Source: OREGON STATE UNIVERSITY submitted to
INFEWS/T2: COLLABORATIVE: IFEWCOORDNET - A SECURE DECISION SUPPORT SYSTEM FOR COORDINATION OF ADAPTATION PLANNING AMONG FEW ACTORS IN THE PACIFIC NORTHWEST
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
EXTENDED
Funding Source
Reporting Frequency
Annual
Accession No.
1011280
Grant No.
2017-67003-26057
Project No.
OREW-2016-09856
Proposal No.
2016-09856
Multistate No.
(N/A)
Program Code
A3151
Project Start Date
Jan 1, 2017
Project End Date
Dec 31, 2021
Grant Year
2017
Project Director
Babbar-Sebens, M.
Recipient Organization
OREGON STATE UNIVERSITY
(N/A)
CORVALLIS,OR 97331
Performing Department
CE Civil Engineering
Non Technical Summary
Overview: Given the increasingly strong evidence for emerging climate change and economic trends, coordination of adaptation decisions for managing limited natural resources - such as water and arable land - in food, energy, and water (FEW) sectors, are expected to become increasingly critical. The goal of this proposal is to establish a novel, intelligent, secure, and human computation-based decision support system (called iFEWCoordNet) that will enable local and regional community actors to coordinate and co-identify robust adaptation decisions for natural resources management in FEW systems, when chronic and/or acute physical and socio-economic perturbations occur. While most studies investigate adaptation at global or regional scale, this study will focus on adaptation to climate-related and policy related perturbations in local FEW systems (where communities are most invested). The PIs will collaborate to investigate five specific interdisciplinary research objectives with pertinent research questions, one outreach objective, and one education objective. In summary, the approach will first develop formulation of interlinked adaptation decisions in interlinked FEW sub-systems, stakeholder-related parameters, and perturbation scenarios for the testbed site in Hermiston, Oregon. Next, novel mathematical and computational approaches for human computation-based Multidisciplinary Design Optimization methods and trust management models will be created, and tested for their effectiveness in enabling coordination of decisions and stakeholder participation in a secure design environment. A tightly integrated plan for research and education is enabled by the participation of undergraduate and high school underrepresented and minority students, along with stakeholder groups in various research tasks.Intellectual Merit: This proposed research is potentially transformative because, once successfully completed, it is expected to make fundamental advancements in human computation principles for multidisciplinary design optimization methods. This research will also improve our understanding of models to simulate human feedback and trust, when a network of FEW actors is engaged in distributed and coordinated planning. For the first time, a DSS will be possible that enable increased and sustained coordination among FEW actors simultaneously adapting to climate-related and other perturbations. Investigation of specific research questions will also lead to improved knowledge on how the scale and type of perturbations in the environment influences effective adaptation, and how uncertainty in stakeholder adoption affects the design of adaptation pathway alternatives and the impact on the natural resources used by FEW sub-systems. Key advancements will also be made in other domains involving the intersection of man-made, environmental, and societal systems, including: (1) improved understanding of interactive links between planning and management decisions in sectors that create a nexus (e.g., food, energy, and water), (2) improved ability to examine questions on aggregating the collective intelligence of humans in search algorithms, and (3) technological approaches for improving shared awareness, governance, and education of adaptation initiatives in multiple interconnected communities.Broader Impacts: Research in the proposed area is critical to the programmatic goals of the National Academies, the Intergovernmental Panel on Climate Change, and the US Environmental Protection Agency, all of whom have called for research on improved decision support methods to address grand challenges on sustainability and human adaptation in multiple sectors. Products - algorithms, design methods, and cyberinfrastructure - developed in this research will be transferrable to other communities and problems for future investigations. The research outcomes and results will be analyzed, interpreted, synthesized, and disseminated via multiple formats. Collaboration with existing pre-college K-12 STEM programs, Women and Minorities in Engineering program, Bioresource research and Bioenergy and Bioproducts minor program at Oregon State University, and Institute for Mathematical Modeling and Computational Science at Indiana University, which all target underrepresented students, will further expand the outreach and impact of this project's education plan for increasing recruitment and retention of minorities in STEM fields.
Animal Health Component
0%
Research Effort Categories
Basic
80%
Applied
10%
Developmental
10%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
11260993030100%
Goals / Objectives
This research aims to create new advanced capabilities for Visualization and Decision Support for Cyber-Human-Physical Systems at the FEW Nexus. Specifically, we propose a novel, intelligent, secure, and human computation-based decision support system that will enable local and regional community actors to coordinate and co-identify robust adaptation decisions for natural resources management in FEW sub-systems, when chronic and/or acute physical (e.g. climatic) and socio-economic (e.g., regional policy changes) perturbations occur.The proposed DSS will, for the first time, enable us to significantly improve our current understanding of how interdependencies among FEW sub-systems can be modeled in a coordinated design environment, and what types of perturbations in the environment necessitate coordination of decisions among local FEW actors.To accomplish the envisioned decision support system (DSS) built for FEW actors responding to perturbations, we propose a suite of five interdisciplinary research objectives with relevant research questions, one outreach objective, and one education objective, which will guide the collaborations in this project"Research Objective #1: Assess and contextualize relevant actors' perceptions of the adaptation problem at the nexus of FEW sectors.Research Objective #2: Formulate quantitative descrip-tions of scenarios of decisions, constraints, criteria, and likely perturbations for adaptation problems relevant to FEW sub-systems.Research Objective #3: Create and evaluate a human computation-based MDO framework in iFEWCoordNet for development of robust adaptation alternatives.Research Objective #4: Create computational learning agents, which are able to discern and learn the interacting FEW actors' preferences, local knowledge, and behavior revealed in the cyber space, and then test approaches for integrating agents as surrogate users in the distributed optimization algorithms.Research Objective #5: Create and evaluate a web-based, trust and security management system for iFEWCoordNet.Outreach Objective #6: Implement and evaluate the proposed iFEWCoord-Net DSS in crowdsourcing the design of community-preferred adaptation pathways for the FEW systems at testbed site in Hermiston, OR.Education Objective #7: Advance the scientific literacy of students at multiple levels on the topic of environmental change and adaptation planning.
Project Methods
Tasks for Research Objective #1 (Tilt, Reimer, and Babbar-Sebens): Assess and contextualize relevant actors' perceptions of the adaptation problem at the nexus of FEW sectors.Task 1.1. Assessment mapping of the existing FEW actors in Hermiston OR: To create plausible scenarios of adaptation problems concerning FEW sub-systems, we first must understand how actors perceive the FEW nexus in Hermiston, OR, and how do they perceive connections and interdependencies. A series of workshops and focus group interviews will be held with relevant FEW actors. Task 1.2: Development of indicators of collaborative decision making: In order to model how FEW actors may collaborate in cases of acute or chronic perturbation scenarios, we will first assess how actors currently perceive risk, and interact and collaborate. Shared social networks and social capital are indicators of collaborative decision-making[75] and can be assessed in focus groups using a variety of qualitative and quantitative tools such as surveys and social network analysis[76], [77]. Measuring the relationship between embeddedness (amount and type of social capital), strength, and density of social networks [78] within the actor's FEW sub-system, among FEW sub-systems, and with the broader local community will provide critical insights on how actors may trust and share knowledge with one another with other actors.Tasks for Research Objective #2 (Murthy, Reimer, and Babbar-Sebens): Formulate quantitative descriptions of scenarios of decisions, constraints, criteria, and likely perturbations for adaptation problems relevant to FEW sub-systems.?Task 2.1. Formulate adaptation decision options: Adaptation decision optionswill be formulated based on a variety of policy and structural strategies for coping with uncertainties (e.g., no-regret strategy, reversible/flexible strategy, soft strategy, strategy with reduced decision making time horizons, etc. as prescribed by Hallegatte[4]). For every considered FEW sub-system, quantitative and qualitative formulations of adaptation decision options will be generated with the help of actors. Consideration will be given to formulations that help to clarify dependencies across multiple sub-systems and spatio-temporal scales (e.g., for both short- and long-term coordination).Task 2.2. Develop decision dependencies and simulation models for sub-systems in FEW sectors: The adaptation planning problems for each sub-system and the overall system will be formulated based on multiple physical-economic simulation models. The simulation models will enable us to determine dependencies between multi-sector decisions identified in Task 2.1 and assess impact of decisions on the physical sub-systems (e.g., wheat farms) related to the FEW sectors. A variety of physical simulation models currently already exist and are used by actors for assessing hydropower production, dam operations, crop growth, hydrology, etc. in this region. These models will be directly used in the related sub-system k's simulation-optimization models.Task 2.3 Develop scenarios of perturbation types. We will use the available regional trends in climate change, land use change, and population trends in the Columbia River Basin and the Hermiston, OR area to evaluate the effect of both acute and periodic natural[86] and anthropogenic disturbances and stresses [87] in the basin.Tasks for Specific Research Objective #3 (Babbar-Sebens, Mukhopadhyay): Create and evaluate a human computation-based MDO framework in iFEWCoordNet for development of robust adaptation alternatives. ?Task 3.1 Develop human computation-based MDO framework. To build the human computation-based Multidisciplinary Design Optimization (MDO)-based framework,adaptation in each sub-system will be first formulated as robust and dynamic optimization problems. Task 3.2. Development of adaptive visualization interfaces for actor participation.In this research, we will expand on our previous work to create adaptive visualization of coupled FEW sub-systems in a strategy-game-like[102], distributed design environment, similar to architectures used for strategic distributed decision making in Gaming literature[103] and Command and Control literature (e.g., Network Centric Warfare[104]).Task 3.3 Validation simulation experiments with different measures of efficiency and effectiveness of adaptation. Factorial simulation experiments will be carried out to validate the proposed coordination mechanisms and resolve research questions.Tasks for Research Objective #4 (Mukhopadhyay and Babbar-Sebens): Create computational learning agents, which are able to discern and learn the interacting FEW actors' preferences, local knowledge, and behavior revealed in the cyber space, and then test approaches for integrating agents as surrogate users in the distributed optimization algorithms.Task 4.1. User modeling techniques to create surrogate actors that represent human actors in cyber space. The problem of computational learning agents to represent human actors has been extensively studied in the human-computer interaction community under the topic of 'user modeling'.In this task we will adapt such user modeling techniques for human-assisted design conducted by optimizers.Tasks for Research Objective #5 (Durresi and Mukhopadhyay): Create and evaluate a web-based, trust and security management system for iFEWCoordNet.Task 5.1. Model trust among actors and user models, and aggregate trust for recommended decisions generated for individual FEW sub-systems and for overall system level: We will use an evidence-based trust management system to derive trust information from actor interactions on the game-like environment of iFEWCoordNet's user interfaces.Task 5.2. System security for human computation-based MDO components in the cyber-human layers: Security in our system will encompass authentication, integrity, and accountability. We assume that each actor participating in the sub-system or system level design tasks is in the possession of a private/public key pair, obtained either using a conventional PKI CA [116], or through identity-based encryption (IBE) [117]. Furthermore, our system will provide accountability by recording securely a synopsis [118] of the traffic that was forwarded by an agent.Tasks for Outreach Objective #6 (Tilt (lead) and all): Implement and evaluate the proposed iFEWCoordNet DSS in crowdsourcing the design of community-preferred adaptation pathways for the FEW systems at testbed site in Hermiston, OR.?Task 6.1: Evaluation and outreach: We will train actors on how to appropriately use the DSS, evaluate the DSS success based on how effectively it is able to gage actors' perspectives, and support a user-friendly environment for coordinated adaptation planning exercises.Tasks for Education Objective #7 (Ganti (lead) and all): Advance the scientific literacy of students at multiple levels on the topic of environmental change and adaptation planning. Scientists and engineers in the 21st century must be proficient in not only science and engineering issues related to the field of Adaptation Planning in FEW sectors, but also to the underlying human dimensions and computational algorithms and DSS issues[119]. To work toward this goal, we plan to continue our active engagement with undergraduate students at both collaborating institutions (e.g., the Women and Minorities in Engineering (WME) group), summer pre-college STEM programs, and with one/two Bioresource research and Bioenergy and Bioproducts minor program's undergraduates every year for fourfor participation in the proposed objectives.

