The implications of self-thinning dynamics on the temporal development of even-aged monocultures of two Lake States conifers

Data from many forest stands may be analyzed by producing a scatter plot of mean tree volume versus number of stems per acre in doubly logarithmic scale. In this scatter plot a linear upper boundary is often prominent beyond which the data do not extend. There is theoretical justification for the existence of this boundary, or self-thinning line. The location of this boundary line is of interest because it contributes to our knowledge of stand dynamics.; It is hypothesized that the level the self-thinning line is influenced by site quality within a species. Statistical methods to fit such boundary relationships are not well defined. A percentile based iterative method is developed to fit such a line to data.; The self-thinning line is a static description of stand development if viewed as an ecological boundary. It is a dynamic depiction of stand development only if the mean tree size and number of stems per acre found in the stand fall exactly on the self-thinning line. A potentially more accurate way to identify the self-thinning line is to fit dynamic models to remeasured permanent plot data. A semi-parametric method is developed to identify the self-thinning line using permanent sample plot data.; To explore individual tree dynamics within a stand, a system of simultaneous differential equations is first developed to describe stand level development consistent with the self-thinning rule. The method of distribution modifying functions is then employed to distribute stand growth among individuals. A logical method is used to disaggregate stand mortality to individuals.; The results suggest that site quality influences the level of the self-thinning line only for certain species. Stand history has as much if not more impact than site quality on the level of the self-thinning line. Using a simultaneous differential equation system to link growth and mortality explicitly recognizes the true simultaneous nature of these processes. The methodology developed for disaggregating stand level growth and mortality to individual trees is a promising simulation tool for furthering understanding.