About

Ann Bostrom is the Weyerhaeuser Endowed Professor in Environmental Policy at the Evans School of Public Policy & Governance at the University of Washington. She joined the Evans School faculty in 2007 and her research focuses on risk perception, communication, and management; environmental policy; and decision making under uncertainty. Prior to her current position, she served on the faculty at the Georgia Institute of Technology from 1992 to 2007, where she was Associate Dean for Research at the Ivan Allen College of Liberal Arts and a Professor in the School of Public Policy. Bostrom co-directed the Decision Risk and Management Science Program at the National Science Foundation from 1999 to 2001, organizing and participating in national and international meetings on research and science policy. She has authored numerous publications and serves on editorial boards for risk analysis journals. Her work has been funded by agencies such as the NSF, EPA, and NIH. Bostrom holds a Ph.D. in Public Policy Analysis from Carnegie Mellon University, an M.B.A. from Western Washington University, and a B.A. in English from the University of Washington. She has completed postdoctoral studies in Engineering and Public Policy at Carnegie Mellon University and in cognitive survey methodology at the Bureau of Labor Statistics. She is an elected fellow of the AAAS, WSAS, and the Society for Risk Analysis, and has served as past president of the Society for Risk Analysis. Currently, she serves on multiple boards and advisory committees related to environmental hazards, climate change, and risk communication, and holds various research leadership roles at the University of Washington.

Research topics

• Computer Science
• Political Science
• Geography
• Computer Security
• Environmental resource management
• Data science
• Engineering
• Business
• Economics
• World Wide Web
• Public relations
• Engineering management
• Telecommunications
• Environmental planning
• Archaeology
• Environmental science
• Economic growth
• Systems engineering
• Medicine

Selected publications

• Partner‑Engaged Workshop, in Assessing Perceptions of Extreme Cold Event Risk and Risk Reduction Strategies in King County, Washington

  Texas Advanced Computing Center · 2026-04-10

  datasetOpen access

  We convened 25 local and regional stakeholders from King County and the broader Puget Sound region for a one‑day in-person workshop on March 24, 2025, to examine local health risks associated with extreme cold. Participants included representatives from agencies that respond to extreme cold, such as emergency management and public health. During a discussion facilitated using the World Café method, workshop participants were asked to consider interrelated questions about extreme cold impacts, vulnerability, and current and potential risk reduction strategies. Each workshop table was instructed to focus on a potentially vulnerable population pre-identified based on existing literature. Notetakers at each workshop table took notes on the discussion. This data publication includes a summary of table notes. To protect participant identities, the research team has redacted most references to specific jurisdictions, agencies, and organizations. A table listing participant characteristics, including Geographical Boundary of Organization, Geography Type Served by Organization, and Organization Type, has also been shared, along with the workshop facilitation guide, and documentation of the University of Washington’s Human Subjects Division exempt determination.

  PublisherDOI

• When Every Second Counts: Parental Decision-Making in Mt Rainier’s Lahar Inundation Zone

  2026-03-14

  articleOpen access

  Mount Rainier, a heavily glaciated stratovolcano in Washington State [USA], has a documented history of producing major lahars. The potential for future high-magnitude flows threatens approximately 90,000 downstream residents and has prompted one of the nation’s most extensive volcanic monitoring systems, including a specialized lahar detection network. Because portions of Rainier’s west flank are composed of hydrothermally altered, unstable rock, the region is especially vulnerable to “no-notice” lahars triggered by sudden, non-eruptive slope failure. In response, schools in at-risk zones have conducted lahar evacuation drills – now a legal requirement – for over two decades, demonstrating that on-foot evacuation is the most effective strategy for student and staff safety. Despite these efforts, many parents report an intention to retrieve their children from school during an emergency lahar evacuation, contradicting official guidance. Such actions could obstruct evacuation routes, delay emergency response, and increase personal risk, especially in areas where modeled lahar arrival times are under one hour. Parent decision-making thus presents a critical, yet understudied, variable in evacuation planning and is considered integral to the success of city-wide evacuations.Here we present the ongoing work from focus groups held with local parents to explore motivations behind their intentions. Topics of discussion within the focus groups include parents’ general understanding of lahar hazards, their intended actions, their confidence in school evacuation plans, and underlying factors in their decision-making. These insights can support more effective communication and preparedness strategies by emergency managers and school officials, while also contributing to broader discussions about protective action decision-making in rapid-onset hazards beyond volcanic settings.

