Capture-recapture for population modelling.

用于群体建模的捕获-再捕获。

• 批准号: 327025-2013
• 负责人: Cowen, Laura
• 金额: $ 0.8万
• 依托单位: University of Victoria
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2019
• 资助国家: 加拿大
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=674729
• 关键词: Capture; recapture; population; modelling.

My research combines the use of statistics and biology to solve population dynamics problems. In particular I employ mark-recapture experiments to estimate population parameters such as survival, abundance, and migration of populations. To do so I develop stochastic models, fit and test these models, and develop computer programs to suit the needs of other scientists and wildlife managers. I study real-world applications because my problems come from biological and sociological settings. This research is important as it tackles current applied significant problems faced by fisheries scientists, wildlife managers, ecologists, and epidemiologists. Three of my projects are as follows:1. I continue to develop methodology to deal with tag loss in mark-recapture experiments. I wish to extend this by studying dependent tag loss where individual behaviour influences tag loss. Tag loss is known to cause bias in estimates and their standard errors, and is often ignored.2. The fit of a model to the data is an important step in modelling. There are currently no goodness-of-fit methods for closed population capture-recapture models which incorporate explanatory variables; there are an infinite number of saturated models to choose from because values of the explanatory variables for individuals not captured are unknown. I will investigate development of a goodness-of-fit test using resampling methods for a two-sample closed population capture-recapture study of injection drug users.3. Fisheries tagging studies often violate the instantaneous-sampling assumption of mark-recapture models (see 2012 SSC Case Study) resulting in biased parameter estimates. I propose to eliminate the instantaneous sampling assumption by developing a continuous-time multistate model.

我的研究结合了统计和生物学来解决人群动态问题的使用。特别是我采用标记重新接收实验来估计种群参数，例如人口的生存，丰度和迁移。为此，我开发了随机模型，拟合和测试这些模型，并开发计算机程序以满足其他科学家和野生动植物经理的需求。我研究现实世界的应用，因为我的问题来自生物学和社会学环境。这项研究很重要，因为它解决了渔业科学家，野生动植物管理者，生态学家和流行病学家所面临的重大问题。我的三个项目如下：1。我继续开发方法，以应对标记 - 重新调整实验中的标签丢失。我希望通过研究单个行为影响标签丢失的依赖性标签丢失来扩展这一点。已知标签丢失会导致估计值及其标准错误，并且通常被忽略。2。模型对数据的拟合是建模的重要步骤。目前尚无封闭群体捕获征用模型的拟合优度方法，该模型包含了解释性变量。有无限数量的饱和模型可供选择，因为未捕获的个体的解释变量值未知。我将使用重采样方法研究开发拟合优质测试，以进行两样本的封闭群体捕获吸毒者的封闭群体捕获率研究。3。渔业标记研究通常违反了标记重新接收模型的瞬时采样假设（参见2012 SSC案例研究），从而导致参数估计。我建议通过开发连续的多态模型来消除瞬时采样假设。