v
WATER POLLUTION CONTROL RESEARCH SERIES 11020 FAQ 03/71
Dispatching System for Control
of
Combined Sewer Losses
ENVIRONMENTAL PROTECTION AGENCY • WATER QUALITY OFFICE
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Reports describe the results and progress
in the control and abatement of pollution of our Nation's waters. They provide
a central source of information on the research, development and demonstration
activities of the Water Quality Office of the Environmental Protection Agency,
through in-house research and grants and contracts with the Federal, State,
and local agencies, research institutions, and industrial organizations.
Triplicate tear-out abstract cards are placed inside the back cover to facili-
tate information retrieval. Space is provided on the card for the user's
accession number and for additional key words. The abstracts utilize the
WRSIC system.
Inquiries pertaining to Uater Pollution Control Research Reports should be
directed to the Head, Project Reports System, Planning and Resources Office,
Research and Development, Water Quality Office, Environmental Protection
Agency, Washington, D.C. 20242.
Previously issued reports on the Storm and Combined Sewer Pollution Control
Program:
11034 FKL 07/70 Storm Water Pollution from Urban Land Activity
11022 DMU 07/70 Combined Sewer Regulator Overflow Facilities
11024 EJC 07/70 Selected Urban Storm Water Abstracts, July 1968 -
June 1S70
11020 — 08/70 Combined Sewer Overflow Seminar Papers
11022 DMU 08/70 Combined Sewer Regulation and Management - A Manual
of Practice
11023 --- 08/70 Retention Basin Control of Combined Sewer Overflows
11023 FIX 08/70 Conceptual Engineering Report - Kingman Lake Project
11024 EXF 08/70 Combined Sewer Overflow Abatement Alternatives -
Washington, D.C.
11023 FD3 09/70 Chemical Treatment of Combined Sewer Overflows
11024 FKJ 10/70 In-Sewer Fixed Screening of Combined Sewer Overflows
11024 EJC 10/70 Selected Urban Storm Water Abstracts, First Quarterly
Issue
11023 --- 12/70 Urban Storm Runoff and Combined Sewer Overflow Pollution
11023 DZF 06/70 Ultrasonic Filtration of Combined Sewer Overflows
11024 EJC 01/71 Selected Urban Runoff Abstracts, Second Quarterly Issue
To be continued on inside back cover
-------
DISPATCHING SYSTEM FOR CONTROL OF
COMBINED SEWER LOSSES
by
Minneapolis-Saint Paul Sanitary District
now the
Metropolitan Sewer Board
St. Paul, Minnesota
For the
Environmental Protection Agency
Water Quality Office
Demonstration Grant 11020 FAQ
March, 1971
NOTE: On August 1, 1970, the Minneapolis-Saint Paul Sanitary District
was absorbed by the Metropolitan Sewer Board. The Metropolitan
Sewer Board carries on all functions of the former Sanitary District.
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price (1.75
Stock Number 5601-0103
-------
EPA Review Notice
This report has been reviewed by the Water
Quality Office, EPA, and approved for publication,
Approval does not signify that the contents
necessarily reflect the views and policies of the
Environmental Protection Agency, nor does mention
of trade names or commercial products constitute
endorsement or recommendation for use.
mv-" ~
-------
ABSTRACT
Impressive reductions in combined sewer overflow
pollution of the Mississippi River in Minneapolis and
Saint Paul have been effected by a regulator control
system. Working entirely within the limits of the
existing interceptor sewer system, and with relatively
minor modifications to selected major combined sewer
regulators, incidence of overflow was reduced by 66%
and duration of overflow by 88% during most of a
rainfall season. Computer simulation techniques using
actual rainfall data indicate that the amount of
overflow volume reduction achieved is the equivalent
of a $200 million separation project. The efficiency of
collection was improved by about 20% at controlled
regulators. The reduction in volume of combined
overflow to the river is estimated to be between 35%
and 70% during the runoff season. The unmodified
combined sewer system captured about 65% of the
urban runoff. Where modified, the system captured
about 77% of the urban runoff.
A mathematical model has been prepared that will,
with rain gage data as input, perform rainfall runoff
analysis, diversion of combined sewer runoff hydro-
graphs at regulators, and routing of diverted hydro-
graphs through the interceptor system. This model will
assist in operation of the system to retain combined
sewer flows and utilize the maximum flow capacity of
the existing interceptor sewer system. Part II of this
report describes the model.
The 1.75 million dollar project includes a computer-
based data acquisition and control system that permits
remote control of modified combined sewage regu-
lators. Data from rain gages, regulator control devices,
trunk sewers and interceptors, and river quality moni-
tors provide real-time operating information. Time
varient quality data from key locations in the sewer
system were obtained by automated analysis of nu-
merous hourly samples.
Work on the "Dispatching System for Control of
Combined Sewer Losses" began in 1966, and the
system has been operated since April, 1969.
This report was submitted in fulfillment of Demon-
stration Grant 11020 FAQ, supported in part by the
Environmental Protection Agency, Water Quality
Office.
in
-------
TABLE OF CONTENTS-PART I
Page
Abstract iii
List of Figures vii
List of Tables x
I. Summary, Findings, and Conclusions 1
II. Introduction 9
Minneapolis-St. Paul Climatology, River System, and Sewer System 9
Minneapolis-St. Paul Sewage Works Studies, 1956-1961 11
Proposed Regulator Control System 12
MSSD Personnel 13
III. Studies of Sewage Strength During Dry and Wet Weather 15
Sampling Program 15
Automated Analysis 16
Data Handling 16
IV. Installation of Gaging System 27
In-Sewer Work by Regulator Crew 27
Contract 664A 27
Rain Gages 27
V. Regulator Modifications 31
Contract 677 31
Post-Construction Evaluation 34
VI. Process Control and Communications Equipment 39
Processing System 39
Leased Line System 41
Communications System 43
Calibration Procedures 44
Inplant Data Acquisition System 44
VII. River Monitoring 47
Site Selection 47
Contract 666 47
Installation, Operation, and Maintenance 48
Findings and Conclusions 52
VIII. Computer Analysis of Storm Runoff and Mathematical Model of the
Regulator-Interceptor System 53
IX. Operation of the System 55
Data Acquisition Programs 55
Modes of Operation 59
X. System Effectiveness 69
Rainfall Events Analyzed 69
Regulator Performance 72
Projected Seasonal Effectiveness 75
1970 Spring Thaw 78
Computer Analysis of Simulated Rainfall 78
Cost-Effectiveness of the Control System 80
XI. Recommendations 81
XII. References 85
XIII. Acknowledgements 87
Appendix A 89
Appendix B 111
-------
TABLE OF CONTENTS - PART II
Page
I. Introduction 123
II. General Concept of Model 125
III. Specific Modeling Techniques 127
A. Hydrology Phase 127
1) Impervious Analysis 127
2) Pervious Analysis 128
B. Diversion Phase 129
C. Routing Phase 132
1) General Discussion 132
2) Method of Characteristics 132
3) Progressive Average Lag Method 134
IV. Discussion of Modeling Techniques 135
V. Model Performance for Natural Events 137
VI. Critique of Model 143
References 145
Appendix A—Operator's Manual 147
Appendix B-Definition of Model Parameters 167
Appendix C-Listing of Model (UROM-9) 173
VI
-------
LIST OF FIGURES-PART I
Figure Page
1. Minneapolis St. Paul Control and Monitoring System 3
2. Drawing of Typical Control and Monitoring Regulator 4
3. Central Station Process Control Equipment 5
4. Performance of Modified Regulators for Simulated One Hour Rains 6
5. Utilization of Minneapolis Northwest lnterceptor-1960 and 1969 7
6. Routes of the Principal Intercepting Sewers 10
7. Utilizatiorrof Joint Interceptor 11
8. Regulator Crew and Vehicle 12
9. Regulator Crew Servicing Automatic Sampler 15
10. Technicon Autoanalyzer used for Analysis of Sewage Samples 16
11. Autoanalyzer Data Key Sheet 18
12. Technicon Autoanalyzer Report of Scaled Values 19
13. STORET Retrieval of Hourly Sewage Analyses 19
14. Machine Plot of Hourly COD Data 20
15. Frequency Distributions—COD at Minneapolis Southwest Interceptor 21
16. Minimum, Maximum, and Mean-24 Hour COD at Minneapolis Southwest Interceptor 21
17. Three-Dimensional Plot of 24-Hour COD Data for Treatment Plant Raw Wastewater 22
18. Sewage Flow Meter Reading Key Sheet 22
19. Comparison of Treatment Plant COD Loadings by Month 23
20. Flow Meter Recorder Chart Showing Effect of Rainfall on the Minneapolis Northwest Interceptor 23
21. Plant Raw Wastewater COD Loadings-All Hours and Stormf low Hours 24
22. Sewage Flow and COD Concentration Compared for a Dry Day and a Wet Day 25
23. Pole-mounted Cabinet 27
24. Transmitting Weighing-type Rain Gage 28
25. Locations of the System's Nine Transmitting Rain Gages 29
26. Location of Control and Monitoring Regulators 32
27. New Venturi Meter Control Gates at Minneapolis Northwest Interceptor 31
28. Hydraulic Power Operated Gate at the Trout Brook Regulator 33
29. Inflated Fabridam Diversion Weir at the Trout Brook Regulator 33
30. Inflated Fabridam and Dry Weather Outlet at Otis and Marshall Regulator 33
31. Inflated Fabridam at St. Peter and Kellogg Regulator 34
32. Hydraulic Power Units 34
33. Control Valves, Piping, and Blower for Inflatable Dam 37
34. Digital Equipment Corporation PDP-9 Computer 39
35. Control Data 853 Disk Drive 40
36. Potter Line Printer 40
37. Interface Cabinet 40
38. Schematic Diagram of Outplant Stations and Leased Line System 42
39. Calibration of an Outplant Address using the CHKOUT Program 44
40. Plant Operator's Logging Equipment 45
41. River Quality Monitor Analyzer Unit 48
42. River Quality Monitor Locations 49
43. River Quality Monitor Pump Screen 48
44. Industrial Molasses River Monitor 48
45. Chloride Data from River Monitor on a Thawing Day 51
VII
-------
46. Rain Report Scale Factor List 55
47. Data File as Stored on the Disk 56
48. River Quality Monitor Report 56
49. Regulator Control Report 57
50. Interceptor Report 57
51. Interceptor Monitoring Stations 58
52. Rain Report 59
53. Use of SCNFRQ Routine to Control Scanning Frequency 60
54. Fabridam and Gate Control List 60
55. Gate Control Report 61
56. Data Analysis Report 62
57. Machine Plotted Rain Gage Data for July 1-2, 1970 63
58. Machine Plotted Interceptor Sewer Flow Depth Data for July 1-2, 1970 64
59. Machine Plotted Trunk Sewer Flow Depth for July 1-2, 1970 64
60. Ogive-1960 Data for Minneapolis East Interceptor 65
61. Ogive-1970 Data for Minneapolis East Interceptor 65
62. Assigning Data Addresses to Recorder Pens Using SELREC 66
63 Estimated Disposition of Seasonal Runoff on Combined Sewered Areas in
Minneapolis and St. Paul. 77
64. Flow Meter Chart-Minneapolis East Interceptor 78
65. Model Predicted Overflow Reduction 79
66. Model Data from Simulated Rainfall 80
VIM
-------
LIST OF FIGURES-PART II
Figure Page
1. Map of Twin City Area Showing Location of Interceptor Sewers, Major Inlets, Diversion
Structures, and Rain Gages 123
2. Map of Twin City Area Showing Major Watersheds and Thiessen Polygons 127
3a. Impervious Loss Function 128
3b. Pervious Loss Function 128
4a. Impervious Loss Rate for 1 In./Hr. Rainfall Rate 128
4b. Pervious Loss Rate for 1 In./Hr. Rainfall Rate 128
4c. Combined Low Rate for 1 In./Hr. Rainfall Rate on a Watershed 1/3 Impervious 129
5. Typical Type 1 Diversion Structure 130
6a. Typical Type 1 Diversion Rating Curve with Fabridam Up 131
6b. Typical Type 1 Diversion Rating Curve with Fabridam Down 131
7. Illustration of Flow Clipping Action of Diversion Structure 131
8. Forward and Backward Characteristics and Definition of Grid Points 133
9. Characteristic at Downstream Boundary 133
10. Results of Routing Irregular Hydrograph by the Method of Characteristics and by the
Progressive Average Lag Method 133
11. Progressive Average Lag Routing 134
12a. Temporal and Areal Distribution of Storm on Day 207 (July 26, 1969) 137
12b. Temporal and Areal Distribution of Storm on Day 211 (July 30, 1969) 137
12c. Temporal and Areal Distribution of Storm on Day 218 (Aug. 6, 1969) 137
12d. Temporal and Areal Distribution of Storm on Day 267 (Sept. 24, 1969) 137
13a. Inlet Hydrograph and Hydrograph Diverted to the Interceptor Sewer at Minnehaha Diversion
for Event of Day 211 138
13b. Inlet Hydrograph and Hydrograph Diverted to the Interceptor Sewer at Minnehaha Diversion
for Event of Day 218 138
13c. Inlet Hydrograph and Hydrograph Diverted to the Interceptor Sewer at Minnehaha Diversion
for Event of Day 267 138
14a. Metered and Model Predicted Flows for Event of Day 207 139
14b. Metered and Model Predicted Flows for Event of Day 211 139
14c. Metered and Model Predicted Flows for Event of Day 218 140
14d. Metered and Model Predicted Flows for Event of Day 267 140
15a. Effect of Fabridam on Model Predicted Flows Reaching Waste Treatment Plant for Event of
Day 207 141
15b. Effect of Fabridam on Model Predicted Flows Reaching Waste Treatment Plant for Event of
Day 211 141
15c. Effect of Fabridam on Model Predicted Flows Reaching Waste Treatment Plant for Event of
Day 218 141
15d. Effect of Fabridam on Model Predicted Flows Reaching Waste Treatment Plant for Event of
Day 267 141
IX
-------
LIST OF TABLES-PART I
Table Page
1. Reduction in Incidence and Duration of Regulator Overflows Accomplished by Regulator
Demonstration Program 2
2. Minneapolis-St. Paul Monthly Precipitation 9
3. USGS Data for Mississippi and Minnesota Rivers through Water Year 1966 9
4. Minimum, Mean, and Maximum River Flows, cfs 10
5. Interceptors in Minneapolis and St. Paul 10
6. Summary of Sewer Sampling Program : . . . . 15
7. Tabulation of Additions and Modifications—Contract 677 35
8. Trunk Sewer Diversion Weir Data 36
9. Number of Measurement and Control Functions 43
10. River Quality Monitor Parameters 51
11. Mississippi River Discharge 52
12. Interceptor Level Measurements, Locations, and Sizes 59
13. Scanning Time Requirements—Minutes 66
14. Historical Precipitation Data 70
15. 1969 Monthly Precipitation Data 70
16. Total Rainfall at Each Gage-16 1969 Rains 70
17. Maximum 60 min. Intensity at Each Gage-16 1969 Rains 71
18. Maximum 30 min. Intensity at Each Gage-16 1969 Rains 71
19. Total Rainfall at EachGage-11 1970 Rains 72
20. Maximum 60 minute Rainfalls at Each Gage-11 1970 Rains 72
21. Summary of 1969 Overflow Incidence Reduction 73
22. Summary of 1969 Overflow Duration Reduction 73
23. Summary of 1970 Overflow Incidence Reduction 74
24. Summary of 1970 Overflow Duration Reduction 74
25. Overflow Volumes at Eleven Regulators From Mathematical Model with Actual Rainfall Data. . . 75
26. Overflow Control—Mathematical Model Prediction versus Real Data Indication 75
27. Project Effectiveness in Reducing Frequency and Duration of Overflow 76
28. Precipitation as Snowfall for Winter of 1969-1970 78
29. Reduction of Combined Sewer Overflow During 1970 Spring Thaw 78
30. Simulated Rainfall Data used for Mathematical Model Analysis 79
-------
I. Summary, Findings and Conclusions
Summary
This project was conceived during the period be-
tween 1956 and 1960 when the sewer and interceptor
systems of Minneapolis and Saint Paul were intensively
investigated as part of a study for the Minneapolis-
Saint Paul Sanitary District on "Expansion of Sewage
Works in the Minneapolis-Saint Paul Metropolitan
Area." These investigations revealed that the inter-
ceptor system was not being fully utilized for con-
veyance of combined sewage during periods of rainfall
and runoff. The net effect of the multiple regulator
settings all over the Twin Cities was that combined
sewage was often needlessly diverted to the Mississippi
River when interceptor capacity was available. More-
over, the studies revealed that rainfall seldom, if ever,
occurs uniformly over the Twin Cities and that this
rainfall diversity could be used to advantage in re-
ducing or eliminating combined sewage overflows if a
means of controlling a variable diversion regulator were
available.
, In 1965 the Minneapolis-Saint Paul Sanitary District
took over the responsibility for maintenance and
operation of combined sewage regulators in Minnea-
polis and Saint Paul from their respective Sewer
Departments in an effort to improve the performance
of the regulators and reduce combined sewer overflow
pollution by more intensive monitoring. A special
Regulator Crew was organized and equipped for this
purpose.
After legislation in the Water Quality Act of 1965
provided the possibility of obtaining Federal funding
assistance, the Minneapolis-Saint Paul'Sanitary District
was prepared and made an application to the Federal
Water Pollution Administration for a "Facilities
Demonstration Grant" to demonstrate an improved
method for controlling the discharge from a sewer
system carrying storm water and sewage. This appli-
cation was accepted in April, 1966, and the Minnea-
polis-Saint Paul Sanitary District obtained the first of
the FWPCA Demonstration Grants, a sum of $870,750,
to constitute 50% funding for a "Dispatching System
for Control of Combined Sewer Losses." The project
was described in the application as follows:
"The Cities of Minneapolis and Saint Paul are
generally served by combined sewer systems. Over-
flows from the sewer system to the Mississippi
River, caused by rainfall or thaw, occur at more
than ISO points in the system. The losses of
domestic and industrial wastewater are estimated to
be up to six per cent of the annual wastewater flow
and possibly from six to fifteen per cent of the
annual average solids and BOD. The losses occur
mainly in the spring, summer and fall. No assess-
ment of the effect of overflows on the bacterial
condition of the river has been made.
i
The proposed system envisions extensive modi-
fications to diversion structures at approximately 15
points along the interceptor system, minor modifi-
cations at most of the remaining diversions, and
construction of five new 'Master Diversions' at key
interceptors, along with telemetry and supervisory
control systems to allow monitoring of overflow
interceptor utilization. Maximum utilization of the
interceptor system and treatment facilities will then
be possible, considering rainfall distribution and
other factors. The routing or dispatching of mixed
storm and wastewater flows can be controlled based
on locations of storms in the system, time of day
and other factors. At diversions with high pollu-
tional load, priority cah be given to avoiding losses
during run-off. The fifteen diversions carried ap-
proximately 80 per cent of the Sanitary District's
load in 1960.
Past estimates of the cost of separation of all
sewer systems in both cities have ranged from
$30,000,000 for a low degree of separation to
$100,000,000* for fairly complete separation which
precludes such a program in the near future. The
proposed system will considerably reduce losses at a
nominal cost.
The pre-construction and post-construction
evaluation will take full advantage of modern data
handling and analysis techniques with development
of correlations, time series analysis and mathe-
matical models. All data will be in accordance with
the USPHS "STORET" system to facilitate use of
data by other agencies. Real time data processing
techniques will be used to direct the operation of
the system, demonstrating its use in the field of
pollution abatement, and providing useful data for
future planning and operations of the proposed
system or those which might be designed for other
areas for reduction of combined sewer system
losses.
The proposed project will demonstrate a new
technique of instantaneous observation and control
of interceptor system performance based on ade-
quate information to drastically reduce losses of
combined wastes. Information gathered in operation
of the system will provide a basis for further
reduction of losses by using trunk sewers for storage
and the facilities constructed will allow such a
measure to be attempted. Post-construction evalua-
tion will provide information which will allow the
method to be adapted to other large combined
sewer systems of differing configuration and clima-
tology.
Since the majority of losses of wastes occurs
during the recreational season, considerable benefit
to the Mississippi River, where it passes through the
populated area, will accrue. "
More recent estimates place the cost for complete separation
at $600,000.000.
-------
The first year of the project was spent in pre-
paration of plans and specifications for the major
contracts involved in the project. Preliminary in-sewer
work for the gaging system was also done by Sanitary
District personnel. Construction phases of the project
began in 1967 and Regulator Modifications were
completed in 1968.
Figure 1 shows, on a map of the metropolitan area
all the elements of the system, including the existing
interceptors. There are 18 control and monitoring
stations on the regulator-interceptor system, nine
monitoring rain gages and five river quality monitors.
Most components of the regulator control and
monitoring system are shown in Figure 2, an artist's
rendition of a regulator typical of those that were
modified. The underground vault provides a suitable
location for all the electronic and motorized equip-
ment which is connected to electric power and
telephone circuits. Depth of flow is measured upstream
from the diversion dam and downstream from the
regulator gate in the outlet pipe leading to the
interceptor. Ths gate is operated by an operating
cylinder and an oil hydraulic power system in the vault
and can be remotely controlled. The diversion dam is
inflated with air and can also be remotely controlled.
Water can flow over the inflated dam, which does not
completely close off the trunk sewer. The dam can be
deflated to afford almost full hydraulic capacity of the
pipe.
The central control station for the system is located
in the Administration Building at the Minneapolis-Saint
Paul Sanitary District Wastewater Treatment Plant. The
central part of the system is the process control
equipment, the functional parts of which are shown on
Figure 3. The same type of equipment has been used to
control electric power systems, refineries, chemical
plants, steel mills and many other processes. In
addition to accomplishing control and monitoring of
the outplant system, the system is capable of logging
treatment plant operating data.
Findings and Conclusions
Real operation of the system began in April, 1969.
t During the 1969 runoff season operation was some-
what impeded by a major flood on the Mississippi
River and later by "hardware bugs" that prevented
complete acquisition of data. Additionally, the rainfall
occurring during 1969 was considerably less than
normal. Nevertheless, enough data that could be used
for purposes of evaluation was accumulated. More
recent data from April and May, 1970, supplements
that from the initial operating season.
The data collected from each regulator allows one
to determine when the depth of flow exceeds the
height of the Fabridam or any other height, such as the
height of the fixed weir that was replaced by the
Fabridam. Based upon the actual data, tables showing
performance as measured by reduction in incidence of
overflow and by reduction in duration of overflow
were prepared. Incidence and duration of overflow
were determined for pre-project conditions with fixed
height diversion weirs and post-project conditions with
fully inflated Fabridams as the diversion weirs.
The results of these analyses are summarized in
Table 1.
Table 1
REDUCTION IN INCIDENCE AND DURATION OF
REGULATOR OVERFLOWS ACCOMPLISHED BY
REGULATOR DEMONSTRATION PROGRAM
% Reduction % Reduction
in Incidence in Duration
16 1969 Rains 65
11 1970 Rains (April and May) 49
Total of 27 Rains 58
88
87
88
A reduction of incidence, obviously, could only be
accomplished when the modifications at the regulator,
primarily the Fabridam, allowed complete retention or
diversion of combined sewer runoff that would not
have been possible with the old fixed weirs. Reductions
in duration were calculated by comparing total regula-
tor hours of duration for the two cases. The duration
figures are high primarily because many of the old
weirs were set only slightly higher than dry weather
flow, and consequently the trailing or falling limb of
the storm hydrograph would cause overflow for several
hours after the bulk of the storm water had passed.
Interpretation was used to prevent distortion of dura-
tion data that could be caused, and usually a depth
flow within a few inches of the weir height was used as
the point of overflow cessation.
For purposes of evaluating performance, 16 rainfall
events were selected from the 1969 rainfall season.
Although more rains occurred during the 1969 rainfall
season, the monitoring system did not acquire data on
all of them, and some data that was gathered was lost.
Particularly troublesome were hardware problems in
the disk handler which were completely corrected on
July 10, 1969.
The 16 selected rains do provide a representative
sampling, since they include 56 per cent of the rainfall
which occurred during the period between April 26
and November 16.
This 1969 data was supplemented by data from
eleven rains occurring in April and May of 1970. When
the 27 means of total rainfall at each of the system
gages are added, the sum is 15.59 inches, so the data
sample analyzed is almost equivalent to a normal
rainfall season, at least as measured by precipitation.
-------
INTERCEPTOR ION I TOR I KG STATIONS
MSSD RIVER QUALITY MONITORS
MSSO REMOTE TRANSMITTING RAIN GAUGES
CONTROL AND MONITORING REGULATORS
l
CO
-------
UNDERGROUND EQUIPMENT VAUL
POWER OPERATED GATf
-------
DISC FILE
PDP-9 COMPUTER
I-O TELETYPE
INTERFACE EQUIP.
Mill Ml
7 LEASED LINES
FROM
OUTPLANT STATIONS
LINE PRINTER
OPERATORS AND LOGGING
TELETYPES
TREND RECORDERS
AND
OPERATORS PANEL
The central part of the system is the process control equipment. The same type of equip-
ment has been used to control electric power systems, refineries, chemical plants, steel
mills and many other processes. This system can control the sewer system and has also
been designed to take readings of treatment plant performance.
PLANT CONTROL ROOM
Figure 3. Central Station Process Control Equipment
-------
Although these analyses certainly indicate that the
control system has great potential and has improved
the capture efficiency of the combined sewer-regula-
tor-interceptor system, it is desirable to be able to
speak in terms of overflow volume kept out of the river
by the system and still entering the river during large
storms in spite of the system. The mathematical model
of the sewer-regulator-interceptor system provided a
convenient method of simulating pre- and post-project
conditions and was used with data from certain rain
days to calculate reduction in volume of overflow
occuring from the operation of the system. Total
overflow volumes from 11 major regulators were
reduced by from 98 to 30 per cent during seven
individual rains by operation of the system according
to this analysis.
The mathematical model was also used with simu-
lated rainfall data in an attempt to evaluate the
effectiveness of the control system in terms of rainfall
intensity and duration. Figure 4 shows how rainfall
intensity effects regulator performance.
20
Analysis of interceptor monitoring data indicates
that full capacity of the Minneapolis Southwest and
East Interceptors has been utilized on several occasions
during rainstorms. The capacities of the Minneapolis
Northwest and Joint Interceptors are being utilized
much more fully than in 1960 when previous gaging
data was 'collected. Evidence suggests that the Joint
Interceptor at the treatment plant has been operated at
near full capacity on several occasions. Figure 5
compares the utilizations of the Minneapolis Northwest
Interceptor during 1969 and 1960.
A system tor controlling overflows from combined
sewer regulators has been installed and operated. Very
significant reductions in combined sewer overflows
have been effected and documented through use of this
control and data acquisition system since the spring of
1969. Overflow control benefits will continue to
accrue so long as the system is maintained and
operated. Moreover, the effectiveness of the system
will increase as operating experience accumulates and
problems encountered are solved. The use of the
.25
1.00
.50 .75
ONE HOUR RAINFALL INTENSITY, IN./HR.
Figure 4. Performance of Modified Regulators for Simulated One Hour Rains
1.25
-------
10
MAX.STORM FLOW
7-8-69 8.25'
MAX. STORM FLOW 5-20-60 6.31
MAX. DRY WEATHER FLOW 1969 4.2'
MAX. DRY WEATHER FLOW 4-22-60. 5-13-60 3.3
9'-6" x 9'-6" SECTION
NORTHWEST INTERCEPTOR
AT V/ENTURI METER
MINNEAPOLIS
40 60
% OF TIME DEPTH OR FLOW EQUALS OR EXCEEDS
Figure 5. Utilization of Minneapolis Northwest lnterceptor-1960 and 1969
100
operating rainfall-runoff-diversion-routing mathe-
matical model developed for the system when available
as a real-time operating aid, will also increase the
effectiveness of the overall system.
Data acquisition and operating experience will make
it possible to up-date, improve, and even expand the
existing mathematical model. A logical extension of
the model, incorporation of flow data feedback, is
discussed in Part II of this report.
The overall flexibility of the computer based system
will permit operation to meet changing conditions in
the tributary urban area. The data acquired by the
system will be of great value in planning future
modifications of the existing combined sewer and
interceptor system. The mathematical model can be
modified and used to test and evaluate new ideas and
alternative solutions to various problems related to the
combined sewer system and the interceptor system.
Considering the present concern about the polluting
effect of separate storm sewer discharges and the
increasing body of evidence supporting this concern,
this system has had an effect on the magnitude of
storm related pollution greater than that which would
occur from a 200 million dollar sewer separation
program which would require years to complete.
Among the accomplishments and findings of this
project can be listed the following:
1. The number of regulator overflow occurrences at
controlled regulators was reduced by 58 per cent. In 27
rainfall events 164 overflow discharges were prevented.
2. The total combined sewer overflow hours at
controlled regulators were reduced by 88 per cent. In
27 rainfall events 1036 overflow hours were prevented.
3. In 7 rainfall events analyzed for overflow volume
at 11 controlled regulators, overflow volumes for
individual rains were reduced by from 98 to 30 per
cent, with a net reduction of 51 per cent. Except for
intense storms occurring only two or three times per
year, overflow volumes at these regulators can be
reduced by about 75 per cent or more.
4. Estimates place the total annual overflow volume
that can be saved by the system at 230 MG. To
duplicate this reduction by a separation program.
-------
10,000 combined sewered acres would have to be
separated at a probable cost of at least 200 million
dollars.
5. Data acquisition and computer analysis indicate
that the runoff capture efficiency of the combined
sewer system was improved by about 20 per cent at
controlled regulators. The reduction in volume of
combined overflow to the river is estimated to be
between 35 per cent and 70 per cent during the runoff
season. The unmodified combined sewer system cap-
tured about 65 per cent of the urban runoff. The
modified system captured 77 per cent of urban runoff
at the modified locations.
6. A network of nine telemetering rain gages have
and are providing urban rainfall information of a
quantity and quality never before available. Rainfall
diversity over the metropolitan area has been shown to
be very common.
7. Monitoring of the existing interceptor system has
shown that it is now being more effectively utilized for
conveying storm water runoff.
8. River quality monitors along the urban reaches
of the Mississippi River have shown that the physical
quality of the water is not appreciably affected by
storm and combined sewer discharges.
9. Analysis of sewage at many points in the sewer
system has permitted the establishment of diversion
priorities to help reduce combined sewer overflow
pollution.
10. The control system was shown to have practi-
cally eliminated overflow from spring thaw runoff at
the controlled regulators.
11. A working, verified, and calibrated mathe-
matical model of the urban hydrology and the
combined sewer regulator-interceptor system has been
prepared and is available for studies and as a sewer
system operating aid. This model is thoroughly de-
scribed in Part 11 of this report.
12. The regulator monitoring system has permitted
continuous surveillance of the major combined sewer
regulators.
13. The program has demonstrated the efficacy and
value of a computer basec control and monitoring
system for overall efficiency of data acquisition and
control and monitoring functions.
-------
II. INTRODUCTION
Minneapolis-St. Paul Climatology, River System, and
Sewer System.
Minneapolis and St. Paul, commonly known as the
Twin Cities, are located side by side along the
Mississippi River in Eastern Minnesota. The climate is
temperate, with temperature ranges between extremes
of 99°F and -32°F. Climatological standard normal
(1931-60) precipitation is 24.78 inches annually. Of
this, 16.07 occurs during the months May through
September. Table 2 shows the maximum, normal, and
minimum precipitation in Minneapolis-St. Paul for each
month.
A recent publication by the University of Minnesota
(1) indicates that series of annual precipitation amounts
are approximately normally distributed. In this report
a somewhat longer period of record was studied for
Minneapolis-St. Paul than the climatological standard
period, 1931-60. St. Paul data from 1837 to 1933 and
Minneapolis data from 1934 to 1965 provided a
combined 129 year record. The range in annual total
precipitation is great, 39.48 inches, between a maxi-
mum of 49.69 inches in 1849 and a minimum of 10.21
inches in 1910. The 129 year mean annual precipita-
tion is 26.85 inches.
The Mississippi River enters the Twin Cities from
the North 1812 miles above its mouth. The river
approximately bisects the Metropolitan area, flowing
within, between, or along Minneapolis or St. Paul for
25 miles. The Minnesota River flows into the Missis-
sippi River near the middle of its urban reach at River
Mile 1799. Lock and Dam No. 1 on the Mississippi
River is about two miles upstream from the Minnesota
River confluence. The Mississippi River watershed is
about 16,000 square miles. The corps of Engineers
maintains a 9 ft. navigation channel in the river, with
the head of navigation being located in the upper St.
Anthony Pool at River Mile 1811.
Upstream from St. Anthony Falls, the Mississippi
River flows through primarily glacial drift. Down-
stream from St. Anthony Falls, the River flows
through a narrow gorge for about 10 miles. Below the
Minnesota River confluence, the Mississippi occupies a
portion of a wider valley formed by a large pre-glacial
river.
Table 3 and Table 4 summarize USGS river data for
the Mississippi River and Minnesota River. Minnesota
River near Carver is 36 miles upstream from the
Mississippi River. Mississippi River near Anoka is about
6 miles upstream from Minneapolis. Mississippi River at
St. Paul is about 6 miles downstream from the
Minnesota River.
Jan.
Table 2
MINNEAPOLIS - SAINT PAUL PRECIPITATION
(1931 -1960)
Feb. Mar. Apr. May June July Aug. Sept. Oct.
Nov. Dec.
MINNEAPOLIS
Maximum
Normal
Minimum
1.65
0.70
0.11
2.66
0.78
0.14
3.37
1.53
0.48
3.53
1.85
0.62
7.87
3.19
0.74
7.80
4.00
1.26
7.10
3.27
0.11
6.60
3.18
0.43
7.53
2.43
0.41
5.64
1.59
0.26
5.15 1.99
1.40 0.86
0.27 0.06
Table 3
USGS DATA FOR MISSISSIPPI AND MINNESOTA RIVERS
THROUGH WATER YEAR 1966
Drainage area, sq. mi.
Period of record, years
Period of record mean
day, cfs
Period of record maximum
day, cfs
Period of record minimum
day, cfs
Mississippi River
Near Anoka
19,100
35
7,053
91,000
586
Minnesota River
Near Carver
16,200
32
3,174
117,000
79
Mississippi River
at St. Paul
36,800
68
10,080
171,000
632
-------
Table 4
MINIMUM, MEAN AND MAXIMUM RIVER FLOWS, CFS
Minnesota River
Near Carver
Mississippi River
Near Anoka
Calendar
Year
1962
1963
1964
1965
1966
1967
1968
Mln.
1700
1750
1820
2900
2200
1500
1500
Mean
8232
5854
5856
13730
11800
7500
7700
Max.
39400
22100
23700
90300
42200
40600
20600
Min.
350
390
385
372
482
245
185
Mean.
6028
2760
2080
7782
3336
3381
4217
Max.
39400
14400
12900
112000
16000
19300
37200
Mississippi River
at St. Paul
Min.
2020
2370
2170
3410
2260
1950
1750
Mean
14590
8546
7820
22380
15730
11100
12400
Max.
56200
31300
32300
171000
52200
52100
25500
Minneapolis, St. Paul, and most of their suburbs are
served by a single large waste water treatment plant
operated by the Minneapolis-Saint Paul Sanitary Dis-
trict (MSSD). The extensive interceptor system leading
to a single large treatment plant was feasible because of
favorable topographic and geological conditions.
Ground elevations in the entire region served permit
the use of gravity flow. A sandstone formation
underlying much of the area permitted the construc-
tion of large interceptor sewers by economical tunnel-
ing methods. With the favorable ground elevations and
the deep interceptors, very little pumping is required.
The total area served is 144,000 acres, of which 54,000
acres are in the core cities of Minneapolis and St. Paul.
Total estimated tributary population is 1,400,000,
with a population equivalent based on BOD of approxi-
mately 2,400,000. Interceptor sewers were constructed
during 1934-1938 by Minneapolis, St. Paul, and the
MSSD. Figure 6 identifies the major existing inter-
ceptor sewers. Flow from the three Minneapolis Inter-
ceptors is metered primarily for purposes of cost
apportionment.
Figure 6. Routes of the Principal Intercepting Sewers
Table 5 indicates the section and capacity of the
interceptors at their downstream ends. Twenty-eight
miles of branch interceptors were constructed by the
two Cities, and a nine mile jointly used interceptor was
constructed by the MSSD. The replacement cost of this
interceptor system is in excess of $100 million, with
the replacement cost of the Joint Interceptor alone in
excess of $25 million.
Table 5
INTERCEPTORS IN MINNEAPOLIS
AND SAINT PAUL
Interceptor Section Capacity, MGD
Minneapolis NW 9'6" x 9'6" 230
Minneapolis SW 6'0" x 6'0" 75
Minneapolis E 6'0" x 6'0" 85
Miss. River Blvd. 3'9" x 5'10" 40
W. 7th St. 5'0" x 6'0" 56
West Side 3'6" x 5'6" 34
Joint 9'6" x 10'0" 620
Double Barrel
The system was designed to provide for the sewage
flow for the year 1970 from a population of
1,400,000. Design capacity for the interceptors was
calculated at 55 percent for sanitary sewage and 45
percent for storm flow.
Sewage regulators accomplish the diversion of flow
from trunk sewers into the interceptor system. They
also control or limit the quantity of combined sewage
which enters the interceptor during periods of runoff.
About 150 separate regulators and diversion structures
are used in the two cities. The regulators are of various
types, float operated gates, tipping gates, orifices,
leaping weirs, and simple diversions. The original design
of the regulators was intended to allow some storm
water to enter the interceptor system. The 1928
Report of the Metropolitan Drainage Commission
describes this "first-flush" of storm water as that
quantity assumed to be caused by a rainfall intensity of
0.04 inches per hour. The rate of runoff was deter-
mined by the Rational Method with assumed coeffici-
ents ranging from 0.2 to 0.8.
10
-------
Most suburbs are served by separate sewer systems
but discharge into either combined sewers or inter-
ceptor sewers which carry combined flow.
Minneapolis has carried on a program of sewer
separation since 1933. Separation is limited to street
catch basins and is not, however, 100% effective. The
separation work has been accelerated during the past
10 years and is being continued. As of 1969 the city is "
nominally 75% separated, that is served to some extent
by separate storm drainage. Saint Paul is still primarily
served by combined sewers, with only about 20% of its
sewered area served by a separate sewer system.
Minneapolis-Saint Paul Sewage Works Studies,
1956-1961
In May 1956, the Board of Trustees of the
Minneapolis-Saint Paul Sanitary District authorized a
five year program of research and development on
future metropolitan area sewage works needs. Because
the existing interceptor system would certainly be
utilized as the backbone of a future expanded system,
engineering investigation was undertaken to determine
its current utilization. A gaging program accumulated
flow depth data at 12 locations during parts of 1958,
1959, and 1960 to supplement the flow records
available from meter stations on the Minneapolis
Interceptors and at the Treatment Plant. Data was also
obtained from about 20 non-recording rain gages read
daily by volunteer observers. Frequency distributions
of recorded flow depths were made and maximum flow
depths during dry weather flow periods and during
storm flow periods were noted.
Figure 7 shows a cumulative frequency distribution
or ogive for the Joint Interceptor at Third St. and
14.
10
a.
Ul
o
MAXIMUM STORM FLOW 10-23-59 9.0*
MAXIMUM DRY WEATHER FLOW 8-19-59 6.8*
13-X12' BOX SECTION
MINNEAPOLIS - SAINT PAUL SANITARY DISTRICT JOINT INTERCEPTOR
AT THIRD ST. t COMMERCIAL ST.
% OF TIME DEPTH OP PLOW EQUALS OR EXCEEDS
Figure 7. Utilization of Joint Interceptor
11
-------
Commercial St. in St. Paul. This location is about 2.4
miles upstream from the Treatment Plant, and only a
few inlets to the Joint Interceptor are downstream
from it. That storm flow conditions occur only a small
percentage of time is evident from the figure. The
volume of storm water flowing in the interceptor
during this small percentage of time is dependent upon
the amount of rain and the functioning of sewage
regulators. At this location and at others in the
interceptor system it was found that the capacity of
the pipe was far from being fully utilized even during
heavy rainstorms.
As part of the studies an inspection was made of
each regulator in the system. About 120 regulators
consisting of float operated gates, tipping gates, ori-
fices, leaping weirs, and simple diversions were known
to be in use at that time. Many were of relatively small
size, on 10 inch and smaller sewers. In general it was
found that the regulators were not functioning as
originally planned. The original design was usually not
conducive to regular maintenance so non-functioning
mechanical devices and obstructed openings were
found. Changing flow conditions and operational prob-
lems had also necessitated modifications at some
regulators. In both cities some regulators were found to
be permitting overflow to the River during dry
weather. Poor hydraulic flow conditions, where velo-
cities and turbulence were great, caused some of these
instances. Other regulators were found to be operating
under flow conditions where only a very small amount
of storm flow would initiate overflow. The net effect
of all the regulator operating conditions was such that
the interceptors seldom or never operated at full
capacity. Conveying capacity available for storm flow
was unused at the same time that overflows would be
occurring at many regulators.
It was determined that 15 major regulators diverted
an estimated 80 percent of the total dry weather flow
arriving at the Treatment Plant. These large regulators
were also the ones causing most problems and at which
most combined sewer losses occurred. The structures at
the metering stations at the ends of the three Minne-
apolis Interceptors also act somewhat as regulators,
since the interceptor can be relieved to the river by
overflow weirs and bypass gates. These 15 regulators
and 3 interceptor metering stations were recognized as
18 key locations for effecting control of combined
sewer losses.
Another recommendation of the report was that
"All regulators which are adjacent to interceptors
should be placed under the control of a single
authority since they play the most important role in
controlling interceptor use and overflows affecting the
river. In addition to permitting readjustment of all
regulators to optimum settings, central control would
also assist in preventing unnecessary by-passing of raw
sewage, thereby improving river conditions within the
Cities."
In 1965 Minneapolis and St. Paul enacted ordi-
nances which transferred to the Sanitary District the
control of and responsibility for the operation and
maintenance of combined storm and sanitary sewer
diversions and regulators from which overflow can be
conveyed directly into the Mississippi River. A regu-
lator crew was established by the Sanitary District to
carry out this newly assigned responsibility. The crew
was equipped with a service truck, necessary tools and
equipment, and extensive safety equipment. Their
initial task was to catalog and describe all regulators
and make physical improvements necessary to permit
them to be safely inspected and maintained on a
regular basis. Figure 8 shows some of the men with
their truck. The ordinances also provided for the
establishment, construction, maintenance, and the
operation of a regulator control system by the Sanitary
District and the payment of costs for the system.
Figure 8. Regulator Crew and Vehicle
Proposed Regulator Control System
The analyses of the interceptors and regulators and
rainfall data resulted in the proposal of a regulator
control system contained in Volume III of the "Report
on the Expansion of Sewage Works in the Minneapolis-
Saint Paul Metropolitan Area". (2) The proposed
system was described thusly:
"It is proposed that a comprehensive control
system be initiated to provide an integrated system
of regulator operations which will take advantage of
the uncontrollable factors affecting the system ...
Any part of the collection system which can be
made variable will provide for more efficient use of
facilities in the future...
It is proposed that a system of supervisory
control be installed at key regulators and other
points in the system to provide regulator operation
based on interceptor utilization.
12
-------
The operation of regulators at present cannot be
based on interceptor level and cannot be varied to
meet actual conditions...
The system of present regulators has a fixed
hydraulic capacity with the variations of quantity
admitted to the interceptor sewer varying with the
head at each particular regulator and sometimes
with the functioning of a gate con tolled by level in
the trunk sewer.
The float regulator shown in old literature can
now be replaced with modem power operated gates
controlled by a supervisory system from one control
point. Gate positions, flows, and levels in major
sewers can be telemetered simply and accurately to
rapidly provide a dispatching operator with a
picture of instantaneous functioning of the inter-
ceptor system.
Jt is suggested that such a system would prove
immediately useful in reducing loss of sanitary
sewage at regulators and provide for day by day
adjustments to meet changing conditions.
Along with the supervisory control system, it is
proposed that modifications be made to non-
automated regulators to provide better operation
and simplify maintenance. It is proposed that a
system of telemetering be installed to provide data
on interceptor sewer and trunk sewer levels for
operation of the supervisory control, determination
of frequency and quantity of overflow of sanitary
sewage in actual operation and to provide data for
design of future modifications. . .
The proposed system will allow full utilization of
existing interceptor sewers. It will also insure a
reduction of the quantity and frequency of sanitary
sewage overflow to the river. ..
The central control of regulators will allow the
choice by an operator of points of bypassing raw
sewage in the system based on river conditions at
the time . . .
The proposed system could also provide auto-
matic readout of data for analysis, either manually
or with computer techniques, to further improve
operation of the system ... In addition, should the
separation programs set forth by the central cities
not fully materialize, the control system would
provide the best means of minimizing loss of
sanitary sewage without major expenditures for
further modification of regulators".
The Water Quality Act of 1965 included legislation
which provided Federal Water Pollution Control Ad-
ministration matching grants for demonstration pro-
jects using new or improved methods for controlling
the discharge from combined sewer systems. The
Sanitary District, being prepared as a result of the
earlier studies, made an application for a Facilities
Demonstration Grant in April of 1966. The Sanitary
District obtained the first such grant for its proposed
project for installation and evaluation of "A Dis-
patching System for Control of Combined Sewer
Losses." The grant amount of $870,750 is 50 percent
of the $1,741,500 total project cost.
MSSD Personnel
Because of the size and nature of the project, which
involved some new approaches, a separate department
or area was essentially created by the Sanitary District.
The Project Engineer for the Regulator Demonstration
Program had been responsible for the previously
initiated Regulator Crew and their duties. This respon-
sibility was retained, and, for the most part, all other
time was devoted to the Demonstration Program.
During the period from July, 1967 to June, 1969, the
Project Engineer had the assistance of another full time
Assistant Engineer. Since June, 1969, only one Engi-
neer has worked on the Project. Since September,
1967, the Project group has included a Data Analyst-
Logger Operator who was familiar with the sewer
system, having worked for three months with the
Regulator Crew, and who had data processing training.
A secretary was assigned to the Demonstration Pro-
gram during most of the Period.
The Regulator Crew was utilized for certain work
and did all of the sewer sampling. The crew contained
four men until the summer of 1968 when week-end
sampling was carried on. Its size was increased to five
at that time and has remained so because of the added
work load.
Two chemists from the MSSD Laboratory were
utilized extensively for operation of the Technicon
Autoanalyzer. They received training on the equip-
ment, did all of the testing preceding the full scale use
of the equipment and operated it during the initial
sampling periods. They later trained part time and
temporary student employees who were used for
operating the equipment seven days per week during
1968.
Five temporary student employees were utilized for
Autoanalyzer operations and other purposes during
1968. During the summer of 1966 one student
employee was used; during the summer of 1969 three
student employees were hired. All temporary help were
students at the University of Minnesota, most engineer-
ing, chemistry, and mathematics majors. Student em-
ployees have performed most of the routine River
Monitor maintenance since the monitors were installed.
One of the electricians from the Maintenance
Department of the Sanitary District was assigned to the
Project in May, 1968, to be used as needed. He has
been used between one-half and three-quarter time for
telemetry maintenance, river monitor maintenance,
calibration and other electrical maintenance work. A
pipefitter was employed on the Project full time for
about three months.
13
-------
III. STUDIES OF SEWAGE STRENGTH DURING
DRY AND WET WEATHER
In the sewer systems of Minneapolis and St. Paul
variations in composition and quantity of flow are
greater than the more well defined variations which
occur at the Treatment Plant. Because of this continu-
ous variability, single grab samples or composite
samples can not describe the conditions existing
adequately enough to assist in real-time operation.
Therefore a combined sewer sampling and analytical
program was initiated.
The intent of the sewer sampling program was to
accumulate a large enough volume of data to enable
statistical methods to be used in describing diurnal
variations in dry weather sewage composition and the
effects of short term transient events, namely rainfall
and runoff events. Knowledge of relative pollutional
character of waste waters by location and time of day
is useful in determining operating procedures from a
pollution control standpoint. When rainfall conditions
will not permit operation for maximum diversion
because of limited interceptor capacity, local differ-
ences in pollutional strength of waste water can
determine priorities for diversion and bypassing. Data
from the runoff events provide information useful in
determining the character of combined sewage over-
flow and evaluation of the magnitude of pollution
from this source.
6 summarizes the sampling program, showing the
number of days of sampling at each location.
Photo By Kent Koberstein
Courtesy Minneapolis Sunday Trioune
Figure 9. Regulator Crew Servicing Automatic Sampler
Table 6-SUMMARY OF SEWER SAMPLING PROGRAM
Sampling Program Regulator Demonstration Program Sampling Data
_„ " November 1966 to September 1968
SERCO automatic samplers were used to obtain .
hourly grab samples. The samplers hold 24 bottles in a ys
rack; each bottle is connected to an individual sample
tube running to the weighted sample head. A portable £ ant Secondary Effluent 12
vacuum pump is used to evacuate the bottles. A clock £|a^ ^nmary E
releases a clamp on one rubber sample tube each hour '
to collect a grab sample. Usually five samplers were ' ' '
used and five different locations sampled at one time. _
Each morning, the MSSD's Regulator Crew would visit _ _ g v ' co
, ot. "eter « NeiioQQ b^
each sampler, remove the twenty-four marked sample We$t Seventh |n^ 43
bottles, drain the sample tubes, install a new set of D. 0 D __
Rice & Rondo 55
bottles, recharge the vacuum and reinstall the sampler Mjssjssippi River B|vd. lnterceptor 35
in thesewer. Otjs & Marsha|| 33
Sampling was initiated in November, 1966, and Wabash & Cromwell 11
some sampling was done through October, 1967, with Wabash & Eustis 65
a total of 633 station days sampled at 14 different Minneapolis East Meter 77
stations in the sewer and interceptor system. Sampling Qak Street 73
began again in March, 1968, and lasted until September Second & Main 67
15, 1968, with a total of 832 station days sampled. 31st & Randolph 61
Figure 9 shows members of the Regulator Crew Minneapolis N.W. Meter 80
withdrawing one of the samplers from a manhole to 26th & Seabury 42
remove the samples and recharge the sampler. Portland & Washington 18
A total of 24 stations were sampled—16 combined Camden 61
sewer regulators, the 3 Minneapolis Interceptors where Minneapolis S.W. Meter 80
storm flow relief can occur, 2 other St. Paul Inter- 38th & Edmund 47
ceptors, and 3 Treatment Plant locations, the raw flow, 39th & Minnehaha 41
the primary effluent, and the secondary effluent. Table TOTAL .... 1465
15
-------
The Regulator Crew usually devoted the first two
hours of each day to servicing samplers. They were met
at the last station serviced and the samples were
transported to the MSSD Laboratory. Sample days
usually ran from 8 A.M. to 8 A.M. Somewhat fewer
than the maximum possible 35,160 samples were
collected, because of occasional equipment problems
leading to loss of a sample. Sampling was resumed
during August and September, 1969, when the gaging
and monitoring system was operating, but little rainfall
occurred during this period.
The samplers have been used frequently within the
plant and by the University of Minnesota on pilot plant
sampling also.
Automated Analysis
To run analyses on an additional 120 samples daily
was beyond the capabilities of the Sanitary District
Laboratory, so automatic analysis equipment was
acquired. The equipment used was the Technicon
Autoanalyzer, a portion of which is pictured in Figure
10. Two MSSD chemists attended a special training
course in its use and were able to work out difficulties
encountered in adapting the equipment for use in
analysis of sewage. The operation of the equipment
and its testing prior to sampling have been reported.(3)
Five or six analyses were made on each sample
collected, at a rate of 30 samples per hour. The
analyses primarily used were chemical oxygen demand,
chloride, phosphate, ammonia nitrogen, and total
Kjeldahl nitrogen, although the equipment has also
been used to measure iron, urea, nitrite nitrogen,
nitrate nitrogen, and ABS. The equipment has also
been used for analysis of treatment plant samples and,
after modification, for analysis of river water samples.
A "common standard" containing all of the com-
ponents measured was developed. Standards at 5
concentration levels were inserted after every 24
samples. Daily plant composites for raw flow, primary
effluent, and secondary effluent were also run in
triplicate. The number of discrete samples analyzed ran
up to 159 each day.
Data Handling
At an average of 5 analyses per sample, up to 795
discrete analyses were made daily. The analysis results
were recorded on continuous strip chart recorders, with
peaks on a continuously traced line representing an
optical density for each sample and standard run.
Calibration curves prepared from the optical densities
observed for various standard solution concentrations
could have been used to manually convert sample
optical density to mg/l. However, this would have
Figure 10. Technicon Autoanalyzer used for Analysis of Sewage Samples
16
-------
been time consuming because of the number of
analyses and because standards were repeated between
sets of 24 samples to account for base line variations.
As an alternative to the manual method a machine
processing method was developed. A key sheet was
designed to be used for keypunching of raw data onto
cards, Figure 11. Sample identification and standard
concentrations and optical density values for standards
and samples were read from the recorder graphs and
entered onto the key sheet. A program was written to
fit a curve to the standard concentrations and optical
densities and to update the curve as standards were
repeated to account for base line drift. The program
calculates the concentration values and generates a
report of scaled values,Figure 12. A program was also
written to punch the data onto cards in a format
acceptable to the STORE! System. The programming
was done for MSSD by a programming and systems
consultant and data processing computer time was
rented from a data processing center.
Now after almost two years of sampling there have
been approximately 200,000 discrete analyses, with
about 150,000 on sewage samples. Over 6,000 samples
and 30,000 analyses were for the plant raw waste. On
the average about 1,200 samples and 6,000 analyses
were obtained from each of 21 other locations in the
sewer system.
Storage within the STORET System was easily
accomplished. An example of a retrieval report, gen-
erated in Washington by the STORET System is shown
in Figure 13. Hourly sewage strength data for a few days
at one of the control and monitoring regulators in St.
Paul are listed. The STORET analysis program was
used to obtain the number of observations, maximum,
minimum, mean, and standard deviation for each
analysis at each station by month, by quarter, by year,
and by period of record.
Variations in concentration occurring on any par-
ticular day are often large as Figure 14,a machine plot
from the STORET system, shows. COD concentration
in mg/l is plotted for a regulator in St. Paul over a 70
day period beginning on March 18, 1968. Wet and dry
weather analysis results are included in the data. There
are up to 24 data points plotted vertically for each day,
and COD concentration on any day varies consider-
ably. Recurring periods of higher plotting density
indicate the lower variability of COD on Saturday and
Sunday at this regulator.
Using data center facilities and the services of our
programming consultant the data were further pro-
cessed. Distributions by hour were made for all data at
each station. Frequency distributions have been plot-
ted for some hours for each analysis at each station.
The diurnal variations in COD shown on Figure 14
were not found to be regular, for most hours of the
day were distributed quite widely. Some frequency
distributions skewed on rectangular coordinate paper
plotted normally on semi-log paper. Figure 15 is an
example of one of these plots, showing frequency
distributions of COD data from the Minneapolis S.W.
Interceptor for four hours.
Plots of maximum, minimum, and mean have been
prepared on 24-hour graphs to illustrate diurnal vari-
ations. These functions for COD data from the
Minneapolis S.W. Interceptor are shown in Figure 16.
The distribution of values within the extremes could be
quite varied, so it is informative to be able to illustrate
both at once. The 24-hour and frequency distribution
plots were combined in a few instances to create
three-dimensional plots such as Figure 17, which shows
the range and frequency distribution of hourly treat-
ment plant raw wastewater COD data. Included in the
plot are both wet and dry weather values.
Continuously recorded flow information was also
available so loadings could be calculated. The plant raw
waste flow and sewage flow from each of the three
Minneapolis Interceptors is metered and recorded at
the treatment plant on 24-hour circular charts. Hourly
readings from these charts were transferred to specially
prepared key sheets. Figure 18. Flow data from
November, 1966, through June, 1969, has been punch-
ed onto cards. Over 180,000 hourly flow values have
been transferred and recorded in this manner.
Distributions of plant and interceptor flows by hour
have been made and plotted. Statistical analysis runs
producing number of observations, maximum, mini-
mum, and mean have also been made and 24-hour plots
have been drawn, and some three-dimensional plots
have been prepared. Diurnal flow variations at four
places in the system have thus been well described.
This will also assist in formulating operational strate-
gies. Flow variations with season have been shown by
distributions by month.
Loadings in pounds per hour were calculated from
the hourly Technicon analysis results and the hourly
flow values and distributed by hour of day, by day of
week, and by month for the plant raw waste. Plant
hourly flow values used in calculating loadings are
those taken through the treatment plant. Primarily
because of a deficiency in grit channels, portions of
high flows caused by storm runoff which exceed safe
limits are bypassed. Bypass flow volumes are estimated,
but are not included in the loading calculations.
Comparison of the by month frequency distribu-
tions of hourly plant loadings indicates that there is no
predictable effect of storm flow on plant loading. In
1967 the months of May and August had 0.61 in. and
2.79 in. of rainfall respectively. In 1968 the same two
months had 3.74 in. and 0.75 in. of rainfall. The
distribution of hourly COD loadings for May, 1968,
with over six times the rainfall of May, 1967, plotted
to the left of May, 1967, indicating lower loadings. The
distribution for August, 1967, with almost four times
17
-------
MINNEAPOLIS-SAINT PAUL SANITARY DISTRICT
00
DATE C
DATE A
Tl
CD
'
C
g
3
D)
Kl
CD
a
Bl
CD
co
I
CARD CODE
OH
0,2
0,3
0|1
0|2
Oi3
OLL
NAL
1 RUN NUMBER
0)1
0|1
on
0,2
0|2
01?
0,1
0|2
0 1 3
0 1 1
0|2
0)1
0|2
0|3
OH
0|2
0|3
0
4
0(3
Oi3
ni3
0,4
0|4
0|4
0|5
0|5
0|5
ECT
YZEI
ED
1
STATION
CODE
N
\
/
\
\
'
0)6
0|6
0|6
n
\
i
\
ANALYSI
UJ
o
8
Q
4
U
>
1
\
,
\
| MONTH
1
' \
o
\
M
J
^ < 11
0 *" >
•<
C
! 1
i
t \
! 1
t
. V
< C
I !
