The Complexity of Urban Water Resources Management

Material Information

The Complexity of Urban Water Resources Management Water Availability and Vulnerability for Large Cities in the United States
Padowski, Julie C
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (216 p.)

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Soil and Water Science
Committee Chair:
Jawitz, James W
Committee Members:
Clark, Mark W
Borisova, Tatiana
Berg, Sanford V
Graduation Date:


Subjects / Keywords:
Available water content ( jstor )
Cities ( jstor )
Distance functions ( jstor )
Groundwater ( jstor )
Natural reservoirs ( jstor )
Urban areas ( jstor )
Water management ( jstor )
Water resources ( jstor )
Water supply ( jstor )
Water utilities ( jstor )
Soil and Water Science -- Dissertations, Academic -- UF
availability -- collaboration -- complexity -- management -- urban -- vulnerability -- water
Tampa Bay ( local )
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Soil and Water Science thesis, Ph.D.


Over the past century, the effects of urbanization in the United States have been profound; urban populations have come to represent the majority of the population, and because of this change we have not only drastically altered the way in which we view and use resources, but how we function as a society. To thrive, urban centers require constant inputs of resources, such as water. However, over the past fifty years, rapidly increasing urban water demands have strained water supplies to the point where serious concerns over urban water scarcity now exist. Despite these concerns, there has been relatively little progress made towards better understanding urban water availability, vulnerability and management at the national level. To address this lack of information, the work presented here collected information on all urban areas throughout the coterminous U.S. with populations greater than 100,000 (n=255) to identify 1) the current state of water availability and vulnerability in urban areas, 2) the relationship between water availability and urban water management responses to stress, 3) the role of complexity in urban utility operations and finances, 4) collaborative participation in urban water management as well as 5) the emergence of complexity over time in urban water systems. Results from this work revealed that 1) approximately 25% of the US population resides in an urban area where low average annual water availability is an issue. Water vulnerability analyses, however, showed that over half of the population experiences at least a moderate level of vulnerability, either due to source inflow variability, or poor water availability. In locations where water availability was low, 2) a positive correlation was found between availability and both the complexity of the water supply infrastructure and water supply management. In addition, research revealed that urban areas who shared sources also had higher levels of infrastructure and management complexity. When assessing 3) costs within the urban water complexity paradigm, results indicated that water provision processes scale across urban utility population size with economies of scale. Using a combined metric to measure total operational complexity of each utility, results showed that in general, the level of complexity was higher than expected, and that higher complexity was correlated to higher overall costs. When utilities were assessed for their 5) level of collaborative participation, results indicated that over the past five years, collaborative participation has risen within urban utilities, and has coincided with a shift away from supply-based management towards a more balanced use of supply- and demand-based management strategies as well as an increase in the importance utilities place on water management plans. Finally, 6) when urban water complexity was examined over time, distinct patterns and trends emerged between disparate case studies. Similar threads of complexity are thought to run though all urban water management systems. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (Ph.D.)--University of Florida, 2011.
Adviser: Jawitz, James W.
Electronic Access:
Statement of Responsibility:
by Julie C Padowski.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Padowski, Julie C. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
Resource Identifier:
818025613 ( OCLC )
LD1780 2011 ( lcc )


This item has the following downloads:

Full Text




2 2011 Julie C. Padowski


3 2010) who inspired generations of students to care about water resources management


4 ACKNOWLEDGMENTS This work would not have been possible without funding and support from the National Water Research Institute ( NWRI ) the Adaptive Management of Water, Wetlands and Watersheds Integrative Graduate Education and Research Traineeship (IGERT) and My sincere thanks goes to my committee for participating in this interdisciplinary project and for all the valuable insights experience, and comments they have given me over the years However, Dr. Jame Jawitz deserves special acknowledgments for his unwavering support of this work from its inception, for the countless hours he committed to guiding and helping me during this time, and for all the excellent advice he has provided. Additional thanks is extended to the water systems representatives who devoted their valuable time to searching through records and completing our questionnaires, the UF Food and Resource Economics Department for access to the su rvey software and the UF Survey Research Center for assisting with survey implementation Thanks is also given to Gareth Lagerwell for his programming expertise to my friends and family for their support, and to our three cats for their constant warmth and attentive companionship during the writing of this manuscript Finally, this dissertation would not have happened without the wonderful encouragement and tireless support provided by Stuart and Jeannie whom have suffered and celebrated though much wi th me. I cannot thank you enough.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 BACKGROUND AND SIGNIFICANCE ................................ ................................ ... 14 The Shifting Management Paradigm ................................ ................................ ...... 14 Contextualizing Urban Water Management ................................ ...................... 14 Integrated Water Resources Management (IWRM) ................................ ......... 16 Issues of Complex ity ................................ ................................ ........................ 18 The State of Urban Water Management: A National Perspective ........................... 19 Water Availability and Vulnerability ................................ ................................ .. 20 Institutions and Infrastructure ................................ ................................ ........... 21 Cost of Water Provision ................................ ................................ .................... 22 Collaborative Participati on ................................ ................................ ................ 23 System Complexity Over Time ................................ ................................ ......... 24 Research Goals and Objectives ................................ ................................ ............. 25 2 A STORAGE BASED APPROACH FOR ASSESSING URBAN WATER AVAILABILITY AND VULNERABILITY ................................ ................................ ... 26 Introduction ................................ ................................ ................................ ............. 26 Current Methods for Meas uring Water Availability And Vulnerability ...................... 28 Estimating Water Availability Using A Storage based Approach ............................ 33 Calculating Water Av ailability and Vulnerability ................................ ................ 34 Data Acquisition and Assumptions ................................ ................................ ... 36 Urban area ................................ ................................ ................................ 37 Urban supply and demand ................................ ................................ ......... 38 Streamflow ................................ ................................ ................................ 38 Reservoir storages ................................ ................................ ..................... 39 Natural lake storages ................................ ................................ ................. 42 Groundwater recharge and storage ................................ ........................... 43 Water Availability Assessments ................................ ................................ ........ 45 Water Vulnerability Assessments ................................ ................................ ..... 47 Conclusions ................................ ................................ ................................ ............ 51 3 URBAN ADAPTATIONS TO WATER SCARCITY: MA NAGEMENT AND INFRASTRUCTURE ................................ ................................ ............................... 66 Introduction ................................ ................................ ................................ ............. 66


6 Management Complexity ................................ ................................ ........................ 69 Methods ................................ ................................ ................................ .................. 72 Power Law Scaling Analyses ................................ ................................ ........... 72 Urban Water Availability ................................ ................................ ................... 74 Infrastructure Metrics ................................ ................................ ........................ 75 Organization Metrics ................................ ................................ ........................ 77 Institutional frameworks ................................ ................................ ............. 77 Governance frameworks ................................ ................................ ............ 78 Adaptation Strategies ................................ ................................ ....................... 79 Results and Discussion ................................ ................................ ........................... 80 Water Availability ................................ ................................ .............................. 80 Power Law Scaling Analyses ................................ ................................ ........... 81 Infrastructural elaboration ................................ ................................ .......... 81 Organizational elaboration ................................ ................................ ......... 83 Adaptation Strategies ................................ ................................ ....................... 85 Conclusions ................................ ................................ ................................ ............ 87 4 COMPLEXITY AND COSTS IN US URBAN WATER UTILITY SYSTEMS ............. 98 Introduction ................................ ................................ ................................ ............. 98 Scaling Relationships in Water Provision Processes ................................ ..... 100 Costs and Complexities ................................ ................................ .................. 104 Discussion ................................ ................................ ................................ ............ 106 5 COLLABORATIVE PARTICIPATION AND WATER MANAGEMENT IN US URBAN WATER UTILITIES ................................ ................................ .................. 116 Introduction ................................ ................................ ................................ ........... 116 Methods ................................ ................................ ................................ ................ 119 Survey Design ................................ ................................ ................................ 119 Survey Content ................................ ................................ ............................... 120 Results ................................ ................................ ................................ .................. 122 Respondent Characteristics ................................ ................................ ........... 122 Utility Management ................................ ................................ ......................... 123 Management planning ................................ ................................ ............. 123 Management strategies ................................ ................................ ........... 124 Collaborative Participation ................................ ................................ .............. 126 Knowledge ex change at conferences ................................ ...................... 126 Knowledge exchange through research ................................ ................... 127 Interactions with other institutions ................................ ............................ 127 Participation within trade organizations ................................ .................... 129 Participation in group based source water management ......................... 130 Collaboration and Management ................................ ................................ ..... 130 Overall management metric ................................ ................................ ..... 131 Overall collaboration metric ................................ ................................ ...... 131 Conclusions and Management Implications ................................ .......................... 133 Planning and Management ................................ ................................ ............. 133


7 Collaborative Par ticipation ................................ ................................ .............. 135 6 MANAGING WATER: THE ROLE OF COMPLEXITY IN AN EVOLVING PARADIGM THE US URBAN WATER EXPERIENCE ................................ ........ 146 Introduction ................................ ................................ ................................ ........... 146 Integrated Urban Water Management and Complexity ................................ ......... 147 A Complexity Framework For IUWM ................................ .............................. 151 Elaboration ................................ ................................ ............................... 151 Complication ................................ ................................ ............................ 151 Redefinition ................................ ................................ .............................. 152 Common Threads of Complexity ................................ ................................ .... 154 Case Study Descriptions ................................ ................................ ....................... 155 Tampa Bay Region, FL ................................ ................................ .................. 156 Current context ................................ ................................ ........................ 156 Management history ................................ ................................ ................ 156 Los Angeles Region, CA ................................ ................................ ................ 158 Current context ................................ ................................ ........................ 158 Management history ................................ ................................ ................ 159 Washington D.C. ................................ ................................ ............................ 160 Current context ................................ ................................ ........................ 160 Management history ................................ ................................ ................ 161 Elaboration ................................ ................................ ................................ ..... 162 Complications ................................ ................................ ................................ 164 Redefinition ................................ ................................ ................................ .... 165 Discussion ................................ ................................ ................................ ............ 166 Conclusions and Implications ................................ ................................ ............... 169 7 CONCLUDING REMARKS ................................ ................................ ................... 175 Chapter 2: Creation and implementation of a storage based water availabi lity and vulnerability metric for assessing US urban areas ................................ ...... 176 Chapter 3: Assessment of the relationship between urban water availability and urban water management in terms of the complexity o f the infrastructure and management organization ................................ ................................ ................. 177 Chapter 4: Determination of the extent to which financial and operational water provision metrics fit into the urban complexity paradigm ................................ ... 179 Chapter 5: Measurement of the degree to which participatory collaboration is being used by urban water utilities to make decisions about water management and planning ................................ ................................ ................ 181 Chapter 6: Reassessment of the historic role of complexity in the development of urban water management ................................ ................................ .............. 182 Final Thoughts ................................ ................................ ................................ ...... 183 APPENDIX A LIST OF URBAN AREAS ................................ ................................ ...................... 186


8 B URBAN WATER UTILITY COLLABORATION AND MANAGEMENT SURVEY ... 193 LIST OF REFERENCES ................................ ................................ ............................. 203 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 216


9 LIST OF TABLES Table page 2 1 Type s and sources of data ................................ ................................ ................. 54 2 2 Observed reservoir and streamgage information ................................ ................ 55 2 3 Water availability estimates risk of water sc arcity ................................ .............. 56 2 4 Water vulnerability of urban areas ................................ ................................ ...... 58 3 1 Water Acquisition and Management (WAM) Scale ................................ ............. 90 5 1 Survey respondents by US region. ................................ ................................ ... 138 5 2 Utility characteristics of US regions ................................ ................................ .. 138 5 3 Management plan requirements by region ................................ ....................... 138 5 4 Average importance of water resource planning components by region. ......... 139 5 5 Su pply and demand based general management strategies options .............. 139 5 6 Fraction of utilities using supply vs. demand based strategies by region. ........ 139 5 7 Conference attendance by region. ................................ ................................ .... 140 5 8 Fraction of utilities participating in studies and surveys by region. ................... 140 5 9 and planning ................................ ................................ ................................ ..... 140 5 10 Number and importance of interactions by region ................................ ............ 141 5 11 Involvement in trade organizations by region. ................................ .................. 141 5 12 Average group involvement characteristics by region ................................ ...... 141


10 LI ST OF FIGURES Figure page 2 1 Location of urban areas and associated water resources. ................................ 60 2 2 Conceptua l diagram describing local vs. captured water sources. ..................... 60 2 3 Relationship between City, Urban Area and Metropolitan Statistical Area populations and urban utility service populations. ................................ .............. 61 2 4 River discharge stream order relationship ................................ .......................... 62 2 5 Location of water resource regions. ................................ ................................ .... 62 2 6 Differences in reservoir inflows using two methods. ................................ ........... 63 2 7 Water availability estimates using a storage based or MAR base d approach. ... 64 2 8 Water vulnerability as a function of inflow varia bility and water availability ........ 65 2 9 Text analysis results for each level of vulnerability. ................................ ............ 65 3 1 Relationship between infrastructural elaboration and organizational elaboration. ................................ ................................ ................................ ......... 91 3 2 Power law relationship of three theoretical parameters and the scaling dynamics of different values. ................................ ................................ ........... 91 3 3 Map of local and captured water availabil ity for each urban area in the coterminous US with a population greater than 100,000. ................................ ... 92 3 4 Local and total water availability measurements for US urban areas. ................ 93 3 5 Elaboration of infrastructure represented by average distance to sources ......... 94 3 6 Percentage of urban areas that occur at each lev el of organizational elaboration. ................................ ................................ ................................ ......... 95 3 7 WAM Scale Percent of urban areas at each level of institutional elaboration. ... 96 3 8 Map of urban areas and overlapping buffers. ................................ ..................... 96 3 9 Proximity issue correlations. ................................ ................................ ............... 97 3 10 Relation ship between governance and institutional elaboration. ........................ 97 4 1 .............................. 110


11 4 2 Power law scaling relationships between urban water service population size and capital expenses. ................................ ................................ ....................... 111 4 3 Power law scaling relationships between urban water service p opulation size and system revenues. ................................ ................................ ...................... 112 4 4 Power law scaling relationships between urban water service population size and finances ................................ ................................ ................................ .... 112 4 5 Power law scaling relationships between urban water service population size and physical water provision ................................ ................................ ............ 113 4 6 Power law scaling relationships between urban water service population size and the price of water. ................................ ................................ ...................... 113 4 7 Total operational complexity and per capita expense ................................ ....... 114 4 8 Total relative complexity and per capita expense ................................ ............. 115 5 1 Boundaries of US regions and locations of urban areas surveyed .................. 142 5 2 Average level to which utilities address various components of their water management plans ................................ ................................ ........................... 142 5 3 Management strategies used by utilities today (current) and five years ago (historic) ................................ ................................ ................................ ............ 143 5 4 Relative reliance on supply and demand based management strategies ...... 143 5 5 Change in utility preference of supply based vs. demand based strategies over a five year period. ................................ ................................ ..................... 144 5 6 Percent of water supply management groups that listed water related institutions are participating in. ................................ ................................ ......... 144 5 7 Relationship between overall water utility management, coll aboration and water availabilit y ................................ ................................ .............................. 145 6 1 Conceptual diagram of IWRM components and their interactions. ................... 170 6 2 Complexity framework For IUWM ................................ ................................ .... 170 6 3 Tampa timelines. ................................ ................................ .............................. 171 6 4 Washington DC timelines ................................ ................................ ................. 172 6 5 Los Angeles timelines ................................ ................................ ....................... 173 6 6 Revised conceptual diagram of IWRM including complexity. ............................ 174


12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE COMPLEXITY OF INTEGRATED URBAN WATER RESOURCES MANAGEMENT: A STUDY OF WATER S UPPLY AVAILABILITY, VULNERABILITY, AND MANAGEMENT OF LARGE CITIES IN THE UNITED STATES By Julie C. Padowski December 2011 Chair: James W. Jawitz Major: Soil and Water Science Over the past century, the effects of urbanization in the United States have been profound; urban populations have come to represent the majority of the population and because of this change we have not only drastically altered the way in which we view and use resources, but how we function as a society To thrive, urban centers require constant inputs of resources, such as water. However, over the past fifty years, rapidly increasing urban water demands have strained water supplies to the point where serious concerns over urban water scarcity now exist. Despite these concerns, there has been relatively little progress made towards better understand ing urban water availability, vulnerability and management at the national level. To address this lack of information, the work presented here collected information on all urban area s throughout the coterminous U.S. with populations greater than 100,000 (n=255) to identify 1) the current state of water availability and vulnerability in urban areas, 2) the relationship between water availability and urban water management responses to stress, 3) the role of complexity in urban utility operations


13 and finances, 4) collaborative particip ation in urban water management as well as 5) the emergence of complexity over time in urban water systems. Results from this work revealed that 1) approxi mately 2 5% of the US population resides in an urban area where low average annual water availability is an issue. Water vulnerability analyses, however, showed that over half of the population experiences at least a moderate level of vulnerability, either due to source inflow variability, or poor water availability. In locations where water availability was low, 2) a positive correlation was found between availability and both the complexity of the water supply infrastruct ure and water supply management. In addition, research revealed that urban areas who shared sources also had higher levels of infrastructure and management complexity. When assessing 3) costs within the urban water complexity paradigm, results indicated that water provision processes sca le across urban utility population size with economies of scale. Using a combined metric to measure total operational complexity of each utility, results showed that in general, the level of complexity was higher than expected, and that higher complexity was correlated to higher overall costs. When utilities were assessed for their 5) level of collaborative participation results indicated that over the past five years, collaborative participation has risen within urban utilities, and has coincided with a shift away from supply based management towards a more balanced use of supply and dem and based management strategies as well as an increase in the importance utilities place on water management plans. Finally, 6) when urban water complexity was examined over time, distinct patterns and trends emerged between disparate case studies. Similar threads of complexity are thought to run though all urban water management systems.


14 CHAPTER 1 BACKGROUND AND SIGNI FICANCE The Shifting Management Paradigm The impact of urbanization within the United States (US) over the past century has been profound. Urban areas hold more than 80% of the total population into less than 3% of the total land area and a s of 2010, were generating nearly 90% of the national Gross Domest ic Product ( GDP ) (US Census Bureau, 2000a; USDA Economic Research Service, 2002; U.S. BEA, 2011) The result of this intense localization of people and productivity is that urbanization has not only changed the way land and resources are acquired and used, but has fundamentally altered the relationship between society and its environment (Grimm et al. 2008; Kates and Parris, 2003) In the past, the strong ethic of growth and a control based approach to natural resource management all owed urban areas to experience major economic development (Gleick, 1998; McMichael et al. 2003) This exhaustive utilization of resources over the last century has taken its toll on environmental health and function and as the effects of over exploitation have become more fully realized, a new m anagement framework of become the dominant discourse in natural resources management, and has redefined the human environment relationship once again (Graedel and Klee, 2002) While sustainability is a concept that can be applied to any discipline, this dissertation project focuses on its applicat ion to the realm of urban water resources management. Contextualizing Urban Water Management To better understand the role of sustainability in urban water system management, the past must be put in appropriate context. Traditionally, water provision in urban


15 areas has been the responsibility of the municipality. The impetus for a permanent, centralized system of water distribution has its roots in the early 19 th century, when US urban centers were first forming. Water pollution due to non existent was tewater management had prompted major public health epidemics in many cities. In response, water and wastewater infrastructure was installed as a public service, of sorts (Melosi, 2000) As urban areas and water demands grew, urban water providers developed this century, urban systems were at the fo refront of water development. For example, urban centers implemented unprecedented technological developments aimed toward appropriating substantial volumes of fresh water for human use, particularly in regions where natural water availability is low (Dynesius and Nilsson, 1994) Billions of dollars were spent building large dams and other types of water infrastructure around the US t o transfer, store and regulate water resources (BOR, 1972) Most urban systems have included either local natural or constructed storages as critical strategies to mitigate the natural variability, and in some cases the insufficiency, of water supply (Graf, 1999 ) The benefits of this type of management strategy have improved the quality of life of millions of citizens through increased flood control, hydropower generation, and a permanent, secure source of water for urban consumption. However, damage to ripar ian ecosystems resulting from alterations in flow quantity, quality and variability, coupled with dramatic and sometimes irreversible decreases in water table levels, have had severe detrimental effects in some locations, creating serious challenges for wa ter managers today.


16 To frame the discussion of present day urban water management, the United States Geological Survey (USGS) estimated a national rate of water withdrawal of approximately 1.544 trillion liters per day (lpd) (Huston et al. 2005) Of this total, the public supply sector, which is mainly comprised of water utilities, was listed as the third largest u ser behind thermoelectric power and irrigation, accounting for 11% of the total national withdrawals. Within this public supply sector, large water systems serving urban populations (>100,000) account for less than 1% of the total number of water systems in the nation, but these large utilities supply approximately 81% of the national population (CBO, 2002) Water demands of this sector have increased steadily since 1950, and have increased more than three fold over the last fifty years. Presently, more tha n three quarters of the national population depend on the public sector to produce a total of approximately 167 billion lpd (Huston e t al. 2005) Today, many urban utilities are faced with the perpetual problem of managing large water demands in the face of climate change, hydrologic non stationarity, increased costs, population growth and environmental regulations (Ginley and Ralston, 2010; Means, West, et al. 2005) T he complexity of this task manifests in myriad physical, economic, and institutional issues. Recognition of these problems, however, has paved the way for the acceptance of a new sustainability based water management paradigm: Integrated Water Resources Ma nagement (IWRM) which endeavors to offer a more holistic approach. Integrated Water Resources Management (IWRM) IWRM embodies a philosophy of sustainability in water resources management, development and management of water, land and related resources, in order to maximize the resultant economic and social welfare in an equitable manner without compromising the


17 (Mitchell, 2005) This management framework is particularly appealing to water managers today, as the growing uncertainty surrounding future urban water availability in the US has become a primary issue of concern for many water providers (Runge and Mann, 2008) In response to these challenges, urban water managers utilize a variety of strategies to deal with all of the issues associated with water resources management. Many urban areas have begun to monitor water usage by their customer base more closely. This increased vigilance provides utilities not only with an assessment of the vulnerability of their systems in terms of adequate supplies, but allows them to more effectively implement demand side management strategies. Demand side management has become increasingly popular as utilities attempt t o control, rather than accommodate, the needs of water users. Demand side management strategies such as water conservation based pricing structures, appliance retro fits, and education all tend to be relatively low cost and easy to implement (Baumann et al. 1998) In addition, urban systems also are utilizing alternative supply based strategies such as water recycling/reuse program s, shared source water planning, and source enhancement from alternate water supplies such as desalination, aquifer storage and recovery and the conjunctive management of surface and groundwater sources (Aichinger, 2009; Gleick, 2000; Miller, 2006; Blomquist et al. 2004) While IWRM has been proclaimed by many to be the solution to the problems of traditional water manageme nt and the inevitable path of the future (Falkenmark, 2004; Gleick, 2003; Serageldin, 1995; Slocombe, 1993) the lack of a specific protocol for impleme nting this new paradigm makes IWRM difficult to use in practice (Bellamy et al.


18 1999; Biswas, 2003; Mitchell, 2005) Additionally, this paradigm represents a quantum leap in complexity, requiring as it does the full conside ration of a vast suite of interconnected and interdependent environmental, economic, and social needs. Despite the challenges associated with the implementation of IWRM, the lure of this new management paradigm remains attractive to many water managers ye t i t is clear that water managers are struggling to identify the best way to successfully implement IWRM. The relatively new field of complexity theory provides a potentially useful framework for understanding the key concepts associated with this effort. Issues of Complexity Complexity theory allows IWRM to be viewed from a new perspective, one in which all of the nuances and complications associated with integrating continual growth with management across environmental, economic and social needs are sim plified. While IWRM has yet to be fully redefined in terms of complexity, his clearer understanding is crucial if IWRM truly will become the paradigm of the future. Complexity can be seen most simply as the relationships between linked components within a system when the dynamics between the individual components are no longer simple. Simple systems are predictable, classic equilibrium based models that rely on linear relationships, scalability, predictability and reducibility (Baynes, 2009) Complex systems represent higher order systems with non linear behavior that reacts actively and anticipate change, with individual system co mponents that are perpetually redefining and revising their inter relationships; such systems have outcomes that are more difficult to predict (Manson, 2001) These dynamics give complex systems emergent properties, with the capacity of the system exceeding the sum of its constituent parts, and an improved ability to self organize or change its internal


19 structure to better adapt to i ts environment (Manson, 2001) As such, constant but rather remains in a dynamic state (Holland, 1992) Since the introduction of complexity theory in the 1970s, concepts of c omplexity have been applied to a wide range of systems the behavior of which once was thought to be too difficult to predict: from biological and chemical, to technological and informational, to economic and social (Boccaletti et al. 2006; Amaral et al. 1998) Com plexity theory also has been used to describe urban systems, describing cities as entities with emergent properties and the ability to self organize (Baynes, 2009) Urban population (Zipf, 1949) land use (Batty, 2008) transportation (Richmond, 1998) innovation (Bettencourt, Lobo, and Strumsky, 2007) economic growth (Glaeser, 1994) and supply networks (Kuhnert et al. 2006) all have been fou nd to conform to the rules governing complex systems. Recently, concepts of complexity also have been applied to natural resources management as a way of better understanding the institutional (Cowie and Borrett, 2005; Ostrom, 1999; Saleth and Dinar, 2004) legal (Ruhl and Ruhl, 1997) and general problems (Allen et al. 1999; Farber, 2003; Geldof, 1995a) associated with its implementation. These efforts have resulted in invaluable additions to the IWRM literature, and provide perspective for a management system once thought too complicated to truly understand or implement. H owever, more study obviously is needed if the full benefits of IWRM are to be realized (Bellamy et al. 1999; Biswas, 2004) The State of Urban Water Management: A National Perspective While there are many water management problems to be addressed, this dissertation project incorporates the con cepts of complexity and scale, and was


20 designed to provide a more comprehensive understanding of water management in the US from a national perspective. This approach utilizes an interdisciplinary framework to examine a comprehensive range of factors that influence urban water management. The overall goal of this dissertation research is to identify and evaluate the major causal factors affecting water management in urban areas across the US. This research provides a new metric for assessing urban water availability and vulnerability and investigates a variety of avenues by which urban utilities have responded to vulnerability, including how complexity has played a role in shaping urban management today. The five sub goals of the dissertation project are discussed below in relation to their context within and contribution to the scientific literature. The major findings of this research, together with the limitations associated with each element of this study, are considered and future avenues of investig ation in each topic area are proposed. Water Availability and Vulnerability At the national level, the US has an abundance of water. However, substantial spatial and temporal variability exists in both water storages and water recharge. Up to 45% of the nation has been affected by severe to extreme drought conditions over the past century (Biswas, 2003) significantly impacting loc al and regional water availability and management in many areas. Hydrologic vulnerability, a situation in which water supplies are at risk either due to source variability or over exploitation, ultimately impacts the entire nation to some extent. Signifi cant water table declines of more than 100 m have been recorded in the Houston, Chicago, and Phoenix Tucson urban areas (USGS, 2003) In some cities, such as Fresno, Las Vegas, and Tampa, groundwater vulnerability also has manifested as land sub sidence, decreased surface water


21 discharges and/or saltwater intrusion (USGS, 2003a) Vulnerability in surface water sources also has been a problem for urban areas, highlighted by conflict over the allocation of flows from Lake Lanier between Atlanta, Florida and Alabama (Sun et al., 2008) and the proportion of Owens Lake that can be used by Los Angel es and the Owens Valley (Reisner, 1993) As water systems have expanded to meet urban demands, the scale at which water is managed also has grown. With this increase in accountability via system expansion, managers have found it more difficult to track and assess the availability and v ulnerability of their water supplies accurately in terms of both urban provision and environmental function (Runge and Mann, 2008) This problem is magnified when urban areas are considered as a homogeneous group. Water availability assessments at the national scale have not been performed since the 1970s (WRC, 1978) making it difficult to track regional scale system performance, since there is no widely available, systematic, centralized benchmarking database for this type of information. Institutions an d Infrastructure The limited understanding of the structural and institutional framework within which urban water management systems exist represents one of the most significant challenges to urban water management at the national level. From an institut ional perspective, the regulation of the US water industry is highly decentralized and heterogeneous. Federal involvement in water management extends only as far as f drinking water quality (Safe Drinking Water Act) and wastewater disposal (National Pollutant Discharge Elimination System) (EPA, 2008; Stakhiv, 2003) One exception, however, is the significant role of the US Army Corps of Engineers in the regulation and


22 control of major water resources for which large supply systems were built and/or managed by the federal government (Davis, 1968; Lee, 1999) This relative absence of federal involvement leaves the major responsibility of policy setting and enforcement, such as the regulation of water rates and water supply coordination, to state or local governments, or in some cases, the urban utility itself (Baumann et al. 1998) In areas in which large scale, capital intensive projects have been created (e.g., Colorado reservoir system, Metropolitan Water District of Southern California), foundations have been laid for inter urban and even regional management solutions for securing water resources. General theories regarding m anagement of natural resources suggest that resource needs drive the complexity of the management system through a series of positive feedback loops (Allen et al. 1999) If these theories are correct, a strong connection between the development of water infrastructure and water management framework would be expected to exist. Such connections could provide insight into how the ma nagement framework supports conflict or cooperation. However these connections have not been studied or characterized at the national level (Ausubel, 1988; Blake, 1956) Cost of Water Provision Water provision, as a general rule, is a highly capital intensive business requiring hundreds of kilom eters of pipeline for distribution and transmission, as well as facilities for water storage, pumping, and treatment (Baumann et al 1998) To better understand how to manage these types of systems optimally, dozens of utility efficiency and productivity studies have been performed over the past two decades (Bhattacharyya et al. 1994; Houtsma and Sackville, 2003; Norton Jr and Weber Jr, 2009; Shih et al. 2006; Tynan and Kingdom, 2005) While these studies constitute a


23 significant body of knowledge on economies of scale in water utility services, the approach to integrating this information into urban complexity studies has yet to be determined. Urban complexity studies use power law functions to relate var ious urban processes to city size (Batty, 2003; Baynes, 2009; Bettencourt, Lobo, Helbing, et al. 2007; Glaeser, 1994; Kuhnert et al. 2006) From the way in which these processes scale, three distinct categories of urban metrics emerge based on the exponents of the power function Bettencourt et al. (2007) Specifically there are metrics that scale supra linearly, which represent primarily social processes th at occur with increasing returns to population size, e.g., innovation and crime. Urban processes also can scale linearly, representing primarily those activities that meet basic individual needs, such as provision of water and electricity or the availabi lity of housing. Finally, there are those processes that scale sub linearly, which are associated most closely with material goods and infrastructure that operate according to economies of scale. Despite the fact that urban water provision is a critical c omponent of urban growth, little research has been devoted to better understanding this crucial factor in the context of urban complexity. Collaborative Participation In response to these challenges, a variety of new strategies have been developed to resp ond to the need for incorporating these socio economic natural linkages into water resources management (Conca, 2006) Urban cente rs have developed a wide array of strategies in an attempt to address these problems. The strategies that are truly integrative in nature, such as conjunctive management and shared source planning, require water managers to coordinate and communicate with other users and agencies.


