EPA-4W3-77-024
Analysis
*rogram of Triangle J
found/ of Governments
A Case History in 208 Water Quality Management Planning
-------
EPA-440/3-77-024
November 1978
NONPOINT SOURCE ANALYSIS PROGRAM OF
TRIANGLE J COUNCIL OF GOVERNMENTS
\
UJ
U.S. ENVIRONMENTAL PROTECTION AGENCY
* Office of Water Planning and Standards • Water Planning Division (WH 554)
401 "M" St. S.W., Washington, D.C. 20460
-------
ACKNOWLEDGEMENTS
This publication contains information prepared for the U.S. Envi-
ronmental Protection Agency, Environmental Research Information
Center, Office of Research and Development; and Water Planning
Division, Office of Water Planning and Standards.
The information in this publication was prepared by Alan E.
Rimer, P.E., Wiggins-Rimer and Associates, Durham, North Carolina
with the assistance of James A. Nissen, P.E., and Roger Schecter,
AIP. Frank H. Chamberlain, III and David E. Reynolds of Triangle
J Council of Governments also assisted in preparing information.
NOTICE
This publication has been reviewed by the Environmental Research Information Center and the
Water Planning Division, Office of Water Planning and Standards, U.S. Environmental Protection
Agency, and is approved for publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection Agency, nor does any mention
of trade names or commercial products constitute endorsement or recommendation for use.
11
-------
NONPOINT SOURCE ANALYSIS
PROGRAM OF TRIANGLE J COUNCIL OF GOVERNMENTS
PROJECT SUMMARY
The Triangle J Council of Governments
(TJCOG) initiated work on its water quality
management planning program in May of
1974. The 1,750-square mile (4,530 km2)
planning area is located in the central Piedmont
section of North Carolina. A total of 22 local
government units participated in the water
quality management planning process. Sixty-
eight percent of the region's population is con-
tained in a three-city urban core and the sur-
rounding area is predominantly rural.
One of the major elements in TJCOG's area-
wide water quality management program was
an in-depth study of nonpoint source pollution
in the region. The approach used was a
comprehensive pollution source analysis which
was designed to assess existing and projected
water quality and to analyze the source, dura-
tion, magnitude, and extent of nonpoint
sources specific to the planning area.
An extensive water quality sampling and
monitoring program was conducted over a 12-
month period. Automatic sampling was con-
ducted under storm event conditions on seven
watersheds, each with a predominant land use,
to determine pollutant loading rates specific to
each land use type. Larger watersheds with
many land use types were also sampled. With
the data gathered in the sampling and mon-
itoring program and the Storm Water Manage-
ment Model (SWMM), pollutant loading rates
for each of four predominant land use types
were determined. To assess in-stream impacts
of nonpoint source pollution, loading rates and
stream hydrographs for 78 catchments were
given as input to the selected receiving stream
model - RECEIV II. The models were run un-
der existing and projected development pat-
terns and the modeled parameters included
BOD, suspended solids, total nitrogen, and
phosphorus.
The results of the model runs were compared
to 1983 water quality goals developed by the
TJCOG staff for particular pollutants. Specific
nonpoint source pollution problems were doc-
umented for suspended solids, phosphorus,
dissolved oxygen, and lead. The nonpoint
source management program was developed to
reduce these pollutant levels through imme-
diate control measures for suspended solids.
The cost for developing and carrying out
this extensive sampling, monitoring, and mod-
eling effort was approximately $400,000. In
addition to establishing an extensive nonpoint
source data base, stormwater runoff and re-
ceiving stream models were developed and
calibrated specifically for the planning area.
These models are now operational and are
serving as a continuing planning and evaluation
tool for TJCOG. This comprehensive nonpoint
source assessment study has produced signifi-
cant input to TJCOG's areawide program. In
addition, the experiences encountered and con-
clusions reached will aid in guiding other agen-
cies responsible for water quality management
planning.
