Sustainable Cities and Society 30 (2017) 128–138

Contents lists available at ScienceDirect

Sustainable Cities and Society

journal homepage: www.elsevier.com/locate/scs

An integrative regional resilience framework for the changing urban

water paradigm
Patricia Gonzales a,b,∗ , Newsha K. Ajami b,c
a
Department of Civil and Environmental Engineering, Stanford University, 473 Via Ortega, Room 314, Stanford, CA 94305, United States
b
ReNUWIt Engineering Research Center, Stanford University, 473 Via Ortega, Room 117, Stanford, CA 94305, United States
c
Woods Institute for the Environment, Stanford University, 473 Via Ortega, Room 204, Stanford, CA 94305, United States

a r t i c l e i n f o a b s t r a c t

Article history: The water sector is going through a paradigm shift. Many communities are incorporating decentralized
Received 25 August 2016 solutions such as water reuse and recycling, stormwater capture, and demand-side management in order
Received in revised form 7 December 2016 to address both short-term and long-term water resources challenges due to population increase, eco-
Accepted 22 January 2017
nomic growth, intensified climate variability, as well as environmental concerns. For these projects to be
Available online 29 January 2017
sustainable, local characteristics including social and institutional contexts must be incorporated in the
planning process. This paper presents a flexible and bottom-up framework that facilitates integration
Keywords:
of such characteristics in evaluation of various water resource management strategies. It incorporates
Urban water
Regional water management
various locally-driven factors such as water use efficiency, stress on existing supplies, and adaptation
Resilience capacity potential, to identify how various local and regional solutions affect resiliency at the utility and
Water supplies regional levels. Rather than defining top-down resilience standards, this framework incorporates quanti-
Water demand tative and qualitative assessments that can help decision-makers tailor adaptation measures to the needs
Supply diversity and opportunities of a given location or community. A case study application of the framework in the San
Francisco Bay Area highlights how community-level characteristics can be used to identify opportunities
and adaptation strategies in order to enhance both local and collective water resource resiliency.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction on supply sources that are themselves sensitive to seasonal

precipitation changes, periodic droughts and infrastructure degra-
The past few decades have seen changes in the scope and dation (Brozovic, Sunding, & Zilberman, 2007; Cayan et al., 2010;
focus of urban water management in many parts of the world Diffenbaugh, Swain, & Touma, 2015; Viviroli et al., 2011). Fur-
(Hering, Waite, Luthy, Drewes, & Sedlak, 2013). Traditional urban thermore, the fragmented nature of today’s water management
water systems were built as a combination of independent com- institutions translates into a disconnect between local priorities,
ponents and managed by multiple governing agencies, typically environmental concerns, and regional water resources that know
comprising extensive distribution pipelines to bring water into no political boundaries (Graymore, Sipe, & Rickson, 2010; Hering
cities, centralized treatment plants, and disposal systems for mov- et al., 2013; Hughes & Pincetl, 2014; Kauffman, 2002; Padowski &
ing wastewater and stormwater away from urban centers. In many Jawitz, 2009). In response to the growing water insecurity, espe-
urban areas, where climate variability, growing populations, and cially in arid and semi-arid regions such as Australia, Israel, and the
degrading ecosystems are prominent, it is now evident that this Western United States, there has been increasing interest in alter-
traditional paradigm will not be suitable to meet future chal- native practices including water recycling and reuse (Arlosoroff,
lenges (Chartres & Williams, 2006; Hering et al., 2013; Padowski 2007), stormwater capture, desalination, water efficiency and con-
& Gorelick, 2014; Viviroli et al., 2011). There is increasing aware- servation measures (Hornberger, Hess, & Gilligan, 2015; Mini,
ness that over-reliance on imported supplies places undue stress Hogue, & Pincetl, 2015), and combinations of these options (Beh,
Dandy, Maier, & Paton, 2014; Luthy & Sedlak, 2015; Newman,
Dandy, & Maier, 2014; Srinivasan, Lambin, Gorelick, Thompson, &
Rozelle, 2012; Tarroja et al., 2014). The adoption of these alterna-
∗ Corresponding author. Present address: 473 Via Ortega, Room 204, Stanford, tive supplies has been accompanied by innovative and adaptive
California 94305, United States. governance configurations in the search for more reliable and
E-mail addresses: [email protected] (P. Gonzales), [email protected]
resilient systems that can withstand variable and uncertain condi-
(N.K. Ajami).

http://dx.doi.org/10.1016/j.scs.2017.01.012
2210-6707/© 2017 Elsevier Ltd. All rights reserved.
 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138 129

Fig. 1. Resilience framework for urban water systems, based on the assessment of supply, demand, and adaptation capacity.

