United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-79-106a
August 1979
Research and Development
Screening/Flotation
Treatment of
Combined Sower
Overflows
Volume II:
Full-Scale Operation
Racine, Wisconsin
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research —
4. Environmental Monitoring
5, Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-106a
August 1979
SCREENING/FLOTATION TREATMENT
OF COMBINED SEWER OVERFLOWS
VOLUME II: FULL-SCALE OPERATION
RACINE, WISCONSIN
by
T. L. Meinholz, D. A. Gruber, R. A. Race
C. A. Hansen, J. H. Moser, M. J. Clark
Envirex Inc.
A Rexnord Company
Environmental Sciences Division
Milwaukee, Wisconsin 53201
Grant No. S800744
Project Officers
Anthony N. Tafuri
Storm and Combined Sewer Section
Wastewater Research Division
Municipal Environmental Research Laboratory (Cincinnati)
Edison, New Jersey 08817
and
Clifford Risley
Great Lakes Region V
Environmental Protection Agency
Chicago, Illinois 60606
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.So Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use«
!i
-------
FOREWORD
The Environmental Protection Agency was created because of increasing public
and government concern about the dangers of pollution to the health and
welfare of the American people. Noxious air, foul water, and spoiled land
are tragic testimony to the deterioration of our natural environment. The
complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solution
and it involves defining the problems, measuring its impact, and searching
for solutions. The Municipal Environmental Research Laboratory develops new
and improved technology and systems for the prevention, treatment, and
management of wastewater and solid and hazardous waste pollutant discharges
from municipal and community sources, for the preservation and treatment of
public drinking water supplies, and to minimize the adverse economic, social,
health, and aesthetic effects of pollution. This publication is one of the
products of that research; a most vital communications link between the
researcher and the user community.
This report describes the planning, design and construction and the operation
over a two-year evaluation period of three full-scale demonstration systems
for the treatment of storm generated discharges using the screening/flotation
principle.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
-------
ABSTRACT
This report describes the planning, design and construction and the operation
over a two-year evaluation period, of three full-scale demonstration systems
for the treatment of storm generated discharges. As part of the evaluation,
the quality of the receiving body was also monitored. Two of the systems -
located at two major points of combined sewer overflow to the Root River in
Racine, Wisconsin - are identical in concept and employ screen ing/dissol ved-
air flotation to treat the overflow prior to discharge. The two systems have
a combined capacity of 222,000 cu m/day (58.5 mgd). The third system utilizes
screening only for the treatment of urban stormwater. It has a capacity of
14,800 cu m/day (3-9 mgd).
This report also describes the verification and modification of the EPA
Storm Water Management Model using the subject drainage area, sewerage and
treatment systems, and receiving body.
Results from the evaluation program indicate that the "satellite plant" con-
cept of locating treatment plants at points of combined sewer overflow dis-
charge is a feasible alternative to comb i ned • sewer separation. Based on the
operating results for these systems, removal efficiencies of 60 to 75 percent
can be expected for suspended solids and 50' to 65 percent for BOD. The chlori-
nation system met the fecal coliform standard for whole body contact specified
by the State of Wisconsin for the Root River. It was concluded that the
operation of the treatment systems had a beneficial effect on the quality
of the River.
Results from the screening of urban stormwater Indicate that this method
will remove 50 percent of the suspended solids and 20 percent of the BOD.
This report was submitted in fulfillment of Srant No. S8Q0744 (formerly
11023 FWS) under the partial sponsorship of the Environmental Protection
Agency. The study program associated with this project was performed by
Envirex Inc. acting as a subcontractor to the grantee, the City of Racine.
Work was completed as of November
IV
-------
TABLE OF CONTENTS
ABSTRACT
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
ACKNOWLEDGMENTS
Page Number
IV
v
vii
x.H
xv i,I
SECTIONS
I CONCLUSIONS
II RECOMMENDATIONS
III INTRODUCTION
I I 1-1 Combined Sewer Overflow Problem
111-2 Project Objectives
I
8
12
12
14
IV TREATMENT SITES
IV-1 Preliminary Studies
IV-2 System Design and Construction
IV-3 Operation and Maintenance Methodology
IV-4 Treatment Results
IV-5 Economic Considerations
V ROOT RIVER HQmTORING STUDIES
V-l General Description of the Root River Drainage
System
V-2 Methods Used to Monitor the River
V-3 River Monitoring Periods
V-4 Monitored Parameters and Results
VI STORM WATER MANAGEMENT MODEL
Vl-l Introduction
VI-2 Runoff and Transport Blocks
VI-3 Storage Block
15
15
39
84
99
148
162
164
16?
205
205
207
220
-------
TABLE OF CONTENTS (continued)
Page Number
VI-4 Receive Block 223
VI-5 Total Program Evaluation 239
VI-6 Remaining Combined Sewer Areas 341
VII REFERENCES 355
VIII *APPENDICES
IV-A Pretreatment Discharge Quality 358
IV-B Demonstration System Costs 370
IV-C Site Operational Parameters 374
IV-D Analytical Results for System Operation 420
1V-E Description of Analytical Techniques 466
IV-F Site Treatment/Bypass Volumes-May to 470
Sept., 1974
IV-G Pollutional Mass Removals Achieved 476
by Treatment Sites
IV-H Total Pollutional Mass Removals from 501
Combined Sewer Overflow Volumes
V-A Root River Quality Data
VI-A Runoff Block Data
VlrB Transport Block Data
VI-C Storage Block Data
Vi-D Receive Block Data
VI-E Rainfall Intensities
VI-F Remaining CSO Area Data
514
622
633
648
652
656
670
*Section VIII - Appendices is available through the National Technical
Information Service, Springfield, Virginia 22161.
VI
-------
LIST OF FIGURES
I Drainage Areas, Combined and Storm Sewer Overflows
and Division of River for Selection of Project Test Reach
2 Schematic Diagram of Simulation Used in Preliminary
River Model
3 Effect of Combined and Storm Water Overflows on the Root
River in Racine, Wisconsin, As Predicted by the
Preliminary River Model
4 Fecal Coliform Concentrations at Root River Sections,
October 28, 1970
5 Alternative Sites for CSO Treatment Plants
6 Michigan and Dodge Streets Overflow Mechanisms,
Discharge No. I
7 Chatham and Dodge Streets Overflow Mechanisms,
Discharge No. 3
8 Last Manhole in Sequence for Storm Sewer, Discharge No. 5,
Main and Dodge Streets
9 Wisconsin and Dodge Streets Overflow Mechanism,
Discharges No. 7 and 8
10 Inside of Discrete Sampler
11 Program Cycle for Automatic Sampler
12 Typical Sampler Installation (Installation at Overflow
No. 5 in Storm Sewer)
13 Typical Installation of Depth Recording Instrumentation
14 Raingage and Treatment Site Locations and Contributing
Areas
15 Site I Treatment Facility
16 Site II and HA Treatment Facilities
17 Schematic of Site I Treatment Facility
18 Schematic of Sites II and MA Treatment Facilities
19 Automatic Operation of Site I and II Treatment Systems
Page Number
16
19
23
2k
26
30
31
32
33
35
36
37
38
44
45
46
49
50
51
vil
-------
LIST OF FIGURES (continued)
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
4f
46
47
Site II Spiral Screw Pump and Bar Screen
Site 1 Parshall Flume and Transmitter
Site II Drum Screens
Site II Drum Screens with Hold-Down Bars
Site II Drum Screen Influent Channel
Site MA Drum Screen
Site 1 Screen Backwash Pump and Wet Cyclone
Site II Screw Conveyor and Flotation Tanks
Site II Pressurized Flow System
Site 1 Pressurized Flow System
Pressurization Tank Level Control and Sight Glass
Pressure Control Valve in Manhole Ahead of Flotation Tank
Mixing Zone Behind Perforated Influent Baffle In
Flotation Tank
Site 1 Sludge Tank
Site 1 Flow Distribution Channel and Screw Conveyor
Site II Flotation Tank Scraper Flights
Site II Sludge Tank
Site II Washdown Water Pump
Site II and 1 IA Supervisory Control Panel
Site II Control Panel
Site 1 Air Compressor and Air Dryer
Site II and 1 IA Recording Flow Meters (Pressurized Flow
and as indicated)
Site 1 Chemical Feed Pumps
Site II Chlorination Equipment
Conceptual Design of SDAF Treatment System
Sketch of Rotating Drum Screen
Schematic of Pressurized Flow System
Percent BOD Removal Versus Duration of Run (Site 1)
v i } 1
Page Number
52
52
53
53
54
5k
55
55
57
57
58
58
59
59
60
60
61
61
62
63
64
64
65
67
73
76
77
•134
-------
LIST OF FIGURES (continued)
48 Sieve Analysis of Combined Sewer Overflow for Storm
No. 6, 1971
49 Sieve Analysis of Combined Sewer Overflow for Storm
No. II, 1971
50 Sieve Analysis of Combined Sewer Overflow and Plant
Effluent at Site I, Run No. 12, 1973
51 Sieve Analysis of Combined Sewer Overflow at Site I,
Run No. 42, 1974
52 Root River Watershed Showing Communities and Principal
Industrial Pollution Sources
53 Location of the Root River Watershed
54 Profile of the Root River and its Major Tributaries
(Data Base 1965, Meters » Feet x 0.3048)
55 Water Quality Conditions in the Root River at Horlick
Dam - (1961-1964)
56 Root River Tests Reach and Monitoring Sites - Site A,
Site B, and Site C
57 Root River Monitoring Station at Site C
58 Computer Printout Showing Relationship of Dissolved Solids
Concentration (Dependent Variable) to Specific Conductance
(Independent Variable)
59 Lower Root River Chlorophyll Profile
60 SWMM Block Diagram
61 Subarea Location and Numbers
62 Transport Elements
63 Run Number 12 Arriving Flow Site I
64 Run Number 16 Arriving Flow Site I
65 Storage Block Printout Before and After Corrections
66 Arriving Flow and Overflow Printout Values Before
Modification
67 Arriving Flow and Overflow Printout Values After
Modification
68 Receive Junctions and Channels (Original)
69 Receive Jlnctions and Channels (Modified)
Page Number
139
139
139
149
150
151
159
160
163
»73
201
206
209
210
217
218
224
226
228
230
235
ix
-------
LIST OF FIGURES (continued)
70 Dummy Elements Connecting Site I and Site II
71 Run Number 21 Arriving Flow Site I
72 Run Number 21 Arriving Quality Site I CBOD 400 mg/1
73 Run Number 21 Arriving Quality Site I CBOD 150 mg/1
7k Run Number 21 Effluent Quality Site I
75 Run Number 21 Arriving Flow Site H
76 Run Number 21 Arriving Quality Site II
77 Run'Number 25 Arriving Flow Site I
78 Run Number 25 Arriving Quality Site I
79 Run Number 25 Arriving Quality Site I
80 Run Number 25 Effluent Quality Site I
81 Run Number 25 Arriving Flow Site II
82 Run Number 25 Arriving Quality Site II
83 Run Number 25 Arriving Quality Site II
8k Run Number 26 Arriving Flow Site I
85 Run Number 26 Upstream Flow Measurements
86 Run Number 26 Effluent Quality Site I
87 Run Number 26 Effluent Quality Site I
88 Run Number 26 Arriving Flow Site II
89 Run Number 26 Arriving Quality Site II
90 Run Number 26 Arriving Quality Site II
91 Run Number 26 Effluent Quality Site II
92 Run Number 26 Effluent Quality Site II
93 Run Number 27 Arriving Flow Site I
Sk Run Number 27 Arriving Quality Site I
95 Run Number 27 Arriving Quality Site I
96 Run Number 27 Arriving Flow Site II
97 Run Number 27 Arriving Quality Site II
98 Run Number 27 Arriving Quality Site II
99 Run Number 27 Effluent Quality Site II
Page Numbers
237
2k\
2kk
2k6
2^9
252
255
257
258
261
263
265
266
269
271
273
27k
276
279
280
282
283
288
289
291
293
23k
296
-------
LIST OF FIGURES (continued)
100 Run Number 27 Effluent Quality Site II
101 Run Number 30 Arriving Flow Site I
102 Run Number 30 Arriving Quality Site I
103 Run Number 30 Arriving Quality Site I
104 Run Number 30 Effluent Quality Site I
105 Run Number 30 Effluent Quality Site I
106 Run Number 30 Arriving Flow Site II
107 Run Number 30 Arriving Quality Site II
108 Run Number 30 Arriving Quality Site II
109 Run Number 30 Effluent Quality Site II
110 Run Number 37 Arriving Flow Site I
III Run Number 37 Arriving Quality Site I
112 Run Number 37 Arriving Quality Site I
113 Run Number 37 Effluent Quality Site I
114 Run Number 37 Effluent Quality Site I
115 Run Number 37 Arriving Flow Site II
116 Run Number 37 Arriving Quality Site II
117 Run Number 37 Arriving Quality Site If
118 Run Number 37 Effluent Quality Site II
119 Run Number 45 Arriving Flow Site I
120 Run Number 45 Arriving Quality Site I
121 Run Number 45 Arriving Quality Site I
122 Run Number 45 Effluent Quality Site I
123 Run Number 45 Effluent Quality Site I
124 Run Number 45 Arriving Flow Site II
125 Run Number 45 Arriving Quality Site II
126 Run Number 45 Arriving Quality Site II
127 Run Number 45 Effluent Quality Site II
128 Remaining Subareas
129 Remaining Transport Network
130 Storage Block Treatment Units
Page Number
297
300
302
303
305
306
308
310
312
314
317
319
320
322
323
324
327
328
331
334
336
337
339
340
343
345
346
348
349
351
354
XI
-------
LIST OF TABLES
Page Number
17
20
20
22
27
40
I Areas Served by Combined Sewers and Separate Sewers
as of 1970
2 Input Conditions for Preliminary River Model
3 Quality Values Used for Preliminary River Model
4 Simulated Overflow Volumes and Characteristics for Racine,
Wisconsin
5 Evaluation of Alternative Sites
6 1971 Overflow Characteristics - Site I - Overflows
No. I and 3
7 1971 Overflow Characteristics - Site MA - Overflow No. 5 41
8 1971 Overflow Characteristics - Site II - Overflow No. 7-8 41
9 Overflow Quality Vs. Time, Overflows No. 7-8 (Site II), 42
June 18 - November 18, 1971, Arithmetic Means for 10
Occurrences
10 1971 Overflow Ve-lume Data 43
II Design Criteria for Combined Sewer Overflow and Storm Water 68
Treatment Systems
12 Process Measurement and Control Instrumentation 79
13 Results of Bench Scale Flocculatlon Tests Performed on 88
Site II Overflows Date: December 4, 1973
14 Average Operational Parameters for Site I 89
15 Average Operational Parameters for Site II 90
16 Average Operational Parameters for Site IIA 91
17 Breakdown of Maintenance Time for May and June, 1974 96
18 Rainfall Characteristics and Combined Sewer Overflow Volumes 105
19 Correlation Coefficients for Regression Analyses Performed 106
on Overflow Volumes and Rainfall Characteristics
20 Rainfall Characteristics and Treatment Sites Bypass Volumes 1Q9
21 Correlation Coefficients for Regression Analyses Performed lo8
on Plant Bypass Volumes During Operation and Rainfall
Characteristics
-------
LIST OF TABLES (continued)
22 Days and Volumes of Combined Sewer Overflows in May
to September, 1974
23 Comparison of Means for 1971 and 1974 Overflow Quality
24 Comparison of Means for Sites I and II Overflow Quality
(1974)
25 Comparison of Quality of Combined Sewer Overflows for
Racine and Various Other Cities (9)a
26 1973 and 1974 Storm Water Characteristics - Site IIA
27 Comparison of Quality of Storm Sewer Discharges from
Racine and Various Other Cities (9)a
28 Rainfall Characteristics and Storm Sewer Discharge
Quality (Site IIA)
29 Correlation Coefficients for Regression Analysis Per-
formed on Storm Sewer Discharge Quality and Rainfall
Characteristics
30 Screened Effluent Characteristics and Average Percent
Removals by the Screens - Site I
3' Screened Effluent Characteristics and Average Percent
Removals by the Screens - Site II
32 Screened (Final) Effluent Characteristics and Average
Percent Removal by the Screens - Site MA
33 Screen Backwash Characteristics (mg/1) (1973)
34 Floated Sludge Characteristics (mg/1) (1973)
35 Final Effluent Characteristics and Average Percent Removal
Site I
36 Final Effluent Characteristics and Average Percent Removal
Site II
37 Percent Removals, for 1973 Compared to Percent Removals for
1974
38 Percent Removals Based on Mass of Pollutants - Site I
39 Percent Removals Based on Mass of Pollutants - Site II
40 Total Mass Removals from Combined Sewer Overflow Volumes-
Site I (1974)
41 Total Mass Removals from Combined Sewer Overflow Volumes-
Site II (1974)
42 Nitrogen Series and Total Dissolved Solids for Run No. 12
After Long Dry Spell - July 20, 1973 (17 Days Since last
Overflow)
Page Number
III
113
114
115
116
117
118
119
121
121
121
122
125
127
128
129
131
132
138
138
141
xi i i
-------
LIST OF TABLES (continued)
43 Operation and Maintenance Costs - August 1974; Enr =• 2078
44 Estimated Future Operation and Maintenance Costs-
August, I972*; Enr - 2078
45 Actual and Estimated Operation and Maintenance Costs for
Racine Compared to 7 Combined Sewer Overflow Treatment
Sites (Enr - 2000)
46 Mean Monthly Precipitation at Milwaukee, Wisconsin-
(1854 - 1964)
47 Water Quality of the Root River at Horlick Dam - Racine,
Wisconsin - (1964 - 1968)
48 Monitored Dry Weather Periods - 1971
49 Monitored Storm Periods - 1971
50 Monitored Storm Periods - 1973
51 Monitored Storm Periods - 1974
52 Rainfall Summary for Racine, Wisconsin for 1971
53 Rainfall Summary for Racine, Wisconsin for 1973
54 Rainfall Summary for Racine, Wisconsin for 1974
55 Monthly Means and Coefficients of Variation for Dis-
charge of the Root River at Racine for Months Monitored
56 Specific Conductance Grand Means for All Sampling
Periods (p mhos/cm)
57 Total Solids - Mean Values for All Monitored Periods
58 Comparison of Total Phosphorus Yearly Means at the Three
River Sampling Sites
59 TOC Yearly Means
60 BOD Yearly Means (mg/1)
61 Root River Bottom Sediment Analyses
62 Concentration of Ortho Phosphate as P and Nitrate as
N Measured at Six River Sites (mg/1)
63 Benthic Organism Survey
64 Light and Dark Bottle Oxygen Determinations (mg/1) (1973)
65 Light and Dark Bottle Oxygen Determinations (mg/1) (1974)
66 Species Classification at Horlick Site (1971) 050M)
(500 ft) Downstream of Dam
67 BOD Values from River Cross-Section (mg/1) (1971)
Page Number
143
145
146
154
157
165
165
166
166
168
169
170
172
175
179
181
183
186
190
195
196
197
199
xiv
-------
LIST OF TABLES (continued)
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
Solids Collected In River Sedimentation Collection
Buckets (gm/day/sq m)
1973 Pesticide Results - Date: September 26, 1971
(ng/1 or ppt)
Pesticide Results for Storm 13 - After Long Dry Spell-
Date: September 4, 1973 (45 days since last overflow)
(in ng/1 or ppt)
Pesticide Results - Date: July 2, 1974 (ng/1 or ppt)
Run Number 12 Arriving Flow Site I
Run Number 16 Arriving Flow Site I
Receive Block Junctions (initial)
Receive Block Channels (initial)
Receive Block Channels (modified)
Receive Block Junctions (modified)
Receive Quantity (initial)
Run Number 21 Arriving Flow Site I
Run Number 21 Arriving Quality Site CBOD 400 mg/1
Run Number 21 Arriving Quality Site CBOD 150 mg/1
Run Number 21 Effluent Quality Site I
Run Number 21 Arriving Flow Site II
Run Number 21 Arriving Quality Site II
Run Number 25 Arriving Flow Site I
Run Number 25 Arriving Quality Site I
Run Number 25 Effluent Quality Site I
Run Number 25 Arriving Flow Site II
Run Number 25 Arriving Quality Site II
Run Number 26 Arriving Flow Site I
Run Number 26 Arriving Flow - Upstream Measurements
Run Number 26 Effluent Quality Site I
Run dumber 26 Arriving Flow Site II
Run Number 26 Arriving Quality Site II
Run Number 26 Effluent Quality Site II
Run Number 27 Arriving Flow Site I
Page Number
200
202
203
204
216
219
231
232
234
234
238
240
243
245
247
250
251
254
256
259
262
264
268
270
272
275
278
281
285
XV
-------
LIST OF TABLES (continued)
97 Run Number 27 Arriving Quality Site I
98 Run Number 27 Arriving Flow Site II
99 Run Number 27 Arriving Quality Site II
100 Run Number 27 Effluent Quality Site II
101 Run Number 30 Arriving Flow Site I
102 Run Number 30 Arriving Quality Site I
103 Run Number 30 Effluent Quality Site I
104 Run Number 30 Arriving Flow Site II
105 Run Number 30 Arriving Quality Site II
106 Run Number 30 Effluent Quality Site II
107 Run Number 37 Arriving Flow Site I
108 Run Number 37 Arriving Quality Site I
109 Run Number 37 Effluent Quality Site I
110 Run Number 37 Arriving Flow Site II
III Run Number 37 Arriving Quality Site II
112 Run Number 37 Effluent Quality Site II
113 Run Number 45 Arriving Flow Site I
114 Run Number 45 Arriving Quality Site I
115 Run Number 45 Effluent Quality Site I
116 Run Number 45 Arriving Flow Site II
117 Run Number 45 Arriving Quality Site II
118 Run Number 45 Effluent Quality Site II
119 Remaining Subarea Data
120 Transport Conduit Elements for Remaining Areas
Page Number
287
290
292
295
298
301
304
309
311
3«3
315
318
321
325
326
330
332
335
338
342
344
347
350
352
XVI
-------
ACKNOWLEDGEMENTS
Many persons from the Environmental Sciences Division of Envirex Inc.
contributed to the successful completion of this study.
Mechanical design and subsequent start-up of the treatment facilities
were carried out fay the Design section headed by Joseph E. Milanowski.
Field operation of the treatment units during more than fifty storm events,
quickly responded to at all hours of the day and night, was performed by
such dedicated individuals as John Moser, Michael Clark, Richard Race,
David Gruber and Thomas Meinholz.
Timely completion of the laboratory analyses for the storm events over the
two year monitoring period was achieved consistently by Richard E.
Wullschleger and his laboratory staff.
The cooperation of the City of Racine in the day to day operation and
maintenance of the treatment facilities is duly recognized.
Deep appreciation is extended to the U.S. Environmental Protection Agency,
especially Project Officers Stephen Poloncsik and Clifford Risley of
Region V and Messrs. Anthony Tafurf, Staff Engineer, and Richard Field,
Chief, Storm and Combined Sewer Section, Edison, NJ, Frank Condon and
William Rosenkranz, Washington, DC for their continued aid and helpful
advice during the project.
XVI
-------
-------
SECTION I
CONCLUSIONS
SECTION IV, TREATMENT SITES
1. Based on storm-generated discharge measurements and on quality determina-
tions of the discharge and of the river during 1971, the discharge of un-
treated combined sewage and storm water was a major source of pollution
of the Root River within the City of Racine. For example, the concentra-
tion of fecal coliform bacteria near the mouth of the river during a
6-hr period after a storm, averaged more than 40 times that during dry
weather.
2. Based on a water quality survey and preliminary river modeling, the maxi-
mum benefit, in terms of the navigable portion of the river, would result
from treatment of discharges in the lower reach of the river. Signifi-
cant points of overflow and potential sites within the reach were inves-
tigated. The most cost-effective potential site was in the area of Main
and Dodge Streets near downtown Racine and the three demonstration sys-
tems were constructed there. Two systems, referred to as Site I
and II in this report, employ screening/dissolved-air flotation for treat-
ment of combined sewer overflow and have design treatment capacities of
53,500 cu m/day (14.3 mgd) and 168,000 cu m/day (44.4 mgdj, respectively.
The third system is adjacent to Site II and is referred to as Site IIA.
At this Site, screening only is used for the treatment of a storm sewer
discharge.
3. The mean quality characteristics found in 1971 were:
CSO QUALITY CHARACTERISTICS
Combined sewer overflow
Item
BOD
TOC
SS
Fee. Coll
Parameter
mg/1
mg/1
mg/1
No./ 100 ml
Si
1971
79
98
298
10,300
te 1
1974
93
95
266
609,000
S
1971
212
238
669
10,100
ite II
1974
110
122
661
416,000 ,
Storm sewer
discharge
Site
1971
39
51
445
23
IIA
1974
15
46
376
580
-------
Using the data collected over the two-year evaluation period, the
combined sewer overflow volumes were related to rainfall by the
following equations:
V, - (15,502 x R) - 3,270
V2 » (31,770 x R) - 8,879
where, V. « overflow volume at Site I, cu m
V_ = overflow volume at Site II, cu m
R « total rainfall, cm.
Overflow volumes bypassing the plants during operation were related
to the rainfall by the following equations:
BV
671R - 517
H 12,737E - 5873
where, BV.
BV,
R
E
plant bypass volume at Site I, cu m
plant bypass volume at Site II, cu m
total rainfall, cm
average rainfall intensity, cm/hr
6. Pollutant removals (concentration basis) by the screening systems alone
Percent removal
Site 1
28
38
32
Site II
42
44
36
Site 1 IA
20
41
50
were:
BOD
TOC
Suspended Sol ids
The screening/dissolved-air flotation process is a feasible method
for abating storm-generated discharges by treatment in full-scale
applications. This conclusion is substantiated by the overall percent
removals (concentration basis) achieved by the screening/dissolved-air
flotation process at the demonstration sites over the two-year
evaluation period:
Percent removal
Site I
S i te II
BOD
TOC
Suspended sol ids
Volatile suspended solids
Total phosphorus
50,
47,
59
64,
46.6
60.4
50.4
66.1
57.0
60.3
These values are the removals achieved during the entire two-year
project. Most of 1973 was a period of startup and shakedown, therefore
-------
8.
the 1973 results were generally lower than the average presented. The
results obtained in 1974 are believed to be more representative of the
efficiency of the screening/dissolved-air flotation process. The 1974
percent removals (concentration basis) were:
Percent removal
BOD
TOC
Suspended solids
Volatile suspended solids
Total phosphorus
Site !
57.5
51.2
62.2
66.8
49-3
Site If?
65.4
64.7
73.3
70.9
70.0
from Site II are
usually lower at
better than Site I because the hydraulic
II than at Site I resulting in
at Site II.
The results
loading was usually lower at Site II than at Site I
lower overflow rates and longer tank detention times
Calculation of the percent removals on a mass basis resulted in the
following values:
Percent removal
Site I
Site II
BOD
TOC
Suspended sol ids
Volatile suspended solids
Total phosphorus
62.4
60.0
67.6
73.6
53.2
69.5
66,6
69,8
67.3
62.4
9.
The reason for the increase over the arithmetic means is that the
overall treatment efficiency was usually better for long duration runs
(large volumes treated) than for short duration runs (small volumes
treated). Therefore, the mass removals are greater than the
arithmetic mean which gives equal weight to each run without regard
to the volumes treated.
From a mass balance the following estimates on sludge production for
typical system operation were made:
Site I
Duration of run, min
Volume of floated sludge, cu m (gal.)
Total sludge volume, cu m (gal.)
Suspended sol ids, %
Backwash water/total sludge volume,
% of tot. si. vol.
Volume of sludge produced /volume of
overflow treated, cu m/1000 cu m
424
21.4 (5,641)
228.4 (60,343)
0.64
91
26.7
S i te II
212
106.4 (28,108)
407.4 (107,635)
1 .29
74
42.6
-------
10. The floated sludge averaged 4.6 percent solids of which 36 percent was
volatile matter.
11. Screen backwash water averaged 0.23 percent solids of which 60 percent
was volatile matter. On the average, the backwash requirements were
2.7 percent of the plant flow.
flt
12. No relationship could be established between the backwash water
volumes used and the screen hydraulic and solids loadings.
13. When operating correctly, the chlorination system produced an effluent
fecal coliform concentration of 113 colonies/100 ml. This number was
below the standard of 200 colonies/100 ml set by the Wisconsin
Department of Natural Resources for the Root River (standard for
whole body contact).
14. Capital costs for the system can be expressed as follows:
Site I Site M
15-
Total cost, $
$/cu m/day of treatment capacity
($/mgd of treatment capacity)
$/hectare of combined or storm
sewer area
($/acre of combined or storm
sewer area)
^36,599
8.16
30,900
16,730
6,779
841,420
5.01
18,950
5,131
2,078
Site HA
25,001
1.69
6,410
3,968
1,613
The operation and maintenance cost for the systems was 6,08^/cu m
(23.0^/1,000 gal.). This cost probably could be reduced to 3.l8
-------
17.
18.
19.
Using the fecal coliform organism as an indication of river water
quality at three points along the Root River for comparing wet weather
and dry weather data over the entire demonstration period, the yearly
geometric means listed below give an indication of quality changes
which have occurred over time:
Fecal CoUform Concentration, No^/100 ml
Point A
Point B
Point C
1971 Dry
353
344
84
1971 Wet
5986
1775
265
1973 Wet
3084
2117
860
1974 Wet
1253
2057
813
Point A, nearest the river mouth is located in the immediate area of
the treatment sites. Point B is upstream of Point A and -Point C
is upstream of both Points A and B. Of the three points monitored,
only Point A showed a significant improvement in water quality during
wet weather over the entire project. There was a 50 percent decrease
in this parameter at Point A from 1971, when no treatment occurred,
to 1973, when treatment started, and a further 50 percent decline
from 1973 to 1974. This improvement is considered to be a direct
effect of treatment on water quality.
Water quality at Point B measured by fecal coliform content, remained
constant during the entire three-day monitoring period following a
storm-generated discharge event. This site is located downstream of
the last combined sewer overflow prior to the test reach.
Point C fecal coliform concentrations increased from 1971 to 1973 but
remained unchanged from 1973 to 1974.
At each specific point there was improvement in water quality as
indicated by dissolved oxygen levels over the duration of the monitoring
program. Listed,below are the mean dissolved oxygen concentrations for
all three points 6ver the three years monitored:
Dissolved Oxygen Concentration, mg/1
1971 Wet 1973 Wet
Point A
Point B
Point C
7.1
5.8
2.8
7.2
6.1
6.9
1974 Wet
8.3
7.4
7.4
This general improvement in water quality cannot be solely attributed
to the treatment system's operations because of other contributing
factors affecting river DO.
During the entire period of monitoring (1971-1974) no change in the
benthic deposits was noted in the areas of Points A and B. Point C,
-------
located in a high energy area, was affected somewhat by scour during
spring flooding. The benthic deposits at Points A and B, being
relatively stable, represent a significant nutrient source and
probably exert a high benthic oxygen demand.
20. The encroachment of Lake Michigan on the Root River had a significant
influence on the water quality at the Points A and B monitoring areas.
It affected most of the parameters monitored (dissolved oxygen, specific
conductance, temperature) and made data interpretation difficult, if
not impossible.
21. The selection of river monitoring sites, although placed at the best
points available during this study, left much to be desired. There was,
for instance, no monitoring site downstream of both treatment systems.
The lack of such a point hindered the assessment of the effect of
operation of the treatment systems on the quality of the river. In
fact, all of the river monitoring locations were located upstream of
the outfall of Site I and, therefore, no positive conclusions could be
reached about the effect of this treatment unit on the river.
It would have been most beneficial, from the standpoint of monitoring
water quality in the. Root River, if the treatment plants could have
been constructed just downstream of the Point C monitoring station.
This would have allowed the placement of monitoring Points A and B
downstream of the treatment plants but far enough upstream to eliminate
the influence of Lake Michigan on the monitored area.
SECTION VI, STORM WATER MANAGEMENT MODEL
22. The Runoff and Transport blocks of the Storm Water Management Model
(SWMM) have been shown to be adequate in predicting the quantity of
arriving flow at the two combined sewer overflow treatment points,
Sites I and II. These blocks have also been shown to be adequate in
predicting the overall quality of the arriving flow at the two sites.
The quality prediction was acceptable in terms of BOD and fecal
coliform and fair in terms of suspended solids. The computed "first
flush" concentrations for suspended solids lacked accuracy but the
prediction of the remainder of the pollutograph was acceptable.
23. The Storage block was used (1) to verify the removals of pollutants
through the treatment units by comparing computed and measured
effluent concentrations, and (2) to design screening/dissolved-air
flotation units for the remaining combined sewer overflows along the
Root River. In both instances this block has yielded acceptable
results.
2k. The application of the SWMM to the modeled areas of this project
produced manpower and cost requirements that may be summarized as
follows:
-------
a. Applying the SWMM to an urban drainage area requires one-man-day
per 2 ha (5 acres) of drainage area to obtain sewer records,
analyze them and produce the needed input data for the Runoff and
Transport blocks.
b. Data debugging then requires one-man-day per kO ha (100 acres) of
simulation.
c. The costs in CPU time of running the Runoff and Transport blocks
with 150 elements and 100 time steps averaged 60 seconds.
-------
SECTION II
RECOMMENDATIONS
SECTION IV, TREATMENT SITES
1. Immediate action should be taken to treat or to otherwise abate the remain-
ing volumes of overflow in the City of Racine. Satellite treatment plants
employing the screen?ng/dissolved-air flotation treatment process should
be considered as a feasible alternative to combined sewer separation.
2. Large volumes of sludge will be generated throughout the City of Racine
if satellite treatment facilities are employed to treat combined sewer
overflows. If this sludge were bled back to the sewer system after the
overflow subsided (as is done at the present demonstration systems), it
would create an excessive load on the dry weather sewage treatment plant
operations. Therefore, for future satellite plants it is recommended
that sludge handling facilities be included. In addition, the volume of
sludge could be significantly reduced by holding the screen backwash water
and the floated sludge in separate tanks. At the demonstration systems
the backwash water accounted for 80 to 90 percent of the sludge volume
but averaged only 0.23 percent solids. Therefore, it is suitable for
bleed-back to the sewer. The floated sludge, on the other hand, while
accounting for only 10 to 20 percent of the sludge volume, averaged 4.6
percent solids. Therefore, if the floated sludge were collected
separately, it might be feasible to treat it at the satellite plant and
to prevent overloading of the dry weather plant.
3. Because capacity flow at Site II was infrequent, and because when capacity
did occur, it was usually at the beginning of a run, and because Site I
was usually hydraulically overloaded at the beginning of a run, a method
of storage followed by the screening/dissolved-air flotation process may
be beneficial at other sites in Racine, or in other cities. This use of
storage would mean less initial hydraulic load on the system, especially
the drum screens, and would reduce the required capacity of the system
to handle a given storm and overflow event. This approach is basically a
problem of optimizing design storage and treatment capacity.
k. The following recommendations are made regarding the equipment design:
a. Use of an alternative source of water such as final effluent, river
or city water for chemical dilution and screen backwash systems.
-------
b. Greater structural support for the drum screen panels.
c. A new design for the drum seals.
d. Complete separation of the drum screen bypass channel for the drum
screen chamber.
e. Inclusion of a method of removing accumulated solids from the drum
screen chamber.
f. Use of a heavy-duty bar screen to eliminate the jamming of the
rakes.
g. Use of air lines that will not deteriorate and are easily
accessible.
h. Placement of flumes or other flow monitoring devices such that
accurate flow measurements may be obtained.
«
i. An automated method of removing deposited solids from the bottom
of the flotation tanks.
j. A different type of air controller for better control of pressuri-
zation-tank pressures.
Achievement of the design changes would significantly reduce the cost of
treatment. It is estimated that the cost could be reduced from 6.08^/cu m
(23.0
-------
7. Fifty percent removal of suspended solids from storm water was achieved
by use of a 50 mesh screen. Increased removals may be possible if
finer-mesh screening media is employed.
SECTION V, ROOT RIVER MONITORING STUDIES
8.
10.
11.
12.
13.
Monitoring of the Root River during both wet and dry weather periods
should be continued at the points monitored during this project. This
period of monitoring should be carried on over at least the next three
years, and,should measure most of the same parameters measured during
this program. This monitoring would build a substantial data base upon
which to evaluate the water quality of the Root River as it passes
through the City of Racine and to provide further input as to the impact
of the treatment systems on the river, !t would also provide:
a. Comparison of dry weather water quality over a long term period to
determine if there is a time related trend or change in the quality
of the Root River.
*
b. Comparison of dry weather to wet weather events in a given year to
show change due to storm-generated discharges.
c. Better comparison of storm events from year to year.
If possible, some method of,analyzing water quality downstream of both
Sites I and II should be devised and implemented to determine the total
impact of the treatment systems on the receiving body.
For any future CSO demonstration project, if part of a project is to
determine the effect of CSO treatment quality, the treatment sites
should be located upstream of any oscillating influence such as Lake
Michigan. Such a location would allow the effects of treatment on the
water quality of the receiving body to be characterized more easily, if
not more graphically.
Flow measurement devices should be installed or stage-flow relationships
should be determined at various points in the test reach of the Root
River.
Benthic productivity should be determined for each of the major areas of
the Root River in the City of Racine. Analysis of the biomass and
species of benthic organisms should also be performed. This research
activity should be conducted during spring, summer, and fall and should
only be done after a prolonged dry period. In each of the areas selected
for benthic studies, a determination
should be made during each season.
of the benthic oxygen demand
In the event that future monitoring work on any type of system utilizes
a constant recording/monitoring system as used in this project, it is
strongly advised that the data collection systems be multiplexed in such
a way that all data collected are compatible with computer systems.
10
-------
The use of manual labor to reduce data from a strip chart is very costly
and time consuming. A computer-compatible data system would allow the
investigator more time to work on the relationships of different events.
H. A system approach to monitoring river-lake interaction areas should
be developed. The Root River - Lake Michigan system, though somewhat
estuarine in nature, must have a monitoring system which takes into
account the subtle river - lake interactions. Such an approach would
ensure that future monitoring efforts on the Root River could make fuller
use 'of all data collected and also ensure that future monitoring efforts
on similar systems could develop more meaningful field data.
SECTION VI, STORM WATER MANAGEMENT MODEL • -
15. A section in the User's Manual (38) is needed to provide information to
calibrate the computed quantity and quality of the SWMM. This procedure
would allow the user to monitor one overflow point and then calibrate
the output so that predictions at other overflow points are accurate.
16. Expression of the coliform concentrations used in the SWMM should be at
the user's option; either membrane filter counts or MPN.
17. The Storage block should provide more options to the user within each
treatment device selected. These options should include variable
chemical dosage rates, screen areas, and design flow rates to give
better simulation at existing treatment units.
18. The evaluation of the other treatment options in"the Storage block is
needed using actual data from full-scale units.
19. The Receive block should be modified to accept smaller and less sophis-
ticated receiving waters such as small rivers and streams. Better
documentation, along with examples of actual application, is needed to
provide the user with a basis to begin the Receive block application.
11
-------
SECTION III
INTRODUCTION
I I 1-1 COMBINED SEWER OVERFLOW PROBLEM
During recent years the discharge of raw untreated sewage as a result of
combined sewer overflows (CSO) has become recognized as a serious pollution
problem. As determined in a 196? survey, approximately 29 percent of the
total sewered population of the United States is served by combined sewers.
Approximately three percent of the total annual sewage flow is discharged in
the overflow which contains as much as 95 percent of the sewage produced
during periods of rainfall (1).
The traditional solution to the problem is to provide separate sewer systems
for storm and sanitary flows. For older, established areas of the city, such
separation involves much expense and inconvenience. In addition, storm water
runoff itself can be highly polluted (2)(3)(*») (5). For these reasons
alternatives to sewer separation have been suggested, principally storage
and/or treatment of the overflow.
There appear to be three alternative methods, or combination of methods,
which can be utilized by a municipality to eliminate or minimize the pollution
associated with CSO:
1. Construction of larger interceptors and expansion of dry weather
treatment plant facilities.
2. Construction of holding tanks with provisions to pump the stored
wastewater back into the system after the overflow subsides.
3. Treatment and discharge of overflows.
In those instances where the location of sewage treatment plants and existing
interceptors make expansion economically attractive, the construction of
larger interceptors and the required accompanying enlargement and/or
modification of dry weather treatment facilities may provide a suitable
solution to the combined sewer overflow problem. This approach has been
successfully used in Kenosha, Wisconsin (6). However, in municipalities
that have widely scattered overflows or where treatment facilities are not
amenable to expansion due to process or area limitations, this approach does
not appear to be economically feasible. Normal design capacity for intercep-
tors is between 1.5 and 5.0 times the dry weather flow (7) (8). During a
storm, the flow in a combined sewer may increase 50 to 100 times the dry
1.2
-------
weather flow (9). If a complex system of interceptors has to be enlarged to
handle flows of this magnitude, or if the sewage treatment plant has to be
relocated because the existing one cannot be modified to treat the total
interceptor flow, construction costs may be prohibitive. Also, the problems
of public inconvenience and lost business, caused by the construction required,
must be considered.
The holding-tank concept has been and is being used as a method of handling
overflows. The disadvantages of the method include the cost of the tank
installation, the physical and economic limitations imposed by required
holding capacities, and the need for returning the entire flow to interceptor
systems for treatment after the storm subsides. In some locations, an "over-
loaded" condition exists at the treatment plant for several days following a
major storm and the overflow would have to be retained in holding basins until
the overloaded condition ceases. This delay would create health and odor
control problems. In addition, any runoff occurring after holding tank
capacity is reached, must be discharged untreated to receiving waters.
For these reasons, treatment and discharge of overflows near the point of
the overflow has generated considerable interest. The Storm and Combined
Sewer Section, Office of Research and Development, U. S. Environmental
Protection Agency (EPA), has sponsored a number of research, development,
and demonstration projects directed to establishing the feasibility of various
processes suitable for "on-site" treatment of combined overflows. Studies
conducted at Fort Smith, Arkansas showed that a system including a gyrating
screen, hydrocyclones, and a total flow pressurization dissolved-air flotation
unit effected an 84 percent removal of suspended solids and a 42 percent
reduction of BOD (9). An 18,900 cu m/day (5 mgd) screen!ng/dissoived-air
flotation (sdaf) pilot plant operated by this contractor achieved suspended
sol-ids and BOD removals of from 70 to 80 percent during the highly pollutional
"first flush" period (10). Based on this performance, the next logical step
was to determine the technical and economic feasibility of utilizing the
sdaf process for full-scale treatment of CSO.
In 19$8 the Wisconsin Department of Natural Resources issued orders to the
City of Racine, Wisconsin to reduce the pollution resulting from the
discharge of raw overflow from the combined sewers in a 284 hectare (700 acre)
area of the central city. The traditional approach at that time was to
separate the combined sewers into storm and sanitary sewers. Estimated costs
of sewer separation for the drainage area at the time were from $10 to $13
million, not including the substantial amount of inconvenience and lost busi-
ness that would result from construction. It was estimated that a system of
small satellite plants utilizing the sdaf treatment process could be installed
at the major overflow points along this stretch of the river for approximately
$4 million. Because such a system could save the City of Racine In excess
of $6 million in construction costs and would be more effective from a pollu-
tion control standpoint than sewer separation, the City of Racine, assisted
by the Environmental Sciences Division of Envirex Inc. submitted an
application for a demonstration project to the U.S. Fedeal Water Pollution
Control Administration (FWPCA). As conceived, the project was to establish
the cost/performance criteria for the full-scale application of the
sdaf process to treatment of combined sewer overflows as an alternative
13
-------
to sewer separation. In June, 1970 a grant offer was made by FWPCA to the
City of Racine. In addition to the federal grant, commitments were made by
the Wisconsin Department of Natural Resources and the City of Racine to
provide additional funds.
II1-2 PROJECT OBJECTIVES
The project had two objectives:
1. To evaluate the screening/dissolved-air flotation process developed
under FWPCA Contract No. 14-12-40 as an alternative to the physical
separation of those combined storm and sanitary sewers that
overflow into the last 6.4 km (4 mi) of the Root River in Racine,
Wisconsin.
2. To evaluate and modify (if required) the combined sewer mathematical
model developed under FWPCA Contract 14-12-502, "Storm Water
Pollution Control Management".
These overall objectives were expected to provide information on:
• The process adequacy of the treatment system as an alternative to
separation of combined sewers in a 284 ha (702 acre) area of
central Racine, Wisconsin.
• The cost/benefit relations to be expected from use of a treatment
system as opposed to sewer separation in the subject area.
• ^alidity of FWPCA combined-sewer mathematical model for application
to problems of any given area.
• Design, operation, and application criteria for the use of the sdaf
treatment method as an alternative to combined sewer separation in
any given area.
14
-------
SECTION IV
TREATMENT SITES
IV-1 PRELIMINARY STUDIES
Evaluation of Candidate Reaches of the Root River
The selection of the project test reach was preceded by division of the
Root River into a number of candidate reaches. Existing information and
results of field observations, sampling and analytical determinations were
used to make objective comparisons between and among the candidate reaches.
Through the use of sewer maps supplied by the City of Racine Engineering
Department, the major drainage areas along the Root River and within the
City of Racine were identified. These areas are identified by number in
Figure 1. For the major area numbers, the total area in hectares was
subdivided into the area served by combined sewers and the area served by
separate storm and sanitary sewers (Table 1).
Initially, the Root River within the City of Racine was divided Into four
possible candidate study reaches as shown in Figure 1. A preliminary survey
was made to determine if any of these candidate reaches were obviously
unsuitable for further consideration. The decision was reached to eliminate
Reach A for the following reasons:
1. Treatment at existing overflow locations in this reach would require
construction of screening/flotation units in the backyards of a
number of single family residences.
2. Within this reach, storm sewers serve 662.2 hectares (1636 acres)
and combined sewers serve only 77.8 hectares (192 acres). Even by
moving the reach upstream to eliminate strictly storm water discharges
Nos. 35 arid 36 (Figure 1) the separate sewered area would exceed the
combined sewered area by a factor larger than 2 to 1. Within the work
and budgeted scope of this project, it was felt that selection of
this reach would result in a disproportionate expenditure of project
funds for treatment of storm water only.
3. The branch in the river at Island Park would complicate the river
monitoring program because of the division of flow in the two channels.
Tfie judgment reached from these three considerations was to reject this
candidate reach and give it no further consideration for this demonstration
project.
15
-------
RACINE, WISCONSIN
STORM AND COMBINED
SEWER DRAINAGE AREAS
AND OVERFLOW LOCATIONS
JVERFLOW LOCATIONS / Jj I
Figure I. Drainage areas, combined and storm sewer overflows,
and division of river for selection of project test reach.
16
-------
TABLE 1. AREAS SERVED BY COMBINED SEWERS
AND SEPARATE SEWERS AS OF 1970
Area
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2k
25
26
27
28
29
30
Total
Hectares
712.8
211.4
17.2
19.0
29.6
26.1
2.96
14.8
14.8
16.8
5.3
29.8
23.7
2.37
18.4
2.7
11.9
2.37
1.9
116.2
8.9
3.0
3.6
5.9
68.2
44.6
502.2
1.0
105.0
127.6
area
(.acres)
1760.0
522.0
42.5
46.9
73.2
64.5
7.32
36.6
36.6
41.6
13.2
72.6
58.6
5.86
45.4
6.6
29.3
5.86
4.7
287.0.
21.9
7.33
8.8
14.65
168.5
110.0
1240.0
2.46
26.0
315.0
Comb i ned
Hectares
144.6
131.2
17.2
19,0
14.4
0.0
0.0
14.8
0.0
16.8
0.0
24.7
14.8
0.0
8.3
2.7
0.0
2.37
1.9
24.8
0.0
0.0
3.6
5.93
0.0
0.0
0.0
0.4
10.5
48.6
sewers
(acres)
357.0
324.0
42.5
46.9
35.6
0.0
0.0
36.6
0.0
41.6
0.0
61.0
36.6
0.0
20.4
6.6
0.0
5.86
4.7
61.2
0.0
0.0
8.8
14.65
0.0
0.0
0.0
1.0
26.0
120.0
Separate
Hectares
568.2
80.2
0.0
0.0
0.0
26.1
2.96
0.0
14.8
0.0
5.3
0.0
8.9
2.37
10.1
0.0
11.9
0.0
0.0
91.4
8.9
3.0
0.0
0.0
68.2
44.6
502.2
0.59
0.0
79.0
sewers
(acres)
1404.0
198.0
0.0
0.0
0.0
64.5
7.32
0.0
36.6
0.0
13.2
0.0
22.0
5.86
25.0
0.0
29.4
0.0
0.0
225.8
21.9
7.33
0.0
0.0
168.5
110.0
1240.98
1.46
0.0
195.0
17
-------
Preliminary Mathematical River Modeling - Because of the complexity of the
drainage systems and of the response of the Root River to the Inputs of
storm water runoff and combined sewer overflow, a modeling consultant
(Hydrosclence, Inc., Leonia, New Jersey) was retained to assist in the reach
selection process. The objective of the river modeling was to determine
which discharges should be treated to yield maximum water quality benefits
to the river.
The following information and procedures were used in the storm discharge
and river quality model:
1. The combined sewer and storm water outfalls were simulated as shown
in Figure 2.
2. Drainage areas, runoff coefficients, dry weather flows and sewer
capacities used are tabulated in Table 2.
3. Dally averages were used for dry weather flow (DWF); the program
made no distinction as to time of day in which the storm event
occurred.
A. Removal efficiencies through treatment devices were as follows:
BOD treated = 0.4 (BOD applied)
SS treated = 25 + 0.06 (SS applied)
5. The program simplified discharge quality variation by assuming either
of two conditions to exist:
a. First Flush Qua11ty Conditions - The highest concentration of
contaminants associated with initial flushing conditions are
assumed to be present the entire first hour of runoff.
Exception - If the antecedent storm ended 12 hours or less prior
to the start of the storm under study, the program assumes that
"first flush conditions" do not exist, and that concentrations
associated with sustained discharge quality will occur in the
first hour, as wel1.
b. Sustained Discharge Quality - The lower concentrations of
contaminants resulting from dilution with rainwater are assumed
to prevail from hour No. 2 throughout the duration of the
discharge event.
6. Table 3 summarizes the quality parameters in the program. In the
opinion of the consultant, there is reasonably good documentation
for using the values and for the use of a 12-hr period to define
"flush first" conditions. The values were considered reasonable
for use in this preliminary evaluation.
18
-------
J.33HJ.S H1XIS
asanoo
HAIdQ
J.334US
J.33dJ.S
NVOIHOIH 3MVH \
re
2!
o.
•o
0)
§
TO
3
••»
I/)
§
Dl
re
o
V)
CM
0)
3
O1
\
19
-------
UJ
o
o
a:
UJ
UJ
ec
o.
cc
£
o
z
o
a.
CM
«
la
OC
UJ
i
UJ
oe
ce
s
LJ
to
<
O
in
o Co
vO
O vO
(A
i
4-1
m
i_ u
0> 4>
§*J
•5
•o
4-> O
10 —
0) >4-
4J 0)
c >
3 O
20
-------
7. To permit Inclusion of stormwater runoff in the preliminary model,
it was assumed that stormwater runoff is similar in quality to
combined sewer overflows with respect to BOD and suspended solids,
the parameters analyzed in this preliminary evaluation.
8. Treatment efficiencies were assumed to apply equally for all
discharges. Treatment units were assumed to be able to handle all
flows at full efficiency.
9. The river water was assumed to have no BOD. Dissolved oxygen was
assumed to be at saturation as the river enters the test section.
The river quality model thus shows only the impact of the storm
generated discharges.
10. River discharge rates obviously increase in response to storms.
Data from the test period and superficial analyses of a few other
periods indicated a lag of a day or two before the full effect is
felt in the test area. These flows will be reduced ?n quality due
to runoff contamination upstream from the test area, although the
preliminary model assumed no contamination to be present.
11. Rainfall records used were those for the Milwaukee Airport Weather
Bureau Station. Annual rainfall distribution and storm patterns in
the project area are expected to be similar.
Three years of actual rainfall data were selected to obtain simulated storm
generated discharges and associated quantities of BOD and suspended solids
for locations indicated in Table k. The three years selected included a wet
year (I960), an average year (1961) and a dry year (1963). This information,
summarized in Table 4, snowed that overflow locations, C, D, L, and M were
responsible for the major portion of the total pollution loads discharged
into the entire reach.
The time variable model used to predict water quality was used to simulate
the river conditions during a 200-hr period in August 1965* The results
of this analysis have been plotted in Figure 3 for four critical stations.
The critical section of the river based on BOD and DO considerations in
Section 9 upstream from sections closer to the large combined discharges
from positions C, D, and E. The influence of dispersion near the mouth of
the river, other lake-influenced causes, low river velocities, and the large
ratio of overflow to river discharge rate were considered responsible for
this effect. The predicted location of minimum DO was confirmed by a survey
conducted on September 23, 1970. It did not appear, however, that the
magnitude of the sag would be sufficient to reduce the DO below 5-0 mg/1
which was established as the minimum permissible value in the State of
Wisconsin Water Quality Standards (11).
The model could not be readily adapted to handle coliform data. Therefore,
it was not possible to model this parameter. However, on October 28, 1970,
during a storm generated discharge event, a number of river samples from
various locations were collected and analyzed for fecal coliform. The
results of this survey are presented in Figure k.
21
-------
2-
CO
o
o
to
>
•h
UJ
z
w«
o
£
a:
o
CO
o
CO
a:
LU
b
o
a
•
*^
o
ll-
CC
UJ
o
a
UJ
I—
5
CO
^
LU
_J
co
men
•3" ^—
O
I in
O 1 -3"
S2
s
a i -3-
to-
rn
o
UJ 1
g*
(9 1 04
• 1 -3-
U. O4
in —
CO O
04 —
, *. S
i — J m
o\o
o. ^r
_°I £
z
00
1
-> 1
0°
f"^
m
t !"»•
men
u\\o
-31
u in — 6
c i ro —
49 *- o c <-
t O \O •— 4}
O {/) i— OC O
S. enSmco SRSm
CM. xO-y CM LA Cnp«*. CO *—
LA\O O CM CM.3- VO r-.
cM IA •— CM cno •— CA
CM
CM. CM LA VO -13* VO LA CO \0
CM -^CO LA •— O t*^ LA LA
ff\ ^i*"*co cn tAcn CTMA
CM LA •— CM cn— — m
CM
LA OCM cn cn OCM vom
O I^CM C^ CM -3*-* •— O
._ CO>CA f^ cn f r** m.^
CM\O — CM CMvO CMlA
— CM
^f P**^ vo cn CMVO o\o
fx. vOcM O CM Cn>— LA LA
CM P*"m Ln ^- cr\»— ^ »—
CAO LA CM cno cno
— in — -3-— CM
*~
o m\o •"• CM cniACOra
m LAlACO ^ LAm <^\CO
\J3 Oif>A'^*m U^CM CA LA
^co co cn cniA *zf ^~
CM^- — • r-*t**.^(v\
o o\o cn \o vop** co *«•
ff\ LA\O cn \o -T^ vo o
\o r*^\o CM o cocn r"*.m
trvcM CM IA OLA mco
— CM^T
m CM p«*. LA»— VOCM cnen
— —CM — -3-0 —
o vo cr\co r*.
Em in 1/1 in
— co in
"s Q 2 ^
i£ 1 1 s
OOOCO O\O '
u\o
— CM
^- o CM
cn
CM
— VO O -STL
LACO CM CO <
en\o
04 CO
Ovo
cncocr. o\o
^- ,— oo r**
o CM vo CM^* o cn enco
CM >—vO LA CM CM CM CM •—
-CT LACO -# — Vfi CO
Is* in
om
>o in
C-..CM r^-co o
CniAlAOn-i
— — CT\ CM
mcA cn o
— -3- LAIA
O P-H LAvO
— ^ cn —
— cn co f— co^r ^•»-
CM CM VO IA ^- P** CO JT
^- •— LA CM ^- CM CO Cn
LA PA CO -3*^ LAvO Cn«—
LA^- \O — •* vO LA rr —
r*. t>~ jf ,—- IAO co cn
— LA -^ O O cn
•"• CM — •"""
vo\o^r •* — f^o ^«
— 3- cniA —co -•
Cn CM CM VD CA CO LA LA CO O CO
• ««M»H««*l
co p-LAO — -a-co mcnoLA
3^- LALACOCn
^— *•* CO *^P^
•— -a-cop^cncMCAcM
JT o —-=r r^.co
Oen
•-CM
CO —
CM\O
o CA -3- cn cn\o vo LA
— CM mco —
v CM LA— CM LAr^vop^ocn
_• .w CM *~ cn CM r** -3f-d* r*«cn
o ^-co — — -a- — CA r-*co
r ^r rAco •— P*. •* \o
• ^* O^-VO^1 r*---
' .3- ^T — LAO 1
CM\O
CM CO
vO f A
s =
o o vo r*- LA mvo vo LA -3- cn
— CM LA
•3" cn LA cn.£f cnr** r**
4-1
i
:l
+J
u
s
M-
ra
' ^
c
is
.s
8
U-
(0
2 £
cn — 3
-e 8,8
in ° " B I
Emu «I
Sco' § t
>-
fl)
i
2-J
• raJ
14- ii-
en c t-
*SI
a>
*J
«*-
to
to
• 4J
i.2
3 §
o tn
22
-------
R1VERFLOW, cfs
8
l/l ID
3 c
o ***
— E
O D-
1. 0)
01 J=
— ID
O T3
4J 0)
I/I 4->
O
-a •—
(A
tu «)
C
3 •—
en
113333(1 N39AXO
23
-------
to
to
a:
o
o
LU
C
r--
cr,
CO
jQ
O
*J
u
o
I/)
§
o
0)
I/)
I
I
4->
10
s
2
4->
1
O
o
8
ID
U
0)
3
cn
u_
NVOIH3IW 3»V1 01
-------
Fecal coil form counts at all sampling positions exceeded the State of
Wisconsin Standard for partial body contact. The high coliform counts at
Memorial Drive and the municipal golf course bridge probably resulted from
large volumes of combined sewage overflowing into the Root River at overflow
Nos, 35 and 36 (Figure 1). At that time separation was in progress in areas
draining to these overflow locations and was expected to be completed before
May, 1971. A large reduction in coliform counts was expected when separa-
tion was completed.
Fecal coliform counts at Main and State Streets appear to be adversely
affected by combined sewage from overflow Nos. 1, 3, 7, and 8 (Figure 2).
This region of the Root River was used intensively by recreational boats
operating from facilities along the Root River and in the inner harbor.
Unintentional whole body contact with water in this region of the Root River
was an occasional experience by employees of the marinas and individuals
operating boats on the river. Personal communications with such individuals
confirmed subjectively the objective evidence of poor water quality in this
reach.
Based on the mathematical model estimates of quantities of BOD and suspended
solids discharged to the Root River, treatment of discharges in the lower
reach of the river would provide the maximum water quality benefit to the
river and harbor area.
The results of the modeling effort and the water quality survey were
summarized as follows:
• The minimum dissolved oxygen concentration occurs in Section 9.
However, this sag, under average flow conditions, is not expected to
reduce the dissolved oxygen concentration of the river below the
recommended standard of 5.0 mg/1.
• The fecal coliform concentration in the lower reach of the river
is greatly in excess of the State of Wisconsin watei quality
standards for partial body contact.
• Protection of Lake Michigan and the recreational uses of the
navigable portion of the river indicates treatment sites should be
located in the lower reach.
Add i t ibna1 Considerations - Factors that have an influence upon reach and
treatment site selection include land availability and cost, feasibility of
construction and relative cost of construction at alternative sites, and
for each site, the effect of water quality. Nine alternative combinations of
sites were evaluated objectively. The locations of these sites are shown
in Figure 5.
The criteria used for making objective comparisons are shown in Table 5. Num-
erical values on a scale of 0 to 4 (least favorable - 0, most favorable - A)
were assigned to land availability and land cost at each site. These two
factors together with the design discharge rates and BOD removal estimates
were then transformed to a scale of 0 to 10 (each number .was multiplied by 2.k
25
-------
RACINE, WISCONSIN
STORM AND COMBINED
SEWER DRAINAGE AREAS
AND OVERFLOW LOCATIONS
Figure 5- Alternative sites for CSO treatment plants.
26
-------
UJ
I-
co
UJ
_
1—
z
az
UJ
r-
_j
^C
u.
o
o
<:
1
UJ
Lf\
UJ
_J
CQ
*J
H S *
^
•_
on ""
-' i
~*
-^ TS —
O C *~
— • 10 «
H
•o
-•* *u a
" £
°u
0) c
Q; fll f
00
-» e c B
^O 3 (0 4)
— ' E 4-1 L.
— IA
«. Q
£
^
Q)
C
-Q 2
i! i
3g
t/»
g
S
*—>*
^S 5
L. l_
« «
15 J
U c
^"^
n
,
— u u
•— «J 4J
--'— Q
< C
o
• o
« •—
3 CM
00
J
•—I I*-
•—1 ff\
*ol
£ |
A
cn on
-*1 CM
1 A
o
I 3
o
•ol r-.
§1 ^
CD) O
•ol o
1 c^
Ol CM
n -
>1
ID O
1 *
3 =
ol
^
4) •>
c r*^
3 so*
§ CO
^
o
4-1
U\
<
—
(A
V "
•*
,
g -
o
o
o
CO
LA
CO
o
LA
f^
O
O
CO
CO
CM
0
o
o
V—
on
CO
r-.
i
>
z"
~~
CM
0
§
LA
VO
LA
-
CO
^.
CM
^
CM
r^
CM
CM
CO
V
oo
-3-
-a-
CO
o
LA
LA
O
_
•*
§
CM
LA
LA
ro
O
O
•t
^
vO PO
ft *k
LA CO
^^
2 Hr
on
CM
LA
0
8
CM
in
CM
PA'
CM
IA
CM
O
PO
V
*
CO
CO
o
VO
o
CO
o
o
CM
on
-a-
o
o
o
S
CM
f^
•—
VD
U\
O
*
««">
>
vo
O
O
O
*
CO
~~
r~
o
o
o
o
CO
C
*~
vO
O
0
r^
u-
o
c
o
4-1
CO
a.
u
Q.
4-1
(U •
1 I
111 —
c u
10 CJ
_•, §
0) CO L.
cn i- a)
•— JI (A
It v^
zs. 1
C VO -Q
-CO 0
(A ^-*
<*-
tn i_ o
O *^N. C
L. 4-1
o "~~* -—
*~ E
t_ •—
S ° *
tn .
4-> J3
•• co
Q) C » 4-< •—
u O w o co
3 -X u <0 >
cn 4-> 3 u- o
U. 4-1 JZ (0 o
C OJ C •— OQ
g u^«- u
-C ^ »*- O
cn co u o vo
cczL.Qj.cn
O O a. u c
— c —
*J -a a) co e
CO Q) JI 4J 3
u in 4J (A tn
o ro — u»
_i co a
-------
to transform the value to a point scale ranging from 0 to 10). Discharge
rate and BOD removal estimates for each site were assigned "points" on a
scale from 0 to 10. The value "10" was assigned to the site having the
largest value for each parameter. Other sites were assigned points on this
scale In direct proportion to the magnitude of each parameter with respect to
the site having the maximum value. A weighting factor was used to^compute
points for the discharge rate to give somewhat more weight to combined over-
flows and less to storm water only.
Point values were obtained for combinations of sites listed in Column 2,
Table 5, and totals were tabulated fn Column 10. Sites were combined into
alternatives listed in Column 1 by consideration of the total estimated cost
for Installations at each site and funds budgeted for construction of treat-
ment facilities. The treatment cost (Column II) is the total estimated cost
of constructing screening/dissolved-air flotation systems at the various sites
for each alternative. These estimates were used as a guide to select the
project test reach.
Three additional factors taken into consideration in selection of the test
reach were:
1
Overflow No. 2, 4, 3, and 13 are primarily the result of a severely
undersized interceptor sewer. This sewer is being relaid and
increased in size to eliminate overflows No. 2, 4, and 9- When
reconstruction of this sewer is completed, there will be no overflow
on the south side of the river from 4th Street, east of the lake.
