Recirculating vertical flow constructed wetland: green alternative to treating both human and animal sewage.Introduction
Conventional septic systems commonly used to treat residential sewage in areas without sanitary sewers are considered a primary contaminant source for surface or underground water supplies (Whitehill, Brian-Tercha, & Davis, 2003), including human drinking water wells (Bhardwaj, 2003). Horizontal gravity flow subsurface constructed wetlands have been used for more than a decade in LaGrange County, Indiana, to remove human sewage contaminants before underground final disposal. Horizontal gravity flow constructed wetlands provide acceptable removal efficiency for the five-day biological oxygen demand ([BOD.sub.5]), total suspended solids (TSS), and fecal coliform bacteria (FC), but they have low efficiency to eliminate nitrogenous compounds because of the limited oxygen transfer. Garcia and co-authors (2006) showed that by modifying the horizontal gravity flow to a vertical recirculating flow, the oxidation process of ammonianitrogen--the predominant form of nitrogen in residential septic tank effluents--can achieve low levels plus improve BOD and TSS removal before land application and surface or underground discharge. The performance of a subsurface constructed wetland using a recirculating vertical flow to treat both human and domestic animal sewage from the LaGrange County (Indiana) Animal Shelter was examined and is discussed in this article.
Materials and Methods
A constructed wetland (6.1 m [20 ft] x 6.1 m [20 ft]; 1.2 m [48 inches] deep; Figure 1) to treat sewage on site at the local animal shelter facility was built in LaGrange County in northeast Indiana. The volume of sewage treated was assumed to be approximately 1,817 L (480 gallons) per day generated by 18 dog runs, 12 cat cages, two isolation rooms, and two full-time employees according to the current commercial regulations from the Indiana State Department of Health (Indiana State Board of Health, 1988). The subsurface recirculating vertical flow constructed wetland (RVFCW) utilized a timer to apply pretreated sewage over the entire top bed area planted with wetland flowering plants, such as water iris (Iris virginica), swamp milkweed (Asclepias incarnata), great blue lobelia (Lobelia siphilitica), and New England aster (Aster novae-angliae) with a density of about four plants per square meter. One plant of conventional garden plants such as morning glory vines (Ipomoea leptophylla), cheddar bath's pinks (Dianthus gratianopolitanus), and fern were also planted to create the visual effect of a conventional flower garden. The wastewater was collected in two 3,800 L (1,000 gallons) septic tanks installed in a series, and then gradually released by gravity flow to the feeding inlet bottom of the wetland cell, which used plastic chambers to spread the incoming effluent at the front end. The wetland cell was built with a 30 mil PVC liner and filled with a bottom layer of 61 cm (24 inches) depth of 13--25 mm diameter stone and a top layer of 61 cm (24 inches) depth of 4 mm diameter gravel. The two layers of stone were separated by a second PVC liner extended over most of the top area of the gravel, leaving 25% of the bottom gravel layer nearest the wetland inlet uncovered. Treated water was collected in the outlet sump pump pit consisting of a 1.52 m (5 feet) section of 122 cm (48 inches) diameter black corrugated drain tile installed vertically. The bottom of the drain tile was filled with concrete to avoid infiltration into the ground and the top was fitted with a concrete block and a metal laminated piece functioning like an opening window. During each 30 minute period, the wetland effluent collected in the sump pit was pumped for five minutes via a 5 cm (2 inches) PVC pipe feeder line to a 2.5 cm (1 inch) PVC manifold pipe onto the top of the gravel and stone bed. The pump, with a maximum recirculation flow of 50 gallons per minute, was controlled by an electronic analog repeat cycle timer built in as standard feature in the electrical control panel.
[FIGURE 1 OMITTED]
Water samples from the effluent collected in the sump pump pit were periodically submitted to a certified U.S. Environmental Protection Agency (U.S. EPA) testing laboratory and analyzed using the methodology described in Standard Methods for the Examination of Water and Wastewater (SM) (Clesceri, Greenberg, & Eaton, 1998) and in Methods for Chemical Analysis of Water and Wastes (U.S. EPA, 1983) for the following parameters: [BOD.sub.5] (SM, 5210-B), nitrate (N[O.sup.-.sub.3] -N, U.S. EPA-353.2), ammonianitrogen (N[H.sup.+.sub.4]-N; SM, 4500-N[H.sub.3]), TSS (SM, 2540), fecal coliform bacteria (E. coli; U.S. EPA, Coliscan), total phosphorus (SM, 4500-P), and TKN (U.S. EPA, 351.2). Total phosphorus (TP) is all of the phosphorus present in the sample regardless of form (organic, inorganic, and hydrolyzable). TKN is defined as the sum of the free N[H.sup.+.sub.4] -4N and organic nitrogen compounds. Total-N (TN) is calculated by the combination of TKN and N[O.sup.-.sub.3]-N. Water analyses parameters collected on site in the outlet sump pit effluent included temperature (air and water) and dissolved oxygen using an oxygen meter, oxygen reduction potential using an ORP meter, and pH using a pH meter. The removal efficiency was calculated according to the equation used by Ebeling and coauthors (2003) as follows:
Removal Efficiency (%) = ([Influent-Effluent]/Influent) x 100
It is important to mention that these calculations were based only in comparison the concentration from influent (septic tank effluent) versus effluent from the wetland cell. Rainfall and evapo-transpiration were not considered in the calculations. A water balance mass (lbs/day) based on a ratio of influent versus effluent flow would be necessary to determine a more accurate removal efficiency.
Results and Discussion
The water quality effluents results for dissolved oxygen, ORP, pH, air and water temperature, [BOD.sub.5], TSS, TP, FC (Table 1), and nitrogenous compounds (Table 2) fluctuated in the RVFCW during the experimental period. Final mean concentrations were relatively low, showing high removal efficiency (Table 3). This constructed wetland provided better contaminants removal efficiency in comparison with the system used by Garcia-Perez and coauthors (2006). One reason could be the presence of the second liner. The second liner creates a completely aerobic second upper compartment with a bottom compartment working as anaerobic environment. These aerobic and anaerobic conditions are well delimited, allowing the microorganisms a better processing of the nitrogenous compounds. The high efficiency reported in this research is similar to data reported by Cooper and Green (1995). They found that vertical flow constructed wetlands can achieve essentially complete [BOD.sub.5] removal because of high amounts of oxygen transfer through the gravel/root hair bed. The final mean [BOD.sub.5] in our study for the animal shelter constructed wetland was 2 mg/L during the two-year experimental period.
The presence of high oxygen content in the filter bed is necessary for bacteria to remove sewage pollutants. Platzer and Mauch (1997) reported that the removal efficiency of vertical-flow constructed wetlands depend on natural aeration to provide oxygen and therefore [BOD.sub.5], COD, and NH+ 4-N removal is high, but total-N elimination is limited. Luederitz and co-authors (2001) suggested that denitrification capacity of vertical flow wetlands is poor. Their study, however, showed TN removal of 71% without using recirculation. To improve nitrogen compound removal in vertical flow wetlands, they recommend intermittently loading the reed beds. In the present study, use of the timer allowed an intermittent loading of the wetland every 30 minutes and the TN performance reached 83% of removal efficiency.
Phosphorus is generally the limiting nutrient in freshwater ecosystems and its removal is important to reduce the high impact of wastewater effluents in receiving surface or underground waters. According to Wallace and Knight (2006), constructed wetlands did not remove significant quantities of phosphorus. Luderitz and Gerlach (2002) found that RVFCW has a lower rate (27%) of phosphorus removal. One reason for the low phosphorus removal efficiency is the shorter flowing distance and retention time in RVFCW. Wymazal (2004) found that TP removal increases 40% to 60% using horizontal flow constructed wetlands where retention time and flowing distance is extended in comparison with RVFCW. Also, the substrate used in the wetland cell has been shown as a key element and significant mechanism in the process of phosphorus removal. Tang and co-authors (2008) did comparison phosphorus removal from vertical flow wetlands using four different substrates. Gravel did show the lowest phosphorus removal efficiency (21%) followed by ironstone and hornblende both with a 33% efficiency. Korkusuk and co-authors (2004) used gravel as substrate for a vertical flow wetland and had a low TP removal efficiency (9%). Prochaska and Zouboulis (2006) removed 45% of phosphates using a mixture of sand and dolomite as substrate for a pilot vertical flow constructed wetland. The RVFCW in our study used gravel as substrate reaching a 33% efficiency phosphorus removal. Gravel as substrate has low treatment performance to remove phosphorus. An alternative to changing gravel as the substrate to improve phosphorus removal, however, could be to find plants needing phosphorus to grow and harvest them. Phosphorus--along with nitrogen and potassium--is a nutrient that plants need in relatively large quantities for normal growth. A future practical application for RVFCW could be to use them to produce agriculture commodities like corn, soybeans, or sunflowers for production of ethanol.
This study shows that the recirculating vertical flow constructed wetland built in the LaGrange County Animal Shelter, Indiana, had high treatment efficiency. The percentage removal after a two-year operation was high for [BOD.sub.5] (99%), TSS (98%), AN (96%), TKN (94%), TN (83%), and FC (95%). Nitrate-nitrogen final mean value was 6.8 mg/L and dissolved oxygen concentration increased from 1.8 to 4.3 mg/L. Removal efficiency for total phosphorus was low (33%), however. An alternative to improve its removal could be to use plants needing phosphorus in large quantities for normal growth and harvesting them or to extend the retention time and flowing distance. The results indicate that a recirculating vertical flow constructed wetland is a viable green technology to pretreating both human and domestic animal sewage before final disposal.
Bhardwaj, V. (2003). Preventing well contamination. Technical Brief On Tap Magazine, 3, 1--4.
Clesceri, L.S., Greenberg, A.E., & Eaton, A.D. (Eds.). (1998). Standard methods for the examination of water and wastewater (20th ed.). Washington, DC: American Public Health Association, American Water Works Association, and Water Environment Federation.
Cooper, P.F., & Green, B. (1995). Reed bed treatment systems for sewage treatment in the United Kingdom--The first 10 years experiences. Water Science Technology, 32, 317-327.
Ebeling, J., Tsukuda, S., Hankins, J., & Solomon, C. (2003). Performance evaluation of a recirculating sand filter and peat filter in West Virginia. Small Flows Quarterly, 4, 27-37.
Garcia-Perez, A., Grant, B., & Harrison, M. (2006). Water quality effluent from a recirculating vertical flow constructed wetland. Small Flows Quarterly, 7(4), 34--38. Retrieved July 27, 2009, from http:// www.nesc.wvu.edu/nsfc/Articles/SFQ/SFQ_f06_PDF/Juried2.pdf
Indiana State Board of Health. (1988). On-site water supply and wastewater disposal for public and commercial establishments. Bulletin Indiana State Department of Health, 13, 1-50. Retrieved July 27, 2009, from http://www.in.gov/isdh/files/se_13.pdf
Korkusuz, E., Beklioglu, M., & Demirer, G. (2004). Treatment efficiencies of the vertical flow pilot-scale constructed wetlands for domestic wastewater treatment. Turkish Journal of Engineering and Environmental Science, 28, 333-344.
Luederitz V., Eckert, E., Lange-Weber, M., Lange, A., & Gersberg, R. (2001). Nutrient removal efficiency and resource economics of vertical flow and horizontal flow constructed wetlands. Ecological Engineering, 18, 157-171.
Luederitz, V., & Gerlach., F. (2002). Phosphorus removal in different constructed wetlands. Acta Biotechnologica, 22(1-2), 91-99.
Platzer, C., & Mauch, K. (1997). Soil clogging in vertical flow reed beds--mechanism, parameters, consequences and ... solutions? Water Science and Technology, 35, 175-181.
Prochaska, C.A., & Zouboulis, A. (2006). Removal of phosphates by pilot vertical-flow constructed wetlands using a mixture of sand and dolomite as substrate. Ecological Engineering, 26(3), 293-303.
Tang, X.Q., Huang, S.L., & Scholz, M. (2008). Comparison of phosphorus removal between vertical subsurface flow constructed wetlands with different substrates. Water and Environment Journal, 1747, 6585-6593.
U.S. Environmental Protection Agency, Office of Research and Development. (1983). Methods for chemical analysis of water and wastes (manual 600/4-79/020). Washington, DC. Staff of the Environmental Monitoring and Support Laboratory of the Environmental Research Laboratory in Cincinnati.
Wallace, S.D., & Knight, R. (2006). Small-scale constructed wetland treatment systems: Feasibility, design criteria, and O & M requirements. Alexandria, VA: Water Environment Research Foundation.
Whitehill, T.J., Brian-Tercha, P.E., & Davis, J.F. (2003). Evaluation of a recirculating sand filter followed by a subsurface-flow constructed wetland to achieve denitrification. Small Flows Quarterly, 49(1), 30-35.
Wymazal, J. (2004). Removal of phosphorus in constructed wetlands with horizontal sub-surface flow in the Czech Republic. Water, Air and Soil Pollution: Focus, 4(2-3), 657-670.
Alfredo Garcia-Perez, PhD
Mark Harrison, PE
Corresponding Author: Alfredo Garcia-Perez, Administrator and Environmental Health Specialist, LaGrange County Health Department, Environmental Health Office, 304 North Townline Road, Suite 1, LaGrange, IN 46761-1319. E-mail: firstname.lastname@example.org.
TABLE 1 Water Quality Variables * Wetland DO pH Air Water Age (mg/L) (SU) ([degrees]C) ([degrees]C) (Days) 60 2.8 7.3 28.1 24.1 108 4.6 7.7 13.5 20.7 152 5.7 7.6 14.0 11.5 205 5.8 7.4 3.5 7.8 295 5.8 7.6 3.6 4.2 338 4.4 7.4 17.1 13.8 428 5.2 7.2 21.9 21.1 488 4.3 7.4 14.6 14.2 663 2.3 7.1 1.5 5.1 765 2.1 7.2 28.4 19.2 Mean 4.3 7.4 14.6 14.2 SD 1.4 0.2 9.7 7.0 Maximum 5.8 7.7 28.4 24.1 Minimum 2.1 7.1 1.5 4.2 Wetland BOD TSS TP E. coli Age (mg/L) (mg/L) (mg/L) (MPN/ (Days) 100 ml) 60 4 2 ND 21,000 108 0 0 ND 11,000 152 0 0 ND 750 205 0 0 ND 3,900 295 0 0 7 38,000 338 2 0 8 45,000 428 2 3 6 44,000 488 0 0 7 33,000 663 7 2 15 63,200 765 0 0 7 2,420 Mean 2 1 8 26,227 SD 2 1 3 21,585 Maximum 7 3 15 63,200 Minimum 0 0 6 750 * MPN: most probable number; ND: no data. TABLE 2 Nitrogen Removal Efficiency Wetland Age Nitrogen Compounds (mg/L) (Days) N[H.sup.+.sub.4]-N N[O.sup.3]-N TKN TN 60 3 4 6 108 1 4 2 7 152 1 4 2 6 205 0 8 3 11 295 1 20 2 22 338 1 15 2 17 428 1 2 2 4 488 1 9 3 13 663 2 2 12 14 765 10 1 2 3 Mean 1 7 3 10 SD 2 6 3 6 Maximum 3 20 12 22 Minimum 10 1 2 3 TABLE 3 Water Quality Efficiency * Water Quality Parameters Recirculating Vertical Flow Wetland Influent Effluent Removal (%) Fecal coliform bacterium (FC [MPN/100 ml]) 2140200 26227 (-) 99 Biochemical oxygen demand ([BOD.sub.5] [mg/L]) 173 1.5 (-) 99 Total suspended solids (TSS [mg/L]) 47 1.0 (-) 98 Total Kjeldahl nitrogenous (TKN [mg/L]) 57 3.4 (-) 94 Ammonia-N (N[H.sup.+.sub.4]-N [mg/L]) 46 3.2 (-) 96 Total-nitrogen ([TN = TKN + N[O.sup.-.sub.3]-N] [mg/L]) 57 10 (-) 83 Total phosphorus (TP [mg/L]) 12 8 (-) 33 Nitrate (N[O.sup-.sub.3]-N [mg/L]) 0 6.8 NA Dissolved oxygen (DO [mg/L]) 1.8 4.3 NA pH (Standard units) 7.0-8.1 7.1-7.7 NA Temperature ([degrees]C) 13.9 14.2 NA Oxido-reduction potential (ORP [mV]) -138 +68 NA * MPN: most probable number; NA: not applicable.