Department of Life Sciences, University of Siena, Via Mattioli 4, 53100 Siena, Italy.
CREA - Research Centre for Agriculture and Environment, Via di Lanciola 12/A, 50125 Cascine del Riccio Firenze, Italy.
Sci Total Environ. 2019 Mar 15;656:659-669. doi: 10.1016/j.scitotenv.2018.11.320. Epub 2018 Nov 22.
Soil plays a fundamental role in many ecological processes, throughout a complex network of above- and below-ground interactions. This has aroused increasing interest in the use of correlates for biodiversity assessment and has demonstrated their reliability with respect to proxies based on environmental data alone. Although co-variation of species richness and composition in forests has been discussed in the literature, only a few studies have explored these elements in forest plantations, which are generally thought to be poor in biodiversity, being aimed at timber production. Based on this premise our aims were 1) to test if cross-taxon congruence across different groups of organisms (bacteria, vascular plants, mushrooms, ectomycorrhizae, mycelium, carabids, microarthropods, nematodes) is consistent in artificial stands; 2) to evaluate the strength of relationships due to the existing environmental gradients as expressed by abiotic and biotic factors (soil, spatial-topographic, dendrometric variables). Correlations between groups were studied with Mantel and partial Mantel tests, while variance partition analysis was applied to assess the relative effect of environmental variables on the robustness of observed relationships. Significant cross-taxon congruence was observed across almost all taxonomic groups pairs. However, only bacteria/mycelium and mushrooms/mycelium correlations remained significant after removing the environmental effect, suggesting that a strong abiotic influence drives species composition. Considering variation partitioning, the results highlighted the importance of bacteria as a potential indicator: bacteria were the taxonomic group with the highest compositional variance explained by the predictors used; furthermore, they proved to be involved in the only cases where the variance attributed solely to the pure effect of biotic or abiotic predictors was significant. Remarkably, the co-dependent effect of all predictors always explained the highest portion of total variation in all dependent taxa, testifying the intricate and dynamic interplay of environmental factors and biotic interactions in explaining cross-taxon congruence in forest plantations.
土壤在许多生态过程中起着基础性作用，通过地上和地下相互作用的复杂网络来实现。这引起了人们对利用生物多样性评估相关指标的兴趣，并证明了它们与仅基于环境数据的替代指标相比具有可靠性。尽管物种丰富度和组成在森林中的共变已在文献中进行了讨论，但只有少数研究探讨了森林人工林中的这些因素，这些人工林通常被认为生物多样性较差，旨在生产木材。基于这一前提，我们的目标是：1）检验不同生物类群（细菌、维管植物、蕈类、外生菌根、菌丝体、步甲、微节肢动物、线虫）之间的跨分类群一致性在人工林是否一致；2）评估由于环境梯度（如生物和非生物因素[土壤、空间地形、树木测量变量]）导致的关系强度。通过 Mantel 和偏 Mantel 检验研究了组间的相关性，同时应用方差分解分析来评估环境变量对观测到的关系稳健性的相对影响。在几乎所有的分类群对中都观察到了显著的跨分类群一致性。然而，只有在去除环境影响后，细菌/菌丝体和蕈类/菌丝体的相关性仍然显著，这表明强烈的非生物影响驱动了物种组成。考虑到方差分解，结果突出了细菌作为潜在指标的重要性：细菌是被预测因子解释的组成方差最高的分类群；此外，它们被证明参与了仅由生物或非生物预测因子的纯效应归因于的唯一情况下。值得注意的是，所有预测因子的共同依赖效应始终解释了所有依赖类群总变异的最高部分，证明了环境因素和生物相互作用的复杂和动态相互作用在解释森林人工林中的跨分类群一致性方面的重要性。
