An Evaluation of On-Site Technology in Indiana

Purdue University, 1997

Catherine Taylor, Joseph Yahner, and Don Jones

Agronomy and Agricultural and Biological Engineering

Introduction , On-Site Systems in Indiana , Public Health Issues

Potential Contaminants of Our Water Resources, Nitrogen Contamination

Effects of Pretreatment on Absorption Field Sizing , Constructed Wetlands , Sand Filters

Aerobic Package Treatment, Recommendations and Conclusions for Indiana

On-Site Systems Serving Individual Residences in Small Communities , Maintenance

Education, Inspection, and Certification , Institutional Changes , References


Soil has a remarkable capacity to purify wastewater using physical, biological, and chemical processes. In an on-site system, soil filters the wastewater, physically removing solids. Microorganisms in the soil decompose organic compounds, forming biomass, carbon dioxide, and soil organic matter. Soil gasses and the soil itself catalyze chemical transformations that detoxify harmful chemicals. However, this system of wastewater disposal has limitations. Biurgan and Sievers (1994) cited internal surveys in Missouri indicating that 70% of the on-site systems are "functioning improperly." They attribute much of this to soil limitations. Forty-five percent of the systems tested in Oakland County, Michigan, all of which served riverside homes, tested positive for "leakage of septic wastes into the river" when fluorescent dye released into the on-site systems was found in the river (EPA, The Water Monitor, 1996). Jacquez et al. (1991) cited a study estimating that upwards of one-half of existing soil absorption systems in the United States did not function adequately. Proper function means that neither untreated nor partially treated effluent is reaching groundwater or surface waters. It is important to remember that effluent need not reach the surface to indicate failure. A system that permits effluent high in N, P, or pathogenic microorganisms to reach water resources is also "functioning improperly." In Indiana, areas of permeable soils over shallow, unconfined aquifers are susceptible to this kind of generally unreported failure.

to top

On-Site Systems in Indiana

It is estimated that over one-third of Indiana’s population utilize on-site systems for their waste disposal, resulting in about 31 billion gallons of effluent per year. This number is growing rapidly as approximately 15,000 on-site wastewater disposal permits are granted each year in Indiana. If on-site systems average $4,000 each, the public is spending $60 million every year on on-site systems. Figure 1 is a map of septic system density by county, compiled by the U.S. 1990 Census.

Figure 1.

On-site systems can be a safe, effective, and economical means for wastewater disposal. Systems must be carefully selected, designed, constructed, and maintained. More stringent regulations (ISDH Rule 410 IAC 6-8.1) and alternative technologies, such as the elevated mound system, have decreased system failure in the last decade. However, even with the improved regulations, on-site systems have a finite lifespan without additional pretreatment and regular maintenance. Furthermore, systems built prior to rule 410 IAC 6-8.1 were frequently undersized, improperly designed or sited, or poorly constructed. Unfortunately, repairing or replacing these systems often requires technological alternatives not currently available or permitted in Indiana.

Purdue University On-site Project surveyed the county health departments to determine the status of on-site wastewater disposal in Indiana in the spring of 1997 (Appendix A and C). County environmental sanitarians were asked about the amount and reasons for on-site system failure. Failure estimates varied widely, with some estimates reaching 70%. Sanitarians ranked wetness problems (seasonal high water tables) as the most common and most important on-site system problem in Indiana. Undersized systems, system age and limited space followed closely. Groundwater contamination from treatment failures could not be estimated. This indicates a need to adapt new technology to overcome these limitations.

Indiana has over 450 small communities without municipal sewage treatment. Often these communities have small lots, each with their own well, and an old septic system often connected to stormwater drainage. On-site systems are normally thought of as intended for large, rural lots, but as shown in Figure 1 on-site systems are being used heavily in counties with rapidly growing suburban populations. A high density of individual on-site systems and private drinking water wells provides little margin for malfunction. Sanitarians regularly report encountering failing systems, tile drained systems, and systems discharging to surface water. Community municipal sewer projects to eliminate these failing individual systems are rare because the cost of building a conventional sewage treatment plant and large diameter gravity sewers frequently exceeds the assessed value of property in the community. Many residents are on fixed small incomes and cannot afford the sewer bill. Attempts to build alternative or experimental systems are often met with resistance from regulatory agencies and established engineering firms. Thus, Indiana continues to expand communities on lot sized on-site systems. When these systems begin to fail, the current options are usually: a connection to a central sewage treatment plant with connection to a municipal system which residents believe they can ill afford; condemn the homes; or, permit the continued existence of the public health hazards and properties that cannot be sold.

Pretreatment is a method of reducing wastewater strength, therefore lessening the burden of wastewater treatment placed on soil absorption fields. Such technologies have been retrofitted to failing systems, successfully renovating the soil absorption field. Pretreatment has potential for the many residences in Indiana which currently do not have adequate wastewater treatment and lack the space necessary for installation of a new absorption field.

In light of this, our project is focusing on evaluation of the impact of pretreatment of wastewater on on-site systems in Indiana soils. This report reviews representative literature and provides an overview of three types of pretreatment systems not currently permitted for widespread use in Indiana: constructed wetlands, sand filters, and aerobic package units.

to top

Public Health Issues

One of the basic tenents of public health is the protection of our water resources from contamination. A cursory look at the history of public health and the consequences of "drinking the water" in developing countries, demonstrates the repercussions and dangers of contaminated water supplies. Wastewater tainted water supplies can lead to diseases such as to meningitis, myocarditis, infectious hepatitis, dysentery, gastroenteritis, typhoid, cholera, and anemia either directly or via a vector organism. Contaminated water may contain pathogenic parasites including ascariasis, hookworm, trichuriasis, amoebiasis, and giardiasis (Harper, 1993, Mancl and Young). Disease causing bacterial organisms include Salmonella, Shigella, Escherichia coli, Vibrio, Leptospira, Mycobacterium tuberculosis, Campylobacter, Yersinia enterocolitica, Francisella tularensis, and Pseudomonas aeruginosa. Preventing contamination is increasingly important as microorganisms become resistant to many drugs relied upon in the recent past. Properly functioning on-site systems are a vital component to preventing disease. This is particularly critical when building occupants rely on a nearby well for their drinking water, a very common situation in Indiana.

"The literature contains many reports of disease outbreaks attributable to ground water contamination by septic system effluent and the pathogenic disease organisms it carries…. Effluent from septic systems is the most frequently cited source of ground-water contamination leading to diseases such as acute gastrointestinal illness, hepatitis A, and typhoid." (EPA, 1986)

Failing on-site systems generate health concerns beyond water contamination and contact with harmful pathogenic organisms and chemicals, such as NO3. They also produce stress on the environment. This may inhibit natural cleansing mechanisms and accentuate health risks of all kinds. Additionally, ponded wastewater from failing on-site systems provides a breeding ground for vector organisms. For example, ponded waste is the preferred breeding sites for the Culex pipiens mosquito, which is the vector for the transmission of St. Louis Encephalitis to man.

Public Health and environmental effects of contamination of our water resources have already been well documented (EPA, 1986). Therefore, the purpose of this paper is to focus on advancing the methods of on-site treatment and disposal to further protect water resources.

to top

Potential Contaminants of Our Water Resources

A common concern about on-site wastewater disposal systems is the release of nutrients and pathogenic microorganisms to the ground and surface water. The major components of potential concern in wastewater are biodegradable organic carbon measured as the five day biological oxygen demand (BOD5), total suspended solids (TSS), nitrogen as ammonia (NH3) and nitrate (NO3), phosphate (PO4) or phosphorus (P) concentrations, fecal coliform counts, and viruses. Other common parameters of interest are dissolved oxygen (DO), chemical oxygen demand (COD), total oxygen demand (TOD), pH, chlorine content (CL-), and iron levels (Fe).

Coliforms are indicator organisms for pathogenic organisms associated with sewage or fecal material. Coliforms include Escherichia, Klebsiella, and Enterobacter. They are measured as total or fecal coliforms. Total coliform measurements can be misleading because many of the organisms are naturally present in soil (Turco, 1994). Fecal coliforms can be distinguished from total coliforms by their ability to ferment lactose at 44.5oC. It is thought that organisms functioning at this high temperature are more likely to be derived from the warmer environment of the gastrointestinal tract of a mammal than the cooler soil environment.

to top

Nitrogen Contamination

Recently there has been increased awareness and concern about consequences resulting from the release of nitrogen to the environment. At concentrations of 10 mg/L in drinking water nitrates are considered to be a human health concern. Nitrites (and aromatic amines, nitros, and chlorate salts) can oxidize hemoglobin to methemoglobin, causing hemoglobin to lose its capacity to bind oxygen and carbon dioxide (Williams and Burson, 1985). Nitrites also combine with amines at the low pH found in our stomach to form nitrosamines, which are carcinogenic. Groundwater usually does not contain sufficient concentrations of nitrites and nitrates to warrant concern over human health. However, if a groundwater aquifer is susceptible to contamination, excessive NO3 may accumulate and become a problem.

