Fate and transport of phosphate from an onsite wastewater system in Beaufort County, North Carolina.
Water Quality Issues Related To Excess Phosphorus
Eutrophic conditions and fish kills continue to be problematic in the Tar-Pamlico watershed in North Carolina (North Carolina Division of Water Quality, 2010). Nutrient-sensitive water management strategies were implemented in 2000 to help reduce nitrogen loads to the watershed and cap discharges of phosphorus from various point and nonpoint sources of nutrient pollution. Despite these efforts, elevated nutrient problems still exist in the river and estuary. More scrutiny is now being placed on potential sources of pollution such as onsite wastewater systems (OWS) that were not addressed in previous regulations.
OWS and Phosphorus Attenuation
OWS process wastewater containing elevated phosphorus concentrations and are used by approximately 48% of residences in the nutrient-sensitive Tar-Pamlico River basin (Pradhan, Hoover, Austin, & Devine, 2007). Therefore, it is important to determine the nutrient reduction efficiency of OWS. While some phosphorus is retained in the septic tank with the settled solids, effluent leaving the tank is still enriched with total phosphorus (5-15 mg/L) and phosphate (1.2-12.1 mg/L) (Corbett, Dillon, Burnett, & Schaefer, 2002; Humphrey & O'Driscoll, 2011; Robertson, Schiff, & Ptacek, 1998; U.S. Environmental Protection Agency, 2002). The soil beneath the drainfield trenches and down-gradient from the system may provide an environment for phosphate attenuation. In the soil beneath drainfield trenches, the dominant phosphate removal processes are mineral precipitation and adsorption (Robertson et al., 1998). Mineral precipitation occurs when phosphate combines with other elements to form solids and thereby limits transport of phosphate away from the OWS. Soil capacity to adsorb significant amounts of phosphate onto reactive surfaces such as clay minerals and iron oxide coatings diminishes over time and can be reversed (desorption), thus releasing phosphate to groundwater. When phosphate adsorption surfaces are completely occupied, phosphate leaching and transport can occur. Thus, mineral precipitation is often considered a more stable sink for phosphate reduction (Harmon, Robertson, Cherry, & Zanini, 1996).
Mineral precipitation beneath OWS is dictated by the pH and redox status of the soils and the availability of other elements such as calcium, iron, or aluminum that combine with phosphate to form various minerals (Zanini, Robertson, Ptacek, Schiff, & Mayer, 1998). For example, the mineral variscite (AlP[O.sub.4*]2[H.sub.2]O) may precipitate when soil pH is below 5.5, anaerobic conditions are present, and aluminum and phosphate are available. When aerobic conditions are present with sufficient aluminum and phosphate and soil pH is below 8, variscite may also precipitate (Robertson et al., 1998). Strengite (FeP[O.sub.4*]2[H.sub.2]O) precipitation can occur when soil or groundwater pH is between 6 and 8, aerobic conditions are present, and phosphate and iron are available (Robertson et al., 1998).
Several studies have indicated that soil texture beneath OWS influences phosphate treatment with fine-textured soils providing more phosphate reduction than coarse-textured soils. The increased surface area of silt-rich and clay-rich soils provides more opportunity for adsorption and also increases the residence time of wastewater in the unsaturated zone, enhancing phosphate attenuation. Robertson and coauthors (1998) reported in a literature review of 10 OWS in Canada that OWS installed in more fine-textured soils such as silt had much shorter phosphate plume lengths than medium- and coarse-grained sands. Karathanasis and co-authors (2006) evaluated the phosphorus removal efficiency of different soil types by leaching wastewater through soil monoliths. They found that in general, clay-textured soils provided better total phosphorus treatment than sandy soils. A recent study in coastal North Carolina (Humphrey & O'Driscoll, 2011) documented significant groundwater phosphate concentrations beneath OWS in sandy, slightly acidic soils (mean: 2.46 mg/L), but much lower groundwater phosphate concentrations beneath systems in more fine-textured sandy clay loams (mean: 0.04 mg/L).
