Mental Models in Urban Stormwater Management
Ines Winz and Gary J. Brierley
School of Geography, Geology and Environmental Management, The University of Auckland
Private Bag 92019, Auckland 1142, New Zealand
Telephone 64 9 373 7599 ext. 88956, Facsimile 64 9 373 7434
i.winz@ auckland.ac.nz, g.brierley@ auckland.ac.nz
Abstract
Different mental models exist regarding environmental problems and solutions. To engender
sustainable approaches to stormwater management, there needs to be understanding of, and
engagement with, these different perspectives. This study used cognitive mapping to elicit and
transparently capture perspectives of 31 stakeholders on problems, solutions and barriers to
the implementation of urban stormwater management. Three core perspectives were
discovered and synthesized as causal loop diagrams: conventional fixes, low impact solutions
and community development.
Analysis confirms perspectives are diverse and conflicting. Each perspective has
shortcomings in providing solutions that can address the perceived challenges. Therefore, the
integration of solutions strategies is proposed. The implementation of low impact solutions
may address environmental degradation and go some way to assist rehabilitation in the short
term. In order to realise sustainable stormwater management, the long term focus must be on
social learning, behaviour change and the creation of effective partnerships between
communities and local authorities.
Keywords: systems thinking, cognitive mapping, soft system dynamics, urban stormwater
management, New Zealand, Project Twin Streams, stream restoration, low impact
development
1 Introduction
An inevitable time delay is evident between shifting perspectives regarding the most
appropriate approach with which to manage environmental systems and the implementation of
measures. While the sustainability agenda now presents a dominant mindset in terms of
desired or even mandated practices, the ongoing deterioration of biophysical conditions is
testimony to the limited effectiveness of transitional processes that promote sustainable
practices. Technological solutions are a fundamental part of reformed approaches to
environmental management. Critically however, they are only a part. Whole of systems
thinking engages all stakeholders in reformation of practices implementing coherent plans of
action that incorporate behavioral/attitudinal participatory approaches through genuine shifts
in societal involvement in lifestyle choices, engagement with environmental repair and
1
maintenance of benefits gained. Such transdisciplinary thinking provides the pivotal
underpinning of sustainable practices.
Water is high on the sustainability agenda particularly in urban areas where huge investments
are required to maintain increasing populations. Balancing the provision of these services,
which import clean water, and export waste and excess rain water, relative to environmental,
social and economic concerns is the primary objective on the journey towards sustainable
urban water systems. We already live with the legacy of past perspectives and actions in
solving stormwater issues, especially those associated with high-cost engineering
infrastructure. As environmental planners and water managers are starting to realize that
engineering infrastructure in itself will not engender sustainable outcomes (Higgs 2003),
contemporary management focuses on providing ‘more natural’, environmentally sensitive
and decentralized engineering solutions. In addition, active public engagement in
environmental repair processes is proposed not only as a means to strengthen decision making
arrangements and increase the likelihood of uptake and implementation of management
activities, but from an understanding that a transition to systemic knowledge and
environmentally friendly behavior can only occur when there are possibilities for exposure to
and active engagement with local natural environments (Butler & Oluoch-Kosura 2006).
However, acceptance of this shift from technological fixes to active public engagement is
lacking among managers as well as the public (Folke 2007).
Pervasive degradation of urban streams and receiving environments is testimony to the fact
that conventional engineering solutions are not working. This has resulted in conceptual shifts
in thinking about environmental management whereby the stream ecosystem as subsystem of
the larger stream network as well as surrounding watershed, are now perceived as complex
adaptive socio-ecological systems for which expert-based approaches are no longer sufficient
to provide effective management (Berkes 2004; Ravetz 2006; Pahl-Wostl 2007). Complex
socio-ecological systems exhibit uncertainty, non-linearities and potentially chaos, a multitude
of feedback processes that interact at different scales and create thresholds of ecosystem
functionality and quality (Grinde & Khare 2008). The very existence of these characteristics
implies a need for flexibility in the management of water resources. Any overly prescriptive
solution lacks reference to stakeholder objectives and thus creates resistance for its
implementation. As a result, solutions have to be developed as part of a long-term adaptive
management strategy. While there are many unanswered questions about what adaptive
management structures and methods are most effective, all build on underlying themes of
commitment to experimentation, the use of iterative modeling processes as well as collective
engagement and ownership.
It is now widely recognized that long-term sustained success in water resource management
will only be achieved through adoption of participatory practices, promoting genuine societal
engagement with the process of environmental repair (Higgs 1997; Gross 2002; Choi 2004;
Naveh 2005; Turner 2005; 2007; Hobbs 2007). Critically, efforts to promote behavioral
change with which to engender sustainable practice and outcomes must first appreciate and
engage with the range of mind-sets that underpin contemporary approaches to environmental
management.
Reported here is a synthesis of results from a case study that investigated key stakeholders’
understanding of stormwater problems and solutions in the Project Twin Streams catchment,
Waitakere City, New Zealand. Cognitive mapping was used to elucidate the different
perspectives on stormwater problems and solutions of 31 research participants with different
backgrounds as stormwater experts, local government officials/managers, scientists, real
estate developers and residents living in the catchment.
2 Stormwater - Problems and Discourses in Management
Stormwater, the flow of water that results from rainfall events, is a disruptive natural force
impacting on urban populations as well as local and regional environments. General problems
caused by stormwater are the flow volume (low flows and high flows/flooding), deteriorating
water quality, and infiltration of stormwater into the wastewater system which can lead to
overflow events. These problems are exacerbated in an urban setting. Urban development
results in an increase in impervious surface areas, e.g. roofs, roads and other paved areas, a
change in vegetation cover, and the compaction of top soil. This greatly reduces infiltration of
stormwater and increases run-off, substantially altering the natural water cycle (Wolman
1967; Amold & Gibbons 1996). Urban pollutants that accumulate on impervious areas are
then carried to urban receiving environments, i.e. streams, rivers, estuaries and harbors.
Associated negative impacts include reductions in habitat quality and availability, amenity
values, ecosystem services, among others (Paul & Meyer 2001; Bunn & Arthington 2002).
Traditionally, stormwater was piped with minimal treatment and disposed of in receiving
environments as quickly as possible. Existing pipe infrastructure is inadequate in filtering the
type and amount of pollutants that exist in urban areas, and insufficient given the current
growth rate of the Auckland region. Overall, this strategy has lead to deteriorating receiving
environments (Hauraki Gulf Forum 2008), a necessity to expensively upgrade existing
infrastructure, and disconnection of citizens with their local streams and other receiving
environments (Peters & Meybeck 2000; Hatt et al. 2004).
In a system with no or minimal anthropogenic influences, natural processes keep stormwater
receiving environments intact by allowing for infiltration of rainwater into the soil, thereby
slowing and detaining flows as well as improving water quality. Modern, water sensitive or
low impact urban stormwater management’ aims to mimic these processes. It has been
defined as “a design approach to site development that protects and incorporates natural site
features into the stormwater management plan” (Auckland Regional Council 2000, p. i-1).
By its very definition LID engineering solutions are decentralized, small-scale and require
only minor if any built structures. LID solutions are shifted off-stream and focus on reducing
imperviousness/drainage connection through the provision of permeable surfaces, e.g.
vegetated swales, raingardens, roofgardens, vegetated buffer zones and house clustering (Hatt
et al. 2004). Permeable areas keep contaminants on-site and prevent them from reaching the
* Common acronyms are LID - Low Impact Development (US), WSUD - Water Sensitive Urban Design (AUS), SUDS —
Sustainable Urban Drainage Systems (UK), LIUDD - Low Impact Urban Design and Development (NZ). This manuscript uses
uD.
stream. Local detention, e.g. in water tanks and small ponds, can reduce run-off volumes and
peak flow. Source control is the reduction of contaminant input into the system. This can be
achieved by, e.g. painting galvanized roofs. LID treatment techniques are associated with the
use of permeable surfaces or small ponds, e.g. flocculation ponds (Auckland Regional
Council 2004a, b; Roy et al. 2008).
