Meta-post

Thanks for all of your wonderful posts and comments! I hope this is useful and fun and not too overwhelming!

Regarding Thursday night exam- I was not able to get a computer lab so we will be meeting in the regular room and writing using pencils and pens!

Some general comments regarding issues that we have touched on but been too time-strapped to explore further:

What is equilibrium? One definition is that it is "the point at which forces cancel each other". To a community ecologist, the "forces" might be competition, predation, mutualisms etc. (biotic factors), which is, of course, a gross oversimplification of nature--- Comm. ecologists may integrate disturbances, as we saw with the Moretti paper and its investigation of fire. An example of a "theoretical equilibrium" is the abundances of two competing species that results in crossing of their zero-growth isoclines-- all other things (prey populations, parasatism, climate, floods, fires, etc.) being equal (never the case), the "community" would be at equilibrium. Also, the community might be locally stable if a small increase in either results in a move back towards the center. Generally however, more studies are debunking the idea of the "balance of nature" or any sort of stable system. Fluctuations and chaotic change in population numbers seems to be more the rule than the exception. I think this topic will continue to be discussed as we move forward in the class.

Species richness and other measures of diversity- Which is the best to use? A benefit of observed species richness is that it is easy to understand and convey. Diversity indices provide numbers that are only interpretable relative to each other. Estimators, while supposedly being more robust, take you one or more steps from your empirically observed data. As you keep fiddling with different estimators, indices, functional descriptions, etc. you can run into the problem of too many analyses (inflating type I error), and ultimate confusion! Best idea is to really read up on the literature beforehand and develop a compelling a priori dependent variable for your a priori hypotheses. Email scientists that have published these papers and ask them for ideas for good measures or diversity to use.

Lemaitre et al 2009- Did they just die of the cold? The authors state that the primary source of mortality was hypothermia and they also state that "the effect of temperature was additive to other variables influencing survival and did not interact with the presence of bot fly (Appendix 1, ESM; Table 1)." I did not see the interaction data in Table 1 and could not access the appendix to evaluate this claim (without seeing mean temp for bot fly + vs. -). They give this data for confinement time in the discussion.

Invasive species? Are they all that bad? Short answer: if they compete with you, no. If they eat you, yes. From Sax and Gaines 2008 (not a required reading):
"Among the 204 vertebrate species with listed causes of extinction, some form of predation (including human hunting, carnivory, and infectious disease) is cited as the sole factor responsible for species extinctions in 69 (33.8%) of extinctions, predation together with other
contributing factors is cited for 85 (41.7%) of extinctions, and competition together with other factors is listed for 18 (8.8%) of extinctions. In no case is competition listed as the sole cause of species extinction."

More evidence for the importance of predation relative to competition?

P-values irrelevant? I think so but keep following the increasing literature on this debate. All the techniques you learn in stats (regression, anova, ancova, etc) can be assessed in an information theory framework so are completely relevant to your development as ecologists. The "Pitfalls" article mentions that you should either use p-values or information theory but not both-- this seems a little dogmatic. Some reviewers will be more familiar with p-values and others with AIC--- if it you have space you may want to provide both. Just keep your inferences locked to one paradigm.

Leamaitre et al. Bot fly parasitism of the red-backed vole.

While most investigations that identify factors that control population dynamics tend to focus on predation and competition, there is little empirical evidence of the role that parasites play on host populations. This study aims to assess the role of parasitism of red-backed vole by bot flies in boreal forests in Quebec, Canada.

The red-backed vole was chosen for this study as it is the most abundant small mammal and primary host of bot flies, in eastern Canada. Density-dependent impacts of bot fly parasitism on host survival and reproduction would ultimately be needed if this parasitism regulated vole populations.

In order to determine the potential impact that bot flies have on regulating populations of red-backed voles, this study:

1. Assessed the impact of the parasites on:
1. Probability of vole survival under stressful conditions, and whether this was density dependent.
2. Reproductive activity in females

2. Indentified main factors driving the relationship between:
1. individual risk of infection and
2. abundance of bot flies in red-backed voles, host abundance, two host life history traits (sex and body conditions) and combination of three habitats variables

3. Impact of bot fly prevalence on growth rate of red-backed vole population between mid-July and mid-August

The probability of vole survival in live traps decreased with bot fly infection. The authors state that this indicates that infection by bot flies on red-backed voles is a recent coevolution since strong host-parasite coevolution should not reduce host survival (as this would impact parasite survival). However some authors argue that the reduced host survival is not so much due to coevolution hypothesis as it is an artifact of mark-recapture studies. Resident voles would be more highly infested (due to bot egg placement near burrow openings) and more resident voles would be recaptured than transient voles which would inflate survival estimates of highly infected resident voles.

Probability of infection and abundance of bot flies in red-backed voles, were both negatively correlated with vole abundance. The authors concluded three possible reasons for this.