Overflow No. 14 has been eliminated by plugging the discharge end
of the pipe.
With a treatment system located at Site
diverted from the interceptor so that no
No. 10, 11 and 15 for storms of intensity
or less.
II, sufficient flow can be
overflows occur at overflow
1.3 cm/hr (0.5 in./hr)
Selection of Project Tgst: Reagh - Evaluation of the river modeling and water
quality survcy, t he objectIve eva1uation table (Table 5) and the above
special consfderations lead to the conclusion that the test reach of river
should be from Lake Michigan upstream to overflows No. 7 and 8. ft was
concluded that overflows in this reach should be treated by screening/
dissolved-air flotation systems at Sites I and II and by a single screen for
storm water at Site IIA (Figure 5).
The preliminary studies, based on a precipitation rate of 1.17 cm/hr
(0.46 in./hr) and antecedent precipitation of 1.17 cm (0.46 in.) total during
the preceding 5 hr, predicted flow rates of 123-5 cu m/min (47 mgd)_at
Site II, 52.6 cu m/min (20 mgd) at Site 1, and 7.89 cu m/min (3 mgd) at
Site IIA (Table 5). Due to land restrictions, Sites I and II could not be
built large enough to handle these flows. Using all the available space at
the sites resulted in capacities of 37.1 cu m/min (14.1 mgd) at Site 1 and
116.7 cu m/min (44.4 mgd) at Site II. The capacity of Site IIA is 10.3
cu m/min (3.9 mgd).
28
-------
The minimum length of river in the selected test reach is 0.6 km (0.4 mi).
For storms of 1.3 cm/hr (0.5 in./hr) or less, without considering any ante-
cedent precipitation as was done for Table 5, the plant capacities were
expected to prevent any untreated discharge between Lake Michigan and Ontario
Streets, a distance of approximately 1.37 river km (0.85 river mi).
Pretreatment Storm Generated Discharge Studies
With selection of the project test reach, combined overflows No. 1,, 3, 6, 7,
and 8 and storm sewer discharge No. 5 were to be treated (see Figure 1).
Overflows No. 1 and 3 were to be treated at the site east of Main Street
(Site I) and overflows No. 5, 7 and 8 at the combined site west of Main
Street (Site II/IIA). Little or no overflow occurred at overflow No. 6 and
it was decided it would be possible to bulkhead the 20 cm (8 in.) discharge
sewer.
During 1971 a study program was conducted to determine the quality and
quantity characteristics of the discharge at the selected discharge points.
Description of Overflow Mechanisms - Overflow No. 1 located at Michigan
and Dodge Streets and overflov7 No. 3 located at Chatham and Dodge Streets
are separate and distinct overflow points. They were grouped together for
contributing combined sewer area because the Michigan and Dodge Streets
outfall serves as a relief overflow for the Chatham and Dodge Streets
combined sewer overflow area.
The combined sewer overflow mechanism at Michigan and Dodge Streets is shown
in Figure 6. It consists of a simple 30.5 cm (12 in.) high concrete weir
in a 91.4 cm (36 in.) interceptor sewer just downstream from the 30.5 cm
(12 in.) concrete relief interceptor which carries normal dry weather flow
west on Dodge Street. The Chatham and Dodge Streets overflow (No. 3) is the
relief for a 137.1 cm (54 in.) interceptor flowing south on Chatham Street
and into the 91.4 cm (36 in.) outfall sewer (Figure 7).
Overflow No. 5 has no retention mechanism as it is only a storm water
collection system servicing a 6.3 ha (15.5 acre) area. Storm water enters
the last manhole in the sequence through a 20.3 cm (8 in.) sewer and flows
into a 30.5 cm (12 in.) line serving as the discharge sewer to the Root
River (Figure 8).
Overflows No. 7 and 8 combined into a single discharge chamber and flow to
the river through two sfde-by-side 167.6 cm (66 in.) outfalls. A sketch of
overflows No. 7 and 8 is presented in Figure 9. Two separate combined sewer
interceptors enter the overflow box. One is a 228.6 cm (90 in.) trunk sewer
which serves the west and central portions of the combined sewer area and
the other is a 91.4 cm (36 in.) interceptor which serves the central area.
Flow entering the chamber drops into a 99.1 cm (39 in.) interceptor flowing
west on Dodge Street by means of 70.0 cm (24 in.) orifices as shown in
Figure 9.
29
-------
TO DRY WEATHER PLANT
COMBINED
SEWER
FLOW
0006E
OVERFLO
TO
RIVER
OVERFLOV
COMBINED
SEWER
FLOW
OVERFLOW DAM
Figure 6. Mlchlqan and Dodge Streets overflow mechanism, Discharge No. T.
30
-------
I
u
COMBINED
SEWER
FLOW
OOOGE
OVERFLOW
TO DRY
WEATHER PLANT
COMBINED
SEWER
FLOW
OVERFLOW DAM
30.5 CM. (12")
91.4 CM.
( 36")
OVERFLOW TO RIVER
DRY WEATHER FLOW
Figure 7'• Chatham and Dodge Streets overflow mechanism, Discharge No. 3-
-------
15.2 CM. (6")
STORM WATER
20.3 CM. (8")
15.2 CM . ( 6" )
STORM WATER
STORM
WATER
\
TO RIVER
Figure 8. Last manhole In sequence for storm sewer,
Discharge No. 5, Mafn and Dodge Streets.
-------
COMBINED SEWER FLOW
TO DRY
WEATHER
PLANT
99.0 CM.
<39")
OVERFLOW ORIFICES
91.4 CM .
(36")
DRY
WEATHER
FLOW
OVERFLOW DAMS
OVERFLOW TO RIVER
152 CM.
(60")
6I.O CM.
(24")
99.0 CM
SECTION "A A" (39-)
Figure 9. Wisconsin and Dodge Streets overflow
mechanism, Discharge No. 7 and 8.
33
-------
Flow Measurement and Sampling Program - Automatic sampling and depth
recording instrumentation was installed at discharges No. 1, 3» 5, 7 and 8.
Since overflows No. 7 and 8 entered a common chamber before discharge, one
composite sample was drawn from the mixing zone each time the sewer sampler
was activated. Flow was measured for these overflows at a common weir.
Sampling occurred at each location automatically during a discharge event.
The samplers were specially designed for this application and were later to
be used for sampling of the treatment processes. The interior of the sampler
is shown in Figure 10 and the sample program cycle is shown in Figure 11.
Each time before a sample is taken, the sample line is purged. During the
preliminary studies automatic startup and shutdown was controlled by a remote
switch located in the discharge chamber (Figure 12). Samples are collected
In one liter bottles to a maximum of 24 discrete samples. Sample interval
and the indexing of the sampling distributor arm is controlled by a timer
which allows a time variation on sampling interval selection of from 1 sample
every 5 min to 1 sample every 60 min. The sample sequence control for all
four of the discharge samplers was set at a ten minute interval. This
allowed for the collection of 24 discrete samples in a 4-hr discharge period.
The discrete samples collected for discharges No. 1, 3, and 5 were composited
by flow using the discharge record at each site. At overflows No. 7 and 8
the discrete samples were collected for individual analysis. Based upon the
duration of the discharge event, discrete samples were isolated every 10 min
for the first 30 min and at carefully selected time intervals thereafter for
individual analysis characterizing the discharge event. All collected
samples were placed in coolers for transportation back to the laboratory
for analysis.
Discharge rates and volumes were originally to be determined at each
location with a float-type liquid level recording instrument. Due to turbu-
lent flow during discharge events, this method of measurment was found
unsatisfactory. An attempt was made to control turbulence using a stilling
well but with limited success. Depth recording instruments operating on a
differential pressure principle were then procured. These instruments were
Installed as shown in Figure 13. By means of an aquarium pump, ambient air
was introduced into tubing which ran from the recorder to the dam in the
sewer. As flow (head) in the sewer increased, pressure built up within the
tubing in increasing increments. These pressure increases were converted to
depth readings and logged on circular charts. The chart was divided into 24
equal segments and rotated electrically at a one-cycle-per-day rate. The
charts were changed once every three days or after every discharge, which-
ever came first.
The depth recorder employed at overflows No. 7 and 8 also operated on the
basis of differential pressure but was mechanically dissimilar. Differen-
tial pressure was measured using a servo-manometer, converted to depth
reading, and recorded with a punched tape recorder. The samples for dis-
charges No. 1, 3, and 5 were composited by flow for analysis. Overflows
No. 7 and 8 were sampled discretely at predetermined time intervals. At
overflows No. 7 and 8, when the samples were taken discretely versus time,
the laboratory data were used to calculate a composite value utilizing the
34
-------
SAMPLER
ARM
Figure 10. Inside of discrete sampler.
35
-------
INDEX TO
FJ LL
i
\
\
\
\
\
\
\
\
PURGE CYCLE ,
2. MINUTES
1 TCR
BbTTLE
PROGRAM
. CYCLE
\l MIN.
\
2 TCR
NEXT
i
1
1
1
1
BOTTLE
i
}
1
I
1
I PROGRAM
/ INTERVAL PURGE CYCLE CYCLE
/2 *
i
3
60 MINUTES 2 MINUTES 1 MIN.
II II m ..,_ , ' • •
TCR 1 TCR 2 TCR
INTERVAL
3 TCR
PUMP OPERATING
PUMP OPERATING
COMPLETE CYCLE
I —H
Figure II. Program cycle for automatic sampler,
36
-------
Ul
UJ
CC
to
O
ID
15
"I
C 0)
O (A
— O
w c
tn
V
(0 «4-
O U
— 0
O. >
>• O
£
O)
IZ
37
-------
I. BELLOWS ASSEMBLY (PRESSURE CONVERTOR).
2. DEPTH RECORDING PEN AND CHART.
3. PRESSURE LINE TO BELLOWS.
k. CONTINUOUS GAS PRESSURE SOURCE
(AQUARIAN PUMP).
5. PRESSURE LINE TO SEWER.
6. INSTALLATION OF PRESSURE LINE ON
.WEIR IN SEWER.
Figure 13. Typical Installation of depth
recording Instrumentation.
38
-------
flows recorded for each corresponding overflow. The average discharge
characteristics for 1971 for discharges No. 1 and 3, 5, and 7 and 8 are
presented in Tables 6, 7, and 8, respectively. These values are compared to
the 1973 and 1974 discharge characteristics later in this section to deter-
mine if there was a significant difference between the 1971, and 1973-1974
data. In addition, averages were calculated for all discrete samples
collected at overflows No. 7 and 8; they are presented in Table 9 on a
quality versus time basis (minutes after start of overflow). The most
concentrated discharge was at overflow No. 3 followed closely by No. 1.
Discharge on a time basis shows the first flush occurring within 10 minutes
followed by a general decrease in concentration. The raw data from the
discharge quality analyses is presented in tabular form in Appendix IV-A,
Tables Al to A10.
TableJO presents a record of discharge volumes as recorded for each discharge
location. The turbulence difficulties discussed previously account for the
absence of some volume determinations apparent in the table. The discharge
volume at overflows No. 7 and 8 (Site It) was for the most part equal to or
greater than the combined volume-for overflows No. 1 and 3 (Site l). In
addition, the overflow rate was usually considerably greater for overflows
No. 7 and 8, although the duration of overflow was shorter than for overflows
No. 1 and 3. These measurements substantiated the need for more treatment
capacity at Site II than at Site I.
IV-2 SYSTEM DESIGN AND CONSTRUCTION
Description of Sites
The sites chosen for the location of the treatment pi ants are shown in Figure
14 at Sites I, II and HA. Sites I and II are screening/dissolved-air
flotation treatment facilities for the treatment of combined sewer overflow.
Site HA, adjacent to Site II, consists of a single rotating drum screen for
the treatment of storm water. The contributing areas for each site are given
in Table II and are also shown in Figure 14. To some extent the contributing
areas for Sites I and 11 overlap due to the interconnection of sewers as
shown in Figure 14. The contributing areas for each site given in Table II
indicate the site to which the largest volume of combined sewer overflow
discharges during an average hypothetical rainfall occurrence. Also, these
numbers represent only areas which overflow directly to the site. Upstream
areas which may contribute to the overflow through surcharge in an inter-
ceptor have not been included if there is an upstream overflow point.
Site I, located on the bank of the Root River approximately one block east of
the Main Street bridge, is shown in Figure 15. Sites 11 and MA are located
directly west of the Main Street bridge on the same parcel of land and also
on the bank of the river. Sites II and IIA are shown in Figure 16. All
three sites are located on the fringe of the downtown area and close to the
mouth of the river. '
Fortunately, sufficient land, a 1 ready owned by the City of Racine, was
vacant close to the sites of the outfalls. The availability of land was
39
-------
TABLE 6. 1971 OVERFLOW CHARACTERISTICS - SITE I
OVERFLOWS NO. 1 AND 3
Michigan and Dodge Street overflow (No. 1)
Parameter
BOD
TOC
Total solids
Suspended sol ids
Dissolved sol ids
Orthophosphate (as P)
Meana
concentration ,
mg/1
79
98
521
298
236
0.64
Fecal Coliform density, No/100 ml 10,300 2
Chatham and
Parameter
BOD
TOC
Total solids
Suspended solids
Dissolved sol ids
Orthophosphate (as P)
Fecal Colfform density, No./lOO
Range
38 - 152
2k - 203
218 -1062
38 - 596
102 - 563
0 - 1.20
,000 - 1,300,000
No. of
events
10
9
11
10
11
11
9
Dodge Street overflow (No. 3)
Mean3
concentration,
mg/1
212
194
943
669
314
2.07
ml 21,900
Range
140 - 406
110 - 340
560 -1649
366 -1506
230 - 544
1.10 - 3.10
1000 - 2,170,000
No.
events
9
9
10
9
10
10
9
a. Means given are arithmetic except for Fecal Coliform density which
is geometric.
40
-------
TABLE 7. 1971 OVERFLOW CHARACTERISTICS - SITE HA
OVERFLOW NO. 5
Parameter
BOD
TOC
Total sol Ids
Suspended sol ids
Dissolved sol ids
Orthophosphate (as P)
Fecal Col i form density
Mean3
concentration,
mg/1
39
51
608
445
HO
0.08
, No. /ml 23
a. Means given are arithmetic except for Fecal Col
is geometric.
TABLE 8.
Parameter
BOD
TOC
Total sol ids
Suspended sol ids
Dissolved sol ids
Orthophosphate (as P)
Fecal Col i form density
1971 OVERFLOW CHARACTERISTICS
OVERFLOW NO. 7-8
Mean3
concentration,
mg/1
212
238
646
669
274
0.75
, No./lOO ml 10100
Range
12 - 29
36 - 84
381 -1012
64 - 801
19 - 317
0 - 0,28
1 - 417
iform density
- SITE II
Range
48 - 282
36 - 184
280 -1348
181 - 847
100 - 501
0.21 - 3.17
No. of
events
6
5
6
6
6
6
5
which
No. of
events
9
9
11
9
11
11
540 - 143,000 9
a. Means given are arithmetic except for Fecal Coliform density which
is geometric.
-------
CO
1
t-«. O
«••
• !•_
O UJ
=? s
o ee
-J <
u_
o; «
UJ —
> f-. co
o on iu
v- O
sraTei
•M »-• fVf'
H- 3
t£. 0
• UJ O
co co o
> 2:
UJ O
tl § oc
-I O
< 1 U_
o"co co
O UJ UJ
u- •=>
f£ —)
UJ
•»>- •*
o *~*»
•"•
Cnuj
f-
UJ •—
-J CO
CO ^"^
«^ CO VO
• «- on
m cj CD r***
tf\ r^ CM o
*~ • *~
^^. "w^ ^^ ^^^
^^w ^^ ^** ^S»
O) D) O) ?)
E £ E E
in
•o
^z
o
••t" *«•
o m
a. o o 4->
1 O O O
0 CO K- |-
CM
CM
in
co
»—
f«^
o
CM
O
vO
CM
«^j-
on
VD
on
on
-a-
co
—
o>
•a
0) (A
-a -a
C **•*
a> _
Q. 0
in in
m
on
CM
*••»
on
CM
CO
PA
fs^
vO
CM
ff\
CO
•—
in O
O
-------
L_
^^
o
Ul
s:
i
o
>
1
li-
Cd
UJ
^>
O
~
~
I
o
»—
UJ
CO
^
CO
1.
•
o
z
2
,2
11-
i.
0)
0
in
O
z
£
(U
>
o
PI
o
z
I
^
>
0
-
Oi
z
J
IZI
0)
O
>.
0} 1 '
cn —
ID in
i- C
iu a)
> -w
< c
^~
(D
C
i
4J
ID
O
•«•
ID
cn
I-V
0
r_
X
E
U
"ro
cn
o
o
E
3
U
ID
cn
r\
o
"x
E
3
O
_
ID
cn
-\
o
X
E
3
0
,_
.C
•^
.-.
L.
r-
"^s.
E
O
C
g
CM SO
sO SO
in r***
o •—
co o
CM CM
i i
i i
i i
i §
i i
i i
i i
t t
i i
i i
i i
i i
i i
t i
i i
i t
i i
i i
cn ^3
CM — •
O O
.3- in
r-. CM
o o
-3- cn
CO CM
O 0
so -f
co r«-
o 6
r- —
x^ f«^
CM CM
r~- co
o
cn
CM
O
CS1
f—
"~
1
1
1
1
1
1
1
1
1
1
1
t
1
1
1
1
*
CO
CM
0
,4.
p^
O
in
CO
o
•2
CM
—
*^
O
OO
SO
fN^
^Hj
^*
CO
00
CO
^
CO
3
o
•
so
in
i
t
1
i
I
i
-3"
in
o
CM
0
CM
-a-
o
r^.
-a-
sO
CM
CM
in
o
CM
CO
^*~
—
*«^
CM
CM
CO
CM
r-.
CM
in
in
cn
—
cn
!^
oo
f*^
so
1
1
1
1
1
CO
-a-
0
_^-
•>
CM
^
0
^_
.3"
0
OO
so
0
£
*~*
—
^^
O
CM
cn
0
o
in
I-.
0
. -3-
co
CM
O
!^r
CO
-3-
SO
1
1
1
,
1
1
t
in
so
CM
o
o
o
i"~' •
0
-3-
CM
O
0
-3-
O
CM
0
"~
—
^s^
CM
01
in
CM
so
CM
so
cn
i
i
i
i
cn
in
0
cn
-3"
CM
CM
O
1— .
SO
SO
CM
§
O
0
CM
d
CM
CM
O
in
o
^
fO
*^
o
so
cn
CM
so
o
co
CO
CM
CO
•a-
in
-a-
CM
CM
in
o
-a-
• —
o
in
o
' o
cn
b n
CO
so
O
0
oo
o
so
CM
CM
_
^^
>^^
*_
-3-
CO
p^
cn
•*;
CM
i
i
i
i
i
CM
_
in
j-
cn
CM
o
so
CO
CO
cn
o
o
ro
CM
O
CM
0
0
o
^
OO
s^
_
-------
OJ
(U
01
C
J-l
3
C
O
O
-o
C
05
C
O
05
O
o
0>
ro
Q)
C
05
TO
P>
C
03
cc
O)
(/>
-------
MAIN STREET
BRIDGE STATION
FRONT WEST
CONTROL BUILDING
FRONT EAST
Figure 15. Site I treatment facility.
45
-------
Figure 16. Site II and MA treatment facilities.
-------
Figure 16 (continued)
-------
a major factor in the selection of these sites. A disadvantage of these
sites is the possible encroachment of Lake Michigan because of the proximity
of the sites to the Lake. Studies performed in 1970, showed that, at that
time, the encroachment of the Lake extended upstream to the State Street
bridge area, approximately 609 m (2,000 ft) upstream of the site. This
encroachment makes it difficult to determine the change in the water quality
of the river brought about by operation of the treatment systems. (Further
discussion of this problem can be found in Section V, ROOT RIVER MONITORING
STUDIES).
Process Equipment^ - The screening/dissolved-air flotation (sdaf) treatment
processes are identical at Sites I and II. Site HA employs a single ro-
tating drum screen for treatment of storm water.
in Figures 17 and 18.
Site schematics are shown
The treatment systems are designed for automatic startup, operation and
shutdown. Automatic operation insures that the system is deployed
immediately at the onset of an overflow regardless of the presence of an
operator. A flow sheet indicating the major supervisory control functions
Is presented in Figure 19.
The presence of a storm generated discharge is detected by a bubble tube
located in the screw pump wetwell. This signal initiates the sequence of
events shown in Figure 19. Shutdown of the system is initiated by a low
level signal in the wetwell. A cleanup cycle begins after system shutdown
to ensure proper operation during the next discharge. This cycle includes
backwashing of the screens and a final skimming of the flotation tansk.
At Sites I and II combined sewer overflow enters the wetwell and passes
through a mechanically cleaned bar screen to the spiral screw pump
(Figure 20). This bar screen has 1.9 cm (0.75 in.) openings between bars
and will remove only large size debris from the flow. After each storm
occurrence this debris is removed and hauled to a sanitary landfill for
disposal. The screw pump discharges into a channel leading to a Parshall
flume (Figures 21 and 22). Flow measurement in the flume provides a measure-
ment of plant flow and is used to provide flow-proportional chemical feed.
After flowing through the Parshall flume, the combined sewage enters the
drum screens (Figure 22-25).
At all three sites, 297 micron opening (50 mesh) screens are employed to
remove suspended matter in the flow. This removal is accomplished by mecha-
ically sieving or straining the overflow or storm water as it flows through
the screen. Each screen is mounted on a 2.4 m (8 ft) diameter drum. The drum
Is designed to be submerged to a depth of approximately 1.8 m (6 ft) and to
rotate slowly so that both straining and screen backwashing can take place
at the same time. When the headless or differential through the screen
becomes excessive, backwash water is drawn from the screen chamber by a
backwash water pump, pumped through a wet cyclone to remove grit, and
sprayed on the outer surface of the screen so as to flush solids from the
inner surface (Figure 26). These solids along with the backwash water are
collected in a trough and flow by gravity to the screw conveyor which
delivers them to the sludge holding tanks (Figure 27). A bypass weir was
48
-------
LU
LU
U
U
re
ID
2!
o
u
V
u
in
£
cr>
u.
-------
o:
MI
Ul
CO
a:
1
UJ
o
a:
UJ
z
* \
to oc s
Q ui {^
<7>
1.
1
u.
Q;
III
o
r
1
t
i
1
L
gui
, [=;>• — ,^_
HUI —,
toto 1
1 W
§ o ill
=~~| QCO *
W § l^^5
•» — — -•
N
CO M
3 ^f
Z
• • o
^ 10
^ CD
s
^ "\ "*N "^
s t t
g ^»^
2J3 < t UJUJ
05 & ^^^
om I f
. 1
uj
0^
i
1
V.
•-
u
tr
<>
(0
i
J
a.
s
a.
o:
1 E
o
l-
z
§
u.
u.
UJ
-^ »i
1 1
(
SUJ
iS
Q. QCO
O
a. ^
o
co-J
ujo:
UUJ
x>
UlO
^ n^
V
Q
><
c
i
*J
ID
0)
re
v>
4J
10
u-
o
u
V
•5
trt,
CO
£
5.0
-------
WET WELL LEVEL
RISE ACTUATES
SYSTEM
MERCHANTS
POLICE
ALARM
PARSHALL FLUME
RECORDER
ACTUATED
ADJUSTABLE
TIME DELAY
STODGE TANK
LEVEL RECORDER
ACTUATED
REX PROCESS
CONTROL PANEL
ACTUATED
SLUDGE DRAIN
VALVES CLOSE
K
DER
' SCREW PUMP
BAR SCREEN
FeCI3 PUMP
ACTUATED
1
LEVEL PROBE IN
ISCREEN PIT ACTUATES
PIPE DRAIN
VALVES CLOSE
SCREW CONVEYOR]
ON
1
SEWAGE I
_ER ON |
1
l_
1
PRESS FLOW
PUMP AND HEATER
NO. 1 ON
1
1
EFFLUENT SAMPLER
ON
j
2.O FOOT WATER
LEVEL IN FLOTATION
TANK NO."N"
1
FLOAT SWITCH
FSB-N ON
1
1 1
MAGNETROL
ON
TANK NO."N"
FULL
>
FLOAT SWITCH
FS-T-N ON
AIR SUPPLY TO
CONTROLLERS
ON
1
1
SKIMMER
ON-OFF TIMER
ON
ON OFF FUNCTIOIV
OF FS-B-N CONTROLl
BY FS-T-N
1
I
AIR SUP
TO PRESS
ON
1
IS
-ED
NOTES
BASIC FLOTATION TANK
START UP FUNCTIONS
FS FLOAT SWITCH
FS-B BOTTOM FLOAT SWITCH
FS-B-N BOTTOM FLOAT
SWITCH TANK NO.
FS-T-N TOP FLOAT SWITCH
TANK NO.
I FLOW TO TANK
NO."N"= 3.2 mgd
rPOLYELECTROLYTE]
PUMP ON
FLOW TO TANK
NO. N«l INTIATES
I CHLORINE FEED
I ON
FLOW TO TANK
NO. N*l AT 2.O ft
CHEMICAL DILUTION
H2O PUMP ON
BASIC FLOTATION
TANK START UP
FUNCTION REPEATS
I FLOW TO TANK
NO.N*I = 3.2 mgd
FLOW TO TANK
Nt2 INTIATES AND
CONTINUES THROUGH
BASIC START UP
FUNCTIONS
Figure 19. Automatic opera-
tion of Site I and II treat-
ment systems.
51
-------
Figure 20. Site II spiral screw pump and bar screen,
Figure 21. Site I Parshall flume
and transmitter.
1:'
-------
BACKWASH HEADER
Figure 22. Site II drum screens.
Figure 23. Site II drum screens with hold-down bars.
53
-------
Figure 24.
DRUM SCREEN
BYPASS WEIR
Site II drum screen influent channel.
Figure 25- Site MA drum screen.
-------
IE
o
LU
CC
O
in
1
o
<0
J3 •
0)
c c
V O
4) •—
u o
o >-
v> u
co c
• a.
vo E
CM 3
Q.
2!
•3
C71
-------
provided in the drum screen influent channel to bypass unscreened,
storm-generated discharge directly into the flotation system or effluent
channel (Site IIA) if all screens became clogged.
Dissolved-air flotation is the second process used to remove suspended
matter from the overflow. Flotation of the solids is accomplished by
introducing millions of microscopic air bubbles into the wastewater at
the bottom of the tank. As these bubbles rise, they attach themselves
to particles in suspension and carry them to the surface. Ferric chloride
and polymer are added to the wastewater to facilitate the coagulation of
particulate matter. Ferric chloride is added in the wetwell ahead of th
the
screw pump. Polymer is added in the drum screen effluent channel. 'Chlorine
is also added at this point for disinfection purposes.
A minimum of 20 percent of the design flow capacity is pressurized in the
pressurization tank (Figures 28 and 29). Operating pressure for the tank
Is maintained at approximately 2.8 kg/sq cm g (40 psig) by the air panel
controller (Figure 39) controlling the position of the downstream pressure
control valve (Figure 31). For the optimum saturation of air in water,
the pressurization tank should contain approximately 0.61 m (2 ft) of
water during operation. This level is maintained in the tank by means of
a level control (Figure 30) which is connected to a solenoid-operated bleed-
off valve.
The air-saturated or pressurized flow is blended with the remainder of the
raw flow in a mixing and flocculation zone at the influent end of each
flotation tank (Figure 32). Air bubbles are formed as the pressurized flow
Is released to atmospheric pressure. The raw flow from the drum screens is
distributed to the flotation tanks by a flow distribution channel (Figure 34).
This distribution channel directs the flow sequentially into the flotation
tanks In staged operation based on the rate of flow. For example, flow will
be directed to tank No. 2 only when tank No. I reaches 70 percent of design
overflow rate.
The floated sludge is periodically skimmed (timer controlled) from the top
of each tank by a flight of scrapers (Figure 35) and is deposited In the
screw conveyor which delivers it to the sludge holding tanks (Figures 27, 33,
36). These tanks are drained back to the interceptor sewer when the water
level In the sewer has decreased to the point where the tank contents can
be drained without causing an overflow at a point farther downstream in the
Interceptor. After the tanks are drained, they are cleansed by washing
down with a firehose using river water. The washdown water pump and header
system for Site II are shown in Figure 37. The Intended method of operation
is to drain and clean the entire system completely (sludge storage tanks
and flotation tanks) after each storm. However, the system is to be deployed
if a second overflow should occur before the draining and cleanup operations
are completed.
Each control building is divided into two sections; one section contains the
control panels, flowmeters, circuit breakers, chemical feed pumps, storage
tanks, and the air compressor and dryer (Figures 38-42). The other section,
56
-------
>-
— tf)
CTi
3
cn
UJ
o: uj
o >
_i
LU <
n. >
ol •
>-
1/1
0)
N
Ifl
2!
Q.
to
CO
CM
0>
3
cn
57
-------
Figure 30. Pressurlzatfon tank level control and sfght glass,
Figure 31. Pressure control valve in manhole ahead of flotation tank.
58
-------
Figure 32. Mixing zone behind perforated
influent baffle in flotation tank.
OVERFLOW WEIR-
SCREW CONVEYOR
SUBMERSIBLE PUMP
Figure 33- Site I sludge tank.
59
-------
Figure 3k. Site I flow distribution channel and screw conveyor.
Figure 35. Site II flotation tank scraper flights.
60
-------
SCREW CONVEYOR
•OVERFLOW WEIR
Figure 36. Site II sludge tank.
Figure 37« Site II washdown water pump.
61
-------
SITE II&IIA PLANT FLOW
RATIO CONTROLLER FOR
CHEMICAL FEED
SITE II WETWELL & FLOOD
GATE LEVEL RECORDER
SITE II&IIA PLANT BYPASS
SITE HA WETWELL &
SLUDGE TANK LEVEL
RECORDER
Figure 38. Site II and IIA supervisory control panel
62
-------
ELECTRICAL CONTROL PANEL
TANK PRESSURE CONTROLLER
AIR FLOW ROTOMETER
Figure 39. Site II control panel
63
-------
AIR DRYER FOR INSTRUMENT AIR
Figure 40. Site I air compressor
and air dryer.
CHLORINE SCREEN BACKWASH
WATER
Figure 41. Site II and HA recording flow meters
(pressurized flow and as indicated).
-------
CONTROLLER
FERRIC CHLORIDE PUMP
POLYMER PUMP
Figure 42. Site I chemical feed pumps.
65
-------
separated by a cement block wall, contains the chlorine storage tanks and
the chlorination equipment (Figure
F 1 ow Heas u remen t and ^Con t ro 1 - At all three sites the plant flow is measured
by means of ¥ "float in a Parshal 1 flume. The level measurement is converted
to a flow rate which is recorded on a circular chart. The volume is
continuously accumulated. Flow in excess of the plant capacity is bypassed
to the river at the wetwell bypass weir. This bypass is measured by
means of a bubble tube. The bypass rate is also recorded on a circular
chart and the total volume accumulated. The screen backwash water flow
is measured by means of a Venturi . The flow rate is recorded and the total
volume accumulated. For each flotation tank the pressurized flow rate is
measured by means of a Venturi and recorded on a circular chart. Sludge
storage tank level is measured using a bubble tube and is recorded on a
strip chart recorder. This level indicates a volume which is the total of
t;he screen backwash and floated sludge. The floated sludge volume is
determined by the difference. Level measurements are also made in the wet-
wells and ahead of the flood gate using bubble tubes and are recorded on
strip chart recorders.
At Site 11 two additional flow control devices have been installed in the
sewers. A large gate, referred to as the flood gate, was installed in the
22A cm (90 in.) trunk sewer bringing combined sewage to the Site 11 overflow
point to utilize in-sewer storage in the contributing area. The position of
this gate is controlled by the plant flow and by the water level upstream of
the gate to utilize sewer storage when the plant is operating at capacity but
at the same time to prevent any basement flooding.
The second flow control device is a sluice gate located on the 76 cm (30 in.)
interceptor downstream of the overflow. This gate closes down when flotation
tank No. 5 becomes full. Use of this. gate is intended to minimize over-
flows downstream of the interceptor by maximizing the treatment rate.
Samp 1 i _ng Eg u i pmen t - Permanent automatic samplers of a special design to
f ac 5 1 i tate the co 1 1 ect i on of storm generated discharge samples were installed
at the influent and effluent end of each treatment site (Figure 10). The
sampler is of the revolving arm type. Both a flexible impeller centrifugal
pump and a submersible sump pump were used with the samplers. A submersible
sump pump was used when the suction lift was greater than 1.8 m (6 ft) for
greater reliability. The program cycle for sampling is shown in Figure 11.
The time sequence can be adjusted as needed. These samplers are capable of
collecting 2*» discrete 1 liter samples on an adjustable time scale from
about once every two minutes to once every 60 min. Where discrete
samples are required, they can be obtained directly from the sampler at the
specific time interval desired. For those tests for which composite sampling
is desired, discrete samples can be collected at regular time intervals and
composited according to flow as recorded on the Parshal 1 flume recorder.
Design Criteria
The design criteria for the combined sewer overflow and storm water treatment
systems are given in Table 11. The Environmental Sciences Division operated
66
-------
Figure ^3. Site II chlorfnation equipment.
67
-------
TABLE 11. DESIGN CRITERIA FOR COMBINED
SEWER OVERFLOW AND STORM WATER TREATMENT SYSTEMS
Contributing area, hectares
acres
Design storm intensity, cm/hr
in./hr
Runoff coefficient (c)
In-sewer storage, cu m
gal .
Design treatment capacity, cu m/day
mgd
Site area, sq m
sq ft
Bar Screens
Channel width, m
ft
Channel depth," m
ft
Flow capacity, cu m/day
mgd
Maximum water depth, m
ft
Bar size, cm
in.
Bar spacing (opening), cm
In.
Travel , m/min
fpm
S'te 1
26.1
64.5
1.3
0.5
0.53
--
53,500
14.13
1,522
16,384
.91
3.0
2.44
8.0
53,500
14.1
1.80
5.90
0.95
3/8
1.90
3/4
2.1
7.0
Site II Site IIA
164.0 6.3
405.2 , 15.5
1.3 1.5
0.5 0.6
0.50 0.50
2,270
600,000
168,000 14,800
44.4 3.9
a (included
l>\l\ in Site II
34'263 area)
2.44
8.0
4.57
15.0
168,000
44.4
2.16
7.10
0.95
3/8
1.90
3/4
2.1
7.0
68
-------
TABLE 11. (continued)
Spiral Screw Pumps
Capacity, I /sec.
gpm
Total head, m
ft
Motor, kw
hp
Inlet fill depth, m
ft
Angle of repose
Minimum diameter torque tube, cm
in.
Minimum flight thickness, cm
in.
Minimum spiral diameter, cm
in.
Parshall Flumes
Throat width, cm
in.
Flow capacity, cu" m/day
mgd
Drum Screens
Number
Length, m
ft
Diameter, m
ft
Site 1
623
9,874
4.82
15.81
45
60-. 3
1.02
3.34
38°
76
30
0.64
1/4
183
72
61
24
53,400
14.1
2
2.1
7.0
2.4
7.87
Site II
1,960
31,066
6.02
19.75
186
249.30
1.51
4.97
30°
107
42
0.64
1/4
244
96
183
72
168,000
44.4
4
3.0
9.84
2.4
7.87
Site MA
170
2,694
4.78
15-68
15
20
0.60
1.97
38°
46
18
0.48
3/16
107
42
30
12
14,800
3.9
1
1.5
4.92
2.4
7.87
69
-------
TABLE 11. (continued)
Screen mesh
Opening size, microns
Screen backwash flow, I/sec.
gpm
Backwash pressure at nozzle,
kg/sq cm gauge
psi
Hydraulic loading rate, 1/min/sq m
gpm/sq ft
Maximum head loss capacity, cm
in.
Drum rotation speed, rpm
Flotation System
No. of tanks
Operating pressure in
pressurization tank, kg/sq cm
psi
Tank dimensions
length, m
ft
width, m
ft
depth, m
ft
Surface overflow rate. at maximum
design flow, 1/min/sq m
gpm/sq ft
Detention time at maximum design
flow, min.
Site 1
50
297
13.2
209.2
2.1
29.8
2037
50
61
24
5
3
2.8-3.5
40-50
16.5
54.2
5.56
18.25
2.4
8.0
135-7
3.33
18.2
S i te II
50
297
42.6
675.3
2.1
2 9. "8
2037
50
61
24
5
8
2.8-3.5
40-50
15-2
50.0
6.1
20.0
2.6
8.5
157.3
3.86
16.5
Site ilA
50
297
4.7
74.5
2.1
29.8
2037
50
61
24
3
70
-------
TABLE 11. (continued)
Pressurized flow, 1/min/tank
gpm/tank
Recycle rate,
Scraper travel
pe rcen t
, m/min
ft/mi n
Site 1
2,460
650
25
0.9
3
Site I!
2,914
770
25
0.9
3
Site IIA
Compressed Air System
Air delivery capacity, cu m/min
Sludge Storage
Capaci ty, cu m
cu ft
(1.5% of design flow for 3 hour
duration)
Chemicals
@ 4.9 kg/sq cm
cu ft/mi n
@ 70 psi
1.53
54.75
98
3,500
2.43
86.72
309
11,030
Maximum dosages at plant capacity
Ferric chloride (40% Fed,
solution, mg/1
Polyelectrolyte, mg/1
(100£ Nalcolyte 607 liquid)
Chlorine, mg/1
50
10
17.0
Chemical dilution water pump capacity,
I'/sec at m TDH 5.7/28
gpm at ft TDH 90/92
Chemical storage,
ferric chloride, liters
gal.
6,813
1,800
80
15
16.2
11.0/28
175/92
11,355
3,000
21
750
71
-------
TABLE 11. (continued)
Site I
Site I I
Site MA
Polyelectrolyte, liters
gal.
570
151
1,320
349
Washdown System
Washdown pump .
capacity
, I/sec at kg/sq cm
gpm at psi
6.9/3.5
110/50
6.9/3.5
110/50
a 19,000 cu m/day (5 mgd) screening/dissolved-air flotation pilot plant for
treatment of combined sewer overflows for approximately two years under EPA
Contract 14-12-40 (10). The design criteria for the Racine systems are based
on the experience gained in the operation of this pilot installation.
A diagram showing the design concept of the screening/dissolved-air flotation
treatment system is given in Figure 44. The various supporting systems of
the Site 1 and 11 treatment plants can be divided as follows:
Pumping System
Mechanical bar screen
Spiral screw pump
Screening System
Drum screens
Backwashing system
Flotation System
Pressurized flow pumps
Flotation tanks
Floated sludge removal
Chemical Addition System
Chem.ical storage
Metering pumps
Disinfection
Instrumentation
Measurement
Control
Flow Control Devices
Flood gate
Sluice gate
Washdown System
Pumping System - At each site a spiral screw pump is used to provide the
necessary head for gravity flow through the treatment units. This type of
pump has a number of advantages for pumping storm generated discharges. It
can operate over a wide range of suction head conditions and can pump at
variable flow rates with a constant speed drive. It is nonclogging and can
handle a wide variety of solids and debris without difficulty. The pump is
self-priming.
72
-------
H- OC
Z UJ
UJ >
u. o
ea — «/» i- ec —• eorsi— o as
V>
4J
C
u.
UJ
O
UJ
o- =3 a
c a. co
73
-------
A self-cleaning mechanical bar screen is utilized just upstream of the screw
pumps at Sites I and II. The bar screen has 1.90 cm (0.75 in.) openings
between bars and removes only large size debris from the flow which could
cause problems downstream in other units of the treatment system. Debris is
removed from the screen by a rake and is deposited in a hopper. The
screenings are disposed of in a sanitary landfill. The type bar screen used
appears to have sufficient strength to handle the sol,ids loadings
encountered.
Screening System" - The screening system consists of two major components,
the drum screen itself and the backwashing system. The critical design
parameters associated with the drum screen include hydraulic loading, solids
loading and the loss of head associated with the hydraulic loading.
Determination of the area of screen needed is based on the allowable
hydraulic and solids loadings. Depending on the waste characteristics as
well as on other factors, either of these variables may be limiting.
The hydraulic loading is a function of the rate of flow and area of wetted
screen. The wetted area varies only slightly as the headloss through the
screen changes during operation. The hydraulic loading is not affected by
rotational speed of"the drum. Generally, a drum screen can operate with up
to 70 percent of the screen surface submerged.
Maximum design headloss or differential across the screen was 70.0 cm
(27.6 in.). In practice, the drum screen was. operated at a maximum headloss
of from 30.5 to 40.6 cm (12 to 16 in.). The level switch controlling
backwashing of the screen was set to initiate backwashing as the upper
limits were approached. Drum rotation speed as well as the solids concentra-
tion in the discharge are determinants of headloss through the screen.
Drum rotation speed can be varied manually from 0 to 10 rpm. Rotation
speed at Sites I and II was generally between 4.5 and 7 rpm. Rotational
speed at Site IIA was between 3 and 4.5 rpm. Drum rotation is continuous
and rotational speed is not varied with-changihg flow rates.
Solids loading 5s directly proportional to solids removal efficiency and
inversely proportional to drum rotation speed and can be calculated using
the following equation:
Ls ^ RFs/rAe
where:
L - solids loading, kg/100 sq m
R = screen efficiency, percent
F s feed solids into screen, kg/min
r = drum rotation speed, rpm
A = effective surface area of screen, sq m
e
-------
The design solids loading was 6.8 kg of dry solids per 100 sq m (1.4 lb/100
sq ft) screen media. This design, figure, was based on findings at the pilot
plant installation. It was found that this solids loading produced a head-
loss of about 33 cm (13 in.) of water.
A sketch of the drum screen configuration is shown in Figure 45. The
inside of the drum is fitted with angle irons to pick up solids that do not
adhere to the screen. The screen cleaning system consists of a pump, a
header system and spray nozzles. Screened water is utilized to clean the
screens. A hopper inside the drum collects the screenings flushed from the
screen. The screenings are carried by a screw conveyor to the sludge
storage tank.
The design screen backwash water rate was 3.1 1/sec/m (15 gpm/ft). of drum
length. Spray nozzle pressure should be about 2.1 kg/sq cm g (30 pslg).
This rate was sufficient to clean a completely blinded screen. The spray
nozzles utilized should provide a low mechanical pressure on the screen
media but still provide good washing capability. The nozzles should have
as large an opening as possible consistent with a goad spray pattern. The
nozzles utilized have ellipitical openings with a minor-axis dimension of
0.64 cm (0.25 in.). The nozzles are positioned about 30.5 cm (12 in.) from
the drum surface. A small wet cyclone, 15.2 cm (6 in.) in diameter, is
utilized to trap solids. This material drains back into the drum screen
influent channel. The screening system has an automatic bypass feature.
When the headless capacity through the screen is exceeded, excess flow is
bypassed around the screens to the flotation tanks.
Flotation System - The flotation system consists of the.following maior
components: pressurized flow pumps, pressurlzation .tanks, flotation tanks,
and floated-siudge collector mechanisms.
The pressurized flow system is the heart of the dissolved-air flotation
process. It includes a pump, pressurization tanks, pressure reduction valve,
source of compressed air and suitable control systems. A schematic of the
pressurized flow system is shown in Figure 46.
The pressurized flow system Is designed to provide a rate of flow equal to
20 percent of the raw flow rate through the system at site capacity. The
pressurization tank should provide maximum air-water.interface to obtain
high rates of air solution. A tank without packing is recommended for treat-
ment of combined sewer overflows and was used in this application. A packed
tank would be more susceptible to plugging as a result of the solids present
In the wastewater. Tanks without packing are generally fitted with an inter-
nal baffle to promote a greater air-water interface. Nominal detention time
in the tank is generally about one minute. ' The tank should also be provided
with a method to control the water level since only about 20 percent of the
tank is full during operation. In this application a level control 0.61 m
(2 ft) from the tank bottom activated an air bleed-off valve if the water
level dropped to this point.
Design operating pressure in the system i-s controlled by an adjustable
pressure reduction valve. The valve is positioned by use of a pneumatic
75
-------
BACKWASH HEADER
LIFTI
ING FLIGHTS FOR SOLIDS
NOT ADHERING TO SCREEN
LEVEL SENSING
BACKWASH SWITCH
SCREEN BACKWASH
HEADER
-SCREENED SOLIDS
HOPPER
RAW WATER
SCREENED WATER
DISCHARGE TO
FLOTATION TANK
SCREENED WATER
Figure
SOLI 6s DISCHARGE TO
SLUDGE STORAGE TANK
Sketch of rotating drum screen
76
-------
0)
0>
N
r-x \n
2
a.
O
u
I
o>
U.
77
-------
controller which allows automatic control of the system pressure. The valve
provides for optimum bubble formation. An approximation of the correct
air flqwrate is 0..0283 standard cu m/min (Iscfm) for each 378 1/min (100 gpm)
of pressurized flow to the air solution tank.
The design size of the flotation tank is based principally upon American
Petroleum Institute (API) standards (12). Skimmers are provided in the
flotation tank to remove the scum. Bottom scrapers are sometimes utilized
in a flotation system to remove any sludge that settles to the tank bottom.
If 50 mesh or finer screening is used in the system, provision for bottom
scrapers may not be necessary because the small amount of sludge expected
can be removed while draining the tank between storms. If flotation is
utilized without screening, bottom scrapers will be required. Removal of
scum should be cyclically controlled by a timer or by sensing the level
of the sludge blanket. This allows sludge to be removed only when required
and minimizes the volume of scum requiring ultimate disposal.
Chemical Addition System - Chemicals are added to the wastewater at the
points shown in Figure 44. Approximately 40 percent liquid ferric chloride
(FeCU) is added ahead of the screw pump in the wetwel 1 . The polyelectrolyte
and chlorine are added downstream of the drum screens at the same point in
the drum screen effluent channel.
The chemical feed rate is automatically varied with incoming flow rate by a
current input/output signal between the Parshall flume transmitter .and the
chemical feeder. The ratio between input and output signal is adjusted to
obtain the desired feed rate.
Instrumentation - There are two basic types of instrumentation utilized in
the demonstration system: measurement and control. The parameters monitored
are indicated in Table 12.
An explanation of the normal sequence of operation illustrates the control
functions of the plant instrumentation.
Sites I and II - The Sites I and II treatment systems are designed for
automatic startup, operation, and shutdown. This ensures that the system
is deployed immediately at the onset of an overflow regardless of the
presence of an operator. The major supervisory control functions were
outlined in Figure 19 previously.
An overflow is detected by a level sensor in the screw pump wetwel 1.
When the liquid level reaches the critical elevation, three
activities are initiated:
1. A telemetry signal is sent to Merchants' Police indicating the
overflow condition. Merchants' Police then notifies the appro-
priate people who are to report to the site and oversee operation.
2. The Envirex Process Control Panel is actuated.
3. The Envirex Process Control Panel controls the following functions:
73
-------
2
O
—
LU
E
CC
h-
co
1
o
ce
o
o
z
1-
Ul
5
Q£
co
LU
r:
CO
CO
tlj
o
o
Q.
CM
UJ
CD
^£
^M
4-1
C
c
3
L.
(/)
C
^^
U
<0
1_
O.
"o
j_
4-1
s
N
**"
(D
o
(U
(0 *-
0 O
— u
-0 (U
c »-
I/)
(U
CO
c
E
(1)
«
V
(0
a
g
u.
XI
3 1
4-» x: co
a) u a.
E <0 u i_ t— 4-1 Q
3 i— 4) 4) (D —
_xi4J4J a)— 3 4>
i i /-» fl) fl) f"» 4iJ (/) jQ
3 E E >- 3 C 3
r— _Q (I) 4-> 4-1 re xiu-uicoxi
u— c c -u xi «— a) a. xi
TO <1) >ia: co^-'cx ca
X XX
XXX X
X X X X X X X
^» «*^ «^
«t A •» »•
z z z n z z _
_____ _ _
?-
V) 4J
ro c
3 ^
\/ ^ g
O O 4)
flj — t.
Xt M- 3 3
i/) O in
w c ^a — N 2: ro
ul xi o *«- c — —
UJ 3 •— 0) 4) 0)
4-14-1 3 CT
C C E W O 4) "O
nj m 3 4> — _i 4J 3
. — — s_ i_ x; 4> • —
a. Q. a a. o 3 co
CO
a.
o
^»,
V.
3
U
V.
Q)
X
(Q
0 O
3 —
-a <4-
o
o
X
—
QJ
XI
E
(D
U
4)
4>
U
CO
x:
o
4-1
'5
tn
4-1
TO
O
Lu
X
X
u
O
4-J
a.
4)
0
4)
C
•—
1
~®
••"•
U
4)
O
CO
79
-------
a. Closing the sludge drain and pipe drain valves and the flap
gate in Tank No. 1.
b. Energizing of the drum screens and backwash system.
4. The raw sewage sampler is energized and the interval timer started.
5. Following an adjustable 0 to 90 sec delay, the screw pump, bar
screen, and ferric chloride feed pump are energized.
Accomplishment of these functions will put the entire treatment system up to
and including the drum screens into operation.
Backwashing of the drum screens is controlled by a pressure switch in the
drum screen influent channel. When the headless through the screens
increases beyond a certain critical value, the pressure switch is tripped,
actuating the backwash pump. The screens are backwashed until the headless
drops, indicating the screens are clean.
Startup of the chemical feeders is initiated by an electrical conductance-
type level probe located in the drum screen chamber. Critical level at this
probe results in actuation of the chemical feed system:
Polyelectrolyte feed pump
Chlorinator
Chemical dilution water pump
Each of the air flotation tanks is equipped with two float switches which
control their operation. These switches are identified by the following
general notation:
FS-B-N and FS-T-N
where: FS-B =• bottom float switch, approximately 0.61 m (2 ft) from
tank bottom
FS-T - top float switch, located at effluent weir level
N » tank number in which the switch Is located
Thus FS-B-2 Is the bottom float switch in tank No. 2 and FS-T-2 is the top
float swtich in tank No. 2.
After the combined sewage passes through the drum screen and is dosed with
polyelectrolyte and chlorine, it flows through the inlet channel and begins
flowing Into tank No. 1. When the water level rises to the 0.61 m (2 ft)
level, FS-B-1 takes control and initiates four activities:
1. Energize pressurized flow pump and meter for tank No. 1
2. Energize magnetrol
3. Supply compressed air (dry) to controllers
4. Supply compressed air (wet) to pressurization tanks
80
-------
When Tank No. 1 becomes full and starts to overflow, float switch FS-T-1
actuates and three activities are initiated:
1. Energize effluent sampler
2. Energize skimmer "ON-OFF" timer
3. FS-T-1 now assumes control of the "ON-OFF" functions previously
controlled by FS-B-1
Flotation Tank No. 1 is now in complete operation.
When the flow rate to Tank No. 1 reaches 70% of the design flow, which Is
12,100 cu m/day (3.2 mgd), a portion of the flow is diverted to tank No. 2
through the weir splitting device in the inlet channel. The flow rate to
tank No. 1 simultaneously continues to increase to the maximum design flow
rate.
The startup of Tank No. 2 is initiated by FS-B-2 when the water level reaches
the 0.61 m (2 ft) level. From that point on, the control is exactly the same
as for tank No. 1.
Startup of all other tanks is according to the procedure outlined above,
i.e., flow to the tank starts when the flow rate to the previous tank
reaches 70% of design flow. The tanks continue to deploy sequentially until
the design flow rate of 53,400 cu m/day (14.1 mgd) at Site I and 168,100
cu m/day (kk.k mgd) at Site II is reached. Any flow in excess of the design
capacity is bypassed to the river. The rate and volume of flow which Is
bypassed is measured by a bubble tube and weir as described previously.
When the level sensor in the wetwell detects the end of the overflow, the
following functions occur:
Screw pump de-energized
Bar screen rake de-energized
Chemical feed equipment de-energized
Ferric chloride feeder
Chemical dilution water pump
Chlorinator
Polyelectrolyte feed pump
Influent and effluent samplers de-energized
A drop in level over the effluent weir in Tank No. 1 is detected by FS-T-1
and signals the end of flow through the treatment system. This signal
initiates four activities:
1. Pressurized flow pump, and air supply and controllers to tank
No. 1 de-energized
2. After a suitable time delay, the Tank No. 1 skimmers are
de-energized .
3. The drum screens undergo a final wash to ensure a clean media and
then are de-energized
81
-------
4. The screw conveyor is de-energized after the drum screen wash has
been completed.
The system is now shut down and ready for draining to the interceptor
sewer. A differential pressure switch located in the interceptor sewer is
utilized to detect when the level in the sewer has dropped to a point where
the tanks can be drained without causing an overflow at a point farther
downstream on the interceptor. When this preset level is reached the
sludge drain valve automatically opens and the sludge storage tanks drain
fay gravity to the interceptor. The flap gate between flotation tank No. 1
and the sludge tanks is equipped with a delay timer so as to prevent the
backflow of concentrated sludge into the flotation tanks. Thus a substantial
portion of the sludge is drained prior to the draining of the flotation tanks.
It is intended practice to completely drain and clean the entire system
(sludge storage tanks and flotation tanks) after each run. However, the
system can and will be deployed if a second overflow should occur before
the draining and cleanup operations are completed.
The differential pressure switch located in the interceptor is designed to
control the sludge drain valve. If an increase in flow in the interceptor
sewer occurs during the tank draining operation, the drain valve will close
to prevent downstream overflows.
Site MA - At Site IIA a simple screen is used for treatment of storm water.
As with Sites I and II, this installation is designed for completely
automatic startup and operation. A level sensor in the wetwel1 is used to
initiate the following functions:
Activate Merchants' Police Alarm
Energize screw pump
Energize automatic influent sampler
Energize rotating drum screen
Energize screen backwash flowmeter.
As at Sites I and II, backwashing of the drum screen is controlled by a
level probe In the drum screen influent channel. The backwash water is
conveyed into the sludge holding tank. The effluent sampler is energized
when the water level in the drum screen chamber trips a float switch.
Shutdown of the system equipment is Initiated by a low level signal in the
wetwel1. A cleanup cycle provides a final backwashing of the screen media
prior to shutdown.
Flow Control Devices - Two additional flow control devices are installed at
Site II to permit control of flow in the sewers. The sluice gate located in
the 76 cm (30 in.) interceptor sewer is intended to provide a means of
controlling the flow in that sewer. When the screw pump at Site II turns on,
this gate closes to a preset level to control the volume of sewage flow
downstream of the system. The use of this gate was designed to prevent the
overflow of combined sewage between the gate and the river crossing on the
south end of Ontario Street for all storms with an intensity of less than
82
-------
1.3 cm/hr (0.5 in./hr). When the screw pump is deactivated, the sluice gate
will open completely.
A flood gate was installed in the 228 cm (90 in.) combined sewer to enable the
use of the sewer for storage. The flood gate operation in the 228 cm (90 in.)
sewer is initially controlled by flow, as measured by the 182 cm (71.8 in.)
Parshall flume. The normal nonstorm setting of the gate is 1.2 m (3-94 ft)
above the sewer invert. At 151,400 cu m/day (40 mgd) the flood gate closes to
a preset level of 0.6 m (2 ft) above the 228 cm (90 in.) sewer invert. At
168,100 cu m/day (44.4 mgd) or maximum plant flow the flood gate closes to
a second position of 0.3 m (0.91 ft) above the sewer invert.
It is desired to maintain the level on the upstream side of the flood gate
between 2.4 m (7.87 ft) and 2.7 m (8.85 ft) above the sewer invert. When the
maximum flow of 168,100 cu m/day (44.4 mgd) is reached, control is transferred
to the sewer level measurement-control system.
During the gate travel motion, all of the sensing devices will be locked out
and not reinitiated until the gate has come to a complete stop. After the
gate stops its travel motion, a resensing will be made of the level in the
stilling well preceding the flood gate, and the gate will be driven 15 cm
(6 in.) closed, or 15 cm (5-9P in.) open, or maintained at its status quo
position if the level is between 2.4 m (7.87 ft) and 2.7 m (8.85 ft).
When the flow through the 182 cm (71-6 in.) Parshall flume decreases below
151,400 cu m/day (40 mgd), the flood gate is again controlled according to
the incoming plant 'flow and the gate opens 15 cm (5-9 in.) regardless of its
position, unless it is wide open. As flow drop below 113,600 cu m/day
(30 mgd), the gate opens to 1.2 m (3-94 ft) above sewer level (fully open).
The position of the gate is indicated on the Supervisory Control Panel.
Regardless of the flow rate to the treatment process or the level in the
screw pump wetwell, when a 2.7 m (8.85 ft) level is sensed in front of the
flood gate, the gate is automatically driven 15 cm (5.9 in.) further open.
Washdown System - After the tanks are drained following a treatment event,
there may be solid material remaining in the sludge tanks and on the bottom
of the flotation tanks. The floor of the sludge tanks and flotation tanks
are sloped (0.691) to drain the sewer and this makes it possible to wash
out the tanks with a firehose. A washdown pump and 60.8 m (200 ft) of
firehose have been provided at each site for this purpose.
In order to completely drain and clean out the sludge tank at Site I, it
was necessary to install a submersible sump pump In the bottom of the sludge
tank due to the surcharged condition of the sewers. To clean out the tanks,
the sludge drain valve to the sewer was closed and the tank water was pumped
with the sump pump into the sludge tank overflow weir discharging to the
sewer.
Design and Construction Costs
One of the objectives of this project is to develop detailed cost information
83
-------
(capital, operating, and maintenance costs) to establish cost/benefit re-
lationships for this method of treatment as compared to other treatment
techniques and/or sewer separation. This information would also be helpful
in evaluating the use of this treatment technique for other sites in Racine
as well as in other cities.
The design, land, construction, and equipment costs for the demonstration
systems are presented in detail in Appendix IV-B and summarized below.
SITE I
SITE II
Land
Equipment & construction
Engineering
Unit prices for the system costs maybe expressed as follows:
SITE I
SITE II
$/cu m/day of treatment capacity
($/mgd of treatment capacity)
based on design intensity of cm/hr
(of ?n./hr)
$/hectare of combined or storm
sewer area
($/acre of combined or storm
sewer area)
8.16
30,885
1.3
0.5
5.01
18,962
1.3
0.5
16,730
6,773
5,131
2,077
SITE IIA
$ 25,100
368,785
42, 714
$ 42,670
705,129
93,621
$ 1,180
21,224
2,597
$436,599 $841,420 $25,001
SITE IIA
1.69
6,396
1.5
0.5
3,968
1,606
IV-3 OPERATION AND'MAINTENANCE METHODOLOGY
Equipment Operation and Modifications
During the course of the project many of the operational parameters and
procedures discussed under SYSTEM DESIGN AND CONSTRUCTION had to be changed
or modified because of problems that were encountered in the operation of
the equipment. In fact, during .the first half of 1973, the common occur-
rence of equipment failures made it necessary to keep the sites shut down
until personnel arrived at the sites. Since the travel time was about 45
minutes, significant volumes of discharge were bypassed before the treatment
sites were manually turned on. It was not until November of 1973 that
major equipment deficiencies were overcome to the extent that the treatment
sites were placed in the automatic mode of operation.
Problems Encountered
The problems encountered during the two-year project may be briefly
classified into three basic types. Type 1 problems were sporadic in occur-
rence and were mainly equipment breakdown that were remedied by repair work.
-------
Examples of these problems were chemical feed-pump breakdowns, broken drive
chains and level recorders. Type 2 problems were related to the design of
the treatment units which could not be overcome and therefore, had to be
worked into the operational procedures. These problems Included
hydraulic overload of the drum screens when the plant flow reached or ex-
ceeded 80% of capacity and the Inability of the drum screen backwash system
to completely remove the solids build-up on the screen panels. Type 3.
problems also were inherent to the sites because of their design, however,
these problems were overcome by making major modifications in the operational
procedures. Typical examples of these problems were plant flow measurements
in excess of plant capacity. This overload was reduced by lowering the spiral
pump wetwell level. Other problems with the drum screens were alleviated
by manually operating the backwash pumps and constantly checking the position
of overlay panels and screens.
Effect of Problems on Operation ~ Many of the operational parameters were
modified during the course of the project because of operating problems en-
countered and the desire to obtain optimum treatment results. The process
operational parameters for each run are given in Appendix IV-C, Tables
C1-C45.
Six parameters were assumed to remain constant throughout the duration of
the project:
Backwash water pressure. The pressure was taken as 2.11 kg/sq cm g
(30.4 psig) for all three sites.
Drum screen depth of submergence was taken as 1.45 m (4.75 ft)
at Site I and 1.37 m (4.50 ft) at Sites II and MA, because at this
height the raw flow would begin flowing over the weir into the
drum screen bypass channel .
The drum screen wetted surface areas were assumed as constant because
they are calculated from the drum screen depths of submergence and
a constant effective screening area. The constants used for Sites I
and II were changed in 1974 because of the addition of the previously
discussed lateral support bars over the drum screen panels. These
support bars reduced the effective screening area and the wetted
surface areas correspondingly decreased. The values for 1973 runs
were 11.5, 31.7, and 4.0 sq m (124.0, 341,2 and 43.0 sq ft) for
Sites I, II and MA respectively. The 1974 values were 30.7, 29.5,
and 4.0 sq m (115-2, 317.5 and 43.0 sq ft). Site MA remained the
same because support bars were not used at this site.
1.
2.
3.
4.
The average headless through the drum screens were considered to
within the range stated in the design, 30-41 cm (12-T.8 in.)- for
Sites I and II and 30 cm (11.8 in.) for Site MA.
be
5. The type of polyelectrolyte used throughout the project was Nalcolyte
607, a liquid cationic polyelectrolyte with a density of 1.2 kg/1
(10.0 lb/gal.). The polyelectrolyte was obtained in 208 liter
85
-------
(55 gal-) drums from the Nalco Chemical Co., Chicago, Illinois
6. Sewage-treatment-grade ferric chloride, obtained from K.A. Steel
Chemicals, Inc., Lemont, Illinois, was used throughout the project.
The concentrated solution was approximately 39 percent ferric
chloride and had a density of 0.55 kg/1 (4*17 lb/gal.).
For each run,average values of the following parameters were determined:
Rainfall characteristics
Flow rates and volumes
Drum screen rotational speed
Drum screen hydraulic and solids loadings
Number of flotation tanks in operation
Flotation tank overflow and pressurized flow rates
Flotation tank detention times
Pressurization tank air pressure
Skimmer flight speeds
Chemical dosages
Effluent chlorine residual
Electrical power used.
The volumes bypassed were originally intended to be an indication of volumes
in excess of the treatment site capacities. However, the values were broken
into two parts (before and after the run, during the run) because large
volumes were being bypassed for two reasons other than exceeding the site
capacities:
1. Sites did not start up automatically or were not set to start
automatically and large volumes were bypassed before they were
started manually.
2. Sewer discharge was occurring up to three days after the rain-
fall event. This situation mainly occurred at Site I in the Spring.
It was considered impractical to keep the site operating that long
because after four to five hours of continuous overflow, the over-
flow pollutional characteristics decreased to a low value and
resulting treatment would be minimal.
The pressurizatlon tank air pressure varied slightly from run to run and
among the eleven individual tanks. The objective of the operating personnel
was to hold the tank pressure in the range of 2.81 to 3-53 kg/sq cm g
(40-50 psig).
The pressurized flow rate was not maintained at a preselected value. Instead
the flow rate was the flow produced by the pressure reduction valve in
maintaining an operating pressure of 2.81 to 3.53 kg/sq cm g (40-50 psig)
and a water level visible in the pressurization tank sightglass.
The drum screen backwash water volumes used during site operation were
theoretically dependent on the drum screen hydraulic and solids loadings
86
-------
However, a relationship between these variables cannot be determined because
the volumes are also greatly affected by how much plugging of the backwash
nozzles, pump, and wet cyclone occurred, and how wel1 the differentia!
pressure switch functioned in controlling the backwash pump.
The timing (on and off) and the flight speed of the flotation tank skimmers
were varied throughout the project in attempts to produce the best possible
results. The final values chosen were 2 min on and F5 min off with a flight
speed of 0.70 to 0.76 m/min (2.3 to 2.5 ft/min).