Shallow groundwater phosphate concentrations adjacent to OWS can exceed 2 mg/L in some coastal areas (Humphrey & O'Driscoll, 2011; Ptacek, 1998; Robertson, Cherry, & Sudicky, 1991). Because phosphate concentrations two orders of magnitude lower (~0.03 mg/L phosphate) may stimulate algal blooms in some surface waters, OWS near coastal waters must be efficient at reducing phosphate transport or surface water degradation may occur (Ptacek, 1998). Research conducted in coastal sandy areas of Rhode Island (Postma, Gold, & Loomis, 1992) and in coastal Virginia (Reay, 2004) indicated minimal groundwater phosphate transport away from conventional OWS. Environmental conditions conducive to phosphate attenuation via mineral precipitation and adsorption must have been present at those sites. Research conducted at three permanent residences on St. George Island, Florida, however, indicated significant groundwater phosphate transport in the sandy surficial aquifer more than 30 m from three different types of OWS (a conventional, an aerobic, and an elevated OWS) (Corbett et al., 2002). Humphrey and O'Driscoll (2011) reported elevated groundwater phosphate concentrations beneath OWS in coastal sandy soils of North Carolina. Therefore, potential exists in some environments for significant phosphate transport down-gradient from OWS. Many coastal North Carolina water resources are impaired because of excess nutrients, so more research is needed to evaluate phosphate transport from OWS in these settings.
Our study objectives were to evaluate the fate and transport of OWS-derived phosphate from a residential system in Beaufort County, North Carolina, and to determine if current OWS setback regulations are sufficient to prevent elevated phosphate discharge to surface waters.
Site and System Location
A residential site in coastal Beaufort County, North Carolina, was chosen for our study because it bordered the nutrient-sensitive waters of the Tar-Pamlico estuary (Figure 1). The system components (3,780 L tank, distribution box, and 3-15 m long drainfield trenches) were located on site using the OWS permit information and a tile drain probe rod. The system was installed in the early 1980s and was utilized by two people during our study. Groundwater flow direction and the orientation of the septic plumes were estimated using electrical resistivity surveying and three-point contouring, with some data from exploratory groundwater piezometers (Heath, 1998; Humphrey, Deal, O'Driscoll, & Lindbo, 2010). Because wastewater contains elevated concentrations of salts relative to fresh groundwater, groundwater affected by wastewater exhibits elevated electrical conductivity. Resistivity surveys allow production of 3-D images that reveal variations in the conductive properties of the subsurface and thus help delineate wastewater plumes (Humphrey et al., 2010). The orientation of the plume image and the three-point water level contouring method (Heath, 1998) were used to determine the groundwater flow direction and to decide the location of groundwater and soil water monitoring points.
The onsite system was installed in soils similar in characteristics to the Seabrook (Mixed, thermic Aquic Udipsamments) and Tarboro (Mixed, thermic Typic Udipsamments) soil series. Seabrook soils are characterized as very sandy and moderately well drained with rapid permeability (15-50 cm/hour), low cation exchange capacity (<3 cmol/kg), and acidic conditions (pH between 4.5 and 6.5) (U.S. Department of Agriculture [USDA], 1995). Tarboro series are somewhat excessively drained, coarse-textured soils with rapid permeability (15-50 cm/hour) and acidic conditions (pH 5.1-6.5) (USDA, 1995).
Groundwater and Soil Water Monitoring Network
Multidepth, nested piezometers constructed of 5-10 cm diameter PVC pipe with 60 cm screen intervals were installed between drainfield trenches, and up- and down-gradient of the OWS flow paths (Figure 2). Hand augers were used to create boreholes to the desired depth, typically 1 m or so below the water table for the "deep" nested piezometer, and 0.4 m below the water table for the "shallow" nested piezometer. Piezometers were cut to the appropriate length and driven into boreholes. Sand was poured into the annular space around each piezometer until the screened section was covered. The remaining annular space was filled all the way to the surface with a mixture of sand and bentonite. Soil samples at depths beneath the drainfield trenches were collected and sent to the North Carolina Agronomic Services Division for descriptive analysis including pH and effective cation exchange capacity. Water table depths were measured manually every two months from the piezometers (five times overall) with a temperature level and conductivity (TLC) meter. The TLC meter consists of a water sensor at the end of a tape measure connected to a reel that allows for quick, manual determination of water table depths. The meter was calibrated before each use. Lysimeters were installed in nests in the unsaturated zone (above water table) at 90, 120, and 150 cm below the surface near background and drainfield piezometers (Figure 2). Lysimeters were used to collect soil water samples.
