Correspondence Alejandro Molina-Moctezuma, Department of Wildlife, Fisheries, and Conservation Biology, University of Maine, Orono, Maine. Email: alejandro.molina@maine.edu Abstract Quantifying the downstream survival of migrating fish past dams is critical for conservation efforts. Regulators require assessments of survival as a condition of operation. Failure to meet an established survival standard may result in required operational or costly structural changes at a facility. Establishing the survival standard, as well as the rules of assessment, is a point of contention between regulators and operators. Management goals are based on biological criteria, but there are inherent statistical and probabilistic trade-offs when choosing a standard value and the method for assessment. We make a distinction between a “biological” goal (the conservation goal) and a “statistical” standard (a function of the biological goal, sample size, assessment method, and years of consecutive evaluation). An effective statistical standard maximizes true positives (passing the standard when the biological goal is being met) and true negatives (failing the standard when the goal is not being met), while minimizing false negatives and false positives. We explored the effects of sample size, true survival, and assessment methods on the probability of passing different statistical standards by simulating survival studies (simulating mark-recapture experiments). We observed a strong influence of assessment methods on the probability of making the right decision (true positive or true negative), especially when sample size, and recapture probability was low. As a support tool, we developed an interactive user interface to explore specific scenarios, and to aid communication among decision-makers.

中文翻译：

---

一种用于评估水电大坝洄游鱼类生存的生物学和统计标准的交互式决策工具

通讯作者 Alejandro Molina-Moctezuma，缅因州奥罗诺市缅因大学野生动物、渔业和保护生物学系。电子邮件：alejandro.molina@maine.edu 摘要 对流过大坝的洄游鱼类的下游生存进行量化对于保护工作至关重要。监管机构要求将生存评估作为操作条件。未能满足既定的生存标准可能会导致设施发生所需的操作或代价高昂的结构变化。建立生存标准以及评估规则是监管者和运营商之间争论的焦点。管理目标基于生物学标准，但在选择标准值和评估方法时存在固有的统计和概率权衡。我们区分了“生物学”目标（保护目标）和“统计”标准（生物学目标、样本量、评估方法和连续评估年数的函数）。一个有效的统计标准使真阳性（在达到生物学目标时通过标准）和真阴性（在未达到目标时未达到标准）最大化，同时最大限度地减少假阴性和假阳性。我们通过模拟生存研究（模拟标记重捕获实验）探讨了样本量、真实生存和评估方法对通过不同统计标准的概率的影响。我们观察到评估方法对做出正确决定（真阳性或真阴性）的可能性有很大影响，尤其是当样本大小、并且追回的概率很低。作为一种支持工具，我们开发了一个交互式用户界面来探索特定的场景，并帮助决策者之间的沟通。