905R79101
00623-
7911 INTERNATIONAL JOINT COMMISSION
MENOMONEE RIVER
PILOT WATERSHED STUDY
DRAFT
FINAL REPORT
VOLUME 5
SIMULATION OF POLLUTANT LOADINGS
AND RUNOFF QUALITY
COOPERATING AGENCIES
WISCONSIN DEPARTMENT OF
NATURAL RESOURCES
JOHN G, KONRAD
UNIVERSITY OF WISCONSIN SYSTEM
WATER RESOURCES CENTER
GORDON CHESTERS
.•j
SOUTHEASTERN WISCONSIN REGIONAL
PLANNING COMMISSION
KURT W. BAUER
Sponsored by
INTERNATIONAL JOINT COMMISSION
POLLUTION FROM LAND USE
ACTIVITIES REFERENCE GROUP
UNITED STATES ENVIRONMENTAL
PROTECTION AGENCY
November 1979
-------
-------
SIMULATION OF POLLUTANT LOADINGS
AND RUNOFF QUALITY
by
V, NOVOTNY
DEPARTMENT OF CIVIL ENGINEERING
MARQUETTE UNIVERSITY. MILWAUKEE, WISCONSIN
D, BALSIGER*
WISCONSIN DEPARTMENT OF NATURAL RESOURCES
D ,S, CHERKAUER
DEPARTMENT OF GEOLOGY
UNIVERSITY OF WISCONSIN-MILWAUKEE
G, V, SIMSIMAN
G, CHESTERS
WISCONSIN WATER RESOURCES CENTER
R, BANNERMAN
J, G, KONRAD
WISCONSIN DEPARTMENT OF NATURAL RESOURCES
Grant Number: R005142
THIS STUDY WAS CONDUCTED IN COOPERATION WITH:
WISCONSIN DEPARTMENT OF NATURAL RESOURCES
UNIVERSITY OF WISCONSIN SYSTEM WATER RESOURCES CENTER
SOUTHEASTERN WISCONSIN REGIONAL PLANNING COMMISSION
CURRENTLY IN THE DEPARTMENT OF STATISTICS, UNIVERSITY
OF WISCONSIN-MADISON
U.S. Environmental Protection AgenCf *
GLNPO Libwy Col'-^ion (PL-12J) A
77 West Jackson B-jJr-.'ard,
i Chicago, IL 60604*3590
-------
DISCLAIMER
This report has been reviewed by the Great Lakes National Program Office
of the U.S. Environmental Protection Agency, Region V Chicago, and approved
for publication. J'fention of trade names of commercial products does not
constitute endorsement or recommendation for use.
ii
-------
PREFACE
Prediction of pollutant loadings from non-point sources is an important
aspect of water quality management. A well-calibrated mathematical model
verified with extensive monitoring data may be applied to other watersheds for
predictive purposes. This volume contains two reports on the application of
the LANDRUN model and a discussion of a simple, empirical model for predicting
runoff quality. The LANDRUN model is utilized to 1. assess sediment loadings
from 48 subwatersheds in the Menomonee River Watershed in an attempt to
identify critical areas that are most cost-effective in terms of pollution
control and 2. obtain unit pollutant loadings for typical land uses to better
understand the processes involved in pollution generation and transport from
urban and non-urban areas.
iii
-------
CONTENTS
Title Page . i
Disclaimer . i i
Preface . ill
Contents iv
*Part I Assessing Pollutant Loadings fron Subwatersheds with
Mixed Land Uses I-i
*Part II Model Enhanced Unit Loading (MEUL) - A Method of
Assessing Pollutant Loadings from a Single Land Use ...... Il-i
*Part III A Simple, Empirical Model for Predicting Runoff Quality
from Small Watersheds ».,....„ IIT-i
*Detailed contents are presented at the beginning of each part,
iv
-------
PART I
ASSESSING POLLUTANT LOADINGS FROM
SUBWATERSHEDS WITH MIXED LAND USES
by
D, BALSIGER
R, BANNERMAN
G, V, SIMSIMAN
J, G, KONRAD
G, CHESTERS
I-i
-------
Stm.nl af> --.ns of --e.-Mirioat loadir,;.;-' for M.;,.C~ \~>\if, \a"d iises : fi 48
subwaf ersheds <>f to/;- M^notf.cnee River Watershed are per toiled using the. lANDRUN
model. In or-v-r ;;••', i.-,1:ermine critical soui'^e areas, simul.ited loadings are
adjusted b^sed on deliver}? ratios estimated for pervJ ous areas in each
sabwater,;<:•.••:•[]« ]-I1r,^ ^ubwatersheds, con.sisting of 35% of the total area of the
Watershed, are idc'i--.; f led as crilicai source are^w with developing lands being
the primary ront r ibutors of sedliDer.Ls, The critical! ty of a subwatershed in
terms of r:\.-npOiot source pollution appears to be enhanced by the extent of
connected imperviousness and proximity co the stream of that area.
-------
CONTENTS - PART I
Title Page I-l
Abstract I-ii
Contents . .. I-ii i
Figures I-iv
Tables . I-v
I-l. Introduction I-l
1-2. Conclusions 1-2
1-3 Methodology 1-3
Source and Form of Data for LANDRUN Simulation 1-3
Manipulation of Land DMS Data Prior to Calibration 1-3
Calibration, Verification and Determination of Degree
of Connected Iniperviousness 1-6
Simulations for 48 Subwatersheds and Determination of
Sediment Delivery Ratios 1-7
1-4. Results and Discussion 1-10
References 1-15
Appendix
1-A. Simulated Loadings for 48 Subwatersheds 1-16
I-iii
-------
FIGURE f,
Fumber Page
'r-l The 48 suhwatersheds in the Menomcmee River Wafershed ... 1-4
l-'l :~*>iul-;ited (S) and monitored (M) sediment Loadings
(kg/ha) from area adjacent: to ma Lnsto-m monitoring
-j L 'j t i o~n s—suTtiTi)ors 1977 •.•••»><»»u >•••••»»*» «•«.« •••.»>••.• I~~ll
1-3 ^isi rJbiiti'-.n of simulated t3fdiment loadings; in the
Mer,:>mon<--.e River Watershed —rummer, 1977 »..,«.„...».,.»,... 1-12
I - i v
-------
TABLES
Number Page
1-1 Land use categories (1975) in the 48 subwatersheds of
the Menomonee River Watershed 1-5
1-2 Estimated sediment delivery ratios for various land
uses (LU) in the 48 subwatersheds of the Menomonee
River Watershed 1-9
1-3 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in the I-fenomonee River
Watershed—summer, 1977 ., .... 1-14
I-A-1 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 12A—summer,
1977 1-16
I-A-2 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 12B—summer,
1977 , 1-16
I-A-3 Water (m ) and sesiment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 12C--summer,
1977 1-17
I-A-4 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 12D—summer
1977 1-17
I-A-5 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 12E—summer
1977 1-18
I-A-6 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed IDA—summer
1977 1-18
I-A-7 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 10B—summer
1977 1-19
I-A-8 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed IOC—summer
1977 1-19
I-v
-------
i-A-9 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 1OD—sum-net
1977 ,....,.... 1-20
I-A-IP Water (m3) and sediment (kO loadings estimated by
LAN DP !l? I for each land use in Subvaterrbed 1 i~>E—Mirror
1977 .....................» 1-20
I-\-!l '".'Hf-"!' (r-,"1) and sediment (Vs) .V.a.iiaps estimated by
L'.MD'RUN for each land rse in Sub-rat ersher) 7A-".sunmor
1977 ,.,., 1-2].
I-A-1? Vatnr (Ti3) and sediment uu;) 1 C'-iJitii-,? e-^ti^ated Hy
LAI- DP r/n for each land use in >I,'!-UH Lerr-die;1 7B-~stinnner
i-:V-l^ V'at-:r i^n ) and sod intent (r'-;^) "" oad niS'S e«:f: i .nateci 'h\
iANnpLN for each land asp in Snb.oter-.1\ ,1 1C—Pun.r^r
1977 . . ............................
[-A--15
1-23
;';ntrr (PI") and sedinu-i'.C (kc> ; io'idin^s psrirnatel hy
LAKFVRi'N for each land use in Sur\Mtershed 7c--^,tnni:r
LQ77 ... ..... ....... ..... .................................. 1-2A
I_4i._ic) Wate-r (mj) and sedinent (kp) loadings cstiiontod by
T.AICP'R^N for each land use in Sub-water^ lied 7H — iamTn(j r
1977 . . . , ..... ................... ....... ... .............. 1-24
1-A-J9 P-iter (m3) and sediment (kg) lojdirgs estinated by
LANDRUN for each land use in Subwatershed MA--sumncr
1977 .,,,,. ____ .. ..... ............. ---- .................. 1~25
I-A-2C Wattr (n3) and sediment. (!'.>T) loadings esrinv-ited by
LANDRUN for each land use in Subwat~ershed 11B— suminer
1977 ......... ..... ...... ................... ............ 1-25
Water (THJ) and sediment O's) loadings est i"ia!'0.d by
T AKDRl'N for each land urse in Suv,^ater;.'bed 1 1C— nunr'fr
1977 . ... r, ................................... .......... , . I-2b
-------
I-A-22 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 8A—summer
1977 1-26
I-A-23 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 8B—summer
1977 , 1-27
I--A-24 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 8C—summer
1977 ., 1-27
I-A-25 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 9—summer
1977 I-2B
I-A-26 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 6A—summer
1977 1-28
I-A-27 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 6B—summer
1977 , 1-29
I-A-28 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 6C—summer
1977 1-29
I-A-29 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 6D—summer
1977 1-30
I-A-30 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 6E—summer
1977 1-30
I-A-31 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 6F—summer
1977 1-31
I-A-32 Water (m3) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 4A—summer
1977 1-31
Q
I-A-33 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 4B—summer
1977 1-32
I-A-34 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 4C—summer
1977 1-32
I-vii
-------
1-A-3S •'•>[-.•.'- (T, '} aa.d s^ii;. :-nt (;:>; U:,.'in;-s est i'\>t,>d S'
< A,NT[V!!L" f, r ear1"1 land iise in "uhwaterFru-d ',}'-—'-.u^rier
1.0 77 ,..,..........,«...,,.=..„............,.., ...... 1-35
'Jjfor ( -!J) and sediment (kf>) U'Winpj est; rated b-,'
LA^'DKi'M for each land use in Suh.^at er^ir-d \V~-MI miner
1977 ..,......,........„,.„,,. L ..„„........, .. I-?'?
Wat<-r (fp") no.d F e'limenc (.'•.:') LOT.'' i. "'p,s estimated hy
LA"DPi(is,' f.Tr ^acli land >i-3c \r. Subv/^lershed 3C —-suMrrc.i~
.LV77 ..,..,.,,,..........,.....«..........,.,,....„...... 1-34
I--A-3'T 1,'ater (m1) and c.ed:iment ('r,^) loaclinps estimated by
LAND!!!!?! for e^cb land H-M- In Subv.;acershc-d 3D~=?unmcjL'
1977 .......>.........,.. ...,.......,..........>.>...... 1-35
I--A-40 S-'ater (m4) and se^rllmerit (i-u) loadings estinate^ by
LAKDRUK for each land use in Subwatersbed 3E~sunir>L-r
1977 .........,..,.,..,,.„.,.,.. ....,,... 1-35
I-A.-41 water (p3) and sediment (k^) loadings cst1n:»tij<] b<«r
i.ANir'UN fr.r earh land ii,:c ?n r,-:1-. /ater^jhed 3F—mmruM:
J077 „„. ............................... 1-36
I-A-'i?. T,'-ir.t-;r (r'} and sedinent (U^) lot.-diiipf estinated by
LAND!'iT*T for each land uso in Subwatershed 3"—Dimmer
1977 ..................................... l-3f
I-A-43 vJatt'T (in") and sedirnent (kc) loadings estinated by
LANPRTN for each land u^e in Subwatersbed J1!7—sintmer
1977 .„„„.„...„...,,,,.. 1-37
l-A-4^ Water (n3) and qediment (i'^,) loadings estinated by
LANDPUN for each land use in Sub^atershed f)-—simmer
1977 ..............,,.,....,.,..,.. 1~37
1--A-45 Water (IP") and sediment (kg) loadinep estimated by
LANDFUr.' tor each land use in Subua !:e r^bed f—surmer
J 977 . .... ............... ... .. 1-38
I-A-46 ''-later (nr') and sediment (l-.g) loadings estii'.ared ^v
LAMDRU?" for earh land u?e in Subx-jatersbed 1 A—summer
1977 . 1-38
I-A-47 Water (m3) and sediment (!y,0 loadings estimated by
LAWDRUN ^"or each land uso i •-. Ji'ibwatersbed IB — Hitmrne!1
-------
l-A-48 Water (m ) and sediment (kg) loadings estimated by
LANDRUN for each land use in Subwatershed 19—summer
1977 1-39
I-ix
-------
1-1. INTRODUCTION
Identifying critical source areas of nonpoint pollution in a watershed is
imperative if economical means of remedial control measures are to be
adopted. Because monitoring of all potential source areas in relatively large
watersheds, like the Menotnonee River Watershed (35,000 ha), incurs extremely
large expense and time, a model capable of predicting pollutant loadt, from
smaller components of the total watershed is very useful.
LANDRUN, a dynamic runoff-sediment overland transport model, after
initial calibration and verification, has demonstrated its capability of
simulating field data for such parameters as runoff, sediment and adsorbed
phosphorus (1). One application of LANDRUN is the prediction of pollutant
loadings from subwatersheds of diverse land uses and physical
characteristics. An attempt was made to use LANDRUN in simulating runoff and
sediment loadings from , 48 subwatersheds in the Menomonee Raver Watershed.
Such application of the model is described in this report and results obtained
should aid in demonstrating what land features, land uses or land activities
contribute to high pollutant loadings. Water and sediment loadings were
simulated during the summer of 1977„
1-1
-------
"i 6 L/-;'.rnRn\" r.odcl wac: capable of s ; PHI! •" : ing ",Tt-Ji and '":>di,"ieat 1 oad r >1"r
tor various tan^ sy>^s in 48 3'ahwaLar'-1v.-'df-;» Deliv?1',"',• r/u; Lo for each land usa
w i,-j nf-cf_fir.-:\ v to id in<;t sedine^t loadings iVo,,i pervious aroay. '~ i niil-i te^i
sc^'iriont londin^s i/er^ round to compare re^scnahJ y v/sll -nth rionl torod dnt.i
fr'jf'i I;be n/iinutetii ^T at inns.
I.'ir;p i.,r It n.'..''1 •""•point sc'irc iiL\T' I ,-••;"'"•. IIP ;!c., ; nns t i t ui in'; 16! of the
total aroa if the rr-it.' fp'.ied, wero ' cu- ~. ^ I i':' ed and or^nt-'i "^'n •' alsiint 50," of the
fcital r,erl f'iK-nr. lo;t'ii !'>,'-» OPVP lopi T; r ;irea;: •• >-:^ the pi-i'-ar}' contributor of
sediments. AT though de'/elcpit:^ iards occ-'py a «na J L j>ort ion of t. lie
•5'i'iufi rors'ned ( L tn ">/'.)> they font r i .in c-?d In'rth aia-_>rnts (30 to ^rj'0 of c> c d i rie n t
loadings. '1 !:•" r "i t i cal i ty of a jiu-arr-? .IT •••a can ba orlian: r.'d b)r tio extent '.if
ronnectod iriiV1'.'vIousMPTS aii'i p1" ^-v '•'' ^'' t-> "-'i- r-rLffar1 o'; th.ii- r-nh'.. j i ar'-aed. Tt
appear,*- iliar .ic'/alopi r,'j aroa.s in ur n-vii •>,: n-> -rahwa torsheds ar;-- fh'- -<"sr e<5t™
ff t-fr t i\'e In ten*7 of" ^"nar.einor.t.
-------
1-3. METHODOLOGY
Source and Form of Data for LANDRUN Simulation
LANDRUN is a mathematical model developed as a method of analysis for
estimating the quantity and quality of runoff and eroded particulates
emanating from watersheds having mixed land uses. The description of this
model and the discussion of its initial calibration and verification are given
in (1).
To perform LANDRUN simulations for the 48 subwatersheds in the Menoraonee
River Watershed (Fig. 1-1.) two types of data are needed, namely, 1. land use
and associated characteristics in each subwatershed and 2. meteorological
information! obtained within and near the Watershed. Data on land use,
soils, slope and degree of imperviousness on the 48 subwatersheds were
provided by the Land Data Management System (Land DMS) described in (2). The
79 land use descriptions were consolidated into 14 land use categories (Table
1-1). The consolidation grouped similar land uses and land uses that have
similar potential for non-point pollution (3). Data obtained from the Land
DMS were in the form of area of each slope category for each soil type found
for each of the 14 land uses in each subwatershed. The Land DMS also provided
the degree of imperviousness for each land use for each of the subwatersheds.
Meteorological data were obtained from two sources. Precipitation data,
in the form of hourly precipitation totals, were furnished by the U.S.
Geological Survey (USGS) from eight precipitation gauges located throughout
the Watershed. Maximum and minimum daily temperatures, as well as daily
evaporation values, were obtained from the National Weather Service at
Mitchell Field.
Dust and dirt data which include dust and dirt fallout, washout
coefficient and sweeping efficiency were obtained from the Chicago study on
pollution from urban areas (4). Information on sweeping frequency was
provided by the Engineering Office of the cities in the Watershed.
Manipulation of Land DMS Data Prior to Calibration
LANDRUN, like other similar overland flow models, is sensitive to the
degree of imperviousness connected directly to storm sewers and streams, and
for pervious areas, to soil permeability, interception and depression
storage. The model requires dividing the Watershed into uniform areas based
on land use and soil characteristics. A land use with two different soil
groups was considered as two sub-areas. For a single land use in a
subwatershed, the many soil types were grouped into hydrologic soil groups B,
C and D (soils under group A are insignificant in the Watershed). An area-
1-3
-------
'—>, ..'•
••"•V,
w
Fig. 1-1. The 48 subwatersheds in the Menoraonee River Watershed.
-------
7 C
12
1-5
-------
weighted mean slope was calculated for each land use-soil group sub-area
(e.g., the row crop land use in a suhwatershed was computed as Row Crop B, Uow
Crop C and Row Crop D; and having an associated area and mean slope).
Saturation permeability and other soLL characteristics could be inputted for
each of the 3 soil groups within a particular land tise.
Land DllS-land use data segregated .'ill streets, freeways and off-street
parking areas from other land uses Into a transportation land use. In order
Lo represent accurately the nature of urban land uses, it was necessary to
integrate these impervious ar^as back into the various land uses. Total area
and degree of imperviousness data werr. adjusted to account for this additional
area. Freeways were retained as a separate land use.
Calibration, Vei'if i rat, Jon and be-: te ruination of
Degree of Connected Inperviousness
Starting with values used in the initial raiibrati.;n and verification of
the model (5), individual events, sequences of events and eventually the
entire 1977 stunner season werf simulated for subwa ter sheds in uhLoh good
monitored data ware available for con;vrr i son.
The hydrology portion of the model was first calibrated on subwater sheds
3 and 9 (Schoonmaker Creek- 41 H 010 and floyes Cree r-41301 1) , each of which had
water quality data and flow it; format ion from a sampling site which nonitored
only that subwatershed. Both ••uhwatersheds are predominantly median density
residential although the Tic-yes Cre.-k area i ,s a nev/cr development. Additional
calibration was performed on the i -:!iL '/atorsheds (11A, 11B and. 11CJ whu'.h
cor.iorise tine area monitored by thi- Coages Bay Read station (463001) and the 4
subwater. sheds (4AS i"B , 4C and 4l.i) nonitored by the roney Creek vai-.nl i -IL- site
(4.i'i006). The Dop:;e.-3 Bay lload subwai ersheds are predominantly rural while the
Roney Creek subi-/atersheds are nostly residential,, but with significant
pervious areas on the southernmost subwatershed (4D). Simulation of these
urban, rural arid mixed land use areas and comparisons of simulated flows with
monitored flows led to the determination of connected j uperviousriess values
for the calibration snbwatersheds.. Calibration of f~he <=pr}inent portion of the
model was done on the Noyes and Schoonmaker Creeks subwatersheds as these
snail urban areas were expected to have a delivery r?t.io nuch closer to unity
than the larger suhwatersheds or rural areas.
Simulations on other subwatersheds for verification showed that the
degree of connected imperviousness could be described as a function of the
extent of storm severing in a subwatershed. The degree of directly connected
imperviousness is the single most important factor influencing simulated
runoff from urban areas. For this reason, it was necessary to obtain detailed
information from maps, conversation with city engineers, etc. concerning the
extent of storm sewering, the precise location of new residential
developments, the usage of grass ditches for drainage, etc. The result of
this exercise was a set of connected imporviousness values for the land uses
modified according to individual differences in each subwatershed. The area
of directly connected impervious surfaces was calculated for each land use in
each subwatershed in the following manner. The model was used to determine
percentages of directly connected iraper viousness for completely sewered and
1-6
-------
unsewered subwatersheds. Values for partially sewered subwatersheds were
derived by prorating on the basis of the land use which was in the sewered
area of that subwatershed. Examples of percentages of directly connected
imperviousness are shown below.
Land use
Industrial
Medium density
Completely
sewered
80
60
Partially
sewered 0>60%)
45
35
Unsewered
8
3
residential
Low density 20 5 1
residential
Parks/recreation 30 15 1
Simulations for 48 Subwatersheds and Determination
of Sediment Delivery Ratios
After calibration and verification was completed, simulations were run on
all subwatersheds. Simulated flow values from the individual subwatersheds
were summed accordingly and were found to compare favorably with measured
flows at the several mainstem river sampling stations. Simulated sediment
valxies corresponded reasonably well with loading estimates calculated from
monitored values for urban areas where a large part of the sediment originates
from impervious surfaces and the degree of connected imperviousness is high.
Calibration in these areas is accomplished by manipulation of the cropping
management factor for developing areas and doubling the literature values for
dust and dirt accumulation values. In more pervious areas and rural areas,
simulated sediment values were much higher than monitored loading estimates
(e.g., as much as 20 to 30 times higher in Donges Bay Road). Thus, there was
a need to develop a series of sediment delivery ratios for the land uses in
each subwatershed.
Proceeding as in the runoff calibration process, it was determined that
sediment delivery ratios were dependent on the extent of storm sewering
(connected imperviousness) in the subwatersheds. Other important factors are
proximity to runoff channels and characteristics of the land use (e.g., parks
vs. small grains, airport vs. shopping center). Again, it was necessary to
collect detailed information to characterize the land uses in each watershed.
Land uses were grouped into three categories and each category was
assigned a sediment delivery ratio for each subwatershed. "Urban" land uses
included industrial, commercial, medium and high density residential. "Rural"
land uses included agricultural areas, parks, low density residential and
landfills. Developing lands (construction), the third category, had such a
high sediment yield compared to the other land uses that it was assigned its
own delivery ratio. Resulting delivery ratios ranged from 1.0 for "urban"
land uses in completely storm sewered urban areas to 0.01 for developing lands
1-7
-------
in non-sewered areas. Table I~2 shows the sediment delivery ratios for the
subwatersheds for the 1977 summer simulations.
1-8
-------
Table 1-2. Estimated sediment delivery ratios for various land uses (LU) in the 48 subwatersl^ds
of the !fenomonee River Watershed '.• '* /, ,*-"'*-^
STORLT No.
673001
683002
683001
463001
413011
413008
413007
41 3006
413005
413010
413009
41J004
Monitoring station
Location
MR at River Lane Rd.
(Ilwy.F)
MR at Pilgrim Rd.
(llwy. YY)
MR at 124th St.
(Hwy. M)
Donges Bay Rd. , Mequon
Noyes Cre^k at 91st St.
Little MR at Apple ton Ave.
(Hwy. 175)
Underwood Cree1' above
Hwy. 45 off "forth Ave.
Honey Creek 140 n above
confluence with MR
MR at 70th St. Bridge
Schooninaker Creek at Vliet St.
MP at Hawley Rd.
MR above 27th St. at Falk Corp.
Adjacent
subwater shed
12A.121.
12B.12C.12D
10A
10B,10C,10D
10E
7A
7B,7D,7F,7G
7C
7E
711
HA.llB.llC
9
8A
8B
8C
f>A
6B
6C,6D,6F
6E
4A , 4B , 4C
4D
3A , 3B , 3C , 3E , 3F , 311
3D
3C
5
2
1 A, IB, 19
Deliverv ratios* L^
LU 1-5
0.34
0.03
0. 80
0.03
0.60
1.0
0.03
0.15
0.53
0.81
0.03
1.0
1.0
1.0
0.03
0.35
0.23
0.03
0. 0 3
1.0
1.0
1.0
0. 52
0.60
1.0
1* 0
1.0
LU 7
0.02
0.013
0.02
0.015
0.02
0.30
0.03
0.04
0.15
0.30
0.03
0. 70
0.40
0.10
0.07
0.10
0.04
0.015
0.01
0.70
0.50
0.70
0.30
0.70
0.70
0. 70
1.0
LU 6,8-13
0.03
0.03
0.03
0.03
0.03
0.06
0.03
0.03
0.03
0.06
0 . 0 3
0. 06
0.06
0.03
0.03
0.10
0.05
0.03
0.03
0.30 ,-
0. 30
0 . 3 0
0.10
0.15
0.30
0.30
1.0
*Foi pervious dreas only; delivery of dust and dirt from impervious areas is assumed to be 1.0,
1-9
-------
1-4. RESULTS AND DISCUSSION
Extensive monitoring at the ma ins tern o!. the Menomonee River reveals that
the more urbanised areas in the lower portion of the Watershed contributed
greater sediment loadings than the rural upper per Iion (Fig. 1-2). Mainsten
monitoring could show general areas of nonpoint cources of pollutants,
however, identification of critical areas is quite difficult because adjacent
areas monitored by the major stations are too large (3,000 to 7,000 ha).
