November 2015

UWIN Management Plan

Urban Water Innovation Network: Transitioning Toward Sustainable Urban Water Systems

i

Table of Contents

I. General Information ............................................................................................................ 1

II. Milestones, performance metrics and evaluation plan .................................................................... 3

III. Goals, strategies, plans, and metrics for engaging the science community in the SRN activities .................. 4

IV. Goals, strategies, plans, and metrics for broadening participation of underrepresented groups ................... 6

V. UWIN Organization ............................................................................................................ 8

VI. Processes to select and integrate research projects with one another and with other SRN activities ............. 10

VII. Processes to allocate funds and equipment across SRN activities and among partners ............................. 13

VIII. Data Management Plan ........................................................................................................ 13

IX. High level risk management plan ............................................................................................ 16

X. Succession plans for the SRN leadership team if it became necessary ................................................. 19

XI. External Advisory Committee (EAC) ...................................................................................... 19

Appendix A: Quality Management Plan Checklist ................................................................................ 21

Appendix B: Research Project Plans ................................................................................................ 22

Appendix C: Research Experiences for Undergraduates (REU) Plan ....................................................... 139

Appendix D: Engagement Plan .................................................................................................... 144

Appendix E: Training Plan .......................................................................................................... 154

Appendix F. UWIN Risk Management Matrix (Risk Register) ............................................................... 162

1

SRN Management Plan (SMP) Urban Water Innovation Network (UWIN)

I. General Information

The vision of UWIN is to create an enduring research network for integrated water systems, and champions of innovation for water sensitive urban design and resilient cities. To implement this vision, our mission will be to: (i) develop technologic and sociopolitical solutions; (ii) build social capital and trusted regional leadership in the regions; and (ii) train scientists and policy makers as champions of innovation for urban sustainability. TheUWIN management plan is a streamlined approach that provides the organizational structure necessary for participative leadership and collaboration across the sciences, while ensuring the highest level of creativity and autonomy possible for network members. The approach supports an agile network that continuously learns and adapts to create new opportunities for teaming, while supporting the individual researchers who make up that network. The UWIN management fosters creating and maintaining highly collaborative research teams by streamlining the process of integrating and synthesizing scientific information. Our goals for integration of research, education and engagement activities are: (1) expanded linkages within the network; (2) transparency; (3) continuous communication; (4) continuous learning; and (5) advancing the understanding of managing complex systems. The management structure of UWIN fully supports the needs of a complex network structure.

Network Management Approaches

The UWIN SRN has an ambitious agenda and includes diverse programs, which are needed to address the sustainability of urban water systems, one of the most important issues of the 21st century. The complex nature of the network is rooted in the need to reconcile interdisciplinary expertise from institutions across the network to fulfill the vision and mission of UWIN. The expertise of each group is unique and essential to achieving our mission. Management of complex scientific networks is a growing field of research, with the recognition that high performance requires smooth coordination. A complex scientific endeavor of this nature and magnitude that supports the integration of a network of researchers and other stakeholders requires management strategies, frameworks, and tools beyond conventional approaches. Managing this work requires skills and knowledge in facilitation, dialogue, conflict resolution, and collaboration on cross-functional teams, and an understanding of complex adaptive systems that exhibit self-organization, complexity, emergence, interdependence, co-evolution, chaos, and self-similarity. Interactions among the researchers comprising the network are complicated by their physical separation, by their wide-ranging disciplines with different norms and terminologies, and other commitments each researcher have. All of these can lead to communication challenges that impede the desired level of collaboration. Personnel transition considerations and related institutional knowledge transference are additional challenges. UWIN utilizes adaptive and inclusive processes and systems to manage the complexity of the research network, including:

Coordinated leadership to increase efficiency and accountability.

Shared ownership to establish trusted regional leadership and promote champions of innovation that link UWIN research with stakeholders.

Integrative and inclusive management to learn from various member institutions in setting the vision, priorities, and approaches in the network; and

2

Agile and adaptive management to ensure that the network can respond to the stakeholder needs. The management of UWIN encompasses two components. The first component is managing the functions of the network and ensuring that UWIN is operated efficiently. The organizational structure of UWIN ensures that roles and responsibilities are clearly designated, desired outcomes and products of the network are met, and financial resources are allocated properly. Additionally, the network will be managed in an inclusive and adaptive manner. The local expertise that UWIN members provide and their existing relationships with stakeholders in the 6 study regions are paramount to the success of establishing trusted regional urban water hubs. UWIN regional leaders are enabled to shape the vision and approaches of the network. This is essential to achieve the desired outcome of the network to develop regional leadership and trusted champions of innovation.

The Organizing Concept: Urban Water Sustainability Analysis Framework

All UWIN activities are conducted according to a conceptual Urban Water Sustainability Analysis Framework (Fig. 1) adopted from the DIPSA causal framework for describing the interactions between society and the environment. This framework is built around the following key components: reducing pressures; building resilience; increasing transition capacity to adapt and integrate; and identifying co-benefits for linked systems. The UWIN conceptual framework guides the development of UWIN research, education and engagement plans, selection of research projects, prioritization of activities, allocation of resources, and development of milestones and performance metrics. Urban water systems are under ever-increasing pressures from climate change, population growth and changes in demographics and land use, water quality degradation, and aging infrastructure. At the same time, the resilience of water systems has been diminishing due to loss of non-renewable resources, critical infrastructure and the natural capacity to recover from extreme events. To alter this trajectory toward a more desired state, transitions are needed. These transitions are most effective when they foster integration across water sectors, taking advantage of the co-benefits and heeding the tradeoffs. The various components of the framework include:

Driving Forces: Baseline and alternative population growth, demographic, land use change, economic development, climate change, and extreme events Pressures: Drought, floods, water quality degradation, loss of natural resources, loss of critical water infrastructure, policy and financing constraints, consumer behavior constraints Assessment Indicators: A set of common indicators that are defined, characterized, and quantified for essential characteristics of urban water and linked systems, allowing classification and comparison of these systems in cities across the U.S. and globally Solutions: Technological, infrastructure and management solutions that facilitate the transition toward integration of water systems, maximizing resource recovery and reuse, and incorporating water-sensitive urban planning, including: green infrastructure and LID, water conservation, source separation, water recycling and reuse, energy management, resource (i.e., nutrient) recovery, sustainable urban drainage networks, socioeconomic, and policy/management

UWIN research projects address at last one or more these components.

Figure 1. The Conceptual Urban Water Sustainability Analysis Framework

3

II. Milestones, performance metrics and evaluation plan

To ensure efficiency and accountability in using resources, a rigorous quality management plan is used. The plan guides annual evaluation of various activities of the network objectively, which could influence allocation of funds. Components of the plan include:

Project Plans: Each project will produce a written initial research plan that is vetted by leadership and advisors for applicability to the mission, efficiency in execution, validity of methods, inter-project coordination, data sharing, and dissemination. Bi-Annual Progress Reports: Project progress reports are due April 1 and October 1 of 2016 and annually thereafter. The project leaders are responsible for timely submission of the reports. Progress reports are reviewed by research thrust leaders, and will be approved by the Associate Director for Research, Claire Welty. Monthly project progress updates: Project teams are required to update the status of project activities on at least on a monthly basis using the Trello web tool. The purpose of project tracking is to ensure that the other UWIN members are informed about the progress of projects, data, and modeling activities. Performance evaluation metrics: The network evaluation plan includes the milestones and metrics described below to enable continuous and formal annual assessment of the research network performance. The assessment includes measures of progress in attaining the SRN program goal of assembling multidisciplinary teams of researchers, educators, managers, policymakers, and stakeholders that collaboratively investigate fundamental sustainability challenges. The project’s impacts and achievements are also quantified. The assessment is conducted via continuous collection and analysis of the milestone and metrics data and annual summaries. The results are used to monitor performance and make recommendations for adaptive adjustments of activities to enhance network performance. Metrics for assessment of the network include:

Inputs o Budget expended o Resources allocated o Projects conducted and concluded

Products o Papers in related peer-reviewed journals o Presentations at conferences and professional meetings o Theses and dissertations o Students graduated o Individuals reached through education and engagement o Data sets created and made public via UWIN data services o Simulation models and data analytics created and made public via UWIN modeling services

Outcomes o Researcher and stakeholder linkages within the network o Website utilization o Impact on literature (citations) o Impact on media (web and media citations) o Impact on policy (testimonials) o Network expansion over the years o Translation of research into practice o New spin-off funding o Quantification of personnel transitions o Retention of network knowledge

4

Initial milestones for evaluation of the progress of UWIN activates include:

Establish the organizational structure of UWIN and staffing by August 30, 2015

Launch UWIN website by August 30, 2015

Develop subcontracts between CSU and partner UWIN organizations by August 30, 2015

Hold kick-off meeting by August 30, 2015

Develop an initial Urban Water Sustainability Analysis Framework that guides the development of research plans by September 15, 2015

Develop initial research, education, and engagement plans by September 31, 2015

Review of project plans by UWIN management team to enhance the synergy among projects by October 30, 2015

Adopt diversity plan for UWIN by November 15, 2015

Develop final project plans by December 31, 2015

Create education and diversity committees by December 31, 2015

Identify Research Thrust Leaders by December 31, 2015

Identify regional stakeholder advisory committee members of the 6 UWIN regions by December 31, 2015

Develop and launch interdisciplinary Urban Water Sustainability seminar series beginning January 2016

Identify, collect, organize and make available data (past, current, and alternative future conditions) that are used consistently in all projects by March 31, 2016

Identify, collect, and make available analysis tools and models that are used consistently in all projects by June 30, 2016

Establish regional advisory groups for regional engagement by March 31, 2016

Hold annual regional stakeholder meetings/workshops

Establish partnerships with leading national and international urban sustainability groups

Develop and streamline the REU program by May 15, 2016

Develop a Citizen Science Program plan by June 30, 2016

Develop the first version of the online Global Urban Water Hub by June 30, 2016

Develop the Urban Water MOOC curriculum/syllabus by June 30, 2016

Develop the UWIN Urban Water Sustainability Blueprint by August 30, 2018

Broadly disseminate the UWIN Urban Water Sustainability Blueprint by August 30, 2020

III. Goals, strategies, plans, and metrics for engaging the science community in the SRN activities

The goal of UWIN engagement with the science community is to better understand challenges that confront urban water systems. These challenges are complex and nuanced, change in time and space, and influence how elected officials and water service utility managers balance multiple, and often conflicting, perspectives in the management of their urban water systems. Specifically, the activities address the following objectives:

i. Document and catalog global urban water research and current practices to increase the use of data analytics and models, improve their accuracy and identify research gaps.

ii. Use a broad range of communication tools to reach the sustainable urban water science community and engage them in UWIN. UWIN uses social media, Water Environment Research Foundation (WERF)—and other partner’s—email lists, and other ‘influencers’ to promulgate information about UWIN activities, tools, and challenges to a large audience.

iii. Implement a program of effective online and live training on UWIN tools and research findings.

5

Because of complex interactions between the natural, built, social and economic environments coupled with changing climate patterns and the fragmented nature of governance, elected officials and managers rely on data analytics and models to navigate uncertainty as they search for feasible infrastructure solutions. However, few utilities have the in-house tools or resources to use the models effectively or to incorporate other effective tools into their decision-making. The complex, often interconnected, challenges already identified as their priorities include, but are not limited to:

Understand realistic climate change scenarios and implications for their systems;

Identify risks to water quality;

Identify risks to water quantity;

Understand how contaminants form and travel through potable water distribution systems;

Understand flooding risk concerns;

Understand the implications of assumptions made in modeling scenarios including boundary conditions and factors of safety;

Analyze trade-offs among risks and the costs of avoided damages associated with flooding or water quality failures;

Integrate social, economic, and environmental considerations with technical analysis, including the financing of infrastructure;

Create scenarios to allow adaptable implementation of infrastructure design;

Enable tailored model development with site specific data to address concerns of communities across UWIN regions;

Address aging infrastructure conditions in light of the risk of failure, land use changes, or other conditions UWIN plans for addressing the science community engagement objectives include:

Annual Regional Meetings: Regional meetings in the 6 UWIN regions bring together more than 250 individuals, including members of the science community, every year to learn about regional urban water sustainability challenges. The Online Urban Water Hub: The success of sustainability research networks also depends on how knowledge mobility and network stability are managed. To foster knowledge mobility and seamless collaboration amongst network members and the global urban water community, an internationally prominent Urban Water Sustainability Hub will be created to facilitate contribution and use of knowledge by members across the UWIN network nodes and the broader urban water science community around the world. The scalable and accessible cloud-computing eRMAS technology will be used to create the hub. The Hub will enable researchers, educators, students, planners, and citizens to contribute and obtain data, models, analysis tools and other information in a collaborative fashion. The Hub serves the following purposes:

Information exchange: data, models, experiences

Contribution to refining the UWIN urban water sustainability framework and blueprint via participation in the definition, characterization, and quantification of sustainability indicators (Fig. 1)

Build teams to respond to societal needs in regions across the U.S. and around the world

Establish a network of networks

Bi-weekly Urban Water Sustainability Seminars: An Urban Water seminar series will be developed with a focus on pressures on water systems in the U.S. and throughout the world. The seminar series will be available via webinars with capabilities for live interactions. As a part of the seminar series, leading scientists working on these topics will be invited to engage with the UWIN researchers and the broader science community as well as interested regional stakeholders.

6

Urban Water MOOC: An Urban Water Sustainability MOOC for graduate student teaching and engagement will be developed. The goal is to reach interested students and the broader science community across the U.S. and beyond, with the added goal of using the resultant network to recruit students and postdoctoral fellows into UWIN activities. CSU has developed a successful MOOC entitled Water, Civilization, and Nature: Addressing Water Challenges of the 21st Century (http://www.online.colostate.edu/free-online-courses/water-civilization-and-nature/), and thus, has the experience and resources to create a MOOC on urban water sustainability.

Metrics for evaluating the success of UWIN engagement efforts include:

Scientist linkages within the network

Website utilization

Network expansion over the years

Translation of research into practice

IV. Goals, strategies, plans, and metrics for broadening participation of underrepresented groups

The purpose of the UWIN consortium diversity efforts is to cultivate talent and promote excellence across the social spectrum, encompassing all race/ethnicity, gender, disability status, nationality, religious affiliation, sexual orientation, or socioeconomic backgrounds. UWIN utilizes best practices to actively recruit and retain diverse undergraduate and graduate students and post-doctoral fellows from throughout the nation to participate in the UWIN research experience for undergraduates (REU) program, UWIN research, and the focused online course. The primary method of achieving UWIN diversity goals and fostering a spirit of inclusion throughout the center is a Diversity Committee that interacts with all of the researchers providing support and best practices for recruitment and retention of diverse students. UWIN consists of diverse institutions and the demographics that they serve are from varied backgrounds. Students are recruited from all backgrounds from the institutions that make up UWIN and other institutions in the network. In addition, the Diversity Committee identify targeted minority serving institutions (MSIs) with departments related to the research areas represented in the center. Once identified, the Committee will develop processes for contacting the departments and develop a rapport through repeated communication with faculty and administrators at these institutions, developing a network to support UWIN diversity in recruitment, participation and retention. Two minority serving institutions are on the UWIN team—Florida International University and Howard University. Team members at these schools have close contacts with colleagues at other minority serving institutions and can help facilitate these contacts. Skilled students, regardless of their sex, race, or heritage, are highly recruited for industry and academic positions. To enhance recruitment by institutions that are not noted by students, relationships and trust must be established. UWIN makes an attempt to build this rapport through UWIN REU and direct contact with programs at MSIs where possible. UWIN fosters a spirit of inclusion throughout the network, evident from marketing materials. The message stresses that there is a place for everyone in the field of sustainability. UWIN promotes a diversity-friendly culture that fosters culturally sensitive ethos throughout the network. Environmental justice is also a theme included throughout UWIN projects. The summer research experience and recruitment of new graduate students are pursued through the UWIN website, the websites of diverse national STEM organizations, through marketing materials distributed to targeted institutions across the country, and through marketing in national publications that reach diverse audiences.

7

One of the main goals of the Diversity Committee is to avoid “schema” – a pattern of thought or behavior that organizes categories of information and the relationships among them. No preconceived beliefs are established in research partners, students, or stakeholders from diverse backgrounds. The vision and mission of UWIN is inclusive of all people. A diversity sub-committee, including UWIN researchers from different research areas, will focus on promoting best broadening participation practices throughout the network. The committee will conduct evaluations of the network activities and will provide recommendations to UWN leadership for improving diversity. Each UWIN institution is committed to participation by women and minorities at various capacities, including students, staff, administrative professionals, and research associates.

Goals: desired outcomes

Recruit and retain diverse undergraduate students, graduate students, post-doctoral fellows, staff and faculty in all UWIN programs.

Develop a culture of inclusiveness across all UWIN institutions and activities that have a positive effect on recruiting and retaining diverse research and education communities.

Train and support a diverse and talented pool of professionals who go on to contribute to the diversity and vitality of the fields associated with the UWIN consortium.

Establish procedures and practices that ensure all UWIN outreach and stakeholder activities strive to be optimally inclusive of the entire diversity of relevant groups and perspectives.

Plans and Strategies: how does UWIN reach out to various communities

Create a Diversity Committee: The committee will meet regularly to development processes and procedures and track progress. The committee’s primary goal is to support the UWIN institutions’ principal investigators in recruiting, matriculating, and retaining diverse students, post-doctoral fellows, research staff, faculty, and stakeholder engagement processes.

Identify departments at MSI colleges and universities in order to develop a network of contacts for recruiting students to UWIN activities and research programs.

Reach out to diversity offices at each of the UWIN institutions to receive marketing assistance in reaching out to students on our campuses and feeder institutions.

Build attention to diversity into project management through a co-Director for Diversity (Glass) and a Diversity Committee.

Establish a consortium-wide vision for diversity supported by guiding documents for how to achieve the vision in the practical activities of the project.

Provide consortium-wide resources to support recruitment.

Incorporate the collection of data and information about recruitment and retention activities, and outcomes into the routine reporting procedures of the consortium, supported by project leaders and staff; and use these data to modify activities adaptively.

Nurture the culture of inclusiveness by incorporating diversity into the trainings provided to UWIN students, staff and faculty.

Document the diversity of people reached in our stakeholder and other outreach and engagement activities, and use these data to modify activities adaptively.

Metrics: how does UWIN measure success

Track our efforts in recruitment of students from diverse backgrounds, both from identifying feeder MSI institutions and working with the Office of Diversity at participating institutions. Centralized data on the

8

diversity of applicants to our REU program, and collated data on the diversity of applicants to all UWIN positions (technicians/staff, students, post docs, etc.). Data on the use and growth of our recruiting tools and resources over time?

Improve on the diversity numbers at UWIN institutions and exceed the aggregate numbers in the STEM fields represented in the network compared with national averages.

Compare our success in retention of students in UWIN fields over the life of UWIN to the national averages of the rates of success of women and underrepresented minorities in STEM.

Track the educational and vocational attainment of all UWIN participants for at least 3 years post-participation to measure their retention in UWIN fields and qualitatively assess the impact and value of UWIN for their subsequent development.

V. UWIN Organization

Organizational relationships and reporting structure among the key areas of responsibility

The UWIN management team includes four areas of responsibility (Fig. 2): (i) associate director of research, (ii) associate director of outreach and engagement, (iii) associate director of education, and (iv) associate director of diversity and broadening participation.

Key members of the SRN Management Team, roles and areas of responsibility

The UWIN organizational chart is shown in Fig. 2. Dr. Mazdak Arabi is the Director with Dr. Pivo as the Deputy Director. The UWIN Management Team consists of the Director, Deputy Director, Associate Director of Outreach, Associate Director of Education, Associate Director of Research, and the Associate Director of Diversity. The Management Team is responsible for ensuring that risk management is part of each of the programs and projects of this effort. Dr. Mazdak Arabi serves as the UWIN director with overall responsibility for developing and maintaining the shared research, education, and goals of the Network. He is the Borland Professor of Water Resources in the department of Civil and Environmental Engineering at CSU with an extensive experience in managing large grants and multidisciplinary projects.

9

Dr. Gary Pivo is a professor of Landscape Architecture at University of Arizona and serves as the Deputy Director for the Network. Dr. Pivo was the Dean of the College of Architecture and served as the Dean of the Graduate College and Director of Graduate Interdisciplinary Programs at the University of Arizona. Dr. Claire Welty, Director of the Center for Urban Environmental Research and Education and Professor of Chemical, Biochemical, and Environmental Engineering at UMBC, has extensive experience with large multidisciplinary research projects such as the Baltimore WSC project. She is also Co-PI of the Baltimore Ecosystem Study LTER. She serves as the UWIN Associate Director of Research. Dr. Welty coordinates delivery of research outcomes working with leaders of UWIN research thrusts. Dr. Alan Berkowitz at CIES, with nearly three decades of experience running a site-based REU program at the Cary Institute, serves as the Associate Director of Education. Dr. Charles Glass at Howard University serves as the Associate Director of Diversity. Dr. Michael Sukop at FIU coordinates outreach activities as the Associate Director of Outreach. He coordinates activities with regional outreach leaders including Dr. Swan (UMBC) in the Mid-Atlantic region, Dr. Meixner (UA) in the Sun Corridor region, Dr. Arabi (CSU) in the Front Range Colorado region, Dr. Jenerette (UCR) in the Los Angeles metropolitan area, and Dr. Hulse (UO) in the Pacific Northwest. Dr. Shirley Vincent from the National Center for Science and the Environment (NCSE) serves as the External Evaluator for the project. Dr. Vincent is a national leader in environmental and sustainability education, with considerable expertise working in higher education and additional experience with large and diverse education program evaluation such as will be involved in UWIN. She has experience evaluating projects supported by NSF, and is familiar with the requirements for documentation, formative and summative assessment.

Responsibilities of the lead and partner organizations

The lead PIs at UWIN institutions are responsible for ensuring the participation of the personnel at their institutions and ensuring their contributions to research projects as outlined in the project plans described in Appendix B,

NSF Program Manager

Director: M. Arabi (CSU) Deputy Director: G. Pivo (UA)

External Advisory Committee

Figure 2. U-WIN organizational chart.

Associate Director for Research C. Welty (UMBC)

Associate Director for Diversity C. Glass (Howard U.)

Research Programs Diversity Programs

Thrust A (Leader to be selected) Thrust B (Leader to be selected) Thrust C (Leader to be selected)

Thrust D (Leader to be selected)

Associate Director for Education A. Berkowitz (CIES )

Associate Director for Outreach M. Sukop (FIU)

Regional Stakeholder Advisory

Committees

Education Programs

Mid Atlantic/Baltimore (C. Swan, UMBC) S. Florida/Miami (M. Sukop, FIU) Sun Corr./Phoenix (T. Meixner, UA) Front Range/Denver (M. Arabi, CSU) S. Cal/Los Angeles (D. Jenertte, UCR)

Pacific NW/Portland (D. Hulse, UO)

10

education activities described in Appendix C, engagement activities described in appendix D, and training activities described in Appendix E. Additionally, institution PIs are responsible for monitoring the expenditure at their institutions according to the research, educational, and engagement activities and outcomes. The lead PIs at UWIN institutions are:

Colorado State University (CSU): Lead PI- Dr. Mazdak Arabi

Arizona State University (ASU): Lead PI- Dr. Matei Georgescu

Cary Institute of Ecosystem Studies (CIES) ): Lead PI- Dr. Alan Berkowitz

Florida International University (FIU): Lead PI- Dr. Michael Sukop

Howard University (HU) Lead PI- Dr. Charles Glass

Oregon State University (OSU) Lead PI- Dr. Roy Haggerty

Princeton University (PU) Lead PI- Dr. Elie Bou-Zeid

University of Arizona (UA) Lead PI- Dr. Gary Pivo

University of California-Berkeley (UCB) Lead PI- Dr. Arpad Horvath

University of California-Riverside (UCR) Lead PI- Dr. Darrel Jenerette

University of Maryland Baltimore County (UMBC) Lead PI- Dr. Claire Welty

University of Miami (UM): Lead PI- Dr. David Letson

University of Oregon (UO) Lead PI- Dr. David Hulse

University of Pennsylvania (UPENN): Lead PI- Dr. Jessica Bolson

Water Environment Research Foundation (WERF): Lead PI- Jeff Moeller UWIN research activities are organized and integrated according the four research thrusts of the network:

Thrust A: Assess baseline conditions for sustainability across ecohydrologic regions under prevailing infrastructure and institutional systems, with current and future economic, demographic, and hydro-climatic projections

Thrust B: Design and discover innovative technological solutions for sustainable urban water across spatial scales Thrust C: Learn to foster innovative transitions (urban form, behavioral, management, policy) for sustainable urban water management Thrusts D: Evaluate the impacts, tradeoffs and co-benefits of innovative technologies and transitions in comparison to the baseline across spatial and temporal scales

See Appendix B for organization and management of research projects by Research Thrusts. Appendix D provides detailed information about plans, organization and movement of engagement activities. The organizational structure of UWIN requires research project personnel to report their activities, outputs, and progress to the PI of the projects, Project PIs report to corresponding research thrust leaders, and thrust leaders report to the Associate Director of Research.

VI. Processes to select and integrate research projects with one another and with other SRN activities

To achieve the goals of the Network and ensure cohesion, our approach includes a suite of activities and tools that have been used successfully on similar projects conducted by team members.

11

Prioritizing SRN Activities: SRN activities are prioritized based on regional stakeholder inputs, feedback and evaluation of external advisory committee, and internal evaluation of the programs. Chartering: Chartering for the network starts with a project kickoff, to align and inspire network team members. Ongoing chartering is required for this multi-year project, to continuously re-align and re-inspire the team to our singular purpose over time. Meetings: The proposal calls for a significant number of Network meetings, many of which include participating stakeholders, decision-makers, and agency technical staff. Annual all-Network member meetings consist of presentations from project PI’s, breakout groups, and graduate student presentations and posters. Table 1 presents a summary of meetings we anticipate to hold.

Table 1. Coordination meetings that enhance coordination and foster integration of activities

Reporting: Reporting requirements are put in place to ensure timely delivery of products:

Senior personnel and Associate Directors report their accomplishments semi-annually.

The Network Director, Deputy Director, and Network Administrator are responsible for developing a draft annual report from all PI contributions prior to the Annual Meeting.

Suggestions during the Annual Meeting are then used to finalize the reports.

Annual reports are uploaded to Fastlane and the project website. Manuscripts are prepared for peer-reviewed publication. The work is presented at various conferences.

Meeting Goal Participants Frequency

Overall project coordination Network Director, Deputy Director, and Administrator Weekly

Student-led project progress All students and any interested participants Monthly

Outreach, education, research, and diversity coordination

Associate Directors, Network Director, Deputy Director, and Network Administrator, Regional UWIN Leaders

Monthly

Project vision External Advisory Committee, Associate Directors, Director, Deputy Director, and Network Administrator, Regional UWIN Leaders

Quarterly

Progress review Associate Directors, Network Director, Deputy Director, and Network Administrator

Semi-annual

Regional workshops Stakeholders, traveling stakeholder outreach team, Regional UWIN Leaders, students and post-docs

Annually

Overall project status review, issue identification, cohort -building

All PIs, students, and post-docs, External Advisory Committee, affiliated researchers, and stakeholders

Annually

12

Collaboration Tools: The eRAMS technology is used as the collaborations tool across UWIN. For tracking the progress of projects, the web-based Trello technology (trello.com) is used. Figure 3 illustrates an example of the use of Trello for organizing UWIN activities.

Selection of Research Projects

Research projects are selected, developed and continuously revised to support the UWIN Urban Water Sustainability Framework (Fig. 1) and Blueprint. The selection process includes the following considerations:

Benefits to the Urban Water Sustainability Framework and the Blueprint

Regional stakeholder inputs and suggestions

Feedback and evaluation of UWIN External Advisory Committee (EAC)

Internal evaluation of the programs by UWIN management team

Figure 3. Tello web-application for tracking the progress of research, education and engagement activities

Integration of Research Projects

13

UWIN management team and participants use systems and tools that allow seamless integration of research and engagement activities. First, Thrust D projects focus on integration and synthesis of information and activities form all other projects. Project D1-1 provides a consistent approach to characterization of baseline and alternative futures for climate, land use, population and urban development patterns in the study regions. Project D1-2 works with all other projects on collection, organization and access to data used in the research, education and engagement activities. Project D1-3 streamlines access to data analytics and modeling services using the environmental Risk Assessment and Management System (eRAMS). Engagement activities described in Appendix D fosters integration of information and synthesis toward characterization of sustainability of regional urban waters systems across UWIN regions. The online Trello web-application is used by all projects to track progress of activities within the network. Members of research teams can follow each other’s progress, including activities that have been completed, are currently being done, or will be done. UWIN projects are highly interconnected. All senior personnel and students are required to continuously update the status of their corresponding projects on a monthly or more frequent basis. All data, analytics, and modeling services will be integrated and streamlined for access by UWIN members, and the broader urban water community when appropriate, using eRAMS.

VII. Processes to allocate funds and equipment across SRN activities and among partners

Similar to selection of research projects, funds and equipment are allocated across UWIN partners to maximize the benefits to the UWIN Urban Water Sustainability Framework (Fig. 1) and Blueprint. Other important considerations include:

Regional stakeholder inputs and suggestions

Feedback and evaluation of UWIN External Advisory Committee (EAC)

Internal evaluation of the programs by UWIN management team

VIII. Data Management Plan

UWIN takes advantage of ongoing efforts of the eRAMS/CSIP Cloud Modeling Integration framework (Fig. 4). Additionally, tools and standards of the open web (http://www.w3.org/standards/) as well as CUAHSI HIS (http://his.cuahsi.org) and LTER (http://www.lternet.edu) networks are used for the management, documentation, and dissemination of data for UWIN projects. A data management strategy is presented in terms of: (1) the types of data and samples that are managed; (2) data and metadata standards; (3) policies for data access; (4) policies for reuse, distribution, and the production of derivatives; and (5) plans for data archival and preservation.

(1) Types of data and samples to be managed in the project

The types of data managed for UWIN activities include static (read-only) data with only annual or periodic updates; spatial-temporal measurements of hydrological, geologic, geochemical, and biotic variables as well as spatial layers (LiDAR, imagery, thematic, DEM); static files that can be stored outside of a database; and write-centric dynamic data such as user session information, dynamic GIS feature classes or raster layers, or other RDBMS data.

(2) Standards for data and metadata format

All data are shared in open, industry-standard formats. Observational data collected in this project are archived in accordance with eRAMS data management practices. Metadata are compliant with data dictionary and ontology

14

consistent with CUASHI, LTER, and WSC projects. Spatial and remotely sensed data are stored in formats that adhere to Open GIS Consortium (OGC) standards. All data and derivative products are documented using Ecological Metadata Language (EML) which is a subset of the Federal Geographic Data Committee (FGDC) standard presented in an XML schema known to be successful for LTER research and currently the de facto standard. Model data inputs are stored in formats which are used in the model. Model results are stored in the format returned from the model or when possible human readable formats (i.e ascii, .txt, .xml, .json).

(3) Policies for access and sharing

Data and metadata are available via the eRAMS website (www.erams.com) and through CSIP modeling services (http://bit.ly/1m0RoT7). All project data, metadata, and results are made available through the project website within 24 months of its collection. This approach provides ample time to generate publications from the data and protects individual intellectual property rights. Special circumstances may exist where the data is constrained by a license agreement. In this case, data are kept internal to project participants. Samples obtained in the course of scientific activities are preserved and housed at each researcher’s home institution.

(4) Policies for re-use, distribution, or the production of derivatives

All hydrologic spatial time series data are made available using eRAMS /CSIP data services. The eRAMS IAAS technology (Lloyd et al., 2013) is used to ensure scalability, accessibility and platform independence of data access and modeling services. Accessibility and security is ensured using the REST (REpresentational State Transfer) protocol combined with token-based authentication where applicable.

Data and derivative products are also made available through these channels and the project website. Options include direct data download via REST or FTP and a query interface for metadata-based data discovery. Finally, the project website is used to publish analysis results and derivative products. Along with the project website, data are available through a number of networks including the LTER and CUAHSI networks.

(5) Plans for archiving data, samples, and other research products and pre servation of access

All CSU data, documents, and results are stored on a cloud infrastructure housed at CSU. All data in the repository are backed up continually. All physical samples are preserved for five years beyond the end of the project.

Desktop Clients

Web Clients

Mobile Clients

RESTful Service

Data

Service Locator

(High Availability

Load Balancer Redirection)

RESTful

Service

Online Urban Water Sustainability Hub Content Management, Code Repository, Wiki, Forums, Blogs,

Theoretical Documentation, User's Guide Documentation

eRAMS Geospatial Services

Decision Analysis

Scenario Analysis Optimization

Uncertainty Analysis System Identification

Visualization

eRAMS Modeling Services

Water Infrastructure Models

Water Socioeconomic Models

Life-Cycle Analysis

Model Development, Enhancement, and Integration using the

Highly Scalable and Accessible eRAMS Platform

Cloud Web Services

CMIP5 climate ICLUS land use

USGS NIWS EPA STORET

CAP LTER BES LTER

UMBC WSC Front Range WSC Areamette WSC S. Florida WSC

Los Angeles WSC CUAHSI

Figure 4. Integration of UWIN data, models and decision innovation system.

15

All data products from UWIN are disseminated between the Network nodes and also via the online Urban Water Sustainability Hub using the data/model integration platform shown in Fig. 4.

eRAMS Information Management Components

The UWIN data analytics and models are developed, deployed and maintained using eRAMS, which works easily at scale, encourages code reuse, fosters innovation, and promotes collaboration. The Platform blends together strengths of powerful and proven open source technologies (Fig. 5): the Codebeamer Application Lifecycle Management (ALM) solution, eRAMS-Version Control System (VCS), and eRAMS-Cloud Service Innovation Platform (CSIP).

The components of the platform can communicate with one another and with external clients through an application programming interface (API) built as RESTful web services. RESTful service catalog of endpoints enables interactions with easily-discoverable and standardized Uniform Resource Locators (URLs) within its infrastructure. REST is a widely-used method to implement client/server architecture. Comprehensive Code Management: The eRAMS team at CSU leverages the Codebeamer ALM system for comprehensive code management. The award-wining system covers all phases of the development process including development, testing, risk management, and requirement management. In addition, the system offers a comprehensive wiki, reporting, issue tracking and extensive document management functionality with versioning. CSIP service hosting, testing, and release management is already integrated using this ALM solution. Project partners are able to transparently manage the model development and service release using Codebeamer and the version control system. More information about the Codebeamer ALM can be found at: http://intland.com/codebeamer/product-overview/. Model Code Repositories and Version Control System: CSU utilizes VCS repositories to host LCC models and service development. The repositories are available to all project partners who are involved in model development during and after the project period. Researchers are able to create their own private code projects including a Subversion-based source code control system and a wiki for notes and other documentation. The distributed nature of this

Figure 5. The architecture of the platform for LCC modeling services

Content Management System: Codebeamer AML

16

version control system ensures that code development always stays current at each location. CSU hosts all repositories and provides their seamless integration into the automated CSIP Model Services update, testing, and deployment. The VCS fosters best practices for reuse, maintenance, and update of LCC models. Model Services in CSUs eRAMS-CSIP infrastructure: The eRAMS-CSIP is an open source Model-as-a-Service infrastructure for providing a modeling and data services using a RESTful protocol. Service management is fully automated and currently serving more than 120 model and data services at CSU. A process has been developed to allow model developers to make models quickly available as web services using the CSIP API and cloud infrastructure. Services might be deployed in a CSU cloud infrastructure or public cloud, such as Amazon with suitable security measures. Geospatial urban water model user interfaces are provided via eRAMS-GIS, while the models are provided as eRAMS-CSIP RESTful services. eRAMS-CSIP is open source and may be deployed at collaborating institutions. Security and Privacy: All servers at CSU, including eRAMS servers, implement multiple firewalls to ensure protection of user data and information. Additionally, eRAMS uses HTTPS (secure-hypertext-transfer-protocol) for communication over the network with added encryption layer of SSL (secure-sockets-layer) to protect the traffic. Other security details at the application level include:

CSRF (Cross Site Request Forgery) Protection - eRAMS checks for nonce values in most of its HTTP POST requests to prevent most kinds of CSRF attacks.

SQL Injection: database queries in eRAMS properly escape user input in SQL queries to prevent users from executing arbitrary SQL on server machines.

HOST Header Validation: Server requests are verified against a list of allowed hosts to prevent CSRF, cache poisoning, and email poisoning.

Apache Web Server: eRAMS is hosted using the latest distribution of Apache which contains many security updates and bug fixes.

Finally, security for the eRAMS cloud is based on Identity and Access management (IAM), an implementation which parallels that used by Amazon's Elastic Compute Cloud (EC2). IAM provides a framework for the management of electronic identities. The framework supports identity management necessary to initiate, capture, record and manage user identities and their related access permissions in an automated fashion. This approach ensures that access privileges are granted according to policy interpretation and all individuals and services are properly authenticated, authorized and audited. All user data on eRAMS are by default private and protected by security protocols and systems. eRAMS offers a secure collaboration component that enables creating “Groups”. Users can create a group and enlist members. Members of a group can share their data and projects with the other members with read-only or full-access privileges. All data and models on eRAMS can be seamlessly published for dissemination to global audiences via a RESTful approach with no further effort, once the user provides necessary permissions.

IX. High level risk management plan

Purpose

The purpose of the risk management plan is to establish the approach that is taken to identify, quantify and manage technical, schedule and cost risks associated with UWIN. The intent is to actively manage all program potential risk areas. In some cases, conflicts may exist between technical issue resolution, cost and schedule. The challenge is in balancing the effects among these three areas. Risk is composed of two components: 1) the probability that a given risk issue will materialize, and 2) the consequences, in terms of cost and schedule, associated with finding a solution to the given risk issue. The product of probability and consequences allows risk issues to be quantified and ranked in terms of potential impact to the project.

17

Definitions

Risk: Risk is a measure of the uncertainty associated with being able to achieve overall program objectives within defined cost, schedule and technical constraints. Risk is composed of two components: 1) the probability that a given risk issue will materialize, and 2) the consequences, in terms of cost and schedule, associated with finding a solution to the given risk issue. The product of probability and consequences allows risk issues to be quantified and ranked in terms of potential impact to the project.

Risk Event: Risk events are events that have the potential, due to unfavorable outcomes, to result in significant delays or cost increases. Risk events should be well enough defined to support accurate estimation of probability and consequence of occurrence. Technical Risk: The risk associated with design evolution, including the solution of both known and unknown technical challenges. This topic also includes the risk of whether performance requirements can be fully achieved. Cost Risk: The risk associated with the program to achieve its cost objectives. Schedule Risk: The risk associated with the program’s ability to meet each scheduled milestone event. Schedule risk includes the accuracy of the original schedule estimates, and whether all required tasks were included in the master schedule. Risk Ratings: The quantitative values assigned to a given risk event based on the probability of the event occurring and the severity of the consequences of the event.

Risk Management Strategy and Approach

The basic approach for identifying, quantifying and managing risk consists of four parts: Risk Identification, Risk Assessment, Risk Reduction and Planning, and Risk Monitoring and Re- Assessment as shown in Fig. 6. By utilizing iterative and adaptive project management frameworks and tools, the Management Team is able to minimize the impact of many risks. Communication and transparency are greatly improved, resulting in a more agile and ‘rapid response’ rather than recouping losses later in the project.

Risk Identification/Planning

A key element in effective risk management is the identification of all possible risk areas. Without thorough identification of the potential risk areas brought under consideration, even the most effective risk management techniques are ineffective. The complete list of project potential risk items is dynamic, and changes with the particular phase and conditions of the project. A UWIN risk matrix or register is used by all projects that are part of UWIN. The UWIN Management Team incorporates the top 3 risks from each project into the UWIN risk register and manages this register. Due to these influences, the list of risk items is reviewed and updated periodically. The Management Team reviews the list periodically in order to monitor the changing risk situations and their potential impact on the overall project. A draft risk register is found in Appendix F.

Risk Assessment

LOW MEDIUM HIGH

1 2 3

IMPACT

LOW

1

PR

OB

AB

ILIT

Y

HIGH

3

Monitor Action RequiredMEDIUM

2

Figure 7. Risk Assessment Table

Figure 6. Risk Management Process

18

Risk assessment provides a means of ranking or establishing the relative importance of risks after they have been identified. Each risk is evaluated in terms of its probability of occurrence and the impact to schedule, cost, and performance. For ease of use, in this project the probability is defined into ordinal categories, which are indicators of percentage ranges and the impact on each cost, schedule, and performance. See Figure 7 for additional detail.

Probability of occurrence: 3 - High probability More likely to happen than not >50% 2 - Medium probability Fairly likely to happen 20%-50% 1 - Low probability Not likely, but not impossible 0-20%

Impact: o Impact on time (schedule):

3 - High impact Causing a large delay to the project 2 - Medium impact Causing a significant slip to the project 1 - Low impact Causing a small slip to the project

o Impact on cost: 3 - High impact Large increase to total cost 2 - Medium impact Significant increase in total cost 1 - Low impact Small increase in total cost

Risk ratings of 3 or above require a specific risk reduction action plan (see Fig. 7).

Risk Mitigation

Risks of significant probability and medium or higher impact should consider a prevention action where feasible to avoid the risk. All risk ratings greater than 3 have an action plan associated with them to reduce and manage the risk throughout the project lifecycle. The risk action plans are an integrated part of the overall risk management effort. The action plans are reviewed at project status reviews and summarized where needed in the Network Management Team Reviews. Risk mitigation can be:

Avoid- Take action now to prevent the risk from occurring.

Minimize- take action which will reduce the impact, perhaps by considering an alternative option

Transfer - Reduce or remove the risk by transferring it in space, time or ownership, as in subcontracting to experts

Accept – Accept the risk

Control and Documentation

Throughout the life of the project it is necessary to keep an up-to-date record of all risks. This record is maintained through the Risk Matrix. It is also necessary to retain a log of the risks, showing how the risks have matured over the project lifecycle. This should record the decisions taken during the planning process and the reasons for taking those decisions. This is useful as a project reference document and as part of the risk management feedback process. For this purpose it is useful to maintain the risks on a database or spreadsheet. The risk matrix is the primary Network documentation. Individual projects however, should maintain further information regarding risks anticipated or incurred to use for future projects and lessons learned.

Monitoring requires:

Identification of new risks as they appear

Identification of old risks that rise to a risk rating of 3 or greater

19

Identification of old risks that have successfully been managed below a risk rating of 3

Identification of action plans that are not working

Modification of action plans to adapt to the changing situation Monitoring risk management requires continual monitoring of the project progress against the success criteria and the risks identified. As such, the UWIN Management Team has identified the following deliverables:

Individual project risk matrix is a deliverable at each review

The full risk matrix is shared at all project and contract reviews.

A summary of the significant risks should be included with each project status report (PSR) and is included in the program risk matrix.

X. Succession plans for the SRN leadership team

Dr. Sybil Sharvelle, an associate professor of Civil and Environmental Engineering at CSU will serve as the UWIN Director should unanticipated conditions hinder Dr. Arabi to serve at this capacity. The changes in other leadership positions will be first evaluated by the management team, including the Director, Deputy Director, and Associate Directors of Research, Educations, Engagement, and Diversity. The management team will solicited inputs from UWIN project leaders and investigators to ensure incorporation of feedback and suggestions in filling leadership positions.

XI. External Advisory Committee (EAC)

The External Advisory Committee (EAC) provides continuous guidance and advice to the management team throughout the life of the Network. The membership of the EAC consists of five to seven peer scientists from the academic, private, non-profit, and public sectors. The basic functions of the EAC are to assist in evaluation of the effectiveness of the Network’s projects and the relevance and importance of the program elements to its goals. UWIN holds a meeting with the EAC annually in consultation with the NSF Program Officer and SRN Director. EAC serve NSF and UWIN by creating a bridge from the research community to the realm of urban water management:

Enable visioning of UWIN

Help focus research and other UWIN activities

Assure the best performance from UWIN programs

EAC Draft Charge

The committee serve the following purposes:

Represent full range of urban water stakeholders

Study results of regional stakeholder workshops

Interpret how results from test bed cities relate to vision for integrated urban water management

Advise project teams on research-to-innovation for integrated urban water systems

Participate in visioning, developing policies for project selection, and evaluation of projects and UWIN outcomes

Assist in creating sets of best management practices for sustainable urban water management

Meet annually with NSF Program Officer and SRN Director to report evaluation and provide policy direction for UWIN

Foster purposeful expansion of UWIN to new stakeholder/researcher teams

20

EAC Operating Plan

6-8 members

Appointed within 60 days after the notice to proceed

Initial organizational meeting via videoconference

First face-to-face meeting during the first six months

A Chairperson will be appointed within six months

Quarterly meetings with the UWIN directors

Annual meetings with NSF Program Director(s)

EAC Preliminary Roster

Selection criteria:

Research Scientists/Academia (international thought-leaders on urban water)

Environmental/Conservation/Environmental Justice Groups

Water Resources Utilities and Leading National Experts

Leading Urban Water Consulting Firms

Private Water Companies

Urban Planners/Community Organizers Members:

Jeanne Van Briesen, Professor, Civil and Environmental Engineering, Carnegie Mellon University

Brian Richter, Chief Scientist, Nature Conservancy

Glen Daigger, Professor, University of Michigan

Charles Moesta, Vice President of Market Development, American Water

Larissa Mark, Environmental Policy Program Manager, National Association of Home Builders

Ben Grumbles, President ,National Water Alliance

Marc Cammarata, Philadelphia Water Department

21

Appendix A: Quality Management Plan Checklist

QA Manager:

Develops Quality Management Plan and reviews annually.

Develops repository for documents and document control system.

Oversees Data Management Plan.

Project Plans (by Project PIs)

Approval: Approval of project plans by signatures on cover page of the Project PI, QA Manager and UWIN Director.

Signed QAPPs archived by QA Manager. QC Reviewers: Project PIs designate project QC Reviewers. Technical System Assessments: QA Manager conducts Technical Systems Assessments of project plans (on-site during first year and once every other year thereafter).

Between on-site TSAs, QA Manager reviews checklists by conference call or in person.

Performance Criteria and audit checklist developed during project plan approval. Assessment Results: QA Manager reports assessment results to UWIN director. QA Manager archives corrective action. Project QA Procedures: Project methods and QA procedures assessed annually by PIs. Staffing: PIs ensure that staff receive training and document qualifications and training. Report hiring and training procedures to QA Manager semi-annually. Data: Data, document and record collection, back-up, collation, transfer and storage handles in accordance with the Data Management Plan. All records maintained in paper or electronic media. Changes in data methods or QA/QC procedures reported to QA Manager immediately. Reviews: Each research product reviewed per the QMP. Logistics: Extramural agreements approved by UWIN Director. PIs are responsible to ensure that supplier products fulfill needs and requirements. PIs document computer hardware and software use in Project plans.

22

Appendix B: Research Project Plans

Thrust A: Assess baseline conditions for sustainability across ecohydrologic regions under prevailing infrastructure and institutional systems, with current and future economic, demographic, and hydro-climatic projections

A1-1: A multi-scale, multi-model dynamical-probabilistic approach to quantifying vulnerability, resiliency and adaptability of US urban water systems to climatic and socio-economic variability

A1-2: Impacts of changes in climate, demographics, and urban form on water supply-demand equilibrium, economic growth, and social equity and equal opportunity

A2-1: Land-atmosphere-hydrosphere interactions in urban terrain

A2-2: Land-atmosphere-hydrosphere interactions: projecting future environmental change in urban areas

A2-3: Assessing the thermal comfort implications of water-supported urban infrastructure at the human scale

A2-4: Assessment and design of innovative building systems and urban infrastructure to mediate impacts on the urban water cycle, heat island, and regional climate

A3-1: Continental Scale Variation in Urban Vegetation Biodiversity – Ecosystem Functioning

Thrust B: Design and discover innovative technological solutions for sustainable urban water across spatial scales

B1-1: Water management solutions to enhance capacity for use of alternative water sources

B1-2: Spatially- and temporally-informed life-cycle assessment of urban water systems

B2-1: Comparative impact of Green Infrastructure impact across UWIN Urban Systems

B2-2a: Flood hydrology and rainfall frequency

B2-2b: Hydrology and hydraulics of urban floodplains Thrust C: Learn to foster innovative transitions (urban form, behavioral, management, policy) for sustainable urban water management

C1-1: Understanding the Adoption of Sustainable Water Solutions

C2-1: Using dynamic information acceleration to understand and forecast homeowner adoption of new technologies for sustainable water management

Thrusts D: Evaluate the impacts, tradeoffs and co-benefits of innovative technologies and transitions in comparison to the baseline across spatial and temporal scales

D1-1: UWIN Envision Modeling of Present and Future Values for Sustainable Water Management Blueprint Indicators

D1-2: Cross-site comparisons and contrasts across ecohydrologic regions

D1-3: UWIN decision innovation system

23

PROJECT A1-1 Research Plan

Project Title A multi-scale, multi-model dynamical-probabilistic approach to quantifying vulnerability, resiliency and adaptability of US urban water systems to climatic and socio-economic variability

UWIN Project number A1-1 Project Lead Jorge A. Ramirez Investigators/Institutions

Thomas C. Brown (RMRS - USFS) Mazdak Arabi (CSU)

Project Period August, 2015 - August, 2018 Project Cost CSU: $370,000 Graduate students funded to work on project

PhD: TBA – Jorge A. Ramirez and Thomas C. Brown Postdoc: TBA – Jorge A. Ramirez and Mazdak Arabi

Project Overview

We will enhance the approach of Foti et al. (2014) to characterize distributions of water yield, water supply, and water demand in terms of spatially and temporally variable and mutually interacting probability distribution functions that will be used to assess water scarcity, water supply vulnerability, and water supply resiliency and capacity for adaptation. Water yield will be determined by coupling an ensemble of downscaled climatic projections from CMIP5 global climate models to a land surface eco-hydrologic model accounting for both surface and ground water processes (e.g., the Variable Infiltration Capacity model – VIC model), which will itself be enhanced to allow for climatically dependent and eco-hydrological optimal vegetation responses (e.g., Quebbeman and Ramirez, 2005). The fully integrated model will be implemented for the entire contiguous US at spatial scales ranging from 10 by 10 km to continental scales, and at temporal scales ranging from daily to monthly to annual over the 21st century. Water supply will be determined using a detailed water allocation model (e.g., WEAP, MODSIM).

Project Summary

Water scarcity, water supply vulnerability, and water supply resiliency and adaptability are dynamic, multidimensional, and scale-dependent. The interactions between water systems and social, economic, and environmental systems occur at multiple spatial and temporal scales; therefore assessment of impacts, tradeoffs, and benefits must include feedbacks across all spatiotemporal scales, and incorporate all multiple-way interactions. We will explore the impacts of urbanization on water scarcity, supply vulnerability and resiliency, answering the following questions:

How does urbanization lead to increased/decreased vulnerability?

How does urbanization lead to increased/decreased resiliency or capacity for adaptation?

How do changes in vulnerability or resiliency depend on the location in a city or the analysis scale?

How do the answers to the above questions vary by region, or depend on the rate of urbanization?

Supplemental Keywords

Dynamic-probabilistic approach; vulnerability; resilience; adaptive capacity; economic trade-offs.

Project description

1. Objectives The objectives of the study are:

Determine water supply sustainability indicators (i.e., water supply scarcity, vulnerability, reliability, and resiliency; water use per capita, water import per capita, reservoir storage, groundwater depletion) over the

24

historical period.

Determine water supply sustainability indicators under alternative future climate, population and land use scenarios.

Assess the effects of water management solutions including source separation, water recycling and reuse, and water policy under current and alternative future climate, population and land use conditions.

2. Intellectual Merit Adaptation to cope with the impacts of socio-economic and climatic variability and change requires strengthening of the resilience of the systems under consideration. Such strengthening can only be achieved after an objective quantification of their current vulnerability and adaptive capacity and resiliency. Because the drivers and determinants of vulnerability and resilience are extremely variable in space and time, and because the urban systems under consideration are extremely complex and interacting across multiple space-time and institutional scales, a probabilistic approach must be taken for such quantification. We will develop a unique probabilistic approach to the objective quantification of vulnerability and resiliency and implement it in a Bayesian decision-theoretic framework in order to develop adaptation measures to enhance and strengthen system resiliency. 3. Approach/Activities

Description of Tasks

A) Estimate Water Supply A.1 - Implement eco-hydrologic model (EM) (e.g., enhanced VIC model) A.2 - Test and calibrate the model based on observations A.3 - Using observations of hydro-meteorological, ecological, and socio-economic drivers estimate water

supply using EM A.4 - Estimate probability density functions (PDFs) of drivers A.5 - Project PDFs of hydrologic, ecologic, and system responses to probabilistic drivers A.6 - Use EM to estimate future water supply for past weather and climate A.7 - Use EM to estimate future water supply based on GCM projections of future climate

B) Water Demand Analysis B.1 - Collect estimates of past U.S. water withdrawals to project future demand B.2 - Project water demand

C) Vulnerability, Resiliency and Adaptability Analysis C.1 - For current climate conditions, quantify the vulnerability and resilience of water supply and other indicators of urban water sustainability C.2 - Based on GCM projections, develop scenarios of stressors: environmental, socio-economic, climatic C.3 - Evaluate vulnerability and resiliency of future water supply in a probabilistic framework C.4 – Bayesian analysis of uncertainty and value of information for adaptation decisions

D) Documentation D.1 - Research reports D.2 - Publish results

4. Expected Results, Benefits, Outputs, and Outcomes

Driving Forces and Pressures:

Climate change

Population growth

Economic development

Extreme events

Indicators:

25

Water supply vulnerability

Water supply resiliency

Adaptation capacity

Groundwater depletion vulnerability

Solutions:

Demand reduction strategies

Building to municipal level water recycling and reuse systems

Green infrastructure

Policy solutions and water law optimization 5. Data Needs The project team will work with Projects D1-1, D1-2, and D1-3 to create and utilize consistent data for the analysis in this project. 6. General Project Information

Project Timeline

Task A: 1/1/2016-12/31/2017

Task B: 1/1/2016-12/31/2017

Task C: 1/1/2017-12/31/2018

Task D: 1/1/2017-12/31/2018

Roles and Responsibilities

Dr. Ramirez will oversee all project activities including development of the models and datasets needed for supply and demand characterization. Dr. Brown will serve as a collaborative in the supply0demand analysis. Dr. Arabi will participate in incorporation of water management solutions in water demand assessment. 7. Interaction with Other UWIN Projects This project will be closely coordinated with projects A1-2 for the socioeconomic assessment, A2-2 for climate products, B1-1 for urban water conservation and recycling solutions, B2-1 for green infrastructure systems, D1-1 for alternative future scenarios, and D1-3 for development of water supply indicators. 8. References Foti, R, J.A. Ramirez, and T.C. Brown, 2014: Response surfaces of vulnerability to climate change: the Colorado River Basin, the High Plains, and California. Climatic Change. DOI 10.1007/s10584-014-1178-0. Foti, R., J.A. Ramirez, and T.C. Brown, 2014: A probabilistic framework for assessing vulnerability to climate variability and change: the case of the US water supply system. Climatic Change. DOI 10.1007/s10584-014-1111-6. Brown T.C., Foti R., Ramirez J.A., 2013: Projected freshwater withdrawals in the United States under a changing climate. Water Resour. Res. 49:1259-1276.

PROJECT A1-2 RESEARCH PLAN

Project Title A1-2: Impacts of changes in climate, demographics, and urban form on water

26

supply-demand equilibrium, economic growth, and social equity and equal opportunity

UWIN Project number A1-2 Project Lead Elizabeth Mack Investigators/Institutions

Jessica Bolson/UPENN (Not funded) Gary Pivo/AU (Not funded) Dave Hondula/ASU (Not funded)

Project Period Years 1-5 (funding available for Years 1-3) Project Cost Will be determined by Dec 31, 2015. Graduate students funded to work on project

RA TBA (4 Mo.: Years 1-3): Supervision, E. Mack

Project Overview

This project evaluates the economic impacts of water price increases on household income, regional income, and regional employment. Impacts on business output will also be estimated. This is important to consider given pressures on urban water systems such as aging infrastructure, growing populations, and climate change. These pressures mean that water costs will rise and place economic strains on businesses and households. From the business side, higher water costs could lead to increases in prices of inputs and salaries, and a lower willingness (and ability to pay) for business outputs by households. This strain on business outputs means that businesses may have to hire fewer workers. This reduction in the number of persons employed, increases the incidence of unemployment in households and reduces household income. Due to this reduction in household income, people will purchase fewer goods and services from businesses, which further reduces firm profits. In other words, there is a vicious cycle between many elements of water: water supply, water demand, water prices, business income, and household income. This means that rising water prices are a sign not only of an environmental problem, but an economic problem. Environmental and economic water issues are interrelated, and have spillover effects. What further complicates this vicious cycle is the concentration of these impacts in people and households The downside of this duality is likely strongest for low-income households who likely use the least amount of water and also pay a larger proportion of their income for water than do high income households. The figure below highlights that key actors (including households, businesses, and government) interacting within the environment cyclically impact water supply and water prices, as well as one another. Government plays a role in this system of impacts through its levels of taxation on businesses and households, but also, and perhaps, most importantly in the context of water, in its attempts to regulate water use through water use restrictions. Water use is often thought of primarily from an environmental point of view. Economic impacts are ignored, and this is particularly the case for low-income and disadvantaged households who have perhaps more fixed demands for water than do high-income consumers who demand is more flexible or elastic for water given the numerous ways in which they use water in their households (pools, hot tubs, watering lawns, etc.). This is distinct from low-income households who do not have pools and hot tubs to fill and perhaps even lawns to water. Their uses of water are more necessity oriented. From a water management standpoint then, the ways in which water use can be reduced, from a price perspective and also from a regulatory standpoint merit evaluation, especially from an equity standpoint. This is because market mechanisms for reducing water use via price are likely to disproportionately impact low-income households who will need to dedicate a larger proportion of household income to pay for water than will middle and high income households. Given this range of issues, the outcomes for this project seeks to provide information and educate a diverse range of individuals including households, stakeholders, and grassroots organizations with information about trends in water prices, water consumption, and an empirical assessment of the impacts associated with water price increases. It will

27

also use this information to work with these groups to construct water use regulation strategies that have more of an impact on regulating water use but also less disparate impacts on vulnerable, low-income households.

Source: E. Mack’s creation

Project Summary

This project will collect historical information about household water demand and water prices to characterize how much water and the price that households pay for water in each of the 6 UWIN regions. This information will be used to generate projections of water consumption and water prices. These past, present (baseline), and future scenarios of water consumption will be used as data inputs to perform a social accounting analysis. The social accounting analysis will be performed for two of the six UWIN regions (South Florida and the Front Range of Colorado) using IMPLAN data that will be purchased for this project. The output of the social accounting analysis will reveal four types of impacts related to changes in water consumption: 1. Variations in impacts for households of varying income strata; 2. Impacts on business output across a range of industries; 3. Impacts on household employment levels for varying income strata; and 4. Impacts on regional output, income, and employment. For all 6 UWIN regions, rough projections of aggregate regional economic impacts will be generated using national input-output information from the Regional Input-Output Model (RIMS) from the Bureau of Economic Analysis (BEA). This will provide some sense of regional impacts in terms of employment, output, and income from changes in household water consumption. To get a sense of how these regional numbers compare with those generated from IMPLAN, the results generated for the South Florida and Colorado Front Range regions using both BEA and IMPLAN information will be compared. Four intellectual products are expected from this project. The first product is an assessment of household income and employment impacts for households of varying income strata for the South Florida and Colorado Front Range regions. The second product is a broader understanding (at the regional level) of industries impacted by water consumption changes for all 6 UWIN regions. The third product is a set of regional estimates of changes in output, employment, and income for all 6 UWIN regions. The fourth and final product is a set of water consumption profiles for each of the 6 UWIN regions. It is hypothesized that the contractions in spending associated with water price increases will be more severe for low-income households and that these reductions in spending are likely to come from household budget categories that are considered necessities (housing and food) rather than luxury items (travel and leisure). It is also hypothesized that reductions in household spending by middle and high-income households as a response to rising water prices will be less severe and in luxury-item budget categories, rather than necessity-oriented budget

28

categories.

Supplemental Keywords

Water price, household income, household employment, business output, regional employment

Project description:

1. Objectives: The objectives of the study are:

Create a database of past, present, and future household water use/consumption and water prices

(WUWP).

Generate projections of water demand based on historical water use data, water supply information from

project A1-1, and the water consumption profiles generated in objective 1.

Analyze past and present (baseline) water prices to generate estimates of future water prices.

Assess the economic impacts of water price increases and decreases on regional economies. This includes

businesses, government, and households of varying socio-economic strata.

Produce information about the economic impacts (in terms of jobs, output, and income) associated with

price-based as well as non-price-based regulation strategies (i.e. water rationing).

Create water consumption profiles of UWIN regions using information from this project and other UWIN

projects based on climate change scenario information, demographic, urban development scenario data.

These profiles will include information about pressures (population growth, water use per capita,

and water prices) as well as relevant indicators including drought propensity, water

demand levels, and urban water sustainability programs.

Generate a topology of UWIN regions, which will incorporate information from other UWIN projects,

that will be based on the water consumption profiles generated in objective 6 and the economic impacts

associated with policy mechanisms for reducing water consumption generated in objectives 4 and 5.

Use information about demographic profiles and water use to collaboratively generate potential green

infrastructure and water conservation solutions for each UWIN region.

Hypotheses:

Household employment will decline as a result of increasing water prices.

The decline in household employment as a result of higher water prices will be greater for low income than for middle and higher income households.

There are regional variations in the economic impacts of rising water prices.

Household water consumption will be greater in dry/arid regions than non-arid regions.

Regions with larger populations of young people will have higher rates of water consumption than region with older population profiles.

2. Intellectual Merit The research and policy communities are aware of water pressures and their impact on water supply. Infrequently however is the interplay between water supply and consumption analyzed in an integrated and comprehensive quantitative framework that considers the impacts and feedbacks of water consumption changes on people, businesses, and government entities. This lack of integration hinders our ability to understand the impacts of various mechanisms for managing water consumption. This project will develop a range of economic impact scenarios (past, present, and future) that characterize a range of impacts (in terms of jobs, income, and output) with various water consumption scenarios. This approach is novel for its integrated approach that incorporate information about

29

climate change assessments, household survey assessments, and input-output data to analyze a range of water supply-demand scenarios. A particular focus of this project is a comparative assessment of the equity implications of price-based strategies to manage water consumption compared to regulation-oriented management strategies. This will provide important information about the economic inequalities associated with price-based management strategies with a focus on low-income households who are likely to see a disproportionate percentage of their income be allocated to pay for water price increases. 3. Approach/Activities:

Research Design:

Construct a database of water use/consumption and water prices (WUWP) which will be derived from secondary data sources. The WUWP database will provide the necessary information to generate time series projections of water demand and water prices to derive the amount of money spent by households on water on an annual basis. It will also be used to derive water consumption profiles for each of the 6 UWIN regions. Estimates of the amount of money spent on water will serve as input information to the social accounting analysis for the Colorado Front Range and South Florida, as well as the more basic input-output analysis that will be used to characterize economic impacts for all 6 UWIN regions in terms of jobs, output, and income. The WUWP database and the results of the economic impact analysis will be used to derive typologies of all 6 UWIN regions based on the economic impact results and their water consumption profiles. Archival research and interactions with stakeholders will serve as the data sources to build the water use and water price (WUWP) database. Time-series econometric techniques will be used to generate projections of demographics, population, water use, and water prices. These projections will provide comparative information about future water use and water price scenarios, as well as the role of demography and population growth projections in driving these scenarios. Water use and water price information will serve as data inputs for social accounting and input-output analyses. Input output analysis is a systems-based approach that uses purchases and sales from businesses, households, and government to estimate growth or decline in three key variables: output, income, and employment. Social accounting analysis represents a more fine-grained level of input-output analysis because it breaks out the household and government rows in input-output tables to provide more fine-grained resolution on the types of households and government actors. This is particularly critical for the evaluation of the impact of water-price increases on low-income households. In standard input-output tables, households are not broken down by income category and thus, the ability to make statements about impacts on low-income households is lost. The social accounting approach also provides the resolution on government activities that permit an evaluation of regulatory attempts to reduce water use on businesses, households, and regional output, income and employment. Geographic information systems (GIS) will be used to identify locales within all 6 UWIN study regions that may be driving demographic and population trends, and that may also suffer from equity issues, as regards water affordability.

Geographic Extent:

Projections of water use and water prices will be estimated and incorporated in water consumption profiles created for all 6 UWIN regions.

Regional economic impacts of water prices increases on employment, outcome, and income will be produced for all 6 UWIN regions.

The social accounting analysis will only be performed for the South Florida and Colorado Front Range regions.

Comparisons of BEA-based impacts and IMPLAN impacts will be made for South Florida and the Colorado

30

Front Range regions.

Locales within all 6 UWIN regions that may suffer from equity issues will be generated using Census block group data.

Temporal Extent of the Project: The goal of this project is to characterize water demand, water prices, and the economic impacts associated with water use in the past, in the present, and in future years. Historical secondary data is available from IMPLAN for the years 2000 to 2011. The cost of purchasing these datasets was included in the original UWIN budget. The projections of water consumption and water use will be made for the next 5 years on an annual basis.

Datasets:

Water use (USGS; EPA);

Water prices (Consumer Expenditure Survey from the Bureau of Labor Statistics);

Household spending (Statistical Abstract of the United States from the U.S. Census Bureau)

Annual population estimates (Bureau of Economic Analysis)

Demographic information (U.S. Census Bureau);

Urban heat island impacts and regional climate (project A2-2),

National input-output tables (Bureau of Economic Analysis)

County input-output information (IMPLAN)

Contingency for obtaining information about obtaining water prices:

o Call local utility companies to get cost estimates for houses of different sizes (one-bedroom apartment, houses of various square footages).

o Include a question about cost of water in household surveys. o Contact Water Resource Departments in Arizona and other locations decided upon once research

funded. The following article demonstrates the type of detailed information these departments will have (http://cronkitenewsonline.com/2014/04/could-price-be-a-tool-for-encouraging-water-conservation-in-arizona/)

Examples of places that will be contacted in Arizona include:

Arizona Department of Water Resources

Phoenix Water Services Department

Scottsdale Water Resources Department

Municipal Utilities

Preliminary Findings from Expenditures Information Based: Bureau of Labor Statistics Consumer Expenditure Survey 2011

Findings:

People spend more on water as they age

More educated people spend more money on water

Households with older children spend more on water

Higher income households spend more money on water

Race: White and other races spend the most on water, African American the least. Hispanic is in the middle

Homeowners with mortgages spend the most money on water

Suburban households spend the most on water followed central city and then rural households

The more income earners in a household the more money spent on water

Managers and professionals spend the most money on water followed by retirees

31

Regions that spend the most on water in order of most to least spent: West, South, Midwest, and Northeast

Input from UWIN Projects:

Water supply scenarios generated in project A1-1. This is a critical item that will be needed as soon as it is

reasonable for it to be produced.

Urban heat island impacts and regional climate (project A2-2), this information will feed into the water

consumption profiles.

Project C-2 Stakeholder engagement information about population trends, water prices, and water

consumption trends. This information is needed to help validate projected water price and water

consumption projections.

Information about the biodiversity of UWIN regions will be incorporated in water consumption profiles

from project A3-2. This will be needed towards the end of the project in year 3.

Information from projects B1-2 about the willingness of households to make investments in water

solutions will prove valuable in the collaborative generation of strategies to mitigate water consumption in

each of the UWIN regions.

Forecasts of the adoption of new technologies for water management from project C2-1 are also critical to

designing collaborative solutions to mitigate water consumption.

4. Expected Results, Benefits, Output, and Outcomes:

Outputs

A diverse range of outputs will be produced from this project. These outputs include: 1. UWIN regional consumption profile: will include pressures and relevant indicators for understanding

dynamics of water consumption and water prices 2. A typology of water consumption regions 3. Estimates of the economic impacts of water price increases on households of varying income strata for

South Florida and the Colorado Front Range 4. Heat maps of locales within all 6 UWIN regions where water equity is a concern from an economic

perspectives 5. Economic impacts in terms of jobs/income/output for all 6 UWIN regions of water price increases

Outcomes

1. Help people understand how changes in water price impact household income. 2. Help stakeholders understand where vulnerable populations to rising water prices are located within the 6

UWIN regions. 3. Help people understand the dynamics of water consumption-who are the primary consumers of water and

what is their willingness to pay for this consumption? 4. Help people understand how low-income households are impacted by rising water prices. 6. Help the research team and stakeholders understand regional difference in economic impacts associated

with rising water prices. 7. Help the research team and stakeholders understand how a variety of regional characteristics impact water

demand and prices across all 6 UWIN regions.

Contributions in terms of Pressures/Indicators/Solutions (UWIN Framework):

Pressures: Population growth, demographic change, public/private financing constraints Indicators: water use per capita, water demand

32

Solutions: Beyond price-based mechanism for reducing water consumption, this project will focus on crafting water conservation solutions to reduce household demand for water, as well as means to influence consumer behavior through education efforts and perhaps incentives to impact water use.

How can this tie into engagement and educational opportunities:

Educational opportunities Mentoring of undergraduate students via the UWIN summer research experiences for undergraduates (REU) program, the development of a module for the urban water sustainability massive open online course (MOOC), the mentoring and preparation of graduate students. Specifically, a MOOC module on water demand and water conservation will be developed. This module will incorporate information from the research project regarding household pressures on water resources, as well as other regional characteristics (demographics and population growth) that increase the demand for scarce water resources. Graduate students will be mentored and trained in input-output modeling, forecasting techniques, and urban applications of geographic information systems (GIS). A summer REU experience could involve undergraduate work on the water consumption profiles and GIS work to identify vulnerable households to price shocks in UWIN communities. Dr. Mack, will work with the Charles Glass (Howard University) to recruit undergraduate students for summer REU opportunities and also graduate students for research assistantships associated with the project. Stakeholder engagement This project would like to work with folks from Theme C to participate in stakeholder engagements efforts. This participation would three-fold. One, it would solicit information from stakeholders about water pressures and water prices in UWIN regions. Two, Dr. Mack would like to share the results of the economic impact, social accounting analysis, and water consumption profiles with stakeholders to receive feedback. Three, this project would also like to work with stakeholders to provide solutions that reduce household water consumption via equitable means. No plans for direct engagement with stakeholders but would like to coordinate with Gary Pivo and other folks interacting with large water groups to get their sense of what will happen to water prices and what they feel are the best mechanism for managing water consumption (price or regulation). Would like to work with folks in other UWIN regions to see if they have access to local information about water use and water prices. Would be willing to work with the LTER folks at Arizona State University and University of Maryland Baltimore for help with data collection. 5. General Project Information:

Facilities

The facilities at Arizona State University and Michigan State University are suitable for carrying out the research outlined above.

Personnel Expertise

Elizabeth Mack has received formal training in input-output modeling and has taught courses on this technique and social accounting analysis at the graduate level. She also has training in and use econometric models and geographic information systems (GIS). From her work in economic development prior to becoming an academic, Dr. Mack also has experience working with Chambers of Commerce and other community groups to solicit information for database building purposes.

Target Dates

Year 1 Fall: Collect water consumption and water price information for the Water Use Water Price (WUWP) database/Solicit data and inputs from UWIN projects and stakeholders Year 1 Spring: Begin generating projections of water use and water prices for all UWIN regions; Share information

33

with stakeholders and get feedback Year 1 Summer: Finish water use and water prices projections. Prepare a journal article for scholarly publication Year 2 Fall: Using information about water prices and consumption, use IMPLAN data to perform a social accounting analysis for select UWIN regions (South Florida and the Colorado Front Range), begin collection of national I-O information for other 4 UWIN regions Year 2 Spring: Estimate projections of impacts (jobs, output, and income) for all 6 UWIN regions for select years (BEA has data for every 5 years) Year 2 Summer: Compare IMPLAN and BEA results for South Florida and Colorado Front Range. Prepare a report and scholarly journal article. Year 3 Fall: Collect remaining data for UWIN regions for water consumption profiles and analyze these data to create water consumption profiles Year 3 Spring: Generate typology of UWIN regions and share water consumption profiles and typologies with UWIN team and stakeholders to get feedback. Brainstorm with these folks as well on proposed solutions. Year 3 Summer: Propose solutions based on profiles and prepare a scholarly article for publication. Prepare a report, based on profiles for UWIN team and stakeholders in each region.

Project Needs

A lot of information will be needed for the proposed research. This may involve getting information from local water providers. Help will be asked from UWIN project D1-2 in locating this information. Extensive efforts have been made to identifying existing source of information, but local information from team members could improve the quality of the database and projections that will be produced in this project. Help will also be needed from Gary Pivo to solicit information and feedback from stakeholders.

34

Project A2-1 Research Plan Project Title Land-atmosphere-hydrosphere interactions in urban terrain UWIN Project number A2-1 Project Lead Elie Bou-Zeid, Princeton Investigators/Institutions (Funded/not funded to work on project)

Claire Welty, UMBC (funded to work on project) Zhihua Wang, ASU (funded to work on project) Matei Georgescu, ASU (Collaborating on this project, mainly funded for project A2.2) Scott Denning, CSU (Collaborating on this project, mainly funded for project A2.2) Forrest Meggers, Princeton (Collaborating on this project, mainly funded for project A2.4) Jim Smith, Princeton (Collaborating on this project, mainly funded for project B2.2)

Project Period 8/1/15 – 7/31/20 Project Cost Will be determined by Dec 31, 2015. Graduate students funded to work on project

Matthew Schley, PhD student (UMBC), supervised by Claire Welty Qi Li, PhD student (Princeton), supervised by Elie Bou-Zeid Hamidreza Omidvar, PhD student (Princeton), supervised by Elie Bou-Zeid TBA, post-doc (Princeton), supervised by Elie Bou-Zeid Chenghao Wang, PhD student (ASU), supervised by Zhihua Wang

Project Overview

Around 3.65 billion people now live in urban areas, accounting for 50% of the world population. By 2050, the number will increase to 6.25 billion, about 67% of the world population at that time (United Nations 2012). When considering economic output and energy use, the importance of cities is even greater: they contribute about 80% of the economic output of the world and account for about 70% of its energy consumption (The Global Commission on the Economy and Climate 2014). Reducing the energy and water needs of cities, understanding their environmental footprint, and mitigating their vulnerability to climate change will therefore be a central challenge for humanity in the coming decades. At present, most urbanization research focuses on individual pieces of this challenge (e.g., building energy consumption, green infrastructure), partly because we do not yet understand all these individual pieces and partly because we lack robust comprehensive and integrated models for cities. The aim of project A2.1 is to integrate the wide range of expertise in the UWIN SRN in order to stimulate groundbreaking advances in our ability to model individual components of the urban system, as well as the integrated system itself. To that end, we will couple existing models of the atmosphere, surface and subsurface in urban terrain into a comprehensive modeling framework. Furthermore, we will simultaneously develop and expand these individual models to more accurately describe the processes they already capture, as well as to allow them to capture additional processes that are central to the urban water challenge, and how it interacts with the energy and climate challenges. More specifically, we will couple the Weather Research and Forecasting (WRF) atmospheric model, the urbanized Noah land surface model where an urban canopy model (UCM) as a special representation for urban terrain, and the ParFlow subsurface flow model. The UCM will be the component where most of the urban processes reside; for example, it is the model layer where urban radiative trapping, anthropogenic heat emissions, and green infrastructure are represented. Through this project, we will expand the UCM into a more general urban model (though not a fully-encompassing one yet) so that it can dynamically represent the interplay between urban water, energy and climate. This will entail adding various features to the model, including but not limited to: i) representation of urban trees and green infrastructure; ii) simple models of urban water demand and how it interacts with energy demand and urban microclimatological change; and iii) urban thermal comfort. The expanded and coupled modeling framework will be validated against a heterogeneous and expansive range of observations, and will then be applied to understand historic extreme events (e.g., heat waves, floods, droughts) and how they influenced the 6 metropolitan regions covered by the SRN, as well to analyze the impact of future extreme events under project A 2.2.

35

Project Summary

Objectives:

The overarching objective of this project is to develop the next-generation urban modeling system by integrating and further developing multiple state-of-the-art components. These components are existing models that have been previously developed by participating PIs to represent the atmospheric (WRF), surface (Urban Canopy Models-UCMs), and subsurface (ParFlow) urban environments. Through this project we will fully coupled them (yielding the first complete urban hydrological/meteorological model), as well as intensively develop their urban representations to capture important processes and indicators such as i) the water demand and cooling impact of canyon trees; ii) total urban water demand and how it is modulated by weather; iii) the complex water-climate-energy repercussions of green infrastructure systems (urban irrigation, rainwater tanks, biofiltration systems, green roofs, etc.), and iv) thermal comfort in the urban outdoor space.

Experimental approach:

The models we will develop and couple are the Weather Research and Forecasting (WRF) model for the atmosphere, an advanced urban canopy model (UCM) embedded in the Noah land surface model, and the ParFlow model for the subsurface. The existing version of the models will be coupled and in parallel the new capabilities described above (under objectives) will be embedded in the UCM. The UCM will be applied offline (uncoupled from WRF and ParFlow) and subsequently online as its new capabilities are ported into the coupled modeling framework. Various existing datasets will be used for validation. The simulations will address historic as well as future (the latter through project A2.2) periods, focusing on extreme conditions such as heat waves, floods, and droughts.

Expected results:

Expected outputs include a coupled ParFlow-UCM-WRF model suitable for urban areas, advancing the scope of the UCM to produce a wider and more general urban model, and applications of the model to the 6 metro regions (but focusing on subsurface coupling only in Baltimore, Denver, and Portland) that would produce key datasets featuring the spatio-temporal variability of multiple indicators of interest to UWIN. The outcomes include an improved ability to model the urban environment to be used in project A2.2, better understanding of the climate-water-energy nexus in cities, and an advanced simulation tool to assess options for mitigating water use and climate impacts in urban terrain.

Supplemental Keywords

Urban Micro climate, Urban Hydrology, Urban Canopy

Project description

1. Objectives The overarching objective of this project is to develop the next-generation urban modeling system by integrating and further developing multiple state-of-the-art components. These components are existing models that have been previously developed by participating PIs to represent the atmospheric (WRF), surface (Urban Canopy Models - UCMs - embedded in Noah), and subsurface (ParFlow) urban environments. But their coupling is not yet achieved and they currently lack representations of many urban water systems that are critical for understanding water challenges and how they interface with energy and climate challenges in cities. The PIs will develop the needed representations of urban water systems, integrate all the components into a unified modeling framework, and evaluate it across various the 6 ecohydrological regions cover by UWIN. The overarching science question we aim to answer is: How do subsurface-surface-atmosphere connections vary across urban regions and during climatic extremes, and what are the implications for urban water sustainability? The modeling framework developed here will also be the basis for the expanded assessment of future climate change impacts on urban water sustainability that will be carried out under project A2.2 (note that the resources listed under personnel and cost on

36

page 1 represent about 2/3rd of the resources allocated to the PIs, with the remaining 1/3rd of the resources (same personnel) contributing towards project A2.2). 2. Intellectual Merit The intellectual merit of this project derives from the “research network” it creates, in the spirit of the larger SRN. The participating PIs have in the past worked on developing improved understanding and modeling capabilities for the various components of the urban hydrologic cycle, energy consumption, and climate. Through this project, these past developments will be integrated and further advanced into a unified modeling framework that can accurately hindcast and forecast the urban climate, and equally importantly how this climate influences and is influenced by urban water and energy use. The efforts can be broadly divided into two main activities: 1) coupling of the ParFlow and the WRF-UCM models, and 2) development of the UCM into a broader and more general urban model (which can be run coupled to WRF (online) as well as in stand-alone mode (offline)) that can simulate not only the urban surface energy budget, but also urban energy and water use and how they interface with the urban microclimate. For activity 1, we will first aim to couple the urbanized hydrological models for the subsurface (ParFlow), surface (Noah+UCM) and atmosphere (WRF)). The coupling will results in the first complete modeling framework that accurately captures the urban environment (coupling over natural terrain has been done). This framework will be validated against various existing observational data sets and will constitute an innovation in and of itself, and will be made available to the wider community to use. Subsequently, we are interested in running the coupled model for periods and in zones where the water table is not too shallow or too deep, such that interactions of atmospheric processes with groundwater processes is expected to time-varying and non-trivial. For this reason, we chose three regions of focus, Baltimore, Denver, and Portland, owing to their climatic variability and the assumed importance of groundwater interaction with the atmosphere in these locations. Historic simulations (for validation, development of baseline, and analysis of past extremes) would be under this project A2.1, while future simulations would fall under project A2.2. For activity 2, we will develop in the UCM simple models for urban water/hydrological and energy systems, including green infrastructure. A version of the energy component was already developed (Salamanca et. al., 2010), we will here integrate it into our UCM. But the main effort will be to develop and integrate a comparable model for water/hydrological systems that captures: i) the water demand and cooling impact of canyon trees, ii) total urban water demand and how it is modulated by weather (increased irrigation during drought for example), iii) the complex water-climate-energy repercussions of green infrastructure systems (urban irrigation, rainwater tanks, biofiltration systems, green roofs, as well as novel ideas developed through project A2.4 led by Forrest Meggers), and iv) thermal comfort in the urban outdoor space. The PIs have already developed and implemented in WRF the most sophisticated urban hydrological model, but the advances needed to robustly address the challenges the SRN aims to tackle remain significant and overcoming these challenges will be a core aim of this project. These advances not only include the representation of hydrologic processes (i to iv above), but also associated model features such as improved parameterizations of surface-air exchanges of heat and humidity and the influence of rainfall on heat transport and temperature in cities. The advanced WRF-UCM will be applied over the 6 metropolitan regions (recall that with ParFlow the applications will be limited to 3 regions). As with WRF-UCM-ParFlow, we will conduct historic simulations as well as future ones, the later fall under project A2.2 and are detailed in the proposal of that project. Evaluation of the integrated WRF-UCM-ParFlow system will be made for simulations against suitable datasets from ground-based and remotely-sensed observations. Sample datasets include surface energy fluxes by eddy covariance (EC) towers from the AmeriFlux network (http://ameriflux.ornl.gov) (Baldocchi et al., 2001), rainfall fields generated in project B2.2 led by Jim Smith, atmospheric boundary-layer dynamics (e.g. boundary layer height, and vertical profiles of temperature/humidity/wind distribution) from the Atmospheric Radiation Measurement (ARM) Climate Research Facility and from the Aircraft Communications Addressing and Reporting System (ACARS), gridded temperature measurement by the Global Historical Climatology Network (GHCN) database and from satellite products such as MODIS and LANDSAT, and USGS streamflow and groundwater observations (http://waterdata.usgs.gov/nwis).

37

For both activities, after model development and validation for historic periods, we will apply the model to better understand subsurface-surface-atmosphere connections in urban regions, how natural (hydrologic) and engineered urban water systems are influenced by past and future climatic extremes, and how the adverse impact of these extremes can be mitigated to yield more sustainable and resilient cities. These goals will aim to aid in the development, in close collaboration with the other projects, of the blueprint for sustainable cities. Examples of the kind of specific questions we wish to address are:

(1) How does the full coupling of WRF-UCM-ParFlow affect hydrological and climatological predictions in urban terrain?

(2) In dry periods, urban vegetation become water stressed and needs irrigation, which places additional demand on water supply. What is the likelihood of such dry periods occurring at present and how will this likelihood be influenced by future climate change?

(3) How do climatic and energy benefits of green infrastructure facilities balance against their potential water requirements across the six metro regions the SRN studies?

(4) What is the impact of impervious surface area and green infrastructure on the water table location and the vulnerability of urban areas to extreme heat and drought?

(5) How much urban water use variability can be attributed to climatic variability and how can water demand increases associated with climate extremes be reduced?

3. Approach/Activities The standard model used by the National Weather Service (NWS) to produce temperature forecasts in the US is the Weather Research and Forecasting (WRF) model, developed through a collaboration centered at the National Center for Atmospheric Research (NCAR) as an operational and research model. WRF is a community model and has become the leading simulation tool used in weather research worldwide, and hence a large number of studies to develop better urban microclimate models have focused on the coupling of these models within WRF. The research team working on this project has a track record in this field: we have developed the new-generation Princeton Urban Canopy Model (PUCM), which includes better representations of urban surface heterogeneity and hydrology (Wang et al., 2012), and subsequently embedded the basic PUCM (Li and Bou-Zeid, 2013; 2014) and more advanced versions (Yang et al., 2015) in WRF. The faithful representation of urban surface physics in WRF-PUCM significantly improved forecasting of urban temperatures in previous studies over the Baltimore-DC metropolitan area, compared to the default single-layer UCM, which over-predicted urban temperatures (Li and Bou-Zeid, 2013; 2014). In this project, we will use the “Advanced Research” dynamics solver of WRF-ARW version 3.7, in a one-way nested mode. ARW solves the non-hydrostatic, fully compressible moist Euler equations for the three-dimensional velocity components, the potential temperature, the geopotential, the inverse density, and the surface pressure. Additional equations can be solved for turbulent kinetic energy or other scalars. The effects of turbulent mixing, spherical projection and earth curvature, earth rotation and physical forcing terms are included in the equations. Time integration is performed using 2nd or 3rd order Runge-Kutta methods; a distinct, reduced time step is used to solve for the high-frequency acoustic modes while a larger time-step is used for the meteorologically significant modes to reduce computational cost. Spatial discretization schemes for the advective terms with accuracies up to the sixth order are available. Grid stretching is allowed on the vertical grid, which uses terrain-following coordinates. The model is initialized using three-dimensional data from reanalysis products or future climate models (which we will obtain in this study from the North American Regional Reanalysis (NARR) historic climatic data of the National Centers for Environmental Prediction (NCEP) and from future climate simulations from various models). Surface information for NOAH, the surface component in which the UCM is embedded, is obtained from the U.S. Geological Survey (USGS) databases: National Land Cover Data (NLCD) for land use and the Shuttle Radar Topography Mission (SRTM) for topography. But for future simulations we will modify these land data to reflect urban growth as detailed in the proposal of project A2.2. Simulation periods will be 2 to 12 week periods selected for

38

the extreme conditions they exhibit (as described before, but notice that longer spin-up times are needed for ParFlow) and the domains will be nested such that the finest WRF domains will have a horizontal resolution of 1 km and cover an area of about 160 km x160 km around the urban areas of interest. ParFlow and the UCM do not need to have the same resolution or domains extent as WRF, and will in fact be run with smaller resolutions. The baseline UCM we will start with is broadly based on the energy exchange scheme described by Kusaka et. al. (2001). PIs Bou-Zeid and Wang have jointly and separately developed it to improve over conventional UCMs by its capability to distinguish various urban materials (concrete, asphalt, grass, brick, gravel etc.) on each urban facet (ground, roof and walls), and by the detailed hydrological module implemented to represent water fluxes and storage (Wang et al., 2012; Ramamurthy and Bou-Zeid, 2014). The improved UCM continues to capture complex processes such as radiative trapping, heat conduction in soil and impervious surfaces, and turbulent exchanges of heat between the surface and the atmosphere as do other UCMs. However, these exchanges are now computed for each sub-facet, consisting of a different material, of a given urban facet. The UCM also models evaporation from soil and vegetated surfaces by numerically solving the one-dimensional Richards equation in a multi-layer soil matrix to obtain surface soil moisture, which is then used to model evaporation using turbulent transfer functions and atmospheric specific humidity (constrained by available energy). Potential evaporation is first computed and then actual evaporation is reduced from its potential rate using a reduction factor based on the moisture in the top layer for bare soils, while a stomatal resistance is added (to the aerodynamic resistance represented by the turbulent transfer functions) for vegetation. Impervious surfaces are assumed to have a variable water storage, capped by a fixed upper limit capacity (bucket model); the storage is replenished by precipitation and depleted by evaporation at a rate reduced from the potential evaporation by a

factor βI = current storage / maximum storage. The model, when used offline, is forced by precipitation and hydrometeorological measurements (e.g. atmospheric specific humidity). The UCM currently does not account for anthropogenic heat sources, anthropogenic water sources, heat advection due to rainfall/runoff, and has a limited representation of green infrastructure and a simplistic parameterization for turbulent transfer functions. To develop it into a tool that can be used the tackle the complex scientific questions targeted by the SRN, further development will be performed and validated in this project (points i to iv listed under activity 2 above, as well as the implementation in the UCM of new turbulent exchange schemes for heat and water vapor in the UCM based on new computational fluid dynamics results recently developed by Ph.D. student Qi Li in PI Bou-Zeid’s lab through a different project, and a parameterization of heat advection with rainfall/runoff through results obtained by Ph.D. student Hamidreza Omidvar also from a separate project of PIs Bou-Zeid and Wang). We will integrate WRF-UCM with the hydrologic model ParFlow. ParFlow is a 3D, fully coupled, distributed surface-subsurface flow model originally developed at Lawrence Livermore National Laboratory and now mainly maintained by R. M. Maxwell at Colorado School of Mines (Ashby and Falgout, 1996; Jones and Woodward, 2001; Kollet and Maxwell, 2006). There is no one “community” groundwater model in current use, but the ParFlow user community continues to grow. ParFlow solves transient, variably saturated flow using a parallel, globalized Newton method coupled to a multigrid-preconditioned linear solver. The model computes the pressure of the water in the subsurface and resulting saturation field over a chosen time step, given initial and boundary conditions (specified as pressure or flux of water). Saturation-pressure and relative permeability-saturation functions are represented by van Genuchten relationships (van Genuchten, 1980). The model includes surface flow in the form of a free-surface overland flow boundary condition consisting of the kinematic wave equation (Kollet and Maxwell, 2006). This equation is discretized using a finite control volume approach in space and an implicit backward Euler scheme in time. ParFlow has been integrated with the land surface model CLM (Dai et al., 2003), allowing interaction between subsurface soil moisture and simulated evapotranspiration. (Maxwell and Miller, 2005; Maxwell and Kollet, 2008). ParFlow has been further coupled to the Weather Research and Forecasting model (WRF) via the NOAH land surface model and tested on a limited basis over natural terrain (Maxwell et al., 2011). The coupled ParFlow-WRF

39

model has not been applied to urban terrains. We have applied ParFlow.CLM to the Baltimore metropolitan region (Bhaskar et al., 2015) as well as WRF separately to the same domain, but we have not coupled these models for application to this region. We propose to use the coupling framework of Maxwell et al. (2011), with modifications for the urban terrain as needed. We will carry out the Baltimore coupled modeling first, since we have previously modeled subcomponents in this region. A hydrogeology database for the Baltimore region was established for the previous ParFlow work. Hydrogeology data have been assembled for the Denver and Portland regions for MODFLOW models of these regions (personal communications, Mazdak Arabi and Roy Haggerty, August 2015); these data could be utilized for the ParFlow modeling effort. We would model the Denver and Portland regions subsequent to the Baltimore effort. We will engage stakeholders through the UWIN framework and activities being set up by Mike Sukop. We anticipate that managers from city departments of public works will be interested in the goals and outcomes of this project. 4. Expected Results, Benefits, Outputs, and Outcomes Expected outputs include: (1) building of a coupled ParFlow-UCM-WRF model suitable for urban areas; (2) application of the model to the 6 metro regions (but focusing on subsurface coupling only in Baltimore, Denver, and Portland); (3) running model scenarios laid out in the scientific questions desired to be answered. For the surface water/ groundwater component, model output will be in terms of pressure head and saturation as a function of gridded spatial scale and chosen time step; stream flow and water levels in aquifers can be derived from this output. For the atmospheric and land surface components, the outputs will consist of surface temperature and soil moisture maps, and near-surface air temperature, relative humidity, wind speed, as well as simple estimates of water and energy demands (focusing on how these demands are linked to the climatology). Outcomes include an improved ability to model the urban environment, which will be used in project A2.2, better understanding of the climate-water-energy nexus in cities, and an advanced simulation tool to assess options for mitigating water use and climate impacts in urban terrain. Blueprint indicators being addressed by this project are related to:

i) The water supply/watershed sector and include the water scarcity category (vulnerability, reliability, resiliency and water demand indicators).

ii) The climate sector and include precipitation-related and heat island and thermal indices. iii) The energy sector and include energy demand and GHG emission per capita

We expect multiple REUs to be involved in the work of this project. 5. General Project Information Welty’s research group has been collaborating with Reed Maxwell and using ParFlow at multiple scales for over 15 years. Bou-Zeid has extensive expertise in modification and application of WRF at a multiple scales, including rewriting WRF subroutines to make them more suitable for application to urban areas and over heterogeneous terrain. Wang has led experimental investigations and modeling developments as well. He focuses on green infrastructure and how to simulate it in UCMs. All PIs have access to parallel clusters at their respective universities on which these models will be run. In addition, Welty and Bou-Zeid routinely apply for and obtain computational time on NSF-supported supercomputers (e.g., Stampede and Yellowstone) for solving large problems. Because UWIN is a NSF project, there should be no problem with continuation of successful application for hours on these machines. Matthew Schley was hired by C. Welty in summer of 2015. He is currently in training to learn ParFlow, by applications to smaller domains supported on other projects. We anticipate that he will begin regional ParFlow

40

work on this project in late spring of 2015, starting with Baltimore. By late spring/early summer of 2015, E. Bou-Zeid plans to hire a new post-doc at Princeton who has expertise in WRF. That postdoc will focus on the development of the urban water use module and the urban tree representation and validation in WRF-UCM, as well as the implementation of the new turbulent transfer and rainfall/runoff heat advection schemes in collaboration with students Qi Li and Hamidreza Omidvar who developed these schemes. The postdoc will also work with Schley on the WRF-UCM-Parflow coupling. We anticipate sending Matt Schley to Princeton for some short period of time (2-6 weeks) to learn WRF from the Princeton postdoc. After that training period, Matt will couple WRF with ParFlow for the Baltimore region, under the guidance of E. Bou-Zeid and the post-doc. Then this expertise will be transferred to set up modeling of the Denver region and the Portland region. The student supervised by Zhihua Wang will focus on developing physical urban irrigation parameterization, the dynamic green infrastructure, and thermal comfort modules in the UCM, and the integration of all new development of the single layer UCM in WRF-ARW 3.7 (Yang et al., 2015). The student will be the bridge to pass these developments to the personnel at ASU who will use them in project A2.2. The UCM development will be a continuous process that is likely to continue into Year 4 of the project. But these developments will be “rolled-out” for use in Project A2.1 and A2.2 as they become available. We anticipate developing the urban trees and green infrastructure capabilities and their inclusion in WRF by the end of year 2 of the project (July 2017), while the development of the urban water use and thermal comfort modules will be completed by the middle of year 4 (Jan 2019). For the WRF-ParFlow coupling, it is anticipated that the model coupling and application to Baltimore will take 1.5 years of the project, and Denver and Portland will take the remaining 1.5 years of the 3-year project. This is a very ambitious schedule, but we only have 3 years of support for Matt. We anticipate staying in close contact with the other projects in this theme/thrust as needed, especially project A2.2. We will also collaborate with the investigators in project B2.2 of the use of the modeling framework to simulate rainfall fields. 6. References Ashby, S. F., and R. D. Falgout (1996), A parallel multigrid preconditioned conjugate gradient algorithm for groundwater flow simulations, Nucl. Sci. Eng., 124, 145–159. Baldocchi, D., E. Falge, L. H. Gu, R. Olson, D. Hollinger, S. Running, et al. (2001), FLUXNET: A new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities, Bulletin of the American Meteorological Society, 82(11), 2415-2434. Bhaskar, A. S., C. Welty, R. M. Maxwell, and A. J. Miller (2015), Untangling the effects of urban development on subsurface storage in Baltimore, Water Resour. Res., 51, 1158–1181, doi:10.1002/2014WR016039. Dai, Y., et al. (2003), The Common Land Model, Bull. Am. Meteorol. Soc., 84(8), 1013–1023, doi:10.1175/BAMS-84-8-1013. Jones, J. E., and C. S. Woodward (2001), Newton-Krylov-multigrid solvers for large-scale, highly heterogeneous, variably saturated flow problems, Adv. Water Resour., 24(7), 763–774, doi:10.1016/S0309–1708(00)00075-0. Kollet, S. J., and R. M. Maxwell (2006), Integrated surface-groundwater flow modeling: A free-surface overland flow boundary condition in a parallel groundwater flow model, Adv. Water Resour., 29(7), 945–958, doi:10.1016/j.advwatres.2005.08.006. Kollet, S. J., and R. M. Maxwell (2008), Capturing the influence of groundwater dynamics on land surface processes using an integrated, distributed watershed model, Water Resour. Res., 44, W02402, doi:10.1029/2007WR006004.

41

Kusaka, H., H. Kondo, Y. Kikegawa, and F. Kimura, 2001: A simple single-layer urban canopy model for atmospheric models: Comparison with multi-layer and slab models. Boundary-Layer Meteorology, 101, 329–358. Li, D., and E. Bou-Zeid, 2013: Synergistic Interactions between Urban Heat Islands and Heat Waves: The Impact in Cities Is Larger than the Sum of Its Parts. J. Appl. Meteor. Climatol, 52, 2051–2064, doi:10.1175/JAMC-D-13-02.1. Li, D., and E. Bou-Zeid, 2014: Quality and sensitivity of high-resolution numerical simulation of urban heat islands. Environmental Research Letters, 9, 055001, doi:10.1088/1748-9326/9/5/055001. Maxwell, R. M., and N. L. Miller (2005), Development of a coupled land surface and groundwater model, J. Hydrometeorol., 6(3), 233–247, doi:10.1175/JHM422.1. Maxwell, R., and S. Kollet (2008), Interdependence of groundwater dynamics and land-energy feedbacks under climate change, Nature Geoscience, 1(10), 665-669. Maxwell, R.M., Lundquist, J.K., Mirocha, J.D., Smith, S.G., Woodward, C.S. and Tompson, A.F.B. Development of a coupled groundwater-atmospheric model. Monthly Weather Review 139(1), 96-116, doi:10.1175/2010MWR3392, 2011. Ramamurthy, P., and E. Bou-Zeid, 2014: Contribution of Impervious Surfaces to Urban Evaporation. Water Resources Research, 50, 2889–2902, doi:10.1002/2013WR013909. Salamanca, F., A. Krpo, A. Martilli, and A. Clappier, 2010: A new building energy model coupled with an urban canopy parameterization for urban climate simulations-part I. formulation, verification, and sensitivity analysis of the model. Theor Appl Climatol, 99, 331–344, doi:DOI 10.1007/s00704-009-0142-9. The Global Commission on the Economy and Climate, 2014: Better Growth Better Climate: The New Climate Economy Report. World Resources Institute, 72 pp. United Nations, 2012: World Urbanization Prospects: The 2011 Revision. "United Nations' Department of Economic and Social Affairs - Population Division, New York, 1 pp. http://esa.un.org/unup/pdf/FINAL-FINAL_REPORT%20WUP2011_Annextables_01Aug2012_Final.pdf. van Genuchten, M. T. (1980), A Closed-form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils1, Soil Sci. Soc. Am. J., 44(5), 892-898. Wang, Z.-H., E. Bou-Zeid, and J. A. Smith, 2011: A Spatially-Analytical Scheme for Surface Temperatures and Conductive Heat Fluxes in Urban Canopy Models. Boundary-Layer Meteorology, 138, 171–193. Wang, Z.-H., E. Bou-Zeid, and J. A. Smith, 2013: A coupled energy transport and hydrological model for urban canopies evaluated using a wireless sensor network. Q J Roy Meteor Soc, 139, 1643–1657, doi:10.1002/qj.2032.

42

PROJECT A2-2 RESEARCH PLAN Project Title Land-atmosphere-hydrosphere interactions: projecting future environmental change

in urban areas UWIN Project number A2-2 Project Lead M. Georgescu/ASU Investigators/Institutions

Z. Wang/ASU (Funded) E. Bou-Zeid/PU (Funded) M. Moustaoui/ASU (Funded) A. Mahalov/ASU (Funded) C. Welty/UMBC (Funded) F. Salamanca/ASU (Not Funded) D. Sailor/ASU (starting Jan. 2016; Not Funded)

Project Period 8/1/2015-7/31/2020 (funding available for Years 1-3) Project Cost Will be determined by Dec 31, 2015. Graduate students funded to work on project

Postdoc TBD (6 Mo.: Years 1-3): Supervision, M. Georgescu Postdoc TBD (2.67 Mo.: Years 2-4): Supervision, E. Bou-Zeid Ph.D. student Chenghao Wang (4 Mo.: Years 1-3): Supervision, Z. Wang Ph.D. student Qi Li (4 Mo.: Years 1-3): Supervision, E. Bou-Zeid Ph.D. student Matt Schley (4 Mo.: Years 1-3): Supervision, C. Welty

Project Overview

This project will quantify hydroclimatic impacts due to the combined and interacting effects of urban expansion and greenhouse gas emissions for the end of century continental U.S. (CONUS). Scenario-based impervious surface expansion projections (as provided by the EPA Integrated Climate and Land-Use Scenarios [ICLUS]) will be used as surface boundary conditions within the Weather Research and Forecasting system (WRF) to approximate the uncertainty associated with future urban expansion. To include the effect of greenhouse gas emissions, we will make use of appropriately selected Global Climate Model (GCM) data, which will be used as initial and lateral boundary conditions for WRF. While initial hydroclimate assessment of decadal timescale simulations will be conducted at medium-range resolution (e.g., 20 km) for CONUS, in order to provide a broad representation of impacts at large-scales, dynamical downscaling at high-resolution (e.g., 1-2 km) will be conducted for each of the UWIN regions of interest to examine and quantify robustness of simulated results across multiple scales. Examination of simulations will focus on local to regional scale hydroclimatic effects (e.g., urban heat island [UHI] effect, regional water cycle) and consequences for energy demand associated with scenario-based climate projections across all UWIN regions. The implementation of designated urban adaptation and mitigation choices will be assessed, and cross-site dis-similarity, implications for tradeoffs, and potential co-benefits, will be quantified. Uncertainty quantification owing to projection differences in spatial distribution of emerging and expanding population centers (as provided by various ICLUS expansion projections) will be accounted for within the modeling framework. Finally, prioritization of geographically dependent urban adaptation strategies will be made.

Project Summary

Objectives: This project has two central objectives. The first objective is intended to quantify the dynamically interactive effect of increased emissions of greenhouse gases (GHGs) and anthropogenic landscape change associated with urban expansion for CONUS. This builds on previous work examining the impact of these two forcing agents, which highlighted the hydroclimatic significance of each for several megapolitan regions, while omitting interactions among the agents (Georgescu et al., 2014). This objective therefore fills an important research gap addressing potential nonlinear impacts associated with GHG-induced climate change simultaneously with urban expansion. The second objective is designed to examine the efficacy of locally deployed urban adaptation solutions through multi-scale numerical simulations with the enhanced urban land surface processes developed in Project A2.1 and implemented within the coupled urban-

43

atmosphere WRF platform. We posit that geographically suitable solutions (e.g., intended to reduce the UHI effect, with implications for changes in energy demand) should comprise a larger portfolio of choices (Georgescu et al., 2015), whose capacity can be enumerated through ensemble-based characterization of the relevant physical processes. Experimental Design: The experimental approach makes use of the advanced research version of WRF for all modeling simulations. WRF is a fully compressible and non-hydrostatic, open-source code that is maintained by the National Corporation for Atmospheric Research (Skamarock et al., 2008). It is a widely used tool, globally, with extensive utility ranging from urban to renewable energy applications. WRF has multiple parameterization options to represent the diversity of physics operating at a spectrum of scales (e.g., microphysics, convection). WRF includes a four-layer land surface model to represent energy and water transport within the soil column, as well as a widely used single layer urban canopy model that accounts for principal physical processes associated with absorption of incoming solar radiation and emission of outgoing longwave radiation (e.g., shadowing from and reflection of buildings due to canyons), anthropogenic heating, and biogeophysical representation of the built environment. Improvements in the hydrological representation of the urban canopy model implemented in Project A2.1 will be used throughout Project A2.2. To ensure consistency and dependability of model simulation results, numerical experiments will be conducted using the same version of WRF on an identical computing platform, for Project A2.1 and A2.2. Deliverables: The deliverables of the project directly tackle several of the blueprint indicators that serve the foundational premise of UWIN. The central outputs of the two principal objectives will inform the efficacy of simulated urban adaptation/mitigation choices, and the degree to which they can successfully reduce targeted pressures, their resilience in the context of changing external forces, and potential co-benefits and unintended consequences that require quantification prior to prioritization of solutions. A key question relates to how, or whether, judicious use of water (e.g., through incorporation of green infrastructure such as green roofs and walls) can be successful in reducing the UHI effect, quantifying this resultant impact on energy demand, with implications for GHG emissions, and importantly, how such targeted solutions may increase population resiliency through, for example, reduced heat related mortality and morbidity (therefore, this project serves as direct input to Project A2.3).

Supplemental Keywords

Urban climate, modeling and simulation, meteorology, surface energy balance, urban sustainability, urbanization, scenario assessment, cooling energy demand, megapolitan.

Project description

1. Objectives This project has two central objectives, with each containing targets to be met that establish the overall purpose of the Project.

a. Quantify the dynamically interactive effect of increased emissions of greenhouse gases (GHGs) and anthropogenic landscape change associated with urban expansion for CONUS.

i. Examine interaction effect of dual forcing agents, and associated non-linearities and land-atmosphere feedbacks.

ii. Calculate sensitivity of interaction effect, determined in i., to varying projections of urban expansion and GHG emissions.

iii. Quantify spatial variability of interaction effect, and its simulated uncertainty, across the six UWIN regions.

b. Examine the efficacy of locally deployed urban adaptation and mitigation solutions. i. Quantify efficacy of proposed urban adaptation and mitigation solutions on stress reduction of relevant

blueprint indicators (e.g., UHI, energy demand). ii. Establish impacts, unintended consequences, co-benefits, and uncertainty of varied solutions across the six

UWIN regions.

44

iii. Institute process of prioritization of locally appropriate, geographically dependent urban adaptation and mitigation solutions.

2. Intellectual Merit There are several aspects that comprise the intellectual merit of this project. First and foremost, the advanced coupled tools to be developed and incorporated into the next-generation of the NCAR-maintained WRF modeling system, shall be made available – through NCAR – to the entire global community. The incorporation of key physical processes that enable and promote use-inspired research emphasizing solutions-oriented approaches makes this tool widely applicable globally, especially for areas already facing or expected to face similar urban challenges in the future (e.g., Asia, Africa). Release of these codes, to be housed and maintained at NCAR, for public and scientific community use is a significant and distinct accomplishment as it pushes the boundaries of intellectual discovery through community distribution of state-of-the-art tools. Second, because the scientific questions posed are themselves linked across several Project Plans, connected projects can have substantially greater impact and lead to enhanced benefits, more than any single project could separately. For example, any distinct urban solution determined to reduce the UHI effect by some quantity is, in and of itself immaterial for societal relevance. The effect of UHI reduction on energy demand, and its consequence for lowering GHG emissions, with direct implications for heat-related mortality and morbidity, illustrates the critical importance of any single process omission that would prevent development of actionable and societally beneficial solutions. Finally, the intellectual approach utilized to initiate prioritization of locally appropriate and geographically relevant urban adaptation and mitigation solutions will provide guidance for other regions of the globe tasked with similar challenges. In other words, local solutions uncovered as part of this work are expected to have broad adoption for other similar (e.g., geographic) regions, and therefore have direct applicability elsewhere. 3. Approach/Activities

Advanced Models

The current inadequacy in representation of urban land surface processes in mesoscale models has resulted in a wide predictability gap, particularly at city and regional scales. To bridge the gap, in this project we will integrate new model development of the single layer urban canopy model (UCM) achieved under Project A2.1 into WRF, and use the enhanced WRF to simulate future urban environmental changes. The implementation of the new physical schemes into WRF-UCM enhances the representation of hydrological processes in urban areas, especially those due to human influence, and improves characterization of the water-energy cycle. Among the added physical processes that serve the basis of the enhanced WRF-UCM include the incorporation of: i) urban trees and their radiative shading effect, ii) total urban water demand modulated by weather, iii) representation of the water-climate-energy repercussions of green infrastructure system deployment (urban irrigation, rainwater tanks, biofiltration systems, green roofs, etc.), iv) measure of outdoor thermal comfort, and v) the coupling of the urban surface and atmosphere to the subsurface (via coupling with ParFlow). Each of these newly developed WRF-UCM capabilities, after thorough offline testing (i.e., uncoupled to the overlying atmosphere), will be implemented into the latest UCM in the public release of WRF-ARW v3.7 (Yang et al., 2015).

Evaluation of the integrated WRF-UCM-ParFlow system will be made against suitable datasets from ground-based remotely sensed observations, including: i) surface energy fluxes by eddy covariance (EC) towers from the AmeriFlux network (http://ameriflux.ornl.gov) (Baldocchi et al., 2001), ii) atmospheric boundary-layer dynamics (e.g. boundary layer height, and vertical profiles of temperature, wind, and humidity) from the Atmospheric Radiation Measurement (ARM) Climate Research Facility and from the Aircraft Communications Addressing and Reporting System (ACARS), and iii) gridded temperature from the Global Historical Climatology Network (GHCN) database and any existing satellite products (e.g., MODIS and Landsat), and iv) USGS streamflow and groundwater observations (http://waterdata.usgs.gov/nwis). The calibrated WRF-UCM (without ParFlow) will then be used for the simulation of future scenarios of urban environmental changes in all the six study regions of the UWIN project. The complete WRF-UCM-ParFlow system, however, will only be applied over three regions (Baltimore, Denver, and Portland; see rationale in Project A2.1 proposal). In both cases, we will force the model with data from future climate simulations (obtained from GCMs; see section below) and we will select specific periods from the climate models to downscale to

45

sub-1km resolution. Graduate students (advised by Z. Wang and C. Welty) and 1 Postdoc (advised by E. Bou-Zeid) will collaborate on several aspects of model coupling and development under Project A2.1. To bridge the developmental aspects from Project A2.1 with the applications elements central to Project A2.2, Z. Wang and his RA will devote portions of their time to lead the implementation aspects of the novel features of the UCM into the online WRF version. Simultaneously, C. Welty and E. Bou-Zeid, and their RAs and Postdocs will assume leadership roles in the development of the online-coupled WRF-UCM-ParFlow system and its application for future scenarios.

Selection of GCM Data for Dynamical downscaling with WRF

WRF CONUS ensemble-based simulations will be conducted at a decadal timescale corresponding to end of century urban expansion (see below) and large-scale climate conditions. The GCM selection criteria will be based on analysis of the Coupled Model Intercomparison Project 5 (CMIP5) dataset as well as statistically downscaled bias-corrected CMIP5 climate and hydrology projections for CONUS. Given the preponderance of available GCMs and Representative Concentration Pathways (RCPs)/emissions pathways available, it is impractical to conduct simulations for the entire range of models and available emissions scenarios. Consequently, four GCMs will be selected to provide initial and lateral boundary condition data (i.e., winds, moisture, geopotential, and temperature) to force WRF with. The first pair of GCMs will correspond to RCP 8.5 (i.e., assumption of greatest CO2-equivalent concentrations by the end of the twentieth century) and will be based on the hydroclimatic sensitivity to changing GHG concentrations (i.e., most sensitive and least sensitive) for CONUS. Similarly, the second pair of GCMs, based on RCP 2.6 (i.e., assumption of least CO2-equivalent concentrations by the end of the twentieth century) and will similarly be based on the hydroclimatic sensitivity to changing GHG concentrations (i.e., most sensitive and least sensitive) for CONUS. In this fashion, the representation of greatest GCM emissions uncertainty will be directly incorporated into the dynamical downscaling approach, allowing for greater fidelity of uncertainty quantification associated with this forcing mechanism. The precise definition of “most” and “least” sensitive can be sensitive to choice of parameter (e.g., temperature or precipitation), or statistical moment (e.g., mean or variance) and will be a critical component of GCM selection criteria. Additional uncertainty associated with model spin-up time (to be varied by model initial time) will also be accounted for. The next step, involving dynamical downscaling for each of the UWIN regions of interest, will be conducted in two stages. First, direct output from CONUS WRF simulations (driven by GCM output) for select case study periods (e.g., summers, or extreme heat events as determined by a shift in the probability density function of specified stressors [e.g., increased frequency of heat waves beyond a particular threshold]) will be used to initialize and force high-resolution simulation (e.g., 1-2 km) conducted with WRF. Alternatively, the utility of GCM output data, according to the criteria defined above will be directly incorporated into the high-resolution simulations for each of the six UWIN regions. The selected periods will therefore represent future average and extreme conditions (e.g., extreme temperature, precipitation, or drought), and simulations for these events will permit evaluation of urban adaptation and mitigation strategies on water and energy use in cities, on the urban microclimate, and on the terrestrial hydrologic cycle.

Integration of GCM Output into WRF

The integration of GCM output data into WRF can be accomplished in two ways. First, WRF simulations can be driven by initial and lateral boundary conditions input data from GCM data using global bias corrected climate model data generated from version 1 of NCAR’s Community Earth System Model (CESM; Hurrell et al. 2013, Bruyere et al., 2015). CESM is one of the state-of-the-art models that participated in the CMIP5 experiment. It has the ability to simulate global patterns of observed temperature and rainfall (Knuti et al, 2013). The CESM simulations used to construct the bias corrected data include both historical simulations and future projections. The future projections are generated under distinct RCPs, span a period up to 2100 (Moss et al., 2010) and are available from NCAR’s Research Data Archive (http://rda.ucar.edu/datasets/ds316.0). The data contains all the two- and three-dimensional variables required for the initial and boundary conditions to force

46

the WRF model. They are distributed on a horizontal grid with a grid spacing of approximately 1 degree and are interpolated to 26 pressure levels. The data are archived with a time frequency of six hours and are given in a binary format that is similar to the format used for the intermediate files in WRF. The WRF Preprocessing System will be adapted to ingest these data by skipping the “ungrib.exe” step in WPS and directly using these intermediate files. Alternatively, if the selection criteria establish that utility of NCAR’s CESM GCM should not be made, the above-described process is not applicable. The global data generated by the selected GCM which are provided on the model vertical levels and on the horizontal model grid can be interpolated directly to a mesoscale grid with vertical and horizontal resolutions that are closer to those desired for the WRF domain. Both interpolations will use cubic monotonic and shape preserving methods to avoid artificial oscillations in the interpolated fields. This is particularly critical for relative humidity as the high-order non-monotonic methods can introduce unphysical negative values for this variable. In addition, special attention is required for the extrapolation of the 3D fields (temperature in particular) simulated by the GCM to levels below the ground. This extrapolation is necessary since the three dimensional data ingested by WRF need to be distributed on constant pressure levels; and also because of possible mismatches between the topography field represented by the coarse resolution grid in the GCM, and the higher resolution topography used in the WRF grid. Thus, a grid point located below (above) the ground relative to the GCM may be above (below) the ground with respect to the WRF grid. The extrapolation of temperature below the ground is based on the moist

hydrostatic balance 𝛿𝛷

𝛿𝑝= −

𝑅𝑇𝑣

𝑝, where𝛷 is the geopotential, p is the pressure and 𝑇𝑣 is the virtual temperature.

M. Georgescu (and a Postdoc) and M. Moustaoui will collaborate on the selection and integration of GCM model output into WRF. In addition, M. Georgescu (and a Postdoc), M. Moustaoui, A. Mahalov and additional (unfunded) personnel will collaborate on the simulations focused on impacts of urban expansion based on available projections of urban growth (see below).

Datasets, Scenarios, and Simulations

GHG-induced climate change and land-use change are major drivers of global environmental change (Pielke et al., 2011). Interactions between climate and land-use change is a significant regional-scale modifier and it is impossible to comprehensively examine the impact of anthropogenic induced climate change without considering changes in land-use. The ICLUS (http://www2.epa.gov/iclus) project, developed at the U.S. EPA, has produced spatially explicit projections of population and urban land-use that are consistent with increases in GHGs. ICLUS utilizes social, economic, and demographic narratives consistent with large-scale climate change that have been adapted for CONUS. Variables of direct utility for assessment of the interactive impacts associated with increasing GHGs and urban expansion include impervious surface fraction, housing density (ranging from rural, to exurban, to suburban, and finally to urban), for CONUS. These projections are available every decade through the end of the century, at 1 ha resolution. A method used to incorporate urban expansion projections at the proposed medium-range (e.g., 20 km; Georgescu et al., 2014) and at high spatial resolution (Georgescu, 2015) on to the WRF-UCM urban landscape configuration (which utilizes three density-dependent classes of urban land use and land cover) has been established. ICLUS scenarios A2 and B1 are consistent with large-scale emissions of the greatest CO2-equivalent concentrations and least CO2-equivalent concentrations, respectively, by the end of the twentieth century. The pair of land use urban expansion scenarios will be used as the basis of scenario-based impervious surface expansion projections, and will be used in conjunction with GCM output used to drive WRF and account for the interactive nature of urban dynamics and greenhouse gas induced climate change. The set of numerical experiments will include:

i. CONUS decadal timescale simulations explicitly incorporating RCP-based representation of GHG-induced climate change, and ICLUS-derived urban expansion scenarios consistent with large-scale climate change storylines.

ii. Nested grid, high-resolution (e.g., 1-2km) simulations for each of the six UWIN regions of interest, with WRF driven by appropriately selected GCM output data (see Selection of GCM Data for Dynamical

47

downscaling with WRF section). iii. As i., above, but with implementation of urban adaptation and mitigation strategies (e.g., the range of

strategies to be investigated includes green roofs and walls, variable irrigation amount for the maintenance and subsistence of biomass growth [based on appropriate plant physiology literature], water tanks, surface relative to roof-level adaptation and efficacy therein, as well as improved technological efficiency of electricity demanding systems (e.g., air conditioners, whose use may become more widespread in the future, and therefore increase GHG emissions), with direct and immediate impact on both the UHI and GHG emissions. Finally, simulations examining the relative fraction of assumed deployment will also be conducted, as additive percentages of deployment cannot be assumed to have linear effects (e.g., 50% deployment of a given strategy cannot be assumed to have twice the impact as 25% deployment of said strategy).

iv. As ii., above, but with implementation of urban adaptation and mitigation strategies (as discussed in iii., above).

4. Expected Results, Benefits, Outputs, and Outcomes The expected results are intended to quantify stressor impact(s) due to global and local scale environmental change while simultaneously addressing the scale of opportunity(ies) associated with available and innovative solutions-oriented adoptions aimed at stressor(s) reduction. Targeted solutions that inform the Urban Water Sustainability Blueprint include:

i. Reduction of the UHI Index (through a variety of scale-variant adaptation and mitigation strategies) ii. Reduction of Energy Demand (resulting from a reduced UHI, lifestyle choices due to decreased anthropogenic

heating of the urban environment) iii. Reduction of GHG emissions associated with technological improvements of electricity demanding systems

due to enhanced efficiency (e.g., Air conditioning systems). Participants necessarily communicating and interacting on the varied ways of approaching large-scale problems, such as this, demand interdisciplinary collaboration that will be of essential value to each of the investigators on the project. Importantly, methods of efficient and effective integration of unique ways of thinking can be elaborated to the broader scientific community, outside of UWIN, as an exemplar of a contemporary approach to science that is solutions-oriented, requires significant cooperation, and provides mutually beneficial partnerships while providing societal value. As a central component of this key outcome highlighting successful interdisciplinary scholarship, the training of undergraduate students through research and mentorship activities made possible by the UWIN Research Experience for Undergraduates (REU) program is anticipated. From an intellectual and community perspective, solutions to existing urban challenges uncovered for CONUS, and more specifically for the UWIN regions of interest, will serve as a test-bed for global cities encountering similar problems. The tools developed and scientific methods established to quantify impacts, obtain solutions, and derive associated uncertainty are expected to provide guidance for similar locales in other parts of the globe. The results and conclusions derived here therefore have direct applicability elsewhere. 5. General Project Information

Facilities

The central task of this project involves computational modeling and requires a high performance computing (HPC) platform that is capable of delivering solutions to solve large problems. Each of the institutions collaborating on Projects A2.1 and A2.2 has their own local HPC platform that permits scalable computing to address the big-data oriented scientific questions posed. However, execution of developed modeling tools on each institution’s HPS platform can potentially lead to uncertainty in output and preclude direct output comparison between institutions. Differences in computing technology (e.g., compiler types, compiler versions, operating systems, etc.) between the HPCs would require elimination. Consequently, all WRF simulations will be carried out on NCAR’s high performance computing Yellowstone HPC environment. Collaborators from Project Plan 2.1 and 2.2 will jointly prepare and submit a Yellowstone computing hours allocation request to NCAR, which will enable remote computing and local post-

48

processing of platform independent (netCDF) simulation output. 6. Interactions with other UWIN Projects/Institutions This project is directly connected to two projects: A2.1 and A2.3 (but is indirectly connected to several others, including A2.4, A1.2, etc.). This project relies on development of advanced modeling tools that will be made available from A2.1, and will inform Project A2.3. To ensure that modeling simulations are initiated, and useful results are available within a timely manner, Project A2.2 will commence with modeling activities using the default WRF (coupled to the existing single layer urban canopy model) system with bulk hydrology implementation. Added urban hydrological physics representation provided by Project A2.1 will be incorporated in a second iteration of simulations in Project A2.2. Differences between the two sets of simulations are expected to reveal important details related to model physics uncertainty (e.g., how does the incorporation of non-bulk urban hydrology implemented in Project A2.1 lead to differences in sensitivity due to urban expansion as simulated in Project A2.2?). Similarly, the technological solutions developed in Project A2.4 will require integration into the offline and eventually the online WRF modeling system, which will subsequently be utilized to conduct simulations using scenario-based projections of future environmental conditions (as described earlier). 7. References Baldocchi, D., E. Falge, L. H. Gu, R. Olson, D. Hollinger, S. Running, et al. (2001). FLUXNET: A new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities, Bulletin of the American Meteorological Society, 82(11), 2415-2434. Bruyere, L., A. J. Monaghan, D. F. Steinhoff and D. Yates (2015). Bias-Corrected CMIP5 CESM Data in WRF/MPAS Intermediate File Format. NCAR Technical Note, NCAR/TN-515+STR. Georgescu, M., Morefield, P. E., Bierwagen, B. G., and Weaver, C. P. (2014). Urban adaptation can roll back warming of emerging megapolitan regions, Proc. Natl. Acad. Sci. (USA), 111(8), 2909-2914. Georgescu, M. (2015). Challenges associated with adaptation to future urban expansion. Journal of Climate, 28(7), 2544-2563. Georgescu, M., Chow, W. T. L., Wang, Z. H., Brazel, A., Trapido-Lurie, B., Roth, M., and Benson-Lira, V. (2015). Prioritizing urban sustainability solutions: coordinated approaches must incorporate scale-dependent built environment induced effects, Environ. Res. Lett., 10(6), 061001. Hurrell J.W., and Co-authors (2013). The Community Earth System Model: A Framework for Collaborative Research. Bull. Amer. Meteor. Soc., 94, 1339–1360. Knutti R., D. Masson, and A. Gettelman (2013). Climate model genealogy: Generation of CMIP5 and how we got there. Geophys. Res. Lett., 40, 1194–1199, doi:10.1002/grl.50256. Moss, R. H., and Co-authors (2010). The next generation of scenarios for climate change research and assessment. Nature, 463, 747Q756, doi:10.1038/nature08823. Pielke, R. A., Pitman, A., Niyogi, D., Mahmood, R., McAlpine, C., Hossain, F., et al. (2011). Land use/land cover changes and climate: modeling analysis and observational evidence, Wiley Interdisciplinary Reviews: Climate Change, 2(6), 828-850. Skamarock, W. C., et al. (2008). A Description of the Advanced Research WRF Version 3 (National Center for Atmospheric Research, Boulder, CO.). Yang, J., Z. H. Wang, F. Chen, S. Miao, M. Tewari, J. Voogt, et al. (2015). Enhancing hydrologic modeling in the

49

coupled Weather Research and Forecasting - urban modeling system, Boundary-Layer Meteorology, 155(1), 87-109.

50

PROJECT A2-3 RESEARCH PLAN

Project Title Assessing the thermal comfort implications of water-supported urban infrastructure at the human scale

UWIN Project number A2-3

Project Lead David Hondula, ASU Investigators/Institutions

J. Vanos/ASU/TTU (Funded) M. Georgescu/ASU (Funded under A2-2) S. Harlan/ASU (Funded under SE&EJ Project) E. Mack/ASU (Funded under A1-2) A. Middel/ASU (Not Funded) D. Sailor/ASU (Not Funded)

Project Period Project years 1–3 are funded Project Cost Will be determined by Dec 31, 2015. Graduate students funded to work on project

RA TBD (12 Mo.: Years 1-3): Supervision, D. Hondula

Project Overview

Health risks associated with heat exposure are receiving increasing attention in national and regional planning and preparedness efforts in the face of projected rises in temperatures. Extreme heat already ranks as a leading weather-related cause of death in the United States (Berko et al. 2014), and climate model simulations suggest that communities across the country—and across the world—will face a future of increasingly intense, long, and frequent heat waves. The potential health consequences of these increases are among the motivating factors for national and international policymaking and implementation of highly localized cooling strategies at households and businesses. Many of the strategies currently in place in urban areas to reduce health risks related to exposure to high temperatures rely, directly or indirectly, on water. At the level of individual households, indoor climates are controlled with central air conditioning, for which a portion of energy is derived from hydroelectric power or other energy sources that require water use. Other households, particularly in dry climates, take advantage of evaporative cooling techniques that require water input. Individuals also reduce heat stress by taking cool showers and baths, utilizing swimming pools and sprinklers, and drinking cool water. Businesses take advantage of small-scale infrastructure improvements to aid the thermal comfort of customers through the installation of water misters. Home and business owners alike maintain water-supported vegetation, which produce thermal comfort benefits related to both shade and evaporation. Increasing vegetation cover on both public and private land is a stated goal of municipal plans in cities across the United States, and the thermal comfort benefits of such vegetation is often cited as a motivator for program adoption. Irrigated landscapes and standing surface water bodies like lakes and ponds can also mitigate heat stress in urban areas. UWIN project A2-3 investigates what will happen to the thermal comfort of urban dwellers, and the associated risks of heat related illness, in the face of changes to urban water systems driven by climatic variability and infrastructure modification. Building from a large literature that has examined how microclimates are sensitive to vegetation and various urban design strategies, the UWIN team will shift the analytical framework for this research to be oriented around people instead of places, making it more directly relevant for human health. UWIN researchers will adopt state-of-the-art microclimate measurement techniques and human energy budget modeling to quantify how changes in the urban water system will affect the thermal properties of the environments in which urban dwellers live, work, and play, and contextualize these changes based on the cumulative daily thermal experience of urban residents. The outputs from project A2-3 will provide rich information for water decision-makers in urban areas to more thoroughly evaluate how alternative configurations for water infrastructure and different future climatic scenarios will impact the health of urban residents as viewed from the lens of thermal comfort and heat stress.

51

Project Summary

Objectives: UWIN Project A2-3 supports the development of the UWIN Blueprint Indicator for Heat Related Illness. The project has two overarching objectives. The first is to measure the microclimatic conditions and thermal comfort experienced by urban dwellers through the course of their daily experiences, integrating indoor, outdoor, and transit-based exposures. This measurement approach is a significant advance for understanding how urban form and urban water infrastructure relate to thermal comfort (and thus, risk of heat related illness) because it transitions the measurement paradigm for environmental conditions from one focused on places to one focused on people. The second objective is to quantify how changes in urban infrastructure, particularly those elements supported by water, as well as changes in urban climate, will impact the thermal comfort experienced by city residents across the UWIN regions. To achieve this objective we will incorporate output from other UWIN projects (e.g., modeling output from A2-2); replacing current conditions examined in baseline assessment of microclimates and thermal comfort with alternative plausible futures for urban infrastructure and climates.

Experimental Approach: The core data set for Project A2-3 is a series of evidence-supported time, location, and activity logs that represent a suite of different lived experiences of urban residents in each of the UWIN regions. The development of these individual activity diaries, constructed for ten-minute intervals over the course of a 24 hour period, will be informed by a series of interviews with health sector stakeholders, discussion with UWIN collaborators and output from other UWIN projects, as well as regional and national-scale data sets such as the National Human Activity Pattern Survey. We will then employ micrometeorological instrumentation (including low-cost, portable sensors) to measure actual environmental conditions in the specific locations “visited” by urban residents in each of the envisioned activity diaries. After data collection, we will apply the COMFA human energy budget model to develop estimates of heat stress (e.g., core temperature, net energy budget) for urban residents who move through the particular environments that have been measured. Subsequently output from other UWIN projects such as recommended changes to urban infrastructure and future climatic scenarios will be utilized as input for the COMFA model—altering the microclimatic parameters relevant for human thermal comfort as dictated by the output from other teams (e.g., changing temperature based on future greenhouse gas and urbanization scenarios, adding shade based on proposed green infrastructure, removing evaporative cooling from vegetation).

Outputs and Outcomes: Project A2-3 will generate several deliverables useful for the construction of the UWIN Blueprint, including a compilation of health official perspectives on heat illness risks related to urban water systems, a set of activity diaries that can be used in the exploration of thermal comfort implications of various urban climate/infrastructure scenarios, a data set of microclimatic conditions at a wide variety of sites in each UWIN region, a modeling framework for relating changing urban climatic conditions and infrastructure to human heat stress, and quantitative indicators of the impact of urban climate change and water uses on human thermal comfort. Collectively, we envision that these outputs enable the full representation of human health concerns related to heat exposure in the evaluation of costs and benefits of innovative urban water solutions proposed to and examined by water decision makers across the UWIN network.

Supplemental Keywords

Core temperature, energy budget, radiation, heat exhaustion

Project Description

1. Objectives Project A2-3 spans several objectives supporting the development of the UWIN Blueprint Indicator for Heat Related Illness:

Understand health sector perspectives on relationships between urban water systems, climate change, thermal comfort, and heat illness

Construct a set of detailed time and activity diaries for representative urban dwellers of each of the UWIN

52

regions that can serve as a foundational data set for this project and others in the exploration of impacts of various urban water scenarios

Measure the microclimatic conditions experienced by urban residents through the course of their daily experiences

Model the physiological stress of urban residents moving through different urban microclimates

Model how changes to water-supported urban infrastructure as well as urban climate may impact individually experienced thermal comfort for urban dwellers across the UWIN regions

Provide training opportunities for undergraduate and graduate students in microclimate measurement and modeling

Enhance coordination across the UWIN sites through the creation of foundational data sets and sharing of project responsibilities

Advance the notion of individually experienced microclimates as a meaningful framework for evaluating health risks associated with heat stress

2. Intellectual Merit This project supports the development of the UWIN Blueprint Indicator for Heat Related Illness, one of the health outcomes of concern with evolving urban water systems. Heat Related Illness is one of the most immediate concerns for the health sector arising from anthropogenically-induced climate change at both the global and urban scales, as articulated in the United States National Climate Assessment and Fifth Assessment Report of the Intergovernmental Panel on Climate Change, and many of the adaptation strategies that cities are enacting to reduce health risks associated with rising temperatures are water dependent. This project will provide the knowledge that enables health impacts related to thermal stress to be included in evaluation of water infrastructure transitions in urban areas across the United States. Further, this project will refine and advance the emerging framework of individually experienced microclimates as the basis for understanding direct impacts of thermal exposures and evaluating how health risks of such exposures may evolve in the future. 3. Approach/Activities The project objectives will be achieved through a mixed methods approach that involves interviewing health sector stakeholders across the UWIN network, systematically reviewing scientific literature, extracting data from existing national and regional-scale data sets related to human activity patterns, making direct measures of microclimates, and modeling microclimates and human energy budgets.

Individual Activity Diaries

The foundational data set for this project will be comprised of a series of high resolution, time, location, and activity diaries that are generated for hypothetical individuals who live in each of the UWIN study regions. These individual activity diaries (IADs) will be co-developed with contextual knowledge of the UWIN cities that already exists within the research teams across the network as well as from perspectives shared from health officials and other key stakeholders. Although the exact number and temporal scope of the IADs will ultimately be based on stakeholder and collaborator input (including perspectives from Project A1-2 by Mack et al. and the Social Equity/Environmental Justice household surveys led by Harlan et al.), we envision approximately ten different hypothetical (but evidence-supported) cases to be generated for each city, spanning a range of occupational, demographical, and behavioral profiles that collectively span a range of representative day-to-day experiences in the lives of residents of UWIN cities. Each IAD will include, at a minimum, for each ten-minute interval of a 24-hour period:

The individuals’ geographic coordinates

An indicator of whether the individual is indoors/outdoors, and/or their mode of transit

The type of activity the individual is performing and associated metabolic demands Each IAD will also be accompanied by a narrative description of an individuals’ lifestyle, age, height, weight, health status, and other relevant contextual information that is necessary input data for human energy budget models and

53

can aid in the evaluation of various “what if” scenarios related to alterations to water infrastructure and changing climate. The construction of these IADs will be, to the maximum extent possible, based on observed or modeled activity log patterns available at national or regional scales. Examples of data sets that will be used to guide the construction of the IADs include the United States National Human Activity Pattern Survey (NHAPS, Klepeis et al. 2001) and activity-based models used by public transportation agencies for route and infrastructure optimization (Bowman and Ben-Akiva 2001). To generate realistic locations and routes throughout cities, researchers will make visits to the locations of interest to map feasible routes as well as use virtual platforms like Google Earth where in-person visits are impractical.

Health Sector Stakeholder Interviews

In each of the UWIN regions we will seek to gain perspectives on connections between water infrastructure, climate change, and the risk of heat-related illness from individuals charged with managing community health risks. These interviews will largely focus on understanding what vulnerabilities health officials see concerning heat related illness and water provisioning in the context of severe medical cases that require ambulatory care, emergency department visit or hospitalization, or result in death. The information and perspectives that will be gleaned from these interviews that are not available from retrospective epidemiological analysis of health outcomes data include greater familiarity with specific cases and medical examiner review notes than available from outcome data, as well as contextual understanding of the various means in which water is delivered to those in need in different geographical settings. For example, the specific communities that might be negatively impacted by water shortages or deteriorating water quality are likely to differ from one city to another, which has implications in terms of the types of strategies these cities might implement to ensure residents in need have access to drinking water, which is especially needed in the case of an extreme heat event. Interviews will be semi-structured and last no more than 90 minutes. Questions will be authored by the A2-3 project team with input solicited from UWIN collaborators. Within each of the UWIN regions we will engage at least three individuals working in the health sector, likely including a practicing medical professional (medical doctor or nurse), a local public health official, and a state public health official. Interviews will be conducted by trained graduate students, postdoctoral researchers, and faculty working in each of the respective sites (as managed by the outreach team). The A2-3 project team is well-positioned to start the interview process in the Sun Corridor region drawing from ongoing collaborations between ASU, the Maricopa County Department of Health, and the Arizona Department of Health Services, and build from these relationships into productive conversations with health sector stakeholders in other regions. The interviews will serve as a platform for the co-development of the activity diaries that serve as the basis for modeling experiments. The interview participants will be asked to imagine a day in the life of an individual (or individuals) whom they believe to be at risk of heat-related illness and describe how interaction with water infrastructure could mitigate or exacerbate this risk. Following the exploration of this “acute,” case, we will also ask health officials how they understand water infrastructure to be related to thermal comfort for the population more generally. These questions will not only aid in the development of alternative activity diaries beyond the acute case, but also provide baseline information regarding these officials’ views on water-thermal comfort connections more generally. As health officials have an especially important position with respect to public communication and policy influence in communities, we may be able to identify opportunities for these individuals to more actively participate in conversations in cities across the country about water infrastructure transitions from a standpoint of thermal comfort and public health. We will continue dialogue with these stakeholders throughout the development of activity diaries and narratives and evaluate results from modeling experiments with these stakeholders to explore the viability of conclusions and imagine additional scenarios or individual activity patterns to be modeled.

Measuring Microclimates

Human thermal comfort is not only based on the air temperatures that an individual experiences, but also

54

atmospheric humidity, wind speed, and shortwave and longwave radiation experienced in a given space at a given time. A large literature has evolved over the past several decades that includes rigorous assessments of urban microclimates in diverse locations across the globe, largely emphasizing the connections between environmental conditions and building and landscape design elements (Ketterer and Matzarakis 2014; Matzarakis et al. 2007). The goal of this portion of the project is to perform a comprehensive evaluation of microclimate parameters for as many instances and locations as possible that are listed on individual activity diaries. The measurement campaign will take advantage of low-cost and portable sensors more appropriate for rapid deployment in a range of urban settings than traditional micrometeorological monitoring stations—emphasizing that the data are collected for the correct contexts (daily lives of urban residents) rather than at the highest level of precision. A comparison of instrumentation precision and accuracy to what is available from research-grade instruments will be performed during the first field measurement period in Phoenix, AZ. Microclimate measurements will be collected at the appropriate times of day as envisioned by the activity diaries and on both an “extreme heat” (hottest 5% of summer days based on each city’s climatology) and “typical summer” (near median of summer days based on city’s climatology) day. Where measurements are not possible (e.g., inside certain private business or residences), estimates shall be drawn from the literature and/or made at analog sites identified by the research team. Particularly important will be the measurement of microclimate variables at those locations in which water is being directly or indirectly supplied to or removed from the landscape or atmosphere (e.g., near open bodies of water, in regions with grass cover or standing vegetation and trees, where water has been used for energy production, in locations where misters or evaporative coolers are being used). These measurements can be made by undergraduate and graduate research assistants at each UWIN location who have received training from project personnel with expertise in meteorological instrumentation. The specific instrumentation needs for this component of the project include a handheld Kestral weather meter to measure relative humidity, air temperature, and wind speed, and a miniature cylindrical radiation thermometer (CRT) in conjunction with an Omega higher accuracy digital thermometer to measure the mean radiant temperature that would be experienced by each individual. After initial testing in Phoenix with this research-grade instrumentation for method validation, microclimate monitoring kits will be mailed to each city to perform the data collection. Undergraduate and graduate UWIN research assistants and REU participants will be able to perform the microclimate assessments in the different study regions with support from in-person visits from the A2-3 project team. At the conclusion of the microclimate monitoring campaigns, individual-based time series of all parameters relevant to the human energy budget will be available at ten minute increments. This data set represents the baseline “thermal experience” of the representative urban dwellers that will be compared to various climate and water infrastructure scenarios explored in other UWIN projects (e.g., modification/creation/removal of urban green spaces, see Brown et al. 2015).

Human energy budget modeling

The human energy budget modeling applied to urban areas will be completed using established methods of the COMFA energy budget model based on biophysical principles of the outdoor microclimate for estimation of thermal comfort (Brown and Gillespie, 1986; Kenny et al., 2009; Vanos et al., 2012). With these energy budget models, developed in part and in current use by Vanos, we will determine indicators for human thermal comfort based on all aspects of the human energy balance. These include gains from long- and short-wave energy absorbed (Rabs), and human metabolism (M), with losses via evaporation (E), convection (C), and long wave radiation emitted from the body (L). The final energy balance equation (that underlies the COMFA model) is commonly written as: EB (W m-2) = M + Rabs –L – C – E. In urban areas where the geometry and thermal properties of buildings create high degrees of microclimatic variability (for example due to water sources, shading and reflection, and absorption and/or emission of radiation), we anticipate high contrasts in estimated energy budgets at high spatial and temporal resolutions. Each component in the EB equation represents a flux to or from the human body, and during extremely warm summer conditions, C becomes a positive flux towards the body with L limited, hence loss through evaporation becomes the body’s naming

55

mechanism for cooling. The human metabolic activity (Mact) will either be estimated using movement data (activity velocity) from activity databases to inform estimates from databases for general energy expenditure ( ) in W m-2, or derivation of more specific estimates for each given subject attributes (e.g., age, height, weight, gender) can be input into a general energy expenditure estimation method (Strath et al. (2000) as demonstrated by Vanos et al. (2012)). This is an important component for as the value of Mact is further used to calculate skin, core, and body temperatures of each subject, which are the physical determinants of heat stress, heat exhaustion, heat stroke, brain damage, and potentially death (Bouchama and Knochel, 2002; Bulcao et al. 2000). The evaporative heat loss calculation will also be used as an indicator of heat stress based on sweat loss. A final integral component of this methodology incorporating observations into the human energy budget model is the radiation absorbed by a human. The component of the radiant exchange is widely cited as the main indicator of personal thermal comfort and heat stress, particularly in warm-hot conditions (Matzarakis et al. 2007; Parsons 2003; Kenny et al. 2009; Vanos et al. 2012), yet it is one of the most difficult variables to accurately obtain. However, the use of a CRT (as tested by Vanos et al. 2012, Kenny et al. 2009) in a miniature form will provide estimations of absorbed radiation and mean radiant temperature. This is a will provide novel microclimate radiation data with new and low cost instrumentation to determine this significant component of heat stress based on accurate human geometry (based on Monteith and Unsworth 1990; Campbell and Norman 1998). In summary, at the end of this stage of the project, we will have generated for each ten-minute interval of each representative urban dweller’s activity log, estimates of net energy gain or loss, sweat loss, and subsequently, core temperature, using the COMFA model. These calculations will be performed for the baseline scenarios as measured from field campaigns as well as for alternative scenarios based on different climatic regimes (driven by model output from other UWIN project teams) as well as adjustments to urban water infrastructure (driven by solutions posed by other UWIN project teams). For each of the scenarios of interest, we will calculate the percentage of time that individuals are experiencing dangerous amounts of heat gain and/or have core temperatures above normal and thresholds of concern for heat illness. 4. Expected Results, Benefits, Outputs, and Outcomes Outputs generated from this project will include: (1) transcripts, analysis, and synthesis of health official perspectives on heat vulnerability related to urban water systems; (2) a suite of representative activity diaries encapsulating daily experiences the lives of residents of the UWIN study cities; (3) comprehensive micrometeorological characterizations of environments utilized by urban dwellers; and, (4) modeled thermal stress of urban residents in current conditions and modeled environments with changing hydroclimatic conditions and water infrastructure. The primary outcome of this suite of results is that they will make possible a richer evaluation of the thermal comfort tradeoffs of various hydroclimatological and water infrastructure scenarios by decision makers across the network. This project is the main research activity that will enable Heat Related Illness to be included as a UWIN Blueprint Indicator. This project requires external engagement with health and water sector stakeholders and will facilitate cross-network educational opportunities for undergraduate and graduate students. We envision the interviews with health sector officials being the start of an iterative process, in which those interviews aid the development of representative activity diaries, which are then used for the exploration of various “what if” scenarios related to climate change and changing water infrastructure, which are then shared back with these stakeholders who may develop new questions and ideas in response. Undergraduate and graduate students from all UWIN sites (as facilitated by UWIN project leaders and cross-site coordinators) will be involved in developing individual activity diaries with contextual information about their own locations, making micrometeorological measurements in urban settings, and calculating heat stress indicators using human energy budget models based on data that they have collected. 5. General Project Information The lead investigators for this project have considerable experiencing executing the proposed research activities. Hondula is co-author of several manuscripts published and in preparation related to heat vulnerability and individually experienced heat stress (Hondula et al. 2015, Kuras et al. 2015). Hondula works closely with health departments in

56

Arizona, including through the CDC-supported Building Resilience Against Climate Effects program, on projects related to heat vulnerability and analysis of health outcome data from which he will draw in the development of stakeholder interview questions and proposing cases of interest for activity diary development. Middel and Vanos have each worked on microclimate assessment projects (Middel et al. 2012, Vanos et al. 2012). Middel and Vanos have each published several manuscripts using different models to evaluate environmental modifications for thermal comfort and implications for the human energy balance (Middel et al. 2014, Vanos et al. 2010, Vanos et al. 2012). This project will closely connect to and draw from the efforts of the urban climate group (e.g., urban climate models from Georgescu et al.), the socioeconomic/policy/institution group (e.g., framing the study for exploration of environmental justice issues with Harlan et al.), the engagement group (e.g., coordinating with health sector stakeholders with Sukop et al.), and the biodiversity and technological solutions group (e.g., understanding the range of possible urban/green infrastructure modifications with Jenerette et al.). It is anticipated that all institutions will contribute to this project in at least a small capacity, whether through the sharing of contextual information about UWIN cities or through the sharing of undergraduate and graduate student time for interview assistance, IAD development, and microclimate measurements. The project team welcomes input from UWIN leadership on how to best facilitate this cross-site, cross-project (and cross-disciplinary) engagement. This project will draw insights and resources from other ongoing efforts at Arizona State University, including those of the Central Arizona-Phoenix Long-Term Ecological Research Program (CAP LTER) and from Hondula, Georgescu, and Harlan’s involvement in an NSF-funded SEES-Hazards project that also focuses on the evaluation of individually experienced microclimates.

Timeline

Year 1 Fall: Develop interview protocol and templates for activity diaries

Year 1 Spring: Interview health officials, begin development of activity diaries

Year 1 Summer: Complete all activity diaries, microclimate data collection in Phoenix

Year 2 Fall: Phoenix microclimate data processing and energy budget modeling

Year 2 Spring: Integration of climate/water use scenarios into Phoenix microclimate/energy budget results, discussion of outputs with Phoenix stakeholders

Year 2 Summer: Microclimate data collection in all UWIN sites

Year 3 Fall: Microclimate data processing and energy budget modeling for all UWIN sites

Year 3 Spring: Integration of climate/water use scenarios into microclimate/energy budget results, discussion of outputs with stakeholders

Year 3 Summer: Final data collection based on stakeholder input, additional data processing and modeling, preparation of final results and manuscripts

6. References Berko, J., Ingram, D. D., Saha, S., & Parker, J. D. (2014). Deaths attributed to heat, cold, and other weather events

in the United States, 2006-2010. National health statistics reports, (76), 1.

Bouchama, A., & Knochel, J. P. (2002). Heat stroke. New England Journal of Medicine, 346(25), 1978-1988. Bowman, J. L., & Ben-Akiva, M. E. (2001). Activity-based disaggregate travel demand model system with activity

schedules. Transportation Research Part A: Policy and Practice, 35(1), 1-28. Brown, R. D., & Gillespie, T. J. (1986). Estimating outdoor thermal comfort using a cylindrical radiation

thermometer and an energy budget model.International journal of biometeorology, 30(1), 43-52.

57

Brown, R. D., Vanos, J., Kenny, N., & Lenzholzer, S. (2015). Designing urban parks that ameliorate the effects of

climate change. Landscape and Urban Planning, 138, 118-131. Bulcao, C. F., Frank, S. M., Raja, S. N., Tran, K. M., & Goldstein, D. S. (2000). Relative contribution of core and

skin temperatures to thermal comfort in humans. Journal of Thermal Biology, 25(1), 147-150.

Campbell, G. S., & Norman, J. M. (1998). An introduction to environmental biophysics. Springer Science & Business Media. Hondula, D. M., Davis, R. E., Saha, M. V., Wegner, C. R., & Veazey, L. M. (2015). Geographic dimensions of

heat-related mortality in seven US cities. Environmental research, 138, 439-452. Kenny, N. A., Warland, J. S., Brown, R. D., & Gillespie, T. G. (2009). Part A: Assessing the performance of the

COMFA outdoor thermal comfort model on subjects performing physical activity. International journal of biometeorology,53(5), 415-428. Ketterer, C., & Matzarakis, A. (2014). Human-biometeorological assessment of heat stress reduction by replanning

measures in Stuttgart, Germany.Landscape and Urban Planning, 122, 78-88. Klepeis, N. E., Nelson, W. C., Ott, W. R., Robinson, J. P., Tsang, A. M., Switzer, P., ... & Engelmann, W. H. (2001). The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental

pollutants. Journal of exposure analysis and environmental epidemiology, 11(3), 231-252. Kuras, E. R., Hondula, D. M., & Brown-Saracino, J. (2015). Heterogeneity in individually experienced temperatures (IETs) within an urban neighborhood: insights from a new approach to measuring heat

exposure. International journal of biometeorology, 1-10. Matzarakis, A., Rutz, F., & Mayer, H. (2007). Modelling radiation fluxes in simple and complex environments—

application of the RayMan model.International Journal of Biometeorology, 51(4), 323-334. Middel, A., Brazel, A. J., Gober, P., Myint, S. W., Chang, H., & Duh, J. D. (2012). Land cover, climate, and the

summer surface energy balance in Phoenix, AZ, and Portland, OR. International Journal of Climatology, 32(13), 2020-2032. Middel, A., Häb, K., Brazel, A. J., Martin, C. A., & Guhathakurta, S. (2014). Impact of urban form and design on

mid-afternoon microclimate in Phoenix Local Climate Zones. Landscape and Urban Planning, 122, 16-28.

Monteith, J., & Unsworth, M. (1990). Principles of Environmental Physics: Plants, Animals, and the Atmosphere. Academic Press.

Parsons, K. (2003). Human thermal environments: the effects of hot, moderate, and cold environments on human health, comfort, and performance. Crc Press. Strath, S. J., Swartz, A. M., Bassett Jr, D. R., O'Brien, W. L., King, G. A., & Ainsworth, B. E. (2000). Evaluation

of heart rate as a method for assessing moderate intensity physical activity. Medicine and Science in Sports and

Exercise, 32(9 Suppl), S465-70. Vanos, J. K., Warland, J. S., Gillespie, T. J., & Kenny, N. A. (2010). Review of the physiology of human thermal

comfort while exercising in urban landscapes and implications for bioclimatic design. International journal of

58

biometeorology,54(4), 319-334. Vanos, J. K., Warland, J. S., Gillespie, T. J., & Kenny, N. A. (2012). Improved predictive ability of climate–

human–behaviour interactions with modifications to the COMFA outdoor energy budget model. International journal

of biometeorology, 56(6), 1065-1074.

59

PROJECT A2-4 RESEARCH PLAN

Project Title Assessment and design of innovative building systems and urban infrastructure to mediate impacts on the urban water cycle, heat island, and regional climate

UWIN Project number A2-4 Project Lead Forrest Meggers, Princeton Investigators/Institutions

Marilys Nepomechie, FIU (funded to work on this project) Elie Bou-Zeid, Princeton (Collaborating on this project, through synthesis with project A2.1) Zihua Wang, ASU (Collaborating on this project, through synthesis with project A2.1) Matei Georgescu (Collaborating on this project through synthesis with project A2.2) Darrel Jenerette UC-R (Collaborating on this project through synthesis with project A3.2) Sybil Sharville, CSU (Collaborating on this project through synthesis with project B1.1) Thomas Meixner UA (Collaborating on this project through synthesis with project B2.1) Gary Pivot, UA (Collaborating on this project through synthesis with project C1.1)

Project Period 8/1/2015 – 7/31/2020 Project Cost

Princeton University: $380K FIU: Will be determined by Dec 31, 2015.

Graduate students funded to work on project

Year 1: Meggers supervising Hongshan Guo – 25% Year 2: Meggers supervising TBA– 100% Year 3: Meggers supervising TBA – 100% Year 4: Meggers supervising TBA – 75% Year 1-5: 2-mo/summer: Meggers supervising 2-month summer 1-3 Federal Work-study funded Architecture graduate students

Project Overview

Our capacity to analyze the built environment has exploded over the past few decades with access to new digital tools, simulation engines, and above all access to novel materials and building processes. These added capabilities have helped increase our understanding of individual aspects of building systems and urban infrastructure, but the added complexity has left large voids in our understanding of the interfaces between the variety of material, computational and technological advances, which are part of modern cities today. One of these key interfaces is at the energy-water nexus. Our expertise spanning building systems, architecture and urban infrastructure will be applied to fill in the knowledge gaps in the city’s energy-water-infrastructure nexus. The project will engage with water sustainability through study of energy, buildings, ecology and people, uncovering and bridging the gaps in knowledge that exist between disciplines of fundamental applied science and creative design research. As shown in Figure A2-4-1 there is much information that can be represented through simple overlay of data onto the form of the city. It can help both designers and scientist better address challenges of urban water and understand critical relationship between water and energy in society.

60

Figure A2-4-1: Overlay of the thermal variations uncovered through thermal imaging of the urban fabric, revealing the hidden thermal exchanges taking place between buildings, micro-climates and surfaces

Project Summary

The objective is to generate new knowledge on the interactions of buildings systems and the urban environment with the water-energy nexus through fundamental modeling of systems, systematic analysis of applications, and design research into solutions.

(1) Baseline analysis to determine role of built environment in urban sustainability through the water-energy nexus and create reduced models of micro-scale technologies to feed up to the urban scale models. Output a robust system for understanding the impact of buildings at the fine scale to inform model inputs and system development. Generation of alternative metrics and modes of addressing impacts created at the energy/water/infrastructure nexus.

(2) Project application to explore overlooked applications of energy-water-infrastructure nexus like evaporative cooling/humidity/condensation and pressure changes/anthropogenic point sources etc. Build simulation of heat and water inputs at the individual building and system scale to interact and aggregate into better representations of the relationships to urban water sustainability

(3) Design opportunities to create alternative solutions through architecture and urban analysis that address the challenges analyzed at the water-energy-infrastructure nexus, and create robust descriptions and presentations of the potentials of these solutions and how they can feed into and help define future more resilient scenarios.

The experimental approach will first employ thermodynamic and heat transfer energy modeling and its interactions with infrastructure, the urban climate and water systems, to calculate the impacts on system performance and urban environment. The approach will include an upscaling and downscaling exchange between climate models and building systems. At the application and design analysis the approach will be more systematic and generative as system evaluations and design scenarios will be informed by both quantitative and qualitative analysis. Meggers will manage the overall research and specifically interactions of 1 and 2 with other fundamental research projects, and Nepomechie will support 2 and 3 guiding objectives starting from applications and focusing on expanding design opportunities spanning architectural and urban design

East facing view from rooftop to neighborhood below, shows

the most extreme and contained temperatures around rooftop

cooling units. Solar gains on building facades appears more

material.

1. Anthropogenic vs Solar Gains

61

The resulting outputs will be analytical metrics for heat and moisture interactions derived from physical analysis and simulation of buildings, systematic analysis of building components at operating at the energy-water nexus, and design scenarios to address the opportunities presented by the first two. The outcomes will increase understanding and improved models of the complex interactions and dependencies of building energy and water usage, which will subsequently facilitate more informed designs and solutions that address these challenges through interactions with culture, society and the environment. The bridge that will be built from fundamental research analysis to design exploration will help address broad environmental challenges like water management and greenhouse gas emission to mitigate the sweeping risks they impose, while simultaneously providing opportunities to adapt to the improved understanding of changes to come.

Supplemental Keywords

Energy chain, Districts, Neighborhoods, Commercial, Residential, Dwellings, Society, Culture

Project Description

1. Objectives The overall objective of this project is to investigate the impacts and interfaces on urban water of the built environment from the scale of individual buildings to neighborhoods. One core area of investigation will be the relationship of the urban water to the energy intensity of urban infrastructure and its operation. In this capacity we define an energy-water-infrastructure nexus that drills down into the specific urban and building aspects of the energy-water nexus. We aim to create new modes and metrics for water and energy interactions that can feed up into the larger scale models of the urban water systems across the UWIN network, and we will reciprocally leverage the results of these larger scale climate and environmental models to better understands the feedback on building systems and their water, energy and environmental impacts. Finally an overarching objective is to extend the science into applications and design opportunities through engagement with architecture and urban studies disciplines. Individual objectives are identified to achieve the overall aim of generating valuable analysis and results at the energy-water-infrastructure nexus that span the basic analysis, application in further projects, and for generation of design opportunities at the urban scale. These objectives are outlined below:

1) Baseline Analysis Objectives: Define relationship of building systems and urban infrastructure to energy-water nexus, micro-climates and local environments

a. Compile an extensive matrix of building systems and urban infrastructure with relevant water and energy inputs and outputs.

b. Determine the direct links between water usage and water cycle with energy demand and utilization crossing the energy-water nexus.

c. Quantify the impacts of the inputs and outputs of the energy and water components at the building and neighborhood scale.

i. Use fine grid CFD at the building scale to explore special cases ii. Use dynamic energy simulation to quantify thermal exchanges

iii. Create latent water exchange model d. Create metrics relevant to parallel ecological and urban climate models

2) Project Application Objectives: Measure, calculate and quantify application metrics to building systems and urban infrastructure to demonstrate opportunities for improvement

a. Measure impacts on urban physical surfaces (both biological and constructed) and infrastructure (Bou-Zeid synergy)

i. Evaporation effects from moisture phase change ii. Radiation effects from thermal emission, absorption and reflection

iii. Convective effects from thermal buoyance and air movement b. Calculate degradation of performance of building systems, particularly air conditioning units and

62

wet cooling towers from influences by resulting heat and moisture changes from urban climate models (Wang, Bou-Zeid, Georgescu synergies)

c. Quantify the impact on thermal comfort at the human scale for micro-scale interactions in the urban climate and inside built form (Wang synergy).

d. Consider interactions with plant surfaces and the resulting measurable evapotranspiration influences on water and thermal exchanges (Jenerette synergy)

e. Determine energy and water evaporation exchanges and losses at specific infrastructure scale relevant to retention and utilization in water-scarce project areas (Meixner synergy)

3) Design Opportunity Objectives

a. Generate alternative system designs or operations to improve performance at the energy-water-

infrastructure nexus. b. Determine capabilities of water-efficiency and other building-scale design innovations to improve

energy performance and contribute to urban heat island mitigation (Sharvelle synergy). c. Compare and optimize the design and utilization of water-based energy transport mechanisms in

the urban environment i. Water-based heating and cooling systems at the building scale

ii. Geothermal exchange, performance and groundwater interface iii. District heating and cooling systems

d. Design of building and urban form to address baseline conditions and application opportunities mitigating urban heat island and improving system performance.

2. Intellectual Merit The intellectual merit can be considered also in the context of generation of 1) baseline information 2) project application, and 3) design opportunities. Each part generates merit from the network aspect of the project through an interaction and exchange with other project aspects within the UWIN SRN. Meggers and Nepomechie will provide a critical bridge to the urban design and applied engineering at the building scale. As part of thrust A, the intellectual merit of the 1) baseline analysis will be in the creation of new inputs and calibrations of thermal exchanges directly connected to the systems and operations at the building scale. By bridging Meggers’ expertise with buildings, thermodynamics and system design with the modeling expertise of the others work in A2 (e.g. Bou-Zeid, Wang, Georgescu), the outputs will create new knowledge and opportunity in the how buildings can react to aspects of the urban and microclimate around them. At the same time the urban climate models and simulations at the larger urban scale will be informed by more resolved and nuanced metrics and refined inputs for the anthropogenic contributions from building design and system operation. The intellectual merit derived from 2) project application will stem from the extension of the results of the baseline analysis in direct application at the building and neighborhood level. A more resolved understanding of how the radiation trapping and surface temperatures that evolve in the building scale and urban fabric can be combined with the work of Jenerette on leaf temperature to create new understanding of interactions and impacts of ecological and constructed surfaces and their relationships to water, energy and the comfort and quality of spaces generated across the wide variety of urban conditions represented within the UWIN group. Additionally, the application of the baseline interactions at the scale of the built environment will provide input and validation for concepts and analysis developed in thrust B in the context of building systems and urban constructions. For example, the expertise from Sharvelle and Meixner will develop new ideas and concepts that can be informed by and combined with analysis done at the building scale in this project to generate new outputs and knowledge. Finally, the connection to the School of Architecture by Meggers and Nepomechie will generate significant intellectual content in the context of 3) design opportunities. In email conversation between Meggers, Nepomechie, and Pivot, there was agreement on the importance of leveraging intellectual merit from design experts into the

63

context of building and urban infrastructure relationships to urban water. This will also play a role in feeding back into the applications at the building scale to validate the economic and practical feasibility as design components at the scale of buildings or urban infrastructure. The expertise of designers to realize graphical representations and illustrations of concepts and knowledge developed within the network will also generate valuable communication and understanding of the outputs of the project. 3. Approaches and Activities The approaches will use systematic analysis and simulation from experiments, data collection and modeling tools as well as generate design explorations and concept development for innovations to address challenges at the energy-water-infrastructure nexus as they are uncovered and studied. We will investigate and catalogue the interactions between building systems and the urban microclimate. These will center on energy infrastructure such as cooling towers, heat rejection from air conditioning units, condensation from cooling systems, and heat emission from building surfaces through transmission of heat and re-radiation of incident solar energy (Dhakal and Hanaki 2002). We will use fundamental thermodynamics and heat transfer to model and predict the heat transfer and the shift in humidity via water content changes from systems like cooling towers and from shifts in partial pressure due to sensible heat additions or removal from air masses, expanding on previous methods of analysis of cooling tower operation (Khan, Yaqub, and Zubair 2003; Meroney 2008). The goal will be creating an understanding of the air masses such that a more accurate and representational aspect can be upscaled for the research on urban climate models and atmospheric models being researched by Bou-Zeid and Wang (Wang, Bou-Zeid, and Smith 2013; Ramamurthy and Bou-Zeid 2014). The air masses will be analyzed at the building scale and in specific instances relevant to particular scenarios of anthropogenic building-climate interactions. This scale will be below the grid scale of the urban climate models and the results will be use to both further understanding of the fundamental phenomena themselves, and also to be developed and upscaled into logical inputs to improve urban climate models. Some work has been done coupling microscale CFD analysis and mesoscale urban models (William J. Coirier 2007), but our analysis will aim toward more generalizable characterization of the operation and heat a moisture exchange from specific building systems. It will also aim for more feedback from urban climate modeling. The temperatures increases determined and spatially resolved by the climate modeling will have a direct impact on building system performance and thermal comfort in the city as demonstrated in thermodynamics and our previous work on building systems (Meggers, Ritter, et al. 2012; Meggers, Baldini, et al. 2012; Meggers et al. 2011). Both performance and comfort impacts can be quantified by incorporation of urban scale feedback into our microscale system analysis. Three examples of building-climate interaction have already been studied in an initial investigation studying potential impacts of micro and macro climate shifts on system performance (Meggers et al. 2015). In this case the cooling energy for New York City was analyzed for its sensitivity to climate phenomena impacting the cooling systems. The results showed a significant energy penalty of 10% to 70%. These are all basic sensible temperature interactions with air conditioning units that can be mitigated through the intelligent use of a wet cooling tower to achieve the web-bulb temperature as we showed in previous work (Bruelisauer et al. 2013). Our project research will extend the analysis to include and connect the relationship between evaporative cooling with water systems, energy demand and urban microclimates. It will expand on the basic analysis of these scenarios with more detailed heat transfer calculations at the system level and heat and moisture exchanges with the surroundings. Additionally we will further develop new systems for evaporative cooling that replaces the cooling tower with a system integrated into the building surfaces as detailed in our recent work (Teitelbaum et al. 2015). The geographic extent will be based on the weather input data from energy and heat transfer simulation and for the base case New York City will be used for agglomerated urban data, but specific build-scale scenarios of interest will be selected for detailed heat and moisture exchange analysis which has historically been analyzed for materials (Kuenzel and Kiessl 1996), but not for the impacts and relationship to climate dynamics. Analysis for variation for each site in the UWIN study will be include through the use of extensively available Typical Meteorological Year

64

(Hall et al. 1978) data for building energy simulation . Additionally scenarios for specific building cases will be coordinated across the six regions in coordination with collaborators working there. Finally some alternative future scenarios based on subsequent project application and design opportunities built on the initial analysis. The main input data for these analyses will be details of urban heat exchanges and specific thermal and operational properties of systems under investigation in specific scenarios, which will include the impacts described from UHI, rooftop heat and convective heat plumes among other to be developed as regionally specific opportunities emerge. We will leverage thermal imaging and sensor deployment to measure the actual conditions in the field. We already have developed a large dataset of thermal imaging in New York City that we have used to demonstrate a variety of phenomena as shown in Figure A2-4-2.

Figure A2-4--2: The impact of heat buildup on rooftops and variations in the urban surfaces (left), the convective stack flows from window units (center), and the micro-climate impacts on people at street level (right). We have also built and are in the process of patenting our own global thermal imaging 3D plotting system. We will investigate the impact of latent moisture on the surface exchange of radiant temperature not just in the context of urban canopy models at Princeton, but also feeding back down to the street level and the perception, comfort and health of people occupying the very micro climates at the bottom of the urban canyons as shown in the Figure above, some of which can be quite extreme as demonstrated by the thermal images in the Figure. In this context we will also engage with the ecological biodiversity project to help understand the role of vegetation evapotranspiration, its evaporative cooling effects, and its radiant heat exchange from its cooler surfaces, building more explicit modeling of thermal interactions and comfort components onto the previous work that will be also further developed in project A3.2 (Jenerette et al. 2011). We will investigate project application potentials also as described in the objectives. These include parallel research in hydronic systems, geothermal ground-source heat exchangers, and district heating systems, and will hinge mainly on the energy-water nexus aspects of these systems. The analysis will center on creating metrics that tie the benefits of these systems with their relationship to water an energy carrying medium and its role and challenges faced in that capacity. Lastly we will leverage design research approaches and interface with architecture students to generate opportunities for innovation in response to the resulting challenges demonstrated by the baseline analysis. Leveraging expertise and access to students in architecture and urban and regional design, the technical analysis will be considered in the context of cultural, social and environmental challenges for energy and water in the fields of urban design and

65

architecture. Design tools and techniques from Nepomechie at FIU will generate a variety of design alternatives and analysis of building system relationships and urban infrastructures. In the context of design we will engage with stakeholders by creating clear representations of the project outputs. Also, at the scale of building and urban infrastructure we will have a clear opportunities engage with water and energy stakeholders to improve system performance. 4. Expected Results, Benefits, Outputs, and Outcomes The outputs from the project analysis of buildings systems and urban infrastructure will be upscaled as inputs for urban climate models that incorporate the outputs of our analysis of thermal and moisture interactions at the building scale. They will also output new basic metrics at the building scale. Additionally, the project will output a standalone set of analyses of building system and urban infrastructure components that quantify the energy and water performance relationship built on outputs generated in building simulation and urban data collection. Finally there will be design scenario outputs that response to the challenges and understanding developed of opportunities for innovation at the building scale. The outcomes we envision will be a better interaction between building scale thermal interactions with the environment and subsequent feedback on performance with the simulation and prediction of energy and water flows in the urban climate. Another outcome will be in the support and development of innovations at the building scale through the analysis of applications at the energy-water-infrastructure nexus. Finally, one of the high-potential outcomes will be the design generation of scenarios that will facilitate a beneficial access through graphical and representational tools of interventions and innovations that can improve urban water sustainability for dissemination to a wide range of stakeholders. Within the context of the UWIN Framework, our results will largely address the pressures of climate change temperature and precipitation feedbacks on urban infrastructure, with additional considerations relating to pressures of land use, population change, demographic change, along with appreciation for how systems have to react to extreme events and aging infrastructure. The key indicators for our project fall principally under the Linked Systems sector with additional consideration for categories under Institutions/Equity, which include social and environmental equity and adaptation capacity in relation to the outputs from our building and urban analysis. Additionally building systems application for innovative solutions at the energy-water nexus will also address indicators relating to specific water systems of indicators such as wastewater and drinking water energy efficiency and energy usage. In the solution space the outputs from or systems analyses and our design research will address all of the aspects of Green Infrastructure, either in terms or direct relationship to models of energy and water exchange, or indirectly through concept development for integrated innovative solutions at the building scale. To a bit lesser extent, but in a similar manner, the relationship of building-integrated Water Conservation systems Water Recycling systems will be addressed in design space for building systems. More specifically the project will characterize and quantify energy efficiency and energy usage at the building scale, that have direct links to urban water sustainability measures, creating opportunities to define better metrics for heat island, micro climate, and greenhouse gas emissions that are interrelated. The research will provide great opportunities for citizen and student science as a component will be the development and deployment of simple sensors to better understand urban climate at the building scale. Meggers is part of an REU proposal is going in at Princeton focused on developing sensors, which could engage students in experimental setups within this project. Additionally Meggers engages in the high school student summer research Laboratory Learning program where data collection and analysis is taught to students through existing research projects. Finally, there will be direct engagement with Masters of Architecture students as researchers and summer interns, where at Princeton federally funded work-study projects are frequently employed to support Masters students on summer design research projects. As a registered architect, associate dean and professor at FUI, Nepomechie is involved in many synergistic activities and programs that will generate many opportunities for external engagement.

66

5. General Project Information At Princeton, the Campus as a Lab program in with Meggers, Bou-Zeid and Smith are all engaged will provide some potentially valuable direct access to fine detail information about the buildings, infrastructure and ecology on campus. Hundreds of thousands of data points are available measuring system temperatures and flows around campus and through the large district heating and cooling system. Additionally, Proximity to New York City and relationships design and architecture firms and engineering consultants there will be useful for data collection in that context. Resources at Princeton include the construction of a new 7000 sq. ft. architectural laboratory with gantry crane for 1:1 prototyping and direct sensing and analysis of facades will be completed in 2016. Additionally dehumidification and evaporative cooling experimentations at the bench scale will take place in the new labs being constructed at the Andlinger Center building at Princeton to be completed in Oct 2015. Suggested specific synergies for the listed unfunded collaborators in other projects:

Elie Bou-Zeid, Princeton, A2.1: Interfaces with buildings and climate models, latent and sensible heat exchanges, novel evaporative cooling.

Zihua Wang, ASU, A2.1: Climate models, building surface temperature interfaces, comfort analysis, and measurement and simulation of heat exchanges

Matei Georgescu, ASU, A2.2: Climate models, measurement and sensor design, development and deployment

Darrel Jenerette UC-R, A3.2: Leaf temperature analysis and measurement for urban influences and thermal comfort impacts

Sybil Sharville, CSU, B1.1: Interface of building systems for water efficiency with energy efficiency aspects

Thomas Meixner UA, B2.1: Interface of urban water management systems with thermal exchanges, perceived temperatures and evaporative cooling or losses.

Gary Pivo, UA, C1.1: Consideration of urban planning and design alternatives to address challenges presented in the energy-water-infrastructure nexus.

6. References Bruelisauer, Marcel, Forrest Meggers, Raphael Engler, and Hansjürg Leibundgut. 2013. “Heat Bus for the Tropics – Exergy Analysis of Coupling Decentralised Chillers with Central Cooling Towers.” In Proceedings of Clima 2013: Energy Efficient, Smart and Healthy Buildings. Prague, Czech Republic. Dhakal, Shobhakar, and Keisuke Hanaki. 2002. “Improvement of Urban Thermal Environment by Managing Heat Discharge Sources and Surface Modification in Tokyo.” Energy and Buildings 34 (1): 13–23. doi:10.1016/S0378-7788(01)00084-6. Hall, I. J., R. R. Prairie, H. E. Anderson, and E. C. Boes. 1978. “Generation of a Typical Meteorological Year.” SAND-78-1096C; CONF-780639-1. Sandia Labs., Albuquerque, NM (USA). http://www.osti.gov/scitech/biblio/7013202. Jenerette, G. Darrel, Sharon L. Harlan, William L. Stefanov, and Chris A. Martin. 2011. “Ecosystem Services and Urban Heat Riskscape Moderation: Water, Green Spaces, and Social Inequality in Phoenix, USA.” Ecological Applications 21 (7): 2637–51. doi:10.1890/10-1493.1. Khan, Jameel-Ur-Rehman, M. Yaqub, and Syed M. Zubair. 2003. “Performance Characteristics of Counter Flow Wet Cooling Towers.” Energy Conversion and Management 44 (13): 2073–91. doi:10.1016/S0196-8904(02)00231-5. Kuenzel, Hartwig M., and Kurt Kiessl. 1996. “Calculation of Heat and Moisture Transfer in Exposed Building Components.” International Journal of Heat and Mass Transfer 40 (1): 159–67. doi:10.1016/S0017-9310(96)00084-1.

67

Meggers, Forrest, Gideon Aschwanden, Eric Teitelbaum, Hongshan Guo, and Marcel Bruelisauer. 2015. “Urban Cooling Potential: System Losses from Microclimates.” In 6th International Building Physics Conference, IBPC 2015. Torino, Italy: Elsevier. Meggers, Forrest, Luca Baldini, Marcel Bruelisauer, and Hansjürg Leibundgut. 2012. “Air Conditioning without so Much Air – Low Exergy Decentralized Ventilation and Radiant Cooling Systems.” In Proceedings of the 5th IBPC: The Role of Building Physics in Resolving Carbon Reduction Challenge and Promoting Human Health in Buildings, edited by The 5th IBPC organizing committee, 529–36. Kyoto, Japan. Meggers, Forrest, Luca Baldini, Philippe Goffin, Matthias Mast, and Hansjürg Leibundgut. 2011. “Improving the Way We Improve Buildings.” In Proceedings of WEC 2011. Geneva, Switzerland. Meggers, Forrest, Volker Ritter, Philippe Goffin, Marc Baetschmann, and Hansjürg Leibundgut. 2012. “Low Exergy Building Systems Implementation.” Energy 41 (1): 48–55. doi:10.1016/j.energy.2011.07.031. Meroney, Robert N. 2008. “Protocol for CFD Prediction of Cooling-Tower Drift in an Urban Environment.” Journal of Wind Engineering and Industrial Aerodynamics, 4th International Symposium on Computational Wind Engineering (CWE2006), 96 (10–11): 1789–1804. doi:10.1016/j.jweia.2008.02.029. Ramamurthy, P., and E. Bou-Zeid. 2014. “Contribution of Impervious Surfaces to Urban Evaporation.” Water Resources Research 50 (4): 2889–2902. doi:10.1002/2013WR013909. Teitelbaum, Eric, Forrest Meggers, George Scherer, Prathap Ramamurthy, Louis Wang, and Elie Bou-Zeid. 2015. “ECCENTRIC Buildings: Evaporative Cooling in Constructed ENvelopes by Transmission and Retention Inside Casings of Buildings.” In 6th International Building Physics Conference, IBPC 2015. Torino, Italy: Elsevier. Wang, Zhi-Hua, Elie Bou-Zeid, and James A. Smith. 2013. “A Coupled Energy Transport and Hydrological Model for Urban Canopies Evaluated Using a Wireless Sensor Network.” Quarterly Journal of the Royal Meteorological Society 139 (675): 1643–57. doi:10.1002/qj.2032. William J. Coirier, Sura Kim. 2007. “Progress towards a Coupled Mesoscale and Microscale Modeling Capability.”

68

PROJECT A3-1 RESEARCH PLAN

Project Title Continental Scale Variation in Urban Vegetation Biodiversity – Ecosystem

Functioning UWIN Project number A3-1 Project Lead Jenerette, UCR Investigators/Institutions

Swan, UMBC (Funded) Shirley Papuga, UA (Funded) Mike Crimmins, UA (Unfunded) Meixner, UA (Unfunded)

Project Period 1/12016-12/31/2018 Project Cost

UCR: $$250,000 UMBC: $$200,000 UA: $$200,000

Graduate students funded to work on project

Peter Ibsen, Supervised by Jenerette (50% effort) TBA, Supervised by Jenerette (50% effort) Dorothy Borowy, Supervised Swan (50% effort) TBA, Supervised by Papuga (50% effort)

Project Overview

Our project will lead to the quantification of urban biodiversity and ecosystem functioning and assessments of potential trajectories for these components in contribution to the UWIN blueprint for each study region. In doing this we will contribute to a growing macrosystem perspective of urban ecosystems and test novel theories of continental scale variation in urban biodiversity-ecosystem functioning relationships. We will conduct this work through a combination of field surveys, analysis of high resolution (1m) remotely sensed imagery, and focused study of existing green infrastructure (GI) across UWIN regions. Surveys of field biodiversity and cover will be conducted at scales of 30 m to 1 ha. These will feature teams of local citizen scientists, local graduate student, and funded members of our project. In addition to direct blueprint outcomes and new scientific discoveries within this project domain, we will also contribute to training and outreach goals of UWIN.

Project Summary

Objectives: 1. Assess vegetation biodiversity distributions and vegetation density distributions throughout each UWIN

region. 2. Evaluate effects of vegetation biodiversity and density on a key amenity, local cooling within UWIN regions. 3. Identify trajectories of changing vegetation biodiversity, density, and ecosystem amenity trade-offs within

each UWIN region. 4. Conduct targeted studies of GI in UWIN regions to evaluate importance of vegetation distributions to

production of ecosystem services and associated water demands. 5. Provide training opportunities for graduate students, undergraduate students, and citizen scientists to better

understand urban ecological concepts associated with vegetation and their connection to water sustainability.

Approach: Our approach includes a combination of in-situ biodiversity assessments, remotely sensed leaf area assessments, and sensor based environmental monitoring. This work will use a combination of project scientists, other UWIN scientists, local citizen scientists, and partner organizations. We will target a geographical subset of each UWIN region, and specific land uses of parks and riparian areas for assessments. Selected GI installations will be instrumented for immediate science applications and as a platform for future field intensive or high resolution

69

modeling activities. Land cover change modeling will be used to assess future states of key blueprint components. Expected Results: We will provide quantitative measures of biodiversity and ecosystem functioning, their relationship to a cooling-irrigation trade-off, and future alternatives for the UWIN blueprint. We will target efforts to GI installations as part of a cross-theme UWIN activity. Our work will help identify city specific practices regarding vegetation, including species, location, and management, that is most conducive to improving urban sustainability. Across all cities, we will evaluate alternate hypotheses of urban macrosystem patterns and advance theories of continental scale ecological dynamics. Supplemental Keywords Ecosystem services, co-benefits

Project description

1. Objectives 1. Assess vegetation biodiversity distributions and vegetation density distributions throughout each UWIN

region. 2. Evaluate effects of vegetation biodiversity and density on a key amenity, local cooling within UWIN regions. 3. Identify trajectories of changing vegetation biodiversity, density, and ecosystem amenity trade-offs within

each UWIN region. 4. Conduct targeted studies of GI in UWIN regions to evaluate importance of vegetation distributions to

production of ecosystem services and associated water demands. 5. Provide training opportunities for graduate students, undergraduate students, and citizen scientists to better

understand urban ecological concepts associated with vegetation and their connection to water sustainability.

2. Intellectual Merit Our project fills a critical component of the sustainable water blueprint, a key intellectual contribution of the UWIN project. The project will also test outstanding hypotheses in how national scale patterns of urban biodiversity and leaf area are distributed within the United States. This work will lead to the development and testing of a new macrosystems theory of urban vegetation diversity – functioning relationships and their influence on a cooling amenity. 3. Activities

Activity 1: Continental Scale Variation in Urban Plant Biodiversity

We will assess patterns of urban tree biodiversity throughout the UWIN regions. We will conduct surveys within randomly distributed plots, parks, and riparian areas. These three foci of analysis will allow us to use randomization and plots that are widely used in ecology and specifically target two ecosystem critical urbanization and hydrological sustainability. Tree biodiversity assessments will be conducted in years 1-3 of the project. Research teams will be unique for each UWIN city. UCR will support Southern California and Pacific Northwest [travel funding is not currently allocated to PN]. UMBC will support Mid-Atlantic and Southeast Florida [travel funding is not currently allocated to SF]. UA will support Sun Corridor and Denver [travel funding is not currently allocated to Denver]. Biodiversity surveys will be conducted by a combination of supported graduate student, local student, and local naturalist partners. Local support by some graduate student effort to enable research (2 month time per year / region?) will be used. Local naturalists within each region will be recruited as citizen science partners who contribute to the field research activities. The inclusion of citizen scientists is useful as these individuals are likely to have local expertise in species identification and the local environment. We plan to work with Earthwatch Institute, a nonprofit citizen science organization with whom Jenerette has developed an urban vegetation based program for the greater Los Angeles region currently in progress. These partners will be engaged closely with other UWIN

70

citizen science activities. We will conduct tree biodiversity surveys using three alternate approaches. Randomized plot tree biodiversity assessments will be conducted within each region using a spatially stratified design. Recently, 30 1 ha plots were deployed in Los Angeles that coincided with 300 0.09 ha plots (Gillespie et al. In Review). These large spatially random plots have provided many benefits over the more common small plots used in much urban tree biodiversity studies, including linkages with remotely sensed data at appropriate scales. Because species identity can be obtained from sight, these surveys can be conducted from streets and alleys without entering private property boundaries. We will randomly place 30 1 ha plots within each study region. In addition to land use independent sampling, we will also target biodiversity assessments for two valued land uses, urban park and riparian. Within parks and riparian areas, we will focus on developing full tree inventories within 20-50 randomized selected parks and riparian reaches. Data from these surveys will be analyzed through standardized community ecology approaches that assess biodiversity from phylogenetic and functional perspectives. We can quantify the phylogenetic diversity for the whole city scale and evaluate how this diversity is partitioned between parks and riparian areas. Similar to other urban phylogenetic diversity analyses we will construct an aged phylogenetic tree from species inventories using Phylomatic (http://www.phylodiversity.net/phylomatic; Webb and Donoghue, 2005). We will quantify phylogenetic community structure at each scale using net relatedness index and nearest taxon index (Kraft and Ackerly 2010). These approaches use both presence and abundance information to characterize community phylogenetic diversity. We will assess functional diversity through a combination of plant functional traits and ecosystem services classifications to create an ecosystem service-based functional classification (Pataki et al. 2013). The goal is to create measurable metrics of the key traits used by urban residents and managers to make decisions about which species to plant and remove in the urban environment. Plant functional trait analysis will be similar to the phylogenetic analysis. First, we will separately calculate functional trait diversity (Laliberté and Legendre 2010) at the neighborhood level for plant visual cues, plant gardening traits, and plant cultural uses. Scaling up, we will calculate functional diversity for each region. These results will allow us to assess relationships between functional and phylogenetic biodiversity within each region. Along with the diversity within a sampling unit (e.g. city), increasingly the turnover in species between locations, or beta diversity, is used as a fundamental linkage between local or regional taxonomic biodiversity (Anderson et al. 2011).

Activity 2: Mapping 1m Resolution Dynamics of Vegetation Cover and Influence of Surface Cooling

For each UWIN study region we will develop 1m resolution maps of leaf area using publically available high resolution (1m) imagery across a decade of change. This work will be conducted in years 1-4 of the project. Increasingly, a high spatial resolution urban classification approach is being used to understand land cover patterns within individual cities but to date no analysis across a network of cities has been conducted (Bigsby et al. 2014; Myint et al. 2015). Our data for classification will be the National Agricultural Imaging Product (NAIP, a free USDA data product) that provides 3 band visible and NIR data (thus allowing full color imagery and calculation of the most common greenness index, NDVI) during the dominant growing season for the conterminous United States. NAIP data were collected at a 5 year interval from 2003-2008, which has then transitioned to a 3 year interval. We will select a 10 year interval based on locally available data to capture current and recent historic imagery. Land cover classification will be conducted using an object oriented image classification (eCognition) that has been found useful in urbanized regions with high heterogeneity at fine scales. From the classified imagery we will extract landscape distributions of both tree and grass/herb for each city. Further, for each land cover object we can quantify its greenness uniquely through NDVI. Combining the vegetated land cover distributions and their NDVI values we will estimate their leaf area distributions at the sub-parcel scales. These analyses will use current and 10 year historic NAIP imagery and subsequent land cover classifications to evaluate trajectories of 1m decadal vegetation change. We will connect the specific land cover patterns to a critical ecosystem service based metric, local land surface cooling through overlays of satellite based (90 m resolution) land surface temperatures (LST) and our high resolution land cover maps. While several studies have linked vegetation as quantified by NDVI and LST (e.g. Jenerette et al. 2011; 2007), coupling high resolution land cover with LST has only been conducted once (Myint et al. 2015) and no

71

studies have examined vegetation LST relationships across a network of cities. The effectiveness of vegetation density and configuration for cooling will be assessed through comparisons with Landsat derived land surface temperature measurements, which are widely used for urban microclimate studies but have not been used in conjunction with high resolution land cover patterns or across a network of cities. While the local assessments are regionally valuable, our efforts are primarily directed to developing within-city metrics as a tool to evaluate cross-city macrosystem hypotheses.

Activity 3: Green infrastructure, Biodiversity, and Ecosystem Functioning: A Focal UWIN Nexus

The opportunity to conduct a networked urban field study is a novel and potentially transformative activity. We are proposing a strategic study of the interactions between biodiversity and ecosystem functioning within GI across all UWIN regions. This activity includes installing an environmental sensor and biodiversity network, and a distributed experimental network across all UWIN regions. With current resources we can deploy the sensor network and species inventory but need to identify new resources for experimental work. Sensor and Biodiversity Observatory Network

Environmental sensor networks are increasingly used to monitor ecosystem changes, although to date no such network has been deployed across a continental scale gradient of urbanization. We will install “phenocams” (e.g. Crimmins and Crimmins 2008; Kurc and Benton 2010) that record daily digital images in six GI setting, three in actively managed and three in unmanaged installations. These imaging sensors will provide key information on vegetation dynamics within GI and serve as a platform for additional sensor deployments or other intensive field measurements or modeling. Pheno-cams provide an invaluable daily archive of information related to plant phenological activity that simply cannot be maintained using more traditional monitoring methods. Furthermore, using image processing software, information such as species-specific greenness and percent cover can be easily extracted from pheno-cam images and monitored over time. Importantly, time series resulting from these images can be directly related to daily time series of environmental drivers such as precipitation or temperature and also to time series of water and carbon fluxes. A daily record of plant phenological activity at GI sites would enable evaluation of the limiting resources in each of the regions, how it relates to biodiversity in the GI and the ecosystem services it can provide and how those might change over the course of the year. Furthermore, data provided by the phenocams would inform decisions regarding how to best instrument GI systems for addressing related UWIN theme areas including Urban Heat Island, Flooding, and infiltration. Such future sensor deployments include mesonets of micrometerological variation, sapflux sensors, and other soil sensors. Biodiversity – functioning experiments

Experiments, which include aspects of treatment randomization, replication and control, are generally considered the most robust approach for the study of biodiversity and ecosystem functioning. Network experiments are increasingly powerful approach to understanding continental scale variation in functional responses to manipulated treatments. The experiment will be planned in year 1, installed in year 2, and actively supported by the project in years 3-5. We expect meaningful results within 2 years of installation. We also expect these experiments to have use for several years after the project concludes. At present we plant to conduct experiments only in Southern California, Sun Corridor, and the Mid-Atlantic (sites with project investigators). Our current concept is based on recently deployed experiments in abandoned lots throughout Baltimore, MD. The focus here is on how native vegetation on abandoned land in cities can function to direct water to desirable sinks, including soil water, groundwater, and evapotranspiration. This study will experimentally identify the role soil plays in establishing native plant communities, as well as how removal of invasive plant species mediates establishment. Plot scale (2 m x 2 m) manipulations of soil type (urban fill vs native topsoil), native plant composition, and weeding will be performed in a full factorial design to learn: 1) The interaction between soil type and invasive competitors on establishment of native plant communities, and (2) the associated consequences for local water cycling. We propose to install 32 raised beds (16” high) in each of three cities, with n=6 allocated to each soil x weeding treatment, plus n=4 unplanted plots per soil treatment to follow ambient colonization. Bimonthly measurements of soil quality (OM, water holding capacity),

72

water using efficiency, specific leaf area will be taken through the growing season in addition to ambient environmental conditions (surface and soil temp, pH, etc.). Net primary production and soil nutrient analysis will be taken at the end of the growing season. We anticipate establishing plots in Year 1, and sampling for three growing seasons. Native species will be selected from available seed sources for each region separately with a targeted goal of 10-12 species established per bed.

Activity 4: Trajectories of biodiversity, vegetation density, and ecosystem service-water demand trade-off

We will develop assessments of future patterns of biodiversity and vegetation. Preliminary efforts have been directed toward developing an agent based modeling framework for individual urban parcel vegetation management. We have some thoughts on how a model could map desires for species traits, species availability, and their capacity to buy/manage the plants in a coupled multinomial logistic framework. An output of the parcel model would be the suite of ecosystem services and associated and water use This project activity will use urban vegetation change modeling we are currently developing and depend strongly on collaborations with the ENVISION and other modeling teams. This project activity will be developed during years one through three and conducted in years four and five.

4. Expected Results, Benefits, Outputs, and Outcomes Our results will show that vegetation diversity and density vary dramatically within and between urban regions. Our hypotheses predict this variation will result from patterns of income and development history within individual cities, and climate and biome distributions between cities. The work will help understand distributions of urban biodiversity and ecosystem functioning and help identify regionally specific species for meeting ecosystem service – water use trade-off goals. Alternative future trajectories will identify how the plasticity in urban biodiversity and ecosystem functioning also varies between regions. The outputs of the research will result in a suite of targeted urban-ecological publications and contributions to the UWIN blueprint. The outcomes will lead to better urban vegetation management. 5. General Project Information We request local support for field campaigns at each of the UWIN study regions (2 month graduate student time per year / region?). We request coordination with citizen scientist activities through education and outreach UWIN teams. A great additional dataset would be high resolution thermal imagery for a subset of each region collected by the MASTER instrument and contracted from NASA. For all six, this would cost about $80,000-100,000 to get to useable / consistent products across each region (Figure 2 provides example of resulting data collected from Phoenix). We are looking for additional local/project support to expand on instrumentation and experimentation in selected GI. We see this project directly collaborating with other research projects directed to GI (B2-1 Meixner), rainfall - flood frequency (B1-1, 2 Smith), Envisions modeling (D1-1 ENVISION team), cross site comparisons (D1-2 Sukop) and Outreach / Education (Sukop and Berkowitz). The project likely has connections that could be expanded with earlier plans directed to embedded sensing (Georgescue) and radiant energy modeling (Forrest). If Colorado State (Arabi) is developing land cover maps the development of vegetation cover maps should be coordinated. Outputs from this project likely could be used by land-atmosphere modeling activities (Bou-Zeid, Georgescue) and social justice research (Harlan). 6. References Anderson, M. J., T. O. Crist, J. M. Chase, M. Vellend, B. D. Inouye, A. L. Freestone, N. J. Sanders, H. V. Cornell, L. S. Comita, K. F. Davies, S. P. Harrison, N. J. B. Kraft, J. C. Stegen, and N. G. Swenson. 2011. Navigating the multiple meanings of beta diversity: a roadmap for the practicing ecologist. Ecology Letters 14:19-28.

73

Bigsby, K. M., M. R. McHale, and G. R. Hess. 2014. Urban morphology drives the homogenization of tree cover in Baltimore, MD, and Raleigh, NC. Ecosystems 17:212-227. Crimmins, M.A. and Crimmins, T.M., 2008. Monitoring plant phenology using digital repeat photography. Environmental Management, 41(6): 949-958. Jenerette, G. D., S. L. Harlan, A. Brazel, N. Jones, L. Larsen, and W. L. Stefanov. 2007. Regional relationships between surface temperature, vegetation, and human settlement in a rapidly urbanizing ecosystem. Landscape Ecology 22:353-365. Jenerette, G. D., S. L. Harlan, W. L. Stefanov, and C. A. Martin. 2011. Ecosystem services and urban heat riskscape moderation: water, green spaces, and social inequality in Phoenix, USA. Ecological Applications 21:2637-2651. Kraft, N. J. B. and D. D. Ackerly. 2010. Functional trait and phylogenetic tests of community assembly across spatial scales in an Amazonian forest. Ecological Monographs 80:401-422. Kurc, S.A. and Benton, L.M., 2010. Digital image-derived greenness links deep soil moisture to carbon uptake in a creosotebush-dominated shrubland. J. Arid. Environ., 74(5): 585-594. Laliberte, E. and P. Legendre. 2010. A distance-based framework for measuring functional diversity from multiple traits. Ecology 91:299-305. Myint, S. W., B. Zheng, E. Talen, C. Fan, S. Kaplan, A. Middel, M. Smith, H.-P. Huang, and A. Brazel. 2015. Does the spatial arrangement of urban landscape matter? Examples of urban warming and cooling in Phoenix and Las Vegas. Ecosystem Health and Sustainability 1:art15. Pataki, D. E., H. R. McCarthy, T. Gillespie, G. D. Jenerette, and S. Pincetl. 2013. A trait-based ecology of the Los Angeles urban forest. Ecosphere 4. Webb, C. O. and M. J. Donoghue. 2005. Phylomatic: tree assembly for applied phylogenetics. Molecular Ecology Notes 5:181-183.

74

Figure A3-1-1 – High resolution imagery used in coordination with 1 ha field surveys. Upper panel shows mapping based solely on image data, lower panel shows changes resulting from subsequent field surveys (from Gillespie et al. In Review).

75

Figure A3-1-2: Representative overlay of land cover data, parcel boundaries, and MASTER LST

measurements for a highly vegetated neighborhood in Phoenix, AZ. The upper left and upper

right show day and night time LST (°C) and parcel boundaries. The lower left shows land-

cover classification and parcel boundaries. The lower right shows a detail subset of the daytime

LST and parcel delineations. Data from Jenerette et al. (In Press).

76

PROJECT B1-1 RESEARCH PLAN

Project Title B1-1: Water management solutions to enhance capacity for use of alternative water sources

UWIN Project number

B1-1

Project Lead

Sybil Sharvelle (CSU)

Investigators/Institutions (Funded/not funded to work on project)

Arpad Horvath (Berkeley) Charles Glass (Howard) Ali Mostafavidarani (UFI) Tom Meixner (UA) Mazdak Arabi (CSU)

Project Period

August, 2015 - July, 2020

Project Cost

CSU ($354,580) Berkeley ($120,000) Howard ($75,000) UFI ($93,740) UA ($5,490)

Graduate students funded to work on project (names, supervisors) If not yet hired, note “TBA”, supervisor

TBA (PhD), Sybil Sharvelle Taylor Bradley (undergraduate), Charles Glass TBA (M.S.), Charles Glass TBA, Arpad Horvath TBA (Ph.D.), Mazdak Arabi (funded from indirect cost recovery) TBA (50% split PhD), Ali Mostafavidarani

Project Overview

Stormwater and wastewater discharges pollute our waterways. Meanwhile, use of these non-traditional water sources has the potential to enhance water efficiency in urban environments. A growing approach to enhance urban water sustainability is to minimize the import and export of water, creating net-zero water communities. Decentralized management of water, wastewater and stormwater may have the potential to enhance capacity for water reuse and create more resilient water systems. Conventionally, treatment of water has been managed at a municipal scale and is not integrated. While centralized management of water has made tremendous strides to provide safe drinking water and maintain water quality in receiving water bodies, there are inefficiencies in this approach. Failures at any point of the system can be catastrophic and water reuse can be more costly and energy intensive when conducted at a large scale. Given issues with addressing aging infrastructure in the US, now is the time to consider innovative alternatives that enable integrated management of water, wastewater and stormwater. While some examples of building and community-scale systems that reuse wastewater and stormwater exist, the role of such systems to enhance resiliency of entire cities has not been determined. Additionally, the most appropriate scales and configurations of integrated water systems is unknown. Optimal solutions will vary by region and likely involve hybrid centralized and decentralized management structures where treatment of some water sources will be at the municipal scale and others will be at the building or community scale. The overarching goal of this project is to investigate the role of alternative water management solutions (including decentralized systems) to create multi-purpose water utilities that

77

are capable of reacting to external pressures and determine most appropriate and robust solutions. The project team will evaluate water management solutions such as use of recycled water, graywater reuse, stormwater capture and use, green infrastructure and reduction of irrigation demand. Configuration and scale of these practices will be integral components of the study.

Building Scale Water Recycling System at San Francisco Public Utilities

Project Summary

Objectives The goal of this project is to investigate the role of alternative water management solutions (including decentralized systems) in response to external pressures on urban water systems in the six study regions. Specific objectives of the study are to:

O1: Assess the effects of various scales (building- to neighborhood-level) and configurations of water management solutions on reliability and resilience of water supply under population growth, land use, and climate pressures. O2: Assess co-benefits of alternative water management solutions across different geographic regions, including reduction of energy demand and GHG emissions. O3: Estimate impacts of alternative water management solutions to improve sustainability indicators across different geographic regions.

These objectives will met through investigation of the following hypotheses:

H1: Use of alternative water supplies and water conservation approaches will increase reliability and resilience of urban water systems under pressures (e.g., population growth and climate change).

78

H2: Configuration and scale of water, wastewater and stormwater systems will change sustainability indicators under various alternative scenarios for population and climate. H3: Net-zero water communities will result in significant co-benefits that extend beyond reliability of water supply.

H4: Effectiveness of different water management solutions in terms of sustainability indicators will vary across geographic regions.

Approach/Activities: Water models including the Integrated Urban Water Model (IUWM) and Storm Water Management Model (SWMM) will be used to evaluate the impacts of alternative water management solutions to improve urban water sustainability under increasing pressures (population growth, land use change and climate change). Alternative water management scenarios to be evaluated include the use of recycled water, graywater reuse, stormwater capture and use, green infrastructure and reduction of irrigation demand. The end goal will be to evaluate impact of studied water management solutions on water demand, energy demand, cost for water supply and wastewater treatment and frequency of CSO/SSO events under the identified pressures across study regions. Scale and configuration of the water management solutions will be a key component of the study. Outputs from the water models will be applied to life cycle assessment (LCA) models to estimate environmental benefits and costs associated with evaluated water management solutions. Outputs from water models and LCA will be applied to estimate sustainability indicators for alternative water management solutions under pressures across geographic regions.

Expected Results, Benefits, Outputs, and Outcomes: This project will study how alternative water management solutions that facilitate the transition toward net-zero water communities would impact identified sustainability indicators under baseline and alternative future scenarios for population, land use, and climate. The outputs will include: 1) calibrated and tested UWIN and SWMM models for the study regions for water demand forecasting; 2) estimates of the selected set of listed sustainability indicators for study cities in UWIN regions under scenarios of alternative water management solutions and increasing pressures (population, land use and climate). The estimated sustainability indicators will be incorporated into the Water Sustainability Blueprint to identify most appropriate solutions to improve urban water sustainability based on region specific considerations.

Supplemental Keywords

One water, water conservation, rainwater capture, stormwater harvesting

Project description

1. Objectives Given the identified issues with addressing aging infrastructure in the US (USEPA, 2002), now is the time to consider innovative alternatives that enable integrated management of water, wastewater and stormwater. While some examples of building and community-scale systems that reuse wastewater and stormwater exist, the role of such systems to enhance resiliency of cities has not been determined and the most appropriate scales and configurations of integrated water systems is unknown. Optimal solutions will vary by region and will likely involve hybrid centralized and decentralized management structures where treatment of some water sources will be at the municipal scale and others will be treated and used at the building or community scale (Daigger, 2011). The goal of this project is to investigate the role of alternative water management solutions (including decentralized systems) in response to external pressures on urban water systems in the six study regions..

79

Specific objectives of the study are to: O1: Assess the effects of various scales (building- to neighborhood-level) and configurations of water management solutions on reliability and resilience of water supply under population growth, land use, and climate pressures. O2: Assess co-benefits of alternative water management solutions across different geographic regions, including reduction of energy demand and GHG emissions. O3: Estimate impacts of alternative water management solutions to improve sustainability indicators across different geographic regions.

These objectives will met through investigation of the following hypotheses:

H1: Use of alternative water supplies and water conservation approaches will increase reliability and resilience of urban water systems under pressures (e.g., population growth and climate change). H2: Configuration and scale of water, wastewater and stormwater systems will change sustainability indicators under various alternative scenarios for population and climate. H3: Net-zero water communities will result in significant co-benefits that extend beyond reliability of water supply. H4: Effectiveness of different water management solutions in terms of sustainability indicators will vary across geographic regions.

2. Intellectual Merit While some examples of building and community-scale systems that reuse wastewater and stormwater exist, the role of such systems to enhance sustainability of cities has not been determined and the most appropriate scales and configurations of integrated water systems is unknown. Optimal solutions will vary by region and will likely involve hybrid centralized and decentralized management structures where treatment of some water sources will be at the municipal scale and others will be treated and used at the building or community scale. This project will seek to understand the role of alternative water management solutions toward creating net zero water communities. A holistic approach will be applied to quantify benefits and co-benefits of these solutions under varying pressures. Ultimately, guidance will be provided on most appropriate solutions to improve urban water sustainability in varying geographic regions. 3. Approach/Activities

Task 1: Assess benefits of various water management solutions under baseline and alternative future scenarios for population, land use and climate

The Integrated Urban Water Model (IUWM) has been developed by Dr. Sharvelle (PI) in 2011 as part of a project funded by Water Environment Research Foundation (WERF; Reichel et al., 2011). It is a mass balance model that simulates water demand, wastewater production and cost of urban water management practices. Management practices currently included in the model are municipal wastewater reuse, graywater reuse, indoor conservation, urban irrigation conservation and stormwater capture and use. IUWM has recently been programmed in JAVA and integrated with a non-proprietary geographical information system (GIS) to estimate water demands and wastewater production based on land use type and estimate infrastructure required for cost estimation. The user selects an area, identifies water management scenarios to evaluate and then is provided with outputs such as water savings, wastewater quantity and quality, energy use and whole life cost including capital and operations costs. IUWM is also currently being expanded to evaluate impacts of the included water management practices on nutrient loading to wastewater treatment facilities and subsequent impacts to receiving water bodies through the recently funded USEPA

Center for Comprehensive, optimaL, and Effective Abatement of Nutrients (CLEAN). IUWM will be interfaced

80

with the USEPA Storm Water Management Model (SWMM5; Figure 1). The following alternative water management practices will be investigated through this project:

Use of treated wastewater effluent (building to municipal scale)

Beneficial use of stormwater (residential to municipal scale)

Graywater reuse (residential, building and neighborhood scales)

Green infrastructure (residential, building and neighborhood scales)

Irrigation conservation via xeriscape and efficient irrigation systems (all scales)

Separate supply of raw water for non-potable uses (municipal scale) The SWMM model will be used to assess the effects of green infrastructure on hydrologic and water quality processes and fluxes in the six study regions. In each region, up to five MS4 areas representing various land use/zoning and urban development patterns will be selected. The SWMM model will be developed and calibrated for each area. The calibrated models will be used to assess changes in water quantity and quality as a result of the implementation of GI under varying climatic conditions. Since developing SWMM models for all MS4 systems in the study cities is not feasible, empirical models and/or neural networks will be developed to explain the variability of the performance of GIs across regions as a function of regional climate, land use/zoning categories, and other geospatial factors. The empirical models will be used to evaluate the effects of GIs on water quantity and quality for the study regions under baseline and future climate, land use and imperviousness scenarios. Interfacing of IUWM with SWMM will provide outputs including whole life costs, water demand, energy demand and impact to combined sewer overflows (CSOs) and sanitary sewer overflows (SSOs; Figure B1-1-1). IUWW will be calibrated in an identified study city in each of the UWIN regions. Data required for calibration will include water use data for each municipality on a monthly basis, separated by commercial and residential/multi-residential use. Upon calibration the model, IUWM will be applied to UWIN team identified scenarios of population growth, land use change and climate change. The end goal will be to evaluate impact of studied water management solutions on water demand, energy demand, cost for water supply and wastewater treatment and frequency of CSO/SSO events under the identified pressures across study regions. Most appropriate water management solutions for each region will be identified through this effort. Scale and configuration of the water management solutions will be a key component of the study.

81

Task 2: Expand IUWM for enhanced capability to estimate benefits from Green Infrastructure

The existing version of IUWM does not include green infrastructure (GI). For this project, it will be critical to include GI as a water management solution to be evaluated under pressure scenarios across UWIN regions. We will work closely with the project team for project B2-1 (Strategic Green Infrastructure to minimize runoff and negative water quality feedbacks; led by Tom Meixner) to incorporate GI benefits into IUWM. Tom Meixner and his project team will provide guidance and review for this work, while modifications to IUWM will be made by Dr. Sharvelle's group.

Task 3: Apply Life Cycle Assessment to quantify environmental benefits of water management solutions

Outputs from IUWM from different regions and scenarios for future conditions will be evaluated using Life Cycle Assessment (LCA) to more thoroughly characterize sustainability indicators throughout the process supply chain associated with alternative water supplies and management (Figure 1). Using the UC Berkeley-developed Water-Energy Sustainability Tool (WEST), we will analyze alternative water supplies using LCA to quantify energy efficiency, greenhouse gas (GHG) emissions, and other environmental impacts and the sustainability indicators to estimate the systemwide effects and co-benefits (e.g., nutrient control, avoided overflow events or infrastructure upgrades) of implementing new technologies and management systems using case studies. The WEST tool was created to evaluate conventional water processes, but will be revised to evaluate new alternative water supplies using site- and regionally-specific energy mixes (see Project B1-3 for more information). In addition, environmental results will be compared to business-as-usual conditions for each regional case study developed in Project B1-3.

Task 4: Estimate sustainability indicators under scenarios of increasing pressures

Model outputs from water models and LCA (Figure 1) will be used to estimate all sustainability indicators developed by the UWIN team. The project team believes that the outputs from models used in this project can be used to estimate the entire set of sustainability indicators listed to date. Sustainability indicators will be estimated for each

Figure B1-1-1. Model Interfacing and Outputs

82

study city without changes to existing water management practices and with alternative water management solutions identified in Task 1. Scenarios will include predictions of increasing pressures of population change, land use change and climate change. 4. Expected Results, Benefits, Outputs, and Outcomes This project will evaluate how alternative water management solutions (see Task 1) that move toward net-zero water communities impact identified sustainability indicators under scenarios of increasing pressures. The outputs will include estimates of the entire set of listed sustainability indicators for study cities in UWIN regions under scenarios of alternative water management solutions and increasing pressures (population, land use and climate). The estimated sustainability indicators can feed into the Water Sustainability Blueprint to identify most appropriate solutions to improve urban water sustainability based on region specific considerations. It is anticipated that results and thus the blueprint will be different in each study region. The outcome is that the blueprint will enable decision-makers to make informed decisions on practices and management strategies to adopt that will improve urban water sustainability. The water management solutions to be evaluated and sustainability indicators to be estimated for this project will be vetted by stakeholder groups across UWIN regions. This project will be heavily guided by stakeholder inputs. In addition, discussion of the water management solutions to be evaluated will ensure stakeholder buy in to these solutions. Integration of this project into proposed stakeholder meetings will be key to the success of the project. Because many of the water management solutions to be evaluated can be implemented at the household scale, these solutions may fit well with the proposed citizen science program. It is also anticipated that studies water management solutions would fit into the proposed Urban Water Sustainability MOOC. All members of the project team will advise REU students interested in participating in Project B1-1. 5. General Project Information

Project Timeline

Run scenarios (solutions and pressures) in Fort Collins and Washington DC: Year 1

Run scenarios (solutions and pressures) in other UWIN Cities: Years 2-4

Integrate GI into IUWM: Year 1

Apply LCA to results from Fort Collins and Washington DC: Years 1-2

Apply LCA to results from other UWIN Cities: Years 2-5

Estimate entire set of sustainability indicators for scenarios: Year 5

Jan 2016- Dec 2016: Calibrated SWMM models for the representative areas in the 6 study region

Jan 2017-Dec 2017: Generalization of the potential effects of GIs for the 6 cities

Interaction with Other UWIN Projects

Interactions among Project B2-1 will be essential to ensure that GI is considered among the water management solutions addressed by IUWM. There will also be strong collaboration amongst all projects in Theme B1. We will work closely with Project C2-1 led by Jessica Bolson and Robert Meyer (UPenn) to evaluate adoption of the water management solutions considered in project B1-1. In addition, the project team will use data and results from the ReNUWIt ERC. Our project team will be in communication with the ReNUWIt team to ensure our work is synergistic and complimentary with their work and not redundant (see Project B1-2). Arpad Horvath is on the Project Team for B1-1 and is the lead for B1-2 and thus will coordinate ties with ReNUWit.

83

PROJECT B1-2 RESEARCH PLAN Project Title Spatially- and temporally-informed life-cycle assessment of urban water systems UWIN Project number

B1-2

Project Lead

Arpad Horvath (UC Berkeley)

Investigators/Institutions

Jennifer Stokes (UC Berkeley)

Project Period

August 2015 - July 2020

Project Cost

$280,000

Graduate students funded to work on project

TBA (Arpad Horvath)

Project Overview

The goal of theme B1 is to support a shift to water- and energy-efficient communities, increasing the quality and quantity of local supplies, and reducing dependence on imports. Since existing water infrastructure in many cities is reaching the end of its useful life, now is an ideal time to reconsider design and management of urban water systems with an emphasis on resiliency, sustainability, and adaptability to long-term supply and demand changes that may result from climate change, population growth, urbanization, and other factors. Given the long life of infrastructure, it is critical that we proactively perform rigorous analyses to evaluate potential negative consequences as well as co-benefits associated with innovative approaches before investments are made as they may affect society for years to come. This project will evaluate UWIN-identified urban water innovations using life-cycle assessment (LCA) to assess energy use and emissions within a city/region and the broader economy. LCA, a methodology defined by the U.S. Environmental Protection Agency and the International Organization for Standardization (ISO), quantifies energy and resource inputs and environmental outputs associated with a product, process, or system throughout its supply chain, providing a holistic “cradle to grave” (or “cradle to cradle” when “waste” products can be recovered and reused) analysis of design decisions (see figure). LCA characterizes tradeoffs associated with technology selection, material and energy consumption, design life, scale, etc. Examples of how LCA has been applied to urban water systems previously include: identification of major contributors to water systems’ energy profile, comparison of the effects of different treatment options, assessment of the impact of changing operational control parameters (e.g., distribution system pressure or pump operation regimes), evaluation of the effects of aging infrastructure (e.g., reduction of leak control/maintenance activities). Some of the past work by Berkeley researchers on urban water system LCA is summarized at http://west.berkeley.edu. Project B1-2 will apply LCA to the urban water infrastructure in the UWIN case study cities (Fort Collins, CO, Washington, D.C., and others to be determined later) to predict performance for the business-as-usual case. These results will provide a benchmark for comparing UWIN technology and management innovations identified and analyzed in Project B1-1. Furthermore, Project B1-2 provides a link between UWIN and NSF-funded urban water research being conducted by the Reinventing the Nation’s Urban Water Infrastructure (ReNUWIt) Engineering Research Center (http://www.renuwit.org), a collaboration between Stanford, UC Berkeley, Colorado School of Mines, and New Mexico State. This connection will allow both teams to maximize synergies, expand knowledge base, avoid redundancies, and leverage industry and policy partnerships to encourage widespread adoption of innovative water approaches.

84

Project Summary

Objectives Project B1-3 has several objectives, including: O1: Providing decision-support tools for evaluating non-traditional water sources under current and future conditions using life-cycle assessment (LCA) with regionally-appropriate data. O2: Characterizing existing centralized gray urban water infrastructure in case-study cities to provide a baseline for comparing alternative technologies. O3: Connecting ReNUWIt and UWIN researchers doing similar research to maximize synergies, expand knowledge base, avoid redundancies, and leverage industry and policy partnerships to encourage widespread adoption of innovative water approaches. The hypotheses to be tested are as follows: H1: The optimal balance of conventional and innovative design approaches and priorities for recovering different valuable resources from urban water systems, given life-cycle effects, will vary between cities based on site-specific conditions. H2: If several diverse urban water systems are analyzed using LCA, findings can be transferred for broader application in those regions and nationwide. Approach/Activities: The project will analyze conventional water and wastewater infrastructure using LCA to compare design and planning alternatives based on a holistic analysis of energy use, greenhouse gas (GHG) emissions, and other environmental impacts, including other air emissions, nutrient discharges, avoided overflow events. Tasks are:

Task 1: Create a decision-support tool to evaluate innovative and emerging water systems and technologies using LCA.

Task 2: Characterize existing centralized gray urban water infrastructure in the six UWIN case study cities.

Task 3: Provide connection for knowledge transfer between UWIN and ReNUWIT. Expected Results, Benefits, Outputs, and Outcomes: This project will provide tools and baselines for evaluating energy and resource consumption, and associated impacts, of alternative water management solutions, under current and expected future conditions. LCA results for non-traditional water sources identified and evaluated in project B1-1 can be compared to the business-as-usual case for each case study to inform the Urban Water Sustainability Framework/Blueprint. The analysis of existing infrastructure will help characterize the pressures of aging urban water infrastructure in UWIN regions and will identify areas for improvement. Benchmark performance for the business-as-usual case will help characterize and quantify urban water energy efficiency and GHG emissions, indicators in the Blueprint. Regional and site-specific factors that affect decisions will be identified. By comparing case study results, the project will provide planning guidelines for innovative water systems. The tools and knowledge developed in this task will be available to assist decision-makers outside UWIN as they consider a comprehensive set of economic and environmental factors in their long-range planning.

Supplemental Keywords

Net-zero water, wastewater, water reuse, graywater, stormwater, water conservation

Project description

1. Objectives

85

The goal of theme B1 is to support a shift to energy-efficient water systems, increasing the quality and quantity of local supplies, and reducing dependence on imports. Since existing water infrastructure in many cities is reaching the end of its useful life, now is an ideal time to reconsider design and management of urban water systems with an emphasis on resiliency, sustainability, and adaptability to long-term supply and demand changes that may result from climate change, population growth, urbanization, and other factors. Given the long life of infrastructure, it is critical that we proactively perform rigorous analyses to evaluate potential negative consequences as well as co-benefits associated with innovative approaches before investments are made as they will affect society for years to come. Project B1-3 has several objectives, including: O1: Providing decision-support tools for evaluating non-traditional water sources under both current and future conditions using life-cycle assessment (LCA) with regionally-appropriate data. O2: Characterizing existing centralized gray urban water infrastructure in case-study cities to provide a baseline for comparing alternative technologies. O3: Connecting researchers in ReNUWIt and UWIN doing similar research to maximize synergies, expand knowledge base, avoid redundancies, and leverage industry and policy partnerships to encourage widespread adoption of innovative water approaches. The hypotheses to be tested are as follows: H1: The optimal balance of conventional and innovative design approaches and priorities for recovering different valuable resources from urban water systems, given life-cycle effects, will vary between cities based on site-specific conditions. H2: If several diverse urban water systems are analyzed using LCA, findings can be transferred for broader application in those regions and nationwide. 2. Intellectual Merit Current work evaluating the life-cycle impacts of urban water decision-making in the United States has been occurring in limited geographic scopes (e.g., California, Florida, and New York City). No studies have been conducted that evaluate these effects using consistent methodology across diverse regions. The broad geographic range of UWIN allows for comparisons of urban water systems to reveal consistencies and disparities in the results. Furthermore, past studies have not evaluated the region-specific tradeoffs and relative values, both economically and environmentally, of mining urban wastewater for different beneficial products or co-products: water, electricity, fuel, heat, and nutrients. These results, combined with results obtained from ReNUWIt work, will provide a strong basis for city, regional/state, and national planning and policymaking for more efficient and sustainable urban water systems. 3. Approach/Activities The project will analyze conventional water and wastewater infrastructure using LCA to compare design and planning alternatives based on a holistic analysis of energy use, greenhouse gas (GHG) emissions, and other environmental impacts, including other air emissions, nutrient discharges, avoided overflow events. LCA, a methodology first defined by the Environmental Protection Agency and the International Organization for Standardization (ISO), quantifies inputs and outputs associated with a product, process, or system throughout its supply chain, providing a holistic “cradle to grave” (or “cradle to cradle” when “waste” products can be recovered and reused) analysis of the environmental effects of a technological decision (see figure at right). The process characterizes tradeoffs associated with technology selection, material and energy consumption, design life, scale, etc. Examples of how LCA has been applied to urban water systems previously include: identification of major

86

contributors to water systems’ energy profile, comparison of the effects of different treatment options, assessment of the impact of changing operational control parameters (e.g., distribution system pressure or pump operation regimes), evaluation of the effects of aging infrastructure (e.g., reduction of leak control/maintenance activities). Some of the past work by Berkeley researchers on urban water system LCA is summarized at http://west.berkeley.edu.

Task 1: Create a decision-support tool to evaluate innovative and emerging water systems and technologies using LCA.

The UC Berkeley research team has already created first versions of LCA-based decision-support tools to aid water utilities (the Water-Energy Sustainability Tool [WEST]), wastewater utilities (the Wastewater-Energy Sustainability Tool [WWEST]), and a streamlined on-line tool, WESTWeb. We will update existing decision-support tools (WEST, WWEST) and/or create a separate but integrated tool to perform LCA of non-traditional water sources and alternative water management. The tool(s) will be updated with regionally-appropriate energy mixes and other emission factors for the U-WIN regions using publicly-available data (e.g., U.S. Energy Information Administration reporting and forecasts) or contacts with local energy providers. In addition, the tool(s) will be expanded to allow scenario evaluation for anticipated future conditions (e.g., energy supplies).

Task 2: Characterize existing centralized gray urban water infrastructure in the six U-WIN case study cities.

Urban water systems in case-study cities will be analyzed using the updated and expanded WEST tools. Case-study analysis will take into consideration factors such as demographics (current and projected population), capacity for gray water systems, and current utility planning and challenges. Existing infrastructure will be inventoried and analyzed using LCA to estimate current energy consumption, GHG emissions, and other environmental impacts. Projected expansion and significant repair required to continue business-as-usual operation over coming decades will be considered. The analytical results will provide a baseline for comparing the infrastructure and resources needed to supply non-traditional water sources in each region. As in Project B1-1, Fort Collins, CO and Washington, D.C. will be analyzed initially. Additional cities in the other four other regions will be identified in collaboration with the B1-1 team. Data will be obtained from publications and communications with municipal and utility staff.

Task 3: Provide contact point for knowledge transfer between UWIN and ReNUWIT.

We will provide a contact point between researchers in ReNUWIt and UWIN doing related research to maximize synergies, expand the knowledge base, avoid redundancies, and leverage industry and policy partnerships to encourage widespread adoption of innovative research approaches. 4. Expected Results, Benefits, Outputs, and Outcomes This project will provide tools and baselines for evaluating energy and resource consumption and associated impacts, and alternative water management solutions, now and with expected future conditions. LCA results for non-traditional water sources identified and evaluated in project B1-1 can be compared to the business-as-usual case for each case study to inform the Urban Water Sustainability Framework/Blueprint. The analysis of existing infrastructure will help characterize the pressures of aging urban water infrastructure in UWIN regions and will identify areas where predicted performance can be improved. Benchmark performance for the business-as-usual case will help characterize and quantify urban water energy efficiency and GHG emissions, indicators in the Blueprint. Regional and site-specific factors that affect planning and design decisions will be identified. By comparing case study results, we will provide planning guidelines for innovative water systems that can be applied nationwide. The tools and knowledge developed in this task will be available to assist decision-makers outside UWIN as they consider a comprehensive set of economic and environmental factors in their long-range planning. 5. General Project Information

Schedule:

Update WEST tools: Years 1-2

87

Evaluate existing infrastructure in Fort Collins and Washington, D.C.: Year 2 Evaluate existing infrastructure in other UWIN Cities: Years 3-5 Provide planning recommendations for UWIN cities: Year 5 Provide connection with ReNUWIt: Years 1-5

Interaction with Other UWIN Projects

We expect there to be strong collaboration among the theme B1 researchers, particularly with theme 1-1, where B1-3 researchers will also contribute. Our work will provide the tools needed and baseline for comparison for that research. In addition, we will continue to monitor ReNUWIt projects for points of connection with UWIN to ensure research is synergistic and complimentary.

88

PROJECT B2-1 RESEARCH PLAN

Project Title Comparative impact of Green Infrastructure impact across UWIN Urban Systems

UWIN Project number

B2-1

Project Lead

Thomas Meixner

Investigators/Institutions Thomas Meixner UA (funded) Phil Guertin UA (funded) Shirley Papuga UA (not funded) Mike Crimmins US (not funded) Brian Bledsoe CSU (not funded) Sybil Sharvelle, CSU (not funded) Mazdak Arabi, CSU (not funded) Jim Smith Princeton (not funded) Andy Miller UMBC (not funded), Darrel Jeanrette UCR (not funded), Chris Swann UMBC (not funded)

Project Period

9/1/2015-8/31/2020

Project Cost

$223,160 at UA

Graduate students funded to work on project

TBD- UA TBA (Ph.D.), Mazdak Arabi (funded from indirect cost recovery)

Project Overview

Green infrastructure (GI) has been proposed as a tool to enable the minimization of water, quanitity and quality hazrds that result from urbanization and also to realize co-benefits through the distribution of water to achieve other desirable ecosystem services (e.g. shade, heat-island mitigation, aesthetic pleasure). While information are available regarding the point scale (e.g single household or parking lot) benefits of GI less is known about how GI implimentation from the individual structure to the catchment and city scales influence the direct and indirect benefits ascribed to GI. In this project our team seeks to understand how the direct benefits of GI might scale from point to city scale using a suite of models to assess background hydrologic conditions. These models will then be implemented with increased GI features and densities for each of the studied cities where possible actual GI implementation will be assessed using available stream flow and water quality data to assess the various models ability to simulate GI implementattion effectiveness under real world conditios.

Project Summary

Objectives: We seek to investigate the following arid to humid dichotomies:

1) GI hazard mitigation will differ in humid versus arid climates a) Arid regions focus on putting water to use to supplement water supplies b) Humid regions focus on minimizing runoff; increasing water quality treatment

2) Co-benefits will differ as well a) Arid regions focus on heat island, shade and the support of natural and landscape vegetation

89

b) Humid regions focus on aesthetic Our experimental approach will emphasize the use of models to investigate the implementation of GI from the parcel (100 m2) to catchment (104 km2).

1) Models will be implemented in three cities minimum (Baltimore, MD; Sun Corridor, AZ; and Front Range, CO) and extended to additional metro areas as data and expertise are developed.

2) Models will investigate scenarios of differing GI practice (e.g. small scale rainwater harvesting, pervious pavement, retention/detention basins) and combinations thereof.

3) Water balances will be computed at different scales to evaluate how water is distributed and how it can be potential used to provide co-benefits.

4) GI implementation will be optimized to meet different societal demands - aesthetics, greenness, biodiversity, equity, etc.

Expected results: We hope to provide a suite of analyses of GI tools that will permit engineers, landscape designers, and policy makers to arrive at best practices of GI implementation beyond the site scale for variable climate and built environment conditions.

Supplemental Keywords

Low-impact development, watershed modelling, ecosystem services, scale

Project description

1. Motivation An effectively integrated biological-physical understanding of the hydrologic, biogeochemical and ecological processes in urban areas should build off of analogous systems where research has revealed insights into connectivity relationships and associated management (e.g. Belnap et al. 2005) as well as past work on urban systems (e.g. Pataki et al. 2011, Grimm et al. 2008a&b). Specifically, research into changes in the vegetation patch distribution in drylands has shown that runoff from bare ground patches provides inflow of water, sediment and nutrients to nearby vegetated patches, supporting more vegetation than would be possible if there were no runoff (Figure 2. Reid et al. 1999, Wilcox et al. 2003, Urgeghe et al. 2010). Similarly in more humid climates research in forested systems has demonstrated the importance of intact and connected hydrologic and biogeochemical systems and their ability to respond to disturbance in such a way as to minimize the impact of disturbance (Bernhardt et al 2005). Patches of impervious and pervious cover could induce similar patterns of water flow and biogeochemical ecosystem services in cities but urban systems are often structured to connect impervious surfaces and route water away from the local environment (Moglen 2009). Promoting disconnection of the flow network through the use of GI can provide co-benefits for build environment by emulating a natural environment. 2. Objectives Hypotheses- the optimal density of Green Infrastructure (GI) to maximize water quality and water quantity ecosystem services as well as other ecosystem services differs based on climate (precipitation regime –amount intensity, frequency) with gentler climates encouraging more centralized GI while more intense climates requiring more distributed GI systems. Similarly, arid climates require a distributed but less frequent GI implementation while wetter climates need a great population of GI sites. 3. Intellectual Merit While GI is being touted as an approach to minimize the water quality and flood hazards related to the urban

90

footprint and much is known about how GI systems work within the context of individual installations less is known about how these systems integrate at the scale of catchments (notable exception Smith et al. 2015). In our proposed work we seek to analyze available data from multiple metro areas and develop simulation models of these setting to investigate the impact of green infrastructure approaches implemented at the catchment to city scale.

4. Approach GI practices aim to increase the capture and infiltration of rainfall and decrease surface runoff. As a result, the design and placement of LID features is a key planning issue for any urban development. An integrated picture of the impact of GI on cities is important for understanding how social and climatological drivers of change in urban systems affect the structure of these systems and influence the relevant ecosystem services and sustainability outcomes of the urban form (Figure 1??). Developing integrated knowledge of the impacts of GI at the neighborhood (~100-1000 ha) and city scale (10,000-1,000,000 ha) requires a coupled observation and modeling framework. We will use this framework to understand the ways in which 1) rain water moves through the urban environment and controls biological, biogeochemical, vegetation growth and shade production processes, 2) solutes in storm water are biogeochemically processed and 3) how GI approaches alter rainfall partitioning and the biogeochemical and biological processes it supports. Our project will initially focus on the Front Range, mid-Atlantic and Sun Corridor regions. These regions differ sharply in their climates varying from semi-arid continental, to humid sub-tropical, to hot desert respectively. Once an approach is developed for these metros we hope to extend it to Southern California, Miami and Portland region as data in those regions comes on line. In our proposed work we will compile data for several neighborhoods/catchments of the scale of 100-1000 ha in each of the three metro areas. Each of these catchments will be characterized in detail for properties related to green infrastructure, hydrologic response, land use, land cover, biological conditions and soils. The sources of the data to conduct these detailed characterizations will include airborne LiDAR, satellite remote sensing, aerial photography, available land cover, land use and soils data, and with local cooperators ground based investigations of GI implementation in these cities. Once these data sets are compiled the focus of investigation will shift to modelling these systems. First in a model calibration/validation framework and then in a synthetic scenario approach. The calibration/validation step will ensure the robustness of model applications in variable situations that the scenarios represent. This work will utilize the Kineros- Opus 2 model (K2-O2) (Fig. 2-1-1)

91

Figure 2-1-1. Schematics of a) model structure and b) representation of urban systems in KINEROS, c) summary of catchment outflow simulation of a Sierra Vista, AZ development, d) event level response for 3 modeled circumstances. The simulations show that rainwater harvesting has the potential of returning the hydrology of the catchment to pre-development conditions.

AGWA Kineros2- Opus 2 (K2-O2)

The individual natural system hydrologic, biogeochemical and ecohydrologic processes can be studied separately and in concert. Such studies cannot capture the full spectrum of possible situations that may arise in the real systems. To understand the integrated response of natural systems to climate variability and human actions related to landscape alteration, simulation models play an important role. Such simulations require an approach that enables models to be constructed and tested based on real world observations before being used to simulate conditions that have not been observed. Modeling the hydrologic, biogeochemical, and vegetative response in urbanize landscapes often requires detailed information about the patterns and structure of urban systems. Past efforts to understand the impact of urban impervious surfaces on runoff quantity and quality impacts have demonstrated specific challenges. First, it is important to adequately capture the complexity of the urban environment (Belt et al. 2009). Second, the particular arrangement of pervious and impervious surfaces is critical to properly represent the generation of runoff in urban settings (Mejia & Moglen 2009). Finally, it is strongly recommended that Geographic Information Systems (GIS) are used to integrate model inputs and provide ready access to decision support tools for a variety of decision maker audiences (Tsihrintzis & Hamid 1997). The Automated Geospatial Watershed Assessment Tool (AGWA) has been developed to robustly handle datasets, and geospatial hydrologic and biogeochemical models. AGWA is a GIS-based hydrologic and erosion modeling support tool jointly developed by the EPA/ORD, USDA/ARS, and the University of Arizona (Miller et al. 2007; www.tucson.ars.ag.gov/agwa) which includes several models (SWAT, KINEROS, OPUS, RHEM, WEPP) suitable for modeling scales ranging from large watershed (HUC 6-8) to individual hillslope scales. AGWA uses existing spatial datasets in the form of digital elevation models, land cover maps, soil maps, and weather data. These inputs are processed to prepare model parameters. The Kinematic Runoff and Erosion Model (KINEROS2) is an event driven, process based model included in AGWA, which simulates small watersheds (Smith et al. 1995; Semmens et al. 2008; Goodrich et al. 2012).

a) b)

c) d)

92

K2-O2 is a combination of KINEROS2, and Opus2, a continuous soil moisture, plant growth, and nutrient cycling model (Smith 1992; Metherall 1993; Parton, et.al. 1988; Massart et al. 2010), which describes the exchange of nutrients (carbon, nitrogen and phosphorus) between atmosphere, plants, soil and microbes. This model coupling enables a robust simulation of the coupled water, sediment, and biogeochemical processes of a catchment system. KINEROS2 also has an “Urban” element which can represent an abstraction of a typical subdivision. This version of KINEROS2 has been used to simulate an urban environment (Figure 1). In a recently completed project with the USEPA/ORD (Korgaonkar et al. 2014) automated parameterization of KINEROS2 for built environments was built into AGWA. These procedures permit the representation of built environments with and without the capture and infiltration of runoff at the home and neighborhood scales. This capability supports the modeling of different built environment design scenarios to investigate the potential impacts of human design actions on the ecosystem services outcomes of runoff, biogeochemical water quality improvement, and vegetation growth. For large urban systems the Storm Water Management Model (SWMM) will be used to route storm water. The SWMM model will be used to assess the effects of green infrastructure on hydrologic and water quality processes and fluxes in the six study regions. In each region, up to five MS4 areas representing various land use/zoning and urban development patterns will be selected. The SWMM model will be developed and calibrated for each area. The calibrated models will be used to assess changes in water quantity and quality as a result of the implementation of GI under varying climatic conditions. Since developing SWMM models for all MS4 systems in the study cities is not feasible, empirical models and/or neural networks will be developed to explain the variability of the performance of GIs across regions as a function of regional climate, land use/zoning categories, and other geospatial factors. The empirical models will be used to evaluate the effects of GIs on water quantity and quality for the study regions under baseline and future climate, land use and imperviousness scenarios.

Integrated Modeling of Urban Runoff Quantity and Quality

The K2-O2 modeling within the AGWA modeling framework will be utilized in two ways to integrate physical and ecohydologic processes. First, K2-O2 simulations will be evaluated using neighborhood and city scale observations of hydrologic and biogeochemical (water quality) response in the three focus regions. These site specific simulations will be compared to observations of water flux, soil moisture and nitrogen and carbon processing to assess the models representation of the hydrologic and biogeochemical response to existing urban circumstances. In each of the regions these simulations will also be done for available GI implementations. We have received initial feedback that research sites with and without GI are available in Tucson, Baltimore and Philadelphia. These GI implementations are of variable types including rainwater harvesting, retention/detention basin implementations. These simulations will also focus on growing vegetation to enable a tie into the project led out of UC Riverside by Darrel Jeanrette. In particular this collaboration between projects will focus on our end on how much is available to grow trees while the Jeanrette led project will develop approaches to estimating plant growth based on water availability and timing. Where possible we will use vegetation observations of tree growth from LiDAR and site specific investigations to confirm the longer term water fluxes that are inferred from the K2-O2 simulations. We will further use the K2-O2 modeling to develop city scale simulations of hydrology and water quality and use the long-term data collected by the city to assess the model’s capability to represent the system, improve the model and then utilize the model to assess alternative scenarios of GI implementation. Scenario development of GI implementation will focus on individual and collective application of GI practices. The practices included in these scenarios will include large scale retention detention, passive and active rainwater harvesting at the lot and street scale, swales and pervious pavement. Realistic scenarios of GI use will be developed with researchers who are expert in the regions being simulated and in consultation with suitable stakeholders. The focus of these simulations will be to develop simulations that assess changes

93

in runoff, water quality, and water available for vegetation. These simulations will result in response curves that enable individual, non-governmental and governmental stakeholders to determine optimal arrangements for maximization of shade, minimization of runoff and water quality improvement. These response curves may differ as spatial scale increases from neighborhood to city scale inducing tradeoffs between street trees and tree lined washes that no longer receive enough water. List of GI systems that will be potentially included in the analysis are: The indicators that will be characterized and quantified include: 5. Expected Products Our project is specifically focused on developing a set of simulations that will help decision makers understand what their options are in terms of green infrastructure to influence flood risk, water quality and vegetation growth in urban areas. Our focus is to develop nomographs of optimal GI implementation based on climate parameters that will provide a sliding scale of deployment so decision makers understand the implications of GI choices and how they scale with the catchment area of implementation. This work will specifically feed into the development of a tool to be placed into water decision makers toolbox for implementation in the urban environment. In each urban region investigated we expect to work with stakeholders to select optimal solutions for their region that meet the region specific goals that exist. We expect to work with REU students to develop the underlying data sets and to do intercity comparisons of hydrologic response and the form and function of GI across the studied cities. 6. General Project Information Partnerships- Municipal and state government agencies focused on heat island, water quantity and water quality impacts.

Project Timeline

Jan 2016- Dec 2016: Calibrated SWMM models for the representative areas in the 6 study region

Jan 2017-Dec 2017: Generalization of the potential effects of GIs for the 6 cities

Roles and Responsibilities

The CSU team led by Mazdak Arabi will develop the SWMM models for representative areas within the 6 cities and will quantify the sustainability indicators under current and alternative future climate, population and land use conditions. 7. References Belnap, J., J. R. Welter, N. B. Grimm, N. Barger, and J. A. Ludwig (2005), Linkages between microbial and hydrologic processes in arid and semiarid watersheds, Ecology, 86(2), 298-307. Belt, K., S. Kaushal, C. Swan, W. Stack, R. Pouyat, and P. Groffman (2009). An Urban Stream Continuum: gutter subsidies, upland riparian zones and engineered “urban karst”…. and lotic Berardo, R., and J. T. Scholz (2010), Self-Organizing Policy Networks: Risk, Partner Selection, and Cooperation in Estuaries, American Journal of Political Science, 54(3), 632-649, doi:10.1111/j.1540-5907.2010.00451.x. Bernhardt, E. S., Likens, G. E., Hall, R. O., Buso, D. C., Fisher, S. G., Burton, T. M., ... & Lowe, W. H. (2005). Can't see the forest for the stream? In-stream processing and terrestrial nitrogen exports. Bioscience, 55(3), 219-230. Goodrich, D. C., I. S. Burns, C. L. Unkrich, D. J. Semmens, D. P. Guertin, M. Hernandez, S.

94

Yatheendradas, J. R. Kennedy, and L. R. Levick (2012), Kineros2/Agwa: Model Use, Calibration, and Validation, T Asabe, 55(4), 1561-1574. Grimm, N. B., S. H. Faeth, N. E. Golubiewski, C. L. Redman, J. G. Wu, X. M. Bai, and J. M. Briggs (2008), Global change and the ecology of cities, Science, 319(5864), 756-760, doi:10.1126/science.1150195 Grimm, N. B., D. Foster, P. Groffman, J. M. Grove, C. S. Hopkinson, K. J. Nadelhoffer, D. E. Pataki, and D. P. C. Peters (2008), The changing landscape: ecosystem responses to urbanization and pollution across climatic and societal gradients, Front Ecol Environ, 6(5), 264-272, doi:Doi 10.1890/070147. Korgaonkar, Y., I. Burns, D. Guertin, D. Goodrich, C. Unkrich, J. Barlow, and W. Kepner (2014), Representing Green Infrastructure Management Techniques in Arid and Semi-arid Regions: Software Implementation and Demonstration using the AGWA/KINEROS2 Watershed Model, edited, U.S. Environmental Protection Agency, Washington, DC. Massart, J., D. P. Guertin, R. E. Smith, D. C. Goodrich, C. L. Unkrich, and L. R. Levick (2010), K2-O2 (KINEROS2-Opus) Spatially Based Watershed Hydrologic and Biogeochemical Model, paper presented at Proceedings of the Joint 9th Federal Interagency Sedimentation Conference and 4th Federal Hydrologic Modeling Conference (CD-ROM), Las Vegas, NV, June 27-July 1, 2010. Mejia, A. I., and G. E. Moglen (2009), Spatial Patterns of Urban Development from Optimization of Flood Peaks and Imperviousness-Based Measures, Journal of Hydrologic Engineering, 14(4), 416-424. Metherell, A. K. (1993), CENTURY Soil Organic Matter Model Environment: Agroecosystem Version 4.0, USDA-ARS. Miller, S. N., D. J. Semmens, D. C. Goodrich, M. Hernandez, R. C. Miller, W. G. Kepner, and D. P. Guertin (2007), The Automated Geospatial Watershed Assessment tool, Environ Modell Softw, 22(3), 365-377, doi:DOI 10.1016/j.envsoft.2005.12.004. Moglen, G. E. (2009), Hydrology and Impervious Areas, Journal of Hydrologic Engineering, 14(4), 303-304, doi:110.1061/(ASCE)1084-0699(2009)14:4(303). Parton, W. J., J. W. B. Stewart, and C. V. Cole (1988), Dynamics of C, N, P and S in grassland soils: a model, Biogeochemistry, 5(1), 109-131, doi:10.1007/BF02180320. Pataki, D. E., M. M. Carreiro, J. Cherrier, N. E. Grulke, V. Jennings, S. Pincetl, R. V. Pouyat, T. H. Whitlow, and W. C. Zipperer (2011), Coupling biogeochemical cycles in urban environments: ecosystem services, green solutions, and misconceptions, Front Ecol Environ, 9(1), 27-36, doi:10.1890/090220. Reid, K. D., B. P. Wilcox, D. D. Breshears, and L. MacDonald (1999), Runoff and Erosion in a Piñon–Juniper Woodland Influence of Vegetation Patches, Soil Sci. Soc. Am. J., 63(6), 1869-1879, doi:10.2136/sssaj1999.6361869x. Semmens, D., D. Goodrich, C. Unkrich, R. Smith, D. Woolhiser, and S. Miller (2008), 5 KINEROS2 and the AGWA modelling framework, Hydrological modelling in arid and semi-arid areas, 49. Smith, B. K., Smith, J. A., Baeck, M. L., & Miller, A. J. (2015). Exploring storage and runoff generation processes for urban flooding through a physically based watershed model. Water Resources Research, 51(3), 1552-1569.

95

Smith, R. (1992), Opus: an integrated simulation model for transport of nonpoint-source pollutants at the field scale: volume I. Documentation, ARS-US Department of Agriculture. Smith, R. E., D. C. Goodrich, and J. N. Quinton (1995), Dynamic, Distributed Simulation of Watershed Erosion - the Kineros2 and Eurosem Models, J Soil Water Conserv, 50(5), 517-520. Tsihrintzis, V. A., and R. Hamid (1997), Modeling and Management of Urban Stormwater Runoff Quality: A Review, Water Resour Manag, 11(2), 137-164. Urgeghe, A. M., D. D. Breshears, S. N. Martens, and P. C. Beeson (2010), Redistribution of Runoff Among Vegetation Patch Types: On Ecohydrological Optimality of Herbaceous Capture of Run-On, Rangeland Ecol Manag, 63(5), 497-504, doi:Doi 10.2111/Rem-D-09-00185.1. Wilcox, B. P., D. D. Breshears, and C. D. Allen (2003), Ecohydrology of a resource-conserving semiarid woodland: Effects of scale and disturbance, Ecological Monographs, 73(2), 223-239.

96

Project B2-2(a) Research Plan

Project Title Flood Hydrology and Rainfall Frequency

UWIN Project number

B2-2(a)

Project Lead

Jim Smith/Princeton

Investigators/Institutions

Tom Meixner/Arizona (not funded), Brian Bledsoe/CSU (not funded), Andy Miller/UMBC (not funded), Darrel Jenerette/ UC Riverside (not funded)

Project Period

September 2015 – August 2020

Project Cost

Approximately $380,000 – Princeton

Graduate students funded to work on project

TBA (Visiting graduate student Joyce Zhou will work on this project for the first two years).

Project Overview

In this project, we will examine the hydrology, hydrometeorology and hydroclimatology of urban flooding through numerical modeling and data-driven studies that focus on urban watersheds in Baltimore, Miami, Denver, Phoenix (and the broader Sun Corridor region), Los Angeles and Portland. The broad objective of this study is to develop a predictive understanding of urban flood hydrology that can be used to assess the effectiveness of “urban water solutions”, especially Green Infrastructure technologies, in reducing flood hazards. We will develop long-term, high-resolution radar rainfall fields for each of the six urban regions for hydrologic modeling studies and for examining the hydroclimatology of urban flash flooding. In previous studies (see, for example, Figure 1), it has been shown that high-resolution radar rainfall fields provide an exceptional resource for assessing urban flooding. The six urban study regions provide a broad range of settings for examining urban flooding, both in terms of land surface process and in terms of rainfall climatology. We will develop and implement hydrologic models for each of the study regions. Modeling analyses will be used to address core scientific questions and assess the effectiveness of urban water solutions in reducing flood hazards.

