RESUMEN
To explore how traits determine demographic performance is an important goal of plant community ecology in explaining the assembly and dynamics of ecological communities. However, whether the prediction of individual-level trait data is more precise compared to species average trait data is questioned. Here, we analyzed the growth and trait data for 11 species collected from October 2018 to October 2020 in a temperate forest, Donglingshan, Beijing. To quantify the relationships between traits and growth rate, we conducted linear regression models at both the species and individual levels, as well as developed structural equation models at both levels. We found there was a clear difference in growth between the warm and cold seasons, with tree growth mainly concentrated in the warm season. Growth rate was positively correlated with the specific leaf area, while negatively correlated with leaf thickness and wood density without considering environmental information. Adding important contextual information in the analysis of species-level structural equation modeling, growth rates were positively correlated with specific leaf area and leaf thickness. However, in the individual-level, there was a negative correlation between growth rate and wood density. Our study showed that individual-level trait data have better predictions for individual growth than species-level data. When we use multiple traits and establish links between traits and tree size, we generated strong predictive relationships between traits and growth rates. Furthermore, our study highlighted that the importance of incorporating topographical factors and considering different seasons to assess the relationship between tree growth and functional traits.