Composite Sampling of Sediments
Contaminated with White Phosphorus
MARIANNE E. WALSH, CHARLES M. COLLINS,
RONALD N. BAILEY, AND CLARENCE L. GRANT
INTRODUCTION
Table 1. White phosphorus concentrations
found in field samples collected from
Determining how successfully a site contami-
Eagle River Flats. (After Racine 1995.)
nated by white phosphorus munitions can be re-
Concentration
mediated is a challenge. Because of the discrete
range
Number of
Percent of
nature of the contamination, concentration esti-
(g/g)
samples
samples
mates can vary by several orders of magnitude
Not detected
1281
66
among individual samples taken over a small
Detection limit to 0.00099
79
4
area. Therefore, to reliably compare concentra-
0.001 to 0.0099
203
11
tions "before and after" remediation would
0.01 to 0.099
185
10
require an unrealistically large number of dis-
0.1 to 0.99
72
4
crete samples to be collected and analyzed. A
1.0 to 9.99
43
2
10 to 99.99
38
2
more efficient way to characterize the contamina-
100 to 999.9
16
1
tion is to composite samples collected over a large
1000 to 9999.9
6
0.3
area (greater then 100 m2). The success of cleanup
activities may be judged from the amount of
white phosphorus found in the composites before
Along transects where white phosphorus was
and after remediation.
detected frequently, concentrations varied widely
(Table 1), with relatively few samples having high
Background
concentrations. When samples were taken at
This study was conducted at Eagle River Flats
close intervals (1 to 5 m) around sample points
(ERF), a wetland artillery and infantry mortar
with high white phosphorus concentrations, we
range on Fort Richardson, Alaska, where dab-
again observed extreme heterogeneity, with non-
bling ducks and swans are acutely poisoned by
detectable concentrations within a few meters of
high concentration samples (over 100 g/g). This
ingesting residues from white phosphorus muni-
tions (Racine et al. 1992a,b; 1993). When this site
pattern of contamination led us to believe that
was initially investigated to determine the spatial
most of the white phosphorus was located in "hot
extent of the contamination, surface sediment
spots" that are less then 1 m in diameter and
samples were collected at approximately 25-m
probably are the impact points of projectiles con-
intervals along transects through sections of this
taining white phosphorus.
wetland. This spacing was chosen because it is
Microscopic examination of high-concentration
the radius of the area estimated to contain most of
samples revealed solid pieces of white phosphorus,
the fallout from the kinds of smoke projectiles
which were mostly 0.5 to 2 mm long; but some
that were commonly fired into ERF during train-
were up to 6 mm long, weighing over 100 mg.
ing exercises (Shin et al. 1985). At each sample
These particles are much larger than the fine-
location, several increments of surface sediment
grained silts and clays (95% finer than 0.02 mm)
were collected from within a 1-m-diameter area.
(Lawson and Brockett 1993) that make up the salt
Most of the samples from these transects con-
marsh sediment and could easily be selected by
tained low or undetectable concentrations of
dabbling ducks and swans searching for food or
white phosphorus.
grit (Trost 1981). The very low concentrations
Previous Page