Estimation of pollutant; loadings on smaller units should provide reasonable
precision for identifying critical source are^s '-'here best management
practices can be applied.
Water and sediment loadings simulated by LA.NDRITV during the summer of
1977 for the 4P> subwatersheds (200 to 1,600 ha) of the Menomonee River
Watershed are shown Ln Tables T-A-1 to I-A-48. Loadings are given for all
land uses identified in a particular sub-watershed. The sediment data were
adjusted accordingly taking into account delivery ratios (Table 1-2) for
pervious areas in the various land uses, Dust and dirt: accumulations on
impervious surfaces were assumed to have 100% delivery. Delivery ratio for a
land use was estimated based on its physical characteristics, extent of
connected imperviousness and proximity to the stream.
Simulated sediment loadings vere found to compare reasonably well with
those monitored at all but one of the ma instera stations (Fig. 1-2). At
station 673001, the simulated data was almost 3 times as high as the monitored
data. The extremely low sediment loading measured at this station could be
due to the trapping effect of a large pond just upstream of the station. The
close agreement between the simulated and monitored data indicates the
validity of the delivery ratios used for each land use and the integrity of
the sediment estimates for each subwatershed.
Results of simulations showed that nine subwatersheds (7H, 7A, 8A, 9, 3F,
3H, 3C, 4C and 4D) contributed significant amounts of sediments (Fig. 1-3).
These high source areas, located in the urbanised lower portion of the
Watershed, constitute 16% of the total area (calculated up to station 413005)
but contributed almost 50% of the total sediment loadings. The high sediment
yields from these subwatersheds can be ascribed mainly to developing areas
and—to a certain degree—to medium density residential areas. Developing
areas were present in almost all of the subwatersheds. However, high amounts
of sediments were transported from developing areas in the critical
subwatersheds essentially because of their short distances to the stream and
extensive connected imperviousness. Although high amounts of sediment can be
eroded in other subwatersheds particularly those in the rural portion of the
Watershed, delivery of sediment to the stream could be impeded as a result of
low connected imperviousness and/or greater distance to the stream. Medium
density residential areas, the predominant land use in the critical
subwatersheds, were significant sources of sediment loadings. Due to
extensive impervious surfaces in these areas, dust and dirt washoff was
prevalent.
1-10
-------
673001
Main-stem station
Menomonee River
and tributaries
Fig. 1-2. Simulated (S) and monitored (M) sediment loadings (kg/ha) from
area adjacent to mainstem monitoring stations—summer, 1977
(monitored data taken from (6)).
1-11
-------
kg/ha
0-150
150-350
^350
Menomonee River
and tributaries
Fig. 1-3. Distribution of simulated sediment loadings in the
Menomonee River Watershed—summer, 1977.
1-12
-------
It is evident from the critical subwatersheds (Tables I-A-11, I-A-18,
I-A-22, I-A-25, I-A-34, I-A-35, I-A-38, I-A-41 and I-A-43) that the majority
of the sediment loadings (50 to 85%) originated from small areas (1 to 5%)
that were under development. This also can be seen in Table 1-3, which is an
integration of the loadings from various land uses in the entire Watershed.
Over 50% of the total sediment loadings was contributed by developing areas
occupying just 3% of the total area of the Watershed.
It has been shown that the model is a useful tool in identifying critical
nonpoint source areas of sediment in the Menomonee River Watershed. Results
indicate that developing areas in urbanizing subwatersheds are tVie raost cost-
effective to manage. The method is applicable to other watersheds. However,
the difficulty of simulating sediment loadings on pervious areas requires some
recalibration and reverification of the model in other watersheds using
monitored data.
1-13
-------
1-14
-------
'
-------
APPENDIX I-A. SIMULATED LOADINGS FOR 48 SUBWATERSHEDS
iLV7.
11 3J
587.
.51
1026G.
S.I I
11711
9.25!
11611.
9 2!
2 1 o n
I .OS
51561.
9.9*
735.
.3%
16812
18.91
152.
.2%
1107.
2 7J
22.
1 ?
II
0
-'13.
1 71
1028.
8.1J
<90.
3 1J
0.
.01
8'126.
fb 1J
3.9J
3150.
5 3Z
31131.
57 6%
3023.
5.5?
11.
. IS
* p E A
PFRV
6.
1 .71
25
7 It
ARL A
IM D r R
33.
7.SS
71.
17.3S
LAND U S E
INDUSTRIAL
COMMERCIAL
MED/PENS/RFS
[0 /DENS/Pff
HI /UEHS/RFS
DEVELOPING
ROW CROPS
PK/Rt C/PAS1 R
FORESTS
WF.TLANLS
FFEDLOT5
WATER
F REtWAYS
WATER
PFRV
11311.
7 6J
15275.
10. 21
20223
1J.5J
1666
1 . 1 J
185S
1 25
1621 3
30.8%
2611
1 .78
29825
19.9!
?311
16110.
1 1 .07
2102.
1 .11
0.
OJ
0.
.05
WAI ER
IHPER
5189
18 55
1662.
16 6J
2512.
9 OS
103.
.1!
126 .
1 5J
96 j
3. 'U
0
.0$
1 1 2
1 55
r
OJ
0
.0!
0.
.0!
937".
33.3?
1)152.
15.85
WATLP
TOTAI
16503.
9 3»
19937
1 1 . 2 %
22765
12 81
1769.
1.0J
2282
1 .31
17176
26.55
2611
1 5J
30237.
17 rs%
2311
1 31
16110.
9.21
2 102.
1 .2?
9371.
5 3J
U152.
2.5*
o|h r,I(.p fjT
PFRV
6, .7.
1 .27,
275.
520 .
1 .OJ
If.
.11
31
11
37 M
71 15
5787.
1 1 U
5153.
9 9!
187.
15
652.
1 25
1688.
3 25
0.
0%
0
.0%
D'JST/I-IRT
I M P E R
520
19.51
166.
16 5J
256.
9.1J
1 u .
15
13
1 .51
97.
3.15
0.
.01
1 1 .
1 .5J
0.
.01
0
.OJ
0.
.OJ
939.
33. 3»
116
15 .85
SEDIMENT
TOTAL
1127
?.'!%
711 .
1 (5
776.
1 .U
56
. 15
71.
. 1J
37155
66.01
5737.
10.5!
5191
1 'U
187
• 3J
052
1 2!
1688
i. 1*
:%
8
9.85
fj
.OS
.OJ
2
2.22
2?,
.-6,6 1
ARFA
101 AL
52
1 . 35
33
3.2%
75
6 ^'J
19.
1 .61
3*
11 .
3.67,
Mi.
jO . 0 j
J 5 1 .
72
' • ™
-., .
.5.
2 .
. ^ a
<•
> ''
1 11839
28123.
177962.
1-16
-------
LAND USF
INDUSTRI Al
COMMERCIAL
MED/DEUE/RES
LO /DFNi/BEr-
DEVELOPING
ROW CROPS
P^/r'EC/?ASl R
FORESTS
WE'lLSNDS
rttPlOTS
WATER
TOTALS
7 > , L' I- *,-u.
LAND <>.,¥
COMMERCIAL
MCD/OEBS/RES
LO /DENS/RES
DLVFL OP ISO
ROW CROPS
PK/pFC/?«STR
TORFSTP
WE HANDS
FEEDLdTS
TOTALS
WA1FR
PtRV
2687.
2.95
7356.
7.9J
22955.
24. 5%
1324.
1 .41
11661 .
15.6?
865.
.9J
J5635
IS. It
2621).
2.8%
4C40 .
1.31
1 495 .
1 6J
0.
.0%
93692.
Summer !377
WATER
PERV
6396.
3.6J
10928.
13.7%
1965.
2.5%
5079.
6.47,
3395.
1.31
38135
17.7?
6381 .
3.05
5922.
7 41
1 168
1 .5?
79869.
WATER
IHPFR
852.
4.7%
1761
9.65
2742
15 0?
83.
.5%
489.
2.1%
0 .
.01
637.
3-5%
0.
.0%
0 .
.0%
0.
.0%
1 1719
64. 1J
18283.
WATER
IMPER
1771 .
47. 3 J
1395.
37.7%
102.
2.8%
135.
3.6%
0.
.01
300.
8.1%
0.
.05
0
.0?
0.
.0%
3703.
WATFR
TOTAL
3539
3.2%
9117.
8.1J
25697
??.9S
1407.
1 .35
15150
13 5%
865.
.8%
36322.
32 4',
2624 .
2.31
4040.
3.65
M9i.
1 .31
11719.
10. 51
111975.
(KF) i -"i a liny
•JATFR
TOTAL
8667.
10.12
12323
11.7%
2067 .
2.5S
5214
6 21
3395.
1.11
38435.
46. OS
6381
7.6Z
5922.
7.1%
1 168.
1 .41
83572.
SF01MFKT
PFRV
12.
.It
57
.5%
468.
4.0t
32.
3J
33%*
214P
13.5%
3007.
26. OJ
199
1 7%
1 13
16fb.
11 6J
0 .
0%
1 1CR2.
rr.iMMEIIT
PbHV
84.
.55
207.
1 3»
52.
.3J
921 .
6.0%
8730.
36 8%
3947.
25.7%
596.
3.9J
166.
1 .1%
664 .
1.3%
15367.
DUST/DIRT
IMPFR
85
1.61
9! 6S
?74.
15. OS
8.
4%
48.
2. 61
0
OS
64 .
0
.OJ
o .
.0?
0
OJ
1 174
hi .'%
1829.
DUST/DIRT
IMPFR
178.
47.35
139.
37 .4%
10 .
2 7%
1 4 .
3.8%
0.
.0%
31 .
8 3J
0.
.0%
0.
.01
0 .
372.
SE DIME (IT
TOTAL
97.
.71
23 i.
1 . 1%
742.
5 5%
40 .
.3%
3903
29 17.
2148.
16.01
3071 .
22. 9t
199
1 5',
1 IS
9J
1656.
12.6%
1 174.
8 3%
13411.
SEflMFNT
TOTAL
262.
1 .7%
346
2.2i
62.
.11
935
5 9%
8730.
55.55
3978.
25 35
596.
3.81
166.
1 1%
661.
4.25
15739.
AFtA
PFRV
1 .
3S
5.
1 0%
36
6.9S
9
5
95
173
3! 35
J9 1%
54 .
i r . 35
-7.
R •-%
3 .
.5%
0
.07,
526.
ARF A
PFRV
19.
2.0%
21 .
2.5S
1 3
1.4%
3
.35
444 .
46 4%
305.
31 .85
1 16.
12 15
31
3.21
3.
.3%
958.
',"E,I 'A
Iff-'. 7,'InL
, -i .'.;
-j . 1 " .
•J . 8 « 1 . 8 «
W.'il 9.6?
;. 11.
3 75 1.9',
3. 3.
7.3S 1 .45
0. 178.
.05 31.1%
13 220.
'i.4J 38.3%
54.
.05 9.45
27.
. % 4.85
.0% .35
5.?% .4%
46. 571.
ArtEA ARFA
IliPER TOTAL
5 24.
19 55 2.45
13. 34.
41.15 3.55
2. 15.
d.95 1.55
1 3.
i.o; .35
C 444 .
OJ 45. 3 J
6 . 311.
''6 OS 31 7%
0. 116.
.OJ 11 35
0. 31
i; 3.2%
3
1", 35
23. J8I.
1-17
-------
16 , j
4.5%
1-18
1 '1 -
6i;
-------
I-A-7. Waiter Cm3) and sediment (kg) loadings estimated by LANDRUM *cr e?ch I-inc! >j-
L-ummer 1^77
I AND USE
INI'USTRIAl
COMMERCIAL
MF1VFNS/RES
10 /DEIIS/RFS
HI /DEMS/RES
OLV-LOPING
ROW CROPS
PK/REC/PASTR
FORESTS
WETLANTS
WATER
FHLFWAYS
TOTALS
WATER
PERV
499.
1)230
6.2S
T5286
55. 2S
1873.
1 .91
5 21
14019.
14. OJ
2315
2.3S
14583.
14.'-?
0 .
01
187
0 .
.0?
0 .
.0?
100185.
'JATER
IMPER
1649.
3.9S
5872.
790'!.
18. 9S
68.
.2}
1245
3.0J
648.
1 .61
r<%
323.
81
0.
.0$
0.
19302.
46. 2S
4753
1 1 4J
41764.
WATER
TOTAL
2148.
1 .51
12102.
8.5J
63110
44. 5J
1941
1 . 4S
4~.5S
1 4667 .
1° 3%
2315.
1 .6%
14906.
10 51
0.
187.
19 ,02.
1 3 .6S
4753
3.3J
14194Q.
SEDlMFNT
PFRV
.01
"1%
1627
12. 71
63.
5S
30
.2S
28 '3.
~- .01
49 7%
I'l 2S
0.
.OS
2
. JS
0%
12-2C5
DUST/CIRT
161
645
14. OS
369
18. 9S
8
,2J
1 (7
3. OS
71 .
1 .51
0 .
.OJ
36
.0?
. 'i%
*l**l
?22.
n . -15
21(9f .
' '1 . 5 »
7 1 .
ur
1 .0?
?R94
""?.
1C ' °f,
1 «
.oi
2122
12 2S
522.
i ~J%
17396
AFcA
pi- J
IS
1 ii
3 1- . 5 «
! ;
i .1?
^ .
i a
r i . ;. **
1 :•?
^. Ti
-1 1
b "•>,%
-t
r%
n
13 . ? .
hATER
PERV
WATER
INFER
WATER
TOTAL
15113-
21 .7%
145.
1 .9%
2105.
?7.?S
11467.
14.85
16572
21.4*
344
1 8J
21 1
27. 25
308.
1 .fS
1 5 . *%
1 u
?9. f
16 ,
J 'S
DLVrLOPING
966 .
1 . 4S
18555
26. 71
32.
.4%
294
3. 81
100
1 3S
]
2635.
34 .0?
18655.
2 4 . 1 S
4114
5 . 1%
980.
1 3!
29.
3 71
1-19
-------
rcr !•
IN [HI T^1AL
,,ni-- :'.!«..
ru L •,. T i-'
:;• -,:r i- MO •
1 1 'at n '• t- ,
rf jtLvir ;-i,.
hew r?i • '
pK',,fc/p^r,;
FORESTS
WET..ANH3
FEEDL01 i
WATFR
FRtEWAl^
roTALt
Table l-'-K-.
I AND USF
COMMERCIAL
MED'DENS/RFS
10 /DttiS'r-t1-
DEVEI OPING
ROW CROPS
PK/REC/P,,TH
FCRESTS
WETLANDS
WATER
FPFEWAYS
ICTALS
PEPV
91 -'.
''f''l%
"5, ^
, 1 1 !t .
''"if
33. 'tt
" 1 .31
64926
ib.n
"?li
~5 61
1594
6?
0 .
0
.0%
,"49137
Wat "I- (mM in
S-iimmrr 1377
WATFR
PEPV
442
.5?
f 3 4 1 8 .
?5.4?
1658.
1.8J
,.-0250.
506C.
5.5%
19328
21 .0?
3264
3.5?
18737.
20.3*
0.
.0?
0.
.0%
92157.
WATER
IMPER
3674 .
r 9J
f-,54.
1.81
1 13.
61;
''?"i
0 .
OJ
1'J2.
.1%
0
.OJ
c.
0?
0.
01
353H5.
28.1%
6920.
5.5%
125991.
cl -:e< imenl
WATER
IHPER
1718.
6.6%
51309.
75 5?
585.
.8?
3689.
5 11
0
.0?
3 i?
o .
.0?
0.
.0?
4582.
6.1%
1562.
2.2?
71920.
WATER
TOTAL
12870.
3.4%
20908.
5.6}
54758.
14.6%
"2:I?
'.'it
151914.
40. 5J
3346.
.9?
651 1»
17.1?
5562.
1.5%
13998.
3.7%
1594.
.4%
35385.
9.4?
6920.
1 .8?
375178.
(kg) load nit.,
WATER
TOTAL
5160.
3.1?
77727.
47.4%
2243.
1.4%
23939.
14.61
5060.
3.1?
21803.
13.3%
3264.
2.0?
18737.
4582.
2.8?
1562.
1.0?
164077.
SEDIMEVT
PERV
165,
.1?
187
.11
^92
2.3?
102
. 1%
•,
8r>698.
63. 31
'9:94
20.81
'5516 .
11.11
597.
41
832.
.6?
1603.
1.11
0.
0?
0.
.01
140191 .
— >
SEDIMENT
PERV
33.
1?
16736.
28. 4S
826.
1 4%
12386.
21 .1%
24973.
42.4%
2966.
5 0?
272
.5?
677.
1 .2?
C.
.0?
0.
,0?
58869.
DUST/DIPT
IMPFR
368.
2.9%
361.
2 9S
6C5.
4.8%
1 1
j
1!
r.t.fi i ,
0.
0%
49.
.4?
0.
.0?
0
.0%
0.
.0?
3545.
28 1?
693.
5.5%
r2620.
LAiBRJN f
DUST/DIRT
IHPER
472.
6.6?
5440.
75.5?
58.
.8?
370.
5.1%
0 .
.0?
248.
3.4?
0 .
.0?
0 .
.0%
459.
6.4?
156.
2.2?
7203.
SttlMEM
ICTAL
533.
.3?
548.
'1%
~2 61
113.
.11
11 .
.0?
93679.
62. 6S
29194.
19.1?
15565.
10.2?
597.
.4?
832.
.51
1603.
1.0?
3545.
2.3?
693.
5%
15.281 1 .
^ cic n land use in
SEDIMENT
TOTAL
505.
.8%
22176.
S3. 6?
884.
1 .3%
12756.
19 3?
24973.
)7.8?
3214.
4.9%
272.
.4?
677
1.0%
459.
.7%
156.
.2%
66072.
ARr A
PERV
19.
1 . <%
9.
1: '".
i.-
AREA
TOTAL
2 .
.21
116.
13.61
10
1 .2%
14.
1 .71
317.
37.1%
224.
26.2?
73.
8. 51
90.
10.6?
1 .
. 1%
6.
.8%
853.
1-20
-------
TaMe I-A-11.
LAND USE
INDUSTRIAL
COMMERCIAL
MED/oFNS/RES
HI /DENS'RES
DEVELOPING
ROW CROPS
?Yf REC/PA3'- R
FORESTS
WETLANDS
hEEDLQTS
WATER
FRFFWAYS
TOTALS
T.J t1 i o 1-A- [ .' .
LAND USE
INDUSTRIAL
COMMERCIAL
M"'D/DENS,'t rr .< : "- <> '
WATER StPIMFNT DUST/DIRT StDIMFt;T AFLA '. = '.A
TOTAL PERV JMPrR TOTAL PrRV Ir' = t''
101252. ^25. 11752 12011. '-"
18.8$ 15 ?0.0$ 3.1$ .35 '.'- J5
156094. '707 17835 19542. 14. J;
29.05 .55 33.35 6.9Z 1 .=5 13 ""*
110981 '-?" 13261 . n49C. 1 !5. 4 =
20 6J 15 22.55 3 55 1^.85 13. '5
52i 0, 63- *3 3 • 5 '
.15 .05 .15 .05 7{ .5
404. 1 "'*• "9
.1% .05 15 -05 "5 .1".
52598 3?S304. 19'"'. 330343. 39. ".
9 85 90.35 3 35 84 »5 5 15 , '5
0. 1 0. 0 '•'
111 °5 -5 .OS 11 55 . '.
10^01 30. 113=.. 1164 >.
1 .95 Of. i 95 3? 5' .85 3 0?
0. -. 0. 0. 68
OJ r5 .05 .05 ") : '%
197. 'i. 'i 7 21
.(5 ."* .07- .05 ' '5 . 'I
?5. 32. 0. ^2.
.05 ''5 .05 05 ;5 «
72114 0 866" "66"-. ",. 15.
13.45 01 14.75 '.SI .05 7.<5
34546. 0. 4148. 4148. C "=.
6.45 OJ 7.05 1 15 .'-"« ?J 95
539P35. 330715 58850 380565. T^f i-.'-
.^\[, , r , ^ - ^ ^ - . 1 : . ' A ^ - 1 h [ ' - ' r ^ - - . - 1 - J -^ p f ' •<"'"',, . ('."
(KgJ l^.ivli., r, IPl.e- ' ul.^.'l - - „„. „-.._ 4^1-
HflEii 'F>MM-_N7 OU^T/'IRI "EO!K.;,T '-'i'- PREA
TOT.',! PtRV I^'PEF rUAL '-M'V -»F-R
993; 34 086. 37r "I. '•
6.15 . <5 6.15 115 '6- I, .45
30 1 cfi ^05 1 1 °6 . 1 ^91 . 1 J 2 4 .
23.95 .°% -'" "'5 4 15 , 'I 33 'i!
1 i»H ?8^ 181 . nf-7 44 13.
8.':'•
46085. 7'.4'' 3-. 797;. 'i ' • 7
2815 -f.25 .75 23-5 jt.-% "-5
.05 05 -% !'l -• ,:?
958. '3 0 18 i-.
65 IS 05 .1"- .-; 5
6062. 53. o 63
3.75 21 .rs ;5 , --,-:.
23219 , 2552 . '-^^^ ~ . ~ .
14.25 OJ 54.45 7 51 ", t. ",
3904. 0 429. 429 '. *, .
2.4$ 05 9.1$ 1-3$ CJ 25 15
163717. 29240 4691. 33931. 733. 36
'i
~6
1 - . 1
.'.r
;
-..'$
- '-,5
^ • 85
-' ?5
r 05
,5
' -5
... '}
7 -70
"*95
' '-, »
' ! -5
i;
' ^7
;'••-.
c. , \"a
£
u
,' ."
, ""J
a^,.
1-21
-------
UI.L. ,j"i
i f'MMf Rf I A
M H> / P Ml S / R F b
_ < / .. F N S / K F S
.11 /I ENS/RLS
DFV'r' r-PlNO
Rf.W "R.IPS
PK/kE n, PAST R
FORLSTS
':\f TL'", HI *>
'! (IT A L ^
LAND UiF
INDUSTRIAL
COMMERCIAL
MED/RE NS/Rf S
LU /DEBS/RFS
HI /DEBS/RES
DEVELOPING
ROW CROPS
Pft/KFC/PASTR
FORESTS
'«LTIAU>S
LUDF II L
WATER
TOTAI S
WATFR
PE PV
3701 .
,' 4%
>*b ?%
713
.51
504.
.31
'17753.
31 .4%
1 .55
"?!ni
1 1
^J
79'. .
1 5 1 S ? 6 .
'Vri t c"" (V 3 ) ''
"'ERV"
7258.
5.6J
6574.
5.01
27006.
20. 1%
.01
344.
• 3%
88197.
67. 5S
.01
0
. 0%
0.
07,
1257
1 .07,
21 .
.0%
n .
.0%
13C657.
WATER
IMPER
8654.
26 0%
'60.75
40
.15
335.
1.01
" o'."ll
0.
.05
'049.
0.
.0%
0 .
.0%
WATER
I1PER
3098.
5.45
14169.
24. 6J
20569.
35.71
16.
.OJ
109
.21
2595.
4.51
0 .
.01
103".
1.8J
0.
.01
0 .
.01
0.
.01
16086.
27.9%
5/676.
WATER
TOTAL
12355
6.7%
90464.
118. 8S
753
.45
839.
.5%
49775.
2292
1.2J
27931
15.11
11 .
.01
"n
185214.
( l'CT ) lOddlt' ~r
-------
-ich lani use ir " i^w
LAM' USE
IN,",srRIAl
CPMMFPfU!
MED/DENS/RES
LO /PFNS/PES
DEVELOP INC
HOW CROPS
PK/REC/PASTR
rCSE^Q
WATFR
FRFEtJAYS
WATER
PEPV
1 .0%
10031 .
18.7%
16385.
30 k%
445 .
1 1452.
21 II J
446 .
1V7B.
26.6%
0.
01
0
OS
WATER
IHPER
14.8%
25003.
27. OJ
36544.
39. 4J
'l«
1426.
1 .5%
0 .
.OJ
1189.
1 . 3%
0.
.OJ
1 1 r 4 o
11 9%
3682
4.0%
WATER
TOTAL
11310.
9.8%
'5031.
23.9%
52929.
36.2%
51 1
12873.
8.31
146
3%
15167.
10.6%
0
1 1040
7.5%
3682.
2.5%
3M1IMFNT DUST/DIF.T
"FfV IPPEB
L < 1511 .
1% 14.81
2011. 2749.
? 6J 27.0%
8711. 4017.
ll.'J 39.4'
"% !J
'},'•"- 157.
82 ,.% 1 5%
9 3 6
1 2J .01
19' <. 130
2.5' 1 3%
1 0
r,t j%
1214.
('- 11.9J
r . 405 .
.0% 4 0%
ilPIMHiT A'lr.A
TOT11 PEKV
1554. 0 .
1.RJ 2J
4760. 11
5.4" '1 ' %
12728. j;.
14 5% 25.21
153. '
.2J ! -".
