High-Resolution Younger Dryas Environmental Variability:

A Comprehensive Assessment from Mid-North America Tree Rings

Collection of Tree-Ring Fossils: We are locating and collecting subfossil wood from mid-N. America and investigating features of their tree rings to determine availability and quality of tree-ring records as a terrestrial high resolution climatic proxy of late Pleistocene - Early Holocene (LP-EH) transition in the North America, centered on the Younger Dryas (YD) abrupt climate change event. Such high-resolution records of the Younger Dryas have been lacking in North America, but this effort is particularly necessary because the Younger Dryas is the last major global climate excursion to very cold conditions, which may have lessons for the past and the future related to marine thermohaline circulation. Thus far we have collected, or obtained from the collections of others, wood samples from almost 20 sites, primarily in Indiana, Illinois, Michigan and Wisconsin, spanning the period from about 14,000 to 8,000 14C years B.P. We have visited about half of these sites in person and our collections in most of those cases are the first major wood collections from these sites. Preservation of recovered wood was primarily through burial in glacial and alluvial deposits along lowlands of lake shores and riversides, and on upland hills and, in two cases, inundation by rising lake levels at sites submerged offshore. Tree-ring network (Map & Photos)

Tree-ring crossdating: To date we have developed seven tree-ring width chronologies and sixteen sequences of tree-ring widths. Additionally, three tree-ring width sequences from two Mid-Holocene sites were established. The tree rings represents tree growth over a highly diverse range of ecological conditions. The dominant species in the tree-ring sequences come from 4 conifer species (spruce, pine, hemlock and red cedar) and 7 deciduous species (oak, ash, hickory, elm, mulberry, maple and butternut). Nevertheless, the tree-ring chronologies are mainly built from spruce of lowland forests. Tree-ring variability of developed chronologies indicates a clear signal of environmental factor(s). However, the captured signatures of tree-ring variability of these sites only represent a local scale of environmental impact on radial tree growth because as of now the floating tree-ring chronologies are spaced widely in time. We found the following features of the studied tree-ring width series: correlation between crossdated tree-ring series, segment length of series (up to 300 years), tree-ring width variance and strength of common signal, which are sufficient to establish a tree-ring record of YD-EH transition. Limited sample depth (total number of overlapped trees for each year of chronology) of the tree-ring chronologies is the most troublesome issue. Total number of crossdated tree-ring series is not high. For example, we examined over 400 tree-ring specimens from 17 sources (possible locations). Tree-ring width measurements were possible for 60% of those samples and only 50% of total number of samples were crossdatable. Tree-ring chronologies (Table and Figures)

Radiocarbon dating: We determined three key sites with highest tree-ring sample replication and used them as pointers for changing climatic conditions through the region (Fig. 3-4). Our radiocarbon dating seems to confirm the first in situ black spruce site identified as being late YD in age (Liverpool). A second site appears to follow a few hundred years after the YD (Gribben Basin). A third site is the previously known Two Creekan buried forest dated ca. 1,000 before the Younger Dryas event (Two Rivers). Besides tree-ring widths, we developed high-resolution radiocarbon chronologies of these three sites and stable-isotope analysis of bulk wood samples from nine sites to address questions of regional environment and global carbon cycle. For the Younger Dryas particularly, high-resolution single-year chronologies of 13C and 18O stable isotopes were developed from the Liverpool site (Fig.5). Calendar ages (Figures)

Environment and climate reconstruction: Besides tree-ring widths, we are developing high-resolution radiocarbon chronologies of 3 of the sites and stable-isotope analysis (carbon and oxygen) to address questions of regional environment and global carbon cycle. The isotopic composition of ancient wood provides information about past environments unlike other climate proxies. Isotopic values of Picea, Pinus, and Thuja species suggest that the 2 sites of near-Younger Dryas age experienced the coldest conditions, although the Gribben Basin site near the Laurentide ice sheet was relatively dry, whereas the Liverpool site 500 km south was moister (Fig. 4). The spatial isotopic variability of 3 of the 4 sites of Two Creeks age shows evidence of an elevation effect, perhaps related to sites farther inland from the Lake Michigan shoreline experiencing warmer daytime growing season temperatures. Thus, despite floristic similarity across sites (wood samples at 7 of the sites being Picea), the isotopes appear to reflect environmental differences that might not be readily evident from a purely floristic interpretation of macrofossil or pollen identification.