1. A one year time lag between population cycles of host and parasite since the bot fly requires two years from egg deposition to adult emergence.

2. When vole populations are at capacity, a larger proportion is transient and these transient voles have lower probability of infection.

3. Dilution effect of all hosts could occur when vole population is high.

Sex and body condition of the host and habitat variability were not significant predictors of infection probability and abundance of bot flies.

Bot fly infection prevalence was linked to a decrease in short term growth rate of vole pop over summer. This study indicates that bot flies can impact host population growth, during peak reproductive season of the host.

Bot flies have the potential to reduce survival of red-backed voles, an effect that may apply to large portions of populations. This study supports the idea that parasites may play a role in regulating population dynamics of hosts, although bot flies may be more of a limiting factor than a regulating one.

Another key point from discussion in class: since most mortality in live traps was due to hypothermia, the conclusion from this would be; don’t freeze the voles in your experiment, it might screw with your data.

Hey All,
Here is the link to the NPR radiolab episode we were talking about in class. Its pretty entertaining! Enjoy!

http://www.wnyc.org/flashplayer/player.html#/play/%2Fstream%2Fxspf%2F133384


ps. the bit about the hook worms curing allergies starts at approximately 30 min, the explanation of crazy cat ladies follows shortly after

Johnson and Omland: Why everything you know about P-values is stupid

Johnson and Omland argue for a new model of evaluating statistical tests instead of the traditional P-value tests against the null hypothesis. While we are used to tests with a null hypothesis, it is typically biologically irrelevant and its rejection is only indirect support for another alternative hypothesis. Furthermore, rejection happens at an arbitrary threshold and can be ineffective at evaluating incredibly complex and dynamic systems that biologists routinely study in nature.




And out of the wasteland of useless null hypotheses, “model selection” emerges as a saving light! Since it is able to evaluate and select a best-fit model out of a number of candidate models the researcher has proposed, it can evaluate complex situations and multiple influences. After using a simple X2 test or a parametric bootstrap procedure to fit the global model to the data, individual models can be evaluated by Least Squares to determine their effectiveness, and the Schwarz and Akaike Information Criterion can synthesize several models simultaneously into a new “best set” of available models.


Johnson and Omland highlight model selection’s successes in biological fields. In ecology, model selection is extremely useful in modeling population cycles, and abundance and survival probabilities. By modeling different survival and encounter probabilities, researchers can separate the probability of marked animal death against the probability of the marked animal surviving but not being recaptured. In addition, model selection is extremely useful for phylogenetic trees, where the order of each mutation is treated as a separate model and evaluated against one another for best-fit. Lastly, they mention that it can also be used for identifying selective pressures and adaptation in the wild.


In their second to last section, they highlight fields that should have converted to the wonders of model selection, but are stuck in the heathen ways of null hypothesis tests. In particular, they cite statistical phylogeography as being able to use current genetic distributions to recreate possible historical populations. In ecosystem science, model selection would be able to evaluate complex trophic systems and multiple food-web models.


Johnson and Omland do acknowledge a few pitfalls about using model selection, but give very little time addressing these issues (apparently leaving it to Anderson and Burnham). First, they note that the conclusions can only be as the multiple models the researcher develops before testing the data. However, I see the eagerness to jump onto the model-selection bandwagon can start from poor initial hypotheses and relying solely on post-hoc conclusions. Secondly, they state that the predictions must actually be biologically plausible and not “merely” statistical significant. If these are not plausible but are “statistically significant” this should indicate either a flaw with this modeling or an unknown factor that is biologically plausible. And lastly, they state that it might not be useful in all situations and null hypothesis testing should be preferred.








…wait what? I thought you were telling us that null-hypothesis tests were terrible and model selection infinitely better? Perhaps they have a little realistic sense about this model after all.

These models do seem like they could be extremely useful, but I would exercise extreme caution about making too many post-hoc conclusions based on the data. And while these models are evaluated as best-fit, there is no objective decision for when the best is good ENOUGH. All models could be relatively terrible at evaluating the natural situation. I would see Anderson and Burnham’s paper for more discussion about the dangers of these tests.

Moretti et. al. 2009: Taxonomical vs. functional responses of bee communities to fire in two contrasting climatic regions

Moretti (et al 2009) examined bee communities in response to fire as a model of animal function trait response to environmental change. The study took place in two different regions; a temperate climate on a mountainside in Switzerland, and a Mediterranean climate in Israel. The climate in Switzerland was wet/warm in the summer, and dry/cool in the winter, while Israel was hot /dry in the summer, and cool/wet in the winter. Both areas are prone to fire but they occur more frequently, of higher intensity/extent, and during the summer in the Mediterranean climate, while in the temperate climate they are of low-medium intensity and quick spreading, furthermore they mainly occur in the winter (coinciding with vegetation dormancy).