In most instances, LID does not explicitly and actively promote behavior change. Note, that
the label low impact design assumes a negative impact on the environment through the
process of development albeit a smaller impact than in conventional development practices.
LID implementation is context driven. For example, water reuse features prominently as a
technique in Australia mirroring water supply concerms, whereas in New Zealand, LID is
mainly driven by stormwater quality objectives (Flynn et al. 2009). Urban areas in developing
countries often lack basic stormwater management altogether. Here, LID is entirely absent
from the management discourse (Silveira 2002; Biswas 2006; Kathuria 2006).
Recent emphasis on stakeholder involvement and social learning has promoted the
establishment of local community stream restoration projects that prevent flooding and
erosion on urban streams while improving environmental values (Kellert et al. 2000;
Bermhardt & Palmer 2007; Rosenberg & Margerum 2008). These projects are often co-
managed, where local authorities engage with communities and provide the support and fund
activities such as clean-ups, re-vegetation, bank stabilization, channel reconfiguration to
improve stream geomorphology and the acquisition of land critical for flood management.
The community then drives most of the stream restoration work mainly focusing on riparian
re-vegetation. Community engagement projects also attempt to raise awareness and facilitate
behavior change. These projects are surrounded by much controversy because the costs are
high and ecological effects in the streams are limited (Kellert et al. 2000; Palmer et al. 2005;
Alexander & Allan 2007; Rumps et al. 2007). However, to date there are no long-term studies
on their effectiveness when it comes to awareness raising and behavioral change. There are
numerous assumptions when engaging in community development processes. For example, it
is often assumed that reconnecting the public with their local environment results in behavior
change or that this behavior change will form the basis for the development of long-term
community ownership. While more research is clearly needed, evidence is emerging to
substantiate claims on biological effectiveness and social change while highlighting numerous
areas of opportunities for improvement (Kellert et al. 2000; Middleton 2001; Purcell et al.
2002; Pahl-Wostl 2006; Alexander & Allan 2007; Rumps et al. 2007).
3 Case Study - Project Twin Streams Catchment
Project Twin Streams (PTS) catchment in Waitakere City is situated between the Manukau
and Waitemata harbours on the westem side of the Auckland region, New Zealand. The
Waitakere Ranges border the catchment in the southwest and most streams’ headwaters
originate there. Streams flow from the foothills of the ranges through the city's urban areas
and into the Waitemata Harbour via Henderson Creek. The catchment area is 10,200 ha with
tural to urban land use types. Population growth and associated urban development is
predicted at 025-35% over the next 20 years (B Osborne 2008, pers. comm. 22 August).
Present population size is 110,000.
Waitakere City stepped up to become New Zealand's first eco-city in 1993 with the objective
of implementing A genda 21 at the local level (Laituri 1996). As a result, programs in different
areas have been implemented to provide environmentally friendly solutions (energy, building,
environmental restoration, etc.). Over the years, Waitakere City Council has become a
proactive local authority with numerous planning documents that actively promote and
encourage LID. In 2003, Waitakere City Council was granted NZ$39.5 million (~US$25
million) in governmental funding for the Project Twin Streams (PTS) stream restoration
program (Waitakere City Council 2008).
The goal of PTS is to create “a sustainable catchment: healthy land, streams and harbours,
and communities who are strong, happy, connected and responsive to the challenges that face
us” (Waitakere City Council 2008). Assumed outcomes included improved water quality and
biodiversity as well as stronger communities through the involvement of local residents and
community organisations in the project. The city council selects well-established and effective
community organizations. Both negotiate a contract and form a partnership to restore the
stream environment by re-vegetating 56km of stream banks (Waitakere City Council 2008).
The rationale for selecting PTS catchment as a case study was that the implementation of
markedly different stormwater management strategies - from conventional to low impact to
project aimed at environmental rehabilitation - created a substantial body of knowledge that I
wanted to tap into. I was particularly interested in finding out about the underlying reasoning
in engaging in different management strategies, their positive and negative outcomes as well
as any problems that were experienced in their implementation. As such, PTS catchment in
the context of stormwater management can be seen as an extreme case which can provide rich
information and allows to “clarify the deeper causes behind a given problem and its
consequences” (Flyvbjerg 2006, p. 229).
4 Methods
Data Collection
Non-probabilistic, purposive/expert sampling including snowballing was used to identify the
31 research participants. These are broadly grouped into stormwater experts (Exp) consisting
of mainly engineers and experienced consultants, local goverment officials/managers (LG)
that were either part of the stream restoration program Project Twin Streams (PTS) or in a
management position, researcher-ecologists (RE) working in universities or research
institutions, private and commercial real estate developers (D) and residents (Res) living in
the catchment.
Semi-structured one-on-one interviews were held between June and September 2007. The
interviews were structured around the following five questions:
e What are the problems associated with stormwater?
e What are the causes and effects of these problems?
5
e What are the solutions to address these issues?
e What are the barriers to implementing these solutions?
e Can you identify the mindsets that are underlying a behavior/decision/solution?
e Can you identify any existing feedback situations?
The interviewees were fully informed of the purpose of the interviews. Care was taken to
minimize interviewer bias by refraining from asking leading questions other than the ones
listed above, and by not introducing any ideas that may form part of the interviewee’s
subsequent answers, as well as being mindful of the interviewer's own verbal and non-verbal
responses.
Answers were mapped out together with the participant utilizing cognitive mapping with
software called Decision Explorer**. Cognitive mapping captures a ‘personal construct
system’, i.e. the elicitation of beliefs, values and expertise of decision makers relevant to the
issue in hand in order to structure, analyze and make sense of written or verbal accounts of
problems (Eden, 1988). The cognitive map is made up of concepts (constructs) that are
interlinked to form chains or hierarchies of cause and effect/ explanations and consequences.
All cognitive maps had a similar structure - from bottom to top: barriers to implementation of
stormwater management techniques, solutions, problems, management goals (Figure 1).
Cognitive maps can provide a medium for problem solving but have been used here as the
basis for structuring and understanding the beliefs of each participant. Maps are coded as
action oriented representations of the participants’ own frame of reference in order to reveal
consequences or implications for all statements made. Arrows that link two concepts thus
show causes or explanations, the implied action as well as its possible outcome(s). As a result,
the map provides meaning not only through individual concepts but the consequences
attributed to them as well as the explanatory concepts that support them.
Cognitive mapping in interviews took on average 40 minutes and resulted in cognitive maps
containing an average of 60 concepts. After each interview, maps were cleaned up to improve
readability and then presented to each participant as a final check. 17 cognitive maps were
confirmed as extensive and correct representation of participants’ own views and thinking
processes. Data saturation occurred approximately after the first 15 interviews. 12 maps are
unconfirmed where mapping could not be performed during the interview (mainly for time
reasons). For these, cognitive mapping was performed based on the interview transcripts. To
avoid bias, the unconfirmed maps were excluded from the quantitative map analysis. The
remaining three interviews were too short to allow for mapping, although answers to the four
questions above were noted. The exercise resulted in 28 cognitive maps consisting of a total
of 1300 concepts.
? Banxia Software Ltd: http://www.banxia.com/dexplore
&
=F
|
S
ES
I
=
El
z
Cognitive Map
09020 81
fe
NY
eg sheep ou)
> reamed es ua yeu suojpe
\ smo jo spedun
eos em op usu cy
zeus
< swoyskso2e omrew
oSpoynowy poy 6e <__pue osqu uo sizeye
= ® ‘ms jo Bupue}siepun
waned jo 3pe1 Zs
srouieg ‘suonnos ‘swogeig qusumaBeuew ssyemuuers dew enuboD
Data Analysis
Individual cognitive maps were analyzed for their structure to determine richness of meaning
for concepts, get a feel for the complexity of each map, and determine similarities between
stakeholder groups. Simple analyses focused on the extent and density of maps with regards
to the number of concepts and links and their relationship to each other to provide indications
of the complexity of the issues (Eden 2004). Further analyses focused on the emergent
properties of maps, i.e. domain analysis to detect central themes, cluster analysis to detect
interconnectedness between themes, and the discovery of feedback loops (virtuous and
vicious cycles), which may indicate dynamic behavior (Eden 2004).
Cognitive maps were combined using several stages of clustering and coded to reveal
common patterns of reasoning. Concepts with identical meaning were combined in the first
clustering stage, e.g. removal of forest and deforestation. In a second stage, concepts with
similar meaning were combined and recoded, e.g. deforestation and removal of bank-side
vegetation. This resulted in 150 clusters. The process was naturally selective and biased:
“Observation is also selective: The researcher is constantly making choices about what to
register and what to leave out, without necessarily realizing that - or why - one exchange or
incident is being noted but another one is not.” (Miles & Huberman 1994, Page 55f).
The 150 clusters were written on post it notes and mapped out on a white board. Links
between the clusters were drawn to form causal loop diagrams reflecting common
perspectives. Causal loop diagrams (CLD) are a key tool in system dynamics methodology for
communicating the feedback structure of systems (Lane 2008). Feedback or the interactions
between system components is the main cause for complex behavior exhibited by social-
ecological systems (Sterman 2000). All dynamics arise from the interaction of self-reinforcing
(positive) or self-correcting (negative) feedback loops.
5 Systems Thinking for Effective Stormwater Management
Structural Analysis of Individual Maps
Results of the structural analysis of the 17 confirmed maps are provided in Table 1. The
average (+SD) number of concepts of all maps is 63.2 +15, the average number of links in all
maps is 87.6 + 27, and the average ratio of links to concepts is 1.4 + 0.2. No statistically
significant similarities between stakeholder groups could be detected. While some participants
had high scores in some of the tests, no group of participants dominated a particular structural
component.
D Exp LG RE Res
Maps a - 5 2 2
Number of concepts 53 59.7418.9 6884129 68417 62+11.3
Number of links 57 84.7432.6 994227 8745.7 58.5+38.9
Links/Concepts 1.1 1440.2 1.4401 1.3404 0.9404
Table 1: Statistics of Structural Analysis of Individual Maps. Abbreviations for stakeholder groups: D -
developer, Exp - stormwater experts/engineers, LG - local government officials, RE -
researcher/ecologist, Res - resident.
Domain analysis argues that if a concept has many ingoing and outgoing links, the concept is
cognitively central and hence dominates a person’s thinking processes (Eden 2004). The
domain analysis for all concepts with more than five links is presented in Table 2. Flooding
was the main concer for about half of the participants followed by LID and Effective
Management of stormwater in general. No one concept dominated the thinking of the
majority of participants. This indicates that different stakeholders perceive different concepts
to be important and provides a justification for including a diverse number of stakeholders in
the research.
Cluster Count D Exp LG RE Res _ AvgLinks
Flooding 8 5 2 1 8.9
UD xz 5 1 iT 10.9
Effective Management 7 t 1 3 2) 8.4
Attitude/Ignorance 5 1 2 2 8
Pollution 5 3 1 a 78
PTS 4 3 i its
Erosion 4 1 1 2 10.3
Community Education 3 2 1 8.7
Imperviousness 3 1 1 1 7.3
Development 2 2 is)
Habitat 2 1 1 8
Species Diversity 2 1 4 a
Delays in consenting process 2 vk 1 7
Behavior Change 2 a, 1 6
Council acceptance 1 >| Ss
Community Buy-in 1 1 8
Table 2: Results of the domain analysis for concepts with more than five ingoing and outgoing links
It was assumed that statistical analysis would show strong similarities in perceptions of
stakeholders within a group and marked differences between groups, for example that
ecologists would be concemed with habitat and engineers with flooding. However, this was
not the case; the similarities in perceptions were largely independent of stakeholder group.
There were a few exceptions when considering domain analysis results for each stakeholder
group. As Table 2 shows stormwater experts (engineers and consultants) are primarily
concemed with LID and flooding. Local government officials thinking prioritized PTS and
pollution. The developer is mainly concerned with delays in the consenting process and
effective management. The highest number of links for a single concept was made by one
expert (21 for LID). Habitat, on the other hand, was a concept that was not considered
important for most stakeholders.
Other map analysis methods were not conclusive although a few feedback loops were
identified. These loops informed the development of the causal loop diagrams introduced in
the next section.
Perspectives in Stormwater Management
Causal loop diagrams were developed based on the saturation of participants’ individual
cognitive maps. Three distinct perspectives emerged: ‘conventional fixes’, ‘low impact
solutions' and ‘community development’. The maps are intended to capture the essence of
these perspectives and include key concepts while avoiding detail. Arrows show causal
relationships and are positive unless indicated otherwise. A negative sign implies a negative
relationship (where an increase in one variable leads to a decrease in the other).
All maps show in black the causes and effects of the three main issues in stormwater
management: flooding, pollution and erosion. Stories underpinning each map, along with
underlying mental models, are supported by quotes.
‘Conventional Fixes’ Perspective
This causal loop diagram (Figure 2) encapsulates the traditional perspective on stormwater
management. The primary management objective at the time was water quantity (flooding)
which was addressed based on a technocratic paradigm by building large scale engineering
infrastructure, particularly underground stormwater pipes. Brown (2005) suggests that this
perspective was most prevalent form the beginning of the 20th century until 1985.
public health
ancnehety damage to urban
= infrastructure
j +, pollution \
y ae 5d ae erosion i
regulatory xf ‘ run-off volume/ }
controls { x velocity/ flooding
end-of-pipe vegetation 4 ; y
treatment = \ \ f
, . Y -
imperviousness iucocaniar
infrastructure
gee
urbanisation Are) 4
Ke . mee
—= ‘
Figure 2: The ‘conventional fixes’ CLD
Urbanisation results in an increase in impervious surface and reduced vegetation. This allows
less entrainment of stormwater into the soil. Water runs off the land and enters the stream
system - often through the stormwater pipe network. The stream has to carry a higher volume
of water at a higher velocity. This increases the likelihood of downstream flooding. Flooding
can damage urban infrastructure including housing, as well as public health and safety.
The damage to urban infrastructure results in an increase in stormwater pipes to prevent
future flooding. However, stormwater pipes increase velocity and volume of water in the
network, which increases the likelihood of future flooding®. This is a self-reinforcing
feedback cycle that may explain the dependence on the ongoing need for pipe solutions (R1).
Whilst initial building costs are very high, maintenance costs for pipes are fairly low when
spread over the working life of the built structure. As a result, urban areas become trapped in
a situation of path dependence. Sunk costs of existing infrastructure are so substantial that a
3 This is because artificial channel networks together with reduced natural drainage density and altered hydrological cycle
contribute to the increase in flow volumes and velocities (Graf 1977; Paul & Meyer 2001).
