Many urban systems worldwide face increasing complexity and uncertainty in water governance. As a result, integrated urban water management (IUWM) has become increasingly common. The IUWM paradigm conceptualizes water supply, conveyance, and treatment as an integrated system to improve water management efficiency and sustainability. This approach often presents new responsibilities that require coordinated efforts and leadership, but factors such as professional protocols, limited time, legal mandates, and budgetary constraints may limit joint implementation. To understand how policy stakeholders translate goals into action, we ask: How do sectoral affiliation of governance stakeholder organizations and leadership influence patterns of joint implementation of programs and policies over time? Using inferential social network analysis, this paper examines how coordination and leadership play a role in the implementation of IUWM. We find that the presence of leaders has a significant effect on joint implementation of integrated programs and policies over time. Counter to our expectations, however, organizations from the same sector (e.g., local government, non-profit, private, etc.) tend to implement IUWM policies and programs together.

  • Institutional setting may explain who leads and the extent of sectoral integration in IUWM implementation over time.

  • Non-IUWM implementation networks are comparatively stable and homogeneous, likely reflecting obligations in regular water and wastewater service provision.

Many urban centers worldwide are transitioning to integrated urban water management (IUWM) to deal with increasing complexity and uncertainty of pressures in water governance, and to promote social, economic, and ecological sustainability (Larsen et al., 2016; Hess & Brown, 2018). The IUWM paradigm views water supply, drainage, and wastewater conveyance as part of an integrated system (Fletcher et al., 2015). IUWM also emphasizes the need for new models of governance that promote coordination and collaboration among diverse water management stakeholders, and that move away from traditional, fragmented approaches to water management (Mukheibir et al., 2014).

Despite the importance of coordination and collaboration in IUWM, it is still largely unclear when and why governance stakeholders coordinate to implement integrated (or non-integrated) policies and programs. Examining coordination in policy implementation, or the carrying out of policy decisions (Pressman & Wildavsky, 1973; Mazmanian & Sabatier, 1983; O'Toole, 2000), is one way to address this gap. While water management literature suggests that diverse stakeholder participation and leadership are cornerstones of transitioning to IUWM (Brown et al., 2013), other work has indicated that factors such as professional protocols, limited time, budgetary constraints, and fear of losing constituent support could limit such action (Brown, 2005; Stoker et al., 2018).

This paper asks: How do sectoral affiliation of governance stakeholder organizations (e.g., local government, businesses, etc.) and leadership (which entails mobilizing others and their resources) influence patterns of joint implementation of programs and policies over time? Current research examines IUWM from a variety of angles, including barriers to implementation (Brown & Farrelly, 2009; O'Donnell et al., 2017) and the role of water stress, heightened public attention, leadership, and venues as catalysts for transitioning to more sustainable water management practices (Gliedt et al., 2018; Garcia et al., 2019). Other work presents more applied considerations including decision-support tools and cost-benefit analyses (Makropoulos et al., 2008; Fattahi & Fayyaz, 2010). Together, these studies offer valuable insight and strategies relating to sustainability transition processes. Yet, there is limited inferential analysis examining how leadership and sectoral heterogeneity affect patterns of joint policy implementation. Systematic examination of these factors can improve understanding of implementation dynamics across different transitioning urban water systems.

Leveraging data from 358 gray literature documents published online, we use exponential random graph models (ERGMs) to estimate the effects of leadership and sectoral affiliation on joint policy implementation for IUWM and non-IUWM water management practices in Tucson, Arizona, USA, in 2007 and 2017. Through this study, we build on current policy implementation research and sustainability transitions literature.

In what follows, we introduce the background for our study. Next, we discuss drivers of joint implementation in the governance networks and introduce our hypotheses. Following a discussion of data, methods, and model results, we descriptively evaluate the network structures. We conclude with the implications and limitations of our work, and considerations for future research.

This study focuses on joint implementation of water management programs and policies in Tucson, Arizona. Tucson is a growing urban center located in the semi-arid Sonoran Desert, which receives about 11 inches of rainfall per year (NOAA, 2019). Water security concerns in Tucson are salient at present because of current climate-related and regulatory conditions. Tucson – like most of Arizona – has endured a drought now for more than two decades (AMWUA, 2020). The drought has also impacted Lake Mead, which supplies a portion of Tucson's annual water resources. Over time, water supply has declined in the reservoir, nearing a level at which the Bureau of Reclamation would declare a shortage in the Colorado River Lower Basin (USDOI, 2007). Should this occur, Tucson would lose its annual allocation to California (Pitzer et al., 2007).

Tucson's water security concerns are rooted in a long history of rapid growth and extensive water consumption. Starting in the early 20th century, Arizona experienced rapid agricultural growth, and later urbanization – both which relied on groundwater resources (Larson et al., 2009; Megdal et al., 2014). Groundwater over-extraction soon became problematic, so Congress approved the construction of the Central Arizona Project (CAP), which would import water over 300 miles uphill to central Arizona, including Tucson. The intent of the CAP was to deliver new water resources to offset groundwater (August & Gammage, 2007). Nonetheless, many consumers instead used CAP to supplement groundwater use.

To address continued over-extraction, the state passed the Groundwater Management Act (GMA) in 1980, which would designate much of the Arizona – Tucson included – as active management areas (AMAs). The Arizona Department of Water Resources (ADWR) designed five consecutive management plans that would introduce increasingly rigorous conservation and management requirements for AMAs to decrease groundwater withdrawals over the next 45 years (ADWR, 2016a). Under these plans, the Tucson AMA must implement conservation measures to achieve ‘safe yield’ standards by 2025, where annual groundwater withdrawals will not exceed the amount recharged (ADWR, 2016a).

ADWR supplemented the management plans with the Assured Water Supply Program, requiring new commercial and residential subdivisions to either demonstrate the availability of 100-year water supply (Arizona State Senate, 2007), or for their AWS-designated, municipal providers (e.g., Tucson) to present a written commitment of service (ADWR, 2016a). Pursuant to the GMA, municipal providers with an AWS designation must also participate in the Total Gallons Per Capita per Day (GPCD) Program, which sets annual municipal use targets (ADWR, 2016b). In 1989, Tucson Water, the City's water department, failed to comply with GPCD standards, so ADWR required the utility to adopt a base conservation program to reduce groundwater consumption. Tucson Water has since introduced price- and incentive-based strategies, regulatory measures, and programs promoting public awareness and compliance, together superseding requirements (City of Tucson & Pima County, 2009). Arizona's experience with long-term drought and recent concerns about future shortages on the Lower Colorado River make these efforts especially timely (Castle & Fleck, 2019).

IUWM networks

Translating vision into action

A key strategy in Tucson's adaptation to long-term drought has entailed adoption and implementation of IUWM. This management paradigm conceptualizes traditional, siloed water management (i.e., supply, stormwater, and wastewater) as an integrated whole (Furlong et al., 2017). The goal of this approach is to improve water-use efficiency and to reduce reliance on potable resources (Mitchell, 2006). There is a wide array of means to implement IUWM, such as demand-based incentives or pricing structures (Sharvelle et al., 2017), passive infrastructure to strategically channel water flows, and technology to promote wastewater reuse (Sharma et al., 2019).

