440/3-77-025
PLANNING METHODOLOGIES FOR
ANALYSIS OF LAND
USE/WATER QUALITY RELATIONSHIPS:
CASE STUDY APPLICATION
U.S. Environmental Protection Agency
Washington, D.C. 20460
NOVEMBER 1977
-------
EPA REVIEW NOTICE
This report has been reviewed by the Environmental Pro-
tection Agency and approved as satisfying the terms of
the subject contract. Approval does not signify that
the contents necessarily reflect the views and policies
of the Environmental Protection Agency, nor does mention
of trademarks or commercial products constitute endorse-
ment or recommendation for use.
-------
TED STATES ENVIRONMENTAL PROTECTION AGENCY
DATE:
SUBJECT:
FROM.
TO:
NOV 8
Wissahickon Case Study
00 JUG,
Walter Groszyk/Dteputyl Director
Water Planning/Division (WH-554)
Regional Water Division Directors
208 Coordinators
Information Memorandum: INFO-78-5
This report by Betz Environmental Engineers presents an application
of a planning methodology for the analysis of land use-water quality
relationships. It is a follow-up to a previous Betz study entitled
"Planning Methodology for Analysis of Land/Use Water Quality Relation-
ships," Water Planning Division, EPA, 1976. The emphasis of this
report is on urban and urbanizing areas, and in particular, the study
emphasizes possible effects from new development. In order of presen-
tation the report outlines some pertinent land use and receiving water
quality relationships, highlights typical problems, and provides a
detailed description of the methodology.
The methodology is applied to the Wissahickon watershed, a "typical"
urban watershed. Technically the report attempts to estimate pollutant
runoff by land use, allows calibration to reflect local conditions, and
determines tnose BMP's which should be most effective in reducing runoff.
This methodology should require less hydrologic and storm data than
dynamic storm modeling. It uses a simple mass balance calculation based
upon the universal soil loss equation. This equation has been found to
have broad applicability but its usefulness for predicting erosion-
related components such as toxic transport is under intensive review.
The methodology does provide a practical way of estimating sediment
loading and does relate field survey observations to the mathematical
model. It does not provide adjustments for hydrologic modifications, or
directly assess water quality impacts. If used for priority ranking of
major sediment contributors the method should be useful for comparing
relative differences between BMPs. It does not provide accurate predictions
for soluble constituents or work well for slow snow-melt or slow rainfall-
events.
A structure like Betz's is needed to begin sorting out NFS problems.
Other methods are of course available such as found in the "Areawide
Assessment Procedures Manual, EPA, 1976" which does offers a much broader
discussion on NPS problem assessment. At this time, however, the state-
EPA For,,, 1320-6 (Rev 3 76)
-------
of-the-art of NPS modeling does not offer any one best method that can
closely predict absolute differences between control strategies. Long
term data should, ideally, be available for a wide range of actual
runoff events. This is seldom the case. Methods like these that permit
an orderly process of estimating the effects of runoff are a practical
necessity.
Please disseminate this handbook to only those agencies who you believe
can make wise use of it. Extra copies may be requested from the Water
Planning Division Library (WH-554), Washington, D. C,, 20460 (FTS 755-6993)
or through Bill Cogger (FTS 426-2522) of this office.
Attachment
-------
PLANNING METHODOLOGIES FOR ANALYSIS OF
LAND USE/WATER QUALITY RELATIONSHIPS:
CASE STUDY APPLICATION
prepared under
EPA Contract No. 68-01-3551
for the
U.S. ENVIRONMENTAL PROTECTION AGENCY
Water Planning Division
PROJECT OFFICERS: WILLIAM J. COGGER
WILLIAM C. LIENESCH
By
BETZ ENVIRONMENTAL ENGINEERS, INC.
PLYMOUTH MEETING, PENNSYLVANIA
PROJECT DIRECTOR: WILLIAM K. DAVIS
PROJECT MANAGER: FRANK X. BROWNE
-------
CONTENTS
I INTRODUCTION
II LAND
A.
B.
C.
USE/WATER QUALITY ANALYSIS PRINCIPLES
Existing Methodologies
Problem of Nonpoint Source Load
Assessment
Overview of Methodology
Ill ANALYTICAL METHODOLOGY
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.
K.
L.
IV CASE
A.
B.
C.
D.
E.
F.
Introduction
Inventory and Evaluation of Existing
Data
Field Surveys and Nonpoint Source
Monitoring
Water Quality and Pollutant Loading
Analysis
Selection of Subareas and Land Use
Categories
Determination of Raw Loading Rate
Determination of Raw Loading Rate
Modifiers
Land Management Controls
Mass Balance Analysis
Water Quality Impact Analysis
Sensitivity Analysis
Control Plan Development
STUDY: APPLICATION OF METHODOLOGY
Introduction
Inventory and Evaluation of Existing
Data
Field Surveys and Nonpoint Source
Monitoring
Water Quality and Pollutant Loading
Analysis
Selection of Subareas and Land Use
Categories
Determination of Raw Loading Rate
Page
1
3
3
5
7
9
9
11
15
17
18
22
26
35
48
55
60
63
68
68
69
72
88
97
107
n
-------
CONTENTS
(Continued)
Page
G. Determination of Raw Loading Rate
Modifiers 111
H. Land Management Controls 128
I. Mass Balance Analysis 137
J. Water Quality Impact Analysis 140
K. Sensitivity Analysis 148
L. Control Plan Development 150
V EVALUATION OF METHODOLOGY 159
A. Advantages and Disadvantages of
Methodology 159
B. Problems in Application of
Methodology 161
GLOSSARY 164
BIBLIOGRAPHY 169
APPENDICES: Available On Request
-------
TABLES
Number Page
1 Summary of Data Needs for the General
Methodology Tasks 12
2 Physical Control Practices for Con-
struction Activities 37
3 Regulatory Controls: Pollutant Load
Reduction Factor Points 42
4 Range of Land Management Controls in
Developed Areas 44
5 Loading Factor Sensitivity - Mass
Balance Points 62
6 Wissahickon Watershed Case Study Data
Sources, Description and Function 71
7 Summary of First Wissahickon Creek Survey 74
8 Existing Point Source Load and Concentra-
tion Summary, Wissahickon Creek, Fort
Washington 89
9 Summary of Storm Loading - Wissahickon
Creek, Fort Washington, 1968 91
10 Annual Sediment Loading, Rainfall eind
Stream Flow, 1964 and 1968 98
11 Definitions of Land Uses 103
12 Wissahickon Creek Case Study - Land
Use by Subarea 105
13 Land Use Projections - Wissahickon
Watershed 106
14 Rainfall Factor (R) for Wissahickon
Watershed 112
IV
-------
TABLES
(Continued)
Number Page
15 Slope Length Factors - Wissahickon
Watershed 11 6
16 Drainage Densities - Wissahickon Watershed 118
17 Sediment Delivery Ratio - Wissahickon
Watershed 121
18 Natural Control Factors (P) for Wissa-
hickon Watershed 124
19 Pollutant Load from Various Land Uses 125
20 Pollutant Loading Analysis for
Wissahickon Watershed 126
21 Storm Decay Rates of Pollutants
(K3) - Wissahickon Watershed 127
22 Municipal Regulatory Controls for
New Land Development - Wissahickon Creek 130
23 Conversion of Municipal Reduction Factor
for New Land Development to Subarea
Reduction Factor 133
24 Existence and Development of Mainten-
ance Controls 133
25 Maintenance Control Program Scores by
Land Use Category 134
26 Load Reduction Due to Maintenance Controls 134
27 Composite Maintenance Control Reduction
Factor by Land Use Category 135
28 Existing Land Management Control Rates
(K2) - Wissahickon Watershed 136
v
-------
TABLES
(Continued)
Number Page
29 Calibration Loads Summary - Wissahickon
Watershed 138
30 Verification Results - Wissahickon
Watershed 139
31 Water Quality Data Summary 142
32 Sensitivity Analysis - % of Load Control 149
33 Sensitivity Analysis - Sediment -
Subarea M3 150
34 Sensitivity, Land Use Controls,
Sediment - Subarea M3 151
35 K2 Rates 154
36 Stream Load by Subarea 157
37 Sediment Load - Subarea M3 158
VI
-------
FIGURES
Number Page
1 Schematic of General Methodology 10
2 Wissahickon Watershed Subarea
Segmentation 19
3 Slope Length Determination Criteria 29
4 EPA Land Use Load Determination Flow
Diagram 51
5 Map of Study Area 70
6 Map of Survey Sites 73
7 Wissahickon Watershed - Sediment Versus
Storm Flow 92
8 Wissahickon Watershed - Sediment Versus
Monthly Flow 93
9 Wissahickon Watershed - Sediment Versus
Storm Rainfall 94
10 Wissahickon Watershed - Sediment Versus
Monthly Rainfall 95
11 Wissahickon Watershed - Sediment Versus
Storm Flow 96
12 Wissahickon Watershed Subarea
Segementation 99
13 Land Use Map for Wissahickon Watershed,
Subarea M3 101
14 Critical Slopes Map for the Wissahickon
Watershed 113
15 Slope Length Determination Criteria 114
-------
FIGURES
(Continued)
Number Page
16 Hydrologic Runoff Potential Map for
Wissahickon Watershed 117
17 Criteria for Determination of
Sediment Delivery Ratio 120
18 Map of Biological Sampling Sites -
Wissahickon Watershed 143
19 Map of Point Source Dischargers -
Wissahickon Watershed 144
vm
-------
I
INTRODUCTION
The purpose of this report is to present a simplified planning
methodology for 208 water quality management agencies. Existing
methodologies were adapted in the development of the present methodology.
The application of the methodology will allow 208 agencies to:
Estimate general values of pollutant runoff coefficients by land
use and type of land
Determine how to calibrate values of pollutant runoff co-
efficients to reflect local conditions, including climate, soil,
slope, nature of water bodies, and ground cover where such
conditions affect water quality
Determine those land use and land management controls which
are most effective in reducing pollutant generation and dis-
charge in the specific 208 area
The emphasis of this report is on urban and urbanizing areas with stress
on single family residential, multi-family, residential, and commercial
land uses. In particular, the study emphasizes residuals discharged
from new development.
Section II outlines some pertinent land use and receiving water quality
relationships and typical problems. Section III provides a detailed
description of the methodology developed to analyze these relationships.
The methodology is applied in Section IV to a case study for a typical
urbanizing watershed. Finally, Section V presents an evaluation of
the methodology, a summary of typical problems encountered in
applying the methodology, and recommendations for application.
A glossary of terms is provided for commonly used jargon of water
quality planners, and for the shorthand abbreviations of the simple
formulas. Supportive date and actual calculations are assembled in
a detached summary. These data can be provided by contacting the EPA
Water Planning Division (WH-554) library in Washington, DC, 20460.
-------
These sections are followed by a glossary of terms used
throughout the report and appendices which include support-
ing data and analysis and sample calculations.
-------
II
LAND USE/WATER QUALITY ANALYSIS
A. EXISTING METHODOLOGIES
There are numerous evolving methodologies to use to evaluate
nonpoint pollution and formulate best management practices
for inclusion in water quality management plans. These
methodologies can be grouped into four major categories:
(1) stormwater modeling, (2) environmental synthesis, (3)
statistical relationships, and (4) others. A brief descrip-
tion of each methodology, including a summary of salient
characteristics and advantages and disadvantages, follows.*
1. Stormwater Modeling
There are essentially two types of stormwater modeling: (1)
Stormwater Management Model (SWMM) developed under the aus-
pices of EPA; and (2) the Storage-Treatment-Overflow-Runoff
Model (STORM), developed by Water Resources Engineers and
later expanded by the U.S. Army Corps of Engineers. Both
are dynamic models which include the hydraulics of storm-
water. SWMM includes the response of receiving water
quality as well as the hydraulics and transport of pollutant
loads. Both would encompass pollutant yields of dissolved
material, suspended material and settleable material, and
both use the universal soil loss equation to simulate pollu-
tant scour and erosion from pervious land cover.
These models present an adequate approach towards solving
water quality management problems. The dynamic nature of
the models allows simulation of storm response, thus pro-
viding criteria for exploring hydraulics, and structural and
non-structural best management practices alternatives. SWMM
has the added advantage of determining recipient water qual-
ity. Both models have a disadvantage: large amounts of data
are required and the simulation of hydraulics is complex.
Simplifications and aggregation to compensate for lack of
* More detailed descriptions of these methodologies can be
found in references 13, 14, 16, 19, 23, 26 and 37 of the
bibliography.
-------
detailed data may result in erroneous results that are no better or
possibly worse than simpler approaches. Applying this methodology is
costly and out of reach of many 208 agencies.
2. Environmental Synthesis
The common characteristic of this group of methodologies is that they
attempt to synthesize multiple environmental objectives to develop
an overall land use and public facility plan. Large regions are sub-
divided into homogenous units based on biologic capability, suitability
and feasibility. The methodology uses overlays of each of these
criterion to highlight problem areas or inconsistencies in capabilities
and objectives.
The methodology is usually less costly than others in its application
but is wanting in answers normally required by planners who are de-
veloping water quality management alternatives. The resulting outputs
are usually insufficient to quantitatively assess the magnitude of
nonpoint loads and the degree of abatement required to meet water
quality goals.
3. Statistical Relationships
The objective of this methodology is to develop statistically sound
relationships between land use and pollutant loads and between
pollutant loads and the quality of receiving waters and hydrologic
variables. It attempts to identify land uses responsible for various
pollutants, thus providing a criteria for quantitative assessment of
pollutant loads and their subsequent control.
This approach is economical and should be sound, but there is no
assurance of success because apparently valid statistical relationships
may not stand up to empirical evidence or data with a low statistical
correlation may hc.ve a practical connection. Pollutants eminate from
a variety of land-soil uses, thus statistically sound criteria may require
a large sample size or highly sophisticated statistics. This may pose
a problem in expense and outstrip practical comparisons and inferential
testing of statistical conclusions. Statistical sorting and con-
solidation may however be the only way of dealing with these complex
conditions. Another problem with statistical analysis is that pollutant
loads from similar land uses vary, depending on slopes, soils and
proximity to stream. Also, the complexities of the hydraulics of one
storm compared to other storms of different intensity or duration presents
obstacles in statistically correlating data. The statistical analysis
can take apparent wide variability into account, but practically it is
limited in its applicability to reasonably similar sets of land use
conditions (land uses with common slopes, soils, stream proximity, etc.).
-------
B. PROBLEMS OF NONPOINT SOURCE LOAD ASSESSMENT
Determining nonpoint pollution problem areas and wasteload allocations
for a 208 area is a complex task. Rarely is the existing data base
adequate to permit a thorough analysis. The manpower and time required
to augment and provide a complete data base are usually beyond the budget
of the 208 planning agency. The agency is usually forced to compromise;
exchanging comprehensive analyses for a limited effort. The following
factors are worthy of consideration when addressing this problem.
1. The presence of nonpoint source loads in streams is caused by runoff from
storms. Samples of receiving stream water should be taken during and
immediately following a storm at strategic locations in the stream
where the presence of nonpoint loads is suspected. The sampling should
be thorough enough to allow for the construction of hydrographs and
pollutographs of each pollutant considered and to relate them to total
storm rainfall. Extrapolation of periodic samples (grab samples) of
water quality data to storm loads should be avoided if possible. The
Wissahickon Watershed Case Study presented in Section IV indicated that
better than 95% of the total annual sediment load occurred within one
or two days following significant storms. In some streams, 95 percent
of the load occurred during less than 6 percent of the total days in the
year. The practice of periodic random sampling increases the chances
that significant storm loadings would be missed; and results extrapolated
from this sampling could be misleading.
2. Nonpoint pollution runoff from various land uses depends not only on
the land use and severity of the accumulated load on the land but
on a complex relationship of other factors such as slopes, soils,
drainage density and degrees of natural and man-made controls. Actual
runoff loads from typical land uses and terrain within the 208 study
area should be measured in streams during and following storms, to
verify desk top projections. If this is beyond the scope of the study,
runoff measurements from similar areas may be used, but care should
be taken that such factors as land use, terrain, soils and climate are
similar. Unverified runoff projections should be used very carefully
for planning purposes or sparingly if qualitative data, for example,
Fish census or on site examination, suggests these projections are in
error.
3. Visual observations of land use and stream conditions before and after
storms are extremely helpful in locating and verifying nonpoint problems
and sources. However, these observations yield only qualitative analysis
-------
of problem areas and need some fixed framework (model or mass balance)
before the observations can be effectively utilized for wasteload allocations.
-------
C. OVERVIEW OF METHODOLOGY
The goals of this study can be summarized in the following
scenario:
Imagine a small to medium sized metropolitan
area in which substantial land development and
suburban growth is expected to take place in the
future. Water quality is known to be affected
by point sources; but the relative importance of
nonpoint sources is uncertain. The 208 funds
available for NFS/IPS (nonpoint source/intermit-
tent point source) analysis will probably not
permit calibration and operation of non-steady-
state receiving water models for all streams
affected by urbanization, or permit a broadly-
based program of storm water modeling. Our task
is to tell them how they might proceed in assess-
ing NFS/IPS problems, setting priorities, evalu-
ating control measures, and developing a cost-ef-
fective control plan which is coordinated with
other planning activities in the area.
Pursuant to these goals, various techniques and methodol-
ogies were examined. The methodology developed is a mix-
ture of what was deemed the best qualities of the method-
ologies previously presented and one which would recognize
and circumvent the highlighted problem areas. Emphasis was
placed on using techniques requiring talents and capabilities
normally found in 208 agencies. Complex hydrologic modeling
algorithms were avoided.
The methodology provides a relatively simplified approach
for the analysis of land use/water quality relationships.
The primary components of the methodology include field
investigations, stormwater runoff monitoring, application
of a simplified form of the universal soil loss equation to
calculate nonpoint loads, mass balance analysis using pol-
lutant generation and in-stream water quality analysis,
and ranking of best management practices.
This mass balance model provides the framework to identify
land uses, terrain and soil conditions, and controls. The
use of a model makes it possible to modify loading factors
-------
based on visual observations. The total mass balance model
is calibrated to conform to the observed stream loads. The
model provides a simple base for exploring promising land
management alternatives and nonpoint wasteload allocations
and their impact on receiving water quality. The model does
not provide a framework for exploring alternative hydrologic
routings or alternatives other than a gross analysis of
their total impact to receiving waters.
The details of the methodology are presented in Section III/
including all applicable criteria. The methodology was ap-
plied to a typical case study "The Wissahickon Watershed."
Details of the case study application are presented in Sec-
tion IV. An evaluation of the methodology is presented in
Chapter V.
-------
Ill
ANALYTICAL METHODOLOGY
A. INTRODUCTION
The methodology presented in this section provides a sim-
plified approach for the analysis of land use/water quality
relationships. The primary components of the methodology
include field investigations, stormwater runoff monitoring,
application of a modified universal soil loss equation to
calculate nonpoint source loads, mass balance analysis using
pollutant generation and in-stream water quality data, and
control plan development. A schematic representation of the
methodology is presented in Figure 1. The total methodology
consists of the following components:
1. Inventory and Evaluation of Existing Data
2. Field Surveys and Stormwater Runoff Monitoring
3. Water Quality and Pollutant Loading Analysis
4. Land Use and Nonpoint Source Loading Analysis
5. Mass Balance Analysis
6. Water Quality Impact Analysis
7. Sensitivity Analysis
8. Control Plan Development
The methodology incorporates three overall classes of tech-
nical activities: problem analysis, source analysis, and
identification of potential control measures. Problem anal-
ysis and source analysis are performed concurrently through-
out most of the methodology.
-------
INVENTORY AND EVALUATION
OF EXISTING DATA
(1)
FIELD SURVEYS AND
NONPOINT SOURCE
MONITORING
(2)
V
IN-STREAM
WATER QUALITY
AND POLLUTANT
LOADING ANALYSIS
(3)
SELECTION
SUBAREA
OF
SELECTION OF LAND
USE CATEGORIES
u
c
0)
c
Ol
c
(D
JJ
c
a)
c
O
a)
CO
MASS BALANCE
ANALYSIS
(5)
WATER QUALITY
IMPACT ANALYSIS
(6)
SENSITIVITY
ANALYSIS
(7)
CONTROL PLAN
DEVELOPMENT
(8)
3
O
Note: Arrows indicate major flow paths of analyses
FIGURE 1
SCHEMATIC OF GENERAL METHODOLOGY
10
-------
B. INVENTORY AND EVALUATION OF EXISTING DATA
Primary and secondary sources of data should be collected
and evaluated. The type of data to be collected, how the
data are to be used, and sources of data are summarized in
Table 1. Specific data requirements are summarized below.
1 f Water Quality and Biological Data
Long-term water quality records and storm-related water
quality data should be obtained. Long term water quality
data indicate seasonal and historical trends associated,
in general, with steady-state conditions. Storm data, if
available, will indicate the existence of transient water
quality problems associated with stormwater runoff. Com-
parison of long-term and storm-related water quality data
indicates the relative impact of nonpoint sources. Existing
water quality should be compared to water quality standards
and the violations noted.
An important point to consider is that the investigation of
problem areas should include more than chemical analysis of
water quality. Existing biological data should be evaluated
along with water quality data. Biological communities are
not as transient as water quality and, as such, are excel-
lent indicators of problem areas. This is particularly
important because most past environmental studies were
performed during dry weather conditions when the impact of
nonpoint sources were not readily apparent from water qual-
ity data. However, the sessile of fixed nature of many
aquatic organisms provides a measure of the long-term impact
of nonpoint sources. This is especially true of benthic
organisms such as macroinvertebrates . Thus, biological data
on algae, fish, and benthic macroinvertebrates should be
evaluated to identify existing or potential problem areas.
The evaluation of water quality and biological data is the
first step in problem analysis.* It is a preliminary anal-
ysis because, in most cases, the existing data are not
* Procedures and criteria for assessing water quality im-
pacts and resulting impacts to biota can be found in the
EPA publication, "Area Wide Assessment Procedure Manual,"
Volume I, No. EPA-600/9-76-014, July 1976, Chapter II,
Data Base Inventory and Problem Identification.
11
-------
tn
(O
*
X
Q
1 ^
§
O
33
H
g
j
*
2
z
33
K
O
fc,
C/l
Q
M
H
2
g
§
^
O
^4
K
<
1
to
H
3
O
to
18
4-1
10
"
rH
18
H
4J
C
0)
4-1
0
cu
o
o
01
3
OJ
n
£
01
H
18
4J
18
C
01
I
O
4J 0
z
4J
3 .*
a n
c id
H EH
§
3SS
3 01 H
rH W O
H 4J O
O 0) 01
a. M
to
M O
01 01 EH
ioag
*»g
01 0) 10
4J -rl-^
a u
4J C <
DI U
O «)
c
10 rH 10
O -O
rH M rH
10 4J O
0 C 0
O O 01
iJ 0 H
0)
o
c
18
rH
01 id
C .CXI
o a
rH 18 01
4-1 M 01
rH OS
rH 4-1
O 01 3 >,
O, 10 rHIH
01 rH -H
14-1 n o u
o <8 a
u e iw
01 0> O T>
Qj rH 0
^Xl C 18
4J O O
V- -H 01
>, Q.4J 4-1
IH IB 18
H o> H w
4J 4-1 (8 Xt
C <0 CL-f-l
(U O 0) rH
a o -H 10
M a a. o
^
to
m
n
o
M
o
o
01
M
01
o
H
>
rH
01
to
u
01
^J
id
S ""
3C
' rt«
tn o
z
e
a
rH
18 q
ts xi o
C -H
cfl tO 4-1
01 (0
01 10 U
£ £l *rl
O< 0)
O In >
rl O
O 4J TJ
>, u c
£ 01 IB 18
4J a c
-H IB o; o
C H -H
3 & 0) 4-1
O C 10
01 4-1 -r! rH
H 3 £ XI
IB rH rH -rl
C^r-H (U m
01 O -P 18
VH a. 01 o
A. a
O 1
g§
rH 01
IH M
4J
rH n
a
M IO
0) U
13 10
01 D
IH
a in
C 01
Id-rl
o
01 C 01
« CTI M
4J 18 O
10 U
r-l 01
-OH
rH rl
18 4J »
0 C O
O OrH
i_3 O UH
01
U
q
18
rH
o <8 e
C XI rH
18 O
tO 4J
01 0! 01
x; IB
a. E TJ
18 C
M >.«
D> IH
0 -H rH
rH H 18
O Oi C
>. > o
j: 01 01
.C T) 18
4-1 a c oi
rl IB 18 01 U)
c u c
3 D> O 01 O
O 4J C -H
01 4-1 IB -rl 4-1
IH 3 H £ 13
18 rH XI H -H
CL -! -H 01 M
01 O rH 4-1
PH U Q
ro
n
4J
H
§,
01
IH
0)
4J
18
4J
tn
T)
C
18
rH
(8
U 0)
rt rrt
U 'U
-H M
O
C7> U
C 01
4J
ui to
H O
x to
W P
01
o
c
18
rH
18
XI
tn
01
18
E
^,
IH
H
VH
01
T3
C
18
01
4J
18
rH
XI
-H
rH
IB
U
^Ji
ro
01
01 H
O>4J 3
C 'H 4J
H rH 10
C rH. 1-1
C 01 01
18 4J 4->
rH IB -H
CX 01 rH 01
tr
C T) rH C
4J C 18 -H
18 IB VH T3
4J 01 18
01 rH C O
13 01 rH
0 -H O
C U 01
IB 01 U
18 IH
rH E 3
IB to IB O
C O M 01
O to BI
Cr> H C
01 - £VH
H 01 O
0) en cu
-H C C
r-l U -H O
« c a. c
o oi a
Sen IB c
10 E 0
o n oi
O rH O
HOB
3 H 18
O 4J rH
in c IB
O XI
4J U
C 01
ri 0! 01
O 0 18
an E
C 3 -
O O >.0!
C m IH M
H 0
r/i-P u 4J
C C IB
4J O H-l
01 C.-O
rl C C IN
X O 18 X
1> C
0) t5
4-1 u-l IB 18
rl 'rH VH
4J 4-1 XI 04
C C -H
01 01 rH -
TJ "v3 IB X
H H U "^
ao
(»
^»
ui
> in
D 4J
M
a o
C 0,
id oi
H
01
4J 01
M 0
O-n
a >
01 rH
H 01
to
^
01 C
> o
H -rl
3 4J
01 >8
>
rl fll
O
C rH
3-H
O O
u to
c
to
^4
o
-p
o
18
IH
S
o
c
18
01
X O
c
01 IB
C rJ.
rl IB
E XI
M
ai in
4-1 01
n e.
- c «
n O H
01 -H 3
O 4J O'4J
5O 01 C
3 rl 01
CM E
rl 4JH -r)
O W IB T3
H C rl 01
O O 0> 01 0)
OtJ M
BI Hi 13 3
C >i HI O 4J
H -H -H
rH j; 01 01 rH.
« 4-1 O
e « a vi IH
o HJ 4-1 ID 18
-rl 18 01 H
O>4J rH 01
01 01 --H C
O c in DI
a c o
q o -H c
10 4.) OrH
>. «! O
rH 4J l-H 01 rl
n c a c 4J
0 3 Oi O C
SO (D-H O
O rl HJ O
n
W tj!
o c
4J -H
0 JiT3 >i Id
4J Id 4> 0!
o e
H rHI rH
O 0) O O
HJ O H-> rH
C M C 4J
oi 3 a> c
> 0 > 0
c in c o
H IH
a
^
n
O
14
O
o
o
to
4J
01 id
aa
id o
3 H
O O
H
H CO
01
4J-O
rH
rH
Q
IH
c
H
£
a
-------
sufficient for a comprehensive identification of the prob-
lem. Nonpoint source water quality problems can be either
transient or steady-state.
Transient problems occur on a localized basis with a time
frame of hours to days. Transient water quality parameters
usually include stream flow, floatables, bacteria, virus,
suspended solids and dissolved oxygen. These parameters
usually cause localized problems during and right after a
storm for periods ranging from hours to days.
Steady-state problems occur on a regional or basin basis
with a time frame ranging from weeks to years. Steady-state
water quality parameters usually include nutrients, dissolved
solids, metals, herbicides and pesticides. These parameters
usually cause regional problems well after a storm during
low to medium stream flow conditions. A good example of a
steady-state problem is eutrophication caused by nutrients
generated by nonpoint sources during the wet season, but oc-
curring during the dry, low flow warm weather season.
Existing data should be reviewed to determine whether the
water quality problems are transient, steady-state or both.
The problem parameters and their probable source should also
be identified.
2. Rainfall Data
Rainfall records corresponding to the stormwater quality
data are required in order to develop unit hydrographs of
stream flow and pollutographs of stream loadings during
storms. These analyses provide an important insight into
the cause-and-effeet relationships between storms and water
quality, thus enabling the determination of a rainfall fac-
tor (R) to be used in calculating the annual nonpoint
source loading. Other analyses which may be performed with
these data include sensitivity analyses relating water
quality to storm intensity, storm magnitude or accumulation
intervals between storm events.
3, Streamflow Records
Records of Streamflow corresponding to those of water qual-
ity and rainfall are also necessary for the completion of
13
-------
unit hydrographs and pollutographs. As with rainfall,
sensitivity analyses can be performed to relate water qual-
ity to stream flow.
4. Pollutograph and Runoff Data
Stormwater runoff data specific to the study area should be
obtained. If the data are not available from past studies,
field monitoring of select storms should be performed. If
possible, at least two storms should be monitored for each
major land use category in the study area. National or
regional nonpoint source runoff data should be used only if
field monitoring cannot be performed. Stormwater runoff
data are necessary for (1) the calculation of the total
annual pollutant loading, and (2) the determination of
pollutant loading rates for specific land uses.
5. Land Use Data
Land use data specific to the study area should be obtained
for the following purposes: (1) to provide a basis for the
logical division of the study area into sub-areas, (2) to
provide a basis for calculating annual nonpoint source pol-
lutant generation from land use loading factors, and (3) to
facilitate the development of control plans for the study
area.
5. Topography and Soils Data
Calculation of site specific runoff loading factors requires
localized topography and soils data such as soil association
maps, critical slope maps, and drainage density maps. These
maps can be readily developed from USGS topographical maps,
aerial photographs, and SCS soil survey data.
7. Nonpoint Source Control Data
Natural or man-made nonpoint source controls applicable to
the study area should be identified and their pollutant re-
moval efficiency and cost effectiveness evaluated. If ex-
isting nonpoint source controls are not available in the
study area, general literature values can be used.
-------
C. FIELD SURVEYS AND NONPOINT SOURCE MONITORING
1. Field Surveys
Field surveys are a key element in the methodology and
should be performed to identify land use trends, specific
areas of development and construction, drainage characteris-
tics, unrecorded sources of pollution, problem areas, and
potential monitoring sites. Reliance on USGS topographical
maps, land use maps, and aerial photographs can be mislead-
ing.