The drum screen rotational speeds were varred occasionally during the
project, The speed was usually increased durina a run when excessive
amounts of drum screen bypass were occurring. The Increased speed was used
to decrease the solids loadings on the screens.
The chemical dosages varied throughout the project. When operation of the
sites began in 1973, the selected dosages based on the results of combined
sewer overflow project, EPA Contract 14-12-40 at Hawley Road in Milwaukee,
Wisconsin were 20 mg/1 ferric chloride and 4 mg/1 polyelectrolyte (Nalco
607). After conducting bench scale tests on the combined sewer overflows
from Racine in December 1973, these values were changed to 40 mg/1 ferric
chloride and 2 mg/1 polyelectrolyte (Nalco 607). The results of these bench
scale tests are given in Table 13. Due to poor operating results, the
ferric chloride addition had already been increased to 40 mg/1 in July 1973
before the bench scale tests were conducted.
The mean values for each operational parameter and the range of the values
throughout the project duration are given in Tables 14, 15 and 16 for
Sites I, II and MA respectively.
From a comparison of the parameters there are some obvious operational
differences between the sites. The ranges and means for the volumes of
overflow treated
of SIte I I is
gpm) at Site
by Sites !. and II are similar even though the capacity
the
440
116.6 cu m/min (30,800 gpm) compared to 36.5 cu m/min (9,643)
On the average Site I ran twice as long as Site II and
maximum run
min at Site
was
I I .
2479 min.
Therefore
This time compares with a maximum run of only
the total volume of overflow at Site I was
nearly the same at Site II.
The volume bypassed during operation of Site II is very large despite the
site capacity of 116.6 cu m/min (30,800 gpm) because of the periods during
site operation when the bar screen rakes would jam resulting in buildup
of material on the bar screen. This material would block much of the plant
flow and cause large volumes to bypass the plant. If this problem had not
occurred, the much larger capacity of Site II would probably have resulted
in less average bypass volume during operation than was experience at
Site I.
It is interesting to note, however, that the solids loading on the screens
at Site II and MA were greater than the solids loadings at Site I. This
difference was caused by the higher average suspended solids concentrations
87
-------
TABLE 13. RESULTS OF BENCH SCALE
FLOCCULATION TESTS PERFORMED ON SITE II OVERFLOW
Date: December 4, 1973
Polymer
1A1
60?
C-31
905-N
Ferric chlori
dosage, mg/1
5
10
25
37.5
50
Type
An i on i c
Cationic
Cationic
Non i on i c
de
Polymer3
dosage, mg/1
0.25
0.50
0.75
1.00
1.00
1.00
5.00
0.25
0.50
0.75
1.00
Floe characteristics
Small, pin floe
Smal 1, pin floe
Better than with 10
Good
Best
Floe characteristics
Good
Better, reforms well
Slightly overdosed
Slightly overdosed
Fair
Good, does not reform well
Good, does not reform well
Good, poor reform
Good, poor reform
Better, reforms well
Overdosed
Using 50 mg/1 ferric chloride.
88
-------
TABLE 14. AVERAGE OPERATIONAL
PARAMETERS FOR SITE I
General
Overflow treated
Duration of run
Average flow rate
Overflow bypassed
During run
Before run
Power used
Drum Screens
Rotation speed
Backwash water volume
Backwash water pressure
Depth of submergence
Wetted surface area
Hydraul ic loading
Solids loading
Head loss
Flotation System
Overflow rate
Detention time
Pressurized flow rate
Pressurization tank
pressure
Skimmer time on
Skimmer time off
Skimmer flight speed
Chemical Addition
Ferric chloride
Polyelectrolyte (Nalco 607)
Chlorine
Effluent chlorine residual
Units
cu m
mtn
cu m/min
cu m
cu m
cu m
KWH
rpm
cu m
kg/sq cm
m
sq m
cu m/min/sq m
kg/ 1000 sq m
cm
cu m/mtn/sq m
hr
cu m/min
kg/sq cm
min
mtn
m/mtn
mg/1
mg/1
mg/1
mg/1
Mean
8,556
424
20.2
5,405
554
4,851
915
5.1
207
2.11
1.45
11.1
1.81
59.77
36
.092
0,50
3.21
2.97
2
12
0.75
32
2
10
2.0
Range
643-43,944
75- 2,479
8.0- 40.3
0-33,255
0-4,122
0-30,378
80- 4,400
4.0-7.0
27- 1,124
10,7-11.50
0.70- 3.50
8.40-224.11
30-41
.045-. 172
0.23-0.91
1.92-4.35
2.25-3.59
1-4
5-15
0.70-0,91
0-136
0-4
0-20
0-10.0
89
-------
TABLE 15. AVERAGE OPERATIONAL PARAMETERS
FOR SITE I I
General
Overflow treated
Duration of run
Average flow rate
Overflow bypassed
During run
Before and after run
Units
cu m
min
cu m/min
cu m
cu m
cu m
Mean
9,572
212
45.2
20,397
2,713
17,679
Range
984-43,376
30-440
11.4-116.4
0-137,774
0-24,034
0-133,232
Power used
Drum Screens
Rotation speed
Backwash water volume
Backwash water pressure
Depth of submergence
Wetted surface area
Hudraulic loading
Solids loading
Head loss
Flotation System
Overflow rate
Detention time
Pressurized flow rate
Pressurization tank
pressure
Skimmer time on
Skimmer time off
Skimmer flight speed
Chemical Addition
Ferric chloride
Polyelectrolyte
(Nalco 607)
Chlorine
Effluent chlorine residuals
residuals
KWH 946
rpm , 5.1
cu m 301
kg/sq cm 2.11
m 1-37
sq m 30.6
cu m/min/sq m 1.39
kg/1000 sq m 66.01
cm 36
cu m/inin/sq m .075
hr 0.76
cu m/min 2.59
kg/sq cm 2.76
min 3
min 12
m/m i n_ 0.72
mg/1 34
mg/1 2
mg/1 6
'mg/1 1.0
160-4,480
3.9-7-0
25-1,640
29.5-31.7
0.39-3.67
8.87-245-70
30-41
.019-.221
0.18-2.12
1.42-4.16
2.60-2.88
1-4
5-20
0.70-0.91
0-147
0-7
0-20
0-5-3
90
-------
TABLE 16, AVERAGE OPERATIONAL PARAMETERS
FOR SITE 11A
General
Discharge treated
Duration of run
Average flow rate
Discharge bypassed
During run
Before and after run
UnUs
cu m
min
cu m/min
cu m
cu m
cu m
Mean
160
29
5.5
0
0
0
Range
19-530
7-102
2.7-9.7
—
—
__
Drum Screen
Rotation speed
Backwash water volume
Backwash water pressure
Depth of submergence
Wetted surface area
Hydraul ic loading
Solids Loading
Head loss
rpm
cu m
kg/sq cm
m
sq m
cu m/min/sq m
kg/ 1000 sq m
cm
3-7
20
2.11
1.37
4.0
1.3**
71.40
30
3.0-4.5
0-91
—
--
—
0.68-2.42
12.11-227.78
—
at Site II and 1IA (280, 515, and 376 mg/1 for Sites I, II, and IIA
respectively).
It should also be noted at this time that the average flotation tank over-
flow rate was greater at Site I then at Site II, and the average flotation
tank detention time was less at Site I then at Site II. This, again, was
due to the fact that on the average, Site I ran at 54% of capacity and
Site 'I ran at only 33% of capacity. The effect of this situation on oper-
ational treatment results will be considered later in thi's section.
Chlorine addition and chlorine residual in the effluent, on the average,
are twice as great for Site I than for Site II because of the many runs
for which the chlorination equipment at Site II was inoperable.
Both Sites I and II have minimum chemical additions of zero because at one
time or another during the project, each particular chemical feed pump was
out of operation because of mechanical difficulties.
91
-------
Normal Maintenance Requirements
The development of the maintenance procedures follows from the many problems
encountered In the operation of the equipment. The purpose of this portion.
of the report is not to be an operation and maintenance manual, but to
establish basic maintenance routines that will be beneficial in eliminating
or minimizing many of the equipment problems that plagued the system. A
separate, detailed operation and maintenance manual was prepared for the
demonstration systems.
Bar Screens - The drive chains for the rakes should be oiled in the Spring
before the sites are put in operation. A sufficient supply of shear pins
should be kept on hand because they are needed when the rakes become jammed
and the pins shear during operation. Experience indicates that three to five
shear pins may be used before the rakes are freed.
If the rakes cannot be freed during operation, maintenance time will be
required to free them when sewer discharge ceases.
The wetwells, especially at Site 11, should be checked frequently for large
pieces of debris that could possibly cause jamming of the bar screen rakes.
After the sites are run for a combined sewer overflow event, the debris
removed from the flow by the bar screens must be disposed in a sanitary
landfill or by another environmentally acceptable method.
Spiral Screw
- The greaser reserve i r for each spiral screw pump should
be filled in the Spring before the sites are put in operation and checked
occasionally throughout the year.
Flow Monitoring Equipment - After each run all of the charts have to be
changed and the totalizer readings should be noted on both the chart taken
off and the new chart that is put on. A sufficient supply of all the
different types of recording charts should be kept on hand.
The air lines to all of the Venturi flowmeters should be drained frequently
to prevent plugging and after they are drained, the recording pen should
be set at zero.
Drum Screens - The drum streens were a major maintenance item throughout
the duration of the project. Sufficient quantities of repair parts should
be kept on hand. These include:
Screening fabric
Screen panel overlays
Seals
Links for drive chains
Bearings for speed reducers
0.95 x 2.5*» cm (3/8 x 1 in.) and
0.95 * 3.18 cm (3/8 x 1 1A in.) bolts
Fine stainless steel wire.
92
-------
The drum screen channels should be checked frequently. If they are lifting
up on the ends, a hole can be drilled through the end of the channel down
through the outer rim of the drum screen frame. The end of the channel can
then be bolted down in position.
Torn screening fabric and broken screen overlays should be replaced immedi-
ately to prevent large holes from developing that greatly reduce the effec-
tiveness of the screening process.
During operation of the drum screens, the seals should be checked. If
water is observed squirting out through the seal, it indicates that the seal
has popped out of position. If possible, the seal should be wired back into
position by drilling a small hole through the seal and the inside lip of
the drum screen. Stainless steel or another type of noncorrosive wire is
placed through the seal and the seal is secured to the lip of the drum
screen. If the seal is torn, however, a new seal should be installed.
Broken drive chains should be repaired immediately because running without
drum screen rotation puts excessive pressure on the drum screen panels.
The chains should be oiled in the Spring before system operation begins.
If the drum screen drive mechanism becomes excessively noisy, the bearings
for the speed reducer and the drive motor should be greased immediately.
If the noise persists, the bearing for the speed reducer should be replaced.
If this does not help, the bearing for the drive motor is probably worn and
the motor will have to be removed for repairs. Frequent greasing of the
drive motor and speed reducer bearings may prevent excessive wear from
occurring.
Screen Backwa sh Systern - The backwash pump should be greased frequently.
because of long operational times, especially at Site 1.
After every run, the backwash valving system should be changed so that all
the backwash flow is pumped through the collection hoppers to flush out
any remaining grit. After this flushing, the valves must be repositioned
so they are ready for the next run.
The backwash water collection piping system should also be flushed out
occasionally with a firehose to prevent it from becoming plugged with grit.
Pressurization System - The pressurized flow pumps should be kept greased
throughout the year.
The solenoid valves for the pressurization tanks' pressure bleed-offs
should be kept clean and the sightglasses for the tanks should be kept
clean also so that the water level in the tanks may be easily observed.
during site operation.
If the pressure bleed-off valves are opening too soon or too late, the
controlling mercury switch in the level control can be adjusted.
9.3
-------
Flotation System - All of the skimmer chains should be oiled and the screw
conveyor motor greased In the Spring before the sites are put in operation.
If grit begins to build up in the screw conveyor channel, it should be
flushed out immediately with a firehose. Otherwise, damage may be done
to the screw conveyor motor due to the excessive load caused by a large
amount of grit.
The high level switch at the effluent end of each flotation tank should be
checked frequently to make sure that it is free to move up and down and
positioned vertically.
Chemical Addition System - A sufficient supply of chemicals should be kept
on hand at all times.
The diffusers at the ends of the ferric chloride and polyelectrolyte feed
lines should be cleaned out occasionally with a small wire or nail to pre-
vent plugging. 'Plugged lines are indicated when the levels of the
chemicals in the storage tanks do not drop during site operation. There-
fore, tank level should be marked before and after each run to insure
correct operation of the chemical feed system.
The entire chlorine injection system should be checked frequently for
chlorine leaks by soaking a piece of cloth with aqueous ammonia and holding
a cloth near all sources of possible leaks (i.e. connections, valves). If
a white smoke appears to come off of the cloth there is a chlorine leak.
Before attempting to stop the leak, one man should be stationed outside the
chlorine room and the man working on the leak should put on the gas mask that
is provided at the site.
Sampling System - The samplers should be checked during dry weather periods
to make sure they are working correctly, and all hoses should be secured.
A good supply of clean bottles should be kept on hand. All hoses and
tubing should be kept clean.
Electrical and Pneumatic Controls - A supply of fuses and heaters should be
kept on hand because blown fuses and heaters were the frequent electrical
problems encountered in the operation of the sites.
A large supply of black 0.95 cm (0.375 in.) I.D. and 0.62 cm (0.25 in.)
i.D. PVC tubing and the corresponding fittings should be kept on hand
because of the frequent cracking of the air lines.
Rubber seats for the solenoid valves in the air control panel should also
be kept on hand because they occasionally crack resulting in air leaks
and loss of air pressure in the system.
All air leaks should be repaired as soon as possible.
Cleanup - Cleanup procedures required a majority of the maintenance time
at the sites.
-------
After a run, the sludge holding tanks and the flotation tanks should be
drained and the solids remaining in the sludge holding tanks washed out using
a f i rehose.
Cleaning deposited solids out of the flotation tanks requires that one man
wash the tanks with a firehose while one or two men push the solids to
the drain channel with shovels. By experience, it was found that during
the tank cleaning, an extra man should be available to keep the drain
channel clear or else solids will build up and prevent sufficient drainage.
This problem was severe at Site 11, where the solids must flow all the way
from tanks No. 8 through 1 to the sludge holding tanks and out the sludge
drain valve. Complete cleaning of all eleven tanks by four men requires
2 to 3 days. The cleaning should be done as soon as possible after draining
the tanks; otherwise odors will soon develop.
If flotation tank cleaning is not possible immediately, 0.6 to 0.9 m (2 to
3 ft) of water should be left in the tanks. This will be helpful in pre-
venting odor problems from developing. It must be remembered that the
solids increase in the bottom of the tanks with each succeeding run. There-
fore, it may be more economical to clean out the tanks frequently when the
volume of solids is small, than to wait until after many runs when the
volume of solids is much larger and harder to remove.
The washdown pump should be greased during cleanup operations because it
will be in constant use for 2 to 3 days.
It was also found that solids are deposited in the drum screen chamber dur-
ing site operation. The only way to remove these solids is for one man to
shovel them into a container and a second man to lift the container out
of the screen chamber, and dump the solids in the screw conveyor channel.
The screw conveyor can then carry them to the sludge holding tanks. Large
solids get in the chamber because of holes in screen panels, leaks through
the seals, and drum screen bypass flowing into the chamber, while fine
solids pass through the screen and accumulate in the chamber.
During freezing weather, part of the cleanup process must be the draining
of all pipes and pumps to prevent ice formation.
Maintenance Costs - Maintenance costs accounted for the largest portion
operating cost. Therefore, they will be discussed here in
being included in the total cost of treatment discussion
of the total
detail besides
later in this section.
The cost of operating and maintaining the sites was 6.08
-------
Table 17 presents a breakdown of the time spent on maintenance of the system
components as previously discussed. These values were determined from the
maintenance log for the months of May and June, 1974. During these two
months the only maintenance on the spiral screw pumps was to check the
grease reservoir and the only maintenance on the flotation system was to
check the operation of the high level switches. Both items took less than
15 min and the time spent was considered as 0 man-hours.
TABLE 17. BREAKDOWN OF TYPICAL MAINTENANCE
REQUIREMENTS FOR SDAF SYSTEM
Item
Bar screens
Screw pumps
Flow monitoring
Drum screens
Backwash system
Pressurization system
Flotation system
Chemical addition system
Samp 1 i ng eq u i pmen t
Electrical and pneumatic controls
Cleanup
Maintenance time
(manrhr)
12
0
86
36.5
8
13.5
0
50.5
44.5
41
226
Percent of
total
man-hr
2.3
0.0
16.6
7.0
1.5
2.6
0.0
9.7
8.6
7.9
43.8
Totals
518
100.0
The difficulty of removing deposited solids from the bottom of the flotation
tanks was responsible for cleanup being the major maintenance item. The
design of the system makes it improbable that the cleanup time can be
significantly reduced.
As the system is constructed, there is also no economical way to reduce
the maintenance time required for the bar screens, drum screens, chemical
addition systems, backwash system, pressurization system, and electrical
and pneumatic controls. These items plus cleanup account for 74.8% of the
time spent on maintenance and this time cannot be reduced without expensive
design changes.
Time spent on the flow monitoring equipment can be reduced because
extensive data collection is no longer necessary now that the study
96
-------
program is over.
Time spent on the sampling equipment could be greatly reduced by purchase of
new samplers or a complete overhaul of the old samplers. The existing sam-
plers are in a deteriorated condition after the two years of the project
and near the end of the project a Jot of time was spent keeping them opera-
tional .
It is believed that the maintenance costs for the bar screens, drum screens,
electrical and pneumatic controls, and cleanup (6\.0% of total time spent
on maintenance) could be reduced greatly for future screening/dissolved-air
flotation satellite plants by changing aspects of the design as recommended
in the following discussion.
Recommended Future Design Considerations
Eleven basic changes should be considered for future installation:
1. An alternative source of water for the chemical dilution and screen
backwash system.
2. More structural support for the drum screen panels.
3. A new design for the drum screen seals.
4. Complete separation of the drum screen bypass channel from the drum
screen chamber.
5. A method of removing accumulated solids from the drum screen chamber.
6. Rakes that will clean the bar screen without jamming.
7. Air lines that will not deteriorate and are easily accessible.
8. Placement of flumes such that accurate flow measurements may be
obtained.
9. An automated method of removing deposited solids from the bottom of
the flotation tanks.
10. Sufficient provisions for site drainage.
M. Different types of air controllers for better control of pressuriza-
tion tank pressures.
Process Water Source - The alternative source of water supply could be the
municipal water system, water drawn from the receiving body of water, or
final effluent. This water would be lower in solids concentrations than
the screened storm water and would be used for the chemical dilution
and screen backwash systems. The use of low solids water would eliminate
the problems of chlorine injector plugging, backwash pump and wet cyclone
97
-------
plugging, and the plugging of the backwash spray nozzles. With a different
source of process water, the wet cyclone may not be needed in the drum
screen backwash system.
Structural Support for the Drum Screen Panels - The lateral support bars
and bolts at the ends of the drum screen channels, or similar changes should
be incorporated into future drum screen designs because the channels them-
selves do not provide sufficient support to the drum screen panels. It
was found that the added lateral support reduced the amount of torn screen-
ing fabric and the number of broken screen panel overlays, resulting in a
reduction of maintenance time spent on the drum screens.
Drum Screen Seals - A new type of seal design is necessary so that the
seal will remain in position despite the water pressures experienced in
the drum screens. If possible, a new seal material that would increase the
life of the seal is desirable because replacement of a deteriorated seal
is a major maintenance project.
Drum Sc_reen_ Bypass Channel - Complete separation of the drum screen
bypass channel and the drum screen chamber is important in order to
reduce the amount of solids entering the drum screen chamber. This solids
reduction would help prevent chlorine injector plugging, backwash pump and
wet cyclone plugging, and backwash spray nozzle plugging if it is decided
to use the screened water for process purposes.
Solids in Drum Screen Chamber - Because it was found that solids will accumu-
late in the drum screen chamber, a method should be provided for removing
them. One possible method is to slope the floor of the chamber to a drain
valve. When cleaning is necessary, the drain could be opened and the
chamber washed out with a firehose.
Bar Screen Cleaning - Jamming of the bar screen rakes by debris coming
into the wetwel.l was a major problem, therefore, it is felt that a heavy duty
bar screen should be specified on future CSO treatment installations. At
the time of design, it was not expected that this type of debris (PVC pipe,
lumber, rubber hose) would be encountered in the overflow.
Air Lines - The air lines that are used should be resistant to deterior-
ation in direct sunlight or completely protected from the sunlight. They
should also be installed in such a way that they are easily accessible
in case they crack or become plugged.
Plant Flow Monitoring - It is very important to have accurate plant flow
measurement because many of the process operations are controlled by the
electrical signal transmitted from the plant flow measurement device. The
problems encountered with the Parshall fumes were due to installation of the
flumes too close to, the spiral screw pump discharges. The channel leading
from the discharge to the drum screens should have sufficient length to
provide nonturbulent, gravity flow before the flow is measured with a
Parshall flumes.
-------
Solids in the Flotation Tanks - An automatic system of cleaning the flotation
needed to reduce the amount of maintenance time spent in manual
solids from the tanks. Two possible methods are provision of
bottom scrapers or a steep floor slope to both the middle of the
the drain end of the tank. With the sloping floor a piping system
around the bottom edges of the tanks to, first, flush the
and, then, to the drain. For both methods, each tank
tanks is
removal of
mechanical
tank and to
could be installed
solids to the middle
should be equipped with its own drain, then the solids would not have to be
pushed from the tank to the sludge holding tank drain. As an alternative, a
screw conveyor or flight of scrapers could be provided in the drain channel
to prevent the deposition of solids.
Dra i nage - Site elevations should be set to provide sufficient slopes for
drainage of the thick sludges that are collected in the flotation tanks and
the sludge holding tanks. During the project, it was impossible to drain
these tanks without the use of a firehose.
Air Controllers - Air controllers that have two pressure sensing devices
should be specified for control of the pressurization tank pressure. One
device is a proportional band which signals the pressure reduction valve
to open or close when the pressure in the tank is above or below the desired
pressure by a certain percentage (i.e, +_ 5%) . The second device is an
automatic reset which will check the pressure once during a given time in-
terval and will open or close the pressure reduction valve in order to bring
the pressure in the tank to the desired level, i.e., 2.81 kg/sq cm g (kQ.5
psig). The air controllers used for this project had only the proportional
band device. This resulted in problems because if the proportional band was
set too sensitive the pressure reduction valve would rapidly open and close
resulting in cycling of the pressurized flow rate and if it was set too
high, sensitivity was lost and the range of pressure variation became too
great. With the automatic reset, the proportional band could have been set
high to prevent rapid cycling and yet guard against very rapid rises or
drops in pressure during the reset time interval. The automatic reset
would adjust the pressure to the desired level at the end of each preset
time interval .
IV-4 TREATMENT RESULTS
The demonstration sites were operated for as many storm-generated discharges
as possible from April 29, 1373 to September 30, 197^. During this period,
45 system runs were achieved, although all three sites were not run for all
45. Site I was run 45 times, Site II 33 times, and Site I IA 36 times.
There were two basic objectives of the project as conducted at the treatment
s i tes:
1. To treat the largest possible volume of storm-generated-d i scharge.
2. To achieve the best treatment possible of a storm-generated-d i scharge
utilizing the screening and screen i ng/d i ssolved-ai r flotation
processes.
99
-------
The evaluation of the project will, therefore, be based on the degree to
which these objectives were achieved and the final recommendations will
cover changes or additions that may be made in order to reach these desired
goals.
A system-run began when the wetwell level began to rise due to a storm-
generated-discharge. When the level reached the preset high level switch,
the switch began automatic site operation and an electrical signal was
transmitted to the office of Merchants' Police Inc., in Milwaukee, Wisconsin.
Merchants' Police, in turn, notified the personnel on-call that a high water-
level condition existed at the sites in Racine.
After this initial check, the personnel remained at the sites until the
end of the discharge and the sites automatically turned off. If any problems
were encountered during operation, every effort was made to correct them
so that the run could continue until the storm-generated-discharges ceased.
After the site cleanup cycle was completed, the personnel changed all
flow recording charts, recorded notes of any problems that occurred, and
collected all the samples. When these chores were completed, the sites
wereagain set for automatic operation in the event of another storm-generated-
discharge, and the personnel returned to Milwaukee.
For every run, the data generated throughout the project included:
• Rainfall characteristics
• Operational parameters for the process equipment
• Treatment results based on the following analyses of influent,
screened effluent, and final effluent samples:
Total BOD
Dissolved BOD
Total organic carbon (TOC)
Dissolved organic carbon (DOC)
Total sol ids
Suspended solids (SS)
Suspended volatile solids (SVS)
Total phosphorus (as P)
Fecal coliforms
pH
Effluent chlorine residual
For specific runs during the project, the following analyses were performed
as special tests:
Pesticide concentrations
Particle size distributions
Nitrogen series (organic N, NH_,
Chloride concentrations
Fecal streptococci concentrations
NO , and
100
-------
The operations parameters for the process equipment were discussed previously.
The operational parameters along with the rainfall characteristics are given
for each run in Appendix IV-C, Table Cl-C^S- The average values for the
operational parameters over the^project duration are presented in Tables 18,
19, and 20 for Sites 1,11 and IIA, respectively.
The results of the laboratory analyses for the influent, screened effluent,
and final effluent samples from each site and individual run results are
given in Appendix IV-D, Table Dl-DtyS.
This portion of the report will cover the storm-generated-discharge
characteristics and their relationship to the rainfall characteristics. The
values obtained will be compared with other published results. The operat-
ing efficiency of the screening, screening/dissolved-air flotation, and
chlorination processes will be discussed. Efficiency is discussed first in
terms of volumes actually treated by the systems and secondly in terms of
total volumes recorded. The latter volume includes the volume bypassed with-
out being treated and is a better indication of the impact of treatment on
the quality of the discharge to the Root River. The results of the
special tests conducted during the project will also be covered.
Data Col lection Methods
Flow Monitoring - At all three sites, flow in excess of the plant capacity
is bypassed to the Root River at the wetwel1 bypass weir. The volume
bypassed is measured by means of a bubble tube. The bypass rate is recorded
on a circular chart and separately totalized. The plant flow rate is measured
by a Parshall flume and is recorded on a circular chart and separately
totalized. The drum screen backwash water flow rate is measured by a
Venturi meter and is similarly recorded on a circular chart and separately
totalized. Sludge storage tank levels are recorded on a strip chart recorder.
This level indicates a volume which is the total of the backwash water and
the floated sludge. The floated sludge volume can be determined by the
difference.
As discussed previously, it was found during the course of the project that
two of the flow monitoring devices were not accurate: the bypass weir and
bubble tube for the Site I plant bypass, and the Parshall flume for Site II
plant flow. Measurements were taken for both and from the measured data,
correction equations were established:
Site I Bypass volume (gal.) =
0.266 (totalized vol., gal.) - 1690 (overflow time, min)
or
Site I Bypass volume (cu m) =
0.001 (totalized vol., gal.) - 6.397 (overflow time, min)
Site ii Plant flow (gpm) =
3-815 (chart reading, gpm) - 16890
101
-------
or
Site II Plant Flow (cu m/min) =
0. IM (Chart Reading, gpm) - 63.93
The Site II plant flow correction Is used only when the chart recorder Indi-
cates flows greater than 22.7 cu m/min (6000 gpm) because the Parshall
flume is accurate up to this point. Using the chart time scales, these
corrected flow rates were converted to volumes which were then added to the
accumulated volumes.
The corrections were applied to all the volumes obtained for Site I bypass
and Site II plant flow throughout the duration of the project.
Occasionally, the backwash water volume totalizer at Site I malfunctioned.
When this occurred, the total volume was calculated from the indicated
circular chart flow rates and the corresponding times.
Sampling - Permanent automatic samplers were used at the influent and
effluent end of each treatment site. The sampler is of the revolving arm
type. Both a flexible impeller centrifugal pump and a submersible sump
pump were used with the samplers. A submersible sump pump was required for
greater sample reliability when the suction lift was greater than 1.8
m (6 ft). The samplers are capable of collecting 24 discrete one liter
samples on an adjustable time scale from once every 2 minutes to once
every 60 minutes. For the length of the project, the sampling Interval was
set at 10 minutes.
When discrete samples were desired, they were obtained directly from the
sampler. When only a composite sample was required, the discrete samples
were collected at the 10-minute Intervals and composited according to the
flow rate as recorded on the plant flow chart.
Exceptions to this procedure occurred when an automatic sampler did not
operate correctly. In this case, grab samples were taken at regular time
intervals, usually once every 20 minutes. These samples were then composited
according to the plant flow rate.
Two methods of sampling the screened effluent (Site I and II) were used.
During 1973, periodic grab samples were taken from the drum screen chamber.
These samples were taken at 20-minute time intervals, if possible. After
the run, the samples were composited according to the plant flow chart. In
I97*t, automatic samplers were installed to sample the screened effluent.
These samplers were set to take a 500-ml sample once every 10 minutes. The
discrete samples were then composited according to the plant flow chart.
The screened effluent samplers did not start automatically, whereas the
influent and effluent samplers did. The former had to be started when
operating personnel arrived at the sites.
Drum screen backwash water and floated sludge samples were obtained manually.
One or two grab samples of each were taken whenever possible during a run.
102
-------
Given the physical layout of the systems, it was very difficult to obtain a
representative sample of the floated sludge-. It was found that the solids
concentration was highly dependent on how and when the sample was taken.
Similarly, it was very difficult to obtain a representative drum screen
backwash water sample.
Two factors caused difficulty in obtaining these two samples:
I. The long'periods of time when the backwash pump was on, especially
at the beginning of a run. Since the only sampling location was
at the discharge of the screw conveyor where both backwash water
and floated sludge discharge, personnel had to be present to obtain
a floated sludge sample when the backwash pump was off and the
floated sludge was being skimmed off the flotation tanks. For
a backwash water sample, they had to be present at the screw con-
veyor discharge when the backwash pump was on, although floated
sludge was not being skimmed off any of the tanks. Because the
backwash pump was usually on early in a run, obtaining of a
floated sludge sample was extremely difficult.
2. The rapid filling of the sludge holding tanks. This problem meant
that the samples had to be obtained in the first hour and a half
of operation, because when the sludge tanks were full, the screw
conveyor discharge was submerged. Since operational problems at the
sites usually occurred at the beginning of a run, time was not
available to obtain a sample at the right instant (backwash pump
off and tank scrapers on, or all scrapers off and backwash pump on)
before the sludge holding tanks were filled. Other contributing
factors were the time required to develop a good floated sludge
blanket and the fact that in many cases the sludge holding tanks
were full or partially full from a previous run.
The purpose of collecting samples of floated sludge and drum screen back-
wash was to provide the input necessary for performing a mass balance
for the treatment system. However, a mass balance could be performed by
using the influent and effluent solids concentrations. It was felt that
the latter method was the better one considering the lack of precision in
sampling. Collection of floated sludge and drum screen backwash water
samples was eliminated during 197^ operations. The values obtained during
1973 will be presented and briefly discussed later.
Sample Preservation - After each run, the samples were removed from the
samplers. When only composites were required, the samples were composited
according to the plant flow.chart to form one composite sample and the
composite sample bottle was then capped and labeled. When discrete samples
were desired, a portion of each discrete sample was taken for the compo-
site sample and then the composite and discretes were capped and labeled.
In addition, a portion of each effluent sample, S'ite I and II, both discrete
and composite, was placed in a sterilized bottle containing sodium
thiosulfate to neutralize the effluent of the chlorine. These samples were
used for the fecal (composite) and total (discrete) conform analyses.
10'3
-------
When all the samples were collected and labeled they were taken back to
Milwaukee. On arrival, the samples were immediately taken to the laboratory.
If it was before 11:00 pm, the sample analyses began immediately. If after
11:00 pm, the samples were refrigerated and the analyses were begun at 8:00
am the next morning. The samples were kept refrigerated at 4 C until all of
the analyses were completed and the results checked by the responsible
personnel.
Sample Analysis - All analyses were performed at the Milwaukee laboratory of
the Environmental Sciences Division with two exceptions:
I. Temperatures and chlorine residuals Jn the effluent were determined
by personnel at the treatment sites.
2. The special pesticide concentration tests were conducted by the
U.S. EPA, Region V Laboratory in Chicago, Illinois, and Llmnetics,
Inc., Milwaukee, Wisconsin.
The analytical methods utilized are referenced in Appendix IV-E,
Characteristics^ of Storm Generated Discharges
Combined Sewer Overflows - The volumes of combined sewer overflow arriving
at the sites are largely dependent on the characteristics of the rainfall
event. This relationship was looked at in great detail during the project,
especially from May to September, I971*- Every attempt was made to obtain
accurate raingage charts, plant flow charts, and plant bypass charts for
every rainfall event that caused combined sewer overflow. This meant that
the rain gage charts and the plant bypass charts were collected even when
the sites wre not operated.
A discussion of the raingages, the gaging network, and rainfall records
for the project duration are presented in Section V, ROOT RIVER MONITORING
STUDIES.
Combined sewer overflow hydrographs and rainfall hyetographs for selected
storms are presented and discussed in Section VI, STORM WATER MANAGEMENT
MODEL.
The runs for which all of the necessary data were available are given in
Table 18. The total overflow volumes are the volumes treated by the
sites plus any volumes bypassed. The rainfall characteristics are based on
the Theissen values obtained from the three raingage locations. The days
since the last system run, that Is, the days since the last rainfall large
enough to cause an overflow, are included to give an Indication of the
sewer capacity available before an overflow will occur.
Regression analyses (13) were run on the data in an attempt to develop re-
lationships between the overflow volumes and the rainfall characteristics.
The resulting correlation coefficients for the different relationships are
given in Table 19.
104
-------
to
to o
o >
tn o
Q£ U.
LU as:
H- LU
o >
< 0£.
Z LU
O 3
LU
_J U-
O L- «
I— i/>
o
3 C
t£. ~Z.
0)
a)
- O
a x
- o
bi "x
•— CM O LA O I xO
CM f-. CM
CM
to
O
— O —
S*J C CMOCMOO— OOO-— CM— O — — CMO — OOO
? —
X 0)
2: c j: —~ropA—ocMCMLA«>oo>i-3'cMcooor^5cMvb
1
u
^ O PA O O O CM ••"• ^ O ^™ O O ^— CM »™" O PA O O O
C OOOOOOOOOOOOOOOOOOOOO
«J
0) 0)
< «= . . ..
OOOOOOOOOOOOOOOOOOOOO
CMvOOCMPALAOCMOO — LAvO»— OCMvDOO — PAOvO
C vOCMLA — -a-JTCMLAr>.CMCMvOCMvOO—-a-PA— CMLA
OO — OOOOOOOOOOO— OOOOOO
fO TO
TO r*s vO »••• ^3 O^\ -«S" ••" CM OO PA ->3" OO PA CM 0"\ *••• CM O"\ PA •"• CM
U E LAvOCOPAO— LAPACALAvOsOLALALA^J-CM
— OPAO — •— o — — oo— o — CMO — ooo —
F— CM — cr>— irv
\o •— CM — r»- m o m t— cr> i — oo CM
CM — — r^i r^»
en
mcno —
oo u\ on — so r~.
-------
TABLE 19- CORRELATION COEFFICIENTS FOR REGRESSION
ANALYSES PERFORMED ON OVERFLOW VOLUMES
AND RAINFALL CHARACTERISTICS
Total
Total
Total
Independent variables
ralnfal 1
rainfall and average intensity
rainfall, average intensity
Total overflow volume
Site 1 Site 1 1
.770a .596a
•775a -629b
.791a .676 b
and maximum intensity
Total rainfall, average intensity,
maximum intensity, and days since last run
Total rainfall and maximum intensity
Total rainfall, maximum intensity,
and days since last run
Total rainfall and days since last run
.814
.779°
.810£
.806C
.688*
.621
.633°
a. Significant at the 99 percent confidence level
b. Significant at the 95 percent confidence level
106
-------
Correlations above the 99 percent confidence level were obtained for all
relationships to volumes at Site I.
The best correlation coefficient, 0.770, was obtained for the relationship
between the total overflow volume at Site I and the total rainfall. This
coefficient is not the largest when numerically compared to others. How-
ever, it is the most significant when the correlation coefficients are
compared to their corresponding critical values. The critical values are
based on the degrees of freedom for the sample group, and the number of in-
dependent variables being considered. The degrees of freedom equal the
sample size, n, minus the total number of variables, both independent and
dependent. The only exception is for a one-to-one relationship, then the
degrees of freedom equal n-I. Using the degrees of freedom and the number
of independent variables, the critical value; for the desired level of con-
fidence can be obtained from a statistical table (14) (17).
In this process, as more independent variables are introduced, the degrees
of freedom decrease and the critical values increase. Only slight increases
in the correlation coefficient when new variables are introduced are actually
reducing its overall significance. This is why the value of 0.770 for the
relationship between rainfall amount and total overflow is the most signi-
ficant.
The linear regression equation for the relationship is:
Vj = I5502R - 3270
where V| = total overflow volume at Site I, cu m
R = total rainfall, cm
This equation predicts that it requires 0.20 cm (0.08 in.) of rain to cause
the combined sewer to overflow at Site I, and that a volume of 3938 cu m
(1,0^0,000 gal.) can be expected at Site I for each 0.25 cm (0.10 in.) of
rain that falls after the first 0.20 cm (0.08 In.).
The application of the above criteria to the correlation coefficients
obtained for the relationships of the rainfall characteristics to the over-
flow volumes at Site I I revealed that the coefficient of 0.596, for the
relationship of total overflow volume to total rainfall was the most
significant.
The resulting linear regression equation is:
V2 = 3I770R - 8879
where V£ = total overflow volume at Site II, cu m
R = total rainfall, cm
This equation predicts that it requires 0.28 cm (O.I I in.) of rain to
cause the combined sewer to overflow at Site M, and that a volume of 8070
cu m (2,132,000 gal.) can be expected at Site M for each 0.25 cm (0.10 in.)
of rain that falls after the first 0.28 cm (O.I I In.).
107
-------
Because overflows are caused at Site I by 0.20 cm (0.08) and at Site M by
0.28 cm (O.I I in.), there were a few storms during the project where Site I
was operated to treat combined sewer overflows but no overflow occurred at
Site II.
In addition to the volumes of overflow generated by rainfall events, the
rate at which these overflow volumes arrive must be considered. These rates,
when greater than plant capacities, will cause plant bypass to the Root River.
Regression analyses were run to determine the relationships, if any, between
bypass occurring during site operation and the rainfall characteristics (in-
cluding days since the last overflow event). The runs for which the
necessary data was available are given in Table 20. The resulting correla-
tion coefficients for the regression analyses are given in Table 21 , below-
TABLE 21. CORRELATION COEFFICIENTS FOR REGRESSION ANALYSES
PERFORMED ON PLANT BYPASS VOLUMES DURING
OPERATION AND RAINFALL CHARACTERISTICS
Independent yartables
Plant bypass volume
Site ISite II
Total rainfal 1
Average intensity
Total rainfall and average intensity
0.616
0.306*
0.628
0.648
0.572
0.756
Total rainfall, average intensity,
and days since last run
Total rainfall, average intensity,
maximum intensity, and days since last run
Total rainfall, and maximum intensity
0.629
0.643
0.636
0.792
0.792
0.686
a. Not significant at the 99 percent level of confidence* all other values
are significant at the 99 percent level of confidence.
The most significant correlation coefficients are 0.616 for the relationship
between plant bypass volumes at Site I and total rainfall; and 0,756 for
the relationship of plant bypass volumes at Site II to total rainfall and
average rainfall intensity. Both these coefficients are significant
above the 99 percent confidence level (Ik).
The resulting linear regression equations are:
BVj - 671R - 517
BV2
I2737E - 5873
108
-------
™J
<
V)
o «/>
— LU
1— ac
CC 0
UJ ',>
f—
C_^ *J*J
2 <
< a,
re >•
O QJ
_-i UJ
LU u>
— l-»
- O 0>
TO C 4-1 4-J
a — in —
to ^^
o c
>- «
fy*
e
O T*
01 >
4-1 O1
— in c
co in —
(D l-
D. D
X* O
CQ
E
D C
^— 3
— O o£
4) cn
._ [/) ._
CO TJ i-
2L 3
>- O
CO
c •
3 O
cc. -z.
U
*N.
c
•—
\.
f~
o
,_
(—
*x.
c
L.
^^
U
^«
(J
• -a-
— o
cn x.
E
3
O
. -a-
— 0
10 "3
cn X
E
O
O f"\ -3" LA -3" -3" O vO LA O ^- f^ r*"! CM »^ CM O LA O ^O \O c*^i CM CM t^~ CM f^\ \O CM
— — ^ _ _ CM — — — — CM r- •— CM
O' CO -3* LA -3" -CJ" O >^ LA O f^ f^\ c^\ CM •— CM O LA O vO O fA CM CM f^^ CM rA vO ^^
— ———— CM— — — CM — —
CM CM CO CD ^"- -3" CM -3" i~~ CO •"" C5 CM O""v -3" CM -3" r^ CA O*\ LA LA CM fA *™" CM CM ^5" 0°^
OOO OOOOCMOCMOO— OOO — CM — O — — CM — O>— O OO
VO r*^ rA CO O -3* **O O >^ CO PA CO LA CT\ CM vD *^D CO PA CT\ r— ~- CM LA -— IA O"\ f*^ CA
LALAO — PA— LAT--3-OPA— OcMCMLAvOCTi-3-CMCOOOl — pn-3"CMvD OCM
O O CM O O ^~ O vO O LA O O CA CM ~~ O PA vO PA CN PA PA LA PA O PO O ~- CM
CM PA vO -3" vO CM CT\ PA CO PA CO -3" **O -3* PA vD VD CM LA LA ~- O \£> r^ LA ~— LA CO CT\
— — OOO— O — OPAOOOCM— — O— OO— CM — OO PA O O O
OOOOOOOOOOOOOOOOOOOOOOOOOOO OO
— PAxOOOOCMCMOLAOOLA— CAO-3-OCMCMCOCM — CTNPACACM O-3"
PAPA— — — PAcNPACMOOCM — — OrA-3- — PA— i — CM LA -31 — — r»- — CMCM
ooooooooooooooooooooooooooooo
CO CD vO f"^ PA -3" PA CM vO CD CM PA LA CD CM CO »~ • LA vO ^~* CD CM ^O LA OO ^~ PA CD VO
*~ CD vO LA LA CA CNJ *^O CM LA "•" -^" -^T CNJ LA P^« OsJ CM vO CM '•O CD ^~ CA ^^ PA *~- CM LA
O — OOOO — OO — OOOOOOOOOOO — OOOOO OO
^lAS^PAcA— LAMDCOPAO — LA. 5^" LASSLALALA^UcM.KrALAS
OCM— — — CMPA— OPAO — — O — — OO — O — CMOCM — OOO —
OOO
OOOOOOLACPiOLAOOOOO — OOCMOOLAOOOOOOO
VO ' o
CM
**o ^o -3" ^r -3* ^— CM co o r**-
OOOOOOCAI^OPAOOOO*— PAP^O-3'OO'— OLAOLAOOO
LA ^3" CD LA ^* CM LA CO *«O f^^*
PA — -3" — — CM -3" OCM
CM — CM
-------
where BVj = plant bypass volume at Sfte I, cu m
B\/2 = plant bypass volume at Site II, cu m
R = total rainfall, cm
E =» average rainfall intensity, cm/hr
When these regression analyses were begun, it was expected that a rainfall
intensity (average or maximum) would be necessary to develop a significant
relationship. This situation was the case at Site II, but not at Site I.
At Site I, the plant bypass during operation, like the total 'overflow volume,
was related only to the total rainfall.
An explanation of this difference is the size of the sewers contributing to
Site I. Through observations, it was found that these sewers were usually
in a surcharged condition during a combined sewer overflow event (this sur-
charge is also predicted by the math model). Even in dry weather, the sewers
were usually running full. In the surcharged condition, the rate of over-
flow could not change with corresponding changes in rainfall intensities
because the rainfall runoff could not reach the site unimpeded.
The total overflow volumes monitored from May to September 1974 are presented
in Table 22. The totals of 266,214 cu m (70,333,000 gal.) at Site I
and 743,677 cu m (196,480,000 gal.) at Site II are the volumes of combined
sewer overflow that would have discharged directly to the Root River had
no treatment been attempted. The treatment of these overflows wi11 be
discussed later in this section.
The large volume of overflow at Site II in September was caused by the inter-
ceptor sewer being blocked with debris for six days causing 171,000 cu m
(45,178,000 gal.) of dry weather flow to overflow.
It should be noted that combined sewer overflows occurred on 56 of a
possible 153 days, and during the wet months of May and June on 35 of 61
days. Rainfalls causing overflow did not necessarily occur this often;
instead, some large rainfalls caused the sites to experience overflow,
especially Site I, for 1 to 3 days after the end of the rain storm. This
overflow happened mainly during the wet months of May and June.
The values obtained for 1973 are separated from the values obtained for
1974 because during most of 1973, the systems were not set to start auto-
matically when an overflow began. Instead, operating personnel traveled
to the sites and manually placed the treatment systems in operation;
sampling began at this time. Due to travel time, the first 45 to 60
minutes of the overflows were missed. A'comparison of the two sets of
data will reveal the effect of this problem on the parameter concentra-
tions in the composite sample obtained for the runs.
As was expected, the quality of the combined- sewer overflows varied widely
from overflow to overflow. For each parameter listed, the value given
represents the arithmetic mean of composite samples for the number of runs
indicated (the only exception is the fecal coliforms; the value presented
is the geometric mean).
110
-------
LU
uj cr\
O Ul
{_> CQ
U. UJ
(/> UJ
UJ - CO
< 3:
o o
u_
• a:
4)
3
*O -
4
U
4)
O
4j
£
3
VH.
O
S4)
O-"-1
U
4)
0
"4-
0
V
n
o
^^
(U
m
£
3
O
*
f—
O
4-1
O
0
0
o
co
0
A
co
co
-a-
CTl
oo
CO
o
o
0
-a-
r-.
*—
•»
CO
-a-
-a-
Sr
CO
vO
o>
•»
^^
10
z:
o
o
0
o
O"\
•t
r**i.
-a-
vO
vO
IA
^_
CO
o
o
o
CM
•IWI
CM
**
0
CM
O
vO
f«**
vO
*—
4>
C
~?
O
o
o
0
VO
CM
*
oo
_^»
vO
CM
_
O
O
o
IA
pv.
O
P"
IA
CTl
O
CM
O*A
«..
f**
^,
^^
3
-j
0
O
O
o
CO
CO
M
IA
vO
LA
CM
CM
CM
O
O
o
CO
CO
CO
*
^*
CO
o-\
vD
-a-
4J
(A
3
O>
3
O
O
O
o
-a-
CO
•«
vO
-a-
f*^.
CJ^
CO
LA
•"•
O^
oo
OA
CO
LA
•—
O
L.
4)
E
4)
Q.
4)
to
o
o
o
o
CO
-a-
•»
vO
O\
1*^
r^*
vO
CO
-a-
o
o
o
JJ-
co
CO
^
o
r^
-a-
CM
vO
VO
CM
VO
LA
«I
^^
t-
O
t-
111
-------
Three important aspects of the quality characteristics will be omitted
here, but will be covered in detail in Section VI, STORM WATER MANAGEMENT
MODEL; they are:
1. Quality variations with time.
2. Discussion of the first flush phenomenon.
3. The relationship between quality characteristics and the rainfall
characteristics.
A comparison of the means for 1973 and 1974 at Site II shows that the
concentrations of the parameters more than doubled from 1973 to 1974; the
probable reason: a change in the overflow sampling. In 1973, sampling was
not begun until personnel arrived at the site. This delay meant that a
significant amount of overflow was missed including the highly polluted
"first flush" of the sewers, and therefore a final composite would not reflect
the true pollutional strength of the combined sewer overflow event but some
lower value. In 1974, automatic sampling began when the site turned on
automatically; therefore, the composite sample was representative of the
entire overflow. This sampling of the entire overflow event may explain the
higher composite sample mean quality concentrations for 1974. This indicated
that the pollutional strength of the first flush may be very significant at
Site II.
A similar change in sampling technique occurred at Site I, but it is not
reflected in a comparison of mean quality parameters between 1973 and 1974.
This observation tends to indicate that the first flush effect is not as
great at Site I as it" is at Site II.
Table 23 presents a comparison of the BOD, TOC and SS concentration means
obtained for 1971 and 1974 at Site I and Site II. The 1974 means at Site I
are compared to two 1971 overflow means because these two overflows both
contributed to Site I when site construction was completed. Using the "t"
statistic for the comparison of two means (15) a significant difference, at
the 95 percent level of confidence, was found between the parameters at
Site I in 197^ and the Chatham and Dodge Streets overflow In 1971. A
significant difference was also found between the BOD concentrations for
Site II In 1974 and the Site II overflow discharge point in 1971.
The difference between the 1974 overflow quality at Site I and the Chatham
and Dodge Streets overflow quality determined In 197' before site con-
struction could be due to three factors. The first factor is that a large
section of the Chatham Street combined-sewer-drainage area was separated
in 1972 which greatly reduced the volumes of storm water discharging at
the Chatham and Dodge Streets overflow point. This reduction of storm water
flow would also reduce the significance of any first flush. The second
factor is that 14 percent of the overflow at Chatham and Dodge Streets is
diverted to Site I I by a flow splitting device in the sewer. These two
factors would tend to decrease the impact of the high pollutional strength
Chatham and Dodge Streets overflow on the composite value for the Michigan
and Dodge plus Chatham and Dodge Streets overflows taken from the Site I
112
-------
TABLE 23. COMPARISON OF MEANS FOR 1971
AND 1974 OVERFLOW QUALITY
Mean, mg/1
Site Parameter
1 BOD
TOO
SS
II BOD
TOC
SS
a. Michigan and Dodge
b. Chatham and Dodge
1971
79*
2Ub
98a
194b
299a
669b
212
238
443
Streets overflow
Streets overflow
1974
93
95
266
no
122
661
(overflow No.
(overflow No.
Significant
difference at
35% Confidence
Level
no
yes
no
yes
no
yes
yes
no
no
1 , Figure 1) .
3, Figure I).
wetwell in 1974. The third factor, and possibly the most important, is the
length of the sampling periods
plugged or malfunctioned after
samples taken probably covered
overflow at Site I was sampled
were usually of long durations,
diluted by the low pollutional
continuous overflow.
During 1971, the automatic'samplers often
one or two hours of sampling; therefore, the
the first flush only. In 1974, the entire
and as previously discussed, the overflows
The composite sample, then, would be
strength samples taken after long periods of
The BOD concentration at Site II was significantly lower in 1974 than 1971.
In addition, there were changes, although not statistically significant, in
the TOC and SS concentrations. The TOC concentration decreased and the SS
concentration increased. The decrease in BOD and TOC concentrations might '
have been due to the same change in sampling techniques that affected the
mean concentrations at Site I between 1971 and 1974. This, however, does not
explain the increase from 1971 to 1974 in the mean SS concentrations.
Possible factors cited are: changes in the drainage area, changes in dry
weather flow rates, and/or addition of inorganic solids to the sewer system,
thereby increasing the SS concentrations but not the BOD and TOC
concentrations.
Table 24 gives a comparison of the BOD, TOC, SS and total phosphorus con-
centration means for Site I and Site II based on the sampling done in
1974. The only significant difference in the overflow quality between the
two sites is in the SS concentrations.
113
-------
TABLE 2k. COMPARISON OF MEANS FOR SITES I
AND II OVERFLOW QUALITY (197*0
Parameter
Mean, mg/1
Site I
Site II
Significant
difference at
95% Confidence
Level
BOD
TOC
SS
Total P
93
95
266
3.13
no
122
661
2.83
no
no
yes
no
The difference in SS concentrations is due to the difference in the con-
tributing areas and the conditions of the contributing sewers during dry
weather. The difference in the contributing areas would account for the
greater SS concentrations at Site 11 because the total gutter length for
the area that contributes to Site II is greater than 610 m (2000 ft) while
the total gutter length that contributes directly to Site I is only 38 m
(125 ft). A good indication of the effect of this difference is the per-
centage of the SS concentrations that are nonvolatile. On the average the
SS at Site M are 73.1 percent.nonvolatile and at Site I the SS are ^9.6
percent nonvolatile. Therefore, Site II is receiving large quantities of
inorganic solids, probably due to much larger volumes of surface runoff than
at Site I.
The dry weather flow in the sewer system is also important. As observed
during the project and predicted by the Storm Water Management Model, the
sewers contributing to Site I were usually running near capacity during
dry weather while the sewers contributing to Site II were running at rates
much lower than capacity. The rates in the Site I sewers would prevent
large deposition of solids during dry weather conditions and when wet
weather flow conditions began, the scouring effect of the increased
flows would be significantly less than at Site II. Conversely, this dif-
ference would indicate a much greater first flush pollutional load at
Site II than at Site I.
These differences in the contributing areas and combined sewer systems
would also be responsible for the larger pollutional strength of the
Site II overflow In 197' when it is compared to the pollutional strength
of the Michigan and Dodge Streets overflow discharge (Table 23).
Table 25 presents a comparison between overflow quality characteristics
(BOD, SS and total phosphorus) as found in Racine and 12 other cities
-------
TABLE 25. COMPARISON OF QUALITY OF COMBINED SEWER
OVERFLOWS FOR RACINE AND VARIOUS OTHER CITIES(9)a
City
Racine (Site I)
Racine (Site II)
Berkeley, CA
Brooklyn, NY
Bucyrus, OH
Cincinnati , OH
Des Moines, IA
Detroit, Ml
Kenosha, Wl
Milwaukee, Wl
Roanoke , VA
Sacramento, CA
San Francisco, CA
Washington, D.C.
Average for other
Years
1974
197^
1 968-69
1972
1968-69
1970
1968-69
1965
197P
1969
1969
1 968-69
1969-70
1969
12 cities
BOD,
mg/1
93
110
60
180
120
200
115
153
129
55
115
165
49
71
118
SS,
mg/1
266
661
100
1051
470
1100
295
274
458
244
78
125
68
622
407
Total
phosphorus ,
mg/1 as P
3.13
2.83
—
—
3-5
—
11.6
4.9
5.9
—
—
—
—
JU5.
5.4
a. Since different sampling methods, number of samples, and other
procedures were used, the data presented here are for general
comparison only.
where combined sewer overflow studies were conducted (16).
The BOD concentration in the Racine combined sewer overflow is similar to
the average value found in the 12 other studies. The SS concentration is
less than the average at Site I and more than the average at Site II. The
total phosphorus concentrations for both Site I and Site II are less than the
average for the five cities that determined total phosphorus during their
studies.
On the whole, the values obtained for the combined sewer overflow quality
in Racine are in agreement with the patterns established by the many studies
U5
-------
conducted on the quality.of combined sewer overflows.
Storm Sewer Discharges - The quality determinations for storm sewer dis-
charges are based on the data collected from Site MA. The wetwell for the
site is the discharge of a storm water collection system servicing a 6.3 ha
(15.6 acre) area.
The volumes of storm water discharge at Site MA will not be discussed in
detail because of the previously covered problems encountered with the
circular flow recording chart and the volume totalizer.
The means, ranges, and 95 percent confidence intervals for the measured
storm water quality parameters are given in Table 26. The quality charac-
teristics are based on the samples collected in 1973 and 1974 because
during the entire duration of the project, Site MA was set to start
automatically when a discharge occurred. Therefore, sampling of the dis-
charge began immediately and continued until the discharge ceased.
TABLE 26. 1973 AND 197** STORM WATER
CHARACTERISTICS - SITE IIA
Parameter
BOD
TOC
Suspended solids
Total phosphorus (as P)
Fecal coliform density
pH
Mean3
concentration
mg/1
15 ± 5
46 ± 13
376 ± 122
0.37 ± 0.12
S80b
__
Range,
mg/1
3-64
7 - 180
55 - 1 ,400
0.10 - 1.65
0-21 ,000
6.90 - 8.35
No. of
events
30
31
31
29
30
29
a. Limits given for mean are for 95 percent confidence interval
using "t" distribution. Means given are arithmetic except for
fecal coliform density which is geometric,
b. Number/100 ml.
The quality characteristics for the storm sewer discharge in 1971 were pre-
viously presented in Table 7. A comparison of the mean values using the
"tu statistic reveals that there is a significant decrease in the BOD con-
centrations between 1971 and 1973-74. Similar!ly, there is a decrease,
although not statistically significant, in the TOC and SS concentrations.
186
-------
Comparison of Quality Characteristics of
Storm Sewer Discharge for 1371 and 1973-74
Parameter
BOD
TOC
SS
1971 Mean,
mg/1
39
51
445
1973-74 Mean,
mg/1
15
46
376
Significant
at the 95
percent
confidence
level
Yes
Mo.
No
The decreases might be due to better street cleaning practices or other
changes that occurred in the storm water drainage area.
Table 27 gives a comparison of the storm sewer discharge quality determined
in Racine and the quality determined in nine other cities that conducted
storm sewer discharge studies (18).
TABLE 27. COMPARISON OF QUALITY OF STORM SEWER
DISCHARGES FROM RACINE AND VARIOUS OTHER CITIES (9)a
City
Racine, Wl
.Ann Arbor, Ml
Des Moines, IA
Los Angeles, CA
Madison, Wl
New Orleans, LA
Roanoke, VA
Sacramento, CA
Tulsa, OK
Wa$hington, D.C.
Average for 9 cities, not
Years
1973-7^
1965
J969
1967-68
1970-71
1967-69
1969
1968-69
1968-69
1969
incl . Racine
BOD,
mg/1
15
28
36
9
—
12
7
106
II
19
~2F
SS
mg/1
376
2080
505
1013
81
26
30
71
247
1697
~6l9"
a.
Since different sampling methods, number of samples, and other procedures
were used, the data presented here are for general comparison only.
The storm sewer discharge in Racine is less than the average of the pollu-
tional strengths found in the nine other cities. The values vary widely,
however, because they are dependent on many factors: land use, average days
between rainfalls, street sweeping practices, etc. Therefore, it is not
possible to definitely state why the Racine storm sewer discharge is lower
in pollutional strength than many other cities. It is probably a combination
of factors that affect the rainfall runoff and the deposition of solids in
the drainage area during dry weather.
117
-------
Regression analyses (13) were run on the storm sewer discharge quality
characteristics (BOD, SS, and fecal coliform concentrations) in an attempt
to relate them to the rainfall characteristics.
Over the two years of the project, Site IIA operated 32 times, but because
of operational problems or problems with the automatic samplers only 21
runs were used for these analyses. Table 28 lists the data obtained from
these 21 runs.
TABLE 28. RAINFALL CHARACTERISTICS AND
STORM SEWER DISCHARGE QUALITY (SITE MA)
Run
No.
5
6
13
14
16
17
18
19
20
21
22
25
27
28
29
30
31
32
34
35
39
BOD,
mg/1
6
3
5
6
5
10
15
64
16
22
4
19
15
9
2k
13
7
6
23
17
15
SS,
mg/1
55
105
159
423
76
383
514
156
102
171
128
223
70
160
841
574
137
121
660
299
304
Fecal
col tform,
no./lOO ml
1
1
17000
70
4800
3500
500
310
720
380
240
1
3300
330
450
390
410
470
800
710
13300
Average
rainfal 1
intensi tyy
cm/hr
0.46
0.25
1.68
0.33
0.33
0.58
0.16
0.10
0.16
0.30
0.22
0.20
0,20
0,10
0,15
0.61
0.33
0.40
0.30
0.12
0.41
in./hr
0.18
0,10
0.66
0.13
0.13
0.23
0.06
0.04
0.06
0.12
0.09
0.08
0.08
0.04
0,06
0.24
0,13
0.16
0.12
0.05
0.16
Maximum
rainfal 1
intensi ty
cm/hr
1.02
2.44
16.00
0.57
1.90
3.61
2,03
0.18
0.30
1.14
0,56
0.46
0.33
0.18
3.05
2.29
1,22
0.56
6.98
3,43
5,72
5n./hr
0.40
0.96
6.30
0.22
0.75
1.42
0,80
0.07
0,12
0.45
0,22
0,18
0,13
0.07
1.20
0,90
0.48
0.22
2.75
1,35
2.25
Days
since
last
run
17
2
42
3
3
4
15
15
4
15
21
5
8
3
3
2
1
2
5
1
2
118
-------
The BOD, SS, and fecal coliform concentrations are the values obtained from
the composite sample for the duration of the storm sewer discharge. The
number of days since the last run is the number since a rainfall of suffi"
cient volume or intensity occurred that could cause storm sewer discharge.
The correlation coefficients for the different relationships are given in
Table 29. The only statistically significant correlations (17) were obtained
between the fecal coliform concentrations and the rainfall characteristics,
and the most significant was the relationship between the fecal coliform
concentration and the maximum rainfall intensity, which resulted in a corre-
lation coefficient of 0.808. The resulting linear regression equation is:
N = 1017 - 3^3 =
where N = fecal coliform concentration, No./IQO ml
M «= maximum rainfal1 intensity, cm/hr
TABLE 29. CORRELATION COEFFICIENTS FOR REGRESSION ANALYSES
PERFORMED ON STORM SEV/ER DISCHARGE .QUALITY & RAINFALL CHARACTERISTICS
i Dependent
variable
BOD
BOD
BOD
BOD
BOD
Suspended
Suspended
Suspended
Suspended
Suspended
Fecal colt
Fecal coli
Feca 1 Co 1 i
Fecal coli
Fecal coli
Independent
variable (xl)
Average
Maximum
in tens 5
intensi
Days since last
solids
sol ids
sol ids
solids
sol ids
forms
f o rtris
forms
forms
forms
Average
Maximum
Average
Maximum
intensi
intensi
intensi
intensi
Days since last
Average
Maximum
Average
Maximum
Days si
Average
Maximum
intensi
intensi
intensi
in tens!
nee last
intens !
intensi
Independent
variable (*2)
ty
ty
run
ty Days since last run
ty Days since last run
ty
ty
run
ty Days since last run
ty Days since last run
ty
ty
run
ty Days since last run
ty Days since last run
Correlation
coefficient
-0
-0
0
0
0
-0
0
-0
0
0
0
0
0
0
0
.313
.138
.027
.kkk
.190
.061
.197
.265
.276
.i»56
.753
.808
.*»92
.753
.809
a
a
b
a
a
a. Significant at the 99 percent confidence level.
b. Significant at the 95 percent confidence level.
119
-------
After obtaining,this relationship, another regression analysis was made using
the log of the fecal conform concentration and the maximum rainfall in-
tensity. The resulting correlation coefficient for this relationship was
only 0.440.
Removal of Pollutants from Storm Generated Discharges
summaries of the
Efficiency of Drum Screens - Tables 30, 31, and 32 present
data on removal of pollutants by the drum screens for Sites I, II and IIA,
respectively. Since only screening was used at Site IIA, these removal
percentages and effluent characteristics are the total removal percentages
and final effluent for the process. Raw data on the operation of the drum
screens is presented in Appendix IV-C on a run-by-run basis. The results
of laboratory analyses on the screen effluent samples are presented on a
run-by-run basis in Appendix IV-D.
All data collected are used in the data summaries. It should be noted that
during 1973, the procedure was to collect samples manually at a predeter-
mined time interval, usually 20 min. During 1974, automatic samplers were
used. They were set to take a sample every 10 min and were started when
personnel arrived at the sites.
The average removals of SS from the combined sewer overflows by the drum
screens at Sites I and II were 32 and 36 percent, respectively. The per-
cent removals, however, varied widely from run to run and, therefore, the
arithmetic mean of the percent removals was not used.
The average hydraulic loading at Site I was 1.81 cu m/min/sq m (44.4
gpm/sq ft) with individual runs ranging from 0.70 to 3.50 (17.2 to 85.9).
The average solids loading was 59.8 kg/1,000 sq m (12.2 lb/1,000 sq ft)
with run values ranging from 8.40 to 22k (1.72 to 45.8). At Site II, the
average hydraulic loading was 1.39 cu m/m?n/sq m (34.1 gpm/sq ft) with a
range of 0.39 to 3.67 (9.6 to 90.1). The average solids loading was 66.0
kg/1,000 sq m (13.5 lb/1,000 sq ft) and the values ranged from 8.87 to 246
(1.8 to 50.3).
