Read more about VES, soil electrical properties, geophysics in forensic and archaelogy, GIS, fractals, and geostatistics at my NEW redesigned website: Landviser, Inc.

A Vertical Electrical Sounding Method for Soil Survey

Electrical methods are intensively used by geophysicists for evaluation of deep subsurface. Measurements of the electrical conductivity or resistivity have been applied for soil salinity surveys in situ for many years (Rhoades and Ingvalson, 1971; Austin and Rhoades, 1979; Rhoades et al., 1990). The most common method is the electrical profiling using four-electrode probes in Wenner configuration. The probes are applied on the soil surface as well as in bore-hole logging (Rhoades and Schilfgaarde, 1976; Halvorson and Rhoades, 1976; Rhoades, 1979). Recently other electrical geophysical methods such as electromagnetic induction (EM) and ground penetrating radar (GPR) become increasingly popular. The methods are still applied preferentially on saline irrigated areas. Some successful applications of the methods were reported on accessing quality of forest soils (McBride et al., 1990), mapping water flow paths (Freeland et al., 1997a), finding perched water locations (Freeland et al., 1997b), and outlining permafrost layers (Arcone et al., 1998). Despite the promising applications, methods of four-electrode profiling, EM, and GPR have some drawbacks when used for shallow soil profiles. Methods of EM and four-electrode probe can not directly measure different resistivities or conductivities of soil horizons and provide only average or bulk electrical conductivity of the soil profile (Corwin and Rhoades, 1984). GPR evaluates profile differentiation in soil electrical conductivity, but its application is limited on soils with high conductivity (salty soils, clay soils). GPR is also not easily modified for shallow subsurface measurements (Liner and Liner, 1997).

Although the method of vertical electrical sounding (VES) is very popular in conventional geophysical studies, such as gas, oil, and coal exploration (Verma and Bandyopadnyay, 1983), it is rarely used in shallow subsurface studies. Vertical electrical sounding was applied to estimate hydraulic conductivity (Mazac et al., 1990) and texture (Banton et al., 1997) of the stratified soils and sediments. Barker (1990) applied VES to a landfill outlining at a 40-m depth. However, the arrays used in these studies can not accurately evaluate very thin (3-30 cm) soil layers. No research has been conducted to evaluate possible applications of the VES method in soil survey for estimating depths and thicknesses of soil horizons. The objectives of our study are to modify the conventional VES method for adequate evaluation of soil horizons and to test the method for soil survey and other possible soil applications.


The vertical electrical sounding and electrical profiling methods are based on the four-electrode principle as shown in Fig. 1. The electrical current (I) is applied to A and B electrodes and the potential (D U) is measured between M and N electrodes. The bulk soil electrical resistivity (ER) is calculated with


where K is the geometric factor.

Fig. 1. Scheme of the vertical electrical sounding (VES) device: (1) auto-canceller, (2) commutator for electrodes AB and MN, (3) netted wires for different distances among electrodes AB and MN, and (4) electrodes.

Some uncertainties exist in the soil literature about the calculation of K and estimation of the measuring depth with different arrays (Kirkham and Taylor, 1949; Banton et al., 1997). As implied in conventional geophysics, the depth of penetration of electrical field in the media is influenced by the array geometry as well as electrical conductivity and layer organization of the media (Beck, 1981; Barker, 1989; Parasnis, 1997). Therefore, the depth of penetration can not be precisely derived from the distances between the electrodes in an array. Theoretical derivations and practical tests have shown that the approximate penetration depth can be considered as 1/6 of [AB] for the arrays of Schlumberger and Wenner types used on wide range of soils and grounds (Barker, 1989; Pozdnyakov et al., 1996; Pozdnyakova et al., 1996; Banton et al., 1997). However, a depth approximation coefficient has been misused (1/3 of [AB]) for the four-electrode profiling with Wenner array (Halvorson and Rhoades, 1976; Rhoades and Schilfgaarde, 1976).

