Several barriers have limited the implementation of climate adaptation solutions in practice over the past two decades. Notably, there are many uncertainties related to human behavior, future greenhouse gas (GHG) emissions, and the direction, timing, and severity of local climate change impacts that can make it difficult to predict how communities will be affected (Stults & Larsen, 2020). Poor communication, inadequate public involvement, and a lack of awareness of local climate risks also act as barriers to adaptation and resilience planning (Lee et al., 2022). To overcome these barriers and promote the successful implementation of climate adaptation and resilience planning in practice, new planning and community engagement tools are needed.

Exploratory Scenario Planning (XSP) is one tool that is increasingly being used to overcome many of the limitations of traditional planning approaches (Goodspeed, 2020; Haigh, 2019; Hurtado & DeAngelis, 2022). XSP is a systematic participatory process where community members work together to create multiple plausible qualitative scenarios or stories that describe what the future could look like under different climate conditions. The outcome of this process can enhance decision-making and improve a community’s understanding of the local impacts of climate change (Haigh, 2019). It can also support a participatory approach to climate adaptation planning, which can lead to the creation of more equitable adaptation and resilience plans.

Although XSP has the potential to support climate adaptation and resilience planning, the scenario planning literature offers little guidance on how to conduct community-engaged XSP (Chakraborty & McMillan, 2015; Hopkins & Zapata, 2007). This presentation will highlight the efforts of an ongoing research project led by researchers from the University at Buffalo (UB) and funded by the New York State Department of Environmental Conservation (NYS DEC). The purpose of this project is to develop resources, guidance, and tools that can support the successful implementation of XSP in communities across NYS.

One of the tools that will be discussed during the presentation includes a serious game and augmented reality (AR) experience that is being developed by UB researchers. Drivers and Debates introduce players to XSP and demonstrates how the process can be used to identify strategies to adapt to multiple plausible climate scenarios. In the first part of the game, players learn about different strategies from the New York State Climate Smart Communities (CSC) program and decide which ones should be prioritized to address extreme heat and flooding in an imaginary community called Anytown, New York. In the second part of the game, the players assume the role of a community stakeholder and discuss which selected strategies make sense to implement in one or more of the scenarios presented.

The game and other resources developed as part of this research project can be used to engage community members in the XSP for climate adaptation and resilience planning process in a more fun and accessible way. The implications and lessons learned from this research can help communities across NYS and beyond utilize XSP to anticipate and prepare for the future impacts of climate change.

References

Chakraborty, A., & McMillan, A. (2015). Scenario Planning for Urban Planners: Toward a Practitioner’s Guide. Journal of the American Planning Association, 81(1), 18–29. https://doi.org/10.1080/01944363.2015.1038576

Goodspeed, R. (2020). Scenario Planning for Cities and Regions: Managing and Envisioning Uncertain Futures. Lincoln Institute of Land Policy. https://www.lincolninst.edu/publications/books/scenario-planning-cities-regions/

Haigh, N. (2019). Scenario Planning for Climate Change: A Guide for Strategists. Routledge, Taylor & Francis Group. https://www.routledge.com/Scenario-Planning-for-Climate-Change-A-Guide-for-Strategists/Haigh/p/book/9781138498402?srsltid=AfmBOop-xnbMdOgwkAka51yyvc-LTmAY3zf8j17m7qcvIrRXjUBsnhA6

Hopkins, L. D., & Zapata, M. A. (Eds.). (2007). Engaging the future: Forecasts, scenarios, plans, and projects. Lincoln Institute of Land Policy. https://www.lincolninst.edu/publications/books/engaging-future/

Hurtado, P., & DeAngelis, J. (2022). The Use of Foresight and Scenario Planning in Hazard Mitigation and Climate Adaptation Planning. 113. https://www.planning.org/pas/memo/113/the-use-of-foresight-and-scenario-planning-in-hazard-mitigation-and-climate-adaptation-planning/#:~:text=Scenario%20Planning%20and%20Hazard%20Mitigation&text=It%20involves%20identifying%20risks%20and,the%20intention%20of%20saving%20lives.

