A PRIMER ON
DECENTRALIZED "WASTE" WATER MANAGEMENT

FOR SOUTH TEXAS COLONIAS

1. INTRODUCTION

The conventional "wisdom" for providing an area-wide wastewater management system is to "regionalize" the system, collecting wastewater from as far and wide as is considered technologically practical to one point, where a "regional" wastewater treatment facility would be installed. This view is based upon a number of assumptions about the fiscal, institutional and technological efficiency of this "regional" strategy. In many situations, these assumptions are highly questionable. In particular, it is noted that a large majority of the investment in the "regional" wastewater system does nothing but move pollution from one place to another. In the case of many South Texas colonias, that transport system poses a severe cost penalty, as these colonias are located at some distance from an existing collection system, necessitating long runs of lines in deep trenches and multiple lift stations.

It is also tacitly assumed that this far-flung collection system does in fact simply move pollution to the treatment plant. However, sewer lines leak, manholes overflow, and lift stations fail with maddening regularity. The result is that a "regional" collection system can actually cause of spread of pollution over the area. Over the long term, this can lead to low-level pollution of groundwater resources over a large area. It has been reported that exactly this result has been observed in the Los Angeles, California, area.

Studies sponsored by the Texas Water Development Board (TWDB) under its Economically Distressed Areas Program (EDAP) indicate that colonias which are not located very close to existing sewer systems would indeed be overly expensive to serve by these kinds of conventional gravity sewer systems leading to centralized treatment plants. This has been confirmed by facility planning, which showed that centralized systems would often be two to three times more costly than an alternative strategy - the decentralized concept of wastewater management. This paper reviews how decentralized systems can be used to address the wastewater management challenges of colonias.

Decentralized management is somewhat of a misnomer because, however dispersed the collection, treatment and reuse components may be, the overall system would be "centrally" managed by a sewer authority with powers appropriate to the technologies and methods employed in each circumstance. Conventional sewers and lift stations-which are also decentralized components-and treatment plants would not continue to function for long if operations and maintenance were left to the sole discretion of individual users. A decentralized concept system is no different in this regard.

In considering the provision of wastewater service, the impact of management strategy on the regional water economy should also be considered. The South Texas area where many colonias are located is an arid region where the adequacy of water supply is a chronic-and growing-problem. Thus, when addressing water resources management in colonias, the term "wastewater" should also be a misnomer. Whenever practical, "waste" water should be managed in a manner that results in reuse of that water for some beneficial purpose. Rather than addressing "waste" water as a nuisance, it should be viewed as a resource that serves the regional water economy. This paper also includes discussion of that issue.

Three generic management strategies may be considered as alternatives to conventional, centralized wastewater collection and treatment systems. These are:

1. Individual on-site wastewater systems, with each system contained entirely on the lot which it serves.
   
   
2. Small-scale collective systems, often called "cluster" systems, in which several lots share common treatment/reuse facilities.
   
   
3. Large-scale collective systems, in which wastewater from all lots in the colonia, or a group of closely-spaced colonias, flows to one treatment/reuse site.

1.1 On-Site Wastewater Systems

Individual on-site wastewater systems are only appropriate if sufficient space is available on the lots to contain a dispersal field, and if soils are suitable for subsurface dispersal of septic tank effluent with a hope of achieving proper treatment in the dispersal field. While many colonias are covered by soils that may be suitable for dispersal of septic tank effluent, widespread use of conventional, gravity-dosed dispersal trenches is typically not a good strategy for colonia-wide wastewater management systems, for three reasons:

1. Many of the lots in a colonia may not contain adequate unencumbered space for a code-compliant drainfield. Typically, if a significant portion of the lots would require some form of collective solution, then the overall system would be more cost efficient if all of the lots are served by the collective system, even though on-site systems may be more cost efficient for individual lots which can accommodate them.
   
   
2. Having many conventional on-site systems in a densely populated area creates a public health liability. Even if most of the systems do not incur problems, any one failure carries the potential for spread of disease throughout the community. Because of this, anecdotal evidence that any particular conventional system has not been problematic is of questionable relevance.
   
   
3. On a small lot, the consequences of a drainfield failure may be disastrous to the owner, because adequate space for a replacement field may not be available.

Despite the objection stated in item 1 above, it may be feasible to use individual on-site systems for part or all of a colonia even when some lots do not have adequate space for drainfields. For example, off-site drainfields might be obtained for lots which cannot accommodate a drainfield while using on-site systems on those which can. Further, if the lots which require off-site drainfields are clustered, these could be served by a small-scale collective system rather than by individual systems, if that would decrease the cost per house. A system of this type is illustrated in Figure 1.

Where it is feasible to employ individual on-site systems, the likelihood of drainfield failure can be minimized by employing shallow, low-pressure-dosed (LPD) drainfields instead of conventional gravity-dosed trenches. This matter is discussed further in Section 3 of this paper, which details the LPD concept. For a fairly modest cost increase, this substitution provides greater assurance of long-term protection of public health and water quality. Employing an LPD field should also preclude the need for a replacement field area as well.

Even if it is determined that conventional on-site systems can be used as a long-term solution for any given lot, this method should be addressed within the context of an organized management system, for two reasons:

• Funding. The aim is to provide service through a managed wastewater system, and financing is to be arranged for implementing that system. If, in order to avoid management fees, an owner were allowed to opt out of that system and use an individually managed conventional on-site system, and then that on-site system subsequently failed, that owner may be left to finance corrective actions from his/her own resources, unless the funding system can somehow provide for such deferred actions.
  
  
• Surveillance. Without routine observation by the management entity, any failure could persist for some time before corrective actions were taken. As noted in item 2 above, this may create a public health hazard to the entire community.

Allowing unmanaged conventional on-site systems to remain in place may largely defeat the purpose of the overall effort, which is to correct wastewater management deficiencies in the colonias in a manner which not only addresses currently obvious deficiencies but also precludes future public health and water pollution hazards.

1.2 Small-Scale Collective Wastewater Systems

Under the small-scale collective strategy, wastewater from a group of lots flows to a small treatment system or to a common dispersal field. An illustration of this concept is shown in Figure 2. The number of lots routed to each treatment/dispersal center would be determined by the treatment and reuse strategy and local conditions, such as topography, nature and density of development, and available space to house the treatment and dispersal systems. In many colonias, it may be cost efficient to define the tributary area to each treatment/dispersal center as those lots which can flow by gravity to that point, eliminating pumps from the collection system. As shown in Figure 2, septic tanks at each house and effluent sewers would be the preferred collection system, a concept which is explored further in the following section.

If soils are appropriate, septic tank effluent may be routed to LPD dispersal fields located on vacant lots or scattered around the margin of the colonia. As illustrated in Figure 2, dispersed treatment plants could also be used, with the effluent either routed to reuse opportunities or piped to a point where it is joined by effluent from other plants for disinfection and discharge. The treatment plant, being installed within a neighborhood, must be essentially "sealed" and tamper-proof, and the technology used must be cost efficient to install and to operate and maintain at a very small-scale. The high performance biofiltration system is the most likely candidate for this duty.

