Watershed Management for Potable Water Supply: Assessing the New York City Strategy (2000)

Treatment and disposal of wastewater in the Catskill/Delaware watershed is a major factor in determining the quality of New York City drinking water. This is because almost all wastewater from the region is discharged either directly into streams that feed the water supply reservoirs or into the subsurface where it can eventually migrate to the reservoirs. Over 30,000 on-site sewage treatment and disposal systems (OSTDS) and 41 centralized wastewater treatment plants (WWTPs) are the major sources of wastewater in the watershed. Although the quality and quantity of wastewater discharged from WWTPs is known from monitoring data, such information is not available for OSTDS. In addition, the aggregate impact of either type of discharge on an individual reservoir, or on the system of water supply reservoirs, is extremely difficult to estimate.

A qualitative approach was used to overcome these limitations and answer the following critical questions:

1. Can the watershed sustain new WWTPs and OSTDS without declines in water quality?

2. Are effluent standards adequate, and are proposed technologies for WWTPs and OSTDS appropriate?

3. Are the rules governing the locations of new WWTPs and OSTDS adequate?

These questions were answered by determining the effects of technology upgrades for WWTPs and OSTDS on overall pollutant loading to the Catskill/Delaware watersheds. Some of these technology upgrades are mandated by the Memorandum of Agreement (MOA), while others are not. For example, all

surface-discharging WWTPs are to be fitted with microfiltration units or their equivalent.

It should be noted that this type of evaluation was conducted for WWTPs in the 1993 environmental impact statement (EIS) created for the Watershed Rules and Regulations (NYC DEP, 1993). However, upgrades to OSTDS were not considered in the 1993 EIS. Rather, that document evaluated the effect of variable OSTDS setback distances on pollutant loading to reservoirs. The similarities and differences between this analysis and that of the EIS are discussed throughout this chapter.

The approach used by the committee views each reservoir in the water supply system as a ''black box" that receives effluent from OSTDS and WWTPs. The volume and pollutant concentrations of OSTDS and WWTP effluent that are currently entering the Catskill/Delaware reservoirs represent a baseline "impact index" of 100 percent. Improved effluent quality related to better treatment technology or regulatory strategies decreases the impact index to a percentage less than 100. Increased effluent volume related to population growth increases the impact index by a proportional percentage. The questions posed above are answered by estimating the impact index produced by various strategies or options. For the purpose of this study, the New York City water supply system was limited to those watersheds west of the Hudson River as well as those supplying the Kensico and West Branch reservoirs.

Six parameters were selected to represent the quality of effluent from OSTDS and WWTPs: total phosphorus, total suspended solids (TSS), fecal coliforms, viruses, Giardia cysts, and Cryptosporidium oocysts. Each of these parameters is considered a priority pollutant by New York watershed planners as evidenced by their inclusion in the EIS prepared for the Watershed Rules and Regulations (NYC DEP, 1993).

The efficacy of removal of these six parameters varies between OSTDS and WWTPs (see Appendix D for more detail). Both processes are efficient in removing TSS. They are also effective in removing fecal coliforms because these bacteria die when exposed to the intolerable environmental conditions presented by either process. WWTPs are much more effective than passive OSTDS in removing dissolved phosphorus. Microorganisms in the WWTP treatment process use the phosphorus in the wastewater as a nutrient, and it is subsequently

deposited in the residuals. Phosphorus may also be removed chemically by tertiary WWTP treatment processes. Dissolved phosphorus passes through the septic tank of a passive OSTDS and enters the drainfield, where it is either used by aerobic microorganisms in the soil or adsorbed to soil particles. In either case, the ability of the soil to retain phosphorus is eventually exhausted, after which phosphorus-rich water can enter the environment. WWTPs are also more effective than OSTDS in eliminating viruses and bacteria because the effluent of WWTPs is chlorinated. Finally, the requirement for microfiltration at WWTPs likely enhances their ability to remove Giardia cysts and Cryptosporidium oocysts in comparison to OSTDS.

Treatment technologies that produce the highest-quality effluent, termed Best Available Control Technology (BACT), are required for WWTPs discharging to surface water under the terms of the MOA. BACT for WWTPs includes sand filtration, disinfection, phosphorus removal, and microfiltration, and all upgrades are scheduled for completion by 2002. Although not mandated in the New York City watersheds, BACT for OSTDS consists of aerobic treatment units (ATUs) that use mechanical devices such as pumps and impellers to create aerobic conditions within the septic tank. Support media may also be provided to increase the relative number of aerobic bacteria colonizing the tank, and a conventional drainfield is used after the aerobic unit. Dead bacteria, and associated phosphorus, are retained in the sludge layer in the tank, and this enhances the ability of ATUs to remove phosphorus. When these systems are properly maintained, the effluent quality of these systems is significantly better than that of passive septic tank and drainfield systems. Although there are other innovative aerobic OSTDS available (see Chapter 5), in the committee's opinion, only ATUs can be considered best available control technology because they have been field-tested and backed by third-party certification.

