Dynamics of regional distribution: the core and satellite species hypothesis.
Author:
Ilkka Hanski
- One of the most influential ecologists of recent years, developer of the metapopulation theory.
- born in 1953 in Lempäälä, Finland, died in 2016.
- Studied biology at the University of Helsinki and gained his doctorate, on the community ecology of dung beetles, from the University of Oxford in 1979.
- Did research around the globe, from Sweden to Borneo.
- Recommended book by Hanski: Messages from Islands: A Global Biodiversity Tour, where he presents his ideas on biodiversity at the time that he gives an account of his field work.
Background:
At the time this paper was published, more attention was starting to be paid to distribution and spatial dynamics of independent populations. Some authors suggested that these spatial dynamics could explain most of the community patterns, contrasting with the approach of MacArthur (competition or other interactions?). Nonetheless, no overall explanation was available for the observations that communities include species widely distributed and species with a patchy distribution, abundant and rare species as well as the occurrence of communities composed of species in which the niche is well spaced-out, while other communities are composed of similar species. In this paper, Hanski tries to explain these observations by modifying a model proposed by Levins for describing species occupancy (number or proportion of sites occupied by a species) and asks whether there is a single equilibrium point for the distribution scale approached by most species (Levins prediction) or if there are two points at both extremes of maximal distribution and regional extinction.
Methods:
Definitions:
vAbundance: the number of individuals at a local population site.
vDistribution: number of suitable population sites occupied by the species.
- Hanski first used four datasets on invertebrate abundances and distribution to show that generally these two parameters are positively correlated.
- When Levins model is modified by including the correlation between distribution and abundances, its predictions change drastically.
- Independent datasets are used to test some of the implications of the modified model.
- Finally, the biological implications are discussed.
Results
- The four analyzed datasets showed a correlation between abundance and distribution.
- The modified model had two implications (predictions). First, extinction probability decreases with the number of occupied sites. Second, the frequency distribution of occupancy (distribution) will be bimodal with peaks at each extreme of the gradient, near maximal distribution and near regional extinction.
- The first implication was assessed using three different datasets on distribution and probability of extinction for insects in mangrove islands, leafhoppers in meadows and molluscs in ponds. All datasets showed a negative relation between these two variables.
- The second implication was assessed by plotting the frequency distribution of occupancy for the same data on island insects, data on forest beetles and plants. The three distributions showed a bimodal pattern, according to the modified model expectations.
Discussion
- The modification of Levins model results in a very different conclusion about the way stochastic variation in extinction and colonization would influence communities (by pushing ones to extinction and others to wide distribution).
- The species at both ends of de distribution frequency are called core (reaching complete distribution) and satellite (near extinction) species.
- Other models might result in the same distribution, thus more testing is required.
- The main difficulty in testing it, is the definition of the population sites, which should include suitable sites for all species.
- The idea of bimodality in species distribution was proposed by previous authors.
- This hypothesis has implications for species interactions and their outcomes. It is predicted that core species will present strong interactions, which will result in more evenly spaced niches among these species.
- More data needed to assess the value of this hypothesis.
I think it's interesting that the authors showed the individual communities in separate figures and that each showed similar relationships between abundance and number of sites occupied. The same goes for the bimodal trend seen in the frequency distribution. Even though the authors claim that more data is needed to validate their hypothesis, the consistency of the results shown is impressive.
ReplyDeleteIt is important to have sites of the same area when making comparisons between core and satellite species. I would like to explore more about how to determine the specific habitats and what factors are used to classify it for each species that share those habitats.
ReplyDeleteI agree with Debra. I was surprised by how the data adjusted to the model. I wonder if further modifications have been done in recent years to the model he proposed, as it was a modification of another model.
ReplyDeleteNear the end of the paper, the author talked about interspecific competition is important in structuring communities so core species should be spaced out than satellite species. The closer the competitors, the more probability one of them will become a satellite species. It's a pretty cool concept I never thought of before.
ReplyDeleteThis is one of those classic papers that describe and discuss population dynamics and their relationships using abundance and distribution data. The general pattern of their relationship is interesting and makes more sense. But the influential factors are still uncertain. Also, I wonder that how much of these ideas are still valid and used for prediction analysis. I would like to read or learn more about these models. I don't know how much other studies support the hypothesis satellite species are at high risk of extinction. I also think that it could naturally evolve into a new species.
ReplyDeleteI though it was neat that the author introduced this bimodal distribution, but also allowed for flow and change under his slopes. He says that satellite species can become core species and vice versa and I would love to see an example of that discussed. Especially since he limits the model by only comparing similar species- what pressures would cause satellite A to become core A when they are of the same genus etc.
ReplyDeleteI agree with Willow that it would be interesting to read about species that transition from core to satellite (or vice versa). I feel like the author's definitions of "core" and "satellite" are different from the ones I've heard before, and I would like to see how species can move between the two. I also would be interested in exploring how people determine the habitat/niche requirements of species. While I agree that closely related species seem to have the same requirements, I also think this varies largely by genus.
ReplyDeleteThe authors threw out ideas then had an explanation following to justify their methods, thoughts and actions. It made this much easier to follow. I appreciate them adding figure four and it makes sense to add a graph as such, but it was rather confusing to identify quickly. I feel like the experiment would have provided more "useful" results had there been either different continents' species tested or completely different locations all together. Overall, great experiment and nice read!
ReplyDeleteI could use a distillation of assumptions (1) and (2) in the right column on page 498. Other than the jargon, I thought it wasn't too bad of a read, interesting topic as well.
ReplyDelete