Modélisation de la dispersion atmosphérique des métaux lourds émis par l'industrie énergétique polonaise.De nombreuses études ont mis en évidence les impacts néfastes de la contamination en métaux lourds de l'atmosphère. Une attention particulière est porté au mercure, au plomb et au cadmium qui font l'objet du protocole d'Aarhus sur les métaux lourds (1998). Les parties signataires (incluant la Pologne) s'obligent à réduire les émissions et à obsrver et suivre les niveaux de contamination dans l'environnement.La Pologne est un des plus gros émetteurs de mercure, de plomb et de cadmium en Europe du fait de l'exploitation de centrale à charbon pour la production d'énergie. Des efforts de recherche sont à réaliser pour étudier la contamination par les métaux lourds. Les objectifs de ce travail sont :• de développer et utiliser un modèle pour représenter la dispersion atmosphérique du mercure dans la plate-forme de modélisation numérique Polyphemus.• d'étudier la dispersion en Europe et en Pologne des émissions du secteur de l'industrie énergétique.Pour satisfaire ces objectifs un nouveau schéma chimique pour le mercure a été implémenté dans la plate-forme de modélisation numérique Polyphemus. Une revue de la littérature scientifique a montré que la chimie du mercure restait mal connue. Cela implique des différences dans les modèles établis jusqu'à présent. Ceux-ci diffèrent également de par leur façon de représenter les processus de retrait de l'atmosphère. Le mercure gazeux élémentaire (GEM) présente des concentrations relativement uniformes dans l'espace. Les concentrations des formes oxydées et les dépôts au contraire présentent de forts gradients dans l'espace.Ce travail a mis en évidence la sensibilité des résultats à plusieurs composantes du modèle. Dans le schéma chimique, les chemins d'oxydation du GEM par le radical hydroxyle, l'ozone et les composés bromés dominent en phase gazeuse. En phase aqueuse les réactions de réduction dominent. Toutes ces réactions ont un impact critique sur les concentrations et les dépôts des formes oxydées du mercure mais faible sur le GEM. Dans ce modèle, le dépôt sec du GEM est un des principaux processus de retrait du mercure atmosphérique. Ce relativement fort flux de dépôt de GEM diffère significativement des précédents modèles qui souvent néglige ce processus. Il a également était mis en évidence que les flux de dépôt humide sont sensibles aux paramètres du modèle de lessivage.Les résultats de simulation montrent que l'industrie énergétique est responsable de 80% des dépôts humides de mercure près des grosses sources ponctuelles. La contribution des émissions nationales sur l'ensemble du pays se limite toutefois à un peu plus de 21% pendant l'hiver, lorsque de grandes quantités de charbon sont brûlées. L'impact des émissions de cadmium et de plomb est comparativement plus faible. Les simulations montrent un impact maximal de 30% et 10% pour le cadmium et le plomb respectivement à proximité des sources.De manière générale de plus nombreuses observations de concentrations et de dépôt de métaux lourds seraient utiles pour une meilleure évaluation des modèles / Modelling of atmospheric transport of heavy metals emitted from Polish power sectorMany studies have been conducted to investigate the atmospheric heavy metals contamination and its deposition to ecosystems. The increasing attention to mercury pollution has been mainly driven by the growing evidence of its negative impacts on wildlife, ecosystems and particularly human health. Lead and cadium are also toxics which are being emitted into the atmosphere by anthropogenic as well as natural sources. The harmful influence of these three heavy metals was underlined in the Aarhus Protocol on Heavy Metals of 1998. The Parties of this protocol (including Poland) are obligated to reduce emissions, observe the transport and the amounts of lead, mercury and cadmium in the environment.Poland is one of the biggest emitter of mercury, lead and cadmium in Europe mainly due to emission from coal combustion processes. Therefore in Poland, research efforts to study the heavy metals emission, atmospheric transport, concentration and deposition are extremely important. The objectives of this work were twofold:• The practical objective was to develop and run a model to represent the atmospheric dispersion of mercury and to implement it in the air quality modelling platform Polyphemus.• The scientific objective was to perform heavy metals dispersion studies over Europe and detailed studies of the impact of the polish power sector on the air quality regarding mercury, cadmium and lead.To meet the declared aim, a new mercury chemical model was implemented into the Polyphemus air quality system. The scientific literature was reviewed regarding mercury chemistry and mercury chemical models. It can be concluded that the chemistry of mercury is still not well known. The models also differ in the way of calculating the dry and wet deposition of mercury. The elemental gaseous mercury ambient concentrations are evenly distributed, on the contrary, high variations in the spatial gradients of reactive gaseous and particulate forms of mercury air concentrations and deposition fluxes were noted.This study shows that many components of the developed model have crucial impacts on the results. In the model of mercury chemistry, the most effective pathway in the gaseous phase are the oxidation of gaseous elemental mercury by hydroxyl radical, ozone and most of all bromine oxide radicals, while in the aqueous phase the reduction reactions of elemental mercury dominate. These reactions have a crucial influence on the mass balance of reactive mercury, but a rather low influence on gaseous elemental mercury (GEM). In this model, dry deposition of gaseous elemental mercury is the prevailing process for removing mercury from the atmosphere. Dry deposition of GEM over land is equally distributed, due to almost uniform ambient concentrations. The relatively high dry deposition flux of gaseous elemental mercury has a huge influence on the presented results and differs significantly from other models where the dry deposition of GEM is often not taken into account. It was also shown that a change in the calculation of the scavenging coefficients or representative raindrop diameter for in-cloud scavenging has a significant impact on the amount of wet deposited mercury.The detailed results show that the Polish power sector can be responsible for up to 80% of wet deposited mercury near large emission sources. The contribution from national sources over whole Poland reaches 21% during the winter heating season when large quantities of coal are burned in the domestic sector and additionally the power sector activity is at its highest. The impact of emissions of cadmium and lead from the power sector is lower compared to the obtained results for mercury. The modelling results showed maximal impacts of 30% and 10% for cadmium and lead near large power sector sources on the ambient concentrations and deposition. Wider measurements of heavy metals are very much needed

No description available.

A large epizootic of the vector-borne disease bluetongue occurred in northern Europe from 2006-2009, costing the economies of the infected countries several hundreds of millions of euros. During this time, the United Kingdom (UK) was exposed to the risk of bluetongue by windborne incursions of infected Culicoides biting midges from the northern coast of mainland Europe. The first outbreaks which occurred in the UK in 2007 were attributed to this cause. Although bluetongue virus (BTV) no longer appears to be circulating in northern Europe, it is widely suggested that it and other midge-borne diseases may emerge again in the future, particularly under a changing climate. Spread of BTV is strongly influenced by the weather and climate however limited use has been made of meteorologically based models to generate predictions of its spread to the UK. The extent to which windborne BTV spread can be modelled at timescales from days to decades ahead, to inform tactical and strategic decisions taken to limit its transmission, is therefore examined here. An early warning system has been developed to predict possible incursion events on a daily timescale, based on an atmospheric dispersion model adapted to incorporate flight characteristics of the Culicoides vectors. The system’s warning of the first UK outbreak in September 2007 was found to be greatly beneficial to the UK livestock industry. The dispersion model is also shown to be a useful post-outbreak epidemiological analysis tool. A novel approach has been developed to predict BTV spread into the UK on climate-change timescales as dispersion modelling is not practical over extended periods of time. Using a combination of principal component and cluster analyses the synoptic scale atmospheric circulations which control when local weather conditions are suitable for midge incursions were determined. Changes in the frequency and timing of these large scale circulations over the period 2000 to 2050 were then examined using an ensemble of regional climate model simulations. The results suggest areas of UK under the influence of easterly winds may face a slight increase in risk and the length of the season where temperatures are suitable for BTV replication is likely to increase by around 20 days by 2050. However a high level of uncertainty is associated with these predictions so a flexible decision making approach should be adopted to accommodate better information as it becomes available in the future.

