United States
       Environmental Protection
       Agency	
Office of Water
(4606)
EPA816-R-97-008
August 1997
&EPA  STATE METHODS FOR
       DELINEATING SOURCE
       WATER PROTECTION
       AREAS FOR SURFACE
       WATER SUPPLIED
       SOURCES OF
       DRINKING WATER

-------

-------
STATE METHODS FOR DELINEATING
SOURCE WATER PROTECTION AREAS
  FOR SURFACE WATER SUPPLIED
  SOURCES OF DRINKING WATER

-------

-------
                         ACKNOWLEDGMENTS
      The authors wish to thank the Federal, State and local officials who contributed,
reviewed and commented on the case studies used in this document. This document would not
have been possible without their assistance.

-------

-------
                                  CONTENTS

                                                                              Page

 ACKNOWLEDGMENTS

 1.0  Introduction and Background	        1

       1.1   Rationale for Source Water Protection	1

       1.2   Summary of TAD Contents	2

 2.0  Source Water Protection Area and Segment Delineation Methods  	3

       2.1   Topographic Boundaries	3

       2.2   Setbacks/Buffer Zones	6

       2.3   Time-of-Travel Calculation	10

       2.4   Modeling to Enhance the Efficacy of Delineating Source Water Protection
             Areas and Segments  	13

Appendix 1
       Examples Of State and Local Source Water Protection Area and Segment Delineation
       Methods

Appendix 2
       Annotated Bibliography of the Technical Literature Addressing Source Water
       Protection Area and Segment Delineation

Appendix 3
       Literature Cited

-------

-------
                       1.0   Introduction and Background

        A review of Federal-, State- and local-government methods to delineate surface Source
 Water Protection areas (SWPAs) in watersheds, or basins, indicates that primarily three
 methods have been used. These methods are:

              Topographic boundary delineation,
        •     Setback/buffer zone delineation, and
        •     Time-of-travel calculation.

 Modeling techniques have been used to enhance these three methods.  A summary of these
 methods is presented in this report.

        In August 1997, the U.S. Environmental Protection Agency (EPA) stated that for
 public water systems (PWSs) relying on surface water, the SWPA will "include the entire
 watershed area upstream of the PWS's intake structure...up to the boundary of the state
 borders."  The method to be used to delineate the SWPA will be the topographic boundary
 delineation method.  The EPA recognizes that the susceptibility of a PWS to any given
 contamination source will depend, to a significant degree,  on the location of that contamination
 source in the watershed area;  that is, in some locations, a source would be a significant threat
 to a PWS while in other locations, particularly within a large watershed,  the source would be
 less significant or even non-significant. Therefore, EPA encourages the  segmenting of
 SWPAs into smaller subunits  for the purposes of source identification and susceptibility
 determinations. Many of the methods/approaches described in this document can be used to
 delineate these subunits of a surface water SWPA.

 1.1   Rationale for Source Water Protection

       The Safe Drinking Water  Act Amendments of 1996 (P.L. 104-182) and many State
drinking water laws include provisions for drinking water protection in SWPAs.  These areas
are the source areas of surface water or ground water that supplies drinking water systems.
Drinking water protection activities fostered under the Safe Drinking Water Act or other
statutes, such as the Clean Water Act,  range  from voluntary efforts to mandatory controls.
As envisioned under the 1996 Amendments, Federal-, State- and local-government and

-------
private-sector activities would help prevent pollution of PWSs from a wide variety of
contaminant sources. Source Water Protection programs can potentially reduce the long-term
costs associated with PWS monitoring and treatment. In addition, these programs can provide
an added degree of public safety for those occasions when human or mechanical errors
temporarily interrupt water treatment.

       Source Water Protection programs for surface water based systems are implemented
through steps similar to those in Wellhead Protection programs for ground water based
systems. That is, both programs have provisions for:  (1) delineating the areas to be
protected; (2) identifying potential contaminant sources; (3) implementing management
measures for those sources; and (4) planning for emergency water supply. Many communities
begin the Source Water Protection process by developing a local program planning and
implementation team.

       In order to protect drinking water supplies, Federal, State and local managers need to
understand the basis for identifying the geographic areas to be protected. Drinking water
programs should delineate surface SWPAs and are encouraged to delineate SWPA segments in
order to facilitate the determinations of susceptibility and identification of the contaminant
sources that may impact drinking water intakes.

1.2   Summary of TAD Contents

       This document is organized into two chapters and three appendices. Following this
Introduction and Background, Chapter 2 describes methods that have been used to delineate
SWPAs (and SWPA segments), and includes State and  local case studies of the methods.
Appendix 1 provides additional examples of the use of delineation methods.  Appendix 2  is an
annotated bibliography of the technical literature describing source water delineation and  some
management methods related to delineation. Appendix 3 is a list of the literature cited in this
document.

-------
                   2.0   Source Water Protection Area and
                            Segment Delineation Methods
        Local, State and Federal agencies have predominantly used three methods to delineate
 Source Water Protection areas (SWPAs) for surface water supplied PWSs:


              Topographic boundary delineation,
        •      Setbacks/buffer zone delineation, and
        •      Time-of-travel calculation.

 Although States are now required to use the topographic boundary delineation method to
 delineate SWPAs, setback/buffer zone delineation and time-of-travel calculations are important
 methods for the delineation of SWPA segments, which can facilitate differential protection
 management.  The delineation methods and example applications are described below.

 2.1   Topographic Boundaries

       Topographic boundaries are, irrespective of scale, defined by the elevation of the land.
 A topographic boundary of a watershed (Figure 1) is the perimeter of the catchment area of a
 stream.  Analogously, a topographic boundary of a  subwatershed is the perimeter of the
 catchment area of a tributary of a stream. The distinction between a watershed and a
 subwatershed is purely one of nomenclature. That is, the catchment area of a tributary is both
 the watershed of the tributary and a subwatershed of the main stream.  Thus, the occurrence of
 one watershed (subwatershed) within another may be thought of as nested watersheds. (Note
 however, that the catchment area of any. stream that drains directly to an ocean is always
 considered a watershed, because, by definition, the  stream is not a tributary of another
 stream.) The topographic boundary of the area contributing surface water to a PWS is the
 perimeter of the catchment area that is upslope of the PWS intake, that is, the watershed area
 (Figure 2).

       A key initial step in Source Water Protection for surface water supplied PWSs is the
 delineation of the watershed area contributing water to the drinking water intake.  This area is
 composed of the land and the surface water (i.e., lake, reservoir, tributaries and streams)
uphill of the drinking water intake.

Method Description

       A watershed area is easily delineated on a topographic map by the drawing of a line
connecting the highest points uphill of the intake, from which overland flow drains to the
intake.

-------
Figure 1. Topographic Boundary of a Watershed

-------
   Figure 2.  Topographic Boundary of a Watershed Area
Legend
5)5- Drinking Water Intake

-------
       Example: The Topographic Boundary Method in Idaho

       Idaho's watersheds have been delineated by the topographic boundary method.  The
State used the U.S. Geological Survey's topographic hydrologic units for Idaho's
basin/watershed/subwatershed boundaries (Seaber, 1987). In its delineation approach, Idaho
uses a three-tiered strategy.  Idaho's first level of topographic delineation includes each of the
State's major river systems,  which are represented by a unique drainage basin (six basins
total). Each basin includes the area of land drained by the major river system. Within these
basins, Idaho delineates watersheds, that is, the area of land drained by a stream or a system
of streams, or a geographic area hi which water, sediment and dissolved materials drain to a
common outlet.  Finally, the State may delineate subwatersheds, which are smaller geographic
management areas delineated to address site-specific concerns.

2.2   Setbacks/Buffer Zones

       Surface water setbacks and buffer zones are often used as a means of reducing the
adverse impacts of runoff on drinking water sources. The primary purposes of
setbacks/buffers are to filter overland flow and, to a lesser extent, slow overland flow and
encourage increased ground water infiltration. Buffer zones ("green areas") may be intended
to serve several functions such as: wildlife habitat, residential or commercial exclusion, or
Source Water Protection.

Method Description

       Determination of the width of buffer zones is often based on consideration of such
factors as:  topography of the land, local land uses, political and legal feasibility of setting
aside such buffers, slope, size of the stream and land ownership rights. A typical
buffer/setback zone (Figure  3) for protecting the water withdrawn by an intake hi  a stream,  is
a strip of vegetated land generally 50 to 200 feet wide, upstream of the PWS intake, along the
shores of the stream. Analogously, a setback/buffer zone can be delineated to protect the water
withdrawn from a reservoir  (Figure 4).

       Setback/buffer zones filter out some portion of sediment-borne contaminants.  In
addition, by slowing down overland-flow velocity, these green areas briefly increase the
exposure of overland flow to such processes as photolysis  and encourage an increase hi
infiltration to the ground water reservoir, where travel tunes are longer (but, where
contaminant cleanup is more difficult).   Forested buffer zones may be effective in reducing
nutrients.

-------
      Figure 3. Watershed with a Buffer/Setback Zone
          for a Drinking Water Intake in a Stream
Legend

9|£ Drinking Water Intake

-------
  Figure 4. Watershed with a Buffer/Setback Zone
     for a Drinking Water Intake in a Reservoir
Legend

$j£ Drinking Water Intake

-------
        The sediment-load carrying capacity of moving water is related to its velocity.  When
 overland flow reaches the edge of a buffer zone, the velocity of the water is reduced and
 sediment is dropped.  Thus, with time, a berm may form along the edge of a buffer zone. A
 berm will cause future contaminant-laden overland flow to travel parallel to the stream, until a
 topographically low area is encountered.  There, the overland flow will cross the buffer zone
 and enter the stream; transported with the overland flow will be its load of contaminants.
 Additionally, if a buffer zone is overtopped during a major precipitation event, any portion of
 the load that flows above the vegetation will not be removed by filtration. (Contaminants may
 also be transported to surface water through the vadose zone or by ground water discharge
 from the saturated zone.)

