Avaliação dos danos e métodos de regulação da floresta submetida à exploração de impacto reduzido na Amazônia oriental.

Abstract

Logging impacts on the residual forest and on the forest floor were assessed in three managed forest areas in the regions of Breu Branco, Novo Repartimento and Paragominas, located in southeastern Pará, Brazil. Damage plots (transects, 10m wide) of c. one ha were established and measured before and after logging viz. 22 in Breu Branco forest management unit (FMU), 28 in Novo Repartimento, and 14 in Paragominas. One plot was laid down at random within each logging unit of c. 100ha of each annual coupe of the FMU. All individuals with DAP≥10cm of all species were measured, except palms. The second assessment was carried out just after logging. At this stage the impact on the forest floor and on the residual forest were evaluated. Logging impacts were indeed reduced as compared to conventional operations which impact c. 25-35% of the trees. Most of the harvesting impacts were concentrated in the smaller sizes classes (10-30 cm). Density (N ha-1) decreased 2,60% in Breu Branco, 5,38% in Novo Repartimento and 12,77% in Paragominas; basal area reduced 2,46% in Breu Branco, 2,97% in Novo Repartimento e 8,55% in Paragominas; volume (V ha-1 decreased 2,45% in Breu Branco, 2,84% in Novo Repartimento e 8,52% in Paragominas. The basal area (G ha-1) was reduced in 2,46% in Breu Branco, 2,97% in Novo Repartimento and 8,55% in Paragominas; volume (V ha-1) reduced 2,45% in Breu Branco, 2,84% in Novo Repartimento and 8,52% in Paragominas. The area of forest floor affected by harvest was comparable to other similar studies. Breu Branco had the biggest area affected (18,37%), followed by Paragominas (17,90%) and Novo Repartimento (8,94%). The precision of the assessment did not comply with the predefined error limit of 10% of the average to estimate the parameters. Thus further research is needed to determine the best size and shape of logging damage plots. Yield regulation simulation carried out in the three FMU revealed that the equations applied which do not use the average mean annual increment (MAI), would not be applicable to estimate the annual allowable cut (AAC) in the conditions of the Brazilian Amazon as they overestimate the AAC. Among the equations which use the MAI the one suggested by Alder (2000) proved more adequate. However, to allow a better effectiveness, it is necessary to estimate with a better precision, the losses in productivity due to logging. It is recommended to utilize the Dawkins (1964) pantropical MAI 1 m³ ha-1 yr-1 as a baseline to determine the AAC in the eastern Amazon while more effective data on MAI from growth and yield studies is available in the region.