Soil Electrical Conductivity Variability

N. J. Hartsock, T. G. Mueller, G. W. Thomas, R. I. Barnhisel, K. L. Wells, and S.A. Shearer


Hartsock, N. J., T. G. Mueller, G. W. Thomas, R. I. Barnhisel, and K. L. Wells and S. A. Shearer, .  2000.  Soil Electrical Conductivity Variability. In. P.C. Robert et al. (ed.) Proc. 5th international conference on precision Agriculture. ASA Misc. Publ., ASA, CSSA, and SSSA, Madison, WI.


ABSTRACT

    Commercial sensors are available that allow rapid field mapping of soil electrical conductivity; however there has been little published research in this area. This study was conducted to determine the nature and causes of soil conductivity variability using a Veris 3100 soil electrical conductivity sensor. The research was done throughout the year on fields in the Pennyroyal and Outer Bluegrass regions of Kentucky. Soil fertility measurements (pH, BpH, P, K, Ca, Mg, and organic carbon), surface soil moisture, surface soil temperature, topsoil thickness, depth to fragipan, depth to clay increase, and depth to bedrock were measured on a number of points in each field. Conductivity varied in both space and time, but the spatial patterns in conductivity were temporally stable. Conductivity was positively related to soil Ca (r2=0.59; Shelby Co.) and Mg (r2 = 0.56; Shelby Co.) and soil moisture (r2=0.72, across locations), and inversely related to depth to a clay increase (r2=0.38, Hardin Co.; r2 = 0.27, r2 = 0.66 Shelby Co.), depth to bedrock (r2 = 0.33, Shelby Co.), and depth to fragipan (r2 = 0.80, Shelby Co). Soil electrical conductivity may be useful in production agriculture because it relates to factors that affect soil productivity, use, and management.  

 

INTRODUCTION

     Soil electrical conductivity (EC) is a property of soil that is determined by standardized measures of soil conductance (resistance-1) by the distance and cross sectional area through which a current travels. Traditionally, soil paste EC has been used to assess soil salinity (Rhoades et al., 1989), but now commercial devices are available to rapidly and economically measure and map bulk soil EC across agricultural fields. The Veris 3100 (Veris Technologies, Salina, Kansas) measures EC with a system of coulters that are in direct contact with the soil. The EM38 (Geonics, Limited, Mississauga, Ontario, Canada) induces a current into the soil with one coil and determines conductivity by measuring the resulting secondary current with another coil. Both sensors have been demonstrated to give similar results (Suddeth et al., 1999)

     The movement of electrons through bulk soil is complex. Electrons may travel through soil water in macropores, along the surfaces of soil minerals (i.e. exchangeable ions), and through alternating layers of particles and solution (Rhoades et al., 1989). Therefore, multiple factors contribute to soil EC variability, including factors that affect the amount and connectivity of soil water (e.g. bulk density, structure, water potential, precipitation, timing of measurement), soil aggregation (e.g. cementing agents such as clay and organic matter, soil structure), electrolytes in soil water (e.g. salinity, exchangeable ions, soil water content, soil temperature), and the conductivity of the mineral phase (e.g. types and quantity of minerals, degree of isomorphic substitution, exchangeable ions). Despite the multiple causes of EC variability, bulk soil EC measurements have been related to individual factors that limit soil use and productivity such as salinity (De Jong et al., 1979; Rhoades and Corwin, 1981), clay content at a depth of 15-m in New Wales, Australia (r2 = 0.78; Williams and Hoey, 1987), depth of sand deposition along the Missouri River (r2 = 0.73-0.94; Kitchen et al, 1996), depth to claypan in Missouri (r2 = 0.73; Doolittle et al., 1994), and soil moisture content (r2 = 0.96; Kachanoski et al., 1988).

    If soil EC maps have utility in production agriculture, 1) EC must be spatially structured, 2) spatial patterns must have temporal stability, and 3) EC must be related to factors of agronomic importance. The objective of this research was to determine the nature and causes of EC variability for several fields in Kentucky. Geostatistical analyses were conducted to examine the spatial and temporal variability of EC variability. Transect studies were conducted to determine the causes of EC variability.

