STEP 3

Delineate Protection Area


Return to: Chapter 3 Steps

Return to: TABLE OF CONTENTS, Water Quality Home Page, DEQ Home Page


Overview:

The delineation process is a fundamental step in developing a Wellhead Protection Plan in that it identifies the area surrounding the well, wellfield or spring that is at the surface directly above the portion of the aquifer that supplies groundwater to the well, wellfield or spring. The delineation allows for the recognition of the area where protection strategies will have the greatest impact on groundwater quality and allows for more efficient use of limited resources. The Wellhead (citizen) Advisory Committee and the Wellhead Rules and Guidance Committee developed an approach to delineation that considered both protection of the source of their drinking water and the community's available resources.

All delineation techniques acceptable in the Oregon program require consideration of the pumping rate for the well(s). The Oregon program requires the use of an adjusted pump rate of 125 percent of the average daily use over the 3-month period of highest water usage during the year in order to accommodate potential growth in the community. The delineation process frequently results in the identification of the well, wellfield or spring's entire zone of contribution or capture zone (Figure 3-1), an area that ultimately supplies water to the source, but may, because of size, pose significant difficulties with respect to management. Accordingly, Oregon requires only that portion of the capture zone that will supply the water over a 10-year period. That portion of the capture zone that comprises a 10-year time-of-travel (TOT) for groundwater is referred to as the wellhead protection area (WHPA). Delineations must be accomplished by an Oregon registered geologist, engineering geologist, or other licensed professional with demonstrated experience in hydrogeology.

The minimum delineation effort that a system must make in order to obtain a state-certified program is dependent upon the population (Figure 3-2). For communities with populations less than or equal to 500 and obtaining their groundwater through use of a well or wells, delineation can be accomplished through the calculated fixed radius method. This technique uses minimum site-specific data and does not consider important aquifer characteristics or groundwater flow. Systems with significant potential threats to groundwater, evaluated through an optional Pre-Plan Assessment, should choose a more technically defensible method for their delineation.

All other systems, e.g., spring-fed or wells serving more than 500 population, must develop a conceptual model of the groundwater system. This model identifies critical characteristics of the groundwater system and provides for a delineation that is more representative of the actual groundwater conditions. For systems serving populations from 500 to 50,000, the minimum method is combined analytical methods and hydrogeologic mapping. Systems serving 3,300 or more must collect site-specific data (e.g., conduct an aquifer test) to be utilized in the process.

Systems serving more than 50,000 will utilize numerical or comparable analytical methods in the delineation process. These methods are able to handle significant heterogeneities within the aquifer, complex boundaries and variable gradient directions to a much greater degree than analytical methods. These models are also capable of being calibrated to the specific site.

Systems are encouraged to conduct susceptibility analyses within the WHPA. The determination of the WHPA generally considers only travel within the aquifer; i.e., it does not address the travel time from the surface to the aquifer. A susceptibility analysis considers factors related to water movement from the surface to the aquifer and the probability that a given contaminant will migrate to the aquifer. This allows the system to prioritize various parts of the WHPA in terms of the susceptibility and to tailor management strategies to specific activities on the surface.

Communities can reduce the overall expense of the delineation process by conducting parts of the investigation themselves, including the following:

The Oregon Health Division (OHD) will work with the water system/community to complete the delineation process. This includes early discussions regarding the general nature of the system, what the community can do up front, assistance in preparing a Request for Proposals from the consulting community, providing technical input during the delineation and approval of the final delineation product.

Additionally, communities that have groundwater sources close together may wish to pool their efforts, obtaining one consultant to delineate their separate wells. Providing the wells are within the same aquifer and same general hydrologic setting, this combined effort may prove to be less costly than each community developing their own independent delineation. It should be noted, that this effort will be successful only if the consultant is able to use a single conceptual model to describe the hydrogeologic setting of both communities.

Delineation Basics:

Delineation of a wellhead protection area (WHPA) for the well(s) in question is a fundamental aspect of a Wellhead Protection Plan. The U.S. Environmental Protection Agency (EPA) has defined the WHPA as "...the surface and subsurface area surrounding a water well or wellfield, supplying a public water system, through which contaminants are reasonably likely to move toward and reach such water well or wellfield." A contaminant released within the WHPA may ultimately reach the water that is being pumped from the well. Therefore, it is within the boundaries of the WHPA that management activities, designed to eliminate or reduce the threat of groundwater contamination from surface or near surface activities, would be concentrated.

Proposed WHPA delineation techniques vary from drawing a circle of arbitrary radius around the wellhead to more sophisticated computer assisted models that account for site-specific characteristics of the aquifer and of groundwater flow (EPA, 1987). Clearly, the utility of the resulting WHPA increases as the level of delineation sophistication increases. However, so does the cost and level of required technical expertise. From a cost-benefit perspective, however, it is in the best interest of the public water supply to invest now in a proactive program of groundwater protection. Experience clearly indicates that the cost and hardships associated with remediating or perhaps losing a groundwater source as a result of contamination far exceeds the cost of implementing a preventative program.

Participation in Oregon's wellhead protection program is voluntary; however, in order to protect their drinking water resource for present and future generations, all public water systems in Oregon should seriously consider developing a WHP program. In order for a water system to have a state-certified program, the system must address all the elements described in Oregon Wellhead Protection Program Guidance Manual. Flexibility is allowed in structuring the WHP program, however, so that the individual water system's characteristics and needs can be accommodated. The State will provide guidance, technical assistance and will review for approval the delineation of each system's WHPA. In compliance with ORS 672.505 to 672.705, the delineation will have to be accomplished by a registered geologist, a registered certified specialty geologist (e.g., a registered engineering geologist), or other registered professional with demonstrable experience in hydrogeology.

NOTE: At present, Oregon's registration requirements for geologists, engineering geologists, or other professionals do not specifically address hydrogeology, e.g., an individual does not have to be proficient in hydrogeology to become a registered geologist in the state. Therefore, prior to contracting with an individual to perform the delineation, the water system should ascertain that the individual being considered actually has experience in the field of hydrogeology.

What Can Water Systems Do Now?

Water systems can facilitate the delineation process as well as become more informed about their drinking water resource if they play an active role throughout the exercise. It is clear that the water system has a more complete knowledge than anyone else of the history of the system and the occurrence of other wells, irrigation practices, etc., that might affect the delineation. In addition to this expertise, the water system can reduce costs to the community by participating in the data collection process.

Well Inventory - An important part of the delineation is the development of a conceptual model of how groundwater occurs and moves in the area. This will need to be accomplished by your consultant. The data your consultant will use will include well reports from the area (see Appendix A). Compilation of the well reports may be time-consuming and expensive if accomplished by the consultant. It may be less expensive for the water system to obtain the well reports from the local watermaster's office.

In addition to compiling the well reports, the consultant will need to have them plotted on a map to show their distribution. Again, it will likely be less expensive for the water system to accomplish this. It is also probable that with the water system's knowledge of the local area, a more complete map will result. The City of Boardman (Morrow County, Oregon) has developed a guide for preparation of a well report inventory. With the City's permission, OHD has made copies and will distribute those to interested water systems. Systems should communicate with OHD, or their consultant, prior to initiating the survey in order to more efficiently design their efforts.

Even after the delineation exercise, the City of Boardman continues to use their well report inventory in day-to-day issues regarding their system. The inventory has proven to be a very useful management tool for the City.

Aquifer Test - Additional data that a water system can begin to collect include static water levels for their wells and an aquifer test for their production wells. Procedures and recommendations for aquifer tests are described in Appendix A. Under any condition, aquifer tests are costly and under no circumstance should a water system conduct one without first contacting OHD or the Water Resources Department or their consultant for technical assistance. It is important to note that the 4-hour pump test requirement for groundwater right holders required by the Water Resources Department generally is insufficient to obtain representative aquifer characteristics.

Public Education - Lastly, water systems can work with the local community to initiate a public education program. This program should be designed to raise the level of awareness of the citizens of the community regarding groundwater and its vulnerability. Water systems can contact the Department of Environmental Quality or OHD regarding educational materials and programs that might be available to the community.