While the depth of penetration for an array varies for the different soils around 1/6 of [AB], the geometric factor (K) can be precisely derived from the array geometry based on the law of electrical field distribution. Using the Laplace equation in polar coordinates, Keller and Frischknecht (1966) derived the electrical potential functions around the source (A and B) and measuring (M and N) electrodes. The geometric factor K can be obtained for four-electrode array of AMNB configuration as


where [AM], [BM], [AN], and [BN] are the distances (m) between the respective electrodes. For central-symmetric array, when [AM]=[BN] and [BM]=[AN], Eq. [2] can be simplified to


The VES array consists of a series of the electrode combinations AMNB with gradually increasing distances among the electrodes for consequent combinations (Fig. 1). The depth of sounding increases with the distance between A and B electrodes. The K factors for the combinations are calculated with Eq. [3] and used to obtain electrical resistivity from measured electrical potential and current using Eq. [1]. The result of VES measurements with central-symmetric arrays is apparent (bulk) electrical resistivity as a function of half of the distance between the current electrodes, i.e. ER=f(AB/2) (Beck, 1981). The relationship between ER and AB/2 can be converted into a relationship between electrical resistivity and actual soil depth through a computer interpretation. Pozdnyakov et al. (1996) developed programs for soil VES interpretations based on an updated R-function. The electrical resistivity measured with the method is shown to be related with salinity, texture and structure, porosity, bulk density, saturation, and hydrological conductivity of the soil (Pozdnyakov et al., 1996; Banton et al., 1997). Thus, the VES profiles can provide information on the geological structures, soil properties, and hydrological conditions in a study area.

We modified the classical geophysical VES array to obtain detailed characteristics of relatively shallow subsurface. Two array configurations are adapted for soil studies. In the first array the [AB/2] distances are fixed as 0.1, 0.15, 0.22, 0.3, 0.45, 0.6, 0.9, 1.2, 1.8, 2.0, 3.6, 4.0, 7.2, 10, and 15 m to ensure a thorough measurement of soil subsurface from 0.02 to 5 m (Pozdnyakov and Chan, 1976). In the second array we increased the [AB/2] distances in a geometrical progression with a coefficient , which results the sounding data distribute with an equal increment in logarithm coordinates. The distances for the second array are set up as 0.1, 0.13, 0.16, 0.2, 0.25, 0.32, 0.40, 0.5, 0.63, 0.8, 1, 1.3 m, etc. The concurrent MN electrodes are placed symmetrically within the center of [AB] for the both arrays (Fig. 1). Resistivity is measured by different combinations of A, B, M, and N electrodes with an automatic switch between the combinations. Since the boundaries of soil layers are often more diffusive than the boundaries of geological strata, we average 2 to 4 replications with different [MN] distances for a [AB] distance to provide a higher measurement accuracy. The second array provides a very high accuracy essential if the soil profile is relatively uniform in electrical resistivity. The accuracy provided with the first array is adequate for most soil applications. Other modifications of the traditional method include the reduced size and weight of electrodes, arrays with the fixed distances among electrodes, and automatic commutator for the electrode combinations. The equipment with such features allows measuring a detailed VES profile within 10 min using the first array and within 20 min using the second array.

To highlight the advantages of VES usage for soil survey we examined soil profiles with highly variable electrical resistivities. The modified VES method was tested in soil horizons outlining in elluvial-alluvial profiles of Spodosols and Alfisols in the humid areas of Russia. Other properties that highly influence the profile distributions of electrical resistivity in soils are salinity, stone or rock content, and pollution by oil or gasoline. Electrical resistivities of stones, rocks, and hydrocarbons such as petroleum, gasoline, bitumen, and oil are about thousand times higher than that of soils, whereas the resistivity of a saline soil can be much lower than that of a non-saline soil. The VES method was applied to evaluate of saline layers and groundwater depths in the alluvial soils of delta Volga, Russia. The profile distributions of stones in soils of Crimea Peninsula, Russia were successfully investigated with the VES method. The pollution by the petroleum mining was revealed in the profiles of Gelisols in North-East Siberia. The VES method was also utilized for in-situ monitoring of soil defrosting and drying processes on cultivated Alfisols and Histosols in Moscow area, Russia.


