Analysing tree species and size diversity patterns in multi-species uneven-aged forests of Northern China

• Multi-species complex forests are ideal data sources for testing the sensitivity of diversity measures.
• We found a relationship between relative size and individual abundances following a saturation curve.
• Mark mingling and mark differentiation functions clearly identified more diversity details than conventional indices.
• Clusters of similar species and diameters with a radius between 5 and 15 m could be identified in the Dongdapo study sites.
• The deviation test was instrumental in asserting the results of the spatial analysis.

The diversity of complex multi-species uneven-aged forests is not easy to describe efficiently and meaningful. In the past, most importance has been assigned to species diversity and there is a wide range of approaches to quantifying this aspect of diversity. In woodlands, size diversity is another important dimension of biodiversity, as trees can be small or large and depending on their size play different ecological roles. Traditionally size diversity has often been neglected in diversity surveys of woody vegetation or considered as difficult to handle. We investigated the size and species diversity of two multi-species uneven-aged forests from monsoon- and subtropical/warm-temperate climate zones in northern China using non-spatial and spatially explicit forestry and diversity characteristics. We found that the analysis of both species and size diversity provided a much better understanding of diversity patterns in complex multi-species forests. Interestingly the relationship between size- and individual-based relative abundances follows a saturation curve. The analysis using non-spatial forestry and biodiversity characteristics led to approximately similar results in all four plots despite the very different woodland communities involved. However, with spatial measures we gained a deeper insight concerning the deviation from random distributions of species and size as well as interaction ranges of trees. We also found an aggregation of similar species and sizes at small inter-tree distances in the monsoon influenced climate and an aggregation of different sizes in the subtropical/warm-temperate climate zone. The use of second-order characteristics was instrumental in elaborating the subtle differences between the diversity of the two study areas and thus made an important contribution to this study.