tions (Beh, Maier, & Dandy, 2015; Connell-Buck, Medellin-Azuara, bottom-up resiliency strategies based on local needs and opportu-
Lund, & Madani, 2011; Ferguson, Brown, Frantzeskaki, de Haan, & nities, as well as the potential for collaborative approaches.
Deletic, 2013; Hughes, Pincetl, & Boone, 2013; Nelson, Howden, & In this paper we present a novel resilience framework that
Smith, 2008; Paton, Dandy, & Maier, 2014; Rijke, Farrelly, Brown, & builds upon previous assessments with two new and significant
Zevenbergen, 2013; Tal, 2006). Despite emerging initiatives, there contributions designed for the emerging challenges in the urban
are still many financial, political, and institutional barriers to the water sector: (1) a focus on the added flexibility that decentralized
widespread adoption of more reliable and resilient urban water approaches can provide to urban water systems; and (2) a bottom-
supplies (Brown & Farrelly, 2009; Hughes, 2013; Moglia, Alexander, up measure that integrates the local socio-hydrologic realities of
& Sharma, 2011; Winz, Trowsdale, & Brierley, 2014). As the water water service providers and assesses their role on regional-scale
use cycle evolves, existing tools for the assessment, planning and resilience. Thus, rather than attempt to define top-down resilience
management of urban water systems need to be re-designed to standards for a region and arbitrary numerical goals to pursue, our
support a more holistic evaluation of the combination of supply systematic framework helps identify strategies that suit each com-
diversification options, system-wide dynamics, and intra-regional munity’s needs and opportunities which can ultimately improve
collaboration called for by the current paradigm shift in urban overall regional resiliency. We follow qualitative and quantitative
water management. indicators related to three pillars: supply, demand, and adaptation
Numerous metrics and assessments that have been developed capacity. We demonstrate the application of this framework to a
in the past can provide a robust backbone to enable such a systems- subset of water utilities in the San Francisco Bay Area of California,
level perspective (Brown et al., 2015; Juwana, Muttil, & Perera, who have recently been under significant pressure to implement
2012; Marlow, Moglia, Cook, & Beale, 2013; Padowski, Gorelick, reliability and resilience measures in the face of a historical drought.
Thompson, Rozelle, & Fendorf, 2015). For example, the popu- The urban water service providers in this case study are intercon-
lar metrics of reliability, resiliency, and vulnerability (Hashimoto, nected not only by hydrologic systems, but also by a representative
Stedinger, & Loucks, 1982; Loucks, 1997) provide an understand- coordinating agent: the Bay Area Water Supply and Conservation
ing of various components of our water supply systems at various Agency.
scales, such as regional hydrologic flows (Ajami, Hornberger, &
Sunding, 2008; Sandoval-Solis, McKinney, & Loucks, 2011), and 2. Methods
local distribution systems (Aydin, Zeckzer, Hagen, & Schmitt, 2015;
Li & Lence, 2007). These and other metrics have been applied in 2.1. Conceptual framework
comprehensive assessments of the resilience of existing systems
when faced with the uncertainties of a range of future scenar- The resilience framework developed in this study integrates a
ios (Milman & Short, 2008; Sullivan, Meigh, & Lawrence, 2006; qualitative and quantitative screening of an urban water system’s
Sandoval-Solis et al., 2011). However, these assessments typically capacity to prepare and adapt to potential stressors based on: the
focus on how robust the engineered or natural systems are with reliability of existing supply portfolios, the capacity to implement
respect to pre-defined standards, rather than their capacity to be demand-side management, and the potential impacts of adapta-
adapted in response to changing needs and priorities of multi- tion strategies on both supply and demand (Fig. 1). The framework
ple water actors and uses (Sivapalan, Savenije, & Bloeschl, 2012; consists of (1) defining indicators and metrics relevant to a given
Sivapalan et al., 2014). Overall, while previous approaches have region’s goals and realities, which are subsequently (2) aggregated
developed useful tools for the analysis of their respective study into a single resilience index. This is followed by (3) an evaluation
goals, they have been tailored to the needs of the conventional of the local and regional characteristics driving the performance
water management perspective, focused on supply-side infrastruc- metrics of water service providers. Finally, (4) we perform a com-
ture and centralized top-down objectives. For example, despite the parative assessment of what these metrics tell us about urban water
growing emphasis on supply diversification as a means to achieve reliability and resilience at the local and regional scales. This sys-
increased reliability and resilience, there is currently no guidance tematic bottom-up approach provides a comparative platform for
on what constitutes a diverse water supply portfolio, or how this utilities to identify their relative strengths and weaknesses with
measure can be used as a guideline to enhance regional resiliency. respect to different metrics of interest, while highlighting oppor-
There is a need for more flexible frameworks that can help identify tunities for collaborative management in the implementation of
adaptation strategies that are feasible and practical in context.
130 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138

2.2. Identification of metrics and assumptions this purpose, we normalize metrics by linearly re-scaling them from
their observed A-B range to a 1–10 range (Eq. (2)), where A and B
Water resource systems are defined not only based on the are the lower and upper bound of the original scale, respectively. In
layout and historic flows of hydrologic systems, but also as a the resulting 1–10 range, each indicator retains its intuitive scale.
function of political decisions, collaborative agreements, and inter- For example, higher values for supply diversity are associated with
dependencies among water providers (Padowski et al., 2015). increased contributions to resilience, whereas higher values for
Consequently, it is important to identify resilience factors on a supply stress have a negative connotation. It is important to notice
region-by-region basis considering on-the-ground realities. Table 1 that this is a comparative index, and should be used as a moving
highlights some baseline metrics related to pre-defined themes. target rather a goal of achieving a certain score. This relative nature
The use of metrics in this framework is intended to provide perspec- means that no single number should be used as a global standard,
tive on the relative performance of water service providers with and is more suited to the bottom-up needs of evolving water sup-
respect to their priorities at any given time. Thus, metrics should be pliers who are all starting from a different reference point, and who
selected to reflect the most pressing aspects affecting water supply may be pursuing different goals.
and demand dynamics in a region of study, and used as a moving
target aligned with the utility’s areas of interest and opportuni- (I − A)
Re − scaled I = 1 + (10 − 1) ∗ (2)
ties as they move forward in their water reliability and resiliency (B − A)
enhancement efforts. Similarly, the possible range of each metric is
defined based on relative performance, and utilities are encouraged The re-scaled indicators are then aggregated into their respective
to improve their performance within those ranges, particularly the theme of supply (S) or demand (D) (Eqs. (3) and (4)), and the two
ones with lower performance values. As utilities make progress, theme components are subsequently aggregated into an overall
the expectations for performance, and thus the range of values for resilience index (Eq. (5)), using the geometric mean. When aggre-
metrics, can be adjusted to become increasingly stringent. gating values, negative indicators are linearly transformed to the
Most of the metrics proposed in this baseline framework have opposite end of the scale in order to follow the normalized 1–10
been defined in the literature or are commonly used in practice range. For example, in Eqs. (3) and (4), the values for supply stress
for urban water planning and management assessments (Juwana and water use are subtracted from 11 (the sum of lower and upper
et al., 2012; Loucks, 1997). The exception is the quantification of the values of the re-scaled range), such that the highest observed value
diversity of water supply and demand portfolios, which, as pointed (10) corresponds to the lowest resilience performance (1) and vice
out in the introduction, has not been yet defined or measured in versa, similar to what previous authors have done with vulnerabil-
the water sector. Water supply diversity is an important indicator ity metrics (Ajami et al., 2008; Juwana et al., 2012; Loucks, 2005;
of reliability and resilience because it demonstrates how flexible a Sandoval-Solis et al., 2011).
supply portfolio is, or the capacity to shift between supply sources We compute the resilience index with and without adapta-
if any particular source becomes compromised. Similarly, demand tion capacity measures to assess their potential effect on supply
diversity indicates flexibility in terms of the capacity of a system and demand indicators. For example, introducing conservation
to incorporate tailored water supplies to meet the different water potential would decrease water use, decrease the percentage of
quality needs of different users, reducing stress on existing high- allocations being used, and potentially shift supply and demand
quality sources. In this analysis we use the Gini-Simpson index diversity. This comparative assessment allows us to identify what
(Simpson, 1949), a measure commonly used in biodiversity, soci- the most effective adaptation strategies are, and which indicators
ology, psychology, economics, and information science (Gibbs & of interest have the most flexibility to be improved.
Martin, 1962; Humprecht & Buechel, 2013; McDonald & Dimmick,
2003; Oksanen et al., 2015; Panniello, Tuzhilin, & Gorgoglione, 1/2
Si = [(11 − supply stressi ) ∗ (supply diversityi )] (3)
2014; van Beers & Zand, 2014) as an indication of the number of
different types (e.g. species, industries, water sources) (n) and their 1/2
relative abundance (computed as the fraction of the total, ranging Di = [(11 − water usei ) ∗ (demand diversityi )] (4)
between 0 and 1) (p) (Eq. (1)). Mathematically, the Gini-Simpson
index indicates the probability that two entities taken at random Local resiliencei = [(Si ) ∗ (Di )]1/2 (5)
from a dataset of interest represent different types (Hurlbert, 1971),
and thus it ranges from 0 (a single type present) to 1 (highly diver- Our index is calculated using the geometric mean to address
sified portfolio). While other tools are available that reflect similar the challenge of assigning specific weights to each criterion. This
measures (e.g. Simpson index, Shannon index, etc.), these have approach has previously been proposed for water resources man-
been found to be generally interchangeable for simple analyses agement (Ajami et al., 2008; Loucks, 1997; Sandoval-Solis et al.,
(Morris et al., 2014), with the Gini-Simpson version being one of the 2011). The multiplicative form of the geometric mean allows us
most popular in the literature. We choose to use the Gini-Simpson to normalize the ranges of performance values being averaged,
index in this framework for its popularity in multiple fields, simple such that a very large performance value can’t negate or substitute
form, easiness of quantification, and intuitive scale. smaller values (Juwana et al., 2012). This form also allows the index
to be easily adapted to different contexts, since additional indica-

