The Use of System Dynamics Simulation
in Integrated Water Resources
Management
Ines Winz, Gary Brierley
School of Geography, Geology and Environmental Science
University of Auckland
Private Bag 92019, Auckland, New Zealand
Tel. +64 (0)9 373 7599 ext. 88956, Fax. +64 (0)9 373 7434
i.winz@auckland.ac.nz, g.brierley@auckland.ac.nz
Abstract
In this paper we discuss the use of system dynamics as a methodology with
which to address dynamically complex problems in water resources manage-
ment. Problems in regional planning and river basin management, urban
water management, flooding and irrigation exhibit important short-term and
long-term effects, and are often contentious issues with high potential for con-
flict. We argue that system dynamics combined with stakeholder involvement
provides an appropriate methodology to address these issues effectively. We
trace the theoretical and practical evolution of system dynamics in these areas
over the past 40 years. From this review of the literature we identify and dis-
cuss a number of best practices and common pitfalls in applications of system
dynamics.
1 Introduction
Widespread recognition of the impact of human activities upon natural sys-
tems is transforming the way we view and manage the earth’s resources. The
quest for sustainable management of the earth’s resources in the light of multi-
ple bottom line considerations has significantly broadened the focus and goals
of management endeavours and is one of the most demanding challenges fac-
ing society today (Chaves and Alipaz, 2007; European Union; World Water
Assessment Program, 2006).
Readily available water resources have already been extensively exploited
across much of the planet, and development pressures, population growth and
climate change place additional stresses upon this vital resource (Baron et al.,
2002; Gleick, 2000; Jackson et al., 2001). Increasingly, we are forced to con-
sider approaches to utilization of remaining marginal water sources, though
such measures can be delayed and potentially avoided by placing appropriate
emphasis upon the demand rather than the supply side of the water balance
equation. In many parts of the world, significant and highly contested discus-
sions are underway regarding future management of water resources, placing
increasing emphasis upon water trading and pricing (Howe et al., 1986; Islam
et al., 2007; Luo et al., 2007).
Traditional ‘command and control’ approaches to water resources manage-
ment emphasized concerns for the provision of adequate water resources to
meet human needs, without giving due regard to the maintenance of ecosys
tem services (Holling and Meffe, 1996). We already live with the legacy of pas
actions, especially those associated with high-cost engineering infrastructure.
Increasing efforts are being made to redress the environmental impacts of
these measures, in programmes that target the maintenance or replacement
of water supply infrastructure, dam/weir removal, ‘naturalisation’ of artifi-
cial channels, etc. One thing is abundantly clear, technical advancements, in
themselves, will not engender sustainable outcomes (Higgs, 2003). Sustainable
practice requires a shift in outlook towards application of proactive measures
that ‘work with nature’. In striving to meet societal needs while maximizing
the life-supporting capacity of the planet, an ecosystem approach to natural
resources management emphasizes concerns for the resilience of living, vibrant
river systems, recognizing explicitly the diversity, complexity, variability and
uncertainty of natural systems (Everard and Powell, 2002).
As legislative and policy frameworks adjust to these altered and evolv-
ing circumstances, personnel with responsibilities to implement these transi-
tions in practice face significant challenges in the identification, development
and application of appropriate management methods, e.g. Brierley and Cul-
lum (in Review); Rogers (2006). The broader range of faces and opinions
at the decision-making table highlight concerns for differing perspectives and
approaches to sustainability, balancing concerns for development and envi-
ronmental/cultural protection. Critically, these concerns are expressed us-
ing widely differing forms of information and insight, whether qualitative or
quantitative, scientific or spiritual. New tools are required to integrate this
breadth of perspectives in endeavours to scope the future. These efforts must
target the key issues of concern, the critical stressors that affect the viability
of sustained resource supply and environmental health, and identify the vi-
tal priorities that can be meaningfully addressed in a strategic and proactive
manner.
Foresighting exercises provide an informed basis for proactive management
strategies, enhancing our prospects to maximise the adaptive capacity of the
system as a whole. Dynamic simulation techniques provide a flexible tool with
which to perform these analyses. Effective approaches to the generation and
use of these modelling tools enhance our capacity to extrapolate and interpo-
late in a meaningful manner, framing system-specific applications in a broader
context. Such exercises present an opportunity to test our understanding,
exploring its implications and contradictions by raising and interrogating hy-
potheses.
In this critical review of the literature, we appraise the suitability of sys-
tems dynamics methodology as a tool for integrative resource management
by tracing theoretical developments and applications for the past 50 years.
We discuss success factors and potential pitfalls of reviewed work, and high-
light prospects for future applications of system dynamics methodology to
approach the sustainability agenda in a genuine and substantive manner. We
then consider factors that hinder the uptake of systems dynamics methodol-
ogy, prompting resistance to more widespread applications.
2 Dynamic simulation in water resources
management
2.1 The uses of dynamic simulation
Due to the complex nature of the problems addressed in water management,
the use of dynamic simulation models has a long tradition (Rogers and Fiering,
1986). Models are representations of a complex reality — a theory of how the
world operates at some level of aggregation. Models are used to test theories,
to explore their implications and contradictions. We constantly form mental
models as we perceive and make sense of the world around us (Doyle and
Ford, 1998). Mathematical models are a more deliberate act of representing
the problem we are concerned with in a scientific form. Their usefulness lies
in the fact that they allow us to test real world behaviour in an artificial
setting, thus being easy and inexpensive to perform in repetition. With ever
increasing computer power we are able to deal with increasingly large and
complex data sets.
Dynamic simulation allows us to observe the behaviour of a modelled sys-
tem and its response to interventions over time. Dynamic simulation models
consist of equations describing dynamic change. If system state conditions are
known at one point in time, the system state at the next point in time can be
computed. Repeating this process one can move through time step-by-step
over any desired interval. Simulation aids our capacity to make predictions of
future states. As long as the model describes reality with a certain accuracy,
the modelling process and its outcomes can be used to improve our under-
standing of the problem as a necessary step towards affecting sustainable and
effective change.
2.2 System dynamics methodology
Numerous analytical procedures have been developed to perform dynamic sim-
ulation. It is beyond the scope of this paper to review these comprehensively.
Rather we briefly contrast statistical forecasting with system dynamics, and
then concentrate on the latter for the remainder of this paper.
In statistical forecasting models equations are developed ex post, i.e. fol-
lowing observation, such that the model output matches available historical
data as closely as possible. This is usually done through regres
The equations are subsequently used to calculate future model states, i.e. the
simulation of future states is based on historical data. As there is no guaran-
tee that these statistical correlations truly fore
measures of error are introduced to quantify uncertainty.
In contrast, system dynamics models are causal mathematical models
(Barlas, 1996). In system dynamics methodology (SDM) the underlying
premise is that the structure of a system gives rise to its observable and thus
predictable behaviour (Forrester, 1968, 1987). The first step in any system
dynamics modelling project is to determine the sy ing
of positive and negative relationships between variables, feedback loops,
tem archetypes, and delays (Sterman, 2000; Wolstenholme, 2004). T!
followed by ex ante projection where future system states are replicated from
this model.
The difference between ex post forecasting and ex ante projection implies
that uncertainties with regards to future changes in system structure can be
more easily addressed as there is better understanding of system structure
in the first place (Sterman, 1994). This understanding of system structure
requires a focus on the system as a whole and we argue that holis
understanding is a necessary condition for effective learning and management
of complex systems as well as consensus building. These are important goals
in their own right. Additionally, systems modelling and simulation supports
sis and evaluation (Morecroft, 1992).
s of qualitative/conceptual and quantitative/numerical mod-
elling methods (Dolado, 1992). Qualitative modelling, e.g. using causal loop
diagrams (Figure 1) or hexagons (Hodgson, 1992), improves our conceptual
system understanding. Quantitative modelling, e.g. using stock-and-flow mod-
els (Figure 2), allows us to investigate and visualise the effects of different in-
tervention strategies through simulation. Quantitative modelling also requires
us to make explicit statements about assumptions underlying the model, iden-
tify uncertainties with regards to system structure, and identify gaps in data
availability. This promotes model transparency.
An SDM project consists of the following phases: problem definition,
system conceptualisation, model formulation, model evaluation/testing, pol-
icy analysis and implementation (Richardson and Pugh III, 1981; Roberts
et al., 1983; Sterman, 2000). These phases are pursued in an iterative fashion
(Homer, 1996). Commonly listed purposes for the development of SD mod-
‘ion analysis.
t future system behaviour,
tem structure con:
Water Users and
Polluters
z
amintetrave or Shortfall in 9
Neoded Construction of Copseiny Pollution Rato
+
Treatment Facilities
Perceived
Excess Capacity
\ Lontying
me aoe
Shortfall in
AS as
Fundiog | Palo
Enforcement and \
NN ton
ae Stream
Discharge Rate
wi Se Awareness.
Figure 1: Causal Loop Diagram describing water quality dynamics. Arcs
describe the directions of influence. A positive arcs reads as “an increase in
variable A leads to an increase in variable B”. A negative arc reads as “an
increase in variable A leads to a decrease in variable B”. Two feedback loops
are highlighted: B1 (thick dark arrows) explains how an increase in water pol-
lution over time leads to an increase in water treatment. Due to system delays
(double line), this balancing loop is likely to result in oscillating pollution l
over time. B2 (thick light arrows) shows how lobbyists (polluters) ut
existing waste assimilation capacity to affect a change in public poli
advantage. The interaction between B1 and B2 gives rise to comple:
behaviour.
els are improved system understanding, the development of a tool to analyse
and evaluate strategies and policies, and the testing of theories (Barlas and
Carpenter, 1990; Richardson and Pugh III, 1981; Sterman, 2000).
SDM explicitly asks for user input during the modelling process (Rouwette
and Vennix, 2006; Vennix, 1996) and is thus well suited for stakeholder partici-
pation. Modelling and simulation are aimed at providing valuable insights into
the problem structure instead of giving precise answers. They are thus suited
‘s that have important short- and
long-term effects. Further advantages of system dynamics methodology have
to investigate dynamically complex proce:
here been categorised under three broad headings (flexibility, ease of uptake
and adaptability, ongoing testing and learning) and are summarised in Ta-
ble 1.
‘Water in Wastewater
Treatment Plants
ry
Indoor
Water Use
Water in Distribution
System
Outdoor
Water Use
&
Treated
Wastewater Flow
Inflow from Other
Water Bodies
Waterin =>
River Basin | —____7___guc>)
Outflow Into Other
Water Bodies
Water Withdrawal Residential
we reewoc— Water Demand
Demand
Nonresidentiat
Water Demand
O——— | Population | —$e
Immigration
Births
Figure 2: Stock-and-flow diagram describing an urban water system. Boxed
variables represent stocks, double arrows represent flows that increase or de-
crease stock levels. Ausiliary variables (italics
can influence any other compo-
nent. The simulation showed, among other things, how outdoor water use has
a more adverse effect on water resources than indoor water demand. Adapted
from Stave (2003).
Category
Explanation
Flexibility
can be used for
a wide range
of applications
and supports
with
bottom
working
multiple
line dimensions
Supports the use of qualitative and quantitative vari-
ables in models: relatio
ships between variables can
be defined on an ordinary scale, e.g. low, medium,
high, as often used in modelling social system compo-
nents.
Cross
oped
Modular object-oriented nature of models: models of-
ten consist of different sub-models (or modules) in-
creasing interchangeability and reusability.
Supports a variety of project goals: the focus of any
project can be on the model development process itself
to support consensus building and team learning, the
final model and its use in simulating
under different scenarios, or both.
alar: a nested scale of models can be devel-
system behaviour
Established. The dynamic nature of the model and its transparency
methodology, allows users to quickly become familiar with modelling
ease of uptake and simulation as they are encouraged to alter the
and adaptability | model structure, parameters and data on their own,
and explore model capabilities and outcomes.
Computer software (e.g. Vensim®, Stella’, Power-
sim, Simile) is widely available and significantly re-
duces the cost of programming and running the model.
Compilation and simulation is fast. There is a wide
variety of model outputs including tables, graphs and
diagrams, wide range of sensitivity analysis capabili-
ties, and in-built error checking capabilities (Eberlein,
1989).
Parameters do not necessarily need to be fixed before
simulation. They can be either manually or dynami-
cally adjusted.
Foresighting, Simulation allows for the continuous testing of as-
ongoing testing sumptions and sensitivity analysis of parameters, with
and learning, few restrictions on problem presentation so long as
stakeholder par- variables can be identified and relationships defined
ticipation (Morecroft, 1988). No simplification is required to
make the model mathematically tractable and no ob-
jective function needs to be specified.
Methods are available to support consensus building
and team learning throughout the different stages of
the model development process (Vennix, 1996).
Table 1: Overview of key strengths of system dynamics methodology.
From our point of view, the key factor influencing the acceptance and
success of models is their practical usefulness. A model is useful when it
serves the purpose for which it was developed: it addresses the right problem
at the right scale and scope, and it represents system response correctly. While
the former refers to a model’s breadth and depth, the latter addresses model
validity. Models are an abstract representation of our limited understanding
of reality and reality in an open system can never be fully defined. Hence, the
concept of validity is flawed and models are never valid (Oreskes et al., 1994;
Sterman, 2002). The challenge becomes to find more appropriate measures
of model quality. Model usefulness and quality are subjective concepts which
do not lend themselves easily to a definition of objective measures. Moreover,
the greater the level of uncertainty and complexity of the problem, the more
superficial objective quality measures become. As a result, model validation
becomes a social process where model structure and outcome is negotiated
until judged valid and useful by all involved parties (Barlas and Carpenter,
7
1990). This concept of model usefulness requires transparency of the model
development process and the model itself. In contrast to system dynamics
models, standard black box models do not provide this level of transparency,
but often require expert knowledge in order to understand and use them.
Although this may increase confidence in the model in the short term, any
dependence on experts will decrease model usefulness either because of the
expense and time required or because of the model’s lack of adaptability to
new parameters, questions, and concerns.
Model developers and users can gain confidence in system dynamics mod-
els through testing (Barlas, 1989, 1996). Three classes of tests are suggested:
structure tests, behaviour tests and policy implication tests. Structure tests
determine how well the structure of the model matches the structure of reality.
