ANALYSIS OF
                WATER AND WATER-RELATED
                 RESEARCH REQUIREMENTS
                IN  THE GREAT LAKES  REGION
              June 1968
     This report is prepared for the
   Director, Office of Water Resource
  Research, Department of the Interior,
   Washington, D.C., in compliance
     with Contract 14-01-0001-1571.
Council on Economic Growth, Technology, and Public Policy
of the Committee on Institutional Cooperation

-------
    Sponsorship and Authorship.  The work reported herein was conducted by
the Council on Economic Growth, Technology, and Public Policy of the Commit-
tee on Institutional Cooperation for the Office of Water Resource Research,
 Department of the Interior, under Contract 14-01-0001-1571 with The Uni-
 versity of Michigan.  This report is being published by The University of
 Michigan in compliance with this contract. It was derived from the delibera-
 tions and written contributions of the individual investigators listed and the
 Project Steering Committee.  The responsibility for the preparation of the
 report rests entirely with the Project Steering Committee.

-------
ANALYSIS OF WATER AND WATER-RELATED
      RESEARCH REQUIREMENTS IN THE
            GREAT LAKES  REGION
        Council on Economic Growth, Technology,

          and Public Policy of the Committee

             on Institutional Cooperation
                        U.S. Environmental Protection Agency
                        Region 5, Library (PL-12J)
                        77 West Jackson Boulevard, 12th Flew
                        Chicago, IL 60604-3590

-------
                        PROJECT ORGANIZATION

Carlisle P. Runge                     Professor of Law, University of Wisconsin
 Director, Council on Economic
 Growth,  Technology, and Public
 Policy of the Committee on
 Institutional Cooperation
James T.  Wilson                      Director, Institute of Science and Tech-
 Chairman, Water Resource Com-       nology, and Professor of Geology, The
 mittee, Council on Economic           University of Michigan
 Growth,  Technology, and Public
 Policy
                   PROJECT STEERING COMMITTEE

William C. Ackermann                Chief, Illinois State Water Survey, and
  Chairman                             Professor of Civil Engineering,
                                       University of Illinois

Lyle E. Craine                       Chairman, Water Resources Committee,
                                       and Professor of Conservation, The
                                       University of Michigan

Richard D. Duke                      Director, Environmental Simulation
                                       Laboratory, and Professor of Regional
                                       Planning, The University of Michigan

J. W. Milliman                       Director, Institute for Applied Urban
                                       Economics; Deputy Director, Indiana
                                       Water Resources Center;  and Pro-
                                       fessor of Business Administration,
                                       Indiana University

Clifford H. Mortimer                 Director, Center for Great Lakes
                                       Studies, and Distinguished Professor of
                                       Zoology, University of Wisconsin at
                                       Milwaukee

Gerard A. Rohlich                    Director, Wisconsin Water Resources
                                       Center,  and Professor of  Civil
                                       Engineering, University of Wisconsin
                            PROJECT STAFF

William D. Drake                     Associate Professor of Regional Planning,
  Staff Coordinator                      The University of Michigan

Spenser W. Havlick                   Assistant Professor of Conservation, The
  Staff Editor                           University of Michigan

James E.  Kerrigan                   Assistant to the Director, Wisconsin
                                       Water Resources Center, University of
                                       Wisconsin

Dale D. Meredith                     Instructor in Civil Engineering, University
                                       of Illinois
                                    ii

-------
                                  ABSTRACT
    Investigators interested in water and water-related research in the Great
Lakes region have considered applicable research requirements and methodologies,
with special emphasis on the applications of systems analysis and modeling.
Representatives of many disciplines, from major midwestern universities, water
resources centers, and federal agencies, met in two working conferences and on
numerous other occasions to discuss a framework for research activities which
appear necessary to comprehensive water management and related development in
the Great Lakes system. Under the auspices of the Committee on Institutional
Cooperation (CIC), the research requirements have been appraised on a region-
wide scale, and this, with an appropriate mechanism for coordination, could pro-
mote  a unique research  collaboration among disciplines and among universities.

    This report indicates the focus placed by researchers of  many disciplines
upon a systems analysis model of the Great Lakes. Early in the study it was
determined that a water-quantity model of the entire system is necessary and
feasible. Attempts at a water-quality  model for the Great Lakes region on a sub-
regional, subsystem basis,  with subregional groupings anticipated as available
data and systems technology permit, are also presented.  The need for  a regional
economic-growth model, water-related information systems,  and a gaming-
simulation model for research on relevant institutions is described.  The research
efforts to supplement and support the water-quantity and water-quality  subsystems
are specified, and priorities among these are suggested.  The appendixes to the re-
port contain papers contributed to the  study, proceedings of the working conferences,
names of conference participants, a listing of responses by conference participants
and their colleagues to a questionnaire on research activities needed in the  Great
Lakes region, and other supplementary materials.
                                     iii

-------
                                  CONTENTS

 Project Organization	  ii

 Abstract	iii
 List of Illustrations	vi
 1. Introduction	  1

 2. Summary and Conclusions	  1

 3. Results to be Achieved from a Regional Analysis of Great Lakes Water
      Research 	  2
      3.1. Guidance for Program planners and Managers                       3
      3.2. A Fuller Context to Which Researchers May Relate Their Work        3
      3.3. Valuable Communication among Individual Researchers and
            Groups Working in this Area                                      3

 4. Summary of the Two Working Conferences	   4
      4.1. The First Working Conference                                      4
      4.2. The Second Working Conference                                     4
 5. Major Water Subsystems and Associated Socioeconomic and Institutional
     Research 	   5
      5.1. The Water-Quantity Subsystem                                      6
      5.2. The Water-Quality Subsystem                                       6
      5.3. A Regional Economic and Demographic Growth Model                11
      5.4. Studies of Social Factors                                          12
      5.5. Studies of Institutions                                             14
      5.6. Data Requirements                                                17
      5.7. Survey of Needed Research Projects in the Great Lakes Region       18
      5.8. A Mechanism for Coordination of Water-Related Research by
            Universities in the Great Lakes Region                           19

 Appendix A: A Report on the First Working Conference:  Consideration of
            Great Lakes Systems  Research, 10-15 September 1967,
            Alpine Valley, Elkhorn,  Wisconsin	23
 Appendix B: A Report on the Second Working Conference:  Consideration of
            Great Lakes Systems  Research, 30 and 31 October 1967, The
            University of Michigan,  Ann Arbor,  Michigan	51

 Appendix C: Water Systems Modeling
            Dale D. Meredith, University of Illinois	61
 Appendix D: A Preliminary Description of Water-Related Information
            Systems for the Great Lakes Region
            Thomas E. Borton and William D. Drake, The University of
            Michigan	78
Appendix E: Preliminary Research Design of the Water-Quantity  Subsystem
            of the Great Lakes
            Rolf A. Deininger, The University of Michigan	81
Appendix F: Preliminary Research Design for the Water-Quality  Subsystems
            of the Great Lakes Region	87

            A.  The Proposition for  a Lake Model
                John C. Ayers, The University of Michigan                    87

            B.  The Proposition for  a Sublake Model
                James E. Kerrigan,  University of Wisconsin                   88

Appendix G: Preliminary Research Design for the Gaming Simulation
            Richard D. Duke, The  University of Michigan	97

Appendix H: An Inventory  of Needed Research Activities in the Great Lakes
            Region
            Compiled by Spenser W. Havlick, The University of Michigan .  . .  100

-------
                               ILLUSTRATIONS
  1.  "Black Box" Model of Average Inputs and Outputs of the
        Great Lakes System	   7
  2.  Initial Subdivided Model of Lake Michigan	   9
  3.  Spatial Relationships Between Local Lake Sectors and
        Associated Water Uses	10
A-l.  A Problem-Solving Approach	47
A-2.  Ayer's Schematic Model	47
B-l.  Schematic for Regional Analysis	58
E-l.  Input and  Output Model   	83
F-l.  Initial Subdivided Model of Lake Michigan	89
F-2.  Free-Body Diagram of Sublake Model	91
F-3.  Block Diagram of Sublake Model	91
F-4.  Modular Arrangement for Sublake Model   	92
F-5.  Time-Dependent Relationships for Sublake Model	95
G-l.  Simplified Model of a Subregion	99

-------
                                             1
                                        INTRODUCTION
    The water-resource problems of the Great Lakes region* are large, diverse, and urgent. Un-
der this contract, the Council on Economic Growth, Technology, and Public Policy of the Committee
on Institutional Cooperation sought to provide program managers of the Office of Water Resource
Research, Department of the Interior, with the guidance essential to effectively  executing the water-
resources research and training program authorized by the Water Resource Research Act of 1964
as amended.  A report identifying water and water-related research requirements in the Great
Lakes region and specifying desirable priorities and schedules for such research was to be the
vehicle for this guidance.
    The procedure followed in preparing the report was to hold two working conferences of people
knowledgeable and experienced in work concerned with the Great Lakes region and to have a project
staff compile the report from the activities and discussions of the two conferences.
    The first working conference was held at Alpine Valley, Elkhorn, Wisconsin, from 10 through
15 September 1967.  Its purpose was to consider the feasibility of developing comprehensive re-
search projects using systems analysis techniques to solve problems affecting the Great Lakes re-
gion.  Some of the problem areas in which systems analysis  could be appropriately used were to be
identified, as were the fundamental investigations required to support these efforts.  The second
working conference was held in Ann Arbor, Michigan, on 30 and 31 October 1967 to consider the
personnel needs and priorities associated with the  investigations identified during the first working
conference.
    A steering committee met periodically before  and during both major conferences to plan and
evaluate the working groups and discussions of the conferences.  This committee was the  main
policy-making body for this program and was vested with the responsibility for general guidance of
the work throughout the contract period.  The membership of the steering  committee was  kept flex-
ible in order to accommodate particular interests and capabilities brought out during the working
conferences.

                                             2
                               SUMMARY AND CONCLUSIONS
    In the light of the previously documented seriousness of water-related problems in the Great
Lakes region, this  multidisciplinary, collaborative effort explored research topics and methodologies
     "Region" here refers to the influenced service area and is not limited to the watershed per se.

-------
which appear necessary and appropriate for universities and other concerned organizations and
groups with a research perspective which can be focused on this region.  The major findings which
emerged frpm the conferences and small group work sessions over a period of six months may be
summarized as follows:
    (1)  Subsystem modeling and related water-resources  research should proceed as soon as is
practicable in each geographic sector of the Great Lakes region with the long-range objective of ob-
taining a region-wide comprehensive model, even though such a model is not feasible at this time.

    (2)  Systems analysis research in at least three aspects,  (a) a water-quantity and lake-level
model, (b) a water-quality model, and (c) an  economic-growth model, is needed for water resources
management in this region.  The relative difficulty involved in each of the proposed models clearly
emerged during this  study, and past and present efforts were  determined to be inadequate in terms
of the innovative experimentation necessary.
    (3)  Data storage and information retrieval should be coordinated, with all public and nonpublic
agencies involved in  water and water-related research activities in this region cooperating, to
promote operational  and  research efficiency and as a basis for planning on a region-wide scale.
    (4)  Analysis and evaluation of the effectiveness of institutions responsible for water-manage-
ment decisions are required. Criteria for appraisal and recommendations with regional perspectives
need to  be established for improved water-resource decision  making.
    (5)  Pioneering research is called for on the socioeconomic ramifications of water-management
decisions at the various governmental and private levels.
    (6)  Increased collection and analysis of data in physical, chemical, biological, and social areas
is recommended.
    (7)  A mechanism to coordinate water-resources  research among academic institutions in this
region and between them and relevant federal and state agencies is needed.
                                             3
                               RESULTS TO BE ACHIEVED FROM A
                     REGIONAL ANALYSIS OF GREAT LAKES WATER RESEARCH
    Until now, research related to water resources in the Great Lakes region has been approached
primarily by several disciplines working independently, with little collaboration among the region's
universities. Any effective procedures for interuniversity cooperation which are developed for
this region may be transferable to other regions.  A regional analysis should also provide guidance
for planners and enable researchers to see their individual contributions in the context of development
and management of the entire region. It should suggest valuable means of  communication among rel-
evant private and public (community, state, national, and international) bodies.
                                             2

-------
3.1.  GUIDANCE FOR PROGRAM PLANNERS AND MANAGERS
    A comprehensive systems analysis model must explicitly treat and interrelate many diverse
variables.  There is no one goal taking precedence over all other considerations for the Great
Lakes region, but, instead, a mixture of important objectives which must be taken into account.
Some of these objectives are complementary, but some are incompatible.  Models which consider the
many feedbacks and time-dependent factors involved would allow a planner to investigate and compare
the effects of alternative policies, which is especially important in the Great Lakes region where  many
different policies already interact.  Such a model would also allow planners and administrators to ob-
serve the behavior of the system over a period of time when subject to their  policies and water-
use projections. If the water-use projections are developed endogenously, then the planner can use
the model to test the sensitivity of the assumed relationship between water use and other variables.
A gaming-simulation or role-playing model would allow a planner to be an intimate part of the model
himself, to make dynamic decisions and react to the impacts  these have upon the system, and to see the
alternative decisions available to him.

3.2. A FULLER CONTEXT TO WHICH RESEARCHERS MAY RELATE THEIR WORK
    A comprehensive systems analysis framework would enable researchers in various disciplines
to see where their work fits into the overall  system and the kinds of results necessary  in terms of
information needed by the model.  It could point up cases where only slight modification to research
in progress would make this work useful to the model. The system-wide analysis should also indicate
the weakest areas of the model, allowing researchers to give these  areas highest priority. As results
are obtained the model could be reformulated and refined.  Thus, the areas requiring greatest effort
would always be kept prominently before researchers and those who fund research.

3.3. VALUABLE COMMUNICATION AMONG INDIVIDUAL RESEARCHERS AND GROUPS WORK-
     ING IN THIS AREA
     Since a multidiscipline approach involving information and data exchanges uncommon to a region
of this  size will be required to develop a comprehensive  systems model of the Great Lakes region, this
will serve to open communications between researchers  in various disciplines.  It  should also open
channels of communication between  individuals and groups concerned with the Great Lakes region be-
cause of the broad and diverse cooperation that will be necessary.
     One requirement of a comprehensive systems analysis  model is for consistent and uniform pre-
sentation of data. Extensive communication among those working on each research project in the  sys-
tem will be essential to obtain and report data in this manner. This, in turn, can lead to the establish-
ment of means for effectively disseminating information to planners and policy  makers as well as re-
searchers.

-------
                                             4
                       SUMMARY OF THE TWO WORKING CONFERENCES
    Two major "study sessions" or working conferences were carried out as recommended in this
contract.  Knowledgeable investigators from the CIC universities and from state water resources
centers and additional systems designers from private industry and nonprofit organizations attended
these conferences.  Papers were presented by these participants and by representatives of selected
federal agencies. High-level personnel from the Office of Water Resource Research of the Depart-
ment of the Interior, the Water Resources Council, and the Great Lakes Basin Commission were
present at both working conferences.  Appendixes A and B are detailed accounts of the conferences
which are summarized here.

4.1.  THE FIRS T WORKING CONFE RE NC E
    This five-day study session held at Alpine Valley, Wisconsin, succeeded in bringing together
researchers and program managers concerned with water problems of the Great Lakes region.
After it was decided that water-quality research could best be implemented on a subregional basis,
the necessary and feasible research projects in modeling and systems analysis were discussed.
Steps were also taken to begin design of a water-quantity model for the entire Great Lakes System.
    Multidisciplinary work groups at this conference set about identifying particular areas for
water-resources research. Presentations from federal agencies detailed data collection,  analysis,
and dissemination capabilities and current programs in the Great Lakes basin. It was generally
agreed that the Lake Michigan basin should serve as one major focus for  research efforts.  The
need to design the relevant subsystem models in a way consistent with the eventual evolution of a
total system model was emphasized.  The potential research projects identified at this first confer-
ence were placed in two broad categories for consideration, physical-biological research and
 institutional-socioeconomic research, with understandable overlap in many cases. Once major
 areas of research had been considered, the task of examining specific research requirements for
 the Great Lakes basin remained for the second working conference.

4.2. THE SECOND WORKING CONFERENCE
    One of the highlights  of the two-day session in Ann Arbor was the decision to focus on the
physical,  economic, political, and biological relationships and possible tradeoffs between waste-
water  treatment and raw water treatment in a given sector of a subregion under study.  This re-
search program would involve (1) the necessity to enlist several CIC universities for each subregion,
i.e., southern Lake Michigan, western Lake Superior, or the southern half of Lake Erie, (2) the  all-
important consideration of present and future water-supply requirements and associated costs,  (3)

-------
the cooperation of industrial and municipal interests, and (4) the collaboration of natural scientists,
social scientists, engineers, systems analysts, and other specialists in an effort of mutual concern.
    A regional economic-growth model, gaming simulation as a research and learning tool, refine-
ment of the water-quantity and water-quality models, and associated requirements for institutional -
social research were major topics of discussion at this conference.  A paper on the state of the art
in system model building was presented (see Appendix C).  The  purpose and structure of nine sys-
tem models and results using them were reviewed.  Eight of these models are mathematical models
evaluated by computer, while one is a. gaming-simulation model in which players are  used in roles to
approximate actual conditions.  Some examples of the types of subsystem models that have been con-
structed were also presented. This review was concluded with a statement on the characteristics re-
quired of a model for the Great Lakes region and a suggested approach for formulating such a model.

    Near the end of the second working conference it was urged that the research capabilities of
the institutions and agencies represented be itemized. An outgrowth of this evaluation was an in-
ventory of projects and a statement of major proposals deemed  important for a systematic manage-
ment strategy in the  Great Lakes region.  Appendixes D through G are research design papers put
into their present form at a drafting session held in Milwaukee, Wisconsin, 17 November 1967.
Investigators interested specifically in institutional and social systems research met in Madison, Wis-
consin,  on 15 March  1968.  In collaboration with investigators who had attended  earlier sessions,
Norman Wengert, Professor of Political Science, Wayne State University, and Keith Warner, Professor
of Rural Sociology, University of Wisconsin, contributed substantially  to these deliberations on socio-
political research requirements (see sections 5.4 and 5.5).

                                             5
                       MAJOR WATER SUBSYSTEMS AND ASSOCIATED
                      SOCIOECONOMIC AND INSTITUTIONAL RESEARCH
    The physical system under consideration is composed of the area covered by the Great Lakes
and their contiguous, contributing subwatershed areas. The surface drainage area is well defined
and serves as a convenient starting point for such study;  This physical system was approached as
two subsystems.  A model of one of these,  the water-quantity subsystem, was determined to be
feasible for immediate preliminary design and construction on a region-wide scale.  Modeling of
the water-quality subsystem was determined to be feasible for one or more subregions within the
Great Lakes region.  It was further judged that both subsystem  models, whether formulated on a
region-wide basis or subregionally, would have political  and social characteristics that should be
considered for research.

-------
    While separating the quantity and quality modeling efforts is helpful, it was recognized that
there are important linkages between the two subsystems. A better understanding of these might be
obtained by studying an element such as chloride over a period of time for the entire Great Lakes sys-
tem. This type of study could provide information about such things as the flushing  rates and the di-
rection of movement through the system by the element being studied.

5.1.  THE WATER-QUANTITY SUBSYSTEM*
     The variation in levels of the Great Lakes affects  shoreline Interests, navigation interests, and
hydropower interests. At present, two of the lakes, Superior and Ontario, are regulated by con-
trolling the outflow according to rule curves  developed by the U. S. Army Corps of Engineers and
modified many times  since they were first developed.  While reducing high lake levels and increas-
ing low levels is certainly one step in regulation, a foot reduction in high levels should not be con-
sidered to have the same associated values as a foot increase in low levels.   Economic data should
be used to properly weigh the two alternatives against each other so that operating rules minimizing
losses or maximizing benefits can be determined.
     Several new techniques generally summarized under the name systems analysis are presently
feasible and ideally suited for developing and exploring new methodologies for optimal control of lake
levels.  It is proposed to investigate  the applicability and  limitations of these techniques in determin-
ing optimal operating policies for the Great Lakes system.  In cooperation with this investigation, it
is proposed to evaluate the effects of (1) land management on runoff, (2) precipitation increases, and
(3) evaporation reduction (cf. Figure 1). To determine  the difference between present day operating
rules and any  new ones derived in the study it will be necessary to construct  a mathematical simu-
lation model of the Great Lakes which will permit testing  of these alternatives. This will be done in
close cooperation with the U. S. Army Corps of Engineers which is presently  in charge of regulating
the Great Lakes.

5.2. THE WATER-QUALITY SUBSYSTEM**
     The difficulties of developing an effective water-quality model for the Great Lakes region are
considerable;  however, the successful application of systems analysis in several complex problem
areas during recent years justifies undertaking this major effort.  The methodology for handling
broad environmental studies by this  means is presently under development, with much of the tech-
nology adapted from uses of systems analysis in industrial, health, defense, and space research pro-
grams.  Recent studies have designed water-quality models for rivers and estuaries to assist in the
selection of operating policies for such systems, and, although lake problems are different from
those of rivers, the methodologies for analyzing both kinds of system are likely to be complementary.
     *This section constitutes a summary of Appendix E.
    **This section constitutes a summary of Appendix F.
                                              6

-------
                        RUNOFF FROM DRAINAGE AREA
FIGURE 1.  "BLACK BOX" MODEL OF AVERAGE INPUTS AND OUTPUTS OF THE GREAT
LAKES SYSTEM.  The data are from the U. S. Lake Survey and are in thousands of cubic feet
per second. These averages are for the period 1950 to 1960 and are based on keeping the
                                  lakes level.

-------
    Because of the complex nature of the numerous elements involved, tt will be useful to approach
the water-quality subsystem from several points of view.  It was determined to try two  approaches,
a lake  model and a sublake model,  initially.  The lake-modeling approach is designed to quantify
the relationship among local areas lake by lake.  The sublake approach seeks to describe the nature
and alternative courses of action within a localized sector,  taking into account the effect of neighbor-
ing sectors.  The two approaches  are not separate and distinct, and results with the sublake model
will allow refinement of the lake model; since the two models will thus be  complementary, they must
be developed simultaneously.
    The lake model is to describe the water quality by subregions within each of the Great Lakes
as a function of water use requirements and the quality of water put into the lake, whether by
natural runoff or return from some  use.  Each subregion is to be  chosen so as to be relatively
homogeneous in its properties, to receive local inputs, and  to provide local withdrawal.  Figure 2
presents, as  an example, the initial  subdivided model of Lake Michigan. This type of subdivision
seems applicable for spring, summer, and fall, though in winter many of these boundaries tend to
disappear.  Information on  the physical nature of currents in each lake and the physical, chemical,
and biological transformations that accompany the currents in each subregion of the lake  must be
collected. Also, the transfer functions which apply to the exchange of water and material between
subregions must be determined.
    The sublake model will not only complement the lake-modeling effort,  but also an economic-
growth model. The sublake model is indicated as a  local sector in Figure  3, which shows the spatial
relationships between the local lake sectors and associated water uses (see Appendix F for def-
initions of the symbols used).  The uses may be such as for municipal or industrial water supply
or for  recreational purposes.  The location, amount, and quality of water put into and withdrawn
from each of the sectors  must be  determined. For some uses  it will be possible to employ con-
trols on the water quality before or after the water is used, while other uses will allow no con-
trol measures.
    Obviously, information must be exchanged between  the two modeling efforts.  The sublake
model must use  the transfer functions between lake subregions and within  each  subregion as deter-
mined from the lake model. The  lake model will  require the location, amount,  and quality of influent
and effluent water for each subregion as determined from the sublake model.  It will be necessary
to identify the water uses in each sector along with  the water-quality limits associated with each
use. After these limits have been established for each quality parameter, studies must be conducted
to determine the methods of water and waste treatment, management, enforcement, etc. that may
be employed to control the  level of water quality within  the limits for the local  sector.  The lake
model will then integrate these sublake models to determine their interdependence and effect on
the entire lake.  The costs  and benefits  associated with  controlling the level of water quality in
selected local sectors, not  only the  direct costs and benefits normally identified but also  those
social  values which are usually not  estimated, should be studied and evaluated.  With a comprehen-
sive assessment of the resources available, the use demands, ideal and practical institutional and

-------
                           \
FIGURE 2. INITIAL SUBDIVIDED MODEL OF LAKE MICHIGAN.
Inputs, withdrawals, and internal transfers under mean conditions
are indicated.  The internal boundaries change seasonally and are
           long-term averages from observations.

-------
              Uk/^tX
                              A4
FIGURE 3. SPATIAL RELATIONSHIPS BETWEEN LOCAL LAKE SECTORS
             AND ASSOCIATED WATER USES
                       10

-------
legal constraints, and the interactions of the system components, alternative procedures for con-
trolling water quality to meet selected social goals  may be suggested.
    It was decided that the major initial effort on the water-quality subsystem would be directed
toward Lake Michigan because of the urgency and magnitude of the problems involved, the rela-
tive isolation of this lake from the other lakes,  and  the interest of available personnel.  It  is an-
ticipated that other individuals and universities will develop the interest necessary to become in-
volved in the study of the other lakes.

