United States
               Environmental Protection
               Agency
               Office of Air
               Land and Water Use
               Washington DC 20460
EPA-600/9-78-035
October 1978
               Research and Development

-------
                                        EPA-600/ 9-78-035
                                        October 1978
      URBAN RUNOFF CONTROL PLANNING
            M.B. McPherson
Director, ASCE Urban Water Resources
          Research Program
        Marblehead, Massachusetts
            Project Officer

            Harry C. Torno
  Office of Air, Land, and Water Use
  Office of Research and Development
 U.S. Environmental Protection Agency
        Washington, D.C. 20460
 OFFICE OF AIR, LAND, AND V1ATER USE
 OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
       WASHINGTON, D.C.  20460

-------
                                   DISCLAIMER


     This report has been reviewed by the Office of Air, Land and Water Use,
Office of Research and Development, U.S. Environmental Protection Agency,
and approved for publication.  Approval does not signify that the contents
necessarily reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or coninercial products constitute
endorsement or recommendation for use.

-------
                                  FOREWORD
     A major function of the Research and Development programs of the
Environmental Protection Agency is to effectively and expeditiously transfer,
to the xoser community, technology developed by those programs.  A corollary
function is to provide for the continuing exchange of information and ideas
between EPA and users, and between the users themselves.  In this latter
spirit, the Office of Air, Land, and Water Use publishes work which, while
not originally supported by EPA, is of sufficient interest and merit to be
useful to engineers and planners working on EPA-supported programs.

     This report, by the ASCE Urban Water Resources Research Council under the
sponsorship of the National Science Foundation, provides insight into the urban
runoff problem, and ashould prove invaluable to those engaged in the planning
for or design of urban runoff pollution abatement or control projects.
                                                Courtney Riordan
                                      Acting eputy Assistant Administrator
                                        Office of Air, Land, and Water Use
                                   111

-------
                                 ABSTRACT
      Section 208 of Public Law 92-500  (Federal Water Pollution Control Act
of 1972) encourages areawide planning for water pollution abatement manage-
ment, including urban runoff considerations where applicable.  Areawide
studies are under way or planned in just about every metropolitan area.
Deadlines for initial areawide reports are not far off, and it is expected
that many of the agencies preparing reports are presently resolving their
projected activities beyond the current first planning phase.  This report
has been prepared to assist agencies and their agents that are participants
in the preparation of areawide plans, from the standpoint of major urban
runoff technical issues in long-range planning.  Emphasized is the importance
of conjunctive consideration of urban runoff quantity and quality and the
need to development a factual basis that will support expected reliability
of performance of proposed actions and programs.  While not intended as a
handbook for urban runoff control planning, this report delves into sane
important technical issues that are often slighted or poorly handled, such
as the utilization of simulation.  Recognizing that the ultimate test of
any plan lies in its implementation, topics are viewed from the perspective
and experience of the local government level where implementation takes place.
                                     IV

-------
                                 TABLE OF CONTENTS
SECTION 1 - INTRODUCTION AND SUMMARY	1
          Synopsis	  1
          Overview	  1
          Problem	2
          Purpose  	  2
          Delivery	„  3
          Intentions and Expectations  	  .  	  3
          Summary, Section 2 - Selected Planning  Considerations   	  ...  5
          Summary, Section 3 - Incentives for Comprehensive Planning ......  5
          Summary, Section 4 - Some Flow Management Considerations  	  .  6
          Summary, Section 5 - Utilizing Simulation  	  7
          Summary, Section 6 - Urban Runoff Models  	  ...  8
          Summary, Section 7 - Some Modeling Considerations  	  8
          Summary, Section 8 - References  	  8
          Acknowledgment	  8
          Principal Information Listings 	  9

SECTION 2 - SELECTED PLANNING CONSIDERATIONS 	  11
          Definition	11
          An Example of Local Government Planning Elements 	  11
          An Illustration of Local Government Planning 	  ...12
          A Broader Perspective  	  ............13
               Figure 1 - Block Diagram of Urban  Runoff Control
                          Master Planning Concept  	  15
          Marshalling for a Plan	....».,	14
          An Emphasis on Flooding and Drainage 	  ....18
          Some Metropolitan Examples	...19
          Two-Direction Communication and Time Horizons  	  21
               Table 1  - Summary Guidelines for  Plan Components
                          of Continuing Planning  Process 	  22

SECTION 3 - INCENTIVES FOR COMPREHENSIVE PLANNING	23
          Synopsis	23
          Land-Use Controls  	  23
          Disparities in Economic Evaluations  	  • 	  25
          Water Reuse	26
               Figure 2 - Uses for Reclaimed Water	27
          Extraneous Sewer Flows 	  28
          Erosion and Sedimentation	29
          A Metropolitan Issue 	 	  29
          Conclusions	30

SECTION 4 - SOME FLOW MANAGEMENT CONSIDERATIONS	31
          Synopsis 	  ......  	  31
          Drainage Versus Flood Control   	  31
          Flood Aspects	32
          Conjuctive Planning  	  33

-------
                             TABLE  OF  CONTENTS  (Continued)
           Detention Storage   	  34
           Other Measures  .  .  . ,	35
           Miscellaneous   ..........  	  .....  37

 SECTION  5  - UTILIZING  SIMULATION  	 ^33
           Synopsis	38
           Why  Simulation?	39
           Background	40
           Performance  Reliability   	  42
               Figure  3 - Subjective Estimate of Increased Reliability
                          and Associated Costs for More Thorough Analysis  .  .  43
           Testing Simulation Models  	 .  .  44
               Table 2 - Physical Characteristics of Representative
                         Catchments Being Gaged in Philadelphia  	  45
           Long-Term Precipitation Record as a Reference  	  46
               Figure  4 - Sample of Yarnell Charts	48
               Figure  5 - Relation for Various Durations 	  48
           Developing Frequency Rationales for Levels of Protection ......  49
           Simulating a Range of Storms and System Loadings 	  49

 SECTION  6  - URBAN RUNOFF MODELS	51
           Contents	51
           Role of Simulation in Planning	51
           Some Reservations  	   .....  52
           Categories of Model Applications 	 ....  53
           Models for Planning Applications 	  54
               Figure  6 - "STORM" Simplified Logic Diagram 	  55
           Models for Analysis/Design Applications  	  57
           Model Comparisons	57
           Receiving Water Modeling 	  58

 SECTION  7  - SOME MODELING CONSIDERATIONS 	  61
           Pitfalls	61
           Land-Use Data	61
           Flood Plain Mapping	63
           Storage Manipulation Via Automatic Control	64
           Ralngage Networks  	  66
           EPA Overview	68

 SECTION 8  - REFERENCES	69

ADDENDUM 1 - METROPOLITAN INVENTORIES  	  88

ADDENDUM 2 - THE DESIGN STORM CONCEPT  	 100

ADDENDUM 3 - NOMOGRAPHS FOR TEN-MINUTE UNIT HYDROGRAPHS FOR
             SMALL WATERSHEDS	119

ADDENDUM 4 - RESEARCH ON THE DESIGN STORM CONCEPT	153
                                        -vi-

-------
                                     SECTION 1

                             INTRODUCTION AND SUMJ&RY

Synopsis

          Section 208 of Public Law 92-500 (Federal Water Pollution Control Act of
1972) encourages areawide planning for water pollution abatement management,
including urban runoff considerations where applicable.  Areawide studies are under
way or planned in just about every metropolitan area.  Deadlines for initial areawide
reports are not far off, and it is expected that many of the agencies preparing
reports are presently resolving their projected activities beyond the current first
planning phase.  This report has been prepared to assist agencies and their agents
that are participants in the preparation of areawide plans, from the standpoint of
major urban runoff technical issues in long-range planning.  Emphasized is the
importance of conjunctive consideration of urban runoff quantity and quality and the
need to develop a factual basis that will support expected reliability of performance
of proposed actions and programs.  While not intended as a handbook for urban runoff
control planning, this report delves into some important technical issues that are
often slighted or poorly handled, such as the utilization of simulation.  Recognizing
that the ultimate test of any plan lies in its implementation, topics are viewed from
the perspective and experience of the local government level where implementation
takes place.  On-site elaboration of the principles presented in this report, and
their adaptations to local conditions, will be made in seminars, workshops and
conferences to be led by the author in a number of metropolitan areas over 1977-19-79.
During that period this report will be updated and supplemented in a series of
subsequent releases.

Overview

          Perhaps never before in our history has there been as rapid a change in the
public's attitudes towards its surroundings as has occurred in the past few years.
This haa beeu fueled by unprecedented world-wide population growth, a resultant
intensified international communication of pollutants as well as information, and a
growing appreciation of the finite limitations of the resources of our planet.
Manifestations are concerns over environmental protection, more rational use of land,
better husbandry of energy and other resources and greater relief from natural disasters.
Because three out of four Americans live in urban settlements that occupy less than
one-fortieth of our land area, it is equitable to say that the winds of change are
predominantly urban in character, or at least that urban dwellers will be the larger
number affected.

          The growth impact of projected areal enlargement of urban areas'1'^'  on
present and planned urban water resource facilities  is almost too stunning a reality
to be comprehended fully at this time.  Even if expected urban growth is checked by a
renaissance of our central cities, the required reconstruction will still be monumental.

          On the basis of past performance, it has been asserted that while  the
construction phase of water resource project implementation takes somewhere  around
1 to 5 years for smaller projects and around 5 to 15 years for larger projects, because
of protracted delays in conflict resolution the true lead-time necessary for
accomplishing the objectives sought has been on the  order of  20 to 25 years,
    references are listed in Section 8.
                                        - 1 -

-------
 particularly for larger projects,  including certain  large  urban water  supply
 developments.(3) Anticipated  geometric  increases  in the impacts on urban water
 resources suggest that  total project  development  time must be  substantially compressed.
 Thus,  means must be found  for  accelerating conflict  resolution, the element most likely
 to cause protracted delays, despite the  fact that  new works will increase steadily  in
 hydrologic complexity because  of a growing interdependence among project components.
 That  is, local  governments  are confronted  with public demands  for  instant solutions
 while  simultaneously there  is  more direct  involvement by the public in decision-making
 as the problems and their  solutions become increasingly more complex.  On top of all
 this  is the shifting of targets as national and State laws and regulations are
 constantly revised  and  augmented in an upward  spiraling of stiffened requirements.
 There  are numerous  instances where elaborate plans have had to be  abandoned or
 drastically altered because the "rules of  the  game"  had suddenly become more demanding,

 Problem

           Considerable  planning activity is being  focused  on the improved management
 of all types of water pollution, including that from urban runoff.  Section 208 of
 Public Law 92-500 (Federal Water Pollution Control Act Amendments  of 1972) encourages
 areawide management planning in areas which, as a  result of urban-industrial
 concentrations  or other factors, have substantial  water quality control problems.
 The 208 studies are resulting  in a many-fold increase in planning  activities on urban
 runoff pollution management, with  studies  under way  or planned for just about every
 metropolis in the nation.  While an opportunity is presented through these studies
 "to plan and manage a comprehensive program based  on integrated planning and control
 over such activities  as municipal  and industrial wastewater, storm and combined sewer
 runoff,  nonpoint source pollutants, and  land use as  it relates to water quality,"^'
 evidence is  accumulating that,  in  order to  meet initial reporting  requirements, many
 of these plans  may  favor short-term solutions  over the meeting of  long-range goals.

           Because knowledge on  quantity and quality  of stormflows  is limited, long-
 range  planning  should be accompanied by programs for the acquisition of local field
 data,  and its analysis  by such  means as simulation techniques.  The 208 study schedule
 does not provide sufficient time to develop a  suitable factual base where it did not
 already exist,  and  as a result  in  many metropolitan  areas  problem  identification,
 and definitive  planning are expected to continue well beyond current initial reporting
 activities.t5'

 Purpose

           The major purpose of  this report  is  to encourage long-range comprehensive
 planning for urban  runoff quality  and quantity  management, as  a supplement to
 applicable current Section 208  planning.

          Another purpose is to stress the  importance of planning not only for
control  of urban  runoff  quality but also quantity, a major economic and environmental
consideration that might be inadvertently slighted in the new national emphasis on
quality management.

          Only a very few instances have been  identified where local government
planning  that was started several years ago  for urban runoff control has conjunctively
addressed both quality and quantity management  aspects.   These and other but narrower
 leading-edge examples are cited whenever methods or procedures that have been used
might be adapted  to advantage elsewhere.
                                        - 2 -

-------
          This report is not a handbook for urban runoff control planning,  and it
is not a definitive exposition of the planning process.  It does delve into some
important technical issues that are often slighted or poorly handled,  such as the
utilization of simulation.

          A synthesis of local government perspectives on planning and associated
problems is presented with the intention of facilitating metropolitan intergovernmental
cooperation and coordination.  Because an intended audience is Section 208 planning
agencies, a central purpose is to bring to their attention some potential technical
impediments that might frustrate effectuation of their proposals.  There are cautions
also, directed at local governments, that are intended to be reminders of some
technical pitfalls that should be avoided in their dealings with areawide plans.
In short, we hope that the uninvited intervention of an outside referee will be of
some assistance in improving the effectiveness of the participants in a very complex
process.

Delivery

          This report is being distributed to area-wide planning agencies and
supportive entities in all metropolitan areas.  On-site elaboration of the principles
presented herein, and their adaptation to local conditions, will be provided in
seminars, workshops and conferences conducted in a number of metropolitan areas over
1977-1979,  As the delivery phase progresses, special topics not treated adequately
or sufficiently elsewhere that are found to need synthesis and summarization will be
reviewed in subsequent Addenda to this report.  The two Addenda included with this
report are examples of what is intended.  Only a fraction of the metropolitan areas
can be assisted via site visits made for the purpose of elaborating the report's
contents, but through the distribution of post-report Addenda to all metropolitan
planning agencies a wider audience will be reached.  Also, the Addenda will serve as
a vehicle for updating the content of the present report.

Intentions and Expectations

          Just about all broad-scale, leading-edge, local government plans for urban
runoff control we have encountered are called "master plans," and although this term
appears to be abhorred by many metropolitan planning agencies we do not propose a
wholesale semantic reform.  This difference in viewpoint probably springs from the
realistic need for areawide planning agencies to preserve the concept of a "living
document" in principle and in fact, whereas local governments cannot afford this
luxury because any planning dealing with capital improvements must lead to acts of
implementation, and once construction starts  the die has been cast.

          The conventional local government master plan has generally dealt with  only
one aspect of water resources such as water supply, or wastewater^ treatment, or
flood control, or pollution abatement.  We favor and encourage the emerging type of
comprehensive planning that provides integrated consideration of flooding protection,
pollution abatement including erosion control, and water conservation.  Even some of
the advanced examples cited in this report have integrative deficiencies.  There are
simply not many local government plans under preparation that are truly comprehensive
because the impact of changing policies and objectives has not been factored into
many such plans.  Moreover, because comprehensive local government urban water
resource planning is a relatively new concept, very few of the broader plans have
been carried to the threshold of Implementation.
                                       - 3 -

-------
           We have no illusions about the possibilities for wholesale change.   Perhaps
 the most we can hope for in the near term, in a majority of metropolitan areas,  is  a
 conjunctive planning approach, at least with regard to the quantity and quality
 aspects of urban runoff.  Comprehensive planning faces a number of obstacles,  including
 fractionalized authority of administrative agencies and balkanized local governments
 in metropolitan areas,(6) specialist approaches taken individually on each water
 service component, and uneven advances in knowledge on the various components,   The
 term "water management," widely used only in recent years, has  not yet acquired  a
 specific meaning.  Although rational control of water is its objective,  there  is no
 common understanding of needed policies and institutional arrangements for achieving
 rational control.  On the other hand,  metropolitan area planning is being conducted
 formally almost nationwide,  although the planning function is normally advisory,  only;
 and in terms of comprehensive water resource development, the planning,  implementation
 and operation of works  is usually fragmented in both the central cities  and in their
 metropolitan districts.  Even water and wastewater services are commonly fragmented in
 metropolitan districts  (e.g.,  our largest metropolis is served  by some 400 separately
 managed water agencies).  A major cause of administrative fragmentation is the
 existence of often gross dissimilarities in geographic areas:   viz.,  differences  in
 hydrologic,  service and revenue areas,  and in political jurisdictions.   One of the
 consequences is that data collection and its analysis  and related research and
 development  is currently pursued independently by individual  local government
 departments  and metropolitan special districts,  with only limited collaboration  on  a
 national or  even metropolitan scale.  Therefore,  early coordination by local governments
 of  data collection and  its analysis is  strongly advised.

           "Although metropolitan-wide planning and coordination is necessary for
 successful urban water  resource management,  the key to the implementation of such
 planning generally lies at the local government level  and at  the State  level in  terms
 of  what the  State enables  or requires local  government to do."(7)   (Emphasis in
 original).                                   '

           While urban water  resource implementation agencies  may do some  long-range
 or  strategic planning,  their basic  functions  as mandated  by authorized responsibilities
 are management activities  that  emphasize  specific  local demands,  a narrow range of
 alternatives,  detailed  rules  for operation and  maintenance, and  the inclusion of
 measures  such  as  service charges and service  extension policies.^)

           "There  have been some attempts  to  define  an  integration  of  regional planning
with local guidance  and  management, but  there have  been few,  if  any,  successful
attempts  to  apply  it."^''

          Mare  comprehensive management approaches would  integrate  or give due
consideration  to all aspects of water in a metropolis,  together with  related
environmental,  energy and other relevant considerations.  The need  for a  comprehensive
overview becomes more crucial as the complexity of urban  areas  increases, caused by
such things  as  population migration and growth, competition over available energy
supplies, rising expectations of urban dwellers in  the face of  inflation  of local
government costs, and greater public awareness of environmental  issues and enlarging
demands  for  public participation.  A means for assuring a comprehensive overview  is
the conduct of  metropolitan water balances, or  inventories, described in Addendum 1
of  this  report.
                                       - 4 -

-------
Summary. Section 2 - Selected Planning Considerations

          Based on experiences with a leading-edge master plan for combined sewer
overflow pollution abatement, a set of planning elements is presented that should
serve as a useful check list.

          The same master planning effort is reviewed in terms of project financing
scenarios for meeting a variety of possible abatement goals.  Illustrated by this
example plan is the urgency of incorporating sufficient flexibility in a plan to
satisfy future shifts in regulationsi_a_s_impleinentation proceeds.

          Six discrete steps can be identified in the typical planning exercise:
problem definition, establishment of goals and identification of constraints; data
assembly and problem redefinition; formulation of alternatives; analysis of
alternatives; identification of more promising alternatives; and evaluation of the
trade-offs among selected alternatives.  These fundamental steps are elaborated in a
detailed block diagram of the urban runoff control master planning concept that
incorporates all features of several less comprehensive depictions.  Noted
parenthetically is that a local agency can become so obsessed with long-range planning
that more immediate, partially mitigative solutions can be overlooked or inordinately
deferred.

          Next, an example is given of a master plan for combined sewer overflow
pollution abatement that was launched from the metropolitan level.  The careful
procedures that were followed in marshalling for the plan are detailed because  they
provide an example worth emulating.

          Examples are then presented that include:  a County master plan for flood
control, erosion control and drainage; the master plan for the largest combined sewer
overflow pollution and drainage control project in the nation; flood control master
planning for a metropolitan area; and inferences of the Urban Studies Program of the
Corps of Engineers.

          The examples given in this Section deserve consideration by many readers
because they are prototypes in many respects of desirable practices and they suggest,
by what they do not include, possibilities for improvements elsewhere.

          Lastly, the cyclic process of reviewing alternatives in a continuing
planning process is briefly considered.  Noted is the need for local government
entities to be in close communication with regional planning agencies.  That is. a
jinn, positive linkage should be forged between metropolitan planning and local
government implementation.  Noting that the detailed local government master plans
alluded to earlier in this Section have required a minimum of three to five years for
their development, it is important to accept that the local government planning time-
frame must be measured in terms of a number of years.

Summary. Section 3 - Incentives for Comprehensive Planning

          The thesis of this Section is that rational planning requires conjunctive
consideration of the quantity and quality aspects of urban runoff within a
comprehensive multiple-use framework.  Some of the more obvious arguments are
explored, in connection with public viewpoints on land-use controls, disparities in
approaches to economic evaluation, potentials for water reuse, opportunities to reduce
extraneous sewer flows, control of erosion and sedimentation, and disposal of solids
from new joint or ad hoc treatment facilities.


                                       - 5 -

-------
           Acting against the conjunctive approach is the tendency of citizen groups
 and Federal laws to dwell on land-use controls as a panacea while minimizing the
 interrelations and interconnections inherent in water resources.  This point surfaces
 recognition of some of the challenges confronted by conjunctive planning and the
 great importance of public information, and even public education, in promoting the
 larger view.

           Economic evaluation of water quality control projects at the local level
 appears to be restricted to some form of minimum cost analysis, as opposed to flood
 control economic justification via cost-benefit analysis at the river basin scale
 Because of the trend towards the blending of river basin planning with metropolitan
 planning,  local agencies will increasingly face the need to reconcile this dichotomy
 of project evaluation criteria.  Further,  these disparities in economic bases for
 project evaluation would seem to suggest rather strongly that  planning for management
 of urban runoff quantity and quality should be conjunctive  to  avoid absolute
 confusion.

           After briefly reviewing various  national developments and trends it is
 concluded  that the reuse of wastewater treatment plant effluents and reclamation  of
 surface runoff is  more likely to be via artificial recharge of groundwater supplies
 in general.   This  eventuality can be provided for only through conjunctive planning*.

           A  national  Symposium concluded that,  because the  majority of publicly-owned
 wastewater treatment  works  will be committed by the time Section 208 plans are  ready
 for review and implementation,  the only questions  that will still  be open  by that
 time will  be nonpoint sources of pollution and  growth management and who pays  for
 growth.  Thus,  land-use control issues will continue to simmer for some  time.

           Erosion  and sedimentation control is  one of several  water quality issues
 that cannot  be divorced from their causative driving force,  urban  runoff.

           With regard to  treatment of  stormwater and combined  sewer flows, jiandljng
 water volumes  is only part  of the  problem,  for  once  solids  conveyed by water are
 removed  they must  be  disposed of somewhere.

           The  point of  resolution  for  inconsistencies,  conflicts and  duplications of
 legal instruments  imposed by  higher  levels  is at the local  level of government.   It
 is  concluded  that  for  this  reason, among several,  local  governments  are forced  to
 define their own problems.

 Summary, Section 4 - Some Flow Management Considerations

          Because urban drainage and flood  plains are almost always  the responsibility
of  separate and different kinds of jurisdictions, their  interconnected behavior is  the
 responsibility of neither type of organization.  A similar  splintering of authority
over water quality aspects  plagues most metropolitan areas.  Given  this institutional
 patchwork, about the only avenue for conjunctive consideration of  these otherwise
disparate  issues is via comprehensive planning and coordination at  the metropolitan
 level.

          Asserted in this Section is that  the guiding principle should be to reduce
the liabilities and increase the assets of urban runoff.

          After reviewing a range of techniques applicable to urban runoff control
the reader is warned that although numerous schemes have been postulated for


                                       - 6  -

-------
controlling the quantity and quality of urban runoff,  very few of these concepts
have been tested in full-size system applications^, and the research on which many of
them are based is fragmented and the findings cannot be sufficiently generalized for
universal applications.  By default, local governments are obliged to assure themselves
of the relevance and reliability of most such schemes before making a commitment to
their wholesale use.  This leads to an extension to the last conclusion above for
Section 3:  local governments are forced to define their own problems and to take the
initiative in finding solutions to those problems.

Summary, Section 5 - Utilizing Simulation

          While simulation is undoubtedly an effective means for defining problems and
analyzing alternative urban runoff control strategies, its most important use can be
in the assessment of expected system performance.  Performance reliability is
ultimately the fulcrum of local government political acceptability.

          This Section commences with reasons for using simulation and a historical
reckoning of practical reservations against incautious acceptance of runoff models.
Emphasized is that simulation techniques adopted should not exceed the level of
mastery of such tools by the user and that tools should be selected on the basis of
their suitability for solving problems.

          Performance reliability is defined in terms of four serially connected
principal considerations, which must be completed in the following order to be fully
effective:  testing simulation models against local field data; using a long-term
precipitation record as the basic reference; developing frequency rationales for
levels of protection; and simulating a range of storms and system loadings.  An
emphasis is placed on the testing of simulation models against local field data
because this _issyie_JLSthe crux of credibility as well as reliability.  As things now
stand, local governments are substantially on their own for the acquisition of field
data.

          The concept of spatial sampling for field instrumentation of catchments
having representative land uses is outlined.  Stressed is the importance of rapid
application of data to models, to check data reliability but principally for
expeditious calibration of models, the primary use of such data in planning.

          Control of flooding and water pollution must be based on probabilities of
occurrence because of the randomness of precipitation.  Rationales are developed for
using long-term rainfall records as the reference for defining urban runoff quality
and quantity control objectives in probabilistic terms.  Addendum 2 is an extension
of this presentation.

          Underscored in this Section is that the only realistic defense for planning
in an atmosphere of ambiguous policy is to employ procedures the results of which have
an inherent flexibility for conversion to meet alternative policy goals.

          Reliability in the employment of calibrated tools for the exploration of
alternatives planned for the future is a function of several things, but the most
important are probably the shrewdness of design of the field data network and__the
extent to which its results approach an ideal set of representative samples.
                                       - 7 -

-------
 Summary. Section 6 - Urban Runoff Models

           Reiterated in this Section is the importance of calibrating any urban
 runoff model against local field data.

           Models are characterized in this Section in terms of their applications.
 Specific models are identified,  with an emphasis on sewered system planning
 applications,  although associable analysis/design applications are duly noted.
 Receiving water modeling is handled separately because of the tendency towards
 customized adaptation of available models for that purpose.  Examples are given of
 some regional  receiving water simulation studies.

 Summary. Section 7 - Some Moclelinfi Considerations

           Models can be used ineffectively if they are selected and employed without
 prior careful  consideration of their data requirements and their place in overall
 information processing.

           While flood plain mapping of 100-year stream stages is a minimum requirement
 the mapping of more frequent occurrences is  required for analysis of the economics of
 flood damage mitigation.

           Permanent field data acquisition systems for operations,  particularly for
 automatic control, can be readily justified,  whereas only temporary installations can
 be  justified for other purposes  except for long-range monitoring.

           The  three themes described above are singled out in this  Section because
 they are particular pitfalls that have been  encountered in past planning efforts.

           Also,  the latest developments in the manipulation of storage via automatic
 control  are described.

 Summary.  Section 8 -  References

           As a convenience to  the reader,  references are divided according to the
 Section  in which they are cited.   Considerable discrimination was  employed in
 selecting these  references from  an initial listing with over three  times  as many
 entries.   Rather than cite,  say,  six related  references,  the writer chose to use
 either the latest  (which  itself  cites  the  ones  omitted)  or  the one  or  two with  the
 most  useful  information.   Some references  are  cited  to  acknowledge  credit for ideas
 or material  while  others  are mentioned  merely  to  document  a point being made.

           Knowing  that  every reader  will have  different  interests and  a unique
 familiarity  with the  body of literature involved,  it would  be folly to  attempt  to
 single out a short list as  the "most  important  reading".  The need  to  cite numerous
 references where each dwells on.a narrow portion  of  the  spectrum merely confirms a
 contention in  this  report that research and development  on  urban runoff control has
 travelled  a  disjointed, almost haphazard journey,  with numerous  efforts  pursued
 independently.

Acknowledgment

           This report has  been prepared on behalf of and with the assistance of the
ASCE Council on  Urban Water  Resources Research by its  full-time  operating arm,  the
ASCE Program bearing  the  same name.  The activities  and  products of the Program are


                                        - 8 -

-------
well known.(°)  Suffice it to say that the central objective of the Council is to
help advance the state-of-the-art by identifying and promoting needed research and
by facilitating the transfer of the findings from research to users.  A Steering
Committee designated by the ASCE Council gives general direction to its Program:
Mr. S. W. Jens (Chairman); Dr. W. C. Ackermann; Dr. J. C. Geyer; Mr. C. F. Izzard;
Mr. D. E. Jones, Jr.; and Mr. L. S. Tucker,  Mr. M. B. McPherson is Program Director
(23 Watson Street, Marblehead, Mass. 01945).  Administrative support is provided by
ASCE Headquarters in New York City.

          Financial support for the preparation and duplication of this report is
entirely from a grant to ASCE by the Research Applied to National Needs program of
the National Science Foundation.  Technical liaison representative for NSF/RANN is
Dr. J. Eleonora Sabadell.  Any opinions, findings, and conclusions or recommendations
expressed herein are those of the author and do not necessarily reflect the views of
the National Science Foundation.  The ASCE Council is greatly indebted to NSF/RANN
for supporting this nine man-month project.

          Formal review of the two drafts of this report was generously provided by
an ad hoc Advisory Panel assisting the ASCE Council:

  Mr. Dennis Athayde, NPS Branch, Water Planning Division, EPA, Washington, D.C.
  Mr. Harold C. Coffee, Jr., Department of Public Works, San Francisco
  Dr. Charles N. Ehler, Office of Research and Development, EPA, Washington, D.C.
  Dr. Neil S. Grigg, UNC-WRRI, North Carolina State University, Raleigh
  Mr. Forrest C. Neil, Metropolitan Sanitary District of Greater Chicago
  Mr. Harry C. Torno, Office of Research and Development, EPA, Washington, D.C.
  Mr. L. Scott Tucker, Urban Drainage and Flood Control District, Denver
  Dr. Stuart G. Walesh, Southeastern Wisconsin Regional Planning Commission, Waukesha
  Dr. Donald H. Waller, Nova Scotia Technical College, Halifax

          Information, assistance and review were also provided by scores of other
persons from among the hundreds of ASCE Program cooperators.  In particular, vital
assistance was provided by officials responsible for some of the more innovative
master planning projects.  However, none of the persons named or alluded  to in this
subsection are to be held responsible for any of the contents of this report.  They
have nevertheless helped insure a better professional consensus on the issues involved
and the ASCE Council and the writer are therefore greatly indebted to them for their
invaluable assistance and generous cooperation.

          A significant portion of this report draws upon relevant parts of a prior
report that was prepared for the U.S. Geological Survey.(*'

          Typing of this report and its drafts was by Mrs. Richard Symmes of
Marblehead, Mass.

Principal Information Listings

          Some readers may be interested in the sources of principal  information
listings in urban water resources research.  Brief descriptions of on-going research
on specific subjects are available from the Smithsonian Science Information Exchange.
One service of the Exchange is the availability of "research information  packages"
that are updated every 90 days.  Of particular interest here are the  packages on  urban
water management, urban runoff, and effects of storm water runoff on  water quality  in
receiving streams,
                                       - 9  -

-------
          For completed research, reference should be made to the selected water
resources abstracts issued twice per month by the National Technical Information
Service that are compiled by the Office of Water Research and Technology^11/  Nearly
all of the reports cited in the abstracts that have been supported by public funds
are available for a cost-recovery charge from NTIS, in which cases the NTIS
identification numbers are included in the abstracts.  Information searches on urban
water publications can be greatly facilitated by referring to the annual cumulated
indexes of the selected water resources abstracts.(12)  In addition, NTIS has packages
of bibliographies with abstracts on some specific subjects, such as urban surface
runoff.  Both NTIS and SSIE also make custom searches of completed and on-going
work, respectively.
                                     - 10 -

-------
                                     SECTION 2

                         SELECTED PIANNING CONSIDERATIONS

Definition

          Comprehensive, areawide planning is conducted by metropolitan planning
agencies.  Implementation of such plans or their variants is by local units of
government.  Moreover, more detailed planning is done by local governments and their
plans are carried through construction into operations.  The term "comprehensive
planning" thus means quite different things at the metropolitan and local levels.
Also, local government plans can follow, be concurrent with, or lead areawide plans.
It would be totally presumptuous and entirely inappropriate to attempt in this
report to instruct regional planning agencies on the nuances of their responsibilities.
Rather, a major purpose is to present a synthesis of local government perspectives
on planning and associated problems which hopefully will facilitate metropolitan
intergovernmental cooperation and coordination.  Because an Intended audience of
this report is Section 208 metropolitan planning agencies, a central purpose is to
bring to their attention some potential impediments that might frustrate effectuation
of their proposals.  There are cautions also, directed at local governments, that
are intended to remind them of technical pitfalls that should be avoided in their
dealings with areawide plans.  In short, we hope that the uninvited intervention of
an outside referee will be of some assistance in improving the understanding of the
participants in a very complex process.

          Regardless of the governmental level involved, when we use the term
"comprehensive planning" we are referring to a multipurpose scope, such as for
integrated management of flooding protection, pollution abatement including erosion
control, and water conservation.  That is, despite the label of comprehensiveness,
we are really addressing only a subset of the overall metropolitan comprehensive
planning matrix.  Just about all broad-scale, leading-edge, local government plans
we have encountered are called "master plans," although this term appears to be
abhorred by many regional planning agencies.  .There are not many local government
plans under preparation that are truly comprehensive because the impact of newer
policies has not been factored into many such plans and, as a result, there are
simply not many broad urban runoff control plans under way and those that are
usually have been initiated to comply with a State directive against a particular
local government.  Further, comprehensive urban water resource planning is a
relatively new concept to local governments and very few plans have been carried to
the threshold of implementation, even for the narrower objective of urban runoff
control.

An Example of Local Government Planning Elements

          What is called for above is essentially a systems approach to facilities
and management planning.  For example, elements needed in water quality control
combined sewer system planning are said to include the following:(•")

. An analysis of implementing governmental entities.
. A public information system.
. A set of goals, consistent with existing and proposed regulations  for the foreseeable
  future.
. A realistic and attainable set of regulations.
. A definite understanding and description of the problems  and  their analysis.
. A complete description of the causes of the problems.
                                                                      (Continued)

                                      - 11 -

-------
 . A finite description of the existing physical system and its inadequacies.
 . A finite description of the receiving-water environment for biota.
 . A finite description of solids disposal locations.
 . A detailed rainfall history perhaps spanning 30 to 50 years.
 . A detailed history of water-demand and water-use factors.
 . A detailed pollutant-constituent history.
 . A complete water quality inventory showing constituents in the water supply,  in
   wastewater prior to treatment, and in effluents.
 . A means for evaluating treatment in terms of storage and overflow options.
 . A series of viable alternatives based on real data.
 . A recommended solution or set of solutions.
 . A flexible facilities and operations staging program.
 . A practical financing program.
 . A source of funds.

 Experience indicates that a steering committee comprised of  outside experts can be
 very helpful in avoiding oversight or slighting of the above elements.   While the
 above elements may be regarded as idealistic by some, the local government that ignores
 them does so at its peril.   Because of unique combinations of local conditions, it
 should be evident that satisfaction of the needs listed above would necessarily vary
 in detail,  methodology and  procedure from one local  jurisdiction to another, even
 in the same State.   However,  the institution of a public information program early  in
 a local government  planning endeavor is very important.

 An Illustration of  Local Government Planning

           One of the most outstanding comprehensive,  imaginative and thoroughly
 developed plans for municipal pollution control has  been prepared by the Department
 of Public Works,  City and County of San Francisco, California.   San Francisco is
 served almost exclusively by  a combined sewer system.   The plan is  founded on a
 data bank of unusual scale,  ranging from runoff-quality field measurements initiated
 in 1965,  through installation of a pilot wet-weather dissolved-air  flotation treatment
 plant  on  a  field catchment,  to the use of a unique automatic  rainfall-runoff
 monitoring  system.   A deliberate step-by-step approach was used in  developing the
 plan,  which was carefully phased with field information acquisition.  The quality of
 the  engineering work in the development  of the plan  was  superb,  and the  plan was
 supported by several DPW studies and  25  ancillary reports  prepared  by consultants.
 Pioneering  development  of automatic control capability by  the DPW is discussed  in
 Section 7.

          Water quality standards  for both dry-weather and wet-weather flows have
 presented a moving  target for pollution  abatement compliance  by local agencies, and
 more changes  in regulations may  be  anticipated in  the  future.   Recognized in the
 development of  the  "master plan"  for  wastewater  and  combined  sewer  overflow control
 in San  Francisco was  that for various  abatement  plans  the  timing imposed by Federal
 and State regulatory agencies  and  subsequently supported by  their funding is the
 critical  factor controlling the  final  selection  of such  a  plan.  The flexible plan
 adopted by  the  City  is  amenable  to  adjustment  during its  implementation  to meet
 future  shifts in regulations,  but  the  feature  to be  discussed  here  is concerned
 instead with  project  financing scenarios  for meeting a variety  of possible abatement
 goals.

          The preliminary master  plan proposed in  1971 provided  four alternates that
would allow an  overflow to occur  from a maximum  of eight times  per  year  to a minimum
of once in  five years.  Because  the amount of  Federal  funds and  the eligibility


                                       -  12 -

-------
requirements for grants had changed almost continuously,  and wet-weather projects  had
a lower priority than dry-weather projects (and were not  scheduled for either
California or Federal grants), the six variables affecting the scheduling of a bond
issue were analyzed:

. Frequency of overflow occurrence (4 alternates).
„ Effluent disposal site for a wastewater treatment plant (3 sites).
. Level of dry-weather treatment (3 levels).
. Number of years for completion of all facilities (10 to 60 years).
. Per cent of total costs obtained from grants (07. to 807.).
. Amount of the City's financial resources that might be  allocated (four values).

          Results of the analysis of the 1,944 combinations for each of the four
alternates were presented in a single ingenious graph displaying all  six variables.
Thus, for a given overflow control level, a given effluent disposal site, a given
dry-weather treatment level, a given number of years for  facilities completion, and  a
given per cent support by grants, the amount of the City's resources  that would be
required can be determined from the correlation graph. Through this  graph the question
of who would pay what share of the costs for meeting a particular water quality
objective under a particular set of circumstances becomes immediately clear.  This is
an excellent illustration of the fulfillment of an engineer's responsibility in this
instance:  to present the range of options available to officials with the authority
to make decisions and recommendations.  In essence, the Department's engineers
provided the City's elected officials with the means for  negotiation with regulatory
agencies having both enforcement powers and any funds for supportive grants.

          While the above description is for abatement in terms of the average number
of combined sewer overflows permitted per year, this could be readily converted into
overflow volumes or pollutant loads or seasonal serial events instead, by reference
to the City's performance simulation results described elsewhere  in this report.

          Parenthetically, although the basic featuresC15) of the preliminary master
plan^  ' have not been altered, some modifications were made in 1973(1'' and again in
1976(18.) when implementation of the first phase commenced.

          Of the 28,000 acres in San Francisco, 24,000 are drained by the sewer system
(plus 2,000 acres in adjoining San Mateo County).  While the City is surrounded on
three sides by receiving waters, the Pacific Ocean and the Bay, strearaflow flooding
does not exist in the City and therefore was not a consideration  in development of
the master plan.  Also, the City is essentially completely developed and thus  is
not involved in the conversion of low-density land into suburban  expansion.  We now
turn our attention to the more general situation where both sewer systems and  streams
drain the land and where both existing and projected drainage systems are involved.

A Broader Perspective

          A review of a number of local government planning projects has revealed a
commonality of six discrete steps, which might be regarded as the framework of the
standard planning process;(19)

. Define problems,  establish goals and identify constraints.
. Assemble data and refine definitions of problems.
. Formulate alternatives.
. Analyze alternatives.
. Identify more promising alternatives.
. Evaluate the trade-offs among selected alternatives.

                                       -  13 -

-------
           Embracing the above  six parts of  the basic planning process, Figure 1
  is a block diagram depicting the urban runoff control planning concept, which attempts
  to accommodate situations with sewer systems and streams and land already developed
  and projected for development.  The term "master planning" appears in the title to
  insure a local government application connotation.  The sequence of steps shown is
  not intended to specify the scheduling of the steps or the degree of their completion
  needed, but the order in which decisions should be reached in order to assure an
  efficient planning progression.  Considerable feedback would be involved from any
  given major step to various preceding steps, but no attempt is made to show the
  raorass of possibilities on the  diagram because it is complicated enough as it is.
  Simulation and related hydrological and water quality field data acquisition aspects
  are discussed in detail in Section 5.

           In the preparation of Figure 1, all the significant features of three
  selected representations were also taken into account,  for water flow control(20)
 and water quality controlv^l»22) local government planning.

           For conjunctive flow control and quality control planning,  water quality
 considerations tend to dominate.  This is primarily because flow control is concerned
 solely with quantity while quality control is concerned with the complex relationships
 between runoff and its pollutant burdens. A compelling reason for engaging in
 conjunctive planning,  however,  is that the overall  flow processes  for quantity and
 quality planning are shared in common,  making it  difficult to  justify two independent
 planning efforts which draw upon the same field  information resources and often use
 the same or similar or related tools of analysis.   Other arguments favoring
 conjunctive planning are presented throughout the remainder of  this report.

           Before leaving Figure 1,  it  is  important  to note that  a  local  agency can
 become  so  obsessed  with long-range planning  that  more immediate, partially
 mitigative  solutions  can be overlooked  or inordinately  deferred.   Zoning ordinances,
 flood  insurance  and incorporation of detention storage  in  new  land development are
 examples  that  come  quickly to  mind.  Preparation  of most  local government  plans
 completed  to date has  required  a minimum  of  three  to five  years.   There  are  interim
 or  initial actions  that  can be  taken early in this period  that do  not require
 significant local government funding and  hence are not  likely to be as highly
 controversial politically.

 Marshalling for a Plan

          Mentioned earlier was that the  first phase of  the San Francisco master plan
 is being implemented, reflecting the effectiveness of its  careful  preparation.  We
 now turn to a planning effort recently completed for metropolitan  Milwaukee.   To be
 discussed here are  the deliberate, careful steps that should be taken  to organize
 and marshall forces for  the conduct of plan  preparation.  While the San Francisco
 effort originated with and was  pursued by an operating department of  a municipality,
 that for the Milwaukee area was  launched  from a metropolitan level.  The description
which follows is a digest of coverage in an  earlier ASCE report.(9)

          A preliminary engineering study leading to a plan for the abatement of
pollution from combined sewer overflows in Milwaukee County was initiated by the
Metropolitan Sewerage District of the County, a special-purpose metropolitan agency.
The combined sewers in the County are owned  and operated by the City of Milwaukee and
one of its adjacent suburbs, the Village of  Shorewood,  both of which are entirely
within the County and the Sewerage District.   Three rivers pass through the area
served by combined sewers:  the  Kinnickinnic, the Monoraonee and the Milwaukee Rivers,


                                      - 14 -

-------
INVENTORY
EXISTING
DRAINAGE
SYSTEMS

CALIG
SIMUL
MOO


IRATE
ATION
ELS


ASSEMBLE
PRECIPITATION-
RUNOFF-
QUALITY
DATA
1



ACHIEVE
ACCEPTABLE
UlCTTIRirAI
PERFORMANCE
SIMULATION



SPECIFY
FORESEEABLE
FLOW AND
QUALITY
OBJECTIVES

ASSEMBLE
LONG-TERM
uiernpfCAl
RAINFALL
DATA




POSTULATE
ALTERNATF
SOLUTIONS


SPECIFY
EXPECTED
TIME, MONEY
AND POLICY
CONSTRAINTS

L
A
SIMULATE
PERFORMANCE
FOR
ALTERNATES


ACQUIRE
AUXILIARY
FLOW AND
QUALITY
DATA

1
DEVELOP
CRITERIA
FOR
ECONOMIC
ANALYSIS


*
SUPPLY
FLOODING
DISBENEFIT
PARAMETERS

EVALUATE
ATTAINABLE
TREATMENT
FACILITY
EFFICIENCIES

REVIEW
UPDATED
OBJECTIVES
AND
CONSTRAINTS

DETERMINE
OPERATION
AND
MAINTENANCE
LATITUDE






EVALUATE
EFFECTIVENESS,
PERFORMANCE
AND COSTS OF
ALTERNATES



SUPPLY
UNIT
COSTS
1,

SCREEN
Al TCBklATT
*»LI CLPfNAI C
APPROACHES


DETERMINE
BEST
LOCATIONS
FOR
TREATMENT



1.
i
SELECT
BASIC
APPROACH
L
1
ANALYZE
A
D AftJfSF
nMN\>t
OF
ALTERNATIVES




REVIEW
OTHER
WATER
RESOURCE
CONSIDERATIONS



DETERMINE
CAPITALIZATION
LATITUDE

1 DEVELOP PLANNING GUIDELINES
DETERMINE
POLITICAL
FEASIBILITY
OF LAND-USE
REGULATIONS


REPORT
FINDINGS
TO
ELECTED
OFFICIALS


NEGOTIATE
ALTERNATIVES
WITH
REGULATORY
AGENCIES
                               REVIEW
                               AND
                               UPDATE
                               PLAN
FIGURE  1 - BLOCK DIAGRAM OF URBAN RUNOFF
         CONTROL MASTER PLANNING CONCEPT
                - 15 -

-------
  all  in  the Lake Michigan Basin.  The three  rivers merge  in  the downtown area of the
  City of Milwaukee and  enter Lake Michigan through a common  channel.  The mouth, the
  confluence and reaches part way up  each river are navigable and used for various types
  of commercial shipping.

           Alternative abatement methods for the master plan were investigated as a
  part of a comprehensive study of water-related problems of  the Milwaukee River
  Watershed conducted by the Southeastern Wisconsin Regional  Planning Commission (SEWRPC)
  at the  request of the Common Council of the City of Milwaukee and the Board of
  Supervisors of Milwaukee County.  After three years of intensive work by SEWRPC
  retained consultants and a technical advisory committee comprised of Federal. State
  and  local citizens, public officials and technical experts, the report(23,24) was
  published and later adopted as a regional guide for development and for solution to
  existing problems.  That part of the report that dealt with the combined sewer
  problem investigated many different methods of abating pollution from combined sewer
 overflows and recommended, based on the pre-feasibility investigation,  the use of a
 combination of methods entitled "Deep Tunnel/Mined Storage/Flow-Through Treatment".
 The  report also recommended that an existing sub-regional agency,  the Metropolitan
 Sewerage District of the County of Milwaukee,  agree to conduct a more in-depth
 engineering study leading to a master plan.

           In April,  1973,  the Metropolitan Sewerage District,  through the agencies of
 its constituent operating bodies,  the Sewerage Commission of the City of Milwaukee, and
 the Metropolitan Sewerage Commission of the  County of  Milwaukee,  requested SEWRPC  to
 prepare a Prospectus for a preliminary engineering study  for combined sewer overflow
 abatement for use in selecting a consultant  to conduct the study.   The  Prospectus  was
 prepared by SEWRPC with the assistance of  a  Technical  Advisory Committee of outstanding
 Federal, State and local specialists in the  water  pollution field and completed  in
 July, 1973.(25)   ^he Prospectus  outlined a proposed  engineering study based on the work
 contained in the Milwaukee River Watershed report  and  directed a more in-depth study of
 possible alternatives  as  well  as further advances  in the  field that  had  occurred  since
 the publication  of the report.   The  Prospectus was used as an  invitation for proposals
 and was  sent  to  many of the most prominent consulting  engineering firms  in the nation.
 Twenty-one  proposals were  received,  as  evidence of the interest  in  such  a  project,
 with  six consultants invited  for oral  presentations.

          The  selection of  the six consultants, as well as the recommendation  to the
 District of the consultant  to be retained, was made  by the same Technical Advisory
 Committee; and the Committee was later  transferred to  the  jurisdiction of  the  District.
 The District  then  entered  into a contract with a consultant  for the  preliminary
 engineering study.   The consultant worked  under the  direct supervision of  the  staff  of
 the Metropolitan Sewerage District assisted  by the Technical Advisory Committee.  The
 primary  jurisdiction of the Sewerage District is the planning, construction  and
 operation of the sanitary sewerage facilities for approximately 420  square miles,
 comprising the Metropolitan Milwaukee area.  The District  operates two regional
 wastewater treatment plants, 240 miles of  interception sewers  and 20 miles of  improved
 stream channels.

          Three basic approaches for abatement and control were considered in the
 Milwaukee River watershed study:  storage of overflows with  their subsequent slow
 release  for conventional treatment at existing wastewater  treatment  facilities; flow-
 through  and inflow treatment of overflows at special treatment facilities during
 peak  flow periods; and complete separation of combined sewer systems.  Ten different
 storage  schemes and three different methods of separation were explored in the total
of fifteen different alternatives considered.^1*/  On the basis of careful economic,
 engineering and legal analysis of the fifteen alternatives,  it was recommended that a


                                        - 16  -

-------
combination of deep-tunnel conveyance, mined storage and flow-through treatment be
utilized for combined sewer overflow abatement for the entire County combined sewer
service area of 27-square miles.(25)  The purpose of the preliminary engineering
study was to verify the feasibility of the recommended plan or the best alternative
thereto, and to determine the needed system configuration with particular emphasis on
the balance to be struck between capacities for conveyance, storage and flow-through
treatment.

          The study area not only encompassed the combined sewer service area of the
three rivers and the lakefront but was extended to include all areas that contribute
to the hydraulic flows in the combined and intercepting sewers, and to permit proper
analysis of related wastewater flows and stormwater discharges and related drainage
and flooding problems.  Provided was all information necessary for the respective
political entities involved to make the public policy determinations required to
proceed with the pollution abatement program.

          Also considered in the study was the potential application of various
combinations of collection, conveyance, storage and treatment facilities for the
additional handling of stormwater runoff from throughout the Milwaukee urbanized
area, should stormwater runoff treatment become necessary in the future to protect
surface water quality, particularly in Lake Michigan and the interconnected Milwaukee
Harbor estuary.

          Tunnel systems have been considered in Boston, Chicago, Detroit, Milwaukee
and Washington, D.C.(26'  A brief description of the Chicago plan is included as a
case study in an ASCE Program Technical Memorandum.(1°)  The Chicago studies were far
more advanced than those for Milwaukee in 1973(2**' and, as noted later in this
Section, a major phase of that master plan is under construction.  Because little has
been published on the Milwaukee master plan'  » '' more description is given here than
would otherwise be justified.

          Local government plans do not evolve overnight and carefully weighed plans
are usually founded on several years of deliberate planning and development, as
epitomized by the initiation of the Milwaukee master plan.

          Work on the Milwaukee Combined Sewer Overflow Pollution Abatement Project
commenced late in 1974 and the first of three phases was completed less than a year
later/28)  The organizational and scheduling plan for conducting the project, by
tasks/29) would be of interest to planning entities embarking on the development of
such plans.  The last phase was completed in mid-1977, and the final report provided
the basic information required to start design for construction.

          "Reliable local municipal stormwater drainage facilities cannot be properly
planned, designed, or constructed except as integral parts of an areawide system of
floodwater conveyance and storage facilities centered on major drainageways and
perennial waterways designed so that the hydraulic capacity of each waterway opening
and channel reach abets the common aim of providing for the storage, as well as the
movement, of floodwaters.  Not only does the land pattern of the tributary drainage
area affect the required drainage capacity, but the effectiveness of the  floodwater
conveyance and storage facilities affects the uses to which land within the  tributary
watershed, and particularly within the riverine areas of the watershed, may  properly
be put."^)  The preceding is given as the guiding principle  for water control
facility development  for the Milwaukee River watershed, and hence  for  the subject
master plan.  Flood mitigation aspects are discussed later  in  this  report.
                                       -  17  -

-------
           It is  important  to note  that  in  the 1970  inventory,specific model
 information needs were not an issue.  In f-".ct, most of  the hydrologic models deployed
 in the master planning were only partially developed, at most, in 1970.  This
 illustrates the  recent phenomenon where information requirements shift in only a very
 few years as modeling capabilities are  improved.

 An Emphasis on Flooding and Drainage

           An outstanding example of a comprehensive local government plan for flood
 control, erosion control and drainage is the program in Fairfax County, Virginia.
 The County lies to the west and southwest of Washington, D.C.  The County's population
 leaped from 100,000 in 1950 to 560,000 residents in 1974.  With this population spread
 over an area of 405-square miles the County can be said to be "suburbanizing" because
 it is being transformed from a predominantly rural area to a residential area for many
 people employed in the District of Columbia,  Such explosive growth has strained
 various public facilities and services.

           Problems associated with stormwater runoff were parried  in November,  1971
 when the voters of the County passed an $ll-million storm drainage bond referendum.
 The objectives of the bond program are to solve  existing flooding  and drainage problems
 to make plans  for avoidance of such problems in  the future,  and to improve the storm
 drainage system operation and maintenance capability of the  County staff.   In the
 spring of 1972,  a consulting firm was  awarded a  contract on  the planning program,
 which will  be  completed late in 1977.   Although  the program  has been described
           /Q  o f\ OT j
 elsewhere,V5»-'"»-'•' some of its  salient features  deserve mention here.

           Briefly stated,  the specific goals  of  the program  are to:   develop
 immediate solutions for existing flooding  and channel erosion problems;  improve methods
 and procedures  for developing budget projections  for storm drainage  capital  facilities;
 establish a standard  baseline from  which to measure environmental  changes;  rewrite
 the drainage code and policy  by  incorporating a balance  of on-site detention and
 off-site  pro rata costs; and  make  full utilization  of stormwater runoff  by regarding
 it as  a resource  out  of  place.(32)

          The thrust  of  the planning program  is on  flooding and erosion.   Because  it
 commenced prior to  the enactment of PL 92-500, water quality  was not considered to be
 a  major issue at  that time.  However,  included among the planning  activities were
 "environmental baseline  studies"'-*3) and environmentally-oriented  design  solutions,
 in all of which water quality was a consideration.  Detailed  study of water  quality
 management in the County commenced with  the Occoquan basin.

          The "environmental baselines"^^) established  for each of  the  29 watersheds
 provided the environmental  framework for developing basin plans and  the necessary  data
 for analysis of projected environmental conditions.  Rather than use a benefit-cost
 approach for evaluating alternative plans, a "problem identification matrix"  was
 chosen to permit better inclusion of benefits and costs  for such usually elusive elements
as visual impact, aesthetics, and creation or destruction of wildlife habitat and  aquatic
 systems.  The matrix approach, which has also been used  elsewhere, was a technique  for
 systematically identifying  the general extent and importance of stormwater-related
problems in the County and evaluating  the adequacy of proposed  solutions to  those
problems.  Its use required three different elements:  the objectives of the  drainage
planning program; the relative importance of each objective;  and the degree  to which
each objective would be met under a given set of circumstances.  Although program
objectives can and may be changed or regrouped as experience by the County in the
application of the method is gained, initial objectives, not necessarily in order  of


                                       -  18  -

-------
importance, were:   freedom of residences from flooding damage;  freedom of  commercial
and industrial facilities from flooding damage;  freedom of  public and  institutional
facilities and equipment from flooding damage; control of bank  and channel erosion;
protection of aquatic ecosystems;  protection of wildlife habitats; freedom of  parks,
recreation and aesthetic areas from flooding damage;  and prevention of traffic
interruptions.  Although protection of health and safety is not specifically included
in this listing, it is an overall  consideration and would take  precedence  over other
objectives as part of any flood control plan.

          By the end of 1977 the consulting firm will make  available a summary report
on the program, encapsulating procedures and findings embodied  in the  dozens of
program reports for the County.

Some Metropolitan Examples

          Chicago Area.  There are more than 5,000-railes of combined sewers within a
375-sq. mi. portion of the territory served by the Metropolitan Sanitary District of
Greater Chicago (MSDGC), which includes the 220-sq. mi. of  the  City of Chicago.   The
District serves more than 5,500,000 people in an 860-sq. mi. area, exercising  control
over 75-miles of open waterways with respect to drainage, pollution control and
navigation, much of which is monitored for water quality.(™)  Lake Michigan is  the
water supply source for the area,  which is drained by three rivers away from the lake.

          An intergovernmental committee resolved a "Tunnel and Reservoir Plan"  for
combined sewer overflow pollution abatement^-*' that  incorporates flood control
features.  In 1972 the cost of implementing the plan  was estimated at 3.3-billion
dollars.^  '  Currently, the first major phase is under construction.   "A unique
feature of the plan is its inherent flexibility to take advantage of discoveries made
during its ten year construction period."(36)  Principal features of the plan  have
been outlined.(1»)

          The primary function of the MSDGC is the collection and treatment of the
water-borne wastes from the City of Chicago and its suburbs.  Because flooding is
related to such collection and treatment, the District has  initiated long-range
planning and local flood plain control.  The District has used its authority over the
granting of sewer permits to require local authorities to enact flood plain ordinances
that safeguard flood plain developments against high water  and requires on-site
detention of excess runoff from new developments.(")  A streamflow model developed
by the District^*' is used in the preparation of watershed master plans.  One of the
completed plans, for a 52-square mile watershed, calls for  measures consisting of:
five floodwater retarding structures; a multiple-purpose structure (flood protection
and recreation); 261-acres of flood plain preserve; about 1.8-miles of channel
modification; and a land treatment program for more than half of the watershed.'
Participants in this planning, in addition to the MSDGC, were the local Soil and Water
Conservation District, the Cook County Forest Preserve District, the State of Illinois,
and four villages and four park districts,(37) illustrating the multiple  jurisdictions
commonly involved in flood control planning.  (The MSDGC spearheaded the  combined
sewer overflow pollution abatement master plan mentioned above, in cooperation with
the State of Illinois, Cook County and the City of Chicago^5').

          Denver Area.  A local government special district, the Urban Drainage and
Flood Control District was established in 1969 to solve growing drainage  problems in
metropolitan Denver.  The District covers about 1,200-square miles of all or  part of
six counties.  A joint project by the Denver Regional Council of Governments  and  the
District, partly supported by the HUD Urban Systems Engineering Demonstration Program,
developed procedural methodology for urban drainage and  flood control master  planning
in the region.(39)
                                        - 19  -

-------
           Major drainageways were targeted in the methodology development,  defined
 as those streams, creeks and gulches that have definable flood plains.   Three types
 of principal activities to be pursued over two decades are called for under the
 Denver regional program:  (1), preventive master planning for areas  where flood
 plain regulation, land use controls and other preventive actions can be utilized;
 (2),  design master planning for areas where problems already exist and  facility
 construction is known to be required; and (3), construction for developed areas where
 preventive measures are not feasible and where channels, culverts, sewers and other
 structures are needed to provide protection.^0'  Activities included are:   delineation
 of flood plains on major drainage channels; regulation of all unoccupied and occupied
 100-year flood plains; 100-year protection on occupied flood plains;  National Flood
 Insurance Program coverage on occupied flood plains  where protection is not economical-
 the provision for limitation of runoff from new real-estate development by  ordinance
 and State law; flood storage capacity and spillway protection for dams  in the region*
 integration of major drainage measures with the regional water resource management
 scheme;  and a flooding early warning system. (^0)  The  first master plan for a major
 drainage area using the methodology developed was  completed in
           Motivation for preventive  master  planning was strong.  The  cost of protection
 for  the quarter of the total  District  flood plain area already occupied, in 1972
 dollars,  was  estimated at $110-million.(39) Use of a preventive approach for the
 remaining three- fourths is proving to  be  comparatively inexpensive.

           Urban Studies Program. Corps of Engineers.  Congress and the Office of
 Management and  Budget  in 1970 expressed an  interest in having the Corps of Engineers
 conduct pilot studies  of wastewater  management for several major metropolitan areas,
 to assist metropolitan agencies  in their  planning efforts.  "The initial objective
 for  each  study  area  was to produce a feasibility report that would identify alternative
 means of  reaching  very high standards of  wastewater quality on a regional basis.  A
 further requirement  was that  the Corps of Engineers consider land treatment as an
 alternative to  advanced plant treatment sys terns. "(^2)  Initial feasibility studies
 were completed  in  1971 for five metropolitan areas:  Merrimack Basin  in New England,
 Cleveland-Akron, Chicago,  Detroit and San Francisco Bay.  Detailed studies for these
 five pilot  areas were  completed in 1973.  By the end of 1974 a total of 37 such
 planning  studies had been  authorized, and the number has since increased.  Several
 studies have been  completed.
          Provided will be a range of urban water resource plans that are compatible
with comprehensive urban development goals of the region under study, and these plans
will provide "an integrated approach to water resources management".'^3'

          Features have been summarized of the Chicago'**2' and San Francisco(^4) area
studies, and the growth of interest in the land disposal of wastewater concept has
been briefly defined.(^^  A significant feature of a study of possibilities for land
disposal in the San Francisco Bay region was the exclusion of places where urban
development might be anticipated by the year 2000.t45'  The important point here is
that not only are the effects of local pollution often felt well outside the metropolis
of origin, but with land disposal the abatement of pollution via this form of treatment
could also take place beyond the metropolis.

          Corps responsibilities under the Urban Studies Program can include:  urban
flood control and flood plain management; municipal and industrial water supply;
wastewater management; bank and channel stabilization; lake, ocean and estuarine
restoration and protection; recreation management and development at Civil Works
projects located in close proximity to urban areas; and regional harbor and waterway
development.  Urban runoff quantity and quality is addressed in several of the Urban
                                       - 20 -

-------
Studies.  Where associated master plans have been adopted by local governments they
are treated mostly as givens in the Urban Studies.

Two-Direction Communication and Time Horizons

          A 1970 analysis of regional land use and transportation planning drew
inferences -from case studies of thirteen metropolitan areas.  At that time the
conventional planning concept was the selection of a metropolitan plan from several
complete and integrated development proposals.  Proposed as a more viable substitute
was a style in common use today, a cyclic use of alternatives.  In the cyclic style,
"alternatives become a means of exploring and understanding the effects and implications
of diverse objectives, assumptions, plans and policies, often in response to a specific
problem.  * o . . . flexible and partial alternatives should be prepared and evaluated for
several types of plans at appropriate geographic scales, time horizons and levels of
detail.
          Table l     characterizes the cyclic employment of alternatives in a
continuing planning process.  The reasons for introducing Table 1 here are to critique
the role of local government master plans therein and to remind the reader of the
long time horizons required for such plans.

          Seemingly overlooked in Table 1 under "Sub-metropolitan Studies" is, in
addition to the impact of metropolitan plans, the identification of potential problems
at the local level, the input of such signals to the metropolitan phases, and the
acknowledgment of these problems in all metropolitan planning.  Consequently, the
allocation of only ten per cent of the total effort in Table 1 to Sub-metropolitan
Studies appears to be too small.  On the other hand, the more detailed planning for
local development generally fails to address the broader considerations involved in
metropolitan planning, with the consequence that there is often an appreciable
planning void between the two levels, with no organized or assured effort made to
accommodate these considerations.  These comments illustrate the common occurrence
of a discontinuity between metropolitan and local planning efforts.  Needed  is not
only the interdisciplinary interaction that is becoming more widely practiced, but
also a firmer interinstitutional approach.  The Consulting Panel on Water Resources
Planning of the National Water Commission indicated in 1973 that there was a long
way to go, claiming that:  "Water-resource planning has rarely been integrated or
coordinated with overall urban planning". W)  Section 208 planning has a great
potential for closing this gap.

          As noted in Section 1, and as illustrated in Table 1, considerable time is
required for the resolution of plans.  The detailed local government master  plans
cited in this report have required a minimum of three to five years for their
development.  Most will require at least  ten additional years  for  their  full
implementation.  Those being implemented have faced some degree of delay for
resolution of conflicts, right up  to the letting of construction contracts.
                                       -  21  -

-------
                                 TABLE 1




SUMMARY GUIDELINES FOR PLAN COMPONENTS OF CONTINUING PLANNING PROCESS(46)

Plan Type
Metropolitan
Development
Framework




Facility-
Service
System
Flans





Capital
Improvements
Program




Metropolitan
Studies and
Alternatives









Sub-
Metropolitan
Studies


_ 	
Purpose Time
Horizon
general framework for urban 20
development and basis for years
metropolitan public facility
and service system plans and
review of local plans and
programs

metropolitan system plans 20
and basis for designing the years
location and characteristics
of services and facilities
and programming their imple-
mentation (transportation,
water-sewer, open space,
sub-regional centers,
education, health, housing)
coordination of the alloca- 10
tion of fiscal resources at years
all levels of government and
the timing of implementation
of projects, both within and
among facility systems

identification and recon- 20-50
ciliation of development years
objectives, policies and
combinations of systems
and examination of long-run
impact of technological,
social and economic change
on urban development, as
basis for the metropolitan
development framework and
facility-service system
plans
impact of plans on the various
development process and
quality of urban environ-
ment at the community to
project scale

Level of Detail
pattern of ur-
banization,
location and
density of ac-
tivities and
layout of fa-
cility systems
sufficiently
detailed for
preliminary
design
studies




large or
composite pro-
jects as basis
for programming
and budgeting
by operating
agencies
highly
generalized to
very specific
as appropriate
to particular
purpose






detailed to
generalized




Proportion
of Effort
257.






30%








15%






20%











10%





                               - 22 -

-------
                                     SECTION 3

                      INCENTIVES FOR COMPREHENSIVE PLANNING

Synopsis

          Emphasized in a 1975 national symposium^"' co-sponsored by the ASCE Urban
Water Resources Research Council was that urban water quality control plans should be
comprehensive, making provision for and fully taking into account future water needs
and withdrawals and recognizing and making provision for appropriate future usage of
waterways and adjacent areas as wildlife habitats and for recreation.  Recognized was
that water supply, flood control, drainage and pollution control are interrelated in
urban planning and that independent development of drainage and flood control plans
might result in higher costs for achieving water quality objectives.

          The thesis of this Section is that rational planning requires conjunctive
consideration of the quantity and quality aspects of urban runoff within a comprehensive.
multiple-use framework.  Some of the more obvious arguments are explored, in connection
with viewpoints on land-use controls, disparities in approaches to economic evaluation,
potentials for water reuse, opportunities to reduce extraneous sewer flows, control
of erosion and sedimentation, and disposal of solids from new joint or ad hoc treatment
facilities.  Admittedly, difficulties and problems are emphasized, but for the sole
purpose of indicating that the only rational way in which a reasonable degree of
order can replace possible chaos is probably through highly skilled comprehensive
planning and policy coordination at the metropolitan level.

Land-Use Controls

          Citizen participation is urged or required in various national laws dealing
with urban runoff management.  While the public is perceived by various observers as
demanding more holistic or overall solutions to problems from its elected officials,
there is a countervailing tendency for responsible public groups to single out
specific parts of the puzzle for detailed scrutiny and simultaneously to oversimplify
the interconnections among closely related issues.

          Exemplifying this inconsistent view, the national Citizens' Advisory
Committee on Environmental Quality has defined planning as "the conscious selection
       c  choices  n land use'C1**) weeas a more  enerall  accetable defntion
of policy choices in land use/'1**  whereas a more generally acceptable definition
would be that planning is the process of deciding what resources to allocate, over
time and space, to achieve a set of specified objectives.  Noted by the Committee is
that many communities begin the planning process with a comprehensive study that
takes into consideration interactions between all present and potential man-made
activities and various natural characteristics, resulting in a document called a
master plan or comprehensive plan for use in guiding community development.
Emphasized by the Committee is the importance of a land-use inventory as "a useful
data collection device with which to obtain a fresh view of an area's relevant
features.  It is a device which will focus attention on the important problems and
proposals that are evolving around us — often gradually, without shock to our daily
lives — but with a tremendous long-range impact on our future.''^*9'  Water aspects
seem to be regarded by the Committee as mostly locational issues in an inventory,
as for other natural resources such as flora and fauna and geologic features.  By_
stressing the location of water aspects, the transitory nature of water resources
and the intermingling of waters in multiple uses is thereby completely ignored, as
opposed to the water orientation to supplement land-use inventories advocated in
Addendum 1 of this report.  Independent of planning issues, it has becm alleged that


                                      - 23 -

-------
 management of urban water resources has been viewed for the most part independently
 from land-use policy and management rather than as an integral element.($0)  These
 points are raised in an attempt to indicate some of the challenges confronted by
 comprehensive planning and the great importance, of public information,  and even
 public education, in promoting the larger view.

           Attempts by Congress to force a marriage between land and water management
 are exemplified by 40 CFR Part 131.11 supporting Section 208,  P.L. 92-500. Although
 intended by Congress to exploit the police power of local governments over land use,
 it is nevertheless expected that in a number of urban areas it may be politically
 impossible to utilize land-use controls for pollution abatement.(51) For this and
 other reasons, there are strong local biases towards capital-intensive  solutions.
 One of the other reasons is that nonstructural solutions,  which seem to  be favored
 by EPA for nonpoint source control,(52,53) ten(j to requ£re additional investment in
 operation and maintenance of affected urban services,  and thus would inflate  already
 burgeoning local agency operating budgets.  Also,  the efficacy of most nonstructural
 solutions has not been adequately demonstrated with the support of field data.   The
 intents of Congress in Section 208 are said to be the establishment of a land-use
 regulatory mechanism to achieve water-quality objectives over  the longer term,  and to
 provide an incentive and a mechanism for States and local  governments to come to grips
 with nonpoint source pollution.(54)  Local governments face difficulties in acceptance
 of land-use regulatory mechanisms in the absence of reliable demonstration of their
 effectiveness.

           EPA has released in three stages a two-volume manual covering  procedures
 available for water quality management,  with particular emphasis  on urban stormwater,
 to assist the conduct of State and areawide planning under Section 208 of P.L.
 92-500.(55)  xhe manual notes that the establishment of an overall wastewater
 management plan calls for examination by State and areawide planning agencies  of the
 wide variety of pollutant sources and corresponding receiving  water impacts in terms
 of the necessity and feasibility of their control.

           Existing and future surface water quantity and quality  problems are
 inextricably tied to existing and future land  use  patterns.  For  example,  because
 land use  significantly influences  the volume and rate  of runoff from the land  surface
 to stream systems,  land use is  a primary determinant of the location and severity of
 stormwater inundation problems  and of riverine area  flooding problems.   The general
 locations  of discharges  of wastewater treatment  facility effluents  to surface water
 systems are likely to be influenced by overall  land-use patterns,  and the nature  and
 density of land  use  will affect the quantity and quality of  pollutants received by
 and  passed through such facilities.   Similarly,  the  type and quantity of pollution
 from diffuse  sources  and the  points of entry of  these  pollutants  into stream  systems
 are  also  largely determined by  existing  and future land-use  patterns.  Engineering
 and  planning  studies  for amelioration of existing  surface water quality  and quantity
 problems and  for the  avoidance  of  future  problems  should be  undertaken within the
 framework  of  an  agreed-upon areawide  land-use  plan.  To  do otherwise  is  to ignore  the
 interdependence  of land  and water  resources.

           Because  the  land-use  control issue is so central  to  the  implementation of
 any urban  runoff control  scheme,  this writer can find no realistic alternative to
comprehensive  planning and policy  coordination at  the metropolitan  level.  Because we
have only  a handful of areawide,  general-purpose metropolitan agencies with operations
authority,  for nearly all metropolitan areas the baton must  go by default  to their
 regional planning  agencies.
                                       - 24 -

-------
Disparities in Economic Evaluations

          Implementation of plans ultimately hinges on political acceptability,
which in turn depends on economic justification.  Explored in this subsection are
disparities in measures of economic efficiency employed,  and again the regional
planning agency appears to be the most likely agent for reconciliation.

          A review(56), of j-^e report^?) from a symposium sponsored by EPA concluded
that, because of the emphasis in the current Federal water pollution control program
on waste control at the source, it appears that environmental decisions  will be  based
on tests of cost-effectiveness and noted that EPA is inclined in that direction.
Symptomatic is the agency's proposed mandatory requirement for the use of value-
engineering in certain wastewater projects.'")

          From a conservationist point of view, there are evaluation hazards in
attempting to assign monetary values to what are usually regarded as non-resources,
such as endangered animals and plants,(59) which is one of the reasons why a cost-
benefit approach finds little popular support as a basis for environmental protection
justification.

          Such terms as "public welfare," "social well-being" and "quality of life"
encountered energetic discussion at a recent international urban water workshop where
agreement on terminology appeared to be better for the concept of "the good life" and
fairly definitive on basic human and social needs requisite to survival.v^O)

          Because aesthetically-oriented water uses can in many ways be incompatible
with utilitarian uses, conflicts have arisen in multiple-objective water resource
management.  There has been a definite trend in recent years towards greater legal
recognition of aesthetic values associated with natural processes, especially with
regard to water, in essentially all areas of the law, including water quality protection,
allocation and planning, and authorization of projects.v"l)  An outstanding example is
the environmental impact analysis requirement of the National Environmental Policy Act
of 1969.

          Because both facts and value judgments are involved, environmental decision
analysis is both economic and political in character, conditions which are said to
lead to the conclusion that models for environmental planning can be no more definitive
than the political processes to which they provide information on the consequences of
alternative decisions.(*>2)  Thus, cost-benefit analysis is truly helpful only in
handling the monetizable assets and liabilities, the economic effects but not the
environmental and social effects, although some of the latter can be assigned subjective
market values.

          Federal flood control planning proceeds essentially on a river basin basis
and uses cost-benefit analysis for project justification.  Thus, we find in urban
runoff management a tendency towards cost-benefit evaluation for water quantity and
a tendency towards least-cost evaluation for water quality.  The traditional approach
for water project evaluation by local governments has been minimum costs, which should
facilitate the water quality side.  However, because of the trend towards the blending
gf river basin planning with metropolitan planning, local agencies will  increasingly
face the need to reconcile the dichotomy of project evaluation criteria.  A  problem
that local governments have always had with cost-benefit analysis is  that, being a
test of economic efficiency, identification of neither beneficiaries nor bearers of
costs is included, whereas the issues of "who benefits" and "who pays" are  central
feasibility questions  in local government projects.


                                       - 25 -

-------
           Disparities in economic bases for project evaluation would seem to suggest
 rather strongly that planning for management of urban runoff quantity and quality should
 be conjunctive to avoid absolute confusion.Federal policy with regard to both flood
 control and runoff pollution abatement in urban areas is biased towards non-structural
 remedial measures, underlining the need for conjunctive planning.   In sum, and  to
 repeat the allegation at the beginning of this subsection,  the metropolitan planning
 agency appears to be the roost likely agent for the reconciliation  of disparities  in
 approaches to economic evaluation mandated by a tangle of Federal  and State lavre.

 Water Reuse

           Water conservation has often been emphasized as one of the advantages of
 wastewater reuse in urban areas.(63,64)  whereas reuse of reclaimed  wastewater  for
 domestic purposes is still controversial,  recycling for industrial uses is not.
 Furthermore,  the requirements of P.L.  92-500 for wastewater treatment in the 1980's
 are expected  to result in such stringent controls on the disposal  of effluents  into
 receiving waters that recycling via groundwater recharge and augmentation of
 industrial supplies may become more cost-effective than one-time use.

           As  local shortages and conflicts over the enlargement of facilities continue
 to grow,  more serious attention to reuse in more places can be expected;  but this
 attention is  now only beginning to emerge.

           There is considerable internal recycling of process  water  by  industry.
 Irrigation and cooling water for industry  are  essentially the  exclusive direct  reuses
 of treated municipal wastewater,  a practice that is  not widespread.(")  Because  of
 its  location  in a semi-arid  region where the bulk of its  water roust  be  imported,  the
 City of  Los Angeles has  been increasingly  leaning towards large-scale reclamation for
 uses such as  aquifer injection as  a seawater barrier,  surface  spreading for water supply
 augmentation,  industrial cooling,  and  park irrigation.   In  1973  it appeared that  the
 only instance of  a carefully engineered system for water  reuse in  a  major U.S.  cit
 was  in the wastewater portion of the City's legally  adopted  master plan.   Figure
 shows the general  types  of reuse that  have  been considered  for Los Angeles.

           Los  Angeles practices  surface spreading of storm and imported water in  an
 area to  the north,  to replenish water  pumped as  part  of the  supply from aquifers
 normally  affected  by river waters.(66'   The hundreds  of recharge basins on Long Island,
 New  York,  are  very effective  in  offsetting  the  effects  of urbanization  on the
 groundwater supply,  according to water  budget studies made of  the  catchments for
 three basins.(&/)   Sooner  or  later,  stormwater will more commonly be reclaimed as a
 source of water supply,  probably mostly through  recharge of  groundwater aquifers.
 However,  in prior-appropriation  doctrine States,  legal  questions may arise when urban
 runoff is captured  and put to a beneficial  use.(^'  There are other formidable
 management  obstacles  to  water  reuse  and conservation,("°' and  groundwater  sources are
 incompletely developed  in  a number of metropolitan areas.(™)

          We can conclude  from  the observations above that reuse of wastewater
 treatment plant effluents  and  reclamation of surface  runoff  is more likely  to be via
 artificial  recharge  of groundwater supplies, in  general.  Where  this occurs, the water
 supply-through-wastewater  treatment  cycle will interconnect with the stormwater
 disposal cycle (an  interconnection shown in Figure 1 of Addendum 1)( particularly for
 combined  sewer systems,  through the  long-term storage replenishment of groundwater.
This eventuality can be  provided for only through conjunctive planning.
                                       - 26 -

-------
                         RESIDENTIAL
                             USE
WASTEWATER
FRESH
WATER
SUPPLY
                                                    PRIMARY
                                                   TREATMENT
                                                   SECONDARY
                                                   TREATMENT
                          PARK
                        IRRIGATION
                        AND LAKES
                    TERTIARY
                    TREATMENT
                                 GROUNDWATER
                                 REPLENISHMENT
                    SEAWATER
                    INTRUSION
                     BARRIER
                                                      OCEAN
                                                     OUTFALL
               FIGURE 2-USES FOR RECLAIMED WATER

                             - 27 -
                                                  (M)

-------
           Considerable capabilities exist for analysis of aquifer performance^!)
 and for monitoring groundwater pollution.(72)  Also, knowledge on groundwater
 pollution has greatly expanded in recent years,(73) partly because of national concern
 for control of water pollution and protection of public water supplies,

 Extraneous Sewer Flows

           Whereas the preceding subsection dealt with beneficial aspects of
 groundwater manipulation, the entrance of extraneous flows that include leakage of
 groundwater into sewers is a liability.

           Plows of stormwater and infiltrated groundwater into wastewater and
 combined sewer systems may considerably exceed domestic flows where poor sewer
 construction and illicit connections are prevalent.(74)  Results from a survey of
 communities in the U.S. and Canada(75) indicated that such infiltration and inflow
 problems are widespread.  More stringent Federal pollution abatement requirements
 (P.L.  92-500) and related increased investment in wastewater treatment facilities
 have necessitated consideration of ways to diminish these extraneous sewer flows  in
 order to reduce the capacities of new or enlarged treatment plants.   Specifically,
 applicants for treatment works grants must demonstrate that each sewer system
 discharging into the treatment works is not subject to excessive infiltration and
 inflow.(76)  xt has been calculated that almost $5.3 billion will  have to be  spent
 to  correct inflow and infiltration problems nationally.(77)  Diminishment of
 extraneous flows would also increase the effective carrying capacity of wastewater
 sewers,  which in some cases would offset future growth in wastewater loads.   Under
 certain  circumstances,  auxiliary capacity could be added  by installing flow-smoothing
 basins at key locations in a system,  as an alternative to the provision of relief
 sewers.(78)

           Considerations for the conduct of infiltration/inflow evaluations have
 outlined.(79,80)

           The connection between stormwater and wastewater is  obvious  for combined
 sewer  systems,  in terms of burdens  of infiltrated  groundwater  that must  pass  through
 wastewater treatment  plants.   Perhaps  less  obvious  is  the fact that  the  promotion of
 consolidation or regionalization of treatment  facilities  to  achieve  economies  of
 scale  may  prove  to  have been a poor policy  when such facilities  are  later called  upon
 to handle  attenuated  stormwater flows  from  combined  and separate sewer systems  from
 dispersed  storages.   Thus,  schemes  for optimizing  the  regionalization  of  wastewater
 treatment  plants(81»82) should be modified  to  accommodate wet-weather  flow
 considerations where  their joint  treatment  at  central  facilities is  projected.
 Further, while regionalization of wastewater collection and  treatment  has  been
 encouraged by various Federal  and State  regulations, diseconomies of scale are
 encountered with increase  in collection  system size, and  minimum-cost  system size is
 highly sensitive to the density of  the population  in the  area  being  served.(83)

          At a national  Symposium on Regional  Solutions to Regional Problems held in
 Portland, Oregon, in  1976,  concluded was  that because  the majority of  publicly-owned
wastewater treatment works  will be  committed by  the  time  Section 208 plans are ready
 for review and implementation, only two  questions will still be open by that time
nonpoint sources  of pollution  and growth management and who pays for growth.(%<*)
Growth management complexities are  compounded by the proliferation of restrictive
 zoning, moratoria on  sewer  or water connections, building bans and dedication
ordinances, adopted as measures to  slow  growth rates and  control the location of
development.(85)  Although  a variety of  motivations have  spurred these actions, local


                                       - 28 -

-------
growth control is viewed as a unifying theme.(86'  Some of the antigrowth mood can be
laid to the higher costs of unlimited urban sprawl,  A study for the U.S. Council  on
Environmental Quality explored the cost of urban sprawl.(87}  On the basis of
quantified results from a supporting analysis of energy consumption related to
alternative metropolitan futures for the Washington,  D.C., area, it was concluded
"that the present concern for land management or growth management can be widened  to
total resource management, including consumption of energy".(^8)  We can expect
increasing pressure for the integration of energy costs into any metropolitan study
of urban water resources versus land use.

Erosion and Sedimentation

          Clearance of land for urban construction exposes it to rapid erosion.'""'
In 1965, Montgomery County, Maryland, adopted the first sediment control program in
the nation.(90)  A Statewide Sediment Control Law was adopted by the Maryland
Legislature in 1970.  Every project involving grading requires a permit from either
the local government or the State Department of Natural Resources.(91'  Each County
and municipality serves as the primary unit of government for administration,
inspection and enforcement for sediment control, the statewide program utilizes the
U.S. Soil Conservation District in each County as technical advisor for erosion and
sediment control, and the program is fully integrated with the State's pollution
control activities.(92)  The statewide sediment control program is gradually being
expanded to include storm water management.  Preliminary evaluations of "individual
site" stormwater detention storage, such as mandated for erosion and sediment control
in Maryland, suggest that such dispersed storage might aggravate downstream flooding
and that a regional approach to stormwater management might be superior.(93)

          Of particular significance here is the fact that erosion and sedimentation
control is one of several water quality issues that cannot be divorced from their
causative driving force, urban runoff.  This is one more reason why conjunctive
quantity and quality planning for urban runoff control is needed.

          A manual has been developed for use in connection with the master plan for
control of flooding and drainage in Fairfax County. Virginia.(94)  Some of the
features in the manual have been summarized.(9-*>9 ^

          For the person concerned with erosion and sediment control there are a
number of important references available.(2,97-102)

A Metropolitan Issue

          Reference is made to Figure 2 in Addendum 1, which shows the annual
quantities of water passing through a hypothetical metropolitan area of one-million
inhabitants.  Total wastewater solids for the million residents would be on the order
of 100-tons/day, dry weight, most of which would be removed in conventional wastewater
treatment.  However, sludge is often transported to disposal sites as a slurry.  Costs
of transportation and handling can be reduced by slurry volume-reduction through
concentration or "thickening" of sludge solids.  Based upon the maximum concentration
that could be hydraulically transported for the hypothetical metropolis producing
100-tons/day, dry weight,  the minimum slurry load would be about 400-tons/day.
However, normal practice, which reflects reliability of operation  as well  as  cost-
attractiveness, would result in about a 1,000-tons/day slurry for  the hypothetical
metropolis, or in the neighborhood of half the weight of  solid wastes handled per
day.  As opposed to solid waste disposal, the disposal of wastewater sludge  encounters
two monumental liabilities:  an extremely large water content .and  a very high organic


                                       - 29  -

-------
 composition.   As  a result,  the feasible avenues  of  disposal  generally differ  from,
 and often are more restricted  than,  those  for solid wastes.   In sum,  handling water
 volumes  is only part  of the problem,  for once solids conveyed by water are removed
 they must be  disposed of somewhere.(69)

           A number of major cities,  such as  New  York and  Philadelphia, transport
 wastewater sludge for disposal outside  their metropolitan areas.  With many of  them
 on  a desist notice to comply with new environmental regulations, they are very  hard
 pressed  to find acceptable  solutions  for alternative disposal.   Wherever urban  runoff
 is  captured for treatment in combined sewer  and  storm sewer  systems in the future
 the sludge disposal problem will be magnified.   Even if the  added annual amounts
 captured  were only about one-tenth  to one-quarter more, their episodic and
 unpredictable accumulation  (because of  the random nature  of  storm occurrences)  would
 tax the  scheduling and operation of sludge disposal systems  that otherwise handle
 fairly uniform loadings throughout  the  year.

 Conclusions

           Explored have been some of  the more cogent  arguments  for comprehensive
 urban runoff  control  planning.  Additional arguments  will  surface in  succeeding
 Sections,  particularly in the  next Section on conjunctive  planning for flow
 management.   There appear to be very  few local government  plans extant that have
 integrated water  quantity management  with water  quality management.   Perhaps  the
 most we can expect, given the  institutional  constraints on planning in metropolitan
 areas, is  an  integrated,  or  comprehensive, or systems approach; but such an approach
 is  subtle  in  a  documentary sense, and its existence would probably be difficult to
 detect from the outside.  Optimistically, much more rational  conjunctive planning
may be going on than  external  indications seem to imply.

          As an experienced observer of the  local water resources operating agency
scene, the writer  is  impelled  to note that any inconsistencies, conflicts and
duplications between  and  among Federal and State laws, policies, guidelines and
regulations must be and are, reconciled, arbitrated and resolved mostly at and by
 the local level of  government.  For this reason,  among several, local governments
are forced to define  their own problems.  Numerous examples abound of dire
consequences where  this responsibility has been abrogated.
                                       -  30 -

-------
                                     SECTION 4

                        SOME FLOW MANAGEMENT CONSIDERATIONS

Synopsis

          This Section is essentially a continuation of the previous Section on
"Incentives for Comprehensive Planning," and has been given separate billing because
the issues involved require a comparatively greater exploration.

          Integrated consideration of convenience drainage systems and urban flood
plains has long been recognized as a logical and efficient basis for their
management,  tore recently, the logic of integrated or conjunctive consideration of
runoff quantity and quality has become widely apparent.  Impeding these ideals is a
fractionalized authority in nearly all metropolitan areas, with responsibilities
for drainage, flood control and their associated water quality aspects splintered
among numerous units of local government.  Additionally, in many cases there are
State and Federal agencies exercising specialized authority.  Given this institutional
patchwork, about the only avenue for conjunctive consideration of these otherwise
disparate issues is via comprehensive planning at the metropolitan level.  While
institutional reform has had strident advocates for decades there have been only rare
indications of wholesale change.

          Asserted in this Section is that the guiding principle should be to reduce
the liabilities and to increase the assets of urban runoff.  Some of the more
promising ideas for accomplishing this are cited.  However, the reader is cautioned
that very few of these concepts have been tested in full-scale system applications,
the research on which many of them are based is fragmented and the findings cannot be
sufficiently generalized for universal application.  Here, as in later Sections of
this report, local governments are advised that they are mostly on their own in the
conduct of the research needed to rationalize fiscal commitments for improved resource
management.

Drainage Versus Flood Control

          Geohydrologic processes have formed natural drainage channels that convey
storm waters to the seas.  Because of an early dependence by commerce on water
transport, most large metropolitan areas originated as urban centers on or near
streams, lakes, estuaries or seacoasts.  Intrusion of urban development on natural
flood plains has resulted in damages to occupying structures and sometimes loss of
life.  Subterranean systems of conduits facilitate human occupancy by draining
sheet-flow runoff from the land surface.  That is, fluvial drainage areas contributing
to urban flood-plain inundation are often of gigantic size compared with individual
underground conduit catchment areas which rarely exceed 10-square miles in extent.
Thus, underground drainage systems are more of a convenience or amenity than a
preserver of public safety.  Structural means for mitigating flood-plain inundation
are designed to provide a much higher level of protection than that for storm
drainage systems because of the much greater threat to human life and more apparent
community-wide economic implications.

          The function of underground drainage conduits is to remove storm water from
urban surfaces (except combined sewers, which in addition convey wastewater on a
perennial basis).  The smallest catchment area (on the order of an acre in size) is
that tributary to a street inlet.  Flow in storm and combined sewer systems is
principally by gravity.  Like natural drainage basins, smaller sewer branches unite


                                      - 31 -

-------
 with larger branches, and so on, until a main sewer is reached.  Thus,  a main sewer
 not only transmits upper reach flow to a receiving watercourse but serves as  a
 collector of surface runoff all along its route.

           Whereas there is a continuum between the subsystems of water  supply,  water
 use and wastewater reclamation (Figure 1, Addendum 1), stormwater has been historically
 regarded as purely a negative good or nuisance and its subsystem (Figure 4, Addendum  1)
 has seldom been deliberately connected to the other urban water subsystems.

           Urban drainage facilities are generally owned,  operated and maintained by
 local governments, and designed and constructed by local  governments and private
 land developers.  Human life is seldom threatened by the  flooding of these facilities.
 Because the principal local detrimental effects of flooding are damage  to the below-
 ground sections of buildings and hindrance of traffic,  the consequences  of flooding
 range from clearly assessable property destruction to annoying inconvenience.   It
 follows that provision of complete protection from flooding can only rarely be
 justified.   Instead,  facilities are designed  which will be overtaxed infrequently.
 However,  because of the marginal level of protection afforded,  storm drainage flooding
 damages are also of considerable magnitude, probably equalling those in  urban flood
 plains.  In addition, indirect damages from local drainage flooding are  much  more
 extensive than for stream flooding and generally recur more often, and direct damages
 are usually much more widely dispersed throughout a community.

           Because urban drainage and flood plains usually are the responsibility of
 separate  and different jurisdictions,  their interconnected behavior is the responsibility
 of  neither  type of organization.   Thus we find  storm sewers  designed for common storm
 events  without an investigation of their  expected performance during rare events, and
 flood plain improvements  selected on the  basis  of rare events but with little
 investigation of flood damages  during  more common events.   This  dichotomy is  epitomized
 by  a reliance on local detention storage  for  common  storm events  and the unexplored
 discounting of the possible effects  of such storage  on receiving waters  during rare
 events.

          The safety  of people  and properties occupying flood-prone areas  is  the
 concern of  every level of  government,  and  Federal, State,  special districts and other
 local government units are involved  in the development of  an  ever-greater arsenal of
 remedial  or mitigative measures  and  policies.  While storm drainage facilities and land
 use  regulation are provided  by  local governments, development of  most major stream
 flood-management  works  is  undertaken by national  agencies.

          Flood  control, drainage and  the  quality of receiving waters are all closely
 related.  (Refer  to Section  3,  "Incentives for Comprehensive  Planning,"  for a review
 of current  trends).   Frequently overlooked is the  fact that precipitation cleanses
 the  land  surface.  However, because pollutants  together with  aesthetically objectionable
 materials are washed  off the land and  transported  to receiving waters in runoff, the
 result  is merely  a  transfer of  land  surface pollution to water pollution despite the
benefits accruing  to  the land.  Considering that  urbanization  increases   the rates and
volumes of  runoff delivered  locally  to receiving waters,  it is evident that the
 conveniences of  surface cleansing and  land drainage are obtained at the  expense of
higher stages and  greater  pollutant burdens in receiving waters.

Flood Aspects

          One out of  six acres of urbanized areas  is in the 100-year natural flood
plain.  There are about 20,000  flood-prone communities and some 16,500 square miles


                                       - 32 -

-------
of urban flood plains in the nation, an area equivalent to the States of Maryland
and Hawaii combined or Massachusetts and New Jersey combined.  Mare than one-half of
the nation's flood plains in urbanized areas have been developed, an area of 8,800
square miles or more than the entire State of Massachusetts.  These estimates,  based
on a 1973 evaluation,'*03' appear in the proceedings of a 1976 seminar on nonstructural
flood plain management measures.(104)

          A supporting study(105) indicated that slightly more than half of all
federal investments in structural flood control works for flood damage prevention can
be credited towards flood protection or reduction of flood damage potential in urban
areas.  While average annual urban area stream flooding damages are not known,  the
U.S. Water Resources Council estimated the rural-urban total at more than $1.7-billion
for 1966, of which the urban portion probably would not be much more than $l-billion.
In contrast, the American Public Works Association estimated 1966 average annual
flooding damages in sewered systems as being in excess of $l-billion.'*^ '  The major
portions of urban areas are served by underground systems of storm drainage.  Thus,
somewhere around half of all urban annual flooding damages occur on the large segment
of urban land outside of 100-year flood plains, a matter of considerable importance
to the National Flood Insurance Program.

Conjunctive Planning

          There was a time when stream flood plains and sewered drainage systems could
be considered separately, and when urban runoff could be regarded as merely an adjunct
consideration in land-use planning.  New public demands and policies require widespread
use of true comprehensive planning, which includes integrated land and water management.
The guiding principle should be to reduce the liabilities and increase the assets of
urban runoff«<.1(^0  The National Flood Insurance Program has been supported by
investigations and analyses concerned mostly with stream flood-plains.  It is to be
expected that the scope will greatly enlarge as more communities become eligible for
insurance against storm sewer flood damages.  There will be severe difficulties,
alone, in defining what is "improper" drainage or "inadequate" storm sewers.
Determination of criteria for defining "construction and land use practices that will
reduce flooding" will be quite difficult for several reasons:  (1), techniques for
hydrologic analysis of rivers and urban streams are at a stage of verification far
beyond that for local drainage, mostly because of a much broader data base for the
former; (2), designs by municipalities, developers and consulting firms, and tools
of analysis developed by universities and other research organizations, have
predominantly emerged as an assortment of unconnected, independent products; (3), the
potential transferability of innovative plans and designs has been drastically impaired
because a range of alternatives was very rarely explored in any given case and a
bias of specialized site conditions was generally suspected; and (4), new requirements
for protecting the quality of water supplies and for abating the pollution of water
bodies add a complex dimension rarely considered in past development projects.

          All the basic engineering methods for in-stream flood regulation are
applicable to flood mitigation works within local urban catchments, but generally
on a smaller scale:  acceleration of flood flows by canalization through threatened
reaches with consequent reduction in stages; isolation of contiguous land in
threatened reaches from flood flows by means of embankments, such as levees and
flood walls; and the attenuation of flood peaks by means of storage, located either
upstream from threatened reaches or at lateral points fed by diverted flows.
Similarly, possibilities also exist for social-economic-administrative-legal controls
to preclude some degree of damages, such as flood-plain zoning, flood-proofing of
structures, flood insurance and related schemes.


                                      - 33 -

-------
          Increased volumes of direct runoff from underground drainage conduits
clearly can aggravate flooding of urban flood plains.  On the other hand, increased
receiving-stream stages can cause or induce flooding of underground drainage systems
because of the intimate hydraulic linkage between them.  Extensive programs of
integrated flood plain management are particularly crucial for metropolitan areas
that have lttle toorahic relef  such as  reater Chicao '^°'    e
 that have little topographic relief, such as greater Chicago.    '  There are large
 cities that early dedicated most of their flood plains over to parkways, such as
 Philadelphia, or developed extensive systems of major drainage channels in step with
 urban development, such as Los Angeles. (109, 110)  Regardless, new suburbs have more
 than occasionally aggravated flooding with subsequent induced damages in many major
 cities located on large streams or bodies of water because their low topographical
 elevation makes them hydro logically subservient.

           It has been suggested that for a growing metropolitan area the thrust of
 drainage and flood control solutions should be in two basic directions:  preventive
 activities in the form of flood plain management together with good planning; and
 remedial actions where flood plains have been improperly occupied and developed and
 where local drainage problems have not been adequately considered and handled. (mJ
 Preventive activities and remedial actions can involve both structural and non-
 structural considerations.  Structural components include storm sewers, inlets, curbs
 and gutters, culverts, and channelization and detention facilities; and non-structural
 activities include flood plain management, flood plain warning and flood insurance.
           Environmental corridors in southeastern Wisconsin encompass  about 18  per
 cent  of the total area served by the regional  planning commission,  486 out  of 2,700
 square miles. (H2)  Asserted is  that riverine  areas  should be viewed as environmental
 corridors  managed in such a way  as to minimize flood problems and  simultaneously
 satisfy a  variety of noneconomic human needs. (H3)  In concept,  an environmental
 corridor is essentially a continuous linear pattern  in the landscape consisting of  a
 composite  of natural resource and natural  resource-related elements that are important
 for maintenance of overall environmental quality.  This concept  is  partly founded on
 the demonstrated unsuitability of riverine areas  for urban development, and favors,
 instead, multiple-purpose floodland management in  terms of flood damage mitigation
 and environmental corridor protection.  From this  perspective, it  is impossible to
 conduct planning separately for  streamflow quantity  and quality.

           Considerations  in water-oriented amenities in the urban  environment have
 been  evaluated  and tested in an  analysis that  included case studies of:   Harrisburg,
 Pennsylvania; Waterloo, Iowa;  Houston and  San  Antonio,  Texas;  Sacramento, California;
 Minneapolis-St.  Paul,  Minnesota;  and the two new  towns  of  Flower Mound,  Texas,  and
 Columbia,  Maryland/114'

 Detention  Storage

           Discharges  from conventional  storm drainage  sewer facilities  and  flood-
 plain  intrusion  by structures  both tend to  aggravate flooding, and  thereby  jointly
 tend  to raise the potential  for  stream  flooding damages.   Revising  storm sewering
 criteria,  such as  by  including much more in-system storage,  can be  an effective
 adjunct in flood  plain management.  While  there is universal agreement  that  the
 planning and development  of  drainage  systems and flood-plain management  programs
 should be  coordinated and  integrated, and  there is evidence  that such efforts are
 increasing, prospects for  accommodation may tend to  diminish rather than  improve
because of an increasing  concern over water quality  considerations  in sewered systems
 and a contemporary neutrality  or  indifference  on water  quality matters by some agencies
                                      - 34 -

-------
dealing predominantly with flood plains.  Often overlooked is that much of the
flood-plain flooding problem as well as the land runoff water quality problem could
possibly be more effectively countered on the land feeding urban watercourses.

          While the advantages of local detention storage in lieu of the traditional
rapid removal of storm flows have long been recognized,^115) such storage has been
only occasionally employed as part of overall flood mitigation, as in Denver'11"'
and Fairfax County (Section 2).  However, specific projects attempting to derive
general storage-requirement parameters have heretofore been necessarily site-specific,
insensitive to storm actualities, limited to one or very few storage options, and
usually have ignored water quality considerations.  For example, while an excellent
study in its own right, the research by North American Rockwell^11?) typifies these
limitations.

          Findings from a general study of stormwater detention usage have been
reported.(H8-120)  Documented in another study were management techniques in the  use
of detention storage.(121)  While the principle of the use of local detention storage
is often championed in lieu of main channel improvements for metropolitan areas,
sharp differences of opinion have been known to arise, partly or perhaps principally
because implementation of the two methods commonly falls within different
jurisdictions of authority.

          There is evidence which indicates that the effects of urbanization on the
magnitudes of streamflows may decrease as rarer floods are approached,'122) to the
extent of a difference that may be undetectable or at least relatively insignificant
at the 100-year level.  That is, the effects of land use changes due to urban
development are more pronounced the more common the occurrence.  Underground conduit
systems for urban drainage and adjunct detention storage are deliberately designed
to be overloaded rather often (such as once in 5 years, on the average), whereas
waterways flooding protection is usually much greater.  Because storm sewers drain to
waterways, they are obviously connected and their performance  is related.  Mare
attention should be accorded the interrelations between land runoff and streamflow,
over a range of occurrence frequencies, in a conjunctive planning sense.  For example,
plans for upstream local detention storage for flow attenuation and/or pollution
control might be found to aggravate unduly streamflow rates and/or pollutant burdens
in a given instance.  A study in the Atlanta, Georgia, area indicated that small
detention storage basins are effective  in holding runoff from  newly developed areas
to their former peaks under natural conditions but become progressively less
effective with increasing watershed size,(123)

          A computer-based procedure for application  in comprehensive flood  plain
information studies'   ) is  in advanced stages of development.  Accommodated  are
land use changes and interrelations between conduit systems, detention storages and
streamflows from the standpoint of both quantity and  quality.

          Through use of simulation, it  is feasible to analyze the  influence  of
alternative urban development on stream hydraulics and  to assess  expected  flood
damages.  Elucidation of incremental increases  in average annual  flood damages  are
an effective means for demonstrating flood damage effects of arbitrary or
indiscriminant urban development.C125)  Also,  there are ways  to analyze via
simulation the influence of  alternative urban development on  storm water  quality.^   '

          Whereas cost information on  local detention storage  has  not been
comprehensively assembled, a manual  is  available  for  estimating construction and
0 & M costs for combined sewer overflow storage and  treatment.^12")  Also,


                                       - 35  -

-------
 some useful data has been assembled on costs of traditional drainage(1^7) an(j
 storage capacities.^ °'

           Urban lakes have been a subject of concern to the U.S. Geological
 Survey(129,130) and the American Geophysical Union. (131)
 Other Measures

           New national urban planning priorities include:  enhancement of urban
 environments; conservation of water resources; and reduction in water pollution.
 These new priorities require use of stormwater management practices that are much
 more comprehensive and complex than generally used at present.  Research on desirable
 practices has been fragmented and diffused, a severe handicap in making deliberate
 efforts to encourage abandonment of historical practice,  where urban settlements  have
 been drained by underground systems of sewers that were intentionally designed to
 remove stormwater as rapidly as possible from occupied areas, with little regard  for
 resultant receiving-stream flooding and ignorance or oversight of lost environmental
 enhancement opportunities.

           There are at least four basic methods for reducing sewered system flood
 peaks and controlling runoff pollution:   (1), retardation of flows by induced
 friction; (2),  control of flow paths and gradients by grading; (3),  induced
 infiltration of stormwater into the ground; and (4),  provision of detention or
 retention storage.  All of these are more effective the nearer they are to the
 sources (individual properties) under the conditions  for  which they are designed.
 However,  as noted in subsections above,  during rarer  runoff  events the benefits of
 such means may  decline substantially by the time  flows reach downstream waterways  and
 other receiving bodies of water.

           Alternatives have been conceptualized that  would cause  a reduction in
 quantity  or improvement in quality of urban runoff, yet could be  implemented by
 investment in other than the construction of major new facilities.  Among the
 possibilities are the following: (132)

                     Source Control Alternatives
                       Roof storage
                       Ponding
                       Porous pavements
                       Erosion control
                       Street cleaning
                       Deicing methods
                      Utilization of natural  drainage  features

                    Collection System Control Alternatives
                       Sewer flushing
                       Inflow/ infiltration
                      Sewer cleaning
                      Polymer injection
                      In-line storage
                      Remote monitoring and control.

These and other alternatives are discussed in an EPA report. (133)  A. "desktop"
procedure  is available for  the comparison of  selected alternative control technologies
for stormwater management.
                                      - 36 -

-------
          "Urban hydrology is one field which has failed to attract major Federal
support in spite of strenuous efforts of professional groups such as the Urban Water
Resources Research Council, ASCE.  The development of acceptable ways to measure
water quality in urban streams and storm sewers has been even more neglected.   The
Federal data-gathering program has concentrated on the larger streams, presumably
because of their traditional use in formulating basin plans.  There has been a
tendency to leave urban data-gathering to the municipalities themselves.  Even so,
Federal research funds have been available only to a very limited extent."(135)

          Thus, the reader is warned that although numerous schemes have been
postulated for controlling the quantity and quality of urban runoff, very few of these
concepts have been tested in full-scale system applications, the research on which
many of them are based is fragmented and the findings cannot be sufficiently
generalized for universal application.  By default, local governments are obliged to
assure themselves of the relevance and reliability of most such schemes before making
a commitment to their wholesale use.  One way to investigate expected performance is
the employment of hydrological simulation methods founded on local field data, the
subjects of the remainder of this report.

Miscellaneous

          A very useful compilation of 800 references has been prepared that emphasizes
nonstructural measures for flood damage abatement, flood insurance and
floodproofing.(136)  Citations span 1928-1976.

          A study of a 29-acre urban catchment in West Lafayette, Indiana, investigated
the economic and environmental impact of alternative drainage systems for varying land
use and population density.v"7)  Drainage alternatives included different levels and
kinds of runoff treatment plants and either open channel or closed conduit drainage,
but consideration of detention storage was not included.  However, the techniques for
cost evaluation of single-family versus multiple-dwelling versus commercial development
are potentially useful for applications elsewhere.

          Several drainage ordinances, including those in Fairfax County and
metropolitan Chicago, have been, examined in terms of the effectiveness of their
specific provisions and the problems encountered in administering them, and a model
drainage ordinance based on the survey was promulgated.
                                      - 37 -

-------
                                      SECTION 5

                                UTILIZING SIMULATION

 Synopsis

           While simulation is undoubtedly an effective means for analyzing
 alternative urban runoff control strategies, its most important use can be in  the
 assessment of expected system performance.  Local government officials  must be able
 to answer pointed voter questions on the exactness of cost estimates versus the
 degree of confidence that may be placed on proposed facilities  and  management
 programs to meet specified abatement goals.  Performance reliability is ultimately
 the fulcrum of political acceptability.

           This Section commences with reasons for using simulation  and  a historical
 reckoning of practical reservations against incautious acceptance of runoff models.
 Emphasized is that simulation techniques adopted should not exceed  the  level of
 mastery of such tools by the user and that tools should be selected on  the basis  of
 their suitability for solving defined problems.

           Performance reliability is defined in  terms of four serially  connected
 principal considerations,  which must be completed in  the following  order to be fully
 effective:   testing simulation models against local  field  data;  using a long-term
 precipitation record as  a  reference; developing  frequency  rationales for levels of
 protection;  and simulating a range of storms and system loadings.   The  first two  of
 these four steps are more  or less inconveniences that must be accepted  to  get  to  the
 threshold of reliability required for undertaking the latter two  steps  where the
 true  payoff  resides.

          An emphasis is placed on the testing of simulation models  against local
 field data because this  issue is the crux  of credibility as  well  as  reliability.
As  things now stand,  local  governments  are substantially on  their own for  the
acquisition  of_field data.   The concept of spatial sampling  of catchments  for  field
 instrumentation having representative land uses  is outlined.  Stressed  is  the
importance of rapid application of data to models, to check  data  reliability but
principally  for expeditious  calibration of models, the primary use of such  data in
planning.

          Control  of  flooding and  water pollution must be  based on probabilities of
occurrence because  of the randomness  of precipitation.   Rationales are developed  for
using  long-term rainfall records  as  the reference  for defining urban  runoff quality
and quantity  control  objectives  in  probabilistic  terms.  (Reasons for using actual
records  rather  than  synthetic storms  are reviewed at  length  in Addendum  2).

          Reference  is made  to  indications  in other Sections of a history of moving
targets  for  flood management  and pollution  abatement  over  the past decade and the
likelihood that  policy targets will  continue  to  shift over the next several years.
Underscored  in  this  Section  is  that  the only realistic defense for planning in an
atmosphere of ambiguous  policy  is  to  employ procedures the results of which have
an inherent  flexibility  for  conversion  to.meet alternative policy goals.

          This Section closes with a  discussion of what was termed "simulate
performance  for  alternates"  in Figure 1, Section 2.  In runoff control planning,
simulation is mostly  for future or extrapolated  conditions.  Reliability in the
employment of calibrated tools  for such extrapolations is a function of several


                                      - 38  -

-------
things, but the most important are probably the shrewdness of design of the field
data network and the extent to which its results approach an ideal set of
representative samples.

          Most of the written record on simulation has been authored by modelers
and data collectors and reflects their research orientation and interest.  Details
over which these specialists fret and expostulate are of only passing interest to
persons responsible for developing comprehensive plans within limited budgets and
rigid schedules.  Therefore, an attempt has been made in this Section to dwell only
on fundamental simulation issues in planning and to sidestep niceties that are the
grist of the researcher's mill.  Thus, some readers concerned with applications may
not necessarily appreciate the full reasons for some of the trenchant assertiveness
of this Section.

Why Simulation?

          The name of the game is system performance.  Readers who are disinterested
in the reliability of performance of plans for projected drainage can skip the
remainder of this report.

          All but a small fraction of storm sewers here and elsewhere in the world
have been sized by means of wholly empirical methods.  Given a lack of evidence of
superior methods, these overly simplistic procedures proved adequate when the primary
purpose of storm sewers was to drain the land and express the accelerated convergence
of surface runoff to receiving waters.  Out of sight, out of mind.  Once restrainment
or containment of flows and their pollutant burdens become added primary objectives,
traditional procedures of analysis are no longer adequate because of added system
complexities for which conventional tools are unsuited.

          Why not use observed discharge variations as a guide?  There are several
compelling reasons precluding this possibility:  (1), very few urban catchments,
particularly sewered ones, have been gaged; (2), a statistical approach requires a
period of record spanning at least ten years, substantial physical changes commonly
take place on most urban catchments over this long a time, and the mixed statistical
series that results is not interpretable; (3), while such a statistical series would
characterize the existing situation, there would be substantial uncertainty over its
extension to differing future situations; and (4), the clinching reason, in the usual
case where no field measurements have been made, is that it would be necessary to
postpone planning and analysis until new long-term field records were accumulated, an
unacceptable option under contemporary imperatives.  An even less acceptable alternative
would be to rely solely on empirical tools and determine prototype system performance
after system changes had been instituted, a procedure that would indicate the overall
errors implicit in the tools used, but would be very expensive experimentation.

          Thus, in order to anticipate future system performance under changed
conditions, because these changes can very rarely be simulated by manipulating
prototype systems, recourse must be made to performance simulation by calculation
or analogy.  Once having accepted simulation as the means for analyzing future
performance, the central issue remaining is the degree of reliability of system
performance that is desirable, under what conditions and at what costs, the central
theme of this Section.
                                      - 39 -

-------
 Background

           Hydrology may be defined as the science that is concerned with the waters
 of the earth, their occurrence, circulation and distribution, chemical and physical
 properties, and their reaction with the environment, including their relationship to
 living things.  Thus, hydrology embraces the full history of water on the earth.^39)

           Because the complex interactions of human activity in concentrated
 settlements with air, water and land must be taken into account,  urban hydrology is
 a distinctive branch of the broad field of hydrology.  As opposed to conventional
 hydrology, because urban development everywhere has been in continuous states of
 expansion and flux, urban hydrology contends with the dimension of dynamic change.
 Also, urban water resources management utilizes the social and biological sciences  as
 well as the physical sciences.  Reflecting the lag in the recognition of the fact
 that America became an urban nation over half a century ago, the term urban hydrology
 gained currency less than two decades ago.  Since then, the term has been tacitly
 expanded to include all urban water resource matters in,  or interfacing with, the
 hydrologic cycle,  including water quality considerations.

           Whereas  generic hydrology is termed a geoscience,  engineering hydrology
 is strictly an applied science.  There are no truly fundamental principles exclusive
 to engineering hydrology, and this is particularly true of urban  hydrology.
 Fundamental laws of soil-water movement are borrowed from soil physics and
 fundamental equations of water motion are adapted from hydrodynamics.   For example,
 the principle of conservation of mass is applicable to all kinds  of systems,  not
 merely the water cycle.   It is important to recognize that urban  hydrology is a
 complex art,  but an art all the same,  because otherwise a more faithful replication
 of system performance would be expected from simulation than warranted by the
 attainable perfection of the art.

           Without  question,  the advent and rapid evolution of electronic computation
 has  accelerated  development of tools  of analysis in urban hydrology as in every  other
 field.  Advances are clearly evident  in a recent review.(l^D)  perhaps more  obvious
 advances  have  been made  in water supply and receiving water  hydrology  simply  because
 these are essentially suprametropolitan.   Another factor  is  undoubtedly the  fact that
 they  have benefited from advances  in  traditional hydrology that came about because  of
 national  and State imperatives.  Federal  legislation on water pollution,  flood
 insurance and  environmental  protection,  has  resulted in a great intensification  of
 urban water model  development  over the last  few years,  with  sewered catchment models
 perhaps very recently eclipsing  urban  receiving-water model  development.(1^1)

           Relatively few runoff-quality  field gagings in sewered catchments have been
made, and  these  have been mostly at outfalls.   Source quality has been investigated
 principally as a function of street surface  pollutants  accumulated  between rainfalls.
In order  to accommodate  cause-effect relationships required  for modeling,  it  is
current practice to  estimate potential  street loadings, separately  for individual
parameters, on the basis  of the  few documented  solids-accumulation  histories.
Arbitrary  allowances  are  then  added to account  for off-street contaminant accumulations
expressed  as multiples of  the  potential street  loadings.  Thus, no  direct verification  '
of the hypothesized  buildup of pollutants  and their  transport to receiving waters is
presently  available.  It  is reasoned that when  "pollutegraphs"  generated by models
reasonably approximate field observations  for a  catchment, that the overall
accumulation and transport hypothesis  is validated.  As a result, it might be
concluded  that model  development has already greatly  outstripped the data base for
model validation,  in  the  sense of bracketing probable reliability.  However,  if  field


                                      - 40 -

-------
research and model testing continue at anywhere near the level of activity of the
past decade, substantial advances in reliability appear to be an inevitable result.

          Or, as summarized in a recent national conference on urban runoff:  "When
we consider the length of time that studies have been made of treating sanitary
wastes and the principal industrial wastes, or of hydrology in general, it is
apparent that water quality analysis of urban runoff is a relative newcomer and a
neglected field".(1^2)  Much more also has to be learned on the quantitative aspects
of urban runoff.

          A recent appraisal of future modeling needs^^O) revealed that,  from the
standpoint of runoff quality simulation, progress is stymied principally by a lack
of sufficient field data to evaluate thoroughly the quality segments of various
models, let alone to evolve improvements.

          A plan for a national program for acquisition and analysis of field data
from urban sewered catchments was designed by ASCE in 1969. (^3)  Despite the best
of intent ions, (144) otliy a token portion of the proposed, critical, field gaging
portion has been implementable.  As things now stand, local governments are
substantially on their own for the acquisition of field data and no integrated or
national program exists or is in sight.

          The fact that three-fourths of our people live in metropolitan America
continually escapes the attention of framers and implementers of national policy.
In the meanwhile, fractionalized, largely independent, fretful but perhaps  impressive
progress is being made in urban hydrology research, and accelerating planning
activities nationwide imply even greater attention in the immediate future.

          Acknowledging that significant advances have been made in many aspects of
urban water resources, research on drainage has long been neglected, particularly
with regard to field data acquisition, and information needs are growing faster than
new knowledge is accumulating.  This neglect can be traced to the continuing
ambivalence of the Federal government on its role in metropolitan areas, an inheritance
from our colonial past.  One view holds that a national program of drainage research
would require the least public expenditure, because the primary objective would be
transferability of findings.  However, the prevailing view is that each of  the more
than 10,000 local units of government in metropolitan areas involved with stormwater
management should "do their own thing" in  such research.  Seemingly overlooked is
the fact that national investment in urban facilities has already constituted seven-
tenths of total investments in all water resource facilities, with its share still
growing.  While it is of small comfort, the U.S. is far from unique in this regard
because the most urbanized countries still exhibit the rural origins of their
institutions, and urban water resources research around the world commonly  has
suffered from inadequate attention and support and from discontinuous and erratic
efforts.

          Against this historical perspective is a viewpoint that deserves  quoting:
"There does not seem to be a 'perfect* model for analysis of stormwater.  The models
are either too complicated, do not allow for distributed  inputs and parameters, do
not simulate continuous streamflow, or have not been  tested extensively on  hydrologic
data	  There remains much uncertainty in stormwater modeling.  There appear
to be enough parametric models available which have been  shown to be feasible
conceptualizations of the stormwater runoff process.  What is needed now  is a
continued and accelerated verification of  the existing models and a follow-up
regionalization of the parameters,"(22)  ^11 this will take some time.  What can  be


                                      - 41 -

-------
 done in individual metropolitan areas in the meanwhile is to be discussed  in  the
 remainder of this Section.

 Performance Reliability

           Users  of the newer simulation models  are confronted with  a  seeming  dilemma.
 While the risk of inadequate representation of  prototype  systems  decreases  as more
 flexible models  are used,  the complexity of the models necessarily  increases  at the
 same time,  as  does the difficulty in obtaining  an  acceptable simulation of  expected
 performance.   A  recommended  rule is  "the premise that  the simplest  model that will
 do  the job  is  usually the best".(132'  Further,  the more  complicated  the model the
 greater the demands on the user for  a mastery of complex  hydrodynamic and
 biochemical relationships.   Indeed,  the technological  sophistication  of some urban
 runoff models  is as great as in any  water resources application.  This may  be seen
 by  reviewing the process details of  modern models.(22) There is  evidence of simulation
 failures that  have been attributed to model inadequacies  where  the  blame more properly
 belonged to improper handling by the user because  of insufficient comprehension of
 the complex processes involved.   Thus,  the first cardinal  rule  is that simulation
 techniques  adopted should not exceed the level  of  mastery of  such tools by  the user.
 Putting it  in  a  more pertinent way,  the range of simulation  possibilities can be
 constrained by the level of  skills of the user.  The finesse  and mastery of the
 technical personnel involved directly in the  simulations  can  therefore be a major
 contributing influence on the reliability of  system performance that  can be attained.

          A second cardinal  rule is  that tools  should  be  selected on  the basis of
 their  suitability  for solving defined problems.  This  is  to say that what is wanted
 from simulation  should  be defined first,  and  the selection of techniques should
 follow,  not lead,  this  decision.  Models  may  be useful as  tools in  solving problems
 but  they have  no inherent capability  for defining  them.  Further, there being no
 single or universal  model, a  stable  of  tools will be involved over  the course of
 comprehensive  planning, starting with preliminary screening models  through post-
 implementation operations models.  It should be evident that model needs should be
 projected through  the  total  planning  epoch, to assure  compatibility (at least for
 data required) and  to avoid  extravagant  costs and almost  irreversible commitments to
 particular data processing schemes.  Also, the package of tools used should permit
 the  incorporation of expected  improved versions because such changes have accumulated
 in the past in the midst of comprehensive plan evolvemenus.

          We now turn to other principal considerations affecting the reliability of
projected prototype system performance.  Four of these principal elements are
 itemized in Figure 3:

 . Testing simulation models against local field data.   Without some adjustment of
  model factors to reflect local peculiarities,  whatever they may be,  any model  is
  necessarily under a cloud of suspicion.  This  item is included in Figure 1,
  Section 2.
. Using long-term precipitation record as reference.  Water quality and quantity
  control objectives must  be defined  in probabilistic  terms,  for which this is usually
  the only way open.  This  item is included in Figure  1, Section 2.

. Developing frequency rationales for levels of  protection.  Flood levels  and
  receiving water quality  goals must  be expressed in several  ways, such as  frequency
  of events or degrees of  protection, as implied in Figure 1, Section 2,
                                      - 42 -

-------
THOROUGHNESS
     OF
  ANALYSIS
                RELYING
                SOLELY ON
                EMPIRICAL
                ESTIMATES
TESTING
SIMULATION
MODELS
AGAINST LOCAL
FIELD DATA
USING
LONG-TERM
PRECIPITATION
RECORD AS
REFERENCE
DEVELOPING
FREQUENCY
RATIONALES
FOR LEVELS OF
PROTECTION
SIMULATING
A RANGE OF
STORMS AND
SYSTEM
LOADINGS
                                /
                              //    J*
                                                        ADDED
                                                        COST
                                     ADDED PERFORMANCE  RELIABILITY
  FIGURE 3-  SUBJECTIVE  ESTIMATE OF INCREASED RELIABILITY  AND ASSOCIATED COSTS
              FOR MORE THOROUGH ANALYSIS

-------
 . Simulating a range of storms and system loadings.  This, too, is implied in Figure 1
   Unless a system is operated (albeit via surrogate simulation) over a reasonable
   range of expected conditions, how can we be assured it will perform as we expect
   and under what circumstances may it be expected to be "overloaded"?

           These elements are discussed at greater length below.  However,  before
 leaving Figure 3 there are important points that should be made.   First, it should
 be noted that the four components are serially connected.  That is,  a "range of
 loadings" depends on the "frequency rationales" developed, which depends on reference
 to a "long-term precipitation record," all of which hinges for collective reliability
 on prior "testing of the simulation models" that are used against local  field data.
 Second, as graphed in Figure 3, the first two elements probably provide  a greater
 payoff relative to their cost than the other two, although the latter are the really
 important ones from the standpoint of comprehensive planning.   That  is,  the first two
 are more or less inconveniences that must be accepted to get  to the  threshold of
 reliability required for undertaking the latter two steps. Figure 3  is  a  purely
 subjective estimate by the writer,  made in an attempt to hypothesize  a nationally
 applicable representation.  It will be some time before supporting evidence
 accumulates.   There does not appear to be a suitable way to accommodate  in Figure 3
 the question  of the use of tools  with complexities compatible  with master  plan
 objectives, discussed at the beginning of this subsection.

 Testing Simulation Models

           In  the analysis  of water  distribution systems,  the first major objective,
 prior to the  investigation of requirements  for the future,  is  the  attainment  of
 satisfactory  simulation of existing system performance.   System simulation is
 considered validated during preliminary analysis  for design when  calculated pressures
 are satisfactorily close to observed  field-gage readings  for given field water source
 send-out and  storage conditions.  "The simulation capability for  the  existing system
 having been verified,  analysis  of the  system or district  is pursued for  conditions
 expected within the  next 5,  10, 25  and possibly 50 years."(l^)  water distribution
 system simulation is singled  out  to illustrate  the fundamental  planning  principle of
 regarding present conditions  as prologues or points  of reference  for  the future.
 Good  information  on  current  actualities  is needed  to proceed confidently with
 extrapolations, where  the  degree  of reliability of projections diminishes  as
 estimates  move  into  more remote time-frames.

           Returning  to  urban  catchments,  the  "systems" involved become considerably
more  complicated.  Even if  there  are parallel similarities  between drainage conduits
and water  distribution  piping, stormwater catchments  are  characterized by  their
large number and  variegated land  use characteristics, and  to this  is added  the
complex  performance  interaction of underground conduits with receiving channels,
streams  and other water bodies.   Thus  the calibration of  simulation model variables
is much more involved.   It  is important that  the field data used be for  catchments
"representative"  of  prevailing land uses.  In principle,  identification of candidate
catchments would  be  on  the basis  of an  inventory of  existing catchments  in  the
metropolitan area, classifying them by predominant land use, size of area, extent of   i
storm sewer drainage, and location of outfall relative to receiving waters.  Detailed
guidelines for  the selection, instrumentation and processing of data of catchments
have been presented  recently.(1^6)

          Table 2(147)  lists the  principal physical characteristics of the nine
catchments presently being gaged  in the City of Philadelphia to obtain regional
model calibration data.  Note that uniformity of land use is deliberately high for


                                       - 44 -

-------
                                                         TABLE 2

                               PHYSICAL CHARACTERISTICS OF REPRESENTATIVE CATCHMENTS BEING
                                                GAGED IN PHILADELPHIA^147'
PREDOMINANT
LAND USE
Residential
Single Houses
Semi-Detached
Row Houses
Commercial
Shopping Center
Industrial
Light Industry
Heavy Industry
CATCHMENT
NAME

Linden
Overbrook
Tustin
Whitaker
Large
Mifflin
Leo Mall

Industrial Park
Erie
UNIFORMITY
OF LAND USE

86
88
100
81
98
90
100

100
100
SEWERAGE

Separate
Comb ined
Separate
Combined
Comb ined
Comb ined
Separate

Separate
Comb ined
SIZE
(Acres)

145
32
82
120
74
80
9

43
100
IMPERVIOUSNESS
(Per Cent)

43
31
63
68
84
98
100

75
76
DWELLING UNITS
(No. /Acre)

1.4
3.4
10.9
8.4
16.0
24.4
-

-
•
I
-p-

-------
 the catchments selected, to ensure a representativeness of land use types for later
 application to the Philadelphia metropolitan area of the models to be calibrated.
 There are about 250 separate storm and combined sewer catchments within the City of
 Philadelphia alone.

           For smaller metropolitan areas it is more feasible to make comparatively
 more extensive gagings simply because of the few catchments involved.  An example
 is the portion of metropolitan Rochester, New York, served by combined sewer systems.
 There, all thirteen combined sewer outfalls involved have been instrumented for
 automatic flow-sensing and water quality sampling.(14°)  The land-use features  of
 the catchments are listed elsewhere.(132)  j^e catchments range in size between 230
 and 2,600 acres,  residential occupancy ranges between 37% and 94% and imperviousness
 ranges between 35% and 80%.  Ten of the thirteen combined sewer outfalls discharge
 overflows to the Genesee River,  which is also monitored for water quality.   Considerable
 modeling work has been undertaken.(132,148)

           A number of considerations  are involved in the deployment of gaging
 stations,(146) such as in the Philadelphia case above,  which may be termed  a spatial
 network design.   The purpose of  these stations is to serve planning and design
 functions.  (The  use of such networks for real-time operation is discussed  in
 Section 7).   Specifically,  short-term records are necessary to calibrate models so
 that long-term performance simulations can be made.   That is,  rainfall variations are
 sampled over a season or two on  catchments  that are samples of predominant  land uses
 (The issue of storm variations over time in discussed in the next  subsection).

           The goal of spatial network design is to attempt an optimization  of the
 best set of  "representative catchments  sariples" that can be realistically funded for
 gaging;  and  the maximum number of station points where  measurements  of flow-quality
 can  be  made,  for  a given total budget,  is then affected by how fast  a point can be
 abandoned and measurements  at some other point can be initiated  with the same
 equipment.   Speed  of abandonment  is a function of how rapidly  the  data are used  to
 calibrate a  model  or models, and  the  deliberate rush to calibrate  may cause an  extra
 non-data cost  for  analysis  which  has  to  be  taken into account  in weighing budget
 schedules.   Because  such field work can  often  be funded  only on  a  year-by-year  basis
 the  staging  of investment  is therefore affected  by the  staging strategy  employed,  in*
 addition .to  considerations  of total project  and  yearly  project budgets.  That is,
 the  staging  over the  project should maximize results  from year to  year,  with  an
 early payoff and sufficient payoff  thereafter  to discourage premature  abandonment by
 officials with fiscal veto  power.  All this  is  to  say that considerable  judgment,
 wisdom and foresight are required.  It is obvious  that  spatial network design is
 presently more art than  science.  On  top  of  all  this  is  the restriction  that  some
 catchments cannot be gaged  at all (e.g., very  complex backwater  flooding of a site
 at a wide  flow range, such as by tides),  and some  can be  gaged only at inordinate
 expense  (e.g., because of difficulties of access).  ASCE  had proposed a  3-year national
 gaging program in 1969,(143) ancj it ^s therefore expected  that any metropolis could get
nine-tenths of its work done within 3 years.

Long-Term Precipitation Record as a Reference

          Control of flooding and water pollution must be based on probabilities of
occurrence because of the randomness of precipitation.  Reasons for using actual
records rather than synthetic storms are reviewed at length in Addendum 2.

          The primary long-term rainfall data base for sewered catchments in any
metropolis is the (usually) single U.S. Weather Service first-order station five-minute


                                      - 46 -

-------
interval data, commonly spanning fifty years or more.   It is necessary to  assume that
short-term rainfall samples from field-gaging programs lasting a season or two
comprise a representative sample of the long-term record.  To repeat,  the  single
rainfall station record is the primary (and sometimes  sole) source of  long-term data
available.  So far, there has been no successful simulation of 5-minute single-point
data via stochastic methods,  so there is no other way.  On top of the  assumption that
our precipitation field gaging is for a representative sample, we must further  assume
that the long-term single-point record would be more or less replicated over a
similar period of time at any other point in the same  metropolis.  The latter is a
generous assumption because we have proof of areal persistence only for pieces  of
storms:  i.e., maximum intensities of a specified duration.  Figure 4  shows examples
of the areal persistence of one-hour maximum average rainfalls of particular
frequencies, after Yarnell'l^'' (only six of the total of 56 charts he prepared).
The same persistence is implicit in later mappings of  the U.S.W.S., including
Technical Report No. 40(1^0)  aiuj ^n recent updates.  In addition, the  Corps of
Engineers studied the Yarnell maps and found that the  depth for any duration was an
approximate function of the same frequency one-hour duration depth for all national
station data, regardless of recurrence interval, Figure 5.(151)  it should be
carefully noted that the "isohyetals," such as in Figure 4, connect equal  values of
precipitation for a given duration and frequency; and  that these values are for
different storms at different years from one location  to another.  (Details on  such
data are given elsewhere'"^).  Thus, there is an implication of long-term persistency
from one point to another over long distances, and hence over relatively short
distances, tenuously based on the documented strong persistence of separate pieces
of storms regarded independently.

          Theory of catchment raingage network design  is relevant only for initial
placement; and optimum or critical placement can be assured only by modeling data as
it accumulates to see if reliability of the objective function (runoff-quality) is
adversely affected by the initial placement.  The underlying assumption in using a
single point (Weather Service gage) long-term record for all points in a metropolis
is that variations are completely random and therefore over a sufficiently long
period (whether 10-years, 50-years, or more, nobody knows), essentially the same
objective function results would be obtained at any point in the metropolis as  at
the single reference point.  However, we have no adequate theory, based on objective
function response and not input response, on the statistical relation between more
than one point of input data and metropolitan-wide points.  Based on our fundamental
assumption, it follows that if more than one raingage is required to characterize a
catchment we will not be able to extend the record as  above-described realistically.
This leads us to the inescapable conclusion that all catchments with a flow-sampler
installation should need only a single raingage.  At first, this would seetn to
restrict consideration to catchments no larger than perhaps 500 acres, but it does
not.  Rather, what it tells us is that subcatchments should be no larger than some
magic limit (say, 300 acres), that each subcatchment of about that size should have
its own raingage, and that each subcatchment should have its own runoff-quality
station at its outlet, so that when long-term records are simulated the simulation
will take account of the separate subcatchments' calibration peculiarities.  This is
still stretching our basic assumption, in the application of long-term single-point
rain data, because we will be assuming perfect joint correlation of objective
function results for the multi-point catchment (i.e.,  between each subcatchment) as
we have for between the reference Weather Service record and some other single  point.
This, however, is the best we can possibly do I

          An  important conclusion that results from the above reasoning is  that,
except where a catchment does not change in character over  a  long  period of  time


                                      - 47  -

-------
On* Hour Rainfall, in Inch**, to
 B* CHp*cl*d One* in 2 Yian
 Ont-Hour Rainfall, in Inchti. lo
 84 Cxpccte! One* in 10 Yon
 On* Hoar Rainfall, m Inch**, lo
 I* CipKttd One* m SOYtart
On*4tour Rtinfall, M Inch**. ID
 •t Eip*cM One* 'n 5 Ytara
One-Hour Rainfall, In Inert**, to
•( E*p*ct*d One* in J5 r«ar»
                  FIGURE 4-  SAMPLE  OF  YARNELL  CHARTS
 On* Hour Rainfall, m Inchn, to
 •* Exp*ct*d One* in 100 Y*ar>

(149)
                 < g
                                                                             fr»qu*ncy
                                                                             2-Yur
                                                                             S-YMr
                                                                             10-YMT
                                                                             25-Y**r
                                                                             50-YMf
                                                                             lOOYMt
                  "0123,56789    10          12
                               R*inlall Intindty for Durations lndicat*d by Param*t«n, in Inc-it* p*r Hour
               FIGURES-  RELATION  FOR  VARIOUS   DURATIONS
                                                                                      (151)
                                             -  48  -

-------
(unchanged land use and hydrology over 10 years or more,  but how can this be
realistically anticipated in advance?) the justifiable duration of data collection
for planning purposes is limited to that amount necessary to calibrate the catchment
response with whatever model or models will be employed to extend the record of that
catchment.  Dallying could be justified on the basis that the "right" model or models
is either not known with assurance at the outset, or that improvements in all models
will be coming, sometime, but such temporizing must be weighed against the cost of
acquiring data only as a "holding action" and the more important fact that the
equipment tied down by equivocation cannot be moved to calibrate another catchment,
and the extent of "catchment sampling" (see the preceding subsection) will
accordingly be shorted.  Of course, there is merit in the perpetuation of a few
stations as a means for sharpening reliability as planning proceeds.  But this is
virtually identical to an extra-cost option on an insurance policy and should be so
regarded.  The only major exceptions are for long-term monitoring and where automatic
control in planned operations is contemplated, discussed in Section 7.

Developing Frequency Rationales for Levels of Protection

          Emphasized throughout this Section, and in Addendum 2, is the importance
of putting results of simulation on a probabilistic footing as a means for maximizing
the reliability of performance of facilities and programs planned for the future.
Noted several times in earlier Sections was the history of moving targets for flood
management and pollution abatement over the past decade, and the likelihood that
policy targets will continue to shift over the next several years.  The only realistic
defense for planning in an atmosphere of ambiguous policy is to employ procedures
the results of which have an inherent flexibility for conversion to meet alternative
jjolicy goals,

          In Section 2 we noted that the policy  goal of the San Francisco master plan
for combined sewer overflow pollution abatement was to limit the average number of
overflows per year.  We also rioted that because  the City had founded  its estimates of
attainable system performance on a long-term precipitation record (62-years), it
would be relatively simple to convert the original simulation results summaries into
their equivalent average annual or seasonal or serial expected occurrences, such as
in terms of overflow volumes or pollutant loads  or seasonal number of overflows or
whatever.  Such a conversion would be facilitated by the existence of field data that
has accumulated over the past six years in San Francisco in support of a total system
automatic control development program (Section 7).  Although the first phase of the
master plan is now committed in terms of physical deployment and capacities of a
about half of the projected facilities, the ultimate incorporation of
automatic control and  interbasin dispatching  features will provide sufficient
flexibility in operating the total system so  that reasonable shifts  in policy targets
could be accommodated.  Had the above elements of flexibility not been included,  the
City would be rather helpless if the policy target was shifted,,.  The  cost of
add-on revisions to new  facilities could quickly escalate abatement  costs  to
completely unrealistic levels,

          While flexibility of analytical tools  is  the  issue  in  this  Section,  it  is
important to note that total project flexibility is much more  important.

Simulating a Range of  Storms and System Loadings

          The preceding  three steps are prologues to  this  step,  called  "simulate
performance  for alternate;;"  in Figure 1.  To  reiterate,  it  is  necessary  to acquire
short-term field dato  records over one or two  seasons  to  calibrate models  for the


                                       - 49  -

-------
 subject simulation.   That is,  rainfall  variations  need to be sampled  over  a  season
 or two on catchments that are  samples of predominant land uses.   Because urban
 catchments,  whether  fully storm sewered or only partially storm  sewered, often  change
 in land use  and drainage system characteristics from year to year,  the  river basin
 analysis approach of deriving  hydrologic indicators  from a long  series  of  field
 observations (such as 20 years or  more) is simply  not feasible.   If reliability of
 the results  of analysis  is a genuine issue,  the only recourse is  a  time-series
 synthesis of runoff-quality events via  simulation  using hydrologic  models.

           Now for the moment of truth.   What are we  going to do with  the resulting
 catchment calibrations?   We are going to extrapolate them.   We are  going to  assume
 that we have a fix on the characteristics  of a  "representative sample"  of  catchments
 and that the model hydrologic  constants,  exponents and other factors  that were  found
 to work on a given class-size-land use  catchment will  apply  to all  catchments of that
 given  character and  (sending up a  prayer)  that  we  can  extrapolate (and  perhaps
 interpolate) these findings for other class-size-land-use combinations  where we did
 not have enough money to  sample for their  characterization.   Obviously, the  level
 of reliability of the tools we have calibrated  is not  merely a function of the
 goodness of  fit of hydrographs and pollutographs in  the  calibrations or in inherent
 instrument or  recording errors,  but also on  the  "representativeness" of the  gaging
 network;  and this latter,  very important point must be brought home to  the so-called
 dec is ion-makers.   That is,  reliability  of  the employment  of  calibrated  tools is a
 function of  several  things,  but  the most important are probably the shrewdness of
 design of the  data network and the extent  to which its results approach an ideal set
 of representative samples.   Let  the modelers lament observed-calculated differences
 and  let  the  field engineers  sermonize on instrument error, but always remember that the
 degree of data  network adequacy  is a much more important measure of the reliability
of  planning or  design conclusions  based on simulations founded on data from the
network.  At this  point we must  qualify the preceding assessment of overall reliability
of  results, downward, but by a necessarily qualitative amount, to allow for the
universal use of  a single-point  rainfall data record within  the metropolis, described
earlier  in this Section, as  the  representative historical rainfall for the entire
metropolitan area.  There  are  a  few cities and metropolitan areas that have raingage
networks  that  have been in operation for five years or longer.  A few of these have
been able to derive empirical  relationships between rainfalls in various sectors of
 their jurisdictions on the basis of selected storms.  While such indicators can
certainly be useful in operations, and perhaps also in design, planning applications
are necessarily referenced  to  the  long-term record of single-point rainfall data.
                                      -  50 -

-------
                                     SECTION 6

                                URBAN RUNOFF M3DELS

Contents

          Reiterated in this Section is the importance of calibrating any urban
runoff model against local field data.  The national dearth of field  data
(particularly for sewered catchments) available for validating models is  cited
as one of the principal liabilities in the use of such models.  However,  there are
a number of potential advantages in the use of models that can be exploited when
certain liabilities, such as a national scarcity of validating field  data, can be
overcome.

          Models are characterized in this Section in terms of their  applications.
Specific models are identified, with an emphasis on sewered system planning
applications, although associable analysis/design applications are duly noted.   Of
the several attempts to compare the performance of various models, the most useful
are those defining extents of application flexibility.  Receiving water modeling
is handled separately because of the tendency towards customized adaptation of
available models for that purpose.  Examples are given of some regional receiving
water simulation studies.

Role of Simulation in Planning

          A special session at an annual meeting of the American Geophysical
attempted to define appropriate rationales and incentives for the more extensive use
of urban runoff mathematical models, for planning, analysis/design and operations.
Among the advantages cited for the use of such models for planning were that:  tests
can be made of alternative future levels of development and their impact on facilities
needed in the future; several models well-suited to master planning are in the  public
domain and are regularly upgraded and made readily available by the Federal agencies
that supported their development; when detailed models are used in advanced stages
of planning the user is able to understand better the physical performance of a
system; the interrelation between land-use projections and planned raitigative
programs and their costs can be made more apparent; revisiting plan assumptions  to
update projects can be done with consistency and relative ease; joint consideration
of quantity and quality of runoff in sewered catchments and in streams can be
accommodated; hydrologic-hydraulic effects of future urbanization can be explored;
and deficiencies in existing facilities and prevailing management programs can be
identified.

          Of the liabilities, outstanding is the dearth of field data on rainfall-
runoff-quality, particularly for sewered catchments, for development of more
acceptable measures of reliability of all types of models.  A recent workshop
conducted by the ASCE Urban Water Resources Research Council  resolved guidelines for
the acquisition of such data by local governments.^^"'  The  spectrum of  investigative
stages utilizing field data for the sewered areas and receiving waters of a
metropolitan area include the following:

. Identification and evaluation of quantity and quality problems.
. Exploration of alternatives for pollution and flooding abatement.
. Analysis of the most attractive alternatives.
. Preliminary design of adopted alternatives.
. Detailed design of adopted alternatives and their  implementation.
. Post-implementation operation via a range of possibilities  extending from simple
  monitoring to automatic control.
                                       -  51  -

-------
 Consistent with  the  comprehensive planning  scope outlined  in Figure 1 (Section 2)
 this  Section will  concentrate on the use of models up  to preliminary design.
 However, because models used in the implementation and operation stages that evolve
 from  a  comprehensive plan rely on extensions of the same data base, some mention
 must  be made of  their characteristics.

           Before embarking on a discussion  of model capabilities, it is necessary to
 establish certain  caveats on the state of the art.

 Some  Reservations

           Because  complex processes, such as in the hydrological response of a
 sewered catchment  to a precipitation occurrence, can never be fully replicated in a
 computation due  to incomplete technical understanding of the processes and the
 infeasibility of detailing the literally myriad pieces involved, resort is made to
 simulation of response of a conceptually equivalent system.  The simulation package
 is commonly called a "model".  Reality dictates that a model should be selected on
 the bases of the type of application involved,  how it is to be used, how much can be
 invested in its use,  how often it would be used, what levels of precision are
 required or desired,  what kinds of outputs are wanted,  how much time can be spent to
 get the model to work,  and how much can be committed to verify and  calibrate the
 model.  Calibration is  the process  of varying model parameters to minimize the
 difference between observed and simulated records.

           We have been  reminded that until each internal module of  an  overall
 catchment model can be  independently verified,  the  model remains strictly a hypothesl
 with respect to its internal locations  and transformations.(154) Because O£ t^e vg  s
 limited amount and kind of field  data available,  just  about all  sewer  applications
 model  validation has  been for total  catchment  response,  at  outfalls.   That is,  under
 contemporary conditions a distributed system model  deteriorates  into a lumped system
 model  for all  practical purposes.   It should therefore  be evident that validation
 using  transferred data  by the model's developer is  not  nearly  enough.  Credibility
 requires at  least token calibration  using some  local rainfall-runoff-quality data
 Unfortunately,  the  acquisition  of such  data  is  commonly regarded as  the  exclusive
 problem of local  governments, and too many planning and analysis exercises  have
 proceeded without benefit  of local  field  data,  using one model or another.

          Calibration and  validation  is further confused by the  fact that much  more
 field  data are  available  for partially  sewered  catchments,  where flow  is measured in
 receiving watercourses,  than  for totally  sewered catchments.   (That water quality
 samples  have been taken  for  only a fraction  of  these gaging sites does not  help).
Adding streamflow hydraulics  to sewer hydraulics hardly simplifies the lumped system
 dilemma  alluded to  above,  yet much of the data  used to verify various models has been
 from such mixed catchments.,  This should add additional incentive for calibration with
 local  data.  "General EPA  guidance to date (1975) has tended to  minimize basic  samplin
 and analysis of urban runoff.  Collection of new stormwater data is severely needed  to
 define the magnitude  of  its  impact on water  quality."(1-55)

          Concluded in a comprehensive Canadian study was that sufficient information
 is not available  on relationships between street surface contaminants, their
pollutional characteristics, and the manner  in which they are transported during storm
runoff periods.  Also concluded was that basically only one type of model exists for
analysis of urban runoff quality,  and that the accuracy of  the water quality
computations using  models extant has not been sufficiently  established to be used
with confidence for prediction purposes, in particular the  formulation relating water
quality with land use.'^6)

                                      - 52 -

-------
          A fundamental objective in water pollution abatement is public health
protection, yet little information is available on the level of pathogenic
microorganisms in stormwater runoff in urban areas.(^7)

          Water resources are not an exclusive major consideration in environmental
protection, and interest is growing in comprehensive or integrated planning,
management and control of all types of pollution sources.   For example,  a model for
metropolitan application has been developed which considers environmental pollution
as a set of interrelated problems, using submodels for land use, residuals,  and
dispersion-disposal of residuals.^"'  While analogous completely comprehensive
modeling of urban water resources processes has been found to be technically
feasible,' 159) conciusions reached in a study of future needs in urban water models
included the opinion that emphasis in urban water modeling in the immediate future
should remain on the simulation of discrete processes (individual subsystems such as
storm runoff) before attempting simulation or optimization of connected subsysterns.
This viewpoint was predicated on the likelihood that the functional responsibilities
of local government water agencies will remain as splintered as they typically are
now.  "It is likely that only where wastewater reclamation, including municipal
reuse, becomes a viable alternative will a single agency require comprehensive urban
water modeling approaches."(140)

Categories of Model Applications

          Mathematical models used for the simulation of urban  rainfall-runoff or
rainfall-runoff-quality can be divided into three different application categories:
planning, analysis/design and operations.  Some particular models have been employed
in both planning and analysis/design, and a few models have been applied  in analysis/
design and operations applications, making it difficult to allocate them  to a single
category.  Additionally, the reader is cautioned that on no account should most of
the models to be mentioned be regarded as typical tools.  Rather, common  practice
still favors rudimentary techniques and only a modest number of offices currently
employ the more advanced tools routinely.

          Planning applications are at a macro-scale, such as  in metropolitan or
city-wide master plans.  Model requirements for planning are less rigorous and require
and permit less detail than for analysis/design because investigation of  a range of
broad alternatives is at issue.  What are sought for planning  tools are general
parameters or indicators for large-scale evaluation of various  alternative schemes.
Hence, the degree of model detail required in  jurisdictional planning is  much less
than in analysis/design.  However, a certain amount of conjunctive intensive detailed
modeling nevertheless is needed to establish parameters and indicators and to provide
an underlying understanding of the governing hydrological processes, so that
simplified expedients are not inadvertently misused.

          Total lengths of underground drainage conduits dwarf  those of open
watercourses  in major cities.  For example, total lengths  in the 97-square miles of
the City of Milwaukee as of  the beginning of 1970 were as  follows:(160)

                    Lakefront length         -         8-miles
                    River  lengths            -        37-miles
                    Combined sewers          -       550-miles
                    Storm  sewers             -       820-miles.

(In addition, there were 685-miles of wastewater sewers).   These drainage conduits
are distributed over 465 drainage catchments having a maximum  size of  1,820-acres  and


                                       -  53  -

-------
 a median size of 25-acres.(I"-*-)  When dealing with so many components  the model  used
 must be as simple and as flexible as possible.  That is,  data processing for planning
 applications becomes a much more important practical consideration than the level of
 sophistication of hydrological process modeling.

           Models used for analysis/design applications are more sophisticated and
 thus are more detailed tools.   They are used for analyzing individual  catchments and
 subcatchments where the simulation of detailed performance of discrete elements  within
 a subcatchment must be achieved.  Whereas hourly rainfall data is  an appropriate input
 for planning models and for simulating flows in larger urban streams,  5-minute
 interval rainfall data (the shortest duration reported by the U.S.  Weather Service)
 is the appropriate input for simulating flows in sewers and small  urban streams  for
 design applications.  That is, the level of sophistication of hydrological process
 modeling for design becomes a  much more important practical consideration than data
 processing,  just the opposite  of the emphasis imposed by  planning  requirements.

           Models used for operations applications are likely to be  more use-specific
 because of wide diversities in management practices,  operating problems and
 individual service-system configurations.  However,  the most potentially transferable
 technology will be for automatic operational control  of total community runoff  a
 capability that is currently receiving intensive development (Section  7).   The
 mathematical models required feature control algorithms that have  to be painstakingly
 derived from numerous indicator applications of both  detailed analysis/design models
 (for generalization of the performance of individual  process components by simulation)
 and planning models (for generalization of community-wide system performance  by
 simulation).   Here also,  analysis/design models are used  as  tactical tools  and
 planning models are used as tools of strategy.

 Models  for Planning Applications

           STORM (Storage,  Treatment  and Overflow Model).   Designed  specifically  for
 urban runoff  and quality evaluation  for total  jurisdiction master planning,  this
 computer model  is  eminently suited  for that  purpose and currently enjoys,  in  one
 version or another,  the  most extensive use of  any urban drainage planning  model.   Not
 only is it non-proprietary but the  computer  program, model documentation^162)  and  a
 users'  manual(163)  are all  readily available.   The original  version of  the model was
 employed in part of the  development  of the Department of  Public Works,   City and Countv
 of  San  Francisco, master  plan  for combined sewer  overflow abatement, described in
 Section 2.

          A simplified logic diagram for  STORM is presented  in Figure 6.(28)  Note
 that  this model  focuses on  structural  means  for  flow and  pollutant containment
 (storage and  treatment).  It is  designed  for use  with many years of continuous hourly
 precipitation records (but can be used  for individual storm  events).  For example,
 a 62-year U.S. Weather Service hourly  precipitation record was used for the San
 Francisco master plan.  Essentially, the model employs an accounting scheme that,  for
 each  storm event, allocates runoff volumes to storage and treatment, noting volumes
 exceeding storage or  treatment capacities (overflows, in  the case of combined sewer
 systems) as these capacities are  exercised from one event to the next.   Water quality
 is handled as a  function of hourly runoff rates, with generated quantities of
 constituents  allocated to storage, treatment and non-capture as for runoff volumes.
Statistics are generated for each event and collectively  for all events processed
 including average annual values.
                                      - 54 -

-------
    INITIALIZE SYSTEM
READ PRECIPITATION DEPTH)
CALCULATE EFFECTIVE PRECIPITATION
CALCULATE RUNOFF FLOW a QUALITY
  FLOW  > TREATMENT
        RATE?
         DECREASE STORAGE
         BY TREATMENT RATE
         MINUS FLOW RATE
            YES
VOLUME = (FLOW - TREATMENT RATE) • A TIMEJ
   VOLUME < RESERVE
   STORAGE CAPACITY
    ADD VOLUME TO
       STORAGE
NO
OVERFLOW = VOLUME MINUS
RESERVE STORAGE CAPACITY
           STORAGE - MAXIMUM
           STORAGE CAPACITY
 FIGURE 6- "STORM"  SIMPLIFIED LOGIC  DIAGRAM*"*

-------
           STORM accommodates non-urban catchments and snowpack accumulation and
 snowmelt, and land surface erosion for urban and non-urban areas can be computed in
 addition to basic water quality parameters.

           Until recently, hydraulic simulation or flow routing was not incorporated
 in STORM.  The latest Corps of Engineers' version includes a capability for routing
 to the outlet of each sub-basin through the use of triangular unit hydrographs based
 on the Soil Conservation Service procedure.(1°3)  Planned for future versions are such
 added capabilities as an expanded procedure for sizing detention reservoirs.   Hence  we
 find a tendency towards enlargement of a model originally designed for broad  scale
 planning to include preliminary analysis/design capabilities.  Simultaneously,
 detailed models for analysis/design are being simplified to make them more amenable
 for planning applications, and both types of models are being interfaced to
 facilitate feedback of information from one to the other.  All this is making it more
 hazardous than ever to categorize a model in terms of its major applicability.

           A modification of the STORM model has been linked with the receiving water
 module of the SWMM analysis/design model (noted below) for continuous simulation of
 receiving water quality,(164,165;

           As part of a nationwide assessment of stormwater pollution control  costs
 a "desktop" procedure was developed,  by streamlining the STORM model, that can be
 used to estimate the quantity and quality of urban runoff in combined sewer and
 storm sewer areas and unsewered portions of a jurisdiction.   Combinations  of  storage
 and treatment for pollution abatement and their costs can be estimated taking
 advantage of generalized results from the nationwide assessment.(166)

           A methodology has been suggested for screening urban development alternatives
 and preliminary design of storage and treatment capacities  for runoff control on the
 basis of water quality impacts  utilizing STORM.(167)

           Other Planning Models.   A simplified stormwater management  planning model
 has been developed  that is  an inexpensive,  flexible  tool  for planning and  preliminary
 sizing of stormwater  facilities.(^32,168)  Time  and  probability  considerations  are
 incorporated in the model.   Joint usage of a complex model  and a simplified planning
 model,  such as  this one,  is  said  to be  not only compatible  but also  complementary.

          A  very  simple  methodology has been advanced for preliminary screening of
 stormwater  pollution  abatement  alternatives.(169,170) while  the  method was conceived
 for  combined  sewer  system applications,  it  could as  easily  be  applied to stormwater
 systems.  Developed for  use at  the  national  or  State  decision-making  level  for  early
 identification of poor candidates  for abatement  project  funding,  it might  as  readily
 be applied  for rough, early assessment of  the maximum pollutional  impacts  of  storms
 at the metropolitan level.

          Reported  verifications of the simple process planning models, including
 STORM, have been limited, although  hearsay  indicates  that the  number  of verifications
 is growing.  Because  suggested magnitudes of model coefficients are based on  the
 sparse amount of field data available nationally, it  is very important that local
 rainfall-runoff-quality  field data be used to calibrate such models for the sake of
 enhanced reliability of  results.  Too many planning exercises have proceeded without
 benefit of local field data, using one model or another.

          The original MIT Catchment Model was predominantly an analysis/design type.
However, a modified version has been employed for estimating sewered  catchment and
 stream runoff in the Fairfax County master plan development,(30,31) section 2.

                                      - 56  -

-------
          Synthetic unit hydrographs ,  based on regional parameter general izat ions,
were employed in planning the drainage for the large "Woodlands" developmental^)
near Houston.  However, unit hydrograph applications have bean mostly in analysis/
design.  For example, a unit hydrograph is in general use for designing drainage and
flood protection works in metropolitan Denver. (173, 174)

Models for Analysis/Design Applications

          There are a number of this type, including the two noted immediately above.

          Although it is essentially an analysis/design type when applied to
smaller-scale catchments, the Hydrocomp Model can and has been used for broad- scale
receiving-water planning applications, described later.  The Hydrocomp Model has also
been employed in several metropolitan sector studies. (175-178)

          There have been some applications of versions of the British Road Research
Laboratory Model. (1'9-lal)  Available to the public is the computer program for a
very simple streamflow and drainage system model used in the Chicago area(l°^) for
preliminary project planning and design.

          The USGS has modified the MIT and ILLUDAS models for continuous simulation,
detention-storage accommodation and water quality simulation.
          As far as drainage system simulation is concerned, the EPA
(Stormwater Management Model) and its variants have enjoyed more applications and been
subjected to more verification than any other model.  For example, a version of SWMM
has been used to supplement and compare results obtained by the Chicago Department of
Public Works from its own models(185,186) as part Of tne raaster plan for combined
sewer overflow abatement in metropolitan Chicago, Section 2.  Another version, used
in connection with the San Francisco master plan, (187*188) has the most versatile
transport module of those extant.  This version is now included as an option in the
computer program for SWMM.  The latter is continually updated. (189)

          For the Milwaukee master planning project, described in Section 2, three
models were used:  STORM, SWMM and Harper's Multipararaetric Water Quality Model. (28)

Model Comparisons

          There have been a rash of projects comparing the merits of various models
on the basis of a variety of criteria. (156, 190-197)  instances where tests had been
published of the performance of various types of models against field data were
reported in 1975. (1^1)  Advances in modeling capability occur almost too rapidly to
keep track, and in 1975 it could be said that mathematical model development for
sewered system applications had already seemingly greatly outpaced the data base for
model            *
          Results of the various tests are mixed, mostly because  there  is no acceptable
basis for multiple-objective comparison.  Peak flow  is the major  consideration  in
sizing conduits, volume and hydrograph shape are critical for sizing storage, and
concentrations and loadings of pollutant emissions are essential  for evaluation of
receiving water impact and helpful  in sizing treatment facilities.  Each model  has  its
strengths, weaknesses and outright  faults for a given application.  Over and above  the
correlation definition problem is the inherent difficulty with  any  runoff model in  the
necessarily subjective separation of abstractions (infiltration,  depression storage,
etc.) from total rainfall to resolve rainfall excess (amount and  pattern),  which  is


                                      - 57  -

-------
 the input from which an equal volume of .direct runoff is generated by models of one
 kind or another.  After analyzing the performance of a variety of models, it was
 concluded that the weakest link is the proper estimation of rainfall excess.(198)
 All this is to reiterate that urban hydrology is still, in the absence of an adequate
 body of field data, more of an art than a science, and that under this circumstance
 the choice of a model for a given application is largely a matter of taste.

 Receiving Water Modeling

           Receiving waters are the common repository of effluents from just  about
 every community and self-supplied industry in a metropolis, constituting perhaps
 the most shared aspect of urban water resources.  Impressive advances have been made
 in receiving water modeling.   Initial attention was on hydrology and hydraulics in
 support of flood control objectives.   Water quality modeling capability has  evolved
 more recently, with a tendency to use tailor-made models for discharge-quality
 simulation in planning applications,  and earlier development was focused on  estuaries.
 The choice of a model or models to be used in any given planning effort therefore
 requires careful and discriminating study.  Consequently,  it is appropriate  to cite
 recent capability summaries and describe a few examples of applications.   Reference
 has been made earlier in this  Section to SWMM capabilities.

          A compendium, (19/)  two  companion reports, (199,200) a jjorth American
 summary/201'' and a text,(202' survey features of large-scale water  quality  models-
 and an annotated bibliography  of  models  for tidal rivers,  estuaries  and coastal
 waters is  available.(20-»)   Tidal  water models have been comprehensively
 classified,(204-,205)  an£j capabilities for modeling estuary and streamflow water
 quality have been assessed.(206)

          Aquatic ecosystem submodels have been delineated  for process  analysis.(207)
 Aquatic ecosystem models were  surveyed in 1974 for the  National Commission on Water
 Quality via  a questionnaire,(208)  and while several  models  reported  are more generally
 applicable,  nearly all have been  developed or tested on a  specific water body and only
 a  fraction of the applications have urban implications.  Computer graphics techniques
 might  be useful  for displaying oceanographic  data for coastal  metropolises.(209)

          Water  quality  modeling  for  systems  containing  rivers and reservoirs has
 been advanced  through  the issuance of a  description  of  a combination of models.(210)
 The Hydrologic Engineering Center  is  having dynamic  flow routing routines  added to
 the model and plans to upgrade the documentation  as  new developments  occur.(211)
 Dynamic or unsteady-flow water quality modeling  is particularly important  in the
 case of  significant pulse loadings from urban runoff or when man-made controls  such
 as dams are  involved.

          Effects of risk and  uncertainty  in  the  application of operations research
 techniques,  including hydrologic modeling,  have been  critically reviewed at  length.

          Although  receiving waters represent  only a  part of the  total  urban  water
 resource, they commonly  traverse entire metropolitan  areas  and  are affected by  the
actions of a multitude of local jurisdictions.  Recent emphasis on regionalized
wastewater treatment and disposal has resulted in some receiving water  simulation
studies on a grand scale, a few of which are noted here as  examples.

     San Francisco Bay.  The study area extends upstream well beyond  the San  Francisco
metropolitan area.  About two-fifths of all the surface runoff  in California  passes
beneath the Golden Gate bridge via the Bay estuary.  About 800-mgd of treated


                                       - 58  -

-------
wastewater effluents from industrial and municipal sources enter the Bay.   Funded by
the California State Water Resources Control Board, the estuary water quality project
is modifying and calibrating an existing set of interrelated hydraulic,  water quality
and ecologic models of the Bay.(213)  xhe hydraulic models were designed to permit
rapid evaluation of alternative wastewater management plans at a reasonable cost and
to serve as a screening tool to reduce alternative plans to a reasonable number.  Once
the screening of plans has been resolved, the ecologic models are to be  used in the
simulation of the ecologic response of the Bay for the more attractive plans.
Descriptions are available of some of the water quality'21^' and flow'^lS)
characteristics of the group of models used.

     Denver Area.  Partly supported by an EPA planning grant, a metropolitan water
quality management planA216) ^s being resolved by the Denver Regional Council of
Governments that encompasses:  a plan for an areawide system of wastewater treatment
facilities; financial arrangements and legislation and institutional guidelines for
implementation of the plan; a water quality surveillance system for regional water
quality management; and an environmental impact assessment of the proposed facilities.
Virtures of alternative treatment plant locations, sizes and groupings were analyzed
using the Hydrocomp model.  Flow and water quality simulations were performed for the
mainstem and principal tributaries of a 750-square mile sector of the South Platte
River basin that essentially includes the metropolitan Denver area.  Numerous flows
enter the study sector from upstream.  Stream sections within the sector were divided
into 33 reaches for the simulations and hourly water quality levels and streamflows
were computed for each reach.  Model calibration was complicated by a necessity to
account for upstream reservoir releases and irrigation diversions and shortcomings  in
long-term water quality data.  Simulations were for anticipated future  land use.  An
earlier DRCOG study(39; (Section 2) had developed  some contemporary runoff-related
land use characteristics for 398 sub-basins, most  of which are  in the study area,
and this information was exploited  in characterizing streamflow inputs  from urban
land runoff and its associated pollution burden.

     Seattle Area.  In response to  a State of Washington directive  to develop water
pollution control and abatement plans for two basins,  the Municipality  of  Metropolitan
Seattle and surrounding King County formed the River Basin Coordinating Committee
(RIBCO), which adopted a coordinated approach to  insure development of  not only an
integrated plan but an integrated planning process as  well.^2!/)  Study elements
included water resource management, water quality management, urban runoff and  basin
drainage, solid waste management, and land use allocation.   Because of  an  emphasis  on
a  continuing planning process, a systems approach was  specified.  This  led to
extensive use of computer  models and provisions  for  their  future updating  and use.

          The water resource management  study(21°' included  use of  the  Hydrocomp
           to develop a regional water  supply plan for municipal and  industrial
usage, navigation  lockage,  stream  flushing  and  fish  flows.   Both streamflow  quantity
and quality characteristics were  included,  together  with  firm yields  and  upper  estuary
considerations.  Seventeen chemical and  biological parameters were  simulated.   Model
input/output  is  responsive for segments  5-square miles or  larger in size.

          A variant of  the EPA SWMM model was used in connection with the urban runoff
evaluation  portion of  the  RIBCO Study.(220)

          Hydrocomp has  since  extended  the  usefulness of  its model  for  non-point
pollution  simulation/221'

      Milwaukee Area.   Responsible  for  finding solutions to areawide developmental
and  environmental  problems in  a  rapidly urbanizing seven county area, the Southeastern

                                       - 59  -

-------
 Wisconsin Regional Planning Commission was created in 1960.   The Commission has
 completed comprehensive studies for the Root River,  Fox River,  Milwaukee  River and
 the Menomonee River watersheds, and is conducting a  similar  study of  the  Kinnickinn"
 River watershed.   The Milwaukee master plan, outlined in Section 2, is  an extension
 of part of the Milwaukee River watershed comprehensive study.

           Among the outputs of the comprehensive  watershed studies are  flood hazard
 maps for computed 100-year recurrence interval  flood stages  and estimated annual
 flood damages for floodprone reaches.   Extensive  simulations  are made using a Corps
 of Engineers  backwater computer program(222' and  a Soil  Conservation Service flood-
 routing computer  program.(223)  SEWRPC also  uses  the Hydrocomp  model.   There are
 several other computer programs suitable for practical  application that are readily
 available.(224)  j^g watershed studies have  included an  examination of  both
 structural  and non-structural  alternative plan  elements  for  the resolution  of
 existing flood problems and for determination of  the effect of  changing land use  in
 an evaluation of  the 1990  regional  land use  plan.(225)  Also  included have  been the
 development of guides for  designation  of floodland regulatory zones, analysis of
 floodland  encroachments, provision  of  bridge design  data and criteria,  provision
 for subsequent updating and refinement of flood stage profiles  and floodland mappine
 preparation of Federal  Flood Insurance reports and dissemination of flood hazard
 information.'2"'   Other modeling has  been undertaken in connection with  Section  201
 and Section 208 planning.

          More  recently, the Hydrocomp water  quality submodel has been  incorporated
 by  its  developers  into  flow simulation capabilities  in a Hydrologic,  Hydraulic  Wate
 Quality and Flood  Economics Model.(153)                                       '     r

     San Diego Area.  Under an Urban Systems  Engineering Demonstration  Program
 comprehensive  planning  grant from the  Department of Housing and Urban Development
 the Comprehensive  Planning Organization of the San Diego Region commissioned a joint
venture  to  formulate water  resource planning methodologies and  technical approaches
 for estimating costs of alternative land use plans.  Five interrelated computer
programs were developed:  Data Management System;  Water Supply Model;  Wastewater
Model; Flood Control Model; and Economic Analysis  Program.^22")
                                     - 60 -

-------
                                     SECTION 7

                           SOME MODELING CONSIDERATIONS

Pitfalls

         •Models can be misused or be used ineffectively if they  are  selected  and
employed without prior careful consideration of their data requirements  and  their
place in overall information processing.  Because the response formulation of
current models is keyed to land-use parameters, the extent and detail of land-use
characterization that can or should be assessed becomes  a function of the intended
uses of model outputs.

          While flood plain mapping of 100-year stream stages is  a minimum requirement,
the mapping of more frequent occurrences is required for analysis of  the economics
of flood damage mitigation.

          Permanent field data acquisition systems for operations, particularly for
automatic control, can be readily justified.  In Section 5, only  temporary
installations could be justified for planning and analysis/design purposes.   On the
other hand, a skeletal instrumentation network should be maintained throughout the
planning through operations cycle, but such long-term foresight has not been well
received in the past.

          The themes described above are discussed in this Section.  They have been
singled out because they are particular pitfalls that have been encountered  in past
planning efforts.

          A special feature of this Section is a brief review of the latest
developments in the manipulation of storage via automatic control.

Land-Use Data

          Land-use information needs for planning and analysis/design models are
similar, differing mostly  in the amount and detail required, the latter model
category being more sophisticated and demanding by far.  For current models, the key
land-use variable for calculating runoff is the imperviousness for each land-use
category being considered.  Runoff rates and volumes rise as the degree of
imperviousness increases.  Because the basic hypothesis for source quality for current
models hinges on street-surface pollutants accumulated between rainfalls, the key
land-use variable for calculating pollutant accumulations is the extent of street
gutters, expressed in terms of length of street curbs.  Indeed,  these two parameters
may be the most sensitive  of the parameters affecting the estimation of runoff and
pollutant loading amounts, respectively, according to the results  from limited
sensitivity tests.

          Intuitively, land-use type, degree of imperviousness and extent of curbs
should be correlatable, using residential  population density as  the  dependent
variable.  Such correlations have been  found for  the State of New  Jersey,^"/)
sectors of metropolitan Washington, D.C.(228,229) ancj for a  few  other samples  in the
U.S. and Canada.*166'  Similar relationships can  be developed  for  any metropolitan
area or major  jurisdiction for use in planning, by characterizing  representative
samples of various land-use population  densities.
                                       -  61 -

-------
           However, because more detailed representation may be justified for later
 preliminary design, recourse to total jurisdiction or metropolitan area assessment
 of current (and perhaps prior) land use might be more consistent.   While aerial
 photographs can certainly be used for land-use surveys,  new technology for  computer-
 aided interpretation of satellite data might be exploited for the  explication of
 imperviousness.(230)  Compared with costs for interpretation of aerial maps   the
 new method appears to be considerably faster and less expensive.(231)   Interpretive
 resolution is of about the same precision as for manual  reading of aerial photographs
 but only for watersheds of a minimum of about one square mile in size.(231)   This     '
 tentative conclusion is based on a case in which comparative model results were
 obtained for both types of mapping.

           At this juncture it is again important to remind  the reader  that planning
 is a dynamic process concerned with alternative futures.  Present  conditions  are
 prologues or points of reference for the future and projections are best made from
 the springboard of a reasonably reliable assessment of current conditions.  Good
 information on current actualities is needed to proceed  confidently with
 extrapolations, where the degree of reliability of  projections diminishes as
 estimates move into more remote time-frames.   That  is, one  of the  tests  of
 contemporary information usefulness relates  to its  amenability for the  projection
 of expected ranges of future conditions or requirements,  a  consideration over and
 above the question of model  response reliability for  a given land-development
 configuration (Section 5).   This is true for  all levels of  planning but  certainly
 most visible at the implementation level.  Land-use maps  show current or recent
 land use,  but planning for  specific facilities  requires projection of  land use
 through  the full  period of  implementation, which would be around two decades  for
 most comprehensive plans.

           Because anticipated  future conditions  cannot be simulated by manipulating
 actual land use,  recourse  is made  to simulation  by  some form of calculation or
 analogy.   Again,  as  in the analysis/design of  urban water resource  facilities, the
 first major requirement, prior  to  investigating  requirements  for the future,  should
 be  attainment  of  a satisfactory simulation of  existing conditions.  This  might be
 called the tool calibration  phase.   Immediately  the question emerges:  to what detail
 should basic  land-use  information  for existing conditions be  acquired?   Because cost
of  information acquisition rises with the  degree of detail  sought,   it is  prudent
 first to derive collaterally an indication of  simulation precision through
 sensitivity  testing.   In the case  of hydrological modeling,  the subject  of the
 preceding  Section,  there is usually  too  little field data for  conducting  a realistic
sensitivity analysis.  On the other hand,  even a year of rainfall-runoff-quality
 field data  for one to  a few catchments would lead to the use of much more reliable
model coefficients than the use of values  transferred from  elsewhere by default.(232)
All  this is  to  say that if local field data is not available  for some degree of
hydrological model calibration, what would be the logic in determining land use
 imperviousness and curb lengths at a scale of, say, 10-acres, when a scale of perhaps
100-acres or  200-acres would be more consistent with the probable errors  in assumed
or  transferred hydrological model  coefficients?  Similarly,  why determine model
information  input  for  every bit of land  involved, when a sampling of perhaps a tenth
of  the planning area would be more consistent?  Conversely,  when land-use projection
of only specious quality are available, why insist on process elegance in the
hydrological tools employed?

          Perhaps a suitable answer to the information scale and detail question
would be to suggest that:  when information on a small scale and with good detail
is available  it should simply be used to the maximum extent possible consistent with


                                      -  62 -

-------
available budgets for simulation;  and when only coarser raw information is  available
the scale and detail used should be weighed against the presence or absence of  field
rainfall-runoff-quality data for hydrological model calibration in tempering the
level of model sophistication adopted.

          There are a number of computer models for land-use projection and land
allocation.  An example of a projection model is one advocated for wastewater system
planning.('  A land-allocation  model was employed in the Fairfax County  master
planning (Section 2) to derive future runoff characteristics for the hydrological
models employed.(9»2J4)  pOr both  of these examples there are computer-graphics
subroutines for display of variables.  Of course, land-use mapping is only  one
component of composite mapping and associated environmental impact analysis.(235)

          Prior to conducting inventories of land-use data and establishing procedures
or designing systems for the storage, retrieval, analysis, and display of such  data,
it is important to consider the potential need, either in the modeling process  or  in
other aspects of master planning,  for other types of areal data.  These other types
of areal data, that is, natural resource and man-made features data having an areal
characteristic as opposed to point location, might include soil type, vegetal cover,
population density, land market values, and flood-prone status.  All such data  types
are similar to the land-use data discussed above in the sense that they have an
areal dimension and, therefore, may be readily inventoried and managed in a manner
that parallels that used for land-use data.  The potential utility of a systematic
approach to the inventory, storage, retrieval, manipulation, and display of land-use
and other areal data has been demonstrated.(236)

          Guidelines for acquisition of information on natural characteristics  of
land for land-use planning are available.(237)

Flood Plain Mapping

          It has been posited that economic criteria can and should be used in
planning nonstructural flood control measures generally, just as they have  long been
used in planning structural measures.(238)  gut little data is available at the local
level for making benefit analyses. (m> 239)  Fundamentally, what is needed  from a
hydrological standpoint  is stage-frequency information, which would have to be
developed by simulation.

          As noted earlier, there is considerable  evidence(122^ which suggests that
the effects of urbanization on streamflows decreases as rarer floods are approached,
to the extent of a difference that may be undetectable or  at least  insignificant at
the 100-year level.  That is, the effects of land  use changes are more pronounced
the more common the occurrence, adding further  substantiation to the need  for  some
sort of regional stage-frequency information over  and above the mapping of  the 100-
year event.  Without a reasonable indication of  stage-frequency relations  for
existing conditions, how can one make  flood plain  planning projections with reasonable
confidence?

          An attempt has been made to  generalize the 2 1/3-year, 5-year, 10-year,
20-year and 50-year frequency of peak  flows for urban waterways on  the basis of
available data.(240)  input variables of  interest  here are drainage area slope and
degree of  imperviousness, akin  to the needs for  conduit  flow  simulation models.
                                       - 63 -

-------
           Over half of natural 100-year flood plains in urban areas  are already
 occupied. (103)  Damages to buildings are a function of the height of flooding level
 at individual buildings in a flood plain.   That is, two identical buildings  located
 at different elevations of a waterway flood plain would suffer differing damages  for
 a given flood, assuming that both were not totally destroyed.   Looking at it another
 way, in any given instance some buildings  are exposed to more frequent flooding
 hazard than others.  There is so much attention accorded the 100-year flood  level (241)
 that more frequent occurrences can be inadvertently overlooked.   There is a  convincin
 indication that potential flooding losses  might be significantly  mitigated in numerou8
 situations if flood protection was provided up to some relatively frequent recurrence
 interval, meaning that conventional protection against a rarer flood level could be
 much more costly.'   '

           In essence, needed are stage-frequency relations  for as many streamflow
 locations as is feasible if benefits are to be assessed realistically.   Clearly
 delineation of 100-year flood plains is not enough.   The National Flood Insurance Act
 of 1968 and the Flood Disaster Protection  Act of 1973 call  for measures implying the
 necessity of using stage-frequency relations.   Whether,  say,  10-year,  25-year and
 50-year flood plains should or could be mapped along with the  100-year  levels is a
 moot point.

           Developers of the Fairfax County master plan (Section 2) were fortunate
 that the USGS had mapped not only the 100-year County flood  plains but  also  the
 25-year and 50-year flood plains.   The Southeastern  Wisconsin  Regional  Planning
 Commission (Section 2) not only mapped the 100-year  flood plains  of  the Milwaukee
 River watershed but also the 10-year highwater surface profiles using a simulation
 model.
 Storage  Manipulation Via Automatic Control

           While we may not  go  that far as a nation, the ultimate solution for the
 problem  of abating pollution from urban storm sewer discharges and combined sewer
 overflows  is  the  treatment  of  such flows prior to their release into receiving
 waters.  Outflow  surface-water hydrographs of sewered catchments exhibit very large
 peaks, on  the order of two  or  more times those of equivalent non-sewered areas.
 Capturing,  transporting and treating all discharges/overflows, unattenuated, would
 require  gigantic  collection sewers, pumping stations and treatment facilities  all
 of which would be used less than the equivalent of about an hour a day, on the
 average, over a typical year.  Furthermore, there will be periods running into several
 weeks or even months where  the discharge/overflow control works will not operate at
 all, because of little or no precipitation.  Therefore, schemes for system-wide
 discharge/overflow collection and treatment incorporate some form of auxiliary storac
 for the  purpose of attenuating sudden inflows to collection and treatment facilities
with the objective of scaling down the size of such facilities to reasonable and    '
manageable proportions.  The complexity added by converting a facility with simple
 gravity  flow into a multiple-component interacting scheme,  where practically
 instantaneous response to flow incidences must be made, requires incorporation of
some degree of automatic operational control.   In addition,  automatic control permits
manipulation of dispersed storage capacities for the purpose of maximizing the
 effective use of their collectively available volumes, and  the diversion of
uncontainable releases to lower impact receiving water reaches, by taking advantage
of the areal non-uniformity of rainfall over large jurisdictions.

          There are three distinct levels of complexity in  automatic control for the
operation of urban water resource systems.   In order of complexity,  they are


                                      - 64 -

-------
automated monitoring, remote supervisory control,  and complete,  or "hands  off,"
automatic control.'  ^^'  So far,  the use of automatic control for  the manipulation
of storage has been considered seriously only for  existing combined sewer  systems,
although the same technology could be applied in the future to existing or new
separate stormwater sewer systems.

          Two basic schemes have been considered:   the exploitation of ambient
storage capacity in existing trunk and interceptor sewers in combined systems  by
adding dams and diversion structures that can be remotely actuated; and the much
more expensive provision of altogether new auxiliary in-line or off-line storage
facilities that can be remotely controlled.  The level of automatic control adopted
has been scaled to the investment in storage capacity added.  Remote supervisory
control is the highest level of automation to which schemes for exploitation of
ambient storage capacity have aspired.  This is exemplified by the metropolitan
Seattle remote supervisory control system.(244,245)  &t least two  other cases  have
not progressed beyond automated monitoring.

          Originally planned features have been outlined of the San Francisco master
plan for combined sewer overflow pollution abatement. (2^'  Implementation of the
master plan (Section 2) represents the exclusive instance where complete automatic
control capability is being sought.  There are fundamentally two distinctive modes
of automatic control, reactive and predictive.  The reactive mode  refers to the
manipulation of facilities in response to the actual occurrence of a rainstorm,  as
the storm progresses over the affected jurisdiction.  The predictive mode refers to
the anticipation of how facilities should be handled just prior to the beginning of
a rainstorm, and blends with the reactive mode once rainfall on the affected
jurisdiction commences, but the predictive feature is retained until the cessation
of incipient rainfall.  Reactive strategies are apt to be site-specific whereas
predictive strategies are likely to be adaptable to any system.

          San Francisco is currently engaged in automatic control  reactive capability
development but plans to advance to a predictive capability soon. (I-')  In cooperative
projects with the City, Colorado State University has advanced the automatic control
capability state of the art with respect to individual catchments  and  the total  City
system.('      Mathematical models employed for developing control  algorithms for
both control levels have had somewhat more specific information requirements than for
STORM but much less detailed requirements than for SWMM.  That is, simpler models
than SWMM were used for developing control algorithms for individual catchments.
However, use of analysis/design models such as SWMM are needed to  monitor and provide
a physical understanding of results from wholesale applications of far simpler models.
That is, a certain amount of intensive detailed modeling  is needed to  establish
parameters and indicators and to provide an underlying understanding of the governing
hydrological processes and system responses so that simplifying expedients are not
inadvertently misused.

          With respect to predictive control capability development, a procedural
scenario can be hypothesized with regard to storm characterization.  The  data source
would be at least two years of rainfall data for a network of raingages surrounding
the jurisdiction, with their ideal density heavier towards  the side  from  which  storms
passing over the metropolis tended to prevail.  Because of  the importance of  time  of
occurrence, synchronization of raingage records would be  mandatory.  By means of
computer-mapping, the depth of rainfall over any point within the  network at  a  given
time interval could be interpolated and the prevailing storm direction at a given
instant would be  indicated by the slope of the maximum tilt of the fitted surface.
Storm path and traversing speed would be  indicated by the progression  of  prevailing


                                      - 65 -

-------
 storm directions across the network from one time interval to the next.   The statistical
 means and a measure of variance would be derived for prevailing direction,  path and
 speed of storms and for total storm depths outside and inside the metropolitan area.
 As an adjunct of total storm depth characterization, similar statistical characteristics
 of the time-pattern of accumulation of total depths would be obtained.   Next  a theorv
 of storm temporal and spatial variations would be postulated, the statistical
 indicators would be applied to the resultant theoretical storm characterization model
 and the validity of the model would be verified by testing it against at least a
 season of data that had not been used in the prior analysis.  It might take some
 adjustment in approach or theory to reach an acceptable predictive capability.  Of
 some help would be the fact that advances continue to be made on the statistical
 characterization of rainfall records for a single point for intervals of an hour or
 more.(2^9'

           Consider now the following capability,  expected to become available in 1977•
 at each node of a metropolitan spatial grid (such as a 64-point by 64-point grid with*
 nodes  3-miles apart) the digital read-out of rainfall intensity every few minutes
 (such  as at a 3-minute interval),  the cumulative  depth of rainfall at each  time of
 reporting (such as at a 3-minute interval),  and rates and depths for each node
 predicted as far in advance as 6 hours.   This capability is emerging from the use of
 special radar equipment in conjunction with the deployment of 200 raingages in a
 recently concluded study in metropolitan St.  Louis  and a subsequently implemented
 study  in metropolitan Chicago involving  a network of 300 raingages.  Storm
 characterization will be possible  by analysis of  the data from these extensive
 networks; however,  in system control applications  later it will  be necessary  to  have
 only a radar facility and only a few telemetry-connected raingages,  with the  emerging
 special  software converting radar  readings into rainfall intensities and depths
 occurring and predicted,  for a grid  of points that  could be as  large as  almost 200-
 miles  by 200-miles  in size.   These exciting  developments  are  part  of the "Chicago
 Hydrometeorological  Area Project," partially  supported  by an  NSF/RANN grant to  the
 Illinois  State Water Survey.t250'

          Criticism  of  the  use of  synthetic  storm patterns  for planning  and analysis/
 design applications  is  elaborated  in Addendum 2, with particular  reference  to  cases
 where substantial new storage  facilities would be  involved.   The  criticism  centers
 about  the point  that  for  these applications  the issue  is  system output characterization
 not rainfall  characterization.  This substantial reservation  does  not apply for        '
 real-time applications, as  in computer-assisted supervisory control or in complete
 automatic control, where storm prediction  capability would be directly interfaced with
 facilities operation simulation capability.  The remaining element  is then  the
 development of predictive mode tactical algorithms for operation of facilities, such
 as rule curves or decision  trees or  the like, as is being done for  the San Francisco
 system.

 Rajngage Networks

          In 1968, fifteen of the largest metropolitan areas had a. network of
recording raingages of between about 5 and 192 instruments, with records  spanning
periods of 2 to about 50 years.(251)  There has since been a growth in  the number
and size of networks, but the most advanced automatic network was installed  in
San Francisco  in 197.1, where each 0,01-inch of rainfall from 30 raingages are
 transmitted at the specific second of occurrence to a central data acquisition and
recording station.(252)  A. very similar,  but smaller, telemetry and automatic data
logging facility has since been installed in Portland, Oregon.(253)  (Both facilities
also automatically log signals from a number of field flow-indicating instruments).


                                      - 66 -

-------
          A computer-mapping program (SYMkP'"') operated off-line has  been employed
by the San Francisco DPW to process a number of storm records obtained from their
network.  For a given storm, the program interpolates between raingage network
readings at any specified clock time and calculates rainfall  depths  for an imaginary
fine grid covering the entire City.  On the basis of the grid values,  the program
prints a contour map showing the variation of depths over the City for the time
interval in question.  Visual inspection of a series of such  maps for  an entire
storm, plotted for some specified time interval such as five  minutes,  provides
dramatic evidence of rainfall variability over time and space and the  movement of
the most intense rainfall sectors across the City.

          Study of storm mappings showed that, for more intensive rainfalls, only
portions of San Francisco are subjected to the higher rainfall rates at any given
time.  Storms of low total depth are more uniformly distributed, but these would not
tax facilities as severely.  The spatial and temporal differences observed in the
occurrence of rainfall has confirmed the presumption of the 1971 master plan that  a
system of interconnected components would result in more efficient utilization of
facilities.  That is, use of real-time computer-actuated control, based on sensing
the direction and likely volumes of rainfall, accompanied by a constant concurrent
updating of system status, would permit fully efficient use of all storage and flow
capacity throughout the system.  The result would be a wider latitude in staging the
master plan implementation, potential improved system performance reliability, and
greater flexibility in meeting water quality standards as they are refined or revised.
In essence, system-use allocations will be balanced between storage vacancies available
and unencumbered treatment capacity, to meet sensed or predicted rainfall loadings.
Further, when overflow occurrences are unavoidable, the sites of their release can be
selected on the basis of minimal environmental impact.

          Use of a raingage network for developing and using some form of remote
supervisory or complete automatic control can be rather readily  justified.  There is
considerable merit in installing networks for surveillance of much less exotic
systems, such as those in several other metropolitan areas.  However, older networks
have had raingages with their own on-site recording charts.  Synchronization of
charts was always a problem, with time correlations within 15-minutes about the best
that could be expected, but the much larger  liability was the onerous and time-
consuming task of reading the inked traces and reducing the results to a digital
tabulation.  The result was that usually only the data  for selected storms of unusual
interest were reduced.  Unless  field data  is reduced and analyzed, it  is difficult to
justify indefinitely the maintenance of a raingage network.

          The above  considerations  lead us  to  the  inescapable conclusion  that  the
principal value  in the use  of raingage networks  is  in  the operation of  systems,
but that full exploitation  of such  networks  requires automatic  data logging and
reporting, an expensive commitment.  However, when  raingage  networks  are  deployed
as part of an automatic control  scheme  their cost  compared with their importance  is
almost  trivial.  (These qualifications apply equally to  flow  monitoring stations,
which are an integral and  indispensible part of  any  automatic control system).

          Installation of  raingage  networks  for  planning and  analysis/design  is more
difficult  to justify.  There are  some obvious  advantages when employed simultaneously
with  the deployment  of temporary  runoff and  water quality  field stations for  the
calibration of  simulation models.   Also,  they  could  prove  invaluable  in the checking
of  post-improvement  performance of  systems,  and  there  is much to be  said for  the
maintenance of  monitoring  stations  on selected catchments  and receiving water sites
from  the beginning of  planning  through  operation,  but  this type of reliability


                                       - 67 -

-------
 insurance has always been one of the most difficult to sell.  Not to be overlooked
 are possibilities for extending features of temporary data acquisition systems into
 permanent urban flood warning systems, such as in metropolitan Melbourne.(254)  jt
 is not entirely inconceivable that the day when real-time bacteriological  and
 chemical warning systems might be added may not be too distant.

           Parenthetically,  mention should be made of recommendations that  have been
 offered for raingage densities in the operation of in-line storage or selective
 discharge systems.(255)  Also, there are new useful references on the effects of
 storage in underground systems(256) aiuj ^n surface systems,(257-259) but mo^ Qf
 these are marred by a reliance on hypothetical cases and  a dependence on synthetic
 storms.  The latter is a practice deplored in Addendum 2  when  used for important
 projects.

 EPA Overview

           An important reference  is the recent review of  EPA's  research and
 development program for urban  runoff pollution control.(133)  Outstanding  features
 are:   estimated  national  costs for  stormwater pollution control;  a summary of
 references on simulation  models;  more numerous  and  more detailed  urban stormwater
 quality management  alternatives than could be justified for inclusion in Section 4
 of  the present report;  and  identification of  174 EPA reports and  74  on-going  EPA
 projects,  plus 15 additional references.

           EPA personnel have provided  annual  literature reviews on urban runoff
 and combined  sewer  overflow papers  and  reports.(260)

           Difficulties with planning called for under P.L. 92-500. particularly  in
 Section  208,  have been discussed by experts from outside of EPA.(261,262)

          Lastly, a methodology has been developed  for assessing  secondary impacts
of wastewater treatment facilities on urban ecosystems  in a project by The
of Ecology for EPA.(263)
                                       -  68  -

-------
                                     SECTION 8

                                    REFERENCES

Section 1

 1.  Godfrey, Kneoland A8J  Jr.,  "ASCE Tackles Land Use Planning," Civil Engineering.
     Vol. 44, No. 7, pp. 66-69,  July, 1974.

 2.  Guy, Harold P., and D. Earl Jones, Jr., "Urban Sedimentation — in Perspective,"
     J.Hyd.Div.. ASCE Proc., Vol. 98, No. HY12, pp. 2099-2116,  December, 1972.

 3.  Hall, Warren A., "The Research Interrelationships of the National Water Commission,
     The Water Resources Council and the Committee on Water Resources Research,"
     Report on the Fifth Annual  Water Resources Research Conference, February 3-4, 1970,
     Washington, D.C., Sponsored by the Office of Water Resources Research, p.  11.

 4.  U.S. EPA, Draft Guidelines  for Areawjde Waste Water Management, GPO, Washington,
     B.C., 126 pp., May, 1974.

 5.  Whipple, W., Jr., and J. V. Hunter, "Nonpoint Sources and Planning for Pollution
     Control," J.WPCF, Vol. 49,  No. 1, pp. 15-23, January, 1977.

 6.  McPherson, M. B., Prospects for Metropolitan Water Management. ASCE, New York,
     N.Y., 250 pp., December, 1970.

 7.  University of North Carolina, Chapel Hill, Promoting Environmental Quality
     Through Urban Planning and Controls. U.S. EPA Socioeconomic Environmental
     Studies Series EPA-600/5-73-015, GPO, Washington, D.C., 441 pp., February, 1974.

 8.  McPherson, M. B., and G. F. Mangan, Jr., "ASCE Urban Water Resources Research
     Program," J.Hyd.Div., ASCE Proc., Vol. 101, No. HY7, pp. 847-855, July, 1975.
     Authors' closure to the 27 discussions:  Vol. 103, No. HY6, pp. 661-663,
     May, 1977.

 9.  McPherson, M. B., Regional Earth Science Information in Local Water Management.
     ASCE, New York, N.Y., 155 pp., July, 1975.

10.  Smithsonian Science Information Exchange, Research Information Services for
     Energy and the Environmental Sciences. Room 300, 1730 M Street, N.W., Washington,
     D.C. 20036, 36 pp., 1977.

11.  Selected Water Resources Abstracts,  issued twice per month, available  from Office
     of Water Research and Technology, Water  Resources Scientific  Information Center,
     Washington, D.C. 20240, and from  the National Technical Information Service,
     5285 Port Royal Road, Springfield, Virginia 22161.

12.  Annual Cumulated Indexes. Selected Water Resources Abstracts.  Water Resources
     Scientific Information Center, Office of Water Research and Technology,
     Washington, D.C. 20240.  (In  two volumes:  Part  1, Author, Organization and
     Accession Number; and Part  2, Subject.   For 1976, there are 700  pp. in Part  1
     and  1,201 pp.  in Part 2).
                                       -  69  -

-------
 Section 2

 13.  (Communication from Harold C. Coffee, Jr., Department of Public Works,  City
      and County of San Francisco, California).

 14.  Department of Public Works, "Preliminary Summary Report," San Francisco Master
      Plan for Waste Water Management,  City and County of San Francisco,  California
      147 pp., September 15,  1971.                                                 '

 15.  Friedland, Alan 0., "City-Wide Master Planning," Urban Runoff.  Quantity and
      Quality, ASCE, New York,  N.Y., pp.  8-12,  1975.                          	

 16.  McPherson, M. B.,  "Innovation: A Case Study,"  ASCE Urban Water Resources
      Research Program Technical Memorandum No. 21, ASCE,  New York,  N.Y.,  59  pp.
      February, 1974.  (Available from  NTIS as  PB 232 166).                     "'

 17.  Department of Public Works, "Supplement I," San Francisco Master Plan for
      Waste Water Management, City and  County of San  Francisco,  California 88 pp
      May 15,  1973.                                                               '

 18.  Roesner, L.  A., "Real Time Automatic  Control of San  Francisco's  Combined
      Wastewater Collection System," a  paper presented at  the Storm  Water  Management
      Model User's Group meeting at Toronto,  Ontario,  October 22, 1976, 26 pp.

 19.   (Communication from Neil  S. Grigg,  University of North Carolina  Water Resources
      Research Institute,  North Carolina  State  University, Raleigh).

 20.   Golding,  Bernard L.,  Discussion of  "Master Planning  Methodology  for  Urban
      Drainage," by Harold F. Bishop, J.Hyd.Djv.. ASCE Proc.,  Vol. 100, No. HY9
      pp.  1301-1303,  September,  1974.                                          '

 21.   Me tealf  and  Eddy,  Inc., Palo  Alto,  California, Urban Stormwater  Management anH
      Technology;  An Assessment,  Environmental  Protection Technology  Series  EPA-670/2-
      74-040,  GPO,  Washington,  D.C.,  447  pp., December, 1974.   (Available  from NTIS  as
      PB  240 687).

 22.  Overton,  Donald E.,  and Michael E.  Meadows, Stormwater  Modeling. Academic Press
     New York,  358  pp.,  1976.                                                       *

 23.  Southeastern Wisconsin Regional Planning Commission, "A Comprehensive Plan for
      the Milwaukee  River Watershed," Vol. One, "Inventory Findings and Forecasts "
     Planning  Report No.  13, Waukesha,  514 pp., December, 1970.

24.  Southeastern Wisconsin Regional Planning Commission, "A Comprehensive Plan for
     the Milwaukee River Watershed," Vol. Two, "Alternative  Plans and Recommended
     Plans," Planning Report No. 13, Waukesha, 625 pp., October, 1971.

25.  Southeastern Wisconsin Regional Planning Commission, "Prospectus, Preliminary
     Engineering Study for the Abatement of Pollution from Combined Sewer Overflow
     in the Milwaukee Metropolitan Area," Waukesha,  36 pp., July, 1973.

26.  Harza, Richard D., and Ramon S. LaRusso, "Deep Tunnel Technology," APWA
     Reporter. Vol. 40, No. 9,  pp. 11-13, September,  1973.
                                      - 70 -

-------
27.  Anonymous,  "Tunnels May Store Storrawater for Later Treatment,"  The American
     City. Vol.  88,  No.  12,  p.  75, December,  1973.

28.  Metropolitan Sewerage District of the County of Milwaukee and Stevens,  Thompson
     & Runyan,  Inc., "Phase One,  Volume One,  Technical Report, Combined Sewer  Overflow
     Pollution Abatement," Milwaukee,  Wisconsin,  114 pp.,  1975.

29.  Metropolitan Sewerage District of the County of Milwaukee and Stevens,  Thompson
     6e Runyan,  Inc., "Phase One,  Volume Two,  Project Design,  Combined Sewer  Overflow
     Pollution Abatement," Milwaukee,  Wisconsin,  114 pp.,  1975.

30.  Michel, Henry L., and William P.  Henry,  "Flood Control and Drainage  Planning in
     the Urbanizing Zone:  Fairfax County, Virginia," Urban Runoff.  Quantity and
     Quality. ASCE,  New York, N.Y., pp. 119-139,  1975.

31.  Michel, H.  L.,  and W. P. Henry, "Fairfax County Works to Solve  Runoff Problems,"
     Water & Wastes Engineering.  Vol.  13, No. 9,  pp. 38-42, September, 1976.

32.  VanWeele, Brian, "Management of Flood Control and Drainage in the Urbanizing
     Zone," Proceedings, National Symposium on Urban Rainfall and Runoff and Sediment
     Control, Office of Research and Engineering Services Publication UKY B106,
     University of Kentucky, Lexington, pp. 229-238, October, 1974.

33.  Allee, Debra C., "Environmental Baseline Relationship Matrix," Urban Runoff,
     Quantity and Quality. ASCE,  New York, N.Y.,  pp. 150-159, 1975.

34.  Lanyon, Richard, and Jerome Kurland, "Monitoring System Boosts Chicago Water
     Data Collection System," Water & Wastes Engineering. Vol. 8, No. 1,  pp. 20-23,
     January, 1971.

35.  Flood Control Coordinatinp Committee, "The Chicago Underflow Plan — Development
     of a Flood and Pollution Control Plan for the Chicagoland Area," State of
     Illinois, County of Cook, MSDGC, City of Chicago, 29 pp., December, 1972.

36.  MSD NEWS, Vol. 3, No. 3, 4 pp., Fall, 1972.

37.  Lanyon, Richard, "Flood Plain Management in Metropolitan Chicago," Civil
     Engineering. Vol. 44, No. 5, pp.  79-81, May, 1974.

38.  Lanyon, Richard F., and James Jackson,  "Flow Simulation  System," J.Hyd.Div.,
     ASCE Proc., Vol. 100, No. HY8, pp.  1089-1105, August, 1974.

39.  Denver  Regional Council of Governments  and Urban Drainage and  Flood Control
     District, "Urban Storm Drainage  and Flood Control in the Denver Region,"  Project
     REUSE Final Report,  219 pp., August,  1972.

40.  Denver  Regional Council of Governments, "Storm Drainage  and  Flood Control for
     Metropolitan Denver," Project REUSE Summary Report,  18  pp.,  May, 1973.

41.  Bishop, Harold F.,  "Master Planning Methodology for  Urban Drainage," J.Hyd.Div.,
     ASCE Proc., Vol. 100, No. HY1, pp.  189-199, January, 1974.   Author's closure to
     discussion:  Vol.  101,  No. HY4,  p.  412, April,  1975.
                                         - 71 -

-------
 42.  Roper, Willard, "Wastewater Management Studies by the Corps of Engineers "
      J.Envir.Engrg.Div.,  ASCE Proc.,  Vol. 99,  No. EE5, pp. 653-669, October,  1973.
      Author's closure to  discussion:   Vol. 101,  No. EE1,  p. 164, February,  1975.

 43.  Department of Defense,  Corps of  Engineers,  U.S. Army, "Urban Studies  Program
      Proposed Policies and Procedures," Federal  Register,  Vol.  39,  No.  130, Part  III
      pp.  24754-24771, Friday,  July 5,  1974.                                         '

 44.  Corps of Engineers,  "San Francisco Bay and  Sacramento-San  Joaquin  Delta  Water
      Quality and Waste Disposal  Investigation, Public Brochure,  Wastewater  Management
      Technical Alternatives,"  U.S.  Array Engineer District, San  Francisco, 30  pp.
      August, 1972.

 45.  Michel, H. L,  P. H.  Gilbert and  H. K. Creed,  "Land Disposal, What's the
      Realistic View?," Water & Wastes  Engineering,  Vol. 11,  No.  6,  pp.  30-35,
      June, 1974.

 46.  Boyce,  David E., Norman D.  Day and Chris McDonald, Metropolitan Plan Making
      "An Analysis of Experience  with  the Preparation and Evaluation of Alternative
      Land  Use and Transportation Plans," Regional  Science  Research  Institute
      Monograph Series Number Four,  Philadelphia,  Petina., 450 pp., 1970.

 47.  Banks,  Harvey  0»,  and Jean  0.  Williams, "Water-Resources Planning," J.AWWA.
      Vol.  65,  No.  11, pp.  665-669,  November, 1973.

 Section 3

 48.  American  Water Resources Association,  Urbanization and  Water Quality Control
      Edited  by William Whipple,  Jr., Minneapolis, Minnesota, 302 pp., 1975.

 49.  Citizens'  Advisory Committee on Environmental Quality,  How Will America Grow?
     A Citizen Guide  to Land-Use Planning,  GPO Stock No. 040-000-00347-9, Washington
     D.C., 39  pp., April,  1976.                                                      '

 50.  Walker, William  R., and William E.  Cox, "Water - An Element of Land-Use and
     Urban Growth Policies," J.Urban Plan.&Dev.Div.. ASCE  Proc., Vol. 102,  No. UP1,
     pp. 81-94a August, 1976.

51.  Lienesch,  William  C., and Gerald A. Emison,  "Planning for Implementation Under
     Section 208," J.Wat.Res.Plan.&Mgt.Div.. ASCE Proc., Vol. 102, No. WR2, pp. 283-
     295, November, 1976.

52.  Pisano, Mark A., "Nonpoint  Sources  of  Pollution:  A Federal Perspective,"
     J.Envir.Engrg.Div., ASCE Proc., Vol.  102,  No. EE3, pp. 555-565, June,  1976.
                                                                               I
53.  Pisano, Mark, "208:  A Process for Water Quality Management," Civil Engineering
     Vol. 46, No. 11, pp.  55-57,  November,  1976.

54.  Billings, Leon G., "The Evolution of  208 Water-Quality Planning," Civil
     Engineering. Vol. 46, No. 11, pp. 54-55, November, 1976.

55.  "Areawide Assessment Procedures Manual," in  three volumes, report EPA-600/9-76-014
     Municipal Environmental Research Laboratory, Office of Research and Development   '
     U.S. EPA, Cincinnati, Ohio 45268, July, 1976, et seq.                           *
                                      - 72 -

-------
56.  Flannery, James J.,  "New Publications," EOS.  Transactions of AGU,  Vol.  57,
     No. 7, pp. 510-515,  July, 1976.

57.  Peskin, Henry M.,  and Eugene P.  Seskin, Cost  Benefit Analysis and  Water
     Pollution Policy.  Urban Institute, Washington,  D.C., 370 pp., 1975.

58.  Gulp, Gordon L., and Leonard S.  Horner, "Coping with EPA's Value Engineering
     Requirement," Water & Sewage Works. Vol. 123, No. 12, pp. 48-51, December,  1976.

59.  Ehrenfield, David W., "The Conservation of Non-Resources," American Scientist,
     Vol. 64, No. 6, pp.  648-656, November-December, 1976.

60.  Lindh, Gunnar, Socjo-Economic Aspects of Urban Hydrology, The Unesco Press,
     Paris, 1977 (in press).

61.  Cox, William E., and William R.  Walker, "Recognition of Aesthetic Values in
     Water Law," J.AWWA.  Vol. 69, No. 1, pp. 31-34,  January, 1977.

62.  Hill, Douglas, "A Resource Allocation Model for the Evaluation of Alternatives
     in Section 208 Planning Considering Environmental, Social and Economic Effects,"
     pp. 401-406 in Environmental Modeling and Simulation. Office of Research and
     Development and Office of Planning and Management, U.S. EPA, report EPA-600/9-
     76-016, 847 pp., July, 1976.  (Available from NTIS as PB 257 142).

63.  Water Resources Scientific Information Center, OWRT, Water Reuse. A Bibliography,
     GPO, Washington, D.C., Vol. 3 (444 pp.) and Vol. 4 (448 pp.), March, 1975.

64.  Dworkin, Daniel M., "Water Reuse:  A Flexible and Efficient  Management Alternative
     for Municipal Water Supply," Water Resources Research. Vol.  11, No. 5, pp. 607-
     615, October, 1975.

65.  SCS Engineers, Long Beach, California, Demonstrated Technology and Research
     Needs for Reuse of Municipal Wastewater. U.S. Environmental  Protection Agency
     report  EPA-670/2-75-038, GPO, Washington, D.C.,  338  pp., May, 1975.  (Available
     from NTIS as PB 249 151).

66.  Bargman, Robert D., George W. Adrian and Donald  C. Tillman,  "Water Reclamation
     in Los  Angeles," J.Envir.Engrg.Div.. ASCE Proc., Vol. 99, No. EE6, pp. 939-955,
     December,  1973.

67.  Seaburn, G. E., and D. A. Aronson,  Influence of  Recharge Basins on the Hydrology
     of Nassau  and Suffolk Counties, Long Island, New York, U.S.  Geological Survey
     Water Supply Paper  2031, GPO, Washington, D.C.,  66  pp.,  1974.

68.  Hulob,  H.,  "Some Legal Problems  in Urban Runoff," in Hydrology  and Water
     Resources  in Arizona and the Southwest, pp.  169-179, Vol.  2, Proceedings of the
     1972  meetings of  the Arizona Section,  American Water Resources  Assn.,  and  the
     Hydrology  Section, Arizona Academy of  Science, Prescott, Arizona,  1972.

69.  McPherson,  M. B., "Management Problems in Metropolitan Water Resource Operations,"
     ASCE  Urban Water  Resources  Research Program  Technical  Memorandum No.  14, ASCE,
     New York,  N.Y., 76  pp.,  September,  1971.  (Available from NTIS  as PB  206  087).

70.  Schneider,  W.  J., and Andrew M.  Spieker, Water for  the Cities — the Outlook.
     U.S.  Geological Survey Circular 601-A, GPO,  Washington,  D.C., 6 pp.,  1969.


                                       - 73  -

-------
 71.  Appelj Charles A., and John D. Bredehoeft, Status of Ground-Water Madeline -in
      the U.S. Geological Survey. U.S. Geological Survey Circular 737, GPO, Washington
      D.C., 9 pp., 1976.                                                              '

 72.  Todd, D. K., R. M. Tinlin, K. D. Schmidt and L. G. Everett, "A Groundwater-
      Quality Monitoring Methodology," J.AWWA. Vol. 68, No. 11, pp. 586-593, November
      1976.

 73.  Todd, David K., and Daniel E. 0. McNulty, Polluted Groundwater. Water Information
      Center, Port Washington,  N.Y., 178 pp., 1976.

 74.  Geyer, J.  C., and J. J. Lentz, "An Evaluation of the Problems of Sanitary Sewer
      System Design," J.WPCF. Vol. 38, No.  7, pp. 1138-1148,  July,  1966.

 75.  American Public Works Association, Control of Infiltration and Inflow into Sewer
      Systems, Water Pollution Control Research Series 11022  EFF 12/70, GPO, Washington
      D.C., 121  pp.,  December,  1970.  (Available from NTIS as PB 200 827).              *

 76.  Office of  Water Program Operations, Guidance for Sewer  System Evaluation.  U.S.
      EPA report EPA-520/9-74-018, GPO,  Washington, D.C.,  10  pp., March,  1974.

 77.  Cesareo, David J.,  and Richard Field,  "Infiltration-Inflow Analysis," J.Envjr.
      Engrg.Div.. ASCE Proc., Vol. 101,  No.  EE5,  pp.  775-785,  October,  1975. Authors'
      closure:  Vol.  102,  No. EE5, pp. 1126-1127,  October,  1976.

 78.  Click,  C.  N.,  and F.  0. Mixon,  "Flow Smoothing in Sanitary  Sewers," J.WPCF
      Vol.  46, No.  3,  pp.  522-531, March, 1974.                                  '

 79.  Beckwith,  R.  A.,  "Infiltration/Inflow  Program for EPA Grants," pp.  R-16 -  R-18-
      P.  A.  Podolick,  "Preparing an Infiltration/Inflow Analysis,"  pp.  R-31 - R-34-
      and D.  J.  Cesareo and R.  Field,  "How to Analyze Infiltration/Inflow," pp.  R-84
      R-90:   Water  &  Sewage Works, Reference Number,  1975.

 80.   Office  of  Water  Program Operations, Handbook for Sewer  System Evaluation and
      Rehabilitation,  U.S.  EPA  Municipal Construction Division Technical Report  MCD-19
      EPA 430/9-75-021, Washington, D.C., 218 pp.,  December,  1975.                     '

 81.   Klemetson,  Stanley L.,  and  William J.  Grenney,  "Development of a Dynamic
      Programming Model for  the Regionalization and Staging of Wastewater Treatment
      Plants," Utah Water Research Laboratory report  PRWA20-2,  Utah State University
     Logan,  124  pp., June,  1975.  (See J.WPCF. Vol. 48, No. 12, pp. 2690-97, Dec. 19*76)

 82.  Weeter,  Dennis W., and  James  G.  Belardi, "Analysis of Regional  Waste  Treatment
     Systems," J.Envir.Engrg.Div.. ASCE Proc., Vol.  102, No.  EE1,  pp. 233-237,
     February, 1976.

83.  Dajani,  Jarir S., and  Robert  S. Gemmell, "Economic Guidelines  for Public
     Utilities Planning," J.Urban  Plan.&Devel.Div.. ASCE Proc., Vol. 99,  No. UP2,
     pp. 171-182, September, 1973.

84.  Anonymous,  "Symposium Studies Solutions  to Regional Wastewater  Problems,"
     "STRIDES,"  Stevens, Thompson & Runyan,   Inc., Portland, Oregon,  3rd Quarter,
     8 pp.,  1976.
                                      - 74 -

-------
85.  Anonymous, "Builders Assault the No-Growth Laws," Business Week,  pp.  44  and  46,
     September 8,  1973.

86.  Rabe, Frank T., and James F. Hudson,  "Highway and Sewer Impacts on Urban
     Development," J.Urban Plan.&Devel.Div..  ASCE Proc.,  Vol. 101,  No. UP2, pp.  217-
     231, November, 1975.

87.  Real Estate Research Corporation, The Costs of Sprawl.  GPO, Washington,  D.C.,
     278 pp., April, 1974.  (An "Executive Summary," 15 pp., is also available
     from GPO).

88.  Roberts, James S., "Energy and Land Use:  Analysis of Alternative Development
     Patterns," Environmental Comment, Urban Land Institute, Washington, D.C.,
     pp. 2-11, September, 1975.

89.  Wolman, M. Gordon, "The Physical Environment and Urban Planning,"
     Geographical Perspectives and Urban Problems. National Academy of Sciences,
     pp. 55-70, 1973.

90.  Ports, Michael A., "Sediment and Erosion Control Design Criteria," APWA
     Reporter. Vol. 42, No. 5, pp. 18-19, May, 1975.

91.  Sachs, Herbert M., "The State Water Resources Planning Program in Maryland,"
     Proceedings. Symposium on Water Resources Needs Facing the District of
     Columbia, Washington Technical Institute, WRRC Report No. 3, Washington, D.C.,
     pp. 17-28, June 20, 1974.

92.  Water Resources Administration, "The Maryland Sediment Control Program,"
     Department of Natural Resources, Annapolis, 8 pp., August, 1972.

93.  McCuen, Richard H., "A Regional Approach  to Urban Storm Water Detention,"
     Geophysical Research Letters. AGU, Vol. 1, No. 7, pp.  321-322, November, 1974.

94.  Parsons,  Brinckerhoff, Quade & Douglas, New York City,  "Erosion,  Sedimentation
     & Debris  Control," Task Order 11.0, Master Plan for Flood Control  and Drainage,
     Fairfax County, Virginia, 183 pp., May, 1975.

95.  Chen, Charng-Ning, "Evaluation and Control of Soil Erosion in Urbanizing
     Watersheds,"  Proceedings, National Symposium on Urban  Rainfall and Runoff and
     Sediment  Control,  Office  of Research and  Engineering Services  Publication
     UKY B106, University of Kentucky, Lexington, pp.  161-173, October, 1974.

96.  Chen, Charng-Ning,  "Planning Tools  for  Erosion Control in Urbanizing
     Watersheds,"  Urban Runoff.  Quantity and Quality. ASCE,  New York,  N.Y.,  pp.  159-
     165,  1975.

97.  Powell,  M. D.,  W.  C. Winter and  W.  P. Bodwitch,  "Community Action Guidebook for
     Soil  Erosion  and  Sediment Control," The American County.  Vol.  35, No. 5, 64 p.
     insert,  May,  1970.

98.  Guy,  Harold  P., Sediment  Problems  in Urban Areas. U.S. Geological Survey
     Circular 601-E, GPO, Washington,  D.C.,  8  pp.,  1970.

99.  Hittman Associates,  An Executive Summary  of Three EPA Demonstration Programs
      in Erosion and Sediment Control.  Environmental  Protection Technology Series,
      Report EPA-660/2-74-073,  GPO,  Washington, D.C.,  50  pp., June, 1974.  (Available
      from NTIS as  PB 239  333).
                                      - 75  -

-------
 100.   ULI, ASCE and  NAHB,  Residential  Storm Water Management. ASCE, New York, N.Y.
       64 pp.,  1975.                                                                '

 101.   Task Committee,  "Urban Sediment  Problems:   A Statement on  Scope, Research,
       Legislation, and Education,"  J.Hyd.Div.. ASCE Proc., No. HY4, pp. 329-340,
       April,  1975.   Committee's  closure  to  discussion:  Vol. 102, No. HY10, pp.'l598-
       1599, October,  1976.

 102.   Dallaire,  Gene,  "Controlling  Erosion  and Sedimentation at  Construction Sites "
       Civil Engineering. Vol. 47, No.  10, pp. 73-77, October, 1976.                '

 Section  4

 103.   Schneider, William J.,  and James E. Goddard,  Extent and Development o£ Urban
       Flood Plains,  U.S. Geological Survey  Circular 601-J, GPO,  Washington, D.C.,
       14 pp.,  1974.

 104.   "Proceedings of  a Seminar on Nonstructural  Flood Plain Management Measures,"
       4-6 May  1976,  The Hydrologic Engineering Center, U.S. Army Corps of Engineers
       609 Second Street, Davis, Cal. 95616, and Institute for Water Resources, U.S.*
      Army Corps of  Engineers, Kingman Building,  Fort Belvoir, Virginia 22060, 193 pp.

 105.   Goddard, James E., "An Evaluation of  Urban  Flood Plains," ASCE Urban Water
       Resources Research Program Technical  Memorandum No. 19, ASCE, New York, N.Y.
       40 pp.,  December, 1973.  (Available from NTIS  as PB 227 337).

 106.   Poertner, H. G., Robert L. Anderson and Karl  W. Wolf, Urban Drainage Practices
       Procedures and Needs. APWA Research Foundation, Special Report No. 31, Chicago
       Illinois, 54 pp., December, 1966.

 107.  Thomas,  H. E., and W. J. Schneider, Water as  an Urban Resource and Nuisance.
      U.S. Geological Survey, Circular 601-D, GPO,  Washington, D.C., 9 pp., 1970.

 108.  Sheaffer, J. R., D. W. Ellis and A. M. Spieker, Flood-Hazard Mapping in
      Metropolitan Chicago. U.S. Geological Survey, Circular 601-C, GPO, Washington,
      D.C.,  14 pp., 1970.

 109.  Wood,  Walter J., "Los Angeles County Flood Control System and the Early 1969
      Storms," Civil Engineering. Vol. 40, No. 1, pp. 58-61, January,  1970.

 110.  Rantz,  S. E., Urban Sprawl and Flooding in Southern California.  U.S. Geological
      Survey,  Circular 601-B, GPO,  Washington, D.C., 11 pp., 1970.

 111.  Grigg,  N. S., "Evaluation and Implementation of Urban Drainage Projects,"
      J.Urban Plan.&Devel.. ASCE Proc., Vol. 101, No. UP1, pp. 61-75,  May, 1975.

112.  Walesh,  Stuart G., "Floodland Management:   The Environmental Corridor Concept,"
      Technical Record, Vol. 3,  No. 6, Southeastern Wisconsin Regional Planning
      Commission, Waukesha, pp.  1-13,  April, 1976.

113.  Walesh,  Stuart G., "Floodland Management:   The Environmental Corridor Concept,"
      Hydraulic Engineering and the Environment.  ASCE, New York,  N.Y., pp. 105-111, 1973

114.  Wendell, Mitchell, and Harvey 0. Banks, "Management of Water Resources for Urban
      Environment Enhancement," Environments for Tomorrow, Inc.,  McLean, Virginia,
      275 pp., July, 1974.   (Available from NTIS as PB 239 301).

                                      - 76  -

-------
115.  Bauer, W. J«, "Economics of Urban Drainage Design," J.Hyd.Div.. ASCE Proc.,
      Vol. 88, No. HY6,  pp.  93-114,  November,  1962.   Author's  closure  to discussion:
      Vol. 90, No. HY1,  p. 257, January, 1964.

116.  Rice, Leonard, "Reduction of Urban Peak Flows  by Ponding," J.Irrig.&Drajn.Div.,
      ASCE Proc., Vol. 97, No. IR3,  pp. 469-482, September,  1971.

117.  Kalvinskas, John J., Wallace A.  Moser and Thomas E. Bullock,  "Water Reservoir
      Systems," Water Resources Bulletin. Vol. 7, No. 2,  pp. 330-342, April,  1971.

118.  Poertner, Herbert G.,  "Better Storm Drainage Facilities  —At  Lower Cost,"
      Civil Engineering. Vol. 43, No.  10, pp.  67-70, October,  1973.

119.  Poertner, Herbert G.,  Practices  in Detention of Urban Stormwater Runoff.  APWA
      Special Report No. 43, Chicago,  Illinois, 231 pp.,  June, 1974.  (Available from
      NTIS as PB 234 554).

120.  Poertner, Herbert G.,  "Urban Stormwater Detention and Flow Attenuation,"  Public
      Works. Vol. 107, No. 8, pp. 83-85, August, 1976.

121.  Becker, B. C., M. L. Clar and R. R. Kautzman,  "Approaches to  Stormwater
      Management," Hittman Associates, Columbia, Maryland, 258 pp., November, 1973.
      (Available from NTIS as PB 228 124).

122.  Hollis, G. E., "The Effect of Urbanization on Floods of Different Recurrence
      Interval," Water Resources Research. Vol. 11, No. 3, pp. 431-435, June, 1975.

123.  Lumb, A. M., J. R. Wallace and L. D. James, "Analysis of Urban Land Treatment
      Measures for Flood Peak Reduction," Georgia Institute of Technology, Georgia
      Environmental Resources Center Report No. ERC-0574, Atlanta,  146 pp., June,
      1974.  (Available from NTIS as PB 236 418).

124.  Hydrologic Engineering Center, Phase I Oconee Basin Pilot Study, Trail Creek
      Test. An Investigation of Concepts and Methods for Broadened Scope Flood Plain
      Information Studies, U.S. Army Corps of Engineers, Davis, California, 98 pp.,
      September, 1975.

125.  Walesh, S. G., and R. M. Videkovich, "Urbanization:  Hydrologic, Hydraulics
      and Flood Damage Effects," a paper presented at  the ASCE Hydraulics Division
      Conference, August 4-7,  1976, West Lafayette,  Indiana,  29 pp.

126.  Benjes, Henry H., Jr., Cost Estimating  Manual —  Combined Sewer Overflow Storage
      and Treatment, Environmental Protection Technology  series  EPA-600/2-76-286,
      U.S. EPA Municipal Environmental  Research Laboratory, Cincinnati, Ohio, 123 pp.>
      December,  1976.   (Available  from NTIS as  PB 266  359).

127.  Rawls, Walter J., and John N. Knapp, "Methods  for  Predicting Urban  Drainage
      Costs," J.Hyd.Div.. ASCE Proc.,  Vol. 98,  No. HY9,  pp. 1575-1585,  September,  1972,

128.  Kohlhaas,  Charles A., "The Optimization of Storm-Holding  Tanks:   A  Problem of
      Water  Pollution Control," Stanford University, California, PhD dissertation,
      321  pp., February,  1970.

129.  Rickert, David A.,  and Andrew M.  Spieker,  Real-Estate Lakes. U.S. Geological
      Survey Circular 601-G,  GPO, Washington, D.C.,  19 pp., 1971.


                                       - 77  -

-------
 130.  Britton, L. J., R. C. Averett and R. F. Ferreira,  An Introduction to the
       Processes. Problems, and Management of Urban Lakes.  U.S.  Geological Survey
       Circular 601-K, GPO, Washington,  D.C., 22 pp.,  1975.

 131.  Anonymous, Man-Made Lakes:   Their Problems and  Environmental  Effects. Geophysical
       Monograph Series Volume 17,  American Geophysical Union, Washington,  D.C.,
       900 pp., 1973.

 132.  Metcalf & Eddy, Inc., Palo Alto,  California,  Development  and  Application of  a
       Simplified Stormwater Management  Model. Environmental Protection  Technology
       Series  EPA-600/2-76-218,  U.S.  EPA Municipal Environmental Research Laboratory,
       Cincinnati, Ohio,  139 pp., August,  1976.   (Available from NTIS  as PB 258 074)!

 133.  Field,  Richard, Anthony N. Tafuri and Hugh E. Masters, Urban  Runoff Pollution
       Control Technology Overview.  Environmental Protection Technology  Series
       EPA-600/2-77-047,  Municipal  Environmental Research Laboratory,  Cincinnati,
       Ohio 45268, 91  pp.,  March, 1977.

 134.  Heaney,  James P.,  and Stephan  J.  Nix,  Storm Water  Management  Model:  Level I  —
       Comparative Evaluation of Storage-Treatment and  Other Management  Practices.
       Environmental Protection  Technology Series EPA-600/2-77-083,  Municipal Environmental
       Research Laboratory,  U.S. EPA, Cincinnati,  Ohio  45268, 88  pp., April, 1977.

 135.  Whipple,  William,  Jr.,  "Water  Quality:  Problems and Research Needs," Water &
       Sewage  Works. Vol.  123, No.  12, pp.  40-41,  December,  1976.                   *

 136.  Weathers,  John  W.,  Editor, Flood  Damages  Prevention. An Indexed Bibliography
       Tennessee Valley Authority, Knoxville,  Tennessee 37902, and University of
       Tennessee Water Resources Research Center,  Knoxville, Tennessee 37916, 61 pp.
       October,  1976.

 137.   Miller,  William L., and Steven P. Erickson, "Systematic Development  of
       Methodologies in Planning Urban Water Resources  for  Medium Size Communities,
       Economic  and Environmental impacts of Surface Runoff Disposal Systems," Purdue
       University, Water Resources Center Technical Report No. 39, West Lafayette,
       Indiana,  74 pp., December, 1973.  (Available from NTIS as  PB  234 315).

 138.  Debo, T. N., "Survey and Analysis of Urban Drainage Ordinances and a Recommended
      Model Ordinance," Georgia Institute of Technology,  Environmental Research Center
      Report No. ERC-0475, Atlanta, 294 pp., February, 1975.  (Available from NTIS as
      PB 240 817).

Section 5

139.  Gray, Donald M., Editor, Handbook on the Principles of Hydrology.  Water
      Information Center, Port Washington, N.Y., 720 pp., 1973.

140.  Water Resources  Engineers, Walnut Creek, California,  Future Direction of Urban
      Water Models, Environmental Protection Technology Series EPA-600/2-76-058,
      U.S. EPA Municipal Environmental Research Laboratory, Cincinnati,  Ohio,  86 pp.
      February, 1976.   (Available from NTIS as PB 249  049).      "                  *'

141.  McPherson, M. B., "Urban Hydrological Modeling and  Catchment Research in the
      U.S.A.," ASCE Urban Water Resources Research Program Technical Memorandum No.
      IHP-1, ASCE, New York, N.Y.,  49 pp., November, 1975.   (Available from NTIS  as
      PB 260 685).

                                        - 78 -

-------
142.  Whipple,  William,  Jr.,  "Conference Summary,"  Urban Runoff.  Quantity  and Quality.
      ASCE, New York,  N.Y.,  pp.  255-258, 1975.

143.  American Society of Civil  Engineers,  Basic Information Needs  in Urban
      Hydrology. ASCE, New York,  N.Y.,  112  pp.,  April,  1969.

144.  Schneider, William J.,  "The U.S.  Geological Survey Urban Water Program,"  Effects
      of Watershed Changes on Streamflow. University of Texas Press, Austin, pp.  165-
      168, 1969.

145.  Committee Report,  "Water-Distribution Research and Applied  Development Needs,"
      J.AWWA. Vol. 66, No. 6, pp. 385-390,  June, 1974.

146.  Alley, William M., Guide for the Collection.  Analysis and Use of Urban
      Stormwater Data. ASCE,  New York,  N.Y., 115 pp., 1977.

147.  Radziul,  Joseph V., and Patrick R. Cairo,  Philadelphia Water Department,  "Case
      Study of Philadelphia,  Pennsylvania," paper presented at Engineering Foundation
      Conference on Instrumentation and Analysis of Urban Stormwater Data, Quantity
      and Quality, November 28-December 3,  1976, Easton, Maryland,  39 pp.

148.  Richardson, Steven L.,  "A Solution to Combined Sewer Overflows," Clearwaters.
      New York Water Pollution Control Association, Vol. 4, No. 4,  pp. 14-15,
      January,  1975.

149.  Yarnell,  D. L., Rainfall Frequency-Intensity Data. U.S. Department of Agriculture,
      Miscellaneous Publication 204, 1935.

150.  Hershfield, David M., Rainfall Frequency Atlas of the United States. Weather
      Bureau, U.S. Department of Commerce,  Technical Paper No. 40, May, 1961.

151.  Hathaway, G. A., "Design of Drainage Facilities," Transactions. ASCE, pp. 697-
      730, Vol. 110, 1945.

152.  McPherson, M. B., "Some Notes on the Rational Method of Storm Drain Design," ASCE
      Urban Water Resources Research Program Technical Memorandum No. 6, ASCE, New
      York, N.Y., 74 pp., January 22, 1969.  (Available from NTIS as PB 184 701).

Section 6

153.  McPherson, M. B., (Editor), "Utility of Urban Runoff Modeling," ASCE Urban Water
      Resources Research  Program Technical Memorandum No. 31, ASCE, New York, N.Y.,
      126  pp., July,  1976.   (Available  from NTIS as PB 261 460).

154.  Gburek, William J., Discussion of  "Hydrologic Consequences of Rainfall
      Augmentation," J.Hyd.Div.» ASCE Proc., Vol. 97, No.  12,  pp.  2114-2115,
      December,  1971.

155.  Black, Crow and Eidsness, Inc., and Jordan, Jones and  Goulding,  Inc., "Study
      and  Assessment  of the  Capabilities and Cost of Technology  for Control of
      Pollutant  Discharges from Urban Runoff,"  a report  for the  National  Commission
      on  Water  Quality, 380  pp., October, 1975.  (Available from NTIS  as  PB 247  391).
                                       -  79  -

-------
  156.   Proctor and  Redfern, Ltd., and James  F. MacLaren, Ltd., Storm Water Management
        Model Study. Volume II, Research Program  for  the Abatement of Municipal Pollution
        under Provisions of the Canada-Ontario Agreement on Great Lakes Water Quality
        Research Report No. 48, Environment Canada, Ottawa, 148 pp., September, 1976.*

  157.   Field, Richard, et al., Proceedings of Workshop on Microorganisms in Urban
        Stormwater,  Environmental Protection Technology Series EPA-600/2-76-244  U.S
        EPA, Cincinnati, Ohio, 120 pp., November, 1976.  (Available from NTIS as
        PB 263 030).

  158.   Department of Economics, Georgetown University, The Integrated Multi-Media
        Pollution Model, U.S. EPA Socioeconomic Studies Series EPA-600/5-74-020  GPO
       Washington, D.C., 259 pp., February, 1974.                             '    *

  159.  American Society of Civil Engineers, Urban Water Resources Research.  ASCE  New
       York, N.Y., 43 pp. + 588 pp.  appendices, September,  1968.*

 160.  Prawdzik,  Ted B., Milwaukee DPW,  "Environmental and Technical Factors for Open
       Drainage Channels in Milwaukee," ASCE Urban Water Resources Research Program
       Technical  Memorandum No. 12,  ASCE,  New York, N.Y.,  37  pp.,  February,  1970
       (Available from NTIS as PB 191 710).

 161.  Tucker,  L.  S.,  "Sewered Drainage  Catchments  in Major Cities," ASCE Urban Water
       Resources  Research Program Technical Memorandum No.  10, ASCE, New York  N.Y.
       71 pp.,  March 31,  1969.   (Available from NTIS  as  PB  184  705).          '   * *'

 162.  Roesner, L. A.,  et al.,  Water Resources  Engineers, The Hydrologic  Engineering
       Center/Corps  of  Engineers,  and Department  of Public  Works/City  and County of
       San Francisco, "A Model  for Evaluating Runoff-Quality  in Metropolitan Master
       Planning," ASCE.Urban Water Resources  Research Program Technical Memorandum
       No. 23, New York,  N.Y.,  73  pp., April, 1974.   (Available from NTIS as  PB  234 312
       or from the H.E.C.,  Davis,  Cal. 95616).

 163.   Hydrologic Engineering Center, Corps of Engineers, "Storage,  Treatment, Overflow
       Runoff Model  (STORM), Users'  Manual,"  Computer Program 723-S8-L7520,  609  2nd    '
       Street, Davis, California 95616, 170.pp.,  July, 1976.

 164.   Winner, Jochen, and Michael Shapiro, Discussion of "Urban Runoff Pollution
       Control ~ State-of-the-art,"  J.Envir.Engra.Div.. ASCiS  Proc., Vol.  102, No. EB1
       pp. 220-223, February, 1976.                                              *     '

 165.   Meta Systems, Inc., Land Use-Water  Quality Relationships. U.S. EPA, Water Quality
       Guidance report WPD 3-76-02,  Water  Planning Division,  233 pp., March, 1976.

 166.   Heaney, J. P., W. C. Huber and S. J. Nix,  Storm Water  Management Model;  Levoi T _
       Preliminary Screening Procedures.,  Environmental Protection Technology Series	"^
       EPA-600/2-76-275, U.S. EPA Municipal Environmental Research Laboratory,
      Cincinnati, Ohio, 77 pp., October,  1976.   (Available from NTIS as PB 259 916).

167.  Jalal, Kazi F., "Water Quality Impacts of Urbanization — A Methodology,"
      J.Envir.EnRrg.Div.. ASCE Proc., Vol. 103, No. ESI, pp.  49-57, February,.1977.

168.  Lager,  John A., "A Simplified  Stonnwater  Management Model  for Planning,"  pp. 259
      279, Short  Course Proceedings. Applications of  Stormwater Management  Models! 197!
      Environmental  Protection Technology  Series  EPA-600/2-77-065,  U?S.  EPA  Municipal
      Environmental  Research Laboratory, Cincinnati,  Ohio,  435 pp.,  March  1977
      (Available  from NTIS as  PB  265 321).                               '
                                      -  80 -

-------
169.  Young,  G. K.,  "Non-Point Source Impact and Urban Holding Capacity,"  pp. 1-98
      to 1-121 in Proceedings^ Urban Stormwater Management  Seminars  (Atlanta and
      Denver), U.S.  EPA,  Water Quality Management Guidance  report  WPD  03-76-04,
      Washington, D.C.,  482 pp.,  January,  1976.  (Available from NTIS  as PB 260 889).

170.  Young,  G. K.,  "Decision Perspectives on Urban Storm Water Pollution," Water
      Resources Research, Vol. 12,  No. 1,  pp. 94-100,  February, 1976.

171.  Espey,  W. H.,  Jr.,  and D. E.  Winslow, "The Effects of Urbanization on Unit
      Hydrographs for Small Watersheds, Houston, Texas," A  Report  to OWRR, TRACOR
      Document No. 68-975-U, Austin, Texas, 70 pp., September 25,  1968.   (Appendix,
      Document No. 68-1006-U, 244 pp.).

172.  Everhart, Ralph C., "New Town Planned Around Environmental Aspects," Civil
      Engineering, Vol.  43, No. 9,  pp. 69-73, September, 1973.

173.  Claycomb, Elmer L., "Urban Storm Drainage Criteria Manual from Denver,"  Civil
      Engineering, Vol.  40, No. 7,  pp. 39-41, July, 1970.

174.  Wright-McLaughlin Engineers,  "Urban Storm Drainage Criteria Manual," Denver
      Regional Council of Governments, Denver, Colorado, 1969.  (In sixteen sections).

175.  Hydrocomp International, Inc., "Simulation of Discharge and Stage Frequency
      for Flood Plain Mapping  in the North Branch of the Chicago River," Palo Alto,
      California, 75 pp., February, 1971.

176.  Crawford, Norman H., "Studies in the Application of Digital Simulation to
      Urban Hydrology," Hydrocomp International, Palo Alto, California, 100 pp.,
      September,  1971.  (Available  from NTIS as PB 206 715).

177.  Crawford, Norman H., "Computer Simulation for Design Criteria for Urban Flow
      Storage  Systems," Hydrocomp International, Palo Alto, California, 109 pp.,
      January, 1973.  (Available from NTIS  as  PB 218 827).

178.  Linsley, R., and N. Crawford, "Continuous Simulation Models in  Urban Hydrology,"
      Geophysical Research Letters. AGU, Vol.  1, pp. 59-62, May,  1974.

179.  Terstriep,  M. L., and  J. B. Stall,  "Urban Runoff by Road Research Laboratory
      Method," J.Hyd.Div.. ASCE Proc., Vol. 95, No. HY6, pp.  1809-1834, November,
      1969.   Authors' closure to discussion:   Vol. 97, No. HY4, pp. 574-579, April,
      1971.

180.  Stall,  John B., and  Michael L.  Terstriep,  Storm Sewer  Design —  An Evaluation
      of the  RRL Method, Environmental Protection Technology Series EPA-R2-72-068,'
      GPO, Washington, D.C., 73 pp.,  October,  1972.   (Available from  NTIS as PB 214 314),

181.  Terstriep,  M.  L.,  and J. B.  Stall,  The Illinois  Urban  Drainage  Area Simulator.
      ILLUDAS. Bulletin  58,  Illinois  State Water Survey, Urbana,  90 pp.,  1974.

182.  Lanyon, R.  F.,  and J.  P. Jackson,  Metropolitan  Sanitary District of Greater
      Chicago, "A Streamflow Model  for Metropolitan Planning and  Design," ASCE Urban
      Water  Resources Research Program Technical Memorandum No. 20, ASCE, New York,
      N.Y.,  49 pp.,  January, 1974.   (Available from NTIS as PB 232  181).   (A truncated
      presentation  is included in  Reference 38).
                                        - 81 -

-------
 183.  Jennings, Marshall E.,  and Harold C. Mattraw,  "Comparison of the Predictive
       Accuracy of Models of Urban Flow and Water-Quality Processes," Proceedings,
       National Symposium on Urban Hydrology, Hydraulics and Sediment Control,
       Lexington, Kentucky,  pp.  1-8,  July,  1976.

 184.  University of Florida,  Storm Water Management  Model User's Manual.  Version II
       Environmental Protection  Technology Series EPA-670/2-75-017,  U.S. EPA,
       Cincinnati, Ohio 45268, 350 pp.,  March,  1975.

 185.  Bureau of Engineering,  City of Chicago,  "Computer Simulation Programs,
       Development of a Flood  and Pollution Control Plan for the Chicagoland Area,"
       Technical Report Part 2,  101 pp.,  December, 1972.  (Available from  NTIS  as
       PB 236 645).

 L86.  Metcalf and Eddy,  Inc., Palo Alto,  California,  "Application of Storm Water
       Management Model to Selected Chicago Drainage Areas - Phase 1," February 16,
       1971;  and "Application  of EPA  Storm Water  Management Model to the Chicago
       River  System - Phase  2,"  April, 1972.

 187.  Roesner,  Larry A.,  David  F.  Kibler  and John R.  Monser, "Use of Storm Drainage
       Models in Urban Planning," National  Symposium on  Watersheds in Transition,
       June 19-22, 1972,  Colorado State University, Proceed ings.  American  Water
       Resources Association, Urbana,  Illinois, pp. 400-405,  1973.

 188.  Water  Resources Engineers,  "The Use  of Stormwater Simulation  Models in
       Developing a Coordinated  Wastewater  Management  Plan for the City of San
       Francisco," Summary Report,  prepared for Department of Public Works, City  and
       County of San Francisco,  Walnut Creek, California,  31  pp.,  September, 1974.

 189.   Torno,  Harry C., "Storm Water Management Models," Urban Runoff,  Quantity and
       Quality.  ASCE,  New  York,  N.Y., pp. 82-89,  1975.

 190.   Sarma,  P.B.S.,  J. W. Delleur and A.  R. Rao, "Comparison of Rainfall-Runoff
       Models  for  Urban Areas,"  J.Hydrology.  Vol. 18,  pp.  329-347, 1973.

 191.   Papadakis,  Constantine N., and Herbert C.  Preul,  "Testing of  Methods for
       Determination  of Urban Runoff," J.Hyd.Djv.. ASCE  Proc., Vol.  99, No. HY9,
       pp. 1319-1335,  September,   1973.  Authors'  closure to discussion:  Vol. 101,
      No. HY3,  pp. 542-544,  March, 1975.

 192.  Keeps,  David P., and Russell G. Mein,  "Independent  Comparison of Three Urban
      Runoff  Models," J.Hyd.Div.. ASCE Proc., Vol. 100, No. HY7,  pp.  995-1009, July,
      1974.,  Authors' closure to discussion:  Vol. 102, No. HY1,  pp.  93-94, January, 1975

 193.  Brown, J, W., M. R. Walsh, R. W. McCarley, A.  J.  Green and H.  W. West, "Models
      and Methods Applicable to  Corps of Engineers Urban  Studies,"  U.S. Army Engineer
      Waterways Experiment Station Misc. Paper H-74-8,  Vicksburg,.Mississippi, 420  pp.
      August, 1974.                                                                  "'

194.  Marsalek, J., T. M. Dick,  P. E. Wisner and W.  G. Clarke,  "Comparative Evaluation
      of Three Urban Runoff Models," Water Resources Bulletin.  Vol.  11, No. 2,
      pp. 306-328, April, 1975.

195.  Huber,  Wayne C., "Modeling for Storm Water Strategies," APWA Reporter. Vol. 42,
      No. 5,  pp.  10-14,.May, 1975.


                                       - 82  -

-------
196.  Chow,  V. T.,  and B.  C.  Yen,  Urban Storrawater Runoff;   Determination of Volumes
      and Flowrates,  Environmental Protection Technology Series  EPA-600/2-76-116,
      GPO, Washington, D.C.,  240 pp.,  May,  1976.   (Available from NTIS  as PB 253 410).

197.  Brandstetter, Albin, Assessment of Mathematical Models for Storm  and Combined
      Sewer Management, Environmental Protection  Technology Series EPA-600/2-76-175a,
      GPO, Washington, D.C.,  510 pp.,  August, 1976.  (Available  from NTIS as
      PB 259 597).

198.  Hossain, A.,  J. W. Delleur and A. R.  Rao,  "Evaporation, Infiltration and
      Rainfall-Runoff Processes in Urban Watersheds," Water Resources Center
      Technical Report No. 41, Purdue University, West Lafayette, Indiana, 63  pp.»
      January, 1974.   (Available from NTIS  as PB 229 642).

199.  Systems Control, Inc.,  Palo Alto, California, "Use of Mathematical Models for
      Water Quality Planning," WRIS Technical Bulletin No.  3, Department of Ecology,
      State of Washington, 212 pp., June, 1974.

200.  Systems Control, Inc.,  Evaluation of Water Quality Models;  A Management Guide
      for Planners, Socioeconomic Environmental  Studies Series EPA-600/5-76-004, GPO,
      Washington, D.C., 176 pp., July, 1976.

201.  Pentland, R.  L., P. J.  Reynolds and Asit K. Biswas, "Water Quality Modeling:
      State-of-the-Art," pp.  481-496, Vol.  3, Proceedings of International Symposium
      on Modeling Techniques  in Water Resources  Systems, May 9-12, 1972, Ottawa,
      Canada.

202.  Canale, Raymond P., Ed., Modeling Biochemical Processes in Aquatic Ecosystems,
      Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan, 500 pp., February, 1976.

203.  Gordon, R., and M. Spaulding, "A Bibliography of Numerical Models  for Tidal
      Rivers, Estuaries and Coastal Waters," Marine Technical Report 32, University of
      Rhode Island, Narragansett, R.I. 02882, 55  pages, 1974.

204.  Hinwood, Jon B.,  and Ian G. Wallis, "Classification of Models of Tidal Waters,"
      J.Hvd.Div., ASCE  Proc., Vol. 101, No. HY10,  pp. 1315-1331, October,  1975.
      Authors' closure  to discussion:  Vol.  102,  No. HY12,  pp. 1776-1777,  December, 1976.

205.  Hinwood, Jon B.,  and Ian G. Wallis, "Review of Models of Tidal Waters,"  J.Hyd.Div..
      ASCE  Proc., Vol.  101, No. HY11,  pp. 1405-1421, November,  1975.  Author*1 closure
      to  discussion:   Vol. 103, No. HY4, pp. 453-455, April,  1977.

206.  Lombardo, P. S.,  "Critical  Review of  Currently Available  Water Quality  Models,"
      Hydrocomp, Inc.,  101 pp., July,  1973.  (Available from NTIS  as PB 222 265).

207.  Lassiter, Ray  R., Modeling Dynamics of Biological and Chemical Components of
      Aquatic Ecosystems.  Ecological  Research Series EPA 660/3-75-012,  GPO,
      Washington,  D.C., 54 pp.,  May,  1975.   (Available  from NTIS as PB 241 987).

208.  Parker, R. A.,  and  D.  Roop,  "Survey  of Aquatic Ecosystem  Models," The Institute
      of Ecology,  Washington,  D.C.,  131  pp., 1974.

209.  Kramer, W. P.,  and  R.  H.  Weisberg, "FORTRAN Graphics Programs for Physical
      Oceanographic  and Time Series Data,1.1 Marine Technical Report 46, University of
      Rhode Island Marine Advisory Service, Narragansett,  92 pp., 1976.


                                      - 83  -

-------
 210.  Hydrologic Engineering Center,  U.S.  Array Corps of Engineers,  "Water Quality
       for River-Reservoir Systems," Computer Programs 401-F2-L2100  and  -L2100A
       Davis,  California,  210 pp.,  July,  1974.

 211.  Willey, R. G.,  "Water Quality Evaluation of Aquatic  Systems," ASCE Convention
       April 14-18,  1975,  New Orleans,  Louisiana,  ASCE Meeting  Preprint  2420,  20 pp.*

 212.  McBean, Edward  A.,  and Daniel P. Loucks,  "Planning and Analyzing  of Metropolitan
       Water Resource  Systems,"  Water Resources  and  Marine  Sciences  Center Technical
       Report No. 84,  Cornell University, Ithaca,  New York,  805 pp., June, 1974.
       (Available from NTIS as PB 235  257).

 213.  Evenson,  Donald E.,  and Donald  J.  Smith,  "Mathematical Modeling of San  Francisco
       Bay," a paper presented at ASCE Annual  and  National  Environmental Engineering
       Convention, October 21-25, 1974, Kansas City,  Missouri,  ASCE  Meeting Preprint
       2380,  30  pp.

 214.  Chen, Carl W.,  "Concepts  and  Utilities  of Ecologic Model," J.San.EnB,rp;.Div.
       ASCE Proc., Vol.  95, No.  SA5, pp.  1085-1097,  October, 1970."'

 215.  Orlob,  Gerald T., "Mathematical  Modeling  of Estuarial Systems," pp. 78-128
       Vol.  1, Proceedings  of International Symposium on Modeling Techniques in Wat-or
       Resources  Systems,  May 9-12,  1972, Ottawa,  Ontario, Canada.             "   "	

 216.   Soice,  V.  Phillip,  "Application  of a Water  Quality Model  to the Denver
       Metropolitan  Area,"  pp. 1165-1182, Volume II,  Symposium  on Modeling
      ASCE, New  York, N.Y.,  1975.                                  "~~

 217.   Benson, Donald  J.,  "A  Coordinated Approach  to  Water and  Waste Management,"
       Municipality  of Metropolitan Seattle, a paper  presented  October 12, 1973  at
       the 40th Annual Convention, Pacific Northwest  Pollution Control Association
      Vancouver, B.C.

 218.  Russell, D. L., J. W.  Poirot and H. A.  Cornell, "A Study of Water Resources
      Under King County's Developing Environmental Management Information System "
      a paper presented on June 20, 1974, at  the AWWA Annual Conference, Boston *Mas
      33 pp.

 219.  Lombardo, Pio S., and  Ronald F. Ott,  "Water Quality Simulation and Application "
      Water Resources Bulletin.  Vol. 10,  No.  1, pp.  1-9, February,  1974.            *

220.  Water Resources Engineers  and Kramer, Chin and Mayo,  "Urban Runoff and Basin
      Drainage Study, Green  and Cedar River Basins of Washington,"  Appendix B of final
      report "Program Documentation and User's Guide for Urban Storm Drainage
      Simulation Models," for Municipality of Metropolitan Seattle  and U.S.  Army Cor
      of Engineers,  Seattle District, approx. 320 pp., July, 1974.                   Pa

221.  Donigian, A. S., Jr., and  N.  H. Crawford, "Modeling Nonpoint  Pollution from th
      Land Surface," a report for the Environmental Research Laboratory, Athens
      Georgia, Hydrocomp,  Inc.,  Palo Alto,  California, 279  pp., February,  1976.

222.  Hydrologic Engineering Center, U.S. Army Corps of Engineers,  "HEC-2,  Water
      Surface Profiles," Computer Program 723-X6-L202A, Davis,  California,  February
      1972.                                                                      ^'
                                       - 84 -

-------
223.  Soil Conservation Service,  Engineering Division,  U.S.  Department of 'Agriculture,
      "Computer Program for Project Formulation — Hydrology," Technical Release No.  20,
      May, 1965.

224.  Eichert, B. S., "Survey of Programs for Water Surface Profiles," J.Hyd.Div..
      ASCE Proc., Vol..96, No. HY2, pp. 547-563, February,  1970.   Author's  closure  to
      discussion:  Vol. 97, No. HY8, pp. 1237-1238, August,  1971.

225.  Walesh, Stuart G., "Simulation in Watershed Planning," J.Hyd.Div.. ASCE Proc.,
      Vol. 99, No. HY9, pp. 1383-1399,  September, 1973.

226.  Montgomery-Water Resources Engineers, "Water, Wastewater and Flood Control   ,'.
      Facilities Planning Model,  Summary Report," LaJolla,  California, 38 pp., May,  1974.
                                                                           . •
Section 7

227.  Stankowski, Stephen J., Population Density as an Indirect Indicator of Urban and
      Suburban Land-Surface Modifications. U.S. Geological Survey Professional f&pet
      800-B, GPO, Washington, D.C., pp. 219-224, 1972.

228.  Graham, P. H., L. S. Costello and H. J. Mallon, "Estimation of Imperviousness
      and Specific Curb Length for Forecasting Stormwater Quality and Quantity,"
      J.WPCF. Vol. 46, No. 4, pp. 717-725, April, 1974.

229.  Gluck, W. R,, and R. H. McCuen, "Estimating Land Use Characteristics for
      Hydrologic Models," Water Resources Research. Vol. 11, No. 1, pp.  177-179,
      February, 1975.

230.  Jackson, Thomas  J., and Robert M. Ragan, "Value of Landsat in Urban Water
      Resources Planning," J.Wat.Res.Plan.&Mgt.Div.. ASCE Proc., Vol. ,103, No. WR1,
      pp. 33-46, May,  1977.

231.  Jackson, Thomas  J., Robert M. Ragan and William N. Fitch,  "Test of Landsat-
      Based Urban Hydrologic Modeling," J.Wat.Res.Plan.&Mgt.Div.. ASC*  Proc., Vol.  103,
      No. WR1, pp. 141-158, May, 1977.

232.  McPherson, M. B., and W. J.  Schneider, "Problems in Modeling Urban Watersheds,"
      Water Resources  Research. Vol. 10, No. 3,  pp. 434-440, June, 1974.

233.  Meier,  Peter M.,  "Land-Use Model  for  Service Area Projection,"  J.Kavit«BnRrg.Div..
      ASCE Proc., Vol. 102, No. KB1, pp. 71-85,  February, 1976.

234   Seader, David, "Runoff  Projections for Fairfax County, Virginia," Urban Runoff.
      quantity and Quality. ASCE,  New  York,  H.Y.,  pp.  139-150, 1975.

235.  Seader, David, "Evaluation and Planning Techniques,"  Environmental Design for
      Public  Projects. Water  Resources Publications, Fort.Coll ins, Colorado,  pp. 521-
      575,  1975.

236*  Walesh, Stuart G.,  Lois A. Kawatski  and  Paul J.  Clavette,  "Land Data
      Management System for  Resource Planning," J.Wat.Res.Plan.& MRt.Div., ASCE
      Proc.,  Vol.  103, No. WR2,  pp.  177-192, November,  1977.
                                       - 85 -

-------
 237.   William Spangle and Associates,  et al.,  Earth-Science Information in Land-lisa
       Planning,  Guidelines for Earth Scientists and Planners.  U.S.  Geological
       Circular 721,  GPO,  Washington, D.C.,  28  pp.,  1976.

 238.   James,  L.  Douglas,  "Role of Economics in Planning Flood  Plain Land  Use,"
       J.Hyd.Div.,  ASCE Proc.,  Vol. 98,  No.  HY6, pp. 981-992, June,  1972.   Author's
       closure to discussion:   Vol. 97,  No.  HY1, pp. 172-173, January,  1974.

 239.   Grigg,  Neil  S.,  and Otto J.  Helwig, "State-of-the-Art of Estimating Flood
       Damage  in  Urban Areas,"  Water Resources  Bulletin. Vol. 11, No.  2, pp.  379-39Q
       April,  1975.

 240.   Espey,  William H.,  and David E. Winslow,  "Urban Flood Frequency  Characteristics  '•
       J.Hyd.Div..  ASCE Proc.,  Vol. 100,  No.  HY2,  pp. 279-293,  February, 1974.         '

 241.   U.S.  Water Resources Council,  Guidelines  for  Determining Flood Flow Frequency
       Bulletin No. 17  of  the Hydrology  Committee, GPO Stock No. 052-045-00031-2,
       Washington,  D.C., 197 pp., March,  1976.

 242.   Jones,  D.  Earl,  Jr.,  "Basis  for Flood  Plain Occupancy Decisions," Urban Runoff
       Quantity and Quality,.ASCE.  New York,  N.Y., pp. 235-246,  1975.           ~L

 243.   McPherson, M.  B., "Feasibility of  the  Metropolitan Water Intelligence  System
       Concept (Integrated Automatic  Operational Control)," ASCE Urban  Water  Resources
       Research Program Technical Memorandum  No. 15, ASCE,.New  York, N.Y.,  110 pp.
       December,  1971.   (Available  from NTIS  as  PB 207 301).                       '

 244.   Gibbs,  C.  V.,  S. M. Alexander  and  C. P. Leiser, "System  for Regulation of
       Combined Sewage  Flows,"  J.San.Engrg.Div.. ASCE Proc., Vol. 98, No.  SA6, pp. 951.
       972,  December, 1972.

 245.   Municipality of  Metropolitan Seattle,  Computer Management of  a Combined Sewer
       System. Environmental Protection Technology Series EPA-670/2-74-022, GPO,
       Washington,  D.C., 487 pp., July, 1974.  (Available from  NTIS  as  PB  235 717).

 246.   Giessner, W. R., R. T. Cockburn, F. H. Moss and M. E. Noonan, "Planning and
       Control of Combined Sewerage Systems," J.Envir.Engrg.Div.. ASCE  Proc., Vol. loo-
      No. EE4, pp. 1013-1032, August, 1974..                                         *

247.   Wenzel, H. G.,  Jr., J. tf. Labadie and N.   S. Grigg, "Detention Storage Control
       Strategy Development," J.Wat.Res.Plan.&Mgt.Div..  ASCE Proc., Vol. 102, No. WRJ.
       pp. 117-135, April, 1976.                                                     '

248.  Labadie, J.  W., N. S. Grigg  and B. H.  Bradford, "Automatic Control of Large-Scale
      Combined Sewer Systems," J.Envjr.Engrg.Div.. ASCE Proc.,  Vol. 101, No. EE1,
       pp. 27-39, February, 1975.

249.  Corotis, Ross B., "Stochastic Considerations  in Thunderstorm Modeling," J.Hyd
      ASCE Proc.,  Vol. 102, No. HY7, pp. 865-879, July,  1976.

250.  Changnon, Stanley A., Jr., and Floyd A. Huff, "Chicago Hydrometeorological Area
       Project:  A  Comprehensive New Study of Urban Hydrome.eorology," First Interim
      Report, Program of Advanced  Environmental Research and Technology, NSF/RANN,
      Atmospheric  Sciences Center  Illinois State Water Survey,  Urbana, Illinois
       61801, 69 pp.,  September, 1976.


                                      - 86 -

-------
251.  Tucker, L. S.,  "Raingage Networks  of the Largest  Cities," ASCE Urban  Water
      Resources Research Program Technical Memorandum No.  9,  ASCE,  New York, N.Y.,
      90 pp., March 17, 1969.  (Available from NTIS  as  PB  184 704).

252.  Friedland, Alan 0., "San Francisco Precipitation  and Flow Measuring Network,"
      Urban Runoff. Quantity and Quality. ASCE, New  York,  N.Y., pp. 205-212, 1975.

253.  Brownson, L. D., City of Portland, Oregon, "Development of  HYDRA (Hydrologic
      Data Retrieval and Alarm System)," paper presented at Engineering Foundation
      Conference on Instrumentation and  Analysis of  Urban  Stormwater Data,  Quantity
      and Quality, November 28-Deceraber  3, 1976, Easton, Maryland,  23 pp.

254.  Earl, C. T., et al., Melbourne and Metropolitan Board of Works and Dandenong
      Valley Authority, "Urban Flood Warning and Watershed Management Advances in
      Metropolitan Melbourne," ASCE Urban Water Resources  Research Program Technical
      Memorandum No. 30, ASCE, New York, N.Y., 72 pp.,  June,  1976.   (Available from
      NTIS as PB 259 549).

255.  Envirex, Inc., Methodology for the Study of Urban Storm Generated Pollution
      and Control. Environmental Protection Technology  Series EPA-600/2-76-145, U.S.
      EPA, Cincinnati, Ohio, 326 pp., August, 1976.   (Available  from NTIS as PB 258 743).

256.  Henry, J. G., and P. A. Ahern, The Effect of Storage on Storm and Combined Sewers,
      Research Program for the Abatement of Municipal Pollution under the Provisions
      of the Canada-Ontario Agreement on Great Lakes Water Quality, Research Report
      No. 34, Environment Canada, Ottawa, 106 pp., February,  1976.

257.  Lott, David L., "Optimization Model for the Design of Urban Flood-Control
      Systems," Center for Research on Water Resources Technical Report CRWR-11, The
      University of Texas at Austin, 217 pp., November, 1976.  (Available from NTIS
      as PB  263 490).

258.  Hardt, Richard A., and Stephen J. Burges, "Some Consequences of Areawide Runoff
      Control Strategies in Urban Watersheds," Department of  Civil Engineering Technical
      Report No. 48, University of Washington, Seattle, 81 pp.,  June, 1976.  (Available
      from NTIS as PB  261 258).

259.  Terstriep, M. L.,  M. L. Voorhees  and G.  M. Bender,  "Conventional Urbanization
      and Its Effect on  Storm Runoff,"  Illinois State Water  Survey, Urbana, Illinois
      61801, 68 pp., August,  1976.

260.  Field, Richard,  Russell Bowden and  Kathy Rozgonyi,  "Literature  Review — Urban
      Runoff and Combined Sewer Overflow," J.WPCF.  Vol. 49,  No. 6,  pp.  1095-1104,
      June,  1977.

261.  Shubinski,  Robert  P.,  and William N. Fitch, "Appraisal of Areawide Wastewater
      Planning,"  J.Wat.Res.Plan.&Mgt.Djv.. ASCE Proc.,  Vol.  103, No.  WR1,  pp.  63-72,
      May,  1977.

262.  Westman,  Walter E.3  "Problems  in  Implementing U.S.  Water Quality Goals,"
      American Scientist.  Vol.  65, No.  2,  pp.  197-203,  March-April,  1977.

263.  Jameson,  David L., and Harold  V.  Kibby,  Ecosystem Impacts  of Urbanization
      Assessment  Methodology.  Ecological Research Series  EPA-600/3-76-072,
       Environmental  Research Laboratory, U.S.  EPA,  Corvallis, Oregon 97330,  227 pp.,
      July,  1976.

                                        - 87  -

-------
                                       ADDENDUM 1

                                METROPOLITAN INVENTORIES

                                   by M.  B.  McPherson

 Hydrologic  Complexity

          Figure 1  is a simplified schematic  summary of  the interconnections of
 water occurrences,  uses and  processes  in the  urban hydrologic system.  Not shown are
 water quality  aspects,  which would be  still more complicated.  Considerable detail
 is  hidden from Figure 1.   For  example, most of our major cities have dozens and even.
 hundreds ptf stormwater drainage catchments  with cumulative conduit lengths often
 exceeding 1,000 miles,  all of  which are  subsumed under "storm drainage" in Figure 1

            for Water  Quantities
          Figure  2'  ^  is a graphical representation of water quantities in a
 hypothetical urban area of one-million  inhabitants.  National averages have been used
 to  calculate the  magnitude of the various components.  These magnitudes are thus
 individually typical,  but their combination in Figure 2 is not.  Nonetheless, two
 important factp are  illustrated:  the bulk of the supplied water originates outside
 of  urban areas; and well over half of all the water handled one way or another is in
 private hands.  This means that there is typically only a partial overlap in local
 government water  supply source planning and metropolitan general planning, and that
 the usual approach of  focusing on public water supply can overlook a much larger user
 group.  The situation  for water pollution control is similar, as over half of the
 volumes of discharges  to receiving waters are from private corporation lands and the
 region impacted by water pollution of tea extends well beyond the metropolis of origin

          Cooling water withdrawals of thermoelectric power plants are essentially all
 self-supplied.  While  such withdrawals in urban areas have not been segregated from
 national totals,'^' they are quite possibly on the order of three times the non-
 thermoelectric industrial withdrawals of Figure 2.
          Figure 3' depicts a breakdown of that portion of Figure 2 termed "public
water supply withdrawals".  Figure 3 is drawn to scale and is again based upon a
composite of national annual averages that resulted from a reconciliation of estimates
for individual components made by various experts.  Thus, the distribution of component
amounts for any given community will vary from those shown.  For example, there are
cases where industrial use is minuscule and cases where it predominates.   Because of
difficulties in quantity accounting generally,  there is considerable uncertainty about
the extent of "unaccounted-for" water and "infiltration and inflow".  This is the
result of the necessity to arrive at amounts for these two elements by determining th
residuals or leftovers remaining after the magnitudes of all other elements have been
taken into account.  Incomplete system quantity measurements and errors in registrati
of measuring devices also contribute to the accounting uncertainties.  Finally,      °n
difficulties in calculation arise from the fact that water and wastewater jurisdictio
often differ in size.                                                                 S

          Figure 4 depicts a breakdown of that portion of Figure 2 termed "urban
runoff".  Whereas Figure 3 is a composite of water and wastewater national annual
                                        - 88 -

-------
00
lO
                                                                    ATMOSPHERE
                                                                                                                                         IMPORTED
                                                                                                                                        LOW-FLOW
                                                                                                                                        AUGMENTATION.
     (PRECIPITATION)
                                                                                                                           NOUSTRIAL
                                                                                                                            COOLING
                                                                                                                             WATER
                                                                                                                                       (SURFACE WATER
                                                                                                                                        OUTFLOW)   ~
URBAN  LAND  SURFACE
    (MANIPULATED)
                                                SURFACE
                                                PULATED)
                                                             ZONE OF
                                                            AERATION
                                                                           (CAPLLARY
                                                                              RISE)
                                   (LAWN 8 CHOP
                                   IRRIGATION)
                                               (SEEPAGE  AND
                                               SPRING FLOW)
                                                      WATER
                                                     SYSTEMS
                                                                          STORM
                                                                         DRAINAGE
                                                                                           (DIRECT
                                                                                           EFFLUENT
                                                                                           DISCHARGES)
                                                                                                                     WAS1EWA1ER
                                                                                                                      SEWERAGE
                                                                                                                         (SEWER
                                                                                                                         LEAKAGE OR
                                                                                                                         ACCRETION)
                                                                                           (EXPORTED
                                                                                          fc INDUSTRIAL
                                                                                           CONSUMPTIVE
                                                                                           USES)
                                                                                          (SUBSURFACE
                                                                                          GROUNDWATER -
                                                                                              OUTFLOW)
(GROUNDWATER
 PUMPAGE)
                                                              ZONE  OF  SATURATION
                                                                  DEEP INTRUSIONS
                                                     (»: Interconnected in Combined Systems)
                                                     Notes: Water Quality Aspects Not Shown.
                                                           Some Connections/Functions Not Typical.
                                                           Salt Water Intrusion  Not Shown.
               FIGURE  1-URBAN  HYOROLOGIC  SYSTEM
                              (Adapted  From:  "Summary of the  Hydrologicol Situation  in  Long Island, N.Y. as a Guide to Water Management Alternatives",
                               by O.L.  Franke  and N.E. McClymonds, U.S. Geological  Survey  Professional taper 627-F, 1972).

-------
         MEAN
        ANNUAL
      PRECIPITATION
       (FROM
     ATMOSPHERE)
                     GROUNDWATER
 .SURFACE
  sWATER
fiu X(FROM
                                   PUBLIC
                                   WATER
                                   SUPPLY
                                   WITHDRAWALS
                                   (1/3 TO
                                   INDUSTRY)
                                                EVAPOTRANSPI RATION
                                                (TO  ATMOSPHERE)
                                     URBAN RUNOFF

                                     (TO RECEIVING
                                     WATER  BODIES)
                                    (TO GROUNDWATER
                                    AND EVAPOTRANSPIRATION)
                                         PUBLIC
                                      WASTE WATER
                                      (TO RECEIVING
                                      WATER BODIES)
  SCALE;
            500,000
          f TONS/DAY
            SURFACE
            WATER
            (FROM
            OUTSIDE)
              SELF-
              SUPPLIED,
              NON-THERMO-
              ELECTRIC,
              INDUSTRIAL
              WATER
              WITHDRAWALS

               (ACTUAL
                 USE
                  IS
                MUCH
               GREATER)
                                        INDUSTRIAL
                                       WASTE WATER
                                      (TO RECEIVING
                                      WATER BODIES)
                                                 (TO EVAPORATION,
                                                 EXPORTS AND
                                                 GROUNDWATER)
FIGURE 2  -
ANNUAL  AVERAGE  WATER QUANTITY BALANCE
FOR  AN HYPOTHETICAL URBAN  AREA  OF  ONE-
MILLION  INHABITANTS,  IN  TONS PER DAY
                                                   (D
                            - 90 -

-------
 PUBLIC
 WATER
 SUPPLIES
 DELIVERED
 TO
 DISTRIBUTION
 SYSTEMS
                               SPRINKLING

                               LAUNDRY a DISHES
                 DOMESTIC
                    USE
               BATHING AND
               PERSONAL USES
                               TOILET FLUSHING
COMMERCIAL  USE
 PUBLIC USE
                INDUSTRIAL
                    USE
               (ACTUAL USE IS
                GREATER BECAUSE
                OF IN-PLANT
                RECYCLING)
PUBLIC
WASTEWATER
COLLECTION
SYSTEMS
PUBLIC
WASTEWATER
TREATMENT
FACILITIES
             'UNACCOUNTED-
             FOR WATER
                CONSUMPTION

               LEAKAGE
               AND WASTE
                             INFILTRATION
                             AND INFLOW
FIGURE 3- COMPOSITE  PUBLIC WATER/WASTE WATER INPUT-OUTPUT BALANCE
                                                      (3)

-------
                          FLOW OVER UNSEWERED LAND
I


»0
   MEAN
  ANNUAL
PRECIPITATION!
  (FROM
ATMOSPHERE)
FLOW
OVER
LAND TO
STREET
INLETS ON
SEWERED
CATCH-
 MENTS
UNDER-
GROUND
OOLLECTTON
SYSTEMS
(SEWERS
AND
UNDER-
 GROUND
STORAGE)
                                                       STORM
                                                       SEWERS
                                            MAN-MADE
                                            OPEN
                                            DRAINAGE
                                            CHANNELS
                                            AND
                                            IMPROVED
                                            NATURAL
                                            CHANNELS
AND
DIRECT
DISCHARGE
                                                 COMBINED
                                                 SEWERS
                                                                               AND

                                                                               SURFACE
                                                                               STORAGE
                                                                               {PONDS,
                                                                               LAKES,
                                                                               DELIBERATE
                                                                               USE OF
                                                                               NORMALLY
                                                                               DRY AREAS
                                                                 RECEIVING
                                                                 BODIES
 WATER
'(STREAMS,
 RIVERS,
 LAKES,
 OCEAN)
   FIGURE  4-COMPOSITE URBAN SURFACE  WATER RUNOFF INPUT-OUTPUT  BALANCE

-------
averages drawn to scale, and there are reservations about how precise some of these
estimates are, we confront much greater uncertainties with urban runoff.  The principal
reason is that there is a substantial body of data on metropolitan water and
wastewater volumes and practically none on storm and combined sewers.  The indication
in Figure 4 that national annual average flows from storm sewers are about twice as
large as from combined sewers is based on an indirect indication^' of the population
served by .combined sewers versus wastewater sewers.  The indication in Figure 4 that
about one-fourth of urban runoff flows over unsewered land is purely conjectural but
within reason.  After leaving sewer systems, runoff may be stored in surface
reservoirs and may pass through artificial and improved channels on its way to
receiving waters, and some of it will enter receiving waters directly.  There is no
basis for estimating the division of national annual averages among these three modes.
Further, there are complex mixtures of open drainage channels and surface storage in
many metropolitan areas and it might be next to impossible to assign a typical
simplified chain of occurrence (vis-a-vis Figure 3) to the flow through these elements,
even for a particular metropolitan area.  The writer's guess is that nationally less
than one-fourth of urban runoff passes through open artificial or improved channels
and that one-tenth or less is subjected to surface storage on its way to receiving
waters.  How much that  is subjected to surface storage also passes before or afterwards
through open artificial or improved channels is anyone's guess.

          Obviously, an input-output balance of urban surface water runoff pollutant
burdens would be even more complicated, and much less is known about urban runoff
quality than on urban runoff quantity.

          No attempt has been made to show the interrelations between groundwater and
surface water in Figure 4.  While these are taken  into account  in Figure  1,  exactly
how to depict them quantitatively in Figure 4 presents quite an enigma because of the
complexities  involved and the lack of data.

          Despite the absence of national indications, as described  above, responsible
management would appear to call for definitive breakdowns of urban  runoff components,
in terms of quantity and quality, in each metropolis.  How can we  solve a presumed
problem when we have not assessed its magnitude?   Such breakdowns would form a part
of a metropolitan water balance inventory,  the subject of the remainder of this
Addendum.

Metropolitan Water Balances

          Comprehensive planning  faces  a number of obstacles,  including fractionalized
authority of  administrative  agencies and balkanized local  governments  in  metropolitan
areas  ^' specialist approaches taken  individually on each water  service  component,  and
uneven advances  in  knowledge on various water  components.  More comprehensive  management
approaches would  integrate or  give due  consideration to  all  aspects of  water in a
metropolis,  together with  related environmental,  energy  and  other relevant  considerations,
The need  for  a  comprehensive overview  becomes  more crucial as  the complexity of urban
areas  increases,  caused by  such things  as  population migration  and growth,  competition
over  available  energy  supplies, rising expectations of urban dwellers in  the face  of
 inflation of local  government  costs,  and  greater  public  awareness of environmental
 issues and  enlarging  demands for  public participation.

          Water balance inventories,  describing  the quantity and quality aspects of
 the  fate of water from its  appearance  as  precipitation through its departure from a
 metropolis  as runoff and evapotranspiration,  have been advocated as a referencing tool
 for  planning in metropolitan areas.'°'  Such inventories provide a basis  for better


                                         - 93 -

-------
 recognition of the interrelation,  interdependence,  and  interconnection of  the  element
 of the water resources of. a metropolis.   "Attempts  to study  any inventory  system ana t-
 from the total system does not enable the investigator  to  see  the  total  impact of
 alternatives."(7'

           In Sweden,  urban water inventories  have been  approached  from two direction
 the "outer system," embodying natural processes  governed by  seasonal and random
 variations; and the "inner system,"  involved  with the conveyance and distribution of
 water for uses within urban areas.   The  collective  average annual water budget for
 urban areas in Sweden is  shown in Figure 5'°^ for the outer  system and in Figure
 for the inner system.  Numerical values  indicated are volumes  in millions of cubic
 meters per year.   The total urban water  budget can  be described as the sum of  the
 outer and inner systems,  Figure 5 plus Figure 6, where  the sole connections would be
 "via combined sewers" (340 x 10 - m /year) and  "receiving waters".

           Inventory methods used in  San  Antonio  for the quantification of soil water
 surface water and groundwater have been  described.'^' Long  Island, New York,  an     '
 instance of a nearly  "closed" hydrologic system  covering about  1,400 square miles
 has been the subject  of extensive water-budget studies.CIO)

           Figure  7^   '  is  a water balance, on an annual average basis, for metropolit
 Chicago (3,714-square miles).  All but the smallest of the seven drainage basins
 involved receive  flows  of water  from outside  the metropolitan area.  Variations withi
 individual  years  were also  inventoried.   "Water  in  the metropolitan area is constantl
 on  the move.   The three principal components  of  the water  resource (atmospheric
 moisture,  surface water and groundwater)  are  interconnected, and water moves in a
 continuous  cycle  between them.  Any  comprehensive study of water resources, therefor
 must  first  recognize  the existence of this cycle and  then  define the total water      '
 system,  even while realizing that only a  limited part of the water moving through th
 system is actually available for use.  The situation is further complexed in that th
 metropolitan area is  part of several  larger systems  which  embrace much of the  United
 States."v11)

 Getting  the Act Together

          In  an effort  to facilitate  the preparation of water balance inventories  a
 attempt  has  been  made in Figure 1 to describe, in flow-chart form,  the urban hydrolo
 system,'  ' of which  stormwater is a part or a subsystem.  Figure 8^^) describes  the °
 urban  stormwater  runoff subsystem,  interpreted from a drainage standpoint,  and deals
with water quantities.  In  contrast, Figure 9'*^' is a schematic diagram of the
 relationship between urban  activities and water quality, where urban runoff is a ma fo
element.  Not many urban runoff control master plans have been developed, fewer still
have been implemented,  and  rare are  the instances where master plans have integrated
quality control with quantity control.  All this  despite the increasing importance of
 their  conjunctive  planning.

          Strategies  for improving water quality  have been delineated from the
perspective of the urban planner concerned with stormwater control.^5'

References

 1. McPherson, M. B., Regional Earth Science Information in Local Water Management-
    ASCE, New York, N.Y., 155 pp.,  July, 1975.                  '               "—'

                                                                       (Continued)


                                         - 94  -

-------
                     PRECIPITATION
                         2820
IMPERMEABLE
SURFACES
/

270 ,rnr
VIA       	
COMBINED (340
SEWERS  v	
                                        1970
                                                EVAPOTRANSPIRATION
                                                   <1270
                                              PERMEABLE
                                               SURFACES
   GROUNDWATER,
   HOUSE DRAINS, ETC.
GROUNDWATER
VIA STORM SEWERS 780
                                      \
FIGURE  5 -  GENERAL  URBAN AREA WATER BUDGET FOR  SWEDEN,
                            OUTER SYSTEM .(8)
         WATER
         SUPPLY
         SOURCES
                      USE IN WATERWORKS 38
                                                         VIA
                                                      340 COMBINED
                                                         SEWERS
                                                      COMBINED
                                                      SEWER
                                                      OVERFLOWS

                                                      35
                                           RECEIVING WATERS
 FIGURE 6- GENERAL  URBAN AREA WATER BUDGET FOR  SWEDEN,
                          INNER SYSTEM.(8)
                            - 95  -

-------
             4410 (24.59 INCHES/YR)
            .EVAPOTRANSPIRATION
                    5867 (33.18 INCHES/YR)
                       PRECIPITATION
                   SURFACE AND SOIL WATER 4088
         :;::x::;PUBLlC WATER SUPPLY 1077 (1961)
                   DIVERSION 646  (1961 ):•:-:•:•:•:•:•:•
            •RUNOFF FROM PRECIPITATION 663
             s^-tMttttMtr&ttW*^^
             p^^^^^^fW**^^wrw*^^w^r^^^Ki
             BASE FLOW 673
         GROUND-WATER PUMRAGE 182(1961)

                                   f
              ANKAKEi: RIVER 2375
                                                         ..GROUND-WATER
                                                          MNFLOW 21
WITHDRAWAL FROM
SUBSURFACE STORAGE40
             WATER IN SUBSURFACE  STORAGE
FIGURE 7-AVERAGE ANNUAL WATER BALANCE, METROPOLITAN CHICAGO
           (MILLION  GALLONS PER DAY,  1961)
                                              (ii)
                                 - 96 -

-------
                PRECIPITATION
                +W44
STREAMFUDW
 VHS)
 zz
 p
 il
          INITIAL
          DRAINAGE
                         INTERMEDIATE
                           STORAGE
                                               OUTFLOW
                                                                I—*
                                            DEPRESSIONS
                                            STREETS
                                            SUMPS
                                               GULCHES
                                               STORM
                                                SEWERS
MAJOR
DRAINAGE
THUNDER
 STORMS

TORRENTIAL
 RAINS

HEAVY
 RAINFALL
 AND
 SNOWMELT
GULCHES
CREEKS
STREAMS
FLOOD
 PLAINS
                                            LAKES
                                            RESERVOIRS
                                            FLOOD
                                             PLAINS
                                               CREEKS
                                               RIVERS
                                                        EVAPORATION
                                                        AND LOSSES
                                                                         •4  NATURAL
                                                                            STORAGE
                                                                                         STREAM FLOW
 AQUIFER
RECHARGE
 STORAGE|
                      POTABLE
                       WATER
                       TREATMENT
                            DISTRIBUTION
                                                                       STORM SEWER
                                                              WASTE
                                                                WATER
                                                                TREATMENT
                                                                             SUBSURFACE FLOW
                                                                                OTHER URBAN
                                                                                 SUBSYSTEMS
                                                          BUILDING AND
                                                           DEVELOPMENT

                                                          PI'^LIC HEALTH
                                                           AND  SAFETY

                                                          TRANSPORTATION

                                                          SOLID  WASTE
                                                          RECREATION/
                                                           OPEN SPACE

                                                          SANITARY WASTE
                                                          IRRIGATION
                                                          WATER SUPPLY
                        FIGURE 8-THE URBAN STORM DRAINAGE  SUBSYSTEM
                                                                         (13)

-------
                                  LAND USE DETERMINANTS
      Population and
         Economic
         Growth
      Private
      Market
       Forces
 Land Use
Regulation
   Public
  Services
and Facilities
  Natural
  Features
ft Constraints
                                          Land Use
                                          Pattern
                                      Form
                                      Density
                                      Use Mix
                                      Open Space
                                      Land Conversion Rate
                              WATER QUALITY  IMPACTS OF LAND USE
       Withdrawal
        of Surface
        and Ground
         Water
            Loss of
            Natural
         Ground Cover
        And Disturbance
       of Environmentally
        Sensitive Areas
   Net Change
    of Stream
      Flow
Construction
  Related
  Sediment
   Loads
          Silvaculture,
           Mining ana
           Agriculture
             Runoff
             Loads
  Urban
  Runoff
  Loads
                          Stream
                       Enlargement
                        and Erosion
                  Domestic
                    Waste
                    Loads
      Industrial
        Waste
        Loads
                        Total
                      Non-Point
                        Load
                   Water
                   Quality
                 Standardsd
                  Effluent
                 Limitations
                   Total
                   Point
                    Load
                                                         Water Quality
                                                          Management
                                                           Practices
                    Quantity
                   and Quality
                   of Receiving
                   Water Prior
                   To discharge
                      Quantity
                     and Quality
                     of Discharge
                                    Land
                                   Disposal
                                 Wastewoter
                                  Reuse,etc.
                              Resulting
                              Quality of
                           Receiving Waters
FIGURE 9-  SCHEMATIC  DIAGRAM  OF THE LAND-USE/WATER-QUALITY
              RELATIONSHIP04*
                                        - 98  -

-------
 2.  Tucker, L. S., J. Millan and W. W. Burt, "Metropolitan Industrial Water Use,"
    ASCE Urban Water Resources Research Program Technical Memorandum No. 16, ASCE,
    New York, N.Y., 45 pp., May, 1972.  (Available from NTIS as PB 212 578).

 3.  McPherson, M. B., "Household Water Use," ASCE Urban Water Resources Research
    Program Technical Memorandum No. 28, ASCE, New York, N.Y.S 25 pp. + 84 p. append.,
    January,  1976,  (Available from NTIS as PB 250 879).

 4.  Sullivan, Richard H., "Inventory of Combined Sewer Facilities," Civil Engineering,
    Vol. 38,  No.  11, pp. 52-53, November, 1968.

 5.  McPherson, M. B., "Urban Water Resources," EOS, Transactions of the AGU, Vol. 57,
    No. 11, pp. 798-806, November, 1976.

 6.  McPherson, M. B., "Need for Metropolitan Water Balance Inventories," J.Hyd.Div.,
    ASCE Proc., Vol. 99, No. HY10, pp.  1837-1848, October, 1973.  Author's  closure
    to discussion:  Vol. 101, No. HY4,  p. 409, April, 1975.

 7.  Koch,  C.  T.,  "Systems Description  for Urban Water Resources," pp.  19-41,
    Proceedings,  16th Annual Conference on  Water  for Texas,  "Urban Water Resource
    Planning  and  Management," Texas A.  and  M. University, College Station,  Texas,
    200 pp.,  September,  1971.   (Available  from NTIS as  PB 210  325).

 8.  Lindh, Gunnar,  "Urban Hydrological  Modeling and Catchment  Research in  Sweden,"
    ASCE Urban  Water Resources  Research Program Technical Memorandum No. IHP-7,
    ASCE,  New York, N.Y., 33  pp., October,  1976.  (Available from NTIS as PB 267 523).

 9.  Koch,  C.  T.,  "Methodology  for Managing Resources,"  J.AWWA,  Vol.  64,  No. 2,  pp.  122-
    126, February,  1972.

10.  Cohen, Philip,  0.  L. Franke and  B.  L.  Foxworthy, An Atlas  of Long Island's  Water
    Resources,  New York State Water  Resources Commission, Bulletin  62, Albany,  N.Y.,
     117  pp.,  1968.

11.   Sheaffer, John R.,  and  Arthur T.  Zeizel,  "The Water Resource in Northeastern
     Illinois:  Planning Its Use," Technical Report No.  4, Northeastern Illinois
     Metropolitan Area Planning Commission, Chicago,  Illinois,  182 pp., June, 1966.

12   Task Committee Report,  "Aspects  of Hydrological Effects of Urbanization,"
     J.Hvd.Div..  ASCE Proc., Vol. 101,  No. HY5,  pp. 449-468,  May, 1975.

13.   Rice,  Leonard,  "Sediment Considerations in Urban Drainage," ASCE National Water
     Resources Engineering Meeting,  January 24-28, 1972, Atlanta, Georgia,  ASCE
     Meeting Preprint 1588,  14 pp.

14.  Shubinskia  Robert P., and Steven N. Nelson, "Effects of Urbanization on Water
     Quality," ASCE Urban Water Resources Research Program Technical Memorandum No.  26,
     ASCE, New York, N.Y., 34 pp., March, 1975.  (Available  from NTIS  as PB  242 297).

15.  Shubinski, Robert P., "Concepts of Urban Runoff Control," Section 5 in "Management
     of Urban Storm Runoff," ASCE Urban Water Resources Research Program Technical
     Memorandum No. 24, ASCE, New York, N.Y., 92 pp., May, 1974.  (Available from
     NTIS  as  PB 234 316).
                                         - 99 -

-------
                                       ADDENDUM 2

                                 THE DESIGN STORM CONCEPT

                                    by M.  B. .McPherson

 Introduction

           Historically,  urban  areas have  been drained by underground  systems of
 sewers  that  were intentionally designed to remove stormwater as  rapidly  as possible
 from occupied areas.   Discharges  from conventional storm drainage  sewer  facilities
 and flood-plain intrusion by structures both  tend to  aggravate flooding, and
 thereby jointly tend  to  raise  the potential for stream flooding  damages.  The
 advantages of local detention  storage in  lieu of  the  traditional rapid removal of
 storm flows  has long  been recognized^' and there is  evidence that the usage of such
 storage is on the rise.'  '  However,  local deteTition  storage has been only
 occasionally employed as  part  of  overall  flood mitigation, as in Denver^' and
 Fairfax County, Virginia.™)
          Detention  storage  is  recognized as one of  the principal means for abatement
 of  pollution  from  stormwater discharges and combined sewage overflows.^)  There are
 opportunities  in new land development  to incorporate detention storage at the ground
 surface.(°'  However,  for existing drainage systems, there may be few opportunities
 to  add detention storage except underground, for both combined sewer and separate
 storm sewer systems.   Because abatement of pollution from the dispersed sources
 served by drainage systems began with  a focus on combined systems, our knowledge on
 storage requirements  for them is greater.  Recognized very early was the need for
 some form of automatic control,''' because of system complexity and the need to
 manipulate flows in  order to insure their containment as a means for reducing
 overflows.

 System Performance

          When rapid removal of stormwater was the primary objective, hydrological
 considerations were mostly restricted  to the sizing of conduits.  Retardation of
 flows and attenuation  of their  peaks is another matter, as is containment of flows
 and their diversion  for treatment to abate pollution.  Whereas simple concepts
 sufficed for conduit sizing, their perpetuation in more complex applications
 inevitably leads to  lowered reliability.

          An attempt is made herein to indicate how precipitation data can best be
 employed in planning and design of stormwater facilities under emerging imperatives
 Responsibilities of  local government operating agencies were formerly restricted1 to*
 the management of underground conduit systems that had been deliberately designed to
be overloaded upstream of their outlets rather often, such as once in five years on
 the average, with little regard for the impact of system flows on receiving waters.
Thus,  the sole criterion of performance was how effectively the land had been
drained.   Once containment of flows and their pollutant burdens become additional
objectives,  performance criteria enlarge to include the meeting of regulations and
 statutes.   As a result, public officials will no longer be able to depend only on
the relatively obvious incidence of system overloading evidenced by flooded
basements and streets.  They will also have to show concrete evidence of compliance
acceptable to regulatory agencies.   This means that they will have to know when
                                       - 100 -

-------
how much and how often allowable limits have been exceeded.   Regulatory limits
necessarily have to be founded on some level of frequency,  or probability,  such as
containment on the average of all but one combined sewer overflow every five years,
or of all but a certain volume of stormwater over a season,  or that would permit no
more than a certain maximum pollutant discharge per storm.

          The only way an event frequency can be determined is by referring to  a
reasonably large series of prior events.  A long history of overflow or storm
discharge amounts is almost never available and, even if it were, it would not  be
suited for systems revised with new storage and controls.  Thus, recourse is made
to simulation of system performance via some sort of calculation, such as the use
of hydrological models.  That is, a historical record of rainfall is transformed by
calculation into a simulated historical record of stormwater flow and possibly
quality.  Simulation of an existing system thereby yields a synthesized historical
record of pre-modification conditions.  System revisions are then targeted to
contain all but some relatively rare events from the synthesized performance record,
as prescribed by regulations.  But how will compliance be judged after a system has
been modified to meet a regulation?  Of course, monitoring all overflow/discharge
points by means of rainfall-runoff-quality instrumentation would indicate when
violations had occurred, but would not indicate why.  Further, reality dictates
that  from an economical standpoint, only a sample of overflow/discharge points could
probably be monitored, and simulation of post-modification performance can be employed
to interpret sparsely monitored systems and explain whether or not violations have
occurred and why.

          At least a decade ago there was widespread acceptance  of the need  for
complete storm hydrograph simulation  if major  system modifications were contemplated,
particularly if new storage was involved.  However, as we enter  an age of regulation
another need arises, for simulation of the performance of systems modified  to meet
regulations.  Unfortunately,  concepts  that were adequate for  sizing  conduits have
been carried into  the sizing  of detention storage and now threaten to be perpetuated
in the future testing of compliance performance  in  flooding  relief and pollution
abatement  facilities planned  for  construction.  For this reason, it  is necessary to
review concepts  that have been used for decades,  to show why they are not  suited for
investigating performance, either expected or  experienced.   These now outmoded
concepts originated in  connection with the  sizing of conduits.

Simplistic Traditions

           The overwhelmingly  used technique  for sizing  combined and  storm sewers  is
known as  the "rational  method".(8)  its  limitations have been discussed  at length
elsewhere,(9^ and  present  purposes  are served  by  concentrating on  its  fundamental
premise.

           Intuition,  theory  and laboratory  tests  indicate  that when a rain of  a
constant  intensity falls on  a catchment,  the outflow  from  the catchment will
ultimately rise to a  maximum rate,  which would be sustained as long as the constant
 rainfall  intensity continued  thereafter.  The time  required to reach the maximum
outflow rate  (to reach a  state of equilibrium) is commonly called  the "time of
 concentration".  The  equilibrium discharge  is merely  a fixed fraction of the
 constant  rainfall  intensity  in the  rational method,  with the ratio of equilibrium
 discharge to  rainfall intensity defined by  a constant that is usually interpreted in
 terms of type of land use.   Because for a given catchment,  equilibrium discharge is
 a f^ed multiple of input constant rainfall intensity,  it  follows  without reservation
                                        - 101 -

-------
 that the frequency* of an equilibrium discharge is identical  with the frequency of
 its associated constant rainfall intensity.

           Storm and combined sewers are usually designed  to  flow full without
 surcharging at some selected frequency, such as on an average of once in  five years
 or once in ten years.   Thus, the only type of rainfall information needed  for use
 of the rational method is an array of intensities  for each of several durations from
 which various frequencies can be interpolated.   The result is called  "intensity-
 duration- frequency curves".   Such curves have been developed  in  a number of
 municipalities and advantage has been taken  of the spatial persistence of  curve
 values in the preparation of national maps that contain lines of equal rainfall for
 given durations and frequencies, such as by  Yarnell,'^)  Hathaway(Il) and
 Hershfield.(12)  Certain characteristics of  such information  must be  reviewed next
 in order to demonstrate some rather severe limitations.

 Intensity-Duration-Frequency Relations

           The almost universal data source is  the  U.S.  Weather Service, from its
 first-order stations in principal cities.