n
tors may be added or re-defined within the supply and demand
Gini-Simpson index = 1 − p2i (1)
components before being integrated into the overall index.
i=1
Finally, to identify the impact of local performance on the
regional scale, we propose a weighted average of resilience indices
2.3. Re-scaling and integration of metrics (Eq. (6)), where weights correspond to the fraction of the total
regional demand (D) that is attributed to a utility’s individual
We develop an aggregate resilience index incorporating the demand (di ).
quantitative metrics of interest into a single number for easier
comparison within a region, building upon mathematical princi- d
i
ples from numerical indices in the literature (Ajami et al., 2008; Regional resilience = ∗ (Local resiliencei ) (6)
D
Juwana et al., 2012; Loucks, 2005; Sandoval-Solis et al., 2011). For i
 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138 131

Table 1
Baseline metrics applied to quantitative resilience analysis.

Theme Indicator Example metrics

Supply Supply stress Proportion of available

supplies currently being used
Water quality level
Supply diversity Gini-Simpson index

Demand Water use efficiency Water use per capita in different sectors
Demand diversity Gini-Simpson index

Adaptation capacity Conservation capacity Projected water savings from conservation measures
Augmentation capacity Projected yield of alternative supply projects

Fig. 2. BAWSCA members map.

Source: BAWSCA Annual Survey for Fiscal Year 2014–15.

3. Results and discussion and water use per capita span the range of 4282–344,000 people
and 53.9–324.2 gallons per capita per day, while the average price of
3.1. Study region: Bay Area Water Supply and Conservation water and median household income vary between 0.41–1.16 cents
Agency per gallon and $50,142–$236,528 per year, respectively (United
States Census Bureau, 2013). BAWSCA has the authority to coor-
To demonstrate the practical application of our resilience frame- dinate water conservation, supply and recycling activities for its
work, we apply our methodology to a case study in the San member agencies, acquire water supplies, finance projects, and
Francisco Bay Area. Specifically, we focus on the Bay Area Water build facilities.
Supply and Conservation Agency (BAWSCA), a coordinating agency
representing the needs of 26 water utilities that are dependent
on the San Francisco Regional Water System (SF RWS) (Fig. 2). 3.2. Resilience indicators in BAWSCA
This region provides a good study site for our framework because
it encompasses several water utilities that are autonomous and To better understand how the socio-hydrologic dynamics of the
unique, yet are highly reliant on a common pool of imported and diverse communities in BAWSCA interact in the transition towards
local water supplies, and who have the capacity to participate in a more reliable and resilient regional water system, we applied
coordinated regional efforts through BAWSCA. Furthermore, Cali- our integrative resilience framework. Table 2 defines the metrics
fornia, as the most populated state in the United States and with and assumptions used in this case study, adapted from the baseline
variable precipitation patterns and unevenly distributed water sup- resilience framework methodology (Table 1). For this application,
plies, faces the need to address the long-term reliability of its the metrics related to supply stress are focused on water use from
state-wide urban water resources (CDWR, 2015; CDWR, 2013; the major water supply in the region: the SF RWS. This water sys-
Grantham & Viers, 2014). The state is currently in its fifth year tem is dependent on pristine snowmelt from the Sierra Nevada
of a historical drought that has left many urban water systems Mountains, and requires little treatment before going to people’s
vulnerable and in search of innovative resiliency solutions. homes. Water quality mostly is not a concern in the system and is
BAWSCA’s 26 member agencies serve 1.8 million residents, thus excluded from our case study application of the framework.
businesses, and community organizations representing a diverse Similarly, we adapt the water use efficiency indicator to focus on
group of service areas. For example, based on data from fiscal year water demand in the residential sector as an even platform for
2014–15 (July 2014–June 2015) (BAWSCA, 2015a), population size comparison. A more detailed description of the importance of each
132 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138

Table 2
Assumptions for quantitative metrics used in BAWSCA resilience assessment.

Theme Indicator Metrics and assumptions Original range Re-scaled range

Supply Supply stress Fraction of allocations from 0.1–1.62a 1–10

the SF RWS currently being
used
Supply diversity
  supply from sourcei 2
Gini-Simpson index: 0–0.9 1–10
1− total supply
i

Demand Water use Water demand per capita 20–340 gallons per 1–10
in the residential sector capita per day
Demand diversity
  demand from sectori 2
Gini-Simpson index: 0–0.9 1–10
1− total demand
i

Adaptation capacity Conservation capacity Projections of the potential 0–0.3 (computed into supply
savings achievable through and demand
conservation and water use indicators)
efficiency measures,
computed as a fraction of
total demandb
Augmentation capacity Fraction of total demand 0–0.3 (computed into supply
that may be satisfied by and demand
feasible alternative indicators)
suppliesc
a
The fraction of allocations being used extends beyond 1 because some agencies have historically purchased more water that their contracted allocated amounts. This is
allowed when supplies are available in excess of the total demands from BAWSCA utilities.
b
Projected water savings from feasible demand management programs to be implemented by 2040. Data taken from BAWSCA’s Regional Demand and Conservation
Projections report (BAWSCA, 2014).
c
Projected yield of feasible supply augmentation projects currently under consideration by BAWSCA. Data taken from BAWSCA’s Long-Term Reliable Water Supply Strategy
report (BAWSCA, 2015b). Local projects identified by agencies comprise several recycled water projects and a groundwater project. Regional projects include local capture
and reuse of rainwater, greywater, and stormwater, desalination of bay water and brackish groundwater, and water transfers.