This is the case when every model component has a real world counterpart
and when every key factor contributing to the problem in the real world has
a model counterpart. As descriptions of system structure are generally not
available, they have to be extracted from the mental models of people familiar
with the system. However, system understanding of different actors is usually
not identical. One goal of participative modelling may thus be to increase the
degree to which overlap occurs, i.e. build consensus. Furthermore, key factors
contributing to the problem may be unrecognised prior to modelling and there
is no guarantee that they will be discovered during the model development
process. Behaviour tests determine how consistently model outputs match
real world behaviour. This can either be based on available time-series data
or the correlation of mental models with established reference modes (Ster-
man, 2000). The usefulness of the former clearly depends on the quality of the
available historical data, while the latter necessitates a substantive and co-
herent overlap in mental models. Policy implication tests determine whether
the observed system responses to policy changes replicate model predictions.
These tests are rarely conducted as they take place after implementation when
the development team’s involvement has usually ceased. This underlines the
need for a transparent model developed in collaboration with the end user.
Statistical tests are commonly not conducted with system dynamics models
as the focus is on the interplay between all model components and model
behaviour rather than certain parts of it.
Operations research type models are able to provide exact, optimal solu-
tions because of the way in which problems are articulated, focussing upon
one-dimensional engineering-based approaches to water supply performed with
little regard for social, cultural and environmental values/implications (Gle-
ick, 2000). Inevitably, the limited regard for social dimensions, sustainability
or biodiversity management is now demanding a shift in perspective (Hjorth
and Bagheri, 2006). Increased recognition and acceptance of complexity and
uncertainty has promoted increased use of flexible simulation based tools, such
as those provided by SDM applications (Sterman, 2002; Vriens and Achter-
bergh, 2006).
Water resource managers need to be aware of a number of limitations of
system dynamics before considering its use. Inherent uncertainties of com-
plex open systems implicate that SDM will not provide exact solutions and
answers. It is thus not suited to address well-defined operational problems.
Concerns for model depth may be evident, reflecting the level of aggrega-
tion. Clearly, in light of existing uncertainty, a detailed system description is
pointless. The level of detail should mirror the problem description and be
effective in addressing the problem in its entirety while striving to be parsi-
monious to aid model transparency and ease of understanding (Saeed, 1992).
The quantification of qualitative variables may be challenging but qualita-
tive data collection and analysis techniques may be utilised (Luna-Reyes and
Andersen, 2003). Indeed, differences in value judgements can dramatically in-
fluence which policies are ultimately recommended (Andersen and Rohrbaugh,
1992). Furthermore, the definition of the problem boundary, i.e. the model
breadth, can be problematic. Modellers are advised to be parsimonious and
only include variables if they contribute to generating the problem behaviour
as experienced in reality (Sterman, 2000). This highlights the fact that sys:
tem dynamics modelling is more of an art than a science. Providing rigour in
the light of complexity and uncertainty indeed seems to be the main challenge
of this approach. The likelihood that two individuals will develop the same
system dynamics model given a complex problem statement is small (Ansoff
and Slevin, 1968).
Given its flexibility and transparency, and increasing recognition of the
complex, multi-dimensional nature of water resource management issues, SDM
provides valuable tools for analysing complex interdisciplinary problems that
inherit uncertainty, aiding efforts at foresighting and guiding decision-making.
Key factors that have assisted these developments have been the capacity to
integrate qualitative and quantitative information, the ability to integrate a
wide range of input parameters in a meaningful way (reflecting their inherent
interactions and feedbacks), explicit recognition of multiple forms of uncer-
tainty, and recognition that the direction of change is the key parameter to
effectively guide management programs and responses in an adaptive fashion.
3 Evolving use of system dynamics method-
ology in water resources management
Over the last 40 years em dynamics applications in WRM have branched
off in many directions. We categorise these by their main problem foci: re-
gional analysis and river basin planning, urban water, flooding, irrigation and
pure process models (Figure 3). For an overview of general dynamic water re-
sources management models the reader is referred to Fleming (1975), Donigian
(1981), Troendle (1985), El-Kadi (1989), DeVries and Hromadka (1993), Singh
(1995) and particularly Wurbs (1994). Only Wurbs (1994) contains a section
on modelling using the object oriented (system dynamics) Stella” modelling
environment and one reference of its application.
The development of system dynamics models to analyse problems and
identify solutions for improved water resources management has a long tra-
dition. The Stanford Watershed Model (Crawford and Linsley, 1966) is com-
monly credited as being the first comprehensive watershed simulation model,
developed shortly after the emergence of SDM (Forrester, 1958). However,
while the Stanford Watershed Model was a hydrological response process
model aimed at simulating physical water flows and stores, the first system dy-
namics models included physical as well as socio-economic factors in order to
improve understanding of long-term systemic issues faced in the region. The
Susquehanna River Basin Model developed by Hamilton during the 1960’s and
published in book form in 1969 aimed to understand the interdependencies
between water resources and their management on the one hand and quantifi-
able social and economic factors (demographics, employment, industry) on the
other (Hamilton, 1969). This additional model complexity, while visionary at
the time given available computer capabilities, came at the cost of increased
data aggregation and larger spatial scale.
The use of system dynamics for integrated regional analysis has contin-
ued to this day. While spatial scales have shifted from regional (Camara
et al., 1986; Cartwright and Connor, 2003; Cohen and Neale, 2006; Connor
et al., 2004; Den Exter, 2004; Den Exter and Specht, 2003; Guo et al., 2001;
Leal Neto et al., 2006; Passell et al., 2003; Sehlke and Jacobson, 2005; Xu,
2001; Xu et al., 2002) to national (Simonovic and Fahmy, 1999; Simonovic and
Rajasekaram, 2004) to global (Simonovic, 2002a,b), so too have the number
of socio-economic factors included, mirroring improved computer capabilities
as well as changing problem foci (global water crisis and social impacts).
Simonovic and Rajasekaram (2004) note a recent trend in the reduction of
spatial scales to basin and watersheds in order to identify regional and local
solutions.
Applications in regional analysis have often had a strong economic fo-
cus examining feedback relationships between industry and available water
resources. River basin and watershed management applications focus more
narrowly on water resources and their interaction with population growth
(Costanza and Ruth, 1998; Ford, 1996; Gastelum Perez, 2006; Huerta, 2004;
Leal Neto et al., 2006; Peterson et al., 2004; Sander et al., 2000; Tidwell et al.,
2004; Van den Belt, 2004). As with regional analysis tools, temporal scales of
these models are typically long-term (50-100 years).
Urban water resources management may be seen as a special case of water-
shed management where concerns are more immediate and more contentious.
This typically increases model complexity challenging the model development
process (Bagheri, 2006; Bagheri and Hjorth, 2007; Grigg, 1997; Passell, 2004;
Stave, 2002; Wallace et al., 1988). Furthermore, spatial boundaries for water
resources management are harder to establish due to the intricate nature of
10
water transfers from the far hinterland.