Septic tank, lysimeters, piezometers, and the estuary were sampled every two months during February-October 2011 (five times). A hand vacuum pump was used to pull soil water from the tension lysimeters and then into sample bottles. A new disposable bailer was used to collect groundwater samples from each piezometer. Septic tank samples were collected using a peristaltic pump connected to rigid tubing inserted into the clear zone of the tank through access manholes. A sample bottle was lowered in the estuary to collect the surface water samples. Groundwater, estuary, and septic tank samples were analyzed for pH, electrical conductivity, dissolved oxygen (DO), and temperature using a multimeter. The multimeter allowed for quick field analysis of most environmental parameters and was calibrated before each sampling event. Water and septic tank samples were kept on ice in a cooler and transported to the East Carolina University Central Environmental Laboratory (CEL) within 12 hours for phosphate analyses. Graduate students assisted with sample collection, transport, and laboratory analysis to maintain the chain of custody. Water samples were filtered at the CEL. Samples were analyzed for phosphate using the U.S. Environmental Protection Agency-approved Smart Chem 200 method.
Statistical Comparison Groups
Septic tank effluent phosphate concentrations were compared to groundwater phosphate between the OWS trenches to assess the effectiveness of OWS in reducing phosphate concentrations before discharge to groundwater. Groundwater phosphate concentrations between the drainfield trenches were compared to background groundwater to evaluate the effects of OWS on shallow groundwater. Soil water phosphate concentrations from background lysimeters were compared to samples collected between drainfield trenches to determine OWS influence on soil water. North Carolina regulations dictate that OWS must be located at least 15-30 m+ from surface waters depending on the surface water classification (15A NCAC 18A .1950d). Down-gradient groundwater phosphate concentrations [less than or equal to] 15 m from the OWS were compared to those between 15 and 30 m. Down-gradient groundwater samples 30-35 m from the system were compared to groundwater samples collected at 35-40 m+ and the estuary. Mann-Whitney tests (Davis, 2002) were performed using Minitab 16 statistical software to determine if significant differences in phosphate concentrations existed between comparison groups. A linear regression of groundwater phosphate concentrations plotted against distance down-gradient from the OWS was used to determine the relationship between setback distance and groundwater phosphate concentrations.
Physical and Chemical Parameters of Water
Groundwater levels beneath the OWS fluctuated by nearly 1 m during our study. The lowest level observed during sampling was in early August 2011 (1.98 m below surface), while the highest water levels occurred in October 2011 (0.98 m below surface). Groundwater flow direction was predominantly toward the south east (1p4 towards 1p16) (Figure 1). Groundwater samples collected from piezometers 1p4 to 1p10, 1p13, and 1p16 were considered to be within the OWS plume and were used to determine concentration reductions with distance from the system.
Mean water sample temperatures, electrical conductivity, and pH levels were highest for septic effluent (20.4 [+ or -] 5.9[degrees]C, 1,314 [+ or -] 95 [micro]S/cm, and 7.3 [+ or -] 0.3), and were elevated in groundwater between drainfield trenches (17.1 [+ or -] 3.8[degrees]C, 809 [+ or -] 213 [micro]S/cm, and 6.6 [+ or -] 0.4) relative to adjacent background groundwater values (16.4 [+ or -] 4.9[degrees]C, 86 [+ or -] 213 [micro]S/cm, and 6.3 [+ or -] 0.8). These data confirm that wastewater was influencing the physical and chemical properties of groundwater adjacent to the OWS. The highest mean DO concentrations (3.3 [+ or -] 1.1 mg/L) were in the background groundwater piezometers. Mean DO concentrations beneath the drainfield trenches were 3.2 [+ or -] 0.4 mg/L, indicating aerobic conditions. Septic effluent had the lowest DO levels (0.3 [+ or -] 0.1 mg/L). The soil pH beneath the drainfield trenches (6.9) was lower than septic effluent pH (7.3), but higher than groundwater between disposal field trenches (6.7), indicating the influence of wastewater on soil chemical properties.
Average septic tank effluent phosphate concentrations (2.97 [+ or -] 0.76 mg/L) were within the range of effluent concentrations reported in other studies (Corbett et al., 2002; Humphrey & O'Driscoll, 2011; Robertson et al., 1998). Groundwater phosphate concentrations between the drainfield trenches (3.05 [+ or -] 0.74 mg/L) were not significantly different at p < .05 than septic effluent concentrations and were elevated relative to background conditions (0.14 [+ or -] 0.12 mg/L) (Figure 3). Soil water phosphate concentrations in background lysimeters (0.02 [+ or -] 0.01 mg/L) were significantly lower than soil water phosphate concentrations between drainfield trenches (1.27 [+ or -] 0.58 mg/L) as shown in Figure 3. Groundwater phosphate concentrations within the plume (1p4 towards 1p16) typically decreased with distance from the OWS. For example, groundwater within 15 m of the OWS had mean phosphate concentrations of 2.22 [+ or -] 0.85 mg/L, while mean groundwater phosphate between 15 and 30 m from the system was 1.19 [+ or -] 0.76 mg/L; groundwater between 30 and 35 m from the system contained 0.84 [+ or -] 0.59 mg/L phosphate; and groundwater more than 35 m from the system had 0.05 [+ or -] 0.05 mg/L phosphate (Figure 3). The estuary had mean phosphate concentrations of 0.06 [+ or -] 0.08 mg/L (Figure 3).