63885. 6
lf.f.% -- 3%
936. 1o
1 .1% 7.2J
207?. 115
2.UJ 51 "•%
.r% '.'I',
1214
1 .4J , ij
405.
.55 ."{
AREA
I1PER
3
10.9*
1 4 .
19.8%
28.
0.
4 .5%
( ,
1 5 J
':'%
^ .
3 T%
1 p .
14 61
LAND USE
INDUSTRIAL
COMHERflfL
MED/DF-fJS/RES
1 r, /TENS/nES
nrvrLOPHK,
WATER
PFBV
l .3%
29641 .
37 OJ
9619
12.0J
3
OJ
39379.
WATER
1MPEP
?827.
9.71
11915
10 .8%
5561 .
19 0%
0 .
.01
15
• 0%
463
.1%
7692
7 0%
109387.
668.
19 '•:%
11.
1'/'
117
3 3J
0 .
."5
oj
924.
26.3%
34^ .
"i.W
1-23
-------
tfATEP
TMPFT
T-24
-------
1-iD 1 t -*- 1 1
LAND USE
COMMERCIAL
MED/: r|,i5/,ui
1 0 /DENS/RES
PFVEL )PING
ROW CROPS
PK'RLC'PASTti
FORESTo
WETLANDS
WATER
TOIALS
j'rlbl C I -A- 2 0 .
LAND I'St
COMMERCIAL
MFD/DF.NS/RFS
',1 /D5NS/PFS
HI /DENS/RES
DEVELOPING
Rr* CROPS
PK/iiFC/PASIR
FO.-STS
WETLANDS
FFFL'LOTS
WAT? R
TOTA;S
*a{f r (ml a
Sum.ni r 19/7
W1TF1
PERV
1111.
2 .55
6985.
15.55
4191.
9.35,
1595.
3.51
6122.
13.65
22187
49.31
109?
2 45
1761
^.95
0.
.05
45049.
S'ratrer 1977
WATER
PtPV
!78.
45
20420 .
19.55
17S9.
1 .71
803
81
33164.
31 61
4925
4.7?
33414 .
31 .91
1298
5391
5.11
j?04 .
"3.15
.05
104786 .
no sedimem 1
WATER
IMPER
249.
4.01
627.
10.25
115.
1.9%
445 .
.75
0 .
.0%
744
12. 15
0
.05
0
.05
43^0.
71 . 15
6150.
WATER
IMPER
698.
1 .15
1907.
2 9%
64.
. 15
176.
.31
725.
1 .15
0 .
. 05
296
0.
.05
0.
05
.«
61998.
94.11
65864.
.Kg) leadings
WAT EH
TOTAL
136,'.
2 71
761,2.
14 95
4309.
8.41
1640 .
3.?.1
6122.
12.05
22931.
44.81
1092.
2 1}
1761 .
3.45
'4^7'j
8.55
51199
WATER
TO' AL
1076.
61
22327
?3. \:
1853
1 . 15
979.
.(,":
33889.
11 ~,%
4925
? 9J
33710
19 85
1298.
.31
5391
3.25
3204.
1 .91
61998.
36.35
170650.
estimated by L^NDRU 4 tot* oT.'h lar.J .j:f i^ . t^cir '.r 1 "•' __ (
SEDIMENT rUST/DIRl SEDIMEM APSA AP:S
PFRV IMPER TOTAL DrcV IM"t^
2P. 24 46 6 1
.11 3.95 .15 i.H 2.71
104 63. 15' 3. 4.
. ?5 10.21 .5' 1.75 16 15
'9. 12. 91 . ' 1 . 2.
.35 1.91 .31 • 2 11 '1.0S
978 5. 983. 1.
3 25 .85 3.21 .15 1 35
27093. 'i. 27093. 353. 1,
89 9} 1,5 38.15 71 . -1 .rl
1745 76 1-VC. ,IQ. 1-:.
5.3J 1?.-I 5.91 I1: "I 64.25
67. 0 67 32. 0.
.21 .CI .25 6 45 .r".
36. 0 30. 3. 3.
.15 .OJ 15 1.51 .05
4-8. 438. 1. 1.
.07. 71.01 1 .445 01 3.31
301^4. 61/. 30f41. 5'= ~,tl
SFP1MENT ri'T/LIRT SrDIMEUT A^EA APIA
P'RV IMPFS TOTAI PU'W :i-'FEn
2 70 72. 1
.0-; 1.15 .11 OS 4.U
953. 191 11441 51 13.
t .3? 2 9J 1.4! ' . jZ 32.'-!
35 7 '42. 1' 1.
3 18 21 . 1 1
os n .r? 15 1.55
43988 V. 44061. 13. ="-.
s3 31 1 15 53.75 2.21 12.2?
17'' 3. " 21793. "»1
?3 07 .0?. P5.6Z '.4.61 . 5
bO?8. ^0. 6058. ^21 6
8 01 '•*, ' 4? -14 =1 '5 i ,;
'.]% ':% .'" 1 -I .'-5
184. 0 184 '4 r
.25 .01 21 '1 25 :r.
?j'!7 3. 2J47 ^
3.15 .05 ".91 .95 .'.5
.05 94 IS 7 61 .", -1 -.5
75416. 6599 32Ci5. 812
1-25
-------
WATER
IMPS-:?;
1-26
-------
Tablr l-A-23 Water (m3) and sediment (kg) loadings estimatf-1 by LANDF', ft fc r <=arh Irinr! >;se
Summer 1977
LAND USE
INDUSTRIAL
COMMtRt IAL
MEn/l'tNi/BEC
LC /DENS/BITS
HI /PENS/RFS
DEVELOPING
BOW (SOPS
PK/REi./PAS1R
FORESTS
WETLANDS
LANDFILL
WATER
TOTALS
Tab1 e ~ - i - '' i, .
[AND USE
INDUSTRIAL
COMMERCIAL
MED/DENS/RES
LO /DENS/RES
HI /DENS/RFS
DEVELOPING
ROW CROPS
PK/REC/PASTR
FORESTS
WETLANDS
FEFDLOTS
JATFR
TOTALS
WATER
PFRV
1878
2 6»
'4152.
5.71
7522.
10.3%
162.
.5?
1867.
2.61
40988.
56. 2J
0.
.0%
13634.
18.7*
0.
.0*
655.
.9*
1905.
2.6J
0.
.01
72963.
Summer 1 9
WATER
PERV
8425 .
"4.1.$
53642.
27.91
8329.
1.3*
1388.
.71
12913
6.71
620P5.
32.3?
6081.
3.'*
29966
15.6J
0.
.p*
9291.
14.8%
71 .
.0*
0 .
.0%
192187.
WAI ER
INFER
31352
9. 61
117919.
"45.11
86638.
26. 4S
753.
.2*
U0752.
12.4*
18633.
5.71
0.
.0*
1713.
5J
0 .
01
C .
.01
0.
.0?
396.
.11
328156.
WATEK
IMFER
3214.
6.5%
22036.
44.4%
1205
2.4J
'10
1%
2540.
5.1?
2096
4.2%
0
0?
202.
.41
0.
.OJ
0.
.0%
.01
18335.
36.9?
49668.
WATER
TOTAL
33230.
8.3*.
152071.
37. 9*
94160.
23.5*
1115.
.3%
42619.
10.61
59621 .
14 9%
0
OJ
15347.
3.8J
0 .
.0%
655.
.21
1905
.5?
196
.1%
401119
WATER
TOTAL
11639.
'1.8%
75678.
31.3*
9534.
3.91
!4?t> .
. 6S
15453.
6.4J
64171
26.5!
6084.
?.5%
30168.
12 5S
0.
0$
9294 .
3.8!
7 1
.0%
18335
7.6J
241855
SF.DIHFHT
PLRV
342.
.21
1793
i.d*
4303.
t.3I
no.
.21
797.
.4%
17774?.
95. 11
.')%
839.
5J
0 .
OJ
34.
.Ot
39.
. 0"
0 .
01
181^225
--DI1ENT
FERV
nO .
.1%
698
.71
349.
.4*
30
0%
111.
.11
68941
72.1%
?;-.'"*
4237 .
•J .IJ
0 .
OJ
199.
.25
24 .
0.
.01
95659.
DUST/DiRT
IMI'ER
3457
9 6*
1631?.
45 . T%
9555
26 4%
83.
.2%
4494.
12.41
2055.
5 11
0
C*
189.
.51
0 .
.0%
n .
.1%
o
44
1%
36190.
DUST/DIPT
IMF ER
354.
6 51
2422.
44 4%
132.
2.4%
. 17
279.
5 U
2"^0
o*
. fl'i
-'3
.41
0
"Z
.0*
.0*
2015.
36.9?
5459.
SEDIMt'ir
TOTAL
3799.
1 7j
'8106.
C IS
13858
6.2?,
413.
• 2%
529 i
1798;.;
8
-------
WA1FR WATER UA~FR
PFRV IMPFR TCJTM
IE "_ 66226. 681C1 .
. •% 9.6% 8.3J
'I892. 1V926 it^Klj.
XA.TE;1' WAI FK r-1 I'T'IT i" U^T /LIFT . t CT.t'FNT
IHPER lOTflL P-FV IMPtR T' TH
19.
Dl
10873 !>16C '503'. -1.1(4 062 infill-
"fit .6? 1 , ^f ?2. U ,6J 12 .*.
7017
p.It
11^13 0. 11313. 5": H
i 8J .OJ 1 .21 C5 '•'.
5867. 5867. '-. >,?',.
.05 .8? .61 'I .»!
0 18058. 18058. 0 213D
.fj 2.5% i 9% .oj 2.5S
752607. 9U66S3. K'9246.
1-28
-------
LAMP USE
INDUSTRIAL
COMMERCIAL
MED/DEHS/RES
HI /DENS/RES
DEVELOPING
PK/REC/PASTR
EORFSTS
WETLANDS
LANDFILL
WATER
FRFEWAYS
TOTALS
TaUe I-A-/S
LAND USE
INDUSTRIAL
COMMERCIAL
MED/DE'IS/RES
LO /DE'IS/RES
HI /DEN-i/
-------
0%
"49.
1-30
-------
Table I-A-31 Wat-r (m5) and sediment (kg) loadings estimated by LANDRUN SLT >-ach land
[AND USE
If.'l'USm/IL
COMMERCIAL
!'FD/DE»S/RES
LO /DENS/RES
DEVELOPING
PK/REC/PASTR
FORESTS
WETLANDS
TOTALS
I.SNb USE
INDUSTRIAL
COMMERCIAL
HED/DFNS/RES
HI /DFMS/RFS
DEVELOP I >,'C
PK/RF C/PASTH
WAI E 3
FRFEWAYS
WATER
65.
. 11
7721 .
16.31
1 1322.
23.91
140
.31
3"6 .
.7?
~46.65
0
.01
5719.
12.11
17168.
WATFP
PERV
1698.
2.2%
5816.
7 61
33928
44.4%
1150
5 4%
1 160
1 .61
29654.
38.85
0 .
.01
0 .
05
WATER
IMPER
238.
5.31
2603.
47.71
1811.
33.21
3.
.11
10.
.21
739.
13 51
0.
.01
0.
.01
5157.
WAFER
IMPER
71106.
8.91
1 1881 1 .
14.31
337468.
40.81
57191 .
6.91
837.
.11
63670.
7 71
752.
.15
175253.
21 .2%
WATER
TOTAL
353
71
10321
19.51
13136.
21.8?
113
-31
356.
.71
22864.
43 21
0 .
.05
6749
10.91
52925.
WATER
TOTAI
75804.
8.4$
124627
13 XI
371396.
11 1 ?
61341
6.8",
1997
93324.
10.31
752.
.11
175253
19.11
SEDIMENT DUST/DIRT SFDJMUl f AREA
PFRV IMPER TOTAL PLRV
0. 33. 3 < r
.05 5 21 1 .35 u5
88 301 . 38Q . 6 .
" 61 17.7} 15.35 2.15
641. -MO 851 09.
33.51 33.31 3-'. 15 22.41
2 0 •> . ' .
.15 .05 .IX 31
15. 1. 96 ii.
5 01 .25 i »! 11
002. '16. 588. 138
4 '.15 13.61 38.81 62.35
0 . 0 0 . 29 .
.05 n% .05 'I 15
186. 0. 186. 31
9 71 .01 7. <5 11 *1
1911. 631 ?5«5. 2f>5.
5EDIMF"JT DUST/DIRF SEDIMEK'I Jt'EA
PEPV IMPEF TOTAI FERV
159 8768. 892,'. l
7% 3.9"- 7 41 .5}
1519. 14f>7. 15676. <'-, .
7.01 143; 13 01 1r.8S
14453. 39926. 54379. '32.
62.81 4'".C5 44.9% '19-31
847 6 "86 7618. V
3 7% f, 91, 6.31 4 35
2110. 99 2209. 1
9 25 .11 1 .85
3821. 7533. 11364. 'i <
16 65 7 75 9 45 V- . 11
C 89. »9. C
05 11 11 •{
0 20735. ?0,'35. :
.0% 2125 171? . " 1
IMPER I-TAL
^ 71 .31
7 . 1 • •
^4,j5 4.31
13. 72.
45 15 2 • TJ
1 .
25 31
.21 11
3 146.
27 51 49.95
C . 2 y .
,01 10.01
OS 10 51
29. 29"
iSEA AREA
7.3". 4 '1
32 6' .
11.6; 11.25
m ", 4<_.6S
6 •<; 6 -~
i .
45 1 -
16 61 2, ,;•
.11
i<"'i '':';
1-31
-------
:,,.:-,-„
UNO liSF
INDUSTRIAL
i-WMf RC1A1
MFLVDFNS/RES
[0 'DFNS/RES
HI /DENS/RES
DEVELOPING
PK/REC/PASTR
WATER
FREEWAYS
TOTALS
Sjmmer- 19?
WATFR
PERV
1 49 1 .
1.4%
9433.
9.1%
69784.
67. 5 %
21.
.OJ
2181 .
2.3J
3300.
3-21
16963.
16 1%
0 .
.0%
0.
.OJ
103373
an
-------
Table I-A-.'S. Water (m3) and sediment (kg) loadings estimated by L4NPKUN fo
Summer 1977
LAND USE
COMMERCIAL
HED/DFHS/RES
LO /DENS/BES
HI ,'DFNS/RES
DEVELOP ING
ROW CBOPS
"K/BEC/PASTR
FORESTS
WETLANDS
LANDFILL
WATER
FREEWAYS
TOTALS
Tibl- I-.\-3f
LAND USE
INDUSTRIAL
COMMERCIAL
MED/DENS/RFS
HI /DEMS/RES
DEVELOPING
ROW CROPS
PK/REC/FASTR
FORESTS
LANbFKL
WATER
FREEWAYS
TOTALS
WATER
PERV
25^6.
2.T.
2^2?1.
26 3J
-jO.
.IS
3735.
3.9%
37596
39. 2%
0.
.0%
23121.
21. IS
51.
.IS
516.
.5%
2970
3 1J
0.
.OJ
0
OJ
95827.
•liter (m3)
Sumnrr - U
WATER
PERV
95
IS
5206.
1.6J
28878
25. 3S
3633
3.2Z
318 .
.31
114.
. 4%
75725.
66 21
0.
.OJ
1 4
OJ
0.
.OJ
0
.OJ
1 11313
WATER
IHPER
108888.
15 1$
361725.
51 IS
18.
.0%
76399.
10.81
23520.
3 3%
0.
.0%
10190.
5.7%
0
OS
0
.0%
0.
.OJ
2115.
• 3J
95117.
13. IS
708032.
dntl sedlirent
7
WATER
t MPER
1357.
7J
89686.
15 1J
279175.
17.9*
11919
7.2J
250.
.0%
0
.C'%
126602.
21 7J
0 .
.0%
0 .
OS
10730.
7. OS
230.
.OJ
582919.
WATER
TOTAL
1 1 1H4.
13. 9S
386951.
18. IS
98.
.0%
80131.
10.01
61116.
7.6S
.OJ
6331 1
7.9S
51.
OJ
516.
.11
2970.
.'4J
2115 .
• 3J
95117.
11 .SJ
803859.
Ug> -o-^np:
WATER
TOTlL
1152.
.61
91892.
13. 6%
308053.
11.2?
6 51
598
.11
411 .
.1?
202327
29.01
.01
11
.OJ
10730.
5.8S
230
.OJ
697292.
SEDIMENT
PERV
1088.
3J
31130 .
22.
.OS
1713.
.5?
313019 .
88 9<
0 .
.01
8163.
2.1J
15
.01
113
0%
~1I
0
OJ
.OS
385985
", L^i.Lu^pJ cy
SEOIMtNT
PFHV
1',
U'23
2, 1J
12i63.
25 5?
1392
7 OS
612
1 31
5015.
1C.3S
26053
53 81
C .
.01
o .
.0?
0 .
.01
0.
.OJ
18163.
DUST/DIRT
IMPER
11602.
15 . 1J
38511 .
51 .1J
5 .
.OS
8110.
10. 8S
2506 .
3.3S
0
OS
4282.
5.75
.OJ
0
."%
.os
229 .
"3*
1C 1 34 .
1 i 41
7513°.
LA'.L-.; ;.,r i
iJUST /riRT
IMPER
515.
.7%
10611.
15 4 J
33030.
I"7 .97,
4^53
7.21
30
.0%
0 .
.0%
11973
21 7J
:>K
0.
'481°
7 OS
OS
68969.
SEDIMENT
TOTAI
12690.
2. 8S
69971 .
15. ?S
27
.0%
9383.
?.1J
315555.
71. q?
.0*
12115
2.7S
15
CJ
113.
OJ
^ 32 .
.IS
229
.OJ
10131.
2 25
4 6 1 4 2 '. .
-,.. h lanJ .1:
SEDIMENT
TOTAL
.1%
1 1631
9.9S
15393
38.73,
8351.
7. IS
642
5015
1 3"
41031
31 .95
0
.0%
OJ
1819.
4. 11
27.
."I
1 17132
3 81
ID
i.'.l
15.
5 91
1-33
-------
I'aMf l-A-3/ Water (m3) and sediment (kg) loadings estimatna by LANDRUN for each land use in Subwaterched
Summc-r 1977
LAND IJSF
INDUSTRIAL
COMMFRCIAL
MED/DFNS/RF S
LO /DESS/FE?
HI /UNb/aKS
DEVriCPING
PK/REC/PASIR
WATER
,REEW«
TOTALS
TaL'-'- I-A--1.
1 AND 1'it
C01MEHC1AL
MFD/DENS/RFS
HI /DrNb/RF-i
DEVELOPING
PK/RtC/PASTH
K, RESTS
HATER
FF^ EWAYS
TTALS
WATFF
PEPV
18«9.
I.2J
>\5H?
3 OJ
103569.
69 41
1 1.
.OJ
7072.
4 7J
3940.
2 61
28 17 1 .
18.9*
.0*
0
.K%
149154.
..'jt.T (I, ')
WATER
"ERV
5463
16.7*
6632.
20.3%
5665.
17 3%
2044.
6.3%
12858.
39.4?
0.
.01,
0.
.01
0 .
.OJ
32662.
WATER
IMPER
39057.
2. 95
144870 .
10.8*
959344.
71.5?
10
.OJ
99972.
=809.
.2?
55442.
4. 1J
31053.
2.3J
9972.
.7*
1 542529.
ind sediment
WATER
IMPER
149847.
44 .OJ
58368.
17.1?
47413.
13.9?
1301 .
.4%
18833.
5.5?
0.
.0?
18301 .
5.4J
46396.
13.6?
340459 .
WATER
TOTAI
40906 .
2,1',
149412
10. J?
10b?913.
71.3?
21 .
1C:70'I4.
7 2J
6749.
.5*
83613.
5.6?
31053.
2.1J
9972.
.7?
1491683.
(Kg) loadings
WATFR
TOTAI.
155310.
41 .6?
65000.
17.4?
53078
14.3*
3345.
.91
31691.
8.5?
0.
.OJ
18301 .
4.97,
46396.
12. 4J
373121 .
SEDIMENT
PERV
157.
n
749
6J
100494.
81 4J
1 .
.0?
195?.
1 .6*
i C 4 4 3 .
8.5J
97H.
7 9J
. 0?
.01
estimated i,y
SEDIMENT
PERV
1481 .
3.6J
8824.
21 .6*
1939.
4 7!
22458.
55. OJ
6123.
15 OJ
0.
.0?
0.
.0?
0.
.0?
40825.
DUST/DIRT
IMPER
4519.
2.9J
16761.
10.8*
111011.
71. 5J
1 .
.OJ
11568.
7.4J
325.
2?
6415.
4.1?
3593.
2 3?
1154.
.74
155350.
LA\"LR!,'N for e
DUST/DIRT
IMPER
17 140.
44 .05
6754.
17.1*
5486.
13.9?
150.
.4?
2179.
5 5J
0.
.OJ
2M8.
5.4?
5369.
13.61
39396.
SEDIMENT
TOT1L
4676.
1 .''%
17512.
6.i?
21 1505.
75.8!
.0?
13520.
4.8!
10768.
3.9?
1 6 1 28 .
5.8J
3593.
1 .3*
1154.
.1*
278856.
-3-1 land use
SEDIMENT
TOTAI
18821 .
23.5*
15578.
19.4*
7425.
9.3J
22608.
28. 2J
8302.
10.3*
0.
.0?
21 18 .
2.6J
5369.
6.7S
80221 .
AREA
PERV
1 .
.3*
4 .
.8?
3C5.
65 5?
C
.0?
17.
3.6%
1 .
3?
137.
,-9.5?
.0!
&
.0?
.n Subwatershed
AREA
PFRV
4 .
3.5*
31.
26. 1J
4 .
3.6J
1 .
.9*
74.
61 .21
5
4. 1%
0 .
M
120.
AREA
IMPER
10.
2. 21
39.
8.2?
344.
72.51
0 .
.0?
31.
6.5J
2 .
.41
40.
8.4*
7.
1 .4*
2.
.5*
4/5.
3C (area
AREA
IMPER
40.
38.5!
21
20. OS
15.
13.9*
1 .
.8?
14
12.9?
M
4 .
3.?*
1 1 .
10.1*
10}.
AREA
TOTAL
12.
1.3J
43.
4.5*
649.
69.0*
0.
.0*
67.
5.0i
3.
.3*
177.
13.8*
7.
.7*
2.
.2J
940 .
in h-i )--
AREA
TOTAL
45.
19.81
52.
23.3*
19.
8.4J
2 .
.8?
87.
38.7?
6.
2 5*
4 .
1.3*
1 1 .
4. n
2^5.
1-34
-------
Table I-A-39. Water (m3) and sediment (kg) loadings estimated by LANDRUN for1 each 1 in
-------
i ule I-A-11
1 AtlD USE
INDUSTRIAL
fOMMFRCIAL
>--
AREA
TOTAL
15.
3.01
86.
17.31
157.
31.81
0.
.11
15.
9.11
12.
2.11
3.
.61
165.
33.21
3.
.61
9.
1.9*
196.
I-A-42. Water (m3) and SGdiment (kg) loadings tbt_ma{ed by LANDRUN for e=ch lard
Summer ID//
atershed 3G
LAND USE
INDUSTRIAL
COMMERCIAL
MED/DEHS/RFS
III /DtNS/BES
"K/REC/PASIh
TOTALS
WATER
PERV
573.
10.51
2852.
52 1!
1668.
30.6*
338.
6.2J
5113.
HATER
IMPER
922.
I 31
31559.
15.0*
21678.
30.9J
11933-
21.3*
1075.
1 5*
WATER
TOTAL
931.
1 .2*
32132.
12.5S
2»53
-------
LAND USE,
1MOUSTRI A[
^ori-"RCUL
MEP/DFNS/HES
HI /DL'IS/fES
iJFVELUPIHG
PK/ ilEC/PAS'I fi
WATER
FKEPrfAr,
I.A1ER
PEFV
103^2
\c>.'l%
18402.
3?. «l
374fi
n 'ii
2113.
3.71
1 2600 .
19 7%
12170
18 81
07,
OS
WATER
IHPER
H9535
17.51
272012.
31. 9%
2H2300.
28.141
5368"4.
6.3!
7902.
.91
71879.
8.8}
20609
2 «
32'4;1
3.8»
WATER
TOT (I
159857
17.11
290111
31.6?
251018
27.31
56079.
6.1J
20702
2. n
3701)9.
9.51
20609.
2 21
32151 .
3.5t
-if UP" NT
= ERV
11*6
^%
5227.
? ,71
6042.
3 IS
•M6.
I*
177102.
91 .°S
/6^6
1 1J
15
.1''%
nN?T/i,I"!
IMPEB
17957
17.51
32669
31.9%
290Q6
6 1 ') b .
f>.?7
9D9.
899J
8.8',
2175.
J.1%
3"97
3.B5
S^filM^ N~ , -F '
TOTAL -"J
ign?^ t-
f.uc ; .11
'-,'896 1^.
12. 8f 5 41
351'S in.
11 97, <:.5".
7192. 1 • .
-.it ).-'%
178351. 'j
60 37, 1 11
11619. 160
3.97, .-<.<;
21)75
*% .<:%
3897
1 . 3i ' J
i, L L ;
u^
1-). V,
25.51
3 - ', «
r *%
1 75
"7 .
1 - . 7 I
5
' .5*
3
.- "?
33935.
1 1 3S
1693.
