DOE received several different estimates of the total volume of soil removed from Kickapoo. The
results for each of these estimates using mixed linear and exponential assumptions are:
Volume Distribution
Areas with

Estimate of

Total Volume Estimate

Subsurface

Total Volume

Contamination

Areas without

Total

Contamination

Removed

Subsurface

cu. yd.

cu. yd.

Activity
Ci

7 July 78, J3
22 July 78,
B Co 84th Engr

4000

1175

2825

0.45

4400

1290

3110

0.50

sample sheets

4500

1320

3180

0.51

22 July 78, FRST-truck

In order to check how much effect the models chosen have on the estimates of total activity
removed, the estimates were repeated assuming only linear gradients. That is, the average for areas
without subsurface contamination was computed as

0.5(131 + 31.8) = 81.4 pCi/g TRU.
In areas with subsurface contamination, the assumption was that the gradient was linear with a
positive slope of 20 em and linear with a negative slope below 20 em. The average concentration
would then be

0.4 [0.5(203 + 194)]+ 0.6

[0.5(194

+ 85.4)]

= 0.4 (198.5) + 0.6 (139.7) = 163.2 pCi/g TRU.

The computed activity removed for the various volume estimates under the all-linear assumptionis:

Estimated

Total Activity

Total Volume

Removed

4000 cu. yds.
4400 cu. yds.
4500 cu. yds.

0.50 Ci
0.55 Ci
0.56 Ci

The differences between the models are far less than the difference between the two methods (IMP

versus truck samples). The IMP methodis preferable for a number of reasons:
1,

2.

3.
4.

5.

The truck samples are biased high.

Truck volumes are difficult to estimate accurately, and are not likely to be consistent.

IMP readings average over a large area, thus taking a larger sample of the population.
IMP readings are unbiased and have much lower variance than soil samples.

Total activity computations are fairly insensitive to errors in fitting a function to the soil
gradient.

Therefore, the values derived by the mixed linear and exponential models are to be considered the
most reliable, and the IMP sampling data is preferable for future computations of total activity
removed.
(Editor's Note:

Following thorough reappraisal of various measurement parameters (ef. Tech Note

23) the final estimates of TRU activity in soil removed from Island Sally are:
Yuma 0.28 Ci; Hustead, 0.16 Ci; Aomon Crypt, 0.93 Ci; Island Total, 2.22 Ci).

B~-10-4

Kickapoo, 0.85 Ci;

shen

Date & Source of

Select target paragraph3