In the context of an increasing production of industrial toxic substances (TIC; Toxic Industrial Compound), the risk of accidental release of hazardous substances is growing in spite of the gradual implementation of safer technological processes and safety improvement measures. To mitigate the consequences of chemical accidents or to prepare preventive protective measures before the accident, it is necessary to know or at least estimate the course of accidents. In particular, it applies to the range of traumatic events and fatal accidents. One of the tools that can express the impact of accidents is modelling programs. This diploma thesis presents a comparison of outputs from two special types of software ? a foreign program the ALOHA and the TerEx developed in the Czech Republic. The purpose of the thesis was to indicate theoretical aspects related to gaseous toxic substances diffusion in the ground atmospheric layer and to describe modelling of their ill effects range. To achieve this objective, scientific literature and consultation with experts were used. The practical section of the thesis aimed at determining to what extent the results of both programs differ when initial conditions were identical. To achieve this goal, the intercomparison of outputs (e.g. hurtful concentration range) of the two programs that provided results for the same input data sets (type and quantity of hazardous substances, environmental temperature, wind speed, degree of cloud cover, weather stability class, type of ground surface) was used. After the assessment of the comparison it is possible to generalize the results stating that the software ALOHA compared to the TerEx is more conservative, which means that the ALOHA software provides longer anticipated ranges of danger. Therefore, it depends on the user (the person responsible ? crisis manager, intervention commander, mayor of the village), which approach they select or recommend ? to prepare a greater or a smaller area for a possible accident (to ensure public awareness, to implement technical measures to mitigate the impact of that accident, to assess the amount of financial resources, etc.). As a subsequent step it would be appropriate to verify the theoretical results experimentally, by field testing, which would be conducted under the same meteorological conditions under which the modelling was made by the mentioned programs. This would thus confirm the legitimacy of the special software use for the purpose of estimating the range of negative effects of chemical accidents.

Conselho Nacional de Desenvolvimento Científico e Tecnológico / In this thesis, we presents a development and validation of a 3D numerical model for the advection-diffusion equation. Models of this kind has been developed for scientific investigations and to support atmospheric emissions control and environmental policy decisions. To develop this model, we used the computational implementation of an explicit numerical scheme for the discretization of the envolved equations. During this procedures, exaustive tests were performed to ensure that the used implementations agrees to the stability, consistence and convergence criterias. As a way for minimizing one of the main deficiences found in almost the major atmospheric dispersion models, i.e. imprecisions in the meteorological input data for initializing this models, we used a realistic atmospheric flow field generated by mesoscale circulation model. As the mesoscale model gives information at scale larger than the necessary for description of a plume trajectory, a weighted linear average proper interpolation was developed for intermediate these distances. Our model considers the assumption that atmospheric turbulence is not isotropic, where diffusion coefficients are variables in time and space and are different for lateral and vertical directions. In our model we estimate this coefficients by the atmospheric boudary layer parameterizations proposed byMoraes (2000). For validation of themodel, we used experimental datasets from field experiment carried near a thermoelectric power plant presidente Médici, in the city of Candiota/RS. These datasets contains surface SO2 concentrations, surface wind velocity measured in meteorological towers as well as turbulence data measured in micrometeorological towers. The results of the validation indicates that the model works well, at least for the source and the terrain were it is located. i.e. continuous emission and homogeneous topography. / Nesta tese, apresentamos o processo de desenvolvimento e validação de um modelo numérico 3D para a equação de difusão-advecção. Modelos como este têm sido desenvolvidos objetivando a investigação científica para dar suporte ao controle de emissões de poluentes atmosféricos e à tomada de decisões no desenvolvimento de políticas ambientais. Para a elaboração deste modelo utilizamos a implementação de um esquema numérico explícito na discretização das equações envolvidas. Durante este processo, exaustivos testes foram realizados para que a implementação garantisse estabilidade, consistência e convergência. Como forma de minimizar uma das principais deficiências encontradas na maioria dos modelos de dispersão atmosféricos (MDA), qual seja, a imprecisão da entrada de dados meteorológicos para a inicialização destes modelos, utilizamos um campo de vento realístico gerado por um modelo numérico de circulação de mesoescala. No entanto, como este modelo de mesoescala fornece informações com uma escala maior do que a necessária para a descrição da trajetória de uma pluma, foi preciso desenvolver um método de interpolação apropriado para intermediar estas distâncias. Nosso modelo contempla, ainda, a hipótese da turbulência atmosférica não ser isotrópica, onde os coeficientes de difusão turbulenta são variáveis no tempo e espaço e, ainda, diferentes nas direções lateral e vertical. Em nosso modelo, estes coeficientes foram estimados utilizando a parametrização de camada limite proposta por Moraes (2000). Para a validação do modelo, utilizamos dados experimentais obtidos em um experimento realizado próximo à usina termelétrica Presidente Médici, no município de Candiota/RS. Estes dados compreendem medidas de concentração superficial de SO2, velocidade do vento em superfície medidos em torres meteorológicas, bem como dados de turbulência medido em torre micrometeorológica. Os resultados da validação mostraram que o modelo funciona bem, ao menos para o tipo de fonte e topografia onde está localizada, ou seja, emissão contínua e topografia homogênea.