5.3. A REGIONAL ECONOMIC AND DEMOGRAPHIC GROWTH MODEL
    While the study of difficult questions such as water quality may require first attempts on a
subregional or local scale, there is strong evidence that final results must be viewed in a  broader
context such as that of an overall regional economic-growth model. Such a model can serve to
unite local problem areas and to provide an appropriate framework for judging the impact of al-
ternative public actions.  It is the essence of the joint effort of the CIC for improving Great Lakes
management that decisions made in a local framework will tend to be less optimum than those
made on a regional basis, and a  regional economic-growth model is necessary for providing the
regional viewpoint.
    In relation to the particular  water-quality model sketched above, a regional  economic-growth
model can serve many purposes. By incorporating  the local water-quality sector in the growth
model, it will be possible to judge not only the feedback from the economy to water, but also the
feedback from water to the economy.  It is widely agreed that sound economic projections  are
needed as a basis for Great Lakes'  management, but no one has constructed a model showing the
feedback from the costs for various lake levels and  water qualities to the region's economic growth.
Yet, this information would be of major importance  in gaining the perspective needed for influencing
economic  growth in the region.   Moreover, while the water-quality link between lakes may be small,
the alternative uses of the lakes  should be tied to a  model of the region's economic growth. It
would be incorrect, for instance, to consider the recreational benefits of water-quality management
in one local  area independently of the total benefits from water-quality management within the area
and within the lake system  as a whole. It is only possible to judge all of the alternative uses of the
lakes within the broad framework of projected population-and economic activity.
    A regional economic-growth model would also have several uses not directly connected with
the water-quality model.  It should be useful in establishing the overall framework for the  water-
quantity model, especially if the  feedback between the  water quantity and the economy is found to
be significant.  More importantly, the model should  be helpful in making explicit the interdependences
for economic growth, that exist among such policy areas as water-resource management and urban
and transportation development.
                                            11

-------
    While the construction of such a model has not gone far enough to permit a statement of its
form, there has been enough consideration to identify some problem areas. For one thing, it is
usually accepted that a general regional economic model considering all relevant questions is an
impossibility.  Decisions must be made on exactly which questions this model will answer.  Also,
as has been indicated,  the Great Lakes region is indeed a combination of a large number of subre-
gions.  Since it is again generally  agreed that there are no "all-purpose" regions, determining the
appropriate subregions for consideration may be difficult.  Furthermore, the  Great Lakes region
constitutes such a large part of the national economy that national economic projections can  not be
regarded as external to this area of investigation.  It will be necessary to specify more completely
the effects of economic growth in the Great Lakes region upon the national economy and subsequently
back upon the region itself.
    Though the economic-growth model will obviously be a large undertaking, it is necessary if the
various CIC research efforts are to have a "specific orientation towards enhancing the means and
capacity for influencing the direction of the regional change." This model  investigation is now being
considered under a separate CIC research project.
5.4. STUDIES  OF SOCIAL FACTORS
    Issues of water quantity, quality, and use ultimately must be resolved by  reference to the in-
terests and actions of people. It is people's use of water resources that is the fundamental basis
of society's concern with the quantity and quality of these resources; water performs  essential
and important functions for people, and the interests, actions, and distributions of people are im-
portant determinants of water quantity and quality. Broadly defined, social factors may be studied
by political scientists, economists, legal scholars, and  social psychologists as well as by sociol-
ogists.  Each discipline will bring somewhat different perspectives and emphases to these studies,
and these must eventually be integrated as research findings are applied to the solution of practical
problems.  This section is intended to suggest some particular areas for needed research by soci-
ologists, political scientists, and others.  The next section, on institutional studies, provides a
framework for integrating diverse social science studies by focusing on water resource institutions.
    First, research to facilitate orderly setting of goals and establishing of priorities for water
resources in the context of society is needed. Although it is not the proper function of the social
sciences to establish goals and values,  systematic study can help to determine the consequences of
attaining or failing  to attain alternative goals and the relationships of decisional processes to end
results. This information can provide the basis for a more rational selection of goals and priorities.
    Research on the social consequences of policies, programs, and technological developments
related to water resources is also necessary.  The effects of such  factors on  patterns of community
growth  and concentration of population, location of industry (and therefore employment), recrea-
tional facilities (especially as metropolitan population pressures increase), and other social
                                             12

-------
 programs (e.g., by preempting tax dollars) are representative of such consequences. Studies in this
 area might include determination of (1) what the specific social and political consequences of partic-
ular water policies, programs, and developments are, (2) the extent to which these consequences
were anticipated and intended, (3) organizational means of assuring greater agreement between
actual and  intended consequences,  and (4) means for more accurately predicting the consequences
of alternative programs and developments.
     Conversely, social changes that have consequences for water resources should also be investi-
gated. Population growth, redistribution, and concentration; changes in the amount of leisure time
available and in recreational patterns; developments in  the location and operation of agriculture
and industry; establishment of voluntary associations interested in promoting conservation of re-
sources; transportation changes that make the population more mobile; and advances in communica-
tion on social problems (e.g. health, discrimination, and pollution) through mass media are just
some of the social changes having  important consequences for water resources.
     The nature and degree of social organization related to water resources in such areas as the
Great Lakes region is yet another  topic meriting investigation. Such relevant questions as the fol-
lowing might be raised: (1) In what way and to what extent can social systems analysis be applied
to the entire region and/or to its parts?  (2) How extensively are the people of the region organized
with reference  to water resources?  (3) Which organizations are more interested and  more power-
ful in water resource decisions?  (4) What are alternatively useful ways of studying parts of society
relevant to particular programs and policies, e.g.,community  power structure,  organizational sets?
(5) What are the region's peculiar  problems of coordination (or rivalry) among  institutions and
agencies concerned with water resources?
    Allied with this research would be that on the advantages and disadvantages of alternative
forms of organization and decision making.  For example, what are the merits of multiple-purpose
agencies or projects relative to multiple agencies or projects  with a single purpose ?  What are the
merits of vesting the authority and responsibility for decisions in units at the local level rather
than the state or interstate level?  What  are the advantages and disadvantages of coalitions between
public and private organizations to deal with water resources as opposed to public organizations
alone for this purpose?
    The organizational patterns and performance of water-resources agencies should  be examined
and evaluated.  How do the organizational patterns of these agencies influence their  performance?
For example, how are administrators and other staff members evaluated and rewarded for their
work, and,  in turn, how does this influence the content and implementation of policies and programs
and the enforcement of laws?  There seems to be a tendency in such agencies to neglect social
goals in favor of economic or physical goals, and it should be determined to what extent this is
true.  If it is  true, what could be done to facilitate the accomplishment of social goals  related to
water resources?

                                            13

-------
    Finally, studies of the beliefs, values, and actions of people In relation to water resources and
agencies dealing with these resources are needed.  There are several ways such information could
be useful. For example, public organizations relying upon laws relating to water pollution and en-
forcement agencies with relatively small staffs cannot observe each citizen and enforce such laws
on an individual basis.  There must be some acceptance, internalization, and self-enforcement of
the norms by the people. Then, how do people come to accept the norms and guide themselves?
What programs for this are effective ?  How long does it take to change the values and attitudes of
people in such matters?  Or, the problem may concern voting a bond for some water facility.  How
can we predict behavior on the basis of beliefs and values and, thereby, predict the success or
failure of such bonding attempts?  What motivates people to  support or reject policies and programs
for water resources, and how much of this motivation involves factors only indirectly related  to
the resources themselves?
    There are, in  sum, two major kinds of work to be done by  sociology and some of the other
social sciences in  relation to water resources. One is to apply accumulated theory and research
findings to particular water resource problems, while the other is to develop basic knowledge about
those aspects of society important to the quality, quantity, and  use of water. The preceding sugges-
tions for research constitute a "first view" of useful work to be done. As more study is given to
such areas, new perspectives will emerge, and focal points will become sharper; consequently,
what research is necessary will have to be redefined.

5.5.  STUDIES OF  INSTITUTIONS
    During the considerations of  a systems  analysis approach to research on the  Great Lakes,
discussion frequently turned to "institutional constraints."  There was general recognition that no
matter how "right" a physical and economic solution might be, there were often institutional ob-
stacles to its adoption  and that research  on such institutional problems is urgently needed.
    The terms "institutions" and "institutional systems" are here used to designate a web of  poorly
defined interdependences among law, agency operations, and public finance as these activate and
constrain federal,  state, and local governments in formulating and implementing water policies
and programs. The relationship of research on these to systems analysis of Great Lakes water
resources is by no means clear.  While it may be said that social, economic, legal, and political
forces do interact  to constitute a type of institutional system or systems which may determine
whether actions indicated by physical and economic considerations will be taken, it is apparent
that these interactions are not readily amenable to quantitative systems analysis. Limitations of
systems analysis in studying institutions have been identified by Professor Donald Michael:*
     * Program Director, Center for Research on the Utilization of Scientific Knowledge, and
 Professor of Psychology and Natural Resources, The University of Michigan.  The points listed
 here are paraphrased from a personal memorandum to Spenser W. Havlick, 17 November 1967.

                                             14

-------
     (1) Setting Goals.  The systems approach does not resolve the prior problem of choosing
among goals.  In the human sphere, particularly the political sphere, goals are not only difficult to
set, they are difficult to maintain. In fact, there is often a question of reestablishing goals when
feedback from the environment requires changes in the original goals.  If goals are not sensitive
to environmental data, the  response from the system tends to be out of phase with a set of goals or
priorities which were previously in effect.
     (2) Interdependency of Goals and "Means". Our political process  always involves compromises
on goals and on the means  for attaining them.  A critical aspect of our  approach to life is often to
eschew goals and make the means ends in themselves. As a consequence, neither system boundaries
nor subsystem relationships can be clearly differentiated.  Also contradictions and ambiguities in
the "system" — conditions  which the systems approach is designed to eliminate — are inherent in
the political-biosphere ecology the program is  designed to examine. Revealing these ambiguities
and contradictions and their political utility would  be very informative  for bio-ecologists, political
ecologists, and systems designers.
     (3) Values That Are Not Quantifiable. The problems of goal setting, interdependences of goals
and means,  and system ambiguity all converge  on and may be symbolized by the problem of how to
incorporate aesthetic considerations into the systems models.  Although  aesthetic factors are not
quantifiable, they are nevertheless central to the political and biological  interplay involved here.
     Acknowledging these limitations, we nevertheless believe that institutional research in the
Great Lakes region can profit from a systems approach. Thinking in terms of systems and general
systems theory can aid in relating a variety of  pertinent  factors,  such as the physical and economic
imperatives in a specific situation and the institutional systems which may be involved in their
accomplishment.  More specifically, a systems approach can serve to
     (1) Alert us to the extent of interdependences
     (2) Alert us to unintended as well as intended and unanticipated as well as anticipated conse-
quences and side effects
     (3) Alert us to direct  or indirect alternatives for changing the system when we may be blocked
from changing certain variables
     (4) Constrain us to search for new and more powerful variables than the numerous traditional
and easily measured ones that often have relatively little relation to decisions or actions on water
resources
    A systems approach in considering water and  related resources in the Great Lakes region will
emphasize that there are institutional values important for reasons other than water-related con-
siderations and that these must therefore be balanced against specific water-related benefits and
costs associated with alternative institutional arrangements.  Research in this area, then, must involve
                                             15

-------
not only water agencies, but the whole range of institutions affecting decisions on water policy and
management.  Institutions directly concerned with water resources are not separable from the en-
tire spectrum of institutions relied upon to achieve a wide range of social purposes.  Particularly
in the humid eastern portion of the country, specific water agencies must be considered only a
small part of the total institutional environment significant in water-resource problems.
    Within this frame of reference, several areas of institutional research emerge as important:
    (1) Analysis of the institutional systems through which water policies and programs  are
formed and implemented. Studies of specific agencies should document differences and relation-
ships among their  objectives,  legal authority, financing methods, and staff needs.  In short,  this re-
search should illuminate these agencies' capacity to deal with problems of the Great Lakes, the
clientele they serve, and the influences  to which they respond and the interactions among  agencies
required in attacking the region's water-resources problems.
    (2) Identification and analysis of the institutional implications of interaction between water-
centered decisions and related environmental factors, e.g., land use, industrial development,  sub-
urbanization, recreation, and transportation.
    (3) Studies of indirect factors affecting water-related decisions, such as the operations of
realtors,  organized industrial and conservation interests, engineering consulting firms, and news
media.
    (4) Studies to develop ways of identifying the "community of interests" which institutional
arrangements should serve by water resource decisions. Research is needed on the decision
structures required by specific water problems and on identifying the beneficiaries, cost  sources,
and principal participants associated with specific actions.
    (5) Studies to illuminate different perceptions of the nature and seriousness of water-related
problems so  that "political demand," as distinguished from "economic demand," may be better ex-
pressed.  Such studies can also guide the design of institutional arrangements to properly articulate
these value judgments along with other  factors with which they must interact.
    (6) Analysis  of the relative roles of the professions and politics in water policies and pro-
grams. There is  general agreement that fundamental water decisions today involve balancing con-
flicting social values and that these decisions must ultimately be made through political processes.
Studies that will clarify the need  and methods of maintaining clear  channels of political responsibility
should be encouraged. Analyses  of existing independent water agencies to determine the  extent and
significance of political responsiveness to them would throw light on the kinds of institutional ad-
justments which would promote more responsible political action.
    (7) Analysis  of emergency legal-institutional problems associated with enforcing water regu-
lations under current federal  law.  For example, enforcement patterns between state agencies and

                                             16

-------
public (municipal) corporations are fairly well established, but those for private corporations, an
area of  considerable importance in the immediate future, are less well developed.
    Research in these seven general areas obviously calls upon the energies and competence of a
variety  of disciplines and professions. In addition to a basic understanding of the technical param-
eters involved in problems of water use, there is a need for high-quality research in economics,
sociology, political science, and law. Much basic legal research related to water use in the Mid-
west has already been done, yet there is still considerable room for new contributions from the
legal profession. Special attention has been given in the preceding section to the needs for  socio-
logical research on water-related problems, and it is also contemplated that many of the above
areas of institutional studies should invite  contributions from sociologists.
    In addition to the important contributions that independent research in any of these general
areas can make to better understanding of  the institutional problems associated with water resources
in the Great Lakes region, specific case studies, properly selected, can provide an appropriate
vehicle  for probing these questions in depth and showing how they interact to form a total institu-
tional environment.  This approach should  identify a relatively discrete decision structure;  for
instance, studies might concentrate on water supply and waste discharge functions.  Decision units
in such  an area would normally be defined  by some relatively homogeneous group of urban jurisdic-
tions tributary to a lake.  For Lake Michigan, the subregional water-quality models  proposed in
this report could further aid in delineating a decision structure in terms of the physical and eco-
nomic parameters of water supply and waste discharge actions.  In such a context, several of the
above general research questions could be examined with the possibility of more immediate results.
    Case studies of this kind are also amenable to  experimental heuristic or gaming simulation of
institutional behavior. The relative roles of institutions involved in water-use decisions,  their re-
sponse to alternative solutions which might be proposed, and the legal  or fiscal constraints  they
might encounter are all important in developing a simulation model of  institutional behavior.
Heuristic or gaming simulation lends itself to the immediate exploration of institutional behavior
in a theoretical and highly abstract way.  This approach may quite soon provide some tentative and
contingent guidelines  for institutional design.  It would permit the introduction of improved physical-
economic information as the  more sophisticated models become operative and thus,  in turn, en-
hance the possibility of developing a more  sophisticated gaming  model. The possibility of such a
gaming  project is described in greater detail in Appendix G.

5.6. DATA REQUIREMENTS
    There exists the  need to examine the present status of and future  requirements  for collection,
storage, and retrieval of water and water-related data on the Great Lakes region so  that water in-
formation programs will be able to fulfill the needs of research, planning, and management programs
                                             17

-------
in the region (cf. Appendix D). Such an examination has two particularly important aspects within
the systems analysis and modeling framework proposed for research in this region.  First, present
data-collection sources, the types of data collected, and the form in which they are available must
be identified.  Then, present and future data.needs should be clarified, the present supply of data
evaluated for its adequacy for future research and information programs, and proposals for making
the necessary data available outlined.

    An accurate description of the data requirements for the proposed modeling effort will remain
unknown until the models are designed.  Nevertheless, during this designing,  a major effort will be
necessary to compile a detailed list of the data sources in subject areas expected to  support the
models developed.  Because of the comprehensive nature of the modeling effort, a broad  set of infor-
mation areas must be evaluated to determine the characteristics of completed data-collection pro-
grams and those in progress and the availability of data from these programs. The following may
be cited as only a few of the diverse types of data that will be required: inlake subsystem  models
will need water-chemistry, plankton, and current data; economic subsystem models will  need data
on water supply costs, sewage treatment costs, and recreation demands; and sociopolitical subsys-
tem models will need data on the political structure, specific governmental units, and special interest
groups in the area.

    A system for collecting and using the diverse kinds of data and information required in the pro-
gram should be developed in three separate parts: (1) a system for handling  selected basic data,
(2) a system for referencing data banks available on file, and (3) a system for processing abstracts
and texts containing analyzed data and information pertinent to the modeling of the Great Lakes re-
gion.  The Departments of Agriculture, Commerce, and the Interior are developing federal informa-
tion systems along these lines and it would be possible, through planning coordinated with both fed-
eral and state agencies, to create a unique collection of basic data for the topics and region under
study. Any  integration of such multisource data into a system which will identify major data sources,
be  coordinated with state and national data storage and retrieval systems, and fill the unique data
needs dictated by the  models developed obviously  requires detailed planning and design.

5.7. SURVEY OF NEEDED RESEARCH PROJECTS IN THE GREAT LAKES REGION
    In addition to the discussions of research efforts indicated in sections 4  and  5, the CIC staff
provided an opportunity for  conference participants and their colleagues to suggest what degree of
personal priority they would assign to the research activities suggested at the working conferences
and drafting sessions and in recommendations sent to staff headquarters In Ann Arbor.  Immediately
                                              18

-------
after the second working conference in Ann Arbor, an "Inventory of Needed Research Activities in
the Great Lakes Region" was mailed to investigators who attended the conferences.  This inventory,
in the form of a multiple-choice questionnaire, was intended to obtain the personal preference of
those responding for several water-related research activities, so the results of this informal
preference tally should not be construed as a group sanction of any particular research project.
The responses  are tabulated in Appendix H.

5.8.  A MECHANISM FOR  THE COORDINATION OF WATER-RELATED RESEARCH BY
     UNIVERSITIES IN THE GREAT LAKES REGION
    A regional  water and water-related research effort of the scope and complexity described in
this report should have continuing guidance and coordination.  Therefore, it is suggested that
a University Water Resources Research Group, which would replace the steering committee now con-
stituted and responsible for  this report, be established.  This  consideration of the research competence
and appropriate organization to further develop, manage, and execute the research design proposed here
has been requested by the  sponsor of this project

    The principal role of universities in an area such  as this is to perform high-quality fundamental
research relevant to the problems and applicable to  policy formulation,  planning,  and program
development by governmental agencies.  These research efforts should be free of the constraints
that are by definition imposed upon planners and program managers, but, on the other hand, the
research community must appreciate that the fruits  of research may only be  accepted and applied
to the extent that these constraints will allow. The fundamental division of effort  between  the re-
search community and the agencies responsible for policies, plans, and  programs is recognizable
at the federal level where  the Office of Water Resource Research is constituted as an agency re-
sponsible for relations between the Department of the Interior and the research community and is
separate from policy, planning, and program agencies  like the Department's Federal Water Pollu-
tion Control Administration or the U.S. Army Corps of Engineers. In the same sense, the Office
of Water Resource Research is distinct from, although obviously associated with, the Water Re-
sources Council, which exercises  centralized supervision over the several river basin commis-
sions. We believe the university research community  should,  in the same manner, be separate
from the operating agencies  of the several states and the federal government in this  region. This
is not to say that research relevant to the problems faced by the operating agencies is not of para-
mount importance to this research community, which should maintain working liaison with state
agencies and with federal agencies through the Great Lakes Basin Commission. The research pro-
grams should support the Commission and its constituent agencies as they study problems of water
and related land resources as necessary for the preparation of comprehensive and coordinated
plans.
                                            19

-------
    The university research community in the Great Lakes region is broader than that represented
by the CIC universities alone.  Other U. S. and Canadian educational institutions in this region also
have substantial research competence.  It is therefore suggested that the CIC or another existing
regional consortium serve as the coordinating vehicle for all of the academic research community
in the region by sponsoring a University Water Resources Research Group consisting of at least the
following  members:
    (1) Directors of the water-resources research centers in the several Great Lakes states
    (2) Representatives from the CIC universities which either do not have water-resources centers
or are not land-grant institutions, i.e., the University of Chicago, Indiana  University, the University
of Iowa, The University of Michigan, and Northwestern University.
    (3) Members-at-large from other major universities in the region, including at least one
Canadian  representative, with these members-at-large to be selected by the sponsoring consortium
    (4) The chairman of the Great Lakes Basin Commission or his designate

Should the industrial community develop a grouping similar to the suggested academic group, it
would then be appropriate to invite the chairman of such a group to sit with the university group.
This group should select a chairman and a small executive committee to conduct business between
meetings of the full group, and limited staff assistance should be provided. Public and private funds
could be solicited to underwrite costs of the group's operation.   This proposed organization would be
built around existing institutions such as the CIC and the  several water-resources research centers
already existing at the land-grant universities, and it is deemed the better part of wisdom to build
on such existing institutional structures. By including the directors of the water-resources re-
search centers, the group would have channels of communication to all universities, public and private,
within the nine-state region; the research center directors could acquaint competent investigators in
any of these institutions with research programs encouraged by the group.
    The broad  mission of the University Water Resources Research Group would be to encourage,
guide, and coordinate academic research activity in the Great Lakes region in support of policy-
makers, planners, and program managers in the responsible governmental agencies.  The
governmental structure in the United States portion of this region is, of course, formalized in the
Great Lakes Basin Commission.  The universities should not presume to preempt or duplicate the
activities  of the governmental agencies, but, rather, should relate  their work to significant problems
while maintaining accepted academic research freedom.  In addition, the following specific activi-
ties are suggested for the University Water Resources Research Group:
                                             20

-------
    (1) Maintaining an overview of relevant and desirable areas for research   an extension of
the comprehensive research design developed in this report
    (2) Encouraging the development of research proposals consistent with (1) above and partic-
ularly  on an interuniversity basis
    (3) Reviewing and endorsing interuniversity research  proposals consistent with (1) above  at the
request of those making the proposal
    (4) Assisting in obtaining the approval of public and private funding agencies for proposals

    (5) Serving as a general channel of communication among the region's universities and be-
tween these and public and quasi-public bodies
    (6) Maintaining immediate liaison with the Great Lakes Basin Commission, the principal
agency for  coordinating plans for the development of water and water-related resources in the re-
gion.  (The chairman of the Commission or his designate has been suggested as a member of the
group.)
    (7) Cooperating with the International Association of Great Lakes Research on symposia de-
voted to Great Lakes investigations. This will allow researchers to review  the progress and re-
sults of CIC projects on a sustaining critical  basis.
    (8) Serving on request as a consultative  body, or in arranging balanced advisory teams for
specific tasks, or in making recommendations to public or private bodies seeking qualified pro-
fessional talent for either permanent or consulting services
    (9) Sponsoring conferences to present problems and challenges of at least 20 years from the
present, which, in the minds of CIC investigators and public agency personnel,  need to be anticipated
and appraised.  Anticipating changes which are expected in the Great Lakes  region will permit re-
search activities to be geared to meeting needs before crises or irreversible developments arise.

    (10)  Making recommendations  concerning water resources research policies on a consensus
basis,  through the CIC, to appropriate bodies
                                             21

-------
                                        Appendix A
                       A REPORT ON THE FIRST WORKING CONFERENCE:
                    CONSIDERATION OF GREAT LAKES SYSTEMS RESEARCH

                                   10-15  September 1967

                              Alpine Valley, Elkhorn , Wisconsin
                                   A.I. INTRODUCTION

    This report is an overview of the events of the First Working Conference, which considered

Great Lakes systems models and related research.  It was held 10-15 September 1967, at Alpine
Valley, Elkhorn, Wisconsin.

    The agenda and timetable of events are listed in Section A.2.  Section A. 3 presents major por-

tions or highlights of presented papers, special work group discussions, and plenary session sum-

maries. In Section A.4, general conclusions  of the week are given in addition, to a "laundry list"

of projects which  emerged as appropriate targets for research by a consortium of CIC member

universities.  It is emphasized that this latter list is a beginning inventory of opportunities which

need attention.  Additional listings of projects and names of accompanying investigators were

welcomed.

    Participants are listed below.  Double asterisk (**) indicates Project Steering Committee, and

single asterisk (*) indicates  staff member of the Council on Economic Growth, Technology, and

Public Policy of the CIC.
    William C. Ackermann
    Illinois State Water Survey
    P. O. Box 232
    Urbana, Illinois  61801
    Sydney N. Afriat
    Krannert Graduate School  of
      Industrial Administration
    Purdue University
    Lafayette, Indiana  47907
    John C. Ayers
    Great Lakes Research Division
    North University Building
    University of Michigan
    Ann Arbor, Michigan  48104

    Shaul Ben-David
    Department of  Agricultural Economics
    Warren Hall
    Cornell University
    Ithaca, New York  14850
H. James Brown
Institute for Applied Urban Economics
Graduate School of Business
Indiana University
Bloomington, Indiana  47401
Henry P. Caulfield, Jr.
Water Resources  Council
Suite 900,  1025 Vermont Avenue  NW
Washington, D. C.  20005
R. F. Clevenger
Great Lakes Basin Commission
c/o Institute of Science and
  Technology
Ann Arbor, Michigan  48105

Charles A.  Dambach
Natural Resources Institute
The Ohio State University
124 West 17th Avenue
Columbus, Ohio 43210
                                            23

-------
B. G. DeCooke
U. S. Army Corps of Engineers
U. S. Lake Survey
Federal Building
Detroit, Michigan 48226

Rolf A. Deininger
School of Public Health
University of Michigan
Ann Arbor, Michigan  48104
William D. Drake
School of Natural Resources
University of Michigan
Ann Arbor, Michigan  48104

William J. Drescher
U. S. Department of the Interior
Geological Survey
1815 University Avenue, Room 234
Madison, Wisconsin  53706

Richard D. Duke**
Urban-Regional Research Institute
Michigan State  University
East Lansing, Michigan  48823

William Flinn
University of Wisconsin
Madison, Wisconsin  53706
Irving Fox
Water Resources Center
University of Wisconsin
Madison, Wisconsin  53706
Robert E. Graham, Jr.
Regional Economics Division
Office of Business Economics
U. S. Department of Commerce
2400 M Street  NW
Washington,  D.  C.  20230

Henry R. Hamilton
Department of Economics and
  Information Research
Battelle Memorial Institute,
  Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201

Spenser W. Havlick*
School of Natural Resources
University of Michigan
Ann Arbor, Michigan  48104

Herbert S. Heavenrich*
Council on Economic Growth,
  Technology  and Public Policy of CIC
2569 University Avenue
Madison, Wisconsin  53705
N. William Hines
College of Law
University of Iowa
Iowa City, Iowa  52240

Charles Holt**
Department of Economics
University of Wisconsin
Madison, Wisconsin  53706

James Kerrigan*
Water Resources Center
University of Wisconsin
Madison, Wisconsin  53706

N. F. Koenig
Bendix Aerospace Systems Division
3300 Plymouth Road
Ann Arbor, Michigan  48107

H. F. Lawhead
U. S. Army Corps of Engineers
Chicago District
536 S.  Clark Street
Chicago, Illinois  60605

Ralph A. Luken
School of Natural Resources
University of Michigan
Ann Arbor, Michigan  48104

Dale D. Meredith*
Department of Civil Engineering
3211 Civil Engineering Building
University of Illinois
Urbana, Illinois  61801

J. W. Milliman**
Institute for Applied Urban Economics
Graduate School of Business
Indiana University
Bloomington, Indiana  47401
A. D. Misener
Great Lakes Institute
University of Toronto
Toronto 5, Canada
Clifford Mortimer**
Center for Great Lakes Studies
University of Wisconsin-Milwaukee
Milwaukee, Wisconsin  53201

Paul Ray
Urban  Regional Research Institute
Michigan State University
East Lansing, Michigan 48823

Roland R. Renne
Office  of Water Resources Research
U. S. Department of the Interior
Washington, D. C.  20240
                                       24

-------
    Gerard A. Rohlich**
    Water Resources Center
    University of Wisconsin
    Madison, Wisconsin  53706
    Carlisle P. Runge*
    Council on Economic Growth,
      Technology, and Public Policy of the CIC
    2569 University Avenue
    Madison, Wisconsin  53705

    Stephen Smith
    School of Natural Resources
    University of Wisconsin
    Madison, Wisconsin  53706

