Calibration and Use of Neutron Moisture and Gamma Density Probes in Rocky Soils

In situ measurement of water content in rocky soils is difficult because rocks hinder insertion of probes and alter physical properties of the soil used to detect water content. We investigated the impact of rock on neutron probe measurements of volumetric water content (theta), and on gamma densitometry measurements of wet bulk density (rho(wet)). Measurements were done in a drum packed with a clay loam having a range of theta and rock density. The annular space between access tube and hole was sealed with expandable polyurethane foam to replicate a field installation in a karst savanna. The relationship between neutron probe count ratios (CRn) and theta, obtained with different fractions of rock, was nonlinear, and nearly identical to that obtained with soil only, indicating that a single calibration curve could be used, regardless of the amount of rock in the soil profile. The nonlinearity was caused by the foam-filled void between tube and hole, which affected measurement geometry. Gamma count ratios (CRd) decreased with increasing wet bulk density, as predicted by theory. However, the foam sealant resulted in higher CRd than predicted by theory because it reduced absorption and increased scattering of gamma particles. A utility of using both methods is that dry bulk density can be estimated by subtracting the density of water in the soil volume from rho(wet), allowing locations of the soil profile to be identified where high rock density may restrict water flow and root water uptake.