The existence of an effective regulatory strategy is an additional element that must be present for an OSTDS treatment process to be considered BACT. Failure rates and the period of time during which failed systems are a public health hazard can both be reduced by rigorous inspection and enforcement strategies. Typically, failures occurs when solids in the tank become too deep and wash into the drainfield. These solids clog the perforated pipe and underlying biomat and restrict the amount of liquid that can percolate through the soil. This results in partially treated sewage surfacing and potentially flowing into surface waters. Such failures often back wastewater into the residence or limit the use of toilets and laundries, which serves as an incentive for the homeowner to have the OSTDS repaired. Regulatory strategies that rely on self-reporting result in a failure rate (defined as an inability of a system to accept wastewater) of about five percent

during the year (Sherman, 1998). More stringent regulatory strategies, such as requiring inspection upon sale of the property, can decrease this failure rate.

In theory, a requirement for annual operating permits that mandate inspection of OSTDS each year can bring the annual failure rate to zero, and this strategy was assumed for the BACT calculations. Annual inspection must accompany the use of aerobic systems because these systems rely on mechanical devices that require electrical power and need preventative maintenance. In addition, aerobic systems produce a far greater volume of sludge than passive systems and require much more frequent pumping of sludge. Some states such as Florida already require annual operating permits for such systems, and these provide both revenue for inspections and a legal tool if enforcement is needed. Florida also requires aerobic systems to be under a maintenance contract to minimize the need for emergency repairs and to make sure maintenance is performed by qualified individuals with timely access to spare parts, rather than by homeowners. Maintenance is particularly important because ATUs function optimally when in continuous use and can experience disruption of function when shut down for extended periods of time. In fact, the addition of recirculating sand filters to ATUs is recommended for those systems operated on a seasonal basis to maintain a supply of aerobic bacteria. It should be noted that not all states have rigorous inspection and maintenance requirements, nor do they always enforce existing requirements, which can lead to significantly higher failure rates for ATUs (P. Miller, Virginia Department of Natural Resources, personal communication, 1999). A zero failure rate was assumed for this analysis in order to represent the greatest possible change that could accompany the use of BACT for OSTDS.

Wastewater Treatment Plants. Determining the contribution of WWTPs to overall pollutant loading in a watershed is relatively straightforward. The daily discharge of each pollutant by a WWTP can be estimated with certainty from monitoring data required under the terms of its State Pollutant Discharge Elimination System (SPDES) permit. Performance testing data can be used to determine the decrease in effluent pollutant concentrations following installation of BACT. Quantitative increases in effluent pollutant concentrations arising from future growth can be estimated using the maximum discharge limits in the SPDES permit and/or estimates of growth and average per capita wastewater production.

On-Site Sewage Treatment and Disposal Systems. There is much less certainty regarding daily discharges of the priority pollutants from OSTDS. Removal efficiency of the six constituents by OSTDS is related to the type and age of the OSTDS, geological conditions, and loading rates. Therefore, it is

necessary to make a number of assumptions in order to estimate the current daily discharge of each parameter from OSTDS. Improvements in effluent quality for OSTDS can be estimated by applying treatment efficiencies for ATUs and by assuming that a failure rate of zero is achieved through a rigorous enforcement strategy. Quantitative increases in effluent from OSTDS caused by population increases can also be estimated using projected rates of population growth and average per capita wastewater production. The assumptions used for both WWTPs and OSTDS and the calculated values derived from these assumptions are given in Appendix D.

Spatial Considerations. Data for effluent quality from both WWTPs and OSTDS do not consider changes between the discharge point and the receiving reservoir. The discharge point for WWTPs is typically a small watercourse, while effluent quality for OSTDS is usually measured 2 ft below the drainfield. In either case, travel of the effluent to the reservoir may result in qualitative improvements in the six parameters of concern. These changes have not been incorporated into these estimates because of a lack of studies on such changes and because of variations in locations and treatment techniques of individual OSTDS and WWTPs. Information from the Septic Siting Study currently being conducted in the watershed (see Box 11-1) may shed light on the fate and transport of

pollutants beyond the drainfield. Our analysis of pollutant removal in setbacks (Chapter 10) and a similar exercise from the 1993 EIS (NYC DEP, 1993, Table VIII.F-12) indicate that subsurface travel of sewage effluent can significantly decrease concentrations of some pollutants, particularly TSS and microbial constituents. On the other hand, overland or surface flow is not expected to decrease effluent concentrations to the same extent (NYC DEP, 1993, Table VIII.F-14). Because almost all WWTPs in the watershed discharge to surface waters, they are less likely than OSTDS to receive the beneficial effects of subsurface travel on effluent pollutant concentration. Thus, we expect this analysis to slightly overestimate the contribution of OSTDS to the overall pollutant loading in those basins with WWTPs.