Outlining of horizons in elluvial-illuvial soil profiles

Soil horizons with highly varied chemical, physical, and electrical properties develop through the soil evolution due to processes of elluviation-illuviation, leaching, calcification-decalcification, lessivage, laterization, etc. (Wilding et al., 1983). For example, a profile of mature Ferrudalf is strongly differentiated in morphology, texture, and chemical properties, therefore, in electrical resistivity (Fig. 2a). The top horizon (A) has a relatively high humus content and cation exchange capacity, hence, a high density of mobile electrical charges and low resistivity. The underlying elluvial (albic) horizon (E) mostly consists of bleached sand grains. The density of mobile electric charges is much lower in E horizon than that in A horizon, therefore, the resistivity of E horizon is higher. The resistivity of the illuvial (spodic or argillic) horizon (B) is the lowest in the profile attributable to enrichment with the fine clay material and Fe+2/+3, Al+3, Mn+5, and other cations.

Fig. 2. Distributions of electrical resistivity measured with four-electrode probe along the soil profiles of (a) Typic Ferrudalf, Moscow area; (b) Typic Kandiudalf, Tula area; (c) Mollic Hapludalf, (d) Typic Argialboll, Kursk area; (e) Typic Calciudoll, (f) Xeric Calcigypsid, Herson area.

As the intensity of the elluviation processes decreased in the row of Ferrudalf-Kandiudalf-Hapludalf-Argialboll-Calciudoll, the differentiation of electrical resistivity in the soil profile vanished (Fig. 2). The distributions of electrical resistivity shown in Fig. 2 were measured along soil profiles in the open soil pits in different areas of Russia. Vertical electrical sounding conducted from the soil surface revealed the same profile distributions of electrical resistivity in Typic Ferrudalf as that obtained with four-electrode probe (Fig. 3, Fig. 2a). Therefore, by observation and interpretation of VES profiles we could obtain exact thicknesses and locations of soil horizons based on different electrical resistivities. The values of the electrical resistivity in E and B horizons have shown to be good indicators of intensity of the elluviation-illuviation process in soils.

Fig. 3. Profile distribution of electrical resistivity in Typic Ferrudalf measured with VES method. The mean values with error bars are shown for 5 measurements on soils under native coniferous-deciduous forest in Kalinin area, Russia.

Therefore, the VES profiles aid soil surveying and mapping, especially in evaluating soils with highly distinguished profiles. Our investigations show a similar three-layer distribution of electrical resistivity within eluvial-illuvial soil profiles in a number of soils, such as Spodosols, Alfisols, Ultisols, and Oxisols (Pozdnyakov et al., 1996). In many other soil profiles, electrical resistivity variations in the layers are attributable to other soil-forming processes, pollution, or differences in parent materials.

Estimation of saline layers and groundwater tables in alluvial soils

One of the applications of the electrical resistivity or conductivity methods is to map and evaluate soil salinity. The VES was used to outline the layers with different salinity in alluvial soil profiles at delta Volga (Halaquaepts). A typical profile of the alluvial soil consisted of thin layers of silt, clay, and sand. Soil forming processes did not develop highly pronounced horizons in these soils as compared with Spodosols, Alfisols, and Ultisols. Water and salt content distributions within the soil profile are the main properties causing considerable variations in electrical resistivity. Since the evaporation in the area was about five times higher than the precipitation, the water content and salt distributions were determined mainly by the saline groundwater.

