PASSIVE ELECTRICAL MONITORING OF AEROBIC AND ANAEROBIC PROCESSES USING SEPTIC SYSTEMS AS AN ANALOG. Abstract. Introduction

Transcription

1 PASSIVE ELECTRICAL MONITORING OF AEROBIC AND ANAEROBIC PROCESSES USING SEPTIC SYSTEMS AS AN ANALOG Jeffrey G. Paine, Bureau of Economic Geology, The University of Texas at Austin Abstract Using anaerobic and aerobic septic systems as analogs, we examined whether the passive electrical potential (SP) method can be used to detect and monitor shallow biochemical reactions. High-resolution SP surveys conducted with nonpolarizing electrodes can reliably detect potential differences as small as a few mv over relatively small areas, suggesting that the technique can find wider use in monitoring biodegradation processes, mapping effluent plumes, and identifying local preferential infiltration areas. Surface SP measurements appear to respond to changes in the oxidation-reduction potential in shallow reactors, correlating best with changes in dissolved oxygen concentration and measured electrical potential within the subsurface reactors. Among the available electrode choices, including those made of carbon, low-grade steel, stainless steel, and copper, only nonpolarizing electrodes such as the copper copper sulfate types provide the measurement stability necessary to achieve high-resolution results. Introduction The spontaneous, streaming, or self-potential (SP) method has enjoyed wide use, particularly in minerals exploration, engineering studies, soil studies, pipeline and other types of corrosion monitoring, dam-seepage investigations, and earthquake prediction and monitoring (e.g. Ogilvy and Bogoslovsky, 1969; Fitterman, 1976; Corwin and Morrison, 1977; Corwin, 1989; Corwin, 1990a, b; Jansen and others, 1996). We investigate using this method at the local scale (high spatial resolution and small differences in electrical potential) to monitor shallow aerobic and anaerobic reactions, using two types of residential septic systems as representative test cases. Septic systems have been studied using active-source electrical methods (EM and resistivity; Amidu and Olayinka, 2006; Lee and others, 2006), but the passive method offers the possible advantages of the ability to monitor near-surface chemical reactions and infiltration patterns. If the passive electrical potential method can be successful in these applications, where measured voltages can be expected to be lower than those observed in more regional studies where signals can reach hundreds of millivolts (mv), then the method can be extended to other possible local applications where small signals (a few to a few tens of mv) and lateral variability are expected, such as infiltration studies, monitoring biogeochemical reactions, and perhaps monitoring gas or fluid migration through near-surface deposits. Methods We modified the passive electrical potential (SP) method (Parasnis, 1986; Corwin, 1990a, b; Telford and others, 1990) to examine whether it can be used to monitor shallow aerobic and anaerobic reactions at the local scale. Standard equipment includes a high-quality (high impedance) voltmeter, insulated copper wire, and two soil-insertable, non-polarizing copper copper sulfate electrodes to enable measurements of electrical potentials between a stationary reference electrode and a roving electrode. In contrast with typical large-area studies, we employed a single reference electrode location and used roving electrode spacings of 0.5 m for individual profiles across features of interest or 1 m in a gridded survey measuring 8 16 m over a septic drain field. We prepared shallow holes for electrode insertion by pushing a pointed metal rod a few cm into the soil to penetrate matted vegetation. Voltage measurements were taken within a few seconds of electrode 255

