Intraspecific trait variation and neighborhood competition drive community dynamics in an old-growth spruce forest in northwest China.

Identifying the factors driving the growth and mortality of trees is important for understanding the mechanisms of forest dynamics. Here, we studied the growth and survival of trees ≥10 cm diameter at breast height (DBH) in a 15-ha temperate coniferous old growth forest plot in northwest China. We examined the relative importance of abiotic (i.e., soil nutrient and topographic) and biotic variables (i.e., tree size, competition intensity, and wood density of each individual) on the growth and mortality Picea schrenkiana, the dominant species in this forest. We found a high mortality rate and a low recruitment rate for P. schrenkiana over a period of six years. The total abundance and basal area of this species decreased, respectively. Overall, nearly 10% of P. schrenkiana individuals died. Our models of mortality had relatively low explanatory power (3% for all trees and 5% for trees <30 cm DBH), while growth models had moderate explanatory power. The growth of P. schrenkiana trees more strongly correlated with biotic factors (i.e., competition and trait) than abiotic factors (i.e., soil nutrients and topography). Overall, DBH, neighborhood crowding index (NCI), wood density (WD), and convexity explained 26% of the variation in the relative growth rate (RGR) of P. schrenkiana trees. The majority of this variation was explained by DBH alone. For trees with DBH <30 cm, DBH, NCI, WD, convexity, and slope) explained 29% of variation in RGR. In contrast, models of the absolute growth rate (AGR) of all P. schrenkiana trees only explained 3% of variation. For trees <30 cm DBH, NCI, WD, and slope explained 21% variation in AGR and the main part was explained by intraspecific variation in WD. Ultimately, our results highlight the importance of intraspecific variation in traits and competition when exploring demographic process in low-density and species-poor forests.