The soil profile was divided into the top unsaturated layer with high resistivity and the bottom layer saturated by saline groundwater with low resistivity. Considering large differences in electrical resistivity between the unsaturated and saturated zones, the VES method was applied to detect the saline groundwater level. The approximate location of the groundwater table was estimated by a visual inspection of the VES curve. The AB/2 value with the sharp change to the low resistivity (3-20 ohm m) was selected from each VES profile and multiplied by an empirical coefficient (Fig. 4). This coefficient was estimated as 0.32 for the investigated soils and varied from 0.28 to 0.34 for other soil types (Barker, 1989). For example, for VES 6 the AB/2 with such sharp change is 7.2 m and the groundwater table is estimated as 7.2x0.32=2.3 m (Fig. 4). In some cases (Fig. 4, VES 3) it was difficult to visually determinate where the VES curve has a sharp change in electrical resistivity. Nevertheless, with the computer interpretation of the VES data we could determine the changes more accurately. The interpretation was based on a representation of VES profile as a numeric function. The function was analyzed to find the extremes and determine the layers with different resistivities using special procedures applied in geophysics (Vanjan and Morozova, 1962; Matveev, 1974; Pozdnyakov et al., 1996). The results of the computer interpretation of the VES data were compared with the real groundwater tables measured in bore-holes and the relative errors of the VES estimation varied from 3 to 11% as shown in Table 1.

Fig. 4. Electrical resistivity distributions in Halaquaepts profile indicated the groundwater depth.

Table 1. Estimation of groundwater table with the vertical electrical sounding

Case number

Groundwater table

Relative estimation error


Real (bore hole)

Estimated (VES)



覧覧覧覧 m 覧覧覧覧

覧覧 % 覧覧





































The differentiation of salinity in the unsaturated zone of the soil profiles was revealed by small fluctuations of electrical resistivity in upper part of the VES profiles (Fig. 4). We thoroughly interpreted the VES results to estimate the layers with different resistivities for 12 soil profiles. The total salt content was measured in soil samples collected from the layers of soil profiles as shown in Table 2 (columns 1 and 2) for one example profile. The VES method outlined three layers with different resistivities for the same soil profile (Table 2, columns 3 and 4). In the column 5 a weighted averages for the outlined soil layers were recalculated from the total salt contents in column 2 (Table 2). Data of recalculated total soil salinity and VES electrical resistivity were combined from the layers of all 12 soil profiles to obtain a relationship between electrical resistivity and total salt content (Fig. 5). The correlation coefficient (r) was calculated for the linearized power relationship as 0.915. Data for the soil layers with resistivity higher than 20 ohm m were excluded from the correlation analyses, because they were mainly corresponded to non-saline and very low saline soils with the total salt content less than 0.3%. For quick delineation and estimation of salinity in a soil profile we can consider that a resistivity of 10-20 ohm m corresponds to a total salt content between 0.3 and 0.5% and a resistivity <3 ohm m indicate that the total salt content in soil is >1%.

Fig. 5. The relationship between the electrical resistivity measured in-situ by VES method and the total salt content in soils of delta Volga, Russia.

Table 2. Illustration to evaluation of soil salinity with the vertical electrical sounding


Total salinity

Results of interpretation

Recalculated salinity



Layer depth


for interpretation layers


覧 % 覧


ohm m

覧覧 % 覧覧
































Stone content evaluation

Layers of different electrical resistivities are observed in stony or rocky soils. The resistivity of rocks or stones is much higher (about 104-1012 ohm m) than the resistivity of soil horizons with any texture (Dohr, 1981; Parasnis, 1997). Based on the principle the VES was applied to locate depths of stone layers in soils of Crimea Peninsula, Russia (Lithic Xerorthent). Most of the soils could be well characterized by three-layer profile. The top layer (I) had the smallest stone content (0.22-0.41 cm3 cm-3) with electrical resistivity about 80 ohm m (Fig. 6). The middle layer (II) had the highest stone content (>50 cm3 cm-3) and electrical resistivity as high as 450 ohm m. Stones in layer II were cemented by carbonate marine deposits. The bottom layer (III) was not always presented in soil profiles. In some profiles layer III was outlined as having lower resistivity (40-200 ohm m) than layer II. This indicated decrease in stone content in the bottom layer compared with layer II and a loose organization of stones in the bottom layer.

