RESUMEN
The Amazon forest contains globally important carbon stocks, but in recent years, atmospheric measurements suggest that it has been releasing more carbon than it has absorbed because of deforestation and forest degradation. Accurately attributing the sources of carbon loss to forest degradation and natural disturbances remains a challenge because of the difficulty of classifying disturbances and simultaneously estimating carbon changes. We used a unique, randomized, repeated, very high-resolution airborne laser scanning survey to provide a direct, detailed, and high-resolution partitioning of aboveground carbon gains and losses in the Brazilian Arc of Deforestation. Our analysis revealed that disturbances directly attributed to human activity impacted 4.2% of the survey area while windthrows and other disturbances affected 2.7% and 14.7%, respectively. Extrapolating the lidar-based statistics to the study area (544,300 km2), we found that 24.1, 24.2, and 14.5 Tg C y-1 were lost through clearing, fires, and logging, respectively. The losses due to large windthrows (21.5 Tg C y-1) and other disturbances (50.3 Tg C y-1) were partially counterbalanced by forest growth (44.1 Tg C y-1). Our high-resolution estimates demonstrated a greater loss of carbon through forest degradation than through deforestation and a net loss of carbon of 90.5 ± 16.6 Tg C y-1 for the study region attributable to both anthropogenic and natural processes. This study highlights the role of forest degradation in the carbon balance for this critical region in the Earth system.
Asunto(s)
Carbono , Conservación de los Recursos Naturales , Bosques , Brasil/epidemiología , Carbono/metabolismo , Humanos , Árboles/crecimiento & desarrollo , Ciclo del CarbonoRESUMEN
For more than three decades, major efforts in sampling and analyzing tree diversity in South America have focused almost exclusively on trees with stems of at least 10 and 2.5 cm diameter, showing highest species diversity in the wetter western and northern Amazon forests. By contrast, little attention has been paid to patterns and drivers of diversity in the largest canopy and emergent trees, which is surprising given these have dominant ecological functions. Here, we use a machine learning approach to quantify the importance of environmental factors and apply it to generate spatial predictions of the species diversity of all trees (dbh ≥ 10 cm) and for very large trees (dbh ≥ 70 cm) using data from 243 forest plots (108,450 trees and 2832 species) distributed across different forest types and biogeographic regions of the Brazilian Amazon. The diversity of large trees and of all trees was significantly associated with three environmental factors, but in contrasting ways across regions and forest types. Environmental variables associated with disturbances, for example, the lightning flash rate and wind speed, as well as the fraction of photosynthetically active radiation, tend to govern the diversity of large trees. Upland rainforests in the Guiana Shield and Roraima regions had a high diversity of large trees. By contrast, variables associated with resources tend to govern tree diversity in general. Places such as the province of Imeri and the northern portion of the province of Madeira stand out for their high diversity of species in general. Climatic and topographic stability and functional adaptation mechanisms promote ideal conditions for species diversity. Finally, we mapped general patterns of tree species diversity in the Brazilian Amazon, which differ substantially depending on size class.
Asunto(s)
Aclimatación , Viento , Brasil , Bosque Lluvioso , BiodiversidadRESUMEN
The Amazon Forest, the largest contiguous tropical forest in the world, stores a significant fraction of the carbon on land. Changes in climate and land use affect total carbon stocks, making it critical to continuously update and revise the best estimates for the region, particularly considering changes in forest dynamics. Forest inventory data cover only a tiny fraction of the Amazon region, and the coverage is not sufficient to ensure reliable data interpolation and validation. This paper presents a new forest above-ground biomass map for the Brazilian Amazon and the associated uncertainty both with a resolution of 250 meters and baseline for the satellite dataset the year of 2016 (i.e., the year of the satellite observation). A significant increase in data availability from forest inventories and remote sensing has enabled progress towards high-resolution biomass estimates. This work uses the largest airborne LiDAR database ever collected in the Amazon, mapping 360,000 km2 through transects distributed in all vegetation categories in the region. The map uses airborne laser scanning (ALS) data calibrated by field forest inventories that are extrapolated to the region using a machine learning approach with inputs from Synthetic Aperture Radar (PALSAR), vegetation indices obtained from the Moderate-Resolution Imaging Spectroradiometer (MODIS) satellite, and precipitation information from the Tropical Rainfall Measuring Mission (TRMM). A total of 174 field inventories geolocated using a Differential Global Positioning System (DGPS) were used to validate the biomass estimations. The experimental design allowed for a comprehensive representation of several vegetation types, producing an above-ground biomass map varying from a maximum value of 518 Mg ha-1, a mean of 174 Mg ha-1, and a standard deviation of 102 Mg ha-1. This unique dataset enabled a better representation of the regional distribution of the forest biomass and structure, providing further studies and critical information for decision-making concerning forest conservation, planning, carbon emissions estimate, and mechanisms for supporting carbon emissions reductions.
Asunto(s)
Biomasa , Bosques , Tecnología de Sensores Remotos , Brasil , Carbono/análisis , Tecnología de Sensores Remotos/métodos , Clima TropicalRESUMEN
Height measurements are essential to manage and monitor forest biomass and carbon stocks. However, accurate estimation of this variable in tropical ecosystems is still difficult due to species heterogeneity and environmental variability. In this article, we compare and discuss six nonlinear allometric models parameterized at different scales (local, regional and pantropical). We also evaluate the height measurements obtained in the field by the hypsometer when compared with the true tree height. We used a dataset composed of 180 harvested trees in two distinct areas located in the Amapá State. The functional form of the Weibull model was the best local model, showing similar performance to the pantropical model. The inaccuracy detected in the hypsometer estimates reinforces the importance of incorporating new technologies in measuring individual tree heights. Establishing accurate allometric models requires knowledge of ecophysiological and environmental processes that govern vegetation dynamics and tree height growth. It is essential to investigate the influence of different species and ecological gradients on the diameter/height ratio.
Asunto(s)
Biomasa , Bosques , Modelos Biológicos , Árboles/crecimiento & desarrollo , Clima Tropical , BrasilRESUMEN
BACKGROUND: Brazilian Amazon forests contain a large stock of carbon that could be released into the atmosphere as a result of land use and cover change. To quantify the carbon stocks, Brazil has forest inventory plots from different sources, but they are unstandardized and not always available to the scientific community. Considering the Brazilian Amazon extension, the use of remote sensing, combined with forest inventory plots, is one of the best options to estimate forest aboveground biomass (AGB). Nevertheless, the combination of limited forest inventory data and different remote sensing products has resulted in significant differences in the spatial distribution of AGB estimates. This study evaluates the spatial coverage of AGB data (forest inventory plots, AGB maps and remote sensing products) in undisturbed forests in the Brazilian Amazon. Additionally, we analyze the interconnection between these data and AGB stakeholders producing the information. Specifically, we provide the first benchmark of the existing field plots in terms of their size, frequency, and spatial distribution. RESULTS: We synthesized the coverage of forest inventory plots, AGB maps and airborne light detection and ranging (LiDAR) transects of the Brazilian Amazon. Although several extensive forest inventories have been implemented, these AGB data cover a small fraction of this region (e.g., central Amazon remains largely uncovered). Although the use of new technology such as airborne LiDAR cover a significant extension of AGB surveys, these data and forest plots represent only 1% of the entire forest area of the Brazilian Amazon. CONCLUSIONS: Considering that several institutions involved in forest inventories of the Brazilian Amazon have different goals, protocols, and time frames for forest surveys, forest inventory data of the Brazilian Amazon remain unstandardized. Research funding agencies have a very important role in establishing a clear sharing policy to make data free and open as well as in harmonizing the collection procedure. Nevertheless, the use of old and new forest inventory plots combined with airborne LiDAR data and satellite images will likely reduce the uncertainty of the AGB distribution of the Brazilian Amazon.
RESUMEN
Tree stem form in native tropical forests is very irregular, posing a challenge to establishing taper equations that can accurately predict the diameter at any height along the stem and subsequently merchantable volume. Artificial intelligence approaches can be useful techniques in minimizing estimation errors within complex variations of vegetation. We evaluated the performance of Random Forest® regression tree and Artificial Neural Network procedures in modelling stem taper. Diameters and volume outside bark were compared to a traditional taper-based equation across a tropical Brazilian savanna, a seasonal semi-deciduous forest and a rainforest. Neural network models were found to be more accurate than the traditional taper equation. Random forest showed trends in the residuals from the diameter prediction and provided the least precise and accurate estimations for all forest types. This study provides insights into the superiority of a neural network, which provided advantages regarding the handling of local effects.
Asunto(s)
Inteligencia Artificial , Ecosistema , Modelos Teóricos , Árboles , Algoritmos , Brasil , Bosques , Geografía , Redes Neurales de la Computación , Árboles/crecimiento & desarrolloRESUMEN
Very few studies have been devoted to understanding the digital terrain model (DTM) creation for Amazon forests. DTM has a special and important role when airborne laser scanning is used to estimate vegetation biomass. We examined the influence of pulse density, spatial resolution, filter algorithms, vegetation density and slope on the DTM quality. Three Amazonian forested areas were surveyed with airborne laser scanning, and each original point cloud was reduced targeting to 20, 15, 10, 8, 6, 4, 2, 1, 0.75, 0.5 and 0.25 pulses per square meter based on a random resampling process. The DTM from resampled clouds was compared with the reference DTM produced from the original LiDAR data by calculating the deviation pixel by pixel and summarizing it through the root mean square error (RMSE). The DTM from resampled clouds were also evaluated considering the level of agreement with the reference DTM. Our study showed a clear trade-off between the return density and the horizontal resolution. Higher forest canopy density demanded higher return density or lower DTM resolution.
São poucos os estudos dedicados a entender a criação de modelo digital de terreno (MDT) para florestas amazônicas. O MDT tem uma importante função quando o escaneamento laser aerotransportado é usado para estimar a biomassa da vegetação. Examinamos a relação da densidade de pulsos, resolução espacial, algoritmos de filtragem, densidade da vegetação e inclinação do terreno com a qualidade do MDT. Três áreas de floresta amazônica foram sobrevoadas usando LiDAR aerotransportado. Cada nuvem de dados original teve sua densidade reduzida objetivando 20; 15; 10; 8; 6; 4; 2; 1; 0,75; 0,5 e 0,25 pulsos por metro quadrado, utilizando um processo de reamostragem aleatória. Os MDTs das nuvens reamostradas foram comparados com o MDT de referência, produzido a partir da nuvem original, calculando o desvio pixel a pixel e resumindo-o por meio do erro padrão da estimativa (RMSE). Os MDTs das nuvens reamostradas também foram avaliados quanto ao nível de correspondência com o MDT de referência. Houve uma clara compensação entre densidade de pontos e resolução horizontal. Dosséis mais densos exigem uma maior densidade de retornos, ou MDT com menor resolução.