Peter Wilf's research

Fossil plants tell the evolutionary story of our green planet and reveal its landscapes from deep time to present. They are also extremely sensitive indicators of past climates, plant-insect interactions, biodiversity, and the effects of major environmental disturbances. These data provide deep-time analogs that uniquely illuminate modern ecosystems and their possible responses to anthropogenic change. The time period I mostly study is the latest Cretaceous through middle Eocene, 67-45 million years ago (Ma), an interval characterized by global disturbances that are closely spaced in geologic time. These include latest Cretaceous warming and cooling (68-66 Ma), the end-Cretaceous mass extinction (66 Ma) and ensuing recovery during the Paleocene (66-56 Ma), and both abrupt and long-term warming across the Paleocene-Eocene boundary (56 Ma). I enjoy collaborations with numerous colleagues around the world and a terrific lab group, with whom I do extensive field work, concentrated currently in Patagonia, Argentina and increasingly in SE Asia. Many of my collaborators are listed and cross-linked in the coauthor section of my Google Scholar page, and our research papers are also listed here.



Here are a few highlights of current projects.


Origins of Southeast Asian Rainforests from Paleobotany and Machine Learning

This is a new (2019-2024) cutting-edge NSF project, in collaboration with Dr. Thomas Serre’s Brown University machine learning lab and Dr. Maria Gandolfo’s paleobotany lab at Cornell, along with collaborators at MEF in Argentina (continuing the Patagonia project discussed below) and in several other countries including China, India, Australia, Indonesia, and Brunei. This tremendously exciting, emerging project aims to develop the powerful tools of machine learning to identify fossil leaves and thus to shed a powerful new light on the evolution of SE Asia’s extremely biodiverse and threatened rainforests.


Coverage of our milestone 2016 PNAS paper included this story in Wired and this interview on Radio NZ This Way Up with Richard Scott. Penn State produced a short video about the work and this press release.

Patagonia Paleofloras Project

(from the NSF award description here, current award DEB-1556666, 2016-2020. This project has been funded by NSF since 2004.)

The fossil record of life on land through time predominantly comes from the Northern Hemisphere. However, the outstandingly rich, relatively little-known fossil beds of Patagonia, southern Argentina, provide an unrivaled opportunity to learn whether life responded differently to mass extinction, plate tectonics, and past climate change on the other side of the world. This project intensively samples and analyzes fossil plants and animals from Patagonia through about 20 million years, from just before the end-Cretaceous dinosaur extinction (66 million years ago), through the early recovery period and the Eocene warming interval. Through groundbreaking field discoveries and state-of-the-art lab techniques, this research aims to transform understanding of the origins of the Southern Hemispheres flora and environments, the role of Patagonia, and the legacy of surviving living fossils now located in vulnerable rainforest areas as far away as Southeast Asia. The team plans to generate the most complete terrestrial record and curated fossil-plant collection for the Southern Hemisphere through a critical interval of Earth history. The work will produce a new, globally significant reference point for studying mass extinction, recovery, and response to climate change from a fossilized living laboratory, with direct importance for conservation and ecosystem management today.

The project uses field paleontology and stratigraphy at a series of exceptional Patagonian fossil localities, dating from the latest Cretaceous to middle Eocene, to elevate understanding of Patagonia's uniquely informative fossil floras. Most of the fossil sites are located in the La Colonia, Lefipán, Salamanca, Peñas Coloradas, and Huitrera formations of Chubut and neighboring provinces. The research focuses on five, broadly related questions targeting significant gaps in knowledge. (1) Was Patagonia a refuge from the end-Cretaceous mass extinction? (2) What are the composition, diversity, and biogeographic affinities of the earliest Paleocene floras and insect faunas? (3) Do the Eocene floras and faunas contain lineages that now survive in South America, Asia, and Africa, as well as Australasia? (4) Do the Eocene floras document the earliest phases of South American isolation by showing a loss of Gondwanan taxa and an increase in New World taxa through time? (5) How much geologic time does the prolific early Eocene site at Laguna del Hunco represent, and what processes formed these unique deposits that preserved one of the most diverse fossil biotas in the world?


