Soil Science Society of America Journal 67:1071-1078 (2003)
© 2003 Soil Science Society of America
DIVISION S-1—SOIL PHYSICS
Soil Solution Electrical Conductivity Measurements Using Different Dielectric Techniques
Yasser Hamed,
Magnus Persson* and
Ronny Berndtsson
Dep. of Water Resources Engineering, Lund Univ., Box 118, 221 00 Lund, Sweden
* Corresponding author (magnus.persson{at}tvrl.lth.se)
 |
ABSTRACT
|
|---|
Accurate measurements of soil solution electrical conductivity (
w) are needed in various applications. One recently developed technique that measures w is the Sigma Probe (SP). The SP is supposed to give accurate readings only slightly dependent on water content (
) and soil type. To test the performance of the SP, it was compared with another dielectric technique, time domain reflectometry (TDR). Both techniques utilize the dielectric constant (Ka) and bulk electrical conductivity (a) to estimate the w. Measurements of w were obtained in a laboratory experiment using nine different soil types with in the range 0.05 to 0.50 m3 m-3. In each soil type, three different w were used (approximately 0.3, 1.2, and 3.0 dS m-1). The linear w–a– model used by the SP contains only one soil specific parameter (K0). Using this model, the SP readings were constant over the encountered range in , whereas the TDR estimation calculated by the same model typically increased at Ka values below the range of 10 to 15. Using the SP with a default K0 value of 4.1 typically gave a w that was ±20% of the true w when w > 1 dS m-1. The error in the w estimation for w lower than 1 dS m-1 can be much larger except in sandy soils. The TDR measurements of w using a conventional w–a– model were more accurate in all soil types at all , with root mean square errors that were lower by about 50% compared with the SP readings. However, this model requires soil specific parameters that have to be obtained during a calibration experiment.
Abbreviations: SP, sigma probe • TDR, time domain reflectometry
|
INTRODUCTION
|
|---|
ACCURATE MEASUREMENTS of water content and solute concentrations are essential in soil science, agriculture, and hydrology. Some examples are: to optimize irrigation and fertilizer application, to study surface runoff and erosion, and to control and evaluate the hazard of soil salinity. The term salinity refers to the presence of the major dissolved inorganic solutes, typically Na+, Mg+, Ca2+, K+, Cl-, SO2-4, HCO-3, and CO2-3, in aqueous samples (Rhoades, 1996). Salinity is quantified in terms of the total concentration of such soluble salts.
The common method for water content and salinity measurement is soil sampling. The water content can be determined by oven drying and the soil extract method is used for salinity measurements. In this method, aqueous extracts of the soil samples have traditionally been made in the laboratory at higher than normal water contents for routine soil salinity diagnosis and characterization purposes. Since the absolute and relative amounts of the various solutes are influenced by the water/soil ratio at which the extract is made (Reitemeier, 1946), the water/soil ratio used to obtain the extract should be standardized to obtain results that can be generally applied and interpreted. Soil salinity is most generally defined and measured on aqueous extracts of saturated soil-pastes (U.S. Salinity Laboratory Staff, 1954). The water content of saturated soil pastes (saturation percentage), as well as the water/soil ratio, varies with soil texture. It is related in a reasonably general and predictable way to soil water contents under field conditions. Crop tolerance to salinity is also generally expressed in terms of electrical conductivity of the saturation extract (Maas and Hoffman, 1977; Maas, 1986, 1990). This method is, however, not practical when many samples are needed. Furthermore, the method is destructive, that is, only a single measurement can be made for a sample soil volume. Another disadvantage is that the estimates made of the w (soil water electrical conductivity) from a (bulk soil electrical conductivity) and the ratio of their water contents will usually be excessively high. This is because salts will often be present in the saturation extract that would not be the case under actual field conditions (Rhoades et al., 1999).
Thus, there is a great demand for new and more accurate methods of measuring water content and salinity. Time domain reflectometry is an electromagnetic technique that has been shown to take accurate water content () and a measurements under both laboratory and field conditions (Topp et al., 1980; Dalton et al., 1984; Ward et al., 1994). The TDR technique utilizes the apparent dielectric constant (Ka) for calculating . To calculate w, a soil specific w–a– relationship has to be determined. Many different w–a– models have been presented (e.g., Rhoades et al., 1976). The most appealing feature of the TDR technique is its ability to take and a in the same soil volume simultaneously. There are several advantages associated with the TDR technique: it is an accurate instrument and it can easily be automated to take scheduled readings. Some important disadvantages are: water content measurements cannot be made in highly saline soils and the initial costs are relatively high compared with other methods.