  PublisherDOI

• Communicating natural hazards science advice: Understanding scientists', decision-makers’, and the public's perceptions of the scientific process

  International Journal of Disaster Risk Reduction · 2025-07-29

  articleOpen access

  How individuals perceive scientific processes impacts their interpretation of, trust in, and use of, science advice particularly when managing uncertain natural hazard risk. We explored a) how diverse stakeholders understand how science of natural hazards is produced, and b) how this relates to their ontological, epistemological, and philosophical views of science. Using inductive analysis of semi-structured interviews with 31 participants involved in the management of natural hazards in Aotearoa New Zealand (including non-scientists), we produced three leading themes describing their views: 1) ‘Science is a way of seeing the world’; 2) ‘Science has limitations’; and 3) ‘Knowledge evolves’. Across Scientist, non-Scientist, and Lay public groups, there was broad agreement on the fundamental steps of the scientific process, aligning mostly with a hypothetico-deductive process. However, many discussed how others may have different perspectives of scientific approaches, truth, and reality. These are informed by training, disciplinary biases, cultural practices, and personal experience of hazards and associated science. We propose that individuals who recognise different worldviews and philosophies of science will experience higher levels of communication and cognitive uncertainty, which encourages information seeking behaviour and can improve communication efficacy, particularly during high pressure events. We conclude with three communication lessons: 1) be transparent about the processes and causes of change in natural hazards science advice; 2) communicate as both trusted individuals as well as through collective Science Advisory Group (SAG) systems; and 3) provide accessible structures and language to help lay people articulate scientific processes they often intuitively understand, rather than just simplifying information.

  PublisherDOI

• An Assessment of How Domain Experts Evaluate Machine Learning in Operational Meteorology

  Weather and Forecasting · 2025-02-17 · 1 citations

  article

  Abstract As an increasing number of machine learning (ML) products enter the research-to-operations (R2O) pipeline, researchers have anecdotally noted a perceived hesitancy by operational forecasters to adopt this relatively new technology. One explanation often cited in the literature is that this perceived hesitancy derives from the complex and opaque nature of ML methods. Because modern ML models are trained to solve tasks by optimizing a potentially complex combination of mathematical weights, thresholds, and nonlinear cost functions, it can be difficult to determine how these models reach a solution from their given input. However, it remains unclear to what degree a model’s transparency may influence a forecaster’s decision to use that model or if that impact differs between ML and more traditional (i.e., non-ML) methods. To address this question, a survey was offered to forecaster and researcher participants attending the 2021 NOAA Hazardous Weather Testbed (HWT) Spring Forecasting Experiment (SFE) with questions about how participants subjectively perceive and compare machine learning products to more traditionally derived products. Results from this study revealed few differences in how participants evaluated machine learning products compared to other types of guidance. However, comparing the responses between operational forecasters, researchers, and academics exposed notable differences in what factors the three groups considered to be most important for determining the operational success of a new forecast product. These results support the need for increased collaboration between the operational and research communities. Significance Statement Participants of the 2021 Hazardous Weather Testbed Spring Forecasting Experiment were surveyed to assess how machine learning products are perceived and evaluated in operational settings. The results revealed little difference in how machine learning products are evaluated compared to more traditional methods but emphasized the need for explainable product behavior and comprehensive end-user training.

  PublisherDOI

• Scientists’ mental models of microplastics: insights into expert perceptions from an exploratory comparison of research methods