Lee, S., Paavola, J., & Dessai, S. (2022). Towards a deeper understanding of barriers to national climate change adaptation policy: A systematic review. Climate Risk Management, 35, 100414. https://doi.org/10.1016/j.crm.2022.100414

Stults, M., & Larsen, L. (2020). Tackling Uncertainty in US Local Climate Adaptation Planning. Journal of Planning Education and Research, 40(4), 416–431. https://doi.org/10.1177/0739456X18769134

Infrastructure systems have been typically designed under the assumption of growing – or at least stable – demand. Yet, cities are experiencing population decline due to deindustrialization, suburbanization, climate migration, and other factors. Urban shrinkage challenges conventional infrastructure planning approaches. This perspective argues that to respond effectively to the uncertainty of urban shrinkage, infrastructure must shift from rigid to flexible systems. Drawing on infrastructure resilience, uncertainty, adaptation, and degrowth literature, we conceptualize flexibility as the capacity for infrastructure to render human-centered affordances (i.e., realized infrastructure services that enable human capabilities) across spatial, quantity, temporal, configurable, functional, and institutional categories. Infrastructure managers may evaluate how to ‘flex’ each of these categories to meet changing demand. We introduce these six flexibility categories, provide definitions, and use the transportation sector to illustrate examples in each category. By reframing infrastructure planning around uncertainty and considering urban shrinkage, this perspective offers a structured lens for evaluating where, how, and for whom flexibility should be embedded in infrastructure systems, with implications that extend beyond transportation to other critical infrastructures such as water and energy. Finally, the perspective's discussion distinguishes between flexibility and adaptation, explores the trade-offs of the interconnected flexibility categories, and considers equity through the lens of flexibility.

This work emerged from the Infrastructure Misfits pre-conference workshop at ISSST 2025. We believe that this perspective is relevant to ISSST participants, particularly those interested in the ‘Sustainability and resilience of infrastructure systems’ and ‘Sustainability and resilience of energy systems’ themes. To achieve impact, we believe this work is relevant to anyone who makes decisions about infrastructure assets (e.g., siting, sizing, financing, maintenance, decommissioning), including municipal planners, utilities, regulators, and private operators. To our knowledge, the existing literature does not present a comprehensive characterization of flexibility dimensions within infrastructure design, suggesting that this framework offers a novel and original contribution. The framing of flexibility within urban shrinkage is novel. The technical quality of the work results from the synthesis of infrastructure resilience, uncertainty, adaptation, and degrowth literature, demonstrating progress toward interdisciplinary research. Finally, the perspective is largely complete. The authors have written everything except the discussion, which is outlined. We suspect having a full draft of the perspective by February 2026. We request that this perspective be presented as a lightning talk at ISSST 2026.

The recent rapid development of artificial intelligence (AI) has enabled advanced solutions across a wide range of applications. However, the environmental sustainability of AI model development and use, particularly the energy use, embodied water consumption, and greenhouse gas (GHG) emissions associated with training and use large-scale models, has raised concerns. While larger machine learning models often offer improved performance, the benefits gained become marginal after a certain point. Depending on the specific application and the frequency of retraining, the environmental benefits gained from the improved performance may or may not exceed the environmental costs from using larger models. A critical question remain unanswered is that for which applications and under which conditions, using larger AI models can provide net environmental benefits. Using transportation routing optimization as a case study, we developed a framework to evaluate whether the energy and emissions reduction achieved from better optimized routes justify the additional resources required by larger AI models. The results can help inform more responsible and sustainable development and use of AI in transportation and other applications.

