Um porto é uma organização dinâmica que possui áreas críticas que abrigam fontes emissoras de poluentes, como indústrias em seu entorno, modais terrestres (caminhões e trens), operações com carga e descarga de granéis agrícolas, fluxo de navios, entre outros. Neste estudo foi utilizada a modelagem desenvolvida em 2009 pela Current Methodologies in Preparing Mobile Source Port-Related Emission Inventories da Agência Americana de Proteção ao Meio Ambiente (US EPA), que estima os fatores de emissão de material particulado por navios considerando a energia, a característica e o tempo de atividade de cada embarcação, o fator de carga de propulsão e o fator de emissão para cada poluente. Para estimar as emissões fugitivas de material particulado provenientes de operações portuárias com milho, grãos de soja e farelo de soja foi utilizada a modelagem desenvolvida em 2006 pelo Group Emissions Factor US EPA, que considera o volume de carga movimentada, velocidade do vento, umidade do material e fator de tamanho das partículas. Os resultados, referente aos meses de maio e agosto de 2017, em conjunto com os dados coletados em 16 de maio de 2017 pela técnica LIDAR, indicam que as emissões de material particulado pelos navios têm sua maior taxa de emissão para MP10 (moda grossa). Para as emissões fugitivas em operações com grãos, os resultados mostraram que as maiores emissões são provenientes de operações de soja em grãos. Em suporte às modelagens, foi utilizada a técnica LIDAR (Light Detection And Ranging) de sensoriamento remoto que utiliza radiação luminosa, e está baseada na emissão de um feixe de radiação eletromagnética que permite a obtenção de medidas a distância através da radiação retroespalhada. O pulso de luz emitido para a atmosfera interage com as partículas e moléculas que agem como espalhadoras. Os dados coletados são inseridos no programa Mathematica, e os valores obtidos inseridos na plataforma PYTON gerando gráficos. Os dados foram coletados em 16 de maio de 2017, com medidas em diferentes horários e identificada a possível presença de material particulado (MP) na região do porto. O porto de Santos, está localizado no centro do litoral do estado de São Paulo, é o maior porto da América Latina e um dos principais conflitos nessa região são as emissões de material particulado. / A harbor is a dinamic organization that has critical áreas that contain pollutants sources such as insdutries in its surroundings, terrestrial modal (trucks and trains), operations with loading and unloading of agricultural bulk, ship flow, among others. This study used the model developed by the Current Methodologies in Preparing Mobile Source of the US Environmental Protection Agency (US EPA), which estimates the emission factors of particulate matter by ships considering the energy, characteristic and time of activity of each vessel, the propulsion load factor and the emission factor for each pollutant. To estimate fugitive emissions of particulate matter from port operations with corn, soybeans and soybean meal was used to model developed by Group Emissions Factor US EPA, which considers the volume of load handled, speed wind, humidity of the material and particle size factor. The results, referrinf to the months May and August of 2017, together with the data collected on May 16, 2017 by the LIDAR technique, indicate that emissions of particulate matter by ships have their highest emission rate for PM10 (coarse mode). For fugitive emissions of particulate matter in grain operations, the results showed that the highest emissions are from soybean grain operations. In support of modeling, It was used the LIDAR technology (Light Detection And Ranging) remote sensing which uses light radiation, and is based on the emission of an electromagnetic radiation beam that allows to obtain measurements at a distance by backscattering radiation. The pulse of light emitted into the atmosphere interacts with the particles and molecules that act as spreaders. The collected data are inserted into the Mathematica program and the values obtained are inserted into the PYTON platform generating graphs. The data were collected on May 16, 2017, with measurements at different times and identified the possible presence of particulate matter (PM) in the harbor region. The harbor of Santos, is located in the center of the coast of the state of São Paulo, is the bigger harbor in Latin America and one of the main conflicts in this region are emissions of particulate matter.

