Modeling of tree recruitment by artificial neural networks after wood harvesting in a forest in eastern Amazon rain forest

Authors:
Open Access
Publisher website
Google Scholar
Abstract
The modeling of recruitment in tropical forests is important for studies of forest management sustainability, for giving adequate subsidies to the recovery of wood stock. The objective of the work was to estimate the recruitment after wood harvest, using a model of artificial neural network (ANN). The study area is located in the Tapajós National Forest (55° 00' W, 2° 45' S), Pará. In 64 ha of the study area, in 1979, an intensive harvest of 72.5 m3 ha-1 was carried out. In 1981, 36 permanent plots of 50 m x 50 m were randomly installed. These plots were measured in 1982, 1983, 1985, 1987, 1992, 1997, 2007, 2010 and 2012. To model the recruitment the variables of target subplot and its neighborhood were considered. The estimates obtained in the training and generalization of ANN were evaluated by statistics: correlation ( R) and root mean square error (RMSE) being obtained RMSE 35.6% and 0.89. It was possible to model the recruitment tendency over the time in tropical forests, after the wood harvest.