.6J
56 11
2 It
1-37
-------
t ">'. D USE
I> ' I^TRIM.
roi" FPCIAI
MIL/Iif HS/'IF.'
in /nn.ViFs
I MM "PItiG
PK' RFC/PASTFi
T'TALS
I AND 'JSE
IMR'JSTRia
CnMMf RCIAL
MED/DEN5/PES
LO /DLMS/RE3
HI /DENS/RES
DEVELOPING
PK/REC/PAS1R
W1TFE
FBFEUAYS
TjTAL^
W (1 7 E ri
PF1.V
1 1
.0%
254=,.
77.6?
1069
3-1%
481
1 . IS
3536.
10.4J
34075.
Cur me? 19 77
ViAl <• F
14695.
9. 1J
26788.
16 .6J
60096
37 3$
252.
.21
•4203.
2.67,
713.
47
54411.
33.8$
0.
.05
.0$
161 158
WATFR
IMPER
234.
.1*
39918.
13. 9J
225380
•"8. 71
1 3224 .
4.6J
334
.1*
7257.
2.58
286847.
WATER
IMPER
630223.
26.11
963993.
37. OJ
557160
21 .41
306.
OJ
55930.
2.1J
505.
.OJ
111711.
4 .4?,
44208 .
1 11,
190438.
7.3*
5607527.
WATER
TOTAL
245.
11
42463.
13.21
252313.
78.61
14293
4.57-
815
-3",
10793
3.U
320922.
WATER
TOTAL
694918 .
25. 11
990731 .
35.31
617256
22. 3J
558
0$
60133.
2.2?
1218
.Of,
169125
6.1$
412C8.
1 6$
190488.
6.9$
2768685.
5FDIM NT
PH.V
.'-%
2.3J
69'5
87 7J
105
1 3$
43C
5 "f,
259
3 31
7889
SEDIhEIIT
PERV
2122
4 4$
4259.
8.9$
23628
49. 2J
15.
C$
685.
1 .4%
1147
2.4J
16195.
33. 7J
0
.05
.0$
48051 .
DUST/DIRT
I M P E R
28
.1J
4723.
13.9$
?6'724
78.7%
1565
4 6$
40.
1$
859
2 5J
33939
DUST/DIRT
IMPER
80479
?6 . 1$
1 14052.
37. OJ
65919.
21 4$
36.
.0$
6617.
2. 1$
60.
.0?,
13572
4 4$
5230.
1.7J
22537.
7.3*
308502.
SEDIMENT
TOT1L
28.
.1$
4903.
11 .7J
33639
80 4?
1670
4 1$
470.
1.1$
1113.
2 7$
41823.
SEDIMENT
TOTAL
8260I.
23- ?»
1 1831 I .
33.2$
"9547.
25 I
5' .
.OJ
7302.
2 .'J
1207 .
.35!
29767
8.3J
523C.
1 .51
22537.
6.1J
356553.
AREA
PERV
r$
J»
68.
72 3$
2 .
2 3$
2J
13.
80 .
AREA
"CUV
13.
3 7J
23
6 7!
173.
50. 8 J
. 1J
9.
2 7J
.11
35.9$
h
n
310.
AREA
K'PER
0 .
.1?
1 1 .
10.6*
32
80. 0»
4 .
4 0$
>,
5.
5.1'.
102.
7.-.e *. I * ( i' <- :
AREA
IMPEP
185.
23. OJ
262.
32.6$
202.
25.1*
.07,
17.
2.2J
'-$
83
10.45
10 .
1 . 2J
44
5 45
803 .
AREA
TJTAL
.OJ
13 .
,'.2J
139.
7a. 6J
6.
3 . 3*
.2J
23.
12.7J
182.
i i r. r - ) - -
AREA
TOTAL
197.
17 .33
285.
24. 9J
575.
32 8!
. U
27
2.3J
1 .
2C5.
ia.cj
10 .
.SJ
14
3 ^2
1 1-.3 .
1-38
-------
Wj-^r (m3) and sediment (kg) loadings estimated by LANDRUN for -a<"h
Summer |077
LAND USE
INDUSTRIAL
COMMERCIAL
MED/DENS/RES
HI /DENS/RFS
I)FV FLOP INC
PK/RECVPASTR
LANDFILl
WATER
FREF.WAYS
TOTALS
Table [-A-13.
LAND USE
INDUSTRIAL
COHMFRUAL
MED/PENS/RbS
HI /DENS/RES
DEV HOP ING
PK/P'C/PCSIR
LANDFI1 1
kATtl!
WATER
PFRV
2386.
1.01
10073
16 91
19 25
3551.
f>.0%
1368.
2 3%
9553.
16.05
3351.
5 61
0 .
.0%
n .
59558.
W,it
Sum-Tier 137
WAI ER
PERV
1176.
2 3%
7026 .
1 3 85
30675.
60.1%
1 H9.
2 2%
1316.
6389
12.5,5
29?.
.6%
0
"*
WATFR
IMPER
1 11205.
15.05
200712.
P7.15
31.0*
16094 .
6.25
919.
.15
15712.
2 1%
0.
.0%
13422.
1.8J
100811.
13. 6J
741215.
and sediment i
WATER
IMPER
52664.
1J.5K
114516.
22 9%
288972.
57.8%
15529.
3.1*
3016.
.6%
14238 .
2 85
0.
.0%
11225
2 2%
WATFR
TOTAL
113591
11 2%
210815.
26.3%
281520.
35 25
19615 .
6.25
2317.
35
25295
3.2%
3351
.17.
13122
1 11
100814.
12 65
800773.
(k,:> leaijnss
WAFER
TOTAL
53810
9 81
121512.
22.15
319617.
58.01
16568.
3.0%
7362.
1 .31
20627
3.7*
29?
. 11
11225.
2.0%
SrPIMEUT
F E R V
222
1 .35
1510 .
9 3*
7811.
47.0%
527.
i.2%
2204.
13.35
3362.
20.2*
965.
5.85
0.
.0%
0 .
01
16631 .
.LtJI-rtfd by
SEDIMENT
PERV
".41
680.
2.85
1 r 7 1 8
14 45
142
.6%
1C771 .
14 6%
7oi
15 .
21
0 .
.0%
DUST/DIRT
IMPEP
13157
15.0*
23750.
27.15
29841.
31.05
5454.
6.2%
112.
.1%
1863-
2.15
0 .
.0%
1588.
t.S?
11928.
13 6*
87696.
LAM! PUN for
DUST/DIRT
IMPER
6231 .
10.51
13649.
2?. 91
34189.
57. 8«
1837.
3. 15
360
.6%
1685.
2.8*
0 .
.01
1328.
2.25
SEDIMENT
TOTAL
13379.
12. 8J
25290.
24.2*
37655.
36.15
5981 .
5.7*
2316.
2.2%
5225.
5.05
965.
1588 .
1 .5%
1 1928.
11 4%
104327
each land use'
SEDIMENT
TOTAL
6322.
7 .'•%
11.229
17-15
14907 .
53. 9J
1979 .
2.4%
11131.
13.4%
3380
1.1%
45.
. 1%
132S.
1 .61
SRfA AREA
PERV IMP-R
2. i! .
1.35 13.2*
15. 55.
9 3* 23.9%
65. 91
40.3* UC.-'J
7. 14.
4.5% 6.3%
0. 1 .
.2% .3%
58 11.
36.0* 5.CS
13. 0 .
/.9* 05
0. 3.
.0% 1.3%
0 73.
.c?, 1C. 15
I'O. P2?.
i- -.bult,r.,N- 1. (J^3
AREA AREA
PERV ' II'PER
1 . 11 .
7% 3.1%
7. 31
1.91 18 .45
93. 1C5.
«S.6J 61.7*
3. 5.
2 01 r ?%
1 . ' .
1.1% 1.11
30. 1C
21.8* 6 1%
1 .
.t)S .r%
.0% 1 45
-'-!'
i ' ',':
H.'.S
4". il
'• 55
25
17'S*
', 2*
3.
51
; 3
5-9*
3i)9.
;- ,)--
AhEA
TO.'IL
1 6 .
5.CS
33
12.41
1 57
64. S*
3
1 . IT.
13 i*
1
.4%
.1",
59179.
83321.
1-39
-------
PART II
MODEL ENHANCED UNIT LOADING (MEUL) - A METHOD
OF ASSESSING POLLUTANT LOADINGS FROM A
SINGLE LAND USE
by
V, NOVOTNY
G, CHESTERS
G, V, SIMSIMAN
ll-i
-------
ABSTRACT
The Model Enhanced United Loading (MEUL) method utilizing the LANDRUN
model has been developed to simulate potential pollutant loadings from urban
and non-urban land uses. The simulations for typical land uses are evaluated
as if the land uses are located on hydrologically different soils
representative of standard hydrologic categories. Pollutant loadings vary
considerably among land uses. Sensitivity analyses indicate that the most
significant factors affecting such differences are extent of imperviousness of
urban areas, portion of the impervious areas directly connected to runoff
channels, depression and storage, length of dry period between rainfall, curb
height for urban areas and soil type, slope and vegetative cover for pervious
urban and non-urban areas. The applicability of the unit loading data
obtained by the MEUL method has been tested on several well-monitored
subwatersheds in the Menomonee River Watershed. The simulated unit loadings
for sediment and phosphate-P are of the same order of magnitude as the
measured values.
Il-ii
-------
CONTENTS - PART LI
Title Page I I-i
Abstract Il-ii
Contents II-lii
Figures ••• H-iv
Tables Il-vi
II-l. Introduction II-l
II-2. Conclusions II-3
II-3. Methodology II-4
Pollutant Transport Process from Non-point Sources II-4
Pollutant Loadings and Transport from Impervious
Urban Areas •.. • II-4
Unit Loadings from Pervious Areas II-5
Rainfall factor, R II-6
Soil erodibility factor, K II-7
Slope-length factor, LS II-7
Vegetative cover factor, C II-7
Erosion control practice factor II-7
Delivery ratio factor, D II-8
Application of LANDRUN Model - Model Enhanced Unit
Loading (MEUL) Simulations Based on Land Use II-8
Surface characteristics II-9
Soils 11-10
Soil erosion data 11-18
Pollutant accumulation in urban areas 11-22
Atmospheric pollutant deposition 11-22
Wind erosion 11-29
Motor vehicles 11-29
Litter deposition 11-29
Effect of vegetation 11-29
Pollutant washout 11-30
Street sweeping practices 11-32
Meteorological inputs 11-32
H-4. Results and Discussion 11-39
Simulated Loadings 11-39
Comparison of Measured Loadings with Estimates
Obtained by the MEUL Method 11-44
References 11-50
Appendices
II-A. Detailed Statistical Evaluation of Street Litter
Accumulation . • • 11-53
II-B. Simulated Loading Diagrams 11-61
II-C. Remedial Measures and Non-Point Pollution Control 11-76
-------
FIGURES
Number Page
II—1 Depression storage capacity in relation to degree of
land slope .............a................ 11-11
II-2 Fraction of impervious areas not directly connected to
channel ....=. 11-12
II-3 Moisture characteristics of selected soils .< 11-15
II-4 Soil particle size distribution accepted by USDA-SCS 11-16
II-5 Relationship between soil permeability arid soil texture ... 11-17
II-6 Determination of soil K factor 11-19
II-7 Pollutant accumulation schematic model 11-23
II-8 Curb length-imperviousue.ss relationship „ 11-25
II-9 Seasonal cunmlative frequency of precipitation ............ 11-35
11—10 Seasonal cumulative frequency of R-factor « 11-36
11-11 Slope correction factor for sediment and phosphate
loadings from pervious urban areas ..«. 11-41
11-12 Loading multiplier for different slope categories IT-42
11-13 Relationship of the size of the area to sediment
loading , 11-43
11-14 Sediment delivery ratio versus drainage area 11-47
II-A-1 Effect of dry periods on the quantity of street litters ... 11-59
II-B-1 Sediment loadings from residential areas 11-62
II-B-2 Sediment loadings from commercial areas 11-63
II-B-3 Sediment loadings from industrial areas 11-64
II-B-4 Phosphate-P loadings from residential areas 11-65
Il-iv
-------
T.T-B— 5 Phosphatp-P loadings from commercial nrcac ................ T]-6f,
II-B-6 Phosphate-P loadings from industrial area^ ................ 11-67
II-B-7 Relationship of sediment load ings and R-f actor in row
crop-woodland areas . . ..................................... 11-68
II-B-8 Probability distribution of sediment loadings in row
crop-woodland areas ..... ...... ...... ................ . ..... 11-69
II-B-9 Relationship of sediment loadings and R-facator in
f eedlots .................................................. 11-70
IT-B-10 Probability distribution of sediment loadings in
f eedlots .......... . .......................... ............. II-7 1
II-B-11 Relationship of sediment loadings and R-f actor in
pastures ... .................... . ....... .. ................. 11-72
II-B-12 Probability distribution of sediment loadings in
pastures ............................................ ...... 11-73
1I-B-13 Relationship of sediment loadings and R-factor in
wetlands ............. . ....... . ............................ 11-74
II-B-14 Probability distribution of sediment loadings in
wetlands .................................................. 11-75
II-C-1 Effect of sweeping interval on pollutant loadings
(sweeping efficiency = 50%) ........ . .................. .... 11-78
II-C-2 Effect of sweeping efficiency on pollutant loadings
(sweeping interval = 7 days) .............................. 11-79
II-C-3 Effect of curb (median barrier) height on street litter
accumulation .............................................. 11-80
II-v
-------
TABL13R
Number Page
II-l Properties of soils used in the simulation ,. 11-13
II-2 Properties of the major soil types surrounding the
Donges Bay station (463001), Menomonee River Watershed 11-14
IT-3 C-value vised to compute erosion 11-20
II-4 Metal concentrations of eurficial materials of the U.S.A. ... 11-21
11-5 Street refuse accumulatior ..». . „ 11-24
11-6 Pollutants associated with street refuse 11-26
II-7 Metal contamination of street refuse 11-27
II~8 Annual and monthly mean deposition rates of particulate
material in Milwaukee County ....I..................... 11-28
11-9 Daily leaf fall 11-31
11-10 Pollutant distribution in various particle sizes 11-33
11-11 Interrelationship of sweeper efficiency and particle size ... 11-33
11-12 Street sweeping removal efficiency of pollutants TI-33
11-13 Urban land use information 11-37
11-14 Non-urban land use information 11-38
11-15 Simulated pollutant loadings for urban land uses under
slope category B (2 to 6%) during an average year (1968) .... 11-40
11-16 Simulated pollutant loadings for land uses on essentially
pervious areas 11-45
11-17 Comparison of simulated and measured sediment and phosphate
loadings in subwatersheds with mixed land uses 11-48
11-18 Comparison of simulated and measured sediment and phosphate
loadings in predominantly single land use areas 11-49
Il-vi
-------
II-A-1 Partial and multiple correlation coefficients between dust
and dirt pollutants and factors affecting their accumulation. 11-57
Tl-vii
-------
II-l. INTRODUCTION
The International Joint Commission, through the Great Lakes Water Quality
Board, established the International Reference Group on Great Lakes Pollution
from Land Use Activities (PLUARG) to study and report the effects of land use
on water quality and recommend remedial measures. Several pilot watersheds
subjected to detailed monitoring were selected throughout the Great Lakes
Basin in Canada and the United States. The Menomonee Ri.ver Watershed located
in the southeastern part of Wisconsin in the Milwaukee metropolitan area was
one of the watersheds selected. The primary task was to establish pollutant
loadings from various land uses and extrapolate these findings to the entire
Great Lakes region.
The investigation discussed in this report presents an effort to develop
unit loadings for typical urban and suburban land uses using a combination of
modeling techniques with measured monitored data. It is true that the best
information on actual loadings can be obtained only from direct field
measurements. However, the applicability of such information is limited by
time and location at which the data were gathered and sometimes by the
sparsity of data. On the other hand, even the most effective models may fail
to provide reliable results if proper calibration and verification is not
guaranteed. Thus, a combination of simulated loadings using a mathematical
model, calibrated and verified by extensive monitoring data and applied to
several hydrologically different seasons and soils, may provide a better
understanding of the variability of the loading figures, their dependence on
meteorological, pedological and environmental factors and may reveal a
possible impact of some remedial measures suggested for reducing pollutant
impact.
Pollution from non-point or diffuse sources originates either from
weathering of minerals, erosion of virgin and forest lands including residues
of natural vegetation, or from artificial or semi-artificial sources. The
latter sources can be related directly to human activities such as fertilizer
application or use of agricultural chemicals for controlling weeds and pests,
erosion of soil materials from agricultural farming areas and animal feedlots,
erosion occurring in urban developments, transportation, atmospheric fallout,
etc. With the gradual elimination of point sources including sewage and
industrial wastewater outfalls, it is becoming obvious that a substantial
portion of surface waters pollution originates from the use of land by man,
i.e. from diffuse sources.
A tendency exists to relate pollutant loadings from non-point sources to
type of land use. In this approach, pollution from diffuse sources is
expressed simply as a value or range of unit loadings (loadings/unit area/unit
time) for the land use. This approach, though justified as an initial rough
approximation may lead to results which deviate markedly from measured
values. More appropriately, it is important to examine and analyze the basic
processes and factors involved in pollutant generation from diffuse sources.
II-l
-------
The tlodel Enhanced Unit" Loading ('!r.ip loadings generated in this way are abstracted fron ,-•
particular location at a particular time and reflect for a typical ar^a mean
pollutant accumulation characteristics and statistically averaged
meteorological conditions subjected to certain land uses. The pollutant
loadings developed in this report do not include background or natural
composition of surface waters caused by its contact with geological layers,
undisturbed soils and natural vegetation.
Limitations of the MEUL method include:
1. The method is intended basically for comparative assessment of
loadings among various land uses.
2. The loadings are related to a few primary variables such as degree of
irnper vioiisnnoS of the arei, rl mnli nf>«q of the area, soil characteristics and
type of land use.
3. The meteorological inputs represent a typical average meteorological
year for the Midwest (Milwaukee). The accuracy of the estimates for pervious
areas was improved by considering the 10 and 90 percentiLe meteorological
seasons selected from 30 years of weather observations in southeastern
Wisconsin.
4. The pollutant accumulation rates on impervious areas represent
average U.S. rates as reported by Sartor and Boyd (1).
5. The loading figures were computed for five typical urban land uses
(residential, commercial, industrial, developing and parks) and five typical
non-urban land uses (row crops, pastures, woodland, wetland and feedlots).
6. The loading figures are not intended to be used for estimating
accurate loadings in areas where no historical or monitoring data are
available.
7. No monitored pollutant loadings from pervious areas and only limited
loadings from impervious areas during winter conditions in Midwestern areas
are available.
II-2
-------
I1-2. CONCLUSIONS
Large amounts of pollutants are washed Into surface waters from non-point
sources. The factors contributing to non-point pollution from various urban
and non-urban land uses have been investigated using a calibrated and verified
hydrologic transport model capable of simulating overland pollutant loading
and transport. The simulated seasonal loadings provide a comparison of the
variability and potential danger to surface waters of typical land use
activities. The model was calibrated and verified using field data from the
Menomonee River Pilot Watershed Study. The simulated loadings for typical
land use areas were evaluated as if the land uses were located on four
hydrologically different soils representative of standard hydrologic
categories. Developing urban, high density urban areas with no cleaning
practices, livestock feedlots and steep-sloped crop lands yield the highest
pollutant potential while parks and recreational areas, low density
residential and most urban areas with good cleaning practices produce much
less pollutants. The differences in pollution potential among the land uses
were several orders of magnitude. Summer rains in Midwestern areas have the
highest erosion potential; however, spring rains on bare soils with frozen
subsurface generate the highest sediment runoff on row cropland. By
sensitivity analyses, various parameters have been tested as to their effect
on loadings. The most significant parameters are extent of imperviousness of
urban areas, fraction of impervious areas directly connected to surface
runoff, depression and interception storage, average length of the dry period
preceding a rain, curb height for urban areas and soil type, slope and
vegetative cover for pervious urban and non-urban areas.
Various control techniques and their impact on non-point sources
pollutant generation have been discussed.
The loading diagrams which relate sediment and phosphate-P unit loadings
to the most important causative factors have been developed and their
applicability tested on several subwatersheds in the Menomonee River Basin.
Estimated and measured loading values were of the same order of magnitude.
II-3
-------
I1-3. MCTHOnOLOCY
Pollutant Transport Process From Non-Point Sourcec
Water is the primary mover of pollutants through the environment from
their sources to the place of final disposal. Unlike pollutants fron point
sources which enter the hydrologic transport route during a late stage of the
hydrologic cycle, (channel or estuary Clow), non-point source pollutants enter
the hydrologic rotitp during its early stage, i.e., in precipitation or by
overland flow. The point where the pollutants enter the hydrologic transport
process depends not only on the type and location of the source but also on
the physical form in which the pollutant occurs. Gaseous, emulsified and
dispersed airborne pollutants enter the water transport route following
deposition on the surface by wet or dry fallout. Soluble pollutants nix with
water directly. Relatively insoluble pollutants either are dispersed and
picked up during rain or snownelt events through subsequent surface runoff, or
are transported by wind and subsequently redeposited. Furthermore, pollutants
can be adsorbed by soil and dust particles and transported by water in the
particulate phase.
It is anticipated that non-point pollutant transport processes in urban
areas may be different from those in non-urban areas because:
1. Large portions of urban areas are impervious resulting in much higher
hydrological activity.
2. With the exception of construction sites most of the pervious
surfaces in residential or city areas are well protected by lawns and are
subject to less erosion.
3. Pollutant loadings in urban areas are affected mainly by litter
accumulation, dry or wet fallout and traffic while in non-urban areas most of
the pollution is due to erosion of soils and soil-adsorbed pollutants.
4. Over a large period of time (season) almost all of the pollutants
deposited on impervious surfaces which have not been removed by street
cleaning practices, wind or decay, eventually end up in surface runoff. On
the other hand, in non-urban areas soil represents an extensive pool of
sediments and pollutants adsorbed by soil and their removal rate depends then
on the energy of rain or runoff which liberates the soil particles and
eliminates surface protection.
Pollutant Loadings and Transport From Impervious Urban Areas
Pollutant accumulation on ground surfaces in urban areas and subsequent
washout by runoff represents a major pollutant contribution from non-point
II-4
-------
urban sources. Since impervious areas are almost fully hydrologically active,
most of the runoff and associated pollutants in highly urbanized areas
originate from these surfaces. The amount of deposited pollutants depends on
various factors and inputs. The major inputs are atmospheric fallout, street
litter deposition, animal and bird fecal wastes, dead vegetation, and road
traffic impacts. The factors which affect the quality of street refuse washed
out to surface waters include land use, population density, traffic flow and
frequency, effectiveness of street cleaning, type of street surface and
condition.
It has been realized that a simple unit loading value related to land use
may not provide an adequate estimation. Instead, the loading values should he
correlated to major causative factors which for various urban land uses can be
listed as follows:
a. Percent impervious area directly connected to a channel (a funrt-jon
of land use or percent of imperviousness).
b. Population density (a factor related to land use),
c. Dry and wet atmospheric fallout.
d. Litter accumulation (a factor related to population density and land
use).
e. Traffic density (a factor related to land use).
f. Curb height and length/unit area (factors related to land use).
g. Percent open area (a factor related to land use).
h. Average wind velocity.
i. Street cleaning practices and effectiveness.
j. Average number of dry days preceding a rain or rain intensity.
k. Depression and interception storage (a factor related to land use).
With the exception of low density residential areas, other factors such as
slope, soil type, are expected to have little effect on pollutant loads from
urban areas because most of the loading originates fron impervious areas.
It can be seen that most—but not all—of the above listed factors are
indeed related to land use. Thus, it may be possible to develop a multi-
dimensional loading factor for various urban land uses which would be a
function of:
a. Dry fallout (primary independent variable).
b. Street cleaning frequency and efficiency. ) parametric
c. Average wind velocity. ) independent variable
d. Average number of dry days preceding a rain. )
Unit Loadings From Pervious Areas
Urban or suburban pervious areas with the exception of those overlain
with heavy clay soils or areas with a very high groundwater table are
hydrologically active only during extreme storms or during spring melt or rain
events when the ground is frozen. Freezing of the surface layers in
Midwestern areas of the United States also provides protection against erosion
and groundwater contamination.
II-5
-------
3edi..:i_.\t and so-: 1-adsorbed po "•.•;( •,,••- .•-'.;;., ;J, in-'-.-" ,••', aU nal , ios;
pesticides) can be modeled by t'ie Universal Soil Lo^1- Ir.<'ju,i r ion (I'SiV,). Tin-1
equation in i t~c. or igjn-il i orn ('?) can be written as:
A - (III (K) (Lei) (C) (F) Eq. (i)
where
A is amount of sediment generated/stern
R is the rainfall energy factor <->f the storm
I' is MIR soil erodibility factor
LS is the length-slope factor
C is the w'^etati^'e cover factor
P is the erosion control factor
In this form the equation represents the amount of soil particles liberated by
rain energy impact. In order to obtain the sediment load to receiving waters
t"he equation rou'M: he multiplied by a delivery ratio:
AS = D * A Eq. (2)
where AS is the sediment load and D is the sediment delivery ratio.
Loadings of some pollutants other than sediment are then estimated by
PL = AS * CP * RP Eq. (3)
where
PL is pollutant loading
CP is pollutant content of the soil
RP is the enrichment factor accounting for the difference
in pollutant concent in soil and the sediment
suspended in water
It is possible now to estimate which of the above variables is land use
related.
Rainfall factor, R
This is a function of storm intensity and volume and is not related to
any land use activity.
