Adjusting Temperature and Salinity Effects on Single Capacitance Sensors
Pedosphere 19(5): 588–596, 2009 ISSN 1002-0160/CN 32-1315/P c 2009 Soil Science Society of China Published by Elsevier Limited and Science Press Adjusting Temperature and Salinity Effects on Single Capacitance Sensors∗1 A FARES1,∗2 , M SAFEEQ1 and D M JENKINS2 Department of Natural Resources and Environmental Management, University of Hawaii at Manoa, 1910 East West Road, Honolulu, HI 96822 (USA) Department of Molecular Biosciences and Bioengineering, University of Hawaii at Manoa, 1955 East West Road, Honolulu, HI 96822 (USA) (Received March 9, 2009; revised July 21, 2009) ABSTRACT Several newly developed capacitance sensors have simplified real-time determination of soil water content Previous work has shown that salinity and temperature can affect these sensors, but relatively little has been done to correct these effects The objectives of this study were to evaluate the effect of media temperature and salinity on the apparent water content measured with a single capacitance sensor (SCS), and to mitigate this effect using a temperature dependent scaled voltage technique under laboratory conditions A column study was conducted containing two media: pure deionized water and quartz sand under varying water contents (0.05 to 0.30 cm3 cm−3 ) and salinity (0 to 80 mmol L−1 ) Media temperature was varied between and 45 ◦ C using an incubator The SCS probes and thermocouples were placed in the middle of the columns and were logged at an interval of minute There was strong negative correlation between sensor reading and temperature of deionized water with a rate of −0.779 mV ◦ C−1 Rates of SCS apparent output were 0.454 and 0.535 mV ◦ C−1 for air in heating and cooling cycles, respectively A similar positive correlation with temperature was observed in sand at different water contents The SCS probe was less sensitive to temperature as salinity and water content increased Using a temperature-corrected voltage calibration model, the effect of temperature was reduced by 98% An analytical model for salinity correction was able to minimize the error as low as ± 2% over the salinity level tested Key Words: calibration, ECH2 O probes, quartz sand, water content Citation: Fares, A., Safeeq, M and Jenkins, D M 2009 Adjusting temperature and salinity effects on single capacitance sensors Pedosphere 19(5): 588–596 INTRODUCTION Soil water content is a major factor that determines plant growth and solute transport in irrigated and non-irrigated systems Besides crop, climate, soil water retention characteristics and soil hydraulic properties, optimum irrigation scheduling requires accurate knowledge of real-time soil water content The most accurate method of measurement of soil water content has been the gravimetric method (Gardner, 1986) However, in the past few decades several nondestructive methods have been derived to monitor soil water content including neutron thermalization (Greacen, 1981), electrical resistance (Spaans and Baker, 1992; Seyfried, 1993), time domain reflectometry (TDR) (Topp et al., 1980; Cassel et al., 1994), and electrical capacitance (Robinson and Dean, 1993; Fares and Polyakov, 2006) By comparison, the gravimetric method is destructive, laborious, and does not allow for real-time measurement of water content Several site specific calibration studies have been conducted for different capacitance soil water sensors to improve the accuracy of their measurement Capacitance soil water sensors respond to the dielectric permittivity (ε) of a soil-water-air mixture, and estimate soil water content from this response ∗1 Project supported by a grant from the U.S Department of Agriculture Cooperative State Research, Education and Extension Service, USA (No 2008-34135-19408) ∗2 Corresponding author E-mail: afares@hawaii.edu ADJUSTING TEMPERATURE AND SALINITY EFFECTS 589 (Fares et al., 2007) The ε of water (78.54 at 22 ◦ C) is large compared with those of the soil matrix (ε < 10) and air (ε ≈ 1), and thus dominates the air-soil-water mixture However, great variability and sensitivity of the ε of soil minerals (4–9) and