Russell W. Graham, Ph.D. - US grants

Floral and Faunal Community Responses to Late-Quaternary Climate Change

John Williams, Stephen Jackson, Eric Grimm and Russell Graham


The PIs propose a study of the responses of species and communities to
late- Quaternary environmental variations and, in particular, the environmental drivers of species turnover, community assembly and disassembly, and the formation of no-analog communities. Their research will demonstrate the power of NEOTOMA, a new community paleoecological database that reduces informatics costs and removes barriers to interdisciplinary collaboration by storing in a single database most of the major late-Neogene paleoecological databases. The proposed research will be the first-ever synoptic analyses of late-Quaternary community dynamics in North America that directly integrate fossil pollen, plant-macrofossil, and faunal records. The PIs propose improving the resolution of the NEOTOMA database by obtaining 100 new AMS radiocarbon dates for key vertebrate fossil localities, adding recent high-quality records to NEOTOMA, revising existing chronologies in NEOTOMA in light of this new data, and mapping all data for 15 time-windows covering the last 21,000 years. These time windows are more finely subdivided than the original FAUNMAP synthesis, and will permit study of species responses to the rapid climate changes of the last deglaciation. The PIs will test such hypotheses by comparing the distributions of the floral and faunal no-analog communities to each other and to the distributions of no-analog climates. They will create plant based and animal based biome maps to see whether such maps are complementary or in contradiction. The PIs will further conduct generalized dissimilarity modeling to reveal patterns of species turnover, along environmental gradients, in both space and time.

Neotoma is a multiproxy relational database that includes fossil data for the past 5 million years (the Pliocene and Quaternary). It is a community database that provides the underlying cyberinfrastructure for a variety of disciplinary database projects. The Neotoma project is an international collaborative effort among individuals from 23 institutions, including domain scientists representing a spectrum of Pliocene-Quaternary fossil data types, as well as experts in information technology (IT). This project builds on the initial development of the Neotoma database, which was funded in 2008 with a 2-year grant from the NSF Geoinformatics program. Neotoma involves domain scientists representing a wide spectrum of fossil groups from the Pliocene and Quaternary teamed with information technology (IT) professionals from the Center for Environmental Informatics (CEI) at Penn State, where the database is hosted. A workshop was held at the National Academy of Sciences in Washington DC on 31 January-2 February 2009 to plan for the future development of Neotoma. This project is the outgrowth of the DC workshop.

During a 2-year development project, the database structure of Neotoma was designed and implemented with the import of vertebrate faunal data from the FAUNMAP database and pollen data from the Global Pollen Database. The North American Plant Macrofossil Database was updated, and data are now ready for import. The Neotoma website was launched, and the Neotoma explorer provides a map-based interface to the data.

During the next phase ( this project) several new constituent database cooperatives from North America, Europe, Africa, and Asia will join with Neotoma, and several new data types will be ingested, including diatoms, ostracodes, insects, and testate amoebae. Neotoma will be exposed to outside developers via Application Programming Interfaces (APIs) and a Software Development Kit (SDK), whereby both desktop and webbrowser- based applications can acquire data from Neotoma over the Internet. Data-steward tools will be developed that will allow stewards to upload and update data remotely. This project establishes working groups for various database activities, including development of software tools, development of age models, integration of various fossil types from packrat middens held in different databases, and organization of a new peatlands constituent database. Two database-wide workshops will be held for coordination, planning, and conceptual development.

This project will provide new data on the paleoclimates, paleoenviroments and the biodiversity impacts of sea level rise on the southern edge of the Bering Land Bridge (BLB), and is intended to facilitate a better understanding of why woolly mammoths survived late into the mid-Holocene only in the environments of Arctic islands of this area. Furthermore, this research will attempt to establish the actual time of extinction of the Holocene mammoth population on St. Paul Island, Pribilof Islands, Alaska, and apply this information to test various proposed causal hypotheses for the extinction.

Cores from Cagaloq Lake, St. Paul, will be sampled for chironomids (aquatic invertebrates), pollen, coprophilous fungi spores, plant macrofossils, charcoal, ancient DNA and cryptotephras. Oxygen isotopes from the heads of chironomids will provide an independent climate record that can be supplemented by paleoecological analyses of the chironomid assemblages. Studies will also test chironomid species for isotope fractionation to disentangle climate signals from ground water effects. Terrestrial community changes will be reconstructed from pollen, spores and plant macrofossils; and charcoal frequency will be used to document fire events. Analysis of ancient DNA will provide data on cryptic plant and animal species that have not been detected by traditional methods of analysis, and can also be used to identify taxa to species.

Spores that grow on animal dung will be used as proxies for the mammoth population size and to document the time of mammoth extinction on the island. Ancient DNA will serve as an important cross-check by helping to resolve specific identifications of these spores as well as providing an independent estimate the time of extinction by the absence of mammoth DNA.

Digital elevations, bathymetric data, sea level curves, and Geographic Information System (GIS) technology will be used to reconstruct island size from the time of its isolation until today. A highly constrained chronology of the Cagaloq record will be achieved by using 14 C dates and tephras. All data can be compared temporally in order to test hypotheses for mammoth extinction.

This project will provide opportunities for two PhD students, a postdoctoral fellow, graduate student assistant and undergraduate student. Results of the study will be disseminated widely in peer-reviewed journals. An exhibit on the results of the project will be prepared by the EMS Museum at Penn State University with a web component. In addition, a special on-line, interactive exhibit will allow participants to reconstruct the island at various sizes, populate it with differing mammoth populations and define different climate and environmental factors to observe how each of these components affects mammoth extinction. The interactive exhibit will be available on the Neotoma database. Members of the group will make presentations on the project at the Alaska Quaternary Center in Fairbanks, which serves as a hub for promoting Alaska Quaternary research and outreach to the public.