2 insertion, typically after the voltmeter reading stabilized. Repeat profiles and grids were conducted using the same holes at intervals ranging from hours to days after the initial surveys. We measured apparent ground conductivity on the drain-field grid using a Geonics EM38 ground conductivity meter, a frequency-domain electromagnetic induction instrument that operates at a transmitter frequency of 14.6 khz. We acquired four conductivity measurements at each location. Orthogonal horizontal- and vertical-dipole measurements were averaged to produce single horizontal- and vertical-dipole apparent conductivity values. Nominal exploration depth for this instrument is less than one meter in the horizontal-dipole orientation and less than two meters in the vertical-dipole orientation, adequate to examine shallow soil properties in a typical drain field. We used a Quanta Hydrolab to determine relevant water properties such as dissolved oxygen (concentration and saturation percent), ph, specific conductivity, and temperature for comparison with ground electrode measurements made over the aerobic septic system. ph and specific conductivity were calibrated with standard solutions. Dissolved-oxygen concentration was calibrated using the saturated-air method and known atmospheric pressure. Electrode Suitability Testing We tested a variety of electrodes (Figure 1) to examine the effects of electrode type, insertion depth, and the variation in the measured voltage with time after insertion. We found that rebar (low-grade steel), copper, and carbon rods all exhibit strong early- and late-time variation in measured potential due to polarization effects, soil electrode reactions, and accumulation of static charge. Stainless steel and carbon rods are less reactive with soil and might be considered usable in passive potential surveys. Stainless rods drift less than rebar and copper rods, but are not as stable as nonpolarizing electrodes such as the preferred Figure 1: Electrodes evaluated for passive soil electrical potential monitoring. Types include rods of rebar (low-grade steel), stainless steel, and machined carbon of varying lengths and burialdepth exposures (left), and portable, non-polarizing copper copper sulfate electrodes such as those manufactured by Tinker and Rasor (right). 256

3 copper copper sulfate electrodes in which a copper rod is bathed in a saturated copper sulfate solution that is allowed to come into contact with the ground through a permeable tip. Carbon electrodes receive a significant charge at insertion that can require hours to dissipate. All experiments described below were conducted with nonpolarizing copper copper sulfate electrodes. Profile Over Conventional (Anaerobic) Septic Chamber At the Coyote Ridge test site in Hays County, Texas, a conventional anaerobic septic system has been in continuous operation since This system consists of a dual-chamber reaction tank that supplies a leach field that is approximately 100 m 2. An 11-m-long profile across the two tanks (Figure 2) shows two distinct voltage highs centered over the two separate tanks. The greater of the two anomalies (measuring about 10 mv) overlies the larger tank, where most of the anaerobic decay of waste matter occurs. The profiles were measured twice; once immediately following hole preparation, and once a day later. The two profiles are similar; potentials measured during the later survey are consistently about 2 mv higher. The low between the two peaks occurs between the separate tanks. Profile Over Aerobic Septic Chamber An example of the newer, aerobic style of residential waste treatment system was studied at the Westview Trail test site, also in Hays County, Texas about 1 km from the Coyote Ridge site. This system has been in operation for about six years. Our tests at this site included measuring passive electrical potentials along an 11-m-long profile centered on the aerobic reaction chamber. This profile was occupied repeatedly Figure 2: Passive voltage measurements along an 11-m-long profile across a shallowly buried, dualchamber anaerobic septic tank. Hour 0 measurements taken within minutes of hole preparation. A second set of measurements were taken in the same holes 25 hours later. 257

4 Figure 3: Repeated passive voltage measurements acquired along an 11-m-long profile across a shallowly buried, single-chamber aerobic (injected air) septic tank. The lower set of profiles was acquired at times ranging from 20 minutes before the supplied air was shut off to 72 hours after the air was shut off. The upper set of profiles was acquired at times between about 40 minutes before the air supply was restored to 62 hours after air was restored. over an 8-day period, beginning with a transect acquired during normal operating conditions. We then shut off the air supply and conducted repeat SP surveys (and measured select reactor water parameters) with no forced air, intending to examine whether we could use the SP method to detect the biochemical change occurring in the reactor as the oxygen supply diminished. We reintroduced injected air on the sixth day, continuing the SP and water parameter measurements as the system readjusted to additional oxygen supply. We completed the measurements 62 hours after the air supply was restored. Repeat SP profiles (Figure 3) and comparisons of water and geophysical measurements over time (Figure 4) suggest that change has been detected over the septic system and that this change is likely caused by chemical changes occurring in the reactor. The initial SP profile, acquired less than an hour before the air supply was removed, shows a 20-mV variation across the profile with a small, 10 mv peak over the 2.5-m long segment directly over the chamber. That peak diminished steadily for a period of about 26 hr, lowering a total of about 15 mv over that time. Voltages recovered about half of that amount over the following 46 hours, and continued to increase over a broad area centered on the reactor until the air injection experiment began at about 130 hr after the air supply was removed. Over the 62 hours of air injection (the upper group of lines in Figure 3), potentials climbed to maximum values of more than 40 mv, measured 38 hr after the air injection resumed. The greatest change was again observed over a nearly 3-m-long segment centered on the reaction chamber. Measured potentials subsequently decreased toward the initial survey values for the remainder of the experiment. 258