Fig. 6. Profile distributions of electrical resistivity measured by VES method in stony soils of Crimea Peninsula, Ukraine.

The approximate stone content of soil profiles was evaluated observing VES profiles. For example in Fig. 6, the stone content in the soil profiles decreased in a row VES 3-7-2-16-10. The stone content estimations based on VES profiles were verified by the direct measurement of the relative volume of stones in the soil layers. Using the VES interpretation data and the direct measurements of the stone content in 16 soil profiles, we developed a relationship between electrical resistivity measured by the VES and soil stone content. The power relationship was obtained for 1/stone content vs. electrical resistivity. The relationship is shown in Fig. 7 and has a correlation coefficient (r) of 0.863. The relationship is useful for a quick judgement about stone content in different soil layers. We developed a rough scale for evaluation of stone contents in soils (Table 3). With the scale, the territory about 400 acres was mapped with the VES and four-electrode profiling methods, the stone content and texture of soils were evaluated, and special recommendations were developed for usage these otherwise unproductive soils under the orchards.

Fig. 7. The relationship between the electrical resistivity measured in-situ by VES method and the volumetric stone content.

Table 3. Scale for evaluation of soil stone content by electrical resistivity measured with VES

Stone content

Electrical resistivity

覧 cm3 cm-3

覧覧 ohm m 覧覧








Estimation of soil pollution by petroleum mining

Besides stones and rocks, petroleum and various products of petroleum manufacture, such as oil, gasoline, bitumen, and kerosine also have very high electrical resistivity compared with soils. Electrical resistivity of petroleum varies from 104 to 1019 ohm m (Fedinsky, 1967), whereas resistivity of petroleum-saturated sand is lower (2200 ohm m) (Znamensky, 1980), which is still higher than that of any non-polluted soil. For any soil type the resistivity is within several hundred ohm m (Pozdnyakov et al., 1996).

Table 4. Electrical resistivity of native and polluted soils in North-East Siberia


Electrical resistivity

覧 ohm m 覧

surface layers of non-polluted Gelisols

2 102 2 103


4 103 8 103

polluted by bitumen and other heavy fraction of oil

1 105 6 105

polluted by gasoline

1 104 4 104

polluted by salty mining water

2 10 2 102

Table 4 shows the average values of electrical resistivity of natural non-polluted soils (Glacic and Aquic Haplorthels) and soils polluted by petroleum and gas mining in North-East Siberia. In this particular case the pollution by petroleum products highly increased the soil electrical resistivity, whereas brine solutions used for the mining considerably decreased soil resistivity. Based on such high differences in electrical resistivities we could evaluate the pollution by petroleum products and salty mining solutions distributed in soil profiles (Fig. 8). Pollution by heavy fraction of petroleum, such as bitumen appeared at the top part of soil profile and was indicated by electrical resistivity as high as 6 105 ohm m (Fig. 8c). The pollution by salty mining solutions, on the contrary, lowered soil electrical resistivity. The resistivity of the soil near the stream where brine from the mining site discharged, varied from 50 to 200 ohm m (Fig. 8b). The surface soil at the brine collector has resistivity as low as 20 ohm m (Fig. 8d), while the electrical resistivity of the native pergelic soils was about 1000 ohm m at the surface (Fig. 8a). Some non-polluted native soils shown increase in electrical resistivity up to 8000 ohm m at the AB/2=2.4 m (about 0.6-m depth) indicated the presence of permafrost in soil profile (Fig. 8a). The depth of the permafrost was verified by the direct soil excavation.

Fig. 8. Profile distributions of electrical resistivity measured by VES method in soils of North-East Siberia polluted by petroleum and gas mining: (a) non-polluted Gelisols, (b) soil near the stream with mining solution discharge, (c) soil polluted with bitumen, (d) soil in a brine collector.