Nonetheless, a key challenge to IUWM is translating vision into action. Such barriers to implementation can be multifaceted. Capacity to carry out a new program, for example, may be limited due to budgetary constraints, lack of expertise with new technology, or unclear responsibilities, or for fear of losing constituent support (Brown, 2005; Stoker et al., 2018).

Coordinated IUWM implementation

One way to understand barriers to IUWM implementation is to examine this in the context of the policy process. According to the Advocacy Coalition Framework (ACF), a policy issue relating to a specific geographic area, or a ‘policy subsystem’, tends to remain fairly stable over time (Sabatier & Jenkins-Smith, 1993). Policy stakeholders within the subsystem are hypothesized to engage in non-trivial degrees of coordination to translate policy goals into action (Sabatier & Weible, 2007). While this coordination is often examined in the agenda-setting stage of the policy process, this can also occur in policy implementation.

Over the past several decades, the policy literature has examined implementation from several theoretical points of departure (DeLeon & DeLeon, 2002). Areas of focus have included (i) why implementation fails following adoption (Pressman & Wildavsky, 1973); (ii) whether it is best to conceptualize variation in implementation through a top-down (i.e., mandated processes) or bottom-up lens (i.e., focusing on actors at the operational level) (Hjern & Hull, 1982; Mazmanian & Sabatier, 1983); and (iii) how to reconcile theoretical gaps of earlier generations of study and employ rigor in scientific research design (Sabatier, 1986; Goggin, 1990; Winter, 2012; Saetren, 2014).

More recent work has shifted focus away from unilateral, hierarchical structures, to governance networks to carry out policy objectives (Nilsen et al., 2013). Within these structures, autonomous stakeholders interact to share financial resources, as well as unique ideas, knowledge, and experience to identify solutions that address outstanding problems (Hartley et al., 2013; Torfing, 2016). Nonetheless, this may include some top-down practices if joint implementation occurs within broader governmental regimes or involves public actors linked to institutional requirements (Koontz et al., 2004; Koontz & Newig, 2014).

Assuming policy stakeholders are not required to coordinate in the implementation of policy (Chisholm, 1992), they may form strategic relationships with others who have complementary resources or capacities, such as unique information, influence, or financial resources (Pfeffer & Salancik, 2003). This resource dependence is a common theme among policy stakeholders that depend on one another to achieve shared management and policy goals (Milward & Provan, 1995; van de Meene et al., 2011; Henry & Vollan, 2014).

Research on the implementation of IUWM points to the importance of coordination between organizations from different sectors to access novel resources and other contributions. For example, it is shown that through these types of relationships, public actors can provide funding to those with expertise to carry out public goals (Mitchell, 2006). In other instances, organizations might rely on one another for unique capacities, such as legal authority, flexibility outside of institutional constraints, or diverse, specialized knowledge or skills (Henry, 2009, 2017; van de Meene et al., 2011).

As a municipal water service provider, Tucson Water is required by ADWR regulations to implement local initiatives for efficient water use, conservation, and use of renewable resources (ADWR, 1999). While the utility is not functionally interdependent with other organizations, its implementation of IUWM has involved a broad range of diverse stakeholders. A key strategy that Tucson Water has leveraged to promote water-use efficiency is the implementation of green infrastructure technologies, which refers to an interconnected system of greenspace (e.g., waterways, wetlands, greenways, natural areas, green streets, swales, etc.), that is designed for sustaining social and ecological health and well-being (Benedict & McMahon, 2012). Green infrastructure, a stormwater management strategy central to IUWM (Brumley et al., 2018), is depicted in empirical research as utilizing cross-sectoral interaction for different dimensions of implementation, such as planning, siting (Meerow & Newell, 2017), and leveraging diverse expertise or skills (Shifflett et al., 2019). In practice, such coordination may be useful for the implementation of a variety of strategies, such as regulation (Fricano & Grass, 2014), market-based approaches (Gómez & Teixeira, 2017), or educational programs (Newell et al., 2013). With new policies and programs, we expect increased operational complexities and needs, and utility reliance on organizations from other sectors.

H1a: In the implementation of IUWM programs and policies, organizations from different sectoral affiliations will tend to work together over time.

Leadership in coordinated IUWM implementation

The policy literature also points to skilled leadership as having a central role in coordination. Such leaders, which the ACF refers to as ‘policy entrepreneurs’, are described as investing time and energy to promote new relationships and mobilize other stakeholders (Sabatier & Weible, 2007). Being centrally embedded in coordination networks is suggested to afford these actors strategic capabilities to access novel information and finances, and to identify opportunities for those in the coordination network to achieve collective goals (Sabatier & Weible, 2007; Christopoulos & Ingold, 2015; Grigg et al., 2018).

Although policy entrepreneurs often receive attention in the policy literature in the agenda-setting process, they may also be instrumental through the implementation phase to ensure the success of a policy innovation (Mintrom & Norman, 2009; Arnold, 2015). In the context of the water management literature, leaders are depicted as necessary for establishing priorities and for visioning after policy adoption (Brown, 2005; Bos & Brown, 2012). They may also support others in implementation by providing needed information, through consultation, or in leading by example (van de Meene et al., 2011).

Empirical research on IUWM has shown that leaders are integral to implementation processes. Leaders are described as persistent and committed, adept in strategic networking, and engaged in directing and motivating others to implement sustainable water management innovations (Taylor, 2009). Prospects for the success of integrated management approaches are also credited to the presence of leadership (Grigg et al., 2018). With this central role of leaders in sustainability transitions – and IUWM implementation more specifically – we propose:

H1b: Leaders will have a clear role over time in driving patterns of joint implementation of IUWM policies and programs.

Non-IUWM networks

Coordination in non-IUWM networks

Policy subsystems that adopt and implement IUWM may also continue traditional strategies for treatment, water supply, wastewater conveyance, and stormwater management (Daigger & Crawford, 2007; Wong & Brown, 2009). Traditional water management systems, unlike IUWM, use large-scale, mechanized approaches and well-established institutional structures such as administrative processes and rules (Brown et al., 2009; Farrelly & Brown, 2011).

Within institutional frameworks, administrators and bureaucrats hold formal responsibility over service provision, with clearly designated responsibilities (Dobbie et al., 2016). One way that government actors may maximize efficiency and responsiveness in implementation is through coordination with one another (Provan & Milward, 2001). Nonetheless, assuming these actors have comparatively narrow professional incentives and responsibilities, they may be more inclined to minimize risk, as well as the potential for liabilities and loss of public support (Farrelly & Brown, 2011; Feiock, 2013; Mullin & Rubado, 2017). Given the predefined roles of responsible actors and the stability of their behavior, we expect that the non-IUWM implementation network might manifest as a relatively stable, homogeneous network over time.

H2a: In implementation of non-IUWM programs and policies, organizations with the same sectoral affiliation tend to work together.

The role of leadership in non-IUWM networks

Compared to an IUWM network, provision of essential public services in traditional water management likely requires less entrepreneurial leadership. Although water professionals may be adept in promoting innovations that address complex problems, routine operations and reliability may remain top priorities (Termeer, 2009). The ACF proposes that, within the policy subsystem, actors seeking to maintain the status quo participate in stable coordination networks over time (Sabatier & Jenkins-Smith, 1993). In the absence of a shared goal to advance policy innovation, entrepreneurial leaders may not be needed to support these groups of actors.