5 Figure 4: Measurements over time (after air shutoff) of (a) ph and (b) dissolved oxygen in the aerobic chamber, (c) voltage between the chamber fluid and a ground reference electrode, and (d) voltage between a reference electrode and ground electrodes in soil directly above the chamber and away from the chamber. 259

6 Water-property measurements made just before and during the first day following air shutoff show a decrease in ph and very low dissolved oxygen content (Figure 4a and b). Low initial dissolved-oxygen concentrations suggest that the system was operating near the anaerobic/aerobic threshhold at the onset of the experiment. During the first 26 hr after air shutoff, ground electrodes directly over the chamber exhibited significant voltage drops (about 15 mv), while electrodes most distant from the reactor (5.5 m away) showed no sustained voltage drop (Figure 4d). ph values in the reactor remained near 7.0 until the air supply was restored at 130 hr after the experiment began, at which time ph rose rapidly to about 7.7 (Figure 4a). Dissolved oxygen remained low for more than 20 hr before rising to 3 to 6 percent of saturated values. Coincident with the rise in dissolved oxygen concentration is a rise in voltage measured between a ground reference electrode and both copper and carbon electrodes immersed in the reactor (Figure 4c). Ground electrodes directly over and most distant from the reactor again show differing trends (Figure 4d). Both show a small increase during the early rise in ph (probably related to changing conditions at the reference electrode), but only the voltage at the electrode over the reactor rises during the rise in potential measured in the tank. This suggests that ph changes had little effect on nearby ground potentials, but that changes in dissolved oxygen (and oxidation-reduction potentials in the reactor) were detected at the surface. Imaging Drain Fields Another setting where the SP method can be tested is in the drain fields of a conventional anaerobic septic system, where the method might be expected to detect such things as preferred infiltration points, extent of effluent plume migration, and aerobic and continued anaerobic reactions consuming the effluent as it migrates through the drain field. We returned to the Coyote Ridge test site to examine the utility of both SP and conductivity measurements at close spacings over the drain field. Horizontal- and vertical-dipole conductivities measured with the shallow-exploring Geonics EM38 are relatively low (9 to 25 ms/m horizontal dipole and 4 to 14 ms/m vertical dipole, Figure 5a and b), consistent with thin and relatively dry clay to clay loam soil over shallow limestone bedrock. Lower conductivities are observed over the drain field in the shallower-exploring horizontal dipole orientation, likely indicating dry and relatively sandy fill material at shallow depths. The conductivities measured in the deeper-exploring vertical dipole mode are generally lower still, but do not clearly reveal patterns suggesting the presence of an effluent plume. In contrast, SP measurements made over the same area at the same time (Figure 6a) do show voltage variations similar in magnitude to those measured over the septic chambers. Relatively high voltages exist over a narrow band near the outlet, broadening toward the left in a pattern suggestive of an effluent plume. Highest voltages were measured near the left edge of the image near the boundary of the drainfield excavation. Three repeat surveys conducted the following day showed a similar pattern. All surveys but the last were made under no-flow conditions. The final survey was completed while introducing fresh water flowing at a rate of 15 l/m into the reactor tank, which forced effluent to flow into the drain field. The image produced from those measurements show relatively little change farthest from the outlet (at the left of the image) and greatest change nearest the outlet (Figure 6b). Conclusions As demonstrated by measurements over aerobic and anaerobic septic systems, passive electrical potential measurements show promise for high-resolution, long-term monitoring of shallow aerobic and anaerobic biochemical processes. This method appears to be more effective than conductivity methods for delineating effluent plumes. By careful attention to choice of electrodes, electrode insertion, and consistency in hole preparation, we were able to achieve reproducible results over voltages ranging from a few to a 260