The rainfall energy factor, R, is computed according to the equation:
R = E1{[(2.29 + 1.15 log Xi)]Di}l Eq. (4)
II-6
-------
where
X. is rainfall intensity, cm/hr
E. is rainfall hydrograph time interval
D. is rainfall depth during time interval i
I is the maximum 30 min rainfall intensity of the storn
in cm/hr
It is evident that the rain energy input/season reduced by the amount of
snowpack on the surface is the^ major independent variable affecting the soil
loss estimation.
Soil erodibllity factor, K
This is purely a function of soil characteristics (2,3). For most
Midwestern soils the K factor is in the range 0.1 to 0.4.
Slope-length factor, IS
This is hased on formula (2):
LS = 1/2 (0.0138 + 0.00*5743 + 0.0013852) Cq. (5)
where
L is length from the point of origin of the overland
flow, m
S is the average slope over the given overland flow
length, %
The equation indicates that soil loss is more sensitive to slope changes than
to the size of the area.
Vegetative cover factor, C
This variable depends on the crop or vegetative cover and the season. It
varies from 0.005 for heavily wooded areas to 1.0 for bare soils. Besides the
rain energy factor and slope this is a variable to which soil loss is very
sensitive.
Erosion control practice factor, P
This factor depends on erosion practices implemented in the Watershed.
In the absence of such practices the value assigned to this factor is unity.
II-7
-------
Delivery ratio factor, D
This is probably the most difficult factor to estimate. For larger
watersheds the delivery ratio seems to be a function of watershed size and
configuration. For smaller areas it may be a function of the lot roughness
(depression and interception storage) and, mainly, permeability. For
relatively homogeneous sites, a study by the Midwest Research Institute (4)
related delivery ratio to soil texture and drainage density which is defined
as the ratio of total channel-segment lengths to the basin area.
If a loading function is to be developed it should be related to the
rainfall energy factor as a primary independent variable, with soil type,
slope and depression storage as parametric variables
Application of LANDRUN Model - Model Enhanced Unit
Loading (MEUL) Simulations Based on Land Use
This method used in the study to develop loading functions relied on
field data and system simulation. It has been realized that although the
field data provide the best information on pollutant loadings from a
particular site the information is limited by time and location at which the
data were gathered. On the other hand, even the most complex simulation model
of a watershed can provide results quite far from reality if the model is not
properly calibrated or verified.
A model developed for this study has the code name LANDRUN (5). It is a
deterministic watershed model capable of simulating the following processes:
a. Snowpack-snowmelt by the Holtan or Philip Models.
b. Infiltration by the Holtan or Philip Models.
c. Excess rain can be computed as the difference between precipitation
and evaporation, evapotranspiration, infiltration and surface
storage.
d. Routing of excess rain by an Instantaneous Unit Hydrograph Method.
e. Dust and dirt accumulation in urban areas and washout.
f. Removal of accumulated pollutants on impervious areas by cleaning
practices.
v' g. Surface erosion by a modified quasi-dynamic USLE which includes
effects of rainfall energy and sheet runoff.
h. Routing of the sediment and sediment-adsorbed pollutants.
The model takes into consideration several parameters including:
a. Land use data.
b. Meteorological parameters.
c. Pollutant input.
II-8
-------
The computer model is capable of estimating:
a. Storm water hydrographs and volume.
b. Sediment transport from pervious areas.
c. Dust and dirt washout from urban impervious areas.
d. Volatile suspended solids in the runoff.
e. Adsorbed pollutant loadings.
A dynamic soil adsorption segment is an optional feature of the nodel which
enables detailed study of pollutant-soil interactions (6).
Following calibration and verification of the LANTiRUI." node! (7),
pollutant loading simulations were conducted for the land uses agreed upon by
PLUARG. The land uses were grouped into urban and non-urban categories:
Urban uses Non-urban uses
Low density residential Row crops
Medium density residential Pasture
High density residential Livestock feedlots
Commercial Woodlands
Industrial Wetlands
Park and recreation
Developing
To simulate pollutant loadings, each land use was assigned typical values
for such variables as degree of imperviousness, fraction of impervious areas
directly connected to a channel, depression storage, permeability of pervious
areas, slope, soil moisture characteristics, etc. In addition, other
variables describing atmospheric fallout, litter accumulation, street sweeping
practices and the USLE inputs were selected. The values were based on
Menomonee River Pilot Watershed data or on literature values typical of
Midwestern urban areas.
Surface characteristics
The model requires a detailed description of the hydrologic
characteristics of the subwatershed surface. Included are: Degree of
imperviousness, depression and interception (surface) storage, subwatershed
slope, surface roughness and extent of impervious areas directly connected to
a channel.
Most of the land surface data was obtained from the SEWRPC Land Data
Management System (Land DMS) (8). Unless otherwise specified default values
were substituted in the model for depression and interception storage and
surface roughness. For combined depression and interception storage
characteristics, default values used are: 6.35 mm (1/4 inch) for pervious
areas and 1.58 mm (1/16 inch) for impervious areas. These values are similar
to those used in the Chicago study (9) and other urban studies. For non-urban
pervious areas a graph developed by Hiemstra (10) served as a guide to
selection of the storage characteristics (Fig. II-l).
II-9
-------
Surface roughness characteristics are necessar" if routing ot pollutants.
Ls required. The value of the Manning roughness factor for pervious areas is
0.25 and for impervious areas is 0.012.
The impervious areas not directly connected to the surface runoff
channels include rooftops discharging through underground drains, paved areas
overflowing on adjacent pervious surfaces, etc. This factor can be related
approximately to the total impe.r viousness of the area as shown in Fig. II-2.
The simulated areas were 1 km for earh land use.
Soil s
For simulation purposes, four soils typical of the Ilenomonee River
Watershed or immediate vicinity were selected. These soils are representative
of each basic hydrologic group ranging from the most permeable hydroiogic
group A to the least permeable group I) (]]).
Table II-l shows the basic soil data used in thp simulation; these data
reflect typical values for soils given in SCS soil maps, More exactly
measured values for ten major soil types in the Donges Bay Road subwatershed
(station 463001) are reported in Table 11-2.
Some of the data such as 0.3-bar moisture tension (field
capacity) and 15-bar moisture tension (wilting coefficient) are ur.r- vailable
from soil maps. In this case, a graph relating moisture characteristics to
median particle diameter of the soils was prepared using data fn..1;1 the
Menomonee River Watershed and literature vali.es (7if>. 11-3). The median
particle diameter in mm was computed using a formula suggested by Horn (13):
d = TH7T t°'3 (% sar>d) + °-01 (% silt) + 0.002 (% clav] Eq. (6)
m 100
The particle sizes (Fig. 11-4) are the averages of the particle size ranges
recommended by the U.S. Department of Agriculture (USDA).
The permeability ranges related to soil mean particle diameter are shown
in Fig. II-5. Known and measured data for some Wisconsin soils indicate that
a lower range of permeability seems to be typical for Wisconsin rather than an
average theoretical curve. However, data measured by Bouma et al. (14)
represent permeabilities of septic tank seepage fields after several years of
operation and may not provide a good approximation of permeability of typical
undisturbed soils. Such values confirm the lower limits of the permeability-
texture relationship.
Soil erosion data
Use of the USLE requires a knowledge of: the rainfall energy factor (R),
soil erodibility factor (K), cropping management factor (C), erosion control
practice factor (P) and the slope-length factor (LS).
11-10
-------
Contour Furrows
25
Fig. II-l. Depression storage capacity in relation to degree
of land slope (10).
11-11
-------
1.0
0.8 -
0.6
O
4-J
O
CJ n ,
p 0.4
0.2 --
20
80
100
Total Impervious Area, %
Fig. II-2. Fraction of impervious areas not directly
connected to channel (12).
11-12
-------
Table II-l. Properties of soils used in the simulation
Soil type
Property
Hydrologic group
Depth of A-horizon, cm
Sand, %
Silt, %
Clay, %
Mean diameter, mm
Organic matter, %
Permeability of A-horizon,
cm/hr
0.3 bar 1^0 content, %
15 bar H20 content, %
Porosity, %
K factor*
PO^-P adsorption,** ug/g
Total P content, ug/g
Boyer Is
A
41
80
15
5
0.415
0.5
40
8
0
30
0.09
243
1,000
Hochheim 1
B
20
45
39
16
0.138
2.0
10
20
7
34
0.24
346
1,500
Ozaukee sil
C
28
15
55
20
0.051
3.0
3.0
30
17
43
0.31
403
1,800
Ashkun stcl
1)
28
5
56
39
0.021
8.0
0.5
36
24
46
0.15
697
3,100
*K is the soil erodibility factor used in USLE.
**Soil adsorption maximum obtained from the Langmuir isotherm.
11-13
-------
c
o
o
e
s
o
o
CO
o
•H
tfl
4-;
fO
05
0)
C
s
Q)
4-1
60
C
•H
C
O
01
cu
CX
•H
O
CD
S-l
O
(C 0)
CO 4J
•H «3
-H US
M
D !-4
P. 0)
O >
M -r)
P-i K
c^]
I
rQ
0)
H
INI co co U
rHOOQO
- u
c ~e
11-14
-------
piieg
pups
UIPO'I
(OUIj) ACT3
AE-[3 3TTS
'A) AE-[D
31TS
13
O
O
-------
Q)
d
•H
fi
CO
05
U
0)
c
cB
CO
0)
C
(3
tfl
CO
c
n)
CO
0)
CO
M
td
o
u
Q)
CO
J-l
ctf
O T3
O C
tO
>•, oo
M
OJ
a)
cfl
S-i
O
o
o
o
CM
o
o
o
o
o
o
o
CM
o
o
o
O O
m
CD
o
•
CXI
o
in
o
o
Particle size, mm
Fig. II-4. Soil particle size distribution accepted by USDA-SCS.
11-16
-------
t~
L
tuui 'j
ooooooooo
o o
o o
o o
o o
a c
a a
C n) C e d
cfl o nj TO nj
w HJ co co co
o (^ o
^J y j_5
a >, >,«
O -H H i-i
J O U CO
CO O CO CO
11-17
-------
The va.lup nf R. is computed by the LANPRUN nodpl from rh«=> rainfall dif.s
and the LS factor is estimated from average slope and area of NM> subwatershf-H
for each land use. However, the remaining Three factors nust he inputt-< d for
each soil arid land cover. Figure II-6 is a nomograph for estimating K. Th»-
factor K is determined from the contents of silt and very fine sand (particle
size 0.01 to 0.1 mm), sand (0.1 to 2 mm), organic matter, soil structure and
permeability. The K factors for the selected four soils are:
Soil K factor
Boyer Is 0.09
Hochheim sii 0.24
Ozaukee sil 0.31
Ashkum sicl 0.11
The factor, C, is dependent on type of groundcover, general management
practices and composition of the soil. For simulation purposes, the values
suggested by Brandt (15) were used (Table II-3). For agricultural cultivated
lands C was 1 during the spring season and adjusted to its tabular value for
summer and fall.
The P factor was 1 for most land uses. Some erosion control was assumed
on croplands.
Organic matter content of soils was selected to reflect typical values in
the Watershed.
Phosphate-P content of soils was based on the known range of P content of
the Ozaukee sil (P = 0.18 %) which was determined from the measured total P-
suspended solids relationship from the spring runoff at the Donges Bay Road
station. The phosphate-P content for other soils was adjusted according to
their adsorption characteristics, Q° (6).
The lead content of average soils is very low. The U.S. Geological
Survey (USGS) has undertaken an in-depth study (16) to determine the elemental
composition of surficial materials in the United States. Soil samples were
collected from 863 sites throughout the 48 conterminous states and analyzed
for 44 elements. The average values for eastern and western parts of the
United States are presented in Table 11-4.
11-18
-------
pups auT,j
11-19
-------
Table TT-3. C-va.l ue used to compute
(15)
Land use C-value
Cropland 0.0.°>
Grassland 0.01
Woodland 0.05
Construction 1.00
Urban 0.01
11-20
-------
Table II-4. Metal concentrations of surt'iclal materials of the U.S.A. (16)
Element
As
Ba
Cd
Ce
Cr
Co
Cu
Fe
Ga
Ge
Au
Hf
In
La
Pb
Mn
Mo
Nd
Ni
Hb
Pd
Ft
Re
Sc
St
Ta
Te
Tl
Tli
Ti
U
V
Yb
y
Zn
Zr
Total
Average, yg/g
. *
554
—
86
53
10
25
25,000
19
—
—
—
—
41
20
560
3
45
20
13
—
--
—
10
240
—
—
—
—
3,000
—
76
4
29
54
240
30,100
Range, yg/g Conterminous U.S.A.
< 1,000
15 to 5,000
< 20
<150 to 300
1 to 1,500
<3 to 70
<1 to 300
100 to 100,000
<5 to 70
< 10
< 20
< 100
< 10
<30 to 200
<10 to 700
<1 to 7,000
<3 to 7
<70 to 300
<5 to 700
<10 to 100
< 1
< 30
< 30
<5 to 50
<5 to 3,000
< 200
< 2,000
< 50
< 200
300 to 15,000
< 500
<7 to 500
<1 to 50
<10 to 200
<25 to 2,000
<10 to 2,000
._
430
—
75
37
7
18
18 , 000
14
—
—
—
—
34
16
340
—
39
14
12
...
—
—
8
120
—
—
—
—
2,500
—
56
3
24
44
200
2,990
Geometric means, Mg/g
West of 97th meridian
__
560
—
74
38
8
21
20,000
18
.._
—
—
—
35
18
389
—
36
16
11
—
—
—
9
210
—
—
—
—
2,100
—
66
3
25
51
170
23,858
East of 97th meridian
300
—
78
36
7
34
15,000
10
—
—
—
—
33
14
285
—
44
13
13
—
—
--
7
51
—
--
—
—
3,000
—
46
3
23
36
250
19,263
* Below detection limit.
11-21
-------
i'ol.lutant" -irrumniat ion in iirKin areas
The basic feature of urban areas is the extent of inpcrviousnr"",.'- of t!ui
land surface. Resides the hydrolog:ieal significance of impervious areas
(higher runoff, shorter duration of high pollutant concentrations, higher
flood peaks), essentially all pollutants are flushed into the receiving waters
whenever runoff takes plan"1.
Pervious urban areas produce pollutant loadings of lesser magnitude
provided that these areas are not steep and are well protected by lawns,
shrubberv and trees. The amount of pollutants deposited on impervious areas
depends on various factors and inputs as mentioned earlier. Pollutants
transported from impervious areas can be carried by wind and traffic impact
and they accumulate near the curb. Thus, it has been reported that street
pollution accumulation rates are related to the unit length of curb (Fig. II-
7; Table. II-5). Reporting street refuse loadings/unit length of curb, instead
of a more meaningful area loading, seems to be justified since it has hern
observed that nlno^f R0% of refuse can be found within 15 en and 97Z within J
rn of the curb (17). The strong correlation existing between curb length
density and degree of imperviousness of residential areas (Fig. II-8) can be
utilized for simulation purposes.
A recently-developed regression formula (9) between curb length of urban
areas and population density is:
CL = 311.67 - (266.07) (0.839)(2>48 PD) Eq. (7)
where
OL is curb length .in ra/ha
PD is population density, persons/ha
Refuse washed from streets by runoff contains many hazardous
contaminants. Significant organic pollutants, toxic metals, pesticides and
bacteria are associated commonly with the dust and dirt fraction (Tables II-6
and II-7). It should be noted that these values, though typical, are not
uniform but represent averages from a wide range of refuse deposition and
contamination from a limited number of municipalities which have been studied.
Atmospheric pollutant deposition
Deposition of atmospheric pollutants occurs as dry or wet fallout. The
deposition rates of particulate atmospheric pollutants in United States cities
vary from 3.5 to >35 Tonnes/km2/month. Higher deposition rates can be
expected in congested industrial areas or business districts while lower
deposition rates are common in residential and rural suburban zones (Table II-
8).
11-22
-------
DUST FALLOUT FROM INDUSTRIAL
AND STATIONARY FUEL COMBUSTION PROCESSES
I
POLLUTANTS CARRIED
AWAY BY WIND AND
TRAFFIC
MEDIAN
BARRIER
LITTER
DEPOSITS
CURB
POLLUTANTS EMITTED FROM
MOTOR VEHICLES
POLLUTANTS ACCUMULATED
AT ROAD SURFACE
Fig. II-7. Pollutant accumulation schematic model.
11-23
-------
Tablr IT-5. Streoi. reiiinr arr.unul.iM on
Land USP
Single family
Multiple family
Commercial
Industrial
Weighted average
Solids accurnu
Chicago*
10.4
34.2
49.1
68.4
22.3
lation, g/curb m/ciay
Eight U.S. cities**
48
66
69
127
*Taken from (9); data is for dust and dirt only.
**Taken from (1); data is for total solids which contain
75% dust and dirt.
11-24
-------
400 -
300-,
CO
E
f\
,c
G
-------
Tahle II-6. Pollutants associated with street refuse
Pollutant
BOD5*
COD
Volatile solids
Total nitrogen
Nitrate-N
Phosphate-P
Total metals
Zn
Cu
Pb
Hi
"8
Cr
p,p'-DDD, ng/g
p.p'-DDT, ng/g
Total coliforms, organisms/g
Fecal coliforms, organisms/g
Concentration, ug/g total solids
Residential Industrial Commercial Total
5,000 3,000 7,700 5,000
3^,800 59,000 31,500
78,000 56,500 77,000 71,400
1,020 870 600 1,570
32 41 314 67
600 800 550 78r>
2,040 1,150 1,800
460
140
4 10
36
32
78
48
43
71xl06
40xl06
*Taken from (9).
11-26
-------
Table TI-7. Metal contamination of street refuse (19)
Contaminant
Concentration, ug/g total solids
Residential
Industrial
Commercial
Total
Cd
Cr
Cu
Ni
Pb
Sr
Zn
3.45
186
95
22
1,468
23
397
2.83
208
55
59
1,339
134
283
3.92
241
126
59
3,924
151
506
2.82
183
101
31
1,324
177
338
11-27
-------
c
•H
r-J
cd
•rl
0)
4-1
cd
S
0)
4-1
cd
rH
3
CJ
•H
4-1
J_j
cd
PI
o
en
01
4-1
cd
l-t
c
O
•H
4-1
•rH
CO
0
d)
(3
ca
01 /-v
S o
CM
rH
r^ >1
4-1 4J
C fl
0 3
e o
u
(3 01
cd CD
^
rH 3
cd cd
3 Ss
C rH
P -rl
cd
H
J.J
J>^
--^
OJ
6
— ^
CD
0)
c
c
0
H
CD
4-1
cd
$-1
C
o
4J
•H
CO
O
O.
0)
T)
rH
cd
3
C
£
-i CO O B "O
60 0) 0 O (3
<3 Pd rJ U rH
O
•^^
CN
3
^^^
CO
0
C
O
M
^•^
ON
xO
ON
rH
O
4J
^
LO
ON
rH
0)
4J
cd
c
•H
4J
•rl
CD
0
ex
(U
^\
rH
_r^
4-1
C
o
Iel3
f j
W
0)
C3
>
•S
j *
a
o
4-1
oT
01
S?
>
CTj
S
a
^_i
fTj
XH
| O
OJ
J^
c
*
"-5
o
CD
CO
rH
CT\
C^l
w
O
rH
v^Q
*
0
rH
xO
•
o
rH
tr,
•
O
rH
CO
CN
rH
LO
-------
Wind erosion
The effect of wind erosion on surface particulate pollutant loadings
seems to be significant only occasionally. Factors important in the
assessment are: climate, soil characteristics, surface roughness, vegetative
cover and length of the eroding surface (21). In urban areas the primary
source of wind eroded materials are open, ungrassed areas and construction
sites.
Motor vehicles
Traffic can contribute significantly to pollutant deposition in urban
areas. High amounts of some metals in storm water runoff are attributed to
motor vehicle emissions and to the breakdown of road surface materials and
vehicular parts. Motor vehicle usage can influence pollutant accumulation in
urban areas and near high density traffic lanes by emission of pollutants, oil
and gasoline spillage, mechanical impact of traffic, tire abrasion, etc.
Therefore, in addition to traffic density, the pavement composition and
conditions are significant in determining traffic impact on pollution.
Streets paved entirely with asphalt have provided total solids loadings of
about 80% higher than all-concrete streets (17). Streets where conditions
were rated "fair to poor" were found to have total solids loadings ~ 2.5 times
greater than those rated "good to excellent" (1).
Litter deposition
Litter deposits in urban areas include solid wastes dropped from garbage
collectors, animal and bird fecal droppings, fallen tree leaves, grass
clippings, etc. The dust and dirt component of litter (material <3.5 mm) is
regarded as having greatest pollution potential; although most of the litter
is orginally larger in size than dust and dirt, the mechanical fracture of
litter increases the amount of dust and dirt. It has been reported that
residential areas had greater amounts of street surface dust and dirt as
population density increased, reflecting increased pedestrian and roadway
traffic (9). It is also expected that the higher the population density, the
greater the street deposition from garbage collections.
Effect of vegetation
Leaf fall and grass clippings in urban areas contribute significantly to
dust and dirt accumulation. For most of the year, the accumulation on
impervious areas arises from erosion of soils from surrounding pervious areas,
atmospheric pollution and litter accumulation and during the fall season, leaf
fall increases the organic solids accumulated at the surface.
11-29
-------
Hf'aney and Huher (22) estiuatrd from the stnrlv (.r Carlisle1 ct al. ( ' V:
that average leaf fall was 14 to 26 kg/free/year. The area investigated was
stocked with trees ranging in age from 40 to 120 years with a 90 to 95'> Hosed
canopy, and 151) trees/ha; species were mainly oak and birch. Typical values
for leaf fall in Minnesota are ~380 Tonnes/km2/year in a forested area with
~420 trees/ha with 65% occurring during the fall season. Fallen leaves are 10
to 97% organic matter and contain about 0.04 to 0.28% P (24).
For loading simulations, values of leaf fall for various land uses were
estimated (Table II-9). Organic and P contents of leaves were assumed to be
90 and 0.1%, respectively.
A detailed statistical evaluation of street litter accumulation is
contained in Appendix II-A.
Pollutant washout
Not all pollutants accumulated during a period preceding a rainfall are
washed off the impervious surface during the initial moments of the rain. The
rate at which loose participate matter is washed from street surfaces depends
on three factors, namely, rainfall intensity, street surface characteristics
and particle size (17). It can be expected that the amount of pollutants
washed off generally will follow the equation:
PL = ~ = - K L Kq. (8)
dt p
where
PL is pollutant washout rate
L is amount of pollutant present on the surface
K is a coefficient depending on rain intensity and
street surface characteristics
The coefficient, K , which was found to be independent of particle size in
the range of 10 to 1000 ura is approximated as follows:
where
Kp = EUR Eq. (9)
E is urban washout coefficient
R is the surface runoff rate, cm/hr
11-30
-------
Table II-9. Daily leaf fall
o
Leaf fall, Tonnes/km /day
Land use
Forest
Parks
Low density residential
Medium density residential
Spring-Summer
2.45
1.22
0.17
0.08
Fall
7.0
3.5
0.35
0.18
High density residential,
commercial and industrial 0.016 0.036
11-31
-------
^ :^ l.di ' h-ivp h 'on reporter1 f-,r Hi
Mot all litter is available for transport by surf.icp runoff. i iit n.-f or*;
spdiment washout rate should be multiplied by an avai labilitv f^rior (25) .>s:
Aq = 0.57 + 0.5 R1'1 Eq. (10)
It is obvious that a limit must be placed on the availability factor as runoff
rate increases, A suggested value for the maximum A is 0.75, wbich implies
that about 25% of urban litter is unavailable for transport.
Street sweeping practices
Street sweeping is a common practice in American cities whereas in
European cities streets are washed. Most of street sweeping is done
mechanically either by brush or vacuum. Removal efficiencies with brush
sweepers are shown in Table 11-10; removal of deposited suspended solids is
~50% with one pass of a sweeper. Some pollutants are associated more with
finer particle fractions (Table 11-11). By cumulative multiplication of
sweeping efficiency for each fraction and pollution concentrations on
particles of the fraction, overall efficiency can be estimated (Table 11-12),
e.g., the efficiency of sweeping for P control would be 22%. compared to 50%
for total solids. Street washing is more effective for fine materials.
Meteorological inputs
The climate of the Milwaukee area is influenced by the general storms
which move eastward across the upper Ohio River valley and the Great Lakes
region.
Annual precipitation is about 762 mm (30 in); two-thirds of which occurs
during the growing season. Thunderstorms, which carry the highest erosion
potential, occur less frequently and with less severity than in areas to the
south and west. The maximum rainfall which occurred in a 24-hr period is 172
mm (5.76 in) in June 1917. As much as 20 mm (0.79 in) has fallen in 5 min, 28
mm (1.11 in) in 10 min, 34 mm (1.34 in) in 15 min, 42 mm (1.86 in) in 30 min,
and 57 mm (2.25 in) in 1 hr.
The average yearly rainfall energy factor, R, for sediment loss
estimation by the USLE assigned for the Milwaukee area is R = 125 (2).
It has been realized that pollutant loadings shall be representative of
an average season, i.e., they express loadings which would be a mathematical
average over a long time period. In order to obtain such averages, at least
20 to 30 yr of data is necessary. In the absence of such a data base, as is
almost always the case, water quality (loading) data time series can be
generated by a properly calibrated and verified model using a measured
meteorological time series as input. Hourly precipitation data for the
11-32
-------
Table 11-10. Pollutant distribution in various particle sizes (17)
Particle
(im
>2000
840-2400
246-840
104-246
43-104
<43
size,
Pollutant distribution, %
Total solids Volatile solids COD TKN
24
7
24
27
9
5
Table 11-11.