Un programme expérimental a été conçu afin d'étudier la dispersion des polluants sur un terrain complexe avec un accent mis sur des conditions stables qui restent délicates pour la modélisation numérique. Ce programme expérimental est mené sur le site du SIRTA dans la banlieue sud de Paris et consiste à mesurer en champ proche la turbulence et la dispersion des polluants. L'objectif de ce programme est de caractériser la structure fine de la turbulence et de la dispersion associée par des mesures à haute résolution temporelle et spatiale. Ensuite, ces mesures permettent de valider et d'améliorer la qualité de simulations CFD pour la turbulence et la dispersion sur un site fortement hétérogène. Le dispositif instrumental comprend 12 anémomètres ultrasoniques mesurant en continu la vitesse du vent et la température à 10 Hz, et 6 détecteurs à photo-ionisation (PID) mesurant la concentration de gaz à 50 Hz pendant des essais de traçage. Plusieurs périodes d’observations intensives (POIs) avec des rejets de gaz ont été réalisées depuis Mars 2012.Tout d'abord, une étude détaillée de l'écoulement du vent sur le site est réalisée, car l’écoulement doit être caractérisé et correctement simulé avant de simuler la dispersion des polluants. Cette étude est basée sur deux ans de mesures en continu et sur les mesures acquises durant les POIs. La forte anisotropie de la turbulence dans la couche de surface est caractérisée à l’aide du calcul des variances, des échelles de longueur intégrales et des spectres des trois composantes de la vitesse du vent. La propagation des structures turbulentes entre les capteurs est caractérisée en utilisant les corrélations de vitesse. Les spectres de vitesse montrent plusieurs pentes dans différentes zones de fréquence. En outre, l’analyse des données montre l’impact de l'hétérogénéité du terrain sur les mesures. La forêt au nord du site expérimental modifie la vitesse et la direction du vent pour un grand secteur nord. Il induit un fort cisaillement de la direction du vent et une décélération en dessous de la hauteur de la forêt. Les simulations numériques sont effectuées avec le code de CFD Code_Saturne en mode RANS avec une fermeture k-ε adaptée pour les écoulements atmosphériques et un modèle de canopée pour la forêt. Ces simulations reproduisent correctement les caractéristiques de l'écoulement moyen sur le site des mesures, en particulier l'impact de la forêt pour les différentes directions du vent et pour la stratification neutre et stable. Les résultats de simulation montrent aussi le cisaillement de direction du vent et l’augmentation de l’énergie cinétique turbulente induits par la forêt. Une étude de sensibilité montre que ces effets sur l'écoulement s’accentuent quand la densité foliaire augmente. L’étude de dispersion est réalisée pour plusieurs POIs. L’analyse des données de concentration montre la cohérence avec les mesures de campagnes précédentes réalisées dans des zones proches de la source. Les fluctuations de concentrations sont caractérisées à travers les séries temporelles, l'histogramme et l'analyse statistique des concentrations. Une zone inertielle peut également être identifiée dans les spectres de concentration. Ensuite, la dispersion des polluants est modélisée par les équations de transport pour la concentration et sa variance. La concentration moyenne est globalement en bon accord avec les mesures pour toutes les POIs étudiées. L’accord avec les mesures sur la position du maximum de concentration dépend de la précision de la rotation du vent simulée en-dessous de la hauteur de la forêt. Les fluctuations de concentration obtenues dans les simulations semblent être affectées de manière significative par la condition initiale et la modélisation du terme de dissipation. Une étude de sensibilité à la paramétrisation est ensuite présentée / An experimental program has been designed in order to study pollutants dispersion at a complex site with a focus on stable conditions, which are still challenging for numerical modelling. This experimental program is being conducted at the SIRTA site in a southern suburb of Paris and consists in measuring, in near field, the turbulence and the pollutants dispersion. The aim of this program is to characterize the fine structure of turbulence and associated dispersion through high temporal and spatial resolution measurements. Then, these measurements allow to validate and improve the performance of CFD simulation for turbulence and dispersion in a heterogeneous field. The instrumental set up includes 12 ultrasonic anemometers measuring continuously wind velocity and temperature at 10 Hz, and 6 photo-ionization detectors (PIDs) measuring gas concentration at 50 Hz during tracer tests. Intensive observations periods (IOPs) with gas releases have been performed since March 2012.First of all, a detailed study of flow on the site is made, because it must be characterised and properly simulated before attempting to simulate the pollutants dispersion. This study is based on two years of continuous measurements and on measurements performed during IOPs. Turbulence strong anisotropy in the surface layer is characterized by calculating variances, integral length scales and power spectra of the three wind velocity components. Propagation of turbulent structures between sensors has been characterized with velocity correlations. Energy spectra show