Basal area and dominant height growth and survival models incorporating general response to silvicultural treatments for loblolly pine stands were developed using data from various silvicultural experiments across Southern United States. Growth models for treated stands were developed by multiplying base-line growth models with modifier response functions/multipliers accounting for effects of thinning, fertilization, and control of competing vegetation. Chapman-Richards functions were used to model the base-line growth. Separate response functions to mid-rotation thinning and fertilization effects were developed. The thinning response function was based on duration and rate parameters and is sensitive to stand age at the time of thinning, time since thinning, and intensity of thinning. The fertilization response functions were based on Weibull distribution and the magnitude of responses varies with time since application of fertilizers, type of fertilizer elements applied, and rate of application. Response functions were integrated as a multiplier to base-line models. Response to early control of competing vegetation was incorporated into base-line models through multiplier factors. Multiplier factors were calculated based on growth difference between treated and untreated stands. A difference function, derived from differential equation with age, initial stand density, and site index served as the base-line survival model. The survival model was adjusted for thinning treatment by including an additional independent variable that represents thinning intensity. No adjustment was required for survival model in response to fertilization and competing vegetation control. All growth models were unbiased and had adequate performance in predicting basal area and dominant height following treatments. Models were developed to represent general growth trends in response to treatments. The response functions developed here can be viewed as general response functions. / Ph. D.

Oaks (Quercus spp.) are an important part of the forested landscape in the eastern United States. Although oak is increasing in standing volume, an oak regeneration bottleneck has occurred throughout its range in recent decades. Subsequently, as oak overstory is being harvested, rarely is oak recruited into the overstory to maintain the historic dominance of overstory oak. In the absence of fire and subsequent canopy closure, mesic species have proliferated, frequently forming a dense understory, inhibiting oak regeneration success. This study was developed to determine species dynamics between oak and oak competitors in response to silvicultural treatments in the southern Appalachian Mountains of North Carolina. The treatments were: a shelterwood treatment (25-30% basal area reduction through mid-story removal with herbicides), a prescribed fire treatment (two late dormant season fires occurred over a 9-year period), a shelterwood and burn treatment (prescribed fire 3-5 years following 30-40% basal area removal), and an unmanaged control. To determine treatment impacts on the regeneration layer, importance value and stems ha-1 were calculated at the species group and individual species level 0- and 9- years post initial treatment. A principal component analysis and an analysis of basal area by treatment 0- and 9-years post-treatment were used to determine the influence of site-specific characteristics on regeneration layer response. The greatest relative increases in importance values were 1401% and 2995% for the red oak group and yellow-poplar (Liriodendron tulipifera), respectively, in the shelterwood and burn (SWB). Change in all species groups were predominantly influenced by the smallest size-class (<0.6 m tall), with the exception of northern red oak (Q. rubra) and yellow-poplar in the SWB. The SWB significantly reduced importance values of all shade tolerant species groups and was the only treatment to decrease red maple (Acer rubrum) importance value and density over the study years. The prescribed fire (RXF) treatment increased red oak group importance value, while simultaneously decreasing yellow-poplar importance value and increasing red maple importance value. Changes in the red oak group in the SWB and the RXF were driven by northern red oak and scarlet oak (Q. coccinea), respectively. Treatments do not appear to change the competitive status of the white oak group. Elevation was closely associated with the red oak group. Yellow-poplar importance value increases, white oak group importance value increases, and site index were closely associated. Decreases in basal area were greatest in the SWB, and the SWB was the only treatment to significantly decrease overstory basal area. The RXF and SWB treatments improved the competitive status of only some oak species, but modifications to these treatments may result in better control of yellow-poplar and red maple competition, further improving oak's competitive status. Site specific factors such as elevation and site index may have impacted the regeneration layer response to treatments. / Master of Science / Oak-hickory (Quercus and Carya spp.) and oak-pine (Quercus and Pinus spp.) forest types occupy approximately 57 million and 11 million hectares of forestland in the eastern United States, respectively. Oaks are considered ecological and economic keystone species throughout the eastern U.S and maintenance