Abstract The topic of uncertainty is of growing interest in the impact assessment (IA) field, due to increases in contextual uncertainty and the awareness of the complexity of advanced analysis. IA practitioners can now draw on maturing theoretical frameworks to manage uncertainty, but questions remain about whether these frameworks align with stakeholder concerns and how their use can benefit IA projects. This article reports on an empirical application of the leading framework for organizing IA uncertainty proposed by Walker et al. in 2003. Twenty-two stakeholders involved in a large water quality modeling project in the U.S. Great Lakes region were interviewed, and their uncertainty-related statements were categorized according to the Walker dimensions. Overall, the framework's three primary dimensions performed well and allowed for the analysis of differences in uncertainty perceptions among the stakeholder groups. In addition, the analysis resulted in useful insights for the project, such as identifying top scenario uncertainties to use for modeling as well as highlighting specific concerns about the assumptions, data, and modeling approach for further exploration. In addition to encompassing the variety of uncertainty concerns raised in the case, the paper illustrates how the Walker framework can support IA practices like stakeholder collaboration and scenario construction which may improve IA outcomes.

中文翻译：

---

分析利益相关者对不确定性的看法以推进协作可持续性科学：对底特律河项目营养负荷的流域评估案例研究

摘要 由于上下文不确定性的增加和对高级分析复杂性的认识，不确定性主题在影响评估 (IA) 领域越来越受到关注。IA 从业者现在可以利用成熟的理论框架来管理不确定性，但这些框架是否符合利益相关者的关注以及它们的使用如何使 IA 项目受益的问题仍然存在。本文报告了 Walker 等人提出的用于组织 IA 不确定性的领先框架的实证应用。2003 年，参与了美国五大湖地区大型水质建模项目的 22 位利益相关者接受了采访，他们的不确定性相关陈述根据 Walker 维度进行了分类。总体而言，该框架' 三个主要维度表现良好，并允许分析利益相关者群体之间不确定性认知的差异。此外，该分析为该项目提供了有用的见解，例如确定用于建模的主要情景不确定性，以及突出对假设、数据和建模方法的具体关注以供进一步探索。除了涵盖案例中提出的各种不确定性问题外，本文还说明了 Walker 框架如何支持 IA 实践，如利益相关者协作和情景构建，这可能会改善 IA 结果。例如确定用于建模的主要场景不确定性，以及突出对假设、数据和建模方法的具体关注，以供进一步探索。除了涵盖案例中提出的各种不确定性问题外，本文还说明了 Walker 框架如何支持 IA 实践，如利益相关者协作和情景构建，这可能会改善 IA 结果。例如确定用于建模的主要场景不确定性，以及突出对假设、数据和建模方法的具体关注，以供进一步探索。除了涵盖案例中提出的各种不确定性问题外，本文还说明了 Walker 框架如何支持 IA 实践，如利益相关者协作和情景构建，这可能会改善 IA 结果。