Posted by: oikosasa | April 23, 2013

The zombie killing spree continues

Look up, zombies are all around us nowadays! Even within science! In the Early View paper  “A critical analysis of the ubiquity of linear local–regional richness relationships“, Thiago Goncales-Souza and colleagues  goes on a zombie-killing adventure. Here is there summary of the paper:

Recently, ecologists have gone on a zombie killing spree, started by a blog post of Jeremy Fox here at the Oikos blog (link here). Dr. Fox defined zombie idea as having “survived decades of attacks from the theoretical and experimental equivalents of chainsaws and shotguns, only to return to feed on the brains of new generations of students.” He featured other zombie ideas such as “neutral = dispersal limitations” and the unimodal diversity productivity relationship. The post on his original zombie idea actually resulted in a peer-reviewed publication (Fox 2013) that, in his own words, has “No zombie jokes or other inflammatory rhetoric in it. I leave it you to judge if that makes the paper better or worse than the blog posts.” (link here).

We think that our publication fits into this zombie killing tradition. One of the most fundamental interests of ecologists since the early development of the ecological theory is the understanding of (potential) processes that drive local community structure. At a fundamental level, communities are structured by a combination of local environmental and regional processes (Ricklefs 1987).  The easiest and most traditional way to test whether regional or local processes affect local community structure consists of regressing local against regional species richness (the famous LSR-RSR relationships). The argument goes that communities controlled by regional processes are considered unsaturated, whereas communities controlled by local processes (such as species interactions) are considered saturated. Sadly enough, this argument survived indeed decades of attack (D. Srivastava, F. He and collaborators, and H. Hillebrand).

Sadder still, this method has kept on feeding on the brains of new generations of students, since it is used and featured prominently in ecology textbooks such as Begon et al. (Ecology: from individuals to ecosystems), Krebs (Ecology: the experimental analysis of distribution and abundance), and Ricklefs (The Economy of Nature and Ecology). Probably every ecology student has heard about the ubiquity of regional processes as drivers of local community structure using this method.

In addition to these fundamental problems, Szava-Kovats and collaborators in Oikos showed that the statistical test for detecting the linearity of the relationship is biased towards linear relationships, and moreover provided an unbiased method (called log-ratio method). In this Forum paper, we reevaluated the evidence for the ubiquity of linear LSR-RSR relationships by comparing the biased conclusions with the unbiased method, and found:

  1. In 113 relationships found in 47 studies, 70% and 30% were considered unsaturated and saturated, respectively, when using the biased method. However, by using the unbiased log-ratio method we found no prevalence of either unsaturated (53%) or saturated (47%) communities (Figure 1);
  2. 40% of the studies using the biased method were misclassified (i.e., mistakenly found an unsaturated pattern when it was saturated or vice-versa) and thus reached wrong conclusions.
  3. 50% of the examples of LSR-RSR relationships used in four (classic) ecology textbooks were misclassified.

Our conclusions thus add a new weapon in the arsenal against the zombie idea of interpreting local-regional relationships based on the linearity of the relationship. We showed that the last argument in favor of the method (its ubiquity) was based on a biased statistical method. We argue instead that future studies should consider the reciprocal interactions between regional and local such as those that take advantage of the metacommunity theory to understand the relative influence of regional and local processes on local community.

While studies of LSR-RSR relationships were instrumental to the development of the ecological curriculum, we hope that we can now finally put this zombie idea to rest and move on. As a final note, one of the reviewers for this manuscript was … non other than Jeremy Fox, zombie slayer par excellence.


About these ads


  1. I’m flattered to be taken as a model to emulate!

    I’ll be posting on Dynamic Ecology tomorrow with my own take on the zombie idea of local-regional richness relationships. Short version: the last good reason anyone had for studying the local-regional richness relationship is “we need to explain the pattern that most local-regional richness relationships are linear”. Now that there’s no pattern anymore, it’s hard to see why anyone would still want to bother studying local-regional richness relationships anymore. After all, there are lots of other ways to study spatial community dynamics, interplay of processes operating at different spatial scales, determinants of local species richness, etc. And many of those other ways aren’t nearly as difficult to interpret as local-regional richness relationships are even in the best of circumstances.

  2. [...] at Oikos Blog, Gonçalves-Souza et al. have a fun summary of their paper which flatters me by placing their work within my own zombie-slaying tradition. I [...]

Leave a Reply

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

You are commenting using your 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



Get every new post delivered to your Inbox.

Join 3,008 other followers

%d bloggers like this: