Search

Drone-based remote sensing is a promising new technology that combines the benefits of ground-based and satellite-derived forest monitoring by collecting fine-scale data over relatively large areas in a cost-effective manner. Here, we explore the potential of the GatorEye drone-lidar system to monitor tropical forest succession by canopy structural attributes including canopy height, spatial heterogeneity, gap fraction, leaf area density (LAD) vertical distribution, canopy Shannon index (an index of LAD), leaf area index (LAI), and understory LAI. We focus on these variables? relationship to aboveground biomass (AGB) stocks and species diversity. In the Caribbean lowlands of northeastern Costa Rica, we analyze nine tropical forests stands (seven secondgrowth and two old-growth). Stands were relatively homogenous in terms of canopy height and spatial heterogeneity, but not in their gap fraction. Neither species density nor tree community Shannon diversity index was significantly correlated with the canopy Shannon index. Canopy height, LAI, and AGB did not show a clear pattern as a function of forest age. However, gap fraction and spatial heterogeneity increased with forest age, whereas understory LAI decreased with forest age. Canopy height was strongly correlated with AGB. The heterogeneous mosaic created by successional forest patches across human-managed tropical landscapes can now be better characterized. Drone-lidar systems offer the opportunity to improve assessment of forest recovery and develop general mechanistic carbon sequestration models that can be rapidly deployed to specific sites, an essential step for monitoring progress within the UN Decade on Ecosystem Restoration. Made available in DSpace on 2020-08-01T11:12:46Z (GMT). No. of bitstreams: 1 27015.pdf: 1386408 bytes, checksum: cce9ebab5fed640e715ad6387e973c5a (MD5) Previous issue date: 2020

Information regarding the spatial distribution of palm trees in tropical forests is crucial for commercial exploitation and management. However, spatially continuous knowledge of palms occurrence is scarce and difficult to obtain with conventional approaches such as field inventories. Here, we developed a new method to map Amazonian palm species at the individual tree crown (ITC) level using RGB images acquired by a low-cost unmanned aerial vehicle (UAV). Our approach is based on morphological operations performed in the score maps of palm species derived from a fully convolutional neural network model. We first constructed a labeled dataset by dividing the study area (135 ha within an old-growth Amazon forest) into 28 plots of 250 m×150 m. Then, we manually outlined all palm trees seen in RGB images with 4 cm pixels. We identified three palm species: Attalea butyracea, Euterpe precatoria and Iriartea deltoidea. We randomly selected 22 plots (80%) for training and six plots (20%) for testing. We changed the plots for training and testing to evaluate the variabilityn, in the classification accuracy and assess model generalization. Our method outperformed the average producer?s accuracy of conventional patch-wise semantic segmentation (CSS) in 4.7%. Moreover, our method correctly identified, on average, 34.7 percentage points more ITCs than CSS, which tended to merge trees that are close to each other. The producer's accuracy of A. butyracea, E. precatoria and I. deltoidea was 78.6 ± 5.5%, 8.6 ± 1.4% and 96.6 ± 3.4%, respectively. Fortunately, one of the most exploited and commercialized palm species in the Amazon (E. precatoria, a.k.a, Açaí) was mapped with the highest classification accuracy. Maps of E. precatoria derived from low-cost UAV systems can support management projects and community-based forest monitoring programs in the Amazon. Made available in DSpace on 2020-08-01T11:12:33Z (GMT). No. of bitstreams: 1 27014.pdf: 15845790 bytes, checksum: 2169ad4041cb991e84ec0efef9ffbeab (MD5) Previous issue date: 2020

Tropical forests are often located in dicult-to-access areas, which make high-quality forest structure information dicult and expensive to obtain by traditional field-based approaches. LiDAR (acronym for Light Detection And Ranging) data have been used throughout the world to produce time-ecient and wall-to-wall structural parameter estimates for monitoring in native and commercial forests. In this study, we compare products and aboveground biomass (AGB) estimations from LiDAR data acquired using an aircraft-borne system in 2015 and data collected by the unmanned aerial vehicle (UAV)-based GatorEye Unmanned Flying Laboratory in 2017 for ten forest inventory plots located in the Chico Mendes Extractive Reserve in Acre state, southwestern Brazilian Amazon. The LiDAR products were similar and comparable among the two platforms and sensors. Principal dierences between derived products resulted from the GatorEye system flying lower and slower and having increased returns per second than the aircraft, resulting in a much higher point density overall (11.3 1.8 vs. 381.2 58 pts/m2). Dierences in ground point density, however, were much smaller among the systems, due to the larger pulse area and increased number of returns per pulse of the aircraft system, with the GatorEye showing an approximately 50% higher ground point density (0.27 0.09 vs. 0.42 0.09). The LiDAR models produced by both sensors presented similar results for digital elevation models and estimated AGB. Our results validate the ability for UAV-borne LiDAR sensors to accurately quantify AGB in dense high-leaf-area tropical forests in the Amazon. We also highlight new possibilities using the dense point clouds of UAV-borne systems for analyses of detailed crown structure and leaf area density distribution of the forest interior. Made available in DSpace on 2020-06-02T04:38:55Z (GMT). No. of bitstreams: 1 27002.pdf: 8268657 bytes, checksum: 92fd75c7b1786acaf00f080246b7eef4 (MD5) Previous issue date: 2020

We are in an unprecedented moment for promoting forest restoration globally, with international and regional pledges to restore at least 350 million hectares by 2030. To achieve these ambitious goals, it is necessary to go beyond traditional plot-scale assessments and develop cost-effective technologies that can monitor the structure and function of restored forests at much broader scales. Lidar remote sensing in unmanned aerial vehicle (UAV) platforms can be an agile and autonomous method for monitoring forest restoration projects, especially under conditions when information updates are frequently needed in relatively small areas or, when using an airplane-borne lidar system may be not financially viable. Here, we explored the potential of an UAV-borne lidar system to assess the outcomes of a mixed-species restoration plantation experiment, designed to maximize aboveground biomass (AGB) accumulation. The experiment was established in Brazil’s Atlantic Forest, with 20 native tree species, by combining two levels of planting density and two management levels, totaling four treatment combinations and one control (plots left over for natural regeneration). We analyzed three structural variables from lidar data (canopy height, gap fraction and leaf area index) and one from field inventory data (AGB). Structural differences between the treatments and the control plots were reliably distinguished by the UAV-borne lidar system. AGB was strongly correlated with canopy height, allowing us to elaborate a predictive equation to use the UAV-borne lidar system for monitoring structural features in other restoration plantations in the region. UAV-borne lidar systems showed enormous potential for monitoring relatively broad-scale (thousands of hectares) forest restoration projects, providing an important tool to aid decision making and accountability in forest landscape restoration.

Sustainable forest management relies on good knowledge of forest structure obtained from ground surveys combined with remote sensing. Capable of detecting both the forest floor and canopy elements, airborne LiDAR can estimate forest structure parameters with accuracy and precision, but is still difficult to acquire due to the lake of service provider in remote regions of developing countries. Alternatively if ground surface elevations are known (e.g., from LiDAR), they can be tied to a canopy surface model derived from stereo photogrammetry using RGB images from unmanned aerial vehicles (UAV). Here we assessed whether such photogrammetric canopy measurements offer aboveground biomass (AGB) and disturbance impact estimates from logging that are comparable to LiDAR, and whether the use of both in sequence can provide an efficient post-harvest monitoring system. Specifically, through a combination of forest inventory ground plots, airborne LiDAR data, and a UAV-RGB camera system we (i) automatically located and measured canopy disturbance caused by logging, (ii) compared AGB models produced by LiDAR alone and the combination of LiDAR (for terrain elevation model) and RGB-photogrammetry (for forest surface model), and (iii) estimated the AGB stock loss from logging. The study was carried out in the Antimary State forest located in the southwestern Brazilian Amazon. Our results demonstrate that the use of RGB-photogrammetry in regions where the terrain elevation has already been estimated can be an effective way to rapidly identify selective logging and to accurately monitor its impact.

International audience; Remote sensors, onboard orbital platforms, aircraft, or unmanned aerial vehicles (UAVs) have emerged as a promising technology to enhance our understanding of changes in ecosystem composition, structure, and function of forests, offering multi-scale monitoring of forest restoration. UAV systems can generate high-resolution images that provide accurate information on forest ecosystems to aid decision-making in restoration projects. However, UAV technological advances have outpaced practical application; thus, we explored combining UAV-borne lidar and hyperspectral data to evaluate the diversity and structure of restoration plantings. We developed novel analytical approaches to assess twelve 13-year-old restoration plots experimentally established with 20, 60 or 120 native tree species in the Brazilian Atlantic Forest. We assessed (1) the congruence and complementarity of lidar and hyperspectral-derived variables, (2) their ability to distinguish tree richness levels and (3) their ability to predict aboveground biomass (AGB). We analyzed three structural attributes derived from lidar data—canopy height, leaf area index (LAI), and understory LAI—and eighteen variables derived from hyperspectral data—15 vegetation indices (VIs), two components of the minimum noise fraction (related to spectral composition) and the spectral angle (related to spectral variability). We found that VIs were positively correlated with LAI for low LAI values, but stabilized for LAI greater than 2 m2/m2. LAI and structural VIs increased with increasing species richness, and hyperspectral variability was significantly related to species richness. While lidar-derived canopy height better predicted AGB than hyperspectral-derived VIs, it was the fusion of UAV-borne hyperspectral and lidar data that allowed effective co-monitoring of both forest structural attributes and tree diversity in restoration plantings. Furthermore, considering lidar and hyperspectral data together more broadly supported the expectations of biodiversity theory, showing that diversity enhanced biomass capture and canopy functional attributes in restoration. The use of UAV-borne remote sensors can play an essential role during the UN Decade of Ecosystem Restoration, which requires detailed forest monitoring on an unprecedented scale.

In high biodiversity areas, such as the Amazon, forest inventory is a challenge due to large variations in vegetation structure and inaccessibility. Capturing the full gradient of variability requires the acquisition of a large number of sample plots. Pre-stratified inventory is an efficient strategy that reduces sampling effort and cost. Low-cost remote sensing techniques may significantly expand pre-stratification capacity; however, the simplest option, satellite optical imagery, cannot detect small variations in primary forests. Alternatively, three-dimensional information obtained from airborne laser scanning (ALS, a.k.a. airborne lidar) has been successfully used to estimate structural parameters in tropical forests. Our objective was to assess to what extent forest plot sampling effort could be reduced, while accurately estimating mean vegetation characteristics in the landscape, by stratifying with ALS structural properties, relative to a random, uniformed conventional approach. The study was developed in an 800-ha area of wet Amazonian forest (Acre, Brazil), including portions of palms, bamboo and dense forest. We estimated relevant structural attributes from ALS: canopy height, openness, rugosity and fractions of leaf area index (LAI) along the vertical profile. We clustered vegetation to define heterogeneity into structural types, employing the Ward method and Euclidean distance. Also, principal component analysis was employed to characterize the groups using field and ALS-derived structural attributes. We simulated sampling intensities to estimate the gain in reducing the field efforts based on pre-stratified and non-stratified forest inventory scenarios. The resulting stratification clearly distinguished the forest?s structural variation gradient and the vegetation density profile. For a fixed uncertainty of 10% in basal area estimation, the ALS-aided stratified inventory reduced the necessary number of field plots by 41%, relative to simple random sampling. The resulting reduction in sampling effort can offset the cost of ALS data collection, significantly enhancing its financial feasibility. In addition, ALS provides broad-coverage quantifications of basal area (or aboveground carbon stock), canopy structure, and accurate terrain characterization, which have an added value for forest management. Made available in DSpace on 2019-11-26T18:10:07Z (GMT). No. of bitstreams: 1 26910.pdf: 2646815 bytes, checksum: 03266a4041ee93c762db333f5828420b (MD5) Previous issue date: 2019

The precision forest management technique still has much to be improved with the incorporation of forest biometric techniques and forest profiling with airborne LIDAR. When planning the cutting of a tree in forest management, the volume to be produced for industry is estimated but not the area impacted by removal of the tree. The objective of the present study was to develop equations for the Amazon rainforest that are able to estimate the impact area of gaps from harvesting individual dominant and co-dominant trees based on the canopy morphology obtained through forest profiling. On two separate occasions profiles were made in an annual forest-production unit in the Antimary State Forest (FEA) in the state of Acre, Brazil. The first was done a few days before the start of logging in 2010 and the second was done after completion of harvest activities in 2011. With field measurements and processing of the cloud of LIDAR points, dendrometric and morphometric variables were obtained for the canopy in order to develop equations for estimating gap areas. After evaluation of the explanatory variables with the highest correlation with gap area, the method used considered all possible models and included 2-4 parameters. The explanatory variables that best represent the impact of clearings are volume of the crown (VCop) and crown-projection area (APC). Ten equations were selected, of which two were chosen for use; these had R2 aj > 75% and Syx 75% e Syx < 23%. O bom ajuste das equações demonstra o potencial do uso do LIDAR visando obter informações para se estimar previamente as clareiras pela extração de árvores de diferentes dimensões.

O manejo florestal madeireiro é uma atividade que promove a manutenção da cobertura florestal em áreas de floresta primária no Brasil. Atualmente, as técnicas de manejo estão sendo aperfeiçoadas pelo desenvolvimento de novas tecnologias da área de geoprocessamento, como o Sistema Global de Posicionamento por Satélite (GNSS), Lidar (Ligth detection and ranging) aerotransportado (ALS) e Aeronaves Remotamente Pilotadas (ARPs). O objetivo geral deste trabalho é apresentar os princípios do manejo florestal de precisão aplicados para a região amazônica e avaliar a capacidade dos dados do Lidar na redução da intensidade amostral no inventário florestal. Para tal, foi utilizada uma área demonstrativa de manejo florestal de precisão localizada no campo experimental da Embrapa, no município de Rio Branco, Acre. O estudo foi desenvolvido em uma área contínua de 800 hectares, com presença de floresta primária do tipo umbrófila, aberta, com presença de palmeiras, bambu e manchas de floresta alta. Foram realizadas duas coletas de dados nesta área: o censo florestal das árvores com DAP >= 40 cm, com o qual calculou-se a área seccional de cada árvore; e o voo Lidar feito a uma altitude de 600 metros, com footprint de 30 cm e densidade de 14 pts.m2. A partir dos dados Lidar foram extraídas os modelos digitais de superfície (MDS), do terreno (MDT) e altura (MAD), e as métricas Altmd (altura média do dossel), rugosidade (desvio padrão da altura média do dossel), Altdossel (abertura do dossel a 15 metros), LAI (índice de área foliar) e LAIsub (LAI do sub-bosque - faixa entre 5 a 12 metros de altura). No capítulo 01, os dados foram analisados de forma a demonstrar a aplicação do Lidar no aperfeiçoamento das atividades técnicas do manejo florestal na amazônia, sendo elas: (i) macroplanejamento da Área de Manejo Florestal (AMF); (ii) microplanejamento do terreno e hidrografia da Unidade de Produção Anual (UPA); (iii) censo e amostragem florestal; (iv) planejamento da infraestrutura de exploração; e (v) monitoramento. Na análise dos dados do capítulo 02, foram feitas: (i) estimativa dos atributos estruturais derivados do inventário de campo e da nuvem de retornos Lidar para três tamanhos de célula; (ii) correlação de Pearson entre os atributos do campo e do dossel derivados do Lidar; (iii) análise de agrupamento para formação dos estratos pelo método de Ward; (iv) validação (ANOVA) e caracterização dos estratos (PCA); e (v) estimativa de redução da intensidade amostral por meio de uma simulação computacional. Como resultado, os mapas gerados mostram a acurácia e qualidade dos dados do Lidar para o mapeamento de unidades de produção anual dentro da área de manejo florestal, microzoneamento da topografia, modelagem da rede de drenagem, identificação de áreas de relevo acidentado, delimitação da APP e planejamento da infraestrutura de exploração respeitando as condições ambientais e potecial madeireiro da área de estudo. A altura média do dossel da floresta (Altmd), obtida com o Lidar, apresentou a maior correlação (r = 0,74) com a área basal e foi utilizada para estratificar a vegetação em quatro padrões de floresta, com diferença significativa entre eles para a variável de campo. A estratificação com a Altmd distinguiu bem a floresta quanto ao seu gradiente de variação estrutural e perfil de densidade da vegetação. A simulação do inventário estratificado versus o inventário inteiramente casualizado evidenciou a capacidade da variável Lidar Altmd em reduzir o número de parcelas em até 41% quando utilizado 4 estratos para uma mesma incerteza relativa de 10%. O uso de uma única métrica de dossel, derivada do Lidar, foi capaz de estratificar uma área de floresta tropical primária, além de gerar informações adicionais, como o índice de área foliar do sub-bosque e o modelo do terreno que podem auxiliar na tomada de decisão para o manejo, conservação e monitoramento de florestas tropicais. Timber forest management is an activity that promotes the maintenance of forest cover in primary forest areas in Brazil. Currently, management techniques are being improved by the development of new geoprocessing technologies such as the Global Positioning Satellite System (GNSS), Airborne Lidar Scanning (ALS) and Remotely Piloted Aircraft (ARPs). The general objective of this work is to present the principles of precision forest management applied to the Amazon region and evaluate the reduction of sampling intensity in the forest inventory from vegetation stratification using Lidar variables. For this, was used a demonstration area of precision forest management located in the experimental field of Embrapa, in Rio Branco, Acre. The study was located in a continuous area of 800 hectares, in open ombrphylous forest with presence of palms, bamboo and dense forest portions. Two data collections were carried out in this area: the census inventory of trees with DBH >= 40 cm, which the sectional area of each tree was calculated; and the Lidar flight made at an altitude of 600 meters, with footprint of 30 cm and density of 14 pts.m2. From the Lidar data were processed the digital surface models (DSM), terrain (DTM) and height (CHM), CHmean (average canopy height), roughness (standard deviation of the canopy), CHopenness (canopy openness of 15 meters), LAI (leaf area index) and LAIunder (LAI between 5 to 12 meters of height). In chapter 01, the data were analyzed in order to demonstrate the application of Lidar to improve the technical activities of the forest management in the Amazon, being: (i) macroplanning of the forest management area (AMF); (ii) micro-planning of the land and hydrography of the annual production unit (UPA); (iii) forest census and sampling; (iv) exploration infrastructure planning; and (v) monitoring. The analysis performed on the data of chapter 02 were: (i) estimation of the structural attributes derived from the field inventory and of the return cloud Lidar for three cell sizes; (ii) Pearson correlation between the field and canopy attributes derived from the Lidar; (iii) cluster analysis for classify strata by the Ward method; (iv) validation (ANOVA) and characterization of stratas (PCA); and (v) reduction of sample estimation intensity by a computer simulation. As result, the generated maps show the accuracy and quality of the Lidar data for mapping of annual production units within the area of forest management, microzoning of the topography, drainage network modeling, identification of rugged relief areas, APP boundary and planning the exploration infrastructure respecting the environmental conditions and timber potential of the study area. The average height of the forest canopy (CHmean), obtained with the Lidar, presented the highest correlation (r = 0.74) with the basal area and was used to stratify the vegetation in four forest patterns, with significant difference between them for the field variable. The stratification with CHmean distinguished the forest well for its structural variation gradient and vegetation density profile. The simulation of the stratified inventory versus the completely randomized inventory evidenced the ability of the Lidar Altmd variable to reduce the number of plots by up to 41% when using 4 strata for the same relative uncertainty of 10%. The use of a single canopy metric derived from Lidar was able to stratify an area of primary rainforest, in addition to generating additional information such as understory leaf area index and the terrain model that can assist in decision making for management, conservation and monitoring of tropical forests.

Airborne Laser Scanning (ALS) has been considered as a primary source to model the structure and function of a forest canopy through the indicators leaf area index (LAI) and vertical canopy profiles of leaf area density (LAD). However, little is known about the effects of the laser pulse density and the grain size (horizontal binning resolution) of the laser point cloud on the estimation of LAD profiles and their associated LAIs. Our objective was to determine the optimal values for reliable and stable estimates of LAD profiles from ALS data obtained over a dense tropical forest. Profiles were compared using three methods: Destructive field sampling, Portable Canopy profiling Lidar (PCL) and ALS. Stable LAD profiles from ALS, concordant with the other two analytical methods, were obtained when the grain size was less than 10 m and pulse density was high (> 15 pulses m -2 ). Lower pulse densities also provided stable and reliable LAD profiles when using an appropriate adjustment (coefficient K). We also discuss how LAD profiles might be corrected throughout the landscape when using ALS surveys of lower density, by calibrating with LAI measurements in the field or from PCL. Appropriate choices of grain size, pulse density and K provide reliable estimates of LAD and associated tree plot demography and biomass in dense forest ecosystems. © 2019 by the authors.

Based on airborne LIDAR data on canopy morphology and height of Amazon forest trees, we developed allometric models to estimate dry biomass stored in the boles of dominant and co-dominant individuals and compared these results with those from equations based on traditional variables such as diameter at breast height (DBH). The database consisted of 142 trees of interest for logging in a forest under management for timber in Brazil's state of Acre. The trees chosen for study were selected through proportional sampling by diameter class (ranging from 45 to 165 cm DBH) in order to properly represent the dominant and co-dominant tree populations with diameters appropriate for harvest. Subsequent to LIDAR profiling of these trees, they were felled, subjected to a battery of dimensional measurements and sampled for wood-density determination. A set of models was generated, followed by model selection and identity testing in order to compare groups of basic wood density (low, medium and high). The morphometric variables of the crown had high explanatory power for bole biomass independent of whether the allometric equations included DBH. When calculating bole biomass with equations that include basic wood density, the best estimate is obtained using variables for both DBH and crown morphology. To obtain an allometric equation that encompasses species in all three classes of basic density, one should either use only independent variables representing crown dimensions or complement these with variables for basic density (BD) and total height (Ht). The study demonstrates the feasibility of using ground-based measurements to calibrate biomass models that include only LIDAR-based variables, allowing much larger areas to be surveyed with reasonable accuracy. The present study is designed to produce data needed for forest management, but the methods developed here can be adapted to studies aimed at reducing the uncertainty in biomass estimates of whole forests (not just harvestable trees) for use in quantifying carbon emissions from forest loss and degradation. © 2016 The Author(s)