Impact Index vs. EIS Analysis. As noted above, the 1993 EIS contains an analysis of upgrades to WWTPs similar to the impact index presented here. The same information on current effluent pollutant concentrations and future concentrations following upgrades was used in both analyses. However, methods used to calculate effluent concentrations in 2010 are different. The EIS applied various methods for calculating the population that would be served by municipal plants, nonmunicipal plants and expanding plants. The impact index applied the growth rate observed for the period 1990–1996, developed in Chapter 2, to the period from 1990 to 2010. The annual growth rate of 0.25 percent for the West-of-Hudson region resulted in a total population increase of 5 percent between 1990 and 2010. The annual growth rate of 1.06 percent for the East-of-Hudson watersheds resulted in a total population increase of 21.2 percent between 1990 and 2010.

Regarding OSTDS, the EIS did not consider the effect of treatment process upgrades on current and future pollutant loading, which is the goal of the impact index. For that reason, a meaningful comparison cannot be made between the two analyses. However, a few issues are worthy of discussion. First, the EIS estimates that 14,600 new OSTDS will be constructed in the Catskill/Delaware watershed between 1990 and 2010. These additional units are taken into consideration in the impact index by estimating future growth (using the same growth rates quoted above). Second, the EIS supports failure rates for OSTDS of 2 percent to 2.5 percent, depending on the type of system. The impact index assumes that all systems without technology upgrades suffer a 5 percent failure rate, while those that install BACT have a zero percent failure rate. Finally, the EIS analysis implies that only those OSTDS residing within proscribed setback distances (between 100 and 500 ft) significantly contribute to pollutant loadings in the reservoirs. (Likewise, the Total Maximum Daily Load calculations for all basins only considered those OSTDS within 100 ft of waterbodies.) The impact index does not exclude any OSTDS based on its proximity to nearby waterbodies.

The impact index for each reservoir in the West-of-Hudson watershed, as well as for the Kensico and West Branch watersheds, is presented graphically in Figures 11-1 through 11-4. These figures provide the total daily discharge of each of the six index parameters under baseline conditions, and they project changes if BACT is implemented and for population growth until the year 2010.

From a simple mass-balance perspective, new sources of wastewater (e.g., population growth or new industry) in a watershed will degrade water quality. This can only be prevented by transporting the effluent to a discharge point outside the watershed or by using treatment technology that does not increase the level of contaminants above the baseline (i.e., zero-discharge). Neither of these methods is practicable in the New York City watersheds; therefore, some degradation of water quality will be associated with new WWTPs and OSTDS.

The magnitude and impact of a decline in water quality depend on the treatment efficiency of the new WWTP or OSTDS receiving the wastewater. First, the impact of new growth can be mitigated by rerouting wastewater flow from relatively inefficient OSTDS to more efficient WWTPs. A second technique is to upgrade existing WWTPs or OSTDS so that the combined impact of the newly upgraded plant and new growth is equal to or less than the baseline impact. This latter technique is currently being used for WWTPs because the MOA requires existing WWTPs to install BACT. The potential for BACT to greatly improve effluent quality is shown dramatically in Figures 11-1 through 11-4.

According to the impact index analysis, the largest reductions attributable to the use of BACT at WWTPs occurred with phosphorus, total suspended solids, and Cryptosporidium oocysts. Reductions for these three parameters exceeded 95 percent in the Cannonsville watershed and were also significant in the Ashokan, Pepacton, and Schoharie watersheds. There were also large decreases in Giardia cysts, coliforms, and viruses in all three of these watersheds. There was no reduction in contaminant levels for the six parameters in the Rondout watershed and little reduction in the West Branch watershed. This indicates that existing WWTPs in those watersheds are already at or near BACT. The Kensico and Neversink watersheds have no WWTPs and therefore do not benefit from any offsets brought about by BACT.

As shown in Figures 11-1 to 11-4, the use of aerobic treatment units as BACT for OSTDS significantly reduces effluent concentrations of the microbial parameters Giardia, Cryptosporidium, coliform bacteria, and viruses. BACT does not substantially change effluent concentrations of total suspended solids from OSTDS because any OSTDS that uses a drainfield already provides excellent removal of total suspended solids. BACT for OSTDS would reduce the loading of total phosphorus to receiving waters as long as residuals are properly managed.

BACT is not specifically required for OSTDS under the terms of the MOA, and there is no evidence that it is being used. Instead, apparent practice sets the minimum acceptable technology to be the passive septic tank and drainfield combination, although ATUs and experimental systems are allowed. This is unfortunate because passive septic tank and drainfield combinations have a limited life and provided inferior treatment of residential wastewater compared to properly functioning ATUs. Requiring existing and new OSTDS to achieve BACT would significantly reduce their contribution of Giardia cysts, viruses, coliforms, and Cryptosporidium oocysts in direct proportion to the number of OSTDS in the watershed. It should be noted that much of the estimated decrease in pathogen loading that would accompany the use of BACT can be attributed to improved enforcement and the associated elimination of surfacing sewage caused by OSTDS failure. Although there is an enforcement presence in the watershed, there is no requirement for an annual operating permit or inspection.

