Changes in soil and vegetation attributes during lowland Atlantic forest succession in Brazil

• Ano de defesa: 2018
• Autor(a) principal: Safar, Nathália Vieira Hissa
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa
• Palavras-chave em Português: Ecologia vegetal; Sucessão florestal; Resiliência (Ecologia); Comunidades vegetais; Sequestro de carbono; Solos - Composição; Formigas; Botânica
• Link de acesso: http://www.locus.ufv.br/handle/123456789/21187

Human disturbances, such as fragmentation and clear-cut logging, cause the loss of biodiversity and carbon stock, affect litter production, cause microclimatic changes at the forest floor, which affect soil nutrient dynamics, and consequently affect the ecosystem functioning. After a disturbance the process of secondary succession begins, which involves changes in plant and animal communities (mainly arthropods), carbon stock, litter production and soil properties. The Brazilian Atlantic forest is one of the most diverse and threatened tropical forest in the world, and its landscapes are mainly composed by small fragments that are in some stage of recovery from past human disturbance. These second-growth fragments have an important role in carbon mitigation and biodiversity conservation. To investigate the changes during forest succession and the resilience of lowland rain forests, we established three main objectives: (i) to assess the effects of whole-tree logging forest clearance on soil properties and their dynamics; (ii) to assess the effects of stand age on tree and ant species richness and composition, aboveground and total C pool, and soil nutrients (organic C, P, Al 3+ and sum of bases) and (iii) to estimate the time secondary lowland forests take to reach the mature forests levels for each parameter. This study was conducted in old-growth and second-growth lowland rain forests from northern Espírito Santo State, Brazil. To accomplish the objective (i) we analyzed and compared soil data (pH, P, Al 3+ , K + , Ca 2+ + Mg 2+ ), collected in two moments (1978, 2017), from an old-growth forest (control) and one adjacent forest that was cleared as part of a long-term experiment performed in 1980, at the Vale Natural Reserve (Espírito Santo State, Brazil). To accomplish the objective (ii-iii) we adopted a chronosequence approach with two old-growth and 11 second-growth forests at different ages, distributed in the northern Espírito Santo and southern Bahia States, Brazil. When assessing the effects of forest clearance on soil properties, we found no significant differences between treatments for any soil component (LMM; p>0.05). Thus, both forests showed the same soil dynamic along the 39 years of interval: increasing pH (LMM; p<0.01), and decreasing Al 3+ over the years (LMM; p<0.001), and no changes in P, K + and Ca 2+ + Mg 2+ (LMM; p>0.05). This result suggest that available P, and exchangeable bases K + , Ca 2+ + Mg 2+ may be resistant to whole-tree removal. Further studies comparing our results regarding the effects of forest clearance on soil properties with the long-term tree dynamic of similar forested areas will help to clarify and understand the pattern found. Moreover, when assessing the effects of forest regeneration on forest attributes, we found positive significant relations (p<0.001) between stand age and tree species richness and composition, ant species composition and carbon pool (total and aboveground), suggesting that these parameters can be used as indicators of forest recovery. However, we found no relationship between soil properties and forest regeneration (p>0.05). This study predicts that secondary lowland rain forests would take several years (57-126 yr) to recover species richness and composition, and much longer, 188 yr to recover C pool, and that these forests are potentially sequestering 1.04 Mg C ha -1 yr -1 , thus contributing to the important ecosystem service of CO 2 sequestration. Our findings suggest that ant species richness and soils of lowland forests can be resistant to human disturbances. Finally, our results indicate that lowland second-growth forests are resilient, but it would take approximately 50 to 100 years to recover the original state of biodiversity and ecosystem functioning, evidencing that we urgently need to develop strategies for selecting areas with highest natural regeneration potential.