Request for Proposal

Water systems or communities who are beginning the delineation step in wellhead protection plan development will work with the Drinking Water Program at OHD. OHD will review the delineation requirements with the water system or community and will provide technical assistance in preparing a "Request for Proposal" (RFP) to be distributed to the consulting community.

Prior to and during the delineation process, OHD will provide input to the system and their consultant as needed. The final delineation and supporting documentation must be submitted to OHD for review and approval. OHD may enlist assistance from other agencies, e.g., Water Resources Department, Oregon Department of Geology and Mineral Industries and the Department of Environmental Quality as needed in the review of the delineation procedures.

Delineation Method Requirements:

Figure 3-1 illustrates the distinction between three areas within the aquifer that can be recognized with respect to a pumping well: the zone of influence (ZOI), the zone of contribution (ZOC) and the WHPA. The ZOI, or the drawdown area, is that part of the aquifer where the hydraulic head is lowered as a direct result of the well discharge. For purposes of identification, it is defined here as the area where the drawdown is greater than 0.05 feet (0.6 inches). The ZOC is that part of the aquifer that will supply water to the well in the future. The WHPA is a subset of the ZOC: that part of the ZOC that will supply water to the well for the next 10 years. It is the WHPA, where groundwater protection policies are implemented, that delineation identifies.

In designing the state's delineation requirements, the citizen's advisory committees sought a balance between factors of potential risk, aquifer sensitivity, and available expertise and funds in designing the state's delineation requirements. The delineation requirements also reflect input from the consulting community through demonstration projects at Boardman, Klamath Falls and Springfield, as well as comments solicited by the committee through consultant review of earlier drafts of this document. In subsequent sections, these will be referred to as the "wellhead demonstration projects".

The result of these efforts is a flow chart (Figure 3-2) that bases the delineation method primarily on the population served by the water system. The delineation techniques in the flow chart are discussed in greater detail below. In order of increasing site-specific character, they are: (1) the calculated fixed radius method, (2) combined application of analytical techniques and hydrogeologic mapping, and (3) numerical methods. It must be emphasized that, in most cases, the delineation technique identified in Figure 3-2 is to be considered a minimum only. A water system should inventory the data available to them (see below), and if the data allow, a more site specific delineation method should be used. In addition, the potential risk of contamination of the system's groundwater source should also influence the community's decision on the level of delineation used. Systems may wish to complete the Pre-Plan Assessment option described in Step 1 of this document in order to perform a preliminary evaluation of the number and types of potential contaminants that occur in the vicinity of their well(s) or spring(s).

NOTE: An important finding of the wellhead demonstration projects was that a more site-specific delineation, though costing more initially, resulted in significant savings later. The savings accrued because the site-specific delineation was more legally defensible and often encompassed a smaller area leading to reduced management costs later in the process.

An important concept to remember, both during and after the delineation exercise, is that the aquifer from which a well derives its water is a three dimensional body that may have considerable regional extent, and from which other water systems, as well as numerous private individuals, derive their water. A truly comprehensive groundwater protection plan should address the entire aquifer rather than targeting a limited portion that contributes water to a well over the next 10 years (i.e., the WHPA). In practice, however, such a comprehensive approach would be difficult to implement and as a result, the delineation of the individual well's WHPA is utilized.

The sections below follow the order of the flow chart in Figure 3-2. Note that various water systems will complete their delineation requirements at different parts of the flow chart. It is not necessary that all systems complete all the phases that are indicated in Figure 3-2. Very little discussion or recommendations are given below regarding the specific model(s) to be used in the more complex situations where analytical or numerical methods are employed. It is accepted that the professionals performing those delineations will be well versed in the available techniques.

More discussion is provided with respect to the Calculated Fixed Radius method because of its limited ability to incorporate even the most basic hydrogeologic data into the model. As a result, recommendations on how to treat some of these more common situations are provided. In addition, some effort is expended on providing minimum expectations regarding the data that will be incorporated into the conceptual model(s). It is hoped that this is viewed as a mechanism to facilitate review by the state as opposed to the state dictating procedures to be followed by the consulting community.

Sources of Information

In addition to the discussion below, the reader may wish to become familiar with the EPA guidance documents that address delineation of wellhead protection areas. These documents include:

Copies of these documents can be obtained free of charge by calling the National Center for Environmental Publications and Information at (513) 489-8190.

Parameters Common to All Delineation Methods

Adjusted Pump Rate (Qa) - All WHPA delineation methods acceptable for use in Oregon require incorporating a pump rate in the equations. In order to provide for protection of the groundwater source for potentially expanded use in the future, the minimum pump rate used in the models should be adjusted as follows: 125 percent of the average daily use calculated over the three months that are traditionally the high-demand period for the system. If the average daily use is not available, the adjusted pump rate should be determined by one of the following: 125 percent of the average pump rate based on a comparable size community in the region, the design capacity of the installed pump, or 90 percent of the safe yield of the well (see Appendix A), whichever is the smallest. As an example, consider a community using a well with a safe yield of 600 gpm. No daily use data is available. The design capacity of the pump in the well is 500 gpm. The estimated usage based on a community with similar residential and industrial characteristics is 390 gpm. A value of 487 gpm (1.25 390) would be used in the delineation exercise. If no other data is available to a water system regarding daily water usage, the system may use a figure derived by multiplying the population served by 125 gallons/day.

The determination of Qa is not, of course, a hard and fast rule. It is in the water system's best interest to delineate a capture zone that will accommodate growth in the future. If for example, the water system's master plan calls for the installation of a larger pump in the future to accommodate projected growth, the delineation should reflect that increased demand.

Time of Travel (TOT) - As discussed above, only a portion of the aquifer is delineated by the WHPA methods. For management purposes, the WHPA is generally limited in size by a criterion that is designed to achieve a predetermined protection level. Examples of these criteria are: (1) drawdown of the water table or potentiometric surface imposed by pumping of the well; (2) the aquifer's assimilative capacity (i.e., the distance groundwater must travel through the aquifer to mitigate the contaminant through dilution or breakdown); and (3) the time of travel, a factor related to the time required to respond to the development of a contamination threat at the WHPA boundary.

In all of the Oregon approved techniques the TOT parameter is utilized to constrain the WHPA. The TOT criteria effectively determines the radius of the calculated fixed radius method and the upgradient distance delineated within the analytical and numerical models. For most delineation techniques, a minimum TOT threshold value of 10 years was chosen based on State review of the time required to remediate and/or develop a new source should a significant contamination event take place at, or arrive at, the WHPA boundary. The 10-year TOT should be regarded as a minimum. Longer TOT thresholds are recommended in those cases where the understanding of the groundwater system is more limited (see below) or significant threats to groundwater quality occur.

The water system will probably also wish to determine additional zones within the WHPA based on shorter travel times. After reviewing WHP plans from other states, Oregon recommends the following TOT zones be delineated within the WHPA:

Figure 3-3: Components of the Calculated FixedRadius Equation

Calculated Fixed Radius

Water Systems Affected - For water systems serving a population of 500 or less, and deriving their drinking water from a well or wellfield, a calculated fixed radius (CFR) is an acceptable delineation technique. Water systems should understand that this technique uses only minimal site specific data and does not account for some important aquifer characteristics (e.g., permeability or hydraulic gradient, which control how fast and in which direction groundwater is flowing). As a result of the limitations of using the (CFR) a TOT of 15 years is used for this technique (Figure 3-2). If the pre-Plan assessment indicates a significant number of potential contaminant sources in the vicinity of their well(s), a more site-specific technique should be considered.

Doing the Calculation - The CFR technique is shown diagrammatically in Figure 3-3. The technique assumes a static system, i.e., no regional groundwater flow, and that all of the water released to the well comes from storage within the aquifer proximal to the well. The technique determines the volume of the aquifer that is needed to supply the demand from the well over the TOT period. Because of the above assumptions, the volume of the aquifer supplying the water is in the shape of a cylinder, the radius of which is determined by the demand.

