Solos e Nutrição de Plantas
URI permanente para esta coleçãohttps://locus.ufv.br/handle/123456789/175
Navegar
2 resultados
Resultados da Pesquisa
Item Harvest residues management and silvicultural operations impact on soil physical and organic matter quality of eucalypt plantations(Universidade Federal de Viçosa, 2017-01-26) Ferreira, Gabriel William Dias; Soares, Emanuelle Mercês Barros; http://lattes.cnpq.br/1555026111282397Brazil is among the world’s largest timber producers and Eucalyptus planted forests are responsible for a significant portion of wood products supply. Furthermore, these forests play an important role on mitigation of increasing atmospheric greenhouse gases. The period between harvesting and new rotation establishment, i.e. its first two or three years, concentrates the major part of forest mechanical operations that may define achievable productivity while causing harmful impacts on soil physical and organic properties. This thesis is divided in three chapters that evaluated and tried to provide a better understanding of these operations impact on eucalypt forest dynamics and soil properties. The first chapter approached how the management of eucalypt harvest residues (HR) could alter its decomposition pattern and the dynamics of two soil organic matter (SOM) fractions (Particulate Organic Matter – POM, and Associated to Minerals – MAOM). The experiment was set up in an area where natural grasslands of Pampa Biome in the Rio Grande do Sul State, Brazil, have been recently converted to eucalypt forests. Removal of all residues (-R), only bark removal (-B) and maintenance of all residues (+B), all of them combined with external 200 kg ha -1 of N addition (+N) or not (-N), were simulated in PVC micro-plots in a 3x2 full factorial with 4 replications. A 10 % 15 N-enriched solution was used as N source to track the role of mineral-N in the process. Whole micro- plots were sampled at 0, 3, 6, 12 and 36 mo. after experiment establishment and taken to the lab for further analysis. All sampling times were used to build decomposition curve, while C, δ 13 C, N and 15 N content associated with both SOM fractions were determined on 36 mo. samples in 0 -1 and 1 -5 cm soil layers. In these same layers, SOM fractions from 12 and 36 mo. samples were characterized with Pyrolysis associated with Gas Chromatography/Mass Spectrometry (Py-GC/MS) to obtain a better understanding of SOM formation pathway under different HR management. Almost 80 % of HR have been decomposed until 3 yr. Bark (p<0.05) and mineral-N (p=0.06) presence slowed down decomposition. Overall, changes in C and N content and δ 13 C due to residues presence were more expressive in 0-1 cm layer and POM fraction. +B tended to increase HR contribution, but its effect was more distinguished in -N treatments. Mineral-N fraction was always higher when HR were present, particularly in +B treatments, but its contribution to SOM fractions was always smaller than 4 %. Py-GC/MS products revealed a direct role of HR on POM formation, but their effect on MAOM seems to be indirect by altering microbial composition and its products. In summary, we showed how HR management drives SOM fractions chemistry and we conclude that a sustainable management of HR can enhance soil C accrual. In the second chapter was evaluated how two different harvest systems would change soil physical properties, soil C content and soil CO 2 efflux, and initial tree growth under two different silvicultural system. To this end, two adjacent stands located in Estrela do Sul/MG were selected. Each stand was harvested and logged with a different system (Feller + Forwarder – F+F; and Feller + Skidder – F+S) and soil physical properties [Soil Bulk Density (Ds), Micro (Mi), Macro (Ma) and Total Porosity (TP), and Penetration Resistance (PR)], SOM properties [Labile- C, C associated with Particulate (C-POM) and Mineral (C-MAOM) fractions] soil CO 2 efflux and stump mortality rate, were assessed after harvesting operations and compared with reference (before harvesting). Afterwards, each stand was divided in coppice and replanting, and we followed soil CO 2 efflux after planting and one year after planting, when trees height (H) and SOM properties were also measured in both areas and system to evaluate how areas would recover from harvesting impacts. All soil variables were assessed at planting and inter-planting row positions. Both systems affected soil density and soil pore configuration, but in different ways. F+F concentrates traffic at inter- planting row position, and therefore causes a slightly higher compaction in this region, while F+S does not follow traffic routes and impacts planting-row similarly, resulting in higher stump mortality. Harvesting operations increased Ds, Mi and PR and reduced Ma and TP at all layers. Soil CO 2 efflux was not affected by harvesting operations. Differences in soil CO 2 were found after planting, when it was higher in F+S system, that also presented higher labile-C and C-POM. One year after planting, mortality rate was still higher under coppice in F+S system, that also presented lower trees. In turn, trees were higher in replanting in this area. Soil respiration behaved similarly, i.e., higher in F+F under coppice and lower under replanting. Overall, after one year coppice system presented higher Labile-C, C-POM and C-MAOM. We concluded that each harvest system affects row and inter-row differently, creating different soil functional zones inside same area, that might be enhanced by the next silvicultural system chosen, and should be observed when assessing ecosystem services and site condition. Lastly, the third chapter evaluated the effect of land use change from natural grasslands of Pampa Biome to eucalypt plantations, as well as N fertilization effects on initial eucalypt growth, fine-root biomass (FRB) and its spatial distribution, and C, δ 13 C and N content associated with SOM fractions (POM and MAOM). 