Papers:
Brooks, J. L. and S.I. Dodson. 1965. Body Size, and Composition of Plankton. Science, New Series, 150(3692): 28-35.
Barneche, D.R., D. R. Robertson, C.R. White, and D.J. Marshall. 2018. Fish reproductive-energy output increases disproportionately with body size. Science 360, 642–645.
Andersen, K.H., N.S. Jacobsen, and D. van Denderen. 2019. Limited impact of big fish mothers for population replenishment. Canadian Journal of Fisheries and Aquatic Science 76:347–349.
Blog Author: Lyndsie Wszola
John Langdon Brooks
Brooks was an evolutionary ecologist who primarily focused on freshwater systems. Throughout his research career, which included undergraduate and graduate education at Yale with G.E. Hutchinson, he focused on ecological and evolutionary interactions between freshwater fauna including fish and plankton. Later in his career, he became a program officer and later director of the division of environmental biology at NSF.
Diego Barneche
Barneche is an independent research fellow and lecturer at the Unviersity of Exeter who focuses mainly on marine systems. He is a quantitative ecologist who seeks to use mathematical, statistical, and theoretical models to answer basic science questions and inform management.
Ken H. Andersen
Anderson is a professor of theoretical marine ecology at the Technical University of Denmark. His work focuses on theoretical and applied models of fish evolution, ecology, behavior, and exploitation. The unifying factor in his work is a focus on size and trait-based modeling.
Brooks and Dodson:
The Brooks and Dodson paper asks what effect the addition of a planktivorous fish, the alewife, has on the community composition and size distribution of the zooplankton in a freshwater lake in Connecticut. They begin with the observation that lakes with naturally populations of the extremely invasive alewife lacked large species of zooplankton. They then establish a natural experiment in which they compared plankton size and community composition form Crystal Lake, which naturally lacks alewife, before and after alewife introduction. They found a marked shift in the community composition and size distribution of zooplankton in the lake towards smaller species. They explained this finding by hypothesizing that the alewife, a planktivorous fish, should prefer larger prey because they are more energetically efficient. They then hypothesize that an alewife preference for larger prey should eliminate those large prey from the food web, releasing competition for smaller zooplankton and causing the observed shift in community composition and size distribution. Essentially, though they don’t use these words, they are claiming to observe a trophic cascade via competition, something which is widespread and powerful in aquatic systems.
Barneche et al.
Barneche et al. begin by identifying the classically held assumption that fish reproductive output scales linearly with body size. This assumption informs many stock and harvest models that are used for both basic science and management. Barneche at al. propose that this assumption is fundamentally incorrect and can lead to harmful mismanagement. Using a large meta-analysis, they demonstrate that fish reproductive output actually increases hyper-geometrically with fish body size. This means, in essence, that one very large fish should have a much higher reproductive output than two smaller fish whose body mass adds up to the same size. Barneche et al. extrapolate this finding to suggest that typical harvest models are vastly underestimating the contribution of large females to fish stocks. They further more suggest, somewhat dramatically, that forces such as warming and over-exploitation, which are driving down fish body size, will have a powerful negative effect on the ability of fish populations to sustain themselves.
Andersen et al.
Andersen at al. wrote their short communication in response to the Barneche et al. article. They counter the claims in the Barneche piece by making two arguments:
1. Barneche et al. over-estimated the relationship between body size and reproductive output.
2. Stock replenishment does not depend solely on reproductive output.
Andersen et al. pursue the first argument by adding some realism to the Barneche model. They make the point that most small individuals are not reproducing, and that including them in models is therefore disingenuous (that is actually debatable). They continue by saying that very large individuals are also very rare, and that they cannot therefore be extremely important to populations (again, there is very good evidence that this was not always true). Andersen et al then replot the dramatic figure from the Barneche et al. paper with what they propose is a more reasonable fit and show that there is a much less significant difference between linear and hyper-geometric assumptions when replotted. Finally, they add more realism to the discussion by saying that reproductive output is not the sole determinant of stock dynamics. Young fish experience density-dependent mortality and few eggs survive to adulthood, so there is often a tenuous relationship between total egg production and recruitment. They conclude by emphasizing that the Barneche et al. paper is responding to an urgent need, but that adding biological realism raises questions between the theoretical work and application.
My thoughts
The Brooks and Dodson paper is an interesting early work on trophic cascades, even though they don’t use those words to describe the dynamics. It is an elegant natural experiment, but I thought they made some very significant assumptions, especially that alewife were actively visually searching for their prey, and that the decline of large zooplankton led directly to a boom in small zooplankton. In a modern context, I would want much clearer mechanistic links between each step in their reasoning process.