of this genus in eastern U.S. forests has been a primary regional focus for decades. Historic disturbance regimes are estimated to have been much different than they are today. Fire was a common disturbance mechanism prior to fire suppression in the early 20th century. Frequent fires maintained much of the oak component historically. In the absence of fire, the species found in the understories of mature oak stands are commonly mesophytic species, such as yellow-poplar (Liriodendron tulipifera L.) and red maple (Acer rubrum L.). Over the last several decades, research has been conducted to investigate the impacts of treatments targeting the promotion of oak regeneration, but results have been varied and valuable long-term studies are rare. To determine the effects of treatments on the regeneration dynamics of oak and its competitors, four treatments were compared in the southern Appalachian Mountains. Treatments included a control, shelterwood harvest (SW), prescribed fire (RXF), and a shelterwood and burn (SWB). Stand structure and composition were monitored over a 9-year period post-treatment. Overall, results indicate the shelterwood and burn treatment has the greatest potential to improve the competitiveness of the red oak group in the regeneration layer, but yellow-poplar competition in the shelterwood and burn will need be addressed, considering its large increases in this treatment. Although increases in the red oak group were not as great as increases in the RXF treatment compared with the SWB, fire does show promise as a method to increase oak regeneration success. Changes in red oak group importance value varied with elevation, emphasizing results of treatments can be affected by site characteristics. Treatments were not successful at enhancing the competitive status of white oak (Quercus alba L.). Silvicultural treatments can be used to improve the competitive status of oak on sites in the southern Appalachian Mountains, but close monitoring of species dynamics throughout the rotation are needed to ensure long-term oak success.

The relative growth response of planted loblolly (Pinus taeda L.), shortleaf (Pinus echinata Mill.), Virginia (Pinus virginiana Mill.), and Eastern white (Pinus strobus L.) pines to intensive silvicultural practices such as woody competition control and fertilization in the Piedmont Province of Virginia is unclear. To address this issue, during 1999, a mixed stand of Virginia pine and hardwoods was clearcut and site-prepared by herbicide application. Three replications containing strips of loblolly, shortleaf, Virginia, and white pines were planted at a 3 m x 1.5 m spacing during February-June, 2000. Four different sources of loblolly pine seedlings were used. The strips were subsequently split across to accommodate four different silvicultural treatments: (1) check (no treatment); (2) woody vegetation control; (3) fertilization; and (4) weed control plus fertilization. The weed control treatment used two directed spray herbicide applications in 2001 (triclopyr and glyphosate) and 2003 (glyphosate) and one mechanical cutting of the remaining hardwoods in 2004. Fertilizer containing N, P, K, and S was applied in 2001, only N in 2002, and N plus P in 2004. Crop tree survival was highest for loblolly pine, decreased in shortleaf and Virginia pines, and Eastern white pine had the poorest survival through age 5. Fertilization without controlling the competing hardwoods decreased survival in all planted pines due to the increased hardwood competition. Loblolly pine was tallest (4.7 m) through the five-year period, shortleaf and Virginia pines were shorter (2.95 m and 3.06 m, respectively), and white pine was shortest (1.7 m). When fertilized, hardwoods were taller than white pine seedlings, almost equal in height to shortleaf and Virginia pines, but shorter than loblolly pine seedlings. Silvicultural treatments had no significant impact on tree height. However, the weed control treatment increased fifth-year pine DBH and stem volume, while fertilization did not. When applied in combination with weed control, there was no additional increase in crop tree DBH and stem volume due to fertilization beyond that from weed control only. The increased availability of soil nutrients due to harvesting the previous stands and allocating them to the crop trees by controlling the competing woody vegetation was enough to meet the nutrient requirements of the young seedlings up to this age. / Master of Science

Existe hoje uma crescente demanda global por ações de restauração de florestas tropicais e um dos principais entraves para a sua concretização é a viabilidade econômica dos projetos. Em cenários em que a restauração ecológica exige a implantação de reflorestamentos de espécies nativas, a produção de madeira torna-se uma alternativa importante para viabilizar economicamente a atividade, tornando-se um meio para atingir o objetivo final da restauração do ecossistema. A carência de conhecimentos técnico-científicos sobre o potencial de produção das espécies nativas em reflorestamentos é, no entanto, um entrave significativo para a sua aplicação em larga escala. Nesse contexto, o presente trabalho teve como objetivo a avaliação do potencial de espécies nativas para a produção de madeira serrada em plantios de restauração florestal. Para isso, avaliou-se o crescimento de 16 espécies nativas com potencial para