O resultado do escaneamento de um Laser Scanner Terrestre (LST) é uma nuvem de pontos com coordenadas geométricas (X, Y, Z), informações de cor (R, G, B) provenientes de uma câmera fotográfica acoplada ao equipamento e, ainda, a informação do retorno da intensidade do pulso laser (I). Esses sistemas de varredura possuem algumas características como, por exemplo, sua rapidez na aquisição de informações, registro de cenas em 3D e coleta de informações sem contato direto que se aplicam de forma importante nas análises florestais. Contudo, a grande vantagem da utilização de um LST na área florestal é a possibilidade de caracterizar alvos remotamente de forma rápida e não destrutiva. Assim, este trabalho teve como objetivo principal avaliar os dados de intensidade de retorno do laser provenientes de um sistema LST para a caracterização de alvos florestais. Metodologicamente foram realizados experimentos controlados que envolveram as seguintes etapas: calibração radiométrica do LST; avaliação da influência da distância nos dados de intensidade de retorno do laser e; análise do efeito de borda em imageamento de alvos florestais (considerado um dos principais problemas que afeta os dados intensidade de retorno quando se utiliza um LST). O equipamento utilizado durante os experimentos foi um laser scanner modelo Ilris 3D da Optech que trabalha no intervalo do infravermelho médio com comprimento de onde de 1535 nm. Os resultados mostraram que para esse comprimento de onda os alvos florestais devem ser imageados a uma distância maior ou igual a 5m e o processamento dos dados com resolução radiométrica de 8 bits foi mais adequado, pois proporcionou uma caracterização geométrica do alvo com efeito visual de melhor qualidade se comparado com o processamento de 16 bits. Os resultados dos experimentos realizados sobre o efeito de borda possibilitaram identificar dois tipos de distorções que ocorrem em dados de nuvem de pontos adquiridos com um LST. O primeiro afetou os valores de intensidade de retorno do laser e o segundo criou um efeito que deslocou os pontos no espaço. Para minimizar este efeito foi desenvolvido um algoritmo, o IRA (Intensity Recovery Algorithm), que possibilitou recuperar automaticamente os valores de intensidade de retorno do laser minimizando em até 35,7% o efeito de borda nos imageamentos do alvo estudado na pesquisa. Assim para o uso de um LST, na caracterização geométrica de alvos florestais, é necessário desenvolver modelos de calibração da intensidade de retorno do pulso laser, pois os sistemas LST são distintos em termos de faixa do espectro eletromagnético que operam. Por fim, no que tange ao efeito de borda concluiu-se que o algoritmo IRA necessita ser aprimorado com outras abordagens computacionais e matemáticas que poderão ser desenvolvidos em estudos futuros. / The result of the scanning of a terrestrial laser scanner (TLS) is a point cloud with geometric coordinates (X, Y, Z), color information (R, G, B) from a camera coupled to the equipment, and also the return information of intensity of the laser pulse (I). These scanning systems have some characteristics, for example, its speed in acquiring information, and of 3D scenes with record of data remotely which applies significantly in the forestry analysis. The advantage of using a TLS in the forestry area is the possibility of remote acquisition of data enabling a fast and non-destructive work. This work aimed to evaluate the laser return intensity data from a TLS system for the characterization of forest targets. Methodologically were performed controlled experiments involving the following steps: radiometric calibration of TLS; evaluating the influence of the distance in the laser return intensity data and; analysis of the edge effect in imaging forest targets (considered one of the main problems that affect the data intensity return when using a TLS). The equipment used during the experiments was a laser scanner Ilris Optech 3D model that works in the mid-infrared range with wavelength of 1535 nm. The results showed that for this wavelength forest targets should be imaged at a distance greater than or equal to 5m and processing of the radiometric data with 8-bit resolution is more suitable because it provided a geometric characteristics of the target with better visual effect quality compared with the 16-bit processing. The results of the experiments on the edge effect possible to identify two types of distortions that occur in cloud data points acquired with a LST. The first affect the laser return intensity values and the second set offset an effect that the points in space. To minimize this effect an algorithm, the IRA (Intensity Recovery Algorithm), was developed which enabled automatically retrieve the laser return intensity values up to 35.7% of minimizing the edge effect in the target imaging surveys studied in research. Thus for use of an TLS, the geometrical characterization of forest targets, it is necessary to develop calibration models of the return laser pulse intensity, for TLS systems are different in terms of the electromagnetic spectrum operating range. Finally, with respect to the edge effect it was concluded that the IRA algorithm needs to be enhanced with other computational and mathematical approaches that may be developed in future studies.

Syn-rift exposures are a prime source of knowledge on the structure and stratigraphy of rift basins and are vital for the petroleum industry as analogues for subsurface reservoirs. Focusing on superb exposures of the Oligo-Miocene rift initiation Nukhul half-graben in the Suez Rift, Egypt, this study investigates applications of 3D digital survey techniques, particularly terrestrial light detection and ranging (lidar), for (i) the analysis of syn-rift fault and facies architecture, and (ii) the building and testing of outcrop-based reservoir analogue models for early syn-rift settings. Lidar-based digital outcrop mapping of the Nukhul half-graben, combined with conventional fieldwork, facilitated quantification of thickness and facies relationships within syn-rift strata, and variations of throw along normal faults. The results provide new insights into the tectono-sedimentary evolution of the intra-block half-graben. At rift initiation, regional/eustatic sea-level and antecedent drainage are interpreted as the dominant controls on accommodation development and deposition, respectively. However, after just c. 2.5 myr of rifting, the structural template imposed by the propagation and linkage of four initially isolated pre-cursor segments of the Nukhul Fault became the dominant control on accommodation development and basin physiography. Progressive SE-NW back-stepping of facies strike-parallel to the Nukhul Fault suggests hangingwall subsidence was locally sufficient to outpace falls in regional sea-level, resulting in a dynamic transgressive system that progressed from a restricted tidal embayment to shallow marine seaway along fault strike. After c. 4.3 myr of rifting, regional drowning of the tidal system provides evidence for progressive localisation of displacement onto the present-day block-bounding structures and declining activity on the intra-block Nukhul Fault during the transition from rift initiation to rift climax. Development of digital outcrop techniques such as point cloud facies classification, 3D deterministic channel modelling and net-to-gross analysis provided reliable geostatistics on the geometry, distribution and heterogeneity of tide-influenced facies of the syn-rift Nukhul Formation. The high volume, reliability and spatial coverage of data reduced uncertainties related to stochastic facies modelling (in this case sequential indicator simulation and object-based techniques), facilitating building and testing of high-resolution analogue models for the complex facies and sequence architecture of early syn-rift tidal reservoirs. Visual analysis of static reservoir connectivity suggests some of the smallest-scale depositional elements, i.e. thin intercalations of mudstone and sandstone lamina in heterolithic facies, have the biggest impact on both reservoir volume and vertical connectivity in this syn-rift tidal system. Heterogeneities at the sequence stratigraphic scale also have significant impact on vertical reservoir compartmentalisation, whereas tidal channel lag deposits and tidal mud drapes have more localised effects. Use of lidar, when combined with conventional fieldwork, offers a powerful tool for quantitative spatial analysis of fault and facies architecture, tightly constraining 3D structural and stratigraphic interpretations and effectively increasing the statistical significance of outcrop analogues for reservoir characterisation. Future developments in survey technologies alongside geoscience-specific software for the integration and analysis of outcrop datasets will provide more diverse and quantitative information on geological heterogeneity, and promote wider-ranging applications in Earth sciences.

