|Schneider, Arland - USDA-ARS, RETIRED|
Submitted to: World Congress of Soil Science
Publication Type: Proceedings
Publication Acceptance Date: February 4, 2002
Publication Date: August 1, 2002
Interpretive Summary: Accurate soil water content measurements are required for irrigation management, measurement of crop water use, conservation of water, and in many fields of engineering and environmental science. Repeated, nondestructive measurements are most easily and inexpensively made by measuring from inside tubes installed vertically in the soil, dam, or other rearthwork. We compared the precision, accuracy and measurement volume, in three different soils, of instruments currently marketed for such measurements. Instruments that measured soil electrical capacitance as a surrogate for water content were typically sensitive to soil temperature changes and measured small volumes, making them susceptible to inaccuracies caused by soil disturbance during tube installation. Also, their accuracies were typically not as good as those obtainable with the soil moisture neutron probe (SMNP), which has a much larger measurement volume (approx. 6 6to 18 inches radius) and is less susceptible to soil disturbance. The stud is ongoing. But, one conclusion is that the SMNP should continue to be used for routine soil water measurements, and its use should be encouraged.
Technical Abstract: Accurate soil water content measurements are required to measure crop water use and soil hydraulic characteristics. We compared the Sentek EnviroSCAN and Diviner capacitance devices, the Delta-T Profiler capacitance probe, the Trime T3 tube-probe, and the soil moisture neutron probe (SMNP). All the devices were used in access tubes. Triplicate soil columns of three soils (silty clay loam, clay, and calcic clay loam containing 50% CaCO3) were each weighed to 50-g precision. Conventional time domain reflectometry (TDR) measurements of water content and thermocouple measurements of temperature were made at several depths in each column. Depth resolution of each device was investigated by lowering a sensor from a height 30-cm above the soil surface and taking measurements at 2-cm increments until the probe was 30-cm below the soil surface. Comparisons of soil water content reported by the devices vs. soil temperature were made before the columns were wetted. The capacitance sensors were significantly sensitive to soil temperature (0.0005 to 0.0010 m**3/m**3/degree C); but TDR and the SMNP were not. The Trime device was most sensitive to temperature (0.009 m**3/m**3/degree C). The conventional TDR and EnviroSCAN systems were most accurate, giving readings within 0.02 m**3/m**3 of actual, with the Diviner being slightly less accurate. The Delta-T and Trime systems were most inaccurate, reporting water contents ranging up to 0.09 and 0.06 m**3/m**3 larger than actual, respectively. The capacitance systems and the Trime T3 generally were not sensitive to soil volumes outside of their respective sensor heights, indicating small measurement volumes, and suggesting that these systems may be susceptible to soil disturbance close to the access tube during installation.