A Novel and Non-Invasive Approach to Evaluating Soil Moisture without Soil Disturbances: Contactless Ultrasonic System

Abstract

Soil moisture has been considered a key variable in governing the terrestrial ecosystem. However, it is challenging to preserve indigenous soil characteristics using conventional soil moisture monitoring methods that require maximum soil contacts. To overcome this issue, we developed a non-destructive method of evaluating soil moisture using a contactless ultrasonic system. This system was designed to measure leaky Rayleigh waves at the air–soil joint-half space. The influences of soil moisture on leaky Rayleigh waves were explored under sand, silt, and clay in a controlled experimental design. Our results showed that there were strong relationships between the energy and amplitude of leaky Rayleigh waves and soil moisture for all three soil cases. These results can be explained by reduced soil strengths during evaporation processes for coarse soil particles as opposed to fine soil particles. To evaluate soil moisture based on the dynamic parameters and wave properties obtained from the observed leaky Rayleigh waves, we used the random forest model. The accuracy of predicted soil moisture was exceptional for test data sets under all soil types (R2 ≥ 0.98, RMSE ≤ 0.0089 m3 m−3). That is, our study demonstrated that the leaky Rayleigh waves had great potential to continuously assess soil moisture variations without soil disturbances.

Contactless estimation of soil moisture using leaky Rayleigh waves and a fully convolutional network

“…Several studies have explored the influences of soil moisture on soil surface waves (Flammer et al., 2001; Lo et al., 2007; Lu & Sabatier, 2009; Woo et al., 2022; Pan et al., 2022). Lu and Sabatier (2009) found that other environmental factors, compared to soil water potential, had less influence on the sound speed in the soil, as sound speeds exhibited a power law dependency on soil water potentials.…”

“…The signals were averaged over 100 consecutive events to minimize unintended bias and error, and we saved the signals every 5 min. Those who require a detailed understanding of the ultrasonic system, including circuit configurations and equations, can refer to previous studies (Hong & Choi, 2021; Hong et al., 2020; Woo et al., 2022). To account for the slight variation in speed of ultrasonic waves between the air and soil, particularly under saturated clay, an acoustic barrier was placed between the transmitter and receivers, although the speed of the leaky Rayleigh wave was contingent upon the degree of soil moisture (Dashwood et al., 2020; Yang, 2002).…”

“…The transmitter was situated 0.3 m away from the receivers, while the distance between the transmitter and the acoustic barrier was 0.2 m. The structural frames, sensor cases, and acoustic barrier were 3D printed using a Sindoh 3dwox 1x. The dimensions of the acoustic barrier were 0.05 m (length) × 0.08 m (width) × 0.045 m (height), and it was composed of four layers located at a distance of 0.07 m. An incident angle of 20° was used based on incident‐angle experiments to maximize the amplitude of ultrasonic waves (Woo et al., 2022). The obtained signals were sent to a data acquisition device (NI USB‐6366) to convert analog signals into digital signals.…”

“…In order to investigate the possibility of utilizing leaky Rayleigh waves for non‐invasive soil moisture estimation, we designed a non‐contact ultrasonic system that built upon a previous study (Woo et al., 2022). The developed system is illustrated in a schematic diagram in Figure 1.…”