COS 76-5
Understanding neighborhood effects on growth and competition in a temperate forest
Fine-scale spatial relationships are fundamental to the structure and function of forests. Neighborhood analysis provides a means to examine these relationships. In particular, it has proven to be a powerful tool to examine the nature and importance of competition on tree growth. Shifts in growth and associated competitive hierarchies are often leading indicators of change. Our goal was to determine whether the neighborhood approach helps inform community dynamics at Hubbard Brook Experimental Forest, a forest responding to novel stressors. Specifically, we conducted a neighborhood analysis to document the role of interspecific competition in regulating tree growth in three community types common in the northern forest (northern hardwood, NH; hemlock hardwood, HEM; spruce-fir, SF). We installed three large plots (6-9 ha) where all adult trees were mapped. Growth rings were collected from ≥90 target trees/species. We used likelihood methods to parameterize growth models that made different assumptions about the effect of competing neighbors: no competitive effects (i.e., the null model), species-equivalent competitive effects, and species-specific competitive effects. Given the importance of spatial scale, we allowed neighborhood size to vary. We used Akaike Information Criteria weights (AICw) to rank the strength of evidence for the alternative models.
Results/Conclusions
The radii of tree neighborhoods varied from 4 to 21m depending on the target species and community type. Of the eleven possible target species-site combinations, nine included the full model as at least one of the best models. The full neighborhood model assumes that each competing species has unique competitive effect on the target species. For six out of eleven tests, the full model was clearly the best model (AICw >0.7). For two (sugar maple in NH; yellow birch in HEM) out of eleven tests, the species equivalence models had the most support in the data (AICw >0.8). Competition is important in these forests: trees in more crowded neighborhoods grew more slowly. More importantly, the species-composition of neighborhood clearly influences growth in addition to simple crowding. Of the three late-successional dominants (sugar maple, hemlock and beech), beech appears to be the weakest competitor. Whereas sugar maple and hemlock interact strongly with each other, their interactions with beech are weak in both community types where they occur. These results suggest that beech has become a weaker competitor possibly due to the impact of beech bark disease.