The water demand over the TOT is calculated by multiplying the adjusted pump rate Qa (in cubic feet per year) by the TOT (15 years). To determine the volume of the aquifer needed to supply the demand, we multiply the area of the circle (3.14r2) times the thickness of the water bearing zone (I). However, only part of that aquifer volume contains water that can drain to a well. That part is determined by multiplying the equation for the volume of a cylinder by the effective porosity (n), that fraction of the aquifer that consists of interconnected pore spaces. We then set the volume expression equal to the demand and solve for "r", the radius of the cylinder (see equation in Figure 3-3 and Calculated Fixed Radius example in Appendix A).

Appropriate values are substituted into the equation and the value of "r" is determined. The adjusted pump rate (Qa) is determined as described above. The effective porosity can be estimated using Table 1 (EPA, 1994). The value of "I" should be obtained as appropriate in one of the following ways:

After the value of "r" is determined by the equation in Figure 3-3 (see also the Calculated Fixed Radius example in Appendix A), a circle with a radius of "r" around the wellhead constitutes the WHPA for that well. This is the area at the surface that directly overlies the cylinder determined above. It is this area where released contaminants might percolate downward to the part of the aquifer supplying the well. It is in this area that protection procedures will be applied.

For examples of determining the CFR, and for special cases where the circles overlap, intersect streams, or when the technique is applied to a wellfield (see Appendix A).

Conceptual Model Development

What is a Conceptual Model? A conceptual hydrogeologic model is a three-dimensional portrayal of the groundwater system in the study area. Within the model, the distribution and geometries of the hydrogeologic units, their hydraulic properties, including variations in hydraulic head, the direction of groundwater flow and the location(s) of hydrogeologic boundaries, and areas of recharge and discharge are displayed.

The conceptual model provides the framework for decision making regarding the groundwater system in an area. It provides a vehicle for determining those hydrogeologic features that are especially important in controlling groundwater flow, It also provides for the testing of assumptions and recognizing where more data are needed. A well-conceived conceptual model is fundamental to developing a WHPA delineation that accurately reflects the groundwater system in the area. The components of conceptual models are discussed in Appendix A.

Water Systems Affected - As a general rule, as the population in a community increases, the potential risks to water quality also increase (population density increases and more industry and businesses are required to serve the community). Because of the greater risk, more detailed information is necessary to provide adequate protection of drinking water resources. Therefore, public water systems that have >500 population must accomplish their delineation using techniques that utilize more site-specific data.

Further, the calculated fixed radius method cannot be used for water systems that derive their drinking water from springs. The discharge of a spring cannot be simply related to circle around that spring. As a result, these water systems must obtain more site-specific data in order to delineate the source of their drinking water. A requirement of all systems that are at this point on the flow chart is the development of a conceptual hydrogeologic model (Figure 3-2).

The flow chart in Figure 3-2 indicates two levels of conceptual model development. The first, required of systems with a population of 500 to 3,299, is developed from existing data, often regional in character. By regional, we mean that aquifer characteristics may have been averaged over a large area and that the hydraulic gradient has been determined over an area of which we are concerned with only a small part. With such an approach, local variations are masked. For a particular WHPA, however, local variations from the norm may be the controlling factors for groundwater flow. For smaller systems, with presumably lower risks to groundwater, the regional approach is considered adequate for the delineation exercise. If the pre-Plan assessment indicates a high risk, the system should consider collecting more site-specific data (see below).

For systems with populations of 3,300 or greater, more site-specific information is required in developing the conceptual model. In the site-specific approach, data are collected through aquifer tests, direct measurement of static water levels and the mapping of spatial variations in aquifer characteristics.

Communities that are located in proximity to one another may wish to pursue the development of a common conceptual model through one consultant. What constitutes proximity varies across the state; however, if communities are within several miles of one another, they should consider the possibility that the data may allow a single conceptual model to be developed that will apply to both areas. As an example, consider two communities separated by three miles. These communities have wells that are within the same aquifer and there is no significant difference in the gradient nor are there any hydrogeologic boundaries that separate them. The application of the analytical models discussed below would be able to delineate the separate wells, using specific pumping characteristics, during the same model run. Such a multiple delineation approach, if appropriate, would certainly be less costly to the communities than if each community approached the conceptual model/delineation step independently.


Porous Media Assumption

What is Being Assumed? For the analytical models discussed below, it is assumed that the openings within the aquifer through which the water moves are such that water movement is directly down-gradient, perpendicular to the hydraulic head contours. This assumption is generally valid in an aquifer where the open spaces consist of pore spaces (voids that occur between individual particles that comprise the aquifer). It may not be valid when the open spaces comprise discrete fractures that occur within the rock material that makes up the aquifer.

In a typical porous material, such as sediment, e.g., sand and gravel, the openings are primary, that is they represent the spaces between grains that were formed when the sediment was originally deposited. Consequently, they are numerous and random in occurrence. As a result, the concentration and orientation of the open spaces tends to be isotropic (uniform in all directions) within the aquifer. Groundwater flow is controlled primarily by gradient direction in porous media (Figure 3-4a).

The Impact of Fractures - Fractures are secondary features. They are generated after the aquifer formed, often as a result of stresses applied to the aquifer. Fractures tend to develop in a specific orientation with respect to the direction of applied stress. As a result, the fractures may not be random in their orientation and the aquifer's secondary porosity may be anisotropic (not uniform in all directions). Groundwater in an anisotropic medium will be driven in a down-gradient direction; however, it may be forced to move along the fractures which often are at some angle from the gradient direction (Figure 3-4b).

Figure 3-4: Effect of Fracture Anisotropy on The Orientation of The Zone of Contribution to A Pumping Well

NOTE: A porous medium can still be anisotropic with respect to other characteristics, e.g., hydraulic conductivity and thickness.

Fortunately, there are a number of settings in which fractured rock behaves as a porous medium with respect to transmitting groundwater. At this point in the flow chart (Figure 3-2), an analysis must be made to determine whether the aquifer can be treated as isotropic with respect to its pore space distribution.

As reported by Long and others (1982), fractured rocks behave like porous media when: (1) fracture density is high, (2) fracture orientation is not uniform, (3) fracture openings are relatively uniform, and (4) the volume of aquifer concerned is large. Recommendations for evaluating the porous media assumption are provided in the Evaluation of the Porous Media Assumption section in Appendix A.

The Impact of Heterogeneities in an Alluvial Aquifer - Even alluvial aquifers are not always uniform in character. Highly permeable channel de-posits may occur at depth within the deposit. These channels are rarely in a straight line; they meander back and forth across what was the valley floor when they formed. If these channels are surrounded by low permeability sediments such as silts and clays or bedrock, groundwater will preferentially flow along the channel even if this direction is not immediately down gradient. If the conceptual model identifies significant channel deposits, their impact on groundwater flow in the vicinity of the well(s) should be considered.

The Impact of Heterogeneities Within Layered Volcanic Rocks - Layered volcanic rocks (e.g., basalts, andesites, etc.) may also exhibit preferential flow directions for groundwater. A typical volcanic section will consist of a sequence of erupted rocks (e.g., lava flows, ash deposits, mudflows, etc). Owing to the character of these rocks and the environment in which they occur, groundwater may not behave as if it was in a porous medium. Lava flows typically have rather dense interiors and highly porous and permeable tops and bottoms. The permeable tops and bottoms originate primarily from the highly porous, i.e., vesicular, nature of these zones within the lava. Therefore, a section of lava flows commonly has very low vertical hydraulic conductivity overall, and high horizontal hydraulic conductivity in these vesicular zones. Groundwater preferentially moves within these interflow zones, even if these zones are not perpendicular to the gradient. For example, a vertical gradient may actually produce significant horizontal flow within an interflow zone.

Additional complications may arise from the fact that in an area where volcanic rocks have been erupted, significant discontinuities may occur on the scale of a delineated WHPA. These result from the fact that the surface upon which the lava erupts may not be horizontal. It is common in a typical volcanic sequence that sufficient time between eruptions occurs, allowing the redevelopment of drainage systems and canyon development. Subsequent eruptions may be confined largely to those drainages. These intracanyon flows, often underlain by stream deposits are linear in character as opposed to the sheet-like form that is often associated with lava flows. Groundwater will preferentially move along these intracanyon flows in the subsurface. Features such as lava tubes will further complicate groundwater flow.