  Microplastics and Nanoplastics · 2025-09-22 · 1 citations

  articleOpen access1st authorCorresponding

  Abstract Microplastics have been studied extensively, yet considerable uncertainty remains about the risks they pose. One way to characterize the state of knowledge about a hazard and the risks it poses is to examine how scientists specializing in that hazard understand and think about it. In two complementary studies our interdisciplinary team examined how microplastics scientists understand and think about the hazards of microplastics accumulation in freshwater systems, and what risks they may pose. Each study used a different approach. Study 1 studied the causal beliefs—that is, the “mental models”—scientists applied in decision contexts. It relied on a mixture of open- and closed-ended questions, and tasks during which microplastics scientists ( N = 15) were asked to think aloud. This approach revealed scientists’ causal thinking about where microplastics come from and about the health and environmental consequences of microplastics. Specifically, in Study 1 microplastics scientists emphasized household consumption as a primary source of microplastics, while acknowledging multiple direct and indirect sources and exposure pathways, and often dwelling on the uncertainties about human health consequences. Study 2 applied the M-Tool, which is a different approach to studying mental models. In Study 2 microplastics scientists ( N = 38) used the M-Tool to draw causal connections between core ideas about microplastics. Top concepts selected in this exercise included waste mismanagement, textiles, plastic degradation, individual littering, and water quality. Across both studies there were commonalities in how scientists understood the sources and exposure pathways for microplastics. Scientists emphasized household consumption of plastics as a direct and indirect source of microplastics, but there were gaps in how they talked about dose–response functions. Together the two studies portray how scientists from diverse disciplines understand the potential risks of microplastics accumulation in freshwater ecosystems. Findings suggest that microplastics risk communication and management strategies can be improved by providing a broader perspective on sources of microplastics beyond household consumption, by sharing information about diverse approaches to managing risks of microplastics, and by addressing uncertainties as well as gaps between knowledge and concerns about human health effects. The novel comparative research approach explored here demonstrates the complementarities of the methods employed, which we hope will be useful for those interested in understanding the social and decision dimensions of microplastics and other environmental problems.

  PublisherOA PDFDOI

• (Re)Conceptualizing trustworthy AI: A foundation for change

  Artificial Intelligence · 2025-02-22 · 10 citations

  articleOpen access

  Developers and academics have grown increasingly interested in developing “trustworthy” artificial intelligence (AI). However, this aim is difficult to achieve in practice, especially given trust and trustworthiness are complex, multifaceted concepts that cannot be completely guaranteed nor built entirely into an AI system. We have drawn on the breadth of trust-related literature across multiple disciplines and fields to synthesize knowledge pertaining to interpersonal trust, trust in automation, and risk and trust. Based on this review we have (re)conceptualized trustworthiness in practice as being both (a) perceptual, meaning that a user assesses whether, when, and to what extent AI model output is trustworthy, even if it has been developed in adherence to AI trustworthiness standards, and (b) context-dependent, meaning that a user's perceived trustworthiness and use of an AI model can vary based on the specifics of their situation (e.g., time-pressures for decision-making, high-stakes decisions). We provide our reconceptualization to nuance how trustworthiness is thought about, studied, and evaluated by the AI community in ways that are more aligned with past theoretical research.

  PublisherDOI

• Leveraging Coproduction to Bridge Research and Operations in Operational Meteorology

  Weather and Forecasting · 2025-05-19

  articleOpen access

  Abstract The benefits of collaboration between the research and operational communities during the research-to-operations (R2O) process have long been documented in the scientific literature. Operational forecasters have a practiced, expert insight into weather analysis and forecasting but typically lack the time and resources for formal research and development. Conversely, many researchers have the resources, theoretical knowledge, and formal experience to solve complex meteorological challenges but lack an understanding of operation procedures, needs, requirements, and authority necessary to effectively bridge the R2O gap. Collaboration then serves as the most viable strategy to further a better understanding and improved prediction of atmospheric processes via ongoing multidisciplinary knowledge transfer between the research and operational communities. However, existing R2O processes leave room for improvement when it comes to collaboration throughout a new product’s development cycle. This study assesses the subjective importance of collaboration at various stages of product development via a survey presented to participants of the 2021 Hazardous Weather Testbed Spring Forecasting Experiment. This feedback is then applied to create a proposed new R2O workflow that combines components from existing R2O procedures and modern coproduction philosophies. Significance Statement This study assesses researcher and forecaster perspectives on the importance of collaboration at each stage of a product’s development cycle. We then incorporate this feedback into a proposed new R2O workflow that combines components from existing R2O processes, idealized practitioner’s cycles, and modern coproduction philosophies.