Bees were collected at 21 sites in each of the study areas. These sites were located in a range of habitats in different succession stages following fire (e.g. from recently experienced fire, to no fires for 20-30 years, to unburnt areas). Bees were identified and functional traits were described. Functional traits, listed in Table 2, are such things as nesting specialization (in the mud, in wood, snail shells etc.), dispersion (measured as ITD, inter-tegula distance, i.e., distance between wings), or other characteristics of an organism which has demonstrable links to the organism’s functions. Functional traits were calculated into mT, an average for a given trait weighted by species abundance, and a measure of functional diversity (FD). *this may be a rough definition, but I understand functional diversity as the diversity or range of things that an organism, or an organisms traits, do in a community*.

The hypotheses of this study were: “(i) fire affects both species and functional composition of bees at both taxonomic and functional levels, that is, through extinctions, immigrations, or dominance redistribution within the community; (ii) the response of bee communities to fire is smaller in the Mediterranean region because of the longer history of fire in arid conditions”.

Following fire disturbance, a greater number of species were present than pre-fire conditions in both climates. However, this diversity of species decreased quickly following fire, likely because habitats immediately following fire support the highest floral diversity compared to later successional stages.

Functional traits selected by fire were similar in both countries, but some traits were identified which were affected differently between the two regions (Table 4). This differentiation in trait responses highlights the effects of the regional climates and fire regime on the functional components within the regional species pool. This can be further explained by looking at Figure 2, which displays the variance of mean traits and functional diversity in each climate. Overall, mean traits and functional diversity change more dramatically in Switzerland than in Israel (Figure 1). Changes in functional diversity in Israel due to fire were not significant, while they were Switzerland (albeit, only be 21.9%).

What this tells us, is that the responses of bee traits to fires are not exclusively a result of the fire regime in the area, rather, the climate of the region as a whole, including landscape structure. The climate and landscape influence the diversity of traits present. In Israel, the landscape is highly heterogeneous as a result of land use, vegetation type, and fire, resulting in higher heterogeneity in species and trait assemblage. Switzerland, on the other hand, dominated by quickly establishing chestnut trees, is less heterogeneous, and therefore, there is less heterogeneity in bee species and trait assemblage. So, despite results showing a significant shift in species composition with fire in both regions, the functional composition of the bees present in the Mediterranean remained stable, likely as a result of the regions characteristics and functions and higher diversity of traits in the system.

Below is a link to a presentation by Moretti. It discusses some of his other work, but the 2009 paper is presented beginning at slide 51. I recommend looking through the entire presentation as it’s pretty interesting/helpful. Also, note page 58, it has a figure of the species response to fire in both the Mediterranean and temperate climates (I think this would have been good to show in the paper, but I guess if he had to cut a figure, this would be the best one).

I’ll update this if any issues/questions arise in class. See ya in a few hours!

http://www.eurac.edu/NR/rdonlyres/836134A5-3468-4888-979C-C32A14370A06/0/Moretti_Marco.pdf

Anderson and Burnham: Avoiding Pitfalls When Using Information-Theoretic Models

Anderson and Burnahm discuss the growing shift away from simple null-hypothesis testing to the more complex and arguably more meaningful models based on information-theory, specifically in wildlife biology. This shift has been prompted by the limitations inherent in null hypothesis testing. However, as the use of information-theory rises, so does the misuse of its methodology. Luckily for us, Anderson and Burnham are here to point out our mistakes and move us in the right direction.

The common mistakes fall into 3 categories: basic science, methodology, and just plain wrong.

Basic Science Mistakes:

1) Poor Science Question - Null Hypotheses are silly! They don't ask relevant interesting questions and since when is science black and white? Having multiple possible models will most likely yield a more descriptive answer. However, models must be chosen carefully because the more specific the models become, the more likely it is that they won't capture reality.

2) Too Many Models - Some scientists are becoming too reliant on computer statistical analysis and don't put enough thought into their data analysis. They may let the computer analyze all possible models rather than reducing the number after some consideration of the science regarding the issue.

3) The True Model is Not in the Set - A model cannot possibly fully represent the issue being investigated. However, information-theory can try to estimate how close a particular model is at representing the truth.

4) Information-theoretic Methods Are Not a "Test" - Null Hypothesis tests (and their resulting p-values) cannot be mixed with information-theory results. These are two different approaches of answering a question that lead to different results with different interpretations.

Methodology Mistakes:

1) Poor Modeling of Hypotheses - Applying mathematical models to alternative hypotheses is difficult and often leads to mistakes or oversimplifications. Linear models are often chosen due to their simplicity but are often not realistic - asymptotes, thresholds, etc should be considered.

2) Failure to Consider Various Aspects of Model Selection Uncertainty - Scandalous! The statistical uncertainty of choosing a best fit model is often not included in the model's precision.