H2b: Leadership will not have a significant effect in driving patterns of joint implementation in traditional water management.

Data

We derived network data from gray literature content to test our hypotheses. Here, gray literature refers to literature resources including news articles, websites, government reports, ordinances, meeting minutes, manuals, presentations, and white papers. These data were most appropriate for the longitudinal nature of our study. Key challenges that come with collecting longitudinal data include high costs and limited availability (Sabatier & Jenkins-Smith, 1993; Macnamara, 2005). While survey respondents may have difficulty with recall (Henry et al., 2012), gray literature can yield resources produced at the time of the event of interest. Data from gray literature may also produce a comparatively holistic rendering of the true network because it is not affected by low response rates.

Our data come from a corpus of gray literature documents sampled with a systematized online search protocol. By combining search terms of topic and location (e.g., ‘stormwater’ and ‘Tucson’), we identified stakeholders and their shared ties through the events they implemented together (i.e., policies and programs) in content published online from 2007 to 20171. Selecting this time range was motivated both by the limited availability of content prior to 2007 and by the theoretical expectation that a decade or more is required for experience with implementation and policy change to occur (Sabatier & Weible, 2007). This does not link change to any major event that would be expected to cause change but provides better information in the transition process that occurs over time.

We selected the first ten unique query results for each search term combination and year. Our heuristic for the ten-result limit was motivated by a preliminary inspection of search results – after the first page (approximately ten results), document content relevance decreased, and redundancy of query results increased. Lastly, with snowball sampling, we also read hyperlinks found in our initial corpus, together yielding 840 gray literature documents. With our interest in examining the governance networks in 2007 and 2017, however, we only relied on 358 of those documents.

Constructing implementation networks

We used network structures to operationalize joint implementation in the policy system. Social networks are mathematical abstractions of the real world (Henry, 2011), comprising actors or ‘nodes’, and the relationships, or ‘ties’ they share. These structures can demonstrate a variety of aspects about joint implementation, such as its prevalence among water governance stakeholders, and whether this mostly occurs among smaller groups.

An event-based approach allowed us to identify relevant network actors and joint implementation (Laumann et al., 1989). This means that, in the absence of survey data (with direct responses about shared ties), we identified ties between two or more actors if they implemented the same policies or programs (i.e., ‘events’). Specifically, we found evidence of joint implementation in the documents either with phrases indicating that two actors had some role in the implementation of a common event or by explicit mentions of collaborative or coordinated efforts in relation to that event. We operationalized implementation as enacting policy, using technology, or carrying out management activity central to IUWM. Collectively, these joint implementation ties comprised unipartite (i.e., actor-actor) networks.

The networks we constructed (Figure 1) represented joint implementation between stakeholder organizations in 2007 and 2017 to evaluate how patterns of actor involvement changed over time. Some organizations only implemented one common program or policy and only shared ties with one another, whereas others were involved in multiple programs or policies and thus shared implementation ties with a broader range of organizations. We further divided the network into two subgraphs (i.e., smaller networks) of organizations involved in IUWM and non-IUWM implementation, respectively2. We conceptualized IUWM as relating to wastewater reuse, demand management, stormwater management (including GI and green building practices), infrastructure management, and reduction of pollution and erosion/sedimentation (Mitchell, 2006; World Bank, 2012).

Prior to modeling, we assessed descriptive statistics of the IUWM network and of the non-IUWM network. By counting nodes and ties, we observed growth and attrition over time. Other statistics we calculated in the network included density, or the proportion of observed ties to all possible ties in the network (Wasserman & Faust, 1994), and the number of transitive triads, where two actors are tied through a direct and indirect path of length two. These triads form when actor A creates ties with actors B and C, increasing the likelihood that B and C also form a tie (Holland & Leinhardt, 1971).

Table 1 shows changes in the network statistics for the IUWM and non-IUWM networks in 2007 and 2017. Membership of the IUWM network increased by 141% and the number of ties more than quadrupled. There was also a slight increase in the triadic closure, but a slight decrease in density. This potentially indicates growth in the number of new policies and programs, and involvement of new actors not necessarily linked to those present in the 2007 network.

Table 1.

Descriptive statistics of the implementation networks, 2007–2017.

StatisticIUWM network
Non-IUWM network
2007201720072017
Number of nodes 54 130 73 111 
Number of ties 128 653 229 822 
Density 0.089 0.078 0.087 0.135 
Triads 0.528 0.536 0.414 0.804 
StatisticIUWM network
Non-IUWM network
2007201720072017
Number of nodes 54 130 73 111 
Number of ties 128 653 229 822 
Density 0.089 0.078 0.087 0.135 
Triads 0.528 0.536 0.414 0.804 

The non-IUWM implementation network showed comparatively less growth – the number of participants increased, but only by 52%. The number of joint implementation ties grew more in the IUWM network, as did the network density and triadic closure. These increases potentially indicate more stability both in organizations and their patterns of joint implementation over time.

Model

We used ERGMs to test our hypotheses. ERGMs estimate the likelihood that a simulated network structure would approximate the observed network we have measured with our selected parameters (Lusher et al., 2013). This network model is preferable to linear regression models because dyadic independent variables of interest violate the assumption of independence (Robins et al., 2012). We fit four models total to estimate the effects of our specified parameters for the IUWM and non-IUWM networks in 2007 and 2017, respectively.

Model parameters

Our model specified two covariates. First, we fit an exogenous nodal covariate that was the sum of media-reported ‘leadership’ roles in the governance network (i.e., not inferred from network position). Specifically, we operationalized this variable as the sum of occupying at least one of three roles mentioned in gray literature reports of brokerage, entrepreneurship, and leadership (with a score ranging from 0 to 3 to account for each reported role occupied). Our coding scheme required only one mention of having occupied a role to receive designation as leader, broker, or entrepreneur.

Brokerage reports were detected as the mention of any intermediary organizations engaged between two or more other organizations. For example, one online report described how a non-profit organization – the identified broker – worked on behalf of the Tucson City Council with the Pima County Regional Flood Control District to address localized flooding and water quality issues (WMG & PCRFCD, 2015). We coded organizations as policy entrepreneurs if they were reported to have championed adoption or implementation of a program, event, or policy. Lastly, we recorded verbatim reports of organizations as ‘leaders’.

The model also fit an edge covariate – the tendency for organizations to form ties with organizations from other sectors. We coded each organization as one of the following: an interest group, or organization from the public, non-profit, private, or academic sectors. Using the nodematch term from the ergm package (Hunter et al., 2008) in R, a statistical computing program (R Core Team, 2014), we estimated the effects of homophily, or preference for similar others in our model (McPherson et al., 2001); a negative coefficient reflected preference for forming ties with organizations from sectors different from their own.

We then included baseline structural terms to help explain the observed network. Density and clustering are two common parameters that are considered baseline features of networks. Given the structure of our data, we fit a Metropolis-Hastings constraint on density to avoid model degeneracy. We did, however, include a parameter for the likelihood of transitive triad formation, discussed earlier.