Project Summary

The objectives of the study are: 1) to demonstrate a predictive understanding of urban flood hydrology through implementation and validation of hydrologic models for urban watersheds in Baltimore, Denver, Miami, Phoenix (and the broader Sun Corridor region), Los Angeles and Portland, 2) to characterize the climatology of flood-producing storm systems in the 6 urban regions and 3) to develop and implement procedures for rainfall and flood frequency analysis that can serve as the foundation for assessing urban flood hazards. The experimental approach will center on: 1) development of high-resolution radar rainfall fields for the six urban regions, 2) development and implementation of hydrologic models for gaged watersheds in the six urban regions, and 3) hydroclimatological studies of urban flooding in the six urban regions. The expected outcomes include: 1) high-resolution radar rainfall fields for the six study regions that can serve as the basis for hydrologic modeling studies and hydroclimatological studies of rainfall and flood frequency, 2) hydrologic models that can be used for examining the consequences of urban water solutions (especially those based on Green Infrastructure technologies) and climate change and 3) advances in fundamental understanding of urban flood hydrology.

97

Supplemental Keywords

Hydroclimatology, urban flooding, urban modification of rainfall.

Project description

1. Objectives The objectives of the study are: 1) to demonstrate a predictive understanding of urban flood hydrology through implementation and validation of hydrologic models for urban watersheds in Baltimore, Denver, Miami, Phoenix (and the broader Sun Corridor region), Los Angeles and Portland, 2) to characterize the climatology of flood-producing storm systems in the 6 urban regions and 3) to develop and implement procedures for rainfall and flood frequency analysis that can served as the foundation for assessing urban flood hazards. This project contributes to the broader objectives of the urban flood hazard projects. A critical objective for these projects will be development and analysis of a flood hazard index that will be used to characterize current (baseline) conditions and that will provide the basis for assessing the effectiveness of “urban water solutions”, especially those based on Green Infrastructure technologies. The flood hazard index needs to reflect the spatial pattern of flood hazards over a drainage network of an urban area, the variation of flood hazards from drainage basin to drainage basin within an urban area and the variation in flood hazards across a diverse collection of urban settings (our 6 study regions). The flood hazard index will be developed through a sequence of modeling and analysis steps, that include hydrologic models, hydraulic models, floodplain mapping and damage assessment. In this project, we will contribute to this broad objective through development of hydrologic models and procedures for characterizing the hydroclimatology of flooding in the 6 urban study regions. 2. Intellectual Merit The hydrology, hydrometeorology and hydroclimatology of urban flooding are still poorly understood. Previous research has provided important insights, but a broad, generalizable understanding of urban flooding is not currently at hand. In this project, we will build on prior research in examining hydrologic and hydrometeorological processes that control urban flooding. We will also place these analyses in a hydroclimatological context that facilitates assessment of climate change impacts on urban flood hazards. As noted above, a key element of the project will be development of a predictive understanding of urban flooding that enables analyses of the impacts of “urban water solutions” on flood hazards. By examining these questions across the range of settings in the six urban regions, we will provide broadly generalizable conclusions. 3. Approach/Activities For this project, we will develop high-resolution “bias-corrected” radar rainfall fields for hydrologic modeling analyses and rainfall frequency analyses (as well as for urban rainfall climatology analyses in the project lead by Elie Bou-Zeid). The spatial resolution for rainfall fields is approximately 1 km and the time resolution is 15 minutes. The procedures use the “Hydro-NEXRAD” algorithms; bias correction will use observations from urban rain gage networks. Procedures for Baltimore have been developed under the previous NSF-WSC project (Smith et al. 2012; see also Wright et al. 2014b and Cunha et al. 2013) and rainfall products have been completed through September 2014 for Baltimore. There are excellent rain gage networks for Phoenix (Maricopa County Flood Control District), the Front Range of Colorado (Urban Drainage and Flood Control District), Miami (South Florida Water Management District), Los Angeles (County of Los Angeles Department of Public Works) and Portland (City of Portland Hydra rain gage network). An early task in this project is developing urban rain gage data sets for each of the six urban

98

regions. Development of rain gage data sets will be a major focus of the first 6 months of the project. “Storm catalogs” of major flood-producing storm events will be developed for each of the 6 urban regions (see, for example, Wright et al. 2014 a and b, Smith et al. 2012, Yang et al. 2013 and Yeung et al. 2015). The period of record for each urban region will be approximately 1995 – 2016 (before 2002, there are gaps in the radar record, so we will miss some events early in the period of record). For Baltimore we will also compile “continuous” 15-minute, 1 km rainfall fields for the period 2000 – 2016 and for the months April – September (2000 – 2014 are currently complete). For Denver, we will develop “continuous” rainfall fields for the period 2000 – 2016 and for the months May – September (2000 – 2014 will be complete in October). For Miami we will also develop “continuous” rainfall fields for the period 2000 – 2016 for all 12 months (complete by the end of the 2015). For Phoenix, Los Angeles and Portland, the current thinking is to just develop the storm catalog data sets. We will update radar rainfall data sets in 2016 to include 2015 data and in 2017 to include 2016 data. We will characterize the hydroclimatology of urban flooding in each of the 6 regions through analyses of USGS stream gaging observations (as in B. K. Smith et al. 2013, Smith and Smith 2015, Yang et al. 2013, Villarini et al. 2001 and Villarini et al. 2010). We will examine the mixtures of flood-generating mechanisms for each of the study regions. Ancillary data sets that will be used to assess flood climatology include cloud-go-ground lightning data from the National Lightning Detection Network (NLDN; see Ntelekos et al. 2007 and Smith and Smith 2015), HURDAT tropical cyclone track data from the National Hurricane Center (Smith and Smith 2015 and Wright et al. 2014), and reanalysis fields (Li et al. 2013 and Yang et al. 2014). Through these analyses, we will characterize the mixture of flood generating mechanisms in each of the study regions. We will carry out numerical experiments using the Weather Research and Forecasting (WRF) model (along the lines of Ryu et al. 2015, Li et al. 2013, Yang et al. 2014, and Ntelekos et al. 2008) to examine the role of urbanization in altering the climatology of flood-producing storm systems. Stochastic storm transposition using storm catalogs will be applied to examine rainfall frequency, and combined with hydrologic models, to examine flood frequency (as described in Wright et al. 2013 and 2014). A major challenge in implementing stochastic storm transposition procedures for the six study regions will be developing procedures for addressing spatial heterogeneity of rainfall associated with complex terrain (directions for attacking this problem are outlined in Wright et al. 2013). Natural sources of spatial heterogeneity, in particular mountainous terrain and land – water boundaries, play an important role in the rainfall climatology of each of the study regions. Mountainous terrain will play a central role in hydroclimatological studies for Denver, Phoenix and Los Angeles. Land water boundaries will be of particular importance for Miami and Baltimore. Hydrologic and hydraulic modeling will play a central role in quantifying baseline flood hazards and evaluating the effectiveness of urban water solutions in reducing flood hazards. The hydrologic modeling platform that we will use is the Gridded Surface Subsurface Hydrologic Analysis (GSSHA) model developed at CSU and supported by COE. We will also work with the Kineros – Opus 2 modeling system, in collaboration with the U. Arizona group. We will build on modeling analyses carried out by Brianne Smith for urban watersheds in Baltimore (B. K. Smith et al. 2015). In particular we will extend the urban modeling capabilities to represent elements of urban water solutions. A key step in hydrologic modeling is mapping “urban water solutions” to model components and model parameters. We will coordinate closely with Tom Meixner and Phil Guertin on their “Comparative Impact of Green Infrastructure Impact” project in which a range of GI solutions are developed and linked to urban hydrology and environmental quality. We will also link closely with Brian Bledsoe and Andy Miller on their “Hydrology and Hydraulics of Floodplains” project in which hydraulic models (2-D HEC-RAS) will be developed for floodplain mapping. Hydraulic modeling analyses will be coupled with hydrologic modeling and rainfall/flood frequency analysis,

99

leading to core floodplain mapping procedures. Characterization of uncertainties in floodplain mapping will be a major challenge for this portion of the project and we will contribute to the hydraulic modeling team in assessing sources of uncertainty. Gaged watersheds will play an important role in assessing our ability to model baseline conditions and to support our analyses that assess changing hydrologic response resulting from urban water solutions. In Baltimore there is a collection of 1 - 20 km2 watersheds (see B. K. Smith et al. 2013) that are natural candidates for modeling analyses. Gwynns Falls provides a Baltimore watershed with nested stream gaging at basin scales ranging from 1 – 200 km2, In the Front Range of Colorado, Denver has a collection of “small” watersheds (Lakewood Gulch, Harvard Gulch, Dry Gulch, First Creek) that are good targets for modeling studies. We will also collaborate with Brian Bledsoe in analyses of larger Front Range watersheds (Cache La Poudre, Big Thompson, St. Vrain, Clear Creek and South Platte). The Maricopa County Flood Control District gaging program provides an exceptional stream gaging network, building on long-term collaborations with USGS. Cave Creek and Skunk Creek are two watersheds in Phoenix with excellent nested stream gaging networks. Miami will be an interesting challenge for assessing urban flooding. On the one hand, there is a significant urban flood problem in the region. There is not, however a very useful stream gaging system to guide our analyses in a very complicated system. We will work closely with our collaborators in the project to determine the appropriate watersheds to use as principal study sites for hydrologic and hydraulic modeling studies.

Data Needs

Hydrologic and hydraulic modeling will rely heavily on land use – land cover data and urban infrastructure data. There are special observational resources for each region, which we will want to utilize. It will also be useful to “standardize” across watersheds and urban areas to the extent possible. We will work with the CSU group in developing land surface data sets for modeling studies. 4. Expected Results, Benefits, Outputs, and Outcome The expected outcomes include: 1) high-resolution radar rainfall fields for the six study regions that can serve as the basis for hydrologic modeling studies and hydroclimatological studies of rainfall and flood frequency, 2) hydrologic models that can be used for examining the consequences of urban water solutions and climate change and 3) advances in fundamental understanding of urban flood hydrology We look forward to working with REU students. 5. General Project Information We will develop close interactions with other projects. Facilities at Princeton are adequate for proposed activities. Analyses are underway. 6. References Ryu, Y.-H., J. A. Smth, M. L. Baeck and E. Bou-Zeid, The influence of land-surface heterogeneities on heavy convective rainfall in the Baltimore-Washington metropolitan area, Monthly Weather Review, in review, 2015. Smith, B. K. and J. A. Smith, The flashiest watersheds in the contiguous United States, J. of Hydrometeorology, to appear, 2015. Yeung, J. K., J. A. Smith, M. L. Baeck and G. Villarini, Lagrangian analyses of rainfall structure and evolution for organized thunderstorms in the urban corridor of the northeastern US, J. of Hydrometeorology, 16(4), 1575 - 1595, 2015. Smith, B. K., J. A. Smith, M. L. Baeck and A. J. Miller, Exploring storage and runoff generation processes

100

for urban flooding through a physically-based hydrologic model, Water Resources Research, 51, 1552–1569, doi:10.1002/2014WR016085, 2015. Wright, D. B., J. A. Smith, and M. L. Baeck, Flood frequency analyses using radar rainfall fields and stochastic storm transposition, Water Resources Research, 50, 1592 - 1615, 2014a. Wright, D. B., J. A. Smith, G. Villarini, and M. L. Baeck, Long-term, high-resolution, radar rainfall records for urban hydrology, J. of the American Water Resources Associations, 50(3), 713-734, 2014b. Yang, L., J. A. Smith, M. L. Baeck, E. Bou-Zeid, S. M. Jessup, F. Tian, and H. Hu,, Impact of Urbanization on Heavy Convective Precipitation under Strong Large-Scale Forcing: A Case Study over the Milwaukee-Lake Michigan Region, J. of Hydrometeorology, 15, 261-278, 2014. Cunha, L. K., J. A. Smith, M. L. Baeck and W. F. Krajewski, An Early Performance Evaluation of the NEXRAD Dual Polarization Radar Rainfall Estimates for Urban Flood Applications, Weather and Forecasting, 28, 1478–1497, 2013. Yang, L., J.A. Smith, D.B. Wright, M.L. Baeck, G. Villarini, F. Tian, and H. Hu, Urbanization and climate change: An examination of nonstationarities in flood frequency, Journal of Hydrometeorology, 14(6), 2013, doi: JHM-D-12-095.1 , 2013. Li, D., E. Bou-Zed, M. L. Baeck, S. M. Jessup, and J. A. Smith, Modeling Land Surface Processes and Heavy Rainfall in Urban Environments: Sensitivity to Urban Surface Representations, J. of Hydrometeorology, 14, 1098 - 1118, 2013. Smith, B. K. J. A. Smith, M. L. Baeck, G. Villarini, and D. B. Wright, The spectrum of storm event hydrologic response in urban watersheds, Water Resources Research, 49(5), 2649-2663, 2013. Wright, D. B., J. A. Smith, G. Villarini, and M. L. Baeck, Estimating the frequency of extreme rainfall using weather radar and stochastic storm transposition, J. of Hydrology, 488, 150-165, 2013. Smith, J. A., M. L.Baeck, G. Villarini, C. Welty, A. J. Miller, W. F. Krajewski, Analyses of a long-term high-resolution radar rainfall data set for the Baltimore metropolitan region, Water Resources Research, 48, W04504, doi:10.1029/2011WR010641, 2012. Villarini, Gabriele, James A. Smith, Mary Lynn Baeck, Paula Sturdevant-Rees, Witold F. Krajewski, Radar Analyses of Space-Time Variability of Extreme Flood-Producing Rainfall in Urban Drainage Basins, J. of Hydrology, 381(3-4), 266-286, 2010. Villarini, Gabriele, James A. Smith, Francesco Serinaldi, Jerad Bales, Paul D. Bates, and Witold F. Krajewski, Flood Frequency Analysis for Nonstationary Annual Peak Records in an Urban Drainage Basin, Advances in Water Resources, 32(8), 1255-1266, 2009. Ntelekos, A. A., J. A. Smith, M. L. Baeck, W. F. Krajewski, A. J. Miller, and R. Goska, Extreme hydrometeorological events and the urban environment: Dissecting the 7 July 2004 thunderstorm over the Baltimore MD Metropolitan Region, Water Resour. Res., 44, W08446, doi:10.1029/2007WR006346, 2008. Ntelekos, A. A., J. A. Smith, and W. F. Krajewski, 2007, Climatological Analyses of Thunderstorms and Flash Floods in the Baltimore Metropolitan Region, Journal of Hydrometeorology, 8(1), 88-101.

101

Figure B2-2(a)-1. Mean daily rainfall (mm) for the 25 largest rain days (2000 – 2010) in Baltimore City (gray

line). Background map illustrates topography of the study region.

102

Project B2-2(b) Research Plan Project Title Hydrology and hydraulics of urban floodplains

UWIN Project number

B2-2(b)

Project Lead

Brian Bledsoe/CSU

Investigators/Institutions

Andy Miller/University of Maryland, Baltimore County (funded), Jim Smith/Princeton (not funded), Tom Meixner/Arizona (not funded), Darrel Jenerette/ UC Riverside (not funded)

Project Period

January 2016-December 2018

Project Cost

Approximately $195,000 – CSU Approximately $259,000 - UMBC

Graduate students funded to work on project

Tim Stephens supervised by Brian Bledsoe TBA to be supervised by Andy Miller

Project Overview

Novel techniques for analyzing and designing urban drainage and floodplain systems for increased resilience to extreme events can ultimately provide an expanded footprint and palette for design of floodplain-greenspace networks that improve water quality, increase biodiversity, moderate temperatures, cleanse air, and enhance human well-being. This project addresses the interactions between flood flows and urban channels, floodplains and riparian zones as influenced by urban infrastructure and efforts to mitigate impacts of urban development (green infrastructure (GI) and low impact development (LID)) on flood response and other environmental consequences in six locations: Colorado Front Range, Baltimore, Phoenix, Los Angeles, S. Florida, and Pacific Northwest (Portland or Puget Lowlands). This effort is fundamentally linked to project B2-2a which provides a suite of hydrologic modeling scenarios that will be used to drive floodplain hydraulic models in the six study regions. Floodplain mapping assisted by modeling of flood hydraulics will represent baseline, development, and various conservation scenarios derived from Envision to examine how integrated low impact development and GI, LID and sustainable urban drainage systems (SUDS) can be strategically designed and positioned to simultaneously enhance flood resilience while providing multiple co-benefits. The resulting floodplain inundation maps will be probabilistic and reflect uncertainty quantified through Monte Carlo analysis of model inputs and parameters, competing methods of linking precipitation and land use scenarios to generate flood frequency distributions, and innovative techniques for portraying channel morphodynamics and erosion hazards.

103

Figure 2-2(b)-1. Explicitly addressing uncertainty in floodplain inundation during extreme events can potentially expand the footprint and palette for design of urban floodplain-greenspace networks that provide many co-benefits. In addition to developing procedures for design of urban drainage and channel/floodplain corridors to increase resilience to extreme flooding, the resulting floodplain maps will provide a platform for examining the efficacy of expanded GI and naturalized floodplains for water quality and biodiversity (project A3-2), moderating temperatures (project A2-3) and improving human well-being.

Project Summary

This project addresses the interactions between flood flows and urban channels, floodplains and riparian zones as influenced by urban infrastructure and efforts to mitigate impacts of urban development on flood response and other environmental consequences. Flood characteristics to be studied include propagation of flood waves, inundation depth and areal extent, velocity and shear stress distributions, hydroperiod, and associated risks to life, property and infrastructure. Project B2-1 will provide hydrologic inputs in the form of competing discharge-frequency relationships for each scenario that will be used to force the hydraulic models. Objectives: Our objectives are to:

Use hydrologic modeling outputs generated in project B2-2a to map urban floodplains under uncertainty in the six study regions.

Compare and contrast the flood resiliency benefits of GI, LID and SUDS among the six study regions

Provide floodplain mapping scenarios that will allow the U-WIN team to explore how novel techniques for designing urban drainage and floodplain systems for increased resilience to extreme events can ultimately provide an expanded footprint and palette for design of floodplain-greenspace networks that also increase biodiversity, moderate temperatures, cleanse air, and enhance health and happiness.

Questions:

104

To what extent are solutions that may be effective in one region translatable to a different hydroclimatic/physiographic/social setting with different urban structure, population density or development pattern?

How does the efficacy of floodplain vs. upland GI for mitigating flood risks vary among regions?

In what ways does sensitivity of flood risks to prospective solutions vary across regions

How does the magnitude of the anticipated urbanization signal on streamflows and geomorphic processes compare with the anticipated climate change signal?

How or to what extent will interactions between climate, land use, physiography and GI/LID produce threshold behavior in flood hazards and risks?

Are there thresholds in fractional spatial coverage of GI/LID that are required in order to produce measurable changes in flood magnitude, inundation extent or risk to property and infrastructure?

Experimental approach: This project is fundamentally linked to project B2-2a which will provide a suite of hydrologic modeling scenarios that will be used to force HEC-RAS floodplain models in the six study regions (See below for a discussion of each study region.) The resulting floodplain inundation maps will be probabilistic and reflect uncertainty quantified through Monte Carlo analysis of model inputs and parameters, competing methods of linking precipitation and land use scenarios to generate flood frequency distributions, and through innovative techniques for portraying channel morphodynamics and erosion hazards. The general steps to be completed in carrying out this research are as follows.

1. Provide input on generation of Envision scenarios representing baseline, “development,” and various “conservation” conditions in the six study regions.

2. Complete data inventory and finalize selection of study watersheds for hydrologic and hydraulic (H&H) modeling in the six regions based on availability of precipitation data, high quality topographic surveys, gaged streamflow / stage data, information on regional performance of GI/LID solutions, and existing H&H models.

3. Develop more detailed representations of land use / land cover within the Envision scenarios by including plausible combinations and spatial configurations of GI/LID practices for study watersheds in each region. These scenarios will include upland only, floodplain only, and upland + floodplain GI scenarios.

4. Collaborate with D. Jenerette and others to translate modeled combinations of floodplain and upland GI into vegetation scenarios that facilitate modeling of GI influence on urban water cycle and temperature regimes, as well as water quality and aquatic / riparian biodiversity through pollutant removal, shading, and organic matter inputs. Feedbacks between vegetation dynamics and channel evolution / floodplain conveyance will also be addressed.

5. Perform hydrologic modeling and gage analysis to generate alternative flood frequency distributions for study watersheds in the six hydrologic regions (project B2-2a).

6. Perform HEC-RAS 5.0 modeling of combined 2-d and 1-d floodplain hydraulics subject to data availability and appropriate boundary conditions. Incremental flows spanning the range of 50-200 year return periods across all scenarios in each region will be modeled.

7. Produce probabilistic maps of floodplain inundation associated with Q2-Q200 return periods for each scenario by performing uncertainty analysis. Manning’s n and geometry inputs will be varied in Monte Carlo simulations that are informed by plausible scenarios of stream restoration, channel adjustment, vegetation encroachment and dynamics, and large wood / debris inputs.

8. Compare and contrast the efficacy of stormwater and floodplain management solutions within and among regions and identify thresholds of response. Some watersheds will contain only small tributaries that will be responsive to GI/LID solutions. Other watersheds will also contain relatively large main stem rivers that are largely unaffected by upland GI/LID solutions in the urban area. Note that this step may be iterative in exploring upland only, floodplain only, and upland + floodplain GI scenarios.

9. Use the resulting floodplain maps to compare to existing 100-yr floodplain maps and examine potential costs

105

and damages associated with extreme hydrometeorological events under the various development and conservation scenarios.

10. Provide flow frequency / floodplain footprint scenarios to other projects. 11. Collaborate with D. Jenerette, E. Bou-Zeid, C. Welty, N.L. Poff, N. Grimm et al. in support of modeling

the influence of stormwater and floodplain solutions on the urban water cycle and temperature regimes, as well as on water quality and aquatic / riparian biodiversity through pollutant removal, shading, and organic matter inputs.

12. Collaborate with project economists and social scientists to examine variable responses to flood scenarios within and among regions to assist with development of Linked Systems and Institution / Equity indicators.

13. Work collaboratively across projects to link the performance of the business as usual, development, and various scenarios of conservation (GI / LID) scenarios with indicators including: Frequency and Duration of Extreme Flows, Nutrients, Connectivity, % Natural land cover, % Artificial drainage area, Road crossing density, Biodiversity, Habitat, Bed/bank stability, and Riparian Losses.

14. Perform literature review and use modeling results to link solutions (GI, LID, SUDS, stream restoration) with novel indicators of flood resiliency, water quality, and aquatic biodiversity.

15. Engage and educate stakeholders through regional workshops and materials prepared for dissemination through the Urban Water Sustainability Hub.

Expected results: We expect to improve our current understanding of magnitude and frequency of flood peak discharges across the diverse range of watersheds within our study domain. We will identify regional thresholds and nonlinearities, and assess the sensitivity of peak discharge to upstream as well as local mitigation efforts within the channel and riparian zone. In addition we anticipate being able to produce metrics for assessing multiple dimensions of flood risk and to compare the relative sensitivity of these metrics to future climate scenarios, future development scenarios, and alternative mitigation scenarios. Finally, this effort will provide insight on how novel techniques for designing urban drainage and floodplain systems for increased resilience to extreme events can ultimately provide an expanded footprint and palette for design of floodplain-greenspace networks that simultaneously provide flood resilience, increase biodiversity, moderate temperatures, cleanse air, and enhance health and happiness.

Supplemental Keywords

Urban flooding, green infrastructure, flood risk, flood mitigation, extreme events, ecological engineering

Project description

1. Objectives and Hypotheses This project addresses the interactions between flood flows and urban channels, floodplains and riparian zones as influenced by urban infrastructure and efforts to mitigate impacts of urban development on flood response and other environmental consequences. Flood characteristics to be studied include propagation of flood waves, inundation depth and areal extent, velocity and shear stress distributions, hydroperiod, and associated risks to life, property and infrastructure. Project B2-2a will provide hydrologic inputs in the form of competing discharge-frequency relationships for each scenario that will be used to force the hydraulic models. Our objectives are to:

Use hydrologic modeling outputs generated in project B2-2a to map urban floodplains under uncertainty in the six study regions;

Compare and contrast the flood resiliency benefits of GI and LID among the six study regions;

Advance fundamental understanding of urban floodplain hydraulics and methods for characterizing uncertainty in floodplain inundation mapping; and

106

Provide floodplain mapping scenarios that will allow our collaborators to explore how novel techniques for designing urban drainage and floodplain systems for increased resilience to extreme events can ultimately provide an expanded footprint and palette for design of floodplain-greenspace networks that simultaneously provide flood resilience, increase biodiversity, moderate temperatures, cleanse air, and enhance health and happiness.

Questions:

To what extent are solutions that may be effective in one region translatable to a different hydroclimatic/physiographic/social setting with different urban structure, population density or development pattern?

How does the efficacy of floodplain vs. upland GI for mitigating flood risks vary among regions?

In what ways does sensitivity of flood risks to prospective solutions vary across regions

How does the magnitude of the anticipated urbanization signal on streamflows and geomorphic processes compare with the anticipated climate change signal?

How or to what extent will interactions between climate, land use, physiography and GI/LID produce threshold behavior in flood hazards and risks?

Are there thresholds in fractional spatial coverage of GI/LID that are required in order to produce measurable changes in flood magnitude, inundation extent or risk to property and infrastructure?

Related questions that highlight potential intersections with other projects:

To what extent can well-connected stream/riparian corridor networks identified through probabilistic floodplain mapping improve resiliency against extreme precipitation events and ameliorate UHI effects?

What combinations of naturalized floodplains, white / green roofs and/or LID can synergistically moderate temperature extremes while increasing aquatic and riparian biodiversity? What is the range of sensitivity of indicator metrics to the range of feasible combinations?

How do interactions between stream corridor network topology and the configuration of built-up areas (e.g., urban canyons) influence UHI effects?

Cities can be richer in upland plant species, including in native species, than rural areas. Can this also be true for riparian / aquatic systems in naturalized floodplains.

How do interactions among floodplain corridor networks, UHI effects, and measures of human well-being vary by region?

What are key incentives for wider adoption of green infrastructure solutions?

What are key institutional agreements and socioeconomic factors that impede adoption of these solutions?

How would sustainable floodplain management influence urban land development patterns? 2. Intellectual merit Floodplain management programs in the US are widely regarded as being “broken.” This project will advance fundamental understanding of urban floodplain hydraulics and methods for characterizing uncertainty in floodplain inundation mapping while challenging the antiquated, deterministic approaches that are entrenched in US floodplain management programs. The investigators in project B2-2 have all previously done flood studies. A key contribution of this effort is conducting the cross-site comparisons given differences in climate and hydrometeorology, geology and geomorphic template, urban form and infrastructure, but with the goal of assessing the relative effectiveness of commonly proposed solutions across different regions. We can ask, for example, whether proposed solutions are effective at all, somewhat effective, or very effective, and whether that judgment will vary either locally within a region or across regions. Figuring out what metrics to use will be an important part of the work.

107

The overarching goal of this project is to use hydraulic analysis of urban floodplains to examine how integrated floodplain networks and sustainable urban drainage systems can be strategically designed and positioned to simultaneously enhance flood resilience, moderate temperatures and improve human comfort, and support biodiversity. 3. Approaches / Activities This project is fundamentally linked to project B2-2a which will provide a suite of hydrologic modeling scenarios that will be used to force HEC-RAS floodplain models in the six study regions (See below for a discussion of each study region.) The resulting floodplain inundation maps will be probabilistic and reflect uncertainty quantified through Monte Carlo analysis of model inputs and parameters, competing methods of linking precipitation and land use scenarios to generate flood frequency distributions, and through innovative techniques for portraying channel morphodynamics and erosion hazards. The general steps to be completed in carrying out this research are as follows.

1. Provide input on generation of Envision scenarios representing baseline, “development,” and various “conservation” conditions in the six study regions.

2. Complete data inventory and finalize selection of study watersheds for hydrologic and hydraulic (H&H) modeling in the six regions based on availability of precipitation data, high quality topographic surveys, gaged streamflow / stage data, information on regional performance of GI/LID solutions, and existing H&H models.

3. Develop more detailed representations of land use / land cover within the Envision scenarios by including plausible combinations and spatial configurations of GI/LID practices for study watersheds in each region. These scenarios will include upland only, floodplain only, and upland + floodplain GI scenarios.

4. Collaborate with D. Jenerette and others to translate modeled combinations of floodplain and upland GI into vegetation scenarios that facilitate modeling of GI influence on urban water cycle and temperature regimes, as well as water quality and aquatic / riparian biodiversity through pollutant removal, shading, and organic matter inputs.

5. Perform hydrologic modeling and gage analysis to generate alternative flood frequency distributions for study watersheds in the six hydrologic regions (project B2-2a).

6. Perform HEC-RAS 5.0 modeling of combined 2-d and 1-d floodplain hydraulics subject to data availability and appropriate boundary conditions. Incremental flows spanning the range of 50-200 year return periods across all scenarios in each region will be modeled.

7. Produce probabilistic maps of floodplain inundation associated with Q2-Q200 return periods for each scenario by performing uncertainty analysis. Manning’s n and geometry inputs will be varied in Monte Carlo simulations that are informed by plausible scenarios of stream restoration, channel adjustment, vegetation encroachment, and large wood / debris inputs.

8. Compare and contrast the efficacy of stormwater and floodplain management solutions within and among regions and identify thresholds of response. Some watersheds will contain only small tributaries that will be responsive to GI/LID solutions. Other watersheds will also contain relatively large main stem rivers that are largely unaffected by upland GI/LID solutions in the urban area. Note that this step may be iterative in exploring upland only, floodplain only, and upland + floodplain GI scenarios.

9. Use the resulting floodplain maps to compare to existing 100-yr floodplain maps and examine potential costs and damages associated with extreme hydrometeorological events under the various development and conservation scenarios.

10. Provide flow frequency / floodplain footprint scenarios to other projects. 11. Collaborate with D. Jenerette, E. Bou-Zeid, C. Welty, N.L. Poff, N. Grimm et al. in support of modeling

the influence of stormwater and floodplain solutions on the urban water cycle and temperature regimes, as well as on water quality and aquatic / riparian biodiversity through pollutant removal, shading, and organic matter inputs.

12. Collaborate with project economists and social scientists to examine variable responses to flood scenarios

108

within and among regions to assist with development of Linked Systems and Institution / Equity indicators. 13. Work collaboratively across projects to link the performance of the business as usual, development, and

various scenarios of conservation (GI / LID) scenarios with indicators including: Frequency and Duration of Extreme Flows, Nutrients, Connectivity, % Natural land cover, % Artificial drainage area, Road crossing density, Biodiversity, Habitat, Bed/bank stability, and Riparian Losses.

14. Perform literature review and use modeling results to link solutions (GI, LID, SUDS, stream restoration) with novel indicators of flood resiliency, water quality, and aquatic biodiversity.

15. Engage and educate stakeholders through regional workshops and materials prepared for dissemination through the Urban Water Sustainability Hub. We will establish and maintain communication the NOAA Advanced Hydrologic Prediction Service and seek to leverage a connection with their work on innovative flood inundation mapping techniques.

For analysis of smaller urban streams and riparian zones we will use topographic data sets derived from airborne LiDAR supplemented with terrestrial laser scans, total-station surveys and Structure from Motion (SfM) using digital images from ground-based cameras and, where feasible, unmanned aerial vehicles to develop high-resolution digital elevation models suitable for input to 2-dimensional depth-averaged hydraulic models. Field observations, aerial photographs and mapped land-cover patterns will be used to derive the spatial pattern of roughness characteristics in the various modeling domains. Hydrologic analyses provided from project B2-2a (Flood Hydrology and Rainfall Frequency) will be used as input to drive hydraulic models at specified nodes in the drainage network and scenarios will be chosen to represent a range of hydrometeorological conditions and recurrence intervals. Actual or simulated hydrographs will be routed through hydraulic model domains for baseline, development, and conservation scenarios from Envision that represent a common set of mitigation options in order to assess the sensitivity of flood response to mitigation efforts. The magnitude of change in flood response can be compared across watersheds within the region with different initial conditions in order to assess whether the modeled changes are of comparable magnitude to existing differences among watersheds. The range of alternative scenarios to be examined include scenarios implemented in the upstream watershed and therefore this part of the experimental design will require close collaboration with the team producing analyses of flood hydrology and rainfall frequency. Cross-site comparisons will be conducted in addition to within-site comparisons. In examining problems associated both with new development and with retrofit of existing urban landscapes, we will accommodate spatial heterogeneity of flooding associated with urban modification of extreme rainfall and time trends in flood hazards. Design protocols will be based on characterizations of both flood frequency of extreme events and their recovery times. The model under consideration for use by both Bledsoe and Miller is HECRAS 5.0, which is a 2-dimensional finite-volume code that solves the shallow-water equations using either a diffusion-wave option or the full St. Venant equations with eddy viscosity coefficients utilized in the turbulence closure. Alternatives include TUFLOW, which is a proprietary finite-difference depth-averaged 2D model that solves the shallow-water equations (Miller has a previous license which can be updated at relatively low cost) and TELEMAC-2D, which is now open-source; both have been used at UMBC before. All are capable of handling complex flow domains, wetting and drying, and subcritical-supercritical flow transitions where needed. There are several other well-regarded 2D hydraulic models that are also publicly available. All of them tend to produce comparable results and therefore the choice of model depends on a combination of features and cost; these comparisons are in progress and a final determination will be made before funds are invested in implementation and training. Datasets including topography and roughness characteristics derived from sources described above; precipitation records, stream-gage records, depth-duration-frequency and flood-frequency analyses carried out for multiple sites in order to assess and compare magnitude-frequency relationships; hydrologic inputs either using selected hydrographs from actual storms (in some cases transposed from one watershed to another) or from hydrologic model output generated using input precipitation fields, including alternatives based on the Stochastic Storm Transposition method. The primary output data include time series of flow depth and x- and y-components of velocity; derived data sets include spatial and temporal patterns of shear stress, Froude number, inundation extent, and measures of flood-wave

109

attenuation vs translation along a study reach; comparative analyses of hydrograph response to alternative scenarios based on the same input storm (or using different input hydrographs based on upstream patterns of implementation of green infrastructure); and derived metrics of flood risk based on a combination of magnitude-frequency analysis, inundation and depth mapping, velocity and shear-stress mapping, hydroperiod analysis, and others as may be identified by the group. We anticipate that Project B2-2a will use the Envision output scenarios, run across the six urban study regions, and representing landscape dynamics at annual time steps for the years 2015-2100, to inform future hydrologic scenarios for each region. A preliminary summary of potential land use / land cover and GI / stormwater innovation scenarios to be considered for hydrologic modeling is provided below:

Baseline / Business as Usual

Developed

Conservation o New development only

Upland GI

Floodplain GI

Upland + Floodplain GI o New development + retrofit

Upland GI

Floodplain GI

Upland + Floodplain GI In addition, the project team will need to develop a set of climate / precipitation scenarios that will be used to drive the hydrologic models. Existing stream flow gage data could also be used for flood frequency analysis where adequate records are available. For example, the range of scenarios could include:

Gauge analysis o LP3 o “fat-tailed” probability distributions, e.g. generalized logistic or GEV distribution

Rainfall distributions generated from storm catalogs and Stochastic Storm Transposition (SST) techniques

Climate change scenarios generated via SST and novel rainfall IDF techniques (M. Arabi and S. Denning, personal communication)

Nonstationary generalized logistic distribution or GEV with time-varying parameters, shifting mean models Clearly these combinations of land use and climate could result in a large number of modeling scenarios and an early task will be to identify the most informative and important combinations and contrasts for achieving project objectives. Deterministic hydraulic modeling can be readily performed across a wide range of streamflow scenarios; however, generating floodplain maps that reflect uncertainty in parameters, geometry, channel change, etc. will be much more demanding and also will require careful prioritization of scenarios by the team. This project addresses several pressures associated with urbanization including:

Land Use Land Cover Change

Economic Development

Climate Change

Extreme Events/Flood

Water Pollution We will explore innovative solutions for mitigating flood risks and degradation of stream and riparian ecosystems.

110

The hydrologic modeling scenarios described above will be designed to encompass a wide range of GI / LID solutions within the bounds of our ability to represent the influence of these practices on hydrological processes using current models:

Permeable Pavements

Green Parking Areas

Green Roofs

Bioswales and Bioretention Areas

Wetlands

Rain Gardens

Disconnected impervious

Green Alleys and Streets

Urban Tree Canopy

Stormwater Capture and Reuse

Stream Restoration

Flood control systems Given the number of different kinds of solutions listed, we will seek a limited set of combinations to represent the extent and spatial configuration of a suite of solutions within an urban watershed and work with project team members to apply similar combinations across different sites in the UWIN network. Finally, the hydrologic and hydraulic modeling outputs will allow us to link the effects of these urban solutions and innovations with previously indicators such as Frequency and Duration of Extreme Flows, Nutrients, Connectivity, % Natural land cover, % Artificial drainage area, Road crossing density, Biodiversity, Habitat, Bed/bank stability, and Riparian Losses. We will also identify new indicators through the course of the study. These novel indicators include fluvial erosion and channel change metrics.

Description of Study Regions

We will work closely with our collaborators in the project to determine the appropriate watersheds to use as principal study sites for hydrologic and hydraulic modeling studies. Decisions about cross-site comparisons and comparative treatments across sites will require further consultation among project investigators prior to completion of the research plan. The following sections provide preliminary thoughts on identifying focal watersheds in the six study regions. Front Range of Colorado The Front Range is located in north-central Colorado on the east side of the Continental Divide. Within the Front Range, Colorado’s interior continental location and the Rocky Mountains combine to produce a complex and diverse climate. The Front Range is composed of the following five primary watersheds:

Cache La Poudre River

Big Thompson River

Saint Vrain Creek

Clear Creek

South Platte River (North and South Forks) Bledsoe already has or has access to existing pre-2013 HEC-RAS models these basins; however, many of these models are being rebuilt after the Sept. 2013 floods based on post flood LiDAR and topographic surveys. Most of these watersheds have large mainstem floodplain corridors traversing Front Range urban areas. This allows modeling focused on both smaller tributaries and large main stem corridors and the potential for setting up contrasts of upland

111

GI vs floodplain GI and stream restoration given the different landscape positions of urban areas. There are three primary causes for floods along the Colorado Front Range that will be considered in project B2-2. Intense localized thunderstorms, which typically occur from May until early September, can generate flash floods essentially anywhere in the state. Floods due to thunderstorms in mountainous areas are most likely to occur at altitudes below 2,300 m as a result of higher rainfall intensities and greater aerial extents of the thunderstorms in the catchments for streams and rivers at lower elevations. Second, intense widespread rainfall events typically occur between April and mid-June and are generally limited to the Great Plains and eastern foothills. Lastly, floods associated with snowmelt are most likely to occur when excessive late winter snowfalls and low temperatures maintain a deep snow-pack over a large range of elevations into late spring. Spring rainfall on snow-pack can often result in floods throughout the United States; however, this condition has generally been less prevalent in Colorado compared to other parts of the country to date. Climate change could increase the frequency of rain on snow events along the Front Range.

Mid-Atlantic The Baltimore metropolitan area is characterized by numerous small urban and suburban watersheds spanning a range of development ages and patterns, from sewersheds without surface channels; to watersheds with >50% impervious cover, buried headwater streams, dense storm drain networks without stormwater management, and mainstem channels encased in concrete with little or no natural riparian zone; to watersheds developed under more recent environmental and stormwater regulations with extensive stormwater management and development on hillslopes with forested riparian zones protected from development. The largest rivers in the metropolitan area, the Patapsco (drainage area = 820 km2) and Gunpowder (drainage area = 1060 km2), drain watersheds with a large fraction of rural land and flow through protected state parks along their lower courses to the northeast and southwest of Baltimore City. Both also have a substantial fraction of their drainage area upstream of water-supply reservoirs serving the metropolitan area. Three streams flow through the city and reach tidewater at Baltimore Harbor: Gwynns Falls (drainage area = 171 km2), Jones Falls (drainage area = 130 km2), and Herring Run (drainage area = 81 km2). All of the regional streams drain areas mostly within the crystalline igneous and metamorphic rocks of the Piedmont province and in their lower courses flow through incised bedrock gorges of the Fall Zone before reaching tidewater near the boundary between the Coastal Plain and the Piedmont. The region has a large number of stream gages in small urban watersheds, several of which have 30-50% impervious cover. There are four gages along the mainstem Gwynns Falls, six in the highly impervious 14.2-km2 Dead Run watershed, three other active gages as well as three inactive gages on small suburban Gwynns Falls tributaries, and ten gages on other urban tributaries in the metropolitan area. The Baltimore area has a humid temperate climate with only modest variation in total precipitation throughout the year, but with significant seasonal differences in characteristic sources and intensity of precipitation. From May through October the largest rainfall events typically are a combination of convective storms of high intensity and relatively short duration or tropical cyclones; from October through April extratropical cyclones with typically longer duration and lower intensity are dominant. Almost all of the annual maximum flood events in small urban watersheds occur during the period between May through October and are controlled much more by precipitation intensity than by total accumulation. Some of the streams in the outer suburbs are influenced more by storms of longer duration and higher total accumulation. Among the gaged tributaries are two of the ten flashiest streams in the continental U.S. based on a national comparison of thousands of USGS stream gages (Brianne Smith dissertation – chapter in prep. for publication), and many others that have experienced flood peaks with 5 to 30-year recurrence intervals that are comparable to floods on the envelope curve of the most extreme events recorded in the mid-Atlantic region. Streams with drainage areas of 1-10 km2 sometimes experience changes in stage of 2-3 m in 15-30 minutes, but with very short hydroperiods. Some of the previous research completed in the region utilizing hydrologic modeling scenarios suggests that large variations in the extent of stormwater management may be responsible for significant differences in flood response, but the

112

effectiveness of traditional stormwater management as a causative factor is still not well-documented by empirical evidence. Stream restoration projects have been or are planned to be installed on several urban tributaries with stream gages, and these present an opportunity to assess whether these efforts affect the magnitude and extent of flood inundation by comparable storm events. Modeling scenarios have already been used to test some of these scenarios but the use of 2-d hydraulic models will be extended in this project to cover a common set of mitigation or green infrastructure scenarios across six study nodes in the UWIN network. In the Baltimore metropolitan area, small urban watersheds are exposed to flood risks associated mostly with intense convective storms of relatively short duration with extremely rapid flood response and short hydroperiods. There are >20 real-time USGS stream gages in small (1-20 km2) watersheds spanning a range of watershed conditions (percent impervious cover, percent controlled by stormwater management, varying land cover, development patterns and age of development, and variations in management of the riparian zone), allowing comparisons of flood response across the metropolitan region. Information on urban infrastructure e.g. road embankments, bridges, culverts, and pipe networks and outfalls, is available partly from GIS datasets already stored on UMBC servers or from field measurements. The number of sites within the Baltimore region to be included in the study has yet to be determined and this will depend on the outcome of discussions about the scope of the project at multiple UWIN nodes and the number and pattern of green infrastructure or mitigation scenarios that are adopted by the group. The two regional analyses provided above are based on extensive experience by Bledsoe and Miller who have worked (along with Smith, Bou-Zeid and Welty) on flood studies in those regions. Meixner and colleagues have detailed knowledge about the Phoenix-Tucson Sun Corridor (see below) but are not funded at levels sufficient to make a substantial investment in this part of the project. For the other three sites listed below we will need to engage with other UWIN partners to determine the extent to which comparable analyses can be conducted with the resources available. Los Angeles Metropolitan We will leverage existing collaborations with the CA State Water Board and the Southern California Coastal Water Research Project (SCCWRP) to conduct an inventory of existing data and models, and to assess stakeholder priorities for focusing our study in this region. Pacific Northwest Cascadia Initial efforts to develop rainfall data for SST analyses have focused on the Portland area. Small watersheds of the Puget Lowlands may also provide an attractive option, as they are relatively well-gaged and encompass a broad gradient of urbanization. There is also relatively large number of geomorphic monitoring stations in this region and precedent for environmental flow studies. Phoenix-Tucson Sun Corridor See discussion of Maricopa County Flood Control District in project plan by J. Smith and T. Meixner. South Florida As discussed for project B2-2a, South Florida will present some interesting challenges for assessing urban flooding because there is not a very useful stream gaging system to guide our analyses in a very complicated system. Complications also arise in that heavy inland rainfall can synergistically interact with coastal storm surge. Current models typically do couple inland rainfall flooding and coastal storm surge. Thus, there is an opportunity to investigate this fundamental source of uncertainty in flood inundation mapping in the Miami area.

Stakeholder engagement

We already work closely with agencies including both USGS and local management agencies responsible for funding

113

green infrastructure solutions. In some cases there are fully-developed watershed plans intended to meet specific regulatory goals. We will communicate with agency personnel on a regular basis in order to exchange information including information needed for our analyses and information we will be able to provide that improves the ability of the agency to assess and manage flood risk. Bledsoe working closely with Urban Drainage and Flood Control District, City of Fort Collins and engineering firms on flooding issues, stream restoration, and floodplain management in the Cache la Poudre, St Vrain, and upper South Platter basins in Colorado. 4. Expected results We expect to improve current understanding of magnitude and frequency of flood peak discharges and floodplain inundation across the study regions. We will also improve our understanding the sensitivity of peak discharge and flood hazards to upstream as well as local mitigation efforts within the channel and riparian zone. In addition we anticipate being able to produce metrics for assessing multiple dimensions of flood risk and to compare the relative sensitivity of these metrics to future climate scenarios, future development scenarios, and alternative mitigation scenarios within and across regional contexts. The investigators in project B2-2 have all previously done flood studies. A key contribution of this effort is conducting the cross-site comparisons given differences in climate and hydrometeorology, geology and geomorphic template, urban form and infrastructure, but with the goal of assessing the relative effectiveness of commonly proposed solutions across different regions. We can ask, for example, whether proposed solutions are effective at all, somewhat effective, or very effective, and whether that judgment will vary either locally within a region or across regions. Figuring out what metrics to use will be an important part of the work. The outputs from this project will ultimately facilitate examination of how integrated floodplain networks and sustainable urban drainage systems (SUDS) can be designed to simultaneously enhance flood resilience, moderate temperatures and improve human comfort, and support biodiversity. This work will provide a framework for quantifying the co-benefits (UHI mitigation, biodiversity, human well-being) of stream networks / GI that provide robust flood management in a nonstationarity world. Through this lens, floodplains are viewed as a connected network of intentionally designed ecosystems that simultaneously reduce vulnerability and renew opportunities for ecological engineering of novel socio-ecosystems and provision of ecosystem services. Sociological research and past experience with flooding and tells us that policies can be reset during times of crisis following floods; however, special interests can take advantage of such situations and push decision-makers toward a return to familiar but antiquated policies. If there is not a well-formulated vision for increasing resiliency through green infrastructure in place prior to the crisis, stakeholders will likely return to the status quo. This research challenges the prevailing narrow and deterministic view of urban stormwater and floodplain management. By educating stakeholders about the many co-benefits integrated green space - floodplain networks, they will be empowered to move beyond deterministic, single factor thinking and proactively forge a vision of serving multiple community benefits and justify it proactive investments in broad terms beyond flood resilience, i.e. clean / cool air and water, wildlife, recreation, health, and happiness. 5. General Project Information

Schedule TBD

We will leverage LTER and extensive existing work along the Front Range. Planned interactions include monthly teleconferences. At this stage, given that there are several interacting projects, we will read the project descriptions and flesh out the interactions as we have more detailed discussions regarding integration. Thus, this is a provisional document that lays out an initial vision that will undoubtedly evolve as the intersections with other disciplinary areas and other projects are worked out.

