Research
Early herbivores
The origin of herbivory during the Carboniferous and Permian is something that has always fascinated me. At least eight times over a period of about 60 million years, terrestrial vertebrates evolved a plant-based diet. Each had their own ways of dealing with the requirements of such a diet, ranging from the toothless beaks of dicynodonts, to the molar-like teeth of diadectids, to my personal favourites the captorhinids, which evolved up to 11 rows of bullet-shaped teeth to form a grinding surface.
During my early career I have looked at rates and modes of evolution in individual herbivore families like the captorhinids and the edaphosaurids. I then received funding s to expand such analyses to the full range of Paleozoic herbivores, comparing their evolutionary patterns to those of the carnivores and examining their responses to plant evolution and mass extinctions.
During my early career I have looked at rates and modes of evolution in individual herbivore families like the captorhinids and the edaphosaurids. I then received funding s to expand such analyses to the full range of Paleozoic herbivores, comparing their evolutionary patterns to those of the carnivores and examining their responses to plant evolution and mass extinctions.
Oudenodon, one of the beaked anomodonts. Photo taken by me at the Museum fur Naturkunde in Berlin
Diversity
My early research was more focussed on examining changes species diversity through time and the completeness of the fossil record. For my master’s project, I examined the fossil record of Mesozoic birds, with particular focus on quantifying the completeness of the specimens themselves. This bias had previously been neglected, but is actually crucial as it not only informs us about how preservation potential changes through time but also will affect how easy it is to accurately identify and name species.
Once you have examined how good your fossil record is, the next (and more interesting) step is to actually try to assess how many species there were and how this number has changed through time. Unfortunately, as palaeontologists, such studies are fraught with uncertainty based around the incompleteness of the fossil record and our incomplete sampling of what is available. A variety of statistical methods exist allowing us to correct for the uneven sampling, and much of my PhD work was focussed on applying such methods to the record of Paleozoic tetrapods. I have also been involved in testing the accuracy and reliability of the various methods used devised a novel simulation approach to test the performance of different implementations of the residual diversity estimate, one of the statistical methods I mentioned for correcting for uneven sampling.
Once you have examined how good your fossil record is, the next (and more interesting) step is to actually try to assess how many species there were and how this number has changed through time. Unfortunately, as palaeontologists, such studies are fraught with uncertainty based around the incompleteness of the fossil record and our incomplete sampling of what is available. A variety of statistical methods exist allowing us to correct for the uneven sampling, and much of my PhD work was focussed on applying such methods to the record of Paleozoic tetrapods. I have also been involved in testing the accuracy and reliability of the various methods used devised a novel simulation approach to test the performance of different implementations of the residual diversity estimate, one of the statistical methods I mentioned for correcting for uneven sampling.
A graph of extinction rates during the early Permian in Texas. The red dotted line marks Olson's Extinction, a mass extinction of terrestrial vertebrates. Modified from Brocklehurst (2017), PeerJ
Olson's Extinction
A particular interest of mine has been the event about 270 million years ago where the fauna dominated by amphibians, large predators like Dimetrodon and herbivorous "pinhead" caseids, was replaced by one dominated by sabre-toothed gorgonopsians, armoured reptiles called pareiasaurs and the beaked herbivorous dicynodonts. Way back in the 1980s, the famous Chicago palaeontologist Everett Olson suggested that this transition was accompanied by a mass extinction (later dubbed Olson's extinction).
Unfortunately the fossil record around this time is particularly poor, hampered by poor-quality fossils, an out-of-date taxonomy and information only being available from a very restricted set of geographic regions. It is probably not surprising, therefore, that the reality of Olson's extinction is a point of great debate, with some suggesting it is an artefact of the terrible record. I've made it one of my missions in life to provide an answer to this dilemma.
So far, all my researches have suggested Olson's Extinction is a genuine extinction event, and was in fact one of the more severe catastrophes during the Permian (at least in the terrestrial realm; no one seems to have noticed it in the marine realm). I have so far been able to put to rest certain claims: it is robust to correction for sampling; it is not an artefact of poor specimen quality, nor of preferentially sampling certain latitudes, nor of using too-large time bins in analyses (and thus lumping several million years worth of extinction into one event). Obviously there are still question marks, but one can only work so fast.
Unfortunately the fossil record around this time is particularly poor, hampered by poor-quality fossils, an out-of-date taxonomy and information only being available from a very restricted set of geographic regions. It is probably not surprising, therefore, that the reality of Olson's extinction is a point of great debate, with some suggesting it is an artefact of the terrible record. I've made it one of my missions in life to provide an answer to this dilemma.
So far, all my researches have suggested Olson's Extinction is a genuine extinction event, and was in fact one of the more severe catastrophes during the Permian (at least in the terrestrial realm; no one seems to have noticed it in the marine realm). I have so far been able to put to rest certain claims: it is robust to correction for sampling; it is not an artefact of poor specimen quality, nor of preferentially sampling certain latitudes, nor of using too-large time bins in analyses (and thus lumping several million years worth of extinction into one event). Obviously there are still question marks, but one can only work so fast.
Biogeography
During my first postdoctoral position, the focus of my research shifted towards examining diversity trends through space rather than through time. For me, one of the most interesting things I have found is that, unlike the present day, where most species are concentrated in the equatorial latitudes, during the Permian peak diversity was actually in the temperate regions. I have also started developing some new methods for examining how provincial or cosmopolitan species were at different points in time. While trying to find a way to assess whether faunas were significantly different from each other, I accidently developed a new method of correct for a important bias in the study of provinciality: that it is easier to sample very abundant species, and so if there are very uneven abundances it is easier to sample the very common species in multiple localities, and thus lower your estimates of provinciality.