Particle size
>2000
840-2000
246-840
104-246
43-104
<43
Overall
Table 11-12.
Pollutant
Total solids
.9 11.0 2.9 9.9
.6 17.4 4.5 11.6
.6 12.0 13.0 20.0
.8 16.1 12.4 20.2
.7 17.9 45.0 19.6
.9 25.6 22.7 18.7
Interrelationship of sweeper efficiency
and particle size (17)
, um Sweeper efficiency, %
79
66
60
48
20
50
18
Street sweeping removal efficiency of
pollutants (17)
Removal efficiency, %
50.0
PO^-P
0
0.9
6.9
6.4
29.6
56.2
Volatile solids 42.5
COD
TKN
PO^-P
31.0
43.9
22.2
11-33
-------
•'1; iwmkf'p -iri'T ,rf> i«/.i i 1 r.^lf anrf , i ! ,n, 'pries cnvr< iny -, vr '-'at- Teparcu.
Tn an ideal case, the simulation period would cover .m entire '57 y of
data, but with more conplex models mirh sinulation periods nay prow '•o ho
prohibitively expensive requiring considerable couputer time and storage
capacity.
To avoid the expensive, long simulation runs, the 37 vr series of
meteorological data was analyzed as to its distribution of seasonal wetness
and erosion potential.
The wetness analysis utilized a simple summation of precipitation per
calendar season; the seasonal erosion potential is based on the USLL R factor
as expressed by Eq. (4). In analyzing the erosion potential, only rain events
were counted, snowfall was omitted.
The probabilistic distributions of seasonal wetness and erosion potential
are shown in Figs. II-9 and 11-10. The arrows indicate the probabilistic
expectancy of season from the monitoring period 1975-1977. It should be
pointed out that the graphs are typical for the storm patterns in the
Milwaukee area and should not be generalized to other areas.
Summaries of the final land data used for simulation are in fables II —l?l
and 11-14.
11-34
-------
400.-
300.-
200--
IOO--
0
'16
1977'
SUMMER
12
\
WINTER
TO
50
(%)
Fig. II-9. Seasonal cumulative frequency of precipitation.
__ —Q
--4
95
11-35
-------
Fig. 11-10. Seasonal cumulative frequency of R factor .
11-36
-------
a
o
•rH
.U
Cd
s
^J
0
UH
G
•H
0)
CO
3
XI
5
rH
s
43
t-l
^
B
m
rH
1
M
H
CU
rH
43
a
H
CO
3
13
a
m
00
•H
a
o
rH
HI
CU
Q
C
"° -H
4-t
* cu
* 5
iH
cd
•H
M
4-1
CO
a
rH
cd
O
•H
M
CU
6
a
0
0
(=-,
4J
•H
CO
G
CU
43
60
rH-
^ "5J
4J ^
7* CD
5 "o
vU
XI g
CO ^
/V* Xt
« cu
4-1
•H
CO
G
CU
XI
3
0
M
O
60
CU
cd
CJ
o in m o
• • • -i- •
no ooo o CNI o o
en o m CM m
o
rH
'-N'->
* -K
* * -H
o -st in o
• • • •
rH O O
rH
>«o oo i — *
o- -* o
• • -)- •
oo o ^ oo r^-o OCNJ o o
C3 ON i^ ^^ CD 1^*1 f*^! *-^
CO O
f-(
OO i-i
(N CO O
* * ~f~ *
sOOiO O O OO r-^O OCN O O
i— I O O^i r-H \O OtnCNLO
/*v CO O
1— t
LT» r-i
CM m o
iHOO iO ^O OO r-^O Ocs) O O
I"*- vO LO ^*D O LO CNI LO
o f^ o
4J iH
m
IT) rH
U-J C^ vO O
o^m o m oo r-^o OCN o o
•tj ON c*>j ON ««^ ^^ in CN) in
o
CO rH
•
O
•X
C
o
•rH CO T3
4J >, CU
Cd *• CO 3s$ 4-t
rH 4J XI O CO
cd 33 CU 3
43 4-l0Orl "XIX l*HO
60 C OrHB OXIGC n} >
C « U d >^ S C 0) -H 0) S-^lH
•H (0 cd cd IH rt 3 0) C CO -H cd (U
rH CUCU^ XICO 3-H4JO CU-O.
rH M tH p 4J ^^^ & Cd (3 «H p 60
CD cd cd •> CU t-iCS XI CU4J-HMH IdCn
3 XI4J -HSC MficdtH -HO
XICdCOCOCU4J Xl^lCd ^H^iHSCU COCXMH
43 3 3 4-1 tH ^, ^^. CC1 3 CU
«.~->. O OOrHcdxItOOO 600XOOCO OCU l-i
600-H-HCU XICOIC G i-HCXIpj-HSO*
(3 r* r* G 4J ^. CdG*HrH •HrHUl "H'tHI ^CO 4JCO
•H -MrlGtUS CMrHP.rHOCl.rH-, COM
3 M OwO.O l-l M w 4J CO fe 0) 3 CL, 0) CO OH CX43>4-ICO
0 3 S 0 *-> 3 3 3 9
ffiOHMCOQ CO CO MCJ
W
jj
^3 • CU
e o
GOG
iH -H Cd
4J C
xi cd cu
OJ 4-J 4-J
CO CU C
3 00 -H
CU CO
H > B
o
4J XI 4H
O G O
cd id
UH CU
rH O
00 rH C
C id cu
•H UH CO
a 43
dc UH Cd
o cd
)-i CU CU
O rH 43
CU CO
43 CU CO
4J XI CU
3 4->
CO rH O
•HOG
C CU
U M O
* * +
*
11-37
-------
i
1
i
t
j
f
(-<
V—
o
•H
4—1
1
s-i
o
M->
C
•H
CU
cn
T3
c
CO
, — |
C
CO
,0
S-l
3
1
C
^o
r^i
•
•vl-
i-l
1
M
M
CU
rH
,0
CO
H
i
!
0)
'0
~3
T3
C
m
hJ
cn
T3
C
cO
rH
T1
O
O
Cfl
T)
rt
cS
rH
(1)
*-^
"J
S-I
3
u
CO
cO
&4
cn
4J
O
, — I
T3
(11
0)
D.
0
S-<
O
s
<§
>>
^
O
M
CU
4-1
Cfl
CJ
"-1 LO
C '—^
o c
» • » •* »
r-1 CO O C
O CN CN O O
rH O O rH rH
cc
O 01 CN O C'
• • « t •
rH O O rH O
^
K^
CO
-a
O ~-~ }J
•Hcsl CU
nee r-
CU ^i C 7H
5-S -C ' — 3 cd
a, cn cc M-J
•> co cu
cfl o d t3 -a
a) s £ c p
VH 4-1 O CO Cd
Cfl CO 4-1
M 60 (_|
CO T3 « C C 01
3 C 4-1 -H rH * -H 6
O c03VJrH n^G
•H O Cu cfl O P. 3
> s-irHcnciw 4-icncn
SJ CU rH CJ
CU 4-1 CO Cfl
D- 4-1 M-J '-M
C -H
M J U
£-
•
rH
O
*
rH
O
^
rH
0
rH
OC
«
C
*
*
S-i
o
4J
CJ
CO
<4-l
C^
to
cd
at
4-1
d
ni
Ol 60 -O
"> -H CC
Cfl 43 -t-1
CU
00
r
•TJ
cu
4-1
O
CO
"C 05
C 4-1
cd d
cj a
>r-4 S-l 4-1
c cu c
ex ,1: co 4-1 o
G to tc 4-1 o
o -H s-i cd
o X o E a,
to
cd
i-i
a)
cfl
•H
o
C/J
c/; -H
P o
•i-H 4-1
CJ
S-i cO
O «-i
4J
0 C
CO O
50 cfl
C >
•i-l SJ
e. ai
a. co
o c
s-i c
o o
cu cu
CO CO
.r-l -H
cj e-,
11-38
-------
RESULTS AND DISCUSSION
Simulated Loadings
The simulation results for each land use and characteristic season
produced loading diagrams which related loadings of pollutants (sediment,
volatile suspended solids and phosphate-P) to the R-factor for mostly pervious
areas and to atmospheric fallout for impervious areas. These loading diagrams
are presented in Appendix II-B.
Loadings for urban areas were related to the degree of imperviousness and
accumulation rates established for relatively clean areas (i.e., areas which
are swept about once a week) and areas with on cleaning. The upper curves
represent loadings from poorly-maintained areas based on a uniform daily rate
of pollutant accumulation which decreases with prolonged dry periods similar
to the rates reported (1,17). The loadings for urban land uses were plotted
separately for impervious and pervious areas. It should be remembered that
the loading from the impervious areas was estimated assuming an atmospheric
fallout rate of 0.8 Tonnes/km2/day and curb litter loadings similar to those
obtained by Sartor and Boyd (1) and Sartor et al. (17). If significantly
different accumulation rates are anticipated the loadings from impervious
areas should be adjusted accordingly to reflect the change in curb loading
rate due to increased or decreased atmospheric fallout.
Since impervious urban areas were simulated for an average year and the
loadings appear to have no correlation with rainfall intensity, the average
loading values can be read directly from the diagram and values are presented
in Table 11-15. In order to obtain average loadings for pervious areas, the
loading diagram related to the R-factor must be transformed to a probability
distribution loading plot using the cumulative frequency chart of the R-factor
as given in Fig. 11-10. The area under the R-factor-probability curve can be
graphically or numerically integrated according to the equation:
L.p.dp Eq. (11)
I is the average loading, kg/ha
L. is the loading function
PJ is the assigned probability of L. being less or equal.
1
It also should be noted that the loading diagrams in Appendix II-B
reflect loadings from a 1 km2 area under slope category B (2 to 6%) for the
impervious urban areas and slope category C (6 to 12%) for pervious areas. To
transform these values to other slopes and areal units, the loadings
corresponding to pervious areas should be multiplied by slope or area
correction factors presented in Figs. 11-11, 11-12 and 11-13. It is clear
11-39
-------
B<
VO
O
IN
PQ
>-.
i-i
O
M
0)
4-J
CO
O
0)
P-
o
rH
0)
"d
C
w
en
3
C
to
i
i^n
cs
•s
3
O
<4-V
•K /-*
W CO
60 ^O
Pj O^
•H rH
CO
0 J-i
rH CO
0)
4-J >,
C
CO 0)
4-1 6C
3 CO
iH J-l
rH 0)
0 >
CX CO
•d c
0) cfl
CO (JO
rH C
3 -H
3 ^
•H 3
i — 1
1
M
(—1
rH
rQ
CO
H
C1
*5
60
,0
2
5?
T"
O
CO
TJ
^_j
0
01
• cfl
o, x:
en •—,
m ,^"
QJ
iH
•H
Tt
O
x;
00
„
•y
OJ
H
-a
C/l
a
g
E
3
L^4
1-4
g
en
ftO
^
a.
1-4
QJ
•H
•^
^
[2
OJ
cn
CO
-H
a.
en
OJ
C
•H
rH
tt-
QJ
§
LO
00
p.
D.
CO
QJ
C
H
^
rH
crj
V-i
d
00
c
a.
V-i
o>
c
•H
^
£-S
i
(U
u
c
CO
c
01
*£
3
B
•a
c
cO
to
•H
0
tn
-r ^ sr S
rvi o - t 0 r^ IH
o o o o do
r^ s.
LJ~1 i ~, U~ L/ , r*^ C
"1 0 CM - - -,
d d o d -id
\& vD
c^i 0 M 0 £ £
O O "I O rH O
to in n
b o r1 0 en (Ni
d f., d o d d
$ in 2 ~ ^ SH;
4J
01 C
O OJ O £N'
0 CN* ^ Cs! r-; £ rH
'
"^
O O O LA o O
1A O ol cl O CN
O 1-1 O O -~, O
u"l vO c-i
•? 3
O -H lj cO
o at rrj 01
P-, P. S j-i
IdOlrHOIoCUnai QJ
oco'o'rarHcorHcTt •n'S
D. S -H -H nJ C
•a •ao'UOTj iJ-H
mc«"c c fi HJJ
O4JOW.O4-IXIJ-' CQ
W-Hm-5
-------
8.0 _,
6.0-
o
•u
o
-------
ion -i
o
i 1
I
ol
o
-H
4-1
e)
0)
M
Vi
O
O
01
CX
o
T-H
to
10 „
1 H
0.1 -
o.or
10
15
20
Slope, %
I
25
Fig. 11-12. Loading multiplier for different slope categories
(for use with Table 11-16).
11-42
-------
6C
C
•H
"d
cfl
O
O
4J
c
1
•H
TD
QJ
01
O
4-J
rt
QJ
M
Ol
0)
x;
4-1
CU
N
•H
. O
a)
4J
K-l
O
cx
•H
C!
O
•H
4J
03
rH
0)
~
O
O
O
00
•H
11-43
-------
that erosion potential of soils in the slope category D (12 to 20%) is about
20 times greater than that for soils in slope category A (0 to 2%).
Table 11-16 shows the average potential loading values for typical
pervious land (non-urban) uses situated on the four hydrologic soil groups.
The loadings for each land use, soil and season are long-term average
simulation results.
It is seen from Tables 11-15 and 11-16 that developing urban, industrial,
commercial and high density residential land uses with poor maintenance and
street cleaning practices, produce the highest potential loadings in urban
areas while low density residential and park and recreation land uses
contribute the least. For non-urban land uses, livestock feedlots are
expected to have the highest pollution potential and woodlands the lowest.
However, simulated loadings for feedlots may be unrealistic and are not
reported because of the impossibility of arriving at reasonable values for the
soil erodibility factor, K, due to the unusually high organic matter content
and unknown compactness of feedlot soils.
Differences between the pollution potentials for various land uses
indicate that pollution control measures should be concentrated intensively on
hazardous land uses; i.e., developing and high density residential areas,
unprotected non-urban areas located on soils with low permeability and steep
slopes and feedlots. Discussion of remedial measures are given in Appendix
II-C.
Comparison of Measured Loadings with Estimates
Obtained by the MEUL Method
One purpose of the Menomonee River pilot project was to establish
loadings from various land use activities. Although at the conclusion of the
research it can be stated that the loadings should be related to various
causative factors such as imperviousness of the area, type and slope of the
soils, vegetative factors etc., some of these factors may indeed be related to
land use. For example, the imperviousness of the area which is one of the
primary factors defining residential land uses can be correlated with housing
density. However, it must be realized that great loading variations should be
expected within one particular land use based upon soil type and slope
category, atmospheric fallout and litter accumulation and type of activities
taking place in the area. This is especially true for such land uses as low
density residential where most of the loadings originate from pervious areas
thereby involving soil type and slope as principal causative factors.
Furthermore, commercial and industrial land categories seem to be too broadly-
defined and need further subcategorization (e.g., type of industry or type of
commercial activities, degree of imperviousness).
Another problem which can arise when comparing estimated and measured
loadings is that each season has a different erosion potential. This is shown
in Fig. 11-10 where cumulative rainfall energy factors defined by the USLE
were arranged on a probabilistic scale of seasons. More than one order of
magnitude of sediment loss can be expected based on whether the season is dry
or has a significant number of high intensity storms. The measured values
11-44
-------
Table 11-16. Simulated pollutant loadings for land uses on essentially
pervious areas
Soil and
slope*
Sediment, kg/ha POt-P
Spring
Summer Fall
Spring
Park and Recreation — SC+ - 0
BMA
BMB
BMC
HfIA
HMB
HtlC
OUA
OUB
OUC
OUD
ASA
ASB
ASC
BMA
BMB
BMC
HMA
HMB
HMC
OUA
OUB
OUC
OUD
ASA
ASB
ASC
BMA
BMB
BMC
HMA
HMB
HMC
OUA
OUB
OUC
OUD
ASA
ASB
ASC
BMA
BMB
BMC
HMA
HUB
HMC
OUA
OUB
OUC
OUD
ASA
ASB
ASC
18
44
120
30
94
275
55
172
501
1,290
61
184
532
a
1.5
14
Summer
Fa 1 1
Pasture — SC - 0.03
0.02
0.03
0.07
0.03
0.06
0.25
0.05
0.09
0.38
1.07
0.08
0.15
0.68
25
102
330
60
252
795
134
487
1,470
3,830
152
522
1,560
54
178
543
142
466
1,420
206
690
2,060
5,300
330
1,000
3,000
21
47
216
48
107
492
60
135
620
1,770
62
140
645
0.02
0.10
0.33
0.09
0.36
1.19
0.23
0.87
2.65
6.8?
0.4/
1.60
4.85
0.05
0.17
0.54
0.21
0.68
2.]2
0.37
1.22
1.71
9.53
1.03
3.11
9.3C
0.02
0 . 1 1 5
0.22
0.117
0.16
0.73
0.11
0. 24
1. 11
3.1*1
0.11
n.4 1
1.99
Wetland— SC » 0.03
<0.001
0.001
0.035
<0.001
0.003
0.012
<0.001
0.011
0.270
2.34
0.009
0.098
1.35
<0.001
<0.001
0.010
<0.001
<0.001
0.027
<0.001
<0.001
0.059
0.52
<0.001
0.007
0.16
26
97
AA
69
256
AA
119
441
AA
AA
140
519
AA
45
144
AA
124
395
AA
248
655
AA
AA
350
1,090
AA
= 1.0 or 0.08
•C0.01
0.30
2.8
<0.01
0.98
8.25
<0.01
3.00
30.6
505
<0.01
4.39
57.9
1.0
1.82
5.89
14,4
7.33
19.2
56.4
29.5
75.6
221
511
21.0
54.8
161
<0.01
0.02
0.56
<0.01
0.05
1.92
0.02
0.18
4.31
37.5
0.14
1.56
16.9
2.97
6.48
17.4
10.8
23.6
63.8
65.5
142
385
882
52.1
117
317
<0.01
<0.01
0.15
•CO. 01
<0.01
0.44
<0.01
<0.01
0.94
8.28
<0.01
0.11
2.50
0.90
2.71
10.9
3.39
10.2
40.7
10.8
32.4
129
360
8.53
25.6
102
830
3,400
11,000
2,000
8,400
26,500
4,500
16,200
49,100
128,000
5,100
17,400
52,200
1,800
5,900
18,100
4,700
15,500
47,200
6,900
23,000
68,700
177,000
11,000
33,500
100,000
4
12
AA
11
34
AA
19
58
AA
AA
25
80
AA
Developing
700
1,600
7,200
1,600
3,600
16,400
2.00C
4,500
20,700
59,000
2,100
4,700
21,500
0.03
0.10
AA
0.10
0.38
AA
0.21
0.79
AA
AA
0.43
1.61
AA
Urban — SC
0.83
3.40
11.0
3.00
12.6
39.7
8.10
29.2
88.4
229
15.8
54.0
161
0.05
0.14
AA
0.19
O.'i9
A,
0.45
1.18
AA
AA
1.09
1.37
AA
- 1.0
1.80
5.9r
18.1
7.05
23.3
71.0
12.4
41.4
123
34.1
104
310
•CO. 001
0. "1
AA
0.02
0.05
AA
0.03
0.11
AA
AA
0.08
0.25
AA
0. 70
1.60
7.20
2.^0
5.40
24.6
3.60
H. . r
37.1
1(6
6.53
14.6
66.7
*BM is Boyer Is, KM is Hochhelm 1, OU is Ozaukee sil, and AS is Ashkum slcl; A is 0 to 2%, B is 2 to 6%, C Is 6 to 127. and D
Is 12 to 20% slope.
**Not applicable.
+SC is the cropping factor used in USLE.
11-45
-------
should he adjusted according to the R-factor for pervious areas to reflect I lie
average meteorological conditions on which the MEUL method is based.
The following correction factors based on Fig. 11-10 should be applied to
sediment loadings from pervious areas.
Season Erosion Correction Factor
Spring 1975 0.44
Slimmer 19 7 r. 4.00
Fall 1975 1.25
Spring 1976 0.31
Summer 197b 2.3
Fall 1976 5.0
Spring 1077 1.0
Summer 1977 0.66
The loading values must be further adjusted by the delivery ratio (DR)
relating loadings at the watershed outlet to those potentially liberated frori
the source area. The DR is still an unknown quantity which includes such
factors as sedimentation and resettling during overland and channel flow,
flocculation and agglomeration of suspended particles and removal of
pollutants by infiltration during overland flow. An inaccurate method of DR
estimation relates DR to the area! size of the watershed as shown in Fig. II-
14. Although the method is inaccurate it is as good as any other available.
Another factor which must be included is type of drainage. Natural drainage
systems with low or no curbs will yield low delivery ratios approximately
proportional to the fraction of impervious (e.g., storm sewer) and pervious
drainage ditches. Areas with no curbs may show loadings reduced as much as
50% or more as compared to typical urban landscapes of impervious areas (i.e.,
streets draining into impervious drainage gutters). The loading figures
presented in this report are based on the assumption that most of the street
pollutants will accumulate near the. curb.
Tables 11-17 and 11-18 present a comparison of measured and estimated
sediment and phosphate-P loadings for some major pilot subwatersheds and for
areas in a predominantly single land use in the Menononee River Watershed. In
almost all cases the estimated values were higher than the measured ones, a
fact partially attributable to assigning a DR-value. For most of the
simulated land uses the DR (ratio of measured:estimated loadings) is within
the ranges indicated in Fig. 11-14. The measured loadings for the fall
seasons were low and do not conform to estimated values. It should be noted
that Fall 1975 and 1976 seasons were very dry with minimal runoff.
It can also be expected that DR for highly impervious areas will be
higher than for largely pervious areas of the same size and DR will be higher
in sewered than in unsewered areas with natural drainage ditches.
Simulated unit loadings agree fairly well with measured values under
similar meteorological conditions and land use characteristics. An exception
has been noted for livestock feedlots where it was impossible to arrive at
reasonable values of the soil erodibility factor, K, due to unusually high
organic content of feedlot soils and unknown degree of compactness. Available
measured loading values from feedlots (28,29) deviate significantly from
simulated ranges; however, more research is necessary to obtain more realistic
data.
11-46
-------
O
•H
•u
CD
p"l M
!-( O
(U -H
> to
•H O
rH >-l
(1) at
Q
•u o
C
•H
13
-------
Table 11-17.
Comparison of simulated and measured sediment and phosphate
loadings in subwatersheds with mixed land uses (measured
loadings are taken from (26))
Land Use
Area, %
Impervious Sedim
areas, % Spring
entj kg/ha
Summer
POk-P, kg/ha
Fall
Spring
Summer
Fall
Donges Bay Rd. (463001), 2144 ha
Commercial
High density residential
Medium density residential
Low density residential
Row crops
Contributing
Pasture A
Pasture B
Wetlands
Feedlots
Developing
Estimated mean
Measured, arithmetic mean
weighted mean
Delivery ratio, weighted
Industrial
Commercial
High density residential
Medium density residential
Low density residential
Park and recreation A
Woodlands A
Developing A
Landfill A
Water
Estimated mean
Measured, arithmetic mean
weighted mean
Delivery ratio, weighted
Industrial
Commercial
High density residential
Medium density residential
Low density residential
Developing A
Row crops
Parks and recreation A
Woodlands
Wetland
Landfill
Estimated mean
Measured, arithmetic mean
weighted mean
Delivery ratio, weighted
Commercial
High density residential
Medium density residential
Low density residential
Developing A
Parks and recreation
Estimated mean
Measured, arithmetic mean
weighted mean
Delivery ratio
2.6
0.05
3.9
4.7
74
32
5
5
2.3
0.5
1.6
1.8
35
3.8
15.8
14.6
23
-
2.7
2.7
0.3
0.9
27.9
3.3
24.2
15.6
1.8
0.07
18.6
0.6
0.3
0.5
26.6
0.5
39.1
27.2
3.0
9.0
200
400
200
120
1,655
134
487
119
2,100
2,800
597
304
107
0.18
Noyes Creek** (413011), 552
60 880(80)***
60 700(100)
70 730(130)
40 '160(160)
10 180(160)
2 55
-
2 3,000
2 3,000
35 547
840
i66
1.0
Honey Creek (413006), 2,803
855(55)
655(55)
655(55)
255(55)
75(55)
3,500
-
55
-
-
2,500
368
417
294
0.80
Schoonmaker Creek+H~f (413010),
90 350(50)
90 350(50)
60 200(50)
25 200(170)
1.6 1,500
5.0 27
54 277
157
120
0.43
400
800
100
250
100
206
690
248
4,525
4,150
212
39
62
0.29
ha
1,020(220)
650(150)
820(170)
470(270)
290(270)
64
-
6,600
6,600
762
389
566
0.74
ha
bt.4(b<4)
3^4(6'. >
714(6'.)
264(641
84(6..)
7,000
64
-
-
5,500
'23
223
287
0.68
179 ha
500(5C)
800(50)
600(200)
280(200)
2,300
32
531
147
210
0.40
200
60
150
50
10
60
135
18
750*
1,200*
50
460(60)
250(50)
350(50)
140(60)
70(60)
30
-
1,260*
1,260*
155
136"H'
153
O.r>9
460(6.').)