Monitoring of soil defrosting

High electrical resistivity of the permafrost layers compared with unfrozen soil can be explained by the mechanisms of soil electrical resistivity or conductivity. Soil primarily conducts electricity by the mobile ions of soil solution. When the soil solution freezes the ions became immobile, which highly increases the electrical resistivity of the soil. Therefore, the freezing and defrosting processes can considerably alter the distribution of electrical resistivity in soil profiles.

We conducted a seasonal monitoring of the electrical resistivity distribution in a profile of intensively cultivated sandy loam Ferrudalf. The electrode array was grounded before soil was frozen in August of 1996 and left in place over the winter. The process of soil defrosting was monitored in about 4-hour intervals during April of 1997. The fixed position of the electrode array insured a high accuracy of measurement and elimination of errors caused by soil spatial variability. Figure 9 shows the changes in the profile distribution of electrical resistivity as the soil was defrosting. The electrical resistivity of the whole soil profile was around 2,000 ohm m, while in the deeper layer (AB/2 = 7.2 m and h 2.2 m) resistivity was as high as 14,000 ohm m on April 2. The deep soil layers started to defrost at the beginning of April as indicated by the changes in electrical resistivity from 14,000 to 8,000 ohm m during 25 hours (Fig. 9a). The top layer was not frozen as much as the deeper soil profile and ER for the top layer was about 3,000 ohm m at the beginning of April (Fig. 9a). The top layer, however, defrosted more slowly than the deeper soil profile, probably, because of snow isolation of the soil surface and water saturation of the topsoil. The electrical resistivity of the top layer changed from 3,000 to 900 ohm m from 2 to 14 of April, whereas soil at 2.2-m depth was completely defrosted during that period with ER about 100 ohm m (Fig. 9b). The snow cover at the soil surface completely melted on April 14 and on April 16 the soil profile was all defrosted and had uniform distribution of electrical resistivity about 150 ohm m (Fig. 9b).

Fig. 9. Changes in profile distributions of electrical resistivity during the spring soil defrosting of cultivated Ferrudalf, Moscow area, Russia.

The distribution of electrical resistivity (as on April 16) was fairly stable for this soil during the growing season. The distribution of electrical resistivity for the cultivated Ferrudalf was different from those for native Ferrudalfs shown in Figs. 2 and 3 because the soil profile was heavily manured and the soil horizons were mixed. However, the slight increase of resistivity in the middle part of the profile reflected remains of the E horizon mixed with A horizon and organic manure (Fig. 9b, VES for April 16). The electrical resistivity increased in August, 1997, and varied from 80 to 350 ohm m within the profile, probably, because of the overall soil drying. Nevertheless, the shape of the electrical resistivity profile was the same as that on April 16. The results demonstrated that the profile distributions of electrical resistivity in soils of humid areas are fairly stable.

Estimation of drying depth in peat soil

Usually water content of cultivated peat soils is close to the field capacity during the whole growing season. The other soil properties, such as bulk density, texture, and ash content, that might influence electrical resistivity, are also practically uniform in the soil profile. Therefore, a typical electrical resistivity distribution in the profile of cultivated peat soil (Hemic Haplosaprist) is uniform and about 100 ohm m during the years with average precipitation (Fig. 10, VES for 4 July, 1994). In extremely dry years as in the summer of 1995 in Moscow area, Russia, the upper 50-cm layer of peat soil dried almost to the wilting point, causing an increase of electrical resistivity up to 1,500 ohm m (Fig. 10). The drying depth was precisely determined with the VES interpretation and verified with direct water content measurements.

Fig. 10. Determination of drying depth on the cultivated Hemic Haplosaprist by the VES measurements. Moscow area, Russia.



This study provided a basis for using the vertical electrical sounding method in soil survey. The conventional VES method commonly used for deep geophysical exploration was modified to make it suitable for soil studies. The specific design of the instrument allows accurate measuring of the electrical resistivity distribution in soil profiles without disturbing soils. In previous studies, electrical geophysical methods were only used to measure bulk electrical conductivity of the soil volume for the evaluation of soil salinity. The paper has shown many other possible applications of the VES method for soil investigation.

