U.S. Geological Survey, Southwest Biological Science Center, Grand Canyon Monitoring and Research Center, 2255 N. Gemini Drive, Flagstaff, Arizona, 86001, USA.
Ecometric Research Inc., 3560 West 22nd Avenue, Vancouver, British Columbia, V6S 1J3, Canada.
Ecology. 2018 Apr;99(4):812-821. doi: 10.1002/ecy.2166. Epub 2018 Feb 21.
Introduced species are frequently implicated in declines of native species. In many cases, however, evidence linking introduced species to native declines is weak. Failure to make strong inferences regarding the role of introduced species can hamper attempts to predict population viability and delay effective management responses. For many species, mark-recapture analysis is the more rigorous form of demographic analysis. However, to our knowledge, there are no mark-recapture models that allow for joint modeling of interacting species. Here, we introduce a two-species mark-recapture population model in which the vital rates (and capture probabilities) of one species are allowed to vary in response to the abundance of the other species. We use a simulation study to explore bias and choose an approach to model selection. We then use the model to investigate species interactions between endangered humpback chub (Gila cypha) and introduced rainbow trout (Oncorhynchus mykiss) in the Colorado River between 2009 and 2016. In particular, we test hypotheses about how two environmental factors (turbidity and temperature), intraspecific density dependence, and rainbow trout abundance are related to survival, growth, and capture of juvenile humpback chub. We also project the long-term effects of different rainbow trout abundances on adult humpback chub abundances. Our simulation study suggests this approach has minimal bias under potentially challenging circumstances (i.e., low capture probabilities) that characterized our application and that model selection using indicator variables could reliably identify the true generating model even when process error was high. When the model was applied to rainbow trout and humpback chub, we identified negative relationships between rainbow trout abundance and the survival, growth, and capture probability of juvenile humpback chub. Effects on interspecific interactions on survival and capture probability were strongly supported, whereas support for the growth effect was weaker. Environmental factors were also identified to be important and in many cases stronger than interspecific interactions, and there was still substantial unexplained variation in growth and survival rates. The general approach presented here for combining mark-recapture data for two species is applicable in many other systems and could be modified to model abundance of the invader via other modeling approaches.
引入物种经常被牵连到本地物种的减少中。然而，在许多情况下，将引入物种与本地物种减少联系起来的证据是薄弱的。无法对引入物种的作用做出强有力的推断，会阻碍对种群生存力的预测，并延迟有效的管理反应。对于许多物种来说，标记-重捕分析是更严格的种群分析形式。然而，据我们所知，目前还没有允许对相互作用的物种进行联合建模的标记-重捕模型。在这里，我们引入了一个两种群标记-重捕种群模型，其中一个物种的生命参数（和捕获概率）允许根据另一个物种的丰度而变化。我们使用模拟研究来探索偏差并选择一种模型选择方法。然后，我们使用该模型来研究 2009 年至 2016 年间科罗拉多河濒危驼背大麻哈鱼（Gila cypha）和引入的虹鳟（Oncorhynchus mykiss）之间的种间相互作用。特别是，我们检验了两个环境因素（浑浊度和温度）、种内密度依赖性和虹鳟鱼丰度与幼体驼背大麻哈鱼的生存、生长和捕获之间的关系的假设。我们还预测了不同虹鳟鱼丰度对成年驼背大麻哈鱼丰度的长期影响。我们的模拟研究表明，在我们应用的具有挑战性的情况下（即低捕获概率），这种方法的偏差最小，并且即使过程误差很高，使用指示变量进行模型选择也可以可靠地识别真实的生成模型。当该模型应用于虹鳟鱼和驼背大麻哈鱼时，我们确定了虹鳟鱼丰度与幼体驼背大麻哈鱼的生存、生长和捕获概率之间的负相关关系。对生存和捕获概率的种间相互作用的影响得到了强有力的支持，而对生长效应的支持则较弱。环境因素也被确定为重要因素，在许多情况下比种间相互作用更为重要，并且生长率和存活率仍存在大量无法解释的变化。这里提出的用于合并两种物种的标记-重捕数据的一般方法适用于许多其他系统，并且可以通过其他建模方法来修改该方法，以对入侵物种的丰度进行建模。