The field surveys should be designed to incorporate the
following activities:
Investigation of specific problem areas identified
from the evaluation of existing water quality and
biological data
Investigation of areas undergoing development and
construction
Investigation of unrecorded point sources or sig-
nificant nonpoint sources (e.g., industrial and
commercial sites with oil spills, refuse and by-
product piles, and junk storage areas)
Investigation of areas experiencing stream modifi-
cations and erosion
Investigation of man-made drainage systems that
alter natural drainage characteristics and bypass
natural buffer zones
Investigation of general land use trends, primar-
ily of development near streams, and topographic
features of the land (slope, soil type, etc.)
Investigation of study area stream conditions dur-
ing or immediately after a rain event to ascertain
and correlate pollution problems with runoff loads
(e.g., stream biota, sediment deposits, scouring
and evidence of land erosion)
15
-------
The field surveys are performed to provide data for both problem analysis
and source analysis. In addition, the information obtained from the field
surveys is used to develop and evaluate nonpoint source controls. Site
specific knowledge of the study area is important when considering the
potential effectiveness and suitability of selected controls.
The level of detail of the field survey will depend on (1) economic and
personnel constraints, (2) time constraints, and (3) the size of the
study area. For a large study area, a limited number of representative
sites should be selected for investigation based on review of the existing
water quality, biological, land use, and storm runoff data.
Detailed notes and photographs should be taken during the field surveys.
Salient observations made during the field surveys should be transferred
to maps of the study area for use in dividing the study area into
subareas and for other applications in the methodology.
2. Nonpoint Source Monitoring
Nonpoint source monitoring should consist of two types: direct land use
runoff monitoring and in-stream stormwater monitoring. Direct runoff
monitoring consists of sampling runoff from typical land use areas to
determine pollutant loading factors specific to land use type. These
factors are used in a modified universal soil loss equation to generate
nonpoint source loadings.
The stormwater runoff or storm sewer monitoring data are land use specific
and are used to determine land use loading factors, whereas in-stream
stormwater monitoring is not usually definitive enough to sort out specific
land use loading factors. The in-stream stormwater data are used to
formulate a relationship between total stream loading and rainfall.
The in-stream stormwater monitoring should be performed during and after
storms at selected summation points in the study area. Summation points
are stream locations selected using the following criteria: (1) maximum
Amount of water quality data are available, (2) stream sites are located at
the downstream end of the study area, and (3) division of the study area
into subareas logically revolves around the summation points (i.e, the
summation point should be the pivot for various subareas; it should not be
included in only one subarea). Subareas and summation points are described
on page 18. The observed in-stream water quality at each summation point
will be used to calibrate the computed nonpoint source loading, calculated
from land use loadings (this mass balance approach is described in detail
further on in the methodology).
16
-------
D. WATER QUALITY AND POLLUTANT LOADING ANALYSIS
The mass balance analysis is used to compare the total in-stream pollutant
loading (calculated by monitoring in-stream dry and wet weather conditions)
to the in-stream nonpoint source loading (calculated by applying pollutant
loading factors to the land uses in the study area). The land use specific
loading factors are then calibrated to provide agreement between the
calculated nonpoint source loading based on land use factors and the
measured in-stream nonpoint source loading.
In calculating the total annual pollutant loading, the total in-stream
loading at a given summation point consists of point source loadings,
nonpoint source loadings and background stream loadings. The in-stream
nonpoint source loading is calculated as shown below:
Total Annual Total
In-stream Annual Point Background
Nonpoint Source = Storm - Source - Loading
Loading Loading Loading
Total annual storm loading can be calculated by monitoring in-stream
water quality at a summation point during a few storm events and
approximating the nonpoint source loading to an annual loading based on
annual rainfall as shown in the following equation:
17
-------
Total Average
Storm = Storm
Loading Loading
Ibs
inch
of
rainfall
Annual
x Rainfall
(inches)
All storms do not yield the same nonpoint loads. The load-
ing depends on the storm intensity, duration arid the time
between storms for land pollutants to accumulate. However,
if there are sufficient rainfall data and storm monitoring,
average storm loadings and annual loadings can be deter-
mined. The methodology for accomplishing this is illus-
trated in the application of the case study under "Water
Quality and Pollutant Loading Analysis."
E. SELECTION OF SUBAREAS AND LAND USE CATEGORIES
The watershed must be divided into subareas. The subareas
should correspond to hydrological drainage areas which are
selected to offer adequate illustration of nonpoint source
loadings and stream loading effects. Figure 2 illustrates a
typical division of a basin into subareas showing receiving
stream and selected summation points. The appropriate se-
lection of subarea boundaries is critical in its effect on
derived sediment yield. A variety of approaches can be ap-
plied and the net evaluation of land use impacts on water
quality is variable, depending on subarea boundaries. Never-
theless, the magnitude of variance among most approaches
should not be so significant that it skews the results be-
yond usefulness.
The following factors should be considered in selecting sub-
areas :
1. Hydrologic characteristics
2. Man-made impacts on natural drainage
3. Land development patterns
4. Common land uses
5. Municipal and hydrologic boundary coincidence
6. Drainage into selected summation points
18
-------
o
c
0)
-------
1. Hydrologic Characteristics
The hydrologic characteristics of the study area, including
natural drainage areas, major tributaries, and ridge lines,
should be considered in delineating subareas. All signifi-
cant tributaries and ridge lines which frame these tribu-
taries should also be included. A significant tributary is
one for which either water quality data have been collected
and are available, or one which is tributary to an urbaniz-
ing or sub-urbanizing land corridor. Also to be considered
are tributary branches which traverse land predominated by a
major land use such as silviculture or mining.
An intuitive assessment of the relative significance of a
tributary in defining subareas is possible by weighing the
relative impact of loadings to the tributary, using data
collected at the nearest nonpoint load summation point in
the mass balance model, the average distance to the summa-
tion point from points along the tributary, and the proximi-
ty of more intensive land uses to the tributary. If the
tributary drains no immediate land uses of the type generat-
ing higher nonpoint pollution loads and if the tributary is
a minor channel far removed from a summation point, its
overall usefulness as a criterion for defining a subarea is
questionable.
2. Man-made Impacts on Natural Drainge
The previous task of assessing hydrologic characteristics
must be qualified by the existence and effects of in-ground
stormwater collection and conveyance systems as well as
above-ground channelization, impoundments, dams, and other
man-made water resource facilities that affect containment,
direction, and flow of runoff. This qualification is im-
portant in the delineation of subareas as well as in the
overall comprehension of land use-water quality relation-
ships. Man-made conveyance systems generally have been
built to fit natural drainage features; nevertheless, these
systems dramatically alter the natural drainage character-
istics of an area and should be considered in the selection
of subareas.
20
-------
3^ Land Development Patterns
Subarea delineation also involves the identification of
development patterns such as standard street grid patterns,
cluster development providing higher densities with high
proportions of open space, central business districts, and
shopping malls. Each of these development schemes are com-
monly associated with community planning techniques of var-
ious eras. Turn-of-the-century planning featured the "City
Beautiful Movement," characterized by broad boulevards,
major civic facilities, and urban park systems. The case
study in this handbook accounted for the effects of Fair-
mount Park in Northeast Philadelphia in the selection of
subareas. This famous development style had a strong in-
fluence on the overall development pattern and is a prime
example of using readily available knowledge of regional
development patterns, coupled with an understanding of urban
planning milestones, to clarify subarea characteristics.
4,
Common Land Uses
The sediment yield effects of common, contiguous land uses
are more evident if associated with the same subareas.
Consideration should be given to those land uses that are
similar and contiguous but which lie in different drainage
areas. A decision must be made whether they should be
combined in one subarea regardless of drainage, or separated
into two or more subareas. Clearly, the effect of drawing
the subarea around the contiguous land area would be to show
a more significant raw yield from that area, particularly
for urban and agricultural areas. Where all such land lies
in one municipality, it may be sensible to follow this
approach because the analysis of nonpoint source controls is
simpler if the land lies entirely in one municipality. The
primary goal is to delineate subareas so as not to lose or
mask information that may be desired at future stages of the
methodology. While the strict hydrologic approach is con-
ceptually valid, it is void of the practical considerations
associated with the response of local government to nonpoint
source control.
5. Municipal and Hydrologic Boundary Coincidence
There is a practical value to using municipal boundaries as
subarea boundaries, in that most cpntrol programs are organized
21
-------
on a municipal basis. In some cases, however, natural fea-
tures such as ridge lines may be chosen as subarea bound-
aries. The practical value of using natural features as
boundaries is that the hydrologic system is more accurately
defined for the analysis. In some cases, a ridge line may
even serve as a municipal border. In general, however, hav-
ing municipalities wholly contained in a subarea is more ap-
propriate for this methodology. The one exception is that
summation points must reflect total drainage of all subareas
above it such that the measured nonpoint loads at these
points accurately reflect the impact of contributing land
uses.
6. Selection of Land Use Categories
After subarea delineation is complete, the next step is to
map the study area by land use categories and measure acre-
age of each land use group in each subarea. The specific
land use categories to be mapped depend on (1) the dominant
land uses in the study area, (2) the level of detail required
or desired in the analysis, and (3) the availability of spe-
cific storm runoff data for each land use. Runoff data may
be available from previous studies, literature sources, or
monitoring performed during the study. The rationale for
land use category selection is addressed on page 100 of the
case study.
The tasks of mapping and calculating acreage are standard
procedures common to all planning agencies. Consequently,
this task will not be expanded upon in this section. An ex-
planation of the mapping and calculation procedures as ap-
plied in the Lower Wissahickon Case Study is found in the
technical appendix accompanying that study.
F. DETERMINATION OF RAW LOADING RATE
Pollutant loading rates should be calculated for each land
use in the study area. These loading factors can be ob-
tained from literature values or from the direct runoff
22
-------
monitoring described later in this report. The sediment
loading function is based on concepts of the mechanisms of
gross erosion and sediment delivery. These loading factors
are in reality net loading rates for a specific land use
with unique characteristics such as soil properties, slope
conditions, land cover conditions, and conservation prac-
tices. The loading factor developed from a specific site
for a given land use category should not be applied direct-
ly to other similar land use areas. Land use modifications
based on the universal soil loss equation should be used to
allow a site-specific application of the loading factor.
This concept is based on principles illustrated in the uni-
versal soil loss equation.
Information pertaining to soil classifications, slope condi-
tions, etc. are required to use the universal soil loss equa-
tion. The present methodology uses a modified soil loss
equation as described below*:
n
Y(S)E = AiKi (R-LS-SD-P-K2) i
i = 1
where :
Y(S)E is the sediment loading in Ibs/year
A is the area of the subarea in acres
K is the average annual raw runoff rate from the
land use in terms of Ibs/acre/yr
i is the type of land use
R is the rainfall factor expressing the ratio
of annual rainfall intensity to the rainfall
intensity of the calibration year
LS is the slope length/steepness factor which re-
flects increased sediment detachment and trans-
port as runoff velocities and volume increase
with increased slope length
* A complete description of the universal soil loss equation
can be found in the publication "Soil Erosion: Prediction
and Control," Soil Conservation Society of America, 1977,
and "Loading Functions for Assessment of Water Pollution
from Nonpoint Sources," McElroy et al, EPA-600/2-76-151,
May 1976.
23
-------
SD is the sediment delivery ratio which reflects
the fraction of sediment runoff from upland
sources actually reaching a contiguous stream.
Factors such as proximity to stream, soil char-
acteristics (porosity, water table, etc.) and
availability of deposition areas affect the sed-
iment delivery ratio
P is the erosion control factor for natural sedi-
ment control practices such as buffer strips
around streams and wetlands which act as a sink
to sediment and other pollutants
K2 is the erosion control factor for man-made con-
trols such as regulatory and physical controls
In the modified soil loss equation, the factors R, LS, SD, P
and K2 are modifiers that relate the raw loading rate to the
net loading rate reaching the stream. The raw loading rate
expresses the raw erosion of sediment from the land, while
the net loading rate is the reduced erosion reaching drain-
age ditches or streams as a result of being reduced by
modifiers. Since the term (R-LS-SD-P-K2) will always be
less than unity, the net loading will always be lower than
the raw loading.
Based on the above principles, the loading rate measured for
a specific land use during the monitoring program is actual-
ly a net loading rate that is site-specific for a particular
land use. To apply this loading rate to other similar land
uses, the raw loading rate must be calculated.
This calculation consists of applying the land use modifica-
tion (factors) to the net loading rate.
The general methodology for calculating raw loading rates
for each land use category is presented below:
Monitor direct stormwater runoff from a site rep-
resenting a specific land use.
Calculate the net loading rate for the specific
site in terms of Ibs/acre/yr or other suitable
units.
-------
Based on site-specific characteristics of the
area, calculate modifier factors, R, L, S, SD, P
and K2.
Using the modifiers, calculate the raw loading
rate as follows:
Raw Loading Rate = Net (Measured) Loading Rate
R.LS-SD-P-K2
The raw loading rate can now be directly applied to other
similar land uses in the study area. The procedure above
should be followed for each typical land use in the study
area. If runoff rates from typical land uses are not avail-
able, literature loading values may be used.* Care must be
taken to assure that the literature values represent runoff
from similar soil characteristics, land use activities and
climatic conditions.
Not all pollutant runoff is sediment-borne (associated with
sediment particles); some pollutants are soluble (dissolved
in water). The Four Mile Creek-Occoquan study, (Northern
Virginia Planning District Commission, October 1976), noted
that as much as 50 percent of runoff nitrogen and phosphorus
was in soluble form for some land uses. Therefore, there
are some inherent inaccuracies in the use of KN ratios for
these pollutants. However, since at least 50 percent is
sediment-borne, this approach can yield a fair approximation
of pollutant loads without resorting to an expensive inde-
pendent analysis of each pollutant.
The above procedure is performed for the sediment loading.
Other parameters are related to the sediment loading by a
factor Kn relating the concentration of a particular para-
meter (nutrient, metal, etc.) to that of sediment for each
land use. The loading for a particular parameter is cal-
culated as follows:
* Typical pollutant runoff rates can be found in referenced
items 12, 18, 19, 21, 38 and 39 listed in the bibliography.
25
-------
Y(S)N = > K1(i)
i = 1
where:
Y(S)N is the pollutant loading rate for parameters
(N) in Ibs/acre/yr
K1 / jv is the net sediment loading for land use (i)
determined previously (A-K'R-LS-SD'P-K2)^
KN/jj is the ratio of the loading rate of pollutant
(N) to the loading rate of sediment for the
land use (i)
The parametric ratios, KN, for each typical land use can be
determined by field monitoring of land use stormwater runoff
or from literature values of similar land uses.
G. DETERMINATION OF RAW LOADING RATE MODIFIERS
The simplified universal soil loss equation employs several
modifying factors to the basic raw runoff loading rate (K
factor). These are:
Rainfall Factor - R
Slope Length - LS
Sediment Delivery Ratio - SD
Natural Control Factor - P
Other Pollutant Ratios - KN
Best Management Control Factors - K2
1. Rainfall Factor - R
The rainfall factor is used to determine the effect of storm
frequency and intensity on sediment erosion from pervious
land and on scour and runoff of pollutants from urban areas.
26
-------
The mass balance model is empirically calibrated to the
rainfall intensity for the calibration (control) year. It
is modified for rainfall/intensity for subsequent years and
alternative iterations.
The rainfall factor for subsequent years can be determined
by: (1) using the ratio of rainfalls of the subsequent
years to that of the control year, or (2) using the ratio of
total annual flow (cfs-days) of the subsequent years to that
of the control year. A high degree of success was attained
using both methods in the case study. Calculations of non-
point loads in the case study showed an error of 10% using
the ratio of total annual rainfall and an error of 4.3% us-
ing total annual streamflow ratios. Both errors are within
acceptable limits; neither represents a preferred method.
The higher error obtained by using the rainfall ratio prob-
ably resulted because the rainfall ratios were based on re-
cords obtained from a basin adjacent to the study area.
Stream flow ratios were based on flow measurements taken
within the study area.
The "R" factor for the rainfall method is simply expressed
as:
R _ Total Rainfall of Year Under Consideration
Total Rainfall of Calibration Year
Some error is implicit in determining this ratio in that not
all of the rainfall occurs during storms. Nevertheless, in
the case study more than 75% of the rainfall in excess of
0.5 inches per day occurred during storms.
Similarly, the "R" factor for the total annual flow method
is simply expressed as:
R = Total Flow (cfs-days) of Year Under Consideration
Total Flow (cfs-days) of Calibration Year
Again, there is a slight implicit error in that not all of
the flow results from storm events. Nevertheless, the storm
flow in the case study for rainfalls in excess of 0.5 inches
per day accounted for greater than 60% of the total flow.
27
-------
2. Slope Length - LS
This factor describes the effect of steep slopes and the run
lengths of these slopes. The detailed data needs and analy-
ses required in the universal soil loss equation were sim-
plified in this methodology. A simplified averaging approach
was developed as follows:
(A ) Topographic maps of the area are examined and the
land area is apportioned into several slope ranges.
For example, a study area could be divided into
three slope characteristics:
(!) Those areas on slopes less than 4 percent
grade
(2) Those areas on slopes between U percent grade
and 8 percent grade
(3) Those areas on slopes greater than 8 percent
The land use areas outlined on topographic maps
are then examined and characteristic run lengths
of slopes are determined. For example, it might
be found that:
(1) Slopes of less than 4 percent grade exhibited
an average run length of 2000 feet
(2) Slopes between 4 percent and 8 percent ex-
hibited an average run length of 1000 feet
(3) Slopes greater than 8 percent exhibited an
average run length of 200 feet
(C ) Using the slope-lengths determined from steps 1
and 2, and the criteria presented in Figure 3, the
slope length (SL) factor is determined for each
land use. For land use areas overlapping two or
more characteristic slope-lengths, a weight
averaged factor must be determined.
28
-------
Slope Length, Meters
o
<0
20.0
10.0
8.0
6.0
4.0
3.0
2.0
1.0
0.8
0.6
0.4
0.3
0.2
3.5 6.0 10
20
40 60 100
200
400 600
0.1
t t I
10
20
40 60 100
200
400 600 1000 2000
Slope Length, Feet
The dashed lines represent estimates for slope dimensions beyond the range
of lengths and steepnesses for which data are available.
Source: EPA, 1976
FIGURE 3
SLOPE LENGTH DETERMINATION CRITERIA
29
-------
SL (mild) XA (mild) +SL (mid) *A (mid) +SL (steep) XA (steep)
SL(D =
where :
= weight averaged slope length for land use (1)
SL/m.:-]j\ = slope length factor for mild slopes
1 ; (less than 4%)
= slope length factor for mid rcinge slopes
(4% to 8%)
SL, . = slope length factor for steep slopes
(steep)
(greater than 8%)
= area
See the case study application of slope-length determination
in Section IV for a detailed explanation of the application
criteria.
3. Sediment Delivery Ratio - SD
This ratio expresses the percentage of eroded sediment that
reaches a stream or drainage ditch. Two factors influence
sediment erosion: (1) the ability of the soil to absorb
water and retard the movement of sediment expressed as the
runoff potential of the land use area, and (2) the distance
or proximity of the eroded sediment to streams or drainage
ditches, expressed as the drainage density of the land use
area.
(A) Runoff Potential: The runoff potential of the land can
be determined by the use of hydrologic groups, categories of
soil type designated by the Soil Conservation Service. All
the known soil series are divided into four classes of run-
off potential (low, moderately low, moderately high and
high) based on infiltration rate and transmission rate.
Infiltration rate is defined as the rate at which water
enters the soil at the surface; the rate is controlled by
surface condi tions. Transmission rate is the rate at which
the water moves in the soil; the rate is controlled by soil
horizons. Pertinent soil properties are particle size com-
position, depth of seasonally high water table, permeability
30
-------
after prolonged wetting, and depth to impermeable layer.
All classifications are based on conditions of prior wet-
ting, opportunity for swelling and the absence of protective
vegetation.
The soil types which make up a particular area can be found
in county soil surveys which are produced by the U.S. De-
partment of Agriculture, Soil Conservation Service (SCS).
These surveys describe the soil series found in the county
and map their location. Using the nation-wide master list
of hydrologic groups, the runoff potential of each soil
series can be determined and a map produced showing runoff
potential within the study area.
The soil surveys are available for most of the country from
county-based SCS offices. The master Hydrologic Group list-
ing can be found in "Hydrologic Guide for Use in Watershed
Planning," National Engineering Handbook, Section 4, Hydrol-
ogy, Supplement A, USDA, Soil Conservation Service, 1964.
For example, a runoff potential map for the Wissahickon
Watershed was produced to determine the percentage of land
in each hydrologic group and to identify problem areas. A
soil map was compiled using the Montgomery County Soil
Survey (1967) and the Bucks and Philadelphia Counties Soil
Survey (1975). The soil series found in the area are then
categorized into hydrologic groups using the nation-wide
master list.
If a locally-named soil series does not appear in the list,
a local list can be consulted, or if such a publication is
not available, the soil type can be compared with character-
istics of soils included on the master list, and its rela-
tive position can be determined. A map overlay is then
produced with four categories: (a) moderately low, (b)
moderately high, (c) high runoff potential and (d) urban or
man-made land. This latter category consists primarily of
impervious surface and thus can also be considered as having
high runoff potential.
( B ^ Drainage Density Determination: A brief literature and
d.ata search should be made to ascertain whether any previous
investigations have provided information on drainage density
within the basin. Such sources as USGS water resources
surveys and local geologic surveys should be consulted. If
31
-------
drainage density data are not available, drainage density
can be calculated as follows:
1. Using topographic maps, measure length of stream
miles, drainage ditches and storm sewer lines
2. Calculate drainage density of each subarea as:
Drainage Density = Miles of Drainage =
Area of Drainage (Mi2)
C. Sediment Delivery Determination; Three steps are used
to determine sediment delivery:
(A) Land use overlays are superimposed on the hydro-
logic overlays, and areas within each hydrologic
soil condition are determined (A^ through AD for
hydrologic groups A through D) .
(B) Using the criteria shown in Figure 3, determine
the sediment delivery ratio of each hydrologic
grouping for each subarea drainage, and summarize
results in a table (SD^ through SDp for hydrologic
groups A through D in each subarea; .
(C) Calculate the sediment delivery ratio of each
subarea/land use as follows:
SD = A xSD +A XSD xA xSD +A xSD
Total land use area within subarea
(land use)
Some simplifications may be employed. In the case study a
relatively narrow range of drainage densities existed;
therefore, a single drainage density was employed for all
subareas. See the case study for more details on applica-
tion of sediment delivery ratio criteria.
32
-------
4. Natural Control Factor - P
This factor expresses the reduction of sediment delivered to
the stream as a function of natural controls, i.e., flood
plain buffer strips or natural storage areas such as wetlands,
In the case study, buffer strips surrounded many of the
major streams; however, the buffer strips were usually by-
passed by storm sewer lines that drained urban areas directly
to the streams. This bypassing of buffer strips must be ac-
counted for in the P factor.
The buffer strip reduction of sediment loads is determined
by:
Mapping an overlay showing significant buffer
strips surrounding receiving streams and super-
imposing the buffer strip overlay on the land use
map.
Each land use category within each subarea is
examined to determine the relative order of mag-
nitude of the buffer effect:
Wide buffer between land use and stream - Maxi-
mum effect
Medium buffer between land use and stream - Medium
effect
Narrow buffer between land use and stream - Mini-
mum effect
The retardation effect of sediment delivery is
estimated for each buffer strip classification:
wide, medium, narrow.
The buffer strip bypass factor (KB) is estimated
for each land use in each subarea. This factor is
a function of the estimated fraction of load from
the land use area that is not bypassed directly to
streams. It is estimated for each land use and
represents the fraction of the sediment load that
is subjected to buffer strip retardation.
33
-------
The P factor is calculated for each land use as
follows:
P = (1 - KA x KB)
where:
P = fraction of load delivered to stream
KA = fraction of load retarded due to
buffer strip
KB = fraction of load received (unbypassed)
by buffer strip.
The P factor determined in step 5 should be modified by the
estimated action of natural detention basins. Swamps, wet-
lands and surrounding streams act as detention basins and
will retard sediment delivery to streams. The P factor is
multiplied by the estimated reduction due to natural deten-
tion to arrive at the final P factor. See the case study
for more details on the application of the natural control
factors (P).
5. Other Pollutant Ratios - KN
The KN ratio provides a means of determining pollutants
loads other than sediment, e.g., BOD, nitrogen, phosphorus,
heavy metals. They are simply expressed as a fraction of
the sediment load. They are determined empirically from
measured land runoff pollutants. There are inherent inac-
curacies in this method in that not all transport of non-
point pollutants are sediment borne. Some nonpoint pollu-
tants are in soluble form; however, with adequate monitoring
of pollutants from typical land uses, a KN ratio can be de-
termined which reflects total pollutant of both soluble and
insoluble forms.
Sediment concentrations typically peak sooner than other
pollutants such as phosphorus and nitrogen. Therefore,
ratios of pollutant concentrations to sediment concentra-
tions should not be used as a measure of the fraction of
pollutant load delivered because this value will vary in
time after a storm and the results may be misleading. Only
the ratio of pollutant loadings should be used to calculate
KN.
34
-------
Pollutant loadings are calculated as a flow-concentration
product of the area under the hydrograph (flow curve) and
the corresponding pollution concentration curve.
The KN ratio for each specific land use within each subarea
is estimated using the average land use KN ratios as a
guide. For instance, if the average phosphorus KN ratio for
an urbanized area was 0.001, and the urban land under con-
sideration is populated to approximately 75% of that of the
average urban area, the phosphorus KN ratio for the partic-
ular land use would be (0.75 x .001) or 0.0075.
H. LAND MANAGEMENT CONTROLS
In this methodology, land management control is a general
term which includes municipal regulatory controls, mainten-
ance controls, and physical controls. It is important to
assess the meaning of "control" in this context. Control in
land development management has historically implied a
responsibility of government, generally on the municipal
level, to regulate the activities of developers and builders
to ensure conformance with community standards and goals.
In this context, the control most often has been an ordin-
ance, regulation, or local law which carries with it local
responsibility for monitoring and enforcement. Such controls
are herein referred to as "municipal regulatory controls."
They must necessarily be differentiated from what will be
termed "maintenance controls" and "physical controls." In
this study, the primary interest is in the provisions of the
land management controls which address runoff from land.
1. Maintenance Controls
Maintenance controls are required for existing developed
sites and new land development sites. For developed sites,
maintenance controls are required where sediment generation
results from the renovation or alteration of existing sites
(building alteration, road repair or widening, and other
structural improvements) as well as from the common daily
accumulation of debris, litter, and dirt. For new land de-
velopment, maintenance controls are applied primarily as a
35
-------
precautionary measure to contain building materials, scraps,
and other wastes. These controls are actual activities in-
volving labor and machine intensive work in possible con-
junction with physical facilities or structures (for further
information on maintenance controls, see pp. 44-45). The
application of this methodology for the case study incorpo-
rated three areas of maintenance controls:
Neighborhood sanitation
Maintenance of on-lot disposal systems
Collection system controls
2. Physical Controls
Physical controls involve structural measures which inter-
face with natural stormwater runoff to reduce the effective
pollutant load. For construction activities, this would in-
volve control of storm-generated runoff, erosion, sedimenta-
tion and subsurface infiltration.
Table 2 illustrates a variety of physical controls which are
typically considered to satisfy some of the following objec-
tives of a pollution control program applied to construction
activities:
(A) Protection of exposed ground cover to minimize
erosion and soil loss
( B) Control of the velocity of stormwater runoff to
maximize natural infiltration, minimize ground
cover and stream bank erosion, and to inhibit
downstream flooding
(C) Promotion of natural infiltration of runoff and
natural filtration of sediment and associated
pollutants
( D) Containment of sediment and associated pollutants
on-site
(E) Containment of stormwater runoff on-site for
either retention, detention, or velocity control
Many of the controls listed in Table 2 appear to serve
multiple objectives. For instance, mulches, nets, and
36
-------
TABLE 2
PHYSICAL CONTROL PRACTICES
FOR CONSTRUCTION ACTIVITIES
I O
5 £
>, OJ tO
-O -P 00 S- 4- 4-
4- C -i- -P O O
O 3 O 4-i
O O O !- 4-> 4->
C S- r 4- C C
O t£3 0) CD C O) d>
i- > C >-H £ 4-> E
P T3 r T- C C C
OO) 4-O S- o3 T-O) -1-4-
CU(/) 4-S- 0) rOE rO4-
-t->O O+-> M-M -M-i- -PO
OQ- CC r-C CTD EC
S-X 3O -r-OI 00) O3
D-LU 0:0 U_E cjon oo;
Channel walls and linings
Chemical soil binders
Culvert risers
Dams and reservoirs
Detentionpasins
Ditches and swales"
Diversion terraces
Downdrains
Filter berms
Grade stabilization structures
Infiltration basins
Inlets
Interceptor channels
[Level spreaders
lulches
Nets
Previous blankets
Ponds
Porous pavements
Retention basins
Rjprappinc[
Rooftop detention
Sandbag or straw bale barriers
Sedimentation basins
Seepage pits
Sodding
Spillways
Note: Most of these control practices may be applied selectively to
new construction sites as well as existing development under
maintenance, renovation, or alteration.
-------
pervious blankets, aside from protecting exposed ground from
erosion, also serve to encourage natural water infiltration
on-site. It is also significant that many of the physical
controls listed require some form of maintenance control.
The best illustrations of this are catch basins and sedimen-
tation basins which require regular maintenance.
The physical controls listed in Table 2 can be explicitly
required or implicitly required by stipulating performance
goals in a municipal ordinance, law, or regulation. In ap-
plying controls within this methodology, attention has been
placed on municipal regulatory controls and supportive
enforcement mechanisms. A list of municipal regulatory
controls and potential maintenance controls are included in
Appendix F. The former list contains the ordinances, laws
and regulations considered in the case study of the Lower
Wissahickon Watershed.
5. Inventory of Existing Land Management Controls
At this stage, the methodology should focus on four primary
objectives:
(A) Identify applicable land management controls on
new and existing land development which might have
beneficial water quality effects, regardless of
their current application in the study area.
Identify land management controls currently im-
posed by regulatory agencies in the study area.
This should be done for both new land development
and existing development.
Assess the "relative" effectiveness of the poten-
tial land management controls on reducing sediment
delivery from new land development sites and ex-
isting, but renovated, sites.
(D) Assess the relative effectiveness of the land man-
agement programs of regulatory agencies as cur-
rently imposed on new land development and exist-
ing, but renovated, development. A management
program is the total combination of land manage-
ment controls in force in a municipality.
38
-------
As the objectives imply, an important distinction in this
methodology is drawn between sediment generation from new
land development and sediment generation from existing de-
velopment with on-going maintenance and renovation. New
land development commonly involves a significant tract of
undeveloped or formerly agricultural land (minimum 2 acres)
which will be developed to accommodate a new, more intensive
use (i.e., residential/industrial/commercial). Sediment
generation occurs primarily as a result of soil cover re-
moval and topsoil disruption. Existing land development, on
the other hand, accounts for a much larger proportion of
total land. It is assumed that sections of each study area
and each subarea were developed at different times. Further,
it is assumed that the older areas require an on-going
schedule of maintenance, renovation and alteration, with
newer sections entering this cycle in years to come. The
total picture becomes one of continual maintenance, ren-
ovation, and alteration in the study area.