        Example:     The Setback/Buffer Zone Method in Georgia

        As part of its Rules for Environmental Planning (Georgia State Statutes for Planning,
 1991) Criteria (Georgia State Statutes for Criteria, 1991), Georgia requires buffers and
 setbacks:

        1)     In watersheds greater than 100 square miles, Georgia requires 100-foot
              buffer zones and 150-foot setbacks within a 7 mile radius of all water
              supply reservoirs upstream of all governmentally owned public drinking
              water intakes; the buffer zone around the water-supply reservoir is
              required to be 150 feet (measured from the normal pool); and

       2)     In water-supply watersheds of less than 100 square miles, in addition to
              the buffers/setbacks in (1) above,  Georgia requires 50-foot buffers and
              75-foot setbacks at distances greater than 7 miles from the water-supply
              intake or reservoir.

No buffers or setbacks are required in water-supply watersheds greater than 100 square miles
that lack a water-supply reservoir.

       Georgia's planning criteria specify that the buffers be naturally or enhanced vegetated
tracts of land with no,  or limited minor, land disturbances.  The State precludes the
development of impervious surfaces and the installation of septic systems within setbacks.
State constraints also limit to 25 percent, the amount of impervious surface in a small
watershed (less than 100 square miles) upstream of a PWS intake.

       Georgia determined the size of its buffer zones through the work of an Advisory Group
to the Governor, which was comprised of 116 representatives from numerous organizations
and interested parties, including the State, local governments, citizens' groups, environmental
organizations, industry, and the general public. This group held twenty public briefings
throughout the State to obtain input and comment.

-------
2.3    Time-of-Travel Calculation

       In addition to the two methods above, drinking water utilities may delineate stream
reaches to facilitate spill- and other emergency-response activities for the protection of source
waters. (In this method, the protection "area" is actually a stream reach, rather than an area.)
The following section describes the use of time-of-travel (TOT) calculations for emergency
planning for Source Water Protection.

Method Description

       This delineation method is based on the time it takes for a contaminant, moving at the
same velocity as a stream, to travel from an upstream monitoring point to the point of interest,
the drinking water intake (Figure 5). In this method, the TOT is calculated between these two
points. It is this travel tune that provides the opportunity for managers to respond to a
contamination event. Use of this method would be of greatest importance for drinking water
utilities tapping streams, or reservoirs on streams, designated  for commercial transport or
other industrial use. Water quality flow models provide a means through which specific
hydrologic, geographic and water quality parameters can be used to estimate the travel time
for a contamination incident such as a spill into a river to reach a drinking water intake, and to
estimate the level of contamination once it is at the intake.

       The TOT method is often used to alert a downstream drinking water supplier that a
contaminant spill has occurred and provides the managers with lead time to close downstream
intakes.  Because stream TOTs are short (generally in the range  of hours or days), the in-situ
remediation that occurs as a stream flows towards an intake may be limited. However, some
remediation does occur; some very volatile contaminants may undergo significant reduction in
concentration as a result of volatilization, the concentration of contaminants from point sources
is often reduced as a result of dilution mixing ha the stream and some processes, such as
photolysis, may reduce contaminant concentrations. In addition, along their journey, some
portion of waterborne contaminants may adhere to clay or other  particles and settle out of the
stream onto the stream bed.

       Although the examples below are part of large scale monitoring and contingency
planning programs, the principles involved can apply equally to  smaller systems.  Specifically,
the TOT estimation techniques (described in 40 CFR 112, Appendix C-III) (U.S. Government
Printing Office, 1995) and the algorithms used in the model described below in the
ORSANCO example, and other water quality models, can be used to determine the TOT from
an upgradient spill site to a PWS intake.  A 1993 pipeline leak to Sugarland Run in Fairfax
County, Virginia, resulted in fuel oil in that stream and in the Potomac River, to which the
stream is tributary.  It took 2 hours to shut down the raw-water supply from the Potomac
River. No oil entered the water system.
                                          10

-------
Figure 5.  Watershed Showing Points at Three Different Stream
      Times-of-Travel (TOT) to a Drinking Water Intake
                                     t3 (e.g., 3 hour TOT)
  Legend

  •3|€- Drinking Water Intake
                            11

-------
       Example:    Ohio River Valley Water and Sanitation Commission (ORSANCO)

       The ORSANCO is an interstate commission established in 1948.  The Commission
coordinates monitoring for specific organic contaminants in the Ohio, Allegheny,
Monongahela and Kanawha Rivers, in cooperation with local water utilities.  The Commission
notifies downstream water utilities of detections so that the utilities can implement protection
measures. The ORSANCO's monitoring locations encompass six States  from Pennsylvania to
Illinois, and include 14 organics sites at industrial plants and drinking water suppliers on the
rivers.

       Personnel at water treatment facilities collect samples directly from raw river water
feed lines, for analysis by gas chromatograph. If contamination above the predetermined action
level for a specific chemical is found, a protocol is activated for increased sampling and
notification of downstream water suppliers, that is,  the individual facility operators, who
decide what action to take. Actions could include shutting down the intake or finding an
alternative water supply.

       The ORSANCO uses a flow model to estimate contaminant concentrations  and travel
times when a notification is issued. The model utilizes river flow data that are provided daily
by the Army Corps of Engineers, Ohio River Division.  Other variables which are input to the
model include: spill location, spill duration, spill amount, initial tune of the spill, and a decay
coefficient.  For a rough estimate of travel time on a daily basis, the river stage is used to
calculate velocity.  Potential time of contaminant arrival is reported to downstream utilities.

       Each of the drinking water utilities has an emergency response plan in case of a
contamination incident.  The City of Pittsburgh, for example, has a spill response plan that
includes provisions for coordinating with a variety of City and State departments,  such as the
Allegheny County Health Department, the State Fish and Game Department, the Coast Guard,
and the State Department of Health.  The City does not have an alternate water supply and,
depending on the potential duration of the spill, the City either shuts down the  water intake
(for shorter spills hi the range of 8 to 12 hours) or treats contaminated water that has entered
the system.

       Example:     The Lower Mississippi River Industrial Corridor:  Baton Rouge to New
                     Orleans, Louisiana

       The Early Warning Organic Compound Detection System (EWOCDS) was established
in 1986 to provide both municipal and industrial water suppliers with early warnings of
contaminant spills on the lower Mississippi River.  Through this monitoring and warning
system, suppliers are able to assess the travel times for a contaminant and take appropriate
measures to avoid an intake of contaminated water.
                                           12

-------
        The EWOCDS is a cooperative program among the Department of Environmental
 Quality (DEQ), five industrial water users and three municipal users.  The State provides the
 monitoring equipment, computers and modems used to transmit, store and analyze water
 quality data. The water users provide volunteer staff time for sampling, and some provide a
 building for the monitoring equipment.
                                                   Figured.  EWOCDS
                                                              EWOCDS
                                             J
                                             Baton Rouge
       Monitoring stations are
 located at each of eight water intakes
 (St. Bernard is no longer a
 EWOCDS site) along the
 approximately 100-mile river
 corridor (Figure 6) (Wagenecht,
 1991). Gas chromatographs are
 used to test for 20 volatile organic
 chemicals identified as priority
 pollutants. Samples are taken twice
 daily at all monitoring stations
 (Wagenecht, 1997).

       Based on river velocity data
 (related to river height) provided
 weekly by the Army Corps of
 Engineers, the water providers and
 DEQ calculate the time of travel for
 the leading edge, peak,  and trailing
 edge of a spill.  The DEQ developed TOT models which were used to predict the probability
 of a spill remaining undetected at the monitoring stations and contaminating downstream water
 intakes. These studies led DEQ, in 1989, to require increased sampling frequency from once
 to twice daily.  The model results have been compared to actual travel tunes for several spills,
 and were found to accurately predict the peak travel tunes over the 100 mile corridor.  In
general, the information supplied by the Army Corps of Engineers is reasonably accurate and
when used in conjunction with twice daily sampling, generally provides sufficient time for
early warnings to be issued (Wagenecht, 1997).
Key

O EWOCDS sites
• Water Works
                                                                New Orleans
                                                                N.O.-C«r. W.W.
                                                                        Pa«.B.rn.«l
2.4
      Modeling to Enhance the Efficacy of Delineating Source Water
      Protection Areas and Segments
       Ground water discharge and surface runoff models can be used to assess the potential
impact of individual contaminant sources, and to identify SWPA segments with the greatest
potential impact on source water quality. Modeling can be used in conjunction with SWPA and
segment delineation techniques to enhance source water quality protection efforts.
                                         13

-------
      Figure 7.  Land Use Parcels
Method Description

       A variety of models has been
developed to assess the impact of
changing land use on surface water
quality. Simpler models require less
detailed, site specific hydrologic
information and provide more
generalized and descriptive output.
More complex models require more
extensive input data and provide output
with greater predictive capability and
site specificity. Site specific output can
provide locations of contamination
sources and yield relatively accurate
predictions of variable flows and water
quality at any point in a watershed.

       Contaminant source loading
models estimate chemical loading rates
to surface water.  These methods are
most useful for estimating variation in
loading rates as a function of changing
land uses within the watershed.  For
example, as shown in Figure 7, land
may be divided into residential-,
industrial-, and agricultural-use parcels.
If agricultural land is subdivided by soil
type, crop type, and land management practice, the nonpoint source loading rates for runoff,
sediment yield, and ground water discharge may be estimated for each parcel type.  These
parcel estimates are summed to obtain the total loading rate for the watershed or watershed
areas (Haith and Shoemaker, 1987).