Field scale research shows concentrated flow paths (CFPs) are prevalent in agricultural watersheds. They are an important source of soil erosion in cropland and significantly contribute to the transport and delivery of agricultural pollutants such as sediment and nutrients to nearby water resources. High resolution LiDAR data have enabled the investigation of the prevalence of concentrated flow at a large geographic extent. This study focused on identifying CFPs in 389 agricultural fields in Jackson County in southern Illinois and estimating the contribution of the CFPs to drainage of the fields. Addressing the lack of literature on factors influencing CFP characteristics, this study also investigates various topographical and soil factors that influence CFP development. LiDAR derived DEMs with a cell resolution of 3 meters were used to identify areas of flow concentration and delineate a drainage basin of each CFP using the Hydrology tools in the Spatial Analyst toolbox in ArcMap 10.3.1. Information on the topographical and soil characteristics were obtained from the DEMs and SSURGO database using the Soil Data Viewer 6.2 extension for ArcMap. Multiple regression analysis in SAS v. 9.4 was used to identify factors influencing CFP characteristics, while CART analysis in R v. 3.3.1 was conducted to detect linear dependencies among predictor variables. An average of 5 CFPs per agricultural field were observed in the study area with a minimum of 0 and maximum of 17 CFPs, but only two fields had no CFPs indicating a high prevalence of CFPs throughout the study area. The mean percent of field area drained by CFPs was estimated to be 81 percent with minimum of 36 percent and ii maximum of 100 percent. The majority of the fields, 85 percent, had more than 70 percent of their area drained by CFPs. Statistically significant regression equations were found for all CFP characteristics with slope, LS factor, K factor and organic matter as the factors influencing CFP characteristics. However, the factors only explained 2 to 22 percent of variation observed. Both multiple regression and CART analysis indicated slope as the most important influencing factor. Variation in CFP characteristics followed regional trends with higher values in the floodplain region and lower values in the rough hills region suggesting residual variation could be explained by other environmental factors along with topographical and climatological factors which were not included in the study. Results from this study highlight the prevalence of CFPs at a regional scale and their substantial contribution to field drainage identifying a need for research in quantifying the impacts of CFP on soil loss and water quality. This study also reports a need of future research to identify important factors controlling CFP formation and development that could help build empirical and physical models to accurately predict CFP locations and morphology. Such information could be useful in designing and targeting conservation practices that protect both soil health and water quality.

Landslides display a wide variety of behaviors ranging from slow persistent motion to rapid acceleration and catastrophic failure. Given the variety of possible behaviors, improvements to our understanding of landslide mechanics are critical for accurate predictions of landslide dynamics. Recent advances in remote sensing techniques, like satellite radar interferometry (InSAR), now enable high-resolution spatial and temporal measurements that provide insight into the mechanisms that control landslide behavior. In this dissertation, I use InSAR and high-resolution topographic data to identify 50 slow-moving landslides in the Northern California Coast Ranges and monitor their kinematics over 4 years. These landslides have similar mechanical properties and are subject to the same external forcings, which allows me to explore geometrical controls on kinematics. Each landslide displays distinct kinematic zones with different mean velocities that remain spatially fixed. Because these deformation patterns are sensitive to subsurface geometry, I employ a mathematical model to infer landslide thickness and find that these landslides exhibit a highly variable thickness and an irregular basal sliding surface. Time series analysis reveals that each landslide displays well-defined seasonal velocity changes with a periodicity of ∼ 1 year. These velocity variations are driven by precipitation- induced changes in pore-water pressure that lag the onset of rainfall by up to 40 days. Despite significant variations in geometry, I find no systematic differences in seasonal landslide behavior. To further explore how stress perturbations control landslide motion, I develop a mechanical model that reproduces both the displacement patterns observed at slow-moving landslides and the acceleration towards failure exhibited by catastrophic events. I find that catastrophic failure can only occur when the slip surface is characterized by rate-weakening friction and its spatial dimensions exceed a critical nucleation length that is shorter for higher effective stresses. These model simulations support my conclusions from the remote sensing analysis but also provide insight into the long-term evolution of landslides. This dissertation includes both previously published and unpublished co- authored material.

