Semantic enrichment and mobilization of data in distributed repositories for taxonomy and ecology: Solving a Gordian knot in the taxonomy of neotropical jumping spiders and trait-based ecological habitat assessment with spider assemblages

将通过以下研究方法提高和展示关联数据库和研究信息系统的效率和潜力：1.基于SMNK收藏库中极富属和物种的Euphryinae亚科(Saltiidae)凭证标本的数字数据，将描述该物种，并制定描述方案，并将其集成在虚拟研究环境多样性工作台中。这些属的修订将有助于克服新热带跳蛛的分类障碍，促进进一步的系统发育研究，并允许在动物群和生态学研究中进行物种鉴定，这对保护管理具有重要意义。研究基础设施使国际社会能够对标准化的形态测量和生态数据进行数字化处理和动员，产生永久更新的交互式确定关键字和半自动物种描述，直接汇编成电子出版物。2.对来自不同来源(收集、未发表的研究、文献)的德国物种出现情况的大量数据进行分析，提供了物种与生境和土地利用类型之间基于特征的联系。选择的栖息地类型的承载能力将使用蜘蛛的生物量进行比较，这些生物量是通过回归的大小测量计算出来的。利用经典的判别分析归纳方法和机器学习的传导性应用(随机神经元网络)，物种出现与生态偏好、环境变量和特征以及定义的生境特定组合相关。在数据选择和这些分析的准备过程中，将在收集和生态研究现有数据的语义丰富的概念下汇编关于德国蜘蛛的大量数字信息。通过定义带有同义词的标准和等级清单(生境类型、景观、土地利用形式、地层--欧洲尺度)，并使用工具从标本到物种进行注释和扩大衡量标准，可以对独立研究的数据进行新的分析。这是(重复)评估栖息地质量及其随时间因土地利用或气候变化而变化的前提条件。所有数据和工具将在几个数据储存库和网站中提供。由此产生的分类和生态信息描述载于物种概况说明书中。栖息地类型清单显示了物种的出现和丰富度。这些描述总是基于参与机构在增加收藏和研究期间更新的数据池。这种代表实际数据池的动态引用对来自保护当局的用户尤其重要。

The efficiency and potential of linked data repositories and research information systems will be enhanced and demonstrated by the following research approaches: 1. Based on digital data for voucher specimens of the extremely genera- and species-rich subfamily Euophryinae (Salticidae) in the collection of SMNK, the species will be described and a description scheme developed and integrated in the virtual research environment Diversity Workbench. The revision of these genera will help to overcome the taxonomic impediment in neotropical jumping spiders, promoting further phylogenetic studies and allowing species identification in faunistic and ecological studies, which is important for conservation management. The research infrastructure enables the digital processing and mobilization of standardized morphometric and ecological data for the international community, produces permanently updated interactive determination keys and semi-automatic species descriptions, directly compiled into an electronic publication. 2. The analysis of a large data package of species occurences in Germany, originating from different sources (collections, unpublished studies, literature) provides trait-based associations of species with habitat and land use types. Carrying capacity of selected habitat types will be compared using biomass of spiders calculated from size measurements via regression. Using classic inductive methods of discrimant analysis as well as transductive appraoches of machine learning (stochastic neuronal networks), species occurences are correlated with ecological preferences, environmental variables and traits and habitat-specific assemblages defined. In the course of data selection and preparation for these analyses extensive digital information on spiders in Germany will be compiled under a concept for semantic enrichment of existing data from collections and ecological studies. By definition of standards and hierarchical lists (habitat types, landscapes, land use forms, strata - on European scale) with synonyms and by the use of tools for annotation and upscaling of measures from the specimen to the species, data from independent studies can be analysed newly. This is a precondition for the (repeated) assessment of habitat quality and its change over time by land use or climate change. All data and tools will be supplied in several data repositories and websites. Resulting descriptions with taxonomic and ecological information are presented in species fact sheets. Checklists for habitat types show the species occurences and abundances. These portrayals are always based on the updated data pool, during the increase of collections and studies of the participating institutions. Such dynamic references, representing the actual data pool, are especially important to users from conservation authorities.