1.3 Large-Scale Collective Management Strategy

Under the large-scale collective management strategy, all of the dispersal area or treatment capacity is moved to one location, and a more extensive collection system is employed. A conventional gravity collection system and conventional treatment options could be used under this strategy, but it will generally be more cost efficient and less problematic for on-going operations to employ small-diameter effluent sewer systems and so-called "innovative/alternative" treatment options (e.g., biofiltration systems or wetlands) well suited to most colonia-sized treatment plants. If soils are suitable for dispersal of septic tank effluent, a large LPD dispersal field as shown in Figure 3 could be used instead of a treatment plant. The flat topography typical of many South Texas colonias generally would require that there be pump stations in the collection system under this strategy-either dispersed effluent pump stations as illustrated in Figure 3, or a central lift station at the treatment/dispersal center. In general, smaller dispersed pumping stations are expected to be the more cost efficient strategy, as explained in the following section on small-diameter sewers.

2. SMALL-DIAMETER EFFLUENT COLLECTION SYSTEMS

Decentralized concept management strategies employ septic (interceptor) tanks at each home (or other source of wastewater generation) to intercept solids, allowing the use of small-diameter effluent sewers in the collection systems. Installing septic tanks at the sources of flow offers the following advantages:

• The septic tanks, if properly designed, provide considerable pretreatment. Typical effluent strengths are less than 100 mg/l of BOD5 and 60 mg/l of TSS out of septic tanks fitted with effluent filters, a significant reduction from the levels generally observed.
• Septic tanks comprise the major sludge handling system. Sludge separation and storage in septic tanks is a passive process, requiring no operator intervention until the tank is pumped.
• Sludge digestion in septic tanks is an anaerobic process, which-relative to aerobic sludge digestion typically employed at conventional treatment plants-greatly reduces the volume of solids to be eventually handled.
• Because the sludge from each generator is segregated, it can be classified by source. Sludge from domestic sources very rarely contains contaminants which render the sludge unusable as a soil amendment, and the input to each tank is relatively "secure" against unauthorized dumping of hazardous materials, unlike a conventional collection system with its very many unsecured access points (manholes). Being readily reusable, this sludge stream is a potential revenue source.
• Sludge from each tank only needs to be handled at very long intervals, as detailed below. Therefore, even though there would be many dispersed sources, sludge management logistics would not be untenable.
• Sludge handling is executed without taking any part of the system out of service. Each tank can be pumped in a very short amount of time, and the users can even load the system while it is being pumped.
• Sludge handling is executed in small increments-one or a few tanks at a time, depending on the capacity of the pumper truck. This allows maximum flexibility in planning for sludge handling.
• Timing of sludge removal is not critical because sludge builds up very slowly. Several months could pass between the time that sludge depth monitoring indicates pumping should be executed and the time that pumping actually occurs without significantly degrading effluent quality. This offers the opportunity to pump at a time of the year when the sludge could be directly landspread, perhaps eliminating the need for intermediate treatment or storage facilities.
• Institutional arrangements for sludge handling are already in place. Existing septic tank pumpers could be contracted to pump the septic tanks.
• Eventually, if these types of systems become widely employed in any given area or jurisdiction, it would probably be more cost efficient to institute some form of unified sludge management program, such as one or more composting centers, and perhaps for the management entity to own and run its own fleet of pumper trucks, but these costs can be deferred for many years. In fact, since it would be several years before the first pumping is required, all costs of a sludge handling system, no matter what its form, can be deferred until the time it is actually needed.

2.1 Septic Tank Design for an Effluent Sewer System

A design for an interceptor tank for an effluent sewer system is shown in Figure 4. Note that septic tanks used with effluent sewer systems must be watertight to preclude excessive flow through the system due to groundwater intrusion. It is typically presumed that a septic tank would be installed to receive flow from each house. However, it may be practical in many cases to route two houses into each tank, as indicated in Figure 3, thereby perhaps significantly decreasing total cost of the system. Routing more flow through each tank would imply shorter pumping intervals, but with fewer tanks to pump, overall operations and maintenance costs should not change significantly.

Two features of this tank should be noted. One is the effluent filter. As stated previously, experience has shown that effluent filters can considerably reduce the level of organics and solids passing out of the septic tank. Thus, an effluent filter is inexpensive protection against needing to ever flush an effluent sewer system, and it assures that a consistently lower strength wastewater would be routed to the treatment/dispersal system.

While the effluent filter must be cleaned periodically, that is a very simple process, consisting of pulling the filter body out of the housing, hosing it off and reinserting it. To accommodate this cleaning, a readily accessible port of the type shown in Figure 4 should be installed. Experience to date indicates that cleaning once every two years would be sufficient. This can be readily executed at the same time that sludge depth is checked. The cost tables shown in Section 6 of this paper presume that the effluent filter is cleaned and sludge depth is checked every other year.

The other feature is the cleanout/drop pipe assembly on the tank outlet. As noted below, septic tank outlets must all be above the hydraulic grade line of the effluent sewer. Installing this assembly at each septic tank outlet not only provides a cleanout for each house connection, but also assures that the tank outlet is elevated several inches above the end of the house connection. This allows that depth of surcharge on the sewer line at the point where this house connection joins it without causing backflow into the septic tank. Effluent sewer systems can be designed to essentially eliminate the likelihood of surcharging the sewer line to this extent, so this feature provides a "safety factor" for the system.

Required pumping intervals for primary septic tanks will depend upon the habits of the users. However, a general idea can be provided by applying an equation based on studies by the U.S. Public Health Service and others. The equation defines the rate of sludge buildup as follows:

Rsl = 8.15 t + 38.82

where t is time between pumpings in years and Rsl is sludge accumulation in gallons per capita. The recommended criterion for pumping is when sludge level rises to within 6 inches of the effluent filter inlet. One model of 1,000-gallon septic tank, considered to be typical, has a plan area of 47.1 sq. ft. and an allowable sludge depth of 17 inches, resulting in a maximum storage of about 66.7 cu. ft., or about 500 gallons. Assuming an occupancy of 5 persons (all the colonias in one very extensive study in Hidalgo County had average occupancy levels at or below 5 persons per house) allowable sludge accumulation is 100 gallons/capita. Substituting this for Rsl and solving for t yields a pumping interval of 7.5 years. This indicates that pumping intervals of several years should be routinely expected. In the cost tables in Section 6, a 4-year pumping interval is assumed.

2.2 Effluent Sewer Concepts

The small-diameter effluent sewers would transport liquid effluent, essentially free of settleable solids, so "normal" sewer line design principles such as minimum scouring velocities do not apply. These effluent sewers can typically be installed at shallow depths, running with the lay of the land, so trench depth is typically quite shallow relative to conventional sewers. The small pipes also allow narrower trenches, and they can be "sprung" around curves, so sewer routing is much more flexible. These factors result in drastically decreased installation costs for effluent sewers relative to conventional sewer pipes. There are three basic types of effluent sewer system:

1. Variable-grade effluent sewers (VGES). In this concept, sewer lines run with the lay of the land, so there may be locally adverse grades, creating isolated low points and high points. The pipe system works like a sink trap through the low sections. Care must be taken that any connections which enter the sewer in these sections are not surcharged to such an elevation that wastewater would flow back into the septic tank at that connection. The system is designed to assure that the hydraulic grade line is below all septic tank outlets. This is done by increasing pipe sizes as required to lower the hydraulic grade line. Air release systems must be provided for isolated high points to prevent "air lock". To prevent odor problems, these should feed into a soil scrubber.
   