Analytical Techniques with Hydrogeologic Mapping

Analytical Techniques - Analytical methods make use of equations that define groundwater flow and contaminant transport and are frequently used in areas with a sloping water table. The analytical equations require that various hydrogeologic parameters be known or can be reliably estimated. When these parameters are substituted into the equations, the resulting solutions provide information regarding the dimensions of the ZOC with respect to the down-gradient divide and the width of the ZOC in the up-gradient direction.

Most analytical models assume that the aquifer is uniform in character, the gradient is constant throughout the modeled area and groundwater flow in the aquifer is two-dimensional in a horizontal plane (vertical flow is not considered). If the conceptual model indicates that any of these assumptions is significantly wrong, the use of a more sophisticated model should be considered.

The WHPA Software - The USEPA, through contractors, has developed software designed to facilitate the determination of wellhead protection areas in hydrogeologic settings that include the above assumptions.

NOTE: There are many programs available commercially and the selection of which one to use will generally be made by the consultant.

Here we briefly review EPA's software as a basis for discussion of parameter requirements.

EPA's programs are inexpensive, easy to use and are well documented. Used in the context of the conceptual model, and within individual program limitations they will provide adequate delineations of wellhead protection areas. Graphic output is limited with the programs, however most of the programs will produce hard copies of the delineation that can be scaled automatically to most common maps. The option to save the model results in an ASCII file that may be used as input to the ARC/INFO proprietary GIS developed by the Environmental Systems Research Institute (ESRI) also exists for most of the programs.

The software package known collectively as WHPA code has the most widespread distribution of EPA's delineation programs. This package consists of four programs: MWCAP, RESSQC, GPTRAC, and MONTEC. All the models are based on an analytical approach and use a particle-tracking routine to delineate the capture zones (Blandford and Huyakorn, 1991). All models assume steady-state groundwater flow. Data input requirements for the techniques are summarized in Table 3-1.

MWCAP. The MWCAP model will produce steady-state, time-related and hybrid capture zones for single or multiple wells within confined or unconfined aquifers. It can accommodate aquifer boundaries (impermeable and stream) but does not consider well interference. Boundaries are modeled as straight lines and fully penetrating. For stream boundaries where the stream is not fully penetrating or if the presence of a clogging layer inhibits the hydraulic connection between the stream and the aquifer, the predicted WHPA will be smaller than the actual capture zone. Forward particle-tracking is an option allowing the path of a given constituent to be determined within the flow field.

RESSQC. RESSQC assumes steady-state groundwater flow conditions and can model multiple discharging and recharge wells. Well interference is determined. The aquifer may be confined or unconfined if the drawdown to initial saturated thickness is less than approximately 0.1 (Blandford and Huyakorn, 1991). The program can also calculate contaminant fronts that migrate away from recharge wells. Boundaries are not addressed automatically and must be modeled through image well placement which must be input by the modeler. Forward and reverse particle-tracking is an option.

Table 3-1: Input Requirements for WHPA Models1

Required Input

RESSQC

MWCAP

GPTRAC

     

Semi-Analytical

Numerical

Units Used

X

X

X

X

Aquifer Type2

   

X

 

Study Area Limits

X

X

X

X

Maximum Step Length

X

X

X

 

Number of Pumping Wells

X

X

X

X

Number of Recharge Wells

X

 

X

X

Well Locations

X

X

X

X

Pumping/Injection Rates (Q)

X

X

X

X

Aquifer Transmissivity (T)

X

X

X

X

Aquifer Porosity (n)

X

X

X

X

Aquifer Thickness (b)

X

X

X

X

Angle of Ambient Flow

X

X

X

 

Hydraulic Gradient (i)

X

X

X

 

Areal Recharge Rate

       

Confining Layer Hydraulic Conductivity

   

X

 

Confining Layer Thickness

 

X

   

Boundary Condition Type

 

X

X

 

Perpendicular Distance from Well to Boundary

 

X

   

Orientation of Boundary

 

X

X

 

Capture Zone Type3

 

X

   

No. of Patholines Used to Delineate Capture Zone

X

X

X

X

Simulation Time

X

 

X

X

Capture Zone Time

X

X

X

X

Rectangular Grid Parameter

     

X

No. Forward/Reverse Pathlines

X

 

X

X

Nodal Head Values

     

X

No. of Heterogeneous Aquifer Zones

     

X

Heterogeneous Aquifer Properties

     

X


1 From Blandford and Huyakorn, 1991.

2 Confined, unconfined or leaky confined.

3 Time-related, hybrid or steady-state.

GPTRAC. GPTRAC contains both semi-analytical and numerical options. The semi-analytical method is similar to MWCAP and RESSQC in that it assumes a uniform aquifer. Consideration of simple straight line fully penetrating boundaries are options in the routine. It has greater flexibility in that unconfined, semi-confined and confined aquifers can be dealt with directly. Discharging and recharge wells can be modeled and well interference is accounted for. Forward and reverse particle tracking is available. Areal recharge is an option for unconfined aquifers. Multiple straight line boundaries can be considered.

A word of caution is appropriate for using the semi-analytical option of GPTRAC for an unconfined aquifer. The user is prompted for the radius of influence of the pumping well. The resulting delineation is very sensitive to the value that is used for this variable. It is recommended that unless the user has specific information from either drawdown measurements or calculations regarding the zone of influence of the well, that this routine not be used for determining the WHPA for a well in an unconfined aquifer. The user should use MWCAP, if there are no interfering wells, or RESSQC instead.

The numerical option of GPTRAC can perform the tasks associated with the semi-analytical mode as well as allow for variations in aquifer characteristics, e.g., transmissivity, and changes in gradient magnitude and direction. Further, the program allows for anisotropic character in terms of the transmissivity (i.e., Tx not equal Ty). The program can be used as a post-processor for numerical programs such as MODFLOW in that GPTRAC will read a head file and calculate flow paths accordingly. The head file can be generated from field data or from published contour maps by interpolating heads on a grid and using the program HEDCON in the WHPA package to generate a file that can be utilized by GPTRAC's numerical option.

MONTEC. The MONTEC routine is similar to MWCAP with the exception that it is based on a stochastic approach, i.e., evaluates the impact of known variations in parameters (e.g., hydraulic conductivity permeability (K), gradient (i), pump rate (Q), etc.) on the delineated capture zone. Input to the program is similar to MWCAP; however, the modeler can enter a range of possible values for parameters of choice during the input. The resulting output consists of a series of capture zones that are presented along with the level of confidence that the actual capture zone is within the delineated area. The program is limited to a single pumping well in a confined or semi-confined aquifer.

Uniform Flow Equation - Though not part of the WHPA package, the Uniform Flow analytical model (Todd, 1980) is advocated in several of EPA's documents (EPA, 1987; 1993). The results of this technique can be fit to known hydraulic head distribution and therefore account for variable gradient direction (Bradbury et al., 1991). Values for "K", "b", "i", and "Q" (see Table 3-1) are substituted into appropriate equations yielding the down-gradient stagnation point and the width of the capture zone at any point "X" along the flow path that intersects the well.

Analytic Element Technique - A disadvantage of the analytical techniques described above is that they cannot be calibrated to existing conditions. As a result, uncertainty regarding the application of the model to the groundwater flow system remains. The analytic element method (Strack et al., 1994; Wuolo et al., 1995) overcomes some of this concern while maintaining the relative simplicity of the analytical model development.

The EPA has recently distributed the program CZAEM (Strack et al., 1994) which is designed to generate capture zones using the analytic element technique. The method generates a uniform flow field based on limited head and conductivity data. Using that flow field the program will predict head values within the area that can be checked against observations to see if the assumptions inherent in the model are appropriate. The program will also predict changes in head values as a function of model elements (e.g., a well) within the field. Comparing these to observed head changes provides for a means of calibration of the model. The EPA program WhAEM (Haitjema et al., 1994) links CZAEM with a Geographic Analytic Element Preprocessor (GAEP) to facilitate the analysis.

