Forest Investment Account (FIA) - Forest Science Program
FIA Project Y081165

    Improving the prediction of species composition of managed aspen and white spruce stands within boreal mixedwoods
Project lead: DeLong, Craig (BC Ministry of Forests and Range)
Contributing Authors: Kabzems, Richard; DeLong, Craig
Subject: Forest Investment Account (FIA), British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
To ensure the sustainability of boreal mixedwood forests it is necessary to predict the impacts of our management decisions upon them (Messier et al. 2003). To do this, we need models, and improving models for boreal mixedwood stands has been identified as a priority for 07/08 Timber Growth and Value Program. Significant efforts have been made to parameterize existing stand-level growth models (e.g., TASS, SORTIE, MGM) for aspen – spruce-pine mixedwoods in BC and Alberta, and this work continues (see Linkages). To project regenerating stands, these models need a specification of the initial stand composition (e.g. per-species stocking density and size-structure). Initial stand composition depends partly on harvesting and silvicultural treatment, but also on the pre-treatment composition, spatial context and essentially random factors (Lieffers et al. 2003). The relations between pre- and post-treatment compositions must be described statistically, e.g. as a “regeneration” matrix of transition probabilities between distinct stand types or compositional strata. However, the variability in stand composition following common management practices (e.g., clearcut harvest of a spruce-leading stand, mechanical site preparation and planting) is not well quantified. Thus, a recent landscape-modeling project in Dawson Creek (Timberline Forest Inventory Consultants and Farnden, 2005) used a regeneration matrix based on expert opinion, not data.
We propose a study that will help to fill this information gap. A side benefit of this project is to determine what elements (i.e., pre-harvest tree species composition, silviculture intervention, spatial context) are required for the development of a stand that over succession will develop certain desired habitat characteristics that may be important for certain organisms. In the most recent Timber Supply Analysis for the Dawson Creek Timber Supply Area all leading species stands that were sufficiently restocked and all leading species stands harvested in the future were assumed to regenerate according to simple stand yield curves (MoF 2002). Although the general principles of forest stand development (Oliver and Larson 1996) and spruce-aspen mixedwood succession are well understood (Lieffers et al. 2003) the developmental trajectory and fate of any particular patch of mixedwood forest remains difficult to predict. Process models in a landscape context may eventually synthesize the needed information in a reliable manner, but many more years of ecosystem-level research will be needed to address all relevant factors. Even when completed, process models will need empirical datasets compiled for purposes of validation and verification. The statistical description of state transition probabilities (e.g. Markov modelling) provides a simple alternative to process modelling. Forest science researchers, modellers and forest managers appreciate having such "real world" information assembled in a useable manner, yet this has never yet been done for B.C.’s northeastern boreal forest.
This proposal would quantify changes in tree species composition with transition probabilities from pre-treatment to current forest types similar to analysis by Weir and Johnson (1998) for boreal mixedwoods in Saskatchewan.


2007/08 Year End Report (0.1Mb)
Improving the prediction of species composition... (Summary) (20Kb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated August 16, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to