Environmental Engineering Reference
In-Depth Information
water in the outer rings relative to less flow in the inner rings
in ring-porous trees, although this explanation would not
hold true for diffusely porous trees.
The technique used to measure the tree-ring metal
concentrations was done so that the concentrations in each
ring could be measured. In brief, after tree-core collection,
the cores were dried in an oven, placed onto a plastic holder
with cyanoacrylate glue, and then shaved with a surgical
blade to a flat surface. The analysis of the prepared core was
by proton-induced x-ray emission (PIXE) spectroscopy.
Because most elements taken up by the xylem in a particular
year are not translocated between rings formed during early
or later years, the concentration detected in a particular ring
reflects the source of water used by the plant at that time. On
the other hand, some elements taken up such as potassium
(K) were found to be translocated and concentrated in the
heartwood of tulip trees (
Liriodendron tulipifera
L.) growing
over K-contaminated groundwater after uptake (Vroblesky
et al. 1992). The source of the potassium was potassium
chlorate used in munitions manufacture. In that study,
depth to groundwater was about 8.2 ft (2.5 m). Potassium
concentrations in the contaminated groundwater from 1985
to 1987 averaged about 9.4 mg/L.
In this study by the same authors, for a few trees there was
no outer ring enrichment of trace elements, as there was
reported for iron, but other trees in which the K concentrations
in groundwater were lower exhibited the outer ring enrich-
ment. In fact, the potassium was higher in the heartwood
relative to the sapwood in two trees growing over
K-contaminated groundwater, opposite what was shown for
iron and chloride. In trees growing over groundwater with
lower levels of K, however, the trees showed higher K in
the sapwood, relative to heartwood. The authors suggested
that trees maintained potassium levels between 700 and
1,300 mg/L in the sapwood, when average groundwater
potassium concentrations were no greater than 10 mg/L. If
true, this would suggest the presence of an ion pump in trees
similar to a sodium-potassium pump.
Nickel contamination of groundwater used by trees
showed similar results of the occurrence of nickel in tree
rings formed during periods of exposure to nickel with little
translocation from sapwood to heartwood (Yanosky and
Vroblesky 1992). Depth to water table was 0.9-8.2 ft
(0.3-2.5 m), and dissolved nickel was as high as 0.2 mg/L
in contaminated groundwater and 0.013 mg/L in uncontami-
nated groundwater. No attempt was made to relate ground-
water trace-element concentration to what was measured in
the tree rings. The tree-ring concentrations for the various
elements were reported as part per million (ppm), similar to
the groundwater concentration. However,
comparison. In a later publication (Yanosky and Vroblesky
1995), these results are reported as
g/g, rather than ppm.
Researchers also have shown that the tree-core technique
for oaks and cypress trees has been a useful indicator of the
presence of chlorinated solvents from groundwater contami-
nation (Doucette et al. 1998; Vroblesky et al. 1999a) and in a
variety of hardwood and softwood species (Vroblesky et al.
2004; Schumacher et al. 2004; Sorek et al. 2008), and petro-
leum hydrocarbons such as benzene, toluene,
trimethylbenzene isomers, and methyl
tert
-butyl ether
(MTBE) in oak trees (Landmeyer et al. 2000; Arnold et al.
2007). In those studies, cores of trunk material were col-
lected using standard increment-borer techniques; core
material was placed in glass vials that were capped with
gas-tight seals, and compounds in the headspace were then
identified using a field gas chromatograph with a photo-
ionization detector (Vroblesky et al. 1999a). This analysis
was done a day or a few days after collection in order to
permit time for gas diffusion from the core to the vial
headspace to occur.
The relative simplicity of this method has resulted in a
variety of similarly-themed publications that confirm that
tree cores are good surrogates for groundwater sampling
from a qualitative point of view, especially for chlorinated
solvents at contaminated sites or as a survey tool to find
suspected contamination (e.g., Sorek et al. 2008; Larsen
et al. 2008). To more rapidly determine the VOC concen-
trations, such as would be needed to direct field studies,
Vroblesky et al. (1999a) demonstrated that heating the
samples in a block heater or water bath for a few minutes
provided similar analytical information about the tree-core
VOC concentration.
In Vroblesky et al. (1999a), nearly 100 trees were cored,
and the species included baldcypress, tupelo, sweet gum,
oak, sycamore, and loblolly pine. The trees were located in
a flood plain of the Savannah River, between South Carolina
and Georgia, that received contaminated groundwater dis-
charge. In Landmeyer et al. (2000), cores taken from the
evergreen live oak (
Quercus virginiana
) and VOCs
extracted with methanol, and the reduced carbon compounds
such as BTEX and the fuel oxygenate MTBE were identified
using gas chromatography/mass spectrometry. This detec-
tion of MTBE in trees at the field scale was later confirmed
to occur in coniferous evergreens by Arnold et al. (2007).
The location on the tree where the core is collected is
important for studies of trees growing above contaminated
groundwater (Vroblesky et al. 1999a). In general, tree cores
should be collected from near ground surface to breast
height, in order to intersect as many annual growth incre-
ments (rings) as possible. This is because growth moves
downward from the stem tips to the base, such that the higher
up a tree a core is collected, the more recent are the annual
rings intersected (Phipps 1985). It has been shown that
m
the results of
PIXE analysis are provided in mg/kg (or
g/g; equivalent
to ppm) such that a direct comparison to groundwater con-
centration is tenuous at best, other than as a relative
m
Search WWH ::
Custom Search