O Laboratório de Aplicações Ambientais a Laser do Centro de Lasers e Aplica- ções do IPEN realiza o estudo de propriedades atmosféricas, tais como os coecientes de extinção e retroespalhamento. Esses coecientes são estimados por meio de um método inverso, cuja qualidade não é facilmente certicada, ou seja não existe um diagnóstico sistemático da qualidade das estimativas. Neste trabalho um método inverso com sólidas bases estatísticas é proposto, para recuperar os mesmos coecientes. O algoritmo proposto oferece vantagens comparada com o método atual em uso pelo grupo, pois: (1) possibilita a incorporação de diferentes tipos de informações de fontes heterogêneas durante a inversão. E (2) o método permite diversos tipos de diagnósticos quanto a qualidade do processo de inversão e do parâmetro recuperado. Fatores que potencializam as chances de estimativas com maior acurácia. v / The Laboratory of Environmental Applications of Lasers at Ipen realizes a study about atmospherics properties, such as extinction and backscattering coecient. These coecients are estimated by an inverse method, whose estimate quality is dicult to measure. This work presents a method with good statistic approach to retrieval the same coecients. The new method, however, oers a number of advantages compared to the rst method in use, including (1) the ability to incorporate dierent kinds of information under a common retrieval philosophy and (2) the method provides number of ways for evaluating the quality of the retrieval. Thus we hope improve the accuracy of estimates. vi

A cobertura de nuvens e os aerossóis são os dois principais fatores que modulam a energia solar que atinge a superfície e é absorvida pela atmosfera. Esses dois fatores, portanto, têm um papel essencial no clima do planeta. Há atualmente um interesse nos efeitos radiativos dos aerossóis, particularmente por causa da atividade antrópica, que aumenta sua concentração na atmosfera, e por sua íntima relação com a formação de nuvens. Partículas que podem ser ativadas e ganhar água para se tornarem nevoeiro ou gotas de nuvens, na presença de supersaturação de vapor de água, são chamadas de Núcleo de Condensação de Nuvens. O estudo das partículas que aumentam de tamanho com o ganho de água conforme aumenta a umidade relativa (higroscopicidade) torna-se então de fundamental importância no entendimento da contribuição dos aerossóis na regulação do clima do planeta. O LIDAR é um instrumento promissor no estudo da higroscopicidade dos aerossóis atmosféricos, porque pode operar em ambiente não perturbado e em condições muito próximas da saturação. O LIDAR Raman apresenta a vantagem de poder obter o perfil de vapor de água e retroespalhamento de aerossóis no mesmo volume atmosférico e sem nenhuma suposição a priori a respeito da razão LIDAR Este trabalho objetiva avaliar o crescimento higroscópico do material particulado urbano em São Paulo, Brasil, e em Washington, D.C, Estados Unidos durante a campanha NASA-Discover-AQ com o uso da técnica LIDAR Raman, obtendo o fator de crescimento por higroscopicidade. Apesar da metodologia que se baseia na determinação de uma atmosfera bem misturada com o uso de radiossondagem já ter sido utilizada na literatura, este trabalho acrescenta importantes informações, já que não se tem notícias de outros estudos extensos com múltiplos casos feitos com o LIDAR Raman para avaliação de higroscopicidade nos Estados Unidos, bem como é a primeira vez que este estudo é feito em São Paulo, como resultado de três anos de aquisição de dados. Os resultados mostram que é possível identificar o crescimento higroscópico dos aerossóis em ambas os ambientes de estudo, cuja detecção depende de condições atmosféricas que estão raramente presentes, tornando o estudo da higroscopicidade com o LIDAR um desafio que exige uma extensa coleta de dados. Mostram ainda que a determinação da origem e o estudo conjugado das propriedades químicas da população de aerossóis são informações que auxiliariam no entendimento do comportamento higroscópico, aprofundamentos estes que podem ser derivados deste trabalho. / The cloud cover and aerosols are two main factors that modulate the solar energy that reaches the surface and is absorbed by