The program CAPZONE (Bair et al., 1991) also provides for calibration of an analytical technique by comparing computed and observed drawdown within a stressed aquifer. The program computes drawdown at rectangular grid points across the area of concern. These drawdowns are compared with observed head values and adjustments to the conceptual model are made if necessary. After calibration is complete, the drawdowns are superposed on the regional head distribution and the resulting hydraulic head map is used as the basis of groundwater flowpath determination.

Hydrogeologic Mapping - Hydrogeologic mapping is performed in conjunction with the application of the analytical technique. With the equations alone, there is no provision for determining the up-gradient boundary of the ZOC. Application of Darcy's law using the TOT value is commonly done to terminate the up-gradient position of the WHPA. There may be, however, geological reasons for terminating the ZOC at positions closer to the wellhead than the WHPA boundary, e.g., a lithologic boundary, groundwater divide or a stream boundary (Figure 3-5). Such features are recognized in the process of hydrogeologic mapping and are incorporated into the conceptual model.

The choice of analytical technique used in the delineation exercise is, of course, up to the modeler. In Table 3-1, we provide a table that lists the required input for the WHPA codes RESSQC, MWCAP, GPTRAC Semi-Analytical and GPTRAC numerical options. This table is presented for the purpose of illustrating extent of flexibility that exists within these analytical techniques and should not be construed as either endorsing or requiring these specific procedures.

Figure 3-5: Wellhead Protection Delineation Using Hydrogeologic Boundaries

Delineation Considerations - There are several decision points encountered by the modeler during the construction of the model. These are often critical in that they may result in very different delineations depending on the decision. It is of importance that the decision made be the one that most accurately reflects the aquifer and well in question. Listed below are several of these decision points with some directions indicated in order to resolve the questions that arise:

Confined Versus Unconfined Aquifer. In many analytical techniques, the modeler must specify directly or indirectly whether the aquifer of concern is confined or unconfined. Further, in developing management strategies appropriate for a given area, it is very useful to know the level of natural protection (i.e., confinement) that characterizes the aquifer (Kreitler and Senger, 1991).

For purposes of classification, this document will adopt the Water Resources Department's definition of "artesian aquifer" [OAR 690-200-050(8)] for that of a confined aquifer. Using that definition, a confined aquifer is one in which groundwater is under sufficient head to rise above the level at which it was first encountered. Clearly, this definition does not make the distinction between various levels of confinement, a characteristic that would have to be addressed in some of the analytical techniques.

An unconfined aquifer is defined as an aquifer in which the upper surface, the water table, is at atmospheric pressure. The systems will be expected to have accomplished an extensive evaluation of existing information in making the distinction of confined or unconfined conditions.

It is important that the distinction between confined and unconfined be supported by the conceptual model and not be based on a single well report. It is not uncommon for a driller to bore through a finer-grained portion of the aquifer and, because of low seepage rate, not recognize that the zone is saturated. The water-bearing zone identified on the driller's log may reflect only a higher permeability part of the aquifer. Because of this, it may appear that the static water level is higher than the water-bearing zone when in fact the entire section is saturated and in hydraulic connection.

Many of the analytical techniques provide for modeling semi-confined situations and such an approach should be used if indicated by the conceptual model.

Granular Versus Fractured Aquifer. The distinction between granular and fractured relates primarily to how the physical character of the aquifer impacts groundwater flow. The flow equations used in the analytical models assume that the aquifer can be modeled as a porous medium, is isotropic in its characteristics and that groundwater flow direction is in a direct and predictable relation to the hydraulic head distribution within the aquifer. This is often the case in a granular aquifer, e.g., unconsolidated and consolidated sedimentary materials. Rocks hosting closely spaced and intersecting fractures may in fact be modeled as granular, but may still be anisotropic in character and therefore not lend themselves to direct application of analytical techniques. Special techniques are available that will allow the use of analytical methods to these water systems (Fetter, 1981; EPA, 1994).

Analytical techniques, however, are not generally applicable to those aquifers that contain widely spaced or discrete fractures, or to rocks that have well developed preferred orientations (e.g., some metamorphic rocks), or in geologic terrains consisting of highly deformed interbedded units of differing conductivities. In these cases, groundwater may be constrained by such anisotropic features to flow in directions other than 90 to hydraulic head contours. As a result, the zone of contributions may be significantly different than that predicted by uniform flow equations, and travel times may be difficult to predict. As stated above, water systems will be required to supply supporting documentation with regard to their assignment of the aquifer of concern.

Aquifer Thickness. All models used to delineate WHPAs require input of the thickness of the aquifer. In the analytical models, this should be the average value, however a sensitivity analysis (see below) should be conducted to determine how much difference in the resulting WHPA would be observed by using the range of observed thicknesses.

The analytical methods are based on the assumption that the well is fully penetrating with respect to the aquifer, i.e., is open to the full thickness of the aquifer. For wells that are only partially screened in the aquifer it is recommended that the combined screened intervals or thickness of water bearing zones (whichever is less) be utilized as the aquifer thickness in the model development. The rationale for this is that groundwater flow to the well tends to be horizontal, owing to the fact that horizontal conductivity generally exceeds vertical conductivity by approximately an order of magnitude or more. A typical ratio of vertical to horizontal conductivity is 0.01.

In areas where the water-bearing zones are in volcanic rock, e.g., basalt, it is not uncommon for the well to be cased to the volcanic rock and then constructed as an open hole for the remainder of the well. Often, there is no information on the well report to indicate the nature of the water-bearing zones. Clearly it is inappropriate to use the entire length of open hole as the aquifer when experience tells us that the water is normally derived from interflow zones, consisting of the more porous and brecciated (i.e., broken up) flow tops.

Observation indicates that the interflow zones within basalt range from <5 percent to >25 percent of the total flow thickness. If there is no information on the well report to indicate the thickness of the water-bearing zones in an open-hole basalt well, a thickness equivalent to 10 percent of the open hole should be used for aquifer thickness or open interval.

Hydrogeologic Boundaries. An important step in adequately determining the ZOC of a pumping well is the recognition of the impacts to groundwater flow of hydrogeologic boundaries. These boundaries can generally be considered to be combinations of two limiting cases: constant head and no-flow boundaries:

1. Constant Head Boundary. The most common example of a constant head boundary is a stream (Figure 3-5). In the case of a perennial stream, largely supported by surface flow through a drainage basin, the water level in the stream may show only minor variations over short time intervals. If the ZOC of a proximal pumping well intersects the stream, a hydraulic connection may be established whereby a portion of the water being derived from the well originates within the stream. The net effect is to reduce the size of the ZOC (i.e., less water is being derived directly from the aquifer). The actual impact will depend on the extent the stream penetrates the aquifer and how well developed the clogging layer is.

Although activities within the watershed upstream from where the hydraulic connection is established could impact the water quality of the groundwater derived from the well, it is unlikely that the community will be able to exert control over the entire watershed as a means of protecting their drinking water supply. As an alternative, we recommend that the community include as part of their management strategy and contingency plan development, a procedure that will protect the aquifer should an event occur upstream that releases a contaminant to the stream. This may be in the form of assuring that the public water system is immediately notified of such a release and has sufficient storage so that the use of the well could be discontinued until the threat of drawing the contaminant into the aquifer is past.

In the WHPA 2.0 package, the programs MWCAP and GPTRAC will simulate a stream boundary directly. The program RESSQC is capable as well; however, the boundary must be simulated by the modeler through the use of image wells. Importantly, the MWCAP and GPTRAC programs assume that the stream is linear and fully penetrates the aquifer, a situation that is rarely encountered in real settings. Also, the programs do not consider the reduced flow to the aquifer from the stream as a result of the presence of the clogging layer, e.g., fines in the steam bed that reduce the permeability of the materials through which stream water must flow to reach the aquifer. Approximation of a partially penetrating stream and the presence of a clogging layer may be accomplished through variable placement of an image well using RESSQC, supplemented by reducing the recharge rate of that well.

