Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #430409
Research Project: From Field to Watershed: Enhancing Water Quality and Management in Agroecosystems through Remote Sensing, Ground Measurements, and Integrative Modeling

Location: Hydrology and Remote Sensing Laboratory

Title: Performance of five common soil moisture sensors using default calibration equations in low-salinity conditions

Author
item AZIZI, SEYED ALI - Texas A&M University
item WYATT, BRIANA - Texas A&M University
item PATRIGNANI, ANDRES - Kansas State University
item OCHSNER, TYSON - Oklahoma State University
item Cosh, Michael

Submitted to: Vadose Zone Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/23/2026
Publication Date: 5/10/2026
Citation: Azizi, S., Wyatt, B.M., Patrignani, A., Ochsner, T., Cosh, M.H. 2026. Performance of five common soil moisture sensors using default calibration equations in low-salinity conditions. Vadose Zone Journal. 25(3). Article e70105. https://doi.org/10.1002/vzj2.70105.
DOI: https://doi.org/10.1002/vzj2.70105

Interpretive Summary: Automated electrical sensors are able to provide an estimate of soil moisture which is a valuable land surface parameter for agriculture. However, these sensors are not all of equal accuracy and the factory calibrations can not account for all of the variability cause by soil textures and compositions. A study was conducted to quantify this variability and provide an estimate of the systematic errors for the most common sensors used in the U.S. to estimate soil moisture. Many sensors were compared and several identified as sufficient for most scientific studies, but all sensors benefit from site specific calibration. Clay soils also proved to be the most difficult to estimate. This research impacts the design and management of large scale monitoring networks, such as state mesonets and regional basin networks, who need to manage water resources at a large scale.

Technical Abstract: Soil moisture monitoring plays a critical role in water resource management, hydrological modeling, and climate studies, making sensor accuracy and reliability increasingly important. This study evaluates the performance of five soil moisture sensors- the CS655, HydraProbe, TDR-315N, TEROS 12, and ThetaProbe- using their default calibrations. Using gravimetric ('v) and neutron probe (NP) measurements as benchmarks, we analyzed sensor performance over two years and used statistical metrics including mean absolute error (MAE), root mean square error (RMSE), Nash–Sutcliffe efficiency (NSE), bias, standard deviation (SD), and mean absolute difference (MAD) relative to another sensor. Sensor performance varied across sites and depths, primarily influenced by soil texture and horizon changes. Across all sites and depths, the TEROS 12, TDR-315N, and CS655 showed the lowest overall errors, with mean MAE values ranging from 0.053 to 0.074 m3 m-3 depending on the reference method. Beyond accuracy, the TEROS 12 and TDR-315N also demonstrated the highest level of correlation and repeatability, as shown by their consistently low MAD and SD values. The ThetaProbe exhibited intermediate performance, while the HydraProbe consistently showed the highest errors on average. Positive NSE values and near-zero biases at shallower depths indicated good agreement with reference methods, while deeper clay-rich layers exhibited reduced NSE and larger positive biases, reflecting consistent overestimation by most sensors. Our results emphasize the need for site-specific calibration to reduce systematic errors and improve measurement reliability. These results can guide researchers and decision-makers in selecting soil moisture sensors that balance accuracy, ease of installation, and cost.