AN EVALUATION OF SOME WHOLE STAND-LEVEL MODELS FOR PREDICTING MULTI-SPECIES BASAL AREA IN A TROPICAL RAIN FOREST

Abstract

Three models, Weibull distribution, stage-structured matrix and multiple regression models, were used to predict the basal area of a tropical rain forest. The results from the models showed that basal area prediction from variables that emerge from whole stand-level of organisation was feasible. The Weibull distribution model predicted 34.82 m² ha^-1 for the forest stand, the matrix model 36.73 m² ha^-1 while the multiple regression model predicted a mean basal area of 26.70 m² ha^-1. There were varying levels of accuracy of basal area estimates among the models. The matrix model produced better fit in the larger diameter classes while the Weibull distribution model consistently under-estimated the basal area. It was noted that significant correlation coefficients between Weibull parameters and stand factors like stem density and mean diameters were partly responsible for the errors in the estimates. An adjustment value for the bias in basal area estimates by the Weibull model was obtained.