10
changeover to other technologies seems prohibitive (Kahneman & Tversky 1979; Sterman
2000). However, where the pace of urban development is rapid, costs for extending
infrastructure to newly developed areas are substantial and hit current and future rate payers,
raising concerns of intergenerational fairness and equity.
Another feedback loop exists between stormwater infrastructure and urban development (R2).
This is explained by the following statement:
Participant 19 (RE): “Once you put in any kind of reticulation system then you'll only get
more development. You want to encourage development. When you put a road in somewhere,
people will build along it.”
High-velocity peak flows lead to erosion of the stream channel. Urban pollutants accumulate
on impervious surfaces and are washed off carried to receiving environments by stormwater.
Traditional approaches to solving pollution problems entail end-of-pipe treatments.
Regulatory controls are expected to reduce pollution. However, stormwater largely carries
small amounts of non-point source pollutants for which there is no clear point of origin. For
these, regulatory controls are generally ineffective. This results in a long-term, slow, and often
umnoticed degradation of water quality in receiving environments:
Participant 7 (Exp): People accept creeping deterioration. People will say, “I remember when
I was a kid, there was a little swimming hole at the beach in Brown’s Bay. Well, it’s all full of
mud now, but that’s progress.” Or they might tell you that they don’t like it, but they don’t
know how it could have been prevented.
Pollution, particularly wastewater overflows also impact on public health and safety. This is
where the boundary of this particular perspective was identified. While pollution, erosion and
changes in the catchment’s hydrodynamics greatly affect receiving environments (streams,
estuaries, harbours), this was not a major concern of stormwater managers at the time.
An ‘out of sight, out of mind’ attitude and a lack of value attributed to the environment itself
are underlying our engagement with stormwater in general.
Participant 31 (Exp): In times past, they [streams] were an open drain to any type of rubbish,
weeds, anything, pollution. People would put it into the streams and stormwater system and it
would disappear.
Participant 23 (Exp): It [stormwater] was seen as something to get rid of as opposed to as a
resource to be used. It was a nuisance to have water on your property.
Participant 19 (RE): The developer’s view was a lot more reductionist and said well, let’s look
at this bit of waterway here and let’s say it has got no value from an ecological perspective,
so then let’s develop it. And this bit here, we will develop that. So gradually you lose bits and
bits of waterway to development.
Removing flowing water from the public eye, e.g. by piping it underground, or creating open
concreted channels for flow mitigation, have created a disconnection of the public with the
stream environment:
11
Participant 16 (Res): Most people don’t understand the relationship between what they are
doing on their section and what happens in the stream and how the stream relates to the sea
and how all of that is interconnected.
Participant 28 (LG): You've got ignorance, pure and simple, with people. It’s the “Ah yeah,
bush heals itself, simple enough. You can throw your rubbish in every waterway. Amongst the
cars and other rubbish it will rot away eventually.” They don’t see what damage they have
done.
‘Low Impact Solutions’ Perspective
Water quality concerns emerged as stormwater management objectives during the 1980's
(Brown 2005) and are currently an important driver for public policy and scientific research.
sustainable social-
{ ny
“— ua
government
public health support
\ ae and safety
ieee
ts)
i
"x
run-off volume/
velocity/ flooding he
Oe Sa} \
ee funding for LID
Fs trials
—
S85 uptake Ee
ions
/ cost,
nowledge
of LID
Figure 3: The ‘low impact solutions’ CLD
In the ‘low impact solution’ CLD (Figure 3), in-stream erosion and deposition of sediment
adversely affects the geomorphology of waterways by creating homogenous, wide and flat
stream channels (Reid et al. 2008). Pollution, vegetation loss, erosion and altered stream
geomorphology reduces habitat quality. A broad range of habitats and healthy environments
12
provide biodiversity and a multitude of ecosystem services. These include among others,
cleaning and inactivating of harmful inputs into the ecosystem, providing harvesting from the
land and water to humans, passive and active recreation including amenity, inspiration, artistic
and spiritual engagement. Underlying the ‘low impact solutions’ perspective is an
acknowledgement of the importance of the ecological health of waterways on our own health
and well-being. Both are perceived as cornerstones of sustainable social-ecological systems.
In the context of Project Twin Streams catchment, substantial funds were provided to
purchase and remove flood-prone properties. As a result, the feedback loop between
stormwater infrastructure, flooding and urban infrastructure damage was broken.
All participants felt that maintaining a clean environment was important. However, for some
participants it lost its importance when it came to actually implementing a stormwater
solution and maintaining environmental values was not practicable.
Participant 8 (Exp): I want my children to go to the park where there are watercourses and
eels and fish etc. 1 am a kingfisherman, I want my harbour to be clean. I want to still be able
to catch fish and my children to be able to catch fish. (......) Because as we found out through
progressive development there is inherent value in keeping some watercourses and keeping
riparian margins. That’s what people want, they want a bit of nature. (...... ) The local
council says ‘We want it piped, it’s pointless.’ This is a residential area. The regional council
is saying ‘Ah, no, we might want to keep it.’ And it’s totally impractical.
As an emerging technology the success of LID follows a diffusion/adoption process (Rogers
2003). Uptake of LID is dependent on the adoption rate, i.e. the probability that a person upon
coming into contact with the technology in some way actually uses it. This probability is high
if the contacts are plentiful, the technology displays significant and valuable results (by
effectively and profitably solving the site’s stormwater problem), the technology is
compatible with existing technologies and there are no problems with training to use the
technology. This is reflected in feedback loop R2 in Figure 3. After deciding that a
technology is good and worth pursuing, it is necessary to spend time and effort to train and
become capable and experienced in its use. There will be a time delay from training and
gaining experience in the field. The last point is represented in feedback loop R1 in Figure 3.
This is the learning loop associated with uptake of LID. The use of LID leads to increased
knowledge and increased quality of work.
Participant 4 (Exp): We were out on Friday on an industrial site where they make concrete.
And those sites are notoriously dirty because of all the concrete dust. We always had pH
problems. So this company put in a wetland swale and it is gorgeous. We went out last Friday
when it was raining and did a pH check at the inflow and outflow. The pH going into it was 9
and the pH coming out was 6.5. That is exactly what we want. So here is the case where the
company paid the money, did this and by doing it came up with an environmental outcome
and he still gets to make his concrete. We need more of those types of stories.
Participant 2 (Exp): Examples like that, if you could get them in front of developers,
particularly for bottom end developers who wanted to put up a square box, you would say
13
“Well, actually if you do it this way, you'll get a motivated buyer and you are making a
similar amount of money.’
LID’s adoption rate appears to be slow, arguably due to perceptions of low profit and high
cost (R1):
Participant 5 (D): You don’t want to cause your business to become less profitable because
you are the only maverick looking for LIDs. That doesn’t make any business sense. (......)
Sometimes costs are a factor, sometimes they are not. Sometimes a more sensible
environmentally friendly solution actually costs you a lot less and is easier to perform. (......)
The relative cost comparisons for LID against standard designs is not an exploratory exercise
that gets any money for a consultant. There is no commercial advantage, therefore why do it.
Participant 8 (Exp): To develop a section of land in a very conventional way with no view on
retaining existing features or stormwater quality or extensive earthworks, your typical
subdivision would be a lot cheaper than an environmentally friendly development. (......) It
comes down to what is the cost of environmentally friendly and what is it that we can afford. I
think that Auckland and New Zealand is going through that battle as our land and house
prices skyrocket up. What is affordable to the general public to buy? This has to be balanced
with what limits of environmental protection we are willing to put up with.
