I already liked E. O. Wilson before reading this paper, but
after reading I have a new found appreciation for Wilson,
Simberloff, and others that put abstracts in the beginning of their papers. As
mentioned in the section introduction, this paper built on the theory of island
biogeography and species equilibrium proposed by Wilson and MacArthur by
actually testing the theory experimentally.
Experimental Design
The design of this experiment is nice in its simplicity,
although I question whether it is a design that researchers would be able to
repeat nowadays. Simberloff and Wilson Performed their experiment on a set of
mangrove islands in the Florida Keys. A survey of the arthropod fauna of each
island was done prior to defaunation, which is a great word I think, by tenting
the tree island and gassing it to kill everything. Periodic surveying was then
done for the next year to document what species colonized the island and when. An
entire section of the paper was devoted to an explanation of why certain
species were not included in their study. This was also their methods section
where terms used in the rest of the paper, and how they recorded their data, are
defined.
Seasonality
I liked this section. Basically Simberloff and Wilson explain
in detail how the Florida Keys completely lack anything that could be
considered seasonality, and because of this they don’t think seasonality is a
factor for colonization of an empty island in their system. They did try to see
how differences in wind patterns might be correlated to colonization, but were
not able to acquire detailed enough measurements for any conclusive results.
Patterns of
Colonization
It is beneficial to have a background in entomology for this
section I think. I don’t have that background so a lot of these names sound
really cool, but I didn’t know what they were talking about. The overall
concept here though is that there is a pattern to colonization of an empty island.
The early colonists are usually strong fliers that can make it to the island
the easiest. The final colonists are ants, which Simberloff and Wilson paid
special attention to. That makes sense since Wilson is an ant guy. The other
notable pattern is that predictability of colonization increased with later
colonists. Early colonists appeared and died, but the later colonizers were
more stable.
Colonization Curves
Ecologists really like math. So, the colonization curves
seen in figures 1-3 are pretty cool. This is what Simberloff and Wilson were
counting in this experiment, how many species colonize an empty island and what
is the time frame for that colonization. The cool thing here is that each curve
approaches the number of species present prior to defaunation and sort of hangs
out around that number. The significance of this finding is that it supports
the proposed concept of an equilibrium in the number of species present on an
island. Here is the first equation in the paper, but all the equations are nice
and simple. I think this is the coolest part and will hopefully garner some
interesting conversations in class.
Dispersal
This is a rather long section for a somewhat simple concept.
Here Simberloff and Wilson discuss the mechanism of dispersal for colonization
to the islands. The take home here is that if you are an arthropod and want to
go to a cool new island do not try to swim there no matter what. Fly there if
you can using the wind to your advantage, and if you can’t fly get on a nice
leaf and still let the wind take you there. If you end up in the water you will
be so dead so fast, and all of your little arthropod dreams of setting up a
cool dive shop on a soon to be discovered awesome island for retirement will
wither away to nothing in the gut of a fish that cares nothing of you or your
little arthropod spawn.
Immigration and
Extinction Rates
This is the crux of the theory under test in this experiment.
The model by MacArthur and Wilson for
island biogeography really comes down the immigration and extinction. Early on
the immigration rate exceeds the extinction rate and the number of species
occupying an island increases with extinction rate not being affected by
interaction between species. As the number of species increases and the
populations of different species also increases then interaction between
species increases and the extinction rate increases. For each island there is a
certain equilibrium dictated by the size of the island and the distance from
the mainland, or source. When below this equilibrium immigration rate exceeds
extinction rate until you approach or go beyond the equilibrium. At equilibrium
the immigration rate and extinction rate are pretty much the same, hence it
being at equilibrium. So there you go. All of your questions about island
biogeography have been answered and we can spend Tuesday eating cake.
A classic paper, though I think the companion paper describing the fumigation treatment is likely better known. This system is novel in its climactic homogeneity allowing seasonality to be largely overlooked. Without this aspect of the system intact this experiment would be very hard to draw conclusions from. It was crazy to me to read about spiders using “balloons” to reach the island through the air. Considering the size of most insects it seems like aerial travel is risky, but certainly better than the other options on water. The E7 case was most interesting to me because the opening of decaying log niche space led to a different outcome than seen on the other islands. Most importantly this illustrated the key concept of this experiment that niche space drives diversity and abundance in this system. Each island has a set number of species and individuals it can accommodate and the only way that can change is through alterations in the type or amount of fundamental niche space available.
ReplyDeleteRemoving fauna from entire islands and then monitoring recolonization is a really nice experimental approach to capture trends in a natural environment. Of course these islands were small and not far from sources of fauna and there was quite a bit of difficulty in measuring and interpreting occurrences. But not only is it an awesome approach but the results themselves are also substantial. The big take away here is that there is an “equilibrium” in species richness that is reached and maintained over time. This is very interesting being that composition and overall diversity (which incorporates abundances with species richness) may vary widely. I also thought it very intriguing that there is a decline in S after the peak at which point it is determined that biotic interactions begin to take place but the decline is slight until the actual equilibrium is reached. Perhaps this is because for a given habitat there is “fixed” number of different species that may exploit the resources until the habitat changes to produce new niches. Since the habitat is constant S can reach an equilibrium but I would suspect that under constant varying conditions or constant disturbances S over time would fluctuate much more. So it seems in this sense that species richness is driven more by the environment than biotic interactions.