Although the solids loadings are similar at both sites, the hydraulic
loading at Site I is greater than at Site II. This may be the reason why
the screening process achieved only 32 percent SS removal at .Site I while
achieving 36 percent SS removal at Site II.
are presented in Table 33 for 1973. Screen
in 1974 for reasons previously explained.
Screen backwash characteristics
backwash samples were not taken
Based on the mean values for the samples collected, the backwash water at
Site I is 0.18 percent solids and the solids are 70 percent volatile. The
backwash water at Site II is 0.28 percent solids and the solids are 50
percent volatile. These results, as well as personal observations, indi-
cate that Site II was receiving much larger quantities of inorganic solids
than Site I during 1973. Sewer construction in the drainage a#ea may
account for the increased loadings.
120
-------
TABLE 30. SCREENED EFFLUENT CHARACTERISTICS AND AVERAGE
PERCENT REMOVALS BY THE SCREENS - SITE I
Screen effluent
mean concentration,8
Parameter rag/i Range
BOD
Dissolved BOD
TOC
Dissolved organic carbon
Suspended soli ds
Suspended volatile solids
66
21
61
20
191
66
± 9
± 5
± 8
± 4
± 33
± 26
16
5 -
14
8 -
18
2 -
- 169
36
- 147
39
- 630
142
Average
percent No. of
removal" events
28
0
38
5
32
55
40
18
39
19
40
11
TABLE 31. SCREENED EFFLUENT CHARACTERISTICS AND
AVERAGE PERCENT REMOVALS BY THE SCREENS - SITE II
Parameter
BOD
Dissolved BOD
TOC
Dissolved organic carbon
Suspended solids
Screen effluent
mean concent rat ion,a
rag/1
50 ± 13
26 ± 10
50 ± 11
23 ± 6
329 ± 101
Range
12 ± 131
2-82
10 - 115
5 - 61
65 - 1038
Average
percent^
remova 1
hi
0
44
12
36
Mo. of
events
26
16
25
18
26
TABLE 32. SCREENED (FINAL) EFFLUENT CHARACTERISTICS
AND AVERAGE PERCENT REMOVAL BY THE SCREEN - SITE I IA
Screen effluent Average^
mean con cen t ra t i on , a pe rcen t
Parameter mg/1 Range removal
BOD
TOC
Suspended solids
Total phosphorus (as P)
Fecal coliform density
pH
12 ± 4
27 ± 6
187 ± 52
0.22 ± 0.05
450
—
2 -
7 -
40 -
0.05 -
0 - 48
6.90 -
34
65
513
0.53
,000
8.40
20
41
50
40
22
--
No. of
events
27
28
29
27
28
28
a. Limits given for mean are for D5 percent confidence interval using
"t" distribution. Means given are arithmetic except for fecal
coliform density which is geometric.
b. Average removal calculated as [(mean raw - mean eff.)/raean raw] x 100.
c. Number/100 ml.
121
-------
^_^
CO
cn
^^
CO
o
1-
co
ui
t-
o
oc
£
o
11
o
ca
z
Ul
ce
o
CO
CO
CO
Ul
1
CD
^£
1-
~~
IB
CO
•w
0)
4-1
CO
•o a)
ca «i- w
C *J •-
—
OL — O
in O 1/1
CO
-o
0)
"a in
c -o
0) 4-
in o
3 in
CO
in
— -a
CO
T3
a) in
•a -c
li
in in
3
CO
0)
ro**j —
•M (D —
,0—0
1- 0 w
. — in
ro T
t- *o
in
li
O O \O
co o*\ *^*
^j* r*^* co
CM" CM —
0 0 LA
-3- -a- co
0 S
CO O
O vO
•* ^
CM —
O O LA
r>. o co
CO CO CO
CO CM -3-
oor^OLAcnocor^vo
r*^ o o c?i ' co LA r**- OA co ••"-
vO>— LAOr^-3-CMCO»— <—
CM CM CM •— CM —
OOCMOO-3-LALA'-vO
OOCM-S^CTv— •— r-.LAf*->
r*« LA co vo cr\ -3" r»» vo co CM
CO CM CM — CO •—
O -a- O fx.
LA co xio r*~
CM VO CT» •—
•t , •» . •»
CO — •—
O OO^--3-LA— — —
O O LA •— CO OO LA !•»• vO
LA ^~ J^* t"1^ vO ^-^ CO LA -3"
-3- — CM — CM CM
CO -3* LA vO r*^ CO O •— CM CO
CA vO
l>^ LA
CM O
vO —
,-T ~
CO «M
VO -3"
cn CM^
^T CM
.3- ^_ co -3- -3- O
O CO LA O CO CT\
IA -3" LA CO LA O
-3" *—
CO -3" VO LA OO LA
r^* r^» t^ -3" co -3*
^^ co c°\ o\ ^— co
vo c^ i^« vo r^** o^
CO — O 1** vO O
^- — CO CM
^r LA vo r-« co cr»
CO
o
-3-
vO
vo
cn
vO
LA
CO
—
CM
-3-
-3-
VO
-3-
—
O
CM
cn
CO
CO
—
r-.
VO
CM
LA
CO
m
~
co
vO
CM
CO
CM
^
•~"
CO
-3"
co
oo
vO
*^"
*
-3-
LA
O
CM
C
(0
122
-------
For the Site I screens, an average volume of 2.4 percent of the total plant
flow was used for backwash ing the screens at a head loss of from 30 to
4l cm (12 to 16 in.). At Site II, screen backwash requirements averaged
3.1 percent of the total plant flow using the same head loss as at Site I
to initiate backwash ing. Backwash requirements varied with hydraulic loading
and the solids loading on the drum screens during the run, with the fre-
quency of backwashing being the greatest during a combination of high hy-
draulic and solids loadings. For the Site I drum screens, backwashing was
continuous during periods of high flow, but intermittent when the flow rate
dropped below 80 percent of the design capacity. Backwashing was generally
intermittent at Site II. It appears that the screen backwashing require-
ments are greater for Site II than Site I because of the high flow - short
duration nature of the overflows at Site II.
A mass balance was performed on the screening process using the average
values found for the raw flow SS concentrations, volumes treated, back-
wash water SS concentrations, and backwash water volumes.used. For Site
I
the mass balance predicted a screened effluent SS concentration of. 221 mg/1
compared to the 95 percent confidence interval of 158 to 224 mg/1 obtained
from the sampling program. For Site II, the mass balance predicted a
screened effluent SS concentration of 574 mg/l compared to the 95 percent
confidence interval of 228 to 430 mg/1 obtained from the sampling program.
Seventeen runs were selected during which no major problems were encountered
with the drum screens or the sampling program. Regression analyses (13)
were run on these selected runs in an attempt to establish relationships
between the SS removals achieved by the drum screens and the hydraulic and
solids loadings. The following correlation coefficients v/ere obtained for
percent SS removals at Sites I and II. Regression analysis was not pel—
formed on the Site 1IA data due to the extreme scatter of the data.
Site
Dependent Variable
Correlation Coefficient
II
Sol ids loading
Hydraulic loading
Solids and hydraulic loading
Sol ids loading
Hydraulic loading
Solids and hydraulic loading
0.154
0.101
0.155
0.335
0.052
0.343
None of the obtained correlation coefficients is statistically significant
and, therefore, no linear relationship could be established for percent SS
removals and the hydraulic and solids loadings.
At Site MA, storm water is treated by screening only and then discharged
to the Root River. An average of 50 percent of the influent SS was removed
by the screen, although the percent removals varied widely for individual
runs.
The average hydraulic loading for the Site IIA drum screen was
123
-------
1.34 cu m/min/sq m (32.9 gpm/sq ft) with individual storm values ranging from
0.68 to 2.42 cu m/min/sq m (16.7 to 59-4 gpm/sq ft). The average solids
loading was 71.4 kg/1,000 sq m (14.6 lb/1,000 sq ft) and the solids loading
ranged from 12.1 to 227.8 cu m/min/sq m (2.5 to 46.6 lb/1,000 sq ft). The
average screen backwash requirement was 12.5 percent of the raw flow. This
value Is high because of a process modification: early in the project, the
differential pressure switch that controlled the backwash pump was dis-
connected and the pump was rewired so that it washed the screen after every
spiral screw pump operation. This change resulted in much more frequent
backwash ing of the screen.
Efficiency of Screening/Dissolved Air-Flotation - Data on the operation
of the flotation system is given on a run-by-run basis in Appendix IV-C.
The results of laboratory analyses of the final effluent samples is given
on a run-by-run basis in Appendix IV-D. All of the data collected during
the project are used in the data summaries although results from some in-
dividual runs may have been affected by operational problems and varying
chemical dosages.
The operational parameters for the flotation process may be summarized as
follows:
Site
Overflow rate,
cu m/min/sq m
(gprn/sq ft)
Detention time,
hr
Pressurized
flow rate,
cu m/min
(gpm)
Mean
0.92
(2.25)
Range
Mean
Range
.045 - .172
(I.10 - 4.22)
II
.075 .019 - .22
(1.84) (0.47 - 5.40)
0.50 0.23 - 0.91
0.76 0.18 - 2.12
Mean Range
3.21 1.92 - 4.35
(858) (513 - 1162)
2.59 1.42 - 4.16
(692) (.379 - 1111)
The pressure in the pressurization tank averaged 2.86 kg/sq cm g (41.2 psig),
the skimmer flight speed was 0.74 m/min (2.4 gpm), and the skimmers were on
for two minutes and then off for 12 minutes. The pressurized flow rate at
Site I averaged 47.7 percent of the raw flow rate and at Site II the
pressurized flow rate averaged 45.8 percent, of the raw flow rate (assuming
all the flotation tanks to be in operation and therefore all pressurized
flow pumps in use).
From this summary of operating data, it should be noted that the Site II
surface overflow rate is less than Site I and the Site II flotation tank
detention time is greater than Site. I. These factors may be responsible
for better removals that were achieved at Site II.
Floated sludge characteristics are presented in Table 34. At Site I
floated sludge ranged from 1.4 to 10.0% solids with a mean value of
solids. The volatile content of the sludge SS averaged 45.1%. At
the
124
-------
"TA
en
O
CO
o
K
CO
"—
cc
LU
f-
o
a:
o
.^— ^
LU —
o -^
3 £
CO
o
LU
^
O
_]
u.
•
•a*
!*"»
LU
i
CQ
^C
t-
~H
M.
CO
XI
0)
Xi in
C XI
4) •*-
0.—
in O
3 in
CO
— in
fD 4-1 •»
1 1 _
o — *5
t— O — XI
C -M —
0) flj r—
CL — O
mow
3 5
CO
XJ
._
,0 — *0
"-;§ «
I/)
— XI
,0*0
H- «n
c •
3 O
1
1
1
I
1
1
1
1
O
0
cn
CO
CM
0
o
•—
-3-
LA
O
0
o
f>
0
o
vO
LA
LA
rr»
O
O
«
1 t t*^ 1 t » I
1 1 — 1 1 1 1
O
LA
VO
1 1 vO 1 1 1 1
1 1 LA 1 1 II
O
O
vO
i i xrT i i i i
i i — i i i i
0
0
vO
i i -3- i i i i
1 1 LA 1 1 1 1
OOO LA O
o o LA r>- LA
0Y CO CT> LA I*~
-3- I"C r-T | fxT vo" 1
— CM — 1 *~1 *~^> «
i «»«^ «^««««M
i cn .a-- «*\ p». i i co P^ vo —
i IA en i i -a- co LA LA
OOLACO— -a-o o
r-^or»-LAcM coo co
r^*. t*\ co cn vo co LA cn
r^t^.^rcnoievivoi icvi
— r<1 — CM i CM CM I I CN)
o o • i — men o-a- o
oo-a-cocsi LAVO -a-
ff\ CD oo oo c^ i cr\ y^^- i ;i c^
•— LT\ r*»* r*"\ -Tj* t ^r «3" -i i i_f\
C3 C3 ^^\ C^ CO vO CO 1"^ CD xO CD
~->f* CD "-3* C3 ~-3" ^^* \O CJ CO l^\ ~ ~f
CT\ c^ ^^ vD CO CO CT\ t^^ •""• O*\ C^
Cn — -a'CMCMCOCA^-LALA —
CM^rcMCM — C4 CM
-------
Site II, the floated sludge ranged from O.A to S.^% solids with a mean value
of k.]% solids. The sludge SS volatile content averaged 27.5%. The percent
volatile suspended solids is greater for the Site I sludge than the Site II
sludge; this same condition was found for the drum screen backwash water.
This also indicates that Site II receives much larger quantities of inorganic
sol ids than Site I.
Frequently, the sludge volumes (backwash water plus floated sludge) exceeded
the capacity of the sludge holding tanks; therefore, accurate measurements
of the total sludge volumes produced by the system could not be obtained.
Also, because the volume of floated sludge produced was to be calculated as
the difference between the total sludge volume produced minus the backwash
water volume used, the floated sludge volume could not be obtained. Instead,
these volumes were estimated from a mass balance on the system later in
this section.
Tables 35 and 36 present the final effluent quality characteristics and
the average percent removals achieved for Site I and Site II, respectively.
The removal efficiencies obtained for each parameter are for the combina-
tion of the screening and dissolved-air flotation processes. Seven time
series studies were also conducted during the course of the project (both
raw and effluent samples were taken). The results of these studies are
presented and discussed in Section VI of this report, STORM WATER MANAGE-
MENT MODEL.
The calculation of the average percent removal by taking the arithmetic
average of the individual percent removals resulted in negative values
for dissolved BOD and dissolved organic carbon removals at Site I and
dissolved BOD removal at Site II. The mean concentrations (mg/1) for these
parameters through the system were:
Site Parameter Raw Screened Final
I Dissolved BOD 22 21 2k
Dissolved organic carbon 22 20 21
I I Dissolved BOD 30 26 23
Using these values, the average percent removal results were calculated to
be: 4.5 percent for dissolved organic carbon at Site I, and 23.3 percent for
dissolved BOD at Site II. The dissolved BOD at Site I, however, still shows
an increase in concentration. Since the dissolved-air flotation process
is not expected to remove dissolved pollutants, it may be assumed, despite
some variation, that no removal of the above pollutants was achieved.
Although the differences are not statistically significant, the data for
Site I reveals that dissolved BOD and dissolved organic carbon may have
126
-------
CO
o
MO
h-
co —
— •
CC LU
LU |—
!__„ ,^_
o t/>
C£ t
«^
nc *
o ^
tw ^*
r^ ^3
LLJ "^~
"IT* ^x
«l ^"*
U- J-
LU Z
UJ UJ
O
<. UJ
2: Q-
-w
U. UJ
IS
^f
• cc
LA LU
PA >
UJ
— i 0
oa s:
^c ^t
f«-
Vj_
o
«
o
o
- 0
O 1-
< Q.
ID
C
o
4-1
*J (D
C L.
<0 4J
3 C
—
0)
_
ID
Q
g
0)
i_
(U
01
c
(0
«..»
*N^
cn
E
i_
(U
+J
m
L-
(D
Cv.
-ST PA JT -3"
^r -3- -a- -r
•—•—•— vO
O PA r*1^ OA
LA CM -31
1 1
CO
-3- VD
•— C"\ ^ PA IA CM
•H +1 +1 -H
O -3" CO •—
I/)
Q l/> CJ m
o .- o .-
CQ e» 1- Q
PA
-3-
f"»
LA
LA
CM
r*^
i
vO
co
CM
co
CM
+1
CA
-3-
-3-
-a
•«•
^«
0
1/1
«~-
u>
•^J
•«
•«.
o
U)
-o
0)
c
PA
-3"
VO
vO
-3"
CM
VO
PA
j
O
-3-
O
LA
CM
O
•H
vO
vO
•
r*.
^.^^
a.
in
(0
*—*
in
3
O
0_
o
a.
^«>
ni
4J
O
H-
CM -3-
-3- -3-
1 1
1 1
O O
O O~%
0
o
-3- 1
cn
0
1 vO
4
0 vO
0
0 1
LA 1
O
^»
•M
•«•
cn
c
(U
E
O
u-
•«•
FM.
O
u
»•.«
flj
0
C O)
•«•
in in
3 *~~
«w f~
(D U
>- 4=
0) 5
c >-
•w
a> in
o c
C Q)
fl) -O
T3
— E
14- U
C O
o >«-
0 —
4J O
c o
o —
u o
a. a)
u-
LA
O
L. U-
0
«<- 4J
a.
a a)
L. 0
CD X
C
.- a)
cn u
(D
in
•«• C}
E >
'—i 01
*
(0
•
to
C
3
i_
'ro
L.
o
M-
to
r—
ro
>
o
E
0)
L.
4_j
C
•—
cn
tt)
3
r—
(D
>
•
-Q
•
r—
E
0
O
*v^
•
o
"Z.
•
a
127
-------
CO
o —
1— ~~
CO LU
"™ l~~
ry* «•
LU CO
i n |
^£
CC CO
h"* ^^
•y 1 1 1
LU CC
Z3
— 1 I—
LU 5
LU LU
LU O
a:
_J LU
< 0.
z
— LU
LU O
2
» Ul
vO >
Ul Q
— 1 3
ea <
!••
**- tft
O 4J
c
* Q)
0 >
2E 0)
Q
0) 4J •—
01 c re
(Q V ^
L. U Q
0 U, g
< o. u
0)
Ol
c
re
a:
re
c
o
JJ
4-1 re
C 1-
0) 4-» —
3 C ^-s.
— 0) 01
•4- 0 E
>»- c
LU O
u
c
re
E
l»
0
4J
Q)
ps
re
re
a.
— O CM —
CA CA CA CA
-^- CM -=r o
O CM O vO
VO La
1
*•"•• CTl *— • CA
La vo in CA
i l i i
ua -^* rs«» r^^
-3- LTV -a- m
+1 -H -H -H
-a- CA vo o~\
CM CM CM —
c
o
-Q
u
re
u
u
•«
c
re
O Ol
O L-
CQ O
"O ^3
<1) 0)
> >
O O
in in
a 1/1 o m
0 — 0 —
OQ Q |— Q
O CM
m CA
vo •—
t~N. VO
rr» vO
f^.
r>. CM
u\ -r
CM
i
11
-a-
VO CM
CM CM
O^ —
CM CM
•H W
O CA
Ul
•w
^_
01 O
"O in
,M
^•~ ID
O
CM
tn
-o
rMB
O
in
•o
D.
in
3
l/t
,2!
•—
+-*
re
O
CM
CA
CA
O
VO
CO
co
r«
1
La
•
o
-a-
-=r
o
+1
CO
*
0
1^^
o.
in
re
**-^
in
u
O
O.
I/)
o
0.
«— •
re
2
1
CA
O
i La
•
O CA
O
O 1
^^* 1
O
?"*
.tf
in
c
0)
-a
E
,2
•«•*
•— •
8
r-w
re
u
LU Q.
in
c
re
*
c
o
•J
-Q
«•— *
U
4-1
in •
— U
T3 —
U
r 4J
*J 0
o
o> a)
C O)
• w
Ul Ul
3 •—
— JZ
re u
^> •— •
L- .C
>
««*
u
a. a)
14-
LT\
OA ^
o
1. 14-
o
tt±lm -IJi
a.
4-
m
r~*
OJ
?>
O
E
O
L.
4-J
C
a)
o
i—
(U
a.
14-
o
c
ro
i
E
O
•—
4-1
(U
E
-C
4-J
•~"
L.
fO
tn
.—
C
(L)
^
O>
-d
•
r—
E
0
O
^~
"^
o
dZ
o
128
-------
actually increased through the flotation tanks. One possible explanation
for the increase in these dissolved pollutants is the presence of sludge
on the bottom of the flotation tanks. If the sludge begins to digest during
dry weather, it would produce some dissolved organics that would then be
picked up by the wet weather flow through the tanks. It is felt that if the
sludge can be removed between system runs, this problem may be eliminated.
However, it is possible that the dissolved fraction of the wastewater in"
creases through the system as indicated by the tests.
The minimum pH value of 3-50 at Site II (Table 36) results from problems
encountered with the ferric chloride feed system. As discussed previously,
the raw flow was frequently overdosed with ferric chloride because the
operating personnel had no control over the gravity feed system. When
overdosing did occur, it resulted in very low effluent pH values (3-5 to
5.5).
For the most part the treatment achieved at Site II was better than the
treatment achieved at Site I because of two factors:
1. At Site II, the flotation tank surface overflow rate was 19 percent
less than at Site I.
2. Correspondingly, the flotation tank detention time at Site II was
52 percent greater than at Site I.
These differences may be explained by the fact that on the average, Site I
ran at 54 percent of its rated capacity while Site II ran at only 39
percent of rated capacity.
Table 37 presents a comparison of the average percent removals achieved
during 1973 and 1974. The overall treatment improved in 1974. The
TABLE 37. PERCENT REMOVALS FOR 1973
COMPARED TO PERCENT REMOVALS FOR 1974
Parameter
BOD
TOC
Total solids
Suspended sol ids
Volatile suspended solids
Total phosphorus (as P)
Site 1
1973
42.7
43.0
25.8
57.1
62.6
A3. 7
1974
57.5
51.2
25.7
62.2
66.8
49.3
Site
1973
52.8
39.2
31.8
56.0
37.7
46.8
II
1974
65.4
64.7
41.1
73.3
70.9
70.0
129
-------
Improvement occurred because 1973 was a period of startup and shakedown
for the system. Many major problems were encountered with the equipment
and standard operating parameters were not yet established. Undoubtedly,
these factors contributed to producing efficiencies that were not truly
indicative of removal efficiencies that could be achieved by the screening/
dissolved-air flotation process. The 1974 results, therefore, give a
better indication of the pollution removals that may be expected from
a well operating screening/dissolved-air flotation unit that is treating
combined sewer overflows: 65 percent BOD removal and 73 percent SS removal.
Tables 38 and 39 give the average percent removals averaged at Site I and
II, respectively, on a mass basis. The mass of pollutants in the influent
and the effluent of the sites for each run are given in Appendix IV-G,
Tables Gl to G12.
Calculation of the percent removals in this manner increased all of the
values, as shown in these data:
Site
Parameter
Average percent removed
Arithmetic mean Mass basis
BOD
Total organic carbon
Total solids
Suspended solids
Volatile suspended solids
Total phosphorus
BOD
Total organic carbon
Total solids
Suspended sol ids
Volatile suspended sol ids
Total phosphorus
50,
47.
25,
59,
64,
46.6
60,
50,
37.6
66.1
57.0
60.3
62.4
60.0
28.1
67.6
73.6
53.2
69-5
66.6
47.2
69.8
67-3
62.4
The reason for this increase is that the overall treatment efficiency was
usually better for long duration runs (large volumes treated) than for
shorter duration runs (small volumes treated). Therefore, the mass removals
are greater than the arithmetic mean which gives equal weight to each run
without regard to the volumes treated.
The principal reason for this phenomenon lies in system start-up time
required (30 to 45 min) after effluent flow began before a good quality
effluent was achieved. Once this quality was established it remained
fairly constant for the entire run, however, the effluent quality did
decrease during this time when the flow significantly increased but
the effluent in quality was still better than in the first 30 to 45 min.
As the run duration increased, the effluent samples obtained during
the first part of the run became a lesser fraction of the total
130
-------
_
LU
1-
CO
1
CO
•z.
j—
_J
o
CL.
U.
O
CO
CO
2:
o
Q
LU
CO
o o
^» E
flj C
E C
O —
•""
O)
u
0}
0)
e
ro
*—
(?
a.
-S^ ^^ ^~« \O
* * * *
CM o co r^.
vO vO Cxi vO
LA LA O O
r-» co •— ^t-
00. 0 ^- LA
co r^- co LA-
CM CM -3- vO
CO
o o o o
•^3™ vo r^*» r***
CO CM r~- LA
O CM LA r—
— — LA CO
O LA o O
— •— ' LA —
-4- VO r- O
co r^ ^^. LA
\o xo r^*« ""*"
-=f CM
O O O O
O~1 I — CO CO
r~^ vo vo LA
oo o xo r — .
csi co r— cri
CM
en
-a
_
in o
"O m
•—
— -a
O a)
w TJ
c
—
3
-------
c «>
o o
0) 0)
C_ L.
LU
H-
to
I
to
tn «
(0 c
~
ca
132
-------
composite sample. Conversely, when the run was short, the samples taken
during the first 30 to 45 min were a large fraction of the final composite.
For that reason, the percent removals for short runs were usually less
than for long runs.
Because of the long duration of the overflows at Site I, the site was
run continuously for as long as 41 hours. This duration was achieved by
allowing the site to run on its own during the low flow periods after the
rainfall had ceased and the overflow subsided resulting in the site running
unattended for up to 17 hours.
Figure 47 presents a plot of the treatment achieved at Site I (percent
BOD removed) versus the run duration for 16 runs in 1974 that were of
duration less than 780 min. The resulting plot reveals the trend of
improving percent removals as run duration increases. The percent removal,
based on the composite effluent sample, increases significantly with time
until the 16-run average of 59 percent removed, is reached after 180 min.
Plant flows and effluent quality are plotted versus time and discussed in
more detail in Section VI of this report, STORM WATER MANAGEMENT MODEL.
As stated previously, the treatment results are based on all of the runs
despite variations in chemical additions, which were due to:
• Changes in the desired dosages based on bench scale flotation tests.
Malfunctions of the chemical feed systems.
In an attempt to establish what effect chemical dosages had on treatment
efficiency, the percent SS removals at Site I! were compared to the
corresponding ferric chloride dosages, as shown in the following table:
Ferric chloride dose (mg/l)
Mean percent removal
Number of runs considered
Use of the "t" statistic for the comparison of two means (15) revealed that
the only significant difference was between the percent SS removals for
additions of zero ferric chloride and 21 to 50 mg/l. No other statis-
tically significant differences were found, mainly because of the small
sample sizes available for analysis. It appears that the ferric chloride
dosage ranges of 1 to 20 and 51 to >70 mg/l all resulted in equal SS
removals, namely 71 percent. The rate of 21 to 50 mg/l improved the
removal to 82 percent, which corresponds with the bench scale results that
predicted the best flotation at a ferric chloride addition of 40 to 50 mg/l,
Using average values obtained during the course of the project and the
entire treated flow, a mass balance was performed on the entire treatment
system as shown in the following table.
133
-------
o
CM
o
vo
NO
o
o
SO
o
-a-
u-\
O
CM
O
vO
O
o
o
-3-
CM
O
CO
o
CM
O
NO
i
o
I
3
cn
000
CO 1^. VO
O
in
o
Q3AOW3U QOS
-------
Site
II
Mass balance data
Suspended solids (mg/1)
Floated
Influent Effluent Backwash sludge
Volumes (cu m)
Treated
d ischarge Backwash
266
661
94
113
1843
2767
50940
4i4so
8556
9572
207
301
The value used for the Site II influent SS concentration was the 1974 mean
because the first flush portion of the combined sewer overflow was fre-
quently missed in 19735 and, therefore, the composite samples did not give
a true indication of the solids entering the site. From the mass balance,
the following estimates on sludge production during the evaluation period
were made:
Estimates of sludge production
Site I
Site II
Volume floated sludge, cu m (gal
Total sludge volume, cu m (gal.)
Total sludge suspended solids, %
Volume backwash/total sludge, %
Sludge production/vol treated,
cu m (gal.)/1000 cu m (gal.)
.) 21.4 (5,641)
228.4 (60,343)
0.64
91
26.7 (26.7)
106.4
407.4
1.29
74
42.6
(28,108)
(107,635
(42.6)
It must be noted that all calculations assumed that no settling of solids
occurred, but in actuality some solids did settle to the bottom of the flo-
tation tanks.
The Site II values of volume of floated sludge produced, percent SS of the
total resulting sludge, and sludge-volume-produced/volume-treated are all
higher than Site I because of the very high concentrations of SS in the
influent. The backwash water is a larger portion of the total sludge
volume at Site I because of the higher hydraulic loadings on the drum
screens at Site I causing the backwash pump to start more frequently and
remain on longer than the backwash pump at Site II.
Efficiency of Chlorination - The concentrations of fecal coliform bacteria
in the effluent from Sites I and II are summarized in Tables 35 and 36.
The chlorination control was generally set to maintain a dosage of 10 to
15 mg/1 of chlorine based on the incoming flow rate. It was difficult,
however, to maintain this rate for the duration of a run because of the
frequent loss of the chlorinator operating vacuum due to plugging of the
injector with solids.
The geometric means for 42 runs at Site I and 29 runs at Site II (Site I:
500 colonies/100 ml; Site II: 700 colonies/100 ml) are both above the
standard (200 colonies/100 ml) set for the Root River by the Wisconsin
Department of Natural Resources (II). The geometric means take into
account nine runs at Site I and eleven runs at Site II when no chlorine
was added or when chlorine was added but no chlorine appeared in the final
effluent because frequent plugging of the chlorine injector. A comparison
135
-------
of runs at Site I and Site II when no chlorine was added and when chlorine
was added in sufficient quantities to produce an effluent residual follows:
No chlorine addition
Chlorine addition with
effluent residual
No. of events
14
37
Coli form
geometric mean
(No./lOO ml)
54,000
113
Thus, whenever sufficient operating vacuum could be maintained during a
run to produce a chlorine residual, the effluent fecal coliform counts
were below the standard set by the Wisconsin Department of Natural Resources,
This condition was achieved by a chlorine residual in the effuent of approx-
imately 0.5 mg/1. The residence of the wastewater in the flotation tanks
appears to be sufficient in terms of manner and time of chlorine contact.
Considering the highly variable concentration of fecal conforms in the
raw waste, secondary control of chlorination based on the residual in the
effluent should be considered for combined sewer overflow treatment facili-
ties, especially In light of the growing concern over the toxic effects of
excessive chlorine residuals on the fauna in the receiving body. Standard
operating conditions produced effluent residuals of 0.4 to 10.0 mg/1 for
trie runs in which an effluent residual was detected.
Impact of Treatment on the Quality of Discharge to the Root River
One of the main objectives of the project is to show an improvement in the
quality of the receiving water as a result of treatment of storm generated
discharges in the test reach. The crucial elements in achieving this
objective are to capture as much of the overflow as possible and to give
It the best treatment possible before discharge to the Root River. The
percent removals presented previously are the removals achieved through
the treatment processes only; no consideration is given to pollution occurr-
ing from plant bypass.
The mass loadings to the Root River due to bypass volumes and site effluent
volumes at Sites I and II in 197** are given on a run-by-run basis in
Appendix IV-H, Tables H-l to H-12. Percent removals are also given using
the values. Site IIA was not included in this analysts for two reasons:
plant bypass rarely occurred at the site and, because of inaccurate plant
flow measurements, the concentration (mg/l) of the pollutants in the in-
fluent and effluent could not be converted meaningfully to a mass basis.
The data collected in 1973 is not included here because it did not take
into account the overflow volumes, if any, that occurred after the site
run was ended. Therefore, the total overflow volumes for each run were
not accurate. However, it should be pointed out that after the first flush
is completed, the pollutional strength of the CSO decreases within two to
three hours to a low, fairly constant value (see plot of discrete sampling
136
-------
events in Section VI, STORM WATER MANAGEMENT MODEL). In addition, screen-
ing/dissolved-air flotation treatment does not result in a significant im-
provement in the quality of the discharge; therefore, the benefits of
treatment of the extended overflows are limited. For this discussion, the
composite sample of the influent is used as the quality for the entire
overflow event. Tables 40 and 4l give summaries of the percent total mass
removals for Sites I and II, respectively.
The low percent removals achieved, especially at Site II, are due to the
problems that occurred resulting in failure to capture large portions of
the combined sewer overflow, especially the "first flush" overflow. Four
recommendations are made to correct this problem:
1. The bar screen rakes at Site II should be kept operational a.. ^11
times, if possible, so that a buildup of material on the bar scrrens
does not block the plant flow and cause bypass.
2. The automatic startup mode for the sites should be kept operational
and the sites should always be set to start automatically.
3. Equipment problems, when they occur, should be corrected while the
sites are running, if possible. The sites should be shut down
during an overflow only if absolutely necessary.
4. Personnel and chemicals should be available so the sites can be
kept running .unti1 the combined sewer overflow has ended.
If these.recommendations are followed, it will be possible to achieve
a much greater percent removal of the pollutants discharging from the
comb i n ed sewe rs.
Special Testing
Combined Sewer Overflow Pesticide Concentrations - The results of the
pesticide testing are covered in Section V of this report, RIVER MONITORING
STUDIES, and are considered in conjunction with the pesticide testing done
on the Root River.
Combined Sewer Overflow Particle Size Distribution - Two separate sieve
analyses were run on sewer samples composited from overflow discharge
points No. 1 and 2 (see Figure l) before site construction was completed.
The analytical results for storm event No. 6 (8/10/70 are presented
graphically in Figure 48, and the results for storm event No. 11 (11/1/71)
are presented in Figure 49.
Sieve analyses were also made for the combined sewer overflow and final
effluent at Site I. Figure 50 graphically presents the results of the
sieve analyses performed on the combined sewer overflow and final effluent
for run No. 12 (7/20/73). Figure 51 presents the results of the sieve
analysis performed on the combined sewer overflow for run No. 42 (7/22/74).
137
-------
^^
•^
cn
^
LU
t
r^
co
i
CO
LU
H
_l
o
>
o
_J
u.
^v
LU
0
fy-
LU
LU
CO
jt i
•—
CO
2:
o
o
s
ft*
Lu
CO
<
^^
LU
C£
CO
s:
_i
^«-
o
t-
•
o
_^«
LU
CO
<
J—
4J TJ
C
i^ E
0} 0)
Q. i-
f\
TJ —
0) l_
CD 0}
*- >
.c o;
u
U) 4-1
— 0
TJ O
C£.
UI
UI O
re ij
2: en
i| o
Vp .
0)
^
o
••* L*
0)
wi S
IA
re ui
2:
TJ
0)
c
— en
^ %^
E
O
U
u
4)
'tj
re
re
0.
-a- co
^ en
-3- cn
o o
CM cn
cn LA
vO -3"
CM cn
o cn
so cn
cn vo
r— IA
_ w-
0 0
— o
— LA
o r«~
LA LA
o cn
cn f^*
1-. 0
CM VO
CM CM
Q U
§ P
cn cn
o -3-
CM -3-
o c
CD r^*
-a- cn
cn CM
CM cn
cn
NO VO
vO —
vO CM
-3- -3-
0 0
0 0
vO vO
La co
o vo
-^J~ /»—
cn -3-
cn cn
cn -3-
cn vo
co f-»
r-w
UI
TJ
UI O
TJ UI
*MB
— TJ
O 0)
IA TJ
C
— Q)
re CL
4-< IA
O 3
1— co
t^. cn
cn LA
-3- cn
o cn
cn r~-
La —
o —
-3"
*•_
•
cn CM
o cn
-3- IA
CO
•_
O -3-
^^ *~~
vO CO
o —
CO
0
•
— cn
en CM
LA CO
vO
cn
IA
TJ
• w
*—
o
IA
TJ
TJ "aT
Q) IA
o. re
IA — '
•3
IA UI
3
0) 1-
*• • o
*••• ~a
4J Qu
re IA
a 1
^
cn
N«^
—
LU
(_.
CO
1
CO
LU
2:
=3
_!
0
>
3:
o
u.
Ql
UJ
0
C£
UJ
UJ
CO
o
LU
•z
"•"
CO
2:
o
o
2:
o
o;
u.
CO
_J
c
o
U
O
0.
L.
O
._
cc
4J
O
o
CC.
o
4-1
8
r«*
*t
!**•
cn
«•—
-------
10
100
0.5 170 175 270
PARTICLE DIAMETER, mm
2.5
Figure 48. Sieve analysis of combined
sewer overflow for storm No. 6, 1971.
0.5 1.0 1.5 2.0 2.5
PARTICLE DIAMETER, mm
Figure 49. Sieve analysis of com-
bined sewer overflow for storm
No. 11, 1971-
100
4- RAW OVERFLOW
V PLANT EFFLUENT
0.5 1.0 1.5 2.0 2.5
PARTICLE DIAMETER, mm
Figure 50. Sieve analyses of combined
sewer overflow and plant effluent at
Site 1, run No. 12, 1973-
I
0.5 1.0 1.5 2.0 2.5
PARTICLE DIAMETER, mm
Figure 51. Sieve analysis of com-
bined sewer overflow at Site 1,
run No. 42, 1974.
139
-------
The drum screens at Site I have an opening size of 0.297 mm. From the
graphs presented, the average fraction of the particles greater than 0.297
mm was 22 percent. The average suspended solids removal by the drum screens
at Site I was 32 percent.
Combined Sewer Overflow Nitrogen Concert!: rat Jons - Total Kjeldahl nitrogen
analyses were performed on both the influent and effluent samples from
Sites I and II periodically during the course of the study project. The
results may be summarized as follows:
II
Mean Concentration
Sample mg/1 (as N)
Influent 8.58
Effluent 5.52
Influent 4.50
Effluent 2.10
Range
2.55 - 13.20
2.95 - 7.40
2.30 - 6.70
1.60 - 2.90
No. of
events
6
6
2
3
These values give average percent removals of total Kjeldahl nitrogen of
35.7 percent at Site I and 53-3 percent at Site II.
Analyses were also performed for nitrogen and total dissolved solids in
Influent and effluent samples from both sites (I and ll) for Run No. 12
(7/20/73). These analyses were performed after ia long dry spell, 17 days
since the last overflow event. The results are presented in Table 42.
Chloride Cohcentrations in Combined SewerOverflows and Storm SeWer Dis-
charges - Twice during the winter months, samples were taken of the combined
sewer overflow at Site I and the storm sewer discharge at Site MA. These
samples were then analyzed for total chloride concentrations. The first
set of samples was obtained on 12/26/73. The chloride concentration at
Site I was 98 mg/1 and the concentration at Site MA was 795 mg/1. Since
Site MA receives mainly street runoff, this high concentration of chlorides
is most likely caused by street salting operations.
The second set of samples was obtained on 3/23/74. The chloride concen-
tration at Site I was 52 mg/1 and the concentration at Site MA was 58 mg/1.
Fecal Streptococci in Combined Sewer Overflows - For Run No. 12 (7/20/73),
the concentrations of fecal streptococci were determined in addition to the
concentrations of fecal col{forms in the combined sewer overflows and final
effluents at Site I and II. The results follow (No./lOO ml):
Raw
Site
I
II
Fecal Colt
2,300,000
1,650,000
Fecal Strep
225,000
160,000
Final Effluent
FecaT Coli
300
Fecal Strep
940
970
140
-------
TABLE 42. NITROGEN SERIES AND TOTAL DISSOLVED SOLIDS
FOR RUN NO. 12 AFTER LONG DRY SPELL
July 20, 1973 (17 Days Since Last Overflow)
Site
No. Parameter
1 NH3
NO,
N02
TKN
TDS
1 1 NH3
NO,
N02
TKN
TDS
Units
mg/1 - N
mg/1 - N
mg/1 - N
mg/1 - N
mg/1
mg/1 - N
mg/1 - N
mg/1 - N
mg/1 - N
mg/1
Raw
2.03
1.01
0.08
13,20
245
0,92
1.34
0.06
6.70
227
Final
effluent
2.42
2.30
0.05
5.00
285
0.81
1 .08
0.02
2.90
U7
It appears that the destruction of fecal streptococci through the system is
much less than the destruction of fecal conforms.
IV-5 ECONOMIC CONSIDERATIONS
As stated previously, one of the objectives of this project is to develop
detailed cost information (capital, operating, and maintenance costs) to
establish cost/benefit relationship for this method of treatment compared
to other treatment techniques. This information would also be helpful in
evaluating the use of this treatment technique for other sites in Racine,
as well as in other cities.
The design, land, construction, and equipment costs for the demonstration
systems are presented in detail in Appendix IV-B. These costs were summa-
rized previously in the subsection, IV-2, SYSTEM DESIGN AND CONSTRUCT I.ON.
The capital cost for Site I was $436,599; Site II was $841,420; and Site
IIA was $25,001. The total capital cost was $1,303,020 (March 28, 1973;
ENR = 1973).
141
-------
Using an interest rate of 8 percent, the following cost table (
-------
TABLE 43. OPERATION AND MAINTENANCE COSTS
August 1974; ENR = 2078
Item
Water
Electricity
Ferric chloride
Polyelectrolyte
Chlorine
Operating personnel
Maintenance Personnel
Utility personnel
Miscellaneous (supplies, etc.)
TOTAL
Chemicals (15% of total)
Utilities (6% of total)
Personnel (78% of total)
Unit cost
$60.00/year
$0.03/kwh
$5-52/100 Ib
$0.4l/lb
$10.25/100 Ib
$10,00/hr
$!0,00/hr
$9.00/hr
$25/month
C/1000 gal.
0.06
1.35
1.84
0.68
0,85
3.08
8.58
6.31
0.25
23.00
3.37
1.41
17-97
$/cu m
0.02
0.36
0.49
0.18
0.22
0.81
2.27
1.67
0.07
6.08
0.89
0.37
4.75
(40 hr/week) and the rate for this overtime was calculated as 1.5 times the
base rate. A base rate of $10.00/hour is used to include the actual pay rate
plus fringe benefits. This is based on an average wage rate for City of
Racine Water Pollution Control Plant operating personnel. The maintenance
cost is based on 30 man-hours/week. The total of 30 man-hours/week was
estimated from the maintenance requirements during the last half of 1974
just before the project ended. The 30 man-hours/week were divided up
into 13-5 man-hours/week for utility personnel who have a lower pay rate
($9-00/hour) and would be assigned to cleanup, and 16.4 man-hours/week
for actual maintenance personnel. This breakup was made because it was
found that 45 percent of available maintenance time was spent on cleanup
activities (see Table 17).
143
-------
Of the total operation and maintenance cost, 6.08$/cu m (23-00^/1000 gal.),
15 percent is for chemicals, 6 percent is for electricity and water, and 79
percent for labor. Of the labor cost, 4.75
-------
TABLE 44. ESTIMATED FUTURE OPERATION AND MAINTENANCE COSTS
August, 197^; ENR - 2078
1 tem
Water
Electricity
Ferric chloride
Polyelectrolyte
Chlorine
Operating personnel
Maintenance personnel
Uti 1 ity personnel
Mt seel laneous
TOTAL
Chemicals (28% of total)
Utilities (\}% of total)
Personnel (60% of total)
Unit cost
$60.00/yr
$0.03/KWH
$5.52/100 Ib
$0.4l/lb
$10.25/100 Ib
$10.00/hr
$10.00/hr
$9.00/hr
$25/month
i/1000 gal
0.02
1.35
1.84
0.68
0.85
1.91
2.15
3.16
0.06
12.02
3.37
1.37
7.22
£/cu m
0.01
0.36
0.49
0.18
0.22
0*50
0.57
0.83
0.02
3.18
0.89
0.36
1.91
145
-------
^*"%
CT"\
^-^
^•^
O
O
o
CM
n
H C£
to z
°ii
Ul to
O Ul
2: f—
•z. to
Ul
I—. l«-
•z. -z.
•M* IjJ
< s:
<
a uj
z cc.
<. }-
•Z. 3:
O O
i i, J^T
S UJ
Ul >
a. o
o
O Ul
ui 3
h- Ul
< to
z:
— o
H~ ul
to -z.
Ul —
oa
o S
z o
< 0
^•1 r**^
So
i- f-
o
< Q
Ul
ino_
o
Ul 0
_j
oa i.t
< z
1
o
-
.tf
U
to
o.
ra
Ol
o
o
o
^~
^s^
"1>
B
3
O
^>
-Q
cr
^»,
ra
•o
•3
u
(/)
(A
1
M-
<0
•o
(A
*-»^*
^iH
ra
J3
4-1
O
(0
^— *
•w
3
•»
Q)
••
U
ID
OL
\0
_
v<—
VO
O
f^
O
*
co
in
o
m
in
en
CM
M-
ro
-a
(A
^^
T3
Q)
4J
ra
e
•—
4J
(/)
4)
*^-*
«-M
3
«k
0)
c
u
ra
C£
co
o
•— •
in
CO
CM
1
1
1
|
«S-
ra
-o
LOk
fl}
C
o
*o
u
u
»•— •
3
£
-O
^»
1C
Di
*^C
«t
f*
.tl
E
to
4^
t_
O
U.
CO
in
p^
in
^_
o
4
in
Lf\
CTi
co
*w
M-
ra
•o
(A
«•
^:
«^
(U
JC
Q.
4)
-o
ra
•— *
f*
a.
—
vO
^_
vO
»_
O
•
VO
O
«k
CM
CM
IH
a>
•™"
^~-
Ol
c
««•
*—
0
L.
J-
_^ ,
z
A
ca
«
o^
c*^
L»
o
•M
u
re
4-1
c
8
•— •
Q
U
4«H
O
_£
5
DQ
«•
^1
A
8
\/
3
ra
^
«««
"Si
o
_
vO
CM
O
O
•
J-".
vO
in
o-»
in
i»
-------
equipment. Table kS also gives the design capacity of the sites reporting
their operation and maintenance costs. Neglecting the Springfield oxida-
tion pond, the Racine sites are more than double the size of any other CSO
treatment site, and 10 times larger than the other dissolved-air flotation
units. Therefore, for the small pilot units, maintenance costs are negli-
gible, but for the full scale application of screening/dissolved-air
flotation in Racine, it was found that maintenance costs will be a very
large portion of the total costs of treatment.
-------
SECTION V
ROOT RIVER MONITORING STUDIES
V-l GENERAL DESCRIPTION OF THE ROOT RIVER DRAINAGE SYSTEM
In October 1965, the Southeastern Wisconsin Regional Planning Commission
issued a comprehensive report titled: "Preliminary Report on a Comprehensive
Development Plant for the Root River Watershed" (18). The following
material is a condensation of the general descriptive information from that
report and from a report issued by the State of Wisconsin Department of
Natural Resources dated November 1967 (19).
Description of the Root River and Its Drainage Basin
The Root River rises near West Allis and flows south and east into Lake
Michigan at Racine, a distance of 64 Hver kilometers (40 river miles). It:
drains about 500 square kilometers (193 square miles) of Kenosha, Milwaukee,
Racine, and Waukesha counties. The watershed basin is bounded on the south
by the Pike and Des Plaines watersheds and on the north by the Menomonee,
KInnickinnIc, and Oak Creek watersheds. The western edge of the Root River-
Basin is bounded by the Fox River (Illinois)-watershed; the western boundary
also marks the Mississippi-St. Lawrence drainage basin divide (Figure 52).
Chief tributaries to the Root River are Hoods' Creek, which intersects the
river about 19 kilometers (12 miles) from the mouth, and the Root River
Canal, which flows into the main stem approximately 42 kilometers (26 miles)
from the mouth. Figure 53 gives the locations of communities and industrial
pollution sources found within the watershed.
In its 64 kilometers (40 miles) of length, the Root River falls nearly
46 meters (151 feet) resulting in an average gradient of about 0.72 meters
per kilometer (3.8 feet per mile). In the final 9-7 kilometers (6 miles) of
its length, however, the river exhibits a much steeper gradient, dropping
about 3.0 meters per kilometer 05..9- feet per mile). The gradient of drop
in the river and its tributaries is presented in Figure 54.
The topographical features of the Root River watershed are a result of
glaciation. Although the area is composed chiefly of ground moraine,
morainal hills and ridges are also encountered in the basin. The hills in
the watershed reach an altitude of about 292 meters (960 feet) or approxi-
mately 116 meters (380 feet) above the level of Lake Michigan. Soils of
the region are complex in pattern, although most can be classified as various
types of silt loams. Extensive areas of poorly drained organic soils occupy
148
-------
WASTE SOURCES
O SEWAGE
A CANNERY
D OTHER
O
Figure 52. Root River watershed showing communities
and principal industrial pollution sources.
149
-------
Figure 53. Location of the Root River watershed,
150
-------
IS)
0)
ID
*J
J3
i. .—.
•M CO
•JT
>- O
O rr\
•— > •
ro O
E x
•a <«-
c
ro it
I- (A
0) t~
> fl>
'cz a)
*j
o •>•
O trv
a: so
en
*J 0)
in
u- ro
o -Q
d) ro
— *j
.- ro
M- "O
O^'
a.
Lrt
en
tZ
133d Nl 13A3T V3S NV3IN 3AO8V NOI±VA313
151
-------
certain areas of the basin. Virtually all of the soils encountered in
the region have severe limitations for residential development utilizing
septic tank disposal. Bedrock underlying the glacial drift consists of
limestone, shale, and sandstone, with the drift, limestone, and sandstone
being water bearing strata of chief importance.
Land Use
The distribution of land use by categories is roughly as follows:
Use category
Residential
Commercial
Industrial
Mi n i ng
Transportation
and utilities
Governmental and
Institutional
Recreational
Agricultural
Water, woodland
and wetland
ha
Area
ac
sq km
52.27
2.36
1.30
3.03
38.28
sq mi
20.18
0.91
0.50
1.17
14-78
5,227 12,916
236 583
130 321
303 749
3,828 9,459
471
1,319
33,883
5,739 14.181 57.39 22.17
51,136 126,357 511.36 197-34
1,164
3,259
83,725
4.71
13-19
338.83
1.82
5.09
130.77
Percent of
watershed
area
10.22
0.46
0.24
0.59
7-50
0.92
2.58
66.27
11.22
100.00
Urban land uses within the basin account for about one-fifth of the total
area and are concentrated primarily within Milwaukee County on the upper
tributaries of the Root River and within the City of Racine.
Approximately two-thirds of the land area in the basin is used for agri-
cultural purposes. Some sections of the region are experiencing rapid
urbanization. There are no natural inland lakes in the basin, and little
recreational use is made of the streams. The relatively low streamflows
frequently encountered, severely limit recreational, industrial, municipal,
and agricultural uses of such waters. Lake Michigan is used extensively
for recreational activities and as a water supply for the larger munici-
palities and industries.
Climate
The watershed has a continental climate characterized by four distinct
seasons. Winter begins in November, lasts through March, and tends to be
cloudy, cold, and snowy. Freeze-up of streams and lakes usually occurs
in early December and does not end until early April; however, there is
often a short-lived mid-winter thaw due to unseasonably warm temperatures.
Soring is slow in arriving, partially because of the cooling effects of the
waters of Lake Michigan, and is a mixture of both summer and winter.
152
-------
Summers are fully developed and generally warm but marked by occasional
hot and humid periods and unseasonably cool periods. Frequent breezes
from Lake Michigan offer relief from high summer temperatures to those
areas of the watershed lying within a few miles of Lake Michigan. Fall may
extend from September to November and is characterized by mild, sunny days
and cool nights. By Fall, Lake Michigan waters have become warm to the
extent that the lake tends to prolong Fall in the watershed a week or so
longer than in areas farther inland. The climate of the watershed can be
understood more fully by examining phenomena of temperature, precipitation,
wind movement, sunshine, and evaporation recorded at the Milwaukee First
Order Weather Station,- which is located within three miles of the watershed
and is considered generally representative of watershed climatic conditions
The mean
(71.35° F)
July, is 21.84 C
of 38.30° C (101° F) .
January, is -5.58° C
(-24° F) . Temperature
daily temperature during the hottest month,
with an official record high temperature
The mean daily temperature during the coldest month
(21.94° F) with an official record low of -31.08° C
conditions within the watershed allow a growing season of from 155 to
175 days. Average dates of the last killing frost in spring and the first
killing frost in fall are May 1 and October 12, respectively, with upland
areas tending to have the most frost-free days.
Annual precipitation on the watershed, including snowfall, averages about
76 cm (30 inches), but annual amounts have ranged from a low of 47-58 cm
(18.69 inches) to a high of 127.91 cm (50.36 inches). Most precipitation
occurs as rain during the growing season (see Table 46). Most summer
rainfall occurs in localized thunderstorms which usually move over the
watershed in a few hours. However, 24-hour rainfall amounts of up to 19-05
cm (7.5 inches) (July 17-18, 1964) have fallen on the watershed as a result
of a thunderstorm which became stationary over the watershed and was kept
active by convergent winds.
Rainfall is often unevenly distributed during the growing season. Con-
sidering agricultural needs of about 2.54 cm (1 in.) of rainfall during each
week of the growing season, the time distribution of rainfall within the
watershed is relatively poor. The probability of this amount of rainfall
occur ing during each summer week ranges from a high of 4 in 10 years in
early June and early August to 2 in 10 years in late July and late August.
Snow is the primary fora of precipitation from late November through March.
Although seasonal snowfall on the watershed averages about 102 cm (40 in.),
individual seasons have ranged from 28 cm (11 in.) to 280 cm (110 in.).
The probability of having snow on the ground reaches a high in mid-February
and then decreases sharply. The actual water content of snowfall on the
watershed varies with the individual storm, but averages about 10 percent,
that is, 25.4 cm (10 in.) of snowfall is equivalent to 2.54 cm (1 in.)
of ra i n .
Prevailing winds are westerly in Winter and southerly in the Summer over
most of the basin; but within 0 to 5 km (0 to 3 mi) of Lake Michigan,
northeasterly winds prevail during the period of April through June.
153
-------
TABLE 46. MEAN MONTHLY PRECIPITATION AT
MILWAUKEE, WISCONSIN
(1854 - 1964)
Month
January
February
March
April
May
June
July .
August
September
October
November
December
TOTAL
Mean
cm
4.75
4.19
6.12
6.91
8.26
8.87
7.62
6.96
7.87
5.82
5.15
4.37
76.89
Precipf tat ion
inches
1.87
1.65
2.41
2.72
3.25
3.49
3.00
2.74
3.10
2.29
2.03
1.72
30.27
Percent of
total
6.18
5.45
7.96
8.98
10.74
11.53
9.91
9.05
10.24
7.57
6.71
5-68
100.00
Wind speeds, neglecting gusts, can be expected to reach 88 km per hour
(55 mi per hour) at the 9.1 m (30 ft) level and 72 km per hour (45 mi per
hour) at the 0.3 m (10 ft) level in at least one out of two years. Speeds
can be expected to reach 161 km per hour (100 mi per hour) at the 9.1 m
(30 ft) level and 137 km per hour (85 mi per hour) at the 0.3 m (10 ft)
level once in 50 years.
Sunshine over the basin occurs 55 percent of the maximum possible time
during the year; 40 percent from November through February; 55 percent
March through May and during October; 60 percent June through September
and about 70 percent of the maximum possible during July.
Annual water surface evaporation is about equal to the mean annual
precipitation of 76 cm (30 in.), but 80 percent of this demand on water
supply occurs during the period May through October. Evapotranspiration
from soils and plants is normally less than water surface evaporation,
averaging about 53 cm (21 in.), most of which is demanded during the
growing season. Depending upon such factors as land use, temperature,
avallable water, and soil conditions, evapotranspiration will vary from
38 to 71 cm (15 to 28 in.).
154
-------
Flow Characteristics
The quantity of streamflow varies widely from season to season and from
year to year responding to variations in precipitation, temperature, soil
moisture conditions, agricultural operations, the growth cycle of vegetation,
and ground water levels. Since the quantity of streamflow is the product
of many interrelated hydrologic factors, the only practical way to determine
streamflow characteristics is to measure streamflow itself. In addition to
the natural factors which affect streamflow mentioned above, it should be
noted that effluent from several sewage treatment plants contributes to the
flow of the Root River.
High streamflows occur principally in the late winter and early spring,
usually associated with melting snow. Low flows persist for most of the
remainder of the year with occasional rises caused by rainfall. Under pre-
sent groundwater conditions, the lowest flows of the river appear to con-
sist almost entirely of sewage disposal plant effluent, without which flows
would probably drop to zero for considerable periods of time. In summary,
river discharge generally responds much more to Winter and Spring rainfall
than to Summer and Fall rainfall.
Water Quality Standards
The Root River has to meet the general use classification standards for
intrastate waters as adopted by the Wisconsin Department of Natural Re-
sources, September, 1973 00. These include standards for fish and
aquatic life and standards for recreational use.
Except for naturally occurring changes, the Root River shall meet the fol-
lowing criteria.
I. Dissolved Oxygen - Concentration should not be lowered to less than
5 mg/1 at any time.
2. Temperature - There shall be no temperature changes that may ad-
versely affect aquatic life. Natural daily and seasonal temperature
fluctuations shall be maintained. The maximum temperature rise at
the edge of the mixing zone shall not exceed 2.8° C (5°F). The
temperature shall not exceed 3I.6°C (89°F).
3. pH - The pH shall be within the range of 6.0 to 9-0 with no change
greater than 0.5 units outside the estimated natural seasonal
maximum and minimum.
4. Toxic Materials - Unauthorized concentrations of substances are
not permitted that alone or In combination with other materials pre-
sent are toxic to fish or other aquatic life. Questions concerning
the permissible levels, or changes In the same, of a substance, or
combination of substances of undefined toxlcity to fish and other
biota shall be resolved in accordance with the methods specified In
"Water Quality Criteria", Report of the National Technical Advisory
155
-------
Committee to the Secretary of the Interior, April I, 1968 (20).
5. Recreational Use - The membrane filter fecal coliform count shall
not exceed 2000 per 100 ml as geometric mean based on not less than
five samples per month, nor exceed 400 per 100 ml in more than 10
percent of all samples during any month.
These standards are used as a basis for comparison in later sections where
the water quality of the river is discussed.
Historical Water Quality of the Root River
Quality of water as conditioned by the natural environment of the watershed
would present no problem for any reasonable possible uses of Root River
systems waters. Most of the potential water uses have been, however, in-
compatible with past water quality factors resulting from human activity;
principally disposal of waste and, to a lesser degree, agricultural and
urban drainage.
A water quality sampling and testing program was carried out as part of the
SEWRPC Regional Land Use - Transportation Study in 1964; a second study was
done by the Wisconsin Department of Natural Resources during 196? and 1968.
Table 47 presents the data obtained from both of these studies. The find-
ings of these investigations indicated that serious pollution problems
existed in the Root River system and were intensifying.
As determined by the above studies, the variation of stream water quality
with respect to location and season is extremely great. Figure 55 depicts
the monthly variation of water quality parameters over the period 1961 to
1964 at the City of Racine. River discharge at the time of sampling has
a strong influence upon the concentration of pollution factors. Since many
pollutants are introduced into the River system at a relatively fixed flow
rate, high streamflows result in a greater dilution than do low flows. From
this historical data base it is easy to see that the natural stream puri-
fication potential is overwhelmed by the pollution load.
Root River Within the City of Racine
The test reach of the Root River, which for this project was defined as
that portion of the river extending from Lake Michigan upstream to the
area of the Horlick Dam (Figure 56), could be classified as a sluggish
stream environment. This reach of the river covered a distance of 9-7 km
(6.0 ml). Internal seasonal variance in Lake Michigan and artificial
external environmental stress caused by heavy industrialization along its
banks appear to keep the river in a state of chemical and biological flux.
The Root River has been dredged from the lakefront to State Street, a
distance of nearly I kilometer (0.6 mi) In an attempt to encourage use of
the waterway as a recreational focal point for private lake traffic. This
abrupt drop in the normal river bottom gradient seems to act as a mixing
156
-------
s:
"*^C
o
u>
.
•C"
c5
!Hd
H
<
OS
UJ
>
a: s:
t~ in
o 2:
o o co
cf, o vo
to CTi
UJ ™ T_
f-«" i»Sj"
* vO
U- LJJ Cn
o rr •—
>- o
!— tfi3"
r^ "•*»
O1
oc
UJ
3
•
-3"
Ul
cn
2
I—
. NOO-^OOsOO.*
E «Jl o o -a- mo o in-»
V 0 I CM CM CM CM
— I \o in o cn o CTI *— in
§CW CT\ *•• -3" CM (T\ P*» CO P^
c I _ » ~.
e, — ^. „
1
_ 1
in 4) {
-a > I
— — — oinoooomo
•— o*>. j-inmor^orncM
o i/> o1. vo cr* p^\o P*» co \o %o
10 in E
2
^*^
-__>>
in —
TO 10 —
— O 17.
0 I- E
«/) *^
in
3
o^ —
C (D *"^
Q.4J 0:
-'
a.
. CMCMin — -3-CMCOP>»
L.~v-i'cM-3'co
a> E|
— O o-»P«-ooinoo
*^ *— O , CMOO^-O*— vO r"%
»2 *Q^4 '^ ^ "
«S d
2
V 1 -3" -^ -T -ST -iT -^
U ^^ 1 vO VO vO ^3" vO A»O vO «^
w «•- df »•»•••» •v.yj >^ ^ -»»va
0 O E I ^3 ^J^ ^^ ^^ ^^ ^™ ^O ^^
a raj cn— CMCO CM — — vo
u>\ ^* *+, -+^ •**. *+* *** -^ •*+.
-a-o — o — ooLaou-voooooooLnoinoo
cr>m— "O — ooocviooinr^vo — rviooco — oocr>o>
— — .— — CM CM CM CM —
*H* 4^\ m co cn cn <^r «^ >«r ••r *~ co co ^3* ^M ^~ o^j r^ ^^ ^« m m
OOCP.OCMCMP>-CMOCM — P»— cMin\o«n
inooomino P'-J'OOJ'— oomcM— co — o«~o^o
vO CM U\ CM O Cr\vDOOCO*~CMO CO vO VO \O .^ •— in
vOP*p~\ovnp^-3-NOP»-o
co p*^co r**> P** r^» p*« p*^ r^» r^ co co co co p*» co r^^ o^> P^ P*^ r^^ oo
cn p^inco -co cn o r- r, c« 0 - P-VO cn ,r j- cn J- r- -
\o inincninj- in cnmj-co pninininj- cncMO in— CM
OOOOOOO — CM OCO— vOOOOOCMCOOO
vo r*^ ^- \o o CM p^-a" i •— — CM P*« o in -3* — \o «*•
CM — •-" CM -"• vO CO cn CM —
in in
vO xO ^O vO t^ \J\ vO \O vO vO vO VO vO VO vO ^^ ^^ vO *O vO vO ^O
***+ ***+ ***** *v. vO vO **««. **s» ^v ***<• ^^ '^'v ^» ^v *^»* OO C*4 *x. \O *^» **^ ^^
w, p^ |j^ ff\ v^ v,^ t/^ ^/^ WIV pv, ^f> 3Q
-------
1)
3
4J
c
lAOOlAlAOUMAOlAOlAUIOiALAOOOOlAOOOOOOOOOOO
r«>—tricocAoo—-T — oocoiM-3-—covo — c^r~»cMcncocMr-. CM -3- vO en CM
~ o •».,
•— in en
O ui E
in —
«--»
•o — —
— H-
u>
3
>. ^-v
5,-!
ex <-> ^-»
in O en
5t6
a.
mo
IAIAOO
CM OO
o m
oo
-3-
— o o
vO
co
CM CO
LTV CO
CM
o
O\OvO-3-3OvO-3--3'COCOeOCMvj3
-------
o o
o> m
1/9W N39AXO 03A1OSSia
!2 Si co * o
IWOOI/HdW OOOI Nl J.NDOO V\H3±OVS
Si
O ^
o o
111
o
X ID
o a:
«1
a:
ui
a.
.—-'
QC
<
UJ
TO
0)
2
*J
I
Q>
JC
C vO
•- en
li
o s
O ID
a
ra ~
D t.
V
ITl
0)
C7)
Sin o m
10 10 *
IN OOl/HdN
o
*
10
m
§22
0001 NI XNOOO vm3j.ova wtjodiioo
1/9W N39AXO OBAIOSSia
o
o>
o
IO
H3J.VM
159
-------
LU
O
UJ
in Q-
UJ
CK
OOD
T> O
ro
4J C
tf) —
0) O
O Q-
vO
cn
160
-------
zone between the waters of Lake Michigan and the river.
material from Ref. 19.)
Root Rfver Mon i tor i ng Sites
(End of condensed
The purpose of the Root River monitoring program was to determine the effects
of the treatment sites on the quality of the river in the test reach. To
achieve this end, three monitoring sites were located at different points on
the river.