Population growth will eventually return the concentrations of the six contaminants emanating from WWTPs to pre-BACT levels. Annual population growth in the West-of-Hudson watersheds for the period from 1990 through 1996 was only 0.25 percent, while annual population growth in the East-of-Hudson region during that same period was 1.06 percent. The calculated increase in total population for the period 1990 to 2010 is 5 percent for the West-of-Hudson watersheds and 21.2 percent for the East-of-Hudson watersheds. Because WWTPs in the Rondout and West Branch watersheds are already at or near BACT, population growth will begin to degrade water quality to below pre-BACT conditions almost immediately in those watersheds. In the Ashokan, Cannonsville, Pepacton, and Schoharie watersheds, where implementation of BACT will significantly improve effluent quality, over 100 years will be required to reach previous contaminant levels. Applying these same growth estimates to OSTDS shows that the impact of each of the six parameters (except phosphorus) on receiving waters would be virtually eliminated if BACT were required for

OSTDS. For phosphorus, about 24 years beyond 2010 would be required before concentrations in OSTDS effluent reach pre-BACT levels.

Technology is chosen with an emphasis on lowest cost, and applicable standards are typically met but not exceeded. Regulatory agencies attempt to influence this decision by requiring the use of BACT. BACT is a moving target that changes in response to technological advances. When upgrades to BACT are warranted, it is easier to retrofit BACT to a relatively few WWTPs than to thousands of OSTDS.

Currently, BACT for new WWTPs discharging to surface waters is defined as sand filtration, disinfection, phosphorus removal, and microfiltration. The following effluent standards are also required, which the SPDES permits for these plants must be updated to reflect:

These effluent standards for WWTPs, which are similar to effluent standards for drinking water treatment plants, are adequate and reflective of BACT. They should be viewed as upper limits that cannot be exceeded, even during large storm events.

At non-city-owned WWTPs, it is likely that the New York City Department of Environmental Protection (NYC DEP) will be substituting Continuous Backwash Upflow Dual Sand Filtration (CBUDSF) for microfiltration (Gratz, 1998). There is continuing controversy as to whether CBUDSF is functionally equivalent to microfiltration in meeting BACT for WWTPs (Marx and Izeman, 1998; NYC DEP, 1998a). In September 1996, NYC DEP presented oocyst removal data collected between 1994 and 1996 to the Environmental Protection Agency (EPA) indicating that the two methods were equivalent. EPA did not support this statement of equivalency, citing the fact that the research had been conducted at different labs, using different detection limits and testing protocols (Gratz, 1998). Thus, a new testing protocol was developed by NYC DEP with input from EPA and the New York State Department of Health (NYS DOH). A series of 12 tests using the new protocol as well as microfiltration was completed in October 1997, and a subsequent NYC DEP report concluded that both technologies ''exceeded the requirements of the Surface Water Treatment Rule (SWTR) and proposed

regulations governing log removals of Giardia cysts and Cryptosporidium oocysts for drinking water filtration plants" (NYC DEP, 1998a). Results from this report have been subject to criticism (Marx and Izeman, 1998).

In the committee's experience, there have been very few demonstrations of the CBUDSF technology. Dual-media filters are more frequently operated in down-flow mode and at drinking water treatment plants rather than at wastewater treatment plants. Indications from wastewater treatment plants using the down-flow technology are that the filters need to be cleaned regularly (beyond simple backwashing), a process that is not well understood and is often overlooked by plant managers (A. Amirtharajah, Georgia Tech, personal communication, 1999).

From a theoretical standpoint, no sand filtration system can be equivalent to properly operating microfiltration (G. Logsdon, Black and Veatch, personal communication, 1999; R. Trussell, Montgomery Watson, personal communication, 1999). However, based on the information available from NYC DEP, EPA, and independent expert scientists, it appears that the CBUDSF is an adequate substitute for microfiltration under relatively low oocyst loading rates and ideal operating conditions. Plants using this technology should be subject to rigorous long-term monitoring of particle counts in the 2- to 30-µm range, of turbidity, and occasionally of oocysts to verify that equivalency is maintained. This monitoring should determine the effectiveness of the filtration process and backwashing, and it should detect operational problems that may occur, such as clogging of the filter.

Allowable treatment techniques for OSTDS in the watershed are contained in Appendix 75A of 10 NYCRR Part 75 (State of New York), which requires that raw sewage from a household sewer must be discharged to an absorption treatment system (i.e., a drainfield) and must undergo treatment prior to such discharge. The law allows only septic tanks or aerobic units for this treatment. These rules may be augmented by local codes, and NYC DEP has delegated regulatory responsibilities for OSTDS to the watershed counties. Section 75A.2(b) states that "local health departments may establish more stringent standards. Where such standards have been established, or approval by another agency is required, the more stringent standard shall apply."

