ABSTRACT

The relationship of lumber and wood waste yields from logs processed in a horizontal headrig sawmill type with three processing variables - log diameter size at level of 0.385+ 0.550+ 0.005 and 0.700+ 0.005 metres; bark thickness at levels of 4, 6 and 8mm, and width of sawkerf at levels of 2.50 and 3.50 mm were investigated in this study. The 72 logs used were sourced from 25 to 33 years old established plantations of the Forestry Research Institute of Nigeria (FRIN), Ibadan and comprised of 4 hardwood species - Gmelina arborea (Roxb.), Terminalia ivorensis (A.Chev.), Triplochiton scleroxylon (K.Schum) and Cedrella odorata (L.Kennedy). As part of preliminary investigation, sawmill operational activities i.n eight selected states of the country - Lagos, Ogun, oyo, Osun, Ondo, Ekiti, Edo and Delta were monitored with a view to collecting up-date baseline statistics on number of existing sawmills, headrig types used, recovery ratio values, volume of log input, sawnwood output and wood wastes generated. Processing of 400 logs, in 200 sawmills, were monitored in the exercise. From the mill experimental investigation carried out, log diameter size, bark thickness and width of sawkerf, at the levels they were applied, significantly influenced lumber and wood waste yields from the processed logs at 1% level Of probability. Processing of bigger logs manifested in increased lumber and decreased wood wastes yields regardless of the thickness of the bark content of the logs and width of sawkerf applied in sawing. At any of the log diameter size, lumber yield decreased with consequent increased wood wastes yield as bark thickness and width of sawkerf increased. Simple regression analysis showed that log diameter size is fairly well correlated with lumber and wood wastes yields from logs (correlation coefficient, R value of 0.768 each). Bark thickness and width of sawkerf are however poorly correlated with yields of lumber and wood waste from logs (correlation coefficient, R values of 0.440 and 0.241 in each case respectively). The multiple regression equation performed relating the two dependant variables with the three independent variables provided a correlation coefficient R value of 0.982 which implies that the three factors jointly explained as much as 96.4% (R2 - 0.964) of the total variations observed in lumber and wood wastes yields. From the field survey exercise carried out, there existed 1,713 sawmills in the sampled states between 2003 and 2004, with 1,853 installed headrig machines. Lumber recovery factor averaged 57 to 63% in the 8 states sampled. Annual log input was estimated at 4,831,400m3 with corresponding lumber (sawnwood) output of 2,921,600m3. Generated wood wastes were estimated at 1,940,000m3 comprising of slab: 1,051,300m3; sawdust: 481,300m3; and bark: 407,400m?. In view of the large quantities of wood wastes generated, workable management strategies that will facilitate use of the waste in value added wood products manufacturing such as ceiling boards and tiles as well as in the production of briquettes and fibreboard at least on small to medium scale basis, were recommended.