Abstract: According to the 60 quadrats data, Picea wilsonii forests in the Luya Mountain Nature Reserve were investigated by using two-way indicator species analysis (TWINSPAN) and redundancy analysis (RDA). The results showed that (1)60 quadrats were classified into 9 groups, representing 9 types of plant communities. (2)The results of RDA ordination clearly reflected the relationships between the distribution pattern of communities and environmental gradients. The distribution pattern of communities was obvious in the first axis gradient, mainly representing elevation and slope position gradients, which resulting in vertical distribution pattern of P.wilsonii forests in this reserve. (3)The statistical significance of the effect of each variable was tested by a Monte Carlo permutation test, and only elevation, slope position and soil pH were significantly related to the species data at the 5% level, indicating that there were significant collinearity effects among aspect of slope, slope and litter layer thickness. Variation partitioning showed that the independent effects of elevation, slope position and soil pH decreased successively, and elevation, slope position were the main environmental factors affecting the distribution of P.wilsonii forests.

Key words: Luya Mountain Nature Reserve, Picea wilsonii forest, TWINSPAN, RDA, variation partitioning