Effluent standards have not been set for OSTDS, and standards relate only to the technology of the system. The current practice allows a choice between a passive septic tank or an ATU prior to the drainfield. Because of the higher cost of aerobic units, the passive septic tank and drainfield combination is invariably chosen. As previously mentioned, passive septic tank and drainfield combinations provide inferior treatment of residential wastewater compared to ATUs. Thus, the most rigorous standards for OSTDS treatment technology are not being applied.

The BACT strategy envisioned by the committee calls for the mandatory use of aerobic systems to maximize effluent quality entering the drainfield, combined with an annual operating permit and other regulatory methods to limit failures and minimize repair time. This strategy was the basis for reducing estimated annual failure rates of OSTDS from five percent to zero percent. Although it is possible to set effluent standards for OSTDS (a requirement for advanced waste-water treatment is now being applied to effluent from OSTDS in the Florida Keys), this strategy is not recommended for the New York City watershed because this technology has not been demonstrated in northern climes. Instead, it is more reasonable to forego sampling and rely on a technology (e.g., ATUs) that has been certified by a third party, such as the National Sanitation Foundation (NSF) or the American National Standards Institute, as being capable of reliably meeting certain treatment efficiencies.

The cost of rapidly achieving BACT for OSTDS in the New York City watersheds would be significant. There are an estimated 38,854 OSTDS in the Kensico, West Branch, and West-of-Hudson watersheds (NYC DEP, 1993). Purchase and installation of an NSF-approved aerobic system is estimated to cost $8,000 each (Ayres Associates, 1998), for a total capital cost of $311 million. Annual operating costs for electricity and maintenance are estimated to add another $300 per year for each household, resulting in an annual total of $11.7 million. The cost of increasing inspection staff and/or implementing a requirement for an annual operating permit have not been estimated. Although the initial capital costs are large, they are likely to be less than the cost of constructing sewers and new WWTPs to serve these homes. In addition, there are strategies that could allow these benefits to be phased in over time. One such strategy is to implement a rigorous enforcement effort to effectively detect and repair malfunctioning OSTDS. Those OSTDS found to need replacement and any new OSTDS required for population growth would use ATUs. In addition, older OSTDS could be retrofitted to ATUs at a pace that would result in BACT being achieved for OSTDS over a ten-year or longer period.

Suitable locations for OSTDS described in the NYS DOH publication "Individual Residential Wastewater Treatment Systems Design Handbook" appear to be adequate (NYS DOH, 1996). There must be 100 ft separating OSTDS from nearby waterbodies and wetlands. In general, these locations require proper separation from groundwater, suitable soils, and control of slope. New York requires a vertical separation of two feet between the bottom of an absorption-trench drainfield and groundwater. The purpose of this separation is to allow the wastewater sufficient time to travel through oxygenated, unsaturated soils that

will support aerobic degradation. Soil suitability is related to the particle size of the soil, and either extreme may be problematic. Soils of very small particle size such as clays cannot accept wastewater rapidly, which leads to ponding of wastewater on the surface. Soils of overly large particle size such as sands will not detain wastewater for a sufficient period of time to allow aerobic degradation.

Separation from groundwater and soil suitability can be overcome by site preparation, such as mounding of the drainfield or the import of less permeable material for the drainfield. However, it is generally not possible to control slope over a large area. Thus, the New York State code does not allow OSTDS on slopes of greater than 15 percent. It should be emphasized that this requirement is based more on practicality than on science. The perforated piping that comprises the upper portion of the drainfield must be level or the wastewater will pond in the lowest area. In addition, because a minimum of 2 ft of aerobic soil is required beneath the distribution piping, and because the shoulders of the mound must be gently sloped to prevent erosion, the mounded area can become quite large.

It appears that waivers from the siting requirement for OSTDS have been authorized in the Kensico and West Branch watersheds (Fox, 1998) and that slopes of up to 20 percent may be used. This is unfortunate because regulatory codes represent minimum requirements. Their lessening may necessitate additional regulatory oversight to ensure conditions are maintained. Waivers are especially ill advised in areas that are environmentally sensitive or are of public health significance.

The Watershed Rules and Regulations include a 60-day travel-time delineation that is used to regulate siting of new or expanding WWTPs in the Catskill/Delaware watershed. The regulations forbid the construction of new surface-discharging plants within the 60-day travel-time boundary and within phosphorus or coliform restricted basins. Outside the boundary, surface-discharging WWTPs are allowed if they discharge into watercourses but not into reservoirs, reservoir stems, wetlands, or controlled lakes. There are no 60-day travel-time restrictions for new or expanding subsurface-discharging plants.

The 60-day travel time refers to the time necessary for water to travel from its point of origin to the distribution system. This boundary is intended to protect the water supply reservoir from pathogenic microorganisms originating in up-stream watersheds. The choice of 60 days, which was based in part on die-off rates for Giardia, has been approved by NYS DOH. Original research supporting the 60-day value is not available (J. Covey, NYS DOH, personal communication, 1998).

