PDF(Sci-hub)
Abstract：
Management or conservation targets based on demographic rates should be evaluated within the context of expected population dynamics of the species of interest. Wild populations can experience stable, cyclical, or complex dynamics, therefore undisturbed populations can provide background needed to evaluate programmatic success. Many raptor species have recovered from large declines caused by environmental contaminants, making them strong candidates for ongoing efforts to understand population dynamics and ecosystem processes in response to human-caused stressors. Dynamic multistate occupancy models are a useful tool for analyzing species dynamics because they leverage the autocorrelation inherent in long-term monitoring datasets to obtain useful information about the dynamic properties of population or reproductive states. We analyzed a 23-year bald eagle monitoring dataset in a dynamic multistate occupancy modeling framework to assess long-term nest occupancy and reproduction in Lake Clark National Park and Preserve, Alaska. We also used a hierarchical generalized linear model to understand changes in nest productivity in relation to environmental factors. Nests were most likely to remain in the same nesting state between years. Most notably, successful nests were likely to remain in use (either occupied or successful) and had a very low probability of transitioning to an unoccupied state in the following year. There was no apparent trend in the proportion of nests used by eagles through time, and the probability that nests transitioned into or out of the successful state was not influenced by temperature or salmon availability. Productivity was constant over the course of the study, although warm April minimum temperatures were associated with increased chick production. Overall our results demonstrate the expected nesting dynamics of a healthy bald eagle population that is largely free of human disturbance and can be used as a baseline for the expected dynamics for recovering bald eagle populations in the contiguous 48 states.
摘要：
应在感兴趣物种的预期种群动态范围内评估基于人口比率的管理或保护目标。野生种群可以经历稳定，周期性的，或者复杂的动力学，因此，未受干扰的人群可以提供评估计划成功所需的背景。许多猛禽物种已经从环境污染物造成的大幅下降中恢复过来，使他们成为不断努力了解人口动态和生态系统过程以应对人为压力的有力候选人。动态多状态占用模型是分析物种动态的有用工具，因为它们利用长期监测数据集中固有的自相关来获取有关种群或生殖状态动态特性的有用信息。我们在动态多状态占用建模框架中分析了23年的秃鹰监测数据集，以评估克拉克湖国家公园和保护区的长期巢穴占用和繁殖，阿拉斯加我们还使用了分层广义线性模型来了解巢穴生产率与环境因素的关系。巢穴最有可能在几年之间保持在相同的筑巢状态。最值得注意的是,成功的巢穴可能会继续使用（被占用或成功），并且在第二年过渡到未占用状态的可能性很小。随着时间的推移，老鹰使用巢的比例没有明显的趋势，巢穴进入或退出成功状态的可能性不受温度或鲑鱼可用性的影响。在研究过程中，生产率是恒定的，尽管4月温暖的最低气温与小鸡产量增加有关。总体而言，我们的结果证明了健康的秃鹰种群的预期筑巢动力学，该种群在很大程度上没有人为干扰，可以用作在连续的48个州中恢复秃鹰种群的预期动力学的基线。
