Scaling from trees to forests: The problem of relevant detail

A central challenge in ecology is understanding the emergence of large-scale community and ecosystem patterns as the result of interactions among individuals (i.e., scaling). I identify the small-scale interactions controlling large-scale community behavior in SORTIE, a mechanistic, individual-based simulation model of forests in the northeastern United States by modifying aspects of the local interactions among model trees. The resulting forests are compared at both large and small scales to understand the changes, if any, in forest structure. Comparison of the forests predicted by these simplified simulations involve many elements of the forest including; total tree biomass, patterns of succession; the distribution of light, and the propagation of errors introduced by the simplification.; SORTIE employs a very detailed measure of local light based on a 216 point light sample around every tree. SORTIE is extremely insensitive to the amount of detail used in the calculation of the local resource, light. A one point sample of light significantly alters the local light environment, but even then the emergent forest dynamics are insensitive to these alterations due to the error-compressing nature of tree growth and mortality.; The success of simplified light simulations raise the question that local information is not needed to predict forest development in SORTIE. However, forests simulated without local spatial structure exhibit significantly reduced total biomass, accelerated successional dynamics, and often predict the wrong dominant species relative to the local model. Local resources must be tracked at a scale of 10 to 20 meters or less to adequately predict forest development.; SORTIE is very sensitive to the exact nature of species-specific functional patterns of growth, mortality, and dispersal. Attempts to simplify species with more general functional forms significantly alter model predictions. These results demonstrate the pivotal interaction between local resource heterogeneity and the functional responses of trees on SORTIE forest dynamics. Successive reformulation of model interactions and subsequent multi-scale comparison of model predictions provides a strong methodology to identify relevant detail and understand the scaling of individual interactions.