  PublisherOA PDFDOI

• National Weather Service (NWS) Forecasters’ Perceptions of AI/ML and Its Use in Operational Forecasting

  Bulletin of the American Meteorological Society · 2024-10-10 · 4 citations

  articleOpen accessSenior author

  Abstract Artificial intelligence and machine learning (AI/ML) have attracted a great deal of attention from the atmospheric science community. The explosion of attention on AI/ML development carries implications for the operational community, prompting questions about how novel AI/ML advancements will translate from research into operations. However, the field lacks empirical evidence on how National Weather Service (NWS) forecasters, as key intended users, perceive AI/ML and its use in operational forecasting. This study addresses this crucial gap through structured interviews conducted with 29 NWS forecasters from October 2021 through July 2023 in which we explored their perceptions of AI/ML in forecasting. We found that forecasters generally prefer the term “machine learning” over “artificial intelligence” and that labeling a product as being AI/ML did not hurt perceptions of the products and made some forecasters more excited about the product. Forecasters also had a wide range of familiarity with AI/ML, and overall, they were (tentatively) open to the use of AI/ML in forecasting. We also provide examples of specific areas related to AI/ML that forecasters are excited or hopeful about and that they are concerned or worried about. One concern that was raised in several ways was that AI/ML could replace forecasters or remove them from the forecasting process. However, forecasters expressed a widespread and deep commitment to the best possible forecasts and services to uphold the agency mission using whatever tools or products that are available to assist them. Last, we note how forecasters’ perceptions evolved over the course of the study. Significance Statement Despite a range of familiarity with artificial intelligence and machine learning (AI/ML), forecasters are open to using AI/ML tools operationally. The extent of this openness ranged from being highly supportive to having some important concerns about how effective AI/ML can be and whether or not it would replace them. Although some forecasters see AI/ML products as the exciting cutting edge of science, others care little of the development approach and more about how well the product verifies and helps them do their job.

  PublisherOA PDFDOI

• Identifying and Categorizing Bias in AI/ML for Earth Sciences

  Bulletin of the American Meteorological Society · 2024-01-22 · 17 citations

  articleOpen access

  Abstract Artificial intelligence (AI) can be used to improve performance across a wide range of Earth system prediction tasks. As with any application of AI, it is important for AI to be developed in an ethical and responsible manner to minimize bias and other effects. In this work, we extend our previous work demonstrating how AI can go wrong with weather and climate applications by presenting a categorization of bias for AI in the Earth sciences. This categorization can assist AI developers to identify potential biases that can affect their model throughout the AI development life cycle. We highlight examples from a variety of Earth system prediction tasks of each category of bias. Significance Statement As artificial intelligence (AI) grows in popularity, its methods are being applied to a wide range of Earth system prediction tasks. Although AI can facilitate more accurate prediction at many tasks, it is not without potential pitfalls, especially if the developers are not as familiar with its potential drawbacks. In this paper, we provide a classification system for the types of bias that one is likely to see in applying AI to Earth sciences. Our classification system will assist current and future AI developers to recognize where their AI system or data are biased so they can take steps to alleviate this bias.

  PublisherOA PDFDOI

• Developing trustworthy AI for weather and climate

  Physics Today · 2024-01-01 · 9 citations

  articleOpen accessSenior author

  By improving the prediction, understanding, and communication of powerful events in the atmosphere and ocean, artificial intelligence can revolutionize how communities respond to climate change.

  PublisherOA PDFDOI

Recent grants

• Testing risk communication strategies designed to address cognitive simplifications that underlie factually erroneous beliefs about scientific issues

  NSF · $649k · 2020–2024

• Collaborative Research: DRU--Warning Decisions in Extreme Weather Events: An Integrated Multi-Method Approach

  NSF · $146k · 2008–2012

Frequent coauthors

• Amy McGovern

  University of Oklahoma

  39 shared

• David John Gagne

  NSF National Center for Atmospheric Research

  32 shared

• Baruch Fischhoff

  Carnegie Mellon University

  32 shared

• Imme Ebert‐Uphoff

  32 shared

• Andrea Schumacher

  29 shared

• Geoffrey Evans

  East Kent Hospitals University NHS Foundation Trust

  28 shared

• Randy J. Chase

  Colorado State University

  27 shared

• L. Ball

  Center for Drug Evaluation and Research

  26 shared

Education

• Ph.D. in Policy Analysis, H. John Heinz III College

  Carnegie Mellon University

  1990

• M.B.A., Business School

  Western Washington University

  1986

Awards & honors

• American Statistical Association/National Science Foundation…
• Fulbright Graduate Research Fellowship (1989-90)
• Lois Roth Endowment Fund grant for studies at the University…
• Patricia Roberts Harris Fellowship at Carnegie Mellon (1988-…
• 2020 Distinguished Educator award from the Society for Risk…