Ammonia (NH3) is formed from the decomposition of organic matter and is the primary soluble nitrogenous compound contaminant in untreated domestic wastewater. The majority of ammonia removal occurs through a series of nitrification and denitrification reactions. Ammonia is in a pH dependant equilibrium with ammonium (NH4+).

NH3 + H2O <-> NH4+ + OH-

Standard procedures for measuring ammonia in wastewater utilize the Nessler reagent, which drives the reaction to the left, measuring the sum of ammonia and ammonium as ammonia (Ogden, 1994). In nitrification, ammonia is oxidized to nitrate in a reaction mediated by Nitrosomonas and Nitrobacter bacteria.

NH4+ + O2 --> NO2 + O2 --> NO3

This oxidation requires aerobic conditions and allows bacteria to assimilate carbon, the foundational atom for biochemical reactions. Nitrate is a frequent contaminant of ground and surface waters because is it mobile in soils. Denitrification, or the conversion of NO3 to N2(gas), is therefore the desired endpoint of the nitrogenous reactions.

organic carbon + NO3 --> CO2 + H2O + N2(gas)

Denitrification is mediated by anaerobic bacteria and requires an organic carbon source, which is readily available as a component of the wastewater.

Nitrogen removal has been evaluated in systems utilizing a wide variety of technologies. Separation techniques include blackwater holding tanks with graywater septic systems and dry toilets coupled with a graywater septic system. Source separation techniques typically achieve 60-90% nitrogen removal but require eventual offsite treatment (Whitmyer et al. 1991). Cation exchange, anion exchange and reverse osmosis, are examples of physical or chemical methods. Physical processes achieve >90% removal, but are typically reserved for large municipal wastewater treatment plans and industrial or hazardous chemical cleanup due to their complexity, expense, and high maintenance requirements. Extended aeration units, anaerobic/aerobic trickling filters, peat filters, RUCK filters, mound/constructed wetland combinations, and single pass and recirculating sand filters with anaerobic filters, carbon sources or rock storage filter attachments, are biologically based technologies. Biological systems achieve 50-60% nitrogen removal (Whitmyer et al. 1991). Extended aeration package plants using cyclical aeration/anoxic periods and recirculating sand filters coupled with anaerobic filters are two biological systems which remove more than 50-60% nitrogen. Biologically dependent systems usually show seasonal variation in nitrogen removal due to changes in temperature.

Traditional trench systems are thought to provide poor nitrogen removal because they first provide an anaerobic environment in the septic tank, and then an aerobic environment in the unsaturated soil. This is the opposite of the sequence desired for nitrification and denitrification. However, preliminary data has shown that even a shallow drainfield (12 inch) may reduce nitrate concentrations by 50% and fecal coliforms by almost 100% (Oakley and Ball, 1994).

to top

Effects of Pretreatment on Absorption Field Sizing

Pretreatment is a technology that has great potential to extend the lifespan of existing or newly constructed soil absorption fields. Our survey showed that system age was the third most common reason for on-site system failure in Indiana. Pretreatment decreases wastewater strength such that it has been shown to actually improve the functioning of absorption fields, instead of merely decreasing the burden and slowing failure (Converse and Tyler, 1994). Renovation of failing on-site systems may be possible because highly pretreated effluent does not form a biomat layer. Work on sand filters has shown that effluent with BOD5 and TSS less than 20 mg/L has not formed a biomat (Converse and Tyler, 1997). Absorption field sizing for high quality effluent should be in relation to the saturated hydraulic conductivity of the soil. References in the literature generally recommend multiplying the "untreated" load rate for septic tank effluent by a factor of 2 to 8 to determine loading rates (Venhuizen, 1995). This provides an option for the numerous small communities in Indiana with no municipal sewage treatment and lot sizes which do not provide space for installation of a new absorption field sized for septic tank effluent.

Effluent pretreatment will not alleviate the need for consideration of soil hydraulic properties throughout the soil profile. The present load rate tables consider soil horizons with greatly restricted hydraulic conductivity as "limiting layers" (rock, compact dense till, fragipans). A zone of 24 to 30 inches of permeable soil is required above a limiting layer. Effluent leaving a trench or other system component may move into the soil at a rather rapid rate and result in disappearance of water from the trenches. However, if it then encounters a more slowly permeable limiting layer, flow within the soil then becomes lateral rather than vertical and may result in a perched water table (Venhuizen, 1995). Systems on slopes utilize flow in the downslope direction to drain effluent from under the system. Horizontal flow to an unintended area, such as downslope to a ditch or water body, may circumvent treatment intended to result from passage through a depth of unsaturated soil. Seasonal high water table is another limitation which must be considered when sizing an absorption field for pretreated effluent. Pretreatment will not alleviated the need for elevated sand mounds, although it may allow for downsized mounds.

Wisconsin (Converse and Tyler, 1997, Tyler and Converse, 1994) has developed a somewhat conservative table based on soil texture and structure which can also be used to determine absorption field sizing of pretreated effluent. Additionally, Oregon state code allows loading rates of 2.1 gpd/ft2 for gravel, sand, loamy sand, and sandy loam, 1.7gpd/ft2 for loam, silt loam, and sandy clay loam, and 1.5 gpd/ft2 for silty clay loam, silty clay, sandy clay, and clay after effluent has been treated by a sand filter. Separation distances are 24in, 18in, and 12in for the three loading rates, respectively. Appendix E explains the basis of the current load rates for septic tank effluent in Indiana and illustrates how soil saturated hydraulic conductivity is related to potential downsizing as a result of pretreatment.

to top

Constructed Wetlands

Reed (1993) defines wetlands as "land where the water surface is near the ground surface long enough each year to maintain saturated soil conditions, along with the related vegetation." Marshes, bogs and swamps are all examples of naturally occurring wetlands. Without disturbances, such as flooding, natural wetlands eventually fill in and become upland areas. Most natural wetlands undergo periodic drying cycles, increasing their productivity. Uniform year-round water levels may actually reduce a marsh’s productivity (Hammer and Bastain, 1996).

Constructed wetlands (CW) are defined as "designed and man-made complex(es) of saturated substrates, emergent and submerged vegetation, animal life and water that simulates natural wetlands for human use and benefits" (Leszczynska and Dzurik, 1994). They have been used for wastewater treatment since the 1960’s in Europe. Other names for constructed wetlands include rock reed filters, vegetated submerged beds, submerged bed flow systems, root zone systems, microbial rock filters, and hydrobotanical systems. CW are used for municipal wastewater treatment, acid mine drainage, industrial process water, agricultural point and non-point discharges, stormwater treatment or retention, and as a buffer zone to protect natural wetlands. They have treated wastewater in environments ranging from the tropics to the colder climates of Northern China, Norway, and Northern Austria. CW treat wastewater using the following processes: filtration, sedimentation, physical or chemical immobilization, chemical and biological decomposition, and absorption and assimilation of excess nutrients by plants (Hammer and Bastain, 1996). Several review articles are available (Hammer and Bastain, Freeman 1996, Leszczynska and Dzurik 1994, Reed and Brown 1995, and Reed, 1993). Most of the following information on constructed wetlands comes from a technology assessment by the Environmental Protection Agency (Reed,1993).

Constructed wetlands are categorized as free water surface wetlands (FWS) or subsurface flow wetlands (SF). Free surface water constructed wetlands are usually earthen basins or channels filled with shallow water and emergent vegetation. They are thought to act similarly to a trickling filter, with a natural ubiquitous microbial population performing the majority of the treatment (Sandretto and Knight, 1991). Subsurface flow wetlands are similar in shape, but often smaller, and filled with porous media such as rock or gravel. Water is expected to stay below the top of the gravel, which provides physical support for plant roots. SF wetlands are often chosen for wastewater treatment because the wastewater is below the rock, decreasing the chance of exposure, odor and insect vectors. However, there is some evidence that CW themselves generate natural checks on insect vectors (Sandretto and Knight, 1991). SF may also be selected because the use of gravel allows smaller wetland cells. FWS wetlands may be preferred for lower strength wastewater (tertiary treatment), stormwater treatment (Green, 1996), or where nitrogen removal is required. In agriculture, FWS are used for high strength wastewater because the high organic matter content plugs the gravel in subsurface flow constructed wetlands. Although larger, FWS may have lower construction costs (and simpler hydraulics) than SF and would be more economically feasible in areas where land is inexpensive. Additionally, they are closer in nature to a natural wetland and provide a habitat for wildlife. FWS have been turned into parks and used for recreation.

It is typically thought that a CW is primarily anaerobic, with oxygenated microsites at the plant root surfaces. However, Gersberg et al. (1996) compared ammonia and BOD5 removal in a SF with and without plants to determine the amount of oxygen available for reactions. They observed removal rates of 94% and 96% in bulrush wetlands, 78% and 81% in reed wetlands, and 11% and 69% in unplanted wetlands, for ammonia and BOD5, respectively. From this they calculated that in bulrush wetlands, 120 mg/L of oxygen was available in the wetland. Other researchers, however, have found the amount of aeration in a CW bed questionable (Freeman 1996, Reed 1993).

There is no consensus historically on loading rates of constructed wetlands. Recommendations for SF have indicated hydraulic loading rates ranging from 0.74 to 3.44 gpd/ft2, hydraulic retention times from 1 to 6 days and organic loading rates from 3.6 to 160 lb./acre/d. Wetlands may lose significant portions of water to evapotranspiration, particularly in the summer. In Missouri, summer water loss averaged 74% with 60%, while winter losses averaged 13% (Burgan and Sievers, 1994).

In subsurface flow constructed wetlands, BOD5 is removed from a combination of the particulate matter settling in the rock media and treatment of soluble organic compounds by microorganisms on the rock and plant root surfaces. Kuehn and Moore (1995) reported that the populations with greatest activity were actinomycetes and fungi. BOD5 is actually produced by wetlands, resulting in a residual concentration of 2-7 mg/L. However, Green (1996) has reported BOD5 concentrations of <1 mg/L when treating tertiary sanitary effluent. Kadlec (1995) suggested typical CW effluent ranges of 5-15 mg/L BOD5 and 30-100 mg/L COD. When reviewing the literature, Reed (1993) found that influent BOD5 concentrations of approximately 7-50 mg/L produced similar effluent concentrations, all < 20 mg/L and most <12 mg/L, under a variety of environmental conditions. BOD5 removal had little correlation with hydraulic residence time after 1.5 days or with ratios of length and width. TSS removal shows a similar pattern with concentrations reduced below 20 mg/L, usually less than 10 mg/L, with loading rates up to 118 mg/L. Additionally, TSS removal was not correlated to hydraulic retention time (after 1 day) or length and width ratio. When evaluating two sites in Louisiana, it was determined that the BOD5 effluent concentrations were a result of BOD5 generated in the systems, instead of residue from wastewater. BOD5 mass removal correlates (r = 0.97) with BOD5 mass loading (kg/ha/d). Burgan and Sievers (1994) found that twice as much BOD5 was removed in the first wetland cell as the second. They cited other researchers reporting that the majority of BOD5 removal is in the first 20-40% of the wetland. Fecal coliforms concentrations typically drop by 1-2 log units with passage through a constructed wetland. Coliform removal can be increased with lower loading and in a SF, with smaller gravel. Phosphorous removal is usually low or erratic, due to the small amount of absorptive surface areas.

When utilizing wetlands for nitrogen removal, one should consider that wetlands themselves produce ammonia and nitrates, by converting a significant proportion of organic N to ammonia. Thus, wastewater containing organic solids and algae will produce ammonia, which then would need to be oxidized to nitrate before it could be reduced to nitrogen gas. Some nitrification occurs in a wetland, but it primarily denitrifies, due to its anaerobic nature. Half of the single pass systems Reed studied had increased concentrations of ammonia nitrogen in the effluent relative to the influent. Freeman (1996) found that reducing the loading from 1.6 gpd/ft2 to 0.55 gpd/ft2 only reduced the effluent ammonia concentrations from 130-150% of the influent ammonia concentration to 125% of the influent concentration. Recycling the water with sprinklers to introduce oxygen did not increase the ammonia removal rate. In a subsurface cell, ammonia removal was adequate (4 mg/L) at 0.2 gpd/ft2, however, designing SF at that low of a loading rate was deemed economically prohibitive. In northern New Mexico, Ogden (1994) explored the relationship between seasonal variations and ammonia removal. Removal rates were 39%, 83%, 63%, and 45% in the spring, summer, fall and winter respectively. Ogden concluded that fall senescence and spring leachates from leaf litter cause unpredictable increases in ammonia nitrogen.

Reed suggested that nitrogen removal can be accomplished, but that it requires available oxygen, a lack of algae, and longer hydraulic retention times (> 4-6 days). When comparing constructed wetland systems that successfully remove nitrogen to those that do not, it was noted that the depth of rooting relative to the depth of the bed was significant. It was found that most plants do not root as deeply as they are capable of, and thus limit the treatment area. Rooting depth may be increased by lowering the water level in the fall, but this has not been extensively tested in the US (Bastain and Hammer,1996). Another solution would be to make the wetlands shallower. SF are usually 1-2 feet deep. Total kjeldahl nitrogen (TKN) removal in 6 systems was not correlated to TKN mass loading (Reed, 1993). FWS demonstrated greater TKN removal than SF systems, probably because FWS may perform nitrification in the upper waters, due to wind aeration. Potential methods for increasing oxygenation include direct mechanical aeration, stages of overland flow, recirculation to an aerating device such as a sand filter, alternating flow between parallel CW cells to allow media aeration and increased root penetration, or deep open water zones the width of the bed. The later provides surface aeration, wastewater mixing and redistribution, and may be used as a habitat for mosquito fish (Sandretto and Knight, 1991).

Surface flow has been reported in several subsurface flow constructed wetland systems, and has been attributed to clogging due to BOD5 loading rates >0.5 kg/m2/day. However, upon closer inspection, it was found that in most cases surface flow is a result of inadequate hydraulic gradient, not clogging (Reed, 1993). Designs should take into account Darcy’s law and the outlet device should be adjustable and low enough to allow drainage of the wetland if necessary. CW are best described by plug flow models. At one site that experienced significant clogging, analysis showed the clogging material to be composed of silica, clay minerals, and limestone dust. Five other sites showed clogging ranging from 2-6% of the pore spaces. Clogging material was >80% inorganic and usually could be easily washed off. Percent clogging typically did not increase over time, leading to the conclusion that it was a result of construction practices or using unwashed granular materials. Due to the misperception of system clogging, larger rock has often been used. Reed recommends smaller rock because it has a greater surface area, is a better rooting media, and will produce a more laminar flow.

The most commonly used CW plants are bulrush (Scirpus), reeds (Phragmites), and cattails (Typha). All of these plants work if plant root depth is considered in the design. Soft tissue ornamentals are not recommended. Ideally, plants will be planted in water (Whalen et al., 1996) in early spring. Initial plant survival has been very low (18-22%) when plants were planted in dry conditions without watering (Sandretto and Knight, 1991). There is a discrepancy in the literature as to whether monocultures remain monocultures. In one study (Hammer and Bastain, 1996), it was found that 2 years after planting 4 species in a CW, 40 species were present. There is a general agreement that plants should be regionally adapted. Annual harvesting is not necessary unless ornamentals, such as lilies, are used. Debris removal every 3 to 5 years ordinarily will suffice. Surface plant litter may also act as insulation in colder climates. Raising the water level after planting usually controls weeds associated with wetland startup.

It takes 2-3 years for a CW to "mature" to its full treatment capacity. Wetlands larger than those serving single family homes are usually divided into two or more cells. The literature is inconclusive as to whether wetlands cells should be in series for more complete treatment or in parallel to allow alternation between cells for treatment and maintenance. Reed’s report describes three sets of design procedures and recommends one theory. Steiner and Freeman (1988) provide an understandable overview of design considerations and Small Flows has a constructed wetland information package intended to help engineers design CW for individual residences (Small Flows).

Five years ago, wetland construction costs averaged about $87,000/acre for a SF and $22,000/acre for a FWS (Reed,1993). Other estimates have included $1.99/gal/day for a SF, a figure which was assumed to be inflated by 10-20% due to inexperience of the contractor (Whalen et al.,1996), $0.61/gal/day for a SF and $0.77/gal /day for a FWS wetland (Reed and Brown,1995), and $1.69/gal/day (Gersberg et al., 1996). FSW systems are sized about 70% larger if sizing is based on BOD5 removal. The major factors influencing cost are availability of appropriate washed gravel and land costs. When sizing CW for nitrogen removal, FWS may be more economical due to the greater retention time desired and the greater surface oxygen exposure.

There are still many unknowns in constructed wetland design and optimization (Sievers, 1996). Large quantities of data have been generated from many sources, but it is often difficult to sort and distill. Inconsistent results are easily found due to the quantity of data (at least 4,000 articles and reports were produced by 1995) and rapidly changing design theory. In light of this, the EPA has developed a "North American Wetland Treatment System Database" to keep track of engineering data on constructed wetlands (Knight, 1994). This Dbase IV database includes parameters such as location, wastewater type, wetland type, number of cells, flow, area, plant species, depth, aspect, costs, and effluent quality from published and unpublished sources. The average reported wetland treatment efficiencies were 71% removal of BOD5 and TSS, 46% NH3, 54% total N, and 46% removal of total P. The database will run under DOS on a PC and can be obtained from the EPA in Ohio by contacting Don Brown at 513-569-7630. It requires 4 MB free disk space, and can be sent by email or on a 3.5 inch floppy disk.

The problem of variable data has been felt by numerous engineers and agencies. In response to concerns by the Louisiana State Department of Health over continued installations of lagoon/rock-reed systems without uniform design specifications or analysis of accumulated data from the present systems, Griffin et al. (1994) evaluated the data from 19 systems within the state. They drew similar conclusions about the inability to compare systems or isolate positive and negative design parameters even within a specific system. They chose to evaluate the systems with respect to their discharge requirements. The criteria for a successful system were that it should not exceed its discharge requirements for TSS and BOD5 more than 20% of the time. Requirements ranged from 10-30 mg/L BOD5 and 15-30 mg/L TSS. Two facilities met the criteria for success. Nine systems violated their BOD5 permits over 20% of the time, five of which violated it at least 60% of the time. Seven systems violated their TSS requirements over 20% of the time and two violated it 60% of the time. Of the systems with TSS and BOD5 discharge requirements of 20mg/L, 3 of 5 violated BOD5 and 4 of 5 systems violated TSS over 20% of the time. The system with the best performance served a hospital. Overall, performance was erratic and poorly described by averages.

Wetland performance variability was studied in Oregon by evaluating adjacent wetland cells receiving the same pulp mill effluent (Kuehn and Moore, 1995). The treatments were the following: a SF with no plants and a 2-day effluent retention time, a FWS with bulrushes and a 2-day effluent retention time, a FWS with cattails and a 2-day effluent retention time, a FWS with cattails and a 10-day effluent retention time, and cattails in a FWS with a 10-day retention time receiving only water. Influent concentrations were low, averaging 20 mg/L BOD5 and 11 mg/L TSS. Bulrush effluent values for BOD5 and TSS were always lower than those of cattails, but significance was not established if all treatments were considered. Data suggested that seasonal variation was greater than the differences in treatment. Matched CW pairs produced effluent with similar quality. Oxygen was usually below detection limits in the summer. They then lowered the influent quality (>40 mg/L BOD5). This resulted in initial data suggesting that the 10-day retention time had significantly greater treatment capabilities in the winter and spring. BOD5 generally increased in the summer at both influent concentrations and retention times, and fall data was not available due to destruction of multiple wetland cells by nutria. TSS removal was greater in cattails than bulrush cells. However, results are difficult to interpret because influent concentrations were so low. On a concentration basis, effluent quality was high.

Mosquito populations in two municipal constructed wetlands were studied in Kentucky (Tennessen). They found that 53.7% (site 1) and 45% (site 2) of the "dipper" samples collected contained mosquito larvae. Similar species dominated both natural and constructed wetlands. A count of approximately 0.25 larvae/dipper sample has been correlated with high levels of "adult (mosquito) annoyance" and requests for insecticide applications in populated areas. Concentrations of 1.7 to 272 larvae/dip were found at the two constructed wetlands evaluated. An experiment was performed where mosquitoes landing on a subject within a 15 minute period were collected and counted. At 7:00 p.m. on the edge of one constructed wetland, 54 C salinarius mosquitoes were collected. Constructed wetlands had higher mosquito populations than natural wetlands, possibly due to constant large inputs of organic matter.

Land area and scale are other important considerations. Wetlands can naturally clean up many types of pollutants. Natural wetlands performing this role may occupy large areas and often have many smaller ecosystems performing a variety of tasks within them. Scaling these down to a uniform, low maintenance system that will fit in a person's backyard can present challenges (Bastain and Hammer, 1996).

Maintenance activities have included cleansing of the influent pipes with a tanker hose, harvesting and composting of plants, replanting, weeding, fertilizing, brushing of inlet and outlet weirs and screens, plastic liner patching, frequent turtle and nutria removal, duckweed control, installation of fences to prohibit entry of deer, fish stocking, insecticide applications, complete draining of the wetland cells, raising and lowering the water level to control weeds, increase root growth, and increase nitrification, and additions of insulation in the fall. Like all treatment systems, maintenance is required.

The advantages of constructed wetlands include inexpensive capital and maintenance costs, ease of maintenance, relative tolerance to changes in hydraulic and biological loads, and ecological benefits. Disadvantages include large land area requirements, lack of a consensus on design specifications, complex physical, biological, and chemical interactions providing treatment, pest problems, and topography and soil limitations (Hammer and Bastain, 1996).

Natural wetlands are a complex, diverse ecosystem. It may be useful to compare wetland performance to farm performance. Both are attempts to manipulate complex, productive, but often unpredictable, natural systems for human purposes. Both have been managed with favorable and varied results. Further research within the state needs to be performed in addition to the development of maintenance procedures and policies prior to widespread permitting.

to top

Sand Filters

Sand filters have been used for wastewater treatment for over 100 years. Mancl and Peeples (1991) looked at data on 26 single pass community sand filters collected by the Massachusetts State Board of Health from the period of 1892-1937. They found that these filters were loaded at approximately 1.01 acres/1000 people to 1.89 acres/1000 people and with rates of 0.48 gal/ft2/day to 2.77 gal/ft2/day. Sand filter depth ranged from 3-8 feet and with sand sizes of 0.06-0.34 mm for 25 of the 26 filters. Seven filters treated raw wastewater, while the other filters had some form of pretreatment, often a septic tank. Of the seven systems discussed in detail, six systems had effluent BOD5 <20 mg/L, with four systems reporting BOD5 <10 mg/L. The one system with higher BOD5 concentrations dropped to below 20 mg/L over time between years 3 and 7. After 39 years, four systems had lower effluent BOD5 than in the earlier years, while three systems showed an increase in BOD5. Significant nitrification occurred, although it did decrease some over time. Sand filter effluent usually contains 5-10 mg/L BOD5 (and TSS) and <1000 counts/mL fecal coliforms (Converse, 1997). Single pass or intermittent sand filters (ISF) typically reduce nitrogen by 30-50% (30 mg/L average), while recirculating sand filters show nitrogen reductions around 60-70%.

Sand filter installations have rapidly increased recently, particularly in the Northwestern states where over 10,000 single pass sand filters and 200 recirculating sand filters (RSF) have been installed since 1976 (Ball, 1995 (a)). Drainfields following sand filters, under observance for 17 years, have shown that "biomat formation does not occur." This allows significant decreases in absorption field size and depth and reduces the need for gravel. One community system has loaded their absorption field at the rate of 16 gpd/ft2 (bottom area) for 10 years.

Loudon reported that an absorption field in sandy loam to clay loam fill material over a poorly drained loam has been receiving recirculating sand filter effluent successfully for 10 years without signs of clogging or biomat formation. A trench at lower elevation was ponded only when the soils were wet due to a high water table or rain. System modifications over the 10 year period included removal of the top layer of sand three times and the addition of a six inch layer of pea gravel, which replaced the sand, increased water infiltration, and decreased icing. Loudon also reported research on a two-year old recirculating sand filter system in a somewhat poorly drained clay loam soil that would not have been approved for septic system installation. Sand replaced gravel as a more economical trench fill material. Sand was adequate because of the lack of biomat formation. Trenches were successfully loaded at 0.8 gpd/ft2. Loudon found that absorption fields loaded at about eight times the recommended rates for septic tank effluent were successfully accepting all recirculating sand filter effluent. Roy and Dube (1994) studied the potential of recirculating gravel filter use in Quebec. Gravel was used instead of sand to minimize heat loss, reduce vegetation growth, and reduce maintenance. BOD5 and TSS concentrations were below 6 mg/L. Total nitrogen removal averaged around 47% and varied with temperature. Trenches located in silty sand were loaded at 6.9 gpd/ft2 and showed no evidence of clogging. The system functioned well even with winter temperatures of -25oC.

Recirculating sand filters are being used in Anne Arundel Co., Maryland, a lowland county on the Chesapeake Bay (Piluk and Peters, 1994). Average RSF effluent concentrations are 5 mg/L BOD5, 8 mg/L TSS and 20 mg/L total nitrogen. Fecal coliforms counts ranged from 240 to 9.5x104 cfu/100mL. RSF are being used as retrofits to failing systems, which are commonly serving homes in waterfront communities on sites now considered unacceptable for conventional systems. RSF costs ranges from $5,850-10,000. Bruen and Piluk (1994) tested the effect of a 50% reduction in sand filter size. The RSF system was loaded at 13.6 gpd/ft2, and trenches equipped with a chamber system were loaded at 5.4 gpd/ft2 resulting in variable ponding and occasional overflow. RSF effluent averaged 13 mg/L BOD5 and 9 mg/L TSS. At another site, gravelless trenches were loaded at 5.7 gpd/ft2 with minimal ponding (0-2 inches). Effluent at this site had 6 mg/L BOD5 and TSS and 2.1 x 104 MPN/100 mL (Bruen and Piluk, 1994). Absorption field size reductions reflecting those successfully tested in the field were recommended.

In a replicated study, Gold et al. (1992) compared recirculating sand filters and RUCK filters (buried single pass filters with alternating layers of sand (25 cm) and gravel (5-10 cm)). RSF received a hydraulic load of 0.9 gpd/ft2 and the RUCK filter 1.9 gpd/ft2. On any filter, intermittent ponding was observed only within a short period after dosing throughout the 3-year observation period. The RUCK filters received no maintenance, while the RSF were seasonally weeded, raked, and winterized to avoid icing. Total N removal was 20% and 8% in the RSF and the RUCK filter respectively, while TKN was about 73-74% in both filters. However, the RSF was almost twice the size of the RUCK filter. Both filters produced effluent with BOD5 concentrations less than 5 mg/L. Fecal coliform removal varied seasonally with greater removal occurring in warmer months and in the buried sand filters in all seasons. RUCK filter effluent fecal coliform concentrations averaged 1.5 cfu/100 mL in the summer and 79 cfu/100 mL in the winter. Fecal coliform concentrations in the RSF effluent averaged 3.1 cfu/100 mL in summer and 3136 cfu/100 mL in winter. F-phage reductions also varied seasonally with reductions of 1.93-5.58 log units. Removal of microbial indicator species was correlated with reductions in pH. Gold suggested that acidity may be the product of enhanced microbial activity.

Three recirculating sand filters were tested in Oregon (Anonymous, Small Flows) using an actual average filter loading rate of 1.45 gpd/ft2 (systems were designed for 3.125 gpd/ft2). They found that the filter surfaces collected debris and served as a habitat for weeds, algae and mosses. Similar growth was found on the distribution troughs when neglected. Litter had to be removed every fall. The filter surface was weeded and troughs cleaned every six months. To determine the effect of negligence, one system was not maintained. In this system, the troughs overflowed, but the sand filter accepted all the effluent. A wooden box with screens and cloth was then constructed and placed over all the systems. Homeowners failed to realign displaced filter troughs resulting in sand filter loading in concentrated areas. Despite this neglect, filter effluent quality was high averaging 2.7 mg/L BOD5, 3.8 mg/L SS, 29.9 mg/L NO3, 0.45 mg/L NH3, 1.1 mg/L TKN, and 8.5 x 103 cfu/100 mL fecal coliforms. Systems were located on soils with "severe" limitations due to shallow a groundwater table or rocky layer, or very slow permeability. Trenches accepted effluent at loading rates of 1.89 gpd/ft2 (silty clay over basalt saprolite and clay), 2.5 gpd/ft2 (silty clay loam over clay) and 2.89 gpd/ft2 (silty clay loam) even during the Oregon rainy season which lasts 3-5 months. No date was located on the paper previously described, however, current recirculating sand filter designs do not use troughs and are often covered by filter fabric and large decorative rock, allowing oxygen to reach the system.

Ronayne et al. evaluated the performance of three types of single pass sand filters using 2 feet of medium sand in Oregon. One design consisted of two filter cells, another utilized one filter cell, and the third was built in an unlined trench dug in saprolite. Effluent was delivered to all filters via pressure distribution. Six of the seven systems tested had design loading rates of 1.23 gpd/ft2. No maintenance was needed. Effluent quality (mg/L) averaged 3.2 BOD5, 9.6 SS, 29.1 NO3, 0.25 NH3, 1.7 TKN, and 407 cfu/100mL fecal coliforms. TSS concentrations measured at 9.6 mg/L may actually have been lower because solids contaminated samples due to an inappropriate sampling area. Trenches located in very limited soils and loaded at rates of 2.3 gpd/ft2 to 7.7 gpd/ft2 performed well. It was noted that fecal coliform removal was about 3 log units in the single pass sand filters, while it was about 2 log units in the recirculating sand filters described in the above paragraph.

Ball (b) tested a system that included a recirculating trickling filter (RTF) attached to septic tank followed by an intermittent sand filter (ISF) in an attempt to reduce nitrogen levels. The RTF, which was 21 inches in diameter and 36 inches deep, was cycled at rates low enough (2.5 gal/min.) to maintain an anaerobic environment in the septic tank. After trickle filtration the effluent concentrations (mg/L) were 23 BOD5, 9 SS, 10 TKN, 4 NH3, and 10 NO3. Total N was reduced by 77%. Continuous recirculation was found to be necessary for nitrification. When the RTF was retrofitted to several septic tanks, nitrogen reduction was not stable and cost of running the pump continuously was high. Replacement of media with an open-cell foam allowed intermittent pumping and resulted in 18 mg/L BOD5, 17 mg/L TSS, 11 mg/L TKN, 5.6 mg/L NH3, and 4.1 mg/L NO3. After passage through an ISF, the BOD5 and TSS concentrations were <1 mg/L with a total N concentration of 10 mg/L. Nitrogen concentrations actually increased with passage through an ISF. To combat this, the IFS was replaced with an upland filter. The upland filter design started with a chamber cell, upon which was layered 12 inches of 13 mm rock, 6 inches of 2 mm sand, 6 inches of medium sand, and a 1 inch slotted distribution pipe in 9mm pea gravel. Filter fabric was placed on top of the pea gravel and pipe. Loamy sand was then backfilled to the ground surface. When proceeded by a RTF, the upland filter effluent concentrations (mg/L) were 8 BOD5, undetectable TSS, 2.9 TKN, 1.2 NH3, and 2.5 NO3. After 12 months of usage, the upland filter media was black but no clogging was found. Rock was replaced with 2mm sand for further research. Research results are promising, but still in the preliminary stages.

Cagle and Johnson (1994) monitored 44 single pass sand filters utilized at sites with shallow soils, fragipans, claypans, fractured rock, and/or high water tables in Placer Co., California. Single pass sand filter effluent characteristics (mg/L) averaged from 30 filters were 2.17 BOD5, 16.2 TSS, 31.1 NO3, 4.6 NH3, 5.9 TKN, 37.4 TN and 11.1x102 cfu/100mL fecal coliforms. These averages included systems that were neglected and abused by the homeowner. A similar study in Oregon produced the following results (mg/L): 3.2 BOD5, 9.6 SS, 29.1 NO3, 0.25 NH3, 1.7 TKN, 30.3 TN, 407 cfu/100mL fecal coliforms. BOD5 (and SS) concentrations in Oregon are presumed to be higher because monitoring began prior to use of screens on the pump vaults.

Regulatory agencies need to incorporate all "stakeholders" into decision making processes regarding new technologies (Cagle and Johnson, 1994). This promotes education and understanding by all parties. Cagle and Johnson pointed out that without this, scientifically sound technologies are often not utilized correctly or at all. Additionally, they recommended instituting operation permits or management zones to provide the required maintenance as "homeowners in Placer County (as a group) are neither schooled nor inclined to provide needed maintenance and operation services for their ISF ‘on their own’." Maintenance was the single most important factor in determining sand filter success.

Sand filter annual maintenance includes the following: (from Converse, 1997)

  1. "Monitor solids and scum build up in septic tank.
  2. Monitor buildup of solids in the retention basin. Sand filters may slough slimes and solids which will end up in the recirculation tank or in the pump chamber. Clean the vault screen in the recirculation tank.
  3. Flush all the laterals
  4. Monitor pressure in laterals. If it is considerable different from initial measurement, unplug orifices.
  5. Observe ponding at media/aggregate interface in the filter through observation tubes.
  6. Observe ponding in the observation tubes for soil dispersal unit.
  7. Monitor water appliances for leaks on a monthly basis and repair as needed.
  8. Protect the gravel filter area and dispersal area from heavy equipment, excavations, etc. Minimize accumulation of organic matter such as leaves on surface of filter."

Technology is advancing but design parameters are now available (Converse, 1997, Bounds, 1994). The Oregon Department of Environmental Quality has worked extensively with sand filters and has relatively up to date regulations. In addition to maintenance, two possible limiting factors for widespread sand filter use in Indiana are the availability of watertight septic tanks and appropriately sized filter media. Accurate dosing and recirculation depend upon a watertight septic tank. Researchers are finding that these tanks are surprisingly rare. Oregon has developed regulation ensuring watertight, structurally sound tanks.

to top

Aerobic Package Treatment

Aerobic package plants or aerobic treatment units (ATU) are umbrella names for various technologies employing aeration. The EPA (1980) describes aerobic treatment as involving suspended or fixed growth processes. Activated sludge is a process by which part of the sludge is recycled back to the unit to "seed" the unit with acclimated degrading microorganisms. Fixed film processes provide a large surface area that becomes a habitat for microorganisms that treat the wastewater they come in contact with. Units may follow a septic tank or incorporate an anaerobic settling area into the design. Aeration units usually discharge to a soil absorption area. However, depending on state regulations, aerobic units may surface discharge after chlorination and/or sand filtration. (Indiana administrative code prevents the discharge from individual residential sewage disposal systems to the waters of the state.) Most available units primarily rely on activated sludge treatment. They may have a filtration component in a clarification or outlet area. The remainder of the discussion will pertain to activated sludge type systems unless otherwise indicated. Specific designs vary and are usually patented or have patented components. The primary advantage to aerobic treatment units is that they are able to provide highly pretreated effluent in a limited space. The primary disadvantage is that they require professional maintenance on a regular basis. Other considerations include the electricity necessary to run the plant, problems with large or small flows, and maintenance expenses.

Aerobic package treatment has potential for allowing renovation of failing systems, in addition their availability for new construction. Converse and Tyler (1994) added aerobic treatment units to failing on-site wastewater disposal systems in Wisconsin. After addition of the 15 units, one system was still overloaded and two others needed pumping for the first 6 or 9 months. The one overloaded system was very undersized. The rest of the systems accepted all the wastewater. They concluded that in general, renovation was successful. Aerobic package units have also been successfully tested and recommended in Illinois (personal communication). In Texas 81 aerobic units were installed around an environmentally sensitive recreational lake area (Carlile, 1994). No soils in the area qualified for a conventional system without restrictions, 40-50% qualified for an alternative system, 34-44% were not recommended for development. After wastewater passed through a septic tank and an aerobic treatment unit, it was chlorinated and dispersed with trickle irrigation or surface irrigation on larger lots. Systems performed well during the 2 years of monitoring, even on 5000 ft2 lots. Sandy soils received 1 to 1.5 gpd/ft2. Chlorine tablets were effective, but required replacement every 3 months. Professional maintenance is required and operational costs are higher than most other systems. Carlile compared data to a study performed by the Trinity River Authority evaluating aerobic unit effluent from systems that had been in use over 2 years. The Trinity River Authority found ATU effluent concentration averages of 8.1 mg/L BOD5, ranging from 1.6 mg/L to 46.8 mg/L and 13 mg/L TSS, ranging from 2.4 to 98 mg/L. Eighty-three percent of fecal coliform samples had concentrations less than 200 cfu/100 mL. Data was also compared to another study of 9 systems by the local county health department indicating an average BOD5 of 4.3 mg/L with a range of 2-8.7 mg/L and an average TSS of 4.2 mg/L with a range of 1 to 7.6 mg/L.

Others have found treatment efficiency to be more variable. Hanna et al. (1995) looked at ATU performance in Virginia. At the time of the study, Virginia did not require a licensed operator and ATU discharged directly to surface waters. When evaluating the performance of 5 ATU, 4 of the 5 units had replacement motors. One of the 4 systems had its motor replaced 6 times due to cotton gauze catching on the motor. Improper landscaping by homeowners resulted in the submersible pump shorting out at one site and the aerator being unplugged for two days at another. One homeowner had removed the alarms and another had removed the outlet filter. Two systems had no dechlorination tablets throughout the study, and two systems did not have chlorination tablets at one of three monitoring visits. In 4 of 15 observations, chlorination tablets were caked and not in contact with the wastewater. BOD5 averaged 70 mg/L and TSS averaged 99 mg/L, with 81% and 64% of BOD5 and TSS samples, respectively, exceeding the discharge requirement of 30 mg/L. Nitrification was variable, and 35% of samples exceeded fecal coliforms discharge requirements of 200 cfu/100 mL. Aerobic treatment unit optimization experiments indicated that shortening the hydraulic retention time from 3 to 5 days to 1 day would increase treatment. They suggested baffled chlorinator chambers, the avoidance of the use of dry chemicals which cake in humid environments, and the use of flow equalization and alarms. Additionally, Hanna recommended policies ensuring maintenance and pumping, increasing the frequency of required grab samples from one to four times a year, and further treatment with either sand filters or absorption fields.

Fixed film and membrane technologies continue to show promise, however, application of these treatments to residential sized units is still being optimized (Chiemchaisri et al. 1993, Velioglu et al., 1988). Japan is subsidizing the installation of a submerged biofilter fixed film technology (Watanabe et al., 1993). The system has two anaerobic compartments or filters, an aerobic filter, a sedimentation tank, and a disinfection tank. In field tests, this system produced BOD5 and TSS less than 20 mg/L. Modifications providing flow equalization and recirculation are being investigated to increase nitrogen removal. Jowett and McMaster (1994) reported on the performance of an aerobic biofiltration system in Canada. They used plastic concave media which provides large surface areas, porosities, and retention times. Wastewater moves through media by capillary action in an unsaturated environment. Initial laboratory experiments loaded filters at 19 gpd/ft2 to compare plugging potential against sand and peat media. No clogging was observed in the synthetic filters after 27 months of operation with domestic and industrial wastewater at 2-19 gpd/ft2. The peat and sand filters plugged within 2-3 weeks when loaded at 19 gpd/ft2. Prior to the addition of forced aeration to the synthetic filter at 7 months, the system’s effluent concentrations averaged 14 mg/L BOD5, 10 mg/L NH4, 25 mg/L NO3, and 2.4x103 cfu/100mL fecal coliforms at the 19 gpd/ft2 loading rate. Forced air dropped BOD5 to 1.7 mg/L, TSS to 5 mg/L, and fecal coliforms to 2.3x102 cfu/100mL. Ammonia decreased to 0.04 mg/L and NO3 increased to 34 mg/L signifying almost complete nitrification. At a loading rate of 2 gpd/ft2 fecal coliforms dropped to 28 cfu/100mL. However, field evaluation of three systems showed difficulties associated with winter operation. Problems involved system freezing and cold or dilute septic tank effluent at air temperatures of -35 and -40oC. Freezing and thawing pipes caused biomat clogging and ponded water. Other systems had decreased or variable performance.

to top

Recommendations and Conclusions for Indiana

On-Site Systems Serving Individual Residences in Small Communities

The continued expansion of residential subdivisions on one or two acre lots should be greatly restricted. As an example, a fifty acre site each with one acre lots places fifty on-site wastewater disposal systems in close proximity to fifty wells all connected to groundwater supplies. Enlarging lot size only creates "sprawl," often spreading rural homeowners into valuable farmland and natural resource areas. This generates land use conflicts, and typically increases the cost of land and local services such as roads, transportation, fire protection, police protection, and schools. Cluster development with community on-site systems accompanied by guaranteed, organized maintenance is a safer, and often a more economical solution.

to top


Programs and policies assuring maintenance should be set up for any system utilizing pretreatment to ensure acceptable operation. Any innovative technology cannot be recommended without the development of programs and procedures which will guarantee maintenance. Attempts to short circuit this requirement will most probably result in increasingly expensive failures and further frustrations by homeowners, contractors, and regulatory personnel in addition to rejection of technologies that would otherwise be appropriate. Although a review of strategies to ensure maintenance is beyond the scope of this report, ideas which may be considered include issuing operating permits, developing regional sewage management districts, tracking maintenance with smart cards similar to ATM cards, and requiring that homeowners keep a record of maintenance, to be part of a normal real estate transaction process when properties are served by individual on-site systems.

to top

Education, Inspection, and Certification

Success of these increasingly complex systems often depends upon contractor understanding and proper system installation. This raises the issue of engineer, installer, and operator training. Contractors must understand not only the basis of how to design a system, but they should also understand the reasons behind the design theory. A learning curve should be expected when any new technology is introduced. However in the long term, failure due to uninformed contractors can and should be prevented. Delaware has instituted a model program for training pertaining to sediment and stormwater management (IECA News, 1996). Contractors attend a four-hour training course every three years. Multiple agencies and associations conduct the training sessions throughout the year. In four years, about 2,800 contractors have attended. The results have been improved designs, greater compliance, and better field implementation throughout the state. Additionally, innovative or alternative on-site systems require increasingly rigorous inspections. Functioning of all system components should be verified at installation and repair for homeowner protection.

