Adaptation of a spatially explicit individual tree-based growth and yield model and long-term comparison between reduced-impact and conventional logging in eastern Amazonia, Brazil

D. Valle, E. Vidal, M. Schulze, J. Grogan, M. Sales, P. Phillips, P. van Gardingen

Research output: Contribution to journalArticlepeer-review


Timber logging is one of the main land uses in the Brazilian Amazon. Despite its recognized potential as a sustainable activity, logging is generally conducted in an unsustainable or predatory manner, with significant negative environmental impact. There is increasing pressure to adopt more sustainable practices and reduced-impact logging (RIL) is gaining acceptance as a more environmentally benign alternative to unplanned, conventional logging (CL). Comparisons of these two harvest systems have largely focused on differences in efficiency (financial) and immediate impacts on stand structure, residual timber stocks and the physical environment.

Growth and yield simulation models allow long-term predictions about the response of forests to disturbance from harvesting, information that is essential to determine the effects of different management systems and to establish systems for the regulation or control of timber that are compatible with sustainable forest management. We describe the calibration of a spatially explicit individual tree-based ecological model within the SYMFOR/SIMFLORA framework for forests located near Paragominas in eastern Amazonia. Data originated from an experiment comparing RIL to CL within an area of 73.5 ha monitored over a period of 10 years. We evaluated biological realism and accuracy of the model, concluding that forest dynamics are adequately represented.

This model was used to evaluate long-term effects of RIL and CL on the forest. Our results suggest that total and commercial volume recovery following RIL are faster than following CL; it takes 10 and 30-40 years to recover total and commercial volume under RIL while under CL it takes 35-40 and over 60 years, respectively. Despite benefits from RIL, as currently practiced this logging system does not result in long-term sustained timber yields, reinforcing results from previous studies indicating that RIL must be combined with appropriate systems for yield regulation.

In the absence of silvicultural treatments, the model suggests that successive RIL harvests at current intensities will produce stands dominated by pioneer species groups, with few trees belonging to the emergent species group, and an increased proportion of defective trees in commercial size classes. These changes in stand composition present challenges to the long-term financial viability of RIL as currently applied in the Amazon region. (c) 2007 Elsevier B.V. All rights reserved.

Original languageEnglish
Pages (from-to)187-198
Number of pages12
JournalForest Ecology and Management
Issue number2-3
Publication statusPublished - 31 May 2007

Cite this