Progress 01/01/19 to 12/31/19

Outputs
Target Audience:In this third year of the project, 1/1/2019 - 12/31/2019, we continued our outreach to a wide range of audience who are direct as well as indirect beneficiaries and stakeholders of this project. These include: Project personnel: During the third year, this project has involved one postdoc (Dr. Samuel Rivera), 6 graduate students (Mr. Nicholas Giles, Ms. Holli Mondo, Mr. Shreyansh Mohnot, Mr. Suleyman Uslu, Ms. Davinder Kaur, and a new PhD student Mr. Sudip Gautam). The students come from different demographics as well social backgrounds, including 4 of them being international students. Similar to previous years, this year we continued to focus on advising students on how their own disciplinary perspective and body of knowledge intersects with those of other disciplines participating in this research. Facilitation of interdisciplinary advising was accomplished via the collaborative learning environment that has been created for students in this project. Students are regularly grouped in smaller teams to help them effectively learn from each other and execute interdisciplinary ideas and concepts into individual sub-tasks of the project. Students also routinely deliver presentations on their own domain topics and their results/discussions during regular project meetings. The presentations were followed by brainstorming discussions within the entire team on how the content of the student's presentation overlaps with and influence the novel interdisciplinary ideas, methods, and body of knowledge being developed in this project. We also conducted "hackathons" to allow students and faculty to work collaboratively on solving pending tasks (e.g., model development and code development) that require everyone's input. The postdoc, Dr. Sammy Rivera, also took leadership roles in some of the sub-tasks of this project and assisted other students in specific aspects of the model development. Dr. Rivera continued to serve as task lead in the development of the simulation-optimization models/framework while mentoring graduate students, mentor undergraduate students via summer Research Experience for Undergraduate (REU) project, and serve as the lead on the cyberinfrastructure development efforts and several "hackathons". Because of his role in such efforts he was involved in the development of research proposals, software development plans, project management tasks, and the writing and presentation of scientific findings. All these activities contributed to his continued training and growth as an academic and will ultimately better prepare him as he pursues a tenure-track faculty position. Food-Energy-Water Stakeholders: In the second year, we continued engagement with stakeholders in the food, energy, and water sectors at the study site (Hermiston, Oregon area) and in the larger Columbia River basin region. Details on this engagement are provided later under Objective 1 discussion on accomplishments in this report. In summary, these stakeholders included private as well as public entities: Calpine Corp, City of Hermiston, Columbia River Inter-Tribal Fish Commission, Confederated Tribes of the Umatilla Indian Reservation (CTUIR), Oregon Department of Environmental Quality (ODEQ), Oregon Water Resources Department (OWRD), Hermiston Agricultural Research and Extension Center, Hermiston Foods, LLC, Hermiston Generating Company/Perennial Power, Hermiston Irrigation District, IRZ Consulting, Madison Farms, Morrow County Commissioner, Morrow County Soil and Water Conservation District, NORPAC, Northeast Oregon Water Association (NOWA), Northwest Food Processors Association, Northwest Natural, Port of Morrow, Renewable Northwest, Sage Center, Umatilla County, Umatilla Electric Cooperative, US Army Corps of Engineers, US Bureau of Reclamation. We used multiple methods to engage with stakeholders, which included presentations to small groups, workshop, field trips and one-on-one meetings with individual stakeholders. These engagements are being continued throughout the project. Cyberinfrastructure and Science Gateway Community: The decision support system being developed in this project (called InterACTWEL or Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land; http://interactwel.oregonstate.edu) will be supported as an application-specific Science Gateway (http://interactwel.org/). A Science Gateway (ScG) is a "community-developed set of tools, applications, and data that are integrated via a portal or a suite of applications, usually in a graphical user interface, that is further customized to meet the needs of a specific community" (source: https://www.xsede.org/web/site/ecosystem/science-gateways). Supported via an effort by National Science Foundation and multiple universities, the experienced and professional staff from the Science Gateway Research Center (SGRI) are assisting our team in developing the cyberinfrastructure for InterACTWEL using state-of-the-art software and hardware resources, including high-performance computation resources (e.g., Jetstream), workflow tools, general or domain-specific analytic and visualization software, collaborative interfaces, job submission tools, and education modules. Over the second year, we continued to closely interact with the staff at SGRI and at XSEDE (eXtreme Science and Engineering Discovery Environment) at Indiana University to develop cyberinfrastructure components of InterACTWEL. We also obtained a startup allocation with SGRI and XSEDE to develop a prototype of InterACTWEL over the next one year. The SGRI is also helping develop graphical user interfaces that will be used to communicate adaptation plans requiring coordination among food, energy, and water sectors to end-users of the DSS. Scientific Community: We continued to disseminate our work to the research community via conference presentations and journal submissions (see details in the later sections of the report). With respect to international collaborations based on this project, we also participated in the NSF and NSFC sponsored 13th Annual US - China Eco-Environmental Symposium "Research and Innovation at the Nexus of Food, Energy, and Water Systems" in Seattle WA, in October 2019. We have also developed new partnerships that advance ongoing work in this project in India on the topic of food-energy-water nexus, and recently leveraged this project to successfully win a new research award on a food-energy-water project in India funded by the Government of India's Ministry of Human Resource Development initiative on Scheme for Promotion of Academic and Research Collaboration (http://www.sparc.iitkgp.ac.in). Finally, a webinar on the work being conducted in this award was also presented to the members and stakeholders of Cascadia Lifelines Program in Oregon. K-12 and Undergraduates: In the second year, we had recruited two high school students via the Apprenticeships in Science & Engineering (ASE) program supported by Saturday Academy (https://www.saturdayacademy.org/ase) to develop a cooperative and strategic board game on the food, energy, and water systems and decision makers in the Umatilla region, the study site of this project. In the third year, we recruited two undergraduates to test the game for multiple scenarios. The game developed in this project will be used by the team to engage stakeholders in future planned activities, as well as new K-12 activities. Additionally, the project research was leveraged to support undergraduate research for a new National Science Foundation funded REU site ("Engineering for Bouncing Back"; https://cce.oregonstate.edu/ebb) at Oregon State University. A minority student from American Samoa was recruited as the REU participant and was mentored to engage in a sub-task related to this USDA-funded INFEWS project. Changes/Problems:None What opportunities for training and professional development has the project provided?Since the beginning, this project has involved two postdocs (Dr. Hossein Tabatabaie and Dr. Sammy Rivera), 9 graduate students (Mr. Sudip Gautam, Mr. Nicholas Giles, Mr. Majid Farahani, Mr. Chris Cody, Mr. Shreyansh Mohnot,Mr. Yefeng Ruan, Ms. Lina Alfantoukh, Mr. Suleyman Uslu, and Ms. Davinder Kaur Ruan, one hourly student (Ms. Holly Mondo), and 2 high school students. Post-doctoral scholars involved in the project have actively mentored participating graduate students, undergraduate students and high school students, lead efforts in sub-tasks of Objectives #1, #2, and #3, lead efforts on team manuscripts, and participating in opportunities for career development. Specific to Dr. Rivera, some of these career development opportunities included: 1) a software development "hackathon" for which he visited the SGRC facilities located in Indianapolis and worked closely with SGRC and XSEDE staff, 2) attendance of conferences to both learn more about current research efforts within the FEW nexus and present the project progress, and 3) serve as a coordinator at Oregon State University's new NSF-funded REU Site on Engineering for Bouncing Back. Via regular group presentations during project meetings (which occur once every two weeks for the entire group, and once every alternating week for sub-teams focused on specific objectives), all students receive feedback on how to improve their approach to design, analysis, discussion, and presentation. Students enhanced their research skills, such as searching for new materials (papers, books, websites), studying research papers, discussing them in class, presenting, summarizing and writing their new knowledge in the form of a research paper. Students were also engaged in assisting the faculty members and the Science gateway staff in the cyberinfrastructure development process. Students also participated in conference presentations, as outlined in the list of publications. How have the results been disseminated to communities of interest?For the public, we have developed the following: Public website for information on the project: http://interactwel.oregonstate.edu The science gateway webportal was developed (http://interactwel.org)for research community's access to InterACTWEL's software modules and components, and high performance computing infrastructure. Video of the InterACTWEL project was developed by an undergraduate student Mr. Finley King, and is posted on the InterACTWEL website. Board game that simulates InterACTWEL decision making space and provides educational and learning resources for future stakeholder activities and support pedagogy in K-12 and university courses focused on topic of adaptation planning and resilience in FEW sectors. Furthermore, individual investigators also disseminated research to discipline-specific research communities. PI Babbar-Sebens and her team presented the work via oral presentations and posters at multiple conferences in 2019, including List of papers and manuscripts are listed in detail in section on Publications. Co-PI Tilt: Dr. Tilt was invited to participate in a High-Level Panel on Food-Energy-Water Systems Transdisciplinary Environmental Research Network (FEWSTERN) at the University of Tennessee in Knoxville, TN in April 2019. Dr. Tilt also presented the FEW stakeholder engagement process, results, and preliminary DSS visualizations at the NSF and NSFC sponsored China-US 2019 Joint Eco-environmental Symposium: Advances in Critical Needs for the Nexus of Food, Energy, and Water systems in October 2019. Dr. Tilt also helped organize this conference. A manuscript based on the stakeholder engagement processed developed for the Umatilla Region Food, Energy, Water Nexus has recently been submitted for peer-review. . Co-PI Reimer: Reimer presented the economic component at Dept. of Applied Economics OSU in Jan. 2019, and at the Western Agricultural Economics Association meetings in Coeur d'Alene Idaho, June 2019. Manuscript submitted to journal in Nov. 2019. Co-PI Durresi: Durresi and his students published four papers, one developing a decision support system based on trust, another using our trust model to detect fake users in the system, another one extending the trust model in cloud, and a paper describing the use of our trust in a game-theoretical approach in our decision support system. List of papers and manuscripts are listed in detail in section on Publications. Co-PI Mukhopadhyay: Mukhopadhyay published three papers, one on interactive machine learning for user modeling, the second on multi-agent reinforcement learning using an MDO approach, and the third on Computational game Theory, in technical conferences. Further, another paper on developing an Multi-Disciplinary Optimization (MDO) approach for reinforcement-based machine learning systems, has been submitted for peer review at the AI Communications Journal. List of conference presentations, papers, and manuscripts are listed in detail in section on Publications. What do you plan to do during the next reporting period to accomplish the goals?Research Objective #1: Assess and contextualize relevant actors' perceptions of the adaptation problem at the nexus of FEW sectors. Develop, disseminate, and analyze a FEW stakeholder survey to further validate the InterACTWEL systems diagram and modeling assumptions Continue to conduct small workshops with key stakeholders to test out and continue the co-development of the GUI-interface of the DSS. Work with the project team Ensure user-needed adjustments are made to the interface. Research Objective #2: Formulate quantitative descriptions of scenarios of decisions, constraints, criteria, and likely perturbations for adaptation problems relevant to FEW sub-systems. Complete the FEW computational modeling framework and validate it using the historical data sets. Conduct sensitivity analysis to identify interdependencies among decisions in FEW sectors in Umatilla region's local communities, for perturbations scenario of new water rights identified in Year 3. Submit new manuscript on problem formulation and sensitivity analysis. Research Objective #3: Create and evaluate a human computation-based MDO framework in InterACTWEL for development of robust adaptation alternatives. Complete development of graphical user interfaces to visualize the adaptation planning problem in local FEW systems of a community, and conduct usability studies with end-users. Test the MDO approach with benchmarks and FEW case study, and submit a manuscript for peer-review publication. Integrate a test adaptation planning problem formulation into the MDO, and test performance. Continue development and testing of initial prototype of InterACTWEL Science Gateway with software components integrated with each other, and conduct workshops to obtain feedback on evaluation of the prototype by stakeholders. Research Objective #4: Create computational learning agents, which are able to discern and learn the interacting FEW actors' preferences, local knowledge, and behavior revealed in the cyber space, and then test approaches for integrating agents as surrogate users in the distributed optimization algorithms. Develop and test neural network surrogate models for various InteractWEL sub-systems to help in the MDO-based design process Develop a multi-level modeling approach for user modeling to deal with the "small data" problem Develop an iterative, interactive machine learning approach for user modeling in a targeted manner using design of experiments principles Develop methodologies for incorporation of the user models in the integrated design process for adaptation planning in interconnected food-energy-water sectors. Research Objective #5: Create and evaluate a web-based, trust and security management system for InterACTWEL. Under this objective, over the next one year, we plan to continue to work on our trust-based multi stakeholders decision making system for InterACTWEL. Specifically, we will study how to better map the trust between pairs of users in the decision making process. Then we will investigate several approaches on how to integrate the historical values of trust between pair in the global trust of each user. Furthermore, we will work on better integration of trust in InterACTWEL's MDO design process. One of the sub-objectives is to study the behavior and effectiveness of our trust based system when used by various types of stakeholders. For this reason, we also plan to develop a simulator for FEW actors that mimics various users' behaviors, based on the findings from workshops and interviews conducted in Objective 1. Furthermore, we will focus on the role of community, individual trust pressure and biases. Our goal is to incentivize cooperation between users in deciding the best solution for the scenario. Outreach Objective #6: Implement and evaluate the proposed InterACTWEL DSS in crowdsourcing the design of community-preferred adaptation pathways for the FEW systems at testbed site in Hermiston, OR. Conduct workshops to validate assumptions in Objective #2 Conduct workshops to perform stakeholder-driven scenario development via InterACTWEL GUI. Education Objective #7: Advance the scientific literacy of students at multiple levels on the topic of environmental change and adaptation planning. Engage undergraduate in Research Experience for Undergraduates on this project.

Impacts
What was accomplished under these goals? Research Objective #1: This year, our engagement with stakeholders (led by co-PI Tilt) shifted focus from gaining a shared understanding of the Food-Energy-Water Nexus in the Umatilla basin to the co-development of the DSS. This co-development process has included workshops held at the end of May 2019 with stakeholders to test out the first version of the DSS and solicit critical feedback. This first version DSS employs a resilience and adaptation scenario in which a water right trading market has been established in the basin and growers are able to trade water rights to increase overall water efficiency--however no additional water rights can be obtained. In the workshop, stakeholders were able to become familiar with this scenario in the DSS and walk through a series of steps where they selected resource management goals, actors who would achieve these goals, and a set of adaptation actions for these actors to implement. Based on the stakeholders' selections, the DSS tool visualized a series of environmental (e.g. nitrogen loading, water use), and economic (e.g. crop production) metrics. Experimenting with the DSS in this workshop setting opened the dialogue for critical feedback regarding the DSS, as well as future scenarios that the DSS should incorporate. Co-development of the DSS has also required several smaller meetings and discussions with specific stakeholders to accurately parameterize water right policies and actions, grower inputs, energy regulations and grid management, among other factors. Research Objective #2: This year, we investigated existing approaches and developed a hierarchical optimization-simulation approach for distributed adaptation planning in multiple sectors of local communities. This approach involves a system-of-systems simulation-optimization model based on a linear and time-variant input-output model that identifies resources allocation decisions for FEW sectors aggregated at the basin scale. The system-of-systems model then communicates aggregated estimates of regional decisions to local actors/decision makers for sub-system-level distributed simulation-optimization of adaptation decisions that represent the socio-economic systems in System Diagram. This year, we accomplished the following list of sub-tasks. System of systems model: Co-PI Reimer finalized development of a new model that captures additional inter-dependencies of the Food-Energy-Water (FEW) system. This model is called WEST (for Water Economy Simulation Tool) and shows how jobs and economic growth interact with surface and groundwater use, food, and energy. A number of scenarios for the study region were developed to illustrate the insights to be gained from the approach and to provide a template for readers to develop their own WEST simulation model, including likely data sources. Sub-system models: In the third year, PI Babbar-Sebens, along with postdoc Dr. Rivera and graduate student Mr. Giles, and undergraduate Ms. Demitrah Mauga (REU Site participant) continued to develop the sub-system models that quantify actor level decisions and impacts in the systems diagram. Specific tasks included calibration and sensitivity analysis of the Soil and Water Assessment Tool (SWAT) model to allow a more spatially explicit representation of actors with different water rights, farmland level crop management operations and regulatory constraints (e.g., stream water quality). To create the SWAT models, a new methodology that was developed last year and uses historical records of crop satellite data (i.e., USDA Cropland Data Layer) was further refined in order to improve the spatial resolution of models and therefore its ability to simulate decisions from the sub-basin to the farmland levels. Additional discretization at the farmland scale was achieved by using spatially explicit water rights' data provided by the Oregon Water Resources Department. Lastly, a database relating land use to the water rights data was developed to allow for a more realistic simulation of scenarios involving either the loss of existing water rights or the acquisition of additional water rights. 3. Simulation-Optimization Framework: In addition to the sub-system modeling efforts, PI Babbar-Sebens and postdoc Dr. Rivera advanced the optimization framework that will be used in this system-of-system simulation optimization approach. Using the concepts and theories of multi-disciplinary design analysis and optimization, actor level decisions, objectives and constraints are coupled with regional objectives and regulations to identify adaptation plans attractive to all actors while improving the overall state of the regional system. Current efforts included the identification and the development of the mathematical equations needed to represent decision of the created scenarios (see Section Research Objective #1) at both, the regional model (see System of systems model section) and the sub-system levels (i.e., actor levels), and the equations needed to couple said decisions (i.e., impact of actor level decisions should match the system level aggregates). Research Objective #3: In the third year, our team and Science Gateway staff focused on refining use cases, developing and testing workflow components/functionalities in InterACTWEL Science Gateway (ScG), and developing and testing source code for MDO for benchmark adaptation pathway problem formulations. Specific tasks included: 1) development of dashboard for typical InterACTWEL user as well as GUI for visualization and evaluation of local adaptation plans in the food-energy-water nexus space, 2) refinement and implementation of use cases for the GUIs used by project managers (global admins) and sectoral leads (local admins) in the ScG to manage different types of users, and data and models of case study region, and 3) continued development and testing of workflows that support the seamless setup and deployment of the simulation-optimization approach given the case study data (i.e., referred to as "Experiments" in the ScG). Research Objective #4: co-PI Mukhopadhyay and his graduate students continued to investigate machine learning approaches to developing surrogate user models for FEW actors, so that real-time learning of user preferences is possible when FEW actors utilize InterACTWEL for adaptation planning. Further, they have also begun an investigation of a solution to dynamic optimization problems, such as those encountered in the FEW sectors during dynamic adaptation planning, using the Multi-Disciplinary Optimization (MDO) approach. Research Objective #5: Co-PI Durresi and his graduate students ( Yefeng Ruan, Lina Alfantoukh, Suleyman Uslu, and Davinder Kaur, developed and tested an experimental trust management system for the InterACTWEL cyberinfrastructure. The trust management system uses two metrics - trustworthiness and confidence - to track the nature of stakeholder interactions in a DSS used for community-wide collaboration and coordination. We used a game-theoretical approach to investigate the interactions among stakeholders in a decision making trust pressure to achieve various tradeoffs among features of solutions such as economic profit and environment damage. Again trust, developed during decision making was used as a pressure factor to select solutions based on tradeoff desirable by the community of stakeholders. Most importantly, we showed that by deploying human in the loop based on trust evaluations our system could include factors beyond the optimization data. Outreach Objective #6: Initiated partially in Objective #1. We initiated preliminary testing in Year 3, and will continue with the process to complete testing end of Year 4 of this project. Education Objective #7: The results of this project are being used to enhance graduate courses at both institutions. This year we recruited two undergraduates to test the board game for the region that was developed in the second year.