AUTOMATIC ANALYZER LOADING AND RESULT
ANALYST
e
4
«t
3
3
C
>
>
>
/
V
1
v
,
\
t
\
f
/
COMMON STANDARD
OPTICAL DENSITY
COMMON STANDARD
CONCENTRATION
ANALYZER READINGS BY HOUR
1 1 1
1 ,1, 1
i i
- 61
! ,! 1
i
i
i l 1
!
i
1 ' 1
, , ,
1
,
i
,
I !
1
i
1 ll '
!
L,l
i
-H
i
-i_
i
I
1
1
|
01
1 1
I |
1
! 1
1 1
1
1 1 l
l 1 1
EU
1 1
I l
i i i
i i
i i
i i
l 1 I
! 1B
t < i
i u
-M-1
i i
i i
1 i.
1 1
H-H
1
l
U
L.
1 1
I
i 1
!
1
i 1
1 1
1
1
1
1 1 1
i i
i
i
i
i
-L-L-
-L-L-
! 1
I) J 1
-L-L.
1 1
1 1 1
1 1
I 1
1 1 1
L_1_L_
I 1
1 1
LJ_1_
1 1
1 !
l 1
C2
1
1
1
J
1
,
1 j 1
-4
i
i j i
u
L-,
M
-.-J-L-
1 1 1
i i l
i i
1 l l
•'.V-1-
1 1 1
1 J '
1 1 1
1 J 1
1 1 1
1 1 1
1 P-
1,1, 1
1 1 1
1 ,1 1
1,1, 1
1 1 1
H-H-
PACE
OF
COMPOSITE
RAW
<;
,
U
PRI
1
,
1
1
SEC
III !
ill 1
III l
l 1
1 1 1
1 1 1
i
,
1
,
Ml 1
1 1 1
i i l
l J. 1 1
,
1
1
1 1
1 1 1
l l
L-
,
|
1 1
III 1 l
1 1 l l
1
1 1
1 1 1
l l
H-H
u,.u,u.u..
-------
lUfiOAt UT Ofc*
PPfi "<*e
Tl'jPiiMt CHICHI 01
P(K CL
-C/l ' "C/l
— TTTT r
?|-CO -^ '* '-I 1.0
21-00 ^/rt.. _£.?N.C
•
tl
J
ii
_L
* . i i r if
u . *. » i ^ •.
•1 r^— #
t .1'*.
I . !M
i .-:;? j
.•'0
"of
X'.l
1 141.0
1 1R.Q
Figure 12. Technicon Autoanalyzer Report of Scaled Values
StnREI RETRIEV1L OtTE 68/12/20
1214C
26<>cia
IBRUT 60 DDK REGUKTOR
26 MINNESOTA
0033^
0«IE TIDE DEPTH COD
FROM OF LOMIEVEL
TO D»T FEET "»C/l
6T/06/0<) 01 00
02 OH
0) 00
0* 00
Of 00
06 30
07 00
67/06/12 09 00
10 00
11 00
12 OC
13 00
15 00
16 00
17 00
18 00
19 00
20 00
21 00
22 00
23 00
?« 00
67/06/13 01 00
02 00
03 00
0* 00
OS 00
06 00
07 00
67/06/1* 06 00
09 00
10 00
U 90
12 00
13 00
It 00
15 OC
16 00
17 00
277.1
222.3
2*6.0
229.3
327.0
186.D
330.0
392.3
»26.0
Ml.O
525.0
298.0
458.0
397.0
»95.0
426.0
41*.0
391.0
440.0
293.0
197.0
216.3
351.0
J57.0
317.0
487.0
244.0
251.0
295.0
273.3
362.0
37S.O
416.0
355.0
324.0
03610
AfMQNK
NH3-N
13.710
12.300
10.600
9.730
B.400
7.50C
8.300
J.600
5.50C
a. 200
11.500
11. COO
9.ZOC
8.800
5. 300
7.600
9.100
9.400
5.300
9.0CO
10.00C
9.400
13.100
a. ceo
5.70C
5 . eoc
4.70"
5.400
3.100
9.500
11.200
9. SCO
11.500
12.30C
15.100
12.30C
11.100
13.200
11.20P
"0653
SOLP04-T
P04
MC/L
15.00
15.30
10.80
10.00
7.90
7.00
7.no
4.6C
7.10
12.00
13.20
21.50
26.80
21.40
12.70
19.70
22.50
21.90
8. 31
20.50
19.On
17.50
17.60
12.60
B.30
8.50
7.70
8.20
4.-SO
14.00
11.10
11.20
15.5C
11.7r
21.80
16.50
n.or
21.BO
14.3C
00940
CHLORIDE
CL
"0/1
65
61
65
77
70
54
85
47
85
102
96
77
89
55
70
66
62
51
71
64
58
65
61
58
62
64
78
73
98
67
95
«2
170
84
95
68
92
SAINT PAUL
I11CR050
Figure 13. STORE! Retrieval of Hourly Sewage Analyses
19
-------
2G0010
PHOLEAI CFCEK fiECULOTOn
ZS MrNNELSdTfl
UPPER MISSISSIPPI
5flIIMT FflUL
o
ED
o
in
on
on
O
a
o
o
r>
o
a
o
o
o
o
o
CM
O
o
d
a
LO
o
a
o
o
o
o
o
o (j
LO
30
SO
60
70
80
STRRTING DRIF; rs
TIME DflYS
G8 3 18
Figure 14. Machine Plot of Hourly COD Data
20
-------
COD ANALYSIS S.W. INTERCEPTOR-August - September '68
40
600
Parts Per Million
1000
Figure 15. Frequency Oistributions-COD at Minneapolis Southwest Interceptor
MPLS. S.W. METER-August - September '68-25-30 Observations/Hr.
1000
900
12 1 2 3456 7 8 9 10 11 12 1 23 45 6 1 8 9 10 11 12
Figure 16. Minimum, Maximum, and Mean-24 Hour COD at Minneapolis Southwest Interceptor
21
-------
NOTE: Data includes wet and dry weather
flows. Flow data available to calculate
pounds per hour.
Number of
occurrences
Figure 17. Three-Dimensional Plot of 24-Hour COD Data for Treatment Plant Raw Wastewater
MINNEAPOLIS - SAINT PAUL SANITARY DISTRICT
D»TEOF MADIIM:
LOCATION
Hi
§
0
5
1
2
1
2
1
2
I
2
1
1
2
1
1
1
2
1
?
' '
' *
STATION NO
,
44-
'
ANALYUS
~
:
1
o
Z Ml
< <
u O
•
O 0
0»Y OF TEAR
1... At'
44.
£
1
L
3
3
SEWAGE FLOW METER
1 2
1
2
|
1
345
3
1 1
| i
, j
4
1
j ,
, ,
S
: 1
READING
READINGS BY HOUR
678
6
7
,
I i
i
9 10 11
9
10
1
f
,
, .
11
1 1
, .
rtst
12
[
. ,
.
12
.
Figure 18. Sewage Flow Meter Reading Key Sheet
22
-------
the rainfall of August, 1968, plotted to the right,
indicating heavier loadings. The months of June in
1967 and 1968 had 7.53 in. and 6.78 in. respectively
and the frequency distributions nearly coincide. Figure
19, shows the comparison of COD loadings for May
and August of 1967 with 1968.
COD-PLANT RAW WASTEWATER,
180
160
20
100
40 60 80
1000's LBS./HR.
MAY 1967 •••• AUGUST 1967 •
MAY 1968 AUGUST 1968— _ .
Figure 19. Comparison of Treatment Plant
Loadings by Month.
120
COD
The effect of rainfall and runoff on the combined
sewer system is readily observable on the flow meter
charts for Minneapolis Meters. Figure 20, shows the
effect on the Minneapolis N.W. Interceptor of 0.26 in.
of rain which occurred on the evening of September
24, 1969. The time during which combined runoff
water and sewage is flowing through the meters can
easily be identified and punched onto cards. The card
format used contains a card code number, the. station
coding, the year, month, day, and the hour combined
sewage flow starts and the day and hour it stops. With
this information, a separation of wet weather Techni-
con loading data at the treatment plant has been made
and distributions made on the wet weather data. The
same has been done at the Minneapolis Meters and
could be done for concentration data at all the other
regulators where sampling was carried on.
The frequency distribution of 274 hourly COD
loading values representing storm flow periods at the
Treatment Plant is compared in Figure 21, with the
frequency distribution of 4138 hourly COD loading
values representing both dry weather and storm flow
periods. In general, most of the COD plant loadings fall
within the same usual range during both dry weather
and storms. Similar dry weather-storm comparisons
were made for calculated hourly plant loadings of
Total Kjeldahl Nitrogen, Total Soluble Phosphate, and
Chloride ion. The results were similar to those seen for
COD. The data representing storm flow almost always
lay within the extremes of the dry weather data. The
frequency distributions for wet and dry weather
Kjeldahl Nitrogen loadings were very similar. Those
comparing wet and dry weather phosphate and chlor-
ide loadings suggested that during storm periods some
dilution of those constituents occurs.
Figure 20. Flow Meter Recorder Chart
Showing Effect of Rainfall on
the Minneapolis Northwest
Interceptor
23
-------
1000
ALL HOURS:
Observ. 4138
Interval
Max
Min.
Mean
S.D.
10.0
176.88*
1.08
43.56
17.96
800
RAIN HOURS:
Observ. 274
Interval 10.0
Max
Min
Mean
600
s.n.
176.88*
14.07
43.78
17.51
'This value is not
significant and
should have been
thrown out.
400
200
FREQUENCY DISTRIBUTION
NO. OF OBSERVATIONS ALL HOURS
NO. OF OBSERVATIONS RAIN HOURS
100
80
60
40
20
15.00 30.00
45.00 60.00 75.00 90.00 105.00 120.00 135.00 150.00
COD LOAD, 1000's of pounds per hour.
Figure 21. Plant Raw Wastewater COD Loadings-All Hours and Stormflow Hours
Concentration data for some specific rainfall events
were reviewed also. An example of some of this data is
presented in Figure 22.- A rainfall amount of 0.67 in.
occurring over a five hour period was measured at the
Weather Bureau gage in South Minneapolis on Septem-
ber 8, 1968, a Sunday. The flow data and COD
concentration data at the Minneapolis N.W. and S.W.
Meters for this day were compared to those occurring
on a dry Sunday two weeks earlier. The concentration
values are unaffected by the storm water in the system.
Similar studies of other concentration data as well as
COD data for other rains at other sampling stations was
inconclusive.
Very likely what is measured may be a combination
of three possibilities. In some parts of the sewer system
storm water is probably diluting wastewater while
elsewhere a first flush of sewer deposits may be
occurring and still elsewhere the stormwater may be
polluted about as much as the wastewater it mixes
with. This would especially seem likely at the Treat-
ment Plant where the arriving flow is such a composite
of wastewater and stormwater from such a large area.
Because the net effect of stormwater on the
strength of the combined flow at the treatment plant is
small, one must conclude that combined sewer over-
flows can affect the river. At least some of the
combined sewer overflows must be as polluted or more
polluted than the dry weather flow that is usually
diverted to the interceptor.
Some of the large mass of data accumulated on hour-
ly wastewater strength and flow have been analyzed. All
the data are, however, stored on magnetic tape and
punched cards and can be analyzed using data
processing techniques. Most of the data is also stored
with the FWQA STORET system. Hopefully important
use of this data is yet to come.
One immediate use of the data by the MSSD has
been in establishing bypass priorities for use in system
operation and flow routing when storm flow condi-
tions are such that interceptor capacity is limited.
24
-------
AUG. 25 I II (DRY DAY)
SEPT. 8
(RAIN DAY)
100
80
Q
U
60
40
20
0
HOOO
cc
Ul
o
o
o
o
o
o
800
600
400
200
0
100
80
O
5
g~ 60
O
u.
40
20
0
,1000
Z 800
O
H
a 600
Z
UJ
u
I 400
U
o
8 200
SW
111****
SW
NW
NW
Illlllllll
METER FLOWS
fill**1
METER COD's
O 111 miniiiimiiiiiiiiiiii I****"
METER FLOWS
METER COD's
•.,.•:.,..•:,,.••.,..
;';•;••:; 0.67" i;V:X
mini
linn
"""•iiiiiiiiiiiiinii
iiiiiniiiiii
"IIIIIIIIIII
iim
ilium
.25 «
u.
.50 I
10 12 14 16 18 20 22
TIME
Figure 22. Sewage Flow and COD Concentration Compared for a Dry Day and a Wet Day
24
25
-------
IV. INSTALLATION OF GAGING SYSTEM
In Sewer Work by Regulator Crew
The Sanitary District's consultants had, as part of
earlier studies, successfully used a gaging system
employing a pressure-conveying tube installed in the
sewer, an air-supply cylinder, a bubbler and a local
recorder. A similar approach was planned for use with
the Regulator Demonstration Program system with a
pressure transducer and telemetry replacing the local
chart recorder. The installation of bubbler elements
and pressure transmitting tubing in the sewers was
accomplished entirely by the Sanitary District's Regu-
lator Crew. Much of the work was done at night to
take advantage of low flow depths, making possible the
installation of most air discharge elements at or near
the sewer invert. Originally, all the elements were made
from 1/4" flexible copper tubing with several rows of
holes spaced at 90° intervals to provide multiple air
discharge ports. Some of these have since been replaced
with galvanized pipe elements.
Forty elements were installed at regulators, usually
with three per regulator: one upstream from the
regulator in the trunk sewer, one downstream from the
regulator in the outlet pipe carrying diverted flow to
the interceptor and one downstream from the diversion
dam in the storm water overflow pipe. Eight elements
were installed in the interceptors. Existing level moni-
toring elements were already in use at three interceptor
locations and three regulator locations. All elements
were terminated in a safe location in a manhole well
above maximum flow depths and 1/4" I.D. plastic
tubing used to transmit air pressure to the surface. The
Regulator Crew has since maintained the tubing and
elements, making repairs and replacements as
necessary.
Contract 664A
In order to avoid expensive construction in existing
congested urban locations and to provide a "clean"
location for electrical and mechanical equipment,
pre-cast concrete vaults, identical to those used by the
power and telephone industries, were specified as part
of Contract 664A. Fourteen vaults were installed near
regulators where revisions were to be made for control
and monitoring.
Seven pole-mounted cabinets (Figure 23) were
installed near interceptor monitoring stations where
less space for equipment was required. One vault was
used for interceptor monitoring equipment to serve a
station located within a city park where an above-
ground cabinet was aesthetically undesirable.
The concept of a sealed "umbilical" connection to
the hazardous environment of the sewer was devel-
oped. The facilities entering the sewer atmosphere were
restricted to non-hazardous low pressure air, oil hy-
draulic and low voltage direct current (five volts at very
low current).
Contract 664A also included the electrical service,
vault to stub pole telephone wires and bubbler panels
for flow depth measurement. The contract was com-
pleted during the summer of 1967. Early installation of
electric services facilitated later work under other
contracts and toy Sanitary District personnel. The
bubbler cabinets contained a high-pressure air mani-
fold, pressure reducing valve and bubbler, an electric
24-hour clock and a purge timer and a solenoid valve to
switch high-pressure purge air into the tubing to purge
the element once every 24 hours. Purge time is
adjustable and a 10 second purge has been used
satisfactorily. Originally compressed air cylinders were
used to supply bubbler air, but these were all replaced
by small compressor-tank outfits. This has eliminated
the problem of changing heavy cylinders periodically in
poorly accessible locations.
Rain Gages
The original rain gage network was comprised of
eight rain gages; a ninth rain gage was added during the
summer of 1969. The rain gages used are Belfort
Model 5915 Rainfall Transmitters. The principle of
operation is weight measurement, with the scale pan
movement transmitted mechanically to a single turn
precision potentiometer. The power supply for potenti-
ometer excitation is supplied from the telemetry
equipment and the voltage analog proportional to
rainfall equivalent weight is transmitted to the central
station. One of the gages is shown in Figure 24 with
collector and case removed to show the collector pail
and weighing mechanism.
Figure 23. Pole-mounted Cabinet
27
-------
Figure 24. Transmitting Weighing-type Rain Gage
Seven of the eight original rain gages constitute
separate remote stations. The eighth original gage is
connected to a river monitor station interrogation
matrix and did not require a separate station telemetry
installation.
Rain gage locations were chosen by the project
staff. Initial letters requesting cooperation regarding
installation of rain gages and telemetry equipment at
various municipal sites were sent during the summer of
1966. Among the requirements for a site were: (1) a
suitable rain gage location where representative rainfall
measurement could be made, (2) an indoor location
within reasonable distance from the gage site where
telemetry equipment could be installed and connected
to 110 volt AC power and telephone wires, (3) the
capability of installing wires between the gage and the
telemetry equipment, and (4) a location where vandal-
ism would not be anticipated. In general, locations
around the periphery of the Twin Cities were chosen so
that storm fronts could be observed as they moved
toward combined sewered areas. Surprisingly, it was
quite difficult to find sites meeting the general require-
ments on publicly owned facilities. An attempt was
made to site a gage on the plant grounds of large
corporation, but approval could never be obtained
from the various corporate departments and the
attempt was abandoned. The locations finally used
were the following: Minneapolis Water Treatment
Plant, St. Paul Water Department Reservoir, St. Paul
Water Department Pumping Station, University of
Minnesota Agricultural Plots, Suburban City Halls (2),
and Metropolitan Airports Commission Maintenance
Garage.
The ninth gage was installed at a regulator control
and monitoring station and connected to the spare
telemetry address there. This location is near the center
of the Twin Cities. The locations of the nine rain gages
are shown on Figure 25.
Installations of the eight initial gages were made by
a local electrical contractor. Direct burial cable and
conduit were used in connecting the rain gage to the
building in which the telemetry equipment was lo-
cated. At one site direct burial cable was replaced by
buried conduit after gophers chewed the insulation off
the wires. The ninth gage was installed by the Sanitary
District electricians.
The gages seemed susceptible to lightning damage
and several of the precision potentiometers were
damaged and needed replacement. Several standard
replacement potentiometers, which were specially
wound for Belforte, were purchased at a cost of about
$90 each. Later, a stock item potentiometer that could
be used with slight modifications was found to be
available for about $20, with no noticeable difference
in precision or linearity. Fuses were installed on all rain
gage circuits and several fuses subsequently have blown
while no damage to the potentiometer occurred, so
they may be a worthwhile addition.
A range of 0 to 5 volts was equivalent to 0 to 6
inches of precipitation. Transmission "noise" appeared
to be causing variations in scaled readings of as much as
± 0.02 inches. A change of 0.04 inches over a
five-minute interval would be equivalent to a rainfall
rate of 0.48 inches/hour, so this noise could produce
rainfall data unacceptable to the mathematical model.
Repeatability tests on the telemetry equipment indi-
cated that its performance was acceptable and that
improvements could not likely be made. Noise could
be introduced by induction in the cable connecting the
rain gage or by the leased telephone lines. Improve-
ments were made by changing the transmission sam-
pling procedure used at the central station to eliminate
the effect of any transient noise. Now 16 readings of
the A/D output are averaged. If any of the readings
differ by more than one per cent from the first reading,
a new set of 16 readings are taken.
To further insure accurate rainfall data the collector
area of the rain gages was increased by a factor of three
by installation of a rain amplifier. Each increment of
rainfall therefore became equivalent to three times as
large a voltage increment as before and any noise
reduced in significance by 1/3. The rain gage data is
now accurate to the nearest 0.01 inch. The rainfall
collector buckets must be emptied more often, but this
has not been intolerable, and a method of remote
control emptying will likely be devised for future use.
The use of remote transmitting rain gages has
elicited interest from meteorologists in the area. The
State Climatologist, along with University of Minnesota
scientists, is studying the micro-meteorology of the
metropolitan area, and we are supplying him with data
from the rain gage system. Data has also been supplied
to the public works departments of the suburbs in
which gages are installed.
28
-------
NJ
(O
(D
ISJ
CJ1
NSSD REMOTE TRANSMITTING RAIN GAUGES
UM ST. PAUL CAMPUS
VAD LAKE VAONAIS
HAZ HAZEL PARK
FIR WEST SIDE
VIL GOLDEN VALLEY
EON EOINA
MAC AIRPORT
WTP MPLS. WATER WORKS
EUS EUSTIS
cc
CO
CD
H
3
3
Co
I
3D
BJ
-r .^v
r4-' •' '/ L ~ r">>* I T'
Vadnais 7-'' / -T jBireh- \ -^
Heights IGem Lake/ White Bear P^^iwtJOdV I r
A / Lake , h '-r
/ 1 If, ^ '
H Htop I
' "1 f"]
ITPcoiutnb,. J-rl
-------
V. REGULATOR MODIFICATIONS
The largest design and construction job involved in
the project was the revisions to the regulators. In
addition to converting existing gates to remotely
controlled power operation and installation of a
variable position diversion device, there were some
modifications to existing concrete structures to im-
prove flow conditions. Before plans and specifications
were completed, the Regulator Crew assisted the
consultant's engineer and a testing laboratory in
obtaining measurements of the forces required to move
existing gates. Field measurements were taken and all
the sites were inspected.
Contract 677
Contract 677 covered modifications to existing
regulators and sewer structures. Work done included
demolition of some existing structures and removal of
existing equipment, concrete construction of new
sewer structures and constructing and providing regu-
lator gates and adjustable diversion devices with neces-
sary fluid power systems for their operation. The
adjustable diversion devices at all locations where
required were inflatable dams under the base bid.
Alternate bids were permitted for the adjustable
diversion devices and either fabricated aluminum crest-
type gates or wood fabricated rotating vane-type gates
could have been bid. No bid on either alternate was
received. Nineteen locations were involved in the
work—eight in St. Paul and eleven in Minneapolis.
Figure 26 shows the locations of modified regulators.
At the Camden regulator, (CAM), a side overflow
weir was raised 1.1 feet and a new gate structure was
constructed at the entrance to the outlet pipe leading
to the interceptor. Two new fabricated wooden gates
with oil hydraulic cyclinder operators were installed in
the new chamber on the dry weather outlet and on the
storm water bypass. Because the sewers have little
cover there, an above-ground structure was necessary
to enclose the hydraulic gate operators. The location is
on Minneapolis Park Board property so an attractive
structure was designed and built.
At Portland and Washington and at Second and
Main, (PW and M2), the entire interceptors are diverted
by master regulators. At these two locations and at
26th and Seabury, 39th and Minnehaha, 38th and
Edmund, and Oak Street (SEA, M39, E38 and OAK,
respectively), hydraulic operators were installed on the
regulator gates and a rubberized fabric "Fabridam"
replaced the existing weir.
At 31st Avenue N.E. and Randolph Street N.E.
(RAN), a deformed corregated metal pipe had been
placed in the invert of the existing trunk sewer as part
of a Minneapolis sewer separation project. A Fabridam
was installed here anyway since, under storm condi-
tions, combined flow exceeding the capacity of the
corrugated metal pipe occurred, and higher head
conditions would increase diversion capacity.
Flow from each of the Minneapolis Interceptors is
metered, primarily for purposes of cost apportionment.
At all three interceptor meter stations, flow is mea-
sured by parallel Venturi meters. Two meters are in use
at all times at the Southwest and East Interceptors. At
the Northwest, prior to the Regulator Demonstration
Program, only one Venturi meter was used. At each of
the three structures overflow relief to the River is
provided as well as by-pass gates. There were gates in
the channels upstream from the Venturi meters at the
Southwest and East also.
Bypass gates were installed in 1938 and were ex-
tremely difficult to operate. A full day was required to
operate one gate using a large crew. As part of Con-
tract 677, new bypass gates with hydraulic cylinder
operators were installed. The gates were fabricated of
creosoted wood with teflon bearing plates to decrease
friction of the gate against the new aluminum guide
channels.
New gates and guide channels were installed in the
channel just upstream of the Northwest meters and the
second, inactive Venturi meter was prepared for
service. Figure 27 shows the two new gates at the
Northwest Interceptor. Now more flow can be con-
veyed from the Northwest Interceptor to the inverted
siphon river crossing during storm flow periods if there
is capacity available downstream in the Joint Inter-
ceptor.
Figure 27. New Venturi Meter Control Gates at
Minneapolis Northwest Interceptor
Modifications to regulators in St. Paul, in general,
were more extensive. At the Rice-Rondo and Trout
Brook regulators (RR and TB), the only modifications
were the hydraulic gate operator and the Fabridam.
The regulator gate and Fabridam at the Trout Brook
31
-------
00
CtM
PW
SEA
M39
E38
RAN
N2
OAK
RR
TB
CONTROL AND MONITORING REGULATORS
CAMDEN PARK
PORTLAND AND WASHINGTON
26TH AND SEABURY
39TH AND MINNEHAHA
38TH AND EDMUND
3IST AND RANDOLPH
2ND AND MAIN
OAK STREET
RICE AND RONDO
TROUT BROOK
PHALEN CREEK
BELTLINE
OTIS AND MARSHALL
ST. PETER AND KELLOGG
EUSTIS - EAST
EUSTIS - WEST
^ Vadnais
-N ' Heights »Gem Lake/ White Bear
'\ V /\ / Lake
~
I 2 3 4MILES
-------
regulator are shown in Figure 28 and Figure 29. The
gate is 32 inches wide and has a stroke of 21 inches.
The Fabridam, installed in an old stone block sewer,
has a height of 75 inches above the concrete base dam
and is seen in this photograph from the back side,
looking upstream. The air inflation pipe, seen along the
left side of the photograph, disappears into the
concrete base dam and enters the Fabridam near the
center of its front attachment.
Figure 28. Hydraulic Power Operated Gate at the
Trout Brook Regulator
At the Phalen Creek regulator (PC) a new, enlarged
regulator access chamber was constructed and in it a
new gate and operator on the dry weather outlet was
installed. In the trunk sewer a Fabridam was installed.
A downstream secondary orifice regulator on the dry
weather outlet pipe was eliminated, so a great improve-
ment was made in the flow conveying structures. At
this particular regulator, the outlet capacity is small in
relationship to the flow arriving in the trunk sewer,
especially in the summer months. Before modifi-
cations, dry weather overflow must have frequently
occurred since, during 1969, high dry weather flow was
completely diverted only when the Fabridam was fully
inflated and a high head condition existed at the outlet
pipe entrance.
At the Beltline regulator (BL), a new manhole was
installed to permit installation of the Fabridam and the
existing diversion channel was widened to improve
flow conditions at the entrance to the outlet pipe. A
new fabricated wooden gate and hydraulic operator
were installed.
At the Otis and Marshall regulator (KD), a new
power-operated gate replaced an existing regulator
gate, the diversion tunnel was enlarged and a Fabridam
was installed in the trunk sewer. Figure 30 shows the
upstream side of the Fabridam in a view looking
downstream. The opening to the dry weather outlet is
visible on the right side of the Fabridam, which is 60
inches high. A power-operated gate is now used to
control the flow.
Figure 29. Inflated Fabridam Diversion Weir at the
Trout Brook Regulator
Figure 30. Inflated Fabridam and Dry Weather Outlet
at Otis and Marshall Regulator.
At the St. Peter-Kellogg regulator (SPK), a Fabri-
dam was installed and work done on a complicated
regulator structure to improve its capacity. This in-
33
-------
eluded lining 73 feet of previously unlined tunnel and
the removal of a cast orifice regulator. The Fabridam is
pictured in Figure 31.
Figure 31. Inflated Fabridam at St. Peter and Kellogg
Regulators
Figure 32. Hydraulic Power Units
Wabash and Cromwell regulator (EUE and EUW) is
a newly constructed regulator provided as part of a
Storm Water Drainage project in St. Paul. Here two
Fabridams were installed in each of two trunk sewers
and equipment added to permit remote control as well
as local automatic control of two existing pneumatic
cylinder operated gates.
Table 7 summarizes the Regulator Modifications.
Table 8 is a tabulation of old diversion weirs and
Fabridam data. In every instance, the fixed concrete
base upon which the Fabridam rests is no higher than
the old fixed height weir so that there is no uncon-
trolled loss of overall capacity to convey flow at any
regulator.
Figure 32 shows package hydraulic pump units,
commonly used in industry, which provide power for
the cylinders which operate gates. These units are all
the same size for easy replacement and maintenance.
As a consequence the smaller gates can be quickly
positioned and the larger gates move more slowly. Gate
travel times for half stroke vary from 10 seconds to 10
minutes. The slowest operating gates are the bypass
gates and Venturi channel gates at the Minneapolis
meter structures. The hydraulic pump units are instal-
led in the vaults serving the regulators and are
connected to the gates in the sewers by stainless steel
tubing.
Fabridam inflation is accomplished by small air
blowers. The blowers used have a rated capacity of
about 40 CFM when operated at 3000 rpm against two
psi outlet pressure. For uniformity, interchangeability
and ease of replacement and maintenance, all the
blowers are identical although different pulley sizes
were used at a few of the smallest dam locations to
reduce blower speed and delivery. Inflation times vary
from about four minutes to about eight minutes to
bring a completely deflated dam to full inflation
pressure. Most dams can be fully inflated in about five
minutes. Recommended maximum working inflation
pressures vary from 2.0 psi to 2.6 psi. The dams
become quite rigid at much lower inflation pressures
and under most flow conditions will function ade-
quately at pressures around 0.5 psi.
The Fabridams are attached to the walls of the
sewer and to the reinforced concrete base constructed
on the floor of the sewer by rows of 12" long anchor
bolts. Bolt diameter varies with the size of the dam;
3/4", 7/8" or 1" diameter bolts are used. The
specification required by Firestone for these bolts is
that the "anchorage develop the full strength of the
bolt." Apparently this is quite important.
Post-Construction Evaluation
In June, 1968, less than a month after installation,
one dam was badly torn during a rain storm when the
fabric pulled from the clamping apparatus and caught
the rushing flow of water like a sail. The center portion
of the fabric was ripped on both sides from front to
rear. If the original installation of the clamping
apparatus was good, and Firestone is quite confident
that the checking procedures used by the installer are
adequate, then the probable cause of this was a slight
34
-------
Table 7
TABULATION OF ADDITIONS AND MODIFICATIONS - CONTRACT 677
REGULATOR GATES
ADJUSTABLE DIVERSION DEVICE
Location
Camden
Portland &
Washington
26th & Seabury
N.W. Meter
S.W. Meter
River Crossing
Outlet
39th &
Minnehaha
38th &
Edmond
31st&
Randolph
2nd & Main
Oak
E. Meter
Rice & Rondo
Trout Brook
Phalen Creek
Belt Line
Otis&
Marshall
St. Peter &
Kellogg
Wabash &
Cromwell
Number
of
Gates
2
-
2
-
1
3
-
-
1
3
-
-
11
11
None
1
1
1
1
1
1
1
1
1
2
-
Type
of
Gates
New wood
New wood
Exist.
Exist.
Exist.
Exist.
New wood
New wood
Exist.
Exist.
Exist.
Exist.
Exist.
Exist
Required
Exist.
Exist.
Exist.
Exist.
Exist.
New wood
New wood
New wood
New wood
Exist.
Exist.
Size Cylinder
of Stroke
Opening (Inches)
4'0" x 5'0" 60"
5'-0%"x 4'0" 48"
28"
28"
28"
66"
7'-iy2"x7'0" 84"
7'-V/2"x7'0" 84"
78"
60"
60"
60"
16"
16"
This Location.
28"
16"
78"
Exist. 36"
21"
3'0" x 3'0" 33"
2'0" x 2'0" 24"
3'0" x 3'6" 42"
6'0" x 2'6" 72"
Exist. 36"
Exist. 36"
No.
of
Devices
None
1
1
None
None
None
1
1
1
1
1
None
1
1
1
1
1
1
2
-
Sewer
Section
Required
Arch.
Cir.
Required
Required
Required
Horseshoe
Cir.
Cir.
Horseshoe
Cir.
Required
Arch.
Arch.
Semi-
Elliptical
Transition
Semi-
Elliptical
Transition
Square
Square
Maxi-
mum
Width
This
7'-6"
8'-0"
This
This
This
11 '-0"
10'-0"
8'-6"
7'-0"
8'-0"
This
8'-0"
10'-0"
9'-3"
9'-0"
5'-11"
7'-6"
5'-0"
Max. Height
of Device
Above Invert
Location
5'-4"
5'-4"
Location
Location
Location
6'-10"
6'-8"
5'-8"
4'-8"
5'-4"
Location
5'-6"
7'-0"
6'-3"
7'-8"
4'-8"
5'-0"
3'-4"
35
-------
Table 8
TRUNK SEWER DIVERSION WEIR DATA
Location
Portland & Washington
26th & Seabury
39th & Minnehaha
38th & Edmund
31st & Randolph
2nd Bt Main
Oak Street
Eustis-1
Eustis-2
Rice & Rondo
Otis-Marshall
Phalen Creek
Trout Brook
Kellogg - St. Peter
Belt Line
Height of
Pre-Project
Diversion Weir
31 1V"
I /2
2'
3' 6"
2' 6"
12"
2' 2"
3'
-
-
-
r 8"
3'
r 8"
r 3"
"
Height of
Fixed Dam
3' 1%"
2'
11'/2"
2' 6"
12"
2' 2"
r 3%"
-
-
-
12"
1.1'
9"
r 3"
12"
Height of
Fabridam
2' 6"
3' 4"
5' 101/2"
4' 2"
4' 8"
2' 6'/i"
3' 7%"
5' 3%"
3' 6"
5' 6"
5' 0"
5' 1-7/8"
6' 3"
3' 5"
6' 8"
Section
Horseshoe 7'6" x 7'6"
Circular 8'0"
Horseshoe 10'6" x 10'3"
Circular 10'0"
Circular 8'6"
Horseshoe 7'0"x 7'0"
Circular 8'0"
Square 7'6"
Square 5'0"
Horseshoe 8'0" x 9'6"
Semi-elliptical 8'2yj"x 9'0"
Semi-elliptical 9'8-11/16"x
9'4-1/32"
Horseshoe 10'0" x 10'6"
Elliptical 5'8"x7'2"
Elliptical 9'0" x 11 '7"
vertical movement of anchor bolts having inadequate
anchorage which allowed the fabric to pull from the
clamping apparatus. The anchor bolts were all removed
and replaced before a new dam was installed. No
further problems with this particular installation have
occurred. During the spring and summer of 1969, three
additional Fabridams were damaged after being re-
leased from the clamping. Two were damaged to the
extent that replacement of the dam was necessary and
one could be repaired and reinstalled. Testing of the
anchor bolts confirmed that bolt withdrawal was likely
the cause of the failures. The three dams were
reinstalled during the spring of 1970 after all anchor
bolts had been replaced.
Some of the Fabridams have developed air leaks
that were not present immediately after installation.
Some of these were caused by a small movement under
the clamping bar. Repairs were made at two locations
where this had occurred, but both dams later failed
catastrophically, so it is safe to presume that the leaks
had developed because of anchor bolt movement.
Another leak that was fixed had originated at the
location where the inflation air enters the dam. During
installation, overly large bolt holes were cut in the
fabric for the flange bolts and the holes elongated
under fabric stress and produced leaks. This leak was
repaired in May, 1969, and the air leaking almost
stopped. Great care was taken on reinstallations to
insure that bolt holes were cleanly cut to the minimum
size so that this problem should not reoccur. The
installer provided by Imbertson and Associates, Inc. of
Burbank, California, engineers for the initial Fabridam
installations, is no longer employed by them.
Revisions to the clamping apparatus have been
suggested by Firestone and Imbertson that should
make installations somewhat less dependent on
strength of bolt anchorage. The revision would permit
clamping together the top and bottom clamping bars
separately from the anchorage of the clamping bars to
the sewer. This should provide more uniform clamping
pressures as well as reduce the likelihood that a small
movement of an anchor bolt would result in cata-
strophic damage to the fabric dam.
Inspection of several dams has revealed no discern-
able deterioration of the rubberized fabric material
after about two years of service in the sewers.
Although there have been some problems, the local
contractor and Firestone have been cooperative in
correcting problems at no cost to the Sanitary District.
The Sanitary District's Regulator Crew provided most
of the labor, under the supervision of the contractor
and the Imbertson installer, for the reinstallation of
three Fabridams in 1970, however. The experience
36
-------
gained by them has increased their ability to handle
maintenance of the Fabridams in the future.
The original installation of the Fabridams included
no provision for draining water from the interior of the
• dam. Water can enter through punctures or other
wounds to the fabric, through leaks or breaks in the
inflation pipe, by leaking through the attachment, and
by condensation. The first problem with water was
encountered following repair of a section of inflation
pipe that was torn out during a rain storm. After the
repair, the dam could be inflated, but not deflated,
because the water in the dam could rise in a vertical leg
of the inflation pipe until the column of water was
equal to the inflation pressure in the dam. Rather than
pierce the fabric to install a drain, a "Y" fitting was in-
stalled near the point where the inflation pipe enters
the concrete base under the Fabridam. With a plug in
the "Y" the dam can be inflated and then, when the
plug is removed, the air pressure in the dam forces the
water out. Several inflation-deflation cycles have been
sufficient to remove almost all the water from a dam
and restore deflation control. The extra fitting was
installed on all the Fabridam inflation pipes so de-
watering can easily be accomplished. This also permits
deflation of the dam without requiring operation of
the solenoid valve in the vault.
In addition to the blower, there are several other
devices installed in the vaults to control and protect
the Fabridams. Figure 33 shows the control valves and
piping at the blower. A one-inch normally closed
solenoid valve is used to control deflation. The valve
can be energized and opened locally or remotely via
the telemetry system. A one-half inch normally open
solenoid valve will cause the dam to deflate automati-
cally upon power failure to insure that, if control is
lost, the dam will be in the safest position. A
pneumatic check valve prevents air flow back through
the blower. The dam must be protected against
over-pressurization, apparently to avoid damage. A
spring-loaded pressure relief valve was provided to
allow air release when pressure exceeds a set limit.
These valves have not been very reliable. Once opened,
they tend to remain cracked open slightly and do not
always seal after pressure has been relieved. They
apparently do not provide fast enough pressure relief
either.
A final precautionary device to prevent over-
pressure is a water column whose height equals the
maximum allowable internal pressure. Ostensibly the
water column was intended to be a redundant device.
However, it was found that severe storms did
occasionally cause Fabridam pressure to be relieved
by the water column rather than by the mechanical
valve. Air entering at the base of the column forms a
piston which discharges most of the water from the top
of column quite suddenly, and then the dam deflates
Figure 33. Control Valves, Piping, and Blower for
Inflatable Dam
entirely. The dam cannot then be reinflated until the
column is again filled with water. This has happened
frequently on only a few dams but is a disadvantage
since post peak runoff often could be diverted if the
dam could be reinflated. On two locations where this
has occurred, a pressure switch has been installed
which, when pressure in the pipe terminated in the
water column exceeds a set value somewhat less than
the height of the water column, will energize and open
the solenoid valve until the pressure is relieved. Since
this more positive acting pressure relief measure has
been in use, several rainstorms have occurred during
which pressure relief was adequate and the column not
discharged. Pressure switches will be installed at other
locations where the water column emptying has re-
duced the effectiveness of the system. Another mea-
sure that could be taken would be to replace the 4-inch
diameter water column with a larger one through
which air could be discharged without causing the
column to be emptied.
A pressure sensing line runs inside the 1-1/2" or 2"
inflation pipe into the interior of the dam. This
pressure is supplied to a pressure switch installed in the
electrical circuit which operates the air blower. The
switch is set to open the circuit and stop blower
operation when pressure in the dam reaches a set limit.
The use of the tubing permits the blower to be
controlled by a near static air pressure because of the
large volume in the dam.
The original installation of the Fabridams did not
provide any means of determining remotely the inter-
nal pressure. Sanitary District personnel later obtained
37
-------
and installed pressure transducers which enable the
dam inflation pressure to be monitored. Small film
type potentiometers manufactured by Computer In-
struments Corporation were used and have been found
to be extremely reliable.
All the piping appurtenant to the inflatable dams is
PVC plastic. There has been no real disadvantage to its
use except that the portions installed in the sewer have
been damaged and torn out at two locations. This has
not recurred since reinstallation included protection
from debris carried by storm flow. The use of securely
anchored metal pipe would probably have made this
protection unnecessary. The inflation pipes enter the
concrete base dams on their downstream sides so that
they are protected by the dam except under severe
storm flow conditions. This also permitted the work to
be done on the dry side of the dam during installation.
Tank gages are used to measure gate position. Gate
movement produces a shaft rotation into a sealed
potentiometer which transduces gate position into an
analog voltage. Several of the gages have been removed
for cleaning and several potentiometers have been
replaced. The service is severe, however—several can be
submerged during rainstorms—and this maintenance
should probably be expected.
38
-------
VI. PROCESS CONTROL AND COMMUNICATIONS
EQUIPMENT
Processing System
Specifications were prepared by Pioneer Service and
Engineering Co. for "Installation of Air Pressure
Transducers, Telemetering and Supervisory Control
Equipment, and Computer-Contract 664B."
The system described in the specifications had to be
capable of collecting, processing and logging data on
sewer flow, gate status, rainfall and river quality. Data
from remote station instruments was to be input to a
computer, scaled to engineering units and logged both
as a printed record and in machine processible form.
Provision for controlling gates at remote stations was
also required. Computer time had to be shared for
logging treatment plant data. The computer had to be
capable of performing other computations during its
idle time in addition to the primary on-line functions.
In other words, the computer was to be provided with
foreground-background time sharing capabilities. The
requirements were essentially for a computer-based
supervisory control and telemetering system.
Specifically, the work under this Contract 664B
consisted of furnishing and installing 44 pressure
transducers, telemetering equipment for 38 remote
stations, supervisory control equipment, and a com-
puter with certain peripheral equipment. The con-
tractor was responsible for the compatibility and
operation of the complete system from pressure
transducer to and including the computer and com-
puter logging.
Bids were received from Badger Meter Manufac-
turing Company, The Bristol Company, General Elec-
tric Company, Automatic Electric Company, Control
Data Corporation and CAE Industries. The contract
was awarded to the low bidder. Badger Meter Manu-
facturing Company.
Certain optional alternates were bid and during
contract negotiations the following were approved:
1. Substitution of a Potter Chain printer for one
output typewriter.
2. Addition of 8000 words to the core memory to
increase its size to 16K.
3. Substitution of a disk file for a Random access
memory file.
Later by change order, two other items were added
to the contract. One was the inclusion of an operator's
panel for the out-plant sewer control system. This
panel would allow manual addressing, remote opera-
tion of gates, and monitoring of outplant data even
though the computer might be out of service. The
operator's panel was also of value for maintenance and
checkout purposes. The other item was the inclusion of
two DEC TC02 magnetic tape transports. This pro-
vided a more reliable system, backup mass memory in
case of problems in the disc file, and easy exchange of
manufacturer prepared programs. Both have proved to
be very valuable additions.
The digital computer used with the system is a
PDP-9 manufactured by the Digital Equipment Corpor-
ation. The PDP-9, pictured in Figure 34, is a single
address, 18-bit word length, parallel binary computer.
It is a completely self-contained machine, not requiring
special air conditioning or humidity control. The
central processor's core memory has a 1.0 micro
second cycle time. Originally the PDP-9 used had
16,384 words; the core memory size was later in-
creased to 24,576 words as part of an agreement
modifying the contractor's software obligations.
Figure 34. Digital Equipment
Computer
Corporation PDP-9
The standard PDP-9 I/O facilities include a 300
character-per-second paper tape reader, a 50 character-
per-second paper tape punch and a 10 character-per-
second console teleprinter. In addition, the Minne-
apolis-Saint Paul Sanitary District system has a high
speed disk, two magnetic tape drives, a line printer,
two additional console teletypewriters and a logging
teletype.
The magnetic tape system, called DECtape, em-
ployes small reels of 3/4", 1 mil mylar sandwich tape
with an information capacity of 2.7 x 106 bits per 260
foot reel.
The disk storage drive is a Control Data 853, shown
in Figure 35. This disk drive uses industry standard
removable disk packs. Each disk pack has six disks with
ten usable disk surfaces and 100 tracks per surface. The
total capacity per disk pack is 24,576,000 bits or
1,365,333 18-bit words. The maximum one-track
access time is 30 milliseconds.
39
-------
Figure 35. Control Data 853 Disk Drive
The line printer, Figure 36, is a Potter HSP-3502
Chain Printer having a printing speed of 315 lines per
minute. Its features include 132 character output per
line and reformattable page control.
Figure 36. Potter Line Printer
The teletypewriters on line with the system are
standard ASR 33 and ASR 35. Both types employ the
10 character-per-second output and compatible inter-
facing. The ASR 35 is simply a heavy duty model of
the ASR 33 which enjoys longer life and quieter
operation.
The controllers and special hardware requirements
were designed and built specifically for the Minnea-
polis-Saint Paul Sanitary District's system. The disk
controller, provided by Digital Logic Corporation as a
subcontractor of Badger Meter Manufacturing Com-
pany, caused serious problems for a while but since
being revised has been highly dependable. Other
interfacing was accomplished by Badger Meter. This
included the line printer handler, the operator's control
panel, digital displays, and strip chart recorders, all
mounted in a single interface cabinet. Figure 37, along
with the disk controller and the base station telemetry
equipment.
Figure 37. Interface Cabinet
The software requirements were to be met by the
systems library software developed by the computer
manufacturer, by special programs supplied by the
contractor, and Sanitary District prepared programs.
Digital Equipment Corporation's standard software
package for the PDP-9 is called Monitor V4B. Its
features include:
1. ASA FORTRAN IV
This compiler meets all ASA specifications and
standards.
2. MACRO-9
A basic assembler language comparable to BAL
(IBM's processors) or OSAS (CDC's processors)
3. PIP
Peripheral Interchange Program for manipulation of
data files and programs.
4. EDIT
Systems program used in program preparation and
modification.
5. DDT9
Dynamic Debugging Technique. A modified loader
which can debug a user's program while in opera-
tion.
6. LINKING LOADER
A program which loads any absolute or relocatable
user program and all called subroutines.
As special options, the POP system also has:
Extended Arithmetic Element
Memory Extension Control
40
-------
Power Failure Protect
Memory Protection
Automatic Priority Interrupt
In addition to its program library, the Sanitary
District has had access to all other services extended by
DECUS the Digital Equipment Corporation Users
Society. Project personnel have attended technical and
educational meetings held by DECUS. Additional
software for the PDP-9 is automatically received as it is
entered into or updated by the DECUS library.
Badger Meter Manufacturing Company was to have
provided an executive time-sharing routine, operator
Input/Output routines and other programs for trend
recorder selection and output, acquisition of outplant
data, and gate selection and control. The Sanitary
District was to supply the real time programs for
scanning, scaling data and logging data, obtaining the
data from contractor supplied subroutines.
The original FWPCA project termination date was
July 1, 1969, with the year of operation for evaluation
complete in 1968. When it was realized that Contract
664B would not be completed in time to start system
operation during 1968, the Sanitary District applied
for and received from the FWPCA a project time
extension of one year with no increase in project
funding. This meant that operation had to begin in the
spring of 1969. In January, 1969, the Project Engineer
felt that the best interests of the Sanitary District and
the FWPCA required that Badger Meter Manufacturing
Company's work toward meeting the requirements of
Contract 664B be interrupted so that the system could
be prepared for operation during the 1969 rainfall
season. The contractor's efforts at this time were
largely directed, through a subcontractor, toward
completion of the executive time-sharing software.
Provisions of the contract clearly permitted the Sani-
tary District to take this action. The contractor, with
the understanding that it would be provided a further
opportunity to complete its obligations when the
system's utilization was less critically needed, coopera-
ted by providing system maintenance and support.
Fortunately, almost all of the system was usable and
the Sanitary District was able to prepare programs in
time to operate in 1969. The system was operated as a
dedicated system and could not, while scanning, be
utilized further.
In May, 1969, negotiations began regarding com-
pletion of remaining contract obligations by Badger
Meter Manufacturing Company. By this time, operating
programs for the system had been prepared and most
of the system hardware had been successfully used.
The contractor proposed to complete all unfinished
hardware related work and have separate hardware
acceptance testing as the criterion for hardware accep-
tability. They also proposed that a soon-to-be-released
DEC foreground/background monitor system would be
better for the Sanitary District than the still not
completed executive time-sharing routines. All pro-
gramming by the Sanitary District since January had
utilized the current DEC monitor system so the use of
DEC's foreground/background monitor had definite
advantages. The proposal also included additional
hardware items, another I/O teletype with associated
logic and 8,000 words additional core memory, and
cash contribution toward writing the handlers and
making other modifications required to adapt the DEC
monitor for use on the Sanitary District system.
On October 15, 1969, an agreement between the
Minneapolis-Saint Paul Sanitary District and Badger
Meter Manufacturing Company was signed. Essentially
the agreement relieved the contractor of the software
obligations in Contract 664B in exchange for the
additional hardware and cash contribution toward
adaptation of the DEC foreground/background system.
Subsequently, the contractor returned, completed un-
finished hardware work and obtained complete hard-
ware acceptance. The Sanitary District received a
preliminary foreground/background monitor system
from DEC and had its systems consultant begin writing
necessary handlers. At the time of writing, June, 1970,
the general and complete foreground/background
monitor had not been released by DEC. This release is
anticipated soon. Based upon experience gained using
the preliminary version, it is expected that only a few
months will be required to implement the foreground/
background time-sharing system once the finished
version is received.
Leased Line System
The telemetry equipment provided under Contract
664B to permit information to be returned from the
outplant sites and control to be exercised from the
central station is dependent upon a system of leased
telephone lines as the medium of communication.
Fairly early in the project a meeting was held between
representatives of Northwestern Bell Telephone Com-
pany, Badger Meter Manufacturing Company and the
Sanitary District. It was agreed that seven circuits
leading to the central station would result in the best
application of the communications equipment. The
telephone company would in the leased lines incor-
porate high impedence bridges with sufficient gain
added to make the lines appear as an 8 db loss at 1,000
cps. The frequency spectrum to be used for the
communications system was 400 to 2700 cycles per
second. Figure 38 shows the leased line arrangement.
Northwestern Bell Telephone Company began work
on providing the leased lines in December, 1967. Many
of the circuit drops were made by January but not
without problems and much communication between
Sanitary District and telephone company personnel.
Many of the remote sites were underground vaults,
some at sites not easily described. The river monitors
41
-------
SEWER CONTROL SYSTEM
DATA POINTS
Figure 38. Schematic Diagram of Outplant Stations and Leased Line System
-------
were not in place by that time so completion of some
of the circuit legs had to be delayed. Rain gages were
located near public buildings and the locations of the
telemetry equipment located in any available space in
the nearby structures, which sometimes caused com-
plications. Street addresses were used to help locate the
various remote stations, but calls from telephone
company personnel asking questions such as, "Are you
sure you want this circuit terminated at the shoe
store?" were not uncommon.
After the circuits had been physically prepared,
they were tested for attenuation losses at various
frequencies and corrections were made where required.
Progress was somewhat hindered by a strike of tele-
phone company workers from April 18 to May 6,
1968. Since telephone company work preceded the
installation of communications equipment in the field
by Badger Meter Manufacturing Company, some equip-
ment connections were not made until the first part of
June, 1968, because the telephone company had to
extend their lines from one termination point to
another within reach of our equipment.
Installation charges for the seven leased line circuits
service 37 outplant sites were $1,000. The monthly
charge for the seven circuits is $548.
Communications System
Part of Contract 664B called for provision of a
communications system consisting of supervisory con-
trol and telemetering equipment. The supervisory
control equipment permits sequential selection, inter-
rogation, and verification of remote station transducer
signals for connection to telemetering channels and
also control of gates at remote stations in the system.
The telemetering equipment transmits data from the
remote stations to the central control station where
analog to digital interface conversion permits data
input to the computer.
The air pressure transducers provided are Robert-
shaw Model 115 Microsen Pressure Transmitters which,
by means of a bellows pressure element, convert air
pressure to an analogous 4 to 20 milliamp signal.
Forty-two pressure transducers in ranges of 0-40",
0-72", 0-120", and 0-144" of water are used, and one
spare in each range was supplied.
The supervisory control and telemetering equipment
is that manufactured by Noller Control Systems, a
wholly owned subsidiary of Badger Meter Manufactur-
ing Company.
The remote station telemetering equipment had to
transmit data from the pressure to current transducers
and from rain gages, gate position transducers and river
monitors. At the present time there are 139 unique
addresses from which information can be obtained.
Remote station addresses are three or four digit
octal numbers which in binary form a selection code
indicating which of the seven leased telephone lines is
to be accessed, which group of two transmitters the
central station will use, and which of three possible
states each of the two transmitters will be in. Table 9
shows the numbers of points of each description.
Table 9
NUMBER OF MEASUREMENT AND
CONTROL FUNCTIONS
1
Function Lo
Level Measurement— Interceptor Sewers . .
Level Measurement — Trunk Sewers . . . . .
Level Measurement— Outlets to Interceptors
Gate Positions and Controls
Rain Gages
River Quality Monitors
Alarms and Spares . ... -
Mo. of
cations
12
15
12
17
9
5
18
(a)
No. of
Points
12
18
12
34
9
30
24
139
(a) Total number of locations of telemetry equipment is 37
due to overlapping functions at certain stations.
The remote stations are of two types insofar as the
telemetry equipment is concerned. One type»has only a
single point such as an interceptor monitoring station
or a rain gage. The other type has up to eight points at
a single location, although most have only five. At the
single point stations only a single receiver and trans-
mitter are used to complete the unique point address
and produce a "confirmation" code back that verifies
selection of the proper point. The other stations
employ a pair of receivers and transmitters. Trans-
mitters are three-state frequency-shift devices, so the
pair arranged as a three by three matrix can produce
nine possible combinations. One combination is not
used, so a pair of tone transmitters and receivers can be
used for up to eight points. Data transmission is analog.
Transducers produce or have signals conditioned to 0
to 5 volt analogs, which are converted to pulses, and
then to a frequency-shift keying tone transmitter
signal. Central station receiving equipment reproduces
the direct current transducer signal. All components
are self-contained on circuit boards which are edge
mounted in card cages.
The Sanitary District has been able to maintain the
operation of the telemetry system without assistance
for over a year now. A short period of instruction was
provided by the contractor and initial checkout and
adjustments accomplished by him.
Some spare components have been bought to assist
in maintaining continuous operation from all outplant
stations. MSSD has not yet performed repair of any
components; components replaced because of prob-
lems have been sent to Badger Meter Manufacturing
Company for repairs.
MSSD has found that maintenance of the communi-
cations system involves more than the ability to
maintain one's own equipment. It has been helpful to
be able to define telephone line problems to some
extent when reporting such problems to the telephone
43
-------
company. All complaints regarding telephone lines are
handled at the closest exchange to the central station.
This has worked well since only a single person need be
contacted rather than someone at each of the many
telephone central offices involved in the system. Some
special equipment has been required to assist in
maintaining the communications system. Two fre-
quency selective voltmeters, two AC voltmeters and
two electronic counters, one each for the remote site
mobile maintenance vehicle and for the central station,
were obtained.
Maintenance of the system has been under the
direction of the Project Engineer who has also handled
the central station end of many calibration and
troubleshooting operations. The outplant end has been
handled by the electrician assigned to the project.
Other plant electricians have been trained, and the
maintenance electricians are now able to handle the
communications equipment alone.
Periodically, perhaps two or three times per year,
adjustments in tone transmitter and receiver sensitivity
levels may be required to maintain the proper levels
and operating margins. This is necessitated by seasonal
variations in the characteristics of the leased telephone
lines.
Calibration Procedures
The use of a computer with the data logging system
makes calibration simple. All calibrations are from the
input to the primary sensing element to the scaled
value in engineering units appearing on the printed
page. An example of a calibration is cited: Personnel
who have reached the remote site to be calibrated
contact someone at the central station via radio. At the
central station, the calibration program, named
CHKOUT for CHECKOUT, is called. This program
samples the A/D output and calculates the analog
voltage equivalent and prints this voltage on the I/O
teletype. At .the remote site, a hand-operated pressure
calibrator is used to apply various levels of pressure to
the pressure transducer. The corresponding voltage is
measured at the central station for each pressure.
Figure 39 illustrates the teletype commands used in
calibrating, with the CHKOUT program. A linear
scaling equation is then calculated from the pressure-
voltage data. The same technique is used for the
inflated dam pressure transducer. Voltages correspond-
ing to gate position are read also. With this data, which
can be obtained in a matter of minutes, four scaling
equations are calculated and entered into the appro-
priate reference list to complete calibration of the
control and monitoring station.
Other stations are calibrated similarly. Rain gages
are calibrated using one-inch equivalent weights. River
monitors are calibrated using the ambient reading
observed at the local readout meters and the zero point
for each parameter. All calibrations are periodically
checked to see if scaling factors need to be updated.
tC
MONITOR V4B
$E CHKOUT
ADDRESS-CXXXX)
1397
(Operator types in request for
CHKOUT program)
(Address 1307 is the regulator gate
at 39th and Minnehaha)
FACTORS-*XXXX.XXX XXXX.XXX)
(Factors used only when
scaled values are to be
reported)
1307 2.845 0.000
ADDRESS-(XXXX)
(0% open - transducer
output is 2.845 volts)
1307
FACTORS-CXXXX.XXX XXXX.XXX)
1307 2*845 0.000
ADDRESS-(XXXX)
1307
FACTORS-tXXXX.XXX XXXX.XXX)
1307 4-640 0.000
ADDRESS-(XXXX)
(100% open - transducer
output is 4.640 volts)
1307
FACTORS-fXXXX.XXX XXXX.XXX)
1307 4.640 0.000
ADDRESS-(XXXX)
Figure 39. Calibration of an Outplant Address using
the CHKOUT Program
Inplant Data Acquisition System
Unrelated to the Regulator Demonstration Program,
but included in Contract 664B, was computer capa-
bility and equipment to permit plant monitoring and
data logging for certain parameters in the sewage
treatment plant as a secondary priority function. This
includes a station located in the Plant Operators' office
44
-------
where plant data can be displayed and logged. The
system provided is hard-wire with all of the analog
signals wired directly to the interface equipment. One
combination of the three most significant bits of the
computer buffer register and the nine least significant
bits is used for address codes for the inplant data
inputs. The output buffer register is decoded using
binary to octal decoders to give 241 individual address
signals which control 241 reed relays which switch the
various analog signals into tne A/D converter. The
common input channel for the thermocouples js a
thermocouple wire referenced to a cold junction and
fed through a high gain amplifier and then to an A/D
converter. All other input signals are conditioned to a
standard DC signal before being switched into the
common input-channel to the A/D converter.
The hardware provided has been tested and ac-
cepted but little use of the plant data logging capabili-
ties has been made to date. It is expected that during
1971, after the new time-sharing monitor system has
been implemented, the inplant data logging programs
will be prepared.
The equipment located in the Plant Operator's
office-the ASR35 logging teletypewriter, the ASR33
I/O teletypewriter, and the inplant interface cabinet-is
shown in Figure 40. The strip chart recorders can be
assigned to either inplant or outplant points. Also
mounted in the cabinet are receivers used to present
telemetering data from the existing interceptor and
plant flow meters to the computer.
Figure 40. Plant Operator's Logging Equipment
45
-------
VII. RIVER MONITORING
Since there has only been intermittant sampling of
Mississippi River water in the metropolitan area at
frequencies varying between once per week and once
per month, with no consideration given to time of
occurrence of rainfall, it was felt that the effects of
sewer overflows were really unknown. To assist in
making an evaluation of the effects of combined sewer
and storm sewer overflows, five river quality monitors
were acquired and sited to intensively monitor the
urban stretch of the river most affected by sewer
overflow.
Site Selection
In selection of sites, several factors were taken into
consideration. First, the general locations had to be
chosen such that the intention, measuring the extent to
which the river is affected by overflows, could be
accomplished. A monitor upstream of most of the
metropolitan area, particularly upstream of any com-
bined sewer outfall, was necessary. About half-way
through the cities is the confluence of the Minnesota
River, which differs considerably in quality, with the
Mississippi River. To permit observation of the effect
on the river of the upstream 13 miles of storm and
combined sewer discharge, a monitor had to be located
upstream from the river junction. The Minnesota River
joins the Mississippi River in a section of rather quiet
flow, and there is no dam, waterfall or rapids to aid in
mixing. The two rivers appear to flow side by side in a
common channel for some distance. To permit obser-
vation of the effect of the Minnesota River on the
Mississippi River, a monitor was needed somewhat
downstream from a point where adequate mixing of the
two waters has occurred. Fortunately, there are only a
few sewer outfalls in the section of river where this
mixing takes place so this monitor location will
constitute a good measure of entering condition,
upstream from a substantial part of St. Paul's sewer
outfalls. The next general location needed is down-
stream from the remaining St. Paul outlets but upstream
of the Minneapolis-Saint Paul Sanitary District Treat-
ment Plant outfall.
The upstream monitor was sited at the Minneapolis
Water Treatment Plant which is located just north of
the Minneapolis city limits in the suburb of Fridley. A
location in the raw water intake pumping station there
was easily the best site available. The total annual
water diversion is over 28 billion gallons, averaging
about 77 MGD or 121 cfs. The river geometry at the
pumping station is such that this diversion should
provide a good representative sampling of the river.
To insure, as much as possible, that representative
samples of river water are provided to the river
monitor, the river was cross-sectioned and sampled
extensively at each of the other general locations. The
contours of the river channel were roughly determined
by soundings. Some measurements of current were
made but were probably not extremely reliable since
they were made with the current meter suspended
from a boat which was not always absolutely station-
ary. Samples were taken at two or three depths at
several stations across each cross-section and also at
any feasible pump-mounting site. Dissolved oxygen,
conductivity and temperature were measured immedi-
ately, and samples were also run through the Techni-
con Autoanalyzer in the Sanitary District laboratory.
The automatic analyzer determined concentrations of
chloride ion, ammonia-nitrogen, total Kjeldahl nitro-
gen, phosphate and urea. In each case the data
indicated that the proposed sampling sites would
provide representative samples.
Contract 666
Specifications were prepared for the Minneapolis-
Saint Paul Sanitary District by Pioneer Service and
Engineering Company to provide water quality instru-
mentation stations. Five river quality monitoring sta-
tions were acquired in 1968 under contract from
Fairchild Camera and Instrument Corporation of
Syosset, New York. Four were mounted in two-wheel
trailers, the type used for temporary construction
offices. The fifth was provided for permanent installa-
tion in the pumping station at the Minneapolis Water
Treatment Plant. The units measure pH, conductivity,
oxidation-reduction potential, temperature, chlorides
and dissolved oxygen. Fairchild Camera and Instru-
ment Corporation has since disposed of its Environ-
mental Systems Operation which is now Automated
Environmental Systems, Inc. of Woodbury, New York.
The trailers measure 8 feet by 22 feet and are
insulated and equipped with sinks, plumbing, electric
heating, ventilating, fluorescent lighting systems, cabi-
nets and a work counter. One sink is provided with a
small water reservoir and a pump as a clean water
source. The other sink is connected to the sample
plumbing and provides a water sample tap at the work
counter. The trailers have one door and no windows in
an attempt to make them as vandal-proof as possible.
The water analyzer unit itself, Figure 41, is quite
compact so a smaller trailer could have been used.
However, the dollar savings in getting a smaller trailer
would have been minimal.and it was felt that a larger
size would increase their later utility for expanded
sampling and analysis work.
The analyzer has local readout meters for each
parameter measured and two electrical signal outputs,
one 0 to 5 volt range for telemetry and one 0-50 mv.
range for local recording. The monitors were bought
without local recorders since a computer-based tele-
metry system was available. Since then, two multipoint
strip chart recorders have been acquired; one provides a
local recording station at the Minneapolis Water Treat-
ment Plant because of their interest, and the other
provides data recording backup in case of telemetry or
47
-------
telephone circuit problems at any of the monitors and
provides a portable recorder for more intensive record-
ing of readout data as an aid to instrument trouble
shooting.
External connections to the trailer plumbing are an
influent pipe and an effluent pipe.
Figure 41. River Quality Monitor Analyzer Unit
Installation, Operation and Maintenance
The analyzers were mounted in the trailers at the
Fairchild factory in New York, transported over the
road without incident and delivered to the Sanitary
District. Delivery of one of the trailers was delayed
because of its instability in high winds at highway
speeds, but trailers have been demonstrated to be fully
transportable without damage to the analyzer. The first
monitor which arrived was put into operation for a test
period on plant secondary effluent. The monitor
operated well there and some effects of rainfall on the
final effluent were recorded.
Peerless Dynaflow positive displacement submersi-
ble well pumps were purchased to provide water flow
to the trailers. The same pump is used by the FWPCA
on their river monitors here. The trailers were delivered
in May, 1968 and were installed immediately although
telemetering was not at that time ready. Figure 42
indicates the locations of the five river quality
monitors.
The pumps are housed in strainers fabricated from
sections of well screen. These strainers confer struc-
tural protection and keep the small screen over the
pump intake from being covered by large debris. Figure
43 shows one of the strainers. The trailers were moved
to their installation sites by commercial movers and
blocked by Sanitary District personnel. Blocking the
four covers of the trailers increased their stability,
allowed them to be closer to the ground and permitted
removal and storage of the wheels and tires which
could otherwise be stolen. Telephone and power
connections were made to the trailers and the pumps
and the appurtenant piping installed.
Figure 43. River Quality Monitor Pump Screen
The downstream monitor, RQ 1, pictured in Figure
44, was installed on a dike at an industrial molasses
terminal. The pump was mounted on a wooden pile
cluster dolphin used in mooring barges at the unloading
dock. Flexible polyethylene pipe was used to convey
the water to the trailer. From the water's edge to the
trailer, the pipe was covered by rubber insulation under
which electric heating tape was installed to prevent
freezing during the winter months if the pump should
be off for a short time. The insulated pipe was
protected from physical damage by a rigid pipe running
up the bank to the trailer. The effluent pipe was run
unprotected to the water's edge. This installation was
accomplished by personnel from the Sanitary District
maintenance shop.
Figure 44. Industrial Molasses River Monitor
48
-------
CO
*
RQI
RQ2
RQ3
RQ4
RQ5
MSSD RIVER QUALITY MONITORS
INDUSTRIAL MOLASSES
FARM BUREAU
HIGH BRIDGE
FT. SKEUING
MPLS. WATER WORKS
l
-------
Because of the inplant work load, the other four
installations were made by a local plumbing contractor.
Monitor RQ 2 was installed at a fertilizer terminal with
the pump mounted from a concrete cell dolphin used
for mooring barges at their unloading dock. Because of
a flood wall between the trailer and the river, the pipe
and power cable were run on the river bottom through
an outlet culvert leading into a storm water pumping
station discharge wet well and up to the trailer. The
pipe above water was again heated, insulated and
physically protected by a rigid pipe where exposed.
The effluent pipe from the trailer discharges to a storm
drain nearby.
Monitor RQ 3 was installed at a cooling water
intake pumping station serving Northern States Power
Company's High Bridge Power Plant. The trailer was
parked next to the pump building and the pump
suspended into the wet well channel behind a traveling
water screen. The pipe was protected outside the
building in the same manner as at the other monitors.
The monitor upstream from the Minnesota River
mouth, RQ 4, is located on City property, right-of-way
from a since-demolished highway bridge. Because of
the narrow width of the river and the barge traffic, the
Corps of Engineers would not permit the installation of
a pile cluster, even a small distance from the river's
edge. Therefore, the pump was mounted on the end of
a private marina boat slip. Because of the low elevation
of the trailer site and the questionable pump location,
no effort was made to heat, insulate or protect the pipe
other than by direct burial.
The pump serving river monitor RQ 5 is mounted
behind a traveling water screen in a pumping forebay at
the Water Treatment Plant. The analyzer was installed
in an old meter repair shop now being used only for
storage.
The primary operating difficulty has been the need
to replace pumps periodically. The pumps run continu-
ously and are subject to wear from silt and sand which
is often encountered in the water. The average life
expectancy appears to be about four to six months.
The pumps are repaired as they are replaced, but the
repairs are fairly expensive. Repairs that have been
made are replacement of motor and replacement of
worn rotor and stator. A few centrifugal pumps were
purchased and used, but proved even less reliable, so
now only the positive displacement pumps are
repaired.
On two occasions, freezing of water in the pipe has
caused problems. After a pump failure at RQ 4, new
pipes and power cable were run over the snow and ice
and the frozen pipe left until spring. We had no
freezing problems with unprotected pipes so long as
the water flow was continuous. On another occasion, a
power outage allowed ice to block a pipe.
RQ 4 had to be moved in October, 1968 when
flooding on the Minnesota River submerged the trailer
site. Three trailers, RQ 1, RQ 3 and RQ 4 had to be
moved in the spring of 1969 during the near record
flood on the Mississippi River. In each of these
instances the pumps and piping remained in the river
and only the trailers were moved. This permitted
speedy reinstallation after water levels receded.
Maintenance of the analyzer itself has been limited
to cleaning and calibration of sensors. Electronically
the equipment has been very reliable. During the
winter months when water temperatures are near
freezing, maintenance visits for cleaning of sensors and
calibration checking at intervals of two weeks were
adequate. Very little biological growth in the sample
cups and on the sensors occurs and, probably because
of the ice cover and absence of river traffic, much less
silt accumulates in the sample chamber. During the
summer months, however, maintenance visits twice
weekly were the rule, primarily to accomplish cleaning
of the sensors and the sample cups and drawers. With
the exception of the chloride parameter, it has been
fairly easy to keep the equipment in calibration, as
long as the sensor are kept clean. The chloride sensor is
installed in a baffled cup because the probe is velocity
sensitive. The baffled cup quite easily becomes fouled
by biological growths and by accumulations of sedi-
ment. This condition is usually followed by abnormally
high chloride ion readings. The chloride sensor is also
the most difficult to calibrate probably because of its
long thermal time constant and its velocity sensitivity.
It does work, however, as evidenced by the increase in
chloride concentrations observed when spring thaws
flush the heavy amounts of ice melting salt into the
river. Figure 45 is an example of this phenomenon.
Maintenance and calibration since the intial start-up
of the monitors has been performed almost exclusively
by part-time and full-time temporary student em-
ployees. Initial calibrations were performed by the
project's engineering staff with the assistance of the
manufacturers local service group. The project's engi-
neers then trained the student employees. Major
calibrations utilizing full-range standards are not be-
lieved to be frequently necessary. The usual procedure
is to measure the ambient status and make whatever
small adjustments necessary. Samples are taken and
titrated for DO and chloride ion in the trailer. A
portable pH meter and a portable conductivity meter
are laboratory calibrated and used as secondary cali-
bration standards. Temperature standard is a measure-
ment by an ordinary laboratory thermometer. Oxida-
tion-reduction potential is always calibrated using
standard solutions prepared in the laboratory and
carried to the trailer. When major calibrations are
performed, the only additions necessary are: zero DO
solution since the river is usually near saturation, high
50
-------
FEBRUARY 22, 1969
'
28
.
RIVES QUALITY MONITOR
AT THE MINNEAPOLIS
WATER INTAKE
30 mg/1 During Thaw
-
'
20 ' S
•• 2
. O
t K
, c
>j;r'
v *
_£*
.*£*
Dissolved Oxygen =12 MG/L
V
"A .
^•v
: "V
i 2 -. Chlorides (salt) concentration = '7"
16 'I H
: Z
HI
. . O
; o
i 0
12 i J"
; O
^»°'&
; o
8
4__
0— -r
. __
_
o
, . - * J
P4
~"9I
0
>M ;
k-i^—O-
^
jC***
^
^ •' '
y
^
jf
if-
i
': i
| 1
Li ,
! 1
i !
1 i
0.4
, 6PM j
i <
; : ' •"•'* •... :
"v ' ' ; • ' • •
• '*
i. \ . *-* • ^ '
"^ • :
i
i
i
ME
*^ 1
'. *"*>• : i '
i . . • _ "t • '^**)fc : ''
. ™ • : ' " •£t'j*r ' * — "~
i i i • l* , "&?•'.
\ Without Thaw 7 mg/1 !
I J a I
j I
1 3PM 12 N
»dnoD loisme
• •
Figure 45. Chloride Data from River Monitor on a Thawing Day
and low chloride standards, two or three pH buffer
solutions, two or three conductivity standards and ice.
Some of the parameters were acquired with dual
range readout to increase accuracy of transmission and
readout. These are DO, temperature and conductivity.
The ranges available are illustrated in Table 10 along
with the normal full range of each parameter observed.
Table 10
RIVER QUALITY MONITOR PARAMETERS
Parameter Instrument Range
Dissolved Oxygen 0-24 mg/1
0-12 mg/1
Temperature 30 - 80° F.
50 - 100° F.
Conductivity 0 - 600 umhos
0- 1200 umhos
Chloride 0-48 mg/1
pH 4-10
Oxidation-Reduction Potential —600 • +600 mv
Approximate
Extreme Values
Maximum Minimum
14.2 mg/1 5.8 mg/1
83.0° F.
620 umho
20.5 mg/1
8.7
585
32.0° F.
220 umho
4.2 mg/1
7.4
310
51
-------
Findings and Conclusions
River monitoring data acquired during and immedi-
ately after rainfall events of sufficient intensity to have
caused appreciable combined sewer overflow to the
river were closely scrutinized. The effort was made in
an attempt to evaluate the significance of the effect of
pollution slugs originating from combined and storm
sewer outlets in Minneapolis and Saint Paul on the
Mississippi River. It was thought that depletion in
dissolved oxygen concentration might occur because of
the BOD entering the river in storm and combined
sewer discharge. Because of the low chloride concen-
tration in the river water (5-10 mg/1), it was thought
that an increase in chloride concentration might be
observed.
During and after most rainfall events, no changes
were observed in any of the parameters measured at
any of the river monitoring stations. There were a few
instances when some reductions in DO were noted
following rain, but they were very slight in magnitude.
Diurnal variations in concentration of DO were often
in evidence so this, plus the possible effect of cloud
cover variability, complicates the situation and in-
creases the uncertainty that any slight change should
be attributed to pollution from combined or storm
sewer discharges.
Extremely low river flow conditions did not occur
during the summer of 1969 despite the fact that
rainfall for the Minneapolis-Saint Paul area was rela-
tively low. Table 11 shows monthly maximum, mean
and minimum Mississippi River flows at St. Paul for
April through November of 1969. The minimum
recorded discharge at the Mississippi River at St. Paul is
632 cfs. At more critically low river flows, the effects
of combined and storm sewer flows would be more
pronounced.
Table 11
MISSISSIPPI RIVER DISCHARGE
APRIL -NOVEMBER, 1969
UNOFFICIAL RECORDS FOR ST. PAUL
Mississippi River Discharge
at Saint Paul, cfs
Minimum
Day
April 32,400
May 20,600
June 10,500
July 9,600
August 3,400
September 2,700
October 3,400
November 3,700
In 1969, data from the river monitoring system
indicates that the dissolved oxygen resources of the
Mississippi River in Minneapolis and St. Paul were not
adversely affected by storm and combined sewer
discharges.
Bacteriological effects were likely more pro-
nounced, but the only data available are from periodic
grab samples that are completely uncoordinated with
rainfall events. Anderson (4) attempted to analyze river
coliform count and BOD statistically for recent years
in an attempt to gage the effect of improved regulator
surveillance and modifications. The limited number of
observations makes it difficult to base any significant
conclusions upon these data.
Mean
88,500
39,000
14,900
14,200
5,900
3,400
5,000
5,300
Maximum
Day
148,600
59,000
20,200
18,500
9,800
4,500
5,700
7,100
52
-------
VIII. COMPUTER ANALYSIS OF STORM RUNOFF
AND MATHEMATICAL MODEL OF THE REGU-
LATOR-INTERCEPTOR SYSTEM
In November, 1967, the Minneapolis-Saint Paul
Sanitary District entered into an agreement with the
University of Minnesota St. Anthony Falls Hydraulic
Laboratory to have prepared a mathematical model of
the existing interceptor sewer system. Work toward
development of this program has been reported in four
status reports (5, 6, 7, 8).
Verification of the program, modification for use on
the PDP-9 computer at the Sanitary District, and the
rainfall loss-rate analysis not complete at the time of
preparation of the earlier reports have all since been
completed. Verification of the program was accom-
plished using real data obtained from the outplant data
logging system during the summer of 1969.
The final report on this project entitlted "Mathe-
matical Model of Urban Storm Runoff and the
Interceptor Sewer System of the Minneapolis-Saint
Paul Sanitary District" has been submitted to the
Minneapolis-Saint Paul Sanitary District and is included
as Part II of this report.
53
-------
IX. UTtMAMUlM Ul- IHt SYSItlVI
Data Acquisition Programs
Data acquisition or scanning is under control of a
group of programs collectively named "RTIME", short
for REAL TIME. The RTIME routine is called up by
issuing a one line command from the I/O teletype. The
operator then initializes the computer internal clock by
typing in the hour of the day, the minute of the hour,
and the day of the year. The computer internal clock
then maintains the time, which is entered on all reports
and stored with all data. RTIME will initiate scanning,
either under control of the computer internal clock, or
by receiving manually initiated signals from the inter-
face control panel. As each address is scanned, an
electrical analog signal from the telemetry is digitized
and read into the computer which referring to the
appropriate list of scaling factors (Figure 46), stores it
in a file, and prepares a written report. At this time
there are five outplant scanning routines used in
RTIME. Normal scanning employs four:
1. RIVMON which scans the river quality monitors.
2. REGCTL which scans the regulator control and
monitoring stations.
3. INCPTR which scans the interceptor monitoring
stations.
4. RAIN RP which scans the rain gages.
PIP V78
>T TT - OK RAINRP LST
0 01 29000 0641 UM-RG .421 -.284
0 01 30030 1144 VAD-RG .449 -.431
0 01 31000 1141 HAZ-RG .460 -.520
0 01 32000 1644 FIR-RD .429 -.421
0 01 33000 1441 VIL-RG .437 -.435
0 01 34000 1641 EDN-RG .436 -.510
0 01 36000 1244 MAC-RG .420 -.416
001 41000 1511 WTP-RG .408 -.151
0 01 42000 0633 EUS-RG .477 -.384
0 00 77777 7777 END-TAP
Figure 46. Rain Report Scale Factor List
The fifth routine, ALARM, can be used to scan
extra addresses at almost all stations which are called
"alarm points." When any of these scanning routines
operate, the data is added to a file stored on the
magnetic disk. These files are usually allowed to
accumulate for 24 hours from 8:00 A.M. to 8:00 A.M.,
closed, and given a distinctive name be which they can
later be accessed. Figure 47 is an example of one of
these files that was printed on the line printer. The
names used are made up of 9 characters. For example
RISR70 140 is a name for a file of river quality
monitoring data (Rl) occurring in a sorted but un-
packed format (SR) for the period in the year 1970
(70) approximately 24 hours in length beginning at
8:00 A.M. on the 140th day of the year (140). Should
this file later be packed, a process by which its
magnetic storage requirements are reduced, it would be
renamed RISP70 140.
182 9 35 1426 CAM-TL 30.620
_J,8 2_1 JL_2_2__.1A2.6. CAMrJL 3 3 .045
182 12 22 1426 CAM-TL 32.970
._18_2_14._L2. 1.4 2 6 _ CAM- TL 3 2 . 3 74 ._
182 14 46 1426 CAM-TL 31.702
_1.8.2__L6._10.._1426 .CAM-TL --30.695 —
182 17 4 1426 CAM-TL 30.770
_1.8.Z_1.8 4 1.426 CAM-TL 30.832 —
182 19 20 1426 CAM-TL 32.672
__18-2..20_2.0 1426 CAM-TL- _... J3 .008 -
182 21 20 1426 CAM-TL 34.350
.-.182 22. 20_. 1426 - CAM-TL - -34.537 -
182 23 20 1426 CAM-TL 34.201
— 132_23 5.4 ._.. 1426 CAM-TL 33.754 .
183 0 4 1426 CAM-TL 35.432
183 018 ..1426 CAM-rlL 38.901--
183 0 28 1426 CAM-TL 42.855
.18.3.... 0...40— 1426-CAM-TL 43.862 -
i83 0 50 1426 C3M-TL 41.922
18.3 1 0._.1426 .CAM-TL 40.207
183 1 10 1426 CAM-TL 38.528
..._18_3.._L_20 1.426. CAM- TL 3 6. -73 8 -
183 1 30 1426 CAM-TL 35.880
183—1-40 1426 CAM-TL- 34.350-
183 151 1426 CAM-TL 33. 791
..— 1 83 2- — 1 142 6- CAM- T4_ 32 . 9 33
183 2 11 1426 CAM-TL 31.963
._..!& 3 — 2-2-1- -1426 CAM-TL 31.478
183 2 31 1426 CAM-TL 31.068
_183 — 2-4 l_.. 1-426 -CAM-TL- — -30 . 285 -
183 3 16 1426 CAM-TL 28.233
... 18.3 3-2-6 1 42-6- CAM-TL 28.494 -
183 3 36 1426 CAM-TL 27.189
183 3 46 1426 CAM-TI ?7 O4O.
183 3 56 1426 CAM-TL 27.152
— 18-3 4 — 6 1426 CAM-TL 26.-965-
183 4 17 1426 CAM-TL 27.040
18_3 4 27 1426 CAM TL 26 704
183 4 37 1426 CAM-TL 25.958
_.183 4-4.7—1426 :CAM-TL 26.406-
183 4 57 1426 CAM-TL 26.033
—1.8 J 5 .7—1426 -CAM-TL 25.883-
183 5 20 1426 CAM-TL 25.846
U3 5 30 1426 fAK-Tl 2S R3V
183 5 41 1426 CAM-TL 25.846
183. .6 11— 1426. CAM-TL 26.182
183 6 24 1426 CAM-TL 27.077
18J-_ 6-34_-1426~CAM-TL 27.077
183 644 1426 CAM-TL 27. 338
183 6-5-4 1-426- CAM-TL 27.450-
183 7 4 1426 CAM-TL 27.562
18J 7-15. ...1426 - CAM-TL 28. 159
183 7 28 1426 CAM-TL 28.308
._ 183 7 38 1426 CAM-TL 28.644
<•;•»• •er»;i^u»»-j-fc*r.-»«,sv»7^r* *?it " • -^rrv-vt' *•-*"'• •*i*-v»«vl.r" •,-•*— **«'~^"*
182 9 36 1435 CAM-OL 28.548
....182 11 23. 1435-. CAM-OL 30.407
182 12 23 1435 CAM-OL 30. 152
Figure 47. Data File as Stored on the Disk
55
-------
The River Quality Monitor Report available immedi-
ately after scanning is shown in Figure 48.
The Regulator Control Report is shown in Figure
49. The station names used are familiar to the
operating personnel and tables are available to assist a
less familiar person in interpreting the data. Camden
Park through Oak Street are regulators in Minneapolis,
and Rice & Rondo through Eustis-West are regulators
in Saint Paul. Their locations are referenced in Figure
26. Trunk level (TL) is the depth of flow in the trunk
sewer just upstream of the regulator. Outlet level (OL)
is the depth of flow in the dry weather outlet pipe
which carries diverted flow from the regulator gate
controlling flow into the outlet pipe expressed as per
cent open. Storm gate (SG) is the inflation pressure
inside the Fabridam. At Camden Park regulator, where
there is no Fabridam, SG represents per cent open for a
bypass gate.
The Interceptor Report is shown in Figure 50. The
names used refer to the locations of the monitoring
stations, which are referenced on Figure 51. Level is
depth of flow in the interceptor in inches. The section
sizes and shapes are referenced in Table 12. Bypass
gates at the ends of the three Minneapolis interceptors
are remote controlled and their positions monitored.
The other two gate positions on the report refer to
gates at the entrances to the two parallel Venturi
meters on the Northwest interceptor. Usually only one
meter is in use, but during rain periods the second gate
can be remotely opened and two meters used. This is
accomplished automatically when flow through the
other Venturi meter exceeds a set point value. The
flow rates in MGD and levels at Bypass and Headhouse
are actually inplant data received from old metering
and telemetry equipment in use before the Regulator
Demonstration Program.
The Rain Report is shown in Figure 52. The rain
gage names used represent their locations and are
referenced on Figure 25. In addition to the data from
MINNEAPOLIS-ST. PAUL SANITARY DISTRICT
RIVER QUALITY MONITOR REPORT
YEAR 70 DAY OF YEAR 164 TIME 20 11
INDUSTRIAL MOL (RQM1)
FARM BUREAU (RQM2)
HIGH BRIDGE (RQM3)
FT. SNELLING (RQM4)
MPLS. WATER WORKS (RQM5)
CL
11.5
10.7
10.6
7.9
2.1
COND
492.0
476.7
409.9
288.0
198.5
DO
7.6
7.7
8.1
8.8
8.6
ORP
513.2
490.6
448.7
387.9
516.0
PH
8.2
7.9
8.3
8.0
TEMP
76.1
76.4
77.1
79.2
74.8
Figure 48. River Quality Monitor Report
56
-------
HIWHEAPOL3ES-ST. PAUL SARITAHY DISTRICT
REGULATOR CONTROL
SEWER SCAN
YEAR 1970 DAY 146 TIME 9 33
CAMDEH PARK
PORTLAND + WASHIHGTH
26TH + SEABURY
39TH 4- MIWHEHAHA
38TH •!• EDMUiJD
31ST 4- RANDOLPH
2ND + MAIN
OAK STREET
RICE + RONDO
TROUT BROOK
PHALEH CREEK
BELT LIKE
OTIS H- MARSHALL
ST. PETER + KELLOGG
EUSTIS - EAST
EUSTIS - WEST
TL
INCHES
29.1
16.4
13.7
20.1
16.7
17.5
28.1
17.0
3.S
2.6
44.7
1.7
10.4
0.9
2.4
1.8
01.
INCHES
29.2
39.7
2?.. 3
19.0
25.8
0.0
15.6
29.3
1.5
4.6
61.7
4.3
5.1
3.8
06
7. OPEN
92.5
95.6
85.9
91.5
91.4
77.8
94.1
102.2
98.0
104.2
101.4
97.5
100.1
SG
PS I •
-0.7
1.4
1.0
0.3
0.4
-0.3
1.6
1.4
0.2
0.1
0.6
0.1
0.5
0.5
0.1
-0.4
Figure 49. Regulator Control Report
WL
INCHES
-2.7
MINHEAPOLIS-ST. PAUL SANITARY DISTRICT
INTERCEPTOR REPORT
YEAR 1970 DAY 146 TIME 9 39
LEVEL BY-PASS GATE 1
GATE 2
MGD
19TH AVE 4- 2ND ST.
FRANKLIN + RIVER
44TH 4- HAWADAHA
10TH + MARSHALL
ROBLYN 4- FAIRVIEW
PINE 4- SPRUCE
ST. CLAIR * 7TH
STEWART + TUSCARORA
OTIS 4- MARSHALL
MPLS. NORTHWEST
MPLS. SOUTHWEST
MPLS. EAST
MPLS. TOTAL
BYPASS
HEADHOUSE
PLANT EAST
PLANT WEST
PLANT TOTAL
26.5
999.9
37.8
30.1
56.9
64.2
12.2
25.7
24.7
18.0
19.7
50.0
-0.2
2.8
999.9
0.8
0.8
99.4
999.9
44.9
30.8
35.7
111.5
122.6
8.3
130.9
Figure 50. Interceptor Report
57
-------
01
00
• INTERCEPTOR MONITORING STATIONS
112 19TH AVE. AND 2ND ST.
FR FRANKLIN AND RIVER RD.
NDA 44TH AND NAWADIHA BLVD.
NE2 10TH AND MARSHALL
JT4 ROBLYN AND FAIRVIEW
ITS PINE AND SPRUCE B
«71 ST. CLAIR AHD «. 7TH
W72 STEWART AND TUSCORQRA
MR OTIS AND MARSHALL
NW MPLS. NORTHWEST
-n S* MPLS. SOUTHWEST
ME KPLS. EAST
;; ,-. <
A - : {
r-'"'
HPLS.S.W.
INTERCEPTOR mil INTERCEPTOR
V $ST+? «7-
' 7-1S
" n L
-------
MINNEAPOLIS-ST. PAUL SANITARY DISTRICT
RAIN REPORT
YEAR 1970 DAY
PAST READNG DAY
284
284
TIME
TIME
15
14
51
51
ST. PAUL CAMPUS
LAKE VADNAIS
HAZEL PARK
WEST SIDE
GOLDEN VALLEY
EDINA
AIRPORT
MPLS. WATER WORKS
EUSTIS
Table 12
INTERCEPTOR LEVEL MEASUREMENT
LOCATIONS AND SIZES
Location Size
MINNEAPOLIS: 4'-6"
19th and 2nd Streets 4'-0" x 4'-6'
Nawadaha 3'-6" x 6'-0"
S. W. Meter 6'-0" x 6'-0'
Franklin 9'-0" x 9'-0"
N. W. Meter 10'-3" x 10'-3'
East Meter 6'-0" x 6'-0'
Tenth and Marshall 7'-0" x 7'-4'
PRESENT
READING
0.064
0.044
0.011
0.460
0.249
0.202
0.132
0.258
0.072
Figure 52.
PAST
READING
0.048
0.030
0.007
0.449
0.215
0.189
0.117
0.235
0.049
Rain Report
RATE
IN /MR
0.016
0.014
0.004
0.011
0.034
0.013
0.015
0.023
0.023
INCHES
TOPAV
0.037
0.055
0.014
0.03*
0.100
0.0*1
0.04?
0.109
0.054
SAINT PAUL:
Mississippi River Boulevard
Tuscarora
Fairview
Pine and Spruce
St. Clair
3'-9"x 5'-10
2'-6"x 6'-0"
3'-9"x5'-10"
the scan run, this report also contains the data from
the last prior scan, the rainfall rate during the
intervening period in inches per hour, and the total
accumulation since the file day began. The readings are
inches of water equivalent to the weight in the
collector bucket.
A routine called SCNFRQ for SCAN FREQUENCY
is a part of RTIME and can be used to vary scan rates.
The operator calls the routine by depressing the
appropriate switch on the operator's console. Any
interval between scans can be indicated in seconds; the
various routines can and often are scanned at different
rates. Figure 53 shows the teletype commands that
would be used to change scan frequencies from hourly
to rates commonly used during rainfall.
Gates and Fabridams can be contolled remotely
both under program control and manually through the
interface cabinet. Any address can be manually ac-
cessed and the analog voltage read so monitoring and
control remain possible even without the use of the
computer. The program by which control functions are
normally exercised is called GATCON for GATE
CONTROL. It can be manually initiated or called by
the computer internal clock. As presently constituted,
the program will access any gate or Fabridam whose
address is written in a list called SETGAT LST
(Figure 54),check the position of the gate or inflation
of the Fabridam, and brings the position or inflation
pressure into agreement with a set point that it also
reads from the list if the actual value differs from the
set point by prescribed limits. A GATE control report
is shown in Figure 55. The remark "INOPERABLE" is
printed if the gate position reading does not reach the
set point value ± 2% within a period of time specified
on SETGAT LST.
Modes of Operation
During dry weather the system is used for data
acquisition, data reduction and processing, and for
programming. Normally, scans run at hourly intervals.
The gate control program, GATCON, runs every three
hours to maintain all outlet gates 100% open and all
Fabridams inflated. While most of the Fabridams are
not effective in or required for diversion of dry
weather flow, they are better kept inflated because of
the rather lengthy time that would be required to
inflate 15 fully deflated dams—about 75 minutes—and
because they need be deflated only during very intense
rainstorms that do not occur frequently.
Each day at about 8:00 A.M. the data files,
RIVMON DLY, REGCTL DLY, INCPTR DLY, AND
RAINRP DLY, are renamed with a unique name
59
-------
MONITOR V49
SE RTIME
ENTER- 1-13 MN DAY
08 45 146
ENTER ID AND RATE-CX XXXXX)
ID PROGRAM
2 RIVh'R
3 REGTR
4 INCPT
5 RAIN •
6 ALARM
7 SETGT
8 SLUDGE
9 EXIT
ENTER ID AND RATE-CX XXXXX)
2 01800
OLD RATE WAS 3600
ENTER ID AND RATE-CX XXXXX)
3 00900
OLD RATE WAS 3609
ENTER ID AND RATE-CX XXXXX)
A 00600
OLD RATE WAS 3600
ENTER ID AND RATE-CX XXXXX)
5 00300
OLD PvATE WAS 3600
ENTER ID AND RATE-CX XXXXX)
PIP V78
>T
1
2
3
A
6
7
8
9
10
11
12
13
IB
19
20
21
22
23
24
31
32
33
35
36
38
39
77
TT CA)
000225
000227
000245
000247
000305
000307
000425
000427
000505
000507
000545
000547
000745
000747
001045
031047
001025
001027
001226
031307
001345
001347
001427
001547
001745
001747
007777
- OK SETGAT LST
M2-SG
M2-OG
PW-SG
PW-OG
OAK-SG
OAK-OG
SPK-SG
SPK-OG
RR-SG
RR-OG
TB-SG
TB-OG
KD-SG
KD-OG
BL-SG
BL-OG
PC-SG
POOG
NWI-G1
M39-OG
SEA-SG
SEA-OG
CAM-OG
RAN-SG
E38-SG
E38-OG
Figure 54. Fabridam and Gate
70 180
98 0.60
65 180
100 063
70 180
96 010
70 180
100 060
70 243
100 060
70 240
85 030
70 180
98 060
070 249
92 060
65 180
95 010
95 060
90 060
68 180
96 030
95 060
70 180
70 180
92 015
Control List
Figure 53. Use of SCNFRQ Routine to Control
Scanning Frequency
60
-------
MIHMEAPOLIF-ST. PAUL SANITARY DISTRICT
GATE CONTROL REPORT
YEAR 1970 DAY 146 TIME 9 45
GATE NO
1
2
3
4
6
7
8
9
10
11
12
13
18
19
20
21
22
23
24
31
32
33
35
36
38
39
ADDRESS
0225
0227
0245
0247
0305
0307
0425
0427
0505-
0507
0545
0547
0745
0747
1045
1047
1025
1027
1226
1307
1345
1347
1427
1547
1745
1747
MM EN ON 1C 7.
M2-SG
f/2-OG
PW-SG
PW-OG
OAK-SG
OAK-OG
SPK-SG
SPK-OG
RR-SG
RR-OG
TB-SG
TB-OG
KD-SG
KD-OG
BL-SG
BL-OG
PC-SG
PC-OG
NWI-Gl
M39-OG
SEA-SG
SEA-OG
CAM-OG
RAH-SG
E38-SG
E38-OG
OPEW
69
97
70
99
71
94
68
98
72
101
70
84
72
98
77
90
67
95
94
89
68
93
95
76
71
89
WAS
67
87
70
96
72
95
86
98
93
101
93
81
86
98
91
90
80
95
94
87
79
93
96
99
87
90
DESIRED
70
98
65
100
70
96
70
100
70
100
70
85
70
98
70
92
65
95
95
90
68
96
95
70
70
92
REMARKS
INOPERABLE
INOPERABLE
figure 55. Gate Control Report
describing the file's content and its date. The next scan
run will begin new daily files. A hard paper copy of
each file is then obtained from the line printer. The
data files are usually quickly scanned to look for data
indicating problems—either real problems or problems
in monitoring equipment. The scaling routines will
record a 9's field, 9999.9, in place of any figure outside
a normal range to make it easier to spot unusual or
abnormal data.
If the telemetry system does not successfully access
a specific address, it is noted by a failure of the address
to "confirm." Confirmation of an address occurs when
the outplant station returns to the base station a pair
of tones corresponding to those sent by the base
telemetry as part of the unique address. The failure of
an address to confirm is indicated by a 7's field,
7777.7, so it is easily recognized, and this failure does
not generate misleading data. The paper copies are
bound serially and retained for later reference.
The data is also reduced using a DATA ANALYSIS
program named DAIMALS. This program reads the
entire data file and, after calculations, prints the
address, the number of times the address was scanned,
the maximum observed value of the parameter at that
address and the time at which it occurred, the
minimum value and the time at which it occurred, the
mean value, the standard deviation, the number of
times that address did not confirm and the number of
values falling outside a set range. Figure 56 is an
example of a DANALS Report. These reports are also
bound serially and retained.
Renamed, closed data files remain on the magnetic
disk surface, "but must be removed so that adequate
room is available on the disk for new data files and
programs. Magnetic disk is too expensive to be used for
data storage and the high speed capabilities of the disk
are not required for data storage.
61
-------
MINNEAPOLIS-SAINT PAUL SANITARY DISTRICT
-P-*4*- A"AL*VIS RfMT
FOR 07/01/70
31
i
CJl
en
a
3
3
a.
3D
n
e
POINT
641
__11A^..
1441
J64V.-
. — 14 .4 I . .
1641 ..
1 2.4.4.. .
1SJ.L.
63J-.
._ . L.42A-.
1AJ5-
_. -1*27
14PS
Z4A_..
2Ji5-
i4i_
-245_
. UL46_
11SS
J347. .
1345
i ins
111 3 ..
1307
1305
NAME
UM-RC
HA2-AG
F 1 R-fiO
VIL-8C
EON-ftC
-HAX-_aC
..UIP-ao
EUS-RC
CJLM-.-U. .
-CAH.-OL-
CAM-.OG
r«N-pp
.. PU-Ti. -
._£U_-01.-.
__EU^)iU_
_-.PU.-SG.
SEAr U_
^F H-QL
_. SEA-OC.
SEfl-SC
B3S.-JL.
Miq-nr.
H39-SC
NO.
- -READ—
. -56- -
56
56
56
56
56
56
.56
56
54
54
54
5t
54
54
St..
. 54
.- 54 -
5^
54
—5V
54
54
. 54.
54
MAX.
0.4
0.6
1.0
0.7
0. 5
0 .-6
0-. 9
0.5
4J.*
39.3
99. 0
1 . 2
- 62 . »—-
- - 96,9 -
- 1.5 -
- . 59.5 -
4), 5--
90. 7
— -4- 2 -
82 9
27.4
114 5— -
1. 1
TIME
OAT -MR HN
>83
183
183
183
183
183
183
183
183
183
183
182
183
183
4-83-
183
183
1-83
183
1-8-J
— HJi-
183
182
7
3
-7-
5
3
7
4
4
0
0
0
1 8
1
1
3
6
0
1
7
A
1
-1^7-
19
47
43
48
40
24
14
2
27
40
40
18
5
11
11
56
12
19
j
5
-15
—31
32
- 6-
22
HIM.
0.5
0.3
0.5
0.2
0.4
o.*-
0.7
0.4
25.8
21.0
V4 .0
-1 .4
15.0
23.9
94. >
0.9
12.7
. ^j — ^_
37.7
0 1
14" 6
14.6
1 0*- 8-
0.5
TIME
OAT HR UN
182 18
182 23
182 23
182 23
182 23
182 23
182 11
182 17
18 3 5
18 3 5
18 3 3
182 21
183 6
183 4
182 18
183 2
183 5
' 18 3 5
182 14
183" 2
- —is) - 7
183 6
183 7
182 14
1 3
40
40
40
49
30
33
14
20
41
36
21
24
28
5
32
21
4 2
48
4 2
1 6" "
56
-16 ~
48
PAGE
MEAN
0.7
0.5
0.8
0.6
O.S
0.6
0.9
0.5
31. 2
26. 3
97.0
-1.0
26.-J
37.9
95. 7
1 .2
21.5
27 9
86.6
0 4
- 39 [
21.2
110.8
0.8
I
S.D
0
0
0
0
0
0
0
0
4
4
1
0
11
11
0
0
10
10
22
4
1
0
. 1
. 1
.2
.2
.0
. 1
.1
. 1
.7
.2
.2
.5
.8
.6
. 7
.1
.8
.1
3
3
.2
. 8
.2
NOT 1 NO.
CON ORG
0
0
D
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
.p
3
0
o
0
0
0
0
0
0
0
c
0
0
0
0
0
0
0
0
0
o
0
-------
If it is desired to retain a file, it can be stored on
magnetic tape or on paper tape. It can be stored in its
initial format or it can be packed. A packing routine
has been prepared which reduces the "space" require-
ment for data storage by about 60%. The magnetic
tape used is referred to as DECTAPE, and the
computer is equipped with two DECTAPE drives. A
dectape library of 128 reels is in use for program
backup storage as well as data storage. Each reel
contains 150,000 eighteen bit words on 260 feet of 3/4
inch magnetic tape. Punch paper tape in use is the 1
inch fan fold type. A paper tape library contains
programs backup and diagnostic and test programs.
Almost all data collected in the first year of
operation was stored on magnetic tape. It is impractical
to consider continuing this practice so, after inspec-
tion, verification that a paper copy exists, and com-
pletion of the data analysis program, some of the older
files have been deleted from magnetic storage. All files
containing data acquired during and immediately after
rainfalls have been retained, however, so further data
analysis could easily be accomplished.
Another data processing program that has been
developed is a data plotting routine named GRAPHP.
This program will plot data on the line printer from all
or any part of any particular file. The time axis
(horizontal) plotting interval and the data axis (verti-
cal) full scale value are selected by the operator, so the
plotting routine is quite flexible. Usually only the
rain-day data was plotted since it was carefully
scrutinized as part of the project evaluation. Figure 57,
Figure 58, and Figure 59 are examples of plotted data
from the rainfall occurring on July 1 and 2, 1970.
1 IS FULL SCALE
o •
line 20
22 o
112
0 0
181
1 0
IB]
) o
tti
Figure 57. Machine Plotted Rain Gage Data for July 1-2,1970
63
-------
100 IS FULL SCALE
HUE 20 o n o
OAT IA2 1*2
22 0 2] 0
182 182
00 10 20 30 40
181 i» 183 193 19;
Figure 58. Machine Plotted Interceptor Sewer Flow Depth Data for July 1-2,1970
130 IS FULL SOLE
fI ME 20 0 210
Dir |»2 1S2
00 10 20
la) 103 183
30 to
183 183
Figure 59. Machine Plotted Trunk Sewer Flow Depth for July 1-2,1970
64
-------
A second plotting program called OGIVE was
developed for our system. An ogive is a type of curve
used to graphically portray frequency distributions in a
cummulative manner. It shows the per cent of time or
frequency that a measurement of a certain magnitude
a
or greater occurs. This technique was used to portray
utilization of interceptors in studies carried out in
1960 (2). Figure 60 is an ogive prepared from 1960
data and Figure 61 is an ogive prepared from 1969 data
at the same location in the interceptor.
MAXIMUM STORM FLOW
4-23-40 4.3'
MAXIMUM DRY WEATHER FLOW
9-23-60 3.3'
6-0" X 6'-0" SECTION
MINNEAPOLIS EAST INTERCEPTOR
VENTURI MCTER I
30 aO ^ 60 80
%OF TIME DEPTH OF FLOW ECUALS OR EXCEft'S
Figure 60. Ogive-1960 Data for Minneapolis East Interceptor
OC1V6
PIUIOO ;OVERCD-«P«ll H70
POIKI' 7J6 KE-II PIPE MEICHT (INCHES!- 1> .0 NO. Of OSSERVtl IONS' 1*54
LMC? io tv _ 7.2.000
75.0
72.C
MAX. LEVEL
E_»M.rt IN INCHES Jb.O
72.0"
0.0
1 OF lint DEPTH OF FLOW EQUALS OR EXCEEDS
Figure 61. Ogive-1970 Data for Minneapolis East Interceptor
65
-------
Mounted in the interface cabinets in the computer
room and in the Plant Operator's office are eight
two-pen miniature strip chart recorders. Data addresses
assigned to these pens are recorded on the strip charts
moving at one inch per hour. The pens are updated at
8-second intervals, so if all 16 pens are assigned, each
address's data value is updated every 128 seconds. The
routine which is used to assign addresses to recorder
pens is named SEL REC for SELECTION-
RECORDER. Figure 62 shows the operator command
string at the teletype which is used to assign addresses
to the 16 recorder pens.
Table 13
SCANNING TIME REQUIREMENTS - MINUTES
tc
MONITOR V4B
SE RTIME
ENTER- HR MN DAY
08 50 146
ENTER RECDR AND POINT NO.-(XX XXXX)
01 1511
NEXT
02 1441
NEXT
03 1641
NEXT
04 1244
NEXT
99 (99 Exits Program)
Figure 62. Assigning Data Addresses to Recorder Pens
Using SELREC
The RTIME program can be operated with or
without the reports being concurrently typed out on
the logging teletype. When the report is typed, more
time is required to scan all points. Table 13 shows the
time required for each scan, with and without output
to the teletype.
RAINRP
IIMCPTR
RIVMON
REGCTL
TOTAL
With Without
Teletype Output Teletype Output
1 0.5
2 1.5
2 1.5
6 4.0
11
7.5
Normal operation of the scanning program, RTIME,
employs the disk as the file device. The flexibility of
the system was demonstrated during the early part of
the summer of 1969 when various equipment bugs led
to use of the scanning program in various ways. When
the disk was unavailable, a modified program was used
which wrote the data file on magnetic tape. When the
Dectapes were unavailable, a version of the scanning
program that used the ASR 33 TTY as the output
device enabled the teletype's paper tape punch to
record the data in processible form.
Modes of Operation
Operation of the system during rainfall events can
be accomplished with or without the use of the
mathematical model. Personnel familiar with the sy-
stem and its operation are usually on duty during
periods of rainfall. The mathematical model program
incorporates RTIME, and can handle the acquisition of
data. The model also produces reports, records data,
and makes recommendations for operator action. The
operator can then change gate positions either from the
teletype keyboard or manually from the operator's
panel. The model then checks the changed gate
positions with the current data and will show the
results of operator action. The model can alert the
operator when excessive flow at any point in the
interceptor system is expected.
Use of the model as an operator aid during a
rainstorm would likely be as follows. The operator has
loaded the model program, and it is scanning rain
gages. The model is requested to use the rain gage data
to calculate a runoff hydrograph anticipated at each
regulator, perform a diversion analysis at each regu-
lator, calculate volumes diverted to the interceptor and
to the river, and route the flows through the inter-
ceptor system, checking for excess flow at each inlet
and at other key places.
The operator can choose to have any or all of this
data printed on the line printer. Most likely, he would
first check the results of the diversion analysis to see if
any overflow to the river is predicted. With this
information he is alerted as to which trunk sewers will
be carrying the largest storm hydrographs and when
66
-------
the storm flow will be arriving. He can then check real
data at appropriate times to verify if Fabridam
deflation is required to prevent surcharging of the
trunk sewer.
The routing portion of the program will print the
flow hydrograph at any point in the interceptor system
where the predicted flow exceeds the capacity of the
pipe. Assume that in this storm, the model warns of
excessive flow at the Minneapolis East Interceptor
Meter Station. The operator, noting the time at which
this will occur, checks a bypass priority table, decides
where flow release is preferred, and tests the result of
lowering the Fabridam using the model.
After performing the diversion analysis and routing
with the changed regulator condition, the model no
longer predicts excessive flow, so the operator now
deflates the Fabridam.
By this time, the model has accumulated more
rainfall data and can update its predictions. The
operator can also check real time flow data from the
regulators and interceptors.
The system can also be operated manually, without
use of the model. The operator has assigned rain gages
to the trend recorders and can check rainfall accumu-
lation every two minutes. Depending upon the rain
gage results, there will be certain trunk sewers where
depth data should be checked. Operating experience
from 1969 has indicated which trunk sewers are likely
to carry large flow hydrographs. Trunk sewer level data
can be assigned to trend recorders so that locations
where Fabridam deflation might be required can be
watched. By watching trend recorders and data logging
reports, the operator can keep abreast of what is
happening and control gates and dams to accomplish
the best operation within limitations of the interceptor
system.
Operation of the system during 1969 was usually
less routine. The mathematical model had not been
calibrated and was not available for use. The trend
recorders were not usable either. Early operation then
consisted of insuring that data scans were run fre-
quently and checking rainfall accumulation, trunk
sewer level, interceptor level, and Fabridam pressure
data as they were logged on the logging teletype.
The only operator actions taken were to deflate
dams when the depth of flow in the trunk sewer
approached the sewer crown. This was not required
very often. During extremely heavy rainfall, flow
conditions at many regulators caused rapid increases in
Fabridam pressure causing them to deflate through the
water column. It is always preferable to relieve air in a
controlled manner using the deflation valve, but the
storm water hydrograph arrives very suddenly at some
regulators after intense thunderstorms. Trunk level data
could be obtained every six minutes, but usually only
about every ten minutes because rain gage and inter-
ceptor scans were also run. Therefore, it was often not
possible to note a rapidly rising trunk level and react to
it in time to prevent the overpressuring of the
Fabridam and its subsequent automatic deflation.
Better operation of Fabridams is now possible, and
fewer instances of uncontrolled deflation should occur.
Controlled deflation from the base station will be more
timely because of the use of the rapidly updated trend
recorders. Also, local pressure sensing switches have
been installed at several regulators to open the de-
flation valve and relieve pressure before it has increased
to the level of the water column.
This type of "automation" could be expanded
somewhat with more local control loops, but at this
time it is not necessary. The computer and supervisory
control system are also used at several key locations to
automatically control gates and Fabridams. The second
Venturi meter on the Minneapolis Northwest Inter-
ceptors is brought into use when flow through the
other meter exceeds a set point, and the Fabridam at
the Beltline Regulator in Saint Paul is deflated when
the flow depth measured at the head end of the dry
weather outlet pipe exceeds a set value. More gates and
dams could be automatically controlled in this manner.
67
-------
X. SYSTEM EFFECTIVENESS
Data accumulated in operation of the system during
rainfall events were analyzed to determine its effective-
ness in reducing combined sewer overflow pollution.
Effectiveness could be evaluated by several approaches.
One used was to enumerate the overflows prevented.
Another was the duration of overflow prevented, and a
third was the volume of overflow prevented. All three
approaches were used.
Three distinct bodies of data were analyzed. A
number of rainfall events occurring during the 1969
rainfall season were well described. This data was
analyzed over the winter season. Overflow reduction
from the 1970 spring thaw runoff was also evaluated.
The third group of data was rainfall events occurring
during April and May, 1970. Data from 1969 and 1970
rains were combined and seasonal figures were extra-
polated.
Rainfall Events Analyzed
The initial operating season used for evaluation of
the project was that of 1969. Considerable effort was
required to ready the system for operation and
acquisition of data, since full-time use of the system
was not acquired from the contractor until January,
1969. Scanning and data acquisition programs were
ready in time but the winter weather had caused
further complications. Heavy snowfall over most of
Minnesota resulted in predictions of record or near
record flood crests on the Mississippi River in Saint
Paul. In 1965 the greatest flood on record had
occurred and inundated the Sanitary District plant site;
there was a very strong possibility that this would
recur. Even flood levels much lower than predicted
would have submerged the access road to the plant and
also submerged the telephone cable leading to the
plant. It was certain then that the flood would
interrupt data acquisition for at least two weeks. With
project plans already behind schedule, and not wishing
to have to wait until 1970 to acquire operating data,
the Sanitary District moved the equipment to a
temporary location on March 25, 1969. Although the
flood had passed and plant operation had restarted on
April 28, 1969, the move back to the Sanitary District
offices was delayed until August 4, 1969, because of
the fear that equipment damage during a move might
prevent acquisition of data or confuse the contract
negotiations with Badger Meter Manufacturing
Company.
The move to the temporary project operating
headquarters had apparently caused some equipment
problems. These problems and other hardware bugs
caused some operating difficulties in the early part of
the summer, and some rains occurred for which no
data was obtained. The flexibility of the system
allowed data acquisition to continue through some
hardware problems, which were all finally resolved on
July 10, 1969.
Unfortunately, the rainfall occurring during the
spring and summer of 1969 was less than normal, so
there were fewer opportunities to operate the system
and obtain operating data. May through September
rainfall was only 58 per cent of normal. August and
September were particularly dry months. Precipitation
data for the last five years compared with the Weather
Bureau Climatological Standard Normals in Table 14
shows that 1969 was the driest year of recent record.
Table 15 shows 1969 monthly precipitation data in
comparison with the normal for each month. All 1969
precipitation amounts are those measured by the U.S.
Weather Bureau's official gage in Minneapolis. Because
of problems caused by the flood related move, opera-
tion in April was incomplete, so the months of
operation were May through November. Only October
had higher than normal precipitation.
Because of the importance of acquiring data and the
shortage of manpower, extensive analysis of data
obtained was delayed until the conclusion of the
rainfall season. A review of all data obtained led to the
selection of sixteen rainfall events to be studied further
and used for the project evaluation. The first of these
sixteen rains occurred on April 24 and the last on
November 16. During this period, 13.00 inches fell at
the Weather Bureau Station and 7.46 inches of this fell
during the 16 rainfall events selected for analysis.
Therefore, the evaluation is based upon 56 per cent of
the rainfall occurring in this period, certainly a
representative sampling. Table 16 shows total precipi-
tation recorded at the Sanitary District remote trans-
mitting rain gages and the Weather Bureau gage for the
sixteen rains. That rainfall amounts vary considerably
over the metropolitan area is apparent. Table 17 shows
maximum 60-minute intensities for each rain gage
during the sixteen storms. The official Weather Bureau
gage is reported only at clock hours, so the figure is not
the actual 60-minute maximum intensity. Table 18
shows maximum 30-minute intensities.
Data acquired during rainfall events occurring in
April and May, 1970, have been similarly analyzed.
Table 19 shows total precipitation recorded at the nine
system gages and at the official Weather Bureau gage.
The average precipitation in all system gages for these
rains varied from a low of 0.20 inches on April 15, to a
high of 1.49 inches on May 21 and 22. Rains varied
from being nearly uniformly distributed (May 27,
A.M.) to having a ten to one ratio in precipitation
totals (April 28). Great diversity in areal distribution of
rainfall such as occurred on April 28 are typical of
thunderstorms. Uniformly distributed rains are more of
a frontal nature. It is interesting to note, in these rains
and the sixteen 1969 rains analyzed, that areal di-
versity in rainfall amount is much more common than
uniform or near uniform distribution. Of the 27 rains
mentioned, only 8 had maximum gage to minimum
gage ratios less than 2 to 1.
69
-------
Table 14
HISTORICAL PRECIPITATION DATA
May-September Per Cent Total Annual Per Cent
Year Precipitations, Inches of Normal Precipitation, Inches of Normal
Normal* 16.07 100 24.78 100
1969 9.34 58 19.39 78
1968 23.89 149 37.93 153
1967 12.92 80 25.44 103
1966 13.53 84 24.34 98
1965 25.50 159 39.94 161
"Climatological Standard Normals (1931-1960)
Table 15
1969 MONTHLY PRECIPITATION DATA
Month Normal*, Inches 1969, Inches Per Cent of Normal
January .70 2.05 293
February .78 .31 40
March 1.53 .90 59
April 1.85 1.55 84
May 3.19 1.98 62
June 4.00 2.93 73
July 3.27 2.95 90
August 3.18 .99 31
September 2.43 .49 20
October 1.59 2.53 159
November 1.40 .65 46
December .86 2.06 240
TOTAL 24.78 19.39 78
MAY-SEPT. TOTAL 16.07 9.34 58
APRIL-NOV.TOTAL 20.91 14.07 67
Table 16.
TOTAL RAINFALL AT EACH GAGE - 16 1969 RAINS
Day
116
141
148
173
176
196
205
207
211
218
242
267
274
285
303
320
Date
4-26
5-21
5-28
622
6-25
7-15
7-24
7-26
7-30
G- 6
8-30
9-24
10-1
10-12
10-30
11-16
US
W.B.
.82
.18
.24
.80
.90
.75
.25
T
.42
.71
.24
.28
.14
.69
.79
.05
Total Precipitation, In.
DM
.58
.23
0
.58
1.64
1.04
.42
.52
.60
.51
.22
.26
.06
.97
1.04
—
VAD
.54
.04
0
.67
2.02
.08
.12
1.52
.53
.56
—
.14
.05
1.30
1.00
.08
HAZ
.37
.22
.16
.66
.80
1.34
.16
.20
.32
—
—
.20
.12
—
.66
.17
FIR
.58
.16
.14
1.02
.92
.69
1.04
.04
.32
.36
.26
.32
.07
1.12
.60
.22
VI L
.62
.20
.10
.56
1.60
1.49
1.08
.46
.74
.06
.22
.24
.06
.96
.68
.18
EDN
—
.14
0
.58
1.18
1.30
.30
0
.56
—
.26
.28
.08
1.11
.66
.17
MAC
-
.10
.14
1.06
1.11
—
.26
0
.62
1.12
.26
.50
.08
1.12
—
.12
WTP
.56
.08
0
.61
1.27
—
.20
.44
.50
—
—
.24
.05
1.16
.80
.08
EDS
.16
.05
.80
.68
.16
70
-------
Table. 17. MAXIMUM 60 MINUTE RAINFALL
AT EACH GAGE - 16 1969 RAINS
Maximum 60 Min. Rainfall, In.