24 As interest in utilizing these integrated, regional management solutions increases, it becomes increasingly important to develop an effective system for information exchange among users to ensure that management decisions are ma de using the best knowledge available (Hackett et al. 1994) Information exchange (i.e., the transfer of knowledge, ideas and v alues across different levels of management) is critical for managing shared resources when multiple stakeholders must collaborate to identify common goals for successful management (Cortner and Moote, 1994; Reitsma, 1996) Access to information depends on a d external to the organization, and can impact the (Anderson, 2008; Cross and Sproull, 2004) Managers who have greater access to information, and subsequently use the knowledge gathered, are more likely to notice trends and problems, under stand their environment, and perform at a higher level than those who have limited access to information (Anderson, 2008) The d egree to which these types of transfers are taking place between urban utilities is still not well understood System Complexity Over Time As urban water systems attempt to make the transition from traditional to more sustainable management strategies, a framework for understanding urban natural resource management as a complex system is required. The application of complexity theory to natural resource management research has yielded positive results (Allen et al. 1999; Cowie and Borrett, 2005; Ostrom, 1999; Rammel et al. 2007) However, the application of complexity theory specifically to issues associated with integrated urban water management has n ot been well studied. The work by Geldof (1995b) represents one of the few publications specifically addressing this topic. While this work is arguably


25 an important contribution, it was somewhat limited in scope and did not consider how complexity changes in urban systems with time. Tainter (2006) and Allen et al. (1999) to some extent have addressed this knowledge gap by providing a framework for assessing changes in urban water management complexity over time for three case studies: Los Angeles, CA, Tampa Bay Region, FL and Washington, DC. Research Goals and Objectives While much research on urban water resources management is currently being performed, few studies attempt to address this topic in the context of complexity. This dissertation project was devel oped to investigate the state of water management in the US from a national perspective by addressing the following objectives: Objective 1. Develop and implement a novel storage based methodology for assessing water availability and vulnerability in US u rban areas based on available hydrogeographic data (Chapter 2). Objective 2. Assess the relationship between water availability and urban water management in terms of the complexity of the infrastructure and management organization (Chapter 3). Objective 3 Determine the extent to which financial and operational water provision metrics fit into the urban complexity paradigm (Chapter 4). Objective 4. Measure the degree to which participatory collaboration is being used by urban water utilities to make deci sions about water management and planning (Chapter 5). Objective 5. Reassess the role of complexity in the development of urban water management over time in three urban case studies (Chapter 6).


26 CHAPTER 2 A STORAGE BASED APPROACH FOR A SSESSING URBAN WAT ER AVAILABILITY AND VULNERABILITY Introduction Over the last half century, the United States (US) has become an increasingly urbanized country with nearly 80% of the population residing in urban and suburban areas (US Census Bureau, 2000a) During this same time period, water demands in the public supply sector have increased steadily, more than tripling since 1950 (Huston et al. 2005) Today, rapidly growing urban demands are straining local and regional water supplies despite the apparent national wealth of water resources, and have resulted in serious concerns over urban water scarcity in the US (Levin et al. 2002) These concerns were highlighted in a survey by the American Water Works Association (AWWA) in 2008, which polled the North American water industry to identify key of uncertainties primary issue of concern for both the short and long term future was that of source water availability, i.e. maintenance of adequate quantities of treatable water available for consu mption (Runge and Mann, 2008) The study determined that this result reflected not only anxiety over recent reports of water sho rtages such as those reported in Atlanta, GA in 2008, and San Francisco, CA in 2006 2007 (NRDC, 2010) but also concerns about oth environmental regulation (Ginley and Ralston, 2010; Means, West, et al. 2005) The concerns reported by urban water managers reflect broader economic, institutional, and regulatory problems faced by the industry, many of which st em in part from the lack of a basic national water policy (Gleick, 2001; Levin et al. 2002; Stakhiv,


27 2003) Currently, regulation of the American water industry is highly decentralized and heterogeneous (Baumann et al. 1998) Although some control over utility operations occurs at the federal and state levels, utilities are largely responsible for cr eating, evaluating, and monitoring their own performance. The flexibility associated with a decentralized regulatory policy has been advantageous in that it has allowed utilities to develop unique management solutions to the variety of water supply and de mand issues experienced at each location. However, the lack of basic, top level oversight has inefficiently and at great cost to society and the environment (Gleick, 2001) While these problems are the product of many factors, part of the blame resides in the fact that historically there has been no cen tralized system from which cross evaluations of water utility management strategies, performance and impacts could be made (Stakhiv 2003) For urban water managers to make effective plans in the face of increasing uncertainty, it is critical that measurements of water availability, or how much water is available for use, and water vulnerability, the ability of available water sour ces to meet all needs under varying degrees of water stress, be made. By understanding how water availability and vulnerability issues affect, and are affected by urban water systems, managers can better address both environmental and human needs without over investing or under supplying (NRC, 2002) While it is obvious that solutions to these issues are desired at the local scale, there are also significant potential benefits to creating a comprehensive, standardized metric for assessing the availability and vulnerability of urban systems. Such a


28 methodology would provide a uniform framework for local evaluations from which more c and national levels, as well as providing a system by which regional collaboration among utilities to resolve these issues could take place. For this metric to be usefu l across many scales, an ideal standardized method should 1) include all hydrologic components that are relevant to the area of study, 2) be executable at a scale that is directly applicable to urban water management, and 3) be general enough to allow for reasonably accurate comparisons with other urban areas. This work attempts to address all of these needs by creating and implementing a metric for assessing water availability and vulnerability that is useful at both the local and national level. Current Methods f or Measuring Water Availability And Vulnerability At the urban utility level, water managers are acutely aware of the need for quality water availability and vulnerability assessments as they plan for growing demands and future investments in inf rastructure (Means, West, et al. 2005) In response to a need for this information, many utilities use internal assessments as pa rt of the planning process. The depth and breadth of these assessments can vary widely between utilities, as can the methods used and results obtained (Viessman and Feather, 2006) This variability exists in part because a comprehensive evaluation of a hydrologic system (which for some urban areas can be quite complex) requires a substantial amount of spatial and temporal informatio n. Assessments of reservoir reliability, groundwater table fluctuations and in stream lake and river water levels all require historical data to monitor and predict the effects of urban use on water resources over time. Problems arise as t he cost of gath ering and evaluating the requisite data for a


29 thorough evaluation each year can be prohibitively expensive (WIN, 2000) In addition, access to these types of utility assessments for any comprehensive analysis can be difficult since there is no publicly available, systematic, centralized benchmarking database for this type of information or any other measure of utility perform ance, such as economic and operational efficiency (AWWA, 2007; Berg, 2002) Together, these factors ultimately render such reports unsuitable for inter city comparisons and prevent standardized assessments from occurring at the regional or national level. Urban water availability and vulnerability are not solely the concern of water utilities. Governmental organizations and academic institutions have been studying various aspects of these problems for decades. The most recent major federal effort to quantify water availability was performed in 197 8 by the US National Resources Council (WRC, 1978) In this comprehensive report, both ground and surface water resources were ch aracterized in terms of their quality and quantity. Water supply, water use, and critical problem areas were identified and evaluated at a variety of scales. Despite significant changes in water use, and population growth and redistribution, no comparabl e national assessment has since been conducted. However, in light of the dearth of information at this national scale, the US Congress promulgated in 2009 the SECURE (Science and Engineering to Comprehensively Understand and Responsibly Enhance) Water Act authorizing the USGS to create and implement a new comprehensive national assessment of water availability and use (Konrad, 2010) Unfortunately, this assessment will not be available for several years. Academic assessments of water availability and vulnerability tend to be broader in scale than utility conducted assessments, and utilize a greater range of methodologies.


30 Many acade mic assessments have been designed to coarsely quantify global water availability (Alcamo et al. 2003; Falkenmark, 1998; McDonald et al. 2011; Sullivan et al. 2003) The Falkenmark Water Stress Index (1989) was one of the first approaches for quantifying water availability as a function of population, accounting for differences t), and management or overpopulation). Subsequent studies such as those reviewed by Rijsberman (2006) have further developed qualitative and quantitative methods for measuring water availability on national, regional and global scales. Although large scale asses sments are informative, many lack sufficient detail and precision to be useful for urban water managers. For example, large scale regional and national assessments such as those by Hurd et al. (1999); Hurd et al. (2004); Roy et al. (2005); and WRC (1978) analyze water availability and vulne rability using units of analysis much larger than those relevant to urban water managers and are usually not adequately refined to account for the individual strategies that urban utilities have adopted to meet demands. Many local scale assessments of ur ban water management have been conducted, focusing on conceptual theory or examining urban water issues as case studies (Brikowski, 2008; Elbakidze, 2006; Jenerette and Larsen, 2006; Reilly et al. 2008; Rose and Peters, 2001; Rosenberg et al. 1999; Saunders and Lewis, 2003; Sun et al. 2008) However, as with assessments made by utilities, the often complex and varied methodologies tend to preclude synthesis or application across individual studies. Other distinct types of water availability vulnerability analyses at the local scale are those using risk based performance metrics developed by Hashimoto et al. (1982)


31 These measure the reliability, resilience and vulnerability (RRV) of wate r resources systems and have been widely applied at the urban scale. RRV analyses have been used to characterize the performance of individual and regional reservoir operations (Hashimoto et al. 1982; Kay, 2000; Vogel, Lane, et al. 1999) and water distribution networks (Farmani et al. 2006) as well as to produce risk assessments of water quality (Maier et al. 2001; Sarang et al. 2008) and impacts of climate change on water resource systems (Fowler et al. 2003; Hurd et al. 1999) While RRV analyses utilize a standardized metric, their usefulness in this context is limited by insufficient records of local, historical data, which are currently limited, localized, or nonexistent at the national scale (Kjeldsen and Rosbjerg, 2004; Taylor, 2009) In addition to these limitations, a critical consideration is that previous studies have relied on of mean annual runoff (MAR) as a measure of availability vulnerability MAR based estimates represent the amount of renewable water av ailable in the form of runoff, calculated as the difference between precipitation and evapotranspiration (P ET) (Taylor, 2009) T his assumes that changes in storage are either negligible or unimportant. The popularity of this measure stems from the wide geographic and historical availability of data; however, the assumptions underlying this type of analysis are potentially problema tic. Specifically, the assumption of stationarity, which states that all water resource systems fluctuate within a predictable/stable/known range of variability and therefore experience no net changes in storage, has come under scrutiny (Cordery et al. 2006; Milly et al. 2008; Taylor, 2009) Recent research suggests that climate change is intensifying the hydrologic cycle, which increases the potential for substantial changes in storages, as well as the amount and variabil ity of water inputs


3 2 and outputs within an urban system (Cromwell et al. 2007) An improved understanding of the impact of climat e change and human development on natural systems has challenged the idea that hydrologic systems are stable and net changes in storage are negligible. Taylor (2009) highlights the issues associated with MAR based assessments by emphasizing the need to s pecifically acknowledge three components typically not included : 1) water available as baseflow vs. stormflow 2) water stored as soil moisture and 3) water in basin storage. All of these have the potential to significantly alter current estimates of water availability and vulnerability The third component, basin storage, is particularly problematic for urban areas. This is because almost all urban areas rely on some form of storage, either below or above ground, as part of their supply portfolio. In man y cases, storage is critical for mitigating the effects of short term (drought) and/or long term (arid) deficits in water availability. In some cases, storage is heavily depended upon to compensate for shortfalls, leading to substantial reductions in tota l storage and user conflicts. Examples of storage based changes in urban water availability and vulnerability have occurred many times over the past century. These imbalances are reflected in the significant water table declines in many cities including Houston (400 ft), Chicago (900 ft), and Phoenix Tucson (400 ft) (USGS, 2003) In some cities, such as Fresno, Las Vegas, and Tampa, storage reduction has also been associated with other problems including land subsidence, decreased surface water discharges and/or saltwater intrusion (USGS, 2003a) Reductions in available surface water storage (i.e. reservoirs and lakes ) are highlighted when conflicts between users becomes apparent The disagreements over the allocation of flows from Lake Lanier


33 between the city o f Atlanta, and the states of Florida and Alabama (Sun et al., 2008) and the propo rtion of Owens Lake that can be used by Los Angeles versus the Owens Valley residents (Reisner, 1993) are two recent examples of these types of conflicts As Taylor (2009) The o bjectives of this study were to examine the need for a more comprehensive understanding of national urban water resources by: 1) producing a national assessment of urban water availability that accounts for the role of storage, and 2) estimating the vulner ability of individual urban areas using storage based water availability calculations. In completing these objectives, the following hypotheses were also tested: 1. Storage based water availability estimates are significantly greater than MAR based estimate s of water availability 2. Storage based vulnerability estimates will accurately reflect known urban water scarcity issues. Estimating Water A vailability Using A Storage based Approach Water availability and vulnerability assessments were performed for all 2 55 major urban areas (Appendix A) with sufficient hydrologic data and populations greater than 100,000 in the coterminous US. Individual assessments were made using a standardized methodology that includes not only renewable flows, but basin storages. Dat a characterizing the local average annual hydrology (including storages) and water usage were collected from publicly available databases (Figure 2 1) and used to produce a modified hydrologic budget from which water availability was quantified. The impli cations of potential change in availability, or vulnerability were estimated based on


34 the volume of water available when the hydrologic system is stressed Water vulnerability was measured as a function of water availability and the variability of those available sources under low flow conditions To verify the water vulnerability estimates produced using this method, an independent qualitative proxy was developed to provide an objective, alternate measure of vulnerability This measure was a media con tent analysis conducted to identify vulnerability as a function of local and national media attention. Calculating Water Availability a nd Vulnerability Water availability estimates for natural systems are most commonly made using a water budget, where the rate of change in stored water is calculated for a given area. The area chosen for the budget depends on the scale and scope of the analysis, but is typically defined by hydrologic boundaries such that the rate at which water flows into and out of an area is balanced. Rather than defining this budget by balanced flows, political boundaries were used to define which sources would be included in the analysis. This was considered necessary as hydrologic boundaries, while convenient, are often at the wrong scale to accurately represent the actual sources and volumes of Of the 255 locations selected, each was first screened to ensure that adequate information was available to perform these assessments. Ur ban areas were rejected from these analyses based on the availability of primary supply source information. If more than 25% of this information was missing from the available databases, the urban area was excluded from the study. In total, adequate info rmation was available for 89% (n=22 6 ) of the urban areas initially selected.


35 Water availability ( Q A ) for each urban area was calculated as the sum of all local and captured flows and storages potentially available on an average annual basis: ( 2 1) where, = average annual available river discharge [L 3 /T], = average annual available reservoir storage [L 3 /T], = average annual available natural lake s torage [L 3 /T], = average annual available groundwater availability [L 3 /T], and the subscripts l and c indicate whether the term is local or captured, respectively. To more clearly identify which urban areas are at risk of potential water scarcit y, a commonly used indicator developed by Falkenmark (1998) was employed. This indica tor defines scarcity according to the ratio of water demanded to water available. Water scarcity occurs when > 0.4 or 40%. While this definition of scarcity has been previously used as a water scarcity reference point for other water availability assessments (Vrsmarty et al. 2005) it is not limited by the type of water source (e.g., parameter for defining a scarcity threshold with the curren t, storage based water estimate the average minimum volume of available water needed by urban areas to avoid scarcity issues ( ) based on their individual average annual demands and is equivalent to = 1459 lpcd Assessments of water availability describe urban water resources for a given location over time but generally are not used to measure the sensitivity, o r the severity, of variability in Q A that may occur due to seasonality, droughts or prolonged over extraction. These types of assessments typically rely on time series analyses; however


36 a lack of historical data for each of the sources at this scale preve nts this type of measurement from being accomplished in this study. Instead, vulnerability was measured as a function of water availability and variability under low flow conditions where water variability ( ) is : ( 2 2) where n = number of input variables used, and average exceedance f requencies ( ), found using methods described in the Data Acquisition and Assumptions section, represent the low flow conditions where Q A inflows are exceeded 95% ( i = 5), 90% ( i =10) and 75% ( i =25) of the time for each local and capture source associa ted with an urban area. In analyses, was normalized by Data Acquisition and Assumptions The urban water availability and vulnerability assessments were performed using an extensive collection of publicly available US hydrologic and water usage databases (Table 2 area. Local sources included all hydrologic features that intersected or bordered the urban area boundary (UAB) (Figure 2 2). Captured sources are those that did not occur within the UAB, but have been identified as a primary source of water for the urb an area. A limitation of this work is that for some input categories, a number of simplifications and assumptions (described in the following section) were required to accommodate the use of limited, localized, inaccessible, and occasionally non existent data.


37 Urban a rea Urban water utilities are the sole providers of water to urban areas and thus are the most suitable scale at which to examine water availability and vulnerability ; however, information regarding the service population and areal extent o f most urban utilities is not readily available. Therefore, to determine the most appropriate substitute scale for representing urban water management, the population sizes of each of the three different urban delineations (City, Urban Area and Metropolit an Statistical Area (MSA)) were compared to the service population size of 100 major urban utilities (Figure 2 3). Results from this analysis indicated MSAs and Urban Areas overestimated utility service population size by 82% (R 2 =0.69) and 47% (R 2 =0.76), respectively. The City scale underestimated utility service population size by 49% (R 2 =0.70). Of the two groups that most closely represented utility service population size, the Urban Area scale was ultimately selected as the unit of analysis due to the slightly closer fit to the desired data, both in terms of the overall approximation of population, and the goodness of fit of the data. While the Urban Area scale was considered the most appropriate of the three options considered for this study, in real ity its usefulness as a proxy for utility service population is limited by the large (47%) difference in the desired population size. and population size of each sampl e selected. Urban area information was collected delineating the boundary of each urban area (US Census Bureau, 2000b) This dataset, originally designed to better distinguish urban areas from rural ones for census purposes, consists of those densely settled territories that contain greater than 5 0,000 people. The urban areas used in this study are a modified subset of the Urbanized


38 each was given a 5km buffer to ensure all adjacent local sources were included. Henceforth, all references to the UAB are assumed to include this 5km buffer. Urban supply and d emand Water utility data were collected for each primary (named) city included in the urban area, totaling to 294 utilities of interest. The information gathered from each Quality/Consumer Confidence Report and included (when available): the name of the primary water utility supplier(s) for each area, its service area, service population size, total annual water provided, and primary sources of supply. Each primary supply source listed by a utility (e.g. reservoir, river, aquifer, etc.) was identified in one of the hydrologic databases described below. Data on urban wa ter demand ( D ) was calculated using an area weighted average of the total public supply demand as reported by county in each urban area (Kenny et al. 2009) Of the 255 urban areas sampled, only two (Barnstable Town, MA and Brooksville, FL) were missing any web based utility information, and were thus excluded from this study. Streamflow Streamflow data were derived from two sources. Spatial data which included the name, location and stream order of over 10,000 rivers and streams were obtained from the National Hydrography Dataset (NHD) (NHD, 2008) River discharge information, including average annual discharge ( ), discharge exceedance probabilities ( P ), length of record, and discharge variability ( ) were obtained from a USGS spatial database of 23,427 streamgages within the coterminous US (USGS, 2008a) To minimize the


39 inclusion of unnecessary data, all streams and riv ers with a stream order less than 4 were excluded from this study, based on the assumption that urban areas do not utilize sources when the equivalent minimum average demand (600 lpcd x 100,000 persons =0.7 m 3 /s) is more than 25% of the average annual rive r flow (2.80 m3/s) (Figure 2 4). Environmental water demands were considered by reserving from local and needed to maintain a desired level of aquatic and riparian health. Minimum flow values were calculated based on work by Smakhtin et al. (2004) who p roposed a general set of guidelines for determining acceptable low flow requirements based on the level of ecological disturbance of different management objectives. Minimum flows for this equirements low flow requirement corresponds to the P 25 exceedance probability, and therefore the maximum flow allowed to each area was measured as: ( 2 3) where Qi = total average annual measured discharge [L 3 /T] and P25 = average discharge that is exceeded 75% of the time [L 3 /T]. For the vulnerability analysis, streamflow variability was simply measured in terms of t he P 5 P 10 and P 25 exceedance probabilities reported with the river discharge information. Reservoir s torages Reservoir information, including dam and reservoir name, location, normal storage volume, source river name, drainage area, purpose, and ownersh ip were obtained from the National Inventory of Dams (NID) (NID, 2009) In the cases where a single


40 reservoir with the sole purp ose of water supply is the source for only one urban area, the available water is equal to the reported normal storage volume ( ) for that reservoir. In the cases where a reservoir serves more than one purpose (e.g. water supply and hydroelectric ge neration) or urban area, conservative estimates of reservoir normal storage allocations were made by first equally dividing the total volume of storage between the assigned purposes, and secondly, equally dividing this modified storage volume between urban areas using the reservoir. Information on stage variability for reservoir sources was not part of the NID database, therefore, variability in reservoir inflows served as a proxy for variability in reservoir normal storage in lieu of any other available in formation. Because discharge and exceedance frequencies were not consistently available for all of the rivers feeding each reservoir, information on reservoir inflows were instead estimated using a regional hydroclimatologic regression model developed by Vogel et al., (1999a) This model had previously been applied by Vogel et al., (1999b) to successfully estimate the mean and variance of annual inflows to 5,392 US reservoirs from the NID database as part of a RRV analysis. Therefore, the following equations were used to estimate the mean annual streamflow ( ) and flow variance ( ) for individual rivers based on hydrologic and climatic information from each of the 18 US water resource regions outlined in that study (Figure 2 5): ( 2 4) ( 2 5)


41 where a through i are empirical model parameters that are listed for each water region in Vogel et al. (1999a), A is drainage area [km 2 ], P is mean annual precipitation [mm/yr], and T is mean annual temperature [ *10] Note that Vogel et al., (1999a) considered small basins where it was assumed th at river flows are unregulated and reservoirs operate independently of each other. Because these assumptions may not apply to the reservoirs considered in this study, the accuracy of Equations 4 and 5 were compared to the small subsample of measured data from reservoirs with gaged river information collected in this study (n = 55). Results from this comparative analysis revealed discrepancies between inflow measurements for reservoirs located in regions 10, 11, 12, 14, 17 and 18, with the largest differen ce between estimated and observed inflow occurring in region 12 (Figure 2 6). Despite the limited subsample, this comparison suggests the largest flow independently manag ed reservoirs may not hold true. Many large rivers in the southern and western half of the US have multiple large dams along their reach, substantially impacting river flows and downstream reservoir operations. To correct for these errors, reservoirs lo 9, 12 and 14) were given a correction factor based on the average distance between the observed and measured data (Table 2 2). Based on the assumption that daily flow distributions for most rivers in the US can be reasonably approximated as log normal (Vogel, Lane, et al. 1999) the adjusted mean annual flow and flow variance were then P 5 P 10 and P 25 exceedance frequencies for reservoir inflow were estimated.


42 Natural lake s torages Information on natural lakes was the mos t difficult to obtain since no reliable database distinguishing natural lakes from reservoirs was available. Therefore, this study only includes lakes specifically identified as water sources by urban utilities. For each lake included, however, informati on on the location, surface area and average depth were collected from independent sources, and when unknown, lake volume was storage could be allocated between users was al so difficult as few studies have attempted to quantify the minimum stage levels necessary for maintaining adequate ecological function in natural lakes. Therefore, urban availability from natural lakes was based on the following two assumptions: 1) lakes are managed to ensure supply over a 50 year planning horizon (Graedel and Klee, 2002) 2) urban areas are the sole users of lakes. Thus, the mean annual lake volume available in any given year was equal to 1/50th of the total lake volume. In instances where more than one urban area utilized a lake as a primary source of water, this modified mean annual volume was equally divided among the users. Variability in lake storage was also unavailable and therefore estimated based on the P 5 P 10 and P 25 Inflow data, including the mean annual flow ( ) and flow variance ( ), were available for the five Great Lakes only. Annual inflow exceedance frequencies for the other 10 lakes considered here were estimated based on measurements of inflow from local precipitation. As basin information for these lakes was not readil y available, these values were simply estimated as: ( 2 6)


43 ( 2 7) where N is number of records [ ], and A L = lake surface area [L 2 ]. In total 19 lakes were included in t his study, of which 4 listed as a source by utilities could not be found. This eliminated two urban areas from this study. Groundwater recharge and s torage Aquifer information obtained from (USGS, 2008b) included location and rock type of the major water supplying aquifers within the coterminous US, while estimates of hydraulic conductivity and transmissivity for each aquifer were c ollected from an auxiliary regional groundwater analysis (USGS, 2003b) Data on aquifer saturated thickness was obtained, when av ailable, from USGS (2008a). For cases where saturated thickness was unknown, the geometric mean of the reported data (61 m) was used as a substitute. Locally available groundwater included all non fractured aquifers that intersected more than 5% of the UAB. Fractured rock aquifers, whose productivity can be extremely variable at the local scale, were excluded since information from this dataset was not of sufficient resolution to quantify whether water was available at specific urban locations. Contrib utions from these types of aquifers were only included when listed as a pri mary source by an urban utility ; otherwise, available water from these sources was assumed to be zero. While 28 urban areas utilize principal aquifers for water supply, 146 also lis ted smaller, surficial aquifers as primary water sources. Information on these non principal aquifers does not exist in any centralized database, and data regarding each is often localized, limited in its coverage, and/or difficult to access. Therefore, in cases where


44 urban areas cited a non principal aquifer as a primary source, yet also intersected a principal aquifer, the principal aquifer data was substituted for the smaller aquifer at the risk of perhaps greatly overcompensating for groundwater avail ability in some areas. When estimating the volume of groundwater available to urban areas, the following assumptions were made: 1) aquifers are unconfined, 2) pumping rates, groundwater recharge and heads are constant, 3) the cone of depression from pumpi ng saturated thickness. Based on these assumptions, the Theim equation (Gupta, 2010) was rearranged to calculate volume of groundwater naturally available to each urban area ( ). Estimation of locally available groundwater was based on the assumption that the drawdown radius was equal to the radius of the UAB intersected aquifer area (Eq. 2 8): ( 2 8) ( 2 9) where b = drawdown saturated thickn ess [L] B = initial saturated thickness [L], R = radius at B [L] r = radius at b [L] = constant pumping rate [L 3 /T] k = hydraulic conductivity [L/T], and w =groundwater recharge rate [L/T]. Captured groundwater availability (Eq. 2 9 )was as sumed to equal urban demand ( = D ) and limited only when was greater than the allowable maximum drawdown as defined by the assumptions above. Groundwater recharge, calculated as the product of mean average annual baseflow and runoff, was o btained from (USGS, 2003b) Individual grid values were


45 averaged by aquifer to calculate an annual mean recharge ( ), and r echarge variability ). These data were then used to calculate P 5 P 10 and P 2 5 exceedance probabilities for inflows to groundwater storages. W a ter A vailability Assessments For each of the 22 6 urban areas with sufficient hydrologic information, a storage based and MAR based estimate of Q A was produced. A comparison of storage based and MAR based Q A estimates was plotted against the cumulative fraction of the total US urban population in Figure 2 7 A The average for all urban areas (so lid line) was bounded on either side using 1 SD (dotted lines), and was considered a region of uncertainty, indicating potential for, but not certainty of, water scarcity issues occurring. Urban areas with Q A values in region ( H ) are considered at high r isk for water scarcity due to their low average annual availability, and urban areas with Q A in region ( L ) were classified as having few, if any, scarcity issues. Because the Q A data are non normal, the median differences between the two estimates (MAR ba sed = 4726.4 lpcd and Storage based = 15574 1 lpcd) were compared using the Mann Whitney Rank Sum Test for non parametric data. Test results indicated a significant difference between the two methods (Mann Whitney U = 10838 .000, P<0.0 0 1), allowing the fir st hypothesis of this study to be accepted. To better understand the role captured storages play in water availability assessments, Figure 2 7B shows the difference between pre and post development of captured water sources. Without additional storage s, 50% of urban ares would face a medium or higher risk of water scarcity, and 20% of urban areas would fall below the threshold representing high water scarcity (Region H). Only those urban areas who


46 have greater than 10,000 lpcd of water locally availab le are those who have not engaged in any efforts to acquire substantially larger volumes of captured water than they already have. These contrasting assessments not only emphasize the importance of the storage based approach, but provide some context for where water scarcity occurs in the US Based on the Falkenmark scarcity indicator, storage based Q A estimates place 7 5 % of the US urban population within region L indicating that the majority of urban areas do not face water availability issues on an av erage annual basis. The remaining 2 5 %, (comprised of 17 urban areas) face some level of water availability risk, including 5 % of the population ( 2 urban areas) which fall within Region H indicating a high risk of mean annual water scarcity. In compariso n, using the MAR based approach, 45 % of the population resides in Region H or M and faces urban water scarcity, and 23% were identified as being at high risk of water scarcity (Region H ). A list of urban areas found in Regions H and M is provided in Table 2 3. Using the storage based method, the urban areas listed in Region H or M are those with low local average annual water availability relative to demand. In some cases, such as Los Angeles, CA and San Diego, CA, low water availability is due to a la ck of close resource s. A rid areas such as these have typically developed substantial infrastructure networks to procure and store w ater resources, however large demands and the high cost of transport mean these urban areas do not have access to supplies t hat are substantially greater than their demands. Other urban areas in th is group, such as Orlando FL and Cleveland OH are those who have local water supplies but only limited access because these sources are relatively small, either in total volume, or in


47 the amount that can be withdrawn under the current assumptions. In these cases, storages may be small, yet frequently recharged or may be already allocated to other purposes, which masks their vulnerability in assessments that use a MAR based approac h. While half of the 17 urban areas in Regions M and H using the storage based approach are also classified as water scarce using the MAR based method 8 of these urban areas, s uch as Miami, FL and Buffalo, NY were uniquely identified as being water sc arce using the storage based calculations. This difference is highlighted in Figure 2 7 which compares the Q A values calculated for each urban area by storage based vs. MAR based assessments. With the MAR based assessment, Miami (yellow dot) is classifi ed as having low risk of water scarcity due to the high levels of recharge over the Biscayne Aquifer; however using the storage based approach, Miami is at risk of water scarcity, because the aquifer is relatively small and shallow. The differences betwee n these storage based and MAR based assessments illustrate the importance of including both flows and storages as components of the urban water budget. Water Vulnerability Assessments Water vulnerability assessments were described as a function of source inflow variability measured as the ratio of the P 5 P 10 and P 25 exceedance probabilities to the and the average annual water availability When analyzing vulnerability outcomes, it was noticed that very few urban areas crossed any thresholds when vulnerability was calculated for any of the exceedance probabilities. In addition, the avera ge percent differences between the exceedance frequency levels were not large ( P 5 and P 25 = 37 21%, P 10 and P 25 = 28 18 %, P 5 and P10=13 13 %). Therefore, to simplify results,


48 the average of the exceedance probabilities reported for each urban area was inste ad used as the final measure of source inflow variability. Results of the vulnerability assessment are shown in Figure 2 8. The Falkenmark scarcity ratio was employed again to help identify where low vs. high water availability exists on the x axis, wh ereas the vulnerability measurements have a built in ratio, as each is defined as the exceedance frequency divided by the The natural break point for the y axis thus existed when this ratio was equal to 1, or when the average exceedance frequen cy volume was equal to the minimum available water required for each urban area based on average annual urban demand. Variability values higher than 1 indicated that exceedance frequencies were larger than the thus having a relatively low measu re of variability. Values that were less than 1 indicated the opposite, and represented sources with high variability. Combined, these two lines delineate the vulnerability plot into four quadrants which represent zones of high availability and low varia bility ( Quadrant 1), high availability and high variability ( Quadrant 2), low availability and low variability ( Quadrant 3), and low availability and high variability ( Quadrant 4). T hose urban areas with high availability and low variability (1) are the l east vulnerable according t o this scale, whereas those in Q uadrant 4 are the most vulnerable. As can be seen in Figure 2 8, the majority of urban areas ( 170 ) reside within Q uadrant 1, suggesting that 43% of the total US population faces little to no water vulnerability issues Quadrant 2 contained 37 urban areas, and 37% of the national population, whereas Quadrants 3 and 4 held the smallest number of urban areas (3 and 6, respectively) and the smallest percentage of the national population (8% and 12%, r espectively ). When comparing the percent of the population with water availability vs