2iO(iO,
350(50)
115(35)
4srss)
1,200
-
30
-
-
1,050**
>58
28
41
0.16
190(10)
210(30)
160(60)
75(60)
660*
15*
120
33
45
0.38
0.30
0.60
0.50
0.35
3.0
0.23
0.87
0.21
5.90
5.67
1.18
0.61
0.20
0.17
0.70
0.50
0.60
0.45
0.30
0.09
-
4.1
4.1
0.56
0.61
0.70
0.80
0.80
0.50
0.18
0.37
1.22
0.45
12 ,89
7.45
0.4T
0.07
0.06
0.15
1.1
0.60
0.80
0.60
0.45
0.13
_
6.2
6.2
0.78
0.36
P. 25
0.70
0.40
0.10
0.00
0.11
0.24
0.03
2.0
2. 5
0.09
0.30
O.lb
0.16
0.15
0.12
0.09
_
3.6
3.6
0.32
0.01
*Corrected for the area used.
**No cleaning in spring, medium maintenance in summer and fall.
***( ) amount contributed by pervious areas.
+Assume that 50% originated fromm pervious areas.
-H-Data for Fall 1976 excluded due to unusually dry weather.
+++Assume good cleaning.
11-48
------- An error occurred while trying to OCR this image.
-------
"REFERENCES-Il
1. Sartor, .1, 0. and G. B. Boyd. Water Pollution Aspects of Street Surface
Contaminants. U.S. Environmental Protection Agency Report No. EPA-R2-72-
081, Washington, D.C., 1<>72.
2. Wischmeier, W. H. and D. D. Snith. Predicting Rainfall-Erosion Losses
from Cropland East of the Rocky Mountains. U.S. Dept. of Agriculture
Handbook 282. Washington, D.C., 1965, 47 pp.
3. Wischmeier, W, II., C. B. .Johnson and B. V. Cross. A Soil Credibility
Nomograph for Farmland and Construction Sites. J. Soil and Water
Conserv. 26(5):189-193, 1971.
4. Midwest Research Institute. Loading Functions for Assessment of Water
Pollution from Non-point Sources. U.S. Environmental Protection Agency
Report No. EPA-600/2-76/151, Washington, D.C., 1976.
5. Novotny, V., II. A. Chin and H. Iran. Description and Calibration of a
Pollutant Loading Model-LANDRUN. Part 1: Description of the Model.
Final Report of the llenomonee River Pilot Watershed Study, Vol. 4, U.S.
Environmental Protection Agency, 1979.
6. Novotny, V., H. Iran, G. Siinsiman and G. Chesters. Mathematical
Modelling of Land Runoff Contaminated by Phosphorus. J. Water Pollution
Control Fed. 50(1):101-112, 1978.
7. Novotny, V., M. A. Chin and H. Tran. Description and Calibration of a
Pollutant Loading Model-LANDRUN. Part II: Calibration and Verification
of the Model. Final Report of the Menoirionee River Pilot Watershed Study,
Vol. 4, U.S. Environmental Protection Agency, 1979.
8. Walesh, S. G. Land Use, Population and Physical Characteristics of the
Menomonee River Watershed. Part I: Land Data Management System. Final
Report of the Menomonee River Pilot Watershed Study, Vol. 2, U.S.
Environmental Protection Agency, 1979.
9. American Public Works Association. Water Pollution Aspect of Urban
Runoff. Water Pollution Control Research Journal WP-20-15, Washington,
D.C., 1969.
10. Hiemstra, L. Frequencies of Runoff for Small Basins. Ph.D. Thesis,
Colorado State University, Fort Collins, Colorado, 1968. 151 pp.
11. Simsiman, G. V., J. Goodrich-Mahoney, G. Chesters and R. Bannerman. Land
Use, Population and Physical Characteristics of the Menomonee River
Watershed. Part III: Description of the Watershed. Final Report of the
Menomonnee River Pilot Watershed Study, Vol. 2, U.S. Environmental
Protection Agency, 1979.
11-50
-------
12. Hydrocomp International. Hydrocomp Simulation Programming Operation
Manual. Hydrocomp International, Palo Alto, California, 1972.
13. Horn, M. E. Estimating Soil Permeability Rates. J. Irrigation and
Drainage Div. Proc., Amer. Soc. Civil Engineers 97(IR2):263-274, 1971.
14. Bouma, J. , W. A. Ziebell, W. G. Walker, D. G. Olcott, E. McCoy and F. D.
Hole. Soil Absorption of Septic Tank Effluents. University of Wisconsin
Extension, Geological and Natural History Survey Information Circular No.
20, Madison, Wisconsin, 1972.
15. Brandt, G. H., E. S. Conyers, 11 B. Ettinger, F. J. Lowes, J. W. rtighton
and J. W. Pollack. An Economic Analysis of Erosion and Sediment Control
Methods for Watersheds Undergoing Urbanization. Final Report, OWRR.
Contract No. 14-31-001-3392, Dow Chemical Co., Midland, Michigan, 1972.
16. Shacklette, H. T., J. C. Hamilton, J. G. Boernagen and J. Tl. Bowles.
Elemental Composition of Surficial Materials in the Conterminous United
States. U.S. Geological Survey Proc. Paper 574-P, Washington, B.C.,
1971.
17. Sartor, J. D. , G. B. Boyd and F. J. Agardy. Water Pollution Aspects of
Street Surface Contaminants. J. Water Pollution Control Fed. 46(3):458-
467, 1974.
18. Graham, D. H., L. S. Costello and H. J. Mallon. Estimation of
Imperviousness and Specific Curb Length for Forecasting Storm Quality and
Quantity. J. Water Pollution Control Fed. 46(4):717-725, 1974.
19. Pitt, R. and G. Amy. Toxic Material Analysis of Street Surface
Contaminants. U.S. Environmental Protection Agency Rep. No. EPA-R2-73-
283, Washington, D.C., 1973.
20. Milwaukee County Department of Air Pollution. Ambient Air Quality
(Particulates and Sulfur Dioxides) in Milwaukee County. Milwaukee County
Department of Air Pollution, Milwaukee, Wisconsin, 1970.
21. Beaseley, R. B. Erosion and Sediment Pollution Control. The Iowa State
University Press, Ames, Iowa, 1972.
22. Heaney, J. P. and W. C. Huber. Storm Water Management Model:
Refinements, Testing and Decision-making. Department of Environmental
Engineering Sciences, University of Florida, Gainesville, Florida, 1973.
23. Carlisle, A. A., H. F. Brown and E. J. White. Litter Fall Leaf
Production and the Effects of Defoliation by Tortrix viridanal in a
Sissile Oak (Quercus petrala). Woodland J. Ecology 54:65-98, 1966.
24. Lutz, H. J. and R. I. Chandler. Forest Soils. John Wiley and Sons, New
York, N.Y., 1976.
25. U.S. Army Corps of Engineers. Urban Storm Water Runoff Model STORM. The
Hydrologic Engineering Center, U.S. Army Corps of Engineers, Davis,
California, 1975.
11-51
-------
26, Konrad, J. G, and G. Chesters. llenoraonee River Pilot Watershed f'tud.v;
Summary Pilot Watershed Report. Submitted to PUJARG Task Group C (U.S.),
Activity 2. Windsor, Ontario, flay 1978. 77 pp.
27, Roehl, J. W. Sediment Source Areas, Delivery Ratios and Influencing
Morphological Factors. J.A.S.H. Commission on Land Erosion Pub!. "o. 59,
1962.
28. Daniel, T, C., W. Wendt and P. K. Mc.Guire. Pollutant Loadings fron
Selected Rural Land Uses. Trans. Amer. Soc. A;;. Hng. (subnitted for
publication), 1979.
29. Coote, n. R. and F. R. Hore. Pollution Potential of Cattle Fcedlots and
Manure Storages in the Canadian Great Lakes Basin. Final Report
Agricultural Watershed Studies Project 21. Submitted to PLUARG, Windsor,
Ontario, 1978,
H-52
-------
APPENDIX II-A
DETAILED STATISTICAL EVALUATION OF STREET LITTER ACCUMULATION
It has been realized that a simple unit loading value may not be
representative of the surface pollution accumulation process. Instead, a mass
balance model can be developed which may better represent the dynamic
character of the street refuse accumulation. The model is based on the
following simple mass balance equation (see Fig. II-7 for more detail):
£ - I* - I* Hq. (A-l)
L is the polllutant accumulation on the surface, g/curb m/day
LQ is the pollutant deposition rate, g/curb m/day
Ln is the pollutant removal rate from the surface, g/curb m/day
The simple mass balance equation presented above can be expanded by
identifying the significant factors which affect deposition and removal from
street surfaces. The primary sources can be related to fallout of atmospheric
pollutants, motor vehicle usage and deposition of street litter.
Traffic can contribute significantly to pollutant deposition in urban
areas. Large amounts of toxic metals in storm water runoff are often
attributed to motor vehicle emissions and to the breakdown of road surface
materials and vehicle parts.
The variables affecting the pollutant deposition rate on impervious urban
areas can be combined to yield the following equation:
LD = (ATFL) (SW/2) + ^ A (SW/2) (POA) + AZ(RD) + AS(TD) (RCC)
Eq. (A-2)
where
ATFL is a coefficient reflecting deposition from stationary
combustion processes and atmospheric fallout, g/ha/day
SW is the street width, m
Aj^ is a coefficient reflecting the effect of open areas on
pollutant deposition
POA is % open area in the vicinity of the site
A is a coefficient reflecting the effect of residential
density on pollutant accumulation
RD is the residential density, dwelling units/ha
11-53
-------
A is a coefficient reflecting the effect of traffic OP
pollutant accumulation
TD is traffic density, thousand axles/day
RCC is road composition and conditions which is a value based on
scale determined from regression analysis
At the same time that pollutants are being deposited on the surface they
are being removed. Factors which should be investigated as affecting the
removal rate include wind speed, traffic speed, and curb and average height of
buildings. The equation for street surface refuse removal can be formulated
as:
f?(WS,TS)]L i:q. (A-3)
where
A is a coefficient reflect ing the rate of pollutant removal due
to the combined effect of wind and traffic speed
II is curb height, cm
WS is average wind speed, kni/hr
TS is average traffic speed, km/hr
The function f (II), describes the effect of curb height on pollutant
removal and can be modeled as:
fjOO = e~3H r.a. (A-4)
where 3 is a statistical coefficient.
The above model was applied to a set of field data. Since the Menomonee
River Watershed data do not yet provide a representative data sample, the data
sample was supplemented by field measurements of street refuse accumulation in
the Washington, D.C. area (A-l).
The solution to Eq. 12 will yield the following formula:
L = (1 - e~Bt) + C Eq. (A-5)
where
t is time from last street cleaning or rain
A and B are variables determined for each constituent
C is a constant
The Washington, D.C. data (A-l) , contain about 73 measurements on 7
different sites. Although the number of sites is probably too low to provide
a sufficient spread of independent variables the statistical analysis did
provide some answers as to the significance of the variables involved.
The best fit equations for four typical constituents, i.e., which were
statistically significant are as follows:
11-54
-------
Dust and dirt suspended solids -
DDSS =4(1 ~ e"Bt) + C i:q. (A-6)
D
A = ATFLO) - 5.02(RD) - 6.29(POA) + 1.15(TD)
B = 0.0116e-°-°88H (TS + WS)
C = 0.0
Multiple correlation coefficient R = 0.86
Similarly:
Dust and dirt chemical oxygen demand -
A — Rt"
DDCOD = |(l-e ) + C Eq. (A-7)
CtJ
A = 2.60(-^-) - 0.28(RD - 0.51(POA) + 0.52(TD)
B = 0.142e~°'98H (TS + WS)
C = 0
Multiple correlation coefficient R = 0.71
Dust and dirt volatile suspended solids -
AI _R j. A« A. _
DDVSS =^-(1 - e ^r) -^(1 - e B2n) +^-(1 - e "s^ + C Eq. (A-8)
Bl B2 B3
= 0.024 e~°-05H (TS + WS)
= 0.25(RD) + 0.31(POA)
B2 = 0.048 e~°'°5H (TS + WS)
A3 = 0.069(TD)
11-55
-------
= 0.105 e °'°SH (TS + WS)
C = 0
Multiple correlation coefficient R = 0.65
Dust and dirt lead -
A A A
DD Lead = ~(1 - e~Blfc ) - ---(1 - c~V ) + -1(1 - e~V > + C
Rl B2 R3
A, - O.U,
= 0.036 e °*°3H (TS + WS)
= 0.027(RD)
0.026 e °'°3H (TS + WS)
A3 = 0.013 (TD)
= 0.053 e °'°3H (TS + WS)
C = -0.825
Multiple correlation coefficient R = 0.80
Table II-A-1 lists the partial correlation coefficients for the above
variables. From the table it can be seen that in all four cases the overall
functional relationship is at a significant level. The dependent variables
which have the most significant effect on the independent variables vary with
the character of the variables. As might be expected, traffic density may
have a very significant effect on the magnitude of the accumulation of dust
and dirt constituents, particularly lead. On initial inspection it may seem
surprising that the regression coefficients have a negative value for POA and
RD. One would expect that quantity of street refuse would increase with
increasing housing density or open area (i.e., area without significant
vegetation). On the other hand, just the opposite can be true if one realizes
that a significant portion of street refuse originates from vegetation — lawns,
trees and shrubs — which are inversely proportional to housing density (RD) or
open area (POA). Thus, it seems that trees and vegetation near impervious
areas may contribute significantly (especially during the fall season) to
pollutant loading.
The above equations represent the best combination of variables which
H-56
-------
4-1
\-t
•H
X>
XI
C
cfl
4-J
CO
3
X5
d
cu
CU
s
4-1
CU
XI
CO
4-1
C
CU
•H
CJ
•r-(
4-1
*nW
o»
0
CJ
C
0
•H
4-1
ca
rH
CU
SJ
i-l
O
O
CU
rH
CX
•H
4J
rH
3
0
13
C
cfl
rH
ca
•rl
4-1
>-l
CB
PH
•
rH
1
<
1
M
M
CU
rH
f>
ca
H
C
o
•rl
4-1
Cfl
rH
3
S
3
O
U
ca
P
•rH
CU
J3
4-1
60
C
•H
4-J
O
CU
m
m
cfl
u;
S-l
0
4J
U
ce
4H
t3
a
CO
CO
4J
0
cfl
4-1
3
i-H
rH
O
cx
Pi
cu
rH
P.
•iH
4-)
i-H
3
^
•K
CU
rH
£>
Cfl
•H
^
cfl
>
4-J
0
0)
XI
C
CU
cx
0)
XI
HH
0
S-i
rH
CO
•r-t
4-1
IJ
C3
Q-,
Q
H
cu
rH
,0
Cfl
•rl
V-I
Cfl
>
4-1
C
cu
xt
C
cu
cx
CU
xt
C
M
vO
oo
•
o
rH
•
O
1
\£>
CN
•
O
Q
O
CJ
m
v£>
•
0
o
CN
•
O
O
CN
•
O
1
01
CN
•
O
1
m
rH
•
0
CO
X)
•H
i— 1
O
CO
•a
cu
X)
C
cu
ex
CO
3
01
CU
rH
•H
4-1
Cfl
r-l
O
>
O
oo
•
0
o
%
4J
•H
CO
C
CU
"a
rH
CU
•H
4J
C
CU
X!
•r-l
CO
CU
X-l
to
•H
a
Pi
A
.c
4-J
X)
•iH
s
4-1
cu
CU
!H
4J
CO
CD
•H
3
CO
*
X!
C
•H
>>
.5
CO
•H
to
-a
C
CO
xi
01
01
ex
CO
o
•H
UH
M i
Cfl
VJ
4-J
CO
•H
CO
H
*
-M
I-C
60
•r-l
CU
^
^3
S-l
3
CJ
CO
•H
33
«\
>•,
4-J
•H
CO
C
cu
•a
CJ
•rl
IH
U-l
ca
M
4-1
CO
•H
Q
H
•
-a
cu
CU
Cu
ca
11-57
-------
were investigated. Other combinations which yielded lower statistical
correlations included the effect of traffic speed on pollutant accumulation
(as in the form of TD x TS or TD x T.S2), excluding some insignificant
variables and others.
Equations (A-6) to (A-9) indicate that as the quantity of deposited
pollutants increases with prolonged dry periods, more particles can be removed
by wind and traffic and the actual differential deposition rate decreases.
This fact was also observed by Sartor et al. (A-2) and is documented in
Fig. II-A-1.
11-58
-------
800
O
Mi
4-1
•H
4J
d
cd
O'
-------
REFCRf,rCLS-\PPLi;DIX I i -A
A-l. Shaheen, p. Contribution of Roadway Usa^e to Water Pollution, I'.S.
environmental Protection Agency Report No. EPA. 600/2-75-004, Washington,
D.C., i975.
A-2. Sartor, J. D. and G. R. Royd. Water Pollution Aspects of Street Surface
Contaminants. U.S. Environmental Protection Agency Report No. EPA-R2-
72-081, Washington, D.C., 1972.
A-3. Sartor, J. D. . 0. B- Royd and F. J. Agarrly. Wafpr Pollution Aspects of
Street Surface Contaminants. J. Water Pollution Control Frd. 46(1):4">S-
467, 1074,
H-60
-------
APPENDIX H-B
SIMULATED LOADING DIAGRAMS
Loadings for impervious urban land uses (Figs. II-B-1 to II-B-6) reflect
values from areas under slope category B (2 to 6%). Average loadings can be
read directly from the loading diagrams. Loading diagrams for volatile
suspended solids and Pb are available but are not presented in this report.
Loadings from pervious areas shown in Figs. II-B-7 to II-B-16 reflect
values from a 1 km2 area under soil slope category C (6 to 12%). To obtain
average loadings, loading diagram related to the R-factor must be transformed
to a probability distribution loading plot using the cumulative frequency
chart in Fig. 11-10. The cropping factor, SC, on all loading diagrams is
0.01. To obtain loadings for each land use with SC other than 0.01 multiply
the values from the graph by 100 and SC factors in Table 11-16. To transform
loadings to other slopes and areal units, values should be multiplied by slope
or area correction factors presented in Figs. 11-11 to 11-13. Loading
diagrams for phosphate-P are available but are not given in this report.
11-61
-------
en
cfl
ca
•H
4-1
c
0)
t3
•H
W
0)
e
o
to
Mi
C
•H
T)
cfl
O
•H
13
0)
w
I
60
•H
11-62
-------
en
cd
QJ
J-i
CO
•H
o
t-l
0)
e
o
o
B
o
0)
•H
•d
0)
CO
CNl
I
PQ
I
6C
•H
11-63
-------
0)
s-i
to
•H
(-4
4-J
05
3
13
C
•H
O
W
00
c
•H
T3
CO
O
4-1
C
13
Q)
CO
ro
PQ
I
6C
•H
II-64
-------
en
n)
0)
•H
4-1
c
0)
T3
•H
w
a)
s
o
en
oo
c
•H
T3
n)
o
P-I
I
0)
p.
en
o
toC
•rl
11-65
-------
co
n)
0)
cfl
•H
o
(J
CU
O
O
e
o
(-1
c
•H
•X3
"
P-i
I
0»
CO
o
LO
PQ
I
11-66
-------
T3
c
•H
e
o
S-4
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 T 1 1
bO
C
•H
T3
cfl
O
PM
I
a)
4-1
cd
a-
w
o
PL,
I
M
•H
ft.
11-67
-------
Ci
rt
i-t
•o
o
o
^!
"l
n,
c
'b u t p e o i } u 3 i
tfl
U-l
I
CO
60
•rl
T)
01
•H
T3
0)
W
w
C
O
I
M
I
&0
•H
H-68
-------
Cfl
n)
01
S-i
re
c
ra
r-i
-a
o
§
i
a.
o
o
c
c
•H
'6 u i p e o ~
u a ui i p 3
•H
T3
0)
W
C
O
•H
•H
(-1
Cfl
•H
•H
t-H
•H
XI
cti
X)
O
»-i
a,
00
I
M
I
M
•H
11-69
-------
c
•H
J-i
C
•H
•a
•H
-a
a/
CO
CO
c
o
•H
4-J
n)
Ml
•H
11-70
-------
CO
4J
0
c
•H
W
M
C
•H
T3
tfl
O
0)
CO
O
•H
J_l
3
,0
•H
)-i
•U
CO
•r-t
•H
r-t
•H
Cfl
O
(-4
I
pa
i
11-71
-------
cn
0,1
w
ID
G.
V-l
I
cfl
6C
C
•H
T3
0)
co
•H
.C
CO
C
o
•H
5 u 3 uj t p a s
i
pq
60
•H
11-72
-------
W
01
>-l
3
4-1
tn
en
D-
c
•H
(0
6C
C
•H
T3
C
0)
e
•H
-a
0)
03
"4n
O
C
O
•H
4-1
3
S-i
4-1
03
•H
-d
>>
4J
•H
rH
•H
,£>
rt
I
pa
i
11-73
-------
en
13
C
cfl
(-1
o
iJ
a
cfl
M-l
•a
c
60
C
•H
t)
n)
o
c!
OJ
•H
T3
01
03
Cu
•H
CO
C
O
•H
4-J
«
I—I
-------
(T.
T3
q
CU
3
C
•rt
cn
oc
C
•H
'6 u L p p 0 -] 3 u a 111 i p a
- - J> O.
QJ
e
•rl
"O
0)
CO
G
O
•H
,
•H
S-i
CO
•H
•H
iH
•H
43
tfl
43
O
-------
Il
REMEDIAL MEASURES AND NON-POINT FOLUiTFON CONTROL
Remedial measures can be categori/pd using a macro or micro scale. The
former may result in better land use practices and zoning, legislation
limiting marketing certain potentially-hazardous pollutants or better farming
practices. These measures are usually long-terra remedies and take longer
periods of time to implement. Micro-scale remedial measures include better
management and control of existing land uses. In urban settings, limiting the
non-point pollution can take place either at the source (maintenance and
cleaning) or at the area outlet (storage and treatment). In non-urban
settings, the control is limited to better farming practices and erosion
control.
A literature review by thp Wisconsin Department of Natural Resources
(C-l) compiled and presented possible management practices to control water
quality of urban runoff. The control techniques mentioned included:
Source control
Increased infiltration
Retention of runoff
Reduction of erosion
Reduction of contaminant deposition
Street sweeping
Outfall treatment and collection control
Reduction in channel erosion
Infiltration and sedimentation basins
Storage basins to equalize flow
Physical, chemical and biological treatment
The study concluded that in low density urbanizing areas the quality of
stormwater runoff is most efficiently handled by systems incorporated into the
development stage such as zoning, control of developing areas, increased
perviousness and optimal design of stormwater conveyance systems. In high
density, developed areas, runoff is handled by good street cleaning practices
and through one of a series of treatment methods subsequent to collection.
Source control of urban-related pollution, which reduces on-site
pollutant generation or prevents pollutants from leaving the small drainage
areas at which a disturbance occurs, is less expensive and more effective than
remedial measures once the pollutants leave the site and move downstream.
Control of runoff pollution by collection systems is more expensive than on-
site source control but less costly than treatment at the outfall.
Treatment of urban runoff may be feasible only for highly developed areas
where source control and collection control are not possible.
11-76
-------
The difference between frequently cleaned and poorly maintained (no
cleaning) urban areas can be seen in Table 11-15. Although Table 11-15
represents simulated pollutant loadings the importance of street cleaning is
evident. Figures II-C-1 and II-C-2 show the simulated effect of street
cleaning frequency and efficiency on sediment loadings. The average
efficiency of street sweepers for the suspended particulate materials (dust)
is about 50% (C-2) but due to the fact that P is associated mostly with the
fine fractions of street dust and dirt the expected efficiency of P removal is
only about 22%. The effects of street sweeping are much higher during a dry
season and when a linear accumulation of street pollutants is assumed.
Other remedial measures include increasing pervious areas within urban
settings and reducing impervious areas directly connected to surface runoff
channels. Installing pervious parking areas, introducing seepage beds and
basins, and disconnecting roof drains from storm sewers can be listed as
possible examples. These measures can be ineffective if the area is located
on impermeable soils or on steep slopes since the conveyance of runoff from
the pervious area would create more erosion and pollutant washout from these
soils. Pervious areas should not be left bare. Permanent or temporal surface
protection, such as lawns, temporary seeding, or application of mulch or
chemicals should be practiced to control erosion and pollutant washout.
Street curbs and highway barriers represent obstacles at which surface
suspended pollutants (dust) can accumulate. Studies by Sartor et al. (C-2)
and Sartor and Boyd >'C-3) indicated that 90% of surface suspended pollutants
are located within 1 m of the curb. One would suspect that the curb hej^ht
can—to some degree-—affect the amount of pollutants accumulated. To provide
some insight i.ito the validity of this hypothesis, a sensitivity analysis of
Eq. (A-6) was performed (Fig. II-C-3). Thus, lower curb heights nay result in
less pollutant accumulation near the curb since some of the deposits can be
removed by wind and traffic and deposited in adjacent pervious areas where
they are less available for transport. Obviously, lowering curb sizes would
be effective only if the streets are surrounded by pervious areas.
11-77
-------
o
ex
60 1
a -H
•H 3
co e
a 3
,
o)
-o
6T
c
•H
a
m
a)
CO
o
e
cu
•H
o
iw
•H
a
cj
a)
w
o
c.
o
o
o
CO
I
o
I
o
00
C
•H
a
cu
0)
ra
o
-------
60
C
•H
O.