the atmosphere. These two factors therefore have a key role in global climate. There is currently an interest in the radiative effects of aerosols, particularly because of human activity, increasing its concentration in the atmosphere, and its close relationship with cloud formation Particles that can be activated and gain water to form fog or cloud droplets in the presence of water vapor supersaturation, are called Cloud Condensation Nucleus. The study of particles that increase in size because of the uptake of water under increasing relative humidity conditions (hygroscopicity) then becomes of fundamental importance in understanding the contribution of aerosols in regulating the global climate. The LIDAR is a promising tool in the study of hygroscopic properties of atmospheric aerosols, because it can operate in undisturbed environment and much close to saturation conditions. The Raman LIDAR has the advantage of being able to obtain the profile of water vapor and aerosol backscatter at the same atmospheric volume and no a priori assumption of the LIDAR ratio. This study aims to evaluate the hygroscopic growth of urban particulate matter in Sao Paulo, Brazil, and Washington, DC, United States during the NASA-Discover-AQ campaign using the technique Raman LIDAR, getting the growth factor by hygroscopicity . Although the methodology - which is based on determination of a well-mixed atmosphere using radiosonde - have already been described in the literature, this work adds important information, since they do not have other news with multiple cases extensive studies made with the LIDAR Raman hygroscopicity for evaluation in the United States, and is the first time that this study is done in São Paulo, as a result of three years of data acquisition. The results show that it is possible to identify the hygroscopic growth of aerosol in both environments, whose detection depends on atmospheric conditions that are rarely present, making the study of hygroscopic properties with LIDAR a challenge that requires extensive data collection. Also, show that the determination of origin and the combined study of chemical properties of the population of aerosols would assist in the understanding of hygroscopic behavior.

Light Detection And Ranging (LiDAR) har under det senaste decenniet utvecklats mycket och används som datainsamlingsmetod vid inventering av skog. Lantmäteriet genomför mellan åren 2009 och 2015 en riksomfattande flygburen laserskanning över hela Sverige och den laserskanningen ska leda till en ny nationell höjdmodell (NNH). Data som genereras från denna höjdmodell kan nyttjas av skogsindustrin för att göra skogliga inventeringar. Programvaruutvecklaren Esri Inc. har utvecklat ett stöd för hantering av denna typ av data i deras nya version av ArcGIS, ArcGIS 10.1. Syftet med denna studie är att undersöka hur skogsindustrin i Sverige använder LiDAR-data, vilka brister och behov som finns, hur ArcGIS används och hur ArcGIS kan utvecklas för att matcha skogsindustrins behov. I denna studie genomfördes dels en jämförelse mellan ArcGIS 10.1 med den tidigare versionen ArcGIS 10.0 och dels intervjuer av sex betydelsefulla personer som är verksamma i branschen. Personer från alla intressegrupper intervjuades; skogsägare, forskare och tekniska konsulter.Resultatet från jämförelsen mellan ArcGIS 10.0 och ArcGIS 10.1 visar potentialen med LiDAR-data och hur enkelt och smidigt det är att hantera LiDAR-data i den nya versionen av ArcGIS. Resultaten från intervjuerna visar att de data som finns tillgängligt från NNH är fullt tillräckliga för att göra skoglig inventering på beståndsnivå. Det är däremot inte tillräckligt för att göra analys på enskilda träd. Några av de intervjuade upplever ArcGIS som ett avancerat och tidskrävande program att lära sig medan andra framhäver att det ska bli intressant med stöd för hantering av LiDAR-data. De konsulter som intervjuats använder sig främst av egenutvecklade programvaror för hantering av LiDAR-data medan övriga intervjuade använder ArcGIS som huvudprogram. Esri Sverige visste inte riktigt hur kvaliteten på NNH var och i hur stor utsträckning skogsindustrin använder sig av NNH innan denna studie. Men det visade sig att NNH-data används i