49 water vulnerability issues, the percent of people at risk increases substantially, from 25% to 57%, respectively. From the list of urban areas found to be vulnerable ( Table 2 4), most of those locations reported to be most vulnerable (Quad 4) were not surprising. Many of the urban areas from California rely on regional water projects for supplies because they do not have ample resources locally available. Urban areas in Quad 3 represent those cities with low water availability, but also with low water variability. Miami, FL, falls into this category beause of its small shallow aquifer, with high rain fed recharge. Brownsville, TX and Chicago, IL make this quadrant b ecause while their storages are relatively large, the volume available to each is limited, as both storages are allocated for many other purposes other than water supply to this city. The largest category of water vulnerable regions was that of Quad 2, wh ich represented urban areas with ample amounts of water available, but whose sources experience a high degree of variability throughout the year. Included in this category are the urban areas Atlanta, GA and Tampa, FL, both of whom have made major headl ines in recent news for facing water scarcity issues despite being located in humid climates. To verify the results of this vulnerability analysis, we employed an alternate, objective method for estimating vulnerability in urban areas. This method was a t ext analysis designed to serve as a qualitative proxy for the degree of vulnerability severity in our vulnerability assessments. The text analysis was comprised of a survey of the Google News Archive database, which was mined to find the frequency with wh ich news articles covering vulnerability issues about specific urban areas occurred. This database search included all forms of news related media between 1/1/1980 8/31/2011


50 for each urban area. Relative vulnerability was measured by searching this dat abase using a combination of terms most commonly found in vulnerability related articles. The number of articles returned for each search string was recorded for every urban area, and the average number of articles across these searches, normalized accordi ng to the urban population size, was then used as a proxy for vulnerability where higher numbers of article hits indicated higher vulnerability To control for the failure of each search to capture the exact number of vulnerability specific articles for each urban area, we used and Drought could be produced using this method. Results from these se arches (Fig. 2 9) showed that indeed, the average number of articles found generally increased with vulnerability severity. A less clear distinction was found between Quads 2 and 3, however in reality, the differences between them are quite nuanced as the different ways. Therefore it was unsurprising that a text analysis would not be able to find significant differences between the groups. Significant differences were measured between the different vulnerability quadrants using the Mann Whitney Rank Sum test, where median values for each quadrant were as follows: Quad 1=1.348, Quad 2=2.876, Quad 1=5.402, and Quad 1=8.598. Significant differences were found between Quads 1 and 2 (U=2068, p<0.001), 1 and 3 (U=298, p=0.021), 1 and 4 (U=136, p=0.001), and Quads 2 and 4 (U=71, p=0.015). These statistical differences verify that the methods used to calculate vulnerability do reflect reality to some degree and confirm the second


51 hypothesis. A larger searc h with more refined terms may bring forth more differences between categories that currently lack significant differences. Conclusions The American water industry has identified urban water availability as top concern for utilities as of 2008. These con cerns reflect the lack of an accurate and standardized method for quantifying urban water availability and vulnerability at the national level. Utilities, the government and academia have all attempted to fill water availability / vulnerability gaps, howev er differences in scales and methods remain vast, and data remains sparse. Of the water availability / vulnerability studies that currently exist, most assessments have measured these parameters in terms of MAR, ignoring changes in storage. In reality, st orages play an important role in nearly all urban utilities supply portfolios, and exclusion of these components seems spurious, at best. As such, this study attempted to fill this gap by creating and implementing a storage based approach for measuring wa ter availability and vulnerability In doing such an assessment, it was found that 25% of the US population resides in an urban area where average annual water availability is an issue, and 5 % of the population lives in an urban area where water availabili ty is highly inadequate for the current demands placed on the system. The differences between MAR and storage based estimates of Q A were compared in this study as well, revealing that significant differences do exist between the two methods, with storage based Q A areas than the MAR based assessment. In response to a lack of historical data for urban water resources at the national level, an alternate measure of urban vulnerability based on the probability of th e inflows being able to add sufficiently to storages when reduced to P 5 P 10 and P 25 was


52 developed. Water vulnerability was analyzed according to average of these three estimates of vulnerability severity, and was validated using a text analysis of Google news articles about water scarcity issues in urban areas. The frequency with which media reported vulnerability issues in each urban area over the past 20 years served as a proxy for severity of vulnerability Results from this analysis showed that the number of news articles returned for each quadrant increased as the relative severity of the vulnerability increased. Furthermore, there was a statistically significant difference between these groups, showing that indeed the vulnerability metric created here reflects reality to some extent. While the text analysis performed here was rather simple in nature, using four search terms to capture vulnerability issues in the news, an expanded search may yield greater differences between each of the groups. F inally, water availability and vulnerability play a critical role in defining and implementing sustainable water management practices. Standardized assessments, such as these, within a national benchmarking framework could be a useful tool for decision m akers not only at the local urban level, but would help water managers at all levels better understand how, and where water scarcity issues arise. To do this, however, more data is required. During the data collection process of this study it became very clear that many of the variables needed to make accurate assessments of water availability and vulnerability were often only locally available, available only in very general terms, or unavailable. As such, these assessments required simplifying assumpti ons that better data would alleviate.. However, this study demonstrated that water availability and vulnerability analysis that accounts for storages is possible, insightful, and often crucial. The authors hope this study will highlight the issues


53 regardi ng data availability and encourage a water availability / vulnerability benchmarking initiative based on better data colle ction, monitoring and sharing.


54 Table 2 1. Types and sources of d ata Data Type Source* Website Aquifer Characteristics USGS http:// Aquifer Recharge USGS Lakes None No available dataset Reservoir Characteristics NID River Flow USGS http://water.usg River Location USGS Urban Area Boundaries USCB Urban Area Population USCB Water Utility Charc teristics Water Utilities Individual Consumer Confidence Reports Urban Water Demand USGS *USGS US Geological Survey, NID National Inventory of Dams, USCB US Census Bureau


55 Table 2 2. Observed r ese rvoir and s treamgage information Region Total Reservoirs Reservoirs with Gages % Reservoirs Represented Correction Factor 1 145 1 0.01 --2 136 11 0.08 --3 76 0 0.00 --4 18 2 0.11 0.77 5 11 2 0.18 0.72 6 2 0 0.00 --7 14 2 0.14 0.87 8 3 1 0.3 3 --9 0 0 1.00 --10 69 3 0.04 2.22 11 68 4 0.06 0.93 12 71 11 0.15 21.54 13 23 0 0.00 --14 26 4 0.15 1.22 15 29 1 0.03 --16 24 0 0.00 --17 46 3 0.07 --18 345 10 0.03 --


56 Table 2 3. Water availability estimates risk of water scarcity Water Availability (lpcd) Risk ( ) ID Urban Area MAR Storage MAR Storage 78904 San Francisco -Oakland, CA 409.7 354.4 H H 88732 Tucson, AZ 560.6 2915.4 H L 18856 Colorado Springs, CO 135.6 3285.9 H L 22042 Dallas -Fort Wor th -Arlington, TX 575.8 3372.1 H L 01171 Albuquerque, NM 532.6 3998.7 H L 69184 Phoenix -Mesa, AZ 479.4 4123.9 H L 65080 Oklahoma City, OK 621.3 4838.2 H L 04384 Austin, TX 404.3 5071.5 H L 95077 Wichita, KS 378.6 5337.2 H L 23527 Denver -Aurora, CO 103.8 7042.7 H L 51877 Lubbock, TX 532.1 8126.1 H L 01927 Amarillo, TX 622.6 9420.0 H L 29089 Fargo, ND -MN 705.3 10106.7 H L 72613 Pueblo, CO 104.2 12758.2 H L 47854 Laredo, TX 497.7 14192.1 H L 00280 Abilene, TX 302.8 14488.0 H L 64864 Odessa, TX 401.9 37702.8 H L 47962 Las Vegas, NV 269.9 45385.1 H L 09298 Boulder, CO 193.2 47008.8 H L 30628 Fort Collins, CO 122.0 56713.0 H L 78661 San Diego, CA 273.7 854.3 H M 51445 Los Angeles -Long Beach -Santa Ana, CA 263.6 880.6 H M 79039 San Jose, CA 4 43.0 1279.4 H M 57709 Mission Viejo, CA 266.0 1426.1 H M 27253 El Paso, TX -NM 582.1 1582.3 H M 78580 San Antonio, TX 391.2 1634.3 H M 75340 Riverside -San Bernardino, CA 597.1 1786.2 H M 78499 Salt Lake City, UT 632.8 1857.2 H M 56602 Miami, FL 2573 .8 644.6 L H 17668 Cleveland, OH 2815.7 1658.4 L M 11350 Buffalo, NY 4357.1 1985.7 L M 65863 Orlando, FL 3415.7 2109.5 L M 15508 Charleston -North Charleston, SC 7302.1 2161.3 L M 19504 Concord, CA 867.4 2191.2 M L 86599 Tampa -St. Petersburg, FL 203 9.5 3038.0 M L 63217 New York -Newark, NY -NJ -CT 1468.6 3110.5 M L 66673 Oxnard, CA 1448.7 3267.7 M L 57466 Milwaukee, WI 2045.9 3314.2 M L 47611 Lancaster -Palmdale, CA 1670.5 4026.8 M L


57 19234 Columbus, OH 2141.3 4093.5 M L 87004 Temecula -Murrieta CA 1281.3 4195.8 M L 92242 Washington, DC -VA -MD 1633.7 4298.2 M L 64945 Ogden -Layton, UT 2027.9 4593.7 M L 44992 Killeen, TX 1013.7 5466.0 M L 41347 Indio -Cathedral City -Palm Springs, CA 1224.8 5603.8 M L 79309 Santa Clarita, CA 1526.1 5685.2 M L 79282 Santa Barbara, CA 1119.5 5820.9 M L 04681 Bakersfield, CA 1657.8 5858.8 M L 23824 Detroit, MI 1112.8 6435.3 M L 87490 Thousand Oaks, CA 2035.1 6447.0 M L 78310 Salinas, CA 2159.2 6519.4 M L 82144 Simi Valley, CA 1485.9 8264.5 M L 57628 Minn eapolis -St. Paul, MN 1125.9 8296.9 M L 40429 Houston, TX 2142.2 9432.0 M L 31843 Fresno, CA 1410.6 10755.1 M L 38215 Hemet, CA 1455.0 11398.8 M L 23500 Denton -Lewisville, TX 1307.4 16567.5 M L 77068 Sacramento, CA 2161.9 19539.3 M L 23743 Des Moine s, IA 1593.6 28016.1 M L 65269 Omaha, NE -IA 786.6 54296.0 M L 88084 Topeka, KS 1910.5 74510.6 M L 18748 College Station -Bryan, TX 1610.5 83245.8 M L 77770 St. Louis, MO -IL 1957.2 108590.3 M L 90028 Vallejo, CA 1086.0 186746.4 M L 28657 Fairfield, CA 1265.7 275694.7 M L 49933 Lincoln, NE 1387.2 761.0 M M 16264 Chicago, IL -IN 1376.5 1404.7 M M 47935 Las Cruces, NM 1517.1 2156.0 M M


58 Table 2 4. Water v ulnerability of u rban a reas Vulnerability Indicators ID Urban Area Quad Availability Va riability 49933 Lincoln, NE 4 Low High 51445 Los Angeles -Long Beach -Santa Ana, CA 4 Low High 57709 Mission Viejo, CA 4 Low High 78661 San Diego, CA 4 Low High 78904 San Francisco -Oakland, CA 4 Low High 79039 San Jose, CA 4 Low High 10972 Brownsvi lle, TX 3 Low Low 16264 Chicago, IL -IN 3 Low Low 56602 Miami, FL 3 Low Low 00280 Abilene, TX 2 High High 03817 Atlanta, GA 2 High High 04384 Austin, TX 2 High High 04843 Baltimore, MD 2 High High 08785 Boise City, ID 2 High High 09271 Boston, MA -NH -RI 2 High High 10162 Bridgeport -Stamford, CT -NY 2 High High 18856 Colorado Springs, CO 2 High High 19504 Concord, CA 2 High High 22042 Dallas -Fort Worth -Arlington, TX 2 High High 22528 Dayton, OH 2 High High 23500 Denton -Lewisville, TX 2 Hig h High 23527 Denver -Aurora, CO 2 High High 29089 Fargo, ND -MN 2 High High 35461 Greenville, SC 2 High High 41212 Indianapolis, IN 2 High High 42346 Jacksonville, FL 2 High High 44992 Killeen, TX 2 High High 47962 Las Vegas, NV 2 High High 60733 M urfreesboro, TN 2 High High 63217 New York -Newark, NY -NJ -CT 2 High High 64945 Ogden -Layton, UT 2 High High 65863 Orlando, FL 2 High High 66673 Oxnard, CA 2 High High 69076 Philadelphia, PA -NJ -DE -MD 2 High High 69184 Phoenix -Mesa, AZ 2 High Hi gh 72505 Providence, RI -MA 2 High High 73261 Raleigh, NC 2 High High 74179 Reno, NV 2 High High 75340 Riverside -San Bernardino, CA 2 High High 78499 Salt Lake City, UT 2 High High


59 78580 San Antonio, TX 2 High High 83548 Spartanburg, SC 2 High High 83899 Springfield, IL 2 High High 88732 Tucson, AZ 2 High High 89326 Tyler, TX 2 High High 92242 Washington, DC -VA -MD 2 High High


60 F igure 2 1. Location of urban areas (yellow) and associated water resources. Figure 2 2 Conceptual diagram describing local vs. captured water sources [Adapted from Carrera, 2010].


61 Figure 2 3 Relationship between City, Urban Area and Metropolitan Statistical Area populations and urban utility service populations.


62 Figure 2 4. River discharge stream order relationship scaled to reflect that 75%is available for urban use. The dotted line indicates the limit of stream order usability to an urban area Figure 2 5. Location of water resource regions used in Vogel et al. (1999).


63 A B Figure 2 6. Differences in reservoir inflows using two methods. (A) O bserved reservoir inflows and (B) R = 0.8238 R = 0.8647 R = 0.734 0.1 1 10 100 1000 0.1 1 10 100 1000 Estimated Streamflow (m3/s) Observed Streamflow (m3/s) R = 0.8319 R = 0.8647 R = 0.734 0.1 1 10 100 1000 10000 0.1 1 10 100 1000 Corrected Estimated Streamflow (m3/s) Observed Streamflow (m3/s)


64 A B Figure 2 7. Water availability estimates using a storage based (filled circles) or MAR based approa ch (open diamonds) (A) Yellow circles show Q A estimates for Miami, FL. Relative change in water availability pre and post addition of captured sources for the storage based method ( B) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 Population (% of Total) Water Availability Q A [lpcd] MAR Storage H M L 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 Population (% of Total) Water Availability Q A [lpcd] Pre Post H M L


65 Figure 2 8. Water vulnerability as a function of inflow variability and water availability. Solid lines represent the minimum available water required based on average annual urban demands ). Figure 2 9. Text analysis results for each level of vulnerability Letters indicate statistically significant differences. 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 1 2 3 4 Avg # of Articles Vulnerability Quadrant a b b,c c,d 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 Normaliz ed Water Variability QA / Q A ( ) Water Availability Q A (lpcd) 1 3 4 2


66 CHAPTER 3 URBAN ADAPTATIONS TO WATER SCARCITY: MANA GEMENT AND INFRASTRUCTURE Introduction Over the past century, the effects of urbanization in the United States (US) have been profound. Urban populations have come to represent the majority of the population (US Census Bureau, 1995) and, in concert with this development, the accelerated growth of urban processes has not only drastically altered the way in which we view and use the land (Alig et al. 2004; Crane and K inzig, 2005; Grimm et al. 2008) but has changed how we function as a society (Kates and Parris, 2003) These effects are part icularly apparent when examining US urban water supply systems, where large water infrastructure systems and complex water institutions are now found in many urban areas. The form and structure of these systems, however, remains poorly understood from a qu antitative perspective. Very few empirical studies have been devoted to resolving how urban water management systems differ from one another, and how they support conflict or cooperation over water resource management. As growth continues, urban water av ailability issues will not only persist, but be exacerbated with additional confounding forces, most notably the uncertainties associated with climate change. It is therefore prudent to invest further efforts in moving past theoretical speculation, and t owards more quantitative analysis in order to better understand these systems. Urban centers have long been considered hubs of innovation and production because they make it possible for economies of scale to exist within infrastructure, social services an d governance (Bettencourt, Lobo, and Strumsky, 2007) In the US, urban areas are extremely productive; they generate approximatel y 90% of the total


67 national GDP (U.S. BEA, 2011) and hold greater than 80% of the total population (US Census Bureau, 2000a) In order to maintain such high levels of productivity, these systems must also be localized centers of mass consumption. Urban areas occupy approximately 3% of the total US land area (USDA Economic Research Service, 2002) and therefore depend heavily on externall y located natural resources. As such urban areas have invested heavily in developing technology to secure the resources needed to meet and maintain these steadily increasing levels of production, although often to the detriment of the environment (NRC, 1999) Despite constantly growing needs, over the years urban areas have continued to successfully exploit resources, despite their seemingly unsustainable rate of consumption (Batty, 2008) In the US however, the growing uncertainty surrounding future urban w ater availability has, for many water providers, become a primary issue of concern (Runge and Mann, 2008) There are many challen ges facing urban systems today with regards to climate change, population growth and environmental regulations (Ginley and Ralston, 2010; Means, West, et al. 2005) yet practical solutions for how to resolve these issues remain obscured. Over the past century, the location of urban development h as been more dependent on a complex mixture of opportunities than on access to resources given the advancements in technology and supply chain efficiencies that have reduced the cost of transporting resources over long distances (Seto et al. 2010) Water is an exception, however. It is a critical resource for all urban development, yet is not easily transported over long distance s, and its local availability can be highly variable. Traditionally, urban areas have relied on technology to secure additional water for growth. For at least the last half of the 20 th century, the development and construction of large infrastructure,


68 suc h as aqueducts, dams, and well fields, has been one of the primary strategies used by urban areas to secure additional water resources (Gleick, 2000) Over this time, thousands of large dams and other types of water infrastructure were built around the US to transfer, store and regulate water resources. These developments have deeply interlinked the development of water infrastruc ture with water management as the scale of supply needs increased in what is referred to as traditional, or supply, management (Ausubel, 1988; Blake, 1956) Concurrently, this type of infrastructure has also helped solidify the top down style of water management associated with many urban areas toda y (Ausubel, 1988) Due to the extremely capital intensive nature of these projects, it became necessary to utilize public investm ents through governmental involvement (Baumann et al. 1998) As large infrastructure solutions to water supply deficits became t he norm, regulation and control of major water resources also came under the centralized control of state and federal institutions (Davis, 1968; Lee, 1999) Where this top down approach has been implemented, the foundations have been laid for inter urban and even regional management solutions for securing water resources. The institutional structure, including the policies, regulations, and rules for water supply management, for these systems is not well documented at the national level, however. This matter becomes particularly important in lig ht of the fact that there currently exists no national policy for managing water (Stakhiv, 2003) The major responsibility for cr eating and enforcing laws, policies and rule for management are left to the state and local governments, yet how they differ between systems is not well understood (Ausubel, 1988; Blake, 1956) Specific knowledge regarding water management strategies used by urban areas is neither well documented n or well


69 classified in the literature at the national scale; individual urban water suppliers alone maintain records of this information. In addition, comparative characterizations of water institutions important to urban water supply at the national level are also non existent, except on a case by case study (Blomquist et al. 2004; Gerlak, 2008; Heikkila, 2004; Scholz and Stiftel, 2005) As urban needs for water continue to grow, knowledge of how different locations have responded to stress could be useful for identifying successful strategies for future problems. While undeniably complex and unique in the specifics of development and adaptation, at a fundamental level urban areas all bear a general resemblance to each other in terms of function and pattern (Harris and Ullman, 19 45) However, to study these general relationships, much research regarding urban form and function has been explored through the use of scaling analyses. Urban scaling has been used to describe many quantitative regularities common to all urban areas s uch as degrees of economic advantage (Glaese r, 1994; Lucas, 1988) levels of innovation (Baldridge and Burnham, 1975; Bettencourt, Lobo, and Strumsky, 2007) and size of supply networks (Kuhnert et al. 2006) This body of knowledge has in turn produced many fundamental quantitative insights about the predictability of underlying urban processes related to production and consumption (Bettencourt, Lobo, Helbing, et al. 2007) yet its application to water management has yet to be thoroughly implemented. Management Complexity To better understand the relationship between natural resource availab ility, urban management and sustainability, the theoretical framework developed by Allen et al (1999) is employed. Based heavily on ideas from The Collapse of Complex Societies by Tainter (1988) Allen et al. (1999) discuss natural resources acquisit ion in the context


70 relationship to renewable resources is essentially a positive feedback loop; as a society grows, it requires more resources to maintain itself. In re sponse to the constantly increasing need for resources, societies have managed growth by developing increasingly elaborate infrastructure to gather, store and provide natural resources. Using this strategy, successive acquisition problems become progressi vely more challenging as each subsequent elaboration adds to the cost a society must pay, either in terms of work or sacrifices that must be made, to maintain growth in this manner. Where at first this type of problem solving stands out as innovative and yields significant benefits to society, over time when re applied to each new problem encountered, the system becomes increasingly complicated, and the efficacy of these solutions diminishes as costs rise. At some point, the burden associated with the gro wth and maintenance of continued elaboration can become too great to bear, and the society collapses as its members can no longer afford, or believe in the worth of, the system. Allen et al. (1999) current natural resource management strategies via two distinct pathways: the el elaboration of organization (which is refer red to as increasing complexity) (Fig ure 3 1). In this work, structural elaboration is very similar to the general concept of elab oration presented by Tainter. Over time, structural elaboration steadily develops into a more complicated infrastructure with increasing costs and diminishing returns within the context of the current management/ organizational framework. These types of elaboration can be seen throughout water management history. The era of large dam


71 construction (1950s 1980s) represents one such example, where the cost to benefit ratio of using large reservoirs to meet water needs diminished as appropriate sites became scarce and environmental damages accrued. The elaboration of organization, or the increased complexity of a system, refers to the increased cost required to run and maintain a more highly organized management system. Elaboration of organization is infre quent relative to elaboration of infrastructure and occurs when the costs of diminishing returns on structural elaboration become too great. This type of with the elabora tion of structure by reframing the resource allocation problem. The new level of organization is an increase in the cost of organization, and through it, resource availability to be redefined such that it reduces the complicatedness of the system by reint erpreting the relationship between the cost and benefit of structural elaboration. Theories about natural resource management, such as those provided by Tainter and Allen, are supported in part by recorded historic examples where the growth of particular US urban areas has been a function of the infrastructural and organizational adaptation strategies to low water availability. There have been no national, quantitative analyses of the types or locations where elaboration, as an adaptive strategy, has occ urred This work attempts to fill this gap by investigating the scalability of urban water availability and management elaboration with regards to three hypotheses: 1. T he degree of infrastructural elaboration in each urban area will be inversely related t o the local water availability, 2. T he degree of organizational elaboration in each urban area will be inversely related to the local water availability,


72 3. H igh levels of infrastructural and organizational elaboration will be found where urban areas must comp ete for water resources. To address these hypotheses, the objectives of this work are to 1) quantitatively assess the degree to which urban areas have adapted to water scarcity using strategies based on infrastructural and/or organizational complexity w ith respect to their local water availability, and 2) examine how the implementation of these adaptation strategies might affect neighboring urban areas. Methods Power Law Scaling Analyses Power law scaling functions have been found to apply to a large nu mber of processes ranging from biological and chemical, to technological and informational, to economic and social (Boccaletti et al. 2006; Amaral et al. 1998) In urban systems studies, this type of scaling effect has been found to occur for a wide variety of urban metrics, including transportation networks (Lmmer et al. 2006) innovation (Bettencourt, Lobo, Helbing, et al. 2007) and economic growth (Glaeser, 1994) No studies have yet attempted to examine how various aspects of urban water provision scale in urban systems This absence in the urban scaling literature has provided the impetus for this more thorough examination of how wate r utilities and water provision fit into the urban processes paradigm. The relationship between the volume of locally available water and the degree of infrastructural and organizational elaboration experienced by an urban area was determined using a pow er law scaling analysis where the locally available water,


73 LWA(t), represents a measure of urban water availability at time t such that power law scaling exists as: (3 1) where is the measure of infrastructural or organizational elaboration, is a normalization constant, and represents the general dynamic rules governing these urban indicators. This method of analysis is us eful not only for identifying patterns between urban processes and size, but provides a means with which to identify and group particular trends. The value has been used by Bettencourt, Lobo, Helbing, et al. (2007) to describe three general types of growth trends in urban processes across urban size; sub linear ( <1), linear ( =1), and supralinear ( >1) (Figure 3 2) Linear scaling patterns were found in processes that reflected individual human needs, such as housing, electricity and water consumption. Sublinear scaling is most closely associated with those t raditional or infrastructure based processes that operate according to economies of scale, while supralinear trends tend to reflect urban processes related to social innovation and information and creation. These trends in scaling represent a means for classifying types of urban processes and provide insight into the sustainability of these processes with respect to resources required. Sublinear processes tend to use resources more efficiently per unit increase, whereas linear and particularly supraline ar processes, due to their accelerated pace of growth use more resource s the larger they become. When resources become scarce, supralinearly scaling (and to a lesser extent linearly scaling) processes must reorganize or collapse. In this work, scaling pa tterns will be used to identify whether infrastructural and


74 organizational elaboration is a scalable process across urban size, and scaling trends will provide insight into the types of processes being studied. Urban scaling analyses require data coverage across an entire system of interest. For this study, the sites under investigation were all those urban areas in the coterminous US, as identified by the US Census Bureau, with populations greater than 100,000 (US Census Bureau, 2000b) While there is information available about water supply and management for specific US urban areas, only data that are easily and commonly availabl e at a national level were applicable. Infrastructural data for this study came from the urban hydrogeographic and water supply information databases created in Chapter 1 Information on the organizational metrics associated with each urban area was mined from utility websites, and previous studies. Urban Water Availability Urban water availability refers to all the potentially usable water sources an urban 5km buffer around the actual urban perimeter, designated as such to unify fragmented polygons and to ensure that adjacent local sources were included. Henceforth, all references to the UAB are assumed to include this 5km buffer. The hydrogeographical data used to calculate urban water availability included spatial information and select hydrologic characteristics of the major rivers and streams, aquifers, lakes, and reservoirs with in the US. Specific information regarding the operation and management most recent Water Quality/Consumer Confidence Report. Local sources for each urban area were de fined as all hydrologic features that intersected or bordered the urban area


75 boundary (UAB). Captured sources were those that did not exist within the UAB, but were specifically listed as a primary source of water for the urban area. Distances to source s were measured radially from the nearest edge of the UAB. The volumes of water available from local sources (LWA) and captured sources (CWA) for each urban area were estimated by the equations developed in Ch 1 (Eq 8) using information from the aforemen tioned datasets: (3 2) and = mean annual local streamflow [L 3 /T], = mean annual captured streamflow [L 3 /T], = mean annual local groundwater availability [L 3 /T], = mean annual captured groundwater availability [L 3 /T], = mean annual captured natural lake availability [L 3 /T], = mean annual local reservoir availability [L 3 /T], = mean annual captured reservoir availability [L 3 /T]. As noted in Ch 1, in some instances, considerable simplifications and assumptions were required to accommodate the use of limited, localized, inaccessible, and occasionally non existent data. The vol umes of captured and locally available water for each urban area are shown in Figure 3 3 Infrastructure M etrics Unlike the hydrologic data described above, national datasets providing detailed, comprehensive information on water infrastructure systems in US urban areas are not available. According to the work by Allen et al. (1999), complexity increases either with time, or with size. As historic information was decidedly more difficult to find, three parameters representing general aspects of current wa ter infrastructure size were quantified for each urban area: 1) the relative volume of captured water available, 2) the


76 overall distance to captured sources, and 3) the total number of sources used by the urban area. For every urban area, an increase in either volume, distance to source, or number of sources represents more infrastructure to be managed, and is therefore assumed to be increasingly complex. Larger total demands, regardless of LWA, require urban areas to have larger volumes of water availab le. Captured water represents the additional volume of water required, beyond that which is locally available, to meet the water availability equations described above However, to more clearly identify the importance of the captured water sources to the total urban system, the volume of available captured water was instead evaluated as the ratio of the total water available (TWA) to LWA, where TWA is the sum of LWA and CWA. For the second parameter listed, distances from urban areas to their water supply sources were used to represent the extent an urban area must reach to maintain adequate supplies of water. Longer distances require more infrastructure, thus mak ing water systems more complex. These distances were measured as the radial length from the UAB to the source using the Near analysis in ArcGIS 10. The overall distance to captured sources was measured as the average distance from an urban area to its so urces. The number of sources used by each urban area was collected directly from utility websites and CCR reports. Only sources specifically listed as water supply were counted; small storage reservoirs for holding treated water (i.e. water towers, treat ed water storage tanks, etc.) and other infrastructure used for water distribution purposes were excluded from this total. The degree of infrastructural elaboration for each urban area was based on the assumption that there is a direct relationship betwee n the level


77 of elaboration and the amount of infrastructure maintained by an urban area. Therefore, urban areas with relatively higher captured water volumes, distances to captured sources and/or number of sources were also assumed to have relatively high er degrees of infrastructural elaboration. Organization Metrics Organizational elaboration in the context of this paper refers to the general water management framework within which an urban area operates, and theoretically includes all of the legal, polit ical, and institutional aspects associated with each. Work characterizing organizational elaboration in US water management systems does not exist, as there is no single or standardized methodology used for managing water, and the role of subjective force s is profound (cultural paradigms, management choices, political climate) (Saleth and Dinar, 2004) While capturing a complete pic ture of the organization framework of each urban area is beyond the scope of this paper, a broad characterization of each function of their 1) institutional and 2) governance frameworks. Institutio nal f rameworks The definition of the institutional framework is therefore focused on two key factors; the way in which water supplies are accessed and operated. A scale for identifying the different levels of institutional organization, the Water Acquisit ion and Managemennt (WAM) Scale, was adapted from work done by (Carrera, 2010) who examined adaptation strategies associated with water scarcity issues by measuring the level of institutional organization for large urban agglomerations in Africa. For each urban area, information describing the number of providers and the extent to which the upplies was assessed from utility websites


78 (Table 3 1). This scale was designed based on the relative complexity of the management system being used. The rationale behind this scale is based on the idea that systems who must share sources must engage in a more complicated set of rules for allocation, than those who control their own sources. It is assumed that as the number of systems sharing a source increases, the complexity of that management system does as well. Therefore, the elaboration, or increa sing complexity, of a system was ranked from 1 (lowest) to 5 (highest) based on the following five characteristics: The urban area 1) takes on complete responsibility of water provision using local resources it manages independently 2) Enters a partnership with another urban area to co own/co operate water supplies 3) buys/contracts water from a local provider 4) buys/contracts water from a regional provider 5) buys/contracts water from more than any water provider that sells distribution system. In cases where the urban area is an agglomeration of cities, there may be more than one utility provider and all ma jor providers were included. Governance f rameworks Similarly, a comprehensive account of individual water governance frameworks for each urban area was beyond the scope of this work. Instead, the governance framework was limited to certain aspects of the legal and political system within which each urban water system operates. A scale for measuring differences in governance (Flood, 1990) Based on the assumption that amount of oversight, as measured by the level of legal and political involvement in the water management process, is proportional to elab orateness of the organization, each state was scored from 1 to 5, with a higher


