Abstract:
The study focused on the development of models for stem volume, diameter distribution and height- diameter for indigenous hardwood tree species in Omo Biosphere Reserve, Omo Elephant Reserve, Akure SNR and Okomu National Park all within Southern Nigeria. Systematic line transect and Multiphase sampling techniques were adopted for sample plots allocation. A total of 26 sample plots of 50 X 50m were laid in line transects for all locations and a total of 1016 tree species of DBh ≥20cm were encountered. The data collected was subjected to both descriptive and inferential statistics. 115 tropical tree species distributed among 35 families with Celtis mildbreadii as the most abundant species and Sterculiaceae and Ulmaceae as the families with the highest number of individuals were observed. The mean DBH and mean height for the study area were 41.22cm and 20.86m respectively. All selected volume models had significant correlation coefficients and coefficients of determination, small standard errors of estimates, and significant F-values for all locations and for species with n≥10. When species were pooled, the weighted model, with the combination of D2H as predictor variable had the highest correlation coefficient of 0.97 and R2 of 0.93 (93 %) The logarithm-transformed model using lnBA as the predictor variable had a high F-ratio of 2295.58 and was rated the best for estimation of volumes of pooled while the model using BA as predictor variable was rated the best for volume for individual tree species .It was also observed that the highest number of trees concentrated in the diameter class of 20-39.9cm while few trees fall within the diameter class of above 100cm. Easy fit Statistical Software was used for the diameter distribution models and General Pareto was adjudged to be the best in modeling diameter distribution for the study area because the pdf had the least D-value of 0.0253 and the calculated D-value was less than the tabulated at p<0.05. Diameter- height relationship was modeled using Curve Expert Software. Gompertz Relation gave the best result in describing the relationship between both total height and DBH and merchantable height and DBH for the study area because they both have the lowest AIC values of 2202.27 and 2131.37 respectively and standard error.
