Scale-dependent effects of landscape pattern on plant diversity in Hunshandak Sandland.

The influence of landscape pattern on plant diversity has strong scale-dependent effects. However, the relationship is still unclear for sandy land, which covers more than one-third of the world's land mass. Aiming at exploring such scale-dependent effects in sandy land, we conducted a case study in Hunshandak Sandland, northern China. Principal component analysis (PCA) and Redundancy Analysis (RDA) was used to disentangle the relations between landscape pattern and alpha and beta plant diversity. Our results show that landscape pattern has an important influence on plant diversity, however, there existed scale effects. Landscape diversity enhanced the alpha diversity, conversely, reduced the beta diversity for all scales. PSSD (Patch Size Standard Deviation) positively related with alpha diversity whilst negatively related with beta diversity on moderate and large spatial scales, same as LPI (Largest Patch Index) on moderate scales. Shape complexity of patches can slightly increase both alpha and beta diversity at large scales. The adjustment of landscape pattern based on different spatial scales can enhance plant diversity. It is useful to improve plant diversity conservation in sandy land. [ABSTRACT FROM AUTHOR]