COLLABORATIVE RESEARCH:Were there "too many" browser species worldwide in local faunas of the Early Miocene? Testing a global hypothesis using the Australian fossil mammal record - Christine Janis, Brown University and John Damuth, University of California, Santa Barbara
Is there evidence for a previously unrecognized worldwide change in terrestrial ecosystems approximately 12 million years ago? The PIs hypothesize that a striking pattern that they observed first in the history of the mammal faunas of North America may be one that is actually observable throughout the globe. This project will test this hypothesis by determining whether the same pattern occurred contemporaneously in the fossil record of Australia. Previous research showed that before 12-10 million years ago in North America, many ecological communities contained exceptionally high numbers of species of browsing ungulates (large leaf-eating herbivores)up to three times the numbers seen in vegetationally comparable modern-day habitats. In fact, the numbers of browser species in some places exceeded even those for modern tropical rainforests the present-day habitats with the highest biodiversity of browsers. No other members of these mammal faunas showed any elevated or unusual biodiversity, and faunas younger than 12-10 million years show only "normal" numbers of browser species (by today's standards). Why the older habitats were sometimes able to support what today would be impossibly high numbers of browsers is currently unknown. However, evidence suggests that some European and African faunas from before 12 million years ago also showed "too many" browsers, which in turn suggests a global cause. This project will answer the question of whether there were also "too many" browsers in the well-documented mammalian fossil record of Australia, and, if so, whether this phenomenon disappeared at about the same time as it did in North America. Study of museum collections from Australian localities will allow determination of the numbers of browser species at different times, based on morphology and wear of the teeth. Australia's distinct fauna and environmental history will provide a robust test of whether "too many browsers" was a global occurrence. If it wasn't, then we must look for regional or local causes; but if the global hypothesis is supported then a worldwide phenomenon becomes more likely and these results will guide and motivate the subsequent search for a global cause. The project's results will have implications not only for global environmental history and climate change, but also for any methods that attempt paleoecological interpretations for faunas older than 10 million years that use modern mammal faunas as an environmental proxy. This project will be a new international collaboration and will involve students and postdoctoral researchers.
Change in global climate is of great concern to us today. One of the ways in which we can understand the possible outcomes of future change is to look to episodes of climate change in the past, and investigate what happened at those points in time. In particular, an understanding of what happened to the climate in the past during periods of elevated atmospheric carbon dioxide could be vital for understanding problems that might face us in the future. How can we look at evidence from the past to gain an understanding of biotic events? Herbivorous mammals (e.g., horses, cattle, kangaroos, etc.) are highly sensitive to their environment, to the quality and productivity of the vegetation, which in turn is strongly affected by climate: not only by temperature and rainfall, but also by levels of atmospheric carbon dioxide. Thus the fossils of these mammals have the possibility to act as a proxy for the vegetation, and hence for the global climate. Our past work (i.e., with Co-PI John Damuth) showed that there was a peak in the diversity of large herbivorous feeding on leafy vegetation (i.e., browsers) in North America in the mid-Miocene (~12 – 18 million years ago), representing several times the numbers seen in similar habitats since then, and today. The same time, other evidence shows a peak in global temperatures and also (more controversially) in the concentrations of atmospheric carbon dioxide. We proposed that the change in mammal diversity was causally linked to these climatic changes: but to test this, we had to see if a similar response was seen at the same time in a different biological situation. In this project we used the Australian mammals as a "test case", to see if there was a similar response: the Australian mammals, being marsupials, are quite unrelated to the North American herbivores. Moreover, the habitat, vegetation, and geological situation of Australia is also different, and an additional important fact is that Australia is (and was then) isolated from the rest of the world. If changes in mammalian herbivore communities during the mid-Miocene in Australia echoed those in North America, then we can establish that the fossil mammal record is indeed a proxy for past climatic events, independent of other types of evidence. This would strengthen and add to the uses of fossil record evidence to examine what happened during past periods of climatic change. To investigate this, the PIs travelled to Australia to study museum collections of both living and extinct herbivorous marsupials. We built up a dataset of measurements of the skeletal bones, skulls, and teeth of living mammals: we then used to determine the correlations bony correlates and aspects of behavior such as diet and type of locomotion. These correlates allowed us to reconstruct the diets and behavior of the extinct species, and to see what had changed in their ecologies over time. This was done in collaboration with Australian colleagues at the University of New South Wales (Sydney), and we also supported a postdoctoral student who worked in close collaboration with the PIs. Our results confirmed a significant change in mammalian communities at this mid-Miocene time period, especially in the sequence at Riversleigh in Queensland, indicative of a more vegetational environment, similar to that observed in North America. However, because of the differences in the Miocene environments between North America and northern Australia at that time (woodland savanna versus tropical-type forest, respectively), and the different biological and environmental histories of the two continents, the signal was quantitatively different. Our studies also generated other data and techniques, such as combining tooth wear with features of the skull anatomy to determine diets, and use of the anatomy of the (calcaneum) a major ankle bone to determine locomotor behavior, which will be of broad use to Australian researchers in future projects. The last time that atmospheric carbon dioxide levels (and global temperatures) were as high as the levels that are currently projected for the end of this century was some time in the Late Eocene (43 million years ago). Current conditions are beginning to resemble those of the mid-Miocene, 12 – 18 million years ago (>400 ppm CO2, several degrees of temperature rise). Thus going backward into geologic time is equivalent to looking into our future, except on a vastly different timescale. Study of Cenozoic climatic changes, vegetation responses, and mammal responses shows us the earth under non-anthropogenic climate change, which gives us perspective on the range of conditions we might experience in this century.