2.  No-Flow Boundaries. These boundaries are those across which groundwater movement is either prevented or reduced to negligible over the period of observation. No-flow boundaries are generally associated with groundwater divides or permeability contrasts (Figure 3-5). A groundwater divide is analogous to a topographic divide that separates surface water drainage basins. A groundwater divide (Figure 3-1) is an elevation high on the water table or potentiometric surface; groundwater moves away from the divide in both directions. Groundwater cannot flow across the divide, consequently, the divide represents a no-flow boundary.

Three types of no-flow boundaries formed by permeability contrasts have been recognized in Oregon. No-flow boundaries resulting from geologic contacts are perhaps the most common. These occur when the aquifer is limited in its lateral extent, either by pinching out, i.e., thins to negligible thickness as in a sand lense, or being in contact with a less permeable rock type:

Modeling Hydrologic Boundaries. In all three of these cases, groundwater flow across the boundary is inhibited because of the difference in the permeability of the materials. Water will not flow across the bedrock-alluvium contact to supply the well. As in the stream boundary, MWCAP and GPTRAC will model the no-flow (barrier) boundaries directly, making the assumption that the boundary is linear and fully penetrating. Because no-flow boundaries do tend to penetrate the aquifer, these assumptions are not as critical in the no-flow case as they are in the constant head boundaries. RESSQC can simulate the barrier boundaries through the use of image wells. Leaking boundaries can also be accommodated through image well modeling (Bair et al., 1991).

Parameter Estimation and Sensitivity Analysis - The key to a quality delineation clearly is the use of data that is representative of the system being modeled. Two weaknesses of the analytical approach are: (1) the models generally assume uniform characteristics within the aquifer (see GPTRAC - Blandford and Huyakorn, 1991), and (2) it is not possible in general to calibrate and validate the model (see CZAEM - Strack et al., 1994) and CAPZONE (Bair et al., 1991). As a result, it is often not possible to test whether the assumptions made in the model development accurately reflect the system.

It is not our intent to detail how one estimates all the parameters utilized within the individual techniques or repeat recommendations made elsewhere (See Appendix A). Rather, we are requiring that a sensitivity analysis be performed to determine the relative importance of each parameter. This may provide information on where to expend limited resources to improve the quality of the data used and make the delineation as representative as possible.

Parameter Uncertainty. In this document, we are using sensitivity analysis to mean the provisional calculations of the WHPA using the range of uncertainty associated with each parameter. Specifically, if estimates or independent measurements of a hydraulic parameter indicate a range of possible values, an assessment of the impact of using one value or the other on the shape, extent or orientation of the WHPA should be evaluated. Unfortunately, as with any estimate, there will be a range of values for most of the parameters of interest. As such, there will be a large number of possible combinations of values that could be substituted into the equations.

As discussed above, the WHPA code MONTEC program is capable of incorporating these uncertainties into the WHPA delineation for single wells in a semi-confined or confined aquifer setting and providing a delineation that gives an estimate of the probability that the true delineation falls within the boundaries of the calculated WHPA. For other settings, or other programs within the WHPA code package, such an evaluation is not available.

As an example, you may have specific capacity data from an alluvial aquifer that indicates a significant range of hydraulic conductivity (K) values or you are uncertain regarding the value to use for the effective porosity of a basalt aquifer (Table 3-1). You should first try to limit the variability through careful analysis of the data in the context of the conceptual model to try to determine which value(s) within the range may be most applicable (EPA, 1994). You should then use the ranges of values in the appropriate models to determine the sensitivity of the area of the WHPA. If there is only small variations here, it may be sufficient to simply choose the most protective version of the WHPA.

If there are large differences in the areas of the calculated WHPAs, it may be more cost-effective to try to gather additional information (e.g., an aquifer test), rather than choosing, and then of course, managing the area within the larger WHPA.

Multiple Water-Bearing Zones. An additional example concerns a situation where a well is screened in two water-bearing zones with significant (i.e., an order of magnitude) differences in hydraulic conductivity. Using an average transmissivity value for the two units may yield a WHPA that does not represent the actual time-related capture zone.

As an alternative, the discharge from the well could be partitioned between the two water-bearing zones (zones 1 and 2) in the following manner (Golder Associates, 1994):

Q1 = w1 Q

where,

1

Q1 represents that part of the total discharge (Q) that can be attributed to water-bearing zone 1. K1 and K2 and b1 and b2 represent the hydraulic conductivities and thickness of water-bearing zones 1 and 2, respectively. The discharge derived from water-bearing zone 2 (Q2) is derived in an analogous manner.

Using the discharge data thus derived and li and transmissivities (Ti = li Ki) of some water-bearing zone (i), the WHPA associated with each zone may be determined using Qi and compared with that derived by assuming average values. In many cases it will be seen that the bulk of the water is derived from the high-K zone and, as a result, the WHPA associated with that water-bearing zone will be larger than that derived from average values. Note that in terms of transmissivities:

2

Modification of WHPA Orientation - The orientation of the WHPA in the analytical models is based on the input data concerning the direction of groundwater flow. This information is derived either from regional considerations or from the direct measurement of static water levels (SWLs) in appropriate wells at some particular time. It has been the observation of many hydrogeologists in the state that the gradient in a given area may change direction dramatically (up to 180?) on a seasonal basis, particularly for shallow unconfined systems. The change in gradient direction (and magnitude) may be the result of a change in recharge pattern or a change in the pattern and amount of withdrawals from the aquifer.

Incorporating Gradient Variations in the Model. Because of the uncertainty attached to the gradient direction, the state is requiring that the water system incorporate the potential variability in one of two ways:

1. Evaluation of the gradient quarterly for at least 12 months. In this step, wells within the aquifer of concern will be monitored for SWLs every 3 months for at least a year. If possible, wells that are screened in the same interval of the aquifer should be used in the monitoring process. The SWLs will be contoured and the direction of groundwater flow will be determined. WHPAs will be generated along each of the gradient directions determined. The area used for wellhead protection will be the entire area encompassed by the quarterly WHPAs.

2. Arbitrary rotation of the gradient direction. In the absence of quarterly SWL determinations, a community may elect to recalculate WHPAs along gradient directions that are rotated an arbitrary amount from the gradient direction determined in the conceptual model development.

The supplementary delineations in practice do not require a significant amount of program development. We anticipate that these extra delineations would only require changing the program input for the gradient and direction, no other modifications would be necessary (unless demanded by the conceptual model). Time investment at this step would be minimal.

As an added precaution regarding changing gradient directions, OHD is requiring that a circular area surrounding the wellhead, with a radius equivalent to the 6-month TOT using the CFR method, be included in the well's WHPA.

Numerical Methods

Numerical methods provide a high degree of accuracy and can be applied to almost all types of hydrogeologic situations. Accordingly, these methods are able to incorporate complex boundary conditions and variations in hydraulic properties within the aquifer. A large number of numerical models are presently available (van der Heijde and Beljin, 1988).

Developing the Model - The process involves dividing the area into cells or elements with each element located uniquely by the coordinates of its corners. Some numerical model codes also allow for vertical division the model into multiple layers. Each cell is characterized in terms of the aquifer's (or bounding material's) hydraulic characteristics, e.g., hydraulic conductivity, thickness, storage, recharge, etc.). Boundaries to groundwater flow can be simulated by specifying cells (elements) with specific boundary conditions (e.g., constant head) or permeability characteristics. The modeler is provided with a significant degree of flexibility in the model development as a result of her/his having the option to design the grid to fit the conceptual model, e.g., making the grid spacing smaller in areas where significant variations occur or where detail is desired.

After the model is designed, it is run (probably several times) until it converges on a solution that is within a specified error criterion. Typically, error criteria for heads and water balance are evaluated. Discrepancies may indicate that the error criterion is too small (e.g., heads) or the model does not adequately characterize the hydrogeologic system being modeled. The numerical model uses the hydraulic data to calculate head and flux values for each cell within the groundwater flow model.

