Abstract:

Minjiang fir (Abies faxoniana) (MJF) is a dominant tree species of sub-alpine forests on the eastern edge of Qinghai-Tibetan Plateau, and is mainly distributed over the upper reaches of the Minjiang, Dadu and Bailong Rivers. The population structure and spatial pattern of MJF was studied in a naturally occurring stand. In a 100 m×60 m plot, the location of every tree was mapped, and the diameter at breast height (DBH), height and canopy area of each individual recorded. Trees were divided into five size classes: seedlings, height <0.33 m; saplings, height ≥ 0.33 m, and DBH<2.5 cm; small trees, 2.5 cm ≤ DBH<7.5 cm; medium Trees, 7.5 cm ≤ DBH <22.5 cm; and big trees, DBH ≥ 22.5 cm. The spatial pattern of MJF was analyzed using four independent methods: the Morisita index (Iδ), variance to mean ratio (V/m), the congregation index (m*/m) and the spatial point pattern analysis (SPPA) (Ripley's second-order- analysis method). The results revealed that MJF was a stable population with an inverse J-shaped size structure indicating good natural regeneration. Seedlings and saplings were very abundant, with densities of 2 217·hm-2 and 2 683·hm-2, respectively. Irregularities in the size structure histogram reflected past disturbances. The spatial analyses revealed that seedlings, saplings and small trees were clumped at most spatial scales studied which ranged from 1 m to 30 m, whereas the medium-sized trees and big trees were randomly distributed. The intensity of assemblage (IA) varied with scale. The first three methods indicated that IA decreased with increasing scale, but the SPPA method showed that the IA of seedlings, saplings and small trees first increased with increasing scale, and then declined at greater scales. We conclude that the spatial pattern of MJF in this subalpine forest resulted from long-term interactions between the MJF and its natural environment and mechanisms of natural regeneration that vary among species. The four different methods were very similar on the whole in their abilities to discriminate spatial patterns, but SPPA was superior in its ability to detect changes of IA with scale. Thus, we recommend SPPA for analyzing spatial patterns of populations. However, a limitation to using SPPA relates to the complexity of sampling and calculation requised and some refinements in Ripley's second-order-analysis are needed in order to better as it detect gaps.