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Babbar-Sebens, M., Rivera, S.J., Giles, N.A., Tilt, J., Reimer, J., Murthy, G., Mukhopadhyay, S., Durresi, A., Marru, S., and Pierce, M.E., Stakeholder-Driven Adaptation Planning of Food-Energy-Water Nexus in Local Communities, American Geophysical Union Fall Meeting, Dec 9-13, San Francisco, CA, 2019, (invited presentation).
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Rivera, S.J., Giles, N.A., Tilt, J., Reimer, J., Murthy, G., Mukhopadhyay, S., Durresi, A., Marru, S., Pierce, M.E. and M. Babbar-Sebens, InterACTWEL Science Gateway for Adaptation Planning in Food-Energy-Water Sectors of Local Communities: Data, Methods, Lessons Learned and Future Directions,American Geophysical Union Fall Meeting, Dec 9-13, San Francisco, CA, 2019.
  • Type: Other Status: Accepted Year Published: 2019 Citation: Babbar-Sebens, M., InterACTWEL Science Gateway for Adaptation Planning in Food-Energy-Water Sectors of Local Communities, Cascadia Lifelines Program Webinar Series, Oct 23rd, 2019, (invited webinar presentation).
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Babbar-Sebens, M., and Tilt J., InterACTWEL: A New Decision Support Tool for Adaptation Planning in the Food, Energy, and Water Sectors, 13th Annual US - China Eco-environmental Symposium, Research and Innovation at the Nexus of Food, Energy, and Water Systems, Seattle, Washington, USA, October 26-29, 2019, (invited presentation).
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Babbar-Sebens, M., A New Decision Support Tool for Adaptation Planning in the Food, Energy and Water Sectors, Second Annual Western Groundwater Congress, Sacramento, CA, September 17-19, 2019, (invited presentation).
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Tilt, J. InterACTWEL: A Stakeholder Driven Approach to Food, Energy, and Water Nexus Adaptation & Decision-Making. High-Level Panel on Food-Energy-Water Systems Transdisciplinary Environmental Research Network (FEWSTERN). University of Tennessee, Knoxville, TN, April 7-9 2019, (invited presentation).
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Babbar-Sebens, M., Rivera, S.J., Abeysinghe, E., Marru, S., Pierce, M., Coulter, E., Farahani, M., Wannipurage, D., and Christie, M., InterACTWEL Science Gateway for Adaptation Planning in Food-Energy-Water Sectors of Local Communities, Proceedings of Practice and Experience in Advanced Research Computing (PEARC) 2019, Chicago Illinois, July 28 - August 1, 2019.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Hoblitzell, A., Babbar-Sebens, M. and Mukhopadhyay, S., 2018, December. Uncertainty-Based Deep Learning Networks for Limited Data Wetland User Models. In 2018 IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR) (pp. 19-26). IEEE.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Mukhopadhyay, S., Tilak, O. and Chakrabarti, S., 2018, December. Reinforcement Learning Algorithms for Uncertain, Dynamic, Zero-Sum Games. In 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA) (pp. 48-54). IEEE.
  • Type: Journal Articles Status: Submitted Year Published: 2019 Citation: Reimer, J. Babbar-Sebens, M., Rivera, S. WEST: Water Economy Simulation Tool to Predict Impacts of Economic and Environmental Shocks. Submitted to Advances in Water Res.
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Rivera, S.J., Giles, N.A., Reimer, J., Murthy, G., Mukhopadhyay, S., Durresi, A., Tilt, J., Babbar-Sebens, M., and Marru, S., InterACTWEL: A Decision Support System for Adaptation Planning in Food-Energy-Water Sectors of Local Communities, 2019 Northwest Climate Conference, Portland, Oregon, October 8-10, 2019
  • Type: Journal Articles Status: Awaiting Publication Year Published: 2019 Citation: Nguyen, T., Mukhopadhyay, S. and Babbar-Sebens, M., 2019. Why the selfish optimizing agents could solve the decentralized reinforcement learning problems. AI Communications, (Preprint), pp.1-17.
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Reimer, J. Water Economy Simulation Tool. Western Agricultural Economics Association meetings, Coeur dAlene, Idaho, June 2019.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Suleyman Uslu, Davinder Kaur, Samuel J Rivera, Arjan Durresi, Meghna Babbar-Sebens, (2020) Decision Support System Using Trust Planning Among Food-Energy-Water Actors. In: L. Barolli, M. Takizawa, F. Xhafa, T. Enokido (eds) Advanced Information Networking and Applications. AINA 2019. Advances in Intelligent Systems and Computing, vol 926. Springer, Cham (pp1169-1180)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Davinder Kaur, Suleyman Uslu, Arjan Durresi, (2019) Trust-Based Security Mechanism for Detecting Clusters of Fake Users in Social Networks. In: L. Barolli, M. Takizawa, F. Xhafa, T. Enokido (eds) Web, Artificial Intelligence and Network Applications. WAINA 2019. Advances in Intelligent Systems and Computing, vol 927. Springer, Cham (pp. 641-650)
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Yefeng Ruan, Arjan Durresi, A trust management framework for clouds, Journal of Computer Communications, Volume 144, 15 August 2019, (pp. 124-131)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Suleyman Uslu, Davinder Kaur, Samuel J Rivera, Arjan Durresi, Meghna Babbar-Sebens, Trust-Based Game-Theoretical Decision Making for Food-Energy-Water Management, In: L. Barolli, P. Hellinckx, T. Enokido (eds) Advances on Broad-Band Wireless Computing, Communication and Applications. BWCCA 2019. Lecture Notes in Networks and Systems, vol 97. Springer, Cham(pp. 125-136)
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Murat Karakus, Arjan Durresi, An economic framework for analysis of network architectures: SDN and MPLS cases, Journal of Network and Computer Applications, Volume 136, 15 June 2019, Pages 132-146
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Murat Karakus, Arjan Durresi, Economic analysis of software defined networking (sdn) under various network failure scenarios, ICC 2019-2019 IEEE International Conference on Communications (ICC), 20-24 May 2019, Pages 1-6


Progress 01/01/18 to 12/31/18

Outputs
Target Audience:In this second year, 1/1/2018 - 12/31/2018, we continued to focus outreach to a wide range of audience who are direct as well as indirect beneficiaries and stakeholders of this project. These include: 1. Project personnel: Over the second year, this project has involved two postdocs (Dr. Hossein Tabatabaie and Dr. Sammy Rivera), 8 graduate students (Mr. Nicholas Giles, Mr. Majid Farahani, Mr. Chris Cody, Mr. Shreyansh Mohnot, Ms. Lina Alfantoukh, Mr. Yefeng Ruan, Mr. Suleyman Uslu, and Ms. Davinder Kaur), one hourly student (Ms. Holly Mondo). The students come from different demographics as well social backgrounds, including 7 of them being international students. Similar to first year, this year we continued to focus on advising students on how their own disciplinary perspective and body of knowledge intersects with those of other disciplines participating in this research. Facilitation of interdisciplinary advising was accomplished via the collaborative learning environment that has been created for students in this project. Students were regularly grouped in smaller teams to help them effectively learn from each other and execute interdisciplinary ideas and concepts into individual sub-tasks of the project. Students routinely delivered presentations on their own domain topics and their preliminary work during regular project meetings. The presentations were followed by brainstorming discussions within the entire team on how the content of the student's presentation overlaps with and influence the novel interdisciplinary ideas, methods, and body of knowledge being developed in this project. We also conducted "hackathons" every 3 months to allow students and faculty to work collaboratively on solving pending tasks that required everyone's input. The postdocs, Dr. Hossein Tabatabaie and Dr. Sammy Rivera, were also actively mentored for taking leadership in some of the sub-tasks of this project and assisting other students in specific aspects of the model development. Dr. Rivera had the opportunity to serve as task lead in the development of the simulation-optimization models/framework while mentoring graduate students, mentor high school students during a summer internship and serve as the lead on the cyberinfrastructure development efforts and several "hackathons". Because of his role in such efforts he was involved in the development of research proposal, software development plans, project management tasks, and the writing and presentation of scientific findings. All these activities contributed to his continued training and growth as an academic and will ultimately better prepare him as he pursues a tenure-track faculty position. 2. Food-Energy-Water Stakeholders: In the second year, we continued engagement with stakeholders in the food, energy, and water sectors at the study site (Hermiston, Oregon area) and in the larger Columbia River basin region. Details on this engagement are provided later under Objective 1 discussion on accomplishments in this report. In summary, these stakeholders included private as well as public entities: Calpine Corp, City of Hermiston, Columbia River Inter-Tribal Fish Commission, Confederated Tribes of the Umatilla Indian Reservation (CTUIR), Oregon Department of Environmental Quality (ODEQ), Hermiston Agricultural Research and Extension Center, Hermiston Foods, LLC, Hermiston Generating Company/Perennial Power, Hermiston Irrigation District, IRZ Consulting, Madison Farms, Morrow County Commissioner, Morrow County Soil and Water Conservation District, NORPAC, Northeast Oregon Water Association (NOWA), Northwest Food Processors Association, Northwest Natural, Port of Morrow, Renewable Northwest, Sage Center, Umatilla County, Umatilla Electric Cooperative, US Army Corps of Engineers, US Bureau of Reclamation. We used multiple methods to engage with the stakeholders, which included one-on-one interviews, field trips and meetings with individual stakeholders, and small group workshops. These engagements are planned to be continued throughout the project. A stakeholder advisory committee is currently being formed to help advice the researchers and the students throughout the project, and after the project is completed. 3. Cyberinfrastructure and Science Gateway Community: The decision support system being developed in this project (called InterACTWEL or Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land; http://interactwel.oregonstate.edu) will be supported as an application-specific Science Gateway (http://interactwel.scigap.org/). A Science Gateway (ScG) is a "community-developed set of tools, applications, and data that are integrated via a portal or a suite of applications, usually in a graphical user interface, that is further customized to meet the needs of a specific community" (source: https://www.xsede.org/web/site/ecosystem/science-gateways). Supported via an effort by National Science Foundation and multiple universities, the experienced and professional staff from the Science Gateway Research Center (SGRI) are assisting our team in developing the cyberinfrastructure for InterACTWEL using state-of-the-art software and hardware resources, including high-performance computation resources (e.g., Jetstream), workflow tools, general or domain-specific analytic and visualization software, collaborative interfaces, job submission tools, and education modules. Over the second year, we continued to closely interact with the staff at SGRI and at XSEDE (eXtreme Science and Engineering Discovery Environment) at Indiana University to develop cyberinfrastructure components of InterACTWEL. We also obtained a startup allocation with SGRI and XSEDE to develop a prototype of InterACTWEL over the next one year. The SGRI is also helping develop graphical user interfaces that will be used to communicate adaptation plans requiring coordination among food, energy, and water sectors to end-users of the DSS. 4. Scientific Community: We continued to disseminate our work to the research community via multiple conference presentations and submissions (see details in the later sections of the report). We also participated in the NSF and NSFC sponsored China-US 2018 Joint Eco-environmental Symposium: Advances in Critical Needs for the Nexus of Food, Energy, and Water systems in October 2018. We have developed new partnerships that advance ongoing work in this project in India and Puerto Rico on the the topic of food-energy-water nexus. Presentations on the project were also presented at the NSF PI meeting in May 2018. 