Ecological Indicators (Nov 2021)
A spatially explicit index for tree species or trait diversity at neighborhood and stand level
Abstract
Knowledge about tree species interactions is important for our understanding of forest ecosystems. Research about such interactions requires a quantitative description of different stand mixture types for example with measures of species- or trait-diversity like Simpson- or Rao-diversity. Many currently available diversity measures are based on relative proportions and do not account for the spatial configuration of a stand. They implicitly assume that all trees are equally likely to interact with one another, irrespective of their spatial distance. However, fundamental ecosystem processes such as competition are strongly influenced by intra- or interspecific interactions that take place between neighboring trees.This study introduces a measure for the neighborhood diversity (NDiv) around individual trees by averaging the dissimilarity between a tree and all its neighbors. NDiv uses maps of the ‘area potentially available’ to trees (APA) to define tree neighborhoods. APA-maps are derived by spatially dividing the entire stand area into adjacent patches that approximate the growing space that is potentially available to each tree. The average NDiv of all trees of a stand weighted by the sizes of their APA-patches provides a consistent diversity measure at stand-level. Depending on the used dissimilarity measure, NDiv may either quantify species or trait neighborhood diversity. Species neighborhood diversity and tree height neighborhood diversity of case-study plots were analyzed to demonstrate the application of NDiv.When tree locations were spatially randomized in a simulation study, NDiv of the case-study plots at stand-level was nearly identical to Simpson-diversity. However, the actually observed NDiv of plots with patch-wise mixing of tree species was considerably lower than their Simpson-diversity. On these plots, NDiv accounted for a limited potential for interactions among species since the tree species were growing spatially separated in patches.NDiv is a useful measure of tree diversity when trees are not randomly distributed across a stand and when mixing effects mainly derive from interactions between neighboring trees. The use of NDiv as explanatory variable may help to account for the spatial configuration of trees in forests which may increase the explanatory power of analyses of relationships between tree diversity and ecosystem processes.