       Several  States, local governments, water suppliers, and watershed management
authorities have begun modeling to identify those land uses that have the greatest potential
impact on source water quality.  Modeling can also be used to assess the impact of differing
land management strategies within a SWPA to foster more effective Source Water Protection.

       Common uses for models include: evaluation of urban and agricultural runoff
scenarios, determination of the impacts of changes in agricultural practices,  and assessment of
the impacts of various point and nonpoint source releases.
                         Drinking Wgjeflntake
14

-------
        Example: Minnesota

        The Agricultural Nonpoint Source Pollution Model (AGNPS) was designed to evaluate
 water quality impacts of land use and land management strategies in agricultural watersheds.
 The primary purposes of the AGNPS model are to prioritize watersheds in terms of water
 quality problems, to identify critical areas within a watershed that are large contributors to
 nonpoint source pollution, and to evaluate and predict the effects of alternative management
 practices.

        Because the model simulates sediment and nutrient transport in steps, water quality
 impacts can be assessed at any point in the watershed. The AGNPS's principal variables are
 hydrology, erosion, sediment, and chemical transport. However, the model also incorporates
 point sources of sediment from gullies and inputs of water, sediment, nutrients, and chemical
 oxygen demand (COD) from such sources as animal feedlots.

        The model operates on a grid cell basis, allowing the user to determine sediment and
 nutrient transport rates for each cell.  Existing soil, topographic, and land use conditions are
 input for  each cell. Based on this data, the model provides erosion, sediment transport,
 nutrient transport, and flow information. In addition to the standard output, the model has the
 capability to perform more specific analyses for any individual cell. These analyses include:
 1) calculation of drainage area, runoff volume and peak runoff rate; 2) estimates of adsorbed
 and soluble nutrients in mass per unit area, and nutrient and COD concentration in runoff; and
 3) for each of five particle size classes, estimates  of upland erosion, sediment yield, percent of
 yield from within a cell and from outside the cell, and cell deposition.

       Currently, the model is used to estimate the results of specific hydrologic events.
 However, with some modification, the model could provide a continuous representation of the
 average annual response of a watershed to a representative annual distribution of rainfall
 events. The model can also be used to evaluate and predict site-specific sedimentation in a
 watershed.

       Minnesota has used the AGNPS model to evaluate water quality hi several watersheds.
 In the Garvin Brook watershed, the model was used to determine sediment and nutrient
 loading in response to a 25-year frequency, 24-hour duration storm. The cell structure of the
 model was used to identify feedlots with high potential pollutant contributions and individual
 cells with excessively high upland erosion rates.  These "critical areas" were then targeted for
 the application of control measures.  The model also identified one subwatershed that
 contributed high sediment loads but that did not contain any high sediment yielding cells.  This
 subwatershed was targeted for a different set of management strategies designed to reduce
 sediment yield over the entire area.  Information from the model was used to encourage
farmers in this subwatershed to sign up for an agricultural cost sharing program.
                                           15

-------

-------
                                    Appendix 1

                Examples Of State and Local Source Water
            Protection Area and Segment Delineation Methods

       This section provides an overview of the delineation methods used in several programs
 that have implemented Source Water Protection activities for surface water supplies.

       The descriptions below illustrate the manner in which Source Water Protection may be
 incorporated into existing drinking water and surface water protection programs.

 Topographic Boundary Delineation Method

 Massachusetts Source Water Protection Delineation Method

       Major delineation efforts in Massachusetts have been made by the State to map
 watersheds for use by the Watershed Management Program, and by the Massachusetts Water
 Resources Authority (MWRA) to delineate the Metropolitan District Commission's (MDC's)
 watersheds at the Quabbin and Wachusett Reservoirs, and at the Ware River. The
 Massachusetts Department of Environmental Protection (MDEP) has mapped the major river
 basins of the State in a Geographic Information System (GIS), which also contains wellhead
 protection area information, municipal boundaries, land uses, and other pertinent planning
 data. The main impetus for this mapping effort was passage of the 1986 Water Management
 Act, which required identification of outstanding resource waters.  The Drinking Water Supply
 Division led the effort, with assistance from Watershed Management.  The basis of the
 delineation of surface water watersheds was the USGS topographic hydrologic units for
 watersheds and subwatersheds. Professional judgment was often necessary when interpreting
 subwatershed data, because tributaries may be split by the presence of gaging stations, and
 small tributaries may be combined.  In general, delineation was based on determining a set of
 upstream points of high elevation  from which surface water drains.  Choosing these points was
 subject to technical judgement.

       The MWRA, which is responsible for treatment and distribution of waters from MDC
 sources, delineated the MDC watersheds. The sole purpose for delineating the watersheds was
 drinking  water source protection.  The MWRA delineated three surface water zones,  mapping
 them in a GIS:  (1) Zone A, a 400 foot buffer around the water supply; (2) Zone B, a half-
 mile buffer beyond Zone A; and (3) Zone C, the remainder of the watershed including
 protection zones.  Watershed boundaries were based on surface water divides,  and ground
 water recharge areas tributary to each supply were included in the ground water zones.

       Once the watersheds were delineated and the data were input into the GIS, MWRA
conducted a threat assessment to determine the level of threat to Zones A and B from various
human-induced activities.  The MWRA rated and prioritized risks to their watersheds using the

-------
potential quantity of pollutants.  After consideration of these criteria, MWRA was better able
to isolate those watersheds that might be most vulnerable to certain types of threats (Chernin,
1992, Willmer, 1992).

       In addition to MWRA GIS mapping efforts, the MDC also maps its watersheds to
comply with the Watershed Protection Act. The MDC used USGS hydrologic units at
1:24,000 scale to map tributaries, and used FEMA maps for floodplain delineation.  The
MDC, in consultation with the MDEP, may adopt more accurate maps.

North Carolina Source Water Protection Delineation Method

       The topographic boundaries of North Carolina's water supply watersheds are delineated
on USGS 1:24,000 scale topographic maps. The location of a surface water intake is the
lowest point on a watershed boundary.  The remainder of the boundary encompasses the land
area draining to the intake as defined by the topography of the land (North Carolina
Administrative Code, 1994; North Carolina Department of Health, 1995; North Carolina
Division of Environmental Management, 1995).

       Within each watershed, the State defines critical areas in which certain land use
restrictions or management practices are applied.  The minimum critical area is defined as:
1) extending to either one-half mile from the normal pool elevation of a reservoir containing a
drinking water intake or to the ridge line (topographic boundary) of the watershed (whichever
is closer to the reservoir); or 2) extending to one-half mile upstream of and draining to a river
intake or extending to the ridge line of the watershed (whichever is closer to the drinking
water intake).  Local governments may extend this area as needed. Analogously, the State has
designated protected areas for certain WS-IV  class watersheds (North Carolina Administrative
Code, 1994; and North Carolina Department  of Health, 1995).

       In classifying the 208 surface water sources in the State, the Division of Environmental
Management (DEM) worked with local governments to determine the location of all surface
water intakes.  The DEM used this information,  coupled with information on existing land use
within the watersheds, and the types and locations of wastewater discharges, to determine
appropriate classifications for each of the water supply watersheds. North Carolina used the
topographic method to delineate water supply watersheds to protect sources of drinking water.

       The State defines five Water supply classes according to existing land uses and the
amount and types of permitted point source discharges (WS-I through WS-V).  State Rules
require that all local governments having land use jurisdiction in watersheds that are classified
as water supply watersheds, implement and enforce nonpoint source management strategies
related to urban development. These strategies must meet minimum standards adopted by the
State and typically include the development of water supply watershed protection ordinances,
maps  and a management plan.  The State then limits the point source dischargers that can
locate within the watershed (North Carolina Division of Environmental Management, 1995).
                                         Al-2

-------
Washington Water Quality Management Areas

       The State of Washington has established twenty three Water Quality Management
Areas (WQMAs).  For planning purposes, WQMAs are considered to be the State's major
watersheds (Figure 8) (Washington Department of Ecology, 1996).  The WQMAs are
generally large drainage areas that include several river basins.  The boundaries of the
WQMAs are partly dependent on the location of smaller Water Resource Inventory areas
(WRIAs), which were delineated in the 1970s as Sewage Drainage Basins for water supply and
sewer system planning. Typically, each WQMA contains several WRIAs.
            Figure 8. Washington State Water Quality Management Areas
                    Water  Quality Management Areas
                                                                           Pend Orellle
                         South
                       Puget Sound-  ( Upper Yaklma
                                                                 WalCT QuaJity Manajj.-incnt.
                                                                 Ana houndiry
                                                                 Walcr Resource Invoniry
                                                             ~*  Area (WRIAJ bouxfary
                                        Al-3

-------
       Public water suppliers are required to delineate a watershed, defined as "the hydraulic
drainage upstream of the intake", within the WQMAs.  Most of the public water supply
systems have watersheds of 300 square miles or less, although systems drawing water from
large rivers may have watersheds of several thousand square miles.

Salt Lake City, Utah Source Water Protection Delineation Method
      Figure 9. Salt Lake City's Water Supply
                     Watersheds
  Sdt loke CHy
  Watershed Uonafement Plan
  WATER BASIN BOUNDARIES
  sn a* vsu
  »*fj,»l
  vrvaa
           M ui at, tu> ua,
                                         UUTDiCFSlimi»llll7W
       The water supply watersheds
for Salt Lake City (Figure 9) (Utah
Automated Geographic Reference,
1989) were originally delineated hi
the early 1900s and have been used
ever since. Each watershed is a
canyon.  A watershed boundary is
defined topographically and
corresponds to the top of the ridge
surrounding the canyon.