dry plant tissue (1–4) make it necessary to calibrate these sensors for a particular soil and if possible for each horizon (Baumhardt et al., 2000) The effect of salinity and temperature on soil water permittivity, which is a constant of proportionality between electric displacement and electric field intensity, is a function of soil mineralogy, texture, and bulk density (Mead et al., 1995) The effect of soil temperature on capacitance systems has been reported for different soil types In a diurnal soil temperature fluctuation experiment of Olton soil, Baumhardt et al (2000) reported a temperature effect on multisensor capacitance probe readings They found a positive relationship between capacitance sensor reading and soil temperature Evett et al (2006) reported variable sensitivity for different electromagnetic based soil water content sensors Two capacitance sensors, EnviroSCANTM and Diviner 2000 (Sentek Sensor Technologies, Stepney, SA, Australia), were moderately sensitive to temperature at the saturated end; however, the Delta-T PR1/6 (Delta-T Devices, Cambridge, UK) and Trime T3 (IMKO Micromodultechnik GmbH, Ettlingten, Germany), a capacitance and a TDR sensor, respectively, were quite sensitive to temperature fluctuations Fares et al (2007) successfully reduced the temperature effect from a multi-capacitance probe (MCP) by 92% using a temperature dependent scaled frequency technique Seyfried and Murdock (2001) showed relatively smaller temperature effects on reflectometer reading in four different soils The effect was strongly positive resulting in a large apparent water content variation across a 40 ◦ C temperature change High salt content in soil has previously been shown to lead to overestimation of water content (Noborio et al., 1994; White et al., 1994) Hence, concerns over sensor accuracy are most likely going to occur under high salinity conditions and variable soil temperature The objectives of this study were to: i) evaluate the effect of media temperature and salinity on the apparent water content measured with a single capacitance sensor (SCS), and ii) mitigate this effect using a temperature dependent scaled voltage technique under laboratory conditions MATERIALS AND METHODS Description of SCS probe Single capacitance sensor ECH2 O probes model EC-20 (Decagon Devices, Inc., Pullman WA, USA) were used in this study The SCS probe sensor has low power requirement (2.5 to 5V-DC; to mA) and the output voltage is 10%–40% of excitation voltage The ECH2 O probes are flat electrode capacitance sensors consisting of copper electrodes positioned in one plane and sealed in epoxy-impregnated fiberglass The electrodes have no direct contact with the medium and the electromagnetic field generated by the electrodes extends through the fiberglass and into the medium surrounding the probe The cm zone of influence extends from the flat surface of the probe, decreasing with distance The probe averages the volumetric water content over its entire length and has little sensitivity at the extreme edges (Fares and Polyakov, 2006) If water content measurement in a narrow soil layer is required, the probe needs to be installed horizontally (Fares and Polyakov, 2006) The SCS probes are easy to install and use and well suited for use at shallow depths; however, soil temperature fluctuations at such depths are higher and could affect their readings The design and principles of operation of the SCS probe are described in the manufacturer’s calibration manual (Decagon Devices, Inc., Pullman, WA, USA) as well as reported by Fares and Polyakov (2006) Laboratory setup and measurement One liter columns containing two media, pure deionized water and quartz sand (hyperthermic, uncoated Typic Quartzipsamments), were used in this study Starting from oven-dried sand, water was added to obtain water contents of 0.05, 0.10, 0.15, 0.20, 0.25 and 0.30 cm3 cm−3 Mixed wet sand of specified volume was then packed into the columns uniformly up to a bulk density of 1.50 to 1.55 g cm−3 590 A FARES et al with the probe inserted in the middle of each column In this study, we assumed that water content of oven-dried