The Grandeur in this View of Life

Eye on Research: Cretaceous Extinction Cascades



We all know that an asteroid impact* ended the age of the dinosaurs roughly 65.5 million years ago. That event, known as the K-Pg extinction**, marks the cataclysmic demise not just of the non-avian dinosaurs (yes, that was a dinosaur you ate on Thanksgiving!), but of a large number of other living things, including the pterosaurs and pleisiosaurs, as well as many lineages of plants and marine and terrestrial invertebrates like insects, cephalopods, bivalves, and echinoderms. In terms of overall destruction***, the K-Pg extinction was one of the worst extinction episodes in the history of life, second only to the “Great Dying” at the Permian-Triassic boundary. The current consensus is that the asteroid impact threw so much debris and ash into the atmosphere that it greatly reduced the incoming sunlight, which in turn hindered photosynthesis and cooled global climate for years to decades. The resulting declines in plant productivity then cascaded through the food chain, leading to the extinction of herbivores and subsequently to the carnivores that depend on them. So long, T. rex!

But like any other event in life, even the outcome of an asteroid impact may depend in part on context. Did the impact, itself, doom the dinos to extinction, or did the particulars of the interactions among species play some role? Recently, Jonathan Mitchell and his colleagues addressed this question by comparing extensive fossil assemblages from the last few million years of the Cretaceous, just before the impact (the Maastrichtian age) to those of previous Campanian age, thirteen million years earlier. They used a mathematical model to investigate whether the structure of Maastrichtian food-webs made them particularly vulnerable to the kind of disturbance produced by the impact (i.e., declines in plant productivity).

As you can imagine, figuring out who-eats-who is a difficult proposition, even in extant communities, and while gut contents do occasionally occur in fossils, ancient food-webs are even more difficult to ascertain, especially when they involve up to 92 different animal species. But while we can rarely be certain that any extinct species ate any other particular species, we can have more confidence in assigning species to particular feeding guilds, based on their anatomy and their size. For example, while we don’t know exactly which plants it found tasty, Triceratops is definitely a Very Large Herbivore (let’s call it the VLH guild).

Once all of the species at a particular site were assigned to guilds, Mitchell and his colleagues used a computer simulation to assign the feeding connections among the species in the different guilds (see schematic above), based on the “connectedness” of existing food-webs. They dealt with uncertainty in the feeding relationships by repeatedly drawing random connections among the particular species in different guilds to make a large number of sample communities for each site. This approach, which is focused on general aspects of food-web structure rather than a detailed characterization of any particular community, allowed them to meaningfully compare the communities despite uncertainty about the particulars of who-eats-who.

For each randomly sampled food-web, they then simulated the asteroid impact by reducing the productivity of plants and algae and tracked the “cascade” of declines and extinction as they wound their way through the complex network of feeding connections. Their approach is particularly compelling because it is not limited overly simplistic linear chains of causation (e.g., plants decline -> herbivores decline -> carnivores decline) and permits a richer set of indirect interactions (e.g., plants decline -> herbivore A declines -> carnivore B eats more of herbivore C -> herbivore A recovers due to reduced competition from herbivore C.) By compiling sets of simulations from seven different Maastrichtian sites and ten different Campanian sites, they could then ask whether differences in food-web structure affected the robustness of a community in the face of a cataclysmic loss of plant productivity.

They found that the later Cretaceous communities were indeed more fragile, suffering greater degrees of simulated extinction at lower disturbance levels. This is certainly not to say that an asteroid impact earlier in the Campanian would not have resulted in a mass extinction – it certainly would have, but the degree of extinction, and the particular taxa that disappeared, may have been different. Interestingly, they also point out that increases in the average diversity of several guilds from the Campanian to the Maastrichtian, including the dinosaur-dominated VLH guild, was actually associated with a decrease in the robustness of the community. They hypothesize further that the importance of the VLH species was due to the fact that they fed many other guilds as they grew, from the small predators that cracked their eggs to the large carnivores unafraid to confront a fully-grown Triceratops.

This study shows that, even in an apocalypse of planetary proportions, context matters. The structure and diversity of ecological guilds, the numbers and functional types of species present, determines, in part, which species survive and which go extinct. This context-dependent complexity of ecological systems is what makes them so difficult to understand and so deeply fascinating. It also means that as we move forward with an extinction crisis of our own making, we are going to have to consider the interactions among species if we hope to mitigate our own impact.

*The evidence is drawn from the global iridium layer, characteristics of the boundary deposit, and the discovery of the Chicxulub crater in present day Mexico. Other accessory causes might include volcanic activity in the Deccan Traps of present day India.

**K-Pg marks the boundaries between two geological periods (or strata), the Cretaceous (K, C was already taken… not really, it is from the German name, Kreidezeit (chalk-time)) and the Paleogene (Pg). It is also known as the K-T extinction, with the T representing the Tertiary period. However, that nomenclature has been discarded by paleontologists and geologists. Bye, Tertiary! Thanks to the International Commission on Stratigraphy for changing the names of periods to make things extra confusing!

***In terms of the proportion of documented taxa going extinct

The research of Jonathan Mitchell and his colleagues, Peter Roopnarine and Kenneth Angielczyk, “Cretaceous restructuring of terrestrial communities facilitated the end-Cretaceous mass extinction in North America”was published November 13, 2012 in The Proceedings of the National Academy of Science of the United States (2012, vol. 109, pages 18857-18861.) A promotional blurb is also available over at Science Daily.

For an up-to-date review of the strong evidence for the role of the Chicxulub impact, see the paper “The Chicxulub asteroid impact and mass extinction at the Cretaceous-Paleogene Boundary” by Peter Schulte and his colleagues in the journal Science (2010, vol. 327, pages 1214-1218.)


3 thoughts on “Eye on Research: Cretaceous Extinction Cascades

  1. I am really digging this blog – It’s easy to find blogs and papers on individual fossils or lagerstattes, but I haven’t seen anything on the work that’s been done around larger scale research into paleo-ecosystems until now. Thanks for the work you’re putting into these posts.

    If you ever want an idea for a post, I’m really interested in the challenges of such reconstructions, particularly as the fossil record is so imperfect (I’m thinking of species that are known only from individual fragments or single speciments) – what are the reasonable assumptions that can be made in when doing such work, and what are the limitations? I imagine that boreal environments would be a challenge.

    • Thanks! I have been on a bit of a paleo kick lately, even though it is not my research area per se. The larger scale perspective really is what interests me, though, so keep checking in.

  2. Thanks for this reference, Drew. What they find is very consistent with some work we’re doing on the K-Pg transition right now – looks like extensive climate change was occurring pre-meteor impact, possibly driving some of this ecosystem reorganization. I’ll hopefully have more for you soon!

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