During the 1990’s, SDM projects increasingly incorporated participatory
methods, particularly in the areas of regional analysis, and regional and ur-
ban watershed management. This reflects increasing demands for stakeholder
involvement and public-centered decision making in environmental resource
management. Critically, SDM has the flexibility and capability to support
deliberative-analytical processes effectively.
Further recent research foci have been the management of flooding (Ah-
mad and Simonovic, 2000, 2001, 2004, 2006; Li and Simonovich, 2002) and
irrigation (Diaz-Ibarra, 2004; Fernandez and Selma, 2004; Saysel, 2004; Saysel
et al., 2002). In these areas, models increasingly aim to investigate spatial out-
comes (Ahmad and Simonovic, 2004) and operational planning over shorter
temporal scales.
Pure SDM process models are rare, for examples see Vezjak et al. (1998)
and Abbott and Stanley (1999), with restricted spatial and temporal scales.
This reflects the limited use and acceptance of SDM for well-defined and de-
tailed problems. Although few applications are evident, use of SDM to as
water quality dynamics are also noteworthy (Albuquerque, 2001; Fasset and
Rostapshov, 2001; Hines and Knight, 1971). The flexibility of this methodol-
ogy is also reflected in the analysis of institutional process
management (Gates et al., 1970; Males and Gates, 1971).
3S
s in water resources
From the outset, system dynamics applications aimed to integrate various
physical, social and economic factors influencing water resources management
with a view to addressing and planning for intra- and inter-sectoral long-term
problems. Indeed, system dynamics appears to be the methodology of choice
for these multidisciplinary and multi-actor problems. Consequently, temporal
scales of these models need to be long-term to be able to reflect system delays.
However, despite many applications in the 1960’s and 1970’s uptake of
SDM subsequently stalled. Initial acceptance of these type of applications
was low and few projects made it to publication. Models were often highly
aggregated and thus did not address the day-to-day operational concerns of
municipal water managers (for example Grigg and Bryson (1975); Hamilton
(1969)). Moreover, the zeitgeist of the time was that everything seemed to be
analysable and solvable, so that complexity and uncertainty could be disre-
garded (Ackoff, 1993, 1974). The strength of operations research type models
to provide exact, optimal solutions for these type of problems and likely the
negative publicity and misunderstanding at the time surrounding the “Urban
Dynamics” (Forrester, 1969) and “Limits to Growth” (Meadows et al., 1972)
studies may have lead to a growing apprehension towards SDM that can still
be felt today (Lane, 2000; Sharp and Price, 1984). Interestingly, underlying
concepts of the systems approach became entrenched in the systems approach
to water resources management (Biswas, 1976; Grigg, 1977). Grigg defines
this systems approach as “a systematic method to conceptualize the water
resources ‘system’ and use the tools of systems analysis (databases, models,
1
“spoyjaut finoqodiayund pasyryn suoynoyddn yorym ayoowpur sMoLs0 psvn
-unog ‘jndyno yosvasas joorjonid pun poorja.oay} fiay fo auyaun, Y 2g aanSLq
ovoz S00 002 S661 066 seer og6r sie) oz6r S961 096 eset
jesfeg ‘ewjag pue zapuewiay ‘evieq)-zeIq « |
| Tere testes «
uo}e6 14] <eeme
‘SINoUOWIS pue PeWILy « |
dinouoWIS pue peWYy «
=
od
ss
B6up < Jouled ¢ = 3
| 78 19 90810 « uoshig pue BUD < Zo
juaweGeuew seoinosai 19}em bean <A AA RRR NT 3 g
| TET PHI 32
12 19 O1ON J207 :2aJeq wnja\se9 ‘STEEN PUE UEYOD q Fac)
Uuosqooer pue exues < gs
Wd USP UPA ‘[eMpil :weseyoseley pue SIAOLOWIS ao
3o}K3 UB "72 19 JOUU0D 3
a
Awyey pie 9
‘Uyny pu ezuE\suo9 <]
Buruuejd uiseq s9Au pue|stsAleue |U0\65_) <a
xuusp puesnomnon a | | |) E |
yea uspluen 4
Weg ep ueh 4
uopieyply pue Ussiepuy “72 79 Wesiepuy «
i XIUUOA 4
USA 4 3
OUIM ¢ 2
ye je xiuuap « g
PPh dais 3
Buyjepow Aovedponied ee ee eee o
ie 18 NO} 4 jeje eiewe5 < 2
BamioH < 9
eye syono7 < BBUD « o3
Semsig < 2,
re 18 Uosiepuy « 8
WAM 01 Yoeoddcle SLs} Ag ss es 3
ewes ‘exeiqeUuIM pue UOIeeG ¢ 3
ploy « Fb
uospieyory < ebues « @
SUJOUUEISION < JB 79 SUEGDH «
ung pue uospieyory «
worry « 191881104 «
_ siepuey <
J91821104 4
SLL
ie 00e, oat
S661
t
066
set ose sl6L
i)
o6L
96t
tT
aie
get
12
GIS) to identify and evaluate management strategies” (Grigg, 1997). These
applications supported the emergence of Integrated Water Resources Manage-
ment in the late 1980’s (Bowden and Glennie, 1986; Da Cunha, 1989; Rogers,
1993). Integrated water resources management not only acknowledges the in-
tegrative nature of water resource problems but also the need to incorporate
multiple objectives and involve multiple stakeholders in decision making.
During the 1990’s SDM applications became more varied. Models either
increased in complexity to make them more useful in operational planning
(c.f. work by Simonovic) and provide expert knowledge in areas such as power
generation, flooding and reservoir control, or models increased in simplicity
and became even more aggregated in order to support learning and problem
understanding (c.f. work by Ford (1996); many other applications not pub-
lished). Models aimed at strategic planning and policy making persisted, but
over the last ten years participatory methods have taken a strong hold. Re-
quests for participative adaptive management were increasingly voiced and
legislation, such as the European Water Directive, now prioritises stakeholder
participation in water management (European Union).
While many initial applications have their origins in the United States,
recent years have seen a global spread with particular uptake in Europe. The
three trends of using system dynamics methodology for operational planning,
learning support and participatory strategic planning are expected to persist.
4 Discussion
Reviewed work illustrates that careful problem definition and project focus
are paramount for a successful project. To a certain degree this is a necessary
condition for any modelling project. However, SDM enables more leeway in
setting model breadth and depth, and thus requires that more attention is
paid to these details at the outset. Similar findings are reported by Eskinasi
and Fokkema (2006), who indicate that the lack of project definition and
model scope in application of a SDM project ultimately lead to its downfall.
SDM provides the unique opportunity to model and test long-term effects
of management decisions and strategies in uncertain and complex s:
while facilitating stakeholder involvement and supporting consensus building.
A focus on these strengths should persist throughout any project to avoid
the modeller or end-user becoming bogged down in details of model structure
or statistical validity. This requires that the selected level of aggregation
corresponds well with the problem of interest and the data available. For
example, Stave (2003) illustrates that a model does not have to be complex
in order to be useful and achieve desired outcomes.
Where outcomes have not been achieved, for example see Grigg (1997)
this was notably due to lack of careful initial scoping which often takes place
independently from the anticipated end-users.
Stakeholder involvement in any SDM project can vary to a great extent.
13
Stakeholders can be fully engaged in the model development process itself,
contribute by suggesting strategies, experiment with a complete model, or
simply provide feedback in an information session. It is considered advanta-
geous that use of SDM does not require any knowledge of the methodology,
modelling or computer simulation, such that this approach can be used with
any group of stakeholders. The degree to which stakeholders are involved will
to some extent contribute to the success of the project and as a result should
be carefully considered. If stakeholders are expected to take ownership of the
model/decision support tool, they have to be included in problem definition
and project scoping processes at the outset.
In contrast to other dynamic simulation approaches, SDM is based on a
better understanding of system structure. However, this does not make sim-
ulation results any more valid and care must be taken not to oversell the
methodology. To date, major applications lie in regional analysis and river
basin planning, urban water management, flooding and irrigation. Resis-
tance to a wider uptake manifests from misconceptions regarding the purpose
and value of systems models, historical animosities with system dynamics,
the notion that SDM is “just another method”, the relative unfamiliarity of
SDM outside North-America and Europe, the notorious lack of time of many
managers to engage in conceptual broad-picture thinking, as well as the com-
plexities and problems surrounding implementation of systemic solutions.
Drawing on our review of the literature, we summarise common pitfalls
and criteria of successful SDM projects in Table 2.
Success criteria Common pitfalls
Careful problem definition, scoping Insufficient definition of the prob-
of project aims and model bound- lem, project purpose and deliver-
aries ables;
Institutional arrangements, bound-
aries and organizational politics
that limit uptake are not addressed
at the outset (Eden et al., 1979);
Misunderstanding or misjudging the
purpose and value of conceptual or
systems modelling and simulation.
Conceptual and numerical models Extending model boundaries during
of high quality, ie. models that modelling;
are parsimonious and able to com- A focus on modelling a system
municate well the system structure rather than modelling a problem
and dynamic behaviour pertaining leading to overly complex models
to the problem at hand. that are difficult to understand for
non-experts
14
Models that are tested to the extent
that end users and/or stakeholders
are confident in the simulation re-
sults
Stakeholder participation to the ex-
tent that it supports defined project
goals
A focus on personal and _ institu-
tional learning and change
Support for implementation of re-
sults at the outset
Models that can be revised and up-
dated in order to support adaptive
project management
Models that are based on individual
untested mental models
Independent model building that
does not lead to insights and may
appear obscure to end users and/or
stakeholders;
Misjudging the conflict potential
among stakeholders and a lack of
s that support consensus
proce:
building;
Underrating the impact of individ-
ual differences in value judgements
by stakeholders;
Targeting and selecting stakeholders
in order to evade conflict
Expectations of exact predictions;
Overrating short-term system ef-
fects over longer term effects
Lack of top-management support
for the implementation of short-
term or long-term solutions, and
their monitoring
Overly complex models that are not
based on modules making revision
and extension difficult
Table 2: Common pitfalls and suc
revi
criteria observed during literature
Numerous research gaps are well suited for system dynamics modelling:
water quality management especially the assessment of spatial effects of land
use change and non-point source pollution; the interconnections of the four
different types of water in urban vs. rural areas; the analysis of institutional
decision processes and stakeholder dynamics; modelling in support of multiple
bottom line reporting as well as virtual water trading dynamics.
5 Conclusions
New agenda items of sustainability, multiple bottom lines, stakeholder partic-
ipation and the efficient management of scarce and contested water resources
pose significant challenges for resource planners and managers. Dynamic sim-
15
ulation methodologies such as system dynamics programming have been sug-
gested and applied to address these issues and scope the future. Despite
various limitations, system dynamics is well suited for multidisciplinary and
multi-actor problems but not operational problems in integrated water re-
sources management.
However, the value of such procedures is limited if outcomes are not be-
lieved or uptake of outcomes is not forthcoming. SDM provides tools with
considerable flexibility with which to approach foresighting exercises. How-
ever, findings from this review indicate that prospects for success are max-
imised when the group itself constrains the definition of the problems to be
addressed, and participatory procedures are applied in scoping, development
and testing of the model. Involvement underpins ownership, providing the
platform for management applications that are not only responsive to group
concerns, but also have greater prospects for effective implementation and
uptake.
SDM offers prospects to enhance the resilience of the system as a whole.
It provides a well-grounded, flexible and realistic approach to identifying and
dealing with inherent uncertainties in water resources management. Hence,
it prospectively provides a critical tool in adaptive management applications,
assisting in derivation and ownership of realistic visions for integrated wa-
ter resources management, and the development of strategies that must be
adopted to achieve these goals. Given the openness and transparency of
participatory processes, SDM also provides an opportunity to meaningfully
test projections of em futures and the reliability/deficiencies in our un-
derstanding. In light of these insights, we are able to progressively adapt
management strategies to changing circumstances. Such flexibility is vital in
responding to prevailing development pressures, climate change and measures
that deal with the long term consequences of our past “command-and-control”
legacies. In addition, application of SDM procedures provides opportunities
to monitor performance indicators and enhance effectiveness in our quest to
manage water effectively and sustainably.
6 Bibliography
M.D. Abbott and R.S. Stanley. Modeling groundwater recharge and flow in an
upland fractured bedrock aquifer. System Dynamics Review, 15(2):163-184,
1999.
R. L. Ackoff. From mechanistic to social systemic thinking. Systems Thinking
in Action Conference, Transcript of the Plenary Address, November 1993.
URL http://acasa.upenn.edu/socsysthnkg.pdf. Retrieved on 30 Jan
2006.
16
R.A. Ackoff. Redesigning the future: A systems approach to societal problems.
Wiley, New York, USA, 1974.
S. Ahmad and S.P. Simonovic. Dynamic modeling of flood management poli-
cies. In Proceedings of the 18th International Conference of The System
Dynamics Society: Sustainability in the Third Millennium, Bergen, Nor-
way, 6-10 August 2000.
S. Ahmad and S.P. Simonovic. Modeling dynamic processes in space and time:
A spatial system dynamics approach. In D. Phelps and G. Sehlke, editors,
World Water Congress 2001 Bridging the Gap: Meeting the World’s Wa-
ter and Environmental Resources Challenges, Orlando, Florida, USA, May
2001. doi: 10.1061/40569(2001)88.
S. Ahmad and S.P. Simonovic. Spatial system dynamics: New approach
for simulation of water resources systems. Journal of Computing in
Civil Engineering, 18(4):331-340, October 2004. doi: 10.1061/(ASCE)
0887-3801 (2004) 18:4(331).
S. Ahmad and S.P. Simonovic. An intelligent decision support system for
management of floods. Water Resources Management, 20(3):391-410, 2006.
doi: 10.1007/s11269-006-0326-3.
S. Albuquerque. A system model for source water assessment in the Las Vegas
valley. Master’s thesis, Department of Civil and Environmental Engineer-
ing, University of Las Vegas, Nevada, USA, 2001.
D. F. Andersen and J. Rohrbaugh. Some conceptual and technical problems in
integrating models of judgment with simulation models. IEEE Transactions
on Systems, Man, and Cybernetics, 22(1):21-34, 1992.
J.M. Anderson, A.A. Anderson, and G.D. Feree. Dynamic models in water
resources planning. Technical Report 72-3, Center for Advanced Planning,
Institute for Water Resources, U.S. Army Corps of Engineers, Alexandria,
Virginia, USA, 1972.
H.I. Ansoff and D.P. Slevin. An appreciation of industrial dynamics. Man-
agement Science, 14(7):383-397, March 1968.
A. Bagheri. Sustainable development: Implementation in urban water systems.
PhD thesis, Lund Universitet, Sweden, 2006. ISBN: 978916286789.
A. Bagheri and P. Hjorth. A framework for process indicators to monitor for
sustainable development: practice to an urban water system. Environment,
Development and Sustainability, 9(2):143-161, May 2007. doi: 10.1007/
s10668-005-9009-0.
17
Y. Barlas. Multiple tests for validation of system dynamics type of simulation
models. European Journal of Operational Research, 42:59-87, 1989.
Y. Barlas. Formel aspects of model validity and validation in system dynamics.
System Dynamics Review, 12(3):183-210, Fall 1996.
Y. Barlas and $8. Carpenter. Philosophical roots of model validation. System
Dynamics Review, 6(2):148-166, Summer 1990.
J.S. Baron, N.L. Poff, P.L. Angermeier, C.N. Dahm, P.H. Gleick, R.B. Jack-
son, C.A. Johnston, B.D. Richter, and A.D. Steinman. Meeting ecological
and societal needs for fresh water. Ecological Applications, 12(5):12471260,
2002.
AK. Bis . Systems approach to water management. Mc Graw Hill, 1976.
ISBN: 0070054800.
M.J. Bowden and J.M. Glennie. Integrated water resource management on the
central plains: the need for irrigation. New Zealand Agricultural Science,
20(1):18-22, 1986.
G. Brierley and C. Cullum. Environmental science and management in a
changing world. Geojournal, in Review.
A.S. Camara, M.G. Viegas, and A. Amaro. Interfacing system dynamics and
multiobjective programming for regional water resources planning. Annals
of Regional Science, 20(3):104-113, November 1986.
L. Cartwright and J. Connor. Collaborative water supply planning: A shared
vision approach for the Rappanhannock Basin. In Proceedings of the 2003
UCOWR Meeting, Washington DC, USA, 2003. Universities Council on
Water Resources.
H.M.L. Chaves and S. Alipaz. An integrated indicator based on basin hy-
drology, environment, life, and policy: The watershed sustainability in-
dex. Water Resources Management, 21(5):883 — 895, 2007. doi: 10.1007/
s11269-006-9107-2.
S. Cohen and T. Neale. Participatory integrated assessment of water manage-
ment and climate change in the Okanagan Basin, British Columbia. Final
report, Environment Canada and University of British Columbia, 2006.
. Connor, L. Cartwright, and K. Stephenson. Collaborative water supply
planning: A shared vision approach for the Rappahannock Basin in Vir-
ginia. In G. Sehlke, D.F. Hayes, and D.K. Stevens, editors, Proceedings
Of The 2004 World Water and Environmental Resources Congress, pages
1-9, Salt Lake City, Utah, USA, 27 June ~ 1 July 2004. Environmental and
18
Water Resources Institute of The American Society of Civil Engineers. doi:
10.1061 /40737(2004)323.
R. Costanza and M. Ruth. Using dynamic modeling to scope environmental
problems and build consensus. Environmental Management, 22(2):183-195,
1998.
N.H. Crawford and R.K. Linsley. Digital simulation in hydrology: Stanford
Watershed Model IV. Technical Report 39, Civil Engineering Deptartment,
Stanford University, California, USA, 1966.
L.V. Da Cunha. Water resources situation and management in the EEC.
Hydrogeologie, 2:57-69, 1989.
M. L. Deaton and J. J. Winebrake. Dynamic Modeling of Environmental
Systems. Modeling Dynamic Systems. Springer, New York, USA., 2000.
ISBN: 0-387-98880-7.
K. Den Exter. Integrating Environmental Science and Management: The Role
of System Dynamics Modelling. PhD thesis, CRC Sustainable Tourism
School of Environmental Science and Management, Southern Cross Uni-
versity, Queensland, Australia, 2004.
K. Den Exter and A. Specht. As:
system dynamics group model-building. In ”Extending extension: Beyond
sting stakeholder decision making using
traditional boundaries, methods and ways of thinking!”. Proceedings of the
2003 APEN National Forum, Hobart, Australia, 2003.
J.J. DeVries and T.V. Hromadka. Handbook of Hydrology, chapter Computer
models for surface water, pages 21.1-21.39. McGraw-Hill, New York, USA,
1993.
M.A. Diaz-Ibarra. A System Dynamics Model of El-Paso County Water Im-
provement District No. 1. PhD thesis, Environmental Science and Engi-
neering, The University of Texas, El Paso, USA, 2004.
J.J. Dolado. Qualitative simulation and system dynamics. System Dynamics
Review, 8(1):55-81, Winter 1992.
A.S. Donigian. Modeling Components of Hydrologic Cycle, chapter Water
quality modeling in relation to watershed hydrology, pages 343-382. Water
Resources Publications, Littleton, Colorado, USA, 1981.
J.K. Doyle and D.N. Ford. Mental models concepts for system dynamics
research. System Dynamics Review, 14(1):3-29, Spring 1998.
R.L. Eberlein. Simplification and understanding of models. System Dynamics
Review, 5(1):51-68, Winter 1989.
19
C. Eden, D. Sims, and S$. Jones. Policy analysis and organizational politics.
European Journal of Operational Research, 3:207-215, 1979.
A.I. El-Kadi. Watershed models and their applicability to conjunctive use
management. Water Resources Bulletin, 15(1):125-137, 1989.
M. Eskinasi and E. Fokkema. Lessons learned from unsuccessful modelling
interventions. Systems Research and Behavioral Science, 23:483-492, 2006.
doi: 10.1002/sres.774.
European Union. The EU Water Framework Directive - integrated river
basin management for Europe. URL http: //ec.europa.eu/environment/
water/water-framework/index_en.html. last updated 22/3/2007.
M. Everard and A. Powell. Rivers as living systems. Aquatic Conservation:
Marine and Freshwater Ecosystems, 12:329-337, 2002.
C. Fasset and O. Rostapshov. Riverweb water quality simulator - creating
an educational model. Poster at the American Geophysical Union Spring
Meeting, 2001.
J.M. Fernandez and M.A.E. Selma. The dynamics of water scarcity on irri-
gated landscapes: Mazarron and Aguilas in south-eastern Spain. System
Dynamics Review, 20(2):117-137, 2004. doi: 10.1002/sdr.290.
G. Fleming. Computer simulation techniques in hydrology. Elsevier, New
York, USA, 1975.
A. Ford. Testing the Snake River Explorer. System Dynamics Review, 12(A):
305-329, Winter 1996.
A. Ford. Modeling the Environment: An Introduction to System Dynamics
Modeling of Environmental Systems. Island Press, Washington DC, USA,
1999.
J.W. Forrester. Industrial dynamics — a major breakthrough for decision
makers. Harvard Business Review, 36(4):37-66, 1958.
J.W. Forrester. Principles of Systems. Productivity Press, Portland, Oregon,
USA, 1968.