quartz sand was 0.0 cm3 cm−3 Thermocouples were carefully placed in the core zone of influence (3 cm away from sensor) of the proposed sensors To insure uniform bulk density throughout the soil column, the soil was added and compacted in small increments The top of the column was tightly covered with plastic wrap to prevent water losses due to evaporation Packed sand columns with probes in the middle at the specified water contents as described above were placed in an incubator where the temperature varied from to 45 ◦ C The SCS probe and thermocouples were logged with a CR-21X Campbell Scientific data logger (Campbell Scientific Inc., Logon, UT) at minute observation intervals The volumetric soil water content can be calculated using either a soil specific calibration or the set default calibration equation supplied by the manufacturer: θ = 0.000424Vs − 0.29 (1) where θ is the volumetric water content (cm3 cm−3 ) and Vs is the sensor output in mV (Decagon Devices, Inc., 2006) Instead of the default manufacturer’s calibration equation that correlates raw SCS voltage output to actual water content, we used a calibration equation that relates water content to a scaled voltage (SV) defined as follows: SV = (Va − Vs )(Va − Vw )−1 (2) where Va , Vw and Vs are the SCS output in air at room temperature, in water at room temperature, and in soil, respectively The dielectric constant of air does not change with temperature; thus, any sensor output changes as a result of temperature increase will be due to temperature effect on sensor electronics Single capacitance sensor measurements were also taken in deionized water with temperature varying from to 45 ◦ C Correction of temperature effect To correct the temperature effects on SCS readings, the sensor output voltage was plotted against the temperature The slope from liner regression between the sensor output voltage and temperature was used to correct the temperature effect by using the following equation: Vc = Vr + (22 − T )Sv (3) where Vc (mV) is the corrected voltage, Vr (mV) is the sensor reading at temperature T (◦ C), and Sv is the slope of curve between Vr and T Since the slope (Sv ) was different for different media the corrected output voltage was calculated for soil, air, and water (Vcs , Vca , Vcw ) The corrected voltage, Vcs , Vca , and Vcw , is then inserted into the Eq (2) to calculate a corrected scaled voltage SVc , which is written as follows: SVc = (Vca − Vcs )(Vca − Vcw )−1 (4) where Vca , Vcs and Vcw (mV) are the temperature corrected SCS output in air, deionized water, and soil, respectively, at room temperature (22 ◦ C) A linear calibration model was chosen to fit the relationship between water content readings and corresponding scaled voltage Measurements in air were made with sensors suspended in air and subjected to temperatures ranging from to 45 ◦ C Saline water measurement Six SCS probes placed in 1-litre cylinders full of water at six salinity levels (80, 40, 20, 10, and mmol L−1 of CaCl2 ) were used to study the effect of salinity on SCS Each sample was subjected to ADJUSTING TEMPERATURE AND SALINITY EFFECTS 591 heating and cooling cycles to evaluate any hysteresis effect RESULTS AND DISCUSSION Temperature effects in deionized water and air The single capacitance sensor response in the deionized water showed a linear negative correlation with temperature (Fig 1a) The regression coefficient indicated a very good correlation (R2 = 0.99) between SCS output and media temperature The rate of decrease in SCS output with increasing temperature was −0.778 mV ◦ C−1 The negative correlation between temperature and apparent water content was consistent with changes in dielectric constant of free water with temperature, as explained by Weast (1986) Similar temperature effects in water have been also observed by Pepin and Livingston (1995) using a dielectric permittivity technique Recently, Fares et al (2007) reported negative correlation between MCP readings in deionized water and temperature However, McMichael and Lascano (2003) reported a positive correlation with temperature in deionized water using a similar sensor type used