Whjle emphasis in the literature has been mostly on the
significant generation of sediment and other pollutants from
new land development, it is the intent of this methodology
to qualify this emphasis by directing attention to the
sediment-generating capacity of existing land development.
The case example of the Lower Wissahickon Watershed is one
in which this approach is applied to a study area with both
significant new development as well as a substantial level
of development pre-dating the year 1900. This latter fact
permitted the investigators to assume a continuing need for
maintenance, renovation, and alteration.
4. Existing Land Management Controls
The land runoff controls generally applied in the study area
must be identified. Although land runoff controls include
regulatory, physical, and maintenance controls, at this junc-
ture assessment of existing use can be limited to regulatory
and maintenance controls. Applied physical controls are
more difficult to catalogue and are generally included as
requirements of regulatory controls. To a certain extent,
maintenance controls are also included as such requirements.
The zoning and subdivision regulatory documents of each
municipality must be reviewed. The same applies to all
building or construction codes as well as special environ-
mental ordinances. The objective of these reviews is to
39
-------
ascertain to what extent these regulations provide controls
on land development activities that minimize water quality
degradation.
Most of these regulatory tools have been created to satisfy
objectives other than water quality improvement. Some
ordinances governing environmentally sensitive areas have at
best made water quality upgrading an objective in addition
to general objectives such as protection of persons and
property. Regulatory tools having the strongest water
quality objectives include up-to-date subdivision or land
development regulations, steep slope or hillside protection
regulations, flood plain or flood protection regulations,
and erosion and sedimentation control regulations.
Examination of municipal codes may show these regulations to
be in a common ordinance, such as a land development ordi-
nance. Occasionally, some of these regulations are mandated
by state or county law. In the case of the Lower Wissahickon,
for example, the Commonwealth of Pennsylvania has adopted
erosion and sedimentation control regulations and procedures
that apply to all municipalities.
Upon identifying the relevant documents, two inventories
should be compiled, one of controls common to all new land
development and the second of controls applicable to all
forms of maintenance, renovation, and alteration to existing
development. The latter inventory will probably be more
difficult to complete because many of the controls on build-
ing renovation or alteration are tied to building permit
processes and their statutory basis in constriiction or
building codes. These are regulatory tools which typically
do not contain provisions for water quality objectives other
than plumbing-related specifications. Some maintenance
activities, e.g., street sweeping, and solid waste manage-
ment functions such as trash removal and landfill opera-
tions, may be provided for under state mandates to local
government and may not require local ordinances. The prob-
lem in these cases is lack of information, such as frequency
of sweeping or trash collection and conveyance, which is
essential to assess control effectiveness properly.
Finally, maintenance construction controls should include
road and highway building, repaving, and widening. In each
state, the responsibility for roads is divided among munic-
ipal, county, state, and federal governments. The controls
-------
on each level, if they exist at all, will differ signifi-
cantly in terms of their water quality implications. In
practice, many of the physical controls applied to new land
development can be applied to road construction. This is
particularly true for sedimentation basins, grade stabiliza-
tion structures, mulches, nets, pervious blankets, and sand-
bag or straw bale barriers. An interview with local, coun-
ty, and state highway or public works agencies would suffice
to identify controls currently being applied. Visits to
road maintenance or urban renewal sites are instructive and
should be included in the design of the field study.
5. Assessing the Effectiveness of Land Management Controls
This methodology requires that the effectiveness (measured
as sediment load reduction) of controls be determined for
each land management control program of each subarea. This
measure is referred to as the K2 factor, derived from the
following equation:
K2 = 1 - (KNLD + KO + KA)
where:
1 = No reduction of load due to control practices
KNLD = Reduction factor for control practices applied
to New Land Development
KO = Reduction factor for general land maintenance
practices, not including that applied to new
land development
KA = Reduction factor for control practices applied
to agricultural (crop) production
The case study explains in detail how K2 factors for each
land use category of each subarea are determined.
(A) Regulatory Control Effectiveness; Table 3 illustrates
a point system for ranking regulatory controls according to
their capability to achieve pollutant load reduction. Three
points represented a control capable of significant load re-
duction, 2 points moderate, 1 point minimal, and 0 points no
load reduction.
-------
TABLE 3
REGULATORY CONTROLS: POLLUTANT LOAD
REDUCTION FACTOR POINTS
Load Reduction Points Description
Zoning
1 General Zoning Ordinance Contents
2 PRO or Cluster Provisions
1 Mixed or Averaged Density Provisions
2 Flood Plain Designation
2 Steep Slope Designation
1 Agricultural District Designation
Subdivision Ordinance
3 General Subdivision Ordinance
1 Surface Drainage Plan Requirement
1 Net Runoff Limitations
1 Structures: Runoff Reduction On-site
1 Structures: Sedimentation/Erosion
Control On-site
1 Maintenance Requirements
1 Tree Protection/Preservation Requirements
1 Connection to Public Sewer Required
Steep Slope Protection Ordinance
Flood Plain Protection Ordinance
2 General Ordinance
1 Runoff Control Criteria
1 Erosion Control Criteria
1 Streambank Alteration Criteria
1 Watercourse Alteration Criteria
2 Stream Setback Requirement
PRO or Cluster Ordinance
2
Average or Mixed Density Ordinance
1
Environmental or Community Impact Statement
2
Erosion and Sedimentation Control Regulations
3
42
-------
Technical information on the effectiveness of such controls
is not currently extensive, although a significant amount of
current research addresses this subject. Whatever technical
information is available to the 208 planning agency should
be coupled with any available, locally-based data on the ef-
fectiveness of controls. Of particular value are the follow-
ing technical documents:
Urban Stormwater Management and Technology, An
Assessment. A report by Metcalf & Eddy, Inc., Palo
Alto, California, for the National Environmental Re-
search Center, U.S. EPA, Cincinnati, Ohio, December
1974.
Water Pollution Aspects of Street Surface Contaminants.
A report by James D. Sartor and Gail B. Boyd for the
Office of Research and Monitoring, U.S. EPA, Washing-
ton, B.C., November 1972.
The Urban Stormwater Runoff Presentation. Printed
text of presentation by Dennis N. Athayde and Andrew
Waldo of U.S. EPA Water Planning Division, Washington,
D.C., to EPA 208 Planning and Implementation Confer-
ence, Boston, Mass., March 1977. Available from EPA.
The data in these documents are applicable to maintenance
construction controls as well as new development construc-
tion controls. They address primarily the relative effec-
tiveness of physical controls. This information can be
extrapolated to determine the effectiveness of municipal
regulatory controls. In so doing, however, it is essential
to include factors for implementation probability, in that
such controls are not by themselves assurances of physical
control implementation.
In order to derive the scores in Table 3, it was necessary
to consider the objective of each control, its relationship
to water quality management, and the types of physical con-
trols required by each. Literature values were used to as-
sess physical control effectiveness and comparisons were
then made between the controls listed. As examples, erosion
and sedimentation control regulations (which had a score of
3) have a direct relationship to water quality management
and physical controls commonly required by such regulations
achieve 50-90% sediment removal effectiveness according to
43
-------
sources just cited. At the other extreme, average or mixed
density ordinances (score of 1) bear only a marginal rela-
tionship to water quality management and may require no spe-
cific physical controls. Its value is more important from
the macro view of overall municipal development through the
preservation of open space and minimization of land disrup-
tion .
(B) Maintenance Control Effectiveness; For maintenance
control effectiveness, a similar point system was also used
in the case study. Table 4 illustrates the types of con-
trols and their associated point ranges. As indicated, the
highest point score possible is 10, which represents a well
developed, generally effective program of maintenance con-
trols for developed areas. The categories of maintenance
controls determined to be significant for the case study
were:
General land use policies
Public cleanliness measures
Controls on state and local road building
Controls on maintenance
TABLE 4
RANGE OF LAND MANAGEMENT CONTROLS
IN DEVELOPED AREAS
Maintenance Controls Point Range
General Land Use Policy
(includes zoning and building controls) 0-1
Public Cleanliness 0-4 (total)
Street cleaning (0-1)
Catch basin cleaning (0-1)
Refuse 6 litter collection (0-1)
Sewer flushing (0-1)
Controls on State and Local Road Building 0-2
(includes street maintenance)
Controls on Maintenance Construction 0-3
Total Control 0-10
44
-------
In applying this approach, the choice of numbers of points
for each category should be predicated on:
Whether the type of control is being applied in
the study area
The adequacy of the control in accomplishing its
objectives
As an illustration of this point system, see pages 132-135
in the case study. As Table 25 illustrates, the urban areas
in the Wissahickon Watershed received 6 out of 10 points.
These six points were associated with a pollutant load re-
duction factor of .41, the derivation of which is explained
in the case study. Using a linear scale of ten, this means
that the maximum reduction capability of the control system
described in Table 4 would be approximately 70% of the total
load.
This approach and rating scheme seemed reasonable for this
case study as well as other studies using this methodology,
assuming the following assumptions:
(1) Land management regulatory programs for newly de-
veloped areas are inadequate in some localities.
(2) Inspection and enforcement of land management reg-
ulatory programs for newly developing areas are
more stringently performed than in developed
areas.
(3) The application of physical controls is more
easily arranged on new sites than on developed
sites undergoing change.
(4) The state-of-the-art of public cleanliness and
maintenance construction control is at a develop-
mental stage in which controls and practices are
being discussed, tested, and evaluated. Neither
a firm approach or consensus on the components of
the approach are available at this time.
(C) Control Effectiveness by Subareas; This methodology
requires that a K2 rate (sediment load reduction) be deter-
mined for the land runoff control program of each subarea.
45
-------
The difficulty in doing this is that each subarea is com-
prised of partial or entire municipalities, each of which
has a distinct and different control program due to any of
the following factors:
Date of municipal incorporation and subsequent
maturity in land management regulation
Amount of land annually undergoing new development
Existence of environmentally sensitive areas in
municipality and corresponding response to pro-
tection of such areas
At the subarea selection stage (see p. 18), it was suggested
that the factor of control be considered and that, conse-
quently, municipal boundaries be considered along with
hydrologic boundaries as determinants of subarea boundaries.
The control effectiveness of the subarea is actually the sum
of the control effectiveness of each of its constituent
municipalities. Although this is the conceptual approach
recommended in this methodology, and was the approach ap-
plied to the case study, there is a subtle limitation: the
land management controls on new development are a factor
only if there is evidence of new development or potential
for such development. It is important to examine the split
of municipalities between subareas to determine if land in a
subarea actually has land development potential. This would
be an issue in a municipality with a sophisticated control
program and an area most prone to receive development. If
that municipality is divided into two or more subareas, the
planner should be careful to assess whether a high reduction
factor is appropriate in each subarea, or just the subarea
containing land most prone to development.
If the land management controls for new development have
been inventoried, then the point system for relative control
effectiveness as presented in Table 3 and explained on p. 43
can be developed for each municipality in the watershed.
This has been done in Table 22 in the case study. The pro-
duct is a list of municipalities, each identified by its
total control points.
The point system in the case study (see Table 22) showed
that the highest point total for a municipality was 18,
-------
corresponding to a total load reduction of 95%. On a linear
scale, the lowest point total of 6 corresponded to a 30%
reduction. These reduction factors for each municipality
applied only to new land development and are based upon:
A literature review of the general effectiveness
of each control for new land development
An assessment of the existence and adequacy of
controls used in the study area
For existing development, a similar approach should be em-
ployed using tables of potential controls, potential con-
trol effectiveness ranking, and controls currently employed
by municipalities. In the case study, it was determined
that controls on existing land development were not suffi-
ciently practiced throughout the watershed but were gener-
ally practiced in certain types of urban and suburban areas.
Consequently, reduction factors were generally derived for
land use categories rather than for subareas. Each subarea
had the identical set of existing land development reduction
factors assigned to it.
The approach recommended in this methodology for determining
new land development control effectiveness of each subarea
is to execute the following equation and sequence of steps:
n
KNLD = y_ X(i) knld(i)
i = 1
where:
KNLD = New land development control factor for subarea
n = Municipalities in subarea (1 through n)
X(i) = Per cent composition of subarea by municipality i
knld = New land development control factor for munici-
pality i
47
-------
1. Determine the percent composition of each subarea
by its constituent municipalities. (i.e., 50% of
subarea is Township X or part thereof, 30% is
Township Y, and 20% is Township Z).
2. Multiply the percent composition of each municipal
part of the subarea by its new land development
control factor.
3. Add the product of each to arrive at the new land
development control factor for each subarea.
I. MASS BALANCE ANALYSIS
The mass balance analysis compares the accumulated nonpoint
source pollutant loadings at summation points calculated in
two ways: (1) from measured water quality, and (2) from the
application of pollutant loading factors to land uses in the
study area. The nonpoint source loadings calculated from
the application of loading factors to land uses are cali-
brated to the measured loadings. Differences between the
measured loadings and calculated loadings are reconciled by
re-examining raw loading factors and modifiers.
The measured nonpoint source loading is calculated from the
data previously obtained in the section, "Water Quality and
Pollution Load Analysis." Basically, the nonpoint source
loading equals the total measured loading (on a per storm or
annual basis) minus the point source loading (obtainable
from NPDES permits).
The nonpoint source loading based on land use is calculated
by applying the land use factors (modifiers) and raw loading
rates previously derived in the section, "Land Use and Non-
point Load Analysis." In-stream removal rate is also ap-
plied to account for the various removal processes that oc-
cur in the stream (e.g., biological uptake, oxidation, hy-
drolysis, sedimentation, etc.)
-------
1 . Application of Modified Soil Loss Equation
The modified universal soil loss equation applied in this
methodology aggregates land uses into major categories and
assigns raw loading rates (K) to these major categories.
The modifying factors R, LS, P and SD are estimated based on
overall determined land use, slopes, hydrologic condition,
drainage density and general levels of observed controls.
Care must be exercised when aggregating land use/subareas.
For instance, a 1976 EPA publication, Areawide Assessment
Procedures Manual (Volume 1, pages 4-48) provides criteria
and caution on the use and application of the universal soil
loss equation. To use the equation correctly the combina-
tion of selected factors must reflect the manner in which
the parameters are associated in each subarea.
Other simplifications in the modified soil loss equation
include the following:
(A) The rainfall factor "R" is set equal to unity "1"
for the calibration year. Other years simply as-
sume a magnitude of R based on the ratio of their
rainfall to that of the calibration year.
(B) The slope factor "S" and the length of slopes "L"
are calibrated to a single "SL" factor.
(C) The soil erodibility factor "K" includes cropping
factors for typical land uses and is expressed as
the basic raw loading rates in Ibs/acre/yr.
To facilitate application of the modified soil loss equation
to typical 208 areas, the following additions were made:
(D) The erodibility factor "K" is also applied as a
raw loading factor to typical urban areas to rep-
resent scour of sediment loads from impervious
surfaces. These are based on measurements of
sediment loads from typical urbanized areas.
(E) The erosion control practice factor (P) was di-
vided into two separate factors: natural controls
(P) and man-made controls (K2). This disaggrega-
tion allows the user more freedom to determine the
effectiveness of man-made controls.
49
-------
2- Calculation of Nonpoint Source Loading
The raw runoff factor (K) developed previously and the
runoff modifers R, LS, SD, P and K2 are assigned to their
respective land uses within the subareas defined.
The resulting loadings from land uses for each subarea are
summed to provide the overall total annual noripoint source
loading at each of the summation points as shown in the flow
diagram of Figure 4.
(S)E
n
Z
K2(i) KN(i)
where:
(S)E
n
R =
K =
LS =
SD =
sediment loading in #/yr for subarea
land use category (1 through n) (see p. 22
for derivation of categories)
area of land use within subarea in acres
(Ibs/acre/yr) is the product of the following
terms: R, K, LS, SD and P.
(Dimensionless Ratio) factor expressing
ratio of annual rainfall intensity to that
for the year of calibration.
(Ibs/acre/year) raw sediment runoff rate for
each land use category. Using measured
runoff from typical land use areas and
literature values as modified by intensity
of land use and construction activity.
(Dimensionless Ratio) expressing the effect
of slope steepness and length.
(Dimensionless Ratio) expressing function of
sediment reaching streams. Factors consider
proximity to receiving streams arid soil runoff
potential. Runoff potential of impervious sur-
faces is assumed very high.
50
-------
0
c
0)
c
CD
C
LU
(U
E
O
>
c
0)
OQ
0)
O
LAND USE (LI)
SUBAREA (!)
I
K-FACTORS
K1 - Sediment Run-off Rate (ibs/acre/yr)
K2 - Land Management Cohtrols U reduction)
KN - Ratio Load (N) to Sediment (Dimensionless)
K3 - Stream Reduction Rate (% Reduction/Mile)
FIGURE 4
EPA LAND USE LOAD
DETERMINATION FLOW DIAGRAM
51
-------
P = (Dimensionless Ratio) reduction factor ex-
pressing percentage of sediment reaching
stream as a result of natural existing sed-
iment or erosion control factors. Best man-
agement practices are encompassed in the
term K2.
K2(i) = (Dimensionless Ratio) reduction of stream
load resulting from existing and planned
future implementation of land management
controls.
KN(i) = (Dimensionless Ratio) ratio of concentra-
tion of parameter (N) such as nitrogen,
phosphorus, or metal to that of sediment
for each land use. This term is set equal
to 1.0 for sediment runs.
The summed annual nonpoint source loadings from each subarea
reaching the stream are reduced by the factor K3, which
represents the rate of reduction of pollutants within the
stream due to settling, biological consumption, etc. All of
the loadings from each subarea are summed to present an
annual pollutant load at each of the three summation points.
A typical mass balance computer program output is shown in
Appendix B. The first two pages show the land use areas and
the loading factors employed for each subarea and land use,
and the third page shows the summation of resulting loads
for each subarea and summation point. This same format is
used to present outputs for each pollutant run and each con-
trol option run, thus simplifying a comparison between load-
ings and loading factors. The methodology for using and ap-
plying the computer mass balance program for a nonpoint
study is explained below. A typical case study, the Lower
Wissahickon Watershed, is presented in Section IV.
3. Calibration of Sediment Loads
Using the loading factors previously developed for the
calibration year, a mass balance of sediment loading is
performed by using the simplified mass balance computer
52
-------
program or desk top calculations. The resultant calculated loadings at
summation points are then compared to the measured loadings at these same
points. Differences are resolved and the land use factors are adjusted to
provide a calculated loading that matches the measured loading. These
adjustments are not arbitrarily made to provide a calibration fit. Rather,
the factors and loadings of each subarea are re-examined with the aim of
uncovering discrepancies in any of the loading factors. In no case should
unrealistic loading factors be employed merely to obtain a fit. When
the calculated and measured loadings at the summation points are in reasonable
agreement (within 10%), all factors should be increased or decreased pro-
portionately to provide an exact match of the calculated-to-measured-year
loading. This mass balance then constitutes the "best-fit" calibration of
sediment loads. See the case study (page 137) for a further explanation
of the application of calibration procedures.
4. Verification of Sediment Loads
Verification is based on matching calibration results to verification
sediment loading data for a year separate from the calibration year or
from data from a separate storm event within the calibration year. Careful
consideration of the margin of safety that surrounds and limits unqualified
verification should be taken into consideration. Although decisions
must be made in the near term an appreciation for and an effort to improve
the margin of safety for the data should continually undergo review and
reevaluation.
(A) Separate Year Verification:
Using the loading factors from the best-fit calibration of sediment loads
and a rainfall factor "R" corresponding to the rainfall of the verifi-
cation year, a mass balance is run. Appropriate adjustments to compensate
for differences in land use between the two years may be required.
The calculated loading at the summation point is compared to the veri-
fication loading measurements. The degree of fit to be expected of
calculated versus measured loadings depends on the quality of the measured
verification data and the confidence the user has in the calibration land
use factors. With reasonable confidence in verification data and land
use factors, a degree of fit within 20% can be expected. If the fit
is outside this range, both the calibration "best-fit" and the verification
data are suspect. Both should be reexamined to determine the cause of the
misfit. Calibration "best-fit" may require a re-adjustment in the loading
53
-------
factors. Both calibration and verification results must be
satisfied with the same loading factors employed, with the
exception of the rainfall factor "R" which is, of course,
tailored to each condition. No unreasonable factor should
be employed to obtain best fit.
(B) Separate Storm Verification: The procedure for verify-
ing separate storms is the same as it is for separate cali-
bration and verification years, with the exception that the
rainfall factor represents the ratio of the single verifica-
tion storm rainfall to the total rainfall of the calibration
year. The degree of fit between calibration and verification
of a single storm rainfall could be somewhat looser than
that derived for separate verification years in that errors
of data for individual storms are averaged out for a total
composite year.
5. Calibration and Verification of Other Pollutant Loads
Using the calibrated and verified sediment loadings model,
other pollutant parameter loadings can be computed as a
function of the sediment loading by using the KN ratios
(ratio of measured pollutant load to measured sediment load)
previously developed for each land use. The computed load^
ing of each pollutant is compared to the measured loading,
and discrepancies are resolved. Adjustments can be made in
the KN ratios, but it is inadvisable to make changes in the
basic sediment calibration model loading factors. Potential
errors in the KN ratio are listed below:
* Errors in determining the KN ratio (calculations
and supporting data)
Existence of significant soluble pollutants (see
previous discussion of KN rates)
Failure to recognize the propensity of pollutant
to be transported by finite size range of sedi-
ment particles
It is advisable to concentrate adjustments on the first two
potential errors. It is beyond the scope of the technology
presented in this report to include the complex analysis of
the latter potential error sources. The basic sediment mass
54
-------
balance model does not have the capability to include this
factor, and adjustment of KN ratios based on this factor
should not be attempted unless a sufficient data base ex-
ists.
J. WATER QUALITY IMPACT ANALYSIS
The primary purposes of this analysis are to (1) verify and
modify the preliminary identification of the water quality
problems, and (2) formulate allowable pollutant loadings to
meet applicable water quality standards. These waste load
allocations should be divided into point and nonpoint source.
Based on these allocations, nonpoint source control options
should be evaluated to determine (1) what controls will ef-
fectively reduce pollutant loadings to the allocated amounts,
or (2) whether nonpoint source control measures will be able
to produce the required pollutant reductions. If nonpoint
source controls are insufficient to produce the required
pollutant reductions, the maximum estimated pollutant reduc-
tions should be applied to the study area, and the water
quality analysis (mass balance, model, etc.) should be rerun
to project the resulting water quality based on application
of the nonpoint source controls.
In many instances a reiterative process will have to be ap-
plied. For example, a given set of controls should be se-
lected. The resulting pollutant loadings based on these
controls should be input to the water quality analysis to
project future water quality. If the projected water qual-
ity does not meet standards, additional controls should be
applied, and the water quality analysis should be rerun to
determine revised water quality projections. This reitera-
tive process should be performed until (1) the projected
water quality meets the standards, or (2) some intermediate
water quality level is projected from the application of the
best available nonpoint source controls.
The impact of nonpoint loadings on water quality and biota
are determined in three steps:
55
-------
Impact on chemical water quality
Impact on stream bed and substrates
Impact on biology of receiving waters
1. Impact on Chemical Water Quality
Nonpoint load impacts on chemical water quality must be
viewed in terms of short-term and long-term effects. In the
short-term during and immediately following storms, signifi-
cant increases occur in total pollutants present; however,
these are coupled with corresponding increases of dilution
waters; therefore, short-term concentrations of pollutants
may not cause short-term water quality problems. The typi-
cal behavior of the streams under consideration are best
determined by examining hydrographs and pollutographs con-
structed as a result of storm monitoring at the summation
points.
Because the modeling technique allows a separation of point
source loads from nonpoint source loads, these hydrographs
and pollutographs can be modified to show the increase or
decrease in concentrations that would occur for future land
use projections and controls. These hydrographs and pollu-
tographs will be at best approximations because new develop-
ment would alter peaks, delays and general shape of the
hydrograph and pollutograph as well as the total overall
loads. Nevertheless, they should be valid to illustrate
relative significance of impacts of different land uses and
alternatives.
The long-term impacts of water quality can be ascertained by
determining the average concentrations of pollutant loads
based on total annual loads (both point and nonpoint) and
total annual flow (see Lower Wissahickon Watershed Case
Study for typical analysis). Accompanying field studies can
reveal specific nonpoint problem areas where the nonpoint
sediment and pollutant loads accumulate and cause benthic
problems and exercise a BOD demand and resulting low dis-
solved oxygen during low flow, steady-state conditions.
Another potentially significant long-term effect of nonpoint
pollutants is in downstream receiving waters where eutrophi-
cation may occur. In these cases, the algal growth and
limiting factors must be determined and mass balance tech-
niques must be employed to determine the effects of the
nonpoint source loads (see Impacts on Biota.)
56
-------
2. Impact on Stream Bed and Substrata
The impact on stream bed and substrata is mainly determined
through field surveys. This impact results from separate
but interdependent effects: (1) scouring of stream banks
caused by increased flow, and (2) deposits of sediment and
other pollutants in dormant pools or basins within the
stream. Both of these are aggravated by increased develop-
ment and impervious cover which causes higher peak flows and
pollutant loads. The effects of stream bed problems on
steady-state water quality characteristics and the destruc-
tion of natural habitat for biota should be studied by field
surveys, preferably following storms (see the results of
field studies conducted during the case study).
3. Impact on Biota
The chemical water quality data and stre
-------
The best way to eliminate these factors is to compare data
from areas comparable in habitat availability, physical
conditions, and data collection techniques, but with poten-
tial nonpoint source pollution affecting only one area. If
the areas are comparable, major differences between them are
probably a result of nonpoint source pollution.
In addition, downstream waters should be analyzed to deter-
mine if eutrophication, or the potential for it, is present.
If eutrophication problems exist as evidenced by algal
blooms, in-depth water sampling and algal identification
should be conducted. Factors and pollutants which provide
limiting ingredients for algal growth should be ascertained.
Finally, mass balances of nutrient enrichment should be con-
ducted using the point and nonpoint loads resulting from the
model. By using model runs for projected land use and con-
trols, the general impact of development and controls on eu-
trophication can be assessed.
4. Summary of Nonpoint Loadings on Water Quality
The observed impacts of nonpoint loadings on chemical water
quality, stream bed and biota are summarized and analyzed
to assess a cause-effect relationship. A discrimination
is made between those problems caused by point source dis-
charges and nonpoint pollutant runoff and wasteload alloca-
tions are made for point sources and nonpoint sources of
pollutants.
The point source impacts are most readily identified by
using steady-state modeling. In many 208 areas, steady-
state models may already exist. These models include the
effects of point sources during low stream flow conditions
when the effects of point sources are most prevalent because
of the absence of dilution waters. The steady-state models
also can offer some insight into nonpoint source pollution
effects by identifying existing benthic deposits, which may
be a result of settled sediments from previous storms.
These benthic deposits exert an oxygen demand during later
low flow periods. Areas of large benthic demand are readily
determined by examining steady-state model results. Steady-
state models will also reveal portions of the stream in
which eutrophication or high algal blooms occur. These,
too, can be the result of nutrients washed off of the land
during previous storms.
58
-------
If there are no steady-state models and budget or time con-
straints prohibit the development of such models, a less
comprehensive mass balance approach can be used wherein a
balance of point source pollutant inputs (determined from
NPDES permits) are compared to the load measured at the sum-
mation points during low flow conditions. Care must be ex-
ercised to include the effects of the loss of point source
pollutants in the stream through consumption or settling.
A description of steady-state modeling techniques and mass
balance techniques and their capabilities and shortcomings
are presented in the EPA publication "Simplified Mathemat-
ical Modeling of Water Quality," March 1971. Point source
control options can then be explored to assess the degree of
improvement possible or practical under point source control
management.
Once the total effects of point source pollutants are ascer-
tained, the remainder of the water quality problems can be
attributed to nonpoint sources. Wasteload allocations of
nonpoint sources can then be assessed. Some water quality
problems can easily be ascribed to the effects of nonpoint
sources, such as scouring of stream beds and sediment depos-
its. Other problems, such as nutrient build-up and areas of
large benthic demand, are more difficult to assess because
they could originate from point sources, nonpoint sources or
a combination of both. Benthic deposits of oxygen demanding
water originating from point sources usually occur a short
distance downstream from the point source discharges. An ex-
amination of the hydraulics, point source locations and ben-
thic deposit locations can reveal a cause-effect relationship,
It is extremely difficult to determine the exact magnitude
of the reduction of nonpoint source loads that would meet
receiving water quality standards. The complexity of the
many stream flow regimes encountered in low flow and storm
flows of varying intensity result in pollutant dispersion
that is often too difficult to assess. For example, a storm
of one intensity might wash off land pollutants to be settled
in a given stream bed location, while a storm of a more in-
creased intensity would result in scouring that deposit and
washing it into receiving waters further downstream. The
answers to such questions are usually beyond the capability
of even the most sophisticated storm modeling techniques.
Therefore, the methodology presented herein is limited to
59
-------
identifying the existence of nonpoint source problems, their
relative order of magnitude and their isolation to given
pollutants and probable sources. The mass-balance modeling
approach provides a tool that enables the user to attribute
problems to specific land use sources and, through the use
of the sensitivity analysis, determine the most cost-effec-
tive nonpoint management alternatives which would reduce the
nonpoint loads to within acceptable limits of the overall
assimilative capacity of the receiving waters.
K. SENSITIVITY ANALYSIS
The sensitivity analysis provides the framework for deter-
mining the land management control techniques to be applied.
Data from the sensitivity analysis are helpful in providing
answers to such questions as:
Which land areas are least sensitive to development?
Which land uses and subareas contribute the most sig-
nificant nonpoint loads?
Which land management controls are most promising?
To gain these sensitivity criteria, several mass balance
variations of the sediment calibration condition must be
run. Separate mass balance rates are made incorporating the
following changes:
(A) The slope lengths (SL) are assumed to be at the
minimum value (i.e. , all slopes theoretically
less than 2%).
(B ) The sediment delivery ratios (SD) are assumed
to be 1.0 for each land use and subarea (i.e.,
all eroded sediment is delivered to the stream).
(C ) Natural erosion control factors (P) are set to
1.0 (i.e., all natural sediment control factors
are eliminated).
60
-------
The man-made control factors (K2) are set to their
best practical value (i.e., all existing controls
are upgraded to their best practical implementa-
tion) . These may be considered separately for
each land use or run in combination with controls
for land use groupings. See Lower Wissahickon
Watershed Case Study for examples.
The loadings from these computer runs are then extracted and
summarized to form sensitivity analysis criteria as shown in
Table 5, which summarizes loads at each mass balance point.
Loads are shown as a function of each of the loading factors
and in terms of total load change in Ibs/yr, loading rate
change in Ibs/acre/yr and the percentage of load change for
each loading factor. The criteria presented in this table
can be used to determine overall basin problem areas.