J.W. Forrester. Urban Dynamics. MIT Press, Cambridge, Massachussetts,
USA, 1969.
J.W. Forrester. Lessons from system dynamics modelling. System Dynamics
Review, 3(2):136-149, Summer 1987.
J.R. Gastelum Perez. Analysis of water resources alternatives to improve
water allocation on the Conchos Basin during drought situations. PhD
thesis, University of Arizona, USA, 2006. ISBN: 9780542528262.
W.E. Gates, R.M. Males, and J.F. Walker. Application of industrial dy-
namics concepts to decision making in environmental management. Water
Resources Research, 6(6):1549-1558, 1970.
P. H. Gleick. The changing water paradigm - a look at twenty-first century
water resources development. Water International, 25(2):127-138, March
2000.
N.S. Grigg. Management of water resources: a systems approach. Colorado
State University, Ft. Collins, Colorado, 1977.
N.S. Grigg. Systemic analysis of urban water supply and growth management.
Journal of Urban Planning and Development, 123(2):23-33, 1997.
N.S. Grigg and M.C. Bryson. Interactive simulation for water system dynam-
ics. Journal of Urban Planning and Development, 101(UP1):77-92, May
1975.
H.C. Guo, L. Liu, G.H. Huang, G.A. Fuller, R. Zou, and Y.Y. Yin. A system
dynamics approach for regional environmental planning and management:
A study for the Lake Erhai Basin. Journal of Environmental Management,
61(1):93-111, January 2001. doi: 10.1006/jema.2000.0400.
H.R. Hamilton. Systems simulation for Regional Analysis: An application
to river-basin planning. MIT Press, Cambridge, Mass., 1969. ISBN:
0262080303.
R.T. Hand. System Structure, Natural History, Dynamic Modeling and Adap-
tive Management of the Mekong Watershed’s Tonle Sap-Great Lake, Cam-
bodia. PhD thesis, Graduate School of the University of Maryland, 2003.
O.J. Helweg. Water resources planning and management. John Wiley & Sons,
1985.
E. Higgs. Nature by Design: People, Natural Process and Ecological Restora-
tion. MIT Press, Cambridge, Massachusetts, 2003.
W.W. Hines and J.E. Knight. Complex systems analysis of water quality
dynamics - the feedback systems structure. Technical Report ERC-0570,
School of Industrial and Systems Engineering and Environmental Resources
Centre, Georgia Institute of Technology, Atlanta, USA, 1971.
P. Hjorth and A. Bagheri. Navigating towards sustainable development: A
system dynamics approach. Futures, 38:74-92, 2006. doi: —10.1016/j.
futures.2005.04.005.
A.M. Hodgson. Hexagons for systems thinking. European Journal of Opera-
tional Research, 59:220-230, 1992.
C. 8. Holling and G. K. Meffe. Command and control and the pathology of
natural resource management. Conservation Biology, 10(2):328-337, 1996.
doi: 10.1046 /j.1523-1739.1996.10020328.x.
J.B. Homer. Why we iterate: scientific modeling in theory and practice.
System Dynamics Review, 12(1):1-19, Spring 1996.
C.W. Howe, D.R. Schurmeter, and $.W. Douglas. Innovative approaches to
water allocation: the potential of water markets. Water Resources, 22(4):
439-445, 1986.
J. M. Huerta. A System Dynamics Approach to Conflict Resolution in Water
Resources: The Model of the Lerma-Chapala Watershed. In Proceedings of
the International Conference of the System Dynamics Society, 2004.
M.S. Islam, T. Oki, S. Kanae, N. Hanasaki, Y. Agata, and K. Yoshimura.
A grid-based assessment of global water scarcity including virtual water
trading. Water Resources Management, 21(1):19-33, 2007.
R.B. Jackson, $.R. Carpenter, C.N. Dahm, D.M. McKnight, R.J. Naiman,
S.L. Postel, and S.W. Running. Water in a changing world. Ecological
Applications, 11(4):10271045, 2001.
D. C. Lane. Should system dynamics be described as a ‘hard’ or ‘deterministic’
systems approach? Systems Research and Behavioral Science, 17:3-22,
2000.
A. de C. Leal Neto, L.F.L. Legey, M.C. Gonzalez-Araya, and S. Jablon-
ski. A system dynamics model for the environmental management of the
Sepetiba Bay watershed, Brazil. Environmental Management, 38(5):879
888, November 2006. doi: 10.1007/s00267-005-0211-5.
I. Li and S.P. Simonovich. System dynamics model for predicting floods from
snowmelt in North American prairie watersheds. Hydrological Process Jour-
nal, 16:2645-2666, 2002.
D.P. Loucks, E. van Beek with contributions from, J.R. Stedinger, J.P.M.
Dijkman, and M.T. Villars. Water resources systems planning and man-
agement: An Introduction to Methods, Models and Applications. Princeton
Hall, Englewood Cliffs, New Jersey, USA, 1981/2005. ISBN: 9231039989.
L. F. Luna-Reyes and D. L. Andersen. Collecting and analyzing qualitative
data for system dynamics: methods and models. System Dynamics Review,
19(4):271-296, 2003. doi: 10.1002/sdr.280.
22
B. Luo, G.H. Huang, Y Zou, and Y.Y. Yin. Toward quantifying the ef-
fectiveness of water trading under uncertainty. Journal of Environmental
Management, 82(2):181-190, April 2007.
R.M. Males and W.E. Gates. Decision processes in water quality management.
Final report submitted to Office of Water Resources Research, Department
of the Interior, Engineering Science, Inc., Research and Development Lab-
oratory, Systems/Behavioral Studies Division, Oakland, California, USA,
April 1971.
D.H. Meadows, D.L. Meadows, J. Randers, and W.W. Behrens. The Limits
to Growth. University Books, New York, USA, 1972. ISBN 0-87663-165-0.
J.D.W. Morecroft. System dynamics and microworlds for policymakers. Eu-
ropean Journal of Operational Research, 35:301-320, 1988.
J.D.W. Morecroft. Executive knowledge, models and learning. European
Journal of Operational Research, 59:9-27, 1992.
N. Oreskes, K. Shrader-Frechette, and K. Belitz. Verification, validation and
confirmation of numerical models in the Earth Sciences. Science, 263:641
646, February 1994.
R. Palmer, A.M. Keyes, and S. Fisher. Empowering stakeholders through
simulation in water resources planning. In K. Hon, editor, Proceedings of the
20th Anniversary Conference of the American Society of Civil Engineers,
pages 451-454, Seattle, Washington, USA, 1993.
H.D. Passell. Hydrogeoecological patterns and trends in the upper and middle
Rio Grande, 1975-2002. PhD thesis, University of New Mexico, USA, 2004.
HD. Passell, V.C. Tidwell, S.H. Conrad, R.P. Thomas, and J. Roach. Cooper-
ative water resources modeling in the Middle Rio Grande Basin. Technical
report, Sandia National Laboratories, Albuquerque, New Mexico, USA,
2003.
T.R. Peterson, A. Kenimer, and W.E. Grant. Mediated Modelling, A System
Dyna Approach to Environmental Consensus Building, chapter Using
Mediated Modelling to Facilitate Collaborative Learning Among Resident
of the San Antonio Watershed, Texas, USA, pages 136-163. Island Press,
2004.
