Vegetation organization and fire frequency in the western subarctic

• Main Author: Johnson, E. A. (Edward Arnold)
• Other Authors: Loucks, O. L.
• Format: Others
• Language: en
• Published: University of Saskatchewan 2010
• Online Access: http://library.usask.ca/theses/available/etd-11232010-112617/

The change of vegetation composition, with particular reference to frequency of fire, in the western subarctic is studied using two theoretical models and empirical data. <p> First, a mathematical model is used to explore the implications of predictable, relatively frequent fire on certain aspects of vegetation, particularly stand mortality. The predicted distribution from the model is found to fit well with stand age data collected in the subarctic forest. The three parameters of the model ecologically describe the lag before reburning can occur, regional resistance to burning and the local susceptibility to burning. The parameters are shown to behave consistently with logically independent empirical evidence related to regional and local climate and topography. The relationship of the hazard of burning function of the model to vegetation composition, particularly r-K selection, is discussed. <p> The second model explains community organization as the product of evolutionary forces which have produced different life history strategies in populations. These evolved differences in populations are the criteria by which populations are assembled into communities by environmental selection. Specifically the model partitions a population's abundance into environmental selection regime due to habitat and fire. Environmental selection consists of (i) populations ordered by their life history strategies (potential success) and (ii) an environmental selection function. The ordered populations can be called the r-K continuum. The environmental selection function transforms a population's potential success (its life history strategy) into the population's realized success (its actual abundance).<p> In order to give empirical content to the model a least square technique is used to partition the vegetation data into the habitat and fire selection regimes and principal components analysis is used to order the species abundance in each partition into r-K continua. The habitat selection regime gives a two dimensional continuum related to heat regime (energy budget) and nutrient regime. The fire selection regime gives a one dimensional gradient which orders species according to their temporal response to fire, from short-lived, fast growing, competitively inferior species to long-lived, slow growth, competitively superior species.<p> The recovery of vegetation after fire is discussed with respect to the selective influence of habitat and fire. The habitat and fire selection regimes are shown to cause, respectively, a resistance and resilience type response in the same community. The classical idea of black spruce-feather moss or white spruce climax is given a new interpretation in which these vegetation types no longer necessarily represent the end-points of compositional change. The speed of vegetation change is explained as the result of the r-K selectedness of the assemblage of species in a stand with respect to its habitat and fire frequency. The often discussed "adaptation" of vegetation to fire is now given explicit form in the model and the ensuing analysis of data.