I really enjoyed the Barneche paper and the Andersen response. Barneche et al. make impressive strides from theoretical work to data mining to management implications. However, I agree with Andersen et al. that the Barneche paper overstates some of its findings and doesn’t adequately address the complexity of aquatic ecosystems. I found the point about density dependence particularly compelling. As someone who seeks to do this kind of work, the discussion between these two authors emphasizes the perils of trying to build workflows that go from theory to data to practice. It’s incredibly difficult to build models that are scientifically interesting, ecologically sound, and practically implementable. In my experience, the only way to actually do this is to have a diverse author team that includes mathematical modeling experts, theoreticians, field biologists and managers. In other words, I would be very interested to read a paper written by both sets of authors together. I’m very curious what others think about these two papers and whether you are more swayed by one argument or the other.
The Brooks & Dodson paper discussed trophic cascades in aquatic systems by looking at the affect of the presence & absence of alewives on planktonic communities. They were able to present clear data showing the very different community dynamics under the two different conditions. I do agree with Lyndsie that they were a bit lacking in their description of the mechanisms behind the conclusions that they drew from this data, though.
ReplyDeleteIt was super interesting to read the two modern papers alongside each other. The contrast between the two papers did an excellent job of highlighting the difficulties of applying ecological and mathematical theory to real-life management of systems. Managing systems such as hunting & fishing harvest levels can prove to be a challenge because of this issue, and, as Barneche et al discuss, this challenge has only been compounded as scientists and policy-makers must also grapple with the effects of climate change.
- Elizabeth
All three papers were good reads. Most importantly, picking two companion papers, one of which responded to the other, was a fantastic idea. It can demonstrate where other authors find gaps in the study.
ReplyDeleteAs far as the Brooks and Dodson paper, I agree with Lyndsie that some broad assumptions were made that would need to be tested. I am not sure how a certain type of fish selectively chooses one type of prey while discounting others (although I am not very familiar about fish or vertebrates in general).
The Berneche et al. and Anderson et al. papers were amazing to read, mostly because Anderson et al. made some excellent points about the previous paper that may not have been super obvious upon the first read. Their commentary on Barneche et al. disregarding density-dependent factors actually made me go back and re-look at the paper. Questions that remain in my mind are as follows: How does climate change affect body size in fish? Does it have to do with survivability, resources, or something else? The acronym "BOFF" is also hilarious. I wonder if this is actually commonly used.
Just jumping into the conversation to say that "BOFF" is very widely and seriously used. And yes, it is hilarious every time.
ReplyDeleteI really enjoyed the older paper, although that may be my personal taxa bias. I would have liked them to consider abiotic factor apart from lake size. In particular, I would have liked a mention about how the salinity of the lake compared to that of the ocean. This is especially pertinent in that the fish is native to the ocean while the zooplankton are mentioned are found in fresh water.
ReplyDeleteI'm always torn about reading paper that respond directly to another. While they typically are civil and make valid scientific point (such as these two) I've read some that do not and seem more personal than scientific. I always feel bad for the first author who was called out.
-Miranda
I enjoyed reading these papers. One thing that I would have liked all papers to address more is the potential for offspring size-fecundity trade-offs. If big plankton are getting eaten, this may suggest that having more, smaller offspring rather than fewer, larger offspring is advantageous. This is reverse to other observed effects of predation on reproduction, such as in a paper published by Zarnette et al in 2011 in Science, where songbirds exposed to predation cues laid fewer, bigger eggs. Likewise, as is mentioned briefly in the newer papers, how does total biomass of offspring change with the mother's body mass? And how does survival rate of offspring change with biomass?
ReplyDeleteThe classic paper conducted an interesting study on how predator would control the outcome of competition. I was impressed by the large sampling size and the detailed identification of zooplankters. One question that puzzled me was that the authors mentioned at the beginning that Alosa primarily avoided shores and fed on “lake species”, leaving the “pond species” that close to the shores alive, which made me think how they sampled. I was wondering if they took samples separately based on “littoral” and “pelagic”, would the results different between the 2 sampling categories?
ReplyDeleteI enjoy reading the 2 companion papers. I like figure 3 in Barneche’s paper showing that most of the studied species have a tendency to be hyper-allometric scaling. I am not familiar with fishery, but I do agree with some of the arguments brought by Andersen. I thought the fishes would turn to reproductive stage after a certain time which was usually the same within species but different among species, and I wonder if there are any fishes that have “adult size” before sexually mature? And how to set the cut-off size based on their reproductivity if it’s different among species? I think one of the caveats on using ecological models is that the assumptions may not be realistic or may be overexaggerated, and I agree with Lyndsie that cooperation with field biologists and empirical biologists would help to avoid these pitfalls.
The Dobson paper presented its ideas thoughtfully, but I agree with other comments about some of the limitations to its scope. The idea that “natural selection will tend to favor the predator that most consistently chooses the largest food morsel available” seems to reflect some of the previous ideas about foraging of the “high ranked prey items” that we’ve discussed earlier in this class. I’m not sure that it is a valid assumption that the fish selectively choose and eat the largest food morsels available (as compared to eating whatever food source is encountered first).
ReplyDeleteSeeing the arguments between the two companion papers side by side is fascinating. Unlike some other scientific papers where there is disagreement between the authors, the Andersen response doesn’t seem to “talk past” the Barneche argument, but instead seeks to further refine the concepts using a different and more informed approach. The response paper highlights that our conception of “science” is not static, and that new information, ideas and perceptions are important contributions to our dynamic understanding of how natural systems function.
I thought the Brooks and Dodson paper had an impressively mechanistic explanation for the shift in size structure induced by predation. They gave good reasons for understanding why larger zooplankton outcompete smaller zooplankton (no resource partitioning and foraging efficacy increases hyperallometrically with size) and why larger zooplankton would be preferentially predated by fish (foraging theory). While the argument would have been strengthened by including experimental support for the idea that fish preferentially eat larger zooplankton, I think their conclusions and reasoning are sound.
ReplyDeleteI thought the pair of new papers were an excellent demonstration of how false modeling assumptions don’t necessarily result in useless models. A common criticism of all models is that the assumptions are unrealistic and that models leave out important biological facts. Here, Banarche et al. make the important finding that reproductive output increases hyperallometrically with body weight. They suggest that this would result in serious bias in model predictions since modellers typically assume proportional scaling. But, as shown in Anderson et al., the false modeling assumption doesn’t matter when accounting for population size structure.
I agree with Stella that studying how offspring size impacts offspring fitness would be needed to get a more complete picture of how size influences fitness.
- David
The Brooks and Dodson paper was concise and introduced me to an ecosystem/group of predators and prey that I have not learned much about (definitely involved some googling to picture these organisms, but I enjoyed it). I like the simplicity of their size-efficiency hypothesis, but as other folks have mentioned in their posts, some of the assumptions made about how trophic cascades/tradeoffs occur perhaps make this hypothesis over-simplified. Also, when the authors brought in evidence from other studies to support their hypothesis, I felt like they may have specifically only mentioned results from studies that supported their idea (e.g., the rate of “photosynthesis of the nannoseston in the presence of Bosmina was about five times greater than in the presence of Daphnia” – about which Brooks and Dodson state “this is precisely predicted by our hypothesis”).
ReplyDeleteAfter reading the Barneche et al. meta-analysis, I was curious – how is egg energy content (J) determined? I had never heard of this as a metric used when looking at fecundity-mass relationships, although I imagine (where relevant and available) it is probably useful in fishery studies. As far as the two papers, having little familiarity with this field, I thought that the conclusions of the Barneche et al. paper seemed reasonable until I read the Andersen et al. paper and realized it overgeneralized these relationships by not accounting for species-specific demography and density dependence! I agree with Lyndsie – not considering density dependence in the Barneche et al. paper was a significant limitation because, as Andersen et al. point out, hyperallometric vs. isometric scaling do not influence recruitment.
The Brooks and Dodson paper was a cool example of trophic interactions. It was interesting to learn an example of an invasive species that actually had the potential to increase local biodiversity, although intense predation resulted in the extirpation of larger crustacean species and changed where the remaining species occupied the water column. As everyone has already mentioned, some of these assumptions might not fly in today's world of peer review and rigorous methods, but I thought it was overall a robust observational study. I especially liked the parts connecting form to function (e.g., size/number of gill rakers on alewives) because I think this was another necessary link between predator and prey besides relative densities, particularly because they did not observe predation events.
ReplyDeleteI like when scientific papers serve as a platform for debate. I actually read the companion papers for an ichthyology course a couple years ago. I think the high stakes management implications for the hyperallometric relationship was one reason why this wound up being such a contentious issue. Fisheries do not want more regulations on their activity, and conservationists argue we need it because we do not want to run out fish. These papers are exemplify how biases can greatly impact scientific work while emphasizing that scientific watchdogs are necessary even in a peer-reviewed system.