5. K-12: In the second year, we also recruited two high school students via the Apprenticeships in Science & Engineering (ASE) program supported by Saturday Academy (https://www.saturdayacademy.org/ase). This program recruits inspired high school students for science, technology, engineering and math (STEM) internships in industry and universities. The students worked over the summer with our team to develop a cooperative and strategic board game on the food, energy, and water systems and decision makers in the Umatilla region, the study site of this project. The students were mentored by the postdoc Dr. Rivera, graduate student Mr. Giles, with overarching guidance from PI Dr. Babbar-Sebens. The game developed in this project will be used by the team to engage stakeholders in future planned activities, as well as new K-12 activities. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has involved two postdocs (Dr. Hossein Tabatabaie and Dr. Sammy Rivera), 8 graduate students (Mr. Nicholas Giles, Mr. Majid Farahani, Mr. Chris Cody, Mr. Shreyansh Mohnot,Mr. Yefeng Ruan, Ms. Lina Alfantoukh, Mr. Suleyman Uslu, and Ms. Davinder Kaur Ruan, one hourly student (Ms. Holly Mondo), and 2 high school students via summer research internships. Post-doctoral scholars involved in the project have actively mentored participating graduate students, undergraduate students and high school students, lead efforts in sub-tasks of Objectives #1, #2, and #3, lead efforts on team manuscripts, and participating in opportunities for career development. Specific to Dr. Rivera, some of these career development opportunities included: 1) a software development "hackathon" for which he visited the SGRC facilities located in Indianapolis and worked closely with SGRC and XSEDE staff, 2) attendance of two conferences (i.e., Digital One-Water Smart Utility of the Future, Arlington, VA & American Geophysical Union, Washington, D.C.) to both learn more about current research efforts within the FEW nexus and present the project progress, and 3) serve as a panelist in the Oregon State University Summer Undergraduate Research Fellowship Program. Via regular group presentations during project meetings (which occur once every two weeks for the entire group, and once every alternating week for sub-teams focused on specific objectives), all students receive feedback on how to improve their approach to design, analysis, discussion, and presentation. Students enhanced their research skills, such as searching for new materials (papers, books, websites), studying research papers, discussing them in class, presenting, summarizing and writing their new knowledge in the form of a research paper. Students were also engaged in assisting the faculty members and the Science gateway staff in the cyberinfrastructure development process. Students also participated in conference presentations, as outlined in the list of publications. ? How have the results been disseminated to communities of interest?For the public, we have developed the following: 1) Public website for information on the project: http://interactwel.oregonstate.edu 2) The science gateway webportal was developed (http://interactwel.scigap.org)for research community's access to InterACTWEL's software modules and components, and high performance computing infrastructure. 3) Video of the InterACTWEL project was developed by an undergraduate student Mr. Finley King, and will be soon posted on the InterACTWEL website by January 1, 2019. 4) Board game that simulates InterACTWEL decision making space and provides educational and learning resources for future stakeholder activities and support pedagogy in K-12 and university courses focused on topic of adaptation planning and resilience in FEW sectors. Furthermore, individual investigators also disseminated research to discipline-specific research communities. 1. PI Babbar-Sebens and her team presented the work via oral presentations and posters at multiple conferences in 2018, including 9th International Congress on Environmental Modelling and Software (June 24-28, Fort Collins, CO), and American Geophysical Union Fall Meeting (Dec 10-14, Washington DC). Babbar-Sebens also participated in panels related to Food-Energy-Water nexus at ASCE EWRI conference (June 3-7, Minneapolis, MN) and Northwest Climate Change Conference (Oct 9-11, Boise, Idaho). Finally, Babbar-Sebens also created a new task committee at ASCE EWRI on "Civil Engineering Perspectives On F-E-W Nexus" (https://www.asce.org/templates/membership-communities-committee-detail.aspx?committeeid=000011507811). List of papers and manuscripts are listed in detail in section on Publications. 2. Co-PI Tilt: Dr. Tilt presented the FEW stakeholder engagement process and results at the NSF and NSFC sponsored China-US 2018 Joint Eco-environmental Symposium: Advances in Critical Needs for the Nexus of Food, Energy, and Water systems in October 2018. Dr. Tilt also attended the NSF PI meeting in May 2018. A manuscript based on the stakeholder engagement processed developed for the Umatilla Region Food, Energy, Water Nexus is currently being prepared. 3. Co-PI Durresi: Durresi and his students published three papers, one validating the trust model we are using in this project with real data from stock market, another validating trust on Internet of Things scenarios, and two papers describing how our trust model is used in our decision support system. List of papers and manuscripts are listed in detail in section on Publications. 4. Co-PI Mukhopadhyay: Mukhopadhyay published two papers, one on interactive machine learning for user modeling and the other on Computational game Theory, in technical conferences. Further, another paper on developing an Multi-Disciplinary Optimization (MDO) approach for reinforcement-based machine learning systems, has been submitted for peer review at the AI Communications Journal. List of conference presentations, papers, and manuscripts are listed in detail in section on Publications. 5. Co-PI Murthy and postdoc Tabatabaie presented their work at two different conferences, details of which are listed in the section on Publications. ? What do you plan to do during the next reporting period to accomplish the goals?Research Objective #1: Assess and contextualize relevant actors' perceptions of the adaptation problem at the nexus of FEW sectors. Develop, disseminate, and analyze a FEW stakeholder survey to further validate the InterACTWEL systems diagram Conduct a series of small workshops with the Advisory Council to test out out the GUI-interface of the DSS. Work with the project team and Advisory Council to ensure user-needed adjustments are made to the interface. Research Objective #2: Formulate quantitative descriptions of scenarios of decisions, constraints, criteria, and likely perturbations for adaptation problems relevant to FEW sub-systems. Complete the FEW computational modeling framework and validate it using the historical data sets. Conduct sensitivity analysis to identify interdependencies among decisions in FEW sectors in Umatilla region's local communities, for a set of perturbations scenarios. Use survey from Objective #1 to validate findings and assumptions based on stakeholder feedback. Develop manuscript on problem formulation and sensitivity analysis. Research Objective #3: Create and evaluate a human computation-based MDO framework in InterACTWEL for development of robust adaptation alternatives. Complete development of graphical user interfaces to visualize the adaptation planning problem in local FEW systems of a community, and conduct usability studies with end-users. Test the MDO approach with benchmarks and FEW case study, and submit a manuscript for peer-review publication. Integrate a test adaptation planning problem formulation into the MDO, and test performance. Complete development of initial prototype of InterACTWEL Science Gateway with software components integrated with each other, and conduct workshops to obtain feedback on evaluation of the prototype by stakeholders. Research Objective #4: Create computational learning agents, which are able to discern and learn the interacting FEW actors' preferences, local knowledge, and behavior revealed in the cyber space, and then test approaches for integrating agents as surrogate users in the distributed optimization algorithms. Develop and test neural network surrogate models for various InteractWEL sub-systems to help in the MDO-based design process Develop a multi-level modeling approach for user modeling to deal with the "small data" problem Develop an iterative, interactive machine learning approach for user modeling in a targeted manner using design of experiments principles Develop methodologies for incorporation of the user models in the integrated design process for adaptation planning in interconnected food-energy-water sectors. Research Objective #5: Create and evaluate a web-based, trust and security management system for InterACTWEL. Under this objective, over the next one year, we plan to continue to work on the our trust-based multi stakeholders decision making system for InterACTWEL. Specifically, we will study how to better map the trust between pairs of users in the decision making process. Then we will investigate several approaches on how to integrate the historical values of trust between pair in the global trust of each user. Furthermore, we will work on better integration of trust in InterACTWEL's MDO design process. One of the sub-objectives is to study the behavior and effectiveness of our trust based system when used by various types of stakeholders. For this reason, we also plan to develop a simulator for FEW actors that mimics various users' behaviors, based on the findings from workshops and interviews conducted in Objective 1. Our goal is to incentivize cooperation between users in deciding the best solution for the scenario. Outreach Objective #6: Implement and evaluate the proposed InterACTWEL DSS in crowdsourcing the design of community-preferred adaptation pathways for the FEW systems at testbed site in Hermiston, OR. Conduct surveys to validate assumptions in Objective #2 Conduct workshops to perform usability analysis of InterACTWEL GUI. Education Objective #7: Advance the scientific literacy of students at multiple levels on the topic of environmental change and adaptation planning. Engage undergraduate in Research Experience for Undergraduates in summer research on this project.