Participant 2 (Exp): There is certainly a real cost limitation to the whole package whether you
use LID or some other method. Doing it properly for the Auckland region is all terribly
expensive. That’s why we get into running battles with some of the city council people.
Because they are trying to run budgets and work out what they can get.
Participant 22 (Exp): If you ask people whether they want to pay an extra $1000 on top of
your rates for stormwater they may say no, unless they are being flooded.
Hence, regulatory support and/or incentives for developers and/or consultants will be
necessary in order to increase adoption of LID. Another reason for slow uptake is a lack of
institutional support:
Participant 2 (Exp): The TP10 [LID guideline] is four years old. It’s probably taken 3 of those
years for us to realise that people who read and even thought they wanted to implement those
ideas on the local scale, we had to watch while they go to the council get frustrated and go
back to a standard design with a stormwater treatment pond at the bottom.
Participant 5 (D): The only difficulty I have with the philosophical aspect of LID is that if the
support network is not there from council or from the market place, intentional or
unintentional is irrelevant. If you are trying to look for low impact solutions for no benefit to
a business it just doesn’t make any sense.
More critically, as another engineering strategy that relies mainly on devices (ponds, swales,
etc.) LID does not raise the public awareness required for voluntary and widespread uptake.
Public behaviour will still be guided by the device paradigm, i.e. the understanding that
problems can be solved by installing devices (Higgs 2003). The widespread disconnection
with the natural environment will still be in place and hence, behaviour change will not occur.
14
Participant 17 (Res): (...) all those new funky technologies like swales, those Hynds sand filter
things. And these are unnatural, there is absolutely nothing natural about them.
Participant 16 (Res): The greatest barrier I think is that people just don’t care enough. It
comes right down to not prioritising it sufficiently. Not seeing it as a real problem. Again, it’s
a cultural thing. It’s like we are safe, we don’t have to worry about the natural world.
In some instances, LID may allow for more urbanisation to occur due to the off-set of
stormwater impacts on the existing stormwater network. For example, in the upper parts of
Project Twin Streams catchment, development is allowed if it maintains hydrological
neutrality, i.e. the increased imperviousness does not result in an increase in stormwater run-
off. The LID paradigm fits in with continued economic growth through increasing
urbanisation. This suggests a similar infrastructure ‘trap’ as was identified in the
‘conventional fixes’ perspective.
‘Community Development’ Perspective
Community engagement projects like PTS aim to strengthen communities by changing
behavior and reconnecting residents with their local environment.
Participant 31 (Exp): The reason for the community to be involved so much is to make the
project most effective really. Without the community being there, the contractors would be
employed to do it. It would get done. It would get done possibly at the same cost, possibly for
less. But once the project was finished it would have the potential to slowly go down. Because
there would be no buy-in from community about changing the value of how they perceive
streams.
Participant 16 (Res): People come; | think it’s a combination of love of the native plants and
the stream and wanting to be part of this neighbourhood. It’s like a sense of identity. (......)
It’s a love of the natural world and it’s a sense of neighbourhood, a sense of coming together
and doing something permanent right there. (......) When people are down there working like
that it’s calm, peaceful, away from the busyness. People start to talk about things. So it’s
there is human development and a human growth - becoming part of something, of a bigger
world.
A focus on behaviour change is anticipated to have positive flow-on effects. Prospectively, it
will not only increase the communities’ willingness to reduce their impact on the streams and
take care of their streams, it will also allow people to re-evaluate their consumption patterns
and decisions regarding mode of travel and energy consumption from which economic
benefits may result. Components and linkages of the community development CLD shown in
Figure 4 promote manifold and intentional reinforcing feedback processes.
The ‘community development’ CLD also describes a diffusion process, in this case the
diffusion of societal norms and behavioural change. Again, government support was
perceived as a key to starting community projects. These projects are the vehicles for
achieving community buy-in which in a favourable political climate would feed back to more
government support (R1). There is a danger that a lack of community buy-in would result in a
45
lack of governmental support. Hence, this relationship needs to be carefully managed.
Community buy-in was a key variable in this CLD with many ingoing and outgoing links.
Community buy-in increases project involvement, in this case stream restoration, with positive
effects on vegetation (R2) and people’s awareness. Awareness is a necessary condition for
behaviour change which will reduce pollution and may increase uptake of LID. Behaviour
change was also perceived to lead to economic benefits which in turn feeds back to community
buy-in and socio-economic status (R3).
PTS has been successful in the past years and this has enabled follow-on projects to be
initiated (R4). These projects use the same model of community engagement and
development to initiate behaviour change in other areas, notably waste reduction, travel mode
and energy consumption (Waitakere City Council 2007). Uptake of these projects has also
been strong (Trotman 2006). Economic benefits at the household as well as community level
are envisaged, further strengthening communities and increasing their socio-economic status.
This in turn will increase the support base for these projects.
sustainable social-
ecological system -
a xo ~ ,
ecosystem rennet
“ a
nn ca publichealth ”
and safety na f
4
a ,
(CV geomorphology
RA . \ ,
“i run-off volume/ of
Nowe word:
vr
ee v4
mn ity
ur & :
imperviousness , fis
TS ‘ Ke -
raf
4
\. 4
UD uptake | involvement p
v
Figure 4: The ‘community development’ CLD
Overcoming initial inertia until a sufficient support base and uptake has been established takes
time (Svendsen & Campbell 2008). Consequently, time delays and their effects are crucial to
this perspective and community engagement in general. This stands contrary to existing and
rapidly progressing environmental degradation.
Participant 28 (LG): It’s going to take another 30 years before people turn around respecting
the streams, the evolution and life in the streams.
16
Participant 16 (Res): I mean people litter and most people don’t really turn their heads. It’s
not culturally unacceptable yet. (...... ) When this area became inhabited by Europeans,
people didn’t really seem to respect these streams, because we find an enormous amount of
junk down there. It’s not been like that when it was Maori living here it was very different,
careful.
Participant 27 (LG): [It] takes away the interaction between people and the natural
environment when you look at something and say “yuk”. This becomes a bit of a cycle, ifa
stream is a rubbish dump people will dump more in there, because they can see it’s a
dumping ground.
The ‘community development’ perspective was shared by half of the participants. While most
of these participants did not come from an engineering background, all had an understanding
of conventional and low impact stormwater solutions. Participants with an engineering
background, on the other hand, showed a lack of understanding for the need of community
development:
Participant 2 (Exp): As I understand it [PTS] at the moment it’s almost entirely riparian
planting. Whether it’s effective or not depends on what else happens in the catchment. It was
always the wrong way around for us. To my mind it is the icing on the cake when you already
taken out the contaminants and sorted out fish passage, and your water quantity problems.
And then you got some streams where you might have a good habitat and then of course with
riparian planting you can get that. So I tend to see it as a cart before the horse. (...) But I do
think it will be good in 20 years time, but by goodness me it’s expensive. It’s very, very
expensive.
Mental Models in Stormwater Management
As part of the interviews participants were asked to identify mindsets (mental models) that are
associated with stormwater management solutions. These are presented in Table 3 as either
supporting or detracting from implementation. Mental models - our perceptions of reality - are
the most powerful underlying driver of our actions. Effective behaviour change requires an
uncovering and engagement with these underlying mindsets (Sterman 2000).
Most participants had a very good understanding of stormwater management strategies and
solutions. However, there was little agreement on what solutions are preferential to address
problems. Direct statements included:
Participant 20 (RE): The system is not in a position where it can really help itself at this stage
because it’s such an impacted system to start with. So there’s got to be a regulatory
environment that says what is and what isn’t permissible.