   
2. Minimum grade effluent sewers (MGES). In this concept, sewer lines generally run constantly downhill, but, with no solids to transport, no set minimum velocity must be maintained. Maintaining a continuously uniform grade is not critical-even some flat sections could be allowed as long as allowance is made for some surcharging (as would be provided "automatically" by the proposed septic tank outlet piping shown in Figure 4), so the line is much easier to lay than conventional gravity sewers. Again, all septic tank outlets must be above the hydraulic grade line. Since there are no high points or low points, the potential for solids deposition is essentially eliminated and air release valves and odor scrubbing systems are not required.
   
   
3. Septic tank effluent pump (STEP) sewers. Again sewer lines run with the lay of the land in this concept. This type of system is used when septic tank outlets are below the hydraulic grade line, so a pump is required to move effluent through the sewers. Since wastewater is pumped, line grade is irrelevant, making STEP sewers very easy to lay. The system may be designed with a pump tank at the outlet of each septic tank, or flow from several septic tanks may be routed through effluent gravity sewers to a collective pump tank. Pump size and pumping frequency are based on the daily flow rate through the pump tank and the total dynamic head against which it is pumped. This is determined by the flow rate, the length and size of pressure sewer, and elevation difference between the tank and the outlet, which is typically into another tank at the treatment/dispersal center.

2.3 Effluent Pump Stations

The two gravity sewer concepts (VGES and MGES) are often collectively referred to as septic tank effluent drain (STED) systems. It has been found in studies that, when pumping is required, generally the most cost efficient system would route effluent from several septic tanks through STED sewers into a collective STEP tank, an effluent pump station. In the very flat terrain typical of many South Texas colonias, this strategy will be more cost efficient than either conventional sewers with a central lift station or a STEP system with individual pump tanks at each septic tank. A typical design for an effluent pump station is shown in Figure 5. The large-scale collective management strategy would typically employ this "STED/STEP" sewer system.

Note in Figure 5 an allowance for surge storage. In a large system, more than one effluent pump station might feed into a pressure sewer line. If more than one pump were to come on at once, the higher flow rate would increase friction head losses and thus decrease the flow rate produced by each pump. In the worst case, if several pumps were to come on at once, one or more pumps might "deadhead" until other pumps turned off. To accommodate the potential for this situation, a suggested practice is to design all effluent pump stations with an equalization volume (surge storage) equal to the peak hour flow. This is in addition to the "emergency" storage volume above the alarm that would signal simultaneous failure of both pumps in the duplex pump system. This equalization volume also provides additional storage for times when power outages occur.

As noted previously, in the small-scale collective management strategy, location of the multiple treatment/dispersal centers could be determined by the limits of a gravity flow collection system. The result is that the collection system effluent pump stations would be replaced with the receiving tank of a treatment/dispersal center. Employing this design concept saves not only additional run of sewer but the extra cost of pump stations and their operation and maintenance expenses.

2.4 Effluent Sewer Pipe Sizing

The flat terrain typical of many South Texas colonias urges use of the MGES concept, a strategy that minimizes operational and maintenance liabilities. For this analysis, it is assumed that the starting depth of the sewer-at the outlet of the most upstream septic tank-is 3 feet at the flow line out of the cleanout/drop pipe. For ease of serviceability and structural reasons, maximum line depth at the receiving tank (effluent pump station or a treatment center tank) should be limited to about 4 feet. In the worst case, the surface elevation at the starting point and the end point would be equal, so only a 1-foot drop in pipe elevation would be allowed. Based upon studies of many South Texas colonias, it is expected that a slope of 0.25% is adequate to carry the required flows, as outlined below. If 1 foot of drop were available, this slope would allow up to 400 feet of pipe run before an effluent pump station would be needed.

Effluent gravity sewer capacity is determined using the Hazen-Williams or Manning equation, with the latter generally found to be more conservative. The Manning equation can be used to calculate flow capacity as follows:

Equation

Q = VA = A(1.486/n)R2/3S1/2

where, Q = flow rate (cfs)
V = flow velocity (ft/sec)
A = pipe cross-sectional area (sq. ft.)
n = Manning coefficient of roughness (dimensionless)
R = pipe hydraulic radius [diameter/4] (ft), and
S = pipe slope (ft/ft).

Performing the calculations using the actual I.D. of Sch. 40 pipe and an n-value of 0.013, it is seen that, with slope set at 0.25%, a 2" pipe can carry about 0.016 cfs, or about 7.3 gpm, and a 3" pipe can carry about 0.047 cfs, or about 21.1 gpm. Note that these capacities assume open-channel flow. Capacity would increase if the line were surcharged, which would be allowable using the cleanout/drop pipe arrangement shown in Figure 4.

Design guidelines for small-diameter effluent gravity sewers suggest assuming a design flow rate of 0.6 gpm per connection. Using this criterion, a 2" pipe could accommodate about 12 houses, and a 3" pipe could accommodate about 35 houses. In many cases for the colonias that have been studied, a 2" pipe would suffice. For conservatism, however, cost estimates shown later in this paper were generated assuming that only the service connections would be 2" pipe, while all main lines would be 3" pipe.

2.5 Permitting of Effluent Sewer Systems

Information obtained from the TNRCC indicates that design standards which would be applied to interceptor tanks and effluent sewers would generally conform to those outlined in the forgoing. Permitting of effluent sewer systems within the context of an organized management system is not anticipated to pose significant problems.

2.6 Maintenance of Effluent Sewer Systems

A suggested maintenance protocol for the MGES or STED/STEP alternative collection system includes:

• Cleaning of effluent filters in the interceptor tanks every two years. The management entity must have right-of-entry to execute this cleaning, and the system users must be agreeable to having a hose connected to one of their hose bibs to provide water to clean the effluent filters, or the management entity must have another source of cleaning water available.
• Observing sludge depth in each interceptor tank every two years and pumping each tank when the sludge depth intrudes into the prescribed sludge clear depth, expected to be in excess of 5-year intervals, as outlined above. The management entity must have right-of-entry to execute the inspection. If the management entity has direct responsibility for the pumping, it must have right-of-entry for that purpose as well.
• Check all pumps and control systems in STEP tanks at six-month intervals to assure that they are operating properly. All pumps would be equipped with alarms to immediately signal a malfunction.
• When they malfunction, pumps or controls must be replaced. Since collective STEP tanks used in the STED/STEP concept would be fitted with duplex pumps, response time is not particularly critical. The pump could be replaced during normal working hours on the working day following notice of the alarm. Simplex pumps would be used in any individual home STEP tanks. In these situations, pump tank volume would be sized to provide one day's flow above the alarm level, so a failed pump must be replaced within one day after the alarm goes off, or the users must stop loading the system until the pump is replaced. Expected average service life of pumps is at least 7 years, so malfunctions should be very infrequent.
• At very infrequent intervals, sewer lines may need to be "snaked" or flushed to clean them. Well designed effluent sewer systems typically have not required routine cleaning. Any cleaning which might be needed is covered under the "general system maintenance" category in the cost tables in this paper.

