3.7 Sampling Strategy
During Phase I, soil, vegetation and surface water sampling showed higher
concentrations for metals and explosives in Jimmy Lake and Shaver River ranges.
During Phase II, these two ranges were re-sampled to verify the results of Phase I.
Jimmy Lake Range was evaluated using both circular and linear sampling strategies,
while for Shaver River Range, only the circular strategy was applied plus a statistical
evaluation using 100 discrete samples. During Phase II, vegetation sampling was
accomplished in Alpha Range using the linear transect strategy that had been omitted
during Phase I. Moreover, the remote areas were visited by helicopter and sampled
for metals. For the remote areas, at each location a different strategy was applied. A
total of 324 soil samples including 100 discrete samples plus 12 composite, 15
duplicate and 12 background samples were collected. 69 vegetation samples were also
collected including 11 duplicate and 19 background samples. 19 surface water
samples including 2 duplicate samples were collected and 28 sediment samples
including 5 duplicates were collected during Phase II. The surface water samples
were collected mainly in Primerose Lake and Jimmy Lake, but also in Shaver River,
in remote areas and in holes containing water in Alpha and Shaver River ranges.
Sediment samples were collected in Primerose Lake and in Jimmy Lake. All samples
were analyzed for metals (440), while a limited number was analyzed for energetic
materials (180 soil and 8 water samples).
Background soil samples are critical to establish the anthropogenic contribution
versus the natural contribution for all metal parameters. Background composite
samples were collected randomly, in circles of approximately 10 meters diameter in
different locations inside and outside the base. A minimum of 30 sub samples was
collected to form each background sample. A statistical analysis was conducted to
identify a mean background concentration and to define a limit for a value that can be
considered normal. Values at the extremities of the lognormal curve were identified.
The limits were chosen for a probability of 97.72% (2 times the standard deviation).
The probability of finding a result with a value higher than this limit is 2.28 %. When
the metals were not detected, a value at half of the detection limit was used for the
data analysis.
The usual strategy for soil sampling was based on systematically sampling around a
representative number of targets in ranges and also around hot spots (broken casings,
UXOs or debris, etc.). Usually, surface soils were collected at a depth of 0-5 cm. This
strategy was used in previous studies on antitank ranges, which showed very distinct
patterns of contamination around targets [2]. This strategy was used mainly in remote
areas. In Bravo Range, only a few samples were collected and most of them were
background samples. In Alpha Range, the linear transect strategy was used to collect
the vegetation samples. This strategy is the same as the one used during Phase I (Fig.
1). This approach was used to evaluate whether the level of contamination by metals
or energetic materials was following a pattern with distance from the target in the
ranges. If firing activities led to the accumulation of contaminants in soils or
vegetation, higher concentrations should be found around targets. Therefore,
composite samples were collected at distances of 20, 40, 60, 100, 120 % and 140% of
the distance from the entrance of the range to the target. Most of the time, an access
road went directly to the targets in the middle of the ranges. This road was used for
maintenance and clean-up. That road was used to build transects (right and left of
8
DRDC Valcartier TR 2004-204
Previous Page