       Drinking water source
protection was not explicitly
considered in the original delineation
of the canyon watersheds.  The
locations of PWS intakes define the
lowest pouits on the boundaries of
Salt Lake City's management areas.
In these areas, ordinances  and land
use controls are implemented for the
sole purpose  of drinking water
source protection. Salt Lake City
uses the entire watershed area
upstream of its surface water intakes
as the basis for its management
decisions.  Because of the
mountainous nature of the region,
drainage areas are smaller, and can
thus be more easily managed, than
the drainage area of a river of
similar size in the coastal plain of
the U.S. The total drainage area for
all seven water supply canyons is
slightly less than 200 square miles
and yields approximately 152,000
acre-feet of water per year.
                                           Al-4

-------
        In developing its land use plan and management recommendations, Salt Lake City
 delineated areas of varying development suitability. The City mapped and analyzed these
 areas using computer modeling and map overlay techniques to generate development suitability
 maps for each watershed.

        The availability of surplus water is one of the more critical development constraints in
 the canyons, because the City is maintaining a moratorium on new surplus water contracts. In
 addition, the Department of Health requires a 50 foot setback from any water source for all
 structures designed to be occupied by people or animals.

        Most other development constraints are related to the hydrologic and soil conditions of
 Salt Lake City's canyon watersheds. In calculating erosion tolerance, the City used the
 universal soil loss equation to estimate soil loss in tons/acre/year.  The equation considers a
 number of variables, including precipitation intensity, soil erosion factors, slope length and
 steepness, and the height, type, and density of native vegetation.  Based on the results of this
 equation, the City assigned erosion tolerance values to different areas, delineating regions  of
 potential  soil quality  degradation. Slope data are included in the City's erosion tolerance
 model. However, the County also considers slope  in setting development restrictions,
 generally prohibiting development on slopes with greater than 30 percent grade.  Finally, soil
 conditions such as: a very shallow water table; very shallow or outcropping bedrock;  high
 shrink-swell potential; very high or high erosion potential; strong salt or alkali effect;
 impermeability; very slow,  or slow permeability; high runoff potential; and susceptibility to
 hillside slippage, can be significant constraints on development.

       Area boundaries based on the constraints were mapped for private lands in each of the
 canyons.  Boundaries were then combined in a series of overlays to construct a development
 suitability map for all private  landholdings (Salt Lake City, 1988).

 Illinois Source Water Protection Delineation Method

       Illinois has divided the State into seven major watershed areas, which were determined
 by matching, to the extent possible, the boundaries  of the Illinois Environmental Protection
 Agency's seven regional office locations, with major river basins (as defined on USGS
 topographic maps). The State has also delineated 860 subwatersheds.  Illinois uses these
 delineations in its watershed approach, which addresses watershed prioritization, monitoring,
point and nonpoint source pollution, and water quality criteria, including both biological and
human health based criteria.  The prioritization process  identifies the watersheds with the most
critical water quality problems.  Drinking  water source  (both surface water and ground water)
quality is a primary consideration in this prioritization process (Illinois Environmental
Protection Agency, 1995).
                                          Al-5

-------
Setbacks/Buffer Zone Segment Delineation Method

Clifton Park, New York, Land Conservation Districts

       The town of Clifton Park, New York, established a Land Conservation District that
protects the wetlands and streams recognized by the State.  The District established 50 foot
buffers on either side of classified streams and 100 foot buffers for certain named stream
confluences.  These buffer zones are not delineated on town zoning maps, but are protected.
Building permits within the District are referred to the New York State Department of
Environmental Conservation to verify locations of buffers. Within these buffer zones,
construction, expansion or alteration of the land is prohibited except for purposes of parks,
residential boat launches, game preserves and similar uses (New York State Department of
Environmental Conservation, 1995).

Austin, Texas, Buffer Zones

       The City of Austin, Texas uses buffer zones to define the protection area in the Town
Lake watershed. The City requires buffers of 100 feet from the center line of "minor"
Streams, 200 feet from the center line of "intermediate" streams, and 300 feet from the center
line of "major" streams.  In addition to these buffer zones, Austin has implemented several
structural best management practices, such as sedimentation/sand filtration basins to mitigate
impacts of urban runoff (Robbins et al, 1991).

Napa County,  California, Setbacks and Zoning Restrictions

       As part of a Napa County county-wide effort to control erosion problems in the Napa
River watershed, the cities of Calistoga, St. Helena, and Yountville passed ordinances on
property creek setbacks, ranging from 35 to 100 feet, as well as zoning restrictions on building
on slopes over 30 percent grade.  Generally, development and any activities which could cause
erosion or other detrimental impacts within the riparian zone, or the "woodlands/watershed"
zoning district, are prohibited.

       Through the Napa County Resource Conservation District, a watershed-wide set  of
protection measures is recommended, including measures targeted at agricultural lands, open
space, urban and rural residential areas, commercial and industrial areas, and recreation space.
General recommendations include developing buffer and setback areas from riparian zones,
sensitive habitats, and critical erosion zones (California Water Resources Control Board, 1995;
and Napa County Resources Conservation District, 1995).
                                         Al-6

-------
                                    Appendix 2

                 Annotated Bibliography of the Technical
            Literature Addressing Source Water Protection
                      Area and Segment Delineation



 1.0   Delineating Watersheds Using Topographic Boundaries

 Black, Peter E. 1991. Watershed Hydrology. Prentice-Hall, pp. 251-270, "Characterizing the
       Watershed."

             This text provides a straightforward description of the hydrologic principles
       involved in delineating a watershed.  The text emphasizes that "watersheds are often
       not immediately discernible from a map or on the ground."  The text discusses the
       distinction between the topographic divide and the phreatic divide to account for the
       contribution of soil water and ground water runoff in watershed delineation.
       Fundamental watershed modeling concepts, such as runoff per unit areas, peak flow,
       time of concentration, and hydraulic  similitude are also described.

 Seaber, Paul R., F. Paul Kapinos, and George L. Knapp.  1987. "Hydrologic Unit Maps,"
       United States Geologic Survey Water-Supply Paper 2294.

             This paper provides an overview of the United States Geological Survey's
       (USGS's) approach for cataloging the 2,149 river-basin units in the United States.  The
       USGS has developed a series of four-color maps that present information on drainage,
       culture, hydrography, hydrologic boundaries and unit codes for (1) the 21  major water-
       resources regions and the 222 subregions designated by the U. S. Water Resources
       Council, (2) the 352 accounting units of the USGS's National Water Data Network,
       and (3) the 2,149 cataloging units of the USGS's "Catalog of Information on Water
       Data."  A complete list of all the hydrologic units, along with their drainage areas,
       their names, and the names of the States or outlying areas in which they reside, is
       contained in the report.

Terrene Institute. 1993.  "Delineating Watersheds - A First Step towards Effective
       Management."

            This document provides step-by-step descriptions of methods for delineating
       watersheds, using topographic boundaries.  The document discusses sources of
       topographic data, methods for determining drainage direction in a watershed, and
       sources of assistance for completing topographic delineations.

-------
2.0   Delineating Segments for Land Use Management: Buffers, Setbacks,
       and Management Districts

Bona, John and James Murray. 1993. "Watershed Selected for National Program," Water
       Environment & Technology, August 1993, p33-34.

             This paper describes a federally funded national demonstration program
       addressing urban runoff and combined sewer overflows (CSOs) in the Rouge River
       watershed in southeastern Michigan.  The program is to establish methods for
       determining the optimal combination of control measures and regulations to provide the
       greatest water quality improvement at reasonable costs. These methods are to be
       transferrable to urban watersheds throughout the U.S. The program began in late 1992
       and is expected to last three years.  The intent of the program is to quantify and define
       contaminant loadings from wet weather sources and identify the amount of reduction in
       contamination levels due to various control measures.  Computer models will simulate
       responses to rainfall and the magnitude and frequency of CSOs  and stormwater runoff
       quantities.  Results will be used to predict the quality of water throughout the
       watershed during and after storms.

Bowerin, Reginald, Kenneth H. Spie, Alfred T. Neal, and William E. Bullar. 1990.
       Watershed Control for Water Quality Management. Pollution Control Council,
       Portland, OR.

             The purpose of this water supply and watershed protection document is to
       demonstrate the basic principles and problems of long-range programs of water quality
       protection and management on forested watersheds in the Pacific Northwest.
       Consideration is given to natural as well as human-caused variations in water resource
       characteristics and their relationships to municipal, industrial and agricultural
       consumptive uses and the non-consumptive uses of water including fisheries and
       recreation. The report summarizes water supply and quality requirements for the
       various water uses in the Pacific Northwest,  indicates how lack of adequate control
       measures and management practices adversely affect these uses, and includes
       recommendations for improvements in watershed protection and management.

Chernin, Philip  R., and Frederick O. Brandon.  1992.  "Protecting Local Supplies:
       Perspectives from a Regional Water Purveyor," WATER/Engineering & Management,
       October  1992, p20-22.

             The Massachusetts Water Resources Authority (MWRA) has implemented a
       water resources management program designed to protect sources of drinking water it
       does not own, for the mutual benefit of its customers and their neighbors.  The
       program was intended to help partially supplied communities protect their local water
       sources and to ensure that MWRA would not have to meet an unforeseen future
                                        A2-2

-------
       demand as a result of contaminated local supplies.  The MWRA funded studies in its 14
       partially supplied communities. These studies assessed the local water supply and
       delineated protection areas for both surface and ground water supplies, identified
       potential hazards, evaluated existing water supply protection mechanisms, and
       recommended a strategy to develop or supplement existing management measures.  The
       GIS datalayers of all the mapped data and a database of potential contaminant sources
       were developed during the study and now serve as useful planning tools, both for the
       communities and for MWRA.  The study's information and  recommendations have
       helped one community  revise its water supply protection overlay district by law, and
       aided another community in preparing a successful application for an exception to
       filtration under the federal Safe Drinking Water Act Amendments.

