Soil material, temperature, and salinity effects on calibration of multisensor capacitance probes

Multisensor capacitance probes (MCAP) are an alternative to gravimetric or nuclear soil water content (theta (v), m(3) m(-3)) measurements. Their theta (v) measurements are more convenient than gravimetric, and don't carry the nuclear regulatory burdens. Previous studies noted potential salinity and temperature effects on MCAP theta (v) determinations. Our objectives were to calibrate and verify MCAP theta (v) measurement accuracy in two soil materials, two water salinities (1.3 and 11.3 dS m(-1)), and with diurnal temperature fluctuations. The surface and calcic horizons of an Olton soil (fine, mixed, superactive, thermic Aridic Paleustoll) were packed into triplicate, 0.5-m-tall, 100-L columns and wetted. We compared theta (v) determined by volumetric measurements, time domain reflectometry (TDR), and MCAPs. The TDR theta (v) were within +/-0.01 m(3) m(-3) of volumetric determinations for air-dry and saturated soil. The factory supplied universal MCAP calibration provided accurate theta (v) estimates for air dry (+/-0.01 m(3) m(-3)) surface and calcic soil materials but not after wetting (approximate to -0.05 m(3) m(-3)). Also, imprecise MCAP sensor positioning during water frequency parameter determination was problematic and biased initial theta (v) measurements. After calibration against TDR, the MCAP theta (v) varied +/-0.01 m(3) m(-3) from measured theta (v) for air-dry and saturated conditions for both soil materials, which were then pooled to obtain one calibration. Column resaturation with saline water affected permittivity and elevated MCAP theta (v) approximate to0.25 m(3) m(-3) above the available pore space. Cyclical soil temperature fluctuations of 15 degreesC induced similar fluctuations in indicated theta (v) throughout the column (0.04 m(3) m(-3) for MCAP and 0.02 m(3) m(-3) for TDR), which was attributed to variations in permittivity.