Evaluation of peat biofilters for onsite sewage management.
Mobile Bay and the coastal waters of Alabama in the Gulf of Mexico receive the discharges of many large river systems that extend into the northern reaches of Mississippi, Alabama, Tennessee, and Georgia. As with other stretches of the Gulf Coast, these coastal waters are impacted by fecal coliform bacteria from inadequately treated septic tank effluent. To protect public health from this and other pollutants, the harvesting of shellfish is frequently restricted in Florida, Alabama, Mississippi, Louisiana, and Texas. More than half of the shellfish-producing areas of the Gulf Coast are permanently or conditionally closed due to fecal coliform contamination resulting from the growing human population in the region (1). Weeks Bay Estuary, situated on the eastern shore of Mobile Bay, and a depository for Fish and Magnolia Rivers, has been temporarily closed to commercial shellfishing for many years due to elevated fecal coliform levels.
Sanitary surveys of the shorelines of Mobile Bay and Weeks Bay Estuary, conducted by the Alabama Department of Public Health (ADPH) in 1991 and 1992, point to the heavy reliance on septic tank systems for onsite wastewater disposal. Almost 65% of Baldwin County's population of 115,000 is served by onsite sewage treatment and disposal systems. More than 1,250 septic tank systems are now being installed annually in Baldwin County.
Soils of the Weeks Bay area exhibit high wet-season watertables. They are physiographically marine terrace, and Azonal (alluvial) and Intrazonal in nature. Parent material in the area is coastal plain sediment. The soils are free-draining during the drier months of the year, providing a deep unsaturated (4-5 feet) zone for percolation of treated wastewater. During the wet season, the high watertable is a major cause of failing onsite sewage systems in the area.
In October 1992, the U.S. Environmental Protection Agency through the Gulf of Mexico Program approved funding for a project in sewage management to demonstrate a reduction in fecal coliform bacteria in shellfish-growing waters. ADPH proposed to replace 20 existing but compromised soil absorption systems with an innovative, peat-based onsite sewage disposal system near Weeks Bay Estuary in south Alabama. Pre- and post-installation sampling of Weeks Bay waters and analysis of effluent quality over a 12-month period were proposed to determine fecal coliform levels. Bord na Mona (BNM), the Irish peat corporation and manufacturer of a commercially available peat biofilter, contributed materials and engineering costs to the demonstration project.
The focus of the study, a modular bio-filtration system called Puraflo[TM], is manufactured in Ireland by BNM and distributed in the U.S. through a subsidiary in North Carolina. The Puraflo [TM] system has been used in Ireland since 1988 for onsite wastewater treatment. More than 400 systems are now in operation in Ireland. The peat biofilter has provided effective onsite sewage treatment and disposal where soil and site conditions preclude the use of conventional soil absorption systems. With the Puraflo[TM] system, BNM achieved 96% and 99+% removal of biological oxygen demand ([BOD.sub.5]) and fecal coliform bacteria, respectively, from septic tank effluent under a temperate maritime climate in Ireland, that has [TABULAR DATA FOR TABLE 1 OMITTED] mild winters (mean range: 4.4 [degrees] to 7.2 [degrees] C or 40 [degrees] to 45 [degrees] F) and cool summers (mean range 15 [degrees] to 16.7 [degrees] C or 59 [degrees] to 62 [degrees] F) (2,3,4). Such efficiencies, and the commercial manufacture of a patented and standardized system, prompted ADPH to evaluate Puraflo's ability to reduce fecal coliform contamination in Alabama's coastal waters.
The Puraflo[TM] system was developed in the early 1980s using a fraction of the peat harvested each year for electric power generation, home fuel, and other applications. This fraction of the harvest is made up of moderately decomposed plant roots. It was found to be suitable as a filtration medium due to its large pore space and surface area, and its ability to withstand excessive compaction over time. Peat has complex physical, chemical, and biological properties. It is formed over several hundreds of years by the gradual accumulation of dead vegetation under anaerobic and waterlogged conditions. Only partial decomposition of the vegetable matter can occur, producing thick deposits of peat in various topographical situations.
The peat used in the Puraflo[TM] system is fibrous and structurally stable. Physical filtering, chemical adsorption, and biological treatment by microorganisms are the principle forces that operate within the system. Chemical adsorption of wastewater constituents to the peat medium is enhanced by the high cation exchange capacity of peat fiber. The large surface area typical of the medium provides for an aerobic environment that supports biological treatment. After several weeks in the ground, the peat biofilter is invaded by a range of microorganisms and invertebrates from the septic tank effluent and the surrounding soil. Such organisms include bacteria, fungi, protozoa, nematodes, earthworms, rotifers, and others. The treatment of septic tank effluent within the peat medium is performed for the most part by acid-tolerant bacteria and fungi and the other biota inhabiting the peat medium. Pathogenic bacteria undergo a significant die-off due to the acidic environment of the peat medium and to the biological treatment process.
BNM installed a demonstration peat filter [TABULAR DATA FOR TABLE 2 OMITTED] bed at a residential site in Ashbourne, County Meath, Ireland, in 1986. The first modularized systems were constructed at seven residences in Clonmel, County Tipperary, Ireland, in 1988 (5) and are still giving very good results without replacement of the peat media. BNM surmises that the peat will last more than 10 years and possibly more than 15 years (6). BNM's seven years of experience along with the longevity of non-modularized, in-ground peat filters installed in Maine in 1978 (7) seem to support the estimated life of the peat media in the Puraflo[TM] system.
A forerunner to the Puraflo[TM] system was developed in the United States by Dr. Joan Lake Brooks of the University of Maine in 1978. The innovative design used milled and air-dried Sphagnum moss peat in a gravity distribution system, built in-place. While the peat medium used in the Brooks design differs in its physical attributes from that used in the Puraflo[TM] system, they share similar biochemistries. Studies of the Brooks peat system since 1978 have shown consistent reductions of 99.9+% in fecal coliform levels from septic tank effluent (8). Effluent [BOD.sub.5] levels of 5-20 mg/1 (9) and 3-11.9 mg/1 have also been achieved. Brooks attributed the reduction of fecal coliform bacteria to the acidic environment of the medium; to antibiotic and phenolic substances inherent in peat fiber; and to the presence of anti-bacteriacidal fungi in the peat medium. These observations were made under the cool maritime climate of Maine. The most efficient treatment of wastewater by the peat system occurred during the winter months when temperatures ranged from 1 [degrees] to 10 [degrees] C (8). Relative respiration rates of fungi versus bacteria indicated that fungi were the dominant component of microbiota living in the peat medium. The metabolic by-products of yeasts and filamentous fungi; the acidic environment of the peat biofilter; and disinfectant properties of peat were cited as key factors for the reductions of fecal coliform bacteria and pathogenic bacteria, and for the high quality of the treated effluent.
The main objective of the demonstration project was to measure the ability of the Puraflo[TM] system to reduce the levels of fecal coliform bacteria entering shellfish-growing waters of the Gulf of Mexico from domestic wastewater. The peat biofilter had never been tested under the hot and humid conditions of south Alabama and the Gulf Coast. Through the project, ADPH and BCHD sought to raise public awareness of the benefits of effective onsite sewage disposal and of the availability of alternatives to conventional onsite sewage systems.
Materials and Methods
Property owners adjacent to the Weeks Bay Estuary were invited to participate in the demonstration project in January 1993, to have their existing septic tank systems evaluated and possibly replaced at no charge. Following a public meeting during which the project's objectives were presented, 33 homesites were investigated to determine the condition of their onsite sewage disposal systems. Measurements were made of existing septic tanks and field lines to determine depth, linear footage, and separation from the wet season watertable, surface waters, wells, and other features Homeowners provided additional data by questionnaire regarding the age of their septic tank systems, number of household occupants, repairs to their septic system, and plumbing problems. Twenty homesites were selected using a scoring procedure that favored those septic tank systems that exhibited the most severe operating conditions. The project's location is shown in Figure 1.
Each Puraflo[TM] system [ILLUSTRATION FOR FIGURE 2 OMITTED] uses a standard septic tank which gravity feeds to a pumping chamber. A one-horsepower submersible pump sends a controlled dose of septic tank effluent under pressure to multiple peat modules. The precast polyethylene modules receive the effluent through a central manifold which is connected to PVC distribution [TABULAR DATA FOR TABLE 3 OMITTED] pipes. The effluent percolates through a bed of compacted peat, two feet deep, and is retained for up to 48 hours. As more liquid enters at the top of the peat bed, it forces treated effluent to exit the modules through holes along the bottom. The peat modules sit on a gravel base, 8 inches deep, that extends laterally for 6 inches beyond the modules. The gravel provides both a firm foundation for the modules, and percolation to the underlying soil for the treated effluent.
The Puraflo[TM] systems, utilized for this project, represented the first North American installations. Published design criteria for sizing of the system based upon percolation rates were not available during the design phase of the demonstration project. Coarse-to-fine textured loamy sands, with estimated percolation rates of less than 20 minutes per inch, allowed for the installation of minimum-sized systems of approximately 320 square-feet for the gravel pad or footprint area. BNM issued design criteria, conditions of installation and a percolation rate sizing chart in June 1995 (6).
To complete the installation, the modules were secured by a soil berm finished to grade with a 3:1 slope. The compact system, with the soil berm, occupies an average of 320 square-feet of yard space for a three-bedroom house (500 g/d). The Puraflo[TM] system uses electrical controls to operate the pump unit in response to rising liquid levels. Audible and visible alarms provide early warning for possible malfunctions.
After site evaluations and installation designs were completed, construction of 20 Puraflo[TM] installations commenced in August 1993. Each construction was supervised and certified by an engineer registered in Alabama. To facilitate the collection of groundwater samples, project staff placed well points (10-ft x 3-in PVC pipes) 25-feet down-gradient toward the canals from 10 Puraflo[TM] systems, where monitoring activities would be conducted.
The monitoring program and performance evaluation began in November 1993. Over a 12-month period, at 10 Puraflo[TM] installations, the following samples were collected: a) septic tank effluent from the pumping chamber; b) Puraflo[TM] effluent from the sampling port; c) groundwater from wells 25-feet down-gradient toward canal; and d) canal water, from two stations on three canals.
Samples b, c, and d were collected monthly from November 1993 through October 1994. Septic tank effluent, sample a, were collected monthly beginning in April 1994, allowing the measurement of removal rates for test parameters. All samples were collected in 500 ml sterilized sample bottles. Measurements of air temperature, sample, temperature, and sample pH were made during sample collection. Water meter readings and watertable elevations were also recorded each month.
All samples were promptly chilled and delivered to the University of South Alabama, Mobile, Alabama, for fecal coliform analysis, within five hours of collection. The laboratory procedure used was the membrane filter method, and the test results were given as fecal coliform colonies per 100 ml. In months 11 and 12, an extra 1-liter sample was collected at each sampling point, except at the canals, and was analyzed to determine the concentrations of organic nitrogen (Org-N), ammonia nitrogen (N[H.sub.3]-N), and nitrate nitrogen (N[O.sub.3]-N), and 5-day biological oxygen demand ([BOD.sub.3]).
During the course of the monitoring program, modifications to the Puraflo[TM] systems were necessary. In the first three months, complications arose involving the electrical control panels, which operate the pump unit. Alarms and pump units malfunctioned at a number of sites, due to the changeover from Irish to U.S. electrical specifications. Such problems were corrected by BNM. A consequence of these malfunctions was the carryover of grease and solids from the septic tank to the peat medium, which is detrimental to the effective operation of the peat biofilter. Further inspections revealed that carryover of solids and grease had occurred in other installations not experiencing control panel malfunctions. The half-baffle wall in the septic tanks used for the project, and permissible under Alabama Rules, was suspected. BNM added a Model A-100 Zabel effluent filter to the septic tank outlet at each Puraflo[TM] installation as a corrective measure.
Puraflo[TM] System Unit Costs for This Project.
Puraflo[TM] systems (hardware, shipping, & handling) $7,413.50 Installation (including engineer) 1,800.00
Total Unit Cost: $9,213.50
Extra peat fiber was also added to the Puraflo[TM] systems during the first six months of the monitoring program. Uneven settlement of the peat fiber was found at several installations. These additions were made to eliminate voids in the medium and prevent short-circuiting of septic tank effluent through the peat. The modifications described above were made during the period of February through June 1994.
Results and Discussion
Fecal coliform concentrations in the biofilter effluent averaged 57,665 colonies per 100 ml over the 12-month sampling period. Because of variability and the sensitivity of the average (mean) to outliers, the median may be a better measure of central tendency. Median fecal coliform concentration for the same period was 3,200 colonies per 100 ml. Influents to the biofilters contained an average of 6.1 x [10.sup.5] fecal coliform colonies per 100 ml. As with effluent concentrations, much variability was found between septic tank effluent samples and between sample months. Overall, the biofilters were responsible for an average reduction of 92.6% in fecal coliform concentrations for the one-year sampling period (median value=97.6%). Average reductions improved with time, reaching 98% in the last sampling period. Fecal coliform concentrations and removal rates are summarized in Table 1.
BOD and various nitrogen species were measured in influent and effluent samples and in groundwater samples during September and October 1994. Biofilter effluent quality was generally good, with BOD values of less than 20 mg/l and ammonia nitrogen values of less than 1 mg/l. Nitrate nitrogen concentrations of the effluent samples ranged from 10 to 40 mg/ l. Some nitrate was detected in almost all groundwater samples, although at low concentrations. Table 2 presents results of sample analysis for BOD and nitrogen species performed in the last two months of monitoring activities.
In Figure 3, the average effluent fecal coliform concentrations for each peat biofilter monitored during the 12-month sampling period are presented. Although overall average effluent concentrations was 57,665 colonies per 100 ml, most of the systems (seven out of 10) performed considerably better than this. The graph of fecal coliform densities shown in Figure 3 follows a pattern similar to that observed for the septic tank effluent samples collected as influent to the 10 Puraflo[TM] systems, with peaks at P1, P2, and P6.
Sludge carry-over onto the peat beds caused problems at sites P1, P2, and P6, resulting from malfunctions of the electrical controls. The absence of a full-baffle wall in the septic tank was also linked to a slight buildup of grease on the medium. The addition of Model A-100 Zabel effluent filters to the septic tanks resulted in improvements in system efficiencies. Values at site P2 were particularly high, with average fecal coliform concentrations of 350,000 colonies per 100 ml. Performance of the systems improved dramatically over the last three months of the study. The best effluent quality was observed just after the start-up (November to January) and from August through October. Most operational problems, related to electrical controls, septic tank design, and pump malfunctions, occurred in the period from February through June 1994. With correction and modification of the systems, including addition of effluent filters, and extra peat fiber, system efficiency improved toward the end of the study. Figure 4 gives the monthly average fecal coliform levels in Puraflo[TM] effluent over the first 12 months of operation.
Septic tank effluent samples were collected beginning in April 1994. Fecal coliform concentrations averaged 6.1 x [10.sup.5] colonies per 100 ml. As with treated effluent, fecal coliform levels, much variation was found in the septic tank effluent samples. With the results of these latter samples, the percentage removal of fecal coliform bacteria achieved with the peat biofilters was determined. As illustrated in Figure 5, average reductions in fecal coliform levels of 93% were achieved by the Puraflo[TM] systems over the first 12 months of operation. Removal efficiencies improved during the last three months of the monitoring program. Average monthly reductions improved to 98%, with fecal coliform levels of 4,911 colonies per 100 ml. Systems P4, P9, and P10 consistently achieved excellent reductions in the order of 99% over the one-year period. Systems P1 and P6 fell below 90% removal over the duration of the study.
The increase in biofilter performance suggested that the peat biofilter was undergoing acclimation to the septic tank effluent and the surrounding soil, with the buildup of biota populations in the medium. This improvement in performance can be seen more clearly by studying effluent quality between the first and the last six months of the study. When the two periods are compared on the basis of cumulative percentage removal, the latter period shows improved effluent quality. While nearly 50% of May to October samples had fecal coliform concentrations of less than 1,000 colonies per 100 ml, only 24% of earlier samples exhibited such quality. The duration of the acclimation process was not clearly defined within the timeframe of the monitoring program. However, maturation of the peat medium and the mix of microflora and invertebrate organisms may be an ongoing process through a longer period of operation. Such a determination would require further study.
The improvement in the peat biofilter performance is also depicted in Table 3. Fecal coliform concentrations in treated effluent are shown at six, nine, and 12 months post installation, as well as for the full year of operation. Removal rates for the same periods are also displayed. Further study would show if the peat biofilters' peak performance level had been reached during the course of this study.
No direct correlation was observed between fecal coliform values of canal and groundwater samples collected during the study and those of the Puraflo[TM] effluent. Long-term data, pre-and post-installation, would be required to define a relationship. Most groundwater samples collected had less than 10 fecal coliform colonies per 100 ml. Significant concentrations were observed, however, at sites 1, 5, 6, and 10. Groundwater wells at sites 5 and 6 averaged more than 400 colonies per 100 ml. Homesites identified by sites 5, 6, and 10 had sewage surfacing on the ground when first inspected in February 1993 during the selection phase of the project. The lots at sites G5 and G6 had a strong odor of septic tank effluent before and after installation of the peat biofilter. These odors dissipated over time; however, a reservoir of fecal coliform bacteria and organic matter remaining in the soil may have impacted the groundwater samples at these sites. A large flock of domestic fowl inhabiting lots at sites 5 and 6 were also considered as a potential source of fecal coliform bacteria detected in the groundwater.
Performance of the Puraflo[TM] systems in terms of BOD and nitrogen reduction was good. Based on data from the last two months of the sampling period, an average reduction in [BOD.sub.5] concentration of 85% was achieved. Organic nitrogen removal averaged 73%, while mean ammonia-nitrogen removal was 96%. In all systems, [BOD.sub.5] concentrations were less than 40 mg/l, most effluent samples having less than 30 mg/l. Ammonia concentrations in the biofilter effluent averaged less than 2 mg/l. The high surface area associated with the peat substrate, the aerobic environment, and the intermittent loading of wastewater on the peat medium provide for good nitrification. Effluents from biofilters P5 and P6 did not perform as well as others in removing BOD or nitrogen. Continued monitoring of these and other biofilter systems would help to answer questions pertaining to efficiencies over time.
The Puraflo[TM] systems evaluated in this study demonstrated a high level of wastewater treatment. Fecal coliform bacteria removal from septic tank effluent averaged 93% over the first 12 months of operation, with effluent concentrations averaging about 57,000 colonies per 100 ml. Because of performance variability, median values better represent central tendency. Median fecal coliform removal through the peat biofilters was 98%, and median effluent concentrations were determined to be 3,200 colonies per 100 ml. Fecal coliform removals improved over time, suggesting a period of acclimation for the peat biofilters before maximum efficiency is attained. After one full year of operation, the 10 biofilters monitored achieved a reduction in fecal coliform concentrations of 99%. The peat biofilters were also very effective in removing BOD and ammonia nitrogen. Effluent BOD and ammonia nitrogen concentrations averaged 18 mg/l and 1.2 mg/l, respectively. Such treatment performance amounts to 85% BOD removal and about 96% ammonia nitrogen removal (nitrification). Effluent nitrate nitrogen concentrations, however, averaged 25 mg/l. No direct correlations between biofilter treatment and groundwater and canal water samples could be determined.
This study successfully demonstrated the efficacy of the Puraflo[TM] system as an alternative onsite sewage treatment and disposal system for the purposes of protecting sensitive coastal environments. In addition to effective treatment performance, the peat biofilter system required only 150 square-feet (320 square-feet with soil berm) of yard space for each installation. Installation costs were higher than conventional onsite sewage systems. Unit costs incurred for the project are shown in Table 4.
The cost of the Puraflo[TM] systems included delivery of the module components from Ireland by ship to Savannah, Georgia, and by land vehicle to Baldwin County, Alabama. Currently, Bord na Mona's unit price for the Puraflo[TM] system is $6,500. Installation costs will vary from site to site, depending on site conditions. Such costs could be as low as $1,500, for a total price tag of $8,000 per system.
Modular peat biofilters offer advantages of size and treatment performance. Such systems would be appropriate for many sites located in sensitive areas, where soil conditions are limiting, or where space or other site limitations prevail.
Further study of the Puraflo[TM] system's performance over a two- or three-year period is recommended for a more precise measure of the Puraflo[TM] system's treatment efficiency.
This project received substantial assistance from the following persons: George B. Allison, Lt. Col. USAF (Ret.), Alabama Department of Public Health; Michael Spector, Ph.D., Associate Professor, Biomedical Sciences, University of South Alabama; Jackie Holliday, Area Environmental Director, Alabama Department of Public Health; and Christie White, B.S., P.S.C., Soil and Site Evaluator, Alabama Department of Public Health.
The project was financed primarily with federal funds from the U.S. Environmental Protection Agency under Grant No. X820884-01 awarded by the Gulf of Mexico Program Office. Frederick C. Kopfler, Ph.D. was the EPA Project Officer.
The contents of this article do not necessarily reflect the views and policies of EPA or the Alabama Department of Public Health, nor does the mention of trade names or commercial products constitute an endorsement or recommendation for use.
1. U.S. Environmental Protection Agency (1993), "Public Health Action Agenda for the Gulf of Mexico," EPA 800-K-93-001, Sept. 1993.
2. Puraflo[TM] Liquid Effluent Treatment System, Product Literature, Bord na Mona, Newbridge, Co. Kildare, Ireland.
3. CNN Newsroom[SM] Global View (1994), CD Software of global articles, atlas, and world facts, Softkey International, Inc., Cambridge, Mass.
4. Microsoft[R] Encara[TM] '95 (1995), CD Software, Interactive Multimedia Encyclopedia with Atlas, Microsoft Corp., Roselle, Ill.
5. Coffey, P., and R. Kavanagh (1989), "Specialized Peat-Based Waste Treatment Systems," In: Proceedings of the International Symposium on Peat/Peatland Characteristics and Uses, S.A. Spigarelli (ed.), May 16-20, 1989, p. 324-334, Bemidji State University, Bemidji, Minn.
6. Walsh, Joe (1995), "Personal Communication," Bord na Mona Environmental Products U.S., Inc., Greensboro, N.C.
7. McKee, J.A., and J.L. Brooks (1994), "Peat Filters for On-Site Wastewater Treatment," In: Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems, Dec. 11-13, 1994, Atlanta, Ga. p. 526-535, ASAE Pub. 18-94, Am. Soc. Ag. Eng., St. Joseph, Mich.
8. Brooks, J.L. (1990), "Peat as a Raw Material for Waste Water Treatment," Proceedings of Seminar at Bord na Mona Peat Research Centre, Newbridge, Co. Kildare, Ireland, May 1990.
9. Brooks, J.L. (1988), "Sphagnum Peat Offers Solution to Maine's Challenging Soils Small Flows, National Small Flows Clearinghouse, June 1988.
Samuel C. Robertson, B.S., M.P.A., Robertson Consulting, Prattville, AL 36067-1617.
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Environmental Health|
|Date:||Nov 1, 1995|
|Previous Article:||Preventing childhood lead poisoning.|
|Next Article:||Environmental health practitioners.|