OVERVIEW OF AREA
TJCOG is the state-designated regional
planning agency for North Carolina's Region J
located in the central Piedmont section. The
Council is governed by locally appointed dele-
gates who are elected officials of member
cities and counties. The primary objectives of
TJCOG are to prepare regional plans and stud-
ies, serve as the project review and comment
agency (A-95), provide a forum for discussion
-------
of regional issues, and assist Council member
governments in various aspects of planning.
In 1973, the Governor designated TJCOG as
the lead agency to undertake water quality
management planning in the region. A total of
22 local governments within the designated
area adopted concurrent resolutions "to
develop and implement a coordinated water
quality management plan for the region." The
Triangle J region and the smaller water quality
planning area are shown in Figure 1. The study
zone is within the drainage area of the Neuse
and Cape Fear River basins. Two multi-purpose
Corps of Engineer reservoirs are proposed for
the area: the B. Everett Jordan and the Falls
of the Neuse, with surface water areas of
12,490 acres (5,050 ha) and 9,400 acres (3,800
ha) respectively.
Three counties (Orange, Durham, Wake)
and portions of two others (Chatham and
Johnston) make up the 1,750-square mile
(4,530 km2) study area. Seventeen municipal-
ities, including the urban triangle formed by
the cities of Raleigh, Durham, and Chapel Hill,
are within the study area. With the exception
of this urban core, the region is primarily rural
and is characterized by small towns and agricul-
tural activity. Population of the planning area
in 1970 was 428,000, and 68 percent of this
population lived in the three-city urban area.
Raleigh is the state capital and although the
governmental sector is the largest employer,
manufacturing is a close second. The Research
Triangle Park, a nationally recognized center
for industrial and governmental research facil-
ities, is located between the three cities of the
urban core.
THE NONPOINT SOURCE
ASSESSMENT PROGRAM
One of the major focal points in Triangle J's
water quality management program was an in-
depth study of nonpoint source pollution in
the region. This comprehensive effort in pollu-
tion source analysis was the first such effort
conducted under the provisions of Section 208.
It was designed to assess existing and projected
water quality and to analyze the source,
magnitude, and extent of pollution specific to
the planning area. In addition to providing
direct input to the water quality planning pro-
gram, the approach also served as a demonstra-
tion effort to determine the feasibility of
characterizing nonpoint source pollution as
it relates to land use through extensive sam-
pling, monitoring, and modeling. Through this
analysis, nonpoint source pollution was ana-
lyzed and computer models were modified to
serve as water quality planning and evaluation
tools for the region. The major objectives of
the assessment program were as follows:
• Conduct sampling of streams under
storm event conditions and utilize data
to develop estimates of nonpoint source
pollution loads for selected, representa-
tive urban and nonurban watersheds.
• Monitor specific stream reaches con-
tinuously in order to evaluate the impact
of pollutant loads on water quality.
• Select appropriate computer models to
predict nonpoint source runoff and its
impact on the major receiving streams in
the planning area.
• Collect watershed data (including land
use, soil type, and slope and stream
channel data) for use in the selected
computer models.
• Calibrate (and verify with locally col-
lected data) the computer models for
nonpoint source runoff and receiving
stream response for the area.
• Assess the probable impact of pollutant
loads on water quality in the proposed
reservoirs.
Water Quality Sampling and Monitoring
Program
An extensive sampling program was con-
ducted over a 12-month period to determine
the nature and extent of nonpoint source
pollution specific to the planning area. The
water quality data for receiving streams was
obtained by using six continuous monitoring
stations. These stations were established on
major streams and located at critical low-flow
dissolved oxygen sag points, arid at points
considered critical under stormwater flow
conditions. Continuous readings were pro-
-------
-a
ti
§
CM
-o
c
c
o
'01
QJ
IT
_
en
c
to
CD
l_
3
01
IDOO
-------
vided for dissolved oxygen, temperature, pH,
specific conductivity, and stream flow.
