Waveform analysis for short time domain reflectometry (TDR) probes to obtain calibrated moisture measurements from partial vertical sensor insertions

Authors: DAKSHINAMURTHY, HEMANTH NARAYAN - Utah State University; JONES, SCOTT - Utah State University; Schwartz, Robert; YOUNG, SIERRA - Utah State University

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/3/2025
Publication Date: 3/24/2025
Citation: Dakshinamurthy, H., Jones, S.B., Schwartz, R.C., Young, S. 2025. Waveform analysis for short time domain reflectometry (TDR) probes to obtain calibrated moisture measurements from partial vertical sensor insertions. Computers and Electronics in Agriculture. 235. Article 110233. https://doi.org/10.1016/j.compag.2025.110233.
DOI: https://doi.org/10.1016/j.compag.2025.110233

Interpretive Summary: Robots have been designed to measure soil moisture by inserting time domain reflectometry (TDR) sensor rods into the soil. However, the force required for insertion is oftentimes greater than that generated by these small robots. Incomplete insertion leads to air gaps which can result in large underestimation errors in measured permittivities required to estimate soil moisture. A new TDR waveform calibration methodology was developed by scientists from ARS-Bushland and Utah State University to account for incomplete insertion on soil moisture estimation during automated measurements. Calculated permittivities using the proposed method were within the accuracy range of plus or minus 4 percent. This new method for short TDR probes enables the estimation of accurate soil moisture contents when rods are incompletely inserted into the soil, thus making automated soil moisture measurements with robotic technologies more reliable.

Technical Abstract: Time domain reflectometry (TDR) probes have been used for measuring soil moisture extensively in agricultural and environmental water management applications. Short (less than 15 cm) commercial TDR probes can be easily installed in the field to accurately measure soil moisture in situ at regular intervals. When TDR probes are installed manually within the soil, complete sensor insertion and contact with the soil are critical for the accurate determination of soil moisture. Recently, ground and aerial robots have been designed to measure soil moisture autonomously by inserting sensors into the soil; however, when using automated systems, proper sensor insertion cannot always be guaranteed. A lack of complete contact with the soil leads to air gaps and underestimation of soil moisture due to differences in the dielectric permittivity between air (' = 1) and water (' = 80). To address the scenario of air gaps near the sensor head due to partial probe insertion, a new TDR waveform calibration methodology was developed to calculate the percentage of the probe exposed to the air during automated TDR sensor insertions by modifying the waveform analysis approach by Schwartz et al. (2014). To validate this new method, TDR waveforms were collected using a 5 cm TDR 305N (Acclima, Meridian, Idaho, US) probe inserted into different liquid mediums (water and methanol) at 10 percent insertion depth intervals. Similarly, TDR waveforms were recorded for different insertion depths in Millville silt loam soil for different moisture conditions (air-dry, 13 percent saturation, 16 precent saturation, 26 percent saturation, and full saturation). Calculated permittivities using the proposed method were within the manufacturer-specified accuracy range (plus or minus 4 percent Full Scale) in all low permittivity mediums. The method was also applied to field data collected from a UAV payload. In 8 out of 11 incomplete sensor insertions, the model was able to reduce the percent error, and in three complete sensor insertions, the model was able to validate the value obtained by the UAV payload. The study also found that the sensor has to be inserted at least 50 percent in the medium to use the proposed waveform analysis. This new calibration methodology for short TDR probes enables the calculation of apparent permittivity from incomplete sensor insertions to determine the actual moisture content, thus making automated soil moisture measurements with robotic technologies more reliable and robust.