Neutron Activation Analysis of Absolutely-Dated Tree Rings

 

Participants:	K. Ünlü, Prof. of Mechanical and Nuclear Engineering
	D. K. Hauck, Ph.D. student at Mechanical and Nuclear Engineering

T. H. Daubenspeck, Activation & Irradiation Specialist

Services Provided:	Neutron Beam Laboratory

Sponsors:	DOE-NEER
	Cornell University
	NSF and RSEC

Introduction

The dendroanalysis project is a search for heightened gold concentration in tree rings with Instrumental Neutron Activation Analysis (INAA).  Large amounts of stratospheric sulfur from volcanic eruptions, forest fires, or industrial pollution produce acid rain, which stresses trees and results in increased nutrient uptake. We are concentrating on gold because of its suitability to INAA and its chemical similarities to copper, a plant micronutrient which is taken up during times of stress. Tree rings with increased gold concentration can be used to pinpoint years of volcanic eruptions. This information can be used to determine the climatic effect of aerosols and volcanic ash injected into the atmosphere during an eruption by correlating it with historical draught and crop records, or ash layers in bedrock and ice cores.  Historians use eruption timelines to date associated historical events, such as plaque, crop extinction, wars, and the end of civilizations, which are sometimes precipitated by volcanic events.

The Dendroanalysis project has been progressing at the Radiation Science and Engineering Center since January of 2003.   Within the last year, a third tree sample named CGRPPG4 was analyzed for its trace constituents.  Tree Sample CGRPPG4 grew in Greece between 1411 and 1988 AD.  Tree samples CTUKLK10B and CTUCAT14C, which both grew in Turkey, were analyzed in 2003 and 2004, respectively.  All three trees have been analyzed for their concentrations of gold, sodium, potassium, zinc and bromine.  The correlations coefficients for isotope concentration between tree samples indicate that the two Turkish samples provide the most correlated data set.  This suggests that the uptake mechanisms for certain elements are regionally dependent.  In addition, Icelandic eruptions are correlated with peaks in tree ring gold concentration more consistently than eruptions in any other geographic area.  This observation is consistent with the fact that due to their latitude and magma type, Icelandic eruptions are climatically more significant than similar sized volcanic eruptions which happen elsewhere.

Experimental Facility

The INAA facility was completed at the Radiation Science and Engineering Center in the previous year (2004).  According, descriptions of the Dry Irradiation Tubes (DITs), the Automatic Sample Handling System (ASHS), and the High Purity Germanium (HPGe) detector are available in the Radiation Science and Engineering Center 49th Annual Progress Report.  Only a summary of the facility and gamma acquisition equipment is offered here.

Dry Irradiation Tubes
An irradiation facility was built for the purpose of analyzing tree ring samples with NAA.  The facility consists of two closed bottom aluminum “dry” tubes which are placed in empty holes in the grid plate.  The positions of the dry tubes correspond to grid plate locations E4 and E13 in the core. Since the dry tubes were put into place in 2003, there has been one core reloading done in late 2004.  The original loading was altered by placing additional fuel elements in close proximity to the dry tubes, thus increasing the flux.

Automatic Sample Handling System
For handling multiple samples quickly and easily, an Automatic Sample Handling System (ASHS) was built for transferring samples one at a time to the detector for counting.  It is capable of holding up to 90 samples at a time and has programmable logic control.

High Purity Germanium Detector
For gamma radiation detection a co-axial High Purity Germanium (HpGe) semiconductor detector was used. The detector purchased for this project has a relative efficiency of 36% and energy resolution of 1.8
keV at an energy of 1173 keV. The crystal is kept in thermal contact with a tank of liquid nitrogen to stabilize temperature and reduce noise. Pre-war lead bricks surround the detector, providing 4” of protection from background radiation.

Gamma Acquisition and Analysis
Gamma acquisition was completed with the use of a PC running Genie-2000 software, a DSA-2000 and a HpGe detector, described above. The DSA-2000 works as the multi-channel analyzer (MCA), amplifier and high voltage power source to the detector.  The Genie-2000 software controls the count settings and allows for automated nuclide identification for qualitative or quantitative NAA.

Sample Description and Preparation
Tree samples were gathered and dated by the Malcolm and Carolyn Weiner Laboratory for Aegean and Near Eastern Dendrochronology at Cornell University.  Members of the laboratory travel to Turkey and Greece to obtain tree samples for dendrochronology and dendroanalysis during last 30 years.  Tree samples chosen for further analysis through INAA are systematically cut into rings to produce samples which range from .1 to .25 grams in size.  The rings are placed into polyethylene heat sealed bags, labeled with a permanent marker and set aside for irradiation.