114

Project C1-1 Project Plan

Project Title Understanding the Adoption of Sustainable Water Solutions (version 2.0)

UWIN Project number

C1-1 This project replaces Project C1-1, C3-1, and C3-2 in the original UWIN proposal

Project Lead

Gary Pivo

Investigators/Institutions Funded: Gary Pivo and Adam Henry, U of AZ; Jessica Bolson, FIU; Neil Grigg, CSU. Not Funded: None

Project Period

2016-2020

Project Cost

U of A: $414,998 (all costs and overhead included) FIU: $187,470 (all costs and overhead included) CSU: approx. $26,000 for Neil Grigg (all costs and overhead included)

Graduate students funded to work on project

At UA, TBA – 2 students, .5FTE, years 2-5 At FIU, TBA – 1 student , .33FTE, years 1-5

Project Overview

We seek to characterize and understand the adoption of sustainable urban water practices in US urban areas and to discover how changes toward more sustainable practices can be deliberately facilitated. Such practices may be adopted by public and private water agencies, public transportation, open space and land use planning agencies, private land developers and property owners, and industrial water users.

Project Summary

Objective: To characterize and understand the adoption of sustainable urban water practices in US urban areas and to discover how changes toward more sustainable practices can be deliberately facilitated. Experimental Approach: Survey research and case study methods Expected Results: A clear picture of the variety of sustainable water practices being employed and what explains differences in achievement.

Supplemental Keywords

Sustainability policy, local government, cities, sustainability transitions.

Project description

115

1. Main Objectives We will develop three types of information about sustainable urban water practices: descriptive knowledge that tells us what is being done and how that varies across time and space (i.e. current status and trends); explanatory knowledge that tells us why some places adopt more practices than others; and instrumental knowledge that tells us what a community can do to increase its level of success with the adoption of sustainable water practices. The following objectives will guide us to our goals:

a. Indicators Lit and Web Review. First, we will review prior research on measuring achievement in sustainable water practices including conceptual frameworks, indicators and metrics.

b. Indicators and Causal Framework. Based on the literature we will develop a conceptual framework to guide our project. This could lead to a recommend index for sustainable policies and practices. Our current framework is as follows:

c. Sustainability Practices Survey. Using our index we will collect information from a sample of public and

private sector players to assess the level of sustainability practice in a representative sample of US urban areas. We are still considering whether it would be best limit our work to the six UWIN regions or to build

a larger, national dataset. We are concerned that the UWIN cases may not include the full range of achievement. We will collect information on both level of achievement and degree of change so we can study both the correlates of achievement and change. The sample frame will be a major challenge insofar as we seek to collect information in practices by both the public and private sectors in our sample

116

communities. This may limit our ability to deal with storm and sewer and drinking water practices, as well as with local institutions.

d. Benchmarking. Based on the indicators data, we will produce an aggregated index score or scores that allows us to summarize the aggregated level of achievement in a given location.

e. Classifying. We will explore whether places can be grouped into clusters or whether a taxonomic system can be created that groups places by level of accomplishment.

f. Explanatory Modeling. Using data collected from additional survey research and secondary sources, we will model the level of accomplishment using existing theories and frameworks including institutional context, social networks and policy learning, organizational capacity, evolutionary theory, and diffusion of innovation. We will work to determine which theories best explain achievement. Our dependent variables will be sustainability practices and outcomes. Our focus independent variables will be things that can be deliberately controlled, like organizational capacity. We will include other controls that may change but lie outside human control, such as drought. Path analysis will be used to test explanatory hypotheses and to determine both what leads to more practices and whether they are associated with sustainability outcomes.

g. Case Studies. We will also conduct case studies of the most advanced or progressive organizations to gain a more in-depth understanding of why they have been so successful.

h. Experimentation. Using lessons from the cases and modeling, we will develop strategies for increasing progress. For example, if we find that leadership, board support, green organizational culture, collaboration, or NGOs explain the adoption of green infrastructure,1 then how does an urban area develop these behaviors or institutions and can we test methods for doing so? For example, do these come about from dedicated staff, green teams, green commissions, collaborative engagement, scenario planning, networking, external advocacy, grants, mandates, regional planning, comprehensive planning, or participation in national events?

i. Self-Assessment tool. We hope to translate what we’ve learned into a self-assessment tool that can be used by those wishing to understand both how they compare to others and what they can do to make progress.

2. Intellectual merit This project will make several valuable intellectual contributions:

a. We will synthesize existing literature on sustainable urban water practices b. We will confirm or improve upon existing indicator systems c. We will explore and describe the pattern of achievement existing today in sustainable urban water solutions d. We will develop and test major hypotheses that explain achievement e. We will create practical tools that communities can use to assess and benchmark their progress and to plan

future improvements. f. We will contribute to social studies in political science, urban planning, geography, and sociology that are

concerned with the dynamics of urban sustainability and the role of social networks in policy learning. 3. Approach

Geographical extent and units of analysis

Within urban areas, there are many different decision makers who are “working on water” including general purpose governments (i.e., states, counties and municipalities), special purpose governments (i.e., water-related special districts), private water and sewer companies2, land development companies (commercial and residential), and existing property managers and occupants. We are interested in the sustainability-related policies, practices, and water system facilities adopted by these public and private decision makers.

Sample

1 Driscoll, C.T., Eger, C.G., Chandler, D.G., Davidson, C. I., Roodsari, B.K., Flynn, C.D., Lambert, K.F., Bettez, N.D.,

Groffman, P.M. 2015. Green Infrastructure: Lessons from Science and Practice. A publication of the Science Policy Exchange. 32 pages. 2 http://www.azcc.gov/divisions/utilities/water.asp

117

We will sample organizations from the urbanized areas in all six UWIN regions and probably from other regions as well.

Hypotheses

Sustainability-related practices vary and that variation can be explained by both social and natural factors such as organizational type, location, socio-economic context, organizational capacity, social networks, NGO advocacy and collaboration3, extreme event history (e.g. floods, drought), hydrology, water availability and vulnerability,4 and state and regional5 governance. Multiple theories have been offered to explain achievement in local sustainability as summarized below. We will organize and test this work empirically. They are contributing to our emerging framework (see appendix). Theory Example Explanatory Factors Key References/Proponents

Evolutionary Development pressure, events Hornberger et al 2015

Capacity Building Politics, finance, technology, management Wang et al, 2012; Lubell et al 2009

Org Change

Champions Leadership Taylor 2009

Policy Implementation Elected officials support Dalton and Burby, 1994

Policy Learning

Social Networks Relationships, Trust

Social capital and stakeholders

mediating organizations, policy brokers, transition labs

Henry et al. 2010; Portney & Cuttler 2010; Nevins et al. 2010

Socio Tech Transition landscape level change Hodson and Marvin 2010

Planning plan making, collaboration Innes, 1995

Institutional knowledge, norms, rules, markets Ferguson et al. 2013

Data

We will require several different datasets

Extreme event history: droughts,6 floods.7

Sustainable urban water policies and practices. We will build on existing datasets and supplement them as needed. AWWA’s annual rate survey includes questions about water conservation programs. They also conducted a survey of sustainability practices in 201and their annual Utility Benchmarking Survey includes questions about sources including reuse, water demand, regulatory compliance, emergency readiness training, shortage contingency planning, leaks, unplanned disruptions, projected year for shortages, stakeholder engagement, energy consumption, sewer overflows, and triple bottom line decision making.8

Literature on the adoption of sustainability practices in the public and private sectors

Agency and company mailing lists. We have been in touch US Census about obtaining the Governments Master Address File (GMAF) which includes all utilities in the US.

3 Driscoll, C.T., Eger, C.G., Chandler, D.G., Davidson, C. I., Roodsari, B.K., Flynn, C.D., Lambert, K.F., Bettez, N.D.,

Groffman, P.M. 2015. Green Infrastructure: Lessons from Science and Practice. A publication of the Science Policy Exchange. 32 pages. 4 Padowski, J. C., and J. W. Jawitz (2012), Water availability and vulnerability of 225 large cities in the United States,

Water Resour. Res., 48, W12529, doi:10.1029/2012WR012335. 5 http://www.psc.state.fl.us/publications/pdf/pai/Water_Management_districts.pdf

6 http://www.ncdc.noaa.gov/temp-and-precip/drought/historical-palmers/psi/201508-201508

7 http://ks.water.usgs.gov/pubs/fact-sheets/fs.024-00.html#HDR1

8 Hornberger, G. M., D. J. Hess, and ,J. Gilligan (2015), Water conservation and hydrological transitions in cities in

the United States, Water Resour. Res., 51, 4635–4649, doi:10.1002/2015WR016943; Landis, AE (2015), The State of Water/Wastewater Utility Sustainability: A North American Survey, Journal AWWA 107(9), E464-E473.

118

Interviews with agency and company informants.

Questionnaires from agency and company respondents.9

Plans, reports, websites, and ordinances.

State mandates.10

Analytical tools

We will use descriptive and spatial analyses to characterize the variety and geography of practices and cluster analysis to test taxonomic and stage theories of sustainable water practices. This work will be followed by the development of explanatory models following a two-step modeling process. First, variation in city-level practices will be modeled as a function of social and natural attributes. Second, emerging network modeling techniques will be used to test hypotheses of why city agents form and maintain network relationships.

Temporal extent

Our work will not involve future scenarios or forecasting. We are interested the present and recent past.

Plans for engaging stakeholders, including water institutions/agencies

We will conduct interviews and surveys with a sample of water institutions/agencies, cities, and private firms. 4. Expectations

Results

We expect to obtain a clear understanding of the variety of sustainable water policies and practices being employed by the various types of organizations that impact urban water, how those organizations can be grouped by level or type of achievement, and what explains differences is achievement.

Potential benefits

This work should allow organizations and others to benchmark their achievement, see how they could do better, and understand why they may be doing less than could be possible.

Contributions to UWIN Framework/Blueprint

Pressures:

Extreme Events – historical drought indices

Water Pollution

Political Pressures

Water Scarcity – Padowski and Jawitz (2012) Indicators:

Management Capacity – AWWA benchmarking, US census of governments

Financial Capacity – AWWA benchmarking, US census of governments

State mandates – Christiansen et al. (op cite)

Social networks – original survey research

Local politics – party affiliations, presidential voting patterns

9 See AWWA Utility Benchmarking Survey; Landis, AE (2015), op cite; and Driscoll, C.T., et al., op cite.

10 Christiansen, W., M. A. Dickinson, A. Schempp, M. Herzog, K. Mirvis, and A. Loftus (2012), The water efficiency

and conservation state scorecard: An assessment of laws and policies, Alliance for Water Efficiency, Chicago, Ill. [Available at http://www.allianceforwaterefficiency.org/.]

119

Education – level of education data from census

Wholesale and retail water prices – survey?

Environmental Attitudes – New Ecological Paradigm Scale11 Solutions:

All via current AWWA and other surveys and new survey work.

Potential to tie back to external engagement and education (citizen science, REU, RET, etc.)

We may be able to use citizen scientists to help complete surveys on their local practices and students to conduct case studies about how important practices were adopted near where they live. This will require the development of a citizen science network and coordination with other PIs to arrange student work. 5. General Project Information

Facilities

No special labs required

Personnel

U of A: Gary Pivo and Adam Henry, FIU: Jessica Bolson, CSU: Neil Grigg

Timeline

2016 – 2020 Interaction with other UWIN projects – we hope to work with the equity team to include practices that are pertinent to equity concerns.2014 Water and Wastewater Rate Survey, American Water Works Association

11

http://umaine.edu/soe/files/2009/06/NewEcologicalParadigmNEPScale1.pdf

120

Additional Contributions FIU Contribution: Salary $77,030+Travel $3,625/ Jessica Bolson, GRA Salary: $93,460 (1/3 student time), GRA Tuition: $13,355, Total FIU: $187,470 CSU Contribution: a portion of Dr. Grigg’s time as indicated on the CSU budget.

8/1/15-7/31/16 8/1/16-7/31/17 8/1/17-7/31/18 8/1/18-7/31/19 8/1/19-7/31/20

ERE Year 1 Year 2 Year 3 Year 4 Year 5 Total

COST CATEGORY Rate Funds Request Funds Request Funds Request Funds Request Funds Request Funds Request

($ amount) ($ amount) ($ amount) ($ amount)

SALARIES AND WAGES

Faculty/appointed personnel: winter/summer supplemental compensation

PI: Pivo, Gary (107,612*.00072=$77.48/hr) 150 hours 30.00% $11,622.00 $11,971.00 $12,330.00 $12,700.00 $6,540.50 $55,163.50

Other: Adam Henry (90,500*.00072=65.16/hr) 120 hours 30.00% $7,819.00 $8,054.00 $8,296.00 $8,296.00 $4,148.00 $36,613.00

Undergraduate students $ 10/hr-10 hours/week-15 weeks 3.50% $1,500.00 $1,590.00 $1,590.00 $0.00 $4,680.00

Pivo/Adams $1,500.00 $1,590.00 $1,590.00

RA AY @$39,420 @ 0.50 FTE - 4 PHD students 59.50% $32,000.00 $32,960.00 $33,948.80 $34,967.26 $133,876.06

Pivo/Adams $32,000.00 $32,960.00 $33,948.80 $34,967.26

Summer Salary for 4 PhDs @ 19/hr (160 hour per RA) 60% $7,372.80 $7,372.80 $7,372.80 $7,372.80 $29,491.20

Pivo/Adams $7,372.80 $7,372.80 $7,372.80 $7,372.80

Total Salaries $19,441.00 $60,897.80 $62,548.80 $63,907.60 $53,028.56 $259,823.76

ERE

Pivo, Gary (107,612*.00072=$77.48/hr) 120 hours 30.00% $3,487.00 $3,591.00 $3,699.00 $3,810.00 $1,962.00 $16,549.00

Adam Henry (90,500*.00072=65.16/hr) 110 hours 30.00% $2,346.00 $2,416.00 $2,489.00 $2,489.00 $1,244.00 $10,984.00

Undergraduate students $ 10/hr-30 hours/week (total)- TBA 3.50% $0.00 $53.00 $56.00 $56.00 $0.00 $165.00

RA AY @$39,420 @ 0.50 FTE - 4 PHD students 13.90% $0.00 $4,448.00 $4,581.44 $4,718.88 $4,860.45 $18,608.77

Summer Salary for 4 PhDs @ 19/hr (160 hour per RA) 3.50% $0.00 $258.05 $258.05 $258.05 $258.05 $1,032.19

Total ERE $5,833.00 $10,766.05 $11,083.49 $11,331.93 $8,324.50 $47,338.96

MATERIALS AND SUPPLIES

computer $1,050.00

TRAVEL (Domestic) $2,500.00 $3,000.00 $3,000.00 $3,000.00 $3,000.00 $14,500.00

Pivo/Henry team $2,500.00 $3,000.00 $3,000.00 $3,000.00 $3,000.00 $14,500.00

OTHER OPERATING

CAPITAL EQUIPMENT

BUDGET $28,824.00 $74,664.00 $76,632.00 $78,240.00 $64,353.00 $322,713.00

tuition remission $0.00 $33,120.00 $33,120.00 $44,160.00 $33,120.00 $143,520.00

Capital equipment $0.00 $0.00 $0.00 $0.00 $0.00 $0.00

Total budget subject to IDC $28,824.00 $41,544.00 $43,512.00 $34,080.00 $31,233.00 $179,193.00

OVERHEAD 51.5% $14,844.00 $21,395.00 $22,409.00 $17,551.00 $16,086.00 $92,285.00

TOTAL BUDGET $43,668.00 $96,059.00 $99,041.00 $95,791.00 $80,439.00 $414,998.00

121

Project C2-1 Research Plan

Project Title Using dynamic information acceleration to understand and forecast homeowner adoption of new technologies for sustainable water management

UWIN Project number

C2-1

Project Lead

Jessica Bolson, FIU/UPenn

Investigators/Institutions

Bob Meyer, UPenn (funded) Kenneth Broad, UM (funded) David Letson, UM (funded) Jeff Moeller, Water Environment Research Foundation (WERF) (Funded)

Project Period 7/31/15 - 7/31/20

Project Cost

UPenn: $179,958 UMiami: $174,997 FIU: $140,493 WERF: $75,000

Graduate students funded to work on project

TBA (GRA), Jessica Bolson TBA (postdoc), Bob Meyer

Project Overview

Strategies for improved flood resilience, stormwater management, distributed infiltration, green infrastructure, and community scale reuse of reclaimed and graywater are at varying stages of development and implementation. Understanding household scale adoption of these different strategies, and the factors that influence it, is critical to the identification of promising pathways toward sustainability. The proposed research uses an innovative approach that guides participants through a computer simulation program, HazSim, to simulate future conditions where choices about technology/policy must be made (Figure 1). By studying individuals’ decision-making behaviors, such as information seeking behaviors, investment choices, support for policy, and perceptions of strategies/technologies, we will improve our understanding of the likelihood of adoption of different urban water management practices and the barriers that might inhibit widespread adoption. Using this information we will suggest strategies for policy and investment that incorporate peoples’ preferences and decision-making biases, making them more likely to be accepted. Figure 1. Example of a HazSim simulation experiment

122

Project Summary

Objectives: The purpose of this project is to clarify how adoption decisions are made by exploring the ability of dynamic information acceleration (DII) to model how future adoption decisions might be reached, and their likely outcomes. DII is an approach to modeling decision making in which potential decision makers engage in a realistic web simulation that “accelerates” them to future time points when a technological adoption decision will be made, at which point they have the opportunity to learn about the technology in a naturalistic way, such as by searching web and print media, talking to other adopters, and viewing simulated television broadcasts (Urban, et al. 1997). The data that emerge from a DII application include not just overall assessments of likely willingness to adopt as a function of experimental manipulations in information content, but also the structure of the information gathering process that leads to that decision (Meyer et al. 2013). In addition to examining adoption behaviors, the research proposed also addresses another fundamental challenge to making cities more sustainable, which is how to get individuals and communities to take a long-sighted view and how to encourage long-term planning and investment. Using computer simulations (HazSim) and other assessment tools including surveys and stakeholder interviews, we will study individuals’ willingness to make large investments in long-term solutions and the different factors that influence willingness to pay across the selected study sites. Experimental approach:

Temporal Orientation

(Acceleration)

Information Gathering Module

Decision Module

123

The research approach is based upon the use of dynamic information acceleration to understand and forecast homeowner adoption of new technologies for sustainable water management. We will design and carry-out experiments using HazSim, an approach where potential decision makers engage in a realistic web simulation that models the decision making environment. Ethnographic methods including online surveys, semi-structured interviews, and observations of stakeholder regional meetings will also be conducted to support the DII research. Experiments will be conducted with homeowners residing across the UWIN study sites and stakeholders engaged through the regional stakeholder meetings in each site. Expected results (outputs/outcomes): The research will provide useful information regarding the likelihood of transitioning to more sustainable urban environments across the UWIN study sites. The research will also allow us to evaluate behavioral biases and perceptions of tradeoffs inherent to decisions that span temporal scales such as investments in long-term solutions vs. spending on maintenance and management (like road repair, for example) that need to be addressed today. Our results will have direct implications for the technological solutions and strategies to educate the public about them that are ultimately included in the UWIN Sustainability Blueprint. With this information, we will craft suggestions for policy instruments that might encourage choices that lead to more sustainable outcomes and reduce water related risks across the network.

Supplemental Keywords

Technology adoption, decision making,

Project description

1. Objectives Technological innovations mean little if they are not widely adopted by homeowners and communities. Understanding how adoption decisions are made, and might be influenced, is challenging for an array of reasons that include the distant time horizons involved, the lack of contemporary understanding of the technology, and the ultimate dependence of adoption on social and media factors that are difficult to forecast. The key questions to be addressed include:

How likely are different technological solutions to be adopted by current and future households?

How can those strategies be refined to increase the chances of large-scale uptake?

What are the most effective methods of education and communication for these different water technologies?

The purpose of this project is to clarify and learn about how adoption decisions are made by exploring the ability of dynamic information acceleration (DII) to model how future adoption decisions will likely be made.

Hypotheses tested:

H1: Individuals exhibit different preferences for specific water saving technologies, information types, and information sources across case study cities and these preferences influence likelihood of technology adoption. H2: Altering communication methods and education strategies can impact likelihood of adoption of water saving technologies.

Data creation:

Qualitative data on individual and group level preferences, acceptability, and likelihood of adoption of water technologies

124

2. Intellectual Merit The continued development and deployment of HazSim represents an innovative DII approach that will provide new information about human decision making. The data that emerge from the DII application include not just overall assessments of likely willingness to adopt as a function of experimental manipulations in information content, but also the structure of the information gathering process that leads to that decision (Meyer et al., 2013). 3. Approach/Activities: Potential decision makers will engage in a realistic web simulation that “accelerates” them to future time points when a technological adoption decision will be made, at which point they have the opportunity to learn about the technology in a naturalistic way, such as by searching web and print media, talking to other adopters, and viewing simulated television broadcasts. After learning about the technology, an online survey is administered to evaluate factors such as the likelihood of adoption and perceptions of the technology and to learn about the rationale behind participants’ decisions. Individuals participate in the experiment remotely through online access. This work will be augmented with observational field studies. Using a blend of ethnographic methods, including surveys, participant observation in meetings, pre- and post-meeting structured and open-ended interviews, and analysis of transcripts of meetings when available, we will be able to investigate the degree to which insights derived from the experimental studies generalize to field settings, as well as gather new insights about group processes not revealed in lab settings. The study population will include household level consumers, water managers, developers, utilities, and other water stakeholders as well as utilities representatives involved in the WERF LIFT program from each of the six case study regions. The online HazSim experiments will target larger sample populations, of no less than 400 participants, of household level water consumers. Engagement with stakeholders, in the form of surveys and semi-structured interviews, will dovetail with the regional stakeholder meetings being conducted as part of the UWIN project.

Analysis will include cross-site comparison of adoption preferences and likelihood of adoption, as well as analysis of barriers to adoption, feasibility of wide-scale community level adoption, and policy incentives that might lead to successful transitions. 4. Expected Results, Benefits, Outputs, and Outcomes: By studying individuals’ decision-making behaviors, such as information seeking behaviors, investment choices, support for policy, and perceptions of strategies/technologies, we will improve our understanding of the likelihood of adoption of different urban water management practices/technologies and the barriers that might inhibit widespread adoption of solutions. The technological solutions that are included in the research will be those proposed in the Urban Water Sustainability Framework/Blueprint. Using this information we will suggest strategies for policy and investment that incorporate peoples’ preferences and decision-making biases, making them more likely to be accepted. 5. General Project Information:

Facilities:

Programming and development of HazSim will be conducted by a Wharton based software development company.

Schedules with associated milestones and target dates:

The first year of the project, we will focus on continued development of the HazSim platform (Figure C2-2-2). We have already started working with a development company based at the Wharton school, to make the tool easily adaptable. Year two’s focus will center around experimental development, including the development of all content and design protocols. We will also conduct our first round of experiments in the second year and begin to conduct

125

ethnographic research with stakeholders. In year three we will continue our ethnographic research and analyze the results from HazSim experiments. Year four will include the completion of all analyses, including ethnographic studies, and during year five we will coordinate research products with other teams and focus on publications.

Figure C2-2-2. Project C2-1 Timeline for Years 1 and 2 6. Interactions with other UWIN projects and institutions: This project requires interaction and planning with several other UWIN research groups. For example, the project team will need to coordinate with the technological solutions research group to understand the technologies that we will be testing for. We will also need to coordinate our stakeholder-based research efforts with the stakeholder engagement team. Finally, once our results from HazSim experiments and ethnographic studies are analyzed, we will need to feedback into the team developing the Blueprint to ensure consistency across the project. 7. References:

Meyer, R., K. Broad, B. Orlove, and N. Petrovic (2013), Dynamic Simulation as an Approach to Understanding Hurricane Risk Response: Insights from the Stormview Lab, Risk Analysis, 33(8), 1532– 1552. Urban, G. L., J. R. Hauser, W. J. Qualls, B. D. Weinberg, J. D. Bohlmann, and R. Chicos (1997), Information Acceleration: Validation and Lessons from the Field, Journal of Marketing Research, 34(1), 143– 153.

126

Project D1-1 Research Plan

Project Title UWIN Envision Modeling of Present and Future Values for Sustainable Water Management Blueprint Indicators

UWIN Project number

D1-1

Project Lead

Roy Haggerty, Oregon State University

Investigators/Institutions

Dave Hulse, University of Oregon (funded) Dave Conklin, Oregon Freshwater Simulations (funded) Maria Wright, Oregon State University (funded)

Project Period

Aug. 1, 2015 – Jul. 31, 2020

Project Cost

OSU: $1,099,948

Graduate students funded to work on project

1 TBA, Roy Haggerty

Project Overview

UWIN Project D1-1 seeks, in collaboration with other UWIN investigators, to quantify baseline values of ”Blueprint indicators” for the six study regions, and, where possible, to compare the impacts on these indicators of future water innovations relative to baseline conditions. The quantification of these Blueprint indicators will be derived from algorithms provided by other UWIN investigators, together with a set of current land use land cover (LULC) representations created using national-scale data. This project will create web services which calculate Blueprint indicators from static input data sets. These web services can then be used by the overall UWIN project to produce Blueprint indicator values for the present and for alternative futures, for all six UWIN regions. This project will also create alternative future scenarios and more refined LULC representations for two of the project’s study regions, the Pacific Northwest (PNW) and the Front Range of Colorado (Colorado work to be done by our CSU colleagues in close collaboration with us). Both the algorithms for the Blueprint indicators and the LULC representation of present and future landscapes will be integrated into Envision, which is a modeling framework developed by Oregon State University (J Bolte, http://envision.bioe.orst.edu/). We will use Envision to work with PNW stakeholders to define future scenarios, simulate water resource use according to these scenarios and quantify the corresponding Blueprint indicator values. Envision will perform spatially and temporally explicit simulations of future urban water resources use within the context of evolving population, climate, and land use. We in Oregon will focus at a greater level of detail and refinement on Portland, Oregon (2.3 million people), in response to the hypothesis that there will be advantages in local stakeholder learning and participation from a finer-grained and more detailed modeling focus on fewer study regions, compared to the results from study regions that use the national scale 'blueprint indices'-driven approach only. The main tasks of the OUWIN project will be to produce the web services for calculating the regional-average Blueprint indicators, and to develop the Envision "plugins" necessary to produce corresponding spatially and temporally explicit metrics for futures scenarios. These plugins will simulate processes such as population growth, urban water demand, water distribution, storm drainage, water discharge, water reuse, urban microclimate, and the feedback between these and other processes. The UWIN-Envision code will be developed for application not just to Oregon but more generally to any of the UWIN regions, provided suitable region-specific map layers and input datasets are available.

127

Figure D1-1-1. Example of the Envision interface showing a portion of Portland, Oregon. Envision enables biophysical and human systems models to communicate via a landscape, a shared data repository that represents instantaneous conditions at specific locations. Each time a model runs, it draws its inputs from the landscape and outputs to the landscape. Envision has been widely used to conduct model-based scenario assessments incorporating both biophysical and sociocultural dimensions of landscape change (Hulse et al., 2009; Santelmann et al., 2012; Bone et al., 2013).

Project Summary

Objectives This project, in collaboration with other UWIN investigators, seeks to quantify baseline values of ‘Blueprint indicators’ for the six study regions, and, where possible, to compare the impacts on these indicators of future water innovations relative to baseline conditions. Approach This project will create web services which calculate Blueprint indicators from national scale data layers. These web services can then be used by the overall UWIN project to produce Blueprint indicator values, for the present and for alternative futures, for all six UWIN regions. This project will also create alternative future scenarios and more refined land use and land cover (LULC) representations for two of the project’s study regions, the Pacific Northwest and the Front Range of Colorado (Colorado work to be done by our CSU colleagues in close collaboration with us). Work in the Pacific Northwest will focus on Portland, Oregon where we will carry out futures modeling with Envision, a modeling framework developed by Oregon State University (J Bolte, http://envision.bioe.orst.edu/). The UWIN-Envision code will be developed for application not just to Oregon but more generally to any of the UWIN regions, provided suitable region-specific map layers and input datasets are available. Expected Results This project will create tools to calculate and compare baseline Blueprint indicators for all six UWIN regions. It will also develop a tool (UWIN-Envision) for calculating Blueprint indicators for finer-grained regional alternative futures analysis. This project will set up the framework to compare Blueprint indicators between UWIN regions, which will

128

provide insight into how different urban water management strategies may mitigate risks and create resilience to pressures from changes in climate change and population. It will also allow for an evaluation of the scale-dependence of UWIN Blueprint indicators, and the tradeoffs involved in conducting indicator-based analysis with national versus regional data sets.

Supplemental Keywords

Land use and land cover, Alternative futures modeling, Water resources, Regional planning

Project description

1. Objectives The objectives of this project are

1. Develop (with other UWIN investigators) and calculate a set of Urban Water Blueprint indicators that assess national strategies to reduce pressures on and increase resilience of urban water systems.

2. Develop web services that will enable quantification of baseline values of the Blueprint indicators for the six study regions using nationally consistent data sets. The web services will also be available to quantify Blueprint indicators for future scenarios, provided suitable region-specific map layers and input datasets are available.

3. Calculate Blueprint indicators at a fine-grained scale for future scenarios for Portland, Oregon using Envision, a platform developed at OSU for alternative futures modeling. We will work with local stakeholders to define realistic assumptions for three futures scenarios and will collaborate closely with our CSU UWIN colleagues as they conduct parallel work for the Front Range region.

This project will test the hypothesis that a suite of Blueprint indices that are regionally-integrated (averaged) over an urban area and that are based on independent and non-evolving landscapes capture the essential features of water resources and their condition. Furthermore, we seek to test the advantages/disadvantages of a greater focus and more detailed modeling of the PNW and Front Range study regions, relative to the national scale 'blueprint indices' -driven approach alone. 2. Intellectual Merit This project will create the tools necessary to calculate and compare Blueprint indicators between UWIN regions. It will set up the framework to test the hypothesis of whether the indicators calculated with national scale datasets provide a reliable estimate of urban water sustainability, or if more dynamic simulations are necessary. The finer-scaled regional analysis with Envision will allow for investigation of scale-dependence of indicators, and provide a case study of how indicators can be used to evaluate and compare alternative water management strategies given a range of possible climate and development futures. 3. Approach The national UWIN team is identifying a suite of Blueprint indicators to evaluate baseline conditions and strategies to improve urban water system sustainability. Under leadership from CSU, the national team will select a priority set of indicators for use in comparing UWIN study regions. We will work with CSU to help guide selection of indicators that can be readily calculated within web services and that will dovetail well with futures modeling. Three priorities will be: 1) functionality with national scale data sets (for example ICLUS and other datasets available on eRAMs) 2) availability of timely, well-defined algorithms and 3) tractability with available coding resources and CPU time. Once indicators, algorithms and data sets have been identified by the national team (target date January 31, 2016), we will work with CSU to begin development of the web services. The Oregon team will implement the web services (hosted at CSU) for up to 15 indicators with CSU implementing additional indicators. Investigators in each UWIN region can then use the web service to calculate baseline indicator values for a) current conditions in their metro region, b) as the basis for feedback and guidance from stakeholders in their region under the national

129

stakeholder efforts led by Florida International University, and c) to compare and contrast results at national UWIN annual meetings. The web services will also be available to quantify Blueprint indicators for futures scenarios in each region, provided suitable map layers and input datasets are available. A second focus of this project will be on modeling future scenarios for urban water systems at a finer-grained, regional scale. To do this, we will adapt Envision, an alternative futures modeling framework developed at OSU, to model a ‘common core’ set of water system phenomena that align with the priority indicators selected by the national UWIN team. This framework, called UWIN Envision, will leverage a set of core plugins developed for previous Envision projects, and be expanded as needed to enable futures modeling of priority indicators. UWIN Envision will perform spatially and temporally explicit simulations of future urban water resource use within the context of co-evolving population, climate, and land use. During development, we will test UWIN Envision first for Portland, Oregon with subsequent expansion, if time and resources allow, to other UWIN cities. As we conduct futures modeling for Portland, we will collaborate with the CSU team who will carry out a similar regional futures analysis for the Colorado Front Range using their own code. We will collaborate with the CSU team to determine consistent assumptions for regional futures modeling such as modeling time period and climate projections, and common definitions for the three futures scenarios: (1) worst case (2) integrated water future, and (3) business as usual. Agreement on these constraints will help guide conversations with stakeholders in each region who will develop region specific water management and policy assumptions for each alternative future. Our stakeholder involvement process will involve as many as six meetings with Portland water managers to build trust and encourage co-learning about pressures, indicators and solutions to urban water management challenges. We will run and analyze the three alternative futures for Portland and contrast Blueprint indicator values calculated with Envision to those calculated from national scale datasets with the CSU-based web services, and to the regional analysis carried out by the CSU team for the Front Range. If resources are available, we will expand our analysis to run alternative futures for other regions using UWIN Envision and the best available LULC future projections for their regions. Prioritization of additional regions will be jointly decided by the UWIN leadership and Oregon teams. 4. Expected Results, Benefits, Outputs and Outcomes The project will generate the following results.

1. Web services running at CSU that will allow calculation of a set of common core Blueprint indicator values for the current conditions in all six regions, using national data sets and indicator algorithms developed and delivered in a timely manner by other UWIN collaborators.

2. Stakeholder-guided, Envision-based future alternative scenarios for the Portland area. 3. Future common core blueprint indices values for three alternative futures for Portland (Oregon Team) and

the Front Range (Colorado Team) and, as resources permit, for best available future projections for the other regions.

4. Web services running at CSU that could be used to calculate blueprint indices for future scenarios for other regions.

Our regional analysis for Portland will compare indicators for different future scenarios and identify linkages and feedbacks among human, climatic, hydrologic and ecologic dimensions of the water system. This comparison will provide insight into how different water management strategies may mitigate risks and create resilience to pressures from climate change and population growth. We will also compare regional and national Blueprint indicator values for Portland in order to evaluate the advantages/disadvantages of a greater and more detailed modeling focus relative to the national scale 'blueprint indices' -driven approach alone. 5. General Project Information and Timeline Within our team, Roy Haggerty (OSU) will lead overall project management and analysis. David Conklin (Oregon Freshwater Simulations, Inc.) will work with CSU to code the web services for analysis of priority national level Blueprint indicators, and will develop code and carry out modeling runs with UWIN Envision. David Hulse (UO) will lead the stakeholder involvement process for Portland, Oregon and will oversee development of the alternative

130

futures scenarios. Table 1 identifies key milestones and responsibilities for the project. Completion of each task requires completion of preceding ones. 6. References Hulse, D., A. Branscomb, C. Enright, and J. Bolte (2009), Anticipating floodplain trajectories: A comparison of two alternative futures approaches, Landscape Ecology, 24(8), 1067– 1090. Santelmann, M. V., J. McDonnell, J. Bolte, S. Chan, A. T. Morzillo, and D. Hulse (2012), Willamette Water 2100: River basins as complex social-ecological systems, in The Sustainable City VII, 1, 575– 586 ed. M. Pacetti, WIT Transactions on Ecology and The Environment, 155. Bone, C., B. R. Johnson, M. Nielsen-Pincus, E. Sproles, and J. Bolte (2013), A temporal variant-invariant validation approach for agent-based models of landscape dynamics, Transactions in GIS, doi: 10.1111/tgis.12016, 1– 22.

131

Table 1. Project timeline.

National Blueprint Indicators (six UWIN study regions)

Regional Futures Modeling (Portland and Colorado Front Range)

Date Project Year

Task Responsible Parties

Task Responsible Parties

8/1/2015 - 1/31/2016

first half year 1

Identify and define priority water sustainability Blueprint indicators, data sets and algorithm sources that will be used to quantify baseline values for the six study regions.

UWIN leadership; Oregon team to provide consultation on tractability of indicators for implementation in web services and futures modeling

Engage Portland stakeholders in necessary conversations for alternative futures development, integrated decision units (IDUs), and access to regional data sets. Begin adapting Envision framework to known UWIN Envision requirements for Portland.

Oregon Team

2/1/2016 - 7/31/2016

second half year 1

Develop framework for indicator web services. Report at 8/16 national UWIN meeting on indicator web services status, Portland stakeholder views on indicators.

Oregon team (Conklin) in collaboration with CSU

Collaborate with CSU team to determine consistent assumptions for regional futures modeling such as time period and climate projections, and common definitions for the three futures scenarios: (1) worst case (2) integrated water future, and (3) business as usual. For Portland, begin development of IDUs and key plugin models that will enable calculation of a subset of national Blueprint indicators.

Oregon Team in consultation with UWIN leadership and CSU

8/1/2016 - 7/31/2017

year 2 Implement web services (hosted at CSU) to enable regional teams to calculate priority indicators for current (static) conditions using national scale data sets such as ICLUS.

Oregon team (Conklin) to implement web services for 15 priority indicators. CSU to implement remainder.

Conduct meetings with regional stakeholders, develop region specific water management and policy assumptions for each alternative future for Portland.

Oregon Team; CSU carries out a parallel process for Colorado Front Range

Regional investigators quantify and evaluate baseline Blueprint indicators for their regions using web services. Compare and contrast results at 8/17 UWIN all scientists (ASM) meeting.

Regional teams under guidance from UWIN leadership

Develop a nearly complete, correctly working UWIN Envision model for the purpose of demonstrating to the broader UWIN team at the summer 2018 UWIN team meeting.

Oregon Team

Date Project Task Responsible Task Responsible

132

National Blueprint Indicators (six UWIN study regions)

Regional Futures Modeling (Portland and Colorado Front Range)

Year Parties Parties

8/1/2017 - 7/31/2018

year 3 Summarize comparison of regional baseline indicators into a publication based on outcomes from 8/17 UWIN ASM.

As designated by national leadership

Continue refining and development of UWIN Envision model for Portland.

Oregon Team

If regional teams choose and have LULC datasets for projected future conditions available, use web services to calculate Blueprint indicators for future conditions. Regions present and compare and contrast results at 8/18 UWIN all scientists (ASM) meeting.

Regional teams under guidance from UWIN leadership

Run and analyze the three alternative futures for Portland and contrast Blueprint indicator values calculated with Envision to those calculated from national scale datasets with the CSU-based web services. Present at 8/18 UWIN ASM.

Oregon Team

8/1/2018 - 7/31/2019

year 4 Summarize comparison of regional baseline and future indicators into a publication based on outcomes from 8/18 UWIN ASM.

As designated by national leadership

Compare regional futures analysis for Portland and Colorado Front Range.

Oregon Team in collaboration with CSU

Prepare publications based on regional futures analysis and comparison to national indicator analysis.

Oregon Team in collaboration with CSU

8/1/2019 - 7/31/2020

year 5 If resources are available, run alternative futures for other regions using UWIN Envision and the best available LULC projections for those regions.

Regional teams under guidance from UWIN national leadership in consultation with Oregon Team

If resources are available, run alternative futures for other regions using UWIN Envision and the best available LULC projections for those regions.

Oregon Team; UWIN leadership advises on selection of priority regions

Compare and contrast results at final UWIN Annual Meeting

UWIN national leadership

Finalize related publications. Oregon Team in collaboration with teams identified in preceding step

133

134

Project D1-2 Research Plan

Project Title Cross-site comparisons and contrasts across ecohydrologic regions

UWIN Project number

D1-2

Project Lead

Mazdak Arabi (Funded)

Investigators/Institutions

Neil Grigg/CSU (Funded) Michael Sukop/FIU (Funded) Gary Pivo/UA (Unfunded) Jessica Bolson/FIU (Unfunded)

Project Period

August 2015-August 2020

Project Cost

CSU: $532,700 FIU: $66,682

Graduate students funded to work on project

1 PhD Student at CSU 2-yr postdoc at CSU 1/3 PhD at FIU

Project Overview

The overall goal of this project is to assess tradeoffs associated with sustainable urban water solutions under climatic, land use, population, economic development, planning, and policy uncertainty. We will first develop a reference case scenario for 2015 - 2100 for each of the regions, based on a comparable assumption of “business as usual”. As part of this development, we will also formulate a suite of sustainability indicators that can be used in the six regions and that measure sustainability in terms of pressures, resilience, and cobenefits. Two alternative scenarios will be developed for each region – “conservation” and “development”. In each scenario and in each region we will then be able to compare and contrast the measures of sustainability. A desired output of this project is a common set of indicators that will comprise the UWIN Urban Water Sustainability Analysis Framework. This project will define and characterize the indicators consistent with the other UWIN research and engagement activities. Project D-3 will develop data analytics and modeling services to quantify the indictors. We will use these services to quantify the sustainability indicators under current and alternative future condition, and will subsequently assess the sustainability of urban water systems in the six study regions.

Project Summary

The overall goal of this project is to assess sustainability of urban water systems under climatic, land use, population, economic development, planning, and policy uncertainty. Specifically, our objectives are to: (i) explore tradeoffs associated with solutions for urban water systems under current and alternative future climatic, land use, socioeconomic, and institutional scenarios; and (ii) identify Pareto-optimal water management solutions that are most consistent with the preferences of stakeholders in the study regions. We will first collect data related to urban water systems in the study regions from past and ongoing efforts, including two NSF-funded urban Long Term Ecology Research Networks (LTERs), five NSF/USDA-NIFA funded Water Sustainability and Climate (WSC) projects, and the NSF-funded ReNUWIt Engineering Research Center, all of which are UWIN partners. Other publically available data will be also collected for the analysis. We will also work closely with projects D1-1 and D1-3 to define, characterize and quantify sustainability indicators under current and alternative future conditions. The indicators will be integrated using top-down optimization and bottom-up scenario building approached to explore tradeoffs

135

associated with sustainable urban water solutions within each region over time, across spatial scales (e.g., building- to municipal-level), and between regions.

Supplemental Keywords

Synthesis, Blueprint, Decision Analysis, Sustainability Indicators

Project description

1. Objectives While Thrusts A, B, and C enable us to assess conditions, discover innovative technological and institutional solutions, and identify needed transitions, the network requires a way to integrate this new knowledge into an outward-looking effort to foster engagement of stakeholders in the decision making process and identify optimal solutions by minimizing pressures, while maximizing resilience and cobenefits. This will be accomplished through creation of a decision innovation dashboard (Thrust D) that tracks and rates options for both top-down and bottom-up optimization of urban water systems via stakeholder engagement processes that facilitate societal assessment. Research activities under Thrust D focus on integration of data, models and products from all other themes and projects to facilitate enhanced decision making. This thrust will enable synthesizing results from assessment projects under Thrust A, technological solution under Thrust B, and socioeconomic and management solutions under Thrust C in order to identify viable solutions that provide maximum benefits at the system level. A system approach will be used to explore impacts, benefits, and tradeoffs associated with various solutions. A desired outcome of the synthesis is to help urban planners find optimal urban form for water sustainability. The “best” form may be one that promotes conservation, which may encourage smaller lot sizes and greater density because that will reduce demand for irrigation. This is also most helpful for reducing vehicle miles travelled, which will benefit GHG emissions, energy conservation, livability and other potential co-benefit. However, it is likely that increased density would increase the likelihood of localized flooding due to imperviousness, and may also result in greater heat island effects. These effects may be sufficiently offset with an adequate overlay of sustainable solutions such as green infrastructure, structures and building arrangements that foster shade, and decentralized rainwater harvesting and low impact development techniques. The overall goal of this project is to synthesize the data, outputs, and findings of all other projects in the study regions to assess tradeoffs associated with various urban water solutions under climatic. Land use, population, economic development, planning, and policy uncertainty. We seek to identify optimal water management solutions in the study regions under uncertainty. Specifically, our objectives are to: (i) explore tradeoffs associated with water solutions for current and alternative climatic, land use, socioeconomic, and institutional scenarios; and (ii) identify water management solutions that are most consistent with the preferences of stakeholders in the study regions. We will use the UWIN urban water sustainability analysis framework to integrate information, synthesize in time and space, and create insight for decision-making purposes. The synthesis will determine the sustainability metrics comprising our Urban Water Sustainability Blueprint for baseline and future scenarios in the six study regions to address the following questions:

How will tradeoffs vary under sustainable infrastructure and technological urban water solutions?

How will tradeoffs vary across spatial and temporal scales within a city or across cities?

What spatial and temporal interconnections exist amongst urban water systems and responses across the US?

What institutional agreements that facilitate or impede sustainable management of urban water systems?

What general patterns can be identified in the relationship between the urban landscape (and its associated

136

changes to the system) and water availability, vulnerability and resiliency? 2. Approach Our research activities will be conducted according to a conceptual Urban Water Sustainability Analysis Framework. This framework is built around the following key components: reducing pressures; building resilience; increasing transition capacity to adapt and integrate; and identifying co-benefits for linked systems. The current state of water systems is unsustainable, with a trajectory toward a feared future, due to ever-increasing pressures from climate change, population growth and changes in demographics and land use, water quality degradation, and aging infrastructure. At the same time, the resilience of water systems has been diminishing due to loss of non-renewable resources, critical infrastructure and the natural capacity to recover from extreme events. To alter this trajectory toward a more desired state, transitions are needed. These transitions are most effective when they foster integration across water sectors, taking advantage of the co-benefits and heeding the tradeoffs.

Driving Forces:

The assessment will include baseline and alternative population growth, demographic change, land use change, economic development, climate change, extreme events, water pollution, loss of natural resources, loss of critical water infrastructure, policy and financing constraints, and consumer behavior scenarios. These scenarios will be developed in close collaboration with Project D1-1 activities.

Pressures:

Leveraging information (data, models, and assessments) from other projects, we will characterize pressures on urban water systems in the study regions for the observed past, current and alternative future scenarios of driving forces.

Assessment Indicators:

UWIN is characterizing and quantifying a set of common indicators that define the essential characteristics of urban water systems, allowing classification and comparison of these systems in cities across the U.S. and globally. The assessment will include indicators for various water systems including: water supply/watershed, drinking water, wastewater, stormwater, streams, and institutions. Project D1-3 will include data analytics and models that will be used to quantify the indicators. In this project, our focus is on assessing tradeoffs associated with various water management solutions under current and alternative futures for the driving forces.

Solutions:

UWIN activities focus on development of technological, infrastructure and management solutions that facilitate the transition toward integration of water systems, maximizing resource recovery and reuse, and incorporating water-sensitive urban planning. These solutions will include green infrastructure and LID, water conservation, source separation, water recycling and reuse, energy management, resource (i.e., nutrient) recovery, sustainable urban drainage networks, socioeconomic, and policy/management. 3. Activities

Task 1. Collect information for the study regions

The U-WIN network will build on an unprecedented amount of hydroclimatic, ecological, socioeconomic and institutional data and models available from existing sustainability efforts across the U.S., including two NSF-funded urban Long Term Ecology Research Networks (LTERs), five NSF/USDA-NIFA funded Water Sustainability and Climate (WSC) projects, and the NSF-funded ReNUWIt Engineering Research Center, all of which are UWIN partners. The first task of this project is to collect, organize, and make available data and information from these and other publically available resources to the entire UWIN research and engagement teams. The data categories will

137

include:

Climate

Population and demographics

Land use and zoning

Urban development patterns

Land cover

Soils

Terrain

Water supply systems

Water providers

Wastewater treatment facilitates

Water infrastructure systems

Urban drainage and MS4 systems

Stream flow and water quality data

Water body impairments

Groundwater data: water table, withdrawals, and quality

Water institutions/organizations

Water law and policy

Municipal water programs An important outcome of this task is the understanding of data gaps in each study region.

Task 2. Explore tradeoffs under uncertainty within the study regions

We will utilize both top-down and bottom-up optimization approaches to explore water sustainability tradeoffs in space and time. For the top-down approach, we will use multi-objective optimization methods (e.g., genetic algorithms) to identify Pareto-optimal urban water management solutions under uncertainty. Since various, and often conflicting, criteria are involved in the assessment of urban water systems and solutions, any solution will likely have desirable effects on some criteria while causing undesirable consequences for other factors. Subsequently, we will use Multi Criteria Decision Analysis methods, e.g. Analytic Hierarchy Process (AHP), to identify solutions from the Pareto-optimal front that are most consistent with the desires and preferences of stakeholders in the study regions. The mathematical optimization approaches will enable us to explore the tradeoff space under alternative scenarios, solutions, in time and space. The challenges that will potentially confront this approach are: 1) availability of models and data that are needed to quantify the effects of conflicting objectives; 2) varying levels of uncertainty associated with characterization and quantification of the objectives; and 3) computational cost. Implementation of top-down optimization approaches may be hindered by failure to engage stakeholder in the decision making process from the outset. As an alternative, a bottom-up approach will be used to identify best solutions from alternatives that are created by stakeholders. Our annual stakeholder meetings are anticipated to bring together approximately 500 stakeholders from various regions in the U.S. These meetings will be used as a means to generate a large number of urban water management scenarios for each of the 6 urban regions. Then, best solutions from the stakeholder alternatives will be identified and compared with the solutions form the top-down approach.