produção de madeira serrada em 13 reflorestamentos mistos, com idades entre 6 e 96 anos, localizados no domínio da Floresta Estacional Semidecidual no interior do estado de São Paulo. Essas espécies foram avaliadas quanto ao crescimento em diâmetro real, diâmetro potencial, altura total e altura comercial através da modelagem destes parâmetros por técnicas de regressão linear múltipla. Tais modelos foram gerados a partir da correlação existente entre a idade do reflorestamento, o fator ambiental mais influente (relacionado ao solo e clima) e o parâmetro analisado, obtendo-se as perspectivas de desenvolvimento de cada espécie em cenários de baixa, moderada e alta produtividade de sítio. Foram estimadas as projeções do crescimento diamétrico considerando-se todos os indivíduos de cada espécie amostrados nos reflorestamentos, bem como a partir dos 25% melhores indivíduos, visando-se estimar o potencial de crescimento das espécies em condições mais favoráveis de manejo silvicultural. Foi realizada uma análise descritiva das variáveis qualitativas avaliadas em campo (número de fustes, qualidade da árvore e nível de exposição à luz), associadas às alturas total e comercial, verificando-se se as relações existentes entre estes parâmetros são capazes de orientar propostas de manejo silvicultural adequado para as espécies. Várias das espécies nativas estudadas, principalmente Cariniana legalis e C. estrellensis, apresentaram potencial satisfatório de produção de madeira serrada em plantios de restauração florestal, principalmente em condições ambientais favoráveis e se submetidas a manejo silvicultural adequado. No entanto, mesmo sob condições favoráveis, algumas espécies apresentaram crescimento excessivamente lento, desestimulando seu plantio para a produção de madeira. O ciclo de produção de madeira esperado para cada espécie foi muito variável em função das condições ambientais da área em restauração, reforçando a importância de escolha orientada de espécies para cada condição de sítio, do preparo adequado do solo e da consideração dessas condições ambientais para definir os modelos de restauração para fins de exploração de madeira. Da mesma forma, a maior parte dos indivíduos encontrava-se dominada na estrutura da floresta, reforçando a importância do espaçamento, da composição de espécies na vizinhança e de desbastes no potencial de produção de madeira. / There is a growing demand today for tropical forest restoration efforts, and one of the main barriers for accomplishing this demand is the economic viability of the projects. In scenarios where ecological restoration requires the implementation of native species plantings, the exploitation of timber can be an important alternative to make restoration financially viable, as a way to achieve the ultimate goal of the restoration of the ecosystem. The lack of technical and scientific knowledge about the potential of timber production by native species in reforestation is, however, an important obstacle for applying this alternative at the large scale. In this context, we evaluated the growth of 16 native species with potential for timber production in 13 restoration plantings, with ages ranging between 6 and 96 years, located at the Seasonal Semideciduous Forest of inland São Paulo State, southeastern Brazil. These species were evaluated regarding their growth in real and potential diameter, as well their total and commercial height, by modeling these parameters through multiple linear regressions. These models were generated based on the correlation among the age of the restoration planting, the most influent environmental factor (related to climate and soil) and the analyzed parameters, allowing us to obtaining the perspectives of growth for each species in scenarios of low, moderate and high site productivity. We established projections of growth in diameter considering all individuals sampled for each species in the reforestations, as well as the best 25% individuals, as means of estimating the potential growth of each species in more favorable scenarios of silvicultural management. We carried out a descriptive analysis of the qualitiative variables evaluated in the field (number of boles, quality of the tree and exposition to sun light), which were associated to the total and commercial heights, to verify if the relationship between these parameters may guide proposals of suitable silvicultural management for these species. Many of the studied species, especially Cariniana legalis and C. estrellensis, showed a satisfactory potential of timber production in restoration plantings, mainly in more favorable site conditions and under adequate silvicultural treatments, However, even under favorable conditions, some species presented growth rates excessively slow, which discourage their planting for timber production. The expected cycle of timber production for each species was highly variable according to the environmental characteristics of the site, thus reinforcing the importance to guide the selection of species for each site condition, to accomplish a suitable soil preparation before planting and to consider these site conditions for designing models of restoration for timber exploitation. In addition, most trees were dominated in the forest structure, which reinforces the importance of spacing, species composition of the vicinity and thinning in the potential of timber production.