One monitoring site was located at Horlick Dam, about 137 meters (150 yards)
north of Wisconsin State Highway 38. This site, designated as Site C, was
used as a control area and data gathered at this point indicated the water
quality of the river just before it entered the City of Racine. In addi-
tion to acting as a source of water quality control data, this site was
chosen because of its close proximity to a U.S.G.S. gauging station which
gives a continuous readout of the stage of the river.
The second monitoring site, Site B, was located downstream of S,ite C, about
8.8 kilometers (5.5 miles) and 2.k2 kilometers (1.50 miles) upstream of
the river mouth. Site B was located on the southern bank of the river on
the property of the Western Publishing Company. This site was selected as
a monitoring point because it was, after the construction of the treatment
sites, located downstream of the last known overflow points before the
river enters the area covered by the treatment sites. Data obtained from
this site indicated the water quality before the river entered the treatment
site area and, through comparison of Site C with Site B, showed the effect
of the upstream contribution to the river from the City of Racine.
The last site, Site A, was located in the test reach about 75 meters (246
feet) upstream of the outfall of treatment Site I and downstream, about
the same distance, from the outfall of treatment Site II. Site IIA was
located upstream of this monitoring site about 40 meters (131 feet).
Figure 56 depicts the relation of the monitoring sites on the river.
Additional Considerations for Monitoring Locations
Because of the nature of the monitoring program it was necessary to locate
monitoring sites in areas which would be fairly concealed from the general
public in order to avoid vandalism. Sites A and B were located using this
as a consideration. Site C was located in an area where the general public
had access, but because the area was frequented by many people, it was
felt that vandalism, generally a private act, would be kept to a minimum.
The location of Site A upstream of one of the treatment sites was necessary
because there were no sites between the Main Street Bridge and Lake
Michigan, which would have allowed monitoring the river without intei—
fering with the shipping channel.
161
-------
V-2 METHODS USED TO MONITOR THE RIVER
Continuous, Analog Data Gathering
The continuous monitoring portion of the project utilized a Honeywell
Water Quality Data Acquisition System at all three sites. The unit used
at Site A was a Model 202F while the two remaining sites each utilized a
Model I-WIOI. These monitoring systems were used to measure, in a con-
tinuous mode, dissolved oxygen, temperature, and conductivity, with the
data being printed out on a strip chart. In addition to these parameters,
wind velocity was also recorded at Site A.
Compos i te Samp11ng
This portion of the water quality monitoring studies consisted of taking
discrete water samples once every hour from time zero of a storm (the time
at which a storm caused an overflow) through hour No. 72. This sampling was
done using Sigmamotor automatic samplers, Model No. WM-124R. Each sampler
was capable of takfng 2k discrete samples of 120 ml each. One of these
samplers was placed at each monitoring site. When an overflow treatment
event occurred, each sampler was turned on and the time recorded as time
zero. Because each sampler only took 25 samples, the samples had to be
collected every 2k hours and the samplers recycled. The 2k discrete samples
were then composited into samples based on the time that they were taken.
Samples taken on day one of a monitored period (0 to 2k hours) were com-
posited Into two 12-hour samples or into four 6-hour samples. Days two and
three of each period had one 2^-hour composite sample made up for each day.
All composite samples were brought back to the laboratory and analyzed for:
Solids
Total
Suspended
Fecal coliform bacteria
Total organic carbon
Phosphorus
Biochemical Oxygen demand
PH
Analyses were done according to the methods given in Appendix IV-D. A
typical monitoring site with the equipment used for monitoring is depected
In Figure 57.
Grab SampI Ing
There were occasions during the monitoring program that grab samples were
taken from the river. This method of sampling was utilized when (I)
samples were taken from places other than the three monitoring sites,
or (2) during the early Spring prior to the placement of the sampling
equipment at the respective monitoring sites.
162
-------
o
163
-------
Miscellaneous SamplIng
In addition to the methods described above, which cover a major portion of
the Root River monitoring effort, various other techniques were used for
additional studies designed to provide supplementary information on water
quality. These techniques will be described when these studies are dis-
cussed.
V-3 RIVER MONITORING PERIODS
Determination of the water quality of the Root River was split into four
different periods: 1971 dry weather studies, 1971 wet weather studies,
1973 wet weather studies, and 1974 wet weather studies.
Studies performed In 1971 provide information concerning the quality
characteristics of the river prior to construction of the treatment systems.
Data gathered during 1973 and 1974 give an indication of the quality of the
Root River after the treatment systems went into operation.
All storm numbers assigned for the river monitoring periods during 1973
and 1974 correspond to the run numbers used in the discussion of the treat-
ment sites.
.19.71 River Monitoring
The 1971 dry weather events consisted of monitoring periods performed after
a period of at least two weeks had passed without a storm-generated dis-
charge. Each event, with one exception, Was monitored for three consecu-
tive twenty-four hour periods. There were seven dry periods monitored
during 1971. Table 48 is a list of these monitored events.
Twelve wet weather events were monitored during 1971. Table 49 Is a list
of the dates on which each of these events started. A wet weather event
was designated as a rainfall which caused storm-generated discharge.
During these events samples were taken while a discharge event was
occurring at one-hour periods. Monitoring of each of these events ran for
three consecutive 24-hr periods from the time that the first samples were
ta ken.
1973 RIver MonI torIng
During 1973, twenty-one storm-generated discharge events were monitored.
Table 50 Is a list of the starting dates of each of these events. Monitor-
ing of these events was carried out in a pattern similar to that used
during the 197! storm events. Of the 22 events, only about 50 percent of
these were monitored completely.
There were no dry weather periods monitored during 1973.
164
-------
(Tv
o
o
CC
LU
a.
e£.
o
to
LU
a:
o
LU
_l
CQ
CEO)
cn v. JD
— OS
(A 4J 3
(A (A C
.E
4-1
«— CM
trv \e
oo
O — CM
I
4)
_ ,- _ cn
3333
V) (A
cn
3
!_
4>
4) 4)
cn a.
3 4)
4J O 4)
0. 4J >
4) O O
H
(01 CO
o| -.
COvOCA
«— CM
CM
OfA
-" CM
O
CM
CM »-
CM CO
^,
en
r-*
i
••
a:
o
a
LU
o
H-
co
-=T
LU
-1
CQ
^f
1-
"O
I. T3
O O
+J •••
••• L>
C 0)
O d-
3E
"O
41 >-
C 0
cn 4J
in c
(A Q
^C c
LB
4>
C
L»
4)
•§
cz
c
j:
4^
0
J(^
re
o
^- CM CO
/-— — -^^— ^-* /""^^^^-N -^w^—.j^«^*'"v
f
*•— CM CO -51" LA vO fx. CO CTi O
U
5
4)4)4) 4) >- >- >.>.>. 4-1
333 333 333 Q
->->—* -»->-» —>—>—> ••— '
T3
4)
{^% C3 *i— * <^^ P««» ^^ BCf" Lj% vO trt
r-» v— C^ T™ ««• *•>• (/)
•«•
z:
-•*}"
**-%
fQ
^
10
T3 «— CM PO
E
o
(A
TO
H"
O
4J
D. 333
Ix* CO O"^
CM CM CM
LTV VD fx.
•™^w^% ^V^^s A
f ,*"">*«' , /"^^»^V^^^
JT LTV vO rxCO V4> E E E
33 *-• 4-1 4J 55 0) S
cn cn c^ Or CL ^ sx >
33 4)4)4) O _O O
<< to in vi ^ 2 -z
C^ O rx.cO CTv O *"• CM
165
-------
TABLE 50. MONITORED STORM PERIODS
- 1973
TABLE 51. MONITORED STORM PERIODS
- 1974
Day
29
1
7
8
25
27
5
15
16
2
3
20
4
17
21
2k
28
12
27
31
14
4
Date
Month
April
May
May
May
May
May
June
June
June
August
August
August
September
September
September
September
September
October
October
October
November
December
Assigned
storm
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Day
28
3
18
28
5
8
11
13
14
16
16
21
21
6
6
9
11
3
10
22
25
10
16
Date
Month
March
April
April
April
May
May
May
May
May
May
May
May
May
June
June
June
June
July
July
July
July
August
Auqust
Assigned
storm
No.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
1974 River Monitoring
Twenty-three storm*>generated discharge events occurred during 1974 which
resulted in the operation of the treatment sites; these are listed in
Table 51. Monitoring of storms during 1974 was 75% complete. Twenty-five
percent of the data was missed due to equipment failure and/or because
the monitoring equipment had not been Installed during the early portion
of the storm season.
166
-------
V-4 MONITORED PARAMETERS AND RESULTS
Rainfall Measurement Program
Collection and characterization of rainfall events during the monitoring
periods was important and necessary to overall project objectives. Data
obtained from the U.S. Weather Bureau Station in Racine, the only station
/oojf ?°°? RIV6r Basln» jnd'cated a mean annual precipitation of 82.45 cm
(32.46 inj. The U.S. Weather Bureau Station in Milwaukee has recorded a
56 year mean precipitation of 75.39 cm (29.68 in.), ranging from a low of
47.7 cm (18.69 in.) to a high of 127.91 cm (50.36 in.).
Three raingages were Installed at selected:locations..within the .Racine area
of the Root River drainage basin to provide detailed rainfall background
data within the perimeter of the selected test reach. The raingage loca-
tions during the 1971 monitoring season are shown in Figure 56 and are
designated as follows:
Gage Rl - on the roof of the Racine Police/Fire Headquarters Building
at Center and 8th Streets;
Gage R2 - on the roof of the Racine Zoological Gardens Main Office
Building one-half block east of Main Street on Walton;
Gage R3 - on the roof of Fire Station No. 2 at North Memorial Drive and
High Street.
The raingages installed were all Bendix gages, Model No. 775-C. The
Model 775-C has a knife edge collector of 20.3 cm (8 in.) inside diameter,
constructed of nonferrous material. The catch is funneled into a bucket
which has a 30.5 cm (12.0 in.) rainfall capacity. The bucket is mounted
on a spring-type weighing mechanism which converts the weight of precipita-
tion Into its equivalent weight in inches of rainfall and actuates a pen
which traces an inked record on a 15.2 cm (6.0 in.) revolving chart. The
record is of the dual traverse type; the pen sweeping across the chart from
bottom to top for 15.2 cm (6.0 in.) of rainfall and then reversing to
return to the bottom for an additional 15.2 cm, thus recording 2.54 cm
(l.O in) of rainfall for each 2.54 cm (1.0 in.) of chart. The chart is
attached to a vertical cylinder which is rotated by an, internal drive
spring once per day for an eight day period. The accuracy of the 775-C gage
is 1/2 of I percent of full scale (+^0.15 cm) (0.06 in.).
Each site was maintained on a seven-day interval or after each rainfall,
depending on which came first. The maintenance included changing the
recording chart, refilling the recording pen, and checking the instrument
calibration. Summaries of the collected rainfall data during the 1971,
1973, and 1974 monitoring periods are included in Tables 52, 53, and 54,
respectively.
The notification and alarm system which initiated a wet weather sampling
period was very closely tied to the rainfall measurement program. Selected
167
-------
TABLE 52. RAINFALL SUMMARY FOR RACINE, WISCONSIN FOR 1971
Date
rainfall
event
6/18 Ma
6/20
7/2
7/8 M
7/16 M
7/19
7/20
7/23 M
8/2 M
8/10 M
8/18
8/22 M
9/5
9/20 M
9/29
9/29
10/3
10/13
10/21
11/1 M
11/18 M
Total
, cm
1.1*
0.25
1.78
2.67
1.07
1.35
0.25
0.86
0.7*
2.16
1.52
1.32
0.89
1.73
1.02
0.*6
2.67
0.56
0.69
0.99
0.31
rainfal 1
(inches)
(0.45)
(0.10)
(0.70)
(1.05)
(0.*2)
(0.52)
(0.10)
(0.3*)
(0.29)
(0.85)
(0.60)
(0.52)
(0.35)
(0.68)
(O.*0)
(0.18)
(1.05)
(0.22)
(0.27)
(0.39)
(0.12)
Duration
hou rs
0.50
0.66
1.16
3.66
0.66
1.33
0.16
1.16
2.83
3.00
2.50
0.50
6.00
*.33
0.16
0.33
1.00
2.66
18.00
1.33
1.33
Average
cm/hr
2.29
0.39
0.53
0.73
1.62
1.01
1.59
0.7*
0.26
0.72
0.61
2.6*
0.15
O.*0
6.35
1.39
2.67
0.21
0.0*
0.75
0.23
intensity
(inches/houry
(0.90)
(0.15)
(0.60)
(0.29)
(0.63)
(0.39)
(0.60)
(0.29)
(0.10)
(0.28)
(0.2*)
(1.0*)
(0.06)
(0.16)
(2.*0)
(0.5*)
(1.05)
(0.08)
(0.62)
(0.29)
(0.09)
a. Indicates a monitored storm event.
168
-------
TABLE 53. RAINFALL SUMMARY FOR RACINE, WISCONSIN FOR 1973
Date of
rainfal 1
event
4/27
5/1
5/7
5/8
5/25
5/27
6/5
6/15
6/16
7/2
7/3
8/20
9/4
9/17
9/21
9/24
9/28
10/12
10/27
10/31
11/14
12/4
Average
Total
cm
1.78
1.78
-
-
0.97
2.54
0.99
0.48
0.58
-
-
4.65
3.73
2.57
0.58
1.52
3.68
1.78
1.58
1.45
2.36
3.25
rainfal 1
(Inches)
(0.70)
(0.70)
-
-
(0.38)
(1.00)
(0.39)
(0.19)
(0.23)
-
-
(1.83)
(1.47)
(1.01)
(0.23)
(0.60)
(1.45)
(0.10)
(0.62)
(0.57)
(0.93)
(1.28)
Duration
hours
5.25
2.00
-
-
1.78
15.00
0.58
1.31
0.50
-
-
3.00
1.72
13.90
2.00
4.67
7.58
10.42
14.53
9.64
7.92
14.36
intensity
cm/hr
0.34
0.89
-
-
0.54
0.17
1.71
0.37
1.17
-
-
1.55
2.17
0.18
0.29
0.33
0.49
0.17
0.11
0.15
0.30
0.23
in/hr
(0.13)
(0.35)
-
-
(0.21)
(0.07)
(0.67)
(0.15)
(0.46)
-
-
(0.61)
(0.85)
(0.07)
(0.12)
(0.13)
(0.19)
(0.07)
(0.04)
(0.06)
(0.12)
(0.09)
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
169
-------
TABLE 54. RAINFALL SUMMARY FOR RACINE,SWISCONS IN FOR 1974
Date of
Rainfall
event
Total
cm
Rainfall
inches
Duration ,
hours
Average
cm/hr
Intensity
fn/hr
Run
No.
3/28
4/3
4/18
4/28
5/5
5/8
5/11
5/13
5/14
5/16
5/16
5/21
5/21
6/6
6/6
6/9
6/11
7/3
7/10
7/22
7/25
8/10
8/16
+• —
1.57
0.66
3.81
0.30
1.09
1.14
0.51
1.32
1.98
0.53
0.64
1.68
0.53
1.52
2.59
0.41
1.22
0.78
0.33
0.51
1.42
__
0.62
0.26
1.50
0.12
0.42
0.45
0.20
0.52
0.78
0.21
0.25
0.66
0.21
0.10
1.02
0.16
0.48
0.31
0.13
0.20
0.56
nM
4.8
3.3
4.5
1.5
11.2
7.5
0.8
4.0
5.0
3.8
2.2
14.0
4.5
5.3
5.0
-1.0
13.0
9.7
1.0
2.8
2.5
6.0
-»•
0.32
0.20
0.85
0.20
0.10
0.15
0.11
0.33
0.40
0.14
0.30
0.12
0.12
0.28
0.52
0.41
0.13
0.79
0.12
0.20
0.24
w w
0.13
0.08
0.33
0.08
0.04
0.06
0.24
0.13
0.16
0.06
0.12
0.05
0.05
0.11
0.20
0.16
0.41
0.31
0.05
0.08
0.09
23
24
25
21
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
170
-------
personnel from the Racine Water Pollution Control Plant were delegated the
responsibility of notifying the appropriate Envirex personnel in the event
of a rainfall occurrence during 1971. Oh notification, the designated
Envirex employee immediately proceeded to Racine to ensure against equipment
malfunction and to collect the initial samples for that period. No alarm
system was necessary for the monitoring of dry weather flow. During 1973
and 197** an automatic telemetered notification system was used to inform
Envirex employees that an overflow event was occurring.
The U.S. Geological Survey maintains a gauging station at the head end of
the test reach, about 200 meters (650 feet) downstream of Site C river
monitoring station. The Madison, Wisconsin office of the Geological
Survey made up-to-date discharge data for this station available for use.
An estimate of the stability of discharge at this site during 1971, 1973
and 197^ wet weather monitoring periods was obtained by computing the mean
monthly discharge from the daily discharge measurements and calculating
the coefficient of variation (21) of the daily measurements around the
monthly mean. The mean monthly discharge and coefficients of variation
for discharge are shown in Table 55. During the 1971 monitoring period the
river discharge was extremely low when compared to 1973 and 197^ data. Dis-
charge measurements during 1973 and 197^ were similar with discharge
measurements during 197^ being slightly higher than those in 1973.
Specific Conductance
All natural waters contain ionized materials and the amount of this material
in a given system can be estimated by measurement of its specific conductance
or conductivity. Units of measure for specific conductance are expressed
in terms of micromhos per centimeter at 25°C. It is the reciprocal of the
specific resistance of a solution. In most bodies of natural waters the
relationship of specific conductance to dissolved solids is linear (22).
Because of this relationship, either parameter may be used as an
approximation of the other. This relationship was confirmed in the Root
River water samples by running determinations of each parameter in the la-
boratory and using the data generated in a statistical analysis. The
analyses used consisted of using "six curves", "best fit" and "regression
analysis" programs, which are part of a statistical package offered by the
Service Bureau Corp. computer center (23). The output from these
analyses is presented in Figure 58. These analyses showed that a linear
relationship existed between the two parameters with the correlation
coefficient having a value of 0.867.
Measurement of specific conductance was done through the use of three
Honeywell Water Quality Data Acquisition Systems, one at each of the three
major sampling sites. Checking the readout of these instruments with ex-
ternal instrumentation (Beckman Instrument Co., Model Mo. RB-3) indicated
that measurements of this parameter by the Honeywell units were very
accurate ( 5%) .
Specific conductance measurements were made during 197' > '973t and
with data from 197' being split into two periods: 1971 wet weather, and
1971 dry weather. All data for 1973 and 197^* were taken during wet weather
171
-------
en
UJ
(3
*T^
mj>~
O
to
o
ae
p o
u. uj
£C
z o
0 J-
P z
g
OC
< CO
>jE
u- 2=
o o
£*
•z. o
UJ U.
O UJ
— z
it ;;;
u. o
UJ <
8*
t—
0 <
5 o:
co£
§ a:
Si-
0
>• s
_j o;
Pa,
§£
Su-
o
•
in
in
UJ
_j
CO
<
P
• I
>|
0
= 1
•
4-»
(J
o
• I
*J]
pi t
51
c/>!
• I
31
O
(D (B -M
•5-5S
.52 .12 *J3
o o —
u-
— e c. «4-
r>> n> nj aj
cn G) o o
*^»
!**•
CM
CM
cn c*^ •
CM
cn
r*«^
CM
CM^I 0
JJ-
f\
in
-*
O vO •
\o m o
o
c Ol O
t_ u
flj ID *J
U U V
«•» ••» CJ
Q Q ._
l*-
inn c c M-
f*v <0 ID Q)
cn fl) fl) O
in
o
cu
co r*. •
CM — 0
. «
CO **"J™ CJ
1 S
E —
^x 4«*
E <^*^ ^
t/i •*••
3 ^^ ^>
O O flj
«J ««*-
cn cn o
u u
ID <0 4J
o °o a)
«- .— *o
Q 0 .-
M-
-T C C «t-
f*^ ID ID 0)
cn JD jo o
— s z: o
172
-------
Coeff icien
c
^O
4J
U
3
O
U
ra
41
c
_i
CO
0
VO
CM
CO
•
O
^"s
*o
c
•H*
X
LA
CM
O
O
•
O
a.
5{
LiJ
CO
o
^f
c^
IA
•
ca
en
0
a.
0)
Q
^•»
(0
•~
4J
4)
C
O
a.
X
LU
vO
CO
•
o
0^
^»
IA
en
CO
vO
en
o
o.
X
LU
•o
c
•^
X
«HO
CM
CO
-a-
•
II
Q.
4)
O
C
O
•«•
*J
u
c
3
U.
u
Q|
5
a.
CM
•»
H
a.
0
o
u
*w
»w«
O
J3
l_
0
Q.
>.
^^
^^
a\
•
o
**^*
•o
c
•M*
X
o
a
o
o
*
o
4
^^
•••-
vO
«
O
*^
c
n
Q.
o
u
*™
1
0
Q.
^
M
—
*o
vt
•o
0
>
"o
(A
M
•••
T3
«4-
o
a.
•«•
JC
(A
C
o
4J
(D
£
o>
1
j:
(A
•M
3
O
4J
c
u
0.
u
0
•j
3
0.
U
CO
IA
0
u
3
cn
iZ
0
u
c
Q
*J
U
3
•a
c
o
u
u
• 1
*^~
"o
0
o.
M
o
*J
'iw
2
re
*u
•5
*j
0
•a
c
0
Q.
0
•o
^^^
c
^O
4J
0
"O
0
o.
0
•o
c
0^^
CM
u-v
173
-------
periods. Measurements of specific conductance for a wet weather period
were run for three consecutive days (day I, day 2, and day 3) with day 1
being the day the rainfall and resultant discharge occurred. During the
1971 dry weather studies only six of the seven three day periods were
monitored for this parameter. Wet weather monitoring during 197' occurred
during all twelve storm events. During 1973 a total of 22 storm-generated
discharge events occurred of which nineteen were monitored for this para-
meter. In 1974, a total of 23 events occurred of which 21 were monitored
for this parameter. Statistical analysis of the data, at the 95? confi-
dence level, was carried out comparing years and sites.
In all four of the monitoring periods there were no significant differ-
ences noted at Point C although there was a decrease in this parameter during
1974 compared to the other monitoring periods. Analysis of data at the
other two points did show a significant increase in this parameter between
1971 and 1973 for wet weather data, as shown in Table 56, as well as a sig-
nificant difference between 1971 wet and dry weather studies. These analyses
indicate that the water quality in the river has decreased between 1971 and
1973? using specific conductance following a discharge event as the indicat-
ing parameter. Comparison of specific conductance values between 1973 and
197^ at Point A indicates that, as happened from 1971 to 1973, the quality of
the River decreased significantly. This same trend was noted at Point B
It can be inferred from the data that degradation of Root River water quality
is being caused by some input upstream of the Horlick Dam.
Temperature
Rivers and streams, as opposed to lakes, have a temperature regime that
fluctuates daily, as well as seasonally. The amount of fluctuation observed
in a running body of water depends a great deal on the depth, width, and
volume of flow. In general, the deeper the body of water the less the
variation will be in its daily temperature cycles, as well as in its annual
temperature cycle. The temperature of a river or stream is related to the
air temperature of the area that it passes through. Due to the fact that
rivers and streams are moving bodies of water, there is little or no
thermal stratification over the course of the year. If a period of stra-
tification does occur, it only lasts for a very short interval because of
the constant mixing action of the moving water.
Temperature affects the biota in a body of water directly and indirectly.
The direct effect Is related to the fact that each organism has a tempera-
ture range that It can tolerate. The indirect effect is related to the
amount of dissolved oxygen (DO) that is soluble in water. The relation-
ship of DO to temperature is an inverse one, that is, as the temperature
increases, the solubility of oxygen In water decreases.
For the stretch of the Root River being studied, the greatest dally varia-
tion in temperature is found at Point C. The physical makeup of the river
at that point is a shallow water area that is not shaded by trees and
therefore receives all of the sun's solation. Point A has a lower value at
the hlqh end of its range, but this fs prooaoiy due to the Influence or
174
-------
TABLE 56. SPECIFIC CONDUCTANCE GRAND MEANS FOR
ALL SAMPLING PERIODS (y mhos/cm)
Samp) ing
Period
Dry 1971
Storm 1971
Storm 1973
Storm 197*
Sampling
Period
Dry 1971
Storm 1971
Storm 1973
Storm ,197*
Sampling
Period
Dry 1971
Storm 1971
Storm 1973
Storm 197*
Day 1
217
2*5
*22
51*
Day )
*l*
50*
760
62*
Day 1
820
803
838
73*
Main Street
Day 2
226
252
*20
528
Western Publishing
Day 2
389
532
723
61*
Horlick Dam
Day 2
819
784
782
7*3
Pay 3
222
258
425
526
Day 1
369
575
69*
620
Day 3
856
77*
781
720
Lake Michigan at this site. Site B is similar to Site A except the "lake
effect" is not as pronounced.
The effect of storm-generated discharges on the temperature regime of a
river system is dependent on two factors: first, the season of the year
which in turn affects the temperature of precipitation received; second,
the temperature of the surface that the runoff water travels across which
175
-------
is also dependent on the reason of the year. These two characteristics
of storm events and their effects on runoff determine the final temperature
of runoff water as it enters a receiving body. The added water, with its
temperature characteristic, will change the temperature of the receiving
body of water. Appendix V-A, Tables A5, A6, A7, and A8 give the daily mean
temperature and daily temperature maximum and minimum of the river at each
of the three sites for each day of the monitored period.
D i sso1ved Oxygen
The amount of dissolved oxygen in a moving body of water is highly dependent
on many different factors. These include size of the body of water being
examined, amount, and type of pollutional load it carries, water tempera-
ture and the biota that live in its waters. It has also been demonstrated
In large rivers that one of the main Influences on the concentration of
dissolved oxygen is the flow rate of the river.
In rivers (and other bodies of water) which receive inputs of organic matter,
such as the inputs from storm-generated discharges, the amount of oxygen
needed on a per day basis is increased above the amount needed to maintain
the Indigenous biotic community. If the Input of organic matter (as well
as other oxygen requiring pollutants) from point sources is great enough,
It may reduce the oxygen concentration in the water below the lower
tolerance of the respiratory needs of the biota, placing them In a stressed
situation.
Data for each of the days monitored Is expressed as a mean for that day
along with the minimum and maximum values recorded for that day. Tables
A9, AIO, AM, and AI2 of Appendix V-A give this ..information for the four
major sampling periods. The grand mean, minimum and maximum values, In
mg/1, recorded over the three major sampling periods at each point are as
follows:
[
Point A
Mean
Minimum
Maximum
Point B
Mean
Minimum
Maximum
Point C
Mean
Minimum
Maximum
1971
)ry weather
8.1
5.2
10.8
5.4
1.0
8.5
2.3
O.I
5.5
Dissolved oxygen content (mg/1)
1971 1973
Wet weather Wet weather
7.1
3-7
10.4
5.3
I.
10,
2.8
O.I
9.0
7.2
0.8
11.8
6.1
1.0
13.6
6.9
O.I
1974
Wet weather
8.3
5.0
11.8
7.4
0.0
16.4
7.4
0.0
17.1
176
-------
Analysis of the data Indicated that there was a decrease in the mean
dissolved oxygen value at Point A from dry to wet weather for 1971. At the
same time,, an increase of this parameter for the same period of comparison
was noted at the other two points. Also, there were no significant differ-
ences between the sampling points when the means for all monitored events
were compared; this comparison included all 1971 and 1973 data.
Comparison of 1971 storm data indicates that, when all points are compared,
a significant difference does exist between points. The trend shows that a
general improvement in water quality occurs from Point C downstream to Points
A and B. When the 1971 and 1973 storm data are compared by point, an overall
improvement in water quality occurred from 1971 to 1973 along the entire
length of the test reach. The greatest improvement in water quality during
the entire monitoring period occurred at Point C where the difference be-
tween 1971 and 1973 storm events proved to be statistically significant.
Statistical comparison of 1971 dry and wet weather events at this site show
no significant differences. Analysis of data gathered at Point B revealed
no significant differences between 1971 and 1973 storm data or 1971 wet and
dry weather data. Dry and wet weather events of 1971 were compared for
Point A. Statistical analysis revealed that the difference between the two
periods at this Point was significant with a decrease in this parameter beincr
noted from dry to wet weather events.
The data gathered in 1974 showed an increase in this parameter at all points
when compared with the three previous monitoring periods. Statistical
comparison of 1973 data to 1974 data at Point A indicated that the differ-
ences during day-I and day-2 of the monitored events was significant while
the differences noted during day~3 were not -significant. These differences
indicated an improvement in water quality from 1973 to 1974. Comparison
of 1973 data to 1974 data for each of the days monitored at the two re-
maining monitoring points indicated no significant differences, although
the general trend was an overall Improvement in water quality. Comparison
of all points to each other during 1974 indicated that no significant
differences existed.
Solids
All natural waters contain a certain amount of solids. These solids mav be
divided into two components, dissolved and suspended. In this
study the difference between these two fractions is based on the abil-
ity of dissolved solids to pass through a glass fiber filter while the
suspended solids are retained. Collectively, these two components make
up the total solids load of a body of water. The material which makes up
the solids load is both organic and inorganic although it has been stated
that the true solids component is only made up of inorganic material (24).
Historically, the dissolved solids content of a water body has been related
to Its fertility as well as to its specific conductance. The relationship
of dissolved solids to conductivity on the Root River has been discussed
previously in this section.
The input of solids to many streams and lakes In the United States is
177
-------
derived, in many cases, from the waste discharged by industry and municipal
waste treatment plants. Effluents from these sources are generally higher
in dissolved solids than they are in suspended solids. A major portion of
the suspended solids load results from soil erosion and/or algae blooms.
The effect of a high suspended solids load on a water body is to inhibit
primary production by decreasing the amount of light entering the system
and/or changing its quality.
During the four monitoring periods (1971 Dry Weather, 1971 Wet V/eather,
1973 V/et Weather, and I97^» Vet Weather) measurement of total solids and
suspended solids were made. Dissolved solids content was calculated from
the difference between the two. Appendix V-A contains the values obtained by
these measurements. Statistical analysis of these results was accomplished
through the use of the Student t Test or an analysis of variance test.
All statistical analyses were carried out at the 35% confidence level.
Total Solids - During the four major monitoring periods, determinations were
made of the total solids content of the Root River. Tables A13 to A16
in Appendix V-A give the values obtained by these measurements for all of
the monitored periods.
Statistical comparison of data for this parameter during the two separate
monitoring periods of 1971 (wet and dry weather) at each of the monitoring
points indicates that there were no significant differences between wet and
dry weather flow. When points were compared to each other for each of the
two 1971 monitoring periods, significant, differences were observed, with
an increase in the parameter proceeding upstream from Point A to Point C
(see Table 57) during both dry and wet weather.
Wet weather data comparisons between 1971 and 1973 at each site indicate
that there was no significant change at Point C, but that significant changes
did occur at Points A a"nd B. The change in this parameter was a 37% and
27% increase for Pbint
-------
TABLE 57. TOTAL SOLIDS - MEAN VALUES
FOR ALL MONITORED PERIODS
Qpg/0
Hours
1971
1971
1973
1974
Dry
Wet
Wet
Wet
0-6
—
230
389
518
7-12
—
219
402
464
13-18
—
205
361
461
19-24
—
238
398
477
0-2^
218
225
385
465
25-48
241
244
387
457
49-72
213
270
396
433
Site
A
A
A
A
1971 Dry
1971 Wet
1973 Wet
1974 Wet
384
626
614
406
615
623
424
586
576
417
620
644
416
407
613
617
435
445
558
622
383
453
622
612
B
B
B
B
1971 Dry
1971 Wet
1973 Wet
1974 Wet
—
647
674
651
—
653
661
681
—
754
672
673
—
687
641
697
730
687
662
680
726
715
672
673
741.
705
618
694
C
C
C
C
Point A, Point A having a lower mean value for this parameter than the
other two points.
When comparisons of 1971 wet and dry weather data are made, there is a
noticeable amount of fluctuation during the wet weather periods which is
not apparent during dry weather periods. This trend is most predominant at
Points A and B which is to be expected because of the input from the city
streets via storm sewers and combined sewer overflows during runoff events,
as well as the influence from the lake.
Suspended Sol ids - Tables A17-A20 (Appendix V-A) present data on the amount
of suspended sol ids found in the river. Point C had the highest amount of
suspended solids and Point A had the lowest amount during periods of low
flow (1971 dry weather). Comparison of 1971 wet and dry weather flow at
Point C and Point B showed ho significant differences when compared using
the Student t statistic. There was a significant difference in this
parameter indicated at Point A. Comparison of the two different wet-weather
monitoring periods at each site reveal that suspended solids were not
significantly different at Point C between 1971 and 1973, but that there was
a significant difference at the other two points. The 1973 values at
Point B and Point A were greater than the 1971 values; the 1973 values were
179
-------
and l\3% greater, respectively, than the 1971 values. The Increases
are probably a result of events whJch are occurring somewhere along the
reach of the river in Racine, such as outfalls from combined sewers and
storm sewers, as well as an overall increase In river discharge.
Events monitored during 197^ showed a significant increase in the parameter
at PointAfrom 1973. Suspended solids during \37k were 1502 higher than
they were during 1973 at this point. Although there were increases noted at
Point B and C from 1973 and 197*1, the increases were not statistically
significant. 'The percent increase at these two points from 1973 to 197't
were 1002 and 1692 for Points B and C, respectively.
Pisso1yed Solids - Dissolved solids, as indicated by the data in Tables
A2I-A24 in Appendix V-A, make up a major portion of the total solids content
in the test reach of the Root River, and reflect similar trends noted for
total sol ids.
Phosphorus
Phosphorus is the element most pointed to when eutrophication of a body of
v/ater is discussed. It is thought that in most cases of eutrophication, if
the amount of phosphorus entering a body of water can be controlled, the
rate of eutrophication can be controlled. Because phosphorus is a common
element found in domestic sewage, different types of industrial waste,
combined sewer overflow and stormwater runoff, it was felt that measurement
of this element was warranted.
In Appendix V-A, Tables A25, A26, A27, and A28 represent the data generated
during the study of this parameter. The units used in measurement are mg/1
of phosphorus. During the 1971 dry and wet weather periods, the ortho form
of phosphorus was measured. Determinations of phosphorus during the 1973
and 197^ wet weather monitoring period were made on the total amount of
phosphorus in the sample.
The mean concentrations of orthophosphorus during 1971 at the Points B and C
for the dry weather monitoring period were very similar. The arithmetic
means of the data for these two areas during the 1971 dry weather monitoring
period were 0.16 and 0.20 mg/1, respectively. The mean value at Point A
for this period was 0.02 mg/1. This shows a definite effect of encroachment
of Lake Michigan waters into the studied reach. As reflected by the low
values, the effect of encroachment is to dilute the phosphate concentration
in the river water at Point A. Seasonal trends in this parameter may exist
in the test reach but due to the effects of the lake, which vary with wind
direction, these trends may be concealed.
During storm events of 1971, there was an increase in this parameter at
Point A but a general decrease at the other two points (see Table A26).
There are probably two reasons why this parameter decreased at these two
pointsdurlng the storm events. First, the total volume of water passing
a given point increased and had a dilution effect. Second, and probably the
180
-------
more important reason for a decrease in this parameter at the two points was
the increase in seston content of the water (see Solids and TOC), which
would tend to sorb this form of phosphorus out of solution. Data for Point B
during day-1 of a storm, especially during the first six hours of a storm.
This is most likely due to the first flush phenomena of combined sewers
occurring during this time and increases the seston load in the river.
Although data for Point C is scant, the same pattern seems to develop as was
demonstrated at Point B.
Due to the kinetics of the seston - orthophosphate interaction, the deci-
sion was made to measure total phosphorus during the 1973 and 1974 wet
weather monitoring program. Because of this decision there is no way dry
and wet periods of 1971 and 1973 and 1974 weather periods can be compared.
The 1973 wet weather data for total phosphorus Is presented in Table A27.
It shows that during the first six hours of the collective events there
was a considerable amount of total phosphorus at Point A, about 200% more
than was at either of the other two monitoring points. When a comparison
of means is made (Table 58) for day-1 (0-24 hours) of the 1973 wet weather
events, no difference1 between the three monitoring points is noted. For
TABLE 58. COMPARISON OF TOTAL PHOSPHORUS
YEARLY MEANS AT THE THREE RIVER
SAMPLING POINTS
Hours After
Overflow
Bepan
0-6
0-24
25-48
49-72
Poi
1973
0.63
0.31
0.33
0.26
nt A
1974
0.33
0.23
0.24
0.23
Point
1973
0.31 .
0.31
0.26
0.29
B.
1974
0.30
0.33
0.32
0.28
Point
1973
0.34
0.33
0.31
0.43
c
1974
0.28
0.33
0.37
0.37
day-2 (25-48 hours) comparison of this parameter, both Point A and Point C
are similar to each other, as well as to day-1 of the monitored events;
Point B shows a decrease in this parameter from day-1 to day-2 of the
monitored events. During day-3 (49~72 hours) of the 1973 storm monitoring
period, there is an increase at Point C when compared to days-1 and-2,
although this increase is not statistically significant. On day-3 of
these events Point B also shows an increase, although the magnitude of
this increase is not as great as the increase indicated at Point C.
181
-------
It is felt that the increase on day-3 at these two points is due to runoff
entering the test reach from farther upstream. Day-3 of the 1973 events at
Point A shows a decrease from days-1 and-2 at this site; this is probably
due to dilution caused by the encroachment of Lake Michigan in the area.
Data gathered during 197^ (see Table A28) , show a decrease in this parameter
during the first six hours of the storm events at Point A when compared to
the 1973 data. Data gathered at the other monitoring points during this
period (0-6 hours after the start of a storm) is similar to the 1973 data at
the respective sites. The decrease in this parameter at Point A is thought
to be caused by (1) the dilution effect of the lake and/or (2) removal of
phosphorus by the treatment sites.
Comparison of data at Point A for days-1 , -2 and -3 during 197^ to the other
two points and to the 1973 data shows a decrease in this parameter while it
remained esentially the same at the other two monitoring points. Again,
as in 1973> statistical comparisons of the data did not reveal any
significant differences.
Tables A29-A32 in Appendix V-A contain the pH values obtained at the three
monitoring points on the Root River. The values obtained indicate that for
the entire monitoring period a highly buffered and therefore very stable
system exists in this reach of the Root: River. The highest recorded pH
value was 8.60 and the lowest value was 6.65. Based on the yearly means
the range between the maximum and minimum values was not more than 0.2 pH
units. This relationship generally held true for the individual storm
events .
Data for this parameter during 1971 showed a general increase from day-l
to day-3 of a given storm event; just the opposite of what occurred
during the 1973 monitoring season. During 1973> there was a decrease in
this parameter at a given point over the three consecutive days of monitoring,
During 1974, the trend was similar to that of 1971; that is, there was a
general decrease in this parameter from day-l of storm events through day-3-
In terms of water quality, pH variations in a range of 6.7 to 8.6 with
extremes of 6.3 to 9-0 will support fish and other forms of aquatic life
without problems. The permissible range of pH that a given organism can
tolerate depends upon many other factors such as dissolved oxygen, tempera-
ture and prior acclimatization. The Aquatic Life Advisory Committee of
ORSANCO (1955) considered changes in pH between a range of pH 5 to pH 9
to be unharmful to fish (25). In view of these recommendations, pH
fluctuation in the Root River may be considered as a noncritical factor in
terms of the river's biota.
Total Organic Carbon
The measurement of total organic carbon is one method used to determine
182
-------
the level of organic enrichment of a water body. The direct measurement of
carbon can be used to replace the more time-consuming BOD and COD determina-
tions.
Tables A33 through A36 in Appendix V-A contain data generated for this
parameter during the four major monitoring periods. The baseline data
gathered during 1971 dry weather monitoring indicated a higher organic con-
tent upstream at Point C with decreasing values downstream. This reduction in
carbon content in the downstream portion of the test reach was maintained
during the 1971, 1973 and 1974 wet weather events. The depression of the
parameter as one proceeds downstream is probably caused by mixing of lake
and river waters.
Comparisons of data over sampling periods and over joints were done using
one of two statistical procedures. The statistical techniques applied
were either an analysis of variance test or the Student t Test.
Analysis of dry weather data obtained during 1971 indicates that there
was a significant difference between all three points. The mean of all
dry weather measurements taken in 1971 for each point is as follows:
Point A, II mg/1
Point B, 17 mg/1
Point C, 22 mg/1
Means were determined for the first twenty-four hour sampling period for
all storms monitored during 1971 (Table 59); comparison of the 1971 wet and
TABLE 59. JOC YEARLY MEANS
(mg/O
Hours after overflow began
Year
0-6
7-12
13-18 19-24
0-24 25-48 49-72
1971 Dry
1971 Wet
1973
1974
10
12
14
9
12
14
8
12
15
8
12
15
11
9
12
15
10
11
14
9
13
12
1971 Dry
1971 Wet
1973
1974
15
17
18
15
17
18
15
16
18
14
16
20
17
14
16
19
14
14
18
16
14
17
1971 Dry
1971 Wet
1973
1974
15
15
17
28
14
18
27
41
16
18
15
16
22
20
15
18
19
15
19
19
18
18
183
-------
dry weather events shows that there Is about a 2 mg/1 reduction in the
amount of total organic carbon for 1971 dry weather to 1971 wet weather
events.
This reduction was analyzed at all sites using the Student t Test and was
found not to be significant. The 1971 wet weather data were analyzed over
all points and indicated that a significant difference existed between them.
When pointwas compared against point, there was a significant difference
indicated between Point A and Point B, but not between Point B and Point C.
Analysis of 1973 data indicated that there was a significant difference
between Point A and Point B but no significant difference between Point B
and Point C. Storm data for this parameter during 1973 had higher values
than the 1971 storm data. Comparison of each site for each of the two
different storm periods using the Student t Test showed that there were no
significant differences between these two periods at Point C or Point B
but that there was at Point A.
When 197^ data was compared for all three points, no significant differences
were observed for day-1 or day-2 data. Differences which were significant
were found for day-3 data. Comparison of point to point for 197^ showed no
significant differences for day-1, -2, or -3. The data show an increase in
this parameter at all three points from 1973 to 197^ although the increase
observed Is not significant at any of the monitoring points.
Biochemical Oxygen Demand
The biological community which inhabits a stream is dependent on the
amount of dissolved oxygen available in the water to carry on respiratory
functions and these populations place a demand on the oxygen available.
Through the normal aeration processes (photosynthesis, turbulence, and
diffusion) of streams this need is met and the amount of oxygen present
is generally greater than the demand. V/hen organic material is contributed
to a water body, especially in sewage, it acts as a food source for the
decomposer portion of the ecosystem. This extra food promotes growth of
this population (the decomposers) which in turn will demand more oxygen
for respiration. This oxygen demand is normally expressed in terms of
biochemical oxygen demand (BOD).
As such, BOD is not a measurement of a specific substance but is rather a
measure of the total impact of the various organic compounds which enter a
body of water and exert an oxygen demand. This impact is based on the
amount of oxygen needed for the breakdown of these organic compounds by the
population of decomposer organisms and those compounds which utilize oxygen
directly. It can only be used as an indicator of water quality and to
detect areas of pollution. The nature of the BOD test lends itself to
detecting areas in a stream where a potential decrease in the amount of
dissolved oxygen may occur.
Tables A37, A38, and A40 in Appendix V-A contain the data gathered
-------
during the four major monitoring periods for this parameter.
Data for the 1971 dry weather period show that the water quality at Point A,
as determined by this parameter, is better than at the other two monitoring
points. The mean for this period at Point A was 2 mg/1 with a maximum value
of 10 mg/1 and a minimum value of 1 mg/1. The maximum value, 10 mg/1, only
occurred once and due to the type of environment being examined at this
point, there is some doubt to the validity of the measurement; the next high-
est value at this point was 3 mg/1. Measurement of this parameter at Point
B had a mean value of 5 mg/1 and maximum and minimum values of 8 mg/1 and
2 mg/1 respectively. Point C had the highest BOD mean value of the three
monitoring sites, 7 mg/1. The maximum value obtained at this point was
20 mg/1 and the minimum value was 1 mg/1.
BOD values, as judged by the mean at each point for the 1971 storm events
(see Table B29), indicate that during a majority of the storm events there
was no change in this parameter above background at Point A and Point B.
The measurement of this parameter at Point C shows an increase above the
background of 1 mg/1 on day-1 of the storm events, as is evident from a
comparison of the means. The mean, maximum and minimum at each point for
this period are listed below for day-1 of the storm events:
BOD values (mg/1)
Mean
2
5
8
Maximum
4
9
20
Mini mum
<1
2
1
Point A
Point B
Point C
The BOD mean values for the 1973 storm season indicate a marked improvement
in water quality at Point C (Table 60). The mean value for day-1 of the
storm events was 3 mg/1. The range of values at this point was from 1 mg/1
to 6 mg/1. Values at Point B also improved in 1973 when compared to the
1971 values. The mean, minimum and maximum values at this point are as
follows: 5 mg/1, <1 mg/1, and 13 mg/1. Values at Point A show little, if
any, change. The mean value for day-1 at this point was 3 mg/1 with a
minimum of <1 mg/1 and a maximum value of 8 mg/1.
BOD measurements during 1974 wet weather events showed a slight increase
when compared with the 1973 measurements. The variations in this para-
meter at Points A and B during 1974 were similar (see Table 60). The
variation at Point C was similar to that of the other two points with the
exception of Run No. 43 (see Table A40). The mean values for the three
days of monitoring along with the minimum and maximum values for each
point during 1974 are listed below.
Table 60 presents the mean values for all wet weather events monitored
during 1971> 1973, and 1974 for the four 6-hr monitoring periods and the
mean values for day-1, -2, and -3 of the same. From the data, it can be
seen that the Root River is a moderately polluted river (26).
185
-------
TABLE 60. BOD YEARLY MEANS, mg/1
Hours
1971
1973
1974
0-6
3
2
4
7-12
2
3
4
13-18
2
3
3
Point A
19-24
2
3
4
0-24
2
3
3
25-48
3
3
3
49-72
3
3
3
Point B
1971
1973
1974
6
it
5
5
5
4
5
4
4
4
5
4
5
5
5
5
3
5
4
3
4
Point C
1971
1973
1974
5
3
5
7
3
5
4
3
5
8
3
4
8
3
5
6
3
5
6
3
5
Feca] Col I form Bacter ia
In this phase of the Investigation fecal coliform bacteria were utilized
as an indicator organism. The indicator organism concept states that,
under a given set of physical and/or chemical environmental conditions, cer-
tain life forms will exist and therefore are an indication of the conditions
that exist in the water body at that time. The presence of fecal colfform
bacteria in the fresh water environment is an indication that an addition
of fecal material to the waterbody has recently occurred.
The organisms are introduced into a water body from a variety of different
sources. These include such sources as discharges from domestic sewage
treatment plants, storm runoff from natural areas, and runoff from feed
lots.
The Wisconsin Department of Natural Resources (27) states that water related
recreation which is classified as secondary contact recreation should not
be carried out on waters which have a fecal coliform count exceeding 2000
organisms per TOO ml as an average. That report classifies secondary contact
recreation as that type of recreation which does not involve significant
risk of ingestion. It states further that on waters "specifically"
designed for recreation, the geometric mean of fecal coliform bacteria
should not exceed 1,000 organisms per 100 ml as an average. For recreation
areas where primary contact is likely, such as swimming, the geometric
mean of the samples should average no higher than 200/100 ml with less
than 10% of the samples having a count exceeding 400/100 ml.
186
-------
Tables A4l - A44 in Appendix A list the fecal collform data gathered during
the 1971 dry weather monitoring period, the 1971 wet weather monitoring
period, the 1973 wet weather monitoring period, and the 1974 wet weather
monitoring period, respectively.
When the data from the 197' dry weather period is examined (Table A4l)
one finds that only the Point C fecal count is below the suggested 200 orga-
nisms per 100 ml and that the other two points have populations which were
considerably above this. This observation leads to the conclusion that
during dry weather periods the contribution of fecal coliform bacteria (or
nutrients for their growth) from various sources was significant in the
City of Racine. Comparing Point A to Point C indicated that a 400% increase
in the number of these organisms occurred as the river passed through Racine
during the dry weather monitoring period.
Storm data from 1971 indicates that during these events the counts of fecal
coliform bacteria at all areas increased over the dry weather data. The
qeneral pattern observed was a rise in the fecal coliform count on day-I
of a storm which steadily decreased over day-2 and -3 of a monitored event.
The geometric mean of this group at Point C over the entire monitoring
period was approximately 200 organisms per 100 ml. . Point A had the highest
total number of these organisms on day-1 of a storm event of all the points,
with a geometric mean of 5986 organisms per 100 ml. At this point the
decline, as measured by the geometric mean for the year, from day-1 to day-2
and from day-1 to day~3 was 64% and 11%, respectively. Point B had a geo-
metric mean for day-1 of all storm events of 1775 organisms per 100 ml, 30%
lower than the value for Point A. This differencewould indicate that the
pollution load in the river is less at Point B than at Point A. The decrease
in this parameter over the second and third day of the sampling when compared
with day-1 of an event is 30% and 57%, respectively, at Point B. One
reason that the pollution load falls off so rapidly at Point A may be due to
the effect of encroachment in the area of Lake Michigan. This effect would
be to dilute the waters in the area and create the illusion that a decrease
in this measured parameter happened faster here than at Point B.
Table A43 lists the data generated for this parameter during the 1973
storm season. Data gathered at Point A during this monitoring period indi-
cated conditions which existed after the treatment sites went into operation.
The geometric mean at Point A for day-1 of all discharge events was
3084 organisms per 100 ml; <497 less than the number of organisms recorded
for the storm period, day-1 of 1971. This reduction in total number may
be due in part to the operation of the two treatment sites in the area.
The reduction in numbers of this organism noted at Point A was not followed
by a similar decrease in this parameter at the other sites.
Fecal coliform data at Point B showed almost a 20% increase from 1971
and 1973 and a 225% increase at Point C during this time. Another factor
which may have a strong influence on the reduction of fecal coliform
bacteria at Point A was the higher lake level in 1973 than in 1971. The
187
-------
higher lake level could have influenced this parameter by causing more dilu-
tion.
Fecal coliform data gathered during 197^ showed a reduction in the bacterial
populations, when compared with 1973 data at Point A. Data gathered at the
other two monitoring points in I972* were similar to the data obtained during
the 1973 sampling effort.
The difference noted at Point A from 1973 to 197^ was not statistically sig-
nificant due to the amount of variation in the data both years, but the
decrease in this parameter at this point was about 60% on day-1 of wet weather
events, 50% on day-2 and 20% on day 3. This decrease was not noted on
day-I of wet weather events at Point B, although there was a decrease of 27%
and 29% during day -2 and~3, respectively, at this point. The decrease at
Point A may be due in part to the continued rise in the level of Lake Michi-
gan, which occurred over the entire duration of the project, and in part,
to the effect of the treatment systems on this parameter. In addition,
there appears to be a substantial decrease in the fecal coliform concentra-
tion during treatment events In which chlorination was performed at Site II
as compared to events during which chlorination was not performed.
Riyer SedJmenit Chem?ca1 Ana1yses
River sediment for chemical analysis were collected once during 1971, four
times during 1973, and twice during I97^« Samples for these analyses were
collected from cross-sections of the river at all three of the major moni-
toring points. Samples at each site were collected at quarter points across
the river; these samples were then composited into a single sample for each
point.
Samples of bottom muds were taken using a Wildco-Ekman grab sampler with
a 15.2 cm (6 in.) square sampler jaw opening. This sampler is designed for
use In soft, finely divided littoral bottoms of lakes, ponds, and streams
which have little vegetation and intermixtures of sand, gravel and other
coarse debris. The sampler is primarily used for quantitative and qualita-
tive sampling of microscopic bottom fauna as well as obtaining, as in this
case, samples of bottom materials for chemical analysis.
The Ekman grab sampler is operated by opening the jaws and settling the
activation springs over their respective retaining bars. The sampler is
then lowered to the bottom of the area to be sampled. To prevent loss of
materials which may have been disturbed by the impact of the sampler with
the bottom, it is raised, via the attached rope, about 0.3 m (I ft) off
the bottom and moved upstream of the initial impact point where it is
allowed to settle freely into the bottom material. A weighted messenger
is then send down the attached line, which upon impact, activates the
springs on the sampler causing it to close, thus taking a sample. To
prevent loss of material upon retrieval from washout, the device is
equipped with lids which are kept closed by water pressure. Sample size is
variable, dependent on bottom composition, but generally"is in the
range of one to two liters per grab.
138
-------
Samples used for chemical analysis were collected directly from the dredge
and put into sample bottles marked for chemical analysis.
After all samples for a given date had been gathered, they were taken back
to the laboratory for analysis of the following parameters:
Kjeldahl nitrogen
Ammonia nitrogen
Total phosphorus
Total solids
Volatile solids
Analyses were performed according to the methods given in Appendix IV-D.
Table 6l lists the results of these analysis for the respective sampling
dates.
Statistical analysis of the data was performed for each of the parameters
using an analysis of variance test to compare points over time. Results of
these tests reveal there was no significant change over time at any of the
points for ammonia nitrogen and Kjeldahl nitrogen. Analysis of total solids,
volatile solids and total phosphorus showed no significant difference
over time, but there was a significant difference indicated"between Point C
and the other two sampling sites. The data generated from these samples
indicate that, because of the high nutrient content, the potential exists
for nutrient loading of the river from the sediments. With the exception
of total solids which increased from Point A toward Point C, all parameters
measured increased from Point C to Point A.
In addition to running chemical analyses on the bottom muds from the river,
analysis of different points in the water column was undertaken to deter-
mine what, if any, nutrient profile existed. These analyses were taken
using the Bacon water sampler. The Bacon water sampler is a mechanically
operated piston cylinder device which selectively takes approximately
500 ml of water at any chosen level. The sampler is lowered to a selected
level and a trip rope, controlling piston activation, is pulled moving the
piston plug upward and opening the sample chamber. The liquid rushes in
discharging air through the top of the cylinder by route of the piston
shaft discharge point. Once the discharged air bubbles reach the surface,
the person sampling releases the piston control rope allowing the piston
plug to settle back into place closing the sample chamber. The sample is
then drawn to the surface and its content discharged into a marked
collection bottle for identification and analysis. No preservatives or
chemicals were added to the water samples collected in this manner. One
composite sample was drawn at each dredge sample location.
Samples were collected at six different sites in the test reach of the river.
These were discrete samples taken from top to bottom In 0.9 «n (0.3ft) inter-
vals. Table 62 contains the data generated from these samples. An increase
is noted In each parameter proceeding upstream from Point A toward Point C.
189
-------
•K
CO
LU
OO
>-
ll
cn
_
L. -3-
r^>
o cn
o •—
LA
4-1
on PO
< cn
CM
CM
4)
c
3 PO
o\
r-.—
CM
tBM
L.
a.—
cn
0 —
CM
4)
C
3 t—
—> r~-
cn
LA —
^f
^j
c
"o
a.
co
CO
LA
CM
LA
CO
Co
CO
,^_
•3-
cn
PO
CM
0^
o
oo
sO
CM
o
0
vO
c
(U
m
o
4-)
• »
c
r—
f~
CD
•a
"4)
"7
s£
^
LA
sO
SO
r-
cn
cn
LA
LA
oo
LA
LA
CM
*••»
C
4)
a>
0
L.
4-1
•—
C
OS
c.
o
f~
CM
p^.
CO
CM
cn
CO
oo
CM
-3"
SO
-3-
LA
OO
*
-3"
CO
o
LA
^^\
O^
CO
in
•a
•—
^—
0
I—
m
-[„>
Q
1 —
CO
CO
CO
*
PO
f>
PO
LA
PO
-3-
PO
_^«
SO
O
LA
SO
O
CO
^:
co
CM
CM
PO
.^
^3-
en
PO
CO
LA
cn
o
o
so
•w
c
4)
cn
O
•M
*.»
C
»—
OQ J=,
4-J "O
c •—
•^ (U
O "~~>
Q. ili
^
CM
-3-
co
CM
co
CM
oo
CM
CM
r-.
r^!
c:
Ol
0
U
4J
•»
C
(U
*»
c
o
E
CM
CM
SO
OO
O
SO
CM
CO
cn
CM
^ ,
cvi
O
o
LA
o
LA
1
1
C
a>
Ol
0
4-J
•_
C
••^
CJ» r*
(U
c •—
•^ d)
o — >
Cu ^
^^
JT
PO
CM
LA
CO
LA
•""
PO
O
1
1
c
4)
Ol
0
4-*
•— «
C
fO
•— .
c
o
^
E
4-1
C
4)
Ul U
E «-
re 4)
i- Q.
Ol
•— in
— <0
•— Ul
E -0
•^
•- "o
in
"2 .»
4) 4)
Ul *"•
(A .—
U (0
a.—
4> >
4) XI
U C
(0 fO
in in
4) "O
3 —
"ra "o
> M
u) rr
3 *"
« <^J
o o
f- 4-»
a. ^
o "2
r- 3
0 E
1 «
m 3
4) O
cn
O E
LB (0
4-1 L-
z?
PM>
*• *
4)
4-> U
i 2.
^
"?>
•
4)
»^
a.
c
flj
W
-^^
3
E
+J
1
m
M—
o
4J
^
Ol
••"
1
>.
J3
190
-------
Cu •—
UJ
I- CO
< UJ
te t-
Q- —•
to to
§*
Q- UJ
O r-
of x
°"
z .
o Q
£ ec
4-1 t- o> en
> c O > -a
O ._!. f- *—• •—
\s •"• ^ •" w~~
< Q. S es ca
CO
4-1 O
a) o>
3 O14-)
g--o c
ffl *«- O
S CO Q_
CQ CO
4-> O cn
C 4-» T3
'O 4J *C
o_ co ca
co o)
en 4-1
c -a c
flj i- O
Z CQ Q-
fA ro
LA LA
O O
LA
\O O
LA
O
R
CM IA LA
OOO
LA LAOO
OOO*
LAO O
— CM CM
O O O
LA r«l O
O P-.VO
000
00000
O LA LA LA LA LA
O O O O O O
O O O O O O
vO ^*O vO CO Lf\ OO
— — O O — LA
. . . O • • •
OOO OOO
£
cn
LAOO
o o
LT\
o vo LA
O O — CM f^-iT LA
191
-------
Benthfc MacroInvertebrates
The benthic population found in a stream is made up of a diverse collection
of fauna and flora. It includes such things as protozoans, algae, rotifers
and macroinvertebrates. By definition, the benthic population is that
community of organisms which live on or in the bed of aquatic or marine sys-
tems. They can also be found attached to aquatic plants and other substrata
found on or near the stream bed. The portion of this community examined
during this study were macroinvertebrates.
The species groups, as well as their numbers, which make up this benthic
community will vary with the physical and chemical conditions of the stream.
Patrick (20) points out that in oligotrophic streams where the nutrient level
is very low, there exists a population which has extremely high diversity
but extremely low population levels. The physical and chemical factors of
such a stream are those of very low nutrient levels. The reverse is true for
streams which have a high nutrient level.
Sources of energy and nutrients for the benthic communities of streams are
quite different than for either the lake or terrestial communities found in
the same types of niche.. The total nutrient load of a stream is continuously
being replaced from the upstream direction and may be of three separate
types: dissolved, suspended,1 and organfsmal. In many lake systems it has
been determined that nutrients are generally in a recycling type of system,
while in a stream, because of its flowing nature, nutrients are being renewed
ponttnually. If this were not the case, the benthic community of a
would be divided into different levels of organism or energy transfer levels*
the decomposers, primary producers, .herbivores, and one or two
levels of carnivores.
The structure of the macroinvertebrate community is commonly used as an in-
dication of conditions in both polluted and unpolluted streams. The prime
reason that this population has been used almost exclusively as an indicator
rests in the fact that the life cycles of many of the benthic species are
long and that this group is primarily confined to the bottom area. If a
perturbation to that system occurs which lasts for only a short time and
it is toxic to the indicator organisms, the investigator who works with the
system can discern that something has happened even if he cannot detect it
through physical or chemical means. Because of this, much research has been
carried out to try and classify benthic organisms according to their pollu-
tion tolerance. A system of this type allows the use of these organisms
as criteria of pollution (29). PresentVy, a number of investigators have
used indices developed from information theory that expresses species
diversity to summarize the community structure of benthic organisms. The
use of the diversity index ?s considered as one of the best and most
sensitive-indicators of ecological change (30), (21). The first use of the
species diversity methods occurred in 1966 when investigations to examine
the effects of organic enrichment on streams was carried out by Wllhm and
Dorris (32).
Samples for biological classification and identification were collected
192
-------
from the river cross-sections at each of the three major sampling points.
Characterization of each point was done by taking quarter point samples along
transects at each point. Samples from the bottom were taken using a Wlldco-
Ekman dredge (previously described).
The method of bottom fauna collection involves collection of a bottom sample
which is then drawn to the surface. Before the dredge Is lifted above the
water line, it is placed into a five gallon plastic wash bucket which has
had the bottom replaced by a 30 mesh screen reinforced by 1.27 cm (1/2 in.)
hardware cloth. The dredge and bucket are then raised slightly above the
water line and the contents of the dredge are discharged into the bucket.
The sample in the collection bucket is then swirled lightly in the water
washing away silt and mud and leaving the benthic organisms and larger
detritus particles. This material is then washed into a collection bottle
containing a 70% alcohol solution and taken back to the laboratory for iden-
tification.
Data comparisons were made using a diversity index, developed by Cairns (33),
over sites for a given sampling date and over dates for a given sampling
point. Data for these studies is presented in Table 63.
The diversity indices generated using data obtained from samples taken at
Point A Indicate an extremely degraded and unstable system. The biota found
at this point were represented by very few toxa; during the June 1974
sampling effort at this po'int, only one type of organism was noted, this
being the species Tubifex, an indicator of an organically enriched system.
As measured using the Cairns Index, Points B and C both were in a healthier
state than Point A. Point C remained fairly stable through the monitoring
program with only one exception, that being the April 1974 sample which
was probably influenced by the scouring actions of the Spring high flow
period. Point B remained at a somewhat degraded level throughout the moni-
toring period.
Light - Dark Bottle Test
The 1Ight-and-dark bottle technique may be used as a simple means of
measuring the total diurnal metabolism of a body of water. It Is also a
starting point to be used for charting energy flow through an aquatic
system.
Lightrand-dark bottle tests were conducted at all three river sites in tri-
plicate on August 13, 1973 and August 30, 1974. The purpose of these tests
was to indicate the potential daytime peaks and nighttime troughs In the
river's dissolved oxygen concentration attributable directly to photo-
synthetic activity. These tests, by design, exclude anchored vegetation,
and concentrate on organisms and/or particles contained In the water column.
Baseline dissolved oxygen determined Immediately upon collection at 9:00 am
on August 13, 1973 was found to be 5.0, 5.4, and 7.3 mg/1 at the Point C,
193
-------
TABLE 63. BENTHIC ORGANISM SURVEY
Date of
Survey
April, 1973
June, 1973
August, 1973
October, 1973
April, 1974
June, 1974
Number of organisms
Number of species
groups
Predominant species
group
Cairns Index Number
Number of organisms
Number of species
groups
Predominant species
group
Cairns Index Number
Number of organisms
Number of species
groups
Predominant species
group
Cairns Index Number
Number of organisms
Number of species
groups
Predominant species
group
Cairns Index Number
Number of organisms
Number of species
groups
Predominant species
group
Cairns Index Number
Number of organisms
Number of species
groups
Predominant species
group
Cairns Index Number
Point A
253
3
Tub if ex
0.020
250
1
Tub if ex
0.000
150
1
Tub i f ex
0.000
300
7
Tub 1 fex
0.07
250
I*
Tub if ex
0.006
275
1
Tub J fex
0.000
Point B
135
2
Tub 1 fex
0.05
55
5
Tub i fex
0.20
150
6
Tub i fex
0.10
300
16
Tub 1 fex
0.40
275
8
Tub ! fex
0.26
275
13
Tub if ex
0.33
Point C
9
5
None
0.89
44
8
Ephemeroptera
0.63
36
7
Ephemeroptera
0.81
300
14
Tub if ex
* *
0.44
300
Tub if ex
0.004
269
Ephemeroptera
0.66
194
-------
B, A, respectively. The Winkler method of oxygen determination was used.
The remaining samples were placed in the direct sunlight at a central loca-
tion and left until 2:00 pm. All were then analyzed for dissolved oxygen
concentration. The data is contained In Table 6k.
TABLE 6k. LIGHT-AND-DARK BOTTLE DISSOLVED OXYGEN DETERMINATIONS (1973)
(mg/1).
Point C
Initial DO Concentration DO Light Bottles DO Dark Bottles
Test 1
Test 2
Test 3
Mean
5.0
5.1
5.0
5.0
9.7
9.9
9.6
9.7
4,3
4. 2
4,3
4.3
Point B
Initial DO Concentration DO Light Bottles DO Dark Bottles
Test 1
Test 2
Test 3
Mean
5-3
5.4
5.6
8,5
8.5
8.5
8.5
4,7
4,8
4,7
4,7
Point A
Initial DO Concentration DO Light Bottles DO Dark Bottles
Test 1
Test 2
Test 3
Mean
7.3
7.4
7.5
7.4
7.1
7.0
7.1
7.1
7.1
7.0
7.0
7-P
Algal activity was indicated at Point Cand to a slightly lesser extent at
Point B. The relative increases and decreases were similar with a slight
overall decrease in magnitude apparent at Point B, probably due to a re-
duction in concentration of photosynthetic material. There was apparently
no algal activity at Point A. This is consistent with 1971 data where
195
-------
extremely limited numbers of algae were reported.
Is Indicated regardless of time of day.
A slight oxygen demand
Data obtained on August 30, 1974 indicated a baseline dissolved oxygen con-
centration of 4.5, 2.1, and 6.1 mg/1 for Points A, B, and C, respectively.
Baseline data was obtained at 5 am on the above date. As in. 1973,
the Wlnkler method of oxygen determination was used. At the time that
baseline conditions were determined, triplicate light and dark bottles were
set up at each point and incubated at a depth of 15 cm (6 In.) for six hours.
After the Incubation period, oxygen determinations were run on all samples,
data from these determinations is presented In Table 65.
Table 65. LI GHT-AND- DARK- BOTTLE DISSOLVED OXYGEN DETERMINATIONS (1974)
(mg/1)
Initial DO Concentration DO Light Bottles DO Dark Bottles
Test I
Test 2
Test 3
Mean
6.1
6.0
6.1
6.1
5.3
5.5
5.8
5.5
5.9
5.4
6.2
5.8
Point B
Initial DO Concentration DO Light Bottles DO Dark Bottles
Test 1
Test 2
Test 3
Mean
2.3
2.0
2.0
2.1
2.7
2.5
2.8
2.7
1.9
1.7
2.1
1.9
Point A
Initial DO Concentration DO Light Bottles DO Dark Bottles
Test 1
Test 2
Test 3
Mean
4.6
4.3
4.6
4.5
4.8
4.7
4.9
4.8
4.2
4.5
4.3
4.3
196
-------
Algal activity was Indicated at Points A and B while there was a net loss of
oxygen at Point C which would Indicate a slight oxygen demand at that point.
Biological CharacterizatIon
During the early summer of 1971, a special study was undertaken at Point C.to
characterize it biologically. Table 66 Is a summary of the results.
TABLE 66. SPECIES CLASSIFICATION AT HORLICK SITE (1971)
[150m (500 ft) DOWNSTREAM OF DAM]
Algae
Sp? rogyra
Mpugeotla
Zygema
Chladophora
Microspora
Scenedemus
Anklstrodesmus
Volvox
Protozoans
IchthyophthIrIrI us
_Sp_|rostomum
Massula
An Isonema
£ron ton I a
Englena
Parameclum
Crustaceans
Cambarus sp. (common Crayfish)
Diaptomus
Mollusco
Strophitls sp.
Sphaerlum sp.
Pleurocera sp.
Miscellaneous Invertebrates
Tubifex
Fishes
Lepomis marchlrus
Notropls sp.
Ictaluru's sp.
Exos luelus
Salmo gairdnerl
Cyprlnus carplo
Castostomus conmerson
Moxos toma rtacro'lep Iodotum
(common bluegi tl)
(common shiner)
(bullhead)
(Great Norther Pike) Rare
(Rainbow Trout) Spring only
(Carp)
(connom White Sucker)
(common Norther redhorse)
The samples collected below the Pointcindicate a bottom configuration
characterized by rock and shale rubble Interspersed with small patches of
197
-------
detrftal materials. It was found to be Ideally suited for harboring diverse
speclatfon and proved to be biologically the richest area sampled within the
test reach. Vast mats of C 1 adophora and Spl rpgyra interspersed with Mougep*'.
tla provided a good deal of natural cover and food for the herb I vqrou's fauns
population. All of the organisms identified in Table 66 were found at
Point C. Several variables were observed, however, which limited the diver-
sity of this point's ecosystem and determined the species dominance at the
point. The water samples taken at Point C were nutritionally rich, espec-
ially in nitrogen, phosphorus, and carbon, which probably contributed signif-
icantly to the high chlorophyll concentrations observed throughout the
period. The combination of nutritionally rich water and the shallow depth
of the river at this point, generally less than 0.3 m (1 ft), resulted in
ideal conditions for the culturing and growth of algae and bacteria.
Special Analysis
Several special tests were conducted during the monitoring period to explore
in more detail areas of concern uncovered by the established battery of
tests. Among these were BOD determinations, sedimentation tests, sieve
analysis In relation to sedimentation, and pesticide analysis.
Biochemical oxygen demand appears to remain unchanged at all three sites
despite the input of storm-generated discharge Into the test reach. This
observation was a cause for concern from its discovery of the phenome-
non early in the monitoring period. A BOD series of 3, 5, 10, 15 and 20
days was run on the sample taken for Run No. 6 (8/10/74) at Site A (Main
Street Bridge). This test revealed a verynormaj BOD curve and indicated
there was no delay in BOD expression in the river. The possibility was
explored that the discharge plume was bypassing the sampling point. A
series of BOD determinations were made from the river cross-section at
point Aand again proved Inconclusive (Table 67), There Is a possibility
that the continuous demand In the river during normal flow conditions Is
sufficient to mask any additional BOD Input as a result of storm-generated
discharge. Determinations to uncover the validity of this possibility have
not yet been made.
Suspended solids, being one of the parameters which exhibited some change
as a result of a rainfall occurrence, was explored further to ascertain
the possibility of sedimentation within the river. Both theoretical cal-
culations and field measurements were done,
Using the formula, V = (1,486) (R) (K[SS-K] d/N) taken from the Storm
Water Management Model (35), a critical velocity to retain 50?! of the
estimated combined sewer input suspended solids (see particle size distri-
bution following) in suspension was calculated. Using values of 0.035 for
Manning's coefficient; an R taken from the river cross-section at Site B;
and the (K) and (Ss) values from the math model; and inserting the (d)
value equal to or less than 50% of the analyzed particle size, a calculated
critical velocity of 0.168 m/sec (0.553 ft/sec) was obtained. The eight
year average river flow at the Horlick Dam is 221 cu m/min (130 cfs),
Based on this flow and a cross-sectional area of 151.0 sq m (1,625 sq ft)
-------
TABLE 67. BOD VALUES FROM RIVER CROSS-SECTION (1971)
(nig/1)
During the period of overflow occuring as a result of Storm 12, 11/18/71,
cross section surface BOD's were drawn at the Main Street Bridge. Starting
on the North Bank and proceeding south, samples were drawn from the surface
at 3-05 M (10 ft) intervals for a total of 12 samples. AH samples were
analyzed for TOC and TIC while only the odd numbered samples were analyzed
for BOD. The resultant values are shown in the table below.
Distance from North Bank
Meters
0
3.05
6.10
9.15
12.20
15.25
18.30
21.35
24.40
27.45
30.50
33.55
Feet
0
10
20
30
40
50
60
70
80
90
100
110
TC
37
33
32
31
38
31
31
32
33
33
32
33
TIC
24
22
22
22
24
22
22
22
23
22
22
21
TOC
13
11
10
9
14
9
9
10
10
11
10
12
BOD
4
2
3
3
4
7
at Point B, a river velocity of 0.024 m/sec (0.08 ft/sec) can be calculated
for Point B. When compared to the critical velocity, sedimentation is
indicated.
To verify downstream sedimentation, sedimentation-collection buckets were
placed at both the Main Street and Western Publishing Sites. The results
of the collections and later analysis for 1971 Storm Nos. 8, 9, and I) are
contained in Table 68. There is indication that sedimentation occurs and
that it Is affected by both wind direction and velocity. An upstream wind
(Storms 9 and II) appears to cause a higher sedimentation rate and the
majority of the increase appears to be ?n .nonvolatile solids.
During 1974, sedimentation characteristics of the river were determined after
a prolonged dry period had occurred, but prior to any discharge. Sedimen-
tation collection equipment was installed at each of the three monitoring
sites and left suspended 0.5 m (18 inches) off the river bottom for a total
of 14 days. This is in accordance with the method described by Edmondspn
(36). The rate of sedimentation at each point during dry weather flow was
estimated to be 7.0 x 10"^ g/mz/day at PointA, 7-3 x |0"4 g/m2/day at Point B
and 3.8 x IO"2* g/m2/day at Point C. These data appears to confirm the facts
established by the chlorophyll analysis and the sedimentation test run
199
-------
TABLE 68. SOLIDS COLLECTED IN RIVER SEDIMENTATION
COLLECTION BUCKETS
(g/day/sq m)
Total Solfds Volatile Solids Wind DIr-
Storm Date Point A PetntB- Ppint..A _ _ Pp i_niLB ectlon Avg.
Average Wind
Velocity, km/hr
• ••«•• I m,^
008
009
Oil
9/20/71
9/27/71
11/1/71
7.4. 2.
18.7 2.
10.28 13.
9
8
4
62.
7.
15.
0
6
5
2.
2.
6.
9
8
1
35°
89°
60°
40.
19.
48.
2
3
3
during 1971, that there is a loss of energy near pQint B and the subsequent
settling out of solids at this point.
Ch1orophy11 Ana1ys i s
During August 1971, two chlorophyll profiles were determined
on the lower Root River at five selected test sites. One uniform composite
of the entire water col-umn was collected. The resultant chlorophyll concen-
trations were determined using an analytical method developed by Yentsch
and Menzel in 1963 (37). This method consists of filtering a measured
volume of water into a glass fiber filter which is then extracted with an
85% acetone solution to free the chloroplast pigments. The extract after
centrifugation is then placed in a fluorometer and fluorescence is measured
using an excitation wavelength of 430 to 450 millimicrons. The readings of
fluorescence values were related to a standard curve for conversion to parts
per billion of chlorophyll and these values were used to calculate total
kg of chlorophyll per one meter cross-section of each selected site. The
results in total kg of chlorophyll per respective cross-section for two
surveys in August 1971 are shown in Figure 59.
One observation made in 197' from the August profile survey of chlorophyll,
and supported by the additional readings available for the biological sur-
vey, was the tremendous decrease in total mass of chlorophyll between the
location of Point B and the Memorial Street Bridge, which is located upstream
from Point A
Pesticide Analysis
Samples for pesticide analysis were collected from the treatment sites and
the River monitoring sites during 1971, 1973, and 1974. The months and
sites sampled are listed below. All samples were taken at the beginning of
a storm which was preceded by a long dry spell.
200
-------
S 2-
« a
3 i
133^15 NIVW
UJ
C9
UJ
I I I I I I I I
WVQ X3I1VIOH
I I I I I I I I I I I
<
UJ
eo
UJ
oo
O
O.
I
u
UJ
3
o
I
D)
o
-------
September, 1971
Site A
Site B
Site C
S i te II raw
Site II effluent
S i te MA raw
Site MA effluent
Site I raw
x
x
x
September, 1973
x
x
X
X
X
X
July, 1974
x
X
X
X
X
X
Analysis of these samples were performed by the E.P.A. Region V Laboratory,
on the 1971 and '974 samples. 1973 samples were analyzed by Limnetics, Inc.,
of Milwaukee, Wisconsin. Results of these analyses are presented in Tables
69, 70, and 71 for 1971, ,1973, and 1974 respectively.
TABLE 69. 1973 PESTICIDE RESULTS
Date: September 26, 1971
(ng/1 or ppt)
Horlick
Pesticide Dam (River)
Llndane3
Heptachlor
Aldrln
Heptach 1 o r Epoxi de
Methozychlor
Dleldrin
Endrln
o,p DDE
p,p' DDE - o,p-DDD
p,p'DDT
o,p-DDT
5
13
16
12
46
<1
20
15
11
44
28
Western
Publ. Co. (River)
7
10
7
15
34
5
19
29
14
39
21
Site II
Raw
<1
<1
14
16
58
<1
<1
30
20
66
34
a. Analyses performed through Region V, EPA, using standard EPA procedure
202
-------
_J '
LU
0.
CO .--.
or *_
^^ i^
U 4>
i5r iT
^5 O
..
ft*" h
LU re
£~
<• g
1 C
«•* *!/> .-*»
3C W Q.
a: >. 0.
o m
H tJ L-
co O
LA
ce -3" •—
O *""•'' **••*
u. en
«^ c
to r--»
H en c
rs • ^
to *
LU -3-
LU OJ
P?> Tj^
O S
H Q.
(O ft)
LU tO
a.
^^
• 9)
0 4J
r*^» re
Q
LU
: 1
OQ
^
*±
— 0)
3
•W H—
00 LU
Lu
m „,
~ 3
0) re
•4-> (**
to
c
^C Q)
— a
14-
«)«*-
4J Lu
CO .
LU
1^.
^»
v /5
to
o "u
,- E «
i°s
4-1 U
et a) «)
— 4) >
c toH
re
^H
>t^
U
.J
*I
4)
Q.
CD O LA O O
^- CM V V V
O O LA O O
CD ro •"* *••• ^*"*
•— •-* V V V
O O LA O CD
•»• CA ^HB ^~ ••••
V CM V V V
O O LA O O
CM V V V
O O LA O O
O LA — — —
O LA LA O O
•— I— «— r— ~—
^
_o u
!c o
o •—
* flJ ^* f^t
C C X U C
.- re o re —
U T3 JC 4J U
•D C *J Q. T3
C •— 4) 4) p—
LU _l S X <
o o LA LA o o
V V V V V
O O LA LA O O
•••* •«• •«• I*H* V~ • p«»
V V V V V V
O O LA LA O O
*•• VM »•» »•» " VO **••
v v v v »— v
O O LA LA O O
_ v v v v v
O O LA LA O LA
O O LA LA O O
•—••*••—•— *— •—
V V V V V V
t- +
O 4) + LU LU
Ct-t-0— T> Q Q Q
u o o o o *x a
tJ 1 1 1 CD O 1 1 1
i— *j ex
q>- - _ a. o -
— Q.Q.Q.4) Q.Q.O.
a o OL a. :c o o o
«^
u
*J
t/1
11)
u
-a
o •
Lu
K-
o
a>
3
•a
0)
o
o
l_
Q.
«^
a_
LU
c
«r
i
i
2
*""1"
••"
z:
.
0
tn
U
4^
C
E
•*•
wJ
^
"O
E
u
O •
^ !****•
Q) rn
O- —
tn «
4) Tl
in tM
*~ 4)
re c
C 3
re
203
-------
- 3:
OJ V
ooooooooo ooooooooo
in.
v v
in.
vvvvvvvvvvvv
o
CM
co
o o o o
O V V V
CM
O O O
V V V
fl) .
4J i-
t/)
oooooooooooooooooo
trv •— -3- ~
o o o o
o -^ ~ —
O V V V
VD
o o o
CM «— V
to
oooooooooooooooooo
o
o o o o
o o o
vOrrvVCNltsJVVVVVVV VVVV V V
IA
CM
O t— •— •—
O V V V
CM
CM
V V
ui r-.
c£ cr>
UJ
CO 3 V.
oooooooo oooooooooo
o o o o
o o o
LTV •— CM Csl V
vvvvvvvvv
o •— IAOO
O V
LTV
• o en
•~ -w c
ooooooooo ooooooooo
o o o o
o o o
Ul
03
v vv>vvvvvvvvvv'vvv
O v
— cr\
r- VO
V V
oooooooooooooooooo
o o o o
o o o
a.
OQ .r»>vvvvvvvvvvvvvvv
o
CM
o v
CO
LAO -"
•— V
O . Q
I- a
UJ
a a
Q Q O
— o i •— a
o o
— Q Q
oouuoooo
•— CO
C O C O X
CO I - TO I - TO I
4J Q. O. 4J O.O.-M Q
O ^»O «»>O ""OiS-l-l-iL.!-!-!-!-
I— O 0.1- O O-l- O O-S<< <<<<<<
c M= -
I Q.
-------
SECTION VI
STORM WATER MANAGEMENT MODEL
Vl-l INTRODUCTION
The Environmental Protection Agency Storm Water Management Model (38) here-
after referred to as SWMM, has been applied to the 335.8 hectare (829.3
acre) drainage area in Racine, Wisconsin which contributes to the two
combined sewer overflow plants described In Section IV. The application,
modification and results of the SWMM for this area will be discussed in the
following pages.
SWMM Description
The SWMM is a packaged computer model available from the EPA which predicts
for a given rainfall event the quantity and quality of storm water runoff
and the resulting combined sewer overflow plus the effects of this overflow
on the receiving body of water. The user of the SWMM supplies the rainfall
intensity, the physical description of the land, the conveyance mechanisms,
any storage-treatment systems within the drainage area and the receiving
body of water. The format for this input data is described In Volume 111 of
the SWMM which is the User's Manual (38). The output of the SWMM is in the
form of hydrographs and pollutographs; that Is, flow versus time and quality
constituents (5-day BOD, total suspended solfds, total conforms) versus
time. The Receiving block also provides velocity, stage, and dissolved
oxygen concentration versus time. This form of the output allows for a time
step analysis of the data as opposed to the overall effects such as total
flow discharged per storm.
The SWMM program consists of over 10,000 Fortran statements which are
divided Into five subprograms or blocks: Executive, Runoff, Transport,
Storage and Receive. The Executive block is used for control and does no
computation as such. The Runoff block computes the quantity and quality of
the Storm water runoff for each subarea. This runoff is then applied to the
various Inlets of the main sewer system. The Transport block routes the
runoff and dry weather flow through the conveyance system and then produces
hydrographs and pollutographs at any selected point within the drainage area.
The Storage block modifies the output of the Transport block according to
the user's selection of various storage and/or treatment facilities provided
in the program. Thus, dissolved air flotation or microstrainers might be
selected as one treatment option. The Receiving block uses the output of
Transport or Storage and computes the effect of the discharge on the receiving
river, lake, or bay. Figure 60 shows the interrelationship and the general
type of input data for each block.
205
-------
UJ
>
5 g
tf §
S£ QD
UJ
4
1
1
I
JL
<
5
0
o
a:
i-
0
r *»
o
V J-
i- S2
OT Si
^^ ^ UJ
^|b
< t <
0 1
^^<
? 5 a:
a: a)
OT
^*N^
•^^ _^~
o>
LJ
0.
?3
Qaf
UJ
DOT
^•n-
~3:
UJ
*rt
^^
UJ O
So-
^
O UJ
a: a
^X-X SJX1 "VX"
o
z N«:
> °
ui S
h^
a:
i
! i
1 1
, 1
! >
* k
11
1-
H
tn O
1-
a:
t-
4
1
I
i
l^jS r^V^i
o
Ll-
CC
111
1-
111
>
o
a.
o
1VTMENT
UJ
!
"*
>
UJ
< 0
o: o
o -J
i— m
«
«9
L*
0)
m
H
O
1
•
o
£
23
U.
206
-------
Data Requirements For Racine
The SWMM requires a large amount of Input data to describe the drainage
area and the receiving body of water. The Runoff block uses the characteris-
tics of the drainage area such as subarea land use, surface slope, and per-
cent imperviousness along with the allocation of rainfall intensities to
each subarea to determine the amount of runoff to the sewerage system. The
Transport block requires the description of the sewerage system and the dry
weather flow for each subarea. Thus, the size, slope, roughness coefficient
and upstream element are required for each sewer and gutter. The dry
weather flow of the area is determined by the population, number of house-
holds and major industrial flows of each subarea. The Storage block re-
quires the description of a storage and/or treatment device selected by the
user to treat an overflow. In Racine, the two screening/dissolved-air
flotation units are used as treatment options with no associated storage
facility. The Receiving block requires a description of the flow, velocity,
depth, stage and loadings for the receiving body of water. The final 10
kilometers (6 miles) of the Root River are used as the receiving body.
Section V of this report describes the Root River monitoring program.
The comparison of the output of the SWMM to actual measured data is an
important part of this report. The two combined sewer treatment plants
provide flow measuring devices and sampling points at each overflow. The
large amount of data that was generated during the k$ monitored overflow
events provide the basis for comparison with the SWMM output. In the
following portions of this section of the report, the data used for each
block will be described, the initial results will be discussed along with
any problems and then the total program output will be evaluated. The
final topic of this section will be the application of the SWMM to the re-
maining combined sewer areas of the city and the results of these dis-
charges on the Root River.
VI-2 RUNOFF AND TRANSPORT BLOCKS
Because of the close relationship between the Runoff and Transport data,
both of these blocks will be described together throughout this section.
Data Acqu i s i .t.I on.
The collection and preparation of the data used as input for these blocks
began shortly after the selection of the outfall locations for the con-
struction of the treatment units. Utilizing maps of the sewer system
supplied by the Racine City Engineer's Office, it was possible to determine
the boundaries of the drainage area which contributed to these overflow
points. Next, the interceptor, trunk (main) and branch sewers were located
in the drainage area and the total area was divided into a number of
smaller areas based on the layout of the sewerage system. These areas
were then examined to determine the direction that the runoff flowed and
the entry point or inlet to the sewer system. The.direction of flow
was determined by using street corner elevations from the sewer maps. If
20?
-------
the runoff was entirely directed to a single inlet, then the subarea remained.
But if the runoff flowed to two different inlet points within the original
subarea, then that subarea was further divided into two separate subareas.
After the runoff patterns were determined, the land use within each common
runoff area was listed. If an area was composed of residential and commer-
cial land uses, then this area was further subdivided along these land
use boundaries. The total drainage area was finally divided into 56 sub-
areas according to the following land use patterns:
Single family residential
Multi-family residential
Commercial
Industrial
Parkland
20 subareas
13 subareas
14 subareas
6 subareas
3 subareas
210.4 hectares
33.1 hectares
40.1 hectares
32.7 hectares
19.0 hectares
(519.9 acres)
( 81.8 acres)
( 99.1 acres)
( 80.8 acres)
( 46.9 acres)
Figure 61 shows the location of these subareas and the numbers used to
identify them throughout the SWMM. The percent imperviousness of each sub-
area was determined by use of an aerial photograph. This photograph was
analyzed for the amount of pavement, roof area or other hard surfaces that
caused runoff to the conveyance system. This area was then expressed as a
percent of the total area. After the percent imperviousness was determined
for each subarea, the sewer maps were again used to determine whether the
surface runoff went directly to an inlet of the main conveyance system or
was transferred by means of a gutter pipe. In most cases a gutter pipe
was the primary means of drainage, with the flow eventually reaching an inlet
to the main conveyance system. It must be noted at this time that subarea
numbers 14, 17, ^7, and 48 are assigned a contributing area of only 0.4
hectares (O.I acre) because these areas do not contribute surface runoff to
the conveyance system but only dry weather flow. Thus, these areas are
assigned minimal runoff contributing area and actual population equivalents
for the dry weather flow contribution.
The elements selected to represent the drainage area were now completely
described. Figure 62 has been constructed showing the conveyance system
within the drainage area. There are 26 gutter pipes in the drainage area
ranging in length from 61 meters (200 ft) to 1,051.5 meters (3,450 ft).
The total length of all gutters is 6,167.7 meters (20,235 ft). A total of
50 sewer elements, ranging in length from 3 meters (10 ft) to 1,024 meters
(3360 ft) with a total length of 119,638 meters (392,515) were modeled.
The diameter of these sewers ranges from 0.2 meters (0.66 ft) to 3-35 meters
(11.0 ft). The slope of these sewers was determined by taking the difference
In elevation between two manholes at the end of each element and dividing by
the length of the sewer. The elevations were obtained from detailed sewer
maps which were later found to be lacking in accuracy during spot checks
of the system. A detailed listing of the slopes was not available and the
present procedure was used as the only alternative. The manholes assigned
to the conveyance system of the Transport block were placed wherever a
change in slope or a branch in the system occurred. There are
68 manholes in the system. The other group of elements used in the con-
veyance system are the flow dividers. These elements are used to route a
208
-------
0)
§
«a
o
o
10
£
n
3
CO
SO
I
O)
209
-------
II)
"a)
o
(X
I/I
c
TO
vO
3
Cl
T
210
-------
portion of the flow from the main system to an overflow point or to divide
the flow between two branching elements. The data used to describe the
flow dividers are the diameter of the contributing sewers, the height
of the dam or weir in these sewers, the weir constant and the number of the
element into which the diverted and undiverted flow is routed. Figure 62
shows the six flow dividers in the conveyance system, numbered 81, 112, 209,
37, 87, and 46. Numbers 81 and 112 are the wetwells of the two treatment
units. The flow that arrives at these wetwells is pumped to the two
treatment units with any flow in excess of the plant capacity bypassed to
the river. Thus, elements 81 and 112 are flow dividers with undiverted
capacities of 42.1 cu m per min (24.8 cfs) and 116.8 cu m per min (68.7 cfs)
respectively. Numbers 37, 209, and 87 are the major flow dividers that
determine when flow is to be routed to the treatment units. During dry
weather, these elements prevent the passage of dry weather flow to the
treatment units. Number 46 routes the flow from the 145-8 hectare (360 acre)
subarea of the drainage system to either the main sewerage system in dry
weather or during wet weather when flows exceed 49.4 cu m/min (29-1 cfs) the
flow is bypassed to Lake Michigan. Since the amount of flow necessary to
cause bypass is relatively large, this element contributes significantly
to the downstream treatment units during wet weather which receive the
undiverted flow.
The Transport block requires the hourly and daily variations in the quality
and quantity of the dry weather flow of the drainage area. This variation
is necessary because the computations of the SWMM are dependant on the real
time of occurrence of the rainfall event. The Racine Water Pollution
Control Plant records provided the hourly and daily variation In the
quantity of the flow which was expressed as a ratio of the mean yearly
flow. The variation in the quality of this flow could not be fully
determined using the treatment plant records since the dry weather samples
are taken as 6 hour composites with no total coliform analysis performed.
For example, during the summer of 1974, dry weather flow entering the treat
ment plant was .sampled every hour for each day of the week during a dry wea-
ther period. The resulting 168 samples (24 samples per day for 7 days) were
analyzed for TOC and the Sunday, Monday, Wednesday, and Friday samples were
also analyzed for five day BOD, total suspended solids, and total col I forms.
The Tuesday, Thursday, and Saturday TOC values were then used to predict
the BOD variation for these days using simple correlation analysis. The
suspended solids and total coliform variations were determined by averaging
the other corresponding values of the week and weighing them according to
the treatment plant's 6 hour composites. Thus, the hourly and daily
variation for the quality of the dry weather flow was completed and could
be used as input to the SWMM. The average quantity of the dry weather
flow for the entire drainage area can be input to the Transport block or
the SWMM can compute this value using the population estimates provided for
each subarea. Initially the dry weather flow for each subarea was estimated
by determining an average dry weather flow per acre for the entire city.
This procedure was later felt to be inaccurate and during the fall of 1973
a portion of the dry weather flow from the drainage area was actually
measured and these data were then used to determine the average dry weather
flow rate. This value, along with the hourly and daily variation in the
211
-------
quantity and quality, provided a complete record of the dry weather flow of
the drainage area.
The amount of rainfall for each overflow event was measured by three rain-
gauges placed throughout the city as shown in Figure 61. A description of
each gage and their operation is presented in Section V of this report.
Each gage recorded the cumulative amount: of rainfall versus time by means
of an inked line on a strip chart. The values from these charts were then
converted into inch-per-hour intensities at five minute intervals. Because
of difficulties in reading the raingage values for certain times of the
rainfall, the intensities from each gage were subject to variations in
the real time of occurrence. This is important to remember when comparing
the SWMM output to actual conditions. The contributing area of each
raingauge was determined graphically by constructing the perpendicular bi-
sectors of the lines joining the location of each raingauge on a map.
The polygons that were formed gave an approximate outline of the contribut-
ing area. The subareas not covered by a polygon were then assigned to the
closest raingage and all subareas east of Main Street were assigned to
raingage I due to the difference in rainfall intensities near Lake Michigan.
input Data_
The data used in the Runoff block to describe the drainage subareas are
listed in Table Al, Appendix VI-A. The first column of this table lists
the subarea number from Figure 61. Note that the numbering system is I to
kB and 60 to 67. This was done to accomodate other modeled areas outside
of this main combined sewer area. The following columns either list the
number of the gutter/pipe used to convey the surface runoff to the main
system or the number of the inlet to the main system that receives this
runoff. The width, area, percent impervtousness, ground slope, raingage
number and land use within each subarea is also listed. The input data
used to describe the gutter/pipes of the conveyance system are shown In
Table A2, Appendix VI-A. The number of the gutter/pipe and Inlet to the
main conveyance system are shown in the first two columns. The width
(diameter), length and slope are also presented in this table. The place-
ment and format of this input data for the Runoff block is listed ?n
Table A3, Appendix VI-A.
The Input data used for the Transport block includes the description of the
conveyance system elements (manholes, sewers, flow dividers, gutter/pipes)
and the data used to characterize the dry weather flow of the area. Table
Bl, Appendix VI-B lists the data used to describe the manholes of the
conveyance system. The upstream elements that, are listed in this table
provide the SWMM with the types of branching and flow routing characteristic
of the conveyance system. Thus, manhole No. 114 has 3 upstream conduits
numbered 89, 93, and 92 that flow into it. Table B2 Appendix VI-B lists
the input data for the 76 sewers and gutter pipes of the area. All of
these elements are circular shaped having their diameters equal to their
widths. Two of these elements ar-e dummy sewers that are used to connect
the treatment units to the bypass channel of the wetwel1. These elements
were added to allow each treatment unit to be a single input to the re-
212
-------
ceivlng water rather than the effluent from the treatment plant plus the
bypass being added separately at each site. This would mean that the re-
ceiving water would need k inputs in less than 305 meters (1000 ft).
The characteristics of the dry weather flow of the drainage area were
obtained from population estimates, and the Racine Water Pollution Control
Plant records. The hourly variations in the quality of this flow were
determined by a week of actual sampling. The average yearly flow, BOD and
suspended solids concentration were then compared to the daily averages
and hourly values obtained from the dry weather sampling.
A ratio of the daily average to the yearly average produces the daily
correction factor and the ratio of the hourly average for the week to the
yearly average provides the hourly variation for flow and quality constitu-
ents. Table B3, Appendix Vl-B lists these ratios as they are used as in-
put to the SWMM. Table Bk, Appendix Vl-B provides the population densities
and actual populations for each subarea that has a residential land use.
Note that subareas kl and kS are the separated areas that only contribute
dry weather flow to the conveyance system. In order to provide an accurate
population equivalent for these areas, the population density used was 405
persons per hectare (9999 per acre) since the contributing area is,only
.Ok hectare (O.I acre). The other method of accounting for the dry weather
flow in the conveyance system was to assign process flows to those points
in the area where the dry weather flow from non-modeled areas enters the
system. Thus, at manholes A6 and hi process flows are added to account
for the contribution of the areas north of subarea 18 and the contribution
from another separated area, No. 19. These process flows contain the
yearly average dry weather BOD and suspended solids concentrations. The
magnitude of these flows was determined by use of data provided in an in-
filtration study performed for the City of Racine (personal communication
from Donohue and Associates, Consulting Engineers, Sheboygan, Wisconsin).
This study provided the flows for each of these areas as they entered
the main interceptor in the conveyance system. During November 1973 a
portion of the sewer system north of Site I was increased in size to pre-
vent problems with surcharging along this line. Sewer number 230 in
Figure 62 shows the repaired sewer location. While this sewer maintenance
was undertaken, the dry weather flow that normally passes through this
sewer was diverted to Site I and treated. The flow measurements taken
during this period provided the basis for modifying the computed dry
weather flows to fit these measured values. The final results of these
modifications are listed in the Transport block input data shown in Table
B5, Appendix Vl-B.
The Runoff and Transport blocks were now operational and the output of
these blocks was investigated to determine where calibrations could be made
if needed. But the application of the SWMM in this project was to an
existing area which provided few such possibilities. The only real cali-
bration that occurred was with the computed dry weather flow. Initial
estimates of this flow for the conveyance system of the area were approxi-
mately 6.8 cu m/min (4.0 cfs) while the measured value was 3-6 cu m/min
(2.1 cfs). By adjusting the total discharge area average sewage flow (ADWF)
213
-------
of the SWMM to be measured value, the computed dry weather flow rate became
closer to actual conditions. Both of these blocks were now ready to be used
in comparison with actual rainfall events and measured values.
Before the initial results with the SWMM are presented, the problems with
the flow measuring devices and data acquisition will be discussed. The
two treatment units discussed in a previous section provide a means for
monitoring the overflows. The wetwells of each unit (element numbers 81
and 112) have pumps that remove the flow at a constant rate equal to the
plant capacity. All flow below this value is removed to a treatment unit
and any flow arriving in excess of this value is bypassed to the Root River.
The plant flow is measured by a Parshall flume which is downstream of the
plant pumps. The bypass flow is measured by a bubbler tube placed at the
bypass weir. Both of the flow measuring devices record the flow in
gallons per minute on a circular time chart. The plant flow values are
obtained from a bulb that rides on the water surface in the Parshall flume.
Because of variations in the water surface through the flume, a range of
values for the plant flow readings at Site I will be presented. Thus,
a typical flow range may be 5.1 to 7-0 cu m/min (3.0 to 4.1 cfs). The
flow measurements at Site II were found to be faulty and appropriate
correction factors had to be applied. Because of these corrections only
a single line will be presented for the Site II plant flow. The bypass
recorder at Site I also required a correction factor. Both of these
factors are discussed in Section IV of this report. Throughout the
following comparisons, the arriving flow at each wetwell will be compared
to the computed flow. This means that the plant flow plus the bypass
flow are added together for each unit of time and compared with the SWMM
output for the arriving flow. This procedure allows better comparisons
since once the arriving flow exceeds plant capacity, the plant flow re-
mains constant at 42.1 cu m/m?n (24.8 cfs) for Site I and 116.8 cu m/min
(68.7 cfs) for Site II. The majority of the comparisons will be done for
Site I since this site provides accurate plant flow measurements and by-
pass records, the simplest flow divider situations and least amount of
mechanical problems that caused variations in the monitoring of the over-
flow. Site II requires a correction factor for plant flow for all runs and
for the first year of operation, no bypass record was obtained. This
site also has two gates in the sewers north of the plant that open and
close during various runs to either store some of the arriving flow in line
or to use this treatment site at or near capacity when flows in the
interceptor are low. Thus, in order to model these different physical
situations would require that certain s'torms be run in sections according
to the configuration of these gates. The operation of these gates is
explained in Section IV under the Design and Construction subsection
When complete data is available, both sites will be used for comparison.
The quality data used for comparison at both sites is obtained from seven
discretely sampled overflows. All other runs h,ave composite samples that
are proportioned according to flow. These values will be presented as
a basis for comparison to the SWMM output with the discrete data providing
a more accurate and detailed comparison.
214
-------
Initial Results
The comparison of the SWMM output to the measured arriving flows at the
treatment units began with run No. 12 which occured on July 20, 1973-
This was the first run which provided accurate raingage data, good flow
measurements at Site I and a significant amount of rain, 3-3 centimeters
(1.3 inches) in 3 hours. The rainfall intensities along with the other in-
put data related to this run are shown in Table El, Appendix VI-E. Only
Site I is used in this comparison because of mechanical problems with Site II.
Because of the large amount of rain, the plant capacity was exceeded from
2:00 to 4:30 pm. The resulting bypass and plant flow record and the computed
values are shown in Table 72. Figure 63 represents the graphical comparison
of the arriving (plant flow plus bypass) flow. As this figure indicates, the
computed flow lags behind the measured at the start of the run and terminates
before the measured. The long duration of the measured flow is thought to
be caused by infiltration since it does not occur for all overflow events.
Outside of the early termination of the computed flow, the flow comparison
was relatively close. The quality of the arriving flow was determined by
a composite sample taken over the duration of the overflow. Using this
value to determine the total kilograms (pounds) arriving for BOD and
suspended solids, and determining the same for the computed, a rough compari-
son can be made. Thus, the following values resulted;
Kilograms (1b)
BOD arriving
Ki lograms (Ib)
suspended solids arriving
computed
1869 (4117)
actual
1703 (3750
computed
2988 (6581)
actual
3691 (8130)
Little, if any comparison can be made between these values at this time.
The next overflow event used for comparison is Run No. 16 which occurred
on September 2k, 1973- Only Site I data are available because of mechanical
difficulties at Site II. Total rainfall for this run was 1.52 centimeters
(0.6 inches). The rainfall data used for this storm is shown in Table E2,
Appendix VI-E. A graphical comparison shows that the computed flow lags
behind the measured at the start of the overflow but then passes above the
measured and remains there for the duration of the overflow (see Fig. 6k).
This run was below plant capacity so that no bypass flow was recorded.
Table 73 lists the measured and computed flows. In Run No. 12, the
majority of the flow was bypassed and the computed flow was less than
the measured. In Run No. 16 with no bypass, the computed flow is greater than
the measured for a majority of the overflow. The quality comparison for
this run is based on the kilogram (pounds) of BOD and suspended solids
arriving at the treatment unit. The results are:
Kilograms (Ib)
BOD arriving
computed
174 (383)
actual
kk5 (980)
Kilograms (1b)
suspended sol ids arriving
computed
1360 (2996)
actual
1026 (2260)
215
-------
TABLE 72. ARRIVING FLOW, SITE I
Run No. 12
Time
hr
1355
1400
1430
1500
1530
1600
1630
1700
1730
1800
1830
1900
1930
Arriving
cu m/min
mm.
0.0
41.7
41.7
94.4
102.2
106.0
106.0
106.0
109.7
107.8
109.7
104.1
100.4
101.0
101.0
83.4
75.7
64.3
41.7
41.7
34.9
34.0
28.1
25.3
24.5
23.5
20.4
21.3
22.8
17.5
15.8
13.6
13.3
13.6
15.8
18.2
max.
0.0
45.4
45.4
98.1
105.9
110.7
110.7
110.7
113.4
111.5
113-4
107-8
104.1
94.7
94.7
87.1
79.4
68.0
45.4
45.4
40.1
40.8
39.9
32.5
28.4
27.2
24.1
25.2
26.5
20.2
19.0
18.0
16.3
16.0
18.4
21.6
flow,
cfs
min.
0.0
24.5
24.5
55.5
60.1
62.3
62.3
62.3
64.5
63.4
64.5
61.2
59.0
53.5
53.5
49.0
44.5
37.8
24.5
24.5
20.5
20.0
16.5
14.9
14.4
13.8
12.0
12.5
13.4
10.3
9.3
8.0
7.8
8.0
9.3
10.7
max.
0.0
26.7
26.7
57.7
62.3
64.5
64.5
64.5
66.7
65.6
66.7
63.4
61.2
55.7
55.7
51.2
46.7
40.0
26.7
26.7
23.6
24.0
20.5
19.1
16.7
16.0
14.2
14.8
15.6
11.9
11.2
10.6
9.6
9.4
10.8
12.7
Computed
cu m/m!n
0.0
3.2
24.0
21.0
38.6
68.7
81.4
97.2
82.8
86.0
95.2
92.5
90.6
80.8
76.5
66.3
61.7
63.1
49.0
34.0
23.1
14.5
7.3
2.0
0.0
flow,
cfs
0.0
1.9
14.1
12.3
22.7
40.4
47.9
57.2
48.7
50.6
56.0
54.4
53.3
47.5
45.0
39.0
36.3
37.1
28.8
20.0
13.6
8.5
4.3
1.2
0.0
216
-------
\
in
I
in
O
O
cn
o
o
co
<
O
o LL
£ O
§
.
cc
ID
o
I
o
o
o
o
-3-
O
o
0)
4J
CO
I
cn
c
ID
csj
!_
0)
E
C
•3
05
— «M
A1ISN31NI
TlVdN
CM
NIW/W HD
217
-------
\ I
VO
-------
TABLE 73. ARRIVING FLOW, SITE I
Run No. 16
Arriving flow,
Time
hours
2300
2330
2400
0030
0100
0130
0200
0230
0300
0330
o4oo
0430
0500
0530
0550
cu m/min
mm.
0.0
8.3
8.2
8.2
7-8
9.9
13-6
19.6
26. A
32.8
25.5
22.4
25.7
23.8
21.3
20.4
20.7
18.4
12.1
16.8
18.7
18.2
27.0
32.6
34.9
34.9
34.9
34.9
18.7
10.5
11.2
6.5
2.7
1.7
0.0
max.
0.0
13.9
13.6
13.3
13.3
15.3
18.9
22.4
28.9
37.7
29.2
28.6
31.0
28.7
28.7
25.7
25.8
23.0
17.3
21.3
23.1
22.6
33.2
39.1
39.8
39.8
39.8
39.8
23.8
14.0
15.1
8.3
4.8
3.4
0.0
cfs
min.
0.0
4.9
4.8
4.8
4.6
5.8
8.0
11.5
15.5
19.3
15.0
13.2
15.1
14.0
12.5
12.0
12.2
10.8
7.1
9.9
11.0
10.7
15.9
19.2
20.5
20.5
20.5
20.5
11.0
6.2
6.6
3.8
1.6
1.0
0.0
max.
0.0
8.2
8.0
7.8
7,8
9.0
11.1
13.2
17.0
22.2
17.2
16.8
18.2
16.9
16.9
15.1
15.2
13.5
10.2
12.5
13.6
13.3
19.5
23.0
23.4
23.4
23.4
23.4
14.0
8.2
8.9
4.9
2.8
2.0
0.0
Computed flow,
cu m/m?n
0.0
0.0
0.0
0.0
0.4
0.0
2.5
2.7
3.4
11.5
17.7
22.6
25.2
26.9 ~
27.5
27.7
27.7
26.5
23.8
20.2
20.7
26.0
27.4
44.7
52.2
53.9
55.8
55.8
50.7
42.5
34.5
27.7
22.3
17.9
14.3
11.4
8.8
6.8
5.1
3.6
2.2
1.0
0.0
cfs
0.0
0.0
0.0
0.0
0.2
0.0
1.5
1.6
2.0
6.8
10.4
13.3
1 •/ • J
14.8
15.8
16.2
16.3
16.3
15.6
14.0
11.9
12.2
15.3
16.1
26.3
30.7
31.7
32.8
32.8
29.8
25.0
20.3
16.3
13.1
10.5
8.4
6.7
5.2
4.0
3.0
2.1
1.3
0.6
0.0
219
-------
These results are the opposite of run Number 12 in that the computed BOD is
less than the measured and the computed suspended solids is greater than
the actual. In order to effectively compare the quality predictive portion
of the SWMM, it was decided at this time that a run was needed with discrete
sampling of the arriving flow at the treatment sites. This run would also
require good .ra'ingage and flow measurement data in order to effectively
compare the computed and actual quality constituents. Run No. 21 was
selected and it is discussed in the portion of this report which discusses
the discretely sampled runs (see Sec. VI-5; TOTAL PROGRAM EVALUATION).
VI-3 STORAGE BLOCK
The Storage block of the SWMM provides the capabilities for storing all or
part of the flow in selected elements and/or treating this flow using one of
several treatment options provided. For purposes of this project, only
the treatment portion of this block was used since there are no storage
facilities in the drainage area in Racine. The Storage block provides the
option of either designing a treatment facility to the maximum arriving flow
or using a design flow rate provided as input to size various processes
within the facility. This latter option was used throughout this application
since the screening/dissolved-air flotation units used for treatment have
already been constructed.
Data Acquisition
A complete description of the two treatment units is presented in Section IV-2
of this report. The data used to describe the characteristics of these
units were obtained from the as-built specifications. The treatment options
which were selected from the User's Manual correspond to the existing
units wjth the following components:
Bar racks
Inlet pumping
Fine screens and dissolved-air flotation
No secondary treatment
No effluent screens
No outlet pumping
No chlorine contact tank (chlorine added in #3)•
Input Data
The Storage block receives the routed flow from the Transport block at ele-
ment numbers 81 (Site I) and 112 (Site II) of Figure 62. The block is run
separately for each element with the design flow rate of the treatment units
provided in the input data. The pump head for the incoming flow to each
site is 6.09 m (20 ft). The dissolved air flotation units are provided with
chemical addition (including chlorine) and a 20% recirculation rate with
a 2.6 m (8.5 ft) depth of the flotation tanks. The design overflow rate
for Site I is 209-3 cubic meters per day per square meter (5131 gallons
per day per square foot) for Site I and 226.6 cu m/day/m (5555 gpd/ft2)
for Site II. This data was then input to the SWMM according to the formats
220
-------
specified in the User's Manual
final form for Site I and II.
Problems
Table Cl, Appendix VI-C lists this data in
After the input data for both sites were prepared, they were submitted with the
needed Runoff and Transport data to the SWMM. The first few runs were used
to debug the data from errors such as undefined element numbers in Transport
which provide data to the Storage blocks. After the correction of these
errors, further runs were needed to determine why this block continued to
terminate before completion along with large amounts of asterisk and nega-
tive^numbers in the printout of the inlet hydrographs and pollutographs.
The input data were ruled out as the cause of this error since different runs
caused the same errors. The program listing was then analyzed and the error
was found to be caused by the incorrect transfer of the hydrographs and
pollutographs from the Transport block to the Storage block. Thus, the BOD
output of an element from Transport was used as the suspended solids input
to Storage. This error was corrected by modifying about six statements in
the program listing. At this time Version II of the SWMM was obtained
which also contained the corrections for these errors. The new ver-
sion was compatible with the original that Is, ft accepts data freely and
has^fewer possibilities of underflow or zero divide errors. It retains all
of its original features, as well as an urban erosion capability, a hydrau-
lic design section, the capability of taking two separate drainage areas and
combining them into a single data set, and the addition of new treatment
options in the Storage block. When the input data were submitted to this
updated version, the program ran to completion without error.
Initial Results and Modifications
The Storage block was used for various runs of the SWMM to determine how
this block sized the treatment units to each overflow event. At this time
it was decided that the results of this block would not be acceptable
because the computed treatment modules were not "sized" the same as the exist-
ing units. For example, the treatment modules of the SWMM utilized screen
areas, submerged areas of screens, number of screens and design flow rates
that were different than the actual conditions. These parameters were im-
portant in determining the total removals for each unit. In order to com-
pare the computed to the actual results, it was decided to modify these
parameters in^the block so that the computations of this block fit the
existing conditions. These changes concerned the capacity of the treatment
units which are used for calculating the size and number of the bar screens
and fine screens.
As was mentioned earlier, the Storage block has been developed mainly for
design purposes. It determines the design flow QDESYN based on the module
size of the unit, QMOD, where QMOD is determined as being as small as
possible but greater than QDESYN. In order to save the actual QDESYN for
further computations, the statement
80 QDESYN - 1.5^7 * QMOD(K) TRTD
221
-------
of the Storage block was changed to:
80 CONTINUE
In this way the module size and the actual QDESYN will be utilized.
The statements used to calculate the number and size of the bar screens
were:
IF NSCRN IS LESS THAN 2, LET NSCRN EQUAL 2
Thus, 2 screens were the minimum possible. The capacity per screen in cubic
feet per second was calculated from:
^APATITY PFR
SCRAP = °-DESYN/NSCRN
SCREEN
The submerged area in square feet of each screen was then
SUBMERGED
SCRAp/3<0
The face area of the screen = 1.4 * SUAREA
These statements are found in subroutine TRTDAT, statement numbers 329
through 333. The changes that were made to these statements to provide the
Input of existing data were:
1200 READ (5,801) NSCRN, SUAREA, FAREAB
CONTINUE
SCRAP * QDESYN/NSCRN
CONTINUE
CONTINUE
A new data card was now- placed immediately after the card group 5 which re-
quires the design flow of the treatment unit. The new card in the data deck
has the following format:
FORMAT
110
2FIO.O
COLUMNS
1-10
11-20
21-30
DESCRIPTION NAME
No. of bar screens NSCRN
Submerged area SUAREA
Face area of screens FAREAB
The fine screens that precede the dissolved air flotation units are also
sized from QDESYN. In order to include the actual screen area (SCREEN), the
statement
222
-------
3400 SCREEN - QDESYN * 449/50 TRTD 420
was deleted and replaced by:
3400 READ (5,522) SCREEN
An input card was then placed Immediately following the level 3 treatment
cards as follows.
FORMAT
FI0.2
DESCRIPTION
SCREEN AREA - FT
Figure 65 presents the printout of the Storage block before and after these
corrections. Note the differences in the number of bar screens, submerged
area and face area of the bar screens and the total fine screen area in
level 3.
The Storage block was now completely operational with data defining both
treatment units used in the calculations of this block. The only other
change to this block was to "clean up" the printout of the "Performance
Per Time Step" section of the output. Here the values associated with the
listings of concentrations for certain treatment levels were either extremely
large or negative when they should have been zero. Thus, when the arriving
flows or overflows are zero or approaching zero (.001), the BOD or sus-
pended solids concentrations were negative or extremely large. Figure 66
shows a typical printout of these values. Note that when the arriving flow
(ARR) or the overflow (OVF) is zero, the BOD or suspended solids leaving
or removed from the treatment level are very large or negative. These
values do not affect any computations since the flows are zero but they do
clutter the printout with unnecessary and incorrect data. These errors
were caused by very small flows being used in the denominator of a calcula-
tion and the results were meaningless. To correct this procedure, the pro-
gram listing was modified so that if the arriving flow or overflow was less
than .01 cubic feet per second, this flow was set equal to zero. Subroutine
TREA contains the calculations of these values. Table C2, Appendix VI C
lists these changes and their location within the Storage block. To imple-
ment these changes, IF statements were added and other variables were set
equal to zero. The results of these changes are shown in Figure 67.
The Storage block output was now ready to be compared to real data. Run
Numbers 12 and 16 were used to debug and modify the data since only compo-
sites of the effluent samples were taken during the operation. Run number
21 used discrete sampling of the effluent from treatment Site I and was
therefore used for the first comparison. The results are discussed in a
later portion of this report.
VI-4 RECEIVE BLOCK
The Root River which flows through the City of Racine was used for the
application of the Receive block. This block consists of two major sections
223
-------
SPECIFIED TREATMENT CAPACITY USED. (Before modification)
DESIGN FLOWRATE = 60.70 CFS.
TREATMENT SYSTEM INCLUDES MODULE UNITS
DESIGN FLOW IS THEREFORE INCREASED TO NEXT LARGEST MODULE SIZE
ADJUSTED DESIGN FLOWRATE = 68.70 CFS = 50,000 MGD.
(KNOD » 8)
NO STORAGE FROM A SEPARATE STORAGE MODEL IS ASSOCIATED WITH THIS TREAT-
MENT MODEL
I)
SUBMERGED AREA
FACE AREA OF BARS
PRELIMINARY TREATMENT BY MECHANICALLY CLEANED BAR RACKS (LEVEL
NUMBER OF SCREENS = 2
CAPACITY PER SCREEN = 3^.35 CFS
11.43 SQ. FT (PERPENDICULAR TO THE FLOW)
16.03 SQ FT
INFLOW BY INLET PUMPING (LEVEL 2)
PUMPED HEAD = 20.00 FT WATER
TREATMENT BY DISSOLVED AIR FLOTATION (LEVEL 3)
MODULE SIZE = 25 MGD
NUMBER OF UNITS = 2
TOTAL DESIGN FLOW = 50.00 MGD = 60.70 CFS
DESIGN OVERFLOW RATE = 5555.00 CPD/SF (5000 SUGGESTED)
RECIRCULATION FLOW = 20.00 PERCENT (IS SUGGESTED)
TANK DEPTH =8.50 FEET
TOTAL SURFACE AREA = 9593.20 SQ FT
CHEMICALS WILL BE ADDED
CHLORINE WILL BE ADDED
TREATMENT OF FINE SCREENS (AHEAD OF DISSOLVED AIR FLOTATION (LEVEL 3)
TOTAL SCREEN AREA = 633. SQ FT
NO SECONDARY TREATMENT INCLUDED (LEVEL 4)
NO EFFLUENT SCREENS (LEVEL 5)
OUTFLOW BY GRAVITY (NO PUMPING) (LEVEL 6)
NO CHLORINE CONTACT TANK FOR OUTFLOW (LEVEL 7)
Figure 65. Storage block printout before corrections.
224
-------
SPECIFIED TREATMENT CAPACITY USED.
(After modification)
DESIGN FLOWRATE == 60.70 CFS
TREATMENT SYSTEM INCLUDES MODULE UNITS
DESIGN FLOW IS THEREFORE INCREASED TO NEXT LARGEST MODULE SIZE
ADJUSTED DESIGN FLOWRATE = 68.70 CFS = 50.00 MGD
(KNOD = 8)
NO STORAGE FROM A SEPARATE STORAGE MODEL
MENT MODEL
IS ASSOCIATED WITH THIS
PRELIMINARY TREATMENT BY MECHANICALLY CLEANED BAR RACKS (LEVEL l)
NUMBER OF SCREENS = I
CAPACITY PER SCREEN - 68.70 CFS
SUBMERGED AREA = 550.00 SQ. FT. (PERPENDICULAR TO THE FLOW)
FACE AREA OF BARS = 125.00 SQ. FT.
INFLOW BY INLET PUMPING (LEVEL 2)
PUMPED HEAD = 20.00 FT. WATER
TREATMENT BY DISSOLVED AIR FLOTATION (LEVEL 3)
MODULE SIZE = 25 MGD
NUMBER OF UNITS =2
TOTAL DESIGN FLOW = 50.00 MGD = 68.70 CFS
DESIGN OVERFLOW RATE - 5555.00 CPD/SF. (5000 SUGGESTED)
RECIRCULATION FLOW = 20.00 PERCENT (IS SUGGESTED)
TANK DEPTH = 8.50 FT.
TOTAL SURFACE AREA = 9593-20 SQ. FT.
CHEMICALS WILL BE ADDED
CHLORINE WILL BE ADDED
TREATMENT BY FINE SCREENS (AHEAD OF DISSOLVED AIR FLOTATION) (LEVEt 3)
TOTAL SCREEN AREA = 550. SQ. FT. -••-.- •
NO SECONDARY TREATMENT INCLUDED (LEVEL 4)
NO EFFLUENT SCREENS (LEVEL 5)
OUTFLOW BY GRAVITY (NP PUMPING) (LEVEL 6) :
NO CHLORINE CONTACT TANK FOR OUTFLOW (LEVEL 7)
Figure 65 (continued). Storage block printout after corrections,
225
-------
— re
O m
0)
1
•—I
0)1
0)1 CS
**~ |C3
103
Ml
Ml
(01
OJ—
>4
(Alt/)
j!
£!§
•^•%j
ICQ
1
1
1
1 U
15
1
I
*—
^^
cr>
E
U)
u
en
E
1
M
IV-
1
§
•v.
f
^
i
^^
9'
M
U-
C
"I
U
T
o
1
o
o
•
o
T
o
1
o
o
o
o
o
o
o
o
0
o
•
o
ec.
<
IA
m
CM
0
O
O
*
O
•
IA
1
1
§
O
Ul
o
o
o
^
0
o
•
o
u.
o
1
o
1
o
o
•
0
1
o
1
0
o
0
4
0
0
o
o
o
o
o
•
o
a:
ec.
<
0
Jf
CM
0
o
o
CD
IA
CM
t
en
i
§
•
o
t£
O
O
o
o
0
•
o
u.
o
MM
1
0
1
o
o
0
o
T
o
1
8
0
fe
0
o
o
o
o
8
'k
o
a:
a:
Uf\
-3'
?*i
O
0
o
o
ff\
IA
CM
1
cr»
o
0
o
i
o
o
o
o
o
0
u.
>
0
1
o
1
0
o
0
T
o
0
o
o
0
o
o
o
o
o
o
0
0£
on
<
o
tA
CM
O
o
0
o
IA
CM
1
cn
o
o
o
i
o
0
o
o
o
0
u.
>
0
— o
o o
1
o
o o
o o
1 tA
CM
1
O CT\
1 1
O O
o o
0 0
£5
o oc
0 0
o o
o o
o o
§0
0 O
DC U.
K. >
< 0
IA
U\
CM
CM O
£ f
— O
^^
O
— o
O 0
CM-*
— IA
0°%
*"*
IA
CM
O O
0 0
=> UJ
fv.
o
u.
CO
vO O
0 0
00 0
vO
IA O
P
0 0
c£ ii-.i
O£ 5*
< 0
o
o
•*
c^
o
ID
U
£
(U
ja
M
8
"rt
O
I
to
CD
C
vO
£
O)
(I
226
-------
I/)
3
p— •
<*-
4J
3
O
U
O
i»-
IZ
o
0
t/>
to
0
o
CQ
„
(0
I*J
o
"i
o
o
x^
^x
C
t
•MV
*>.
9
fim
•*+.
I1
m
u-
u
0
o
•
•M*
1
•
o
1
o
o
•
0
o
0
•
•M
1
•
O
1
o
o
•
o
o
0
•
«•
1
»
o
1
o
o
•
o
o
o
•
••»
1
»
o
o
o
•
o
o
o
•
•M
1
•
o
o
•
o
-3-
p
Ul
cs
\0
o
•
f>4
•"•
•
•^
CA
o
^
p
•o
4)
C
II
C5 C5
— o
ei—
tot o
o
•
o
o
•
o
o
•
0
Ul
CM
vO
S
O)
Ul
mica
4->IO
ClQQ
Ol
Ul
oi
2:1—
I «D
t-M
I O
II-
O)
o o
o
o o
O O
I
o
o o
o o
o
o o
o o
o o
I
o o o o o o
o o
o o
o
op
— o
o
— o
o o
a: u. a: u.
oc > a: >
< o < o
LA O
•< o
CM
GC tt*
< O
O
LA
• •
CM
OC Ix*
< o
LA
LA
a: u.
ec >
< o
o
o
227
-------
s
vu
r
i
i1
0
a >•
S I
•J 3
m vi
o o
^ i
*• „,
U. tf
C1 <
• RCCAKOLESS
TANX1
ENT SCREENS)
u uu.
m x
VED IN XEVEtS 2«
rEL 1 (CHLORINE CO
(AT HAR RACKS AM
r£»
u-l-l
Ul M O
•« VI Ul
VI <
a ui ^
-
X
CO O
X
-IV)
H O
a
t—
U. *H
O X
VI *^
a -j
§5
Ul U.
h- O
z
X
a:
o
o
o
o
0 0
o o
o o
* *
0 0
0 O
0 0
00
o o
o o
o o
o o
W y
71 UJ
c a
o o
o o
o o
oo
o o
00
11
in
(M
o
i
o
o
o
0 0
00
0 0
o o
o o
o o
o o
0 0
o o
O 0
0 O
»- X
SUJ
a:
O 0
o o
o o
o o
o o
00
or a.
< o
o
CM O
-• o
o o
t£ U.
IA
CM
o
UJ
o
**•
CD
o
o
0
UJ
f-
o
eo O
0
O 0
o o
in o
-t
CO O
o
0 O
in o
CO O
m
o
0 O
=> UJ
a •*.
0
UJ
CO
IA O
0 0
eo o
^
CM 0
o o
or a.
ss
0
m
o
s
o
fM
N
O
O
m
•o
•4- O
fM O
t •
in o
c*l O
sO
•* o
CM O
IA O
m o
O 0
en o
in
in o
-* 0
.0 0
CM
in
in o
rM
^ Ul
a e;
o
Ul
CM
— i O
O 0
eo o
CO
IA O
CO
in o
or u.
a: >
< a
in
rt
o
UJ
o
0
-a
CO
in
o
U)
o
en
CO
r* o
in o
CM
o<- o
CO
r- o
in o
PC
0
O 0
m
CO
f- O
CM
•D LU
Cj QZ
o
Ul
-4 O
0 O
0* O
f- 0
i\j O
c£ >
< T
in
rg
2
o
UJ
IA
O
CM
CO
in
0
UI
in
o
r\j o
00
in o
in o
fl O
r\J O
fl
CO
in o
tn o
3°
•M O
ro
in o
in o
*- 3"
O cc
0
UJ
* O
0 O
^ 0
E°
•D O
tn o
oi'u,
a: >
o
m
2
UJ
0"
o
•0
»M
O
UJ
0
•* o
>0 0
CM o
r- o
•c
* O
CM
CM
•O O
CM O
r- o
>4- o
CM
* •
r; uj
C a:
o
tu
f- O
0 O
r*- o
in
rM o
a: U,
3§
1A
CO
O
UJ
IA
O
CM
o
CM
-o
0
o
UJ
IA
O
0 O
fM
sO
O O
0 O
n o
CM
o o
M
•£>
0 0
0 0
m o
CM
•T
O O
CM
•0
O 0
0 O
^ UJ
UJ
-* o
0 0
CT- o
CM
cr-
0 0
0 O
C£ U.
ce: >
< 0
0
fi
0
0
o
o
o
o
0 0
0 0
0 0
0 O
o o
0 C
o o
o o
0 0
0 0
0 0
t- f.
z> '-It
o o
0 0
0 O
00
o o
0 0
2: a.
f£ >
< o
in
m
o
0
0
o
o
o
o o
,o o
0 0
O 0
0 O
c o
o o
o o
o o
•=> ui
C CC
0 O
o o
0 0
0 0
o o
0 O
1 -
o
0
o
o
o
0
0 0
0 O
O 0
0 0
o o
o o
0 O
0 O
0 O
0 G
i- s:
tj. 0,
O T
0 0
O 0
o o
o o
0 0
a: LL
*->
M-
(0
U)
3
"ro
o
c
t_
Ct
2
p
u-
t_
o
TJ
c
1_
)_
D)
u.
228
-------
which may be run together or separately. One section, designated QUANTITY,
determines the hydraulics of flows for the receiving body while QUALITY
determines the concentration of selected constituents at points within the
modeled area. The major effort in the application of this block was with
the QUANTITY section; once operational, the data obtained from Section V
Root River Monitoring_ StudJies^ was used in the QUALITY portion.
Data Acquisition
The collection of data used to define the receiving body for the SWMM was
initiated shortly after the Storage block was completely operational. The
data needed as input included water surface elevations, depths, widths, flows,
velocities, any head relationships and the loadings of selected constituents
and their decay or reaeration rates. The data defining the water surface
elevations and flows were obtained from the Root River Watershed Report (18)
and reports from the Wisconsin Department of Natural Resources (19). From
thesedata,the Root River from Horlick dam to the harbor entrance of Lake
Michigan was selected as the modeled area. This 10 kilometer (6 mile)
section of the river provides a source for upstream water quality monitoring
before entering the City of Racine at Horlick dam and contains all the
combined sewer overflow locations that discharge to the river.
The Receive block requires the modeled area be sectioned into a series of
channels or triangles which are connected by node points or junctions. These
elements are assigned numbers which are used throughout the simulation to
identify the inflow points, head relationships and changes in the physical
layout of the receiving body. Figure 68 presents the layout of the receiving
waters that was used for the initial runs of this block. There are 16 junc-
tions and 18 channels used to describe the area. The junctions along the
river correspond to the location of the combined sewer overflow points and
major storm sewer discharges to the river. The channels inside the harbor
area define three triangles which are used for open bodies of water. The
depths and surface elevations of the junctions were obtained by actual
measurements conducted by project personnel and the Racine City Engineer's
Office. The data used to describe the junctions and channels are presented
in Tables Ik and 75.
The Receive block provides three options to the user to define the head re-
lationship of the system; tidal influence, dam, or specified inflow and
outflow. Therefore, a tide or dam could be applied at the harbor entrance
to Lake Michigan to simulate the effects of the lake on the mouth of the
river. It was decided that the input data would use a dam at Junction 16.
The lake effect would be simulated by placing the elevation of the dam
slightly higher than the water surface elevation of the harbor. This would
tend to "hold" a small portion of the river flow in the harbor area. The
total inflow that is assigned to the system was the flow that was recorded
on the day in which the water surface elevations and depths were obtained.
This flow was 127-5 cu m/min (75 cfs) and is applied at junction I.
229
-------
t/1
"flj
C
ro
•s
T3
C
ra
in
8
o
0)
8
cc
VO
a
CD
iZ
230
-------
TABLE 74. RECEIVE BLOCK JUNCTIONS (INITIAL)
Junction
Mo.
1
2
3
4
5
6
7
$
9
10
11
12
13
14
15
16
Initial
meters
187.8
185.6
183.5
1 82 . 0
180.4
179.2
177.7
177.7
177.4
176.8
176.5
176.4
: 176.4
176.3
176.3
176.2
head,
feet
616.0
609.0
602.0
597.0
592.0
588.0
583.0
583.0
582.0
580.0
579.0
578.8
578.7
578.4
578.4
578.0
Depth,
meters
186.5
184.7
182.6
180.4
179.2
177.4
176.2
175.6
174.3
172.2
169.5
167.6
174.3
173.7
173.7
167.0
feet
612.0
606.0
599.0
592.0
588.0
582.0
578.0
576.0
572.0
565.0
556.0
550.0
572.0
570.0
570.0
548.0
Cord mates,
X
0.0
0.8
1.8
3.2
5.6
5.3
5.0
7.0
8.6
10.7
10.8
12.6
12.4
13.5
13.3
4.2
Y
12.0
9.5
6.1
4.4
4.6
2,9
0.0
1.8
3.8
4.6
5.6
5.5
3,8
4.5
6.1
5.5
231
-------
(A
•3
•o
ee
O
•—
or>^r>O^LA'— *—
or-r-«— -*-=rcoc^
-------
Input Data
The input data used for the initial runs of the Receive block are
shown in Table Of, Appendix VI-D. This block was run separately from the
other blocks with input of storm water quantity and quality from cards rather
than the tape transfer from Transport. This procedure was followed until
the Receive block was completely operational, then the entire SWMM was
run at one time to include the outflows from the Transport or Storage block.
P rob1 ems
In
The data describing the receiving water were input to the SWMM with the pro-
per job control language. The first few runs were used to debug some of
the input errors that occurred because of the difficulty in interpreting
the user's manual. The output of this block continued to terminate before
completion of the QUANTITY section because of errors at the printout of the
hydrograph inputs to the system. Numerous runs were then made with slight
modifications of the input data. The output always terminated in the same
location but different errors were the cause. In some cases the termina-
tion was caused by a divide by zero error or a system error. The test
data from Lancaster ran without error in previous runs so the SWMM itself
was not at fault. Next the program listing was consulted to determine the
exact location of the errors. The SWFLOW subroutine contained the state-
ments causing the error but the reason for the errors was not clear. At
this time in the project, the overflow events of 197^ were occurring and
the resulting data were used in the Runoff, Transport and Storage blocks.
addition to these runs, the Receive block data was modified in a stepwise
manner to try to duplicate the test data. The changes in the input data
were: increase the length of the channels by removing..some of the junctions,
increase the depths, use smaller time steps, remove the storm water inputs,
and modify the data used to describe the dam at junction 16. All of these
changes produced errors in the same location which again terminated the out-
put. After contacting and providing the University of Florida with
the input data, the following changes were recommended: remove the dam
at junction 16, use the specified input-output flow and use 30 seconds for
the hydraulic time step. When these changes were implemented, the output
progressed past the previous errors and then terminated when the velocities
in the channels exceeded 20 feet per second. The Receive block contains
this "check" to insure that the computed results that follow are reasonable.
The mechanics of this check are the following: if the velocity that is
computed within any time step for any channel is greater than 20 feet per
second, the output terminates and prints out the junction number and other
hydraulic data where the error occurs. The standard fixup for this error
is to increase the length of the channel so that the calculation of the
velocity i-s taken over greater distances. Again, various modifications of
the input data were undertaken to decrease the existing slope in
the river and determine if the velocity decreased with this change. V/hen
this change was implemented, no significant change in the velocity resulted.
At this time various suggestions received from other SWMM users included
increasing the length of the hydraulic time step or increasing Mannings
233
-------
coefficient by a factor of 10. Most of the suggestions were to modify the
existing data to fit the SWMM. This was not the purpose of this project
and further modifications of this type were dropped. The program itself
was analyzed to determine if modifications could be made to allow the
existing data to be input to obtain as reasonable and consistant as the
output. The experience with the velocity errors indicated that the calcu-
lations of the flows for certain channels converged rapidly to zero flows
and the resulting "dry channel" is then used to determine the resulting
velocities which are extremely large at these small heads. The physical
layout of the junctions and channels was then changed to a system consisting
of only 7 junctions and 6 channels as shown in Figure 69. The data used
to describe these modified junctions and channels are presented in Tables
76 and 77- The junctions of Figure 69 correspond to the major combined
TABLE 76. RECEIVE BLOCK CHANNELS (MODIFIED)
Channel
No.
1
2
3
4
5
6
Length,
meters feet
945
1433
91^
914
1829
4115
3100
4700
3000
3000
6000
13500
Width
meters
61
53
61
46
19
23
t
feet
200
175
200
150
63
75
sq.
485
236
91
49
15
28
Area,
meters sq. feet
5220
2537
980
525
158
300
TABLE 77. RECEIVE BLOCK JUNCTIONS (MODIFIED)
Junction Initial
Mo.
1
2
3
4
5
6
7
meters
176.2
176.5
177.4
177.7
177.7
180.4
187.8
head ,
feet
578.0
579.0
582.0
583.0
583.0
592.0
616.0
Depth,
meters
167.0
169.5
174.3
176.5
176.5
179.2
186.5
feet
548.0
556.0
572.0
576.0
579.0
588.0
612.0
Cordlnates ,
X
15.0
10.8
8.6
7.0
5.0
5.6
0.0
Y
5.5
5.6
3.8
1.8
0.0
4.6
12.0
234
-------
o;
c
(0
u
(G
O
U
C
3
4)
U
&
cn
VD .
0)
0)
-------
sewer overflow points of the area. The harbor area of Lake Michigan was
represented as a wide channel instead of a series of triangles. To prevent
confusion between the two layouts, the numbering system was changed from
junction 1 as the start of the modeled area to that of the end or lowest
junction. Where previously there were two junctions relatively close to-
gether, these were now combined into one junction. The downstream head
relationship was now changed to a tidal effect having a small constant
influence on the lower reaches of the river.
When the modified input data for this block were submitted to the SWMM, the
output ran to completion without error. The head and velocity profiles
along the*system were not very consistent but the quantity portion of this
block was now operating. The quality portion of the block was then added
to simulate the entire impact of the combined sewer overflows.
The first few runs with the Receive block were obtained from running this
block alone and using cards to input the stormwater characteristics. Later
runs used the Runoff, Transport,and Storage blocks previous to the Receive
block. This procedure required that the outfalls from the two treatment
units at Sites I and II be combined into one junction in the river since
two separate junctions less than 152.4 m (500 ft) apart would cause problems
with the velocity determinations in the program. Thus, the wetwells at each
site were connected by two dummy elements as shown in Figure 70. In order
to keep the total number of Transport elements less than 150, the two
elements at each treatment unit which routes the specified plant flow from
each wetwell, (numbers 217, 77, 83, and 84) were dropped. Therefore, when the
Storage block is run, it is applied to the wetwells at each site and any
flow in excess of the design flow is bypassed as overflow. The dummy
elements carry the treated outflows and overflow from each site to the new
common element numbered 2, which corresponds to junction 2 in the Receive
block.
Results
The entire Receive block was run using different stormwater inputs from
cards or transferred by tape from storage or transport. The quantity re-
sults were extremely variable with large differences in stage between time
steps for the same junction as shown in Table 78. The velocities and
flows listed for each channel varied in magnitude with some being positive
and others negative. Again, changes were made to the pertinent input data
to rectify these problems, but no improvements resulted.
The quality determinations that were obtained from these runs were relatively*
constant for all parameters except 00. The initial concentrations of the
four constituents at each junction were added to the input data. Thus,
the initial BOD averaged 2.5 mg/1, suspended solids 30 mg/1, coliforms
500 MPN/IOO ml and a DO of 7.5 mg/1. During the first day of simulation
when the overflows to the river were occurring, the BOD, suspended solids
and coliform concentrations did not change appreciably but the DO dropped
to less than I mg/1. During the second day of simulation, the BOD and
suspended solids concentrations started to build up in the middle of the
236
-------
O
(0
u
(T
111
h-
O
O
s-
u
I/)
A)
ro
2!
3
237
-------
"3
—
—
z
£
p
5
o
LU
^>
LU
O
LU
OC
CO
r-.
LU
CO
|2
__±
VI
s
<£> *O
O
n Q)
—
c
n o
_) Ij
m|
-a- j)
•—
c
c 5
n o
o IT
4J
•V O
ra>
o
c
,
*J
|V% U
o
*"* s>
"3
c
n B
£ IZ
X
CM|
•—
*~ >
—
c
TO B
o tZ
^
o
01
Q
^~
,_
in
^i
(/>
^_,
14-
o
in
Q
^—^
M-
O
, — .
O.
— •
'uT
4-
O
's
ill.
'in
^
- — •
r^%
^,
r
O
o
t--.
,_
o
o
CO
CM
O
o
o
0
0
o
o
in
o
o
o
o
o
0
o
m
in
,_
i
o
f\
in
i
ft
o
i
o
vO
f
{^
O
o
VO
CM
CM
O
O
CM
^*
*-
O
o
CM
in
in
o
o
-------
receiving body (junctions 4, 5, and 6), the conform concentrations increased
slightly throughout the system and the DO concentrations dropped to zero in
all junctions. These same patterns were found for all the runs of the Receive
block which used the inflows to the system which were transferred by tape
from Storage or Transport or listed on cards at hour intervals for all
junctions of the river. Actual river monitoring did not show these trends
for any overflow events. The low DO readings that were computed were found
for various decay rates, reaeration rates and initial loadings. Although the
output of this block was questionable, an attempt was made to determine
if the two treatment units had any effect on the computed receiving water
quality. To accomplish this simulation, the receive block was run using the
untreated outflows from the Transport block. The resulting water quality
effects were then compared with the Receive block output when the Storage
block v/as used to treat the overflows to the river. No difference in the
quality of the river for each run was found during the three days of simu-
lation.
VI-5 TOTAL PROGRAM EVALUATION
The following portions of this report will present-the results of seven
discretely sampled runs at both sites. These runs provide discrete quality
data of the arriving and effluent flow from each treatment unit. In order
to evaluate how well the SWMM predicts the quantity and quality of these
flows, two procedures will be followed. The computed and actual hydrographs
will be integrated to determine the total actual and computed flow arriving.
The ratio of the actual to the computed flow will then be listed to give an
indication of the SWMM's ability at predicting the total arriving flow.
The manner in which these computed values follow the peaks and slopes of
the measured graphs will be expressed by visually rating the goodness of
fit of the two curves. If the actual hydrograph is closely paralleled by the
computed, then the goodness of fit is excellent. When little correlation
in the trend of the two records is found the goodness of fit is poor. Thus,
even though the ratio of the total arriving flows may be 1.0, the goodness
of fit may be poor. The pollutographs for each run will be compared solely
on the goodness of fit of the two curves. These two methods have been
selected since other methods such as correlation analysis have proved un-
satisfactory due to the differences in the start and finish times between
the computed and actual hydrographs and the small number of actual samples
used in each analysis.
Run Number 21
The first run that provided discrete samples of the arriving flow was number
21 which occurred on November l*f, J973. The average total rainfall for
the entire area was 1.9^ centimeters (0.76 inches) over a duration of 400
minutes. There were ]k days prior to this run when no rain fell and
21 days for which the cumulative rainfall was less than 2.5^ centimeters
(I inch). Table 79 lists the actual and computed arriving flows at Site I
and Figure 7' presents the graphical comparison of the data. The ratio of
the total measured flow arriving to the total computed flow is 0.80. The
239
-------
TABLE 79. ARRIVING FLOW, SITE i
Run No. 21
Time
hours
1650
1700
1730
1800
1830
1900
1930
2000
2030
2100
cu
mln.
4.9
5.3
6.1
7.5
12.1
14.3
15.8
19.7
24.1
25-7
24.1
25.7
31.8
32.5
33-8
40.7
35.8
32.8
31.8
31.5
29.4
24.1
23.8
25.0
20.4
19.2
18.5
Arriving
m/min
max.
9.9
10.5
10.5
11.7
15.8
18.5
22.3
25.0
30.3
32.1
32..1
31.8
41.8
42.9
43.2
46.5
44.2
41.5
39.1
36.2
33.7
29.4
28.7
30.0
25.3
23.8
23.0
16.20 20.7
2130
2200
2230
2300
2330
2400
Q030
18.5
18.2
18.9
15.1
18.5
22.6
26.5
25.0
25.0
24.1
23.1
23.5
24.1
22.3
18.9
17.3
22.3
22.6
23.1
19.7
22.6
28.7
31.1
30.0
29.6
28.7
26.9
28.0
28.7
26.9
24.1
22.0
flow,
min.
2.8
3.1
3.6
4.4
7.1
8.4
9-3
11.6
14.2
15.1
14.2
15.1
18.7
19.1
19.9
23.9
21.1
19.3
18.7
18.5
17.3
14.2
14.0
14.7
12.0
11.3
10.9
9.5
10.9
10.7
11.1
8.9
10.9
13.3
15.6
14.7
14.7
14.2
13.6
13.8
14.2
13.1
11.1
10.2
cfs
max.
5.8
6.2
6.2
6.9
9.3
10.9
13.1
14.7
17.8
18.9
18.9
18.7
24.6
25.2
25.4
27.3
26.0
24.4
23.0
21.3
19.8
17.3
16.9
17.6
14.9
14.0
13.5
12.2
13-1
13.3
13.6
11.6
13.3
16.9
18.3
17.6
17.4
16.9
15.8
16.5
16.9
15.8
14.2
12.9
Computed
cu m/min.
2.0
2.2
2.4
2.7
3.6
5.3
6.8
7.7
8.5
13.4
23.3
30.8.
42.0
56.1
64.8
66.8
66.8
65.1
59-7
53-9
48.8
44.2
40.1
36.7
33.5
30.9
29.9
29.8
30.3
31.5
33.0
34.5
35.0
35.4
35.2
34.7
33.8
32.8
31.1
29.4
26.9
24.3
21.6
19.4
17.3
15.8
14.5
flow,
cfs
1.2
1.3
1.4
1.6
2.1
3.1
4.0
4.5
5.0
7.9
13.7
18.1
24.7
33-0
38.1
39-3
39.8
38.3
35.1
31.7
28.7
26.0
23.6
21.6
19.7
18.2
17.6
17.5
17.8
18.5
19.4
20.3
20. f.
20.8
20.7
20.4
19.^
19.3
18.3
17.3
15.8
14.3
12.7
11.4
10.2
9.3
8.5
240
-------
o
f\
fl
CM
O
lf\
CM
CM
O
-------
goodness of fit of the two curves is generally acceptable since there is only
s slight lag of the computed peak behind the measured.
The series samples collected during this run were taken manually since
the treatment unit at Site I was operating automatically on the diverted
dry weather flow when this rainfall started. Because of this operational
problem, no discrete samples were taken during the first two hours of the
overflow. After this delay, samples of the arriving flow and plant effluent
were taken at 10 to 15 minute intervals. Table 80 lists the computed
and measured quality values of the arriving flow. No coliform analyses were
performed on these samples. Figure 72 presents a graphical comparison of
the computed and actual concentrations and shows that the computed
BOD values were much higher than the measured throughout the -run. A
review of the pertinent input data was conducted to determine where calibra-
tions could be made. The only possibility was with the data describing
the contents of each catchbasin in the area. The BOD of the stored contents
of each catchbasin was initially estimated as 400 mg/1. After this run with
the SWUM, a few of the catchbasin in the area were sampled and composited
and the results indicated a BOD value of 150 mg/1. It was then decided to
change the BOD value that is used as input to Runoff block to 150 mg/1 and
to rerun the SWMM. The results of this change are listed in Table 81
and plotted in Figure 73. This change brought the BOD values closer to
the measured and the resulting goodness of fit is now excellent for both
the BOD and suspended solids pollutographs.
The effluent from the treatment unit at Site I was also discretely sampled
for comparison with the output of the Storage block. Table 82 lists the
computed and actual effluent concentrations while Figure Ik presents the
graphical comparison. Since the arriving quality comparisons were so close
during this run, the resulting Storage block output provides a good check
of the SWMM's ability to simulate the removals from the treatment units.
As Figure 75 indicates, the goodness of fit of the two curves is excellent
throughout the simulation. The Storage block is, therefore, accurate in its
predictive capabilities.
The computed arriving flow at Site II was not affected by the sewer main-
tenance at Site I since the Transport network of the SWMM was modified to
fit the existing conditions. Table 83 lists the actual and computed flows
arriving at Site II and Figure 75 presents the graphical comparison of
the flows. The ratio of the total measured flow arriving to the total
computed arriving flow is 1.06. The computed peak is much lower in magni-
tude than the measured peak. This difference could be due to the mechanical
problems in determining when and if the sluice gate in element 146, Figure
62, closed to channel more flow into this site. The goodness of fit of
the two curves is acceptable throughout the run.
The arriving quality at this Site is determined by eight series samples
taken during one hour of the overflow. Table 84 lists the measured and
computed values for this run and Figure 76 presents the graphical comparison.
The measured values show a definite peak in the BOD and suspended solids
concentration which is preceded by the computed peaks of almost equal
magnitude. The goodness of fit of the two curves is generally acceptable.
242
-------
TABLE 80, QUALITY OF ARRIVING FLOW, SITE I
Run Number 21
Catchbaslrs BOD - 400 mg/l
Time
hours
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
2330
2400
Actual
BOD
mg/l
74
78
99
84
107
112
76
72
77
54
103
59
56
53
SS
mg/l
300
160
280
220
340
310
275
190
173
115
288
230
86
90
Computed
BOD
mg/l
232
235
230
222
234
232
230
232
240
240
236
229
220
210
202
194
187
180
174
168
160
155
152
149
147
145
143
141
138
135
132
130
129
128
125
123
120
118
116
115
115
SS
mg/l
297
313
292
355
453
418
384
395
407
408
399
370
329
286
24?
212
180
153
130
112
100
96
94
94
95
97
96
95
94
93
91
89
86
83
78
73
67
61
56
51
47
243
-------
CATCHBASIN BOD « 400 mg/1
_J
•s
o
s
o
o
m
X
o
s
o
o
(O
FENDED
(O
z>
300 —
250 -
200 -
150 -
100 -
50 -
600 -
500 -
400 _
300 _
200 -
100 -
p~ jfy^^^
i ^^" "Q
/ ^
/ "Q
^ ^°"O--a.
•f^\*f\
9 9 0 — ~£—AA'~^
i i i i
1830 2030 2230 0030
\ - COMPUTED
\ s&
\ pS** \
\ f \&f\ *
\ / W 4^ V V- MEASURED
^ J^ \ ^k \
V^ j[ \ At
^-^^^^^^
T
1830 • 2030 2320 0030
TIME-HOUR OF DAY
Figure 72. Run Number 21 arriving quality Site I
-------
TABLE 81. ARRIVING QUALITY, SITE I
Run Number 21
Catchbasln BOD = 150 mg/1
ACTUAL
Time
hours
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
2330
2400
0030
BOD
mg/1
74
78
99
84
107
112
76
72
77
54
103
59
156
53
50
52
53
SS
mg/1
300
160
280
220
340
310
275
190
173
115
288
230
86
90
80
148
105
COMPUTED
BOD
mg/1
118
117
112
114
113
108
106
107
108
107
104
100
95
91
87
83
79
76
74
70
68
67
66 -
65
64
63
62
61
60
58
57
57
56
55
54
53
52
51
51
SS
mg/1
310
306
302
422
444
397
389
411
415
415
395
358
315
272
234
200
170
144
123
106
97
95
94
94
96
96
95
94
93
91
89
86
83
79
74
69
63
57
52
245
-------
_J
X
(D
Q
O
m
O
(O
O
LU
O
z
LU
D.
(0
D
tn
150-
125-
100 _
75-
50-
25-
CATCHBASIN BOD = 150 mg/1
COMPUTED
MEASURED
1830
2030
2230
0030
525-
450-
375-
300—
•
225-
150-
75-
\
COMPUTED
\
\ I
\ I
I
I
I
T
1830 2030 2230 0030
TIME-HOUR OF DAY
Figure 73. Run Number 21 arriving quality Site 1.
246
-------
TABLE 82. EFFLUENT QUALITY, SITE I
Run Number 21
Time
hours
1300
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
2330
2400
Actual
BOD
mg/1
54
49
46
37
29
31
37
34
31
37
38
32
32
25
21
36
28
25
31
Effluent
SS
mg/1
154
158
112
84
72
86
75
72
64
70
90
71
60
49
41
63
53
52
55
BOD
mg/1
49
49
48
46
47
47
47
45
46
46
46
45
43
41
39
37
35
33
32
30
29
28
28
27
27
26
26
26 '
25
25
24
24
24
23
23
22
22
1
Storage
SS
mg/1
50
57
56
55
91
123
142
139
146
147
147
141
130
116
102
89
77
63
50
40
34
31
30
31
32
34
34
34
34
33
32
30
28
26
23
, 20
17
247
-------
N.
o
5
o
o
m
o
S
v>
o
o
V)
o
LU
O
z
LU
Q.
tn
3
V)
175-
150-
125-
100-
75-
50-
25-
MEASURED
1900
I
2000
I
2100
T I
2200 2300
C— MEASURED
i r i I i
1900 ' 2000 2100 2200 2300
TIME-HOUR OF DAY
Figure 74. Run Number 21 efflluent quality Site I.
-------
o
o
-3-
CM
O
O
-------
TABLE 83. ARRIVING FLOW, SITE II
Run Number 21
Time
hours
1700
1730
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
2330
2400
Arriving
cu m/min
6.0
8.3
13.4
21.1
29.9
38.4
40.8
41.3
42.2
51.7
59.5
65.6
105.9
114.1
101.5
86.7
78.4
61.9
38.8
31.3
26.4
20.6
16.5
12.8
12.6
12.9
16.0
18.2
21.8
28.6
31.3
33.0
28.7
24.8
23.5
23.0
23.0
21.9
17.6
9.9
8.5
6.3
4.8
flow,
cfs
3.5
4.9
7.9
12.4
17.6
22.6
24.0
24.3
24.8
30.4
35.0
38.6
62,3
67.1
59.7
5KO
46.1
36.4
22.8
18,,4
15.5
12,, 1
9.7
7.5
7.4
7.6
9.4
10.7
12.8
16.8
18,4
18.4
16.9
14.6
13.8
13.5
13.5
12.9
10..4
5.8
5.0
3.7
2.8
Computed
cu m/min
0.0
0.0
7.8
19.7
34.0
60.7
74.8
88.4
94.7
98.6
105.4
143.1
170.0
181.9
169.3
105.6
109.7
71.9
50.2
27.9
15.0
1.9
1.7
0.0
0.0
2.2
3.4
7.7
10.7
15.5
20.1
24.3
23.8
23.1
21.4
14.6
7.3
3.6
3.4
1.0
0.0
0.0
0.0
flow,
cfs
0.0
0.0
4.6
11.6
20.0
35.7
44.0
52.0
55.7
58.0
62.0
84.2
100.0
107.1
99.6
62.1
64.5
42.3
29.5
16.4
8.8
1.1
1.0
0.0
0.0
1.3
2.0
4.5
6.3
9.1
11.8
14.3
14.0
13.6
12.6
8.6
4.3
2.1
2.0
0.6
0.0
0.0
0.0
250
-------
TABLE 8k. ARRIVING QUALITY, SITE II
Run Number 21
Tfme
hours
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
Actual
BOD
mg/1
79
33
35
30
21
18
21
20
SS
mg/1
705
188
105
110
120
140
133
72
Computed
BOD ~ SS
mg/1 mg/1
122
119
114
110
110
105
96
3k
91
82
73
69
66
59
56
55
53
51
49
48
48
50
48
46
45
44
44
44
42
41
40
890
486
285
280
369
470
510
506
843
488
541
534
521
560
555
496
435
392
314
221
173
145
120
88
76
67
69
78
82
92
106
251
-------
_J
X
ID
S
0
o
03
O
s
in
o
o
to
o
UJ
o
Z
Ul
Q.
(O
W>
105-
90-
75-
60 _
5~
30-
15-
700-
600-
500-
400
300-
200-
100-
^OOD^ COMPUTED
^? '
\
o
Q
o
Q_
X'
—MEASURED
\
,
Vbs^b
1800 1900 2000 2 loo 2200 2300 2^00
\ t
|
i
i <
1 A A1^
1 A '
A/ * '
! ^ ^
1 /
i /
i '
i /
4i
^.MEASURED
\
$ — COMPUTED
\
\
\
\
\
i
\
\
t ^
W^* ^
^^ V ^ AA'
180
.20*00 2*100 2200 2300 2«»00
TIME- HOUR OF DAY
Figure J6. Run Number 21 arriving quality Site II
252
-------
Run Number 25^
The next rainfall event that was used to evaluate the SWMM output was run
25 which occurred on April 18, 1974. The total rainfall of this storm was
0.66 centimeters (0.26 inches) over 185 minutes. Table E4 of Appendix VI-E
presents the rainfall intensities at five minute intervals for each site
during this run. There were four days prior to this run in which the cu-
mulative rainfall was less than 2.5^ centimeters (1.0 inch). Table 85
presents the actual arriving flow at the wetwel1 of Site I and the computed
flow for the same location. Figure 77 shows the graphical comparison of
these two flows. The actual flow starts almost one hour earlier than the
computed, and peaks 30 minutes later than the computed. The ratio of the
total actual flow arriving to the computed total flow Is 1.34. The close-
ness of fit of the two curves is generally fair. The reason the computed
flow peaks early could be due to the slopes of the conveyance system used
in the SWMM, or to the inaccuracies of the timing mechanisms of the rain
gauges. The input variable which describes the percent of the impervious
area with zero detention was increased to determine if this would bring
the peaks closer together. Because the remaining runs varied in the time
of arrival of the peak, some were early and other behind the actual, and
because run No. 21 was accurate, it was decided to leave the default value
of 25 percent.
The quality of the arriving flow was determined by nine samples taken over
the duration of the overflow and analyzed for BOD, suspended solids and
total coliforms. The number of total coliforms is expressed as number per
100 ml as determined by the membrane filter technique while the computed
value is expressed as MPN per 100 ml. Although the two methods are different,
it was found, by running both methods on the same samples, the two results
were within the same acceptable range. The remaining coliform analyses
were run using the membrane filter technique. Table 86 presents the actual
measured quality and the computed quality of the arriving flow at Site I.
The actual samples were taken as the overflow began (1500 hours) and at
various times throughout the overflow. Figure 78 presents the graphical
comparison of the BOD and suspended solids data. The measured BOD and
suspended solids are relatively constant at 40 and 150 mg/1 respectively.
The absence of a "first flush" could be due to the occurrence of over 2.54
centimeters (I inch) of rain four days earlier. The computed BOD is
reasonably close to the measured although a difference of 20 mg/1 is pre-
sent at the peak of the computed pollutograph. The computed suspended
solids pollutograph peaks well above the measured, but without this initial
peak the computed values are close to the actual measured concentrations.
Figure 79 presents the arriving coliform data which is very close in com-
parison. In this case, the computed arriving flow for this run is 30
minutes earlier than the actual, therefore if it were brought closer to
the actual, it might change the quality predictions which show acceptable
goodness of fit.
The effluent from the Site I treatment unit was also discretely sampled
during the overflow and the measured results along with the computed out-
put of the Storage block are shown in Table 87. No coliform analyses
253
-------
TABLE 85. ARRIVING FLOW, SITE I
Run Number 25
Time
hours
1500
1530
1600
1630
1700
1730
1800
1830
Arriving
cu m/min
mm.
0.9
1.5
0.9
K4
1.7
2.0
3.4
2.9
2.9
3.4
5.3
12.8
21.9
26.5
27-9
27.9
24,1
21.9
12.8
8.3
9.7
6.8
max.
1.7
2.2
1.2
1.7
2.4
2.9
3.7
6.0
6.0
6.8
8.3
15.1
22.6
29.6
31.8
31.8
26.5
23.5
15.1
11.6
14.3
9.0
flow,
cfs
min.
0.5
0.9
0.5
0.8
1.0
1.2
2.0
1.7
1.7
2.0
3.1
7.5
12.9
15.6
16.4
16.4
14.2
12.9
7.5
4.9
5.7
4.0
max.
1.0
1.3
0.7
1.0
1.4
1.7
2.2
3.5
3.5
4.0
4.9
8.9
13.3
17.4
18.7
18.7
15.6
13.8
8.9
6.8
8.4
5.3
Computed
cu m/min.
1.2
1.0
7.1
13.1
17-7
20.2
21.4
20.7
18.0
15.1
12.2
9.7
7.7
5.8
4.3
2.9
1.7
0.7
flow,
cfs
0.7
0.6
4.2
7.7
10.4
11.9
12.6
12.2
10.6
8.9
7.2
5.7
4.5
3.4
2.5
1.7
1.0
0.4
254
-------
v
m
LTV
Q "•&
U. "
Q g
UJ
i
on
->
o
X
0)
c
1.
u
• n
ir\
CM
0)
C
O)
IZ
A1ISN3.INI
MO"ld
255
-------
TABLE 86. ARRIVING QUALITY, SITE I
Run Number 2*)
Actual
Time
hours
BOD
mg/1
SS
mg/1
Coliforms
No./lOO ml
BOD
mg/1
Computed
SS
mg/l
Conforms
MPN/100 ml
1455
1500
1530
1600
1630
1700
1730
1800
1830
56
52
30
56
43
31
45
53
33
93
134
129
84
91
98
171
230
1.4x10
l.OxlO
2.6x10?
8.5x107
2.7x10?
2.4x10?
1.5xlO&
212 1.4x10c
50.9
77.0
76.0
65.8
56.3
46.6
38.8
33.1
29.4
27.0
25.5
25.0
444.0
190.5
166.8
182.3
186.7
175.6
150.9
123.4
99.1
79.9
64.2
58.2
1.4xlo5
5.1x10?
3.5x10?
2.3x10?
1.8x10?
1.9x10?
2.2x10?
2.7x10?
3.3x10°
4.1x10?
5.1x10?
6.1x10°
256
-------
o
Q
O
CO
O
en
o
o
Q
LU
o
**
ttl
CL
100-
75-
50-
25-
1500
600-
650-
300-
150 -
COMPUTED
MEASURED
1600*
*''
1700
1800
1900
COMPUTED
MEASURED
1500 1600 1700 1800
TIME-HOUR OF DAY
Figure 78. Run Number 25 arriving quality Site I.
1900
257
-------
a.
in
I
-j
8
10
5x1 Oy
8
10' --
_J
s
8 5x1O5
10° --
5x103 --
10P --
5x10
— MEASURED
COMPUTED
1300
1500
1700
1900
HOUR- TIME OF DAY
Figure 79. Run Number 25 arriving quality Slt4 I,
-------
TABLE 37. EFFLUENT QUALITY, SITE I
Run Number 25
Actual
Time
hours
150Q
1530
-
1600
1639
1700
1730
1800
1830
1900
BOD
£s£L
126
33
4Q
32
37
35
48
46
32
3Q
ss
mg/ 1
*"f"™«VP™"'
267
59
97
46
58
62
124
110 . •
81
55
Computed
BOD
yafl
26.9
23.4
21.0
19.4
18,5
17.1
27.9
19.9
17.4
14,5
11,7
9.3
7.3
5,8
4.6
3.8
3.4
3.1
3.0
3.0
3,0
ss
mg/l
176.3
165.9
155.4
148.3
143.6
134.6
57.8
57-5
70.8
77.3
76.5
70.0
60.6
49.9
38.5
28.9
20.0
12.9
8.0
5.9
4.5
259
-------
were performed on these samples. Figure 80 presents the graphical comparison
of these data. The first measured value of the effluent BOD and suspended
solids is extremely high because of the scouring of solids from the bottom
of the flotation tanks and the poor flotation at the very start of the
effluent discharge. These high values may be discarded since they are
higher in concentration than the influent throughout the dura-tion of the
overflow. To compare the remaining values, the influent quality must serve
as a function of the computed effluent quality. Thus, large variations
between the actual and computed influent values will make the comparison of
the effluent quality difficult. The measured BOD of the effluent is gener-
ally higher than the computed, but since the arriving BOD is generally less
than 50 mg/1, the treatment units could not remove a substantial part of the
incoming BOD. The arriving computed BOD was greater than the measured value,
while the computed effluent concentration was lower. The suspended solids
data follow the same general pattern with the computed effluent values being
lower in concentration than the measured values after being greater in
concentration in the arriving flow. Again the low suspended solids concen-
tration in the arriving flow could be the reason for the variations.^ In
summation, the overall computed effluent quality shows good correlation
throughout the overflow.
The actual and measured flows at Site II for run No. 25 are listed in Table
88. The computed values of the arriving flow are taken from two points in the
transport system of the SWMM. This procedure is used whenever the sluice
gate in element 146 of Figure 62 is closed. When this occurs, all flow in
the interceptor is routed to Site II and the computed arriving flow is
taken as element 87, not 112 as is the usual case. When, the gate opens,
the computed arriving flow is taken as element 112. The graphical comparison
of the arriving flow is shown in Figure 81. Both the computed and measured
hydrographs start and stop at the same time. Although there is no definite
peak in the computed flow, the highest computed value occurs at about the
same time as the measured peak. The area under each curve is very close
with the ratio of measured to the computed areas being 0.96. The closeness
of fit of the two hydrographs is poor even though the total flows arriving
are almost equal.
The arriving flow at Site II was sampled ten times during the duration of
the run. Table 89 lists the measured quality characteristics and the com-
puted values. Figure 82 presents the graphical comparison of the measured
and computed quality. The measured BOD and suspended solids are relatively
constant except for the large peak at 1715 hours. At this point in time
the corresponding hydrograph is also at a maximum. The computed values of
the BOD and suspended solids decrease from the start of the overflow and
do not show the peak which the measured values do. This peak could corres-
pond to a "flush" phenomenon, but it has not been seen in other overflow
events at this site., Figure 83 presents the arriving coliform data in
graphical form and again the computed and measured values are very close
throughout the overflow. Thus, the overall comparison of the arriving
quality for this run is generally poor in the suspended solids and BOD
constituents and very good with the col 5 form data. No discrete effluent
260
-------
120"
X
o
s
o
o
CD
Q
_j
o
-------
TABLE 88. ARRIVING FLOW, SITE II
Run Number 25
Time
hours
i4oo
1430
1500
1530
1600
1630
1700
1730
1800
Arriving
cu m/min
0
0
0
0
0
1.7
3.4
7.5
8.7
9.5
10.5
11.2
10.9
10.2
9.5
9.5
13.1
44.2
109.0
110.8
80.2
58.7
18.9
11.4
11.4
9.5
7.5
0
flow,
cfs
0
0
0
0
0
1.0
2.0
4.4
5.1
5.6
6.2
6.6
6.4
6.0
5.6
5.6
7.7
26.0
64.1
65.2
47.2
34.5
11.1
6.7
6.7
5.6
4.4
0
Computed
cu m/min
0
0
0
0
0
0
0
10.5
10.7
10.7
10.7
11.6
19.0
32.2
42.0
42.7
44.2
48.3
51.0
58.1
56.4
45.6
27.2
21.6
19.4
17.7
15.0
14.6
0
flow,
cfs
0
0
0
0
0
0
0
6.2
6.3
6.3
6.3
6.8
11.2
21.3
24.7
25.1
26.0
28.4
30.0
34.2
33.2
26.8
16.0
12.7
11.4
10.4
8.8
8.6
0
262
-------
I I
in •—
trv
CM
0)
-Q
Z3
C
tz.
CO
£
o>
tH
A1ISN31NI
wiw/wno Moid
263
-------
TABLE 89. ARRIVING QUALITY, SITE II
Run Number 25
Actual
Time
hours
BOD SS Conforms
mg/1 mg/1 Nb./lOO ml
Computed
BOD
mg/1
SS
mg/1
Col i forms
MPN/100 ml
1500
1530
1600
1630
1700
1730
1800
1830
57
42
62
50
42
43
58
104
45
235
233
142
156
143
134
195
293
546
6.8x10;
9.0x10.
1.2x10
2.5x10;
1.7x10°
8.0x10?
2.1x10°
162 1.7xlOc
77
75
72
70
58
49
45
39
39
25
20
17
21
16
104 <
101
161 l
321
352
247
239
260
244
236
225 :
218 :
207 !
166 !
i. 9x1 of}
7.3x10?
t.6xlO^
.9x10,.
.1x10°
.8x10?
.4x10°
.0x10?
.3x10°
.7x10°
J.3xlO?
J.SxlO?
5.0x10?
5.1x10°
264
-------
X
o
o
o
CO
V.
e>
2
(ft
Q
o
UJ
o
?z
UJ
0_
120-
80-
40-
o-a
•o-cx,
1500
1600
MEASURED
COMPUTED
-o-o-o-o.-o-o
1800
1960
MEASURED
1500 1600 1700 1800
TIME-HOUR OF DAY
Figure 82. Run Number 25 arriving quality Site H.
265
1900
-------
5x107 --
107 --
o
o
Z
CL
|06 _.
in
ft: 5xl o5
o
u.
o
o
o
105 --
1500
MEASURED
p-o
/
P" COMPUTED
1600 1700
1800
1900
HOUR- TIME OF DAY
Figure 83. Run Number 25 arriving quality Site II
266
-------
samples from this site were taken due to mechanical problems with the chemi-
cal pumps of the treatment units.
Run Number 26
The third rainfall event in which discrete sample data were obtained is
run Number 26 which occurred on April 28 and 29, 1974. The total rainfall
was 3.46 centimeters (1.36 inches) contributing to Site I and 4.2? centi-
meters (1.68 inches) to Site II over a duration of 275 minutes. Table E5,
Appendix VI-E presents the rainfall intensities at five minute intervals
for each site. Prior to this run there were fifteen days in which the cu-
mulative rainfall was less than 2.54 centimeters (1.0 inch) and seven days
in which no rain fell. Table 90 presents the measured arriving flow and
computed flows for Site I during this run. The measured arriving flow at
this site is shown as ending at 0800 hours but the actual flow continued to
arrive for^more than 16 hours after this point. Figure 84 shows the graphi-
cal comparison of the computed and measured data. Note that the computed
flow terminates shortly after 0400 hours. The large amount of rain could
have caused excessive surcharging along the transport system but not enough
to cause the more than 24 hours of overflow. The flow divider in subarea 18
was-investigated to determine; if its submergence was caused by the high levels
of Lake Michigan during the Spring of 1974. No mechanical problems were
found with this element and since the arriving flow at Site II did not
show this same long duration, the possible surcharging of the Augusta
Street subarea could be the cause. During this overflow event another
agency was conducting flow measurements in selected sewers in the area and
these measurements (40) for manhole 47 of the SWMM transport system are
shown in Table 91. Figure 85 graphically presents the computed flows
for this element during run Number 26. Note that the actual flows continue
at high levels until well after the computed flows and that the computed
flows surcharge from 0140 to 0320 hours at the full capacity of sewer
number^72. Without this surcharging the computed flow would peak at
approximately the same time as the measured flow,.. Thus, the goodness of fit
of the flows for this run are very poor and the ratio of the actual to
computed flow is less than 0.20.
The arriving quality for this run is not presented because of mechanical
problems with the automatic samplers. The effluent quality is available
and can be used to get a fair estimate of the overall quality predictive
capacity of the SWMM for this run. Table 92 presents the computed and
measured effluent quality for this run. The graphical comparison shown in
Figure 86 shows the computed BOD and suspended solids data to be in the
general range of the measured but the trends differ in the time the computed
values terminate and peak. The coliform data are graphical1y presented in
Figure 87. Again, the computed values are reasonably close to the measured.
The goodness of fit of the quality data is fair.
Table 93 lists the measured and computed arriving flow for run No. 26 at Site
II. The arriving flow ends shortly after 0600 hours. Figure 88 shows the
graphical comparison of the measured and computed flows which are reasona-
bly close throughout the duration of the run. The ratio of the total actual
2,67
-------
TABLE 90. ARRIVING FLOW, S!TE
Run Number 26
Time
hours
2350
2400
0030
0100
0130
0200
0230
0300
0330
0400
0430
0500
0530
0600
0630
0700
0730
0800
cu ml
min.
27.2
29.6
29.6
28.1
30.9
31.8
34.0
37.7
43.7
47-8
50.3
50.8
50.3
48.8
47.8
46.8
46.4
45.2
44.7
43-7
42.7
41.7
42.5
41.7
40.8
38.8
36.7
34.0
34.0
34.0
34.0
34.0
26.6
22.7
21.9
26.6
23.8
26.0
26.5
26.9
28.
28.
28.
28.
28.
28.
28.
28.
28.
26.5
Arriving
min
max.
30.3
32.3
32.3
33.3
34.9
34.9
38.6
45.6
51.7
55.8
58.3
60.5
58.3
56.8
55.8
55.3
54.4
53.2
52.3
51.0
50.1
49.6
50.5
49.6
48.8
46.8
44.7
42.0
42.0
42.0
42.0
42.0
30.3
28.1
25-7
26.5
27.5
30.3
30.3
30.9
31.8
32.
32.
32.
32.
32.
32.
32.
32.
30.3
flow,
cfs
min.
16.0
17.4
17.4
16.5
18.2
18.7
20.0
22.2
25.7
28.1
29.6
29-9
29.6
28.7
28.1
27.5
27.3
26.6
26.3
25-7
25.1
24.5
25.0
24.5
24.0
22.8
21.6
20.0
20.0
20.0
20.0
20.0
15.6
13.3
12.9
13-3
14.0
15-3
15-6
15.8
16.5
16.5
16.5
16.5
16.5
16.5
16.5
16.5
16.5
15.6
max.
17.8
19.0
19.0
19.6
20.5
20.5
22.7
26.8
30.4
32.8
34.3
35.6
34.3
33.4
32.8
32.5
32.0
31.3
31.0
30.0
29.8
29.2
29-7
29.2
28.7
27-5
26.3
24.7
24.7
24.7
24.7
24.7
17.8
16.5
15.1
15.6
16.2
17.8
17.8
18.2
18.7
18.9
18.9
18.9
18.9
18.9
18.9
18.9
18.9
17.8
Computed
cu m/min.
0.0
18.7
37.9
60.5
75.8
85.2
87.2
85.2
79.6
75.8
71-9
70.4
66.3
64.4
60.5
56.8
53.0
56.1
53.0
49.3
41.7
34.0
22.1
0.0
0
0
0
0
0
0
0
0
0
0
0
0
flow,
cfs
0.0
11.0
22.3
35.6
44.6
50.1
51.3
50.1
46.8
44.6
42.3
41.4
39.0
37.9
35.6
33.4
31.2
33.0
31.2
29.0
24.5
20.0
13.0
0.0
0
0
0
0
0
0
0
0
0
0
0
0
268
-------
o
•s
o
o
o
o
•s
o
o
-a-
o
o
o
U.
O
LU .£
O
4-1
V)
Ol
c
i
a:
1 o
o
o
o
o
o
O
O
CM
V.
U
ro
\o
CM
L.
(U
C
CO
C
A1ISN31NI
MOld
263
-------
TABLE 91. UPSTREAM FLOW MEASUREMENTS
Run Number 26
Time
hours
2230
2300
2330
2400
0030
0100
0130
0200
0230
0300
0330
0400
0430
Measured
cu m/min
4.1
11,6
11.6
19.0
48.8
91.6
113.2
52.7
46.8
42.7
38.9
21.1
19.3
Flow
cfs
2.4
6.8
6.8
11.2
28.7
53.9
66.6
31.0
27.5
25.1
22.9
35.9
32.8
Computed
cu m/mln
6.1
8.8
12.6
22.3
29.8
34.2
36.7
38.3
40.6
53.7
62.1
71.4
73.
73.
73.
73.
73.
73.
73.
73.
73.
73.
73
/ j •
78.2
65.0
53.4
45.6
19.9
17.2
16.7
Flow
cfs
3.6
5.2
7.4
13.1
17.5
20.1
21.6
22.5
23.9
31.6
36.5
42.0
43.0-
43.0
43.0
43.0
43.0
43.0
43.0
43.0
43.0
43.0
43.0
46.0
38.2
31.4
26.8
11.7
10.1
9.8
1
« 1
O)
ro
O
3
CO
270
-------
o
o
o
o
•a-
o
o
o
o
o
CM in
f
o
in
CM
A1ISN31NI
Moid
271
-------
TABLE 92. EFFLUENT QUALITY, SITE I
Run Number 26
Time
hours
2300
2330
2400
0030
0100
0130
0200
0230
0300
0330
0400
BOD
mg/1
238
103
69
127
68
63
90
70
Actual
SS Coliforms
mg/1 NO./100 ml
A
61 TNTC
(L
3^ 1.0x10
12 2 4xl05
c
13 l.SxlO3
c.
7 4.4xlOb
,
8 7.6x10
12 4.7xl03
8 2.0xl03
Computed
BOO
mg/I
0
39
36
37
32
27
22
33
34
32
28
24
23
22
20
17
14
10
8
5
4
3
2
1
1
SS
mg/1
0
27
75
142
114
123
117
64
50
68
83
94
169
210
218
207
184
150
115
84
66
42 ,
25
14
9
Col i forms
MPN/100 ml
l.Sxlo!
4.1x10,
1.4x10,
2.2x10,
2.0x10,
1.9x10,
9.4x10:?
7.5x10^
4.8x10^
3.1x10^
2.1x10;!
2.9x10^
1.5x10^
l.lxlOJ?
1.0x10?
8.6x10?
7.8x10?
7.6x10*
8.8x10^
7.7x107
7.4x107
... 8.9x10?
8.9x10?
7.0xlX>?
3.2x10
A Too numerous to count
272
-------
_J
;X
O
Q
O
CO
_J
X
O
V)
O
O
o
in
o
H
UJ
0.
(/)
D
V)
0100
0200
0300
0400
0500
300-
200-
100-
0100
0200
10300
0^00
0500
TIME-HOUR OF DAY
Figure 86. Run Number 26 effluent quality Site I.
273
-------
o
o
Q.
in
2
(£
O
U.
O
O
10 -
5x105--
5
5xl04-f-
10 --
5x10'
„• COMPUTED
MEASURED
—I—
0200
0600
HOUR- TIME OF DAY
Figure 87. Run Number 26 effluent quality Site I
-------
TABLE 93. ARFUVfMG FLOW, SITE If
Run Number 26
Time
hours
2230
2300
2330
2400
0030
0100
0130
0200
0230
0300
0330
0340
0400
0430
0500
0530
0660
Arriving
cu m/min
0
0
0
0
0
0
73.3
80.6
80.6
73.3
45.7
66.1
80.6
382.7
382.7
382.7
382.7
382.7
382.7
382.7
200.8
200.8
200.8
200.8
200.8
102.2
95.0
80.6
80.6
80.6
73.3
66.1
66.1
66.1
58.8
58.8
51.7
44.4
37.2
22.8
18.9
18.9
18.9
14.4
7.8
1.7
flow,
cfs
0
0
0
'. 0
0
0
43.1
47.4
47.4
43.1
26.9
38.9
47.4
225.
225.
225.
225.
225.
225.
225.1
118.1
118.1
118.1
118.1
118.1
60.1
55.9
47.4
47.4
47.4
43.1
. 38.9
38.9
,38.9
34.6
34.6
30.4
26.1
21.9
13.4
11.1
11.1
11.1
8.5
4.6
1.0
Computed
cu m/min
(
0
r
n
0
0
2.6
52.4
• 72.1
81.1
80.0
75.7
78.7
119.3
202.1
274.0
295.5
271.7
244.6
234.1
326.1
235.8
224.4
150.0
106.6
85.7
70.6
62.2
53.9
41.0
22.3
20.2
18.0
16.7
16.0
14.6
13.6
13.1
12.9
12.8
10.2
5.1
1.7
flow,
cfs
0
0
0
0
0
0
1.5
30.8
42.4
47.7
47.0
44.5
46.3
70.2
118.9
161.2
173-8
159.8
143.9
137.7
138.9
138.7
132.0
88.2
62.7
50.4
41.5
36.6
31.7
24.1
13.1
11.9
10.6
9.8
9.4
8.6
8.0
7.7
7.6
7-5
6.0
3.0
1.0
275
-------
^HWfffJ
n
r-.
\
LA
T
\D
o
o
VO
o
o
o
tf\
o
o
o
-3-
O
o O
•o
co ,.
o U_
o
4)
UJ
§ JZ
CM r-
0
>
•-
u
u
ro
SO
2
Ol
o
o
o
o
CNt CO
in \o
co
A1ISN31NI
TIVdNlVd
MOU
276
-------
flow to the total computed flow is 0.75. The goodness of fit Is excellent.
The arriving quality for run Number 26 Is.listed in.Table.94 and graphically
represented in Figures 89 and 90. The measured values show a definite
first flush effect at 0100 hours. The measured suspended solids peak at
1538 mg/1 while the computed values peak at over 600 mg/1. The computed
BOD values are very close to the measured with both showing peaks of about
60 mg/1. The overall comparison of the BOD data is very good while the
suspended solids data show only fair correlation because of the large
measured peak at 0100 hours. The computed coliform data closely parallels
the measured values as shown in Figure 90. The goodness of fit on the
quality data is acceptable.
The effluent quality data for Site I! are presented in Table 95. Figures 91
and 92 present these data in graphical form. The computed effluent BOD is
close to the measured now that the large peak after 0100 hours is dampened.
In the arriving quality the computed values were lower and this trend Is
carried through the Storage block output. The effluent suspended solids
pollutograph magnifies the differences of the arriving quality. Thus,
the computed suspended solids are generally much lower than the actual as
shown in Figure 91. The effluent coliform comparison of Figure 92 shows
little correlation which could be the result of over chlorination during
the operation of the treatment units. The overall goodness of fit of the
quality data is poor.
Run Number 2J
The next discretely sampled overflow event was run Number 27 which occurred
on May 5, 1974. The total rainfall contributing to Site I was 0.29 centi-
meters (O.I I inch) and 0.39 centimeters (0.16 inch) to Site II. There
were five days in which the cumulative rainfall was less than 2.54 centi-
meters (I inch) of rainfall. Table E6, Appendix VI-E lists the rainfall
intensities for each site at five minute intervals throughout the run. Be-
cause of the small amount of rain that was recorded, the computed flows
for Site I did not reach a high enough magnitude to pass over the over-
flow weirs contributing to Site I. Thus, in Table 96 and Figure 93 no
computed flow arrived. Figure 93 presents the measured arriving flow which
peaked briefly at 21.9 cu m/min (7.6 cfs). There are two possible reasons
for zero computed flow. One is the inaccuracy of the raingages and the
other is the difficulty in modeling the flow dividers upstream of Site I
(element numbers 37, 209, and 46). Data are avallable on the characteristics
of these structures but field inspection proved this data to be inaccurate.
High flows in these structures during dry weather make accurate measure-
ments difficult. The most likely possibility is that the raingauge measure-
ments at such low intensities are Inaccurate. The goodness of fit is there-
fore unacceptable.
During this run discrete samples of the influent and effluent were obtained.
Because there was no computed flow for Site I, it was decided to use the
quality of the flow above the main flow divider for Site I (Number 37) to
277
-------
TABLE 94. ARRIVING QUALITY, SITE II
Run Number 26
Actual
Time
hours
2400
0030
0100
0130
0200
0230
0300
0330
0400
0430
BOD
mg/1
32
58
20
16
21
14
16
5
ss
mg/1
427
1538
677
473
457
298
242
64
Col i forms
No./lOO ml
f
6.6xlOb
f.
8.3xlOb
e.
1.4xlOb
c
5.8xl03
3.4xl05
f
I.lxl0b
j™
9.5xl05
C
8.5xl05
BOD
mg/1
71
60
51
42
34
23
14
13
12
12
11
10
8
7
6
5
4
3
3
2
2
2
2
2
Computed
SS
mg/1
653
612
516
418
360
280
182
197
198
199
193
172
152
130
113
102
81
61
44
35
28
17
15
15
Col i forms
MPN/100 ml
3.5x10?
2.6x10:1
1.7x1o£
8.3x10?
2.8xlo£
1.5xlOb
8.3x10-?
2.7x10^
1.9xlo£
1.7xlo£
1.7x10-!
1.8x10^
1.7x10^
2.3x10^
2.7x10^
3.0x10^
3.2x10^
3.4x10^
4.0x10^
5.1x10^
5.9x10^
7.5x10^
1.0x10^
1.3x10^
278
-------
6
-------
-I
5
I07--
8 5xlo6--
0.
2
o:
o
u_
8
i
I06 --
5x105 --
I05«
5x10^--
MEASURED
COMPUTED
0100 0200 0300 OlfOO
HOUR- TIME OF DAY
0500
Figure 90. Run Number 26 arriving quality Site II.
280
-------
TABLE 95. EFFLUENT QUALITY, SITE I!
Run Number 26
Time
hours
0030
0100
0130
0200
0230
0300
0330
0400
0430
BOD
mg/1
\2
9
17
9
10
7
6
3
5
2
Actual
SS Col i forms
mg/1 No./lOO ml
62
93
493
321
A
479 <9
i.
279 8.2x10^
160 6.5x10
, D
94 30B
92 40B
41 <9A
Computed
BOD
mg/1
16
11
10
10
10
9
8
7
6
5
4
3
2
2
1
1
1
1
SS
mg/1
72
84
125
127
147
137
131
H7
104
91
74
41
23
14
8
7
7
6
Col i forms
MPN/100 ml
1.3x10^
7.1xlof
9.9x10^
1.1x10^
9.6x10^
8.6xloJ
7.*2xloJ
7.1x107
£1
4.6x10?
1.2x10,
1.6x10,
1.9x10,
2.2x10,
2.5x10^
A. No growth
B. Less than 20 colonies
281
-------
;J
0
2
o
o
CO
30-
25-
20-
15-
10-
-J
V.
to
Q
o
uj
o
z
UJ
Q.
(O
tn
5-
600-
• MEASURED
"b7
00
200-
02oT 0300 O^'OO
TIME-HOUR OF DAY
Figure 9t. Run Number 26 effluent quality Site M
01'OO
0500
282
-------
s 10-
O L
2 5x10
X
z
Q_
in
O
UL
O
u
IS
O
10
5x10
3..
5x10-
2400 0100 0200 0300 0400
HOUR- TIME OF DAY
Figure 92. Run Number 26 effluent quality Site H,
0500
283
-------
o
o
0>
Q J2
U.
O
LU r
en
c
">
tmm
L.
l_
TO
o
o
o
o
LU
o.
o
o
o
tr
o
I
E
C
C£.
CTl
£
3
LU
o
o
I
Cvl
CO
01 Lf\
A1ISN31NI
MOld
28%
-------
TABLE 96, ARRIVING FLOW, SITE I
Run Number 27
Arriving flow,
Time
hours
101 5
1030
1045
1100
1115
1130
cu
min.
0
1.4
3.2
4.6
6.6
7.7
10.2
12.9
11.2
10.5
9.2
5.3
4.6
3-9
2.7
1.5
0.0
m/min
max.
0
2.4
5.4
7.8
11.2
13-0
17.3
21.9
19.0
17.8
15.6
9.0
7.8
6.6
4.6
2.5
0.0
mm.
0
0.6
1.3
1.5
3.2
3.6
5.0
6.9
6.0
5.4
4.5
3-1
2.1
1.7
1.3
0.7
0
cfs
max.
0
0.8
1.9
2.7
3.9
4.5
6.0
7.6
6.6
6.2
5.4
3.9
2.7
2.3
1.6
0.9
0
Computed
cu m/min.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
flow,
cfs
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
285
-------
compare with the actual measured data. This procedure is acceptable because
the flows in excess of the flow divider are normally transported this short
distance to the Site I wetwell without any other inflows. This procedure
allows full use of the available discrete data. Table 97 lists the actual
and computed quality for this run while Figures 9^ and 95 present a graphi-
cal comparison of the data. The upstream computed BOD remains relatively
constant at 100 mg/1 which is-near the dry weather flow average v/hile the
measured values decline from 100 mg/1 to less than 25 mg/1 in the hour of
sampling. The suspended solids comparison is very close throughout the
duration of the run both in magnitude and trend. The col?form data of
Figure 95 are close in magnitude, but the measured values decrease after
the initial peak is reached while the computed values remain relatively
constant. The goodness of fit of the quality data is acceptable.
The problem with zero computed flow at Site I was not found for the flows
at Site II as listed in Table 98 and graphed in Figure 96. The measured
flow peaks above 22 cubic meters per minute (13 cfs), because during this
small rainfall event the sluice gate was down for the duration of the run.
This negated the dampening effects of the flow divider (element No. 87). Gen-
erally, the computed flow is much higher than the measured flow throughout the
duration of the run. The overall predictive capacity of the SWMM and good-
ness of fit is only fair for this run. The ratio of the total actual
flow to the computed flow is 0.5^.
Table 99 lists the quality of the arriving flow for this run at Site II.
Figures 97 and 98 present the graphical comparison of the measured and
computed data. The BOD data shows only fair correlation and the computed
remains relatively constant at 70 mg/1. The suspended solids data corre-
late very well and the computed values follow the same trend as the
measured. The coliform data are within the same range but the computed
values do not follow the measured trend. The goodness of fit of the
arriving quality data is acceptable.
Table 100 lists the measured and computed values of effluent quality at
Site II during run 27. Figures 99 and 100 present the graphical
comparison of this data. The BOD data is now close in magnitude because
the peaks of the arriving pollutograph are dampened through the treatment
units. The suspended solids data show that the Storage block removed
more suspended solids than was actually measured. This removal might be
due to the low concentration of the measured arriving flows. The coliform data
show good correlation. The goodness of fit of the effluent quality data is
acceptable based on the difference of the influent concentrations.
Run Number 30
The fifth discretely sampled run was Number 30 which occurred on May 13,
197^4. The average total rainfall for the area was 0.59 centimeters (0.23
inches) with a duration of 60 minutes. There were two days prior
to this run when no rain fell and six days when the cumulative rain-
fall was less than 2.5k centimeters (I inch). Table E7, Appendix VI-E,
286
-------
TABLE 97. ARRIVING DUALITY, SITE I
Run Number 27
Time
hours
1000
1015
1030
1045
1 100
1115
1130
1145
Actual
BOD
mg/1
114
114
86
54
38
31
29
17
SS
mg/1
261
215
201
121
81
66
42
60
Col i forms
No./lOO ml
7
1.0x10'
7
2.1x10'
f
8.5xlOb
f.
7.5x10°
A
2.7xlOb
(L
1.8xlOb
6
1.7x10
e.
1.4xl06
Upstream Computed
BOD
mg/1
113
1.06
104
104
103
103
106
109
115
119
120
121
120
119
117
115
112
110
108
106
104
102
100
SS
mg/1
223
224
275
306
317
300
256
199
158
132
117
108
100
94
89
84
80
76
73
69
66
64
61
Col i forms
MPN/100 ml
3.5x10?,
2.7x10'
2. 3x1 0;
2.3xlO_
2.2x10;
2.3x10^
2.6x10^
2.9x1o(
3.2x10^
3.1x10^
2.9x10^
2.7x10^
2,5x10;
2.3x104
2.1x10^
2.0x10;
1.8x10'
i.7xio;
1.6x10^
1*6x1 Q'
1.5x10;
I. 4x10;
1.2xTG'
287
-------
V.
o
S
O
o
CO
X
o
2
tn
B
o
to
o
iu
o
LU
Q.
z>
in
150
125
100
50
25
300 _
250 _
200 _
150 _
100 _
50 _
COMPUTED
1000
1030
1100
1130
A
/ s COMPUTED
/ X
/ 4^
MEASURED
T
T
1000 -1030 1100 1130
TIME- HOUR OF DAY
Figure 94. Run Number 27 arriving quality Site I.
1200
1200
288
-------
o
o
X
a.
in
2
ce
o
u.
o
o
e
10
5x10
8
10
7.
5x10
6.
10
6-
5x10
5_
I ' 1 1
1000 1100
HOUR- TiME OF DAY
Figure 95. Run Number 27 arrfvfng quaHty.STte F.
—I—
1200
289
-------
TABLE 98. ARRIVING FLOW, Slit II
Run Number 27
Time
hours
05100
H930
1000
1030
1100
1130
1200
1230
Arriving flow,
cu m/min
-
0.
0.8
1.7
22.7
19.0
22.7
19.0
17.0
7.6
5.6
5.6
1.8
0
cts
0
0.5
1.0
13.4
11.2
13.4
11.2
10.0
4.5
3-3
3.3
1.1
0
Computed
cu m/min
0
3.0
3.5
4.7
6.3
8.6
11.6
20.2
32.3
38.4
35.5
27.0
21.0
16.6
15.0
13-4
12.0
11.2
10.3
9.7
9.1
8.8
8.3
flow,
cfs
0
1.8
2.1
2.8
3-7
5.1
6.8
11.9
19.0
22.6
20.9
15,9
12.4
9.8
8.8
7.9
7.1
6.6
6.1
5.7
5.4
5.2
4.9
290
-------
o
o
CM
O
O
o
o
o
o
LU
I
01
o
I
-------
r
TABLE 99. ARRIVING QUALITY, SITE I I
Run Number 27
Actual
Time BOD SS Conforms
hours mg/1 mg/1 No. /1 00 ml
0900
0915
0930
0945 ,
80 412 2.0x10
1000
-»
1015 33 105 8.1x10'
1030 43 151 4.4xl07
1045
•t
77 148 3.4x10'
1100
1115 87 64 1.3xl07
1130
1145
BOD
mg/1
64
63
68
69
67
71
72
75
79
78
76
75
77
80
80
79
77
Computed
SS
mg/1
1108
1320
1080
634
300
172
142
198
268
337
353
289
212
148
106
81
66
Col t forms-
MPN/100 ml
2.8 x 107
2.6xl07
7
2.4x10'
2.0xl07
1.6xl07
7
1.3x10'
9.1xl06
5.7xl06
6
3.7x10°
3.3xl06
3.9xl06
f.
7.4xlOb
^
l.OxlO7
1.2xl07
•J
1.3X107
1.3xl07
1.3x107
292
-------
-J
X 100-
o
80-
0
CD 60 -
40.
20-
X
2 700 -
600.
to
0
-J 500 -
O
O
IU
0 300.
UJ
8> 20°-
01 100 -
COMPUTED
• X0-o~0-o^~0"°-0
n-O- ^O\P"°' ~Of^
0-0' \ /
\ J MEASURED
i i i i —
0900 1000 1100 1200
A
i
1
i
l
t
i
\\ COMPUTED
l \ A*A
4\ f V
\ \ '
V^^ \.XAN
MEASURED Jk ^&
1 1 1 1
0900 1000 1100 1200
TIME-HOUR OF DAY
Figure 97« Run Number 27 arriving quality Site U,
293
-------
10I
5Xl(r .-
O
o
Q.
5
5
o:
£
O
u
£
g 5X105
5X10 4-
•-OH
COWUTED
MEASURED
0900
1000
1100
HOUR- TIME OF DAY
Figure 98. Run Number 27 arriving quality Site II.
23k
-------
TABLE 100. EFFLUENT QUALITY, SITE I I
Run Number 27
Time
hours
1000
1015
1030
1045
1100
1115
1130
Actual
BOD SS Conforms
mg/1 mg/1 No./lOO ml
o
39 109 8.0x10^
iiO 99 1.2x10
r
37 99 2.8xl04
i
36- 87 2.2x10^
,
32 75 8.2xl04
,
Computed
BOO
mg/1
36
40
36
36
32
32
32
31
31
30
28
28
27
26 '
26
26
24
SS
mg/1
140
92
65
48
41
49
56
62
64
64
64
60
61
57
57
56
54
Col i forms
MPN/100 ml
6.7x10:*
6.0x10^
5.2x10^
4.2x10^
4.2x10^
3.6x10^
3.2x10^
3.1x10^
3.2x10^
3.9x10^
4.8x10^
6.2x10^
8. 2x1 Of
2.6x10?