Participant 26 (LG): The overriding goal of PTS is to engage local residents in behaviour
change and that is the most effective way of good stormwater management.
Participant 17 (Res): So it’s process as well, it’s not just about the streams; it’s about the
people who manage the streams. Until there is a forum or a project team where there is a
17
reciprocal respect for the Maori culture and dimension to that, and the technical expertise,
there isn’t an effective management of that stream as far as I am concerned.
Supporting Mindsets
Preventing Mindsets
Conventional
Fixes
Fear of uncertainty, of losing control
Uncertainty about how to deal with new ways of
Cost avoidance
Changing perceptions
thinking
Control nature and people
Acceptance of creeping deterioration
Human predominance and superiority
Low importance of stormwater compared to
wastewater and water supply
Right to develop and make own decision for private
land
Stormwater is a nuisance rather than a resource to
be used
Disconnection from environment in general
Want council to take care of everything rather than
share responsibility
Low Impact Acknowledging the value of water and waterways Frustration with bureaucracy
Solutions Outcomes encourage uptake Cost/Profit perception
Fear of uncertainty, of losing control Disregard of maintenance
Human predominance and superiority Stormwater is not interesting politically
Engineering perspective with ecological objective Maximisation of the development area
Profit maximisation
Stuck in old ways, hard to change
Community Caring attitude Acceptance of environmentally
Development | Changing perceptions
Enjoy green areas and get upset by damage
Desire to learn more council
Pride in achieving positive impact Stuck in old ways, hard to change
damaging behaviour
Perception/suspicion of policy and
Table 3: Supporting and preventing mindsets
Participants also expressed uncertainty as to how much management is required.
Participant 11 (Exp): There is a lack of knowledge about the natural system, about how much
we should manage the natural system. And how much we should let the system be natural, we
haven’t made that decision yet.
Participant 19 (RE): In a way we think we go and do a few actions and [the stream] then takes
care of itself. We are not really seeing that.
Conflict between the three different perspectives was apparent. This conflict may indicate
uncertainty and possibly a lack of a common vision and well-defined objectives. Achieving
agreement on the most effective solutions and measures of involvement requires careful
collaboration, communication and deliberation strategies.
Analysis of the Differences in Perspectives
Results based on the analysis of participants’ cognitive maps show that perspectives in
stormwater management are diverse and potentially conflicting. Stakeholders conceive
different solutions for stormwater management which are driven by substantially different
mindsets. The main differences between perspectives are summarized in Table 3.
18
Conventional Fixes
Perspective
Low Impact Solutions
Perspective
Community Development
Perspective
Priorities of
management goals
Public health and safety,
safeguard built structures
Public health and safety,
safeguard built structures,
Provide ecosystem function
Behavior change,
Community buy-in and
development
Solution drivers
Capacity, reticulation
infrastructure
Permeable surfaces,
maintenance
Behavior change, public
ownership of local
waterways and their
management
Spatial focus
Focus on the stream as
well as all drainage
connections to the
streams, city scale
Off-stream, lot-scale with a
catchment wide integration
Local stream environment,
neighborhood scale
Timescale of Short term Short term Long term
implementation
Feedback Minimal, linear thinking, Minimal predominantly Diverse and manifold, long
short linear thinking, short-
medium as part of the
diffusion process
Inclusivity of Scientific and Scientific and engineering, Public (to a large extent),
stakeholders
engineering, government
officials, public (minimal)
government officials, public
(to a small extent)
government officials
Use of available
knowledge
Scientific knowledge
(Hydrology, Civil
Engineering, Public
Health)
Scientific knowledge
(Hydrology, Geomorphology,
Ecology, Public Health)
Local and community
knowledge, social
marketing, social
development
Ownership of the
problem
Local authority
Local authority
Public (to a large extent)
and local authority in a
supporting role
Responsibility for
implementing
solutions
Regional and local
authority
Local authority and public
particularly with respect to
maintenance
Public
Main shortcoming
of this perspective
Disregard for stream
ecosystem
Lack of behavior change and
public ownership
Substantive delays and
long-term implementation
Table 3: Overview of the main differences between the three perspectives in stormwater management
While balancing feedbacks are clearly present, we have concentrated our analysis on self-
reinforcing feedback that serves to increase and perpetuate implementation. Most participants
were not familiar with the concept of feedback and could not identify feedback situations on
their cognitive map. This reflects the prevalence of linear thinking processes (Sterman 2000).
Feedback in the conventional fixes perspective is nearly absent. The implementation of
solutions is triggered by localised problem events. For example, a pollution event such as a
spill will trigger a clean-up, and for recurring or non-point sources possibly the installation of
a treatment device. Similarly, the low impact perspective shows few and predominantly short
feedback cycles, even though the diffusion process for LID uptake follows a longer feedback
cycle. Short feedback occurs when variables influence each other directly. For example in
Figure 2, a localised flood event may lead to building of a pipe or a concrete channel, which
in turn will prevent flooding to occur at this place in the catchment (but will increase
possibilities for flooding downstream). The ‘community development’ perspective comprises
a multitude of short and long feedback cycles. As such, the community development
perspective starts to transition from our existing anthropocentric worldview by purposefully
targeting and incorporating feedback processes.
19
Each perspective has shortcomings in providing solutions that can effectively address the
challenges posed by stormwater. The conventional fixes perspective lacks an integration of
ecosystem health in the management goals, and it can be argued that conventional solutions
are inherently conflicting with environmental restoration goals. The main shortcoming of the
low impact design perspective is the lack of public ownership of problems and solutions. In
essence, low impact design provides a similar set of engineering solutions which may be more
localised and environmentally friendly, but still maintain the status quo of a public that is
inherently disconnected with their local environment. This disconnection causes a lack of
understanding, a lack of care and environmentally damaging behaviour. It can be expected
that with future population growth and existing development patterns, water quality problems
in streams will continue to deteriorate. This will further increase the lack of respect of this
resource to the point where they cannot be offset by low impact solutions:
Participant 11 (Exp): There is the issue of impaired aesthetics of the streams from gross
pollutants. There is a lack of ownership and the gross pollutants and littering causes a lack of
respect in the stream. So, if 1 see lack of respect it enables me to have lack of respect. If I see
that things are respected it enables me to respect things.
The main drawbacks of the community development perspective are the long delays and
resulting long-term implementation of solutions. Community development postulates social
learning, behaviour change and the creation of public ownership, and as such has, in theory,
manifold flow-on effects that can engender uptake of more sustainable technologies and
strategies, e.g. co-housing, use of public transportation, reduced energy consumption.
However, environmental degradation is a pressing concern and existing stormwater problems
have to be addressed quickly and effectively while at the same time steps ensuring the
restoration of impacted ecosystems. This however, is not identified as a priority in the
community development perspective. As a result, none of the perspectives in themselves will
lead to sustainable stormwater management outcomes.
6 Integration of Efforts: Transition Culture
The convergence of pervasive challenges, including environmental degradation and social
discontent with prevalent political processes and outcomes, has brought us to a point where
we are increasingly challenged to act. Thus, a concerted and integrated effort is required to
address the multiple and pressing problems experienced in Project Twin Streams catchment.
