Use of the dual-probe heat-pulse technique to monitor soil water content in the vadose zone

The dual-probe heat-pulse (DPHP) technique is emerging as a useful technique for measuring soil volumetric water content (theta). However, few published data are available regarding the performance of the DPHP technique under field conditions. The objective of this study is to evaluate the effectiveness of the DPHP technique for measuring theta under field conditions. We used 24 DPHP sensors to monitor theta in a soybean [Glycine max (L.) Merr.] field during the 2001 and 2002 growing seasons. The DPHP sensors demonstrated durability in field conditions and clear sensitivity to temporal and spatial variations of theta at the scale of measurement. The mean theta measured by the DPHP sensors (theta(DPHP)) was on average 0.040 m3 m-3 larger than the mean theta measured by soil sampling (theta(SS)). The response of the DPHP sensors was linear. Regressions of theta(DPHP) vs. theta(SS) yielded r2 values of 0.949 and 0.843 at depths of 7.5 and 37.5 cm. The DPHP technique showed good resolution with RMSE values for the regression of 0.009 and 0.011 m3 m-3 at the two measurement depths. The slopes of the regressions were 0.75 rather than 1.0. Errors in theta(SS) are a likely cause of this low slope. We shifted all the theta values for each sensor up or down by a constant value to make the first theta measurement from each sensor equal theta determined from soil sampling near that sensor at the time of installation. This simple matching point procedure improved the accuracy of the DPHP technique, resulting in a -0.024 m3 m-3 average difference between theta(DPHP) and theta(SS). Also, the matching point procedure markedly reduced the variability between sensors, reducing the average SD from 0.063 to 0.026 m3 m-3. This procedure requires no additional soil sampling and is recommended for field applications of the DPHP technique.