Because of the complexities of system operation, varying water levels in the reservoirs, and temperature gradients, the delineation of the 60-day travel-time boundary was a difficult undertaking (Klett, 1996). The boundary was calculated using flow data, reservoir-stage data, and bathymetric information for the years 1983–1991. The end points for the calculation are the two locations where water discharges from the Kensico Reservoir (CATLEFF and DEL 18). As discussed in Chapter 6, these points are heavily monitored for compliance purposes.

To be conservative, the boundary calculation represents the 95th percentile case. That is, water originating in areas outside the boundary has a travel time of greater than 60 days in all but 5 percent of the cases. Because of the relatively long residence times experienced by the reservoirs, the time needed for water to travel overland before reaching reservoirs was not considered in the calculation. A map of the Catskill/Delaware watershed showing the 60-day travel-time boundary is presented in Figure 11-5.

The appropriateness of a 60-day travel time for source protection must first be framed in the context of what objective is to be achieved. In the absence of precise written documentation justifying the 60-day value, it is assumed that this concept is based on the belief that 60 days will provide sufficient time for decay of microorganisms from a properly treated wastewater, so as to have de minimus impact on microbial levels in the source water.¹ In other words, travel time from the WWTP to the reservoirs is considered a barrier to pollution, much like chlorination or other treatment processes. Analyzed in this context, the literature suggests that 60 days may not be appropriate for protozoa, nor indeed for all enteric viruses. This point is elaborated upon below, with particular emphasis on Giardia and Cryptosporidium (of which the latter is of special concern because its reduction by chlorination alone is minimal).

One way of evaluating the travel time is to consider the log reduction of microorganisms that would occur over that period. This is a convenient way to compare the travel-time barrier with other conventional barriers such as filtration and disinfection. Disinfection criteria under the SWTR require a 3-log removal of Giardia and a 4-log removal of enteric viruses. For systems that filter, the Enhanced SWTR will require a 2-log reduction of Cryptosporidium.

Viruses. There is limited information on viruses with which to evaluate the 60-day travel time. In the case of some viruses, such as adenovirus, 60 days of holding in dechlorinated tap water has been found to produce only a 2-log reduction at 23°C (Enriquez et al., 1995), significantly less than the 4-log reduction required of disinfection processes. The same viruses have greater resistance to disinfection by UV than does poliovirus (for which most disinfection data has been obtained) (Meng and Gerba, 1996), and their resistance to chemical disinfection in water does not appear to have been investigated. These facts suggest that the 60-day travel time may be inadequate to protect against adenovirus.

Giardia. Treated wastewater effluents in the Catskill/Delaware watershed have been found to have average Giardia concentrations ranging from less than 1 to over 400 cysts per 100 L (NYC DEP, 1999). These sources have had detectable levels of Giardia in 40.8 percent of all samples taken between 1993 and 1998. Over the same time period, Kensico Reservoir effluent locations CATLEFF and DEL 18 have had mean concentrations (substituting zero for measurements below the detection limit) of about 0.2 cysts per 100 L. These values imply that a 1- to 3-log die-off of cysts from the treatment plant effluent discharge to the distribution system would have to occur during the 60-day travel time.

As is the case for viruses, few studies have investigated log removal of cysts over long time periods in conditions similar to those found in the watershed. DeRegnier et al. (1989) suspended Giardia muris in river water (Mississippi River at Minneapolis) and lake water at ambient temperatures and monitored viability using propidium iodide and animal infectivity. As measured by infectivity, cysts remained viable at least up to 40 days. It should be noted, however, that the small number of animals did not likely permit measurement of inactivation beyond 1 log. The authors concluded, "G. muris cysts suspended in environmental water remained viable for 2 to 3 months, and their survival was enhanced by exposure to low water temperature, despite the fact that the cysts were suspended in the fecal biomass within the sample vial." Other studies have shown that Giardia cysts can survive as long as three months in pit latrines and sewage sludge (Deng and Cliver, 1992), but log removal was not calculated. Based on this information, a 60-day travel time for protection against Giardia is not supported.

Cryptosporidium. Wastewater effluents discharging into the New York City watershed have substantially lower oocyst levels than do other WWTP effluents across the country. For example, Madore et al. (1994) found that an average wastewater with conventional activated sludge treatment had final effluent oocyst levels of 14,000–396,000/100 L, while NYC DEP studies consistently show concentrations several orders of magnitude lower (Table 11-1). Based on the data in Table 11-1 and the source water oocyst levels measured in Kensico

Reservoir, at least a 1-log (for advanced treated wastewater) to 4-log removal during 60 days might be appropriate benchmarks for Cryptosporidium.