Similar tables expressing sensitivity criteria for subareas
and land use levels can be derived to identify land uses and
subareas which are significant contributors to pollutant
loadings. These criteria can then be related to pollutants
and land uses which contribute heavily to identified water
quality problems. Together, these criteria provide the
basis for determining:
Basin-wide problem areas
Isolation of problems to subareas
Isolation of problem land uses within each subarea
Which areas, if developed, will provide the least
nonpoint impacts
Where development on steep slopes would be less
conducive to nonpoint problems
Where sediment basins might most effectively be
used to reduce sediment delivery
Where and which controls will be most cost effective
in reducing pollution loads for existing and projected
futures
61
-------
tn
EH
H
O
w
u
CQ
cn
CN
Q)
CP
C
U
O
in
W
CQ
EH
H
H
EH
H
CO
W
CO
o
EH
U
<
O
S3
H
P
O
CU
4->
U
0)
QJ
EH O
CU rd
to
XI
to
XI
rH
0)
rd
rH
rd
CM
X
X
CM
XXX
X
X
X
X
X
X
X
X
X
rd
to
-^ rd
H
-" tO
C
0
H
4-»
&!
H
^
U
to
Q)
P
Q)
tr>
£
td
r]
O
rQ
rd
O
1
Q)
O
rH
to
<4-l
O
c
O to
H x;
4J 4J
U CP
c c
3 Q)
U-l H
a
iment
to -d
rd 0)
to
to
QJ y-i
tP O
fl
rd C
XI O
U -H
H-)
Ti U
id C
O 3
O
H
_p
rd
M
>i
rH
QJ
>
H
rH
(D
rd
H
to rd
rd M
10 4-> 10
C ro M
O fi O
H 4J
4-> 4-1 O
U O rd
!3 4-1
r(3 fIJ
Q) O rH
VH -H O
4-1 rH
T! 0 4->
rd C C
O 3 O
J m u
rd
to
rd
to
£
O
H
4J
U
3
rfj
Q)
rH
rrj
rd
O
a
-H
4-1
to
H 4->
X C
Q) Q)
e
4H QJ
O Cn
rd
a c
O rd
H S
O 4J
C to
d Q)
M-l CQ
to
QJ
U
H
4J
O
rd
rH
Pu
1
U
Ti
0)
rH
t3
rd
O
rH
g
3
g
H
X
rd
S
rH 4-1
Q) O
c to
d 4J
H
Q) e
X! rH
H
tO rH
to rd
O 0
Qj-H
4J
C 0
O ro
H 'H
4->
C
g
QJ
CP
rd
C to
rd rH
S 0
rH
4J 4J
to c
QJ O
PO U
tn
P
to
CU
CM
CN
to
CU
rH
rd
C
H
rd
-P
0
O
-P
O
x:
m
o
c
o
-H
4J
rd
c
rd
X
CD
fj
rd
S-i
O
4H
!>-,
T)
j3
-P
to
QJ
to
rd
O
Q)
4J
14.1
o
o
in
rH
I
00
rH
CO
0)
rd
p.
Q)
QJ
tn
rH
-P
c
H
O
£4
£
O
-H
-P
rd
g
g
rj
CO
-P
rd
T)
rd
O
rH
rH
rd
-P
O
4J
O
rH
QJ
-P
QJ
e
rd
rH
rd
m
o
4J
u
QJ
4H
4-1
w
CN
rd
Q)
rH
rd
r*1
TJ
4-1
to
-P
rj
O
tP
O
J_J
r;
4J
to
Cu
jg
CQ
Q)
rH
Q
rd
rH
H
>
rd
4-)
to
Q)
XI
D1
g
-H
to
4H
0
4->
0
Q)
ii i
ii i
w
ro
62
-------
L. CONTROL PLAN DEVELOPMENT
The value of the model and methodology in the development of con-
struction and other nonpoint source control plans should be apparent
after examining its output. The output information includes summation
point loads, total in-stream loads associated with land use categories of
each subarea, and the sensitivity of these loads to any variable changes
in the USLE-modified equation. Of primary interest is the sensitivity
of the model to changes in the load reduction factor, which takes into
account controls placed on new land development, existing development, and
agricultural activity.
The model can be used to assess the effectiveness of the existing control
mechanisms on in-stream pollutant loads. It can also be used to evaluate
which land use categories are the major contributors of such loads under
controlled and uncontrolled conditions. The data can be produced as
total load associated with each land use for the study area or for each
subarea. It can also identify the nonpoint source load per unit of land
(e.g., Ibs/acre/yr) such that the relative pollutant load generation of
one land use can be compared to another.
If information on the effectiveness of physical controls is available,
it is possible to test a "promising control system" in the model to ascertain
its in-stream effects. The marginal benefits in effectiveness between
promising alternative systems can then be compared with the direct and
marginal costs associated with each system.
Although this methodology is based conceptually on a relatively simple
equation, sensitivity of the model to the quality of data always leads
to questions of accuracy and degree of error. Of particular importance
are data on loading factors and effectiveness of controls. If such data
are inaccurate, the identification of major contributing land uses, and
further, the development of control systems for these land uses, may be
inappropriate. As a tool for policy analysis and program development,
caution is necessary in judging the accuracy and appropriateness of
the methodology. It should be stressed that even the best analyses will
only yield useful approximations which may allow better management of
solutions in the near term. More precise tools will evolve if continued
technical emphasis is given to calculating land use water quality
relationships. High quality data will often need to be collected over
a protracted period to permit a high degree of confidence with the data
base.
63
-------
Literature on new land development controls and their effec-
tiveness is expanding. Nevertheless, where possible, data
from available literature should be supplemented by actually
testing the effectiveness of controls as applied in the
study area. This also applies to controls associated with
agricultural production and maintenance construction.
1. Use of Physical Controls
The range of alternatives for control plan development is
broad. A myriad of plan alternatives can be created for
controlling pollutants generated by new land development,
existing development, or agricultural production. The con-
trol plan should provide a blend of physical controls to
achieve the following objectives:
* Protect existing ground cover
Control runoff velocity
Filter runoff
Contain sediment and runoff
As an aide to understanding how alternative plans (using
different physical controls) may accomplish similar objec-
tives, refer to Table 2. This table indicates the objec-
tives accomplished by a representative list of physical
controls. A choice can usually be made of the control to
accomplish one of the objectives. Of course, the effec-
tiveness of each control varies and the selection of plan
components should be based in each case upon the require-
ments of the site, degree of sediment (or other pollutant)
removal required, and marginal benefits and costs of alter-
native plans.
2. Use of Regulatory Controls
The use of on-site physical controls is usually required
by a regulatory control such as, in the case of new land
development, a subdivision ordinance or erosion and sedi-
mentation control ordinance. However, there is a problem
in specifying physical control usage in ordinances and
other regulations. To achieve a high degree of effective-
ness in pollutant load reduction, physical controls should
be chosen and designed for the actual site under considera-
tion. While certain standard physical controls are commonly
64
-------
used, each must be tailored to suit the topographic and
hydrologic features of the site. In an effort to provide
the flexibility required in choosing physical controls, a
trend has developed in which subdivision and other envi-
ronmentally sensitive ordinances have required the appli-
cation of "performance standards." These standards trans-
late objectives of the ordinance into quantitative goals.
An example of a performance standard would be no net in-
crease in runoff from newly developing sites, or a mini-
mum tonnage of sediment loss per unit of time during con-
struction. The performance standard approach leaves the
choice of physical controls and their implementation to the
developer, who is responsible for meeting the performance
standard.
3. Implementation of Regulatory Controls
In order for regulatory controls (and the physical controls
which they may require) to be considered effective in reduc-
ing loads, they must first be implemented and enforced. For
this reason, implementation is a key variable to be consid-
ered in this methodology. Proper implementation of regula-
tory controls by municipal governments implies that the
controls are consistently applied when necessary with a
minimum of waivers and variances and that developers imple-
ment the required physical controls. When using this meth-
odology, the planner must take into account the probability
of implementation in determining the effectiveness of regu-
latory controls. As an example of such a consideration, a
sedimentation and erosion control ordinance, specifying
physical controls determined to achieve 95% reduction of
sediment loss, is less effective as a regulatory control if
implemented by the municipality in only 75% of necessary
cases. In fact, this regulatory control, under these con-
ditions, might be 71% effective (.95 x .75 = .71).
As a guide, regulatory controls on the municipal level are
more apt to be implemented if they contain provisions such
as the following:
Performance standards based on water quality
goals
65
-------
Water quality monitoring requirements
Requirements for site inspection by knowledgeable officials
Procedures for enforcing the regulations, including imposition
of sanctions
In addition to such ordinance provisions, implementation is generally
encouraged where technical and program assistance is offered by appropriate
government agencies such as planning commissions, soil conservation
districts, health departments, and special municipal review boards such
as environmental and architectural review commissions. Where such assistance
is received by local regulatory authorities, the technical capabilities
of municipal managers and their ability to understand control increases,
leading to a better understanding of environmental problems, appropriate
solutions, and consistency of implementation.
In determining the effectiveness of regulatory controls in the study
area, thought should be given to the qualifications noted above. While
these qualifications do not lend themselves to quantification, the use
of good judgment should be sufficient to assess their impact on control
implementation.
4. Regulatory Controls and Best Management Practices (BMP)
For construction, agriculture, and other pollutant-generating activities
addressed by this methodology, the 208 planning agency should prescribe
best management practices to be applied for the study area, county, or
watershed affected. In some municipelities, the recommerded BMPs will
already be in effect. In many cases, BMPs already exist in many munic-
ipal regulatory controls. This is particularly true in the case of
regulating construction activity, where many well designed subdivision
and land development ordinences have required physical control techniques
fitting the definition of BMP.
As a means of implementing BMP, 208 planning and management
agencies should encourage municipal, county, and state gov-
ernments to include their BMP recommendations as amendments
or changes to existing regulations. Since BMPs are most ef-
fective when consistently applied throughout the study area,
these efforts should be directed at all municipalities or
through county government if statutory authority permits
such regulation at that level of government. This issue is
particularly important in study areas composed of large
numbers of municipalities where inter-municipal cooperation
is often difficult to achieve.
66
-------
Two types of regulatory controls appear well suited as BMP
mechanisms. The first are municipal ordinances which can be
adapted to include BMP as requirements. If included in any
of the types of ordinances listed on page U2, the municipal-
ity retains strong control through ordinance enforcement.
State legislation permitting municipalities to pass local
ordinances should be reviewed to determine limitations or
qualifications in this area.
The second type of regulatory control suited for BMP are the
state permit issuance programs. On the state level, permits
are often required for wastewater discharge facilities,
water supply, sewer extension, on-lot disposal systems,
watercourse encroachments, erosion and sedimentation con-
trol, mining activities, and various other activities. Most
state agencies which issue permits have the authority to
attach conditions to permits which could include BMP for
activities requiring them. In such cases, guidelines for
permit issuance, generally published by such agencies as
instructions to permit applicants, could specify types of
BMP to be required as permit conditions. The enforcement of
BMP, as elements of permit programs, is the role of the
state government, or other state designated levels of govern-
ment.
67
-------
IV
CASE STUDY: APPLICATION OF METHODOLOGY
A. INTRODUCTION
The methodology presented in Section III was applied to a
watershed in Pennsylvania to present a case study of the
methodology. The purpose of this case study is twofold:
To illustrate the applicability of the methodology
for the analysis of land use-water quality re-
lationships
To demonstrate how the methodology can be applied
to a specific watershed
To accomplish this it was necessary to select a small
urbanizing watershed having a wide range of water quality
and nonpoint source problems. The Wissahickon Watershed in
Pennsylvania was selected because it fit the selection
criteria of (1) an urbanizing area, (2) single and multi-
family residential area, and (3) considerable new develop-
ment. Its salient characteristics are:
It covers a broad spectrum of land use. Its head-
waters lie in rural areas that exhibit a mixture
of agricultural use and suburban growth.
Its mid-section is suburban, with a significant
mixture of heavy and light industry, a high growth
rate and accompanying new construction and road
building.
Its lower section lies within the city limits of
Philadelphia in a region of high population den-
sity and old established neighborhoods where
little new land development occurs, but where a
large share of the city's urban renewal and re-
development takes place. This section does have
one distinguishing feature that may not be typical
of most city environments: the stream threads
68
-------
through Fairmount Park, where a relatively wide
adjoining area of natural woodlands serves to re-
tard nonpoint pollutant runoff. The buffering
zone occurs to a much lesser extent in the mid and
upper watershed.
The watershed drains an area of 64 square miles
and has a main branch 22 miles in length. Promi-
nent features include its urban-suburban interface
and high rate of development from the mid-section
of the watershed to its head waters. The Wissa-
hickon watershed is presented in Figure 5.
B. INVENTORY AND EVALUATION OF EXISTING DATA
A description of the case study data sources and their
functional uses are presented in Table 6. These data were
evaluated to determine land use characteristics and all of
the land use loading factors required to apply the mass
balance model to the Wissahickon Watershed. Calibration and
verification data sets were also derived from these data,
along with assessment of water quality problems. Deficien-
cies in these data were augmented by field surveys and a
literature search of parameters and criteria from studies
conducted in similar urbanizing areas. The applicability
and use of these data are explained in detail in the subse-
quent paragraphs.
1 Water Quality Problem Analysis
Review of existing water quality and related data indicates
that one of the primary water quality problems in the study
area is steady-state nutrients. Although the high phospho-
rus levels in the Wissahickon Creek produce some localized
or transient aquatic plant problems, the primary problem is
produced in the Schuylkill River downstream of the Wissa-
hickon Creek.* High phosphorus loadings from the Wissahick-
on Creek and the Schuylkill River combine to form eutrophic
* The Wissahickon Creek is tributary to the Schuylkill
River, which is used as a water supply for the City
of Philadelphia.
69
-------
FIGURE 5
MAP OF STUDY AREA
70
-------
TMJLE 6
WI3SAHICKON WATERSHED CASE STUDY DATA SOURCES, DESCRIPTION AND FUNCTION
Da t ft Sourca
Philadelphia Water Department
1970-1974
Wissahickon Water Quality Hodel
C. Gabriel 1974
United States Geological Survey
Water Resources Data for
Pennsylvania: Part 2.
Water Quality Records 19C4
and
United States Geological Survey
Kater Resource* Data for
Pennsylvania : Part 1.
Surface Water Records 1964
and 1968U)
U.S. Depsrtrnent of Com-
merce Pr« cipltJtion Pecords-
Pennsylvania Annual Summary :
1963, 1964, 1967, 1963
Northern Virginia Planning Dis-
trict Commission Quarterly
Report Oqcoquan/Four Mile
Ho n point Source Correlation
Study, October 1976
Vissahickon Watershed Waste-
water Managenent Study,
betz Environmental Engi-
neer* , 1976
COWAMP/208
Critical Slopes Map, Water
Quality Management Plan
Study (Chapter IV)
USD*
Montgomery County Soil
Survey (196?)
Philadelphia County Soil
Survey (1975)
COW AMP/203 Storm Drainage
Facilities Map
Montqomery County Planning Cora-
issioni Montgomery County
Land Uie Map
t?is««hickon Watershed Develop-
ment Guide
Delaware Valley Regional Plan-
ning Commission Year 2000
population forecasts
Kunlcipal zoning, subdivision,
id other ordinances
0.8. Geological Survey
Topographic Maps, 1965,
1972 revisions
Aerial photography, 1973-75
photographs
Description
Water quality dataj monthly grab samples;
IS water quality parameters, Bells Mills
station and mouth of Wissahiciton Creek
Steady-state Water Quality Model: Dis-
solved Oxygon, BOD and listing of point
sources and their effluent characteris-
tics* 10 parameters
Sediment load during storms at Fort
Washington Station, Wissahickon CreeJc
Daily flow records: Wissahicken Creek-
Fort Washington, Bells Kills and Mouth
Rainfall Records, North Philadelphia Air-
port (1967, 1968) and Phoenixvillc, Pa.
(1963, 1964)
Storm loads for various land uses and
rainfall: sediment, BOD, pnosphorus,
arimonia, TKN, nitrogen
Biological survey of Wisaahickon Creek,
6 stations (1972!
Kap of slopes in excess of 12% grade
Hydrologic groupings of soils, composi-
tion of soils
Hydrologic grouping of soils
Storm sewer service areas, location and
description of facilities
Land use categories
Land use and development control guide-
lines for Philadelphia portion of Wis-
ahickon Watershed
Year 2000 population forecasts by
municipality
Local regulatory controls addressing the
management of land
Maps showing topography, existing and new
development* and natural features
Aerial photographs of land use and
topography
Functional Use
Establish steady-state water quality
conditions
1. Establish steady-state water quality
conditions
2. Determine point source total load and
effects
1* Establish sediment calibration data set
(1968)
2. Establish sediment verification data set
(1S64)
1. Correlate flow to sediment load
2. Extrapolate storm tlow sediment load to
total annual sediment load
3. Establish rainfall factor (P.) baseline
for calibration (1968)
4. Establish "R" factor for verification
(1964)
1. Correlate rainfall to sediment load
2. Develop "H" factor base for 1968 cali-
bration
3. Develop "P" factor for verification (1964)
Develop *KN* ratio for various pollutants
Determine existing biological conditions
Develop slope length (SL) factor
1. Develop sediment delivery ratio (SO)
2. Develop cropland sediment load rates
1. Develop sediment delivery ratio (SO)
2. Develop cropland sediment load rates
Identification of storm sewer drainage
area and outlets
To broadly outline major land uses
To broadly outline major land uses
TO determine population and land use
changes to year 2000
To determine the relative effectiveness of
local regulatory controls
1. To develop hydrologic divisions of »ub-
basins
2* To Identify areas of new land development
3. Develop slope length factors (SL)
4. Determine drainage density
S. Develop sodimerit delivery ratios (50)
To identify land use and development
(1) Correspond* to USGS water year beginning Novoieber of preceding year and ending in September of year listed.
Sourcei Bets Environmental Engineers, Inc.
71
-------
conditions in the downstream regions of the Schuylkill
River. Excessive algal growths in this reach of the river
cause treatment and taste and odor problems in the City of
Philadelphia's water treatment plants. Nutrient-induced
eutrophication problems occur on a regional basis during
dry, low flow warm weather conditions.
A second and equally troublesome condition is sedimentation,
Sedimentation from extensive development and construction
within the watershed has destroyed much of the stream bed
and biota.* In addition, the increased storm flows caused
by increased impervious cover in the watershed has scoured
the stream banks and uprooted trees. The stream surveys
subsequently support these observations.
C. FIELD SURVEYS AND NONPOINT SOURCE MONITORING
1. Field Surveys
Two field surveys were performed during the case study. The
first was a cursory survey to provide an overview of land
uses, stream conditions and existing runoff controls prior
to establishing loading factors for subareas and land uses.
The second survey was performed after a heavy storm to
determine the effect of nonpoint loads on water quality and
biological conditions in the stream. Photographs of visible
problem areas taken during these surveys were helpful in
documenting problems and would be valuable for use in later
public meetings to demonstrate these areas.
2. First Survey
The first survey was performed on January 31, 1977. Figure
6 presents a map of the survey sites. A summary of the
findings appears in Table 7. This survey was conducted to
* Regional Science Research Institute, Environmental Study
of the Wissahickon Watershed within the City of Philadel-
phia, June 1973.
72
-------
o
c
C7)
c
c
a)
c
O
c.
UJ
0)
oa
. \/
Q 1st Survey
A 2nd Survey
FIGURE 6
MAP OF SURVEY SITES
73
-------
J3
a
e &
o TI
n -H
1 VH
0 Cl
rH rH
rH TI
TI 10 13
C X! O
HE K
3 -5
X 10
C
[K C
r- o & 10
r- o x:
rH C 3
10 v
* a ui
r-l E rH 3
rH 0) QJ 01
EH -H
>i VH I
n * a
10 VH ij C
3 10 O 3
C OJ £
10 rH
ID U O >
'O
c
3 in
OJ
VH rf.
HO rt
QJ -,&
QJ P r-l QJ
P lO QJ p
10 QJ -H H
Q [£ [i* tn
tn U) JJ
C X C
>> O G QJ S «
rH rH 3 3 O 3 TJ
rt 01 10 X) rH ( rH UH O C tO
g X IHrl 0) O VJ 10
me 01 tnUH at at jC 8
UHIO 3 C 01 > 01 JJ B W JJ O
xi oo o xi -H 3 30 n
OJ rHr-tjj at e on uj
rHtP ,
CO 01 OJ rH E OJ 3 Vj H
HrH > U) JJi 10 VH TI XI 01 JJ
0110 OVH 01 rt > 10 C
3 JJ In 01 01 X O 01 V
*TJ VJOCJJ CX tn C r) Vt
oioi oiooio) a, IH 10 10 a
JJJJ -PB '0 * XI CX
OO 10 UH Jj 01 |J 010) aUH
rHC 3 rH 10 JS 01 & X -rl oxoj oi-H jj OT w o TJ uj *o
10 01 10 QJ 10 OjrH O a 0) O « 41
en oj 3 -Hjj x: UH tojj x a < H
IP z >, x; c 01
OJ 10 C XI JJ QJ TJ
C 3 -H O (0
O g 10 TI til C O
VH ^- tji oj G O PS
>, C g < 01 X,
rH P Oil QJ 3 E
P OJ tN
O OJ -H
S3 H
o n
VH
'.' OJ
01 g X
01 3 -H
0 Z ft
rH C
a rH
C -H
01 3 X
g « 01
0 01 E
01 > QJ -H
T: O rJ
C C 10
H 10 0, 1
in ui
01 JJ
X U-l c C
C O "J rt
10 Cu rH
X) 0] O ft
c
OJ O 01 P
p -p p 01
VH E
O 0 UH p
J-J Oj O 10
o
rji tr ol VH
C C JJ EH
rl 0 0
C t 'IH tr
0) IM 10
p TJ 3J VH
X rH 01
0 3 r, 3
tj O 0)
C 'H
01 rH 3 C
E rH XI 0
O* TJ 4_i
O X) OJ IT
H x: c
0) TJ P -rl
> c .q
aj rj QJ «;
01 i 0) >i
UI p p TJ
TJ rH 'C TJ
01 .0 rH tfj
M -H 3
H p c «:
rH C 3 fN
OJ OJ it
c TI x;
C H o 01
10 I) -H P
T: a1 c -H
U W 3 10
rH 10 0)
X) O TJ
O OJ 10
VH rH OJ u
H
P S. >i C
C n) 3
01 tn E 0
EC P
H -H C
O rH JJ (0
0 rH C g
01 o 01 VH
HE cu
OJ TI O
rH 01 C
X! H 3 H
H 0 10
01 >, Q. 01
oi x; g c
Q, 10 C
IH rH 3
ClrH OC
com
o Q. g :»
H O Jl T!
P JJ C
O 10 10
H 10 3
-P VH o x;
t/J 0) rH J-1
C C OJ -H
o a xi s
o o
TJ 01
rl OJ U
0) P C
TJ QJ O OJ
C en C 3
3 .0 r-l
3 01 UH
0 0
O X!
0) rH 01
a o o
0) O t-3
Id X! TI
TJ -H 10
01 EG
D* QJ
0 -P
rH . P
01 01 OJ
> TI >
01 -H 10
as UH
C 10
3 P J
OJ
01 01 VI
3UH 15
O OJ
IH 0 c
3 JJ 3
23 K
o
XI >
10 T)
01 C
X 0) 10
C XI W
10
xi o x:
p p
x: -i-t
O Cy
XI VH 0)
0 O
IP OJ C
C. Gj OJ
3 QJ (0 C O QJ 3
O 01 3 0
c u
0) O 10 OJ
X O >
C VH PK C
nj O XI
0
tn O X
C tn o 0
o c. ;> a:
rH O rl
(0 VH 01 U
^ s
10 01 O C
rH 10 rH
g "c
01 10 O
rj OJ O
VJ p p
TI X
u) OJ
rj rj Q
O 01 Vj
OP O
^ 0
C G
OJ iH Ol - 0
VH -a UH x e >. x
10 10 VH O! id r U
£ QJ -H
H > x:
0) 0) -rl
'-J "C .C
10 s o ro JJ (0 ^
rH 10 01
P OJ 10 W
J) U UI
10 01
G VH 3 -nl TJ H
CJ P O ^ OJ £ 2
TJ to VH
o; rt
u) fi in
oi d *
VH ol c
X Q)
3 c p P
^ XI XI «
N O
H VH r
OJ »
C 01
C 3 OJ
10 0 P
VH X 10 01 H O
01 0> 01 VH rH U
01 C > 10 *H PrH
0-H 0 P. C« X-O
CrH >l 01 -H C S
OIH O rH gJJ 10 10
o o z -o) a, c xi
H VH 10 > O 0 rH
01 0) -rt rH TJ S lO
0) > 01
VH rH VH rH
p o ra
>l C rH TI
rH OJ O 01
H tn O C
g 10
rj 01 01 1-3
UH -rl rH
XI C
QJ >, a -H
,H x; oj
en rjj Cl CJ
c io -H en
H VH P 13
ol en O ol -H
0 c VH VH
NO. O O
rH O X TI
JJ EH P O TI
01 -H 10
0 3 01 0
E rH K
VJ g X)
» 0 lO 10 Cl
P iH 0) QJ p.
CO HOT
0.1 U OJ P -rH C
R eH 01 jj x
0. >i «' O
OX UJ
rH ri
OJ 3 TI
OJ O
TJ Ol d
rH
O XI 01
01 13 rH
HO rH
ID U -H
to* p
O 01
rH
01 x; 01
E 4J a
to's P
p -
01 QJ X
T) OJ
x: ! u
-P 3 Vi
XI P °
OJ C
CO 0
H UH X
O 0
I-.O -H
-j ^ x;
(TV 10
P ir
01 3 01
P
.<" -H CO
U UH ui | s E 10
fl 22 p
C X
O OJ i.
E tn 0
10 VH
3 en o
0
rH UH C
01 O O
X
>i B u
rH 10 -H
0) OJ Xi
> VH re
H -P 0)
p 01 OJ
10 CX H
H 3 S
0)
VH rH UH
a) o
rH G
H C rO
H 10 C1
P Xi JH
K O <
" rH n
01 10 VH
X VH CJ
C P P
it! C ra
XI CJ S
O TJ
TI 13
OJ C Cl
TS -H r:
0
O JJ <
3 C co
0) »
ae
QJ -H Q)
OJTI JJ
4J GJ ri
to oj to
VJ *J QJ -H O -H
fl a > 01 VH p
i-H 01 01 "H C
0) 01 Q rl 01
H H tn*O
x; uj c -ri
JJ o) . o >o
H Xi TI
JJ U JJ 01
3 >! 10 O C
OX: 01 rH >,
x; a, ri 3
en nj « o
3 rl CO
O tn -H QJ H
vi o x: xt u
P o' T) a
H QJ
JJ TJ
O 0] ol
rH X C
c 10
OJ 10 1-1
HX1
O !
10 tr
1 C 0)
TJ 0 tr
VH --H 'U
H lO -H
Xi H
P TJ CQ
1 C
QJ 10 T)
C 10
o « o
01 K
JJTJ X
C-H 0
01 VH O
E XI 0
a. c
0 VH aj
rH OJ K
> C .P
CJ 3 10
TI
O X
r-l OJ QJ
0 JJ OJ
-HO VH
JJ C O
c
oj 01 s:
TI 3 O
-H O X
01 rH O
OJ UH
n H x;
OJ ffl
rH O 01
H -H
g H B
iO QJ
UH JJ <
10 Ol
QJ 3 w
rH E
CO P
to 01 to
JJ rH O
3 3
O 0 C
XiO -H
3 01
0 v, tr
H JJ 01 TJ
x; 10 jj -H
P rH C rl
UH QJ ft)
-~ u
0) >* 0)
P rH tJl X
o -H r; -H
rH <0-H ft
M (X
iy o) £X P
VH c: O 3
u QJX; o
1 >
H rH >, 01
a) rH xt c
P-H n H
In jJ iO 3
10 01 oj in
3 C
tr >, H
1 Xi g 10
OJ CX O 0)
C rO H C
O VI UH
" Ol X
OrH o;
p a 10 oi
C O -H U
C' EH O w'
E H
£1« OJ Ci
o g o
H OJ g X
OJ rH O O
>XI O -H
oj H x;
TJ 01 TJ 10
-H c o:
rH -H
rl 01 rH &
n oj 10
10 Dj H r<
UH H JJ O
CX C rH
OJ O/ ^
rH VJ TI
Oi 0) -H 0)
ex; o) jj
H p 0) H
to O VH tn
H 0)
O 01
c n
01
JJ O
X JJ
0)
0>
VjrH
01 XI
JJ H
C JJ
oi o,
0 0)
o
C 3
H a
a
ax:
o o
x: «
to oi
IH
H -rl
c
o e
n «
UH 0)
IH
to JJ
0) 0)
CX
rl O
!X JJ
o>
c
Xi-H
O tn
P H
01 1)
n .c
JJ O
0) OJ
H
0) T)
H
Xi U>
JJ-rl
vi c
01 O
TJM
VH JJ
O 10
XI JJ
0)
0)
TJ tn
c c
«J-rl
H CX
jj e
3 a
74
-------
TO
3
E
*^,
rt
XI
o
u
a
»
01
tn
TO
H
M
111
c
H o
01
S.S
J-1 >i
3 B
O
ft) TO
H
0) S
t
O r<
01 01
C
- C
01 10
a x
H U
CO
&
4^ W
c c
3---I
o tn
U flj
.Q
,H
* .c
M U
.3 **
M n)
U
in >i
^ nj
c ?
XI tr>
H
nx
01
TO G
o o
O r-l
£ 4-1
O
H
*
N
rH
"^
10
c
01
P
i
rH O JJ
C rH C
E 01 3
m > o
a, 01 o
o
H rH
10 >t a
E 0) 3
10 VI
E o
3 H >,
K « rH
S co en
0 0
rH
S X JJ
C 01
n£ (0 01
IN XI t*-l
rH
11 TO O
01 V£>
Ol TO 1
p O O
H O U"t
W S
W
xi a
to 3
co TO
a
o x
rO P
a> e
JJ 0
H 14
3
10 a
to
X 01
c
10
XI I-l
JJ O
to B
10 3
0) M
E rH
O 10
H
-~ JJ
01 B
01 10
JJ p
01 01
iO XI
S 3
to
to
ss
tn 3
01 TO
XI 0
CO h
n) a
OTO
O O
P U
TO rO
01 01
p H
I-l H iO
Q O
rH CU CO
XI 01 -H
S §
1 CO
01
Ol JJ X
tJ ME
TO 10 10
H SXI
PQ rH P
10 tn
H H 01
01 01*
>H rH
P O 10
X EC
01 3-H
01 O C
u E B
0 3 «1
O 14 B
C O> O
O HUH
X 10 O CO
0 rH p 3
H a rH
x a: 0) o
10 U E
to no
10 it CJ
-H 01 JJ
S X C TO
(4 H C
11 XI
iH 01
if TO B P
0) 10 rH
01 TO 0) 3
P OH in
-r-l OP 01
u} rs w ft
91
rH JJ
(0
< C
O-r<
M
01 10
1 01
O H
u a
!£
o a
CO O
TO TO
B B
10 n)
-TO
CO 01
JJ 0.