    Harry Steele
    Water Resources Council
    Suite 900, 1025 Vermont Avenue  NW
    Washington, D. C.  20005
    Phillip L. Taylor
    Chief Data Operations Branch
    Division of Pollution Surveillance
    Federal Water Pollution Control
      Administration
    Washington, D. C.
William C. Walton
Water Resources Research Center
University of Minnesota
2675 University Avenue
St. Paul, Minnesota 55114
Matthew E. Welsh
International Joint Commission
United States Section
1711 New  York Avenue NW
Washington, D. C. 20440

James T.  Wilson**
Institute of Science and Technology
University of Michigan
Ann Arbor, Michigan  48105

Colin Wright
Department of Economics
31 Kresge Hall
Northwestern University
Evanston, Illinois  60201
                           A.2.  THE CONFERENCE TIMETABLE
Sunday, September 10
3:30 p.m.    Arrival of conference participants

5:30 p.m.    Steering Committee and CIC staff gathering

6:30 p.m.    Dinner and keynote address by Dr.  Henry P. Caulfield, Executive Director, Water
            Resources  Council

9:30 p.m.    Steering Committee and staff meeting

Monday, September 11

8:45 a.m.    Presentation of conference format and program task, Professor William C. Acker-
            mann, Conference  Chairman

10:00 a.m.   Coffee pause

10:15 a.m.   Open  discussion
            Informal comments solicited from Dr. Roland Renne, Director, Office of Water Re-
            sources Research, and Harry Steele, Water Resources Council

12:00        Lunch

1:15 p.m.    Problem and subsystem consideration in work groups selected by discipline, with the
            following membership under the following "subsystem" titles:
                                           25

-------
        Physical-Biological         Economic-Demographic         Social-Institutional
            Subsystems                   Subsystems                   Subsystems

          Rohlich, Chmn.               Milliman, Chmn.               Duke, Chmn.
          Deininger                    Alriat                         Smith
          Ayers                        Holt                           Hines
          Dambach                     Wright                        Luken
          Mortimer                    Hamilton                      Flinn
          Walton                       Koenig                        Ray
          Meredith (staff)               Ben-David                     Havlick (staff)
          Kerrigan (staff)               Brown (staff)
                                       Drake (staff)

                 The following were at liberty to visit among the three groups:

                                       Ackermann
                                       Renne
                                       Caulfield
                                       Steele
                                       Misener

3:30 p.m.    Coffee pause

3:45 p.m.    The three groups reassembled and then later reported in plenary session

4:30 p.m.    Issues of primary importance were reported by group leaders Professors Gerard
            Rohlich, University of Wisconsin, Jerome Milliman,  Indiana University, and Richard
            Duke, Michigan State University

5:15 p.m.    Open discussion and questions

6:30 p.m.    Dinner, followed by observations and  comments by Dr. Roland Renne, USDI, and Dr.
            Henry Caulfield and Harry Steele,  Water Resources Council

8:30 p.m.    Open discussion focused on conference objectives and problem identification
10:30 p.m.   Steering Committee and staff meeting

Tuesday, September 12

8:30 a.m.    Overview of comprehensive regional systems modeling for water resourced manage-
            ment, Professor J. W. Milliman, Indiana University, presiding
9:00 a.m.    Presentation by Henry R. Hamilton, Battelle Memorial Institute-Columbus Labora-
            tories:  "The Susquehanna Experience"
10:00 a.m.   Discussion
10:15 a.m.   Coffee pause

10:45 a.m.   Presentation by Rolf A. Deininger, School of Public Health, University of Michigan:
            "Physical Modeling of the Great Lakes System"

11:30 a.m.   Discussion

12:00 noon   Lunch

1:15 p.m.    Presentation by Charles Holt, University of Wisconsin:  "Economic-Demographic
            Subsystem Modeling"

2:00 p.m.    Comments by Professors Sydney Afriat, Purdue University, and Colin Wright,
            Northwestern University
3:00 p.m.    Coffee pause


                                           26

-------
3:15 p.m.    Open discussion

5:00 p.m.    Adjournment

10:00 p.m.   Steering Committee and staff meeting

Wednesday, September 13

9:00 a.m.    Consideration of social subsystems including institutional arrangements in Great
            Lakes  region, Gerard Rohlich, moderator for the day

9:15 a.m.    Panel discussion headed by Dean Stephen Smith, University of Wisconsin School of
            Natural Resources
            Other panel participants:
               Fox
               Havlick
               Hines

10:30 a.m.   Coffee pause
10:45 a.m.   Discussion and questions re social subsystems

12:00 noon   Lunch  and Steering Committee-staff meeting

1:30 p.m.    Open discussion and assessment of progress. Five multidisciplinary groups were
            formed to consider geographical and problem-oriented research possibilities. Group
            members:

Social Goals
 and Values

   Brown
   Drake
   Duke
   Flinn
   Koenig
   Smith
   Wright

3:45 p.m.    Discussions in interdisciplinary groups led by moderators: Ayers, Duke, Hines,
            Milliman, and Walton
4:30 p.m.    Free time
6:30 p.m.    Dinner

8:00 p.m.    Remarks by Gov. Mathew E. Welsh, Chairman, International Joint Commission, U. S.
            Section
9:30 p.m.    Discussion and questions

Thursday, September  14

9:00 a.m.    Presentation of reports from the five multidisciplinary work groups

10:30 a.m.   Coffee pause

10:45 a.m.   Continuation of reports and discussion after each

12:00 noon   Lunch

1:00 p.m.    Remarks by Hon. Raymond F. Clevenger, Great Lakes Basin Commission

1:45 p.m.    Questions
                                            27

Quantitative
Afriat
Deininger
Drescher
Meredith
Walton


Quality
Ayers
Ben-David
Dambach
Heavenrich
Holt
Mortimer

Economic
Graham
Hamilton
Kerrigan
Milliman
Steele

Legal-
Institutional
Ackermann
Clevenger
Havlick
Hines
Ray
Runge

-------
2:00 p.m.    Series of presentations on existing data sources for Great Lakes Research
            Robert E. Graham, Department of Commerce
2:40 p.m.    William Drescher, U. S. Geological Survey
3:00 p.m.    Coffee pause
3:15 p.m.    H. F. Lawhead, U. S. Army Corps of Engineers
3:45 p.m.    B. G. DeCooke, U. S. Lake Survey
4:10 p.m.    Phillip L. Taylor, Federal Water Pollution Control Administration
4:30 p.m.    Discussion
5:00 p.m.    Free time
6:30 p.m.    Dinner
8:00 p.m.    Informal discussions
Friday, September 15
9:30 a.m.    Plenary session,  William Ackermann presiding
            Review of reports from problem-centered multidisciplinary work groups
10:30 a.m.   Comments by Dr. James T. Wilson, Chairman of the Project Steering Committee
11:30 a.m.   Discussion; summary of the conference results
            Adjournment for general participants
12:00 noon   Lunch
1:15 p.m.    Steering Committee-staff meeting
3:00 p.m.    Adjournment for Steering Committee and  staff

                  A.3. ABSTRACTED HIGHLIGHTS OF THE CONFERENCE
    Inasmuch as the major objectives in this section are to illustrate the evolution of thinking and
present highlights of the group effort, the names of individuals making specific remarks have not
been included.  Abstractions  of many remarks have been made from tape recordings and notes as-
sembled by the CIC staff. Due apologies are made because more complete renditions of the dis-
cussions and presentations were deemed inappropriate in this summary.

                            10 September 1967 — Evening Session
    The formal program of the conference began with a keynote address by Dr.  Henry P. Caulfield,
Executive Director of the Water Resources Council.  A useful history of federal  involvement in
water-management activities set the stage for a  discussion of present and future  expectations in
water-planning activities by the governmental organizations which now exist.  For the sociologists,
economists, engineers, biologists, and systems analysts who were not intimately familiar with the
posture of the federal establishment in water-related research, the remarks provided a necessary
and enlightening beginning for the week's deliberations.
                                            28

-------
    The administrative operation of the Water Resources Council was detailed in terms of member-

ship, responsibilities, and ongoing functions. It was pointed out by the speaker that from his vantage

point the will of both the Congress and the Executive Branch regarding overall management func-

tions in natural resources is coordination of water- and related land-resources matters as pro-

vided by the Water Resources Planning Act of 1965.  Those functions are:

        (a)  To maintain a continuing study and to prepare biennially a national assessment of the
             adequacy of water supplies,

        (b)  To appraise the adequacy of administrative and statutory means for  coordination and
             implementation of the water- and related land-resources policies and programs of
             the several federal  agencies and to make recommendations to the President with re-
             spect to federal policies and programs.

        (c)  To establish, after appropriate outside consultation  and with the approval of the
             President, principles, standards, and procedures for federal  participation in the
             preparation of  comprehensive regional or river-basin plans and for  the formulation
             and evaluation  of federal water- and related land-resources projects.

        (d)  To coordinate schedules, budgets, and programs of  federal agencies in comprehensive
             interagency regional or river-basin planning.

        (e)  To carry out Council responsibilities with regard  to the  creation, operation, and ter-
             mination of federal-state river basin  commissions.
        (f)  To receive and  review comprehensive regional and river-basin plans and transmit
             them, together with  Council recommendations, to the President for consideration and
             transmittal to the Congress.

        (g)  To assist the states  financially in developing and participating in the development of
             comprehensive water- and related land-resources plans.

        To perform these functions the Council has been organized as set forth in its Rules and
    Regulations.  Certain highlights are  significant for your understanding of this new  institution.

        First, the organization is designed  to assure that Council members themselves, who in
    effect constitute a Cabinet Committee, "will meet at least quarterly and consider and decide
    major  matters ..."

        Second, the  organization  provides for Representatives  of Council members together with
    the Executive Director to "take action when necessary and  appropriate and, after consideration,
    submit recommendations to Council  members on matters requiring their action."  The Execu-
    tive Director is  chairman of  meetings of Council Representatives, and they are held at least
    biweekly. Decisions at this level are by unanimous agreement of the Representatives and the
    Executive Director.  Thus all unresolved matters automatically go to the Council members
    for decision.

        Third, the organization includes provision at headquarters for administrative, technical,
    and consultative committees.  Three administrative committees — for policy, planning, and
    state grants — have been established. Each is chaired by the appropriate  Assistant Director of
    the staff.  The Council's technical committees,  so  far,  are  the old technical subcommittees of
    the Inter-Agency Committee on Water Resources,  popularly known as "Icewater,"  that was
    abolished in April  1966.  By consultative committees, the Council means committees of persons
    from outside the Federal Government.   To date, the Council has not given consideration to es-
    tablishment  of any consultative committees.

        "Icewater" field committees  also came  under  the aegis of the Council in April  1966. It is
    now expected that these  committees will continue to function until, as in New England, the
    Pacific Northwest, and the Great  Lakes  they are superseded  by Title II river basin commis-
    sions.


                                             29

-------
    At this point the discussion began to zero in on the replacement of interagency committees.

Then a special reference is made to the River Basin Commission arrangement with the Great Lakes

Basin Commission (GLBC), held up as a case in point:

        We have, as you know, established the Great Lakes Basin Commission.  The functions of
    a river basin commission are what you need to be concerned about here.  Let's get to the struc-
    ture now and functions later.  What we have here applies to the Great Lakes Basin Commis-
    sion, the chairman being appointed by the President.  Now he is not just a chairman of a com-
    mission; he is the coordinating officer for the federal member, with the additional functions
    of the chairman of the commission.  This is not an unimportant role. In consultation with the
    Vice Chairman, he picks the time and place of meetings, sets deadlines of  submission of annual
    and other reports, establishes subcommittees, and so forth. With the concurrence of the Vice
    Chairman,  he appoints commission technical staff. He himself is responsible for the use and
    expenditure of funds of the commission, so  that the chairman has some powers of consultation.
    The two of  them have definite power of respective appointment of personnel.  I might say, how-
    ever, that the details of the chairman's functions that are not indicated  in the by-laws can be
    circumvented by agreement of the commission.
        The River  Basin Commission members are, of course, the eight states in the Great Lakes
    region; and the federal  members are, I believe, from nine departments including the Justice
    Department, which was added because of the litigation or the degree of  legal action that has
    taken place in this basin.  The United  States section of the International Joint Commission could
    be a member, but this was decided against for two reasons.  One, because  the United States
    section of the IJC is made up of three people who could not possibly  divide themselves for
    actually attending meetings of all the commissions.  That's a bureaucratic reason.

        The second reason is probably more important.  We had to consider the role of the GLBC
    as opposed to the role of the IJC. The IJC is supposed to be an advisory body,  not a negotiating
    body between American and Canadian  members, considering problems without respect to
    nationality.  If the same people were  members of both commissions  and participated in making
    IJC policy, their membership in  the GLBC would in a sense be prejudicial  in making their de-
    cisions in the IJC.  We therefore decided (Governor Welsh was particularly influential here)
    against having IJC membership on the GLBC.  The IJC considered this  question also.

    Four functions  of the Great Lakes Basin Commission seemed to stand out in the presentation.

First, the Commission is to be the principal agency for coordinating all water planning in the re-
gion.  This  coordination function included "water- and land-resources planning" at all levels of

government. It remains an open question how far down the ladder this in fact can go and how in-

volved private industry can be.

    A second function is to prepare and keep up to date a comprehensive, coordinated joint plan

for water and related land resources in the region. The language in Title II of  the Water Resources
Planning Act suggests that the plan include federal, state, interstate, local  and nongovernmental

resources development.

    A third function, although not stressed, deals with recommendations of  long-range schedules

of priorities for the collection and analysis of basic data. Equally important is a fourth function,

which encourages the Commission to support and "undertake such studies of water and land re-

sources problems ... as are necessary" for comprehensive river-basin planning and management.

    Most of the above points are fairly obvious  if one  is at all familiar with Senate Document No.

97 and the Federal philosophy in this field. Dr. Caulfield gave due emphasis to the objective of

                                            30

-------
economic development, but he went beyond that consideration to stress the social and amenity ob-

jectives which are awkward to deal with in economic analyses.  The example of preserving wild

rivers in their natural state for the purpose of open space and natural beauty was used.

        At the time this was put in Document 97 it was a brand new idea. Now we have Scenic
    Rivers Bills, etc., and it is accepted almost as "old hat." Well-being of people is  a general
    term which has social significance,.and as indicated here it is very difficult at times to spell
    out more precisely in a general way,  applying all over the country, exactly what one would
    mean by well-being.  It certainly  is a reminder to indicate that people, after all, are what we
    are concerned about here in connection  with the use of resources,  in many instances.   How-
    ever, in a group like this I should remind you  that this also means particular policies which
    the Federal Government has adopted which are believed to be of social importance, and are
    matters of law.

    The final thrust of Dr. Caulfield's remarks concerned  challenges and opportunities which are

in need of serious consideration.

        Procedures for getting the principles of welfare economics into planning have not  yet been
    developed adequately to get such procedures into use by people actually employed in the work.
    In the Delaware Study in some rough way  Marshallian considerations were incorporated in
    what that study thought was a "Marshall Plan."  The Delaware Study people were conscious of
    this in time,  but their procedures were  very crude indeed compared to what might have been
    rigorous procedures leading to an optimum solution. As you know, we take into consideration
    the constraints of social goals (which are intangibles and include any kind, even, shall  we say,
    a wild river) when we are thinking of  modifying an efficiency model. However, there has been
    no project formulated in the United States that has ever actually done this. This is an  impor-
    tant point.  You may know that we have developed a joint Army and Interior project on  the
    Upper Missouri which originally had eleven alternatives and came  down to, in the public  sense,
    five meaningful alternatives.   But it certainly did not start out as an efficiency model in such
    a way that you could indicate the development  suggestions by means of the constraints  on the
    model . . .

        Now as far as I am aware, there  has  never been a model rigorously developed for  a  river
    basin or ever practically carried out. The hard work on the Harvard and  Lehigh River models
    fell short [sic] as many of you know, and you can understand that one can't take the approach,
    "Well, this just takes a few more of us bright fellows and it will get done." But I need  not
    labor that point; I'm sure you're all aware of this.  So a real research function exists,  real
    things aren't known and real inventions  are needed in this field.  It is in these inventions in
    river basin planning, where systems analysis can work, that we are looking forward to, and
    the possibility of your making a real contribution to this phase. This is a unique situation in
    the Great Lakes, not quite the same thing as in the Missouri Basin.

        Now I want to call your attention  beyond these two areas into specific areas  of research.
    One is on economic matters.   Those of you who know more about this than I can carry  on from
    here, but I will remind you that there  may be something substantial here.  Senator  Nelson has
    introduced a bill — S2123 — for investigating alewife and other fish in this area, and so there
    may be something more going on  than just what we are talking about. I need not  mention to
    those from Wisconsin that we're concerned with the eutrophication problem.  The area that I
    would think of first is cost sharing research.  This is of more general interest.  I think our
    friend over here from the University of  Indiana knows more about that than we do,  but  I haven't
    seen any research paper on it  yet.

        Then the next question is the  whole  problem of flood-plain management.  The flood-plain
    management problem includes the problem of cost sharing of flood protection, flood insurance,
    and flood-plain use regulation.  These are areas of needed research. Those of you who are
    hydrologists may know that the task force of flood-control policy singled out flood-frequency
    problems, methods of determination of uniform methods of determining  flood frequency.  This

                                            31

-------
    is a very difficult problem.  We have had our Hydrology Committee working on it, through out-
    side consultants.  This is an area in which (if we are going to really go ahead on any kind of
    flood insurance, let alone do a good job with flood protection) we certainly need to have much
    better methods or the best methods as the basis of determination of calculation of damage and
    risk.
        I hope I have given you some kind of a conception of the  government and other ongoing ac-
    tivity on the one hand and our interest in research on the other hand, and I hope,  if your group
    decides to go forward, there can be a  marriage between these efforts in such a way that neither
    member of the marriage is stifled by the association with the other, and that something fruitful
    can come from the marriage.

                                 Monday, 11 September 1967
    Moderator-for-the-day Professor William C. Ackermann led off with the basic challenge that
the conference was intended to "advance the state of the art as a  multidisciplinary group."  It was
made clear that no constraints should exist on areas and approaches  to be considered.  Definitions
were offered about the tasks at hand, the concept of the model, and the system and subsystems un-
der discussion. A problem-oriented approach was emphasized.
    Whether the development of a model for the  entire Great Lakes drainage  basin was realistic
came up as a serious question.  The Canadian experience illustrated  how the Lake Ontario basin
has been "divided" seven ways for planning purposes (water, quasi-political, economics, terrain,
population, industrial, and "potential for development").  Then a caution was issued about the
"falsity of boundaries."  The 1967 National Assessment by the Water Resources Council is a first
attempt to overcome particular boundaries.  In fact a considerable effort seems needed in correct-
ing the distortions created by economic and physical externalities (economies and diseconomies)
in the system.
    Another tack was taken by the question 'What are the problems we want to solve?"  Official
objectives (via Senate Document No. 97) are stated as economic development, recreation, flood
protection, etc.  Problems specific to the Great Lakes are the difficulties with "pollution, alewives,
lake levels, hydro-power capacities, recreation development, soil erosion, water supply, wildlife
considerations, massive urban proliferation,  maintenance of wild areas, waste disposal procedures
and ineffective institutional arrangements."
    At one point a participant expressed his surprise  about the magnitude of the CIC  contract.  It
was suggested then that the ultimate goal of this  experiment in systems models was "to produce a
proposal, not a result."  The notion of a problem basis with a geographical subsystem for each
major problem began to gain support.  Several participants suggested a "blue sky" approach; others
favored a research design which could  produce results with high practical utility in terms of cur-
rent organizational problems.  Still other questions raised were to whom the research is to be di-
rected, how much money is available, and  "will the people with the money have to define the
problem."

                                            32

-------
    After some confusion at this point, a suggestion was made that a basic set of interrelationships
could be put together rather as a farmer integrates and manages his unit of the landscape.  The
farmer interfaces the market conditions with his environmental capabilities and manages the farm
as a relatively simply system.  Of course the analogy is weakened when one considers the complexity
of even one of the Great Lakes' basins.  It was agreed that the model of a municipal or metropolitan
operation might make a better analog:  the elements have  to be identified and a  time scale has to
be agreed upon.
    Before the three subsystems groups (physical-biological, economic, and social) went into work
sessions, a series of questions which expressed the interests and concerns of many conferees was
propounded.  The questions listed below serve as  illustrations:
    	 If our mission is to lay out a regional research design or proposal instead of solving
          local problems per se, don't we have to look at  the interrelationships between all three
          basins  (Upper Mississippi, Ohio, and Great Lakes)?
    	What was  the Susquehanna experience ?
    	 At what points does the water sector actually feed back on the economic sector as a
          whole ?
    	 Is  some research needed on what the social goals really are?
    	 Is  the pollution problem in the Great Lakes so long term and subtle that we would have
          trouble identifying action proposals?
    	Wouldn't it be a good idea to try to study  feedback from the water sector in terms of
          critical point identification?
    	 Is  there really a public demand for pollution abatement?
    	 Is  there really a water shortage in the Great  Lakes basin?
    	What are the  marginal products vis a vis tradeoffs between navigation and power?
    	What are the  restraints from the Supreme Court "diversion" activity?
    	 How much social value and economic value is placed on different levels of pollution?
                                       ************
    Reports  of the three groups are paraphrased below  in a condensed form.
    Professor Rohlich's report on the physical-biological subsystems began the afternoon session.
Attention was directed principally to water pollution and lake levels.
    It was indicated  that lake-level investigation lends itself very well to systems analysis.  Some
question exists about whether the total lake-level picture is truly a system.  Is it worth while to
go into more  sophisticated models of lake levels at  the present time ? Can existing works be  better
used? Could Lake Superior be used as an upstream regulatory body?  Models can be  built to
answer these last two questions and others.
    The eutrophic problem was a critical issue. Factors  which affect the metabolism of the lakes
were  discussed, and  it was agreed that we need more research on indices of eutrophication, better
monitoring or eutrophication, and research on sediment-water interfaces.   The ground-water
                                            33

-------
"basin" needs to be established and compared with the surf ace-drainage area.  Improvement of

tributary water quality was discussed in terms of total lake quality. More needs to be known about

coastal and offshore waters in terms of "mechanisms of exchange."

    Availability and retrieval of data at the present time needs study. How little data can we get

along with to do the necessary modeling?

    Questions  raised as a result of the group's report included one concerning a possible benefit-

cost analysis of the U. S. Army Corps of Engineers' dredging and disposal operations.  Other state-

ments  indicated that very little relationship exists between lake levels and pollution.  Also it was

pointed out that eutrophication and  pollution are not synonymous.

    Professor  Milliman's report from the economic-demographic subsystems group was next.

        Do we need a large complex economic model of economic growth of the Great Lakes re-
    gion, perhaps in a systems sense, to answer the  kinds of questions that we are dealing with;
    and when will we know if that kind of economic model is justified?  A large complex economic
    model  could be used to generate possible growth patterns and factors that condition economic
    growth in the area and to generate an idea of demands of the economy  (under various assump-
    tions) to deal with the various  kinds of water and water-related problems. In addition, that
    model  could show strong feedback between changes in water costs and the economy; and per-
    haps it would show if there is  any feedback and how  the economy would be influenced by various
    changes in water costs ....  If changes in water costs do not directly and greatly affect eco-
    nomic  growth, then it is fairly doubtful on this basis  alone that a large complex economic
    model  (very difficult to construct) would be justified.  We don't know the answer to that, but
    we suggest as a strategy that some preliminary research needs to be done to find out. As a
    starter some likely critical points, in terms of both products and spatial configurations should
    be picked out, where we think  that the water-economy feedback might be evident.  If the pre-
    liminary research strategy shows that strong feedback does exist, then the case for a complex
    economic systems model depicting economic growth, within which you look at other factors
    affecting economic growth as well as water,  might be strong.

    You could take some simple, straightforward projections of economic  activity and from them
see the demands for water.  The field of recreation was  mentioned.  What changes in water costs
will affect  "sensitive" industries?   "We feel very strongly that additional work needs to be done on

price policy, on financing and on allocation ....  We need to know the efficiency implications of
putting water into one use as opposed to another  ..." Could we not build a simple model to moni-

tor the economic effects that changes in the water-quality regulations will  bring about, at least at a
few critical points, to give us a handle on economic effects of changes in water-quality management?

     The message from this group  to the CIC is that the complex systems approach perhaps can be

justified in examining the water sector, but only  after the initial models and research have been

undertaken.  Also, the group agreed that the CIC can take the leadership to begin the very, very

difficult task of doing some quantitative research on  social goals and values.

     Professor  Duke reported for the social-institutional subsystems group.

     Three  considerations were presented as a prelude to the main points of the report: (a)  "There

is relatively slow growth in water  technology in terms of the problems which are evident; (b) with

                                            34

-------
the greatest supply of fresh water in the world, people are going to congregate in megalopolitan
regions on a scale that society has not encountered before; and (c) water as a free good is going to
become increasingly scarce ". . .  which will call for more political demands for water policy, for
further managerial intervention, for further planning of water usage, and for reduction of such
negative externalities as shipping  your  sewage down to the next lake." The discussion centered on
a review of present and future organizations including the Great Lakes Basin  Commission and the
need some day for an effective international organization, which will ultimately be required.
    Perhaps more limnological data are needed before institutional arrangements can be devised.
Are the localized pollution problems really a concern for the entire Great  Lakes system? If  not,
what subsystems should be identified and what linkages built?
    Public involvement in the water-planning process was discussed.  The examples of Texas,
Illinois, and California were given. In the  Great  Lakes region some experimentation could be done
with "anticipation of problems" in addition to the traditional approach of planning to resolve prob-
lems.  Some research is needed in organizational effectiveness in local and regional water-
management bodies.  "The theory  of planned social intervention and planned social  change is akin
to the notion of strategic theory, and this has been . . . very badly developed in the  social sciences.
The other kind of model (naturalistic theory) has been very good; but it does not help us to be  good
change agents and perhaps an  elaboration of that would be helpful."
    As an appendix to the social subsystem report, the "annotated outline" prepared by Lyle E.
Craine, Spenser W. Havlick, and Ralph  A. Luken, which was discussed in the social-institutional
subsystems group, is included here.

                             An  "Annotated Outline" of the Social
                             Subsystem in the Great Lakes Region
    The destiny of the Great Lakes region depends in part on its ability to  utilize its resources
for regional development.  The available water resource is one key ingredient for development.
Water of the Great Lakes Basin, to greater or lesser extent, constitutes  a  physical system. How-
ever, decisions regarding water use and development involve interaction with  other systems.  For
the purpose of this conference we  are conceiving the entire Great Lakes  Regional System as being
composed of three functional subsystems:  physical, economic-demographic, and social.
    Our outline focuses on the social subsystem.  This  system is less precise then the others and
has been given relatively less  attention by  systems analysts. We suggest that social subsystems
be delineated by the interdependencies of social change  agents.  These agents  of social change are
often, but not always, identifiable organizational entities.  They are the formal and  informal mecha-
nisms of social interchange which articulate the collective needs of the region and mobilize re-
sources for meeting these needs.
    In order to provide a basis for discussion, we raise several questions  in outline form which
seem germane to the consideration of the role of social systems to water resources of the Great
Lakes.  In this exploratory approach we propose to raise and briefly comment on a succession of
questions.  We anticipate several answers to our questions, many of which  will differ from our
answers.
                                            35

-------
I.   What is a system?

    A system is a set of objects, together with relationships between the objects and between the
attributes [1].

    Subsystems are primary divisions of a system.
    Objects are parts of components (physical or abstract) of the system.
    Attributes are properties of the objects.  For example, stars have temperature, velocity, etc.
    Relationships are the interdependencies which tie the system together. Any set of objects has
      relationships.  The question is, which are the relevant relationships?

    Whether a set of relationships is a system or subsystem depends on the complexity of the prob-
lem and the objectives of systems analysis.

II.  What do we mean by systems analysis ?

    Two possible definitions are:

    A.  Systems analysis is a strategy for problem solving. In this situation it has been defined as
an effort "to make comparisons systematically and in quantitative terms and to use a logical se-
quence of steps which can be retraced and verified by others" [2].

    B.  Systems analysis aids in defining the relevant components of any system and the relevant
linkages or interdependencies among the components (objects).  The approach is more descriptive
than the problem-oriented approach. In many cases, a systems approach may be as important as
a method of defining the relevant problem as  it is a process of problem  solving.  It is broader
gauged research in that it deals with quantitative and qualitative factors.

III. What is regional development?

    A.  What is a region?

        We suggest one set of criteria for defining a  region. These criteria are geographic (the
physical environment), an awareness of regional problems and opportunities, and an anticipated
capacity to solve these problems [3].

    B.  What is development?

        Economists distinguish between economic growth and development for a region.  Economic
growth is usually defined as a change in real  per capita income as a result of capital investments.
Economic development is both a change in real per capita  income and a  change in technical and in-
stitutional arrangements which generate income  [4].

    C.  What is the role of the social subsystem in this process ?
        In the ultimate analysis evidence suggests that whatever gets accomplished does so as a
result of the social subsystem.  Social interaction emerges in decisions to locate industry, to develop
recreation facilities, to modify zoning regulations, or to dredge harbors. One possible interpreta-
tion is that the social subsystem generates support for the particular objectives of regional develop-
ment and mobilizes resources for achieving these objectives.
IV. What is the Great Lakes Region?
    One possibility is the region described by hydrologic boundaries.  We question whether the hy-
drological region is the relevant region for problem  solving and planning in the Great Lakes area.
Very few aspects of water management require a Great Lakes  management program.  Problems of
pollution abatement, flood control, and provision of water-recreation facilities can be solved on
subregional bases.  For example, pollution abatement does not require a Great Lakes solution, but
rather programs in specific independent areas, such  as southern Lake Michigan, Detroit River, and
eastern Lake Erie.

V.  What is the social subsystem of the Great Lakes region?

    The social subsystem comprises all components which engage in social  interaction  for resolv-
ing water-management problems or which utilize water for promoting regional development.  Social
interaction is defined as the "process in which individuals (organizations) relate to their own minds

                                             36

-------
or the minds of others — the process, that is, in which individuals take account of their own or their
fellows' motives, needs, desires, means and ends, knowledge, and the like" [5].  Social organization
exists to the extent that actors engaged in social interaction make some effort to restrict entrance
into or withdrawal from participation by present actors.

VI. What are the components or who are the social change agents in  the Great Lakes Region ?

    Some of the components are;
    (1)   Formal political organizations ^-national, international, state, county governments
    (2)   Formal administrative agencies — U.  S. Army Corps of Engineers, Water Resources
         Councils, F.W.P.C.A., U. S. Geological Survey, Great Lakes Basin Planning Commissions,
         state water agencies
    (3)   Formal private organizations — industries
    (4)   Planning bodies
    (5)   Interest groups
VII. What is the role of research in analyzing the social subsystem ?
    Perhaps the most important role is identification of linkages or interdependencies among the
components, individually and collectively as they respond to challenges.  For example, what are
the linkages among the formal political organizations in the region and the Federal Government
for pollution abatement; which are strong and activated frequently, and which can be controlled in-
directly by policy guides instead of administrative decisions?

    A.  What are linkages among change agents in the region?
    B.  What are the  linkages between the Great Lakes and other regions?
[1]  These definitions are taken from A. B. Hall and R. E. Fagen, General Systems Yearbook
    Vol. 1, 1956.

[2]  E.S. Quade, Military Analysis (Santa Monica, Calif., The Rand Corporation, 1965), p. 2.

[3]  "DesignFor a Worldwide Study of Regional Development," (Washington, D. C., Resources
    For the Future,  1966), p. 4.

[4]  This distinction  is taken from Charles Kindleberger, Economic Development, 2nd ed.,
    New York: McGraw-Hill, 1965, p. 3.

[5]  Guy E. Swanson, "On Explanations of Social Interaction," Sociometry, (June 1965), p. 102.
                                         *****

    On Monday evening, Dr. Roland Renne presented his hopes and expectations about the confer-

ence and an overview from the Office of Water Resources Research (OWRR).

    What with 26 federal agencies dealing with water resources in some way, OWRR sees its role

as "extramural." A  "relevancy test" has been developed as a result of about three years of expe-

rience (700 projects  under way). It is applied by the water-resource center directors. For exam-

ple, you in the Great Lakes know what the problems are and we are trying to stimulate research

projects which "fit" the problems.

         The CIC proposal appealed to us very much not only because of the approach and the sig-
    nificance of the problem but also because of the powerhouse that is represented by the eleven
    universities here in the Midwest that are involved. In Washington it is pointed out that one of
    the areas of greatest  Congressional concern is the pollution of the Great Lakes (particularly

                                            37

-------
    Lake Erie).  We are asked what we expect to get from our investment in water-resources
    research.

        By dependence  only on mission-oriented agencies, many significant projects may be
    missed.  Thus the OWRR brings together through competition the best competence in the
    country to discharge water-resources research; 107 universities are in the program.

        The type of research  that will come out of your deliberations will probably be considered
    under Title II.  It now has 31 projects operating in FY 1968 out of 255 submitted projects ask-
    ing $29 million versus $2 million appropriation available.  Many  of the  systems analysis
    projects were disappointing.  Shortage of time may have been a factor.  Many of the projects
    were naive and exploratory.  They fell short in terms of depth and  were not related to specific
    major problems.  The projects are due on 15 November 1967 for FY 1969.  We don't  know how
    much money will be available but we want to explore longer range planning horizons.

        We're in the market for good projects in systems analysis of a set of problems that need
    further examination and study and more research, like the Great Lakes region. We're certainly
    interested in getting further work done that will reveal these areas, and we think that this
    $27,000 is a good investment in terms of getting from you more specific, detailed projects that
    we'll be very much  interested in helping to support.  Now of course the relevance of this type
    of work to the problems that are significant up in this area will have national significance;
    they will certainly fulfill the purpose of the Water Resources Research Act in trying  to provide
    a more adequate supply of water both in quantity and quality for the United States and its grow-
    ing needs, which purpose was written into the Act.  But in addition, your work should be very
    helpful to the planning programs of this region; and while we are not interested solely in fund-
    ing research that will be of benefit to the Water Resources Council (although  it is important),
    we feel that in our overall program  there should be projects of this kind that will be of primary
    interest and relevant to the planning problems of the area.  So the usefulness of the results,  if
    they are  good in the case of water resources planning, will help develop a ready, good, and
    highly desirable market for further  research along these lines. How we can weld together
    more effectively the research work we are doing with the planning  needs  of the Council is a
    very important matter.

        The  amount of  money available  for this undertaking might be about 10% at the outside limit
    ($400,000 or $500,000) if a $4 million appropriation is available for FY 1969.  The Budget
    Bureau has emphasized Category Six in the 10-year program—"Research for Water Planning"
    (about 20%).  One-fifth of our allotment projects are in the water-resources planning area,
    which is  very encouraging.

        In the matching fund projects the proportion is 30-32%.  In Title II we've put 76% of FY  1968
    money on research  related to water-research planning (of a $10.1  million program).  After these
    short-term projects (9-24 months) bear specific fruit, we hope to go up on the Hill and show
    what early efforts have yielded in order to obtain additional help.
        Dr. Renne in his closing remarks offers an inspiring challenge:
        Looking down the road it appears individual projects are not currently being made for ten
    years, but I'm sure, if this group comes up with a solid program of research needs and can
    develop the necessary projects that will stand up under our evaluation process, there isn't any
    reason why this office along with other agencies of the Federal Government can't take a more
    active role in helping to finance these for the next 5 to 10 years and get the kind of information
    and facts that we need for much more effective planning in water-resources development.

    Mr. Harry Steele of the Water Resources Council noted the hard work that had preceded the

present priority for research  in the water-planning area.  It was made  clear that  "extramural"

research was going up and that "in-house" research is going down.  Nevertheless, a strong in-house

capacity is required to translate research results into action programs in the planning community,

and this is the big gap.


                                            38

-------
    Then a summary was given about the national water assessment presently under way.  A hope

was expressed that the figures  in the Great Lakes Basin, for example, can be tested in order to

improve the national ability to translate the economic projections into water requirements "realis-

tically and more accurately."  The present data are poor, e.g., current data on water use for the

1967 assessment.  Better techniques for obtaining such data can improve the quality and reliabil-
ity of the statistics.
                                Tuesday, 12 September 1967

    Moderator-for-the-day Professor J. W. Milliman opened  with  comments about the inadequacy

of the benefit-cost analysis for dealing with larger systems, particularly in river-basin planning.

There is a need to fit public-investment decision  making into schemes of regional growth models

and models of regional economics.  Then, within that framework, benefit-cost analysis does make

some sense.

         I started out making my  living as an economist using  the tools of benefit-cost analysis in
    the design and analysis of projects.  Over the years I have become impressed with inadequacy
    of benefit-cost analysis for dealing with larger systems and with problems inherent in systems
    of public investment, particularly in the case of river-basin planning. I feel more and more
    strongly that the tools of benefit-cost analysis are not adequate until we have public-investment
    decision making fitted into schemes of regional growth models and models of regional economy;
    and once within that  framework benefit-cost analysis does make sense.  Prior to that, for the
    larger more basic questions  of development and growth we need to explore much more thor-
    oughly the questions of regional growth models and regional economics.

         River-basin planning and regional economics developed in the economic literature more
    or less  as two separate animals.  River-basin planning  in the economic  literature (in  the
    1930's with the Tennessee  Valley Authority, Columbia River Basin Project, Colorado  River
    Project) should have showed  that these regional problems would have been a source of theo-
    retical inspiration for regional economics.  The fact of  the matter seems to be  that river-basin
    planning at least until recently has not been responsible for new developments in regional
    economic theory.  Nor has river-basin planning achieved its earlier promise of being truly re-
    gional in character.  The best survey of regional  economics [John R. Meier, Amer. Econ.
    Review, March 1965], however, did not include a single  reference to river-basin planning. In
    river-basin planning we believe that there are no generally accepted procedures for forecast-
    ing in the planning process (regional forecasting, regional planning); however, both factors
    should be mutually supporting and  proceed in a joint fashion.  Initial planning guidelines must
    be established, then  a preliminary forecast model be constructed.
         Preliminary projections then will reveal a need for additional information about additional
    variables, which in turn will  modify the planning assumptions.  The refinement  of plans and
    forecasts usually begins from nature, and as  a process it  is continually winnowing and expand-
    ing.  Because of this we are now coming more and more to believe that it is desirable to con-
    struct a projection model in mathematical form which can be manipulated on a computer in
    order to illustrate the effects of alternative assumptions about goals, plans, and projections.
    Mathematical formulation serves to make alternative assumptions explicit and provide a sys-
    tematic framework for quantitative analysis.  In this sort  of framework, the technique of com-
    puter simulation seems particularly well suited for the iteration of planning and projection.

         The model is not a device for producing  single-valued projections or not even for produc-
    ing optimal solutions.  It may, more importantly, be a means of facilitating understanding of
    complicated systems of relationships that are relative to policy making.  An important existing
    illustration of regional model building by simulation technique is perhaps the Harvard model
    and the successor Lehigh model.  The Hawaii planning model appears to be the  most flexible

                                            39

-------
tool but is still in need of vast improvement.  It has an unusual feature: unlike any other
model, when it grinds out projections of likely future income and employment under alternative
public investment assumptions, it does distribute income by income class.  It gives us a feeling
for income redistributional consequences of various kinds of public policy measures.  It's one
of the first to  get at some of the social questions  more directly and explicitly within the model.
Another feature of the Hawaii planning model is that it is kept  as a planning tool and is contin-
ually revised,  updated, and rerun in light of new information and new policies, etc.  The model
is used as a planning tool not to grind out single-valued projections which you are stuck with
but continually gives revised projections in light of new technical information. This gives a
feeling for policy.

    Turning to a river-basin model based upon simulation, we can look at the Harvard and
Lehigh models, which start out with an attempt to specify optimal design  in optimizing operat-
ing procedures.  These operating procedures are based on linear programming techniques.
The programming techniques necessarily require simplification of the original problem.  They
use target outputs and benefit functions (highly restricted and  simplified), and a rigid system
of priorities had to be imposed. In many cases operating procedures for the facilities were
fixed, and there are many sorts of dynamic feedback that necessarily had to be omitted. The
Lehigh model  simulates mean monthly flows and 3-hour flow at flood peaks measured at 6
reservoirs and 9 hydroelectric power plants.  With systems design there is an associated
unique operating policy and a schedule of priorities for the use of minimum flow.  Economic
benefit functions were associated with targets for water supply, flood control,  recreation, and
power.  These economic  benefit functions were used to evaluate the results and point toward
optimal designs and optimal operating procedures. A regional economist using this model
doesn't have much work to do.  The economic projections and  economic benefit functions which
were developed outside the model were merely specified exogenously, and the simulation was
largely engineering and hydrologic with predetermined economic benefit functions, predeter-
mined targets, and economic projections.

    There are differences of philosophy regarding river-basin planning and simulation vis  a
vis the Susquehanna model and the Lehigh model. The use of optimizing and programming
techniques may limit the researcher in the complexities of the problem he can tackle.  The
mathematics for programming under many complex situations simply does not exist.  There
are an extremely large number of basic difficulties of public policy in attempting to specify
a social welfare function.

    For a large region (the kind that really makes sense to  plan for), there are many publics;
there are  conflicts between these publics.  These conflicts are not  easily resolved.  It's diffi-
cult to know what a social welfare function might  look like.  It  may be impossible to construct
a meaningful social welfare function in many  cases.  For this  region, simulation which does
not require an optimizing solution is a useful technique. Another major philosophical point is
the relation of regional model building and river-basin planning. The two strands of analysis
in regional economic growth, river works construction and river works planning, are part or
should be  viewed as part of the same general system.  Regional economic analysis and river
basins should  take account of the effects of the economy upon the water and possible feedback
from the water sector upon the economy.  It is incorrect to plunge  into river-basin planning,
certainly on any large scale, by designing optimum management systems for the water varia-
bles without first having  a general regional economic  model of the  basin which includes the
water sector.

    A general regional economic model can be fairly  complex and  sophisticated if necessary,
or fairly crude and highly simplified. But even when the feedback is not  strong from the water
sector back upon the economy, we need to start from the framework of a  general economic
model, crude or sophisticated, and have this model include a water sector,  in order to engage
in a kind of river-basin planning that ties the hydrology and the water variables to the economic
and social systems that we really want to make it a part of.  Also,  we must make public in-
vestments in the water sphere congruent with other important  kinds of public investment vari-
ables with which we must deal, and which we  must make more sensible in a planning spirit.


                                         40

-------
        And it is in this spirit that I believe the planning organization might view the guidelines
    offered by the Lehigh simulation model. You may include some skepticism, or at least worry
    about optimizing procedures for design and optimizing operations of those design variables
    after they are placed within the larger framework of regional economic growth and regional
    economic  systems.   Ron Hamilton, manager of the Susquehanna model, took a group of in-house
    professionals and outside  academic consultants and  managed to produce a simulation model
    which is in some respects unique.  The simulation involved for the first time an economic
    model that has direct ties with a demographic sector (economic sector with direct interdepen-
    dent ties between them).  All other large regional models have computed  the supply of labor
    independently, by standard demographic techniques.  Here the two sectors are brought together
    ad hoc and sort of rationalized. This  model does have, however crudely,  the economic and
    demographic sectors.  Then you grind up assumptions about age groups, migration rates, etc.,
    directly tied to the economic sector.  It is also a river-basin model which generates demands
    upon the water within the model, directly related to  economic activity.  The demands upon the
    water are not computed once the economic projections are done.
    [In the original Proceedings report, Rolf Deininger's paper, "Systems Analysis of the Great
Lakes Area — The Physical Subsystems,"  was reproduced in its entirety.   The reader is referred
to the updated version,  Appendix E.]  Professor Deininger summarized his remarks by indicating
that a complete modeling of water quantity and quality of the Great  Lakes  region is a desirable un-
dertaking. Based on currently available knowledge, the water-quality modeling will present diffi-
culties due to the fact that the  technological relations are not known very well. The modeling of
the water quantity seems to hold more promise  and is the area where systems analysis  can contri-
bute immediately.
    The following list of questions and summary remarks (in the order they were raised) indicates
the sort of thinking that  characterized the  afternoon proceedings.
    	  Has anyone analyzed major economic implications of regulating the levels of Lake
          Superior or the other lakes?
    	  Do we know enough about the lake processes to model Great Lakes water quality?
    	What about ground-water boundaries and influences?
    	What data are really significant in terms of our objectives?
    	Where shall the emphasis of new data collection be put?
    	What shall our priorities be for research proposals?
    	  Not  enough data are available on coastal currents, stratification, and the nutrient budget
          of the  lakes.
    	  Modeling should be tried to compare demographic influences on nutrient levels in the
          lakes.
    	  Lake-level problems are an appropriate target for systems analysis.
    	What is lacking in lake-level regulations is a sound management strategy.
    	  The quality dimension deserves  a high priority for physical and biological research.
    	  Some crude model building should be attempted by use of nutrient parameters.
    	  A critical area for analysis is the southern end of Lake Michigan.
    	We need both optimizing models and simulation models to do our job.
                                            41

-------
    	  Optimizing models are used for making the search more efficient.
    	  Nonlinear systems have greater sensitivity in certain regions, but this approach is very
          difficult to devise. We go into this with extreme humility.
    	  There are optimum levels of pollution vis a vis utilities versus cost.
    	  Consider water demand as a function of the price people are willing to pay for it.
    	Water law in the Great Lakes raises questions amenable to research.
    	  Development of "social indicators" may be a new area of discourse.
    	  The demand for cleaning up Lake Erie  may come from nonusers who express a need even
          though they have no direct economic or geographical contact.
    	We're concentrating almost too much on water and neglecting land-related problems.
    	  A "water-use" orientation will be helpful for focusing on research proposals.
    	  The real problem is what do we want to use the Great Lakes for.

                               Wednesday, 13 September 1967
    Moderator-for-the-day Professor Gerard Rohlich turned the floor over to Dean Stephen Smith,
who headed a panel discussion,  "Institutional Systems for Water Resources Management of the
Great Lakes."  The remarks summarized below include those of the other panel members (I. F.
Fox, N. W. Hines,  and S.  W. Havlick).
    Questions were raised about the way situations of public conflict are resolved into operative
decisions, how various interests of a region are represented in institutions, and how information
can be generated to provide a basis for public action.  Data are needed on existing patterns of in-
terest and influence in the Great Lakes. Private  interests, for example, are very strong and often
have very different objectives (vis a vis public interests) as to what should be done.
    Environmental corridors* which tend to be concentrated around water resources are in need
of investigation in terms  of legal, political, ecological, and social constraints.   Pollution of shore
lines and urban areas needs to be recognized as a significant organizational and jurisdictional
problem.  There are public and private, written and unwritten arrangements operating through law,
public policies, and administrative devices.
    Work needs to be done on how a public agency can be motivated to generate information on the
range of choices and alternative strategies of managing a part of the environment.  How are the
potentially vast number of choices sifted down to  an appropriate number for public consideration
and who, in the final analysis, makes the decisions?
    In order for practical research projects to result it was urged that focus be directed to specific
problem areas. Effects of various decisions would be fed back to the  institutions involved. Areas
    *Areas in urban or rural areas that provide quality opportunities for persons in contiguous land
units.  A corridor may contain scenic open space or structures of visual quality, for example.
                                            42

-------
of present and expected external diseconomies should help establish research priorities.  It was
argued that poor water-management decisions were not the result of apathy but rather "a  matter of
not knowing what the real alternatives are for the water decision-makers to examine."
    We should be mindful of how lawyers approach the problems; often the point where the critical
decisions are being made is influenced by lawyers or by individuals influenced by lawyers who are
client oriented. Important differences were stressed between "legal structure" and unwritten per-
formance which through custom becomes common usage.
    It was  mentioned that pollution-abatement agencies should consider the Ontario Water Resources
Commission arrangement and the relevant applications of the Genossenschaften activities  of the
Ruhr  Valley in West Germany.  The current study of "Institutional Design for Water Quality Manage-
ment" in the Wisconsin River basin was discussed.  A major difficulty is our  lack of precision in
quantifying the benefits from improvements  in water quality; e.g., for each increase in oxygen in
parts per million, what are the incremental  costs and benefits ?
    The suggestion that certain problem areas in the region be specific research targets came up
again as it  had in previous days and as it did throughout the rest of the week.  Work can hardly be
done on all of them  at once,  so some priority needs to be established in light of payoffs. The de-
velopment of a research strategy seems called for here instead of prescription in any detail of an
"optimum" solution to specific problems.
    Special efforts  seem needed to determine what institutions are effective and why,  who the cli-
entele are for various possible research activities, and who the beneficiaries are of pollution con-
trol and abatement.
    At one point in  the open discussion, considerable attention was given to flow of information be-
tween the private and public sector in water-management decisions.  Part of the difficulty in achiev-
ing desirable programs for environmental enhancement is that of defining the range of choice. It
was often mentioned that, with market mechanisms as they are, strong economic motivation exists
for investors and private entrepreneurs to propose only particular courses of action.  Even certain
public agencies have shown partiality to structural measures  as  solutions to water problems which
were  brought to particular agencies for resolution.  The need in almost every case is  to explore a
wide range of options designed to solve particular problems.  Incentives must be provided, and be-
fore additional governmental apparatus is erected, simpler pathways toward solution of subsystem
or subregion problems need testing.
    The following summary, presented in outline form, is extracted from a discussion agenda by
Smith and Fox. It covers the principal remarks of the Wednesday session.
                                             43

-------
I.   Social Systems
    A.  Systems which structure private and public action.
    B.  Systems which establish working rules for private action, e.g., market mechanisms, water
        rights.
    C.  Systems which make decisions for taking action (public and private).
    D.  Informal processes of behavior concerned  with definite courses of action.
II.  Research Orientation
    A.  Orient research around defined problems — existent and prospective. (The patterns es-
        tablished by Lakes Erie, Michigan, and Superior give us limnological and institutional
        planning opportunities of the past, present  and future.)
    B.  Identify the decision processes, formal and informal, relevant to the defined problem —
        understand the processes of behavioral response.
    C.  Understand system and intersystem interaction, which is a part of pluralistic adjustment—
        guard against unrealistic comprehensiveness on the one hand and excessive judgments on
        the other.
    D.  Identify, and quantify where possible, effects of  system performance—also express it in
        gross and net monetary terms if this  can be done.
    E.  Develop performance criteria.
        1.  Test effects of selecting alternative ranges of criteria.
        2.  Examine systems for determining criteria, e.g., generating information, achieving con-
           sensus, implementing capability.
III. Research Task
    A.  Define problems, e.g., pollution, commercial and sport fisheries, transportation, lake
        levels.
    B.  Define primary systems as they relate to the problem.  Define secondary and tertiary sys-
        tems as they relate to the problem, including feedback relationships.
    C.  Develop models of systems to test hypotheses and develop insight.
    D.  Relate performance to criteria.
    E.  Develop policy judgments for decision units such as LJC, states, Great Lakes Basin Com-
        mission, FPCA, Water Resources Council, operating agencies.

                 Thursday, 14 September 1967, and Friday,  15 September  1967
    Reports on Thursday and Friday of work groups and subsequent discussions are presented in
Section A.4 of this report.  The summary nature of the final day and a half permits a logical divi-
sion of the proceedings in this manner.  Presentations and questions on Thursday afternoon were
hosted by  Dr. Wilson.
                                            44

-------
    Mr. Raymond Clevenger, Chairman of the Great Lakes Basin Commission, brought to the con-
ferees fresh information and insights regarding the operation of the Commission.  The remarks were
especially instructive inasmuch as several efforts of the CIC and the Commission may provide  re-
ciprocal benefits.
    The existence of the GLBC is a demonstration of Congressional intent for a single agency to
be charged with the responsibility for the formulation of a single,  comprehensive plan for the basin.
The Commission's efforts will continue only as long as a majority of the eight participating states
want it to continue.
    Funding and staffing of the Commission was discussed.  States contribute 50%, and the Federal
contribution is  50%  (amounting to $300,000-$500,000 per year).  Major challenges will be to (1) de-
crease the  "perception of the problem" time lag, (2) formulate a comprehensive plan (within 2 to 3
years), (3)  obtain a  legitimate expression of the particular problems to be solved, and (4) provide
for a closer dialogue at the technical level (between universities and agencies), which should ex-
pedite  and integrate the participation in the work that needs to be done.
    After Mr. Clevenger's remarks, the representatives of federal agencies (see Section A.I) gave
data papers on respective retrieval procedures, difficulties presently faced, and programs or ex-
pectations in the near future. All of the discussions concerned the Great Lakes region.  Many of
the points may have specific value to particular CIC researchers in the immediate future. Individ-
uals who desire verbatim copies of the agency presentations are asked to correspond directly with
the CIC staff based in Ann Arbor, where the substantial "permanent appendix" of these and other
conference papers is maintained.

                     A. 4. CONCLUSIONS AND CONFERENCE SUMMARY
    Condensed reports of the work groups on Thursday and Friday have been incorporated into
this section.  The reader will discover a slightly elaborated  "laundry list" of problems which are
framed as possibilities for research among the CIC member universities. More important, how-
ever,  is a presentation of specific  researchable projects which emerged as reasonable targets for
the investment of scholarly energy and federal monies.  It was anticipated that, by the end of October
1967,  some effort would be directed to the preparation of firm proposals (withnames of investigators,
techniques, target problems, etc.)  by interested conference participants  or academic colleagues.
                                       ************

    Eventually specific research project proposals will be written by the respective project leaders
or principal investigators. Nevertheless an identification of useful systems  studies  and other
Great Lakes research which comprise specific problems is a present goal.   Consensus was reached
that the Lake Michigan basin will be a primary target for research, with special emphasis on the
                                             45

-------
southern portion of the lake and its environs.  International impediments, although not serious, dis-
couraged immediate  consideration at critical subregions of the St. Mary's River, the Detroit River,
and the Niagara River.
    The  idea that the southern portion of Lake Michigan be given hard scrutiny in terms of re-
searchable projects should not be construed as a sign to discourage CIC universities from other
research efforts in other problem-ridden subregions.  Problems in need of attention, data availa-
bility, interest  among a large  number of institutions, and several probable researchers who indicated
a personal commitment to the Lake Michigan region were leading factors in the consideration of
that sector of the Great Lakes basin.  Extreme care should be taken to design the subsystem models
as part of an overall strategy  because eventually a model of the entire basin will probably be feasi-
ble when the data and modeling tools are improved. In the summary reports,  several of the work
groups included diagrams and outline models to  symbolize some problem-solving approaches.
Two are  given here.
    The  subgroup on legal and social institutions (Professor Hines) came up with Figure A-l.  It
was suggested that institutional problems occur  on least at two levels:  (1) policy establishment by
the institutions involved in the political process, and (2) operating efficiency at the level of the
action agency.  It was agreed that the first  step would be to identify the  public  and private agencies
having some impact on the management of the region's water resources, and begin with a  familiar-
ization of their rules, processes, and actions.
    A schematic  model from the Ayer's "quality" group looked something like Figure A-2. To
utilize this model, the following tasks are indicated:
    Task #1.  Define the inputs, outputs, and state-of-the-art controls.
    Task #2.  (a)  Develop transfer functions between input, controls, and input pollution.
              (b)  Develop transfer functions between the elements of pollution and the general
                  pollution level and controls.
              (c)  Develop relation between pollution level and  effects.
    Potential research projects which gained recognition in the final days of the conference are
clustered into two broad areas — the institutional-socioeconomic and the physical-biological —
with understandable overlap in many instances.  The lists which follow are not intended to be ex-
haustive.  There is the implicit hope that it may suggest refinements of these  and other projects
which can be accommodated under the auspices of the CIC institutions.

Institutional and Socioeconomic Research Acitivities and Topics
    1. An analysis of the way water-management decisions are made in the (Lake Michigan) region
and what techniques  can be applied to assist the  decision-making machinery in producing and select-
ing from a wider  range of alternative programs  of resource management.
                                             46

-------
                                                         Feedbacks Affecting
                                                         Policy Formulation
               Economic-Demographic
                    FIGURE A-l. A PROBLEM-SOLVING APPROACH
   Causes
                       Man-Caused and Natural Pollution
                                 (algae, etc.)
Natural     -
Industrial   —»•
Urban (people)
Agricultural -
Elements
   of
Pollution
  Circulation
    Natural
decomposition
	Flushing
General
Pollution
Level
                                  Pollution
                                   Effects
            Control prior
            to Pollution
         Control on
            Lake
               Social and
               Political Control
    Effects

 Fish
 Bacteria
•Recreational
-Purification
••Industrial
 Natural Beauty
       L
            Economic Model Interlinked
            with the Above Boxes
                                      on a Regional,
                                Local, or Subregional
                                Scale
                      FIGURE A-2. AYER'S SCHEMATIC MODEL
                                          47

-------
    2. An analysis of the flow of information for planning and policy activities between and among
the private and public sectors of the subregion.
    3. An identification of water-management influences and influential factors as a part of the
regional or subregional political power structure.
    4. Operational efficiency and effectiveness  at various levels of water-related organizations.
    5. How individuals, groups, organizations, and other public agencies express themselves in
the process of arriving at a consensus about what should be done with the available water resources.
    6. What media and vehicles can be applied most effectively (and how) in water-related infor-
mation dissemination.
    7. An evaluation of the prospect that traditional lines of authority and influence may be re-
routed from Congressional delegations to regional agencies of national and international stature,
particularly by state and metropolitan governments.
    8. How values are expressed through our present institutions, including the International Joint
Commission and the Great Lakes Basin Commission.
    9. How an organization such as the Great Lakes Basin Commission needs to be assessed for
its strengths and weaknesses in terms of  reflecting social values, resource planning program
strategies, etc.
    10.  A study of conflict situations among competing uses of Lake Michigan with respect to
tradeoffs between (a) recreation and waste disposal (pollution), (b) commercial fishing and pollu-
tion, (c) one form of recreation versus an incompatible form of recreation (power boats vs. duck
hunters  or canoeists), and (d) power and maximum shoreline use and development.
    11.  Design of a feasibility study to determine an approach to a Lake Michigan model which
can eventually be plugged into a Great Lakes regional model.
    12.  The social (as well as economic) implications of water-resource management and develop-
ment  in this subregion.
    13.  How frequently water management  and development decisions are made by persons or or-
ganizations not directly involved as beneficiaries and/or contributors to the project cost.
    14.  How to monitor at a few critical  points  economic effects of changes in water quality as a
result of putting water-quality standards into effect.
    15.  What social change can be brought  about through planned intervention in conjunction with
water-resource projects.
    16.  Determine the process of problem  definition within watersheds,  major lake basins, and
the entire Great Lakes  region and the priorities of the problems.
                                             48

-------
     17.  What heuristic research techniques may be applied to the decision-making performance
of operating agencies.
     18.  Determine the process by which public opinion on water-resource questions is formed.
     19.  How a concern for the environment is formed among individuals or groups, how it is mea-
sured, how it is expressed, and how the governmental sector incorporates this concern into political
and social action.

Physical-Biological Research Activities and Topics
     1.  Develop a quantitative model of the Great Lakes system with special attention directed to
lake levels and to associated flows and losses including augmentation and diversions resulting
from structural works.
     2.  Develop a water-quality model of Lake Michigan (or others) and its contiguous drainage
basin in terms of nutrient budget,  rate, and effects of fertilization.
     3.  Study the effect of evapo-transpiration  rates, ground-water gains and losses, and the neces-
sary degree of accuracy required  for these and other variables in a quantitative  model.
     4.  Establish relationships between various pollution levels and their effects in biological as
well as economic terms.
     5.  Define inputs, outputs and  state-of-the-art  controls with respect to a cause-and-effect
pollution model.
     6.  Develop transfer functions between input, controls, and input pollution, and also between
elements of pollution and the general pollution  level and controls presently in use in the Lake
Michigan situation.
     7.  Relate physical observations of lake levels to resultant values (and to beneficiaries).
    8.  Determine the output from simulation or  any physical model can be used most effectively,
with the eventuality that value judgments will need  to be made which force numerical rating tech-
niques.
     9.  Study irrigation opportunities in the region and the consumptive implications.
     10.  Determine spatial requirements  for recreation developments.
     11.  Consider physical, biological  (as well  as legal  and economic) effects of further diversion
or diversions of Lakes Michigan-Huron water.
     12.  Predict future water-supply needs and expected levels of pollution.
     13.  Determine the status and future of eutrophication in Lake  Michigan.
     14.  Determine the effects and mechanics of  coastal and offshore interchanges and mechanics.

                                             49

-------
    15. Collect more data on pelagic and benthic conditions in Lake Michigan.
    16. What simplified water-quality models can be applied to the entire Great Lakes system.
    Priorities of research in the physical-biological area emphasizes the question of institutional
arrangements and the way decisions are made.  "There are some techniques that could be applied
to this problem which would make sense to the researcher.  I think that as a matter of strategy in
writing our proposal, we might make sure that we don't stay with generalities. We keep falling
into the traps of economic and engineering approaches that we know how to do pretty well."
    There is a growing consensus, and it is highly critical, that we need to do some first-rate re-
search on legal, social, and institutional problems.  In writing this report we want to go farther
than just calling for more  research.  We need to be explicit in suggesting case studies. We want
to be explicit in suggesting particular kinds of institutions to look at. More important, we need to
stress in some detail the kind of analytical techniques and tools that are available.   There has been
reputable, scholarly work that has been done in some other fields that are now beginning to  grow.
The question of nonmarket decision-making studies on power structures, etc., hasn't been applied
to water organizations.
          The problem is  getting enough "meat" here to entice some research in this field, and to
    entice the kind of scholars involved.  It would be a good  strategy since the present steering
    committee does not have the kind of competence on it to "flesh this out." Someone should give
    some examples of research  approaches.  The examples  of scholarly techniques should be
    listed and how they can be applied to some given kinds of institutions and organizations.
    In conclusion, emphasis was placed on generating research projects that would  involve  cooper-
ation between schools, agencies, and the respective research personnel.  If the areas of research
are extensive enough to give an appropriate problem base, an advance will have been made in the
state of the art and a regional service can be rendered with wider national and international applica-
tion.  In order to sustain a degree of cooperation, coordination, and informational feedback, some
thought needs to be given about periodic appraisals and conferences by the steering  committee and
project leaders.  Some effort should also be made to involve additional colleagues and agencies to
achieve a more comprehensive dimension to the  Great Lakes research effort.
    Edmund Burke once said, "Those who would carry on the great public schemes  must be proof
against the most fatiguing delays, the most mortifying disappointments, the most shocking insults
and, worst of all, the presumptuous judgment of the ignorant upon their designs."  Even though this
quotation was never used or suggested at the conference it may serve to dramatize the  immensity
of the challenge for which the stage has been set and at the same time remind the university
scholar and resource  administrator that answers are needed that will have consequences (whether
we intend them or not) for the megalopolis of the Great Lakes well into the 21st century.
    Grateful acknowledgment is expressed to Alice Bond, Pat Nemacheck, Marilyn Schmits, Herbert
Heavenrich, and James Kerrigan for their help with the conference proceedings.

                                            50

-------
                                        Appendix B
                     A REPORT ON THE SECOND WORKING CONFERENCE:
                    CONSIDERATION OF GREAT LAKES SYSTEMS RESEARCH

                                   30, 31 October 1967
                                 The University of Michigan
                                   Ann Arbor, Michigan


                                   B.I.  INTRODUCTION

    The Second Working Conference, held on 30 and 31 October 1967, in Ann Arbor, Michigan, was

designed to advance the concepts of systems analysis and related water-research efforts in the

Great Lakes region.  The session was designed to serve as a follow-up of the First Working Con-

ference at Alpine Valley, Wisconsin, 10-15 September 1967.

    This appendix  is a summary of the Ann Arbor meeting.  Section B.2 compresses an annotated

agenda. Section B.3 traces the discussion and thought during the plenary sessions.  Section B.4 is

a summary and a look toward the final report.

    Participants in the Ann Arbor Working Conference are listed below.  The double asterisk (**)

indicates Project Steering Committee chaired by Professor Ackermann, and the single asterisk (*)

indicates staff member of the Council on Economic  Growth, Technology, and Public Policy  of the

CIC directed by Professor Carlisle P. Runge.

    William C. Ackermann**                   Raymond F. Clevenger
    Illinois State Water Survey                  Great Lakes Basin Commission
    P. O.  Box 232                              c/o Institute of Science and Technology
    Urbana, Illinois 61801                      The University of Michigan
    John C. Ayers                              Ann Arbor' Michigan  48105
    Great Lakes Research Div.                  Lyle E. Craine**
    North University Building                  School of Natural Resources
    University of Michigan                     The University of Michigan
    Ann Arbor, Michigan  48104                 Ann Arbor, Michigan  48104

    Robert C.  Ball                             Thomas D. Crocker
    Institute of Water Research                 Department of Economics
    Michigan State University                  University of Wisconsin-Milwaukee
    East Lansing,  Michigan 48823              Milwaukee, Wisconsin  53201

    David C. Chandler                         Rolf A. Deininger
    Great Lakes Research Division              School of Public Health
    North University Building                  University of Michigan
    University of Michigan                     Ann Arbor, Michigan  48104
    Ann Arbor, Michigan  48104
                                            51

-------
    William D. Drake*
    School of Natural Resources
    The University of Michigan
    Ann Arbor, Michigan 48104

    Richard D. Duke**
    Urban-Regional Research Institute
    Michigan State University
    East Lansing, Michigan  48823

    E. D. Eaton
    Office of Water Resources Research
    U. S. Department of the Interior
    Washington, D. C.  20240

    Carl Eifert*
    Council  on Economic Growth, Technology,
      and Public Policy of the  CIC
    2569 University Avenue
    Madison, Wisconsin  53705

    Robert S. Gemmell
    The Technological Institute
    Northwestern University
    Evanston, Illinois 60201

    George P. Hanna
    Water Resources Center
    The Ohio State University
    Columbus, Ohio  43210

    Spenser W. Havlick*
    School of Natural Resources
    The University of Michigan
    Ann Arbor, Michigan 48104

    Herbert S. Heavenrich*
    Council  on Economic Growth, Technology,
      and Public Policy of the  CIC
    2569 University Avenue
    Madison, Wisconsin  53705
    N. William Hines
    College  of Law
    University of Iowa
    Iowa City, Iowa 52240
James E. Kerrigan*
Water Resources Center
University of Wisconsin
Madison, Wisconsin

Dale D. Meredith*
3211 Civil Eng. Bldg.
University of Illinois
Urbana, Illinois  61801
Jerome W.  Milliman**
Institute for Applied Urban Economics
Graduate School  of Business
Indiana University
Bloomington, Indiana  47401

Clifford Mortimer**
Center for Great Lakes Studies
University of Wisconsin-Milwaukee
Milwaukee, Wisconsin  53201

George L. Peterson
The Technological Institute
Northwestern University
Evanston, Illinois 60201

Gerard A. Rohlich**
Water Resources Center
University of Wisconsin
Madison, Wisconsin  53706

Carlisle P. Runge*
Council on Economic Growth, Technology,
  and Public Policy of the CIC
2569 University Avenue
Madison, Wisconsin  53705

Stephen Smith**
School of Natural Resources
University of Wisconsin
Madison, Wisconsin  53706

William C.  Walton
Water Resources Research Center
2675 University Avenue
University of Minnesota
Minneapolis, Minnesota  55455
                 B.2.  THE CONFERENCE TIMETABLE AND ANNOTATIONS

Sunday, October 29

9:00 p.m.      Steering Committee and CIC staff meeting

Monday, October 30

9:00 a.m.      Opening  remarks by Professor William C. Ackermann: a review of the charge
              spelled out in the OWRR contract and a discussion of tasks which need consideration
                                            52

-------
9:15 a.m.      The Alpine Valley Report summary and review by Professor Spenser W. Havlick
              The reporting procedure was described with a listing of researchable projects.
              Three general conclusions were reported:
                  (1)  A system-wide quantity model can be designed.
                  (2)  A crude beginning can be made on a quality model for a subregion, e.g.,
                      southern Lake Michigan or Lake Erie and western Lake Superior.
                  (3)  Arrangements with other organizations, including the Great Lakes Basin
                      Commission, need to be considered and related to (1) and (2).
              The report was received without objections.
9:35 a.m.      Dale Meredith presented a report which  reviewed systems engineering, modeling,
              and gaming in water-resources research.  This was the result of a  literature search
              of major efforts in water-resource modeling and systems analysis.
10:00 a.m.     Coffee break
10:30 a.m.     General discussion on water-quantity model
12:15 a.m.     Luncheon
2:00 p.m.      General discussion on water-quantity model (continued)
3:45 p.m.      Coffee break
4:15 p.m.      Remarks by E.  D. Eaton, Associate Director, Office of Water Resources Research,
              U.  S. Department of the Interior
5:30 p.m.      Discussion
6:00 p.m.      Dinner
9:00 p.m.      Steering Committee and CIC staff meeting
Tuesday, October 31
8:45 a.m.      General discussion on a water-quality model
10:15 a.m.     Coffee break
10:45 a.m.     Discussion on research strategy
12:30 p.m.     Lunch
1:45 p.m.      Quality-model considerations
2:30 p.m.      Individual commitments of research interest
4:00 p.m.      Adjournment

                         B.3. HIGHLIGHTS OF THE CONFERENCE
    The Second Working Conference on Great Lakes systems research began with  a statement of
purpose by Professor Ackermann.  The group was charged with considering the feasibility of a
systems model for the  Great Lakes region. The  obligation to prepare a final report may require a
total of three meetings, but the task for October meeting included making a judgment on the feasibil-
ity of  various models and exploring what university personnel and facilities are interested and
available.
    An abstract of the  Alpine Valley meeting was presented as a  review for participants who were
at the first conference  and as an introduction for the Ann Arbor conferees who had not been present
                                            53

-------
at Alpine Valley.  It was suggested that Professor Havlick's report on the proceedings (after con-
densation) be included as Appendix A to the final report.  Mr. Meredith's "Survey of Systems
Modeling" will also be attached to the final report as Appendix  C.  Therefore an elaboration of
these two papers  is not provided here.
    The discussion reviewed and picked up the considerations which were left unresolved at the
meeting at Alpine Valley.
    Liaison with  the Canadian counterpart Of the  CIC and appropriate governmental units needs to
be established.
    The water-quantity model, which is under the guidance of Professor Deininger, was discussed
in considerable detail.  A two-year project is anticipated for the preliminary model of lake stage
and water quantity.  Dr. Deininger was invited to  proceed with the proposal formulation.
    The work under way and anticipated in Ohio was reported by George Hanna. The remarks in-
dicated primary research interest in water recreation and biological areas of the Great Lakes
basin with special focus on Lake Erie as the target subregion.
    Professor Duke called attention to the gaming-simulation research capabilities of the METRO
model which was  discussed in Mr. Meredith's paper.  The simulation of the METRO model should
be recognized as  an educational device to  sensitize decision-makers to effects of decisions, whereas
mathematical models  running on a computer are usually  more helpful as predictive tools for water
and land management. Interest was generated in  the capabilities of the gaming research for look-
ing at what Dr.  Craine called "the knotty institutional and organizational questions" of the southern
Lake  Michigan subregion.
    A suggestion was made that the CIC staff investigate the research capabilities  and efforts of
some of the  more sophisticated regional planning commissions, particularly those of southeastern
Wisconsin, northeastern Illinois,  and southeastern Michigan.
    Then the discussion returned to the quantity model.  Ways  of dealing with chlorides and other
conservative elements were mentioned. The difficulty of estimating evaporation and ground-water
inputs and losses was emphasized.  Whether or not the quantity model would "drive" a water-quality
model was a point raised by Professor Gemmell.  This ultimately launched a mild controversy
about the reasonableness of the kind of model that should come first, how the needs of a second-,
third-, and fourth-generation model could best be met, and where  the early payoffs might be ex-
pected.
    Before the coffee break, a "roll call" of back home interest was requested. Several partici-
pants felt that after the Ann Arbor meeting they would be in a better position to obtain reactions
about possible research interests and about who might be available during the next  academic year.
The majority of responses showed an interest in research related to the water-quality model.  It

                                             54

-------
was here that the first serious suggestions of interuniversity collaboration were indicated.  It was
suggested that specific research projects be sent to Ann Arbor for compilation.
    Mr. Eaton's observations and comments concluded the formal session of the first day.  The
message which was extracted from a recent paper  given at a Symposium on Water Resources Re-
search restated the general challenge of transferring research results to practice; of identifying
the multitude of physical, social, and economic interactions; and of improving the decision-making
mix of water-resources management through methods of systems analysis, with special emphasis
on a development of planning  methods in metropolitan situations.
    Informal discussions proceeded into the evening.
    On Tuesday,  31 October  1967, significant progress was  initiated by Professor Rohlich's sug-
gestion of a water-quality model which articulates  the possible tradeoff between waste-water treat-
ment methods or  costs and raw-water treatment costs.  The analysis would focus on the intermediate
uses which could  be made of the receiving water under alternative degrees of waste treatment and
treatment for water supply. Interest and approval  were immediate on the part of the participants.
An array of reasons were offered which supported  a water-quality model whose major emphases
were on water-supply costs and efficiencies which  would accrue from various water-management
practices.  The research design calls upon the talents of the engineers, the economists, the biolo-
gists and limnologists, the  sociologists, and the lawyers.
     The same analysis was thought to be appropriate for "microsystem" studies for various
municipal operations for water supply, and for industrial water users.  Furthermore, several sub-
regions could be studied simultaneously throughout the region. After the costs associated with
water supply have been investigated, other water uses and water-quality needs  could be studied.
This of course would increase the list of parameters under consideration over  time and presumably
sophisticate our understanding of water quality in each of the Great Lakes.
    Researchers interested in the current relationships between municipalities, states, and
other units of decision-making would have a major task, according to Dean Stephen Smith. The
interdependencies which are in need of identification would be a major  research activity.
    Even though work is slated to begin on a southern Lake Michigan model, Robert Ball  and Dr.
John Ayers urged that both the total system and the subsystem could proceed at the same time.
William Drake felt that a modification of the METRO model could start in a relatively crude  way
as well.
    Consensus was reached that an analysis begin  with single urban and/or industrial units or
sub-subsystems.  Suggestions which marked forward steps in the discussion  included the  following:
    1. Values of  water supply in the Great Lakes need to be established.
    2. What economic data can be obtained on the  costs of water and waste treatment which  are
       attributable to eutrophication?
                                            55

-------
    3.  The "faraway" goal of water-quality management can best be approached through submodel
       systems which can be plugged together ultimately for planning and operational reasons in a
       region as complex and extensive as the Great Lakes.
    4.  An assessment of research interests and availabilities indicated where active research is
       presently in operation or anticipated.  A linkage  of these operations appears to be the  unique
       experiment in systems research which OWRR might be willing to support.
    The Ann Arbor conference adjourned with the understanding that the participants from member
CIC institutions and representatives of state water-resources centers identify how their inputs
might best be fit into the water-quality and water-quantity  models and other related Great Lakes
research activities.

                           B.4. SUMMARY AND FUTURE PLANS
    Several CIC staff meetings have  taken place as follow-up activities growing out of the Ann
Arbor working  conference. Major energies have been directed to the preparation of the final re-
port on terms established  by the Office of Water Resources Research.  A report of events which
have taken place since the Ann Arbor meeting will be reported to a gathering of the Water Resources
Steering Committee on December 6 and 7, 1967, scheduled at Des Plaines,  Illinois.
    However, it is appropriate to include here two interim efforts which may give the reader  a
clue about the direction  which is being taken as a result  of the Ann Arbor deliberations.  The first
is a suggested format for the final report. The combined efforts of Messrs. Meredith, Kerrigan,
Runge, and Havlick have produced these suggestions.

                                   A.  Suggested Format
                Report of a Consideration of a Comprehensive Systems Analysis
                               Model of the  Great Lakes  Region
1.  Introduction
    1.1.  Agreement between CIC and OWRR
    1.2.  Steering Committee and staff
2.  Results to  be achieved from macroscale and subsystem analysis
    2.1.  Guidance for planners
    2.2.  Enable researchers to  relate their efforts in advancing the entire system
    2.3.  Provide a valuable means of communication
3.  Summary of Working  Conferences
    3.1.  Procedure of research design
    3.2.  Abstracts of First and Second Working Conferences
    3.3.  Major water subsystems considered
                                            56

-------
         3.3.1.  Quantity subsystem
         3.3.2.  Quality subsystem
4.  Needed framework-research activities
    4.1.  Institutional arrangements
    4.2.  Physical-biological studies
    4.3.  Gaming-simulation
5.  Inventory of needed research in the Great Lakes region by institutions and potential researchers
6.  Organization required to coordinate water research considerations related to the Great Lakes
    system
7.  Appendixes
    Appendix A.   Minutes of First Working Conference, 10-15 September 1967, Alpine Valley,
                  Elkhorn, Wisconsin
    Appendix B.   Minutes of Second Working Conference, 30-31 October 1967,  Ann Arbor,
                  Michigan
    Appendix C.   A survey of systems modeling in consideration of a Great Lakes area systems
                  model
Appendixes D, E,  Working papers from the Systems Drafting Meeting, 17 November 1967,
           TJt  f~t
             '  '  Milwaukee, Wisconsin

                         B.  An Overview and a Preliminary Sketch of
                                   a Water-Quality Model
Systems Analysis Model of the Great Lakes Region
    One approach to developing a system for study and analysis of the Great Lakes region and its
associated problems would be to construct a diagram within which the major elements and sub-
elements were organized.  During the Alpine Valley conference a generalized scheme (Figure B-l)
was suggested to identify the major elements and to illustrate their general interrelationships.
    Social Values and Goals.  The foundation of the schematic is the box, "Social Values and Goals"
shown in Figure B-l.  Social values, as used in the scheme, represent the values used in develop-
ing objectives in the planning and legislative processes.  Among those values are respect for per-
sonal and public property  rights, the  market system, distributive equity, preservation of unique
natural environments and  historic monuments, as well as aesthetic values.
    In the past only benefits and costs  that could be measured by the market system or the "shadow"
benefits and costs that could be generated were used in quantifying the values associated with a
program or project under  consideration for government support. Other intangible values,  which
were often major elements in the decision process, were set through fixed constraints or direct
assignments. Several programs are under way to develop methods by which some of those intangible
values can be structured into some quantifying system.

                                            57

-------
3

-------
     Fundamental studies will be required to take advantage of the work now under way, to develop
new approaches and to make recommendations that may be incorporated into the planning process
for the Great Lakes.
     Resources. The  water resources of the Great Lakes region among the greatest on earth; the
land surrounding these mighty lakes help provide the great strength of both the United States and
Canada.  In Figure B-l the resources for the region are  separated into two categories: land and
water.  Land includes any physical resource that is not directly attributed to the water resources
of the  Great Lakes basin.  The water sector is divided into the  use and value areas of water quantity
and water quality.
     It can be shown that there is a relatively minor interrelationship between the quantity and
quality aspects of  the water resources in the basin.  That is quite different from the relationships
found in a river system where the stream flow has a major influence on the degree of water quality.
     Because of the nature of the water-quantity  system,  it is best analyzed on a multilake or total
basin basis. The quantities of water in surplus and in  demand in  connected lakes will require a
total system management scheme  if  optimal operating  procedures are to be developed.  The U. S.
Corps of Engineers is currently studying the question of  lake-level regulation for the basin.  To
complement that study, the CIC universities are developing a number of optimizing schemes that
may be useful.
     On the other hand, the  water-quality problems facing the basin may better be viewed as local
problems from the physical standpoint. It is the opinion  of many that generally there is a weak
correlation between the water quality of connected lakes  within the  Great Lakes system.  The major
effect  is  localized  along the inshore waters. It is primarily the shore waters that receive urban
waste and drainage while being used  as water-supply sources.
     Economic-Demographic Development.  The component of Figure B-l titled "Economic-
Demographic" includes the general system where the economy of  the region is modeled and where
projections of development are made and analyzed. A  strong interacting element to the economy
will be the demographic characteristics of the region.  Both the population and economy will depend
upon the  social values selected for the region and the land and water resources available for alloca-
tion for use and for preservation.
     This component in the overall structure will act as the accounting system to determine whether
the resources are  being allocated in  the best manner and conform to the social values and goals
identified for the basin.
    Institutional.  The organizational framework under which the  region is operated and governed
is the next component, titled "Institutional" in  Figure B-l. Within the region two countries, eight
states, one province, and several hundred local governments have responsibilities  associated with
the  Great Lakes.  The interrelationships between formal  governmental units will shape the  policies
                                             59

-------
controlling the management of the region's resources.  In addition, the commercial, industrial,
labor, and recreational associations will influence the system within their respective areas of in-
terest.  Studies are needed to identify the current status of those interactions and to detail altera-
tions that will enhance the present system.
    Legal.  In every workable framework there must be a system of rules that establishes rights
duties.  The component labeled "Legal" will cover the system of laws  and rules governing the over-
all system.  Supporting studies will be required in this area.
    Water-Quality Model.  During the Ann Arbor meeting a number of suggestions were made as
to how a water-quality model might be approached.  A few of these ideas were combined into  a uni-
fied scheme, a discussion of which, with illustrations, is presented in  Appendix F, Section F.2.
    It was suggested that a first-cut water-quality model should define the municipal use of lake
water.  This would include the costs associated with treating lake water to meet given drinking
water quality standards and the costs associated with treating the municipal wastes before they are
discharged into the lake.  It would be useful to determine the cost of waste treatment that would
meet given effluent standards or water-quality levels.
    Similar studies could be conducted to define the cost and direct benefits associated with treat-
ing or controlling the influent and effluent waters associated with other use sectors, such as urban
runoff, industrial use,  thermoelectric power generation, fishing, and recreation.
    The difficulties of establishing the relationships between the degree of water  quality in the
various parts of the lake and the influx of degrading  and enrichment materials are fully appreciated.
However, as more information is gained in this area it should be incorporated into a useful model.
                                              60

-------
                                         Appendix C
                                WATER SYSTEMS MODELING
                                      Dale D. Meredith
                                     University of Illinois

                                   C.I.  INTRODUCTION
    A model is a formalization of the relationship which may be seen as existing among data, and
a system is a set of objects, together with relationships between the objects and between the proper-
ties of the objects.
    Simulation as used here means a particular type of model which pictures the functioning of
complex, dynamic systems as they change in time.
    To be acceptably accurate, quantitative behavior models of  a system the size of the Great
Lakes area system must inevitably be complex, yet must be feasible to operate.  These require-
ments were not compatible until high-speed computers were available.
    The ideal model would specify completely the properties of the processes that occur in all
relevant components of a system.  Our  knowledge and techniques do not permit more than a rough
approximation to this ideal.
    A review of the state of the art in system  model building that would be applicable to a Great
Lakes area system model is presented.  This is followed by some alternative approaches to the
development of a Great Lakes area systems model itself.  The author leaned heavily upon material
from a forthcoming book, Systems Simulation for Regional Analysis:  An Application to River Basin
Planning [1] for parts of this paper.

                  C.2.  STATE OF  THE ART IN SYSTEM MODEL BUILDING
    Sections C.3 and C.4 review the  state of the art in system model  building as it would influence
building a model of the Great Lakes area system. Two schools  of endeavor working on modeling
techniques may be distinguished and will be discussed as follows:
    (a)  Complete system modeling,  i.e., modeling of the entire  system.
    (b)  Subsystem modeling, i.e.,  complete specification or complete modeling of each subsystem
        or each element in the system.
                                            61

-------
                             C.3.  COMPLETE SYSTEM MODELS
    The state of the art in building complete system models for economic study can possibly be
best illustrated by describing some of the more important models which have been developed in
the last few years. This survey is not complete either in terms of coverage or depth, but does  show
what can and has been done in model building for complex systems.
    The models discussed are: (a) the New York Metropolitan Region Study, by the Graduate School
of Public Administration, Harvard University, for the Regional Plan Association; (b) the Upper
Midwest Economic Study, jointly undertaken by the Upper Midwest Research and Development Coun-
cil and the University of Minnesota;  (c) the Ohio River Basin Study, by Arthur D.  Little for the
U. S. Army Corps of Engineers; (d) the California Development Model, for the State of California;
(e) the Oahu, Hawaii, Model, for the State of Hawaii; (f) the Lehigh Basin Simulation Model, by the
Harvard Water Program, Harvard University; (g) the Susquehanna River Basin Model, by Battelle
Memorial Institute; (h) a Regional Economic Simulation Model, by the Southeastern Wisconsin Re-
gional  Planning Commission; and  (i) the METRO Model, developed for the Tri-County Regional
Planning Commission, Lansing, Michigan.
    A  1965 survey by Abt Associates, Inc. [2] includes descriptive typologies of over  50 repre-
sentative current social, political, and economic models, computer simulations, and human player
games. The Abt study includes a  brief summary of the staffing, time, and money requirements for
some of these  model projects.

C.3.1.  THE NEW YORK METROPOLITAN REGION STUDY [3, 4]
    This study analyzed the major economic and demographic features of the New York Metropoli-
tan region, a 22-county expanse covering 7000 square miles (in parts of three states), and made
projections for 1965, 1975,  and 1985.
    The model makes two independent projections of population and the labor force.  Taking into
account birth rates, deaths, and migration by age groups, population projections were first made
by standard demographic techniques.  Employment and population projections were then derived
from a separate model of economic activity.  The two projections were reconciled by  allowing the
economic  projections to stand and increasing the in-migration in the demographic model.
    The complete model contains 47 linear equations for 47 variables and is designed to forecast
employment, output, and value added for 43 industrial groups for the years 1965,  1975, and 1985.
In addition to the outputs of 43 industries, it generates estimates of disposable personal income,
total population, and employment for domestic servants and government employees.
    The projection model was designed to utilize data and special projections made before the
study,  so that there was little opportunity to have the special projections fitted into the overall
model  before they were made.

                                            62

-------
    The employment in national industries in the region was assumed to be some constant "share"
of the projected total employment in the United States by industry for 1965,  1975, and 1985.  These
derived employment demands for the national industries was then used to derive an input-output
matrix for the region.  Total employment was derived by use of multipliers in the matrix, based
upon assumed local input demands, local consumption patterns, and local labor-force participation
rates.  Then the output and employment for each industry and estimates of  disposable personal in-
come and  total population were derived from the total employment.
    National  coefficients were used for the purchase of interindustry inputs for the local market
for the input-output framework. Consumption purchases from each of the 43 industry sectors were
computed  by a simple linear consumption function based upon population and disposable personal
income. Government expenditures were treated as a linear function of population with government
purchases from each local  market industry considered to be a fixed percentage of total government
expenditure in the region.  Disposable personal income was taken as a direct function of total em-
ployment,  and total population was derived from total employment by a single parameter.

C.3.2.  THE UPPER MIDWEST ECONOMIC STUDY [5]
    The Upper Midwest regional study was designed to develop basic data for  1960 and to make
projections of employment, income, population, and migration for 1975 for the region of the Ninth
Federal Reserve District.  This region includes Montana, North Dakota, South Dakota, Minnesota,
26 counties in northwestern Wisconsin, and the Upper Peninsula of Michigan.
    The research team assumed that extensive region-wide policies could be implemented on the
basis that  component states have shared many similar growth experiences, with employment closely
tied to the processing of natural resources.
    The formal model  is of the interregional multiplier type designed to provide estimates of in-
come and  employment under various independent assumptions about population increase,  migration,
and unemployment.  There  are no  feedback relationships between the demographic and employment -
and-income sectors.  The basic set of projections  for 1975 are  labeled as "neutral" projections
and are (a) that the National Planning Association projections for regions outside  the Upper Midwest
will be realized; (b) that the 1960 shares of the regional and national markets will remain the same
for the Upper Midwest  sectors; (c) that labor productivity will increase at nationally projected rates
in the Upper Midwest sectors; (d)  that income will increase at nationally projected rates  in the
Upper Midwest sectors; and (e) that the labor force will increase at the same rate as total employ-
ment in each  Upper Midwest state.
    The model concentrates upon  interregional trade flows to specific regions instead of  interin-
dustry sales to specific industries.  The assumed size and  stability of the flow coefficients govern-
ing trade flows to the various regions are now important rather than the size and  stability of inter-
industry coefficients needed in the New York model.
                                            63

-------
    Thirty-eight income-generating sectors plus exogenous estimates for agricultural and military
income are used in the model. Sales for each of the 38 sectors are computed for each of 15 regions
(the 6 states and 9 regions encompassing the rest of the world).  Income for each state and for each
income-generating sector is derived from  total sales by a parameter specifying the proportion of
sales accruing as income payments.   There were 690 equations, 921 variables, and 2784 parameters
in the model. Total state incomes expressed as direct functions of demands  by external regions
were derived by a greatly simplified model containing only 6 equations. The expanded model was
then used to derive employment and income for each of the employment sectors from the total state
incomes.

C.3.3.  THE OHIO RIVER BASIN STUDY [6]
    The Ohio River Basin model was to provide guidelines and data to the U. S. Army Corps of
Engineers and to public agencies interested in various kinds of water-resource-related investments
in the region.  The study concentrated on developing equilibrium estimates of the supply of and de-
mand for labor  by decades over a 50-year  period for the entire Ohio River Basin, which covers
parts of 10 states and 400 counties.  The county was used as the basic political unit for purposes
of data sources, so that the region boundaries would coincide with  the drainage-basin boundaries
as closely as possible. Difficult constraints upon data availability resulted because of the size of
the area and the use of the county as the basic unit.
    The  supply of and demand for labor were projected separately and then reconciled by averaging
the two sets of projections for each preceding projection year and  checking back to make all the
numbers consistent with the averaged projections.  The supply of labor was determined by assump-
tion made concerning labor-force participation rates and unemployment rates of the population es-
timates from a  demographic model based upon cohort-survival techniques involving estimates of
birth, deaths, and migration.
    The demand for labor was based upon a modified input-output  model by use of 29 separate sec-
tors (27 major industry groups, 1 government sector and 1 nonclassified sector). The total output
in the basin was taken as the sum of demands for each industrial sector; the  demands were derived
from the final demands for consumption, investment, government,  defense, and net exports, and
for interindustry demand.  A set of simultaneous linear equations  describing the demand sectors
was then solved for the total demand for labor.  The estimates for investment,  government expendi-
tures, defense expenditures, and net exports in the basin were derived exogenously from national
projections.
    The  modified input-output model used only 81 interindustry coefficients instead of the 729 re-
quired by a full 27-industry matrix.  The use of 81 coefficients allowed for substitution among in-
puts within each subtotal as long as the sum of inputs remained a fixed proportion of total output.
Also, national coefficients were used at the regional level.

                                             64

-------
    Separate projections were made for 19 subareas in the basin by fitting least-squares regression
equations to each subarea's share of total employment for each of the employment sectors deter-
mined from historical observations. The effect on growth implied by historical trends was damped
by employment of time lag.  Thus the projections did not take into account internal factors influenc-
ing subarea growth because they were not made independently of the projections for the entire
basin.
    The model assumes that water  will be available at all times in sufficient quantities and qualities
to support the project economy.

C.3.4.  THE CALIFORNIA DEVELOPMENT MODEL [7-9]
    The purpose of the  California Development model was to forecast personal income and employ-
ment  on a quarterly basis by major industry groups for the State of California  to 1975. The Phase
II model discussed here projected personal income and employment for 59 industry sectors by
quarters for the period  1963 through 1975.
    Multiple regression for income and employment for each of the sectors were derived from a
sample of 52 quarterly observations for the period 1950 to 1963. Wages and salaries served as
proxy variables for output.
    Export demands, interindustry  demands,  and local final demands are the demand sectors used.
The national economic projections by the National Planning Association are used to determine ex-
ports for 7 categories.  The interindustry and local final demands are then developed endogenously
from  the exports.
    This is a one-region model.  It depicts California and the rest of the world without feedback
from  California to the rest of the world.  The state economic growth is  generated by  exports to
the rest of the world. The model contains 130 equations, is linear in variables and  coefficients,
and is largely recursive.  Thus, the structural relations are expressed in lags  which are unidirec-
tional with respect to time.
    The sensitivity of the model to important assumptions and the meaning of forecast results be-
yond  1964 have not been explained.

C.3.5. THE OAHU, HAWAII, PLANNING MODEL [10,  11]
    This study began in 1963 to develop a model for economic planning and growth for the island
of Oahu, Hawaii. The first phase of the program was completed in 1965, and it was decided to ex-
tend the model to the entire  State of Hawaii.  There are no major structural differences between
the  Oahu model and the extended model.
    The model projects economic growth and the types and levels of certain kinds of exogenous
spending necessary to achieve  four planning goals. Specified in quantitative terms, these goals

                                            65

-------
were related to desired levels of population, to the shape of income distribution, to external pay-
ments outside the island, and to budgets of state and local government agencies. No discussion is
provided on the derivation of the planning goals themselves, but the state report indicates that pri-
mary attention was given to the achievement of certain levels of population in 1985 and the ability
of the economy to produce a sufficient number of jobs. The exogenous spending which is considered
to be driving the local economy are Federal defense expenditures, tourist expenditures, research
and development expenditures, and investment (public and private). A household sector plus 16 in-
dustry sectors which included a service sector and state and local expenditures, but excluded public
investment, are considered to represent the local economy.
    Exports to the  rest of the world are given a rather minor role.  Exports of sugar and pineapples
were determined outside the model based upon trend projections over the past two decades, and the
values of other exports  are determined within the model under  the assumption that given proportions
of total outputs would be exported.  Coefficients for most of the equations were determined on the
basis of time-series estimates.
    Total population was derived from a sum of total employment adjusted by coefficients measur-
ing labor-force participation rates and unemployment. This population is then compared with side
assumptions and calculations dealing with demographic factors such as birth rates, deaths, and net
migration.
    Plans are to keep the model in operation and to revise and keep it up to date in terms of new
information.

C.3.6.   THE  LEHIGH BASIN SIMULATION MODEL [12-14]
    The Lehigh model is a river-basin model which emphasizes the river and its hydrology and
relies upon simulation techniques.  The model is an outgrowth of  an earlier study of the Harvard
water program.
    The earlier study involved a hypothetical river system  involving 12 design variables consisting
of reservoirs, power plants, irrigation works, target outputs for irrigation water and hydroelectric
power,  and specified allocations of reservoir capacity for active, dead, and flood storage. Fixed
operating procedures, in conjunction with predetermined benefit and loss functions, were used to
generate outputs and the resulting net benefits by routing hydrological data through the reservoirs,
power plants, and irrigation systems for various periods of stream flows.
    The simulation of a simplified river basin on a digital computer and the development of mathe-
matical models for programming river systems were the major contributions of this study.
    The Lehigh Basin study extends the methodology developed in the earlier Harvard study to a
real situation—that of  the Lehigh River Basin in Pennsylvania.  The simulation analysis utilizes
long synthetic stream-flow traces derived by statistical analysis  of the historical record to trace

                                             66

-------
the behavior of the system over time, given a certain set of priorities for flow use and a set of tar-
gets  for outputs.  Economic benefit functions associated with the target for water supply, flood con-
trol, recreation, and power (the only four purposes studied) are composed of benefits which result
when the system exactly meets a predetermined target, and also of losses associated with being
below or above the target in any time period.
     The study attempted to specify optimal design and optimizing procedures despite the emphasis
on simulation.  The optimizing procedures are based in large part on linear programming tech-
niques and specified objective functions.  The target outputs and benefit functions were highly  re-
stricted and simplified, a rigid system of priorities was imposed,  operation procedures for facilities
were fixed, and many sorts of dynamic feedback were omitted.
     The Lehigh model is one of hydrology and engineering design which takes the economy of  the
region as given.   The compiled or binary object program for the simulation model uses almost all
of the memory locations of an IBM 7094 computer.

C.3.7.  THE SUSQUEHANNA BASIN STUDY [15]
     The Susquehanna model is a dynamic, mathematical model of the economy of the Susquehanna
River Basin.   The basin is first divided into economic regions which are modeled separately but in
like  manner.   The economic regions are then divided into  sectors. Each sector is then further
divided into subsectors. The sectors are demographic, employment,  water, and income. An
electric-power sector was used in the first version of the study but was not illustrated in the final
report. The income sector does not feed back on the economy; it only computes a version of "per
capita  income" based on wages, salaries, and selected transfer payments. The demographic,  em-
ployment, and water sectors are tied together in each subregional  model. The demographic and
employment sectors are separate from one subregion to the next but are tied together within each
subregion. The water sector in a subregion is dependent on the water sectors in the other subre-
gions upstream.
     The model is modular, such that the sectors are connected by simple relationships; thus one
sector can be changed without extensive modification of another.
     The demographic sector is composed of three factors — births, deaths, and migration.  The
two major ties between this and the  employment sector are migration and the labor-force participa-
tion  rates.  Migration occurs when the subregional unemployment rate differs from an assumed,
long-run national  rate and this alters the age-class structure  of the population.
     The employment sector  consists of export industries, business-serving industries,  and house-
hold-serving industries.  The market demand  for exports, operating through export industry em-
ployment, "drives" the model. The  export industries' growth  is determined by the relative attrac-
tiveness of the subregion to industry in relation to other areas and the demand for goods in relevant
                                            67

-------
market areas that can be supplied economically from the subregion.  The attractiveness is deter-
mined by a relative cost concept involving transportation and labor. The market demand for exports
is specified to the model.  Thus the model computes employment, given the export demand.
    The water sector is to simulate conditions relating to water quality  and quantity.  Based on a
lack of evidence of a basinwide water-problem, potential feedback on economic growth caused by
investments in flood control, irrigation, power, or navigation were  considered to be minimal or
nonexistent. The hydrological model used considered only low flows at "critical points" in the
basin.  The water sector did not feed back to the demographic sector, and the only feedback to em-
ployment is through river works construction and the recreation subsector.

C.3.8.  A REGIONAL ECONOMIC SIMULATION MODEL (SEWRPC) [16]
    The Southeastern Wisconsin Regional Planning Commission is  interested in providing a series
of forecasts of future regional population and employment levels that are sensitive  to alternative
public and private development policies, to 1990, for seven counties in southeastern Wisconsin.
    This is a dynamic input-output feedback simulation model with investment  being the dynamic
force in operation.  The model is organized into a number of sectors that are interconnected by an
input-output matrix.
    The flow relationships between respective sectors of the economy are mathematically expressed
as a series of balance equations.  One balance equation is required for each industry in the model.
This balance equation relates the output of the sector to the current purchases  and  capital invest-
ment of the other regional sectors.  There are 32 sectors at the regional level  (30 industries, re-
gional household, and government).  In addition, each of the regional sectors  has an input-output
and investment relationship with each of the 9 sectors at  the national level (6 industries, national
household, federal government, and foreign purchases).  This results in  1312 Input-output coeffi-
cients and the same number of investment coefficients.   Many of these coefficients  are zero.
    National household, federal government, and foreign purchases are forecast outside the model.
In addition, internal resource coefficients relating material purchases, capital  spending, employ-
ment, and wages in each industry and the input-output coefficients relating sales and purchases  in
each industry are needed before  the model  can be operated.  Sampling of individual firms, state
corporate tax records, state industrial employment and payroll records, and household survey data
were used to estimate the coefficients.
    Government investment is a programmed exogenous  variable since one of the primary purposes
of the model is to determine the  effects of public works investment spending on the regional
economy.  Investment is the dynamic force in the model operation and allows the balance equations
to be solved sequentially using the investment lag to generate a synthetic series of  employment
forecasts with time.

                                            68

-------
    In forecast runs, the present input-output matrix is assumed to remain constant over time.
The productivity is assigned an annual rate of increase based on recent historical trends. Employ-
ment forecasts are obtained and used to estimate population and land requirements.
    The model has no internal water sector.

C.3.9.  THE METRO MODEL [17]
    The METRO model represents a different type of study than those presented above.  The other
models are all in the form of mathematical equations and relations to be solved by a computer.
The METRO model cannot be solved by computer alone, but is a gaming-simulation involving the
participation of people to take care of the relationships which cannot be expressed in mathematical
terms with definite coefficients.  The people participating then become a part of the model.
    In the METRO game simulation, there is created a miniature world that represents the  typical
metropolitan area, problems and all.  The METRO model is designed to demonstrate to decision-
makers the effects of a series of individual decisions on  the metropolitan growth pattern.  It does
this by using a simulated, abstracted environment, with a reduced time span and dynamic  interplay
of current decisions  with fixed policies.  In addition it illustrates the kinds of data available to
dec is ion-makers and informs the decision-makers about the techniques that are available to evalu-
ate the implement decisions.
    The METRO model deals with ideal types of governments, budgets, issues, and policies, all
abstracted from Lansing,  Michigan, socioeconomic, demographic,  and political data for the period
1960-1965. There are three basic firms in the model: a large, heavy-industrial manufacturing
firm that is tied to the fluctuations of the national economy and has primarily blue-collar  workers;
a large,  slow-growing, and highly stable firm of clerical  and white-collar workers; and a  high-
growth, technologically oriented firm whose major product is innovation and which hires adminis-
trative, professional, and  technical personnel.  There are 5 household types in the region  which
are correlated with consumption patterns, residential mobility, and voter responses on political
candidates and issues by multiple regression from historical data.
    The number of people involved is large (for the Lansing model about 20 plus  operating staff).
The present model has two types of teams, a team of people representing an areal unit and a team
of people representing particular roles.  The areal teams are central city,  suburbs, and urban
townships.  The roles are politicians, planners, school people, and land people, with one judge.
The game starts at 1963 and goes from there.  It takes about 90 minutes to simulate 1 year.
    The relationships are dynamic, requiring heavy emphasis to be placed upon proper analysis
of previous results.  The controlling rules in the simulation are fairly  complex,  and the players
can at  best only approximate these relationships.  The extent to which they analyze previous re-
sults will affect their success in the future.
                                            69

-------
                                 C.4.  SUBSYSTEM MODELS
    The ideal model would specify completely the properties of the processes that occur in all the
relevant components of a system.  Subsystem modeling can be thought of as modeling certain parts
of the system before a trial at model the entire system.  Here the division of subsystem models is
between those which are not water centered and those which are water centered.

C.4.1.  NONWATER-CENTERED SUBSYSTEM MODELS
    The work has ranged from the modeling of individual household units to the modeling of the
national economy and population in a highly aggregative fashion.
    The reference for microanalysis might well be the work by Orcutt et al.  [18], in which individual
household units are simulated.  This is a disaggregation that is almost impossible to try to work
with as a component of a larger system  model, especially if that system were very large (such as
the  Great  Lakes area).
    Forrester [19] treats  the central framework underlying industrial activity. The approach he
presents is one of building models of companies and industries to determine how information and
policy create the character of the organization.
    Isard [20] sets  forth the techniques  of regional  analysis which have been  proved to have at least
some validity.  The virtues and limitations of the techniques are presented to allow the worker to
judge the applicability of the technique for a particular regional situation.  Techniques are pre-
sented for population projection, migration estimation, regional income estimation and social ac-
counting, interregional flow analysis and balance of payments statements, regional cycle and mul-
tiplier analysis, industrial location analysis, interregional and regional input-output analysis,  in-
dustrial complex analysis, interregional linear programming, and gravity, potential, and spatial
interaction models.
    A computerized model has been operated for assessing  indirect impacts  of water-resource
projects [21].  The strategy of the model is to  compare the  most efficient locational solution under
the  existing conditions with the most efficient solution under the new conditions, as altered by the
investment program for water-resource projects.  The locational efficiency is measured by average
delivered  cost of the commodity.  The difference in efficiency between the actual location patterns
before and after the water-resource investment is  assumed to be measured by the difference in
efficiency between the most efficient solution with and without the investment. The working model
is highly idealized.  It is for a single industry  with a standardized product and no  cross-hauling or
overlapping market areas.

C.4.1.  WATER-CENTERED SUBSYSTEM MODELS
    Amorocho and  Hart [22] have presented a general survey of contemporary methodologies in
hydrologic research.  This account gives a clear exposition of systems analysis and  synthesis as
                                            70

-------
applied to both linear and nonlinear systems. Perhaps the most widely known nonlinear modeling of
basin behavior for use with a digital computer is that developed by Crawford and Linsley [23].  This
model aims at representing the entire land phase of the hydrologic cycle for basins 10 to 90 square
miles. The model uses  a period of the record to determine the parameters for the watershed (soil-
moisture storage, stream-flow recession constants, ground-water discharge constants, infiltration
capacity, etc.).  Then, from daily evapotranspiration and hourly precipitation data, the model pro-
duces hourly stream-flow.
    Or lob and Woods [24] have outlined a general hydrologic model and presented two deterministic,
dynamic models for simulation of the operation of irrigation systems which determine, at any de-
sired point within the system, the proportion of water  which has been in prior contact with the sur-
face of an irrigation  plot.
    To indicate the fact that operations research could be used in water-quality management, sev-
eral simple models were developed at Harvard [25].  These included a queueing model for pollution
transport  in streams, a  model to minimize economic risk in sanitary engineering design, and a
model for determining optimal sizes of water-treatment plants.  These are simple models  to be
used in optimizing procedures, are restrictive in application, and are similar to the subsystem
models discussed in  the other Harvard Water Program publications [12, 13].
    The Aerojet-General Corporation [26] made a systems study for waste management in California.
This report outlined  the structure of waste-management systems and indicated some of the govern-
mental functions and organization required to implement the  systems approach to waste manage-
ment.  The study  was aimed at minimizing the cost of  waste disposal by "acceptable" procedures.
A lot of effort went into  indicating the  cost of various units in waste disposal and the various al-
ternatives.  A more detailed analysis was made of the Sacramento area.
    Bramhall and Mills [27] have presented  a computer program the data for which are taken be-
ginning at the head of a stream and proceeding down to its mouth, calculating at each specified
station the stream-flow, the dissolved-oxygen level (DO) and the biochemical oxygen-demand level
(BOD). The DO and BOD are computed by the Streeter-Phelps equation. As each tributary to the
stream is reached, the program begins at the tributary head  and follows it down to the confluence,
performing the  same calculations.  The model is deterministic, to minimize the cost of meeting
predetermined standards of DO and BOD.
    If the DO falls below or the BOD above the predetermined standards,  the wastes at the source
of pollution immediately above the violation are treated to a level sufficient to eliminate the viola-
tion.  After the  entire system has been treated where necessary, the stream-flow  is augmented at
a station predetermined to be a potential damsite. At  this higher flow level, all downstream sta-
tions are re-evaluated and treatment levels are correspondingly reduced.  Flow augmentation pro-
ceeds in this manner by 5% increments until all violations are eliminated  without treatment.
                                           71

-------
    At each flow level the annual dollar cost of waste treatment and the annual dollar cost of stor-
age to augment flow are computed and the minimum cost can be determined.
    The 86-mile stretch of the Delaware Estuary between Trenton, New Jersey, and Liston Bay,
Delaware, has been modeled by the Federal Water Pollution Control Administration [28].  The
economic and demographic sectors of the model were of the trend projection type and are not dis-
cussed here. The concern here is with the water-quality model of the estuary. The estuary is re-
sponsive to an approximate and semidiurnal lunar tide.
    Eight municipal waste-water treatment plants are assumed to be the waste sources.  DO,
alkalinity, coliform bacteria, chlorides, water temperature, and nitrogen constituents were the
water-quality parameters  monitored by weekly sampling runs during 1964.  The first four param-
eters are the primary indicators illustrated.  The study stops at the bay and is a one-dimensioned,
deterministic system.
    The basic quality system considered is for the DO and is composed of two subsystems, one for
BOD and one for DO.
    The estuary is divided into 30 sections, and a linear differential equation is  used to repre-
sent the mass balance for the BOD in each sector. A similar equation Is used for DO.  This results
in 2 series of 30 simultaneous equations each. The first solution is  made on the assumption that
the equations do not vary with time.  This assumption allows the utilization of matrix manipulation
techniques to obtain a set of transfer functions from the coefficients of the  equations.  This set of
transfer functions details the transformation from a waste-load input in any section to the stream-
quality output in any other section.   The total  effect at any section is then formed by summing for
that section the effects caused by inputs anywhere in the estuary.
    The solution of the equations for the time-vary ing situation were then  solved, and the model
was verified with past data which were  not under steady-state conditions.
    The cause-and-effect  relationships for nonconservative variables such as bacteria concentra-
tions are obtained by solving only those equations used for the BOD, with the proper decay rates
for these variables.  The decay mechanism is then eliminated in the BOD equations and used with
such conservative variables as alkalinity — pH and chlorides.
    Loucks  [29] has formulated deterministic and stochastic linear programming models to deter-
mine reservoir releases and allocations of water to meet some management objective; maximization
of total expected net benefits, minimization of total expected net losses, or  minimization of total ex-
pected deviation from each use "target."
    These are examples of the kind of  subsystem models that have been developed.
                                             72

-------
                  C.5.  GREAT LAKES AREA SYSTEMS MODEL SYNTHESIS
    This section discusses some of the important concepts to be kept in mind while the model is
developed, and then suggests some alternative approaches.