4 N-levels were tested (24, 36, 48 e 108 kg ha^-1 of N) on initial tree growth (until 2 yr.). Afterwards, representative trees were chosen to evaluate FRB until 40 cm depth, and soil samples also until 40 cm depth were collected for SOM evaluation. Positive effect of N on tree growth (diameter and height) was seen initially after fertilization, i.e., 1.5 yr., whereas at 2 yr. N effects were seen only at trees height. The FRB fitted regression showed increase of FRB until 56 kg ha^-1 of N, and after that level a decrease in FRB was observed, and the highest N level used resulted in the lowest FRB. Both horizontal and vertical anisotropy in fine-root distribution were observed, and besides differences among N levels, we could not see a clear relation between N fertilization and fine-root spatial distribution. Overall, land use change to eucalypt plantations increased soil C content, particularly in top-soil layers. 36 kg ha^-1 resulted in higher C-POM in 0 -10, while 48 kg ha^-1 resulted in higher C-MAOM in this layer. C and N dynamics were tightly correlated, especially in MAOM fraction. C-POM was positively correlated with FRB. Tillage had a strong control on soil C and N stocks, enhancing C deposition and turnover at row (ridge) region, most likely for favoring roots development in this region. Therefore, it is shown that N fertilization may alter initial tree growth, but its effects don’t seem to last longer. Nevertheless, N effects can be reflected on fine-root biomass and distribution and C and N of SOM fractions. We hope that our findings could guide the adoption of proper management practices in eucalypt plantations in Brazil, enhancing the productivity and sustainability of these forests.Item Resíduos da colheita do eucalipto: ciclagem de nutrientes e uso da espectroscopia do infravermelho próximo para predição da composição química(Universidade Federal de Viçosa, 2012-06-26) Ferreira, Gabriel William Dias; Silva, Ivo Ribeiro da; http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4799432D0; Barros, Nairam Félix de; http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4783694P8; Soares, Emanuelle Mercês Barros; http://lattes.cnpq.br/2680869902625278; http://lattes.cnpq.br/1555026111282397; Vergütz, Leonardus; http://lattes.cnpq.br/1282294478259902; Neves, Júlio César Lima; http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4783076D4; Paiva, Haroldo Nogueira de; http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4788177J6A grande demanda por produtos oriundos das florestas de eucalipto faz com que se busquem formas adequadas de manejá-las, a fim de se obter elevadas produções em ciclos cada vez mais curtos. A grande exportação de nutrientes na colheita dos povoamentos gera preocupações sobre a sustentabilidade do ambiente e da produção dos ciclos subsequentes. A manutenção dos resíduos na área surge como alternativa para minimizar essa exportação, liberar gradualmente os nutrientes à próxima rotação e favorecer os estoques de matéria orgânica do solo (MOS), de papel fundamental nos solos de regiões tropicais. Esses processos são governados por fatores de clima, solo e qualidade do material vegetal. A avaliação da composição do material vegetal é feita por análises químicas trabalhosas e com grande consumo de tempo e reagentes. O uso de métodos espectroscópicos pode tornar esse processo mais simples, com resultados de boa confiabilidade. Os objetivos desse estudo foram avaliar a liberação de nutrientes pela decomposição dos resíduos da colheita do eucalipto e testar a viabilidade da espectroscopia do infravermelho próximo (NIRS) para predizer a composição química dos mesmos. As amostras utilizadas foram obtidas de dois experimentos distintos. Experimento 1: oito tratamentos definidos por um fatorial 2x2x2, sendo presença ou não da casca, com ou sem aplicação de N, mantidos em superfície ou incorporados, coletados em cinco tempos, até três anos em dez locais diferentes do Brasil, distribuídos por quatro estados. Foram determinados os teores de Ca, Mg, N, P, K e S. Não foi observada influência significativa da aplicação de N na liberação de nutrientes. A presença da casca aumentou os estoques de nutrientes na área e acelerou a liberação. A incorporação também aumentou a velocidade de liberação dos nutrientes. Os nutrientes tiveram comportamentos semelhantes nos locais avaliados, contudo, as taxas foram diferentes. Em geral, K foi o nutriente mais rapidamente liberado, enquanto o Ca teve liberação mais lenta. Mg, S, N, e P tiveram liberação mais gradual, tendendo a acompanhar a taxa de decomposição dos resíduos. Para todos os nutrientes, com exceção do K, foram observadas algumas fases de imobilização. Experimento 2: litter bags alocados em cinco povoamentos de eucalipto localizados em regiões distintas do Extremo Sul da BA, sendo quatro tratamentos, resultantes da combinação de resíduos de árvores submetidas ou não a altas doses de N, e presença ou ausência de casca, coletados em cinco tempos, até um ano após a instalação do experimento. Foram determinados os teores de C, N, lignina insolúvel Klason, lignina solúvel, extrativos e holocelulose. Foram selecionadas, com base na distribuição de frequência, 308 amostras que foram utilizadas na análise NIRS. Foram testadas 21 formas de pré-processamento espectral, variando entre a correção do espalhamento e aplicação de derivadas, até o segundo nível. Em cada uma das combinações, aplicou-se a técnica de análise de componentes principais (PCA) para identificar outliers e definir os grupos de calibração e validação. Com uso da mPLS (regressão modificada dos mínimos quadrados parciais) relacionou-se os espectros obtidos com os valores determinados pelos métodos químicos. Os modelos foram avaliados pelos resultados da validação cruzada e validação externa. Os modelos obtidos foram considerados eficientes na predição de todas as características, em todos os tempos, sendo para N e extrativos os mais acurados. Quando se utilizou formas diferenciadas de pré-processamento espectral para cada componente, ajustaram-se modelos com maior capacidade preditiva. O uso de apenas uma forma de pré-processamento (StdMSC 2a d.) possibilitou predizer o comportamento de todos os constituintes avaliados com boa precisão, apesar de não ter sido identificado como a melhor forma de pré-processamento para cada constituinte individualmente. Assim, verificou-se que é viável a utilização do NIRS para predição das transformações sofridas por diferentes componentes dos resíduos da colheita do eucalipto em decomposição.