This section covers the model results and interpretation. While we find support for some of our hypotheses, we also observe coefficients with significance and directions that are counter to our expectations. Here, we discuss these outcomes in greater detail and potential empirical drivers for we observe.

Transitive triads

Table 2 shows us the ERGM coefficients for transitive triads, shared sector, and the presence of leadership. The positive, significant transitive triad coefficient suggests that the log likelihood of tie formation between two actors that share ties with a mutual connection is greater than would happen by chance. This means that in both the IUWM and non-IUWM networks, organizations directly linked to one another also have multiple shared partners. This coefficient may be inflated by the event-based coding (i.e., assuming all implementers of a common policy or program share a tie), but it also potentially suggests that some organizations were involved in the implementation of more than one policy or program in 2007 and 2017, respectively.

Table 2.

ERGM resultsa

ParametersIUWM network
Non-IUWM network
2007201720072017
Transitive triads 2.113*** (0.290) 7.023*** (0.442) 3.887*** (0.373) 7.427*** (0.571) 
Shared sector 0.260 (0.135) 0.241*** (0.063) 0.055 (0.082) −0.028 (0.056) 
Leadership role 0.221*** (0.064) 0.061* (0.028) 0.065 (0.048) −0.001 (0.034) 
ParametersIUWM network
Non-IUWM network
2007201720072017
Transitive triads 2.113*** (0.290) 7.023*** (0.442) 3.887*** (0.373) 7.427*** (0.571) 
Shared sector 0.260 (0.135) 0.241*** (0.063) 0.055 (0.082) −0.028 (0.056) 
Leadership role 0.221*** (0.064) 0.061* (0.028) 0.065 (0.048) −0.001 (0.034) 

aGoodness of fit simulations in Supplementary Material, Appendix B.

***p < 0.001; **p < 0.01; *p < 0.05; standard errors in parentheses.

In practice, triadic closure can be beneficial to joint implementation. These structures have been theoretically linked to social dynamics such as trust, norms of reciprocity, cooperation (Simmel, 1950; Krackhardt, 1999), and lower risk of defection in collective action scenarios (Berardo & Scholz, 2010). Triads can also improve access and distribution of resources network-wide (Henry & Vollan, 2014).

Leadership

The IUWM network

We also see a positive coefficient for leadership in the IUWM network over time. This lends support to Hypothesis 1b that leaders shape patterns of joint implementation in IUWM policies and programs. This finding suggests that leaders attract new ties – and increase their own centrality in the network structure – through actions such as initiating programs, garnering resources, and mobilizing others. Without organizations in leadership roles, these ties may have not otherwise formed. The effect of leadership on the network structure decreases over time, however, potentially indicating that new actors not reported in the gray literature as ‘leaders’ also began to introduce new IUWM approaches. Such shifts might reflect how organizations other than key actors like Tucson Water were taking the initiative to champion integrated approaches to urban water management.

In the non-IUWM network, we see – as proposed in Hypothesis 2b – no significant effect of reported leaders on joint implementation. This may indicate the relative stability of traditional water management approaches over time. In this case, both leaders and non-leaders may be less active in activities such as gaining support and resources, or mobilizing others into action.

Given these results, we consider (i) the validity of the gray literature reports, (ii) the types of organizations that assume leadership roles, and (iii) why we might observe change over time. We informally validate leadership by assessing degree centrality3, extent of program or policy participation, and how these statistics corresponded to gray literature reports of occupying at least one leadership role (i.e., ‘broker’, ‘entrepreneur’, and/or ‘leader’). Figure 2 shows these three metrics for 2007 and 2017. Organizations with more leadership nominations in the gray literature (‘Leadership’) also tend to have higher degree centrality (‘Centrality’) and participate in more policies and programs (‘Participation Frequency’).

Fig. 1.

IUWM and non-IUWM joint implementation networks for 2007 and 2017.

Fig. 1.

IUWM and non-IUWM joint implementation networks for 2007 and 2017.

Close modal

Tucson Water – the city's main water utility – was the top leader in the implementation of IUWM in both 2007 and 2017. Tucson Water's prominence may have stemmed from efforts to fulfill water conservation and reuse goals to meet GMA safe yield goals by 2025 (discussed previously). The utility pioneered, for example, a program to support residential installation of passive and active rainwater harvesting systems and an ordinance requiring commercial facilities constructed after 2010 to install rainwater harvesting features to capture 50% of the estimated annual irrigation needs for the property (Huckelberry & Letcher, 2009). Tucson Water also began promoting conservation through home water audits and by providing water-saving fixtures to customers (Tucson Water, 2013). Around this time, the utility also began offering incentives and support for retrofits and installation of high-efficiency appliances, rainwater harvesting, and residential graywater use (Tucson Water, 2013).

The University of Arizona also maintained a top leadership role over the study period. Since 2007, the University has both implemented a workshop with the Pima County Cooperative Extension and has pioneered multiple education- and information-based initiatives. By 2017, the University participated in 24 different IUWM programs and policies. Continuing an emphasis on water management education, the University introduced an instructional program and an online tool to inform consumers and aid in decision-making relating to rainwater harvesting and conservation strategies (WRRC & CALS, 2017; WRRC, 2019). It also partnered with local government to reduce stormwater pollution and wetlands restoration (PAG, 2016).

In 2007, the Tucson Department of Transportation was a leading organization, institutionalizing curb cut standards, allowing streetside capture of stormwater runoff for trees and shrubs (Kraker, 2014). This action was one of the several codified implementation strategies included in the Department's Stormwater Management Program, which provided a framework for city staff to work in ‘an effort to reduce discharge of pollutants to the municipal storm sewer system and into the navigable waters’ (TDOT, 2014).

By 2017, the Watershed Management Group surpassed Tucson Department of Transportation in both project involvement and its connectedness in the network. The Watershed Management Group has supported Tucson Water's incentives and financial support for residential rainwater harvesting by leading installation and maintenance workshops (WMG, 2019). The Group also promotes rainwater harvesting in Tucson through group job training and by hosting demonstration sites (WMG, 2018).

The non-IUWM network

As proposed in Hypothesis 2b, we find no significant effect of leadership on joint implementation in the non-IUWM network. Just as the descriptive network statistics suggest stability in network members over time (Table 2), we also see stability in the leading public organizations.

The two top leading organizations – Tucson Water and the Pima County Regional Wastewater Reclamation Department – are both obligated to provide water and wastewater services. Tucson Water is responsible for supply in the core metropolitan area and has led siting and maintenance of major infrastructure to store and distribute potable resources through traditional, piped systems. Examples of the utility's programs to ensure quality service provision include its lead pipe detection programs and water dye testing program for potable and reclaimed resources (Tucson Water, 2010, 2014).

The Pima County Regional Wastewater Reclamation Department was the second most central leader involved in the most non-IUWM programs and policies over time. The Wastewater Reclamation Department is responsible for Tucson's wastewater conveyance system and reclamation facilities (PCRWRD, 2014). Examples of its other tasks include a subsidy program offering low-income support with rising sewer fees, and its grease collection campaign to reduce drain blockage (PCRWRD, 2019).