in this study Fig Adjusted and raw SCS output in deionized water (a) and air (b) at varying temperature In contrast, we observed a positive correlation between air temperature and SCS readings in air (Fig 1b) In the heating cycle, approximately 99% of the sensor response is attributed to temperature variation The increasing rate in sensor reading due to air temperature increase was 0.587 mV ◦ C−1 This positive correlation concurs with data reported by Fares et al (2007) for MCP, and Campbell (2002) for SCS A portion of the effect of temperature on dielectric-permittivity based water sensor is caused by temperature effects on the instrument electronics (Seyfried and Murdock, 2001) Since the dielectric constant of air did not change with temperature, the increase in sensor reading could be attributed to the temperature effects on sensor electronics Effect of salinity and water temperature Results of the response of SCS to water temperature at the five non-zero salinity levels are shown in Fig The rate of temperature effect increased from to 10 mmol L−1 and then decreased with increase of the solution salt concentration Thus, the temperature effect ranged between 0.557 and 1.150 mV ◦ C−1 for the 80 and 10 mmol L−1 treatments, respectively Almost 80% of the SCS response to solution salinity occurred at concentration below 20 mmol L−1 CaCl2 (Fig 3) The positive correlation between SCS reading and saline water temperature could be explained based on the relationship between permittivity and voltage output for capacitor The dielectric constant of pure and saline water is temperature dependent in low frequency range because the dielectric loss of the orientational polarization results from the thermal effects as explained by Grant et al (1957) The static dielectric constant decreases with increasing temperature in low frequency range (Klein and Swift, 1977), resulting in an increase in output voltage of SCS since capacitance is directly proportional to dielectric constant 592 A FARES et al and inversely proportional to voltage The temperature effect corrected using the Eq (3) is shown in Fig A regression mathematical model was developed to explain the combined effect of salinity and temperature on SCS response The SCS voltage (V ) in the form of polynomial equation is given as follows: V = 3.90359 × 10−5 S + 3.9571 × 10−5 S T + 1.2275 × 10−7 S T + 1.5703 × 10−9 ST − 1.3741 × 10−6 T − 5.0699 × 10−3 S − 5.1866 × 10−3 S T − 1.0935 × 10−5 ST + 1.3758 × 10−4 T + 1.5232 × 10−1 S + 1.7319 × 10−1 ST − 4.6163 × 10−3 T + 1.1496 × S − 2.8813 × 10−1 T + 1812.605 (5) where T is the temperature (◦ C) and S is the salinity (mmol L−1 ) The root mean squared error (RMSE) of the regression was 7.09 mV (n = 54) Fig Single capacitance sensor response to temperature in water at different salinity levels Fig Single capacitance sensor response to salinity in water at 25 ◦ C Fig Fig Adjusted SCS output to temperature in water at different salinity levels Single capacitance sensor response to temperature in non-saline sand at different water content levels Temperature and salinity effect in saline quartz sand A two-parameter linear regression model was used to describe the correlation between media temperature (i.e., quartz sand) and sensor response voltage as follows: Vr = aT + b (6) ADJUSTING TEMPERATURE AND SALINITY EFFECTS 593 where Vr (mV) is the sensor reading, a and b are fitted parameters, and T (◦ C) is temperature Table I shows highly significant correlations between media temperature and sensor response for different water contents at mmol L−1 salinity (the lowest, R2 is 0.957) A positive correlation between sensor reading and temperature was observed (Table I) for all water content levels (0.05, 0.10, 0.15, 0.20, 0.25 and 0.30 cm3 cm−3 ) The temperature effect was particularly pronounced at the low water contents The rate of temperature effect decreased with increased water content; thus, the highest effect (9.38 mV ◦ C−1 ) was observed in the lowest water content treatment (0.05 cm3 cm−3 ) Results also indicated that at high salinity and water content levels SCS was less affected by the temperature of