Task 3. Compare and contrast pressures, sustainability indicators, and optimal solutions across study regions

Various decision analysis tools, including graphical and statistical methods (e.g., Fig 1), will be used to compare the sustainability of water systems within each region over time, across spatial scales (i.e., building- to municipal-level), and between regions. The indicators will be directly linked to the components of the sustainability framework:

138

pressures, resiliency, co-benefits, and transition capacity. Using these indicators, we will be able to compare areas within a city, cities within a region, or regions across the country. We will be also able to compare sustainability of urban water systems over past, present and projected future conditions. Fig. D1-2-1 illustrates how our research themes will systematically explore and assess various indicators. The indicators will define the essential characteristics of urban water systems, hence providing guidance on collection of data. They will also point decision makers toward best practices for integration of urban water systems. Consistency in data collection will allow sharing experiences, peer learning and peer pressure to take actions, and also to prepare for future challenges.

Figure D1-2-1. Graphical illustration of sustainability indicators 4. Expected Results, Benefits, Outputs, and Outcomes The desired outcome of the project is to help communities in the study regions understand pressures on their waters systems and assess effects of various solutions. The products of the projects include:

The framework: o Wiki for collecting inputs from the UWIN team and ultimately the broader community

Solutions for urban waters systems in the study regions o A database of best practices (Pareto-optimal solutions) o A database of current solutions and programs (water connect website)

Urban water gaming system

Publications in peer reviewed journal

Conference proceedings and presentations

Materials for the Stakeholders’ workshops

139

A Book titled “Risk Analysis of Water and Environmental Systems”

Data Needs

This project will need data and modeling products for all other projects, which will be provided as services from Project D1-3. 5. General Project Information

Project Timeline

8/1/2015 – 12/31/2015: Define urban water sustainability indicators

1/1/2016 – 5/30/2016: Characterize indicators

8/1/2015 – 5/31/2016: Collect available data for the study regions

1/1/2016 – 8/15/2020: Continuously revise indicators based on stakeholders’ feedback

1/1/2017 – 8/15/2020: Identify optimal solutions using mathematical procedures

1/1/2017 – 8/15/2020: Develop stakeholder driven scenarios using the gaming system

1/1/2018 – 8/15/2020: Synthesize the information for cross-site learning

Roles and Responsibilities

Colorado State University (CSU): The team at CSU will lead the definition and characterization of the urban water sustainability framework indicators. The team will also apply the framework to assess the effects of urban water solutions in the study areas under prevailing climatic, socioeconomic and policy conditions as well as alternative future scenarios that will be developed in collaboration with project D1-1. Florida International University (FIU): Drs. Sukop and Bolson will assist with integration of the engagement activities with the urban water sustainability framework and blueprint. University of Arizona (UA): Dr. Pivo will examine how urban development patterns influence household water demand. 6. Interaction with Other UWIN Projects This project serves as a means for integration of all other projects and synthesis of findings. The project team will be continuously in communication with the personnel working on the other projects. All project activities will be closely coordinated with Projects D1-1 and D1-3 teams.

Appendix C: Research Experiences for Undergraduates (REU) Plan

Lead PI: Alan Berkowitz

Basic model

Pairs of students working with mentors at 5 of the six cities each summer – one social science and one biophysical or engineering student per site.

Students join research teams and, where possible, programs with other undergraduate researchers at each site

Projects are interdisciplinary at each site

The whole group convenes at the beginning of the program for a 2-3 day kickoff meeting

140

Virtual interaction takes place across sites all summer, culminating in a virtual symposium at the end of the summer.

Virtual interactions with mentors and the whole cohort continue during the following academic year.

Selected students attend the UWIN All Scientists’ Meeting in the spring following the summer of their participation

Budget (assume $8000/student for 10 students for each of 5 years)

Student stipends - $500/week for 8 weeks ($4000/student)

Housing - $1000/student

Food - $400 per student

Kickoff Meeting - $1400/student

Research expenses - 200

Travel to and from site and/or to UWIN Annual Meeting the following year - $1000

Leadership Team

Alan Berkowitz, Director

Charles Glass, Co-Director

Elizabeth Mack, Co-Director

Matei Georgescu, Co-Director

Program Coordinator

Overall Schedule (first cohort)

Date Activity

10-30-15 Receive project descriptions from all UWIN researchers

11-10-15 Advertise Program

2-5-16 Application Deadline

2-12-16 Applications distributed to mentors

TBD Virtual meeting with mentors: common vision for reviewing applications

3-4-16 Mentor choices to Director

3-11-16 Students selected

5-1-16 Pre-program correspondence (research preparation, pre-assessment)

TBD Virtual meeting with mentors: effective strategies for mentoring

6-6-8-16 Kick-off Meeting

6-20-16 Student research proposals

7-28-16 Final symposium (virtual)

7-29-16 Final day of program

Fall 2016 Monthly virtual meetings

W/S 2017 Monthly virtual meetings

May 2017 Students participate in UWIN All Scientists’ Meeting

Student Recruitment, Selection and Preparation

Solicit short project descriptions from each UWIN researcher to be used for program announcement and recruitment.

Broad advertisement via internet, email blasts, word of mouth. Use recruitment database being developed by project (Diversity Committee, etc.)

141

Presentations by WIN scientists and staff at appropriate meetings, seminars, etc. using a generic, editable slide show

On-line application process hosted by Cary Institute, collecting academic and work background and experience, project/site preference, essays about interest and career plans, contact information for 3 references

Applications reviewed by a small team of program leaders (to be determined)

Suitable applications forwarded to mentors for review and ranking

Program leaders make final decisions about cohort, contact students and fill 10 slots

Students meet in pairs (virtually) with mentors before the program begins

Welcome packet distributed to students that includes logistics, forms, and key readings in Urban Waters, research and their specific research topic

Students complete pre-program training, if needed

Mentor Preparation and Support

Virtual meeting with all mentors reviewing student applications each year to develop a common approach to reviewing and selecting students

Compile materials from other REU and related programs, and the UWIN Diversity Committee, to support mentors

Virtual meeting with all mentors hosting students to share effective strategies for mentoring, especially in light of the interdisciplinary and distributed nature of the program.

Debrief with mentors at the end of each year

Student Program Activities

Pre-program o Virtual meeting with mentor

o Background reading o Preliminary definition of questions and hypotheses o Pre-assessment

Kick off meeting (associated with UWIN All Scientists’ Meeting or a regional stakeholder meeting)

o Team / Collaboration Training

o Interdisciplinary research training

o Group building

o Communications workshop

o Emersion in urban water sustainability research (half day authentic experience), followed by

reflection

o Project resources, tools, practices and logistics

Week 1 (late) – arrive at research site/institution

o Begin schedule of 1-2 meetings/week

o Stats/R Workshop 1 – introduction to statistics and R

o Seminar - theory and urban water sustainability

Week 2 – project planning, field exploration, literature review

o Stats/R Workshop 2 – statistics and study design

o Writing Workshop 1 – writing strong proposals

o Virtual presentation of research plans by all students (End of week)

Week 3 – Data collection underway for all students

142

o Students submit written proposal (includes lay-friendly summary)

o Peer review across sites

o Workshop – Ethics in Environmental Science and Engineering

o Seminar – UWIN interdisciplinary research

Week 4 – data collection and analysis

o Forum on Opportunities in Urban Water Sustainability Research and Application

o Workshop – interdisciplinary research techniques

Week 5 – data collection and analysis

o Sharing science – students present about their research to a non-scientist audience (students,

general public, etc.)

o Stats/R Workshop 4 – data archiving and metadata – students learn how to place their data in the

UWIN data repository

Week 6 – data analysis and synthesis

o Workshop – Giving strong presentations

o Seminar – Future Pathways Discussion – graduate school, etc.

Week 7 – data analysis, synthesis and writing

o Writing Workshop 2

o Stats/R Workshop 5 – data analysis and synthesis

o Peer and mentor review of paper sections

Week 8 – data synthesis and writing

o Peer and writing-mentor review of paper sections

o Final symposium with full participation of students, mentors, other UWIN researchers, as well as

family and friends.

o Final day:

Seminar – Virtual discussion of Urban Water Sustainability and the REU program.

Draft of written paper submitted

Minimum requirements for hosting a UWIN REU student:

o Committed mentors, one in biophysical science/engineering and one in social sciences willing to

work together on an interdisciplinary project

o A suitable pair of coordinated projects that are engaging and accessible to undergraduate students,

while also contributing to the Blueprint and other UWIN goals/products

o Work/lab/office space for the students where they would work with other undergraduates,

graduate students, post docs, and staff.

o On-campus or nearby housing, with the assistance needed to help the students, etc.

o A coordinator or “point person” who can look after the students’ well being

Evaluation

o Metrics

Prior experience, academic

Knowledge in environmental science, social science and/or engineering

Skills in writing, statistics, research (study design, data analysis/interpretation),

collaboration, systems thinking

143

Interest, motivation and self-confidence in research, collaboration, graduate study and

careers

Academic and career attainment

Sustained engagement in interdisciplinary, urban waters work

Sophistication of reasoning

Productivity – presentations, posters, papers

o Tools and artifacts for evaluation (per cohort)

Application

Pre-program survey of skills, confidence, interest

Pre-assessment writing tasks: systems thinking, study design

Research plans presentation, proposals and peer reviews

Final presentation

Final paper

Post-program survey of skills, confidence, interest

Mentor survey

Alumni surveys (annual for 3 years, then every 5 years)

o Implementation

Collaboration between Leadership Team, mentors and UWIN Evaluator

144

Appendix D: Engagement Plan

Project Title UWIN Stakeholder Engagement Plan UWIN Project number E-1 Project Lead Mike Sukop Investigators/Institutions

Jessica Bolson/FIU (funded) - Co-PI Alicia Lanier/FIU (funded) - Lead Facilitator Jeff Moeller/Water Environment Research Foundation (WERF) Regional Engagement Leaders (unfunded)

C. Swan, UMBC - Mid Atlantic/Baltimore Gary Pivo, UA - Sun Corr./Phoenix Mike Sukop, FIU - S. Florida/Miami Mazdak Arabi, CSU - Front Range/Denver D. Jenerette, UCR - S. Cal/Los Angeles D. Hulse, UO - Pacific NW/Portland

Consultants (unfunded) Alan Berkowitz/Cary Institute (unfunded) - UWIN Senior Education Advisor Meagan Smith/Colorado State (unfunded)- UWIN Project Coordinator Collaborator Shirley Vincent/National Council for Science and the Environment (unfunded) - Educational and Outreach Evaluator

Project Period September 2015 through July 2020 Project Cost

Project costs are included for each primary institution involved with the project (fringe and indirect costs included). Funding for Regional Engagement Leaders is not included.

FIU: $207,337 WERF: $25,000

Expenses funded through CSU/Master budget: Travel to stakeholder meetings: $225,000 Stakeholder meeting costs (meeting room, food, supplies, audio and visual equipment): $100,000

Graduate students funded to work on project

1/3, TBD, Jessica Bolson, Mike Sukop

Project Overview

Stakeholder-centered research is critical to the development of relevant and usable science and technologies that reduce pressures, enhance resilience, maximize co-benefits and foster transitions. To this end, our stakeholder engagement strategy will take a three-pronged approach. First, we engage stakeholders to gather feedback and input to ensure that our science is relevant to decisions and drivers of concern. Second, we strive to compare findings from observations and analyses of stakeholder interactions across regions throughout the course of the project. Third, the stakeholder engagement fora will serve as testbeds for the tools and products developed through the project. Finally, we will measure the change in network composition and extent throughout the duration of the project (Figure E1-1, Plastrik and Parzen, 2012).

145

Figure E1-1. Growth in connectivity among USDN members (Plastrik and Parzen, 2012) (http://www.sustainablecitiesinstitute.org/Documents/SCI/Report_Guide/Guide_USDN-INC-Regional-Networks-Guidebook-2-0.pdf)

Project Summary

The purpose of the meetings is to develop a regional stakeholder network that will actively contribute to project research and outreach. Feedback elicited from stakeholder participants during meetings will be central to the development of the Urban Water Sustainability Blueprint. The regional stakeholder groups will also serve as a vehicle for inter-regional horizontal information exchange. As a result of the process of developing this network, we also anticipate increased social capital (expanded professional network, relationships with scientists, development of a common language, etc.) emerging among participants. The extent to which improvements in information exchange and social capital are occurring will be measured through network analysis and pre- and post-workshop surveys at each meeting. Network analysis tools capable of producing diagrams like those shown above are available from our UWIN collaborator the Urban Sustainability Directors Network (USDN, Plastrik and Parzen, 2012).

Supplemental Keywords

Stakeholders, participation, social network analysis, social learning

Project description

1. Objectives Stakeholder participation is central to the success of the UWIN project. The goals of the stakeholder engagement project are to: 1. Engage stakeholders to foster social learning and assessments that guide the research programs of UWIN including

the development of the Urban Water Sustainability Blueprint; 2. Facilitate sharing of information and experiences on sustainable management of urban water systems among

research groups and regional stakeholders; 3. Vet the Urban Water Sustainability Blueprint against desires, priorities, preferences, and concerns of stakeholders

from various regions in the U.S.; 4. Analyze the role of the participatory fora in the development of usable science and compare perceptions of usability

across case study cities; and 5. Measure the changes in the network throughout the duration of the project.

146

2. Intellectual Merit Information gathered from both stakeholder evaluation surveys and the process of interaction itself can contribute new knowledge on methods for developing and deploying usable science. Additionally, engaging stakeholders in the planned activities is expected to lead to social learning and increased adaptive capacity among participants. Another key impact of our stakeholder engagement plan is the participatory development and vetting of the Sustainable Urban Water Management Blueprint. 3. Approach/Activities

Stakeholder Selection and Hierarchy

The UWIN research team from each urban region, led by the Regional Engagement Leaders listed above, will work with the Outreach team to identify a representative group of stakeholders selected from existing local networks. Stakeholders will be selected with the intent of balanced representation across all regions. The plan is to primarily select city, county, and state government representatives considered regional thought leaders on issues including water, wastewater, stormwater, and related hazards such as Urban Heat Islands and flooding. The Appendix lists preliminary ideas for potential stakeholders in each of the regions. At the project outset, a smaller 6-8 member stakeholder advisory committee will be convened in each region (Figure 2). Together with the Outreach team and the local research team, the advisory committee will select a larger group of stakeholders of approximately 25 people as part of in-person meetings held during the first year. This larger group will be assembled for annual in-person meetings with the core Outreach team and the local research team each of the four remaining project years.

Stakeholder Network: Meeting Format

We envision a series of stakeholder meetings conducted at each of the study sites each year. A core traveling team including members of the UWIN team (Jessica, Mazdak, Gary, Mike, Alicia) will attend all of the regional meetings along with the research team from the host urban area. The travelling team will moderate the meetings to ensure uniform meeting structure and data collection for comparative purposes. With the guidance of the Regional Stakeholder Advisory Committee and the host urban area research team, we will organize one-day meetings at neutral locations. Meetings will be structured to gather and/or disseminate information pertinent to the greater UWIN project. Observations and pre- and post-meeting surveys will be conducted at each meeting to contribute to research on the development of usable science. Facilitator, meeting organizers, and the research team will meet the day before the stakeholder meeting to engage in training and preparation as described in the UWIN Training Project Plan.

Stakeholder Meeting: Organization

Day 1: Setup/organize/logistics: Outreach team will participate in setting up and preparing for the meeting to ensure a seamless meeting flow and productive environment, as each meeting will have unique requirements and space limitations. Some minor logistical assistance from the local research team could benefit the Outreach team. Day 2: Student/researcher training: Outreach team will provide a training module of up to 4 hours on facilitation and scientific project management skills for students and researchers. We will also provide training in leadership and communication. The remainder of the day will be dedicated to stakeholder meeting planning. Day 3: Stakeholder meeting: The Outreach team and the Regional Research Team will meet with the regional stakeholders for a full day meeting. The nature, goals, and specific activities of these meetings will change as the project progresses and the needs evolve. However, each stakeholder meeting will include in-meeting observations and pre- and post-meeting surveys aimed at fulfilling social science research objectives. This ‘research arc’ envisioned for the project is described in the next section.

147

Day 4: Retrospective, debriefing, analysis, and archiving of results, outcomes, meeting notes: The project outreach team will facilitate the retrospectives and capture all results. Too often, inadequate time is allotted for this crucial step; here we explicitly allocate the day after the meeting. The retrospective and debriefing will be focused on what was learned that could be used to improve the next meeting, and what challenges and opportunities presented themselves. Results and outcomes will review what data were obtained and how they contribute to the Blueprint. The social research data on the group will be compiled.

Stakeholder Meetings: Research Arc

Year 1: Building the network and learning about concerns: Convene Stakeholder Advisory Group to build group network cohesion and sense of shared purpose. Conduct group activities that help build trust, including but not limited to team building exercises and development of shared project vision. Conduct group activities and discussions focused on identifying challenges in urban water management and impacts of concern experienced by stakeholders, building an inventory of ongoing urban water sustainability projects, and establishing clear objectives for the group and for individuals involved. Engage Stakeholder Advisory Group to expand the network. We will assess the baseline network characteristics amongst the stakeholders and the research team. We will conduct stakeholder needs assessments through pre- and post-workshop surveys of participants to understand goals for participation and to gather data for the Blueprint. Year 2: Understanding decisions: Stakeholder meetings in year two will focus on improving our understanding of water sustainability decisions in testbed cities. In each meeting, we will hold presentations on sustainable water management practices across cities and the science of water depending on interests expressed by stakeholders in the prior year of meetings and follow-on discussions. We will focus group discussions on decisions being considered in each city and barriers to implementing sustainability strategies. Pre- and post-surveys will be used to gather data on decisions under consideration and responses to materials discussed during the meeting. This information will be used in further development of the Blueprint. Year 3: Developing actionable science: Based on meetings from Years 1 & 2, we will develop presentations on science, case studies on sustainable water management from other testbed cities, and focus discussions on adapting information to host city. Pre- and post-surveys and interviews will focus on barriers to using science and perceptions of usability of science. We will also update our network analysis by surveying stakeholder participants to learn about changes in network composition and character. Year 4: Best management practices: Identify best management practices collaboratively (including technical water management strategies, communication and implementation strategies, adoption strategies, etc.), based on findings from all workshops. Share iteration of Blueprint and gather feedback on product. Pre- and post-surveys and interviews will focus on further understanding of the best management practices. Year 5: Presenting updated Blueprint and extending the network: Blueprint based on integrated activities of all stakeholder group and scientist interactions will be vetted. Pre- and post-surveys will focus on barriers to using Blueprint, perceptions of usability, and changes in network composition and character.

Stakeholder Meetings: Ongoing network engagement

Due to the substantial time between regional stakeholder meetings (one year) and the wide dispersal of project participants, a strategy for maintaining engagement will be designed. This strategy will rely upon regular updates to the Urban Water Sustainability Hub, a knowledge management site which will be created and managed by the UWIN research team.

Data

Qualitative data will be collected through pre- and post-meeting surveys, interviews and observations at each study site each year. Social network analyses measuring changes in group dynamic will also be conducted.

148

4. Expected Results, Benefits, Outputs, Outcomes As demonstrated in the research arc, the goals and expected outcomes from stakeholder engagement differ slightly each year of the project. The specific contributions of stakeholder engagement strategies to the greater UWIN project goals will differ accordingly (Table 1). In Years 1 and 2, general information about the Pressures and Indicators will be assessed to build our understanding of relevant Indicators and the decision contexts in each region. The feedback that is collected in meetings during early project years will help to define and characterize the Pressures and Indicators. In Year 3, the scientific basis for Solutions will be the focus of meetings, and in Year 4 and 5, we will present and vet the blueprint.

● During stakeholder meetings, we will also be collecting qualitative data on individual expectations from participation, learning outcomes, perceptions of information, and characteristics of the network using ethnographic methods including survey tools and semi-structured interviews. Network analyses will provide useful information about the changing nature of the relationships among group participants, allowing us to measure the impact of the UWIN. 5. General Project Information

Facilities:

Stakeholder meetings will be held at a neutral location in each of the case study sites.

Personnel expertise/experience

Mike Sukop - Project Lead PI, Project lead and Director of Stakeholder Engagement: Dr. Sukop is lead Principal Investigator of the South Florida Water, Sustainability, and Climate Project, which is a 5-year $5M complex interdisciplinary project with a nationally-distributed team and a significant stakeholder interaction component. This has provided him with first-hand experience interfacing with a broad range of stakeholders. Working with Ms Lanier and Dr Bolson, he is engaging with the most prominent stakeholder groups and serves as the principal liaison between the 4-county Regional Climate Change Compact and the Florida Climate Institute.

Jessica Bolson - Co-PI: Dr. Bolson is an environmental anthropologist whose research primarily focuses on the intersection of environmental governance and climate science, with particular emphasis on water resource decision-making under uncertain conditions. She’s especially interested in the co-production of scientific information and policy within the hydroclimate context, including the roles of integrated models, stakeholder/scientist interactions, and climate information use in decision-making. Jessica also has extensive experience in education, including a Masters in Secondary Biology Education and course development in interdisciplinary environmental research approaches.

Alicia Lanier - Lead Facilitator: Alicia Lanier has owned and operated the certified woman-owned firm Lanier Consulting, LLC, in the State of Oregon (#3670) since 2002. She has 18 years of specialized experience in project management and facilitation in a career that spans over 28 years in civil engineering, water resources engineering, and biological and agricultural engineering. As a registered engineer and Human Systems Dynamics Professional, Alicia brings the group process skills and team science awareness to help teams develop ‘light-weight’ management structures for complex endeavours. Light-weight in this context means selecting just enough structure for the project at hand. Her experiences with the South Florida Water, Sustainability and Climate project, Greenwood Resources, GSI Water Solutions, the City of Portland Bureau of Environmental Services, Oregon Department of Transportation (ODOT), and other organizations illustrate her ability to work with various team environments, from co-located mono-disciplinary to distributed inter-disciplinary teams, to identify and implement creative management solutions. Through her training and coaching, Alicia focuses on supporting collaborative learning-teams in complex projects by building awareness of team science skills, consensus building and other facilitation practices, and iterative and adaptive management principles and tools.

Schedule:

149

The high-level roadmap in Figure E1-2 shows the integration of the stakeholder and training projects with milestones and target dates for the next 4 quarters of the project.

Figure E1-2. High-level roadmap for stakeholder project. Table E1-1 shows the major project activities for the duration of UWIN. Through the end of 2015, we will be conducting pre-meeting planning activities, which will include coordinating with Regional Engagement Leaders and developing a common approach across the regions. Activities include but are not limited to identifying members for the regional Stakeholder Advisory committees, proposing a preliminary meeting schedule for Year 1 for each region, and further developing the stakeholder ‘message’ for Year 1. Pre-meeting planning activities will also include developing research activities, and working with other project leaders on the integration of relevant science into the stakeholder meeting goals. Table E1-1. Meeting Arc by Year, Stakeholders, and Blueprint Status

150

As described, for each region, we plan to work with the Stakeholder Advisory committees in Year 1 meetings (2016), to present the UWIN project, the concept for the blueprint, and general indicators. The advisory group will be asked to provide a list of at least 25 people to form the larger stakeholder group (Group). At the end of Year 1, we plan to document what we have learned for presentation and discussion at the Annual Meeting and for inclusion in the NSF Annual Report. We will then begin pre-meeting planning for Year 2. This general sequence will be followed in subsequent years, with:

Stakeholders increasing in number (from the Stakeholder Advisory Committees in Year 1 to the full Stakeholder Group thereafter),

Bi-directional information flow on the Blueprint progressing from concept to draft and final vetting,

The framework indicators progressing from exploratory and general to quantified for each region,

Pressures and solutions focusing first on existing conditions and ultimately on future scenarios. The plan includes integration of results across UWIN and reporting back to the UWIN annual all-hands meeting each year. Refer to the Meeting Arc description for more detail on the anticipated approach for each year. 6. Interactions with UWIN research projects: This is an overarching project that will interact with all UWIN projects and institutions. Drs. Berkowitz and Vincent will serve as consultants and the Regional Engagement Leaders will interact with the project to provide insight into local circumstances, connection to local stakeholders, and to participate in stakeholder meetings. As the project progresses, stronger connection with the science relevant to a particular region will need to be incorporated into stakeholder interactions, and more participation of project scientists is anticipated. The Team Science Training Project is directly tied in, since the training will be conducted largely just prior to the Stakeholder meetings. 7. References: Plastrik, P. and J. Parzen, 2012, Guidebook for Building Regional Networks for Urban Sustainability 2.0, A product of The Urban Sustainability Directors Network and The Innovation Network for Communities. Supported by the Summit Foundation. 8. Potential stakeholders

Urban Area Potential Agency/Stakeholder Groups

151

Mid-Atlantic region urban agglomeration (New York tri-state area to the Baltimore-Washington, DC area (Baltimore Ecosystem Study LTER site)

— Baltimore County Department of Public Works — Blue Water Baltimore — Parks & People Foundation — US Forest Service — USGS — Maryland Water Monitoring Council — Baltimore City Office of Sustainability — Baltimore County Department of Environmental Protection and

Sustainability — Center for Watershed Protection — Chesapeake Stormwater Network — Chesapeake Bay Program — Montgomery County Department of Environmental Protection — Metropolitan Washington Council of Governments — DC Water and Sewer Authority — Washington Suburban Sanitary Commission — Anacostia Watershed Society

South Florida urban area includes the cities of Miami, Ft. Lauderdale, and West Palm Beach, and counties of Miami-Dade, Broward, and Palm Beach

— Florida Water and Climate Alliance — Miami Beach SLR Initiative — Southeast Florida Climate Compact — Miami-Dade Water and Sewer Department — South Florida Water Management District — US Army Corps of Engineers, Jacksonville District — South Florida Regional Planning Council

Phoenix-Tucson Sun Corridor, centered on Phoenix, AZ

— Water CASA (Conservation Alliance of Southern Arizona)

— City of Tucson (Tucson Water, Stormwater, Office of Sustainability) — Pima County Department of Environment Quality (Wastewater, Drinking

Water, Stormwater) — Pima Association of Governments, Sustainable Environment Program — Town of Marana Water Utility — Oro Valley Water Utility — Community Water Company of Green Valley — Watershed Management Group — Tucson Clean and Beautiful — Sonoran Institute — Sustainable Tucson — U.S. Green Building Council, Arizona Chapter — Sonoran Permaculture Guild — Desert Harvesters — Habitat for Humanity, Tucson

Front Range of Colorado — Denver Water Department — Fort Collins Water Utility — Denver Metro Reclamation — Fort Collins Wastewater Utility — Urban Drainage and Flood Control District (Denver) — Fort Collins Stormwater Utility

152

— South Denver Metro Chamber of Commerce — City of Fort Collins Advance Planning — Denver Latino Commission — Community Foundation, Fort Collins — Denver area developer/real estate — Northern Colorado developer/real estate — South Platte Greenway Foundation — Big Thompson Watershed Forum — Denver Public Works Department — Fort Collins Fire Authority — Central and Northern Water Conservancy Districts — Colorado Department of Health and Environment (CDPHE).

Willamette River Basin in the Pacific Northwest Cascadia contains the urban areas of Portland, Salem, Corvallis, and Eugene

— City of Hillsboro, Oregon — South Lane School District — City of Salem, Oregon — Oregon Department of Environmental Quality — Columbia River Inter-Tribal Fish Commission — US Forest Service, Willamette National Forest — Farmers, Benton County — Intel — Oregon Department of Agriculture — Oregon Water Resources Department — City of Eugene, Oregon — Tualatin Valley Water District — Benton County, Oregon — Eugene Water and Electric Board — City of Portland Water Bureau — US Army Corps of Engineers, Portland District — Greenberry Irrigation District — Multnomah County, Oregon — Clackamas Water Providers — Association of Oregon Cities — Clean Water Services (Tualatin, Oregon) — Oregon Association of Clean Water Agencies — Oregon State University — U.S. Fish and Wildlife Service — Starker Brothers Forests — U.S. Geological Survey — Meyer Memorial Trust — Oregon Watershed Enhancement Board — Oregon Dept. of Fish and Wildlife — University of Oregon

Los Angeles Metropolitan region: complex organization of many cities in four counties

— Cities of Los Angeles, and Riverside, Los Angeles, Riverside, Orange, and San Bernadino Counties Counties. We have good relationships with municipal governments throughout the region and could add more as of interest.

— Metropolitan Water District, Los Angeles Department of Water and Power, we have ongoing relationships with many more of the local water

153

and electricity districts. There are a huge number of local boards in the region.

— Amigos de los Rio – community group / nonprofit that works with hundreds of residents and other stakeholders around urban greening. They have developed a regional greenbelt plant.

— Climate Resolve – policy directed non-profit working in Los Angeles metroregion, they work with many officials throughout the region.

— Natural History Museum of Los Angeles — The Nature Conservancy — Los Angeles Community Garden Council — Community Garden Council — Friends of Los Angeles River, Friends of Santa Ana River — Earthwatch – citizen science organization that is making a push into urban

science opportunities — 11. UCR has a prominent extension program directed to urban types of

issues (weeds, gardening, turf, invasive species, etc.)

154

Appendix E: Training Plan

Project Title UWIN Team Science Skills Training (Including Tools for Adaptive Management of Complex Projects)

UWIN Project number T1

Project Lead Mike Sukop Investigators/Institutions

Alicia Lanier/FIU (funded) - Lead Trainer Jessica Bolson/FIU (funded) - Co-Trainer Alan Berkowitz/Cary Institute (unfunded) - UWIN Senior Education Advisor Meagan Smith/Colorado State (unfunded)- UWIN Project Coordinator Shirley Vincent/National Council for Science and the Environment (funded through subcontract with Cary Institute) - Educational and Outreach Evaluator

Project Period September 2015 through July 2020 Project Cost

Project costs for each institution involved with the project are included here. FIU costs include fringe benefits and indirect costs. Travel costs included for Jessica and Mike are for conference travel only.

FIU: $207,500 Cary Institute: $25,000/Shirley Vincent, Subcontract

Expenses funded through CSU/Master budget:

Travel to training/stakeholder meetings Training meeting costs (meeting room, food, supplies, audio and visual equipment):

Graduate students funded to work on project

None

Project Overview

Management of complex scientific networks is a growing field of research, with the recognition that “high performance requires smooth coordination” (Ren et al., 2008). A complex scientific endeavor of this nature and magnitude that supports the integration of a network of researchers and other stakeholders requires management strategies, frameworks, and tools beyond conventional approaches (Cummings and Kiesler, 2005). In addition, interactions among the researchers comprising the network are complicated by their physical separation (Cummings and Kiesler, 2007), by their wide-ranging disciplines with different norms and terminologies, and other commitments each researcher will have. All of these can lead to communication challenges that impede the desired level of collaboration. Building and maintaining trust is critical for success of such teams as well (Bennett and Gadlin, 2012). Managing this work requires skills and knowledge in facilitation, dialogue, conflict resolution, and collaboration on cross-functional teams, and an understanding of these types of complex adaptive systems that exhibit self-organization, complexity, emergence, interdependence, co-evolution, chaos, and self-similarity (Eoyang and Holladay, 2013). To paraphrase from a recent study by Sprauer et al (2015), relationships and interactions may be as important to the success or failure of these projects as technical competency. The landscape diagram below illustrates these concepts quite well for complex and chaotic projects (those far from certainty and far from agreement). Providing training to the UWIN project team and students will bolster UWIN collaboration needs, build competency in the next generation of scientists to successfully address interdisciplinary sustainability challenges of the future, and enable graduate students and post-docs to sustain and increase the UWIN research network as they advance through their careers.

155

Figure T1-1. Michael Quinn Patton, 2009, from: https://tschofen.wordpress.com/

Project Summary

A major risk in inter-disciplinary projects is that the vision of the overall project may not be realized because of lack of communication, transparency, adaptability, and trust. One way to minimize this type of risk is to improve communication (both quality and frequency), increase transparency, and help develop a high level of trust through light-weight management tools. A training curriculum for project graduate, undergraduate, post-docs, and project PIs will be provided, based on successful project management and facilitation skills training modules that have been developed and deployed over the past several years by Lanier Consulting (Lanier, 2015). Lanier Consulting’s training is oriented around the philosophy of agile/adaptive project management, informed by team science literature, and targeted to cross-functional, inter-disciplinary, and distributed teams. The objective of the training will be to train project graduate and undergraduate students and post-docs as well as project PIs in adaptive management, team science skills, and facilitation, in order to develop the next generation of inter-disciplinary researchers. We envision this community of researchers to have the effective communication and management skills for managing complex projects to reduce risks associated with individual project work and research efforts. Learning outcomes include but are not limited to the following:

Participants will learn the components and team science skills required for a successful project, and a light-weight planning technique for immediate application

Participants will practice a brainstorm method that quickly and easily helps teams come to consensus on their priorities; this method can be used to continuously build backlog of tasks and adjust priorities.

Participants will learn an iterative and adaptive workflow method that will help their entire team see their commitments and where they are in their work as the project proceeds.

Participants will be introduced to a technique for sustaining the momentum and learning and adapting as changes occur.

Tangible metrics and evaluation methods will be used to observe learning outcomes throughout the training process. One method that we will use will include an exercise that helps participants reflect on past experiences prior to training, followed by an evaluation of what they would do differently after training. Other methods will be developed

156

as training modules are finalized and in collaboration with the educational evaluation development (See Project Evaluation).

Supplemental Keywords

Group process, facilitation, team science, collaboration, consensus, pattern language, agile, lean, iterative, research, scientists, business, municipal, water, complexity

Project Description

1. Objectives The primary objectives of this training to the UWIN project team and students are to support UWIN collaboration needs, build competency in the next generation of scientists to successfully address interdisciplinary sustainability challenges of the future, and enable graduate students and post-docs to sustain and increase the UWIN research network as they advance through their careers. 2. Intellectual Merit Team science awareness and tools are being advanced by this training that incorporates human systems dynamics, agile management practices, and adaptive project management concepts. 3. Approach/Activities

What: Project Management (Team Science) Skills Training

We will provide training and experience for graduate students and researchers in project management, leadership, communication, team building, and facilitation skills applicable to large-scale interdisciplinary, collaborative, stakeholder-based research projects. Adaptive project management skills and facilitation training, including training in selected tools currently used by UWIN management (e.g., Trello) will be a focus, and will be based on an approach developed by Lanier (2015). The training will be evaluated continuously for relevance and updated as needed.

Who: Graduate students, Post-docs, Undergrads, Researchers

UWIN PIs, undergraduate and graduate students, and postdoctoral researchers will be trained in the project. The potential number of trainees in various categories from different institutions is shown in Table T1-1. Each year’s training will be an independent, self-contained module to accommodate likely shifts in the cohort of participants, although the training can become more advanced if the cohorts persist over the course of the project. Table T1-2 summarizes the number of individuals in the training cohorts of each region. Note that the Undergraduate Research Assistants (UGRAs) in these tables are in addition to potential trainees from the cohort of REU program participants. Other non-project personnel could be included as appropriate for UWIN needs and as space allows.

Table T1-1. Trainees , Summary by Institution

Institution PIs GRAs UGRAs Postdocs

Arizona State University 8 4 1

157

Colorado State University 10 4 3 1

Florida International University 3 2 1

Howard University 1 1

Oregon State University 2 1

University of Oregon 1 1

Princeton University 3 3

University of Arizona 6 4 3

University of California-Berkeley 2 1 1

University of California-Riverside 1 1

University of Maryland Baltimore County 3 3 3

University of Miami 2 1

University of Pennsylvania 1

Total 43 23 9 7

Table T1-2. Project Trainees , Summary by Region

Region PIs GRAs UGRAs Postdocs Training Cohort

Baltimore, MD 8 7 3 0 18

Miami, FL 5 2 0 2 9

Denver, CO 10 4 3 1 18

Phoenix, AZ 14 8 3 1 26

Los Angeles, CA 3 2 0 1 6

Portland, OR 3 0 0 2 5

Total 43 23 9 7 83

How: Face-to-face training

The face-to-face training will be half-day sessions immediately preceding the stakeholder meetings in all six regions, and will be led by Alicia Lanier, Jessica Bolson, and Mike Sukop. All regional project team members will be invited to this training in each region (Table T1-2). Additional training and support for team science could be provided

158

throughout the year via other project meetings, in person and virtual, as budget allows.

Retrospectives: Learning and adapting

We will conduct retrospectives (Kerth, 2001; Derby and Larsen, 2006) after each training, to assist in iteratively and adaptively improving the training, along with programmatic evaluations described in the section below titled “Project Evaluation”. 4. Expected Results, Benefits, Outputs, Outcomes Twenty-three graduate students, nine undergrads, and 50 researchers will receive training in team science, in particular in managing (participating in and leading) complex projects. UWIN PIs will also receive training as needed to support the management and expansion goals of UWIN. Specifically, project management, facilitation, leadership, communication, and team building will be incorporated in the training. Potential benefits are that graduate students, post-docs, and undergraduates will be poised to address the increasingly complex, interdisciplinary sustainability challenges of the future. Another potential benefit is that exposure to team science will help to sustain and increase the UWIN research network as students advance through their careers. An expected benefit of providing training to UWIN PIs is that collaboration, transparency, and trust will be increased, helping accomplish the mission of UWIN at the highest level of creativity. 5. General Project Information

Facilities:

Facilities for training will be those used for stakeholder meetings. We assume the overall UWIN project will provide funding for facilities, food and refreshments for participants, audio and visual equipment (projector and screen), and supplies such as large flip charts. Similarly, training/stakeholder meeting-related travel is part of the master budget.

Personnel expertise/experience:

Mike Sukop - Project Lead: Project lead and Director of Stakeholder Engagement: Dr. Sukop is the lead Principal Investigator of the South Florida Water, Sustainability, and Climate Project, which is a 5-year $5M complex interdisciplinary project with a nationally-distributed team. This has provided him with first-hand experience in managing a large project. Working with Ms Lanier and Dr Bolson, he has adaptively managed the project and incorporated new project management, facilitation, leadership, communication, and team building strategies. Alicia Lanier - Lead Trainer: Alicia Lanier has owned and operated the certified woman-owned firm Lanier Consulting, LLC, in the State of Oregon (#3670) since 2002. She has 18 years of specialized experience in project management and facilitation in a career that spans over 28 years in civil engineering, water resources engineering, and biological and agricultural engineering. As a registered engineer and Human Systems Dynamics Professional, Alicia brings the group process skills and team science awareness to help teams develop ‘light-weight’ management structures for complex endeavors. Light-weight in this context means selecting just enough structure for the project at hand. Her experiences with the South Florida Water, Sustainability and Climate project, Greenwood Resources, GSI Water Solutions, the City of Portland Bureau of Environmental Services, Oregon Department of Transportation (ODOT), and other organizations illustrate her ability to work with various team environments, from co-located mono-disciplinary to distributed inter-disciplinary teams, to identify and implement creative management solutions. Through her training and coaching, Alicia focuses on supporting collaborative learning-teams in complex projects by building awareness of team science skills, consensus building and other facilitation practices, and iterative and adaptive management principles and tools. Jessica Bolson - Co-Trainer: Dr. Bolson is an environmental anthropologist whose research primarily focuses on the intersection of environmental governance and climate science, with particular emphasis on water resource decision-making under uncertain conditions. She’s especially interested in the co-production of scientific information and policy within the hydroclimate context, including the roles of integrated models, stakeholder scientist interactions, and climate information use in decision-making. Jessica also has extensive experience in education, including a Masters

159

in Secondary Biology Education and course development in interdisciplinary environmental research approaches.

Schedule:

The training project schedule is tied to the Stakeholder meetings. Figure T1-2 shows this integration and the training project milestones for the next 4 quarters. We plan to kick the stakeholder meetings and associated training off in Miami early in the first quarter of 2016, followed by meetings in Phoenix, Baltimore, Portland and the Front Range (ideally before the UWIN annual meeting to leverage travel).

Figure T1-2. High-level roadmap for training activities. 6. Interactions with Other Projects: This is an overarching project that will interact with all UWIN projects and institutions. More specific interactions will be with the Stakeholder Project. Other interactions include UWIN’s Education project through the REU students and the Education Evaluations of the training as described below under “Project Evaluations”. 7. Project Evaluations: Evaluation of the project will use an Outcome Mapping (OM) process led by Shirley Vincent. OM is a methodology developed by the International Development Research Center for planning and assessing programs that focuses on a specific type of result: outcomes as behavioral changes that will contribute to the vision and mission of the UWIN program (Earl, Carden and Smutylo 2001). OM includes continuous self-assessment to assist project leaders in thinking systematically and practically about the outcomes and training activities of the project to adaptively manage strategies for improving the project as it progresses. The OM process is also participatory—participants in the training projects collaborate with the program leaders to articulate their intentions and personal goals, and provide feedback on

160

the project training strategies and evaluation methods and tools. This process has been widely used for evaluation in complex systems projects such as the UWIN education and outreach projects (Patton 2011). There are three primary components in Outcome Mapping:

1. Intentional Design: This component of the project is to clarify and establish consensus on the change the project hopes to bring about in the participants and how the project will define the its goals, activities, and progress toward the anticipated results.

2. Monitoring (Outcome and Performance Monitoring): The second component is a framework for the ongoing monitoring of the project’s activities and the participants’ progress toward the achievement of the desired outcomes. Data collection tools are developed for: (1) tracking the progress of the participants’ toward the desired outcomes, (2) performance of the project activities used to support the participants’ progress, and (3) effective organizational practices for the project.

3. Evaluation Planning: The third component describes evaluation priorities and the development of an evaluation plan that makes good use of resources and provides strategic benefit to the project.

The three components will be developed in collaboration with the project leaders as project details are finalized. Participants will be involved in the process once the project is implemented. Diversity of the participants in the project and their achievement of the desired outcomes will be integral aspects of the evaluation.

8. References: Bennett, L. M., H. Gadlin, S. Levine-Finley (August 2010), Collaboration and Team Science: A Field Guide. National Institutes of Health. NIH Publication 10-7660. https://ccrod.cancer.gov/confluence/download/attachments/47284665/TeamScience_FieldGuide.pdf?version=2&modificationDate=1285330231523&api=v2 Bennett, L.M., and H. Gadlin (June 2012). Collaboration and Team Science: From Theory to Practice. Journal of Investigative Medicine, (60(5): 768-775, DOI: 10.231/JIM.0b013e318250871d. Cummings, J. N., and S. Kiesler (2005), Collaborative research across disciplinary and organizational boundaries, Social studies of science, 35(5): 703– 722, doi: 10.1177/0306312705055535. Cummings, J. N., and S. Kiesler (2007), Coordination costs and project outcomes in multi-university collaborations, Research Policy, 36:1620– 1634. Derby, E. and D. Larsen. (2006) Agile Retrospectives. Making good teams great. The Pragmatic Bookshelf. Earl, Sarah, Fred Carden and Terry Smutylo. 2001. Outcome Mapping: Building Learning and Reflection into Development Programs. International Development Research Centre: Ottawa, Canada. Eoyang, G. H., and R. J. Holladay (2013), Adaptive Action: Leveraging Uncertainty in Your Organization, Stanford, California: Stanford Business Books, an imprint of Stanford University Press, 253 p. Kerth, N.L. (2001) Project Retrospectives. A handbook for team reviews. Dorset House Publishing. Lanier, A.L. (2015). Corral Project Chaos - Simply, Effectively, Immediately. An Adaptive Project Management Approach. Project Management Skills training delivered by Lanier Consulting, LLC. (unpublished) Ren, Y., S. Kiesler, and S. R. Fussell (2008), Multiple group coordination in complex and dynamic task environments: Interruptions, coping mechanisms, and technology recommendations, J. Management Information Systems, 25(1):105– 130.

161

Sprauer, W., T. Blackburn, P. Blessner, and B.A. Olson (2015). Self-Organization and Sense-Making in Architect-Engineer Design Teams: Leveraging Health Care’s Approach to Managing Complex Adaptive Systems. American Society of Civil Engineers. Journal of Management in Engineering. DOI: 10.1061/(ASCE)ME.1943-5479.0000405. early access Patton, Michael Quinn. 2011. Developmental Evaluation: Applying Complexity Concepts to Enhance Innovation and Use. Guilford Press: New York.

162

Appendix F. UWIN Risk Management Matrix (Risk Register)

UWIN Risk Management Matrix (Risk Register)

Project Project # Project manager Sponsor NSF Project artifacts Updated

ID

Risk Description P

roba

bilit

y

Impa

ct

Ris

k R

atin

g

Category

Trigger Event/Indicator

Risk Response and Description

Action Plan

Owner

Status

Date Entered

Date to Review

Ex What is this risk? What act or event initiates either the risk occurrence or precipitates the response strategy?

How will you respond to this risk and what actions will you take to match that response?

If the risk becomes a reality, what will you do in response, as a backup, or alternative/ workaround?

Who monitors this risk?

1 Delayed project start due to delay in receiving input data from predecessor project

3 2 6 Project Management Contingent project is behind schedule; Anticipated output data is unavailable

Mitigate - Insert buffer between projects determined with input from PIs and team

Project will start non-linked tasks where possible; Assessment of alternatives will be performed

PM team; Project PI

Open

2 Identified data unavailable for one or more regions

3 3 9 Technical Initial data search does not yield necessary data for all regions

Mitigate - Determine if other available datasets can be used as proxy; perform sensitivity analysis on existing data

Determine if feasible to collect/generate missing data; Assessment of alternatives will be performed

Project PI Open

163

3 Delayed project progress due to personnel unavailability

2 2 4 Human Resources The personnel designated in the project proposal becomes unavailable (e.g., quits or graduates), and the project falls behind schedule until new student/staff is found

Mitigate - Identify well ahead student graduation time or staff career change

Hire new student/staff Institution PI; Project PI

Open

4 Data are not comparable between regions

3 3 9 Technical Data collected not same (different time period, etc.)

Mitigate-Do we limit the data to bits that are comparable?

Determine if feasible to collect/generate missing data; Assessment of alternatives will be performed

Thrust Leader, Project PI

Open

5 Gap in team capabilities that reduces project momentum

3 3 9 Project Management Critical team member transitions off project temporarily or permanently

Mitigate- Personnel transition plan

PIs will be aware of the possibility and bring new team member up to speed prior to the transition if possible.

Project PI Open

6 Survey respondents provide inaccurate information

2 3 6 Technical Survey distribution Validate - Use triangulation methods to test accuracy

Employ multiple methods for assessing community achievement; Apply robustness checks in analysis

Project PI Open

7 Undergraduate student(s) and mentor(s) during REU-UWIN program

1 2 2 Project Management Mentors, responses on mid-program survey indicate student/mentor incompatibility.

Pre-program assessment of student interests, knowledge and skills; pre- program communication

Project under a new mentor, either at the host institution or within UWIN.

Education, REU- UWIN

Open

164

8 Unable to gain participants for water-heat- health sector interviews

1 2 2 Human Subjects/Data Collection

Response rate to interview requests is low, interviews are not scheduled in timely manner, interviews not available in all regions

Mitigate - work with other institutions (UWIN and beyond) to take advantage of existing relationships; minimize "cold contacts"

Rely more heavily on scientific literature to understand regional variations in water- heat issues for health sector; expand contact lists for interview requests

Project PI

9 HEC-RAS 5.0 2-d model may not adequately capture influence of bridges / other floodplain infrastructure.

2 2 4 Technical Sensitivity analysis of bridge geometry on inundation depth and comparison with 1- d models indicates lack of fidelity to actual water surfaces

Mitigate - Determine if other models (e.g. TUFLOW) provide necessary accuracy

Assessment of alternatives will be performed

Project PI Open

10 Calibration of models proves difficult

3 3 9 Technical Models do not accurately predict baseline conditions

Mitigate - Work with partners in study regions to determine local anomalies, modify models appropriately to address those anomalies

Evaluate simple model modifications to adequately address local considerations and value added via those modifications

Project PI Open

11 Hydrologic model validation is poor

3 2 6 Technical Initial model implementation shows deficiencies for one or more regions

Mitigate - Additional data sets can be integrated into the modeling system or alternative model systems can be employed

Determine if additional data are available for model implementation; Assess alternative models

Project PI Open