Submitted by ANDRÉ CACERES CARRILHO null (carrilho.acc@gmail.com) on 2016-04-01T02:33:57Z No. of bitstreams: 1 d_carrilho_ac.pdf: 6862099 bytes, checksum: fa858752c5f46a9fc8acf83fee1ed2cc (MD5) / Approved for entry into archive by Ana Paula Grisoto (grisotoana@reitoria.unesp.br) on 2016-04-05T13:36:03Z (GMT) No. of bitstreams: 1 carrilho_ac_me_prud.pdf: 6862099 bytes, checksum: fa858752c5f46a9fc8acf83fee1ed2cc (MD5) / Made available in DSpace on 2016-04-05T13:36:03Z (GMT). No. of bitstreams: 1 carrilho_ac_me_prud.pdf: 6862099 bytes, checksum: fa858752c5f46a9fc8acf83fee1ed2cc (MD5) Previous issue date: 2016-02-16 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / As nuvens de pontos provindas de Sistemas de Varredura a LASER (Light Amplification by Stimulated Emission of Radiation) Aerotransportados (SVLA) representam a superfície com um grande detalhamento, sendo registrada a vegetação e as feições antrópicas, como edificações e rodovias, por exemplo. A automação da tarefa de classificação da nuvem de pontos é importante para a redução de custos em aplicações práticas relacionadas a atividades de mapeamento. Esta dissertação tem por objetivo principal o estudo, implementação e avaliação de técnicas de processamento e análise de imagens visando a classificação da nuvem de pontos LiDAR (Light Detection and Ranging), e a obtenção, de forma automática, do MDT (Modelo Digital do Terreno) e dos pontos pertencentes a edificações e vegetação. Inicialmente são apresentados os conceitos necessários para o entendimento do método proposto, o qual é detalhado na sequência, assim como os conjuntos de dados utilizados para as avaliações. Além da revisão conceitual sobre tópicos como Morfologia Matemática, Análise de Componentes Principais e algoritmo RANSAC, estudou-se a necessidade de redução da quantidade de iterações do RANSAC e uma alternativa para obtenção de alguns dos parâmetros do método a priori. Foram realizados experimentos sobre a escolha do interpolador (para a geração da grade regular) e para a seleção da forma e dimensão do elemento estruturante (usado nas operações morfológicas). A classificação final foi avaliada em termos de completeza e nível de acerto, calculadas com relação a referências geradas manualmente. / The point clouds acquired by ALS (Airborne LASER Scanning) systems represent the surface with great level of detail, recording both natural vegetation and anthropic features, such as buildings and roads, for example. The automation’s task of point clouds classification is important to reduce costs in practical applications related to mapping activities and production. This work is primarily focused in the study, implementation and evaluation of image processing and analysis techniques, aiming the automatic cloud point classification, generation of the DTM (Digital Terrain Model) and the detection of roof plane and vegetation points. Besides the conceptual review about Mathematical Morphology, Principal Component Analysis and RANSAC (Random Sample Consensus), it was studied the need to reduce the quantity of iterations and an alternative to obtain some of the parameters required by RANSAC algorithm considering one preliminary classification based on Principal Component Analysis. Some experiments were conducted in order to select both the interpolation method (to generate the grid) and the shape and size of the structuring element (used in the morphological operations). The final classification was evaluated in terms of completeness and correctness, computed with respect to a manually generated reference.

Disposal of the UK's legacy nuclear waste is the biggest challenge facing the industry at present. There is currently no long term storage facility in the UK and the inventory is continually growing. This project investigates the role that digital geoscientific data collection, analysis and modelling techniques play in the search for, and development of, a Geological Disposal Facility (GDF), critically analyses classical techniques and new, digital methodologies to assess what their impact would be on any site investigation. The Borrowdale Volcanic Group outcrop in Cumbria, NW England was chosen as it provides an analogue to a higher-strength crystalline basement setting for a GDF. Terrestrial lidar and photogrammetric surveys were conducted at four locations around the study area. These provided information on the fracture geostatistics which are the main fluid migration pathways in the subsurface in the BVG. The mechanics of deformation are identified by analysing the clustering of data points via digital stereonet analysis. The analysis shows the rocks sampled are highly fractured and their orientations and dips reflected the extensional tectonism experienced in the area. These are in the form of adjacent sets trending broadly NNE-SSW and NNW-SSE at very high angler dips (~70 degrees). A new workflow developed for this work demonstrates how a potential site's fracture statistics, and indeed the 3D geology, should be investigated as part of future GDF site investigations. Areas of complex geology such as the BVG present many difficulties in interpretation and analysis due to the poorly constrained polyphase nature of the deformation. These complexities make characterisation and modelling highly problematic, and as such, areas of simpler geology should be investigated first. Assessments which were based on early geological studies using traditional field data collection techniques underestimated the impact of heterogeneity on fluid flow migration modelling within the subsurface. This suggests that, should a GDF should be developed in such a geological setting, huge difficulties may be encountered. These will be associated with the development of performance assessments and safety cases which are typically based on geological models that should use such complex data. In addition to this, datasets collected using digital methods are a powerful visualisation tools for communication of complex geology, that can be utilised in stakeholder engagement activities that will form a key part of any GDF development process.