C.5.1.  MODEL CHARACTERISTICS
    A model is designed to answer specific questions.  The questions to be answered determine
the completeness required in the model.  The types of system behavior anticipated are important
in the selection of the factors to be included  in the model.
    A model is an abstraction from the real world and not an exact duplication. Models are neither
true nor false.  The value of a model is determined by the contribution  it makes to our understand-
ing of the system it represents.
    The ideal model would specify completely the properties of the processes that occur in all the
relevant components of the system.  The specification would be given in terms of physical param-
eters and would involve all behavior relationships within the system.
    The first major task is, then, to define the questions we want to ask and then see if we can
develop one  model that will answer them or whether we need several models.  Two purposes for
which the model might be designed are (1) forecasting the future of the  system, and (2) evaluating
the impact of alternative policies on the system [30].  Let us first look  at how we might approach
the problem of developing a complete system model to answer  our questions.

C.5.2.  COMPLETE SYSTEM MODEL
    After we have the questions we want to ask the model, we must then define the boundaries of
our model to correspond to boundaries of the real system so that these questions can be answered.
One of the first characteristics that is noticed of the Great Lakes area  is that it is an open system.
No matter how we define the boundaries of the system it still exchanges materials, energies, and
information  with its surrounding area.  The model will require the analysis of a large, complex
system  involving feedback,  nonlinearities and lag structures.
    The next step is to make a "rough cut" of the model to determine the areas needing further
research.  This could be followed by data gathering and further analysis so as to refine the model.
    Two avenues are open for  the model development.  One approach would be for a complete
mathematical formulation so that the desired answers to our questions  could be obtained by com-
puter calculations.  This is the approach used in the Susquehanna study. The other approach would
be a gaming-simulation study similar to the  METRO project in which players are involved to serve
as the "equations" which define the relationships we are unable to put into  mathematical form or
to which we  are unable to assign parametric  values.  Both approaches have merit.
                                            73

-------
    The complete computer model would allow a sensitivity study of various variables but would
be difficult to formulate in terms of the decisions that must be made by people in the actual system.
    The gaming-simulation permits the  use of greater realism in the analysis and evaluation of a
given set of policies in that the test environment can contain all of the characteristics that seem to
be important to the systems functioning. It puts people into the model who can, by using creative
abilities, design decision rules that help to advance the defined goals of the system and who can
provide reactions to the model builder regarding the realism of the simulation.  It is more flexible
than real-world tests or all-computer simulation because we can more easily change either the
computer or player parts.  It  is expensive because of the long time required to  make a run and the
number of participants involved. This often precludes sensitivity testing of results.

C.5.3.  SUBSYSTEM MODELS
    Subsystem models can be viewed as either subregion models or sector models. A complete
system model  could be developed by developing models for each of the subregions in the Great
Lakes  area and then tying the subregion  models together. These subregion models could be devel-
oped by different researchers as long as there was enough communication between them to include
the factors necessary to make the ties.  This approach would allow certain "critical" areas such
as the  area of  lower Lake Michigan to be modeled before the entire Great Lakes area was modeled.
    A  complete system model could also be developed by developing subsystem models for each of
the sectors in  the area. Many states are now involved in developing a systems  model of their water
sectors.  These "water models" might be evaluated and an effort made toward tying these together
to form a larger subsystem model of the Great Lakes area. This is an example of what might be
done in the other sectors as well. It has already been agreed that the water-quantity sector can be
readily modeled and offers an opportunity for quick results  in terms of an operating model (see
Appendix E).   The important point in the development of the individual sector models is that there
should be enough flexibility to tie them all together or at least be able to  use them together to an-
swer the questions we desire  of the comprehensive model.

                            C.6.  SUMMARY AND CONCLUSIONS
    Some contemporary attempts at developing models, for both systems and subsystems, are
reviewed.
    A  systems model of the Great Lakes area can be developed.  The degree of refinement required
in the model will be determined by the questions the model is to answer,  and the degree of refine-
ment will determine the amount of time and effort that must be expended  in developing the model.
     Two basic approaches seem to be available for a complete system model:  (1)  complete com-
puter simulation and  (2) a gaming-simulation.  Either approach could be formulated by working

                                             74

-------
with subsystem models which will be tied together, although the complete computer simulation
seems to lend itself more favorably to this procedure.
     The method to follow in developing a Great Lakes system model may be as follows:
     (1) Define questions to be answered by the model.
     (2) Make a "rough cut" of the model and determine the relevant subsystems to be included.
     (3) Examine the subsystems to determine the areas of the model which are well defined and
        those which are  least well defined.
     (4) Give highest priority for research to those  areas  of greatest weakness in the model.
     (5) Maintain a series of working conferences which meet periodically to keep  the focus on the
        overall model and its development. This step is important in that it enables the researchers
        in the various disciplines to see where their work fits into the model and also to see the
        kinds of results  they need to make their work compatible with the information needed by
        the model.
     It might be desirable to proceed with modeling of the well-defined subsystems  simultaneously
with step (4) in order to have part of the system model working as soon as possible.
     The above procedure could be used for either a model of the entire Great Lakes area or for a
subregion.
     The model could be modular, so that different sectors could proceed at different rates of de-
velopment  (e.g., the water levels in the lake could be modeled before the model to evaluate the
effects of these levels is completed).  The critical factor is that the various models must eventually
be compatible.
                                       REFERENCES
  1.  Arnold Zellner, J. W. Milliman, and H. R. Hamilton, Systems Simulation for Regional Analysis:
     An Application to River Basin Planning, in press.
  2.  Abt Associates, Inc., Survey of the State of the Art:  Social, Political and Economic Models
     and Simulations, made for the National Commission on Technology, Automation, and Economic
     Progress, Washington, D. C.,  November 1965.
  3.  Raymond Vernon, Metropolis 1985:  An Interpretation of the Findings of the New York Metro-
     politan Region Study, Harvard University Press, Cambridge, Mass., 1960.
  4.  B. R. Berman,  B. Chinitz, and E. M. Hoover, Projections of a Metropolis: Technical Supple-
     ment to the New York Metropolitan Region Study, Harvard University Press, Cambridge, Mass..
     1961.
  5.  James M. Henderson and Anne O. Krueger, National Growth and Economic Change in the Upper
     Midwest, University of Minnesota Press, Minneapolis, Minn., 1965.
                                            75

-------
 6.  Arthur D. Little, Inc., Projectlve Economic Study of the Ohio River Basin, Appendix B, Ohio
    River Comprehensive Survey, Vol. Ill, prepared for the Corps of Engineers, U. S. Government
    Printing Office, 1964.

 7.  John W. Dyckman, "State Development Planning:  The California Case," J. Am. Inst. Planners,
    Vol.  XXX, No.  2, May 1964, pp. 144-152.

 8.  John W. Dyckman and Richard P. Burton, "The Role of Defense Expenditures in Forecasts of
    California's Economic Growth," Western Econ. J., Vol. HI, No.  2, Spring  1965, pp.  133-141.

 9.  Richard P.  Burton and John W. Dyckman, A Quarterly Economic Forecasting Model for the
    State of California, Center for Planning and Development Research; Institute of Urban and  Re-
    gional Development, University of California, January 1966.

10.  Department of  Planning and Economic Development, The Hawaiin Economy:  Problems and
    Projects, A Report on the Economic Foundations of the General Plan Revision, Honolulu,
    Hawaii, March 1966.

11.  Roland Artie, "External  Trade, Industrial Structure,  Employment Mix and the Distribution of
    Incomes: A Simple Model of Planning and Growth," Swedish J.  Econ. 1965, pp. 1-23.

12.  Arthur Maass, et al., Design of Water-Resource Systems:  New Techniques for Relating
    Economic Objectives, Engineering Analysis, and Governmental  Planning,  Harvard University
    Press, Cambridge, Mass., 1962.

13.  Maynard M. Hufschmidt  and Myron B. Fiering, Simulation Techniques for Design of Water Re-
    source Systems, Harvard University Press, Cambridge, Mass., 1966.

14.  Maynard M. Hufschmidt, "The Harvard Program:  A Summing Up," pp. 441-445 in Water
    Research, ed. by Allen V. Kneese and Stephen C. Smith, The Johns Hopkins Press,  Baltimore,
    Md., 1966.

15.  H. R. Hamilton et al., A  Dynamic Model of  the Economy of the Susquehanna River Basin,
    Battelle Memorial Institute, Columbus, Ohio, 1966.                                ~

16.  Southeastern Wisconsin Regional Planning Commission, A Regional Economic Simulation
    Model, Tech. Report  No. 5, Waukesha, Wis., October 1966.

17.  Tri-County Regional  Planning Commission, M.E.T.R.O., A Gaming Simulation, M.E.T.R.O.
    Project Tech. Report No. 5, Lansing, Michigan, January 1966.

18.  Guy  Orcutt et al., Microanalysis  of Socioeconomic Systems: A  Simulation Study, Harper,
    New York,  1961.
19.  Jay W. Forrester, Industrial Dynamics,  The MIT Press, Cambridge, Mass., 1961.
20.  Walter Isard, Methods of Regional Analysis:  An Introduction to Regional  Science, MIT Tech-
    nology Press, Cambridge,  Mass., 1960.
21.  Edgar M. Hoover, Computerized  Location Model for Assessing  of Indirect Impacts  of Water
    Resources  Projects,  Working Paper CWR 1, Institute for Urban and Regional Studies,
    Washington University, St. Louis, Mo., May 1966.

22.  J. Amorocho and W. E. Hart, "A  Critique of Current Methods in Hydrologic Systems Investiga-
    tions," Transactions, American Geophysical Union, Vol. 45, No. 2,  1964.
23.  N. H. Crawford and R. K. Linsley, The Synthesis of Continuous  Streamflow Hydrographs on a
    Digital Computer, Tech. Report No. 12, Dept. of Civil Engineering, Stanford University,
    Stanford, Calif., July 1962.
24.  Gerald T. Orlob and Phillip Woods, "Water-Quality Management in Irrigation Systems," jJ.
    Irrigation Drainage Div., ASCE, Vol. 93, No. IR2, June 1967, pp. 49-66.
25.  Harold A. Thomas, Jr., and Robert P. Burden,  Operation Research in Water Quality Manage-
    ment, Final Report for Contract PH86-62-140,  Harvard University Div. of Engineering and
    Applied Physics, Cambridge, Mass., February  1963.


                                            76

-------
26.  Aerojet-General Corp., California Waste Management Study, Report No. 3056 to State of
    California Dept. of Public Health, August 1965.

27.  D. F. Bramhall and E.  S. Mills, A Computer Model of Stream Quality, for Maryland State
    Planning Dept., Publication No. 132  C, Baltimore, Md., June 1966.

28.  Federal Water  Pollution Control Administration,  Delaware Estuary Comprehensive Study:
    Preliminary Report and Findings, U. S. Department of the Interior, 1966.

29.  Daniel P. Loucks, Management Models for Water Resource  Systems,  Technical Report 1,
    Water Resources Center, Cornell University, Ithaca, N. Y., June 1967.

30.  Sheldon R. Simon et al., The Usefulness  of Computer Simulation for River Basin Analysis,
    Battelle Memorial Institute, Columbus, Ohio, March 1967.
                                            77

-------
                                        Appendix D
           A PRELIMINARY DESCRIPTION OF WATER-RELATED INFORMATION SYSTEMS
                                 FOR THE GREAT LAKES REGION
                            Thomas E. Barton and William D. Drake
                                 The University  of Michigan

                                      D.I.  PURPOSE
    To conduct a feasibility study for a coordinated information system to supply data require-
ments of the various systems models under way or proposed for the Great Lakes system, and to
prepare a proposed basic design for the structuring and implementation of such a system.  This
will include a statement of the needs, criteria, and guidelines for the long-range programming for
the proposed  system.

                            D.2.  STATEMENT OF THE PROBLEM
    There is growing concern among various specialists about the current condition of the Great
Lakes.  Speculations are that intensive management and control programs will be required if
deteriorating conditions are to be stopped or reversed.  Various research teams are proposing a
set of linked systems modeling studies which will  utilize simulation as a research tool in conjunc-
tion with the study of the Great Lakes.  While conceptually these will consist of a set of linked
models, they will be pursued by separate teams of individuals.
    Models such as these are prodigious consumers of data and,  as they become increasingly
sophisticated, the need for a thorough rationalization of data requirements and sources becomes
increasingly important. It is clear, then, that  CIC universities involved in this research program
will be collecting data on both primary and secondary source  levels for utilization in their  own
particular systems modeling efforts.  A thorough  and coordinated understanding of available data,
its sources, its limitations, and the needs and  characteristics for potential additional data, will be
of value to these research efforts and to other present or potential operating programs oriented
towards the Great Lakes problems.

                                D.3. RESEARCH PROPOSAL
    In order  that the fullest possible level of coordination and compatibility of these CIC systems
modeling efforts can be maintained, a clear need is present to provide a parallel and continuing
examination of data types and forms used in these research programs which would lead to  a more
standardized  methodology in data collection and handling.  This would be aimed at increasing the

                                             78

-------
potential effective utilization of data on an interstudy basis and would eventually increase the likeli-
hood of a compatibility and comparability of the resulting systems models.
    This proposal is to identify and clarify the data requirements for the specific set of systems
models which are being developed that pertain to the Great Lakes quality and quantity conditions
that may be used for management and control procedures in the Great Lakes basin.
    Several steps are proposed to examine the feasibility of data systems to accomplish the opera-
tionalization of such systems models:
    1. The identification of the characteristics of data collected  and used for the systems models
       which are proposed for the  Great Lakes area.
    2. The definition of already available data bases (present programs, the types of quantity of
       data, and the forms of storage  of this data, etc.).
    3. A review of the  characteristics of proposed or operating information systems.
    4. A review of the  feasibility of an organized data program for use with these models and a
       suggested structure for such a system. The study should follow an information  require-
       ments approach as the focus for the research effort concentrating on the specific require-
       ments of the models being proposed by the several  studies conducted under CIC's coordina-
       tion.

                                  D.4. RESEARCH PLANS
    The  identification of the characteristics of data collected and used for the systems  models
which are proposed for  the Great Lakes area.  Systems models referred to in this proposal are in
embryonic form. They  will change  in complexity and style as  the researchers learn more about
the Lakes and as more information  becomes available as input to the models.  The research teams
preparing the systems models will  recognize basic needs in the type and form of data for their
own particular  research program.  However, factors such as time and proximity normally will
prevent an effective common effort  at coordinating data collection and maintenance programs
among these teams.  For this reason, one of the primary objectives of this proposal is the identifica-
tion of common properties in data used for present or proposed systems modeling programs
through discussions with the research  teams and the maintenance of  a continuing liaison with these
various projects.
    The definition of already available data bases.  The investigation will identify present data
collection sources, the types of data collected, and the form in which it is available. This proposed
research acknowledges  the significant quantity of data generated in the normal course of activities
of various individuals, firms, and agencies, but which is  nevertheless not available to the research-
ers because of  conditions of: lack of awareness of need for the data, its format  or style, level of
                                             79

-------
detail, and numerous other barriers of this type.  For this reason, the study will be oriented toward
the purpose and activities of various Great Lakes oriented organizations, with a particular emphasis
toward rationalizing these considerations of the barriers to data utilization.
    A review of the characteristics of proposed or operating information systems. There are two
aspects to this portion of the research.  First is the review of experiences in certain other informa-
tion system design and implementation programs.  This entails  the selection of several most rele-
vant examples and studying the experience of procedures and problems experienced in these cases.
Second is the evaluation of the already functioning data supply programs which are relevant to the
Great Lakes research program in terms of their adequacy for future developments for research,
planning, and management programs for water resources. From this evaluation, recommendations
will be made for the improved compatibility and coordination among agencies collecting and dis-
seminating water and water related data important to the Lakes systems modeling research.
    Review of the feasibility of an organized data program for use with these systems models and
a proposed structure for such a data system.  The ultimate aim of the research is to propose a
feasible scope and design for an information support system for the present and future research,
planning, and management programs for the Great Lakes area.  This involves a number of aspects.
    1.  The investigation of organizational arrangements to implement and utilize improved infor-
        mation programs.
    2.  The exploration of the application of new information-gathering,  -handling and -dissemina-
       tion technologies which can assist in both research and decision-making.  This phase would
        include the investigation of the use of remote sensing devices, high-speed computers  and
        electronic display hardware.   An example of the use of new technology is embodied in a
        computer software system being developed at The University of Michigan which stores, re-
       trieves,  manipulates, and displays mapped regions in a variety of ways.   It is designed to
       be used in an on-line or remote terminal interactive mode, provide rapid feedback in
        graphic form, and allow exploration of many planning alternatives.  Designers may examine
        arbitrary areas of interest, address complex multiple questions about the area to the sys-
        tem, and perform mathematical analyses of the area.
    3.  Propose improvements for the "quality" (timing, sequencing, feedback) of information for
        research.
    4.  Provide plans for a system which will enable more efficient use of the limited number of
        highly trained professionals available to agencies through the alleviating of much of the
        effort they must expend upon locating, recording, transferring, and collating data records
        in order to carry out their analyses and dec is ion-making requirements.
    5.  Design and establish procedures to demonstrate the utilization of expanded water informa-
        tion programs  to decision-makers and government agency personnel.
                                            80

-------
                                         Appendix E
                        PRELIMINARY RESEARCH DESIGN OF THE WATER-
                         QUANTITY SUBSYSTEM OF THE GREAT LAKES
                                      Rolf A. Deininger
                                 The University of Michigan

                           E.I.  STATEMENT OF THE PROBLEM
    The Great Lakes region comprises an area of roughly 300,000 square  miles with a total popula-
tion of 28 million, which in the next 50  years is expected to grow to close to 60 million with a five-
fold increase in water demand.  Although in the northern part of the region there is an abundant
supply of good-quality water in the interior basins, the southern half of the region is meeting part
of its demands already from the  Great  Lakes and will  do so increasingly.  The major problems in
water-resource management will become more serious, one related to water quantity and one re-
lated to water quality.  This research design will deal mainly with the problem of water quantity
and its management over time and  space, particularly the questions of water supply and lake-level
control.  The hydrologic subsystem of the Great Lakes area consists of five interconnected lakes.
In order of decreasing  lake elevations these are Lake  Superior, Lake Michigan, Lake Huron, Lake
Erie, and Lake Ontario.  The  total  drainage and water area is about 295,000 square miles, of
which about 95,000 square miles are water  surface. Of the total area, roughly 60% is under the
jurisdiction of the United States, the remaining 40% is under Canada.
    The principal source of water  supply to the Great Lakes area is the precipitation over its
water and land area. The average  precipitation is about 31 inches per year, about two-thirds of
which is returned to the atmosphere by evaporation and transpiration.  The average net annual  sup-
                          12                    13
ply is therefore about 7 x 10   cubic feet, or 5 X 10    gallons.  The amount of ground water enter-
ing or leaving the Great Lakes area is  considered to be insignificant.  The only additional source
of water supply is from the Long Lake  and Ogoki diversions, and this water enters the system via
Lake Superior.
    Considering one lake alone,  a mass balance equation may be written as follows:
where  S = change in lake storage
       I = inflow from upper lake
      O = outflow to lower lake
                                               R = runoff from the drainage area
                                               p = precipitation on the lake surface
                                               E = evaporation from the lake
                                               D = diversion in or out of lake
                                            81

-------
    Figure E-l is a flow chart of the average inputs and outputs of the system of lakes, based on
data from the U. S. Lake Survey.  These average inputs and outputs, for the period of 1950 to 1960,
assume that the lakes have a constant level and indicate the relative magnitudes of the flows.
    If these flows occurred continuously over time, the levels of the Great Lakes would actually
stay constant and there would not be any problem with the  hydrologic system.  However, the very
fact that the supply varies introduces a variation in the lake levels which is undesirable to almost
all users. Excluding any water-quality considerations, three major water-use categories can be
identified which relate closely to the lake levels and volume of water stored.  These are;  (1)
riparian or shoreline  interests, (2)  navigation interests, and (3) power interests.
    The shore property owners have an interest in reducing the high lake levels both in magnitude
and duration, since they will cause erosion and damage buildings and lands.  Similarly, low levels
will expose land areas normally under water, which may look objectionable from an aesthetic point
of view, and which may make it difficult to launch small boats.  Most significant in economic terms
is probably the damage due  to erosion and wind action at high lake levels.
    The navigation interests require high minimum levels on the lakes to support greater draft for
the  ships and also high minimum flows in the connecting channels. According to the U. S. Army
Corps of Engineers studies, high levels  and high flow velocities  in the channels are of less
importance.
    The power interests are generally seeking higher lake levels to provide more head at their
turbines and an equalizing in the outflow to increase their  firm power capacity.
    Summarizing, then:  with the exception of power interests which seek high lake levels, all other
interests are best served if the lake-level variation and the outflow variations are reduced and
maintained close to their long-term average.
    At the present time two of the lakes are regulated:  Lake Superior and Lake Ontario.  Lake
Superior, the uppermost of the Great Lakes, has been regulated since 1921 when the natural river
bed of the St. Mary's River  was dammed and the outflow controlled by a number of gates.  At the
same time a lock system was installed to permit the movement of ships to and from Lake Superior.
At the present time, Lake Superior is regulated according to the "modified rule of 1949." This
rule curve was developed over the years by the Corps of Engineers and is a modification of some
earlier rule curves. It provides for lowering the maximum stages and raising the minimum stages
by regulating the outflow.
    Lake Ontario has been regulated since 1960 in accordance with 1956 orders of the International
Joint Commission (IJC).  Again, these orders essentially provided fixed rule curves with the aim
                                             82

-------
                    NET  RAINFALL
                RUNOFF FROM DRAINAGE AREA
FIGURE E-l.  INPUT AND OUTPUT MODEL. xlOOO cubic feet per second.
                             83

-------
of lowering high stages and raising low stages.  The rule curves were originally developed in 1958
by the Corps of Engineers; in a Corps report in 1965 it is stated that they had undergone three re-
visions  in three years.  This points up the need for continuous revision of these regulations to adapt
to changing needs and also raises the question whether these rigid rule curves should not be re-
placed by a more dynamic decision process.
    In 1965 the Corps  of Engineers published a report on the regulation of all the  Great Lakes in
which, for the first time, the rule curves vary depending on the  antecedent rainfall and  net water
supply to the basin.  These plans, which include regulation of Lakes Michigan-Huron, and Erie,
were  developed considering U. S. interests alone and are presently under study by Canadian
authorities.
    It appears that there is a definite need for mathematical modeling of the water quantity and
lake levels.  The regulation rules should be determined for the total system, and not just for regu-
lating one lake alone.  For example, Lake Superior could be operated as a  "reservoir"  to help im-
prove the levels of  Lake Michigan-Huron.  In the model one would have to weigh the damages due
to low lake levels on Superior versus the benefits to be derived  from increasing the lake levels of
Lake  Michigan-Huron.
    Similarly, although reducing high lake levels and increasing low levels is a step in the right
direction, a "foot" of low-level regulation should have a value different from a "foot" of high-level
regulation, and these should be properly weighed against each other.
    The objectives of this proposed research are to explore and develop new methods for optimal
lake-level control.  In this  respect several new techniques  generically summarized under the name
of "systems analysis" are  ideally suited.  Specifically, it is proposed to investigate  the applicability
and limitations of the use of linear, nonlinear and stochastic programming for  the determination of
optimal operating policies  and regulating structures of the Great Lakes system.  The research will
proceed along two major areas of models:  deterministic models and stochastic models.
Both types should add to our understanding of the lake-level control problems.

                                    E.2.  RESEARCH PLAN
    The research leading to an improved method for lake-level regulation will proceed in the fol-
lowing three phases:
    (1)   Collection and analysis of the hydrologic and economic  data.
    (2)   Formulation of deterministic and stochastic models.
    (3)   Verification of the proposed models by simulation of the system.
                                             84

-------
E.2.1.  COLLECTION OF DATA
    The first phase of the study will be a collection and critical review of all the data available at
present regarding the hydrologic and economic systems, as related to lake levels, of the Great
Lakes.  Some preliminary review of the hydrologic data obtained from the Lake Survey of the
Corps of Engineers is  in process. Data on the net basin supplies to all of the Great Lakes have
been obtained on IBM cards and are  currently being analyzed for their statistical properties.  Fur-
ther data will be gathered on observed lake levels,  rainfall, evaporation, and transpiration.  Many
of these data are  available from the  offices of the Corps of Engineers, the U. S. Geological Survey,
the Weather Bureau, and state agencies.   In order to establish benefit-and-loss functions for vary-
ing lake elevations, considerable data collection activities are necessary.  At the present time the
Lake Level Board is conducting a study on these aspects, and it is proposed not to duplicate this
study but to use its data.

E.2.2.  FORMULATION OF MATHEMATICAL MODELS
    The mathematical models and techniques which will be explored in this study can be divided
into two classes:  deterministic and  stochastic models.  In the first category, models will be es-
tablished beginning first with one lake and then expanding to the entire chain of lakes.  For example,
in the absence of  economic data,  criteria  for lake-level regulation could be;  (a) to  minimize the
square of the deviations from a proposed  rule curve; (b) to minimize  the sum of the absolute devia-
tions from a preselected level; or (c) to find  rule curves which minimize the maximum absolute
deviation. If economic data are available, then one would wish to find operating rules which would
minimize the total losses or maximize the net benefits.  These models will be studied under present
limitations by existing engineering works and then under conditions with these restraints removed.
    Once these deterministic models have been explored to the fullest extent, a refinement to
stochastic and chance-constrained models will take place.  The amount of water supply in any one
month to any one  lake is a function of the  previous month's rainfall, temperature, and lake level.
Operating policies will therefore be  established which will be optimal taking into consideration the
hydrologic events of  the past and the predicted supplies of the future.   Typical of these types of
models  are, for example, the models Loucks  has developed for the regulation of the Finger Lakes
in New York.  The Markow chain model, where the  inflow each month is dependent only on the pre-
vious month's  inflow, is a step in the right direction; however, the Great Lakes have more than one
significant lag period and the several lakes have to be accounted for at the same time.  It is there-
fore planned to extend this method and also to explore the use of some recent developments in the
area of chance-constrained programming.

