Inferring dispersal patterns in aquatic insects from Bayesian gene flow analysis and model selection

尽管扩散是理解生物分布的一个重要的生活史特征，但人们对淡水生物的扩散速率和方式知之甚少。这对于长距离扩散（10 ~ 10公里）尤其如此，这对于（重新）殖民新的或重新储存的栖息地非常重要。与全脑无脊椎动物相比，大多数水生昆虫不仅限于溪流传播，而且在携带翅膀的成虫阶段也有可能积极地在陆地上传播。本研究选择了三种不同目水生昆虫（蜉蝣目、翼翅目、毛翅目）作为模式物种，验证了在成虫远距离（约10 ~ 10公里）传播过程中，种群间通过基因流动在河流流域有效连接的假设。目前正在进行的基因流动的速率和方向将在具有良好特征的Fulda/Werra溪流系统和邻近集水区的溪流内部和溪流之间进行调查。首先，我们将利用高分辨率微卫星分析研究集水区内和集水区之间的种群遗传分化。为了区分现在和过去的基因流动，我们将使用基因型数据来检查最近的移民并将他们分配给源种群。这种方法将允许估计目前正在进行的基因流动的所有三个模式物种在我们的研究区域。此外，我们将实现历史丰度数据和遗传信息的各种模型在不同的速率和扩散模式。将使用模型选择来选择每个物种的最佳模型。收集到的三种模式物种的扩散模式（流内和跨流）和速率的信息将有助于了解水体系统中水生昆虫种群的连通性，并有助于我们理解恢复的淡水生态系统中的再定殖过程。

Although dispersal is a central life-history trait for the understanding of organismal distributions, little is known regarding dispersal rates and modes in freshwater organisms. This is especially true for long-distance dispersal (>10km) which is important for (re-)colonization of new or re-stored habitats. In contrast to hololimnic invertebrates, most aquatic insects are not solely restricted to instream dispersal, but are potentially capable of active over-land dispersal in the wing-carrying adult stage. We chose three species of aquatic insects from different orders (Ephemeroptera, Plecoptera, Trichoptera) as model species to test the hypothesis that populations of merolimnic insects are effectively connected across stream catchments by gene-flow through adult long-distance (>10km) dispersal. Rates and directions of currently ongoing gene flow will be investigated within and among streams in the well-characterized Fulda/Werra stream system and adjacent catchments. First we will investigate genetic differentiation of populations within and among catchments using high-resolution microsatellite analysis. In order to distinguish between present and past gene flow, we will use genotypic data to check for recent immigrants and to assign them to source populations. This approach will allow estimating currently ongoing gene flow for all three model species within our study region. In addition, we will implement historical abundance data and genetic information in various models differing in the rates and modes of dispersal. Best models for each species will be chosen using model-selection. Gathered information on dispersal modes (in-stream vs. across-stream) and rates of the three model species will help to understand connectivity of aquatic insect populations in lotic systems and will contribute to our understanding of recolonization processes in restored fresh-water ecosystems.