several slopes in different frequency regions. Also, data analyses show impact of terrain heterogeneity on the measurements. The forest to the north of experimental field modifies wind velocity and direction for a large northerly sector. It induces a strong directional wind shear and a wind deceleration below the forest height. Numerical simulations are carried out using the CFD code Code_Saturne in RANS mode with a standard k-ε closure adapted for atmospheric flows and a canopy model for the forest. These simulations are shown to reproduce correctly the characteristics of the mean flow on the measurements site, especially the impact of the forest for different wind directions, in both neutral and stable stratification. Simulations results also show the directional wind shear and the turbulent kinetic energy increase induced by the forest. A sensitivity study has been made for various values of forest density and shows that the typical features of canopy flow become more pronounced as canopy density increases. Pollutants dispersion study are made for several IOPs. Concentration data analysis shows a consistency with previous measurements made in a near-source region where the plume scale is smaller than the large-scale turbulence eddies. Concentration fluctuations are characterized through concentration time series, histogram and statistical analysis. The internal subrange can be observed in the concentration spectra. Next, pollutants dispersion are modelled by transport equations for concentration and its variance. The mean concentrations show a good agreement with measurements in values for all the IOPs studied, except that the position of the concentration peak depends on the accuracy of simulated wind rotation below the forest height. The concentration fluctuations obtained from simulations seem to be affected significantly by the initial condition and the modelling of the dissipation term. A sensitivity study to the parameterisation is then presented

An emission inventory tool for estimating SO2, NO2, and PM10 emissions from brick clamp kiln sites was developed from investigations performed on three representative South African clamp kiln sites in order to facilitate application for Atmospheric Emission Licenses (AELs) from these sources. The tool utilizes readily available site-specific parameters to generate emission factors for significant activities that emit the aforementioned pollutants. PM10 emission factors for significant processes were developed using empirical expressions from the Compilation of Air Pollutant Emission Factors (AP-42) documents. SO2 emission factor for clamp kiln firing was obtained from “reverse-modelling”, a technique that integrates ambient monitoring and dispersion modelling (using Atmospheric Dispersion Modelling System software) to “standardize” actual emission rate from an assumed rate of 1 g/s. The use of multiple point sources proved to improve the simulation of the buoyancy-induced plume rise; therefore, a “bi-point” source configuration was adopted for the kiln. The “reverse-modelling” technique and “bi-point” source configuration produced SO2 emission rates differing from -9 % to +22 % from mass balance results, indicating that the “reverse-modelling” calculations provide reliable emission estimates for SO2. An NO2 emission factor could not be obtained from the “reverse-modelling” technique due to experimental errors and the significant effect of NO2 emissions from other onsite air emission sources such as internal combustion engines. The NO2 emission factor was obtained from previous comprehensive study on a similar clamp kiln site. The emission factors obtained from this study were utilized in developing an “emission inventory tool” which is utilized by clay brick manufacturers in quantifying air emissions from their sites. Emissions quantification is a requirement for brick manufacturers to obtain an AEL which is regulated under South African environmental laws. It is suggested that the technique used here for SO2 emission confirmation could be used to estimate emissions from a volume or area source where combustion occurs and where knowledge of the source parameters is limited. / Dissertation (MSc)--University of Pretoria, 2013. / gm2014 / Chemical Engineering / unrestricted

In the event of a radioactive release in an urban environment the consequent response mustbe swift and precise. As soon as first responders have correct information, they can make anaccurate risk assessment. However, if the position, release rate and time of the radioactiverelease is unknown it is hard to know how the pollutant will spread. This thesis aims to testa model which approximates these three unknowns using weather data (wind and rain) as wellas measurement data collected at sensors placed around an urban environment. An atmospheric dispersion model based on an existing Reynolds Averaged Navier-Stokes modelis set up in two geometries of different complexity to create forward mode synthetic depositiondata and adjoint mode concentration fields resulting from a fixed dry deposition velocity andscavenging effect for wet deposition. Variations of time- and space-dependent rainfall is simu-lated. The resulting data is used in an existing optimization model, where a parameter studyis conducted regarding regularization coefficients. This thesis shows that the optimization model accurately estimates position and its approximaterelease rate of a 2D geometry of radioactive releases using a logarithmic optimization approach,and fail to do so using a linear optimization approach. The logarithmic optimization model alsoapproximately estimates position and release rate in a 3D geometry. Regularization parametersshould be within the range of 0.1 and 1.2 depending on rain. More rain requires smallerparameters and will estimate a lower release rate. Time-dependent rainfall is shown to have amajor negative effect on simulation time.iii