The principal Argentine institution in this project is the Museo Egidio Feruglio in Trelew, directed by Co-PI Dr. N. Rubén Cúneo. Our Cornell collaborators, led by Dr. Maria Gandolfo, have posted a very nice web page about the Patagonia project here.

The project has a Google Scholar page here. We are fast approaching 130 peer-reviewed papers under NSF support since 2004, including several in Science, Nature, and PNAS.


Here are some of many news stories, interviews, and commentaries about our work in Patagonia regarding:

The first Fagaceae (Beech family) fossils from the Southern Hemisphere | First fossil fruit and oldest fossils of the nightshade family, fossil tomatillos | Fossils that challenge molecular dating | Exceptional diversity in Eocene floras | The connections between SE Asian orangutans and Patagonian fossils | The first fossil kauri trees (Agathis) ever found in South America | Patagonia’s petrified forests | Fate of Patagonian insects at the time of the dinosaur extinction | Eocene insect diversity | First Papuacedrus, a living New Guinea conifer genus, ever found in South America.


Evolution of plant-insect associations
A long-term, cross-cutting research focus with Dr. Conrad Labandeira, Smithsonian Institution, and many students and colleagues, including the Ph.D. and subsequent work of Dr. Ellen Currano, now Associate Professor at Univ. Wyoming, and Dr. Michael Donovan, now of Cleveland Museum of Natural History.

Terrestrial plants and insects today account for the majority of the Earth's biodiversity, and almost half of insect species are herbivores. Today's ongoing global warming and habitat loss will impact insect life cycles, distribution, and diversity. A key development in my research program was recognition of the breakthroughs in knowledge available from the abundant, diverse, and well-preserved insect-feeding damage found on fossil leaves. Dr. Conrad Labandeira and I examined the responses of insect herbivores to Paleocene-Eocene warming, from a time interval when body fossils of insects are lacking, and we discovered a marked increase in insect feeding that matched predictions based on modern observations. Subsequent work on the K-Pg fossil leaves from North Dakota provided the first evidence for a mass extinction of insects at the K-Pg as well as evidence for the Late Cretaceous origin of a diverse insect group (hispine beetles). Our work on the Laguna del Hunco flora from Patagonia shows a much greater diversity of insect feeding than in contemporaneous, comparable floras from North America, revealing another important, previously unknown component of high diversity in Eocene South America. We have also used tropical canopy access cranes in Panama to quantify modern herbivory and thus calibrate fossil herbivory, with former student Dr. Mónica Carvalho (now at STRI, Panama), entomologist colleagues at STRI, and several other paleontologists.

Here are some of many news stories, interviews, and commentaries about our work on insect damage through time, regarding:

Increased insect feeding during the Paleocene-Eocene thermal maximum and early Eocene climatic optimum | Insect extinction and recovery across the K-Pg boundary in the Great Plains and Patagonia | Discovery of 66 million-year-old feeding marks of hispine beetle larvae on fossil gingers | Canopy crane work in Panama


Other projects

Precolonial leaf mats from southeastern Pennsylvania and their value for landscape reconstruction and stream restoration- with Drs. Robert Walter and Dorothy Merritts, Franklin and Marshall College. This work provides the first landscape reconstructions based on leaf fossils for the pre-colonial Pennsylvania Piedmont. It was the topic of two MS theses and subsequent journal articles in PLoS One and Palaeontologia Electronica by Sara Elliott and Christy Grettenberger. There were several news articles on this work (including here and here).