indicator selected, and its relevance to local and regional resilience have permanent contracts or an ISG, and thus show the highest
assessments is discussed in the following sub-sections. supply stress in Fig. 3. Currently, BAWSCA agencies are allowed to
The results of quantifying these relevant metrics for each purchase water beyond their allocated ISG when supplies are avail-
BAWSCA agency are shown in Fig. 3, based on data reported in able in the SF RWS, but water availabilities from the SF RWS could be
the 2014–15 BAWSCA Annual Survey (BAWSCA, 2015a). Each indi- impacted by competing demands, infrastructure damage, climate
cator in the graph reflects re-scaled values according to Eq. (2). change and the ensuing hydrologic conditions, policy decisions, and
The results show that overall, the strongest performance for most regulatory actions (SFPUC, 2013).
BAWSCA utilities is related to low per capita water demand, while
the weakest performance is related to low diversity of supply port-
folios. Furthermore, the adaptation capacity indicators show that
most utilities still envision a high capacity to expand their conser- 3.2.2. Supply diversity
vation programs, and a few utilities are also planning to incorporate Overall, two thirds of BAWSCA agencies’ water supplies come
additional augmentation projects that could provide them consid- from SF RWS. While some utilities rely solely on the SF RWS,
erable local resilience benefits. Analyzing each individual indicator others have strategically diversified their water supply portfolios
allows us to see that BAWSCA member agencies have several oppor- to include additional sources such as groundwater, local surface
tunities to enhance the reliability and resilience of their water water, recycled water, and imported water from the State Water
supplies. Project and Central Valley Project. However, it is important to
note how this diversity is distributed among BAWSCA agencies.
Although several of the larger water agencies (Alameda County
3.2.1. Supply stress Water District, Santa Clara, Milpitas, Sunnyvale, and Daly City) have
To understand the stress on BAWSCA water supplies, it is impor- diverse water supplies (reflected in high scores in Fig. 3), most ser-
tant to understand not only the hydrology but also the governance vice areas in our study region have a small enough demand that
dynamics of the region’s main water source: the SF RWS. Each their imported SF RWS supply is sufficient to cover their present
BAWSCA member agency is subject to an individual water sales needs and growth projections. This lessens the sense of urgency
contract from the SF RWS, which specifies an Individual Supply for these utilities to collaborate or develop alternative supplies.
Guarantee (ISG) for that agency. The permanent ISGs (Fig. 4) were Nevertheless, on an agency-by-agency basis, the long-term relia-
agreed to and adopted by the governing bodies of all of SFPUC’s per- bility of BAWSCA’s traditional supply portfolios is highly uncertain
manent wholesale customers in 1994 and reflected the conditions (BAWSCA, 2015b). The need to diversify water supply portfolios
and known projections at the time, but they may not be adequate is exacerbated by the potential economic consequences of sup-
for present conditions, and water deliveries are subject to cutbacks ply shortfalls. Economic losses to businesses and residential users
in times of shortage (BAWSCA, 2015a). For example, both Purissima under a 20% water supply deficiency on the SF RWS are estimated
Hills Water District and the city of Menlo Park are solely depen- up to $7.7 billion annually (BAWSCA, 2015b). This number could
dent on water from the SF RWS, but Purissima Hills is currently almost double under a supply interruption scenario in the RWS
using up the entirety of its allocated supply, while Menlo Park’s such as an earthquake, with economic loss estimates of up to $14
demand accounts only for 80% of their allocation, indicating Menlo billion in damage costs depending on the extent and location of the
Park could potentially be less vulnerable in their projections for disruption (Brozovic et al., 2007; Khater et al., 1993). Based on the
future needs as shown by the difference in scores between the two potential risks of future shortages, even if BAWSCA member agen-
agencies. Similarly, San Jose MWS and the city of Santa Clara don’t cies have not outgrown their allocated supplies from the SF RWS,
 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138 133

Fig. 3. Resilience indicators and metrics computed for each BAWSCA member agency as of fiscal year 2014–15. Numbers are already scaled such that higher values reflect
better performance with respect to the baseline (see Table 2).

they could considerably enhance their reliability and resilience by shortage, even when the reliability of the supplies themselves is
diversifying their water supply portfolios. uncertain.
Utilities in the BAWSCA service area have maintained a steady
water demand for the past two decades thanks to investments
3.2.3. Water use on conservation and efficiency measures, which have provided
Demand-side characteristics influence the reliability and some added resiliency to population growth and drought in the
resilience of water supplies. Water demand is a result of intrin- region (BAWSCA, 2014). However, water use among some of these
sic water use behaviors in a community, as well as a function of agencies may not be sustainable in the long-term. For exam-
land use, population density, industrial activities, outdoor irriga- ple, Purissima Hills Water District, the Town of Hillsborough, and
tion needs, etc. (Hornberger et al., 2015; Mini, Hogue, & Pincetl, California Water Service-Bear Gulch, some of the higher income
2014; Panagopoulos, 2014). Managing water demands effectively communities in the region, still have demands that exceed 150 gal-
by taking into account these local characteristics can extend avail- lons per capita per day (reflected in high water use scores in Fig. 3).
ability of water supplies and limit the potential impacts of water
134 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138

Fig. 4. (a) Reliance on SF RWS as a percentage of total supplies, (b) distribution of individual supply guarantee among agencies, normalized by population, and (c) unused
supplies (individual supply guarantee minus actual water purchases) normalized by population. Data as of fiscal year 2014–15.

Fig. 5. Linear regression of median household income (x-axis) on residential water use (y-axis). Relationship with water pricing (circle size) and population density (color
scale) is also shown. Insert displays a closer look at the relationship when outliers are removed (corresponding to Hillsborough, Purissima Hills Water District, and CWS-
Bear Gulch).