Robertson et al. (1992) used the DAPI-PI dye inclusion assay to monitor decay of viability of oocysts held in membrane diffusion chambers in river water under ambient conditions. It has recently been determined that the survival of oocysts in fecal material as measured by vital dyes correlates well with the ability of the oocysts to excyst (Jenkins et al., 1997). Results of the Robertson et al. experiments are shown in Figure 11-6. From these results, 1-log inactivation is estimated to occur at 100 and 180 days for the two strains examined.

FIGURE 11-6. Inactivation of C. parvum oocysts in river water. Source: Reprinted, with permission, from Robertson et al., 1992. © 1992 by the American Society for Microbiology.

These results should be tempered by the observation that vital dyes and excystation appear to be less sensitive indicators of oocyst inactivation by chemical disinfectants than is animal infectivity (Black et al., 1996). That is, greater kill is noted if oocysts are tested for animal infectivity as compared to other would viability methods. Although it is not known whether this same differential would exist for oocysts inactivated over time in surface waters, it may be that dye and excystation tests generally overestimate viability and thus underestimate inactivation. (From a public health point of view, this makes them conservative assays.)

In any event, consideration of both Giardia and Cryptosporidium information raises questions as to whether a 60-day travel time is sufficiently protective. Studies using the waters under question and sensitive viability assays (animal infectivity or perhaps cell culture in the case of Cryptosporidium) should be conducted to establish more site-specific guidelines. If the criterion of no significant adverse impact is used and inactivation is the sole removal mechanism considered (as opposed to dilution and sedimentation), then it would appear that 60 days may be insufficiently conservative, and perhaps a zone of 180 days or more might be required, especially for Cryptosporidium. An explicit scientific rationale for such a zone of protection should be developed.

1. The upgrades to WWTPs mandated by the MOA should be effective in reducing effluent concentrations of phosphorus, TSS, coliforms, viruses,Giardia cysts, and Cryptosporidium oocysts. The requirement that these upgrades use BACT is an important component of New York City's watershed management strategy. Because BACT is a "moving target," it is important that these requirements be reevaluated regularly to ensure additional improvements are implemented.

2. Current technologies being used for OSTDS are not adequate and do not represent Best Available Control Technology. Passive systems consisting of a septic tank and drainfield are allowed instead of requiring ATUs, which maximize the destruction and inactivation of microorganisms. Implementation of BACT for OSTDS, including a substantial enforcement effort, could drastically reduce effluent concentrations of Giardia, Cryptosporidium, and viruses in all Catskill/Delaware watersheds, although BACT will not significantly alter OSTDS effluent phosphorus concentrations.

3. Aerobic treatment units should be mandated for new or replacement OSTDS, and enforcement efforts should include annual inspections. This recommendation is especially important for the Kensico watershed, because of its critical location in the water supply and because OSTDS serves the entire popu-

lation. An evaluation should be made of allowable time to upgrade other OSTDS to BACT. Requiring effluent standards for OSTDS is not absolutely necessary, although a representative sample of ATUs should be performance-monitored to ensure their functionality.

In the future, other technologies may become candidates for BACT. To explore their use, New York State should consider conducting demonstration projects and establishing a rigorous experimental permit program for these technologies (as has been done in Virginia, Florida, Washington, and elsewhere). If ATUs, or some equivalently tested and approved technology, are not required for repairing existing OSTDS and installing new OSTDS, NYC DEP should consider converting users of OSTDS to wastewater treatment plants.

4. Effluent standards for WWTPs are adequate. BACT should be used for all WWTPs whenever possible to counteract declines that occur because of population growth. The implementation of BACT at WWTPs in the Ashokan, Schoharie, Cannonsville, and Pepacton watersheds will significantly reduce contaminant concentrations in plant effluents. BACT for WWTPs will have little present-day impact on effluent concentrations in the Rondout, West Branch, Kensico, and Neversink watersheds. Maximum effort should be made to have the upgrades mandated by the MOA installed as quickly as possible at all existing facilities.

5. To make sure that the Continuous Backwash Upflow Dual Sand Filtration units represent BACT, wastewater treatment plants using this technology should be subject to rigorous long-term monitoring of particle counts and turbidity to verify that equivalency is maintained. This monitoring should determine the effectiveness of the filtration process and backwashing, and it should detect operational problems that may occur, such as clogging of the filter.

6. Waivers that allow placement of OSTDS on slopes of greater than 15 percent should not be permitted. Although slopes between 15 percent and 20 percent are not fundamentally less able to provide adequate treatment, slopes greater than 15 percent can make proper construction of OSTDS more problematic. In addition, such systems may require additional regulatory oversight.

7. The current 60-day value used for siting WWTPs does not appear to be supported by available knowledge. The particular criteria to be achieved by such a barrier must be explicitly defined prior to stipulation of a duration that might be acceptable, and an explicit scientific rationale for such a zone of protection should be developed. Limited research suggests that 60 days may be inadequate for significant inactivation of Cryptosporidium oocysts, which are known to be resistant to disinfection. The 60-day limit may not allow adequate

protection against Giardia, particularly at low temperatures. Sixty days may not be sufficient to substantially reduce virus levels, although their sensitivity to chlorination makes this less problematic than for the protozoans. NYC DEP should pursue further studies of pathogen transport and fate using sensitive viability assays and local source waters to refine this value.