Submitted by Erika Demachki (erikademachki@gmail.com) on 2017-01-06T18:59:59Z No. of bitstreams: 2 Tese - Fernanda Gomes Ferreira - 2016.pdf: 3057846 bytes, checksum: 5cdd832b55313b3b126d066db50d3548 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-01-09T09:48:21Z (GMT) No. of bitstreams: 2 Tese - Fernanda Gomes Ferreira - 2016.pdf: 3057846 bytes, checksum: 5cdd832b55313b3b126d066db50d3548 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-01-09T09:48:21Z (GMT). No. of bitstreams: 2 Tese - Fernanda Gomes Ferreira - 2016.pdf: 3057846 bytes, checksum: 5cdd832b55313b3b126d066db50d3548 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-10-14 / This study had as objective to evaluate diametric growth aspects and the dynamics in a Semideciduous seasonal forest, twelve years after the conduction of a silvicultural management. The main objects were: to analyze the arboureous community density during the period that ranges from 2003 to 2015, after silvicultural interventions (Chapter 1); to develop Markov models and test its predictive capacity for different silvicultural managements (Chapter 2). The experiment was carried out in 2003, in Fazenda Vagafogo, Pirenópolis, Goiás. Experimental area was divided in four blocks, and, inside each one, three plots were allocated (25 30 m – 750 m 2), with the following treatments: Treatment 1 – control (T1), without interventions; Treatment 2 – all woody species were removed at a distance of one meter, considering the wanted trees (T2); Treatment 3 – T2 was repeated, joining to large lianas removal all over the plot (T3). Inside the plots, in 2003, 2007 and 2015, all the individuals (DBH > 3 cm) were measured. The community dynamics, considering the period ranging from 2003 to 2015, was described according to mortality taxes, recruitment, basal area gain and loss, net change rate to the number of individuals and basal area. Collected data were also used to develop Markov models that describe the seasonal forest dynamics of the diameter distribution after silvicultural treatments application (GT1, GT2, and GT3). Treatments did not negatively affect long term the community dynamics, because, in all treatments, recruitment was higher than mortality. Growth taxes and annual periodic increment (API) between 2003- 2007 and 2007-2015 analysis allow us to conclude that silvicultural treatments lost its effect, because taxes drop considerably in the second period compared to the first one. Models GT1, GT2 and GT3 were tested for an 8 year period (two consecutive projections), and all of them were able to efficiently simulate the diametric structure. This way, Markov chain is an efficient tool to project seasonal deciduous forests dynamics upon silvicultural interventions, which strengths its importance as a tool for forest management. / O presente estudo teve como objetivo geral avaliar aspectos da dinâmica e do crescimento em diâmetro de uma floresta estacional semidecidual, doze anos após a aplicação de tratamentos silviculturais. Os objetivos específicos foram: analisar a dinâmica da comunidade arbórea durante o período de 2003 a 2015, após intervenções silviculturais; desenvolver modelos markovianos e testar sua capacidade preditiva para diferentes tratamentos silviculturais. O experimento foi implantado em 2003, na Fazenda Vagafogo, em Pirenópolis, Goiás. A área experimental foi dividida em quatro blocos, e, dentro de cada bloco foram alocadas três parcelas de 25 30 m (750 m 2) com os seguintes tratamentos: Tratamento 1 – testemunha (T1), sem intervenções; Tratamento 2 – retirada de todas as espécies lenhosas em um raio de um metro em relação às árvores desejáveis (T2); Tratamento 3 – repetiu-se o T2, incorporando a retirada de cipós de grande porte em toda a parcela (T3). Dentro das parcelas, nos anos de 2003, 2007 e 2015, foram tomadas medidas de todos os indivíduos com DAP > 3 cm. A dinâmica da comunidade, considerando o período de 2003 a 2015, foi descrita por meio das taxas de mortalidade, recrutamento, ganho e perda em área basal, taxa de mudança liquida para número de indivíduos e área basal. Os dados coletados também foram utilizados para o desenvolvimento de modelos