Ehrman, Richard L., Martha L. Link, and Jeffrey J. Gottula.  1990.  "Special Protection
       Areas:  A New Nonpoint-Source Management Option in Nebraska," Journal of Soil
       and Water Conservation, March-April 1990, p263-264.

              Extensive nonpoint source contamination prompted Nebraska to enact legislation
       creating a Special Protection Area Program.  Under this program, the State Department
       of Environmental Control evaluates applications for special protection area
       consideration for areas  with ground water contamination. After setting a priority list,
       department personnel work with local officials to determine the cause or causes of the
       contamination, a study that includes surface water sampling and consideration of
       ground water/surface water interaction. If there is evidence  of nonpoint source
       contamination, the department designates a special protection area and the local district
       develops an action plan and a monitoring program for the protection area to address the
       contamination.  The State has experienced some difficulties with insufficient funds to
       implement action plans  and with setting the boundaries of the protection areas.

Eichner, Eduard M.  1993. "Watershed Protection: A Cape Cod Perspective on National
       Efforts," Environmental Science & Technology, 27(9): 1736-1740.

             In response to rapid development, the Cape Cod Commission established a
       Regional Policy Plan (RPP) containing a strategy for watershed  delineation and
       protection.  Because Cape Cod's hydrologic system is dominated by ground water,
       watersheds are delineated based on the configuration of ground water lenses rather than
       on topographic configurations.  Once delineated, the RPP classified the watersheds and
       identified water resources of concern, developing  watershed-specific standards and
       frameworks for development.  High nitrogen concentrations limit land use
       development.  This strategy is dependent on having land use  information and GIS
       technology to assess the amount of nitrogen loading to each watershed.  Cape Cod has
       found the concept of single contaminant regulation and land use  planning to be fairly
       effective. However, as  new technology reduces nitrogen loading, the need for land use
       controls based on additional criteria is becoming apparent.
                                         A2-3

-------
Goldstein, Kenneth J., Anne B. Benware, and David M. Kutner.  1994.  "A New York State
       of Mind,"  Water Environment & Technology, January 1994, 34:34-39.

             Municipalities served by a Schenectady, NY aquifer have joined together to
       create a wellhead and watershed protection strategy.  Tools used in this strategy
       include: intermunicipal watershed rules, uniform land use regulations for the protected
       areas, and wellhead protection areas around each municipal well field.  The
       mtermunicipal watershed rules and regulations have three main objectives:  minimum
       standards, uniform application, and accountability.  Limitations on land uses are most
       stringent within the wellhead protection zone and become progressively less stringent in
       the primary recharge zone (zone 2), the general aquifer recharge zone (zone 3), and the
       tributary watershed zone (zone 4). The tributary watershed zone is defined as the
       surface water drainage basins located in the valley uplands where runoff flows overland
       and into defined stream beds until it reaches the general aquifer recharge zone.  In the
       tributary watershed zone, snow or ice collected off-site from roadways, agricultural
       chemicals or pesticides, and coal or chloride salts are prohibited within 100 feet of any
       body of water.

Harryman, M. B. M. 1989.  "Water Source Protection and Protection Zones," Journal of the
       Institute of Water and Engineering Management, December 1989, p548-550.

             This paper discusses United Kingdom Government policy to August 1989 on the
       use of protection zones to protect drinking water sources from pollution by nitrate.
       The concept of protection zones was established to  provide a mechanism to limit
       pollution from nonpoint sources.  The paper outlines  the use of the powers in the Water
       Act 1989 to declare Nitrate Sensitive Areas,  and the procedures that would be followed
       in declaring protection zones.  It outlines the circumstances in which compensation
       would be paid in Nitrate Sensitive Areas and presents the reasons for departing from
       the  'polluter pays' principle. In addition, it discusses the similarities between the  UK
       policy and the proposed European Community Directive on the control of nitrate.

Machorro, Eric. 1994. "Portland Prepares New Watershed Plan," Water Environment &
       Technology, February,  1994, p33.

             Portland,  Oregon and several neighboring communities are developing a
       second-generation watershed management plan for  Johnson Creek, a tributary of the
       Willamette River, which drams a 54 square mile partially urbanized watershed with a
       population of 200,000.  The plan outlines a more comprehensive approach to watershed
       protection, linking point and nonpoint source control programs.  The plan's goals
       include reducing flood frequency, restoring salmon and steelhead populations, and
       developing the creek corridor  as a refuge for wildlife. To meet these goals, the plan
       may include construction of passive treatment facilities, temporary storage facilities for
       stormwater, and revegetation of the stream corridor.
                                          A2-4

-------
Osterman, Douglas, Frederick Steiner, Theresa Hicks, Ray Ledgerwood, and Kelsey Gray.
       1989.  "Coordinated Resource Management and Planning:  The Case of the Missouri
       Flat Creek Watershed," Journal of Soil and Water Conservation, September-October
       1989, p403-406.

              This paper describes a working method of coordinated resource management
       and planning that has evolved from,  and is being tested through, the Missouri Flat
       Creek Watershed Plan. The ultimate goal is to reduce and finally eliminate sediment
       and other agricultural runoff in Missouri Flat Creek.  The strategy includes planning
       and implementing actions on a watershed basis; taking into consideration the social,
       political, economic, institutional, and biophysical processes of the area; implementing
       farm conservation plans on all of the cropland in the watershed; stabilizing the
       streambank and channel of the creek; and supplementing all actions with area wide
       education and information programs. Addressing the problem on a watershed basis
       provided a clear picture to farmers and planners  of the dynamic natural and social
       interactions affecting soil  erosion and water quality in the watershed. The watershed
       approach, combined with  excellent communication, enabled planners to effectively
       target their efforts and united farmers in their attempts to reduce soil erosion.  In
       addition, increased awareness, education and cooperation has enabled the farmers to
       reach consensus on goals for general resource management in the watershed.

Porter, Keith.  1994. "Development and Watershed Protection: Finding the Middle Ground,"
       Cornell Engineering Quarterly, Spring 1994, p8-13.

             An interdisciplinary team at Cornell University is researching several issues
       related to proposed protection measures for New York City's water supply, which were
       contained hi regulations proposed by New York City's Department of Environmental
       Protection (NYCDEP) in 1990.  Although it was intended to promote public debate,
       the document created a public uproar by residents of the nearly 2,000 square mile
       watershed.  Degradation of the Croton reservoir system, located in Westchester and
       Putnam counties, has required  filtration. To avoid the costs of filtration for the
       Catskill-Delaware part of the system, strict watershed management plans are required.
       One of the major threats to water quality are the phosphates used as fertilizers by
       farms, which cause eutrophication. Also, pathogenic protozoa may be spread by farm
       animals. The Cornell Pathogen Group is focusing on farm animals as vectors for
       water-borne disease. The  Cornell group is also focusing on "whole farm planning", by
       bringing together specialists from a variety of fields to focus on farm management.
       New York City has provided $3.9 million for the first phase of a long-term
      demonstration program involving ten dairy farms, and  has committed $35 million to the
       second phase.  The theory behind "whole farm planning" has also been applied more
      widely,  using the concept of "whole community planning" within the watershed.  This
      effort was led by the Coalition  of Watershed Towns and the NYC DEP. Although
      political issues have made progress difficult, a Cornell  group and the City Department
                                         A2-5

-------
       of Regional Planning plan to use the research issues, which emerged from these
       studies, to continue work.

Robbins, Richard W., Joseph L. Glicker, Douglas M. Bloem, and Bruce M. Niss.  1991.
       Effective Watershed Management for Surface Water Supplies. Journal of the American
       Water Works Association.

             This report is designed to assist water utility managers and local governments in
       developing effective watershed protection programs for their surface water supplies.  It
       presents a detailed discussion of the management options and control measures that are
       available to limit the water quality impacts of land-use activities. It describes general
       control measures that apply to several types of land use, such as land acquisition,
       trespass control, watershed inspection programs, reservoir-use restrictions, stream and
       reservoir buffers, plan review, written agreements with landowners, and public
       education and involvement.  The report also describes best management practices
       (BMPs) for agricultural land and silvicultural activities and structural and nonstructural
       nonpoint source controls. In addition, the report provides information on the
       implementation of watershed management programs.  Finally, the report documents
       twenty-four case studies of effective watershed management programs used by water
       utilities or their cooperating jurisdictions to protect raw water quality.

U.S. Environmental Protection Agency. 1993.  Guidance Specifying Management Measures
       For Sources of Nonpoint Pollution in Coastal Waters. USEPA, Office of Water.  EPA-
       840-B-92-002, January 1993.

             This document provides guidance on the implementation of management
       measures for the control of sources of nonpoint pollution.  It addresses five source
       categories of nonpoint pollution:  agriculture, silviculture, urban, marinas, and
       hydromodification.  A variety of management practices are discussed for each source
       category.  The guidance provides information on the applicability, selection,
       implementation, and effectiveness of each practice.  In addition, one chapter discusses
       other management tools that are available to address many source categories of
       nonpoint pollution.  These tools include the protection, restoration, and construction of
       wetlands, riparian areas, and vegetated treatment systems.

U.S. Environmental Protection Agency, Chesapeake Bay Program. 1993. "Role and Function
       of Forest Buffers in the Chesapeake Bay Basin for Nonpoint Source Management."

             This document addresses the impact that urbanization and deforestation have
       had on water quality in the Chesapeake Bay.  The EPA summarizes research from a
       variety of sources on the effectiveness of the riparian forest in reducing nonpoint
       source load in runoff and ground water.  Most of the research has been done in
       agricultural watersheds or hi connection with silvicultural activities. The document
                                         A2-6

-------
       also describes the use of buffer strips as a management practice.  Forest buffers are
       also recognized for their high value in wildlife and fish habitat and maintaining
       ecosystem integrity.

Whipple, William Jr.  1993.  "Buffer Zones Around Water-Supply Reservoirs," Journal of
       Water Resources Planning and Management, 119(4):495-499.

