s:2226:"TI Allometric equations, wood density and partitioning of aboveground biomass in the arboretum of Ruhande, Rwanda 
AU Mukuralinda, A. 
AU Kuyah, S. 
AU Ruzibiza, M. 
AU Ndoli, A. 
AU Nabahungu, N.L. 
AU Muthuri, C.W. 
AB There is growing interest in plantation forests throughout Africa because of their role in environment, economy and people's livelihoods. However, the contribution of planted forests to climate mitigation is poorly understood, partly due to lack of allometric equations for biomass estimation. This study aimed to determine wood density and biomass fractions in aboveground components, and to develop biomass estimation equations for multispecies plantation forests in the arboretum of Ruhande in Rwanda. Allometric equations were developed by regressing diameter at breast height (DBH) alone or in combination with height or wood density or age of trees against the biomass of 45 trees harvested from a 200-ha site. Biomass estimates obtained from destructively sampled trees were up-scaled to estimate the amount of carbon stocked in the arboretum of Ruhande, assuming a stem density of 250 stems per ha. Wood density varied among the species but not tree size. The greatest fraction of aboveground biomass was allocated to stems (71–77%) compared to branches (19–27%) and leaves (1–8%) and varied by species. Equations developed fit the data well with DBH explaining over 90% of the observed variation in aboveground and stem biomass. Including height or wood density as supporting parameters reduced the relative error for aboveground biomass by 6.4 and 8.0% and improved model fit by 2.1 and 2.9%, respectively. Akaike information criterion (AIC) showed that wood density (AIC=63.6) and height (AIC=48.2) were the most suitable parameters to support DBH as a proxy for aboveground and stem biomass, respectively. Allometric equations developed in this study are useful tool for estimating carbon stocks of plantation forests in Rwanda and can enhance the accuracy of biomass predictions where site-specific equations rather than generalized models are recommended. Further studies focusing on development of allometric equations on belowground biomass in such systems are recommended. 
";