Forest management and regeneration of tree species in the Eastern AmazonNew Generic article

Schwartz, G. (2013)
PhD thesis, Wageningen University, Wageningen, the Netherlands
ISBN: 978-94-6173-466-2, 132pp.
With references, with summaries in English, Dutch and Portuguese

The Amazon encompasses approximately 60% of all remaining tropical rainforests of the world and most of its area is in Brazilian territory, where nearly 20% of the forest cover has been lost due to deforestation.  Forest management in the Brazilian Amazon started experimentally in the 1970s, when the Malayan Uniform System was adapted to local conditions by being made a polycyclic silvicultural system; later, elements of the CELOS system were included.  The harvesting techniques were also improved with the introduction of reduced-impact logging (RIL) techniques.  RIL is more environmentally-friendly than conventional logging, having low or no impact on wildlife and populations of harvested tree species.  Some studies, however, show that forests managed under the current practices may not have enough regeneration of commercial species and that the harvested species will have lower timber yields in future cutting cycles.  Among the possible options for mitigating these problems is the application of post-harvesting silvicultural treatments.  Promising candidates are the tending of the naturally established regeneration by diminishing competition with lianas and other commercially less valuable trees, and enrichment planting in gaps created by RIL.  These treatments, however, require long-term financial investments, which might reduce their profitability.  On the other hand, they may also add value to the forests, thereby enabling managed forest to compete better with other land uses, which have a more negative impact on forest cover.

In this thesis the following questions were addressed: 1) How does the regeneration of commercial tree species respond to the different levels of disturbances created by RIL? 2) How does the regeneration of commercial tree species respond in the medium term to disturbances caused by RIL? 3) Which of the two post-harvesting silvicultural treatments applied in logging gaps is most suitable: tending the natural regeneration, or enrichment planting followed by tending? 4) To what extent are the post-harvesting silvicultural treatments of tending natural regeneration and enrichment planting in logging gaps profitable in the long term?

The short-term effect of RIL on the regeneration of seven commercial tree species was assessed in Chapter 2, through an experiment carried out in an intensively studied plot in the 600,000 ha Tapaj√≥s National Forest, eastern Amazon, Brazil.  The species studied belong to three functional groups: the long-lived pioneers (LLP) Bagassa guianensis and Jacaranda copaia; the partially shade-tolerant (PST) Hymenaea courbaril, Dipteryx odorata, and Carapa guianensis; and the totally shade-tolerant (TST) Symphonia globulifera and Manilkara huberi.  All individuals of these species that were < 20 cm in diameter were inventoried and measured three times (in 2002-2003) before logging and then twice after logging (in 2003-2004).  The logged area was sampled through 28 transects consisting of 10 x 10 m plots, covering 24.39 ha in total.  The average intensity of harvesting was 22 m3 ha-1 (21% of the original commercial volume).  In Chapter 3 it is evaluated the medium-term effects of RIL, part of these harvested transects were measured twice more (in 2006 and 2009) and compared with another three transects established in an unlogged adjacent forest and also measured in 2006 and 2009 (total area measured was 2.37 ha).

Chapter 4 describes the research carried out in the forest management area of the Orsa Florestal forestry company in Jari Valley, in eastern Amazon, Brazil.  Here, 64 2-year-old logging gaps with an average size of 427.2 m2 were sampled.  In 34 of these gaps the enrichment planting treatment was applied, 15 gaps underwent the tending treatment, and the remaining 15 were used as control.  The mortality and growth rates of the monitored individual trees were measured for four years.  In Chapter 5, the responses of individual trees to the tending and enrichment planting treatments were extrapolated in the long term (60 and 90 years) to evaluate timber production and profitability.  The two silvicultural treatments were compared to a control treatment representing the standard current procedures of RIL (when gaps have no post-harvesting treatments), assuming different long-term interest rates, rising prices for round- and sawnwood, and increasing harvesting costs.

The results show that RIL did not have a destructive effect on the harvested forest: 52.5% of the area remained untouched and only 9.2% was highly disturbed by logging operations.  All species except C. guianensis increased in density after logging.  The LLP species mainly increased in density in highly disturbed plots, whereas the PST and TST species increased in density in logged plots under low disturbance levels.  The positive effect of increasing densities of the harvested species was ephemeral and had disappeared two years after logging, but the modified forest structure was still discernable after 6 years.  RIL had a positive effect on the height growth rate of S. globulifera and on the diameter growth rate of C. guianensis.  Plants growing in disturbed plots had faster height growth rates than plants growing in the undisturbed plots, but the same pattern was not observed for the diameter growth rates.

Of the 10 species analysed, five (the LLP Goupia glabra and Laetia procera, and the PST Dinizia excelsa, Tachigali myrmecophila, and Trattinnickia sp.) responded positively to silvicultural treatments.  Based on these results, the treatment of tending is recommended for areas of logged tropical forest with sufficient natural regeneration of commercial species.  The enrichment planting treatment should be applied using only species with high initial growth rates and commercial value.  Neither tending nor enrichment planting was profitable under the long-term interest rates of 3%, 6%, and 9% and only sawnwood produced through tending under a 750% increase of timber prices in 60 years would make the treatment profitable.  Reduction in costs and improvements in the tending and enrichment planting techniques would nonetheless make these post-harvesting silvicultural treatments profitable for both round- and sawnwood.  Once having these treatments more profitable, managed forest will be a more economically competitive land use in the Amazon