J. Randers, editor. Elements of the System Dynamics Methods. Productivity
Press, CAmbridge, Massachussets, USA, 1980.
G.P. Richardson, editor. Modelling for Management - Simulation in Sup-
port of Systems Thinking, volume I and II of The International Library of
23
Management. Dartmouth Publishing Company Ltd, Hants, England, 1996.
ISBN 1 85521 697 3.
G
Richardson and A.L. Pugh HI. Introduction to System Dynamics Mod-
elling. Productivity Press, Cambridge, Massachussets, USA, 1981.
N. Roberts, D. Andersen, R. Deal, M. Garet, and W. Shaffer. Introduction to
Computer Simulation: A System Dynamics Modeling Approach. Produc-
tivity Press, Portland, Oregon, USA, 1983.
K. H. Rogers. The real river management challenge: integrating scientists
stakeholders and service agencies. River Research and Applications, 22(2):
269-280, 2006. doi: 10.1002/rra.910.
P. Rogers. Integrated urban water resources management. Natural Resources
Forum, 17(1):33-42, 1993.
P.P. Rogers and M.B. Fiering. Use of systems analysis in water management.
Water Resources Research, 22(9):146S-158S, 1986.
E.A.J.A. Rouwette and J.A.M. Vennix. System dynamics and organizational
interventions. Systems Research and Behavioral Science, 23:451-466, 2006.
doi: 10.1002/sres.772.
K. Saeed. Slicing a complex problem for systems dynamics modeling. System
Dynamics Review, 8(3):251-262, Fall 1992.
K. Sander, E. Kaneshige, and M. E. Verona. An evaluation of an online
water quality simulator and its success at integrating curriculum. Technical
report, 2000.
A.K. Saysel. System dynamics model for integrated environmental assessment
of large-scale surface irrigation. Technical Report 2, The System Dynamics
Group, Department of Information Science, University of Bergen, Norway,
2004.
A.K. Saysel, Y. Barlas, and O. Yenigiin. Environmental sustainability in an
agricultural development project: a system dynamics approach. Journal of
Environmental Management, 64(3):247-260, March 2002. doi: 10.1006/
jema.2001.0488.
G. Sehlke and J. Jacobson. System dynamics modeling of transboundary
systems: The Bear River basin model. Ground Water, 43(5):722-730,
September-October 2005. doi: 10.1111/j.1745-6584.2005.00065.x.
P. M. Senge. The Fifth Discipline. Random House, Australia, 7th ed. edition,
1992.
J. A. Sharp and D. H. R. Price. System dynamics and operational research:
An appraisal. European Journal of Operational Research, 16:1-12, 1984.
S.P. Simonovic. Global water dynamics: Issues for the 21st century. Water
Science and Technology, 45(8):53-64, 2002a.
S.P. Simonovic. World water dynamics: Global modeling of water resources.
Journal of Environmental Management, 66:249-267, 2002b. doi: 10.1006/
jema.2002.0585.
S.P. Simonovic and H. Fahmy. A new modeling approach for water resources
policy analysis. Water Resources Research, 35(1):295-261, 1999.
S.P. Simonovic and V. Rajasekaram. Integrated analyses of Canada’s wa-
ter resources: A system dynamics approach. Canadian Water Resources
Journal, 29(4):223-250, 2004.
V.P. Singh, editor. Computer Models of Watershed Hydrology. Water Re-
sources Publications, Highlands Ranch, Colorado, 1995.
K.A. Stave. Using system dynamics to improve public participation in envi-
ronmental decisions. System Dynamics Review, 18(2):139-167, 2002.
K.A. Stave. A system dynamics model to facilitate public understanding of
water management options in Las Vegas, Nevada. Journal of Environmental
Management, 67:303-313, 2003.
J.D. Sterman. Business Dynamics: Systems Thinking and Modelling for a
Complex World. McGraw-Hill, Boston, MA, 2000.
J.D. Sterman. Learning in and about complex systems. System Dynamics
Review, 10(2-3):291-330, 1994.
J.D. Sterman. All models are wrong: reflections on becoming a systems scien-
tist. System Dynamics Review, 18(4):501-531, 2002. doi: 10.1002/sdr.261.
V.C. Tidwell, H.D. Passell, S.H. Conrad, and R.P. Thomas. System dy-
namics modeling for community-based water planning: Application to the
Middle Rio Grande. Aquatic Sciences, 66:357-372, 2004. doi: 10.1007/
s00027-004-0722-9.
C.A. Troendle. Hydrological Forecasting, chapter Variable source area models,
pages 347-403. John Wiley & Sons, 1985.
M. Van den Belt. Mediated Modelling - A System Dynamics Approach to
Environmental Consensus Building. Island Press, Washington, USA, 2004.
M.J. Van den Belt. Mediated modeling: A collaborative approach for the de-
velopment of shared understanding and evaluation of environmental policy
scenarios, with case studies in the Fox River Basin, Wisconsin and the Ria
Formosa, Portugal. PhD thesis, University of Maryland College Park, USA,
2000. ISBN: 0493184716.
J.A.M. Vennix. Group-Model Building: Facilitating Team Learning Using
System Dynamics. John Wiley & Sons, Chichester, UK, 1996.
M. Vezjak, T. Savsek, and E.A. Stuhler. System dynamics of euthrophication
processes in lakes. European Journal of Operational Research, 109:442-451,
1998.
D. Vriens and J. Achterbergh. The social dimension of system dynamics-based
modelling. Systems Research and Behavioral Science, 23(4):553-563, 2006.
doi: 10.1002/sres.782.
S.D. Wallace, F.H. Sancar, and H. Fahriye. An integrative approach to water
resources management: An application in Middleton, Wisconsin. In N.B.
Forrester, A. Ford, and J. Homer, editors, Proceedings of the International
Conference of the System Dynamics Society, pages 448-459, La Jolla, Cal-
ifornia, USA, 1988.
E. Wolstenholme. Using generic system archetypes to support thinking and
modelling. System Dynamics Review, 20(4):341-356, 2004. doi: 10.1002/
sdr.302.
E.F. Wolstenholme. System Enquiry: A System Dynamics Approach. Wiley,
Chichester, 1990.
World Water Assessment Program. Water: a shared responsibility. World
water development report no. 2, United Nations, Paris, France, 2006.
R.A. Wurbs. Computer models for water resources planning and management
(sudoc d 103.57/5:94-nds-7). Technical Report IWR Report 94-NDS-7,
US Army Corps of Engineers, Institute for Water Resources, 1994. ISBN:
BOO010OTE2.
H.G. Xu. Exploring effective policies for underground water management in
artificial oasis: a system dynamics analysis of a case study of Yaoba Oasis.
Journal of Environmental Sciences - China, 13(4):476-480, October 2001.
Z.X. Xu, K. Takeuchi, H. Ishidaira, and X.W. Qhang. Sustainability analy-
sis for Yellow River water resources using the system dynamics approach.
Water Resources Management, 16(3):239-261, 2002. doi: 10.1023/A:
1020206826669.
