their idea of measuring the residuals from expected dynamic equilibrium to be an interesting, applicable way to conceptualizing resilience. However, I thought it was contradictory to pursue the idea of quantifying resilience in dynamic equilibrium while also emphasizing the idea of return to a previous state, which they never actually seem to document.
The Davis paper was by far my favorite older paper we have read this semester. I thought is was extremely well written and easy to follow. And for the first time, her methods were repeatable, as shown by the newer paper. My only complaint is that it didn't have nice illustrated timeline figure showing the change in vegetation.
ReplyDeleteTo me, the newer paper had a tone similar to some the first older papers we read with all the talk of recovery and resilience as thought an ecosystem has some ideal stable state. Ecosystems are dynamic and should be treated as such. Understanding how systems change overtime is important, but one state shouldn't be held above another, especially looking back in history.
-Miranda
Since I don’t know much about paleoecology or plant communities, I thought the most interesting parts of the readings were the methods since they were totally foreign to me. I thought Davis’s approach to using pollen abundances was an amazing way to reconstruct the community composition of past plant communities over long time scales. I also thought her method of quantifying observation error of pollen by introducing known quantities of Eucalyptus pollen was neat. I do wonder how important lake “sampling error,” that is, changes in water circulation that alter sediment deposition and consequently pollen, affects the inferences researchers using these data can make.
ReplyDeleteI thought the questions Calder and Shuman were asking about resilience of vegetation to disturbance was interesting. I especially liked how they operationalized resilience and clearly quantified them which made the results easier to follow. With that being said, I my lack of familiarity with paleoecological studies was a barrier to any deeper understanding or assessment of the paper. For instance, they present results that are in units of per thousand VPDB which is apparently an abbreviation of “Vienna Pee Dee Belemnite.” I have no idea what that means. Moreover, it’s unclear to me why they only used pollen percentages without considering abundances like Davis did. I would think the abundance of pines, not just their percentage, would be of interest to understanding community responses to disturbance.
- David
The Davis’ paper is impressive. It linked the sedimental pollen records and the climate history together and studied the changes in vegetation in response to climate changes. I am not familiar with plant paleontology the methods in this paper is clear and reasonable to me. The results gave a clear image of how the vegetation shifted. I am surprised that the pollens preserved for so long could be identified to family level (table 2) and even genus level (fig 3-5), which makes me wonder how precise the identification could be. Another question is whether pollens of different genera/families last equally, and whether the fractions of pollens being preserved are the same among plant taxa.
ReplyDeleteThe companion paper is a nice follow-up study that compared the pollen records in Rogers Lake (the study area of Davis’ paper) and records from Summit Lake that have high resolution. Although the authors called for high-resolution climate proxies, the sediment age of the Summit Lake was much younger than that in Rogers Lake. I don’t know if there is a way to get the high-resolution data for the old lakes.
While I don't have a strong background in paleoecology, I found Davis's paper to be extremely readable and her discussion to be quite clear. The methods especially were laid out in a way that came across as much closer to modern papers, going into detail about how exactly she conducted her study and expressed in a way that could be repeatable for further study. This stood in contrast to last Wednesday's paper, where we had many questions raised in our discussion regarding how the authors accomplished their data collection.
ReplyDeleteThe Calder paper paired nicely with the classic paper. Although it seems like a small detail, one point that I appreciated was their acknowledgement of how definitions of "resilience" may vary and that they outlined precisely what definition they were working under to frame their study. They both gave a brief definition, as well as citing the paper that they used as the framework for their definition. "Resilience" is one of those words that certainly has different definitions to different ecologists, so laying out their definition right at the beginning of the paper helped a lot with readability and interpretation of their paper.
- Elizabeth
I am not familiar with paleoecology, but the methods in the Davis paper seemed pretty straightforward. There were a few parts that I did not quite understand. Davis referenced the effect of “ancient carbon apparently from groundwater” that added 730 years to age determinations, but I was confused as to what the “ancient carbon” actually is. Also, is the deposition of the pollen record strictly limited to freshwater aquatic systems? If so, does this introduce bias into the fossil record?
ReplyDeleteIt is interesting to compare the use of “climate change” between the two papers. The Davis paper looked at what might be considered more “natural” climate changes (e.g. glacial retreats), while the Calder and Shuman paper specifically considered how the previous “natural” climate changes might be used to predict resilience for ecosystems when considering the current anthropogenic changes to the climate. It was nice to some repeatability in the Davis work when the same Rogers Lake was used in the companion paper.
I have no experience with paleoecology or plants outside of the few papers we have read in this class, so the methods from both papers left me with lots of questions. For the Davis paper, are two cores from one lake sufficient for robust results? How do they determine how thick to cut the core samples? There was a lot of jargon, so I appreciated the glossary at the beginning of the result section. I agree with David that the abundances of pollen as a measure for community assemblages was clever, but I also wondered whether these results could be skewed depending on pollen materials. This is an uncertainty I have always had about paleoecology as an outsider of the field. Is there just always an assumption in paleoecological studies that the materials recovered are representative of communities in the absence of other evidence? What if there were species that were not well preserved but were incredibly important to the community? Or is the fact that we didn't find any specimens representative of its importance? This seems particularly relevant to pollen because of its small size and flimsy composition.
ReplyDeleteFor the Calder and Shuman paper, I agree with Lyndsie's comments. I felt that the authors did well laying out the pitfalls of misdefining a baseline equilibrium state and arguing for a shifting or dynamic equilibrium. However, I was unsure about the methods of defining the expected equilibrium state. The authors also define recovery in the intro as the time it takes to return to equilibrium, which can change, but then argue that it did not recover because it didn't return to its previous equilibrium state. This was contradictory to their definitions and to the arguments they made in the introduction.