3. LOW-PRESSURE-DOSED DISPERSAL SYSTEMS

The low-pressure-dosed system concept is illustrated on Figure 6. This shows that any number of houses can flow to a dosing tank, which would disperse septic tank effluent to LPD fields. The general design of LPD trenches is shown on Figure 7.

3.1 Advantages of LPD Fields over Conventional Drainfields

Design features of LPD systems provide protection against either treatment or hydraulic failure of the drainfield. Hydraulic failure occurs when the soil cannot accept effluent at the rate it flows into the trench, typically resulting in the effluent surfacing in or near the drainfield area. This is the failure mode which is generally recognized by users and local regulatory authorities-the effluent fails to go "away". Treatment failures occur when the effluent does indeed percolate "away" but has not been adequately treated to prevent water pollution whenever it gets to wherever "away" is.

Conventional trenches are prone to both types of failures in any soil. One reason is that they are installed fairly deeply, so most of the effluent flows into the soil below the root zone. As illustrated in Figure 7, LPD trenches are very shallow, so that effluent is introduced into the biologically active upper soil horizons. This not only provides better treatment of percolating effluent, but also maximizes the potential for effluent to be lost from the surface soils through evapotranspiration into the atmosphere rather than by deep percolation, thus minimizing the amount of effluent which percolates into groundwater. The shallow, narrow trenches act as a "pseudo-drip" irrigation system, providing some beneficial reuse of wastewater system effluent as well as protection against drainfield failure.

Another reason that conventional trenches are prone to failure is that they are gravity-dosed. Uniform distribution over the entire trench is not achieved until the entire trench is on the verge of hydraulic failure. Typically, a small portion of the field accepts all the flow until it becomes clogged and forces effluent to flow further down the trench. This cycle repeats until the entire trench becomes clogged-the "creeping failure" phenomenon-and effluent surfaces. This is the major reason that conventional systems are expected to have a limited service life and would eventually fail hydraulically and have to be replaced. While hydraulic failure creeps along, soil below the parts of the trench where effluent is flowing are typically overloaded, and-depending on the depth and quality of soil in the field area-treatment failures can result.

In an LPD system, effluent is pumped into the trenches in a manner which provides uniform distribution over the entire field area every time the field is dosed. This prevents localized overloading and creeping failure. Also, the dosing and resting cycle enforced by the pressure distribution system allows the trench face to aerate between doses, which further minimizes the likelihood of clogging.

It is also significant that, because they are gravity-dosed, conventional trenches are only appropriate for very small flows-e.g., a single house-even under the best of circumstances. As total flow rate increases, the localized overloading problem becomes more severe and pressure distribution is required to prevent early hydraulic failure regardless of whether treatment failure is a concern.

3.2 LPD Field Design

In the collective LPD systems following the concept of Figure 6, the LPD fields would be on easements or land purchased to house them. These fields could be on vacant lots within the colonia, if these lots could be obtained for a competitive price, or on surrounding land. In the latter case, a long, narrow easement along back lot lines could house the fields while creating minimal disruption of agricultural practices on the adjacent land. Alternatively, the field area could be configured more as shown in Figure 3 to create an area that could be used to grow a cash crop or as a neighborhood park, perhaps engaging the neighborhood in maintenance of the grounds.

The LPD fields could enhance some agricultural practices by providing an irrigation benefit. The value of a crop could be used to defray system operating costs, or to reduce or eliminate the cost of an easement. Since the pipes would be as shallow as 6" from the surface, any plantings would have to be a perennial standing crop, such as trees or hay, since the soil could not be tilled to any significant depth, and heavy machinery could not be allowed to run over the field.

LPD dispersal fields would be designed in accordance with provisions of Chapter 285 of the Texas Administrative Code, "On-Site Sewage Facilities". The intertrench spaces are counted as field area, since it is expected that evapotranspiration losses would cause effluent to "wick" into these spaces. Chapter 285 stipulates that a field area of 3 square feet be credited for each linear foot of LPD trench; that is, lateral spacing of trenches is 3 feet center to center in the most compact configuration. Actual trench spacing would be largely determined by construction practices.

Field size is based on soil type. As an example, most of the soils covering the colonias in one study classify as clay loam. The maximum loading rate stipulated in Chapter 285 for this soil type is 0.2 gallons/sq. ft./day. Total field area is determined by dividing this loading rate into the design daily flow rate. Studies conducted on South Texas colonias to date have assumed a flow rate of 300 gallons/house, deriving from a per capita flow rate of 60 gallons/day and an occupancy of 5 persons/house. (This provided a design flow rate at or above that estimated from actual occupancy and observed water use in these colonias.) At this rate, a field area of 300/0.2 = 1,500 sq. ft. per house would be required. A greater land area may be required if trench spacings were made greater than 3 feet for construction efficiency.

3.3 LPD Field Dosing System

As shown in Figure 6, effluent sewers would flow into a dosing station tank located adjacent to the dispersal field area. This tank, illustrated in Figure 8, would have a septic tank chamber and a dosing tank chamber. A suggested practice is to size the septic tank chamber to provide a volume equal to 1/4 to 1/2 of the daily design flow running to this dosing station. Fitted with an effluent filter, this septic tank would provide additional clarification of effluent before it is dosed into the dispersal field. It will also settle out solids that may be generated in the effluent sewer pipes, where slimes may grow and be sheared off and washed down to the dosing station.

As indicated in Figure 6, the field would be "zoned" into areas which are dosed sequentially, using an automated flow switching system. By splitting the field up like this, instantaneous flow rate required is limited. This will avoid the need for large and costly pumps, thus reducing system operations and maintenance liabilities. A "human scale" pump can be used, one which can be readily changed out by a single operator rather than requiring that a crew be assembled to do this job.

A simple control system for this flow switching is also desired to minimize maintenance liabilities. This urges the use of a passive, mechanical switching system, such as the Hydrotek automatic switching valve. A strategy used in planning these types of systems for colonias in Hidalgo County is illustrated in Figure 9. The Hydrotek valve can switch among 2, 3 or 4 zones. When more than 4 zones are required, one "master" valve would be used to feed multiple "slave" valves. This would allow each duplex pump system to feed up to 16 zones (4 outlets on the "master" valve being in turned switched to 4 zones by each "slave" valve). Where more than 16 zones are required to accommodate the total flow rate, more than one set of duplex pumps could be installed in the dosing tank, each feeding a portion of the dispersal field, again to avoid the use of much larger and more costly pumps.

Advantage can be taken of this situation by routing flow to more than one dosing station and feeding the dispersal field from more than one point, as illustrated in Figure 3. This strategy would allow effluent pump stations to be avoided or reduced in number, and would also reduce the total length of sewer line required.

