Location: Watershed Management ResearchTitle: Comparison of two dielectric sensors for soil water measurements) Author
Submitted to: Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 6/1/2011
Publication Date: 6/29/2011
Citation: Kargas, G., Kerkides, P., Seyfried, M.S. 2011. Comparison of two dielectric sensors for soil water measurements. VI EWRA International Symposium Water Engineering and Management in a Changing Environment, Catania, Italy, June 29 – July 2, 2011. Interpretive Summary:
Technical Abstract: Soil moisture is an important factor in several hydrological processes and a key variable for land degradation, flood generation, and drought mitigation, especially under present conditions of eminent climatic perturbations and frequent occurrences of extreme events. Soil moisture has also an important effect on the partitioning of precipitation in runoff, infiltration and groundwater recharge. Recently new dielectric devices, estimating soil water content,', from the dielectric constant of the soil, e, have been manufactured and promoted, which cheaply, easily and non-destructively provide estimates of ' on a routine basis. Before using these devices, a calibration procedure is strongly recommended and also an investigation is needed for their safe '-estimation and accuracy limits under various conditions, regarding soil water content uniformity, presence of salts, soil structure mineralogy etc. Two such dielectric devices, namely the WET and the ML2-theta probe have been studied, which operate at frequencies 20MHz and 100MHz respectively. The results concerning their calibration and performance are presented. From the results it is shown that the WET-measured e increases with the increase of salinity. On the contrary ML2-measured e is found to depend less on salinity. Prediction of ' by the WET sensor appears to be accurate with errors falling in the prescribed by the manufacturers’ limits. This is not so for sandy soil where there is an overestimation of '. For the ML2 theta probe, again, prediction of ' appears to be accurate with errors falling in the prescribed limits, except for the case of sandy loam soil where there is an overestimation. The 'e versus independently measured 'mis strongly linear for the two sensors studied. Specific calibration improves the results.