We need to enter a culture of transition among all facets of society. In order to realise a
sustainable future, the long term focus must emphasize social learning, behaviour change, and
the creation of effective partnerships with local authorities with a focus on increasing
community resilience and capabilities for adaptation. Institutional learning and change must
go in hand and support this long-term focus on behaviour change and community
development. Low impact solutions can be implemented in the short term in new
developments while retrofitting of old developments can occur over time as old infrastructure
is phased out and more sustainable building practices are implemented. Figure 5 synthesises
our argument by integrating the low impact development and community development
20
In addition, the deliberate use of regulatory controls should guide
implementation of low impact development solutions and community development.
perspectives.
auom yo Aaenb
aanonaaseajul pue
Auadoud 01 a3ewep
\.
Aaayes pue
uueay oygnd
Suyuueyd wawyores paresSaiu)
suayjng aneya8an pue
saoeyins ajqeauuad 31e319
anjenreugey ~
| ~\
Aysranpo1g
Saoynsas
aquaytsad /
wiaasAs 12130}033
~ JeIp0s ajqeujersns
different levels and timescales.
21
Figure 5: Integrating community engagement and LID serves multiple objectives and creates change at
This transition culture incorporates respect for diverse perspectives, personal responsibility
and co-management. It requires understanding and acceptance of the need to live within the
limits imposed by nature and each other by contributing to the common good. Genuine well-
being and security arise from connections to people and place. The overarching goals of this
culture must be the creation of sustainable mechanisms that can internalize anthropogenic
environmental impacts.
This transition is hindered by the existence of manifold barriers which may be classified as
social, institutional, educational, attitudinal, logistical, communication and internal barriers.
These barriers can be understood as an indication or manifestation of underlying reductionist
and risk-averse mental models among government officials and the public. We have
purposefully steered away from a discussion barriers of implementation as this warrants a
more in depth analysis. However, it must be clear that the existence of diverse perspectives on
stormwater management is a contributing factor in our current failure to live and use our land
without eroding its capacity for sustained provision of environmental services.
Understanding stormwater management as a complex social-ecological system is the pre-
requisite for developing and implementing integrated solutions. Acknowledging and
exploring the different perspectives on problems and solutions in stormwater management has
far reaching implications for the relationships we build and the processes we put in place to
manage stormwater. Under this new paradigm, stormwater is not considered to be purely a
technical issue that requires a technological fix to deal with. Rather, there is a need to
acknowledge that stormwater is multi-faceted: different stakeholders have different priorities
for dealing with stormwater quantity, quality and erosion problems, and opinions on effective
solutions range as far and wide so as to render the promotion of a single solution strategy such
as low impact development ineffective. Social learning, stakeholder buy-in and ownership of
any solution are paramount for successful implementation. Establishing buy-in calls for the
development of new methods and processes focused on engagement and relationship building.
Institutional structures need the capacity to adapt to this new focus, while our scientific
research endeavors are challenged by different data needs and analytical practices. This
presents considerable challenges for management and institutional structures, engagement
with stakeholders, socio-cultural considerations that promote scientific/engineering agendas.
Recent moves towards LID which are now advocated by public institutions across the world,
emphasize concems for a technical focus, working from a premise that notionally healthy
‘solutions’ can be imposed through revised engineering applications. Implicit to these
applications is the assumption that an engineered solution exists and that technology can be
used to provide the answers. The premise is appealing to authorities given their desire to
invest in measurable outcomes. As such, there has been a proliferation of design solutions that
apply engineered infrastructure. Ultimately, however, it must be asked whether these
measures address the key issues in stormwater management and are appropriate in the context
of sustainability. Many pressing contemporary issues, including environmental degradation
from urban stormwater run-off, can be traced back to a lack of understanding and adverse
behavioral choices. A genuine commitment to sustainability requires that effective solutions
extend beyond the device paradigm and place emphasis on socio-cultural transition.
22
At present, there is no wide-spread acceptance and ownership of low impact solutions and no
understanding of what will be required to promote behavior change. There are inherent time
delays in the process of behavior change. Given the scale and pace of change at which
problems occur, windows for opportunities to address pollution problems are fast closing.
There is an urgent need to develop and implement community oriented approaches to
stormwater management that meet human needs while reducing harmful impacts and
repairing stream environments.
7 Conclusions
Our perspectives of environmental issues frame the way we develop and implement solutions
and actions to manage our natural environment. The most dominant perspectives in
stormwater management have been based on a linear and technocratic worldview that often
ignored perspectives of other stakeholders. In order to learn from past mistakes,
understanding and engagement with diverse perspectives are required to develop more
integrated and sustainable stormwater management practices.
Cognitive mapping and systems thinking were used successfully in this research to elucidate
mental models and perspectives on stormwater management in Project Twin Streams
catchment. Results show that problems and solutions regarding stormwater are conceptualized
in substantially different ways. This provides evidence that there are multiple ways of
knowing and that differing values are associated with stormwater management. The three
different perspectives that were uncovered have profound influence on management goals,
time scales and solutions for stormwater management. Critically, results show that none of the
perspectives in themselves will lead to sustainable stormwater management outcomes.
Opportunities for future research include the application of cognitive mapping other areas to
create a multiple case study which may provide a general theory. Furthermore, insights can be
modeled quantitatively in a stock-and-flow or agent-based model in order to test some of the
assumptions drawn from the CLD. Another interesting project would be a group modeling
session which outcomes could be tested against this work. This could provide an
understanding of the power relationships between stakeholder groups.
Our recommendations first and foremost are to create processes that serve to understand
different perspectives of stakeholder groups as they pertain to council related business. This
understanding is most effectively created in joint participatory action research projects.
Second, we propose to better understand barriers of implementation of LID and community
development projects, and to work to remove some of those barriers with the aim of better
integrating different approaches. Community development projects rely on long-term funding
and at present Project Twin Streams future is uncertain with funding streams to cease in 2012.
While market-based instruments are a possible option for funding LID, their increased
compliance and administrative costs as well as the total unfamiliarity with these approaches in
city councils make their use unlikely. Development contributions are currently widely
implemented to offset some of the cost associated with infrastructure provision. Clearly,
23
providing incentives for LID might go a long way towards ensuring uptake, but numerous
institutional barriers to implementation will have to be addressed first.
Understanding the diversity of perspectives and their wider impacts on research and
management is only truly useful if it adds to mutual acknowledgement and respect of the
different positions among stakeholders. This is a necessary step in reducing existing conflict
and creating a common vision for a sustainable future. Different perspectives may lead to
different solutions, but there are opportunities for these to complement each other. A
transition towards this integration is crucial if we are to address present challenges in
stormwater management and create sustainable solutions.
8 Acknowledgements
I would like to thank all research participants for their time, interest and encouragement.
Helen Haslam, Clare Feeney and three anonymous reviewers thank you for helpful comments.
References
Alexander GG and Allan JD (2007) Ecological Success in Stream Restoration: Case Studies
from the Midwestem United States. Environmental Management, 40, 245-255.
Amold CL and Gibbons CJ (1996) Impervious Surface Coverage: The Emergence of a Key
Environmental Indicator. Journal of the American Planning Association, 64, 243-258.
Auckland Regional Council (2000) Low Impact Design Manual for the Auckland Region.
Auckland, New Zealand.
Auckland Regional Council (2004a) Auckland Regional Stormwater Project - an Action Plan
to Deliver Improved Stormwater Outcomes - Final Report. Report prepared by the Boston
Consulting Group, Auckland Regional Council, New Zealand.
Auckland Regional Council (2004b) Stormwater Treatment Devices: Design Guideline
Manual. Auckland, New Zealand.
Berkes F (2004) Rethinking Community-Based Conservation. Conservation Biology, 18, 621-
630.
Bernhardt ES and Palmer MA (2007) Restoring Streams in an Urbanizing World. Freshwater
Biology, 52, 738 - 751.