Groundwater phosphate concentrations showed an inverse relation ([r.sup.2] = .83) to distance from the OWS (Figure 4). Based on the regression equation (y = -0.0618x + 2.5538) and mean background phosphate concentration (0.14 mg/L), a 39 m setback distance would be required to reduce OWS-derived phosphate concentrations to background levels.
Onsite System Phosphate Attenuation
The dominant phosphate removal processes in soils beneath OWS are mineral precipitation and adsorption (Robertson et al., 1998). Precipitation of the minerals strengite or variscite would be possible at this site during the study period given the near-neutral pH of soil (6.9) and groundwater beneath the drainfield (6.7) as well as aerobic conditions (DO: mean 3.2 mg/L) (Robertson et al., 1998). Vivianite ([Fe.sub.3][[P[O.sub.4]].sub.2] x 8[H.sub.2]O) is another iron-based mineral that could precipitate at this site. Given the soil and groundwater pH, vivianite formation would be possible if anaerobic conditions occurred between the drainfield and estuary and if phosphate and iron were available (Ptacek, 1998). Mean phosphate concentrations in groundwater beneath the OWS were not significantly different, however, than septic effluent and groundwater phosphate concentrations more than 30 m down-gradient from the system and were still elevated relative to background levels. These data indicate that phosphate attenuation processes were not sufficient to prevent groundwater enrichment of phosphate at distances greater than the minimum setbacks from OWS to surface waters (15-30 m depending on water classification).
The OWS in our study was installed in sandy, permeable, low-reactive surface area Tarboro and Seabrook soil series. The OWS had been in use for more than 25 years. It is possible that the phosphate adsorptive capacity of the sandy soils was exceeded over the past few decades allowing for significant phosphate transport more than 30 m down-gradient. It is also possible that phosphate minerals did not form because not enough iron was available for precipitation of variscite, strengite, or vivianite. The Seabrook soils at the research site were located near the front of the drainfield trenches and between the OWS and the estuary. The Seabrook series soils in Beaufort County, North Carolina, typically have a light gray color (chroma less than 2) starting at depths 1 m below the surface (USDA, 1995). The gray colors indicate a lack of iron-coated sand grains and potential iron deficiency at that depth (Richardson & Vepraskus, 2001). Even if phosphate is abundant and pH and redox potential are conducive to formation of variscite, strengite, or vivianite, precipitation is unlikely if iron is not present.
Onsite System Setback Distances
The OWS was contributing elevated concentrations of phosphate to soil water and shallow groundwater beneath and down-gradient from the system. While groundwater phosphate concentrations decreased with increasing distance from the system, elevated concentrations (0.84 [+ or -] 0.59) were still discovered more than 30 m down-gradient. Research conducted in various areas of Canada (Robertson et al., 1998) and St. George Island, Florida (Corbett et al., 2002) also found elevated phosphorous concentrations more than 30 m down-gradient from OWS installed in sandy, coastal soils. For our site, it was determined that a 39 m setback would be necessary to reduce phosphate concentrations to background levels. Thus, 30 m may not be a sufficient setback distance from OWS to surface waters in some settings.
Our study has shown that OWS installed in some sandy coastal environments can contribute significant phosphate to shallow groundwater more than 30 m down-gradient from the OWS. The OWS in our study was located more than 40 m from the estuary and thus phosphate concentrations were reduced to background levels before reaching surface waters. In eastern North Carolina and other coastal regions, setback distances from OWS may need to be increased, however, to protect nutrient-sensitive waters or more advanced technologies employed to reduce phosphate transport in some areas.
Acknowledgements: The authors would like to thank Dr. Max Zarate-Bermudez for facilitating the project, and the Centers for Disease Control and Prevention and NEHA for funding and assisting with the project. We would also like to acknowledge the efforts of several East Carolina University students including Shawn Thieme, Keaton Henry, Rob Howard, Sarah Hardison, Katie Supler, and others.
Corresponding Author: Charles Humphrey, Assistant Professor of Environmental Health Sciences, East Carolina University, 3408 Carol Belk, Greenville, NC 27858. E-mail: firstname.lastname@example.org.