E O
g 01
a >
O 01
rH TO
at E
> 3
0)
TO p
B
rH 10
10 O
-H -H
E -H tn
01 B E
TO tr O
H -H -H
01 01 p
0) H
k 0 0
01 a
E H
10 10 H
xi oi
H V -P
a H m
xi 10 »
3 TO
CO 0) 10
%4 01
H OX
ox
H 01
rH JJ
H
H P H
10 C 10
6 0) «)
H g E
H a
a o TO
rH B
TO 0) n>
0) > rH
10 0) g
H TO H
H 10
R ffl
O -H TO
U H B
P 10
01 10
-H 3 0)
TO P
TO E ta
0) -H 01
X H
to TO O
r4 01 P
x 01 a
E-iTO *
rH
ao o
-H H
U 01 CO
P E JJ
n 3 0
B 01
a H tP
NO
rH > rH
POX)
E 9-rf
51 o O
3 K-rJ
Cl (D
H > C
h IH ai
iHXI
O
B O
TO 3X
M
H P 0
rH 10 P
01 E 10
E-H D>
E 10 01
10 tn B
X <3
0 O
01 JJ
r-l E
H O TO
01 N B
H P
>. 01
rH IH TO
H
>i O O
TO a
B 0
CJ » B
ta a a
C CO
a n
x a «
0 X
01 01 E
r4 X 10
U 4JXI
B to 01
O V4X
x a) p
OTO
H H tn
X 0 E
d XI O
tO rH
to TO it!
H B
S n) B
rH 01
D> 01
ETO to
H 0)
a TO at
H 0X1
r4 O
a? B
g i
OHO
U O
to
M 01 S
X X O
H M rH .
O d IH 10
Sana;
P OB
E13 O N
B
6 10 rl H
10 rH O O
Q> P 3
to a E xi
0 »
0) O O
XIH JJX
EH O 01 JJ
,
IN
H
O
o
o
u
H
B
rO
S1
o
t
Q*
H
a
§
a
to
0)
5
o
01
B
0
d
p
O
P,
,
-------
provide a preliminary assessment of land uses exhibiting the
potential for generating significant nonpoint source loads
and to assess water quality problems. The investigation
studied construction sites, storm sewer outfalls, use of
buffer strips and use of controls to restrict nonpoint
source runoff. Before conducting the survey, steady-state
water quality model results and reports of biological studies
of the stream were examined to provide an insight into
probable areas of concern. On the basis of these preliminary
analyses and knowledge of existing construction site loca-
tions, the survey sites were chosen.
3. Second Survey
The second survey was performed on March 24, 1977, two days
after a rainfall of approximately 2.0 inches. Figure 6 also
presents a map of the survey sites. These survey sites were
selected on the basis of (1) potential problem areas un-
covered in the first survey; (2) results of further data
analysis; and (3) suspected areas of concern revealed by
information gained from applying the mass balance analysis.
The timing of the survey was purposely chosen: two days
after a significant storm when the water level of the stream
was still high. Assuming that the water level would have
subsided to near steady-state conditions approximately five
days later, a third, more limited survey would have produced
greater insight into the effects of stormwater damage of
sediment and nonpoint pollutants on stream biological condi-
tions. Time and budget constraints prevented such a survey.
A summary of the findings is presented in the following
paragraphs.
4. Summary of Second Wissahickon Creek Survey
Date: March 24, 1977
Weather: Clear, Temp. 50°F, Wind 10-30 mph, Rainfall
March 22, 1977 1.97 inches, recorded at
Phoenixville
Stream Flow: Higher than normal, but subsided substantially
from peak
Field Crew: C. Gabriel and J. White
76
-------
Site #1B Headwater Area
The headwater area above known point discharger is bio-
logically unproductive, but does not appear to be seriously
affected by nonpoint source pollution. The main reason for
this conclusion is the absence of siltation in the area (see
Photo 1). The probable reasons for the low productivity is
the intermittent flow of the stream in the headwaters and
possible scouring from a recent storm (scouring would be
increased by nonpoint source runoff).
Photo 1 - Wissahickon Creek Headwater Area (Site 1B)
Site #2B Ambler Area
Downstream, in the Ambler area, the stream does not appear
to be seriously affected by nonpoint source pollution. In-
vertebrate diversity and algae growth did not appear to be
reduced and invertebrate organisms characteristic of good
water conditions were present. The mainstream does not ap-
pear to have a siltation problem; however, a feeder stream
showed evidence of significant siltation (as shown in photos
2 and 3). It was obvious that heavy siltation loads were
77
-------
Photo 2 - Silted Tributary to Wissahickon, near Ambler
Photo 3 - Silted Tributary to Wissahickon below Ambler
78
-------
received by the mainstream of the Wissahickon Creek and were
transported further downstream.
A second noteworthy fact was a large pile of residue of as-
bestos from a nearby plant. Additional waste asbestos is no
longer stored there but there is evidence that the existing
pile runs off to the stream (see photos 4 and 5).
Photo 4 - Asbestos Pile at Ambler; Wissahickon Creek
in Foreground
Site #3B Sandy Run Upstream from Sewage Discharge
The Sandy Run tributary upstream from the known sewage dis-
charge appeared to be nutrient enriched. Algae growth ap-
peared relatively thick for the time of year and for follow-
ing a recent heavy rainfall. The invertebrate community was
greatly reduced and consisted of tolerant organisms. The
source of the enrichment was not obvious and little sedimen-
tation was apparent. However, the stream was channelized,
which is probably attributable to the increased runoff from
79
-------
Photo 5 - Asbestos Pile, Erosion and Runoff to
Wissahickon Creek (foreground)
the heavily urbanized areas adjacent to the stream. There
are large storm drains directed into the stream (see photo
6) .
Site
Sandy Run Downstream from Sewerage Treatment Plant
Downstream from the sewage treatment plant on Sandy Run, the
stream appeared heavily degraded; however, the absence of
siltation indicated in this reach of the stream that non-
point source pollution was not a significant problem. There
80
-------
Photo 6 - Stormwater Runoff and Channelized Stream,
Sandy Run
was evidence that large sediment loads from a construction
site adjacent to the stream had entered the water (see photo
7) as evidenced by washouts to the stream from the recent
storm. The sediment loads were evidently transported down-
stream. Photo 8 shows evidence of thwarted best management
practices in that a large six-foot-high pile of wood chips
used to control erosion at the construction site was washed
out and deposited along the stream bank.
Site #5B Wissahickon Creek-Bells Mills Road
At both the Bells Mills and Valley Green area the water was
a gray, murky color and biota were reduced, indicative of a
degraded environment. The stream in this reach was a
series of riffles, then pools. There is evidence of sedi-
ment deposits in the pools (see photo 9).
81
-------
Photo 7 - Construction Site near Sandy Run Creek
(Sandy Run is located approximately 100
feet beyond excavation in background
along tree line)
Sandy Run
(flows under
bridge)
Photo 8 - Wood Chips (used for Erosion Control) Washed up
on Bank of Sandy Run Creek from Construction
Site (see Photo 7)
82
-------
Photo 9 - Sediment Deposits Below Bells Mills Road
Site #6B Ivsey Lane, below Confluence of Cresheim Creek
Below the confluence of Wissahickon and Cresheim Creeks and
the dam, biological conditions appeared to be quite good
where the stream flow was rapid. Invertebrate and algal
conditions were indicative of good water quality and no sed-
imention was apparent. Above the dam, habitats were reduced
and some sedimentation was apparent, indicating the dam was
providing an area that collected sediment. Further below
the dam was an area of collected sediment deposits from a
swale running through Fairmount Park (photo 10). These
sediment deposits were largely due to runoff from a site
about 1/8 mile from the stream where a single home is being
built (photos 11 and 12). No erosion controls were visible.
83
-------
Excavated soil was piled high along the steep slopes adja-
cent to the swale. Philadelphia County employs what has
been judged excellent land management controls for the
Wissahickon Watershed. Apparently, this construction site
escaped these controls because enforcement is directed
mostly to larger developments. However, it can be seen that
even a small construction site can cause significant non-
point loads.
,tf
Photo 10 - Sediment Deposits (foreground) from Construction
Site Transported via Swale (background)
84
-------
Photo 11 - Construction Site near Ivsey Lane, Front View
Photo 12 - Same Construction Site (Rear View)
Sediment Erodes into Swale on Right
Side of Photo and is Transported to
Wissahickon as shown in Photo 10
85
-------
Site #7B Cresheim Creek (Tributary to Wissahickon)
Cresheim Creek, which drains an urban area but has no point
sources, appeared to be in fairly good biological condition.
The invertebrate community was quite diverse, and there were
no signs of storm scouring. There was evidence of a small
amount of sediment (see photo 13).
*»*
J
Photo 13, Cresheim Creek, Some Sediment Deposits
Site #8B Mouth of Wissahickon
At the mouth of the Wissahickon and appproximately 1 mile
upstream from the mouth, near a series of dams, there was
heavy siltation and the biota were greatly reduced (see
photos 14 and 15). Available habitats were heavily silted,
which was the probable reason for the absence of aquatic
biota. This appeared to be the area that received the bulk
of the sediment load. Conversations with members of a canoe
club located at the mouth of the Wissahickon revealed that a
bar area approximately 150 feet long and 50 feet wide re-
sulted from sediment deposits washed downstream during major
storms.
86
-------
Photo 14 - Wissahickon Creek Upstream of Mouth,
Large Sediment Deposits and Mucky
Bottom
Photo 15 - Mouth of Wissahickon Creek, Very
Large Sediment Deposits
87
-------
5. Nonpoint Source Monitoring
Monitoring of stormwater runoff from various land uses dur-
ing storms was beyond the scope of this study. Pollutant
loading rates and other data, therefore, were obtained from
local and regional reports, local USGS data, and the scien-
tific literature. The sources of information and the ra-
tionale and assumptions used in this case study are ex-
plained in detail in Section F so that the user of this
manual may also apply this methodology without performing a
land runoff monitoring program.
It is recommended, however, that field monitoring programs
be instituted to obtain localized data on loading factors.
The cost of a modest field monitoring program to ascertain
typical land use loading rates, in light of the ultimate
costs for point and nonpoint source controls, is relatively
low.
D. WATER QUALITY AND POLLUTANT LOADING ANALYSIS
Water quality, stormwater, stream flow, and rainfall data
were analyzed to provide steady state pollutant loadings,
calibration data, and verification data.
1 Steady-State Pollutant Loadings
Steady-state pollutant loadings were calculated from point
source data and from monthly water quality data. The
steady-state loading and concentration of suspended solids,
BOD5, total phosphorus, total nitrogen and total heavy
metals are presented in Table 8. Values are shown in terms
of concentration during average low flow and in total load
in Ibs/year.
A portion of the point source sediment loading settles out
a short distance downstream from the wastewater discharge.
These loads, which are resuspended during storm flows and
washed downstream, were considered in the annual point
source loading. The sediment loading includes estimated
suspended solids from point sources which have settled out
as benthic deposits. All the point sources discharge above
88
-------
W
w
I
2
w
^^
2H -->
O^ rtj
p m -P
O 0) CD
B W 5!
(0
rH O
"3 ^
P a
O M
EH 0
A
CU
rH CJ1
ra o
-P H
0 -P
EH -H
2
H
-------
the Ft. Washington gauging station; therefore, the calcu-
lated point source loading is representative of total point
sources in the Wissahickon Watershed.
2. Calibration Data
The United States Geological Survey (USGS) flow and sediment
yield data at the Ft. Washington gauging station for water
year 1968 (October 1967 through September 1968) were the
most complete data available for calibration of sediment
loads. Corresponding rainfall data were taken from National
Oceanographic and Atmospheric Administration (NOAA) records
for an adjacent watershed near North Philadelphia Airport
(approximately 15 miles from the geometrical center of the
Wissahickon Watershed). Table 9 presents a summary of the
loading and rainfall conditions for 1968. During 1968, 12
storms yielded more than 40 tons of sediment per storm. The
total sediment loading from these 12 storms which occurred
during only 6.3% of the year, accounted for 15 million
pounds or 94.3% of the total annual sediment loading. The
annual nonpoint sediment loading at the summation point was
12.25 million pounds.
Regression analysis of sediment loading versus storm flow
and monthly stream flow indicated excellent correlation with
coefficients of 0.93 and 0.92 as shown in Figures 7 and 8,
respectively. Figures 9 and 10 show a similar relationship
of sediment load versus storm rainfall and monthly rainfall.
This regression analysis exhibited acceptable correlations
of 0.76 and 0.61, respectively. The rainfall correlations
with sediment are believed to be less than those of flow
correlations because rainfall was measured in an adjacent
watershed while flow was measured within the watershed.
3. Verification Data Set
Sediment loadings from six of the storms occurring during
1964 were measured by the USGS at Ft. Washington. These
data were selected to serve as the base for verification
analysis. Regression analysis of the sediment loading
versus storm flow, presented in Figure 11, yielded a good
correlation of 0.96.
90
-------
U
HH
w
PQ
-P
C
0)
B w
H X>
-0 H
<1) »-
W
Oi
-P C
C! -H
-H 13
O
ro
rH
ro
in
ro
m
m
VD
Q)
H
XI
id
EH
14H
0
m
C
id
*3*
H.
ro
01
g
0)
4J
e
o
M
'M
-d
0)
^
H
M
Q)
Q
id
0)
MH
0
U]
^i
id
d
rH
id
4J
0
4J
MH
0
r^P
ro
rl
O
Ul
£*1
id
d
ro
CM
d
0)
P!
(0
0J
in
ui
g
^
O
-p
co
st of load consisted of nonpoint runoff
o
fi
0)
Ul
3
id
o
0)
Q
Ul
XI
rH
^0
0
rH
X
rH
ro
in
rH
c
id
^£j
-P
01
U)
0)
rH
^
rH
-P
^
CP
H
i-H
co
ding minus point source loading and
ading
id o
O rH
rH
(1)
rH O
id M
-P 3
O O
4J W
0 4J
p a
rl
H O
3 C
D1 O
0) G
U) C
H O
H
tP+J
G id
rH ^-)
-d xi
rtJ -rH
O rH
rH (d
O
0)
0 0
M -p
£j
o w
Ul TJ
C
-P O
rl Ul
O (U
ft M
EH M
0 0
13 U
a
o
C
H
V
Ul
M
GJ
(j)
C!
H
C
W
rH
id
-P
C
ci
2
0
^
^
G
W
N
+J
0)
pQ
(J)
O
CM
o
w
91
-------
10 J.
200
1200
1AOO
FLOW CFS - DAYS
(Accumulated Flow)
FIGURE 7
WISSAHICKON WATERSHED - SEDIMENT VERSUS STORM FLOW
WATER YEAR 1968; FORT WASHINGTON GAUGE; CORRELATION = 0.93
92
-------
25
50 75 100 125
MEAN MONTHLY FLOW (CFS)
150
FIGURE 8
WISSAHICKON WATERSHED - SEDIMENT VERSUS MONTHLY FLOW
WATER YEAR 1968; FORT WASHINGTON GAUGE; CORRELATION = 0.92
93
-------
RAINFALL - INCHES
FIGURE 9
WISSAHICKON WATERSHED - SEDIMENT VERSUS STORM RAINFALL
WATER YEAR 1968; FORT WASHINGTON GAUGE; CORRELATION = 0.76
94
-------
»-
z
UJ
o
ui
to
10.000
9
8--.
7
6
S
4
3 ..
2
1,000
9
8
7
6
5...
4
3 .
100
9.
8.
7.
6.
5.
4.
3.
2.
10-
0
I -
_,
^ r ^
"H'i-"
r. -:-..:.
~Z.TT-
~. _
i. : ~ ..
~. i~
-.-r_::.
i '--.--'
-T--: : v
U_L--_
| r* -t"
. . i ,
- i i .
.-: : .
- - T
~
_J_. _
._" .
~r
- : :
i : :-;
- l
t u-f.-. -_
, .-
_~t-^-
i
' i '
.j : ' 1 : ; '
-O \
Tf: . ;
~it: T"
-. . r
:- -...
-.-'r '
T". "".-
' i .
:M:
. .
....
y
N
/
1
,
,
: ' t ' :
" : ' : r
. -^.
" nT *
. . --
. .-:-
:-i : .
-: r:- 1 -: .- -
.r,-:-
r*1
~ -^cj-
LU"1"
[
:- ;
..-4
I}-i_-;-
1 i i
: _ - - .--.-.
: j^
/
* '
f- ' '
!/
yf:
f
M
-
. t
'--. :
-
.
!
i
. j
t
. . - »
: : ._-
.
_ . .1.
;~;r:r :1
^~
- ; ; - . -
_,
-"-) *y
'"J^
:::
- - -
/
^ - '"'"
. ._--. ( : : i :
.... j - ...
':":: t" : :
. . :
' -
-
i ^
;
- ;
-. : : ! -
L_ ._
::--!' /
'/
r "^'
4-J
^
^^
-i _
- i -
..-::.
Q
i
. . j
.--I .
i
i . :: /
i . . .
.
-
' *
>
, /
i/
/
/ ''
^--i
: i
-
., .^
|_r. '
~t
- . r .
_
.y.:=
"X"-
. . \ ...
. - 1 . .
1
!
j .
' .
.
' i '
.1.:..
-
.
:t
/
IIL'J
mom
;-:
.
7eur
hty
i :
-: i . - :.-.:.
- :-T
: -.
T :
-: :
' - : ;--
1 ' ;
r
' j . _
' : - H "
/
1
1
.: . 1
i
;
....
_
-abr
-
Sv
I
- . i
:
. . ;.L
i i -
1.0 2.0 3.0 k.O 5.0 6.0
MONTHLY RAINFALL - INCHES
FIGURE 10
WISSAHICKON WATERSHED - SEDIMENT VERSUS MONTHLY RAINFALL
WATER YEAR 1968; FORT WASHINGTON GAUGE; CORRELATION =0.61
95
-------
10,000.
9.
8.
7.
6-
5.
i/>
Z
Q
O
Ul
4...
3...
2-.-
1,000..
9..
8__
7..
6..
5__
100
9.
8.
7.
6.
S.
3.
2.
10-
0
. !
q::U±t
y
^L
12c
,003K
*
I
200
400
600
800
1000
1200
FLOW CFS - DAYS
(Accumulated Flow)
FIGURE 11
WISSAHICKON WATERSHED - SEDIMENT VERSUS STORM FLOW
WATER YEAR 1964; FORT WASHINGTON GAUGE; CORRELATION = .96
96
-------
Based on the equation derived from the regression analysis,
the sediment load of all of the storms for the USGS water
year 1964 were calculated. This resulted in a total annual
load of 13.84 million Ibs/year (see Appendix C for full data
set). Table 10 shows the relationship of sediment loads,
rainfall and flow of the verification year 1964 and the
calibration year 1968. The nonpoint source loading (total
loading - point source loading) was 9.84 million Ibs. This
represents the verification load at summation point number
one.
E. SELECTION OF SUBAREAS AND LAND USE CATEGORIES
1. Subarea Delineation
The most prominent feature of the subarea delineation is its
urban-suburban interface. Most of the developed land in the
study area is suburban. The southernmost portion of the
study area lies in the City of Philadelphia where the Wissa-
hickon Creek threads its way through Fairmount Park.
The primary considerations in subarea selection were hydro-
logic features, man-made influences on runoff drainage, and
the existing land development patterns.
(A) Hydrologic Characteristics; The criteria used to
delineate subareas were primarily hydrologic characteristics
of the study area. Each subarea represents a drainage area
to the Wissahickon Creek. Subarea segmentation is shown in
Figure 12. Drainage areas M2, M3 and M4 each have creeks
the Trewellyn, Prophecy and Sandy Run, respectivelywhich
are tributary to the Wissahickon.
Seven subareas were delineated along with three summation
points. The summation points are at Ft. Washington, Bells
Mills and at the confluence of the Wissahickon Creek and the
Schuylkill River. Basically, all data derived from subareas
Ml through M5 are summed at the Fort Washington summation
point. At Bells Mills, data cover M1-M5 as well as Nl. At
the confluence, data include information from all subareas,
including 01, which is largely Philadelphia.
97
-------
W
PQ
EH
o
(_^
Pu
y
3
W
EH
W
n
£5
(3*
^3
pu OO
2 10
H CT\
(Cj* ,|
K
*O
K (^
O «J
2
H TT
Q W>
rij (Ti
O rH
EH
M
^
H
Q
W
05
^
*^4
D
£5
i^3
4J Qi
Pi d
H -H
O -O
o, id
C O
O f-^
2
~- 0)
o
W3 3
O\ O
* 1 1
00 I |
1 1
0> 1 1
^^
a
o
H
00 -P
vo id
M
H X)
H
r4
id
O
in
^^
i* H in
oo ro ^f
^\
CO > rH
i-l ID CO
rH
vo
0
H
^
to
X!
-I
*-^
^^
CP W
d Q)
a;
4J
e
(0
^4
nj
H A
-a u
id C
0 li
"in
4J >*-l H
C U rH
0) *- (0
g ^ MH
H S C
T3 O -rH
(1) rH Id
W ^ K!
0)
0
c
id
r-t
m
X)
tQ
(0
id
g
o
(0
4J
£j
O
+J
c
H
^4
a.
M
0)
-P
a.
g
o
o
T3
d
n)
vo
t
g a d
0 D. W
M-l N
d -p
d -H iu
d) (H
> d
H ?
}H O "
(U X! (I)
Q W 0
M
*-^ ^
H O
W
98
-------
u
c
Ol
c
(0
4-1
c
o
0)
CO
3
O
to
E PT. 1
E PT. 2
E PT. 3
FIGURE 12
WISSAHICKON WATERSHED SUBAREA SEGMENTATION
99
-------
In this study, hydrologic boundaries outweighed municipal
boundaries due to an overriding interest in reflecting
drainage impact at the summation points. The incidence of
municipal boundaries based on ridge lines was minimal.
Since this would predominately have an effect on land man-
agement controls, a methodology was developed to overcome
this limitation (see pp. 129 to 132).
(B) Man-Made Influences on Runoff Drainage: Existing
stormwater conveyance systems were confirmed by reviewing
available engineering reports and performing a field study
to locate outlets and overflows. The significant extent of
storm sewers in Philadelphia was the reason the Philadelphia
portion was placed in one drainage subarea.
(C) Land Development Patterns: Efforts to protect Fair-
mount Park by the City of Philadelphia have included regu-
latory measures which influenced land use development and
land management in the southern part of the study area.
While the park has been protected since its origin by city
regulatory measures, the land adjacent to the park has
recently been subject to land management measures (most
notably, the Wissahickon Development guide) designed to
protect Wissahickon Creek from encroaching development.
North of Philadelphia, the study area includes two smaller,
older urbanized areas: the Borough of Ambler and the
Borough of Lansdale. The remainder of the study area is
mostly suburban, combined with agricultural and undeveloped
land. An extension of higher density development from
Philadelphia outward has occurred and is accounted for
predominately in subarea Nl.
2. Selection of Land Use Categories
Figure 13 illustrates the level of detail required for a
land use map in this methodology by delineating land use
categories found in subarea M3. In this study, a pre-
screening examination was made of land use maps prepared by
the Montgomery County Planning Commission and the Philadel-
phia City Planning Commission, as well as land use data
prepared by the Delaware Valley Regional Planning Commis-
sion. This provided a broad outline of the major land uses
to be considered. The land use map developed for the study
was based on aerial photography and USGS topographic maps.
The method used to develop the land use map is presented in
Appendix G.
100
-------
o
c
0)
c
en
c
c
o
c
o
r
UJ
0)
CO
o
00
FIGURE 13
LAND USE MAP FOR WISSAHICKON
WATERSHED-SUBAREA M3
Urban
Suburban
Commercial
Undeveloped
Agriculture
Suburban
51 Residential
New Development
101
-------
The delineation of land use categories and sediment loading
rates in this study were complementary tasks. Since the
pre-screening of land use information made more apparent the
predominant characteristics of land uses in this study area,
it was possible to review the literature for similar loading
rates of land uses. Some of the characteristics of interest
in this review included the specific type of housing, in-
dustry, or commercial use to which loads were attributed;
the degree of impervious cover; the frequency of rainfall;
and the level of urbanization, population, or other parame-
ter indicating intensity of activity.
The land use categories for the study area were not chosen
until after the literature review was made. It was apparent
that land use could be categorized in a number of forms, by
varying the definitions of categories. As much as possible,
an attempt was made to use compatible definitions of land
use for which loading rates were contained in the literature
by such definitions.
There were a number of constraints in categorizing land use
and arriving at sediment loading rates. In the first place,
the characteristics -of land uses associated with loading
rates in the literature were often not well identified,
leaving open the question of how applicable the rates were
to this study area. Second, the available literature in-
cluded studies performed throughout the United States, most
of which were not readily applicable to the Philadelphia
region due to soil, topographic and rainfall condition dif-
ference. Finally, most of the studies measured loadings in
Ibs/acre/year or some convertible form. Those studies which
used another form of measure such as concentrations without
accompanying flow which could not readily be converted to
Ibs/acre/year limited the overall choice of values.
The land use categories chosen, based on this approach, are
presented in Table 11. These categories are representative
of land use in the study area, but general enough in their
definition to be equated with loading rates from the litera-
ture reviewed.
Monitoring of stormwater runoff to determine study area
specific loading rates was beyond the scope of this case
study. Undoubtedly, field inventory of stormwater runoff
would have greatly enriched this data base. It is recom-
mended that monitoring of representative land use be em-
ployed to achieve study area specific data where possible.
102
-------
rH
W
ft
rij
M
M
W
w
0
0
1
0
t/5
85
O
M
f
H
z
H
fc,
W
Q
neral Assumptions
01
o
impervious cover
ally turned over
* 0
O -H
rH 'O
o
C -H
<0 rl
x; 01
01 id
01 C
$ S
01
p
-p
rH
3
o
rl
M
Cn
«)
C
O
H
4J
a
rl
rH
0
0]
01
Q
H
O
3 V CO rl
O Cn rH
g CO
O 4J
u c
0)
-id
rH -rl
cd 01
H 0)
rl rl
4J
Ul >i
3 rH
C g
H CO
KH
1-1 1
O -H
p
X r-i
H 3
S g
X
rl
J£
C
CO
XI
rH
O
size
us cover
o
4J -rl
O >
rH H
0)
01 ft
rH g
*Q -H
CO
rt dP
H +
10 m
> t-
i
o -
(0 .P 01
C H
rH 0) O
CO g O
o o o
fc rH 01
0) Q)
g ^ *
g 0) 01
o "d x
0 M
ft CO
IH -H ft
O >H
4J rH
Ul 01 CO
-p -H
01 01 rl U
X CO P 4J
U CJ1 01
O £ 3
3 C Ul
CD 01 -H rH
P CO
0) >. -P
V* P 0) -H
CJ -H rH ft
Ul > rH Ul
-rl -rl ClJ Q
Q -P g J3
rH
CO
O
f_I
0)
c~
0
U
c
<0
H
3
"I
W
4 acre but generally smaller than 1 acre
^^
rH
A
0)
N
-rH
01
-P
O
rH
01
01
0
c
01
d
-H
01
01
rl
^
rH
-rl
c3
01
r-t
Cn
c
H
Ul
r-,
p
01
O
5C
rH
CO
rl
4J
C
01
-d
H
01
01
«
c
CO
n
3
t/3
ious cover
j>
rl
01
t
H
dP
O
in
I
o
rH
impervious cover
tructure per acre
dP Ul
o
H rH
c c
CO CO
4J P
01 01
01 01
0 01
rH rH
01
c
o
rH
id Cn
C 01
CO >H
01 'O
d CD
iH +J
0) CO
i« 0)
- 01
01 >
rl T3
CO 01
pt p,
o
Ul cy
CO >
C1J Q)
M Td
CO C
d
01 r-1
d 01
o .c
O P
z o
-d
01
ft
o
rH
0)
01
d
c
o
(4
4J
Ul
01
Vj
O
size
tegory up to maximum 4 years
01 CO
4J 0
H
01 C
rl
0)
rH 01
XI C
CO -rH
iM £
"3 0)
1
§ ,
rl
H 01
° rH*l
O -P O
M n u
3 01
P r< rl
r-H rl O
3 3
0 0 rH
rl <0
H 01 -H
Cn 0) 4J
CO-H C
H 0) K
C O >d O
rl Sl-H -H
111 Ul P
>i4-> 0) O
rH CO ^ 3
>4 U U
tti tn4-i
g -3 c !d co
CO 01 C -H
r^ Tl 3 U
c
01
(3
ft
o
rH
o
CD
Q
o
CO
t-5
0)
gement varies by municipality
10
rt
i
01
in
3
d
c
cO
rH
sediment loss potential
0)
rH
8
01
d
rH
(Q
c
o
o
103
-------
1968 Land Use; In this study, water quality and non-
point source data were most complete for 1968; therefore 1968
was selected as the base year for analyzing existing condi-
tions including land use and raw loading rate determination.
Land use categories used for the study area are presented in
Table 11. The categories include cropland, urban mix,
suburban commercial, suburban residential, forested undevel-
oped, and new land development.
Urban mix includes an integration of residential, commer-
cial, industrial, public and institutional land uses at a
sufficient density to indicate the central core of an area.
New land development was delineated as a separate land use
category because of its high erosion potential.
A summary of land use by subarea based on 1968 data is
presented in Table 12. Forest and undeveloped land ac-
counted for the highest percentage (36.2%) of the 40,960
acres in the study area. Urban mix accounted for 26.9% and
all suburban categories accounted for 20.2% of the total
land.
Land Use Projections; Projections of land acreage for
each of the six categories were derived for the years 1985
and 2000. This information is presented in Table 13.
Land use data for the year 2000 were based on population
forecasts of the Delaware Valley Regional Planning Commis-
sion (DVRPC) . The DVRPC data were available by minor civil
division. The data were used to obtain an estimate of
population increase for each subarea. The specific land use
changes in each subarea were based on the following assump-
tions: 50% of the population increase would settle in
multi-family units at a density of 25 persons/acre; 50%
would settle in single family units at a density of 1 0
persons/acre .
In 1970, the U.S. Bureau of the Census in Detailed Housing
Characteristics (Pennsylvania) reported for Montgomery
County, Pennsylvania (where almost all new residential
development in the study area will most probably occur) a
split of 60% single family units 40% multi-family units for
1970. However, housing unit construction trends nationwide
(Source: Joint Center for Urban Studies of M.I.T. and
Harvard University, February 1975) indicated that single-
104
-------
W
r~^
VI
0)
I.
M
u
N_» <**
?4 «i
Q
^^
E-..-H <0
to--' CU
< >-l
W W TCI
to r« XI
< -< 3
O P3 W
P
X 01
W
S PQ
U
W
Z en
OD
«
U Q
H &
W<
< ^
to
to
H
5
i-H
r-H «*
nj o
O rH
fH rH
VO
r-H VO
01 CO
n
, 0
rH 00
"Z.\ 00
CN
m
in oo
Si vo
, **
*#\ o\
sl "i
, °°
m o
SI CN
r-l
0>
CN r-
SI o
r-H
rH
rHl IT)
SI CN
r-H
, .