It was of particular interest in Triangle
J's assessment program to establish relation-
ships between land use, pollutant loading
rates, and resultant stormwater runoff char-
acteristics. Samples were taken in watersheds
with a predominant land use by automatic
sampling units. Samples were analyzed for
BODs, COD, suspended solids, total Kjeldahl
nitrogen, nitrate nitrogen, total phosphorus,
total organic carbon, and in some instances,
heavy metals. Generation of this input data
was critical to the application and calibration
of computer models which were used in pre-
dicting pollutant loading rates and receiving
stream water quality.
Eleven sites were selected for the automatic
sampling program based on predominant land
use typical to the study area and other cri-
teria. Density, type of development, and de-
gree of activity (primarily related to traffic)
in the watershed were considered in the
selection process. Analysis of maps and data
generated from LANDSAT satellite imagery,
supplemented with aerial photography, pro-
vided ground information necessary for de-
termining appropriate locations for the auto-
matic sampling units. The seven land use
types sampled were as follows: low activity
rural; high activity rural; low activity residen-
tial; high activity residential; low activity
commercial; high activity commercial; and
urban (central business district).
In addition to these seven watersheds with
predominant land use types, four total-load
stations were established to determine runoff
effects of larger drainage areas with multiple
land use types. Drainage areas ranged from
120 to 49,000 acres (50 to 20,000 ha) for pre-
dominant land use stations and from 36,000
to 730,000 acres (14,500 to 300,000 ha) for
total-load stations. When a storm event oc-
curred, the samplers were automatically acti-
vated and samples were taken at prescribed
intervals throughout the storm event. Depend-
ing on location of the station, between five
and eleven storm events were sampled over the
12-month period. Locations of the sampling
and monitoring stations and the rainfall
gauging stations are shown in Figure 2.
Related Nonpoint Source Studies
Concurrent with the water quality sam-
pling and monitoring program, several other
potential nonpoint source pollution problems
were studied. For example, a study was made
of runoff from a parking lot to determine the
potential impact of a large impervious surface
on water quality in a nearby stream. Also, a
study was undertaken to determine the pos-
sible impact of storm flows that could resus-
pend sludge deposits below wastewater treat-
ment facilities and cause increased oxygen de-
mand in the stream. Because of the signifi-
cance of the proposed Corps of Engineers
multi-purpose reservoirs, analyses were also
conducted of nutrient loading (particularly
phosphorus) from point and nonpoint sources
to assess the potential for eutrophication.
NONPOINT SOURCE
ASSESSMENT RESULTS
Determination of Loading Rates by Land Use
Type
Triangle J utilized the SWMM, modified for
specific application in the study area, as a
means of estimating runoff quantity and qual-
ity. Through a series of sensitivity analyses,
model calibration tests, and verification checks
on model runs using collected data, pollutant
loading rates for land use types were estab-
lished. It was concluded that nonpoint
source pollution potential was closely related
to density of development. On the basis of
data collected in the field during the sampling
program and the loading rates generated by
SWMM, it was determined that four primary
land use categories most accurately reflected
nonpoint source pollution potential. These
categories are as follows:
• Urban — predominantly Central Business
District (CBD).
• Commercial - predominantly high and
medium density commercial and indus-
trial development other than CBD.
• Residential — predominantly single and
multi-family residential areas.
• Rural - predominantly agricultural and
forest areas with associated rural
development.
-------
o
C
to
D)
C
"5.
E
di
c
c
o
E
o
^
03
O
O
CM
(D
-------
These land use categories and their respec-
tive pollutant loading rates are shown in Table
1. SWMM provided watershed (catchment)
hydrographs and pollutographs (BOD, sus-
pended solids, nitrogen, and phosphorus load-
ings over time) which were given as input to
the selected receiving stream model, RECEIV
II. Seven major receiving streams were mod-
eled by RECEIV II. With SWMM and
RECEIV II, the data collected during the
sampling and monitoring effort could be
utilized for catchments throughout the plan-
ning area, since the models were calibrated
for each of the predominant land use types.
Based on SWMM predictions, it was deter-
mined that no single land use was responsible
for generating the highest pollutant loading
rates for all constituents studied. In Table 1,
for example, it can be seen that urban land
use generated the highest BOD loading rate
but the lowest suspended solids loading rate.
