The forest cover rate of Taihang Mountain had dropped from 20~40% (8000 B.C.) to < 5%
(A.D. 1950). Compared to the forest higher up on the mountain, the forest on hilly land of Taihang
Mt. has been seriously destroyed. We investigated 24 quadrats (20×20 m) in JiYuan City, which
covers typical secondary forests types of southern Taihang Mt. We used a principal components
analysis (PCA) to analyze environmental data and Hellinger pre-transformed species data, and used
the Kaiser-Guttman criterion to decide how many axes are worth representing and displaying on
the basis of the amount of variance explained. Two biplots were produced in scaling 1 (optimal display
of quadrats) and scaling 2 (optimal display of variables). All 13 variables were sorted by their
contributions: elevation > electric conductivity > soil pH > bulk density > pore space > total N >
organic C > topography > moisture content > aspect > slope > soil depth > gravel volume percent.
Ward’s minimum variance clustering results showed that vegetation types varied along elevation
and soil nutrient gradients. Gleditsia microphylla, Quercus baronii, Cotinus coggygria, Quercus
variabilis, and Koelreuteria paniculata were 5 species with the largest contributions among all 31
tree species to community variety. This result agrees with reality as these 5 species were dominant
species in 17 quadrats. The species composition has been largely affected by the topographical
gradient, and has been less affected by the nutrient gradient.