PERFORMANCE OF THE WATERMARK. GRANULAR MATRI X SENSOR IN SANDY SOILS

Abstract

The Watermark
granular matrix sensor (GMS) is relatively inexpensive and provides a continuous and indirectmeasurement of soil matric potential (SMP). By using five different calibration equations, the performance of model 200SSWatermark GMS for estimating SMP from the sensor resistance was evaluated for two different sandy soils (Dothan loamyfine sand and Astatula fine sand). The performance of the calibration equations was compared with the mercurymanometertensiometer measured SMP. A simple optimization procedure was used to reparameterize existing calibrationequations for estimating SMP from sensor resistance in two selected sandy soils. An approximate relationship between sensorresistance versus soil water content (SWC) was determined for the Dothan soil.<br><br>All calibration equations published to date significantly overestimated (less negative or wetter soil) SMP in both soils.The poor performance of the calibration equations was found to be more severe in the Astatula soil, which had a coarsertexture than the Dothan soil. The SMP calibration equations developed in this study for the two soils showed high correlationbetween sensor resistance and mercury manometertensiometer measured SMP. A linear resistanceSMP relationship wasobserved within the ranges of approximately 10 to 80 kPa and 11.5 to 23 kPa for the Dothan and Astatula soils,respectively. The GMS did not respond to changes in SMP at potentials higher than approximately 10 and 11.5 kPa in theDothan and Astatula soils, respectively. Optimization results showed that calibration equations with optimized parameterscan successfully be used to estimate SMP for the soil in which the parameters were optimized. However, applying the sameequation with the optimized parameters to estimate SMP in the other soil type resulted in poor SMP estimates. The WatermarkGMS was calibrated to estimate SWC from the sensor resistance for Dothan soil between the SWC range of approximately15 to 41.2% on volumetric basis. The relationship between SWC and GMS resistance was described by an exponential decayfunction.<br><br>It was concluded that the upper limits of the Watermark GMS (10 and 11.5 kPa) obtained in this study present a limitationof using Watermark GMS for irrigation scheduling in sandy soils where 10 kPa is often used as a point for triggeringirrigations. However, the results showed that the GMS can be used to estimate SMP from sensor resistance at potentials lowerthan 10 and 11.5 kPa SMP in Dothan and Astatula sandy soils, respectively, by using the calibration equations developedin this study.