Although the Pima Association of Governments and the Pima County Department of Environmental Quality do not carry out the core water and wastewater service provision obligations, they were more involved with other organizations and programs than most in the non-IUWM network. Nonetheless, the Department of Environmental Quality was not a reported leader in the gray literature, and the Pima Association of Government's reported leadership was related to a stormwater management program. These findings suggest that the gray literature sufficiently detects outstanding leaders with high involvement but that this approach is not always accurate.

Sectoral affiliation

We observe mixed results for sectoral heterogeneity as a driver of joint policy or program implementation. In the IUWM network, there is no significant effect in 2007, but there exists a significant, positive effect for shared sectoral affiliation in 2017. While the latter result aligns with our expectations that sector indeed plays a role in joint implementation patterns, it is counter to our proposition in Hypothesis 1a, that organizations with different sectoral affiliations tend to work together to implement IUWM policies and programs.

The patterns of sectoral growth offer potential clues to help us understand these findings. In the IUWM network, there was a 500% increase in the number of private organizations, a 118% increase in the number of non-profit organizations, and a 73% increase in the number of public organizations partaking from 2007 to 2017 (Figure 3). Assuming public organizations – comprising the largest sectoral proportion in the network – also worked with Tucson Water to meet GMA requirements, there may have been relatively fewer connections with organizations not obligated to achieve regulatory benchmarks. Nonetheless, the extensive growth in the number of new ties, decrease in density, and clear emergence of more auxiliary network actors (Figure 2) suggest new emerging programs not associated with the public sector.

Fig. 2.

Leadership in the IUWM and non-IUWM networks, measured by degree centrality, gray literature nominations, and frequency of project participation.

Fig. 2.

Leadership in the IUWM and non-IUWM networks, measured by degree centrality, gray literature nominations, and frequency of project participation.

Close modal
Fig. 3.

Sectoral distribution of organizations in the IUWM and non-IUWM networks.

Fig. 3.

Sectoral distribution of organizations in the IUWM and non-IUWM networks.

Close modal

Lastly, coefficients for non-IUWM joint implementation among organizations from the same sector are insignificant. Thus, our model findings do not support Hypothesis 2a, which proposes that organizations tend to form ties with others from the same sector. The coefficients also show us that the effect is positive in 2007, but negative in 2017. This indicates that organizations in the non-IUWM implementation network tended to form fewer ties with others from the same sector over time. One possible explanation may be that there was a core of stable ties for ongoing water management approaches (especially among governmental organizations), but also new programs that increased the sectoral diversity. Initiatives such as the Pima County's Dispose-A-Med, for example (introduced in 2009), involves 33 partners, including the City, as well as local organizations from the non-profit, private, and academic sectors.

Data limitations

Our study illustrates how the joint implementation of policy innovations does not always rely on heterogeneous networks of policy stakeholders and that – in this case – leadership roles remain fairly stable, despite having a decreasing effect on patterns of join implementation over time. Yet, these findings should be interpreted with some caution because of data collection and coding challenges.

Data collection

As mentioned previously, we collected data by systematically searching key terms that reflect different dimensions of water governance. However, we realized post hoc that by not searching Google in a private browsing session (i.e., ‘incognito mode’), the search engine may have biased results. In 2005, Google personalized searches for all users with Google accounts, such that the search results would be tailored based on previous searches (Google, 2018). A private search would likely reduce the extent with which the browser dynamically updated results, as previous work has found that Google Chrome searches in private mode do not raise major concern for storage of web browser artifacts (Flowers et al., 2016).

Data coding and biases

Rendered social networks do not perfectly represent the real world. This principle also applies to our event-based approach to code network data. Namely, our assumption of direct engagement between all organizations participating in the same event likely led to overestimation of network density, degree centrality, and relatedly, modeled effect of leadership and heterogeneous ties in the implementation networks. Our cross-sectional analysis also likely limited our ability to detect attrition or new organizations over time.

Notwithstanding efforts to resolve biases from data collection and coding, the quality and availability of the gray literature information can still be uneven. For instance, gray literature produced regularly through government agencies may lead to better coverage of public programs and policies (Maclaren, 1996), whereas media reports may be comparatively sparse or have targeted focus on noteworthy events or well-known stakeholders (Yi & Scholz, 2016). This was evident in our study: the public sector published 49% and media sources (i.e., news and magazines) were responsible for 37% of the gray literature that reported on IUWM policies and programs in 2007 and 2017. In addition to reporting bias, our time frame of one decade likely provides but a narrow window of information regarding the policy process in Tucson's water governance system.

The validation of network measures with additional data sources would indubitably strengthen findings. Future data collection strategies to support such efforts might entail, for example, qualitative interviews of representatives from identified leader organizations. Accounts from organization representatives could offer deeper insight on social processes and on the relevance of the identified network leaders and patterns (Rice et al., 2014).

Extending systematic studies of implementation

Lastly, Tucson's strong role of the public sector brings into question the generalizability of our study's findings. Our models suggest that – unlike other systems – joint implementation of IUWM strategies is not significantly predicted by sectoral heterogeneity. This directs our attention to a related line of inquiry: how does implementation vary across systems with different natural and human features more broadly? Some systems have different governance regimes, where organizations such as corporatized water systems, private or competing public water utilities, or separate statutory authorities yield different processes or outcomes relating to IUWM implementation (Mukheibir et al., 2014; Furlong et al., 2016). Another area to consider is the nature of context-specific needs. Who leads implementation, for example, could depend on whether an urban system is responding to the needs of rapid peri-urban development, such as that occurring in developing countries (Díaz et al., 2016), or to climatic pressures such as extreme flooding and waterlogging (Wang et al., 2018). We thus encourage more work that applies network analytic techniques to build a broader, systematic understanding of implementation processes in sustainability transitions.

The body of the literature on policy implementation has made great strides over the past 30 years, but there is little systematic examination of what motivates joint implementation processes. This paper uses inferential social network analysis and content analysis to identify drivers of joint policy implementation in adaptive water governance systems. We focus on implementation in the context of system transition to IUWM. The sustainability transitions literature has cited factors such as administrative responsibilities, maintaining constituent support, and budgetary constraints as keeping organizations from undertaking new tasks associated with the IUWM approach. Yet, joint implementation efforts between organizations from different sectors (e.g., public, private, non-profit) and leadership have also been described as instrumental in advancing transitions to IUWM. We find that the institutional setting may play a key role in when, and the extent to which, leadership affects joint implementation and when sectoral integration occurs.

Our work offers theoretical contributions to the sustainability transitions and policy implementation literature. While we find that actors outside traditional public service provision increasingly participate in joint policy implementation over time (especially for GI), public organizations are highly involved through institutional and professional obligation. At the same time, we see that – counter to many sustainability transitions studies – government actors are leading niche innovations pursuant to regulatory structures. This potentially signals a need for more work to understand when and why stakeholders endogenous to a current regime may implement conscious or planned efforts to make transitions (Quitzau et al., 2013). Future work to address this area of focus may gain traction by examining how local governments spearhead and continue policy innovations when embedded in different state- and federal-level institutions. This, in conjunction with a systematic assessment of social, economic, and ecological capacities and vulnerabilities, can paint a better picture of how some cities are overcoming barriers to IUWM implementation.