Day
116
141
148
173
176
196
205
207
211
218
242
267
274
285
303
320
Date
4-26
5:21
5-28
6-22
6-25
7-15
7-24
7-26
7-30
8- 6
8-30
9-24
10-1
10-12
10-30
11-16
US
W.B.
.20
.07
.24
.20
.56
.59
.18
T
.23
.69
.12
.12
.08
.13
.10
.03
UM
.23
.10
0
.23
.79
.92
—
.50
.36
.50
.12
.11
.05
.25
.16
—
VAD
.26
.04
0
.20
1.06
.06
—
1.53
.37
.53
—
.09
.04
.22
.18
.04
HAZ
.16
.05
.17
.19
.65
1.15
—
.20
.15
—
—
.12
.12
—
.17
.07
FIR
.27
.05
.14
.49
.63
.61
-
.04
.27
.36
.13
.14
.07
.12
.11
.08
VIL
.30
.05
.11
.18
.38
.83
-
.30
.59
.06
.15
.11
.04
.20
.10
.05
EDN
.23
.06
0
.16
.93
1.06
-
0
.42
—
.14
.14
.08
.16
.12
.08
MAC
—
.07
.15
.39
-
-
-
0
.45
1.12
.13
.21
.08
.22
—
.07
WTP
.25
.05
0
—
-
-
-
.44
.34
-
-
.12
.05
.22
.16
.04
EUS
.09
.05
.18
.10
.05
Table 18. MAXIMUM 30 MINUTE RAINFALL
AT EACH GAGE - 16 1969 RAINS
Maximum 30 Min. Rainfall, In.
US
Day Date W.B. UM VAD HAZ FIR VIL EDN MAC WTP EUS
116 4-26 _______ __
141 5-21
148 5-28
173 6-22
176 6-25
196 7-15
205 7-24
207 7-26
211 7-30
218 8-6
242 8-30
267 9-24
274 10-1
285 10-12
303 10-30
320 11-16
71
.06
0
.12
.50
.49
.50
.25
.49
.10
.06
.05
.13
.07
.04
0
.14
.70
.05
1.53
.22
.53
—
.07
.03
.17
.07
.04
.17
.11
.37
1.10
.20
.09
—
—
.07
.11
.07
.14
.03
.07
.30
.38
.48
.04
.14
.35
.08
.08
.05
.06
.06
.04
.08
.09
.24
.44
.30
.45
.05
.10
.06
.02
.12
.07
.04
0
.12
.58
.81
0
—
0
.12
.07
.08
.09
.06
.04
.14
.24
.76
-
0
.28
1.12
.09
.15
.05
.13
—
.05
—
-
—
0
.44
.21
—
—
.06
.05
.12
.09
.05
.04
.12
.08
-------
Table 19. TOTAL RAINFALL AT EACH GAGE
11 1970 RAINS
Total Precipitation, In.
Day No.
102-103
105
109
112
118
120-121
141-142
147
147-148
149
151
Date
April 12-13
April 15
April 19*
April 22
April 28
April 30-May 1
May 21-22
May 27
May 27-28
May 29
May 31
U.S.W.B.
.34
.11
1.38
.31
.14
.67
1.10
.44
.58
.41
.77
UM
—
—
—
—
-
.72
1.37
.43
.44
.83
.24
VAD
.42
.21
1.34+
—
1.56
.69
1.19
.41
.27
.85
.15
HAZ
.59
.11
1.36+
.31
.36
.82
1.73
.51
.51
.39
.11
FIR
.58
.20
1.02+
.27
.52
-
1.57
.47
.38
.57
.20
VI L
.70
.29
1.04+
.55
.73
.57
1.78
.57
.56
.35
.54
EDN
.58
.26
1.13+
.52
.63
.76
1.33
.44
.59
.11
.45
MAC
.48
.14
1.36+
.48
.16
-
1.25
.56
.69
.51
.84
WTP
.45
.31
1.46+
.50
1.45
.73
1.35
.49
.55
.17
.42
EUS
.26
.10
1.42
.62
.25
.67
1.84
.46
.42
.83
.43
"Rainfall weight in 7 gages exceeded transmitter range so exact total amount was not measured.
Table 20 shows the maximum 60 minute accumula-
tions at the nine system gages. Intensity influences very
greatly the overflow losses that occur. A short high
intensity rainfall can cause more combined sewer
overflow than a less intense rain having a much greater
total accumulation of rain.
Regulator Performance
For each selected rainfall event, a compilation of
data was arranged. Listings of data files for regulators,
interceptors, rain gages, and river monitors were
generated on the line printer. A plotting routine has
been written for the system, and the data was plotted,
making review and interpretation easier. Knowing the
depth of flow in the trunk sewer and the height of the
dam or diversion weir, one can determine whether
overflow occurred and its duration. Volumetric flow
rate of the overflow can be calculated using a weir
equation. The analysis of data proceeded approxi-
mately as follows:
1. Gate position and Fabridam inflation pressure were
checked since both have an effect on depth of flow
in the trunk sewer during a storm flow period.
2. If the Fabridam was inflated, the duration of the
overflow, if any, was determined and noted.
3. Assuming that the old diversion weir was still in
place, the duration of overflow that would have
occurred was determined and noted.
4. In instances where the Fabridam was not in an
inflated state during the storm flow or where it
deflated during the storm flow, an attempt was
made to arrive at an estimate of overflow duration
based upon the trunk level data, the rainfall data,
the flow at other regulators, and the past behavior
of the particular regulator. Estimates were made and
Table 20. MAXIMUM 60 MINUTE RAINFALL
AT EACH GAGE - 11 1970 RAINS
Maximum 60 Min. Rainfall, In.
Day No.
102-103
105
109
112
118
120-121
141-142
147
147-148
149
151
Date
April 12-13
April 15
April 19
April 22
April 28
April 30-May 1
May 21-22
May 27
May 27-28
May 29
May 31
U.S.W.B.
.05
.05
.38
.23
.11
.11
.88
.22
.32
.26
.40
UM
—
-
-
—
—
.17
.56
.16
.20
.54
.16
VAD
.12
.09
.56
—
—
.12
-
-
.14
.57
.07
HAZ
.12
.06
.51
.18
.21
.16
-
.18
.30
.29
.05
FIR
.06
.08
.42
.14
.42
-
.77
.21
.12
.52
.10
VI L
.16
.12
.40
.34
.69
.06
.61
.18
.23
.13
.45
EDN
.16
.12
.38
.41
.32
.12
.70
.20
.24
.04
.24
MAC
.13
.06
.44
.40
.13
-
1.01
.25
.38
.33
.41
WTP
.10
.10
.46
.31
1.34
.12
.59
.15
..22
.06
.15
EUS
.04
.05
.38
.52
.13
.14
.67
.16
.15
-
.32
72
-------
used in overall data compilation only if conditions
indicated that accurate interpretation could be
made.
When an inflated dam was the diversion weir,
volumetric flow calculations for the old weir condition
could not be calculated using the weir equation.
Applying an orifice equation to the gate opening could,
however, produce a figure for additional flow volume
diverted because of the higher head condition in the
trunk sewer.
by reduction of incidence of overflow during the 16
1969 rains. The figures cited for incidence of overflow
are totals for all monitored regulators. Table 22 shows
regulator performance as measured by reduction in
duration of overflow during the same rains. That many
of the per cent reduction figures are high confirms that
in the past, with diversion weirs only slightly higher
than dry weather flow depth, regulators overflowed for
a fairly long period following most rains. Table 23 and
Table 24 show the reductions in overflow incidence
and overflow duration during the eleven 1970 rains
analyzed.
Day
116
141
148
173
176
196
205
207
211
218
242
267
274
285
303
320
Table 21 shows regulator performance as measured
Table 21
SUMMARY OF 1969 OVERFLOW INCIDENCE REDUCTION
Number of Regulator Overflows
Post-Project
Conditions
Date
April 26
May 21
May 28
June 22
June 25
July 15
July 24
July 26
July 30
August 6
August 30
September 24
October 1
October 12
October 30
November 16
TOTALS
SUMMARY
Pre-Project
Conditions
13
8
6
14
15
15
13
11
12
13
8
10
5
9
8
7
167
1
0
1
6
13
11
3
4
6
7
1
1
1
2
1
1
59
Day
116
141
148
173
176
196
205
207
211
218
242
267
274
285
303
320
Date
April 26
May 21
May 28
June 22
June 25
July 15
July 24
July 26
July 30
August 6
August 30
September 24
October 1
October 12
October 30
November 16
Pre-Project
Conditions
37 00
33 00
7 15
50 00
51 45
43 15
21 30
15 45
41 00
31 30
17 15
20 15
8 00
69 00
61 30
19 30
TOTALS
Table 22
OF 1969 OVERFLOW DURATION REDUCTION
Total Regulator Overflow Hours
Post-Project
Conditions
1 30
0
1 00
10 30
11 45
12 45
1 30
2 45
7 30
6 15
0 30
1 30
1 00
2 30
1 00
0 15
527 30 62 15
i Reduction
92
100
83
57
13
27
77
64
50
46
87
90
80
78
87
86
65
i Reduction
96
100
86
79
77
71
93
83
82
80
97
93
87
96
98
99
88
73
-------
Table 23
SUMMARY OF 1970 OVERFLOW INCIDENCE REDUCTION
Day
102-103
105
109
112
118
120-121
141 - 142
147
147-148
149
151
Date
April 12-13
April 15
April 19
April 22
April 28
April 30 - May 1
May 21 - 22
May 27
May 27 - 28
May 29
May 31
TOTALS
Number of Regulator Overflows
Pre- Project
Conditions
8
8
11
12
11
8
13
11
11
10
11
114
Post-Project
Conditions
0
0
8
6
8
3
11
4
7
6
5
58
> Reduction
100
100
27
50
27
62
15
64
36
40
55
49
Day
102-103
105
109
112
118
120-121
141 - 142
147
147-148
149
151
Table 24
SUMMARY OF 1970 OVERFLOW DURATION REDUCTION
Total Regulator Overflow Hours
Date
April 12-13
April 15
April 19
April 22
April 28
April 30
May 21 •
May 27
May 27 - 28
May 29
May 31
TOTALS
• May 1
22
Pre-Project
Conditions
67 00
46 15
111 15
66 45
37 00
102 45
66 00
45 30
47 30
36 30
29 15
655 45
Post-Project
Conditions
0
0
17 30
4 15
11 00
7 00
18 15
8 45
9 45
5 15
3 15
85 00
% Reduction
100
100
84
94
70
93
72
81
79
86
89
87
Probably of greater interest than the effectiveness of
the program in reducing overflow incidence or over-
flow duration, is a measure of its effectiveness in
reducing overflow volume. To avoid the laborious
calculations that would be necessary to produce the
additional volumes diverted and prevented from over-
flowing, use was made of the mathematical model. As
part of the model, gate movements may be simulated
to evaluate effects of operations.
The actual rainfall data from four 1969 rains and
three 1970 rains were used as input for the model with
diversion weir height varied to simulate both fully
inflated Fabridams and the old fixed height weirs.
Table 25 shows the results of this analysis for eleven
regulators.
The overall volume reduction predicted by the
model for these seven rains is 51 per cent. Reductions
for individual rains varied from 30 per cent to 98 per
cent. It is to be expected that success in reducing
overflow would be inversely related to the total runoff
generated, and this is confirmed by the model. The rain
of Day 218 was very intense and caused a compara-
tively large runoff volume. Rains of this nature occur
only a few times a year. The other rains analyzed by
the model are also of higher than average amount, so
the gentle, long duration rainfall events are not
represented in this analysis. It is in rains of this nature
that the overflow can be completely controlled at the
modified regulators. Excluding the Day 218 rain, the
overflow volume reduction for the other six rains is 73
per cent.
74
-------
Table 25
OVERFLOW VOLUMES AT ELEVEN REGULATORS
FROM MATHEMATICAL MODEL WITH ACTUAL RAINFALL DATA
Volume of Overflow, CF
Additional Volume
Rainfall Day Old Weirs Inflated Dams Diverted, CF
207-1969 181,100 2,900 178,200
211-1969 1,095,700 474,000 621,700
218-1969 3,161,600 2,216,000 945,600
267-1969 220,700 27,100 193,600
147-148-1970 568,500 108,000 400,500
149-1970 397,600 10,400 387,200
151-1970 522,300 176,300 346,000
TOTALS 6,147,500 3,014,700 3,072,800
TOTALS
(excluding Day 218)
2,985,900
798,700
2,127,200
Per Cent Reduction
of Volume
98
57
50
88
81
97
66
51
73
It is interesting to compare the model's predicted
results for the eleven regulators with the data accumu-
lated during the actual rainfall event. Table 26 shows
this comparison. For some rains, duration reduction
closely parallels overflow volume reduction, while for
other rains, duration reduction is both greater than and
less than volume reduction.
Projected Seasonal Effectiveness
A sufficient number of rains were analyzed that
projections of normal seasonal effectiveness of the new
regulator control system can be made. The greatest
probable source of error involved in scaling up rainfall
related overflow data is in evaluating how repre-
sentative is the data available of the entire normal
season. Very likely the rains for which data were
analyzed fall on the heavy side of the distribution of
rainfall events. This system can admittedly effect good
control only for frequent return interval, low intensity
rainfall events. Therefore, scaling up measured to
seasonal overflow in the same proportion as measured
to seasonal rainfall, will indicate poorer than actual
overflow reduction success on a seasonal basis. Statis-
tical studies on rainfall return interval primarily deal
with one year return interval storms and greater and
little information was available for high frequency
rainfall events.
So it should be emphasized that overflow control
data presented on a seasonal basis are estimated.
Because of the low number of rains analyzed volu-
metrically with the mathematical model, the projection
of seasonal overflow volumes and reductions is even
more tenuous. Other methods of estimating overall
season overflow volumes were used in an attempt to
improve the accuracy of the projections.
The previously mentioned 27 rains occurring in
both 1969 and 1970 represent a total of 15.59 in. of
rain. This figure is the sum of the averages of all system
gages in use for each rain. Average total precipitation
amounts for the 27 events ranged from 0.07 in. to 1.32
in. The normal precipitation total for Minneapolis-St.
Paul during April through October is 19.51 in.
Table 26
OVERFLOW CONTROL
MATHEMATICAL MODEL PREDICTION
VERSUS
REAL DATA INDICATION
Per Cent Reductions
Predicted by Model
Rainfall Day
207
211
218
267
147
149
151
Overflow Volume
98
57
30
88
81
97
66
Incidence of
Overflow
67
43
22
50
71
89
60
Overflow Duration
81
83
75
85
80
86
89
Per Cent Reductions
Indicated by Real Data
Incidence of
Overflow
67
60
37
83
36
40
55
75
-------
Table 27 indicates that 164 incidents of combined
sewer overflow discharge and over 1000 combined
sewer overflow hours were prevented during the 27
rains, reductions of 58 and 88 per cent over what
would have occurred had pre-project conditions existed
at the major regulators. The overflow incidents and
durations were multiplied by the ratio of seasonal to
measured rainfall to produce estimated seasonal
figures.
Table 27
PROJECT EFFECTIVENESS IN REDUCING
FREQUENCY AND DURATIONOF OVERFLOW
Total Normal
27 Rains Season
15.59
281
1183
19.51
352
1480
147
184
58
88
Average Total Rainfall, In.
Pre-project overflows
Number
Duration, hours
Post-project overflows
Number 117
Duration, hours 147
Reduction (per cent) in overflow
due to project
Number 58
Duration 88
a measured values
k estimated values
Results from computer analysis of actual rainfall
data using the mathematical model routines have
already been presented in Table 25. At eleven regu-
lators during these seven rains a total of 3,072,800
cubic feet of combined sewage was prevented from
overflowing and was intercepted. Comparison of
routed interceptor flow hydrographs near the end of
the Joint Interceptor for simulated pre- and post-
project conditions suggest that a total of about
4,850,000 cubic feet of combined sewage was captured
by the system during these rains. Scaling up this value
by the simple ratio of seasonal total rain to measured
rain results in an estimate of 31 million cubic feet or
230 million gallons combined sewage overflow volume
prevented by the system on a seasonal basis.
The mathematical model, described in greater detail
in Part II of this report, attempts to account for all
flow which reaches the interceptor system, so that the
routing phase of the program will produce hydrographs
at the end of the interceptor system that are repre-
sentative of actual conditions. Since only the major
regulators and interceptor inlets were directly modeled,
the other smaller regulators and inlets were accounted
for by using coefficients representative of the ratio of
total tributary area to modeled tributary area for each
of the 15 catchments modeled. The model calculates
the total inlet volume for any rainfall data and then
calculates how each hydrograph is diverted at each of
the 15 modeled regulators.
For the seven rains analyzed, the model calculated a
total inlet volume of 426 MG. Scaled up to a seasonal
figure by the ratio of rainfall amounts, the total inlet
volume becomes 2700 MG. This represents the total
runoff volume for the combined sewered areas of
Minneapolis and St. Paul, since the base or dry weather
flow was subtracted. This volume is distributed among
approximately 150 combined sewered regulators, but,
based upon studies of areas involved, it is estimated
that about 70 per cent of the runoff arrives at the
controlled major regulators for diversion. Figure 63
shows post-project and pre-project dispositions of the
estimated total seasonal combined sewer runoff vol-
ume. It should be emphasized that the figures are
estimated by the model program which was used in a
rough state of calibration. Certain of the numbers
shown in Figure 63 are not consistent with those
produced by other manipulations of data. For ex-
ample, the reduction in overflow going to the river
from controlled regulators of 230 MG represents a
reduction of only 34 per cent. This is much lower than
the 51 per cent figure in Table 25. This table was
prepared from analysis of eleven controlled regulators
for which model predictions seemed most consistent
with measured data, so this data is perhaps more
reliable.
In defense of the combined sewer system it should
be pointed out that a large fraction of storm runoff
occurring each year is intercepted and conveyed to the
treatment plant. Based on 2700 MG of storm water
runoff from combined sewered area, 2160 MG runs off
the separate storm sewered area. Only 720 MG of
combined sewage reaches the river, 1/3 of the volume
discharged from the separate storm sewers. So com-
bined sewage would have to be 3 times as polluted as
separate storm sewer water to equal the effect of the
storm sewer discharges on the receiving water.
Unfortunately the treatment plant is unable to
handle flows in excess of about one half of the
capacity of the interceptor at the plant. Therefore
most of the additional volume intercepted is bypassed
untreated at the plant to protect the plant. Correction
of this deficiency will be accomplished in future
construction at the plant, and in the meanwhile the
pollution of the urban stretch of the Mississippi River
is much less frequent and of shorter duration to
increase dramatically the acceptability of the river for
boating, hiking and other recreation.
76
-------
Post-Project Conditions
2700 MG
Combined Sewer System
(70%) 1900MG
800 MG (30%)
Pre-Project Conditions
2700 MG
i
Combined Sewer System
(70%) 1900 MG
800 MG
(30%)
Controlled
Regulators
Uncontrolled
Regulators
(77%) 1460 MG 520 MG (65%)
440 MG
280 MG
1980 MG
Interceptors
I
720 MG
i
Mississippi River
Large Regulators
Small Regulators
1230 MG
520 MG
670 MG
1750 MG
Interceptors
I
950 MG
280 MG
Mississippi River
Figure 63. Estimated Disposition of Seasonal Runoff on Combined
Sewered areas in Minneapolis and St. Paul.
77
-------
1970 Spring Thaw
The control and monitoring system was maintained
year-round and hourly scanning continued through the
winter. Thaws occurring during the winter and spring
were capable of causing combined sewer overflow, so it
was interesting to note the improvements effected by
operation of the system. Operation to increase diver-
sion of combined thaw runoff and sewage is somewhat
simpler. Thawing in this climate is usually restricted to
daylight hours and usually occurs between 12:00 noon
and 5:00 P.M. Thaw flows slowly and predictably rise
and fall, there being no equivalent of a "cloudburst,"
so there is no need to be concerned about deflating the
Fabridams to provide needed flow capacity.
The heavy snowcover accumulated in December
remained on the ground most of the winter. Table 28
compares December, 1969-March, 1970 snowfall as
inches of water with the climatological standard
normal precipitation. Figure 64 is the Venturi meter
chart for the Minneapolis East Interceptor for Febru-
ary 23, 1970, illustrating a typical thaw runoff pattern.
Table 28
PRECIPITATION AS SNOWFALL FOR
WINTER OF 1969-1970
Precipitation Normal Precipitation
Month Inches-Water Inches-Water
December 2.06 0.92
January 0.47 0.81
February 0.16 0.86
March 2.05 1.52
Total 4.74
4.11
Figure 64. Flow Meter Chart-Minneapolis East
Interceptor (one of two identical
parallel meters)
There were four days on which thawing was
extensive enough to cause at least one overflow to the
river. There were 26 days, however, on which thawing
would have caused overflow at one or more of the
same regulators had pre-project conditions existed.
Table 29 indicates the amount of thaw overflow that
was prevented during the spring snowmelt of 1970.
Most of the overflow incidences prevented were prob-
ably small and the volumes not great. Seven thaw days
produced considerable runoff, however, and substantial
volumes of combined flow were kept out of the river
by the system.
Table 29
REDUCTION OF COMBINED SEWER OVERFLOW
DURING 1970 SPRING THAW
Pre-Project Post-Project
Regulator Regulator
Conditions Conditions
Number of Days
Having Thaw
Overflow 26 4
Number of Incidences
of Regulator Over-
flow 99 8
About 70 per cent of the incidences of overflow
that would have occurred, had pre-project conditions
existed, were at four regulators. There were a few
regulators for which no pre-project overflows would
have occurred at all.
Computer Analysis of Simulated Rainfall
Further use was made of the mathematical model
for evaluation of system effectiveness. Rather than use
real rain data, rainfall data were simulated. As a
simplification, the synthetic rainstorms all were of
uniform distribution and intensity. While this type of
rain may never in fact actually occur, the analysis
provides a means of quantifying what the system may
be capable of doing and pointing out the variability of
what can be done at various regulators. Each synthetic
rain data file was used as input to the model
twice—once with the inflated dams as the diversion
weirs, and once with the old weirs as the diversion
devices. Rainfall lastings 1/4, 1/2, 1, 2, and 3 hours was
simulated. Rainfall intensities from 0.10 inches/hour to
1.50 inches/hour were used. Table 30 indicates the
various combinations of intensity and duration that
were used.
Figure 65 and Figure 66 summarize the results of
this analysis. Figure 65 shows that reduction in
overflow volume is influenced by both intensity and
duration. As duration increases, some limiting overflow
78
-------
Table 30
SIMULATED RAINFALL DATA USED
FOR MATHEMATICAL MODEL ANALYSIS
OF THE COMBINED SEWER SYSTEM
INTENSITY, INCHES/HOUR DURATI'
1/4 1/2
0.10.
0.20 ....
0.30
0.40
0.50
0.75
1.00
1.25
1.50
X
X
X
X X
X
X X
ON HOURS
1 2 3
XXX
XXX
X
X
XXX
X X
X X
X X
X
reduction curve would be reached as steady state con-
ditions develop and continue. Figure 66 shows, for
rains of one-hour duration, the total overflow occurring
at 12 controlled regulators for the two diversion weir
conditions. Somewhere between intensities of 0.75 and
1.0 inches/hour, the additional increment of flow that
can be diverted appears to become constant, which is
what would be expected.
In these analyses, interceptor capacity has not been
considered to be a factor. The model, in its routing
section, does indicate where it calculates excessive flow
in the interceptor; that is, where an input or a routing
results in a flow rate exceeding that considered to be
the maximum for the interceptor at that location. In
processing some of the heavier simulated rainfall, the
model did indicate flow volumes exceeding the capa-
city of the interceptor, but these conditions are not
reflected in the figures.
100
80
5
D
_1
O
I
S60
O
O
Q40
O
EC
\
20
.25
.50 .75
RAINFALL INTENSITY, IN./HR
1.00
1.25
Figure 65. Model Predicted Overflow Reduction
79
-------
.25 .50 .75
ONE HOUR RAINFALL INTENSITY, IN./HR
Figure 66. Model Data from Simulated Rainfall
i.oo
1.25
Cost-Effectiveness of the Control System
Using the extrapolated, estimated figures obtained
from the mathematical model, approximately 30,000
combined sewered acres produce approximately 720
MG of combined sewer overflow each year. To
duplicate the reduction in overflow accomplished by
the control system, 230 MG, by separation would then
require complete separation of almost 10,000 acres.
Estimates of the cost of separation in St. Paul alone are
about 400 million dollars, at an average cost of about
$20,000/acre. At that cost, the 1.7 million dollar
demonstration program is accomplishing the same
effect as sewer separation work costing about 200
million dollars. Additionally, this hypothesized separa-
tion project would increase the separate storm sewer
discharge by about 350 MG each year, thereby
increasing storm sewer pollution. Possibly little head-
way would be made by separating at this very great
expense.
The separation cost of $20,000/acre is probably not
an overly high estimate either. Guarino et. al. (23)
have reported a detailed cost analysis for combined
sewer separation in Philadelphia. Their analysis in-
cluded costs of internal plumbing conversions as well as
costs of separation of the collection system in the
streets. The total cost of complete separation on a per
acre basis was $55,000 to $66,000 per acre.
So it should be assumed that the combined sewer
overflow control approach demonstrated here on a
small number of major regulators is capable of more
than duplicating the results of complete separation at
less than one one-hundredth of the cost.
This is even more impressive when one considers
that these results can be accomplished in two years. It
would be practically impossible to accomplish com-
plete separation of 10,000 acres in 2 years regardless of
cost. And whatever the time schedule, separation work
would involve infinitely greater inconvenience and
commercial interference for the citizens of the cities.
80
-------
XI. RECOMMENDATIONS
At a meeting in Washington in the Fall of 1968,
among other things, the District and FWPCA personnel
agreed that it would be worthwhile to empanel a
technical review committee representing various fields
of engineering interest applicable to the project. The
committee's purpose was, as stated in the Board
Minutes of October 14,1968, "to analyze project goals
and other program aspects which might lead to
suggested modification." The committee, authorized
by the Board on February 24, 1969, with members
jointly selected by FWPCA and District staff, met at
the District's offices on March 17 and 18, 1969. The
committee consisted of nationally known experts in
research, design, and operation of pollution control
facilities.
The committee members were most penetrating in
their questioning of the project personnel. In addition
to the District project personnel, representatives of the
St. Anthony Falls Hydraulics Laboratory who de-
veloped the mathematical model, and the District's
consulting systems analyst participated in the dis-
cussion. A number of changes in the mathematical
model and in data management resulted. Several new
ideas for evaluation of the effectiveness of the system
were obtained.
Not all of the recommendations of this committee
were followed because of the need to proceed with the
projects' limited objectives. Most of the recommen-
dations are still very pertinent now, as the demonstra-
tion period of the project ends and a final report is
prepared, and as the continued use and improvement
of the system is anticipated. The entire report of the
committee follows.
REPORTBYEVALUATING COMMITTEE
IN SESSION MARCH 17-18, 1969
The Evaluating Committee named below met on
March 17-18, 1969, to review the project entitled
"Dispatching System for Control of Combined Sewer
Losses" financed jointly by the Federal Water Pollu-
tion Control Administration and Minneapolis-St. Paul
Sanitary District.
The following members of the Committee partici-
pated in the deliberations:
Mr. Srifel W. Jens
Rein & Jens,
Consulting Engrs.
Ill So. Meramec Ave.
St. Louis, Mo. 63105
Dr. E. Benjamin Wylie
Associate Professor of
Civil Engineering
University of Michigan
Department of Civil Engrg.
A /m A rbor, Mich. 48104
Mr. David R. Dawdy
Research Hydrologist
U.S. Geological Survey
345 Middlefield Road
Menlo Park, Cat. 94025
Mr. Carmen F. Cuarino
Deputy Commissioner
Ciry ofPhila. Water Dept.
15th and John F. Kennedy
Blvd.
Philadelphia, Pa. 19100
Mr. V. A. Koelzer
Vice President
Harza Engineering Company
400 West Madison Street
Chicago, III. 60605
Mr. M. B. McPherson
Program Director
ASCE Urban Water Resources
Research Program
Harvard University
Engineering Science Lab.
40 Oxford Street
Cambridge, Mass. 02138
Ben Sosewitz, Acting Chief
Maintenance and Operation
The Metropolitan Sanitary
District of Greater Chicago
100 E.Erie Street
Chicago, III. 60611
The Committee elected as its Chairman, David R.
Dawdy, and as its Secretary, Ben Sosewitz.
The following project description was reviewed and
is taken from the information sheet prepared by the
Office of Research and Development, Federal Water
Pollution Control Administration:
DESCRIPTION OF PROJECT: "Dispatching System
for Control of Combined Sewer losses. "
"The project includes preliminary studies to update
historic data, a four-phase construction project
consisting of installation of a gaging system, a data
logger, five river monitors, telemetering rain gages,
regulator modifications, and a post-construction
program evaluation to include special studies by the
University of Minnesota. Existing regulators will be
replaced with modem power operated gates at 18
key diversion locations. A supervisory system will
be provided to telemeter gate positions, flows, and
levels in sewers to be controlled to a central point
where a dispatching operator can observe conditions
and regulate flow accordingly. Maximum utilization
of interceptor sewer capacity would be assured and
overflow to the river will be minimized. "
The above mentioned description defined initially
the tools to be used in achieving the objectives of this
program which are articulated in the last sentence of
the project description. It is the consensus of the
Evaluating Committee that the approach to achieve
this end, as well as the construction work completed
and the instrumentation provided, have been coordi-
nated sufficiently to demonstrate successfully the
capability of operating remotely and automatically the
regulating devices on the interceptor sewer system.
It has also been demonstrated that the collection,
storage, and analysis of data describing sewer elevation,
gate position, rainfall occurrence and waterways quali-
ty monitoring, are on line in accordance with the
project description.
The Committee has chosen for its purpose to divide
the basic objectives of this project into the following
three categories:
1. operational benefits accruing to the Minnea-
polis-St. Paul Sanitary District;
2. continuity of operational results and data
accumulation for future analysis, evaluation
and study; and
81
-------
3. program concepts and results usefully trans-
ferable to other municipalities facing similar
problems.
It is the consensus of the Committee that opera-
tional benefits may be anticipated by the Minneapolis-
St. Paul Sanitary District in the utilization of this
system to reduce overflows to the river. No numerical
results can be associated with this until after an
operational season is completed. The full impact of this
system's value can only be expected after the mathe-
matical model being developed is placed into service
and coordinated with the system and computer in
performance tests. Several suggestions regarding the
above follow:
(a) Development of the modeling program should
be continued.
(b) It would be desirable to obtain some measure-
ments of discharge inside the system to verify
the model.
(c) Other auxiliary field measurements should be
obtained to verify the data being monitored
by the system.
(d) Pre-project conditions should be simulated
and evaluated in order to determine relative
comparisons.
(e) Economic evaluation to determine relative
effectiveness of the system should be planned
for and pursued.
Of greater concern to the Committee was the need
for crystalizing data and conclusions which may be
transferable and of benefit to other communities. The
Committee cautions the Project Director and the
Federal Water Pollution Control Administration not to
allow the unique opportunity, available as a result of
this project, to be dissipated because of lack of
completeness or followup. It is the opinion of the
Committee that the following steps should be included
in the present program or an extension of the program
in order to guarantee the value of the transferable data:
(a) The model theory applied in both routing and
analysis should be verified operationally and
by field measurement.
(b) Simulation ofpre- and post-project conditions
should be accomplished.
(c) Considerable effort should be placed in re-
porting out the details of the system used in
achieving the operational capability demon-
strated.
(d) Any sophisticated economic evaluations
should be reinforced by actual field verifi-
cation. The Review Committee recommends
that funds be sought for a new program or an
extension of the present program to make
such economic evaluations.
It is the feeling of the Committee that prior to
final submittal of the report as outlined, a second
project review committee should evaluate performance
over the operational season, which will commence
shortly, together with the responsiveness to the recom-
mendations outlined herewith. Such a committee could
bring additional insight if it included independent
participants such as a hydraulic modeler, hydrologic
modeler, planner, instrumentation specialist and speci-
fically operators from Seattle and Detroit because of
the similarity of the project to their activities. In
addition, it is suggested that the final report author
avail himself of an ASCE report to the USGS (dated
April, 1969) which will shortly be available.
In conclusion it is the consensus of the Committee
that the project direction has been handled expertly
and that in the Committee's opinion, if additional
funds are required to establish the continuity and
guarantee the value of the project, such expenditure is
strongly justified by the opportunity available to meet
the industry's needs at the present time.
DRD:rc
April 17, 1969
END OF REPORT
Regarding the four steps recommended in the report
to be included in the present program or an extension
of the program, the following remarks can be made.
1. The model theory has been verified operationally
with real data. This is reported upon in detail in Part
II of this report, the report of the Saint Anthony
Falls Hydraulics Laboratory to the Minneapolis-
Saint Paul Sanitary District. There is, nevertheless,
need for additional improvement, calibration, and
further verification with more data that has and will
be collected. At this time, this is not possible
because of lack of funding.
2. Simulation of pre- and post-project conditions has
been accomplished to some extent using the existing
mathematical model. This is reported upon in both
Part I and Part II of this report. One benefit derived
from this simulation, besides evaluation of system
effectiveness, is that areas of the model requiring
improvement are readily identified in analysis of
simulation results.
3. Hopefully, this report has adequately reported
details of systems used in the Sanitary District
Regulator Demonstration Program. If not, personal
inquiries and inspection visits will be as welcome in
the future as were the many occurring in the past.
4. Sophisticated economic evaluations have not been
made.
Many recommendations regarding the data collec-
82
-------
tion aspects of the system have been received. The
following are selections from personal correspondence
to the Executive Director-Chief Engineer and the
Project Engineer at the Minneapolis-Saint Paul Sanitary
District from review committee members and others.
"Both you and your staff are to be congratulated on
the progressive policy which you are following
concerning water pollution control. There are a lot
of words being spoken these days about necessary
advances in our field, but not as many deeds. Your
organization is matching your words with deeds."
Carmen F. Guarino, Deputy Commissioner,
Water Department, City of Philadelphia. March
19,1969.
"You are indeed fortunate in having pursued so
vigorously this inovation in the control of overflow
from a combined sewer system. The detail to which
this project has been carried and the enthusiasm
expressed by its staff were particularly impressive."
Vinton W. Bacon, General Superintendent,
Metropolitan Sanitary District of Greater
Chicago. March 28, 1969.
"The opportunity for collecting the additional data
by proper supplementary instrumentation and net-
work design for one or more of the catcnments
involved in the Minneapolis-St. Paul demonstration
project ... is so great that to ignore it appears
indefensible . . .
Local operational and probable design benefits will
accrue from a verified or modified mathematical
model."
Stifel W. Jens, Reitz and Jens, Consulting Engi-
neers. April 12, 1969.
"The approach taken here certainly does have
applications in the other metropolitan areas. If the
older cities are required to eliminate or minimize
stormwater overflows, they will have to choose
between various alternatives and combinations. The
experience of the Minneapolis-St. Paul area will be
extremely helpful in this regard.
To attempt separating sewers in large cities would
be a tremendous undertaking and, needless to say,
very expensive .. .
The information which will be collected by the
system that Minneapolis-Saint Paul has installed .. .
will be helpful to determine the best methods of
minimizing stormwater pollution, proper design of
collector systems, and the practicality of routing
storms through a collector system."
Carmen F. Guarino, Deputy Commissioner,
Water Department, City of Philadelphia. April
13,1969.
"Although Mr. Sosewitz's report emphasized the
need for contining the data collection program, I am
so convinced of the unique opportunity available in
your project to obtain data that are otherwise sadly
lacking, that I feel continued re-emphasis of this
aspect is essential. This alone would justify future
additional grants from FWPCA or Office of Water
Resources Research."
V. A. Koelzer, Chief, Engineering and Environ-
mental Sciences, National Water Commission.
May 1, 1969.
"... a point ... was made by our evaluating
committee ... that, in the light of the very meager
amount of data on quantity and quality of waste-
water flows, the unique installation which you have
should continue to generate information which will
be of widespread public value. It would be very
disappointing, in my view, for the information
system to be used for purely operational purposes."
V. A. Koelzer, Chief, Engineering and Environ-
mental Sciences, National Water Commission.
December 22, 1969.
To the words of these people, the Project Engineer
can only add the statement that the Minneapolis-Saint
Paul Sanitary District has been deeply impressed by
these recommendations. Certainly, efforts will con-
tinue to be made to accumulate much needed data, but
to insure transferability of data, guidance and direction
must come from without.
A report to the United States Geological Survey by
the American Society of Civil Engineers also em-
phasized the importance of data similar to that which
the Sanitary District system can continue to collect.
"Extensive construction of new and replacement
urban drainage systems and facilities for reduction
of pollution from storm discharge and combined
sewer overflows over the remainder of this century
will involve significant national expenditures. Exist-
ing knowledge is inadequate for efficient and
optimal planning and development of such facilities.
Better methods of analysis and more realistic design
criteria are needed for engineering design of new
drainage systems."
"Improved methods are evolving very slowly be-
cause of a dearth of field measurements of rainfall-
runoff from which logical, reliable models and
quantification of their principal parameters can be
developed. No municipality has had the financial
and manpower resources to mount a research
program suited for national transferability of re-
sults, and currently no prospect is in evidence of a
consortium of local governments financially pre-
pared to undertake the task." (9)
Very likely the Sanitary District system would have
to be modified to increase its value as a source of basic
data needs. Again, impetus must come from without.
Other recommendations can be made. In all proba-
bility, systems similar to the one used at the Minnea-
polis-Saint Paul Sanitary District can be used with
similar success elsewhere.
83
-------
"Of the fourteen largest U.S. cities, ten are partially
or wholly served by combined systems of sewerage:
New York, Chicago, Philadelphia, Detroit, Cleve-
land, Washington, St. Louis, Milwaukee, San Fran-
cisco and Boston. Of these, the following are served
almost exclusively by combined systems: Chicago,
Detroit, St. Louis and San Francisco; and more than
half of New York, Philadelphia and Cleveland are
served by combined systems." (9)
Several years ago, the American Public Works
Association (21) estimated the national construction
cost for eliminating combined sewer systems by sepa-
ration at $48 billion. Because of the extreme cost and
the impracticality of separation, along with evidence
that urban storm water discharges from separate storm
sewers is signficantly polluted, the FWQA has pro-
moted research and demonstration studies of alterna-
tive methods of combined sewer overflow pollution
abatement, of which the Minneapolis-Saint Paul Sani-
tary District Regulator Demonstration Program is one.
The approach used here, one of control and routing,
would seem to be applicable elsewhere to achieve
limited but more immediate results at proportionally
low cost and to complement other alternative methods
by which more complete abatement might be attained.
Any treatment facility built on a controlled regula-
tor system, such as that in Minneapolis and St. Paul,
would accrue these benefits from the existing system:
Its size could likely be smaller; it could be accurately
sized because of data already available; and operating
and maintenance costs would be lower because of less
frequent utilization.
Another fact worthy of mention is that this was
really a "first of a kind" project. Much has been
learned here, and in other similar projects elsewhere,
that should permit subsequent systems, having greater
capabilities, to be installed elsewhere at lower cost.
Engineers at Badger Meter Manufacturing Company
have indicated, in private conversation, that the system
they provided the Sanitary District with could be
acquired at a lower cost today both because of what
was learned during this project and because of ad-
vancing technology. The central processor, or com-
puter power, of this system could be provided today at
a much lower cost than three years ago.
84
-------
XII. REFERENCES
1. Baker, D. G., Haines, D.A. and Strub, J.H., Jr.,
"Climate of Minnesota-Part V. Precipitation
Facts, Normals and Extremes," University of
Minnesota-Agricultural Experiment Station Tech-
nical Bulletin 254, 1967.
2. Toltz, King, Duvall, Anderson and Associates, Inc.,
St. Paul, Minnesota, "Report on the Expansion of
Sewage Works in the Minneapolis-Saint Paul Met-
ropolitan Area," Sponsored by the Minneapolis-
Saint Paul Sanitary District, Volume 3, September,
1960.
3. Anderson, J.J., Bruss, O.E., Hamilton, L.A. and
Robins, M.L., "Application of the AutoAnalyzer
to Combined Sanitary and Storm Sewer Prob-
lems," Presented at the Technicon Symposium
entitled Automation in Analytical Chemistry, New
York, October 3, 1967.
4. Anderson, J.J., "Computer Control of Combined
Sewers," Presented at American Society of Civil
Engineers Annual and Environmental Engineering
Meeting, Chicago, Illinois, October 13-17, 1969.
5. Bowers, C.E., Harris, G.S. and Pabst, A.F., "The
Real-Time Computation of Runoff and Storm
Flow in the Minneapolis-St. Paul Interceptor Sew-
er," University of Minnesota, St. Anthony Falls
Hydraulic Laboratory Memorandum No. M-118,
Prepared for Minneapolis-St. Paul Sanitary District
and FWPCA, December, 1968.
6. Harris, G.S., "Real-Time Estimation of Runoff in
the Minneapolis-St. Paul Metropolitan Area,"
University of Minnesota, St. Anthony Falls Hy-
draulic Laboratory Memorandum No. 119, Pre-
pared for Minneapolis-St. Paul Sanitary District
and FWPCA, December, 1968.
7. Harris, G.S., "Mathematical Models of Major
Diversion Structures in the Minneapolis-St. Paul
Interceptor Sewer System," University of Minne-
sota, St. Anthony Falls Hydraulic Laboratory
Memorandum No. 120, Prepared for Minneapolis-
St. Paul Sanitary District and FWPCA, December,
1968.
8. Harris, G.S., "Development of a Computer Pro-
gram to Route Runoff in the Minneapolis-St. Paul
Interceptor Sewers," University of Minnesota, St.
Anthony Falls Hydraulic Laboratory Memoran-
dum No. M-121, Prepared for Minneapolis-St. Paul
Sanitary District and FWPCA, December, 1968.
9. "An Analysis of National Basic Information Needs
in Urban Hydrology, A Report to the U.S.
Department of Interior," by the American Society
of Civil Engineers, April, 1969.
10. "Dispatching System for Control of Combined
Sewer Losses," Interim Report to the Federal
Water Pollution Control Administration Submitted
by the Minneapolis-Saint Paul Sanitary District,
May, 1969.
11. Anderson, J.J., "Dispatching and Routing of Com-
bined Sewer Storm Flows for Maximum Inter-
ceptor Utilization," Presented at American Society
of Civil Engineers Water Resources Conference,
New York, October 18, 1967.
12. Anderson, J.J., "Real-Time Computer Control of
Urban Runoff," Presented at the ASCE Hydraulics
Division Conference, Cambridge, Massachusetts,
August 21-23, 1968.
13. Gallery, R.L., "Management of Pollution Data
Using Modern Techniques," Presented at American
Society of Civil Engineers Annual and Environ-
mental Engineering Meeting, Chicago, October 15,
1969.
14. Request for Facilities Demonstration Grant Form
Submitted by the Minneapolis-Saint Paul Sanitary
District to the Federal Water Pollution Control
Administration, April, 1966
15. Green, S.R., "The Storage and Retrieval of Data
for Water Quality Control," U.S. Department of
the Interior, Federal Water Pollution Control
Administration, Public Health Service Publication
No. 1263, August, 1966.
16. Freeny, A.E. and Gabbe, J.D., "A Statistical
Description of Intensive Rainfall," The Bell Sys-
tem Technical Journal, Volume 48, Number 6,
Page 1789, July-August, 1969.
17. Horner, W.W. and Jens, S.W., "Surface Runoff
Determinations from Rainfall Without Using Co-
efficients," Transactions, AM. Soc. C.E., Volume
107, Page 1039, 1942.
18. Hicks, W.I., "A Method of Computing Urban
Runoff," Transactions, AM. Soc. C.E., Volume
109, Page 1217, 1944.
19. Mick, K.L., "The Minneapolis-Saint Paul Sanitary
District Operation and Expansion," Journal of the
Water Pollution Control Federation, Page 1684,
October, 1967.
20. McKee, J.E., "Loss of Sanitary Sewage Through
Storm Water Overflows," Journal of the Boston
Society of Civil Engineers, Volume XXXIV, Num-
ber 2, April, 1947.
21. "Report of Problems of Combined Sewer Facilities
and Overflows, 1967," Federal Water Pollution
Control Administration by the American Public
Works Association, December 1, 1967.
85
-------
22. Anderson, J.J., "Analysis of Operating Data from 23. Guarino, C.F., Radzuil, J.V., and Greene, W.L.,
a Full Scale Primary Sedimentation Plant at the "Combined Sewer Considerations By Philadelphia,"
Minneapolis-Saint Paul Sanitary District," A thesis Journal of the Sanitary Engineering Division,
submitted to the University of Minnesota in partial ASCE, Vol. 96, No. SA 1., February, 1970.
fulfillment of the requirements for the Degree of
Master of Science in Civil Engineering, April,
1967.
86
-------
XIII. ACKNOWLEDGEMENTS
The author of the report wishes to acknowledge the
help and cooperation of the Board and Staff of the
Metropolitan Sewer Board and the former Minneapolis-
Saint Paul Sanitary District.
Metropolitan Sewer Board Members
MILTON C. HONSEY, Chairman, New Hope
JULIUS SMITH, Vice Chairman, Chaska
HERMAN J. ARNOTT, Treasurer, Minneapolis
JOSEPH C. COOK, Coon Rapids
BERNON STENSENG, Stillwater
WARREN SCHULTZ, Eden Prairie
GEORGE RUTMAN, Saint Paul
DANIEL R. BAKER, Saint Paul
Metropolitan Sewer Board Staff
RICHARD J. DOUGHERTY, P.E., Chief Ad-
ministrator
MAURICE L. ROBINS, P.E., Deputy Chief Ad-
ministrator
GEORGE W. LUSHER, Chief Superintendent of
Operations
Former Minneapolis-Saint Paul Sanitary District Board
Members
ARTHUR E. NAFTALIN, Minneapolis
JACK E. NEWTON, Minneapolis
DONALD P. RISK, Minneapolis
GEORGE J. VAVOULIS, Saint Paul
ROBERT A. OLSON, Duluth
ARNOLD J. IMSDAHL, Saint Paul
THOMAS R. BYRNE, Saint Paul
GERARD D. HEGSTROM, Minneapolis
ROBERT F. PETERSON, Saint Paul
CHARLES P. McCARTY, Saint Paul
JENS CHRISTENSEN, Minneapolis
ROBERT F. SPRAFKA, Saint Paul
DONALD R. WATKINS, St. Cloud
LOUIS P. COTRONEO, Saint Paul
NATHAN HARRIS, Minneapolis
VERN R. ANDERSON, Minneapolis
Former Minneaplis-Saint Paul Sanitary District Staff
KERWIN L. MICK, P.E., Chief Engineer and Super-
intendent, MSSD (Retired July, 1968)
MAURICE L. ROBINS, P.E., Executive Director-
Chief Engineer, MSSD (July, 1968 to Present)
Principal Project Personnel
ROBERT L. GALLERY, P.E.
Project Engineer, July, 1969-Present
Assistant Engineer, June, 1967-July, 1969
JAMES J. ANDERSON, P.E., Project Engineer,
April, 1966-July, 1969
LOUIS J. BARSCHER, JR., Data Analyst
EPA Project Officers
DARWIN R. WRIGHT, Washington, D.C.
LOUIS J. BREIMHURST, Minneapolis, Minn.
Consulting Programmers
STEPHEN F.CARLSON, Systems Analyst
RICHARD S. CARLSON, Programmer
Mathematical Model Program Personnel
C. EDWARD BOWERS, Professor, University of
Minnesota
GARTH A. HARRIS, Post Doctoral Fellow, Uni-
versity of Minnesota
ARTHUR F. PABST, Research Fellow, University
of Minnesota
Plans and Specifications
Contract 664A-Pioneer Service & Engineering
Co., Chicago
Contract 664B - Pioneer Service & Engineering
Co., Chicago
Contract 666 —Pioneer Service & Engineering
Co., Chicago
Contract 677 — Toltz, King, Duvall, Anderson &
Assoc., St. Paul
Contractors
Contract 664A - George F. Cook Construction Co.,
Minneapolis
Contract 664B - Badger Meter Mfg. Co., Milwaukee
Contract 666 — Fairchild Camera and Instrument
Co., New York
Contract 677 - Acton Construction Co. and Ash-
bach Construction Co., a joint
venture, St. Paul
Special Consultants
Mathematical Model Program—Saint Anthony Falls
Hydraulics Laboratory, University of Minnesota,
Minneapolis
Cooperation by others
The District has enjoyed excellent cooperation from
many local, state, and federal agencies, numerous
private firms, and private individuals. A complete list
and all the details here would be too lengthy, but a
brief summary follows with apologies for unintentional
omissions.
Federal Water Pollution Control Administration
This project has been authorized, supported, and
financed in part, by the Department of Interior
pursuant to the Federal Water Pollution Control Act.
The fine assistance and cooperation from the
personnel of the FWPCA is greatly appreciated. Dis-
cussions on various matters with Washington and local
personnel always provided quick resolution to pro-
blems and yielded information requested immediately.
The river quality monitors are as developed by the
FWPCA, with considerable help locally and from
Cincinnati. STORET personnel have been very knowl-
edgeable and helpful, and the effectiveness of the
system which the FWPCA has developed can hardly be
overstated.
87
-------
Without the financial assistance from the FWPCA,
the project might not have existed.
U.S. Geological Survey
The USGS has furnished data for the project, both
locally and in Washington, promptly and exactly as
requested.
ESSA
The U.S. Weather Bureau has assisted the daily
regulator surveillance by providing direct rainfall in-
formation and the state climatologists have assisted in
the placement and checking of rain gages. They have
also made publications and rainfall data available, some
of which has been placed on punch cards for corre-
lation with MSSD data.
Corps of Engineers
The Corps of Engineers has provided data and maps,
and computer programs developed by them have been
used in the project.
Cities of Minneapolis and Saint Paul
Both Cities are providing financing for the District's
share of works constructed exclusively for one city and
the cost of the District's share of jointly used project
facilities is financed by them through normal District
procedures. The engineering departments in both cities
have provided plans and drawings and obtained neces-
sary approvals, in addition to maintaining close liaison
on other matters.
The Water Departments in both cities have allowed
installations of rain gages on their property. One river
quality monitor is installed in a building at the
Minneapolis Water Department. The Saint Paul Fire
Department has cooperated in the installation of a rain
gage, as has the Metropolitan Airports Commission. A
river quality monitor is located at one of Saint Paul's
storm water pumping stations.
Private Firms and Individuals
Northern States Power Company has allowed instal-
lation of a river monitor on their property at the High
Bridge Plant. They also have been most cooperative in
providing electrical services at all of the locations,
including some in unusual locations.
Northwestern Bell Telephone Company personnel
have been extremely helpful and have maintained close
liaison with District and contractor personnel. When
required, all parties have worked very effectively as a
team.
Industrial Molasses Corporation and Svoboda Boat
Works have allowed river monitors to be placed on
their property.
Manufacturers, vendors, sales representatives and
contractors in all fields have taken an interest in this
unusual project and have provided help even when no
direct benefit to them could be realized.
American Society of Civil Engineers
The ASCE Urban Hydrology Research Council has
made available the results of extensive research and
publications concerning combined sewer overflow pro-
blems sponsored by the FWPCA and concerning urban
hydrology and water resources management sponsored
by the USGS and OWRR, respectively. The interchange
of information at meetings and in the Division Journals
has been invaluable to the project.
88
-------
APPENDIX A
An Example of Data, May 27-28,1970
Presented as an appendix is a nearly complete set of
plotted data for a single rainfall event. The data are
typical of the quality and quantity of data available for
27 rainfall events. By the end of the 1970 rainfall
season, this body of data will likely have grown to
include over 50 rainfall events.
With this large collection of high quality data, the
calibration of the existing mathematical model can
certainly be improved. This data, which because of the
procedures that have been set up can continue to
accumulate, will be of great value to. the Public Works
Departments of the cities of Minneapolis and St. Paul
or to their consultants as they face the continuing
problems of controlling pollution from their combined
sewer systems.
The data are presented in the form of plots or
graphs that were generated on the project system. The
time axis on all plots is 16 hours long, from 1600 hours
on Day 147 (May 27) to 0800 hours on Day 148 (May
28). Each vertical grid line represents one hour. The
vertical axis on all plots is labeled from 0 to 100,
representing 0 to 100 percent of the full scale value
that is printed at the top center of each plot. Raingages
and inflatable dam pressures have full scale values of 2
in. and 2 psi respectively. Full scale values for depth of
flow may be 50, 100, or 120 in.
On plots of trunk level (TL) the horizontal row of
2's represents the height of the old pre-project diver-
sion weir and the horizontal row of 1's represents the
height of the new inflated dam.
The rain for which the data are presented was fairly
uniform. Rain began falling at about 2100 hours on
May 27 and lasted, on the average, about 3 hours. The
total rainfall amount measured at the 9 system rain
gages varied from 0.27 in. to 0.69 in.
Review of this data indicated that there were 7
overflows from control and monitoring regulators and
that there would have been 11 overflows had pre-
project conditions existed. Total overflow duration
hours for pre-project conditions would have been 47.5
and the actual total overflow hours were only 9.7, so at
controlled regulators 37.8 overflow hours were pre-
vented.
When the rainfall data were input to the mathe-
matical model, the overflow volume reduction attri-
butable to the program at eleven well modeled regu-
lators was from 568,500 cubic feet to 108,000 cubic
feet, a reduction of 81 per cent.
89
-------
RAIN GAGES
The total amount of precipitation
at the Water Treatment Plant rain
gage was .55 inches.
The total amount of precipitation
at the Golden Valley rain gage was
.56 inches.
The total amount of precipitation
at the Edina rain gage was .59
inches.
90
-------
The total amount of precipitation
at .the University of Minnesota St.
Paul Campus rain gage was .44
inches.
The total amount of precipitation
at the Eustis rain gage was .42
inches.
The total amount of precipitation
at the Airport rain gage was .69
inches.
91
-------
mt V
1*1 1*1 !*•
The total amount of precipitation
at the Lake Vadnais rain gage was
.27 inches.
The total amount of precipitation
at the Hazel Park rain gage was .51
inches.
The total amount of precipitation
at the West Side Fire Station rain
gage was .38 inches.
92
-------
REGULATORS
Camden Regulator
The increased depth of flow caused
by rainfall is apparent in this plot.
The pipe height is 48 inches.
CAM
I'M Elft-Bl l»l IV (BU l(*ll
I I I.I 1 I.I I I.I I I.I I I.I I I.I I 1.1 I 1.1 I I.I I I.I I I.I I 1.1 I I.I I I.I I 1.1 1 1.1 1 •
The height of a side overflow weir
was raised from 35" to 48". No
overflow occurred, but before this
modification, overflow would have
occurred for 1.5 hours.
93
-------
Portland and Washington
This is termed "Master Regulator"
because it is built into the N.W.
Interceptor, and regulates all flow
: entering the interceptor upstream.
Although increased flow was re-
ceived here, no overflow would
have occurred even before regulator
modification. The pipe height is 90
inches.
The same proportionate increase in
flow is apparent in the interceptor
down stream from this regulator.
The vertical lines represent auto-
matic inflation of the 'Fabridam'
The pressure losses due to leakage
cause no problems. Within the
range of pressure shown, there is no
change in height.
94
-------
26th and Seabury
This regulator would have overflow-
ed for 3.5 hours prior to installa-
tion of the Fabridam. The pipe
height is 96 inches.
This depth is measured in the dry
weather outlet chamber. The outlet
pipe size is 36 inches, and submerg-
ed entrance conditions existed for
over 1.5 hours.
ifi-tc • tin mi scut
This plot shows the automatic in-
flation pressure adjustment at three
hour intervals and the increased
pressure exerted upon the
'Fabridam' during high flow
periods.
95
-------
39th and Minnehaha
The depth of flow here exceeded
the height of the old weir for 2.25
hours. The trunk sewer height is
123 inches.
S fULl SCILE
The increased flow in the outlet
pipe leading to the interceptor is
shown. The level sensor is located
within a parshall flume, so the
discharge can be determined here.
M39
Due to damage the 'Fabridam' was
not functional.
96
-------
m» 111-11 loons fyn
0:11 FUll iC»L(
t?*i tH-iC 1. IS Ftlll SCUt
38th and Edmund
During this rainfall event, flow ex-
ceeded the height of the inflated
'Fabridam' for four hours. A much
greater volume would have been
lost over the old weir which was
overflowed for six hours. The
height of the pipe is 120 inches.
The 36 inch outlet pipe here did
not flow full, indicating that the
size of the gate opening controls
the maximum diversion rate.
The re-inflation every third hour
and the increased flow pressure
against the 'Fabridam' can be seen
in this plot.
97
-------
31 stand Randolph
Runoff would have caused overflow
for the duration of the rainfall
event prior to modification. The
pipe height is 102 inches.
• RAN.
The plot shows re-inflation every
third hour and the result of flow
pressure on the 'Fabridam'. There
was a rapid leak at this location
which is shown in this plot.
98
-------
100'. IS fUll SClLt
2nd and Main
This is a "Master Regulator" for the
Minneapolis East Interceptor. Over-
flow would have occurred for at
least 4 hours prior to modification
at this regulator. Flow did exceed
the height of the 'Fabridam' for
approximately 45 minutes. The
pipe height is 84 inches.
The trend of increased flow is
shown in the interceptor down-
stream from this regulator. The
pipe size is 42 inches.
Re-inflation and increased pressure
on the 'Fabridam' is shown. The
loss of pressure is very slow here.
99
-------
Oak Street
Runoff flow exceeded the height of
the 'Fabridam' here for approxi-
mately 3 hours. (The old weir
would have overflowed for over five
hours and much more volume would
have been lost.)
EAST INT.
Increased pressure due to the storm
flow and deflation of the 'Fabri-
dam' is shown in this plot.
Normally this dam is almost leak
free, but the pressure relief valve
apparently released air during the
runoff event.
100
-------
Rice and Rondo
The pattern of increased runoff is
apparent in this plot. Because of
high velocities, this data may not be
correct. The pipe height is 114
inches.
This plot of the level in the outlet
section shows the effect of rainfall
runoff. The dry weather outlet pipe
size is 36".
JOINT INT
\
Increased pressure from runoff flow
and re-inflation every third hour is
shown here. This dam has a known
leak and rapid loss of pressure.