Recently, Hilhorst (2000) presented a linear w–a–Ka model. He claimed that this model was only slightly dependent on the soil type. Based on his findings a new dielectric sensor was developed. This sensor is now commercially available and it is called the Sigma Probe (SP). The potential advantages of the SP are: it is easy to use and handle because of its small volume and weight and the method is only slightly dependent on soil type. Furthermore, the SP manual states: ‘The SigmaProbe measures pore water conductivity with a high degree of independence from both soil moisture content and degree of contact between the probe electrodes and the soil’. Persson (2002)and presented an evaluation of the linear w–a–Ka model using TDR data from three sandy soils. He showed that using TDR, there was a larger dependency of the linear model on soil type than predicted by Hilhorst (2000).
The objectives of this study are to further evaluate the linear w–a–Ka model in several different soil types using TDR data and to compare the w measurements of TDR and SP. Furthermore, the SP performance is evaluated in several different soil types with varying and w. For this purpose, a series of laboratory experiments was conducted in nine different soil types ranging from loamy sand to heavy clay using three different w levels over a wide range of .
|
THEORY
|
|---|
Water Content and Bulk Electrical Conductivity Measurements with TDR
The dielectric properties of a material can be described by the complex dielectric constant K*. The dielectric constant of a material consists of a real part K', and an imaginary part K'', or the electric loss. The use of TDR to measure the complex dielectric constant of liquids has been presented by Fellner-Feldegg (1969). The TDR instrument sends a high frequency (20 kHz–1.5 GHz) electromagnetic signal through a probe buried in the soil. The signal is reflected at the end of the probe and the traveling time (t) of the signal can be measured. Topp et al. (1980) introduced the apparent dielectric constant Ka, which can be estimated by
 | [1] |
where c is the velocity of an electromagnetic wave in free space and L is the length of the TDR probe. The electrical loss in soils is normally small but it does affect the estimate of K*. Therefore, the measured dielectric constant is called the apparent dielectric constant, Ka (Topp et al., 1980). In the frequency range transmitted from the TDR instrument, the K' of most soils is almost independent of frequency. The dielectric constant is about 80 for water (at 20°C), 2 to 5 for dry soil, and 1 for air. Thus, Ka is highly dependent on .
Dalton et al. (1984) showed that the bulk electrical conductivity could be calculated from the attenuation of the TDR trace. The bulk electrical conductivity, a, of moist soil is influenced by the following parameters; the electrical conductivity of the pore water (w), the water content (), the tortuosity of the electrical flow paths depending on the geometry of the soil matrix, and the series coupled surface conductivity of the soil particles. Several different models have been developed for the w–a– relationship, for example, Rhoades et al. (1976), Mualem and Friedman (1991), and Heimovaara et al. (1995). All these models contain one or several soil specific parameters. A comparison of different w–a– models can be found in Persson (1997) and Amente et al. (2000).
The linear w–a–Ka relationship and the Sigma Probe
Malicki et al. (1994) and Malicki and Walczak (1999) found that when w was held constant the relationship between Ka and a was linear when Ka > 6. An empirical w–a–Ka model was also presented;
 | [2] |
where S is the sand content in percentage by weight. Inspired by this work, Hilhorst (2000) recently presented a theoretically based linear w–a–Ka relationship:
 | [3] |
where Kw is the dielectric constant of the pore water and K0 is the Ka value when a = 0 (see Hilhorst [2000] for details). The parameter K0 is not, however, the Ka value of dry soil, but it appears as an offset of the linear relationship between a and Ka. Hilhorst (2000) found that the parameter K0 was dependent on soil type and that it was in the range of 1.9 to 7.6. This value has to be determined experimentally for each soil type; however, a value of 4.1 should fit most soils. Note that in Hilhorst (2000), Ka, Kw, and K0 represent the real part of the dielectric constant only. However, Heimovaara (1994) concluded that the TDR measured Ka represents the real part of the complex dielectric constant at the highest effective frequency of the TDR set-up. Thus, Eq. [3] should be applicable to TDR measurements. Hilhorst (2000) found that Eq. [3] was valid in most soils when was higher than about 0.10 m3 m-3. Below this limit, the a–Ka curve should theoretically deviate from Eq. [3], however, this was not verified experimentally in the study by Hilhorst (2000).