Calibration - An important characteristic of numerical models is the ability to calibrate the model to fit observed data. In the calibration process, the modeler identifies specific calibration targets, usually known values of hydraulic head or fluxes into and out of the groundwater system. The modeler compares the head field and fluxes generated by the numerical model to the observed water levels and fluxes. The comparison is generally done statistically so that the degree of agreement between the modeled and observed conditions can be evaluated quantitatively. It is also useful to plot the areal distribution of the differences. When an acceptable comparison is achieved, the model is said to be calibrated.

Calibrated values for hydraulic conductivity, storage, boundary conditions, etc., are then systematically changed during a process called a sensitivity analysis. The sensitivity analysis is done to quantify the uncertainty in the calibrated model caused by the uncertainties in the estimates of aquifer parameters, boundaries and stresses (e.g., pumping) on the aquifer. A sensitivity analysis is usually done by changing one parameter value at a time and observing how much change in the model result occurs, i.e., how sensitive the model is to that parameter. The results of the sensitivity analysis should be reported as part of the model development summary.

The calibrated model can then be used to delineate the WHPA, through the use of particle tracking routine. The particle tracking routine calculates groundwater flow as a function of the head distribution developed by the numerical method. Capture zones as a function of time-of-travel can be delineated.

Validation - The numerical model can be further refined through the validation process. In this step, the impact of some perturbation to the model is explored, e.g., increased pumping from one or more wells, etc. The resulting new head field simulation can be compared to the actual situation when it occurs. Modifications to the input data or to boundary conditions can be made in order to reach agreement between predicted and observed head distributions.

Once a numerical model has been calibrated and validated, it becomes not only a tool for WHPA delineation, but also a planning tool for the community. Through the use of the model, the impact of new wells, infiltration lagoons, potential contaminant releases, etc., can be evaluated with much greater certainty.

Evaluation of Delineation Techniques

In this section, advantages and disadvantages of the above three techniques are described.

Calculated Fixed Radius:

Advantages - Simplicity, low cost, does not require significant amount of data acquisition.

Disadvantages - Generally not representative of the groundwater system, prone to legal challenges, tends to over protect downgradient and under protect upgradient, often yields larger area than other techniques.

Analytical Techniques/Hydrogeologic Mapping:

Advantages - Incorporates hydrogeologic characteristics of the aquifer, groundwater flow and hydrogeologic boundaries into the model, provides for a defensible delineation of the WHPA, is based on site-specific information. Often produces a WHPA that is smaller than the one produced using the calculated radius method.

Disadvantages - Assumes a uniform aquifer (note that some exceptions to this do exist), requires significant expertise and is moderately costly.

Numerical Models:

Advantages - 3-D modeling of groundwater flow, can account for evapotranspiration and recharge, groundwater-surface water interaction can be quantified, provides a more accurate delineation of the WHPA, accounts for variation in hydraulic parameters and boundary conditions, can be used for predicting the impact of natural and human-related activities on the flow field. Often produces a smaller area to manage than other techniques.

Disadvantages - Costly relative to other techniques, requires significant amount of data collection and high level of expertise to set up the grid.

Factors That May Influence the Delineated Area in the Future:

A WHPA is delineated based on a set of conditions with regard to hydraulic properties of the hydrogeologic units, fluxes into and out of the system and stresses applied to the aquifer. Although it may seem unlikely that the hydraulic properties of the aquifer could change appreciably, it is possible that surface-related activities could lead to a significant change from conditions assumed to exist in the initial conceptual model.

Prolonged drought and/or over-drafting of an unconfined aquifer could result in a decrease in the thickness of the saturated zone or variations in gradient. Both parameters could cause a change in the size and orientation of the WHPA.

Changes in recharge, either in amount or pattern of application may alter the dimensions of the WHPA. Recharge could be a function of natural precipitation or result from changes in irrigation or application of waste water in the area.

Any addition of high-volume wells in the WHPA or in proximity to its boundary may alter the shape of the WHPA. Significant changes in the pump rate will obviously have an impact on the size of the well's capture zone. Figure 3-6 portrays the change in shape of the original delineation (a) as a result of the addition of a single well immediately upgradient (b) and two wells in proximal positions (c). Details of the models are given in the figure caption.

After the WHPA has been delineated, the community should have as part of their management strategy a method by which the above influences can be recognized and evaluated.

Susceptibility Analysis

The delineations discussed above are based on water movement within the aquifer only. In fact, a contaminant released at or below the surface must travel or be transported across the unsaturated zone to the aquifer. Absent a direct communication to the aquifer, e.g., an improperly constructed well or a through-going fracture system, the characteristics of the unsaturated zone control the time and probability that the contaminant will reach the aquifer. Accordingly, it is recommended that a susceptibility analysis be conducted within the WHPA.

Use of the Analysis - The utility of the susceptibility analysis is that it indicates areas within the WHPA in which the aquifer is most susceptible to contamination, i.e., has the highest potential of being impacted by surface activities. When the susceptibility analysis is combined with the results of the potential contaminant inventory (Step 4), the highly vulnerable areas can be recognized. A high vulnerability would be indicated where a high-risk surface practice occurs in an area where the aquifer has a high susceptibility.

Several methods exist for determining aquifer susceptibility on a site-specific scale. The EPA has developed a technical document on "Managing Ground Water Contamination Sources in Wellhead Protection Areas" that incorporates both susceptibility and risk rankings into its determination. Not all potential sources of contaminants are considered; however, in some cases the risk triggers are somewhat high.

Oregon State University has developed a decision aid that is designed to help producers consider groundwater protection when using various agricultural chemicals in crop production. The Oregon Water Quality Decision Aid utilizes soil characteristics such as sorption capacity and permeability and chemical characteristics of the contaminant such as its sorption potential and persistence to arrive at an estimate of the aquifer's vulnerability to that chemical. Much of the data has been compiled by OSU and the decision-making process is very user-friendly. (More information can be obtained by calling OSU at 541-737-5713.) The OHD developed a susceptibility analysis in conjunction with a monitoring reduction program for public water systems (OHD, 1992). The determination of susceptibility makes use of information regarding the hydrogeologic characteristics of the area and the chemical characteristics of the contaminant of concern (Figure 3-7). Decision points are assisted through the use of matrices.

Hydrogeologic Characteristics - The OHD process begins by using available soil data from Soil Survey Reports. Within the area of a mapped soil type, the depth to the aquifer and weighted hydraulic conductivity would be estimated from well reports (or other available information). These two parameters would be compared in a matrix to yield a traverse potential score. The traverse potential score is then utilized in conjunction with the hydraulic surplus, the difference between water applied, as irrigation or precipitation, and the water lost through evapotranspiration and runoff, to yield an infiltration potential. The infiltration potential is an estimate of the probability that water will migrate from the surface downward to the aquifer.

Chemical Characteristics - Consideration of the chemical characteristics begins with using the tendency for the contaminant to sorb (attach itself) to organic matter along with the amount of organic matter in the soil (as reported in the soil survey). These parameters combine to yield the mobility potential, the probability that the chemical will move through the soil zone.

The mobility potential is linked to the infiltration potential (Figure 3-7) to derive the leach potential, an estimate of the probability that water migrates to the aquifer carrying the contaminant. Combining the leach potential with the persistence of the contaminant, i.e., how long the contaminant "survives" before processes in the subsurface cause it to break down. This latter step indicates whether or not the aquifer is susceptible in that area.

Example - An example of the application of this method is illustrated in Figure 3-8. The WHPA for a well has been delineated and the soil survey of the area consulted to determine the principle soil types. In this example, only two types dominate, soil A and soil B. A cross section has been drawn from A to A in the lower figure. Note that there are significant differences in soil thickness and depth to the aquifer. The question being addressed is how does the aquifer susceptibility beneath soil A differ from that below soil B? The contaminant of concern here is trichloroethylene (TCE).

The comparison is illustrated in the lower table of Figure 3-8. The Traverse Potentials beneath the two soils are quite different based on the different hydraulic conductivities, the potential of movement of water from the surface to the water table being lower beneath soil A. Hydraulic surpluses are considered the same for both soil types, but the lower traverse potential beneath soil A leads to a lower infiltration potential beneath that soil as well.

