Calibration and Evaluation of a Frequency Domain Reflectometry Sensor for Real‐Time Soil Moisture Monitoring

Ojo, E. RoTimi; Bullock, Paul; L’Heureux, Jessika; Powers, Jarrett; McNairn, Heather; Pacheco, Anna

doi:10.2136/vzj2014.08.0114

Cited by 70 publications

(51 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an effort to understand what soil physical property variable(s) is (are) best to use for calibration of SWC ground sensors (Rowlandson et al 2013;Ojo et al 2015) or to enter into remote-sensing algorithms to interpret the soil dielectric constant (Mladenova et al 2014;Manns et al 2015), it would be helpful to know the amount of variance that is expressed equally by all variables (same r value between any pairwise combinations) separately from the amount to which individual variables express unique variability in specific environmental conditions. The objective of this paper is to analyze the relationships among soil physical variables and in particular to assess the role of SOC as related to surface SWC when averaged over multiple fields and sampling times.…”

Section: Introductionmentioning

confidence: 99%

Evidence of a union between organic carbon and water content in soil

2016

View full text Add to dashboard Cite

Soil organic matter, comprising ∼58% soil organic carbon (SOC), is attributed with increased water holding capacity in the surface horizon of agricultural soil. This paper addresses the role of SOC as a component of a common functional unit in soil from analysis within a single field and over multiple fields. Soil data measured on the fields during the SMAPVEX12 satellite prelaunch algorithm development campaign exhibited high correlation among SOC, field-mean soil water content (SWC), bulk density, and soil texture. The analysis extended over a wide range of soil texture and wetness in the top 5 cm of soil over 50 agricultural fields covering ∼400 km 2 of southern Manitoba. Data collected over a much smaller area from Ontario silt loam soils at the Elora Research Centre demonstrated a similar correlation between SOC and SWC in intensive field sampling. This intercorrelation of SOC and SWC is examined with partial least-squares regression, principal component analysis, and geostatistical semivariograms. A model is proposed to interpret the feedback process between SOC and SWC to explain the persistent correlation. Further work to substantiate the strengths and limits of the relationship between SOC and SWC may be beneficial for estimating SWC for remote sensing, agriculture, hydrology, and ecosystem function.Key words: soil organic carbon, soil water content, organic matter molecule, partial least-squares analysis, carbon feedback.Résumé : On estime que la matière organique, qui se compose d'environ 58 % de carbone organique (CO), retient l'eau davantage dans l'horizon superficiel des sols agricoles. Cet article examine le rôle du CO en tant que composante d'une unité fonctionnelle commune du sol d'après l'analyse du sol d'un ou de plusieurs terrains agricoles. Les données relevées au champ lors du développement de l'algorithme SMAPVEX12 avant le lancement du satellite montrent qu'il existe une étroite corrélation entre le CO, la teneur en eau moyenne du sol (TEM) au champ, la masse volumique apparente et la texture du sol. L'analyse a porté sur un large éventail de textures et de concentrations d'eau dans la couche supérieure de 5 cm d'une cinquantaine de parcelles cultivées couvrant une superficie de ∼400 km 2 dans le sud du Manitoba. Les données recueillies sur une superficie plus modeste constituée de loams limoneux, au centre de recherche d'Elora, en Ontario, montraient une corrélation analogue entre le CO et la TEM pour les parcelles échantillonnées, cultivées intensivement. Les auteurs examinent la corrélation entre le CO et la TEM du sol au moyen de la méthode de régression partielle par les moindres carrés, de l'analyse en composantes principales et de semivariogrammes en géostatistique. Ils proposent un modèle permettant d'interpréter les mécanismes de rétroaction entre le CO et la TEM en vue d'expliquer cette corrélation tenace. Des recherches plus poussées pour étoffer les forces et les limites du lien entre le CO et la TEM du sol auraient leur utilité pour estimer la TEM en vue de son util...