The study demonstrated that although soil electrical resistivity depends simultaneously on many soil properties, such as salt, water, humus or stone content, texture, and temperature, in many applications one or two highly variable properties can be considered as main factors influencing the profile distribution of electrical resistivity. Thus, the profile distribution of electrical resistivity measured with the VES method reflects the profile distribution of such key soil properties. For example, as the variation in stone content influences the soil electrical resistivity much stronger than variation of any other properties in soils of Crimea Peninsula, the VES method was able to accurately outline the layers with different stone contents in these soils. Salt and water contents were two properties considerably influencing the electrical resistivity of alluvial soils in Astrachan area, therefore, VES was used to outline groundwater table and layers with different salinity levels in the soils. Pollution by petroleum products highly increased the electrical resistivity of Gelisols in North-East Siberia, while salty mining solution decreased resistivity of the soils. Soil horizons with different electrical resistivities formed in Spodosols, Alfisols, Ultisols, Oxisols, and some Aridosols; thus, the VES method was successfully used to outline horizons in elluvial-illuvial soil profiles.

Soil water content and temperature are highly spatially and temporally variable soil properties. Our study has shown that these properties considerably influence soil electrical resistivity only if they change around some extreme values. Thus, the variation of soil temperature around freezing point causes high changes in soil electrical resistivity and allows monitoring of defrosting process in soils. Extreme dryness of Histosol in some seasons highly increased the electrical resistivity at the top of the profile, whereas, variation of soil water content around field capacity usually does not alter the typical profile distributions of electrical resistivity in the soils. Generally, the VES method can be used for in-situ soil monitoring when the monitored property alone highly influences the distribution of electrical resistivity in a soil profile. The VES method can be useful for monitoring saline solution movement, defrosting, drying, or compaction of soils.


Arcone, S.A., D.E. Lawson, A.J. Delaney, J.C. Strasser, and J.D. Strasser. 1998. Ground-penetrating radar reflection profiling of groundwater and bedrock in an area of discontinuous permafrost. Geophysics. 63:1573-1584.

Austin, R.S., and J.D. Rhoades. 1979. A compact, low-cost circuit for reading four-electrode salinity sensors. Soil Sci. Soc. Am. J. 43:808-809.

Banton, O., M.K. Seguin, and M.A. Cimon. 1997. Mapping field-scale physical properties of soil with electrical resistivity. Soil Sci. Soc. Am. J. 61:1010-1017.

Barker, R.D. 1989. Depth of investigation of collinear symmetrical four-electrode arrays. Geophysics. 54:1031-1037.

Barker, R.D. 1990. Improving the quality of resistivity sounding data in landfill studies. In S.H. Ward (ed). "Geotechnical and environmental geophysics. V. 2. Environmental and groundwater applications". 245-251.

Beck, A.E. 1981. Physical principles of exploration methods. John Willey & Sons, NY.

Corwin, D.L., and J.D. Rhoades. 1984. Measurement of inverted electrical conductivity profiles using electromagnetic induction. Soil Sci. Soc. Am. J. 48:288-291.

Dohr, G. 1981. Applied Geophysics. Introduction to geophysical prospecting. In H. Beckmann (ed.). Geology of petroleum, Vol. 1. Halsted Press, New York, Wiley Ltd., Chichester.

Fedinsky, V.V. 1967. Exploration geophysics. (In Russian.) Nedra, Moscow.

Freeland, R.S., J.C. Reagan, R.T. Burns, and J.T. Ammons. 1997b. Noninvasive sensing of near-surface perched water using ground-penetrating radar. Proceeding of the ASAE Annual International Meeting. Minneapolis, Minnesota. August 10-14. N 973073.