The data developed relates to the build-up of
pollutants on the land and the wash-off
potential to a receiving stream under a partic-
ular storm event. Other factors such as the
extent of impervious surface areas, total area
of a particular land use, hydrological charac-
teristics of the catchment, and contribution of
upstream catchments were also considered in
assessing the wash-off rates and impacts of
pollutants on receiving streams.
Use of Models to Predict Receiving Stream
Impacts
To predict water quality throughout the
study area, the region was subdivided into a
number of individual catchments, each of
which drained to a node point on a receiving
stream. SWMM was then used to generate
runoff data and pollutant loadings from
these catchments. Input data on land use
type and total area, loading rates for the land
use, soils, topography, and various other
physical characteristics were developed for
each of 78 catchments. For the selected de-
sign storm having a recurrence rate of about
one year, runoff hydrographs, pollutographs,
and average pollutant concentrations were
calculated as input to the receiving stream
nodes. Figure 3 shows the catchments
modeled and the location of the nodes for
the SWMM runs.
SWMM was run with input data for exist-
ing and future land use patterns to determine
differences in loading rates and pollutant con-
centrations under alternative growth patterns.
Table 2 shows output data for selected nodes
on the Neuse and New Hope river systems
for existing and future conditions. In almost
all cases, loading rates and pollutant concen-
trations increased between existing and future
land use conditions.
SWMM relates only to the characteristics of
stormwater runoff from various land uses and
its associated pollutant potential. To predict
the impact of stormwater runoff on receiving
streams, RECEIV II was utilized. SWMM was
an input to RECEIV II. Data from the sam-
pling and monitoring program were used to
test and calibrate the outputs of RECEIV II.
To assess the relative significance of each
parameter on water quality, outputs of
RECEIV II were compared with the 1983
Table 1. — Pollutant loading rates based on SWMM model predictions
Land use
Pollutant loading rates (Ibs/acre/day) = (kg/ha/day)
Urban (CBD)
Commercial
Residential
Rural
BOD5
0.42
0.29
0.17
0.12
Suspended Solids
11.6
21.5
18.5
15.0
Total Nitrogen
.027
.026
.016
.008
Total Phosphorus
.008
.010
.004
.003
-------
r )?
LU
u
LU
DC
-o
c
CD
OJ
-a
o
-------
1
^;
4!
c
I
^*"
S
£
patterns
^t
§
5
&
S
§
£
a
J3
H^
-Q
to
0)
.c
.">
i
v^
«
§
g
a
Ul
.§>
8
•o
^
CO
!
(U
C
0
V
-J\
a
1
><
8
Q.
Q.
<0
1
£
H^,
-8
0
1
5
Q. -X
52
o —
h-
•*-<
c:
0)
EQJ
-o
•£ O
U 7
<-• ^
10
O
&m~
o" °
^
~ x
z ^
11
!5 -~
"^ CO~
o A
w o
"2 x
03
? 5
a-^
u> £
CO ~"
•€
«"
_ X
& 0)
% •?
03 1?
£
+<
C
0)
E
-g
•M
Q
0)
3
3
U.
0)
C
'&
vt
'x
01
i_
a;
^>
DC
h»
4-*
3
a.
o>
1
'x
LU
0)
3
+•<
3
U-
01
«
'x
LU
0}
>
oE
0}
(/}
D
CD
Z
(O
CO
o
00
LO
CN
S
LO
•*
O
o
o
in
8
CO
40
2300
CN
CO
P-.
gj
CN
CN
CO
CN
00
CO
CO
CO
CO
CO
O3
CO
00
CO
CO
CN
CN
LO
O
CO
740/340
CO
CO
LO
LO
CN
oo
00
0)
CN
S
(U
a
o
I
s
CO
o
q
io
o
S
o
o
^~
co
o
o
CN
400/1090
CO
00
IO
LO
CO
00
o
I
o
820
80
CO
O)
CN
CO
CO
CN
(U
•—
<£\
CO
LO
00
CO
00
CO
LO
00
o
^~
LO
o
o
00/950
CN
CN
8
-------
water quality goals shown in Table 3. By com-
paring SWMM and RECEIV II output with
these goals, the magnitude of nonpoint source
loadings was assessed. It was concluded that
several 1983 goals would not be met under
existing land uses because of the upstream
oxygen demand and phosphorus loadings.
