Dendrochemistry: Seeing the forest through the trees
Open Access
- Author:
- Hauck, Danielle Kristin
- Graduate Program:
- Nuclear Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- April 04, 2008
- Committee Members:
- Kenan Unlu, Committee Chair/Co-Chair
Margot Wilkinson Kaye, Committee Member
Yousry Azmy, Committee Member
Peter I Kuniholm, Committee Member
Sturt W Manning, Committee Member
Jack Brenizer Jr., Committee Member - Keywords:
- trace elements
eruptions
tree rings
neutron activation analysis
dendrochemistry - Abstract:
- Dendrochemistry is a term used to describe the compositional analysis of tree-rings and its application to biomonitoring of the tree’s environment. The central hypothesis of this study is that the concentrations of silver and gold in tree-rings in the Mediterranean region record the history of African dust storms. Although there have been previous studies of tree-ring composition using both NAA and other analytical techniques, to the author’s knowledge, this is the first dendrochemical study to consider the effects of dust storms on tree-ring chemistry. However, the results, as with many other dendrochemical studies, are promising at best. The tree-ring concentrations in a single tree in west-central Turkey were highly correlated with the local aerosol index obtained from the Total Ozone Mapping Spectrometer (TOMS). This suggests that silver and gold concentrations in carefully selected trees may serve as a proxy indicator of dust storm activity and indirectly, the phase of the North Atlantic Oscillation. Three other trees from the same forest did not have any silver or gold concentration peaks and possible reasons for this discrepancy are discussed. The correlation with dust storms is still uncertain because it was impossible to gather appropriate information about soil characteristics or additional trees for sampling to prove or disprove the hypothesis. However, the chemistries of silver and gold are discussed to elucidate why they may be uniquely good at recording environmental signals in tree-rings. Some previous dendrochemical studies indicated that tree-ring chemistries record volcanic eruptions. Although some data were obtained in this study that support this view, the four long (300+ year) tree chronologies that were analyzed were not correlated with each other or with the timing of major volcanic eruptions. The concentrations in these trees were not accurately recreated when they were re-analyzed for the current study. Therefore, the failure to correlate the tree-ring concentrations may be the result of poor analysis methodology and contamination problems early in the dendrochemistry project. Some promising results for detecting eruptions in tree-rings were obtained on tree samples of a shorter duration (20-30 years) from Catacik Turkey that were prepared for irradiation under improved conditions of cleanliness. These tree samples were analyzed later in the project after sample preparation procedures were altered to reduce the possibility of sample contamination. Concentration peaks in the Catacik samples were correlated with the timing of the Pinatubo eruption (1991) and the El Chichon eruption in 1982. However, many of the concentration peaks were in single trees and could not be verified by comparison to other trees that grew nearby. The relative success of correlating concentration peaks in these shorter tree sequences may have been due to the improved sample preparation procedures. Another possible explanation is that concentration peaks do not remain in the tree-rings over time due to diffusion or mobilization by the tree. This would make it more difficult to find or identify possible environmental signals in the older rings of a long tree sequence. This study establishes a methodology for using Neutron Activation Analysis (NAA) for dendrochemistry. In the process of analyzing more than 2200 tree-rings, in some cases using multiple irradiation schedules, a range of NAA procedures and a set of tools for analyzing dendrochemical data were developed. During the course of the study many improvements were made to the NAA procedures, including 1) reducing the possibility of sample contamination through better sample preparation, 2) refining the NAA methodology to produce more data with higher accuracy and 3) determining and minimizing the major sources of errors in the concentration data. Unfortunately, this study contributes to the overwhelming number of feasibility studies that have been performed since the 1950s. There is still no consensus regarding the mechanisms responsible for recording environmental events in tree-rings beyond passive uptake. The feasibility of dendrochemistry will be more firmly established by moving forward with thoughtfully planned control studies. The refined NAA procedures developed here are necessary for future experiments which should strive to understand the actual mechanisms responsible for uptake. There are several different aspects of dendrochemistry that could be tested with control studies and that would be useful for interpreting tree-ring chemistry. For instance, dendrochemical studies could focus on the variation in specific elements between trees and the creation of composite forest chronologies for the appropriate elements. If this was successful, it would indicate that tree-ring compositions faithfully record environmental conditions in some cases and for some elements. A possible correlation was found between the Ag and Au concentration in tree-rings and the intensity and frequency of dust storms. However, the promising results were only found for one tree out of four sampled in the Catacik forest. A future study could be dedicated to verifying the presence of the Ag and Au signals in other trees in the same forest. Future studies could also analyze the ground water near sample sites to monitor the changes in soluble concentrations of metals, especially silver and gold in response to dust storms.