Biswas AK (2006) Water Management for Major Urban Centres. International Journal of
Water Resources Development, 22, 183-197.
Brown RR (2005) Impediments to Integrated Urban Stormwater Management: The Need for
Institutional Reform. Environmental Management, 36, 455-468.
24
Bunn SE and Arthington AH (2002) Basic Principles and Ecological Consequences of Altered
Flow Regimes for A quatic Biodiversity. Environmental Management, 30, 492-507.
Butler CD and Oluoch-Kosura W (2006) Linking Future Ecosystem Services and Future
Human Well-Being. Ecology and Society, 11, 30.
Choi YD (2004) Theories for Ecological Restoration in Changing Environment: Towards
"Futuristic" Restoration. Ecological Research, 19, 75-81.
Choi YD (2007) Restoration Ecology to the Future: A Call for New Paradigm. Restoration
Ecology, 15, 351-353.
Eden C (2004) Analyzing Cognitive Maps to Help Structure Issues or Problems. European
Journal of Operational Research, 159, 673-686.
Flynn J, Beecham S, Shaver E, Kennedy P and Hodges P (2009) Panel Discussion. NZAWWA
Stormwater Conference, Auckland, New Zealand.
Flyvbjerg B (2006) Five Misunderstandings A bout Case-Study Research. Qualitative Inquiry,
12, 219-245.
Folke C (2007) Social-Ecological Systems and Adaptive Govemance of the Commons.
Ecological Research, 22, 14-15.
Graf WL (1977) Network Characteristics in Suburbanizing Streams. Water Resources
Research, 13, 459-463.
Grinde J and Khare A (2008) The Ant, the Grasshoper or Schroedinger's Cat: An Exploration
of Concepts of Sustainability. J ournal of Environmental Assessment Policy and Management,
10, 115 - 141.
Gross M (2002) New Natures and Old Science: Hands-on Practice and Academic Research in
Ecological Restoration. Science Studies, 15, 17.
Hatt BE, Fletcher TD, Walsh CJ and Taylor SL (2004) The Influence of Urban Density and
Drainage Infrastructure on the Concentrations and Loads of Pollutants in Small Streams.
Environmental Management, 34, 112-124.
Hauraki Gulf Forum (2008) Tikapa Moana - Hauraki Gulf State of the Environment Report.
Higgs E (2003) Nature by Design: People, Natural Process and Ecological Restoration. MIT
Press, Cambridge, Massachusetts.
Higgs ES (1997) What Is Good Ecological Restoration? Conservation Biology, 11, 338-348.
Hobbs RJ (2007) Setting Effective and Realistic Restoration Goals: Key Directions for
Research. Restoration Ecology, 15, 354-357.
Kahneman D and Tversky A (1979) Prospect Theory: An Analysis of Decision under Risk.
Econometrica, 47, 263?291.
25
Kathuria V (2006) Controlling Water Pollution in Developing and Transition Countries -
Lessons from Three Successful Cases. Journal of Environmental Management, 78, 405-426.
Kellert SR, Mehta JN, Ebbin SA and Lichenfeld LL (2000) Community Natural Resource
Management: Promise, Rhetoric and Reality. Society & Natural Resources, 13, 705-715.
Laituri M (1996) Cross-Cultural Dynamics in the Ecocity: Waitakere City, New Zealand.
Cities and the Environment, 13, 329-337.
Lane DC (2008) The Emergence and Use of Diagramming in System Dynamics: A Critical
Account. Systems Research and Behavioral Science, 25, 3 - 23.
Middleton JV (2001) The Stream Doctor Project: Community-Driven Stream Restoration.
BioScience, 51, 293-296.
Miles MB and Huberman AM (1994) Qualitative Data Analysis. Sage, Thousand Oaks, CA.
Naveh Z (2005) Epilogue: Toward a Transdisciplinary Science of Ecological and Cultural
Landscape Restoration. Restoration Ecology, 13, 228-234.
Pahl-Wostl C (2006) The Importance of Social Learning in Restoring the Multifunctionality
of Rivers and Floodplains. Ecology and Society, 11, 10.
Pahl-Wostl C (2007) The Implications of Complexity for Integrated Resources Management.
Environmental Modelling & Software, 22, 561-569.
Palmer MA, Bernhardt ES, Allan JD, Lake PS, Alexander G, Brooks S, Carr J, Clayton S,
Dahm CN, Follstad Shah J, Galat DL, Loss GS, Goodwin P, Hart DD, Hassett B, Jenskinson
R, Kondolf GM, Lave R, Meyer JL, O'Donnell TKOD, Pagano L and Sudduth E (2005)
Standards for Ecologically Successful River Restoration. Journal of Applied Ecology, 42,
208-217.
Paul MJ and Meyer JL (2001) Streams in the Urban Landscape. Annual Review of Ecology
and Systematics, 32, 333-365.
Peters NE and Meybeck M (2000) Water Quality Degradation Effects on Freshwater
Availability: Impacts of Human Activities. Water International, 25, 185-193.
Purcell AH, Friedrich C and Resh VH (2002) An Assessment of a Small Urban Stream
Restoration Project in Northern California. Restoration Ecology, 10, 685-694.
Ravetz JR (2006) Post-Normal Science and the Complexity of Transitions Towards
Sustainability. Ecological Complexity, 3, 275-284.
Reid HE, Gregory CE and Brierley GJ (2008) Measures of Physical Heterogeneity in
Appraisal of Geomorphic River Condition for Urban Streams: Twin Streams Catchment,
Auckland New Zealand. Physical Geography, 29, 247-274.
Rogers EM (2003) Diffusion of Innovations, 5th ed edn. Free Press, NY, USA.
26
Rosenberg S and Margerum RD (2008) Landowner Motivations for Watershed Restoration:
Lessons from Five Watersheds. Journal of Environmental Planning and Management, 51,
477 - 496.
Roy AH, Wenger SJ, Fletcher TD, Walsh CJ, Ladson AR, Shuster WD, Thurston HW and
Brown RR (2008) Impediments and Solutions to Sustainable, Watershed-Scale Urban
Stormwater Management: Lessons from Australia and the United States. Environmental
Management, 42, 344-359.
Rumps JM, Katz SL, Bamas K, Morehead MD, Jenkinson R, Clayton SR and Goodwin P
(2007) Stream Restoration in the Pacific Northwest: Analysis of Interviews with Project
Managers. Restoration Ecology, 15, 506-515.
Silveira AL (2002) Problems of Modem Urban Drainage in Developing Countries. Water
Science & Technology, 45, 31-40.
Sterman JD (2000) Business Dynamics: Systems Thinking and Modelling for a Complex
World. McGraw-Hill, Boston, MA.
Svendsen E and Campbell LK (2008) Urban Ecological Stewardship: Understanding the
Structure, Function and Network of Community-Based Urban Land Management. Cities and
the Environment, 1, 4.
Trotman R (2006) Sustainable Communities - Sustainable Catchment (Sc2), Stage Two
Evaluation, (July 2005 - June 2006). Sustainable Communities, Department of Intemal
Affairs, Auckland Regional Council.
Tumer RE (2005) On the Cusp of Restoration: Science and Society. Restoration Ecology, 13,
165-173.
Waitakere City Council (2007) Project Twin Streams Strategic Plan 2007. Waitakere City,
New Zealand.
Waitakere City Council (2008) Project Twin Streams Strategic Plan Summary. Waitakere
City.
Wolman MG (1967) A Cycle of Sedimentation and Erosion in Urban River Channels.
Geografiska Annaler, 49A, 385-395.
27