stor utsträckning redan och att kvaliteten är fullt tillräcklig för den information skogsindustrin vill ha. ArcGIS upplevs ibland som ett avancerat program och det har ofta att göra med tidsbrist. Finns tiden kan det mesta lösas. LiDAR, och framför allt NNH är bra och användbart nu, men frågan är vad som händer när det är dags att göra nästa inventering för skogsbolagen? / Light Detection And Ranging (LiDAR) has in the past decade developed a lot and is used as a data collection method for inventory of forest. The Swedish National Land Survey is between 2009 and 2015 carrying out a nationwide airborne laser scanning throughout Sweden, and this laser scanning process will lead to a new national elevation model called NNH. Data generated from this height model can be used by the forest industry to make forest inventories. The software developer ESRI Inc. has developed a support for handling LiDAR data in their new version of ArcGIS, ArcGIS 10.1. The purpose of this study is to investigate how the forest industry in Sweden are using LiDAR data, identifying gaps and needs, how ArcGIS is used and how ArcGIS can be developed to match the forest industry. This study was founded by comparing the new version of ArcGIS with the previous version and by interviewing six relevant people who are active in the industry. People from several stakeholders were interviewed: foresters, researchers and technical consultants.The results of the comparison between ArcGIS 10.0 and ArcGIS 10.1 show the potential of LiDAR data and how easy it is to deal with LiDAR data in the new version of ArcGIS. The results of the interviews show that the data available from the NNH are fully sufficient for forest inventory at stand level. It is however not sufficient for analysis of individual trees. Some of those interviewed experienced ArcGIS as an advanced and time-consuming program to learn while others emphasize that it will be interesting with support for managing LiDAR data. The interviewed consultants mostly use software that they have developed by themselves for managing LiDAR data, while other interviewees use ArcGIS as the main program. ESRI Sweden was not sure of the quality of the NNH and how much the forest industry uses NNH, before this study. But it turned out that the NNH data are widely used already, and that quality is adequate. ArcGIS is sometimes perceived as an advanced program and it has often to do with time constraints. If time is available the problem often can be fixed. LiDAR, and NNH is good and useful now, but the question is what will happen when it's time for the forest companies to make their next inventory?

To assess information about the optical, microphysical, and radiative properties of aerosol particles the lidar technique and sun photometers are commonly used. Information that result from both lidar and sun photometer data can provide a distinct image of the vertical aerosol properties. The algorithm developed at the Institute of Physics of the National Academy of Science of Belarus (IPNASB) uses lidar measurements at the three wavelengths 355, 532, and 1064 nm and mean backscatter and extinction coefficients retrieved from radiometric data to obtain profiles of fine-mode and coarse-mode concentrations. Within the master thesis the IPNASB algorithm was tested for specific aerosol situations. Three cases are considered representing Saharan dust, smoke and industrial aerosol from East Europe, and volcanic aerosol from the Eyjafjallajokull eruption. The retrieved microphysical aerosol properties are in good to acceptable agreement with findings of well-established methods.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

