Variation Partitioning of Species Data Matrices: Estimation and Comparison of Fractions

Abstract

Establishing relationships between species distributions and environmental characteristics is a major goal in the search for forces driving species distributions. Canonical ordinations such as redundancy analysis and canonical correspondence analysis are invaluable tools for modeling communities through environmental predictors. They provide the means for conducting direct explanatory analysis in which the association among species can be studied according to their common and unique relationships with the environmental variables and other sets of predictors of interest, such as spatial variables. Variation partitioning can then be used to test and determine the likelihood of these sets of predictors in explaining patterns in community structure. Although variation partitioning in canonical analysis is routinely used in ecological analysis, no effort has been reported in the literature to consider appropriate estimators so that comparisons between fractions or, eventually, between different canonical models are meaningful. In this paper, we show that variation partitioning as currently applied in canonical analysis is biased. We present appropriate unbiased estimators. In addition, we outline a statistical test to compare fractions in canonical analysis. The question addressed by the test is whether two fractions of variation are significantly different from each other. Such assessment provides an important step toward attaining an understanding of the factors patterning community structure. The test is shown to have correct Type I. error rates and good power for both redundancy analysis and canonical correspondence analysis.

Distribution patterns of plant communities and their associations with environmental soil factors on the eastern shore of Lake Taihu, China

“…This was consistent with the finding that soil conductivity was one of the most important factors in structuring plant communities (Peres-Neto et al 2006;Corriale, Picca, and Francescantonio 2013) and that A. donax, growing in Community I, tended to occur on clay soils. Only certain species, such as Z. latifolia and T. orientalis, were able to grow in Community III, and A-P was important in structuring this community; A-P was usually an important factor for vegetation development and distribution in wetlands and therefore it was significant in vegetation restoration in such areas (Ladányi et al 2016).…”

“…Quantifying the variation in species diversity and community composition with location provided the opportunity to infer possible mechanisms of vegetation community assembly (Kadowaki et al 2014;Reid, Reid, and Thoms 2016;Yamaji et al 2016). Similar results were found in lowlands, temperate forests, arid grasslands, beech forests, and natural forest (Peres-Neto et al 2006;Dwirek, Kauffman, and Baham 2006). We found that the main factors affecting vegetation distribution were SOM, Cond, pH, TP, and A-P. SOM provides nutrients to plants, ensures plant production and development, and plays an important role in the sustainable development of the wetland.…”

“…The nature of the relationships is shown in the ordination diagram by vectors with lengths proportional to their importance and directions showing their correlations with each axis. Soil variables were centered, standardized, and submitted to the Canoco "forward selection of explanatory variables" to avoid any artificial increase in the explained variation (Peres-Neto et al 2006). A Monte Carlo permutation test was performed to determine the accuracy of the relationship (999 randomizations) between the two data sets.…”

Vertical Distribution of Functional Potential and Active Microbial Communities in Meromictic Lake Kivu

“…We further quantified the influences of environmental variables (pH+DO+NO x +POC+CH 4 +Chl-a), depth, and season on bacterial community variation using variation partitioning analyses [46], with the varpart function of the vegan package [47,36]. If any colinearity was present, it would be revealed by the analysis.…”

A Molecular Survey of the Diversity of Microbial Communities in Different Amazonian Agricultural Model Systems

“…To investigate which soil attributes best explained the variability in ARISA profiles, a Non-metric multidimensional scaling (NMDS) was performed in Canoco 4.5 (Biometrics, Wageningen) with Hellinger transformed data [15]. This multivariate analysis was supplemented with environmental data.…”