Soil Water Sensor Performance and Corrections with Multiple Installation Orientations and Depths under Three Agricultural Irrigation Treatments

Performance evaluations and corrections of soil water sensors have not been studied using different installation orientations under various irrigation treatments in the Texas High Plains. This study evaluated the performance of four sensors using factory calibration and derived field corrections as compared to calibrated neutron moisture meters (NMMs). Sensor performance was assessed using horizontal insertion, laid horizontal placement, and vertical insertion at 15.2, 45.7, and 76.2 cm depths in a clay loam soil with three irrigation treatments. Results indicated the factory-calibrated Acclima 315 L performed satisfactorily using horizontal insertion as compared to NMM measurements at 45.7 and 76.2 cm depths with a ±2% mean difference (MD) and <3.5% root mean square error (RMSE). The factory-calibrated Acclima 315 L using horizontal insertion also performed satisfactorily across all irrigation treatments according to soil profile water storage (MD = 0.36% and RMSE = 3.25%). Generally, the factory-calibrated Decagon GS1 and Campbell Scientific 655 using vertical insertion agreed more closely with NMM measurements compared with other installation orientations. There was a significant underestimation of water storage (>60 mm) in the 0.9 m soil profile using the Watermark 200SS. In summary, field corrections are required for Decagon GS1, Campbell Scientific 655, and Watermark 200SS sensors.

. Illustration of the multiple sensor installation orientations including laid horizontal, vertical insertion, and horizontal insertion at the study site. Note the extensive amount of wiring leads in the trench with the sensor array.

Neutron Moisture Meter (NMM)
Campbell Pacific Nuclear (CPN) model 503DR NMMs were used in this study (InstroTek, Inc., Raleigh, NC, USA) to measure θv (m 3 m −3 ). The meter consists of a radioactive source and detector. The source is lowered into the soil through an access tube, and measurements are recorded for desired depths as determined by stops installed on the source cable. Measurements at depths of 10,30,50,70,90,110,130,150,170,190,210, and 230 cm below the ground surface were recorded in this study although a maximum rooting depth of 150 cm was determined for corn. The use of a depth control stand served to control probe depth relative to the soil surface and allowed for successful use at shallow soil depths [1]. Fast neutrons are scattered into the surrounding soil and lose energy and slow after colliding with hydrogen nuclei. The detector counts the low energy "slow" neutrons, and the Neutron Count is compared to a Standard Count to estimate θv using a linear calibration equation with a slope of "a" and intercept of "b".
The Standard Count is also used to check NMM performance and verifies that the detector is operating correctly. Typically, the NMM is not sensitive to variations of soil temperature and salinity [2]. However, it can be affected by soil clay content, organic matter content, texture, and chemical elements [3]. Therefore, a site-specific calibration with the thermogravimetric method is recommended. The NMMs used in this study were previously field calibrated for the Pullman clay loam soil at Bushland using the thermogravimetric method for the soil horizons of Ap (0-10 cm), Bt (30-110 cm), and Btk (130-230 cm) with a 1.0% accuracy [4]. Following on-site calibrations, NMMs can be used to evaluate other soil water monitoring devices [5].

Acclima 315L (ACC-315L)
The Acclima 315L (Acclima, Inc., Meridian, ID, USA) is a Time Domain Reflectometer (TDR) sensor with three parallel rods serving as the waveguide. The sensor head contains the necessary electronics and firmware to generate an electromagnetic (EM) pulse and construct a waveform to calculate the travel time of the EM wave, which is used to estimate the soil apparent permittivity (Ka; unitless) [6]. The sensor also measures soil temperature. The ACC-315L measures apparent electrical conductivity (EC) according to the Giese and Tiemann method [7]. The sensor outputs θv (m 3 m −3 or %) from 0 to 100% with a resolution of ±0.1%, soil temperature from −40 to 60 °C with a resolution of ±0.1 °C, and soil EC from 0 to 5000 µS cm −1 with an accuracy dependent upon the reading range. The following equation is used to determine θv from Ka [8]. The CS-655 also measures soil temperature. The sensor can measure θv ranges from 5% to 50% with an accuracy of ±3%. Soil EC measurements vary from 0 to 8000 µS cm −1 with an accuracy of ± (5% of reading + 50 µS m −1 ). Soil temperature measurement range is −10 to +70 °C with an accuracy of ±0.5 °C.

Watermark 200SS (WM-200SS)
The Watermark 200SS (Irrometer, Inc., Riverside, CA, USA) is an electrical resistance device. The resistant electrodes are embedded within a granular matrix, and an electrical current is applied to obtain a resistance value. The water content resistance (SWR) changes as the water in the granular matrix changes in response to soil water tension changes. This resistance is measured with a voltage divider circuit with known resistances and correlated to resistance in centibars (cb) or kilopascals (kPa). The sensor outputs soil tension over a range of 0 to 200 kPa. To compare WM-200SS sensor with other sensors, a fitting equation (θv = 38.14 × SWR −0.14 ) adopted from Varble and Chávez [10] for a loamy sand soil was applied to estimate θv from soil water resistance from the WM-200SS. It is acknowledged that there is uncertainty by using this fitting equation. Figure S2. Graphical and statistical comparisons of sensor and neutron moisture meter-derived soil profile water storage values in the upper 0.9 m soil profile under the 100% crop evapotranspiration treatment. MD indicates mean difference; RMSE indicates root mean square error; * indicates a significant difference at p < 0.05; ** indicates a significant difference at p < 0.01.