markovianos que descrevem a dinâmica da distribuição diamétrica da floresta estacional após a aplicação dos tratamentos silviculturais (GT1, GT2 e GT3). Os tratamentos aplicados não afetaram negativamente a dinâmica da comunidade no longo prazo, pois, em todos os tratamentos, o recrutamento foi maior do que a mortalidade. A análise das taxas de crescimento e incremento periódico anual (IPA) entre os períodos 2003-2007 e 2007-2015, nos permite concluir que os tratamentos silviculturais perderam o efeito, pois as taxas caíram consideravelmente no segundo período com relação ao primeiro. Os modelos GT1; GT2 e GT3 foram testados para um período de oito anos (duas projeções consecutivas), e todos eles foram capazes de simular a estrutura diamétrica de forma satisfatória. Assim, a cadeia de Markov é uma ferramenta eficiente para projetar a dinâmica de florestas estacionais semideciduais sob intervenções silviculturais, o que reforça sua importância como ferramenta para o manejo florestal.

Existe hoje uma crescente demanda global por ações de restauração de florestas tropicais e um dos principais entraves para a sua concretização é a viabilidade econômica dos projetos. Em cenários em que a restauração ecológica exige a implantação de reflorestamentos de espécies nativas, a produção de madeira torna-se uma alternativa importante para viabilizar economicamente a atividade, tornando-se um meio para atingir o objetivo final da restauração do ecossistema. A carência de conhecimentos técnico-científicos sobre o potencial de produção das espécies nativas em reflorestamentos é, no entanto, um entrave significativo para a sua aplicação em larga escala. Nesse contexto, o presente trabalho teve como objetivo a avaliação do potencial de espécies nativas para a produção de madeira serrada em plantios de restauração florestal. Para isso, avaliou-se o crescimento de 16 espécies nativas com potencial para produção de madeira serrada em 13 reflorestamentos mistos, com idades entre 6 e 96 anos, localizados no domínio da Floresta Estacional Semidecidual no interior do estado de São Paulo. Essas espécies foram avaliadas quanto ao crescimento em diâmetro real, diâmetro potencial, altura total e altura comercial através da modelagem destes parâmetros por técnicas de regressão linear múltipla. Tais modelos foram gerados a partir da correlação existente entre a idade do reflorestamento, o fator ambiental mais influente (relacionado ao solo e clima) e o parâmetro analisado, obtendo-se as perspectivas de desenvolvimento de cada espécie em cenários de baixa, moderada e alta produtividade de sítio. Foram estimadas as projeções do crescimento diamétrico considerando-se todos os indivíduos de cada espécie amostrados nos reflorestamentos, bem como a partir dos 25% melhores indivíduos, visando-se estimar o potencial de crescimento das espécies em condições mais favoráveis de manejo silvicultural. Foi realizada uma análise descritiva das variáveis qualitativas avaliadas em campo (número de fustes, qualidade da árvore e nível de exposição à luz), associadas às alturas total e comercial, verificando-se se as relações existentes entre estes parâmetros são capazes de orientar propostas de manejo silvicultural adequado para as espécies. Várias das espécies nativas estudadas, principalmente Cariniana legalis e C. estrellensis, apresentaram potencial satisfatório de produção de madeira serrada em plantios de restauração florestal, principalmente em condições ambientais favoráveis e se submetidas a manejo silvicultural adequado. No entanto, mesmo sob condições favoráveis, algumas espécies apresentaram crescimento excessivamente lento, desestimulando seu plantio para a produção de madeira. O ciclo de produção de madeira esperado para cada espécie foi muito variável em função das condições ambientais da área em restauração, reforçando a importância de escolha orientada de espécies para cada condição de sítio, do preparo adequado do solo e da consideração dessas condições ambientais para definir os modelos de restauração para fins de exploração de madeira. Da mesma forma, a maior parte dos indivíduos encontrava-se dominada na estrutura da floresta, reforçando a