SAMPLE HETEROGENEITY
co
Z
713
r
&
5 10
Fig. 1— Variations in the
12
1
vertical distribution of Sr
and *9’Cs in soil from the
profiles of three fields,
November 1962. @, field 4;
‘
A, field 5; C, field 6.
16
18
{|
0
20
{ fof
0
20
fj
40
j
60
4
80
MEAN MC/SQUARE MILE
picocuries per kilogram of soil are directly proportional to the millicuries per square mile and inversely proportional to the kilograms of
soil per square meter. Variation in the radioactivity of a sample of
several soil specimens is a complex function of the variation of the two
independent variables plus variation introduced among specimensbythe
Sampling and nuclide-extraction procedures. It is unfortunate that the
sampling method requires that the millicuries per square mile be com-
puted from the picocuries per kilogram of soil.Soil Samples were taken
to find the amount and variability in fission-product deposition from
Site to site, but the real variability in the field was confounded with
other variations among the specimens analyzed.
Equipment for weighing and for linear measurements was quite
accurate. Figure 2, a scatter diagram, shows the variability among
weights of soil specimens taken from field 4. Measurementsof soil
mass per unit area tended to diverge for specimens taken above and
below the plow pan.
Dispersion in picocuries of "Sr per kilogram of soil is shown in
Fig. 3.
The variation was greatest within the plowed layer and de-