|Evett, Steven - Steve|
Submitted to: World Congress of Soil Science
Publication Type: Proceedings
Publication Acceptance Date: 1/3/2002
Publication Date: 8/2/2002
Interpretive Summary: Several new devices for measurement of soil profile water content were compared with the soil moisture neutron probe (SMNP), which has been the standard for undisturbed measurements for many years. Measurements were made in soils in the United States, Austria, Australia, France, and Tunisia. The new devices all work on the idea of measuring the electrical properties of the soil. These properties vary as soil wets and dries, but are also affected by soil salinity and temperature. All the new devices were less precise than the SMNP, in many situations, and were sensitive to temperature and salinity effects that do not affect the accuracy of the SMNP. The SMNP responds to changes in the concentration of hydrogen in the soil. Because most of the changes in hydrogen are the result of changes in water content, the SMNP can be calibrated to give accurate water content values. The SMNP is still the most practical device for soil profile water content measurements. The SMNP employs a very small radioactive source sealed in doubly welded stainless steel cylinders. Over 50 years of use have proven this device entirely safe for routine use.
Technical Abstract: Neutron scattering, capacitance, and time domain reflectometry (TDR) methods of soil water content measurement were compared using measurements made in a soil in Australia, three soils in Austria, five soils in France, two soils in Tunisia, and three soils in the United States. Except for conventional TDR, the devices were used in access tubes. Devices tested included the soil moisture neutron probe (SMNP), Sentek EnviroSCAN and Diviner 2000 capacitance probes, the IMKO Trime tube probe, and conventional TDR systems from Soil Moisture Inc., Tektronix, Inc. and Dynamax, Inc. The Sentek, IMKO, Delta-T and SMNP devices all required soil specific calibration. The conventional TDR systems were reasonably accurate without calibration. Due to their small measurement volume, installation of access tubes without soil disturbance outside of the tube was critical for success with the Sentek, IMKO, and Delta-T devices, but not for the SMNP. Successful tube installation was difficult with the smaller diameter devices (<30 mm), and was difficult in some soils for all devices except the SMNP. Preliminary calibration vs. gravimetric sampling resulted in coefficient of determination values of 0.42 and 0.53 in two Austrian soils for the Diviner. The Sentek, IMKO and Delta-T devices produced values with more scatter than the SMNP. All devices, excepting SMNP and TDR, were sensitive to soil temperature. Both the Diviner and EnviroSCAN devices were highly sensitive to the electrical conductivity of soil water, with a 5% change in water content caused by an EC increase of 5 dS/m. Tests in Australia showed the EnviroSCAN to overestimate water content near saturation and to underestimate near wilting point. The SMNP is still the most practical device for soil profile water content measurements.