              This paper outlines the use of buffer zones to protect reservoir water quality
       from the effects of development and presents an approach to preclude development in a
       front buffer zone adjacent to a reservoir and the lower portions of the tributaries
       draining to the reservoir.  Development in a wider zone up-slope from the front buffer
       would be required to implement special controls for nonpoint source pollution. This
       special control buffer zone would provide for reduction in pollution from the runoff
       that reaches the reservoir from channels and tributaries.

3.0    Modeling Land Use Impact on Source Water Quality

American Society of Civil Engineers Task Committee on Definition of Criteria for Evaluation
       of Watershed Models of the  Watershed Management Committee, Irrigation and
       Drainage Division.   1993.  "Criteria for Evaluation of Watershed Models," Journal of
       Irrigation and Drainage Engineering, 119(3):429-442.

             This paper does not provide an overview of delineation methods; however, it
       does provide criteria and recommendations for evaluating site-specific hydrologic
       models. The ASCE  Committee is seeking to promote a standard set of reporting
       criteria for descriptions of all watershed models:  The report recommend that the
       following be addressed in all watershed model documentation:

             •     a complete description of model parameters, parameter selection, or
                   discussion of the range of parameters describing hydrologic flow in the
                   watershed;

                   availability of the types of data needed to set up  and run the model  (e.g.,
                   readily available USGS gaging data);

                   reports of model validation and testing and the range of conditions over
                   which  the model was tested; and

                   the model results in relation to other models or modeling approaches.
                                        A2-7

-------
Bou-Saab, Jamil F.  1993.  "Runoff as a Resource," Civil Engineering, October 1993, p70-71.

             This article examines the BMPs used by one university to reduce pollutant loads
       in storm water runoff and to use the runoff to recharge ground water sources of
       drinking water. Storm water flows typically contain significant quantities of the same
       pollutants found in wastewater and industrial discharges, making storm water an
       important factor in source-water and public-health protection. This article explores
       how BMPs can be used to treat runoff as a resource, controlling it on the surface rather
       than building treatment plants at the discharge point. The BMPs were designed to both
       reduce pollutant loads in runoff and to conserve water through ground water recharge.
       In addition, these BMPs had the effect of reducing storm water flows,  increasing flood
       control and erosion protection, and lowering costs for storm water storage and
       transport.

Cooper, A. Bryce and Adelbert B.  Bottcher. 1993.  "Basin-Scale Modeling as Tool for
       Water-Resource Planning,"  Journal of Water Resources Planning and Management,
       119(3):306-323.

             This paper describes the development of a basin-scale model designed to
       simulate long-term average losses of water, sediment,  and nutrients from large, rural
       watersheds.  The CREAMS model (Chemicals, Runoff, and Erosion from Agricultural
       Management Systems) was used.  This model incorporates diffuse and point sources
       and riparian and stream channel processes. The model was accurate in predicting the
       direction and magnitude of changes resulting from implementation of a riparian pasture
       retirement scheme (a BMP used in New Zealand). Model predictions of soluble
       nutrient losses were sensitive to variation in stream attenuation coefficients, and users
       are cautioned that uncertainties in watershed and riparian processes can affect model
       results.

DeVantier, Bruce A., and Arlen D. Feldman. 1993. "Review of GIS Applications hi
       Hydrologic Modeling," Journal of Water Resources Planning and Management,
       119(2):246-261.

             This paper describes applications  of GIS technologies to support hydrologic
       modeling. The GIS data architecture provides a digital representation of watershed
       characteristics.  This paper summarizes applications of GIS-based digital terrain models
       to support hydrologic analyses. Three methods of geographic information storage are
       discussed: raster or grid, triangulated irregular network, and contour-based line
       networks. The computational, geographic, and hydrologic aspects of each data-storage
       method are analyzed.  The use of remotely sensed data in GIS and hydrologic modeling
       is also reviewed. Lumped parameter, physics-based, and hybrid approaches to
       hydrologic modeling are discussed with respect to then: geographic data inputs.
                                         A2-8

-------
 El-Kadi, Aly I. 1989. "Watershed Models and Their Applicability to Conjunctive Use
       Management," Water Resources Bulletin, 25(1): 125-137.

              This paper assesses the applicability of typical watershed hydrograph models to
       provide data on ground water recharge to support conjunctive use management. The
       paper assesses the degree to which 28 models, including HEC-1, CREAMS, and HSP,
       allow for estimations of the conjunctive use of ground water and surface water
       resources.  The author concluded that the large number of processes that these models
       simulate prohibits detailed analysis of subsurface flow, due to excessive computer and
       data requirements.  The models emphasize surface flow and include only that portion of
       water lost to the subsurface and the portion returned to the stream as baseflow.  These
       models do not easily accommodate the impacts of conjunctive use, although the paper
       does describe a framework for including these considerations in watershed models.

 Engel, B. A., R. Srinivasan, J. Arnold, C. Rewerts, and S. J. Brown.   1993.  "Nonpoint
       Source (NFS) Pollution Modeling Using Models Integrated with  Geographic
       Information Systems (GIS),"  Water Science  & Technology, 28(3-5):685-690.

             This paper assesses three nonpoint source pollution models which were
       integrated with a GIS and used to simulate a  watershed response  to a series of rainfall
       events.  Simulated responses were compared with actual runoff and sediment data from
       observed events, and were found to match reasonably well. ANSWERS (Areal
       Nonpoint Source Watershed Environmental Response Simulation), AGNPS
       (Agricultural Nonpoint Source Pollution), and SWAT (Soil and Water Assessment
       Tool) were  used and integrated with the GRASS GIS. ANSWERS was developed
       primarily for agricultural watershed analysis, AGNPS analyzes nonpoint source
       pollution in agricultural watersheds, and SWAT is another watershed model that is
       flexible in watershed configuration. The match between observed and simulated results
       was noted, especially  since inputs were estimated using GIS data and were  not
       calibrated for the particular watershed.

Etzel, Ronald A., and Ginger K. Ellis. 1990.  "Comprehensive Watershed Management
       Planning - A Case Study - The Magothy River."   In Riggins, Robert E., et al.,
       editors.  Watershed Planning and Analysis in Action, Proceedings of the Symposium
       sponsored by the Committee on Watershed Management of the Irrigation and Drainage
       Division of the American Society of Civil Engineers, Durango, Colorado, July 9-10
       1990.

             This paper examines existing hydrologic, hydraulic, and environmental
       conditions in the Magothy River Watershed to estimate impacts of future development.
       The objectives of the Magothy River Comprehensive Watershed Management Plan
       Study were to: identify existing water quality, flooding and sedimentation problems;
       evaluate  the interaction of upstream flows and water quality with  estuarine flows and
                                        A2-9

-------
       water quality of the river to reduce upstream pollutant loadings; and identify and
       evaluate existing and potential management strategies, recommend appropriate
       strategies, and compose a master plan for overall guidance. Estuary dynamics were
       modeled using a mathematical simulation and calibrated by comparing results with
       those of a dye study. From this, it was concluded that concentrations of pollutants
       were diluted in the river and the effects of pollutant loads were seen at the mouth of the
       tributaries. Historic data and Nationwide Urban Runoff Program data were used to
       assess existing water quality, which met state standards, although water uses were
       restricted in certain locations. Soil Conservation Service methodology was used to
       determine existing and future stormwater runoff rates and volumes, erosion,
       sedimentation, and flooding potential.  Of the 18 subareas  analyzed, 11  showed reaches
       of stream channel erosion. No sedimentation problems in  the streams were noted,
       although sedimentation has become a problem at the tidal interface.  Twelve flood
       prone areas were identified; these were usually  roadways with undersized culverts.
       Common watershed management approaches, such as stormwater facility management
       projects, were easily cited, while more difficult management approaches,  such as land
       use zoning changes and BMP ordinances, were  only suggested.

Gburek, William J. 1983. "Hydrologic Delineation of Nonpoint Source Contributing Areas,"
       Journal of Environmental Engineering,  109(5): 1035-1048.

              This paper describes a model that determines the area of a watershed that
       contributes surface runoff to a stream after a rain storm. The paper describes steps
       taken to develop the return period within a watershed, and the potential for nonpoint
       source pollution to enter the stream.  Design rainfall data and initial watershed soil
       moisture content are used as inputs.  The model is demonstrated using a small,
       agricultural watershed hi east-central Pennsylvania.  Areas of the watershed having the
       potential to deliver surface-applied contaminants to the stream channel for one-hour
       storms at different periods of occurrence (1 to 25 year occurrence periods) are
       delineated, and travel distances are compared between different areas of the watershed.

Griner, Axel J. 1993. "Development of a Water Supply Protection Model in a GIS," Water
       Resources Bulletin, 29(6):965-971.

              This paper describes a water  supply protection model developed using the
       Southwest Florida Water Management District's ARC/INFO GIS software.  Several
       hydrologic and hydrogeologic layers were overlaid to develop maps showing ground-
       water supply suitability, protection areas for surface water supply, protection areas for
       major public supply wells, susceptibility to contamination, and recharge to an aquifer.
       Protection areas for surface water supply were  derived for 11 existing and potential
       withdrawal points. The most critical areas were those immediately adjacent to rivers
       and their tributaries upstream from the withdrawal point.  Actual delineations of these
                                         A2-10

-------
       areas were determined by generating buffers around the stream courses. Wellhead
       protection areas were also delineated.

Houlahan, John, W. Andrew Marcus, and Adel Shirmohammadi.  1992.  "Estimating
       Maryland Critical Area Act's Impact on Future Nonpoint Pollution Along the Rhode
       River Estuary," Water Resources Bulletin, 28(3):553-567.