79 score representing a more elaborate organization. Each state started the assessment with one point, and additional points were assigned for each of the following four characte ristics: 1) permits are required for surface water or groundwater use 2) participation in one or more treaties and/or compacts occurs, 3) use of a special, water focused judicial system to mediate allocation issues 4) legislative involvement in the water r ights system. Adaptation Strategies While urban growth manifests itself in a variety of ways (i.e. increased wealth, demands, size), their tendency to expand in total area can create potential problems with regards to water resource acquisition and allo cation. This is particularly true in locations were urban areas are in close proximity, are low in natural water availability, or both. When they occur, these proximity issues create opportunities for either cooperative or conflictive water management so lutions. Resolving issues such as these has often required new solutions to water management in which agreements between users must be created to define how resources will be divided among users. In cases where urban areas have introduced to implement th e infrastructure and management organization needed to operate and maintain a cooperative system. To identify where potential management conflicts over resources might be occurring, the average distance to water sources utilized by each urban area was ma pped in ArcGIS 10. For each case where captured water was a part of an urban radius equal to the average distance to these sources was created. This buffered area was u sed to represent the current spatial range needed by each urban area to meet


80 current demands. Areas of overlap were found where intersecting distance based urban buffers occurred. Results and Discussion Water Availability For each urban area, the local (LWA) and captured (CWA) water availability were measured and used to calculate the total water available (TWA) to each location. The relationship between TWA and LWA was then used to examine the institutional and organizational elaboration each locatio n has used to seek additional supplies ( Figure 3 4 ). Of the 29 urban areas with LWA less than the national average of 600 liters per person per day (lpcd), nearly all have used captured water to augment supplies when considered in terms of TWA (those poin ts above the 1:1 and horizontal line ) For the 45% of urban areas with an excess of 10,000 lpcd, there are few if any additional inputs to local sources required (points directly on the 1:1 line). As LWA decreases below this threshold however, more urban areas are found to have developed strategies to secure water captured outside of their UAB. Though the data express substantial variability in the lower range of the x axis, a best fit trend indicates that urban areas with low LWA supplement local suppli es with enough captured water to maintain, on average, 10,000 lpcd. This 10,000 lpcd limit exhibited in this analysis provides some interesting insights into urban water management in terms of the inherent boundaries defined by urban water systems. At n o time during the data collection process did any mention of lower or upper limits of water availability occur in the urban systems studied. Yet Figure 3 4 suggests that such limits do exist in reality, despite the hydrogeographic and organizational variab ility within the dataset. At nearly double the national average, this


81 lower limit in acceptable TWA is significant, particularly for regions where local supplies are scarce. Also of interest is the broad TWA range for which urban areas have added capture d sources. Some urban areas have increased their TWA by more than two orders of magnitude, like Brownsville, TX whose substantial increase can be attributed to their access to large federally controlled dams. Smaller increases tend to be correlated to ur ban areas in Colorado and California who receive water from regional supply systems, such as the Metropolitan Water District of Southern California, who produces approximately 19 million liters of water per day for 26 member agencies (MWD, 2011) Power Law Scaling Analyses Infrastructural el aboration Before these analyses took place, steps were taken to minimize the amount of var iability associated with the datasets used to measure infrastructural elaboration. As these measurements represent the averages of the variable of interest (i.e. average distance from an urban area to each of its sources), it is expected that there will b e a larger degree of variability within each dataset than if a single parameter was measured exclusively. Variability such as this is most commonly reduced using one of the many data discretization techniques available, which converts continuous data into a finite set of intervals with minimal loss of information. Due to the highly skewed nature of these data, the equal frequency (equi depth) binning method was selected, where data were binned into categories with an equal number of data points, thus pres erving the skew of the data better than equal distance (equi distance) binning methods, which bin by interval with unequal numbers of data points in each. For this step, the average value for each bin was based on the smallest sample count allowable, n=5 Therefore data


82 presented in Figure 3 5 have been smoothed using this process before being analyzed for scaling relationships with water availability. All three parameters used to define infrastructural organization were compared against the local urba n water availability to determine scaling behavior ( Figure 3 5 ). Scaling patterns revealed that the level of elaboration is relatively simple when LWA is large, but increases exponentially as LWA approaches 0, confirming the first hypothesis. In additi on, in all three cases, a clear sublinear trend ( <1) was identified indicating that the relationship between infrastructural elaboration and LWA follows patterns of decreasing returns to scale. The trends in the data show that drier urban areas have to exponentially increase their infrastructural capac ities to secure acceptable volumes of water for their customers. The strength of the correlations between these parameters and LWA was less than is typically found in studies scaling against urban population size, measuring at = 0. 94 ( Figure 3 5 A ), = 0. 2 1 ( Figure 3 5 B ) and = 0. 66 ( Figure 3 5 C ), respectively. While lower, this overall decrease is not unexpected, as LWA is a complex measure incorporating multiple inputs of varying degrees of informational quality, unlike population size, which is a more st raightforward metric. Scaling patterns in Figure 3 5 C differed slightly from those in Figure 3 5 A and 5 B other two parameters. Since Fig 4c was a ratio representing the amount of local to captured water used, urban areas with little or no captured water all had the same ratio Figure 3 4 and in Figure 3 5 C this occurred primarily for urban areas who had >10,000 lpcd locally available. While this information is useful, show ing the range of LWA over which urban areas do not utilize captured


83 storages ( ~ 10 4 10 6 lpcd), they were excluded from the regression analysis so that only significant increases in the elaboration of captured volumes (TWA/LWA > 1.0) only occur at LWA of app roximately 10,000 lpcd or less. Organizational e laboration Urban areas were characterized according to the institutional elaboration metrics. Results revealed that the majority of urban areas (76%) exist at the simplest level of elaboration, where water provision, operation and distribution are a self contained service ( Figure 3 6 A ). The remaining 23% of urban areas were distributed with exponentially decreasing frequency among the remaining levels, with the highest institutional elaboration occurring i n the fewest urban areas ( Figure 3 6 A ). In contrast to this exponential trend, the degree of elaboration in governance displayed an approximately normal distribution, with the largest number of urban areas operating at a medium level of elaboration. Of t he components used to create the governance metric, the use of permitting was the most frequently cited component across all urban areas : 74% were found to use this as a type of governance mechanism. Over half of the urban areas were found to use special courts and legislative activity to control water management, whereas only a minimal percentage of urban areas adhered to or participated in compacts or treaties. Organizational elaboration metrics were compared to LWA ( Figure 3 6 B and 3 6 C ) and demonstrat ed similar trends to those seen for the metrics representing infrastructural elaboration in Figure 3 5 Both organizational metrics, institutional and governance elaboration, increased in complexity as LWA volumes diminished ( Figure 3 6B and 3 6 C ). For e ach metric, the mean LWA for urban areas reporting management frameworks at each level of complexity is displayed, with error bars defining the


84 standard deviation associated with this mean. While the variability in the data was large, a one way ANOVA rev ealed that there was a significant difference across the five levels of institutional elaboration (F(2, 235) = 11.220, p <0.001). Significant differences between pairs were found using the Holm Sidak method identifying significant differences (p<0.05) bet ween Level 1 and each sequential level only. Results of the governance elaboration metric failed to meet the normality criteria required for the one way ANOVA; instead significant differences between levels for these data was tested using the Kruskal Wall is method. Results from this analysis also indicated a statistical difference between the five levels of governance elaboration (H= 18.768 d.f.=4 p <0.001). A post significant differences (p<0.05) betwee n Levels 1 and 4, 1 and 5 and 2 and 5. While the differences between sequential levels is not statistically significant in most cases, the results of the ANOVA and Kruskal Wallis tests indicate that differences do exist between the maximum and minimum en d of each scale, at the very least. This significant trend of increasing elaboration with decreasing LWA partially confirms the second hypothesis, by showing the predicted trend in the data, but not meeting criteria for being statistically significant bet ween all levels measured. The fact that this hypothesis displays the expected trends is quite significant, however. While an analysis such as this has been avoided in the literature because of the nuances associated with the data collected, these results show that generalizations can be made, and do show logical trends in relation to water availability. Where water resources are scarce, urban areas have developed and implemented more complex organizational structures, both institutionally and in terms of governance. These trends further support the idea that


85 national assessments of organizational components can indeed be made, and with a certain amount of accuracy. The results from the organizational elaboration also provide corroborative evidence for the work done by Carrera (2010) While the WAM scale was shown to be usef ul for distinguishing differences in institutional complexity in foreign/developing countries, its successful application here shows that it is an appropriate tool for using in developed nations as well, and that the indices chosen can be used to place the institutional structure of water management in context, both within a single assessment, and cross studies. Given this, a comparison of institutional elaboration between the two studies ( Figure 3 7 ) shows the differences between water provision services in African vs. US urban areas. Many African areas have yet to reach their full water provision potential, however this comparison shows how complexity in African agglomerations has the potential to shift towards more elaborate management systems in the fu ture as urban areas work towards a reality of full water provision to urban residents and sustainable water management, assuming centralized management remains the dominant method for water provision. If this assumption holds true, then African urban area s would have much to learn from how water is managed in the US, both in terms of the successes and limitations that urban areas have faced throughout the course of their development, and in the face of calls for sustainable water management. Adaptation St rategies Finally, a spatial analysis was performed to better understand how infrastructural and organizational complexity is affected by proximity issues. For this study, proximity issues referred only to potential management problems due to the nearness of urban areas to each other, or the reliance of more than one urban area on a given source. To


86 identify urban areas with proximity issues, a buffered area extending from the edge of each UAB with a radius equal to the average distance to these sources wa s created. This buffered area was used to represent the current spatial range needed by each urban area to meet current demands. Areas of overlap were found where intersecting distance based urban buffers occurred ( Figure 3 8 ). In total, 1 18 urban areas in 26 groups were identified as having potential proximity issues with the number of urban areas in each ranging from 2 to 17 The three largest overlap areas with potential proximity issues were northern and southern California, containing 13 and 17 ur ban areas, respectively, and the New York City, NY region in which 14 urban areas overlap. The average radius of a proximity issue area was found to generally increase in proportion to the number of urban locations within the issue area, ranging from 8 t o 119 the number of overlapping sources confirmed this positive trend, al though showing substantial variability ( Figure 3 9 A ). The level of institutional and governance elaboration was also examined as a function of overlap ping area ( Figure 3 9 B ). On average, there was a relatively high correlation between higher levels of institutional elaboration and proximity issue areas containing larger numbers of urban areas (R 2 =0. 47 ). This was expected as the WAM scale defines institutional elaboration based on degrees to which sharing of a source or set of sources occurs. The relationship between the elaboration of governance and the size of the overlap is less clear Th e low s lope and weak R 2 limits the conclusions that can be drawn about governance and overlap


87 The relationship between the two organizational elaboration metrics, institution and governance, reveal a fairly strong positive correlation (R 2 =0.50) for these data h owever (Figure 3 10 ). These results indicate that areas with increased institutional elaboration also tend to have higher levels of governance elaboration While direct results from the elaboration of governance and overlapping proximity issues were not strong enough to resolve the third hypothesis, this relationship between the two organizational metrics provides suggest that overall organizational elaboration does provide some explanatory power for how urban systems have adapted to water stress. Conclus ions This paper sought to identify and quantitatively validate theoretical relationships between water availability and water management. According to historical information and natural resource management theory, urban water systems adapt to low water av ailability by elaborating (increasing the complexity of) their water management frameworks. In this work, two metrics, infrastructure and organization, were developed to evaluate urban water management. When compared against the local water availability of each urban area, both metrics were found to become increasingly elaborate in locations with lower local water availability. Data from this study suggests that elaboration is one mechanism urban areas use to mediate high competition for resources when w ater capture areas overlap. By mapping the results of the infrastructural and organizational elaboration analyses, the data were given a spatial framework from which comparisons with current management frameworks could be made. Mapping confirmed the clus ters of urban areas that are known to exhibit more complicated management frameworks, such as the State Water Projects and


88 Metropolitan Water Districts in California, as well as large networks of shared resources such as tho s e found among the urban areas i n Colorado. There are many studies in the literature that emphasize the important role of water management for a sustainable future, however little work has been done to quantify even the most general relationships between water availability and urban wa ter management systems. This work produced a current, national assessment of two types of adaptation strategies used by urban areas to counter problems of low water availability, and provided a simple scale for measuring the varying levels of complexity f ound within each type of strategy. Further, these metrics were found to support ideas regarding how societies adapt to problems related to natural resource management through an acceptable degree of correlation. While it is useful to be able to validate the theoretical relationships between water availability and water management at a national level, there are still many questions The quantification of any information related to US urban water management is difficult as no national database or benchmar king system currently exists to centralize and standardize (for benchmarking) these types of data. In addition, the uniqueness of each system adds much variability within each metric, reducing the analytical power of these assessments to reflect only the most general of trends, but also makes identifiable trends all that more important by showing that similarities between distinctly different systems do exist. In addition, there is no repository and a dearth of adequate data related to the adaptation stra tegies used by urban areas in the US. This not only prevents historical assessments of urban adaptations to water scarcity, but also impedes analysis of newer strategies, such as alternative water supplies and demand


89 management strategies, despite the fac t that both strategies are playing increasingly important roles in water management today. Despite the challenges inherent to executing this study, it lays important groundwork for further analyses of this nature and provides a basic database on current u rban water management to which more information can be added. While the relationship between growing cities and growing infrastructure may appear to be obvious, the results of this work imply that there is much more to learn, and that these types of analy ses offer a valuable new perspective. Future work developing more substantial metrics for measuring the urban processes driving consumption and production would potentially provide policymakers with a better understanding of how urbanization affects water resource management. More work is also needed to better identify not only how traditional adaptation strategies are used by urban areas, but how new types of management strategies, such as demand management, alternative supplies and collaboration are sha ping future responses to increased water needs. These questions are important at both at a local and a national scale, especially in light of the predicted hydrologic uncertainties due to climate change, environmental regulations and growing concerns over sustainable urban practices.


90 Table 3 1. Water Acquisition and Management (WAM) Scale US Scale International Scale Description -1 Individual collects own water from a local source -2 Group of individuals builds/invests in infrastructure to a llow easier access to a local source (self collection) -3 Urban area takes on partial responsibility of water provision using local sources distribution system for a centralized area (<50% access) 1 4 Urban area takes on complete responsibility of w ater provision using local resources distribution system for entire area (100% access) 2 5 Urban area enters partnership with another urban area to co own/co operate water supplies 3 6 Urban area buys/contracts water from a local provider (manages wat er, but no domestic distribution system) 4 7 Urban area buys/contracts water from a regional provider 5 8 Urban area buys/contracts water from 1+ regional providers -9 Urban area buys/contracts water from a national provider -10 Urban area buy s/contracts water from an foreign provider Description of terms: Urban area represents the local government or equivalent; Provider manages water supplies but operates no domestic distribution system; Local manages supplies for many smaller populati on centers, but only one urban area; Regional manages sources for more than one urban area within a defined region; National manages sources for more than one urban area in more than one region; International manages sources for urban areas in more than one country


91 Figure 3 1. Relationship between infrastructural elaboration and organizational elaboration. Adapted from Allen et al. 1999. Figure 3 2. Power law relationship of three theoretical parameters and the scaling dynamics of different v alues Power law dynamics A) showing B) supralinear, C) linear and D) sublinear patterns


92 Figure 3 3 Map of local (light blue) and captured (dark blue) water availability for each urban area in the coterminous US with a population greater than 100,000


93 Figure 3 4 Local and total water availability measurements for US urban areas. The dashed line represents the 1:1 line, and the dotted lines demarcate the average national water use rate of 600 lpcd. R = 0.9769 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 Total WA (LWA+CWA) [lpcd] LWA [lpcd]


94 A B C Figure 3 5 Elaboration of infrastructure represented by average distance to sources (A), number of sources used (B), and ratio captured storage volume (C) as a function of LWA. y = 71647x 0.94 R = 0.4066 100 1000 10000 100000 1000000 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03 Mean Dist [km] LWA [lpcd] y = 20.747x 0.211 R = 0.4772 1 10 100 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 No. Sources LWA [lpcd] y = 721.52x 0.657 R = 0.6994 1 10 100 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 TWA / LWA [ ] LWA [lpcd]


95 A) B ) C ) Figure 3 6 Percentage of urban areas that occur at each level of organizational e laboration. (A) Elaboration of organization via institutional (B) and governance (C) elaboration when compared to the amount of local water available (LWA). 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1 2 3 4 5 Urban Areas (%) Level Institutional Governance y = 73.513x 2.895 R = 0.8995 0 1 2 3 4 5 2.0 3.0 4.0 5.0 6.0 Insitutional Elaboration Log LWA [lpcd] y = 8048.1x 5.776 R = 0.9774 0 1 2 3 4 5 2.0 3.0 4.0 5.0 6.0 Governance Elaboration Log LWA [lpcd]


96 Figure 3 7 WAM Scale Percent of urban areas at each level of institutiona l elaboration. Figure 3 8 Map of urban areas and overlapping buffers. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2 3 4 6 7 8 Urban Areas (%) WAM Scale Africa US


97 A B Figure 3 9 Proximity issue correlations. Average A) distance to sources and B) elaboration for overlapping areas. Figure 3 10 Relationship between govern ance and institutional elaboration. y = 3.9689x + 6.2736 R = 0.4116 0 50 100 150 200 0 2 4 6 8 10 12 14 16 18 Avg. Distance to Sources (km) Number of Urban Areas Overlapping R = 0.4747 R = 0.1038 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0 2 4 6 8 10 12 14 16 18 Avg. Degree of Elaboration Number of Urban Areas Overlapping Institution Governance y = 0.9225x 1.7962 R = 0.5021 0.0 0.5 1.0 1.5 2.0 2.5 3.0 2.5 2.7 2.9 3.1 3.3 3.5 3.7 3.9 4.1 4.3 Elaboration of Institution Elaboration of Governance


98 CHAPTER 4 C OMPLEXITY AND COSTS IN US URBAN WATER UT ILITY SYSTEMS I ntroduction As urban areas have become prominent geographic, economic and social features of modern society, science has sought to better understand how underlying structures and processes affect the overall dynamics of urban growth. The idea that cities were complex entities was first introduced by Jacobs (1961) who proposed that cities were a integral part of urban studies. Today, the concept of complexity encompasses a wide array of characteristics, including nonlinear behavior and self organization (Baynes, 2009) Howev er, while studies in urban complexity strive to understand how cities operate as a whole, urban dynamics cannot be properly interpreted without understanding the individual processes driving growth. While many urban processes such as city size (Batty, 2008) income (Glaeser, 1994) and transportation networks (Alig et al. 2004) have been considered within the urban complexity framework, little research exists on how water supply fits into this paradigm. This work seeks to evaluate fill this gap. A n increasingly common method for assessing complexity in urban systems is scaling. Power law scaling functions can be found throughout a wide array of systems, ranging from biological and chemical, to technological and informational, to economic and social (Boccaletti et al. 2006; Amaral et al. 1998) Included in this spectrum of scalable systems are urban processes. Urban processes, or those functions that allow an urban area to perpetuate and grow, have been studied in earnest since the early 20 th (Zipf, 1949) which found that population sizes


99 of cities are inversely proportional to their rank. Since then, other research has found that scaling laws apply to a large number of physical and social urban processes when compared against u rban size (Batty, 2008; Bettencourt, Lobo, Helbing, et al. 2007; Pumain et al. 2006) generally following the power law relationship: ( 4 1) where Y = any urban metric, Y 0 = a normalization constant, N = population at a given time, and = general growth pattern across urban systems. Bettencourt et al. (2007) were among the first to perform a meta analysis on this broad spectrum of data, and identified three distinct categories of urban metrics based on their exponents. Specifically, metrics that scale supra linearly ( >1) show increasing returns with population size, and are typically represented by social processes such as innovation and wealth generation. Metrics that scale linearly with population ( =1) follow constant returns to scale and are represented individual needs s uch as water and electricity provision, housing, etc. Finally, processes that follow sub linear trends ( <1) are most closely associated with material goods and infrastructure that operate according to economies of scale. Recent work on urban scaling, su ch as that of (Bettencourt, Lobo, Helbing, et al. 2007; Bettencourt, Lobo, and Strumsky, 2007; Bettencourt et al. 2010) has focused mainly on the supralinear components of urban processes, as until recently, these processes have been previously harder to quantify and incorporate into urban theory and were thus found to be novel. These types of processes are inherently interesting because they are seen as compelling drivers of growth (Bettencourt, Lobo, Helbing, et al. 2007) In contrast, urban growth also relies on processes that follow linear and

PAGE 100

100 sublinear trends, or economies of scale. While econo mic efficiency literature suggests that water provision and supply networks operate according to economies of scale (Abbott and Coh en, 2009) research on the scaling effect of urban supply networks has included transportation (Lmmer et al. 2006) power generat ion (Kuhnert et al. 2006) and communications (Xie and Kumar, 2004) however water supply systems have only received a cursory glance ( Bettencourt, Lobo, Helbing, et al. 2007) This absence in the urban scaling literature has provided the impetus for this more thorough examination of how water utilities and water provision fit into the urban processes paradigm. The primary objectives of the current study were two fold. The first objective was to examine how urban water provision services fit into the existing urban processes paradigm. In this objective, a range of provision processes, including capital costs, revenues, and productio n were compared across water provision systems of different sizes to test for scalability. The second objective was to combine urban process information into two aggregate measures of complexity, allowing examination of system performance from both opera tional and financial perspectives. Scaling Relationships in Water Provision Processes Data for these analyses were 2000 Community Water System Survey (2000 EPA CWS Survey), a federal survey sent to commu nity water systems serving small (<500) to very large (>500,000) populations (AWWA) 2004 Water and Wastewater Rate Survey From the EPA dataset information was obtained from 1246 respondents on: 1) systems operations, including water production, storage, treatment, a nd pipe networks; and 2) systems finances, including billing structure, capital

PAGE 101

101 expenses, system revenues, and funding sources. Information on the price of water was collected from the AWWA su rvey. One limitation of using the 2000 EPA CWS Survey dataset to better understand urban scaling for water supply is that while water systems were asked to provide their system names, there was no record of which communities each system supplied. To justi fy the use of water system service population in lieu of urban population size, urban invariance based on population size ( Figure 4 1). A lower boundary for urban water sy stems (or systems that would be serving the equivalent of an urban area with a population >100,000) for this dataset was defined as those serving populations >47,000 based the relationship between a subset of urban and service populations (Fig 2 3). The data from the 2000 EPA CWS Survey, Figure 4 2 were organized into eight cost categories that were each scaled against service population size. In contrast to previous economic studies previously performed on this dataset (Norton Jr and Weber Jr, 2009; Shih et al. 2006) which examined the full r ange of water system data (as capital costs scale sublinearly within the urban range. Economies of scale ( <1) exist only for half of the costs investigated: storage (0.81), employees (0.87), routine operations (0.79) and water sources (0.83) ( Figures 4 2A 4 2D). Costs associated with improvements, repairs or expansion of distribution and transmission system s (0.99), land (0.97), treatment (1.03), and debt services (1.04) all scaled at 1 ( Figure 4 2E 4 2H). Comparison between the average values for these two groups (0.83 and 1.01 respectively) reveals that these two grou ps are significantly different (t test, p=0.009).

PAGE 102

102 While the data shown in Fig ure 4 2A 4 2D suggest economies of scale, the other capital costs ( that for the population range measured there is no increased econ omic benefit to adding additional customers. It is possible those differences in economies of scale may be insignificant over this portion of the population spectrum, or that such economies may have been exhausted. This idea is not implausible according t o a recent review of economies of scale studies (Abbott and Cohen, 2009) In addition to capital expense data, limited amounts o f revenue and operational data were also available, and were incorporated into this urban scaling study. Figure 4 3 shows select data on revenues reported. Again, differences in values were observed. Urban utilities benefit from near constant returns to scale with respect to revenue generated from connection fees ( Figure 4 3A., =1.02). Larger utilities saw smaller revenue returns from residential and commercial/industrial customers ( Figure 4 3C and 4 3D., 0.85 and 0.93), yet were also privy to la rger revenue returns from water sales to other suppliers ( Figure 4 3B., =1.17) in proportions almost equal to those missed A comparison of total costs and total expenses is shown in Figure 4 4, where the value for total expenses ( 0.91) is substant ially lower than that associated with total revenues ( 1) This relatively large gap between revenue and expense suggests that large urban utilities are not suffering from revenue deficits as smaller systems often do (EPA, 2002) According to the EPA survey report, many of the large urban providers also sell water to other smaller providers. As revenues from these types of sales w ere the only revenue type with > 1, this may explain the observed profit. Finally, scaling effects for overall water production ( Figure 4 5A and 4 5B) best

PAGE 103

103 described by 1 category (i.e. the amount of water collected and produced increases in near unit proportion with population gr owth). Two other non cost variables in this group, number of employees, and volume of treated storage, were found to scale sublinearly. Fewer employees per person served as service population size increases is direct evidence of economy of scale in manag ement, but the same for treated storage is less obvious. One explanation may be that larger providers often have the financial ability to build and maintain water treatment plants with flow capacities that well exceed average daily demands, rendering trea ted storages unnecessary. Using the AWWA survey information, information on water pricing, including the cost per 42,475 liters used, and the pricing structure implemented, w ere collected for 1 06 urban water utilities. Three types of pricing structures we re investigated, including those urban utilities following decreasing block rates (DB), increasing block rates (IB) and those using uniform pricing schemes. When regressed against service population size, no correlation between urban utility size and pric e was apparent for any of the three pricing structures examined, nor for the consumer cost of water as a whole (Figure 4 6 ) While many factors are used to determine how water prices are set and may vary according to different state and local rules, t his lack of distinction between the cost of water and the size of the utility show that neither decreasing or increasing returns to scale exist within the US urban water framework Rather, these data suggest that the cost to secure, treat and deliver water d oes not translate directly to the urban consumer nationally speaking. With the exception of water price, b ased upon the above analyses, there is reason to believe that power law relationships apply to all of the various aspects of water

PAGE 104

104 provision across system size. This concept has not been specifically addressed in the past, but is nonetheless important as part of a comprehensive understanding of urban growth and development. While urban scaling metrics were found to apply across the current data se t, it should be noted that due to the high degree of variability in the data, the rules for scaling along size may not apply in all cases. Costs and Complexities Much of the motivation behind studying urban dynamics and patterns in emerging urban form an d structure lie in the desire to better understand system complexity (Batty, 2003) Scaling principles, and in particular the exp onential function, have provided a mechanism by which complexity can be transformed into simple laws (Baynes, 2009; Chen and Zhou, 2008) While scaling principles in the urban studies literature are more frequently used for identifying urban dynamics, they are also useful for describing system co mplexity across time or size. Given a very simple definition of complexity that the act of operating and managing more infrastructure, people, and resources makes a system more complex, this portion of the work examines how complexity varies with urban w ater system size, and how it relates to the total expenses incurred. Tynan and Kingdom (2005) conducted a similar investigation on international water utility size using a set of commonly measured metrics (length of distribution network, volume of water produced, number of connections, population served, and annual costs). Using th e available data, this study investigated how operational complexity, or the complexity associated with those factors that are required for utility operation, scales in urban systems. To measure operational complexity, the following four size metrics (sim ilar to those used in Tynan and Kingdom (2005) ) were combined, rather than

PAGE 105

105 analyz ed individually, to into one measure of complexity: total length of distribution network, total water produced, total volume of treated storage and total number of employees. In doing so, the number of usable respondents dropped from the total of 1246 dow n to 142. Using these data, two measures of operational complexity were produced, total and relative, to better understand how operational complexity scales to urban utility size, and where deviations in expected complexity may be occurring, respectively. To calculate total operational complexity ( C T ), each of the per capita operational metrics were normalized from 0 to 1 according to the maximum and minimum values observed in each category, summed for each water system, and reported in relation to the g eometric mean ( C m = 0.76). The distribution of ranked C T values is shown in Fig Figure 4 7 A with 63% of the systems measuring greater than the geometric mean, and 37% measuring below. At the extremes, the systems with the relative highest and lowest C T values were 3.6 and 0.08 times the mean, respectively This particular distribution is interesting, in that it suggests that the majority of large water utility systems are operating above the expected level of complexity, or in other words, with more op erational requirements In Figure 4 7 B, a power relationship suggests that per capita costs increase with C T although the strength of this trend is only moderate (R 2 =0.45). While this result shows that urban water systems with relatively low complexity have low per capita expenses, the variability in the results increases with increasing complexity. Of particular interest are those systems with low complexity but high expenses, however based on the information provided in this dataset, reasons for these inefficiencies are currently unknown.

PAGE 106

106 As an alternate method for examining complexity and its costs in urban water systems, the relative operation al complexity ( C R ) of these systems was also analyzed. The metric C R was calculated as the sum of the ratio of the per capita value ( Y /N ) to the rate of change in each category with respect to population size. (4 2) Final values for C R were reported in relation to the total possible complexity ( C R =4) in order to present these data in approximately the same manner as C T for comparative purposes. One point to re iterat e here is that this method assumes that the reported trend characteristics for each category accurately reflect the pattern of the observed data; however from the scaling work done earlier, the large amount of variability associated with the categories mea sured suggest that these findings be interpreted with care. Given these assumptions Figure 4 8 A shows the distribution of C R values as a function of standard complexity. In contrast to the results shown in Fig ure 4 6A, the percent of urban water systems above (85%) and below (15%) average were more extreme compared to the distribution found using C T On the far ends of the C R spectrum, the highest and lowest C R values were measured as being 6.4 and 0.3 times the standard complexity, respectively. In Fig ure 4 8 B, a similar trend to Figure 4 7 B is seen in the relationship between total per capita costs and C R however the data in this relationship are even more variable (R 2 =0.30), suggesting again that expenses associated with higher system complexity can be much greater than would be expected. Discussion The study of urban dynamics relies in part on identifying trends and patterns that show scale invariance. It has been noted that many urban metrics, such as road

PAGE 107

107 density (Alig et al. 2004) income (Glaeser, 19 94) patent generation (Bettencourt, Lobo, Helbing, et al. 2007) follow power law relationships when compared across urban size However, little work has been done to describe how urban supply networks, particularly those associated with water provision, fit into the urban growth framework. The current study used a subset of data from two national survey s of water utility finan ces and operations Th e EPA survey was used to examined 1246 US community water systems to better define how water provision services vary within the urban sector. Of the three types of data analyzed (capital expenses, revenues, operations), nearly all w ere found to scale sublinearly or linearly with service population size, indicating that indeed many water system processes occur with varying degrees of constant or decreasing returns to scale. The exception to this pattern was seen only in the amount of revenues collected from sales of untreated water to other suppliers. These revenues were found to scale supralinearly, indicating that these types of water sales are one of the most profitable sources of revenue generation for water provision systems who engage in these types of transactions. Information on the amount urban utilities charge consumers for water services was obtained from an AWWA Water and Wastew ater Survey. When examined across a range of urban utility service population sizes, no relat ionship between urban size and price was found. The lack of a correlation between these two variables implies that the average amount spent by urban consumers is not a function of the size of the utility, despite the differences in amount of infrastructur e managed, and the costs incurred across urban utilities as seen in data from the EPA survey. While the price of water is determined in different ways by different utilities, this uniform trend suggests that urban

PAGE 108

108 utilities, must compensate for these unif orm rates in other ways. The reasons for this pattern in costs across differing utility sizes may benefit from further investigation into the impact of subsidies or other pricing controls/ alternate revenue generating mechanisms to keep water costs low, p articularly in areas where large scale water supply infrastructure has been created to move water supplies over great distances In addition to identifying scaling laws in urban water provision, this work also and costs of water operations by region. Two metrics, total operational complexity ( C T ) and relative operational complexity ( C R ) were applied to a subset of the survey data. In both cases, analyses that utilized t hese metrics revealed that the majority of urban water systems have higher than anticipated operational complexity (63% and 85%, respectively). Compared to total expenses incurred, greater per capita costs were associated with increasing levels of complex ity, implying that system growth can be costly for providers who must make major changes to their operating infrastructure. While a metric for combined operational complexity has great potential usefulness, creating a meaningful measure was difficult to d o. The aims of the 2000 EPA CWS survey were quite specific, which meant that there was data in depth, but not necessarily breadth. As such, this made a comprehensive assessment of total operational complexity quite difficult. This lack of information al so extended beyond operational data, making it difficult for survey respondents to be identified by any characteristic (i.e. city served,). In addition, respondents to this survey were not required to answer all the survey questions, leaving many partial ly complete surveys which was particularly limiting for this part of the study. Complicating these matters

PAGE 109

109 further, the source type data (i.e. Purchased, Self Obtained) from this survey lacked ancillary data that could identify those systems who are mainl y in the business of selling water to other providers, vs. those systems that buy water. any measure of integrated water supply system evaluation as a whole (i.e. from source to user). This last factor has the potential to seriously complicate this analy sis as there is no easy way to measure the cost incorporated into the total embedded complexity for the integrated system as a whole vs. just the utility. These factors are, and will remain remains serious limitations to this work. The results of this s caling analysis contribute to current understanding of urban growth and dynamics by identifying power law relationships in previously unstudied urban processes. As complexity theory is integrated into urban studies, methods for evaluating complex interact ions within systems become increasingly important. Therefore, novel combined metrics such as operational complexity provide potentially useful and necessary tools for investigating urban growth from a more holistic perspective. Such efforts will allow ur ban systems to be studied as discrete entities rather than simply a collection of individual processes. Ideally, complexity metrics will integrate multiple aspects of water provision, including not only more data on operational components, but environment al, economic and management information as well. The two metrics developed in this effort, although simple in nature, could be used successfully to analyze system components collectively, including contributions from other providers. For this reason, metr ics such as these have significant potential as benchmarking indicators for urban systems, and may be invaluable as water management become s a more holistic process.

PAGE 110

110 Figure 4 y = 3E+07x 1.019 R = 0.9799 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+00 1.E+01 1.E+02 1.E+03 Service Population Rank

PAGE 111

111 Figure 4 2. Power law scaling relationships between urban water service population size and capital expenses.

PAGE 112

112 Figure 4 3. Power law scaling relationships between urban water service population size and system revenues. Figure 4 4. Power law scaling relations hips between urban water service population size and finances. A ) total revenues and B) total expenses.

PAGE 113

113 Figure 4 5. Power law scaling relationships between urban water service population size and physical water provision Figure 4 6. Power law scali ng relationships between urban water service population size and the price of water. y = 0.0144x + 3.025 R = 0.0019 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 10 11 12 13 14 15 16 17 18 LN Water Price (42475 liters) LN Urban Population DB IB Uniform

PAGE 114

114 Figure 4 7 Total operational complexity and per capita expense. A) distribution of total operational complexity ( C T ) and B) its relation to total per capita expenses

PAGE 115

115 Figure 4 8 Total relative complexity and per capita expense A) distribution of relative operational complexity ( C R ) and B) its relation to total per capita expenses

PAGE 116

116 CHAPTER 5 COLLABORATIVE PARTIC IPATION AND WATER MA NAGEMENT IN US URBAN WATER UTILI TIES Introduction The growing uncertainty surrounding future urban water availability in the United States (US) has become a primary issue of concern over the last decade for many water providers (Runge and Mann, 2008) This concern has surfaced as managers have increasingly realized that water is a common pool resource with real limitations that are not only geographic, but economic and environmental as well. As such, numerous challenges face water systems today with regards to climate change, population growth and environmental regulations (Ginley and Ralston, 2010; Means, West, et al. 2005) In response to these challenges, a variety of new strategies have been developed to deal with the need to incorporate these socio economic natural linkages into water resources management (Conca, 2006) Demand s ide management has become increasingly popular as utilities attempt to control, rather than accommodate, the needs of water users. Demand management strategies such as water conservation based pricing structures, appliance retro fits, and education all te nd to be relatively low cost and easy to implement (Baumann et al. 1998) Other strategies currently being used include water rec ycling/reuse programs, regional water planning, and source enhancement from alternate water supplies such as desalination, aquifer storage and recovery and conjunctive water management (Aichinger, 2009; Gleick, 2000; Miller, 2006; Blomquist et al. 2004) While some of these demand management approaches are relatively simple to implement because they only require local changes strategies such as conjunctive

PAGE 117

117 management and shared source planning require water managers to be aware of the system needs beyond the local scale (i.e. other user demands, ecosystem needs, etc.). An effective system for information exchange among users therefore is required to ensure that management decisions are made using the best knowledge available (Hackett et al. 1994) Inf ormation exchange (i.e., the transfer of knowledge, ideas and values across different levels of management) is critical for managing shared resources, particularly when managing for a sustainable future (Cowie and Borrett, 2005) The nature and extent of communication and cooperation depend upon the complexity of the shared resource. Ideally, water managers can negotiate a variety o f legal, political, environmental and operational constraints, and work collaboratively with multiple stakeholders to ensure that management decisions set appropriate precedents and support broader goals related to long term water policy (Cortner and Moote, 1994; Reitsma, 1996) Many water utilitie s frequently interact and exchange information with governmental and state agencies, scientific and political organizations and various stakeholder groups. The impact of information exchange may differ depending on the context of the two participants; in some cases, the knowledge shared may only be peripheral to the decision making process, while at the other extreme it may be directly related to, and have significant influence over, decision outcomes (Cowie and Borrett, 2005) Therefore, bearing in mind the collaborative nature of the decision making process, it is not surprising that communication has been identified as an importan t mechanism for facilitating the cooperation of these common pool resources users (Gerlak and Heikkila, 2007; Hackett et al. 1994)

PAGE 118

118 From an organizational perspective, information not only acts as a catalyst for forming alliances, which can result in significant competitive advantages (Dyer and Singh, 1998; van Wijk et al. 2008) but plays a significant role in economic growth and fostering innovation (Agrawal et al. 2008; van Wijk et al. 2008) Access to information (Anderson, 2008; Cross and Sproull, 2004) Managers who have greater access to information and subsequently use the knowledge gathered are more likely to notice trends and problems, understand their environment, and perform at a higher level than those who have limited access to information (Anderson, 2008) Communication and collaboration can lead to increased legitimacy in environmental decision making processes and provide a mechanism through which multiple actors can identify common goals for successful management. However, it has been noted that by allowing collaborative decision making, utilities must access and maintain more connections and synthesize more information by collaborating with other actors. This may require more time and energy on the part of the utility, and can make communication, cooperation and management more difficult (Agrawal, 2001; Cowie and Borrett, 2005) With the exception of individual case reports (Cowie and Borrett, 2005; Gerlak and Heikkila, 2007) few studies have attempt ed to specifically quantify the impact of knowledge exchange, achieved through communication and collaboration between water management entities, on water management approaches or policies. While there is much evidence for why such studies have not been performed previously (i.e., difficulties in quantifying knowledge exchange, lack of pre existing or historic data, etc.),

PAGE 119

119 this information is crucial for understanding how water managers are collecting and disseminating information and how these processes may be impacting water resource management. The present study was designed to investigate the relationship between water utility management and collaborative participation, defined as the level and importance of participation in external groups or activ ities that facilitate exchange of knowledge between relevant entities. To better understand how these two components of utility operations relate, a survey was created to measure how frequently mechanisms for knowledge transfer between water utilities and other entities are being used, how such knowledge transfer impacts relevant aspects of management, and ultimately the importance placed upon these issues by utilities. Finally, the level of collaboration reported by each utility/respondent was compared to the local water availability (as derived in Chapter 2) to examine the possibility of water scarcity as a driver of utility collaboration. The survey results were used to test the hypothesis that utilities that are more engaged in collaborative participat ion will place more importance on their water management plans. Methods Survey Design To measure the role of collaborative processes in US urban water management, an internet based survey was developed using Qualtrics online survey software (Qualtrics, 2011) The survey consisted of fixed answer and open ended questions about water management strategies and participation in collabo rative activities. The survey was administered to 294 water utilities that served urban areas with populations exceeding 100,000 in the coterminous US. As the survey requested broad information

PAGE 120

120 about how the utility entity operates, surveys were sent, wh en possible, to the general manager or executive officer of each utility. Using standard surveying techniques, participants received two follow up reminders via either telephone and /or email to participate in the survey after the second and fourth weeks h ad passed (Dillman, 2007) Endorsement from the National Water Research Institute added legitimacy to the survey, and was expected to increase the response rate. Of the 294 surveys sent, 47 were returned, and 37 included complete, useable responses, representing a 13% response rate. This response rate is more than one standard deviation below the mean range of response rates (351 7%) typically obtained when polling organizations or executive managers (Baruch and Holtom, 2008; Cycyota and Harrison, 2006; Tomaskovic Devey et al. 1994) but is comparable to the 15% response rate to the 2008 water indus try survey published by the American Water Works Association (Runge and Mann, 2008) Survey Content The survey was composed of th ree sections that addressed the general utility characteristics, management strategies, and degree of collaborative participation of each entity (Appendix B). The general characteristics section included basic questions about utility size and water produc tion. The management strategies section explored utility operations more deeply, asking questions about the content and length of the planning process, as well as previous and current strategies used for managing water supply. The strategies listed were based strategies were categorized as those that were typically implemented proactively to manage and control existing demands (Baumann et al. 1998) and included the following: water conservation and outreach, alternate pricing

PAGE 121

121 schemes (including the use of block rate designs, low cost reclaimed supplies, etc.), water conservation rebates (retrofit, efficiency, irrigation, etc), water use restrictions (both voluntary and mandatory), and water rationing. Supply based strategies encompassed those actions taken by a utility that do not proactively control de mands, and were grouped as: creation, expansion or improvement of surface water supplies; creation, expansion or improvement of groundwater supplies (including aquifer recharge); creation or expansion of recycled/reclaimed water supplies; contracts to purc hase water; and regular infrastructure maintenance. The third section of the survey was designed to measure the type and level of participation by utilities in different collaborative activities. Participants were asked to estimate the degree to which the ir utility engaged in the following five types of activities: 1. Attendance at conferences Participation in conferences, meetings and seminars increases the amount of contact employees from a utility may have with other individuals and companies in the wate r industry, and provides an opportunity for those attending to learn and exchange ideas. 2. Participation in surveys and studies Contributing to and sharing results of studies regarding water management and operations signifies an increased interest and wi llingness to exchange ideas and information with other interested parties. 3. Association with trade organizations Trade organizations provide relevant operational and managerial information, as well as news, to utilities who subscribe. Depending on the org anization, utilities have the opportunity to follow and present opinions, news and related research by other groups through conferences, journals and other organization sponsored events. 4. Interactions with other institutions A larger number of interactions with other water related institutions provides a greater forum from which utilities can collect and exchange potentially useful information for improving water management. 5. Involvement in water supply related groups Engaging and cooperating with other ins titutions to manage water supplies places utilities in direct contact with external sources of information and requires a shared outlook on how the implementation of various management strategies affects all the users involved.

PAGE 122

122 Results Respondent Character istics After receiving survey responses, utilities were grouped by similar climate types (Karl and Koss, 1984) so that differences in collaborative participation could be examined across different hydroclimatic conditions ( Figure 5 1). According to these designations, participants in this survey were located in the Southeast (SE), West (W), Northeast (NE), Central (C), South (S), Ea st North Central (ENC), and Southwest (SW) regions of the US. Response rates from each region are listed in Table 5 1. The highest percentage of respondents came from the SW, ENC and SE regions of the US, respectively. The NE, and S had response rates le ss than 10%, and no responses were received from the NW and WNC regions. As part of the survey, each respondent was asked to provide general information about the utility at which they were employed (Table 5 2). Based on the responses received, the aver age service population was 343,059 individuals within an average service area of 612 km 2 The average volume of water sold was 184 million liters per day (MLD), average water consumption was 658 liters per capita daily (lpcd), and utilities were staffed o n average by 1.04 employees per thousand people served. When characterized by region, the NE had the highest average service population size, totaling over 1,000,000 customers, while the W region had the lowest, serving on average 123,260 customers. Th e largest and smallest average service areas were found in the C (933 km 2 ) and SE (321 km 2 ) regions. The highest and lowest average water production belonged to the SW (427 MLD) and SE (84 MLD) regions, whereas the largest and smallest average demands wer e experienced in the W (788 lpcd) and

PAGE 123

123 NE (494 lpcd), respectively. The region with the least number of employees per daily water produced was the W (4.53), and the region with the highest was the C (15.04). Utility Management Management p lanning Of the utilities polled, 87% of the respondents indicated that their system utilized some form of a water management plan. Of those utilities with plans, the majority (56%) reported that water management plans were required by local or state government or anoth er controlling water institution. By region, mandatory water management plans were most commonly found in the W, S and SE (Table 5 3). The remaining 44% of respondents indicated that their utilities created plans voluntarily. Voluntary plans were most f requently cited by respondents from the SW region. Survey responses from the NE and ENC regions indicated an equal number of mandatory and voluntary plans in existence. None of the responses from the C region reported information about water management p lans. The average time frame across which planning horizons occur was found to be 21 (14) years, with the S region having the shortest average planning time frame (12 yrs) and the SW having the longest (36 yrs). In terms of plan content, the respondents were asked to use a scale of 1 (minimally addressed) to 3 (thoroughly addressed) to indicate how comprehensively availability, and customer demands, as well as infrastru cture, security and financial needs. Using this simple scale, the average level of importance at which components of urban utility plans are addressed was recorded ( Figure 5 2). Across all responses, the most thoroughly addressed components were those r elated to source water availability (2.8 0.4) and customer demand (2.80.5), followed closely by source water

PAGE 124

124 quality (2.50.7) and infrastructure needs (2.20.8). The two components addressed least frequently by utilities were those concerning financial (1.81.1) and security (1.61.0) needs. A small number of respondents cited other components to their plans that were not listed in the survey, such as alternative futures planning, customer outreach and education, and environmental/watershed sustainabil ity. In each of these water management plan. Regional responses regarding water management planning are found in Table 5 4. Excluding the C region, the average level of importance given to management planning was lowest in the NE (1.40.6) and highest in the SW (2.70.4). Management s trategies Survey respondents were asked to provide information about the current primary and secondary strategies used for managing water supply, as well as those strategies that are under development. Each then was asked to compare current strategies to those implemented five years ago. A list of ten commonly used general management practices was given in the survey, along with the optio n to input other strategies that were not provided. The types of management practices used by respondents are shown in Figure 5 3. The majority of respondents indicated that across both reporting periods the three most commonly used strategies for manag ing supplies were regular infrastructure maintenance, water conservation education and outreach, and water restrictions. The least commonly employed strategy was water rationing. A comparison between the two reporting periods showed that in nearly every case, with the exception of contracts for water purchases, there was an increase in the number of utilities using each type of strategy. A small percentage of the respondents (8%) reported strategies

PAGE 125

125 not provided on the survey list; these strategies includ ed drought management planning, source substitution, water banking, and desalination. In addition to reporting the types of strategies used, respondents also were asked to identify those strategies that were most important to their operations by categoriz ing each as a primary, secondary, or unused strategy during each reporting period. Strategies were divided into two groups based on whether they were inherently designed to manage either water supply or demand (Table 5 5), and the percent of all utilities utilizing each was examined across time ( Figure 5 4). For both time periods examined, a larger percentage of utilities used supply based, rather than demand based, primary strategies. The opposite was true for strategies of secondary importance; more u tilities implemented demand based strategies as supplemental and/or alternative options to their primary management strategies. Supply based strategies in development during each time period also were more heavily favored by utilities than demand based st rategies. When compared across regions (Table 5 6), all groups reported that their primary strategies for water management were supply based rather than demand based. However, the C, SE, S and W regions indicated that they relied more heavily on demand ba sed strategies when considering secondary management options. The relative ratio of utilities using each strategy type as a primary or secondary management option changed over time. As can be seen in ( Figure 5 5), the focus of primary strategies shifted over the past five years; the number of utilities using supply based solutions decreased by 23%, while those using demand based solutions have increased by 38%. The percent of utilities using supply based and demand based

PAGE 126

126 strategies has increased by 9% an d 17%, respectively. The frequency of supply and demand based strategies in development by a utility has been the opposite of that found in the primary strategy case. Results show a substantial decrease (29%) in the number of demand based strategies be ing developed by utilities over the past five years, while the number of supply based strategies has increased by 13%. To better understand the overall change in water management strategies over the reported time period, the two types of data collected in this question were combined as the product of the number of strategies used and the average importance of those strategies for the current and historic reports. Differences in the median current overall practices (18.3) were compared to those used five ye ars ago (14.5) using a Wilcoxon Signed Rank Test. Results from this test confirmed that a greater emphasis is placed on current overall management strategies than was five years ago (Z=5.120, p < 0.001). Collaborative Participation This survey measured th e degree to which urban utilities participated in the following five collaborative activities: 1) knowledge exchange at conferences, 2) knowledge exchange through research, 3) interactions with other institutions, 4) interactions with trade organizations, and 5) interactions in groups specifically focused on source water management. Knowledge exchange at conferences Of those who responded to the survey, 86% reported that employees from the utility they represent had attended at least one conference with in the past two years. Utilities reportedly attended 6 (5) conferences on average over this time period, sending 12.1 (13.3) employees. In addition, respondents also were asked to provide information on the number of conferences at which their firms pr esented material. The

PAGE 127

127 average participation rate in this case was 2.7 (2.6) conferences with 5.3 (10.7) employees. By region, the W claimed on average the highest number of conferences attended and at which material was presented. However, the SW sent the most employees to attend and present at conferences (Table 5 7). Knowledge exchange through research Collaborative participation requires that entities be open to sharing information and exchanging ideas regarding their operations and management practi ces. When asked to report how frequently each utility participated in individual, local, state, regional, or national water utility performance studies, reports or surveys (excluding this one), 84% reported having engaged in such activities within the pas t two years, 11% said they had not participated, and 5% did not know if their utility had participated in these types of activities. Respondents also were asked to describe, in general, the availability of results from these activities to the general publ ic. Of those who took part in surveys, studies or reports, 54% said their results were freely available to the public, 27% indicated they were usually semi private results accessible to interested parties who contact the study participants for information and 3% said the results of their participation in such activities were closed to the general public. When explored according to region, utilities in the ENC and C participated most frequently in publically available studies, whereas the highest percent of utilities participating in semi private and private studies were found in the S and NE, respectively (Table 5 8). Interactions with other institutions Interactions through regulatory compliance or supply management may also provide an avenue for urban u tilities to collaborate. Therefore, participants were asked to report the number of water related institutions with which they interact, and the

PAGE 128

128 the past five years there appeared to be no change in in the average number of institutions with which each utility interacted. When examining the types of interactions, however, the most frequently interacted with institutions were state and local governments, as well as other water utilities; 100% of utilities reported these groups as being involved in their planning and management. Five years ago, 97% of utilities reported interacting with state and local governments when dealing with issues related to utility planning and management; however, water utilities also reported interacting more frequently with private consultants (97%) than with other water utilities (89%) at for eith er time period occurred with state based and water industry based associations, with only 46% and 65% of utilities reporting interactions in these categories. The perceived importance of each institution to utilities was related to the involvement of util ities with those institutions (Table 5 9). On a scale of 1 (minimally important) to 3 (very important), the following three institutions were ranked highest for current and historic reporting periods: state government (2.80.4/2.80.4), local government (2 .50.8/2.40.7) and system customers (2.41.0/2.41.0), respectively. The largest average net change in institutional importance to urban utilities over this time period occurred between utilities and other water utilities (+13%), state based associations (+6%) and local government (+5%). Table 5 10 summarizes the current and historic institutional interactions and their relative importance by region, with the SW reporting the highest mean level of interaction and importance across both periods.

PAGE 129

129 The overa ll change in institutional interactions for all water utilities over the reported time periods was calculated as the product of the number of interactions and the average importance of those interactions for the current and historic reports. Differences in the median current overall interactions (15.5) were compared to those experienced five years previously (16.4) using a Wilcoxon Signed Rank Test. Results from this test indicated that more emphasis was placed on overall institutional interactions histo rically (Z=2.159, p = 0.003). Participation within trade organizations Participants in the survey were asked to report their current level of involvement with related trade organizations. Six of the most common trade organizations were listed in the sur vey: the American Water Works Association (AWWA), the Association of Metropolitan Water Agencies (AMWA), the American Public Works Association (APWA), the National Association of Water Companies (NAWC), the Water Environment Federation (WEF), and the Water Utility Benchmarking Association (WUBA). Participants also were given the option of listing other trade organizations. For each organization, respondents were asked to report how active their utility was on a scale of 1 (minimally engaged) to 3 (active ly engaged). Of the utilities responding to the survey, 100% reported involvement with at least one organization. On average, 46% of respondents reported that their utilities were actively engaged with trade organizations, 51% reported moderate engagemen t, and 24% reported minimal engagement. The sum of these percentages does not add to 100% due to the fact that many utilities are involved to different degrees with more than one trade organization. Among the trade organizations listed, utilities were mo st frequently involved in AWWA (100%), WEF (70%) and APWA (54%). Most utilities were actively involved in AWWA,

PAGE 130

130 and moderately involved in WEF and APWA. When analyzed by region, the SW participated most, and C participated least. With respect to level of engagement, ENC hd the highest number of utilities actively involved, the SW had the highest number of utilities moderately involved, and the S had the highest number of utilities minimally involved with these organizations (Table 5 11). Participation in group based source water management One of the most straightforward measurements of supply collaboration within the water industry is based on how utilities interact with other groups to directly manage water sources. Respondents were asked to indicate w hether their utility was a member of any local or regional board, agency, compact or other similar type of group that manages, makes decisions about, or controls any or all of their current water supply sources. Of those who responded, 73% reported partic ipating in at least one such group; on average, a utility participated with 2.8 (1.6) groups. Within a collaborative group, the average number of other institutions involved was 5.2 (0.7). The most frequently listed institutions in these water manageme nt groups were other water utilities, who were mentioned 84% of the time, the local government (76%), private consulting firms (59%), and state government (54%) ( Figure 5 6). By region, the SW participated in the highest number of groups on average, and w ith the highest number of institutions per group. The utilities in the NE participated in the lowest number of groups, with the lowest number of institutions per group (Table 5 12). Collaboration and Management To test the hypothesis that utilities that a re more engaged in collaborative participation will place more importance on their water management plans, survey responses were combined into two comprehensive metrics: overall management and

PAGE 131

131 overall collaboration. For those questions involving measureme nts of historic and current practices, only current responses were included. For questions with multiple attributes, responses were consolidated into a single measurement. The following information was used to calculate each overall metric: Overall mana gement m etric Average importance of management plan components Current management strategies: (Strategies used x Average importance) Overall collaboration m etric Attendance at conferences: Participation Points for participating: Yes =1, No= 0 Participat ion in Studies or Surveys: (Participation x Sharing) Points for participating: Yes =1, No= 0 Points for sharing: Public = 2, Semi private =1, Private =0 Average level of engagement in trade organizations: Score based Points for engagement: High=3, Modera te=2, Low=1 Current institutional interactions: (Interactions made x Average importance) Involvement in water supply related groups: (Participation x (Number of groups x Institutions per group)) Points for participating: Yes =1, No =0 For each metric described above, measurements were normalized using the following equation: (5 1) where x = individual value, x min = minimum value in the dataset x max = maximum value in the da taset, and then summed across response categories.

PAGE 132

132 The relationship between the two overall metrics was assessed with linear regression ( Figure 5 7 A ). A positive trend between the overall management and collaboration metrics was apparent in the data, sugg esting that utilities who are more engaged in collaborative activities place a greater importance on the quality and composition of their management plans. T he explanatory power of the relationship between these two variables as currently measured was low (R 2 =0.19) and did not achieve statistical significance however high variability in this assessment is inevitable, as the samples being studied here operate within a diverse and dynamic system. As such, this ultimately makes tight correlations between ge neral metrics for samples such as these unlikely, yet these types of assessments are still useful in that regression slopes are able to provide information about how two variables relate at a more general, national scale. Results of overall collaboration w ere also compared to regional local water availability (Figure 5 7B) to test for a potential effects on management due to water scarcity. A linear regression was again employed showing a weak correlation (R 2 =0.17) and slight negative trend between overal l collaboration and average regional local water availability Currently, t he lack of a clear relationship between these two variables suggests that local water availability plays only a small role in affecting a aborative activities This may be due in part to the small sample size used here, or the fact that survey only focused on recent management decisions and did not ask respondents to comment on past decisions or collaborative actions that may have occurred more than five years ago. Collaboration may have been more important in the past when projects to find and acquire additional

PAGE 133

133 water sources was a higher priority for water utilities. The importance of historic collaborations over the past several decades may be hidden by management reorganization over time, for example, where joint efforts to secure water may have been transformed into a new entity, rather than existing as a compact between two utilities. Conclusions and Management Implications Collabor ation is considered a key component in the sustainable management of common pool resources, yet little quantitative evidence exists for determining if collaboration occurs on a national scale and what impact such collaborations have on management. The cur rent study is one of the first national assessments of US urban water utility collaborative efforts. This descriptive survey collected basic information regarding the range of possible mechanisms used by utilities to disseminate and collect information. Such data have been lacking at the national level, and provide insight into knowledge transfer between utilities and other water institutions. These data also represent how these exchanges may impact water management. Planning and Management Water manage ment planning is crucial for comprehensive, sustainable development, however there have been few assessments made about how planning is occurring at the national level. As of the time this study was performed, the last known assessment of water managemen t planning at national level was a state based survey performed in 2005 (Viessman and Feather, 2006) Results from this survey fo und that the majority of utilities had plans, and for all regions except the SW, most of these plans were mandatory. The reasons for requiring utilities to have management plans were not investigated here, however, the high percentage of mandatory plans i n both the eastern

PAGE 134

134 and western portions of the US, suggest that, if it was not before, mandatory planning may be a truly national trend in the 21 st century. The importance of plan components to each utility varied widely, and as was the case with many of the individual metrics evaluated, the SW region consistently scored the highest in terms of the importance of plan components. While the plan importance was subjectively based on the opinion of the respondent, higher scores for this relatively drier regi on could be the result of a long history of managing for water in a water scarce area from early on in the utilities histories. This rationale does not apply to the regions reporting the second (SE) and third highest (ENC) importance of planning componen ts, and further investigations into the driving factors for management planning and implementation would be insightful and useful. In terms of the importance and types of strategies used for managing water supplies over the past five years, utilities appe ar to be placing a greater emphasis on the use of a varied array of supply and demand management options. Supply based strategies were still clearly heavily favored by utilities, despite the increased prominence of demand based options in recent literature However, this study did find a reversing trend with regard to the overall emphasis on utilities place on supply based strategies suggesting that demand based strategies may become a higher priority in the future. Based on the results from the managemen t portion of the survey, further investigations into regulatory and institutional policies might provide crucial information as to why some regions place more importance on planning than others, and such information also would help define what kinds of col laborative activities can and do take place under given sets of policies and rules.

PAGE 135

135 Collaborative Participation At the national level, the vast majority of utilities allow representatives to attend and present at conferences, and 100% of the respondents re ported engagement in at least one professional trade organization. Most utilities also participate in outside research activities in the form of studies, reports, or surveys, with the majority indicating that the results of these efforts were freely avail able to the public. The number of utilities reporting activity in this area was somewhat unexpected, as low response rates are supposed to be typical for industries such as these (Tomaskovic Devey et al. 1994) and may reflect an unexpected response bias (i.e., those utilities responding to this survey are theoretically more likely to respond to other research initiati ves as well). With regards to interactions with other water based institutions, state and local net change over the past five years was an increase in the level of in volvement by other water utilities in planning and management operations, which lend credence to the idea that water utilities are becoming more open to collaboration and integrated management, particularly when it comes to managing urban water demands (Ginley and Ralston, 2010) When asked to provide information about group participation in water supply management, only 59% of utili ties reported information on involvement in such a group, and local and state governments again were key players. Reasons for the low response rate to this question were clarified by comments from respondents. The nature of the question was thought to be insufficient to capture the variety of group interactions in which utilities engage when managing water supplies, particularly resources, infrastructure and costs between use rs. These helpful comments from

PAGE 136

136 survey participants suggest that a more in depth study of institutional/group arrangements may be needed to more accurately describe collaborative efforts when managing water supplies. When individual metrics were combin ed to create overall indicators of management and collaboration, the results of this study are consistent with the hypothesis that the quality and composition of the management plan is related to the level of collaborative participation with a positive lin ear regression trend. Correlations between overall collaboration and regional local water availability were less apparent, however. In each of the cases mentioned above, the statistical significance of the assessment was low with R 2 =0.19 and R 2 =0.17, re spectively possibly because of the relatively low number of responses to the survey. Further work with an in depth survey of a smaller number of carefully selected utilities could possibly help develop more substantive conclusions about this relationship Finally, the purpose of this study was to establish a quantitative understanding of the processes and mechanisms driving collaboration in US urban water utilities. While this effort has provided new insight into the issue, further investigation clearly would be beneficial. Collaborative participation in general, is a poorly defined and difficult behavior to monitor, and the development of a single indicator, or set of indicators, to measure this behavior and its impact is inherently problematic. Ass essment of the cost of collaboration, in terms of the human and financial investment that must be made, would demonstrate which collaborative activities are more helpful and the degree to which utilities must be engaged to benefit maximally from participat ion.

PAGE 137

137 The results of this study indicate that the majority of responders evidenced collaboration. While this type of metric has never previously been quantified, these results support ideas presented in Chapter 3 about how groups form and function to pr ovide local and even regional solutions to urban water management problems. The successes and limitations of this work all clearly indicate a need for a more cohesive understanding of how information exchange and collaborative participation affect water u tility management. Utilities often are encouraged to participate in financial and efficiency benchmarking activities. These activities not only help utilities identify weaknesses in relation to their peers, but also provide regional and national database s of information on urban water management. If used to collect information regarding collaborative processes, these data could be used to better identify those strategies that are important for successful management and those that will provide only minima l impact.