Corbett, D.R., Dillon, K., Burnett, W., & Schaefer, G. (2002). The spatial variability of nitrogen and phosphorus concentration in a sand aquifer influenced by onsite sewage treatment and disposal systems: A case study on St. George Island, Florida. Environmental Pollution, 117(2), 337-345.
Davis, J.C. (2002). Statistics and data analysis in geology (3rd ed.). New York: John Wiley & Sons.
Harman, J., Robertson, W.D., Cherry, J.A., & Zanini, L. (1996). Impacts on a sand aquifer from an old septic system: Nitrate and phosphate. Groundwater, 34(6), 1105-1114.
Heath, R.C. (1998). Basic ground-water hydrology (Water Supply Paper No. 2220). Reston, VA: U.S. Geological Survey.
Humphrey, C., Deal, N., O'Driscoll, M., & Lindbo, D. (2010, May). Characterization of onsite wastewater nitrogen plumes in shallow coastal aquifers, North Carolina. Paper presented at Proceedings of the World Environmental and Water Resources Congress, Providence, RI.
Humphrey, C.P., & O'Driscoll, M.A. (2011). Biogeochemistry of groundwater beneath onsite wastewater systems in a coastal watershed. Universal Journal of Environmental Research and Technology, 1 (3), 320-328.
Karathanasis, A.D., Mueller, T.G., Boone, B., & Thompson, Y.L. (2006). Nutrient removal from septic effluents as affected by soil thickness and texture. Journal of Water Health, 4(2), 177-195.
North Carolina Division of Water Quality. (2010). Nonpoint source management program. Retrieved from http://portal.ncdenr. org/web/wq/ps/nps
Postma, FB., Gold, A.J., & Loomis, G.W. (1992). Nutrient and microbial movement from seasonally used septic systems. Journal of Environmental Health, 55(2), 5-10.
Pradhan, S.S., Hoover, M.T., Austin, R.E., & Devine, H.A. (2007). Potential nitrogen contributions from onsite wastewater treatment systems to North Carolina's river basins and sub-basins. North Carolina Agricultural Research Service Technical Bulletin 324. Retrieved from http://www.soil.ncsu.edu/ publications/TB324Finalmay29.pdf
Ptacek, C.J. (1998). Geochemistry of a septic system plume in a coastal barrier bar, Point Pelee, Ontario, Canada. Journal of Contaminant Hydrology, 33(3-4), 293-312.
Reay, W.G. (2004). Septic tank impacts on ground water quality and nearshore sediment nutrient flux. Groundwater, 42(7), 1079-1089.
Richardson, J.L., & Vepraskus, M.J. (2001). Wetland soils: Genesis, hydrology, landscapes, and classification. Boca Raton, FL: CRC Press.
Robertson, W.D., Cherry, J.A., & Sudicky, E.A. (1991). Groundwater contamination from two small septic systems on sand aquifers. Groundwater, 29(1), 82-92.
Robertson, W.D., Schiff, S.L., & Ptacek, C.J. (1998). Review of phosphate mobility and persistence in 10 septic system plumes. Groundwater, 36(6), 1000-1010.
U.S. Department of Agriculture. (1995). Soil survey of Beaufort County, North Carolina. Retrieved from http://soils. usda.gov/survey/printed_surveys/
U.S. Environmental Protection Agency. (2002). Onsite wastewater treatment systems manual (Doc. No. EPA/625/R-00/008). Retrieved from http://nepis.epa.gov/Adobe/PDF/30004GXI.pdf
Zanini, L., Robertson, W.D., Ptacek, C.J., Schiff, S.L., & Mayer, T. (1998). Phosphorus characterization in sediments impacted by septic effluent at four sites in central Canada. Journal of Contaminant Hydrology, 33(3-4), 405-429.
Charles Humphrey, MS, PhD, REHS
Department of Environmental Health Sciences
East Carolina University
Mike O'Driscoll, MS, PhD
Department of Geological Sciences
East Carolina University
Nancy Deal, MS, REHS
North Carolina Department of Health and Human Services
David Lindbo, MS, PhD
North Carolina Cooperative Extension
North Carolina State University
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ADVANCEMENT OF THE SCIENCE|
|Author:||Humphrey, Charles; O'Driscoll, Mike; Deal, Nancy; Lindbo, David|
|Publication:||Journal of Environmental Health|
|Date:||Jan 1, 2014|
|Previous Article:||Effects of centralized and onsite wastewater treatment on the occurrence of traditional and emerging contaminants in streams.|
|Next Article:||Place-based exposure and cataract risk in the Beaver Dam cohort.|