              The authors present the results of a study assessing the impact of the Maryland "*
       Critical Area Act on the generation of nonpoint source loads of phosphorus, nitrogen,
       and sediment to the Rhode River Estuary.  Three different nonpoint simulation models
       (CREAMS, Simple Method, and Marcus/Kearney regressions) were used to estimate
       the generation of annual areal nutrient and sediment loadings under four development
       scenarios. The Simple Method Model was used to estimate generation of nitrogen and
       phosphorus loads from developed and forested areas.  Sediment erosion from non-
       agricultural lands was estimated using Marcus and Kearney's regressions for suspended
       sediment data derived from continuous suspended sediment collected at seven sites
       between 1975 and 1987. The CREAMS model was used to model nitrogen,
       phosphorus,  and sediment from farmland.  Development conditions modeled included:
       present conditions, maximum land use development allowable with the Critical Area
       Act, and two development scenarios without the Act's restrictions in place.  Results
       indicate that the Act can reduce the present generation of nonpoint nutrient and
       sediment loadings 20 to 30 percent from the regulated area, while preserving
       agricultural lands and allowing limited development.  This reduction is primarily due to
       agricultural BMPs. Impacts of the Act are even greater when compared with
       unregulated future development, primarily through limiting uncontrolled woodland
       cutting.

Kimball, Kathleen, and Doug Beyerlein.  1990.  "Intergovernmental Agreements in Watershed
       Planning." In Riggins, Robert E., et al.,  editors.  Watershed Planning and Analysis in
       Action. Proceedings of the Symposium sponsored by the Committee on Watershed
       Management of the Irrigation and Drainage Division of the American Society of Civil
       Engineers, Durango, Colorado, July 9-10, 1990.

             Cooperative efforts among three communities in Washington are described. A
       detailed inventory of the drainage network was performed, and slopes of culverts were
       noted for use in modeling. An existing USGS model for Scriber Creek was used as a
       starting point of the inventory;  the watershed was split into subbasins, the model was
       calibrated using two years of streamflow  and rainfall data, and peak stream flows were
       used to assess the impact of land use changes on flood peaks.  Recommendations in the
       watershed plan included structural and non-structural solutions to increased flood peaks
       caused by land-use changes.  Structural solutions included regional detention facilities
       and local conveyance-systems improvements, while nonstructural solutions included
       public information programs, ordinances  and guidelines for BMPs. Specific BMP
                                        A2-11

-------
       recommendations included:  new policies regarding revegetating roadside ditches,
       expansion of existing regulations for clearing and grading, and ordinances for buffers
       and nature-protection growth areas adjacent to riparian zones.

Kuo, Chin Y., Kelly A. Cave, and G. V. Loganathan.  1988.  "Planning of Urban Best
       Management Practices," Water Resources Bulletin, 24(1): 125-132.

*             This paper describes  a user friendly computer model that was developed to
       compare the feasibility and cost effectiveness of alternative urban nonpoint source
       pollution control measures.  The model determines the effectiveness of individual
       BMPs and addresses concerns about management strategies for combining facilities into
       an effective basin-wide control program.  The model is  a tool that allows planners to
       select the size, location, and type of BMP and to plan land uses to meet stormwater
       quantity  and quality requirements.  The model generates hydrographs and
       pollutographs at the basin and sub-basin outlets for the present and the post-
       development conditions, and with or without BMP measures.  The model includes
       detention basins, infiltration trenches, and porous pavements.  In general,  it was found
       that the extended wet ponds  were the most cost effective controls of the measures
       evaluated.

Lackaff, Beatrice B., Bruce J. Hunt, and Ian E. Von Essen.  1993.  "The Development and
       Implementation of a  GIS-Based Contaminant Source Inventory over the  Spokane
       Aquifer,  Spokane County, Washington,"  Water Resources Bulletin, 29(6):949-955.

              This study describes the use of GIS in the development of a protection program
       for a large and contaminant sensitive underground source  of drinking water. Because
       of its extremely high transmissivity, all wells hi the Spokane-Rathdrum Aquifer are
       included  in one wellhead protection zone  called the Aquifer Sensitive Area (ASA).
       Because  of the large area and population within the ASA,  the use of GIS technology
       and existing datasets was required to create a Contaminant Source Inventory (CSI) for
       the ASA. Datasets listing businesses and agencies within  the ASA were imported into
       the GIS from State, county, city, and local agencies.  These datasets were selected,
       joined, and  sorted using GIS relational database capabilities into one ASA "business
       master file." Map files were projected and transformed into common coordinates.
       Next, business sites within the master file were spatially related by address to the
       digital map  files.  Likely Critical Materials Users (CMU)  were identified by sorting on
       selected  Standard Industrial Codes (SIC). Additional files of CMUs were imported into
       the Contaminant Source Inventory.  Geographic Information System queries were
       performed to locate specific  materials, quantities, and storage facilities,  and to analyze
       CMU activity within selected buffer zones. The GIS technology was critical in this
       project in the development, management, maintenance,  and analysis of the vast
       quantities of data associated  with this aquifer protection program. The GIS proved to
       be a valuable tool for future  resource and land use planning.
                                         A2-12

-------
Summer, R. M., C. V. Alonso, and R. A. Young. 1990. "Modeling Linked Watershed and
       Lake Processes for Water Quality Management Decisions," Journal of Environmental
       Quality, 19:421-427.

             This paper describes the use of a watershed, agricultural nonpoint source
       model, AGNPS, and a lake model, FARMPND, to link watershed and lake processes
       and to assess effects of various land use practices and weather conditions.  Parameters
       used in the AGNPS model were: hydrology, erosion, nitrogen and phosphorus
       transport, sediment transport, and chemical oxygen demand.  These variables provided
       input for the lake model, which simulated temperature stratification, mixing by wind,
       sedimentation, inflow density current, and algal growth.  Outflow from the lake
       provides input to downstream segments in the watershed model. Simulations were
       conducted on Eagle Lake watershed in west central Minnesota. Changes in water
       quality were examined by modifying land area hi wetlands and land area with retired
       cropland (permanent grass cover with  terraces). Less volume of runoff and less
       transported sediment were noted under wetland versus non-wetland conditions, and
       improvements in water quality of both in-lake and outflow were noted with more
       credible  cropland taken out of use.

Tim, Udoyara S. and Robert Jolly. 1994. "Evaluating Agricultural Nonpoint-Source Pollution
       Using Integrated Geographic Information Systems and Hydrologic/Water Quality
       Model,"  Journal of Environmental Quality, 23:25-35.

             This study describes integration of the AGNPS model with a GIS to analyze
       nonpoint source pollution hi an agricultural watershed. The GIS organized the data
       spatially, while the AGNPS model was used to predict soil erosion and sedimentation
       within a watershed. The model and GIS were used to evaluate the impact of BMPs
       implemented in a watershed in Iowa, where agricultural runoff was creating
       sedimentation.  Several land management strategies were simulated, including the use
       of vegetative filter strips along the primary streams (strategy 1), using  contour buffer
       grass strips on all cropland areas (strategy 2), and combining the first two strategies
       (strategy  3). Strategy 3 yielded the highest sediment load reduction (71 percent), while
       contour buffer strips alone would not make an appreciable difference in sedimentation
       improvement.  The study concludes that the GIS and the model are powerful
       interpretive tools, and are useful in providing a framework for assessing the
       effectiveness of several land management strategies; however, the tune investment may
      be considerable.
                                       A2-13

-------
U.S. Environmental Protection Agency.  1984.  "Modeling Water Quality and the Effects of
       Agricultural BMPs (Best Management Practices) in the Iowa River Basin,"
       EPA-600/D-84-112.

             This paper describes a demonstration application of comprehensive hydrology
       and water quality modeling on a large river basin, to evaluate the effects of agricultural
       nonpoint pollution and proposed BMPs. The model application combines detailed
       simulation of agricultural runoff and soil processes, calculates surface and subsurface
       pollutant transport to receiving water, and simulates instream transport and
       transformation. The result is a comprehensive simulation of river basin water quality.
       The investigation of the Iowa River Basin described in this paper was part of a large
       study which included application, and evaluation, of the Hydrological Simulation
       Program - FORTRAN (HSPF). The HSPF was applied to, and evaluated at, the
       data-intensive Four Mile Creek watershed and at the Iowa River above Coralville
       Reservoir.  This study allowed the exploration of problems associated with modeling
       hydrology,  sediment transport, and chemical fate and transport hi a large river basin
       with varying meteorologic  conditions, soils and agricultural practices.

4.0    Time-of-Travel Calculation

Lindsley, R.K.  1985.  Hydrology for Engineers. McGraw-Hill. New York.

             This text describes methods for calculating stream velocity based on channel
       size, channel roughness,  gaged discharge rates, and regional topography, to assist hi
       determining time-of-travel  estimates for spills or other releases.  This text also
       describes available stream stage and gaging data, and methods for interpolating
       between gaged data hi order to estimate discharge and velocity in ungaged reaches.

U.S. Government Printing Office. 1995. Code of Federal Regulations, 40 CFR Section 112:
       Oil Pollution Prevention, Appendix C-ffl: Calculation of the Planning Distance.

             This regulation provides guidance for determining if an oil storage or  transport
       facility is located at a distance such that a discharge from the facility would result in
       the shutdown of a public drinking water intake. The method discussed takes  into
       account river velocity, the material discharged,  and a planning "time-of-travel."  This
       method may be applied to determining travel tune from points upgradient of the intake.
                                         A2-14

-------
                                    Appendix 3

                                 Literature Cited
 American Society of Civil Engineers Task Committee on Definition of Criteria for Evaluation
       of Watershed Models of the Watershed Management Committee, Irrigation and
       Drainage Division.  1993.  "Criteria for Evaluation of Watershed Models," Journal of
       Irrigation and Drainage Engineering,  119(3):429-442.