           In principle,  one  would search the  recorded  rainfall for each storm at a
 given station for the  largest catch over a particular  duration.   Because engineers
 have been more interested in the rarer events,  and  to  reduce  the  data processing and
 reporting burden,  the  USWS processes  and reports data  for  only those  storms where  at
 least some rainfall  depths exceed a specified  threshold level.  The level  is
 sufficiently low  to  include  1-year frequency and somewhat  lower values.  Data from
 storms  having depths above the threshold are  termed "excessive precipitation".
 Because for a given  duration the approximately  largest  depth  of rainfall is reported
 it  is therefore the  maximum  amount for that duration for that  storm at that gage.    '
 Because there is  considerable variability  in rainfall  over time,  the  selected amount
 when expressed  as  an intensity,  is  necessarily  the average intensity  for the given
 duration.   Hence,  we will hereinafter  use  the term maximum average when describing
 excessive  precipitation.

           In  deriving  intensity-duration-frequency relations,  rainfall values for
 each  duration are  regarded independently from other durations, the first step being  a.
 separate  ranking of  values for  each duration in descending order of size.  A
 mathematical  fit  is  made  to  the  array  of depths or intensities for each duration and
 the  line of best  fit values  are  computed.  In the last step,   fitted values for  each
 duration  for  a  specified mean  recurrence  interval (or  frequency) are  plotted with
 intensity as ordinate  and duration  as  abscissa; and smooth fitted lines are drawn
 through points  of  equal  frequency.

          Starting in  1936 the USWS published maximum average excessive precipitation
 for various durations,  presumably  to facilitate rational method applications, but
which makes it necessary  to refer  to the National Climatic Center, NOAA Environmental
Data Service, Asheville, N.C.,  for  the  true sequence of occurrences.   From 1895
 through 1935  the USWS  published maximum  average excessive precipitation as it had
accumulated over time,  mostly  for a 5-minute interval.  The report of a study of
 frequency analysis methods includes a  listing of such data for 1913-1935 for
Chicago,(13) USed  in the  succeeding discussion.

*:  (The terms  "frequency," "return period" and "recurrence interval" are used
     interchangeably  in drainage practice, as  they will be here.  Frequency is the
    reciprocal of  probability).


                                       - 102  -

-------
          Listed in Table 1 are the top 14 values of maximum average excessive
precipitation for Chicago,  1913-1935„   Casual inspection of Table 1 will reveal the
fact that maximum average depths for a given average return period are not necessarily
from the same storm, e.g. see Rank No. 5.  Inspection of actual records of
accumulation would reveal that 5-minute maximum average depths did not necessarily
occur in the first five minutes, and so on for other durations.  Also, the record
includes storms that lasted as little as five minutes, with a majority of the 130
storms reported lasting less than 35-minutes, and with very few lasting more than an
hour.

          Further, in order to insure a complete data matrix for regression analysis
to develop frequency curves, a dummy value is inserted for all durations beyond the
cessation of excessive precipitation for a given storm.  For example, for the storm
of 7/16/14, to be examined later, 1.61-inches had accumulated at the end of the
storm, which lasted 35-minutes, and for frequency analysis purposes a fictitious
value of 1.61-inches would be inserted for durations from 40 to at least 120 minutes
for that storm prior to the ranking of maximum averages.  The insertion of dummy
valufes (called "extended duration" values) is consistent with the underlying concept
of the rational method, but further distorts intensity-duration-frequency curves from
the actual history of occurrences.  This distortion is illustrated in Figure 1, where
maximum average values of rainfall for a T£ (the average return period) of 4.6-years
and 5.8-years from Table 1 have been converted into intensities.  Compared is a
"5-year" curve for Chicago attributed to Eltinge and Towne.^1^  The  falloff of
plotted points beyond a duration of 40-minutes, compared with the uninterrupted
smooth curve, reflects the exclusive use of actual data for the former and the
traditional use of extended duration values for the latter.  Note that the plotted
points in Figure 1 cease at 80-minutes because only one storm from the record used
had lasted any longer but was a rarer value.

          In referring to a frequency curve  in the context of  the rational method,
some engineers erroneously use  the term  "storm frequency".  For example, values  taken
from a 5-year curve or the resulting computed flow rate are often stated to be  for
"a five year storm".  Considering  that a given frequency curve can represent values
from different storms, in a time sequence  different than actually occurred, and  might
include non-existent dummy values, it is obvious  that  any  reference  to  the term
"storm" in rational method applications  is at least misleading and is  certainly
technically  imprecise.

          Further,  it should be noted that the time of concentration varies  from
point to point in a catchment  area.  Thus, using  a  given frequency curve, various
portions of  a drainage area may be designed  on the basis of  pieces of different
storms that  might have occurred several  years apart.   Hence,  the  presentation of
water surface profiles or hydraulic gradients from upstream  extremities  to  the
outlet for "design  conditions"  can be greatly misleading.

           In summary,  the  term "storm"  should never be used  in connection with the
rational method, whereas the  term  "rainfall"  is  adequately vague  and innocuous.  In
fact, an outstanding  limitation of the  rational  method stems from its complete
independence from storm  pattern.   To  reiterate,  the maximums of the several
durations  from a given  storm  record as  used  in  the  rational  method are not
necessarily  in their  original  sequential order;  and  the resulting tabulation of
maximuras ordered by size of duration  may bear  little  resemblance to the original
storm pattern.
                                        - 103 -

-------
TABLE 1 - HANKING BY MAXIMUM DEPTH FOR VARIOUS DURATIONS,  CHICAGO,  1913-1935
BANK
1

2

3

4

5

6

7

8

9

10

11

12

13

14

AVERAGE
RETURN
PERIOD,
YEARS
23.0

11.5

7.7

5.8

4.6

3.8

3.3

2.9

2.6

2.3

2.1

1.9

1.8

1.6

MAXIMUM DEPTH
5-min.
0.55
(6/26/32)
0.53
(6/29/20)
0.51
(8/11/31)
0.50
(6/20/28)
0.48
(8/8/27)
0.45
(8/11/23)
0.42
(8/27/31)
0.41
(10/5/19)
0.41
(8/13/25)
0.41
(8/2/33)
0.39
(9/5/20)
0.39
(7/7/21)
0.39
(7/20/24)
0.38
(5/18/26)
10-min.
0.96
(8/11/31)
0.94
(6/20/32)
0.88
(6/20/28)
0.80
(8/11/23)
0.80
(7/2/33)
0.76
(7/7/21)
0.74
(10/5/19)
0.74
(8/27/31)
0.71
(8/2/35)
0.68
(8/13/25)
0.68
(5/18/26)
0.68
(6/29/33)
0.67
(6/18/35)
0.66
(7/16/14)
15-min.
1.22
(8/11/31)
1.16
(6/20/28)
1.16
(6/26/32)
1.12
(8/11/23)
1.08
(7/7/21)
1.00
(7/2/33)
0.96
(8/27/31)
0.93
(10/5/19)
0.87
(6/13/26)
0.87
(8/2/35)
0.86
(7/20/24)
0.86
(6/18/35)
0.84
(8/13/25)
0.83
(8/9/14)
20-min.
1.39
(8/11/23)
1.39
(8/11/31)
1.38
(6/20/28)
1.35
(7/7/21)
1.21
(8/27/31)
1.16
(7/2/33)
1.13
(6/13/26)
1.10
(10/5/19)
1.05
(7/20/24)
1.05
(8/13/25)
0.97
(7/16/14)
0.96
(6/29/33)
0.96
(8/2/35)
0.95
(5/18/26)
2 5-min.
1.58
(6/20/28)
1.49
(8/11/23)
1.45
(7/7/21)
1.34
(6/13/26)
1.29
(10/5/19)
1.27
(8/27/31)
1.25
(7/20/24)
1.21
(7/16/14)
1.21
(8/13/25)
1.18
(6/29/33)
1.07
(8/15/34)
1.05
(9/5/20)
1.05
(7/1/27)
1.04
(5/18/26)
IN INCHES
30-min.
1.70
( 6/20/28 )_
1.61
(8/11/23)
1.55
(6/13/26)
1.49
(7/7/21)
1.48
(7/16/14)
1.44
(10/5/19)
1.43
(7/20/24)
1.38
(6/29/33)
1.31
(8/13/25)
1.20
(8/29/28)
1.19
(8/15/34)
1.12
(9/5/20)
1.11
(7/1/27)
1.05
(8/2/35)
FOR STATED
35-min.
1.86
(6/20/28)
1.77
(6/13/26)
1.71
(8/11/23)
1.61
(7/16/14)
1.59
(7/20/24)
1.53
(10/5/19)
1.53
(6/29/33)
1.38
(8/13/25)
1.30
(8/29/28)
1.21
(8/15/34)
1.17
(7/19/31)
1.16
(9/5/20)
1.09
(6/25/26)
1.05
(7/19/16)
DURATIONS;
40-min.
1.92
(6/20/28)
1.85
(6/13/26)
1.81
(6/29/33)
1.77
(8/11/23)
1.75
(7/20/24)
1.24
(7/19/31)
1.23
(8/15/34)
1.22
(9/5/20)
1.13
(6/24/25)
1.10
(9/10/22)
1.08
(7/19/16)
1.08
(8/5/24)
1.06
(8/2/35)
1.03
(6/18/35)
AND DATE OF RAINFALL
45-min.
2.04
(6/20/28)
2.03
(6/29/33)
1.80
(8/11/23)
1.31
(7/19/31)
1.25
(6/24/25)
1.14
(9/10/22)
1.12
(8/5/24)
1.06
(7/11/22)
1.05
50-min.
2.06
(6/20/28)
1.85
(8/11/23)
1.40
(7/19/31)
1.23
(9/10/22)
1.17
(8/10-11/31)
1.15
(8/5/24)
1.12
(7/11/22)
1.02
(6/23/31)
0.87
(8/10-11/31) (6/15/25)
1.00
(6/23/31)
0.98
(6/12/15)
0.82
(6/15/25)
0.74
(9/17/27)
0.73
(4/10/22)
0.80
(9/17/27)
0.79
(3/31/29)
0.78
(4/10/22)
0.78
(6/24/28)
0.74
(8/19/21)
60-min. 80-min. 100-min. 120-min.
2.30 2.30 1.36 1.53
(6/20/28) (8/11/23) (6/23/31) (6/23/31)
2.02 1.55
(8/11/23) (9/10/22)
1.53 1.54
(9/10/22) (8/5/24)
1.27 1.45
(8/10-ll/31)(8/19/21)
1.22 1.23
(8/5/24) (3/31/29)
1.21 1.22
(7/11/22) (6/23/31)
1.13 0.96
(6/23/31) (7/5/30)
1.12
(8/19/21)
0.99
(6/24/28)
0.97
(9/17/27)
0.89
(3/31/29)
0.82
(7/5/30)
0.81
(9/23/26)



-------
   .  3
i- CO

g 2
.  UJ
  <
  o:
     0




  FIGURE 1
                     TE = 5.8-YEARS (1913-1935), FROM TABLE 1


                     TE = 4.6-YEARS (1913-1935), FROM TABLE 1
                                           ,TC « 5-YEARS  (ELTINGE AND TOWNE)
                               I
                             I
                          I
   20
40
60      80      100

DURATION, MINUTES
                                                      120
                                                    140
                                                 160
180
- INTENSITY-DURATION-FREQUENCY COMPARISONS, CHICAGO  DATA

-------
5
30-35
20-25
10
30-40
20-30
15
20-35
20-35
20
20-40
10-30
25
15-40
5-30
30
10-40
0-30
35
5-40
0-35
~~40 — ~~
0-40

 Volumetric  Considerations

          Because  the  sizing,  deployment  and operation of detention storage facilities
 is  determined  by the time  patterns  and  amounts of rainfalls selected, the use of
 actual  precipitation records  is greatly preferred,  for reasons already given.  Listed
 in  Table  2  are the top 14  storms  in terms of total  rainfall depth for Chicago, 1913-
 1935,  for the  same record  from which Table  1 was derived.  Shown are the recorded
 depths  for  successive  intervals of  time.  It may be noted that the larger amounts of
 rain occur  early,  late, and in the  middle of storms, and some have double peaks
 such as Rank No. 5.

          In order to  illustrate  the mixture of individual frequencies of separate
 parts of  storms, the actual depths  for  two  storms given in Table 2 have been traced
 through Table  1 (and its extension,  not shown), and the result is given in Table 3
 Note the  wide  disparity in individual storm component frequencies.  Also to be
 considered  is  the  distortion in timing of the maximum average intensities:

                                       Occurrence During Storm of Maximum
                             	Aver;
 Rank           Storm

 4           6/13/26
 5           7/16/14


which may be checked by inspection of Table 2.

          Hopefully, some  indication has been presented of the folly of attempting
 to define a design storm by using concepts applicable solely to the rational method

Another Example

          As a companion to Table 2, listed in  Table 4 are the top 14 storms in terms
of total rainfall depth for Philadelphia,  1913-1935.  What has been said about the
Chicago data is generally applicable for the Philadelphia data.   In addition,  because
the writer has been supplied a copy of the complete USWS  record by the Philadelphia
Water Department,  certain features from that record can be cited —

.  For the storm of August 7,  1921,  0.65-in. fell  between 3:40 p.m.  and  5:05 p.m.
   0.81-in.  between 6:10 p.m.  and 7:00 p.m., and  the 1.91-in.  indicated  in Table 4
   followed  between 10:15  p.m.  and 11:40 p.m.

.  For the storm of May 27, 1918,  1.28-in. fell between 7:38  p.m.  and  8:45  p.m.,
   followed  by a 0.25-in.,  and then  the  1.62-in.  indicated  in  Table  4  followed betwe
   10:18 p.m.  and 11:33 p.m.                                                         Sa

.  For the storm of June 27,  1914, the 1.60-in. indicated  in Table 4 fell  between
   3:21  p.m. and 4:15 p.m.  and was followed  by 0.56-in. between  5:40 p.m.  and  8:10

.  Note  that in Table 4 the storm of July  10, 1931 with a  total  depth  of  1.57-in.  wa
   preceded  only three  days earlier  by a storm with  a  total depth of 1.43-inches
  Also,  0.91-in. fell  in  a little over  half an hour on July 14,  1931, followed bv
   1.03-in.  on July 18, 1931, which  fell mostly in a 26-minute period.
                                       -  106  -

-------
                              TABLE 2 - BANKING BY DEPTH OF STORM, CHICAGO, 1913-1935
RANKING BY MAGNITUDE
OF TOTAL STORM DEPTH

RANK
1

2

3

4

5

6

7

8

9

10

11

12

13

14


AVERAGE
RETURN
PERIOD,
YEARS
23.0

11.5

7.7

5.8

4.6

3.8

3.3

2.9

2.6

2.3

2.1

1.9

1.8

1.6


TOTAL
STORM
DEPTH,
INCHES
2.30
(6/20/28)
2.30
(8/11/23)
2.03
(6/29/33)
1.85
(6/13/26)
1.61
(7/16/14)
1.55
(9/10/22)
1.54
(8/5/24)
1.53
(10/5/19)
1.53
(6/23/31)
1.49
(7/7/21)
1.45
(8/19/21)
1.40
(7/19/31)
1.39
(8/11/31)
1.38
(8/13/25)
-. 1 •! " —

0-5
0.22

0.10

0.31

0.08

0.27

0.08

0.12

0.09

0.18

0.32

0.20

0.15

0.26

0.10
•^^VH^H


5-10
0

0

0

0

0

0

0

0

0

0

0

0

0

0


.50

.27

.37

.22

.24

.23

.12

.19

.23

.39

.16

.21

.45

.21




10-15
0

0

0

0

0

0

0

0

0

0

0

0

0

0


.38

.35

.12

.21

.19

.08

.23

.33

.26

.37

.22

.32

.51

.16


RECORDED
15-20
0.28

0.45

0.05

0.21

0.12

0.04

0.10

0.41

0.09

0.27

0

0.23

0.17

0.27


DEPTH
20-25
0.20

0.32

0.25

0.31

0.33

0.16

0.12

0.17

0.12

0.10

0

0.07



0.41


IN INCHES FOR STATED TIME INTERVAL IN MINUTES
25-30
0.12

0.09

0.28

0.23

0.33

0.15

0.11

0.19

0.08

0.04

0

0.06



0.16


30-35 35-40 40-45 45-50 50-60 60-80 80-100 100-120
0.16 0.06 0.12 0.02 0.24

0.13 0.06 0.03 0.05 0.17 0.28

0.15 0.28 0.22

0.33 0.26

0.13

0.13 0.01 0 0.21 0.44 0.02

0.08 0.20 0.04 0.03 0.07 0.32

0.15

0.03 0.01 0 0.02 0.11 0.09 0.14 0.17



0 0 0.07 0.09 0.38 0.33

0.13 0.07 0.07 0.09



0.07
	 __ 	

o
•H

-------
TABLE 3 - RETURN PERIODS FOR MAXIMUM DEPTHS
          OF TWO STORMS, CHICAGO, 1913-1935
RANK
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
AVERAGE FROM
RETURN RANKING
PERIOD, BY DEPTH
YEARS OF STORM
23.0
11.5
7.7
5.8 6/13/26
4.6 7/16/14
3.8
3.3
2.9
2.6
2.3
2.1
1.9
1.8
1.6
1.5
1.4
1.3
1.3
1.2
1.2
1.1
1.0
1.0
1.0
0.9
0.9
0.8
OCCURRENCE OF MAXIMUM DEPTHS FOR GIVEN STORMS
5-min. 10-min. 15-min. 20-roin. 25-rln. 30-min. 35-min. 40-min.

6/13/26 6/13/26
6/13/26
6/13/26 7/16/14
7/16/14

6/13/26
7/16/14
6/13/26

7/16/14


7/16/14



7/16/14

6/13/26





6/13/26
7/16/14

                -  108  -

-------
TABLE 4 - BANKING BY DEPTH OF STORM,  PHILADELPHIA,  1913-1935
RANKING BY MAGNITUDE
OF TOTAL STORM DEPTH

RANK
1

2

3

4

5

6

7

8

9

10

11

12

13

14


AVERAGE
RETURN
PERIOD,
YEARS
23.0

11.5

7.7

5.8

4.6

3.8

3.3

2.9

2.6

2.3

2.1

1.9

1.8

1.6


TOTAL
STORM
DEPTH,
INCHES
2.70
(6/26/30)
2.09
(8/16/17)
1.91
(8/7/21)
1.82
(7/15/26)
1.82
(4/21/27)
1.76
(7/3-4/26)
1.74
(5/24/33)
1.63
(7/13/19)
1.62
(5/27/18)
1.60
(6/27/14)
1.57
(7/10/31)
1.48
(7/7/31)
1.43
(6/20/13)
1.39
(8/28-29/23)





RECORDED DEPTH IN INCHES FOR STATED TIME INTERVAL IN MINUTES
0-5
0.38

0.41

0.12

0.06

0.07

0.12

0.38

0.06

0.09

0.15

0.25

0.32

0.34

0.08


5-10
0.

0.

0.

Oc

0.

0.

0.

0.

0.

0.

0.

0.

0.

0.


32

30

02

04

08

11

49

05

08

22

18

40

19

10


10-15
0

0

0

0

0

0

0

0

0

0

0

0

0

0
--
.32

.37

.08

.05

.13

.18

.29

.14

.03

.21

.20

.46

.20

.11

15-20
0.29

0.23

0.24

0.05

0.09

0.15

0.23

0.28

0.04

0.21

0.16

0.30

0.22

0.13

20-25
0.08

0.49

0.24

0.17

0.08

0.15

0.12

0.04

0.11

0.14

0.16



0.38

0.12

25-30
0.02

0.27

0.17

0.14

0.11

0.09

0.03

0.02

0.23

0.09

0.03



0.10

0.11


30-35 35-40 40-45 45-50 50-60 60-80 80-100 100-120
0.04 0.05 0.10 0.08 0.33 0.69

0.02

0.18 0.04 0.02 0.02 0.06 0.62 0.10

0.13 0.26 0.24 0.13 0.15 0.40

0.07 0.01 0.20 0.24 0.31 0.20 0.23

0.08 0.16 0.15 0.24 0.28 0.05

0.02 0.01 0 0.02 0.05 0.06 0.02 0.02

0.20 0.36 0.23 0.18 0.07

0.20 0.11 0.15 0.08 0.05 0.45

0.01 0.02 0.22 0.20 0.13

0.06 0.02 0.10 0.21 0.20





0.08 0.12 0.15 0.15 0.24



-------
    On August 3,  1898,  5.43-inches  were recorded for the period between  10:50  a.m.
    and 12:35 p.m.,  only 105 minutes;  and on September 14 and 15,  1904,  an  equal
    amount was recorded, 3865-in.  in 149 minutes preceded by almost  3-in. which had
    ended only about seven hours earlier and most of which occurred  in a period of
    only one hour.

           These  examples should give  a clear indication of why synthetic storms  can
 be a liability in detention storage evaluations.  Only by referring to  actual records
 as a basis for simulating performance can multiple events be realistically taken
 into account. Additionally,  a synthetic storm approach neglects  surficial and
 subsurface effects  of  prior storms on the magnitude of runoff and pollutants  from a
 given storm (commonly  called  antecedent conditions).

 Synthetic Storms

           Graphed in Figure 2 is a "5-year" synthetic storm developed in Chicago.(14)
 Also graphed thereon are the  two storms from Table 2  with average return periods of
 4.6-years and 5.8-years on the basis  of total storm depth.   It  is difficult to decide
 just how the two actual storms should be placed with  respect  to time for comparison
 and the positions shown are quite  arbitrary.   The  extension of  the  synthetic  storm
 to 180-minutes and  its  overstated  total depth (2.28-inches  versus 1.61-inches and
 1.85-inches  for the  actual storms)  reflect its  origin as  an intensity-duration-
 frequency curve containing extended duration values described earlier.

           The method of storm synthesis for Chicago reported  in 1957 was criticized
 on the  grounds that  it  retained too many of the fallacies  and empiricisms  inherent
 in the  rational method  to  recommend its principle  for general use by others. *• *'
 The authors(14) were asked(l^) why  could not  actual storms  be used  instead, in the
 later development of general  design criteria.(16)  «£0  be  fair, all  this occurred before
 high-speed,  general-purpose computers  were readily available  and the processing of a
 number  of storms rather than  only one  was  simply not  practicable at that time.
 However,  when it came time to  prepare  a master  plan for  combined sewer overflow
 abatement  in  metropolitan  Chicago  that incorporated extensive underground  storage (17,18)
 hourly  precipitation records  for 1949  through 1969 for  twenty raingages in  the
 metropolitan  area were  used in conjunction with over  a  score  of computer models as a
 means for determining system  component  sizes, and  necessary waterway improvements and
 waterway  quality for each  plan alternative.(19)

          Perhaps because  the  Chicago  synthetic storm was  featured in a widely used
 handbook  ^ that had been  first issued  in  1960  and last  in  1969, there have been a
 number of reports and published papers  that either adopt the method or allude to its
 use.  This paper has attempted to indicate why  synthetic storms are a poor choice for
 simulating system performance, particularly where  detention storage of any significance
 is  involved.

          Availability of Chicago data makes possible the illustration of another
 synthetic storm procedure.  A  computer model has been developed for storm sewer design
 called ILLUDAS (Illinois Urban Drainage Area Simulator),(20'  that had been described
 earlier in an ASCE journal.(2i)  A design  storm procedure for Illinois sites is
 recommended in the users' manual,(20)  Graphed  in Figure 3 are the two storms from
Table 2 with average return periods of 4.6-years and 5.8-years on the basis of total
 storm depth.  Also graphed are "5-year" synthetic  storms for ILLUDAS criteria, setting
 the total duration in each case equal  to the duration of the actual storm compared.
                                       - 110 -

-------

8


7
^
i 4
O
I
tr
•s
t-1 •»
Eeo 3
1 1
O
—
> 2
i-
co
z
LJ
1-
2 I

0
	 1 	 T 	 1 	 1 1 1 I >
I

ll
M^« j^J 	
I STORM OF JULY 16
TE =4.6-YEARS(19
ON BASIS OF TOT
DEPTH (1.61-IN.) v



__



—
-r^'^i
) 20 40
, 1914,
13-1935)
'AL
h
•" ••:

1
/
/
f
1
nrVr
i i
1 \
1
[ •»«






^
STORM OF JUNE 13, 1926,
: ': x TE = 5.8- YEARS (1913-1935)
• : xv^ *•
: ^ ON BASIS OF TOTAL
! i DEPTH (1.85-IN.)
!l"
\
\
\
\
\
\
"~n v
\ M5-YEAR" SYNTHETIC STORM
/ \ y (TOTAL DEPTH = 2.28-INJ
r*/ ^^
^Z ^^* — 	
• ^'^ —
• — — 	 ^^^





^










^~ 60 80 100 120 140 160 18O
                      TIME, MINUTES
FIGURE 2-ACTUAL  VERSUS SYNTHETIC STORM PATTERNS,  CHICAGO

-------
U.D
0.5

[30.4
X
o
Z n -a
( J T
^~ XX* W
t 0.2
LJ
o
0.1
n




1
	 ^"5-YEAR," ILLUDAS CRITERIA
f (TOTAL DEPTH = 1.33- IN.)
1
1
1
	 T

	 1
^^
L.
—

0 10
STORM OF JULY 16, 1914,
/TE-4.6-YEARS
((1913-1935) ON BASIS OF




	





TUTAL DEPTH

t ^_
1.1.61-IN.)


4 	 ' 	 , 	 L___

20 30


-
1
40
                  TIME,  MINUTES
U.D

0.5

S04

^
-0.3
X
fc
gjO.2
Q
0.1
0
1 1
"5-YEAR" ILLUDAS
_ f (TOTAL DEPTH = .
i i
CRITERIA
39- IN.)
STORM OF JUNE 13, 1926,
TE = 5.8-YEARS (1913-1935)
1 ON BASIS OF TOTAL DEPTH
1 	 1
i
I
i
, 	 ,
I
[__--,
, i
4/1 on


-IN.)
	 \






-

_

    0        10        20       30       40
                  TIME,  MINUTES

FIGURE 3-COMPARISON  WITH  ILLUDAS,  CHICAGO
                   - 112 -

-------
The synthetic storms yield lesser volumes.  They have an advanced type of pattern,
with decreasing depth over time, the characteristics of distributions from intensity-
duration-frequency curves such as the one illustrated in Figure 4.  It seems evident
that the ILLUDAS synthetic storm criteria also evolved from the rational method
concept.

          Because there are inherent non-linearities in most methods for processing
inputs (subtracting assumed abstractions from total rainfall to derive the net
rainfall associated with runoff) for linear models, and dynamic models are non-linear
by definition, the statistics of the input array may differ appreciably from those of
some or all of the arrays for ruroff and quality characteristics.  Attempting to
assign a mean frequency of probable occurrence to a "design storm" is meaningless
because of statistical nonhomogeneity of rainfall, runoff and quality.

          In an attempt to test the design storm transformation hypothesis, a
graduate student at Purdue University calibrated the ILLUDAS model for a 13-acre
Chicago catchment for which synchronous rainfall-runoff field data had been collected,
and input both a 35-year rainfall record for Chicago(•"' and ILLUDAS synthetic
storms of various frequencies.(22)  Concluded was that the ILLUDAS synthetic storms
of given return periods produced peak runoffs of essentially the  same return period
as those generated from the long-term rainfall record, at least  for  return periods of
25-years or less.  Unfortunately, while the student reproduced the table of
accumulated depths for the 35-year record in an appendix of his  report, he had
mistakenly used, instead, another table listing maximum values for each duration,
skewing all storm data into advanced patterns.  Therefore, the only  conclusion  that
can be reached  is that advanced rainfall  patterns give very similar  peak discharge
arrays.  What happens with actual storm pattern data  thus remains to be demonstrated,
because the well-intended Purdue experiment appears  to have been the first  reported.

          However, an  inkling of what might be  found  is  suggested in a  test  of  a
synthetic storm used in a preliminary study of  stormwater pollution  control  for a
major metropolitan area.(23)  Because a planning  model(2^>25) was used  for  the  study
that normally operates on hourly precipitation  data,  one-hour  interval  rainfall data
was used  for  the  test.  The model was calibrated  for  a  catchment draining  into  the
Charles River,  and  then run with a  long-term  precipitation record for Boston and with
the synthetic  storm.   From  the  test  it was  found  that storage  sized  to  contain  either
90% or 95% of  the  total runoff  volume resulting from the synthetic storm would  be
capable  instead of  containing only  about  607.  or 807.,  respectively,  of  the  total
runoff volume.

          So  far,  we have restricted discussion to the shorter return periods
associated with storm  drainage  system sizing.   A  synthetic  storm expected  to yield
the peak 100-year urban  streamflow  was  developed  in a study of the Four Mile Run in
northern Virginia^26'  following the scheme developed in Chicago.('  However,  the
 100-year peak flow obtained from an extrapolation of a series  of peak flows in
 simulations  with a long  period  of actual storm data was used to check the validity
of the synthetic storm.   The 100-year peak flow simulated  using the synthetic storm
was only 11% lower,  and  the synthetic storm was enlarged accordingly for its
 application.in subsequent simulations using a much more elaborate model.  Demonstrated
 in this study was that a simple continuous simulation model can be used to develop
 a synthetic storm for  later use in  a complex single-event simulation model for more
 reliable estimation of a rare event such as a 100-year peak flow.  Use of this
 aoproach for more frequent events would be even more reliable,  provided the simple
 simulation model was first calibrated with local rainfall and runoff data as in the
 case above.

                                        - 113 -

-------
   0.5
to
LJ
I
o
Q.
LJ
0
   0.4
   0.3
   0.2
   0.1
T-j-R
1
1
- 1
1
1
1
1


1
1

1


1 1 \ \ 	 1 	 -j
DR ELTINGE AND TOWNE, "5-YEAR"












STORM OF JULY 16, 1914,
/TE = 4.6-YEARS (1913-1935)
k ON BASIS OF TOTAL DEPTH (1.61-IN.)
/ ^/STORM OF JUNE 13, 1926,









	

t

	 1
j
i
j
1
i 	


	 £, TE=5.8-YEARS (1913-1935)
ON BASIS OF TOTAL DEPTH
j (1.85-IN.)
i
i 	 ,




™"~" i
	 1
L._, J
	 1





"""»— L— ,
(CONTINUES)
i


	 . 	 i
     0
                   ELAPSED TIME,  MINUTES
60
   FIGURE 4- STORM  PATTERN COMPARISONS, CHICAGO
                           - 114 -

-------
Some Preliminary Planning Implications

          Preferred inputs for modern planning models are reasonable lengths  of
actual rainfall records, perhaps spanning at least 20 years.   It seems  more
reasonable to route rainfall data of local record through a model to arrive at
output parameter frequencies than to synthesize a storm of some assumed probability.
Use of meteorological expedients can conceivably be unnecessarily hazardous and the
results obtained thereby can be extremely misleading.

          In the development of the San Francisco Master Plan for abatement of
pollution from combined sewers, 62-years (1907-1968) of hourly rainfall from the
local U.S. Weather Service gage were used(27,28) ^n conjunction with an early
version of an uncomplicated planning model that has since been iraproved'24>25) an(j
extensively used elsewhere.

          As mentioned earlier, for the development of the Chicagoland Master Flan
for abatement of overflow pollution and reduction of flooding, 21-years (1949-1969)
of rainfall data from the local network of twenty raingages was used.'^)  Some 104
individual drainage basins in the 375-sq. mi. study area were involved, and thirty-
two computer simulation models were developed as part of the planning effort.

          Features of both of the above-mentioned master plans are outlined in an
ASCE Program report.(29'

          Another example is the use of a 60-year rainfall record with a rainfall-
runoff model.  The model was calibrated using 4-year to  10-year  periods of concurrent
rainfall  and runoff observations for  each of 26  partly sewered catchments ranging  in
size from 1/2 to 88 square miles, and an imperviousness  of 2  to  35  per cent.''^'
Primary  interest was on peak discharges.

Some Design Implications

          Because there is  insufficient evidence to  justify reform,  it is difficult
to  fault  the use of synthetic  rainfall distributions and simplistic  methods  in routine
design of storm  sewers  and  detention  reservoirs  of  small size.   Major  facilities  that
cost substantially more are another matter  entirely.

          Because  the  cost  of  running the more  elegant design/analysis models per
storm  event  is high, many defenders of these models  champion  acceptance of a
svnthetic storm  as an  expedient to  save  time  and money,  but at the  expense of
credibility of results.   This  is not  to  suggest that all catchments of a  jurisdiction
where  large  capital  investments are  contemplated should  be analyzed using 20 years or
wore of  continuous rainfall records  on a complex model.   Rather, as some  of  the
leading  practitioners  do,  such a long record  should be  applied to a calibrated
catchment near  the reference  weather  station  to segregate those storms of design
 •mportance.   Because only the unusual occurrences are of design interest,  there  may
be perhaps  only  two  dozen or  so actual storms  of concern.   To be consistent,  any
 oroject sufficiently important to  call for the use of a complex model should also be
 important enough to  apply a few storms rather than a single synthetic storm.  Thus,
 the handful of storms  selected on the basis of simulated catchment  response become
 a family of design storms for use in connection with other catchments in the
 Jurisdiction.   Officials  in charge of urban drainage facilities are hard-put to
 explain an artificial  synthetic storm's frequency to irate citizens who have been
 flooded or to a State  official regulating overflows.  Defense against storms of
                                        - 115 -

-------
 record is rather direct,  and in the opinion of the writer the only  realistic  option
 open to a local government official.  The temptation to use artificial  confections
 as input data should be resisted.   In addition,  the only way to  check the  expected
 performance of a facility is to simulate not only the design loading but greater
 possible loadings as well.

           Even if very simple procedures are used,  e.g.  to  determine only  a peak
 flow rate,  the collateral, occasional monitoring of computations by means  of  one  of
 the more complete models  can serve as an auxiliary guide to sharper judgment.

           In the United Kingdom, a national study has been  undertaken that
 demonstrates that economic analysis-planning-design of storm sewer  systems is
 technically feasible,  on  a limited scale.^'  However,  needed in particular  would
 be development of damage  cost-curves and risk  cost-curves.   Underpinning damage
 cost curves would be performance evaluation, which would have as its objective to
 simulate conditions  of expected system performance "well enough  to  be able to
 measure how well the thing works and thence,  if  possible, to get an estimate  of
 value before the design is committed	the degree  of  stress that  a system can
 cope with is a measure of its  performance."(31)   This would be a considerable
 departure, from conventional evaluation where,  for example,  pipe  sizes are  selected
 on the basis of negligible hydraulic stress  (no  surcharge,  flowing  full:   no
 flooding).   Concerning performance,  it is important  to realize that the adoption  of
 a  particular design  frequency  imputes  acceptance of  a particular risk of system
 capacity or water quality level being exceeded.   Thus, employment of a  "design storm"
 freezes the level of risk,  and  very  little  information is available to  assess the
 reality of  presumed  levels  of  risk in a  cost-benefit  context.  Unfortunately, little
 data is available even in the U.S.  for making  suitable benefit analyses.(32")

 Conclusion

          Current storm drainage analysis procedures  too often emphasize the use  of
 a  synthetic  storm, which  is  a device for facilitating analysis but  at the  expense of
 reliability.   For small projects such  an approach is  useful  and can be  appropriate
 in helping  to  define marginal costs  among alternatives, but  may be acceptable only
 when gross  differences  in levels of  protection from  flooding or pollution  are sought.
 However,  for more important  projects,  local officials  should make reference to
 actual  rainfall histories  for the  planning, design and operation of new facilities.
 This  recommendation  is  particularly  appropriate where detention storage is involved
 and  even more  so when detention storage  is  interconnected as  in schemes for reduction
 of combined sewer overflows.  For  the  latter, and especially where outflow from
 storage would be  governed by treatment plant capacities, dewatering may take a
number of hours or days.

        : When  flooding or pollution occur  from system overloading,  local officials
 should have a  firm rainfall data base on which to defend themselves  against
criticism and  liability,  and the use of synthetic storms would leave them all too
vulnerable.  Similarly, they should be encouraged to make some field rainfall-runoff-
quality measurements to sharpen the validity of whatever analytical  methods and
models are used.  Lastly, crudities used in preliminary planning should not be
carried into advanced stages of design or into the operation of complex interacting
systems.
                                       - 116 -

-------
     Bauer,  W.  J.,  "Economics  of  Urban Drainage Design," J.Hyd.Djv.. ASCE Proc.,
     Vol. 88,  No.  HY6,  pp.  93-114,  November,  1962.
 2.   Poertner,  Herbert  G.,  Practices  in  Detention of  Stormwater  Runoff. APWA
     Special Report No. 43,  Chicago,  Illinois, 231  pp., June,  1974.  (Available
     from NTIS  as  PB 234 554).
 3.   Rice, Leonard, "Reduction of Urban  Peak  Flows  by Ponding,"  J.Irrig. and Drain
     Div., ASCE Proc.,  Vol.  97, No. IR3, pp.  469-482, September,  1971.
 4.   Michel, H. L., and W.  P.  Henry,  "Fairfax County  Works  to  Solve  Runoff  Problems,"
     Water & Wastes Engineering,  Vol. 13,  No. 9, pp.  38-42,  September,  1976.
 5.   Field,  Richard and John A. Lager, "Urban Runoff  Pollution Control  — State-of-
     the-Art," J.Envir.Engrg.Djv..  ASCE  Proc., Vol. 101, No. EE1, pp.  107-125,
     February,  1975. Authors' closure  to discussion:  Vol.  102,  No. EE4, p. 863,
     August, 1976.
 6.   Poertner,  Herbert  G.,  "Better  Storm Drainage Facilities —At Lower Cost,"
     Civil Engineering, Vol. 43,  No. 10, pp.  67-70, October, 1973.

 7.   Grigg,  Neil S., and John  W.  Labadie,  "Computing the Big Picture,"  Water &
     Wastes Engineering, Vol.  12, No. 5, pp.  37-39  & 86, May,  1975.

 g^   Design and Construction of Sanitary and  Storm Sewers,  ASCE Manuals and Reports
     of Engineering Practice No.  37 (WPCF Manual of Practice No. 9), ASCE,  New  York,
     N.Y., 332 pp., 1969.
 9.   McPherson, M. B.,  "Some Notes  on the Rational  Method  of Storm Drain Design,"
     ASCE Urban Water Resources Research Program Technical Memorandum No.  6, ASCE,
     New York, N.Y., 74 pp., January 22, 1969.   (Available  from NTIS as PB 184  701).

10.   Yarnell, D. L., Rainfall  Frequency-Intensity  Data. U.S. Department of Agriculture,
     Miscellaneous Publication 204, 1935.

11   Hathaway, G. A., "Design of Drainage Facilities," Transactions. ASCE,  pp.  697-
  *   730, Vol. 110, 1945.
12.   Hershfield, David M., Rainfall Frequency Atlas of the United States,  Weather
     Bureau, U.S. Department of Commerce, Technical Paper No. 40, May, 1961.

13.   Chow, Ven Te, Frequency Analysis of Hydrologic Data With Special Application to
     Rainfall Intensities. University of Illinois  Engineering Experiment Station
     Bulletin No. 414, Urbana, 80  pp.,  1953.

14   Keifer, Clint J., and Henry Hsien  Chu,  "Synthetic Storm Pattern for Drainage
     Design," J.Hyd.Div.. ASCE Proc., Vol. 83, No. HY4, pp. 1-25, August, 1957.

15.   McPherson, M. B., "Discussion of Synthetic Storm Pattern for Drainage Design "
     J.Hvd.Div.. A.SCE Proc., Vol.  84, No. HY1, pp. 49-57, February, 1958.        '

16.   Tholin, A. L., and C. J. Keifer, "The Hydrology of Urban Runoff," Transactions
     ASCE, Vol. 115, pp. 1308-1379,  1960.                              	L-ai>

17.  Pikarsky, Milton, and Clint J. Keifer,  "Underflow Sewers for Chicago " Civil
     Enp-lneering. ASCE, Vol. 37, No. 5, pp.  62-65,  May, 1967.               	

18.   Flood Control Coordinating Committee, "The Chicago Underflow Plan - Development
     of a Flood and Pollution Control Plan for the Chicagoland Area,"  State of
     Illinois, County of Cook, Metropolitan  Sanitary District of Greater Chicago
     City of Chicago,  29 pp.,  December, 1972.                                    *
                                       - 117 -

-------
 19.  Bureau of Engineering, City of Chicago, "Computer Simulation Programs,
      Development of a Flood and Pollution Control Plan for the Chicagoland Area "
      Technical Report No. 2, 101 pp. plus 37 exhibits, December,  1972.   (Available
      from NTIS as PB 236 645).

 20,  Terstriep, M. L., and J. B. Stall, The Illinois Urban Drainage Area Sjmulator,
      ILLUDAS.  Bulletin 58, Illinois State Water Survey,  Urbana, 90 pp.,  1974""*"

 21.  Terstriep, M. L., and J. B. Stall, "Urban Runoff by Road Research Laboratory
      Method,"  J.Hyd.Div.. ASCE Proc.,  Vol. 95, No.  HY6,  pp. 1809-1834, November,
      1969.  Authors' closure to discussion:   Vol. 97, No. HY4,  pp. 574-579 April
      1971.

 22.  Veres,  James S., "Testing the Design Storm Concept  on Oakdale Basin," a paper
      presented to Professor J. W.  Delleur in partial fulfillment  of the  course
      requirement for CE 697, Department of Civil Engineering,  Purdue University
      West Lafayette, Indiana, 87 pp.,  August 8,  1975.

 23.  New England Division, "Wastewater Engineering  and Management Plan for Boston
      Harbor  —  Eastern Massachusetts Metropolitan Area,"  Technical Data Vol. 8,
      "Urban  Stormwater Management," Department of the Army,  Corps of Engineers
      Waltham,  Mass., 210 pp., October,  1975,

 24.  Roesner,  L.  A., et al., Water Resources  Engineers,  The Hydrologic Engineering
      Center/Corps of Engineers,  and Department of Public Works/City and  County  of
      San Francisco,  "A Model for Evaluating  Runoff-Quality in Metropolitan Master
      Planning," ASCE Urban Water Resources Research Program Technical Memorandum
      No.  23, New  York,  N.Y.., 73  pp., April,  1974.   (Available from NTIS  as
      PB  234  312 or from the H.E.C.,  Davis, Cal.  95616).

 25.  "Storage,  Treatment,  Overflow,  Runoff Model (STORM),  Users'  Manual,"  Computer
      Program 723-S8-L7520,  The Hydrologic  Engineering Center, Corps  of Engineers
      U.S.  Army, 609  Second Street,  Davis, California 95616,  170 pp.,  July, 1976.'

 26.   Shubinski, Robert  P.,  and William  N.  Fitch,  "Urbanization  and  Flooding, an
      Example,"  pp. 69-73  in  Environmental Modeling  and Simulation.  Office  of Research
      and Development and Office  of  Planning and  Management,  U.S.  Environmental
      Protection Agency  report  EPA 600/9-76-016,  847  pp., July,  1976.  (Available
      from  NTIS as  PB 257 142).

 27.   Department of Public  Works, "San Francisco  Master Plan  for Waste Water
      Management," City  and County of San Francisco,  September 15,  1971.  (Three
      volumes plus  an appendix).

 28.  Department of Public Works, "San Francisco  Master Plan  for Waste Water Management
     Supplement I,"  City and County of  San Francisco,  88 pp., May  15, 1973.            '

 29.   McPherson, M. B.,  "Innovation:  A  Case Study," ASCE Urban  Water Resources
     Research Program,  Technical Memorandum No.  21, ASCE, New York,  N.Y.   59
     February,  1974.  (Available from NTIS as PB  232  166).             *'     P**

 30.  Johnson, Steven L., and Douglas M. Sayre, "Effects of Urbanization on Floods
      in the Houston, Texas,  Metropolitan Area," U.S. Geological Survey Water-Res
     Investigations  3-73, 50 pp., April, 1973.  (Available from NTIS  as PB 220     CeS

31.  Green, J., A. King and  K. Bowden,   "Economics of Sewerage Design  " Local
     Government Operational  Research Unit, Royal Institute of Public'Administrati
     Report No. C218, 201 Kings Road, Reading, Berkshire, U.K., 77 pp.,  April,
     1975.

32.  Grigg, Neil S., and Otto J. Helwig, "State-of-the-Art of Estimating Flood
     Damage in Urban Areas," Water Resources Bulletin. Vol. 11,  No. 2  DD
     April, 1975.                                                     ' VVt

                                        - 118  -

-------
               ADDENDUM 3
NOMOGRAPHS FOR TEN-MINUTE UNIT HYDRDGRAPHS
       FOR SMOL URBAN WATERSHEDS
ASCE Urban Water Resources Research Program
        Technical Memorandum No. 32
                    by
William H. Espey, Jr., and Duke G. Altman
    Espey, Huston & Associates, Inc.
                   and
         Charles B. Graves, Jr.
             City of Austin
              Austin, Texas
              December,  1977
     American Society of Civil Engineers
            345 East 47th Street
            New York, N.Y. 10017
                        - 119 -

-------
                                      PREFACE

                                by M.  B.  McPherson

 Background

           The following technical memorandum is Addendum 3 of a recent ASCE
 Program report on "Urban Runoff Control  Planning11.'1-)  Addendum 1, "Metropolitan
 Inventories," and Addendum 2,  "The Design Storm Concept," were appended to the
 latter report.  The present technical memorandum  is  the first of several  that
 will  contain additional,  individual Addenda  over  the period 1977-1979.

           The principal intended audience of the  recent ASCE Program report was
 the agencies and their  agents  that are participating in the preparation of
 areawide plans for water pollution abatement management pursuant to Section 208
 of the Federal Water Pollution Control Act Amendments of 1972 (P.L. 92-500).
 While the presentation  which follows  is  also directed to areawide agencies and
 their agents,  it is expected that it  will be of interest and use to many  others
 particularly local governments.                                                '

 ASCE  Program

           The  American  Society of Civil  Engineers' Urban Water Resources  Research
 Program was  initiated and  developed by the ASCE Urban Water Resources Research
 Council  (formerly  the Urban  Hydrology Research Council).  The basic purpose of the
 Program is to  help establish coordinated long-range  research in urban water
 resources  on a national scale.

          Abstracts of  the twenty-eight reports and  technical memoranda of
 Program for  the  1967-1974  period  are  included in a readily available paper.
 The two  reports  and the six  technical  memoranda of the regular series completed
 since are  identified in a  recent  publication.^)  Also included in the latter is
 a.  listing of all but one of  the  twelve national reports in the special technical
 memorandum series  for the  International Hydrological Programme; and the last
national report(^) and an  international summary^) have been released since.

          A Steering Committee designated by  the ASCE Council gives general
direction to the Program:  S. W. Jens  (Chairman); W, C. Ackermann; J. C. Geyer-
C. P.  Izzard; D. E. Jones, Jr.; and L. S. Tucker.  M. B. McPherson is Program '
Director (23 Watson Street, Marblehead, Mass. 01945).  Administrative support is
provided by ASCE Headquarters  in New York City.

Unit Hydrographs

          Unit hydrographs have been developed from rainfall-runoff field data
for completely sewered drainage catchments in Louisville,  Kentucky,(6) and Atla
Georgia.(7)  However, there is considerable doubt over the transferability or    *
universality of  the findings from a single jurisdiction.  Moreover,  most of the
unit hydrographs developed across the nation have been for partially sewered
catchments where the streamflows measured had included a significant contributio
 from non-sewered sectors.  An example  is  the testing of unit hydrographs using  **
urban streamflow data from large catchments  nearby by the  Georgia Institute of
Technology.(8)
                                      - 120 -

-------
          Dr. William H.  Espey,  Jr.,  the senior author of the following Technical
Memorandum, pioneered in the regionalization of unit hydrographs  (the regional
synthesis of unit hydrograph characteristics) for urban streams.   His initial work
dealt mainly with streamflow data for urban watersheds located in Texas, Kentucky,
Illinois, Indiana and Ohio. (9)  This  initial unit hydrograph analysis was
subsequently expanded utilizing the USGS data for Houston, Texas. ^10)  His  first
syntheses of characteristics were for thirty-minute duration unit hydrographs,
suitable only for direct application to larger-sized urban watersheds.  In  contrast,
the characteristics reported herein are for ten-minute duration unit hydrographs.
The synthesized equations and associated nomographs are derived from data for
forty-one urban watersheds of which eighteen are in Texas.  Flow gaging in  nearly
all instances was in streams, a limitation discussed below.

          Unit hydrographs have been employed in investigations of the effects  of
urbanization or in the improvement of prediction techniques. (H~16)  Field data
from urban catchments in Indiana and elsewhere have been used in tests of various
linear process methods, including "instantaneous" unit hydrographs.' I''  A series
of papers has been devoted to the development of unit hydrographs for gaged urban
Local Validation

          Users of the equations herein, or of the nomographs representing them,
are very  strongly urged  to check their local validity by deriving as many unit
hydrographs  as  possible  using rainfall and runoff data from local catchments for
comparison.  This is  particularly  important where applications will be for wholly
sewered catchments, a condition not truly represented by the formulations
presented.   One's first  reaction might be that if local unit hydrographs can be
derived   why do we need  the nomographs?  What the nomographs provide is a
generality of  characteristics so that local unit hydrographs for particular
catchment physical characteristics can more reliably be extrapolated as part of
a family  of  characteristics for application in catchments where the physical
characteristics differ.  Also, the fact  that the nomographs are founded on data
from around  the country  should reinforce the credibility of any indications
obtained  using local  data.

          An excellent manual on unit hydrograph analysis  is readily available^ )
together  with  a computer program user's  manual^")  for development of unit
hydrographs  from  field data and for  routing flows  from one point to another.  The
computer  program  has  been used extensively  in urban projects of the Corps of
Engineers.   Alternative  methods for  deriving unit  hydrographs  (linear programming
versus  least squares) have been compared. (2*)

          As is the  case for  some  of the newer  runoff  planning tools, access  to  a
digital computer  is  not  required  for the use of the information in  this  report.
 The user can calculate synthetic  unit hydrograph parameters  directly  from the
 eauations with a  pocket  calculator of modest  capability or simply use the
 nomographs.   Mareover,  complete  program listings have  been published  for use of
 eleven different  small programmable calculators for:  deriving a unit hydrograph
 from an observed  hydrograph;  converting a unit  hydrograph of a given duration to
 its equivalent for another duration; and synthesizing a hydrograph using a given
 rainfall pattern and a unit hydrograph. W2-* Also included among the 40 complete
 program  listings are three different methods  for routing hydrographs from one point
 to another.  The cost of the calculators for which programs are listed span
 essentially the full price range.   A sample of the programs available,  but for
 water distribution system analysis,  is provided in a technical magazine.' "'
                                         - 121  -

-------
          We are not advocating ,the use of unit hydrographs as a replacement for
more versatile or comprehensive tools of analysis but as a supplementary element
among the range of tools needed in the development of urban runoff control plans.
For example, use of locally validated synthetic unit hydrographs based on
indicators from the following report would be an improvement over employment of
hypothetical triangular unit hydrographs as descriptors of subcatchment inputs in
some of the more comprehensive runoff models.

Validation Example

          The Colorado Urban Hydrograph Procedure (CUHP) has been used extensively
in the design of flood abatement facilities throughout the Denver metropolitan area
When the CUHP was introduced in 1969 it was founded on a very small amount of
field data.  On the basis of much more subsequent data, the Procedure was modified
in 1975(2^' and again in 1977.^25'  The Executive Director of the Urban Drainage
and Flood Control District of the Denver metropolitan area, Mr.  L. Scott Tucker
was asked to check the equations in the following technical memorandum against
local data.  (Mr. Tucker is also a member of the ASCE Program's  Steering Committee
and Chairman of the ASCE Council).  Mr. Ben Urbonas of his staff reported their
findings,' °' summarized below:
Stream Gaging Station



Area, A, mi2
L, mi
S, ft/ft
I, %
4
Hydrographs checked



Hydrograph
CD
(2)
(3)
(4)
(5)
(6)
(7)
S. Platte R.
Tributary
at Englewood
1.03
1.93
0.0076
46
0.8
(1) 8/19/71 (3)
(2) 8/24/72

Peak Discharge
Via This Tech. Memo.
185
204
99
57
100
218
170
S. Platte R.
Tributary
at Denver
0.75 0.51
2.02 1.56
0.007 0.006
42 60
1.0 0.8
8/7/73 (4) 7/21/76


, cfs, Generated and
Via CUHP
164
180
93
53
105
223
176
Sand Creek
Tributary
at Denver
0.270
0.80
0.0069
43
0.6
(5) 7/24/73
(6) 8/7/73
(7) 7/30/74
Actual
Recorded
122
122
59
36
104
252
250
                                        - 122  -

-------
          For the South Platte River Tributary at Denver,  rather high values  of 
were selected to account for induced surface storage expected in that particular
case whenever storm sewer capacity is exceeded.  (Considerable care should be
exercised in the selection of ^-values because of their profound influence on the
magnitude of unit hydrograph peak discharge and the time to the peak discharge).

          No special attempt was made to modify the originally selected values of
J, to force a better agreement between observed and generated peak discharge for any
of the above cases.  The tests reported above were for different streams than the
two in the Denver area from which data was used in the development of the
nomographs herein.

          For the Sand Creek Tributary at Denver, the "Hydrograph (6)" storm of
8/7/73 had a maximum average 15-minute intensity close to the once in 100 years
on the average historical value.  The 100-year design peak flow would be about
335-cfs  for this location, using the CUHP, only about one-third higher than the
peak flow recorded for storm "Hydrograph (6)".

          Reiterating the point made-earlier, users of this Technical Memorandum
are emphatically urged to check the  local validity of the relationships presented
by deriving as many unit hydrographs as possible using rainfall and runoff data
from local catchments for comparison.  The exercise of such care will not
necessarily be easy, because there is an inherent difficulty with any runoff model
in the necessarily subjective  separation of abstractions (infiltration, depression
storage, etc.)  from total rainfall to resolve  rainfall excess  (amount and  pattern),
which  is the  input from which  an equal volume  of direct runoff is generated by
models of one kind or another.  After analyzing  the performance of a variety .of
models,  it was  concluded  that  the weakest  link is the proper  estimation of
rainfall excess/27'

Acknowledgments

          The empirical  hydrograph  equations  presented  in  this Technical  Memorandum
were developed  as  part of the  City  of Austin  Master Drainage  Study under  the
direction of  Charles B.  Graves,  Jr., Director of Engineering.   Pursuant  to this
study,  a drainage  criteria  manual was written which contains  design  discharge
curves  to be  used  in hydrologic studies  in and near the Austin,  Texas,  area.
These  curves  were  developed utilizing synthetic  unit  hydrographs  determined  from
the  empirical equations  along with  local  rainfall  data  for application in the
Austin area.

          The ASCE Urban Water Resources  Research Council  is indebted to the
 authors for their generous contribution of this  report as  a public service.

           Processing,  duplication and distribution of this Technical Memorandum
 was supported by grants to ASCE from the Research Applied to National Needs  program
 of the National Science Foundation.  Technical liaison representative for NSF/RANN
 is Dr. J.  Eleonora Sabadell.  Any opinions, findings, and conclusions or
 recommendations expressed herein are those of the contributing authors or the
 writer and do not necessarily reflect the views  of the National Science Foundation.

 New References

           To the references of Section 4 of the parent report^' should be added
 a useful document containing brief  summaries of 18 selected publications dealing
 with nonstructural measures for reduction of flood losses.^ °'
                                          - 123 -

-------
           In the report' ' to which this Technical Memorandum is Addendum 3
 reference 55 on page 72 states that the EPA "Areawide Assessment Procedures
 Manual" is in two volumes, whereas a third volume on "Best Management Practices"
 has since been released.  Also, reference 99 on page 75 is available from NTIS
 as PB 239 333, and reference 221 on page 84 is available from NTIS  as PB 257  089
 (July, 1976).
           Reference 236 on page 85 of the parent report    has  been published in
 the Journal of the Water Resources Planning and Management Division. ASCE
 Proceedings, Vol. 103,  No. WR2, pp. 177-192,  November,  1977.  Other papers  in the
 same issue that are related to the subject of urban runoff control  planning are
 on:  "Urban Flood Management:   Problems and Research Needs,"  by D.  H.  Howells
 pp. 193-212; "Innovative Management Concept for 208 Planning,"  by J. w. Bulkley
 and T.  A.  Gross,  pp.  227-240;  and "Flood Management for Small Urban Streams "
 by W.  Whipple, Jr.,  pp. 315-324.   Overlooked  in a previous issue was a reference
 on control of sedimentation via detention basins. (29)

           The report  of a nationwide evaluation of combined sewer overflows and
 urban  stormwater  discharges (including the cost of control or abatement of
 receiving  water pollution from such sources)  has been released  in three volumes
 The first  volume  is an  executive  summary. C3")  Assessments were for all 248 "Urba
 Areas"  in  the US., of  which about 46 per cent is undeveloped.   Land uses for th **
 54 per  cent that  is developed  are approximately as  follows:   residential, 58%-
 industrial,  15%;  commercial, 9%;  and other, 18%.  Also,  about 14% of the 'developed
 urban  land is served  by combined  sewers and 38% by  separate storm sewers, with
 the balance containing  unsewered  storm drainage.   "Average annual dry-weather
 flow is  significantly greater  than average wet-weather  flow only in the arid  are
 However, in most  parts  of the  country,  dry-weather  flows represent  30  to 50 per  8*
 cent of  the total  (wet  plus  dry)  runoff from  urban  areas. "OO)

           Mr.  Carl F. Izzard,  member of the ASCE  Program Steering Committee  ha
 re-analyzed  the original  experimental data for  full-size street  inlets and
 developed  a  graphical solution for the hydraulic  design  of curb-opening inlets  (31}

          An  overview and  assessment  of current catchbasin technology has been
 reported recently/32''   Included  are  evaluations  of hydraulic and pollutant
 removal efficiencies.