PAGE 138

138 Table 5 1. Survey respondents by US r egion. Region Surveyed RR (%) SE 59 19 W 49 14 NE 47 6 C 38 13 S 38 8 ENC 25 20 SW 15 33 NW 14 0 WNC 5 0 Table 5 2. Utility characteristics of US r egions Region Service Population (thousands) Se rvice Area (km 2 ) Avg Water Sold (MLD) Demand (lpcd) Employees/ 1000 Served NE 1,038 ( 1,300) N/A 344 ( 103) 494 ( 234) 0.59 ( 0.25) ENC 256 ( 209) 792 ( 221) 117 ( 17) 637 ( 523) 0.74 ( 0.62) C 481 ( 584) 933 ( 347) 240 ( 68) 737 ( 566) 3. 42 ( 4.66) SE 163 ( 92) 321 ( 95) 84 ( 17) 530 ( 225) 1.04 ( 0.68) S 207 ( 211) 629 ( 293) 153 ( 52) 593 ( 217) 0.92 ( 0.68) SW 663 ( 540) 853 ( 201) 427 ( 97) 662 ( 104) 7.36 (10.43) W 123 ( 64) 524 ( 391) 119 ( 27) 788 ( 573) 4.5 3 ( 2.83) Average: 343 ( 490) 612 ( 258) 184 ( 59) 658 ( 379) 1.04 ( 4.41) Data reported as mean ( SD) Table 5 3. Management plan requirements by region Region Mandatory (%) Voluntary (%) Avg. Length (Yrs) NE 50 50 25 ENC 50 50 15 C --SE 73 27 18 S 67 33 12 SW 20 80 36 W 57 43 24

PAGE 139

139 Table 5 4. Average importance of water resource planning components by region. Region Finances Source Quality Source Availability Customer Demand Infrastructure Security Avg(SD) NE 0.7 1.7 2.0 2.0 1.0 1.0 1.4( 0.6) ENC 2.0 2.3 3.0 2.5 2.0 2.0 2.3( 0.4) C 0.0 0.0 0.0 0.0 0.0 0.0 0.0( 0.0) SE 2.0 2.3 2.7 2.8 2.4 1.8 2.3( 0.4) S 1.3 2.3 2.7 2.7 2.0 1.3 2.1( 0.6) SW 2.6 3.0 3.0 3.0 2.8 2.0 2.7( 0.4) W 1.0 2.6 2.7 2.6 1.9 0.9 1.9( 0.8) Average Im portance based on a scale of 1 (minimally addressed) to 3 (thoroughly addressed). Table 5 5. Supply and demand based general management strategies options Supply based Strategies Demand based Strategies Regular infrastructure maintenance Water conser vation education and outreach Contract to purchase additional water Alternate pricing schemes Investments in surface water supplies Water use restrictions Investments in groundwater supplies Water rationing Recycled/reclaimed water Water conservation rebates Table 5 6. Fraction of utilities using supply vs. demand based strategies by region. Supply based Demand based Region Primary Secondary Developing Primary Secondary Developing NE 0.33 0.67 0.33 0.33 0.33 0.33 ENC 0.50 0.50 0.25 0.25 0.25 0 .50 C 1.00 0.20 0.00 0.00 0.40 0.00 SE 0.45 0.18 0.18 0.27 0.27 0.18 S 1.00 0.00 0.33 0.00 0.67 0.33 SW 0.60 0.00 0.20 0.20 0.40 0.00 W 0.43 0.57 0.29 0.29 0.14 0.14

PAGE 140

1 40 Table 5 7. Conference attendance by region. Avg. Conferences Avg. Employees Region Attended Presented Attended Presented NE 2.3 1.0 1.7 2.0 ENC 4.3 2.5 13.7 4.0 C 4.8 0.5 13.5 0.5 SE 5.9 2.5 9.3 3.2 S 1.7 0.7 NA NA SW 5.0 3.3 23.3 19.0 W 8.0 3.3 7.4 2.5 Table 5 8. Fraction of utilities participating i n studies and surveys by region. Public Semi Private Private None Do Not Know Region NE 0.67 0.00 0.33 0.00 0.00 ENC 0.75 0.25 0.00 0.00 0.00 C 0.75 0.00 0.00 0.25 0.00 SE 0.55 0.36 0.00 0.09 0.00 S 0.00 0.67 0.00 0.33 0.00 SW 0.60 0.40 0.00 0.00 0.00 W 0.43 0.14 0.00 0.14 0.29 Table 5 and planning Institutional Groups Current Historic Change (%) State Government 2.8 2.8 0.00 Local Government 2.6 2.5 0.06 System Customers 2.4 2.4 0.03 Other Water Utilities 2.2 1.9 0.13 Federal Government 2.1 2.0 0.03 Private Consultants 1.9 1.9 0.03 State based Associations 1.4 1.4 0.06 Water Industry Associations 1.4 1.4 0.02 Professional Trade Organizations 1.4 1.3 0.05 Academia 1.4 1.3 0.05 Other 0.4 0.4 0.04 Averages measures on a scale of 1 3 (1 minimal, 2 moderate, 3 considerable)

PAGE 141

141 Table 5 10. Number and importance of interactions by region Current Historic Institutions Importance Institutions Importance Region Mean SD Mean SD Mean SD Mean SD NE 9.0 1.5 1.85 0.4 8.7 1.5 1.76 0.3 ENC 9.3 2.4 1.68 0.6 9.3 2.4 1.68 0.4 C 8.8 1.3 1.82 0.4 8.8 1.3 1.82 0.4 SE 8.3 1.6 1.46 0.4 7.8 1.6 1.39 0.3 S 8 .7 1.2 1.76 0.2 8.3 1.2 1.70 0.2 SW 9.8 0.4 2.22 0.3 9.8 0.4 2.22 0.3 W 9.4 2.5 1.74 0.6 7.9 2.5 1.47 0.4 Table 5 11. Involvement in trade organizations by region. Avg. Groups Involvement Activity (%) Region Involved Actively Moderately Minimally NE 3.0 30 30 40 ENC 3.0 75 13 13 C 2.0 42 50 0 8 SE 3.3 29 43 28 S 3.7 25 31 44 SW 5.2 13 55 17 W 2.3 33 38 29 Table 5 12. Average group involvement c haracteristics by region Number of Groups Institutions Per Group Region Mean SD Mean SD NE 1.0 N/A 3.0 N/A ENC 3.3 1.7 6.3 2.1 C 2.0 0.0 5.0 0.0 SE 3.0 1.8 6.3 3.5 S 2.0 1.0 7.0 2.6 SW 3.5 1.3 8.3 2.9 W 3.0 2.3 5.8 2.6

PAGE 142

142 Figure 5 1. Boundaries of US r egions (black line) and locations of urban areas surveyed (grey). Figure 5 2. Average level to which utilities address various components of their water management plans. Values are the average out of a maximum score of 3. 2.4 2.4 2.1 1.9 1.5 1.4 0 0.5 1 1.5 2 2.5 3 Water Availability Customer Demand Water Quality Infrastructure Finances Security Components Addressed (1 minimal, 2 moderate, 3 thoroughly))

PAGE 143

143 Figure 5 3. Management strategies used by utilities today (current) and five years ago (historic) A) B) Figure 5 4. Relative reliance on supply and demand based management strategies A) Currently used strategies and B) strategies used five years ago. 0 10 20 30 40 50 60 70 80 90 100 Other Water rationing Contract to purchase additional water Water conservation rebates Investments in groundwater supplies Recycled/reclaimed water Investments in surface water supplies Alternate pricing schemes Water use restrictions Water conservation education and outreach Regular infrastructure maintenance Respondents (%) Current Historic 0.00 0.10 0.20 0.30 0.40 0.50 0.60 Primary Secondary Developing Utilities (%) Management Strategy Type Supply-based Demand-based 0.00 0.10 0.20 0.30 0.40 0.50 0.60 Primary Secondary Developing Utilities (%) Management Strategy Type Supply-based Demand-based

PAGE 144

144 Figure 5 5. Change in utility preference of supply based vs. demand based strategies o ver a five year period. Figure 5 6. Percent of water supply management groups that listed water related institutions are participating in. -0.40 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 0.40 0.50 Utilities (%) Management Strategy Type Supply-based Demand-based 84 76 59 54 44 40 38 36 33 21 7 0 10 20 30 40 50 60 70 80 90 100 Other Water Utilities Local Government Private Consulting Firms State Government Prof. Trade Organizations Academia State-based Associations Non-Governmental Organizations Federal Government General Public Other Involvement (%)

PAGE 145

145 A B Figure 5 7 Relationship between overall water utility management collaboration and water availabilit y. ( A) Water utility management and local water availability and (B) w ater utility management and overall collaboration y = 0.7169x + 2.0888 R = 0.1857 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0.0 0.5 1.0 1.5 2.0 Overall Collaboration Metric [ ] Overall Management Metric [ ] y = 5E 06x + 2.9912 R = 0.1701 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 1000 10000 100000 Overall Collaboration Metric [ ] Local Water Availability [lpcd] SW W SE ENC NE S C

PAGE 146

146 CHAPTER 6 MANAGING WATER: THE ROLE OF COMPLEXITY IN AN EVOLVING PARADIGM THE US URBAN WATER EXPERIENCE Introduction Sustainability has become the dominant discourse in natural resources management over the past 30 years, and has fundamentally changed the way society views its relationship to the resources it uses (Graedel and Klee, 2002) While the precise definition of sustainable resource management remains elusive, a loosely agreed upon understanding of the term recognizes sustainability as the equitable developme nt of environmental, economic and social demands that meet current needs without compromising the needs of future generations (Gleick, 1998; Loucks et al. 1998) In the realm of water management, these concepts have been combined into a coordinated development and management of water, land and related resources, in order to maximize the resultant economic and social welfare in an equitable ma nner Figure 6 1) (Mitchell, 2005) This new management paradigm has come forth as society has realized that traditional management strategies, which have relied almost exclusively on physical, infrastructure based solutions to water supply needs, are not sustainable (Postel et al. 1996) While the traditional water management paradigm has improved the lives of millions through the transfer, storage and control of water resources, these improvement s have caused a multitude of unintended and costly social and environmental problems (Gleick, 2000) IWRM has been proclaimed by many to be the

PAGE 147

147 solution to the problems of traditional water management and the inevitable path of the future (Falkenmark, 2004; Gleick, 2003; Serageldin, 1995; Slocombe, 1993) however the lack of a specific protocol for implementing this new paradigm makes IWRM difficult to use in practice (Bellamy et al. 1999; Biswas, 2003; Mitchell, 2005) Despite the general definition o f IWRM and the difficulties associated with its implementation, the lure of this new management paradigm remains attractive. In trying to develop a better understanding of how to successfully implement IWRM, concepts of water resource management are being more clearly understood using ideas and principles from the relatively new field of complexity theory. Complexity allows IWRM to be viewed from a new perspective, where all of the nuances and dynamics associated with integrating continual growth with man agement across environmental, economic and social needs are simplified. This clearer understanding is crucial if IWRM is truly the paradigm of the future. However, the ideas regarding the application of complexity theory to IWRM are scattered and range f rom the quantitative to the qualitative. In this work, literature regarding complexity and IWRM in urban water systems is summarized and collected into a coherent framework that describes Integrated Urban Water Management (IUWM) in the language of complex ity. This framework is then used to identify similarities in the progression towards IUWM in three different urban water management systems over time. Integrated Urban Water Management and Complexity As a developing field of study, no one identifiable complexity theory currently exists, however some generalities appear to hold across all interpretations (Manson, 2001) In general complex systems can be understood as a balance between order and chaos that is maintained through a dynamic and prescriptive framework in which

PAGE 148

148 uncertainty and unpredictability dominate and surprise and structural change are inevitable (Holland, 1992; Holling, 2001; White and Engelen, 1994) In order to better understand what complexity means for IURM, it is worth Complexity can most simp ly be seen as the relationships between linked components within a system where the dynamics between the individual components are no longer simple, but represent higher order, non linear behavior (Manson, 2001) The components are the pieces of the system that grow, change, and interact with each other to varying degrees. Components of urban water management systems are represented b y the consumers, water providers, external regulatory, legal or political institutions, as well as the environment, water infrastructure. Complex systems are not static entities, they actively react and anticipate changes, where components are perpetually redefining and revising their relationships to each other. These dynamics give complex systems emergent properties, where the capacity of the system is greater than the sum of its constituent parts and the capacity to self organize, or change its interna l structure to better adapt to its environment (Manson, 2001) As such, the behavior of these systems is far from optimal, as const ant evolution prevents the (Holland, 1992) This form of behavior preven ts the complex system from conforming to classic equilibrium based models that rely on linear relationships, predictability and reducibility (Baynes, 2009) While the application of complexity to natural resource management research has been fruitful (Allen et al. 1999; Cowie and Bor rett, 2005; Ostrom, 1999; Rammel et al. 2007) literature regarding complexity theory specifically in the context of IURM is less

PAGE 149

149 common. The following section attempts to reconcile the gaps in the IUWM literature by applying concepts found in the natura l resource management literature to IUWM. The work by Geldof, (1995b) is one of the few wo rks specifically addressing this topic, providing insight into how IUWM systems should be assessed, and outlines problems that occur when complexity is not a part of the assessment. The ideas proposed in Geldof, (1995b) state that IUWM systems can be viewed from both internal and external perspectives. Internal assessments yield information a bout the systems components, where interactions within each component are simple, follow linear, equilibrium based models, and can be summed to represent the component as a whole. External assessments approach IUWM as a complex system, where complexity as the balance between order (maximum structure), and chaos (maximum entropy). In complex systems, assessments examine the relationships between components, and the management system itself with regards to the various aspects (i.e. environmental, economic, organizational which it must manage. These assessments assume that components maintain complicated relationships, and thus produce positive feedbacks with non linear behaviors. While simple water management systems with few components and straightforwar d relationships may be adequately described using internal assessments, problems arise when an internal approach is used to assess complex IUWM systems. Geldof, (1995b) described these problems in terms of scale, level and assessment. Problems of scale exist because many components in complex systems exist at different scales, both temporally and spatially, rendering them incomparable within an equilibrium based system. Problems of level exist because complex systems account

PAGE 150

150 for more than physical and chemical processes, but biological, social and intellectual processes as well, where the hig her the level, the greater the positive feedback. Finally, Geldof, (1995b) described prob lems of assessment, which reiterates the concept of change with time. An important limitation of this work, however, is that it does little to explain how complexity evol ves over time. The gap in the IUWM framework proposed by Geldof, (1995b) is more thoroughl y addressed in the natural resource management literature by Tainter and collaborators. Previous assessments of the relationship between society and resource use has shown that the ability to access and utilize resources has played a critical role in huma n growth and evolution (Tainter, 1988) As populations congregated into urban areas, the problems of resource acquisition became greater as more complex solutions were required to meet needs (Tainter, 1988) However, history has also shown that complexity is costly. It requires time, energy, financial capital and labor to create and maintain systems that grow both in a tangible sense (i.e. number of parts and people), as well as intangibly (i.e. increase in knowledge, information, and regulation) (Tainter, 2006) Allen et al. (1999) can be defined by two processes: elaboration of structure and elaboration of organization. Elaboration of structure refers to al l those processes and components that steadily add and change complexity incrementally, whereas elaboration of organization occurs only occasionally and suddenly, and is the response to imbalances in complexity when over elaboration of structure has strain ed the current resource base. These increases in organizational elaboration serve to reframe resource consumption

PAGE 151

151 and the cost of problem solving by either redefining the relationship of the users to the resource, or through the implementation of efficien cy, which reduces burden of complexity by streamlining, reducing or completely removing components, or whole processes, from the system (Allen et al. 1999) As such, one sees system behavior becomes simpler as the levels of organization increase, but becomes more elaborate as the amount of structure increases (Allen et al. 1999) A Complexity Framework For IUWM Based on these two different understandings of how complexity ties to resource management, a more complete understanding of how complexity evolves o ver time in urban water management systems is proposed. This framework breaks the evolution of urban water management into three phases: 1) Elaboration 2) Complication and 3) Redefinition Elaboration The first phase, Elaboration, refers to the incremen tal gains in complexity via infrastructural elaboration. In this context, Elaboration describes the major devices used to resolve water supply issues, typically represented as infrastructure based projects constructed to access new sources or enhance curr ent system productivity, or the incorporation of new environmental or institutional levels to management. While infrastructure projects are added as direct solutions by the water provider, additional levels of management are more frequently either imposed or otherwise indirectly added to the current set of management responsibilities and restrictions over time. Complication The second phase is Complication, and represents the decrease in the marginal returns on progressive expansions of elaboration, whe re additional infrastructure

PAGE 152

152 produces a relatively small per capita increase in resources procured. Complication can also occur where incremental elaboration within the current, traditional, management framework leads to problems of scale, level and/or as sessment. Complication occur s where the continued assumption of linearity, equilibrium based dynamics, and optimal, fixed outcomes, no longer adequately reflects the positive, non linear behavior of the complex system being managed. In a deviation from t he work of Geldof, (1995b) however, here the problem of level is represented by three asp ects of management, rather than four processes. While similar in nature, levels described as organizational, environmental and institutional fit the narrative of this work better. The organizational level represents those processes that a provider is dire ctly in control of, including decisions about routine operations, staffing, and capital improvements. The environmental and institutional levels describe processes a utility must account for that are not under the direct control of the provider. These pr ocesses take the form of external interactions providers must have with other regulatory, governmental, or fellow provider groups. Redefinition The third phase is Redefinition, and represents the critical point at which systems either fail, or redefine t hemselves within the complexity framework. Redefinition can happen by way of elaboration of organization where resource consumption associated with local problem solving is far below the new expanded universe of resource availability. Alternately, Redef inition can occur via elaboration of efficiency, where the burden of complexity is streamlined or eradicated by reducing the amount of elaboration.

PAGE 153

153 Applying this progression of complexity to urban water systems evolution over time should show that when u rban areas were small and demands were relatively low, the complexity of urban water systems was low. In the US, increasingly dense urban clusters, combined with no formal distribution system to deliver clean water and remove wastewater led to a rash of w ater borne illnesses and deaths in the 1800s (Melosi, 2000) To reduce health related issues, municipal water systems were design ed. Once systems were installed, attempts to streamline and increase water production efficiency and service area size occurred, the problems associated with the management of more infrastructure and demands, as well as less local available water, also gr ew. Increasing the volume of water available perpetuated growth, such that system complexity increased to the point where an equilibrium based, static method to water management was no long available and problems with management arose. Once management has reached the point where these three problems exist, components can no longer be managed individually or be expected to not influence one another. The events leading up to this point may be hard to define, and often cannot be detected until the complexit This framework is conceptualized in Figure 6 2, which shows how each component changes over time. For a given system constraint, such as the amount of water available at any given t ime, the benefits from each elaboration event (i.e. per capita water available) slowly decreases (1) until a new elaboration is introduced. However, the gains associated with each successive elaboration event become smaller as time progresses (2). These reduced gains continue to occur until further elaboration is no longer an acceptable solution, or when some minimum tolernance threshold is

PAGE 154

154 reached (dotted line). At this point, a major shift, or redefinition (3), of how management occurs is required for further system gains to be made. Common Threads of Complexity As discussed earlier, previous assumptions about the traditional urban water management framework implied that management as a whole could be described as the sum of its components. Planning an d system adjustments could be made based on concepts of linearity, additivity, equilibrium dynamics, and optimal, fixed outcomes (H olland, 1992) Ongoing research however indicates that traditional assumptions are no longer applicable. In the past, urban water systems could address resource needs within an operational context using a small local infrastructure, with supplies and de mands operated predictably and in isolation from the greater system within which they reside. Today, these same systems have expanded in size, operating large systems within an integrated operational environmental institutional context, where interactions and responses between the linked levels of context occur are complex and often unpredictable. Within this new integrated framework, multiple contexts are connected by feedbacks to create a complex system that traditional equilibrium based models of manag ement no longer fit. Rather, these systems must now be understood within a non equilibrium context, such as that encompassed by complexity theory, where control and order are no longer hierarchical in nature, but emergent (Dooley, 1997) In order to reinterpret urban water management through US history in this new language of complexity, we assess three cities not using internally focused, detailed, historical assessments of the components of each as many other authors have done, but rather using an external assessment of the patterns between urban water systems (Geldof, 1995b) By defining system complexity using the framework created above, it

PAGE 155

155 is possible to see the elaboration of complexity through the accumulation of solutions to problems of resource acqui sition. This external assessment allows us to follow the thickening thread of complexity through urban water management systems over time. The insights thus drawn allow us to identify similarities that exist between very different urban systems while nev er neglecting the idiosyncratic contexts of each.. An abbreviated narrative of each case study is given below to outline the major factors affecting the three phases of management as they occurred over time. While these phases are in reality both multi dimensional and often interrelated, an attempt was made to clearly track major Elaborations to the system, as well as identify the Complications management attempted to continue operating as a simple system despite the added complexity, and the Redefinitio n that ensued. Further discussion about each phase of complexity within the context of the case studies will be described at the end. Case Study Descriptions To examine the evolution of complexity in urban water management systems, three major urban areas in the United States were selected as case studies. Each case involves distinct problems and stakeholders; however they are similar in that the goal of each management system was to support urban growth through the continual acquisition of water supplies Three highly publicized cases were selected because much information about them is readily available and yet, despite their prominence, no direct comparisons have yet been made either within or across these cases to evaluate water supply system complexi ty. Thus, the goal of this study is to provide a collective understanding of complexity from multiple, different perspectives.

PAGE 156

156 Tampa Bay Region, FL Current c ontext The Tampa Bay region lies on the west coast of Florida and is comprised of two neighborin g cities, Tampa, and St. Petersburg that were both established in the late 1860s and together represent one of the largest urban areas in the state with approximately 3 million people within its 2000 km2 boundary. The Tampa Bay region is sub tropical, rec eiving on average1100 mm/yr of rainfall. Water provision within the urban area occurs at two levels. Distribution services are maintained under municipal control as the Tampa Water Department and the St. Petersburg Water Resources Department. Water supp ly sources and services are primarily the responsibility of the regional provider, Tampa Bay Water (TBW). While St. Petersburg obtains all of its water from TBW, Tampa maintains its own surface water supplies, periodically purchasing water from TBW to su pplement these sources. This regional supplier manages a combination of ground, surface and alternative water sources, providing 210 billion liters of water per year to approximately 2.5 million people in 2010. TBW is the product of a regional agreement between three counties and three cities to ensure environmental standards in the region are maintained. Management h istory Tampa Bay region was focused primarily around local ground water resources. Due to coastal proximity of the two major cities in the area, Tampa and St. Petersburg, and the relatively small volume of water locally available, over pumping in the area eventually led to salt water intrusion problems and the so urces were abandoned (City of St. Petersbur g, 2011; City of Tampa, 2011)

PAGE 157

157 St. Petersburg bought and started developing wellfields in southern Pasco and northern Hillsborough Counties (City of St. Petersburg, 2011) whereas the City of Tampa built and designed infrastructure to capture water from the Hillsborough River, and only later, groundwater in northern Hillsborough Count y (City of Tampa, 2011) As growth adjace nt to the urban wellfields started complaining about declines in local lake, wetland and groundwater levels. At the time, it was unclear whether the environmental desiccation taking place was due to groundwater pumping or low rainfall, however the citize n anger and stark evidence of negative environmental impacts were strong enough to compel the Florida Legislature to create the West Coast Regional Water Supply Authority (WCRWSA) in 1974 (Scholz and Stiftel, 2005) This group consisted of the three cities and three counties involved in the water supply dispute, and was originally intended to resolve the problems associated with the resources. Member governments were supposed to buy water from the WCRWSA, who was to control the existing wellfields (Scholz and Stiftel, 2005) while regulation of groundwater withdrawals was to be performed by the Southwest Florida Water Management District (SWFWMD). The success of the WCRWSA was limited as an unequal power distribu tion and general unwillingness to cooperate prevented any practical regional management from occurring. As such, groundwater withdrawals continued unabated, and after another series of citizen complaints about local environmental and supply conditions the SWFWMD started in 1993 to revise its groundwater permitting framework to include issues regarding over pumping (Scholz

PAGE 158

158 and Stiftel, 2005) The ensuing volley of lawsuits and water restrictions between water providers, SWFWMD, and citizen environmental groups fervently continued until 1998, when all groups agreed to the Northern Tampa Bay New Water Supply and Ground Water Withdrawal R eduction Agreement (the Partnership Agreement) (Scholz and Stiftel, 2005) As part of the new agreement, each member of the former WCRWSA forfeited their source ownership rights and transferred all water production responsibilities to the newly formed Tampa Bay Water, which has been supplying water to the Tampa Bay Region ever since (Scholz and Stiftel, 2005) Los Angeles Region, CA Current c ontext The city of Los Angeles, California was established in 1850 after having been a missionary outpost for approximate ly 70 years (LADWP, 2011) Today, the entire city is served by one utility, the Los Angeles Department of Water and Power which d elivers water to a population of over 4 million people within a 1200 km2 area (LADWP, 2011) The municipal utility itself was foun ded in 1902 and collects water from over 360 km away using one of the largest water supply infrastructure systems in the US. The water sources for Los Angeles are varied and include multiple groundwater sources, aqueducts, and water purchased from a regio nal provider, who collect and distribute water from the Colorado River and San Joaquin Valley (MWD, 2011; SWP, 2011) The utility was delivering approximately 760 billion liters per year as of2009, with average per capita consumption of approximately 500 liters per day (lpcd) (LADWP, 2011) Located in the semi arid desert of southern California, and receiving average rainfall of 900 mm/yr, Los Angeles, and the state of Calif ornia, have developed an impressive

PAGE 159

159 array of legal and institutional frameworks, compacts and agreements to allocate water supplies. Ma nagement h istory The local Los Angeles River served as the sole source of water for the City ter shortage problems, due to population growth and water pollution, forced the City to seek new supplies (LADWP, 2011) Subsequen t system expansion occurred under the direction of William Mulholland. Initially, Mulholland focused on increasing the capacity and quality of the local system by adding storage capacity to the Los Angeles River Basin as well as modernizing and extending the early, existing distribution system (LADWP, 2011) prodigious growth, however, Mulholland sou ght a larger source to anticipate increased water demands. Mulholland proposed building an Aqueduct to utilize water from the Owens Valley, nearly 360 km away (LADWP, 2011) Despite the enormity of the project, both physically and financially, it was approved in 1905 and completed in 1913 (LADWP, 2011) The Los Angeles Aqueduct has the capacity to transport nearly 1.2 billion lpd, however continued growth and needs for water led to the construction of the Colorado River Aqueduct in 1939, and a second Los Angeles Aqueduct to Owens Valley in 1941 (LADWP, 2011) While the second Aqueduct was the sole implementation of the City, the Colorado Ri ver Aqueduct was built under the auspices of the Metropolitan Water District of Southern California (MWD), as suggested by Mulholland (Reisner, 1993) As one of the 26 members of the MWD, Los Angeles obtains its water from this regional project, as well as from the state built water conveyance system (SWP), which currently delivers water to nearly two thirds of the on, including Los Angeles (State of California, 2011)

PAGE 160

160 While the engineering feats produced by the City of Los Angeles are insp iring, the manner in which the City appropriated water from distant sources was not. To accomplish the acquisition of water rights in Owens Valley for the initial Aqueduct, agr eeing to a new federal irrigation project. Once the ruse was discovered, violent tactics were used to secure the remaining water rights needed, while equally violent tactics were used by the residents of Owens Valley to prevent the export of their water r esources (Reisner, 1993) Courts upheld Los Angeles claim to Owens Valley water rights, and over the next decade, diversions woul d completely drain Owens Lake. Further conflict was introduced when Los Angeles constructed their second Aqueduct, which led to the severe degradation of the Mono Lake Basin, reducing the lake volume by 30% and crippling an important feeding stop for mill ions of migratory birds. The environmental damages incurred with the second aqueduct sparked a suite of lawsuits against the City, and the eventual reductions in withdrawals from both aqueducts were mandated by court orders (Reisner, 1993) Washington D.C. Curr ent c ontext The city of Washington, DC was established in 1790 by Congressional decree. It is a federal district nestle d between the states of Maryland and Virginia, receiving on average 1000 mm/yr of rainfall. In 2010, 600,000 residents in the 1878 km2 DC service area used 150 billion liters of water delivered by the D.C. Water and Sewer Authority (DCWASA). The service p rovider for the area has undergone restructuring three times since 1935, becoming the DCWASA in 1996. DCWASA does not maintain any of its own sources, but rather purchases its supplies from the US Army Corps of Engineers

PAGE 161

161 (USACE) owned Washington Aqueduct (Singhal, 2010) The Aqueduct utilizes the Potomac River along with several other regional water providers. User agreements contr ol the allocation of the river flows, particularly in times of drought. Management h istory management is based largely on the work produced by Ways, (1993) Originally, the water needs of the city were met using local springs and wells, however growing water needs for munic ipal consumption and fire protection motivated Congress to authorize the US Army Corps of Engineers (USACE) to construct the Washington Aqueduct. The Aqueduct was the brainchild of Lt. Montgomery C. Meigs and transports water from the Potomac River throug h 20 km pipeline into three main reservoirs. The Aqueduct was completed by 1864, and was intended to deliver 260 million liters per day (lpd). Based on his estimates of population growth and water demands, Meigs predicted that this supply would last the city for the next 200 years. In reality, the supply was only large enough to last 68 years, and in 1927 another intake was added to the system, bringing the total deliverable volume up to 380 million lpd. While this Aqueduct was capable of providing ampl e amounts of water, the poor quality was highly criticized by the citizens was installed in 1905 and a second in 1927 along with the new intake. Additions to t he pumping capacity at all intakes, and further development of treatment facilities. While control over the Aqueduct has always remained with the USACE, its supply is primarily used by the District of Columbia Water And Sewer Authority (DCWASA). the

PAGE 162

162 Fairfax County Water Authority (FCWA) and the Washington Suburban Sanitary Commission (WSSC). Between these three users, a substantial volume of the local water resources was being allocated to urban water provision. Vulnerability of the USACE to start studying low flow forecasts and drought contingency plans while any further water supply development was halted. In 1963, the USACE released a drought contingency plan recommending the implementation of 16 new reservoirs within the Potomac River Basin. This plan was ultimately rejected, however, not only for operational and financial considerations (both Congress and the general public opposed this plan), but because scientific studies of the basin had indicated that in times of drought, coordinated operation of a few strategically placed reservoirs could provide the equivalent protection of the 16 proposed reservoirs (Hagen et al. 2005) In 1978, the Low Flow Allocation Agreement (LFAA) was implemented, which bound the USACE, State of Maryland, Commonwealth of Virginia, District of Columbia, WSSC and FCWA to the coordinated operation of all system sources during which each would receive a vailable water in proportion to their relative current consumption. After several modifications to the initial LFAA, the 1982 Water Supply Coordination Agreement created the Co operative Water Supply Operations on the Potomac (CO OP), which currently mana ges the regional water supply coordination. Elaboration Figure 6 3 6 4 and 6 5 provide a series of timelines of the major events and components of each system through time in the context of Elaboration, Complication and Redefinition. These figures are m eant to support the text described in the following sections to show how various components of urban water management create complex

PAGE 163

163 interactions that systems were small and complex ity was low. Regions had modestly sized service populations whom they supplied water to from local sources; groundwater in the Tampa Bay Region, springs in Washington DC and a river in Los Angeles. Large infrastructure was not needed at this point; howev er major improvements in distribution systems were increasing the total amount of infrastructure each region had to manage. During this period few institutions were in place to regulate the way in which water resources were procured, distributed or dispos ed of, allowing these early systems to operate in relative isolation from other water management systems in any context other than urban supply. It is worth noting that from the 1850s through the next century, urban areas, and indeed all of the US, were firmly set in a strong ethic of growth ( Figure 6 3A 6 4A and 6 5A ). Growth was essentially the equivalent of productivity, and productivity could not be gained without ample access to resources. Therefore, in response to the growing need for water ( Fi gure 6 3B 6 4B and 6 5B ), each region set about on a strategic plan for first maximizing, and then moving beyond locally available resources. The dates on which these trends begin span nearly a century, with Washington DC completing its first major feat of infrastructural water supply in 1864 ( Figure 6 4C ), the completion of Los Angeles first aqueduct in 1913 ( Figure 6 5C local river and non local groundwater in the 1920s ( Figure 6 3C ). In each of these cases, tappin g a new source was a major decision requiring signficant financial and labor investments, in the cases of Los Angeles and Washington DC, and substantial funds and work to complete the expansion in the Tampa Bay region. After the initial major investement in infrastructure, the amount of additional physically based

PAGE 164

164 infrastructure solutions tapers off. Additional projects are incorporated into the system of each in the form of increased pumping capacity, new treatment facilities, and additional distributio n lines, however the incremental benefits derived from each addition of infrastructure produce smaller and smaller returns in terms of the water produced per capita. At the same time elaboration was also occurring in the form of regulatory controls, inst itutional rules and regional user responsibilities ( Figure 6 3D 6 4D and 6 5D ). In the Tampa Bay Region, the formation of the SWFWMD in 1961 saw the introduction of permits on groundwater withdrawals and the independent, yet similarly focused, expansion of groundwater wellfields and pumping rates put Tampa Bay water providers in close proximity to each other. In Washington DC, the 1960s saw increasing concerns waterbody, and the number of regional providers that were now tapping the same river Angeles grab for Owens Valley water placed the city in direct contact with the residents of Owens Valley. Complications The Complications in this framework represent the impacts of decreasing marginal returns on the elaborations achieved prior. This phase represents where system imbalances were at their greatest, and where frustrations associated wi th problems of scale, level and/or assessment become a burden. In the case of the Tampa Bay Region, complications arose as uncoordinated, regional water production and low rainfall led to a rash of citizen complaints about environmental degradation. Man agement did not have the technology, no the

PAGE 165

165 motivation to accept environmental responsibilities at the time. Problems were management paved the way for more serious p roblems for water providers, as citizen, government and utilities all became embroiled in years of litigation. Los Angeles had similar problems to Tampa, in that they too were faced with forcefully embracing management for the environment. A sour histor y with the Owens Valley residents, and an insatiable need for water had left the sources of both of Los Angeles Aqueducts in poor condition and litigation followed. However, Los Angeles lack of desire to interact with Owens Valley residents was near compl etely reversed in the context of their interactions with the MWD and SWP. Cordial agreements and financial support showed that Los Angeles could cooperate, given the right circumstances. Washington DC was the only case in this study that took on the cha llenges of environmental and cooperative management gracefully. The proactive realization over low flow conditions and the ability to meet with and come to user agreements with the other providers in the area shows a very different, and milder, set of co mplications. While this may have been in part due to the involvement of the federal government (USACE), there is reason to believe that this was not the major reason for the minimal amount of complications leading up to Redefinition. Redefinition The poi nt of Redefinition for each case represents a major change in the urban water systems operate upon dealing with their Complications. As mentioned earlier, Redefinition can be through a reduction in complexity via efficiency, or through a reorganization of management by which complexity is redefined.