Recently, a new dielectric sensor based on the findings of Hilhorst (2000) has been presented. The sensor consists of a 0.105-m rod, 0.005 m in diam. At the end of the rod there are two electrodes (0.015 m long) separated by an isolating material. The rod is connected to a handle with built in electronics. The measuring frequency of the sensor is 30 MHz. The sensor is commercially available and it is called the Sigma probe, SP (type EC1, Delta-T devices Ltd., Cambridge, UK). The SP is connected to a hand held data logger (Psion Workabout) and gives readings of temperature and w (calculated by Eq. [3]).
|
MATERIALS AND METHODS
|
|---|
Soil Sampling
Soils from six different locations were used in this study. Four of these locations are situated in southern Sweden. At three sites, both the topsoil (0- to 0.1-m depth) and the subsoil (0.4- to 0.5-m depth) were collected, while at the fourth site only the topsoil was collected. Two samples were also collected in the catchment of M'Richet el Anze, located 110 km southwest of Tunis, Tunisia. These samples were collected from the topsoil at the hollow and at the slope of the catchment. Further information about the soils in this catchment can be found in Öhrström et al. (2002). Thus, a total of nine different soil samples were used. A description of some selected soil properties is found in Table 1.
Laboratory Experiments
The soil samples were oven dried and passed through a 2-mm sieve. Then, the soil was packed into Plexiglas soil columns, 0.076 m in diameter and 0.1 m long (Soil Measurement System, Tucson, AZ), to the bulk densities encountered in the field. One three-rod TDR probe with a length of 0.08 m and a wire spacing of 0.03 m (model 6111, Soilmoisture Equipment Corp., Santa Barbara, CA) was inserted in the center of the column. Time domain reflectometry measurements were performed using a Tektronix 1502C cable tester with an RS232 interface connected to a laptop computer (Tektronix, Beaverton, OR). Estimates of Ka and a were calculated from the TDR trace using the WinTDR99 software (developed by the Soil Physics Group at Utah State University).
First, the Ka– relationship was determined by performing an upward infiltration experiment (Young et al., 1997). Previous experience with this method shows that the sharp boundary between oven dry and almost saturated soil that exists during the experiment will make the end reflection of the TDR trace difficult to identify, especially at the beginning of the experiment when this boundary is situated at the end of the probe. Furthermore, oven dry soil might be water repellent, which will lead to an unstable wetting front. To avoid these difficulties, after measurements were taken in the oven dry soil, the samples were removed from the column and mixed with 9 mL of water and then repacked into the columns. To prevent clay particles from adhering to one another, the soils with high clay content (i.e., the Värpinge and Tunisian soils) were not oven dried. Instead, they were air dried before packing them into the soil columns. A small subsample of the air-dry soil was oven dried to calculate the initial water content. The upward infiltration experiment was performed by pumping water with a syringe pump from the bottom of the column at a flow of 0.015 m h-1. Distilled water with KBr added to increase the w to 0.30 dS m-1 was used. Water with low w was used because several studies have shown that high a might affect the Ka– relationship (e.g., Dalton, 1992; Sun et al., 2000; Persson et al., 2000). After application of 4 mL, three TDR measurements were taken and averaged. The upward infiltration was continued until the sample was saturated, resulting in approximately 50 pairs of Ka and for each soil sample.
After the Ka– relationship had been established, the w–a–Ka relationship was examined. For this, it is important that the w is known and constant in the soil sample. Thus it was necessary to leach the soil with water of constant w. The upward infiltration was continued until the w of the effluent and the TDR measured a reached a constant value using the same water as in the Ka– calibration experiment. Approximately five pore volumes of water were added since this proved to be more than sufficient for the soils used in this study. The flow rate was 0.008 m h-1 during daytime and 0.002 m h-1 during the night. After leaching, the SP was inserted into the column so that the tip of the SP was at the same level as the midpoint of the TDR probe at 0.04-m depth. Then, suction was applied at the bottom of the column to drain water. The drainage experiments were always started after a >15-h period of low flow, since the difference between mobile and immobile water w is likely to be small during low flow. Suction was increased stepwise from 0 to 70 kPa (system limit) with an increment of 10 kPa. The TDR measurements were taken automatically every half-minute at the beginning of the experiment; the measurement frequency was then decreased gradually to once every 30 to 60 min as decreased more slowly when the drainage experiment progressed. The drainage experiment was continued for 24 to 48 h, at this time there was no change in the TDR measured Ka and a. The SP measurements of w were taken manually during the drainage experiment, once each minute in the beginning and more seldom at the end of the experiment; each time three readings were taken and averaged. The w of the drained water was measured several times during the experiment.