This may be in part because, even though BAWSCA-agency retail non-potable uses (Hering et al., 2013). These opportunities can
water prices have increased by 58% from 2005 to 2013 (BAWSCA, be leveraged further through more efficient sharing of common
2014), water bills are still only a small percentage of household resources. For example, the city of San Francisco has already estab-
income (less than 1% for the majority of BAWSCA agencies), and lished a program that promotes the development of in-building
therefore many consumers may not be properly aware of the true and district-scale, non-potable water reuse systems (SFPUC, 2015).
value of water in times of shortage (Pacific Institute, 2013). As With this program, San Francisco provides small grants and tech-
shown in Fig. 5, there is a strong relationship between household nical assistance that incentivize new developments to be more
income and water use in the BAWSCA region (R-squared = 0.83), self-reliant and to reduce their toll on high-quality water sup-
which overshadows the correlation between water use and other plies where alternative sources may suffice. In our assessment,
drivers of demand such as population density (R-squared = 0.25) many BAWSCA utilities scored high in the demand diversity metric
and the price of water (R-squared = 0.10). Structuring water rates, (Fig. 3), indicating a capacity for similar programs.
water use efficiency policies, and conservation incentives that take
into account each service area’s diverse characteristics can help
manage water demand as well as provide an opportunity for fund- 3.2.5. Adaptation capacity
ing mechanisms to continue expanding resiliency measures. The adaptation capacity of a utility depends on technical and
financial constraints, as well as the willingness of each service
area to undertake new projects. As changing supply and demand
3.2.4. Demand diversity dynamics affect people’s perspectives on water use, the capacity to
Though BAWSCA serves mostly residential customers, other implement different conservation measures is also likely to evolve.
sectors including commercial, industrial, governmental uses, and For example, because of the ongoing drought, the state of California
dedicated irrigation account for 40% of the total water demand in has issued executive orders to make water conservation a way of
the region. Having a diverse set of customers is important because life extending beyond drought periods, and individual behavioral
using water in different demand sectors offers an opportunity changes have contributed to decreasing water use in response to
to match alternative supply sources of different quality levels to increased drought awareness (SWRCB, 2016). For this study, the
 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138 135

Fig. 6. Change in local resilience index among BAWSCA member agencies. Bars show comparison between scores in fiscal year 2010–11, 2014–15, and potential resilience
improvements with the addition of adaptation measures through conservation and supply augmentation projects.

metrics shown in Fig. 3 quantify the potential to conserve water 2014; Stern, 2010). This type of collaborative management solution
and to augment supplies based on realistic projects self-reported is particularly relevant for BAWSCA given the existing overlap in
by BAWSCA member agencies in recent studies (BAWSCA, 2014; natural resource and political boundaries among agencies, as well
BAWSCA, 2015b). While many utilities may not have the capacity as established coordination layers.
to implement significant adaptation strategies locally (e.g. small
utilities such as Hillsborough have low scores in the adaptation
metrics), there are opportunities for adaptation capacity projects
3.3. Responding to the drought
in neighboring utilities and for regional-scale projects, all of which
would reduce stress on existing supplies and enhance the reliability
3.3.1. Local resilience
and resiliency of utilities small and large.
BAWSCA water agencies, along with the rest of California, have
The diversity of supply and demand characteristics and existing
been struggling with a severe drought from 2012 to 2015 and still
agency interconnections within BAWSCA present several opportu-
ongoing. We applied our resilience framework to evaluate whether
nities for the efficient implementation of additional conservation
recent changes in water management and community behavior
programs, supply augmentation, and targeting these projects to
in response to the drought have resulted in enhanced resiliency
suit different demand sectors as discussed previously. For exam-
and reliability. Fig. 6 shows that overall utilities have made slight
ple, areas with low population density and low cost of land may be
improvements in increasing their resilience metrics from fiscal year
better equipped to implement new water infrastructure projects,
2010–11 (prior to the drought) to 2014–15 (the fourth continuous
while areas with higher water use and faster growth rates are
drought year). The distributions of supply portfolios and demand
the ones in greater need of securing new supplies. However, the
sectors haven’t changed much in this period, and utilities have not
locations of opportunity and need do not necessarily overlap. As
added significant new supply sources. Instead, the improvements
BAWSCA agencies work on diversifying their supply portfolios, the
observed from 2010 to 11 to 2014–15 are reflective of the empha-
existing infrastructure of the SF RWS is an opportunity for shar-
sis on water efficiency and conservation in the region. This progress
ing these supplies, acting as a bank account. For example, many
has not only demonstrated the capacity for demand-side manage-
agencies are starting to outgrow their SF RWS allocations, whereas
ment to create resiliency in the short-term, but also a potential for
others like Palo Alto and Menlo Park have contracts for significantly
lowering demand in the long-term.
more water than they are currently using (excess amounts that
Many utilities also foresee the possibility to continue enhanc-
translate into 84 and 61 unused gallons per capita per day for these
ing the observed resiliency moving forward (2014–15 + adaptation
agencies, respectively) (Fig. 4). This unused water remains in the
in Fig. 6) by continuing to pursue conservation and augmenta-
SF RWS until claimed (assuming 100% level of service). Given this
tion projects as recently reported by BAWSCA (BAWSCA, 2014;
discrepancy, the distribution of water supplies could be more effec-
BAWSCA, 2015a). Potential augmentation projects included in this
tive through the development of flexible market mechanisms that
comparison include several recycled water projects, a groundwa-
take advantage of common supply pools (Anderson, 2015; Grafton,
ter project, local capture and reuse of rainwater, greywater and
Libecap, McGlennon, Landry, & O’Brien, 2011; Palazzo & Brozovic,
stormwater, desalination of bay water and brackish groundwater,
136 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138

Fig. 7. BAWSCA-wide resilience index and supply portfolios from fiscal year 2002–03 to 2014–15.

Fig. 8. Regional resilience indicators under 2014–15 conditions, a local adaptation scenario, and a local + regional adaptation scenario.

and water transfers from other regions, all of which are being con- water supplies and demands. Adaptation capacity scenarios were
sidered as part of BAWSCA’s Long-Term Supply Reliability Strategy. added as either agency-specific, in which case local resilience is
directly enhanced and correspondingly weights into the resilience
of the region, or regional projects meant to benefit BAWSCA as a
3.3.2. Regional resilience whole. Fig. 7 shows the performance of BAWSCA as a whole from
We explored the regional impacts of utilities’ individual indica- fiscal year 2002–03 to 2014–15. The time series shows a steady,
tor performance by computing a BAWSCA-wide resilience index. though very slow, increase in resilience over this period attributed
Water utilities with low resilience scores are at high risk in terms mainly to water conservation and efficiency measures, as supply
of growth and the challenges of water shortage. While these utili- and demand portfolios have not changed significantly. This trend
ties could improve their performance through independent supply is particularly visible in the change between 2013 and 14 and
augmentation and demand management efforts as shown in Fig. 6, 2014–15, when water agencies in California started implementing
higher reliability and resilience could also be achieved through significant conservation measures in response to the drought.
collaborations with neighboring utilities. For example, while the Using our framework, we compare this business-as-usual
Town of Hillsborough has one of the lowest resilience indices in our scenario to alternative cases that include adaptation capacity mea-
framework, it only accounts for 1.5% of the total water use among sures (Fig. 8). There are clear performance benefits of including
the BAWSCA agencies, giving it some flexibility to share alterna- both local and regional supply augmentation and diversification
tive supply projects with other small utilities or to negotiate small projects, all of which can help reduce stress on existing supply
portions of a larger project. sources and thus benefit agencies that rely on the same common
For the purposes of regional integration, each component was pool of water resources. The regional resilience index has the poten-
calculated as a weighted average of individual utilities, with tial to increase if all local and regional projects continue in the
weights corresponding to water use. This reflects the fact that each pipeline, while individual utilities could not reach this the same
water service provider acts as a small piece of a regional pool of
 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138 137