3.3 Permitting of LPD Systems

Discussions with TNRCC indicate that collective LPD systems could be permitted using the design assumptions outlined above. In all cases where the total design flow rate of the overall system is in excess of 5,000 gpd, this process would fall under "municipal" system permitting procedures (Chapter 317) rather than under rules governing "on-site" systems (Chapter 285). The main difference is the municipal permitting process would require explicit consideration of the potential impact of these systems on groundwater, while the Chapter 285 process merely requires documentation of a minimum soil depth. Detailed soils analysis would be required, and all wells within 1/2 mile of the field area would have to be identified and evaluated for potential impact. Assuming that soil and site conditions are appropriate for this strategy, permitting of these systems is not expected to encounter significant problems.

3.4 Maintenance of LPD Dispersal Systems

Operations and maintenance (O&M) of an LPD system is relatively simple and straightforward. The suggested maintenance protocol includes the following activities. This protocol forms the basis for the estimates of maintenance effort shown in cost tables in Section 6.

• Check all pumps and control systems in dosing tanks at six-month intervals to assure that they are operating properly. All pumps systems would be equipped with alarms to immediately signal a malfunction.
• When they malfunction, pumps or control devices must be replaced. Since collective LPD system dosing tanks would be fitted with duplex pumps, response time is not particularly critical. The pump could be replaced during normal working hours on the working day following notice of the alarm. For individual home LPD systems employing a simplex pump system, dosing tank volume would be sized to provide one day's flow above the alarm level, so a failed pump must be replaced within one day after the alarm goes off, or the users must stop loading the system until the pump is replaced. Expected average service life of pumps is at least 7 years, so malfunctions should be very infrequent.
• Mow the field area as needed, if the field area is not being actively cropped. This activity would be covered by the "general system maintenance" category in the cost tables.
• If the area is not routinely observed during mowing or tending of a crop, a walkover inspection should be performed at least every three months to look for evidence of surfacing effluent. In any case, the dosing system of a multi-zone field should be checked at this frequency to assure that effluent is being properly distributed among the zones.
• Repair of broken drainfield pipes, malfunctioning switching valves, etc., as needed. This activity would be covered by the "general system maintenance" category in the cost tables.
• At very infrequent intervals, field laterals may need to be flushed. This has not been observed to be routinely required in LPD systems but may be necessary on occasion, especially if the pretreatment system is not serviced properly. This activity would be covered by the "general system maintenance" category in the cost tables.
• Clean screens in the in-line filters on the pump discharge lines periodically. It is expected that cleaning every 6 months would be the maximum frequency required.
• The dosing tank should be cleaned whenever a layer of sludge builds up to such a level that the dosing pump might suck it in. The pump can be set on a block so that some sludge could build up on the bottom of the tank without being sucked into the pump. If the pretreatment system is properly serviced, dosing tank pumping should be required very infrequently. A 4-year interval is assumed in all the cost analyses in Section 6.

4. INNOVATIVE/ALTERNATIVE TREATMENT TECHNOLOGIES

Since treatment plants would be small and dispersed throughout the jurisdiction of the management entity, they should employ technologies that will incur low operational and maintenance liabilities. These plants should also be cost efficient to install at small scale. The two treatment technologies which best meet these criteria are biofiltration systems (sand filters being the most familiar type) and constructed wetlands. Biofiltration technology offers the advantages of a long track record of reliable, consistently high-quality performance, a smaller land area requirement, more predictable operations and maintenance liabilities, and capability of being installed in covered, essentially sealed housings, which is critical for installation close to residences.

Wetlands are likely to require less labor to operate and maintain, unless the plants are overly stressed by heat, predation, parasites, or disease. Having a much shorter-and spotty-operating history, design requirements to consistently produce discharge quality effluent are less well defined. Also, wetland beds will evapotranspirate a large portion of the effluent, especially during the hotter months. This may increase concentrations of various constituents in system effluent. If an irrigation reuse opportunity is available, this consumptive use would reduce the amount of water available for reuse most severely just at the time when irrigation is typically most needed. Further, the bed will intercept rainfall, which would then flow to the system's effluent tank. This will increase the effluent flow just when rainfall reduces the need for irrigation water. Therefore, wetlands are a poor match for systems where effluent is to be dispersed in drainfields or irrigation systems to accomplish true beneficial reuse. They may, however, be a good low-cost treatment method for systems that will discharge effluent to surface waters.

As noted, these two technologies are chosen for decentralized concept systems because they are relatively easy to operate and maintain. Typically, as long as routine O&M activities are attended to in a timely manner, biofiltration and wetland systems basically operate trouble-free and do not require any further attention. As outlined below, the routine O&M activities are fairly straightforward and easy to execute. This ability to operate without intensive oversight is a key to being able to deploy dispersed treatment systems without incurring untenable O&M liabilities.

These technologies consume far less land than lagoons-the low-tech, easy-to-operate technology of choice for more centralized systems. Biofilters and wetlands are also capable of consistently producing a much higher effluent quality than lagoons. This is not only important to obtaining a discharge permit, but also enhances the prospects for beneficial reuse of the "waste" water resource. Also, biofiltration plants and subsurface flow wetlands can be installed with far smaller buffer zones than lagoons.

4.1 Biofiltration Treatment System Design

The biofiltration treatment system concept is displayed in Figure 10. An example of an actual system layout for a small-scale collective biofiltration plant system was illustrated in Figure 2. Granular media ("sand") filters are the most familiar form of this process. It has been known for about 50 years, since the landmark studies at the University of Florida, that-when using larger media and more frequent dosing-"sand" filters can be loaded quite a bit more heavily than is quoted in EPA literature. Over 25 years ago, it was shown that employing recirculation enhanced the efficiency of the process. These characteristics have been confirmed in many efforts over the past several years. Recently, alternative types of media-textile and foam-have been researched. These media can be loaded much more heavily than a granular media filter, offering the potential for a much smaller and more cost efficient biofiltration bed. (A review of biofiltration technology can be found on the web at www.venhuizen-ww.com - click on the "Biofiltration Technology" button.)

Small-scale biofiltration systems to serve colonias should be designed on the basis of this body of knowledge, rather than being constrained by outdated notions about the capabilities of this technology. An average design loading rate of 7 gallons per square foot per day is used to size granular media filter beds for the purpose of generating the cost estimates shown in the tables in Section 6. It is reported that textile or foam media filters could be loaded at around 20 gallons per square foot per day. It is not known at this writing whether this would result in material cost savings, as these media are much more costly than granular media. However, it would significantly decrease the land area required to house the biofiltration treatment plant and perhaps allow significant savings on installation cost.

As shown in Figure 10, the filter bed should be designed to be covered rather than open to the air. This is required for odor control if the beds are located close to residences, as they are likely to be in a decentralized concept system. Covers should also be used to exclude rainfall and windblown debris and weed seeds to minimize maintenance in any case, but a less expensive cover might be used if filter beds are located well away from any residences. If textile or foam media filters are used, this offers the possibility of using pre-manufactured units with integral covers.

4.2 Permitting of Biofiltration Systems

Discussion with TNRCC indicates that the "advanced" design concepts for filter bed sizing and/or the use of "alternative" media filters would be considered on their merits. TNRCC also indicated that buffer zone requirements could be waived if the filter beds were sealed to preclude odor problems. Similar biofiltration treatment systems have been approved by TNRCC over the last decade, under both the on-site and municipal permitting procedures. Therefore, it is not expected that permitting of biofiltration treatment systems would encounter significant regulatory barriers.

