Soil Structure and Texture Effects on the Precision of Soil Water Content Measurements with a Capacitance-Based Electromagnetic Sensor

摘要

The physical properties of soil, such as structure and texture, can affect the performance of an electromagnetic sensor in measuring soil water content. Historically, calibrations have been performed on repacked samples in the laboratory and on in situ soils in the field, but little research has been done on laboratory calibrations with intact (undisturbed) soil cores. In this study, three replications each of disturbed and undisturbed soil samples were collected from two soil texture classes (Yutansilty clay loam and Fillmore silt loam) at a field site in eastern Nebraska to investigate the effects of soil structure and texture on the precision of a METER Group GS-1 capacitance-based sensor calibration. In addition, GS-1 sensors were installed inthe field near the soil collection sites at three depths (0.15, 0.46, and 0.76 m). The soil moisture sensor had higher precision in the undisturbed laboratory setup, as the undisturbed calibration had a better correlation [slope closer to one, R~2 _undisturbed (0.89) > R~2 _disturbed (0.73)] than the disturbed calibrations for the Yutan and Fillmore texture classes, and the root mean square difference using the laboratory calibration (RMSD_L) was higher for pooled disturbed samples (0.053 m~3 m~(-3)) incomparison to pooled undisturbed samples (0.023 m~3 m~(-3)). The uncertainty in determination of volumetric water content (theta_v )was higher using the factory calibration (RMSD_F) in comparison to the laboratory calibration (RMSD_L) for the differentsoil structures and texture classes. In general, the uncertainty in estimation of soil water depth was greater than the uncertainty in estimation of soil water depletion by the sensors installed in the field, and the uncertainties in estimation of depthand depletion were lower using the calibration developedfrom the undisturbed soil samples. The undisturbed calibration of soil water depletion would determine water demand mth better precision and potentially avoid over-watering, offering relief from water shortages. Further investigation ofsensor calibration techniques is required to enhance the applicability of soil moisture sensors for efficient irrigation management.