Location: Northwest Watershed Management Research
Title: Response of three soil water sensors to variable solution electrical conductivity in different soils Authors
|Kargas, G -|
|Kerkides, P -|
Submitted to: Vadose Zone Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: July 7, 2014
Publication Date: N/A
Interpretive Summary: In recent years a number of soil water sensors have been developed. One important application is in irrigation management. By irrigating only when soil water status indicates the need, it is possible to increase both efficiency and yield. A critical factor that strongly affects instrument performance is the soil electrical conductivity, which is often directly related to the soil salt content. This is an important consideration in irrigated soils, where salt accumulation is a common problem. We evaluated the performance of three different soil water sensors in a combination soils and salt concentration. We found sensor sensitivity to salt concentration only in sand soils, which have essentially no natural conductivity. However, only one sensor type was both insensitive to soil type and robust with respect to the calibration equation used. The other two sensors were either highly sensitive to soil type or required the use of a special calibration equation. These results also illustrate the importance of instrument testing.
Technical Abstract: Commercial dielectric soil water sensors may improve management of irrigated agriculture by providing continuous field soil water information. Use of these sensors is partly limited by sensor sensitivity to variations in soil salinity and texture, which force expensive, time consuming, soil specific calibration. The impacts of both salinity and texture diminish with sensor measurement frequency so that a single calibration equation (TOPP), applies to a wide range of soils and salinities. We evaluated the calibration response of three commercial sensors with measurement frequencies ranging from 20 to 100 MHz using three salt solutions with electrical conductivities (ECs) ranging from 1.2 to 6 dS m-1 in six soils ranging in texture from sand to clay. Standard, multi-point (6 to 8 water contents) and the much simpler two-point (CAL) calibration procedures were also compared. The only ECs response was in the sand using the lowest measurement frequency sensor. This sensor was also most sensitive to variations in soil type (textures). The remaining sensors were relatively insensitive to soil type. The highest frequency sensor data were in close agreement with the TOPP calibration equation. The CAL procedure was highly effective for all soils and sensors. In general, we can expect all three sensors to be effective in most soils if the ECs is 6 dS m-1 or less. In terms of calibration, measurement frequency alone is not sufficient to predict sensor calibration so that some empirical evaluation remains critical.