Abstract

A statistical analysis was performed on hydrochemical and hydrobiological data obtained from samples of macrozoobenthos, meiobenthos, and zooplankton in 2013 and 2018 from different biotopes of the five highly mineralized rivers of the basin of the hyperhaline Lake Elton. The general trends in the formation of the species structure of the metacommunity of aquatic organisms are considered based on the Lybold-Mikkelson concepts. Randomized null models are used to test hypotheses on the consistent occurrence of species (coherence), the formation of compact species associations (boundary clumping), and the regular change in the species composition along environmental gradients (turnover). The nature of changes in biotic diversity were determined to be nonstationary and intermittent, and the taxonomic structure was highly mosaic, which is characteristic of stochastic-type ecosystems. A close correlation was established within individual mixed associations of benthic and planktonic species due to biotic interactions and the coordinated response of organisms to environmental factors. Statistical models Random Forest and subsequent multivariate analysis were used to identify groups of species by the degree of tolerance to external abiotic influences, and the sets of factors having a significant impact were established.

中文翻译：

---

埃尔顿湖盆地河流中浮游生物和底栖生物的元群落结构分析

摘要

对2013年和2018年从高盐湖Elton盆地的五个高度矿化河流的不同生物群落的大型动物，足底鱼类和浮游动物样本中获得的水化学和水生物数据进行了统计分析。基于Lybold-Mikkelson概念，考虑了水生生物元群落物种结构形成的总体趋势。随机零模型用于检验以下假设：物种一致发生（一致性），紧密物种关联的形成（边界聚集）以及物种组成沿环境梯度的规律变化（周转率））。生物多样性变化的性质被确定为非平稳和间歇性的，并且分类结构是高度镶嵌的，这是随机型生态系统的特征。由于生物相互作用和生物体对环境因素的协调反应，在底栖生物和浮游生物的个体混合协会之间建立了密切的联系。统计模型通过对外部非生物影响的耐受程度，使用随机森林和随后的多变量分析来识别物种组，并建立了具有重要影响的因素集。