Methodologically, this paper shows promise for future text analysis – much of which is already gaining traction in studies of policy and political science (e.g., Nowlin, 2016; Gao et al., 2018). Automated web scraping and rule-based extraction, for example, is a promising direction to improve the breadth and accuracy of content analysis (Valenzuela-Escárcega et al., 2015). Furthermore, this will enable large-scale, comparative analyses for theory building and to gather new, timely knowledge for understanding adaptation.

We thank the National Science Foundation for support of this project, Award #1518376.

The Supplementary Material for this paper is available online at https://dx.doi.org/10.2166/wp.2020.153.

1

See Supplementary Material, Appendix A for a list of search terms.

2

In this case, there were actors that were active in both IUWM and non-IUWM implementation networks.

3

Degree centrality is a common statistic in social network analysis to identify leadership. All ‘leader’ organizations selected from the 2007 and 2017 networks had a centrality score at least two standard deviations above the mean.

ADWR
(
1999
).
Third Management Plan for Tucson Active Management Area, 2000–2010. Available at: http://infoshare.azwater.gov/docushare/dsweb/Get/Document-10035/TucsonAMA_3MP.pdf (Accessed 2019)
.
ADWR
(
2016a
).
Active Management Areas
. .
ADWR
(
2016b
).
Fourth Management Plan – Tucson Active Management Area. Available at: http://infoshare.azwater.gov/docushare/dsweb/Get/Document-10038/TAMA_4MP_Complete.pdf (Accessed 2019)
.
AMWUA
(
2020
).
Drought & Shortage
. .
Arizona State Senate
(
2007
).
Amended Fact Sheet for S.B. 1575: water adequacy amendments. Available at: https://www.azleg.gov/legtext/48leg/1r/summary/s.1575nrra_aspassedbysenate.doc.htm (Accessed 2020).
Arnold
G.
(
2015
).
Street-level policy entrepreneurship
.
Public Management Review
17
(
3
),
307
327
.
August
J. L.
Gammage
G.
Jr.
(
2007
).
Shaped by water
. In:
Arizona Water Policy: Management Innovations in an Urbanizing, Arid Region
.
B. G. Colby & K. L. Jacobs (eds).
Resources for the Future
,
Washington, D.C
, p.
10
.
Benedict
M. A.
McMahon
E. T
. (
2012
).
Green Infrastructure: Linking Landscapes and Communities
.
Island Press
,
Washington DC
.
Berardo
R.
Scholz
J. T.
(
2010
).
Self-organizing policy networks: risk, partner selection, and cooperation in estuaries
.
American Journal of Political Science
54
(
3
),
632
649
.
Bos
J. J.
Brown
R. R.
(
2012
).
Governance experimentation and factors of success in socio-technical transitions in the urban water sector
.
Technological Forecasting and Social Change
79
(
7
),
1340
1353
.
Brown
R. R.
Farrelly
M. A.
(
2009
).
Delivering sustainable urban water management: a review of the hurdles we face
.
Water Science and Technology
59
(
5
),
839
846
.
Brown
R. R.
Keath
N.
Wong
T. H.
(
2009
).
Urban water management in cities: historical, current and future regimes
.
Water Science and Technology
59
(
5
),
847
855
.
Brown
R. R.
Farrelly
M. A.
Loorbach
D. A.
(
2013
).
Actors working the institutions in sustainability transitions: the case of Melbourne's stormwater management
.
Global Environmental Change
23
(
4
),
701
718
.
Brumley
J.
Marks
C.
Chau
A.
Lowrance
R.
Huang
J.
Richardson
C.
Acree
S.
Ross
R.
Beak
D.
(
2018
).
The Influence of Green Infrastructure Practices on Groundwater Quality: The State of the Science
.
U.S. Environmental Protection Agency
,
Washington, DC
,
EPA/600/R-18/227
.
Castle
A.
Fleck
J.
(
2019
).
Featured Paper: The Risk of Curtailment Under the Colorado River Compact
.
Chisholm
D
. (
1992
).
Coordination Without Hierarchy: Informal Structures in Multiorganizational Systems
.
University of California Press
,
Los Angeles
.
Christopoulos
D.
Ingold
K.
(
2015
).
Exceptional or just well connected? Political entrepreneurs and brokers in policy making
.
European Political Science Review
7
(
3
),
475
498
.
City of Tucson & Pima County
(
2009
).
City of Tucson and Pima County Water Conservation Technical Paper
.
Daigger
G. T.
Crawford
G. V.
(
2007
).
Enhancing water system security and sustainability by incorporating centralized and decentralized water reclamation and reuse into urban water management systems
.
Journal of Environmental Engineering and Management
17
(
1
),
1
.
DeLeon
P.
DeLeon
L.
(
2002
).
What ever happened to policy implementation? An alternative approach
.
Journal of Public Administration Research and Theory
12
(
4
),
467
492
.
Díaz
P.
Stanek
P.
Frantzeskaki
N.
Yeh
D. H.
(
2016
).
Shifting paradigms, changing waters: transitioning to integrated urban water management in the coastal city of Dunedin, USA
.
Sustainable Cities and Society
26
,
555
567
.
Dobbie
M. F.
Brown
R. R.
Farrelly
M. A.
(
2016
).
Risk governance in the water sensitive city: practitioner perspectives on ownership, management and trust
.
Environmental Science & Policy
55
,
218
227
.
Farrelly
M.
Brown
R.
(
2011
).
Rethinking urban water management: experimentation as a way forward?
Global Environmental Change
21
(
2
),
721
732
.
Fattahi
P.
Fayyaz
S.
(
2010
).
A compromise programming model to integrated urban water management
.
Water Resources Management
24
(
6
),
1211
1227
.
Feiock
R. C.
(
2013
).
The institutional collective action framework
.
Policy Studies Journal
41
(
3
),
397
425
.
Fletcher
T. D.
Shuster
W.
Hunt
W. F.
Ashley
R.
Butler
D.
Arthur
S.
Trowsdale
S.
Barraud
S.
Semadeni-Davies
A.
Bertrand-Krajewski
J. L.
Mikkelsen
P. S.
(
2015
).
SUDS, LID, BMPs, WSUD and more – the evolution and application of terminology surrounding urban drainage
.
Urban Water Journal
12
(
7
),
525
542
.
Flowers
C.
Mansour
A.
Al-Khateeb
H. M.
(
2016
).
Web browser artefacts in private and portable modes: a forensic investigation
.
International Journal of Electronic Security and Digital Forensics
8
(
2
),
99
117
.
Fricano
R. J.
Grass
A. K.
(
2014
).
Evaluating American rainwater harvesting policy: a case study of three US cities
.
Journal of Sustainable Development in Africa
7
(
6
),
133
.
Furlong
C.
Gan
K.
De Silva
S.
(
2016
).
Governance of integrated urban water management in Melbourne, Australia