101
-------
Trout Brook
Overflow occurred here for approx-
imately 15 minutes. Prior to 'Fabr-
idam' installation, this regulator
would have overflowed almost 4
hours. The pipe height is 126
inches.
This data indicated a problem with
the bubbler element which was
later repaired.
JOINT INT.
102
-------
Phalen Creek
Runoff flow exceeded the height of
the 'Fabridam' for a total of 1.25
hours. Dry weather flow would
have been occurring before the
event prior to modification. Pipe
height is 112 inches.
A submerged entrance condition
occurred continuously before,
during, and after the rainfall event.
Deficient outlet capacity is appar-
ent. The pipe height is 30 inches.
JOINT INT.
The increased pressure of additional
flow and the re-inflation of the
Fabridam' is shown in this plot.
103
-------
Belt Line
Because of high velocities, trunk
sewer level data was not reliable.
This level is in the outlet chamber.
The entrance to the 24 inch outlet
pipe was submerged for 3Vi hours.
JOINT INT.
A very rapid pressure loss is evi-
dent. An air leak in the attachment
caused this. The function of the
dam is therefore somewhat un-
certain.
104
-------
: A
Kittsondale
Overflow would have occurred for a
total of 2.25 hours at this regulator
prior to modification. No flow pass-
ed over the inflated dam. Pipe
height is 108 inches.
Increased flow in the outlet section
is shown in this plot.
• KD
MISS. RIVER
BLVD INT.
Inflation pressure adjustment every
third hour is shown in this plot.
105
-------
INTERCEPTORS
Minneapolis Northwest Interceptor
The rainfall almost filled the Minne-
apolis Northwest Interceptor at this
location at 1 a.m. The pipe height is
54 inches.
The runoff from the rainfall event
in the northwest area of Minne-
apolis that was captured is shown in
the interceptor level at the end of
the Northwest interceptor. The
pipe height is 123 inches.
!«• • It*
106
-------
Minneapolis Southwest Interceptor
Increased runoff filled the Minne-
apolis Southwest Interceptor at this
location for well over two hours.
The pipe height is 72 inches.
Increased runoff flow in the Minne-
apolis Southwest Interceptor is
shown in this plot of level at its
end. The pipe height is 72 inches.
NDA
107
-------
Minneapolis East Interceptor
The plot of level in the Minneapolis
East Interceptor at Tenth and
Marshall shows the effect of in-
creased rainfall. The pipe height is
88 inches.
The effect of rainfall runoff on the
flow in the Minneapolis East Inter-
ceptor is shown in this plot of level
at its end. The pipe height is 72
inches.
EAST INT.
MISS. RIVER
BLVD. INT.
St. Paul Mississippi River Blvd.
Interceptor
The St. Paul Mississippi River Blvd.
Interceptor was filled to capacity
around midnight due to the in-
creased runoff. The pipe height is
70 inches.
108
-------
St. Paul W. 7th St. Interceptor
The level in St. Paul West Seventh
St. Interceptor shows the increased
flow during this runoff event. The
pipe height is 70 inches.
Joint Interceptor
Runoff flow at the Joint Inter-
ceptor at Pine and Spruce reached
almost 130 inches. The pipe height
is 157 inches.
JOINT INT.
JT3
109
-------
APPENDIX B
Selected Project Materials
a. Tables of Inputs to Interceptors
Several tables of data abstracted from that
accumulated during the sewer sampling and
analytical phase of the program are presented
to show examples of preliminary analysis of
this information. Only average values are incor-
porated into the tables. Part C of this appendix
illustrates what the average value obscures.
b. Figures of Inputs to Interceptors
Information of the type included in the
tables has been applied to a schematic of the
interceptor system to more graphically illus-
trate the sources of sewage load under average
conditions.
c. Three-Dimensional Plots
The best way to illustrate the diurnal and
day to day variations in wastewater flow or
composition without resorting to statistical
parameters was found to be graphical. These
are photographs of three-dimensional models
constructed by student-employees.
111
-------
HOURLY CONCENTRATIONS OF
CHEMICAL OXYGEN DEMAND
In sewers
Tributary to the following:
STATION
SAMPLE PERIOD
Minneapolis Northwest Interceptor
Camden April, 1968-October, 1968
Portland and Washington July, 1967-January, 1968
Northwest Meter October, 1966-July, 1967
NUMBER OF
SAMPLES
1005
262
651
CHEMICAL OXYGEN DEMAND MG/L
MAX. MIN. MEAN
1340 173 344
1790 211 673
1700 150 578
II Minneapolis Southwest Interceptor
39th and Minnehaha January, 1967-January, 1968
38th and Edmund January, 1967-January, 1968
Southwest Meter October, 1966-July, 1967
III
IV
Minneapolis East Interceptor
31st-Randolph April, 1968-October, 1968
Second-Main October, 1967-October, 1968
Oak Street April, 1968-October, 1968
East Meter October, 1966-July, 1967
Joint Interceptor
Eustis
Wabash-Cromwell
Kittsondale
Rice and Rondo
West Seventh Street
St. Peter and Kellogg
Trout Brook
Phalen Creek
Belt Line
January, 1968-July, 1968
January, 1968-April, 1968
April, 1968-July, 1968
January, 1967-October, 1967
April, 1967-October, 1967
January, 1968-July, 1968
January, 1967-October, 1967
January, 1968-July, 1968
January, 1967-January, 1968
612
741
629
1630
1920
1420
181
183
174
426
574
466
811
1059
967
678
864
24
351
654
632
853
723
981
570
1990
1980
1910
1910
1950
1910
1950
1940
1870
1810
1240
1940
1660
184
177
169
168
187
504
186
227
178
179
179
187
180
593
606
408
622
625
1132
613
890
479
523
367
533
402
Treatment Plant
Plant Raw Waste
Plant Primary Effluent
October, 1966-January, 1969
April, 1968-July, 1968
4719
501
1930
1310
148
181
600
405
NOTE: Data contains wet weather as well as dry weather flows.
112
-------
III
IV
AVERAGE INPUTS OF CHEMICAL OXYGEN
DEMAND FROM KEY REGULATORS TO INTERCEPTORS
STATION
SAMPLE PERIOD
Minneapolis Northwest Interceptor
Camden May-August, 1968
Portland-Washington August-October, 1967
Seabury January, 1967-August-October, 1967
Northwest Meter November, 1966-May, 1967
Minneapolis Southwest Interceptor
39th-Minnehaha January, 1967/August-October, 1967
38th-Edmund January, 1967/August-October, 1967
Southwest Interceptor November, 1966-May, 1967
Minneapolis East .Interceptor
Randolph May-August, 1968
2nd-Main May-August, 1968
Oak April-July, 1968
East Meter November, 1966-May, 1967
Mississippi River Boulevard Interceptor
Otis April-May, 1968
Mississippi River Blvd.
AVERAGE
FLOW, MGD
10.5
35
12
50
AVERAGE CHEMICAL
OXYGEN DEMAND
MG/L LB/DAY
Joint Interceptor
Eustis
Rice-Rondo
W. 7th Interceptor
St. Peter-Kellogg
Trout Brook
Phalen Creek
Belt Line
345
675
500
580
March-May, 1968
January—August, 1967
May-August, 1967
March-May, 1968
January-August, 1967
March-May, 1968
January—August, 1967
NOTE: Hourly concentrations and flow depths are available.
30,000
197,000
5,000
242,000
8
13.5
37
14
27
8
37
430
575
470
595
605
410
625
29,000
65,000
145,000
69,000
136,000
27,000
193,000
3.5
11
15
6
5
0.8
11.5
15
9.3
613
625
890
480
525
370
530
400
18,000
78,000
44,000
20,000
4,000
35,000
66,000
31,000
113
-------
PRELIMINARY ANALYSIS OF VOLUMETRIC AND COD LOAD
INPUTS INTO THE MINNEAPOLIS-ST. PAUL INTERCEPTOR SYSTEM
Some flow data is estimated. COD concentration values are from Technicon AutoAnalyzer. COD concentrations
in parentheses are calculated from differences in flow COD load.
I Minneapolis Northwest Interceptor
Camden
Other
Portland and Washington
Seabury
Other
TOTAL
11 Minneapolis Southwest I nterceptor
39th-Minnehaha
38th-Edmund
Other
TOTAL
III Minneapolis East Interceptor
Randolph
Other
Second and Main
Oak Street
Other
TOTAL
IV Mississippi River Blvd. Interceptor
Otis and Marshall
Other
TOTAL
V Joint Interceptor
Eustis
Rice and Rondo
West Seventh
St. Peter and Kellogg
Trout Brook
Phalen Creek
Belt Line
Other
TOTAL
Plant
AVERAGE
FLOW, MGD
10.5
24.5
35
12
3
8
13.5
15.5
14
13
27
8
2
3.5
7.5
50
37
37
11
AVERAGE
COD MG/I.
345
(820)
675
500
580
430
575
(395)
595
(620)
605
410
(1800)
613
(1150)
470
625
981
15
6
5
0.8
11.5
15
9.3
7.4
625
890
480
525
370
530
400
(1250)
TOTAL
70.0
205
608
600
AVERAGE
COD # /DAY
30,000
167,000
197,000
50,000
-5,000
29,000
65,000
51,000
69,000
67,000
136,000
27,000
30,000
18.000
72,000
78,000
44,000
20,000
4,000
35,000
66,000
31,000
77,000
242,000
145,000
193,000
90,000
355.000
1,025.000
114
-------
AVERAGE INPUTS OF CHLORIDES FROM
KEY REGULATORS TO INTERCEPTORS
STATION
SAMPLE PERIOD
I Minneapolis Northwest Interceptor
Portland-Washington August — October, 1967
Seabury January, 1967-August-October, 1967
Northwest Meter November, 1966-May, 1967
II Minneapolis Southwest Interceptor
39th-Minnehaha January, 1967/August-October, 1967
38th-Edmund January, 1967/August-October, 1967
Southwest Interceptor November, 1966-May, 1967
AVERAGE AVERAGE
FLOW,MGD CHLORIDES, MG/L LB/DAY
35 170 50,000
12 80 8,000
50 165 69,000
8 110 7,000
13.5 180 20,000
37 175 54,000
III Minneapolis East Interceptor
East Meter November, 1966-May, 1967
37
90
28,000
IV Joint Interceptor
Rice-Rondo
W. 7th Interceptor
Trout Brook
Belt Line
January—August, 1967
May-August, 1967
January—August, 1967
January—August, 1967
6
5
11.5
9.3
110
65
90
90
6,000
3,000
9,000
7,000
NOTE: Hourly concentrations and flow depths are available.
STATION
AVERAGE INPUTS OF AMMONIA
NITROGEN FROM KEY REGULATORS TO INTERCEPTORS
SAMPLE PERIOD
I Minneapolis Northwest Interceptor
Portland-Washington August - October, 1967
Seabury January, 1967-August-October, 1967
Northwest Meter November, 1966-May, 1967
II Minneapolis Southwest Interceptor
39th-Minnehaha January, 11J67 /August-October, 1967
38th-Edmund January, 1967/August-October, 1967
Southwest Interceptor November, 1966—May, 1967
AVERAGE AVERAGE AMMONIA
FLOW, MGD NITROGEN, MG/L LB/DAY
35 15 4,400
12 8 800
50 17 7,000
8 14 900
13.5 16 1,800
37 20.5 6,300
III Minneapolis East Interceptor
East Meter November, 1966-May, 1967
IV Joint Interceptor
Rice-Rondo
W. 7th Interceptor
Trout Brook
Belt Line
January—August, 1967
May-August, 1967
January—August, 1967
January-August, 1967
37
6
5
11.5
9.3
NOTE: Hourly concentrations and flow depths are available.
21
4.5
11
15
2,500
1,100
200
1,100
1,200
115
-------
MINNEAPOLIS NORTHWEST
MINNEAPOLIS EAST
182,000 Ib./day
625 mg/l
35MGD
50MGO
580 mg/l
242,000 Ib./day
JOINT
| SWI | ]
35MGD
470 mg/l
137,000 Ib./day
MINNEAPOLIS SOUTHWEST
11MGD
MISSISSIPPI RIVER BLVD.
PLANT
Note: Hourly concentrations and
flow depths available.
, I |PRw|
205MGO
600 mg/l
1,025,000 Ib./day
Input of Chemical Oxygen Demand from
Minneapolis Meters to Joint Interceptor
under Average Conditions
10.5MGD
-OD 345 mg/l
CUD 30,000 lb./day
MINNEAPOLIS WEST
8MCO
430 mg/l
29.000 Ib./day
MINNEAPOLIS SOUTHWEST
Note: Hourly concentrations and
flow meter readings available.
Input of Chemical Oxygen Demand
from Major Regulators to Interceptors
under Average Conditions
116
-------
MINNEAPOLIS WEST
MINNEAPOLIS EAST
28.000 lbs./day
90 mg/l
37MGD
50MGD
165 mg/l |NWI
69,000 Ibs./day
|SWI
37MGD
175 mg/l
54,000 Ibs./day
MINNEAPOLIS SOUTHWEST
JOINT
MISSISSIPPI RIVER BLVD.
PLANT
Note: Hourly concentrations and
flow depths available.
Input of Chlorides from
Minneapolis Meters to Joint Interceptor
under Average Conditions
MINNEAPOLIS WEST
MINNEAPOLIS EAST
35MGD
170 mg/l
50.000 Ibs./day
9,000 Ibs./day
6.000 lbs./day 90 '
110 mg/l
6MGD
RR
SEA
12MGD
80 mg/l
8,000 Ibs./day
[39-MJ-*
8MGD
110 mg/l
7,000 lbs./day
MISSISSIPPI RIVER BLVD.
' | E38 |
13.5MGD
180 mg/l
20,000 Ibs./day
MINNEAPOLIS SOUTHWEST
9.3MGD
BL | 90 mg/l
7,000 Ibs./day
5MGD
65 mg/l
3,000 IbsVday
PLANT
Note: Hourly concentrations and
flow meter readings available.
Input of Chlorides from
Major Regulators to
Interceptors under Average Conditions
117
-------
MINNEAPOLIS WEST'
.MINNEAPOLIS EAST
37MGD
20.5 mg/l
6,300 IbsVday
50MGD
17 mg/l
7,000 lbs./day
PLANT
MISSISSIPPI RIVER BLVD.
MINNEAPOLIS SOUTHWEST
Note: Hourly concentrations and
flow depths available.
Input of Ammonia Nitrogen from
Minneapolis Meters to Joint Interceptor
under Average Conditions
I MINNEAPOLIS EAST
MINNEAPOLIS WEST
P-W
35M6D
15 mg/l
4,400 Ibs./day
1,100 Ibs./day
21 mg/l
6MGD
1,100 Ibs./day
11 mg/l
11.5MGD
JOINT
|39-M
8MGD
14 mg/l
900 lbs./day
MISSISSIPPI RIVER BLVD.
E38|
13.5MGD
16 mg/l
1,800 lbs./day
MINNEAPOLIS SOUTHWEST
5MGD
4.5 mg/l
200 IbsVday
9.3MGD
15 mg/l
1,200 IbsVday
PLANT
Note: Hourly concentrations and
flow depths available.
Input of Ammonia Nitrogen from
Major Regulators to Interceptors
under Average Conditions
118
-------
NOTE:
Data includes wet and dry weather
flows. Flow data available to calculate
pounds per hour.
Hourly Frequency Distributions of Chlorides
in Plant Raw Waste, Approximately 1800
Observations.
600
• ~\ 300
v 400 -rinNMG'
*" co«*»1BKI
NOTE: Data includes wet and dry weather
flows. Flow data available to calculate
pounds per hour.
Hourly Frequency Distribution of Chemical
Oxygen Demand at Minneapolis Northwest
Meters. Approximately 600 observations.
119
-------
NOTE: Data includes wet and dry weather
flows. Flow data available to calculate
pounds per hour.
Hourly Frequency Distributions of Ammonia
Nitrogen in Plant Raw Waste, Approximately
1800 observations.
NOTE: Data includes wet and dry weather
flows. Flow data available to calculate
pounds per hour.
Hourly Frequency Distributions of Chemical
Oxygen Demand at Minneapolis East Meter,
Approximately 600 observations.
120
-------
Part 11
MATHEMATICAL MODEL OF URBAN STORM RUNOFF AND THE
INTERCEPTOR SEWER SYSTEM OF THE
MINNEAPOLIS-ST. PAUL SANITARY DISTRICT
by
Arthur F. Pabst, C. Edward Bowers
and Garth S. Harris
Prepared at the
St. Anthony Falls Hydraulic Laboratory
University of Minnesota
for
The Minneapolis-St. Paul Sanitary District
and
The Federal Water Quality Administration
121
-------
I. INTRODUCTION
The cities of Minneapolis and St. Paul are served by
a combined sewer system which is designed to carry
both normal dry-weather sanitary sewage and storm-
water runoff. Originally the main trunk lines dis-
charged the combined flow directly into the Mississippi
River. However, in the 1930's interceptor sewers were
constructed to collect the flow that would normally be
discharged into the river and convey it to a treatment
plant. Diversion structures were built into the trunk
sewers to divert the flow into the new interceptor
sewers. The interceptor sewers were designed to carry
the normal dry-weather sanitary flow plus a small
amount of storm-water runoff, but excess storm-water
still had to be discharged into the river during large
storms. Figure 1 shows the location of the interceptor
sewers, the main diversion structures, and the inlets to
the interceptor lines.
Originally most of the major diversion structures
had fixed-height dams. Then in 1968 remotely-
controlled fabridams and gates were installed ^t the
diversion structures. These controllable gates and fabri-
dams allow an operator to alter the system according
to current conditions in the sewer system. With the
variable-control diversion structures, advantage can be
taken of the non-uniformity of rainfall distribution to
obtain the maximum capture of runoff. Larger
amounts of runoff can be directed into the sewer
system at certain locations if it is known that little or
no rain is occurring in other areas of the cities.
However, it is difficult for an operator to make
quantitative decisions about the amount of rainfall and
storm runoff in the system. A mathematical model of
the Minneapolis-St. Paul interceptor sewer system [1] *
was therefore developed to aid the operators in
determining the expected flow at critical points in the
system so that gates and fabridams could be manipu-
lated to retain the maximum flow in the system and
release the least overflow to the Mississippi River.
'Numbers in brackets refer to references.
V MISSISSIPPI RIVER
LEGEND
«- Major Inlet
• Diversion Structure
"i • Rain Gage
MMNEAPOLIS- ST. PAUL
SEWAGE TREATMENT PLANT
Figure 1. Map of Twin City area showing location of interceptor sewers, major inlets, diversion
structures, and rain gages.
123
-------
II. GENERAL CONCEPT OF MODEL
The mathematical model (UROM-9, Urban Runoff
Model PDP-9) is a deterministic runoff model capable
of continuous real-time operation. Input to the model
consists of rainfall readings and postulated gate and
fabridam settings. Output can range from a simple
message to the operator, to complete rainfall and loss
rate analysis, diversion analysis, and predicted hydro-
graphs at points in the sewer system. The operator can
interact with the model while it is operating in real
time to alter the sequence of operations, postulate new
gate and fabridam settings, and change the model
output. In addition the model can be set to operate on
historical or design rainstorms. The operation of the
model is discussed in detail in Appendix A, Operator's
Manual for UROM-9.
To make the model easier to understand and to
work with it has been broken down into a series of
simple tasks, each of which is carried out by a
subroutine. As each subroutine completes its task it
passes the results to another routine until the entire
model has been executed. This process can be illustra-
ted as follows: The rainfall data that are given to the
model may contain various errors of electronic,
mechanical, or human origin. Thus the model's first
task is to screen the rainfall data for errors and make
reasonable corrections. This task is performed by a
subroutine entitled RAINF. Since the rainfall data may
have been recorded at any random time interval, the
next task is to space the data at some regular interval
(usually 5 minutes); this task is performed by sub-
routine EVNTL. The next step is to determine what
portion of the rainfall can be expected to be "lost" to
the ground and what portion will become runoff.
Subroutine SLOS performs this task. These are only a
few of the steps in the overall model.
It can be seen that changes in methods are easily
implemented. If a different loss analysis is desired, for
example, the change can be made by simply replacing
the current version of SLOS with a new version
containing the new loss analysis. The subroutines are
not concerned with the particular area modeled, but
simply perform their functions on the data given them.
The model is applied to a given area by supplying it
with input data describing the physical features of the
area. These input data include the number of sub-
watersheds the total area is broken into, their areas,
their times to peak, and the base times of their unit
hydrographs. All input data is given in Appendix A. In
the preceding example, subroutine RAINF would be
fed the number of rain gages in the system, subroutine
EVNTL would be told at what regular interval to space
the rairrfall data, and subroutine SLOS would be given
several parameters for each sub-watershed which would
relate the amount of pervious and impervious area, the
initial depression storage, the infiltration constants,
and other variables depending on the particular loss
analysis used. Thus the methods used are governed by
the subroutine instructions, while the application to an
area is governed by the input data to the subroutines.
Some restrictions on the general applicability of the
model are as follows:
1. Storage requirements set an upper limit on the size
of input data arrays which in turn limits such things
as the number of sub-areas that can be modeled.
The upper limit can change depending on computer
storage available.
2. Variations in the sizes of data arrays may require
minor changes in the format of the output data.
3. Certain aspects of diversion structure modeling are
not general in nature, but apply only to the area to
which the model is being applied.
125
-------
III. SPECIFIC MODELING TECHNIQUES
For purposes of discussion the model can be divided
into three phases: the hydrology phase, the diversion
phase, and the routing phase. The hydrology phase is
the period between the time the storm water appears as
rainfall and the time it arrives at the inlets to the
interceptor sewer system. The diversion phase deter-
mines, for given positions of the rubber fabridam and
variable gate, how much of the arriving flow will enter
the interceptor sewers and how much will be diverted
to the river. Finally, the routing phase predicts the
movement of the flow that enters the interceptor
sewers. Further discussion of each of these phases is
given below.
A. Hydrology Phase
In this phase rainfall from a network of recording
gages is converted to runoff at the major inlets to the
interceptor system. The first step in the process deals
with the mechanics of input of the rainfall data (For
additional information on these steps see Appendix C.)
This step provides corrected rainfall readings at each
gage in the network at regular time intervals, usually 5
minutes. From these raingage readings the average
rainfall over each sub-watershed is determined, using
the Thiessen method [2]; this consists of weighting
each raingage by a factor dependent on the location of
the gage relative to the sub-area. During the 1969
runoff season a network of eight recording raingages
supplied data to the model. Figure 2 shows the
Thiessen polygons used to determine the weighting
factors for an eight-gage network with 15 sub-areas. At
the end of the year (1969) a ninth recording gage was
added. Future changes may also be made in the
locations of some existing gages.
LEGEND
— WATERSHED BOUNDARY
• RAIN GAGE *
— THIESSEN POLYGON
- INTERCEPTOR SEWERS
Figure 2. Map of Twin City Area showing major water-
sheds and Thiessen polygons.
Next the losses and the excess precipitation to be
expected in each of the sub-areas must be found. The
"loss" is that part of the total precipitation which is
retained in the sub-area, while the "excess" is the
portion that becomes runoff and enters the sewer
system. Predicting the losses that will occur during a
particular rainfall event over a particular area is
probably the most difficult task of the model. Several
approaches have been considered. The current method
determines separately the losses on the impervious
surfaces (i.e., paved areas) and those on the pervious
surfaces (grassy areas) and then combines them, taking
into account the relative amounts of pervious and
impervious surface in the sub-area.
These relative areas were estimated using aerial
photographs of representative areas of the Twin Cities.
Photos on a scale of either one inch equals 100 ft or
one inch equals 200 ft, covering areas of approximately
0.1 sq mi and 0.4 sq mi, respectively, were overlaid
with a 1/4 inch or a 1/8 inch uniform grid, and grid
squares enclosing impervious surfaces were counted.
Only effective impervious surface area was included;
where runoff from rooftops would drain across lawns,
for example, the rooftops were not considered imper-
vious area. The proportion of impervious area was
found to vary from 96 per cent for a downtown
business area to 26 per cent for a single-family
residential area. The average for Minneapolis and St.
Paul combined is 35 per cent.
Impervious Analysis—The excess from impervious
areas is assumed to depend on (1) the rainfall, (2) an
antecedent index parameter, and (3) three constants.
The antecedent parameter is the accumulated loss—i.e.,
the moisture retained—on the impervious surf aces; this
is presently taken to be zero at the beginning of an
event. The first constant is the value of the accumu-
lated loss that must be reached before any runoff will
occur. This value may take into account interception
by trees and wetting of land surfaces. Zero is presently
used for this initial loss constant. Once the initial loss
has been satisfied, the loss is determined as follows:
Loss = Rainfall x loss coefficient (1)
where
Loss coefficient = Ce~k (accumulated loss) (2)
and C and k are constants for a sub-watershed
(assumed to be 1.0 and 2.0, respectively.) Figure 3a
shows this loss coefficient as a function of accumulated
loss on the impervious area. The loss coefficient
decreases asymptotically from an initial value of 1.0
for completely dry conditions to a value of zero for
wet conditions. Thus for dry conditions the entire rain
is retained as loss on the impervious surfaces. As
moisture builds up, depression storage is filled and the
loss coefficient decreases, allowing a portion of the
127
-------
rainfall to become runoff. Figure 4a shows the actual
loss rate when this same function is applied to a
uniform one-inch-per-hour rain.
The impervious excess is then
Impervious Excess = Rainfall — loss
(3)
Loss
Coefficient
0 0.5 1.0 1.5 2.0 2.5
Accumuloted loss on impervious oreo (in)
Figure 3a. Impervious loss function.
0.5 1.0 1.5 2.0
Accumuloted loss on pervious Oreo (in)
2.5
Figure 3b. Pervious loss function.
Pervious Analysis—The excess from pervious areas is
assumed to depend on (1) the rainfall, (2) an antece-
dent index parameter, and (3) four constants. The
antecedent parameter is the accumulated loss on
pervious surfaces—i.e., the moisture retained by the soil
in a given layer. This value is presently taken as zero at
the beginning of an event. The first constant is an
initial value of the accumulated loss that must be
satisfied before any runoff can occur; a value of 0.25
in. has been used here. Once the initial loss has been
satisfied, the loss rate is determined using the Holtan
infiltration equation as modified by Muggins and
Monke: [3]
, , . /T - (accumulated loss)\k IA.
f=fc + Al - 1 (4)
where f is the potential infiltration rate (in. perhr); fc
is the constant equilibrium infiltration rate, taken to be
0.2 in. per hr; A is the increase in infiltration rate
under dry conditions, assumed to be 2.0 in. per
hr—giving a total infiltration rate of 2.2 in. per hr when
the soil is dry; T is the total available storage in the soil
above the restricting layer, assumed to be 2.0 in.; and k
is a constant taken as 0.7.
Figure 3b shows the potential loss rate as a function
of the accumulated loss on the pervious area. The loss
rate decreases from an initial value of 2.2 in. per hr for
completely dry conditions to a value of 0.2 in. per hr
for wet conditions. This same loss rate applied to a
one-in-.-per-hour rain is shown in Fig. 4b. Under these
conditions it may be noted that for the first 1-2/3 hrs
of the storm the rainfall rate is lower than the potential
loss rate, so the actual loss rate is equal to the rainfall
rate. During the remainder of the storm the potential
loss rate is lower than the rainfall rate. At 2-1/3 hrs the
potential loss rate has decreased to its equilibrium
value (0.2 in. per hr). Thus when the rainfall is less
than the potential loss rate, the actual loss rate is equal
to the rainfall rate; but when the rainfall equals or
exceeds the potential loss, the actual loss rate is equal
to the potential loss rate, which in turn equals the
equilibrium value fc. The excess on the pervious areas
is then given by
Pervious excess = rainfall — loss
and the sub-watershed excess is given by
(5)
Basin excess = —= I impervious excess) + ——-— (pervious excess)(6)
where x is the percentage of sub-area that is im-
pervious.
1.0
Loss rate
(in/hr)
.5
( Rainfall rote = 1.0 in/hr)
0123
Time (hrs)
Figure 4a. Impervious loss rate for 1 in./hr. rainfall rate.
i.o
Loss rate
(in/hr)
.5
\
PERVIOUS AREA
Time (hrs)
Figure 4b. Pervious loss rate for 1 in./hr. rainfall rate.
128
-------
Figure 4c shows the results when the complete loss
rate is applied to an area that is 1/3 impervious with a
one-in.-per-hour rainfall. The area between the time
axis and the loss curve represents the total loss over the
sub-watershed, while the area between the loss curve
and the dashed one-in.-per-hour line represents the
total excess over the sub-watershed. This excess will
become runoff at the inlet to the interceptor sewer.
1.0
Loss rate
(in/hr)
.5
COMBINED AREA
(33% IMPERVIOUS)
0 I 2 3
Time (hrs)
Figure 4c. Combined loss rate for 1 in./hr. rainfall on
a watershed 1/3 impervious.
An important feature of this loss analysis is that the
independent variable is accumulated loss (moisture
retained in the watershed), whereas in other loss
analyses—Morton's [2], for example—time is the inde-
pendent variable. Since under real conditions there are
often periods of low or zero rain during a prolonged
storm, an analysis based on time requires additional
techniques, such as time condensation. In addition to
simplifying the process, the use of accumulated loss as
the independent variable allows for a recovery or
draining feature to be incorporated into the analysis
which would describe the return to dry conditions
during periods of no rain. Presently no recovery feature
is included in the model.
The determination of the hydrographs at the inlet,
is next. The excess water, determined in the above loss
analysis, arrives at the interceptor sewer inlet; it is
assumed that this quantity of water is spread uniformly
over the sub-watershed. Several approaches to deter-
mining the hydrograph, or time distribution, of this
runoff have been proposed. The unit hydrograph
approach used here is based on the theory that when
the water drains to the watershed outlet (i.e., the
interceptor inlet) it does so according to a given time
pattern which is assumed to depend on the physical
features of the area—its size, shape, slope, etc. The
theory further assumes that if different amounts of
water occur as runoff, the magnitude of the runoff
changes proportionately, but the time pattern does
not change.
The shape of the unit hydrograph is most reliably
determined from recorded events that have occurred
on the watershed. Since this information was not
available in the present case, some characteristics of the
unit hydrograph were estimated from the physical
features of the sub-watersheds, and a triangular shape
was assumed. The estimated characteristics were the
time to peak Tp and the base
proximation Tp and
relationship given by Kent [4]:
Tb. As a first ap-
Tb were determined from the
0.8
TD(hrs) = -
9000 Sa
(7)
where D is the duration of rainfall increment (hr), L is
the length of mainstream to farthest divide (ft), CN is
the SCS hydrologic soil-cover complex number, Sa is
the average slope of the watershed (per cent) along the
mainstream, and
Tb (hr) = 2.67 Tp
(8)
Kent provides a nomograph for the solution of the
second term of Eq. (7). The peak discharge of the unit
hydrograph would then be given by
AR
Qp (cfs) = 0.0000464 ^~
(9)
where A = area (ft^), R = rainfall excess (in.), and Tb =
base time (hr).
The initial values of Tp and Tb given by Eqs. (7)
and (8) have been modified in some cases as recorded
data have been obtained by the data acquisition system
at the MSSD treatment plant. The Tp and Tb given by
Eqs. (7) and (8) have agreed reasonably well with
values obtained from recorded data.
The duration of the unit impulse is selected as 5
minutes. Thus each 5-minute rain period yields a
runoff hydrograph. These incremental hydrographs
are combined linearly and base flow is added to form
the total inlet hydrograph. The mean annual base flow
is used at each inlet with average hourly variations
imposed on it.
The formation of the total inlet hydrograph, com-
pletes the hydrology phase of the model. Additional
discussion of hydrology phase may be found in
reference [5].
B. Diversion Phase
In this phase the total inlet hydrographs, as deter-
mined in the hydrology phase, are passed through the
simulated diversion structures to provide an estimate of
how much of the flow will enter the interceptor sewer
and how much will enter the Mississippi River. The
amount of flow that enters the interceptor sewer
system will then be passed on to the routing phase of
the model.
129
-------
To accomplish the diversion simulation each major
diversion structure was analyzed and then represented
by a combination of orifices, weirs, and pipe sections
for which discharge equations were prepared. Several
structures are similar in geometric configuration, while
others are of unique design, occurring only once in the
sewer system. The most common type of diversion
structure is composed of (1) an outlet from the trunk
sewer to the interceptor sewer controlled by a hy-
draulically operated gate and (2) a tunnel to the river
controlled by a pneumatically operated fabridam. Such
a structure is shown in Fig. 5. This configuration is
referred to as a TYPE 1 diversion structure. The gate
opening is modeled as a variable-area orifice and the
fabridam as a variable-height weir. [6]
Figure 5. Typical Type 1 diversion structure.
130
-------
The flow coming down the trunk sewer is assumed
to take one of two paths: either through the gate to
the interceptor or over the fabridam to the river.
Actually, some of the flow must go into temporary
storage in the trunk sewer, but this flow is not
modeled. The portions of the trunk flow which go to
the interceptor and to the river can be determined
from the gate opening and the fabridam height. The
discharge Q0 that will pass through the orifice with a
driving head equal to the fabridam height can then be
found. If Qo is greater than the flow coming down the
trunk, all the flow will go through the orifice to the
interceptor sewer. If Qo is less than the trunk flow,
part of the flow will go through the orifice and part
will go over the fabridam. In this latter case a common
head must be found such that the sum of the
discharges over the weir and through the orifice equals
the trunk flow. This result is found using an iterative
procedure (method of bisection). Figure 6 shows two
complete solutions of a TYPE 1 diversion with the
fabridam both down and up, and each with three
different gate openings (0.25 ft, 0.75 ft, 1.25 ft). For
the fabridam up and the 1.25 ft gate opening, it can be
seen that when the trunk flow is less than 50 cfs the
entire flow will go through the orifice to the inter-
ceptor. For discharges greater than 50 cfs only part of
the flow goes to the interceptor, and the remainder
goes over the fabridam to the river. Thus for a trunk
flow of 100 cfs the interceptor flow will be 60 cfs and
the river flow will be 40 cfs. Once flow has started over
the fabridam only small increases in interceptor flow
occur for large increases in trunk flow. This is due to
the difference between the exponent for flow over a
weir (1.5) and that for flow through an orifice (0.5)
and the difference in datum for the heads.
The diversion structure thus acts to clip off the
higher flows, sending only a regulated amount to the
interceptor sewers.
The clipping action of the diversion structures
increases the accuracy of the model prediction. If the
loss rate analysis predicts rainfall excess equal to or
greater than the actual values, the inlet hydrograph will
be either correct or over-estimated. The diversion
structure will then cause the excess flow to be
discharged into the river. Thus the amount that enters
the interceptor, which is of primary interest, is more
nearly correct, with the errors going to the river. Figure
7 illustrates this effect. Curve 1 represents a computed
inlet hydrograph that may be in error. Curve 2 shows
the portion of this hydrograph that would enter the
interceptor, Curve 3 represents the actual inlet hydro-
graph, and Curve 4 shows the portion of this actual
hydrograph that would enter the interceptor. There-
fore, while there are large differences between 1 and 3
(that is, before diversion), there are only small differ-
ences between 2 and 4, after diversion. This, of course,
holds only when the true inlet hydrograph is large
enough to cause diversion to the river.
120
80
Fabridam up
Gole Opening (ft.) 1.25
0.75
40
160
200
80 120
Trunk Inflow (cfs)
Figure 6a. Typical Type 1 diversion rating curve with
fabridam up.
I20
o
£40
Fabridam down
Gate Opening (ft.) 1.25
40 80 120
Trunk Inflow (cfs)
160
200
Figure
6b. Typical Type 1 diversion rating curve with
fabridam down.
M^ Computed inlet hydrogroph
(?) Computed diverted hydrograph
(3) Actual inlet hydrograph
(4J Actual diverted hydrogroph
80
60
6 20
0 I 2 3 4 5 6
Time (hrs)
Figure 7. Illustration of flow clipping action of diversion
structure.
The TYPE 1 diversion structure occurs ten times in
the sewer system. The model includes nine other
locations where flow may be diverted to the river.
Three of these are modeled as sharp clipping devices
(i.e., all discharge above a fixed value is released to the
131
-------
river). Another three occur at the Minneapolis dis-
charge meters and are modeled as variable bypass gates
only. The remaining three are each special cases and are
handled individually. A complete discussion of each
structure is given in Ref. [7].
The land area that contributes storm runoff to the
interceptor system at the 15 major inlets represents 61
per cent of the total area contributing runoff to the
system. The remaining area contributes storm water to
the system through about 47 minor inlets. Diversion to
the river may occur at the 15 major inlets plus 4 more
structures such as venturi meters.
To allow for the minor inlet flow the contri-
bution of each minor inlet is assumed to enter the
sewer system at the nearest major inlet. All model
computations through the diversion phase are based on
the actual area contributing at a major inlet. After
diversion is complete the diverted hydrographs are
corrected by multiplying the discharges by the inlet
lumping factor, which is taken as the ratio of the area
assumed to contribute to a major inlet to the actual
area that contributes to the inlet. In other words, the
minor inlets are assumed to behave in a manner which
is hydraulically similar to that of the major inlets. The
largest area represented in this manner is at the Phalen
Creek inlet. In this case the actual area contributing to
the inlet is 2.06 sq mi, and multiplying by the lumping
factor of 3.96 gives an assumed or lumped area of 8.15
sq mi. Most of the lumped area here is from the
Beltline trunk sewer. The dry-weather connections
from the trunk sewer to the interceptor sewer for both
the Phalen Creek inlet and the Beltline inlet have low
capacities. Only a small portion of the total trunk
sewers capacity can enter the interceptor. For this
reason the lumping of such a large area at the Phalen
Creek inlet does not adversely affect the accuracy of
the model prediction.
C. Routing Phase
General Discussion—The third phase of the study
covers the routing of the flow through the interceptor
system. Available methods permit a much better
prediction of the routing process than they do of
runoff or diversion. The "method of characteristics"
was selected for initial routing of the hydrographs. The
usual methods of hydrologic routing are based primar-
ily on the equation of continuity with constants or
simple relations used to account for momentum
effects. The method of characteristics is one of several
methods of mathematical routing which solve the
equations of momentum and continuity. This method
has been proven in the literature to accurately repre-
sent unsteady, gradually varied flow in open channels.
It was found that for the large system of open
channels which makes up the interceptor system, the
method-of-characteristics solution requires a consider-
able amount of computer time, and therefore a method
was sought which would produce the same answers
using less computer time. It was found that the
"progressive average-lag method" from Ref. (8]
accomplished this objective if the constants used were
determined by comparing input and output hydro-
graphs with those obtained from the method of
characteristics.
Method of Characteristics-The equations of con-
tinuity and momentum for unsteady, gradually varied
open channel flow are
A 8
1 d-A
A 3t
3V
(10)
and
where
V is the mean velocity at any instant t and position
x
y is the depth of flow at any instant t and position
x
A is the corresponding cross-sectional area
x is the distance measured along the channel
t is the time
S is the energy slope
So is the bed slope.
g is the acceleration due to gravity.
These equations can be written in finite difference
form for integration along what are known as the
characteristic curves as follows:
(12)
(15)
The first two equations refer to integration along the
C+ characteristics shown in Fig. 8, while the second
two refer to the C_ characteristic.
The reader is referred to Ref. [9] for the derivation
of these equations. The subscripts in the equations
refer to the points P, R, and S in Fig. 8. The values of
velocity and depth at R and S are found by interpola-
tion between known values at the grid points A, B, and
C. As indicated in Ref. [9], these equations can be
132
-------
"
A
_
Figi
t
L
t
c/
P
\c_
O A R C S B
jre 8. Forward and backward characteristics an
definition of grid points.
solved to determine the velocity and depth at all points
like P which are At later than the points like C, at
which conditions are known from previous calcula-
tions. In other words, the flow can be represented on
the x — t plane, as shown in Fig. 8. At every point on
the grid the velocity and depth must be calculated
using a procedure which is similar to extrapolation.
The boundary conditions are important in this
process, and they give some information in the x — t
plane about the depth and the velocity at all points on
the lines x = 0 and x = L, where L is the total length of
the channel. The boundary conditions, which are
discussed in Ref. [10], are
a. Critical flow at the downstream end of a channel
b. Inflow hydrograph at the upstream end of a
channel
c. Inflow hydrograph in the middle of a channel
d. Junction of a number of channels.
At the end of the channel only one pair of character-
istics equations applies (i.e., either a C+ or a C_
characteristic—see Fig. 9). The other equation in the
case of critical flow, for instance, is
(16)
Thus in the case of critical flow Eqs. (12), (13), and
(16) can be solved to determine Vp and yp. The details
of the derivation of the boundary conditions are given
in Ref. [10].
It was found that the interceptor sewer system
could be broken up into a series of 23 open channels,
each with one of the above boundary conditions at its
two ends. The critical flow condition commonly occurs
at the downstream end of the channel, while input
hydrographs commonly enter through a diversion
structure and drop shaft either at the upstream end of
the interceptor or in the middle of it. The junction
conditions occurs only once.
A typical solution for a channel is shown in Fig. 10.
It can be seen that the hydrograph is attenuated as it
flows downstream and is displaced along the time axis.
Ditcharga
ivo Avvroga Log
had ol
Chorocttrilticf
Figure 10.
Results of routing irregular hydrograph by the
method of characteristics and by the progres-
sive average lag method.
A R C
Figure 9. Characteristic at downstream boundary.
Many different numerical tests have been conducted
using the characteristics routing method on various
channels; a resume of the results is as follows:
1. There is a maximum allowable rate of rise or fall in
the input hydrograph for a given channel subdivided
into certain reach lengths. This rate is at present
found by trial and error.
2. The program has been compared with numerical and
actual results from work at Colorado State Univer-
sity (sponsored in part by the Bureau of Public
Roads) as reported by Mitchell in Ref. [11]. The
numerical results calculated in the present study are
indistinguishable from those calculated by Mitchell,
and both of these agree reasonably well with
measured values.
3. A comparison has been made between results for a
circular cross section and for a rectangular cross
section in a channel of fixed length using the same
input hydrograph for both configurations. The
rectangular channel width was the diameter of the
circle, and the hydraulic radii of the two were the
same when the circular channel was half full. The
outflow hydrographs in the two cases are so similar
133
-------
that the routing process does not seem to depend
too closely on the cross-sectional shape for rela-
tively minor differences in shape.
4. The attenuation of the peak of the hydrograph
depends on the shape and size of the input
hydrograph.
5. The rate of travel of the peak of the hydrograph is
dependent on the shape and size of the input
hydrograph.
Progressive Average Lag Method
As previously mentioned, the method of charac-
teristics requires a comparatively large amount of
computer time, and a simpler method which would
produce the same results was sought to facilitate rapid
computation of hydrographs while rain is still falling.
As mentioned in Ref. [10], quite a number of
different methods were tried, including the Musk-
ingum. The method initially selected as a substitute is
the progressive average-lag method (Ref. [8]), and this
has been incorporated in program UROM 9.
The method has several variations, but it is best
described as shown in Fig. 11. The value of qnj on the
routed hydrograph depends on the average of a number
of equally spaced values of q on the inflow hydro-
graph. In the figure qnj is given by
Generally
nl = 1/3(q,.2+qM
Qnl = - I
(18)
DISCHARGE
TIME
Figure 11. Progressive average lag routing.
where n is the number of points to be averaged. The lag
time tg is determined empirically, from known flow
times, or from the method of characteristics.
Successive use of Eq. (18) will determine the routed
hydrograph at the end of a certain reach. The process
can be repeated over the next reach downstream with.
the first routed hydrograph becoming the input hydro-
graph.
In this project the values of n and tg and the
number of reaches required for a given channel have
been determined by trial and error. The input and
routed hydrographs calculated by the method of
characteristics for the particular channel have been
taken as correct and different values of n and tg tried
until the routed hydrographs coincided. Again refer-
ence is made to Fig 10, where the points plotted
represent the routed hydrograph which best conforms
to the hydrograph calculated by the method of
characteristics. It can be seen that there is good
agreement between the two methods. Similar results
were obtained for other conduits with diameters
ranging from 4.44 to 13.83 ft and lengths from 2000 ft
to 23,500 ft. As many as three input hydrographs have
been used as input data for a particular channel, and
the same routing constants gave good answers in each
case.
It was concluded that the progressive average-lag
method of routing produces the correct routed hydro-
graphs for the channels investigated if the appropriate
routing constants are used. The progressive average-lag
method is considerably faster than the method of
characteristics, and hence it has been used in the
real-time program UROM 9.
In addition to routing flow, the routing phase of the
model must perform two additional functions: the
combining of flow and the diversion of flow. Com-
bining of flow takes place either where additional flow
enters an interceptor at an inlet or where two
interceptors meet. Combining is accomplished by
simple addition of the flows; this follows from the
superposition principle of unit hydrograph theory.
Diversion of flow sometimes occurs after the flow
has already entered the interceptor lines. This kind of
diversion may occur at five points in the interceptor
system to prevent overloading of the main interceptor
lines.
The routing phase of the model contains the
subroutines necessary to perform each of the three
operations—routing, combining, and diversion. A table
of input data determines which of the three operations
should be carried out and provides an index as to
which hydrographs are involved. The steps listed in the
table are carried out in sequence until all operations
have been completed. The program instructions are
thus independent of the area modeled. The data table
is in effect a map that shows the layout of the sewer
system and the locations of the inlets and diversion
structures. Thus by adding entries to or removing
entries from the table, sewer lines can be added or
removed from the system. Inlets and diversion struc-
tures can be added or removed in the same way.
134
-------
IV. DISCUSSION OF METHODS
Throughout the model various methods are used to
simulate the runoff process. Some of these methods are
linear and some are non-linear. Linear methods are
used where the results obtained are believed to be
within the degree of accuracy desired or where the
information is such that only linear methods can be
used.
Initially the precipitation data are given for each
raingage as a function of time at some irregular time
spacing. Linear interpolation is used to determine
equivalent rainfall data spread at regular intervals
(usually 5 minutes). Since the model must operate in
real time up to the latest rainfall reading available, the
interpolation can be based only on readings taken prior
to the current readings—the future trend being un-
known—and a linear interpretation is justified.
The area! distribution of precipitation over each
sub-watershed is found by the Thiessen method. An
alternate method based on isohyetal maps of the
precipitation, according to which the isohyetal lines are
determined by linear interpolation between stations,
has been found by Kwan [12] to give results which
agree well with those of the Thiessen method, and this
agreement served to substantiate the use of the more
rapid Thiessen method.
Determination of the sub-watershed losses and the
resulting excess precipitation is a non-linear process
and has been modeled as such.
The unit hydrograph method is used in this program
to determine the inlet hydrograph resulting from
excess precipitation. It has been shown by Machmeier
and Larson [13] and others that the runoff process is
actually non-linear. With increased excess precipitation
the magnitude of the peak discharge increases non-
linearly. In addition, the time to peak discharge de-
creases. Most of these studies were concerned with
basins possessing "natural" characteristics. For urban
sewered areas other conditions might cause the runoff
process to behave quite differently. Changing control
at different discharges can cause marked changes in the
runoff distribution. This may result from constrictions
in the system, clogging of inlets, or various other
factors. Also, a sewered area is actually served by two
drainage systems, the primary system (i.e., the storm
sewers)—which is designed to handle flows up to a
given design value-and the secondary system (i.e.,
natural or manmade surface drainage), which should be
available to handle flows above the design value of the
primary system. Factors such as these may well mask
out or reverse the commonly expected "non-linear
effects." Thus it was felt that at least for the initial
approach to the formation of the inlet hydrograph, the
unit hydrograph method is justified.
The diversion structure analysis, as shown pre-
viously, is non-linear when the discharges in the trunk
sewer cause overtopping of the fabridam. The use of
the lumping factor to account for area not actually
contributing to a major inlet assumes a linear relation-
ship between the response of the actual contributing
area and the total lumped area. This assumption
provides an easy method of accounting for flow
entering at major inlets. The use of the lumping factor
tends to overemphasize the effect of some of the
variable diversion structures.
The progressive average-lag routing method is en-
tirely linear.
135
-------
V. MODEL PERFORMANCE FOR NATURAL
EVENTS
Rainfall for the months of June, July, August, and
September of 1969 was 7.36 inches. The normal
expected for this period would be 12.88 inches. The
dryness of the 1969 runoff season, in combination with
hardware and software problems of a new computer
system, limited the number of runoff events that were
recorded during 1969. From the data that were
available, four events were selected on which to test
and fit the runoff model. These four events occurred
on Julian days 207, 211, 218, and 267, which
correspond to July 26, July 30, August 6, and
September 24, 1969 respectively. The precipitation
characteristics for each of these storms are shown in
Figs. 12a, b, c, and d, respectively. The areal distri-
bution of precipitation for the entire storm is shown
by the isohyetal lines in the figures. The time distri-
bution is given for four of the recording gages in the
network.
Figure 12a. Temporal and areal distribution of storm
on day 207 (July 26, 1969)
Figure 12b. Temporal and areal distribution of storm
pn day 211 (July 30, 1969)
Figure 12c. Temporal and areal distribution of storm
on day 218 (Aug. 6, 1969)
Figure 12d. Temporal and areal distribution of storm
on day 267 (Sept. 24, 1969)
137
-------
The first event (day 207) was centered north of the
Twin Cities. Intensities of 2 in. per hour were recorded
at gage number 8. The southern-most portions of the
cities received no rain at all. The significant rainfall
lasted about 20 minutes. Rainfall data were recorded at
about 5-minute intervals.
The second event (day 211) was centered to the
west. Moderate rainfall occurred throughout the area.
Intensities reached 0.8 in. per hour in most areas. The
rainfall lasted about 1-1/2 hours. The rainfall data were
recorded at about 5-minute intervals.
The third event (day 218) was centered to the south
of the cities. Intensities as great as 3.8 in. per hour
were recorded at gage number one. The northwest
corner of the cities received only light showers. The
significant rainfall lasted 20 minutes or less. Data were
recorded at about 5-minute intervals.
The fourth event (day 267) was also centered to the
south of the cities. Very mild intensities occurred over
a period of three hours. Data for the early part of the
storm were recorded at one-hour intervals, and there-
fore 5-minute intensities are shown as uniform over a
one-hour period.
The effect of each of these storms on the Minne-
haha inlet will be shown. This inlet is located in the
southwest corner of the Twin Cities, as shown in Fig.
1. Only a portion of the sub-watershed is serviced by
combined sewers contributing to the interceptor sys-
tem. The area serviced is 1.14 sq mi. The farthest point
in the watershed is about 3.7 mi from the interceptor
inlet.
Figures 13a, b, and c show the rainfall and runoff at
the Minnehaha inlet. The average precipitation and
calculated excess (darkened area) is shown at the top
of each figure. The computed hydrograph at the inlet
before diversion and after diversion is shown at the
bottom of each figure.
On day 207 no rain occurred over the southern
portion of the cities. Since only sanitary base flow
would appear at the Minnehaha inlet, a figure was not
prepared for this day.
Inlflt njdrogroph
Diverted nydrograph. lob OOv
OivtrlM ri.droOrOpri, fob up
20 21 22 23
Figure 13b. Inlet hydrograph and hydrograph diverted to
the interceptor sewer at Minnehaha diversion
for event of day 218.
inlet hydrogropri
Divined hydrogioph, lob do*i
Figure 13a. Inlet hydrograph and hydrograph diverted to
the interceptor sewer at Minnehaha diversion
for event of day 211.
Figure 13c. Inlet hydrograph and hydrograph diverted to
the interceptor sewer at Minnehaha diversion
for event of day 267.
On day 211 (Fig. 13a) significant precipitation
excess occurred in two bursts over a period of about 1
hour. All of the excess for this event came from
impervious surfaces. During the storm the initial loss
on the pervious surfaces was satisfied, but rainfall
intensities did not exceed the computed pervious
infiltration rate. In the impervious analysis the loss
coefficient had been reduced to 0.44 by the end of the
storm.
On day 218 (Fig. 13b) a very intense short storm
occurred. In this case excess came from both pervious
and impervious surfaces. The pervious contribution
came during two 5 minute periods when the rainfall
rate of 3.42 in/hr exceeded the infiltration rate of 1.97
in/hr giving a pervious excess of 1.46 in/hr, and the
rainfall rate of 3.79 in/hr exceeded the infiltration rate
of 1.83 in/hr giving a pervious excess of 1.94 in/hr. The
impervious surfaces contributed excess during all per-
iods of rainfall. At the end of the storm the impervious
loss coefficient had been reduced to 0.28.
For the storm of day 267 (Fig. 13c) all the excess
came from the impervious surfaces. At the end of the
storm the impervious loss coefficient had been reduced
to 0.50. The final pervious loss rate was 1.84 in/hr,
much greater than any measured rainfall intensities.
The inlet hydrograph produced from the total
excess is shown by the solid line in figures 13a, b, and
138
-------
c. The precipitation excess for days 211 and 267
occurred over a period of 1-1/2 hours or more and
produced hydrographs with peak discharges of 40 cfs
and 33 cfs, respectively. In contrast, the large excesses
for day 218 occurred in only a 10 minute period, and
produced a peaked inlet hydrograph with a maximum
discharge of 105 cfs.
The clipping action of the diversion structures is
also seen in these figures. Two postulated conditions
are shown in each figure, fabridam up and fabridam
down. On days 211 and 267 (Figs. 13a and 13c) with
the fabridam up, the inlet hydrograph ( ) and
the diverted hydrograph (A A) coincide, indicating
that no flow was released to the river. On the other
hand with the fabridam down ( ) a portion of
the flow would be discharged to the river. Fig 13b (day
218} shows an inlet hydrograph with a peak discharge
of 105 cfs. For this large discharge in the trunk, flow
was released to the river even with the fabridam up.
The interceptor flow from the City of Minneapolis
is metered at the Minneapolis-St. Paul city line. The
location of the meters are shown in Figure 1. The
meters are designated NW meter, SW meter, and E
meter for the respective interceptor sewers they serve.
The metered ( ) and predicted ( )
discharges for the SW and E meters are shown in
Figures 14a, 14b, 14c, and 14d. The NW meter is not
shown because on day 211 and 218 it was found that
150
100
Discharge
(cfs)
50
S.W. Meter
19 21 23
Time (hrs)
150
100
Discharge
(cfs)
50
East Meter
19 21 23
Time (hrs)
Figure 14a. Metered and model predicted flows for event of day 207.
ISO
100
Discharge
(cfs)
50
S.W. Meter
150
100
Discharge
(cfs)
50
East Meter
19 21 23
Time (hrs)
19 21 23 I
Time (hrs)
Figure 14b. Metered and model predicted flows for event of day 211.
139
-------
Discharge
(cfs)
150
100
50
S.W. Meter
150
100
Discharge
(cfs)
50
East Meter
19
21 23 I
Time (hrs)
19 21 23 I
Time (hrs)
Figure 14c. Metered and model predicted flows for event of day 218.
ISO
IOO
Discharge
(cfs)
50
S.W. Meter
Discharge
(cfs)
ISO
IOO
50
East Meter
20
22
0
Time (hrs)
20
22
0
Time (hrs)
Figure 14d. Metered and model predicted flows for event of day 267.
an abnormal backwater condition affected the N\N
meter discharge. This condition was caused by the then
unknown slippage of a gate that controls the flow of
the sewage from the west side of the river to the east
side. The SW meter discharges were effected by this
same condition but to a lesser extent.
The results in Figures 14a, b, c, and d indicate that
generally the runoff process is being modelled ade-
quately but that further adjustment of coefficients or
parameters is necessary. The results shown were ob-
tained with only a few changes to the original values of
the model parameters.
The predicted hydrograph at the headhouse (near
the MSS6 treatment plant) is shown in Figures 15a, b,
c, and d. It is of interest to note that a numeric average
over the two cities of the raingage catches are 0.52"
and 0.51" for days 211 and 218 respectively. Thus, the
same amount of rainfall occurring in different time and
spatial patterns (see Figures 12b and 12c) produced
significantly different runoff patterns and volumes.
The storm for day 211 actually produced a slightly
smaller volume of excess precipitation than day 218.
However, for day 211 a significantly larger amount of
the excess precipitation reached the interceptor and
was retained in the system. This illustrates the obvious
effect that the shape of the hydrograph at the diversion
structure has on the amount diverted to the river. For
the same volume of hydrograph much less is lost to the
river if a flatter peak occurs.
In Figures 15a, b, c, and d the predicted hydrograph
at the headhouse is shown for two conditions. The
solid line ( ) shows the expected discharge
with all gates open and fabridams up. The dashed line
( ) shows the expected discharge with all gates
open and fabridams down. For the storms on days 207,
211, 218 and 267, the percent increases of volume
retained in the sewer system were 16%, 25%, 41%, and
30%, respectively. As stated earlier the lumping of
some areas at major inlets increases the apparent
impact of the variable diversion structures. Thus the
percentages given above are probably somewhat higher
than the true values, but it may be stated that there is a
significant increase in the volume retained in the
system when the fabridams are raised.
140
-------
800
600
Discharge
(cfs)
400
200
Day 207
1200
900
Discharge
(cfs)
600
20
300
Day 211
22 0 2 4 19 21 23 I 3
Time (hrs) Time (hrs)
Figure 15a. Effect of fabridam on model predicted flows Figure 15b. Effect of fabridam on model predicted flows
reaching waste treatment plant for event of day reaching waste treatment plant for event of day
207. 211.
1200
900
Discharge
(cfs)
600
300
Day 218
Fab. up
Fab. down
800
600
Discharge
(cfs)
400 -
200
19
21
23
Time (hrs)
Fab. up
Fob. down
Figure 15c. Effect of fabridam on model predicted flows Figure 15d. Effect of fabridam on model predicted flows
reaching waste treatment plant for event of day reaching waste treatment plant for event of day
218. 267.
141
-------
VI. CRITIQUE OF MODEL
The material in the previous sections describes the
initial methods used in the model and the overall
performance of these methods. Several assumptions
have been made in developing the model. These
assumptions will be discussed in this section and in
addition suggestions for improvement of the model will
be given.