The leaching/drainage experiment was repeated twice using water with w of about 1.2 and 3.0 dS m-1. These w levels should cover the range of interest in most non-saline agricultural soils. First, the samples were leached using upward infiltration. Again, about five pore volumes proved to be sufficient for reaching a constant a. Then, suction was applied as described above. Thus, for all soil samples the Ka– relationship and the Ka–a for three different w were determined. Furthermore, SP readings were taken at a range of Ka (or ) for three different w. All experiments were performed at a constant temperature of 20°C, thus the data were not temperature corrected.
|
RESULTS AND DISCUSSION
|
|---|
Water Content Calibration
An example of an upward infiltration experiment is given in Fig. 1 (Revinge subsoil). The Ka– data were used to estimate the parameters of a third-order polynomial equation
 | [4] |
where a, b, c, and d are empirical parameters. Several studies have shown that this approach gives the most accurate estimations compared with other commonly used Ka– models (Jacobsen and Schjønning, 1995; Young et al., 1997; Persson et al., 2001). The parameters of Eq. [4] for the different soil types are given in Table 2.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1. The dielectric constant (Ka) measured during the upward infiltration experiment in the Revinge subsoil plotted against water content (). The solid line represents Eq. [4] with parameters according to Table 2.
|
|
The TDR Measured w–a– Relationship in Different Soil Types
In most leaching/drainage experiments the w of the inflowing water was slightly different from the effluent above the soil surface after the TDR measured a reached a constant level. However, the effluent and the extracted water had similar w. Therefore, it was assumed that the w was constant within the samples and equal to the effluent and extracted water. Thus, this value was used in the analysis. All soil samples showed a strong linear relationship between Ka and a for any given, constant w. The only exception was the Värpinge topsoil using the lowest w, for which the Ka–a relationship was strongly nonlinear. Thus, these data were not included in the further analysis. One example of the Ka–a data during the drainage experiments is given in Fig. 2. By rearranging Eq. [2] and [3], the slope and intercept of the Ka–a relationship can be calculated. In Table 3, the result of a linear regression analysis for each experiment is given along with the predicted slope and intercept of the Hilhorst (2000) and Malicki and Walczak (1999) models. In the Malicki and Walczak (1999) model, both the slope and intercept depend on both soil type (sand content) and w. In the Hilhorst (2000) model, the slope depends only on w, whereas the intercept is a soil specific constant and should be in the range of 2 to 9. The calculated slopes for both models are similar and in many cases they agree well with the ones calculated by the TDR measurements. The slopes for the Tunisian soils, however, were lower than the Hilhorst (2000) model and much lower than the Malicki and Walczak (1999) model. This is probably because the soils in the study of Malicki and Walczak (1999) all had a clay content lower than 20%. The intercept of our data is fairly close to the ones predicted by Malicki and Walczak (1999) and followed the predicted behavior, that is, the intercept increased with increasing w. The only exception was the Odarslöv soil, which showed a fairly constant intercept as w increased.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2. Measured bulk electrical conductivity (a) plotted against the dielectric constant (Ka) measured during one drainage experiment in the Odarslöv topsoil with a soil solution electrical conductivity of 2.90 dS m-1. The solid line is the regression line with parameters according to Table 3.
|
|
Even though the Ka–a relationship exhibited a linear behavior, none of the linear models gave satisfactory w predictions. Therefore, the Rhoades et al. (1976) model was also fitted to the TDR data.