CO
vo
en
rH
'
rH
n3
rH
O
IH
CU
£5
s
0
CJ
c:
ITS
ja
>-i
3
XI
3
to
00
o\
rH
VO
H
ro
CN
00
o
vo
n
VD
n
CT>
r~
cr>
rH
rH
CN
in
H
en
^r
n
rH
r-
r-H
rH
rH
a
H
4J
C
0)
O
rl
W
CU
rt
a
n3
ja
M
3
XI
3
W
H
CN
00
^J1
rH
00
CN
VO
CN
H
r~
0
*r
0
CTI
CN
O
CO
O
m
tr>
rH
CO
H
H
r*1
in
CN
CO
o
*!
O
0)
a,
o
rH
CO
>
CU
'0
c
p
00
T?
m
in
CN
*
VO
CN
rH
m
in
in
CN
00
CN
l>
in
00
H
in
*
vo
0
a
O
iH
O
\
(1)
in
3
4J
rH
3
U
rH
)H
(T>
<
VO O
00 VO
CN O\
rH 0
f
r~ vo
vo m
rH VO
VO
r> CN
vo a\
H H
00
"* H
o r»
CN 00
H
a\ -*r
o m
CN O
CTi
vo
1- H CO
m m rH
rH f- X)
vo n)
E-i
O
C C
H H
^r r« r^ -
0) CO r-H 0)
n in M
vo T3 (1)
C
e c
-P O W
G H
(U 14-1 N
R -P
ft -do)
O oo 0) «
rH VO >
O CTl -rl
> r-H H
CO CU CO
Q rH Q O
KJ H
£ P ~ 0
CU O rH O
2 EH W
105
-------
CO
In
S
Q u
w <
SB --
cn
P< id
W cu
rij Id
S X!
3
2 cn
o
o
H
cn
cn
H
5
I
cn
&
o
H
£H
0
in ro
cn CN
O co «3"
CO CO
in o
co en
Z cn cn
CN CN
O VO
in co r-
S r-r-
rH CO
If** ^J*
vor-
CO CO
ro vo CN
S CN CO
rH rH
*r oo
CN f~"* VO
S rH CN
rH rH
rH 0
rH m r-
g CN CM
H rH
w
o
>H in o
W ID co o
cn Q) cn o
D JH rH CM
Q
a
cu
CO
d
C
id
i-l C
id
A
[D
O CO
in r-
CN CN
co in
CO VO
CO *3*
ro r~
ro co
*3* rH
r-- co
"* m
Tf CO
co r-
CN ro
in H
CO rH
I-- cn
rH r-
VO CO
in o
CO O
cn o
rH CM
rH
id
H
o
U
CU
0
g
o
o
C!
JQ
H
3
8
cn
co m
r- CN
CN CO
co r*
ro ro
r-- co
cn *
H r^
* ^
VO CM
rH in
rH CN
CN CN
in cn
CO **
vo r^
rH rH
r- T
o vo
in vo
rH rH
O CO
in co
rH rH
in o
CO O
cn o
rH CN
rH
id
.1-4
.p
C
0)
d
H
CO
(U
cd
C
id
»Q
M
0
A
3
cn
VO CO
CN CN
VO VO
CN CN
oo r~
vo vo
cn co
co ro
^ CO
m co
CN CN
in o
*^* ^j*
cn co
CN CN
CO CO
cn o
VO VD
H rH
VO rH
in cn
31 (M
CM CM
ro co
vo cn
ro CM
in o
CO O
cn o
rH CN
d
0)
o
H
0)
cu
d
a
p
T>
§
^j
10
cy
M
0
rH 0
CM
ro O
vo
f~- O
CN
cn in
rH CO
r- in
CN CN
r^ CM
vo in
( VO
i-H rH
r- in
CM 0
m <*
0 0
m o
00 O
cn o
rH CM
CO
ft
O
U
\
cu
4J
rH
3
O
H
M
fd
VO **
00
0 CO
O CO
rH
CO H
O rH
rH
cn co
CM vr
H
cn rH
cn *
CO CO
CO T
H
^3* r~H
CM rH
in o
CO O
cn o
H CM
-P
a
(U
e
ft
o
rH
0)
Q)
Q
d
C
id
i«3
CU
2
(U
o
r<
3
O
cn
106
-------
family construction accounted for 55% of all units in the
1965-1969 period and 47% for the 1970-1973 period. Based on
these trends, the percent of unit types previously cited
were applied for this projection.
Additional acreage counts were determined for urban, subur-
ban residential, and suburban commercial, taking into account
that approximately ten percent of all additional developed
acreage was considered to be new land development in the
year 2000. The increases in the above categories were
compensated by an equal loss of land in the agriculture/crop
and undeveloped land categories.
Land use for the year 1985, selected as a mid-point between
1968 and 2000, was developed based on a straight-line extra-
polation of acreage counts for the categories of urban mix,
suburban commercial and suburban residential. The other
categories were similarly changed to balance loss of land to
development. The value of the 1985 land use projection was
to illustrate how the methodology could be used to assess
sediment reduction capability of control systems over time,
given changing land use conditions. It is important to
qualify, however, that land use projection was not the
intent of this study. More sophisticated methods of pro-
jection than that applied here should be used when attempt-
ing to assess control system capabilities over such periods
of time.
F. DETERMINATION OF RAW LOADING RATES
The land use runoff sediment loading rates are based on the
modified runoff equation and include the following factors:
Raw runoff rates (K)
Rainfall factor (R)
Slope length factor (LS)
Sediment delivery ratio (SD)
Natural controls (P)
Land Management Controls (K2)
A list of typical loading ranges (Ib/acre/yr) was developed
from a general literature search, the purpose of which was
to determine the characteristic orders of magnitude and to
107
-------
rank the land uses based on typical loading ranges. The
following ranges were found:
Sediment Loading
Land Use Range, Ib/acre/yr,
Urban Mix 174 - 6,617
Agricultural (crops) 255 - 3,750
Forest and Undeveloped 40 - 117
However, these rates often included some or all of the
modifying factors listed above. As such, many of the values
found in literature are low.
The range of rates found for both urban mix and agriculture
are so wide as to render selection of an average runoff rate
of questionable value. It is believed that the widely
varying rates found in the literature search resulted from
the varying degree of intensity of land use (degree of
urbanization and farming) and from the fact that many of the
runoff values reported implicitly included many of the
modifying factors listed above and as such were much lower
than the characteristic raw runoff rate. Therefore, the raw
runoff rate of each typical land use was determined through
selective use of literature rates. Only those literature
values for similar land use intensity and geographical and
climatic conditions were used. In addition, estimates were
made as to the effect of runoff rate modifiers and raw
runoff rates back-calculated.
1. Urban Mix Loading Rate
Since urban runoff data were not available for the study
area, selection of a raw loading rate was based on the
following approach developed in the EPA publication, "Report
on Loading Functions for Assessment of Water Pollution from
Non-point Sources" (November 1975):
(A) Selection of a region-wide urban loading rate
(Ibs/curb mile/day) from data collected from a
nation-wide study. A mean loading rate of 103
Ib/curb mile/day of raw runoff was reported for
the northernmost portion of the southeast cli-
matological region. However, since the City of
Philadelphia and older urban areas such as Ambler
108
-------
and Lansdale comprised most of the study area's
urban land, it was decided to use a rate twice as
great (206) as the documented value of 103 for
this study area to account for the greater in-
tensity of urban activity exhibited here.
(B) Estimation of the average curb mile density (curb
miles/acre) based on a graphical solution relating
curb mile density to population density (see Ap-
pendix I for graphical solution and source). The
population density of the Ambler area was 14.8
people per acre based on the 1970 census. This
population density reflects a curb density of 370
ft/acre (0.07 mile/acre).
(C ) Using the factors generated in the two previous
steps, compute a projected annual loading rate:
K = (206 Ib/curb mile/day)(0.07 curb mile)(365 days/yr)
acre
= 5,263.3 Ib/acre/yr.
The value generated above was then compared with
the list of typical loading ranges from the lit-
erature. This list (page 108) showed a range of
174-6617 for the urban mix category. In comparing
the generated value to the literature values, a
determination was made, based on an urban charac-
teristics of the study area, to choose an approxi-
mate value of 6000 Ib/acre/year.
2. Agricultural Loading Rate
Since runoff data were not available for the study area, the
cropland loading rate was based on an analysis of soils
association maps for Montgomery and Philadelphia Counties.
Typical soils in the study area are comprised of shale and
sandstone with some clay deposits. The soils have the fol-
lowing properties:
Permeability range: slow to moderately rapid.
Erosion hazard: severe where slopes exceed 7%.
Overall runoff rate: rapid.
109
-------
A draft watershed work plan for the Wissahickon (Montgomery
County Soil and Water Conservation District, October 1965)
reported that the annual sediment production in the water-
shed is generally high, averaging 4 acre feet per square
mile (32,670 Ib/acre/yr). Because of the high potential for
runoff, the extreme end of the agricultural range (see page
108) was selected (3,750 Ib/acre/yr). This value was
rounded off to 4,000 Ib/acre/yr.
3. Forest Loading Rate
Little forest runoff data were available on which to make a
selection; however, the literature search produced a raw
sediment loading range in the low hundreds. To be on the
conservative side, a rate of 500 Ib/acre/yr was selected.
4. Suburban Commercial and Suburban Residential Loading
Rates
The aid of literature values for loading rates of these two
categories was not possible since only one value could be
found for each of these land use categories and each was
felt to be too low for this study area and probably not
representative of a range for these categories. Since, in
addition, specific runoff data were not available for the
study area, the sediment loading rate for these two land
uses was determined to be less than the urban mix or crop-
land rates, but higher than the forest. A loading rate of
1000 Ib/acre/yr was selected for suburban residential. A
rate of 4000 Ib/acre/yr was assigned to suburban commercial
to reflect the large amount of impervious surface and the
on-site commercial activities.
5. New Land Development
Extreme variations in raw sediment loading rates for land
under new construction were found in the literature. In
many instances, the estimated rate was as much as 100 times
that for undeveloped, undisturbed land, with extremes rang-
ing to 60,000 Ib/acre/year. A conservative estimate of
30,000 Ib/acre/day was selected; this represented a 6000%
increase over the rate applied to undeveloped land.
110
-------
G. DETERMINATION OF RAW LOADING RATE MODIFIERS
"1 Rainfall Factor (R)
The rainfall factor was calculated using the ratio of total
rainfall of a given year to total rainfall of the calibra-
tion year, or alternately, the ratio of total flow (cfs-
days) of a given year to the total flow of the calibration
year. Both methods were utilized in the Wissahickon Water-
shed Case Study and are shown for comparison purposes. They
are summarized in Table 14. The calibration year rainfall
factor is normalized to 1.0; others are related to it.
2- Slope Length Factor (LS)
The slope length factor for each land use within each sub-
area is determined by a simplified three-step process: (1)
determine steep-slope factors, (2) determine mild-slope
factors, and (3) determine composite weight averaged slope
length factor.
(A) Steep Slope Factors: The areas with slope gradients
greater than 12% were provided in a gradient map overlay
shown in Figure 14. Steep slope factors were calculated as
follows:
This steep slope overlay was superimposed with the
land use categories map and USGS Topographical
maps.
Examination of steep slope areas revealed an
average slope length to be approximately 200 feet.
The average of the slope grades in excess of 12%
was approximated at 15% grade.
An average slope length factor of 3.5 was derived
from the criteria presented in Figure 15. This
factor was multiplied by the area of each land use
overlaying steep slopes.
Ill
-------
CU
cn
id
«
O
4-)
U
fa
ro
00
i
H id
fa Q
rH 1
4-> CO
O fa
EH U
C^
ro
(N
O
(N
W
ffi
CO
w
EH
ri]
[2
^^
53 0)
O CO
« id
U ffl
K rH
co id
co IH
H C
!S -rl
id
PH On
O -'
1
. ,
PH PH
1
Pi
O
EH
J-l
O
4->
U
id
fa
PH
O
rH
H
iH
(d
n i ^^
C co
rl 0)
id A
PH U
C
rH -H
id
4->
o
EH
in
^J*
.
m
ro
U
fa
rl
H
g
0
H
-P
id
M
Q
rH
rH
id
U
(Ti
ro
CM
VD
H
in
00
o
o^
o^
H
G
O
H
-P
id
0
rl
MH
-rl
in
cu
!>
H
rl
CU
0
rH
C
O
H
4-)
rd
rl
XI
H
1 1
id
o
CU
f-t ^_l
AH rl
4-> rH
id
MH m
0 CJ
, J
cu id
CO M
o
4-1 CU
4->
rH O
Id -H
Cr1 cu
(U rl
a
0
1 \ ^j
o
T)
CU T)
g cu
3 rl
cn -H
cn co
id a)
CU
cu C
£ id
CO rl
C O
O »H
H
P C!
U 3
(U rl
n
O CU
M X)
CX
C
in id
O 0
rH CO
H a
id O
ij I .^
C 4-)
H U
id cu
M -n
0
CH fV)
id
cn
15 &H
O id
r-H CU
fa >,
CM
O
e
cn
>H
(U
Q)
a
-H
w
id
4->
C
(U
o
rl
rl
N
4J
CU
CQ
CU
o
rl
3
O
CO
112
-------
Slopes greater
than 12%
FIGURE 14
CRITICAL SLOPES MAP FOR THE WISSAHICKON WATERSHED
113
-------
20.0
3.5 6.0 10
Slope Length, Meters
20 40 60 100
400 600
Slope Length Factor (Steep)
Slope Length Factor (Mild)
40 60 100 200 400 600 1000
Slope Length, Feet
The dashed lines represent estimates for slope dimensions beyond
the range of lengths and steepnesses for which data are available.
2000
FIGURE 15
SLOPE LENGTH DETERMINATION CRITERIA
114
-------
B. Mild Slope Factor; All land use areas not overlaying
steep slopes were found to be on relatively flat grounds.
They were estimated to have a slope of 2.5% grade and an
average length of 1000 feet. These estimates were made by
observing topographical grade line features. A slope length
factor of 0.4 was derived from the criteria previously shown
in Figure 15. This factor was multiplied by the area of
land use overlaying mild slopes.
C. Weight Averaging: The weight averaged slope length
factor (LS) for each land use was calculated as follows:
(Steep Slope Factor x Steep Area) +
LS _ (Mild Slope Factor x Mild Area)
(Weight Averaged) Total Land Use Area
The resulting slope length factors employed for all land
uses/subareas are shown in Table 15.
3. Sediment Delivery Ratio
The sediment delivery ratio for each land use within each
subarea was determined by a four-step process:
Determine runoff transport potential
Determine drainage densities
Determine sediment delivery of each hydrologic group
Determine weight averaged sediment delivery ratios
(A) Runoff Transport Potential: The runoff transport po-
tential of soils was determined by:
(1) Developing a map overlay of soils runoff potential
based on hydrologic subdivision of soils as pre-
sented in the soil survey publications of the U.S.
Department of Agriculture for Montgomery and
Philadelphia Counties (see Figure 16). The hydro-
logic descriptions of runoff potential were previ-
ously presented in Section III under description
of methodology for sediment delivery ratio.
115
-------
a
u
u
I
8
u
H
U M
r
X
EH
O
Z
a.
s
CQ
c
a
o
0)
>
a
d
0)
a
0
01
0)
a
c
~ D
CO
M *O
0 C
4-> D
O f-H
tu U
h O
M
£ O
4J
C
3 H
n>
01 "H
o. -p
0 C
i-H 01
CO *O
H
n
0)
»
r-l
C )0
(d -H
J3 O
3 Q)
"§ I
CO O
CJ
c
IB
£
vo vo in in
o o o o
vo r- vo in
o o o o
T in in in
o o o o
en vo in in
0000
^" vo » ^"
o o o o
vo vo in in
o o o o
«« VO f»
000
p. o
0 0 r-l
« vo *
o o o
V VO 00
o o o
«» VO VD
o o o
9 VO VO
000
m
0)
C
H
O
0)
4J
O
n
a>
(0
0)
n)
§
to
S
in
l
M
U
s
18
a
EH
c
H m
1-1
o *»
Dr-t
U
I
iH
a
4)
c
H
H
N
4J
01
m
o
CO
116
-------
u
c
0)
c
(1)
CO
O
oo
FIGURE 16
HYDROLOGIC RUNOFF POTENTIAL MAP
FOR WISSAHICKON WATERSHED
f
N
RUNOFF POTENTIAL GROUPS
B I Moderate
Impervlous
to Average
Moderate to
High
High
117
-------
(2) Overlay land use map and determine percent of each
land use within each hydrologic group.
(B ) Drainage Density; The drainage density was determined
by measuring the miles of stream in each subarea (using
topographic map and measuring wheel) and calculating the
miles of drainage per square mile of area as shown below:
Drainage Density = miles of stream
square miles of area
The measured drainage densities for each subarea are pre-
sented in Table 16.
TABLE 16
DRAINAGE DENSITIES - WISSAHICKON WATERSHED
Drainage Area Drainage
(2)
Segment Stream Miles mi2 Density Vpp_
M-1 2.7 2.88 0.94 1.063
M-2 14.8 10.30 1.44 .69
M-3 18.4 10.55 1.74 .57
M-4 20.3 14.5 1.43 .70
M-5 4.45 2.92 1.52 .66
N 15.9 12.80 1.24 .81
0 13.3 10.39 1.28 .78
Mean 1.37 .73
(1) Derived from item 16 of Table 6.
(2) Drainage density is equal to stream mile divided by
drainage area in mi2.
(3) This value (1/DD) is equal to the reciprocal of the
drainage density.
Source: Betz Environmental Engineers, Inc.
118
-------
(C )Sediment Delivery Ratio of Hydrologic Groupings; The
sediment delivery ratio of each hydrologic grouping of each
subarea are determined using the criteria presented in
Figure 17. Since the measured drainage densities of the
subareas are grouped towards the flat ends of each soil type
curve, the process was simplified and a single drainage
density (the mean value) was employed for all of the study
area. The resultant sediment delivery ratios for hydrologic
groups in the Wissahickon are:
Group A - 0.2
Group B - 0.35
Group C - 0.52
Group D or U - 0.75
( D )Determination of Weighted Average Sediment Delivery
Ratio: The weighted average of the sediment delivery ratio
is calculated as follows:
SD.
LU
x SDA)
x SDB)
x SDC)
D or U x SDD or U)
where :
SD
LU
SD
= Sediment delivery ratio of land use within
subarea
= Percent of land use within subarea in hydro
logic group A
= Loading factor for group A soils.
The final weighted average sediment delivery ratios employed
are presented in* Table 17. Appendix H presents a sample
calculation of sediment delivery ratios. The sediment
delivery ratios are relatively high in that there are few
Group A hydrologic soils in the study area, and there is a
significant amount of impervious cover (type U hydrologic
group) . The Group A soils have the lowest sediment delivery
ratio and the Group U have the highest sediment delivery.
119
-------
o
<
>-
LU
DC
o
LU
00
U.
O
z
o
C£.
O
s_
0)
M (/I
U C
O
CL
C3
<-> O
1/1
C CD
O
U \-
1- -O
03 >-
2 Q-
T3 O
-a
2: <
o
CO
120
-------
r-
!-)
W
,J
9
FH
IT1
tH
M
K
CO
K
H
j|
2
O
0
i
CO
CO
BE
|
CO
o
H
1
X
OS
3
H
Q
|J
W
S
H
Q
H
CO
S
Cfl
0
i-l
i
l ..... i
r» CM a> o oo < o
n « m in vo « r-
iH T *§
H (8
VI
(U vo
QH
,^
i-H
ID
a)
M
10
A
to
C
Jj
c
CO
V
VI
a
jj
o
c
flj
to
^
a
c
a
_
CM
U
c
W
0)
0
c
H
Cn
U
H
S
c
c
g
0
VI
H
C
N
OJ
QJ
a
..
0)
u
3
co
121
-------
4. Natural Controls Factor (P)
Natural controls consist of buffer strips adjacent to streams,
and wetlands and marshes surrounding streams. In the Wissa-
hickon Watershed there are no wetlands or marshes along the
streams; however, there is significant buffering along the
stream channels, especially in the Philadelphia area where
the stream threads through the wooded areas of Fairmount
Park. Unfortunately, these natural controls are bypassed to
a large degree by storm drainage systems from urbanized
areas which transport the nonpoint runoff pollutants direct-
ly to the stream. The effect of bypassing of buffer strips
was included in the determination of the P factor. The P
factor was determined as follows.
(A) The degree of buffering in each subarea was
categorized and assigned load reduction rates as
follows:
Wide Buffer . - 0.5 load reduction
Medium Buffer - 0.25 load reduction
Narrow or No Buffer - 0.1 load reduction
Using the criteria from above a buffer factor (KA)
was determined for each subarea.
(8 ) The bypass factor based on field observations and
storm drainage maps was estimated in three classes
for each land use in each subarea:
High Degree of Bypass - 0.9 bypassed or 0.1
unbypassed
Medium Bypass - 0.5 bypassed or 0.5 unby-
passed
Low Bypass - 0.1 bypassed or 0.9 unbypassed
The factor (KB) was set equal to the unbypassed
load.
(C ) The P factor for each land use in each subarea was
determined as follows:
- KA*KB)
122
-------
where:
PLU = P factor for land use within subarea
KA = Load reduction that would result if
runoff were directed over the buffer
strip
KB = Unbypassed fraction of runoff that
is directed over the buffer strip.
The resultant P factors for each subarea/land use are pre-
sented in Table 18. A sample calculation of the P factor is
presented in Appendix H.
5. Ratio of Other Pollutants to Sediment (KN)
If possible, the KN ratios should be based on measured
runoff loads from typical land uses within the study area.
There were no such measurements made in the Wissahickon
Watershed; therefore, the KN ratios are based on measure-
ments made in watersheds similar to the Wissahickon. The
Occoquan/Four Mile Creek Study conducted by the Northern
Virginia Planning Commission measured runoff loads from
several sites during three storms. The mean value of pollu-
tant loads from these storms are summarized in Table 19.
Based on the criteria presented in Table 19, the KN ratios
for land uses in the Wissahickon Watershed were developed;
they are presented in Table 20. Also shown for reference
are the raw sediment loads (K) in Ibs/acres/yr and the
resultant equivalent raw pollutant load of each land use in
Ibs/acre/yr.
6. in-stream Consumption/Decay of Pollutants (K3)
The in-stream consumption and decay of pollutants varies
with the pollutant and stream characteristics. For example,
organic matter (measured as BOD loading) is a non-conserva-
tive material which decays with time and stream mile.
Phosphorus and nitrogen are other non-conservative sub-
stances. Available phosphorus and nitrogen are readily
consumed by algae. Effluents from domestic sewage treatment
123
-------
I
U]
1
I
U
M
1
10
§
E-i
Z
O
o
I
z
£
~
GO
t-l
I*
4J
r.
o
u
t-l
W
M
3
4J
10
Z
4J
C
c
Q)
U
Q
0)
2
0
V
a
o
^^
r- P- a> «n in
(N (M VO VO en en oo vo
VD VO CO 00 VO O O
en en oo oo en oo \o
>1
3
a
s
01
H
VH
-H
|
H
n
o>
o
H
4J
a
o
,
vo O
^ D>
O
O A>
C «
« U
-I O
3
u
(0
3
CO
in
2
a
!
O
0)
-H
V4
a>
a
a
10
+3
M-t
o
S
o
u
c
a
w
0)
v
rl
I
i-4
5
o
H
H
U
N
4)
O
O
M
s
(0
124
-------
PfM
C
01 O
rH-H
Id 4-1
5S
m
nH
id
P Z
o
*
rH
id
.00.
g
vo
o
o
*
£T>
O
'
441
O
O
*
rH
O
*
m
o
o
*
"*
0
'
fr|
0
o
*
ffS
VO
vo
CM
CM
o\
CO
vo
o
H
0
EH
^
-H
N^
rH
rH
IH
C
H
M
O
q
tH
X
0)
M
U
a
a
rH
O
id
&
e
0)
rH O
id o
4J H
Z
n
2
rH O
* .c
+j a
O (D
EH 0
C
o ^ -w
Ot CM rH
in r^ *
co id
JS >! £
o x S
000
J
* r- vo
rH in CA
CM in rH
a
01 m
o-o
C-H
ao
n CQ
3
co
en co CM
vo r** o
rH CO r-t
00
CM
O
At
01
0)
-H
M
3
O
8
0)
m
o
o
id
tJ* *H
O TJ
U 0)
U 10
O
MH
C O
H
4J
Di id
g 5
o
4J 0
0) 4J
0)
0)
M
JZ
41
a
id
o
4J
id
CO
in
I
e. oi
id m
fo 3
O
0> K
?7* C«
c ?
-H O
CO t<
n
C
0)
a
X!
»
W
a
01
3
rH
id
>
0)
Oi
id
M
01
>
id
c
O
a
0)
m
a
rH
3
rH
rH
O
a
MH
0
o
H
4J
id
v<
c
o
o
0>
01
id
CQ
CM
9
O
C H
sa
?&
u
OrH*
C 01
M
id
. 3
S0
II
M n
CU n
id g
P g
3S
c o<
o c
H
*o c
01 C
to id
Id rH
J3 O.
~ (8
H
u e
C-H
M O<
M
»-H
«J>
01 C
0) M
e o)
H.C
CK.P
e M
H O
z
rH
id -
Jj >1
CTJ
0) 3
2 co VP
O r-
iH 01 rH
> 0>
C M H
MOO)
N O 0
0-H O
mso
o
to
125
-------
§
a o
X JJ
in id
M
H Z
rtj ££
^
Z
0
u
H
s
w
w
M
«
O
b
U)
"^
^
*>^
0)
M
U
id
X
U)
Q
rH
«^>
tn
C
H
a
id
,3
5
id
a
c
0)
rH
>
H
3
CT
n
rH
id
4*
1
^
id
0)
8
C
i
rH
rH
VD
O
O
VO
H
rH
in
o
0
T O
o
r-
t*}
o
in
ro
en
o
0
o
JVC
m
, °-
N
H
O O
o m o
q- o o o
o
r**
rH H
O O
o in o
t
CO O
^t
H
cn fo
0 0
o in o
vo o <* o
ro o
°L °
o r
f^
EH .
10 b
a >
rH 01
o k
u u
A
u n
a J3
CrH
01
Oi-O
to id
,55
W i-J
o
o
o
o
o
o
o
o
o
o
o
o
o
o
in
o
o
o
o
CO
0)
m
D
a
10
id
M*
rH
id
H
u
M
0)
S
6
u
i
c
id
A
M
3
.a
3
w
rH
id
H
^J
r
)
a
H
n
to
i
C
id
a
M
3
A
3
U)
o
S
rH
a.
o
u
0)
M
3
jJ
rH
3
U
H
)H
Cn
<
««!,
CN
*«^
o
a)
a
o
rH
0)
^
0)
o
c
D
^
ro
^-*
JJ
e
0)
E
a
0
tH
0)
0)
a
3
0)
udy, October 1976
jj
CO
V
rH
,-j
£
\4
3
O
h
G
id
3
tr
o
u
u
0
c
H
O
01
JJ
e
(U
VI
0)
M
a
10
a
id
O
rH
C
0
-o
a)
in
«
a
^.^
rH
in
r-
o
rH
«
H,
idl
jj|
OM
rH
rH
(U
M
O
O
C
H
"3
0)
JJ
C
Q)
U)
O
M
a
ia
o
n)
O
rH
C
0
o
0)
in
fl
PQ
^.
rs
r-
r>
en
H
&
i
I
JJ
to
0)
rH
H
S
M
3
O
X.
c
id
3
8*
u
0
o
c
rt
a
o
jj
c
CO
01
M
a.
n
a
id
0
H
C
O
TJ
0)
Dl
id
CO
^-^
ro
>
»O
O*
H
*
K^
3
(9
C
H
a
a*
c
0)
10
4)
a
m
a
o
H
C
o
o
0)
V)
id
JO
n
H
id
jj
0)
e
S^
p
«
Q)
*-*.
*3«
*-»
U
g
H
.
a
0)
0)
lr»
s
w
rH
id
JJ
a
Q)
c
2
0
M
^
C
U
N
JJ
0)
a
0)
o
M
3
O
(Q
126
-------
plants are usually higher in available phosphorus and nitro-
gen while nonpoint sources typically have lower percentages
of available nitrogen and phosphorus. However, the nutri-
ents attached to or in the form of suspended solids can
settle in the slower-moving portions of the streams and re-
sult in subsequent nutrient enrichment as the particles are
dissolved.
Conservative substances such as heavy metals typically have
low removal rates.
Based on the above rationale the removal rates of pollutants
were estimated as fractional parts of those determined in
the steady-state water quality model for the Wissahickon
(Gabriel, 1972) and on available literature values for
typical stream decay rates.
Table 21 presents the K3 rates employed in the calibration
model (1968 conditions). These rates were applied uniformly
through the receiving streams.
TABLE 21
STORM DECAY RATES OF POLLUTANTS (K3) -
WISSAHICKON WATERSHED
Decay Rate
Parameter %/mile of stream
Sediment 0.0
BOD5 0.0125
Nitrogen 0.003
Phosphorus 0.003
Heavy Metals 0.001
127
-------
H. LAND MANAGEMENT CONTROLS
Land management controls include municipal regulatory con-
trols, maintenance controls and physical controls. A table
of physical practices for construction activities was pre-
sented in the methodology section (Table 2). These physical
controls are applicable to most regions of the country.
Municipal regulatory controls and maintenance controls are
listed in Appendix f. The controls are applicable to other
study areas. For the Wissahickon Watershed, the existence
of these regulatory controls in municipal codes was fairly
representative.
The effect of land management controls are manifested in the
K2 rate, which represents the reduction of the sediment
yield to the stream resulting from existing land management
practices. The practices included each of the three types
of controls noted above. The following equation represents
the logic of the management practice reduction factor (K2):
K2 = 1 - (KNLD + K0 + KA)
where:
1 = No reduction of load due to management
practices
%ILD = Reduction factor for management practices
applied to New Land Development
KO = Reduction factor for general land maintenance
practices, not including that applied to new
land development
KA = Reduction factor for management practices
applied to agricultural (crop) production
The following pages describe the derivation of the KNLD, KQ,
and K, reduction factors by subarea and land use category.
128
-------
1. New Land Development Control Effectiveness (KNLD)
A point system for relative effectiveness was devised for
potential new land development controls and is summarized
below:
Points Effectiveness of Regulatory Control
3 Significant load reduction
2 Moderate load reduction
1 Minimal load reduction
0 Zero load reduction
Table 22 lists each municipality in the study area and the
existence of new land development controls. Scores of rela-
tive effectiveness, as noted above, were assigned to each
control (see pp. 132-135 for derivation of effectiveness
scores). The final two columns show the total score for
each municipality and an associated "highest sediment re-
duction capability." As the table indicates, the City of
Philadelphia and the Townships of Springfield and Whitpain
in Montgomery County were associated with a highest sediment
reduction capability of 95% from new land development based
on the potential of their controls.
