APPLICATION 



The following example shows how the re- 

 sults of this laboratory study apply to field 

 studies. Assume that soil moisture measure- 

 ments are made at a certain depth within an 

 access tube when the soil is saturated. If these 

 measurements show a value of soil moisture 

 which is greater than the maximum pore space 

 of the soil, then a void adjacent to the tube 

 should be suspected. Figure 12 illustrates such 

 a hypothetical situation. The assumed maxi- 

 mum pore volume is 44 percent and the soil is 

 assumed to be a coarse sand similar to that used 

 in this study. This figure indicates that the soil 

 moisture measurements, taken from late March 

 to late June, are inaccurate because the meas- 

 ured soil moisture exceeds the maximum pore 

 volume of the soil. If measurements are made at 

 different depths in this tube during the time 

 when the soil is saturated (in May for this 

 example), the depth of the suspected void can 

 be determined. Further measurements at 

 1-inch intervals may define a soil moisture 

 curve similar to the solid line in figure 9 A. A 

 hypothetical field curve is shown as a solid line 

 in figure 13 and illustrates the type of curve 

 that might be encountered if there was a void 

 partly filled with water. This field curve, which 

 corresponds to the "control," is, of course, 

 unknown; but a control curve may be drawn in 

 after a study of the shape of the control curves 

 in figures 7 through 10. An assumed control 

 curve is shown as the dashed line in figure 13. 

 The error curve is derived in the usual way from 

 the horizontal displacement between the two 

 curves and is shown at the right of figure 13 ; the 

 total measurement error is the area ABC A. The 

 44 percent moisture ordinate is extended 



vertically until it intersects the hypothetical 

 field curve at point D. Then a line is drawn 

 horizontally from D to point E on the error 

 curve. If the error curve is redrawn as the hori- 

 zontal displacement between the control curve, 

 the field curve, and the 44 percent ordinate, 

 then this defines the error curve AEFA. Thus, 

 soil moisture measurements may be made from 

 a depth of 12 inches to a depth of 33 inches 

 including a maximum estimated error of about 

 +8.5 percent (at point E) instead of a maximum 

 error of about 16 percent (at point B). The 

 region of questionable data is shortened from 

 the 12-inch level to about the 25-inch level 

 instead of from the 12-inch level to the 33-inch 

 level. The maximum measurement error is cor- 

 respondingly reduced from +16 to +8.5 percent 

 and the total error is reduced from 0.96 to 0.22 

 inch. 



Figure 14 was derived from a plot of the 

 maximum positive and maximum negative er- 

 rors for both fine and coarse sand — a total of 

 eight points taken from table 2. This graph 

 can be used to estimate the volume of a sus- 

 pected void provided that a field curve of neu- 

 tron measurements can be made while the soil 

 and void are saturated and provided that the 

 saturated pore volume of the soil can be de- 

 termined. In order to use figure 14, it is neces- 

 sary to determine the maximum measurement 

 error from field data. The field data are plot- 

 ted, the saturated pore volume ordinate . is 

 superimposed, as in the left hand portion of 

 figure 13, and the maximum measurement er- 

 ror is taken from this graph. Assume, for 

 example, that the maximum measurement er- 

 ror for a completely saturated void in a coarse 



16 



