ERRORS AND ERROR PROPAGATION IN COMPUTED TRU ACTIVITY
DATED: April, 1980

DOE/ERSP TECH NOTE NO. 22.0
AUTHOR: W. John Tipton, EG&G, Ine.
Introduction

Conversion factors relating measured photopeak count rate data (as obtained with the IMP system)
to source activity in the ground depend on certain properties of the soil in which the radioactivity is

distributed. In particular, it is necessary to know the in situ soil density and soil moisture as well as
the elemental composition of the soil, These parameters are required to obtain the linear
attenuation coefficient (the inverse of the gamma ray mean free path) in soil for a given energy
gamma ray. The soil density is also required to convert activity per unit volume to activity per unit
mass.
A series of measurements were made between November 28 and December 11, 1979 over 9 islands to
expand the rather limited data base which previously existed for these parameters. Using a nuclear
density/moisture gauge, in situ measurements were taken at 182 locations in 73 areas over the 9
islands. A total of 124 soil samples were also obtained and sent to LLL for elemental composition
analysis.
An additional 11 samples were returned to EG&G in Las Vegas, NV for direct
measurementsof the linear attenuation coefficient.
Procedures

Direct in situ soil density and soil moisture measurements were made using a Troxler Model 3411
nuclear density/moisture gauge. The instrumentation and procedures employed were those specified

by the American Society for Testing and Materials (ASTM). Standard procedures for measuring soil

density by nuclear methods are given in ASTM D 2922-71 and for soil moisture in ASTM D 3017-72.
Briefly, the in situ or wet density of soil is determined by measuring the attenuation of 662 keV
gamma rays from a !37Cs source through a given depth of soil. The moisture content, by weight, of
soil is determined by measuring the moderation or slowing of fast neutrons from an Am-Be neutron
source.

Dry density is obtained by subtracting the moisture content from the wet density.

percent moisture is obtained by dividing the moisture content by the dry density.

The

In the Troxler Model 3411 gauge both the 137Cs and the Am-Be sources are located in a probe which

can be inserted to a given depth in the soil. The gamma ray and neutron detectors are placed on the

surface at a fixed lateral displacement of 25 em from the sources. After placing the sources at a

given depth, gamma ray and neutron counts are accumulated for a period of one minute.

The

resulting counts are converted to wet density and moisture content using calibration curves supplied
by the manufacturer.
Four independent measurements were made at each of the 182 locations sampled.

Measurements

was repeated after rotating the detectors through an angle of 90°. Each measurement gives the
average wet density and moisture content for that volume of soil lying between the sources and the
detectors.

ow

were made with the sources located at a depth of 15 em, 10 em and 5 em. The 5 em measurement

The standard procedure was to measure three locations within a given area to obtain an area
average. Measurements were made 5 meters N, 5 meters SE and 5 meters SW of a given reference
point, generally chosen to be one of the IMP measurement locations. This procedure was followed

for 54 of the 73 different areas which were measured. Only a single location was measured in the
other 18 areas,

density~-the sand-cone method. In the sand-cone methodsoil is carefully removed down to a given

depth. The resulting hole is then filled with fine sand having a known density. Measuring the weight
of sand required to fill the hole gives the total volume of soil removed. The apparatus used
B-22-1

we son

Of the 18 areas where only a single location was measured, 13 were areas where a cross-calibration

was performed between the nuclear density/moisture gauge and another technique for measuring soil

Select target paragraph3