Impacts
What was accomplished under these goals? Research Objective #1: This year we continued efforts led by co-PI Tilt on engaging with stakeholders in Hermiston, OR and nearby communities in the Umatilla River Basin (URB) to help them understand the overall goal of this DSS that is being developed for local communities. The feedback from stakeholder engagement was used to then develop a systems diagram (see Figure 1) that communicates how a perturbation can lead to ripple effects of decisions in a local FEW community because of the interdependencies among natural systems, socio-economic systems, and societal drivers. The system diagram developed by the community was then used to guide the research efforts of Objective 2 (i.e., development of scientific models, data gathering and the formulation of optimization problem). See Figure 1 uploaded here: https://drive.google.com/open?id=1s3jN5Eg9bt9xkieNnup09bUxXAOVsmtJ Research Objective #2: With InterACTWEL, it is envisioned that community leaders, citizens, and government agencies will be able to understand the impacts of threats to the food, energy, and water sectors before they happen, and be prepared for them with a "menu" of diverse options for response actions. These response actions would consist of a suite of management actions (or, decision alternatives) and their pathways (i.e., sequence of actions over time). This year, we accomplished the following list of sub-tasks that will finally create the overall computational approach for using the systems diagram (Figure 1) to solve for decision pathways in InterACTWEL. 1. System of systems model: Co-PI Murthy and postdoc Tabatabaie extended the quantitative linear time-invariant model developed last year to a linear time-variant model. A quantitative, input-output (IO) framework commonly used to study interdependencies of various sectors in Economics was developed for the regional scale analysis of the FEW sectors. Additionally, a detailed life cycle assessments of the FEW subsectors was conducted in OpenLCA using Ecoinvent 3.1 database to assess the environmental impacts. Co-PI Reimer progressed in the development of a theoretical economic model that translates changes in water or energy scenarios to changes in activity across different economic sectors (e.g., food processing, manufacturing, construction and mining, education). Lastly, PI Babbar-Sebens and postdoc Rivera worked on identifying potential models that could be used for the planning of regional energy portfolios. Projections developed by the Northwest Power and Conservation Council for future energy demands and energy portfolios under different economic incentives and technology innovations will be used to guide our modeling efforts. 2. Sub-system models: PI Babbar-Sebens, along with postdoc Rivera and graduate student Giles led the development of the sub-system models used represent actor level decisions. Specific tasks included: 1) updating the Soil and Water Assessment Tool (SWAT) model to allow a more spatially explicit representation of actors with different water rights, farmland level crop management operations and regulatory constraints (e.g., stream water quality), and 2) the identification of variables, and associated values, used to represent the potential actions of different actors. Current efforts include the calibration of the SWAT models using streamflow, stream temperature, crop yields, and various environmental outputs. 3. Simulation-Optimization Framework: In addition to the sub-system modeling efforts, PI Babbar-Sebens and postdoc Rivera advanced the optimization framework in InterACTWEL. Using the concepts and theories of multi-disciplinary design analysis and optimization, actor level decisions, objectives and constraints are coupled with regional objectives and regulations to identify adaptation plans attractive to all actors while improving the overall the state of the regional system. Additionally, in collaboration with Co-PI Murthy, postdoc Tabatabaie, co-PI Mukhopadhyay, and student Mohnot, the current framework is being adapted to use a dynamic planning approach that considers the growing uncertainty of future predictions (i.e., 15 to 20-year planning periods). Current challenges associated with said efforts include the development state transitions equations for the regional scale model and the coupling of different time scales used for decision variables between the different models. Research Objective #3: In the second year, PI Babbar-Sebens, postdoc Rivera, graduate students (Giles, Farahani, Mohnot), and Science Gateway staff focused on developing use cases, developing and testing workflow components/functionalities in InterACTWEL Science Gateway (ScG), and developing and testing source code for MDO for benchmark adaptation pathway problem formulations. Specific tasks included: 1) development of use case associated with the GUI used for visualization and evaluation of adaptation plans, 2) use cases for the GUIs used by project managers and sectoral leads in the ScG to manage different types of users, and data and models of case study region, and 3) development and testing of workflows that support the seamless setup and deployment of the simulation-optimization approach given the case study data (i.e., referred to as "Experiments" in the ScG). Current efforts are focused on benchmarking the simulation-optimization approach and developing the back-end that will support the developed GUIs within the ScG (see Figures 2). SeeFigure 2 uploaded here: https://drive.google.com/open?id=1UfgfxJ4HVKYa_H5OAqCBncHSYsOVk9Qp Research Objective #4: co-PI Mukhopadhyay and his graduate students continued to investigate machine learning approaches to developing surrogate user models for FEW actors, so that real-time learning of user preferences is possible when FEW actors utilize InterACTWEL for adaptation planning. Further, they have also begun an investigation of a solution to dynamic optimization problems, such as those encountered in the FEW sectors during dynamic adaptation planning, using the Multi-Disciplinary Optimization (MDO) approach. Research Objective #5: Co-PI Durresi and his graduate students (Ruan, Alfantoukh, Uslu, and Kaur) developed and tested experimental trust management system for the InterACTWEL cyberinfrastructure. We used a game theoretical approach to investigate the interactions among stakeholders in a decision making trust pressure to achieve various tradeoffs among features of solutions such as fairness and individual gains. Furthermore, an initial approach for Multi-stakeholder Decision Making based on Trust was also developed. Outreach Objective #6: Initiated partially in Objective #1. Full testing with stakeholders will occur in Year 3 of this project. Education Objective #7: The results of this project are being used to enhance graduate courses at both institutions. This year we also recruited two high school students to develop a board game that simulates the complexities of the FEW nexus and the InterACTWEL decision making environment. The objective of the board game (See Figure 3) were two-fold: (1) create an educational tool that could be used to teach a wide range of audience about the complexities of the FEW nexus and its different actors, and (2) to help engage stakeholders as well as students in adaptation planning exercises and capture negotiation strategies and trust concerns amongst intra- and inter-sectoral actors. The students also presented their game to an undergraduate class on the topic of Sustainability, and tested their game via participation of undergraduate students. Initial results of testing the board game in an undergraduate level course were also disseminated by the high school students in a project report and a symposium poster presentation. See Figure 3 ongame board and scoring card by clicking on weblinks https://drive.google.com/open?id=1mNjZpm5kpUleQs46l20wjzGbk10YMQCb and https://drive.google.com/open?id=1eMt_ZP2Yj2erZS_tUcMQ6wSCIxlLZWFq

Publications

  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Tilt, J.H., Babbar-Sebens, M., Murthy, G., Reimer, J., Mukhopadhyvay, S., Durresi, A. (2018). iFEWCoordNET: A secure decision support system for coordination of adaptation planning among FEW Actors in the Pacific Northwest. Innovations at the Nexus of Food, Energy, and Water Systems (INFEWS): 2018 Principal Investigators Workshop. National Science Foundation, Alexandria, VA. May 16-18, 2018.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Tilt, J.H., Babbar-Sebens, M., Giles, N., Mondo, H., M., Murthy, G., Reimer, J. (2018). Lessons from the field: Creating a collaborative, stakeholder-driven approach to food, energy and water adaptation and resilience planning. Joint Eco-environmental Symposium: Advances in Critical Needs for the Nexus of Food, Energy, and Water system. National Science Foundation and National Science Foundation of China. Yixing, Jiangsu Province, China. October 24-28, 2018.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: N. A. Giles, M. Babbar-Sebens, Samuel J. Rivera, Hossein Tabatabai, Jeff Reimer, Ganti Murthy, Snehasis Mukhopadhyay, Arjan Durresi, Jenna Tilt. A Secure Decision Support System for Coordination of Adaptation Planning Among FEW Actors. Presented at: American Geophysical Union 2018 Fall Meeting, Washington D.C., 14 December, 2018.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: N. A. Giles, M. Babbar-Sebens, J. Tilt, H. Mondo. Towards a Stakeholder-driven Planning Approach for Adaptation and Resilience in Food, Energy, and Water Sectors. Presented at: 9th International Congress on Environmental Modeling and Software (iEMSs), Fort Collins, Colorado, 27th June 2018.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: N. A. Giles, A Multidisciplinary Design Optimization Approach for Developing Integrated Adaptation Pathways for Food, Energy, and Water Sectors Susceptible to Drought, Presented at: Oregon State University College of Engineering Graduate Research showcase, Corvallis, Oregon, 8th February, 2018.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Tabatabaie, S.M.H., and Murthy, G.S. 2018. Development and application of a quantitative framework for food-energy-water nexus in the Pacific Northwest, USA. ASABE Abstract No. 1800228. ASABE, St. Joseph, MI.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Tabatabaie, S.M.H., and Murthy, G.S. 2018. Developing an integrated model for food-energy-water nexus in the Pacific Northwest, USA. iEMSs. 24-28 June, 2018. Fort Collins, CO.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: A. Hoblitzell, M. Babbar_sebens, and S. Mukhopadhyay. Uncertainty-Based Deep Learning Networks for Limited Data Wetland User Models. Accepted and published in Proceedings of the IEEE International Conference on Artificial Intelligence and Virtual Reality (IEEE AIVR), 2018.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: S. Mukhopadhyay, O. Tilak, and S. Chakrabarti. Reinforcement Learning Algorithms for Uncertain, Dynamic, Zero-Sum Games. Accepted and published in the Proceedings of IEEE International Conference on Machine Learning and Applications (IEEE ICMLA), 2018.