show abstract

Section: Introductionmentioning

confidence: 99%

Evidence of a union between organic carbon and water content in soil

2016

View full text Add to dashboard Cite

show abstract

“…Similar results have been observed in previous investigations of the performance of a frequency domain reflectometry sensor. Ojo et al (2015) found that the results of field calibration of the sensor were superior (r 2 = 0.95) to the laboratory calibration (r 2 = 0.89). On the contrary, Gabriel et al (2010) found that the accuracy of a capacitance probe (EnviroScan, Sentek Pty Ltd., Kent Town, South Australia) was slightly better under field conditions using laboratory calibration equations (RMSD = 0.019 m 3 m -3 ) rather than field conditions (RMSD = 0.023 m 3 m -3 ) and recommended the use of laboratory conditions because they are easily reproducible, facilitate work planning, and minimize uncertainties.…”

Section: Analysis Based On Variable Soil Structurementioning

confidence: 89%

“…The performance of EM soil water sensors under various soil conditions has been investigated extensively (Geesing et al, 2004;Mittelbach et al, 2012;Singh et al, 2018;Varble and Chávez, 2011;Vaz et al, 2013), and some studies have proposed correcting for non-water influences on  v by developing soil-specific calibrations. For soil moisture sensors based on capacitance and frequency domain technology, the sensor response over a large  v range has been captured in the laboratory (Adeyemi et al, 2016;Goswami et al, 2019;Ojo et al, 2015;Provenzano et al, 2016;Santhosh et al, 2017) and in the field (Datta et al, 2018;Huang et al, 2017;Lea-Cox et al, 2018;Ojo et al, 2014;Rudnick et al, 2015;Sui, 2017).…”

mentioning

confidence: 99%

Soil Structure and Texture Effects on the Precision of Soil Water Content Measurements with a Capacitance- Based Electromagnetic Sensor

Singh

Heeren²,

Rudnick³

et al. 2020

Transactions of the ASABE

View full text Add to dashboard Cite

Collection HIGHLIGHTS Capacitance-based electromagnetic soil moisture sensors were tested in disturbed and undisturbed soils.  The uncertainty in estimation of soil water depth was lower using the undisturbed soil sample calibrations.  The uncertainty in estimation of soil water depletion was lower than the uncertainty in volumetric water content.  Undisturbed calibration of water depletion quantifies water demand with better precision and avoids over-watering.ABSTRACT. The physical properties of soil, such as structure and texture, can affect the performance of an electromagnetic sensor in measuring soil water content. Historically, calibrations have been performed on repacked samples in the laboratory and on in situ soils in the field, but little research has been done on laboratory calibrations with intact (undisturbed) soil cores. In this study, three replications each of disturbed and undisturbed soil samples were collected from two soil texture classes (Yutan silty clay loam and Fillmore silt loam) at a field site in eastern Nebraska to investigate the effects of soil structure and texture on the precision of a METER Group GS-1 capacitance-based sensor calibration. In addition, GS-1 sensors were installed in the field near the soil collection sites at three depths (0.15, 0.46, and 0.76 m). The soil moisture sensor had higher precision in the undisturbed laboratory setup, as the undisturbed calibration had a better correlation [slope closer to one, R 2 undisturbed (0.89) > R 2 disturbed (0.73)] than the disturbed calibrations for the Yutan and Fillmore texture classes, and the root mean square difference using the laboratory calibration (RMSD L ) was higher for pooled disturbed samples (0.053 m 3 m -3 ) in comparison to pooled undisturbed samples (0.023 m 3 m -3 ). The uncertainty in determination of volumetric water content ( v ) was higher using the factory calibration (RMSD F ) in comparison to the laboratory calibration (RMSD L ) for the different soil structures and texture classes. In general, the uncertainty in estimation of soil water depth was greater than the uncertainty in estimation of soil water depletion by the sensors installed in the field, and the uncertainties in estimation of depth and depletion were lower using the calibration developed from the undisturbed soil samples. The undisturbed calibration of soil water depletion would determine water demand with better precision and potentially avoid over-watering, offering relief from water shortages. Further investigation of sensor calibration techniques is required to enhance the applicability of soil moisture sensors for efficient irrigation management.