importância do espaçamento, da composição de espécies na vizinhança e de desbastes no potencial de produção de madeira. / There is a growing demand today for tropical forest restoration efforts, and one of the main barriers for accomplishing this demand is the economic viability of the projects. In scenarios where ecological restoration requires the implementation of native species plantings, the exploitation of timber can be an important alternative to make restoration financially viable, as a way to achieve the ultimate goal of the restoration of the ecosystem. The lack of technical and scientific knowledge about the potential of timber production by native species in reforestation is, however, an important obstacle for applying this alternative at the large scale. In this context, we evaluated the growth of 16 native species with potential for timber production in 13 restoration plantings, with ages ranging between 6 and 96 years, located at the Seasonal Semideciduous Forest of inland São Paulo State, southeastern Brazil. These species were evaluated regarding their growth in real and potential diameter, as well their total and commercial height, by modeling these parameters through multiple linear regressions. These models were generated based on the correlation among the age of the restoration planting, the most influent environmental factor (related to climate and soil) and the analyzed parameters, allowing us to obtaining the perspectives of growth for each species in scenarios of low, moderate and high site productivity. We established projections of growth in diameter considering all individuals sampled for each species in the reforestations, as well as the best 25% individuals, as means of estimating the potential growth of each species in more favorable scenarios of silvicultural management. We carried out a descriptive analysis of the qualitiative variables evaluated in the field (number of boles, quality of the tree and exposition to sun light), which were associated to the total and commercial heights, to verify if the relationship between these parameters may guide proposals of suitable silvicultural management for these species. Many of the studied species, especially Cariniana legalis and C. estrellensis, showed a satisfactory potential of timber production in restoration plantings, mainly in more favorable site conditions and under adequate silvicultural treatments, However, even under favorable conditions, some species presented growth rates excessively slow, which discourage their planting for timber production. The expected cycle of timber production for each species was highly variable according to the environmental characteristics of the site, thus reinforcing the importance to guide the selection of species for each site condition, to accomplish a suitable soil preparation before planting and to consider these site conditions for designing models of restoration for timber exploitation. In addition, most trees were dominated in the forest structure, which reinforces the importance of spacing, species composition of the vicinity and thinning in the potential of timber production.

Esta tese compara as dinâmicas de uma floresta explorada com exploração de impacto reduzido (EIR) com outra explorada convencionalmente (EC) numa propriedade denominada Fazenda Agrosete (3º S, 50º W), em Paragominas, no nordeste do Estado do Pará. Os estudos foram realizados numa área de 210 ha, dos quais 75 ha foram explorados convencionalmente, 105 ha foram explorados com exploração de impacto reduzido e 30 ha não explorados (testemunha). A área foi explorada em 1993 e monitorada até 2000. A avaliação da dinâmica das espécies madeireiras nesses três tratamentos foi feita em 10 clareiras selecionadas em cada tratamento. A análise dos efeitos dos tratamentos silviculturais sobre a regeneração de valor madeireiro foi feita em 60 clareiras selecionadas aleatoriamente em cada tratamento. A análise da diversidade de espécies arbóreas foi feita numa área de 5,25 ha delimitada em cada tipo de exploração. E, finalmente, para analisar a dinâmica de espécies arbóreas estabeleceram-se 49 parcelas de 0,5 ha em cada tratamento. Os