Certification of all on-site system service providers should be considered. Site evaluations by soil scientists are the foundation of subsequent on-site system design and installation. An inaccurate soil evaluation can negate all other attempts to construct and maintain an effective treatment system. After the site evaluation, proper system design and installation are essential. Purdue regularly receives calls asking for help with systems that are failing due to poor work performed by on-site professionals. Highlights of recent conversations include a curtain drain which has been installed so that it "drains" uphill, a repair system constructed so the trenches are in a depression, a new system which has level trenches not constructed on the contour, and a lot no longer suitable for a system because the location of the house and well were changed without regard to the need for an on-site absorption field. Certification should also be considered for individuals who provide for the operation and maintenance of the on-site systems. The more complex systems will require that the operator be well versed in the biological, as well as the mechanical, aspects of system performance for each type of system.

Certification will not overcome all of Indiana’s problems, but it does provide evidence that on-site professionals meet a minimum level of expertise. It also serves as an avenue to inform and train personnel. Many counties have already recognized this need and require certification of on-site system contractors and soil scientists. This may be done more effectively on a statewide basis. Recently the Indiana Land Improvement Contractors Association (LICA) board has voted to consider a voluntary certification proposal. For soil scientists, Indiana has a voluntary certification program through the Indiana Soil Classifiers Association already in place. The Soil Scientist Society of America also provides nationwide certifications through ARCPACS (American Registry of Certified Professionals in Agronomy, Crops, and Soils). The Indiana Soil Classifiers Association currently requires an examination for certification, and ARCPACS will begin requiring an examination on January 1, 1998.

to top

Institutional Changes

A significant barrier to on-site wastewater disposal reform is the fact that regulation is split between state agencies. At the state level, individual residential and privately owned commercial and cluster systems are under the jurisdiction of the Indiana State Department of Health (ISDH). Discharging systems and all on-site systems which are proposed by municipalities, sewer districts, or conservancy districts, are regulated by the Indiana Department of Environmental Management (IDEM). This system worked fairly well when on-site technology consisted primarily of conventional soil absorption systems, which governmental entities typically did not propose for use. However, new technologies and their accompanying abilities are blurring the lines of separation. In a report to Congress outlining the barriers to decentralized systems, the EPA outlined regulatory difficulties as one of the five major barriers.

"Legislative and regulatory constraints -- State laws usually divide oversight of centralized and decentralized systems between two or more agencies, resulting in confusion about and less emphasis on decentralized systems. And many state and local governments have codes for decentralized systems that allow only conventional septic systems, or have complicated processes for approving alternative onsite systems. Solution- States should consider consolidating all legal authority for centralized and decentralized systems under one state agency, and state and local regulatory coded for allowing decentralized systems should be revised." -Small Flows, Summer 1997

Indiana provides a textbook example of these difficulties. For instance, current regulatory organization makes it difficult to repair on-site systems serving small communities if the repair includes the use of innovative or alternative wastewater treatment technologies. Purdue’s survey revealed 470 small communities without municipal sewage facilities in the 70 counties responding. These communities usually do not have the economic resources to build a conventional municipal sewage plant. Purdue knows of at least one town that has a community "on-site system" with no absorption field. Due to the expense of building a conventional sewage treatment plant and the regulatory difficulties associated with alternative technologies, this community and others feel that for now, meeting wastewater treatment standards is unattainable. Instead of taking a sequence of small steps to achieve the desired level of treatment, an option available with alternative technologies, nothing is done.

On-site technologies researched for residential use often perform as well or better on a system scaled for a cluster or small community. One technology which might be appropriate for community on-site systems is spray irrigation. Pennsylvania has permitted this technology on sites unsuited to other technologies (McIntyre et al., 1994). Wastewater is applied in the morning at rates ranging from 10mm/week to 2.54 mm/wk depending on the soil conditions. Wastewater is pretreated extensively prior to irrigation, passing through either dual septic tanks or an aeration tank, a sand filter, and chlorination. Large buffer zones are required for citizens piece of mind, however "there have been NO documented cases in which the spray effluent of properly treated wastewater has caused ANY health problems" despite numerous studies by several public and private institutions (including the EPA) to determine health risks (McIntyre et al., 1994). Although this technology may be a feasible economic solution for small communities (Canody, 1997), it is currently not utilized largely because of regulatory barriers. The treated wastewater is applied to a soil area for final polishing, but is classified as a discharging system because of the spray irrigation component in Indiana. Wastewater would meet discharge requirements after treatment by the soil, but the technology is often nullified because it does not meet discharge requirements in the middle of the treatment process.

The institution of management and maintenance plans, required for effective use of pretreatment, provides another example of the need for regulatory consolidation. One of many effective ways of providing on-site system maintenance is to develop regional management districts. Small flows has an information package on management districts. Residences and companies within the district pay into a fund which in turn pays contractors to maintain the systems. Examples of this principle at work are found in regional sewer and trash removal districts. A district committee makes decisions about expansion, upgrades, and technologies appropriate for their area. If these districts were instituted, regulatory questions would certainly arise such as: What agency would organize and oversee the district? Would both agencies try to regulate those districts containing both residential and larger commercial on-site wastewater disposal systems? On a practical level, ISDH and IDEM have different missions, perspectives, regulations, and procedures for on-site systems, resulting in little cooperation. Ignoring this reality will result in further stagnation, ineffective regulatory policies, and inadequate funding for on-site regulatory staff.

Purdue strongly recommends combining regulation and oversight of all on-site wastewater disposal systems that do not discharge directly to surface waters to one agency. This agency should have the experience and expertise to design, understand, evaluate, and implement these systems. Additionally, this agency should have a commitment to research and application of innovative systems, and on-site operation, maintenance, and educational programs.

Development of programs to support the necessary reforms outlined above, especially in the area of ensuring maintenance, will require resources. Consolidating the on-site wastewater disposal program to one agency will reduce overlapping regulation, research, duties, support, and overhead, thus reducing the economic burden. Reforms will be ineffective unless the agency selected to execute the on-site wastewater program is properly funded. Certification and user fees are one way of providing some of the needed funding.

to top


Anonymous. Oregon Chapter 2: Recirculating Sand Filter. WWPCRE04. National Small Flows Clearinghouse, West Virginia University.

Ball, H.L. (b) Nitrogen Reduction in an On-Site Trickling Filter. Orenco Systems, Inc., Sutherlin, Oregon. Revised from On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Ball, H.L. 1994. Sand Filters: State of the Art and Beyond. Orenco Systems, Inc., Sutherlin, Oregon.

Ball, H.L., 1995(a). Sand Filters and Shallow Drainfields. Orenco Systems, Inc., Sutherlin, Oregon.

Bastain, R.K. and D.A. Hammer (a). The Use of Constructed Wetlands for Wastewater Treatment and Recycling. Constructed Wetlands: General Information Package. 1996. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Bounds, T.R. 1994. Design Criteria for Recirculating Sand Filters. Orenco Systems, Inc., Roseburg, Oregon.

Bruen, M.G. and R.J. Piluk. 1994. Performance and Costs of On-Site Recirculating Sand Filters. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Burgan, M.A. and D.M. Sievers. 1994. On-Site Treatment of Household Sewage Via Septic Tank and Two Stage Submerged Bed Wetland. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Cagle, W.A, and L.A. Johnson. 1994. On-Site Intermittent Sand Filter Systems, A Regulatory/Scientific Approach to their Study in Placer County, California. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Canody, J. 1997. Spray Irrigation is Alternative for Homeowners, Communities. Small Flows Newsletter, Vol.11. p.1.

Carlile, B.L. Performance Evaluation of Alternative Waste Systems for Lake and Coastal Developments in Texas. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Chiemchaisri, C., K. Yamamoto, and S. Vigneswaran. 1993. Household Membrane Bioreactor in Domestic Wastewater Treatment. Water Sci. Tech. Vol. 27 No. 1 p. 171-178.

Converse, J.C. 1997(a). Intermittent Sand Filters for On-Site Treatment of Domestic Wastes.

Converse, J.C. 1997(b). Recirculating Sand Filters for On-Site Treatment of Domestic Wastes.

Converse, J.C. and E.J. Tyler. 1994. Renovating Failing Septic Tank-Soil Absorption Systems Using Aerated Pretreated Effluent. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Converse, J.C. and E.J. Tyler. 1997. Soil Dispersal Units for Aerobically Treated Domestic Effluent.

Converse, J.C., E.J. Tyler, and S.G. Litman. 1994. Nitrogen and Fecal Coliform Removal In Wisconsin Mound System. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

D. Leszczynska and A. Dzurik, 1994. Constructed Wetlands for Simultaneous Treatment of Wastewater and Stormwater in Small Communities PROCEEDINGS, American Society of Agricultural Engineers, Annual Meeting and Symposium on Wastewater Management for Small Towns, Atlanta.

