Numerical Classification of Forest Community on Qingcheng Mountain

A series of methods including the tabular comparison of similarity coefficient, the nearest neighbor method and the groupaverage method of hierarchical agglomerative classification are used to deal with the forest communities on Qingcheng Mountain. From the results of the classification, the 11 forest plots on the Qingcheng Mountain are divided into two vegetation subtypes and five formations according to the important value in the tree layer. The most formations of evergreen broadleaved forests are in a stable stage, except a few indicated secondary characters, because deciduous trees have intruded communities under man's slight interference. If the management is strengthened, these forests will restore to the former state. Cunninghamia lanceolata forests are principally distributed the border districts of evergreen broadleaved forests that have been disturbed by human being. The trees of Cunninghamia lanceolata don't grow very well because many evergreen trees form shade, for this reason, the warm conifer forests of the region will turn to evergreen broadleaved forests.