The Liverpool site revealed the first glimpse of N. American Younger Dryas environment at annual resolution between 12,100 and 12,000 Cal years BP. The black spruce forest existed for almost 120 years with two waves of recruitment punctuated by intervals of stressful growth. Interannual variability of tree-ring records (isotopes along with tree-ring widths and ring anomalies) suggests a cold and wet climate, with trees experiencing frost events, tilting, drowning and burial in lacustrine sands of a rising paleo-Lake Michigan as the Laurentide ice sheet deteriorated, and ultimately shearing of their tops after death, perhaps by lake ice. This rise in water levels may evidence a distinct Post-Calumet (Algonquin) high-water phase of ancestral Lake Michigan (Fig.5, 6). The isotope records show various trends and fluctuations associated with temperature. However, tree-ring 18O does not show any unusually light values as might be expected if pulses of glacially derived melt-water were a major component of the water, as suggested in very low ostracod 18O (implying lake water 10-12‰ lighter than modern) from Lake Huron at ca. 10,600 to 10,000 14C y BP.

The developed set of tree-ring climate proxies suggests that high-frequency of the climatic system operated at ENSO-like mode during the late Pleistocene - Early Holocene (LP-EH) transition in the North America.The wavelet spectra of the tree-ring records showed ENSO-like signature in the interannaul and interdecadal variance of tree growth over the studied transition interval (Fig. 7). Environmental and climatic variability (Figures)

This systematic study provides the first high-resolution portrait of the Late Glacial/Early Holocene transition in mid-N. America.  The events and changes of YD time (related to temperature, water-use efficiency, and hydrologic cycle-related parameters of precipitation, relative humidity, and soil saturation) are measured against those of the periods immediately before and after, and with modern conditions.  The results is better characterize environmental/vegetation changes and responses in mid-North America, place the changes into the context of global changes during deglaciation, and use them to better understand wide-ranging air-ocean changes in this period.  The study conforms closely with current Earth System History program goals to reconstruct paleoclimate variability for intervals of the Holocene in N. America, especially related to (1) abrupt climate change events, (2) climate variability on decadal through millennial time scales (including linkages to ENSO), and (3) atmosphere-ocean-land interaction such as associated with internal forcing by thermohaline circulation.

We are grateful to NSF-ATM support through grant # 0213696 (ESH) for resources to carry out this extensive volume of work. We thank all participants, contributors, collaborators (see bellow) and volunteers for their hard work and valuable insights in our research.

Participants: I. Panyushkina, S. Leavitt, L. Cheng, P. Garcia, K. Lombardo, G. Goral, D. Petry, S. Noggle.

Collaborators: Numerous contacts were developed that resulted in sharing data and information exchange between the LTRR working group and Midwestern universities. It greatly increased collaborative interactions of the LTRR. The project has been immeasurably enhanced by joint work of the dendrochronological group and colleagues having expertise in radiocarbon measurements, geology, botany and forestry.

University of Arizona, Tucson, AZ: T. Lange (NSF AMS Facility)
Northern Arizona University, AZ: K. Cole
Northern Michigan University, Marquette, MI: J. Hughes
North Eastern Illinois University, Chicago, IL: F. Pranschke
Forest Products Lab, USDA, Madison, WI: A. Wiedenhoeft
Michigan Tech University, Houghton, MI: K. Pregitzer, B. Regis, D. Reed, S. Shetron
Oakland University, MI: R. Douglas Hunter
University of Wisconsin-Oshkosh, WI: W. Mode
DuPage County Forest Preserve, IL: Maggie Zoellner
Lincoln Quarry, Material Services Corp., IL: R. Brown
University of Winnipeg, Canada: J. Tardif
University of Iowa, IA: A. E. Bettis
University of Wisconsin-Green Bay, WI: R. Stieglitz, J. Moran
Illinois Geological Survey, Champaign, IL: B. Curry, L. Follmer
Illinois State Museum, Springfield, IL: E. Grimm
Indiana Geological Survey, IN: T.A. Thompson