3) Failure to Consider Overdispersion in Count Data - Overdispersion occurs when data independence is assumed but may actually be dependent. This results in underestimating sampling variance.

4) Post hoc Explanation of Data Not Admitted - Data analysis should never be influenced by the outcome of the results.

5) Statistical Significance versus Quantitative Evidence - P-values must be euthanized! Information-theory has no predetermined alpha levels that magically make results significant or not. Instead, significance is based on the strength of the evidence and biology that is being addressed.

6) Goodness of Fit Should Be Assessed Using the Global Model - GOF should not be assessed for each model, but on the most highly parameterized model

7) Failure to Provide All the Needed Information - Many papers using information-theory leave out values for important information in their results, which can hinder the interpretation of their models.

Outright Mistakes:

1) The Incorrect Number of Estimable Parameters (K) - There are many variables that influence the value of K, which are not always included.

2) Use of AIC Instead of AICc - AICc should be used for smaller sample sizes.

3) Information Criteria Are Not Comparable Across Different Data Sets or Different Response Variables - pretty self-explanatory. Data sets must be analyzed separately and response variables must be kept consistent.

4) Failure of Numerical Methods to Converge - Computer software can fail to converge on the maximum of the log-likelihood. When this happens seek professional help.

So, is null hypothesis testing dead and should p-values be "euthanized"? If it's so terrible and outdated why is it still so common and taught as a ritual to students?

Gonzales and Arcese: Herbivory more limiting than competition on early and established native plants in an invaded meadow

Here is it, the big question! Which is more important: Competition or Predation? According to Gonzales and Arcese...in this case...predation wins! (kinda)

Here's a short review of what happened in this study. This study was conducted in a fragmented oak meadows in Canada. In these meadows, there was a native grass that is declining in abundance compared to an invasive grass that is increasing. On its face, this appears to be an open-and-shut case of competitive exclusion of the native by the invasive. However, the researchers also mention that this habitat is fragmented due to human activities, which along with general disturbance often have the effect of decreasing the local carnivores that would keep the herbivores in check. This has lead to an explosion of the deer population in these areas, which could have increased the effect of herbivory (predation on plants). And so with all this in mind, Gonzales and Arcese propose 3 hypotheses: 1) The invasives are out-competing the natives 2) Deer are consuming too many of the natives compared to the invasives or 3) some combination of both. They then conclude by saying they think that herbivory has the greatest effect

(completely unrelated to ComEco, this sort of reminded me what was asked in Stats last night about just starting with multiple hypotheses and then following the one that's the most right....does anyone else see that?)

In the experimental design, there were several. First with the grasses, they had a 2x2 factorial where they have transplanted native grasses and native grass seed addition that varied by + or - herbivory and + or - competition (either by clearing or severing the roots of neighbors). And this design was repeated for the flowering plant seed addition experiment. In a 3rd experiment, they did a bulb addition experiment of lilies, but how they set this up was a little hard for me to follow. And just to quickly summarize their results; all of the graphs for all of the experiments show that when you remove herbivory, the biomass for all examined species increased.

So what does it all mean? Well, after re-examining their hypothesis, they decide that the dominant effect on the decrease of the native plants is due to herbivory. When they first approached the problem, the usual suspect to blame in these cases seems to be competitive exclusion. However, after this experiment, this seems to be a case of "apparent competition" where because the natives where being preferentially fed on by the deer, we could only see that the natives were declining and the invasives were increasing.

And discovering that herbivory (plant predation) was having a larger effect on the natives than competition seems to be a complete 180 degrees from Errington's Hypothesis. Remember that Errington thought that predation would only have a small effect on organism abundance. Yes, the full hypothesis is meant for larger, motile organisms, but if you squint and turn your head to the side, you can see how part of it could be shown to apply (and then fail) for this situation. For example, Errington says that the survivors would have increased fecundity, presumable because their kin are not there taking up resources. However, this leads into the second part of Gonzales and Arcese's finding that herbivory isn't the whole story. There is also a competition effect! Sure the native grasses should be able to re-fill the spaces where their kin had been consumed from, but the invasive is just better at filling those empty spots.

So, in conclusion, Gonzales and Arcese show that herbivory can have dramatic effects on the welfare of native plant species in disturbed ecosystems that can lead to giving the invasive a competitive advantage. And another theme that we keep seeming to come back to; deer are evil and the cause of a lot of ecological damage. And if you needed more proof that deer are evil, I submit the following picture as my evidence. See you in class tonight!


Figure 1: Aren't they cute with that maniacal twinkle in their eyes?You have to click on the picture to see the whole thing. The blog has cropped it so you can't see all of it. I did a google image search for "evil deer" and this was the best shot. Seems like we keep coming back to how evil deer are with their tendency to mess up plant communities when their numbers get too high.