Changes in species composition and diversity in the restoration process of sub-alpine dark brown coniferous forests in western Sichuan Province, China

Abstract

By adopting the concept of space as a substitute for time, we analyzed the dynamics of species composition and diversity of different restoration sequences (20, 30, 40, 50 years) in two secondary forest types in western Sichuan Province, distributed in a northerly or northwesterly direction. The analysis was based on the results of measurements of 50 plots located at elevations between 3100–3600 m. The forests originated from natural regeneration in combination with reforestation of spruce when the old-growth bamboo-dark brown coniferous forests and moss-dark brown coniferous old growth forests were harvested. Similar old-growth dark brown coniferous forests at ages ranging between 160 and 200 years were selected as the reference forests for comparisons. We recorded 167 species of vascular plants from 44 families and 117 genera. There was no significant difference in terms of the number of species among secondary forests. But the importance values of dominant species varied during the restoration processes. The dominant species in the secondary forests is Betula albo-sinensis, while Abies faxoniana is the dominant species in old-growth dark brown coniferous forests. Species richness increased significantly with restoration processes. It increased quickly in secondary forests during the period from 30 to 40 years, but decreased significantly in the old-growth dark brown coniferous forests. The species richness among growth forms decreased in the following order: herb layer > shrub layer > tree layer. The maximum value of the evenness index occurred in secondary forests at age 40 and remained relatively stable in the bamboo-birch forests, but the evenness index tended to decrease in moss-birch forests and slightly increased in the old-growth mossdark brown coniferous forests. There was a statistically significant difference in the eveness index between the tree and shrub layers as well as between the tree layer and the herb layer, but there was no significant difference between the shrub layer and the herb layer. The value of the Shannon index increased over restoration time. In bamboo-birch forests, the maximum value of the Shannon index was 3.80, recorded at age 50. In moss-birch forests, the maximal value was 3.65, reached in this forest at age 30. The value of the Shannon index of old-growth dark brown coniferous forests was recorded between younger secondary and older secondary forests. The value of the dominance index of communities varied. At the first stage of restoration, it increased, and at the end it was decreased. The dominance index of the tree layer had a similar trend as that of the community dominance index, but was more variable. The minimum value of the dominance index of the tree layer in the moss-birch forests reached 20 years earlier than that of the bamboo-birch forests. There was a significant difference among restoration sequences in the α diversity indices except for the dominance index. No significant differences between the two secondary forest types were detected. Over age, the value of the Bray-Curtis index between secondary forest and old-growth dark brown coniferous forest increased.