A Wood Density Variation Among Dominant Tree Species of Comberetum-Terminalia Woodland in Northwest Dryforests of Ethiopia

Abstract

Wood density is among the key variables in biomass and carbon stock estimation models.
Hence, better understanding of wood density variations within and between species is
critical for enhancement of forest biomass and carbon flux estimating models. This study
examine the inter and intra species wood density (ρ) variations to reduce sources of
uncertainty in biomass estimations of Combretum-Terminalia woodlands of Northwest
Ethiopia. For the determination of stem and branch densities, destructive sampling of 101
trees of the selected nine dominant tree species were taken. Three diameters at breast
height and branch diameter classes (5–10, 11–15, and > 16 cm) represented tree
population between 5 and 65 cm. The mean ρ of the studied tree species ranges from 0.308
to 0.612g cm-3 . While the branch and stem wood density of mixed‒species was estimated
at 0.498 and 0.509g cm-3, respectively. Despite the absence of general trends within a tree,
ρ decreased with stem height and branch thickness. Statistically significant variation in ρ
was observed among the selected tree species signifying the need for applying species
specific ρ in biomass models. The stem had higher ρ than the branch suggesting ρ derived
from the branch with a diameter of 11‒15 cm explained 92.3% variability of stem ρ. Some
wood densities adopted to our dataset overestimated the biomass of mixed‒species, by
27.5 to 33.6%. However, the best ρ based on averaging technique can be reliable to assess
biomass and carbon stock dynamics Combretum-Terminalia woodlands and comparable
vegetation formation with considering validation. Yet applying species specific ρ is
recommended for accurate biomass estimation models given the significant inter-species
variation.