Implications of natural history traits to system-level dynamics: comparisons of a grassland and a forest

Abstract

The importance of life-history traits to vegetation pattern and system-level behavior was evaluated for two North American ecosystems, a semiarid grassland and an eastern deciduous forest. Responses of the systems to spatial and temporal variability in environmental conditions were compared using a common modeling framework, a class of individual-based simulators known as ‘gap models’ that incorporate life-history mechanisms to simulate mixed-species, mixed-lifeform and mixed-age communities. A grassland (STEPPE) and a forest model (ZELIG) of similar structure were used to provide a comparative analysis between the two ecosystem types. Life-history traits were important in distinguishing dynamics between the simulated grassland and forest. Responses of the two systems to spatial extent and temporal variability were not simply scaled by the size or rate of processes associated with individual grasses or trees. Two important characteristics of a system were the form of growth by the dominant species (clonal or incremental) and the mode of competition (symmetric, asymmetric). Perennial grasses with clonal growth were responsive to a broader range of environmental conditions relative to the more restrictive response of individual trees with incremental growth. The endogenous rhythm associated with the lifecycle of individual trees was important to forest dynamics, but was not important for systems dominated by clonal grasses with an indefinite lifespan. Asymmetric competition for light by individual trees resulted in different system-level dynamics than for a grassland, where symmetric competition for belowground resources by individual grassland plants is the primary mode of competition.