PAGE 166

166 In the case of the Tampa Bay Region, redefinition occurred with the 1998 Partnership Agreement and the formation of Tampa Bay Water. This resolution represented a major reorganization in the way water resou rces were managed in the area, placing all of the groundwater within the control of TBW. Regional water provision brought IUWM complexity to the next level, where individual complications were erased as management of environmental and cooperative responsi bilities became the major responsibility of the regional provider, not the individual municipal systems. The example of Los Angeles revealed a system that chose to redefine itself through austerity. This redefinition came about as a court ordered mandat e to reduce withdrawals from Owens Valley, but can also be seen in the decisions made by Los Angeles earlier in its management history to encourage other entities to take on mega water conveyance projects, saving them the resources. Today, the Los Angeles region has become one of the lowest per capita users of water in the nation. Washington DC faced a similar style of redefinition, although less obvious than the formation of the TBW. While the Aqueduct and provision systems remained independently contr olled, the forward thinking agreements for operation during low flow events place water management, occasionally, within a new framework called the Co operative Water Supply Operations on the Potomac (CO OP) As the complications experienced in this syste m were mild at best, it is therefore not surprising that no drastic redefinition had to take place. Discussion It is worth pointing out that the historical path and specific outcomes of water management for each case is unique. The initial factors that defined how each city grew and the specific interactions between environmental, economic, and social

PAGE 167

167 distinctly different. No two cities experienced the exact same popula tion growth rates, had the same amount of available local water or funds for development, nor did they grow under the same institutional oversight or within exactly the same cultural paradigms. By stepping back and making an external assessment of each c as e, the patterns associated with the emergence of complexity become clear ( Figure 6 3E 6 4E and 6 5E ). In all three cases there was evidence of an inevitable and steady increase in elaboration through the creation of water supply infrastructure as each case sought large, and more distant sources to meet needs. As infrastructure developed, so did organizational frameworks (i.e. SWFWMD) to deal with the growing large scale use of water resources. Concurrently, although not necessarily obviously, growing complications mirrored each step of elaboration until the level of the complexity was no longer manageable within the current framework. In the cases of Los Angeles and Tampa, these critical junctures were punctuated by litigation, and open stakeholder fr ustration. Critical junctures of complication were not always marked by dissent complications in management was with the redefinition of how management occurred (dark line in Fi gure 6 3E 6 4E and 6 5E ). In the Tampa case, this was most obvious in no new management entities were formed, rather an earnest and intense move to reduce water usage through conservation forced the system to fundamentally redefine how it managed its resources. Finally, Washington DC was perhaps the most subtle

PAGE 168

168 and successful of the three cases studied. Management is temporarily redefined in times of drought, where lo cal users collaborate to fairly allocate supplies between users and the environment. Thus, complexity, despite the differences in space and time, can be seen as an ever present force, guiding and reshaping management according to the needs of society. Fro m this study, it becomes clearer that complexity is present in all water resources management systems. While its influence is not always directly observable in the form of Complications or Redefinitions, it steadily increases with time. In a sense, the p aradoxical nature of complexity also becomes clearer through this study. Increased complexity is not only the solution to our problems of water management, but the problem that needs to be solved. Therefore the amount of resources invested into resolving issues of complexity can only ultimately result in the need for more investment; however the concept of Redefinition helps temper this. However, as Allen et. al (1999) and others in the natural resources management literature have concluded, by managing at the right scale, namely the watershed, or ecosystem scale, rather than for the resources, the severity of the Complications and the effectiveness of the management system would greatly increase. For this to occur, however, a better understanding of wa ter management systems is required. This means knowing how The role of interdisciplinary research in this case cannot be emphasized strongly enough. To understand systems in their entirety, specialists studying the parts and generalists synthesizing these parts into the larger picture become necessary, as complex systems make it difficult for any individual to accomplish this on their own.

PAGE 169

169 Conclusions and Implications Face d with the imminent arrival of the new water management paradigm, IWRM, this study used current literature to more closely investigate the role of complexity in the evolution of urban water management systems over time. The transition from a small, simpl e water supply system to a large, complex entity that manages water not only for urban development, but also in the context of environmental health and regional sustainability, needs to be more completely understood if sustainable IWRM is to occur in the f uture. This study re translated the history of three urban management systems using the language of complexity and found evidence of, a direct role for, complexity in all cases. The framework presented and applied appears to offer new insights into both the unique manifestations of complexity in each individual case, and the common general trends shared between cases, which describe the evolution of complexity in IUWM through inveterate elaboration that leads to emergent complications, which eventually fo rce a critical redefinition. Given this evidence for the importance of complexity, both as a characteristic of these systems and as a force that directs them, a revised framework for IWRM as a whole is proposed ( Figure 6 6 ). This new framework explicitly recognizes the existence and influence of complexity in IWRM. It is the component that converts IWRM into a web, and by placing it above the other three, it is intended to indicate that complexity is in fact the directing force in the evolution of enviro nmental economic social systems. This is not to say that these components do not direct as well, but is to say that the do not direct in isolation, but always in integration, and the emergent forces of complexity collectively capture all their directions into a greater one that is, likely, greater than the sum of the directing parts.

PAGE 170

170 Figure 6 1. Conceptual diagram of IWRM components and their interactions. Figure 6 2. Complexity f ramework For IUWM R epresented by 1) e laboration, 2) c omplication and 3) r edefinition.

PAGE 171

171 Figure 6 3 Tampa t imelines.

PAGE 172

172 Figure 6 4 Washington DC t imelines

PAGE 173

173 Figure 6 5 Los Angeles t imelines

PAGE 174

174 Figure 6 6 Revised conceptual diagram of IWRM including complexity.

PAGE 175

175 CHAPTER 7 CONCLUDING REMARKS As populations in crease, the growing demand for water means sustainable water management is ever more important (Gleick, 2000) This is particular ly so in urban areas, where intensive localized demands can place large strains on local resources (Levin et al. 2002) While muc h research on sustainable water management issues is being done, there are few studies examining these problems from a national perspective. This broad scale assessment is necessary if we are to fully understand how best to manage for the future. This pr oject was interdisciplinary in nature, and was designed to span the gap between elements of the physical and social sciences in order to provide a more comprehensive understanding of the urban water resources system. The overall goal of this dissertation research was to identify and evaluate the major causal factors affecting water management in urban areas across the US. This research, in aggregate, provided a new metric for assessing urban water availability and vulnerability and investigated a variety of different ways in which urban utilities have responded to vulnerability, and how complexity has shaped urban management today. The five sub goals of the dissertation project are discussed below in relation to their context within and contribution to the scientific literature. The major findings of this research, together with the limitations associated with each element of this study, are considered and future avenues of investigation in each topic area are proposed.

PAGE 176

176 Chapter 2 : Creation and implement ation of a storage based water availability and vulnerability metric for assessing US urban areas Accurate assessments of urban water availability are crucial for understanding where vulnerability exist, and are reportedly a major concern of the water ind ustry (Runge and Mann, 2008) While many individual utilities have strategies for ensuring they have adequate supplies of water f or urban areas in the near future (Means, Ospina, et al. 2005) the only national assessments of water availability and vulnerabi lity that have been completed fail to account for the role of storage, a critical (Hurd et al. 2 004; Roy et al. 2005; Sun et al. 2008; Taylor, 2009) In this work, a significant effort was made to create and implement a new quantitative and storage based methodology for measuring water availability and vulnerability in urban areas, based on the mo st recent hydrogeographical data and information about water utility supply systems. Results from this work showed that the water availability assessments that included storages were significantly greater than those that did not. Despite this general pat tern, storage based assessments were also capable of identifying urban areas where average annual water availability could be an issue. The water vulnerability of each urban area was quantified using this national availability assessment, revealing a subs et of urban locations who were most water vulnerable. These results were corroborated using a text analysis of the frequency with which news articles about urban water scarcity appeared for each location. A national assessment such as the one provided in this work fills a current gap in the literature; how can we quantify and assess at risk populations with regards to water availability and vulnerability. This effort was hampered, however, by a paucity of quality

PAGE 177

177 data with which to make such assessments. The coarse resolution of some of the data available, and the lack of data for other components of the analysis mean that this assessment was forced to make many assumptions, particularly regarding the storage components of each urban area. In addition, b ecause a storage based approach has not previously been attempted, there was no way to verify the results of this analysis against other studies. The same holds true for assessments of water vulnerability. Previous assessments of urban vulnerability are based on a variety of different methods that normally do not account for storages, making a validation of the method presented here difficult. In response to this, an attempt was made to verify the results of urban vulnerability assessments based on the f requency with which news articles about urban water scarcity in the respective locations appeared. While this approach proved useful, a more refined and rigorous method would be needed to distinguish between the different types of vulnerability made in th is assessment. Such a method would in turn require substantial additional data. In the future, a centralized database of higher resolution hydrologic information, including flows and storages, for the entire nation is needed if integrated water resources work is to continue at the national level. This includes creating a centralized data repository for the storage of critical elements within the human system, such as those water supplies owned and operated by water utilities. Chapter 3 : Assessment of the relationship between urban water availability and urban water management in terms of the complexity of the infrastructure and management organization Understanding urban water vulnerability requires understanding the context in which water systems are ma naged. While many studies have examined the structure and operations of individual or groups of utility systems (Galloway; Hagen et al. 2005;

PAGE 178

178 Means, West, et al. 2005) there currently exists no synthesizing literature on how urban water systems as a whole have adapted to water stress. Recent work in the natural resources management literature has promoted the idea that complexity is the major mechanism by which systems adapt to increasing needs for resources (Allen et al. 1999; Tainter, 1988) In the management of resources, such as water, this is yielding more and more complex management systems, yet application of these ideas had yet to be embraced. This work attempted to measure complexity in urban water management systems by evaluating system complexity as a function of their infrastructure and management organization, and to use the information to see how complexity has evolved with water availability. Results revealed that increasing complexity is indeed one mechanism by which urban areas deal with low wa ter availability. Even when using very general data about urban water systems, increases in complexity both in terms of infrastructure and management were observable in locations were water was more scarce, and in regions were urban areas were sharing s ources. This work was limited in part by the fact that national assessments require the use introduce much variability into the analysis. While variability can be dealt with using various data discretization methods, it is believed that more information about each metric, infrastructure and management, would help refine these results by explaining some of outlying points in the current dataset used and providing a more co mplete context within which management and infrastructure complexity could be understood.

PAGE 179

179 For future work to occur, it will be necessary for researchers to organize more information about urban utilities at the national level. Once this has been done, a more precise analysis of complexity could be performed to better separate some of the general, yet nuanced, details of urban water management. In addition to issues regarding data, a more robust method for analyzing the effects of proximity issues on ur ban area complexity would open the way for further work regarding management, collaboration, and complexity. Chapter 4 : Determination of the extent to which financial and operational water provision metrics fit into the urban complexity paradigm While m uch economic research has been done regarding urban water provision services (Abbott and Cohen, 2009) how these analyses tie into the national urban water complexity paradigm remain unclear. Urban analyses, to date, have studied a range of urban processes, including transportation networks (Lmmer et al. 2006) urban size (Batty, 2008) and innovation (Bettencourt, Lobo, Helbing, et al. 2007) to name a few. These analyses have only briefly studied how complexity impacts the form and function of supply networks i n urban areas, however (Kuhnert et al. 2006) Urban water supply is a critical component of urban systems, and related research from both economic and complexity studies suggests that attributes of water provision should follow similar patterns of non linear scaling seen in other urban processes when compared across urban size. This work tested the theory that urban water provision systems fit into the developing urban complexity paradigm. Financial and operational data from a subset of urban utilities were used to assess how various aspects of utility costs and operations scale against population size. Results of these scaling as sessments indicated that both

PAGE 180

180 of these components of utility systems scaled similarly to other work regarding supply networks, showing indications of economies of scale across population size. In addition, two metrics were designed and tested look at relat ionships between urban scaling and collective operational complexity. The two metrics developed in this effort, although simple in nature, could be used successfully to analyze system components collectively, including contributions from other providers. Urban water systems are more than the sum of their financial and operational parts. While the data used in this study lead to use insights about these two components of urban water provision, more information is necessary for better understanding how othe r aspects of urban provision systems scale with complexity. This is especially so if we wish to further advance the application of a combined complexity metric. The further study of urban water provision systems in the context of urban complexity has th e potential to yield some insightful results. The combined complexity metrics offered in this work necessarily focused on only some elements of complexity. For this reason, metrics such as these have significant potential as benchmarking indicators for ur ban systems, and may be invaluable as water management become a more holistic process. This approach therefore produced insights into only those aspects of how urban utilities function. Additional development of a more rigorous and comprehensive combined complexity metric would help to provide a clearer picture of how complexity is expressed in urban water provision systems.

PAGE 181

181 Chapter 5 : Measurement of the degree to which participatory collaboration is being used by urban water utilities to make decisions about water management and planning It is well known that information and system knowledge play an important role in the growth and adaptability of organizations (Anderson, 2008; Cortner and Moote, 1994; Dyer and Singh, 1998; Hackett et al. 1994) With water utilities facing a multitude of problems in an increasingly uncertain future, their adaptability becomes a key component of their ability to meet t he demands of the future. When considering US water utilities as a whole, the avenues by which they communicate, and with whom they exchange information, has been unclear. A survey was implemented in this work to better understand the degree of collaborat ive participation; how information is being exchanged and used by utilities, how it affects utilities management decisions, and where these processes may be more or less important within the nation. Results from the survey showed that for the types of coll aborative participation explored, the majority of respondents reported some minimal amount of participation in a wide range of groups and activities. When the levels of collaborative participation were compared to the self reported importance of their man agement plans, there was an indication that those utilities who collaborated more also had higher quality management plans. While the results from this work provide some of the first quantitative evidence of collaboration and its effects on management in urban water utility systems, the robustness of the study was limited by the low response rate (13%). While this response rate was comparable with other studies of the same sample population, its usefulness for being able to draw conclusions about the wate r industry as a whole is more limited.

PAGE 182

182 It is notoriously difficult to compel respondents to participate in surveys, therefore future work would benefit from the development of other metrics to measure the level of collaboration between utilities that did not involve self reporting. Additionally, further work examining the reasons for why urban systems participated as much or as little as they did would be useful for better understanding why collaboration works better in some places than others. Other wo rk that this study touched on, but could not examine in depth was the extent to which participation in collaborative activities becomes a burden to utilities. End with something to the effect of this survey being intended to set the stage/be a foundation for future work on this subject. Chapter 6 : Reassessment of the historic role of complexity in the development of urban water management All of the other studies done in this work show how utilities are responding to vulnerability in the present. However, if our goal is to manage urban systems using Integrated Water Resource Management (IWRM) strategies, it would be useful to know how urban systems have coped with vulnerability in the past and what some of the challenges of increasing complexity have been for urban water systems under earlier management paradigms (Bellamy et al. 1999; Biswas, 2004) From general knowledge, we know responses to vulnerability have generally led to increased management complexity (Allen et al. 1999) Therefore, in this work a general framework was developed for following and evaluating the evolution of water management complexity over time for three case studies: Los Angeles, CA, Tampa Bay Reg ion, FL, and Washington DC. The framework sought to consolidate some fundamental complexity theory as applied to resource management, and to then serve as a new language in which urban water management has evolved.

PAGE 183

183 From the application of this framework o n the three case studies we saw that a common thread of complexity could be identified running through each. The progression of management over time in each case following roughly the same patterns in evolution elaboration, complication, redefinition d espite the very different contexts under which they developed. Each urban area experienced an incremental increases in management complexity as urban areas, and thus water provision services, grew. Increased complexity was followed by an increase in the co mplicatedness of management, which at some point reached a critical juncture. When the complications of management were too great, each urban area redefined its relationship to how it used and managed water, allowing the system to internalize and successf ully incorporate some of the increased complexity using a new management framework. From these results we see that complexity is inherent to urban water management systems, both as a characteristic of such systems and as a force within them. To understand about how the system has progressed over time. This type of information may not be However, it als o appears that the general patterns are shared between cases, which is promising for a more general understanding of how urban water systems evolve. Final Thoughts However, during the course of this work, it became obvious how unorganized urban water man agement is at the national level. No single federal agency exists to regulate or create policy for urban areas, and therefore there also exists no national database of information relevant to US urban water management. This lack of an organized informati on collection system in some ways reveals why the work produced

PAGE 184

184 here is so necessary. Without a centralized, national database or benchmarking system, it is almost impossible to be able to quantify any aspect of water management; how much water exists, ho w much has been used, how much is expected to be available in the future, and how urban areas are adapting to changing hydrologic conditions. Without the ability to measure the current state of urban water resources, there is no way to review the past, or how availability and management have changed over time. The idea of such a centralized repository for information is not new, utilities have long been encouraged to participate in financial and efficiency benchmarking activities, which not only help utili ties identify where weaknesses in their operations may be occurring with regard to their peers, but also provide regional and national databases of information on urban water management. If used at a national level to collect information regarding all asp ects of water management, these data could be used to better identify which variables are important for benchmarking urban water sustainability, which would be invaluable to all those working in the water related industry. On a separate note, the work done here emphasizes the fact that there is still much to be learned about IWRM as a complex system; about how complexity manifests itself in the new paradigm, and what consequences these manifestations may have on the evolution of IWRM in practice. A better understanding of complexity in IWRM would help flesh out the work done here, potentially helping urban areas become more sustainable by showing them what can be expected, and how other urban areas have overcome similar difficulties. Having said this, this work has advanced the knowledge of

PAGE 185

185 IWRM and sustainable water management through the assessment of IWRM with regards to complexity. In doing this, we have found that: 1) that complexity is a necessary internal lens if we wish to truly acknowledge and man age the interconnected web of physical components comprising IWRM, and 2) that complexity is the most promising external lens through which to look down on the whole picture, in order to see the emergent properties not apparent from the internal view. This significance of the complexity lens for looking at the problem from both within and without may seem paradoxical, but paradox is itself the foundation of complexity: that something collective is more than the sum of its parts. This is the fundamental rec ognition we must hold to because its describes both why our management systems have failed in the past (we have not grasped the reality of the complexity inherent to the resource systems we have tried to manage), and why they might fail in the future as we continue to grapple with our complex resource systems but do so now with a management system that is itself a profoundly complex problem in its own right. Complexity is the problem and the solution accounting for complexity is insufficient to solve all the problems, but discounting complexity both exacerbates the problems and hampers the solutions.

PAGE 186

186 APPENDIX A LIST OF URBAN AREAS ID Urban Area Population 00199 Aberdeen -Havre de Grace -Bel Air, MD 174598 00280 Abilene, TX 107041 00766 Akron, OH 570 215 00928 Albany, NY 558947 01171 Albuquerque, NM 598191 01495 Allentown -Bethlehem, PA -NJ 576408 01927 Amarillo, TX 179312 02602 Ann Arbor, MI 283904 02683 Antioch, CA 217591 02764 Appleton, WI 187683 03358 Asheville, NC 221570 03763 At hens Clarke County, GA 106482 03817 Atlanta, GA 3499840 03898 Atlantic City, NJ 227180 04222 Augusta Richmond County, GA -SC 335630 04384 Austin, TX 901920 04681 Bakersfield, CA 396125 04843 Baltimore, MD 2076354 05167 Barnstable Town, MA 24 3667 05680 Baton Rouge, LA 479019 06058 Beaumont, TX 139304 07705 Billings, MT 100317 07732 Binghamton, NY -PA 158884 07786 Birmingham, AL 663615 08407 Bloomington -Normal, IL 112415 08785 Boise City, ID 272625 08974 Bonita Springs -Naple s, FL 221251 09271 Boston, MA -NH -RI 4032484 09298 Boulder, CO 112299 09946 Bremerton, WA 178369 10162 Bridgeport -Stamford, CT -NY 888890 10729 Brooksville, FL 102193 10972 Brownsville, TX 165776 11350 Buffalo, NY 976703 11755 Burlingto n, VT 105365 13375 Canton, OH 266595 13510 Cape Coral, FL 329757 14752 Cedar Rapids, IA 155334 15211 Champaign, IL 123938

PAGE 187

187 15481 Charleston, WV 182991 15508 Charleston -North Charleston, SC 423410 15670 Charlotte, NC -SC 758927 15832 Chattan ooga, TN -GA 343509 16264 Chicago, IL -IN 8307904 16885 Cincinnati, OH -KY -IN 1503262 17317 Clarksville, TN -KY 121775 17668 Cleveland, OH 1786647 18748 College Station -Bryan, TX 132500 18856 Colorado Springs, CO 466122 18964 Columbia, SC 4 20537 19099 Columbus, GA -AL 242324 19234 Columbus, OH 1133193 19504 Concord, CA 552624 19558 Concord, NC 115057 20287 Corpus Christi, TX 293925 22042 Dallas -Fort Worth -Arlington, TX 4145659 22096 Danbury, CT -NY 154455 22366 Davenport, IA -IL 270626 22528 Dayton, OH 703444 22636 Daytona Beach -Port Orange, FL 255353 23311 Deltona, FL 147713 23500 Denton -Lewisville, TX 299823 23527 Denver -Aurora, CO 1984887 23743 Des Moines, IA 370505 23824 Detroit, MI 3903377 24850 Dulu th, MN -WI 118265 25228 Durham, NC 287796 27253 El Paso, TX -NM 674801 26794 Elkhart, IN -MI 131226 27766 Erie, PA 194804 28117 Eugene, OR 224049 28333 Evansville, IN -KY 211989 28657 Fairfield, CA 112446 29089 Fargo, ND -MN 142477 2944 0 Fayetteville, NC 276368 29494 Fayetteville -Springdale, AR 172585 29872 Flint, MI 365096 30628 Fort Collins, CO 206633 30925 Fort Smith, AR -OK 106470 31060 Fort Walton Beach, FL 152741 31087 Fort Wayne, IN 287759 31519 Frederick, MD 11914 4

PAGE 188

188 31843 Fresno, CA 554923 32167 Gainesville, FL 159508 32653 Gastonia, NC 141407 34300 Grand Rapids, MI 539080 34813 Green Bay, WI 187316 35164 Greensboro, NC 267884 35461 Greenville, SC 302194 35920 Gulfport -Biloxi, MS 205754 36190 Hage rstown, MD -WV -PA 120326 36892 Harlingen, TX 110770 37081 Harrisburg, PA 362782 37243 Hartford, CT 851535 38215 Hemet, CA 117200 38647 Hickory, NC 187808 38809 High Point, NC 132844 40375 Houma, LA 125929 40429 Houston, TX 3822509 40753 Huntington, WV -KY -OH 177550 40780 Huntsville, AL 213253 41212 Indianapolis, IN 1218919 41347 Indio -Cathedral City -Palm Springs, CA 254856 42211 Jackson, MS 292637 42346 Jacksonville, FL 882295 43210 Johnson City, TN 102456 43723 Kalamazo o, MI 187961 43912 Kansas City, MO -KS 1361744 44479 Kennewick -Richland, WA 153851 44506 Kenosha, WI 110942 44992 Killeen, TX 167976 45451 Kissimmee, FL 186667 45640 Knoxville, TN 419830 46018 Lafayette, IN 125738 46045 Lafayette, LA 178 079 46531 Lake Charles, LA 132977 46828 Lakeland, FL 199487 47530 Lancaster, PA 323554 47611 Lancaster -Palmdale, CA 263532 47719 Lansing, MI 300032 47854 Laredo, TX 175586 47935 Las Cruces, NM 104186 47962 Las Vegas, NV 1314357 49096 Leo minster -Fitchburg, MA 112943 49582 Lexington Fayette, KY 250994

PAGE 189

189 49933 Lincoln, NE 226582 50392 Little Rock, AR 360331 51364 Lorain -Elyria, OH 193586 51445 Los Angeles -Long Beach -Santa Ana, CA 11789487 51715 Louisville, KY -IN 863582 518 77 Lubbock, TX 202225 52822 Macon, GA 135170 53200 Madison, WI 329533 53740 Manchester, NH 143549 55333 Marysville, WA 114372 52390 McAllen, TX 523144 55981 Medford, OR 128780 56116 Memphis, TN -MS -AR 972091 56251 Merced, CA 110483 5660 2 Miami, FL 4919036 57466 Milwaukee, WI 1308913 57628 Minneapolis -St. Paul, MN 2388593 57709 Mission Viejo, CA 533015 57925 Mobile, AL 317605 58006 Modesto, CA 310945 58330 Monroe, LA 113818 58600 Montgomery, AL 196892 60733 Murfreesboro, TN 135855 60841 Muskegon, MI 154729 60895 Myrtle Beach, SC 122984 61165 Nashua, NH -MA 197155 61273 Nashville Davidson, TN 749935 61786 New Bedford, MA 146730 62407 New Haven, CT 531314 62677 New Orleans, LA 1009283 63217 New York -Newark NY -NJ -CT 17799861 63838 North Port -Punta Gorda, FL 122421 64135 Norwich -New London, CT 173160 64567 Ocala, FL 106542 64864 Odessa, TX 111395 64945 Ogden -Layton, UT 417933 65080 Oklahoma City, OK 747003 65242 Olympia -Lacey, WA 143826 65269 Omaha, NE -IA 626623 65863 Orlando, FL 1157431 66673 Oxnard, CA 337591 67105 Palm Bay -Melbourne, FL 393289 67294 Panama City, FL 132419

PAGE 190

190 68482 Pensacola, FL -AL 323783 68509 Peoria, IL 247172 69076 Philadelphia, PA -NJ -DE -MD 514907 9 69184 Phoenix -Mesa, AZ 2907049 69697 Pittsburgh, PA 1753136 70993 Port Arthur, TX 114656 71479 Port St. Lucie, FL 270774 71263 Portland, ME 188080 71317 Portland, OR -WA 1583138 71803 Poughkeepsie -Newburgh, NY 351982 72505 Providence, RI -MA 1174548 72559 Provo -Orem, UT 303680 72613 Pueblo, CO 123351 73153 Racine, WI 129545 73261 Raleigh, NC 541527 73693 Reading, PA 240264 73774 Redding, CA 105267 74179 Reno, NV 303689 74746 Richmond, VA 818836 75340 Riverside -San B ernardino, CA 1506816 75421 Roanoke, VA 197442 75664 Rochester, NY 694396 75718 Rockford, IL 270414 76474 Round Lake Beach -McHenry -Grayslake, IL -WI 226848 77068 Sacramento, CA 1393498 77149 Saginaw, MI 140985 78229 Salem, OR 207229 78310 Salinas, CA 179173 78499 Salt Lake City, UT 887650 78580 San Antonio, TX 1327554 78661 San Diego, CA 2674436 78904 San Francisco -Oakland, CA 2995769 79039 San Jose, CA 1538312 79228 San Rafael -Novato, CA 232836 79282 Santa Barbara, CA 196 263 79309 Santa Clarita, CA 170481 79336 Santa Cruz, CA 157348 79417 Santa Maria, CA 120297 79498 Santa Rosa, CA 285408 79606 Sarasota -Bradenton, FL 559229 79768 Savannah, GA 208886 80227 Scranton, PA 385237 80362 Seaside -Monterey -Marin a, CA 125503

PAGE 191

191 80389 Seattle, WA 2712205 81739 Shreveport, LA 275213 82144 Simi Valley, CA 112345 82225 Sioux City, IA -NE -SD 106119 82252 Sioux Falls, SD 124269 83116 South Bend, IN -MI 276498 83332 South Lyon -Howell -Brighton, MI 106139 83548 Spartanburg, SC 145058 83764 Spokane, WA -ID 334858 83899 Springfield, IL 153516 83926 Springfield, MA -CT 573610 83953 Springfield, MO 215004 77770 St. Louis, MO -IL 2077662 85087 Stockton, CA 313392 86302 Syracuse, NY 402267 86464 Tallahassee, FL 204260 86599 Tampa -St. Petersburg, FL 2062339 87004 Temecula -Murrieta, CA 229810 87490 Thousand Oaks, CA 210990 87868 Toledo, OH -MI 503008 88084 Topeka, KS 142411 88462 Trenton, NJ 268472 88732 Tucson, AZ 720425 88948 Tu lsa, OK 558329 89110 Tuscaloosa, AL 116888 89326 Tyler, TX 101494 89785 Utica, NY 113409 90028 Vallejo, CA 158967 90406 Vero Beach -Sebastian, FL 120962 90541 Victorville -Hesperia -Apple Valley, CA 200436 90892 Virginia Beach, VA 1394439 9 0946 Visalia, CA 120044 91027 Waco, TX 153198 92242 Washington, DC -VA -MD 3933920 92485 Waterbury, CT 189026 92593 Waterloo, IA 108298 95077 Wichita, KS 422301 95833 Wilmington, NC 161149 96670 Winston Salem, NC 299290 96697 Winter Haven, FL 153924 97291 Worcester, MA -CT 429882 97507 Yakima, WA 112816 97750 York, PA 192903

PAGE 192

192 97831 Youngstown, OH -PA 417437 indicate urban areas excluded from this study

PAGE 193

193 APPENDIX B URBAN WATER UTILITY COLLABORATION AND MA NAGEMENT SURVEY 1 Genera l Information a. What is the name of your water system? (no abbreviations, please) b. Where is your s ystem located? City: State: c. Using the most recent 12 month record of data available, please ans wer the following: Population Served (number of individuals): What is your: 12 month record used (ex. 06/2008 06/2009) Service area (sq. miles): Average volume of water sold (MGD): Number of employees:

PAGE 194

194 2 Water Supply Management a. Thinking about your CURRENT SYSTEM, please identify the degree to which your system relies on the following management strategies to deal with potential or actual supply shortages/droughts. Do Not Use Primary Strategy Seco ndary Strategy Currently Developing Creation, expansion or improvement of surface water supplies Creation, expans ion or improvement of groundwater supplies (inc. aquifer recharge) Creation or expansion of recycled/reclaimed wat er system Regular infrastructure maintenance Contract to purchase additional water W ater conservation education and outreach Water conservation rebates (retrofit, efficiency, irrigation, etc.) Alternate pricing schemes (increasing block rates, low cost reclaimed supplies, etc.) Water use restrictions Water rationing Other If Other, please specify:

PAGE 195

195 b. Thinking about your SYSTEM FIVE YEARS AGO, please identify the degree to whi ch your system relies on the following management strategies to deal with potential or actual supply shortages/droughts. Did Not Use Primary Strategy Secondary Strategy Was Developing Creation, expansion or improvement of surface water supplies Creation, expansion or improvement of groundwater supplies (inc. aquifer recharge) Creation or expansion of recycled/reclaimed water system Regular infrastructure maintenance Contract to purchase additional water Water conservation education and outreach Water conservation rebates (retrofit, efficiency, irrigation, etc.) Alternate pricing schemes (increasing block rates, low cost reclaimed supplies, etc.) Water use restrictions Water rationing Other If Other, please specify:

PAGE 196

196 c. Does your system currently have a water management plan? No (Skip to Section 3) Yes it is mandatory. An external a gency requires it. Yes it is voluntary. It is not required, it was our choice. d. To what degree are each of the following management components addressed in this plan? I f Other, please specify: e. How many years into the future does your plan extend? Not Addressed Minimally Addressed Somewhat Add ressed Thoroughly Addressed Finances Source Water Quality Source Water Availability Fu ture Customer Demand Infrastructure Security Other*

PAGE 197

197 3 Level of External Participation a. Within the past 12 m onths, how engaged was your water system with the following nationally based professional trade organizations? Examples of engagement: Minimally paid dues, subscribe to newsletter, etc. Moderately attended a conference/workshop, buy literature, e tc. Actively participate in surveys, serve on a board, edit or author articles Not a Member MINIMALLY ENGAGED MODERATELY ENGAGED ACTIVELY ENGAGED American Water Works Association (AWWA) Association of Metropolitan Water Agencies (AMWA) American Public Works Association (APWA) Association of Water Companies (NAWC) Water Environment Federation (WEF) Water Utility Benchmarking Association Other* If Other, please specify:

PAGE 198

198 b. Within the PAST TWO YEARS, have any of your system's employees attended or presented in any professional trade organization or academic events (conferences, meetings, seminars, etc.) about managing water availability and/or supply. No, no such events were attended. (Skip to 3d) Yes, our system attended or presented at these types of events I don't know (Skip to 3d) c. For the events (conferences, meetings, seminars, et c) mentioned above, please indicate the following: Total number of events Total number of employees Attended only Attended and presented d. Within the past five years, has your system participated in any individual, local, state, regiona l, or national water utility performance studies, reports or surveys (excluding this one)? No, our system did not participate in any of these types of projects (Skip to Section 4) Yes, our system did participate in projects like these. I don't know (Skip to Section 4) e. Please select answer which best completes this statement. "The results from the studies, reports or surveys our system participated in are MAINLY..." private, and available to interested parties at the discretion of the participants."