ReplyDeleteA great paper - although I agree with what Sami wrote in questioning the feasibility of it happening today. The experimental design covers a pretty large scope, and that's one of the reasons the paper is so influential! I'm amazed that this experiment happened and had successful findings, because it's kind of like killing everything on Earth and seeing what colonizes first (obviously not that extreme, but the same concept). In addition to reading the paper, I loved looking at the appendices and the visualization of what was present on the island before and after the defaunation. A lot of the species that had been present before the defaunation either didn't come back at all or took a long time to repopulate the same island. The same goes for species that hadn't been there before - they ended up establishing a population where there hadn't been one before. I love this study for the ambitious scope the authors took on, and think that the findings made a lot of progress for the field of Ecology.
ReplyDeleteSomething I found unique and really cool about this paper is how Simberloff and Wilson seemed to be both superb naturalists who know their system and species very well, while also being strong theoreticians, bringing in equations and drawing out conceptual insight from their findings. This is a quality that I don't feel like we see as often as we should in ecology, with the natural history often downplayed.
ReplyDeleteIt's interesting to consider the fact that while S & W do observe an equilibrium in terms of species richness, there doesn't appear to be an equilibrium in terms of community composition, at least yet. Community composition is of course harder to deal with quantitatively, particularly in such a diverse system. In plant succession, there is a fairly defined community turnover, at least in temperate systems, and I suspect you'd see some stabilization in faunal communities over the course of colonization/succession as well, but is this the case?
Martina poses an interesting hypothesis about the role of biotic interactions in succession/colonization. I'd extend it and suggest that early in succession/colonization, particularly following a severe disturbance, abiotic factors (e.g. microclimate, substrate) and individual species traits (e.g. dispersal, stress tolerance) are more important, whereas later in the process biotic interactions and traits like competitiveness and predation/avoidance ability become more important.
The implications of this article seem very broad, since they can extend to any system with geographic barriers, like mountain valleys, coral reefs separated by currents, or forests separated by deserts. It made sense to me that the authors excluded species that treated the islands as if they were not islands. While reading about the colonization patterns of the various species, I kept wondering about what was happening with the associated communities of microorganisms. Assuming that methyl bromide would kill any microorganisms not protected by layers of mangrove bark or in some other way, would these all find their way back to the islands via the other species? How would the colonization curves look for the microbiomes?
ReplyDeleteI'd like to compare Simberloff and Wilson's colonization curves to the humpbacked curve of the intermediate disturbance hypothesis (http://www.jstor.org/stable/3546874?origin=crossref&seq=1#page_scan_tab_contents),
since the axes are comparable. In the empty island experiment, the graphs of diversity begin with maximum disturbance, and progress toward less and less disturbance with time. Could we consider the peak in number of species as a point of intermediate disturbance, since the populations have had time to recover, but have not recovered to the point of equilibrium?
Great summary Schuyler! I too question the feasibility of this kind of experiment today. I feel like you would have to find an island that was already defaunated instead of just doing it yourself. The predictability of the colonizations was the coolest thing to me. It kind of makes sense and doesn't at the same time. It seems like it should be easier to establish yourself if there is already an established order but it also seems like the idea of unlimited resources when you get somewhere would imply that getting there first is best. But I guess this just goes to show how important community is in these systems.
ReplyDeleteAnother study that I fondly remember from my early undergrad and one that is in large part responsible for my interest in biogeography. This paper and others that came from this study have largely shaped our understanding of biogeography as it relates to dispersal and colonization and with such a simple premise that the rate of colonization depends on the distance from the colonizing population and the colonizer’s vagility.
ReplyDeleteThe lack of seasonal effect has definitely made this study simpler. However, I don’t think seasonal influences would necessarily make it harder to draw conclusions; we would still see a predictable pattern of colonization by arthropod taxa and although the timing would be correlated to seasonal hatches, those that are more able to disperse would appear first.
It is also interesting, although not unexpected that the taxonomic composition of the recolonized islands differs from pre-defaunation composition. I would be surprised to see identical taxonomic communities before and after but I would expect to see similar ecological communities.
It is interesting how fast recolonization of arthropods occurred on the "islands." I foresee many figures being drawn in class today. After reading the intro again, I too question how this study would change if competition and predation were included. Maybe somebody has already done this...
ReplyDeleteI second Schuylers statement in saying that we all eat cake today!
As stated above the feasibility of this experiment happening today is questionable. I can understand why the species that were excluded were. However, I wonder how their presence could influence immigration and survivability of colonizing species. I loved ballooning spiders for obvious reasons. The fact they do that is awesome, to watch it happen, even more so and of course many people never heard of it beyond “Charlotte’s web” yes, spiders can ‘fly’ . Reading the patterns of colonization and what species ultimately established in comparison to the original inhabitants was intriguing .
ReplyDeleteI really liked this paper because it bring a lot of the concepts that I learnt when I was an undergraduate in my biogeography class. For instance i was thinking about natural episodes of colonization on island worldwide and the importance of the emergence of new island or defaunation after and eruption event.Nice seminal paper
ReplyDeleteI really liked this paper because it bring a lot of the concepts that I learnt when I was an undergraduate in my biogeography class. For instance i was thinking about natural episodes of colonization on island worldwide and the importance of the emergence of new island or defaunation after and eruption event.Nice seminal paper
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThis is an interesting study to look at how unoccupied spaces become colonized. While I do see it as having broader applications like others mentioned before, I also wonder how this would take form if it was done to Earth as Ali suggested. How would this impact Earth's evolutionary trajectory? I think this study could be developed further (if it can be done at all today) by monitoring the adaptations of the colonizers.
ReplyDeleteThis study also has implications for how species come to "invade" islands. A study on Kodiak Island in Alaska has looked at how a small mammal (I forget which one) came to colonize the island, and how long it had been there. Although many look at this animal as an invasive species the data suggests that the species may have found its way to the island without the assistance of people.