          To  the  transport module  of  the version  of the Storm Water Management
 Model embodying the fundamental hydrodynamic  equations of motion has been added
 capability for analyzing alternatives  for  the abatement of deposition and scour
 storm and combined sewers.  The project report'33'  includes listings of the new  *
 computer program subroutines that have  been added for solids transport
 characterization.  A related reportv3^' describes procedures for estimating dr
 weather pollutant deposition in combined sewers.                              ^~
          Four papers presented at a recent national meeting, under the auspice

single document.' •"'
of the ASCE Urban Water Resources Research Council,  have been preprinted in SL
          A critique of traditional land-use control mechanisms is included tn
report on economic incentives for such control.(3°)  An extensive bibliograoh  *
occupies almost a third of the content.

          National state-of-the-art indications on urban stormwater runoff co
possibilities have been blended with local  information in a comprehensive ret>o   °^
for the metropolitan Milwaukee region.'3''
                                        - 124 -

-------
References

 1.  McPherson, M. B., "Urban Runoff Control Planning," ASCE,  New York,  N.Y.,
     118 pp., June, 1977.  (Available as PB 271 548,  at $5.50  per copy,  from the
     National Technical Information Service, 5285 Port Royal Road, Springfield,
     Virginia 22161).

 2.  McPherson, M. B., and G. F. Mangan, Jr., "ASCE Urban Water Resources Research
     Program," J.Hyd.Div.. ASCE Proc., Vol. 101, No.  HY7, pp.  847-855, July,  1975.

 3.  McPherson, M, B., and G. F. Mangan, Jr., Closure to Discussion of "ASCE Urban
     Water Resources Research Program," J.Hyd.Div.. ASCE Proc., Vol. 103, No. HY6,
     pp. 661-663, May, 1977.

 4.  Ramaseshan, S., and P. B. S. Sarma, "Urban Hydro logical Modeling and Catchment
     Research  in  India," ASCE UWRR Program Technical Memorandum No. IHP-12, ASCE,
     New York, N.Y., 21 pp., May, 1977.  (NTIS No. PB 271 300).

 5.  McPherson, M. B., and F. C. Zuidema, "Urban Hydrological Modeling and
     Catchment Research:  International Summary," ASCE UWRR Program Technical
     Memorandum No.  IHP-13, ASCE, New York, N.Y., 48 pp., November, 1977.

 6  Eagleson, Peter S.,  "Unit Hydrograph Characteristics for Sewered Areas,"
     J.Hvd.Div*. ASCE  Proc., Vol. 88, No. HY2,  pp. 1-25, March,  1962.  Author's
     closure to discussion:  Vol. 89, No. HY4,  pp. 193-203, July,  1963.

 7  Black,  Crow  and Eidness, Inc.,  Storm andjCpmbined  Sewer  Pollution  Sources  and
     Abatement. Atlanta,  Georgia, Water Pollution Control Research Series  11024,
     ELB  01/71, GPO, Washington, D.C.,  181  pp.,  January, 1971.

 g.  Wallace,  James  R.,  "The Effects of Land Use Change on  the  Hydrology of  an
     Urban Watershed," Georgia  Institute of Technology Report ERC-0871,  Atlanta,
     Georgia,  66  pp.,  October,  1971.  (NTIS No.  PB  206 426).

 9  Espey,  W. H., Jr.,  C.  W. Morgan and F. D. Masch,  "A Study of Some  Effects  of
     Urbanization on Storm Runoff  from a Small Watershed,"  Center for Research  in
     Water Resources, Technical Report No.  HYD07-6501/CRWR-2, Department of  Civil
      Engineering, The University of Texas,  Austin,  109 pp., July, 1965.

 10   Espey, W. H., Jr.,  and D.  E.  Winslow,  "The Effects of  Urbanization on Unit
     Hydrographs for Small Watersheds, Houston, Texas, 1964-1967," TRACOR Document
     No.  68-975-U, Austin,  Texas,  70 pp.,  September 25, 1968.  (NTIS  No. PB 183 049),
      ("Appendices — Data Compilations," TRACOR Document No. 68-1006-U,  244 pp.,
      September 25, 1968. NTIS  No.  PB 183  050).

 11   Dracup, John A., Thomas J. Fogarty and Sharon G. Grant,  "Synthesis aud
      Evaluation of Urban-Regional  Hydrologic Rainfall-Runoff Criteria,"
      Environmental Dynamics, Inc., Los Angeles, California, 105 pp.,  February,
      1973.  (NTIS No. PB 220 965).

 12   Johnson, Steven L., and Douglas M. Sayre, "Effects of Urbanization on Floods
      in the Houston, Texas, Metropolitan Area," U.S. Geological Survey Water-
      Resources Investigations 3-73, Austin, Texas, 50 pp., April, 1973.   (NTIS No.
      PB 220 751).
                                        - 125 -

-------
 13.   Hamra,  D.  W.,  C.  W.  Morgan  and H. A. Reeder, "Statistical Analysis of
      Hydrograph Characteristics for  Small Urban Watersheds," Tracer, Inc., Report
      No.  T73-AU-9559-U,  Austin, Texas,  155 pp., October, 1973.
      (NTIS  No.  PB  228 131).

 14.   Chien, Jong-Song, "Urban Runoff by Linearized Subhydrograph Method,"
      J.Hyd.Div.. ASCE Proc., Vol. 100,  No. HY8, pp. 1141-1157, August, 1974.

 15.   Schulz, E.  F., and  0. G. Lopez, "Determination of Urban Watershed Response
      Time," Colorado  State University,  Hydrology Paper 71. Fort Collins, Colorado
      41 pp., December, 1974.                                                     '

 16.   Brater, Ernest F.,  and James D. Sherrill, Rainfall-Runoff Relations on Urban
      and  Rural Areas.  Environmental  Protection Technology Series EPA-670/2-75-04"6
      GPO, Washington,  D.C., 98  pp.,  May, 1975.                                   '

 17.   Rao, A. R. , J. W. Delleur  and P. B. S. Sarma,  "Conceptual Hydrologic Models
      for  Urbanizing Basins," J.Hyd.Div., ASCE Proc., Vol. 98, No. HY7, pp. 1205-
      1220, July, 1972.

 18.   Stubchar, James M. ,  "The Santa Barbara Urban Hydrograph Method," pp. 131-141.
      Hann, C. T., "Comparison of Methods for Developing Urban Runoff Hydrographs "
      pp.  143-148; and Delleur,  J. W. , A. R.  Rao and H. Hassain, "On Modeling the'
      Runoff Process in Urban Areas," pp. 193-208:   Proceedings. National Symposium
      on Urban Hydrology and Sediment Control, July 28-31, 1975, Office of Research
      and  Engineering Services report UKY BU 109,  University of Kentucky, Lexington
      314  pp., November, 1975.                                                     '

 19.  Hydrologic Engineering Center,  Corps of Engineers, "Hydrologic Engineering
     Methods for Water Resources Development:  Volume 4,  Hydrograph Analysis "
     Publication HEC-IHD-400, Davis, California,  122 pp., October, 1973.  (NTIS
     No. AD 774 261).

20.  Hydrologic Engineering Center,  Corps of Engineers, "HEC-1, Flood Hydrograph
     Package," Generalized Computer Program 723-010, 609  2nd Street, Davis,
     California 95616, 186 pp.,  January, 1973.

21.  Singh,  Krishnan P.,  "Unit  Hydrographs —A Comparative Study," Water
     Bulletin.  Vol. 12, No. 2,  pp. 381-392,  April,  1976.
22.  Croley,  Thomas E., Hydrologic and Hydraulic Computations on Small ProgrammaM*.
     Calculators.  Iowa Institute of Hydraulic Research,  The University of Iowa -- "
     Iowa City, 837 pp., 1977.  ($15.95).                                     '

23.  Croley,  Thomas E., "Hydraulic Computations for Small Programmable Calculators "
     Water & Sewage Works.  Vol.  124, No.  11,  pp. 64-71,  November,  1977.          S*

24.  Schulz,  E. F., Ben Urbonas  and Bill  DeGroot,  "improved Version of the Colorad
     Urban Hydrograph Procedure," Flood Hazard News (Urban Drainage and Flood     °
     Control District, Denver),  Vol. 5, No.  2, pp.  1-2,  August,  1975.

25.  Urbonas, Ben, "Colorado Urban Hydrograph Procedure  Revisions," Flood Hazard
     News (Urban Drainage and Flood Control  District,  Denver),  Vol. 7   ito — T -
     p. 7, June, 1977.                                               '   *  »
                                      - 126 -

-------
26.  Urbonas, Ben, Urban Drainage and Flood Control District,  Suite 156B,  2480 West
     26th Avenue, Denver, Colorado 80211,  letter and enclosure dated July 1,  1977,
     to M. B. McPherson, 32 pp.

27.  Hossain, A., J. W. Delleur and A. R.  Rao, "Evaporation, Infiltration and
     Rainfall -Runoff Processes in Urban Watersheds," Water Resources Center
     Technical Report No. 41, Purdue University, West Lafayette, Indiana, 63 pp.,
     January, 1974.  (NTIS No. PB 229 642).

28   Hydrologic  Engineering Center, UoS. Army Corps of Engineers, "Annotations of
     Selected Literature on Nonstructural Flood Plain Management Studies," Davis,
     California, GPO No. 1977-789-006/7333, 95 pp., March, 1977.

29   Curtis, David C.,  and Richard H. McCuen, "Design Efficiency of Stormwater
     Detention Basins," J^Wat.Resour.Plan.k Mgt.Piv.. ASCE Proc., Vol. 103,
     No. WR1, pp.  125-140, May,  1977.

r»n   American Public Works Association and University of Florida, Nationwide
     Evaluation  of Combined Sewer Overflows and Urban Stormwater Discharges,
     Volume  I:   Executive Summary. Environmental Protection Technology Series
     EPA-600/2-77-064a, Municipal Environmental Research Laboratory, U.S.  EPA,
     Cincinnati,  Ohio  45268,  95  pp.,  September, 1977.  (NTIS No. PB 273  133).

31   Izzard, Carl F.,  "Hydraulic Design of Curb-Opening Inlets," Flood Hazard
     News  (Urban Drainage and Flood  Control District, Denver),  Vol.  7, No.  1,
     "p~pT~8-ll,  June,  1977.
 •*2   Metcalf and Eddy,  Inc.,  Palo Alto,  California,  Catchbasin Technology Overview
      and Assessment,  Environmental Protection Technology Series EPA-t>00/2-77-051,
      Municipal Environmental  Research Laboratory,  U.S.  EPA,  Cincinnati,  Ohio
      45268,  128 pp.,  May,  1977.   (NTIS No.  PB 270  092).

 33.   Sonnen, Michael  B.,  "Abatement of Deposition  and Scour in Sewers,"  Final
      Report to U.S. EPA Storm and Combined  Sewer Section, Water Resources
      Engineers, Inc., 710 South Broadway, Walnut Creek, California 94596, 114 pp.,
      June 1, 1977.

 •>4   Energy and Environmental Analysis,  Inc., Boston, Massachusetts, Procedures
   *   for Estimating Dry Weather Pollutant Deposition in Sewerage Systems.
      Environmental Protection Technology Series EPA-600/2-77-120, Municipal
      Environmental Research Laboratory,  U.S. EPA,  Cincinnati, Ohio 45268, 92 pp.,
      July, 1977.  (NTIS No. PB  270  695).

  _   Lytiam, B. T., F. C. Neil and F. E.  Dalton, "Managing Storm Runoff  in the
 ^    Chicagoland Area," pp. 1-34; W. B.  Whipple, Jr., "Urban Runoff Pollution and
      Water Quality Planning," pp. 35-59; B. Urbonas and L. S. Tucker, "Stormwater
      Quality - What  is the Problem?," pp. 60-89; and P. G. Collins and  L. A.
      Roesner,  "Data Analysis for Nonpoint Pollution Control,"  pp. 90-111:   "Urban
      Runoff Quality - Measurement and Analysis," ASCE October  17-21, 1977,  Fall
      Convention and  Exhibit, San Francisco, California, ASCE Preprint No. 3091,
      New York, N.Y.  ($4.00).

  f.   CONSAD Research Corp., Pittsburgh, Pennsylvania,  Economic Incentives  for
      Land Use Control, U.S. EPA Office  of Research  and  Development,  Report
      EPA-600/5-77-001, Washington,  D.C. 20460,  359  pp.,  February,  1977.  (NTIS
      No. PB 265 468).
                                       - 127 -

-------
37.  Southeastern Wisconsin Regional Planning Commission,  "State of the Art of
     Water Pollution Control in Southeastern Wisconsin,  Volume 3,  Urban Storm
     Water Runoff," SEWRPC Technical Report No. 18,  P.O. Box 769,  Waukesha
     Wisconsin 53186, 63 pp., July, 1977.  ($5.00).                        '
                                   - 128 -

-------
 NOMOGRAPHS  FOR TEN-MINUTE UNIT HYDROGRAPHS
         FOR SMALL URBAN WATERSHEDS
(Addendum 3 of the ASCE Program report  "Urban
    Runoff Control Planning,"  June,  1977)
                     by
    William H. Espey, Jr., President, and
       Duke G. Altraan, Staff Engineer,
      Espey, Huston & Associates, Inc.,
                3010 S. Lamar
             Austin, Texas 78704
                     and
           Charles B. Graves, Jr.,
          Director of Engineering,
               City of Austin,
               P.O. Box 1088,
             Austin, Texas 78704
                December,  1977
                  - 129 -

-------
                                TABLE OF CONTENTS

                                                                             Pa^e

Introduction ...... 	 ...... ...... •  «••••«..  131

     Figure 1 - Definition of Unit Hydrograph Parameters „ „	^33

Watershed Data	.	134

     Table 1  - Location and Physiographic Description of Watersheds  .  „  „  .  135

     Figure 2 - Watershed Conveyance Factor, ^, As a Function of
                Per Cent Watershed Impervious Cover, I,  and Weighted
                Main Channel Manning 'n* Value	138

     Table 2  - Classification of Watershed Drainage Systems	140

Development of Empirical Ten-Minute Unit Hydrograph Equations  	  142

     Table 3  - Ten-Minute Unit Hydrograph Equations	144

Nomographs and Examples  ........ 	 . 0 	  143

     Figure 3 - Time of Rise (TR) Nomograph  ...oo....	145

     Figure 4 - Peak Discharge (Q) Nomograph .........  	  146

     Figure 5 - Time Base (Tg) Nomograph	  147

     Figure 6 - Hydrograph Width (W^Q)  Nomograph ..............  143

     Figure 7 - Hydrograph Width (Wy^)  Nomograph	•  .  .  149

     Figure 8 - Example of Ten-Minute Unit Hydrograph Construction .....  152

References	>..  151
                                  - 130 -

-------
                   NOM3GRAPHS FOR TEN-MINUTE UNIT HYDROGRAFttS
                           FOR SM&.LL URBAN WATERSHEDS

Introduction

     Due to rapid runoff response and the resulting flood potential characteristic

of many relatively small urban and surrounding area watersheds,  a need is evident

for adequate and efficient description of the dynamic process  involved.  In some

instances, procedures such as the "rational method" may be sufficient in

describing a hydrologic process in a small watershed but at other times a more

definitive method is required.  A short-duration unit hydrograph has considerable

potential for describing the dynamic runoff process of small watersheds.

     The basic theory of the unit hydrograph appears to have been suggested first

by Folse (1929).  The Boston Society of Civil Engineers (1930) stated  that "the

base of the  flood hydrograph appears  to be approximately constant  for  different
              i
floods, and  peak flow tends  to vary directly with  the total volume of  runoff".

Three  years  later,  in 1932,  Sherman  introduced the basic concept of  the  unit

hydrograph.  The unit hydrograph  has  found wide  acceptance as an outstanding

contribution to engineering  hydrology.  The  general  concept of the unit  hydrograph

has been  summarized as  follows  (Jforgan and Johnson,  1962):

      "for a  given drainage area,  the time-base of surface-runoff hydrographs
       resulting  from similar storms  of equal  duration are  the same
       regardless of the intensity of rainfall;

      "for a  given  drainage area,  the ordinates of the surface-runoff
       hydrographs  from similar storms of equal duration are proportional
       to  the volume of  surface runoff; and

      "for a  given  drainage area,  the time distribution  of surface runoff
       from  a particular storm period is  independent of  that  produced by
       any other  storm period."

 Because the unit hydrograph  is such a valuable hydrologic tool,  many investigations

 have been undertaken to develop synthetic unit hydrograph relationships based on

 watershed features.
                                    - 131 -

-------
     Because it is generally accepted that the best unit period should be

somewhat smaller than the watershed lag time (time from the center of mass of

rainfall to the peak of the hydrograph), the need for short-duration unit

hydrographs to describe the runoff process for small, quick-responding watersheds

becomes, obvious.  Empirical equations given subsequently were developed to aid

users in describing or synthesizing the shape of a ten-minute unit hydrograph for

any small watershed for which adequate rainfall-runoff relationships had not been

previously developed through use of site-specific data.

     The following hydrologic parameters were chosen to describe the shape of the

ten-minute unit hydrograph:

          T         -     the time of rise, in minutes;
           H
          Q         -     the peak discharge, in cfs;

          T_        -     the time base, in minutes;
           B
          W _       -     the time, in minutes, between the two points on
                          the unit hydrograph at which the discharge is
                          half of the peak discharge; and

          W         -     the time, in minutes, between the two points on
                          the unit hydrograph at which the discharge is
                          three-fourths of the peak discharge.

For clarification, these parameters are illustrated in Figure 1.  Once the five

parameters have been determined by use of the empirical equations, and-an

auxiliary means for distributing a runoff volume of one inch has been employed,

the ten-minute unit hydrograph can be easily constructed for a watershed.

Dimensionless unit hydrographs, such as those found in the Soil Conservation

Service hydrology manual (USDA-SCS, 1971), can be used as an aid in determining

the basic shape of the ascending and descending portions of the unit hydrograph.

Generally, the lower portions of the ascending and descending unit hydrograph

limbs can then be adjusted to obtain the one inch runoff volume.

     From the empirical equations, parallel, scaled nomographs have been developed

to provide a straightforward, simple procedure enabling graphical determination

by the user of the five hydrologic parameters.  The nomographs are featured in

this document.

                                  - 132 -

-------
UJ
o
OL
<

O
                                  TIME
       FIGURE  I - DEFINITION  OF  UNIT   HYOR06RAPH   PARAMETERS
                                -  133 -

-------
Watershed Data

     The five hydrologic parameters that describe the fundamental shape of the

ten-minute unit hydrograph were developed from rainfall-runoff data collected

for forty-one small watersheds throughout the United States.  Of the forty-one

watersheds, eighteen are located in Texas.  Listed in Table 1 are the name,

location, and physiographic characteristics of each of the forty-one watersheds

which have a range in size of drainage area between 0.014 and 15.0 square miles.

As indicated in Table 1, size of drainage area (A), main channel length (L)

main channel slope (S), extent of impervious cover (I), and a dimensionless

watershed conveyancy factor (
-------
                                                        TABLE 1
                                   LOCATION AND  PHYSIOGRAPHIC DESCRIPTION OF WATERSHEDS
Watershed Name
Bachman Branch
Hunting Bayou at
Cavalcade Street
Stoney Brook Street
Ditch
Sims Bayou at Carlsbad
Street
Cole Creek at Guhn Rd.
Bering Ditch at
Woodway Drive
Berry Bayou at Forest
Oaks Street
Berry Bayou at Gil pin
Street
Berry Bayou at Globe St.
Brickhouse Gully at
Clarblak Street
Halls Bayou at
Deertail Street
Keegans Bayou at
Keegan Road
Willow Waterhole
at Landsdown St.
Watershed Drainage Area, A
Location (Square miles)
Dallas, Texas 10.0
Houston, Texas 1.03
" 0.50
" 4.99
" 7.05
11 2.59
11 11.10
" 3.26
" 1.58
" 2.01
11 5.27
11 5.77
" 1.15
Main Channel
Length, L
(Feet)
28,512
5,800
3,700
12,400
18,500
11,400
18,300
4,500
7,900
12,700
20,600
31,200
1,700
Main Channel
Slope, S
(Ft. per Foot)
0.006
0.002
0.0006
0.0008
0.001
0.0007
0.0016
0.0015
0.0006
0.0011
0.0011
0.0005
0.0006
Impervious
Cover, I
(%)
30.0
27.0
33.0
4.0
4.0
17.0
13.0
9.0
15.0
2.5
2.0
2.0
33.0
Dimensionless
Conveyance
Factor, 
0.8
1.25
0.60*
1.30
1.30
0.85
0.65*
1.00
1.00
1.05
1.30
1.30
0.95
*Modified i selection process
                                                                                                    (Continued)

-------
TABLE 1 (Continued)
Watershed Name
Waller Creek at
38th Street
Waller Creek at
23rd Street
Helotes Creek at
State Hwy. 16
Brown Creek at
State Fairgrounds
Three Mile Creek
Crane Creek
Walton Run
Little Sugar Creek
at Brookcrest Dr.
Little Sugar Creek
at Burnley Road
Paw Creek at
Allenbrook Drive
Briar Creek at
Sudbury Road
Briar Creek at
Shamrock Drive
Irwin Creek
Tributary
Main Channel
Watershed Drainage Area, A Length, L
Location (Square miles) (Feet)
Austin, Texas 2.31
4.13
San Antonio, 15.00
Texas
Nashville, 11.80
Tenn.
Jackson, Miss. 1.10
0.45
Phila., Penn. 2.17
Charlotte, 0.84
N.C.
11 0.44
" 0.62
0.56
" 0.52
11 0.27
23,080
27,560
35,600
28,512
9,504
4,224
14,784
9,029
5,544
7,022
5,808
5,650
5,914
Main Channel
Slope, S
(Ft. per Foot)
0.009
0.009
0.01
0.0072
0.0079
0.0067
0.0068
0.0175
0.0157
0.0143
0.0112
0.0148
0.0193
Impervious
Cover, I
(7.)
27.0
37.0
6.0
15.7
19.7
27.5
24.7
21.0
14.0
18.0
16.0
20.0
19.0
Dimensionless
Conveyance
Factor, 
0.80
0.80
1.00
0.80
0.75
0.80
1.00
0.80
0.80
0.80
0.85
0.85
0.90
                                           (Continued)

-------
                                                       TABLE 1 (Continued)
Watershed Name
Silas Creek at Pine
Valley Road
Tar Branch at Walnut
Street
Brushy Creek at U.S.
Hwy. 311
Sanderson Gulch
Tributary
Tuck Drain
Western Outfall
Sewer
17th Street Sewer
Northwest Trunk
Southern Outfall
Southwestern Outfall
Freeman Field A
Freeman Field B(l)
+ Apron
Freeman Field B + Taxi
St. Anne Auxiliary
Field
Godman Field No. 1
Watershed Drainage Area, A
Location (Square Miles)
Winston-Sal em,
N.C.
ii
it
Lakewood,
Colorado
Northglenn,
Colorado
Louisville,
Kentucky
ii
ii
ti
ii
Indiana
ii
ii
Ind iana
Kentucky
0.89
0.59
0.55
0.47
0.067
2.77
0.22
1.90
6.43
7.52
0.015
0.0128
0.014
0.114
0.0205
Main Channel
Length, L
(Feet)
8,554
6,706
5,808
4,752
1,584
22,000
4,900
16,000
34,000
34,200
900
555
1,200
2,600
1,300
Main Channel Impervious
Slope, S Cover, I
(Ft. per Foot) (%)
0.0193
0.0295
0.0271
0.0139
0.0163
0.0009
0.0038
0.0012
0.0014
0.0015
0.009
0.0058
0.004
0.008
0.015
19.0
28.0
37.0
51.0
33.0
70.0
83.0
50.0
48.0
33.0
21.6
100.0
100.0
5.1
22.1
Dimensionless
Conveyance
Factor, $
0.90
0.80
0.90
0.70
0.80
0.60
0.60
0.60
0.60
0.60*
1.00
1.00
1.00
1. 00
1.00
u>
•vj
   *Modified i selection process

-------
.01  02  03   04  05  .06  .07   .08   .09  .10
.12  .13   .14   .15  .16  .17
              WEIGHTED  MAIN  CHANNEL  MANNING n  VALUE

FIGURE  2-WATERSHED  CONVEYANCE FACTOR,  $, AS A FUNCTION OF
          PER CENT WATERSHED IMPERVIOUS  COVER, I, AND WEIGHTED
          MAIN  CHANNEL MANNING V VALUE. (Austin/Drainage Manual, 1977)
                           - 138 -

-------
     The effects of urbanization on the unit hydrograph can generally be




accounted for in the percentage of impervious cover (I) and the watershed




conveyance factor (tf).  Man-made improvements such as buildings, paved streets,




and parking lots effectively reduce infiltration into the ground and often




increase the volumes and peak rates of runoff.  The I and  values are related to




these improvements and are indicators of a watershed's runoff response




characteristics.




     A brief history concerning the evolution of the 4 factor is appropriate at




this point.  Espey e£ &!• (1965) found that the introduction of watershed impervious




cover as an index of urbanization was not sufficient for describing  the  runoff




characteristics of some urban watersheds.  It was noted in  this early study that




in most  cases when the channel had been  improved, and/or a  storm sewer  system




existed, the predicted values of  the  time of  rise  (TR) were high compared to




measured values.  Therefore, a new urban factor (^)  was  introduced  to account for




the  reduction in  the  time of rise due to channel  improvements  or  storm  sewers.




Further  studies by Espey e£ al^  (1968) refined  the determination of the  factor




to  include channel vegetation  conditions.   In the latter  study,  for several




watersheds,  predominately  in  the Houston area,  a  significantly shorter  response




time was indicated than  for watersheds in  other areas.   The only apparent variable




that could account for the shorter response times was channel  vegetation.   To




explain the effect of channel  vegetation on the time of rise,  the watershed




conveyance factor (
-------
      *1
    CLASSIFICATION
     0.6
EXTENSIVE CHANNEL IMPROVEMENT AND
STORM SEWER SYSTEM,  CLOSED CONDUIT
CHANNEL SYSTEM.
     0.8
SOME CHANNEL  IMPROVEMENT AND STORM
SEWERS; MAINLY CLEANING AND ENLARGE-
MENT OF EXISTING  CHANNEL.
     1.0
NATURAL CHANNEL  CONDITIONS.
$2
0.0
0. 1
0.2
0.3
CLASSIFICATION
NO CHANNEL VEGETATION.
LIGHT CHANNEL VEGETATION.
MODERATE CHANNEL VEGETATION,
HEAVY CHANNEL VEGETATION.
                    = <£,
TABLE 2 - CLASSIFICATION  OF WATERSHED DRAINAGE SYSTEMS
          (Espey et  gl., 1968)
                       - 140 -

-------
that study the 4> factor was determined  for each watershed  shown In Table  1 and




used as one of the watershed physiographic characteristics in a statistical




analysis to determine the ten-minute unit hydrograph parameter equations.




Figure 2 was then developed to aid users in the determination of the 4 factor




for urban watersheds.



     It must be pointed out that the normal process for selecting a watershed's 4




value (Figure 2) may not be applicable for watersheds with atypical drainage




characteristics.  For  that reason, judgment must be exercised in the use of



Figure 2 because the data base used in the development of the figure was limited




(Table 2).  Future study and collection of additional data may indicate  a need




for modification or refinement of  the  selection process.  If flow-modifying




conditions  exist  in a  watershed, such as  excessive  surface detention  storage,



combined  sewer  systems, or  overtaxed storm sewer systems  combined with a lack of




surface escape  routes, additional  hydrologic  procedures may be required  for  the




description of  the rainfall-runoff process.




      Rainfall-runoff data  for numerous  watersheds  were reviewed  for applicability



 in describing the composition of the ten-minute unit hydrograph equations.   The



major share of this  data was obtained  from a  TRACOR report (Espey,  et aJU,  1968)




and a report  to the  Texas  Water Commission (Espey, e£ al.,  1965).  Reported in




 these two documents  were reduced runoff data  from fifty  watersheds providing



 average thirty-minute unit hydrographs for the respective watersheds.  Data




 compilation8 reported by the U.S. Geological  Survey were also reviewed, from



  hich a few more watersheds were found with adequate hydrologic data that could




 be incorporated into  the data base.  The previously derived thirty-minute unit




 hvdrographs were transformed into ten-minute unit hydrographs by the S-curve




 procedure, which  is described in most comprehensive hydrology texts.  The results




   £ £he S-curve reductions  combined with  new unit hydrographs  were  then  studied




 and  the  resulting ten-minute unit hydrographs  having  the best shape were chosen
                                         - 141 -

-------
 for further analysis.  From  this  process  the  final  forty-one watersheds listed
 in Table  1  were  selected  to  determine  the relationships between physical watershed
 characteristics  and  the shape characteristics of the ten-minute unit hydrograph to
 be generalized.  Definitions of the  five  hydrograph parameters discussed earlier
 were derived  from among the  variables  affecting the ten-minute unit hydrographs for
 the forty-one selected watersheds.
 Development of Empirical  Ten-Minute  Unit  Hydrograph Equations
     The  objective of developing  empirical equations to describe the runoff
 response  characteristics  of  watersheds (i.e., a unit hydrograph) is to provide a
 means  for taking known or relatively easily obtained information, such as rainfall
 and watershed physical characteristics, and to describe synthetically the runoff
 hydrograph  for a particular  rainfall event.  Of course, there are other
 considerations that must  be  accounted  for along with the synthetic unit hydrograph
 in describing runoff patterns.  These considerations include such items as
 infiltration  rates, rainfall surface-storage, base  flow, and other flow-modifying
 conditions.
     The  problem of statistically synthesizing a unit hydrograph is by definition
 one of approximating the  hydrograph  shape.  Empirical equations allow the
 description of a hydrograph  in terms of typical hydrograph dimensions (parameters)
 avoiding  the  difficulties that would have to be overcome in attempting to derive
 a mathematical function for  the hydrograph.  Having described the hydrograph for
 each of the forty-one watersheds in  terms of five parameters, each parameter was
 statistically described as a function of the physiographic features and/or other
 unit hydrograph parameters of the watershed.
     To describe the five parameters mathematically, a multiple non-linear
regression formula of the following form was used:
                                  A
                                V  -2
Y = KX,   Xn   . . . AI,  ,
                                      - 142 -

-------
where Y is one of the five hydrograph parameters,

      X  (for i = 1, ..., N) are physiographic characteristics or hydrograph
       1  parameters for the watershed, and

      K and A  (for i = 1, ..., N) are regression coefficients.

The above equation was expressed in the following logarithmic form:

                   log Y » log K + AI log Xx + . . . A^ log X^

and the method of least squares was then applied to evaluate the regression

coefficients.  This method, instead of minimizing the sum of the squares of the

deviations of the function itself, minimizes the sum of the squares of the

deviations of the logarithm of  the function.

      Table 3 presents  the results of  the regression analysis  in  the form of the

ten-minute unit  hydrograph equations.  For easy reference, brief definitions are

included  in Table 3 of the watershed  physiographic characteristics and the  five

hydrograph parameters.  Table 3 also  presents  the  statistical accuracy of  the

empirical ten-minute  unit hydrograph equations obtained using the data from the

forty-one watersheds,  by giving the  total  explained variation of the  logarithmic

oredictions  for each of the five equations, where the value  1.000 would be for

a perfect fit.

            and Examples
      To simplify tne process of synthesizing ten-minute unit hydrographs, a set

 Of parallel, scaled nomographs were constructed using the empirical equations, to

 facilitate user determination of values for all five of the unit hydrograph

 parameters.  A ten -minute unit hydrograph can then be constructed resembling the

 ofle  shown in Figure 1.  These nomographs are presented in Figures 3 through 7.

 A description of the respective nomographs shown in each of these five figures

 will now be given along with an example of their use.

      Determination of  the unit hydrograph parameters using  the  developed nomographs
                                          - 143 -

-------
                                      TABLE 3

                       TEN-MINUTE UNIT HYDROGRAPH EQUATIONS

                                                                    Total
            Equations                                          Explained Variation


 T^.IL^V0'25!-0-18*1-57                                     0.802


 Q  ,  31.62xl03 A0-9V°7                                         0.936

 T_  •  125.89xl03 A  Q~°'95                                            0.844
  B

 W5()-  16.22xl03 A0'93 Q-°-92                                         0.943


 W75»  3.24xl03 A0'79 Q-°'78                                          0.834


 L    is  the  total  distance (in feet)  along  the main channel from the point being
      considered to the upstream watershed boundary.

 S    is  the  main channel slope (in  feet per foot) as defined by H/(0.8L), where L
      ia  the  main channel length as  described above and H is the difference in
      elevation between two points,  A  and B.  A is a point on the channel bottom
      at  a distance of 0.2L downstream from  the upstream watershed boundary.  B is
      a point on the channel bottom  at the downstream point being considered.

 I    is  the  impervious area within  the watershed (in per cent).

 4    is  the  dimensionless watershed conveyance factor as described previously in
      the text.

A    is  the  watershed drainage area (in square miles).

 T_    is  the  time of rise of the unit  hydrograph (in minutes).
  j\

 Q    is  the  peak flow of the unit hydrograph (in cfs).

 T0    is  the  time base of the unit hydrograph (in minutes).
  B

 W50   is  the width  of the hydrograph at 50% of the Q (in minutes).

 W?5   is  the width  of the unit hydrograph at 75% of Q (in minutes).
                                         -  144  -

-------
50,000


O
^ 20,000<
•*-
— 10.OOO
_j
-"
x
o ^o00

z
UJ
_l 2,000
_J
2 1000
Z
X
0 400
Z
2 200

IOO

f-
r c
*- «
^- S
r ^
b <—
H_ *"^
— "*•*'
r oc
r uj
r >
= o
r u
r co
r o
^- >
— cc.
=- UJ
9- O.
i_ .^
r^ 2
^
h S
= X
r- <"
E_ uj
£T
*

a,






=-
L oc
= o
1-3 <-»
— 4
r u.
L4 w
= ""Z
= <
=-5 >
=. UJ
1-6 Z
E °
E-8 0 <
= UJ
—9 X
S-K) ^
- ui
— <
E ^
—
E-
— 1.5


—
•>
~ 3
-10 •-
^ ' C
— ~~- ^_^
E~ -^ *~
=-0.9 "" ^H?
~ ""^
=". UJ
- t CO
— *
— u.
— O
=Lo.7 uj
— «
r »-
~ /
— 06 ..^
=-50O
—
^
^_
|-
=-2OO
—
^ .
—
^100
5-
~ ^^ — ^
fso "
_ ^~ 	 -^
=• / %
5-20 li

r z
I z
1-
i-20 /
=- X
= /
1. /
1-30 /
[/
t~4O
|-50
1- 60












e/
p __
UJ
z
_i

0
z
z
(T
t-







R











c
' - _^_
~~












E-7O
E-80
i-90 FIGURES- TIME OF RISE (TR) NOMOGRAPH
e-100
UJ
Q.
O
_J
CO
U)
o

z
   .0005


==.001


|-002

t


I-.005


   01
EE_b

|-.02

t

S-.04

-------
                   Q = 31.62 xl03(A-96)(TR~L07)
                                                   i-IO
                                                   r'5



r-15
C\J I
0) ~
r -10 •£
E -9 "
< L; §
r6 _, -m-" <
LLl ; 	 	 —""" """ Q:
< : 4 x
r o
-3 co
UJ ; _
CO r- Q
z --z ^,
< ~ "*
2 f S
-
-1.0
- 9
=- 30,000
=-20,000
=-
z
^10,000
—
1-5,000
1- 4,000 __ ^- — " "
j~ ^T^^^^
1- 2,000
i: 1,000
h 500
|- 400
1- 300
i-
|- 200


— 100
- .8
— .7
6'










^_ — "
£
3
c
E
"^r
K
Ul
CO
o:
u.
o
Ul
3s
(-







E-20
|-
|-25
~
r30
-
t*
|-40

|-50
1-60
5-70
E-80
L-90
^-100



— 150


E-200


§-250
L300
E-400
-500
FIGURE 4-  PEAK DISCHARGE  (Q) NOMOGRAPH

-------
                                      PEAK  DISCHARGE  (Q)   cfs
                    Hinniuinuipiinim|mn i |i p|i| i  n 'l""l
               o
               o
                          to
                                         o» -J CO 
                                        TIME   BASE   (Tg)   minutes
                                                            iyni|iiM|i|i|i|ipiyiii|iiiin

                                                   JSSM  S    gSS^SSiS  S
                                                   ao 5 o p  o    °^
O

o
                                      DRAINAGE  AREA   (A) miles
                                                                                                       m
                                               0»
                                         bbo  b o  o  o
                                                                          (0 OD --j  »

-------
o

O


UJ
O
IT

X
O

—
5
ui
a.
     -4
    E-3
    — 2
      30,000




     -20,000
     -10,000
     •9
     -7


      6

     ._ |E
0>
O
     -1,000
     -9

     -8

     -7

     -6

     -5


     -4
    =-3
    — 2
                                  — ! 0
                                  =-2.0


                                  !-30

                                  E-4.0

                                  i- 5.0
                                     60
                                     70
                                     8.0
                                     9.0
                                     10.0
                                  E- 20 0
                                                                    evj
                                  — 100.0
=- 200.0
j- 300 0

   400.0

   500.0
   600.0
   700.0
   8000
   9000
   1000.0
                                                                     UJ
                                                                     a:
                                                                     ui
                                                                     c
                                                                     o
                                                                        F^ .5
                                        5- .7

                                        ^ 8

                                        I 9

                                        ^ 1.0
                                        r 1.5



                                        L 2.0



                                        1- 2-5


                                        I 3.0
                                        - 5.0


                                         • 6.0

                                        - 7.0

                                        -8.0

                                        -9.0

                                        E: 10 o
                                                                        — 15.0
      • 100
         FIGURE 6- HYDROGRAPH WIDTH (W5O) NOMOGRAPH
                                      - 148  -

-------
c- 30,000
                      W75=3.24xl03(A79)(Q~78)
- 20,000











*—
a

'i
"" — -^ 'To
^ ^ --•
X
H
9
^g

X
a

a:
o
o
oc
a
X








- 10
^ 20


=•3.0
^•40
1-5.0
L6.0
-7.0
rs.o
r9.0
710.0



r200
=-30.0
1-40.0

i-so.o
^-60.0
-70.0
rSO.O
: 106.0
•
-
~
-200.0
•
- 300.0
-400.0

- 500.0
^- 600.0
L 700.0
L8OO.O
                                                     p-.5
                                                     - .6
                                                     - .7
                                                      .6

                                                      1.0
                                                       20
                                                     =- 25
                                                       40
                                                       50
 FIGURE?- HYDROGRAPH WIDTH (W75) NOMOGRAPH
                         - 149 -

-------
 is shown below for a watershed  having  the  following physiographic characteristics:
                       A   =»    4.1  square  miles
                       L   -    21,000  feet
                       S   -    0.01  ft/ft
                       I   m    40 per  cent and
                       4   »    0.8
      The  unit hydrograph  time of rise (Tj.) determination is a multiple-step
 process that can be  followed In Figure 3.  Align the scale values for L (21,000
 at  point  a) and S  (0.01 at point b) with a straightedge and mark point c on
 turning line 1.  Align point c and the scale value for I (40 at point d) along a
 straight  line and  mark point e on the intersection with turning line 2.  Align
 point e and the scale value for $ (0.8 at point f) with a straightedge and read
 the time  of rise (TR) value at the intersection with its scale (35 at point g).
      The  peak discharge (Q) of the unit hydrograph can be obtained from Figure 4
 by simply aligning the drainage area scale value (4.1) and the time of rise scale
 value (35) with a  straightedge and reading the Q value from the peak discharge
 scale in  the middle (2700).
      Figure 5 provides the unit hydrograph time base (TR) by aligning the peak
discharge scale value (2700) and the drainage area scale value (4.1) with a
straightedge and obtaining the value from the time base scale in the middle (285)
     Hydrograph widths W   and W _  can be respectively found by aligning the
drainage area (4.1) and peak discharge scale values (2700) with a straightedge in
Figures 6 and 7 and reading the appropriate values from the hydrograph width
scales in the middle of each nomograph (W   - 42 and W   = 21).
     The above determinations are summarized below for the five ten-minute unit
hydrograph parameters:
                       TR   -   35 minutes
                       Q     -   2700  cfs
                       T0    »   285 minutes
                       a
                       W.    »   42 minutes
                       W    m   21  minutes
                                        - 150 -

-------
       8 shows a constructed ten-minute unit hydrograph satisfying these five
hydrograph parameter values.   The dimensionless  unit hydrograph shape presented
in the USDA-SCS (1971) hydrology manual was used  to approximate initially the
shape of the ascending and descending portions of the constructed hydrograph in
Figure 8.   ^ne un*-t hydrograph's ascending and descending portions below 50
per cent of peak discharge were then adjusted until the represented runoff volume
equalled one  inch.
inferences
Boston Society of Civil Engineers, 1930.  Report of the Committee on Floods:
     journal  of  the  Boston  Society of Civil Engineers, V. 17, no. 7 (Sept.)
     p.  285-464.
City of Austin Engineering  Department,  1977.  Drainage Criteria Manual, First
     Edition.
Espey, Jr., w- H-» C*  W*  ^rg*11*  and F- D-  Masch,  1965.  A Study  of Some  Effects
     of  Urbanization on Storm Runoff from a Small  Watershed:   Technical Report
     HYD 07-6501,  CRWR-2, Center for Research in Water Resources,  Department of
     Civil Engineering, University of  Texas,  Austin,  Texas.
 Espey   Jr., W.  H., and D. E. Winslow,  1968.  The Effects of  Urbanization on Unit
     Hydrographs for Small  Watersheds, Tracor Document No. 68-975-U,  including
      appendices.
 Folse,  J. A., 1929.   A New Method of Estimating Streamflow:   Carnegie Institution
      of Washington,  Publication 400.
 Hanm  D. W.,  C. W. Morgan and H. A. Reeder, 1973.  Statistical Analysis of
      Hydrograph Characteristics for Small Urban Watersheds,  Tracor Document
      No. T73-AU-9559-U.
 Morgan, P. E«i and  s» M* Johnson, 1962.  Analysis of Synthetic Unit-Graph Methods:
      Journal of the Hydraulics Division, Proceedings, American Society of Civil
      Engineers, V.  88, No.  HY5,  pt. 1  (Sept.),  p. 199-220.
 U S. Department of  Agriculture,  1971.  National Engineering Handbook,  Section 4,
      Hydrology, USDA-SCS.
                                           - 151 -

-------
                          TR = 35 min
en
ro
              2750 f*
                           40
80
  120       160


TIME- minutes
                                                                  200
                                      240
280
                                                   300
                         FIGURE 8- EXAMPLE OF TEN-MINUTE UNIT HYDR06RAPH  CONSTRUCTION

-------
   RESEARCH ON THE DESIGN STORM CONCEPT
ASCE Urban Water Resources Research Program

        Technical Memorandum No. 33

                    by

               Jiri Marsalek
     National Water Research Institute
           Burlington, Ontario,
                  Canada
              September, 1978
    American Society of Civil  Engineers
           345 East 47th Street
           New York, N.Y. 10017
                   - 153 -

-------
                                     PREFACE

                               by M. B. McPherson

 Background

           The following Technical Memorandum is Addendum 4 of a 1977  ASCE
 Program report on "Urban Runoff Control Planning".'*'  Addendum 1,  "Metropolitan
 Inventories," and Addendum 2. "The Design Storm Concept," were appended to the
 latter report.  Addendum 3^2' was the first of several additional,  individual
 Addenda to be released over the period 1977-1979.

           The principal intended audience of the ASCE Program's June,  1977,
 report was the agencies and their agents that are  participating in  the preparation
 of areawide plans for water pollution abatement management pursuant to Section 208
 of the Federal Water Pollution Control Act Amendments of 1972 (P.L. 92-500).
 While the presentation which follows is also directed to areawide agencies and
 their agents, it is expected that it will be of interest and  use to many others
 particularly local governments.

 ASCE Program

           The American Society of Civil Engineers' Urban Water Resources  Research
 Program was initiated and  developed by the ASCE Urban Water Resources  Research
 Council.   The basic purpose of the Program is  to promote needed research  and
 facilitate the transfer of findings from research  to  users on a national  scale.

           Abstracts of the twenty-eight reports  and technical memoranda of the
 Program for the  1967-1974  period  are included  in a readily available paper.(3)
 The  two reports  and the six technical  memoranda  of the  regular series  completed
 since are  identified in a  companion publication.'^'  Also  included  in  the  latter
 is a listing of  all  but one of the twelve  national reports in the special  technical
 memorandum series  for the  International  Hydrological  Programme;  and the last
 national report^5)  and  an  international  summary™) have  been  released  since.

          A  Steering Committee designated  by the ASCE Council  gives general
 direction  to  the Program:   S.  W.  Jens  (Chairman); W. C. Ackermann; J.  C. Geyer-
 C. F.  Izzard;  D. E.  Jones,  Jr.; and  L. S.  Tucker.  M. B. McPherson is Program
 Director (23 Watson  Street,  Marblehead,  Mass. 01945).  Administrative support  is
 provided by ASCE Headquarters  in New York City.

 Design Storms

           In our Addendum  2, "The Design Storm Concept," in our June,  1977,
 report^1' we attempted  to  indicate the hazards that might be encountered in
 interpreting results of analyses based on that concept.  However, we were forced
 to dwell almost exclusively on the characteristics  of rainfall because demonstrate
of the effects of transforming synthetic rainfalls  into runoff were lacking.  Xhe
 report that follows provides the first such demonstration.  While the  findings ar
 site-specific and are certainly not universal, they effectively show the potential
 liabilities in the adoption of synthetic storms for the planning and design of
 important works.

          We have long advocated reference to a long period of historical rainfall
for simulation of important sewered systems.  It was  not proposed that the entir


                                         - 154 -

-------
record would be used in every simulation exercise.   Rather,  we suggested  that
the about two-dozen actual storms of analysis  interest be identified from total
record simulations of token catchments,  and that the resultant set of "design
storms" be applied thereafter in analyses for  other catchments in the jurisdiction
involved.  The report herein describes an alternative method for accomplishing
essentially the same objective.  A somewhat similar procedure was reported in
1972(7) for segregating storms of primary interest to facilitate simulation of
urban catchment flows for a long period of record.   More recently, a storm
screening procedure applied to the rainfall data for several Canadian cities has
been described. (8)

          An alternative to the above approaches has been sought in an attempt to
oreserve the statistically attractive features of continuous hydrological modeling
while reducing the extensive computer costs of such simulations.  To this end, a
"fixed recurrence interval transfer technique" has been developed by the
Southeastern Wisconsin Regional Planning Commission.^9)  Using this technique,
once a full continuous simulation has been completed for a given catchment
condition,  it  is possible to explore revised conditions via simulation of a few
selected meteorological intervals of data.  In this way, alternative plans and
oro lections can be compared.  The technique was developed "for preliminary
screening and  assessment of the impact of alternative land use conditions and
structural  water control measures". (9)  While the technique  is illustrated  in the
reference with flood  flow simulation examples,  it is noted that  the concept has
been used already in  water quality simulations. (10)

          Testing of  the design  storm concept is also under  way  overseas,  for
example  in  Sweden. (H)

          Results have been  reported  from a preliminary  test of  design storms
versus actual  rainfall data  in Denver. (12)  The 190-acre catchment is drained by
separate storm sewers and concrete  lined  channels.  A unit hydrograph derived
from field  data  for  the catchment was used in the  tests. Concluded was  that
design  storms  can produce results that  can vary significantly from the probability
distribution of  runoff  simulated using  actual rainstorm data.  Planned are similar
analyses for about  fifteen other catchments for which rainfall  and runoff data  are
available.

           Initial  indications from  an investigation at the  University of Illinois
    Urbana of  the design storm concept will be reported in December 1978. (13)
This  writer was  advised  in  private  correspondence  with the  University  investigator
that preliminary findings,  for a large  urban  watershed for  which the model used
  AS calibrated against field data,  were analogous  to those  in Denver.

           Both the Denver and University of Illinois correspondents have warned
  Hat the magnitudes of assumed rainfall abstractions used in simulations can  have
    large an effect on the results  as the use  of 'design storms.

           We may state with little  fear of contradiction that exploration of the
   sign storm question has only barely begun.   This is not too surprising when we
   nsider that most of the more comprehensive simulation tools emerged in the 1970's,
           This report draws on and is an extension of findings from papers
     ented at four international conferences.  Mr. Marsalek has assembled  in this



                                             - 155 -

-------
 report  the salient results of his research on  implications  in the use of the desien
 storm concept  in urban sewered  system planning and design.  The Hydraulics Research
 Division of the National Water  Research Institute at the Canada Centre for Inland
 Waters  has spearheaded a substantial research  effort on urban hydrology, resultina
 in products of considerable  international value, mostly under the Canada-Ontario^
 Agreement on Great Lakes Water  Quality.  In 1976, Mr. Marsalek assembled a
 national state-of-the-art report on urban hydrological modeling and catchment
 research in Canada(14) for the  ASCE Program, which was reprinted in Canada(15)
 and included in a Unesco publication.(*•&)

          The ASCE Urban Water  Resources Research Council is indebted to
 Mr. Marsalek and the Hydraulics Research Division of the National Water Research
 Institute for their generous contribution of this report as a public service.

          Processing, duplication and distribution of this Technical Memorandum
was supported by Grant No. ENV77-15668 awarded to ASCE by the U.S. National
Science Foundation.  However, any opinions,  findings, and conclusions or
recommendations expressed herein are those of Mr. Marsalek or this writer and
do not necessarily reflect the views of the National Science Foundation.

References

 1.  McPherson,  M.  B.,  "Urban Runoff Control Planning," ASCE,  New York,  N.Y.
     118 pp.,  June,  1977.   (Available as PB  271 548,  at $5.50 per copy,  from
     the National Technical  Information Service,  5285 Port Royal  Road,
     Springfield, Virginia 22161).

 2.   Espey, W.  H.,  Jr., D. G. Altman and C.  B.  Graves,  Jr.,  "Nomographs  for
     Ten-Minute Unit  Hydrographs for Small Urban Watersheds,"  ASCE UWRR  Program
     Technical  Memorandum  No. 32, ASCE,  New  York,  N.Y.,  22 pp., December   1977
     (NTIS:   PB 282  158).                                                '

 3.   McPherson,  M. B., and G. F. Mangan, Jr.,  "ASCE Urban  Water Resources  Reseav
     Program."  j;Hyd.Div.. ASCE  Proc.,  Vol.  101, No.  HY7,  pp.  847-855, July,  1975

 4.  McPherson,  M. B., and G. F. Mangan, Jr., Closure to Discussion of "ASCE Urb
    Water Resources  Research Program,"  J.Hyd.Djv.. ASCE Proc., Vol.  103  NO  avR0
    pp. 661-663, May, 1977.                                            '    *   *»

 5.  Ramsseshan, S.t  and P. B. S. Sarma, "Urban Hydrological Modeling and Catchmen
    Research in India," ASCE UWRR Program Technical  Memorandum No. 1HP-12  ASCII   *
    New York, N.Y.,  21 pp.,  May, 1977.  (NTIS:  PB 271 300).              '      •

 6.  McPherson, M. B., and F. C. Zuidema, "Urban Hydrological Modeling and
    Catchment Research:  International Summary," ASCE UWRR Program Technical
    Memorandum No. IHP-13, ASCE, New York, N.Y., 48 pp., November, 197?
    (NTIS:  PB 280 754).

 7.  Lee1ere,  Guy, and John C. Schaake, Jr.,  "Derivation of Hydrologic Frequeac
    Curves," Ralph M. Parsons Laboratory for Water Resources and Hydrodynamics
    Report No. 142, M.I.T., Cambridge, Mass. 02139, 151 pp., January, 1972.

8.  Howard, Charles D. D., "Theory of Storage and Treatment-Plant Overflows '•
    J.Env.Engrg.Div.. ASCE Proc.,  Vol. 102,  No. EE4, pp. 709-722, August
    Discussions:  Vol. 103,  No.  EE3, pp. 514-520, June, 1977.  Closure:  '
    Vol. 104, No. EE2, pp. 369-371,  April,  1978.


                                        - 156  -

-------
 9.  Walesh, Stuart G.,  and Daniel F.  Snyder,  "Reducing the Cost  of Continuous
     Hydrologic-Hydraul ic Simulation," Water Resources  Bulletin,  (in press),  1979.

10*  Geiger, W. F., S. A. LaBella and  6.  C.  McDonald,  "Overflow Abatement
     Alternatives Selected by Combining Continuous and  Single-Event Simulations,"
     pp. 71-79 in Proceedings. National Symposium on Urban Hydrology. Hydraulics
     and Sediment Control, Report UKY BU111, University of Kentucky, Lexington,
     386 pp., December,  1976.

11.  Arnell, Viktor, "Analysis of Rainfall Data for Use in Design of Storm Sewer
     Systems," pp. 71-86 in Preprints. International Conference on Urban Storm
     Drainage. 11-15 April 1978, University of Southampton, England, U.K.

12.  Urbonos, Ben, "Some Findings in the  Rainfall -Runoff Data Collected in the
     Denver Area," Flood Hazard News.  Urban Drainage and Flood Control District,
     Denver, Vol. 8, No. 1, pp. 10, 11 and 14, July, 1978.

13.  Wenzel, Harry 6., "Evaluation of the Design Storm Concept," a paper to be
     presented at the session on Urban Hydrologic Systems, Fall Meeting of the
     AGU, San Francisco, December 4-8, 1978.

14.  Marsalek, J., "Urban Hydrological Modeling and Catchment Research in Canada,"
     ASCE UWRR Program Technical Memorandum No. IHF-3, ASCE, New York, N.Y.,
     52  pp., June, 1976.  (NTIS:  PB  262 068).

15.  Marsalek, J., Urban Hydrological Modeling and Catchment Research in Canada,
     Technical Bulletin  No. 98, Canada Centre for Inland  Waters, Burlington,
     Ontario, 52 pp., 1976.

