How to predict marine biodiversity

26 07 2011

One of the most important components of conservation ecology is arguably the focus on robust methods to predict ‘biodiversity’. This covers everything from detection issues (whether or not a species is in a particular area), species distribution models (to predict where a species should be given habitat and/or physical attributes), climate change predictions, to reserve design algorithms (to assess whether we are protecting what we think we are protecting).

It might seem a bit strange to the uninitiated that we have to spend so much time trying to figure out what’s there. Surely, one just goes to the area of interest and does a few quick surveys? Wouldn’t that be lovely; the truth is that most species are, in fact, rare, and the massive areas we must usually survey tend to preclude complete coverage. This is why experimental design and statistical techniques are so advanced in our discipline – to account for the probability of missing what’s actually there, and to estimate what should be in areas we haven’t even looked in.

One way we deal with some of these issues is through the use of what we call ‘surrogates’. In an ecological context, a ‘surrogate’ is a component of the entire biodiversity that one can more easily measure than others, that is used as an indicator of the greater biodiversity in a particular area. For example, let’s say a particular species is relatively easy to survey, say, a tree that you could identify from aerial photos. If you could establish that many types of species live on or near such trees, then you’d have a relatively easy way of estimating biodiversity patterns over an entire forest just from inventorying that particular tree’s distribution. It’s rarely that straight-forward, but you get the general idea.

So it’s with great (and modest ;-) pleasure that I highlight a paper on this very subject of surrogacy in the marine realm. Interestingly, surrogacy studies have largely focussed on terrestrial questions to date, probably because data are generally more forthcoming compared to marine ecosystems; however, good surrogates are generally much more essential in marine ecosystems for that very same reason.

My post-doc, Camille Mellin, recently published a meta-analysis paper on marine surrogacy to determine if any generalities exist (see also a previous post on Camille’s exciting work). The paper appeared in PLoS One last month: Effectiveness of biological surrogates for predicting patterns of marine biodiversity: a global meta-analysis.

Using a Bayesian approach (incorporating expert opinion as informative priors), we determined which types of surrogates perform the best (i.e., give the best predictions of overall biodiversity patterns). We examined four major categories of influence on surrogate effectiveness: spatial scale, habitat, surrogate type, and statistical approach.

  • spatial scale categories ranged from small (< 10 km), to medium (10 – 100 km), to large (> 100 km)
  • habitat types were: tropical reefs, temperate reefs or soft bottoms
  • surrogate types were: cross-taxa, higher-taxa or subset-taxa (see figure)
  • statistical approaches were: univariate congruence, multivariate congruence, or representation

Overall, and rather unsurprisingly, we found that surrogate effectiveness is generally less than that assumed, but that there a few things one can do to improve predictive capacity. First, using higher-taxa surrogates (i.e., an entire class or genus for a group of species) was much better than using cross- or subset-taxa surrogates (again, see figure). One should also avoid representation-based surrogates (i.e., those that select a set of sites based on a surrogate and sum the representation of the target within the selected set), and those applied across broad spatial scales. Finally, tropical reefs had the poorest performance, most likely because of the high diversity and functional complexity found there relative to most other marine habitats.

So, if you’re designing a reserve based on surrogate data, make sure you account for the uncertainty in your biodiversity predictions based on the type of surrogates used. It could make the differences between extinction and persistence.

CJA Bradshaw

ResearchBlogging.orgMellin, C., Delean, S., Caley, J., Edgar, G., Meekan, M., Pitcher, R., Przeslawski, R., Williams, A., & Bradshaw, C.J.A. (2011). Effectiveness of Biological Surrogates for Predicting Patterns of Marine Biodiversity: A Global Meta-Analysis PLoS ONE, 6 (6) DOI: 10.1371/journal.pone.0020141



3 responses

9 05 2012
No more ecology «

[…] with – the use of ‘surrogates’ in marine ecology – that is, using a species/taxon to predict the distribution of many more species. We have also done some work to predict coral reef fish diversity using little more than the […]


1 08 2011
Camille Mellin

Thanks Hedley for this great insight into other, probably more applied studies about biological surrogates. Indeed the practicality of surrogates was initially included in the scope of our paper, in particular their cost-effectiveness – however there was so much to say with the fundamental aspect of their definition and rationale that we chose to restrict the scope of the paper.

Grantham et al (2010) is probably the most relevant of these citations in our case, unfortunatly our study was submitted before its publication. A great piece of work nevertheless.

Camille Mellin.


26 07 2011
Hedley Grantham

I did a fair amount of thinking on this and produced some studies that might be of interest. Their focus is on non-marine areas but the message is likely to be the same. Grantham et al. (2010) found that differences in the way surrogates are assessed will have an influence on their perceived effectiveness, questioning the value of all representation-based surrogacy tests. Grantham et al. (2008; 2009) argue that, actually, surrogay testing methods are not asking the right question, from a practical perspective. Much more useful is the question, “should I invest in more data or not given the data I currently have.” Surprisingly Mellin et al. don’t mention any of these studies.

I still think these more theoretical surrogacy studies are interesting and important and congratulate Mellin et al. on a great study. This was really needed for marine areas. Sorry to plug my own resesarch.

Hedley Grantham

Grantham H.S., Moilanen A., Wilson K.A., Pressey R.L., Rebelo T.G. and Possingham H.P. (2008) Diminishing return on investment for biodiversity data in conservation planning. Conservation Letters, 1, 190-198.

Grantham, H.S., Wilson, K.A., Moilanen, A., Rebelo, T. and Possingham HP (2009) Delaying conservation actions for improved knowledge: how long should we wait? Ecology Letters. 12, 293-301.

Grantham H.S., Pressey R.L., Wells J.A., Beattie A.J. (2010) Effectiveness of Biodiversity Surrogates for Conservation Planning: Different Measures of Effectiveness Generate a Kaleidoscope of Variation. PLOS One 5: e11430. doi:10.1371/journal.pone.0011430


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