Search

Since the mid-1990s the forests of central British Columbia have undergone an unprecedented Mountain Pine Beetle (Dendroctonus ponderosae Hopkins) (MPB) epidemic that has resulted in extensive mortality of canopy lodgepole pine (Pinus contorta var. latifolia Engelm.). This study investigated how seed-source availability, seedbed substrate, overstory structure, and time since MPB attack interact to affect post-MPB seedling recruitment of the dominant tree species of these forests. In addition to post-MPB recruitment, these forests may be regenerated by trees established in the understory prior to MPB disturbance. Accordingly, we examined abundance and patterns of all regeneration less than 130 cm tall. We found post-MPB recruitment was sparse. Subalpine fir (Abies laciocarpa (Hook.) Nutt.) comprised the majority of the post-MPB recruitment. It increased with local parent tree basal area and increased strongly with proximity to a major seed source. This resulted in a patchy distribution for subalpine fir post-MPB regeneration. Lodgepole pine post-MPB recruitment was limited by overstory shading. Recruitment of pine decreased as the total overstory basal area increased. Interior spruce (Picea glauca × engelmannii) post-MPB recruitment was similarly limited by total overstory basal area. Seedbed substrates were uniform and dominated by moss. Substrate type distribution did not change as time since MPB disturbance increased. The overall low post-MPB recruitment observed was likely due to a lack of disturbance to the moss-dominated forest floor. Moss is known to be a poor substrate in northern forests. The distribution of all regeneration less than 130 cm tall showed the same trends as the post-MPB regeneration. We believe the post-MPB seedling recruitment dynamics of these forests was not substantially changed from conditions prior to MPB disturbance. There was no pulse of regeneration up to 10 years post-MPB disturbance. Unless this changes, future stand structure will be dominated by the seedling bank established prior to the MPB epidemic. Subalpine fir dominated the seedling bank (68%) and post-MPB recruitment (94%). This suggests that MPB-disturbed forests are undergoing a substantial shift in landscape-level species composition.

The topic of model complexity is fundamental to model developers and model users. In this study, we investigate how over- and under-fitting of a driving function in a simulation model influences the predictive ability of the model. Secondly, we investigate whether model selection approaches succeed in selecting driving functions with the best predictive ability. We address these issues through an example with the forest simulator SORTIE-ND. Utilizing maximum likelihood methods and individual tree growth data we parameterize five growth functions of increasing complexity. We then incorporate each growth function into the simulation model SORTIE-ND and test predicted growth against independent data. Compared to the independent data, the simplest and the most complex growth functions had the poorest predictive ability while functions of intermediate complexity had the best predictive ability. The poor predictive ability of the simplest model is caused by poor approximation of the system while the poor predictive ability of the most complex model is caused by biased parameter estimates. A growth function of intermediate complexity was the most parsimonious model where error due to approximation and error due to estimation were simultaneously minimized. The model selection criteria AIC and BIC were found to select complex functions that were over-fitted according to the independent data comparison. BIC was closer to choosing the model that minimized prediction error than AIC. In this example, BIC is the more appropriate model selection criterion. It is important that both model developers and models users remember that more complex models do not always result in better predictive models.