Samples are irradiated for 4 Megawatt-hours and allowed to decay for 15 hours before counting.  This allows for adequate activation of the gold without producing excessive amounts of radioactivity.  Tree rings are irradiated in groups of 40. Therefore, counting samples for 1 hour each insures that they will all be counted before the gold has decayed by 1 half-life.

Results

Three tree samples covering approximately the last 400 years were analyzed for their trace constituents. Six elements were consistently identified in the wood samples including sodium, potassium, bromine, zinc, lanthanum and gold.  It is possible to define a cross correlation coefficient for each of these isotopes as

where [Ml] is the concentration of isotope M in the ith tree sample. The cross correlation coefficient, r, indicates a perfect correlation for two time series when it is equal to one, and a perfect anti-correlation when it is equal to negative one. Figure 1 shows the cross correlation coefficients for element concentrations between two trees.  The element with the highest correlation is potassium for all three comparisons.  Zinc and Bromine follow, and are consistent with relation to each other for each comparison. Gold is either slightly negative or slightly positive, indicating that the correlation is very close to zero.    The tree samples CTUKLK10B and CTUCAT14C, which both grew in Turkey, are the best correlated.  The low correlation coefficients for gold indicate that individual tree samples do not show a complete record of environmental events that may be indicated by gold. Therefore, several tree samples may be needed to identify significant events in history.

The size of an eruption is often designated by the Volcanic Explosivity Index (VEI) which is an order of magnitude estimate of the amount of tephra released into the atmosphere. Generally speaking, the larger the VEI for the eruption, the greater its effect on the environment will be. Eruptions with a VEI of 4 or greater may inject aerosols into the stratosphere and have widespread effects as a result. Figure 2 lists the volcanoes that have a VEI of 4 or greater that had correlated gold peaks in any one of the three trees.

The largest volcanic eruption in recent history was the eruption of Tambora in 1815 with a VEI of 7.  Tambora released over 8 cubic kilometers of ash and has substantial signals in both the artic and Antarctic ice cores. However, the only tree in which a visible gold concentration peak appears in 1815 is CTUKLK10B. The peak in 1815 appears at the end of an active gold region which begins in approximately 1809. There has been evidence from other sources of an eruption of unknown origin in 1809. Along with Tambora, its signal is evident in both the northern and southern ice cores, and sometimes the two peaks are referred to as the “Tambora doublet”.

Figure 1. Cross Correlation Coefficients for Isotope Concentrations

Many of the eruptions which had correlated peaks in gold concentration occurred in Iceland.  Katla, Hekla and Grimsvotn are currently the most active volcanoes in the region and have been identified multiples times in the data. The fact that Icelandic eruptions are well correlated with gold concentration is consistent with the fact that the characteristics of Icelandic eruptions make them climatically significant.

The volcanoes that have the largest atmospheric effect are those which have basaltic magma, which contains large concentrations of sulfur. The composition of magma is characteristic of the tectonic plate which it originates from in the volcanic rift or hotspot. Icelandic magma is very basaltic in nature and eruptions in that area often make large contributions to the global sulfur budget. Also important is the ability to inject aerosols into the stratosphere.  Since the tropopause is lower at high latitudes, a smaller plume height is required to reach the stratosphere.  Near Iceland, the tropopause is at a height of 8-10 km,  and even relatively small explosions can break through it.

Conclusions

The volcanoes which were most consistently identified in the gold peak data occurred on Iceland.  This is consistent with the fact that Icelandic magma is very basaltic in nature and the island is at low latitudes.  However, some Icelandic eruptions which are known to have a large atmospheric effect, such as Grimsvotn in 1783, do not show up in the data.

There are several variables at play which effect the correlation between volcanic eruptions and gold concentration in tree rings.  In order for a tree to be stressed by acidic fallout from an eruption, volcanic aerosols must pass over the sample site and be correlated with abundant rainfall. The complexity of Earth’s dynamic system makes it impossible to predict on a theoretical basis which regions will experience acid aerosol washout.  Contrarily, it is hoped that a statistical analysis of the eruptions that did stress trees in the Near-East can help quantify the effects of climate dynamics from an empirical stand point.

Dendrochemistry has the possibility of supplying an independent dating scheme for volcanic eruptions, with use to both environmentalists and historians.  However, more research is needed which is aimed at understanding the mechanisms behind growth patterns and heavy metal ion uptake. Studying multiple trees over the past 20-30 years would allow for a better statistical analysis of the correlations between them. During this period detailed atmospheric data is available from research satellites. The effects from climatically important eruptions may then be mapped and compared to satellite data of the volcanic sulfur clouds produced by recent eruptions.

Figure 2. Gold Concentration int three tree samples