8. Although declines in water quality related to WWTPs and OSTDS will occur with population growth, in most cases 40–100 years will pass before contaminant contributions from WWTPs and OSTDS reach pre-BACT levels. It should be noted that these results are likely to be conservative because the impact analysis did not attempt to quantify any improvements to wastewater quality occurring between the wastewater's point of discharge to a subsurface drainfield and its point of entry into a surface water reservoir. Other factors that may affect the accuracy of these predictions include the fact that no attempt was made to differentiate seasonal and permanent populations in the watershed, and BACT for OSTDS was assumed to be ATUs.

Ayres Associates. 1998. Technology Memorandum #7: Technology Assessment of On-site Wastewater Treatment Systems. Prepared for Monroe County Sanitary Wastewater Master Plan.

Black, E. K., G. R. Finch, R. Taghi-Kilani, and M. Belosevic. 1996. Comparison of assays for Cryptosporidium parvum oocysts viability after chemical disinfection. FEMS Microbiology Letters 135: 187–189.

Bott, T. L. 1973. Bacteria and the assessment of water quality. Pp. 61–75 in Biological Methods for the Assessment of Water Quality. ASTM STP 528. Philadelphia, PA: American Society for Testing and Materials.

Deng, M. Y., and D. O. Cliver. 1992. Degradation of Giardia lamblia cysts in mixed human and swine wastes. Applied and Environmental Microbiology 58(8):2368–2374.

DeRegnier, D. P., L. Cole, D. G. Schuff, and S. L. Erlandsen. 1989. Viability of Giardia cysts suspended in lake, river and tap water. Applied and Environmental Microbiology 55(5):1223–1229.

Enriquez, C. E., C. J. Hurst, and C. P. Gerba. 1995. Survival of enteric Adenoviruses 40 and 41 in tap, sea and wastewater. Water Research 29(11):2548–2553.

Fox, J. 1998. Letter to Ronald Tramontano, New York Department of Health from EPA Region II. Dated April 10, 1998.

Gratz, J. 1998. Letter to Interested Parties. March 20, 1998. Re: NRDC's Letter on Microfiltration. Dated March 20, 1998. Region 2 EPA, New York, N.Y.

Jenkins, M. B., L. J. Anguish, D. D. Bowman, M. J. Walker, and W. C. Ghiorse. 1997. Assessment of a dye permeability assay for determination of inactivation rates of Cryptosporidium parvum oocysts. Applied and Environmental Microbiology 63(10):3844–50.

Klett, B. 1996. Delineation of a Sixty-Day Travel Buffer for the Protection of the New York City Water Supply. In New York City Water Supply Studies. Proceedings of a Symposium on Watershed Restoration Management. American Water Resources Association. Herndon, VA.

Madore, M. S., J. B. Rose, C. P. Gerba, M. J. Arrowood, and C. R. Sterling. 1994. Occurrence of Cryptosporidium oocysts in sewage effluents and selected surface waters. Journal of Parasitology 73(4):702–705.

Marx, R., and M. Izeman. 1998. Memorandum to Jeff Gratz, EPA Region 2, on April 21, 1998.

Meng, Q. S., and C. P. Gerba. 1996. Comparative inactivation of enteric adenoviruses, poliovirus and coliphages by ultraviolet irradiation. Water Research 30(11):2665–2668.

New York City Department of Environmental Protection (NYC DEP). 1993. Final Generic Environmental Impact Statement for the Proposed Watershed Regulations for the Protection from Contamination, Degradation, and Pollution of the New York City Water Supply and its Sources. November 1993. Corona, NY: NYC DEP.

NYC DEP. 1998a. DEP Pathogen Studies on Giardia spp. and Cryptosporidium spp.—Protocol Development for testing equivalency of two technologies to remove pathogens from wastewater effluent. Valhalla, NY: NYC DEP.

NYC DEP. 1998b. DEP Pathogen Studies of Giardia spp., Cryptosporidium spp., and Enteric Viruses. July 1998. Valhalla, NY: NYC DEP.

NYC DEP. 1999. DEP Pathogen Surveillance Report of Giardia spp., Cryptosporidium spp., and Enteric Viruses. Semi-Annual Report. Valhalla, NY: NYC DEP.

New York State Department of Health (NYS DOH). 1996. Individual residential wastewater treatment systems design handbook. Albany, NY: NYS DOH.

Robertson, L. J., A. T. Campbell, and H. V. Smith. 1992. Survival of Cryptosporidium parvum oocysts under various environmental pressures. Applied and Environmental Microbiology 58:3494–3500.

Sherman, K. 1998. OSTDS Research Coordinator for the Florida Department of Health. Presentation given at the Third NRC Committee Meeting. May 13–16, 1998, Oliverea, New York.