Chilling events are infrequent but important disturbances in subtropical Florida. When temperatures drop to near freezing, significant mortality often accrues in mangrove forests. Chilling events play a role in maintaining structural diversity in mangrove forests, and in limiting mangrove poleward distribution. I examined the spatio-temporal distribution of chilling events in mangrove forests of southern Biscayne Bay by using Landsat TM5 images since 1989. Damage was usually confined to dwarf mangrove forest, especially when chilling temperatures were moderate and short in duration. However, damage from extended and severe freezes such as in January 2010 impacted larger trees as well. Recovery is gradual, often extending over multiple years, depending on disturbance severity. Plant communities respond to repeated chilling with increase in the dominance of black mangrove. In the absence of chilling events, patch level dynamics might lead to prevalence of a more homogenous tall red mangrove canopy in these wetlands. Such a trajectory may result with increasing temperatures expected under current global climate change scenarios

abstract: With a ground-based Doppler lidar on the upwind side of a wind farm in the Tehachapi Pass of California, radial wind velocity measurements were collected for repeating sector sweeps, scanning up to 10 kilometers away. This region consisted of complex terrain, with the scans made between mountains. The dataset was utilized for techniques being studied for short-term forecasting of wind power by correlating changes in energy content and of turbulence intensity by tracking spatial variance, in the wind ahead of a wind farm. A ramp event was also captured and its propagation was tracked. Orthogonal horizontal wind vectors were retrieved from the radial velocity using a sector Velocity Azimuth Display method. Streamlines were plotted to determine the potential sites for a correlation of upstream wind speed with wind speed at downstream locations near the wind farm. A "virtual wind turbine" was "placed" in locations along the streamline by using the time-series velocity data at the location as the input to a modeled wind turbine, to determine the extractable energy content at that location. The relationship between this time-dependent energy content upstream and near the wind farm was studied. By correlating the energy content with each upstream location based on a time shift estimated according to advection at the mean wind speed, several fits were evaluated. A prediction of the downstream energy content was produced by shifting the power output in time and applying the best-fit function. This method made predictions of the power near the wind farm several minutes in advance. Predictions were also made up to an hour in advance for a large ramp event. The Magnitude Absolute Error and Standard Deviation are presented for the predictions based on each selected upstream location. / Dissertation/Thesis / Masters Thesis Mechanical Engineering 2014

Submitted by Automação e Estatística (sst@bczm.ufrn.br) on 2018-07-03T11:34:03Z No. of bitstreams: 1 AlineKellyDeSouzaMelo_DISSERT.pdf: 2929215 bytes, checksum: 6f73161823fbbfb7b658938cba4f48c9 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2018-07-09T15:24:38Z (GMT) No. of bitstreams: 1 AlineKellyDeSouzaMelo_DISSERT.pdf: 2929215 bytes, checksum: 6f73161823fbbfb7b658938cba4f48c9 (MD5) / Made available in DSpace on 2018-07-09T15:24:38Z (GMT). No. of bitstreams: 1 AlineKellyDeSouzaMelo_DISSERT.pdf: 2929215 bytes, checksum: 6f73161823fbbfb7b658938cba4f48c9 (MD5) Previous issue date: 2018-04-27 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este trabalho tem como objetivo analisar aerossóis atmosféricos sobre Natal/RN por meio de suas propriedades ópticas. Os dados utilizados foram do satélite CALIPSO (2007-2016), que possui um Lidar denominado CALIOP e do DUSTER. Utilizaram-se dados do CALIPSO obtidos no máximo a uma distância de um raio de 120 km em volta de Natal, para se obter a profundidade óptica de aerossóis (AOD), a razão lidar (RL) e a altitude da base e do topo da camada de aerossóis. Realizou-se um estudo de caso, para o mês de setembro, usando as máscaras de subtipo de aerossóis geradas pelo CALIPSO. Também foram feitas trajetórias de massas de ar, com o modelo HYSPLIT, para identificar a região de procedência dos aerossóis medidos. O objetivo de escolher dados sobre a região foi para ser efetuado um estudo de comparação de dados do DUSTER/Lidar terrestre na UFRN em Natal, com os dados de dias coincidentes de passagens do CALIPSO. Gerou-se perfis verticais do sinal de retorno a 532 nm, e perfis de retroespalhamento total atenuado a 532 nm do CALIPSO. A análise da profundidade óptica de aerossóis (AOD) mostrou, que sua variabilidade média mensal não excedeu valores maiores que 0,16. Os meses que apresentaram menores concentrações apresentaram uma AOD média de 0,04. Os histogramas de frequência de AOD mostraram que os tipos de aerossol continental limpo, marinho e poeira, possuem valores pouco variáveis. Os tipos de aerossóis que apresentaram maior variação foram o do tipo poluição continental/fumaça. De acordo com a análise da RL, Natal apresentou valores que caracterizam aerossóis do tipo poeira, queimadas, continental poluído e principalmente marinho limpo. O topo e base da camada de aerossóis nas menores altitudes ocorrem em maio, junho e julho, sendo estes os meses em que ocorre a estação chuvosa, e novembro. Os valores mínimos sempre se apresentam abaixo de 1 km de altitude, os meses de fevereiro e setembro obtiveram 25% das altitudes acima de 2,5 km, com altitudes de até 4,9. Os valores de RL de 70 sr e os maiores valores de altura máxima e mínima das camadas de aerossol, destacados para o mês de setembro sugerem a hipótese de casos de detecção de aerossol de queimada em altitudes elevadas acima da camada limite, motivando o estudo de caso, que mostrou, nos dias 07/09/2009, 13/09/2011, 07/09/2013 e 15/09/2016, aerossóis de fumaça localizados a uma altitude de 3 a 4 km. Com trajetórias modeladas pelo HYSPLIT, para 7 dias, pode-se concluir, que essas massas de ar são advindas da África central, para a atmosfera de Natal. A comparação dos dados DUSTER/Lidar com os dados CALIPSO dos dias 06/12/2016 mostrou camadas de aerossóis entre 1 e 3 km, coincidentes nas duas medidas. O perfil de retroespalhamento do CALIPSO ainda permitiu observar uma camada de aerossóis de poeira à um altitude incomum de 15 km, sugerindo futuros estudos específicos para interpretação. / This work aims to analyze atmospheric aerosols over Natal/RN by means of its optical properties. Ten years (2007-2016) of data were obtained from the CALIPSO satellite, which has an on-board Lidar called CALIOP. CALIPSO data obtained at most within a radius of 120 km around Natal were used to obtain the optical Aerosol Depth (AOD), the lidar ratio (RL) and the altitude of the base and top of the aerossol layers. A case study was carried out for the month of September using the aerosol subtype masks generated by CALIPSO. Air mass trajectories with the HYSPLIT model were also used to identify the region of origin of the measured aerosols. The objective of choosing data about the region was to carry out a data comparison study of the terrestrial DUSTER/Lidar at UFRN in Natal, with data from coincident days of CALIPSO passages. Vertical profiles of the return signal were generated at 532 nm, and backscatter profiles attenuated at 532 nm of CALIPSO. The analysis of the aerosol optical depth (AOD) showed that their monthly mean variability did not exceed values greater than 0,16. The months that presented lower concentrations had an average AOD of 0,04. The AOD frequency histograms showed that the types of clean continental, marine and dusty aerosols have little variable values. The aerossol types that presented the greatest variation were the type of continental/smoke pollution. According to the RL analysis, Natal presented values that characterize dusty, burnt, polluted continental and mainly clean marine aerosols. The aerosol layers top and base are at their lowest altitudes during May, June and July, these being the months in which the rainy season occurs, and also during November. The minimum values are always below 1 km of altitude, the months of February and September obtained 25% of the altitudes above 2,5 km, with altitudes of up to 4,9 km. The RL values of 70 sr and the highest values of maximum and minimum height of the aerosol layers, highlighted for the month of September, suggest the hypothesis of cases of biomass burning aerosol detection at high altitudes above the boundary layer, motivating a case study, which showed, on 09/07/2009, 09/13/2011, 09/07/2013 and 09/15/2016, smoke aerosols located at an altitude of 3 to 4 km. With trajectories modeled by HYSPLIT, for 7 days, it can be concluded that these air masses are coming from central Africa, over the Natal atmosphere. Comparison of the DUSTER/Lidar data with the CALIPSO data of 12/06/2016 showed aerossol layers between 1 and 3 km, coinciding in both measurements. The CALIPSO backscatter profile also allowed to observe a layer of dust aerosols at the unusual height of 15 km, suggesting further future studies for interpretation.

O resultado do escaneamento de um Laser Scanner Terrestre (LST) é uma nuvem de pontos com coordenadas geométricas (X, Y, Z), informações de cor (R, G, B) provenientes de uma câmera fotográfica acoplada ao equipamento e, ainda, a informação do retorno da intensidade do pulso laser (I). Esses sistemas de varredura possuem algumas características como, por exemplo, sua rapidez na aquisição de informações, registro de cenas em 3D e coleta de informações sem contato direto que se aplicam de forma importante nas análises florestais. Contudo, a grande vantagem da utilização de um LST na área florestal é a possibilidade de caracterizar alvos remotamente de forma rápida e não destrutiva. Assim, este trabalho teve como objetivo principal avaliar os dados de intensidade de retorno do laser provenientes de um sistema LST para a caracterização de alvos florestais. Metodologicamente foram realizados experimentos controlados que envolveram as seguintes etapas: calibração radiométrica do LST; avaliação da influência da distância nos dados de intensidade de retorno do laser e; análise do efeito de borda em imageamento de alvos florestais (considerado um dos principais problemas que afeta os dados intensidade de retorno quando se utiliza um LST). O equipamento utilizado durante os experimentos foi um laser scanner modelo Ilris 3D da Optech que trabalha no intervalo do infravermelho médio com comprimento de onde de 1535 nm. Os resultados mostraram que para esse comprimento de onda os alvos florestais devem ser imageados a uma distância maior ou igual a 5m e o processamento dos dados com resolução radiométrica de 8 bits foi mais adequado, pois proporcionou uma caracterização geométrica do alvo com efeito visual de melhor qualidade se comparado com o processamento de 16 bits. Os resultados dos experimentos realizados sobre o efeito de borda possibilitaram identificar dois tipos de distorções que ocorrem em dados de nuvem de pontos adquiridos com um LST. O primeiro afetou os valores de intensidade de retorno do laser e o segundo criou um efeito que deslocou os pontos no espaço. Para minimizar este efeito foi desenvolvido um algoritmo, o IRA (Intensity Recovery Algorithm), que possibilitou recuperar automaticamente os valores de intensidade de retorno do laser minimizando em até 35,7% o efeito de borda nos imageamentos do alvo estudado na pesquisa. Assim para o uso de um LST, na caracterização geométrica de alvos florestais, é necessário desenvolver modelos de calibração da intensidade de retorno do pulso laser, pois os sistemas LST são distintos em termos de faixa do espectro eletromagnético que operam. Por fim, no que tange ao efeito de borda concluiu-se que o algoritmo IRA necessita ser aprimorado com outras abordagens computacionais e matemáticas que poderão ser desenvolvidos em estudos futuros. / The result of the scanning of a terrestrial laser scanner (TLS) is a point cloud with geometric coordinates (X, Y, Z), color information (R, G, B) from a camera coupled to the equipment, and also the return information of intensity of the laser pulse (I). These scanning systems have some characteristics, for example, its speed in acquiring information, and of 3D scenes with record of data remotely which applies significantly in the forestry analysis. The advantage of using a TLS in the forestry area is the possibility of remote acquisition of data enabling a fast and non-destructive work. This work aimed to evaluate the laser return intensity data from a TLS system for the characterization of forest targets. Methodologically were performed controlled experiments involving the following steps: radiometric calibration of TLS; evaluating the influence of the distance in the laser return intensity data and; analysis of the edge effect in imaging forest targets (considered one of the main problems that affect the data intensity return when using a TLS). The equipment used during the experiments was a laser scanner Ilris Optech 3D model that works in the mid-infrared range with wavelength of 1535 nm. The results showed that for this wavelength forest targets should be imaged at a distance greater than or equal to 5m and processing of the radiometric data with 8-bit resolution is more suitable because it provided a geometric characteristics of the target with better visual effect quality compared with the 16-bit processing. The results of the experiments on the edge effect possible to identify two types of distortions that occur in cloud data points acquired with a LST. The first affect the laser return intensity values and the second set offset an effect that the points in space. To minimize this effect an algorithm, the IRA (Intensity Recovery Algorithm), was developed which enabled automatically retrieve the laser return intensity values up to 35.7% of minimizing the edge effect in the target imaging surveys studied in research. Thus for use of an TLS, the geometrical characterization of forest targets, it is necessary to develop calibration models of the return laser pulse intensity, for TLS systems are different in terms of the electromagnetic spectrum operating range. Finally, with respect to the edge effect it was concluded that the IRA algorithm needs to be enhanced with other computational and mathematical approaches that may be developed in future studies.