PAGE 199

199 4 Role of Outside Organizations a. Please indicate how important each group CURRENTLY IS to your system when consideri ng issues related to your source water supply operations and planning. No Interaction Minimally Important Moderately Important Very Important Federal Government Agencies State Government Agencies Local Government Agencies Professional Trade Organizations Academic Researchers Private Consultants System Customers Other Water Utilities State based Associations (ex. Texas Water Utilities Associati on) Water Industry Associations

PAGE 200

200 b. Please indicate how important each group was FIVE YEARS AGO to your system when considering issues related to your source water supply operations and planning. No Interaction Minimally Impor tant Moderately Important Very Important Federal Government Agencies State Government Agencies Local Government Agencies Professional Trade Organizations Academic Researchers Private Consultants System Customers Other Water Utilities State based Associations (ex. Texas Water Utilities Association) Water Industry Associations 5 Water System Collaborations a. Is y our system a member of any local or regional board, agency, compact or other similar type of group that manages, makes decisions, or controls about any or all of your current water supply source(s)? No, we do not parti cipate in these types of groups (Skip to the end) Yes, we participate in groups like these.

PAGE 201

201 b. Please list the name of each group and how long you have been a member. If you are involved in more than five groups, plea se list those that are most important to your system. Group 1: Group 2: Group 3: Group 4: Group 5: c. Please select al l other members that participate in each of the groups listed: Group 1 Group 2 Group 3 Group 4 Group 5 General Public Professional Trade Organizations Local Government State Government Federal Government Private Consultants University/Academics Other Water Utilities Non Governmen tal Organizations State Water Associations Other If Other, please specify:

PAGE 202

202 d. For each group listed, please indicate how involved YOUR WATER SYSTEM is: Group 1 Group 2 Group 3 Group 4 Group 5 Your system is informed of decisions and processes when the group is eva luating decisions, but does not provide any input Your system provides informat ion regarding preferences, goals, and/or perceptions when the group is evaluating decisions Your system identifies issues, sets decision agendas, and/or generates and evaluates alternatives when the group is evaluating decisions Your system is actively involved in collectively making substantive decisions with the group If desired, please use this space to add any additional comments about the survey in the space provided below:

PAGE 203

203 LIST OF REFERENCES Abbott, M. and B. Cohen, 2009. Productivity and Efficiency in the Water Industry. Utilities Policy 17:233 244. Agrawal, A., 2001. Common Property Institutions and Sustainable Governance of Resources. World Development 29:1649 1672. Agrawal, A., D. Kapur, and J. McHale, 2008. How Do Spatial and Social Proximity Influence Knowledge Flows? Evidence from Patent Data. Journal of Urban Economics 64:258 269. Aichinger, C., 2009. Institutional Structures for Water Management in the Eas tern United States. The Water Environment of Cities:217 234. Alcamo, J., P. Doll, T. Henrichs, F. Kaspar, B. Lehner, T. Rosch, and S. Siebert, 2003. Global Estimates of Water Withdrawals and Availability Under Current and as ons. Hydrological Sciences Journal Journal Des Sciences Hydrologiques 48:339 348. Alig, R.J., J.D. Kline, and M. Lichtenstein, 2004. Urbanization on the US Landscape: Looking Ahead in the 21st Century. Landscape and Urban Planning 69:219 234. Allen, T.F.., J.A. Tainter, and T.W. Hoekstra, 1999. Supply side Sustainability. Systems Research and Behavioral Science 16:403 427. Amaral, L.A.N., S.V. Buldyrev, S. Havlin, M.A. Salinger, and H.E. Stanley, 1998. Power Law Scaling for a System of Interacting Units wit h Complex Internal Structure. Physical Review Letters 80:1385 1388. Anderson, M.H., 2008. Social Networks and the Cognitive Motivation to Realize Journal of Organizational Behavio r 29:51 78. Ausubel, J., 1988. Cities and Their Vital Systems: Infrastructure Past, Present, and Future. National Academies Press. AWWA, 2007. Benchmarking Performance Indicators for Water and Wastewater Utilities: 2007 Annual Survey Data and Analyses Repo rt. American Water Works Association. Baldridge, J.V. and R.A. Burnham, 1975. Organizational Innovation: Individual, Organizational, and Environmental Impacts. Administrative Science Quarterly 20:165 176. Baruch, Y. and B.C. Holtom, 2008. Survey Response R ate Levels and Trends in Organizational Research. Human Relations 61:1139.

PAGE 204

204 Batty, M., 2003. The Emergence of Cities: Complexity and Urban Dynamics. Working Paper Series, Center for Advanced Spatial Analysis, London. Batty, M., 2008. The Size, Scale, and Sh ape of Cities. Science 319:769 771. Baumann, D.., J.J. Boland, and M.W. Hanemann, 1998. Urban Water Demand Management and Planning. McGraw Hill Company Inc. Baynes, T.M., 2009. Complexity in Urban Development and Management. Journal of Industrial Ecology 13:214 227. Bellamy, J.A., G.T. McDonald, G.J. Syme, and J.E. Butterworth, 1999. Evaluating Integrated Resource Management. Society & Natural Resources 12:337 353. Berg, S.V., 2002. U.S. Water and Wastewater: Are There Lessons for Developing Countries? Uni versity of Florida, Gainesville, Florida. Bettencourt, L., J. Lobo, and D. Strumsky, 2007. Invention in the City: Increasing Returns to Patenting as a Scaling Function of Metropolitan Size. Research Policy 36:107 120. Bettencourt L., J. Lobo, D. Helbing, C. K hnert, and G.B. West, 2007. Growth, Innovation, Scaling, and the Pace of Life in Cities. Proceedings of the National Academy of Sciences 104:7301. Bettencourt, L.M.A., J. Lobo, D. Strumsky, and G.B. West, 2010. Urban Scaling and Its Deviations: Reveal ing the Structure of Wealth, Innovation and Crime Across Cities. PloS One 5:e13541. Bhattacharyya, A., E. Parker, and K. Raffiee, 1994. An Examination of the Effect of Ownership on the Relative Efficiency of Public and Private Water Utilities. Land Economi cs:197 209. Biswas, A.K., 2003. Water Resources of North America. Springer, New York. Biswas, A.K., 2004. Integrated Water Resources Management: a Reassessment. Water International 29:248 256. Blake, N.E., 1956. Water For the Cities A History of the Urban Water Supply Problem in the United States. Syracuse University Press, Syracuse, New York. Blomquist, W.A., E. Schlager, and T. Heikkila, 2004. Common Waters, Diverging Streams: Linking Institutions to Water Management in Arizona, California, and Colorado. Rff Press. Boccaletti, S., V. Latora, Y. Moreno, M. Chavez, and D. U. Hwang, 2006. Complex Networks: Structure and Dynamics. Physics Reports 424:175 308.

PAGE 205

205 BOR, 1972. Repayment of Reclamation Projects. Bureau of Reclamation. United States Government Printin g Office, Washington D.C. Brikowski, T.H., 2008. Doomed Reservoirs in Kansas, USA? Climate Change and Groundwater Mining on the Great Plains Lead to Unsustainable Surface Water Storage. Journal of Hydrology 354:90 101. Carrera, L., 2010. Integrated Approac hes for the Study of Water Vulnerability of Major African Urban Agglomerations in a Global Change Context. Universita di Bologna and University of Florida, Bologna, Italy. Chen, Y. and Y. Zhou, 2008. Scaling Laws and Indications of Self organized Criticali ty in Urban Systems. Chaos, Solutions & Fractals 35:85 98. City of St. Petersburg, 2011. City of St. Petersburg Water Treatment and Supply. City of Tampa, 2011. Tampa Water Department History Conca, K., 2006. Governing Water Contentious Transnational Politics and Global Institution Building. The MIT Press, Cambridge. Cordery, I., R. Mehrotra, and M.J. Nazemosadat, 2006 How Reliable Are Standard Indicators of Stationarity? Stochastic Environmental Research and Risk Assessment 21:765 771. Cortner, H.J. and M.A. Moote, 1994. Trends and Issues in Land and Water Resources Management Setting the Agenda for Change. Environm ental Management 18:167 173. Cowie, G.M. and S.R. Borrett, 2005. Institutional Perspectives on Participation and Information in Water Management. Environmental Modelling & Software 20:469 483. Crane, P. and A. Kinzig, 2005. Nature in the Metropolis. Scienc e 308:1225. Cromwell, J.E., J.B. Smith, and R.S. Raucher, 2007. Implications of Climate Change for Urban Water Utilities. Association of Metropolitan Water Agencies. Cross, R. and L. Sproull, 2004. More Than an Answer: Information Relationships for Actiona ble Knowledge. Organization Science 15:446 462. Cycyota, C.S. and D.A. Harrison, 2006. What (not) to Expect When Surveying Executives. Organizational Research Methods 9:133. Davis, R.K., 1968. The Range of Choice in Water Management A Study of Dissolved O xygen in the Potomac Estuary. Johns Hopkins Press, Baltimore.

PAGE 206

206 Dillman, D.A., 2007. Mail and Internet Surveys The Tailored Design Method. John Wiley & Sons, Inc., New Jersey. Dooley, K.J., 1997. A Complex Adaptive Systems Model ofOrganization Change. Nonli near Dynamics, Psychology, and Life Sciences 1:1 29. Dyer, J.H. and H. Singh, 1998. The Relational View: Cooperative Strategy and Sources of Interorganizational Competitive Advantage. Academy of Management Review 23:660 679. Dynesius, M. and C. Nilsson, 19 94. Fragmentation and Flow Regulation of River Systems in the Northern Third of the World. Science 266:753 762. Elbakidze, L., 2006. Potential Economic Impacts of Changes in Water Availability on Agriculture in the Truckee and Carson River Basins, Nevada, USA. Journal of the American Water Resources Association 42:841 849. EPA, 2002. 2000 EPA Community Water System Survey. Environmental Protection Agency, Washington D.C. EPA, 2008. Investing in a Sustainable Future 2007 Annu al Report. Office of Water, U.S. Environmental Protection Agency, Washington D.C. Falkenmark, M., 1998. Dilemma When Entering 21st Century rapid Change but Lack of Sense of Urgency. Wate r Policy 1:421 436. Falkenmark, M., 2004. Towards Integrated Catchment Management: Opening the Paradigm Locks Between Hydrology, Ecology and Policy making. International Journal of Water Resources Development 20:275 281. Farber, D.A., 2003. Probabilities B ehaving Badly: Complexity Theory and Environmental Uncertainty. U.C. Davis Law Review 37:145. Farmani, R., G. Walters, and D. Savic, 2006. Evolutionary Multi objective Optimization of the Design and Operation of Water Distribution Network: Total Cost Vs. R eliability Vs. Water Quality. Journal of Hydroinformatics 8:165 179. (Editor). Water Right of the Fifty States and Territories. American Water Works Association, Denver, CO p. Change and Variability on the Reliability, Resilience, and Vulnerability of a Water Resource System. Water Resources Research 39. Galloway, G.E., Goals, Institutio ns and Governance: The US Experience. :499 511.

PAGE 207

207 Geldof, G.D., 1995a. Policy Analysis and Complexity a Non equilibrium Approach for Integrated Water Management. Water Science and Technology 31:301 309. Geldof, G.D., 1995b. Adaptive Water Management: Integ rated Water Management on the Edge of Chaos. Water Science and Technology 32:7 13. Adaptive Management Along U.S. Rivers. Society and Natural Resources 21:538 545. Gerlak, A.K. and T. Heikkila, 2007. Collaboration and Institutional Endurance in US Water Policy. Ps Political Science & Politics 40:55 60. Ginley, J. and S. Ralston, 2010. A Conversation with Water Utility Managers. Journal American Water Works Association 102:11 7 122. Glaeser, E.L., 1994. Cities, Information and Economic Growth. Harvard Institute of Economic Research, Harvard University. Gleick, P.H., 1998. Water in Crisis: Paths to Sustainable Water Use. Ecological Applications 8:571 579. Gleick, P.H., 2000. The Changing Water Paradigm A Look at Twenty first Century Water Resources Development. Water International 25:127 138. Gleick, P.H., 2001. Global Water Threats and Challenges Facing the United States. Environment: Science and Policy for Sustainable Develop ment 43:18 26. Gleick, P.H., 2003. Global Freshwater Resources: Soft path Solutions for the 21st Century. Science 302:1524 1528. Graedel, T.E. and R.J. Klee, 2002. Getting Serious About Sustainability. Environmental Science & Technology 36:523 529. Graf, W ., 1999. Dam Nation: A Geographic Census of American Dams and Their Large Scale Hydrologic Impacts. Water Resources Research 35:1305 1311. Grimm, N.B., S.H. Faeth, N.E. Golubiewski, C.L. Redman, J. Wu, X. Bai, and J.M. Briggs, 2008. Global Change and the E cology of Cities. Science 319:756. Gupta, S.K., 2010. Well Hydraulics & Test Pumping. Modern Hydrology and Sustainable Water Development. John Wiley and Sons, p. 99. Hackett, S., E. Schlager, and J. Walker, 1994. The Role of Communication in Resolving Comm on Dilemmas Experimental Evidence with Heterogeneous Appropriators. Journal of Environmental Economics and Management 27:99 126.

PAGE 208

208 Hagen, E.R., K.J. Holmes, J.E. Kiang, and R.C. Steiner, 2005. Benefits of Iterative Water Supply Forecasting in the Washingto n, D.c., Metropolitan Area1. JAWRA Journal of the American Water Resources Association 41:1417 1430. Harris, C.D. and E.L. Ullman, 1945. The Nature of Cities. The Annals of the American Academy of Political and Social Science 242:7. Hashimoto, T., J.R. Ste dinger, and D.P. Loucks, 1982. Reliability, Resiliency, and Vulnerability Criteria for Water Resource System Performance Evaluation. Water Resources Research 18:14 20. Heikkila, T., 2004. Institutional Boundaries and Common pool Resource Management: A Comp arative Analysis of Water Management Programs in California. Journal of Policy Analysis and Management 23:97 117. Holland, J.H., 1992. Complex Adaptive Systems. Daedalus 121:17 30. Holling, C.S., 2001. Understanding the Complexity of Economic, Ecological, and Social Systems. Ecosystems 4:390 405. Houtsma, J. and N. Sackville, 2003. Water Supply in California: Economies of Scale, Water Charges, Efficiency, and Privatization. ERSA 2003 Congress, August. Hurd, B.H., M. Callaway, J. Smith, and P. Kirshen, 2004. Climatic Change and US Water Resources: From Modeled Watershed Impacts to National Estimates. Journal of the American Water Resources Association 40:129 148. Hurd, B., N. Leary, R. Jones, and J. Smith, 1999. Relative Regional Vulnerability of Water Resour ces to Climate Change. Journal of the American Water Resources Association 35:1399 1409. Huston, S., N.L. Barber, J.F. Kenny, K.S. Linsey, D.L. Lumia, and M.A. Maupin, 2005. Estimated Use of Water in the United States in 2000. United States Geological Surv ey. Jacobs, J., 1961. The Death and Life of Great American Cities: The Failure of Town Planning. Penguin books. Jenerette, G.D. and L. Larsen, 2006. A Global Perspective on Changing Sustainable Urban Water Supplies Global and Planetary Change 50:202 211. Karl, T.R. and W.J. Koss, 1984. Regional and National Monthly, Seasonal, and Annual Temperature Weighted by Area, 1895 1983. National Climatic Data Center, Asheville, NC. Kates, R.W. and T.M. Parris, 2003. Long ter m Trends and a Sustainability Transition. Proceedings of the National Academy of Sciences 100:8062.

PAGE 209

209 25:617 623. Kenny, J.F., N.L. Barber, S.S. Hutson, K.S. Linsey, J.K. Lovelace, and M.A. Maupin, 2009. Estimated Use of Water in the United States in 2005. U.S. Geological Survey Circular, United States Geological Survey. Kjeldsen, T.R. and D. Rosbjerg, 2004. Choice of Reliability, Resilience and Vulnerability Estimators fo r Risk Assessments of Water Resources Systems. Hydrological Sciences Journal Journal Des Sciences Hydrologiques 49:755 767. Konrad, C., 2010. A National Assessment of Water Availability and Use. Journal of the American Water Worrks Association:30 34. Kuhne rt, C., D. Helbing, and G.B. West, 2006. Scaling Laws in Urban Supply Networks. Physica A: Statistical Mechanics and Its Applications 363:96 103. LADWP, 2011. History of Water and Power in LA. Lmmer, S., B. Gehlsen, and D. Helbing, 2006. Scaling Laws in the Spatial Structure of Urban Road Networks. Physica A: Statistical Mechanics and Its Applications 363:89 95. Lee, T.R., 1999. Water Management in the 21st Century: The Allocation Imperative. Edward Elgar Pub lishing Inc., Cheltenham, UK. Levin, R.B., P.R. Epstein, T.E. Ford, H. Winston, E. Olson, and E.G. Reichard, 2002. U.S. Drinking Water Challenges in the Twenty First Century. Environmental Health Perspectives 110:43 52. Loucks, D.P., Task Comittee on Susta inable Criteria, and Working Group of UNESCO/IHP IV Project M 4.3, 1998. Sustainability Criteria for Water Resource Systems. ASCE Publications, Baltimore. Lucas, R.E., 1988. On the Mechanics of Economic Development. Journal of Monetary Economics 22:3 42. M aier, H.R., B.J. Lence, B.A. Tolson, and R.O. Foschi, 2001. First order Reliability Method for Estimating Reliability, Vulnerability, and Resilience. Water Resources Research 37:779 790. Manson, S.M., 2001. Simplifying Complexity: a Review of Complexity Th eory. Geoforum 32:405 414. McDonald, R.I., P. Green, D. Balk, B.M. Fekete, C. Revenga, M. Todd, and M. Montgomery, 2011. Urban Growth, Climate Change, and Freshwater Availability. Proceedings of the National Academy of Sciences 108:6312.

PAGE 210

210 McMichael, A.J., C .D. Butler, and C. Folke, 2003. New Visions for Addressing Sustainability. Science 302:1919 1920. Means, E.G., L. Ospina, and R. Patrick, 2005. Ten Primary Trends and Their Implications for Water Utilities. Journal American Water Works Association 97. Mea ns, E.G., N. West, and R. Patrick, 2005. Population Growth and Climate Change Will Pose Tough Challenges for Water Utilities. Journal American Water Works Association 97. Melosi, M.V., 2000. Pure and Plentiful: The Development of Modern Waterworks in the U nited States, 1801 2000. Water Policy 2:243 265. Miller, W.G., 2006. Integrated Concepts in Water Reuse: Managing Global Water Needs. Desalination 187:65 75. Milly, P.C.D., J. Betancourt, M. Falkenmark, R.M. Hirsch, Z.W. Kundzewicz, D.P. Lettenmaier, and R .J. Stouffer, 2008. Stationarity Is Dead: Whither Water Management? Science 319:573 574. Mitchell, B., 2005. Integrated Water Resource Management, Institutional Arrangements, and Land use Planning. Environment and Planning A 37:1335 1352. MWD, 2011. At a Glance The Metropolitan Water District of Southern California Facts and Figures. Metropolitan Water District of Southern California, Los Angeles. NHD, 2008. U.S. Geological Survey National Hy drography Dataset. Accessed 7 Aug 2011. NID, 2009. National Inventory of Dams. Accessed 7 Aug 2011. Norton Jr, J.W. and W.J. Weber Jr, 2009. Water Utility Efficiency Asse ssment Using a Data Envelopment Analysis Procedure. Journal of Infrastructure Systems 15:80. NRC, 1999. Our Common Journey: A Transition Toward Sustainability. The National Academies Press, Washington, D.C. NRC, 2002. Estimating Water Use in the United Sta tes A New Paradigm for the National Water Use Information Program. National Academy Press, Washington D.C. NRDC, 2010. Thirsty for Answers: Preparing for the Water related Impacts of Climate Change in American Cities. nswers.asp. Ostrom, E., 1999. Coping with Tragedies of the Commons. Annual Review of Political Science 2:493 535.

PAGE 211

211 Postel, S.L., G.C. Daily, and P.R. Ehrlich, 1996. Human Appropriation of Renewable Fresh Water. Science 271:785 788. Pumain, D., J. Lobo, C. V acchiani Marcuzzo, and F. Paulus, 2006. An Evolutionary Theory for Interpreting Urban Scaling Laws. Cybergeo: European Journal of Geography. Qualtrics, 2011. Qualtrics Online Survey Software. Qualtrics. Rammel, C., S. Stagl, and A Co evolutionary Perspective on Natural Resource Management. Ecological Economics 63:9 21. Reilly, T., K. Dennehy, W. Alley, and W. Cunningham, 2008. Ground water Availability in the United States. U .S. Geological Survey Circular 1323. Reisner, M., 1993. Cadillac Desert: The American West and Its Disappearing Water. Penguin Books, New York, NY. Reitsma, R.F., 1996. Structure and Support of Water resources Management and Decision making. Journal of Hyd rology 177:253 268. Richmond, J.E.D., 1998. Simplicity and Complexity in Design for Transportation Systems and Urban Forms. Journal of Planning Education and Research 17:220 230. Rijsberman, F.R., 2006. Water Scarcity: Fact or Fiction? Agricultural Water M anagement 80:5 22. Rose, S. and N.E. Peters, 2001. Effects of Urbanization on Streamflow in the Atlanta Area (Georgia, USA): a Comparative Hydrological Approach. Hydrological Processes 15:1441 1457. Rosenberg, N.J., D.J. Epstein, D. Wang, L. Vail, R. Srini vasan, and J.G. Arnold, 1999. Possible Impacts of Global Warming on the Hydrology of the Ogallala Aquifer Region. Climatic Change 42:677 692. Roy, S., P. Ricci, K. Summers, C. Chung, and R. Goldstein, 2005. Evaluation of the Sustainability of Water Withdra wals in the United States, 1995 to 2025. Journal of the American Water Resources Association 41:1091 1108. Ruhl, J.B. and H.J. Ruhl, 1997. The Arrow of the Law in Modern Administrative States: Using Complexity Theory to Reveal the Diminishing Returns and I ncreasing Risks the Burgeoning of Law Poses to Society. UC Davis Law Review 30:405 482. Runge, J. and J. Mann, 2008. State of the Industry Report 2008 Charting the Course Ahead. Journal of the American Water Works Association 100:61 74.

PAGE 212

212 Saleth, R.M. and A Dinar, 2004. The Institutional Economics of Water. MPG Books Ltd., Cheltenham. Sarang, A., A. Vahedi, and A. Shamsai, 2008. How to Quantify Sustainable Development: A Risk based Approach to Water Quality Management. Environmental Management 41:200 220. S aunders, J.F. and W.M. Lewis, 2003. Implications of Climatic Variability for Regulatory Low Flows in the South Platte River Basin, Colorado. Journal of the American Water Resources Association 39:33 45. Scholz, J.T. and B. Stiftel, 2005. Adaptive Governanc e and Water Conflict: New Institutions for Collaborative Planning. Resources for the Future. Serageldin, I., 1995. Toward Sustainable Management of Water Resources. World Bank Report #14910, Washington, D.C. Seto, K.C., R. Snchez Rodrguez, and M. Fragki as, 2010. The New Geography of Contemporary Urbanization and the Environment. Annual Review of Environment and Resources 35:167 194. Shih, J.S., W. Harrington, W.A. Pizer, and K. Gillingham, 2006. Economies of Scale in Community Water Systems. Journal Amer ican Water Works Association 98:100 108. Singhal, V.C., 2010. Washington Aqueduct Financial Report FY 2010. U.S. Army Corps of Engineers, Washington D.C. Slocombe, D.S., 1993. Implem enting Ecosystem Based Management. BioScience 43:612 622. Smakhtin, V., C. Revenga, and P. D \ ll, 2004. A Pilot Global Assessment of Environmental Water Requirements and Scarcity. Water International 29:307 317. Stakhiv, E.Z., 2003. Disintegrated Water Res ources Management in the US: Union of Sisyphus and Pandora. Journal of Water Resources Planning and Management 129:151. State of California, 2011. California State Water Project. Sullivan, C.A., J.R. Meigh, A.M. Giaco mello, T. Fediw, P. Lawrence, M. Samad, S. Mlote, C. Hutton, J.A. Allan, R.E. Schulze, and others, 2003. The Water Poverty Index: Development and Application at the Community Scale. Natural Resources Forum., pp. 189 199.

PAGE 213

213 Sun, G., S.G. McNulty, J.A.M. Myers and E.C. Cohen, 2008. Impacts of Multiple Stresses on Water Demand and Supply Across the Southeastern United States. Journal of the American Water Resources Association 44:1441 1457. SWP, 2011. California State Water Project Overview. California State Wa ter Project Overview. Tainter, J.A., 1988. The Collapse of Complex Societies. Cambridge University Press. Tainter, J.A., 2006. Social Complexity and Sustainability. Ecological Complexity 3:91 103. Taylor, R., 2009. Rethinking Water Scarcity: The Role of Storage. Eos 90:237 248. Tomaskovic Devey, D., J. Leiter, and S. Thompson, 1994. Organizational Survey Nonresponse. Administrative Science Quarterly:439 457. Tynan, N. and R. Kingdom, 2005. Optimal Size for Utilities? http://www al_Size_for_Utilities.pdf. Accessed 19 Oct 2011. U.S. BEA, 2011. Economic Growth Widespread Across Metropolitan Areas in 2010. Bureau of Economic Analysis, U.S. Department of Commerce. h ttp:// on_0211b.pdf. US Census Bureau, 1995. Urban and Rural Population: 1900 to 1990. Accessed 6 Aug 2011. US Census Bureau, 20 00a. U.S. Urban/Rural and Metropolitan/Nonmetropolitan Population: 2000 (GCT P1). geo_id=01000US& _box_head_nbr=GCT P1& ds_name=DEC_2000_SF1_U& format=US 1. Accessed 7 Aug 2011. US Census Bureau, 2000b. Census 2000 Urbanized Areas Cartographic Boundary Files U.S. Census Bureau. Accessed 6 Aug 2011. USDA Economic Research Service, 2002. Major Uses of Land in the United States, 2002. ublications/EIB14/. Accessed 17 Sep 2011. USGS, 2003a. Ground water Depletion Across the Nation. Fact Sheet 103 03. United States Geological Survey. 103 03/#pdf. USGS, 2003b. Estimated Mean Annual Natural Ground water Recharge in the Conterminous United States. USGS Water Resources NSDI Node.

PAGE 214

214 Accessed 7 Aug 2011. USGS, N., 2008a. USGS Water Resources NSDI Node. tesdd.xml. Accessed 7 Aug 2011. USGS, 2008b. Principal Aquifers of the 48 Conterminous United States, Hawaii, Puerto Rico, and the U.S. Virgin Islands. Accessed 7 Aug 2011. Viessman, W. and T.D. Feather (Edito rs)., 2006. State Water Resources Planning in the United States. American Society of Civil Engineers, Reston, VA. Vogel, R.M., M. Lane, R.S. Ravindiran, and P. Kirshen, 1999. Storage Reservoir Behavior in the United States. Journal of Water Resources Plann ing and Management Asce 125:245 254. Vogel, R.M., M. Lane, R.S. Ravindiran, and P. Kirshen, 1999. Storage Reservoir Behavior in the United States. Journal of Water Resources Planning and Management 125:245 254. Vogel, R.M., I. Wilson, and C. Daly, 1999. Re gional Regression Models of Annual Streamflow for the United States. Journal of Irrigation and Drainage Engineering 125:148 157. Vrsmarty, C.J., E.M. Douglas, P.A. Green, and C. Revenga, 2005. Geospatial Indicators of Emerging Water Stress: An Applicatio n to Africa. Ambio:230 236. Ways, H.C., 1993. The Washington Aqueduct 1852 1992. Washington Aqueduct Division, US Army Corps of Engineers, Washington D.C. White, R. and G. Engelen, 1994. Urban Systems Dynamics and Cellular Automata: Fractal Structures Betw een Order and Chaos. Chaos, Solitons & Fractals 4:563 583. van Wijk, R., J.J.P. Jansen, and M.A. Lyles, 2008. Inter and Intra organizational Knowledge Transfer: A Meta analytic Review and Assessment of Its Antecedents and Consequences. Journal of Manageme nt Studies 45:830 853. 2000. U.S. Government Printing Office, Washington, D.C. Xie, L.L. and P. Kumar, 2004. A Network Information Theory for Wireless Communication: Scaling Laws and Optimal Operation. Informati on Theory, IEEE Transactions On 50:748 767.

PAGE 215

215 Zipf, G.K., 1949. Human Behavior and the Principle of Least Effort. Addison Wesley Press, Cambridge, Mass.

PAGE 216

216 BIOGRAPHICAL SKETCH Julie Padowski was born in North Tonawanda, New York in 1981. She received a Bach lor of Science (B.S.) degree in e nvironmental s cience in 2003 from the University of Rochester and a Master of Science (M.S.) degree in s oil and w ater s cience in 2005 from the University of Florida. Upon being accepted to the UF AMW3 IGERT program in 2006, she decided to remain in Florida to pursue a Ph.D. degree with an interdisciplinary focus. She has learned much from this experience, and hopes all t he lessons learned will pay off in the future.