Nondestructive assessment of rangeland soil depth to petrocalcic horizon using electromagnetic induction

The distribution of soil depth to cemented subsoil horizons is important for land evaluation and management, particularly iii arid and semiarid rangelands, where plant growth is limited by soil depth. Traditional methods of quantifying soil depth distribution are destructive, time-consuming, and costly. Therefore nondestructive, efficient, and inexpensive methods for assessing soil depth are needed. We investigated the effectiveness of electromagnetic induction for nondestructively assessing the distribution of soil depth to petrocalcic horizon in a semiarid rangeland landscape in central Utah. Apparent conductivity was measured with the EM31 and EM38 electromagnetic induction meters in an 11.5-ha survey area and in four 0.1-ha plots. Soil depth was measured in hand-excavated pits or in hand-augered holes. Rangeland soil depth was positively and significantly correlated, with apparent conductivity. Apparent conductivity measurements made by different operators were not significantly different, indicating that different operators should not affect precision. Apparent conductivity measurements made on different days about 3 weeks apart were significantly different, indicating that regressions of apparent conductivity and soil depth should be recalibrated if data are collected over extended time periods. Plant cover measured in 1-m-diameter subplots was not significantly correlated with apparent conductivity and therefore does nor affect the ability of the EM38 to measure apparent conductivity. In a validation subset, apparent conductivity did not accurately predict actual soil depth, nor did it predict the correct distribution of soil depth classes, but it provided rudimentary information about soil depth distribution (<50 versus greater than or equal to 50 cm) that could be useful to land managers.