 | [5] |
where e and f are empirical parameters and s is the surface conductivity of the soil matrix. This model was used since several studies have shown that it is able to predict w in different soil types. The best-fit parameter values are presented in Table 4. The fit was good for all soil types. As seen from the table, the parameter values obtained using one w can be very different for another w. Therefore, it is important that the parameters are calculated using different w that cover the range of interest. In Table 4, the root mean square error (RMSE) for the w estimation using all w for each soil type is also presented. The reason for the relative high RMSE was that Eq. [5] typically gave erroneous w estimations when was low (in most cases the w was underestimated at low ), especially at the lowest w level. This can be seen in Fig. 3, where the TDR measured a and the calculated a using Eq. [5] along with the TDR measured and known w for the Värpinge subsoil are plotted. The w estimation from the TDR measurements can also be made using another more sophisticated w–a– model, however, only the Rhoades et al. (1976) model was used because of its simplicity and the fact that it has been widely used for TDR measurements.
Comparison of SP and TDR Measured w
The TDR measured a and Ka during the drainage experiments were converted to w using the model by Hilhorst (2000) with K0 = 4.1. The SP also uses this model. The soil parameter (K0) was kept at the default value of 4.1. When the is too low, the SP will not give any readings. In the Löddeköpinge soil SP readings were not possible when Ka was smaller than 7. In the Revinge soil, however, SP readings were possible even though Ka was as low as 6. In all other soils the Ka was not lower than about 10.
In Fig. 4, an example of the w estimated using TDR and SP during the drainage experiment in the Revinge topsoil is plotted. The Revinge and Löddeköpinge soils all exhibited the same behavior. That is, the TDR estimated w was slightly higher than the SP estimated w at saturated conditions. When Ka was in the range of 10 to 15, the TDR estimated w increased significantly while the SP readings were more or less constant throughout the entire Ka range. In the Odarslöv soil, both the TDR and SP estimations were fairly equal and constant through the entire Ka range. In the Värpinge soil the differences between the TDR and SP readings were larger and the difference increased as the Ka decreased, but still the SP readings were fairly constant throughout the experiment. In the Tunisian soils the TDR estimations were always much higher than the SP readings, but also here the SP readings were constant. A summary of the results of the TDR and SP estimated w are presented in Table 5. In this Table, the TDR estimations are calculated at saturated conditions and the SP estimations are the average of all readings except the ones that deviated from the others at the lowest Ka (this only occurred in some soils). It can be seen that the SP gives fairly good results in the sandy soils (Revinge and Löddeköpinge) but as the clay content increases, the accuracy decreases, especially at low w. In average, the SP measured w was within 20% of the correct value for all soil types when the w > 1 dS m-1. The presented root mean square errors (RMSE) are slightly higher than those presented by Hilhorst (2000), 0.24 dS m-1. He used the SP in seven different soil types with w ranging from 0.8 to 3.1 dS m-1. In his study, however, the K0 parameter was calibrated for each soil type individually to obtain the best fit.
The differences between the TDR and SP readings can probably be explained by the different sampling volume and the different measuring frequencies used by the instruments. The TDR uses a wide band (20 kHz–1.5 GHz) signal while the SP only measures at a single frequency, 30 MHz. This will mostly affect the Ka reading. The a readings of the two techniques were compared in salt solutions and were found to be almost identical (data not shown).
The accuracy of the SP readings could, of course, be improved by changing K0. In the SP manual it is suggested that the soil specific K0 can be determined in a saturated soil sample using another dielectric sensor to measure Ka and a. This was also done in this study using the TDR measurements. The calculated K0 were close to the intercepts presented in Table 3. However, the SP software only allows K0 in the range 2 to 9. Our calculated K0 were often negative and they increased with w. Therefore only the default K0 value of 4.1 was used in the drainage experiments.
