PARENT SESSION
Contributed Oral Session 100: Modeling: Forest Systems
Wednesday, August 10, 1:30 PM - 5:00 PM, Meeting Room 519 B, Level 5, Palais des congrès de Montréal

Modeling tree size effects on growth-resource relationships.

MacFarlane, David^*,1, Kobe, Richard^*,1, ¹ Department of Forestry, East Lansing, MI, USA

ABSTRACT- Trees are the largest of all types of plants and, from seed to senescence, may pass through many orders of magnitude in size. Plant size affects performance, including growth; thus variation in size could obscure growth factors of interest. Size can be incorporated into the analysis of growth in several different ways, which have not been systematically compared. In this research, we explored a suite of different models characterizing sapling radial growth rate as a function of light, water and nitrogen availability that: (1) ignored size effects on absolute growth rate (AGR) -resource relationships, (2) rescaled AGR to relative growth rate (RGR) and related RGR to resources, and (3) related AGR and RGR to size and resource availability. For four different tree species: red maple (Acer rubrum, L.), sugar maple (Acer saccharum, Marsh), American Beech (Fagus grandifolia, Ehrh.) and northern red oak (Quercus rubra, L.) we found that AGR-based models that included size as a predictor variable provided the best fits to the data, the least bias, and the most unambiguous interpretation of growth-resource relationships across all growth model types examined. Models that used RGR or that did not include size variability over-predicted growth for slower growing saplings and under-predicted growth for faster growing ones, resulting in poorer model fits and increased bias. Non-proportional size effects on growth rates were observed for 3 of the 4 species examined, but lack of proportionality in size effects on growth had a negligible effect on the predictive power of growth models used in this study. Overall, growth models that explicitly considered tree size as a predictor of AGR had more favorable statistical properties (with regard to model fit and bias) and allowed for greater insight into the biological mechanisms of growth.

Key words: growth modeling, model selection, ontogenetic drift, size scaling