Objective  The mechanisms of species abundance distribution is one of the central issues in community ecology, and there has been an increasing interest in the association between species abundance distribution and functional traits in recent years. The primeval forests in Changbai Mountains have been severely damaged for a long time, creating large areas of secondary forests, resulting in a serious loss of species diversity and a significant decrease in ecosystem service. This study aims to reveal the species abundance distribution pattern and its driving process, to provide better guidance for local forest management and restoration strategies.

Method In this study, we used observational data collected from three 5.2 ha and one 5 ha forest plots with different forest types (secondary Populus-Betula mixed forest, secondary coniferous and broadleaved mixed forest, primary Pinus koraiensis-Tilia amurensis mixed forest & primary broadleaved-Pinus koraiensis mixed forest). Six functional traits, including leaf area, specific leaf area, leaf thickness, maximum tree height, leaf nitrogen concentration, and leaf phosphorus concentration were measured to examine the distribution and functional traits. The relationship between species abundance and functional traits distribution was also assessed in this study.

Result The types of models that passed the statistical test varied among the forest types, but the statistical models had the best fit in all forest types. The maximum tree height, specific leaf area, and leaf phosphorus concentration were normal distribution in all the 4 plots. However, the species abundance distribution converted from the functional traits cannot fit the real species abundance distribution well.

Conclusion The findings suggest that random process is not the main mechanism driving the assembly of community species diversity in Changbai Mountains. Instead, competition determines the existence of common species during forest succession. There is a correlation between functional trait distribution and species abundance distribution, but community-level trait distribution cannot be applied to infer species abundance distribution directly.