1. The modeling of the diversion structures is com-
plicated by the rubber fabridams. This device is
raised and lowered pneumatically. A record is
maintained of the air pressure in the fabridam. This
gives an indication of the position of the fabridam
which is modeled as a variable height weir. Presum-
ably a given fabridam pressure would be associated
with a given fabridam height. This is not necessarily
true as water build-up behind the fabridam also
causes a change in the fabridam pressure. Thus, it is
difficult at best to estimate the fabridam height
from the fabridam pressure. In addition the rubber
fabridam is deformable. The fabridam crest is often
not horizontal and may actually be depressed in a
vee shape on the sides or in the center under partial
inflation conditions.
2. The present spacing of the remote indicating rain-
gages could be improved. The gages are generally
spaced around the Twin City area. For modeling
purposes a network centered within the area would
be more desirable. Presently gage spacing averages
about 5 miles, allowing significant variations in areal
precipitation to go undetected.
3. The present analysis does not properly model the
system for relatively large rainfall events. An urban
area is actually served by two drainage systems. The
primary one, the sewer system, is designed to handle
a certain amount of runoff. If the runoff exceeds
the capacity of the sewer system, local ponding and
surface runoff will occur. This surface runoff will
follow natural or man-made surface channels until it
finally enters another sewer or some other surface
drainage facility. In the model all runoff is assumed
to enter the sewer system and arrive at the diversion
structure. This is a valid assumption for small and
moderate amounts of runoff (up to design values).
For larger amounts where ponding and "natural
diversion" occur the model will predict excessive
flows in the trunks. While the diversion structures
do tend to limit the flow to the interceptor there is
still no upper limit on the amount they will predict
that goes to the interceptor. Thus, for unreasonably
large amounts of trunk flow the model will predict
reasonable but still too large values of interceptor
flow.
4. Each time the model is executed the initial moisture
conditions in the area are taken to be zero. This
would be true only when all surface depressions
were dry and the upper soil moisture was at the
wilting point. Actual initial conditions should be
established before each real storm event. For moder-
ate rains where only runoff from impervious areas is
significant, a dry initial condition is still a reason-
able assumption.
5. The use of separate pervious and impervious loss
functions is desirable. Although the constants in
each function have been selected by judgement
only, the analysis indicates that the most significant
runoff contribution in an urban area comes from
the impervious area. The pervious areas contribute
little runoff during the less intense rainstorms. The
constants in the loss functions should be sub-
stantiated by runoff data.
6. The present model may be used for various pur-
poses; real time operation aid, system evaluation,
design evaluation, etc. As an operational tool, the
accuracy and reliability of the model may be
increased by the addition of a feedback correction
based on measured water levels in the sewer system.
The model in its existing form does not utilize the
recorded water level data to aid in its prediction.
This feature would be of use during real time
applications. When evaluating postulated storm
events, recorded data would not exist and the
prediction would not include feedback correction.
7. If excessive flow is indicated in the interceptor
system, the console operator must decide what gates
or fabridams should be manipulated to alleviate that
condition. A gate and fabridam operation scheme
may be added to allow the machine to decide what
the best remedy would be.'The operator could still
decide whether or not to implement the recom-
mended gate or fabridam movements.
8. The addition of storage in the trunk sewer behind
the fabridams or at other locations would bd
another desirable revision to the model. Presently
storage in the trunk sewers is reflected only in the
shape of the unit hydrograph. Accounting for
storage in the trunk sewers would be of importance
only where the storage volume is significant com-
pared with the volume of the runoff hydrograph.
9. Only the 15 major inlets to the interceptor sewer
are modeled. The flow that enters at other points in
the system is assumed to contribute at the nearest
major inlet. It is assumed that the additional flow
lumped at the major inlets will occur in direct
proportion to the relative land areas contributing.
That is, if the actual area contributing to a major
inlet is x sq mi, and an additional area of y sq mi is
lumped at this inlet, the computed discharges based
on the actual area x will be increased by the ratio
-—¥ to represent the flows from the total lumped
x
area (x + y). This lumping correction is applied im-
143
-------
mediately following the diversion analysis at each
inlet.
10. For verification and fitting purposes it is important
to have recorded runoff data corresponding to given
rainfall events. Presently water levels are recorded in
the trunk and interceptor sewers in addition to
venturi metered flows in the interceptor sewers at
the Minneapolis-St. Paul city line. These recorded
values are needed to assess the discharges entering or
within the interceptor system. They are of less use
for assessing the runoff discharges in the trunk
sewers. This is due to the problem of modeling the
fabridam which acts as a control section in these
lines. For this reason it would be very desirable to
install a fixed control section (weir) at some point
in the trunk sewer that would enable the total
runoff to be measured. This would allow significant
improvement in the loss rate and runoff prediction.
144
-------
REFERENCES
1. Bowers, C.E.; Harris, Garth S.; and Pabst, Arthur
F., Status Report on "The Real-Time Computa-
tion of Runoff and Storm Flow in the Minne-
apolis-St. Paul Interceptor Sewers," Memorandum
No. M-118, St. Anthony Falls Hydraulic Labora-
tory, University of Minnesota, December, 1968.
2. Handbook of Applied Hydrology, Ven Te Chow,
Editor, McGraw-Hill, 1964.
3. Huggins, L.F.; and Monke, E.J. "The Mathematical
Simulation of the Hydrology of Small Water-
sheds", Technical Report No. 1, Water Resources
Research Center, Purdue University, August 1966.
4. Kent, K.M.; "A Method for Estimating Volume
and Rate of Runoff in Small Watersheds," USDA,
SCS-TP-149, January 1968.
5. Harris, Garth S., Status Report on "Real-Time
Estimation of Runoff in the Minneapolis-St. Paul
Metropolitan Area," Memorandum No. M-119, St.
Anthony Falls Hydraulic Laboratory, University
of Minnesota, December 1968.
6. Anwar, H.O.; "Inflatable Dams", Journal of the
Hydraulics Division, ASCE, Vol. 93, No. 3, May
1967.
7. Harris, Garth S., Status Report on "Mathematical
Models of Major Diversion Structures in the
Minneapolis-St. Paul Interceptor Sewer System,"
Memorandum No. M-120, St. Anthony Falls
Hydraulic Laboratory, University of Minnesota,
December 1968.
8. Hydrograph Combining and Routing, Hydrologic
Engineering Center, U.S. Army Engineer District,
Sacramento, California, August 1966.
9. Streeter, V.L. and E. B. Wylie, Hydraulic Tran-
sients, McGraw-Hill, New York, 1967.
10. Harris, Garth S., Status Report on "Development
of a Computer Program to Route Runoff in the
Minneapolis-St. Paul Interceptor Sewers," Memo-
randum No. M-121, St. Anthony Falls Hydraulic
Laboratory, University of Minnesota, December
1968.
11. Mitchell, James S., "Comparison of Mathematical
vs. Experimental Flood Wave Attenuation in Part-
Full Pipes," Colorado State University Engineering
Research Center, March 1967:
12. Kwan, J.Y.; Riley, J.P.; and Amisial, R.A., "A
Digital Computer Program to Plot Isohyetal Maps
and Calculate Volumes of Precipitation," Pub. No.
80, IASH Symposium at Tucson, Arizona, Decem-
ber 1968.
13. Machmeier, Roger E. and Larson, Curtis L.,
"Runoff Hydrographs for Mathematical Watershed
Model," Journal of the Hydraulics Division, ASCE,
Vol. 94, No. 6, November 1968.
145
-------
Appendix A
UROM-9 OPERATOR'S MANUAL
Section I - Introduction
Section II - Operation of CREATE
1. Input data file preparation
2. Execution of CREATE
— Operation of Mathematical Model —
Section
UROM-9
1. Input data
2. Sense switch options
3. Program control
4. Gate editor
5. Execution of model
6. Loader Memory map
(UROM-9)
of CREATE and Model
147
-------
I. INTRODUCTION
UROM-9 (Urban Run-Off Model) is a digital com-
puter program written for execution on the Minne-
apolis-St. Paul Sanitary District's Digital Equipment
Corporation PDP-9 computer. The purpose of this
program is to predict discharges in the Minneapolis-St.
Paul Interceptor Sewers given rainfall readings at
various points around the Twin Cities.
UROM-9 acquisitions its own data when operating
on a real time basis. Instructions pertaining to the
operation of the data-gathering section of UROM-9
(RTIME) are not given in this manual. References are
made to RTIME only to clarify its relationship to
UROM-9.
1. Method
The rain that occurs at the raingages is first operated
on by a subroutine that determines rainfall losses. The
excess rainfall is then applied to a unit hydrograph for
each of the catchments modeled.
The resulting composite hydrograph represents the
flow that would arrive at the major diversion structures
of the inlets to the interceptor sewer system.
The flow then enters the interceptor. Any flow not
able to pass into the interceptor is discharged to the
Mississippi River. The flow entering the interceptors is
increased by a lumping parameter to account for flow
that does not really enter at one of the 15 major inlets
modeled.
The flow in the interceptor is then routed and
combined with other entering flow until it reaches the
treatment plant.
2. CREATE Routine
The working section of UROM-9 requires many
items of input data that describe the hydrologic,
diversion and routing constants of the modeled area.
These data are fixed for a given area and are handled
separately from information (such as rainfall readings)
that pertain to a given storm event. CREATE is a
one-chain-execute file, run prior to the actual model
program UROM-9, that prepares the fixed data in a
form that is easily used by the model.
From the given input data file on the disk (DAT 3)
CREATE lists the input data on the line printer (DAT
5) and then writes 4 binary output files on the disk
(DAT 8). Each of these 4 binary files is used by a
different chain of the model.
CREATE also does a small number of calculations.
The data are not screened for errors other than those
detected by the system routines.
When new input data are to be run by the model, or
if old data have been changed, it is necessary to then
execute CREATE before running the model. Once
CREATE has been executed the model may be run any
number of times.
3. Structure of Model, UROM-9
The model itself (UROM-9) is composed of 6
chains, or sections. (See Fig. 1) At any given time only
one chain may be loaded into core storage. The
remaining chains are held on disk storage until called
by the chain in core. The data in unlabeled common
storage are retained in core from chain to chain.
a. Chain 1
This chain initializes the model and is always
executed first. Its functions include initializing vari-
ables and constants, loading the common storage area,
setting the real time clock, setting initial loss rate
parameters to zero, inputting the rainfall file to be
processed, and printing instructions pertaining to sense
switches used by the model. Control is always passed
to Chain 2 upon completion of Chain 1.
Use of sense switches is explained in detail in III.
b. Chain 2
Chain 2 is the basic control chain for the model.
After any other chain is completed, control is always
passed to Chain 2. Chain 2 then determines the proper
chain to be called next. This decision is based on the
previous chain processed and the positions of the
various sense switches selected by the operator. In
addition the PROGRAM CONTROL routine within
Chain 2 allows the operator to override the normal
processing sequence. Use of the PROGRAM CON-
TROL options is explained in detail in Section 4.
If real time data acquisition is desired, the scanning
chain (RTIME) (Chain 9) is called each time Chain 2 is
entered. Upon completion of scanning, control is
returned to Chain 2 for further model processing.
149
-------
This chain also contains the GATE EDITOR rou-
tine. This routine is explained in Section 5.
c. Chain 3
This chain begins the actual model calculations.
When processing under real time conditions, initially
data 2 hours prior to the current real time are used.
Once this initial block is processed-. Chain 3 updates
the inlet hydrographs based on rainfall readings that
are being acquisitioned in real time.
When historical data is being processed, processing
begins at the time specified by the operator, and each
prediction is updated by exactly one hour of rainfall
data.
Output from Chain 3 may include the loss rate
calculations and the inlet hydrographs at the 15 major
inlets to the interceptor sewers.
d. Chain 4
Chain 4 performs the diversions of inlet hydro-
graphs to determine the flow to the interceptor sewers.
Fourteen of the inlet hydrographs are diverted by this
chain. The 15th inlet hydrograph and the flow at 4
additional locations within the interceptors are di-
verted during the routing process of Chain 5.
Chain 4 always uses the hydrographs produced by
the last run of Chain 3 for input data. Chain 4 may be
run any number of times with different gate and
Fabridam settings without re-running Chain 3.
Output from Chain 4 may include the diverted
hydrographs at the major inlets, the diverted hydro-
graphs after lumping of additional flow at the major
inlets, and the volumes of the undiverted and diverted
hydrographs.
e. Chain 5
Chain 5 generally performs the routing and com-
bining operations. In addition it may divert flow at 5
points within the interceptor sewers. Its output is the
interceptor flow at various locations.
f. Chain9 (RTIME)
This chain handles the acquisition of data. It is run
each time control transfers from one model chain to
the next. A detailed discussion of the operation of
Chain 9 is not given in this manual.
150
-------
II. OPERATION OF CREATE
1. Input Data File
The input data file (INDAT8 LSI, INDAT9 LSI or
others) resides on the disk (DAT 3). Changes to these
files should be made with the system EDITOR Pro-
gram. A sample listing of INDAT9 LST is given in
exhibit 1. In general, the constants pertaining to
hydrology are given first, then diversion, and finally
routing.
Each time CREATE is executed a complete listing
of input data and associated headings is given on the
line printer (DAT 5). This listing, given in the next
part, contains the same information that appears in the
input data file, but in a different format, and includes
titles and other identifying information.
When making changes to the input data a copy of
the output from CREATE will aid in identifying the.
parameters, but an actual listing of the data file is
necessary where the proper format is in question.
INDAT8 LST is the standard input data file for the
8 raingage network.
INDAT9 LST is the standard input data file for the
9 raingage network.
Exhibit 1
Sample input data file used by CREATE (INDAT9LST)
900.
IB 00.
810.
900.
1000
ueo
1060
1200
1200
1800
1800
1600
11.00
3600
2000
IS 100.
7200.
16200.
2230.
3300.
10800.
3960.
2880.
7800.
18000.
7200.
moo.
12000.
10000.
7200.
6600.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0 .00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
0.00 0.02 2.00 0.
57200000.
98900000.
29600000.
1)6400000.
75900000.
49100000.
25 100000.
5)400000.
2940000C.
36800000.
76600000.
31600000.
64600000.
61800000.
12800000.
1244 641
1641 1144
1441 1141
1511 1644
1144 1441
1 141 1641
1644 1244
641 1511
633 6)3
95 0.31 0.
95 0.55 0.
95 0.30 0.
95 0. 33 0.
95 0.33 0.
95 0.36 0.
95 0.30 0.
95 0.52 0.
95 0.30 0.
95 0.39 0.
95 0.30 0.
95 0.38 0
95 0.29 0.
95 0.3 1 0
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
70 0.20
2.00 0.
2 . 00 0.
2.00 0.
2.00 0.
.00 0.
.00 0.
.00 0.
2.00 0.
2.00 0
2.00 0.
2.00 0,
2.00 0
2.00 0
2.00 0.
2.00 0
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
0.25 0.00
1 .00
1 .00
1 .00
1 .00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
).96
1.06
I. 58
1.52
1.0
1.6
2.85
1.0
1.82
2.73
1.2
1. 13
1.33
1.48
1.56
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
2.00 0.00
0.0
.2
0.0
0.4
0.6
0.2
.2
0.0
1.
0.)
0.4
1.0
0.8
0.6
0.4
0.1
.3
0.5
151
-------
.00
0.0
0.0
PH AL N
III GUI
HELLO
RICE
HARSH
EUST 1
OAK
IUNOO
18IH
HINNE
26TH
6IH
42 ND
2ND
EASin
NU N
SU M
1
200
0
168
)
0
J.
0.
0. b
.026
0.
6.
1.5
0.
0.
2.5
0.
0.
-1 .7
0.
0.
1.5
1. 7
9. 67
0.
0.
48
4
1
1
2
3
1
2
1
4
1
2
1
I
)
2
1
4
1
]
1
2
1
3
1
2
1
2
1
4
1
2
1
2
I
i
I
2
1
2
1
5.97
9.66
5. 17
6. 7
22.4
5. 8
1. 02
0. 70
1.20
1.11
.05
0.9
0.5
2 4
200 200
0 «4
272 242
1 0
0 0
].
1.
0.6
.026
.026
6.
1.5
1.5
0.
i.»
3.
0.
-2.S5
0.
0.
5.5
9.25
10.
10.
0.
15
15
15
15 14
15 0
15
IS 1}
15
12
li
12 1 1
12
12
15
15 12
15
10
10
10
10
10 9
10
10
10
10 «
10
15 10
15
7
I
1 6
7
15 7
IS
15 S
15
4
4
3
J
) 4
4
15 4
15
15 2
15
15 1
15
0.0007)
0.0007
0.00095
16.0
11 .5
0.91
0.69
I.2J
1.08
.05
.5
.0
5 6
7 }
0 0
10 40
2 1
J.I
3.
0.6
3.
.026
».
.6
1.5
0.
3.
2.75
0.
0.
0.
0.
9.5
2.5
5.T
10. 5
0.
15
15 2
15 2
15
15 12
15 6
15
15 2
12
12 6
12
12 a
12 17
15 16
15
15 2
10
10 6
10 14
10 6
10
10 15
10 1 5
10 6
10
10 )
15
15 2
7
7 5
7
7 2
15
15 2
15
15 2
2
4
3
15
15 B
15
15 2
15
15 8
0.012
0.012
0.012
1.8
10.2
0.43
0.73
1.25
1 .04
1 .
8
0
0
).
).
1.
1.
).
4.
-6
0.
4.
9.
8.
5.
-11
1
1
-7
0
-24
-I
7
7
1
-10
1
-4
1
7
0
0
0
1
1
-10
1
1
- 14
0
1
1
1
b.
16
0.
0.
1 .
0
9
0
1
026
026
5
5
.5
1
25
3
0
6
1
10
2
1
2
1
12
5
1
3
2
2
1
2
1
120
1
1
6
1
1
1
1
1
.0
98
82
26
10 1 1
18 0
0 1
3.
1.
.026
.026
1.5
1.5
3.
8.
-0.7
-0.6
4.25
9.75
7.5
0.
10.0
150.
250.
1290.
0290.
4290.
5360.
6360.
20.0
710.
8165.
9165.
10165.
11360.
12530.
13530.
5.0
1460.
1S60.
1660.
17125.
18125.
19125.
20125.
21200.
22200.
23600.
24600.
2519.3
2690.
2790.
28690.
29690.
30690.
31690.
3270.
3355.
34170.
3S170.
36820.
37820.
38820.
39820.
40900.
41900.
8.9
32.0
0.83
0.94
1.24
12 13 14 15 200
0000 209
11101
3. 3.
4.1 3.
1.5 .026
28.5 .026
.171 1.5
.6 1.5
0. 2.750
4. 4.
0. 0.
0. 0.
0.6 6.
7.5 9.0
0. 7.
42U-I
4 T Hill
6TMUA
6tMU«
6TWI
26S4U
26S4A
NUICT
nlNMi
38 IH
38 (•
SUM 1
SUNED
HUnfD
HE ICO
rflUH
RKLf 1
2NIMA
2NDHA
OAK
0»
ENEI
E1EIO
EUSI 1
EUSI 1
nEJUN
tIEJUC
HRBJN
jEFf 1
HUSH
HARSH
HR8JN
HRBJC
RICf
RICE
UIBIN
HELfll
UABJN
OAMII
VHOJN
UtBJC
UAB1N
UABIC
mriR
INIIC
IMIPH
IHIPC
INI«H
3.1 1.8
14.5 24.0
0.19 0.74
I.O5 1.14
1.20 1.15
46 44 39 37 34 30 29 24 20 17 10 9
3 1
152
-------
2. Execution of CREATE
CREATE is an execute file that normally resides on
the disk unit. It is run from the system monitor by the
command
$E CREATED)
(where ) indicates a carriage return).
The message then appears on the console;
ENTER NAME OF INPUT DATA FILE
one of the input data file names should then be typed
by the operator as,
INDAT9LSO
If the name given by the operator is valid the
following message appears, otherwise another file name
is requested.
ENTER TIME AND DATE (15AS)
The time, date, and any other information may then
be entered on one line. Whatever is entered appears on
the Line Printer listing for identification purposes.
Upon completion the following appears on the
console:
STOP 000000
and control then returns to the system monitor.
An example of an execution of CREATE is shown
in exhibits 2 and 3.
Exhibit 2
Sample teletype output from CREATE
MONITOR V4B
SE CREATE
ENTER NAME OF INPUT DATA PILE
RA
ENTER NAME OF INPUT DATA PILE
INDAT9 LST
ENTER NAME OF INPUT DATA PILE
INOAT9LST
ENTER TIME AND DATE C15AS)
12S42 OCT 8/1970
STOP 000000
Exhibit 3
Sample line printer output from CREATE
121 «2 OCI S.I970
INPUT DM* FILE USl 0 - INDAT9LSI
NO OF GAUGES 9
NO OF CACHHENIS 15
II IE INCKEnENI OF RAINFALL HEADINGS
TIKE INCREMENT OF HTOROCRAPHS ) Ou .
100. SECS
iECS
II HE 10 PEAK
900.
1800.
»3u.
too.
1000 .
1480.
1060.
\ 100.
woo.
laoo.
1 BOO.
)600.
11.00.
lt.00.
200(1.
BASE TIME
7200.
16100.
2230.
1)00.
10800.
I960.
2S80.
7AOO .
18000.
7200.
14400.
12000.
10000.
7200.
b600.
CACn
;
10
1 1
1 2
1 J
1 4
1 5
LOSS DATE VARIABLES
NO. TSTLOS RECOV All OECA! PE5V PfRVCF cHOK I BMC B.1L SILOS t'.ILO<°
SCLUSP COEF POKER ACCUSS
1 0.00 0.02
2 o.oo o.o;
) 0.00 0.02
4 0.00 0.02
5 0.00 0.02
2.00
2.00
2.00
2.00
2.00
0. 95
o.v5
0.95
0.95
0. 95
0. )1
O.-S
J. 10
0. ))
0. 31
0. 10
0. 70
0. '0
J. 70
o. ro
3.20
J.20
0.20
2. 00
;. oo
•. oo
2. 00
2. 00
0.00
J.OO
0.00
0.00
0.00
0.00
3.00
o.oc
0.00
0.00
3. 25
0.2S
0. 25
0. 25
0. 25
0.00
0.00
0.00
0.00
0.00
1.00
1.00
1.00
1.00
1.00
2.CO 3.00
2.00 3.00
2.00 0.00
2.00 0.00
2.00 0.00
153
-------
6 0.00 0. 02 2.00 0.11 0.16 0.70 0.20 2.00 0.00 0.00 0.25 0.00
7 0.00 0.02 2.00 0.95 J.JO O./O 0.20 t . JO 0.00 0.00 3.25 0.00
8 0.00 0.02 2.00 9.95 0.52 O.fO 0.20 2.00 0.00 0.30 0.25 0.00
9 0.00 0.02 2.00 0.95 0.10 d. 70 0.20 £ . UO 0.00 0.00 0.2S 0.00
10 0.00 0.02 2.00 0.9V 0.39 0.70 0.20 2.00 0.00 0.00 0.25 0.00
11 0.00 0.02 2.00 3.95 0.30 0.70 0.20 .'.00 0.00 0.00 0. 2S 0.00
12 0.00 0.02 2.00 0.95 0.38 0.70 0.20 I . 00 0.00 0.00 0.25 0.00
11 0.00 0.02 2.00 0.95 0.29 0.70 0.20 .'.00 0.00 I). 00 : . 25 0.00
1* 0.00 0.02 2.00 0.95 0.31 O./O 0.20 J . 00 0.00 0.00 0.25 0.00
15 0.00 0.02 2.00 0.95 0.33 O./O 0.20 ' . OC 0.00 0.00 0.25 0 00
IREASISO FTI FKIOHS CUCH.NJ
57199488. .90 i
98899488. . 06 2
295997*4. .58 ]
1)6400000. .52 4
75899488. .00 5
49099744 . .00 b
25099872. .85 7
5 J 399744 . . 00 8
29199744. .32 9
1 6799744. . 7J 10
'6599488. .20 11
11599744. .1) 12
64599488. . JJ 11
61199488. .48 14
1 2799936. .58 15
RMNGACE ADDRESSES IN ORDER USED BT MODEL AND SCANNED
1 1244 641
2 1641 114
1 1441 114
4 1511 164
5 1144 144
6 1141 164
7 1644 124
8 641 1511
9 631 6))
RUN GAUGE PROPORTIONS
GAUGE NO 12 ] 4 5 6 7 B 9
0.000 0.000 0.000 0. 000 0.000 0.900 0.100 0.000 0.000 1
0.000 0.000 0.000 0.000 0.000 0.100 0.400 0.100 0.000 2
0.700 0.000 0.000 0.000 0.000 0.000 0.111(1 0.000 0.000 3
0.000 0.000 0.000 0.000 0.000 0.000 0. '00 0.000 O.JOO 4
0.000 0.000 0.000 0.000 0.000 0.000 0.500 0.100 0.400 5
0.000 0.000 0.000 C.OOO 0.000 0.000 0.000 0-001) 1.000 0
0.200 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.800 7
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.400 0.600 8
0.000 0.000 0.000 0.000 0.000 0.000 U.OOO 0.600 0.400 9
0.000 0.000 0.000 0. '00 0.000 0.000 C.OOO 0.200 0.1UO ;0
0.200 0.300 0.000 0.000 0.000 0.000 0.000 0. 000 0.500 11
1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 12
0.000 0.050 0.050 0.003 0.000 3.000 0.000 0.000 0.900 13
0.000 0.000 0.500 0.000 0.000 0.000 D.OOO 0.000 0.500 14
0.000 0.000 0.000 1. 00.) 0.000 0.000 0.000 0. OOu 0.000 15
HTDKOCRAPH NO AND DIVERSION NO
1 1
2 2
4 1
5 4
6 5
8 6
9 7
10 a
11 9
12 10
11 11
14 12
15 1)
200 14
200 15
200 16
200 17
7 18
3 19
INDEX OF DIVERSION POINT
0 1
0 2
94 J
0 4
0 5
0 6
0 7
18 8
0 9
0 10
0 11
0 12
0 13
209 14
16> 15
272 16
242 17
30 18
40 19
.00 2.00 0.00
.00 2.00 0.00
.00 2.00 0.00
.00 2.00 0.00
.00 2.00 0.00
.00 2.00 0.00
.00 2.00 0.00
.00 2.00 0.00
i.OO 2.00 0.00
1.00 2.00 0.00
154
-------
ORIFICE ITPE
) 1
I 2
0 )
2 4
1 S
0 6
1 7
0 8
1 9
1 10
1 11
1 12
0 1 I
1 14
0 15
0 16
0
1
2
3
4
5
6
7
8
9
1 1
12
1 3
15
16
1 7
Cl i
17
2
2
3
2
2
2
2
Cl
99998
29998
99998
9999B
99998
99998
0.00000
2
2
2
4
0
0
0
99998
99918
99998
09996
60000
60000
60000
C2
0.02»00
2.19998
0
0
1
0
0
0
0
0
28
5
5
5
1 .69999
F«UR IDA
1
2
3
10
11
12
13
14
1 5
16
17
7
7
0
9
10
02600
02600
49998
02600
00000
02600
02600
02600
49957
99996
99V96
99993
C 14 9
I
0
\
i
0
1
0
1
1
1
0
0
0
0
C3
49999
60000
49999
41999
1 7800
49999
00000
49999
49999
41999
60000
00000
00000
00000
24984 C15
C.
J. 49997
. .99998
2.9)998
2 .99998
0 .00000
2.7*995
0.00000
2 . 99998
i. 9 99 95
7 .99990
1.9)995
0.00000
0 .00000
0.03000
2 .49997
-1
0
-5
-0
0
0
0
0
0
-0
0
0
0
0
C5
69998
00000
4999 3
70000
00000
00000
00000
00003
00000
60000
00000
COOOO
COOOO
00000
Co
1 . 49999 5
C7 ca
. 499 90 9 . 49990
OUnETEHS
66990
99184
69996
99990
74984
49990
99990
00000
99984
49984
7.99990
7.49190
0
6
0
0
0
00000
99990
00000
00000
00000
UPERAMON TABLE.
OPE
R
1
2
3
4
5
6
7
a
9
10
1 1
1 2
1 3
14
15
16
1 7
IB
19
20
21
22
2 J
24
25
26
27
28
29
30
31
32
) 3
34
35
36
37
38
31
40
41
42
4 3
44
45
46
4 7
48
HTOROCRAPH
Hit) ENTER
4
1
I
2
3
1
2
1
4
1
2
1
3
3
2
1
4
1
1
1
2
1
1
2
1
2
1
4
1
2
1
2
1
2
1
4
1
4
1
2
1
2
1
2
1
2
1
NOS AND KG'JIIHG CUKSIS
HID COMB
15
15
15
15
15
15
15
15
12
1 1
11
U
12
15
!<;
IS
13
13
10
10
19
10
13
13
10
10
15
15
J
I
r
i
15
15
15
15
,;
/,
1
t
1
i.
IS
s
«,
i;
'i
!i
HTO LE A Vc NST RL
0
3
3
14
0
0
1 3
0
0
0
11
0
0
0
12
0
0
0
0
0
9
0
0
0
H
0
13
0
0
U
6
0
/
•J
5
3
0
0
0
0
4
0
4
0
2
0
1
0
.5
15
|5
15
1-5 I
15
15
15
12
1 2
i2
1 2
1 2 1
1 5 1
15
i 5
10
13
10 1
10
10
10 1
0
2
2
0
2
6
0
2
0
n
0
a
7
6
0
2
0
o
i
6
0
•,
.0 15
.0
lO
10
15
1 5
r
r
t
i
15
1 •>
IS
15
4
*
1
I
4
4
15
15
15
t 5
15
;5
6
0
t
0
2
0
5
0
2
0
J
0
J
0
2
0
4
0
3
0
a
0
2
0
d
LAC
0
- 1 I
1
0
1
- 7
0
0
0
-2 <.
0
7
J
0
1
0
-10
1
- 4
0
I
7
J
0
0
3
0
0
I
0
1
0
- 1 0
0
1
0
1
0
-14
0
0
0
1
0
1
C
1
N?C HS
0
25
1
0
0
6
0
1
0
I 0
0
2
1
2
0
1
0
1 2
0
5
0
1
3
2
0
2
0
1
0
j
0
1
0
120
0
1
0
1
0
6
0
J
0
1
0
1
0
1
Ofl AX
I0.042N-I
50.04THNI
50.06IMUA
2)0. 06THUA
250. 06 1HUA
2S0.02t>SEA
360.026SEA
36U.ONUHEI
20.0MINMI
70. 0 J 8 1 h
165.038TH
165 . OSUME T
165.0SUKED
)LO. ONwntD
5 lO.OrtETCO
5 10.0METON
S.ORMM
60. 02 NOnA
60.02NCHA
60. OOAK
I25.00AK
l<5.0f IE!
125.0EME10
125. OEU^>; 1
200.0EUSI 1
200. Or.E JON
600.0ME JUC
600. OrtRB JN
0. OjEff I
19. } MAHS
lO.OMAR'.M
°0. OMRdJN
690. O.lfiHJC
610. OK ICE
6HO.OR ICE
6*rO. OUAfll N
0. OkELUI
'0. OWAUJN
O.OOAVIll
•)5.0UABJN
1 ro.OUABJC
170. OUABIN
SJO.OUABIC
820. 0 INT TR
820.0IN1IC
820. 01 NIPH
900. 01 NTPC
900. 01 NIHH
155
-------
d«SE HOU
1
2
3
4
5
6
7
a
i
10
ii
12
11
It
IS
OIAH : 5.97 SLOPE = O.U007J BANKINGS N = O.OU
OIAM = 9.66 ^LLOPE = 0.00070 MANNINGS N = O.U12
01 AH = 5.97 SLUPt = 0.00095 HANNINCS N = O.Sll
6.7
16.0
1.8
6.1
«.»
).l
1.6
Ji.«
11 .S
10.2
16.0
32.0
14. S
24.0
S.B
HOURL* BASE FLOU f AC TORS
NA KE
10
II
12
1)
14
15
HP
IIB
1
I
)
t.
5
6
;
a
9
10
11
12
13
14
15
16
17
16
19
20
21
22
23
24
INOEI
46
44
39
Jl
34
30
29
24
20
17
10
9
6
3
1
3
24
1.02
0.98
0.93
O.B8
0.91
0. 7f
0. 74
0. 70
0.69
0. 73
0.«2
0.94
I.OS
1 . 14
1 .20
1.23
1.25
1.26
1.24
1.20
1. 15
1.11
1 .06
1 .04
6
54
3 10 S 4 4 II 6 6 12 S 12 7
II 30 13 10 26 60 24 48 40 31 24 22
156
-------
OPERATION
UROM-9
OF MATHEMATICAL MODEL-
1. Input Data
All input data that are fixed for all events are
processed by CREATE as explained in Section I. These
data are prepared as 4 binary files. The first file
(STARTLIST) is used by Chain 1 to initialize the
model. The remaining files, HYDROLIST, DIVER-
LIST, ROUTELIST, are used by Chains 3, 4, and 5
respectively.
In addition to the fixed data the model also requires
input data consisting of the rainfall at the remote
gages, the gate and Fabridam settings to be used, and
the initial moisture conditions before the storm.
When processing real time data the model uses the
file RAINRP DLY to obtain its rainfall data. New data
are continually being added to this file by chain 9
(RTIME).
When historical data are being processed, rainfall
data are obtained from any file specified by the
operator. If new real time rainfall data will also be
acquisitioned, they will be put in the file RAINRP
DLY.
The gate and Fabridam settings that the model will
use are stored in a file GATESETTG. This file contains
the times and positions of gates and Fabridams, as well
as the maximum and minimum values allowable. All
positions are in feet. This file may be changed by the
GATE EDITOR routine (part of the model) or by the
systems EDITOR program.
The current values of basin moisture and starting
loss are found in the file BMCXSTLOS. This file
contains for each of the 15 major catchments, the
Basin Moisture Content (BMC) and the starting loss
(STLOSP) for the pervious areas and accumulated loss
(ACLOSS) and starting loss (STLOS) for the imper-
vious areas. These are given in inches. The last entry of
the file is the accumulated time in seconds since the
DECAY (a recovery factor) has been applied to update
the moisture values.
2. Sense Switch Options
Accumulator switches located on the lower right
side of the central processor are used as sense switches
by the model. All sense switches should be off (down)
at the beginning of execution of the model unless the
operator is familiar with the program. The use of each
sense switch is described below.
Sense Switch Number
# 0 Not presently used
# 1 Normally if all sense switches are down the only
printed output on the line printer will be a
message indicating when the model predicts that
there will be excessive flow at a given point. If
this switch is up this message will not be printed.
# 2 If this switch is up the model will return to
"Program Control" if excessive flow is predicted
anywhere in the system. This then allows the
operator to postulate possible corrective changes
or to continue calculations as he desires.
#3 If it is desired to see the rainfall rates at the
raingages for each 5-minute period, the rainfall
rates over each catchment, the loss rates, basin
moisture content, and the rainfall excess, this
switch should be up.
# 4 If on, this switch allows the latest inlet hydro-
graph for each major watershed to be printed.
These represent the flow arriving at the structure
before diversion.
#5 If on, the diverted hydrographs for each major
watershed and the total lumped flow entering the
interceptor sewers will be printed.
# 6 If on, a tabulation of the volumes of water
before diversion, after diversion, and diverted to
the river will be shown.
#7 If on, the hydrograph at any point in the
interceptor system where the predicted flow is
greater than the allowable flow will be printed.
# 8 If on, all hydrographs will be printed as the flow
is routed through the interceptor system.
# 9 If on, only the inlet hydrographs before diversion
will be computed. Model will cycle through
chains 1, 2, 3, and 6 only (no diversion or
routing).
#10 If on, only the inlet hydrographs and diverted
hydrographs will be computed. Model will cycle
through chains 1, 2, 3, 4, and 6 (no routing).
#11 Not presently used.
#12 If on, the program will scan real time data only
(RTIME).
#13 If on, the program will not scan any real time
data.
#14 If on, the program will interrupt its normal
processing and return to program control for
operator intervention. The computations inter-
rupted will be lost.
#15 If on, the program will process a historical
rainfall file specified by the operator. The data
will be processed in one-hour blocks, starting at
the time given by the operator.
#16 If on, the instructions normally printed out on
the teletype will not appear.
#17 If on, the program will return to program control
for operator intervention after completing the
chain of the model that it is currently working
on. No computations will be lost.
The sense switches may be reset at any time.
Depending on the specific function of the switch it
may have an immediate effect on the program or a
delayed effect. When controlling or interrupting the
157
-------
sequence of processing by switches #9, #10, #14, or
#17, the operator must be careful to re-enter the
processing sequence at the proper point.
3. Program Control
The model is so constructed that an operator may
intervene in the normal processing sequence. This is
accomplished by the routine PROGRAM CONTROL.
At any time during processing the operator may
turn on sense switch (SS) #14 or #17 to return to
PROGRAM CONTROL.
PROGRAM CONTROL is also entered at other
times when information is needed by the program.
The options within PROGRAM CONTROL avail-
able to the operator are given below.
#1. The model is reentered at its origin. This gives the
operator a chance to process a new historical
rainfall file or to process real rainfall in file
RAINRP DLY. Control is transfered to Chain 1.
#2. This allows the operator to restart computing inlet
hydrographs. When processing real data the pro-
gram always computes starting 2 hours prior to the
present real time or at the beginning of the
RAINRP DLY file, whichever is later. For a
historical file the model starts at the starting time
given by the operator. This time is entered by the
operator at the console. If rain has occurred prior
to the starting time given by the operator, the first
two hours of prediction will not reflect this rain.
Control is transfered to Chain 3.
#3. This causes the model to continue the processing
by adding on the latest rainfall readings for real
time data or readings for one hour for historical
data. Control is transfered to Chain 3.
#4. When reentering at this point the program takes
the last determined inlet hydrographs and diverts
them. Control is transfered to Chain 4.
#5. When reentering at this point the program takes
the last determined diverted hydrographs and
routes them. Control is transfered to Chain 5.
#6. This allows the program to scan if it is the proper
time. PROGRAM CONTROL will always be re-
entered following this selection unless sense switch
#12 is kept on. Control is transfered to Chain 9.
#7. This calls the gate editor program. The gate editor
allows the operator to enter gate settings to the
model.
#8. This causes the program to stop and return control
to the monitor.
4. Gate Editor
The GATE EDITOR is an option called from
PROGRAM CONTROL that allows the operator to
delete, insert or replace gate and Fabridam settings that
are used by the model. The file that contains the model
gate settings is GATESETTG. If desired this file may
be changed with the systems EDITOR program, but
this requires that the model be stopped.
The GATE EDITOR has 5 operations, explained
below.
1. The first operation (L) provides a listing of all the
gate and Fabridam settings. In addition the maxi-
mum and minimum settings for each structure are
shown.
2. This operation (D) will delete a setting from the file.
The settings for day zero and day 365 may not be
deleted.
3. This operation (I) will insert a new setting.
4. This operation (R) will replace a given setting with a
new value.
5. This operation (E) causes the routine to return to
PROGRAM CONTROL.
Format:
NGTHRMNDAY X.XX X.XX
where
N is operation to be completed (L, D, I, R, E)
GT is gate number
HR is hour of day (0-23)
MN is minute of hour (0-59)
DAY is day of year (1-364)
X.XX is gate opening in feet
X.XX is fabridam height in feet
Exhibit 4 shows the teletype message when GATE
EDITOR is used.
Exhibit 4
Teletype message when Gate Editor is used.
BATE EDITOR
TYPE L FOR L1ST1N6 OF PRESENT SATESETTINGS
TYPI D KltH 6ATC NUMBER AND TIME- TO DELETE A SETTING
TYPE I KITH GATE NUMBER. TIME AND SETTING- TO INSERT SETTING
TYP.C R KITH DATE NUMBER. TIME AND SETING- TO REPLACE SETTING
TYPE C TO CXIT TO PROGRAM CONTROL
N OT HR MN DAY X.XX X.XX
L
N GT KR MN DAY X.XX X.XX
158
-------
Exhibit 5 shows the list of gate and Fabridam
settings.
Exhibit 5.
List of gate and Fabridam settings.
DIVERSION
NUKfltB
CAIE «ND FASflDAn SEIIINGS
line
HUN
CAIE FABRI£AH
SEIIIHC SETTING
nix
GATE
0
1
1
I
I
)
1
4
I
S
s
6
6
7
7
9
9
10
10
11
11
12
\2
11
11
It
14
1%
IS
It
It
17
17
IB
18
19
19
PHAL
P HAL
IROUt
IRQUI
KELOG
KELOG
RICE
II ICE
HARSH
HARSH
EUS1E
EUSIE
OAK
OAK
J8IH
jaTH
H INNH
n INNH
26SEA
26SEA
PUASH
PUASH
42CAM
42CA1
2 HA IN
2NA1N
N UN
Nun
Sun
sun
E AS1N
EASin
EUSTU
EUSTU
42-8P
42-BP
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
1
0
0
0
1
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
165
0
165
0
165
D
165
0
165
0
165
0
165
0
105
0
165
0
165
0
165
0
165
0
165
0
J65
0
165
0
165
0
165
0
165
0
2
2
1
1
6
6
1
1
}
1
1
3
1
1
1
1
1
1
2
I
I
2
4
4
2
2
0
0
0
0
0
0
1
1
00 0.00
50 1.00
50
75
75
00
00
00
00
75
75
00
00
25
25
25
2S
25
25
.25
.67
.00
.25
.67
.00
.50
.67
.00
.50
.67
.00
. 1)
.50
.67
.50
.82
10 2.00
10 5. 11
10 1.12
10 5.67
00 0.00
00 0.00
10 2.16
10 4.67
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 0.00
00 1.50
00 1.50
0.00 0.00
0.00 0.00
0
2
2
1
1
6
6
1
)
1
1
2
2
2
2
4
4
2
2
0
0
0
0
0
0
1
1
0
0
00
50
50
15
f5
00
00
00
00
75
75
00
00
25
25
25
2 5
25
25
10
10
10
10
00
00
10
10
00
00
00
00
00
00
00
00
00
00
0.00
6.25
6.25
7.00
7.00
4 .67
4.67
5.50
5.50
6.00
6.00
5.67
5.67
5 .11
5.31
6.67
6.67
6.e;
6.82
5. 13
•>. 13
5.67
5 .67
0.00
0.00
4 .67
4 .67
0 . 00
0.00
0.00
0.00
0 . 00
0.00
1 .50
1.50
o.co
0.00
c. Gather real time data and process historical data
d. Process historical data only
A description of each of these modes follows.
a. To use the model to acquisition data only, SS
#12 should be turned on. The operator will be
requested to ENTER REAL TIME. Then control
will pass to Chain 9 and will remain there as long
asSS#12 is on.
b. To both gather and process real time data SS #12
and #13 should be off. The operator will be
requested to ENTER REAL TIME. Then control
will pass to Chain 9 to allow data acquisition and
on to Chain 3 to allow data processing.
c. To gather real time data and process historical
data SS #12 and #13 should be off and SS#15
should be on. The operator will be requested to
enter the name of the historical rainfall file to be
processed.
ENTER NAME OF HISTORICAL RAINFALL FILE
(FILNAMEXT)
At this point the operator should type the 9
character file name containing the rain data to be
processed followed by a carriage return. If the file
is not present on the input device (DAT. 10) an
error will appear:
5. Execution of Model (UROM-9)
MODEL is an execute file that normally resides on
the disk. It is run from the system monitor by the
command:
SE MODEL )
K
The processing that follows depends on the selec-
tion of the various sense switches. As noted above
all sense switches (accumulator switches are used as
sense switches and are located on the lower right side
of the central processor) should be down (off) at the
beginning of execution unless the operator is familiar
with the program. When the sense switches are down
the program will print out important instructions at all
points where the operator may interact with the
program.
The following message will appear to inform the
operator of the possible choices he has available for
output and program branching.
The MODEL may be used in any of the following
modes:
a. Gather real time data only
b. Gather real time data and process real time data
FILE XXXXXXXXX NOT PRESENT ON DEVICE 8
and the operator will have the opportunity of
entering another file name.
After allowing a scan to occur in Chain 9 control
will return to PROGRAM CONTROL. At this
time the operator should exercise option 2. This
will allow him to then ENTER MODEL START-
ING TIME. Processing will then be initiated at
the specified time in the historical rainfall file.
d. To process historical data only, the procedure
outlined in 3 should be followed except that SS
#13 should be on.
When acquisitioning data the scanning reports will
appear on the line printer. Output from the model will
appear depending on the settings of the sense switches.
Exhibit 6 shows the teletype output for a run made
using an historical rainfall file (RASR70219). After
processing two hours of data the operator interrupted
the program, made a listing of the gate and Fabridam
settings and returned to the system monitor. Exhibit 7
shows part of the rainfall data used.
159
-------
Exhibit 6
Teletype output for sample model run.
MONITOR \
SE MODEL
Exhibit 7
Partial listing of input rainfall data
(RASR70219)
Shown is the time of reading (day, hr, min.), raingage
number.and mnemonics and accumulated rainfall (inches)
SENSE SWITCHESfACCUM. SHS> »l -MM 7 ARE USED IN THIS PROGRAM
IF ON(UP) THEY HAVE THE FOLLOWING FUNCTIONS
•I SURPRESS PRINTOUT OF EXCESSIVE FLOW LOCATIONS
It PAUSE IF EXCESSIVE FLOW
• 3 PRINTOUT RAINFALL LOSS-RATE DATA
*« PRINTOUT UNDIVERTED HYDROGRAPHS
•5 PRINTOUT DIVERTED HYDROGRAPHS
16 PRINTOUT VOLUMES OF HYOROGRAPHS
n PRINTOUT HYDROORAPHS AT POINTS OF EXCESSIVE FLOW
IS PRINTOUT HYOROGRAPHS AT ALL POINTS
•9 COMPUTE UNDIVERTED HYOROGRAPHS ONLY
lie COMPUTE UNDIVERTED AND DIVERTED HYDROGRAPHS ONLY
•12 SCAN REAL TIME DATA ONLY
•I] 00 NOT SCAN REAL TINE DATA
• M RETURN TO PROGRAM CONTROL
•15 PROCESS HISTORICAL RAINFALL FILE
•16 SUPPRESS PRINTOUT OF PROGRAM INSTRUCTIONS
•17 PAUSE AT END OF LATEST COMPUTATIONS
SET SWITCHES TO DESIRED POSITION AND THEN TYPE
CONTROL P AFTER THE PAUSE
THE SWITCHES MAY SE RESET AT ANY TIME
PAUSE
tPENTER NAME OF HISTORICAL RAINFALL FILE (FILMANEXT)
RASR7B 219
FILE RASR7B 81 NOT PRESENT ON DEVICE 8
ENTCfl NAME OF HISTORICAL RAINFALL FILE (FILNANCXT)
RASR702I9
ENTER REAL TIME
HR MN DAY
12 12 281
PROGRAM CONTROL PREVIOUS CHAIN I MODEL TIME £81 IE IB
TYPE 1 TO RESTART WITH ALL INITIAL CONDITIONS
TYPE 2 TO RESTART COMPUTING UNDIVERTED HYDROORAPHS
TYPE 3 TO CONTINUE COMPUTING UNDIVERTED HYDR06RAPHS
TYPE 4 TO CONTINUE COMPUTING DIVERTED HYOROGRAPHS
TYPE J TO CONTINUE ROUTING
TYPE « TO CONTINUE SCANNING REAL DATA
TYPE 7 TO CALL GATE SETTINGS
TYPE 8 TO STOP
ENTER MODEL STARTING TIME
HR MM DAY
19 86 219
IOP34I
50
0
11
2J
11
44
56
6
16
27
50
2
13
21
11
641 UH-RC
641 Un-RG
641 un-RG
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
64
l>4
64
64
64
64
64
64
UF1-RG
UM-RG
un-HG
un-RG
UM-RC
Utl-RC
UM-RC
UN-DC
un-RC
UN-RC
UM-RC
UM-RC
un-RC
un-RC
un-RC
UM-RC
un-RC
un-RC
UM-RC
UM-RC
un-RG
un-RU
UM-RC
UM-RC
un-RC
un-RC
un-RC
un-RC
un-RC
un-Rc
un-RC
UM-RC
UM-RC
UM-RC
un-RC
UM-RC
UM-RC
UM-RC
un-RC
un-RC
UM-RC
un-RC
UM-RC
UM-RC
UM-RC
UM-RC
UK-SG
UM-RC
un-RC
un-RC
641 un-RG
641 UM-RC
641 UM-RC
641 UM-RC
-0.
-0.
064
008
-0.010
-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
021
004
Oil
008
002
006
021
058
065
099
128
265
291
124
111
1BI
413
418
449
457
476
484
494
492
500
518
520
515
519
521
545
512
544
S49
SIB
540
540
517
542
SIS
546
512
492
0. 529
o
0
0
0
0
0
0
0
0
0
541
S42
5!»
540
511
519
546
S44
541
541
TYPE NUMBED AGAIN
9
TTPI NUMBER A8AIN
PR06RAM CONTROL PREVIOUS CHAIN
B
STOP 777777
2 MODEL TIMC 219 17 «
MONITOR V49
160
-------
Exhibits 8-12 show part of the line printer output for the same run shown in Exhibit 6.
Exhibit 8
Partial output of rainfall loss analysis.
The rainfall file name is shown at the upper right. The nine raingages and their corresponding 5 minutes intensities
for the period ending at 16:11 day 219 are given next. The remaining table shows the rainfall loss and excess on
each of the 15 basins. Unless specified all values are given in inches for the 5 minute period ending at the time
shown. The definition of the column mnemonics are given below.
I
RAIN
ISTLOS
PSTLOS
IINLOS
PINLOS
ILOSCOE
PPOTLOS
I LOSS
PLOSS
IEXS
PEXS
BASLOS
BASEXS
BMC
ft A I NCACE
GAGE
READINGS
1
1
I
>
4
5
6
7
8
9
10
1 1
1 !
1 1
11
15
READ INGS
1
1
I
)
4
5
6
7
8
9
10
1 1
1 t
1 1
It
15
«EAO INGS
1
I
2
1
*
5
b
7
a
9
10
— basin index
— Average rainfall over basin
— Accumulated value of initial loss on impervious surfaces.
- Accumulated value of initial loss on pervious surfaces.
- Contribution to initial loss on impervious surfaces.
— Contribution to initial loss on pervious surfaces
— Impervious loss coefficient-dimensionless.
— Potential loss on pervious surfaces.
— Loss on impervious surfaces.
— Loss on pervious surfaces.
- Excess on impervious surfaces.
— Excess on pervious surfaces.
- Average basin loss.
— Average basin excess.
— Accumulated losses on pervious surfaces.
ACCLOSS — Accumulated losses on impervious surfaces.
no.
AND HUE
HA IN
0.0092
0.0196
0.0090
0.0195
0.0217
0.02)5
0.0191
0.0278
0.0299
0.0092
0.01*9
0.0052
0.0222
0.0186
0.0000
AND IIME
RAIN
0.0092
0.0198
0.0090
0.0195
0.0217
0.02)5
0.019B
0.0278
0.0299
0.0092
0.01*9
0.0052
0.0222
0.0156
0.0000
AND llnE
RAIN
0.0087
0.0176
0.0075
O.Olbl
0.0192
0.0235
0.0198
0.0275
0.0294
0.0090
12 1 4 5 b 7
HAC-
0.0628b 0.
ISTLOS PSTLOS
0.0000 0.0792
0.0000 0.1198
0.0000 0.0690
0.0000 0.1495
0.0000 0.1817
0.0000 0.2500
0.0000 0.2498
0.0000 0.2500
0.0000 0.2)99
0.0000 0.1192
0.0000 0.1749
0.0000 0.0552
0.0000 0.2500
0.0000 0.2186
0.0000 0.0100
MAC-
0.06286 0.
ISILOS PSILOS
0.0000 0.0884
0.0000 0.1)91
0.0000 0.0180
0.0000 0.1689
0.0000 0.20)4
0.0000 0.2500
0.0000 0.2500
0.0000 0.2500
0.0000 0.2500
0.0000 0.1284
0.0000 0.1899
0.0000 0.0605
0.0000 0.2500
0.0000 0.2)72
0.0000 0.0700
hAC-
0.06285 0.
ISTLOS PSILOS
0.0000 0.0971
0.0000 0.157)
0.0000 0.0855
0.0000 0.1850
0.0000 0.222b
0.0000 0.2500
0.0000 0.2500
0.0000 0.2500
0.0000 0.2500
0.0000 0.1374
EON-
08512 0
1 1 «LOS
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
o.oouo
0.0000
0.0000
EON-
»IL-
. 16500
PINLOS
0.0092
0.0198
0.0090
0.0195
0.0217
0.0000
0.0198
0.0200
0 .0299
0 . 0 09 2
0.0149
0.0052
0.0000
0.0186
0.0000
V IL-
08572 0.16500
1 1 NLOS
0.0000
0.0000
0.0000
0 . 0 00 0
0.0000
0.0000
o.oouo
0.0000
0 . 0 00 0
0.0000
0.0000
0 .0 000
0.0000
0.0000
0.0000
EON-
PINLOS
0.0092
0.0198
0.0090
0.0195
0.021 '
0.0000
0.0002
0.0000
0.0101
0.0092
O.OH9
0 .005 2
o.oooo
0.0186
0.0000
VII-
0851 1 0. 14000
1 1 NLOS
0 .0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
PINLOS
0.0087
0.0176
0.0075
0.0 161
0.0192
0.0000
0.0000
0.0000
0.0000
0.0090
BTP-
0.00000
-------
Exhibit 9
Partial listing of calculated inlet hydro-graphs.
Points on hydrograph are given at 5 minute intervals from -34 to +50 periods relative to
point zero which corresponds to 16:06day 219. Read horizontally left to right. Discharge
in cfs.
UNDIVERTED HVDROCftAPHS
Y U IN 1
INTPH
-34
-22
-10
2
1 4
26
) 8
so
INIIR
-34
-22
-10
2
14
26
3 8
SO
DAVN 1
-14
-22
-10
2
14
26
3 8
SO
KELOI
-22
-10
2
14
26
38
50
HYODOCBAPH NO. 1
8.0-11
a.
8.2-21 8.
(
1
.4 -9
.3 3
9.6 IS
8.0 27
7 7 19
8.
11 .
9.
a.
j
7.4
HTDROCBAPH
19.2-31
1
2(
1.7-21
.0 -9
22.2 )
22.9 IS
21.3 27
18.0
H
H
OROGRAPH
.6-3)
.7-21
.7 -9
.2 3
.7 IS
.6 27
.4 39
.2
DROCUPH
.)-))
.5-21
.6 -9
62.9 3
7.6 IS
7.3 27
.0 19
NO
19.
19.
20.
22 .
22.
21.
NO
4.
4 .
4.
7 .
4.
4 .
NO
7.
7.
7 .
62.
7.
7.
7 .
0-)2
2-20
4 -8
5 4
S 16
0 2B
7 40
2
2-32
7-20
o -a
a 4
8 16
2 28
)
6-12
7-20
8 -8
9 4
7 16
6 28
4 40
4
3-12
5-20
6 -8
9 4
6 16
3 28
0 40
.0-1 1
.2-19
.4-7
1 .4 S
.117
.0 29
T 741
19.2-31
19.7-19
20.0 -7
23.) 5
22. 7 17
21.1 29
-) 1
. -19
- /
5
. 1 7
. 29
7. 3-3 1
7.5-19
7.7-7
54.4 5
7.617
7.) 29
7.041
a
8
a
II
9
6
19
19
20
21
22
21
7
7
7
45
7
7
I
0-30
2-18
4 -6
2 6
2 18
0 30
742
2-10
7-18
0 -6
7 6
6 18
1 10
6-10
7-18
a -6
a 6
7 18
6 10
3-10
5-18
8 -6
8 6
6 la
3 30
042
a
a
a
11
9
a
19
19
20
21
22
21
7
7
a
)7
7
/
7
0-29
2-17
4 -S
1 7
0 19
0 11
2-29
7-17
0 -5
8 7
6 19
0 31
6-29
7-1
8 -
2
7 1
: -
3-29
5-17
0 -S
4 7
6 19
1 31
0 41
PUINTS ARb AT
8.0
8.2
8.5
10.9
8.9
a.o
19.2
19.7
20.0
21.7
22.5
20.9
4.6
4. 7
4.8
4.8
4.7
4 .6
7.1
7.5
a. )
29.1
7.6
7.3
7.0
-2 8
-16
- 4
8
20
12
-28
-16
- 4
a
20
)2
-28
- lo
- 4
B
20
32
-28
-16
- 4
8
20
32
44
8
8
8
10
a
8
19
19
10
23
22
20
7
7
8
21
7
7
7
S MlNUTt
0-27
2-15
5 - 1
7 9
7 21
0 ) 3
2-27
7-15
1 -3
7 9
4 2 1
8 11
6-27
7-15
9 - 1
8 1
7 2 1
6 11
3-27
5-15
9 -1
1 9
6 2 1
1 11
0 45
8
a
8
10
8
8
19
19
20
23
22
20
7
7
INTERVALS
.0-26
.2-14
.6 -2
.6 10
.6 22
. 0 34
.2-26
.7-14
.2 -2
.6 10
.) 22
.7 )4
.6-26
.7-14
.2 -2
.8 10
.7 22
.6 14
.1-26
.5-14
10 .9 -2
14
7
7
7
.6 10
.6 22
.3 )4
.0 46
.0-25
.2
.8
1 .4
. 4
.0
19.2
19.7
20.)
23.5
22.2
20.7
.6
.7
.7
.8
.7
.6
7. 3
7.5
15.)
10.0
7.6
7.)
7.0
-1 1
- 1
I 1
2)
35
-25
-1 3
-1
1 1
2)
)5
-25
-1 )
- 1
11
23
35
-25
-1 )
-1
11
23
15
47
1
19
20
20
21
21
19
7
7
25
7
7
7
6
.2-24
.4-12
.2 0
.112
. 1 24
.7 36
.7-24
.0-1 2
.7 0
.1 12
.5 24
.8 16
.7-24
.7-1 2
.4 0
.712
.6 24
.436
.5-24
.6-12
.1 0
.6 12
.3 24
.0 36
.848
a
8
.2-23
.4-1 I
9.8 1
10
a
7
19
20
21
2)
21
19
.0 1 3
.0 25
.7 37
.7-23
.0-1 1
.1 1
.0 1 3
.4 25
.7 37
4. 7-2 1
7
7
38
J
7
7
6
.7-1 1
.1 1
.7 1 J
.6 25
.437
2 49
.5-23
.6-1 1
.7 1
.6 1 1
.3 25
.037
.8 49
.2
.4
1 .7
.1
.0
.7
19.7
20.0
21 .7
21.0
21.4
H.6
4.7
4.7
t .0
4.7
4.6
4.4
I.S
7.6
5*. 8
r.6
7.3
1.0
4.1
6.8
Exhibit 10
Partial listing of hydrographs diverted into the interceptor at
major inlets before lumping parameters are applied.
