INTRODUCTION 



The neutron method of measuring soil 

 moisture is widely accepted by hydrologists 

 and soil scientists because the method can be 

 fast, easily performed, and accurate. Accuracy 

 is due primarily to the large volume of soil 

 sampled, nevertheless it is essential that the 

 sample be truly representative of the soil 

 within the measured zone. If a soil void is ad- 

 jacent to an access tube an error in measure- 

 ment will result. The magnitude of the error is 

 a function of the size and shape of the void as 

 weU as the volume of water within the void. 

 Soil voids, due primarily to poor installation 

 techniques, cause the greatest errors in neu- 

 tron moisture measurements. 



Many installation techniques will give satis- 

 factory access tube installation in rock-free 

 soils. However, in rocky wildland soils, special 

 access tube installation procedures must be 

 used to prevent voids between the soil and 

 the tube (Richardson 1966). Unfortunately, 

 literature that pertains specifically to the use of 

 neutron meters in wildland soils is scarce; fur- 

 thermore, literature dealing with the effect of 

 voids on neutron moisture measurements is al- 

 most non-existent. Koshi (1966) tested various 

 methods for drilling access holes in rocky soils 

 and concluded that the holes must be drilled 

 oversized to compensate for irregularities in the 

 dimensions of the drilled hole. These holes 

 were then backfilled with screened soil to re- 

 move any air gaps in the soil next to the access 

 tube. Koshi also tested the effect of 1/8 inch to 

 3/8 inch air gaps next to access tubes installed 



in loamy soil and concluded that there was no 

 effect. Troxler (1963) has suggested that 0.15 

 inch is the maximum air gap that can be toler- 

 ated without error. 



If the soil profile never becomes saturated, 

 then the effect of small air gaps is negligible. If 

 the soil moisture regime is such that the ground 

 water level fluctuates seasonally, then a small 

 air gap next to the access tube introduces a 

 small error in the summer when the soil is rela- 

 tively dry. However, this fluctuation causes an 

 unknown but much larger error when the water 

 table rises and saturates a soil void (Burroughs 

 1967). In a study of soil moisture on wildland 

 soils. Burroughs concluded that backfilling can- 

 not be relied upon completely to fill all voids 

 left by the installation process. Another con- 

 clusion was that the first indication of a serious 

 void problem occurs when the void becomes 

 saturated. Abnormally low neutron counts 

 ca mot be taken as prima facie evidence of a 

 \u:ge air filled void. Koshi (1966) showed that a 

 large rock next to the access tube may also 

 cause a drop in count rate. 



This paper describes a laboratory study of 

 the error caused by voids adjacent to access 

 tubes in a uniform porous medium when the 

 soil moisture content changes from a fully 

 saturated to a drained condition. The objec- 

 tives are to define quantitatively the neutron 

 errors so that scientists may have guides to 

 detect the presence of voids and estimate the 

 magnitude of the error associated with them. 



1 