the medium (Table II) A similar saturation effect was also reported by Fares et al (2007) with the MCP The SCS was positively correlated with temperature in the non-saline sand treatment at different water contents The rate of temperature effect increased with increasing water content Thus, the highest effect (5.72 mV ◦ C−1 ) was found for 0.30 cm3 cm−3 water content The temperature effect in non-saline sand at saturation was much higher than corresponding rates at all salinity levels Single capacitance sensor response to temperature in non-saline sand (Fig 5) was similar to that reported by Baumhardt et al (2000) using the MCP and TDR systems and by Fares et al (2007) using MCP In saline sand, the temperature effect decreased with increasing water content, whereas in non-saline sand it was the opposite (Table II) TABLE I Result of the linear regression (Eq where a is the slope and b is the intercept) between temperature and sensor response voltage at mmol L−1 salinity level and different water contents in quartz sand Water content a b SEEa) P R2 cm3 cm−3 0.05 0.10 0.15 0.20 0.25 0.30 9.38 7.94 7.21 4.03 4.94 1.50 854.30 098.61 151.80 432.16 494.00 719.63 7.7503 19.2632 8.8410 9.0738 6.9666 4.6528 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.996 0.973 0.993 0.976 0.990 0.957 a) SEE = standard error of estimate TABLE II Temperature sensitivity of single capacitance sensor (SCS) probe output at different water contents and salinity levels in quartz sand Water content Increase in SCS probe output at different salinity per unit increase in temperature (mmol L −1 ) cm3 cm−3 0.05 0.10 0.15 0.20 0.25 0.30 10 20 40 80 4.73 3.75 4.06 2.23 1.62 2.56 3.07 2.34 2.18 1.57 1.67 1.65 2.18 2.53 1.54 1.14 1.33 0.81 mV ◦ C−1 1.29 1.14 1.94 4.69 4.89 5.72 9.38 7.94 7.21 4.03 4.94 1.50 8.81 5.78 4.87 3.45 3.73 2.03 The sensor response voltage (Vr ) in sand at different water content was corrected for temperature as described by Eq The temperature effect was reduced by 98% on average within the given water content range The temperature corrected voltage (Vcs ) was then used to describe the salinity effect in quartz sand The regression fit to explain the water content (θv ) as function of sensor corrected voltage (Vcs ) and salinity (S) is given by: θv = −3.1993×10−7 S −3.100036×10−7 S Vc −1.9492×10−8 SVc2 +5.0149×10−10 Vc3 +6.115×10−4 S + 1.0734 × 10−4 SVc − 2.05272 × 10−6 Vc2 − 1.343 × 10−1 S + 2.8813 × 10−3 Vc − 1.1142 (7) 594 A FARES et al The RMSE of the regression was 0.017 cm3 cm−3 (n = 36) Soil water calibration The linear two-parameter models used to fit calibration data gave better fits than the nonlinear three parameter power models (data not shown) The former model correlates SV c and volumetric water content (θv ) as follows: θv = c + d × SVc (8) where c and d are fitted coefficients Temperature adjusted calibration equations were developed for each salinity level (Fig 6) These new calibration equations gave better correlation between scaled voltage and water content than isothermal calibration equations at the specified salinity levels Temperature effect was significantly reduced at higher water content However, in lower water content, the new calibration equation over-predicted water content by 0.01 cm3 cm−3 for a 40 ◦ C temperature increase between and 45 ◦ C Whereas, the calibration equation using isothermal scaled voltage for the same temperature range over-prediction was almost ten times higher (0.10 cm3 cm−3 ) Polyakov et al (2005) found different temperature effects on multi capacitance sensor system For SCS probe, the relationship between temperature and sensor response was linear Seyfried and Murdock (2001) reported relatively smaller temperature effects on measurement by reflectometer sensors in different soil types Fig Calibration curves for quartz sand at different salinity levels after temperature adjustment Fig Coefficients of salinity correction for quartz sand at different water content levels Salinity affected the values of the fitted coefficients, c and d, of the calibration equation There was positive relationship (Table III) between the salinity and the absolute value