DIVERTED HTDHOCRAPHS
POINT 0 REPRESENTS PRESENT MODEL lint
INTPH HrOROCRAPH NO. 1
-14
-22
-10
2
1 4
26
3 8
50
INIIR
-34
-22
-10
2
1 4
2 6
38
50
DAVN 1
-22
-1 0
2
1 4
26
1 8
50
KELOI
-14
-22
-10
2
1 4
26
18
50
8.0-11
8.2-21
8.4 -9
11.) 3
9.6 IS
8.0 27
7.7 19
7.4
8.0-)!
8.2-20
8.4 -B
11 .5 4
9.5 16
8.0 28
7.7 40
HTDRDGRAPH NO.
19.2-13
19. 7-21
20.0 -9
22.2 J
22.9 IS
21.3 27
19.5 19
18.0
19.2-12
19 .7-20
20.0 -8
22 .» 4
22.8 16
21.2 28
19.5 40
xrDROCRAPH NO.
4.7-21
4.7 -9
B.2 )
4.7 IS
4.6 27
4.4 19
4.2
4 .7-20
4.8 -S
7.9 4
4.7 16
4 .6 28
HIDROCRAPH NO.
7.1-3)
7.5-21
7.6 -9
41.2 1
7.6 15
7.) 27
7.0 )9
6.8
7.3-1;
7 .5-20
7.6 -8
43.3 4
7.6 16
7.) 2»
7.0 4D
ft
e
8
11
9
8
2
19
19
20
23
22
2 I
19
3
4
4
6
4
4
4
7
7
7
37
7
7
7
0-> 1
2-19
4 - 7
4 5
1 1 7
0 29
2-1 1
7-19
0 - 7
1 5
7 1 7
1 29
4 4 1
7-1
B -
7
7 1
6 29
1-1 I
5-19
7 -7
4 5
6 1 7
1 21
0 4 1
16
6.0-10
U
a
11
9
d
19
19
20
21
22
2 1
19
7
7
7
17
7
7
t
2-ia
4 -6
2 6
2 ia
0 10
2-JO
1-18
0 -6
7 6
6 18
1 10
1 42
7-18
6 -6
B 6
' IB
6 30
)- 10
5-H
8 -6
2 6
6 1 8
. 1 10
.0 42
6 219
8.0-
8.2-
8.4
II. 1
9.0
a.o
19.2-
19. 7-
20.0
Jl.t
12. t,
21.0
19.2
4.7-
4.8
5.2
4. 7
4.6
7. )-
DISCHARGE
29
1 7
-5
7
19
11
29
1 7
-5
7
19
3 1
4 1
1 7
-S
7
19
11
29
.5-1 7
.0
I .0
. 6
. 3
.0
-5
7
19
11
4 1
tt . 0
8.2
8.5
10.9
8.9
8.0
19.2
19.7
20.0
21.7
22.5
20.9
19.2
4 . 7
4.8
4. B
4 . 7
4.6
7. 1
7.5
8.3
29. 1
7.6
7. 1
7.0
-28
-1 6
- 4
8
20
12
-28
-1 6
- 4
a
20
1 2
4 4
-16
- 4
8
20
32
-28
-16
- 4
a
20
12
4 4
AT 5 nlNUIE NTEHVALS
1!
8
8
10
8
8
11
19
20
21
It
2 0
11
7
7
B
21
7
7
7
0-! 7
2-15
5 -1
7 9
7 2 1
0 1 1
2-27
7-1 5
1 - I
7 9
4 I I
8 31
1 45
7-15
9 - 1
a 9
7 2 1
6 13
1-27
5-15
9 - J
1 9
6 2 1
1 13
.0 45
8
8
B
10
8
a
19
19
20
2 1
22
19
7
7
10
14
7
7
7
0-26
2-14
6 -2
6 10
6 22
0 14
2-26
7-14
2 -2
b 10
1 22
0 46
7-14
2 -2
a 10
7 22
6 14
1-26
5-14
9 -2
6 10
6 22
1 )4
0 46
8.0-25 8
8
B
10
B
8
19
19
20
23
22
2 0
IB
t
5
4
4
4
7
7
15
10
7
7
7
2-11
a - 1
4 1 1
4 23
0 35
2-2S
7-1)
1 - 1
S 1 1
2 21
947
7-11
7 -1
B 1 1
7 23
6 35
3-25
5-11
) - 1
0 11
6 21
3 35
0 47
8
9
10
8
7
19
20
20
2)
21
18
7
7
25
7
7
7
6
2-24
4-1 2
2 0
1 1 2
1 24
7 16
7-24
0-1 2
7 0
1 1 !
S 24
I 4B
7-24
7-1 1
4 0
7 1 2
6 24
4 36
2 48
5-24
t-12
1 0
6 1 2
1 24
0 16
8 48
1
19
2-2 3
4-1
a
0 1
0 2
7 3
7-2 3
20.0-1 1
21
21
21
18
4
7
7
36
7
7
7
6
1 1
0 1 1
4 25
1 49
7-21
-1 1
1
1 1
25
)7
2 49
5-2 3
6-1 1
1 1
6 1 1
3 25
0 3 7
849
8.2
a. 4
10.7
9.8
a.o
7.7
19.7
20.0
21 .7
21 .0
21 .4
18.0
4.7
8 .0
4 .7
4.6
4 . 4
4 . 2
.5
.6
1 .7
.6
. 3
.0
.8
162
-------
Exhibit 11
Partial listing of flow diverted into interceptor
after lumping factor applied.
DIVERTED H1DROGRAPHS
POINT 0 REPRESENTS PRIISENT MODEL Tint
INTPH HyOROCRAPH NU.
DISCHARGE AI 5 M1NUIE INTERVALS
14
22
10
2
14
26
18
11.8-11
12.6-21
11.2 -9
44.9 1
38.2 15
11.8 27
10.5 19
12
1!
45
17
11
10
.8-12
.2 -8
.7 4
. b 16
.8 ii
.5 40
11 .8-1 1
12.6-19
11.2 - 7
45.0 5
36.9 17
31.8 2?
)0.5 41
1 1.8-10
12.6-18
11.1 -6
44 . 4 6
Ib. 1 13
11.! 30
30. 5 42
11.8-29
12.6-1 7
31. 1 -5
41.8 7
15.7 \^
11. t 11
30.5 41
11 .8-28
12.6-16
11.5 -4
41.1 B
15.1 20
11.8 12
10.5 44
31 .0-27
12.6-15
11.8 -3
14.5 21
11.8 13
10.5 45
11 .8-26
32 .6-14
14.1 -2
14 .0 22
11 .« )4
10 .5 46
11.8-25
32.6-1 1
34.9 -1
41.1 11
11 .8 15
10.5 47
126-24
11.2-1 2
16.4 0
40.1 12
10.5
29.5
36
43
32.6-
13.2-
19.0
19.5
11. B
30.5
29.5
•21
-1 1
1
1 1
25
1 7
49
13.2
W .5
18.8
11.8
10.5
29.5
lltTTR HtOROCRAPH NO.
-J4
-22
-10
£
1 4
26
18
50
0«VN 1
-14
-22
-10
2
14
26
1 8
50
1.110 1
-14
~2 £
-1 0
t
\ 4
26
!8
50
20
20
21
2)
24
22
20
19
.4-31
.9-2 1
.2 -9
.6 3
.2 15
.6 27
.7 19
.0
20
20
21
24
24
22
20
MYDROGRAP
7
7
7
12
7
7
b
.2-11
.4-21
.5 -9
.9 1
.4 15
.227
. 7
1
HTDROCRAPH
11
11
65
11
11
10
10
.1-31
.6 -9
.7 J
.5 15
.1 27
.7 19
.1
11
1 1
11
65
11
1 1
10
4-12
9-20
2 -8
2 4
2 16
5 28
6 40
NO.
2-1J
4-20
5 -8
4 4
4 16
2 28
NO.
1- 32
6 -R
7 4
5 16
1 28
7 40
20
20
21
24
24
22
20
1
;
7
7
10
7
7
4
II
1 1
11
56
II
11
10
4-31
9-1
2 -
7
1 1
4 2
6 4
2-3 1
4-1 '
5 - 1
7 5
4 1 7
2 29
1-3 1
7 - 7
8 5
5 1 7
1 2^
7 4 I
20
20
21
25
24
22
20
Y
7
7
9
7
•
11
1 1
1 1
56
II
11
10
4- 10
fl-ia
I -6
1 6
a is
1 10
5 42
2-10
4-18
5 -b
2 b
> 18
2 10
1- 30
4-18
9 -b
b b
5 18
1 10
7 42
20
20
21
25
2 1
22
20
7
j
j
s
/
/
u
1 1
12
11
II
| ]
10
4-2*
9-17
2 -•>
1 7
1 11
2 31
4 4 1
2-29
4-1/
5 -•>
2 1
4 11
? 11
1-29
1 -5
1 7
5 19
1 11
7 41
20
20
21
25
21
22
20
7
7
7
7
7
7
1 I
12
&4
11
II
10
-28
-1 b
- 4
8
20
12
1 44
2-28
4-16
6-4
6 8
4 20
2 12
I -28
b -4
2 B
5 20
I 12
7 44
20
20
21
25
21
22
20
7
11
11
12
11
11
10
. -27
. -15
. - 1
9
. 2 I
. 3 1
.2 45
.2-27
.4-15
.6 - 1
.b 9
.4 2 1
.2 33
.1-27
.6 - 1
.*. 9
.5 21
.1 11
.7 45
20
20
21
25
2 1
22
20
7
7
8
7
7
7
11
16
22
II
11
10
4-26
9-14
4 -2
0 10
' 22
0 14
1 46
.2-26
4-14
2 -2
6 10
4 22
2 14
. -26
-2
10
. 22
14
46
20
20
21
24
21
21
20
11
23
15
II
11
10
4-25
9-1 3
5 - 1
9 11
6 2 1
9 35
1 47
2-25
4-1 3
9 - 1
6 11
423
2 15
1-25
1 - 1
2 1 1
5 2 1
1 15
7 47
20
21
2i
24
22
21
19
7
7
10
;
7
b
II
1 1
18
11
11
10
10
9-24
2-1 2
9 0
5 12
1 24
0 16
1 48
4-24
5-12
1 0
4 1 2
2 24
9 16
4-2 I
6-12
1 0
5 1 2
1 24
7 16
1 48
20
21
22
24
22
20
19
7
7
11
7
7
6
11
1 1
55
11
11
10
10
9-2 ]
2-1 1
1 1
4 1 1
7 25
9 37
2 49
4-2 3
5-1 1
3 1
4 I 1
225
9 1 7
7 49
4-21
6-1 1
2 1
5 I 1
I 25
7 1 7
349
20.9
21 .2
21.0
24.1
22 .7
20.8
19.1
7.4
7.5
12.7
7.4
7.2
6.9
6 . 7
11 .4
11.6
57. 1
11.5
II .1
10. J
10.1
Exhibit 12
Summary of volumes before and after diversion.
INLET INOE1
1
2
1
4
5
b
7
8
9
10
11
12
1)
14
15
TOTAL VOLUME
TOTAL VOLUME
LUMPED INLET VOL
85991 1.
S5151 7.
192652.
456168.
1 5 78 6 9 .
171116.
49(705.
16811 6.
7 19502.
901750.
697681.
11 10914.
f. 1695 1.
1249695.
645588.
CF 10? 1661 4.
HC 76.42
LUMP ED 0 IV V GL
859911.
551517.
192652.
416791.
15(870.
164082.
354566.
720288.
719502.
901750.
695514.
1044109.
816-15 1.
1249695.
645588.
9912878.
74. 15
LUHPf D RIVER VOL
0.
0.
0.
19577.
-0.
7014.
14011 7.
4802 7.
-0.
0.
2167.
66809.
0.
0.
0.
101751.
2.27
INLET VOL
217151.
520126.
1 2191 1.
10024 4.
157868.
211950.
171582.
76811 4.
406322.
111045.
S8I407.
981127.
614256.
844199.
1801] 3.
6632271.
49.61
DIVERTED VOL
217151.
520126.
12193).
274206.
157B68.
227554.
12441 1.
720286.
406322.
111045.
571602.
924002.
614256.
844199.
18013 1.
6441706.
4 8.20
Rl VE ft VOL
0.
0.
0.
26018.
-0.
t 396.
49 172.
4B027.
-0.
0.
1806.
59124.
0.
0.
0.
188561.
1.41
163
-------
Exhibit13
6. Loader Memory maps of
CREATE and MODEL (UROM-9)
MONITOR V48
SA OKA -6.-4.-1.3.19
ICHAIN
CHAIN V2A
>8UILD CREATE
.-CHAIN 1
*CREAT2. 1NPUTD. INPUTH.ROUTJN
>END
CREAT2 55492
INPUTO 54051
INPUTH 52373
ROUTJN 51460
45462
45074
44541
44530
44461
41432
41 I 56
41076
41063
40770
40230
40134
OKA.
LPA.
FILE
FLOAT
• DA
8CDIO
BINIO
• SS
STOP
SPNSG
FIOPS
OTSCR
INTEGE 37661
REAL 36625
•CB 40114
CHAIN* |
LOWEST 31537
COMSZE 00000
BONI TOR V48
IB PR
MONITOR V4S
ItJOS 8U1LO MODEL.
6UILO MODEL
>CHAIN I
>CHAIN1.SENSE.SET.TMADD
>ENO
CHAIN! 54545
SENSE 54514
SET S4I67
TMADD 53723
OKA. 47725
47372
47323
FILE
.DA
SCDIO 44374
81NIO 44020
43740
43725
• S3
STOP
PAUSE 43711
SPHSO 43616
FIOPS 430S6
OTSER 4Z762
INTEGE 42643
REAL 41607
.CB 41567
CHAIN* I
LOWEST 41567
COMSZE 07385
>CHA[N 2
>CHAIN2.CONTL.G«TEOT.EWNTL.TMAOD,SENSE.RAINF
.END
CHAIN2 56410
CONTL 55421
BATEDT 5341 I
EVNTL 52246
TMADO 520Q2
SENSE SI 751
RAINF 471 I I
OKA. 43113
LPA. 42525
FILE 42172
IABS 42156
•DA 42107
BCD 10
BINIO
.SS
GOTO
STOP
SPMSG
FIOPS
OTSER
INTEGE
REAL
.CB
34751
41633
41553
41523
41512
41 41 7
40657
40563
40444
33715
40424
CHAIN* 2
LOHEST 33265
COMSZE (7305
»CHA1N 3
>CHAIN3.HrDR.SLOS.SENSE.THADD
>END
CHA1N3 55681
HYDR 53435
52152
52121
51655
45657
4S271
44736
44725
44672
44657
44566
44450
44404
44335
41306
41032
40752
40737
40644
40104
37704
SLOS
SENSE
TMAOD
OKA.
LPA.
FILE
FLOAT
• BE
EXP
• EE
.EF
• EC
• DA
BCDIO
BINIO
• SS
STOP
SPMSG
FIOPS
OTSER
INTEGE 37565
REAL 36531
.CB 40064
CHAIN* 3
LOWEST 34777
COMSZE 07305
>CKAIN 4
>CHAIN4.DlV,Tri,Tr2.TY4.TY5.ARE4.BICC.TOP.CUES,SENSE.TRAPOZ,
>THADD»FA88TS
>ENO
CHA1N4 54732
53232
52601
52202
51543
50436
50065
47703
47566
47464
47433
DIV
TYI
TY2
TY4
TY5
AREA
B1CC
TOP
CRES
SENSE
TRAPOZ 47286
TMADO 46762
FABGTS 46521
OKA. 42523
LPA • 42 I 35
FILE 41602
41564
41510
41455
41367
41354
41341
41237
41150
41057
40741
40675
40626
34751
40352
40272
40244
40231
40136
34211
ASS
• BC
• BE
SORT
SIN
ATAN
• EB
• ED
.EE
• EF
• EC
BCOIO
BINIO
.SS
GOTO
STOP
SPMSG
FIOPS
OTSER
INTEGE 34072
REAL 33036
•CB 40022
042
CHAIN« 4
LOWEST 32046
COMSZE 37305
164
-------
>CHA1N 5
>CHAIN5.ROUTB.COMB,DI WlN.WALY,AREA,SENSE.TYI.TY7.FABOTSrTMAOO.
>CRES.TOP,BICC.TRAPOZ
>ENO
CHAINS 55715
ROUTS 54576
COMB 34414
OIVIN 53422
VALY 53035
AREA 52464
SENSE 38433
TYI 52808
TY7 51636
FABGTS SI 315
TMADO SI 131
CRES 51027
TOP 56712
8ICC 59508
TRAPOZ 50333
OKA. 44325
LPA. 43737
FILE 434B4
ABS 43366
IABS 43353
HOC 43326
.BC 432S2
.BE 43217
SORT 43131
SIN 43116
ATAN 43183
.EB 43801
.ED 42712
.EE 42621
.EF 42503
.EC 42437
.DA 42370
BCDIO 347SI
BIN10 42114
.55 42034
GOTO 45006
STOP 41773
SPMSG 41708
FIOPS 41140
OTSER 41844
INTEGE 40725
REAL 33715
•CB 40705
CHAIN* S
LOWEST 32364
COHSZC 07305
•CHAIN 9
>SCKDL2«SCALER,RIVI10N,REGCTL.INCPTl)»RAlNRP.ALARHR.SETaAT.
MNTRPT.SCNFRO.TINR.POINT.CONSII.TINE
H) GATE/SENSE
• IMD
SCNOLP. 56432
SCALD) SSS7S
RIVMON S46B7
REGCTL 53384
INCPTH 52333
RAINRP 51225
ALARM) 50322
SETBAT 47594
INTRPT 47411
SCNFRB 47B63
TIM) 46727
POINT 46455
CONSW 46401
TIME 4631B
RGATE 46183
SENSE 46051
OKA. 42053
LPA. 41465
FILE 41132
IABS 41116
.BB 41040
.DA 40771
BCDIO 34751
.SS 40711
STOP 40676
SPMSG 40603
FIOPS 40043
OTSER 34655
INTECE 34536
REAL 33592
.CB 40023
CHAIN* II
LOWEST 33502
COHSZE 07303
165
-------
APPENDIX B
Description of Labels Used in CREATE and UROM-9
If the same label is used in different routines to
represent different variables, the name of each routine
is shown with the appropriate description. An asterisk
(*) indicates that the label may indicate a type other
than that implied by the label name.
(TYPE 7)
Label
A
ACCUM
AG
AJJ
AJL
ALOSC
ALPH
ALPHA
AN
ANS
AO
AR
AREA
AVG
AVG1
A5
A6
BA
BB
BC
BD
BFFAC
BMC
BMSTL
CHECK
COUNT
CZ
C1
Routine(s)
(TYPE 4)
(HYDR)
Description
Not used
See AO
See ARE A
Not used, accumulated time
since loss rate recovery para-
meter has been applied.
Temporary value
Index in PAL routing
Index in PAL routing
Loss rate constants
See NAMES
See NAMES
Manning's "n"
SeeNSTRL
Area of orifice opening (sq ft)
(CHAINS) See AREA
Combined sewer area actually
contributing to given inlet (sq
ft)
Average initial rainfall reading
Average rainfall reading
Area of orifice opening (sq ft)
Area of orifice opening (sq ft)
Temporary value
Temporary value
Temporary value
Temporary value
Base flow factors—hourly fac-
tors to determine hourly base
flow from mean daily
Basin moisture content (total
loss on pervious surfaces)
File name of antecedent con-
dition parameters
(BMCXSTLOS)
'Flag
Number of valid rainfall
readings
Thiessen polygon factors
(RAINF)
C10
C11
C12
C13
C14
C15
C2
C3
C4
(TYPE 1)
(TYPE 2)
(TYPE 4)
(TYPE 5)
(TYPE 7)
(TYPE 1)
(TYPE 2)
(TYPE 4)
(TYPE 5)
(TYPE 1)
(TYPE 2)
(TYPE 5)
(TYPE 1,
TYPE 4)
(TYPE 2)
(TYPE 5)
CQ of Fabridams
CQ side channel weir
CQ weir A
C5
C6
C7
C8
C9
D
DDT
DELT
DF
DFA
OFF
DIFF
DIVC
DIVCC
DIVE
DR
DT
DTT
E
CQ orifice
Length of lower side weir
(Type 5)
CQ orifice (Type 5)
Elev. lower weir (Type 5)
Drop from orifice to intercep-
tor (Type 5)
Length of upper side weir
(Type 5)
Width of orifice (Type 5)
1/2(32.2) (CD
Length side channel weir
Fab. expon.
Length of Weir A
Width orif.
Fab. expon.
CQ gates
Discharge ratio
CQ orifice 0
Width orif.
Elev. weir
CQ Fab.
Elev. orifice invert (Type 1)
Expon. Fab (Type 5)
Area orifice D (Type 5)
Elev. upper weir (Type 5)
CQ side weir (Type 5)
(AREA,
CRES)
(GATEDT)
(TOP)
(CONTL)
Nominal opening of orifice
Hollerith "D" Delete
Nominal fabridam diameter
See DTT
Difference in time (hrs)
Fabridam diameters (ft)
Fabridam diameter (ft)
Temporary value fabridam di-
ameter (ft)
Difference between current and
last reading; average rainfall in-
crease in time period
Diversion constants (C1 through
C5)
Diversion constants (C6
through C15)
Common block name
(EVNTL) Difference of values
Time interval of points on
hydrograph (sec)
Time interval of equally spaced
rain data (sec)
(GATEDT) Hollerith "E" Exit
167
-------
F
FACT
GAGE (RAINF)
GATES
(TYPE 1)
H
HA
HF
HFA
HFF
HFFF
HFL
HFMAX
HFMAXL
HG
HGG
HGGG
HGL
HGMAX
HGMAXL
HS
H1
H5
H6
I
IA
IADDA
IADDR
IB
1C
ICN
ICONTA
ICONTB
ICONTC
ICONTD
ID
IDA
IDAY
IDAYH
IDD
(DIVIN)
(EVNTL)
Residual in bisection method
Lumping factors to account for
areas that are assumed to enter
interceptor at one of 15 major
inlet points
Gage name
File name for gate and Fabri-
dam setting file (GATESETTG)
Fabridam height (ft)
Gate opening (ft)
Gate opening (ft)
Fabridam height (ft)
Fabridam height (ft)
Fabridam height (ft)
Fabridam height (ft)
Fabridam height (ft)
Max Fabridam height (ft)
Max Fabridam height (ft)
Gate opening (ft)
Gate opening (ft)
Gate opening (ft)
Gate opening (ft)
Max gate opening (ft)
Max gate opening (ft)
Temporary value
Temporary value
Gate opening (ft)
Gate opening (ft)
Index
Index of diversion type
Temp index
Rain gage address
Valid rain gage addresses in
order used by model
Index of meter
No longer used
Number of additional lag peri-
ods if NSTRL is even
Flag indicating (0—don't print
instructions to operator, 1 —
print)
Index of current operation in
operation table (JR)
Total number of steps in opera-
tion table (JR)
Index giving last chain executed
Time-days
Time-days
Real time day
Diversion model time (days)
Time-days to begin forming
even intervals
IDT Time interval (sec)
IDTM Time interval minutes
IDTT (RAINF) See IDT
IE Index of operation to be done
(1—routing, 2—combining, 3—
diversion, 4—begin on new line)
IGT Gate number
IH Time-hours
(CHAINS) Index of hydrograph to be
operated on
IHA Time-hours
(CHAIN 5) Index of hydrograph to be
combined
IHH Index of where to store results
of operation
(ENVTL) Time-hours to begin forming
even intervals
IHR Realtime-hour
IHRH Diversion model time (hrs)
MB Sense switch flags (0 through
17)
IJ Flag for excessive flow
IK (ROUTB) Index
IM Time minutes
IMA Time minutes
IMIN Real time minute
IMINH Diversion model time (minutes)
IMM Time-minutes to begin forming
even intervals
IN Index
IN LAG Index of hydrograph points to
be logged
IO INDEX
IR (HYDR) See NNN (RAINF)
IRI Index
IS Time seconds
IS (HYDR) Index
ISEC Real time seconds
ISS Time-sec to begin forming even
intervals
1ST AT File status flag
IT Index
ITB Number of time intervals for
base length of unit hydrographs
ITEST (CHAIN 1) File status flag
ITIME (RAINF) Temp, storage oftimes of rain-
fall readings
(TIMES Used by RTIME (to keep com-
mon size equal)
ITM (GREAT 2) Largest unit hydrograph base
time (DT time intervals), no
longer used
168
-------
ITP Number of time intervals for
time to peak of unit hydro-
graphs
ITQI (GREAT 2) Not used
ITT Index
IX Index
IXD Time days
IXH Time hours
IXM Time minutes
IXS Time seconds
IXXX (RAINF) Dummy label
IY Flag (=), no excessive flow;
@ 1, return to program control
at completion of routing)
IZ (CONTL) Parameter indicating where
control should be transfered in
Program Control
Index
J Index
(GATEDT) Flag indicating (0 file closed, 1
file open)
JADDR Valid raingage addresses in or-
der scanned by RTIME
JDAY Time days
JHR Time hours
JIDA Time days
JIHA Time hours
JIMA Time minutes
JISA Time seconds
JJ Index
JL Index
JMN Time minutes
JR Operation table for controlling
routing, combining and in line
division
JSEC Time seconds
JTD Time days
JTH Time hours
JTM Time minutes
JTS Time seconds
JX Index
K Index indicating number of
times file has been searched in
GATEDT
KA Flag
KD Time days
KDAY (TMADD) Days before adding
KF Index
KH Time-hours
KHR (TMADD) Hours before adding
Kl Index
KK
KKD
KKM
KKMM
KKS
KM
KMIN
KS
KSEC
KTD
KTH
KTHH
KTM
KTS
L
LAG
LD
LDAY
LF
LGT
LH
LHR
LIST
LIST A
LISTB
LISTC
LISTD
LISTX
LISTY
LL
(TMADD)
CHAIN 3)
(TMADD)
(GREAT 2)
(TYPE 1)
(TMADD)
(TMADD)
(GREAT 2)
(CHAIN 1)
(GREAT 2)
(CHAIN 1)
Number of points needed on
hydrograph for diversion
Time days
Number of time intervals base
flow factor should be applied
Number of time intervals base
flow factor should be applied
first time
Time seconds
Time minutes
Minutes before adding
Time-seconds
Index
Seconds before adding
Time days
Hour of day
Time hours
Time minutes; time into the
past that base flow must be
established
Time seconds
Not used
Index
*Fabridam with (ft)
Number of 5 minute periods
flow should be lagged (Lag <0,
lag time = |LAG| x 5 minutes;
Lag >0, lag time = (LAG +
k) x NRCHS x 5 minutes)
where k = 0 if NSTRL is odd,
k = 1/2 if NSTRL is even
Time days
Days before and after adding
*Fabridam width (ft)
Gate Number
Time hours
Hours before and after adding
'File name of constants for
Chain 1 (STARTLIST)
'File name of constants for
Chain3(HYDROLIST)
'File name of constants for
Chain 4 (DIVERLIST)
'File name of constants for
ChainB (ROUTELIST)
'File name of valid raingage
addresses (RAINGAGES)
'Not used
*Not used
Index
169
-------
LLL Dummy
LLOCI Index of names (NAMES) for
inlet hydrographs
LM Time minutes
LMIN (TMADD) Minutes before adding, minutes
after adding
LQ Index of hydrograph related to
given diversion structure
LS Time seconds
LSEC (TMADD) Seconds before adding, seconds
after adding
LSX (CHAIN 1) Time in seconds which model
back dates when starting to
compute real time hydrographs
(CHAIN 4) Time in seconds which model
back dates when starting to
compute diversions
M Index
MIDAY Model day
MIHR Model hour
MIMIN Model min
MISEC Model sec
MQ Index of maximum discharge at
diversion structure
MX Dimension limit
N (CONTL) Dummy parameter
(EVNTL) Number of points unequally
spaced being past to EUNTL
(TRAPOZ) Number of points on hydro-
graph
NA Index of orifice opening type
(rectangular, crescent, segment)
(EVNTL) Temp index
NAM (GATEDT) *Name in gate setting file
NAMES *Names of various points in
system
NC Number of catchments in
system
NCC Index of catchments
ND Numberof diversions in system
ND Time days
NDTT Number of points on hydro-
graph
NDTTB Index of hydrograph base
length
NDTTP Index of hydrograph peak
NDTTP1 Index of hydrograph peak + 1
NG Number of raingages in system
NH Time hours
NM Time minutes
NN (EVNTL) Number points generated so far
(RAINF) SeeNNN
NNG Index of raingages
NNGG Raingage index
NNN (EVNTL) Max number of points to be
generated by EUNTL; actual
number of pts generated
(RAINF) Number of rainfall intensities
NP Number of points that can be
used in current interval
NPA Number of points that can be
used in current interval based
on one gate record for a
structure
NPB Number of points that can be
used in current interval based
on 2nd gate record for a
structure
NPP (EVNTL) Accumulated value NP
NPT Index of last point on hydro-
graph for which diversion is re-
quired
NP1 Index of first point on hydro-
graph for which a given gate
and Fabridam setting holds
NP2 Index of last point on hydro-
graph for which a given gate
and Fabridam setting holds
NQ Index giving hydrograph num-
ber corresponding to diversion
number
NQI (GREAT 2) No longer used
NQQ Index of max discharge
(QMAX) for Type 6 diversion
structures
NRCHS Number of subreaches section
should be broken into
NSTRL Number of pts on hydrograph
to be averaged
NT (EVNTL) Time interval of evenly spaced
points (sec)
NTER Flag indicating (1-model is
starting over start from base
flow, GT. 1—add runoff to pre-
vious hydrographs)
NUMB (RAINF) Number of readings for each
rainfall gage
NX (RAINF) SeeNNN
N1 Index
p Wetted perimeter
Q (GREAT 2) Not used
Discharge before diversion (cfs)
(ROUTE) Temporary storage of discharge
hydrograph
QM Max discharge at diversion
170
-------
QMAX
QMIN
QN
QO
QP
QPB
QPP
QPPDP
QPPDX
QPPDZ
QS
QXXXDP
QZ
Q2
Q2S
03
Q5
R
(GREAT 2)
(GREAT 2)
(CHAIN 3)
(GATEDT)
(EVNTL)
(HYDR)
(VALY)
RA
RAIN
RAINDP
RAINDT
RDG
REX
RF
(RAINF)
Maximum discharge at various
points in system, additional
base flow at beginning of line
from suburbs. Upper limit on
discharge for bisection method
Lower limit on discharge for
bisection method
Last estimate of Q2
Discharge for which depth is
desired (cfs)
Peak discharge of hydrograph
(cfs)
Mean daily base flow (cfs)
Hydrographs at all points for
all times
File name of hydrographs at
inlets before diversion
(QPPXXDUMP) saved for fu-
ture additions
File name of hydrographs at
inlets after diversion
(QPPDIVERT) passed to
routing
File name of hydrographs at
inlets before diversion
(QPPZZVERT) passed to
diversion
No longer used
No longer used
Seed
Discharge to interceptor after
diversion
Discharge to interceptor after
diversion 90" line
Portion of flow from 60" line
SeeQ2
Not used
Hollerith "R"
Values at uneven intervals
SeeRF
Hydraulic radius
Average rainfall intensity over
watershed
File name of rainfall file to be
processed
No longer used
File name of rainfall intensities
at even intervals for HYDR
(IRAINDATA)
Raingage reading
Excess intensity (in/hr)
Rainfall intensity for (DTT)
time period
RO
RR
S
SAVE
SDATA
SIMPSN
SLOPE
SO
STLOS
SUM
T
TB
TEMP
TEMPX
TF
TH
TMM
TP
(HYDR)
(HYDR)
(RAINF)
(RAINF)
(CHAIN 4)
(GATEDT)
(GATEDT)
(GREAT 2)
TTB
TTP
TTT
TTTT
V
VOL A
VOLB
VOLC
X
XDECAY
XI
XINTL
XKD
XKH
XKKD
XKKS
XKM
XKS
XL
(GREAT 2)
(GREAT 2)
(CHAIN 1)
(HYDR)
(GATEDT)
(EVNTL)
(GATEDT)
(EVNTL)
XLS
See RF
Values at even intervals
Slope of interceptor at meters
Fraction of peak discharge on
unit hydrograph
Temp, storage of raingage read-
ings
File name of input data file to
GREAT
Temporary value
Rainfall rate (intensity, in/hr)
Bed slope at meters
Starting value of initial loss
parameter for impervious sur-
faces
Total of valid initial rainfall
readings; total of valid rainfall
readings
Total volume to river (cf)
Hollerith - operation
Base time of unit hydrographs
(sees)
Temp file name (TEMPOFILE)
Not used
Interpolation factor between
two values
Temporary value
No longer used
Time to peak of unit hydro-
graphs (sec)
Temp, variable ITB
Temp, variable ITP
No longer used
Common block name
Flow velocity (fps)
Volume before diversion (cf)
Volume after diversion (cf)
Volume to river (cf)
Temporary value
Loss rate recovery parameter
Hollerith "I" Insert
Spacing of points
Time days
Time hours
Time days
Time seconds
Time minutes
Time seconds
Hollerith "L" listing
Index
Time seconds
171
-------
XNP Number of points that can be
used in current interval
XNT Time interval - seconds
XRAIN File name for real time rainfall
data file (RAINRPDLY)
XSUM Total volume to river
XXKS (EVNTL) Time seconds
XXXI Raingage name
XXX2 Raingage name
XI C1 for one structre
X1Q C1-
X11 C11
X12 C12
X2 C2 for one structure
X3 C3 for one structure
X4 C4 for one structure
X5 C5 for one structure
X6 C6
X7 C7
X8 C8
X9 C9
Y (DIV) Depth (ft)
Gate opening (ft)
YMAX Max depth for bisection
method
YMIN Min depth for bisection method
YY Gate opening (ft)
Y1 See Y (DIV)
ZZ Volumes of hydrographs before
and after diversion and vol. to
river (cf)
172
-------
APPENDIX C
Listing of the Model
The following is a complete listing of the model program. It is recognized that the
programs in the form listed are not always accomplishing maximum efficiency. The
programs have not undergone refinement but are logically correct.
PROGRAM TO GREAT MAG TAPE OF fiATA FILES 27 SEP f,B
REftL LISJ<2),LISTA(2),LIST3<2),LISTC<2).LISTD(2>
REAL LISTX(2),LISTY<2>
DIMENSION RAINDP<2),DF(19),NQ( 19>,NQO( 19 > ,01 VCC4 10). 11 81 18) .
11TP<15>.ITB(15>
DIMENSION QS<15),A(17),NA< 17), QPPDPt 2) .QXXXDP<2* ,L,_OCI (15)
COMMON/BLOCKS/J«(* 8, 7) ,QHAX( 48 ) , D< 3 ) ,S < 3 ) ,QP B( 15 , 1) . &Ni, ALPH A < 4 8 ),
1 ICONTC.LLOC1,BFFAC(24)
COMMON/BLOCKL/L,ITQI,MQI,a<100,4)
COMHON/BLOCKO/DF.NQ.NA.NQQ,A,DIVCC
1,DIVC<17.5)
COMMON/BLOCKH/ C2( 15 ,9 ).TP(15) ,TB<15),ALOSC( 15.15),A.
1 FACT(15).IADDRC9). JADDR(9)
DATA LISTA(1),LISTA(2)/5HHYDR0.4HLIST/
DATA L1STB( I),L ISTB(2)/5HDIWER . 4HL1ST/
DATA LISTC(1).LISTC(2)/5HROUTE .4HLIST/
DATA LISTD(1),LISTD(2)/5HRAING .4HAGES/
DATA L1ST( 1),LIST(2)/5HSTART,4 ML IST/
DATA QXXXDPC 1) ,QXX XDP ( 2 ) /5HQ XX XX. 4HDUMP/
DATA QPPOP( 1),QPPDP(2)/5HQPP-XX .4HOUMP/
DATA RAINDP(1).«AINDP<2)/5HRROIU,4HDUMP/
DO b 1=1,18
6 IIB(I)=0
CALL IMPUTHCNG.NC,DT.DTT)
CALL (MPUTD(ND)
CALL ROUTJN(TMM)
DO 107 NCC=1,NC
TFP-=TP(NCC)/DT
IFP (NCC)=TTP
1F( (TTP-FLOAT( ITP(NCC) I ) .LT.O. S»GO TO 104
103 ITP(NCC)=ITP(NCC)*1
104 TTB-(TB(NCC)-FLOAT(IT?(NCC))»DT)/DT
1TB(NCC)=TTB
IF(tTTB-FLOAT(iTB(NCC))).LT.O.S)GO TO 106
105 ITB(NCC)= ITB(NCC)*!
106 ITB(NCC)=ITB(NCC)*ITP(NCCI
107 CONTINUE
URITE<5. 108) ITP.ITB
108 FORMAT«5H ITP 15I5/5H ITS 15151
ITM-0
DO 2 I=1,NC
IF< IT8( I)-ITM)2,4,4
4 ITM=IT8(I)
2 CONTINUE
CALL CLOSE ( 3)
CALL CLOSE (8)
CALL DLETEC8.L 1ST, I)
CALL ENTER(8,LIST)
WRITE(8)DTT,DT,IIB.LLOCI,ALPHA.L1ST A.LI STB.LISfC,LISTD.NG.NC
CALL CLOSEJ8)
CALL DLETE(8,LISTA,I)
CALL ENTER<8,LISTA)
WRITE(8)ITP,ALOSC,AREA,CZ,QP8,ITB,QPPDP.BFfAC
CALL CLOSEC8)
173
-------
CALL DLETE<8.L IST8, I)
CALL ENTER<8,LISTS)
WRITE(8)ND,NC,DF,NO,NA,NQQ,DIV£.DIVCC, A.QMAX.FACT
CALL CLOSE(8)
CALL OLETE(8,LISTC.1)
CALL ENTER (8.LISTC)
WRITE <8)JR,OMAX,D,S, AN, I CONTC.OF,0 IVC,D I VCC
CALL CLOSE(8)
CALL DLE-TE(8.LISTO.I)
CALL ENTER(8.L ISTO)
URITE(S) IAOOR, JADO R
CALL CLOSE(8)
STOP
END
SUBROUTINE INPUTDtND)
DIMENSION C1(17).C2( 17J.C3U7) ,C4< 17).C5<17)
CONMON/BLOCKQ/ OF( 19),NQ(19),NA<17),NQQ(19).A(I?>,01VCC(10)
1 ,DIVC(17,5)
REAO(3.8)A
ND=19
8 FORNAKA5)
REAOO.DNQ
REAOO.DNQQ
REAOO.DNA
1 FORHATC14I5)
WRITE(5,2)
2 FORMAK31H HYOROGRAPH NO AND DIVERSION NO)
WRITE(5.3) (NQ( I),I .1 = 1,19)
3 FORHATt 215,5X)
URITE(5.4)
4 FORMAT(/25H INDEX OF DIVERSION POINT)
WRITE(5,3) (NQQ(I), I, 1=1,19)
URITE(5.5)
5 FORMAT(/13H ORIFICE TYPE)
WRITE(5.3)(NA( I),I . 1 = 1,17)
READ(3,6)C1
REAO(3.6)C2
READ(3.6)C3
REAO(3,6)C4
REAO(3.6)C5
6 FORMAT(7F10.5>
REAO(3.6)C6.C7,C8,C9,C10,Cll,C12,C13.C14,C15
WRITE(5.7)
7 FORHAT(/9X ,3H C1.7X.3H C2.7X.3H C1.7X.3H C4,7X,3H C5,7X,3H C6
U7X.3H C7.7X.3H C8.7X.3H C9,7X,4H C10.6X.4H C11.6X4H C12)
00 25 1-1,17
25 URITE(5.24)I,C1(I).C2(I),C3(I) ,C4(I) ,C5( I)
24 FORMATC 15.5F10.5)
9 FORHATJ I5.5F10.5)
WRITE(5.10)C6.C7.C8.C9.C10.C11.C12
10 FORMAT(55X,7F10.5)
WRITE(5.11)C13 .C14.C15
11 FORMAT(4H C 1 3 , F 10 . 5 , 5 X . 4H C 1.4 , Fl 0. 5, 4H C15.F10.5)
REAO(3,6)DF
WRITE(5, 12 )
12 FURMATI/19H FA8RIOAM DIAMETERS)
WRIT£< 5.13)( I,DF(I ) , 1 = 1,17)
13 FORMATl IS.F10.S)
DO 99 1=1, 17
DIVC< I.D-CK I >
174
-------
DIVC(I.2)-C2( I J
DIVCCI.3)-C3( 1)
DIVC(I,4)=C4( I J
99 DIVC(I.5-) = C5( I >
DIVCCt I) = C6
DIVCC12) =C 7
DIVCU 3)=C8
DIVCC(4)= C9
DIVCC(5)-C10
DIVCC(6) -Cll
OIVCC( 7) =C 12
DIVCC( 8) =C 13
DIVCC(9)=CU
OIVCC(10)=C15
RETURN
END
SUBROUTINE INPUTH< NG,NC,DT,0TT )
DIMENSION SDATA<2)
COMMON/BLCCKH/ CZ< 15 , 9 ),TP(15) , T8<15).ALOSC( 15.15),A(15) .F4CJ( 15)
1 IADDRC9),JADDR<9)
C READ QUANTITIES UHICH ARE CONSTANT FOR WHOtEl PROBLEM
C NG-NO OF RAIN GAUGES.NC-NO OF CACHMENTS,DT-T I ME INTERVAL FGR
C RAIN GAUGE READINGS,OTT-TI ME INCREMENT FOR HYDROGRAPHS
12 URITE(b,10)
10 FORMATOOH ENTER NAME OF INPUT DATA FILE)
READ(b,ll)SDATA
11 FORMAT
C LOSS COEFFS.XINLOSS-INITIAL LOSS,FC-FINAL LOSS.RATE.FO INITIAL
C LOSS RATE.XK-EXPONENT.A A-A NTECEDANT INDEX
READ13, 105)((ALOSC(I.J),J-1, 15 >. 1-1,NC)
105 FORMAT( 15F5.2)
WRITE(5.113)
113 FORMAH//20H LOSS RATE VARIABLES/
175
-------
193H NO. TSTLOS RECOV AX DECAY PERV PERVCF ENOK T8MC BMC STLOS TSTL
20SP STLCJSP COEF POWER ACCLOSS /)
WRITE(5.106)(I.ULOSC(l.J>,J=l.l5).I=l,NO
106 FORMATU5, 15F8.2)
READ(3.3) (A( I),FACT< I ) , 1 = 1, NO
WRITE( 5,107)
107 FORMATU3H AREAS(SQ FT ) , 5 X , 7HF ACTORS , 3 X , 8H CACH.NO)
WRITE<5.108)(AU).FACT<[),I.I=1.NC)
108 FORMAT(1X,F10.0.6X,F10.2, 16)
C VALID RA1NGAGES ADDRESSES IM ORDER USED BY MODEL AND RTIME
READ ( 3,115) ( IADOR( I ),JADDR( I) ,.I = l.NG)
WRITE (5,116)
116 FORMAT(//54H RAINGAGE ADDRESSES IN ORDER USEO 3Y MDDEL AND SCANNED
1 //)
WRITE(5,114)(1,1ADDR(I),JAODR( I), 1=1,NG)
114 FORMAT(2X,15,217)
115 FORMAT(215)
C RAIN GAUGE PROPORTIONS FOR EACH CACHMENT
DO 109 1=1,NC
109 READ(3,1)(C2(I,J),J=1,NG)
WRITEC 5,110)(J,J=1 ,NG)
110 FORMAT
RETURN
END
SUBROUTINE ROUTJN(TMM)
COMMON/8 LOCKS/J«(48,7),QIAX «.8),D(3),S(3),QP8t15,1) . AN .ALPH( 48),
11CONTC,LLOCI(15>.BFFAC(24)
READC3.1) ICONIC
WRITE(5,5)
DO 2 I=1.ICONTC
READ ( 3,1 )( JR( I,J), J-l, 7),QMAX( I),A|_PH( I)
2 WRITE(5.6)I.(JR(I,J),J=1,7).QMAX(I),ALPH(I)
1 FORMAT(7I5,5X.F10. 2,A5)
3 FORMAT(7F10.0)
READ(3,15)(D(I),S(I),AN.I=1,3)
15 FORMATC3F10.1)
READ(3.3»QPB
READ(3,3)BFFAC
6 FORHAT(8I8,F10.1. A5)
5 FORHATtSOH OPERATION TABLE.HYDROCRAPH NOS AND ROUTING CONSTS
1/56H OPER H1O ENTER HYD COMB HYD LEAVE NSTRL LAG
24X.18H NRCHS QMAX >
WRITE(5,9)(D(I).S(I),AN,I=1,3)
8 FORMAT(15,F10.1)
9 FORMAT*//(10X.7H DIAM =F10.2,9H SLOPE =F10.5,
113H MANNINGS N -F9.3))
WRITE(5,10)
10 FORMAT(//10H BASE FLOW)
WRITE(5.8)(I,QP8CI.I),!=!, 15)
WRITE (5. 12 )( I, 8FFAC( I). I = 1,2 4)
176
-------
12 FORMAT(//5X,24HHOURLY BASE FLOW FACTORS/ /< 5X , I*. F6. 2 »
READ(3.20)LLOC I
20 FORHATU5I3)
URITE(5.11>( I.LLOC Id >, 1 = 1,15)
11 FORHATI//11H NAME INDEX/( 1 5, 15 I >
RETURN
END
CHAIN 11
REAL LIST! 2) .L ISTA.LISTB.L ISTC.L1STD.L ISTX(2),LlSTY(2) .NAMES
DIMENSION QPP( 15, 1 20 > , I1B< 18 ) , LLOCI ( 15 ) .NAMES (48) ,LISTA( 2) .
lRAIN(2>.XRAIN(2),LISTB<2).ulSTC(2),lISTO(2)
DIMENSION OF( 19),NQ( 19) ,
1NA(17).NQQ(19) ,OIVC(17,5),DH/CC(10),A(17).SAINOP(2),0(3),S(3)
COMMON DTT,NDTT,IDT,NTER,DT, II 8.LLOC I,
1 NAME S, L I ST A, L I STB . L I STC , L I S T 0 , KK , I CON T A , I CON TB ,
2ICONTC.ICONTD, 1C, I Y, MI HR , M IM IN .M IDAY ,M I SEC , R AI N. NG. NC . QPP ,
3ITIMES(12)
DATA LtST( 1) ,LIST( 2 ) /5HSTA «T .4 ML IST/
DATA XRAINl 1), XRA I N ( 2 > / 5HR A I NR .4HPDL Y/
DATA RAINDP( 1) ,«A I NDP( 2 ) / 5HRRO XX ,4HDUMP/
RAIN( 1)-XRAIN( 1)
RAIN(2)=XRAIN( 2)
CALL SEEK (8, LIST)
REAO(8)DTT,OT, 118, LLOC I ,.MAME.S. LISTA, L I SI B,L [STC. LISTD, NG,NC
CALL CLOSE (8)
IDT=300
NDTT=120
KK=85
MISEC^O
900 FORMATtSX. A5, A4>
NTER=1
IY = 0
ICOWTA=1
ICONTB=1
ICONTC=1
ICONTD=1
CALL SENSEdb. I1B( 17) )
IF(IIB(17) .EQ. 1>CO TO 13
WRITE16.2)
> FORMATC61H SENSE S W I TCH E5< ACCU M. SJS) "1-10. 17 ARE UiED IN THIS P
1ROGRAM/45H IF ON(UP) THEY HAVE THE FOLLOWING FUNCTIONS//
250H «1 SURPRESS PRINTOUT OF EXCESSIVE FLOW LOCATIONS/
328H «2 PAUSE IF EXCESSIVE FLOW/
437H «3 PRINTOUT RAINFALL .OSS-RflTE DATA/
536H «4 PRINTOUT UNDIVERTED HYOKOGRAPHS/
634H »5 PRINTOUT DIVERTED rtYDROCRAPHS/
736H »6 PRINTOUT VOLUMES OF HYOROGRAPHS/
853H «7 PRINTOUT HYDROGRAPiHS AT POINTS OF EXCESSIVE FLOW/
939H «8 PRINTOUT HYDROGRAPHS AT ALL POINTS)
WRITE(6.4)
4 FORMATCVOH «9 COMPUTE UNDIVERTED HYDROGRAPHS OWLY)
5 FORMAT!
154H »10 COMPUTE UNDIVERTED AND DIVERTED HYDROGRAPrfS ONLY)
WRITE(b.20D
201 FORMATiiOH «12 SCAN REAL TIME DATA ONLY)
WRITE<6«200>
177
-------
200 FORMATC32H «13 00 NOT SCAN REAL TIME DATA/
131H "14 RETURN TO PROGRAM CONTROL'
238H »15 PROCESS HISTORICAL RAINFALL FILE/
347H «16 SUPPRESS PRINTOUT OF PROGRAM. INSTRUCTIONS)
URITEC6.6)
6 FORHAT(
241H "17 PAUSE AT END OF LATEST COMPUTATIONS)
WRITE(6. 7)
7 FORMAM/
347H SET SWITCHES TO DESIRE} POSITION AND THEN TYPE)
WRITE(6,8)
8 FORNATt
427H CONTROL P AFTER THE PAUSE)
URITE(b,9)
9 FORHAT(
538H THE SWITCHES MAY BE RESET AT A. MY-. TIME)
PAUSE
12 CALL SENSEC15, 118(16))
IF( 118(16) .EQ.O)GO TO 26
102 URITE(6.22)
22 FORMATUX.51H ENTER NAME OF HISTORICAL RAINFALL F!L£ (FILNAMEXT))
READ (6t23)RAIN
23 FORMAT(A5. A4)
GO TO 25
26 CONTINUE
WRITE(b. 10)
10 FORMATdX, 15HENTER REAL TIME/LX.10H HR MN DAY)
READC6.11) IHR, 1NIN. IDAY
11 FORMAT(2I3 ,14)
ISEC=3600»IHR*60»IMIN
CALL SETdSEC.IOAYJ
MISEC=0
MIDAY= IDAY
IF(II8(16) .ECJ. II GO TO 23
21 CONTINUE
LSX=((OTT-KK-1I»IOT
CALL TMADO(MISEC.MIMIN.MIHR,MIOAY,LSX.O,0.0)
20 TTT= 0.0
GO TO 100
25 CONTINUE
CALL FSTAT(8.RA1N. ITEST)
IF(ITEST.NE.O)GO TO 26
WR1TE(6, 10DRAIM
101 FORHAT(5X,4HFILE,2X, A5.A4, 2X.2 3HNOT PRESENT ON DEVICE 8/ )
GO TO 102
100 CONTINUE
CALL CMAIN12)
STOP
13 ICONTA=0
GO TO 12
END
SUBROUTINE TMADD(L SEC ,LMI N ,L HR .LDAY, KSEC . KM I N , K«R,
-------
LDAY=LDAY «• KO AY
IF(LSEC.GE .O.ANO.LSEC. IT. 60)30 10 I
LMIN^LMI N*LSEC/60
LSEC=LSEC-(LSEC/60)»60
IFUSEC.GE .0)00 TO 1
LSEC=LSE060
LMIN=LMIN- 1
IF(LMIN-GE .O.AND.LMIN.LT.6J) GO TO 2
LHR-LHR *• LMIN/60
LMIN=LMIN-'60
IFUMIN.CE .0) GO TO 2
LM1N=LMIN + 60
LHR^LHR-1
IFUHR.CE. O.AND.LHR.LT.24) GO ID 3
LOAY^LDAY * LHR/24
IF(LHR.GE. 0) GO TO 3
LHR-LHR * 2^
LDAY=LDAY- 1
3 RETURN
END
C CHAIN2
C GET RAINGUAGE READINGS, SET GATES, CHECK LEVELS
REAL LISTA.L IST8.L ISTC ,L IS TD . N AMES
COMMON OTT,NDT T.IDT. NTER.D T, II S< 18) ,LL OC [(15 ) ,NAMES(4a ) ,
1LISTA(2).LISTB(2).LISTC(2».LISTD(2).KK. ICONTA, ICONTB.
2ICONTC.[ CONTD, 1C. I Y. MI HR , M I M IN . M [DAY , M I SEC. R A I Nl 2 ) . NG, NC ,
30PPI 15.120), ITIMES( 12)
CALL SENSE(1«. IIb( 15))
IF( I 18(15) .EQ. 1) GO TO 3
CALL SENSE (13. I I3( 14 >>
IF( 118(14) .EQ. 1>CO TO 6
IF( ICONTO-LT.OGO TO 6
CALL CHAIN (9)
STOP
b CONTINUE
ICONTDMABS( ICONTD)
CALL SENSE (9, I 18(10) )
CALL SENSE( 10, IIB< 11 »
CALL SENSE (17. IIB( 18) )
CALL SENSE (?., 118(3 ) )
IF( IIBC 16) .EQ. 1.ANO.NTER.EQ..1) GO TO 3
IFUIBU8) .EQ. l.OR .1 IBU5) .EQ.l . OR . I CO NT D . EQ . 2)CO TO 1
IF
-------
CALL CHAIN ( 5)
STOP
END
SUBROUTINE CONTL
REAL NAMES,LIST*,LISTB.LISTC.LJSTD
DIMENSION BMSTH2)
COMMON OOT.NOTT.IDT.NTER.OT. I IB(18).LLOCK15).NAMES ( 48).L1ST A(2 ).
1LISTB(2).L1STC(2).LISTD(2),KK,JCONTA,ICONTB, ICOflTC,ICONTD,IC, IY,
2MIHR,MIMIN,MIDAY,MISEC,RAIN(2) ,NG,NC,QPP(15, 120J , IT I MESI 12>
DATA BMSTL(1),BMSTL(2)/5HBMCXS .4HTLOS/
4 7 WRITE(6.99 HCONrO, MI DAY, MI HR , HIM IN
99 FORMATd6H PROGRAM CONTROL,15H PREVIOUS CHAIN, I4.12H MODEL TIME
1, 15,213)
ICONTD=2
27 IF( ICONTA.EQ.O) GO TO 46
WRITE(6.28)
28 FORMATC//
146H TYPE 1 TO RESTART WITH ALL -.INITIAL CONDI TtOflS/
251H TYPE 2 TO RESTART COMPUTING UNDIVERTED HYDROGRftPHS/
352H TYPE 3 TO CONTINUE COMPUTING UNDIVERTED HYDROGRAPHS/
450H TYPE 4 TO CONTINUE COMPUTING DIVERTED HYDROGRAPHS/
527H TYPE 5 TO CONTINUE ROUTING/
638H TYPE 6 TO CONTINUE SCANNING REAL DATA/
729H TYPE 7 TO CALL GATE SETTINGS/
815H TYPE 8 TO STOP/)
ICONTA=0
46 CONTINUE
45 FORMATI12H TYPE NUMBER)
48 READ(6.35) IZ
35 FORMAT(Il)
IFUZ.LT.l .OR. I2.GT.8) GO TO 50
GO TO 191, 92,93,94,95,96,97.98 ), 12
50 WRITEC 6.40)
40 FORMATdBH TYPE NUMBER AGAIN)
GO TO 48
91 CALL CHAIN ( 1)
STOP
92 NTER=1
URITEC 6.100)
100 FORMATtlX,25HENTER MODEL STARTING TI ME/I X.10H MR MM DAY)
READ(6.101>MIHR.MIMIN.MIDAr
101 FORMATC2I3,14)
C WRITE(5,102)MISEC,MIMIN,MIHR, Mi DAY
C 102 FORMAH8H TIME 1 ,415)
CALL RAINF
C URITE(5.103)MISEC.MIMIN.MIHR,MIDAY
C 103 FORMAT(8H TIME 2 ,415)
BMC^O.O
STLOS^O.O
CALL DLETE(7,BMSTL,N)
CALL ENTER(7.BMSTL)
DO 37 1=1.NC
WRITE(7,36) BMC.STLOS.BMC.STLOS
37 CONTINUE
36 FORMATdX, 4F5.2)
WRITE(7.38) BMC
38 FORMATdX, F8.II
180
-------
CALL CLOSE(7)
CALL CHAIN ( 3)
STOP
93 NTER=2
CALL RAINF
CALL CHAIN (3)
STOP
94 CALL CHAIN (4)
STOP
95 CALL CHAIN (5)
STOP
96 ICONTD=2
CALL CHAIN(9)
STOP
GO TO 46
97 CALL GATEDT
GO TO 47
98 STOP 777777
END
SUBROUTINE GATEOT
REAL NAM
DIMENSION GATES(2).TEMP(2),IIB(18>
DATA GATESi 1) , GATE 5>( 2 ) / 5HG AT ES .4HETTG/
DATA XL. XI ,R,D,E/1HL, 1HI ,1HR ,1 HO.IHE/
DATA TEMP( 1) , T EMP( 2 ) / 5HTEMPO , 4 HF I LF. /
J = 0
WRITE<6.2)
2 FORMATU2H GATE EDITOR/)
CALL SENSE(16, I IB< 17) )
IF< I 18(17) .EO. 1) GO TO 3
URITE<6,4>
4 FORMAT(43H TYPE L FOR LISTING UF PRESENT CATESEIT INGS/
154H TYPE D WITH GATE NUMBER AND TIME- TO DELETE A SETTING/
261H TYPE I WITH GATE NUMBER. TIME AND SETTING- TO INSERT SETTING/
361H TYPE R WITH GATE NUMBER. UME AND SETING- Tfl REPLACE SETTING/
434H TYPE E TO EXIT TO PROGRAM CONTROL/)
3 WRITE(6.5)
5 FORMATd X. 24HN GT HR MN DAT X.XX X.XX)
REflO(6,6)T,IGT,JH, 1M.ID.HG.HF
6 FORMAT(A1,313.I4.2F5.2)
10 FORMAT(25X,26HGATE AND FABRIDAM SETTINGS//
110X.9HDIVERSION8X4HT IMEIOX4HGA TE4X8HFABR I DAM 3X3MMAX6X3HMAX/
211X6HNUMBER7X2HMR3X3HHMN2X3HDA Y<» X7HSET T I NG3 X 7HS€ T TI NG3 X4 HGAT E
32X8HFABRIOAM/ )
K=0
1^0
IFU.EO.XL ) 1=1
IF(T.EQ.D)1=2
1FCT.EQ.XI ) 10
IF(T.EQ.R)1^4
IF(T.EQ.E)1=5
IFU.LT.l.OR.I .GT. 5)GO TO 606
IF(I.EQ.5)GO TO 500
IF
-------
100 IF (J.EO.l) GO TO 600
CALL SEEK (7,GATES)
WRITE (5.10)
13 READ (7.11) LGT.NftM, LH.LM, LD , H£L , HFL , HGM AXL, HFKAXL
IF (LGT.EQ.20) CO TO 14
WRITE (5.23) LGT.NAM.LH.LM.LD,HCL.HFL,HGMAXL,HFHAXL
GO TO 13
1* CALL CLOSE (7)
GO TO 3
300 IF
-------
SUBROUTINE E VNTLC N, NNN, NT, ISS, 1MM.IHH, [DO)
C TAKE N UNEQUALLY SPACED EVENTSRAND GENERATE NN EtfENTSRR WITH SPACING 0
C STARTING AT TIME ISS . I MM. I HH, I DO.
