Spatial and environmental components of variation in the distribution patterns of subarctic plant species at Kevo, N Finland — a case study at the meso-scale level

This study presents a quantitative partitioning of the total variance in the patterns of occurrence of 231 vascular plant taxa in 362 1 × 1 km grids in the Kevo Nature Reserve into four independent components: purely spatial variation, spatially structured environmental variation, non-spatial environmental variation, and unexplained variation. This partitioning is done with (partial) constrained ordinations (canonical correspondence analysis) and associated Monte Carlo permutation tests. The numerical results suggest that most of the biological variance captured by the external explanatory variables is related to 'local' meso-scale environmental factors, as 12.6% of the variation in the species data is explained solely by the environmental variables. Part of the variance (6%) represents a spatially covarying environmental component, but only a very small part, ca 2%, is related to purely spatial variation. The amount of unexplained variation is very high (>75%). The results are compared and discussed in relation to the relative amounts of these four variance components at broader- and finer-scales and to the concepts of domains and transition zones of scales in biological patterning.