trophiCH v1 - a food web for Switzerland

Environmental pressures on species can cascade within food webs and even extend beyond individual ecosystems to interconnected systems at large spatial scales. To facilitate the exploration of these dynamics, we constructed trophiCH: a data-based national trophic meta-food web (henceforth the metaweb), that includes vertebrates, invertebrates, and vascular plants within Switzerland's national boundaries, and drawing from literature published between 1862 and 2024. Our comprehensive dataset catalogues 1,112,073 trophic interactions involving 23,151 species and 125 feeding guilds (e.g., detritivores, fungivores, etc). Interactions were primarily documented at the species level from the literature. Additional species-level interactions were inferred by resolving coarser taxonomic records (e.g., using “species A feeds on genus B” to infer interactions between species A and species within genus B) using taxonomic and habitat co-occurrence information.
We provide seven datasets: 1) the metaweb, 2) the taxa checklist, 3) the data source meta-dataset, 4) the list of generalist basal and predator families and polylectic species with citations and 5) a dataset with citations for the inferences of missing predators, 6) a dataset with citations for the parallel inference of diets from similar species and 7) a list of other existing metawebs. The empirical dataset is 00_empirical_metaweb.csv. To reproduce the final metaweb (01_final_metaweb.csv), start the R Project file: trophich.Rproj. We provide five scripts, accompanying functions, and the raw data required to run these scripts to reproduce the taxonomic expansion and validation of the datasets. In the first script (01_inferring_interactions.R), we infer interactions using genus and family level interactions and for basal feeding groups (see Methods: Taxonomic expansion). In the second script (02_inferring_interactions_special_cases.R), we infer further interactions for a few special cases with detailed explanations. In the third script (03_metaweb_comparisons.Rmd), we provide the statistical comparisons between our metaweb and other empirical metawebs as an R Markdown document. We additionally provide a a Python Jupyter Notebook document, outlining the error validation of the data extraction process (04_error_validation.ipynb and an accompanying .html file). Finally, we provide a script to reproduce figure 1 from the associated data paper (05_metaweb_summary_figure_1.R).
This is a first step towards a comprehensive food web for Switzerland, but data gaps remain (see Reji Chacko et al 2024, Fig.1C). We encourage addition of new datasets to this metaweb and welcome any collaborations to contruibute to future versions of this dataset. Please contact: [e-mail].

物种所承受的环境压力可在食物网内发生级联效应，甚至可突破单个生态系统的边界，蔓延至大空间尺度下的互联生态系统中。为推动此类生态动态的探索研究，我们构建了trophiCH：一个基于数据的国家级营养级元食物网（以下简称元食物网），涵盖瑞士国界内的脊椎动物、无脊椎动物与维管植物，数据来源为1862年至2024年间发表的文献。本综合数据集共收录了1112073条营养级交互关系，涉及23151个物种与125个取食功能群（如食腐动物、食真菌动物等）。此类交互关系最初主要从文献中以物种水平进行记录。其余物种水平的交互关系则通过解析更粗粒度的分类学记录（例如，基于"species A feeds on genus B"的记录，结合分类学与栖息地共现信息，推导出物种A与属B内所有物种间的交互关系）得到补充。我们共提供7个数据集：1）元食物网；2）物种分类清单；3）数据源元数据集；4）附带引用信息的广食性基位类群、捕食者类群与多食性物种列表；5）附带引用信息的缺失捕食者交互推导数据集；6）附带引用信息的基于相似物种的食性平行推导数据集；7）其他现有元食物网列表。经验数据集为00_empirical_metaweb.csv。若需复现最终元食物网（01_final_metaweb.csv），请启动R项目文件trophich.Rproj。我们还提供了5个脚本、配套函数以及运行这些脚本所需的原始数据，用于复现数据集的分类学扩展与验证流程。首个脚本（01_inferring_interactions.R）通过属、科水平的交互关系以及基位取食类群推导交互关系（详见方法部分：分类学扩展）。第二个脚本（02_inferring_interactions_special_cases.R）针对若干特殊场景推导额外交互关系，并附带详细说明。第三个脚本（03_metaweb_comparisons.Rmd）以R Markdown文档形式，提供了本元食物网与其他经验元食物网间的统计学对比分析。此外，我们还提供了一份Python Jupyter Notebook文档（04_error_validation.ipynb及配套.html文件），用于阐述数据提取流程的误差验证方法。最后，我们提供了一个脚本（05_metaweb_summary_figure_1.R），用于复现相关数据论文中的图1。本数据集仅是构建瑞士综合食物网的初步尝试，目前仍存在数据缺口（详见Reji Chacko等人2024年发表的图1C）。我们欢迎向该元食物网添加新数据集，并期待与各方合作，共同推进本数据集后续版本的完善。联系方式：[e-mail]。

搜集汇总

背景与挑战

背景概述

trophiCH v1是一个瑞士国家级的营养级元食物网数据集，涵盖脊椎动物、无脊椎动物和维管植物，基于1862年至2024年间的文献构建。该数据集包含1,112,073条营养相互作用，涉及23,151个物种和125个摄食功能群，旨在支持食物网动态和生态系统互连性研究。