Growth modeling of Carapa guianensis and Tetragastris altissima for improved management in native forests in the Amazon

Abstract In forests of the Amazon biome, Sustainable Forest Management Plans are based on technical guidelines. Such legislation provides for a cutting cycle that can vary from 25 to 35 years and a minimum cutting diameter of 50 cm. In view of the above, the present research aimed to evaluate the accuracy of regression models for the projection of growth in diameter and to calculate the t by diametric class for trees of the species C. guianensis and T. altissima. Four models with fixed effects were tested and best model was selected as the base model for the incorporation of random effects. The best fixed-effect model was Pienaar and Schiver. Thus, structures of variance and autocorrelation were added to this model to correct heteroscedasticity and autocorrelation. Finally, the time of passage for each species studied was calculated. The average annual increment in diameter estimated with the Pienaar and Schiver model with fixed effect was 0.38 cm.year−1 for C. guianensis and 0.46 cm.year−1 for T. altissima. Using the Pienaar and Schiver model with a random effect, the average annual diameter increase varied from 0.31 to 0.58 cm.year−1 for C. guianensis and from 0.37 to 0.65 mm.year−1 for T. altissima. Results showed that the estimated cutting cycle varied from 24 to 45 years for the species C. guianensis and from 21 to 38 years for the species T. altissima. Thus, using the cutting time of 25 to 35 years and a minimum cutting diameter of 50 cm for these species, can lead to incorrect decisions about the intensity of logging or the appropriate length of the cutting cycle. The growth and production models depict a synthesis of the growth dynamics of the forest, allowing to providing fundamental information for the definition of planning strategies, such as the establishment of a cutting cycle and an exploration intensity more compatible with the growth rate of the forest and for each species.