The living legacy of Gondwanan rainforest history and its conservation in Australasia and Southeast Asia- with Dr. Robert Kooyman, Macquarie University, and many others. This work explores the living analog side of the Patagonian Floras and SE Asia projects and has involved several field trips in rainforests of subtropical Australia, Tropical Australia, and Mt. Kinabalu in Malaysian Borneo. Several papers have been published, on paleo-neo floristic connections across Gondwana (Kooyman et al. 2014, Am. J. Bot); site-specific living analog comparisons of Australian rainforests and our flagship Patagonian fossil site Laguna del Hunco (Lisa Merkhofer MS thesis and subsequent 2015 paper in Am. J. Bot.); fossil Agathis in Patagonia, including its living analogs in Malaysia (Wilf et al. 2014, Am. J. Bot); the ecology of leaf teeth in Australia (Royer et al. 2009, Am. J. Bot.), and a review paper on the origins of SE Asian rainforests (Kooyman et al., 2019). This research thread emerged from my 2005-2010 participation with Dr. Kooyman and others in the ARC-NZ Research Network for Vegetation Function, directed by Dr. Mark Westoby, Macquarie University.

Cenozoic paleobotany of Borneo and Sumatra- Cenozoic plant fossils from these extremely biodiverse and threatened regions have barely been studied since the 1800s. I participated in a 2014 field trip to West Sumatra and South Kalimantan led by Dr. Gregg Gunnell, Duke University, and colleagues at Bandung Institute of Technology, wherein we discovered and collected several new Eocene and Oligocene plant-fossil localities. In 2015, doctoral student Michael Donovan and I traveled to Brunei and, in collaboration with colleagues at Universiti Brunei Darussalam, discovered and collected the first plant-fossil sites ever found in that country (Miocene-Pliocene). Work is in progress and feeds into the new Origins of SE Asian Rainforests project listed above.


What about leaf physiognomy and climate?

Correlations of leaf shape and climate and their applications to paleoclimate reconstruction were a significant component of my early work with many colleagues. Several of our papers demonstrated that considerably less labor-intensive, univariate methods (i.e., leaf-margin analysis), while not problem-free, would generate paleotemperature and paleoprecipitation estimates at least as reliable as multivariate approaches (i.e., CLAMP). I applied these methods to several Cretaceous through Eocene fossil floras to investigate climate changes during critical time intervals. I also developed an alternative multivariate method, digital leaf physiognomy, that introduced computerized measurements to improve reproducibility. This approach came out of a productive collaboration with University of Michigan (Peter Huff) and Penn State (Dave Janesko) undergraduate researchers and Penn State then-postdoc Dr. Dana Royer (now Professor, Wesleyan University).

In 2010, this line of research forever changed when then-postdoc Dr. Stefan Little demonstrated the significant influence of evolutionary history (“phylogenetic signal”) on leaf-climate correlations in this PloS One paper. Leaf-climate correlations can still be used to evaluate climate change in a relative sense, but they are influenced both by the environment and by history. The core challenge now is the great difficulty of disentangling these effects quantitatively, especially in fossil floras where very few species can be placed phylogenetically. Ultimately, the “taxon-free-leaf” approach to paleoclimate reconstruction is much more limited than was long thought. As a result, I have become more interested in paleoenvironmental reconstructions that make better use of combining systematic, biogeographic, floristic, anatomical, and/or geological data from fossil floras, and in using traits that are less influenced by evolutionary history. The Patagonia Paleofloras project has become a showcase of this sort of approach. For example, Lisa Merkhofer et al. (Am. J. Bot., 2015) did a detailed investigation of how genera found as fossils at Eocene Laguna del Hunco (Argentina) are distributed today among a network of hundreds of Australian rainforest plots, and how the leaf sizes of the same genera compare between ancient and modern. To estimate leaf size, Merkhofer et al. were the first to use on fossils the Sack et al. (2012, Nature Comm.) secondary vein-scaling method, which seems to be largely free of phylogenetic effects and represents the state of the art for leaf-size reconstruction in fossil floras.

back to Peter Wilf's homepage