Conservation catastrophes

22 02 2012

David Reed

The title of this post serves two functions: (1) to introduce the concept of ecological catastrophes in population viability modelling, and (2) to acknowledge the passing of the bloke who came up with a clever way of dealing with that uncertainty.

I’ll start with latter first. It came to my attention late last year that a fellow conservation biologist colleague, Dr. David Reed, died unexpectedly from congestive heart failure. I did not really mourn his passing, for I had never met him in person (I believe it is disingenuous, discourteous, and slightly egocentric to mourn someone who you do not really know personally – but that’s just my opinion), but I did think at the time that the conservation community had lost another clever progenitor of good conservation science. As many CB readers already know, we lost a great conservation thinker and doer last year, Professor Navjot Sodhi (and that, I did take personally). Coincidentally, both Navjot and David died at about the same age (49 and 48, respectively). I hope that the being in one’s late 40s isn’t particularly presaged for people in my line of business!

My friend, colleague and lab co-director, Professor Barry Brook, did, however, work a little with David, and together they published some pretty cool stuff (see References below). David was particularly good at looking for cross-taxa generalities in conservation phenomena, such as minimum viable population sizes, effects of inbreeding depression, applications of population viability analysis and extinction risk. But more on some of that below.

David’s former post-doc supervisor, the renown conservation geneticist, Emeritus Professor Richard (Dick) Frankham (and one of my co-authors), has just recently co-written (with George Gale and Charles Fox) an obituary in Animal Conservation, and I’d like to share a few thoughts from that article here.

David’s most recent position was Associate Professor and the Wallace Chair of Conservation at the University of Louisville, Kentucky, USA (from 2009), and he was acting editor of Animal Conservation. At the time of his death, he had published 47 peer-reviewed articles, with many of the most highly cited written while he was a post-doc in Dick Frankham’s lab at Macquarie University from 1999-2001. David had most recently invested a lot of time in Thailand conservation education, mentoring many Thai academics and students in quantitative conservation biology. He was also actively involved with the IUCN/SSC Conservation Breeding Specialist Group (CBSG), and had an active and growing lab developing at the University of Louisville. He is survived by his wife and 6-year-old daughter.

While I have cited many of his papers, one in particular is something I refer to regularly and still marvel at why and how it works. His 2003 Animal Conservation paper entitled The frequency and severity of catastrophic die-offs in vertebrates, is nearly what I’d label a Conservation Classic. I say that because I find it particularly useful when coding a new population viability analysis because it’s the only paper I know that quantifies the probability that a given species will succumb to a ‘catastrophic’ mortality event (catastrophe).

Why is this important? Well, if you’re trying to predict an unbiased and accurate probability of extinction risk, if you forget to include the probability that some whopping die-off that you couldn’t normally predict might occur, you’re going to under-estimate extinction risk by quite a lot, and potentially lose your species of concern well before you thought possible.

Reed and colleagues examined the Global Population Dynamics Database and looked for time series indicating where species had suddenly dropped by 50 % or more in abundance (a whopping ‘mass die-off’ by any standard). They found just such phenomena in 88 vertebrate species and discovered that the probability of such a catastrophe occurring scales with generation length (i.e., how long it takes to start breeding, or the average age of all breeding mothers – there are several definitions of this). In fact, they determined that the chance of a mass die-off is about 14 % per generation; thus, long-lived species tend to experience fewer of them. They also provided a clever severity function where the magnitude of the die-off is modelled as a modified power function.

Most population viability analyses need to incorporate what we call ‘stochasticity’ – a measure of variance around parameters such as survival, fertility and migration that arise due to variation in environmental conditions (process error), demographic ‘chance’ (e.g., the uncertainty of finding a mate and breeding) and measurement error. But predicting the big die-offs, which can often be the deciding factor in whether a species or population goes extinct, hadn’t until the Reed et al. paper, had an underlying theoretical relationship.

If you’re writing a population viability analysis, and you don’t have an incredibly detailed and long abundance time series that specifically identifies catastrophes, then you’d better be invoking Reed & colleagues’ catastrophe-generation relationship, or your predictions will be wrong. I use their results all the time.

So, my sincerest condolences to David’s family and friends, and my professional congratulations on some seminal work. He’ll be remembered for it for a long time, I’m sure.

CJA Bradshaw

References – Some of David’s key papers



3 responses

27 02 2012
Zhigang Jiang

We will remember Dr. David Reed for his contirbution to Conservation Biology. Tribute to David!


23 02 2012

Thank you, Onnie. If you haven’t yet read the Gale et al. obituary, it’s worth the read too (link provided above).


23 02 2012
Onnie Byers

This is a lovely tribute to David. I have shared it with the entire CBSG membership and am sure that many will be as moved by it as I am.Thank you.


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 )

Facebook photo

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

Connecting to %s

%d bloggers like this: