Parts a whole do not make

17 02 2012

I’m particularly proud of our latest paper for three main reasons:  (1) Salva Herrando-Pérez, lead author and contributor-extraordinaire to CB, has worked extremely hard to get this one out; (2) it is published in a really good journal; and most importantly, (3) it’s the very first empirical demonstration over hundreds of species that just because you have a density effect on some vital rate (e.g., survival, fertility, dispersal), this in no way means you have any evidence at all for density dependence at the population level. Let us explain.

Quantifying variation in population size is an important element for explaining and predicting population dynamics. In models where a vital (demographic) rate responds to change in population size, those ‘density-dependent’ relationships are ecologically understood as being demographic signals of trophic and social interactions, such as parasitism, predation or competition for shelter, because the intensity of those interactions varies with population size.

In fact, density-dependent effects reflect the theoretical capacity of populations to adjust growth and rebound from low or high numbers – and so this concept has become an important metric in population management and conservation  (Eberhardt et al. 2008).

For populations where time series are available for both population size and vital rates, we know that the overall population growth rate (r, log of the ratio of population sizes over two consecutive time steps) can be decomposed into the contribution of age-structured fertility and survival rates (Coulson et al. 2005), and that fertility and survival rates do track change in population size in some mammals and bird populations (Owen-Smith et al. 2005; Paradis et al. 2002).

Yet recently in a series of experiments on a perennial plant, Kolb et al. (2010) found that the effects of the number of flowering individuals on a range of fertility rates such as the number of seeds per plant had no effect on the population growth rate. They cautioned that we can make wrong management decisions (so waste limited resources) by measuring vital rates which have no relevance to the dynamics of a population.

So in our (Herrando-Pérez, Delean, Brook & Bradshaw) latest paper entitled: Decoupling of component and ensemble density feedbacks in birds and mammals just published online early in Ecology, we enquired whether Kolb and colleagues’ results would hold for vertebrates (65 bird, 32 mammal species). Interestingly, we found so that indeed they did: the magnitude of density feedbacks on r correlated poorly with the magnitude of density feedbacks on fertility and/or survival rates. In 77 populations whose growth rates were compensating (decreasing with a population boom), 30 % of the survival rates and 40 % of the fertility rates were actually depensating (increasing with a population boom).

Our take-home message is that the management of anthropogenic impacts on populations using change in demographic rates alone is ill-advised, just as managing on the basis of broad population growth rates (from census data) without a mechanistic understanding of the contributions made by its components and environmental variability can lead to suboptimal decisions.

We paraphrase the sentence of one of our anonymous reviewers: “The overall thrust of the work being reported in this paper is important, even though it is largely a message that comes to a conclusion which most of us would feel safe in assuming at the outset. This work turns that assumption into a hypothesis, tests it, and finds it to be a valid assumption – good science”.

Salva Herrando-Pérez & CJA Bradshaw

P.S. We congratulate the Editorial Office of Ecology; it has taken them 3.5 working days from receiving our final manuscript, suggesting and incorporating format changes, and putting the manuscript online. Certainly this is an example of diligency in peer review, setting the standards pretty high across the ecological literature.


References


Actions

Information

One response

17 09 2012
Ecology is a Tower of Babel « ConservationBytes.com

[...] To appreciate the importance of such distinction, bear in mind the theoretical principle that an ensemble density feedback is the result of the compound interplay of all component density feedbacks. Thus, detectable component feedbacks might contribute nothing to ensemble density feedbacks, and [...]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




Follow

Get every new post delivered to your Inbox.

Join 6,496 other followers

%d bloggers like this: