UBC Theses and Dissertations
Growth of British Columbia coastal species in response to thinning and fertilization treatments Rathbun, Leah C.
The successional processes of the mixed-species Pacific coastal temperate rain forests of British Columbia (BC), Canada, are defined by gap dynamics, where small-scale disturbances, mainly due to windthrow, create openings in the canopy necessary for regeneration. Douglas-fir (Pseudotsuga menziesii var. menziesii (Mirb.) Franco) is the dominant, pioneer species in this area and western hemlock (Tsuga heterophylla (Raf.) Sarg.) and western redcedar (Thuja plicata Donn) are the late-successional, shade-tolerant species. Silvicultural systems such as variable retention systems have been applied to many of the secondary growth mixed-conifer forests. Variable retention in this area is designed to differ dramatically from stand to stand. This approach differs from the traditional even-aged management applied to forests of the Pacific Northwest coast. In this study, a model-based approach was used to investigate how multiple treatment interventions as a part of active management across a landscape affect mortality and growth within actively managed stands. There is a need for this information as current growth and yield models used in this area are limited by either the number of species which can co-exist in a stand (e.g., the model TASS of BC) or are limited by the need for data not commonly obtained in inventory databases (e.g., the models FVS and ORGANON of USA). Additionally, no growth and yield models have been developed to include variable retention systems, where a variety of thinning intensities and spatial patterns, timing of thinning and fertilization treatments, and number of treatments are used. Mortality, diameter increment, and height increment models were developed and the effects of thinning, fertilization and the combination of thinning and fertilization were examined for Douglas-fir, western redcedar, and western hemlock. For each species, shade-tolerance was found to impact the possible predictor variables included in model development. The use of a generalized logistic survival model resulted in accurate estimates for larger trees, but poor results for smaller trees. To model the effects of fertilization, additional fertilization effect variables were included in the models; conversely, thinning effects were modeled using the immediate change in state variables such as basal area of larger trees which occurred immediately following thinning.
Item Citations and Data
Attribution-NonCommercial-NoDerivatives 4.0 International