全球变化知识资源中心

Individual tree growth and mortality drive forest stand dynamics and are universally important metrics of tree success. Studying factors that affect growth and mortality is particularly challenging in mixed-species, unevenaged systems due to their defining heterogeneity and strong temporal and spatial variability. The goal of this study was to determine the relative importance of individual tree attributes (i.e., species, size, neighborhood crowding, crown position) and environmental characteristics (i.e., soil moisture) in driving tree growth and survival in an uneven-aged, mixed species forest. In particular we tested if the factors regulating growth were the same as those regulating mortality, as is often assumed. Due to its large size and intensive sampling, the 3-ha, stem-mapped plot (established in 1989) at Howland Research Forest in central Maine, USA, allowed us to address additional questions regarding the influence of sapling crowding, neighbor species identity, and legacies of past disturbance. Growth and survival of over 3000 plot trees was assessed after 25 years and modeled using multiple linear regression (growth) and binary logistic regression (survival). As expected, species, neighborhood crowding, and tree diameter were top predictors of growth and survival. Specifically, growth and survival decreased with greater crowding, and increased with larger diameters. We also found that the identity of neighbors influenced focal tree growth: growth generally improved in neighborhoods comprised of species different from that of the focal. However, this general finding did not hold for all species: eastern hemlock grew better in hemlock neighborhoods, and northern white-cedar showed no response related to neighbor identities. In contrast to growth, neighborhood identity was not related to survival. Crowding from saplings did not explain any additional variability in growth; however, unexpectedly, individuals with greater sapling crowding were more likely to survive. For both growth and survival, we found an interaction between crowding and soil moisture, suggesting that within a single stand, individual success can be limited by both excess and insufficient moisture. We found no relationship between neighborhood cut stumps (legacy of past disturbance) and recent growth or survival. These results highlight the many variables driving growth and survival in uneven-aged, mixed-species forests. The top predictors for growth were identical to those for survival; however, other predictors differed in their relative importance. Given the recent emphasis on promoting uneven-aged, mixed-species forests, we suggest that studies addressing a full range of predictors of individual tree success are necessary to better manage and maintain these complex systems.