Volume prediction through form factor and regression models by age class for Pinus taeda L.

Abstract

The objective of this research was to evaluate the use of form factor and volume models by age class for the volume prediction of a Pinus taeda L. forest located in the city of Telemaco Borba - PR. For this, 302 trees were measured and scaled by the Smalian method, which were distributed among the following age classes, obtained by Sturges formula: class I: 4.10 – 7.10; class II: 7.11 – 10.10; class III: 10.11 – 13.10; class IV: 13.11 – 16.10 and class V: 16.11 – 19.10. Four volume estimation methods were tested, such as the mean artificial form factor, form factor by age class, overall fitting of regression models and by age class. The selection of the best method was based on the fitting statistics of the models. The best fitted model for all data was Schumacher-Hall, while the best model by age group was Spurr, with the exception of class I. The comparative analysis among the techniques tested revealed that the mean form factor and by age group did not provide accurate estimates. The mean artificial form factor showed results superior to those obtained by the application of the average form factor for classes I, II and III. For the other classes, the average form factor per class performed best. For the volume equations, both the overall and per-class adjustment were satisfactory. There were no significant differences between treatments, however, the method that best estimated the volume of Pinus taeda L. was the fitting of volume models by age class