 

METHODS

 Site Description

     This research was conducted in a field in Hardin Co, KY (Field 1), and three fields in Shelby Co, KY (Field 2, Field 3, and Field 4). Field 1 consists of the Vertrees series (Fine, mixed, mesic Typic Paleudalfs), which formed in residuum from limestone, the Nolin series (Fine-silty, mixed, mesic Dystric Fluventic Eutrochrepts), which formed in mixed alluvium, and the Crider series (Fine-silty, mixed, active, mesic Typic Hapludalfs), which formed in loess over limestone residuum. Field 2 consists of the Nicholson series (Fine-silty, mixed, mesic Typic Fragiudalfs) and the Lowell series (Fine, mixed mesic Typic Hapludalfs), both of which formed in residuum from limestone and shale and have a loess cap. Field 3 has both the Nicholson series and the Shelbyville series (Fine-silty, mixed mesic Mollic Hapludalfs), which formed in loess over residuum from limestone. Field 4 contains the Cynthiana series (clayey, mixed, mesic Lithic Hapludalfs) and Faywood series (fine, mixed, active, mesic Typic Hapludalfs), both which are shallow to bedrock and formed in residuum from limestone.

Soil EC Data Collection

      A Veris 3100 Soil EC Mapping System was used to measure soil EC. The sensor consists of six coulters, two of which introduce an electrical potential into the soil. The remaining four coulters are spaced to measure EC over two approximate depths, 0-30.5-cm (EC30.5) and 0-91.5-cm (EC91.5). When used in conjunction with a DGPS receiver, EC data can be geo-referenced to create a map.

Transect EC Measurements

     Transects were selected from Field 1, Field 3, and Field 4 (Fig. 1). At selected points on each transect, several measurements were taken in addition to EC. Volumetric water content to a depth of 12-cm was determined with the HydroSenseTM (Decagon Devices, Inc, Pullman, Washington), which uses transmission line oscillation. Depth to a clay increase was assessed using the �texture by feel� method. Penetrometer resistance was measured on all Field 1 transects. Depth measurements to fragipan (Field 3) and to bedrock (Field 4) where measured only in the fields where these factors affected crop production. Soil EC values for each point on the transect were determined by driving directly over the sampling point, stopping, and recording the EC30.5 and EC91.5 values. In addition, the transect EC values in Field 3 were measured on two consecutive days.

 

Whole Field EC Measurements

     Whole field EC data were collected for a location in Hardin Co. (Field 1) on three dates: Oct. 8, 1999, March 7, 2000, and May 5, 2000. A second location in Shelby Co. (Field 2) was measured once on July 7, 1999. The fields were traversed with approximately 7.5-m between passes on each date. Field 3 was traversed in the north-south direction and the east-west direction. Data were recorded every second, and groundspeed was maintained at approximately 10.5 km hr-1. Soil samples were collected for the top 15-cm using a 30.5-m grid pattern (Fig. 2). Soil analyses, including pH, buffer pH, organic matter content, and Mehlich III extractable P, K, Ca, Mg, and Zn, were performed by the Department of Regulatory Services, at the University of Kentucky. At each sample site, all EC values falling within a 4.6-m radius were averaged and related to EC30.5 and EC91.5 using simple linear regression.

 

RESULTS AND DISCUSSION

Nature of the Variability

     Day-to-day variability was greater for EC30.5 than for EC91.5 (Fig. 3). While collecting EC data, extreme values were encountered on occasion, as can be seen by the high value for May 4th EC91.5 in Figure 3. The small-scale temporal variability also gives an indication of measurement error. The relative nugget variances for both depths also give an indication of measurement error (EC30.5, 46%; EC91.5, 27%) and were large.

     The larger scale temporal variability reflects changes in EC associated with different environmental conditions (Fig. 4). While the EC values were substantially lower during the drought of 1999 (October 8th, 1999) than in the spring of 2000, the general spatial patterns in EC were similar across all three dates.

     Both EC30.5 and EC91.5 tended to be higher on the Vertrees series, which is located in the lower right, lower left, and upper right regions of Field 1 (Fig. 4). The values tended to be lower for the Nolin series, which are mainly located in the depressions. The remainder of the field was the Crider series. Soil EC was a good indicator of soil type for this field.

     The Vertrees soils have a red, Bt horizon near the surface, which increases conductivity (Fig. 4). The EC values for the Vertrees series were not as high on October 8th, 1999 (during the drought) as on the later two dates. This suggests that soil moisture enhances the conductivity of clay. This did not, however, greatly change the overall appearance of the maps on the different dates because each was based on an equal number of observations in each category rather than equal sizes of mapping intervals (note the ranges in Fig. 4). Date of the measurement did, however, change the spatial statistics of the data (Fig. 5). Anisotropic behavior was not apparent on October 8th, 1999, but it was on the latter two dates. The dark diagonal lines indicate the direction of the anisotropic axis with of minimum spatial variability (here the northwest-southeast direction). Orthogonal to this direction indicates the direction with the maximum spatial variability. [For more detail on semivariogram surfaces see Isaaks and Srivastava (1989) and Goovaerts (1997)] Anisotropy was present in this field because the Vertrees soil, which occurred in southwest, northeast, and southeast regions of the field, had very large EC values on the latter two dates when soil moisture was greater. Therefore, variability was much greater in the southwest-northeast direction. In all cases, the anisotropy does not seem to be a great issue within the first 50-m. Anisotropy was less important with EC91.5. This may be because at greater depths, the soil was more uniform in clay and moisture content. The impact of anisotropy in this field would depend upon the depth of interest, whether the data would be used for interpolation, and the sampling interval.

     Users should be aware that spatial-temporal interactions, such as with soil moisture and depth to clay increase, can affect variability and the appearance and

Causes of the Variability

     Soil Ca and Mg explained a large amount of the variability in both fields. The relationship was much stronger in Field 2 than in Field 1 (Table 1), which may be explained by the large ranges of concentrations in Field 2 (Fig. 6). The results for Field 2 are consistent with McBride et al. (1990) for forest soils in Ontario.

     The relationship between soil moisture and EC was much stronger for Field 4 than Fields 1 and 3 (Table 2; Fig. 7). Although Kachanoski et al. (1988) found a linear relationship for a study in Canada between these variables up to 25% volumetric water content, above which there was little change in soil EC, Field 4 exhibited a linear relationship up to at least 45%. The Canadian study may not be directly comparable because the soils were sandy whereas our soils tend to be silty. Kachanoski et al. (1988) found that when clay content was low, soil moisture had a greater impact on EC, which may explain why moisture governed EC in Field 4 but not in Field 1. The soils in Field 1 were much more clayey than in Field 4. 

     As the temperature of nearly all metals increases, the metal�s conductance decreases. Electrolytes, however, exhibit the opposite relationship. Increased solution temperature decreases the liquid�s viscosity, which increases the ease with which ions can move (McNeill, 1980). We did not find any relationships with temperature and EC (Table 2, Fig. 7) most likely because the ranges of temperatures within the fields were small (i.e. 1.7 C, Field 1; 3.5 C Field 2). However, seasonal fluctuations in soil temperature may affect EC.

     Clays greatly impact EC because of their exchangeable cations and the water film associated with them (McNeill, 1980); however there was little relationship between surface clay content and soil EC. There was, however, a relationship between soil EC and depth to clay increase, consistent with other research studies (Doolittle et al., 1994; Suddeth et al., 1999) that related EC to depth of claypan.

     Soil EC to evaluate depth at which clay increases occur may be a valuable tool in Kentucky for assessing soil use limitations and for management decisions [e.g. variable-rate nitrogen and seeding prescriptions, Barnhisel et al. (1996)]. This is because depth to clay is highly spatially variable in Kentucky and because it affects the effective rooting depth and plant available water. In many Kentucky soils, this can be considered the depth to the argillic horizon. Depth to clay is variable because of the original depth of the loess deposit over the limestone residuum and differential erosion across the landscape.

     Fields 3 and 4 contain soils with restrictive layers at depths of less than 2-m. Field 3 contains a fragipan soil (Nicholson series) and Field 4 contains soils that have shallow bedrock (Cynthiana and Faywood series). We found an inverse relationship between EC and both depth to fragipan and depth to bedrock (Fig. 8). One possible explanation for this behavior is that water may sit above these layers thereby increasing conductivity. If water is so important in the detection of these layers, then soil moisture contents at the time of sampling may be critical. 