DIMENSION R( 35 ).RR(35) .IS( 35). 1M( 3'3 ) . I H( '35) . IDl 3 5 )
C R IS AVERAGE RATE FOR LAST PERIOD UP TO TIME OF R.
COMMON/BKTL/R. IS. I M , I H , 10 , RR
C GO BACK IN RECORD TO READIN3 NEJ(T PREVIOUS TO ISS , IrtM, I HH , IOD
NN-0
2 J
M^N-J+H
IFXKM»60.*XKH»3600.+XKD»at>400.
KS=IS( IA>
KM-IM( IA>
KH= IH( IA)
KD= I0< 1A)
CALL TMAOD(KS,KM, K H ,KD,- I S S , - I MM.- I HH. -I DO)
C WRITE( 5. 198)KS.XM, KH,KD
IFtKD.LT -0)GO TO 3
XKS-KS
XKH^KH
XKD=KD
XNT-NT
XKS=XKS*XKM»60.*XKH»3600. + XKD» 8*600.
NP-XKS/XNT*! .
XNP=NP
XXKS-( XNP-1. 0) »XNT
IF« XKS-XXKS) .EQ.XNT)NP = NP*1
NPP = 0
DO 4 L=l .NP
NPP-NPP+1
XL=L-l
183
-------
NN-NN+l
DR=RUA)-R< I )
TF = (XLS-XKS«-XNT»XL )/Xl_S
RR ,NN, I ,R< l),R( U)
C 11 FORMAU20X.I4, 3F10 .5.2I4.2F10. 5)
IF(NN.EQ.NNN)GO TO 12
4 CONTINUE
11 CONTINUE
XNP-NPP
XKKS=XNP»XNT
C HRITE<5.199)XKKS
C 199 FORMAT(1X,7H XK.KS ^,F15.2)
IFUKKS.LT. 131000. ) GO TO lb
KKD=XKKS/86400.
XKKD=KKD
XKKS=XIUS-XKKD»86400.
6 KKS-XKKS
17 CflLL TMADD ( ISS. IMM, IHH. IDD, KKS. 0,0. KKDJ
C WRITE <5. 10)XLS,XKS,NP. NN.KKS. M.J.I, NPP
C 10 FORHAT(2F1 0.1, 216, 110.4I&)
IF(NN.EQ.NNN)GO TO 13
3 CONTINUE
13 NNN=NN
KKS=-NT
CALL TMADDUSS.IMM, IHH.IDO.KKS .O.C.O)
C HRITE(5.102) ISS.IMM. IHH.IDO
C 102 FORMAT(8H TIME 6 ,515)
RETURN
16 KKS=NPP»NT
GO TO 17
95 NNN=0
RETURN
END
SUBROUTINE RAINF
INTEGER CHECK
REAL LISTA.L IST8.L I S TC , L I T, TO , N AMES
DIMENSION RAINOT(2).SAVE(33,9) . I TIME ( 4 . 35, 9 ) . I ADDR( 9) .
UAOORt 9).NUMB(9)
COMMON /8KTL/R( 35), IS( 35) . I ri( 35 ).IH( 35).IOl35 ).RR(35)
COMMON/TTTT/GAGE(9 ) ,RF (9)
COMMON DTT.NDTT.IDT.NTER.DT.'I IB< 18).LLOC I I 15),NAMES< 48 ),
1LISTA(2).LISTB(2).LISTC(2),LISID(2).KK, 1 C UNT A, ICONT8,
2ICQNTC.ICONTD, IC.IY,M[HR.M1M.IN.MIDAY,MISEC,RAIN<2).NG.NC,
3 SAVE, I TIME .IXXX<17 10), ITIMtSl 12)
DATA RAINOT(1).RAINOT(2)/5HIRA1N.4HOATA/
CALL SEEM 8.LISTO)
CALL OLETE(7,RAINDT. I)
CALL ENTER(7,RA/NDT)
READ(8) IAOOR.JAOOR
CALL CLOSE (8)
CALL SEEK( 8,RA IN)
DO 20 1=1. NG
20 NUM8( I )=1
CHECK=1
IDTT=IDT
184
-------
C WRITE( 5,505)MIS£C, MIMIN.MI HR.M 1DAY
C 505 FORMAT(20X .6181
1-1
NNG=NG
201 READ (8. 10) IDA, IHA. IMA.IAODA.XXK1.XXX2.RDG
10 FORMAH3I3.I6. A5.A2.F9.3)
IF(IADDA.CE.7777) GO T 'J 210
IF( 1AODA.EQ. JADOR(NNG) ) GO TO 203
208 NNG-NNG+1
CHECK=1
IF1NNG.GT. NG) NNG = I
00 101 1=1. NG
NNGGM
IF( IAOOR( I ).EQ.JADOR(NNG) > GO TiJ 11
101 CONTINUE
11 CONTINUE
GAGE(NNGG) -XXXI
C HRITE(5.500)NNC,NNGG, JADDR
C 500 FORMAK 318 .5X, AS)
1 = 1
GO TO 221
202 REAO(8.10) IDA, IHA, IM A, I A DO A, X X Ml , XXX 2 , RDG
IF(IADDA.GE.7777) GO TO 210
IF( IAOOA.NE. JAOOR(NNG) ) GO TO 208
221 IF(ROG.LE. ( - . 5 ) .OR . ROG . GE . 3.0) ROG-99999999.
207 ITIME( l.l.NNGOMOA
ITIHE( 2. l.NNGO-IHA
ITIHEO. l.NNGG) = IMA
ITINE(4.1.NNGG>=0
SAVE(1.NNGG)=ROC
1 = 2
GO TO 201
203 IFIRDG.LE. (- .5 > .OR .RDG .GE. 3.0) GO TO 204
IFCSAVEt I-l.NNGO.GT .3.0) GO TO 205
DIFF=ROG-SAVE( I-l.NNGG)
IF(OIFF.LT .(-.02)) GO TO 204
JIOA=-ITIME( 1. I-l.NNGG)
JIHA = -ITIH£(2. I-l.NNGG)
JIHA=-ITIME(3. I-l.NNGG)
JISA=0
CALL THAOO(JISA,JIMA, JIHA. JIDA.O.INA .IHA. IDA)
OELT=JIOA»2**JI«A*JIMA/60»JISA/3600
SLOPE=DIFF/OELT
IF(SLOPE .GT.6.0) CO TO 204
GO TO 205
204 ROG=SAVE( I-l.NNGG)
IF(CHECK.EQ.O) RDG =99999999.
205 IF(CHECK.EO.O) GO TO 206
JIHA=-IHA
JIMA=-IMA
CALL T(1AOD(0, J IMA, JIHA, JIDA.O, MIMIN.MI
IF( JIDA.GE.O) GO TO 207
CHECK-0
206 ITIHE< 1. I. NNGG) = IOA
ITIME12, 1, NNGG)=IHA
ITIMEt 3. I.NNGO-IHA
ITIHE(4, I,NNGG)-0
SAVE( UNNGG)=RDG
185
-------
NUMB GO TO 210
IF( IADDA.EQ. JAOOR(NNG) ) GO TO 209
GO TO 208
210 IF( I.LE.2) GO TO 300
MX=35
DO 211 I=1.NC
N=NUMBd)
C WRITE(5.503)GAGE( I ),NUMB( I )
C 503 FORMATC5X, A5.5X.I5)
DO 212 J=1,N
R(J)=SAVE< J, I)
ID( J>- IT IMEU, J.I )
IH< J)=ITIHE(2, J.I)
IM( J)=ITIME<3, J.I)
IS(J)=0
212 CONTINUE
C URITt(5.501>R
C 501 FORMAT<1X.10F13.3)
CALL SENSE(3, IIB(4))
NN=NDTT-KK
IF( IIB(16) .EQ. I) NN=13
IXS-MISEC
IXM-HIMI N
IXH-MIHR
IXD=MIDAY
CALL EVNTL(N,NN.IOTT. I XS , I XM , I XH, IXD )
NUMB(I)=NN
IF(NX.GT.NN) MX-NN
00 213 J-l.NN
213 SAVEU. I >-RR( J>
C URITE(5.502)NN
C WRJTE( 5.50DRR
C 502 FORHATtllO)
211 CONTINUE
IF(NN.EQ.MX) GO TO 214
ISEC=(MX-1)»IDTT
CALL TMADOCMISEC. HIM IN , Ml HR, M I BAY . ISEC .0.0,0)
C URITE( 5. 505) MI SEC, HIM IN. MI HR , M ID AY. I SEC
GO TO 215
214 MISEC=IXS
MIMIN=IXH
MIHR-IXH
MIDAY=IXO
215 SUM^O.O
C URITE<5.505)MISEC. MI M IN. MI HR , M iOAY
COUNT=0.0
NN = HX
DO 216 1=1 ,NG
IF(SAVE(1. I) .GE.3. 0) GO TO 216
COUNT=COUNT*1.0
216 CONTINUE
AVG-SUM/COUNT
00 217 [=1.NG
186
-------
IF(SAVE(ltI) .GE.3.0) SAVE(1,I> =
217 CONTINUE
DO 218 1=2,NN
SUM=0.0
COUNT-0.0
DO 219 J=1,NG
IF(SAVE(I.J).GE.3.0) GO TO 219
SUM=SUM*SAVE(1.J>
COUNT-COUNT* 1.0
219 CONTINUE
AVG1=SU«/COUNT
DIFF=AVG1-AVG
DO 220 J=l,NG
IF(SAVE(I,J).GE.3.0) SAVE(b,J) -SAVE ( I- 1. JJ+0 IFF
220 CONTINUE
218 AVG^AVGl
IF(IIB(4).EQ.O) GO TO 70
WRITE( 5,71)(I,1=1,NG)
71 FORMATiIH1/1X.8HGAGE NO./9I9)
DO 222 J=l,NN
222 WRITE(5.72XSAVE
IF(RF(I).LT.O.OJRF(IUO.O
18 CONTINUE
URITE(7) RF
C WRITE<5.100) RF
C 100 FORMAH IOX.9F12.6 )
19 CONTINUE
CALL CLOSE(7)
CALL CLOSE (8)
RETURN
END
C CHAIN «3
REAL I.ISTA,LISTS,L ISTC , L ISTD . N AMES
DIMENSION ITP(15), ITB(15),ALOSC( 15,15) ,CZ( 15,9)
DIMENSION QPB(15.1),QPPDP(2).Ri8),JJ(85)
DIMENSION QPP( 15, 1 20 ) , 1 IB ( 18 ) , LLOCI ( 15 ) .NAMESCrS) ,LJSTA( 2),
1LISTBC2),LISTC(2),LISTD(2),QPP02(2)
COMMON DTT.NOTT.IDT.NTER.Dr.II 8,LLOCI,
1 NAMES, LIST A. LI STB, L I STC , L I STD, KK, ICONTA, ICONTB
2, ICONTC. 1C ON TO, 1C, IY.MIHR, MI MI N,MIDAY, M I SEC , RA I>N( 2 ) , NC , NC , QP P,
3 I TIMES* 12 >
COMMON/BLOCKH/ITP,ITB.CZ,AR(15),8FFAC(2*).QP8
COMMON/BLOCKT/ALOS C
DATA QPPDZ(1),UPPDZ(2)/5HQPP2Z,^HVERT/
ICONTD-3
DO 203 I=1,KK
203 JJ(l)--(NDTT-KK)*!
CALL SEEK (8.LISTA)
READ (8) ITP,ALOSC,AR,C2.QPB,I 18,QPPDP,BFFAC
CALL CLOSE(8)
C READ RAIN GAUGES AND CALCULATE HYDROGRAPHS
187
-------
IF(NTER-l) 100.101,100
101 CALL DLETE(8.0PPDP, IK)
CALL ENTERCB.QPPDP )
KTM=-(NOTT-KK) '(IDT/60)
KTS=KTHH=KTD=0
CALL TMADD(KTS,KTM,KTHH,KTD,MIS£C,MIMIN, MIHR.MIDAY)
KKMMM 60-KTH)/ t IDT/ 60)
DO 71 I=1,NC
KKM-KKHM
KTH=KTMH+1
KS=1
32 KF=KS*KKM-i
DO 30 KI=KS,KF
QPP(I.KI)-QP8( I.1)»BFFAC(KTH>
IFUI.EQ.120) CO TO 31
30 CONTINUE
KS=KF+1
KKM=(3600/IDT)
IF(KTH.GE.25)KTH=KTH-24
GO TO 32
31 CONTINUE
71 CONTINUE
WRITE <8)QPP
CALL CLOSE (8)
100 CONTINUE
112 CALL HrDR>
IFUIB15) .EQ.OJGO TO 80
WRITEC5, 75 )MIHR,MI MIN, MIDA Y
75 FORMftT(//lX.22HUNDIVERTED HYDROGRAPHS/
139H POINT 0 REPRESENTS PRESENT MODEL T I ME, 2 X, 2.1 3 , U . 5 X,
232HPOINTS ARE AT 5 MINUTE INTERVALS)
DO 74 1-1, NC
LL-LLOCM1 >
WRITE(5,72 JNAMES(LL) . I
WRITE (5. 73 > ( JJ< J) ,QPP( I, J) ,J=1 ,KK)
72
73
1
74
FORMATt/
FORMATI 1
13, F7. I,
CONTINUE
/I
X.
13
X, A5
13, F
,F7.
. 15H
7.1,
1.13
HtDROGR
I3.F7.1,
•F7.1.I3
APH
13. F
,F7.
NO., I
7. 1, I
1.13.
5)
3.F 7.
F7. 1.
1,1
13.
3.F7. 1.I3.F7.
F7. 1>
1
80 CONTINUE
CALL CHAIN(2)
STOP
END
SUBROUTINE HYDR
REAL L1STA.LISTB.LISTC,LISTO,NAMES
DIMENSION OPP( 15,120).TEMPX(2) ,QPPOP(2 I,XOECAr(15)
DIMENSION ITP( 15) . IT B( 1 5 ) , RA I NO! ( 2 ), GAGE ( 9)
188
-------
DIMENSION RO(9).RF (9).CZ(15,9) , R(9),BMSTL(2)
COMMON DTT,NDTT,IDT,NTER,DT,IIB<18).LLOCI(15)
1, NAMES(48) ,LISTA(2),LISTB(2),HSTC(2),LISTD(2).
2 KK, I CO NT A. ICOMTB, I CO NIC.ICONTD. IC.IY.MIHR.MIMIN.MIDAY
3,MISEC.RAIN(2).NG,NC,OPP
COMMON/BlOCKH/ITP,ITB,CZ,A(15) ,BFFftC(24) ,QPB(15. 1)
COMMON/BLOCK!/ALOSC(15,15)
DATA BMSTL(1),BMSTL(2)/5HBMCXS,4HTlOS/
DATA RAINDTl 1) .RAINDTt 2J/5HIRA [N.4HDATA/
DATA QPPDP(l),QPPDPl2)/5HOPPXX.4HDUMP/
DATA TEMPX( 1), TEMPX(2)/5HTEMPO.4HRARY/
C SETINIT1AL RAIN GAUGE READING
CALL SEEM 7.RAINDT )
98 FORMATU8H READINGS AND T I ME , 9 F 10 .5 , 5 X , 2 I 3, I 4/>
C
C TIME
C TTT-TIME AFTER START OF STROM, NTER-CORRESPONDI NC RAINFALL
C INCREMENT
C NO
CALL SEEK(8, BMSTL)
DO 798 1=1,NC
READC8.799) ALOSCU , 9 ) , ALOSU I , 10 ), AL OS Ct 1.12 ) . AL OSC( I , 15 )
XDECAY(I)=ALOSC(I.4)*»(.25>
799 FORMATC4F5 .2)
798 CONTINUE
READ(8.796)ACCUH
796 FORMATIF8. 1)
CALL CLOSE(8)
1DTT=IDT
CALL SENSEU, I 13< 4 ) )
IF(IIBU).EQ.O)GO TO 200
URITE(S.201)RAIN, ( I , I - 1 , NG )
201 FORMATdHl/51H RAINGACE READINGS AT 5 MINUTE INTERVALS IN (IN/HR)
134H FOR INTERVAL ENDING AT TIME SHOHN, 10X,16HRAlNFALL FILE - ,
2A5.A4//5X.8HGAGE NO.,3X,9110,8X.9HHR MN DAY)
200 CONTINUE
C READ RAINGAUGES
READ(7)IR.GAGE
CALL SEEM 8. OPPOP)
READ (8)OPP
DO 129 IRI=1, IR
ACCUM=ACCUM*DT
199 READ ( 7)R
IF(IIB(4).£Q.O)GO TO 114
JSEC=-(IR-IRI)»IDTT
JMN=0
JHR=0
JDAY^O
CALL TMADO(JSEC.JMN,JHR,JDAY.M 1SEC.MIM IN.MIHR,Ml DAY)
WRITE* 5,97)(GAG£(I ), I = 1.NG>
97 FORMAT(/18X.9(5X.A5))
URITE< 5.98) (R( I),I = 1,9).JHR,JMN.JDAY
114 DO 4 NGG=1.NG
RF(NGG)=R(NGG>
RO(NGG)=R(NGG)
4 CONTINUE
DO 21 NCC= KNC
189
-------
ALOSC
6 CONTINUE
1TT=NDTT-1
DO 700 IT = 1, ITT
IS=IT*1
700 QPP/<3600.0*12.0 >»DT«2.0/(FLOAT(ITB(NCC)-l).OT)
NDTTP=1*ITP(NCC)
NDTTB=1*IT8
QPP(NCC.IT)=QPP(NCC.IT)*S»QP
9 CONTINUE
21 CONTINUE
129 CONTINUE
CALL CLOSE(8)
130 CONTINUE
CALL CLOSE(7)
NTER=2
CALL DLETE(8,BMSTL,LLL>
CALL ENTERC8.BHSTL >
DO 795 l=l,NC
WRITE(8,794)ALOSC(It9),ALOSC(I.10).ALOSC<[.12).ALOSCCf.l5>
794 FORMAT(1X,4F5.2»
795 CONTINUE
WRITE(8,793) ACCbM
793 FORMATilX, F8.1 I
CALL CLOSE(8)
RETURN
END
SUBROUTINE SLOS(R, RE X, I, DT , 1 1)
COMMON/BLOCKT/TSTLOS(15),RECOV < 15 ) ,AX( 15 ) ,DE CAK15) .XI HPERdSI .
1XIMPCFC15),ENDK<15),TBMC(15),BMC(15),STLOS(15).TSTLSP<15),
2STLOSPU5) .COEFdS ).POWER( 15) , AC LOSS ( 15)
RA=R»DT/3600.
IMPERVIOUS AREA
Z=STLOS(I)
STLOS(l)=STLOS(i)*RA
190
-------
IF(STLOS( I ) .GT.TSTLOS( I) > GO Tfl 3
X^O.O
GO TO 19
3 X=STLOSi l)-TSTLOS< I)
STLOS( l)-TSTLOS(I)
19 XILIMP=STLOS( I )-Z
XLSCOF = 1 .0/-TSTLSP( I )
4 XILPRV = STLOSP( 1>-U
SLOSX=ENOK( I)+AXU ) * ( (TBMC ( I )-BMC(I ) ) / TBMC ( I ) > » • 0 . 70
POTLSS = SLOSX»DT/3600.
ACTLSS=POTLSS
IF(ACTLSS.GT. Y) ACTLSS^Y
= lf-ACTLSS
BMC1 = BMC( I )
BMCU )=BMC ( I J + TLSPER
BASLSS=TLSPER»XIMPER(I )*TLS I MP Ml .0- XI MPER( I »
BSEXCS= BMC ( I ) - TB MC ( I )
IF( I I .EQ.01GO TO 14
IF( I.GT. 1) GO TO 11
URITE( 5. 12 )
12 FORMAU128H I RAIN ISTLOS PSUOS IINLOS PINLOS ILOSC
IDE PPOTLOS ILOSS PLOSS I E XS PEXS BASLOS BASEXS BMC
2ACCLOSS. // )
11 WRITE(5.13)I,RA,STLOS
EQUIVALENCE < D I VC < 1 ) . Z Z( 1 . 1 ) ) , ,S S( 1) )
DATA QPPDX ( 1) , QPPDXl 2>/5HQPPOI .4HVERT/
191
-------
DATA OPPDZ ( D.QPPDZm /5HQPPZZ .4 HVERT/
ICONTD=4
DO 200 1=1, KK
200 JJ( I)=-(NOTT-KK)«-1
CALL SEEK (8,L ISTB )
READ(8)ND, NC.DF,NQ,N A, NQQ, DIVC ,D IVCC.FACT.SU M.'.iUM,QMAX, FACT
CALL CLOSE (8)
CALL SEEM 8.QPPDZ)
REAO(8)OPP
CALL CLOSEC8)
DO 32 1=1, ND
IF(NQ( I) .G6.16JCO TO 32
CALL D1V( I )
CALL SENSEU4, 118(15) )
IF( IIBC15) .EQ. 1>GO TO 500
32 CONTINUE
CALL SENSE(5. IIB<6»
IF(IIB(6).EQ.O) GO TO 42
UR I IE (5. 201) HI HR, HIM IN, MI DAY
DO 42 1=1, NC
LL-LLOCI (I )
WRITE15, 72 )NAMES
-------
IFdC.EQ. 2)GO TO 303
CALL SEEK(8,QPPDX)
GO TO 302
303 K=K*1
00 44 1=1, NC
44 ZZ(1, I )-ZZ(l, I )»FACT( I )
WRITE(5»307)
307 FORMAT<13H INLET INDEX .18H LUMPED INLET VOL ,
118H LUMPED DIV VOL . 18H LUMPED RIVER VOL .
118H INLET VOL ,18H DIVERTED VOL ,14ri RIVER VOL )
DO 304 (=1 ,NC
ZZ(K,I)=ZZ(1,I >-2Z (2,1 )
zz(4,D=zz(i. n/FAcni >
ZZ(S,I)=22(2,I )/FACTlI )
22(6,1)^22(3, I )/FACT(I )
DO 11 J=l,6
SS( J)-SS(J) + ZZU, I >
11 CONTINUE
304 WRITE(5.305) I. (22( J, I>,J=1,6)
305 FORMAU5XI2, 10XF 10 . 0 , 8XF 1 0 . 0 , 8 XF 10 . 0 . 8 XF 10. 0 . 8XF 10. 0 .
18XF10.0I
WRITE (5. 306MSSO) ,J=1.6>
306 FORMAT(17H TOTAL VOLUME CF F 10 .0, 8XF 10 . 0 , 8XF 10.0 ,
18XF10.0.8XF10. 0.8XF10. 0)
308 FORMAT(17H TOTAL VOLUME MG 4 XF 8 . 2, 10XF8 . 2, 10XF8. 2 ,
110XF8.2.10XF8.2.10XF8. 2)
DO 12 1=1.6
SSU >=SS([ )» 7. 48/1 000000.0
12 CONTINUE
WRITE(5.308) (SS ,J=1.6)
211 CONTINUE
500 CALL CHAIN12)
STOP
END
SUBROUTINE OIV(ND)
C AT EACH INLET POINT THE HYDROGRAPH IS CALCULATED FOR THE AMOUNT OF
C RAIN WHICH HAS FALLEN. THIS SUBROUTINE THEN DETERMINES THE AMOUNT
C OF THE HYDROGRAPH WHICH ACTUALLY FLOWS TO THE MAIN INTERCEPTOR AT
C THAT POINT. (I.E. THE FLOOD WAVE WHICH IS THEN ROUTED THROUGH THE
C SYSTEM)ND-DIVERSION NO, DF-DIAM OF PIPE IN WHICH
C FA8RIDAM IS FIXED, H-HGHT OF F A BR1DAM , NQ- NO OF HYDROGRAPH FOR
C DIVERSION ND, NT-TIME NO.NA-NO DES IGNAT ING GE OMET RIC AL SHAPE OF
C ORIFICE. NQQ-IND£X OF POINT AT WHICH FLOW IS DIV£RTED(EG METERS),
C ALSO INDEX OF INFLOW POINT FOR ST PETER DIVERSION
C
C
REAL LISTA.LISTB.L ISTC ,L IS TO , N AMES
DIMENSION C1(17),C2(17),C3(17).C4(17).C5(17)
DIMENSION IIB( 18), L LOCI (15 I .NAMES (48 ) , L I STA ( 2 ) .L I STB( 2 ) ,
1LISTU2) ,LISTD(2)
DIMENSION DF(19),NQ( 19),NA(17) .OPPl 15 , 12 0 ) , NQO( 1 9 )
C FOR SPECIFIED DIVERSION STRUCTURES GO TO APPROPRIATE SUBROUTINE
COMMON/TY/X1.X2.X3 ,X4,X5,X6.X7.X8,X9,X10.X11,X12,C13,C141C15
COMMON DTT. NDTT.IOT.NTER.DT, MB. L LOCI. NAMES,
1L ISTA.LISTB.LISTC, LISTD.KK, I CO NT A. I C ON TB , I CON TC. ICONTD,
2IC,IY,MIHR,MIMIN, MIDAY.MISEC,RAIN(2),NG.NC,QPP
193
-------
COMMON /O I WE/ OF, NQ. NA,NQQ,C1,C2 .C3.C4.C5. QMAX<48) ,FACT< 15)
Xl^CHND)
X2-C2(NO)
X3=C3(ND)
X4=C4
QPP
-------
32 CONTINUE
IF(NP2.EO.KK»GO TO 33
NP = NP»IDTM«-IDTM
CALL TNADD(0, IMINH, IHRH.IDAYH, O.NP.0,0)
GO TO 34
33 CONTINUE
GO TO 100
4 LQ=NQ
-------
6 LQ-NQ(NO)
MO=NQQ
00 181 K-l.KK
IF(QPP
-------
GO TO 24
C3 WRITE(6,20l) NC,C4,HG
23 CONTINUE
C01 FORMAT(2X.I2,2F10.5)
AO=CRES(C4,HG>
C02 FQRMAT<2X,F10.5>
24 IFUO.LE.0.0)AO-0.00001
IF (Q.LE. 0.00) 0 = 0.000 001
IF(C2»0»»2 ./AO»»2«-HG/2 .-H + C5M 1.11,2
2 QMAX=Q
L=TOP
-------
SUBROUTINE T YPEM ND , Q, Q2 , H , HF , H A , HF A, OF ,DfA )
REAL LF
COMMON/TY/C1.C2.C3
C EUSTIS DIVERSION
C Q-APPROACH ING FLOW , Q2-D I VE S F ED F LOW, C 1 -COEF F OF DISCHARGE OF
C FABR IDAH.C2-EXPONENT OF FA3RIDAM 01 SC H AR GE , H-H&H T GF GATE.HF-HGHT
C OF FABRIDAM,EQN F- C 1 »L F »( 0 2 » • 2 / ( 3 2 . 2 * A » » 2) + H/2-MF ) » »C2
C A-AREA OF OR IF ICE, LF-L GHT OF FAfcRlOAM
C THE ABOVE APPLIES TO DIVERSION OF 90INCH PIPE.HiE SAME E QN APPLIES
C TO DIVERSION OF 60 INCH PIPE. THE LETTERA REFERS fO THE LATTER DIVER
C SI ON
C Q3-APPROACHING FLOW FOR 60INCH PIPEIASSUMED TO BE C3»Q
C
C
C
C
C WRITE( 5.400)ND.a,C 1. C2.H.HF, C3.HA.HFA, DF.DFA
C 400 FORMAT(39H TYPE4 NO. 0, C 1 , C 2 , H , HF . C 4 . H A , H F A , DF , OF A/ I 5 . 1 OF 10. 5 )
Q3=C3»Q
Q=Q»(1 .0-C3)
Kfl-0
7 LF=TOP(DF, HF)
A = CRES(3.,H)
C , HRITE15.40DLF.A
C 401 FORMAT111H TYPE4 LF,A,2F10.5)
IF(Q»»2/(32.2»A»»2)*H/2.-HF)l,l,2
2 QMAX=0
QMIN-0.
K=l
3 Q2=(QMAX+QMIN) 12.
BB=Q2*»2/( 32.2»A»»2)*H/2.-HF
IF(BB) 30.8,8
30 86=0.000001
8 F=Q-C1»LF»BB»«C2
C WRITEC 5. 402)02, F.K
C 402 FORMAT(13H TYPE4 Q 2 , F , K , 2 F 1 0 . 5 , I 5 )
CALL BISECC(F,Q2,QMAX,QMIN,K,ND.QN)
IF(K-50> 3, 4,4
I Q2 = Q
4 IF(KA)5.5.6
5 Q2S=Q2
C 60 INCH DI VERS ION
HSAVE^H
DFF=DF
DF^DFA
HF-HFA
KAM
GO TO 7
6 Q2=02+Q2S
H=HSAVE
HF^MFSAVE
DF^DFF
RETURN
END
198
-------
SUBROUTINE TYPE5CND,Q,DF,HF
l,C13,C14,H.Yl.Q5,C 15)
REAL LF
COHMON/TY/Cl,C2.C3.C<>,C5,Cb.C7 .C8.C9.C10.Cll.C12
C ST PETER KELLOGG
C NC-CACHMENT NO.Q-APPROACHING FLOW.Cl-COEFF OF DISCHARGE OF WEIR AT
C A.C2-LENGTH OF ME IR AT A,C3-CO£FF OF DISCHARGE OF ORIFICE AT 0,
C C7-AREA OF ORIFICE AT D.C4-COEFF OF DISCHARGE Of F ABR I DA M, L F-L GTH
C OF FABRIDAM CREST,C6-EXPONENT FOR FABRIDAM DISCHARGE EON
C 1ST DIVERS ION EQN
C F=Q-Q2-C1»C2»<02»» 2/l2'32.2»C3»'2»C7»»2-5>»»1.5-<4»LFMQ2'»2/2»
C 32.2»C3»»2»C7»»2-HF)»»C6
C Q-APPROACH1NG FLOW RATE.Q2-FLOW GOING TO 2ND 01 WERSION.DF-0I AM OF
C FABRIDAM P1PE.HF-HGHT OF FABRIDAM
C
C 2ND DIVERSION
C C9-COEFF OF DISCHARGE OF SIDE WEIR.C8-HGHT OF UPPER WE IR CREST.
C C10-LGTH OF LOWER SIDE WEIR CREST,Cl1-COEFF OF DISCHARGE OF
C ORIFICE,C12-HGHT OF LOWER WEIR CREST,C13-D IFF IN ELEV BETWEEN
C ORIFICE AND INTERCEPTOR.Cl*-LGHT GF UPPER WEIR CREST,AG~AREA OF
C ORIFICE. Yl-DEPTH OF WATER IN INTERCEPTOR,H-HGHT OF ORIFICE GATEt
C Q2-AS ABOVE.OS-FLOW DIVERTED TO INTERCEPTOR
C CIS-WIDTH OF ORIFICE
C DETERMINE BRACKETTED EXPRESSIONS
C WRITE(5.500)ND,Q.C1,C2,C3,C4 .C6.C7.C8.C9.C10.C11.C12.C13.C14.
C 1H.Y1,C15.DF,HF
C 500 FORMAH39H TYPES NO, Q, Cl, C 2, C 3 .C4, C6 ,C 7, C8, C9, CIO/15,1 OF 10 . 5
C 1/31H C11.C12.C13.C14.H.Y1.C15.DF.HF/9F10.5)
8A=Q»*2/(64.4»C3»»2*C7'»2)-5.
BB=BA*5.-HF
C IS THERE FLOW OVER FABRIDAM
IF(8B) 1. 1.2
1 88 = 0.
GO TO 3
2 BB=1.
3 IF(BA)4.4.9
4 BA-0.
GO TO 10
9 BA=1.
10 IF(BA*B8)5,5. 11
11 CONTINUE
LF=TOPCOF,HF)
C WRITE( 5.50DLF.8A, BB
C 501 FORMATU5H T YP E5 L F , BA ,BB. 3 F 1 0 . 5 >
QMAX-Q
QMIN^O.
K = l
7 Q2MQMAX+QMIN) /2.
BC = Q2»»2/C64.4»tC3»C7)'»2) -5.
BD = QZ»»2/(64.4 »(C3»C7)»»2) -HF
IF(BC)22,23,23
22 BC^O.OOOOOOl
23 IF(BD)24.25,25
24 BD-0.0000001
25 F = Q-Q2-C1» C2»B A*(8C) • »1.5-C4.Lf *B8'(BD)*«C6
C WRITE(5.502)02.F.K.BC.BD
C 502 FORMATU3H T YPE5 02 . F . K, 2F 1 0 .5 . J 5 .2F 1 0 . 5 »
CALL BISECUF, Q2, QMA X, QM1 N , K , NO, QN)
199
-------
IF(K-SO) 7. 8.8
C 2ND DIVERSION
5 Q2 = U
8 IFCY1-C13-H/2. ) 12, 12,13
12 HS=H/2.
GO TO 14
13 HS=Y1-C13
C IS THERE DISCHARGE OVER WEIRS
14 AG=C15»H
8A = 1.
88=1.
IF(Q2»*2/( 64.4. (Cl 1»AG> »»2 ) -Cl 2+HSJ15, 15,16
15 8A=0.
16 IF(Q2»»2/( 64.4»18, 19, 18
18 QMAX=Q2
QMIN=0.
K-l
C WRITE<5,503)HS. AC, 8A.BB
C 503 FORMATU8H TYPE5 HS . AG . BA , BB , 4 F10 .'5 )
20 05=(OMAX*OMIN> /2.
BC = Q5»»2/<64.4*(Cll»AG)»»2)-C12-»-HS
BD=Q5»»2/t 64.4»
-------
FUNCTION AREA < D, Y Y )
Y = Y Y
IF> )
20 IF-0.10)3,2,2
2 K-K+1
ON=QZ
IFIK-50) 13, 14, 14
13 IF(F)8.7,7
7 QMIN-Q2
RETURN
8 QMAX-QZ
RETURN
3 K = 50
RETURN
14 WRITE(5,15)NC
15 FORMAT127H NOT CONVERGING CACHMENT NO, 15)
RETURN
END
FUNCTION TOP(D,Y)
IF(4.0»(0»Y-Y»YJ)1,2,2
C 1 WRITE(6,3)Y,D
J FORMAT15X.5H Y,0 . 2F 10.2)
I Y-D
2 CONTINUE
TOP=SORU4 .0*( a»Y- Y»Y) >
C TOP IS TOP WIDTH,D-DIAM.Y-3EP1H
RETURN
END
FUNCTION CRES(D.H)
C AREA OF CRESCENT
C USES FUNCT ION AREA
H1MD-H) 12 .
A=AREA(D.H1)
CRES=(D/2.)»'2'3.14159-2.»AREA<0,Hi)
RETURN
END
201
-------
FUNCTION TRAPOZCN, XINTl., J)
REAL NAMES.L ISTA.L ISTB.LISTC.L IS TO
COMMON DTT.NDTT.IDT.NTtER.Df, 118(18) ,LLOC I (15). NAME S«;8>,
1LISTA<2> ,L IiTe(2).L!STC(2) ,LIS TD(2).KK, ICCNTA. ICONT8.
2ICJNTC. ICON TO, 1C, I Y, Ml HR, Ml M [ M ,M I Dfi Y , M ['.,EC , R A I .M< 2J.NG.NC.
JQPP<15.120)
SIMPSN-(QPP(J, l)*QPP(j.N)) 12 .0
N1=N-1
00 2 1=2.Nl
2 SIMPSN-SIMPSN+QPP(J,I)
TRAPOZ-SIMPSN'XINTL
RETURN
END
SUBROUTINE FABCTS(ND, IMINH. 1HRH, IDAYH,HGC.HFF.NP)
DIMENSION GATES(2).DF(19),NQ(I'M
COMMON/DIVE/OF.N3
DATA GATES( I> , GATES(2>/5HGATES.4 HE TTG/
LS^O
CALL SEEK (7.GATES)
4 READ(7,5) J.LH,LN,LD,HG.HF,hGMAX,HFMAX
IF(J.NE.ND) GO TO 2
CALL TMADD(LS,LM,LH,LD,0,-IMINH.-IHRH,-IDAYH)
IF(LD.GE.O) GO TO 3
2 HGG=HG
HFF-HF
5 FORMATC 13,5X,21.3, I4,4F!>.2)
GO TO i,
3 NP ••- LO'28d*LH»12>LM/5
CALL CLOSE (7)
RETURN
END
C »»» »»CHAIN 5
C PAL METHOD OF ROUT ING
C JR-CONSIANTS,ICONTB-OPERATION NO,ICONIC NO OF OPERATIONS
REAL LISTA.LISTB.LISTC.LISTD,NAMES
DIMENSION JR(A8.7),OMAX(48),D( 3),S(3),QPPDX(2).Q(120),TEMP(2).
10PP(15.120)
COMMON DTT,NDTT.IDT,NTER.DT,tIB(18).LLOCI(15).
1NAMES(48).LISTA<2),LISTB(2),LISTC(2) ,L IS TO(2),KK,ICONTA. ICONT8,
2ICONTC.ICONTD, IC.IY.MlHR.MlMlN.MIDAY.MlSEC.kAtNt 2).NG,NC,OPP
3, IT1MESC 12 )
COMMON/BLCK5/JR.Q, IJ,0 MA X, JJ(120)
COMMON/6KDI/O,S.AN,DF( 19),DlVC{17,5) .DI VCC<10)
DATA QPPDX(1),QP?DX(2)/5HQ?PDI.4HVERT/
DATA TEMP{ 1),TEMP( 2)/5HTEMPO,4 HFIL6/
ICONTO=5
IY = 0
DO 70 1=1,NDTT
70 JJ( I)=-(NDTT-KK»*I
80 CALL SEEM 8.LISTC)
READ(8)JR,QMAX.D.S.AN,ICONIC.DF.DIVC.DIVCC
CALL CLOSE(8)
ICONTB=0
CALL SEEM 8.QPPDX)
60 READ(8)QPP
202
-------
CALL CLOSE(8)
5 ICONTBMCONTB*!
1H=JR( ICONTB,2»
IHH=JR(ICONTB,4)
IJ-O
IE=JR( ICONTB,1 )
GO T0(1,2.3,401.IE
1 CALL ROUTB( 1H. I HH )
GO TO 4
2 CONTINUE
IHA= JR< ICONTB, 3)
CALL COHB( IH, I HA,IHH)
GO TO 4
3 CALL DIIHN(IH)
GO TO 4
40 00 42 1-1,NDTT
42 QPP( I H.I)-QPP( IH, I )+QMAX
IF(IIB(2).E0.1)GO TO 82
WR1TE(5.12)NAMES(ICONTB)
82 IF(IIB(8).EQ.O.AND.1 18(9).EQ .0 ) CO TO 41
IFUIB(8).EO.1.AMD.IJ.EQ.O.AND . JIB(9).EG .0)GO TO 41
URITE(5.8)NAME SU CONTB).IE,I CO NTB.QMAX( ICONTB).MIHR.MIMIN.MIDAY
8 FORMATI//14H HYDROGRAPH AT2X,A 5 , 5X,1 OH OPERATION 13,7H NUMBER 13,
15X,5HQMAX .F10.2,3X, 11HTIME PI. 1 ,213,14)
URITE(S.9>(JJ( 10).QPP( IHH, 10), IU=1,NOTT)
9 FORMAT(IX.I3.F7.l,I3,F7.1,I3,F7.1.I3.F7.1,I3.F7.1.I3.F7.1,
1I3.F7.1,13,F7.1,I3,F7.1,I3,F7.1,I3,F7.1.I3,F7.1J
12 FORMATU18H EXCESSIVE FLOW AT 2X ,A5, 2X ,9H»»» ».»»«» /
146H BOB CALLERY'S BASEMENT WILL BE FLOODED»»*»»»» )
41 CONTINUE
IF( ICONTB- ICONTO5.6.6
6 CONTINUE
CALL CHAIN (2)
STOP
END
SUBROUTINE ROUTB( I rt. IHH )
REAL NAMES.L 1STA.L[STB,LISTC.LISTO
DIMENSION 0(120),JR(48,7),QMAX(48),QPP(15.120)
COMMON DTT.NDTT,IDT.M'ER.OT,1I8<18),LLOCH15),
1NAMES(48),LISTA{2),LISTB(2).LISTC(2),LISTD(2),KX,ICQNTA.ICONTB.
2ICONTC.ICONTO, 1C, IY,MIHR,MIMIN , MI OhY,M1 SEC,RA IN(2),NG,NC,QPP
COMMON/BLCK5/JR.Q,IJ.QMAX
NRCHS^JR( ICONTB.7)
LAG= JR( ICONTB,6)
NSTRL= JRU CONTB.5)
DO 56 1=1,NDTT
IF(QPP
-------
C 40 FORMAK8I6)
ANS=NSTRL
C PERFORM AVERAGING
00 20 IZ=1.NRCHS
DO 7 IX =l.NDTT
J=IX-NSTRL 12
JX=J+NSTRL-1
0(1X1=0.0
1F(J-l)2.3.3
3 IF . t 'j . 0) GO TO 6
00 50 L-l,NDTT
50 QPP( IH,L ) = 0(L)
GO TO 33
6 DO 51 L=2,NDTT
51 QPPl IH.L-1)-Qt L)
QPP(IH.NDTT)-Q(NDTT)
33 CONTINUE
20 CONTINUE
DO 32 1=1,NDTT
32 Q(1) = QPP(tH. I )
41 FORMATt10F10.2 )
C PERFORM LAGGING
ICN = 0
I F.< LAG) 3 8, 2 1,2 1
2 1 IF(MOD(NSTRL,2).60.0)ICN=NRCMS/2
GO TO 10
38 CONTINUE
NRCHS=1
30 LAG=IABS(LAC)»NRCHS+ICN
IF(LAG.GE. 0)GO TO 44
LL=NDTT*LAG
DO 42 IN=1.LL
INLAG= IN-LAG
42 OPP( IHH. IN)-Q( INLAC)
LL=LL*1
43 QPP( IHH.IN)=Q(NDTTI
RETURN
44 LL=LAG*1
DO 9 IN=LL,NDTT
INLAG=IN-LAG
9 QPP( IHH. IN)=Q( INLAG)
DO 10 IN=1.LAC
10 QPP( IHH, IN)=Q( I)
RETURN
END
204
-------
SUBROUTINE COMB ( I H , I Hfl . I HH )
REAL NAMES.L ISTA.LISTB.LISTC.L.ISTD
DIMENSION JRI48.7)
COMMON OTT.NDrT, IDT, NT ER. OF. 118(18). LLOCK15).
1 NAMES ( 48>.LISTA(2).LlSTBl2),LlSIC<2).LlSrD(2).KK,lCONTA.lC3NTB.
2 1 CONK. I CO NT 0. I C, 1 Y , M IHR , MI M I N , M 1 DA Y , M I SEC, R A I N< Z ) , NG . NC ,
3QPP< 15.120)
COMMON/BLCK5/JR, 0(120) , IJ , QMAX148)
00 1 1=1 .NOTT
1 QPP( IHH, I) -QPPUH, I ) +QFP( I HA , I )
RETURN
END
SUBROUTINE DIVIN(IH)
REAL NAMESrLlSTA, LISTB.LlSrC.LlSTO
DIME MS ION JRU8,7)
COMMON /BLCK5/JR, 0(1 20) , U.QMAXC^S)
COMMON DTr.NDTUlDT.NTER.DT.IIBUaj.LLOCKl1;),
INAMES< 48),LlSTA(2),LlST8{2).UISfC(
VOLA=TRAPOZ(NDTT,DTT. IH)
NP2 - 0
NPT^NDTT
IOTM- IDT/60
LSX=-(NDTT-KK-1)»IDT
IS=0
IMINH- MI MI N
IHRH^MIHR
IDAYH-MIDAY
CALL T MA DO (IS. IMINH. IHRH.IOAYH.LSX. 0,0.0)
4 CALL FABCTS(ND, IMI NH, IHRH. 1DAY H. HCG.HFF. NP)
IF(NPT .LE.NP) CO TO 10
NP2=NP
GO TO 11
10 NP1=NP2*1
11 DO 22 K-NP 1.NP2
GO TO (1 . 1,1, 1.1,1. 7) , IA
I CALL TYPE1 ( NO,DF( NO) .HFF.HCG, OV'P( Iri.K ) .02.1 )
QPP( IH.K) =02
GO TO 22
7 IF(HGG.LE.O. 1 ) GO TO 22
CALL VALY(0( IB) ,Y, S( IB),QPIJ( IH ,K).flN)
CALL TYPEMND,Y,HGG,D1VC(NL>,1) ,DIVC(ND,2).QPP( l.H.K))
22 CONTINUE
IF(NP2.EO. NDTT) GO TO 13
205
-------
NPT=NDTT-NP2
NP=NP»IDTM«-IDTM
CALL TMADD(IS, IMINH, IHRH,[OAYH,0,N?,0,0>
GO TO 14
13 CONTINUE
IF(IIB(7).EQ.O)CO TO 15
VOLB=TRAPQZ(NDTT.OTT, IH)
VOLC-VOLA-VOLB
WRITE(5.16>VOLA,VDLB,VOLC
16 FORMAT124H VOLUME BEFORE 0 IVERSI ON,F10.0/
122H VOLUME TO INTERCEPTOR,F10.0/
216H VOLUME TO RIVER,F10.0)
15 CONTINUE
RETURN
END
SUBROUTINE V AL Y( D. Y . SO . Q'J, AN )
C GIVEN D-D1 AM.SO-BED SLOPE,QO-D ISCH.AND-MANNINGS N, ROUTINE FINDS
C NORMAL OEPTH-Y FOR CIRCULAR SE CIIOiM, US IN C BISECTION. AREA FUNCTION
C USED.
K=0
YMIN=0.0
YMAX=D
8 Y = (-YMIN*YMAX)/2.0
K = K + 1
10 FORMAH21H NORMAL DEPTH USED ft S. F10 . 3 , 34 H DISCHARGE IS.GT.SECTI
ION CARRY >
IF
-------
BIBLIOGRAPHIC:
Minneapolis-Saint Paul Sanitary District, Dispatching Sys-
tem For Control of Combined Sewer Losses, Final Report
FWQA Grant No. 11020 FAQ, Maich, 1971.
ABSTRACT
Impressive reductions in combined sewer overflow pollu-
tion of the Mississippi River in Minneapolis and Saint Paul
have been effected by a regulator control system. Working
entirely within the limits of the existing interceptor sewer
system, and with relatively minor modifications to selected
major combined sewer regulators, incidence of overflow was
reduced by 66% and duration of overflow by 88% during
most of a rainfall season. Computer simulation techniques
using actual rainfall data indicate that the amount of
overflow volume reduction achieved is the equivalent of a
$200 million separation project. The efficiency of collection
was improved by about 20% at controlled regulators. The
reduction in volume of combined overflow to the river is
estimated to be between 35% and 70% during the runoff
season. The unmodified combined sewer system captured
about 65% of the urban runoff. Where modified, the system
captured about 77% of the urban runoff.
ACCESSION NO.
KEY WORDS:
Combined Sewer
Regulator
Interceptor
Overflow
Control and Monitoring
Variable Diversion
Real Time Operation
Urban Hydrology
Stormwater Runoff
BIBLIOGRAPHIC:
Minneapolis-Saint Paul Sanitary District, Dispatching Sys-
tem For Control of Combined Sewer Losses, Final Report
FWQA Grant No. 1 1020 FAQ, March, 1971.
ABSTRACT
Impressive reductions in combined sewer overflow pollu-
tion of the Mississippi River in Minneapolis and Saint Paul
have been effected by a regulator control system. Working
entirely within the limits of the existing interceptor sewer
system, and with relatively minor modifications to selected
major combined sewer regulators, incidence of overflow was
reduced by 66% and duration of overflow by 88% during
most of a rainfall season. Computer simulation techniques
using actual rainfall data indicate that the amount of
overflow volume reduction achieved is the equivalent of a
$200 million separation project. The efficiency of collection
was improved by about 20% at controlled regulators. The
reduction in volume of combined overflow to the river is
estimated to be between 35% and 70% during the runoff
season. The unmodified combined sewer system captured
about 65% of the urban runoff. Where modified, the system
captured about 77% of the urban runoff.
ACCESSION NO.
KEY WORDS:
Combined Sewer
Regulator
Interceptor
Overflow
Control and Monitoring
Variable Diversion
Real Time Operation
Urban Hydrology
Stormwater Runoff
BIBLIOGRAPHIC:
Minneapolis-Saint Paul Sanitary District, Dispatching Sys-
tem For Control of Combined Sewer Losses, Final Report
FWQA Grant No. 11020 FAQ, March, 1971.
ABSTRACT
Impressive reductions in combined sewer overflow pollu-
tion of the Mississippi River in Minneapolis and Saint Paul
have been effected by a regulator control system. Working
entirely within the liinils of the existing interceptor sewer
system, and with relatively minor modifications to selected
major combined sewer regulators, incidence of overflow was
reduced by 66% and duration of overflow by 88% during
most of a rainfall season. Computer simulation techniques
using actual rainfall data indicate that the amount of
overflow volume reduction achieved is the equivalent of a
$200 million separation project. The efficiency of collection
was improved by about 20% at controlled regulators. The
reduction in volume of combined overflow to the river is
estimated to be between 35% and 70% during the runoff
season. The unmodified combined sewer system captured
about 65% of the urban runoff. Where modified, the system
captured about 77% of the urban runoff.
ACCESSION NO.
KEY WORDS:
Combined Sewer
Regulator
Interceptor
Overflow
Control and Monitoring
Variable Diversion
Real Time Operation
Urban Hydrology
Stormwater Runoff
207
-------
A mathematical model has been prepared that will, with
rain gage data as input, perform rainfall runoff analysis,
diversion of combined sewer runoff hydrographs at regu-
lators, and routing of diverted hydrographs through the
interceptor system. This model will assist in operation of the
system to retain combined sewer flows and utilize the
maximum flow capacity of the existing interceptor sewer
system. Part II of this report describes the model.
The 1.75 million dollar project includes a computer-based
data acquisition and control system that permits remote
control of modified combined sewage regulators. Data from
rain gages, regulator control devices, trunk sewers and
interceptors, and river quality monitors provide real-time
operating information. Time varient quality data from key
locations in the sewer system were obtained by automated
analysis of numerous hourly samples.
Work on the "Dispatching System for Control of Com-
bined Sewer Losses" began in 1966, and the system has been
operated since April, 1969.
This report was submitted in fulfillment of Demon-
stration Grant 11020 FAQ, supported in part by the
Environmental Protection Agency, Water Quality Office.
A mathematical model has been prepared that will, with
rain gage data as input, perform rainfall runoff analysis,
diversion of combined sewer runoff hydrographs at regu-
lators, and routing of diverted hydrographs through the
interceptor system. This model will assist in operation of the
system to retain combined sewer flows and utilize the
maximum flow capacity of the existing interceptor sewer
system. Part II of this report describes the model.
The 1.75 million dollar project includes a computer-based
data acquisition and control system that permits remote
control of modified combined sewage regulators. Data from
rain gages, regulator control devices, trunk sewers and
interceptors, and river quality monitors provide real-time
operating information. Time varient quality data from key
locations in the sewer system were obtained by automated
analysis of numerous hourly samples.
Work on the "Dispatching System for Control of Com-
bined Sewer Losses" began in 1966, and the system has been
operated since April, 1969.
This report was submitted in fulfillment of Demon-
stration Grant 11020 FAQ, supported in part by the
Environmental Protection Agency, Water Quality Office.
A mathematical model has been prepared Ihat will, with
rain gage data as input, perform rainfall runoff analysis,
diversion of combined sewer runoff hydrographs at regu-
lators, and routing of diverted hydrographs through the
interceptor system. This model will assist in operation of the
system to retain combined sewer flows and utilize the
maximum flow capacity of the existing interceptor sewer
system. Part II of this report describes the model.
The 1.75 million dollar project includes a computer-based
data acquisition and control system that permits remote
control of modified combined sewage regulators. Data from
rain gages, regulator control devices, trunk sewers and
interceptors, and river quality monitors provide real-time
operating information.- Time varient quality data from key
locations in the sewer system were obtained by automated
analysis of numerous hourly samples.
Work on the "Dispatching System for Control of Com-
bined Sewer Losses" began in 1966, and the system has been
operated since April, 1969.
This report was submitted in fulfillment of Demon-
stration Grant 11020 FAQ, supported in part by the
Environmental Protection Agency, Water Quality Office.
208
-------
Accession Number
w
Subject Field &. Group
04A,056
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Minneapolis-Saint Paul Sanitary District, St. Paul, Minnesota
(NOW METROPOLITAN SEWER BOARD)
Title
DISPATCHING SYSTEM FOR CONTROL OF COMBINED SEWER LOSSES
10
Authors)
Robert L. Gallery
16
Project Designation
EPA/WQO Demonstration Grant 11020 FAQ
2] Note
22
Citation
Water Pollution Control Research Series 11020 FAQ 03/71, 206 pp,
81 fig, 30 tab, 36 ref
23
Descriptors (Starred First)
*Storm Runoff, *Sewers, *0verflow, *Pollution Abatement, *Mathematical
Models, Diversion Structures, Routing, Sewerage, Sanitary Engineering,
Water Management (APPLIED), Simulation Analysis, Model Studies, Water
Pollution Control
25
Identifiers (Starred First)
*Combined Sewer, *Regulator, *lnterceptor, *Control and Monitoring,
Urban Hydrology, Computer Model, Sewage Sampling and Analysis
27
Abstract
Results of initial operation of a computer-based data acquisition and control system
for a major urban combined sewer system are presented.
Impressive reductions in combined sewer overflow pollution of the Mississippi River
in Minneapolis and St. Paul have been effected. Working entirely within the limits of
the existing interceptor sewer system, and with relatively minor modifications to selected
major combined sewer regulators, incidence of overflow was reduced by 66% and duration of
overflow by 88$ during most of a rainfall season. Computer simulation techniques using
actual rainfall data indicate that the amount of overflow volume reduction achieved is the
equivalent of a $200 million separation project. The reduction in volume of combined
overflow to the river is estimated to be between 35% and 70% during the runoff season.
The combined sewer system captured about 77% of the urban runoff at the modified locations.
A mathematical model has been developed ttiat will, with rain gage data as input,
perform runoff analysis, diversion of combined sewer runoff hydrographs at regulators,
and routing of diverted hydrographs through the interceptor system.
Data from rain gages, regulator control devices, trunk sewers and interceptors, and
river quality monitors provide real-time operating information. Time variant quality
data from key locations in the sewer system have been acquired from analysis of numerous
hourly samples.
L. Gallery
Watermation, Inc.,2304 University Ave.,St. Paul 55114
WR:I02 (REV. JULY 1869)
WRSIC
SEND WITH COPY OF DOCUMENT. TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
"~ WASHINGTON. D. C. 20240
209
• GPO: I 970-360-930
-------
Continued from inside front cover....
11022 --- 08/67 Phase I - Feasibility of a Periodic Flushing System
for Combined Sewer Cleaning
11023 --- 09/C7 Demonstrate Feasibility of tfie Use of Ultrasonic
Filtration in Treating the Overflov/s from Combined
and/or Storm Sewers
11020 --- 12/67 Problems of Combined Sewer Facilities and Overflows,
19C7, (WP-20-11)
11023 — 05/68 Feasibility of a Stabilization-Retention Basin in Lake
Erie at Cleveland, Ohio
11031 — - 00/68 The Beneficial Use of Storm l/ater
11030 DNS 01/65 Water Pollution Aspects of Urban Runoff, (UP-20-15)
11020 Dili OG/69 Improved Sealants for Infiltration Control, (UP-20-13)
11020 DES 06/69 Selected Urban Stonn Water Runoff Abstracts, (KP-20-21)
11020 — 06/6S Sev/er Infiltration Reduction by Zone Pumping, (DAST-9)
11020 EXV 07/69 Strainer/Filter Treatment of Combined Sewer Overflows,
(WP-20-16)
11020 DIG 08/69 Polymers for Sewer Flow Control, (WP-20-22)
11023 DPI 08/69 Rapid-Flow Filter for Sewer Overflows
11020 DGZ 10/69 Design of a Combined Sewer Fluidic Regulator, (DAST-13)
11020 EKO 10/69 Combined' Sewer Separation Using Pressure Sewers, (ORD-4)
11020 — 10/69 Crazed Resin Filtration of Combined Sewer Overflows, (DAST-4)
11024 FKM 11/69 Storm Pollution and Abatement from Combined Sewer Overflows-
Bucyrus, Ohio, (DAST-32)
11020 DWF 12/69 Control of Pollution by Undenvater Storage
11000 — 01/70 Storm and Combined Sewer Demonstration Projects -
January 1970
11020 FKI 01/70 Dissolved Air Flotation Treatment of Combined Sewer
Overflows, (WP-20-17)
11024 DOK 02/70 Proposed Combined Sewer Control by Electrode Potential
11023 FDD 03/70 Rotary Vibratory Fine Screening of Combined Sewer
Overflows, (DAST-5)
11024 DMS 05/70 Engineering Investigation of Sewer Overflow Problem -
Roanoke, Virginia
11023 EVO 06/70 Microstraining and Disinfection of Combined Sewer
Overflows
11024 — 06/70 Combined Sewer Overflow Abatement Technology
------- |