|
SUMMARY AND CONCLUSIONS
|
|---|
In the present study, the linear w–a– model was evaluated and the SP and TDR techniques were compared. Measurements were taken in nine different soil samples including loamy sand, sandy loam, loam, silty clay, and clay soils. In each soil sample, three different w were used. The results showed that the Hilhorst (2000) model did not describe the TDR data well. The slope of the Ka–a relationship was reasonably well predicted by the Hilhorst model but the intercept found in the experiments was different from the study by Hilhorst (2000). In the Hilhorst (2000) model, the slope is only dependent on w and the intercept is only dependent on the soil type. However, this study showed that both the slope and intercept depend on both soil type and w. Similar results have previously been presented by Malicki and Walczak (1999) and Persson (2002). Instead, the model by Rhoades et al. (1976) was used for calculating w from the TDR readings. This model gave good w predictions for all soil types. The SP measured w was constant over the entire range of encountered in this study. However, the accuracy of the measurements was not as high as the TDR measurements. The results could be improved using a soil specific K0 value. However, the SP software does not allow for negative K0 values, which were typically encountered at low w. Furthermore, the K0 value depends on w, thus using a single K0 value will only give accurate w readings when changes in w are small. Using the default K0 value of 4.1 will give a w that will be ±20% of the true w when w > 1 dS m-1 in most examined soil types. The error in the w estimation for w lower than 1 dS m-1 can be much larger, except in sandy soils. We believe that SP users would appreciate if the SP software would be modified so that negative values of K0 can be used. Furthermore, not only w but also a and Ka should be displayed, then both w and measurements would be possible also with the SP.
In conclusion, though the SP is relatively inexpensive and easy to handle and it can be used for different soil types without calibration, it is not very accurate. The TDR technique though it is relatively expensive and more difficult to handle, can accurately measure w, if a soil specific calibration experiment is conducted.
|
ACKNOWLEDGMENTS
|
|---|
This study was funded through the Swedish Research Council for Engineering Sciences and the Swedish Natural Science Research Council. Experimental equipment was purchased through a grant from the Royal Physiographic Society in Lund. The first author wishes to thank the Egyptian government for financing his stay at Lund University.
Received for publication November 19, 2001.
|
REFERENCES
|
|---|
- Amente, G., J.M. Baker, and C.F Reece. 2000. Estimation of soil solution electrical conductivity from bulk soil electrical conductivity in sandy soils. Soil Sci. Soc. Am. J. 64:1931–1939.[Abstract/Free Full Text]
- Dalton, F.N. 1992. Development of time-domain reflectometry for measuring soil water content and bulk soil electrical conductivity. p. 143–167. In G.C. Topp et al. (ed.) Advances in measurement of soil physical properties: Bringing theory into practice. SSSA Spec. Pub. No. 30. SSSA, Madison, WI.
- Dalton, F.N., W.N. Herkelrath, D.S. Rawlins, and J.D. Rhoades. 1984. Time-domain reflectometry: Simultaneous measurements of soil water content and electrical conductivity with a single probe. Science 224:989–990.[Abstract/Free Full Text]
- Fellner-Feldegg, J. 1969. The measurement of dielectrics in the time domain. J. Phys. Chem. 73:616–623.
- Heimovaara, T.J. 1994. Frequency domain analysis of time domain reflectometry waveforms 1, measurement of the complex dielectric permittivity of soils. Water Resour. Res. 30:189–199.
- Heimovaara, T.J., A.G. Focke, W. Bouten, and J.M. Verstraten. 1995. Assessing temporal variations in soil water composition with time domain reflectometry. Soil Sci. Soc. Am. J. 59:689–698.[Abstract/Free Full Text]
- Hilhorst, M.A. 2000. A pore water conductivity sensor. Soil Sci. Soc. Am. J. 64:1922–1925.[Abstract/Free Full Text]
- Jacobsen, O.H., and P. Schjønning. 1995. Comparison of TDR calibration functions for soil water determination. p. 25–33. In L.W. Petersen and O.H. Jacobsen (ed.) Proceedings of the symposium:Time-domain reflectometry applications in soil science. Research Center Foulum, Denmark, 16 Sep. 1994. SP report no. 11. vol. 3 Danish Institute of Plant and Soil Sci., Lyngby, Denmark.
- Maas, E.V. 1986. Salt tolerance of plants. Applied Agric. Res. 1:12–26.
- Maas, E.V. 1990. Chapter 13. Crop salt tolerance. p. 262–304. In K.K. Tanji (ed.) Agri. Salinity Assessment and Management. ASCE Manuals & Reports on Engineering No. 71. ASCE, New York.
- Maas, E.V., and G.J. Hoffman. 1977. Crop salt tolerance-current assessment. J. Irrig. Drain. Div. Am. Soc. Civ. Eng. 103(IR2):115–134.