resultados demonstraram que: i) considerando a regeneração abaixo de 10 cm de DAP não existiram diferenças significativas na dinâmica da regeneração entre as clareiras de exploração de impacto reduzido e as de exploração convencional; ii) os tratamentos silviculturais causaram efeitos positivos sobre o crescimento de espécies de valor comercial presente nas clareiras crescendo de 75% a 115% a mais do que as não tratadas; iii) considerando as espécies arbóreas acima de 10 cm de DAP, a diversidade de espécies reduziu com a colheita e tem apresentado uma boa recuperação, inclusive na área explorada de madeira convencional; iv) a mortalidade de árvores foi menor e o recrutamento foi maior no tratamento de exploração de impacto reduzido em comparação com os tratamentos de exploração convencional e com a testemunha; v) a taxa anual de mortalidade na área de exploração convencional foi 70% superior ao da área com exploração de impacto reduzido; e vi) o crescimento das árvores no tratamento exploração de impacto reduzido foi 5,5 vezes maior do que o crescimento no tratamento de exploração convencional. Esses resultados demonstram que a exploração de impacto reduzido reduz bastante os efeitos negativos sobre a diversidade de espécies, além de promover o crescimento das árvores. Dessa maneira, as florestas exploradas com técnicas de exploração de impacto reduzido se recuperarão num período relativamente menor do que as florestas exploradas convencionalmente, ao mesmo tempo em que estarão mantendo a conservação da biodiversidade florestal. Nossa conclusão é que a exploração de impacto reduzido pode ser uma alternativa ao modelo de exploração atualmente praticado na Amazônia. / This thesis compares the dynamics of forest harvested using reducing impact logging (RIL) and conventional logging forest (CL) at the property known as Fazenda Agrosete (3º S, 50º W), in Paragominas, in the north-east of the State of Pará. The studies were carried out in an area of 210 ha, of which 75 ha were conventional logging, 105 ha were logged using reducing impact logging (RIL) and 30 ha was an unlogged control. The area was harvested in 1993 and monitored until 2000. The evaluation of the dynamics of timber species in these three treatments was carried out in 10 gaps selected in each treatment. The analysis of the effect of silvicultural treatments on regeneration of timber value was done in 60 randomly selected gaps were for each species. The analysis of tree species diversity was done in an area of 5.25 ha delimited for each type of logging. And, finally, to analyse the dynamics of tree species 49 lots, each measuring 0.5 ha were established in each treatment. The results demonstrate that: i) considering regeneration below 10 cm DBH there were no significant differences in regeneration dynamics between the gaps in the reducing impact logging (RIL) area and those in the conventional logging area; ii) the silvicultural treatments caused positive effects on growth of commercially valuable species present in the gaps; grew 75% a 115% more than non-managed regeneration; iii) considering tree species above 10 cm DBH, species diversity was reduced with harvest and has shown good recovery, including in the conventional harvested area; iv) tree mortality was lower and recruitment was higher in the reducing impact logging (RIL) treatment in comparison with conventional logging and witness treatments; v) the annual mortality rate in the conventional logging area was 70% higher than that of the area under reducing impact logging (RIL); and vi) growth of trees in the management treatment was 5.5 times higher than growth under the conventional logging treatment. These results demonstrate that reducing impact logging (RIL) considerably reduces negative effects on species diversity, besides promoting tree growth. In this way, forests logged using reducing impact logging (RIL) techniques will recover in a relatively shorter period than forests that are conventional logging, while at the same time maintaining forest biodiversity. Our conclusion is that reducing impact logging (RIL) can be an alternative to the logging model currently practised in Amazon.