A estimativa volumétrica, a partir do escaneamento digital de florestas por meio do uso do LIDAR, potencializa o emprego de técnicas de manejo de precisão no planejamento da exploração nas florestas tropicais. A utilização dessa tecnologia de sensoriamento remoto permite a incorporação de variáveis da morfometria de copa, ainda pouco empregadas e menos conhecidas em decorrência da dificuldade de coleta em campo. O objeto deste estudo foi construir equações capazes de estimar o volume do fuste de árvores individuais dominantes e codominantes, a partir da morfometria da copa obtida por meio do LIDAR aerotransportado, considerando duas situações de inventário florestal: a) com a coleta do DAP, conjuntamente com as variáveis morfométrica da copa obtidas pelo LIDAR e b) apenas com os dados de morfometria de copa. Para seleção dos modelos foram considerados: a matriz de correlação das variáveis preditoras e a combinação das variáveis que geraram os melhores resultados estatísticos pelos critérios Syx , Syx(%) e Press p, e que foram homocedásticos e com disposição dos resíduos normais e independentes. Para as melhores equações foram realizadas análise de influência. Os resultados estatísticos do ajuste dos modelos para as duas situações permitiram selecionar equações com e sem DAP, com resultados R²aj. (%) de a) 92,92 e b) 79,44; Syx(%) de a) 16,73 e b) 27,47; e, critério de Press p de a) 201,15 m6 e b) 537,47 m6, respectivamente. Por meio das variáveis morfométricas, foi possível desenvolver equações capazes de estimar com precisão o volume do fuste de árvores dominantes e codominantes em florestas tropicais. The volumetric estimate from digital scanning of the forests through the use of LIDAR increases the precision of forest management techniques in planning tropical forest logging operations. The use of this remote detection technology allows the incorporation of crown morphometric variables which are still little known and little used due to the difficulty of collecting field data for volume equations. The objective of this study was to build equations capable of estimating the stem volume of dominant and codominant individual trees from the crown's morphometry obtained by airborne LIDAR, considering two forest inventory situations: a) with the collection of diameter at breast height (DBH), and crown morphometric variables obtained from LIDAR data and b) using only the crown morphometry variables. For the selection of models the factors considered were: the correlation matrix of predictor variables and the combination of variables that generates the best results by statistical criteria Syx, Syx(%) and Press p, and that were homoscedastic and had a normal and independent distribution of errors. The influence analysis was performed for the best equations. The results for the statistical fit of the equations to the two situations allowed the selection of models with and without DBH, with R²aj.( %) values of a) 92.92 and b) 79.44, Syx(%) values of a) 16.73 and b) 27.47, and, Press p criterion values of a) 201.15 m6 and b) 537.47 m6, respectively. Through morphometric variables it was possible to develop equations capable of accurately estimating the stem volume of dominant and codominant trees in tropical forests

O manejo florestal madeireiro é uma atividade que promove a manutenção da cobertura florestal em áreas de floresta primária no Brasil. Atualmente, as técnicas de manejo estão sendo aperfeiçoadas pelo desenvolvimento de novas tecnologias da área de geoprocessamento, como o Sistema Global de Posicionamento por Satélite (GNSS), Lidar (Ligth detection and ranging) aerotransportado (ALS) e Aeronaves Remotamente Pilotadas (ARPs). O objetivo geral deste trabalho é apresentar os princípios do manejo florestal de precisão aplicados para a região amazônica e avaliar a capacidade dos dados do Lidar na redução da intensidade amostral no inventário florestal. Para tal, foi utilizada uma área demonstrativa de manejo florestal de precisão localizada no campo experimental da Embrapa, no município de Rio Branco, Acre. O estudo foi desenvolvido em uma área contínua de 800 hectares, com presença de floresta primária do tipo umbrófila, aberta, com presença de palmeiras, bambu e manchas de floresta alta. Foram realizadas duas coletas de dados nesta área: o censo florestal das árvores com DAP > 40 cm, com o qual calculou-se a área seccional de cada árvore; e o voo Lidar feito a uma altitude de 600 metros, com footprint de 30 cm e densidade de 14 pts.m2. A partir dos dados Lidar foram extraídas os modelos digitais de superfície (MDS), do terreno (MDT) e altura (MAD), e as métricas Altmd (altura média do dossel), rugosidade (desvio padrão da altura média do dossel), Altdossel (abertura do dossel a 15 metros), LAI (índice de área foliar) e LAIsub (LAI do sub-bosque - faixa entre 5 a 12 metros de altura). No capítulo 01, os dados foram analisados de forma a demonstrar a aplicação do Lidar no aperfeiçoamento das atividades técnicas do manejo florestal na amazônia, sendo elas: (i) macroplanejamento da Área de Manejo Florestal (AMF); (ii) microplanejamento do terreno e hidrografia da Unidade de Produção Anual (UPA); (iii) censo e amostragem florestal; (iv) planejamento da infraestrutura de exploração; e (v) monitoramento. Na análise dos dados do capítulo 02, foram feitas: (i) estimativa dos atributos estruturais derivados do inventário de campo e da nuvem de retornos Lidar para três tamanhos de célula; (ii) correlação de Pearson entre os atributos do campo e do dossel derivados do Lidar; (iii) análise de agrupamento para formação dos estratos pelo método de Ward; (iv) validação (ANOVA) e caracterização dos estratos (PCA); e (v) estimativa de redução da intensidade amostral por meio de uma simulação computacional. Como resultado, os mapas gerados mostram a acurácia e qualidade dos dados do Lidar para o mapeamento de unidades de produção anual dentro da área de manejo florestal, microzoneamento da topografia, modelagem da rede de drenagem, identificação de áreas de relevo acidentado, delimitação da APP e planejamento da infraestrutura de exploração respeitando as condições ambientais e potecial madeireiro da área de estudo. A altura média do dossel da floresta (Altmd), obtida com o Lidar, apresentou a maior correlação (r = 0,74) com a área basal e foi utilizada para estratificar a vegetação em quatro padrões de floresta, com diferença significativa entre eles para a variável de campo. A estratificação com a Altmd distinguiu bem a floresta quanto ao seu gradiente de variação estrutural e perfil de densidade da vegetação. A simulação do inventário estratificado versus o inventário inteiramente casualizado evidenciou a capacidade da variável Lidar Altmd em reduzir o número de parcelas em até 41% quando utilizado 4 estratos para uma mesma incerteza relativa de 10%. O uso de uma única métrica de dossel, derivada do Lidar, foi capaz de estratificar uma área de floresta tropical primária, além de gerar informações adicionais, como o índice de área foliar do sub-bosque e o modelo do terreno que podem auxiliar na tomada de decisão para o manejo, conservação e monitoramento de florestas tropicais. Timber forest management is an activity that promotes the maintenance of forest cover in primary forest areas in Brazil. Currently, management techniques are being improved by the development of new geoprocessing technologies such as the Global Positioning Satellite System (GNSS), Airborne Lidar Scanning (ALS) and Remotely Piloted Aircraft (ARPs). The general objective of this work is to present the principles of precision forest management applied to the Amazon region and evaluate the reduction of sampling intensity in the forest inventory from vegetation stratification using Lidar variables. For this, was used a demonstration area of precision forest management located in the experimental field of Embrapa, in Rio Branco, Acre. The study was located in a continuous area of 800 hectares, in open ombrphylous forest with presence of palms, bamboo and dense forest portions. Two data collections were carried out in this area: the census inventory of trees with DBH > 40 cm, which the sectional area of each tree was calculated; and the Lidar flight made at an altitude of 600 meters, with footprint of 30 cm and density of 14 pts.m2. From the Lidar data were processed the digital surface models (DSM), terrain (DTM) and height (CHM), CHmean (average canopy height), roughness (standard deviation of the canopy), CHopenness (canopy openness of 15 meters), LAI (leaf area index) and LAIunder (LAI between 5 to 12 meters of height). In chapter 01, the data were analyzed in order to demonstrate the application of Lidar to improve the technical activities of the forest management in the Amazon, being: (i) macroplanning of the forest management area (AMF); (ii) micro-planning of the land and hydrography of the annual production unit (UPA); (iii) forest census and sampling; (iv) exploration infrastructure planning; and (v) monitoring. The analysis performed on the data of chapter 02 were: (i) estimation of the structural attributes derived from the field inventory and of the return cloud Lidar for three cell sizes; (ii) Pearson correlation between the field and canopy attributes derived from the Lidar; (iii) cluster analysis for classify strata by the Ward method; (iv) validation (ANOVA) and characterization of stratas (PCA); and (v) reduction of sample estimation intensity by a computer simulation. As result, the generated maps show the accuracy and quality of the Lidar data for mapping of annual production units within the area of forest management, microzoning of the topography, drainage network modeling, identification of rugged relief areas, APP boundary and planning the exploration infrastructure respecting the environmental conditions and timber potential of the study area. The average height of the forest canopy (CHmean), obtained with the Lidar, presented the highest correlation (r = 0.74) with the basal area and was used to stratify the vegetation in four forest patterns, with significant difference between them for the field variable. The stratification with CHmean distinguished the forest well for its structural variation gradient and vegetation density profile. The simulation of the stratified inventory versus the completely randomized inventory evidenced the ability of the Lidar Altmd variable to reduce the number of plots by up to 41% when using 4 strata for the same relative uncertainty of 10%. The use of a single canopy metric derived from Lidar was able to stratify an area of primary rainforest, in addition to generating additional information such as understory leaf area index and the terrain model that can assist in decision making for management, conservation and monitoring of tropical forests. Made available in DSpace on 2019-01-11T00:02:20Z (GMT). No. of bitstreams: 1 26685.pdf: 9461143 bytes, checksum: c28dd3c7998999ceade4de9e3db4052c (MD5) Previous issue date: 2018-09-20 Dissertação (Mestrado em Ciências: Silvicultura e Manejo Florestal) - Programa de Pós-graduação Recursos Florestais, Escola Superior de Agricultura "Luiz de Queiroz", Universidade de São Paulo, Piracicaba. Orientador: Luiz Carlos Estraviz Rodriguez.

Based on airborne LIDAR data on canopymorphology and height of Amazon forest trees, we developed allometric models to estimate dry biomass stored in the boles of dominant and co-dominant individuals and compared these results with those from equations based on traditional variables such as diameter at breast height (DBH). The database consisted of 142 trees of interest for logging in a forest under management for timber in Brazil's state of Acre. The trees chosen for studywere selected through proportional sampling by diameter class (ranging from45 to 165 cm DBH) in order to properly represent the dominant and co-dominant tree populationswith diameters appropriate for harvest. Subsequent to LIDAR profiling of these trees, they were felled, subjected to a battery of dimensional measurements and sampled for wood-density determination. A set ofmodels was generated, followed by model selection and identity testing in order to compare groups of basicwood density (low, medium and high). The morphometric variables of the crown had high explanatory power for bole biomass independent of whether the allometric equations included DBH. When calculating bole biomass with equations that include basic wood density, the best estimate is obtained using variables for both DBH and crown morphology. To obtain an allometric equation that encompasses species in all three classes of basic density, one should either use only independent variables representing crown dimensions or complement these with variables for basic density (BD) and total height (Ht). The study demonstrates the feasibility of using ground-based measurements to calibrate biomass models that include only LIDAR-based variables, allowing much larger areas to be surveyed with reasonable accuracy. The present study is designed to produce data needed for forest management, but the methods developed here can be adapted to studies aimed at reducing the uncertainty in biomass estimates of whole forests (not just harvestable trees) for use in quantifying carbon emissions from forest loss and degradation.

A estimativa volumétrica, a partir do escaneamento digital de florestas por meio do uso do LIDAR, potencializa o emprego de técnicas de manejo de precisão no planejamento da exploração nas florestas tropicais. A utilização dessa tecnologia de sensoriamento remoto permite a incorporação de variáveis da morfometria de copa, ainda pouco empregadas e menos conhecidas em decorrência da dificuldade de coleta em campo. O objeto deste estudo foi construir equações capazes de estimar o volume do fuste de árvores individuais dominantes e codominantes, a partir da morfometria da copa obtida por meio do LIDAR aerotransportado, considerando duas situações de inventário florestal: a) com a coleta do DAP, conjuntamente com as variáveis morfométrica da copa obtidas pelo LIDAR e b) apenas com os dados de morfometria de copa. Para seleção dos modelos foram considerados: a matriz de correlação das variáveis preditoras e a combinação das variáveis que geraram os melhores resultados estatísticos pelos critérios Syx, Syx(%) e Pressp, e que foram homocedásticos e com disposição dos resíduos normais e independentes. Para as melhores equações foram realizadas análise de influência. Os resultados estatísticos do ajuste dos modelos para as duas situações permitiram selecionar equações com e sem DAP, com resultados R2 aj.(%) de a) 92,92 e b) 79,44; Syx (%) de a) 16,73 e b) 27,47; e, critério de Pressp de a) 201,15 m6 e b) 537,47 m6, respectivamente. Por meio das variáveis morfométricas, foi possível desenvolver equações capazes de estimar com precisão o volume do fuste de árvores dominantes e codominantes em florestas tropicais.