The means of deriving "sediment reduction capability" in-
volved the analysis of municipal regulatory controls in
effect in the study area for new land development. This
analysis focused on the types of physical controls required
by the ordinances which would in effect reduce sediment
movement from the construction site. With an understanding
of which physical controls were required (many of which are
the same as those listed on Table 2), the literature was
reviewed to determine documented evidence of control effec-
tiveness. Since the primary emphasis was on controls which
maintain sediment on-site, the controls of interest were
those which protect exposed ground, control runoff velocity,
filter sediment from runoff and contain sediment.
The highest reduction capability of 95% was associated with
a regulatory control score of 18 in three municipalities.
A review of each control program indicated a well developed
program to control sediment through what can be described as
129
-------
II H 1 1 H
H in
H H n 1-1
1 1 1 1 N 1 1
s a
s *
* g
s
il?
INI n I rl if r-IHIH IIIHM
II I I I I I I I H I I I I I I I n
1 1 1 1 1 1
t i I i I
I
M
*n ) i | i i i
i i i i i i
* i i i n I
C
0
ts)
sl g
C D
C 0
8>-l 1
P.
0) X
O *J C
S H 0
«d « -H v
C C J1 O
tJ D C <^
o v -S* -5
U' U) tl
C 2 Q 0
H QJ
§> Z C.
< -H 0
N * -H
M rH n
M Tj T3 -«
C X 8 ^
$ E £ W
* -0 *
§s §
U *> C C
C U O U -H
3 S -H 0 U -D
« Oj «J -^ SO
M U C t M C
O C C *w t 3 '-^ M c o> ^
4 [3 -H ^ P
w Q o S 2 .2 £ S -S -
M O H U
QJ
c
r^ C
-HO 1C
CM K
4J C *O
CO U
o o o
0
coo o
0 -H ^ -i
^t 4J 4J *J
JJ « C U
(O 1* O V
H 01 U 4J
^H (U 4J O
tH 0 *J «H Ji )
*J C QJ n U
O U 5 H Q) O
(J fl 3 W 1
VM O b U S
«M -H nj ^ fl n
O « 0'- 01 0) 0*
g o ^ -u n a>
3 M -P «J +J 4->
pi u w ;e w m
P
c
i
a
o
H
>
S
rH
rtj
r(
4)
0
n
c:
V
I
w
o
c
a
c.
rH
O
*J
C
J
a>
0)
Q
4-'
rH
G
g
C
C
^
Ul
4
>
>
Penns
n
s
^
q
o
41
0
§
m
0
w
V
o
rH
ii g
§*
**
U U
o al e
« -M U
§, i s,
a -H
a -o
« -H
-4 u -H
o « .a
1 1!
2 S
a -5
I I
I 3
V 5
s i
H
i i
130
-------
a system's approach, whereby a number of physical controls
are required as lines of defense. Each line of defense may
meet a certain objective and in combination the controls
reinforce one another. This systems approach to sediment
control has been documented in two EPA publications:
Comparative Costs of Erosion and Sediment Control,
Construction Activities/ Office of Water Programs
Operations, (EPA Contract No. 68-01-0755), July 1973.
Nonpoint Source Control Guidance, Construction Activi-
ties, Office of Water Planning and Standards, (Tech-
nical Guidance Memorandum: Tech-27), December 1976.
The original source of data on control effectiveness for
these two documents were the following:
Dow Chemical Corporation, An Economic Analysis of
Erosion and Sediment Control Methods for Watersheds
Undergoing Urbanization (C-1677), Final Report,
February 15, 1971-February 14, 1972.
County of Fairfax, Virginia, Erosion-Siltation Control
Handbook, Chapter 13 and Appendix, August 1972.
Each of these documents outlines promising systems for
sedimentation control employing various forms of physical
controls. The control systems documented were similar to
requirements of the regulatory controls of the three munici-
palities cited in this study area. The literature sources
indicated control effectiveness between 90 and 99%.
Based on this documentation, the most promising regulatory
controls in this study area were assigned "highest achiev-
able sediment reduction capability" of 95%. This figure, in
turn, was used as a benchmark against which the controls of
other municipalities were measured. A linear scale was used
whereby 18 points achieved 95%, 9 points achieved 48%, 6
points achieved 30% and so forth.
A number of qualifications have to be made regarding this
approach and the use of documented data. The benchmark
figure of 95% is probably high for this study area and most
others. The primary reason for such skepticism is that its
use for measuring "regulatory control" effectiveness does
not account for the probability of implementation, which is
a variable of municipal and developer responsibility. None-
theless, as a measure of control "capability," it was deter-
mined that the absolute number was not as significant to the
131
-------
methodology as the relationship of control effectiveness
between municipalities. In an effort to illustrate varia-
bility between municipalities, the benchmark of 95% was
useful. As guidance for other study areas, documented data
should be reinforced by study area-specific data on the
effectiveness of controls. The development of site-specific-
data was beyond the resources of this case study.
(A) Derivation of %fLD by Subareas: Since the methodology
is developed on a subarea rather than municipal basis, the
load reduction factors for new land development by munici-
palities were converted to a subarea reduction factor (see
Table 23). Basically, the subarea factor is derived by
determining the percent composition of each subarea by
municipality. For example, in M4, since Lower Gwynedd is
80% of the subarea land area, its reduction factor (.68) is
multiplied by .8. Since North Wales is 20% of the subarea
land area, its reduction factor (.30) is multiplied by .2.
Thus, for subarea M4, the new land load reduction factor is
calculated as follows:
Lower Gwynedd (.68) x (.8) = .544
North Wales (.30) x (.2) = .060
Total Subarea 600
2. Maintenance Control Effectiveness (KO)
Table 4 in Section III illustrated the maintenance controls
and point ranges associated with each control. In this case
study, the sediment load reduction factor for maintenance
control effectiveness in developed areas was not determined
for each minor civil division. Instead, scores were deter-
mined for five land use categories, excluding new land
development. Table 25 illustrates the finding of a higher
incidence of such controls in urban areas and a lesser
representation in each of the other land use categories.
Refer to Table 4 in Section III for the point range of each
type of control.
132
-------
TABLE 23
CONVERSION OF MUNICIPAL REDUCTION FACTOR FOR
NEW LAND DEVELOPMENT TO SUBAREA REDUCTION FACTOR
New Land Development
Subarea Subarea Composition Reduction Factor
M-1 Abington 90% .74
Upper Dublin 10%
M-2 Upper Dublin 100% .74
M-3 Upper Dublin 50% .78
Whitpain 30%
Whitemarsh 20%
M-4 Lower Gwynedd 80% .60
North Wales 20%
M-5 Lansdale 50% .59
Montgomery 20%
Upper Gwynedd 30%
N1 Whitemarsh 50% .76
Springfield 40%
Philadelphia 10%
01 Philadelphia 100% .95
TABLE 24
EXISTENCE AND DEVELOPMENT OF MAINTENANCE CONTROL PROGRAMS
Control
Program Type of Control U SC SR
A General Land Use Policy 1111
B Public Cleanliness 2 2 1
C Controls on State and
Local Road Building 1 1 .5 .5 .5
D Controls on Maintenance
Construction 2 1 .5
133
-------
An assumption was made that control programs A and B would
typically be applied to activities generating low levels of
sediment and accounting for only 25% of the non-new con-
struction sediment load. Programs C and D together were
assumed to address major activities generating about 75% of
this projected load. These activities would include road
building and maintenance as well as structural renovations
and alterations.
Table 25 illustrates the distribution of the points for the
control programs A-B and C-D for each land use category.
Table 26 converts these points into a percent sediment load
reduction factor.
TABLE 25
MAINTENANCE CONTROL PROGRAM SCORES BY LAND USE CATEGORY
Suburban Suburban Agricultural/
Urban Commercial Residential Cropland Forest
Programs A and B 3 2 2 1 0
Programs C and D 3 2 1 .5 .5
TABLE 26
LOAD REDUCTION DUE TO MAINTENANCE CONTROLS
Suburban Suburban Agricultural/
Urban Commercial Residential Cropland Forest
Programs A and B 15% 10% 10% 5% 0%
Programs C and D 50% 33% 17% 8% 8%
134
-------
Table 27 takes into account the percent sediment load
reduction factor and the percent of total sediment generated
by activities under that program. The addition of the pro-
ducts WX and YZ yields a single composite reduction factor
for existing development.
TABLE 27
COMPOSITE MAINTENANCE CONTROL REDUCTION FACTOR
BY LAND USE CATEGORY
% Reduction A + B
Programs: W
% of total sedi-
ment generated
by activities
under A + B: X
Suburban
Suburban Agricultural/
Urban Commercial Residential
15
10
10
Cropland
05
Forest
00
25
25
25
25
25
% Reduction
C+D programs: Y .50
% of total
sediment gener-
ated by activi-
ties under C +
D: Z .75
,33
.17
.08
.08
.75
.75
.75
.75
Composite Reduc-
tion Factor
(WX x YZ)
.41
.28
.15
.07
.06
3- Agricultural Control Effectiveness (KAG)
Farmers in the study area, in cooperation with the County
Conservation District and the USDA Soil Conservation Ser-
vice, were engaged in a relatively successful program of
erosion and sedimentation control in 1968 (the calibration
135
-------
year). Consequently, a reduction factor of .60 was derived
for all agricultural land in the study area. The .60 figure
represents a land conservation program of 90% sediment
reduction existing on 2 out of every 3 acres of agricultural
land and 1 out of 3 acres not under any conservation program.
4. Land Management Control Effectiveness (K2)
Results of K7 derivation for study area: Using the equation
= 1 - (K
NLD
+ K) and based on the preceding deriva-
tion of each variable, Table 28 was constructed to summarize
K2 values for each land use category in each subarea.
TABLE 28
EXISTING LAND MANAGEMENT CONTROL RATES
WISSAHICKON WATERSHED
(K2) -
K2 Rates: 1968 Conditions
Subarea
M1
M2
M3
M4
M5
N1
01
U
59
59
59
59
59
59
,59
SC
72
,72
72
72
,72
,72
72
SR
.85
.85
.85
.85
.85
.85
.85
.26
.26
.26
.26
.26
.26
.26
A
.94
.94
.94
.94
.94
.94
.94
NLD
.29
.26
.22
.39
.41
.24
.05
Legend:
U
SC
SR
F
A
NLD
Urban
Suburban Commercial
Suburban Residential
Forest
Agricultural/Cropland
New Land Development
Source: Betz Environmental Engineers, Inc.
136
-------
I. MASS BALANCE ANALYSIS
Based on the loading factors, calibration data sets and
verification data sets previously presented, a mass balance
model was constructed. The mass balance model uses the
factor involved in the universal soil loss equation as ex-
plained previously in Section III. The model was calibrated
to the 1968 land use, rainfall and water quality conditions
and was verified to the 1964 land use, rainfall and water
quality conditions.
1 . Calibration
Using the raw loading factors and the modifier previously
described, a computer mass balance model was used to calcu-
late sediment loads at the summation points. The resultant
loads, calculated by applying the various factors to the
land uses in each subarea, agreed to within 10% of the
measured sediment load at summation point 1 (Ft. Washington
Station) . Minor adjustments were then made in loading
factors to obtain exact agreement.
Next, the KN ratios were employed along with the 1968 sedi-
ment calibration computer run to obtain loads for other
pollutants, i.e., BODc , nitrogen, phosphorus and heavy
metals. The results of the calibration runs for sediment
and other pollutants are presented in Table 29. The actual
computer runs showing details of the total loads are con-
tained in Appendix B and sample calculations are shown in
Appendix H.
^' Verification
The land use acreage was adjusted to approximate the ex-
pected 1964 (verification year) land use. The 1968 rainfall
factors based on flow and rainfall were incorporated and the
computer model was run. The results of the verification
runs are presented in Table 30. Printouts of the computer
runs are presented in Appendix C.
137
-------
Cft
CM
a
i
EH
r J
X
W
w
EH
jrf
S
Z
O
X
o
H
*T1
^
CQ
(A
H
S
X
rf
g
w
to
Q
S
25
O
H
EH
£
(Q
H
l-l
U
n
4J
C
H
O
C
O
4J
n)
S
3
w
(N
4J
rl
O
C
O
iH
4J
n)
g
g
^
CO
rH
-p
c
-H
O
ft
0
H
4J
Id
5
3
CO
M
0) >
0) -PX
O* id Q)
id « Vi
M U
o> -o
0) -P\
O^ fd Q)
(S A H
^ O
a) -o IB
^ id *^x
< o w
ij rj
iH
*+*
ff^
»dvo
rd 0
O »H
^
X
fH
(d w
EH '*^'
"57
a) 4J\
0> (0 Q)
Id « r4
Wl U
a) -a id
t> id \.
< o m
i-^ *Q
rH
«-^
.^^
ovo
id o
O rH
X
rH
id ifl
-P .Q
°d
CM
rH
00
CM
ro
CM
CM
^*
vo
in
r-
co
in
(N
CM
rH
4J
C
0)
e
a
0)
CO
r** o\ f> o
^O PO t** rH
CM
P»
00 O rH CO
n 01 r» CM
O O rH O
O
VD
r> CM ^- o
H
r^
CQ
X
H
f£
T rH CO IU
o ^* co vo fx
VO VD CM rH Oi
*3* O rH O |4J
O
0)
c
3
M
0)
-P
3
* ft
o r^ CM * g
0
(N rH 01 O O
rH
0
rl
^j
n)
Q
H
rH
id
«3- 00 ^ O
^* CO CO rH
rH ^ 00 rH i m
in rl >
Q O 4^ id ^^
O £ -H 0) rH
o
c
H
CO
M
0)
C
H
0»
u
rH
id
4J
a
Q)
g
2
o
M
H
>
C
pa
N
P
(1)
m
01
u
n
3
O
CO
138
-------
o
m
W
1
rH
""£j
w
SB
CO
«
H
EH
S
3
O
0
K
rij
CO
CO
H
5
l
CO
EH
l^]
D
CO
W
05
Z
O
H
CJ
H
fc.
M
O
M
M
H
vo
o
X
(0
rH
O
10
s
o
0)
Ifl
rH
3
O
rH
Ifl
O
4J
C
o
(Xi
c
o
H
4->
Ifl
§
W
i
i
i
CM
n
CM
CM
CM
4J
C
O
A.
§
-rl
4J
2
g
3
CO
t^
00
vo
rH
rH
4J
C
-rH
O
a.
c
o
4J
|
co
in
CM
«
CM
r-l
00
oo
00
at
o
in
fM
o
oo
o
03 U
n)
n
03
p-
oo
oo
0)
X
T
vo
en
§
(0
4J
10
o
l-t
^J
Ifl
M-l
C
H
<0
o
c
(0
^^
u
X
rl
o
c
0)
C&4
pt
rtj
^-*
to
c
3
>H
^
(U
4J
3
g
O
c
o
H
JJ
10
o
H
MH
H
14
0)
>
MH
O
vo
QJ
rH rH
3X1
01 10
Q) EH
^1
M-l
C 0
o
in
o
Q
EH
0
4J
^
vo
en
rH
M-l
o
s
O
^H
g
R3
a)
CO
rH
ia
OJ
Q
f-t
MH
O
O
r-l
4J
10
V^
C
O
*o
(U
(0
(0
,Q
L4
o
4J
O
(0
Pn
]
Di
^^
rH
H
rH
10
MH
C
rl
&
rH
(0
4J
O
MH
O
O
rl
4J
10
^|
C
o
o
(1)
in
10
XI
l^
o
Jj
o
(0
[i*
1
«
X-o.
ro
*-^
I
3
n
4J
Ifl
vo
0
rH
X
,J,
00
(J\
*~*
d
ifl
o
rH
0
10
rl
c
o
o
0)
to
10
^^
HrH
O
M 4-1
M C
W-H
o
* a
^^,
^-
u
c
H
^
n
^
0)
c
u
N
0)
03
0)
0
IH
3
O
CO
139
-------
The verification results agreed closely with the 1964
measured sediment load. There was a 10.7% error when using
an R-factor based on ratios of total rainfall and a 4.3%
error when using an R-factor based on ratios of total stream
flow.
It is expected that the R-factor based on flow ratios pro-
vided better results because flow measurements for both 1968
and 1964 were made within the Wissahickon Watershed at the
Ft. Washington station, whereas rainfall measurements were
made in watersheds adjacent to the Wissahickon Watershed.
Rainfall measurements for 1968 were collected at the North
Philadelphia Airport (approximately 15 miles east of the
center of the Wissahickon Watershed). Rainfall measurements
for 1964 were collected at Phoenixville (approximately 12
miles west of the center of the Wissahickon Watershed). The
NOAA precipitation records represented the best available
rainfall measurements for the area.
J. WATER QUALITY IMPACT ANALYSIS
The effects of nonpoint source pollutants on aquatic biota
were analyzed based on (1) available biological data (2)
water quality data, and (3) results of the field surveys.
The basic steps followed were:
0 Obtain water quality data indicative of both low
flow and high flow conditions.
e Determine parameters present in concentrations
potentially harmful to biota during storm runoff
(high flow periods).
9 Analyze existing biological data to determine
degraded areas that appear to be caused by non-
point sources.
9 Conduct field survey of areas identified in Step 3
and determine whether they are affected by non-
point sources.
140
-------
1 Chemical Water Quality Analysis
A comparison of available water quality data on the Wissa-
hickon Creek for steady-state (low flow) and average (high
flow) conditions (Table 31) provides an approximate indica-
tion of the impact of storms on water quality. The steady-
state suspended solids concentrations are much lower than
the average concentration. This pattern indicates that high
sediment loads occur during high flow periods. The concen-
trations do not differ between stations, indicating that a
relatively consistent amount of sediment enters along the
course of the stream. Nutrient concentrations are lower or
approximately the same at high flow as compared to low flow,
indicating a dilution factor. Although absolute pollutant
loadings in Ibs/day may in fact be higher at high flow, the
immediate impact of a substance on the biota is primarily a
function of concentration. The BOD data follow no consis-
tent flow pattern, but concentrations decrease downstream as
a result of biological consumption.
This cursory examination of the Wissahickon Creek chemical
data indicates the greatest nonpoint source biological
effects are from sedimentation and scouring during storm
periods, and possibly from a reduction in dissolved oxygen
concentrations as a result of BOD inputs. In order to
determine whether the nonpoint source pollution indicated by
the previous data is in fact affecting the aquatic biota, it
is necessary to compare the existing biological data col-
lected at stations affected by point sources with data
collected in areas not affected by point sources. Figure 18
illustrates the stations sampled in a previous survey and a
summary of the general biological conditions at the time the
data were collected. Figure 18, in conjunction with the
point source locations shown in Figure 19, indicates those
areas and conditions influenced by point source discharges.
These data indicate organic enrichment is present at Station
2, above any known point source discharges. From Stations
3-5 point source pollution is so severe it masks any effects
nonpoint sources may have. Nonpoint source organic pollu-
tion undoubtedly exists in this area; however, the precise
nonpoint source pollution effects will not be apparent until
the point sources are eliminated or reduced.
The effects of nonpoint source pollution are also difficult
to assess further downstream because of the upstream point
source loads. There is some observed improvement in bio-
logical quality from Stations 6 to 8, in the absence of
141
-------
§
\
Cn
G
C
O
rl
x>
(0
VI
4J
C
0)
o
c
o
u
n
X)
c
H
O
Oi
c
o
rl
I
CO
(N
-P
C
H
O
Pk
G
0
H
4J
(0
g
g
rj
CO
M
X>
C
H
0
C
o
rl
4J
|
3
CO
(1)
en
10
VI
V
*c
o O * CM CM t-
co r-
in CM r-i CM
r- oo H in
ro
rH
Cft
CN rH in VO
lo
CO 00 rH p- CM *3" CM
p» oo
o
vo m en co
O ^ rH t- O
CO rH
"^
00
r-
0)
JJ
01
4J
a)
>rj
O
O
CO
0)
3
M
O
a
CO
o
rl
Z
K O
t^
o
CM
CM
CO
O
(0
I-l
(d
4J
0)
s
£*,
>
01
E£
of calibration
0)
jj
H
9
n
o)
d
§
vo
«)
rH
a
EH
O
CM
O
10
rH
10
e
0)
4J
H
C
rl
o
0)
c
rl
rl
HI
a
^_*
rH
*.^
m
X
rl
d
c
a,
^
c
H
to
rj
3
^
Vl
4J
rj
a
E
O
o
CO
c
o
rl
4J
<0
4J
c
Q)
O
q
o
o
S
o
rH
IW
S
0
rH
at
-U
rd
4->
D)
>1
a
cu
XI
CO
, ,
CM
**
rH
10
4J
c
cu
o
c
o
o
rH
10
y
c
c
n)
V
m
Vl
0)
>
^^
CO
0)
o
p
m
S1
H
4J
C
rl
a
0)
(0
o
c
0
rl
P
14
4J
C
0)
o
c
o
u
£
.,-t
J5
n
u
XI
iO
u
rl
O
c
rl
g
1)
4J
(0
,«*
^"
^»
u
H
«
n
Vl
0)
4)
a
rH
O>
C
H
rH
10
P
Q)
g
2
o
Vl
H
>
c
u
N
4J
s
0)
u
Vl
3
O
10
142
-------
o
c
en
c
to
4->
C
c
o
c
Ul
N
0)
00
O
1-
3
O
t/1
«U. rinding*
1 little organic degradation
2 heavy algae growths
3* badly polluted by eraanic wastes
.. a. tome toxic and silt pollution
( some recovery
tussive algae growths
7 more recovery, healthy
heavily enriched
considerable algal growth
8 healthy less heavily enriched
alqal growths
9 lower quality through Station 8
heavy organic enrichment
sediments and nild toxic
pollution
better than Station 3, 4, and 5
Findlnfi
10 degration from Sandy Run
11 considerable recovery
from organic pollution
12 no further recovery
sediment deposits
overflow from Paper Mill Creek
interceptor
13 very polluted
heavy organic pollution
14 some recovery but still
severely degraded
15 s»re recovery but still
severely degraded
/"- 9.5OO'
HGURE 18
MAP OF BIOLOGICAL SAMPLING SITES
WISSAHICKON WATERSHED
143
-------
o
c
-------
severe point source pollution. It is difficult to determine
whether the extent of recovery is being affected by nonpoint
source effects. Wissahickon Creek from Station 8 through
10, including the Sandy Run area, exhibits a pattern similar
to that discussed above: severe biological degradation as a
result of point source pollution, following by partial
recovery.
It is not until Station 11 that point sources do not dis-
charge into the stream. No noticeable improvement is ob-
served between Stations 11 and 12, indicating nonpoint
source pollution may be affecting the biological quality of
the lower reaches of the stream.
Based on this assessment of the available data, it appears
that significant parts of Wissahickon Creek are affected by
nonpoint sediment loading. Portions of the stream are
probably affected by nonpoint source oxygen demanding wastes
and other pollutants but their effect is masked by the point
source loads.
2. Biological Analysis
A cursory study of the watershed was performed to investi-
gate the stream biota. Areas expected to be affected by
nonpoint sources were emphasized. The results of the
survey are presented below.
(A) The headwater area above known point discharges is
biologically unproductive, but does not appear to
be seriously affected by nonpoint source pollu-
tion. The main reason for this conclusion is the
absence of siltation in the area. The probable
reasons for the low productivity is the intermit-
tent nature of the stream in the headwaters and
possible scouring from a recent storm (this would
be enhanced by nonpoint source runoff).
(B) Downstream, in the Ambler area, the stream does
not appear to be seriously affected by nonpoint
source pollution. Invertebrate organism diversity
and algae growth did not appear to be reduced.
Invertebrate organisms characteristic of good
water quality conditions were present. The main-
stream in this area does not appear to have a
siltation problem, although it was apparent that
145
-------
heavy sediment loads are washed into the stream
from feeder streams and stormwater runoffs. The
sediment loads are apparently carried downstream
by later storm events.
(C) The Sandy Run tributary upstream from the known
sewage discharge appeared to be nutrient enriched.
Algae growth appeared relatively heavy for the
time of year and for following a recent heavy
storm. The invertebrate community was greatly
reduced and consisted of tolerant organisms. The
source of the enrichment was not obvious and no
sedimentation was apparent.
( D) The stream appeared heavily degraded downstream
from the sewage treatment plant on Sandy Run.
The absence of siltation indicated nonpoint source
pollution was not a significant problem; however,
there was evidence of high sediment loads being
washed in from construction sites and storm
drains. Like the Ambler area of the Wissahickon
these sediment loads were washed downstream.
( E) At both the Bells Mills and Valley Green area
below the confluence of Sandy Run with the Wissa-
hickon Creek, water was a gray, murky color and
biota were reduced and indicative of a degraded
environment. In this reach, the stream consists
of a series of riffles and pools. The pools
collect large deposits of sediments which destroy
the biological habitat. These sediment deposits
are attributed to the upstream nonpoint source
sediment loads in the Wissahickon Creek and Sandy
Run.
( F ) Further downstream below the confluence of Wis-
sahickon and Cresheim Creeks and the dam, bio-
logical conditions appeared to be quite good.
Both invertebrate and algal conditions were in-
dicative of good water quality and no sedimenta-
tion was apparent. Above the dam, habitats were
reduced and some sedimentation was apparent,
indicating the dam was providing an area that
collected sediment.
146
-------
(G ) Cresheim Creek, which drains an urban area, but
has no point sources, was in fairly good bio-
logical condition. The invertebrate community was
quite diverse, and there were no signs of storm
scouring. There was a small amount of sediment
present.
(H ) At the mouth of the Wissahickon and approximately
1 mile upstream from the mouth, near a series of
dams, there was heavy siltation and the biota were
greatly reduced. Available habitats were heavily
silted which was the probable reason for the
absence of aquatic biota.
3. Summary of Analysis
In summary, the results of the water quality analysis indi-
cate that:
(A ) The temporary effects of nonpoint pollutant runoff
do not have any noticeable effect on biological
conditions. Stream concentrations of pollutants
are not increased because nonpoint pollutant loads
are accompanied by increased dilution flow.
(B ) The Wissahickon Creek from the confluence of Sandy
Run to approximately 1/2 mile above the mouth
suffers from nonpoint source sediment loads. The
habitat in pools and directly above impoundments
are destroyed by settling of sediment loads.
These loads result from upstream runoff from
construction sites and storm drains.
{ C ) The biological condition at the mouth of the
Wissahickon (last 1/2 mile of the stream) are
completely destroyed by sediment loads. The slow
stream velocities in this reach result in settling
of large deposits of sediment.
( D) Even if point source pollutants were eliminated a
significant portion of the Wissahickon Creek from
the confluence of Sandy Run to the mouth will not
recover biologically. Siltation in pools, areas
above impoundments and the mouth region will
continue since these problems result from nonpoint
runoff loads.
147
-------
(E) The total nutrient loads from nonpoint sources are
large enough to significantly contribute to down-
stream eutrophication problems as indicated by the
concentration of phosphorus and nitrogen resulting
from runoff.
K. SENSITIVITY ANALYSIS
The sensitivity of land use, land management controls and
loading factors were explored for all subareas. Table 32
presents a summary of the sensitivity analysis of the total
study area. Results indicate that under existing conditions
50% or more of the loading is controlled by natural factors
such as slopes, sediment delivery ratio and natural con-
trols. The existing BMP controls provided approximately 50%
reduction in pollutant loadings.
Subarea M3 was analyzed to determine the sensitivity of land
uses to runoff loads. This subarea was selected because it
represented an area in which significant development was
taking place; thus.it would reveal sensitivity to urbanizing
factors. The results of this analysis are summarized in
Table 33. Fifty percent of the urban land use sediment load
is controlled by existing land management practices while
only 26% of suburban commercial and 12% of suburban residen-
tial loads are controlled. Only 34% of construction loads
resulting from new development are controlled. As such, the
table indicates that the effectiveness of controls in sub-
area M3 is below the average of other subareas. This raised
the question of how much improvement could be obtained in
subarea M3 if the controls were upgraded. Consequently, the
best controls in the study area were applied to the land
uses in subarea M3 and the mass balance program was run.
The results are summarized in Table 34. As can be seen,
under existing land management practices 693 Ibs/acre/year
of sediment loading are controlled in the urban land loads.
An additional 323 Ibs/acre/year could be controlled by
implementing best management practices in urban areas. New
development land uses (areas under construction) showed the
most promise for improvement. Under existing management
practices, 968 Ibs/acre/year are controlled, whereas using
best management practices for new development (construction
BMPs) would result in an additional 2,255 Ibs/acre/year
being controlled (reduced).
148
-------
0)
rd
CU
fd
00
CN
00
n
o
o
C
H
A
-P
H
CN
CO
W
cn
H
cn
>H
EH
H
>
H
EH
H
cn
w
cn
rti
MH
m
CO
&
P
o
fd
&
H
in
O
dP
rd
cu
rd
cn
10
rd
cu
CO
cu
-P
cu
g
rd
SH
rd
a)
en
C
O
-rH
O
in vo U
Pi ft rd
~ V-l
>1 ft
M CQ
CU rH -p
> o c
H M CU
rH -P g
cu c cu
Q O &i
U rd
-P G
C rH rd
CU rd g
en H
H 3 -P rd
T3 -P W -P
CU rd 0) O
cn s PQ EH
d
cu
d
rd
0
rH
C
o
CD
n
Q)
-p
cu
g
rd
n
fd
c u
CU G
> H
H
Cn
tn
MH M
O CU
cu
-P C
O -H
fd C71
PJ C
g H
-H
rH
CU rd
> -P
H G
-P CU
rd rd S
rH CU G
CU >H O
M rd in
A -H
CU rj j>
.G 0) G
-P W
Tl
03 CU N
CU -P -P
X! 03 CU
H -H cq
(l 1
u
03 CU
CU A CU
Q -P D
!q
^^ 3
rH O
-^ cn
149
-------
TABLE 33
SENSITIVITY ANALYSIS - SEDIMENT - SUBAREA M3
% of Impact on Runoff Load
(1)
Parameter
Slope Length (LS)
Sediment Delivery (SD)
Natural Controls (P)
Best Management
Practices (K2)
Total
Suburban
Commer-
Urban cial
9
31
10
50
100
7
62
5
26
100
Suburban
Residen-
tial
18
48
22
12
100
Forest
18
51
27
100
Agricul-
ture/
Cropland
8
46
17
29
New
Develop-
ment
14
42
10
34
100
100
(1) Describes the relative impact of given parameters on
load reduction within the listed land use.
Source: Betz Environmental Engineers, Inc.
In terms of impact to total load from the subarea, BMP
controls applied to construction would reduce the load from
2.447 million Ibs to 2.093 million Ibs/year. This repre-
sents a 15% load reduction for controls applied to only 2%
of the land use within subarea M3. This indicates that
cost-effectiveness results could be obtained by implementing
construction controls. The analysis previously presented
for construction controls in subarea M3 illustrate the use
of sensitivity analysis in developing criteria for formula-
tion control plans. Similar analysis could and should be
applied to other controls and other subareas.