.
Utilities Policy
43
,
48
58
.
Furlong
C.
Brotchie
R.
Considine
R.
Finlayson
G.
Guthrie
L.
(
2017
).
Key concepts for integrated urban water management infrastructure planning: lessons from Melbourne
.
Utilities Policy
45
,
84
96
.
doi:10.1016/j.jup.2017.02.004
.
Gao
Y.
Church
S. P.
Peel
S.
Prokopy
L. S.
(
2018
).
Public perception towards river and water conservation practices: opportunities for implementing urban stormwater management practices
.
Journal of Environmental Management
223
,
478
488
.
Garcia
M.
Koebele
E.
Deslatte
A.
Ernst
K.
Manago
K. F.
Treuer
G.
(
2019
).
Towards urban water sustainability: analyzing management transitions in Miami, Las Vegas, and Los Angeles
.
Global Environmental Change
58
,
101967
.
Gliedt
T.
Hoicka
C. E.
Jackson
N.
(
2018
).
Innovation intermediaries accelerating environmental sustainability transitions
.
Journal of Cleaner Production
174
,
1247
1261
.
Goggin
M. L.
(
1990
).
Implementation theory and practice: Toward a third generation
.
Scott Foresman
,
Glenview, IL
.
Gómez
D. Y.
Teixeira
L. G.
(
2017
).
Residential rainwater harvesting: effects of incentive policies and water consumption over economic feasibility
.
Resources, Conservation and Recycling
127
,
56
67
.
Google
(
2018
).
Search Settings
. .
Grigg
N. S.
Connor
T.
Maas
A.
(
2018
).
Financing integration of urban water systems: from service provision to resource management
.
Public Works Management & Policy
23
(
2
),
186
198
.
Hartley
J.
Sørensen
E.
Torfing
J.
(
2013
).
Collaborative innovation: a viable alternative to market competition and organizational entrepreneurship
.
Public Administration Review
73
(
6
),
821
830
.
Henry
A. D.
(
2009
).
The challenge of learning for sustainability: a prolegomenon to theory
.
Human Ecology Review
16
(
2
),
131
140
.
Henry
A. D.
(
2011
).
Ideology, power, and the structure of policy networks
.
Policy Studies Journal
39
(
3
),
361
384
.
Henry
A. D.
(
2017
).
Network segregation and policy learning
. In:
The Oxford Handbook of Political Networks
.
J. N. Victor, A. H. Montgomery & M. Lubell (eds).
Oxford University Press
,
Oxford
, pp.
559
588
.
Henry
A. D.
Vollan
B.
(
2014
).
Networks and the challenge of sustainable development, annual review of environment and resources
.
Annual Reviews
39
(
1
),
583
610
.
Henry
A. D.
Lubell
M.
McCoy
M.
(
2012
).
Survey-based measurement of public management and policy networks
.
Journal of Policy Analysis and Management
31
(
2
),
432
452
.
Hess
D. J.
Brown
K. P.
(
2018
).
Water and the politics of sustainability transitions: from regime actor conflicts to system governance organizations
.
Journal of Environmental Policy & Planning
20
(
2
),
128
142
.
Hjern
B.
Hull
C.
(
1982
).
Implementation research as empirical constitutionalism
.
European Journal of Political Research
10
(
2
),
105
115
.
Holland
P. W.
Leinhardt
S.
(
1971
).
Transitivity in structural models of small groups, comparative group studies
.
SAGE Publications
2
(
2
),
107
124
.
Huckelberry
C. H.
Letcher
M.
(
2009
).
Integrating Land Use Planning with Water Resources and Infrastructure Technical Paper
. .
Hunter
D. R.
Handcock
M. S.
Butts
C. T.
Goodreau
S. M.
Morris
M.
(
2008
).
Ergm: a package to fit, simulate and diagnose exponential-family models for networks
.
Journal of Statistical Software
24
(
3
),
54860
.
Koontz
T. M.
Steelman
T. A.
Carmin
J.
Kormacher
K. S.
Mosely
C.
Thomas
C. W
. (
2004
).
Collaborative Environmental Management: What Roles for Government
.
Routledge
,
Washington, DC
.
Krackhardt
D.
(
1999
).
The ties that torture: simmelian tie analysis in organizations
.
Research in the Sociology of Organizations
16
(
1
),
183
210
.
Kraker
D.
(
2014
).
Rain Man: how one Tucson resident harvests the rain. MPR News, 24 May
. .
Larsen
T. A.
Hoffmann
S.
Lüthi
C.
Truffer
B.
Maurer
M.
(
2016
).
Emerging solutions to the water challenges of an urbanizing world
.
Science
352
(
6288
),
928
933
.
Laumann
E. O.
Marsden
P. V.
Prensky
D.
(
1989
).
The boundary specification problem in network analysis
.
Research Methods in Social Network Analysis
61
,
87
.
Lusher
D.
Koskinen
J.
Robins
G
. (
2013
).
Exponential Random Graph Models for Social Networks: Theory, Models, and Applications
.
Cambridge University Press
,
New York
.
Maclaren
V. W.
(
1996
).
Urban sustainability reporting
.
Journal of the American Planning Association
62
(
2
),
184
202
.
Macnamara
J. R.
(
2005
).
Media content analysis: its uses, benefits and best practice methodology
.
Asia Pacific Public Relations Journal
6
(
1
),
1
.
Makropoulos
C. K.
Natsis
K.
Liu
S.
Mittas
K.
Butler
D.
(
2008
).
Decision support for sustainable option selection in integrated urban water management
.
Environmental Modelling & Software
23
(
12
),
1448
1460
.
Mazmanian
D. A.
Sabatier
P. A
. (
1983
).
Implementation and Public Policy
.
Scott Foresman
,
Lanham, MD
.
McPherson
M.
Smith-Lovin
L.
Cook
J. M.
(
2001
).
Birds of a feather: homophily in social networks
.
Annual Review of Sociology
27
(
1
),
415
444
.
Meerow
S.
Newell
J. P.
(
2017
).
Spatial planning for multifunctional green infrastructure: growing resilience in Detroit
.
Landscape and Urban Planning
159
,
62
75
.
Megdal
S. B.
Dillon
P.
Seasholes
K.
(
2014
).
Water banks: using managed aquifer recharge to meet water policy objectives
.
Water
6
(
6
),
1500
1514
.
Mintrom
M.
Norman
P.
(
2009
).
Policy entrepreneurship and policy change
.
Policy Studies Journal
37
(
4
),
649
667
.
Mukheibir
P.
Howe
C.
Gallet
D.
(
2014
).
What's getting in the way of a One Water approach to water service planning and management?
Water: Journal of the Australian Water Association
41
(
3
):
67
73
.
Newell
J. P.
Seymour
M.
Thomas
Y.
Renteria
J.
Longcore
T.
Wolch
J.
Shishkovsky
A.
(
2013
).
Green alley programs: planning for a sustainable urban infrastructure?
Cities
31
,
144
155
.
Nilsen
P.
Ståhl
C.
Roback
K.
Cairney
P.
(
2013
).
Never the twain shall meet?—a comparison of implementation science and policy implementation research
.
Implementation Science
8
,
63
.
NOAA
(
2019
).
Monthly and Daily Normals (1981–2010) Plus Daily Extremes (1895–2019) for TUCSON, ARIZONA, Tucson Monthly and Daily Normals