TNRCC rules stipulate the frequency of sampling for water quality parameters, mainly BOD5 and TSS. Plants of the size discussed in Section 6 are expected to require two samples per month. It is suggested that more continuous monitoring of turbidity-a good "indicator" parameter for effluent quality-and much less frequent sampling for lab analysis would provide a more cost efficient way to monitor small dispersed plants, while offering perhaps superior assurance of continuous proper operation. (As noted previously, biofiltration plants typically produce consistent quality effluent in any case. The current rules were adopted in recognition of the variability of effluent out of activated sludge plants.) As the cost tables in Section 6 show, sample collection and analysis are expected to be a significant portion of routine operating costs. When permitting these small biofiltration treatment plants, TNRCC should examine the merit of this alternative monitoring strategy.

4.3 Maintenance of Biofiltration Treatment Systems

As noted, operations and maintenance of a biofiltration treatment system is fairly simple. A suggested maintenance protocol includes:

• Check dosing pumps and control system every month. Observe that wastewater is being properly distributed over the bed area. Pump systems would be fitted with an alarm to immediately signal a malfunction.
• When they malfunction, pumps or controls must be replaced. Since collective biofiltration treatment system dosing tanks would be fitted with duplex pumps, response time is not particularly critical. The pump could be replaced during normal working hours on the working day following notice of the alarm. Expected average service life of pumps is 7 years, so malfunctions should be very infrequent.
• Observe condition of filter bed every month. If water ponds on the bed surface for a significant amount of time after a dosing event, the bed is beginning to clog. Raking the surface may alleviate clogging temporarily.
• If a granular media filter bed begins to clog, it is cleaned by flooding it, agitating the bed with an air compressor, and suctioning water off the top to remove solids which "float" out when the bed is agitated. If the pretreatment system is being properly serviced, bed cleaning is not expected to be required more than once every 5 years. This frequency is assumed in the cost analyses for this study. A textile or foam media filter bed is cleaned by simply washing the media.
• Mow/maintain landscaping on the plant site as needed. This activity would be covered by the "general system maintenance" category in the cost tables.
• Repair of broken feed pipes, spray heads, malfunctioning switching valves, etc., if needed. This activity would be covered by the "general system maintenance" category in the cost tables.
• Clean any filters on pumps periodically, if required. It is expected that cleaning every 3 months would be the maximum frequency required.
• The dosing tank should be cleaned whenever a layer of sludge builds up to such a level that the dosing pump might suck it in. The pump can be set on a block so that sludge depth must be significant before it would be sucked into the pump intake. If the pretreatment system is properly serviced, dosing tank pumping should be required very infrequently. A 4-year interval is assumed in the cost analyses in Section 6.
• Change UV disinfection system bulbs at required intervals. A 1-year replacement cycle is assumed in the cost tables. Clean bulb sleeves at required intervals. This activity would be covered by the "general system maintenance" category in the cost tables.
• Collect and analyze water quality samples at the interval stipulated by the system permit.

4.4 Wetland Treatment System Design

The wetland system concept is essentially the same as the biofiltration treatment system, with the wetland bed replacing the biofiltration bed in Figure 10, except that wetland systems probably would not employ recirculation concepts. Wetland bed size is typically determined using methods available in several texts on this subject. Many design innovations could be entertained to create more stable and robust wetland systems. In general, it is presumed that a subsurface flow wetland would be preferred for dispersed treatment centers, as this precludes a free water surface, thus minimizing the prospect for problems with mosquitoes or other disease vectors. Additional study would be required to definitively determine design parameters appropriate for small-scale treatment systems serving the colonias.

4.5 Permitting of Wetland Treatment Systems

Discussions with TNRCC indicate that wetland system permitting would be reviewed on a case-by-case basis. Past experience indicates that subsurface flow wetlands could be permitted for small systems installed fairly close to residences.

4.6 Maintenance of Wetland Treatment Systems

As noted, operations and maintenance of a wetland treatment system should be fairly simple, as long as the wetland plants remain in good health. A suggested maintenance protocol includes:

• If system design requires pumps to distribute wastewater into the wetland beds, check dosing pumps and control system every month. Observe that wastewater is being properly distributed among the bed areas, if multiple parallel beds are employed. Pump systems would be fitted with an alarm to immediately signal a malfunction.
• When they malfunction, pumps or controls must be replaced. Since collective wetland system dosing tanks would be fitted with duplex pumps, response time is not particularly critical. The pump could be replaced during normal working hours on the working day following notice of the alarm. Expected average service life of pumps is 7 years, so malfunctions should be very infrequent.
• Observe condition of wetland bed every month. If water is ponding on the bed surface, this indicates that the gravel bed is becoming too highly clogged to transmit the flow it is receiving without flowing over the surface of the gravel. This should be precluded by design, but the bed can clog excessively if the pretreatment system is not being adequately serviced. The bed should be drained and flushed if surface flow is constantly occurring.
• Opinion is divided on the need to remove dead plant tops from the system. If this is determined to be needed, plant tops would be chopped off and removed when they go dormant in the winter, if the plants do go dormant in this climate. Otherwise, chop and remove at any convenient time during the winter.
• Mow/maintain landscaping on the plant site as needed.
• Repair of broken feed pipes, malfunctioning switching valves, etc., if needed.
• Clean any filters on pumps periodically, if required. It is expected that cleaning every 3 months would be the maximum frequency required.
• If a dosing system is used, the dosing tank should be cleaned whenever a layer of sludge builds up to such a level that the dosing pump might suck it in. If the pretreatment system is properly serviced, dosing tank pumping should be required very infrequently.
• Change UV disinfection system bulbs at required intervals. A 1-year replacement cycle is assumed in the cost tables. Clean bulb sleeves at required intervals.
• Collect and analyze water quality samples at the frequency stipulated in the system permit.

5. EFFLUENT REUSE STRATEGIES

Typically the focus of the "waste" water management system is indeed on how to waste this water in the most cost efficient and institutionally acceptable manner. However, as noted in the foregoing sections, "waste" water could serve irrigation demands, which abound in South Texas. Other non-potable demands, such as toilet flushing, could also be supplied with reclaimed water, assuming proper controls are in place. What is often looked upon as a nuisance to be gotten rid of can instead be addressed as a resource with revenue potential. Various strategies can be employed to beneficially reuse the effluent, including:

• A redistribution system back to the lots generating the flow. Usage would be restricted to lawn and garden watering, and perhaps other exterior uses such as washing cars. The amount of post-treatment that would be required to allow such usage is within the capabilities of biofiltration treatment concepts, with perhaps a post-polishing process added on. There is of course a liability issue to consider. This might be blunted by restricting on-lot reuse to subsurface drip irrigation systems.
• Growing crops on land purchased or leased by the operating entity to serve as the "disposal" area. This strategy was suggested in the discussion of LPD systems. Utilizing biofiltration or wetland system effluent in spray or drip irrigation systems would maximize beneficial reuse of this resource.
• Selling the water to an adjacent property owner for agricultural irrigation reuse.
• Discharging system effluent into a supply canal instead of a drainage canal. This returns the water to the irrigation district. It must be determined if there is a mechanism by which the wastewater system operator can assess a charge for making this water available. However, since supply canals also serve as the source for potable water systems in this area, a higher level of treatment might be required. (Given the pollution level of Rio Grande River water, it is highly questionable if biofiltration system effluent could create any additional hazard.)
• Organizing the management system for a commercial center as a flush water recycling system, with any residual flow routed to drip irrigation of landscaping on the project.