The PI will develop and pilot a scalable strategy for involving educators in polar research and making data from a polar research program accessible to students. Plans include engaging grades 6-12 and undergraduate science educators in a long-term, ongoing research project in Greenland that examines changes in the phenology (e.g. seasonal life cycle events - in this case, the onset of springtime plant growth) of arctic plants and how this relates to climate. Educators and students will conduct similar observations and experiments in their local environment and will contribute their data to the ongoing research efforts. Participants will consist of approximately 24 middle school, high school, and undergraduate educators from schools across a range of geographical locations and school districts serving underrepresented and underserved students.

The PI will conduct a workshop each year with the educators and will develop online resources that will be available to the educators. Educators will be engaged in an ongoing study of plant phenology at a field site near Kangerlussuaq, Greenland, and they will learn observational and experimental methods of studying plant phenology that they can use with their students in their local environment. Educators from Fort Lewis College, Colorado, which is a primarily undergraduate institution; Blue Hill Consolidated School, Maine; and St. Paul Island, Alaska, will participate during this project. Additionally, the development of online resources will provide access to data from long-term research on arctic plant phenology, related lesson plans and media representations of research activities in Greenland to an expanded network of educators. These on-line resources will serve educational needs beyond the duration of this project. The proposed activities present outstanding opportunities for the involvement of student groups, including Native American and Alaskan Aleut students, in research through hands-on training and inquiry-based educational experiences.

Paleontologists provide data about the past distribution and diversity of life. These data are useful both to geologists, because they can help determine the age of rocks, reconstruct past environments, and constrain models of the Earth system; and to biologists interested in the evolutionary history of organisms and the behavior of ecological systems during past global changes. Currently, data about fossils are dispersed across thousands of scientific publications, and dozens of small to large databases, only some of which are publicly available via the Internet. Even publicly available databases can be difficult to access because each stores different kinds of data with different conventions, requiring researchers to individually harmonize searches and their outputs. This project brings together six paleobiological databases so that they share a single set of Internet-based commands by which researchers and the public can easily access fossil records from all of Earth history. By coordinating with other emerging efforts in geological and biological data sharing, best practices, and protocols, we ensure that data will be freely available to all, enabling new scientific syntheses and discovery, more powerful educational opportunities, and general exploration of the history of life on Earth.

The paleobiological sciences sit at the nexus between geosciences and the biosciences, with close interdependencies in both domains. Within the geosciences, information about the past spatiotemporal distribution of organisms, species, and assemblages of species is essential to a wide array of allied disciplines: to sedimentologists and economic geologists studying facies relationships and employing biostratigraphic controls for correlating rock strata, to structural geologists and geophysicists seeking biogeographic constraints on reconstructions of former tectonic plate positions, to paleoclimatologists extracting paleoclimatic signals from paleoecological data, and to earth system modelers seeking to understand how biospheric dynamics have shaped, and continue to shape, the history of the Earth-Life system. Within the biosciences, the fossil record is essential for understanding how contemporary ecological systems are shaped by historical legacies of slow-acting processes, for testing climate-driven models of species distribution and diversity that are being used to project the impacts of 21st century climate change, for constraining phylogenetic models of species divergence and rates of evolution, and for understanding the fundamental drivers of biodiversity (i.e. species extinctions and originations). In an era of global change, when stewarding biodiversity is an urgent societal concern, conservation biologists, global change ecologists, and earth system scientists are all looking to the past to study the behavior of the Earth-Life system during rapid transitions. Paleobiological data are currently served by a wide array of databases that vary in structure, composition, temporal scales, types of data and metadata. To conduct ?global? or holistic analyses of the paleobiological record it is necessary to retrieve data from a variety of these databases - requiring queries of each database to retrieve the types of data needed. The purpose of this project is to make six different paleobiological databases interoperable so that they can be accessed via a common Application Programming Interface (API) to query the data from these and other databases. Towards that end, five key records of North American Pleistocene lakes will be uploaded and become available through this integrative project. This project also will increase the interoperability between these paleobiological resources and contemporary databases of species distributions and diversity, enabling continuous time-series analyses (e.g., of biodiversity) from the beginning of life on earth to today. Integration of the paleobiological databases with databases of the stratigraphic record (Macrostrat) will enhance the value of both types of data. New R packages will facilitate retrieval and analysis of data from all of the databases. Finally, this proposal establishes a Paleobiological Data Consortium, consisting of leaders of cyberinfrastructure resources in the paleobiosciences and allied disciplines, with the goal of sharing best practices and protocols among the geoinformatic and bioinformatic communities.

This Division of Earth Sciences Geoinformatics collaborative award supports continued development of Neotoma, a web-accessible centralized data clearinghouse for paleoecological data for which initial development support was provided by the EAR Geoinformatics Program (EAR- 0948652). This effort will enhance the current Neotoma platform by: 1) maximizing rates of data ingest through on-going data mobilization campaigns, recruitment and training of new data stewards, and development of facilitation software; 2) expanding the Neotoma data model to facilitate new proxy acquisition and secondary data including organic biomarkers, isotopic data from faunal specimens, and faunal taphonomic information; 3) developing new tools for data visualization and scientific exploration; 4) fostering engagement and growth of the Neotoma research community through support for a series of workshops intended to train data users, data stewards, and teachers, and by development of an on-line blog.

The Neotoma geoinformatics platform will promote integrative studies of the biotic impacts of climate change over the past five million years by allowing greater access to previously collected geoscience data sets of relevance and the tools to support analysis at marginal cost. The project will promote community engagement, support ease of data set ingestion, address PI team succession and promote educational module development. This award directly addresses OMB Memo M-13-13 ?Open Data Policy-Managing Information as an Asset.?

***