show abstract

“…Numerous sensor performance and performanceinfluencing factors experiments have been conducted under laboratory and field conditions (Leib et al, 2003;Plauborg et al, 2005;Chandler et al, 2004;Ojo et al, 2014Ojo et al, , 2015Miller et al, 2014;Soulis et al, 2015;Visconti et al, 2014;Jabro et al, 2017;Kargas et al, 2019). Vaz et al (2013) evaluated eight commercially available electromagnetic water content sensors (TDR100, CS616, Theta Probe, Hydra Probe, SM300, Wet2, 5TE, 10HS) in soils, ranging from sand to clay, including an organic soil.…”

mentioning

confidence: 99%

Time-domain and Frequency-domain Reflectometry Type Soil Moisture Sensor Performance and Soil Temperature Effects in Fine- and Coarse-textured Soils

Zhu¹,

Irmak²,

Jhala³

et al. 2019

Applied Engineering in Agriculture

View full text Add to dashboard Cite

The performances of six time-domain reflectometry (TDR) and frequency-domain reflectometry (FDR) type soil moisture sensors were investigated for measuring volumetric soil-water content ( v) in two different soil types. Soilspecific calibration equations were developed for each sensor using calibrated neutron probe-measured  v. Sensors were also investigated for their performance response in measuring  v to changes in soil temperature. The performance of all sensors was significantly different (P<0.05) than the neutron probe-measured  v , with the same sensor also exhibiting variation between soils. In the silt loam soil, the 5TE sensor had the lowest root mean squared error (RMSE) of 0.041 m 3 /m 3 , indicating the best performance among all sensors investigated. The performance ranking of the other sensors from high performance to low was: TDR300 (High Clay Mode), CS616 (H) and 10HS, SM150, TDR300 (Standard Mode), and CS616 (V) (H: horizontal installation and V: vertical installation). In the loamy sand, the CS616 (H) performed best with an RMSE of 0.014 m 3 /m 3 and the performance ranking of other sensors was: 5TE, CS616 (V), TDR300 (S), SM150, and 10HS. When  v was near or above field capacity, the performance error of most sensors increased. Most sensors exhibited a linear response to increase in soil temperature. Most sensors exhibited substantial sensitivity to changes in soil temperature and the  v response of the same sensor to high vs. normal soil temperatures differed significantly between the soils. All sensors underestimated  v in high temperature range in both soils. The ranking order of the magnitude of change in  v in response to 1°C increase in soil temperature (from the lowest to the greatest impact of soil temperature on sensor performance) in silt loam soil was: SM150, 5TE, TDR300 (S), 10HS, CS620, CS616 (H), and CS616 (V). The ranking order from lower to higher sensitivity to soil temperature changes in loamy sand was: 10HS, CS616 (H), 5TE, CS616 (V), SM150, and TDR300 (S). When the data from all sensors and soils are pooled, the overall average of change in  v for a 1°C increase in soil temperature was 0.21 m 3 /m 3 in silt loam soil and-0.052 m 3 /m 3 in loamy sand. When all TDR-and FDR-type sensors were pooled separately for both soils, the average change in  v for a 1°C increase in soil temperature for the TDR-and FDR-type sensors was 0.1918 and-0.0273 m 3 /m 3 , respectively, indicating that overall TDRtype sensors are more sensitive to soil temperature changes than FDR-type sensors when measuring  v .

show abstract

Calibration and Evaluation of a Frequency Domain Reflectometry Sensor for Real‐Time Soil Moisture Monitoring

Cited by 70 publications

References 26 publications

Evidence of a union between organic carbon and water content in soil

Evidence of a union between organic carbon and water content in soil

Soil Structure and Texture Effects on the Precision of Soil Water Content Measurements with a Capacitance- Based Electromagnetic Sensor

Time-domain and Frequency-domain Reflectometry Type Soil Moisture Sensor Performance and Soil Temperature Effects in Fine- and Coarse-textured Soils

Contact Info

Product

Resources

About