When a high quality effluent meeting discharge standards is produced, it should be possible either to discharge or to irrigate, depending on the presence or absence of irrigation demands. The strategy of discharging effluent when there is no irrigation demand, typically during wet weather, and irrigating when practical is termed "opportunity reuse". This strategy can be employed to obviate the need for expensive storage facilities, making reuse a more financially feasible option. The regulatory constraints on this option currently require further investigation.

6. COST ESTIMATES FOR DECENTRALIZED CONCEPT SYSTEMS

A preliminary planning effort, conducted as part of a study funded by the Texas Water Development Board, generated system plans for four colonias in Hidalgo County for the small-scale collective and large-scale collective strategies employing both treatment plants and LPD dispersal fields. These plans were used to prepare cost estimates for implementing these strategies in those colonias. The cost estimates for one of those colonias are reviewed here as a guide to expected implementation costs. Cost estimates for system operation and maintenance were also generated in the study, and those are also reviewed for each option.

The colonia reviewed here is Engleman Estates, a development containing 83 lots, 64 of which were occupied at the time of the study. It is located approximately 8 miles east-northeast of Edinburg. To hook this colonia into the nearest treatment plant at San Carlos would require over 4.5 miles of force main and 3 lift stations. The terrain is flat, drainage is poor, and the colonia is surrounded by open land. Engleman Estates is typical of many South Texas colonias that are located at considerable distance from an existing sewer system.

Four system plans are evaluated here. These are the small-scale collective strategy employing several dispersed LPD fields, the small-scale collective strategy employing dispersed biofiltration treatment plants, the large-scale collective strategy employing one large LPD field, and the large-scale collective strategy employing a single larger biofiltration treatment plant. For the treatment options it was assumed that the treated water would be discharged into an adjacent irrigation drainage canal.

Assumed design flow rate for all options was 300 gallons per day (gpd) per house. This derives from an actual occupancy of 5 persons and a per capita flow rate of 60 gpd. Using this criterion, total design flow rate for all 83 lots would be 24,900 gpd. For the LPD dispersal fields, a loading rate of 0.2 gallons/ft2/day was used, which is appropriate under current rules for the soil type in the fields surrounding Engleman Estates. The TWDB-mandated discount rate, used to convert future costs into a net present worth, was 7% at the time of this study. For all options, it was presumed that 7 additional lots would be built on in 2 years, 6 additional lots would be built on in 7 years, and the remaining 6 lots would be built on in 12 years.

6.1 Small-Scale Collective Strategy Employing Dispersed LPD Fields

The system plan for this strategy is shown in Figure 11. As noted in the discussion of the small-scale collective strategy, LPD dosing stations are placed so that all lots tributary to each one can flow to it by gravity. At Engleman Estates, this resulted in four LPD systems, two serving 20 lots each (6,000 gpd), a third serving 21 lots (6,300 gpd), and the fourth serving 22 lots (6,600 gpd). The LPD fields would be installed on strips of land along the boundary of the colonia. For the small-scale collective strategy, it was determined that LPD field area required for full development should be installed during initial construction. With the field area being split up and the vacant lots being randomly distributed among the individual systems, it would be logistically difficult to add field area at the proper time.

The estimated construction and O&M costs of this strategy are shown in Table 1. The estimated initial construction cost to serve the 64 occupied lots works out to $3,380 per lot. Adding engineering (surveying, design, construction management and permitting) and land costs brings the total cost to $4,590 per lot. Also shown in Table 1 are the projected costs of accommodating the 19 additional lots if they were to be built upon. Adding these costs, discounted to derive a present value, to the initial costs and dividing all the costs over all 83 lots, the projected total cost at buildout would be $3,698 per lot. The projected O&M cost to run this system is calculated to be $7.20 per month per lot served initially using the assumptions shown in Table 1. This drops to $5.93 per lot served at buildout. Total net present worth of both installation and 20 years of O&M for this system is $368,046.

Table 1 Small-Scale Collective Management Strategy Using LPD Dispersal Fields

Initial System Installation Costs for 64 Lots

Future Construction Costs for 19 Additional Lots

OPERATIONS AND MAINTENANCE COSTS

6.2 Small-Scale Collective Strategy Employing Dispersed Biofiltration Treatment Plants

Figure 12 shows the system plan for this option. Similarly to the LPD option, the treatment plants are located so that all lots tributary to each can flow to it by gravity, resulting in there being four treatment centers, two serving 20 lots each (6,000 gpd), one serving 21 lots (6,300 gpd), and one serving 22 lots (6,600 gpd). The treatment plants would be contained within small tracts on the boundary of the colonia. Effluent from all four treatment centers would be piped to a common disinfection unit at a single outfall point, even though as shown in Figure 12 this would require an effluent pump station to route the flow from one side of the colonia to the discharge point. This is done so that there would only be one outfall to monitor; otherwise monitoring costs would be excessive. (As discussed previously, it is to be hoped that in the future the regulatory system would be amenable to remote monitoring of easy-to-detect indicator parameters like turbidity instead of requiring frequent testing for BOD5 and TSS at each outfall, thus obviating the need to create a single outfall.) As was the case for the LPD strategy, it was determined that all of the filter bed capacity that would be required at buildout should be installed during initial construction.

The estimated construction and O&M costs of the small-scale collective strategy using biofiltration treatment are displayed in Table 2. The estimated initial construction cost to serve the 64 occupied lots works out to $3,732 per lot. Adding engineering (surveying, design, construction management and permitting) and land costs brings the cost to $4,687 per lot. Note that, while the construction cost of this option is estimated to be somewhat higher than construction cost of the small-scale LPD system, the lower land requirement results in the overall estimated cost being much closer to the LPD system cost. Also shown in Table 2 are the projected costs of accommodating the 19 additional lots if they were to be built upon. Adding these costs, discounted to derive a present value, to the initial costs and dividing all the costs over all 83 lots, the projected total cost at buildout would be $3,773 per lot. The projected O&M cost to run this system is calculated to be $13.68 per month per lot served initially using the assumptions shown in Table 2, dropping to $10.93 per lot served at buildout. Total net present worth of both installation and 20 years of O&M for this system is $426,691. While the installation cost of this option appears to be competitive with the LPD system, the higher O&M costs, driven mainly by the sampling and monitoring requirements, result in a somewhat higher total net present worth.