Figure 3-7: Aquifer Susceptibility AnalysisFigure 3-7: Aquifer Susceptibility Analysis

Source: OHD, 1992


Figure 3-8: Example of Using The Susceptibility Analysis inWellhead Protection Strategies

Potential Contaminant: TCE with Log Koc = 2.0

Characterics - Type

Soil A - Silt

Soil B - Silty Sand

Thickness (in.)

60

12

% Organic Matter

3

0.5

Organic Matter Score

1

10

K (gal/day/ft2)

0.04

45

Depth to Aquifer (ft)

<50

<50

Hydraulic Surplus (in.)

15

15

Traverse Potential

2

9

Infiltration Potential

3

8

Mobility Potential (TCE)

5

9

Leach Potential

3

9

The Mobility Potential for TCE through soil B is higher because of the lower organic matter content, which in this process is recorded by a higher organic matter score, in that soil. The combination of the higher infiltration and mobility potentials beneath soil B lead to a higher leach potential for TCE in that area. This coupled with the high persistence of TCE indicates that the susceptibility of the aquifer is greater under soil B than under soil A.

It is apparent from Figure 3-8 that the value of the infiltration potential provides an estimate of the general susceptibility of the aquifer while the leach potential, coupled with the persistence of the contaminant, will provide an estimate of the vulnerability of the aquifer to contamination from a specific chemical. It is recommended that data from soil surveys and well reports be used to determine the infiltration potential at individual wells selected to provide good coverage throughout the WHPA(s). These values can be contoured to indicate areas of high versus low infiltration potential. General management strategies, particularly those designed to address future land use, can be developed from this data base. Individual potential sources can be further evaluated using chemical specific data to determine the potential of the contaminant of concern migrating to groundwater. Specific management strategies, designed to minimize risks associated with a specific existing land use, can be developed from this information.

General Application - Within a wellhead protection area, communities will want to focus their limited resources on those areas where the risk of contamination is greatest. Determination of the infiltration potential (see Figure 3-7) across the WHPA will indicate those regions where rapid infiltration of water from the surface is most likely. A contaminant release in these areas would pose a greater risk than in areas where the infiltration potential is low.

OHD's guidance manual suggests selecting representative wells within the WHPA that penetrate the aquifer of concern and calculating the traverse potential for each based on well logs and soil survey reports. Ideally the distribution of wells will be sufficient to provide at least one value of the traverse potential per quarter-quarter section (i.e., 16 wells per square mile). Using precipitation data and/or irrigation data (see OHD's guidance), the hydraulic surplus could be calculated and combined with the traverse potential to yield the infiltration potential (Figure 3-7). If data is sufficient, the WHPA area could be contoured with respect to the infiltration potential to more readily identify those regions where the risk of infiltration of a contaminant release to the aquifer is greatest. Communities could use this information to prioritize the area for purpose of protective management.

Individual sites could be more specifically evaluated by considering further the chemical characteristics of the contaminant of concern. This would be done by combining the infiltration potential at the site with the leach potential of the specific chemical, i.e., the tendency for the chemical to attach itself to the organic matter present in the soil, and the chemicals persistence in the environment. OHD's guidance manual provides pertinent data for those chemicals that are routinely monitored for drinking water purposes. The procedure will result in an estimate of the susceptibility, high, moderate or low, of the aquifer for that specific chemical. A facility where the susceptibility is low will obviously need less oversight than a facility where the susceptibility is high.

Submitting the Delineation for OHD Certification - The final delineation should be submitted to the groundwater coordinator of OHD's Drinking Water Program. The report submitted to OHD should include a map showing the delineated area and a report that provides documentation of the model parameters used and justification for the assumptions made in the modeling effort. Well locations and delineated areas must also be submitted digitally, in DXF or ArcInfo GIS compatible format. For those delineations involving the development of a conceptual model, the report should include well logs with locations keyed to a map, cross-sections (fence diagram), aquifer test results and a discussion of the hydrologic units, boundaries, groundwater flow direction and gradient, etc. OHD will expect that the issues, concerns and required tasks and decision points discussed in this chapter will be addressed in the report.

It is recommended that a line of communication between the local wellhead protection committee, or their consultant, and OHD's groundwater coordinator (503-731-4010) be established early on in the delineation process. This provides OHD the opportunity to provide technical assistance and will facilitate the submission and review process.

OHD will make every effort to review the delineation report in a prompt and timely manner. In the event that the report cannot be reviewed within 60 days, OHD will provide the submitter in writing with an estimate of the date in which the review will be completed.

The final delineation should be submitted to:

Groundwater Coordinator

Drinking Water Program

Oregon Health Division

800 NE Oregon Street

Portland, OR 97232

Delineation References:

Bair, E. S.; Springer, A. E.; and Roadcap, G. S., 1991. Delineation of Travel Time-Related Capture Areas of Wells Using Analytical Flow Models and Particle-Tracking Analysis. Ground Water, 29:387-397.

Blandford, T. N.; and Huyakorn; P. S., WHPA: A Modular Semi-Analytical Model for The Delineation of Wellhead Protection Areas, Version 2.0. U.S. Environmental Protection Agency, Contract Number 68-08-0003. [Distributed by the International Ground Water Modeling Center.]

Bradbury, K. R.; Muldoon, M. A.; Zaporozec, A. and Levy, J., 1991. Delineation of Wellhead Protection Areas in Fractured Rocks. U.S. Environmental Protection Agency, EPA 570/9-91-009.

Domenico, P. A. and Schwarz, F. W., 1990. Physical and Chemical Hydrogeology, John Wiley & Sons, New York, 824p.

Golder Associates, 1994. Wellhead Protection Area Delineation Report, Project No. WHPA-2, Prepared for Springfield Utility Board and Rainbow Water District, Springfield, Oregon. Golder Associates, Inc., Redmond, Washington.

Haitjema, H. M.; Wittman, J.; Kelson, V. and Bauch, N., 1994. WhAEM: Program Documentation for The Wellhead Analytical Model. U.S. Environmental Protection Agency, EPA/600/R-94/210.

Fetter, Jr., C. W., 1981. Determination of the Direction of Groundwater Flow. Ground Water Monitoring Review, 1:28-31.

Kreitler, C. W. and Senger, R. K., 1991. Wellhead Protection Strategies for Confined-Aquifer Settings. U.S. Environmental Protection Agency, EPA/570/9-91-008.

Long, J.C.S.; Remer, J.S.; Wilson, C.R. and Witherspoon, P.A., 1982. Porous Media Equivalents for Networks of Discontinuous Fractures. Water Resources Reserved Research, V. 18, pp. 645-658.

Oregon Health Division, 1992.

Guidance Document for Phase II/V Use and Susceptibility Waiver Applications. Drinking Water Program, 92 p.

Strack, O.D.L.; Anderson, E.I.; Bakker, M.; Olsen, W.C.; Panda, J.C.; Pennings, R.W. and Steward, D.R., 1994. CZAEM User's Guide: Modeling Capture Zones of Ground-Water Wells Using Analytic Elements. U.S. Environmental Protection Agency, EPA/600/R-74/174.

Todd, D. K., 1980. Ground Water Hydrology. John Wiley and Sons, Inc., New York.

USEPA, 1987. Guidelines for Delineation of Wellhead Protection Areas. U.S. Environmental Protection Agency, EPA-440/6-87-010.

USEPA, 1993. Wellhead Protection: A Guide for Small Communities. U.S. Environmental Protection Agency Seminar Publication, EPA/625/R-93/002.

USEPA, 1994. Ground Water and Wellhead Protection. U.S. Environmental Protection Agency Handbook, EPA625/R-94/001.

van der Heijde, P. and Beljin, M.S., 1988. Model Assessment for Delineating Wellhead Protection Areas. U.S. Environmental Protection Agency, EPA440/6-88-002.

Wuolo, R.W.; Dahlstrom, D.J. and Fairbrother, M.D., 1995. Wellhead Protection Area Delineation Using The Analytic Element Method of Ground-Water Modeling. Ground Water, 33:71-83.

Return to Top of Step 3


Return to: Chapter 3 Steps

Return to: TABLE OF CONTENTS, Water Quality Home Page, DEQ Home Page