0)
a)
en
oc
C
•r-(
Cu
4)
dl
tn
6C
C
•p-<
T3
0)
O
O
ft,
00
C
-------
90 -
>,
cfl
id
I
42
U
4 7° -
C
o _
•rH
.u
cti
i-t
p
§ 50 _
o
o
O
V-i
a)
^j
4-1
•rj
t-3
^ 3° -
0)
OJ
n
w
1 II
/^~
/ Medium density residential
^r street width = 10 m
f percent open area = 40
residential density = 10 units/h
traffic density = 1000 ax/day
wind and traffic speed = 60 km/h
to 0 "i 1 1 1 1 , — 1 1 1
0 40 80
Curb (Medium Barrier) Height, cm
Fig. II-C-3.
Effect of curb (median barrier) height on street litter
accumulation.
11-80
-------
REFERENCES-APPENDIX II-C
C-l. Oberts, G. L. Water Quality Effects of Potential Urban Best Management
Practices. Dept. of Natural Resources Tech. Bull. No. 97, Madison,
Wisconsin, 1977.
C-2. Sartor, J. D., G. B. Boyd and F. J. Agardy. Water Pollution Aspects of
Street Surface Contaminants. J. Water Pollution Control Fed. 46(3):458-
467, 1974.
C-3. Sartor, J. D. and G. B. Boyd. Water Pollution Aspects of Street Surface
Contaminants. U.S. Environmental Protection Agency Report No. CPA-R2-72-
081, Washington, B.C., 1972.
11-81
-------
PART III
A SIMPLE/ EMPIRICAL MODEL FOR PREDICTING
RUNOFF QUALITY FROM SMALL WATERSHEDS
by
D, S, CKERKAUER
Ill-i
-------
ABSTRACT
A single model for calculating the time distribution of suspended solid
loads in a runoff event is presented. Instantaneous solids concentrations are
related to discharge per unit drainage area, rainfall intensity, antecedent
dry period, and stage of urban development. A set of empirical curves devel-
oped from observations on small watersheds within the Menomonee and Milwaukee
River watersheds allows calculation of suspended solids concentrations for any
percentage of urbanization. These concentrations can then be combined with
discharges predicted by some standard means to provide loading. The model
has been tested in watersheds from a variety of climatic, geologic and topo-
graphic regions. For storms within the calibration limits of the model, it
predicts loads with reasonable accuracy.
EIT-ii
-------
CONTENTS - PART III
Title Page , Ill-i
Abstract . IH-ii
Contents ... Ill-iii
Figures Ill-iv
Tables . ...... Ill-vi
III-l. Introduction „ III--1
III-2. Conclusions . III-2
III-3. Methods and Procedures ....... . III-6
III-4, Results and Discussion III-9
References 111-22
Ill-iii
-------
FfCURES
Number_ Page
III--1 Regression coefficients for model for total suspended
solids . , , IIT-S
III-2 Comparison of observed and predicted suspended solids
concentrations for Brown Deer Creek on 6/8/77. Total
precipitation was 1,32 cm, the antecedent dry period
was 3 days and rainfall intensity was 0.25 cm/hr . 111-10
III-3 Comparison of observed and predicted suspended solids
loads for Brown Deer Creek on 6/8/77. Total precipi-
tation was 1.32 cm, antecedent dry period was 3 days
and rainfall intensity was 0.25 cm/hr III-ll
III-4 Comparison of observed and predicted suspended solids loads
for Underwood Creek on 4/23/76. Total precipitation is
5.4 cm, antecedent dry period was 1 day and rainfall
intensity was 0.30 cm/hr 111-12
III-5 Comparison of observed and predicted suspended solids
loads for Event 32 at Third Fork Creek, Durham,
North Carolina. Total precipitation was 2 cm, the
antecedent dry period was 2.5 days and rainfall
intensity was 0.48 cm/hr 111-14
III-6 Comparison of observed and predicted suspended solids
loads for Event 29, at Third Fork Creek, Durham,
North Carolina. Total precipitation was 6 cm, the
antecedent dry period was 5 days and rainfall
intensity was 0.86 cm/hr 111-15
III-7 Comparison of observed and predicted suspended solids
loads for Event 27 at Third Fork Creek, Durham,
North Carolina. Total precipitation was 3.8 cm,
the antecedent dry period was 11.3 days and rainfall
intensity was 1.14 cm/hr 111-16
III-8 Comparison of observed and predicted suspended solids
loads for Bloody Run, Cincinnati, Ohio on 9/25/70.
Total precipitation was 1.7 cm, antecedent drjr period
was 1 day and rain fall intensity was 0.73 cm/hr 111-17
-------
Number PjLSJ?.
III-9 Comparison of observed and predicted suspended solids loads
for Bloody Run, Cincinnati, Ohio on 10/20/70. Total
precipitation was 2.3 cm, antecedent dry period was 6 days
and rainfall intensity was 0.45 cm/hr 111-18
111-10 Comparison of observed and predicted suspended solids loads for
Baker Street Basin, San Francisco, California on 11/5/69.
Loads have been predicted using the general model (x) and
also a modified model which reduces the importance of
antecedent conditions (o). Total precipitation was 1.6 cm,
antecedent dry period was 19 days and rainfall intensity
was 0.33 cm/hr III-L9
III-ll Comparison of observed and predicted suspended solids
concentrations and flow for Baker Street Basin,
San Francisco, California. Total rainfall was 1.6 cm,
antecedent dry period was 19 days and rainfall intensity
was 0.33 cm/hr 111-20
III-v
-------
TABLES
Number Page
III-l Comparisons of predictive capabilities of model for
suspended solids loads , III-3
III-2 Coefficients for final regression equations for various
degrees of urbanization . , III-8
Ill-vi
-------
III-l. INTRODUCTION
One of the objectives of the Menomonee River Pilot Watershed Project was
to synthesize the collected data into a form useful to planners and others
concerned with the effects of runoff quality from future urban development.
Models, calibrated with data gathered from the Menomonee River Study can be
extrapolated to project the effects of developemnt. The LANDRUN digital model
represents the primary modeling effort and like most available digital runoff
models for calibration, it requires detailed input of the hydraulics of the
Watershed and its channels. When precise inputs can be provided, the model
produces precise results. However, in many urban areas in the Great Lakes
Watershed, either the necessary input data is not available or time and
budget constraints do not allow development and/or calibration of a digital
model.
With these concerns in mind, a methodology is presented for development
of a simple empirical model for predicting runoff quality from small watersheds.
This model is less precise than LANDRUN in its final product, but it is one
which can be calibrated for a particular urban area with data which is easily
obtainable.
III-l
-------
III-2. CONCLUSIONS
Table III-l summarizes the investigation and provides a comparison of the
observed and predicted suspended solids loads for each event discussed. After
calibration in an area, the model is able to predict suspended solids loads to
about + 20%, It cannot be used on watersheds (such as Underwood Creek) which
are substantially larger than those used for calibration without introducing
a substantial error (Table III-l). In addition, the model is valid only for
the range of rainfall intensities and totals for which it is calibrated. It
would probably be advisable to calibrate it locally for small, intermediate
and large storms, but insufficient data has been analyzed to determine the
value of multiple calibrations.
Extrapolation of the model to areas of vastly different climatic, geo-
logic and topographic conditions produced surprisingly good results. Admit-
tedly, predicted solids loads were generally substantially different from the
observed ones (error range of 8 to 80%, Table III-l). However, within the
constraints of its calibration, the model was always within the proper order
of magnitude for watersheds and events that produced from 1,100 to 46,000 kg
suspended solids/km2. In addition, it cannot be determined from the published
watershed descriptions the extent of active construction in these areas. Such
construction is not accounted for in the model.
Two conclusions can be drawn from the apparent flexibility of this
statistical model. First, the regression coefficients developed for the
Menomonee River Watershed are valid for a wide range of conditions. Local
calibrations should be made to refine the coefficients for local conditions.
Secondly, it can be inferred that rainfall conditions (intensity and duration
of antecedent dry conditions), amount of runoff and degree of urbanization
are much more important in determining suspended solids in urban areas than
are such local conditions as topography, geology and vegetation. If this
were not the case, the regression information transferred from one area to
another would bear no relationship with reality.
Furthermore, it has been suggested that the model produces reasonably
accurate estimations of suspended solids loads after it has been calibrated
for local conditions. The principal value of the model is the ease with
which it can be calibrated. Runoff samples must be collected from a variety
of small streams for which the following is known:
a. Intensity and quantity of rainfall capable of producing runoff.
b. antecedent rainfall conditions,
c. discharge at the time of sample collection, and
d. land usage information for the sampled watersheds.
Multiple regression relations are then developed for suspended solids
concentrations and Items a. and c. for each stream. The regression coeffi-
cients are plotted as functions of urban development (Fig. III-l). The
III-2
-------
CO
ctf
o
en
•H
o
05
0)
G
0)
0,
0)
3
to
.
0
M-l
"cu
o
e
0
CO
CU
•H
•H
H
•H
__Q
Ct)
ft
oJ
0
OJ
•H
O
•H
CU
M
HH
0
CO
CO
4-1
C
QJ
O
0
J
•u
c
c
QJ
aj P--S
ii-t
UH
a
i -,-T
co
rd
C
-1
OJ CN
u ,2
•H ~--»
Ij] $
VH
TD
> e
GJ ---.
W OC
i— i
i — i
rt
•H
as
CO
QJ *
T3 -0
O O
O >H
>,
s
„
>1
4-i
•H in
CO -C
4-1 O
£
*
;-j
3 5
0 U
fi
O
CO
H
rt
p..
0
0
•
1 — I
l__j
^
0)
3
C
co
.a
OJ
bO
rt
-H
03
B-5
c3
rt
S-i
S
n)
01
^
O
V-j C
o
0) C
4-1 GJ
03 >
a a)
S c
o o o o
•i-l O -H ._» H
•LJ O -U 4_t 4_t
•H w to e; -
•one s-i L, L.
CO O 4J ,a _o ^:
O -H -H 3 -H -H -•
U 4J 4J ,JD iH — « ~
CT3 T3 fl ^ "3
rH )-l J-J T3 O U O
rH ,Q Xl CU
Cl] -H 'H ,£ ^j !U O
i— 1 rH U) 13 "C T3
CO TO Cfl J-i QJ -H --I -i-t
ucj ucuoc cort xrt tr "3
OJ 4-l^-i UQJ U 11 1CJ
CUU-i Cc^CC 3V-i 3U 3 —
SO KHS-H OTO Ot C~
Cvl vC ^rrO-^- ,i"r-. OLO|
1 I 1 + + + ^ 4-
-f
O O OOO CO '--C
o^ i-n OOO ceo vc-^
CN CO -
co ui o - •—
S ^ S — i 3'— imJ ^
U 0 ^ T3] "i
P 3 ° ol i
V- -1- O IH
C (J i— i *_j
3 T3 Tj CC J^
C C JH
CO ,C O
cr
-.
^
H ^ O-
1/1 O vj" 'C CO ""I L" c^ """"i ^~>
(N CO rH 00
"^
T
v;
^
LO -t a> o ' Z a
.
O
O l«
E w
O C
S~ Oj
Qj
^ Or- O> o^ c C
^- J^ (^--^vCvDaJCLj
"^ --- -k -K •« -^ o \ --- -^ ---
"^- f"> r^cj\cN m rj mf-i-^c
CO ^-l Csicjro c-j-^. ~-^.-^__<;
^c
-------
suspended solids concentration model is then interfaced with whatever method
is used locally to predict runoff quantities.
Relatively few samples are needed; only 15 to 20 from each of 5 to 8
watersheds as a minimum should be collected from a range of storm events.
However, it is unnecessary to monitor the runoff events continuously. As
long as discharge is known spot sampling is adequate because each sample is
treated independently by the model. If continuous monitoring data is avail-
able, the precision of the model should be markedly enhanced by separate
consideration of the rising and falling limbs of the hydrograph.
III-4
-------
"O
•H
--I
O
en
"O
-------
TTI-3. METHODS AND PROCFDURFS
Efforts have- been concentrated on small watersheds (
-------
start of runoff to an average response time for the watershed. Response
time was defined as the time elapsed between the start of runoff and the
crest of the hydrograph. As a result, samples on the rising limb had relative
time ratios < 1.0, those on the descending limb were > 1.0. Separation of
samples into rising and falling limb categories improves the statistical
significance of the multiple regressions. However, this separation has not
been included in the model because it may reduce the availability of data
for calibration at other sites.
After the initial determination of primary independent variables,
multiple regressions were run in each watershed. The regression coefficients
for each independent variable were plotted as a function of the extent of
the watershed which was urbanized (Fig. III-l). The extent of urbanization
is the sum of residential, commercial, industrial and transportation land
uses. This factor was used—rather than extent of imperviousness—because
it is more readily obtainable from literature or from local or regional
planning agencies.
The graphs in Fig. III--1 can be used to create a multiple regression
equation for a small watershed for which degree of urbanization is known.
Table II1-2 lists equations for several levels of urban development. Thus
a user need know only the following to operate the model:
a. Watershed drainage area (km2),
b. area urbanized (%) ,
c. instantaneous discharge for the time suspended solids concentration
is desired (m3/sec),
d. rainfall intensity (cm/hr), and
e. antecedent rainfall period, i.e., number of days since preceding
rain which produced runoff rain (days).
The degree of urbanization determines which equation to use (Table III-2;
Fig. III-l), and the equation provides the instantaneous suspended solids
concentration after Items a, c, d and e are entered in the model. The
instantaneous discharge values can be obtained from any runoff predicting
system available to the user, from the basic "Rational Method" to the more
sophisticated digital models. Any error inherent in discharge prediction
will be additive in this water quality model.
However, a word of caution is essential for developing the suspended
solids model. It has been found that the regression coefficient for
instantaneous discharge is sensitive to active construction. For those
watersheds where construction is underway (Brown Deer) coefficients are
produced which fall above the line in Fig. Ill-la. Data were insufficient
to determine the extent to which construction activity affects the coeffi-
cient, but it is known that the model will produce erroneous results under
such conditions.
III-7
-------
Table IIJL-2. Coefficient for f >naJ "egression equations for various
degrees of urbaniza t i on '•
Watershed
urbanized, %
0
20
40
60
80
100
Coefficient
for QA,
m3 /sec/km2
(a)
+700
+550
+400
+250
+100
-50
Coefficient
for I, cm/hr
(b)
0
+80
+200
+520
+1420
+3000
Coefficient
for A, days
(c)
-12
-3.5
-4.5
+12.5
+21
+29
Regression
constant
(d)
+160
+80
0
-120
-400
-820
*SS = a(QA) + b(I) + c(A) + d, where SS is suspended solids concentration
(mg/L), QA is discharge/unit drainage area (m /sec/km ), I is rainfall
intensity (cm/hr), A is antecedent dry period (days).
-------
III-4. RESULTS AND DISCUSSION
In an attempt to determine the reliability of the model and limitations
of its use, several tests have been tried. The model has been used to predict
suspended solids loads for streams in the study area, one of which was used in
calibrating the model. Also, it was tested against published data for small
watersheds outside the Great Lakes Watershed. It would have been desirable to
also test in the Great Lakes area outside of southeastern Wisconsin, but data
for small watersheds were not available. The model also was tested on watersheds
having different geological and hydraulic conditions from those used to calibrate
it and for storms of different magnitudes and intensities from the studied storms.
For each test, measured flow rather than predicted flow was used because
the model provides no method of flow prediction, and the use of any runoff
predictor introduces an error in the final load calculations. That error com-
pounds with any error due to the suspended solids prediction. Separation of
these two errors is difficult and clouds the validity of the test of the
empirical model. Thus, it is assumed that each user will interface the sus-
pended solids model with his own method of obtaining flow.
Comparison of observed suspended solids loads with those predicted by
the model for the Brown Deer Watershed for a storm event on 6/8/77 is shown in
Fig. III-2. The Watershed is one used for calibration of the model, but data
fro a this event were not used in the calibration. The Watershed is 65%
urbanized and the equation derived from Fig. III-l is:
SS = 200 QA + 6801 + 14.5A - 170,
where: SS is suspended solids concentration (mg/L)
QA is discharge/unit area (m3/sec/km2)
I is rainfall intensity (cm/hr)
A is antecedent dry period (days)
The agreement is obviously good. The comparative suspended solids loads/unit
area (QA x SS) are shown in Fig. III-3, and agreement again is good. All
further tests compare loads because they are more reliable indicators of
average stream conditions during an event. Concentrations tend to fluctuate
dramatically in the early and late stages of an event when discharge is very
low. However, these fluctuations are of little importance because the stream
does not carry large quantities of suspended solids at these times. Comparison
of loads attaches more importance to the bulk of the sediment transported.
A second test (Fig. III-4) was run on Underwood Creek, one of the larger
(49.7 km ) tributaries to the Menomonee River. In this case, agreement is
poor likely because the Watershed is outside the size range of watersheds for
which the model was calibrated. Because of its size, Underwood Creek is not
simply a single stream with ephemeral tributaries, but has two main branches
which complicate its hydraulics. The model does not work well on complex or
large stream systems.
IH-9
-------
(30
6
01
•O
0)
•O
250 •
200 '
150 -
of 100 -
50 -
800
1000 1200 1400
Time, hr
1600
3000
3200
Fig. III-2. Comparison of observed and predicted suspended solids
concentrations for Brown Deer Creek on 6/8/77. Total
precipitation was 1.32 cm, the antecedent dry period
was 3 days and rain fall intensity was 0.25 cm/hr.
111-10
-------
o
ra
to
"O
(0
o
o
ra
•a
a)
•a
a
0!
a
tn
3
CO
20.0 •
10.0 .
800
3200
Fig. III-3.
Comparison of observed and predicted suspended solids
loads for Brown Deer Creek, Milwaukee, Wisconsin on 6/8/77,
Total precipitation was 1.32 cm, antecedent dry period
was 3 days and rainfall intensity was 0.25 cm/hr.
III-ll
-------
I
o
a)
w
(JO
T3
cd
o
rH
CD
TJ
•H
rH
O
CO
03
a)
0)
3
CO
50
40
30
20 '
10 "
11 15 19 23 27 31 33
Time, hr
39 43 47
51
Fig. III-4. Comparison of observed and predicted suspended solids
loads for Underwood Creek on 4/23/76. Total precipi-
tation is 5.4 cm, antecedent dry period was 1 day and
rainfall intensity was 0.30 cm/hr.
111-12
-------
Other tests were run using data from Durham, North Carolina (1),
San Francisco, California (4) and Cincinnati, Ohio (5). The purpose of these
tests was to determine whether the coefficients established in Wisconson could
be transferred to other urban areas where topographic, climatic and geologic
conditions were different. It was anticipated that these conditions would
each play major roles in defining the coefficients and consequently the degree
of transferability that could be achieved.
The Durham, North Carolina data is most complete, providing runoff and
suspended solids for a wide range of rainfall events on a 4.3 km watershed
which is 80% urban. The terrain is steeper than that in Milwaukee (average
land slope of 6 to 7% in Durham, 2% in Milwaukee) and geologic conditions are
entirely different, However, for storms which fall within the range of intens-
ity and total precipitation of storms used to calibrate the model, there is
remarkably good agreement (Figs. III-5 to III-7).
The model was calibrated in the Menomonee River Watershed using storms
which had intensities > 0.25 cm/hr and total precipitation > 1.0 cm. With
the Durham data, the model was used to predict suspended solids for each of
the 34 events for which rainfall data was available (1). It was found that
the model did not agree with observed data for events of intensity < 0.25
cm/hr (19 events). Of the remaining 15 events, 7 had precipitation of < 1.0
cm, and were not handled well by the model. However, for the 8 events which
had intensity > 0.25 cm/hr and total precipitation > 1.0 cm, the model worked
well (Figs. III-5 to III-7). It seems that rainfall conditions and percentage
development may play a larger role in controlling the sediment regression
coefficients than local topography and geology.
Data from the Bloody Run Watershed in Cincinnati (5) also provided an
opportunity for investigating the transferability of the model. This Water-
shed is 9.63 km in size, is 80% urban and has an average slope of about 5%.
Again it is topographically and geologically different from the Menomonee
Watershed. Data for several events are published, but only four fall within
the total precipitation and intensity range valid for the model. Use of the
model to predict suspended solids loads for Bloody Run, Cincinnati are shown
in Figs. III-8 and III-9. Agreement with observed values is not particularly
good. The results for the 9/25/70 event (Fig. III-8) reveal a major short-
coming of the model, i.e., the model is extremely insensitive to changes in
suspended solids during events when discharge remains relatively constant.
The Bloody Run flow response to a rainfall of 1.65 cm (intensity of 0.73 cm/hr)
on 9/25/70, varied only from 0.27 m3/sec/km2 to 0.30 m3/sec/km2 over a 3.5 hr
period. Consequently, the model, which is discharge dependent, predicted a
relatively constant solids load while observed values were variable. Such a
response from an urban watershed is probably anomalous, but nonetheless, the
model does not handle it well.
The San Francisco data (4) provides a less comprehensive test than
Durham or Cincinnati. Only one storm fits in the intensity and total rainfall
conditions for the model, and it has an anomalous antecedent dry period of
19 days. For a watershed of 0.73 km2 which is 100% developed, the model
greatly overpredicted suspended solids (Fig. 111-10). However, it does
properly predict for this Watershed the unusual conditions where suspended
solids concentrations increase when runoff decreases (Fig. III-ll). This
dilution effect is anomalous for suspended solids. In fact, if an antecedent
111-13
-------
4
o
dl
CO
•a
to
o
01
•a
O
CO
•a
cu
t>
d
(i>
a
CO
200
160 -
120
80
40 •
Predicted
26
Time, hr
Fig. III-5.
Comparison of observed and predicted suspended solids
loads for Event 32 at Third Fork Creek, Durham, North
Carolina. Total precipitation was 2 cm, the antecedent
dry period was 2.5 days and rainfall intensity was
0.48 cm/hr.
111-14
-------
5000
u
-------
o
0)
w
cfl
O
CO
13
•H
fi
0)
CU
W
CO
8000
60 6000
4000 •
2000
Fig. III-7. Comparison of observed and predicted suspended solids
loads for Event 27 at Third Fork Creek, Durham,
North Carolina. Total precipitation was 3.8 cm, the
antecedent dry period was 11.3 days and rainfall
intensity was 1.14 cm/hr.
111-16
-------
I
•H
EH
- oo
>. C
•d a)
O -3
O OJ
rH CJ
03 (D
4J
H C
O CO
14-1
cn 6
T3 8
n)
O r^
cn ~-~.
o) 6
S o
C m
O r-
•H .
•a
g
a,
en
w
T3
CU
•H
cu cn
cfl QJ
o u -w
•HOC
T3 H -H
0)
1-1 rH
D, • M
O CO
a ~- c
co m -H
CN CO
a) a\
> T3
t j p4 £*
CD o oj
to
fi O fx,
O -H CO
j: T)
M o
en
03
O
O
in
o
c!
O
CO CO
•H C
(-1 C TJ
cfl -H O
a o -H
6 e *-i
O -H 0)
000,
oo
I
spjjos pspuadsng
txO
•H
111-17
-------
• 00
o
ro
0)
•H
H
O
•ro
O
o
o
.-I
o
o
in
iH
L/1
"3
r; to
O T3
C
•H O
u
w
- cfl
C &
PS 13
O
>,-H
T5 >-(
O OJ
o a
iH
PQ >-.
O
M-( 4-1
a
en 01
•O TJ
cfl
-------
u
SJ
CO
a
o
CO
T3
•H
<—!
O
CO
-a
cu
T3
C
0)
P-.
CO
1000
Predicted
A
Predicted
Observed
»9<
Fig. 111-10.
Time, hr
Comparison of observed and predicted suspended solids
loads, Baker Street Basin, San Francisco, California on
11/5/69. Loads have been predicted using the general
model (x) and also a modified model which reduces the
importance of antecedent conditions (o). Total precip-
itation was 1.6 cm, antecedent dry period was 19 days
and rainfall intensity was 0.33 cm/hr.
111-19
-------
1 -
- 7SO
- 250
to
E
h- 500 £
c
(D
O
C
O
O
-------
dry period of 1 day is entered into the equation, the model produces very
reasonable results. Exactly what this means is not understood. Perhaps the
model does not work for such a steep (average slope 8 to 10%) watershed or for
such long antecedent dry periods. Or perhaps on steep watersheds, the
antecedent dry conditions become unimportant or the model is unaffected after
1 or 2 days. The interpretation of the test remains unresolved.
111-21
-------
REFERENCES-III
1. Colston, N. V. Jr. Characterization and Treatment of Urban Land Runoff.
U.S. Environmental Protection Agency Report No. EPA 670/2-74-096, 1974.
157 pp.
2. Konrad, J. G., G. Chesters and K. W. Bauer. Menomonee River Pilot
Watershed Study: Semi-Annual Report. IJC-Pollution from Land Use
Activities Reference Group. Sponsored by U.S. Environmental Protection
Agency. April 1976. 135 pp.
3. Cherkauer, D. S. The Hydrologic Response of Small Watersheds to Suburban
Development: Observations and Modeling. In: Urbanization and Water
Quality Control, W. Whipple Jr., ed., American Water Resources Association,
Minneapolis, Minn., 1975. pp. 110-119.
4. Yen, B. C., V. T. Chow and A. 0. Akran. Stormwater Runoff on Urban
Areas of Steep Slope. U.S Environmental Protection Agency Report No.
EPA 600/2-77-168, 1977. 91 pp.
5. University of Cincinnati, Department of Civil Engineering. Urban Runoff
Characteristics. U.S. Environmental Protection Agency Report No. EPA
11024 DQU 10/70, 1970. 340 pp.
IIT-22
------- |