of the coefficients, c and d, of the calibration equation for quartz sand The calibration equation developed for mmol L−1 CaCl2 over-predicted apparent water content by as high as 0.23 cm3 cm−3 if used for 80 mmol L−1 salinity To correct these salinity effects, we developed an empirical relationship between the measured water contents using the above non-saline sand calibration equation and actual water content Salinity corrected TABLE III Parameter for the calibration equation (Eq 8) of single capacitance sensor (SCS) probe at different salinity levels in quartz sand Fitted coefficienta) c d a) c Salinity level (mmol L−1 ) 0b) 10 20 40 80 0.0045 0.5269 −0.1568 0.4613 −0.3980 0.6879 −0.9171 1.1871 −1.3915 1.6071 −1.6110 1.7719 is the intercept and d is the slope in Eq calibration used for salinity effect correction in this study b) Reference ADJUSTING TEMPERATURE AND SALINITY EFFECTS 595 actual water content (θa ) was calculated using the following equation: θa = αS −β θv (9) where S is the salinity (mmol L−1 ), θv is the apparent water content calculated at different salinity using the calibration curve developed for non-saline sand (cm3 cm−3 ), and α and β are two fitting parameters A calibration equation for zero salinity level was developed and used as reference calibration for calculating the apparent water content at different salinity levels Ideally, if the sensor reading is not affected by the salinity, the ratio of (θa ) to (θv ) should be equal to one The SCS probe was affected by salinity; thus, this ratio varied between 0.09 and 1.2 The (θa /θv ) ratio for all water contents was correlated with different salinity levels between and 80 mmol L−1 Fig shows the correlation between water content in one hand and α and β in the other hand using the following models: 0.4121 α = 1.0463θv0 (10) β = −0.9146θv0 + 0.2929 (11) where θv0 is the water content at zero salinity The use of temperature and salinity corrected scaled voltage calibration equations substantially mitigated the effect of temperature and salinity on SCS probe readings (Fig 8) The analytical model for salinity correction was able to minimize the error to as low as ±2% at lower water content level and ±1% near saturation Using the scaled voltage calibration with voltage correction, the SCS probe outputs were accurate over the varied temperature ranged from to 45 ◦ C at all water content levels For almost all water content levels, the use of temperature corrected scaled voltage calibration equations reduced the temperature effect by 98%, which is similar to what Fares et al (2007) reported for MCP using a temperature dependent scaled frequency technique Fig Relationship between actual and apparent water contents after temperature and salinity effect correction CONCLUSIONS Single capacitance sensor measured water content was sensitive to both soil temperature and salinity, which were attributed to related soil permittivity changes Apparent water content increased for saline and non-saline treatments as temperature of the medium increased Temperature effect in non-saline sand increased with increasing water content; however, in the case of saline sand, the response was opposite A ‘saturation’ effect was observed in response to salinity, with almost 90% of the increase in the output voltage occurring for salinity levels below 20 mmol L−1 CaCl2 The temperature correction model was able to mitigate the temperature effect to as low as 98% The empirical model for salinity corrections provided good estimates for the salinity range we tested (5–80 mmol L−1 ) The two-parameter linear model described the relationship between the soil water content and the scaled voltage as measured by the SCS Isothermal scaled voltage with temperature correction was able to mitigate the temperature 596 A FARES et al effect on SCS probe output Ignoring the media temperature and salinity may result in an over prediction of water content up to 0.23 cm3 cm−3 , particularly at lower water content The analytical model for salinity correction was able to mitigate the error to as low as ±2% REFERENCES Baumhardt, R L., Lascano, R J and Evett, S R 2000 