Generative AI (GenAI) has become an increasingly important part of everyday interactions with the digital world. In academic research and higher education, GenAI-powered tools are now used across many stages of the research process, including idea generation, literature discovery, data collection and analysis, programming, and writing. As researchers consider the adoption of GenAI tools, it is essential that they understand both the strengths and limitations of these technologies, including their sustainability implications.

To date, research has largely focused on assessing the environmental impacts of AI models themselves, while the implications of using GenAI tools to conduct research and perform literature reviews remain limited. This project aims to investigate the environmental impacts associated with using GenAI-powered literature search tools and AI-powered literature reviews.

To achieve this, electricity use and water consumption associated with performing various literature review tasks using AI are quantified, including resource use at the end users and within data centers. Life cycle inventory (LCI) data sources, including the U.S. Electricity Baseline and the USLCI databases from the Federal LCA Commons, are used to estimate direct and indirect energy consumption, greenhouse gas emissions, and water use associated with AI-supported research activities. Differences among AI platforms, such as Scopus AI, Consensus, and Keenious, as well as geographic location, are also considered.

The results will inform higher education administrators, researchers, and publication venues in the development of responsible policies and practices for the use of GenAI in research. The findings can also contribute to the development of sustainability-aware research policies and best practices that balance the benefits of GenAI with environmental responsibility.

Building heat is a significant portion of Canada's greenhouse gas (GHG) emissions. In 2022, residential and commercial buildings emitted 103 Mt CO₂e, which corresponds to 21.8% of Canada’s total emissions. Of this total, 76.8% is estimated to comes from building space heating. Natural gas is the most prevalent fuel for building heating in many regions in Canada and, therefore offers a significant opportunity for emissions reduction and advances Canada’s 2050 net-zero goal. While hybrid heating and electrification have been explored in some regions, studies at a national scale are still scarce. Regional assessments, when extrapolated nationally, often overestimate or underestimate results because of underlying data gaps and computational limits. This study provides a high-resolution national perspective that helps identify regions with the most significant emissions reduction potential. It assesses the technical potential of hybrid heating systems across 293 census divisions in Canada using a high-resolution modelling framework. The study combines building stock characteristics such as floor area, housing type, heating equipment or HVAC systems, and hourly temperature data from NASA’s MERRA-2 T2M dataset. The data is processed into an energy model that estimates hourly heating demand for each census division which are simulated under three technology setups: conventional gas furnaces, hybrid heating systems (heat pumps and existing gas furnaces), and fully electrified heat pump systems. Building and heating equipment assumptions are calibrated to provincial statistics to ensure representativeness nationally. The results are expected to identify census divisions where hybrid heating adoption achieves the most emissions reductions, specifically in cold-climate regions with dense buildings. Census divisions with warmer climates and lower building density will see smaller emissions reductions with some electrification or hybrid heating adoption. In conclusion, this study provides a national assessment of hybrid heating potential, which offers insights for utilities and policymakers to target impactful regions for emissions reduction. It also provides a better resolution on potential changes in natural gas and electricity demand under different deployment scenarios, which supports effective decarbonization planning.

Agricultural bioenergy systems are increasingly recognized as critical components of decarbonization, rural energy resilience, and circular economy strategies. In New York State (NYS), anaerobic digestion (AD) offers a promising pathway for valorizing agricultural residues, yet adoption at facility scale remains limited. A key barrier for deployment is uncertainty in feedstock availability, composition, and performance. This challenge is important for specialty crop residues, which are often treated as homogeneous inputs despite substantial chemical and temporal variability

Grapes (Vitis vinifera) contain high concentrations of soluble sugars and other fermentable organic compounds, making them ideal for biogas production systems. Approximately 80% of global grape production is used for winemaking, an industry valued at more than $6.65 billion in NYS. However, 20-30% of the total grape weight becomes waste, primarily as grape pomace, a nutrient-rich residue from pressing the grapes, but whose characteristics vary due to environmental conditions. According to Wine Folly, the U.S. is the world’s fourth-largest wine producer, and within the U.S., NYS ranks third, with most wineries distributed across eleven federally recognized American Viticultural Areas (AVAs). Despite substantial production, little research has investigated the valorization of NYS grape pomace. Moreover, this approach directly supports NYS’s 2025 Draft Energy Plan for increased renewable natural gas (RNG)/biogas from on-farm waste sources.

