versión impresa ISSN 0044-5967versión On-line ISSN 1809-4392
VASCONCELOS, Sumaia S.; HIGUCHI, Niro y OLIVEIRA, Marcus V.N.. Projection of the diameter distribution in a selective logging forest in the western Amazon. Acta Amaz. [online]. 2009, vol.39, n.1, pp.71-80. ISSN 0044-5967. http://dx.doi.org/10.1590/S0044-59672009000100007.
The diameter distribution of an experimental forest stand in the Western Amazon was projected using a stochastic model after selective logging. The study was developed using data from five permanent plots located in the colonization project Pedro Peixoto, in the state of Acre. Initial measurements of diameter at breast height (DBH) were taken in 1996. The forest was selectively logged in 1997 and DBHs were re-measured in two different occasions, 1999 and 2001. A probabilistic transition matrix (Markov Chain) was used to project the diameter distribution of the number of surviving trees in each diameter class. The model was first tested to project the diameter distribution in 2001, based on DBH measurements from 1997 and 1999. When the projected diameter distribution for 2001 was compared with the field data from the same year, a Chi-squared test (α = 0.05) showed that there was not significant difference between the expected and observed diameter distribution. After that, a projection for 2005 (four years in the future) was run using DBH measurements from 1997 to 2001, indicating that mortality rate was similar to 2001. If repeated the rate of recruitment of 2005, the total number of trees will be higher than observed in 2001. The dynamics of the studied forest suggests that there is not a definitive pattern to changes in diameter distribution and mortality, which indicates a stochastic or probabilistic pattern. This pattern is better modeled by the Markov Chain to project the forest dynamics of studied area, and can help on determination of timber harvesting or the tendencies of forest dynamics in a near future.
Palabras clave : Selective logging; Diameter distribution; Markov Chain; Amazon.