  • Type: Journal Articles Status: Under Review Year Published: 2018 Citation: T. Nguyen, M. Babbar-Sebens, and S. Mukhopadhyay. Why the Selfish MDO Agents Could Solve the Decentralized Reinforcement Learning Problems, Submitted (Under Review), AI Communications, 2018
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Yefeng Ruan, Arjan Durresi, Suleyman Uslu, Trust Assessment for Internet of Things in Multi-access Edge Computing, Published in: 2018 IEEE 32nd International Conference on Advanced Information Networking and Applications (AINA), 16-18 May 2018, Krakow, Poland
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Suleyman Uslu, Yefeng Ruan, Arjan Durresi, Trust-Based Decision Support System for Planning Among Food-Energy-Water Actors, In: Barolli L., Javaid N., Ikeda M., Takizawa M. (eds) Complex, Intelligent, and Software Intensive Systems. CISIS 2018. Advances in Intelligent Systems and Computing, vol 772. Springer, Cham.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Lina Alfantoukh,; Yefeng Ruan, Arjan Durresi, Multi-Stakeholder Consensus Decision-Making Framework Based on Trust: A Generic Framework, Published in: 2018 IEEE 4th International Conference on Collaboration and Internet Computing (CIC)
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Todd Meng, Adrian Cerny, N. A. Giles, Samuel J. Rivera, M. Babbar-Sebens. Umatilla Basin Board Game. Presented at: Apprenticeships in Science and Engineering (ASE) Symposium, Portland, OR, August, 2018.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Adrian Cerny, Todd Meng, N. A. Giles, Samuel J. Rivera, M. Babbar-Sebens. Umatilla Basin Resource Management Game. Presented at: Apprenticeships in Science and Engineering (ASE) Symposium, Portland, OR, August, 2018.
  • Type: Journal Articles Status: Under Review Year Published: 2018 Citation: Chen, D. and Babbar-Sebens, M. (in review) Coupling causal-loop of system dynamics and centrality of network analysis to identify key elements for water-energy-food nexus, Journal of Water Resources Planning and Management.
  • Type: Websites Status: Published Year Published: 2018 Citation: InterACTWEL project website: http://interactwel.oregonstate.edu InterACTWEL Science Gateway website: http://interactwel.scigap.org/


Progress 01/01/17 to 12/31/17

Outputs
Target Audience:In this first year, we focused on outreach to a wide range of audience who are direct as well as indirect beneficiaries and stakeholders of this project. These include: 1) Project personnel: Over the first year, 1/1/2017 - 12/31/2017, this project has involved one postdoc (Mr. Hossein Tabatabaie), four graduate students (Mr. Nicholas Giles, Mr. Chris Cody, Ms. Lina Alfantoukh, and Mr. Yefeng Ruan), two hourly students (Mr. Amir Javaheri and Ms. Holly Mondo), and two undergraduate students (Mr. Nicholas Giles (before he started graduate school in Fall 2017) and Ben Rietmann (an OSU honors college student)). The students come from different ethnicities as well social backgrounds, four of them being international students. These students also have varied technical backgrounds including Social Science, Computer Science, Economics, and Engineering. We also recruited Mr. Patrick MacQuarrie as an outreach coordinator and consultant to assist with stakeholder engagement in Objective 1. Since the focus of the first year has been to advise students on how their own disciplinary perspective and body of knowledge intersects with those of other disciplines participating in this research, a collaborative learning environment has been created for them. Students have been grouped in smaller teams to help them effectively learn from each other and execute interdisciplinary ideas and concepts into individual sub-tasks of the project. Students routinely delivered presentations on their own domain topics and their preliminary work during regular project meetings. The presentations are followed by brainstorming discussions within the entire team on how the content of the student's presentation overlaps with and influence the novel interdisciplinary ideas, methods, and body of knowledge being developed in this project. The postdoc, who is working with Co-PI Murthy Ganti, is also being actively mentored for taking leadership in some of the sub-tasks of this project and assisting other students in specific aspects of the model development. In addition to graduate students and postdoc, we also involved two undergraduates in the first year via undergraduate research- Mr. Nicholas Giles was recruited as a Summer Undergraduate Research Fellow (http://cce.oregonstate.edu/surf2017) in the summer of 2017 to work with PI Babbar-Sebens on development of a conceptual and quantitative approach for integrated design of adaptation plans for local community of Hermiston, OR in the Umatilla River Basin (within the larger Columbia River Basin), and Mr. Ben Rietmann, an OSU honors college student majoring in agricultural business management, who is being advised by co-PI Jeffrey Reimer on development of economic models for the project for his honors thesis. 2)Stakeholders: During this first year, we extensively focused on engagement with stakeholders in the food, energy, and water sectors at the study site (Hermiston, Oregon area) and in the larger Columbia River basin region. Details on this engagement are provided later under Objective 1 discussion on accomplishments in this report. In summary, these stakeholders included private as well as public entities: Calpine Corp, City of Hermiston, Columbia River Inter-Tribal Fish Commission, Confederated Tribes of the Umatilla Indian Reservation (CTUIR), Oregon Department of Environmental Quality (ODEQ), Hermiston Agricultural Research and Extension Center, Hermiston Foods, LLC, Hermiston Generating Company/Perennial Power, Hermiston Irrigation District, IRZ Consulting, Madison Farms, Morrow County Commissioner, Morrow County Soil and Water Conservation District, NORPAC, Northeast Oregon Water Association (NOWA), Northwest Food Processors Association, Northwest Natural, Port of Morrow, Renewable Northwest, Sage Center, Umatilla County, Umatilla Electric Cooperative, US Army Corps of Engineers, US Bureau of Reclamation. We have used multiple methods to engage with the stakeholders, which includes one-on-one interviews, field trips and meetings with individual stakeholders, and small group workshops. These engagements are planned to be continued throughout the project. A stakeholder advisory committee is also being created to help advice the researchers and the students throughout the project, and after the project is completed. 3) Cyberinfrastructure and Science Gateway Community: The decision support being developed in this project (called InterACTWEL orInteractiveAdaptation andCollaborationTool for managingWater,Energy andLand) will be supported as an application-specific Science Gateway(https://sciencegateways.org/). A Science Gateway is a "community-developed set of tools, applications, and data that are integrated via a portal or a suite of applications, usually in a graphical user interface, that is further customized to meet the needs of a specific community" (https://www.xsede.org/web/site/ecosystem/science-gateways). Supported via an effort by National Science Foundation and multiple universities, the experienced and professional Science Gateway staff are available to researchers in helping them develop cyberinfrastructure for applications in non-computer science domains. In this manner, discipline-specific applications are able to have access to a wide variety of state-of-the-art software and hardware resources, including high-performance computation resources, workflow tools, general or domain-specific analytic and visualization software, collaborative interfaces, job submission tools, and education modules. Over the first year, we closely interacted with the staff at SGRI (Science Gateway Research Institute) and at XSEDE (eXtreme Science and Engineering Discovery Environment) at Indiana University to develop a plan for the cyberinfrastructure for InterACTWEL. We are currently in the process of applying for a startup allocation with SGRI and XSEDE to develop a prototype of InterACTWEL over the next one year. We are also working with the Center for Applied Systems and Software (CASS) at Oregon State University to help develop graphical user interfaces that will form the front end of the DSS. 4) Scientific Community:Besides conference presentations and submissions (see details in the later sections of the report), we also reached out to teams of other INFEWS projects, including two of which are also based in the Columbia River Basin. We also helped organize and participate in the two Tri-State Food Energy Water Workshops supported by Oregon State University, Washington State University, and University of Idaho on April 10-11 and on October 23-24, 2017. Presentations on the project were also presented at the NSF PI meeting in March 2017. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project has involved one postdoc (Mr. Hossein Tabatabaie), four graduate students (Mr. Nicholas Giles, Mr. Chris Cody, Ms. Lina Alfantoukh, and Mr. Yefeng Ruan), two hourly students (Mr. Amir Javaheri and Ms. Holly Mondo), and two undergraduate students (Mr. Nicholas Giles (before he started graduate school in Fall 2017) and Ben Rietmann (an OSU honors college student)). Post-doctoral scholar involved in the project has been provided mentoring opportunities to participating graduate and undergraduate students, lead one of the sub-tasks of Objective #2, lead efforts on a team manuscript the describes the systems model being used in this research, and participating in opportunities for career development. Via regular group presentations during project meetings (which occur once every two weeks for the entire group, and once every alternating week for sub-teams focused on specific objectives), all students are being trained and receive feedback on how to improve their approach to design, analysis, discussion, and presentation. Students enhanced their research skills, such as searching for new materials (papers, books, websites), studying research papers, discussing them in class, presenting, summarizing and writing their new knowledge in the form of a research paper. Students are also engaged in assisting the faculty members and the Science gateway staff in the cyberinfrastructure development process. How have the results been disseminated to communities of interest?The information gathered through the in-depth interviews, as well as background data and information about the study site (e.g. demographics, agricultural economy, water usage, etc.) were presented at an invited presentation at the Stockholm Resilience Center, in Stockholm, Sweden in June, 2017. Meeting summaries from the scenario meetings were sent to all meeting participants. Results have been and are also planned to be soon presented to peers at conferences (Two abstracts will be submitted to the ASABE annual international meeting) and at stakeholder meetings. Some of the presented work include that by co-PI Mukhopadhyay at 2017 IEEE International Conference on Machine Learning and Applications (ICMLA), and also by co-PI Durresi's team that presented and published one conference paper in the Advances on Broad-Band Wireless Computing, Communication and Applications, BWCCA 2017. Furthermore, co-PI Durresi has collaborated to organize the following international activities, where our work is discussed: Co-Chair of BalkanCom 2017, First International Balkan Conference on Communications and Networking, Tirana, Albania, May 30-June 2, 2017 Co-Chair of The Tenth International Workshop on Bio and Intelligent Computing (BICom-2017), Tamkang Universit, Taipei, Taiwan, March 27 - 29, 2017. In conjunction with the 31-st International Conference on Advanced Information Networking and Applications (AINA-2017) Co-Chair of The 13-th International Workshop on Heterogeneous Wireless Networks (HWISE-2017), Tamkang Universit, Taipei, Taiwan, March 27 - 29, 2017. In conjunction with the 31-st International Conference on Advanced Information Networking and Applications (AINA-2017) Co-Chair of The 8th International Workshop on Trustworthy Computing and Security (TwCSec-2017), August 24-26, 2017, Ryerson University, Canada. In Conjunction with NBiS 2017 International Conference. Co-Chair of The 11-th International Workshop on Advances in Information Security (WAIS-2016), July 10-12, 2017, Citta della Politecnica, Torino, Italy. In Conjunction with the 11-th International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing (IMIS-2017) Besides conference presentations and submissions (see details in the later sections of the report), we also reached out to teams of other INFEWS projects, including two of which are also based in the Columbia River Basin. We also helped organize and participate in the two Tri-State Food Energy Water Workshops supported by Oregon State University, Washington State University, and University of Idaho on April 10-11 and on October 23-24, 2017. Presentations on the project were also presented at the NSF PI meeting in March 2017. In addition, the team is also working towards journal