Diachronic analysis of individual-tree mortality in a Norway spruce stand in the eastern Italian Alps

Understanding tree mortality processes across time requires long term studies. Spatiotemporal patterns of mortality in a 200 years-old mono-layered Norway spruce stand were evaluated to determine what factors affected individual-tree mortality. We performed an analysis on two surveys (1993 and 2005) in a 1-ha permanent plot in the Paneveggio forest (Eastern Italian Alps). Tree diameter and age distribution between surveys were compared. We examined spatial patterns of living and dead trees before 1993, in 1993 and in 2005 using univariate and bivariate Ripley's K(d) function, and a kernel estimator of local crowding. A logistic model was used to assess the effects of diameter, age, recent growth and competitive pressure on tree mortality. Spatial pattern analysis indicated mortality was associated to tree neighbourhood (neighbour effect at 2-5 m). An increment of regularization of tree spatial pattern occurred due to density-dependent mortality. Logistic regression showed tree diameter and recent growth were determinant on mortality risk during the monitoring period. Even if the stand is relatively aged, mortality dynamics are those typical of stem exclusion stage. Mortality was related to competitive dynamics, and small suppressed trees with slow growth rate had higher probability to die.