EPA, Michigan: Detecting Leaking Septic Systems in the Rouge River Basin. The Water Monitor, November-December 1996. EPA(United States Environmental Protection Agency) 841-N-97-001

EPA. 1980. Innovative and Alternative Technology Assessment Manual. 430/9-78-009. MCD-53. Office of Water Program Operations, Washington D.C. and Office of Research and Development, Cincinnati, OH.

EPA. 1986. Septic Systems and Ground-Water Protection: An Executive’s Guide. Office of Ground-Water Protection, U.S. Environmental Protection Agency. Washington D.C.

Franzmeier, D.P. 1991. Estimation of Hydraulic Conductivity from Effective Porosity for some Indiana Soils. Soil Sci. Soc. Am. J. 55:1801-1803

Franzmeier, D.P., B.B. Worstell, and B.R. Brasher. 1991. Bulk Density and Water Holding Properties of Indiana Soils. Proceedings of the Indiana Academy of Science. 100:187-196.

Freeman, R.J. Constructed Wetlands Experience in the Southeast. Constructed Wetlands: General Information Package. 1996. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Gersberg, R.M., S.R. Lyon, R.Brenner, and B.V. Elkins. Integrated Wastewater Treatment Using Artificial Wetlands: A Gravel Marsh Case Study. Constructed Wetlands: General Information Package. 1996.WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Gold, A.J., B.E. Lamb, G.W. Loomis, J.R. Boyd, V.J. Cabelli, and C.G. McKiel. 1992. Wastewater Renovation in Buried and Recirculating Sand Filters. J. of Environ. Quality. Vol. 21 No. 4 p. 720-725.

Green, M.B. Growing Confidence in the Use of Constructed Reed Beds for Polishing Wastewater Effluents. Constructed Wetlands: General Information Package. 1996.WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Griffin, D.M., B.E. Price, G.P. Pennison., Assessment of Rock Reed Filter Performance in the State of Louisiana. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Hammer, D.A. and R.K. Bastain. Wetlands Ecosystems: Natural Water Purifiers? Constructed Wetlands: General Information Package. 1996.WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Hanna, M.K, J.L. Kellam, and G.D. Boardman. 1995. Onsite Aerobic Package Treatment Systems. Water Resources., Vol 29, No. 11, p. 2530-2540.

Harper, W. 1993. Septic Systems: General Primer. Campground Management. Woodall.

Hinson, T.H., M.T. Hoover, and R.O. Evans. 1994. Sand Lined Trench Septic System Performance on Wet, Clayey Soils. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

IECA News. 1996. Educate while You Regulate. September/October. International Erosion Control Association. p. 16-17.

Jacquez, R.B., F. Vora, N. Kareem, X. Wang. 1991. Onsite Treatment of Septic-Tank Effluent by a rotating Biological Contractor. On-Site Wastewater Treatment. Volume 6. Proceedings of the Sixth National Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Jowett, E.C. and M.L. McMaster. 1994. Absorbent Aerobic Biofiltration for On-Site Wastewater Treatment-Laboratory and Winter Field Results. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Kadlec, R.H.. 1995. Overview: Surface Flow Constructed Wetlands. Water Sci. Tech. Vol. 32 No.3. p.1-12.

Knight, R.L. Treatment Wetlands Database Now Available. Constructed Wetlands: General Information Package. 1996. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Kuehn, E. and J.A. Moore. 1995. Variability of Treatment Performance in Constructed Wetlands. Water Sci. Tech. Vol.32 No.3. p.241-250.

Loudon, T.L. Performance of Trenches Receiving Sand Filter Effluent in Slowly Permeable Soils. Sand Filter Information Package. 1996. WWPCGN29. National Small Flows Clearinghouse, West Virginia University.

Mancl, K, and K. Young. Giardiasis. Agricultural Engineering Fact Sheet. SW-22. The Pennsylvania State University, College of Agriculture, Cooperative Extension Service.

Mancl, K. M. and Peeples. 1991. One Hundred Years Later: Reviewing the Work of the Massachusetts State Board of Health on the Intermittent Sand Filtration of Wastewater in Small Communities. On-Site Wastewater Treatment. Volume 6. Proceedings of the Sixth National Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

McIntyre, C., C.D’Amico, J.H. Willenbrock. 1994. Residential Wastewater Treatment and Disposal: On-Site Spray Irrigation Systems. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Oakley, S. and H. Ball. 1994. Subsurface Absorption Trench Design and Monitoring for Protection of Groundwater Quality. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Ogden, M.H. 1994. Ammonia Removal in Constructed Wetlands as a Function of Water Temperature. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Piluk, R.J. and E.C. Peters. 1994. Small Recirculating Sand Fitlers for Individual Homes. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Reed, S.C. and D.S. Brown. Constructed Wetlands Design-The Second Generation. Constructed Wetlands: Design. 1995. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Reed, S.C., EPA.1993. Subsurface Flow Constructed Wetlands for Wastewater Treatment: A Technology Assessment. Municipal Technology Branch. Office of Wastewater Enforcement and Compliance.

Ronayne, M.P., R.C. Paeth, and T.J. Osborn. Intermittent Sand Filter Design and Performance- An Update. Oregon State Department of Environmental Quality. In WWPCRE04. National Small Flows Clearinghouse, West Virginia University.

Roy, C. and J.P. Dupe. 1994. A Recirculating Gravel Filter for Cold Climates. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Sandretto, D.A. and Knight, R.L. 1991. Constructing wetlands-Do beavers do it better? Constructed Wetlands: General Information Package. 1996 WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Shaw, B. and N.B. Turyk. 1994. Nitrate-N Loading to Groundwater from Pressurized Mound, In Ground, and At-Grade Septic Systems. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Sievers, D.M. A Submerged Wetland for On-Site Sewage Treatment. Constructed Wetlands: General Information Package. 1996. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Small Flows, General Design, Construction, and Operation Guidelines: Constructed Wetlands Wastewater Treatment Systems for Small Users including Individual Residences. WWPCDM65. National Small Flows Clearinghouse, West Virginia University.

Small Flows. 1997. Response to Congress: EPA Outlines Barriers, Solutions to Use of Decentralized Systems. Small Flows Newsletter, Vol.11. p.4.

Steiner, G.R. and Freeman, R.J. 1988. Configuration and Substrate Design Considerations for Constructed Wetlands Wastewater Treatment. Constructed Wetlands: Design. 1995. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Tennessen, K.J. Production and Suppression of Mosquitoes in Constructed Wetlands. Constructed Wetlands: Design. 1995. WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Turco, R.F. 1994. Detection and enumeration of Coliform bacteria in soil. In: Methods of soil analysis: Microbial and biochemical properties. p 145-159. R. Weaver et al. (ed). Soil Science Society America. Madison WI.

Tyler, E.J. and J.C. Converse. 1994. Soil Acceptance of Onsite Wastewater as Affected by Soil Morphology and Wastewater Quality. On-Site Wastewater Treatment. Volume 7. Proceedings of the Seventh International Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Velioglu, S.G. K. Curi, and M. Kerestecioglu. 1988 A New Package Treatment Unit. Water Resources. Vol. 22 No. 1 p. 1-13.

Venhuizen, D. 1995. Soil Treatment Mechanisms: An Analysis of the Potential Impacts on Groundwater Quality of On-Site Wastewater Management Using Alternative Management Methods. Submitted to the Wisconsin Department of Industry, Labor, and Human Relations.

Watanabe, T., K. Kuniyasu, and H. Ohmori. 1993. Anaerobic and Aerobic Submerged Bio-Filter System for Small Scale On-Site Domestic Sewage Treatment. Water Sci. Tech. Vol. 27 No. 1 p. 51-57.

Whalen, K.J., Lombardo, P.S., Wile, B.D., and T.H. Neel. Constructed Wetlands: Design, Construction, and Costs. Constructed Wetlands: General Information Package. 1996.WWPCGN54. National Small Flows Clearinghouse, West Virginia University.

Whitmyer, R.W., R.A. Apfel, R.J. Otis, R.L. Meyer. 1991. Overview of Individual Onsite Nitrogen Removal Systems. On-Site Wastewater Treatment. Volume 6. Proceedings of the Sixth National Symposium on Individual and Small Community Sewage Systems. ASAE. Michigan.

Williams P.and J. Burson. 1985. Industrial Toxicology, Safety and Health Applications in the Workplace. Van Nostrand Reinhold, New York.

to top

An Evaluation of On-Site Technology in Indiana: Table of Contents