Predicting Diameter Distributions of Longleaf Pine Plantations: A Comparison Between Artificial Neural Networks and Other Accepted Methodologies

constructed from poor Weibull estimates will produce biased volume estimates. Thus, general applicability of a procedure Artificial neural networks (NN) are becoming a popular estimation tool. Because they require no assumptions about the form of a fitting does not mean that the procedure is always best, as growthfunction, they can free the modeler from reliance on parametric and-yield modelers are always searching for procedures to approximating functions that may or may not satisfactorily fit the improve d.b.h. distribution estimates. Artificial neural obierved data. T o date there have been few applications in forestry networks (NN) may provide better estimates of d.b.h. science, but as better NN software and fitting algorithms become available, they may be used to solve a wide variety of problemsdistributions that do not rely on assuming an imperfect particularly problems where the underlying relationship between underlying probability model. predicted and predictors is unknown. We benchmark tested an glternative to the traditional Weibull probability distribution function, diameter-at-breast-height moment, and direct parameter prediction models for approximating stand-diameter distributions. Using a feedforward backpropagation network, we demonstrated that NN are a somewhat better option. Unlike Weibull approximations, NN solutions cannot easily be mathematically constrained to match known reality constraints, but this difficulty is easy to overcome in practice.