.
Available at: https://www.wrh.noaa.gov/twc/climate/tus.php (Accessed 2019)
.
Nowlin
M. C.
(
2016
).
Modeling issue definitions using quantitative text analysis
.
Policy Studies Journal
44
(
3
),
309
331
.
O'Donnell
E. C.
Lamond
J. E.
Thorne
C. R.
(
2017
).
Recognising barriers to implementation of blue-green infrastructure: a Newcastle case study
.
Urban Water Journal
14
(
9
),
964
971
.
O'Toole
L. J.
(
2000
).
Research on policy implementation: assessment and prospects
.
Journal of Public Administration Research and Theory
10
(
2
),
263
288
.
PAG
(
2016
).
Clean Water Starts With Me!
Available at: http://www.pagregion.com/tabid/766/Default.aspx (Accessed 8 July 2019)
.
PCRWRD
(
2014
).
Strategic Plan FY14/15–FY18/19
.
PCRWRD
(
2019
).
Sewer Outreach Subsidy (SOS) Program
. .
Pfeffer
J.
Salancik
G. R
. (
2003
).
The External Control of Organizations: A Resource Dependence Perspective
.
Stanford University Press
,
Stanford, CA
.
Pitzer
G.
Eden
S.
Gelt
J.
(
2007
).
Layperson's Guide to Arizona Water
. .
Pressman
J.
Wildavsky
A.
(
1973
).
Implementation
.
University of California Press
,
Berkeley
.
Provan
K. G.
Milward
H. B.
(
2001
).
Do networks really work? A framework for evaluating public-sector organizational networks
.
Public Administration Review
61
(
4
),
414
423
.
Quitzau
M.
Jensen
J. S.
Elle
M.
Hoffmann
B.
(
2013
).
Sustainable urban regime adjustments
.
Journal of Cleaner Production
50
,
140
147
.
R Core Team
(
2014
).
R: A Language and Environment for Statistical Computing
.
R Foundation for Statistical Computing
,
Vienna
,
Austria
.
Available at: http://www.R-project.org/ (Accessed 2019)
.
Rice
E.
Holloway
I. W.
Barman-Adhikari
A.
Fuentes
D.
Brown
C. H.
Palinkas
L. A.
(
2014
).
A mixed methods approach to network data collection
.
Field Methods
26
(
3
),
252
268
.
Robins
G.
Lewis
J. M.
Wang
P.
(
2012
).
Statistical network analysis for analyzing policy networks
.
Policy Studies Journal
40
(
3
),
375
401
.
Sabatier
P. A.
Jenkins-Smith
H. C.
(
1993
).
The advocacy coalition framework: Assessment, revisions, and implications for scholars and practitioners
. In:
Policy Change and Learning: An Advocacy Coalition Approach
.
P. A. Sabatier & H. C. Jenkins-Smith (eds).
Westview, Boulder, CO
, pp.
211
236
.
Sabatier
P. A.
Weible
C. M.
(
2007
).
The advocacy coalition framework
. In:
Theories of the Policy Process
, Vol. 2.
P. A. Sabatier, (ed.).
Westview, Boulder, CO
, pp.
189
220
.
Sharma
A. K.
Rashetnia
S.
Gardner
T.
Begbie
D.
(
2019
).
WSUD design guidelines and data needs
. In:
Approaches to Water Sensitive Urban Design
, pp.
75
86
.
doi:10.1016/b978-0-12-812843-5.00004-6
Sharvelle
S.
Dozier
A.
Arabi
M.
Reichel
B.
(
2017
).
A geospatially-enabled web tool for urban water demand forecasting and assessment of alternative urban water management strategies
.
Environmental Modelling & Software
97
,
213
228
.
Shifflett
S. D.
Newcomer-Johnson
T.
Yess
T.
Jacobs
S.
(
2019
).
Interdisciplinary collaboration on green infrastructure for urban watershed management: an Ohio case study
.
Water
11
(
4
),
738
757
.
Simmel
G.
(
1950
).
Individual and Society. The Sociology of Georg Simmel
.
Free Press
,
New York
.
Stoker
P.
Pivo
G.
Stoicof
A.
Kavekewitz
J.
Grigg
N.
Howe
C.
(
2019
).
Joining-up Urban Water Management with Urban Planning and Design. Available at: https://www.waterrf.org/resource/joining-urban-water-management-urban-planning-and-design (Accessed 2020).
Taylor
A. C.
(
2009
).
Sustainable urban water management: understanding and fostering champions of change
.
Water Science & Technology
59
(
5
),
883
891
.
TDOT
(
2014
).
Stormwater Management Program (SWMP)
. .
Termeer
C. J.
(
2009
).
Water professionals and public leadership
.
Irrigation and Drainage
58
(
S2
),
S212
S216
.
Torfing
J.
(
2016
).
Collaborative Innovation in the Public Sector
.
Georgetown University Press
,
Washington, DC
.
Tucson Water
(
2010
).
Dye Testing: Ensuring the Separation of Potable and Reclaimed Water Systems. Available at: https://www.tucsonaz.gov/files/water/docs/dyetesting.pdf (Accessed 2019).
Tucson Water
(
2013
).
2012 Update, Water Plan: 2000–2050. Available at: https://www.tucsonaz.gov/files/water/docs/2012_Update_Water_Plan_2000-2050.pdf (Accessed 2019).
Tucson Water
(
2014
).
Lead and Copper
. .
US Department of Interior
(
2007
).
Record of Decision Colorado River Interim Guidelines for Lower Basin Shortages and the Coordinated Operations for Lake Powell and Lake Mead. Available at: https://www.usbr.gov/lc/region/programs/strategies/RecordofDecision.pdf (Accessed 2019).
Valenzuela-Escárcega
M. A.
Hahn-Powell
G.
Surdeanu
M
. (
2015
).
A domain-independent rule-based framework for event extraction
. In:
Proceedings of ACL-IJCNLP 2015 System Demonstrations
.
Beijing, China
,
July 26–31
, pp.
127
132
.
van de Meene
S. J.
Brown
R. R.
Farrelly
M. A.
(
2011
).
Towards understanding governance for sustainable urban water management
.
Global Environmental Change
21
(
3
),
1117
1127
.
Wang
H.
Mei
C.
Liu
J. H.
Shao
W. W.
(
2018
).
A new strategy for integrated urban water management in China: Sponge city
.
Science China Technological Sciences
61
(
3
),
317
329
.
Wasserman
S.
Faust
K
. (
1994
).
Social Network Analysis: Methods and Applications
.
Cambridge University Press
,
New York
.
Winter
S. C
. (
2012
).
Implementation perspectives: status and reconsideration
. In:
The SAGE Handbook of Public Administration
.
B. G. Peters & I. Pierre (eds).
Sage
,
Los Angeles, CA
, pp.
265
278
.
WMG & PCRFCD
(
2015
).
Solving Flooding Challenges with Green Stormwater Infrastructure in the Airport Wash Area
.
Wong
T. H. F.
Brown
R. R.
(
2009
).
The water sensitive city: principles for practice
.
Water Science and Technology
60
(
3
),
673
682
.
World Bank
(
2012
).
Blue Water Green Cities: An Initiative from the World Bank for Integrated Urban Water Management
.
WRRC
(
2019
).
Water Harvesting Assessment Toolbox
.
Available at: https://wrrc.arizona.edu/DWHI/toolbox (Accessed 2019)
.
WRRC & CALS
(
2017
).
C2E: Conserve to Enhance
.
Available at: https://conserve2enhance.org/ (Accessed 2019)
.

Supplementary data