With these broad conclusions and with
specific catchment data, TJCOG developed a
control program which reflected the need to
reduce suspended solids from all land uses.
As a means of developing the program TJCOG
sought input through: literature review and
technical advisory committee discussions, a
series of public workshops, and analysis of the
existing institutional capabilities.
A relationship was observed between in-
creases in suspended solids and phosphorus.
Lead was projected as a potential problem.
No temperature, nitrate, mercury, or dis-
solved solids problems were projected.
Future land development conditions were
modeled. Some differences were seen in the
relative contributions of individual catchments.
Overall differences in pollutant concentrations
between existing and future conditions did
not appear startling except for a few catch-
ments. Increases in average sediment con-
centrations ranged from 10 to 50 percent.
The results of other aspects of Triangle J's
nonpoint source assessment program pro-
vided additional information on water quality
problems. For example, the analysis of
benthic material indicates that its resuspen-
sion during storm events is insignificant com-
pared to the loading from stormwater runoff.
Nonpoint source phosphorus loadings to the
proposed reservoirs are predicted to effect
early eutrophication. Stormwater runoff was
projected to contribute approximately 65
percent of the calculated annual phosphorus
load.
Identification and Applicability of Best
Management Practices (BMP's) for the Region
A wide range of best management practices
was identified through a literature survey and
in conjunction with technical advisory com-
mittee input. An annotated list of manage-
ment practices was prepared by TJCOG and
each practice was assessed on the basis of
effectiveness in reducing nonpoint source loads
which had been documented as problems by
Table 3. — Triangle J 1983 water quality goals
Parameter
Planning goal
Dissolved Oxygen . . .
Suspended Solids. . . .
Total Phosphorus as P
5.0mg/l
80 mg/l
Temperature
pH
Nitrate-Nitrogen as N .
Dissolved Solids . . . .
Mercury
Lead
1.0 mg/l in free flowing streams
0.5 mg/l in streams above reservoirs
0.1 mg/l in reservoirs
Always less than 84°F with no change 5°F above natural conditions
6.0 to 9.0
10 mg/l
250 mg/l
0.002 mg/l
0.05 mg/l
-------
the sampling and modeling effort. Each man-
agement practice was analyzed in terms of its
utility in reducing stormwater runoff and sus-
pended solids levels, as well as other consider-
ations such as cost and the effectiveness of the
control practice. Table 4 lists the BMP's which
were considered applicable in the study area.
The applicable practices were incorporated
in the nonpoint source control strategy as a
major element of the water quality manage-
ment plan. Of the available techniques, source
controls and surface transport controls were
assessed to be more effective and less costly
than collection and treatment techniques. The
modeling results indicated that the severity of
nonpoint source problems did not warrant the
expense of collecting and treating stormwater
runoff.
In conjunction with developing the detailed
list of BMP's, a series of eight workshops was
conducted throughout the region. These
workshops were co-sponsored by local govern-
ments, civic environmental groups, citizens,
and special interest groups such as realtors,
agri-business representatives, Soil and Water
Conservation District representatives, and
others. At each workshop, Triangle J staff pre-
sented findings of the pollution source analysis
with particular emphasis on how water quality
problems affected the interests of those at-
tending the particular workshop. Workshop
participants were asked to draw on their own
perception of water quality problems and
suggest solutions to these problems. After
suggestions were tallied, the group was asked
to rate the effectiveness of each suggested con-
trol measure, taking into account the feasi-
bility of implementation. Although many sug-
gestions were broad in scope, the control mea-
sures identified during these workshops were
closely allied with and supported those which
were developed by the staff and technical com-
mittees. Input from the workshops was used as
supporting information for developing the non-
point source control strategy in the draft Water
Quality Management Plan.
