© 2005 Heron Publishing—Victoria, Canada
Carbon and biomass partitioning in balsam fir (Abies balsamea)
Zisheng Xing (1), Charles P.-A. Bourque (1, 2), D. Edwin Swift (3), Christopher W. Clowater (1), Marek Krasowski (1) and Fan-Rui Meng (1)
1. Faculty of Forestry and Environmental Management, University of New Brunswick, Fredericton, NB E3B 6C2, Canada / 2. Corresponding author (cbourque@unb.ca) / 3. Natural Resources Canada, Canadian Forest Service, Atlantic Forestry Centre, P.O. Box 4000, Fredericton, NB E3B 5P7, Canada / Received August 11, 2004; accepted January 21, 2005; published online July 4, 2005
Summary
Balsam fir (Abies balsamea (L.) Mill) was extensively sampled to investigate the effects of forest management practices, site location, within-crown
position, tree component (i.e., stem, foliage, branches and roots), and tree social classes on biomass and carbon (C) partitioning
at the individual tree level and across ecological regions. The sites were located in three ecologically distinct forest regions
of west-central New Brunswick, Canada. There were no significant differences in %C content of trees across ecological regions
or across tree social classes. However, at the individual tree level, significant differences were evident in biomass and
C allocation between different parts of the tree, between treatment types (i.e., unmanaged and pre-commercially thinned stands)
and between within-crown positions, indicating the need for separate estimates of biomass and C content of tree components
to obtain more precise estimates of quantities at the stand level. Calculating stand C content based on constant allocation
values, as is commonly done, produced errors of up to 15% compared with the values calculated in this study. Three allometric
equations of biomass and C that account for partitioning among different parts of the tree were developed and compared: (1)
a third-order polynomial, (2) a modified inverse polynomial and (3) a modified Weibull equation. Diameter at breast height
(DBH) was used as the only explanatory variable to describe fresh biomass, dry biomass and C content. All regressions derived
showed a high correlation with DBH, with most r2 values > 0.95. A comparison of the equation results showed that the modified Weibull equation gave consistent results with
the best overall fit and was the simplest of the three equations investigated. The regressions can be used to estimate forest
biomass and tree C content at the stand level, given specific information on DBH.
Keywords:
allometric equations, carbon content, dry biomass, eco-region, fresh biomass, unmanaged stands, pre-commercially thinned stands,
regression equations, social classes.