Table 2 Small-Scale Collective Management Strategy Using Dispersed Biofiltration Treatment Plants

Initial System Installation Costs for 64 Lots

Future Construction Costs for 19 Additional Lots

OPERATIONS AND MAINTENANCE COSTS

6.3 Large-Scale Collective Strategy Employing an LPD Dispersal Field

The system plan for this option is illustrated in Figure 13. Under this strategy, effluent pump stations are required in the collection system and the STED/STEP sewer concept is used to route septic tank effluent to the dosing station for the single large LPD field. The field would be contained on one larger tract. As noted previously, this area could be used to grow a cash crop, or it could be a neighborhood park for colonia residents. Under this strategy, it is considered reasonable to defer some of the field installation until additional houses are added. The initial installation is sized to serve the existing 64 houses plus the 7 houses that were presumed to be added after 2 years (21,300 gpd total capacity), and the remaining field area would be installed in the 7th year. To maintain the concept of "human scale" pumping systems discussed previously, the field dosing system would utilize 3 independent sets of duplex pumps, each serving a portion of the field area.

The estimated construction and O&M costs of the large-scale collective strategy using an LPD dispersal field are displayed in Table 3. The estimated initial construction cost to serve the 64 occupied lots works out to $3,536 per lot. Adding engineering (surveying, design, construction management and permitting) and land costs brings the cost to $4,767 per lot. Also shown in Table 3 are the projected costs of accommodating the 19 additional lots if they were to be built upon. Adding these costs, discounted to derive a present value, to the initial costs and dividing all the costs over all 83 lots, the projected total cost at buildout would be $3,898 per lot. The projected O&M cost to run this system is calculated to be $8.13 per month per lot served initially using the assumptions shown in Table 3. O&M cost falls to $6.65 per month per lot served at buildout. Total net present worth of both installation and 20 years of O&M for this system is $391,842.

Table 3 Large-Scale Collective Management Strategy Using LPD Dispersal Field

Initial System Installation Costs for 64 Lots

Future Construction Costs for 19 Additional Lots

OPERATIONS AND MAINTENANCE COSTS

6.4 Large-Scale Collective Strategy Employing a Biofiltration Treatment Plant

Figure 14 illustrates the system plan for this option. Under this strategy again, effluent pump stations are required in the collection system to route septic tank effluent to the dosing station for the single larger plant. The plant would be contained on one larger tract, as shown in Figure 14. Effluent would be disinfected and discharged into the drainage channel. Under this strategy, it is considered reasonable to defer some of the filter bed installation until additional houses are added. The initial installation is sized to serve the existing 64 houses plus the 7 houses that were presumed to be added after 2 years (21,300 gpd capacity), and the remaining filter bed area would be installed in the 7th year. To maintain the concept of "human scale" pumping systems discussed previously, the filter bed dosing system would utilize 4 independent sets of duplex pumps, each serving a portion of the filter bed area.

The estimated construction and O&M costs of the large-scale collective strategy using biofiltration treatment are displayed in Table 4. The estimated initial construction cost to serve the 64 occupied lots works out to $3,666 per lot. Adding engineering (surveying, design, construction management and permitting) and land costs brings the cost to $4,607 per lot. Here again, while the construction cost of this option is estimated to be higher than the cost of the large-scale LPD system, the lower land requirement results in the overall estimated cost being a little lower in this case. Also shown in Table 4 are the projected costs of accommodating the 19 additional lots if they were to be built upon. Adding these costs, discounted to derive a present value, to the initial costs and dividing all the costs over all 83 lots, the projected total cost at buildout would be $3,783 per lot. The projected O&M cost to run this system is calculated to be $14.44 per month per lot served initially using the assumptions shown in Table 4. This falls to $11.51 per month per lot served at buildout. Total net present worth of both installation and 20 years of O&M for this system is $431,862. As for the small-scale strategy, the installation cost of this option appears to be competitive with the LPD system, but the higher O&M costs, driven mainly by the sampling and monitoring requirements, result in a somewhat higher total net present worth.

Table 4 Large-Scale Collective Management Strategy Using Biofiltration Treatment Plant

Initial System Installation Costs for 64 Lots

Future Construction Costs for 19 Additional Lots

OPERATIONS AND MAINTENANCE COSTS

6.5 Cost Comparisons

For ease of comparison, the various cost measures for each of the options are shown in Table 5, along with each of those cost measures for a hooking Engleman Estates into the nearest conventional centralized collection system. The latter were derived in the same TWDB-funded study which produced the cost estimates for the decentralized concept options. This information indicates that the total cost at buildout of a decentralized concept strategy is estimated to be about 40% of the total cost of a centralized system at buildout. Add to this the value of the water which is lost to the community and the O&M and pollution problems inherent in a centralized collection system, and it is clear that the decentralized concept is a far more cost and resource efficient approach to management for many South Texas colonias.

Table 5. Comparison of Costs Among System Options

There is little difference in the installation costs of the various decentralized concept options. Due to the higher O&M costs associated with effluent sampling requirements, total net present worth of the treatment plant options are about 15% greater than for the LPD dispersal field options. Clearly, if the LPD field represents a beneficial reuse of the wastewater, the LPD options would be preferred. However, it is to be expected that, in general, a much greater level of beneficial reuse could be accomplished by treating the water to high quality, then delivering it to points of use. The additional cost of the treated water distribution system should be covered by the value of the water at the point of use if the use is truly beneficial.

7. A VISION OF THE FUTURE

The basic idea of the decentralized concept is to treat, and beneficially reuse where possible, the "waste" water as close to where it is generated as practical. This readily lends itself to a management system focused on maximizing reuse potential, thus reducing stress on regional water resources as the population of this area continues to grow. One can envision an array of dispersed small treatment centers supplying irrigation water, toilet flushing, and other non-potable demands. Highly stable biofiltration plants would be installed to receive flow from all the nearby users that could flow by gravity to it through an effluent sewer system. Each of these plants would be, in essence, a little "water factory". If no local reuse opportunity were available to utilize this reclaimed water, the effluent might be piped to a point of beneficial reuse, or it might be discharged into a supply canal, as discussed previously. Irrigation reuse could supply some combination of private greenspace, agricultural land, and habitat enhancement, the latter being an important adjunct to the growing bird-watching tourist industry. Opportunity reuse could be accommodated, simply allowing the treatment system effluent tank to overflow to a discharge if it filled up because water was not being demanded at the reuse sites. The system could respond quickly and cost efficiently to changing demands and opportunities.

The many dispersed plants, since they employ a robust, very stable treatment concept, would require minimal oversight. A "circuit rider" operator would do routine inspections. A remote monitoring system would provide continuous observation of effluent turbidity, and an alarm would sound if a predetermined limit were exceeded. All pump functions would also be remotely monitored. Highly reliable ultra-violet (UV) disinfection systems would be used, with alarms to indicate loss of required light intensity.

Septic tank pumpage would be routed to composting centers, and the system operator would realize a revenue stream from production of a high quality organic fertilizer, closing the nutrient cycle. Private enterprises might compete for pumping contracts to supply their own composting operations, or this business might be conducted directly by the management authority.

This vision will not appear overnight. But there are very good reasons to begin to work toward it. A reuse-focused decentralized concept management system promises to be the most cost efficient and water resource conserving means of providing high quality "waste" water management for South Texas colonias. This strategy offers the potential for systems that are more fiscally reasonable, societally responsible, and environmentally benign than the conventional "wisdom".