Collaborative Research: Mechanisms of tree population collapses in eastern North America: Disentangling causes of abrupt ecological change during the Holocene

University Of Maryland Center For Environmental Sciences, Cambridge MD

Investigators

Abstract

Trees are dying at alarming rates worldwide, raising concerns about forest health in the face of increasing droughts and fires. In the US, eastern forests have been less affected by droughts and fires than western forests. However, several eastern tree species -- notably eastern hemlock and American beech -- have experienced big population collapses over the last several thousand years. Hence, eastern forests may be at greater risk of collapse than commonly thought. This project will collect new data and apply powerful mathematical tools to test hypotheses about why hemlock and beech forests have suffered big declines. Researchers will focus on the roles of severe droughts, bigger and more frequent fires, and greater competition among tree species. Ultimately, this project seeks to 1) better understand the fundamental processes that trigger abrupt collapse of tree populations and 2) thereby inform and reduce current risks to forest health. The project will analyze long-term records (>6,000 years, at decadal resolution) of forest composition, fire regime, and drought, based on lake sediment cores to be collected at four sites -- two at places where beech and hemlock are abundant today and two where only beech is abundant. At each site, the timing of past tree population declines will be established by records of fossil pollen and radiocarbon dates, while past fire events will be identified through analysis of the abundance of charcoal fragments in sediments. Because plant water stress affects the kinds of carbon isotopes in plant tissue, this project will identify past drought episodes through an innovative approach that requires precise measurements of the carbon isotopic content of individual pollen grains. Because all records will be co-located from the same sediment cores, the relative sequence of events will be precisely known. At two sites, reconstructions of changes in lake water balance, based on characteristics of the lake sediments, will provide an independent check on the drought and water-stress signals inferred from carbon isotopes. A new community-level statistical model will integrate data and be used to test hypotheses about the relative importance of drought, fire events, and competition on past declines of tree populations. All data collected will be archived in the Neotoma Paleoecology Database, an open public repository of long ecological records, curated by experts. All source code for statistical software developed here will be made publicly available, and a workshop will be held to train students and early career scientists in the new methods that will be developed. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.