En atmosphère naturelle, l’eau est l’un des facteurs contribuant fortement à la masse des particules d’aérosol. Ceci va fortement modifier les propriétés optiques et radiatives des aérosols. Cet impact a été calculé à partir de plus de 2 ans de mesures sur le site ACTRIS/GAW du Puy de Dôme (PdD, 1565m). La distribution en taille, l’extinction et l’hygroscopicité mesurées au PdD ont été combinées pour calculer les propriétés optiques et radiatives, sèches ou humides. Pour chaque propriété, le facteur d’accroissement hygroscopique (f) a été estimé à l’aide d’un code de Mie. La longue série temporelle a permis de paramétrer l’évolution en fonction de l’humidité des propriétés optiques de différents types d’aérosols. Pour un aérosol d’origine océanique qui s’est mélangé avec des aérosols anthropiques, le coefficient de diffusion augmente plus de 4.4 fois si il est placé à 90% d’humidité. Le forçage radiatif va évoluer en conséquence et sera 2.8 fois plus élevé à 90% d’humidité que pour une atmosphère sèche (1.8 à humidité ambiante). Cette longue série de mesures a également permis de montrer, pour la première fois à notre connaissance, la forte variation saisonnière de ce paramètre. Ce manuscrit présente également une méthode originale pour calculer la masse d’aérosols volcaniques à partir de mesures LIDAR et in situ. Lors de l’éruption du volcan Islandais Eyjafjalla en Mai 2010, la masse d’aérosols volcaniques au dessus de Clermont a été estimée de 655±23μg.m-3. / Water contributes significantly to the aerosol mass under ambient conditions of relative humidities, and thus may significantly impact their optical and radiative properties and direct effect. In the present work, the impact of the aerosol hygroscopicity on its optical properties is evaluated from a set of instrumentation located at the ACTRIS/GAW Puy de Dôme station (PdD, 1465m) over two years of measurements in 2010 and 2011. In situ size distributions, extinction and hygroscopicity measurements are combined to retrieve the aerosol refractive index, both dry and wet. For each optical property the enhancement factor (f) due to hygroscopicity can be computed using Mie calculations. The long data set available enables us to generate parameterisations of optical properties enhancement as a function of relative humidities for different aerosol types. At 90% humidity, fσsca is more than 4.4 for marine aerosol that have mixed with a pollution plume. Consequently, the aerosol radiative forcing is estimated to be 2.80 times higher at RH=90% and 1.75 times higher at ambient RH when hygroscopic growth of the aerosol is considered. For the first time this study highlight the high seasonal variability of this impact. The manuscript also presents an original method of mass inversion for volcanic aerosols with the synergy of in situ and LIDAR measurements. The calculated mass of volcanic particle transported over Clermont during the eruption of Eyjafjalla in May 2010 was up to 655±23μg.m-3.

O Laboratório de Aplicações Ambientais a Laser do Centro de Lasers e Aplica- ções do IPEN realiza o estudo de propriedades atmosféricas, tais como os coecientes de extinção e retroespalhamento. Esses coecientes são estimados por meio de um método inverso, cuja qualidade não é facilmente certicada, ou seja não existe um diagnóstico sistemático da qualidade das estimativas. Neste trabalho um método inverso com sólidas bases estatísticas é proposto, para recuperar os mesmos coecientes. O algoritmo proposto oferece vantagens comparada com o método atual em uso pelo grupo, pois: (1) possibilita a incorporação de diferentes tipos de informações de fontes heterogêneas durante a inversão. E (2) o método permite diversos tipos de diagnósticos quanto a qualidade do processo de inversão e do parâmetro recuperado. Fatores que potencializam as chances de estimativas com maior acurácia. v / The Laboratory of Environmental Applications of Lasers at Ipen realizes a study about atmospherics properties, such as extinction and backscattering coecient. These coecients are estimated by an inverse method, whose estimate quality is dicult to measure. This work presents a method with good statistic approach to retrieval the same coecients. The new method, however, oers a number of advantages compared to the rst method in use, including (1) the ability to incorporate dierent kinds of information under a common retrieval philosophy and (2) the method provides number of ways for evaluating the quality of the retrieval. Thus we hope improve the accuracy of estimates. vi

Ce travail de thèse a pour objectif la prédiction des paramètres structurels des forêts à grande échelle, grâce aux images de télédétection. La démarche consiste à étendre la précision des données LIDAR spatiales, en les utilisant là où elles sont disponibles, en tant que donnée d'apprentissage pour les images radar à synthèse d'ouverture polarimétriques et interférométrique (PolInSAR). A partir de l'analyse des propriétés géométriques de la forme de cohérence PolInSAR, nous avons proposé un ensemble de paramètres susceptibles d'avoir une corrélation forte avec les profils de densité LIDAR en milieu forestier. Cette description a été utilisée comme données d'entrée de techniques SVM, de réseaux de neurones, et de forêts aléatoires, afin d'apprendre un ensemble de descripteurs de forêts issus du LIDAR : la hauteur totale, le type de profil vertical, et la couverture horizontale. L'application de ces techniques à des données réelles aéroportées de forêts boréales en Suède et au Canada, et l'évaluation de leur précision, démontrent la pertinence de la méthode. Celle-ci préfigure les traitements qui pourront être appliqués à l'échelle planétaires aux futures missions satellites dédiées à la forêt : Biomass, Tandem-L et NiSAR. / The objective of this thesis is to predict the structural parameters of forests on a large scale using remote sensing images. The approach is to extend the accuracy of LIDAR full waveforms, on a larger area covered by polarimetric and interferometric (PolInSAR) synthetic aperture radar images using machine learning methods. From the analysis of the geometric properties of the PolInSAR coherence shape, we proposed a set of parameters that are likely to have a strong correlation with the LIDAR density profiles on forest lands. These features were used as input data for SVM techniques, neural networks, and random forests, in order to learn a set of forest descriptors deduced from LIDAR: the canopy height, the vertical profile type, and the canopy cover. The application of these techniques to airborne data over boreal forests in Sweden and Canada, and the evaluation of their accuracy, demonstrate the relevance of the method. This approach can be soon be adapted for future satellite missions dedicated to the forest: Biomass, Tandem-L and NiSAR.

Seit 2015 gibt es Doppler-LIDAR-Messungen der Windgeschwindigkeit über Leipzig. Diese Messungen werden zum einen vom HALO Photonics Streamline Doppler-LIDAR am Leipziger Institut für Troposphärenforschung (TROPOS) sowie vom Leosphere Windcube 8 Doppler-LIDAR am Leipziger Institut für Meteorologie (LIM) aufgenommen. In dieser Arbeit werden insbesondere die Daten des Windcubes bezüglich der horizontalen Windgeschwindigkeit, der Windrichtung und der Turbulenzintensität bis in eine Höhe von ca. 500 m ausgewertet. Der Vergleich mit dem HALO Photonics Streamline Doppler-LIDAR zeigt gute Korrelationen zwischen beiden Geräten. Die Ergebnisse beinhalten unter anderem eine gute Übereinstimmung mit der Ekman-Theorie. Auch die erhöhte Rauhigkeit der Stadtfläche gegenüber dem ländlichen Raum wird in der Auswertung deutlich. Es wurde ein Verfahren getestet, das es erlaubt, den turbulenten Diffusionskoeffizienten und die aerodynamische Rauhigkeitslänge aus den Messwerten abzuleiten und erste Abschätzungen dieser Größen vorzunehmen. Diese Arbeit legt nahe, den Einsatz der Fernerkundungsmessung zur Bestimmung der urbanen Grenzschichtdynamik fortzusetzen und wenn möglich zu erweitern. / Since 2015 Doppler-LIDAR measurements above Leipzig are available. The Leibniz Institute for Tropospheric Research (TROPOS) operates a HALO Photonics Streamline Doppler-LIDAR while the Institute of Meteorology Leipzig uses a Leosphere Windcube 8 Doppler-LIDAR. In this study mainly meausrements of the Windcube for horizontal wind velocity, wind direction and turbulence intensity below 500 m are evaluated. The comparison to the HALO Photonics Streamline LIDAR shows good correlations between both devices. Among others, the results include good agreements with the Ekman theory. Additionally the increased roughness of the city surface compared to rural areas becomes apparent. A way to determine characteristic quantities like the turbulent diffusion coefficient and the aerodynamic roughness length is described and initial estimations were conducted. This study suggests to continue and if possible expand such remote sensing measurements for analyses of urban boundary layer dynamics.