7 (a) (b) Figure 5: Apparent ground conductivity measured at 144 locations over a drain field supplied by a conventional anaerobic septic system. Two orthogonal measurements were taken at each site using a Geonics EM38 ground conductivity meter in (a) horizontal and (b) vertical dipole orientation. Gridded values are the average of the two orthogonal measurements for each dipole orientation. Effluent flows into the drain field from the outlet at right. 261

8 (a) (b) Figure 6: Passive voltage measured at 144 locations over a drain field supplied by a conventional anaerobic septic system. Image (a) was obtained at steady state (no inflow); image (b) was obtained while supplying the septic system with water at a rate of about 15 l/min. 262

9 few tens of mv. Lateral variability in measurements over relatively small distances suggests that spatial aliasing would prevent using the SP method effectively over large areas unless the anticipated voltages are much higher than a few tens of mv. References Amidu, S. A., and Olayinka, A. I., 2006, Environmental assessment of sewage disposal systems using 2D electrical-resistivity imaging and geochemical analysis: a case study from Ibadan, Southwestern Nigeria: Environmental and Engineering Geoscience, v. 12, no. 3, p Corwin, R. F., 1989, Data quality for engineering self-potential surveys: in Merkler, Georg-Paul, Militzer, Heinz, Hoetzl, Heinz, Armbruster, Heinrich, and Brauns, Josef, editors, Detection of subsurface flow phenomena by self-potential/geoelectrical and thermometric methods, Springer-Verlag, Berlin, Lecture Notes in Earth Sciences, 1989, Vol. 27, pp Corwin, R. F., 1990a, The self-potential method for environmental and engineering applications: Ward, S. H., editor, Geotechnical and environmental geophysics, Soc. Explor. Geophys., Investigations in Geophysics No. 5, Tulsa, OK, volume I, p Corwin, R. F., 1990b, Applications of the self-potential method for engineering and environmental investigations: Proceedings of the Symposium on the Application of Geophysics to Engineering and Environmental Problems (SAGEEP), EEGS, Englewood, CO: Corwin, R. F., and Morrison, H. F., 1977, Self-potential variations preceding earthquakes in Central California: Geophys. Research Letters, v. 4, no. 4, p Fitterman, David V., 1976, Calculation of self-potential anomalies generated by Eh potential gradients: U.S. Dept. of the Interior, Geological Survey, Open-file report no , 32 p. Frischknecht, F. C., Labson, V. F., Spies, B. R., and Anderson, W. L., 1991, Profiling using small sources: in Nabighian, M. N., editor, Electromagnetic methods in applied geophysics applications, part A and part B: Society of Exploration Geophysicists, Tulsa, p Jansen, John, Billington, Ned, Snider, Fred, and Jurcek, Patrick, 1996, Marine SP surveys for dam seepage investigations: evaluation of array geometries through modeling and field trials: Journal of Environmental and Engineering Geophysics, v. 1, no. 1, p Lee, Brad D., Jenkinson, Byron J., Doolittle, James A., Taylor, Richard S., Tuttle, J. Wes, 2006, Electrical conductivity of a failed septic system soil absorption field, Vadose Zone Journal, v. 5, p McNeill, J. D., 1980a, Electrical conductivity of soils and rocks: Geonics Ltd., Mississauga, Ont., Technical Note TN-5, 22 p. McNeill, J. D., 1980b, Electromagnetic terrain conductivity measurement at low induction numbers: Geonics Ltd., Mississauga, Ont., Technical Note TN-6, 15 p. Ogilvy, A. A., Ayed, M. A., and Bogoslovsky, V. A., 1969, Geophysical studies of water leakages from reservoirs: Geophysical Prospecting, v. 17, no. 1, p Parasnis, D. S., 1986, Principles of applied geophysics (fourth edition): Chapman and Hall, London, 402 p. Telford, W. M., Geldart, L. P., and Sheriff, R. E., 1990, Applied Geophysics (second edition): Cambridge University Press, New York, 770 p. 263