level of benefits independently. Moreover, the comparison of sce- Carr, Michael Hurley, and Andree Johnson for providing us with
narios in Fig. 8 allows us to visualize how sensitive each indicator is constructive insights throughout this research.
to changes in local and regional adaptation strategies. In this case,
the metrics of supply stress and supply diversity show the highest
sensitivity to additional adaptation capacity measures, reflective of
References
the high dependency of the region on a single supply source and
the considerable potential for enhancing resilience by focusing on Ajami, N. K., Hornberger, G. M., & Sunding, D. L. (2008). Sustainable water resource
these measures. management under hydrological uncertainty. Water Resources Research, 44,
These numbers are relative to the indicators and assumptions W11406. http://dx.doi.org/10.1029/2007wr006736
Anderson, T. L. (2015). Dynamic markets for dynamic environments: The case for
of this case study, and should be used only as a snapshot in time, to
water marketing. Daedalus, 144, 83–93.
assess intra-regional decisions. The consideration of all these com- Arlosoroff, S. (2007). Wastewater management, treatment, and reuse in Israel,
ponents in a systematic framework provides important insights wastewater reused—Risk Assessment. Decision-Making and Environmental
Security, 55–64. http://dx.doi.org/10.1007/978-1-4020-6027-4 6
for the region, integrating information available into the decision-
Aydin, N. Y., Zeckzer, D., Hagen, H., & Schmitt, T. (2015). A decision support system
making process at a level that decision-makers can understand, and for the technical sustainability assessment of water distribution systems.
preventing the loss of any relevant information. As the region pro- Environmental Modelling & Software, 67, 31–42. http://dx.doi.org/10.1016/j.
gresses towards more reliable supplies and the priorities shift, the envsoft.2015.01.006
BAWSCA. (2014). Regional water demand and conservation projections, Bay Area
baseline metrics used for initial assessment may be adjusted and Water Supply and Conservation Agency.. Web site, available at: http://www.
re-scaled accordingly. bawsca.org/docs/
BAWSCA%20Demand%20and%20Conservation%20Projection%20FINAL%20REPORT.
pdf
4. Conclusions BAWSCA. (2015a). Annual Survey FY 2014-15, Bay Area Water Supply and
Conservation Agency.. Web site, available at: http://www.bawsca.org/docs/
BAWSCA FY14-15 AnnualSurvey.pdf
Water shortages are a prevalent risk in many parts of the world BAWSCA. (2015b). Long-Term Reliable Water Supply Strategy Phase II- Final Report.
as a result of a changing climate, rapidly growing populations, Bay Area Water Supply and Conservation Agency.. Web site, available at: http://
bawsca.org/water-supply/long-term-reliable-water-supply-strategy/
and over-reliance on imported supplies. Regional coordination, Beh, E. H. Y., Dandy, G. C., Maier, H. R., & Paton, F. L. (2014). Optimal sequencing of
defined by common socio-hydrological water issues and opportu- water supply options at the regional scale incorporating alternative water
nities as exemplified by our framework, can greatly contribute to supply sources and multiple objectives. Environmental Modelling & Software,
53, 137–153.
better resource management, promote supply diversification, and
Beh, E. H. Y., Maier, H. R., & Dandy, G. C. (2015). Adaptive, multiobjective optimal
help address the risks associated with water shortages. This work sequencing approach for urban water supply augmentation under deep
develops a generalized socio-hydrologic framework that can pro- uncertainty. Water Resources Research, 51, 1529–1551. http://dx.doi.org/10.
1002/2014wr016254
vide a platform for more holistic bottom-up assessments of urban
Brown, R. R., & Farrelly, M. A. (2009). Delivering sustainable urban water
water systems as these move from centralized to decentralized management: A review of the hurdles we face. Water Science and Technology,
approaches. In this way, the framework can help decision-makers 59, 839–846. http://dx.doi.org/10.2166/wst.2009.028
identify and evaluate solutions in order to enhance reliability and Brown, C. M., Lund, J. R., Cai, X., Reed, P. M., Zagona, E. A., Ostfeld, A., . . . & Brekke, L.
(2015). The future of water resources systems analysis: Toward a scientific
resilience, given the diversity of local and regional characteristics, framework for sustainable water management. Water Resources Research, 51,
criteria, and preferences 6110–6124.
The application of our integrative framework to the case study of Brozovic, N., Sunding, D. L., & Zilberman, D. (2007). Estimating business and
residential water supply interruption losses from catastrophic events. Water
the Bay Area Water Supply and Conservation Agency demonstrates Resources Research, 43, W08423. http://dx.doi.org/10.1029/2005wr004782
how, rather than being competitors for a limited resource, intercon- CDWR. (2013). California water plan update. Volume 1—The strategic plan.
nected utilities have the opportunity to be partners in a dynamic Sacramento, CA: California Department of Water Resources. Web site,
available at: http://www.waterplan.water.ca.gov/docs/cwpu2013/Final/0a-
system of adaptation capacity building. While our framework helps Vol1-full2.pdf
assess the opportunities and potential impact of adaptation strate- CDWR. (2015). California’s most significant droughts: Comparing historical and recent
gies, future work should also consider how to best guide the conditions, report.. Available at: http://www.water.ca.gov/waterconditions/
docs/California Signficant Droughts 2015 small.pdf
selection of these strategies among available options to make sure
Cayan, D. R., Das, T., Pierce, D. W., Barnett, T. P., Tyree, M., & Gershunov, A. (2010).
the strategies are feasible and sustainable in practice. For exam- Future dryness in the southwest US and the hydrology of the early 21 st
ple, it is important to have a more diversified supply portfolio, but century drought. Proceedings of the National Academy of Sciences of USA, 107,
21271–21276.
also to account for the cost, drought-proof, interdependency, and
Chartres, C., & Williams, J. (2006). Can Australia overcome its water scarcity
limitations of different supply sources. This study introduces an problems? Journal of Developments in Sustainable Agriculture, 1, 17–24.
innovative approach to measure supply diversity, bringing the Gini- Connell-Buck, C. R., Medellin-Azuara, J., Lund, J. R., & Madani, K. (2011). Adapting
Simpson index, commonly used in diversity studies in other sectors, California’s water system to warm vs. dry climates. Climatic Change, 109,
S133–S149. http://dx.doi.org/10.1007/s10584-011-0302-7
to the field of water management. This metric can be used as a Diffenbaugh, N. S., Swain, D. L., & Touma, D. (2015). Anthropogenic warming has
first step in evaluating the resiliency of a region and the impacts of increased drought risk in California. Proceedings of the National Academy of
potential adaptation scenarios. Similar innovative metrics could be Sciences USA, 112, 3931–3936.
Ferguson, B. C., Brown, R. R., Frantzeskaki, N., de Haan, F. J., & Deletic, A. (2013). The
further adapted and added to our proposed framework to support enabling institutional context for integrated water management: Lessons from
decision-makers in the transition to a more holistic water man- Melbourne. Water Research, 47, 7300–7314. http://dx.doi.org/10.1016/j.watres.
agement paradigm. More importantly, emerging decision-support 2013.09.045
Gibbs, J. P., & Martin, W. T. (1962). Urbanization, technology, and the division of
tools should take into account bottom-up perspectives as identi- labor: international patterns. American Sociological Review, 27, 667–677.
fied in this work, and enable knowledge-sharing and collaboration Grafton, R. Q., Libecap, G., McGlennon, S., Landry, C., & O’Brien, B. (2011). An
in the pursue of more reliable and resilient water supplies. integrated assessment of water markets: A cross-country comparison. Review
of Environmental Economics and Policy Journal, 5, 219–239.
Grantham, T. E., & Viers, J. H. (2014). 100 years of California’s water rights system:
Patterns, trends and uncertainty. Environmental Research Letters, 9, 084012.
Acknowledgements
http://dx.doi.org/10.1088/1748-9326/9/8/084012
Graymore, M. L. M., Sipe, N. G., & Rickson, R. E. (2010). Sustaining human carrying
This work was supported by the National Science Foundation capacity: A tool for regional sustainability assessment. Ecological Economics, 69,
(NSF) Engineering Research Center for Reinventing the Nation’s 459–468. http://dx.doi.org/10.1016/j.ecolecon.2009.08.016
Hashimoto, T., Stedinger, J. R., & Loucks, D. P. (1982). Reliability, resiliency, and
Urban Water Infrastructure (ReNUWIt) EEC-1028968, and an NSF vulnerability criteria for water-resource system performance evaluation. Water
Graduate Research Fellowship. We would like to thank Adrianne Resources Research, 18, 14–20. http://dx.doi.org/10.1029/WR018i001p00014
138 P. Gonzales, N.K. Ajami / Sustainable Cities and Society 30 (2017) 128–138