Esta tese compara as dinâmicas de uma floresta explorada com exploração de impacto reduzido (EIR) com outra explorada convencionalmente (EC) numa propriedade denominada Fazenda Agrosete (3º S, 50º W), em Paragominas, no nordeste do Estado do Pará. Os estudos foram realizados numa área de 210 ha, dos quais 75 ha foram explorados convencionalmente, 105 ha foram explorados com exploração de impacto reduzido e 30 ha não explorados (testemunha). A área foi explorada em 1993 e monitorada até 2000. A avaliação da dinâmica das espécies madeireiras nesses três tratamentos foi feita em 10 clareiras selecionadas em cada tratamento. A análise dos efeitos dos tratamentos silviculturais sobre a regeneração de valor madeireiro foi feita em 60 clareiras selecionadas aleatoriamente em cada tratamento. A análise da diversidade de espécies arbóreas foi feita numa área de 5,25 ha delimitada em cada tipo de exploração. E, finalmente, para analisar a dinâmica de espécies arbóreas estabeleceram-se 49 parcelas de 0,5 ha em cada tratamento. Os resultados demonstraram que: i) considerando a regeneração abaixo de 10 cm de DAP não existiram diferenças significativas na dinâmica da regeneração entre as clareiras de exploração de impacto reduzido e as de exploração convencional; ii) os tratamentos silviculturais causaram efeitos positivos sobre o crescimento de espécies de valor comercial presente nas clareiras crescendo de 75% a 115% a mais do que as não tratadas; iii) considerando as espécies arbóreas acima de 10 cm de DAP, a diversidade de espécies reduziu com a colheita e tem apresentado uma boa recuperação, inclusive na área explorada de madeira convencional; iv) a mortalidade de árvores foi menor e o recrutamento foi maior no tratamento de exploração de impacto reduzido em comparação com os tratamentos de exploração convencional e com a testemunha; v) a taxa anual de mortalidade na área de exploração convencional foi 70% superior ao da área com exploração de impacto reduzido; e vi) o crescimento das árvores no tratamento exploração de impacto reduzido foi 5,5 vezes maior do que o crescimento no tratamento de exploração convencional. Esses resultados demonstram que a exploração de impacto reduzido reduz bastante os efeitos negativos sobre a diversidade de espécies, além de promover o crescimento das árvores. Dessa maneira, as florestas exploradas com técnicas de exploração de impacto reduzido se recuperarão num período relativamente menor do que as florestas exploradas convencionalmente, ao mesmo tempo em que estarão mantendo a conservação da biodiversidade florestal. Nossa conclusão é que a exploração de impacto reduzido pode ser uma alternativa ao modelo de exploração atualmente praticado na Amazônia. / This thesis compares the dynamics of forest harvested using reducing impact logging (RIL) and conventional logging forest (CL) at the property known as Fazenda Agrosete (3º S, 50º W), in Paragominas, in the north-east of the State of Pará. The studies were carried out in an area of 210 ha, of which 75 ha were conventional logging, 105 ha were logged using reducing impact logging (RIL) and 30 ha was an unlogged control. The area was harvested in 1993 and monitored until 2000. The evaluation of the dynamics of timber species in these three treatments was carried out in 10 gaps selected in each treatment. The analysis of the effect of silvicultural treatments on regeneration of timber value was done in 60 randomly selected gaps were for each species. The analysis of tree species diversity was done in an area of 5.25 ha delimited for each type of logging. And, finally, to analyse the dynamics of tree species 49 lots, each measuring 0.5 ha were established in each treatment. The results demonstrate that: i) considering regeneration below 10 cm DBH there were no significant differences in regeneration dynamics between the gaps in the reducing impact logging (RIL) area and those in the conventional logging area; ii) the silvicultural treatments caused positive effects on growth of commercially valuable species present in the gaps; grew 75% a 115% more than non-managed regeneration; iii) considering tree species above 10 cm DBH, species diversity was reduced with harvest and has shown good recovery, including in the conventional harvested area; iv) tree mortality was lower and recruitment was higher in the reducing impact logging (RIL) treatment in comparison with conventional logging and witness treatments; v) the annual mortality rate in the conventional logging area was 70% higher than that of the area under reducing impact logging (RIL); and vi) growth of trees in the management treatment was 5.5 times higher than growth under the conventional logging treatment. These results demonstrate that reducing impact logging (RIL) considerably reduces negative effects on species diversity, besides promoting tree growth. In this way, forests logged using reducing impact logging (RIL) techniques will recover in a relatively shorter period than forests that are conventional logging, while at the same time maintaining forest biodiversity. Our conclusion is that reducing impact logging (RIL) can be an alternative to the logging model currently practised in Amazon.