Ecological analysis, 75 million years in the making

The fossilised bones of a diverse assemblage of herbivorous dinosaurs provide clues about the feeding ecology of these extinct creatures, but suggests more evidence is needed to find out how such diversity was able to be maintained.

Parasaurolophus walkeri restoration_Steveoc 86_Wikimedia Commons ccThe fossil record is a capricious thing; we take from it only what chance dictates is preserved through time. Such patchiness poses unique problems for paleoecologists like Jordan Mallon from University of Calgary and Canadian Museum of Nature, who is attempting to infer ecological relationships among species that coexisted millions of years ago. Unlike conventional ecological analysis of living specimens, studying the fossilised remains of dinosaurs means that quantitative analysis on large sample sizes is extremely difficult:

This is especially true of terrestrial vertebrate fossils, which suffer from severe preservational biases.

Mallon, together with colleagues from Royal Ontario Museum and Cleveland Museum of Natural History outline their latest research in BMC Ecology today, investigating whether the huge diversity of herbivorous dinosaurs found in the late Cretaceous landmass Laramidia – present day North America – can be explained by differences in feeding height. Here, they utilise the most diverse dinosaur assemblage currently known, the Dinosaur Park Formation in Canada, to create a statistically robust dataset from which to test ecological hypotheses:

The upshot is that many taxa are known from just a handful of specimens–sometimes only one–whose variation cannot be studied statistically. With respect to dinosaurs, only a handful of localities throughout the world yield enough fossils to facilitate quantitative palaeoecology.

This huge variation is something of a conundrum, since compared to what we understand of modern animal populations, such high diversity within a spatially restricted area would have imparted huge resource pressures on the vegetation base that would be needed to support them.

Two possible explanations have been put forward to explain this: one posits that plant resources may simply not have been limiting in this ancient environment, either because of the low nutritional requirements of dinosaurs, non-limiting availability of plants in an atmosphere of elevated CO2, or because predation from carnivorous species kept herbivore densities low. Alternatively, a second hypothesis suggests that dietary niche partitioning – the separation of competing species into different patterns of feeding – allowed such a large number of species to coexist in this region.

Such partitioning of resources is seen among extant species of browsing ungulates in the African savannah, with giraffes and antelope species effectively feeding at different heights within the tree canopy. Although this all appears very polite, this separation of the resource base reduces competition in harsh seasonal environments when access to food may be scarce.

But how do you quantify whether this also occurs in your study organism, when it has ceased to be extant for the last million-or-so years?

Adapted from Fig 2 Mallon et al BMC EcologyFor this, Mallon and the team took anatomical measurements from fossil specimens prised from the same geological strata, and from these inferred the maximum height at which each species could feed. For bipedal species such as the duck-billed hadrosaurs, the application of simple trigonometry could be used to infer just how high these creature could stretch for a meal.

Among the three Families studied, it was only these tall bipeds that displayed any significant differences in feeding heights, being able to browse generally at a height of around two meters, or rise up to a maximum of five meters if they needed to reach taller vegetation – in much a same way that modern day giraffes might. Although the more squat ceratopisids were restricted to browsing scrub at a height of around one meter, they only did so at a slight height advantage to their quadrapedal competitors the ankylosaurs.

Despite this, evidence from this study that differences in feeding height may have helped facilitate coexistence among these families is not strong, although the possibility remains that other proxies of dietary partitioning etched in the fossil record – such as differences in skull morphology or dental microwear – may yield more clues.

The old and the new

Evidence from fossilised dental remains also informs the only other study published by BMC Ecology on the ecology of ancient species, applied to investigate dietary partitioning in extinct horses. Although traditionally the fields of conventional ecology (or neoecology) and paleoecology have  rarely mixed, the broad scope of the BMC-series lends itself well to consideration of unconventional connections. We certainly welcome robust quantitative studies that try to understand the interaction of organisms with their environment – regardless of the timescale.

Asked why he thinks such a divide might exist between these two fields, lead author Mallon said:

Conceptually, I think neoecology and palaeoecology are strongly connected in that they both seek to understand the patterns and processes that shape the biosphere. The only difference is that the former involves the study of present ecosystems, whereas the latter involves the study of past ecosystems. This inevitably leads to methodological differences, but the underlying intent is really the same.

Practically speaking, I think neoecology and palaeoecology are still worlds apart. On this point, I can do no better than to cite a recent paper by Valentí Rull, who suggested a variety of psychological, methodological, and semantical reasons why this might be. The upshot is that neoecologists and palaeocologists tend to frequent different conferences and publish in different journals, which only deepens the divide. Certainly, there’s more collaboration between Quaternary palaeoecologists and neoecologists because missing data and non-analogous ecosystems aren’t such a problem. But these issues only become increasingly more prohibitive as one goes further back in time.

View the latest posts on the BMC Series blog homepage