 Black, P. E.  1991. Watershed Hydrology. Prentice Hall.  Englewood Cliffs, New Jersey.

 Bona, John and James Murray.  1993.  "Watershed Selected for National Program,"  Water
       Environment & Technology, August 1993, p33-34.

 Bou-Saab, Jamil F.  1993.  "Runoff as a Resource," Civil Engineering, October 1993, p70-71.

 Bowerin, Reginald, Kenneth H. Spie, Alfred T. Neal, and William E. Bullar. 1990.
       Watershed Control for Water Quality Management. Pollution Control Council,
       Portland OR.

 California Water Resources  Control Board, California Environmental Protection Agency.
       1995.  California's Rivers and Streams - Working Toward Solutions.

 Chernin, Phil and Frederic O. Bandon.  1992. "Protecting Local Supplies: Perspectives from
       a Regional Water Purveyor",  WATER/Engineering & Management, October:20-22.

 Cooper, A. Bryce and Adelbert B. Bottcher.   1993. "Basin-Scale Modeling as Tool for
       Water-Resource Planning," Journal of Water Resources Planning and Management
       119(3):306-323.

 DeVantier, Bruce A., and Arlen D. Feldman.  1993. "Review of GIS Applications in
       Hydrologic Modeling," Journal of Water Resources Planning and Management
       119(2):246-261.

Ehrman, Richard L., Martha L. Link, and Jeffrey J. Gottula.  1990. "Special Protection
       Areas:  A New Nonpoint-Source Management Option in Nebraska," Journal of Soil
       and Water Conservation, March-April  1990, p263-264.

Eichner, Eduard M. 1993.  "Watershed Protection: A Cape Cod Perspective on National
       Efforts," Environmental Science & Technology, 27(9): 1736-1740.

-------
El-Kadi, Aly I. 1989. "Watershed Models and Their Applicability to Conjunctive Use
       Management," Water Resources Bulletin, 25(1): 125-137.

Engel, B. A., R. Srinivasan, J. Arnold, C. Rewerts, and S. J. Brown.  1993.  "Nonpoint
       Source (NFS) Pollution Modeling Using Models Integrated with Geographic
       Information Systems (GIS)," Water Science & Technology, 28(3-5):685-690.

Etzel, Ronald A., and Ginger K. Ellis.  1990. "Comprehensive Watershed Management
       Planning  - A Case Study - The Magothy River." In Riggins, Robert E., et al.,
       editors. Watershed Planning and Analysis in Action, Proceedings of the Symposium
       sponsored by the Committee on Watershed Management of the Irrigation and Drainage
       Division of the American Society of Civil Engineers, Durango, Colorado, July 9-10,
       1990.

Gburek, William J. 1983. "Hydrologic Delineation of Nonpoint Source Contributing Areas,"
       Journal of Environmental Engineering, 109(5): 1035-1048.

Georgia State Statutes for Planning 50-8-7.  1991.  Rules for Department of Natural Resources
       Chapter 110-3-2.

Georgia State Statutes for Criteria 12-2-8.  1991. Rules for Department of Natural Resources
       Chapter 391-3-16-.01.

Goldstein, Kenneth J., Anne B. Benware and David M. Kutner. 1994.  "A New York State of
       Mind," Water Environment & Technology, January:34-39.

Griner, Axel J. 1993. "Development of a Water Supply Protection Model in a GIS," Water
       Resources Bulletin,  29(6):965-971.

Haith, D. A. and L. L. Shoemaker. 1987.  "Generalized Watershed Loading Functions for
       Stream Flow Nutrients," Water Resources Bulletin, 23(3):471-478.

Harryman, M. B. M.  1989. "Water Source Protection and Protection Zones," Journal of the
       Institute of Water and Engineering Management, December 1989, p548-550.

Houlahan, John,  W. Andrew Marcus, and Adel Shirmohammadi. 1992.  "Estimating Maryland
       Critical Area Act's Impact on Future Nonpoint Pollution Along the Rhode River
       Estuary," Water Resources Bulletin, 28(3):553-567.

Illinois Environmental Protection Agency, Division of Water Pollution Control. 1995.  Water
       Pollution  Control Program Plan. IEPA/WPC/95-017.
                                        A3-2

-------
 Kimball, Kathleen, and Doug Beyerlein.  1990.  "Intergovernmental Agreements in Watershed
       Planning." In Riggins, Robert E., et al., editors.  Watershed Planning and Analysis in
       Action. Proceedings of the Symposium sponsored by the Committee on Watershed
       Management of the Irrigation and Drainage Division of the American Society of Civil
       Engineers, Durango, Colorado, July 9-10, 1990.

 Kuo, Chin Y., Kelly A.  Cave, and G. V. Loganathan. 1988.  "Planning of Urban Best
       Management Practices," Water Resources Bulletin, 24(1): 125-132.

 Lackaff, Beatrice B., Bruce J. Hunt, and Ian E. Von Essen.  1993.  "The Development and
       Implementation of a GIS-Based Contaminant Source Inventory over the Spokane
       Aquifer, Spokane County, Washington," Water Resources Bulletin, 29(6):949-955.

 Lindsley, R.K. 1985. Hydrology for Engineers. McGraw-Hill. New York.

 Machorro,  Eric. 1994.  "Portland Prepares New Watershed Plan," Water Environment &
       Technology, February, 1994, p33.

 Napa County Resource Conservation District.  1995.  Napa River Watershed Owners Manual
       - An Integrated Resource Management Plan.

 New York  State Department of Environmental Conservation - Division of Water.  1995.
       Survey and Compendium of Local Laws for Protecting Water Quality from Nonpoint
       Source Pollution.

 North Carolina Administrative Code Section: 15A NCAC 2B .0100 - Procedures for
       Assignment of Water Quality Standards and  15 A NCAC 2B .0200 - Classifications and
       Water Quality Standards Applicable to Surface Waters of North Carolina. 1994.

 North Carolina Department of Health. 1995.  Report of Proceedings Proposed Revisions to
       the Water Supply Watershed Protection Rules C15A NCAC 2B .0100. .0200 & .0300X

 North Carolina Division of Environmental Management, Water Quality Section. 1995.
       French Broad River Basinwide Water Quality Management Plan.

 Osterman, Douglas, Frederick Steiner, Theresa Hicks, Ray Ledgerwood, and Kelsey Gray.
       1989.  "Coordinated Resource Management and Planning:  The Case of the Missouri
       Flat Creek Watershed," Journal of Soil and Water Conservation, September-October
       1989, p403-406.

Phillips, Jonathan D.  1989.  "Nonpoint Source Pollution Risk Assessment in a Watershed
       Context," Environmental Management, 13(4):493.
                                        A3-3

-------
Porter, Keith.  1994.  "Development and Watershed Protection:  Finding the Middle Ground,"
       Cornell Engineering Quarterly, Spring 1994, p8-13.

Robbins, Richard W., Joseph L. Glicker, Douglas M. Bloem, and Bruce M. Niss.  1991.
       "Effective Watershed Management for Surface Water Supplies," Journal of the
      American Water Works Association, December(34-44).

Salt Lake City, Department of Public Utilities and Salt Lake City Planning Division.  1988.
      Salt Lake City Watershed Management Plan.

Seaber, Paul R., F. Paul Kapinos, and George L. Knapp.  1987. "Hydrologic Unit Maps,"
      United States Geological Survey Water-Supply Paper 2294.

Slimmer, R.  M., C. V. Alonso, and R. A. Young.  1990.  "Modeling Linked Watershed and
      Lake  Processes for Water Quality Management Decisions," Journal of Environmental
       Quality, 19:421-427.

Terrene Institute.  1993.  Delineating Watersheds - A First Step towards Effective
      Management.

Tim, Udoyara S. and Robert Jolly.  1994.  "Evaluating Agricultural Nonpoint-Source
      Pollution Using Integrated Geographic Information Systems and Hydrologic/Water
      Quality Model," Journal of Environmental Quality, 23:25-35.

U.S. Environmental Protection Agency, Office of Water.  1997. State Source Water
       Assessment And Protection Programs Guidance. Final Guidance.  EPA 816-R-97-009.

U.S. Environmental Protection Agency, Office of Water.  1993. Guidance Specifying
      Management Measures for Sources of Nonpoint Pollution in Coastal Waters.  EPA-
      840-B-92-002.

U.S. Environmental Protection Agency, Chesapeake Bay Program.  1993.  "Role and Function
      of Forest Buffers in the Chesapeake Bay Basin for Nonpoint Source Management."

U.S. Environmental Protection Agency, Office of Water.  1987. Delineation of Wellhead
      Protection Areas.

U.S. Environmental Protection Agency.  1984.  "Modeling Water Quality and the Effects of
      Agricultural BMPs (Best Management Practices) in the Iowa River Basin." EPA-
      600/D-84-112.

U.S. Government Printing Office.   1995.  Code of Federal Regulations, "40 CFR Section 112:
      Oil Pollution Prevention, Appendix C-JH: Calculation of the Planning Distance."
                                        A3-4

-------
Utah Automated Geographic Reference.  1989.  Watershed Canyons Reference Map.

Wagenecht, David.  1997.  Written communication. Louisiana Department of Environmental
      Quality.

Wagenecht, David.  1991.  Early Warning Organic Compound Detection System. Map.
       Louisiana Department of Environmental Quality.

Washington Department of Ecology, Water Quality Program. 1996.  Water Quality
       Management Areas. Map.

Whipple Jr., William.  1993.  "Buffer Zones Around Water-Supply Reservoirs," Journal of
       Water Resources Planning and Management. 119(4)-.495-499.

Willmer, Ralph R.  1992.   "The Role of Watershed Protection Planning and the Surface Water
       Treatment Rule," The Environmental Professional, 14:125-135.
                                       A3-5

-------

-------