Freeland, R.S., J.D. Bouldin, R.E. Yoder, D.D. Tyler, and J.T. Ammons. 1997a. Mapping preferential water flow paths beneath loess terrains using ground-penetrating radar. Proceeding of the ASAE Annual International Meeting. Minneapolis, Minnesota. August 10-14. N 973074.

Halvorson, A.D., and J.D. Rhoades. 1976. Field mapping soil conductivity to delineate dryland saline seeps with four-electrode technique. Soil Sci. Soc. Am. J. 40:571-574.

Keller, G.V., and F.C. Frischknecht. 1966. Electrical methods in geophysical prospecting. Pergamon Press. Oxford, New York, Toronto, Sydney, Braunschweig.

Kirkham, D., and G.S. Taylor. 1949. Some tests of a four-electrode probe for soil moisture measurement. Soil Sci. Soc. Proc. 14:42-46.

Liner, C. L., and J. L. Liner. 1997. Application of GPR to a site investigation involving shallow faults. The Leading Edge. November. 1649-1651.

Matveev, V.K. 1974. Interpretation of electromagnetic sounding. (In Russian.), Nedra, Moscow.

Mazac, O., M. Cislerova, W.E. Kelly, I. Landa, and D. Venhodova. 1990. Determination of hydraulic conductivities by surface geoelectrical methods. In S.H. Ward (ed). "Geotechnical and environmental geophysics. V. 2. Environmental and groundwater applications". 125-131.

McBride, R.A., A.M. Gordon, and S.C. Shrive. 1990. Estimating forest soil quality from terrain measurements of apparent electrical conductivity. Soil Sci. Soc. Am. J. 54:290-293.

Parasnis, D.S. 1997. Principles of applied geophysics. Chapman & Hall, 2-6 Boundary Row, London SE1 8HN, UK.

Pozdnyakov, A.I., and K.Y. Chan. 1979. The technique of the electrical sounding and profiling in soil investigations. (In Russian.) Vestnik MGU 17. 1:46-54.

Pozdnyakov, A.I., L.A. Pozdnyakova, and A.D. Pozdnyakova. 1996. Stationary electrical fields in soils. (In Russian.) KMK Scientific Press, Moscow.

Pozdnyakova, L.A., A.I. Pozdnyakov., and L.O. Karpachevsky. 1996. Study hydrology of valley agricultural landscapes with electrical resistance methods. Proceeding of XXI Assembly European Geophysical Society, HS16 "The Hydrology of Small Agricultural Catchments", The Hague, Netherlands. 341-352.

Rhoades, J. 1979. Inexpensive four-electrode probe for monitoring soil salinity. Soil Sci. Soc. Amer. J. 43:817-818.

Rhoades, J.D., and J. van Schilfgaarde. 1976. An electrical conductivity probe for determining soil salinity. Soil Sci. Soc. Am. J. 40:647-650.

Rhoades, J.D., and R.D. Ingvalson. 1971. Determining salinity in field soils with soil resistance measurements. Soil Sci. Soc. Amer. Proc. 35:54-60.

Rhoades, J.D., P.G. Shouse, W.J. Alves, N.A. Manteghi, and S.M.Lesch. 1990. Determining soil salinity from soil conductivity using different models and estimates. Soil Sci. Soc. Am. J. 54:46-54.

Vanjan, L.L., and G.M. Morozova. 1962. Calculation of the theory curves of electrical sounding. (In Russian.), Gostoptehizdat, Moscow. Practical Geophysics, 34:135-144.

Verma, R.K., and T.K. Bandyopadhyay. 1983. Use of the resistivity method in geological mapping case histories from Raniganj coal field, India. Geophysical Prospecting, 31:490-507.

Wilding, L.P., N.E. Smeck, and G.F. Hall (ed.). 1983. Pedogenesis and soil taxonomy. I. Concepts and interactions. Developments in Soil Science, 11A. Elsevier. Amsterdan, Oxford, New York.

Znamensky, V.V. 1980. Field geophysics. (In Russian.) Nedra, Moscow.