Soil material, temperature, and salinity effects on calibration of multisensor capacitance probes Soil Sci Soc Am J 64: 1940–1946 Campbell, C S 2002 Response of ECH2 O Soil Moisture Sensor to Temperature Variation in Water Content, Soil Type and Solution Electrical Conductivity Decagon Devices, Inc 950 NE Nelson Court Pullman, WA 99163 USA Cassel, D K., Kachanoski, R G and Topp, G C 1994 Practical considerations for using a TDR cable tester Soil Technol 7: 113–126 Decagon Devices, Inc 2006 ECH2 O Soil Moisture Sensor Operator’s Manual for Models EC-20, EC-10, and EC-5 Version 3.0 Decagon Devices Inc., Pullman, WA Evett, S R., Tolk, J A and Howell, T A 2006 Soil profile water content determination: Sensor accuracy, axial response, calibration, temperature dependence, and precision Vadose Zone J 5: 894–907 Fares, A and Polyakov, V 2006 Advances in crop water management using capacitive water sensors In Sparks, D (ed.) Advances in Agronomy Vol 90 Elsevier Science, USA pp 43–77 Fares, A., Hamdhani, H and Jenkins, D M 2007 Temperature-dependent scaled frequency: Improved accuracy of multisensor capacitance probes Soil Sci Soc Am J 71: 894–900 Gardner, W H 1986 Water content In Klute, A et al (eds.) Methods of Soil Analysis Part Physical and Mineralogical Methods 2nd Ed ASA and SSSA, Madison, WI pp 493–544 Grant, E., Buchanan, T and Cook, T 1957 Dielectric behavior of water at microwave frequencies J Chem Phys 26: 156–161 Greacen, E L 1981 Soil Water Assessment by the Neutron Method CSIRO Melbourne, Australia Klein, L A and Swift, C T 1977 An improved model for the dielectric constant of sea water at microwave frequencies IEEE T Antenn Propag 25: 104–111 McMichael, B and Lascano, R J 2003 Laboratory evaluation of a commercial dielectric soil water sensor Vadose Zone J 2: 650–654 Mead, R M., Ayars, J E and Liu, J 1995 Evaluating the influence of soil texture, bulk density, and soil water salinity on a capacitance probe calibration ASAE meeting, 18–23 June, ASAE Paper 95-3264 ASAE, St Joseph, MI Noborio, K., McInnes, K J and Heilman, J L 1994 Field measurements of soil electrical conductivity and water content by time-domain reflectometry Comput Electron Agr 11: 131–142 Pepin, S and Livingston, N J 1995 Temperature-dependent measurement errors in time domain reflectometry determinations of soil water Soil Sci Soc Am J 59: 38–43 Polyakov, V., Fares, A and Ryder, H M 2005 Calibration of a capacitance system for measuring water content of tropical soil Soil Sci Soc Am J 65: 914–921 Robinson, M and Dean, T J 1993 Measurement of near surface soil water content using a capacitance probe Hydrol Process 7(1): 77–86 Spaans, E J A and Baker, J M 1992 Calibration of watermark soil moisture sensors for soil matric potential and temperature Plant Soil 143: 213–217 Seyfried, M S 1993 Field calibration and monitoring of soil-water content with fiberglass electrical resistance sensors Soil Sci Soc Am J 57: 1432–1436 Seyfried, M S and Murdock, M D 2001 Response of new soil water sensor to variable soil, water content, and temperature Soil Sci Soc Am J 65: 28–34 Topp, G C., Davis, J L and Annan, A P 1980 Electromagnetic determination of soil-water content measurements in coaxial transmission lines Water Resour Res 16: 574–582 Weast, R C (ed.) 1986 Handbook of Chemistry and Physics 61st Edition CRC Press, Boca Raton, FL White, I., Knight, J H., Zegelin, S J and Topp, G C 1994 Comments on ‘Considerations on the use of Time-Domain Reflectometry (TDR) for measuring soilWater content’ by W R Whalley Eur J Soil Sci 45: 503–508 ... six salinity levels (80, 40, 20, 10, and mmol L−1 of CaCl2 ) were used to study the effect of salinity on SCS Each sample was subjected to ADJUSTING TEMPERATURE AND SALINITY EFFECTS 591 heating and. .. between media temperature (i.e., quartz sand) and sensor response voltage as follows: Vr = aT + b (6) ADJUSTING TEMPERATURE AND SALINITY EFFECTS 593 where Vr (mV) is the sensor reading, a and b are... Single capacitance sensor response to temperature in non-saline sand at different water content levels Temperature and salinity effect in saline quartz sand A two-parameter linear regression model