Prior studies have demonstrated the biogas and biomethane potential of winery residues; however, most existing work is geographically limited, conducted at a laboratory-scale, or based on snapshots of a single feedstock. As a result, variability arising from grape variety, vintage conditions, and winemaking practices is frequently ignored, limiting the applicability of results for real-world decision-making. Notably, the same chemical differences that drive wine style and regional identity could also influence anaerobic digestion performance, suggesting that variability in feedstock could directly affect waste-to-energy system design and performance.

This research establishes a system-level framework that treats grape pomace feedstock variability as a factor in agricultural bioenergy systems’ capacity. Focusing on NYS-specific characterization of grape pomace as a bioenergy feedstock and evaluating AD potential for micro-scale techno-economic analysis as it relates to on-farm operational benefits (capital and operating costs, co-product value, payback time, and net present value, etc.). Expected results include identification of key drivers of variability in bioenergy performance, improved understanding of uncertainty bounds in energy and economic outputs, and preliminary thresholds for system viability under realistic operating conditions. These results can provide the foundation for a proposed study that expands analysis across multiple years and integrates techno-economic analysis, lifecycle assessment, and artificial intelligence/machine learning -enabled modeling to develop uncertainty-aware decision-support tools. By reframing feedstock variability as a central design parameter, this work aims to bridge the gap between laboratory bioenergy research and operational deployment. While the current research includes a case study on NYS winery waste, its applicability extends well beyond, offering the potential to generate significant economic and environmental benefits for U.S. and global agricultural systems and stakeholders. Furthermore, it contributes to the advancement of national sustainable energy infrastructures, and the enhancement of rural energy security.

Sustainable food systems should support public health while minimizing environmental impacts, both of which are significantly influenced by food choice. Current food systems and consumer behaviors result predominantly in unsustainable and unhealthy diets, as documented in a wide range of existing literature. Such research efforts could be used to guide improved decision-making across consumers, policy-makers, and industry, but relevant publications lack a clear visualization that simultaneously addresses both concerns in a simple and intuitive format. Therefore, this work presents a new visualization of the environmental and health impacts of thirty commonly consumed food groups in the form of a color-coded 3x3 matrix to help inform such decisions.

Food groups are classified in the visualization matrix according to their carbon footprint along the x-axis and associated relative risk for all-cause mortality and common chronic diseases along the y-axis. Data for both components are aggregated from meta-analyses and systematic reviews, then classified as low, medium, or high carbon footprint and favorable, neutral, or unfavorable to health. The combination of these two classifications determines the arrangement of food groups on the visualization matrix.

The resulting color-coded 3x3 chart simultaneously communicates each food group’s typical environmental impacts and health implications in a single figure. Overall, the visualization indicates plant-based and less processed foods are both preferable to animal-based and more processed foods in both categories. The 3x3 format facilitates more nuanced classification than a traditional dichotomy while maintaining a simple format. For example, foods with tradeoffs between environmental impacts and health outcomes include fish (high carbon footprint, but favorable for health) and candy and snacks (low carbon footprint, but unfavorable for health). Meat substitutes are also shown to have a medium carbon footprint and neutral overall effect on health. The matrix design emphasizes the exceptionally large carbon footprint of beef by splitting the lower-right cell vertically into two halves.

Food group classifications presented in this new visualization are consistent with results presented in previous studies. However, the color-coded matrix format additionally facilitates the rapid identification of carbon footprint and expected health outcomes for thirty different commonly consumed food groups. This information should help quickly and intuitively communicate tradeoffs made when choosing between different food groups, helping improve such decisions.