En cas d’accident impliquant des produits chimiques, il peut être fait appel à des experts pour évaluer les effets générés par cet accident. Ces experts fournissent des distances d’effets à l’aide de modélisations informatiques et sont confrontés à une difficulté majeure : peu d’éléments à leur disposition pour caractériser la situation.L’objectif de cette thèse est de proposer une méthodologie permettant de prendre en compte les incertitudes relatives aux données d’entrée dans les modélisations effectuées en situation d’urgence tout en restituant de manière explicite cette incertitude au gestionnaire de la situation d’urgence.Une première phase a consisté à évaluer, pour une situation à même de générer un nuage toxique ou explosible, la dispersion des résultats des modélisations. Il a été établi une hiérarchisation des variables d’entrée en fonction de leur influence sur le résultat final. Cette phase a été réalisée au moyen d’une analyse de sensibilité dont la stratégie a été spécifiquement développée.Une seconde phase a eu pour but d’établir une méthodologie d’estimation des distances d’effets en situation d’urgence en tenant compte du niveau d’incertitude des variables d’entrée. Une méthodologie de classification opérationnelle des données d’entrée a été réalisée. Elle s’appuie sur deux critères : la sensibilité du modèle au paramètre d’entrée et l’incertitude sur sa valeur (imprécision ou variabilité). Sur cette base, une nouvelle manière d’utiliser ces variables a été proposée. Enfin, différentes façons de restituer de manière opérationnelle les résultats des modélisations ont été proposées. / During accidents involving chemicals, experts can be asked to assess the effects generated. These experts provide distance effects using computer modeling and are faced with a major difficulty: little (or no) information available in order to assess the situation.The objective of this thesis is to suggest a methodology able to take into account uncertainties in the input data for the modeling carried out in emergency situations and to return explicitly these uncertainties to the manager of the crisis.A first step was to evaluate, for a given situation generating a toxic or flammable cloud, the dispersion of modeling results. A ranking of the input variables according to their influence on the final result was established. This phase was carried out on the basis of a sensitivity analysis with a specifically developed strategy.A second phase aimed to establish a methodology for estimating distance effects (in crisis situations), which takes into account the level of uncertainty in the input variables. A methodology for the classification of input operational data was carried out. This methodology is based on two criteria: the sensitivity of the model to the input parameter and the uncertainty about its value (imprecision or variability). On this basis, a new way of using these variables was suggested. Finally, several methods aimed to restore explicitly the results of this modeling were suggested.

[pt] Um dos grandes desafios da computação gráfica é conseguir representar ambientes virtuais que assemelham-se aos ambientes reais observados pelos seres humanos em seu dia-a-dia. A fim de alcançar essa representação, diversos estudos foram realizados na área de renderização fotorealística e, principalmente, na que diz respeito à modelagem física e representação de fenômenos naturais. Esse estudo propõe explicar e analisar técnicas que representam o fenômeno de dispersão atmosférica; responsável por definir a cor da atmosfera/céu. Para alcançar esse objetivo, serão apresentadas duas técnicas: uma técnica baseada em um modelo físico e outra baseada em um modelo analítico e de aproximações. Esse estudo apresenta os detalhes de cada uma das técnicas e apresenta um comparativo para auxiliar na escolha da técnica de acordo com a necessidade e os requisitos da aplicação que irá representar o fenômeno de dispersão atmosférica. / [en] One of the greatest challenges of computer graphics is to represent virtual environments that resemble real environments observed by humans in their day-to-day. In order to achieve this representation, several studies have been conducted in the area of photorealistic rendering, and especially in regards to physical modelling and representation of natural phenomena. This study aims to explain and analyze techniques that represent the phenomenon of atmospheric scattering; responsible for setting the color of atmosphere/sky. To achieve this goal, two techniques are presented: one based on a physical model and the other based on an analytical model and approximations. This study presents the details of each technique and a comparative to help in choosing the technique according to the need and requirements of the application that will represent the phenomenon of atmospheric scattering.