- Malicki, M.A., R.T. Walczak, S. Koch, and H. Flühler. 1994. Determining soil salinity from simultaneous readings of its electrical conductivity and permittivity using TDR. Proceedings of the Symposium on Time Domain Reflectometry in Environmental, Infrastructure, and Mining Applications, Evanston, Illinois, 7–9 September, U.S. Bureau of Mines, Spec. Pub. SP 19–94, NTIS PB95–105789, 328–336. U.S. Bureau of Mines, Washington, DC.
- Malicki, M.A., and R.T. Walczak. 1999. Evaluating soil salinity status from bulk electrical conductivity and permittivity. Eur. J. Soil Sci. 50:505–514.
- Mualem, Y., and S.P. Friedman. 1991. Theoretical prediction of electrical conductivity in saturated and unsaturated soil. Water Resour. Res. 27:2771–2777.
- Öhrström, P., M. Persson, R. Berndtsson, S. Nasri, P. Zante, and J. Albergel. 2002. Field scale variation of preferential flow indicated from dye coverage. J. Hydrol. (Amsterdam) 257:164–173.
- Persson, M. 1997. Soil solution electrical conductivity measurements under transient conditions using TDR. Soil Sci. Soc. Am. J. 61:997–1003.[Abstract/Free Full Text]
- Persson, M. 2002. Evaluating the linear dielectric constant-electrical conductivity model using time domain reflectometry. Hydrol. Sci. J. 47:269–278.
- Persson, M., R. Berndtsson, S. Nasri, J. Albergel, P. Zante, and Y. Yumegaki. 2000. Solute transport and water content measurements in clay soils using time domain reflectometry. Hydrol. Sci. J. 45:833–847.
- Persson, M., R. Berndtsson, and B. Sivakumar. 2001. Using Neural networks for calibration of time domain reflectometry measurements. Hydrol. Sci. J. 46:389–398.
- Reitemeier, R.F. 1946. Effect of moisture content on the dissolved and exchangeable ions of soils of arid regions. Soil Sci. 61:195–214.
- Rhoades, J.D., P.A.C. Ratts, and R.J. Prather. 1976. Effects of liquid-phase electrical conductivity, water content, and surface conductivity on bulk soil electrical conductivity. Soil Sci. Soc. Am. J. 40:651–655.[Abstract/Free Full Text]
- Rhoades, J.D. 1996. Salinity: Electrical conductivity and total dissolved solids. p. 417–435. In Methods of Soil Analysis. Part 3. SSSA Book Series no. 5. SSSA, Madison, WI.
- Rhoades, J.D., F. Chanduvi, and S. Lesch. 1999. Methods and interpretation of electrical conductivity measurements. In FAO Irrigation and Drainage, Paper 57. Food and Agriculture Organization of the United Nations, Rome.
- Sun, Z.J., G.D. Young, R.A. McFarlane, and B.M. Chambers. 2000. The effect of soil electrical conductivity on moisture determination using time-domain reflectometry in sandy soil. Can. J. Soil Sci. 80:13–22.
- Topp, G.C., J.L. Davis, and A.P. Annan. 1980. Electromagnetic determination of soil water content: Measurements in Coaxial Transmission lines. Water Resour. Res. 16:574–582.
- U.S. Salinity Laboratory Staff. 1954. Diagnosis and improvement of saline and alkali soils. USDA Handb. 60. U.S. Gov. Print. Office, Washington, DC.
- Ward, A.L., R.G. Kachanoski, and D.E. Elrick. 1994. Laboratory Measurements of solute tranSPort using time domain reflectometry. Soil Sci. Soc. Am. J. 58:1031–1039.[Abstract/Free Full Text]
- Young, M.H., J.B. Fleming, P.J. Wierenga, and A.W. Warrick. 1997. Rapid laboratory calibration of time domain reflectometry using upward infiltration. Soil Sci. Soc. Am. J. 61:707–712.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
C. M. Regalado, A. Ritter, and R. M. Rodriguez-Gonzalez
Performance of the Commercial WET Capacitance Sensor as Compared with Time Domain Reflectometry in Volcanic Soils
Vadose Zone J.,
April 9, 2007;
6(2):
244 - 254.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Persson, D. Bendz, and P. Flyhammar
Time-Domain Reflectometry Probe for Water Content and Electrical Conductivity Measurements in Saline Porous Media
Vadose Zone J.,
November 1, 2004;
3(4):
1146 - 1151.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