Hering, J. G., Waite, T. D., Luthy, R. G., Drewes, J. E., & Sedlak, D. L. (2013). A Padowski, J. C., & Gorelick, S. M. (2014). Global analysis of urban surface water
changing framework for urban water systems. Environmental Science & supply vulnerability. Environmental Research Letters, 9 http://dx.doi.org/10.
Technology, 47, 10721–10726. 1088/1748-9326/9/10/104004
Hornberger, G. M., Hess, D. J., & Gilligan, J. (2015). Water conservation and Padowski, J. C., & Jawitz, J. W. (2009). The future of global water scarcity: Ppolicy
hydrological transitions in cities in the United States. Water Resources Research, and management challenges and opportunities. The Whitehead Journal of
51, 4635–4649. Diplomacy and International Relations, 10, 99–114.
Hughes, S., & Pincetl, S. (2014). Evaluating collaborative institutions in context: The Padowski, J. C., Gorelick, S. M., Thompson, B. H., Rozelle, S., & Fendorf, S. (2015).
case of regional water management in southern California. Environment and Assessment of human-natural system characteristics influencing global
Planning C, 32, 20–38. freshwater supply vulnerability. Environmental Research Letters, 10 http://dx.
Hughes, S., Pincetl, S., & Boone, C. (2013). Triple exposure: Regulatory, climatic, and doi.org/10.1088/1748-9326/10/10/104014
political drivers of water management changes in the city of Los Angeles. Cities, Palazzo, A., & Brozovic, N. (2014). The role of groundwater trading in spatial water
32, 51–59. http://dx.doi.org/10.1016/j.cities.2013.02.007 management. Agricultural Water Management, 145, 50–60.
Hughes, S. (2013). Authority structures and service reform in multilevel urban Panagopoulos, G. P. (2014). Assessing the impacts of socio-economic and
governance: The case of wastewater recycling in california and Australia. Urban hydrological factors on urban water demand: S multivariate statistical
Affairs Review, 49, 381–407. http://dx.doi.org/10.1177/1078087412458762 approach. Journal of Hydrology, 518, 42–48.
Humprecht, E., & Buechel, F. (2013). More of the same or marketplace of opinions? Panniello, U., Tuzhilin, A., & Gorgoglione, M. (2014). Comparing context-aware
A cross-national comparison of diversity in online news reporting. recommender systems in terms of accuracy and diversity. User Modeling and
International Journal of Press-Politics, 18, 436–461. http://dx.doi.org/10.1177/ User-Adapted Interaction, 24, 35–65. http://dx.doi.org/10.1007/s11257-012-
1940161213497595 9135-y
Hurlbert, S. H. (1971). Nonconcept of species diversity—Critique and alternative Paton, F. L., Dandy, G. C., & Maier, H. R. (2014). Integrated framework for assessing
parameters. Ecology, 52, 577–586. http://dx.doi.org/10.2307/1934145 urban water supply security of systems with non-traditional sources under
Juwana, I., Muttil, N., & Perera, B. J. C. (2012). Indicator-based water sustainability climate change. Environmental Modelling & Software, 60, 302–319.
assessment—A review. The Science of the Total Environment, 438, 357–371. Rijke, J., Farrelly, M., Brown, R., & Zevenbergen, C. (2013). Configuring
Kauffman, G. J. (2002). What if. . . the United States of America were based on transformative governance to enhance resilient urban water systems.
watersheds? Water Policy, 4, 57–68. Environmental Science & Policy, 25, 62–72. http://dx.doi.org/10.1016/j.envsci.
Khater, M., Scawthorn, C., & Rojahn, C. (1993). A model methodology for 2012.09.012
assessment of seismic vulnerability and impact of disruption of water supply SFPUC. (2013). Water availability study for the City and County of San Francisco. San
systems. In K. J. Tierney, & N. J. M. Cent (Eds.), Earthquake hazard reduction in Francisco Public Utilities Commission.
the Central and Eastern Unites States: A time for examination and action SFPUC. (2015). Non-Potable Water Program, San Francisco Public Utilities
(Socioeconomic impacts: 1993 National Earthquake Conference) (pp. 605–614). Commission.. Web site, available at: http://www.sfwater.org/index.
Memphis, Tenn: U.S. Earthquake Consortium. aspx?page=686
Li, Y., & Lence, B. J. (2007). Estimating resilience for water resources systems. Water Sandoval-Solis, S., McKinney, D. C., & Loucks, D. P. (2011). Sustainability index for
Resources Research, 43 http://dx.doi.org/10.1029/2006wr005636 water resources planning and management. Journal of Water Resources
Loucks, D. P. (1997). Quantifying trends in system sustainability. Hydrological Planning and Management, 137, 381–390.
Sciences Journal, 42, 513–530. Simpson, E. H. (1949). Measurement of diversity. Nature, 163, 688–688.
Loucks, D. P. (2005). Decision support systems for drought management. In A. Vela, Sivapalan, M., Savenije, H. H. G., & Bloeschl, G. (2012). Socio-hydrology: A new
G. Rossi, J. Andreu, & F. Vagliasindi (Eds.), Drought management and planning for science of people and water. Hydrological Processes, 26, 1270–1276.