1$.  Unesco,  Research on Urban  Hydrology. Volume  1, Technical  Papers in
     Hydrology 15,  Imprimerie Beugnet, Paris, 185 pp.,  1977.
           (Areawide planning agencies  should have the  following report In their
 libraries:   Setting the Course for Clean Water,  "A Citizen's Guide to the
 Section 208 Water Quality Management Program," National Wildlife Association,
 Education Division, 1412 16th St.  N.W.,  Washington, D.C. 20036, 64 pp.,  March,
 1978).
                                            - 157  -

-------
     RESEARCH ON THE DESIGN STORM CONCEPT
(Addendum 4 of the ASCE Program report "Urban
     Runoff Control Planning," June,  1977)
                      by
                 Jiri Marsalek
         Hydraulics  Research Division
      National Water Research Institute
              867 Lakeshore  Road
      Burlington, Ontario, Canada L7R 4A6
               September, 1V78
                        - 158 -

-------
                               TABLE OF CONTENTS

                                                                          Page

SECTION 1 - SUMMARY AND CONCLUSIONS	160

SECTION 2 - SYNTHETIC DESIGN STORMS	162
          Background  	  •••*..*........  162
          Chicago Design Storms 	  162
               Figure 1 - Synthetic  and Actual  Storm Hyetographs   ....  164
               Table 1  - Values  of  Parameters  Used for Chicago Method  .  163
          Illinois State Water Survey  Design Storms 	  163
               Table 2  - Maximum Hourly  Rainfalls of Various
                          Return Periods	165
               Table 3  - Median Rainfall Distribution of
                          Predominant Storms  	  165
          Selection of Actual Events	165
               Table 4  - Characteristics of Top-Ranked Actual Storms . .   166
          Discussion	. . •	   167

SECTION 3 - SIMULATED PEAK FLOWS	168
          Background  ..... 	   168
          Actual Catchment Simulations  	 .....   168
               Figure 2 - Recurrence Intervals of Runoff Peaks Simulated,
                          Malvern Catchment, Actual and Design Storms . .   169
          Simulation Time Step	170
          Hypothetical Catchment Simulations   ...... 	   17Q
               Figure 3 - Comparison of Simulated Peak Flows,
                          Hypothetical 26-Ha Catchment   	   171
               Figure 4 - Comparison of Simulated Peak Flows,
                          Hypothetical 130-Ha  Catchment  	  172
          Discussion of Results  ,.	173
          Caveat	•  175
SECTION  4  -  EFFECTS OF DETENTION STORAGE   ..........  	  176
          Background	17 g
           Simulation of Storage  Effects  ..........  	  176
               Table 5   - Detention Reservoir  Characteristics  ......  177
               Figure  5  - Simulated Effects of Storage,
                          Hypothetical Catchment   ............  178
           Discussion  ..........................  177
               Figure 6  - Runoff Volumes  Produced by Selected
                          Actual and  Design Storms  ...........  177
               Table 6  - Ranks  of  Selected Actual Storms with
                          Respect to  Runoff Peaks and Volumes
                          Produced  by These Storms  ......  	   180
                Table 7  - Examples  of Transformation of
                          Frequencies by Storage  .......  	   180

 SECTION 5 - WATER QUALITY SIMULATION CONSIDERATIONS 	   182
           Background	   182
           Antecedent Rainfall Effects 	   182
                Figure 7 - Effects of Dry-Weather Period,
                           Malvern Catchment	  183
                Figure 8 - Probability of Dry-Weather Periods,
                           Burlington, Ontario  ......... 	  185
           Discussion	184

 SECTION 6 - REFERENCES	  186

                                       -  159 -

-------
                                     SECTION 1

                              SUMMARY AND CONCLUSIONS


           This report draws  on and extends  the findings  presented  in  four  recent
 papers,(1~^) assembling in one document the salient results  of research on
 implications in the use of the design storm concept in urban drainage design and
 planning.   Analysis was restricted to sewered catchments in  a southern Ontario
 municipality.

           Urbanization substantially increases the volumes and peak flows  of
 surface  runoff.  The cumulative effect of increased runoff volumes and localized
 peaks then contributes to  flooding of downstream areas.   Such adverse effects of
 urbanization are particularly pronounced in the case of  catchments in which the
 downstream part is  developed and the upstream part is undergoing development.
 This  type  of catchment is  fairly common in  southern Ontario.

           To control the increase in stormwater runoff due to urban development
 many  government regulatory agencies  have introduced criteria for urban drainage*
 design.  Such criteria require various degrees of control of runoff from areas
 undergoing development.

           In order  to  meet requirements  stipulated by runoff control  policies
 the frequencies of  runoff  flows  of various  magnitudes need to be determined.  in
 streamflow flood frequency studies,  the  frequency of occurrence of various
 floods is  often determined directly  from a  flow record.  Since adequate flow
 records  are virtually  nonexistent in the case  of  urban catchments, runoff models
 are employed to produce  surrogate flow records  from which the frequencies of
 occurrence can  be determined.   Such  an approach was taken here and the frequeueie
 were  determined in  two ways.   Firstly,  runoff  flows were simulated for design
 storms of  assigned  frequencies of occurrence and  it was assumed that  the
 frequencies  of  the  runoff  peak flows  were identical to those ascribed to the
 design storms.   Secondly,  runoff flows were simulated for a  large number of
 historical  storms and  the  frequencies of  occurrence of various runoff flows were
 determined directly  from the simulated runoff  flow record.  Since most sewer
 systems are  typically designed for return periods of from one to 10 years, such
 a range of return periods  is of  primary  interest  here.

          To establish the frequency of occurrence of runoff peak flows,  continuous
 runoff simulation was approximated by a series of single-event simulations for 27
 selected actual  storms.  The selection of these storms,  which effectively replaced
 a 15-year rainfall record,  was based on the ranking of all of the storms  accordin
 to their peak rainfall intensities for several durations.                        8

          In Section 3, runoff peak flows simulated for synthetic as  well as
actual rainfall events are compared for an actual catchment and for several
hypothetical catchments patterned after typical urban  developments in southern
Ontario.   Although the results obtained are valid only for the conditions
studied,  the comparisons give a general indication of  the relationship between
synthetic and actual storms and demonstrate  some shortcomings of an approach
based on a synthetic rainfall hyetograph.  The analysis  is restricted to  runoff
peak flows on small  and intermediate catchments (a maximum of 130-ha).
                                          .160 -

-------
          Comparison of runoff peaks simulated for two types of synthetic design
storms and for actual storms of nominally equal return periods produced widely
varying results.  One type of synthetic storm produced runoff peak flows much
larger than those simulated for actual storms of corresponding return periods.
The use of another type of synthetic storm resulted in runoff peak flows that were
also larger than those simulated for the actual storms of corresponding return
periods.

          The use of design storms simplifies preparation of rainfall data for
runoff calculations and drainage design.  It appears plausible that these
subjectively defined storms could be replaced by historical storms which produced
runoff flows of certain frequencies of occurrence on similar catchments in a given
locality.  These historical storms would then be used in future drainage design.
To identify such events, one needs to carry out either a continuous simulation of
runoff for a reasonable time period (10 to 20 years), or to carry out single
event simulations for a number of selected events.  In the latter case, the
initial catchment conditions may be adjusted as necessary to account for antecedent
precipitation.

          Selection of historical storms to be used for design  is affected to some
extent by the characteristics of the urban catchment on which such a storm would
be applied.  It is therefore recommended that a wide range of catchment  parameters
be covered  in runoff simulations serving for  the  selection of historical design
storms.  For the design of  storage, runoff volumes also have  to be considered in
such an analysis.  On  the basis of  simulated  performance for  a  catchment, described
in Section  4, synthetic storms were found to  be  incapable of  representing the true
volume,  timing  and multiple-peak nature of actual hyetographs.  For  the analysis
of storage,  it  is preferable  to employ historical storms producing runoff events
of known  frequencies of occurrence  rather than to use a design storm.   Implications
of the  concept  requiring  that urban runoff be held to  pre-development  levels were
explored  through a  series of  runoff simulations  for different degrees  of detention
storage.

          The last  Section  of this  report is  devoted  to  some water quality
simulation  considerations.  Under  the conditions explored,  it is  concluded  that
 the  design  storm concept  and  single-event runoff simulation are of  limited  use
 for  water-quality oriented  design,  because of statistical  nonhomogeneity of runoff
 quantity and quality events.
                                           - 161 -

-------
                                    SECTION 2

                             SYNTHETIC DESIGN STORMS

 Background

           Sizing of storm drainage conduits is universally based on the concept
 of a design storm.   In application,  it is assumed that the frequency of
 occurrence of a runoff peak flow is  identical to that assigned  to the associated
 rainfall.   Design storms have been traditionally represented as a. block rainfall
 and more recently by synthetic hyetographs.  Such hyetographs are typically
 derived by synthesizing generalizations  from a large number of  actual events.

           The concept of a design storm  and its use in urban drainage design
 and other  applications has been a subject of considerable  criticism.   Particularly
 criticized have been attempts to assign  mean frequencies of probable occurrence
 to storms  of various intensities and durations,  and the presumption of identical
 frequencies of occurrence of coupled rainfall and runoff events has been
 questioned because  of the inherent statistical nonhomogeneity of rainfall and
 runoff processes.(5)  Although such  criticism seems to be  generally justified
 the shortcomings of the design storm concept have been only tenuously
 demonstrated with actual rainfall data,  and only indirectly in  conjunction  with
 runoff calculations.   Included herein are findings from a  study of the design
 storm concent.   Two synthetic storm  configurations were analyzed,  termed the
 "Chicago"(6> and "Illinois State Water Survey (ISWS)"(7) design storms.  The
 Chicago design storm was singled out for attention because several Canadian
 municipalities  have adopted this type of storm representation as  part of their
 design criteria for urban drainage.   The ISWS  design storm was  included  because
 of its association  with a well-known computer simulation model  developed for the
 sizing of  storm drains.

 Chicago Design  Storms

           Formulation  of the  Chicago  synthetic hyetograph  was presented  over
 twenty years  ago  in August, 1957.(°)  Adaptation  of  the method  to  local  rainfall
 data  elsewhere  has  been  reported,  for example,  in India,(8)  the U.S.(9)  and
 Canada.(10)   jhe Chicago method  has been rather wid3ly incorporated  in North
 American practice because it  can be readily derived  from available  rainfall
 intensity-duration-frequency  relationships and partly  because of  limited
 alternative approaches.   A contributing  factor was undoubtedly  its  inclusion
 in a widely used handbook. (H)   When  the method was  presented it was  criticized
 on the  grounds  that  it retained  too many of  the fallacies  and empiricisms inherent
 in the  Rational Method to  recommend  its  principle  for  general use by others.(12)
 The fact remains that  several Canadian municipalities  have  included this type of
 design  storm  in their design  criteria  for urban drainage.

           In an attempt  to  preserve correspondence with actual rainfall  events
 the Chicago method  takes  into account the maximum rainfalls of individual     '
 durations,  the average amount of rainfall antecedent to the peak  intensity  and
 the relative timing of the peak  intensity.  The first  step  in applying the
method  is determination of  the time antecedent to the peak  intensity, expressed
as a dimensionless ratio.  This  time, tr, divides the hyetograph  into  two parts
 and is defined as                                                               *

                                   tr = tp/T


                                         -162 -

-------
where tp is the elapsed time from the onset of rainfall to the peak intensity and
T is the total storm duration.  Values of tr are determined individually for a
number of historical storms and their mean value is used for the design hyetograph.
The hyetograph intensities on either side of the peak are obtained from applicable
local intensity-duration-frequency relationships in. the form
                                  1av
where
is the average maximum rainfall  intensity over a  duration
                                                                        and the
constants a, b and c satisfy a fit to the data.  Typically, one to six hours is
selected as the total storm duration, I.  (However, the choice of T does not
affect  the magnitudes of the peak rainfall intensity or the dimensionless time
to peak).

          Synthetic hyetographs of the Chicago method type have been derived for
various assigned  frequencies for a 15-year rainfall record from the station at
the  Royal Botanical Gardens in Hamilton, Ontario. (13)  These hyetographs were
adopted for the studies to be reported herein.  Figure l(a) is an example, for
a "Two-Year Storm".  Figure l(c) illustrates the typical departure of variations
in an actual storm from those represented in a synthetic formulation.   (The
computed  peak  discharges for Storm No. 47 in Figure l(c) were for about a two-year
frequency).  Parameter values used for the Chicago method are listed in Table  1.
                                      TABLE 1

                             Values  of Parameters Used
                                for  Chicago Method
Return Period, Years
Parameters
tr:
a:
b:
c:
T, rain.:
1
0.48
22.6
0.78
5
180
2
0.48
29.9
0.80
6
180
5
0.48
43.7
0.84
7
180
10
0.48
55.9
0.86
8
180
 jinnots State Water Survey Design Storms

           In this procedure, maximum hourly rainfall depths are taken from local
 data or from intensity-duration-frequency relationships for various return periods.
 Individual hourly depths are then distributed over the selected storm duration in
 accordance with normalized relations developed from Illinois data.O*1^,15)  yOT
 application elsewhere, actual storms are first divided into a number of groups in
 accordance with the relative timing of the peak intensity.  Distributions over time
 are next determined for the predominant group of storms and their median
 distribution is used for the design storm.
                                           -  163  -

-------
  150--
E
E
  100- •
ID

K-
   50--
z

<
DC
                                                       JL
             60
120         0        60


   TIME .MINUTES
60
   (a) CHICAGO "2-YEAR STORM"    (b) ISWSM2-YEAR STORM" (c) ACTUAL STORM

                                                        No. 47
      FIGURE 1-SYNTHETIC  ANQ ACTUAL STORM  HYETOGRAPHS
                                - 164 -

-------
          As for the Chicago method,  the data used to develop ISWS design storms
were from a 15-year rainfall record from the station at the Royal Botanical
Gardens in Hamilton, Ontario.  Maximum hourly rainfall depths that had already
been identified in this record(13) were used, from which values for various
return periods were abstracted, Table 2.
                                     TABLE 2

                Maximum Hourly Rainfalls of Various Return Periods
                        (Royal Botanical Gardens, Hamilton)
                     Maximum Hourly
                      Rainfall, mm:
Return Period^ Years
_1    _2    _5    10


22    26    33    39
          About 30 heavy recorded storms, which are further described in the next
 subsection, were used to determine the temporal rainfall distribution.  These
 storms were divided  into three groups in accordance with the distinctive part of
 each storm in which  the burst of peak rainfall intensity had occurred.  The peak
 intensity for a majority of the storms occurred within the last  third of their
 total duration.  A median rainfall distribution was determined for  this group and
 expressed as

                                  Rep - f(TCp)

 where Rcp is  the cumulative per cent of rainfall,  T    is  the cumulative per cent
 of storm time,  and  f is an empirical function.  Numerical values of this
 distribution, which was adopted for ISWS design hyetographs, are listed in
 Table 3  and an  example hyetograph  (two-year return period)  is  shown in Figure  l(b).
 Again,  there  are  important departures  from actual  storm variations, such as  in the
 example  of an actual storm  in Figure l(c).
                                      TABLE 3

                Median Rainfall Distribution of Predominant Storms
                        (Royal Botanical Gardens, Hamilton)
Cumulative Per Cent of Storm Time, Tcp

Cumulative
Per Cent of
Rainfall, R^t
0 10 JO


0 5 10
3CI 40 M)


15 22 30
60 70


39 56
80


86
90.


96
100


100
 Selection of Actual Events

           For analysis of certain types of urban runoff problems, particularly
 those related to water quality, a very strong case can be made  in support of
 continuous rainfall-runoff-quality simulation.  In some types of analysis,  such
                                           -165 -

-------
as for peak flow determination, continuous simulation can be effectively
approximated by using a series of single-event simulations.  For the studies
reported here, simulations were made for selected actual storms that would be
likely to cause high runoff peak flows on urban catchments, and the findings are
therefore limited to that realm of analysis.

          Selection of actual storms that would be most likely to produce high
runoff peak flows was facilitated by screening the rainfall record to segregate
all storms with either a total rainfall depth larger than 1.25-cm or a ten-minute
intensity larger than 1.5-cni/hr.  A total of 34 storms met one or both of these
criteria.  Next, the top 20 storm depths were identified for durations of 5-
10-,  15-, 30- and 60-mlnutes.  Because a number of the storms contained multiple
martima, the segregation process yielded only 27 storms that net all the selection
criteria.  For the purpose of establishing the frequency of occurrence of runoff
peaks on the catchmeats studied, these 27 storms were regarded as a suitable
replacement for the 15-year rainfall record.  The basic characteristics of the 27
selected storms are summarized in Tabla 4.
                                     TABLE 4

                   Characteristics of Top-Banked Actual Storms
                      (Royal Botanical Gardens,  Hamilton)
Storm
Number
44
2
46
10
25
36
47
20
23
6
1
8
39
54
31
29
37
22
35
11
15
53
17
9
32
43
26
Means:
Total
Rainfall, mm
37.8
57.7
31.2
14.2
44.7
20.8
15.3
46.5
22.9
28.7
30.0
30.7
17.0
78.5
27.7
26.4
24.9
32.8
24.4
80.3
26.4
27.2
20.6
25.9
27.9
37.3
23.4
32.6
Duration,
hours
0.5
10.3
1.5
5.4
4.8
1.0
1.3
6.5
0.6
6.3
9.2
0.7
4.5
18.4
2.4
3.4
1.9
5.6
5.6
19.5
3.8
6.6
9.1
2.9
5.2
14.1
6.2
5.8
Antecedent
Dry Weather
Period, Days
8
2
2
6
3
1
1
3
1
6
3
1
3
8
0
10
1
7
2
4
4
5
6
8
18
2
0
4
5-Day Antecedent
Precipitation, mm
0.8
46.2
10.9
15.2
4.8
18.8
8.9
19.1
3.0
8.4
16.3
17.5
19.8
0.5
21.3
0.5
13.7
0.3
13.5
5.3
5.8
3.6
0.3
0
0 i
36.3 *
18.5
11.5
                                         -166 -

-------
          In the segregation of storms,  the minimum inter-event time was  taken as
three hours.  That is, a storm event was defined as one where at least three hours
without rainfall occurred before and after the event.   On this basis, the average
total rainfall depth was about 33-nm and the average storm duration was about six
hours for the storms selected, Table 4.

          Of interest is the relationship between the antecedent dry-weather
period and the antecedent five-day precipitation of these heavy storms.  Because
the values of these parameters indicated that catchments in the area studied would
have been fairly dry at the beginning of heavy storms, neglecting the effects of
antecedent precipitation on runoff from the associated storms appeared to be a
safe approximation.  This observation contradicts to some extent one of the
objections to the use of design storms but at the same time removes a possible
limitation from the results to be presented.

Diecussion

          Explained above were the reasons why it was assumed to be a safe
approximation to neglect the  long-term effects of antecedent precipitation on
runoff simulations, for the rainfall record involved.  Further  justification  for
ignoring such antecedent effects in this study is given  in the next Section.
However, long-term antecedent conditions indicated by the raingage record from
Hamilton in Table 4 may well be unusual.  Also, for small rural or very  sparcely
urbanized catchments, not considered here, it would probably be hazardous to
neglect long-term antecedent  influences.

          The most  important  objections to the use of design  storms  of the  type
involved here are that  they:  attempt to summarize variegated  storm patterns  in
a  single hyetograph shape; assemble components of equal  individual  probability of
occurrence  that have  a  quite  different  joint probability; and  ignore the
possibility that antecedent conditions may vary considerably  from  storm  to  storm.
Described  in  the next Section is the computer model used in the simulations,  which
has  a  single-event  capability in the version employed.   Therefore,  it was
serendipitous to be able  to omit the simulation of all  antecedent  rainfall
required in continuous  simulation.  That  is, because  minor long-term antecedent
influences  were  indicated in  the rainfall  record used,  it was deemed reasonable
to use the single-event model without substantial  modification. In what could be
the  more usual  case where long-term antecedent  conditions are more pronounced, it
would  be prudent  to rely on continuous  simulation.   In sum,  it must be recognized
that the effect of  accounting fully for the  influence of antecedent precipitation
on runoff  for recorded  storms,  as  opposed to  its omission in design storms, was
not  Investigated  in this  study.  However,  antecedent conditions were taken into
account in this study,  when required, by adjusting the initial detention storage
and  infiltration.   That is,  because the version of the computer model used does
not  maintain  water balances between storms,  this  was done externally to  set
 initial conditions when required.   Little guidance for further refinement is
provided in the literature because the  effects of antecedent conditions  on urban
 runoff are frequently discussed on the  basis of conceptual models rather than on
 the  basis  of actual observations.
                                           - 167 -

-------
                                     SECTION 3

                               SIMULATED PEAK FLOWS

 Background

           To avoid the shortcomings of the design storm approach,  several
 researchers have proposed converting the rainfall record into  a runoff  record
 and then determining the frequencies of occurrence of various  runoff flows  from
 the latter record.  Such a conversion can be performed by means of urban runoff
 models.  In the study described here, a version of the Storm Water Management
 Model (SWMM) was used for runoff simulation.  SWMM is essentially  a single-event
 urban runoff model,  although it has recently been run in a continuous simulation
 mode on a test catchment.(16)  Its  development was first reported  in 1971 for
 the U.S. Environmental Protection Agency,(17) it is continuously upgraded by the
 U.S. EPA,(18) and it has been studied at length in Canada.(19)

           The 27 top actual storms  and the synthetic storms described in the
 preceding Section were used in SWMM simulations for an actual  catchment and a
 group of hypothetical catchments.  Simulations were performed  in a single-event
 mode.

 Actual Catchment Simulations

           The test catchment,  known as the Malvern catchment,(2°)  is  a  modern
 residential area of  23-ha with single-family housing units.  The catchment  is
 fully developed,  its imperviousness is 30  per cent,  and it is  drained by separate
 storm sewers.   For modeling purposes,  the  catchment was subdivided into ten
 subcatchments  varying in size  between 0.9-ha and 4.0-ha.   The  sewer system  was
 represented by 21  pipes  varying in  size from 0.3-m to 0.83-ra.   The Malvern
 catchment was  instrumented in  1973  and a large number of  rainfall-runoff events
 have since been recorded there.

           Before  proceeding with the simulations  to  be reported in this
 subsection,  SWMM was calibrated  and verified for  the Malvern catchment.  About
 25  rainfall-runoff events  were available for this  purpose,  for which  the return
 periods of  the  two highest  observed runoff peak flows  were about 1-year.  On the
 average,  the calibrated  model  underestimated observed  runoff volumes by 1%  and
 observed  peak  flows  by 5%,  with  standard deviations  about  the  means of  12%  and
 16%  respectively.  The raingage  serving  the  Malvern  catchment  is located about
 10-km east  of  the  raingage  at  the Royal  Botanical Gardens,  the source of the
 15-year record  from  which  the  27  storms  cited  in Table 4 were  obtained  and  from
which  the design storms were derived.

          Design storms determined  according to the Chicago method  and  the  ISWS
method were then used with  the calibrated  SWMM, for assigned return periods of
 1-year, 2-years, 5-years and 15-years.   By definition,  the return periods of the
 resulting runoff peak  flows were  assumed to be  identical to those assigned  the
design storms.  For  the actual storms, the simulated runoff peak flows were
 ranked and  their frequency of occurrence determined therefrom.   The results of
 the Malvern catchment  simulations are presented in Figure 2.   There are large
discrepancies between  the runoff  peaks for the historical storms and for the
design storms having the same nominal frequencies.  This discrepancy will be
discussed after the  results from the simulations for the hypothetical catchments
are presented.
                                          -168 -

-------
  0.2
 o
 o>
  I I
             2     3  456789 10
       AVERAGE  RETURN  PERIOD, YEARS
                                       15
               LEGEND - Chicago Design Storms +
                       ISWS Design Storms  x
                       Actual Storms       •
 FIGURE 2- RECURRENCE INTERVALS OF RUNOFF
           PEAKS SIMULATED, MALVERN CATCHMENT,
           ACTUAL AND DESIGN  STORMS
                       - 169 -

-------
 Simulation Time Step

          The simulation time step was selected on the basis of previous
 studies on the Malvern catchment with SWMM.^20^  As a part of those simulation
 studies, the effect of the time step on the reproduction of observed hydrographs
 was investigated.  For high intensity storms, the best reproduction was
 obtained for simulations with time steps of one and two minutes.  Consequently,
 these two time steps were adopted for all the simulations discussed in this
 report.  Specifically, a one-minute time step was used for storms with a
 duration of two hours or less and a two-minute time step was used for storms of
 longer duration.  Use of such small time steps makes possible the determination
 of runoff peak flows from generated hydrographs without any need for interpolation

           Hyetographs for the design storms were accordingly discretized into
 one or two minute intervals.  For the one-hour duration ISWS design storm a
 one-minute interval was used, whereas a two-minute interval was used for the
 longer Chicago design storm, which was assigned a duration of three hours as
 specified by its proponents.  For all simulations reported, the simulation time
 step was therefore the same as the time interval for the input rainfall  data.

 Hypothetical Catchment Simulations

           Physical catchment parameters strongly influence runoff simulations and
 can to some extent influence the selection of rainfall inputs.   In order to
 explore the effects of such parameters,  runoff flows were simulated for  a series
 of nine hypothetical catchments  of widely varying characteristics.  These
 catchments  were patterned after  some typical  urban catchments  located in modern
 residential developments  in Ontario.  Three catchment sizes were used:   26-ha
 52-ha  and 130-ha.   The drainage  was maintained at about the same density for all
 three  cases.   Catchment imperviousness  was set at three levels:   15%,  30% and
 45%.   The last two levels are typical for modern residential areas in Ontario.
 Because of  the more extended sewer system,  the lag time for the 130-ha catchment
 was longer  by about 7  to  14 minutes than that for the 26-ha catchment.

          Two  types of rainfall  inputs  were used in runoff simulations:   design
 storms  obtained via the Chicago  and ISWS methods described earlier,  for  four
 return  periods; and the 27  selected actual  storms.

          Return periods  and the associated peak runoff flows simulated  for the
various  actual  and  synthetic storms  are  plotted  in  Figure  3  for  the  smallest
hypothetical catchment  (26-ha),  separately  for each of the three  levels  of
 imperviousness.  Figure 4  is a companion illustration,  for the  largest
hypothetical catchment  (130-ha).   These  and similar plottings for  the  intermediate
catchment size  revealed an  attenuation  in peak flows  per unit area with  an  increas
 in  catchment size.   This attenuation in  peak  flow was  fairly consistent  and
represented about a  13% reduction  for the  largest (130-ha)  catchments over  the
smallest  (26-ha) for otherwise identical characteristics.   It is conceivable that
even larger differences could be  encountered  in  practice,  depending on the
relation of the lag  times of the catchments involved.  The  attenuation in peak
flows with  increase  in  catchment size is attributed  to  the  larger  concentration
times noted above.  The degree of  this attenuation  increased with  the amount of
imperviousness.

          Comparison of runoff peaks simulated for actual and synthetic  storms
yielded  interesting results.  For all four  return periods, both design storms


                                         .170 -

-------
10 . 1MPERVIOUSNESS, 15%
.00
    IMPERVIOUSNESS, 30°>5
.20
 .10 •
    IMPERVIOUSNESS, 45%
8   10
               2       3    4   5   6

        AVERAGE  RETURN PERIOD, YEARS

                 LEGEND - Chicago Storms  •
                         ISWS  Storms   •
                         Actual Storms   A
  FIGURE 3 -COMPARISON  OF SIMULATED PEAK FLOWS,
              HYPOTHETICAL 26-HA CATCHMENT
                         - 171 -

-------
      IMPERVIOUSNESS
      IMPERVIOUSNESS, 30%
O
      IMPERVIOUSNESS, 45%
                2       34568

         AVERAGE RETURN PERIOD, YEARS


                  LEGEND — Chicago Storms  •
                          ISWS Storms    •
                          Actual  Storms   A
   FIGURE 4-COMPARISON OF SIMULATED  PEAK FLOWS,
               HYPOTHETICAL J30-HA CATCHMENT
                          - 172 -

-------
produced flows larger than those produced by the actual storms for corresponding
return periods.  This overestimation was  particularly large for the design
storms derived via the Chicago method,  which produced peak flows for all three
sizes of catchments that were three-fourths larger,  on the average, than those
produced by the corresponding actual storms.  The design storms derived via the
ISWS method produced better results, with simulated  peak flows only one-fifth
higher, on the average, than those for the corresponding actual storms.

          Recall that the ISWS design storms used had a delayed pattern, dictated
by the predominant characteristics of the actual storms from which they were
derived.  In an attempt to explore the effect of storm pattern, some simulations
were also performed using the advanced pattern recommended by the ISWS for
drainage design.(^  In contrast to the delayed pattern, starting at the 2-year
level the simulated peak flows for the advanced pattern were lower than for the
actual storms and corresponded, In general, more closely to the actual peak
flows, except at the ten-year level, where they were about one-fifth below the
actual storm rates.  The effect of pattern shift was non-linear, as indicated
below, for simulations with 307. tmperviousness:

                               Ratio of Peak Flow for Advanced Pattern
        Return Period,            to Peak Flow  for Delayed Pattern
            Years	        26-ha Catchment         130-ha Catchment

               1                    0.91                    0.82
               2                    0.84                    0.80
               5                    0.69                    0.69
               10                    0.61                    0.62

As would be expected,  storm pattern evidently has an important influence  on  the
magnitude of  catchment peak flow.

Discussion of Results

          Three  shortcomings affect simulated  hydrographs  obtained through use of
Chicago design storms:

   All the values  represented by a  particular  intensity-duration-frequency curve
    are implied to  belong to the same storm, whereas such curves are typically
    obtained by a synthesis of data  from a large number of storms.

    The intensity-duration curves are extrapolated beyond the smallest reported
    data duration of 5-minutes, yielding a peak rate exceeding the 5-minute
    intensity  by about 607..  (However,  these high intensities are reduced
    somewhat when the hyetograph is divided into constant time intervals).
    The description of the time of the peak intensity by a single tr-value,
    which is  an average of all the tr-values derived from selected storms, is
    questionable in view of the probabilistic nature of this parameter.  Analysis
    of the selected storms reported here yielded a mean tr-value of 0.48 with a
    standard deviation about the mean of 0.32,  Indicating a large scatter.

           While better results were obtained with  the ISWS method, there  is some
 degree of arbitrariness in the definition of these  design storms,  particularly
 in the choice of the  storm duration, which affects  the magnitude  of  rainfall
 intensities.   A one-hour duration had been recommended by the ISWS on the basis
                                          - 173 -

-------
 of findings  from some  simulations  of runoff from several  urban catchments.
 The effect of ISWS  design  storm duration  was investigated in the  present  study.
 For the delayed  pattern, when  the  design  storm duration was  reduced  to  one-half
 of an hour,  the  runoff peak flows  increased about one-third  over  those  obtained
 for a one-hour duration, such  as the cases  plotted  in  Figures  3 and  4 for a
 one-hour duration.  When the design  storm duration  was raised  to  five hours,  the
 runoff peak  flows were much smaller  than  those obtained for  a  one-hour  duration.
 The durations of the actual storms used here (e.g.,  the ones in Table 4) do not
 support the  hypothesis of  a fixed  value of  one-hour duration.   It is evident  that
 the comparatively better performance of the ISWS  design storm  for a  one-hour
 duration reported here may be  largely coincidental.

          Findings  from this study strongly suggest  that  much  more attention
 should be given  to  the rainfall  input than  has  been  the normal practice.  Results
 obtained for the two synthetic design storms  differed  from each other and from
 those for actual storms.   The uncertainty in simulated runoff  peaks  caused by
 the choice of rainfall  input appeared to  be larger  than the  uncertainty inherent
 in the simulation process.

          Recall that  the  actual storms used  for  runoff simulations  were selected
 on the basis  of  the rank of peak intensities  for  durations of  5-,  10-,  15-, 30-
 and 60-minutes.  As a  means of investigating  the  efficiency  of this  process,
 correlations  were examined  between the ranks  of peak rainfall  intensities for
 individual durations and the associated runoff  peak  flows using the  Spearman
 rank correlation coefficient.  Using  the  values for  all 27 actual  storms, the
 correlation  coefficients were larger  than 0.545,  which indicates  a rank
 correlation  significant at  a 0.01  level of  confidence.(21)   The segregation of
 peak intensities therefore  appeared  to have been  a good selection  criterion for
 the identification of  important historical  storms which would  be  associated with
 larger runoff peaks.

          Peak flow and peak storm intensity  are  assumed  to  be  directly related
 in  the  Rational  Method.  The highest  correlation  coefficients  obtained  for
 simulated runoff peak flows versus peak rainfall  intensities (5-,  10-,  15-, 30-
 and  60-minutes duration) of actual associated storms varied between  0.629 and
 0.734.  This means that, at best, only about half of the  linear variation in the
 runoff  peaks could be explained by the linear variation in the rainfall intensity.
 Evidently, other parameters of the rainfall distribution are also  important in
 the generation of realistic runoff peak flows.

          Although the results presented  here are valid only for the conditions
described, the methodology used in the selection of actual storms and the use of
 frequency graphs  of runoff  flows may have a general applicability.  We plan to
apply these methods to studies of other situations.  Graphs of runoff flow
 frequencies,  analogous to  those in Figures 3 and 4, could be used to obtain quick
estimates of runoff peaks  from new urban developments or for checking design values,

          Effects of storage reservoirs on runoff peaks were not considered in
this Section.  As will be  shown in the next Section, such storage transposes the
runoff peak flow frequency curve into a series of smaller values.
                                         -  174  -

-------
Caveat

          The version of SWMM employed does not accommodate surcharging
realistically in a hydraulic sense.  Prior simulations of catchments of the
type involved in this study revealed surcharging of some reaches at the 2-year
to 5-year peak flow level.  Surcharging of entire systems would be expected
somewhere between the 5-year and 15-year peak flow in this type of system.
Because incorporation of the WRE transport module, which accommodates surcharging
realistically, would have required considerable reprogramming for the computer
available, and its use would have been much more costly, an expedient was
adopted instead.  Conduit diameters, Including those of the Malvern catchment,
were artificially enlarged to sizes that would handle all flows to be simulated
without surcharging.  Use of this expedient naturally casts some doubt on the
reliability of the results.  However, because all simulations reported here
employed this expedient, it is felt that the comparative results are nevertheless
valid.
                                          -  175  -

-------
                                    SECTION 4

                          EFFECTS  OF DETENTION STORAGE

 Background

           Various  methods of runoff control are  employed  in modern  innovative
 design of urban drainage.  Use of detention storage  is one of  the most  important
 methods and can be used to achieve  almost  any degree of runoff control, provided
 that the costs  of  such facilities are not  prohibitive.  In Canadian urban
 drainage practice,  use of stormwater detention ponds has  become popular in
 recent years and many  storage facilities of this  kind have been constructed.
 Some of these ponds have a multiple-purpose use,  serving  not only for runoff
 control and reduction  of drainage costs but also  for recreation.  These
 detention ponds are frequently constructed as  small  reservoirs with gently
 sloping embankments and drop-inlet  spillways.

           The hydrological  design of detention ponds  is not well established.
 Regardless,  in  urban conditions where real damage would occur  as a  result of
 flooding,  detention ponds should  be designed  in a manner  analogous  to the design
 of  flood control projects.(22) por SOme urban conditions, the damage would be
 little more than a  nuisance and often a rather arbitrary  decision is made to
 limit the  probability  of this  nuisance to  some acceptable level.  In either case,
 however,  the hydrological techniques employed  should  yield information  adequate
 for the definitions of flood frequencies before and  after control.v22)  As
 discussed  later, an approach based  on the  design storm concept does not meet
 this  criterion.

          One of the most stringent  runoff controls  is imposed by ordering that
 there shall  be  a zero  increase in stormwater runoff due to urban development,  in
 other words,  the runoff  peak flows  from a  developed catchment have  to be
 restricted  to their predevelopment  level for rainfalls of certain frequencies of
 occurrence.   In one case  reported in the literature,  such a standard was enforced
 for rainfalls having a 2-year or shorter return period.(23)

 Simulation of Storage Effects

          The concept of a  zero increase in runoff is not as well defined as it
would  appear.  Technical  implications of this runoff control policy are discussed
 in this Section.  Using historical storms,   the recurrence interval of runoff peak
flows was analyzed  for the hypothetical catchment of 26-ha and an imperviousness
of 30%, described earlier.  Ten selected runoff hydrographs for this hypothetical
test catchment were then routed through detention storage and, by providing a
sufficient storage capacity, runoff peak flows were reduced to the predevelopment
level for a range of return periods.

          Characteristics of the three levels of reservoir storage employed in
the simulations  are listed in Table 5.   In all three cases, the reservoir was
assumed to be located in the vicinity of the catchment outlet.
                                         - 176 -

-------
                                    TABLE 5

                       Detention Reservoir Characteristics

Reservoir Designation
R-l


R-2


R-3


Volume,
cu.m.
200
680
1310
500
1650
3020
660
2160
3900
Depth,
m.
0.3
0.9
1.5
0.3
0.9
1.5
0.3
0.9
1.5
Outflow,
cu.m/sec.
0.13
0.74
0.96
0.13
0.74
0.96
0.13
0.74
0.96
          Routing of runoff hydrographs  through the  detention reservoirs  was
accomplished by using the modified Puls  method.(2*)   The outflow hydrograph peaks
were then used to establish new peak flow versus recurrence interval values,
plotted in Figure 5.  Included also in Figure 5 are  the values from Figure 2,
designated in Figure 5 as "urbanized, no control".   For reference,  results for
"non-urbanized" conditions are also given in Figure  5.

          The peak outflow discharges from storage were affected not only by the
magnitudes of the inflow peaks and the storage volumes, but also by the total
volume of inflow and its time distribution.  It can  be deduced from Figure 5 that
the criterion of a zero increase in runoff with urbanization is essentially
satisfied for the cases of the two larger reservoirs, where the peak rates are
practically identical with those for the undeveloped catchment, particularly for
recurrence intervals of design interest.

Discussion

          Among the various means of runoff control, only detention storage and
its effects on catchment runoff were considered.  By providing a sufficient
storage capacity, runoff from an urbanized catchment was reduced to about the
predevelopraent level for a wide range of recurrence intervals (see Figure 5).

          The design-storm concept was  found to be of questionable use in this
analysis.  Multiple-peak storms may  produce higher outflow peaks from storage,
if a  second consecutive peak arrives when  the  storage facility  is essentially full,
The outflow hydrograph of the Chicago design storm will be affected by the
tr-value.  Peak runoff rates  from storms with  smaller tr-values will be  more
attenuated than those with larger tr-values because  for smaller tr-values the
inflow peak arrives  when  the  available  storage capacity  is larger than it would
be  for larger  tr-values.  Special tests were performed  using Chicago design
storms with 1-year,  2-year and  5-year average  return periods for all three
detention reservoir cases.  The resulting runoff peak  flows  were all higher  than
those indicated  in  Figure 5 for the historical storms  ("urbanized,  with  storage")
with departures of  more  than  100% for the 1-year design storm and  as little  as
about 107. for the 5-year design storm.   These  results  were not unexpected, as may
be  seen in Figure 6, which Is a plot of frequency of flow volumes  without storage
 for the related  simulations (middle graph of Figure 3).  In Figure 6 the

                                            -  177 -

-------
CO
        o>
        V)
        O
          0.05-
        LU
        CL
 _
O
z
Z)
cr
        LEGEND
         Urbanized, No Control    A
         Urbanized, With Storage—

                 Rl  A     R2 •
                          R3 0
                             f
                 Non- Urbanized

                                        1    8
•
                             2       3    4   56789 10

                     AVERAGE  RETURN PERIOD,YEARS
                                                                       15   20
           FIGURE 5-SIMULATED EFFECTS OF STORAGE, HYPOTHETICAL CATCHMENT

                                (26-ha, 30% Imperviousness)

-------
  300
fO


~ 200 f
O
>
£ 100

UJ
u.
O
o:
       LEGEND

         Actual Storms

         Chicago Design Storms

         ISWS Design  Storms
     0.5
                                             H	1   I  I  I  I
5  6  7  8 9 10
                     AVERAGE RETURN  PERIOD,YEARS
                                                               15   20
 FIGURE 6-RUNOFF VOLUMES PRODUCED BY SELECTED ACTUAL AND DESIGN STORMS

                        (26-ha, 30% Imperviousness)

-------
 departures in magnitude of runoff volumes between those for the Chicago  design
 storms and those for the historical storms also decrease with rarer return period
 even though no routing through storage is involved.

           Peak flow rates and magnitude of flow volumes are not highly correlated,
 as indicated in Table 6, where the top 15 peak flow  rates from the 27  actual
 storms are associated with a seemingly random ranking of runoff volumes.  During
 this study it was noted in about half of all cases that storm hydrographs  are
 transformed differently by storage, with the result  that frequencies of  occurrence
 of Inflows to storage and outflows from storage do not agree.  Two of  the  most
 obvious examples of this change in frequencies are given in Table 7.
                                      TABLE 6

              Ranks of Selected Actual Storms with Respect to Runoff
                    Peaks  and Volumes  Produced by These  Storms
Storm
Number
44
2
46
10
25
36
47
20
23
6
1
8
39
54
31
Rank of the Runoff
Peak Produced by
the Storm
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Rank of the Runoff
Volume Produced by
the Storm
4
2
6
15
3
12
14
5
8
7
9
10
13
1
11















                                     TABLE 7

                          Examples of Transformation of
                             Frequencies by Storage
Storm
Number
(Table 4)
10
54
Average Return Period. Years
Without
Storage
4
1
Outflow From
Reservoir R3
1
8
          The above indications substantiate the view that the design storm concept
was found to be of questionable use in this analysis of storage effects.
                                          - 180  -

-------
          The requirement of a zero runoff increase due to urban development is
a political solution to flooding problems created by progressive urbanization,
and under some circumstances this requirement may have little technical
justification.  Typically, such a requirement is specified for a design storm
of a certain assumed frequency.  While some control of urban runoff from
urbanizing catchments is mandatory where increased runoff would contribute to
flooding of downstream areas, this purpose may not be best served by the concept
of zero increase in runoff due to urban development.  As interpreted by various
governmental agencies at the present time, this concept appears to be vaguely
defined and impractical.  Runoff control measures required under this policy
apply to events of more or less arbitrarily specified frequency of occurrence
typically expressed in terms of design storms.

          Urban runoff can be controlled by means of storage to a more Or less
arbitrary degree, including the predevelopment level.  For the analysis of
storage, it is preferable to employ historical storms producing runoff events
of reliable frequencies of occurrence rather than to use a design storm.  The
Chicago storm, in particular, is not capable of representing the true volume,
timing, and multiple-peak nature of actual hyetographs.

          It is recommended that the zero increase in runoff concept be replaced
by a runoff policy in which the degree of runoff control  is determined from a
cost-benefit analysis of the control measures.  The entire basin and the main
stream have to be considered in such an analysis, since localized runoff control
by detention and uncontrolled, poorly-timed releases of stored  runoff may still
lead to flooding in the main stream.
                                           - 181 -

-------
                                    SECTION 5

                     WATER QUALITY SIMULATION CONSIDERATIONS

 Background

           Consideration of the water quality aspects  of  urban drainage discharges
 represents a new facet of urban drainage design.  As  "point" pollution sources
 are abated and become less threatening,  "non-point" sources of water  impairment
 become relatively more significant.   Among such non-point  sources, urban runoff
 has been found to be particularly important.  For example, a recent study sub-
 stantiates that urban runoff can convey  high pollution loads, and under certain
 conditions can control water quality in  receiving waters for extended time
 periods.(25)

           Some water quality considerations  have to be recognized early in  the
 planning of urban drainage.   Such considerations are  related to pollution control
 at the source and include the planning of land  use, type of development and
 extent of natural drainage.

           Water-quality oriented design  of urban drainage  is somewhat hindered by
 a lack of understanding of runoff-quality processes and  by a lack of
 authoritative water  quality criteria for use  in such  a design.  Ideally, water
 quality criteria should be defined as receiving water criteria.  In that case,
 a given level of water quality (e.g., pollutant concentration) has to be
 attained during critical storms  and  under critical conditions in the  receiving
 water  body.   Extensive data  needs  and complex modeling of the environmental
 response retard wider application of receiving  water  criteria.  More  often,
 effluent control criteria are imposed in Canada.  Such criteria can be stated in
 terms  of annual allowable loads,  number  of events (e.g., annual average number of
 overflows from a combined sewer  system),  controls for specified events, or maximum
 allowable pollutant  concentrations,  or by prescribing specific control measures
 for an area  or at an effluent.(2°)

          Although the best  estimates of runoff quality  are obtained  from
 extensive local field  data,  such data are virtually nonexistent and quality
 simulation models have to  be  employed.   Such simulation  models again  require
 precipitation  input  data.  These data, however, differ from those used in the
 hydraulic design of  urban  drainage.  Before discussing the nature of  these
 differences, mention should be made  of the relative accuracy of quantity and
 quality  models.   In  each case, the uncertainty  in the simulated output depends
 upon the  uncertainties  in  the  input  and  in the  model  description of the processes
 which  transform inputs  into outputs.  Uncertainties in precipitation  inputs are
 comparable for  both  cases.  However, uncertainties in contaminant inputs and in
 the descriptions  of  the  actual processes  lead to far  greater errors in simulated
 quality outputs  than  in  simulated quantity outputs.

Antecedent Rainfall  Effects

          The magnitude  of pollution loads conveyed by urban runoff events depends
 strongly  on  the  initial  conditions of the event, which are thought to be
 characterized by  the amount of pollutant  accumulated on  the catchment surface
during the antecedent dry-weather period.  This point  is illustrated  in Figure 7,
which contains  runoff hydrographs and pollutographs from the Malvern catchment
 for two storms  of  similar volume and intensity of rainfall.  The two hydrographs
                                         -  182  -

-------
2 0
i 10
-§ 20
2 J


J]


•



                                    2  0
4
OC
   4
u.
O
   .2
400
200
    oc
    t-
    ui
    o

    8
U.
O
Z
          20    40    60

          TIME, MINUTES
              .2
                                                              800
                                                              400
z
UJ
U
Z
o
U
                     20    40   60

                    TIME, MINUTES
 lo) ONE-DAY DRY-WEATHER  PERIOD

                           Legend
                              SS
                             COD
          lb) SIXTEEN-DAY DRY-WEATHER PERIOD
             FIGURE 7-EFFECTS OF  DRY-WEATHER PERIOD,

                          MALVERN CATCHMENT
                                   - 183 -

-------
 agree quite well  but  the  storm with the  longer antecedent dry-weather period,
 Figure 7(b),  produced pollution loads about  five  times higher than those conveyed
 by the other storm, Figure  7(a).  Further  changes  in the magnitude of event
 pollution loads might be  introduced by street sweeping that effectively reduced
 the accumulation  of pollutants on the catchment surface between storms.

           One possible way  to  estimate the antecedent dry-weather period is to
 analyze the historical precipitation record  and to derive therefrom
 probabilities of  dry-weather durations for individual events.  An example of
 such a relationship is given in Figure 8.  For a catchment in Burlington, Ontario,
 a  five-year precipitation record was analyzed for  the duration of dry weather
 period considering only storms with rainfall sufficient to wash off the catchment
 surface (in this  case, storms with  rainfall  larger than 2.5-mm).  If the
 pollutant accumulation rates are known or  determined from street sweeping
 experiments,  the  total pollution loads per event and their probability of
 occurrence can be determined from Figure 8 by multiplying the accumulation rates
 by the number of  dry  days.  Note that this approach assumes that the distribution
 of dry periods does not depend on storm characteristics and that any antecedent
 storm would have  completely washed  off the catchment.  Such assumptions appear to
 be acceptable when considering the  large uncertainties involved in runoff quality
 computations.

 Discussion

           Under the circumstances described above, the design storm concept and
 single event  simulation are of limited use for water-quality oriented design.
 Storms  producing  high runoff flows  may produce relatively low pollution loads and
 vice versa.   Consequently, continuous simulation of runoff quality, quality
 control and associated costs should be employed using historical precipitation
 data.   Such simulations offer a good basis for the selection of the most cost-
 effective  control alternative meeting water quality criteria.  As for design peak
 flows,  discussed  earlier, continuous simulation might be reduced to indicator
 catchments, with  abbreviated simulations indicated thereby successively applied
 to  the more numerous other catchments in the jurisdiction involved.

           In summary,  water-quality oriented drainage design is a new idea which
 has not yet gained wide acceptance.  Quality considerations related to source
 control have to be undertaken in the planning process.   Among these considerations
 one could name land use,  type of development and extent of natural drainage.
 Precipitation data requirements for water-quality oriented design are virtually
 identical  to those for the planning of urban drainage.   Subsequently, the quality
design  is  finalized by means of detailed simulations for selected events of which
 the initial conditions have been determined.   The design storm concept has little
application in quality design because of statistical nonhomogeneity of runoff
 quantity and quality events.  Such nonhomogeneity is illustrated by the data
 plotted in Figure 7.  Typically,  historical rainfall events are used in quality
design.  Additional data,  such as rainfall chemistry and atmospheric fallout
rates may be useful.
                                         - 184 -

-------
                                    EVENT RETURN
          5  1      1/12       1/52   PERIOD, (YEARS)
          t    i       t
  25i
£20
o
o
DC 15
UJ
Q_
a:
UJ
x  _
ui
 i
cc  5
        I I  I  I  I  l  I   I   III ii I   I   ill
          l  l l  l   l  l   l  l l  I  I l  l   i  l  II
    0.1   .1    1      10       50      90    99  9999
             EXCEEDANCE  PROBABILITY (%)
  FIGURE 8-  PROBABILITY OF DRY-WEATHER  PERIODS,
                  BURLINGTON, ONTARIO
                            ' 185 -

-------
                                   SECTION 6

                                  REFERENCES


  1.   Marsalek,  J.,  "Runoff Control on Urbanizing Catchments," pp. 153-161 in
      Proceedings, Amsterdam Symposium on Effects of Urbanization and
      Industrialization on the Uydrological Regime and on Water Quality,
      October, 1977,  International Association of Hydrological Sciences, IAHS-
      AISH Publication No. 123, 1977.

  2.   Watt, W. Edgar, and Jiri Marsalek, "What the Practising Urban Hydrologist
      Needs from the Hydrometeorologist," pp. 15-23 in "Preprint Volume,"
      Second Conference on Hydrometeorology, October 25-27, 1977, Toronto,
      Ontario, American Meteorological Society, Boston, Mass.

  3.   Marsalek,  J., "Comparison of Runoff Simulations for Actual and Synthetic
      Storms," pp. 100-116 in special release of "Storm Water Management
      Modeler,"  the news release bulletin of the SWMM Users Group, U.S. EPA,
      Washington, D.C., January, 1978.

  4.   Marsalek, J., "Synthesized and Historic Design Storms for Urban Drainage
      Design," a paper presented at the 11-15 April 1978 International Conference
      on Urban Storm Drainage, University of Southampton, England.

  5.   McPherson, M. B., "Special Characteristics of Urban Hydrology," pp. 239-255
      in Prediction in Catchment Hydrology. Australian Academy of Science,
      Canberra, ACT, 1975.

  6.   Keifer,  Clint J., and Henry Hsien Chu, "Synthetic Storm Pattern for
      Drainage Design," J.Hyd.Div.. ASCE Proc., Vol. 83, No. HY4, pp. 1-25,
     August,  1957.

  7.  Terstriep, M. L., and J. B.  Stall,  The Illinois Urban Drainage Area
     Simulator. ILLUDAS.  Bulletin 58, Illinois State Water Survey, Urbana,
     90 pp.,  1974.

 8.  Bandyopadhyay, M.,  "Synthetic Storm Pattern and Run-off for Gauhati,
     India,"  J.Hyd.Div..  ASCE Proc.,  Vol.  98,  No. HY5, pp. 845-857,  May, 1972.

 9.  Preul,  H. C., and C. N.  Papadakis,  "Development of Design Storm Hyetographs
     for Cincinnati, Ohio," Water Resources Bulletin. Vol. 9, No. 2, pp. 291-300,
     1973.

10.  Mitci, Constantin,  "Determine Urban Runoff the Simple Way." Water & Wastes
     Engineering.  Vol. 11,  No.  1,  pp. 24-26 and 35-36, January,  1974.

11.  Design  and Construction of Sanitary and Storm Sewers. ASCE Manuals and
     Reports  of Engineering Practice  No. 37 (WPCF Manual of Practice No. 9),
     ASCE, New York, N.Y.,  332  pp.,  1969.

12.  McPherson,  M. B., "Discussion of Synthetic Storm Pattern for Drainage
     Design," J.Hyd.Div.. ASCE  Proc., Vol.  84,  No. HY1,  pp. 49-57, February, 1958.

13.  M. M. Dillon  Ltd.,  "Storm  Drainage  Criteria Manual for the City of Burlington,"
     unpublished report,  Toronto,  Ontario,  1977.

                                        -  186 -

-------
14.  Huff, F. A.,  "Time Distribution of Heavy Rainfall in Storms," Water
     Resources Research. Vol. 3, No. 4, pp. 1007-1019, 1967.

15.  Vogel, John L., and Floyd A. Huff, "Heavy Rainfall Relations Over Chicago
     and Northeastern Illinois," Water Resources Bulletin. Vol. 13, No. 5,,
     pp. 959-971,  October, 1977.

16.  Jewell, T. K.,  T. M. Nunno and D. D. Adrian, "Methodology for Calibrating
     Storrawater Models," J.Env.Engrg.Div.. ASCE Proc., Vol. 104, EE3, pp. 485-
     501, June, 1978.

17.  Field, Richard, and John A. Lager, "Urban Runoff Pollution Control —
     State-of-the-Art," J.Env.EngrR.Div.. ASCE Proc., Vol. 101, No. EE1,
     pp. 107-125,  February, 1975.

18.  Torno, Harry C., "Storm Water Management Models," pp. 82-89 in Urban
     Runoff. Quantity and Quality. ASCE, New York, N.Y., 1975.

19.  Proctor and Redfern, Ltd., and James F. MacLaren, Ltd., Storm Water
     Management Model Study. Volume II, Research Program for the Abatement of
     Municipal Pollution under Provisions of the Canada-Ontario Agreement  on
     Great Lakes Water Quality, Research Report No. 48, Environment Canada,
     Ottawa, 148 pp., September, 1976.

20.  Marsalek, J., Malvern Urban Test Catchment. Volume I,  Research Program  for
     the Abatement of Municipal Pollution under  Provisions  of  the Canada-Ontario
     Agreement on Great Lakes Water Quality, Research Report No. 57,  Environment
     Canada, Ottawa, 55 pp., 1977.

21.  Siegel, S., Nonparametrlc Statistics  for the Behavioral Sciences.  McGraw-
     Hill, New York, N.Y., 1957.

22.  Linsley, R., and N.  Crawford,  "Continuous Simulation  Models  in Urban
     Hydrology," Geophysical Research Letters. AGU,  Vol.  1, pp.  59-62,  May,  1974.

23.  McCuen, Richard H.,  "A  Regional  Approach to Urban  Storm Water Detention,"
     Geophysical Research Letters. AGU,  Vol. 1,  No.  7,  pp.  321-322, November,
     1974.

24.  Chow, V. T., Editor, Handbook of Applied Hydrology.  McGraw-Hill, New  York,
     N.Y.,  1964.

25.  Colston, Newton  V.,  and Anthony N.  Tafori,  "Urban Land Runoff Considerations,"
     pp.  120-128  in Urbanization and Water Quality Control. American Water
     Resources Association,  Minneapolis,  Minnesota,  1976.

26.  Weatherbe, D.  G.,  "Urban  Drainage Planning," Chapter 2 in Manual of
     Practice on  Urban Drainage (Draft),  Canada-Ontario Agreement on Great
     Lakes Water  Quality, Environment Canada, Ottawa, March,  1977.
                                          -  187 -
                                                    «U.S. GOVERNMENT PRINTING OFFICE: 1078 620-007/3737 1-3

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instruction* on the reverse before completing)
1. REPORT NO.

 EPA  -  600/9  - 78 - 035
              3. RECIPIENT'S ACCESSION>NO.
4. TITLE AND SUBTITLE
                                                          5, REPORT DATE
                                                            October 1978
 Urban Runoff Control  Planning
              6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 Murray  B.  McPherson
              8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 ASCE Urban  Water Resources Research Program
 23 Watson Street
 Marblehead,  Mass.   01945
              10. PROGRAM ELEMENT NO.
               \. £6NtRACT/GRANT NO.


                 N/ A
12. SPONSORING AGENCY NAME AND ADDRESS
 U.S.  Environmental  Protection Agency

 Office of  Research  and Development
 Washington,  D.C.  20460
              13. TYPE OF REPORT AND PERIOD COVERED
              14. SPONSORING AGENCY CODE
                EPA 600/Z
 B. SUPPLEMENTARY NOTES
      Section  208 of Public Law 92-500  (Federal  Water Pollution Control Act  of 1972)
 encourages  areawide planning for water  pollution abatement management, including
 urban  runoff  considerations where applicable.   Areawide studies are under way or
 planned  in  just about every metropolitan  area.   Deadlines for initial areawide
 reports  are not far off, and it is expected  that many of the agencies preparing
 reports  are presently resolving their  projected activities beyond the current
 first  planning phase.  This report has  been  prepared to assist agencies  and their
 agents that are participants in the preparation of areawide plans, from  the stand-
 point  of major urban runoff technical  issues in long-range planning.  Emphasized
 Is the importance of conjunctive consideration  of urban runoff quantity  and quality
 and the  need  to development a factual  basis  that will support expected reliability
 of performance of proposed actions and  programs.  While not intended as  a handbook
 for urban runoff control planning, this report  delves into some important technical
 Issues that are often slighted or poorly  handled, such as the utilization of simu-
 lation.  Recognizing that the ultimate  test  of  any plan lies in its implementation,
 topics are  viewed from the perspective  and experience of the local government level
 where  Implementation takes place.
                               Key WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED.TERMS  C. COSATI Field/Group
Urban  Runoff
Master Planning
 Urban Hydrology
 Storm Water  Runoff
 Combined  Sewer Overflows
                                                                          13 B
Release  to  Public
                                             19. SECURITY CLASS (This Report)
                                              Unclassified
                           21. NO. Oi- PAGES
20. SECURITY CLASS (Thispage)
 Unclassified
22. PRICE
      aaao-t it-73)

-------