2.6x10?
2.7x10?
2.4xlOH
295
-------
o
o
CO
to
o
o
UJ
Q
z
UJ
Q.
40 -
20 .
10 .
1*0-
100-
80-
60-
40.
20-
MEASURED
COMPUTED
1000
1100
tIEASURED
COMPUTED
1000
1100
TIME- HOUR OF DAY
Figure 99. Run Number 27 effluent quality Site U
296
-------
10
.5 ..
o
o
X
0.
5
io --
1 5X103
O
u.
O
U
104-
5X10 --
MEASURED
p-o-o-o
COMPUTED
1000 1100 1200
HOUR- TIME OF DAY
Figure 100. Run Number 27 effluent qualfty Site H.
297
-------
TABLE 101. ARRIVING FLOW, SITE I
Run Number 30
Time
hours
1200
1230
1300
1330
1400
1430
Arriving
cu m/min
mm
0.0
0.0
0.0
0.0
17.3
21.9
28.7
31.8
31.8
20.4
12.1
9.0
5.8
4.4
3.9
3.1
max.
0.0
0.0
0.0
0.0
25.7
28.1
34.0
36.4
35.7
25.7
16.7
13.6
9.9
8.3
7.8
6.8
flow,
cfs
mm
0.0
0.0
0.0
0.0
10.2
12.9
16.9
18.7
18.7
12.0
7.1
5.3
3.4
2.6
2.3
1.8
max.
0.0
0.0
0.0
0.0
15.1
16.5
20.0
21.4
21.0
15.1
9.8
8.0
5.8
4.9
4.6
4.0
Computed
cu m/min
0.0
0.0
0.0
1.7
9-5
8.8
6.8
20.7
29.4
26.0
31.5
28.4
26.0
27.0
19.7
9.0
3.4
0.0
flow,
cfs
0.0
0.0
O.Q
1 .0
5.6
5.2
4.0
12.2
17.3
15.3
18.5
16.7
15.3
15.9
11.6
5.3
2.0
0,0
298
-------
lists the rainfall intensities for each site at 5-minute intervals through-
out the run. Table 101 lists the computed and measured values of the arriv-
ing flows to Site I. Figure 101 presents a graphical comparison of the flows
which shows the computed flows to lag behind the measured by about 30
minutes. The ratio of the actual to the computed total flow flow arriving is
1.05. Because of the time difference between the peaks, the closeness of fit
of the two hydrographs is acceptable, but could be very good without the
lag.
The arriving quality of the flow at Site I was determined by 10 samples
of the flow taken at 10-minute intervals throughout the run. Table 102 lists
the actual and computed quality values and Figures 102 and 103 present the
graphical comparison. The BOD values are close throughout the run although
the "first flush" value from the measured graph is much higher than the
computed. The suspended solids data are close throughout the duration and
the initial computed and measured values are very close. The coliform
comparison in Figure 103 are generally close except for the large drop in
the actual values at 1330 hours. Although the coliform pollutograph may
appear to be out of phase because of these drops, such is not the case be-
cause the other pol1utographs for this site are in phase. The goodness of
fit of the quality data is acceptable.
The effluent quality from the treatment unit at Site I and the computed
effluent quality are listed in Table 103 and graphically compared in
Figures \Qk and 105. The BOD comparison maintains the trend shown in the
arriving quality, in that the actual values are higher in concentration
than the computed. The initial high peak shown by the measured values at
1300 hours relates to the first flush concentration of the influent and
the fact that poor removals are experienced from the treatment units
during the first few minutes of operation. The Storage block has simulated
the BOD removals relatively well for the remaining values. The effluent
suspended solids comparison is very close throughout the run. The high
concentrations of the influent values (700 mg/1) are reduced to 150 mg/1
range at the start of the effluent values. The remainder of the run is very
close showing good correlation between the computed and actual values. The
coliform comparison shown in Figure 105 shows little similarity between
the computed and the actual. But the low concentrations shown at the start
of the computed values may be neglected since the arriving computed flow
is very low at this same time. When the computed flow to the treatment
units increases, the concentrations stabilize at about 10^ MPN/IOO ml. The
actual coliform numbers show the opposite tendency. Initially the measured
values are very high and then stabilize. The high values may be traced to
poor chlorination at the start of the treatment unit operation or from the
discharge of the flotation tanks to the effluent channel of the initial
non-chlorinated flows. After these modifications are taken into considera-
tion, the results maintain the small differences shown in the arriving
quality and the overall goodness of fit is acceptable.
The actual and computed flows arriving at Site II during this run are
299
-------
\
I I I
Jf !•>.•—
vo ^^ o*^
• • *
o — .—
^ Q
U.
O
UJ
en
<0
O
0)
JD
t£ §
I
15
o
X
a
CtL
0)
3.
On
co
A1ISN31NI
MOld
300
-------
TABLE 102. ARRIVING QUALITY, SITE 1
Run Number 30
Actual
Time
hours
1230
1300
1330
1400
1430
1500
BOD
mg/1
>225
137
145
111
119
103
72
66
66
67
SS
mg/1
725
458
400
340
390
375
214
181
146
127
Col 1 forms
No./lOO ml
7
2.0x10'
1.4xlOi
6.4x10?
6.5x10?
3.1x10?
5.0x10^
2.5x10?
1.2x10?
2.8x10?
5.0x10
BOO
mg/1
137
104
96
85
99
95
72
76
69
62
56
64
62
60
Computed
SS
mg/1
347
726
1179
681
353
355
308
358
343
319
291
123
81
43
Col i forms
MPN/100 ml
5.0xl07
1.2xlo'
4.1x10-:?
1.3xl07
1.8xl07
1.5xlOi
9.6x10
8.1x10^
5.0x10?
3. 9x1 0,
3.5x10?
2.0x10°
1.6x10?
1.4xlOb
301
-------
Q
O
CD
Q
_j
o
o
LU
Q
z
UJ
Q.
(A
200-
150-
100-
50-
MEASURED
COMPUTED
I » I
1240 1300 1320
I ' I
13^0 1400 1420
800-
600-
400-
200-
COMPUTED
-&-.
1240
r
1340
1300 1320 1340 1400
TIME-HOUR OF DAY
Figure 102. Run Number 30 arriving quality Stte f.
1420
302
-------
OL
S
to
O
LL
O
O
10 --
5x10' __
10' --
-J
S
8 5x1 O6
10 --
SxlO3 --
5x10 --
"Q
1230
T
1300
1^00
HOUR- TIME OF DAY
Figure 103. Run Number 30 arriving quglity Site I.
303
-------
TABLE 103. EFFLUENT QUALITY, SITE I
Run Number 30
Actual
Time
hours
1240
50
1300
10
20
30
40
50
1400
10
20
30
40
50
1500
10
20
30
BOD
mg/1
144
96
90
62
55
51
59
53
50
48
SS
mg/1
58
263
259
195
171
164
144
136
126
105
Col i forms
No./lOO ml
8.9xl05
1.5x10
4.2xl03
2.0x103
t
7.7x10*
1.3X101*
3.3xl03
l.lxlO3
l.lxlO3
1.2xl03
BOD
mg/1
31
36
35
28
—
44
41
37
34
31
28
26
23
21
18
17
16
15
Computed
SS
mg/1
71
128
160
138
—
45
44
43
41
38
35
31
27
26
24
23
20
19
Col i forms
MPN/100 ml
3.9X101
4.0xl02
f\
2.6x10
2
2.7x10
i
2.1x10
j.
1.3x10
9.4xl03
o
7.1x10^
5.6xl03
4.6xl03
4.0xl03
3
3.2x10J
7
2.8x10^
2.5*10
2.3xl03
2.1x103
304
-------
X
o
s
Q
O
CO
-J
X
to
Q
O
UJ
O
z
111
Q.
150 _
100 -
50 -
150 -
100 -
50 -
COMPUTED
1300 1320 1340 1400 1420 1440
COMPUTED
1300
1320
1340 1400 1420
TIME-HOUR OF DAY
Figure 104, Run Number 30 effluent quality Site. U
1440
3P5
-------
10
Sxicr--
10*4-
»
*J
s
o
0
z
0_
s
2
cc
o
u_
-J
0
U
_J
o
4
5xl03
103
5xl02
io2
SxlO1
MEASURED
COMPUTED
1 j 1 1 T"
1200 1300 1400 1500 1600
HOUR- TIME OF DAY
Figure 105. Run Number 30 effluent quality Site I.
306
-------
listed in
the smal1
faci1ities at as
after the start of
Table 104 and graphically represented in Figure 106. Because of
amount of rain and because it was desired to use the Site II
high a flow as possible, the sluice gate was closed shortly
the overflow. The computed values of the flow peak before
the measured and at a slightly higher flow.
The ratio of the actual total flow arriving to the completed total flow
is 1.13- The slightly longer duration of the actual flow results from the
sluice gate being closed. The goodness of fit of the hydrograph is
acceptable.
The arriving quality at Site H is determined by seven discrete samples taken
at various intervals throughout the run. Table 105 lists the computed
and actual concentrations for this run and Figures 107 and JOtf present the
graphical comparisons of these values. The BOD comparison in Figure 107
shows the "first flush" of the measured values while the computed .concen-
trations remain relatively constant. The computed concentrations do not
start until 1240 and no "first flush" was found. The suspended solids
comparisons also show high initial measured values and a peak in the
computed values. The goodness of fit of the suspended solids data is
generally poor because of the computed peak at 1300 hours. The coll form
concentrations in Figure 108 show that the computed values are lower
throughout the run and show a large decrease at 1300 hours. Generally,
the coliform comparisons are poor and the overall goodness of fit of the
quality data is fair.
The effluent concentrations of BOD and suspended solids are listed in
Table 106. Because of chlorination problems at Site II, no coliform data
were obtained. Figure 109 presents the graphical comparison of the effluent
quality. The BOD comparison shows the measured values to be generally
constant at 20 mg/1 while the computed values are higher with some
variation. Because of the low effluent concentrations, the variations of
less than 10 mg/1 are Insignificant. The suspended solids data shows very
little correlation and even the trend of each Is different. The computed
influent suspended solids concentrations were well above 600 mg/1 with
the measured values at 300 mg/1. The same trend is found in the effluent
concentrations with no difference in the closeness of fit. Because of
the large differences in the arriving quality, little if any estimation
of the Storage block effectiveness is possible. The goodness of fit of
the effluent quality is poor.
Run Number 37
The next discretely sampled run was Number 37 which occurred on June 6, 1974.
The average rainfall from the three raingauges of the area was 1.62 centi-
meters (0.64 inches) with a duration of 330 minutes. There was one dry day
prior to this run and 16 days in which the cumulative rainfall was less
than 2.54 centimeters (I inch). Table E8, Appendix VI-E, lists the rain-
fall intensities used for this run. The computed and measured arriving
flows for Site I are listed In Table !07 and graphically presented in
307
-------
:i£v"V;:v ' , I_
£ j;.;;..;:?
tWiilM.!.^ I
VD CM
-------
TABLE 104. ARRIVING FLOW, SITE II
Run Number 30
Time
hours
1145
1200
1215
1230
1245
1300
1315
1330
13^5
1400
1415
1430
Arriving
cu m/min
0.0
0.0
0.0
27.7
34.9
48.8
34.9
27.7
20.9
19.2
12.2
10.4
7.0
3.4
2.7
1.7
1.7
flow,
cfs
0.0
0.0
0.0
16.3
20.5
28.7
20.5
16.3
12.3
M.3
7.2
6.1
4.1
2.0
1.6
1.0
1.0
Computed
cu m/min
0.0
0.9
15.3
34.5
63.6
63.8
54.9
^3.9
34.9
28.1
22.1
14.5
8.7
4.6
3.4
3.2
2.9
2.8
2.4
2.4
2.2
2.0
1.7
flow,
cfs
0.0
0.5
9.0
20.3
37.4
37.5
32.3
25.8
20.5
16.5
13-0
8.5
5.1
2.7
2.0
.9
.7
.6
.4
.4
.3
.2
.0
309
-------
0
2
Q
O
03
a
>j
o
o
Ui
a
z
UJ
a.
10
300 _
200 -
100 -
1200-
900-
600-
300-
1200
COMPUTED
./>-
•~o
MEASURED
--.0—O^ /
-0--O—°~-o«--0
- 1-^ - \ -- 1
1ZOO 1230 1300 1330
1^00 UJ30
COMPUTED
MEASURED
T
T
1230 . 1300 1330 1400
TIME-HOUR OF DAY
Figure 107. Run Number 30 arriving quality Site M
1430
310
-------
TABLE 105. ARRIVING QUALITY, SITE I I
Run Number 30
Time
hours
1200
1230
1300
1330
1400
1430
Actual
BOD
mg/1
328
86
65
57
119
66
40
SS
mg/1
1169
496
310
220
265
198
156
Col i forms
No./lOO ml
7
2.1x10'
2.2x107
7
2.1x10'
2.7xl07
1.3x107
2.8x10'
l.SxlO7
BOO
mg/1
80
70
100
89
80
80
65
68
64
61
58
Computed
SS
mg/1
695
608
969
896
661
496
352
351
338
325
310
Col i forms
MPN/100 ml
7.3x10^
1.7x10?
1.1x10?
1.8x10?
5.9x10?
7.1x10?
5.3x10?
4.9x10°
4.0x10^
3.8xl06
311
-------
•
_J
s
8 5X1 07.
o
f\
Q.
2
io7-
2 cYln6
Q* P A 1 U •
o
u.
j
8
i '°6'
MEASURED
• • — • ^^v^^ / >^
°\ 0"'°^
^ ; >°"''O<~^«
\ / ~o — o — <•)
\ / °
\ /
AV /
°\ /° COMPUTED
.
1 , , -, 1 • i —
1200 1300 1^00
HOUR- TIME OF DAY
Figure 108. Run Number 30 arriving quality Site II
312
-------
TABLE 106. EFFLUENT QUALITY, SITE II
Run Number 30
Time
hours
1200
1215
1230
1245
1300
1315
1330
1345
1400
1415
1430
1435
BOD
mg/1
19
20
17
16
16
16
18
23
Actual
SS Conforms3
mg/1 No./lOO ml
339
102
97
101
57
62
66
53
BOD
mg/1
28
35
30
39
42
37
37
33
33
31
32
29
26
27
30
27
26
24
24
22
Computed
SS Coliforms3
mg/1 MPN/100 ml
10
18
28
87
133
163
173
170
171
170
170
146
138
141
141
143
147
151
150
91
No Coliform Data, Chlorination at Site II Inoperative.
313
-------
o
Q
O
CD
40 -
30-
20 .
10 -
1200
N.
O
2
10
O
in
o
z
IU
Q.
(A
300-
200,
100,
COMPUTED
MEASURED
1300
1400
1500
MEASURED
COMPUTED
1200 1300 1400 1500
TIME-HOUR OF DAY
Figure 109. Run Number 30 effluent quality Site II
-------
TABLE 107. ARRIVING FLOW, SITE !
Run Number 37
Time
hours
1700
1730
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
2330
2400
Arriving
cu m/min
min.
0.0
0.0
10. 4
43.2
42.8
40.8
37.
37.
37.
37.
37.
37.
24.1
20.7
19.7
17.0
16.0
13.6
12.1
10.7
9.2
8.3
8.0
7-3
6.8
6.5
5.8
5.3
7.1
9.2
ll.it
11.9
12.4
12.8
12.4
12.1
12.1
10.5
9.2
7.8
8.0
8.2
8.3
max.
0.0
0.0
10.4
46.2
47.4
49.1
45.4
45.4
45.4
45.4
45.4
45.4
27.2
25.5
23.8
22.1
22.1
20.1
16.7
15.5
13.9
12.1
11.9
11.1
9.9
9.4
8.8
8.3
11.4
14.3
15.1
14.3
16.8
17.3
16.7
15.6
15.1
13.9
12.6
11.4
11.6
11.7
12.1
flow,
cfs
min.
0.0
0.0
6.1
25.4 ,
25.2
24.0
21.8
21.8
21.6
21.8
21.8
21.8
14.2
12.2
11.6
10.0
9.4
8.0
7.1
6.3
5.4
4.9
4.7
4.3
4.0
3.8
3.4
3.1
4.2
5.4
6.7
7.0
7.3
7.5
7.3
7.1
7.1
6.2
5.4
4.6
4.7
4.8
4.9
max.
0.0
0.0
6.1
27.2
27.9
28.9
26.7
26.7
26.7
26.7
26.7
26.7
16.0
15.0
14.0
13.0
13.0
11.8
9.8
9.1
8.2
7.1
7.0
6.5
5.8
5.5
5.2
4.9
6.7
8.4
8.9
8.4
9.9
10.2
9.8
9.2
8.9
8.2
7.4
6.7
6.8
6.9
7-1
Compu ted
cu m/min.
0.0
0.0
0.0
4.1
17.0
11.2
14.3
18.4
38.1
51.3
56.1
56.6
53.9
50.5
47.3
44.5
42.8
42.3
43.2
43.7
44.0
43.0
41.7
37.1
31.3
25.7
20.9
17.3
14.6
12.8
11.7
11.4
11.7
12.4
12.2
10.9
9.2
7-3
5.6
3.9
2.6
1.7
0.0
flow,
cfs
0.0
0.0
0.0
2.4
10.0
6.6
8.4
10.8
22.4
30.2
33.0
33-3
31.7
29.7
27.8
26.2
25.2
24.9
25.4
25.7
25.9
25.3
24.5
21.8
18.4
15.1
12.3
10.2
8.6
7.5
6.9
6.7
6.9
7.3
7.2
6.4
5.4
4.3
3.3
2.3
1.5
1 .0
0.0
315
-------
Figure 110. This figure shows that the computed flow again lags behind
the measured by about 30 minutes. The computed flows also remain higher
after the initial peak is reached., po'ssibly because of the surchargi ng .of
some elements within the transport network. The ratio of the total measured
arriving flow to the total computed flow is 1.3- The goodness of fit of
the two hydrographs is fair.
The computed and arriving quality is listed in Table 108 and graphically
represented in Figures I I I and 112. As Figure III indicates, the measured
BOD and suspended solids concentrations increase as the run progresses
while the computed values decrease. Since there was only one dry day prior
to this run, the computed BOD values are very low initially and decrease
rapidly to near zero. The suspended solids data are initially high and
again decrease to near zero. There is little, if any correlation between
the computed and actual values.
The coliform data presented in Figure 112 shows the computed coliform
concentrations to be low initially and then stabilize at higher levels.
The measured values show just the opposite trend with much variation
between values. The overall goodness of fit of the arriving quality
graphs is poor because of the differences in trend prediction.
The effluent values for Site I during this run are listed in Table 109 and
graphed in Figures 113 and II1*. Because of the large differences in the
actual and computed arriving BOD, suspended solids and coliform concentra-
tions, these differences are carried through the Storage block and make
any analysis of the output difficult. The coliform data in Figure 115
shows little correlation or similarity.
During the operation of the Site II treatment unit, the recording pen
of the plant flow chart did not operate for the first hour of operation.
Since the sites start-up automatically, the pen was not fixed until
operating personnel arrived after one hour of the overflow had passed, but
samples of the incoming flow were automatically obtained. Table 110
lists the computed and actual flows for this run and Figure 115 presents
the graphical comparison. No evaluation of the arriving flow is possible
because of the mechanical problems but the quality data is useful.
Table III lists the computed and actual concentration of the arriving
flow at Site II and Figures 116 and II? present this data in graphical
form. Note that no computed values are presented after 1900 hours since
the computed flow at this time was approaching zero. The BOD values
are relatively accurate in predicting the arriving quality while the
suspended solids concentrations are variable and show little correlation.
The coliform data presented in Figure II? shows the computed values to
be one order of magnitude lower than the measured values. This difference
is not due to the methods of analysis between the computed and the actual
coliform numbers inasmuch as the two methods (membrane filter and MPN)
were not found to differ appreciably during this run. The two curves are
very similar in trend and show good correlation. The overall goodness of
316
-------
vO CM
-------
TABLE 108. ARRIVING QUALITY, SITE I
Run Number 37
Actual
Time
hours
1700
1730
1800
1830
1900
1930
2000
2030
2100
2130
2200
BOD
mg/l
52
136
179
64
80
71
131
135
126
195
ss
mg/l
213
405
474
365
314
243
347
463
400
664
Col i forms
No./lOO ml
2.4xl07
7
2.4x10'
0
2.4xlOb
,
3.9x10
/•
4.3x10°
.
9.3x10
7
2.4x10'
f.
4.3x10°
/•
9.0x10
•7
2.4x10'
BOD
mg/l
93
69
61
43
68
75
63
49
37
27
20
16
13
11
9
8
7
6
6
6
6
6
Computed
SS Coliforms
mg/l MPN/100 ml
769 7-7xloJ
565 2.5x10*
713 4.0x10^
568 7.4x10?
387 5.6x10°
402 4.5x10,
388 2.9x10°
330 2.1x10°
260
196
143
105
78
61
50
39
32
25
19
16
.7x10?
.5x10?
.4x10°
.3x10°
.2x10?
.1x10°
.1x10°
.1x10?
.1x10°
.1x10°
.2x10°
.4x10?
15 1-5x10;
13 1.8x10
318
-------
X
e>
'S
Q
O
03
to
Q
-J
O
O)
O
Ui
Q
z
in
Q.
tn
180 -
150-
120
90-
60-
30-
600-
500-
Aoo
300-
200-
100-
1700
1800
1900
MEASURED
COMPUTED
O-O-OOO-O
I
2000
2100
MEASURED
1^700 1800 I960
2000
210C
Figure III
TIME-HOUR OF DAY
Run Number 37 arriving quality Site U
319
-------
5 x 10'--
J 1074-
S
8 5 x 106t
106+
5 x 103--
o
u_
5 x 10.
—, 1—, p—i j r—fi |—
1730 1800 1900 2000 2100 2200
HOUR- TIME OF DAY
Figure 112. Run Number 37 arriving quality Site I
320
-------
TABLE 109. EFFLUENT QUALITY, SITE I
Run Number 37
Actual
Time
hours
1800
1830
1900
1930
2000
2030
2100
BOD
mg/1
42
19
11
11
25
30
31
33
35
70
SS
mg/l
210
76
41
32
75
103
72
73
102
223
Col i forms
No./JQO ml
8.6 x 103
c
4.8 x 105
•)
4.8 x 103
l.
4.8 x 104
L
1.9 x 10^
1,
4.2 x 10
r
2.2 x 105
c
4.8 x 10s
c
4.8 x 105
r
4.8 x 10^
BOO
mg/1
38
28
25
18
28
39
38
30
23
16
11
8
6
5
5
4
4
3
3
Computed
SS
mg/1
142
104
132
105
108
187
215
190
150
108
75
52
36
27
22
17
14
10
7
Col i forms
MPN/100 ml
7.6 x 10^
2.5 x 10,
4.0 x 10,
7.3 x 1-0,
5.5 x 10^
8.7 x 10;?
9.6 x lo£
7.5 x lo£
5.8 x 10J!
4.5 x 10;?
3.4 x lo£
2.5 x lo£
.8 x 10;?
.5 x 10J?
.5 x lo£
.7 x lo£
.9 x 10J?
.8 x 10J?
.7 x 10P
321
-------
o
o
00
X
o
o
.j
o
LU
O
z
LU
CX.
80-
60_
20-
300-
200.
IOQ.
1300
1900
—I
2000
MEASURED
COMPUTED
2100
IROO 1300 2000 2100
TIME-HOUR OF DAY
Figure 113- Run Number 37 effluent quality Site I
322
-------
icr
5x10'
5x10^ --
O
O
CL
•g
o:
o
u.
8
5x10^ --
SxlO -
MEASURED
COMPUTED
T
T
Figure
1700 1800 1900 2000
HOUR- TIME OF DAY
Run Number 37 effluent quality Site I.
2100
323
-------
I
IL
-I 1 1-
tr\ vD O^
VO CXJ CO
S)
CT)
c
ro
E
c
o;
0)
'3
cn
A1ISN31NI
iva
NIW/WHD MOld
324
-------
TABLE 110. ARRIVING FLOW, SITE- IT
Run Number 37
Time
hours
1700
.1730
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
cu
mm.
13-9
13.9
12.2
11.4
11.4
10.3
8.7
8.7
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
10.4
15.6
15.6
19.2
19.2
15.6
15.6
15.4
8.6
7.0
Arriving
m/min
max.
20.9
20.9
19.2
19.2
19.2
17.3
15.6
14.8
13.9
12.2
12.2
12.2
12.2
12.2
12.2
12.2
22.6
24.3
24.3
26.0
26.0
24.3
22.6
22.4
14.0
12.2
flow,
cfs
mm.
8.2
8.2
7.2
6.7
6.7
6.
5.
5.
4.
4.
4.
4.
4.
4.
4.
4.
6.
9.2
9.2
11.3
11.3
9.2
9.2
9.1
5.1
4.1
max.
12.3
12.3
11.3
11.3
11.3
10.2
9.2
8.7
8.2
7.2
7.2
7.2
7.2
7.2
7.2
7.2
13.3
14.3
14.3
15.3
15.3
14;3
13.3
13.2
8.2
7.2
Computed
cu m/min.
8.3
103.7
137.5
132.4
102.0
70.4
33.2
16.0
7.3
2.0
.8
.3
.0
.3
.7
.8
1.0
0.6
0
0
0
0
0
0
0
1.2
3.7
3.7
2.2
0
0
0
0
0
0
flow,
cfs
4.9
61.0
80.9
77.9
60.0
41.4
19.5
9.4
4.3
1.2
1.1
0.8
0.6
0.8
1 .0
1.1
0.6
0.4
0
0
0
0
0
0
0
0.7
2.2
2.2
1.3
0
0
0
0
0
0
325
-------
TABLE 111. ARRIVING QUALITY, SITE II
Run Number 37
Actual
Time
hours
1700
1730
1800
1830
1900
1930
2000
2030
BOD
mg/1
56
116
45
•»•%
51
42
29
27
29
26
SS
mg/1
494
780
655
355
428
396
361
345
331
314
Col i forms
No./lOO ml
4
4
9
9
1
9
4
2
1
.6
.3
.3
.3
.5
.3
.3
.4
.5
X
X
X
X
X
X
X
X
X
IO7
,
io6
f
io6
e.
io6
-7
IO7
10?
io6
I07
10'
BOD
mg/1
53
67
50
38
30
28
28
25
20
20
Computed
SS
mg/1
79
414
460.
457
382
376
436
466
254
650
Col i forms
MPN/100 ml
2.1
1.0
5.3
4.3
7.9
.0
.3
.5
.7
.4
X
X
X
X
X
X
X
X
X
X
|»|
10r
105
106
106
106
106
106
IO6
326
-------
150-
X
o
5
Q
O
CD
100.
50.
1800
X>0
COMPUTED
1900
2000
2100
CO
Q
O
O
IU
O
z
IU
OL
(A
800_
600-
400-
200-
COMPUTED
MEASURED
Figure 116.
1800
1900
2000
2100
TIME-HOUR OF DAY
Run Number.-37 arriving quality Site II.
327
-------
s
o
o
X
2
Q.
5
10 --
5x10' --
10' --
in
I 5x1O6
O
u.
o
u
i
r-
10
6 _.
5x10
5 ..
\ r
v
MEASURED
O COMPUTED
—I 1 1 1 T-
1700 1800 1900 2000 2100
HOUR- TIME OF DAY
Figure 117. Run Number 37 arriving quality S?te II
328
-------
fit of the quality curves is therefore fair.
The actual and computed values of the effluent from the Site II treatment
unit is listed in Table 112 and graphically represented in Figure 113. These
comparisons show very little similarity because of the low arriving con-
centrations. The resulting goodness of fit is, therefore, poor.
Run Number 45
The seventh and final series sampled run was Number 45 which occurred on
August 16, 1974. The average rainfall for the entire drainage area was
1.75 centimeters (0.69 inch) over 350 minutes of noncontinuous rainfall.
There are three small rainfall events within this 350 minute duration.
This pattern of rainfall provided an opportunity to simulate the storms that
are typical for the late Summer and Fall seasons in the V/i scons in area.
The rainfall intensities at 5-min intervals for this run are listed
in Table E9, Appendix VI-E. The computed and measured arriving flows to
Site I are listed in Table 113 and plotted in Figure 119. The computed
and measured flows are very close throughout the first four hours of the
run. After this initial low flow, the computed flow peaks well above the
measured at generally the same point in time. The ratio of the total
measured flow arriving to the total computed flow is 0.53. The closeness
of fit of the two curves is generally good. Although the magnitudes of
the final peaks are different, the two records are similar in trend.
The quality of the arriving flow was determined by eight discrete samples
taken during the final portion of the overflow where a majority of the
total flow for this run occurred. Table 114 lists the actual and computed
concentrations for this run and Figures 120 and 121 present the graphical
comparison. The arriving BOO of Figure 120 show that the computed values
decrease rapidly from the start of the overflow. The measured values are
generally higher during the later stages of the overflow. The suspended
solids concentrations show the same pattern but the computed values show
a slight peak at the final stages of the run. These comparisons tend to
indicate that the SV/MM predicts most of the arriving BOD and suspended
solids during the first hours of the overflow. Because no actual samples
were taken during the first hours, it is hard to predict what the total
actual pollutographs would be. Thus, the goodness of fit of the two
pollutographs is poor. The coliform data plotted in Figure 121 show the
measured and computed values to be relatively constant -throughout the
overflow. The measured values are greater in number than the computed.
The closeness of fit is therefore, good and overall goodness of fit of the
quality data is fair.
The effluent from the treatment unit was also discretely sampled during the
final portion of this run. Table 115 lists the measured and computed
effluent concentrations and Figures 122 and 123 present these results in
graphical form. Because of the large differences in the computed and
actual influent concentrations, little if any comparison of the effluent
values can be made. The actual coliform concentrations are extremely low
329
-------
TABLE 112. EFFLUENT QUALITY, SITE I I
Run Number 37
Time
hours
1740
1800
1830
1900
1930
2000
2030
9JOO
BOD
rag/1
24
16
17
21
22
23
18
20
14
Actual
SS Conforms BOD
mg/1 No./lOO ml mg/1
*
89
199
107
136
235
228
107
92
101
28
25
18
12
12
12
10
9
6
6
6
7
6
5
5
5
7
Computed
SS Coliforms
mg/1 MPN/100 ml
172 2.08X10^
233 1.00X1 Of
195 8.29X10?
120 5.30X10,
83 7.70X10;
60
48
3^
22
18
18
20
15
13
10
14
21
.00X10^
.20X10^
.48x10*
.63x10*
.32x10::
.29x10:*
.30x10^
.40x10^
.27x10:?
.22x10^
.24x10^
.48x10-*
* No coliform analysis run due to chlorfnation problems.
330
-------
Q
O
CD
X
O
5
o
_i
o
o
IU
o
z
UJ
Q.
30-
25-
20.
15-
10-
5
6-O-Q
•o-o cf
1800
1900
2000
2100
300
250
200
100
50
A
A \
1800
1900 2000 2100
TIME- HOUR OF DAY
Ftgure 118. Run Number 37 effluent quality Site II
331
-------
TABLE 113. ARRIVING FLOW, SITE I
Run Number 45
Time
hours
Arriving flow,
cu m/min
cfs
mm.
max.
nun.
max.
Computed flow,
cu m/min.
cfs
1230
1300
1330
1400
1430
1500
1530
1600
1630
1700
1730
1800
3-
3-
3-
14.3
13.6
9.8
4.6
3.7
3.7
3.7
,0
.0
.0
3.4
3.7
4.6
4.9
5.2
4.6
3.0
2.2
4.6
6
8
6.8
5.2
6.1
7.5
6.8
6.8
11.4
15
26
26
28
,1
.3
28.7
19.7
18.8
12.9
8.3
8.3
4.6
4.6
3.7
3.7
3-7
4.0
4.6
5.2
5.6
6.1
5.2
3.7
3.0
5.2
6.8
9.0
7.5
6.1
6.8
8.3
6.8
11.4
13-6
16.6
30.9
30.9
33-3
33.3
8.4
8.0
5.8
2.7
2.2
2.2
2.2
1.8
1.8
1.8
2.0
2.2
2.7
2.9
3.1
2.7
1.8
1.3
2.7
3.6
4.9
4.0
3.1
3.6
4.4
4.0
4.0
6.7
8.9
15.6
15.6
16.9
16.9
11.6
11.1
7.6
4.9
4.9
2.7
2.7
2.2
2.2
2.2
2.4
2.7
3.1
3.3
3.6
3.1
2.2
1.8
3.1
4.0
5.3
4.4
3.6
4.0
4.9
4.0
6.7
8.0
9.8
18.2
18.2
19.6
19.6
3.6
3.4
1.5
0.0
5-1
9.3
10.3
9.3
7.8
6.0
4.4
3.5
2.9
3-2
4.0
5.1
6.4
6.8
6.3
5.1
3.9
2.5
1.5
2.2
6.6
10.7
18.0
28.9
40.6
50.6
58.1
63.6
67.8
70.4
2.1
2.0
0.9
0.0
3.0
5.5
6.1
5.5
4.6
3.5
2.6
2.1
1.7
1.9
2.4
3.0
3.8
4.0
3.7
3.0
2.3
1.5
0.9
1.3
3.9
60
10.6
17.0
23.9
29.8
34.2
37.4
39.9
41.4
332
-------
TABLE 113- (continued). ARRIVING FLOW, SITE I
Run Number 45
Arriving flow,
Time
hours
1830
1900
1930
2000
2030
2100
2130
CU
mm
28.7
18.8
14.2
14.2
6.1
6.1
6.1
3.0
3.0
3.0
2.2
m/min
max.
33.3
24.1
15.1
15.1
9.0
9.0
9.0
6.8
6.8
6.5
5.2
min.
16.9
11.1
8.4
8.4
3.6
3.6
3.6
1.8
1.8
1.8
1.3
cfs
max.
19.6
14.2
8.9
8.9
5.3
5.3
5.3
4.0
4.0
3.8
3-1
Computed flow,
cu m/min.
67.5
64.7
62.7
54.2
44.9
36.2
29.2
23.4
18.9
15.3
12.2
9.7
7.5
5.8
4.0
2.9
1.7
0.7
cfs
39.7
38.1
36.5
31.9
26.4
21.3
17.2
13.8
11.1
9.0
7.2
5.7
4.4
3.4
2.4
1.7
1.0
0.4
333
-------
Lu
O
Id
i
o:
8>
4-'
to
en
u
(0
JQ
E
•z..
CD
1-
3
CT.
1
in
CO
t
CO
vD
1
U\
1
-3-
f\
1
r»-
O —
UtJ /'UJO
A1ISN31NI
Moid
-------
TABLE 1U. ARRIVING QUALITY, SITE
Run Number 45
Actual
Time BOO SS Coliforms
hours mg/1 mg/1 No./ 100 ml
1400
1430
1500
1530
1600
1630
1700
122 109 2.8x10'
1730 7
59 200 1.2x10'
1800
115 353 8.6x10°
7
1830 77 268 2.2x10'
,
78 227 8.3x10
1900
67 197 2.5x10'
f.
1930 79 156 8.6x10
,
40 96 7-5x10
2000 ,
38 87 6.9x10°
2030
2100
BOD
mg/1
83
75
69
63
57
50
44
41
38
37
35
33
31
30
29
29
26
20
19
18
17
16
15
14
13
12
12
11
10
9
9
9
9
9
10
11
12
12
13
13
14
15
16
18
Computed
SS
mg/1
228
212
193
174
149
143
137
128
113
110
110
106
100
94
84
80
80
120
138
127
123
133
148
159
165
164
158
149
137
124
109
94
81
70
61
54
48
43
39
36
34
33
31
33
Col i forms
MPN/100 ml
7.29x10^
5.75x10°
4.75x10°
4.10x10?
3.69x10;?
3.31x10°
2.89x10?
2.57x10°
2.40x10°
2.22x10^'
2.08x10?
2.01x10?
2.02x10?
2,24x10?
2.58x10?
2.92x10?
2.88x10?
1.54x10?
.37x10°
.61x10?
.61x10?
.46x10?
.27x10?
1 .12x10?
l.OOxl 01
9.70x10^
9.61x10;?
9.51xlQr
9.60x1o5
1.01x10?
1. '08x10?
1.18x10°
1.29x10?
1.37x10?
1 .48x10?
1.63x10
1.82x10^
2.05x10?
2.18x10°
2.53x10°
2.95x10°
3.39x10°
3.85x10?
4.32x10
335
-------
N.
O
2
o
O
CD
-J
•*v
O
2
.J
O
O
LU
O
Z
IU
OL
V)
tn
150-
100.
50-
400-
300-
200-
cr
COMPUTED
MEASURED
1400
1600
I I
1800 2000
2200
MEASURED
I I I I
1400 1600 1800 2000
TIME-HOUR OF DAY
Figure 120. Run Number 45 arriving quality Site I.
2200
336
-------
O
U.
8
10 --
o
2 5x10
X
Q.
2
7- •
10'--
5X10 --
10 --
5X10-
.a'
--
1700
1800
1900
2000
HOUR- TIME OF DAY
Figure 121. Run Number 45 arriving quality S?te I.
337
-------
TABLE 115. EFFLUENT QUALITY, S!TE I
Run Number 45
Time
hours
1300
1330
1400
1430
1500
1530
1600
1630
1700
1730
1800
1830
190C
1930
2000
2030
Actual
BOD SS Coliforms
tng/1 mg/1 No7100 ml
45 70 <100
67 84 80
33 89 <80
39 66 <30
22 49 <10
23 5** <20
22 43 <10
26 6k <50
33 98 <10
41 126 <10
BOD
mg/1
35
28
0
'lit
40
37
34
31
28
26
23
20
18
16
15
15
14
14
13
12
12
11
10
8
7
7
7
7
C,
~s
9
9
8
8
7
7
6
5
4
4
4
4
4
k
4
5
k
Computed
SS
mg/1
160
138
0
45
kk
43.
41
38
35
31
27
26
24
23
20
20
19
18
17
16
14
14
14
21
24
27
33
57
80
96
105
108
106
98
88
78
63
46
30
22
16
12
9
7
6
5
Col i forms
MPN/100 ml
o
2.61x10^
2. 71x107
2.10x10?
1.61x10?
1.12x10,
8.21x10^
6.38x10,
5.14x10^
4.34x10^
3.83x10^
3.34x10^
2.94x10^
2.60x10^
2.37x10^
2.19x10^
2.03x10^
1.93x10^
1.89x10:?
1 .99x10^
2.26x10^
2.61x10^
2.75x10^
2.12x10^
1.32x10^
1.43x10^
1.56x10^
1 .47x10;?
2.72x10^
3.89x10^
4,18x10^
4.32x10^
4.48x10^
4.53xlO;>
4.47x10^
4.61x10^
4.39x10^
3.32x10^
4.53x10^
4.47x10^
4.61x10^
4.39x10^
3.32x10=
1,15x10^
1.37x10^
1 .49x10;?
1.28x10^
338
-------
X
0
s
o
o
m
80-
60 _
40-
20-
A
^
COMPUTED
^
h
XX
1400
600
1800
1
2000
2200
200-
— !
N
S
w> 15°-
Q
-I
O
V)
Q ~
Ul
a
UJ
S) 50-
^
t
,' \
/ ( A
/ \ COMPUTED /
' ^ A-A /
/ ^ /. vv 4 MEASURED
i \ 4i V \ /
* ' \ Q 7*
\ ' \ A /
^^ x ^Sp
~^» / y»
J^-y\ /X A
A^V/V /^ \
£iA A,
AA-A
,i i i ,
1^00 1660 ]8'00 ,
TIME- HOUR OF DAY
Figure 122. Run Number 45 effluent quality Site I
2200
339
-------
Q.
s
a:
o
u.
8
5X10
1800 1900 2000
HOUR- TIME OF DAY
2100
2200
Figure 123. Run Number ^5 effluent quality Site I.
-------
due to an overdose of chlorine during the run.
The arriving computed and measured flows for this run at Site II are listed
in Table 116 and graphed in Figure 12^. The same general flow pattern as
Site I results with three separate rainfalls within the 350-min duration.
The ratio of the total actual flow arriving to the total computed flow is
0.99. The actual peak at 1500 hours does not contain a computed peak while
the other peaks correlate relatively well. Again the sluice gate was closed
initially and then opened as the flow increased. The goodness of fit of
the two curves is fair because only two of the three actual peaks have
corresponding computed peaks.
The computed and actual arriving quality concentrations are listed in
Table 117 and graphed in Figures 125 and 126. Since there is no computed
flow between 1330 and 1630 hours, no arriving quality is listed but the
initial values are connected to later values to show the pollutograph
pattern. The actual and computed BOD values differ by only 25 mg/1 but the
trends are variable. The measured suspended solids concentrations are
higher than the computed but tend to follow the same trend. The coliform
data of Figure 126 shows no similarity at all. The overall goodness of fit
of the quality graphs is therefore fair.
Table 118 lists the computed and actual effluent concentrations during this
run at Site II. Figure 127 presents the graphical comparisons of thase data
No coliform analysis were obtained during this run. The same trends of
the arriving flow are carried through the Storage block with the overall
goodness of fit of the curves being fair.
VI-6 REMAINING COMBINED SEWER AREAS
The combined sewer areas outside of the previously modeled drainage area
that contribute to the Root River have also been modeled to determine the
loadings to the river during wet weather. The computed quantity and quality
of the overflows were later used to provide a more accurate representation
of the loadings to the river during the Receive block simulation.
Da ta Acq u i s i t i on
The data used to define the remaining combined sewer areas were acquired
from the Racine City Engineer's Office. The boundaries of each drainage
area, the location of the transport elements and the location of each sub-
area was determined in the same way as the main drainage area which was
described earlier. Figure 128 shows the resulting nine subareas along the
river. The total drainage area is 75 ha (175 acres). Table 119
presents the area, land use and percent imperviousness of each subarea.
The elements used to describe the transport system within these subareas
are shown in Figure 129- Table 120 lists the data describing each conduit
element. The five overflow points to the river are shown as manhole numbers
6, 5, ^, 3, and 10 in Figure 129. Points 6, 5, b, and 3 correspond to the
-------
TABLE 115. ARRIVING FLOW, SiTE II
Run Number ^5
Time
hours
1250
1300
1330
1400
1430
1500
1530
1600
1630
1700
1730
1800
1830
1900
Arriving flow,
cu m/min.
22.8
26.2
16.8
0.0
15.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
15-1
16.8
0.0
0.0
15.1
11.6
8.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
22.8
28.4
0.0
34.2
49.3
28.4
9.0
22.8
19.0
19-0
0.0
17-0
15.1
0.0
0.0
22.8
11.4
0.0
0.0
cfs
13.4
15-4
9-9
0.0
8.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
8.9
9.9
0.0
0.0
8.9
6.8
4.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
13.4
16.7
0.0
20.1
29.0
16.7
0.0
13.4
11.2
11.2
0.0
10.0
8.9
0.0
0.0
13-4
6.7
0.0
0,0
Computed
cu m/min.
4.3
45.2
24.5
0.0
4.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
6.0
0.0
0.0
0.0
12.4
28.6
35.0
0.0
36.9
37.4
37.0
0.0
37.0
37.0
36.0
0.0
38.3
18.7
2.2
0.0
0.0
0.0
0.0
0.0
flow,
cfs
2.5
26.6
14.4
0.0
2.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
3.5
0.0
0.0
0.0
7-3
16.8
20.6
0.0
21.7
22.0
21.9
0.0
21,8
21.8
21.5
0.0
22.5
11.0
1.3
0.0
0.0
0.0
0.0
0.0
-------
o
_o
en
o
o
-co
o
o
• r».
LJL
O
!£ LJ
o
o
LTV
O
O
I
o:
ID
O
X
4)
4-1
CO
c
>
(D
ur\
-3-
E
3
(1)
D
C71
L.
o
.o
ff\
Lf\
\O
in
A1ISN31NI
IVd
MOld
343
-------
TABLE 117- ARRIVING QUALITY, SITE II
Run Number 45
Actual
Time BOD SS Conforms
hours mg/1 rag/1 No./lOO ml
1250
1300
1330
1400
1430
1500
1530
1600
1630
1700
o
1730 51 256 1.0 x 10
47 344 5.6 x 10'
42 297 5.0 x lo;
1800 43 311 6.7 x 107
44 306 8.8 x 10;
45 416 8.0 x 10°
1830 38 265 3.5 x 10'
63 232 6.0 x 10'
1900
Computed
BOD
mg/1
92
81
85
85
59
48
45
46
32
25
25
22
18
16
14
13
12
11
11
12
9
SS
mg/1
914
682
828
934
367
144
139
123
125
127
168
172
167
162
159
156
152
147
145
145
84
Col i forms
MPN/100 ml
7-3 x 10^
3.2 x 10?
1.2 x 10?
1.8 x 10?
7.0 x 10
4.8 x 10^
3.6 x 10?
2.8 x 10°
- 6
2.3 x 10?
1.1 x 10°
8.6 x 10^
8.6 x 10^
8.3 x 10^
8.0 x 1o£
7.7 x 10^
7.5 x 10'
8.0 x 10:!
8.6 x 10£
8.8 x 10?
1.1 x 10?
1.6 x 10
344
-------
\
i
o
0
00
_J
CD
5
CO
Q
_J
O
o
LU
O
UJ
£L
CO
CO
90-
80-
60-
40-
20-
900-
80O-
600-
400-
2OO-
x
\
\
\
\
\
\
\ COMPUTED «
\ r
\ I MEASURED
\ • I
N\ *"^1
\
^
\
\
\
b-o-o
i i i i
1400 1600 I80O 2000
\
\
^
\
\
\
^COMPUTED
\
\
\
\
\ AM MEASURED
\ /w\
\ A
>^^A
\
A
1400 I60O 1800
TIME- HOUR OF DAY
2000
Figure 125. Run Number 45 arriving quality Site II.
-------
o
o
5x10
108-
7
in
S
01
o
u_
_J
i*
5x10J_
10-L
5x10--
MEASURED
O-° COMPUTED
1700 1800 1900
HOUR- TIME OF DAY
Figure 126. Run Number J»5 arriving quality Site II
-------
TABLE 118. EFFLUENT QUALITY, SITE II
Run Number 45
Actual
Computed
Time BOD SS Conforms BOD
hours mg/1 mg/1 No./lOO ml mg/1
1250
1300
1330
1400
1430
1500
1530
1600
1630
1700
1730
38 70
30 61
1800 24 47
23 56
24 70
1830 25 93
23 51
19 59
1900
38
33
35
35
24
20
18
19
13
10
, 10
9
8
6
6
5
5
5
4
5
4
SS
mg/1
169
126
153
172
67
26
25
21
22
23
30
31
30
29
29
28
27
26
26
26
15
Col i forms
MPN/100 ml
7-2 x 10,
3.1 x 10*
1.2 x 10^
1.8 x 10^
6.9 x 10-3
4.6 x 10^
3.4 x 10,
2.7 x 10-3
2.1 x 10^
1.1 x 10*
8.4 x 10,
8.3 x 10,
8.0 x 10,
7.7 x 10,
7.4 x 10,
7.5 x 10,
7.6 x 10,
8.2 x 10,
8.5 x 10,
1.1 x 10^
1.5 x 10-3
347
-------
O
2
O
O
CO
X
i
v>
0
o
-------
3*9
-------
TABLE 119. REMAINING SUBAREA DATA
Subarea
No.
70
71
80
91
92
93
94
95
96
Area
Hectares
11.1
12.2
2.2
4.3
7.9
2.3
6.1
23.0
5.9
,
Acres
27.4
30.1
5.4
10.6
19.4
5.7
15.0
56.9
14.5
Percent
Land Usea Impervious
1 60
1 60
1 65
5 15
4 90
k 90
2 70
4 90
3 65
a1« single family residential
2- multl family residential
3« commercial
4» Industrial
5<* park land
350
-------
U)
1C
«
ro
.n
3
I/)
£
i
U
tn
c
(Ti
2!
3
351
-------
TABLE 120. TRANSPORT CONDUIT ELEMENTS FOR REMAINING AREAS
Element
Number
302
304
306
308
310
312
314
16
315
319
21
22
324
326
329
400
402
404
331
33
325
406
41
43
345
347
348
351
53
54
355
357
359
61
363
365
71
73
016
442
441
408
446
Length
meters
80.7
84.1
91.4
73.2
44.5
154.6
41.8
128.0
97.2
304.8
127.4
365.8
97.8
22.9
3.0
913.2
455.7
632.8
114.3
67.1
12.2
390.1
121.0
59.1
116.4
31.1
61.0
320.3
376.1
3.0
5.8
158.8
63.7
265.2
85-3
152.4
85-3
109-7
128.0
3.0
349.3
609.6
609.6
Diameter
feet
265
276
300
240
146
507
137
420
319
1000
418
1200
321
75
10
2996
1495
2076
375
220
40
1280
397
194
382
102
200
1051
1234
10
19
521
209
870
280
500
280
360
420
10
1146
2000
2000
meters
0.30
0.38
0.46
0.46
0.91
0.46
0.38
0.30
0.91
0.25
1.12
0.38
1.12
1.52
3.05
0.76
0.76
0.91
0.25
0.30
0.76
0.99
0.38
0.20
0,38
0.46
1.07
0.38
0.61
0.76
0.76
0.46
0.46
0.30
0.38
0.30
0.30
0.30
0.30
1.80
0.99
0.99
0.99
feet
1.0
1.25
1.5
1.5
3.0
1.5
1.25
1.0
3.00
0.83
3.67
1.25
3.67
5.0
10.0
2.5
2.5
3.0
0.83
1.0
2.5
3.25
1.25
0.67
1.25
\ 1.5
3.5
1.25
2.0
2.5
2.5
.5
.25
.0
.25
.0
.0
.0
1.0
6.0
3.25
3.25
3.25
Slope
m/100 m
0.66
0.66
0.50
0.30
0.10
0.50
0.66
0.66
1.82
0.60
1.82
0.88
1.82
2.0
3.0
0.083
0.083
0.090
0.77
0.83
1 .70
0.090
1.00
1.00
1.00
1.20
2.00
4.00
4.20
1.00
0.90
2.90
0.54
0.94
6.10
0.66
1.10
2.46
0.66
3.00
0.08
0.90
0.90
352
-------
junctions of the same number within the Receive block. The elements numbered
greater than kOO and represented by a dashed line are the Interceptor ele-
ments of the area. The two branches of the interceptor meet at element
A09 and are then transferred to the sewage treatment plant. Manhole number
206 is the common element between these supplementary areas and the main
transport system described previously and shown in Figure 6l. The input
data for these areas is shown in Table Fl of Appendix VI-F.
P rob 1ems
The interceptor elements that flow through these areas contain large volumes
of dry weather flow from other nonmodeled areas. In order to account for
these flows in the transport network, process flows of the quantity cal-
culated for the area were added at manholes II and 206 with the BOD and sus-
pended solids concentration of the average yearly dry weather flow at the
main sewage treatment plant.
Results
Because no flow monitoring or quality determinations were obtained at any
of the overflow points in these areas, no comparisons with actual data
can be used to check output. The small contributing areas for each over-
flow point and the large capacity of the Interceptor system tend to dampen
any large computed overflow volumes. Field investigations during wet
weather have verified these facts. The computed output from these areas is
important because it provides more accurate data needed to simulate the
total inflow to the river during wet weather.
The Storage block was used with the Runoff and Transport blocks for these
areas to determine the size of a screening/dissolved-air flotation unit for
each overflow and to determine the effects on the river with and without
treatment of these computed overflows. Various runs with the Storage
block were used to "size" a treatment unit for each overflow. The results
indicate'd that the smallest size unit available in the Storage block was
best suited to these overflows. Figure 130 shows the printout of a typical
treatment facility for one of the overflow points.
The Receive block was now run using the inflows to the river from these
remaining areas with the 18,925 cu m/min (5 mgd) treatment unit at each
overflow and with the untreated outflow from the Transport block. The
results of these runs are discussed in the Receive block discussion.
353
-------
RUN NO. I
INPUT DATA FOR TREATMENT PACKAGE FOLLOWS
CHARACTISTICS OF THE TREATMFNT PACKAGE ARF
tEVEL MODE PROCESS
01
12
2?.
33
41
51
61
71
NO SFP. STORAGE
BAR RACKS
INLFT PUMPING
FINE SCR t D.A.F
(BYPASS)
(BYPASS)
(RYPASS)
(BYPASS)
IPRIMT = 0, ICHST
PES1CN STORM USED. TREATMENT CAPACITY HILL BE SELECTED TO SUIT.
OESUN FLOHRAT6 » 7.I* CFS.
f* 1.000 TIMFS MAXIMUM ARRIVAL RATE OF 7.14 CFS.)
TREATMENT SYSTEM INCLUDES MODULE UNITS
PFSIGN FLOW IS THEREFORE INCREASED TO NEXT LARGEST MflDULF SIZE
ADJUSTED DESIGN FLOWRATF = 7.I* CFS., = 5.00 MOD.
IKMCD * 1)
NO STORAGE FROM A SEPARATE STORAGE MOOFL IS ASSCCIATEO WITH THIS TREATMENT MODEL
PRELIMINARY TREATMENT BY MECHANICALLY CLEANED BAR RACKS (LEVFL t)
NUMBER OF SCREENS = 1
CAPACITY PER SCREEN = 0.00 CFS
SUBMERGED AREA = 17.70 SQ.FT. (PERPENDICULAR TO THF Finn)
FACE AREA OF BARS = 24.00 SO.FT.
INFLOW BY INLET PUMPING (LEVFL 2)
PUMPEO HFAD = ?0.00 FT. HATER
TREATMENT BY DISSOLVED AIR FLOATATION (LFVEL 31
MODULE SIZE « 5 HGO
NUMBER OF UNITS = 1
TOTA1 DESIGN FLOW = 5.00 MC.D, = Q.OO CFS
DESIGN OVERFLOW RATE - 5555.00 GPD/SF, (5100 SUGGESTED)
RECIRCULATIfN FLOW = ?0.00 PERCENT (15 SUGGESTED)
TANK OFPTH = 8.50 FEET
TOTAL SURFACE ARFA > 0.05 SO.FT.
CHEMICALS WILL BF ADDED
CHLORINE HILL BF ADDED
TREATMENT BY FINE SCREENS (AHEAD OF DISSOLVED AIR FLOATATION) (L>=VFL 3)
TOTAL SCREEN ARFA = 206. SQUARE FEFT
NO SECTNDARY TREATMENT INCLUDED (LEVEL 4)
NH FFFLUENT SCREENS (LEVFl 5)
OUTFLOW BY GRAVITY (NO PUMPING) (LFVEL 6)
N1 CHLORINE CONTACT TAK'K FOR OUTFLOW (LEVEL 7)
Figure 130. Storage block printout for remainina areas,
354
-------
SECTION VII
REFERENCES
I. "Problems of Combined Sewer Facilities and Overflows", Federal Water
Pollution Control Administration, U.S. Department of the Interior
Report No. WP-20-II, 1967. '
2. Palmer, C. L., "Feasibility of Combined Sewer Systems", Journal
Water Pollution Control Federation. 35:2:162, 1963. "
3. Dunbar, D. and Henry, J., "Pollution Control Measures for Stormwaters
and Combined Sewer Overflows", Journal Water Pollution Control
Federation. 38:1:9, 1966. "-"
4. Benzie, W. J. and Courchaine, R. J., "Discharges from Separate Storm
Sewers and Combined Sewers", Journal Water Pollution Control
Federation. 38:3:410, 1966. ———'
5. "Pollution Effects of Stormwater and Overflows from Combined Sewer
Systems", U.S. Department of Health, Education and Welfare, 1964.
6. Agnew, R. W., et.al.. "Biological Treatment of Combined Sewer Over-
flow at Kenosha, Wisconsin" trough draft), EPA Project H023 EKC.
7. Benjes, H. H., et.al.. "Stormwater Overflows from Combined Sewers",
Journal Water Pollution Control Federation. 33:12:1252, 1966.
8. Camp, T. R., "Overflows of Sanitary Sewage from Combined Sewer
Systems", Journal Water Pollution Control Federation. 3l:4":38l, 1962.
9. "Dissolved-Air Treatment of Combined Sewer Overflow", Federal Water
Pollution Control Administration, U.S. Department of the Interior,
Report No. WP-20-17, 1970.
10,
II
Mason, D. G., and Gupta, M.K., "Screening/Flotation Treatment of
Combined Sewer Overflows", EPA Office of Research and Monitoring.
Report No. 11020 FDC 01/72, 1972.
State of Wisconsin Administrative Code, Register, September 1973
No. 213, Chapter NR 102 and NR 103.
355
-------
12. "Screening/Dissolved Air Flotation Treatment as an Alternate to Com-
bined Sewer Separation", detailed post construction plan of operation
for EPA demonstration project performed by Ecology Division, Rex
Chainbelt Inc., December 1971.
13. Service Bureau Corporation, Ca.11/360, "Statpack Statistical Package",
Form No. 65-2208-1, New York, NY, 1969.
14. Rohlf, F. J., and Sokal, R. R., Statistical Tables, W. H. Freeman
and Co., San Francisco, California, 1969.
15. Miller, 1., and Freund, J. E., Probability and Statistics for Engi-
neers, Prentice-Hall Inc., Englewood Cliffs, NJ, 1965.
16. Metcalf and Eddy, Inc., "Urban Stormwater Management and Technology:
An Assessment" (in preparation), EPA Report for Contract No. 68-03-0179.
17. Thomann, R. V., Systems Analysis and Water Quality Management.
Environmental Science Services, New York, NY, 1972.
18. Southeastern Wisconsin Regional Planning Commission, "A Comprehensive
Plan for the Root River Watershed", July 1966.
19. State of Wisconsin Department of Natural Resources, "Report on the
Investigation of the Pollution in the Root River Basin Made During
1966 and 1967", November 27, 1967.
20. United States Federal Water Pollution Control Administration, Water
QuaJIW Criteria. Report of the National Technical Advisory Committee,
U.S. Department of the Interior, 1968.
21. Sokal, Robert R., and Rohlf, F. J., Biometry. W. H. Freeman and Com-
pany, San Francisco, California, 1969.
22. Reid, G. K., Ecology of Inland Waters and Estuaries, Van Nostrand
Reinhold Company, New York, 1961.
23. Service Bureau Corporation, Ca11/370, Data Analysis, Form No.
65-2^91.2; San Francisco, California, 1973.
2k. Hutchlnson, G. E,v "A Treatise on Limnology", John Wiley & Sons,
Inc., New York, 1967.
25. Aquatic Life Advisory Committee of the Ohio River Valley Water
Sanitation Commission, "Aquatic Life Water Quality Criteria, First
Progress Report", Sewage and Industrial Waste, 27;321-331.
356
-------
Nemerow, N. L., Scient i fie Stream Pol jut ion Ana lysis , Scripts Book
Company, Washington, D.C.; McGraw-Hill Book Company, St. Louis,
26.
27. Schuettpelz, D. H., "Fecal and Total Coliform Tests in Water Quality
Evaluation", State of Wisconsin Department of Natural Resources,
Research Report No. 42, 1969.
28. Patrick, R., "Biological Measure of Stream Conditions", Sewage and
Industrial Waste, 22:926-938., 1950.
29. Wurtz, C. B., "Stream Biota and Stream Pollution", Sewage and In-
dustrial Wastes. 27:1270-1278.
30. Wilhm, J. L., and Dorris, T. C., "Biological Parameters for Water
Quality Criteria", Bioscience. 18:477-481, 1967.
31. Hooper, F. F., Eutrophication Indices and Their Relation to Other
Indices of Ecosystem Change, In Eutrophication: Causes. Consequences.
Correctives, National Academy of Sciences, Washington, DC, I9&9.
32. Wilhm, J. L. and Dorris, T. C., "Species Diversity of Benthic Macro-
invertebrates in a Stream Receiving Domestic and Oil Refinery
Effluents", American Midland Naturalist, 76:427-449, 1966.
33« Cairns, J. Jr., and Dickson, K. L., "A Simple Method for the Biologi-
cal Assessment of the Effects of Waste Discharges on Aquatic Bottom-
Dwelling Organisms", Journal Water Pollution Control Federation,
43:755-772, 1971.
34. Odum, E. P., Fundamentals of. Ecology, W. B. Saunders Company,
Phi Iadelphia, Pennsylvania, 1971.
35. Stormwater Management Model, Volume III, Metcalf 6 Eddy Inc., Palo
Alto, California, University of Florida, Gainesville, Florida, Water
Resources Engineers, Inc., Walnut Creek, California, Environmental
Protection Agency, Publication No. 1102^ DOC 09/71, July 5971, p. 359.
36. Edmondson, W. T,, and Wivberg, G. G (ed.), A Manual on Methods for
The Assessment of Secondary Productivity in Fresh Watersf B1ackwe11
Scientific Publications, Oxford, England, 1971.
37. Yentsch, C. S., and Menzel, D. W., "A Method for the Determination
of Phytoplankton Chlorophyll and Phaeopljytin", Deep-Sea Research.
10:221-231.
38. Stormwater Management Model, Volume III, Metcalf & Eddy Inc., Palo -
Alto, California, University of Florida, Gainesville, Florida, Water
Resources Engineers, Inc., Walnut Creek, California, Environmental
Protection Agency, Publication No. 11023 DOC 07/71, Final Report;
11024 DOC 08/71, Verification and Testing; 11024 DOC 09/71, User's
Manual; 11024 DOC 10/71, Program Listing.
357
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-106a
I. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
SCREENING/FLOTATION TREATMENT OF COMBINED SEWER
OVERFLOWS; Volume II - Full-Seale Operation
Racine, Wisconsin
5. REPORT DATE
August 1979 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
T.L. Meinholz, D.A. Gruber,
J.H. Moser, M.J. Clark
8. PERFORMING ORGANIZATION REPORT NO.
R.A. Race, C.A. Hansen,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Envirex Inc. (A Rexnord Company)
Environmental Sciences Division
Post Office Box 2022
Milwaukee, Wisconsin 53201
10. PROGRAM ELEMENT NO.
1BC822. SOS 1, Task 14
11. CONTRACT/GRANT NO.
S-800744
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Anthony N. Tafuri (201) 321-6679, FTS 340-6679
See also Volume II Appendices. EPA-600/2-79-106b. available from NTIS.
16. ABSTRACT
This study involved the planning, design, construction and operation of a two-year
evaluation period, of three full-scale demonstration systems for the treatment of storm
generated discharges. As part of the evaluation, the quality of the receiving body was
also monitored. Two of the systems, located at two major points of combined sewer
overflow to the Root River in Racine, Wisconsin,, employed screening/dissolved-air
flotation to treat the overflow prior to discharge. The two systems had a combined
capacity of 222,000 cu m/day (58.5 mgd). The third system utilized screening only
for the treatment of urban stormwater, having a capacity of 14,800 cu m/day (3.9 mgd).
Results indicated that the "satellite plant" concept of locating treatment plants
at points of combined sewer overflow discharge is a feasible alternative to combined
sewer separation. Based on the operating results of these systems, removal efficien-
cies of 60 to 75 percent can be expected for suspended solids and 50 to 65 percent for
BOD. The chlorination system met the fecal coliform standard for the whole body con-
tact specified by the State of Wisconsin for the Root River. It was concluded that
the operation of the treatment systems had a beneficial effect on the quality of the
River.
Results from the screening of urban stormwater indicated that this method would
remove 50 percent of the suspended solids and 20 percent of the BOD.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
*Combined sewers, *0verflows-sewers,
*Waste treatment, Sewage treatment, Sewage,
Waste water, Mathematical models
Screening/flotation
treatment, Dissolved air
flotation, Stormwater
runoff, Urban hydrologic
models, Combined sewers
overflows
13B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
376
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
358
ft U.S. GOVEBNMENT PRINTING OFFICE; 1979 -637-060/5454
------- |