Research – Old Page

We link fossil records, historic specimens, and modern surveys.

Species, assemblages, and landscapes have histories. Fossil records and historic specimens in natural history collections can enrich studies of the modern day by augmenting the temporal and spatial extents of datasets. Together, they provide more robust statistical power and can uncover otherwise hidden evolutionary patterns (eg, island biogeography) and ecological experiments (eg, mesopredator release, defaunation). We can assess the accuracy of models built on modern data by looking to the fossil record for alternative training datasets. These long-term datasets serve as windows into past responses to an array of climatic and anthropogenic perturbations, yielding multiple baselines that can be used in a “conservation paleobiology” framework.

A governing theme of our work is using lessons from the past to inform conservation decision-making in a changing world.

Research Theme 1. Evaluating ecological mechanisms of species persistence in the past and present, with an eye towards the future.

The fossil record tells us that today’s vertebrate communities are a non-random subset of  communities that existed only ~15,000 years ago. Many now extinct species persisted through dramatic climate changes associated with deglaciation, only to disappear several thousand years later, often following the arrival of humans on continents and islands. Historic records further document the extinction and range contraction of species even less than 100 years ago. While many paleontologists and conservation biologists focus on why species went or are going extinct, I am driven by the question: why did some species survive, and what makes them ecologically unique?

I study the fossil and historic records of extant “survivors” in combination with modern assessments to reconstruct their ecological trajectories through time. Dietary flexibility, omnivory, and a generalist niche are qualities often associated with survival. Are these new qualities a response to life in the Anthropocene, or are they carried over from Pleistocene and Holocene niches?


Reconstructing the diet of California’s lost grizzly bears. In partnership with UCSB and the California Grizzly Research Network, I am using stable isotope analysis and radiocarbon dating to describe the Holocene ecology of grizzly bears (Ursus arctos) in California. Grizzly bears survived megafaunal extinctions in North America, and were only recently extirpated from the state in the 1920’s due to targeted persecution. Carbon and nitrogen isotopes from bear bones and their potential foods can uncover trophic interactions of the past. I am particularly excited to investigate evidence of niche contraction between pre- and post-European bear populations.


Dietary flexibility, a generalist niche, and a medium body size promote survival from the Pleistocene to present in the Hispaniolan Solenodon. The islands of the Caribbean experienced the largest post-glacial mammal losses, with >80% of terrestrial mammals now extinct. One survivor is the Hispaniolan Solenodon, an evolutionarily distinct insectivore that diverged 70 million years ago from all other living mammals. Using the fossil record, I have shown that solenodons passed through two body-size biased extinction filters, being “just right” in  size to avoid predation. Using modern surveys and diet studies, I have shown that solenodons can persist in mosaic agriculture-forest environments typical of disturbed habitats in the Caribbean. A combination of stable isotopes and DNA metabarcoding confirm that the solenodon is a flexible insectivore and feeds across multiple trophic positions.

Research Theme 2. Investigating the causes and consequences of species extinctions. 

I establish baselines of diversity to inform our understanding of fundamental ecological and evolutionary processes and contextualize modern conservation priorities. I do so through the excavation of Pleistocene/Holocene deposits, radiocarbon dating of faunal remains, and literature meta-analyses. Without knowing who went extinct when it is impossible to accurately identify the extinction filters that shaped modern fauna. Further, knowledge of past community structure and function sets the foundation for asking how extant species responded to the ecological consequences of extinction.


Excavations in the Caribbean of caves and asphaltic sediments. I work closely with the Museo Nacional de Historia Natural of the Dominican Republic to excavate caves on Hispaniola. Our excavations of Cueva del Mono have yielded newly described monkey remains, ground sloths, tortoises, giant birds of prey, and remains of still extant solenodons and hutias. Radiocarbon dates place this cave at 40,000 years to present.


Humans emerge as a leading cause of Holocene Caribbean extinctions. As part of a collaborative interdisciplinary team, I have worked to synthesize existing radiocarbon dates from the literature for all Caribbean mammals and archaeological sites. Using Bayesian hierarchical modeling in a meta-analysis, we found that anthropogenic impacts caused two size-biased extinction waves. This information is the foundation for exploring the impacts of megaherbivore loss and invasive species gain on Hispaniola.

IMG_1326 2

From three bears to one: the ecology of California’s lost bear assemblage. Rancho La Brea (Los Angeles, California) is one of the world’s most important Late Pleistocene fossil deposits, exhibiting a disproportionate representation of megafaunal carnivores due to “entrapment” in asphalt. Only black bears (Ursus americanus) exist in California today; however, fossils show co-existence of three large bears, including grizzly bears (Ursus arctos) and the terrifying short-faced bear (Arctodus simus). How did these large carnivores partition resources in an already carnivore-packed ecosystem? Did black bears today expand to fill the niche of their extinct relatives? Stay tuned!

Research Theme 3. Testing assumptions of niche conservatism and realized niches using fossil datasets.

Anticipating species responses to future climate change is a key priority for conservation planning. Species distribution models that rely on bioclimatic information and occurrence data only from the present yield projections that are likely biased by anthropogenic impacts, as present-day distributions for most species are a realized niche shaped by both humans and climate, not climate alone. Further, these projections assume that species’ niches are conserved, though evidence abounds that some species are more flexible than others. How can fossils quantitatively improve our models? I use the pre-human fossil record to test the validity of niche conservatism assumptions by projecting past relationships into the present day, creating a ‘null model’ of distribution in the absence of human impacts.

These tests can also be done in reverse. Many phylogeographic studies pair genetic data with species distribution models that hindcast present distributions into past climatic conditions. Such models could be strongly enhanced by comparing hindcast distributions with the known distributions from fossil occurrences.

Screen Shot 2018-08-15 at 3.34.46 PM

Humans, not climate, determine the present and future distributions of the Hispaniolan Solenodon. Working with an undergraduate honors student, Lauren Gibson, we developed species distribution models that incorporated the fossil record as well as points from our modern surveys for feces and camera trapping. Using the Last Glacial Maximum as a pre-human baseline, we found that modern distributions significantly depart from the expectations of those based on climate alone. Instead, solenodon occurrence is predicted exclusively by human population density – which likely reflects predation by commensal village dogs. While climate change is a significant concern for Caribbean biodiversity, predation by invasive species is a more urgent priority for this species. These results further bolster arguments that humans caused the Holocene extinction of many Caribbean mammals.

Research Theme 4. Providing quantitative reconstructions of climatic and anthropogenic extinction drivers. 

I am interested in how species responded to past extinction drivers; therefore, a relevant line of research is to develop quantitative paleoecological and archaeological datasets as context. I actively seek collaborations with Quaternary scientists who study pollen and sediment cores, and contribute to the Neotoma Paleoecology Database.


Paleoecological windows from packrat middens. Rodents of the genus Neotoma are known as ‘packrats’ because of their distinctive hoarding behavior, in which they create giant “middens” or nests comprised of material foraged from the local environment. These middens can be preserved for hundreds to thousands of years, creating time-capsules of the past. I am currently describing the vegetation of Rancho La Brea from a 50,000 year old packrat midden, using stable isotopes of fecal pellets and analysis of plant macrofossils. These data provide one of the oldest references of climate for Southern California.


Spatio-temporal population dynamics of humans in South America. I developed a comprehensive database of all radiocarbon dates for South American archaeological sites. This database has been used to model population growth trends of humans across continental South America. These data were collected to provide context for Late Pleistocene megafaunal mammal extinctions.

%d bloggers like this: