HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Hall, L. M. Articles by Hendrickx, J. M. H. Search for Related Content PubMed Articles by Hall, L. M. Articles by Hendrickx, J. M. H. GeoRef GeoRef Citation Agricola Articles by Hall, L. M. Articles by Hendrickx, J. M. H. Related Collections Vadose Zone Processes and Chemical Transport Field-Scale Studies Electromagnetic Induction, EMI

SPECIAL SECTION: HYDROGEOPHYSICS

Determination of Solute Distributions in the Vadose Zone Using Downhole Electromagnetic Induction

^a New Mexico Institute of Mining and Technology, Department of Earth and Environmental Science, 801 Leroy Place, Socorro, NM 87801
^b Sandia National Laboratories, Geohydrology Department, P.O. Box 1101, Sandia Park, NM 87047
^* Corresponding author (bowman{at}nmt.edu). Published in Vadose Zone Journal 3:1207–1214 (2004)

¹ Product identification in this manuscript is informational only; it is not a guarantee and does not imply that other products would not also be suitable.

Received 29 January 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
We conducted a field experiment to evaluate the ability of downhole electromagnetic induction (EM) measurements to track the migration of a salt plume in the vadose zone. A 6.9 g L^–1 slug of NaCl was applied to a 3 by 3 m area for 85 d at a flux of 2.7 cm d^–1. Electrical conductivity measurements were collected weekly, using 13 12-m boreholes installed in a 15 by 15 m area. The measured bulk soil electrical conductivity (EC_a) was converted to soil water electrical conductivity (EC_w) using water contents from neutron probe measurements and clay contents from soil cores. The calculated mass of salt in the profile agreed well with the known mass of salt infiltrated when the appropriate immobile water content was assumed. The low water content, _w, (<15% by volume) and low bulk soil electrical conductivity, EC_a, (<100 mS m^–1) measured at the test site presented more resistive conditions than previous studies of this type and were a cause of uncertainty in the results. Good agreement between calculated and known EC_w resulted from the ability to measure _w and EC_a in the same locations. Sensitivity analyses showed that the calculated EC_w was strongly dependent on the assumed upper limit of immobile water content, and less sensitive to soil temperature and clay content. The results of this study demonstrate that downhole EM methods can accurately characterize water and solute distributions in the vadose zone.

Abbreviations: EM, electromagnetic induction

	INTRODUCTION

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
VADOSE ZONE FLOW and transport processes are inherently complex because of the interactions between the liquid, gas, and solid phases in unsaturated soils. In particular, textural heterogeneities have a profound effect on the distribution of water and solutes as a result of the nonlinear dependence of the hydraulic properties on the water content and pressure head. Therefore, spatially variable water contents are the rule and not the exception. Adding to this complexity is the potential for preferential flow paths through root casts, insect and animal holes, fractures, and cracks. Preferential flow paths also occur across contacts of contrasting textures in basin fill deposits. As a result, sparse point measurements of water contents and solute concentrations can fail to provide the information needed to identify vadose zone flow and transport mechanisms and can lead to grossly inaccurate understanding of solute distributions. Demand for accurate mapping of the vadose zone is on the rise as it becomes clear that remediation of contaminants is usually much more time- and cost-effective if they can be treated before reaching the groundwater.

We present the results of efforts to identify subsurface distributions of water and solutes using EM. There is a close relationship between EM-sensed bulk soil electrical conductivity (EC_a) and soil water content, clay content, and soil water conductivity. Detailed descriptions of EM methods and their applications are provided by Hendrickx and Kachanoski (2002), Hendrickx et al. (2002), and McNeill (1980a)(1980b). Electromagnetic induction measurements are often taken with handheld probes near the surface (using, e.g., the EM31 or EM38 from Geonics, Ltd., Mississauga, ON, Canada), but can also be taken aerially from a plane, with both methods offering a quick means of mapping over a large area.¹ The EM39 is a slimmer version of the EM31 and EM38 probes, allowing it to be lowered down boreholes to take measurements of bulk soil electrical conductivity vertically instead of horizontally. Due to its design the EM39 is most commonly used to verify surface EM measurements in a few locations, or in well logging to detect changes in stratigraphy. The EM39 shows promise for subsurface mapping, but is not commonly used because it requires a suitable distribution of boreholes across the area of interest.

We therefore designed an experiment to test the utility of the EM39 for determining solute distributions in a desert vadose zone. We applied water and solute (NaCl) under controlled conditions at the surface and measured changes in EM to a depth of 12 m in 13 boreholes. Specific objectives of this study were to (i) image an evolving solute plume with time, (ii) evaluate stratigraphic controls on plume development, (iii) compare the calculated solute mass in the profile to the known solute mass applied, and (iv) evaluate the sensitivity of the plume distributions and calculated solute mass to media characteristics and environmental factors.

	THEORY

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The bulk electrical conductivity of a soil is the combination of conductivity in the aqueous and solid phases. Rhoades et al. (1989) defined the three pathways for conductance as (i) the alternating layers of soil particles and interstitial soil solution (a solid–liquid coupled element), (ii) the surfaces of soil particles in contact with each other (a solid element), and (iii) the soil pore water (a liquid element). Assuming the pathways acted in parallel, Rhoades et al. (1989) derived the following expression relating bulk (measured) electrical conductivity, EC_a, to the conductivities of the individual pathways:

[1]

where EC_wc is the conductivity of the mobile soil water, EC_ws is the conductivity of the immobile water, EC_s is the conductivity of the soil particles, _s is the ratio of soil bulk density (_b) to soil particle density (_s), and _wc and _ws are the mobile and immobile volumetric water contents, respectively (_wc + _ws = _w, the total volumetric water content). Rhoades et al. (1988)( 1990) determined by soil saturation experiments and sensitivity analyses that the parameters in Eq. [1] can be related to other soil properties as

[2]

[3]

[4]

[5]

[6]

where SP is saturation percentage. The results of the Rhoades et al. (1989) analysis show that the exact relationship between EC_a and soil water electrical conductivity is dependent on soil structure and clay percentage, and they suggested that similar saturation extract measurements would be appropriate when applying these equations to a specific site.

The electrical conductivity of the soil water, EC_w, is obtained from the solution of Eq. [1], with the assumption EC_ws = EC_wc:

[7]

[7a]

[7b]

[7c]

Since temperature affects the conductivity of a material, it is necessary to standardize field-measured EC_a values to an equivalent electrical conductivity at a reference temperature, typically 25°C. Sheets and Hendrickx (1995) gave the following temperature standardization equation:

[8]

where EC₂₅ is the standardized EC_a and T is the soil temperature (°C).

Rhoades et al. (1989) acknowledged that the assumption EC_ws = EC_wc is a potential source of error in the model, since EC_ws and EC_wc cannot always be assumed equal, especially under conditions where a concentrated solution is infiltrating into a soil. In most cases assuming EC_ws = EC_wc is reasonable because EC_wc plays a much more significant role in determining the overall EC_a than does EC_ws, particularly in moist agricultural soils (Rhoades et al., 1988, 1990). In a study conducted in dry soils, Lesch and Corwin (2003) suggested that to apply this model the water content should be at least 65% of field capacity, where field capacity (_fc) is defined as

[9]

As will be shown below, the model of Rhoades et al. (1989) remains applicable at low _w as long as the threshold water content (_t), defined as the cutoff between the _wc and _ws, is chosen appropriately.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Research Site
The study was conducted at the Sandia-Tech Vadose Zone test facility located adjacent to the New Mexico Institute of Mining and Technology campus in Socorro, NM. The subsurface consists of poorly consolidated, heterogeneous alluvial deposits comprised of interbedded sands, gravels, and clays exhibiting heterogeneity on scales ranging from microstructures through small-scale cross-bedding to fairly homogenous beds as thick as 7 m. Depth to water at nearby wells was about 35 m. Cuttings from 41 instrumentation boreholes and four continuous core samples located axisymmetrically around the center of the site and drilled to depths up to 12 m provided data for developing a stratigraphic model of the deposits. Four simplified stratigraphic columns of the continuous core samples are depicted in Fig. 1 , along with likely correlations among mappable geologic units. A notable feature is the seemingly continuous clay-rich layer at the 4- to 6-m depth, identified as Units 5 through 7 in Fig. 1. Electromagnetic induction and natural logs from all 41 boreholes aided in locating and verifying contacts between units of contrasting electrical conductivity, while two trenches excavated to 1.5 m on a hill slope adjacent to the site and exposures in a nearby sandpit provided additional geologic information on the character of the deposits. The additional EM logs supported the presence of one or more clay-rich layers of varying thickness at the 4- to 6-m depth across the site.

View larger version (41K): [in this window] [in a new window]   Fig. 1. Correlated geologic stratigraphic columns of the subsurface sedimentary deposits below the STVZ site (Brainard et al., 2004). Core locations are indicated in Fig. 2.

View larger version (45K): [in this window] [in a new window]   Fig. 2. Site layout, with the arrays of the infiltrometer numbered 1 through 9. Arrays 2, 6, and 9 received NaCl solution, while the remaining six arrays continued to receive tap water, all at a flux of 2.7 cm d–1.

Water Application
The infiltrometer was divided into nine 1 by 1 m arrays, numbered 1 through 9 in Fig. 2. Each array incorporated 100 18-gauge stainless-steel syringe needles as emitters, installed on a 0.1-m grid. Flow to each array was controlled by a pressure regulator and monitored with a flow meter. Well water was piped to three 909-L (200 gal) storage tanks in a nearby insulated data acquisition building. Infiltration events were timer-controlled, and flow rates and tank levels were monitored with flow meters and pressure transducers equipped with data loggers. Infiltration with untreated municipal drinking water (average conductivity 80 mS m^–1) began 11 Mar. 1999 (Day 0). Five-minute irrigation events occurred every 12 h, resulting in an average flux of 2.7 cm d^–1 during the course of the experiment. At this flux the subsurface remained unsaturated, as verified by tensiometers installed at the site. After 1095 d, a 6.9 g L^–1 NaCl solution (1300 mS m^–1) was added to one of the storage tanks and applied to infiltrometer arrays 2, 6, and 9, while the other arrays continued to receive untreated water and the same flux of 2.7 cm d^–1 was maintained at all arrays. The salt pulse continued for 85 d (until Day 1180), after which untreated water was supplied to all nine arrays for an additional 57 d (Day 1237). Infiltration ceased on Day 1237 and subsurface monitoring continued for 21 d as the site drained.

Collection of Bulk Soil Electrical Conductivities
Bulk soil electrical conductivity data were collected using an EM39 borehole conductivity probe. The EM39 probe is most sensitive to material at radial distances of 0.05 to 0.3 m and a vertical distance within 0.3 m of the probe center, resulting in an effective measurement volume of about 0.1 m³. Before calibration and data collection, the probe was suspended at the bottom of the easternmost borehole for temperature equilibration. The probe was then lowered with an electric winch down each of the 13 boreholes and a conductivity reading taken every 0.10 m. The velocity of the probe was 0.100 m s^–1, with a ±0.010 m s^–1 error due to variability in the speed of the winch motor. Complete data sets for all 13 boreholes were obtained in 3 h, including the 1-h temperature equilibration. Data were obtained weekly during the early stages of the salt pulse, and then biweekly as the conductivity profiles approached steady state. Data collection began 3 d before the salt application and continued for 21 d after the end of infiltration (Day 1258). The repeatability of measurements taken in the same location at the beginning and end of each data set was within the reported instrument precision of 0.5 mS m^–1 (McNeill et al., 1990).

Volumetric water content was measured monthly in all access tubes at 0.25-m intervals using a CPN 503DR neutron probe (Campbell Pacific Nuclear, Martinez, CA). Temperature readings from the thermocouples were logged hourly.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
EM39 Measurements
Figure 3 shows the raw EM-39 data collected in the center borehole (directly below the infiltrometer) during the salt pulse. The conductivity increased to a depth of 8 m during the 81-d salt application and showed peaks in conductivity between 4 and 6 m, corresponding to the finer grained layers suggested by the core data of Fig. 1. The raw EM-39 data from the other 12 boreholes showed similar trends but with lower absolute conductivities as the lateral distance from the infiltrometer increased.

View larger version (23K): [in this window] [in a new window]   Fig. 3. EM39 measurements in the center borehole during the salt pulse application. A measurement made before any water application to the site is labeled "pre."

The clay percentage in the soil was estimated by particle size analyses of samples from the continuous core sample collected in the northwest quadrant (Fig. 1 and 2). The soil properties of the remaining three cores were assumed similar, and the distribution of each unit determined by stratigraphy. Volumetric water contents were estimated from the monthly neutron probe readings.

Interpretation of the Three-Dimensional Images
Figure 4 is a representative image of the low water content conditions from neutron probe readings observed during the salt pulse. The shape and extent of the wetted area closely followed the stratigraphy, with the wettest areas directly below the infiltrometer and across the site at the 4- to 6-m depth corresponding to the clay-rich layer. The shape of the wetted area remained stable during and following the salt pulse, as was expected since infiltration had been occurring for almost 1100 d at the same flux before the start of the pulse. The maximum standard deviation of water content measurements from the center borehole during the year before salt pulse application was 0.65%. Neutron probe and EM images showed a decrease in water content and electrical conductivity between the near-surface zone and the clay layer at 4- to 6-m depth. The low water contents and conductivities were most likely due to the coarser deposits that dominated this interval.

View larger version (65K): [in this window] [in a new window]   Fig. 4. Interpolated three-dimensional image of neutron probe data collected 17 d before the salt pulse began.

View larger version (160K): [in this window] [in a new window]   Fig. 5. Time series of difference images during the salt pulse application showing the difference in measured ECa (mS m–1) relative to 3 d before salt application. The number of days following salt pulse initiation is indicated in the top right corner or each image.

View larger version (75K): [in this window] [in a new window]   Fig. 6. Time series of difference images during the salt-free pulse application showing the difference in measured ECa (mS m–1) relative to 3 d before salt application. The number of days following initiation of the salt-free pulse is indicated in the top right corner of each image.

View larger version (38K): [in this window] [in a new window]   Fig. 7. Time series of difference images during drainage showing the difference in measured ECa (mS m–1) relative to 3 d before pulse application. The number of days following the end of infiltration is indicated in the top right corner or each image.

From the series of EM and _w images shown in Fig. 5 through 7, it appears that the infiltrated water and salt moved downward from the infiltrometer until spreading at the clay layer at approximately the 5-m depth. The water then appeared to leave the instrumented area, most likely along the clay layer in regions that were not well imaged by either the EM or neutron probe because of their limited measurement volumes and the sparsity of boreholes.

Mass Balance Calculations
By applying the Rhoades et al. (1989) model to the interpolated data, EC_w was calculated for each cell in the three-dimensional grid. The calculated EC_w was first converted to a concentration of NaCl and then finally to a mass of NaCl using the volume of water for that cell. The total mass of salt in the soil water was calculated by summing the mass of salt in each of the cells in the three-dimensional volume. This calculated mass was then compared with the known cumulative mass of salt infiltrated at the surface.

The results of sensitivity analyses correcting for temperature effects or varying the estimated clay percentage had relatively little effect on the calculated mass of salt in the profile (data not presented; see Hall, 2003, for details). In contrast, the _t had a strong effect on the calculated mass. We performed mass balance calculations assuming _t values of 4, 6.5, 8, and 10%. The result of varying _t is shown in Fig. 8 for temperature-corrected conductivity data, where it is clear that as _t decreases (i.e., more water is considered mobile) the calculated total mass of salt increases dramatically. In fact, the mass calculated with 4% threshold water content is two orders of magnitude larger than the mass calculated with 10% threshold water content.

View larger version (22K): [in this window] [in a new window]   Fig. 8. Change in mass from 3 d before the start of the salt pulse using temperature-corrected data for 4, 6.5, 8, and 10% threshold water contents.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The EM39 proved to be a fast and reliable method for measuring soil salinity and water content changes in the vadose zone. The stability of the shape of the wetted area (as measured with the neutron probe) increased our confidence that the changes in conductivity measured with the EM probe were due to the movement of the salt pulse. Water content images derived from the neutron probe data showed that infiltrated water moved directly down from the infiltrometer until spreading laterally at about the 5-m depth. Difference images of EM readings, while clearly showing the growth of the salt plume, did not highlight any specific pathways to suggest how the water was exiting the study volume. The trend of increasing salt mass in the vadose zone with increased salt infiltration, however, suggests that the EM39 was accurately capturing the majority of the salt plume.

Due to the very low water contents measured at the test site, the mass calculations were strongly dependent on _t, and much less sensitive to estimates of clay percentages. The chosen value of _t is critical, since calculation of EC_w is not possible if all of the soil water is assumed immobile (see Eq. [7]). Because of this sensitivity of the model, it is usually only recommended for use with soils at or near field capacity. While the soils at our test site were only 35% of field capacity, data from laboratory soil analyses and the large number of water content measurements made in the same location as EM39 measurements allowed a close approximation of the actual threshold water content. The laboratory analyses of soil samples showed that the threshold water content for most samples, and hence for the site in general, was near 6.5%, the value that provided an almost exact match with the known infiltrated salt mass with time.

The results of this study suggest that the downhole EM39, combined with downhole neutron probe measurements, can accurately determine soil water salinity at much lower water contents than previous studies using EM techniques.

	ACKNOWLEDGMENTS

 
The authors thank Doug LaBrecque and Melissa Stubben at MultiPhase Technologies for help with funding and with the interpolation in TecPlot, as well as the numerous people who helped collect data in the field. Support for this work was provided by the USDOE under the Environmental Management Science Program, Project 70267. Sandia National Laboratories is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the USDOE under Contract DE-Ac04-94AL85000.

	REFERENCES

TOP ABSTRACT INTRODUCTION THEORY MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Baker, K.E. 2001. Investigation of direct and indirect hydraulic property laboratory characterization methods for heterogeneous alluvial deposits: Application to the Sandia-Tech Vadose Zone Infiltration Test Site. M.S. thesis. New Mexico Institute of Mining and Technology, Socorro.
• Brainard, J.R., R.J. Glass, D.L. Alumbaugh, L. Paprocki, D.J. LaBrecque, X. Yang, T.C.J. Yeh, K.E. Baker, and C.A. Rautman. 2004. The Sandia-Tech Vadose Zone Facility; experimental design and data report of a constant flux infiltration experiment. Sandia Rep. Sandia Natl. Lab., Albuquerque, NM. (in press).
• Davis, J.C. 1986. Statistics and data analysis in geology. 2nd ed. John Wiley & Sons, New York.
• Hall, L.M. 2003. Measuring salinity changes in the vadose zone using downhole electromagnetic induction. M.S. thesis. New Mexico Institute of Mining and Technology, Socorro.
• Hendrickx, J.M.H., B. Borchers, D.L. Corwin, S.M. Lesch, A.C. Hilgendorf, and J. Schlue. 2002. Inversion of soil conductivity profiles from electromagnetic induction measurements: Theory and experimental verification. Soil Sci. Soc. Am. J. 66:673–685.[Abstract/Free Full Text]
• Hendrickx, J.M.H., and R.G. Kachanoski. 2002. Nonintrusive electromagnetic induction. p. 1297–1306. In J.H. Dane and G.C. Topp (ed.) Methods of soil analysis. Part 4. SSSA Book Ser. 5. SSSA, Madison, WI.
• Lesch, S.M., and D.L. Corwin. 2003. Predicting EM/soil property correlation estimates via the Dual Pathway Parallel Conductance Model. Agron. J. 95:365–379.[Abstract/Free Full Text]
• McNeill, J.D. 1980a. Electrical conductivity of soils and rocks. Tech. Note TN-5. Geonics Ltd., Mississauga, ON, Canada.
• McNeill, J.D. 1980b. Electromagnetic terrain conductivity measurement at low induction numbers. Tech. Note TN-6. Geonics Ltd., Mississauga, ON, Canada.
• McNeill, J.D., M. Bosnar, and F. Snelgrove. 1990. Resolution of an electromagnetic borehole conductivity logger for geotechnical and ground water applications. Tech. Note TN-25. Geonics Ltd., Mississauga, ON, Canada.
• 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.[Abstract/Free Full Text]
• Rhoades, J.D., P.A.C. Raats, 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., P.J. Shouse, W.J. Alves, N.A. Manteghi, and S.M. Lesch. 1990. Determining soil salinity from electrical conductivity using different models and estimates. Soil Sci. Soc. Am. J. 54:46–54.[Abstract/Free Full Text]
• Rhoades, J.D., B.L. Waggoner, P.J. Shouse, and W.J. Alves. 1988. Determining soil salinity from soil and soil-paste electrical conductivities: Sensitivity analysis of models. Soil Sci. Soc. Am. J. 53:1368–1374.
• Sheets, K.R., and J.M.H. Hendrickx. 1995. Noninvasive soil water content measurement using electromagnetic induction. Water Resour. Res. 31:2401–2409.

This article has been cited by other articles:

				H. Vereecken, S. Hubbard, A. Binley, and T. Ferre Hydrogeophysics: An Introduction from the Guest Editors Vadose Zone J., November 1, 2004; 3(4): 1060 - 1062. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Hall, L. M. Articles by Hendrickx, J. M. H. Search for Related Content PubMed Articles by Hall, L. M. Articles by Hendrickx, J. M. H. GeoRef GeoRef Citation Agricola Articles by Hall, L. M. Articles by Hendrickx, J. M. H. Related Collections Vadose Zone Processes and Chemical Transport Field-Scale Studies Electromagnetic Induction, EMI