papers that are specific to their scientific community, and an overarching journal paper that is focused towards the larger FEW community. What do you plan to do during the next reporting period to accomplish the goals?Research Objective #1: Assess and contextualize relevant actors' perceptions of the adaptation problem at the nexus of FEW sectors. The stakeholder and outreach/engagement team will continue to solicit feedback on the FEW Nexus scenarios with stakeholders. As research Objective # 2 team refines and quantifies the scenario, the outreach team will develop an online scenario to help assign appropriate weights to variables in that model. The team will convene an "InterACTWEL Advisory Council" to work with the research team to field test the first version of the InterACTWEL tool. The results from the scenario meetings will be analyzed and developed into a manuscript that will be sent to a peer-review publication for peer-review. Research Objective #2: Formulate quantitative descriptions of scenarios of decisions, constraints, criteria, and likely perturbations for adaptation problems relevant to FEW sub-systems. Complete the FEW modeling framework and validate it using the historical data sets. Prepare scenarios for communication and validation during the second stakeholder meeting. Research Objective #3: Create and evaluate a human computation-based MDO framework in InterACTWEL for development of robust adaptation alternatives. Develop graphical user interfaces to visualize the adaptation planning problem illustrated earlier in Figure 2. Test the MDO approach with benchmarks, and submit a manuscript for peer-review publication. Integrate a test adaptation planning problem formulation into the MDO, and test performance. Develop initial prototype of InterACTWEL Science Gateway with software components integrated with each other. Research Objective #4: Create computational learning agents, which are able to discern and learn the interacting FEW actors' preferences, local knowledge, and behavior revealed in the cyber space, and then test approaches for integrating agents as surrogate users in the distributed optimization algorithms. Develop and test neural network surrogate models for various InteractWEL sub-systems to help in the MDO-based design process Develop a multi-level modeling approach for user modeling to deal with the "small data" problem Develop an iterative, interactive machine learning approach for user modeling in a targeted manner using design of experiments principles Develop methodologies for incorporation of the user models in the integrated design process for adaptation planning in interconnected food-energy-water sectors. Research Objective #5: Create and evaluate a web-based, trust and security management system for InterACTWEL. Under this objective, over the next one year, we plan to continue to work on the our trust-based multi stakeholders decision making system for InterACTWEL. Specifically, we will study how to better map the trust between pairs of users in the decision making process. Then we will investigate several approaches on how to integrate the historical values of trust between pair in the global trust of each user. Furthermore, we will work on better integration of trust in InterACTWEL's MDO design process. One of the sub-objectives is to study the behavior and effectiveness of our trust based system when used by various types of stakeholders. For this reason, we also plan to develop a simulator for FEW actors that mimics various users' behaviors, based on the findings from workshops and interviews conducted in Objective 1. Our goal is to incentivize cooperation between users in deciding the best solution for the scenario. Outreach Objective #6: Implement and evaluate the proposed InterACTWEL DSS in crowdsourcing the design of community-preferred adaptation pathways for the FEW systems at testbed site in Hermiston, OR. Conduct workshop in 2018 to test the prototype Education Objective #7: Advance the scientific literacy of students at multiple levels on the topic of environmental change and adaptation planning. Engage K-12 and undergraduates via Research Experience for Undergraduates in summer research.

Impacts
What was accomplished under these goals? Please note that in the proposal this DSS was named as iFEWCoordNet, and it has now been renamed to InterACTWEL (Interactive Adaptation and Collaboration Tool for managing Water, Energy and Land) for facilitating an easier understanding of the DSS's scope by the stakeholder community. In summary, the first year was spent on Recruitment of project personnel Multiple team discussions and brainstorming on research objectives and tasks, in order to ensure that all members of the research team have a clear understanding of their individual roles and the overall research project vision and plan. Development of smaller sub-teams for individual research objectives, and detailed plan of tasks along with their milestones. Initiation of research tasks. Initiate partnerships with stakeholder community, end users, and the Science Gateway personnel and CASS staff to help develop plans for use cases (i.e. description of how users will perform tasks in InterACTWEL), and DSS's software architecture. Accomplishments under specific objectives include: 1) Research Objective #1: Once developed, InterACTWEL is envisioned to empower land, water, energy managers and food producers to coordinate their efforts in identifying long-term management of resources, and for adaptation to threats that they do not have control of. Initial effort in this specific research objective focused on engaging with stakeholders and end users in Hermiston, OR and nearby communities in the Umatilla River Basin (URB) to help them understand the overall goal of this DSS that is being developed for local communities. We then partnered with them to help co-identify the different types of threats that could impair the water, energy, and land resources that these stakeholders (i.e. actors) heavily depend on, and the consequential impact on services provided by the food, energy, and water (FEW) sectors. Co-PI Tilt, who is the stakeholder and outreach/engagement lead for the project, conducted in-depth interviews with seventeen stakeholders that could be identified as "FEW stakeholders" because their scope of work or land management choices inherently involve tradeoffs that affect food, energy and water management. In-depth interviews included questions to understand the current decision-making space of actors ("What factors do you consider when making decisions regarding land, water and energy management?"); pressures or changes to the decision-making landscape ("What potential changes/threat/pressures worry you the most as you make food, energy and water decisions?"); adaptation strategies ("How have you coped, managed, or adapted to changes in the past?"); sources of credible information and need for additional information ("What types of information do you find to be 'credible' when making decisions" and "What kinds information would you like to have to help make your decisions?") and, finally, potential for collaborative adaptation ("Describe the level of detail you feel comfortable sharing information about your decision-making processes? Does this change according to specific groups or people?"). Data from these interview questions were coded for thematic elements related to trade-off choices between food (e.g. land), water and energy choices, external constraints and threats on the FEW system and existing collaborations and future collaborations. This information was then developed into a series of "storylines" that illustrated the complexities of food, energy and water management and tradeoffs in the study area. These storylines are being used as a foundation to inform tasks under Research Objective #2, #3, #4, and #5. The stakeholder and outreach/engagement team worked with Research Objective #2 team to develop these storylines into future scenarios complete with a series a feasible actions that food, energy, and water actors may take under conditions of chronic or abrupt changes. A series of four meetings were conducted, one for each major FEW actor group: 1) Energy providers; 2) Large grower operations (>10,000 acres) and food processors; 3) Regional land-use planners; 4) small grower operators and state environmental regulators. A total of 21 people participated in these meetings. In these meetings, we asked specific questions about the scenarios including: "What groups could be impacted by this scenario?", "What potential actions seem feasible to you?", "What other actions would you take?", "What are the trade-offs between potential actions?", "What impacts (e.g. indicators/metrics) concern you when considering these potential actions?". 2) Research Objective #2: With InterACTWEL, it is envisioned that community leaders, citizens, and government agencies will be able to understand the impacts of threats to the food, energy, and water sectors before they happen, and be prepared for them with a "menu" of options for response actions. These response actions would consist of a suite of management actions (or, decision alternatives) and their pathways (i.e., sequence of actions over time). For InterACTWEL to be useful for adaptation planning in local communities, FEW actors would want to identify and envision reliable (fail-safe), resilient (safe-to-fail), and sustainable (cater to social, economic, and environmental consequences) alternatives for managing water, energy, and land affected by a range of probable threats of concern. Hence, we used the feedback from stakeholders in Objective 1 to identify scenarios of threats (i.e., chronic threats, sudden threats, external threats, and internal threats) at the study site, and document the potential adaptation alternatives and their impacts, and interactions among actors, decisions, and impacts. As a first step, led by PI Babbar-Sebens, all team members co-developed a conceptual model of the adaptation planning problem, including a URB watershed model based on the Soil and Water Assessment Tool. Co-PI Murthy is working on a linear time-invariant model of the FEW nexus consisting of nine food, six energy, and three water sectors/sources. This model is being developed based on the input-output paradigm that is commonly used in economics to analyze country wide sectoral dependencies, and identify quantitative formulations for the various outcome indicators and resiliency scores for FEW actors. Co-PI Reimer initiated the development of a theoretical cropping decision model for Northwest Oregon. The model is being developed to allow for commodities including: jumbo onions, red onions, potatoes, vegetables, etc. He has also been compiling data regarding the incentives and constraints facing agricultural producers in the region. These include data regarding gross net revenue per year, water needs, yields, output prices, labor costs, electricity costs, and other input costs. 3) Research Objective #3: In the first year, PI Babbar-Sebens and her students have focused on developing use cases, DSS components/functionalities, reviewing the core source code for MDO for adaptation pathway problem formulation. 4) Research Objective #4: co-PI Mukhopadhyay and his team are investigating machine learning approaches to developing surrogate user models for FEW actors, so that real-time learning of user preferences is possible when FEW actors utilize InterACTWEL for adaptation planning. 5) Research Objective #5: This objective is being led by co-PI Durresi. Durresi's previous work for trust management system is being extended in the InterACTWEL cyberinfrastructure. The trust management system uses two metrics - trustworthiness and confidence - to track the nature of stakeholder interactions in a DSS used for community-wide collaboration and coordination. An initial approach for Multi-stakeholder Decision Making based on Trust was also developed in the first year. 6) Outreach Objective #6: Initiated partially in Objective #1. 7) Education Objective #7: The results of this project are used to enhance graduate courses at both institutions.

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2017 Citation: Thanh Nguyen and Snehasis Mukhopadhyay (2017). Multidisciplinary Optimization in Decentralized Reinforcement Learning. Accepted for publication, 2017 IEEE International Conference on Machine Learning and Applications (ICMLA).
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Lina Alfantoukh, Yefeng Ruan, Arjan Durresi, Trust-Based Multi-stakeholder Decision Making in Water Allocation System, In: Barolli L., Xhafa F., Conesa J. (eds) Advances on Broad-Band Wireless Computing, Communication and Applications. BWCCA 2017. Lecture Notes on Data Engineering and Communications Technologies, vol 12. Springer, Cham, pp 314-327|
  • Type: Conference Papers and Presentations Status: Other Year Published: 2017 Citation: Tilt, J. MacQuarrie, P., Babbar-Sebens, M. "Identifying Adaptive Capacity for local Food-Energy-Water (FEW) nexus systems within a global marketplace." Invited Presentation given to the Stockholm Resilience Center, June 15, 2017, Stockholm, Sweden.