water resources (pp. 119–131). CRC Press. Sivapalan, M., Konar, M., Srinivasan, V., Chhatre, A., Wutich, A., Scott, C. A., . . . &
Luthy, R. G., & Sedlak, D. L. (2015). Urban water-supply reinvention. Daedalus, 144, Rodriguez-Iturbe, I. (2014). Socio-hydrology: Use-inspired water sustainability
72–82. science for the anthropocene. Earth’s Future, 2, 225–230.
Marlow, D. R., Moglia, M., Cook, S., & Beale, D. J. (2013). Towards sustainable urban Srinivasan, V., Lambin, E. F., Gorelick, S. M., Thompson, B. H., & Rozelle, S. (2012).
water management: A critical reassessment. Water Research, 47, 7150–7161. The nature and causes of the global water crisis: Syndromes from a
http://dx.doi.org/10.1016/j.watres.2013.07.046 meta-analysis of coupled human-water studies. Water Resources Research, 48,
McDonald, D. G., & Dimmick, J. (2003). The conceptualization and measurement of W10515. http://dx.doi.org/10.1029/2011wr011087
diversity. Communication Research, 30, 60–79. http://dx.doi.org/10.1177/ Stern, J. (2010). Introducing competition into England and Wales water
0093650202239026 industry—Lessons from UK and EU energy market liberalisation. Utilities Policy,
Milman, A., & Short, A. (2008). Incorporating resilience into sustainability 18, 120–128.
indicators: An example for the urban water sector. Global Environmental Sullivan, C., Meigh, J., & Lawrence, P. (2006). Application of the water poverty index
Change, 18, 758–767. at different scales: A cautionary tale. Water International, 31, 412–426.
Mini, C., Hogue, T. S., & Pincetl, S. (2014). Patterns and controlling factors of SWRCB. (2016). State water resources control board water conservation portal
residential water use in Los Angeles, California. Water Policy, 16, 1054–1069. –Emergency conservation regulation. State Water Resources Control Board. Web
Mini, C., Hogue, T. S., & Pincetl, S. (2015). The effectiveness of water conservation site, available at: http://www.waterboards.ca.gov/water issues/programs/
measures on summer residential water use in Los Angeles, California. conservation portal/emergency regulation.shtml
Resources, Conservation and Recycling, 94, 136–145. Tal, A. (2006). Seeking sustainability: Israel’s evolving water management strategy.
Moglia, M., Alexander, K. S., & Sharma, A. (2011). Discussion of the enabling Science, 313, 1081–1084. http://dx.doi.org/10.1126/science.1126011
environments for decentralised water systems. Water Science and Technology, Tarroja, B., AghaKouchak, A., Sobhani, R., Feldman, D., Jiang, S., & Samuelsen, S.
63, 2331–2339. http://dx.doi.org/10.2166/wst.2011.443 (2014). Evaluating options for Balancing the water-electricity nexus in
Morris, E. K., Caruso, T., Buscot, F., Fischer, M., Hancock, C., Maier, T. S., . . . & Rillig, California: Part 1-securing water availability. The Science of the Total
M. C. (2014). Choosing and using diversity indices: Insights for ecological Environment, 497, 697–710.
applications from the German Biodiversity Exploratories. Ecology and United States Census Bureau. (2013). American community survey.. Web site,
Evolution, 4, 3514–3524. http://dx.doi.org/10.1002/ece3.1155 available at: http://www.census.gov/
Nelson, R., Howden, M., & Smith, M. S. (2008). Using adaptive governance to van Beers, C., & Zand, F. (2014). R&D cooperation, partner diversity, and innovation
rethink the way science supports Australian drought policy. Environmental performance: An empirical analysis. Journal of Product Innovation Management,
Science & Policy, 11, 588–601. http://dx.doi.org/10.1016/j.envsci.2008.06.005 31, 292–312. http://dx.doi.org/10.1111/jpim.12096
Newman, J. P., Dandy, G. C., & Maier, H. R. (2014). Multiobjective optimization of Viviroli, D., Archer, D. R., Buytaert, W., Fowler, H. J., Greenwood, G. B., Hamlet, A. F.,
cluster-scale urban water systems investigating alternative water sources and . . . & Woods, R. (2011). Climate change and mountain water resources:
level of decentralization. Water Resources Research, 50, 7915–7938. Overview and recommendations for research, management and policy.
Oksanen, J., Bergman, C., Sainio, J., & Westerholm, J. (2015). Methods for deriving Hydrology and Earth System Science, 15, 471–504. http://dx.doi.org/10.5194/
and calibrating privacy-preserving heat maps from mobile sports tracking hess-15-471-2011
application data. Journal of Transport Geography, 48, 135–144. http://dx.doi.org/ Winz, I., Trowsdale, S., & Brierley, G. (2014). Understanding barrier interactions to
10.1016/j.jtrangeo.2015.09.001 support the implementation of sustainable urban water management. Urban
Pacific Institute. (2013). Water rates: Water affordability, report.. Available at: http:// Water Journal, 11, 497–505. http://dx.doi.org/10.1080/1573062x.2013.832777
www.pacinst.org/wp-content/uploads/2013/01/water-rates-affordability.pdf