 

CONCLUSIONS

     Soil EC was temporally variable, but overall spatial trends were consistent between dates. Temporal variability and measurement error both affect EC stability, but the collection of a large amount of data overrides these problems.

     There was an inverse relationship between EC and factors that restrict rooting depth and limit yields of crops (depth to clay increase, depth to bedrock, and depth to fragipan). The only chemical properties that were strongly related to EC were levels of exchangeable Ca and Mg in Field 2, where average contents of both Ca and Mg were high.

     An EC sensor may be useful in assessing the spatial variation in productivity in agricultural fields. In order for the EC data to be useful, however, field-specific calibration is necessary because several soil characteristics can affect EC simultaneously.

     Soil EC has value in delineating within-field variability. Because EC is controlled by ionic concentration, clay, and soil moisture, relationships may vary depending on the soil series, timing of data collection, and soil moisture status. If sufficient investigation is used to determine the causes of field EC variability, then EC maps may be useful in soil and crop management.

 

ACKNOWLEDGEMENTS

    Much appreciation is extended to Mike Ellis and Charlie Stuecker for allowing this research to be conducted on their farms. In addition, we would like than the Shelby Co. extension agent, Brittany Edelson, and the Hardin Co. extension agent, Rod Grusy.

 

REFERENCES

Barnhisel, R.I., M.J. Bitzer, and J.H. Grove. 1996. Agronomic benefits of varying corn seed populations: a central Kentucky study. p. 957-966. In P.C. Robert et at. (ed.) Proc. 3rd international conference on precision agriculture. ASA Misc. Publ., ASA, CSSA, and SSSA, Madison, WI.

De Jong, E., A.K. Ballantyne, D.R. Cameron, and D.W.L. Read. 1979. Measurement of apparent electrical conductivity of soils by an electromagnetic induction probe to aid in salinity surveys. Soil Sci. Soc. Am. J. 43:810-812.

Doolittle, J.A., K.A. Suddeth, N.R. Kitchen, and S.J. Indorante. 1994. Estimating depths to claypans using electromagnetic induction methods. J. Soil Water Cons. 49(6):572-575.

Goovaerts. P. 1997. Geostatistics for natural resource evaluation. Oxford University Press, New York. 483 pp.

Isaaks, E. and R. Srivastava. 1989. An introduction to applied geostatistics. Oxford University Press, New York, NY.

Kachanoski, R.G. E.G. Gregorich, and I.J. Van Wesenbeeck. 1988. Estimating spatial variations of soil water content using no contacting electromagnetic inductive methods. Can. J. Soil Sci. 68:715-722

Kitchen, N.R., K.A. Suddeth, and S.T. Drummond. 1996. Mapping of sand deposition from 1993 midwest floods with electromagnetic induction measurements. J. Soil Water Cons. 51(4):336-340.

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.

McNeill, J.D. Electrical conductivity of soils and rocks. Tech. Note TN-5. Geonics Ltd., Mississauga, ON.

Rhoades, J.D. and D. L. Corwin. 1981. Determining soil electrical conductivity-depth relations using an inductive electromagnetic soil conductivity meter. Soil Sci. Soc. Am. J. 45:255-260.

Rhoades, J.D., N.A. Manteghi, P.J. Shouse, and W.J. Alves. 1989. Soil electrical conductivity and soil salinity: new formulations and calibrations. Soil Sci Soc. Am J. 53:433-439 

Suddeth, K.A., N.R. Kitchen, S.T. Drummond. 1999. Soil conductivity sensing on claypan soils: comparison of electromagnetic induction and direct methods. p. 971-990. In. P.C. Robert et al. (ed.) Proc. 4th international conference on precision agriculture. ASA Misc. Publ., ASA, CSSA, and SSSA, Madison, WI.

Williams, B.G, and D. Hoey. 1987. The use of electromagnetic induction to detect the spatial variability of the salt and clay contents of soil. Aust. J. Soil Sci. 25:21-27.