L. CONTROL PLAN DEVELOPMENT
The next step in the methodology is to evaluate nonpoint
source controls. As noted previously, nonpoint source
controls should be selected and the resulting pollutant
150
-------
*3>
n
S
0
tt
H
m
2
H
B!
S
3
in
i
EH
[V]
H
Q
CO
«
W
§
g
o
u
[V]
CO
Q
^
^i
M
M
M
CO
Z
U
tn
C
o
rl
0
3
O
01
B
O
10
3
jj
c
0)
S
a,
o
01
01
Q
J
0)
Z,
50
rH
C X
O Ll
ox
a
33
n
jjj X^
So
L4
u
v a
35
f-i
Is
O
t) id
S3
0)
co
B i-l
U M
O-s
10 U)
S3
(U ?*
4JV.
10 01
U
O rO
Sen
.a
CN
in
i-l
o
00
\f)
o>
m
m
o
o
m
M
in
f*>
rj1
O
o
m
<^i
CM
CN
CO
.
o
rn
(Ti
V£>
tn jJ
01 tn en
emu
ro -H
C 0-1 JJ
C
o
jj
a
H
M
O
u)
0>
D
u o o
(0
T3 C f-f
(0 O ft
O-H
.H -U 4J
u c
tT1 £ 1)
C 3 E
-H I4H QJ
JJ CP
tn re to
H C
x tn (0
W 10 E
u
01
JJ
Q)
P
s
id
(X
r~>
rH
*^
CN
*
1
0
3
01
Ll
a
10
o
rH
a
c
o
rH
JJ
H
T3
Tl
<
^M.
CN
^^
CN
«
rl
01
a
c
3
01
rH
rj
H
tn
tn
o
a
c
o
H
01 Ll
01 o
CQ IIH
IM co
O rH
0
01 Ll
JJ JJ
H C
6 0
H CJ
r-H
rH C
10 (1)
0 E
H 0)
JJ tn
U to
rO C
VH ro
as
^1
m
m
o
in
in
CN
cs
m
CN
rH
i
O
in
CO
n
m
0
0
CM
r-l
en
m
o
m
CN
ro
S
01
C
o
c c
10 O
rH
T3 JJ
C Ll E
C -H O
jj u c oi
rH 3 O >
3 -a e oi
tn o o a
3J Li C1
Ll rO C
O C id
C 10 10 rH
O 0 E CU
H H
JJ JJ rH
O rH M O
3 «rj 0) Li
a c « -P
0 O C
U -rl O
JJ (_>
10 t3 S -
O -O ^
rH 10 Id
OI.C
tn 01 LI a<
o -a JD M
H -H 3 &
> > 01 id
O O Ll
Ll LHH Id
a. a. o o.
rH fN
u
c
H
n
Ll
01
B
H
en
B
r-l
Q
JJ
C
g
2
o
Ll
H
B
H
N
JJ
0)
A
l
O
U
3
0
en
151
-------
reductions obtained. Nonpoint source and point source
controls should be run through the water quality analysis in
a reiterative process until the required pollutant levels
are obtained, or until the best controls have been applied
and additional improvements in water quality are not prac-
tical or obtainable. This reiterative process was not per-
formed in this case study because it was beyond the scope of
the study. The first step in this reiterative process,
selecting and evaluating the most cost-effective nonpoint
source controls, is presented in the following sections.
The water quality impact analysis indicated that sediment
and nutrients (nitrogen and phosphorus) are the principal
water quality pollutants associated with nonpoint sources.
Therefore, the control plan must address these problem
areas. The results of the sensitivity analysis were helpful
in exploring and determining alternative control plans to
accomplish this.
In the development of a control plan for the Wissahickon
Watershed, the choice of types of controls for varying
settings had to be understood. Urban, suburban commercial
and suburban residential land use categories were subject
only to "maintenance controls." These controls are exer-
cised at developed sites undergoing renovation or altera-
tion. They include labor and machine-intensive work in
conjuction with on-site application of physical facilities
or structures. The control plans for each of these three
land uses were assumed to improve sediment reduction from
M68 to 1985 and from 1985 to 2000. (Consequently, the K2
rates associated with each connotes this change.) The 1985
control plans were assumed to include improved controls on
state and local road building and repair as well as munici-
pal regulatory controls on maintenance construction. By the
year 2000, improvements in public cleanliness technologies
(i.e., street cleaning, sewer flushing) employed as controls
would be of sufficient worth to achieve the .40, .50 and .50
K2 rates for urban, suburban, commercial and suburban resi-
dential land uses, respectively.
Forest and undeveloped land were subject to no controls
except for maintenance control programs for state and local
road building. Agricultural lands were subject primarily to
sedimentation and erosion control practices. These lands
also were affected by maintenance control on state and local
road building. The control plans for undeveloped and agri-
cultural lands were assumed to be constant for the years
152
-------
1968, 1985 and 2000. Consequently, the K2 rates associated
with each remained constant as well.
For new land development, the K2 rates were derived from
documented data on land development controls of each munici-
pality (see Table 22). This was then converted to a subarea
rate (see Table 23). In all instances, except for subarea
0, the K2 rate (identical in this case to KNLD) was postu-
lated to improve for the years 1985 and 2000. At the year
2000, all controls on new land development were assumed
equal at a K2 rate of .05. The types of controls necessary
to achieve this K2 level are exemplified in section L.1.
This section notes that the highest rating of 95% reduction
(equivalent to a K2 of .05) was achieved in Springfield
Township. The controls in place in Springfield are listed
in that section. A comparable combination of controls are
postulated to be present throughout the study area by the
year 2000.
In Table 35, the existing and projected K2 rates are pre-
sented for each land use category. The 1968 rates are based
on existing control potential, subject to certain judgmental
considerations relative to implementation probability. The
years 1985 and 2000 K2 rates were arbitrarily selected as
performance goals for those time periods. As discussed in
Section III, there are innumerable combinations of new land
development controls, maintenance controls, and agricultural
management controls which, in relation to one another, could
achieve these desired levels of in-stream sediment and other
pollutant yields. The physical controls listed in Table 2
can likewise be combined in innumerable ways to achieve a
desired response.
1. Year 1968 Control: New Land Development
The most basic municipal regulatory controls (for new land
development) instituted in this study area in 1968 included
flood plain designation and management provisions, general
subdivision ordinances, and sedimentation and erosion con-
trol regulations. In the Commonwealth of Pennsylvania, the
regulations on sedimentation and erosion control are man-
dated and administered directly by the state through a
permit system and through County Conservation Districts on a
plan development basis. All municipalities are subject to
this control.
153
-------
TABLE 35
K2 RATES
(1)
Land Use 1968 1985 2000
Urban .59 .50 .40
Suburban Commercial .72 .63 .50
Suburban Residential .85 .68 .50
Undeveloped .94 .94 .94
Agriculture/Crops .26 .26 .26
New Development Ml .29 .17 .05
New Development M2 .26 .16 .05
New Development M3 .22 .14 .05
New Development M4 .39 .22 .05
New Development M5 .41 .23 .05
New Development N .24 .15 .05
New Development 0 .05 .05 .05
Source: Betz Environmental Engineers, Inc.
(1) These reductions indicate no change in either undeveloped or AG/Crops, K2
rates, however, different strategy and actual water quality needs may
necessitate or be more cost-effective than this control mixture.
With a few exceptions, each of the other two regulatory controls existed in each
municipality. Consequently, a basic level of total load reduction, associated
with having these controls, was assumed. This basic level was a 40% reduction
factor for the flood plain, general subdivision, and sedimentation and erosion
control regulations. Only North Wales, which did not have a subdivision ordinance,
scored below this level. The 40% factor took into account the assumption that
implementation of these basic controls was highly probable. While each of these
controls actually differ (with the exception of sedimentation and erosion control)
in terms of procedures by municipality, they were reasonably similar so as not to
differentiate at that level of detail.
Higher reduction factors were achievable if other ordinances and ordinance com-
ponents were in place in these municipalities which would have beneficial water quality
impacts. Examples of these components in zoning ordinances included mixed or
average density provisions, steep slope designation or agricultural district designatic
In subdivision ordinances, these provisions included surface drainage
154
-------
plans, net runoff limits, structural requirements on-site
and maintenance requirements. Finally, the use of other
ordinances were accounted for, including steep slope ordi-
nances and PRD or cluster ordinances. The highest rating of
95% reduction, in Springfield Township, for example, in-
cluded, beyond the basics:
Mixed or average density zoning provision
Flood plain designation zoning provision
Net runoff limit subdivision provision
Required structures for runoff reduction on-site
Required structures for sedimentation and erosion
control on-site
Required maintenance
Required connection to public services
Watercourse alteration criteria - flood plain provision
Stream setback requirement - flood plain provision
2- Year 1968 Control; Existing Development
For existing development, emphasis was on the use of main-
tenance controls for state and local road building and
maintenance construction. The use of such controls in this
study area was prevalent in the urban areas where renova-
tion, alteration and overall improvements were most promi-
nent. In some of the developed suburbs, where sediment
generation was not relatively great, the existence of such
controls was minimal, particularly for agriculture and
undeveloped land uses.
3. Projected Controls - 1985 and 2000; New Land Development
As stated in Section III, two approaches for projecting
future levels of control are to either delineate the use of
specific physical controls in a municipal ordinance or to
set performance standards on the ordinance which leave the
perogative for control choice to the developer. In this
case study, the performance standard approach was selected.
155
-------
Since some municipalities in the study area had shown evi-
dence of significant sediment reduction based on existing
controls, it was assumed that all municipalities could
achieve the K2 level of .05 by the year 2000, with an in-
terim goal set for the year 1985. The technology for
achieving these goals in this study area were believed to be
available or potentially implementable. In the section on
loading factor sensitivity it was made apparent that the
land management controls to be placed on new land development
by the year 2000 would result in the most significant in-
stream sediment load reduction of all land use areas. Con-
sequently the most cost-effective concern for control devel-
opment, from the point of view of sediment reduction, would
be controls placed on new land development. Once again, the
level of regulatory control development cited for Springfield
Township is indicative of the control package necessary to
achieve performance standards as postulated here.
4. Projected Controls - 1985 and 2000: Existing Development
Unlike new land development controls, the information and
technology for addressing pollutant generation from existing
development is not as far advanced, although both forms
require similar physical controls. The problem in this case
study was one of assessing the pollutant generation prob-
lems, addressing the problem, and implementing the control
program. The pollutant generation problem of existing
development was associated most prominently with urban
areas. The raw sediment production rates used for urban
areas (6000 Ibs/acre/year) is evidence of that assessment.
In this area of maintenance controls, the need for per-
formance standards was determined to be the most practical
approach, given the preliminary characteristics of the
technology and problem assessment. At the same time, the
performance goals for the years 1985 and 2000 were reason-
ably set at K2 rates of .50 for suburban areas and .40 for
urban areas. It is assumed that these goals are achievable
by instituting selected public cleanliness programs (i.e.,
street sweeping and litter collection) in conjunction with
stricter controls on road building and street maintenance.
The technology to reach these goals is currently available.
When more advanced technology becomes available and the
problems are better understood, the performance standards
can be raised.
156
-------
There is one additional qualification to be made regarding
year 2000 K2 rates for existing land uses. Many of the
controls on new land development have a carry-over effect as
new construction sites become existing development sites
over time. Some physical controls, such as detention
basins, ditches and swales, and grade stabilization struc-
tures as examples, are maintained as controls beyond the
construction phase. Moreover, the effect of implementing
physical controls during construction may have a lasting
effect on sediment generation from these developments over
time which currently has not been well documented. Taking
this into consideration, the K2 rates of .40, .50, .50 for
urban, suburban commercial, and suburban residential,
respectively, (see Table 35) could be increased, assuming
the carryover effect of construction controls in conjunction
with maintenance controls.
5. Impact on Stream Loading
The improvement in performance standards for the land
management controls in conjunction with changing land use
patterns between 1968 and 2000 resulted in a significant
water quality response. Table 36 illustrates the stream
loading at three points representing the summation of loads
from the subarea. Subarea M includes the subareas M1-M5.
TABLE 36
STREAM LOAD BY SUBAREA
(Ibs/acre/yr)
Subarea 1968 1985 2000 % Change 1968-2000
M 469 376 304 35.2
N 564 477 387 31.3
0 819 705 571 30.3
In order to focus on the water quality response associated
with a single subarea, sediment loads (Ibs/acre/yr) for each
land use category in Subarea M3 are presented in Table 37 as
an example. As indicated, the greatest decrease in sediment
157
-------
load was associated with new land development where the K2
rate changed from .22 in 1968 to .05 in 2000.
TABLE 37
SEDIMENT LOAD - SUBAREA M3
(Ibs/acre/year)
1968 1985 2000 % Change 1968-2000
Urban 1003 850 680 32.2
Suburban Comm. 627 548 435 30.6
Suburban Res. 206 165 121 41.2
Undeveloped 101 101 101 0.0
Agriculture/Crops 182 182 182 0.0
New Development 1771 1127 402 77.3
No change in sediment load (Ib/acre/year) was achieved for
undeveloped and agricultural lands where the K2 rates re-
mained constant during this time period. For urban, sub-
urban commercial and suburban residential land use cate-
gories, the percent change in sediment load between 1968 and
2000 varied from 32.2% to 41.2%. This change is based on
achieving performance standards for maintenance controls on
existing development.
Complete print-outs of the 1985 and year 2000 land use
computer runs are contained in Appendix E.
158
-------
V
EVALUATION OF METHODOLOGY
The methodology presented in this report provides a simpli-
fied approach to the analysis of land use/water quality
relationships. The following sections present advantages
and disadvantages of the methodology for this type of analy-
sis and discusses problems encountered in the application of
the methodology.
A. ADVANTAGES AND DISADVANTAGES OF METHODOLOGY
1.
Data Requirements
An advantage of the methodology is that it uses data normal-
ly available from federal, state and local sources. These
sources include, among others:
Data
Precipitation
Soils Surveys
Stream Flow
Water Quality
Land Use
Source
National Oceanic and Atmos-
pheric Administration, U.S.
Department of Commerce
Soil Conservation Service,
U.S. Department of Agriculture
United States Geological Survey
U.S. Environmental Protection
Agency - STORET
County and Local Planning
Agencies
The methodology also requires minimal collection and analysis
of hydrologic and storm data, in contrast to dynamic storm
modeling which requires more extensive data of this type.
Overall, there are no significant disadvantages since most
existing methodologies require similar or more data.
159
-------
2. Ease of Implementation
The methodology uses a simplistic mass balance model based
upon the universal soil loss equation. This equation has
been found to have broad applicability and is suited to
modification. The mass balance model can be calibrated and
run without difficulty. Furthermore, parameter inputs to
the mass balance model are easily derived from the data
base. Although the model requires significant quantities of
input data, it is much more manageable than that required
for dynamic storm modeling.
3. Capability
The following are major capabilities of the methodology:
Provides an assessment of the relative magnitude
of nonpoint pollutant loads from varying land uses
and subareas
Permits the correlation and translation of visu-
ally observed problems (through field surveys)
into quantitative assessments
Provides an analysis of the sensitivity of varying
land uses to in-stream loads to identify which
land uses are major pollutant contributors
The impacts of land uses and land management con-
trols can be independently analyzed.
Allows for exploration of alternative control
programs for cost-effective analyses
The methodology cannot perform the following functions:
~ It does not provide the detailed predictions of
hydrological modifications produced by dynamic
modeling.
9 Water quality impact is not a direct output of the
model; supplemental assessment using model output
loads is required.
160
-------
The measurement of runoff pollutant loads is less
accurate for soluble pollutants than for insoluble
(sediment-borne) pollutants.
Loads resulting from slow snow melt and long dura-
tion, low intensity rainfall are not adequately
assessed.
The universal soil loss equation is not well
suited to measuring pollutant generation from
impervious surfaces. A high sediment delivery
ratio was assigned to impervious surfaces to
approximate urban runoff conditions.
B. PROBLEMS IN APPLICATION OF METHODOLOGY
]. Rainfall Data
While the primary source for rainfall data was the National
Oceanic and Atmospheric Administration, no such data were
collected specifically for the Wissahickon Watershed. As a
result, precipitation data for neighboring land areas were
used. These data collection problems did not significantly
alter results of the analysis since good correlation existed
between rainfall data and stream flows.
2. Municipal and Hydrologic Coincidence
Resolution of differences between municipal and hydrologic
boundaries was necessary in the selection of subareas. It
was important to develop subareas based largely on hydro-
logic characteristics in order to permit the mass balance
model to adequately reflect pollution conditions. However,
once pollutant generation can be related to specific sub-
areas, the step of designing land management controls is
based largely on municipal programs and consequently varies
by municipal boundary. In the process of moving from a sub-
area base to a municipal base, there is a problem of data
translation. A method for addressing this problem is pro-
vided in the case study (see p. 133).
161
-------
3- Man-Made Impacts on Natural Drainage
Where separate storm sewer systems exist in portions of a
study area, much of the urban runoff is channeled through
conveyance systems to outfalls and into streams. This
changes what would otherwise be a nonpoint source generator
into a point source generator. For this methodology, such
man-made drainage systems are not adequately accounted for
in the USCE-modified equation. In the case study of the
Wissahickon Watershed, such systems were predominant in the
City of Philadelphia. To account for this condition the
P factor (buffer effect) and SD factor (sediment delivery)
were evaluated to determine how closely actual conditions
are being modeled.
4. Loading Rates
Resources did not permit field monitoring and sampling to
determine study area-specific loading rates for urban mix,
surburban commercial, suburban residential, agriculture, and
undeveloped land. As a result, reliance was placed upon
loading rates found in the available literature. While it
was difficult to find complete documentation of loading rates
for site description and storm conditions, the rates chosen
were based on environmental conditions similar to that of
the Wissahickon Watershed, such as:
Intensity of land use
Topographical features
Climatic conditions
Soil characteristics
5. Pollutant Ratios
In this methodology, pollutant loadings other than sediment
are expressed as a ratio of the sediment load, since all
are sediment-borne pollutants. Study area-specific data on
pollutant ratios, however, were not available and were be-
yond the scope of this effort. Consequently, ratios were
derived from areas in other studies exhibiting common land
use and climatic characteristics. This is a problem which
can be overcome by collecting data fitting local conditions
through field monitoring, sampling and analysis.
162
-------
6. Control Effectiveness
As in the case of loading rates, resources did not permit an
examination of control effectiveness in the Wissahickon
Watershed. Again, information from available literature was
used. This information does not demonstrate much agreement
in the effectiveness of physical controls, and less so for
regulatory controls. Consequently, in this study, more
emphasis was placed on the relative differences in control
effectiveness, as opposed to absolute differences. BMP
control effectiveness criteria are constantly being expanded
and examined. Therefore, the state-of-the-art is increasing
rapidly and this technology gap is being closed.
7. Implementation of Controls
In assessing the implementation of controls, this method-
ology focuses primarily on regulatory controls. Implemen-
tation affects both the municipality which enforces its land
management regulations and the developers who must place
controls on the land. In determining the probable effective-
ness of land management controls for the model, the implemen-
tation factor must be accounted for in order to avoid pro-
jecting higher levels of effectiveness than can be realisti-
cally expected.
163
-------
GLOSSARY
Abbreviation and Definitions of Key Variables
K factor: raw loading rate (Ibs/acre/yr)
fraction of load retarded due to buffer strip
fraction of load received by buffer strip
KA factor:
KB factor:
KAG factor:
KN ratio:
KO factor:
K2 factor:
P factor:
R factor:
LS factor:
SD Factor:
reduction factor for management practices
applied to agricultural (crop) reduction
other (n) types of pollutant loadings, de-
rived as a ratio of the sediment load
fraction of load reduced due to general land
maintenance practices
reduction of sediment load to stream from
land management control practices
reduction of sediment delivered to the stream
as a function of natural controls such as
buffer strips and marshlands
(Rainfall factor) expresses the intensity and
quantity of rainfall as a ratio of the rain-
fall conditions being considered to those
encountered during the calibration conditions
(Slope Length) A dimensionless ratio expres-
sing the effects of sloped terrain and its
run length on the erosion and delivery of
sediment to streams
(Sediment Delivery Ratio) A dimensionless
ratio expressing the function of sediment
reaching streams. Factors consider proximity
to receiving streams and soil runoff poten-
tial
164
-------
Best Fit
The best agreement obtainable between model-generated non-
point loads and measured stream loads for either calibration
or verification conditions.
Calibration
The process of comparing and resolving differences between
model-generated nonpoint pollution loads and measured stream
loads. The model is simulating the same environmental and
runoff conditions existing at the time of measured stream
loads.
CFS - Day
The volume of water represented by a flow of 1 cubic foot per
second for 24 hours (expresses total accumulated flow at a
stream gauging station).
Curb-Mile
The distance of one mile of single curb length, or the
length of one side of a street or other thoroughfare.
Drainage Density
The miles of drainage (streams, creeks and other tributaries)
into which runoff flows, divided by the area of drainage.
Eutrophication
The aging process by which a lake or other water body evolves
into a bog or marsh and eventually assumes a terrestrial
state. Eutrophication may be accelerated by the impact of
human activities.
Hydrograph
A graph of the quantity of water flow at a specific stream
location as a function of time.
Land Management Controls
General term for category of controls on the use and treat-
ment of land, including municipal regulatory controls,
maintenance controls and physical controls.
165
-------
Land Management Program
The combination of regulatory, maintenance and physical
controls required by a municipality or other political
jurisdiction for the management of land undergoing construc-
tion or alteration.
Loading Rate
The rate of discharge of a pollutant from its source to the
stream.
Net Loading Rate
The rate of discharge of a pollutant to a stream which has
been reduced by land management and natural controls.
Raw Loading Rate
The rate of discharge of a pollutant to a stream, excluding
the effects of on-site land management and natural controls
(expressed in this report as Ibs/acre/year).
Maintenance Contruction
The renovation or alteration of a site or building to main-
tain or preserve its use. Buildings, bridges, roads and
sidewalks are subject to maintenance construction.
Maintenance Controls
Controls required where sediment generation results from
the renovation or alteration of existing developed sites, as
well as from common daily accumulation of debris, litter and
dirt.
Mass Balance
The measure and mathematical balance of accumulated pollu-
tant loads washed off from land mass as compared to those
loads measured at summation points. Considers the principle
of conservation of mass.
Municipal Regulatory Control
Regulation, ordinance or local law covering land use and
management requirements.
166
-------
New Land Development
Land formerly in agricultural or undeveloped state undergoing
construction for residential, commercial or industrial use.
Performance Standards
Specific, measurable levels of on-site control effectiveness,
required as objectives of municipal regulatory controls. The
means for achieving the required level of effectivness is not
specified.
Physical Control
A structural or non-structural measure, which in its inter-
face with the land, serves to alter or control natural
processes such as erosion and sedimentation.
Pollutograph
A graph of accumulated pollutant load at a specific stream
location as a function of time over the runoff period.
Sensitivity Analysis
A test of the significance of variables of the USLE - modi-
fied equation, in terms of effect on pollutant load.
Subarea
A part of a watershed, composed of a number of municipalities,
used as a base-area for land use-water quality data collec-
tion and analysis.
Summation Point
A stream sampling station at a point on a major drainageway
at which cumulative pollutant loadings and other factors are
measured.
Verification
The process of comparing and resolving differences between
model-generated nonpoint pollution loads and stream measured
loads for storm and runoff conditions different than those
expereienced for calibration conditions.
167
-------
BIBLIOGRAPHY
1. American Public Works Association. "Nationwide Charac-
terization, Impacts and Critical Evaluations of
Stormwater Discharges, Nonsewered Urban Runoff and
Combined Sewered Overflows," Monthly Progress
Report to the U.S. EPA, August 1974.,
2. Betz Environmental Engineers. "Wissahickon Watershed
Wastewater Management Study," for Montgomery
County, Pennsylvania, 1976.
3. Coughlin, R. E. and Hammer, T. R. "Environmental Study
of the Wissahickon Watershed within the City of
Philadelphia," Regional Science Research Insti-
tute, June 1973.
4. County of Fairfax, Virginia. "Erosion-Siltation Con-
trol Handbook," 1972.
5. Delaware Valley Regional Planning Commission. City of
Philadelphia aerial photographs, 1975.
6. Delaware Valley Regional Planning Commission. "COWAMP/
208 Storm Drainage Facilities Map," 1973.
7. Dow Chemical Corporation. "An Economic Analysis of
Erosion and Sediment Control Methods for Water-
sheds Undergoing Urbanization," for U.S. Depart-
ment of the Interior, 1972.
8. Engineering-Science Inc. "Comparative Costs of Erosion
and Sediment Control, Construction Activities,"
for U.S. EPA (EPA 68-01-0755), 1973.
9. Gabriel, Claude. Wissahickon Water Quality Model
(Unpublished report).
10. Grizzard, T. J. et al. "Assessing Runoff Pollution
Loadings for 208 Planning Programs," A Study for
the Northern Virginia Planning District Commis-
sion, 1977.
168
-------
11. Hammer, Thomas R. "Planning Methodologies for Analysis
of Land Use/Water Quality Relationships," A Study
by Betz Environmental Engineers for the U.S. EPA
(EPA 68-01-3551), 1976.
12. Harms, Leland L. e_t al. "Physical and Chemical Quality
of Agricultural Land Runoff," Journal of the Water
Pollution Control Federation, Vol. 46, No. 11,
November, 1974.
13. Heaney, J. P. et. al. "Stormwater Management Model,
Level 1, Preliminary Screening Procedures," for
the U.S. EPA, 1976.
14. Huber, Wayne C. et al. "Storm Water Management Model
User's Manual - Version II," Prepared for U.S.
EPA, 1975.
15. Hydroscience, Inc. "Simplified Mathematical Modeling
of Water Quality," prepared for U.S. EPA, 1971.
16. Hydroscience, Inc. "Storm Water Management Model,"
(in progress).
17. Lager, J. A. and Smith, W. G. "Urban Stormwater Man-
agement Technology: An Assessment," Metcalf and
Eddy, Inc. for U.S. EPA (EPA 670/2-74-040), Cin-
cinnati, Ohio, 1974.
18. Loehr, Raymond C. "Characteristics and Comparative
Magnitude of Non-Point Sources," Journal of the
Water Pollution Control Federation, Vol. 46, No. 8,
August 1974.
19. McElroy, A. D., Chui, S. Y., Nelogen, J. W., Aleti, A.,
and Bennett, F. W. "Loading Functions for Assess-
ment of Water Pollution from Nonpoint Sources,"
for U.S. EPA (EPA-600/2-76-151), Washington, D.C.,
1976.
20. Montgomery County Planning Commission. "Montgomery
County Land Use Map," 1972.
21. Philadelphia City Planning Commission. "Wissahickon
Watershed -Development Guide," 1976.
169
-------
22. Real Estate Data, Inc. "Real Estate Atlas of Mont-
gomery County, Pa.," 1976.
23. Roesner, L. A. et al. "A Model for Evaluating Runoff
Quality in Metropolitan Master Planning," ASCE
Urban Water Resources Technical Memorandum No. 23,
1974.
24. Sartor, James D. and Boyd, Gail B. "Water Pollution
Aspects of Street Surface Contaminants," A study
by the URS Research Company for the U.S. EPA (EPA-
R2-72-081), 1972.
25. Soil Conservation Society of America. "Soil Erosion:
Prediction and Control," 1977.
26. U.S. Army Corps of Engineers, Hydrologic Engineering
Center. "Storage, Treatment, Overflow, Runoff
Model (Storm Computer Program)," 1976.
27. U.S. Department of Agriculture, Soil Conservation
Service. "Montgomery County Soil Survey," 1967.
28. U.S. Department of Agriculture, Soil Conservation
Service. "National Engineering Handbook," 1964.
29. U.S. Department of Agriculture, Soil Conservation
Service. "Philadelphia County Soil Survey," 1975.
30. U.S. Department of Commerce, National Oceanic and
Atmospheric Administation. "Precipitation Rec-
ords -Pennsylvania Annual Summary," 1963, 1964,
1967 and 1968.
31. U.S. Department of the Interior, Federal Water Quality
Administration. "Storm Water Pollution from Urban
Land Activity," Water Pollution Control Research
Series, 11034 FKL, July 1970.
32. U.S. Department of the Interior, Geological survey.
"Water Resources Data for Pennsylvania: Parts 1
and 2," 1964 and 1968.
33. U.S. Environmental Protection Agency, "Areawide Assess-
ment Procedures Manual," Volume I (EPA 600/9-76-
014), Cincinnati, Ohio, 1976.
170
-------
34. U.S. Environmental Protection Agency. "Land Use In-
formation for Water Quality Management Planning,"
Technical Guidance Memorandum No. 13, 1976.
35. U.S. Environmental Protection Agency. "Methods for
Identifying and Evaluating the Nature and Extent
of Nonpoint Pollution," (EPA-430/9-73-007), 1973.
36. U.S. Environmental Protection Agency. "Nonpoint Source
Control Guidance, Construction Activities," Tech-
nical Guidance Memorandum No. 27, 1976.
37. U.S. Environmental Protection Agency. "Storm Water
Management Model: Level 1 - Comparative Evalua-
tion of Storage, Treatment and Other Management
Practices," (EPA-600/2-77-083), 1977.
38. Wanielista, Martin P. "Nonpoint Source Effects," State
of Florida Department of Environmental Regulation,
Technical Series, Vol. 2, No. 3, June 1976.
39. Weibel, S. R., Anderson, R. J., Woodward, R. L. "Urban
Land Runoff as a Factor in Stream Pollution,"
Journal of the Water Pollution Control Federation,
Vol. 36, No. 7, July 1964.
*U.S GOVERNMENT PRINTING OFFICE 1978 0-7?0-33V61?5
171
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA 440/3-77-025
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Planning Methodologies for Analysis
of Land Use/Water Quality Relationships : case
Study Application
5. REPORT DATE
November 1977
6. PERFORMING ORGANIZATION CODE
WH-554
. AUTHOR(S)
Betz Environmental Engineers
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
One Plymouth Meeting Hall
Plymouth Meeting, PA 19462
(215) 825-3800
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. EPA
Water Planning Division, WH-554
401 M Street S.W.
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Case Study
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The objective of this report is to present a detailed methodology and a case
study analysis of how runoff from urban and urbanizing areas affect water quality
The study outlines existing and projected land uses in the relatively small
Wissahickon Basin (Pennsylvania) and its related water quality problems. The
methodology employs a simple mass balance determination of sediment and'absorbed
materials based on a modified universal soil loss calculation. It is adopted
to field study data and other local conditions, such as climate, slope;
and patterns of residential and commercial development.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Land Use
Land Use Management
Runoff
Runoff Coefficients
Water Quality
Sediment
Universal Soil Loss Equation Projections
18. DISTRIBUTION STATEMENT
Limited to Professional Uses
19. SECURITY CLASS {ThisReport)
None
21. NO. OF PAGES
172
20. SECURITY CLASS (Thispage)
None
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue,'date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Insert appropriate code.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented'at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI HELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. 8.20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (Rev. 4-77) (Reverse)
------- |