An increase in atmospheric CO2 levels and global climate changes have led to an increased focus on CO2 capture mechanisms. The in situ quantification and spatial patterns of forest carbon stocks can provide a better picture of the carbon cycle and a deeper understanding of the functions and services of forest ecosystems. This study aimed to determine the aboveground (tree trunks) and belowground (soil and fine roots, at four depths) carbon stocks in a tropical forest in Brazil and to evaluate the spatial patterns of carbon in the three different compartments and in the total stock. Census data from a semideciduous seasonal forest were used to estimate the aboveground carbon stock. The carbon stocks of soil and fine roots were sampled in 52 plots at depths of 0-20, 20-40, 40-60, and 60-80 cm, combined with the measured bulk density. The total estimated carbon stock was 267.52 Mg ha-1, of which 35.23% was in aboveground biomass, 63.22% in soil, and 1.54% in roots. In the soil, a spatial pattern of the carbon stock was repeated at all depths analyzed, with a reduction in the amount of carbon as the depth increased. The carbon stock of the trees followed the same spatial pattern as the soil, indicating a relationship between these variables. In the fine roots, the carbon stock decreased with increasing depth, but the spatial gradient did not follow the same pattern as the soil and trees, which indicated that the root carbon stock was most likely influenced by other factors.
Volumetric equations is one of the main tools for quantifying forest stand production, and is the basis for sustainable management of forest plantations. This study aimed to assess the quality of the volumetric estimation of Eucalyptus spp. trees using a mixed-effects model, artificial neural network (ANN) and support-vector machine (SVM). The database was derived from a forest stand located in the municipalities of Bom Jardim de Minas, Lima Duarte and Arantina in Minas Gerais state, Brazil. The volume of 818 trees was accurately estimated using Smalian’s Formula. The Schumacher and Hall model was fitted by fixed-effects regression and by including multilevel random effects. The mixed model was fitted by adopting 14 different structures for the variance and covariance matrix. The best structure was selected based on the Akaike Information Criterion, Maximum Likelihood Ratio Test and Vuong’s Closeness Test. The SVM and ANN training process considered diameter at breast height and total tree height to be the independent variables. The techniques performed satisfactorily in modeling, with homogeneous distributions and low dispersion of residuals. The quality analysis criteria indicated the superior performance of the mixed model with a Huynh-Feldt structure of the variance and covariance matrix, which showed a decrease in mean relative error from 13.52% to 2.80%, whereas machine learning techniques had error values of 6.77% (SVM) and 5.81% (ANN). This study confirms that although fixed-effects models are widely used in the Brazilian forest sector, there are more effective methods for modeling dendrometric variables.
The increasing awareness of global climate change has drawn attention to the role of forests as mitigators of this process as they act as carbon sinks to the atmosphere. Understanding the process of carbon storage in forests and its drivers, as well as presenting consistent models for their estimation, is a current demand. In this sense, the aim of this study was to evaluate the performance of machine learning techniques: support vector machines (SVM) and to propose a new nonlinear model extracted from the training of an artificial neural network (ANN) in the modeling of above ground carbon stock in a secondary semideciduous forest. SVM and ANN construction and training process considered independent variables selected by stepwise: minimum DBH (diameter of breast height - 1.3 m), maximum DBH, mean DBH, total height and number of trees, all by plot. SVM and the model extracted from ANN were applied to the data set intended for validation. Both techniques presented satisfactory performance in modeling carbon stock by plot, with homogeneous distribution and low dispersion of residues and predicted values close to those observed. Analysis criteria indicated superior performance of the model extracted from the artificial neural network, which presented a mean relative error of 6.94 %, while the support vector machine presented 13.52 %, combined with lower bias values and higher correlation between predictions and observations.
This study aimed at evaluating the performance of different models based on Artificial neural networks (ANN) to estimate the total height of eucalyptus trees (Eucalyptus spp.), reducing the number of measurements in the field. Forty-eight ANN were tested, different from each other by the number of trees used as training sample, number of trees used to calculate the dominant height and use of variables (a) categorical, (b) categorical and continuous and (c) continuous, except for the diameter at 1.30 meters above the ground (DBH), used in all combinations. Estimates of height obtained by ANN were compared with values observed and estimates obtained by a hypsometric model. The ANN that showed the best results were used for the height estimation in forest inventory data for further application in the Schumacher and Hall volumetric model. The proposed models were efficient to estimate the total height of eucalyptus trees and allowed the expressive reduction of the number of trees to be measured in forest inventory. The best model found is composed of five trees as training sample, one as test sample and one as validation sample; dominant height coming from the height of the tallest tree in the plot; categorical variable Clone and continuous variables DBH, DBH dominant and basal area of the plot.
This study aimed to analyze relations between the distribution of tree species and variations in soil texture, fertility and organic matter levels in a rupestrian cerrado area located in Diamantina-MG. Individual trees with circumference at soil level (PC) ≥ 3 cm were sampled in 10 plots of 20x50m. Samples of surface soil (0.00-0.20 m of depth) were collected from each plot and analyzed for their chemical and physical properties. Through canonical correspondence analysis, Eremanthus incanus, Kielmeyera lathrophyton, Terminalia sp and Vochysia discolor show a stronger relation to higher potassium and remaining phosphorus levels; the species Xylopia sericea, Protium spruceanum and Protium heptaphyllum show a stronger relation to higher levels of organic matter and total cation exchange capacity; the species Roupala montana, Qualea grandiflora, and Erythroxylum suberosum grew uniformly in all plots. There is relation between species distribution and chemical and physical soil variables in the area.
O inventário florestal constitui uma prática fundamental para quantificar e qualificar o potencial de uma floresta. Intensidade de amostragem, tamanho e forma das parcelas são pontos relevantes no planejamento dessa atividade, pois afetam diretamente os custos e tempo de coleta de dados, além de influenciar na sua precisão. Este estudo teve como objetivo avaliar, em termos de precisão, custos e tempo, a utilização de diferentes tamanhos, formas e arranjos de parcelas amostrais na realização do inventário florestal pré-corte de Eucalyptus spp. A pesquisa foi realizada em uma fazenda no norte de Minas Gerais, Brasil, onde foram realizados o censo florestal e a cubagem rigorosa. Foram considerados 16 cenários: dois processos de amostragem (amostragem sistemática e amostragem casual simples), quatro tamanhos de parcela (200, 300, 400 e 500m²) e duas formas (circular e retangular). A intensidade amostral foi fixada em 5 parcelas para todos os cenários. Realizou-se 10.000 simulações pelo método de Monte Carlo, verificando a probabilidade de os valores estarem próximos à média. A análise combinada entre erro, custos e tempo de medição apontou que, para um plantio de eucalipto aos 8 anos de idade, a amostragem casual, com parcelas circulares de 300m² representa o cenário amostral ótimo.