E.2.3.  VERIFICATION OF THE PROPOSED  MODELS
    In order to establish the significance and the benefits to be derived from integrated lake-level
control rules,  it will be necessary to establish the difference between present-day operating  rules
                                             85

-------
and any new ones derived in the study.  To accomplish this, it will be necessary to construct a
mathematical simulation model of the Great Lakes which will permit a testing of these alternatives.
This latter part will be executed in close cooperation with the U. S. Army Corps of Engineers which
is now charged with the  responsibility of regulating the Great Lakes.  There is a possibility of using
their regulation model and extending it  where appropriate.

E.2.4.  HYDRO-METEOROLOGICAL MANAGEMENT
    Another aspect of water-quality  modeling is evaluation of management practices which can
change runoff from the drainage area, precipitation on the lakes and in the drainage area, lake evap-
oration, and diversions. These management possibilities as well as the  purely engineering regulatory
works need to be tested. For example, runoff from tributory drainage areas is not necessarily fixed,
but may be subject to increase, decrease, or change in timing, depending upon land management
practices.
    Weather modification is receiving  considerable attention, and it may become feasible to  in-
crease precipitation in the  relatively near future.  Such increases directly on the lake surfaces and
in the form of increased land runoff  can be  evaluated in the water-quantity model.
    Lake evaporation through the use of monomolecular film is now feasible for small ponds and
lakes.  An  evaluation of these and other techniques for suppressing evaporation should be tested for
their effects on lake levels and discharges as well as their economic consequences.
    It is considered likely that numerous proposals for diversions into and out of the Great Lakes
will be made in the interests of water supply and waste disposal.  The water-quantity model  will be
well suited to test the physical  and  economic consequences of such proposals.
    Finally, it should be pointed out that water-quality models are highly dependent upon a work-
able quantity model in matters of flushing rates and residence times of nutrients and pollutants.
                                             86

-------
                                        Appendix F
                             PRELIMINARY RESEARCH DESIGN FOR
                  THE WATER-QUALITY SUBSYSTEM OF THE GREAT LAKES REGION

                             A.   The Proposition for a Lake Model
                                        John C. Ayers
                                   The University of Michigan
    The lake model envisioned for the Great Lakes region divides each lake internally into a series
of subregions.  Each subregion is chosen to be relatively homogeneous in its properties.  Successive
summations of the inputs, outputs, and transfers of the subregions define water quality within each
subregion, provide a mean level  of water quality in the lake, and at the outlet give a level of water
quality that serves as one of the  inputs into the next lake.
    The objective of the model is to describe  water quality by subregions within all the lakes of
the Great  Lakes system as a function of population size and industrial activity and of the level of
control  imposed upon the sources of pollution  input.

Steps in developing the model:
    (1)  Pollution inputs will have to be defined by location, types, and amounts.  Location is de-
        fined as being the points of direct discharge  into the lake.  Type means identified by
        origin,  e.g.,  industrial,  municipal, or agricultural; and identified by inorganic nutrient
        contents, contents of  conservative (nonbiological)  inorganic  compounds, and contents of
        organic compounds.   Amount is defined as weight  per unit volume.  All the above should
        be determined as a function of population size and as a function of industrial activity in
        order that they may be meaningfully related  to historical trends and to projections from
        the economic growth  model.
    (2)  The best possible determination of present and past water-use (volume) demands and
        water-quality demands by subregions will be needed as functions of population size and
        industrial activity in  order, again, that they may be related  to projections from the
        economic growth model.
        [Note:   The assembly of the data listed in steps (1) and (2) should be coordinated with
                FWPCA, state, and all other data-gathering activities.]
    (3)  Information on the physical nature of lake currents in each  lake and each subregion of each
        lake must be collected and organized for use in the model.
                                            87

-------
    (4)  The present information about physical, chemical, and biological transformations that
        accompany the currents in each subregion of each lake must be collected and organized
        for use in the  model.
    (5)  An abstract model, which divides the lake into subregions that are reasonably homogeneous,
        must be formulated for each lake.
    (6)  Transfer functions that apply to the exchange of water and material between subregions
        must be determined for each subregion of each lake.
    (7)  The biological and/or chemical transformations which describe the water quality in param-
        eters which are critical at water-intake points  must be established as a function of inputs
        and withdrawals within the subregion and exchanges between subregions.
    (8)  The mean water quality at the lake outlet, which is one of the inputs into the next lake or
        the St. Lawrence River, can be determined by combining the model components developed
        in steps (1) through (7).
    Figure  F-l  presents, as an example, the initial subdivided model of Lake  Michigan.  This type
of subdivision seems to be applicable in spring, summer, and fall; in winter these divisions tend to
disappear.
    Similar subdivided models of the other Great Lakes can be made from existing knowledge.

                           B.  The Proposition for a  Sublake Model
                                      James  E. Kerrigan
                                   University of Wisconsin
    The modeling of the water-quality aspects of the Great Lakes will necessarily take more than
one form.  Because of  the complex nature of the numerous elements  involved within the system, it
will be useful to approach the water-quality system from several points of view. The previously
lake-model  approach described above is designed to quantify the interrelationships between local
areas lake by lake.  The sublake approach seeks to describe the nature and alternative courses of
action that  may take place within a localized region or local sector, taking into account the effect
of neighboring sectors. By narrowing the scope of the physical region under investigation, it will
permit a closer  investigation of the parameters identifying water quality and the environmental
factors which influence their occurrence.  Also, it will permit manageable studies  to be undertaken
that will indicate possible methods for treating or controlling the levels of water quality within the
local  sector for  specified uses.
    The role of  the sublake model will be to complement the other modeling undertaken to describe
and identify the water-quality system, as well as the overall modeling that will be pursued under
the framework of the study, such as  the regional economic-demographic growth model.  Within
this context the sublake model will seek to satisfy a series of specific objectives which are uniquely
suited to its design.
                                             88

-------
                       \
FIGURE F-l. INITIAL SUBDIVIDED MODEL OF LAKE
MICHIGAN.  Dotted lines indicate conceptual boundaries.
    Dashed lines indicate hypothetical subregions.
                        89

-------
    The objectives of the lake model include the following points:
    (1)  To test and evaluate alternative methods and levels of control, including budget constraints
        and enforcement costs. Among the alternatives are:
        (a)  Effluent fees and subsidies
        (b)  Water-quality standards (local, state, regional, national)
        (c)  Information dissemination
    (2)  To provide realistic inputs representative of the local sectors into the lake model.
    (3)  To simulate or stimulate a control system capable of achieving and maintaining the re-
        sults of a perfectly competitive market.
    The first task of the sublake model will be to identify the system.  A simplified free-body dia-
gram, shown in Figure F-2, could be used as a basis for the system.  The free-body diagram may
be converted into the block diagram (Figure F-3) to show the general network of the sublake model.
Figure F-4 illustrates the modular arrangement that could be used to build the sublake model to
represent the interaction between uses within a sector, adjoining local sectors, and inshore and
offshore water zones.
    In these figures, the following symbols are used:
    Al, A2, A3, A4 = the local sectors
                 B = benefits
                 C = costs
                 D = the sediment that  forms at the bottom of the  lake
                  I = the exchange of the substance or energy between the points within local sec-
                     tors and between the  surrounding areas of the lake
                 L = the interaction between Lj and L-
                 L = the interaction between the inshore water of  adjoining local sectors
                 I, = the relationships between the water quality of the inshore and offshore water
                  O
                     zones
      i (subscript) = specific use parameter
      ) (subscript) = water-quality parameter
      k (subscript) = local sector parameter
                 L = the degree of water quality for a given parameter in  the lake
                L. = the characteristics of the lake water at the point where the water is collected
                     for a particular use
                L  = the quality of the water following use
                  £i
                     (Note:  for some uses, L- and !_,„ would be the same)
                                             90

-------
           J2  / L4

         -ti2^	LakeA.^:'
          L,  \
Lake B
                            A4
FIGURE F-2. FREE-BODY DIAGRAM OF SUBLAKE MODEL
  FIGURE F-3.  BLOCK DIAGRAM OF SUBLAKE MODEL
                     91

-------
LOCAL SECTOR #1
LOCAL SECTOR #2
            FIGURE F-4. MODULAR ARRANGEMENT FOR SUBLAKE MODEL
                                     92

-------
                L_ = the mean concentration of a given substance or physical state for the local
                 o
                    sector

                L. = the interaction between the inshore water zones of the local sectors
                 4

                 O = the diversion of water from the lake to a neighboring watershed

                 R = the characteristics of water discharging into the lake sector and not previously

                    collected from the lake

                T  = the treatment,  management, or enforcement controls on influent  water

                T_ = the controls on the effluent side of the specific use
                 i
                 U = use, and associated benefits and costs attributed to such use

    Explanations of the complexities of some of the factors identified in these figures  are given

below.

    D. Sediments tend to concentrate, stabilize, and exchange and release nutrients and other sub-

stances under various  environmental conditions.  An understanding of the interactions  between the

sediments and the overlying lake water will be required for the model.

    i, j.  The model is based on the  identification of degree of water quality based on a specific

use.  Some of the major uses and quality parameters are listed below.

                                          USES (i)
              Commercial fishing                Irrigation
              Cooling water                      Recreation
              Domestic waste  carrier               Swimming
              Domestic water  supplies              Wading
              Industrial waste carrier               Fishing
              Industrial water supplies              Boating
                                                   Sightseeing

                            WATER-QUALITY PARAMETERS (j)

              Bacteriological indicator tests       Phenols
                (coliform, enterococci, etc.)       Phosphorous compounds
              Carbon dioxide                     Plankton (number, diversity)
              Color                              Toxic substances
              Dissolved oxygen                   Suspended solids
              Dissolved solids                    Tastes
              Hardness                          Temperature
              Nitrogen compounds                Toxic substances
              Odor                               Turbidity
              Organic material

    L.  It is well recognized that the water quality varies greatly in large lakes such as the Great

Lakes.  The variation is caused in part by the natural environment and physical configuration of

the lakes, the lake currents, and localized shoreland uses.  During the spring, summer, and fall

seasons, a natural barrier develops between the inshore and offshore waters.  The separation

breaks down in the winter season, thus permitting a greater interchange of shore  and deeper  waters.


                                            93

-------
As shown in Figure F-2, the lake is divided into inshore and offshore water zones and boundary
lines are used to separate "local sectors."  Local sectors are areas in which strong water-quality
interrelationships exist and have weak interaction with neighboring sectors.
    R.  Examples of water-discharge characteristics would be rural and urban runoff, discharges
of used cooling water obtained from ground water, and ground water entering the lake  from the
adjoining land.
    T,, T0.  The water-flow pattern between the lake and the water-use zones is generally cyclic.
      1   £i
Two situations  can exist within a local sector where the water is not cycled and their influence on
the  system can be significant.  (1) Lake water  can be diverted from the lake for some use outside
the  original watershed, as in the case of the Chicago diversion for stream-flow augmentation.  The
lake water is used to  improve the quality of the river system that drains away from the lake with-
out  affecting the quality of the lake water.  An alternative in improving the quality of the river water
may be to discharge the treated municipal waste into the lake.  (2)  The other case involves the in-
flux of great quantities of pollutants and nutrients from the inflow of urban and agriculture drainage.
    U.  Figure F-2 indicates the spatial relationships between the local lake sectors and associated
water uses. The uses may be municipal or industrial water supply, thermoelectric power  genera-
tion, cooling water, recreation,  commercial fishing, or irrigation, to mention only a few.  For some
uses it will be possible to employ management  or treatment controls before and after  the water is
used, whereas other uses will permit only one, and possibly no control measure on the influent or
effluent stream.  For in-lake recreation uses,  such as swimming, water quality control measures
may be unavailable.  However, the degree of water quality required for this use is high.
     In Figure F-5, the I  box may be defined as a time-dependent relationship which  describes
the material balance for a specific constituent  by
                                          dM.
                                          _
                                          dt
where M is the mass of the substance stored in the L,  element, t is the time, and F represents the
mass rate  of flow of the substance.  The rate could be determined by
where v  is the mean current in the kth local sector, A,  is the effective cross -sectional area in
the kth sector, V,  is the volume in the local sector, and  M.  is the mass of the substance in a
                K.                                      1
specific use.  The concentration of the substance, C, ., would be
                                         C, . = V. M.
                                          ki    k  i
     In developing a useful scheme for selecting the kinds of control that may be used to manage
the quality of the water within a local sector, it will be necessary to identify the water uses in the

                                             94

-------
FIGURE F-5.  TIME-DEPENDENT RELATIONSHIPS FOR SUBLAKE
                        MODEL
                          95

-------
sector along with the limiting levels of water quality associated with each. After the limits have
been established for each parameter, studies must be conducted to determine the types of water
and waste-treatment methods, management-control measures,  enforcement schemes, etc., that
may be employed to control the level of water quality within the limited range.  As a rough-cut
model, it would be desirable to define the controls suitable to the municipal use of lake water.
This would include the urban runoff and diversion for stream-flow augmentation.  A  series of in-
vestigations into the cost and benefits associated with controlling the level of water quality in
selected local sectors  should be undertaken.  The objectives  of such studies would be to evaluate
the direct costs and benefits which are normally identified, as well as to make  professional esti-
mates of those social values which usually escape quantification.  With the comprehensive assess-
ment of the resources  available, the use demands, ideal and practical institutional and legal con-
straints, and the interaction of the components of the system, alternative control  schemes will be
suggested to control and manage the water-quality scheme to meet selected social values.
                                             96

-------
                                         Appendix G
                  PRELIMINARY RESEARCH DESIGN FOR THE GAMING SIMULATION
                                      Richard D. Duke
                                 The University of Michigan
     The purpose of the envisioned gaming-simulation model is to design a heuristic micro-
environment, replicating in its salient aspects key policy problems of water-basin planning and
control.  Objectives of this approach are listed below.
     (1)  It would clarify the technical problems of researchers for politically oriented decision-
         makers.
     (2)  On the other hand, it would clarify the political problems of decision-makers for tech-
         nically minded research workers.  (Too often technical advice that is given to politicians
         and administrators falls on deaf ears because they are unable to see the political  rele-
         vance of the researchers' recommendations:  (a) because politicians are not conversant
         with the technical rationale,  and (b) because researchers are  not attuned to the needs of
         politicians and give  data and indicators that miss the real needs of decision-makers.)
     (3)  It would allow researchers to quickly and efficiently evaluate alternative models of
         politics, economics, and water resources but with a knowledge of policy.
     (4)  It would allow researchers to project alternatives for the future of the basin by using
         several different policy assumptions.
     (5)  It would allow key decision-makers to experiment with alternative innovations and to see
         their long-run impact on a basin, but in a risk-free setting.
     (6)  It would allow the linking together of families of models in one package so that researchers
         and decision-makers alike could see the entire universe in the reduced time span.
     (7)  It would train future professionals, both researchers and administrators, in the policy
         problems of water resources.
     (8)  It would allow key federal  personnel to plan and evaluate alternative decisions altering
         their roles and relations within the game.  Thus they could see the impact of various
         decisions on a future water-resource system.
    One  game that might be devised would be a free-wheeling structure that  mimics international
relations.  This could point up problems and mechanisms of cooperation, conflict, and competition
in a water-resource  basin context.  Issues would be developed for decision-makers to respond  to
("polluters versus users" "up and down" stream sources and uses, multiple-use conflicts with
                                             97

-------
varying quality and quantity criteria, etc.) Within such a game, each participating unit (city or
other government) would have several roles, played by individuals, that might be considered to be
typical for water-resource problems.  These might include such roles as politicians representing
constituencies at different levels of government; businessmen who are involved in water problems
like use and consumption; technical specialists and administrators; health experts; economists;
water engineers; government administrators; and, finally, representatives of the mass media.  In
addition, there would be computer simulation of population, influential natural events, firms, house-
holds, agencies, and water quantity and quality.
    Critical problems of research supporting such a game are to determine what the power struc-
ture is, what characterizes its key positions and what the linkages between them is; and also what
key organizations exist (with what power) and what their linkages are.  This  is essentially unre-
searched for large  basins.  Also, basic research is required on public opinion formation about
water resources issues, the role of the mass  media, and the relation between economic and political
models  of water resources.
    The research design would include the identification of major social and political institutions
(e.g., municipalities, townships, counties, states, private users)  in the Great Lakes  region, and
their  representation in modular form.  These modules could be combined appropriately to repre-
sent the existing decision-making organizations in a particular water- and land-use area.  A hy-
pothetical water- and land-use area  is shown  in sector A of Figure G-l. After institutional con-
straints have been  identified, possible alternatives to existing conditions would be formulated.  Ex-
periments could then be run to evaluate these various alternative institutional forms.
                                             98

-------
                                                            Hypothetical
                                                            Areas of
                                                       ~~   Institutional
                                                       """   Jurisdiction
                            111.  I  Ind.
FIGURE G-l.  SIMPLIFIED MODEL OF A SUBREGION.  Sector A represents a
hypothetical water- and land-use area that could serve as a sample area for model-
ing.  The interaction between the water-quality modeling and the gaming-simula-
tion model can be visualized from the figure.  Pairs of facing arrows indicate inter-
     relationship. Directional arrows show hypothetical current patterns.
                                    99

-------
                                         Appendix H
                      AN INVENTORY OF NEEDED RESEARCH ACTIVITIES
                                IN THE GREAT LAKES REGION
                                         Compiled by
                                     Spenser W. Havliek
                                  The University of Michigan
    As an indication of personal interest and, in an effort to obtain an indication of the personal
interest of others, an informal survey was made of faculty at Water Resources Centers and participat-
ing universities.  The objective of the survey was to provide an opportunity for investigators at
earlier CIC gatherings to indicate their personal research preferences as shown below.
    Basically two categories of research  activities were identifed by participants  in the working
conferences.   (Specific suggestions for interuniversity research projects which came from inves-
tigators were  kept on file by the CIC staff.)  Within these categories,  "institutional and socioeconomic
research activities" and  "physical-biological research activities," items were arranged randomly
in a questionnaire which  was  sent to investigators.
    Part of the questionnaire asked the respondent to indicate, which projects, in his judgment,
were in need of "immediate research attention" or "research attention very soon." The respondent
was also given options to mark projects which needed research effort "ultimately" or "no attention
at all."  A space following each potential research activity permitted persons  to register their
names as an indication of personal interest.
    Of the working conference investigators, 70% completed the questionnaire. Persons who in-
dicated personal interest included many conference participants and their colleagues in CIC univer-
sities, and researchers in cooperating institutions including Cornell, Indiana,  and Ohio Water Re-
sources Centers.
    The results of this modest inventory do not lend themselves readily to statistical analysis be-
cause of the variables which may have influenced respondents and because of their individual biases.
Nevertheless, the replies concerning the areas where research attention was desirable "immediately"
or "very soon" seemed of enough interest to merit publication.

                                     QUESTIONNAIRE
    In the randomly arranged list below, the number in the left column refers to the number of
"early attention votes" which were cast for each research activity.  Evidence  was  obtained that
multi-university interest exists in each of the projects listed.  Considerable overlap is found be-
tween several projects in the two categories.  In both categories it is of interest to note that the
                                            100

-------
research activities which received the highest number of "votes" are essentially the same research

projects which have been stressed in this report.

    The categories of research activities are arranged for cursory observation, with projects

which received the highest number of "votes"  listed first; but undue weight should not be placed on

the number of indications of interest ("votes") attached to each potential project.  This tabulation

includes responses received before December 12, 1967.

Indications of Interest         Category of Institutional and Socioeconomic Research Activities

    17	Collection of economic data on water and waste treatment costs as re-
                           lated to eutrophication.

    16	An analysis of how water management decisions are made in the (Lake
                           Michigan) region  and what techniques can be applied to assist the
                           decision-making machinery in producing and selecting from a wider
                           range of alternative programs of resource management.

    16	Construction of a regional economic growth model for the Lake  Michigan
                           subregion.
    15	Conflict situations among competing uses of Lake Michigan are  in special
                           need of study.  Tradeoffs, if any, need to be understood between  (1)
                           recreation and waste disposal (pollution); (2) commercial fishing, and
                           pollution; (3) one form of recreation versus an incompatible form of
                           recreation (power boats vs. duck hunters or canoeists); and (4) power
                           and maximum shoreline use and development.
     15	How does the price of water affect its recreational, domestic, and  in-
                           dustrial uses ?

     13	How does the political process  affect decisions relevant to water-
                           resource usages  in Lake Michigan or the Great Lakes system?

     13	A feasibility study designed to determine an approach to a Lake Michigan
                           model which can  eventually be plugged into a Great Lakes regional
                           model.

     13	An  analysis of the flow of information for planning  and policy activities
                           between and among the private and public sectors of a subregion.

     13	How do individuals, groups, organizations, and other public sectors ex-
                           press themselves in the process of arriving at a consensus on what should
                           be done with the available water resources?
     13	  How are values expressed through our present institutions, including
                           the International Joint Commission and the Great Lakes Basin
                           Commission?

     12	   What does the law say, and, what is the actual application in water
                           resources?

    12	What are the social implications (as well as economic) of water-
                           resource management and development in this subregion?

    11	Data storage and  retrieval system from the water-research activities
                           in the Great Lakes basin.

    11	  What are the organizations in the Great Lakes area ?
                                            101

-------
Indications of Interest       Category of Institutional and Socioeconomtc Research Activities (Cont.)

    11	How does a concern for the environment form among individuals or
                           groups, how is it measured, how is it expressed, and how does the
                           governmental sector incorporate this concern into political and social
                           action?

    11	Identification of data-collection sources, the types of data collected,
                           and the form of available data.

    11,	What is the process by which public opinion on water-resource questions
                           is formed ?

    11	Monitoring of economic effects of changes in water quality at a few
                           critical points as a result of putting water-quality standards  into
                           effect.

    11	How are social values for different goods and services (from the water)
                           related to different uses in different parts of the basin?

    11	What is the process of problem definition within watersheds, major lake
                           basins, and the entire Great Lakes  region?  What are the priorities?

    11	An  institutional arrangement such as  the Great Lakes Basin Commission
                           needs to be assessed for its strengths and weaknesses in terms of
                           reflecting social values, resource-planning program strategies, etc.
    11	In light of the externalities in the Lake Michigan situation, how can the
                           differences and impacts between social values and economic  values
                           be determined ?

    10	An  identification of water-management influences  and influentials as a
                           part of the regional or subregional  political power structure.

    10	How can the generation and presentation of information for making public
                           decisions be  improved?

     9	A determination of operational efficiency and effectiveness at various
                           levels, in water-related organizations.

     9	What is the capability to implement decisions after consensus on water-
                           resource developments or -management schemes ?

     8	What heuristic  research techniques may be applied to the decision-
                           making performance of operating agencies ?
     8	Attitude survey of water-using or water-polluting  industries.
     8	How frequently are water-management and -development decisions
                           made by persons or organizations not directly involved as beneficiar-
                           ies and/or contributors to the project cost?

     7	What media and "vehicles" can be applied most effectively (and how) in
                           water-related information dissemination?

     7	What about new  institutional structures ?

     7	An  evaluation of the prospect that traditional lines of authority  and  in-
                           fluence may be rerouted from Congressional delegations to regional
                           agencies of national and international stature, particularly by state
                           and metropolitan governments.

     7	What social change can be brought about through planned intervention
                           in conjunction with water-resource projects?

     6	Role-playing type of simulation.


                                            102

-------
Indications of Interest      Category of Institutional and Socloeconomtc Research Activities (Concl.)

     6	Production of documentary film on Great Lakes basin problems.

     4	While the subsystem is under study, the  effects of the study or the
                           "client" need to be identified and separated from any changes in be-
                           havior, etc., that would have proceeded in the absence of an investiga-
                           tion.  Thus the need  for monitoring and evaluation in this  area.
     4	Time lag from occurrence of actual problem to perception of the
                           problem.

                                  Category of Physical-Biological Research Activities

    13	  Water-quality model of Lake Michigan or others and  its contiguous
                           drainage basin in terms of nutrient budget, rate, and effects of fertili-
                           zation.

    13	Relationships need to be established between various  pollution levels
                           and their effects in biological as well as  economic terms.

    12	An investigation of transfer functions between input controls and input
                           pollution; they also need to be developed  between elements of pollution
                           and the general  pollution level and controls presently in use in the
                           Lake Michigan situation.

    12	The effects and mechanics of coastal and offshore interchanges and
                           mixing.

    11	A determination of the status and future  of  eutrophication in Lake
                           Michigan.

    11	What are significant parameters and methods of measuring water
                           quality, sedimentation, and oxygen demand?

    II	Quantitative model of the Great Lakes  system with special attention
                           directed to lake levels, associated flow,  and  losses including
                           augmentation and diversions resulting from structural work.
    11	The definition of inputs, outputs, and "state-of-the-art" controls on a
                           cause-and-effect pollution model.

    11	What simplified models can be applied to the entire Great Lakes sys-
                           tem  in areas of  water quality?
    11	Improved indices and monitoring techniques for eutrophication.
    10	Future water-supply needs and expected levels of pollution.
    10	A study on pelagic and  benthic conditions in Lake Michigan or other
                           Great Lakes.

    10	How pollutants modify water quality.
    10	Sport and commercial fishery resources.

     9	A sediment-water interface study.

     9	Is improvement in waste-treatment processing gaining on the increased
                           pollution added by increased population?
                         Retention times in the Great Lakes.
                         How can the output from simulation or any physical model be used
                           most  effectively with the eventuality that value judgments will need
                           to be  made which force numerical rating techniques.
                                            103

-------
Indications of Interest         Category of Physical-Biological Research Activities (Concl.)

     7	Precipitation-evaporation on lake surfaces of the Great Lakes.

     7	Physical, biological (legal and economic),  effects of further diversion
                           or diversions of  Lake Michigan-Huron water.

     7	Spatial requirements for recreation developments.

     7	The effect of evapotranspiration rates,  ground-water gains and losses,
                           and the necessary degree of accuracy required for these and other
                           variables in a quantitative model.

     7	Regulation of  lakes with and without existing structures.

     7	Physical observations of lake levels need to  be related to resultant
                           values (and to beneficiaries).

     6	What engineering structures are  needed to better regulate the system,
                           what are the benefits, and to whom do they accrue?

     5	Systemswide model of quality using a conservative element (e.g.,
                           chloride).

     4	Irrigation opportunities in the  region and the consumptive implications.
                                             104

-------