The detailed list of BMP's and the work-
shop recommendations were reviewed with
regard to the magnitude of identified non-
point source problems and the potential for
implementing control programs. Existing
and potential management systems were
analyzed by TJCOG with particular empha-
sis on the institutional ability and legal au-
thority to implement control mechanisms.
Although the nonpoint source assessment
concluded that the control of sediment was
of primary importance, other nonpoint
sources of pollution were identified, and
actions to abate these problems were also
developed. The TJCOG nonpoint source
control program focuses on correcting prob-
lems through existing institutional and legisla-
tive authorities and strengthening those au-
thorities where possible, rather than attempt-
ing to control all potential sources.
Table 4. — Best management practices developed for nonpoint source pollution control
Source control
Land use planning
Minimization of stripped areas
Buffer zones along streams and channels
Porous pavement
Street sweeping
Grade stabilization
Seeding and mulching
Terraces and diversion ditches
Lattice blocks
Cover crops
Contour plowing and tillage practices
Surface transport control
Street and channel design
Grass-lined waterways and outlets
Channel stabilization and stream bank protection
Collection
Detention basins (short term storage)
Treatment
Gravity settling
Filtration
10
-------
NONPOINT SOURCE PROGRAM
AND IMPLEMENTATION STRATEGY
Focus of Nonpoint Source Control Program
Through the process of developing manage-
ment practices that are applicable to the plan-
ning area, TJCOG involved groups and agencies
which would have an impact on implement-
ing the proposed nonpoint source program.
By studying the conclusions of the pollution
source analysis and focusing on identified
problems which were documented and sup-
ported by modeling efforts, these groups
gained an awareness of the complexities of
nonpoint source problems. Furthermore,
TJCOG established the need to take imme-
diate corrective action on significant prob-
lems as a first step in the nonpoint source
control program.
Even with extensive sampling, monitoring,
and modeling, some nonpoint source prob-
lems cannot be adequately documented.
Pesticides, for example, were not modeled be-
cause of the large number of different chem-
icals used and the cost effectiveness of pur-
suing sampling and analysis for this parameter.
Immediate controls were recommended for
implementation if they were practical and in-
expensive. Improved water quality can be ex-
pected by instituting better street sweeping,
passage of ordinances, and the employment of
construction site inspectors.
Major Actions and Implementation
Before the draft Water Quality Management
Plan was published, all levels of government
and other agencies who would be affected by
the nonpoint source control program were
given the opportunity to review the proposed
program. Since publication of the 1977 draft
Water Quality Management Plan, some of the
proposed actions have already been imple-
mented.
A prime element in TJCOG's nonpoint
source program is the control of suspended
solids. As a result of the water quality manage-
ment study, local governments have indicated
they will consider adoption of sedimentation
and erosion control ordinances. Annual pro-
gram costs range from less than $1,000 for a
small town to over $65,000 for a countywide
program.
Steps are being taken, in conjunction with
the U.S. Department of Agriculture Soil Con-
servation Service and appropriate state agen-
cies, to encourage all active farms in the area to
develop voluntary conservation management
plans to reduce soil loss to four tons per acre
per year (43,500 kg/ha/yr). Related actions
include securing county financial support and
increasing the staff to help prepare plans for
soil conservation with area farmers.
TJCOG's sedimentation and erosion control
program includes erosion control during all
phases of construction. Better land manage-
ment of construction sites is necessary to
reduce suspended solids runoff and associated
pollutants, such as phosphorus and lead.
Local governments are currently providing
funds and services to Triangle J who in turn
will provide technical assistance to home-
owners, county inspectors, etc. TJCOG will
advise on all aspects of water quality manage-
ment, especially operation and maintenance of
individual septic systems.
Some elements in Triangle J's nonpoint
source management program would not
and could not have been proposed without
extensive monitoring and modeling. This
effort provided one way of characterizing and
assessing general water quality conditions.
Because of this work, local government and
industry mostly supported the early imple-
mentation of the plan. These models are now
functioning as an ongoing planning tool.
11
-US GOVERNMENT PRINTING OFFICE 1979-657-060/1535
------- |