No description available.

In this dissertation, we investigate the n=0 Geodesic Acoustic Modes (GAM) in the framework of both magneto-hydrodynamics and kinetics. In MHD, the purpose is to understand the numerical results out of the CASTOR code (1). Effects of energetic particle are ignored. The leading perturbation is the density perturbation, which leads to a local GAM. The coupling of density perturbation to the magnetic perturbation, which is treated to be smaller, leads to global a GAM. We recover the numerical results from the CASTOR code and obtain and analytical solution to the radial eigen-mode equation though asymptotic matching. To understand recent experimental results on DIII-D (2) a kinetic theory is constructed in which magnetic perturbations are neglected and energetic ions are treated on the same footing as the thermal species based on drift kinetics. Not only do the energetic particles destabilize the local GAM induced by thermal species, but they are also crucial to establish the global GAM due to their large orbit shifts. Polarization of thermal ions is included. A mechanism for fast GAM excitation through NBI is proposed, based on our local kinetic GAM theory when there exists a loss boundary in pitch angle. / text

Quasi-periodic disturbances have been observed in the outer solar atmosphere for many years. Although first interpreted as upflows (Schrijver et al. (1999)), they have been widely regarded as slow magneto-acoustic waves, due to their observed velocities and periods. Here we conduct a detailed analysis of the velocities of these disturbances across several wavelengths using the Atmospheric Imaging Assembly (AIA) on board the Solar Dynamics Observatory (SDO). We analysed 41 examples, including both sunspot and non-sunspot regions of the Sun. We found that the velocities of propagating disturbances (PDs) located at sunspots are more likely to be temperature dependent, whereas the velocities of PDs at non-sunspot locations do not show a clear temperature dependence. This suggests an interpretation in terms of slow magneto-acoustic waves in sunspots but the nature of PDs in non-sunspot (plage) regions remains unclear. Finally, we found that removing the contribution due to the cooler ions in the 193 wavelength suggests that a substantial part of the 193 emission of sunspot PDs can be attributed to the cool component of 193. Phase mixing is a well known and studied phenomenon in the solar corona, to enhance the dissipation of Alfvén waves (Heyvaerts and Priest (1982)). In this study we run numerical simulations of a continuously driven Alfvén wave in a low beta plasma along a uniform magnetic field. We model phase mixing by introducing a density inhomogeneity. Thermal conduction is then added into the model in the form of Braginskii thermal conduction. This acts to transport heat along the magnetic field. A parameter study will be carried out to investigate how changing the density structure and other parameters changes the results. We go on to consider the effect of wave reflection on phase mixing. We found that wave reflection has no effect on the damping of Alfvén waves but increases the heat in the system. We also consider a more realistic experiment where we drive both boundaries and study how the loop is heated in this situation. We also study what effect changing the frequency of one of the drivers so there is a small difference between the frequencies (10%) and a large difference (50%). We find the general behaviour is similar, but the heat is tilted. We have investigated basic phase mixing model which incorporates the mass exchange between the corona and the chromosphere. Chromospheric evaporation is approximated by using a non dimensional version of the RTV (Rosner et al. (1978)) scaling laws, relating heating (by phase mixing of Alfvén waves), density and temperature. By combining this scaling law with our numerical MHD model for phase mixing of Alfvén waves, we investigate the modification of the density profile through the mass up flow. We find a rapid modification of the density profile, leading to drifting of the heating layers. We also find that similar results are own seen in the propagating Alfvén wave case when we incorporate the effects of reflection.

O ciclo solar é um dos fenômenos magnéticos mais interessantes do Universo. Embora ele tinha sido descoberto há mais de 150 anos, até agora permanece um problema em aberto para a Astrofísica. Há diferentes tipos de observações que sugerem que o ciclo solar corresponde a um processo de dínamo operando em algum lugar do interior solar. Parker foi o primeiro a tentar explicar o dínamo solar como um processo hidro-magnético acerca de 50 anos atrás. Desde então, embora tenha havido avanços significativos nas observações e investigações teóricas e numéricas, uma resposta definitiva para o dínamo solar ainda não existe. Acredita-se que no caso do Sol, pelo menos dois processos são necessários para completar o ciclo magnético observado: a transformação de um campo poloidal inicial em um campo toroidal, um processo conhecido como efeito , o qual se deve ao cisalhamento em grande escala ocasionado pela rotação diferencial; e a transformação do campo toroidal em um novo campo poloidal de polaridade oposta ao inicial. Esse segundo processo é menos conhecido e motivo de intensas discussões e pesquisas. Duas hipóteses principais foram formuladas para explicar a natureza deste processo, usualmente conhecido como efeito : a primeira, baseada na idéia de Parker de um mecanismo turbulento onde os campos poloidais resultam de movimentos convectivos ciclônicos operando em tubos de fluxo toroidais em pequena escala. Esses modelos se depararam, no entanto, com um serio inconveniente: na fase não-linear, i.e., quando a reação dinâmica do campo magnético ao fluido torna-se importante, o efeito pode ser amortecido de forma catastrófica, levando a um dínamo pouco efetivo. A segunda hipótese é baseada nas idéias de Babcock (1961) e Leighton (1969) (BL), que propuseram que o campo poloidal forma-se devido à emergência e decaimento posterior das regiões bipolares ativas. Neste modelo a circulação meridional tem um papel fundamental pois atua como mecanismo de transporte do fluxo magnético, de tal forma que a escala de tempo advectivo deve dominar sobre a escala de tempo difusiva. Por essa razão essa classe de modelos é comumente conhecida como modelo de dínamo dominado pelo transporte de fluxo, ou dínamo advectivo. Os modelos de dínamo dominados pelo transporte de fluxo são relativamente bem sucedidos em reproduzir as características em grande escala do ciclo solar, tornando-se populares entre a comunidade de Física solar, no entanto, também apresentam vários problemas amplamente discutidos na literatura. O objetivo principal deste trabalho foi identificar as principais limitações dessa classe de modelos e explorar as suas possíveis soluções. Para tal, construímos um modelo numérico bi-dimensional de dínamo cinemático baseado na teoria de campo médio e investigamos primeiro os efeitos da geometria e da espessura da tacoclina solar na amplificação do dínamo. Depois, consideramos o processo de bombeamento magnético turbulento como um mecanismo alternativo de transporte de fluxo magnético, e finalmente, incluímos a reação dinâmica do campo magnético sobre a difusividade magnética turbulenta, um processo conhecido como amortecimento de . Verificamos que é possível construir-se um modelo de dínamo dominado pelo transporte de fluxo capaz de reproduzir as observações ao considerar-se uma tacoclina de espessura fina localizada abaixo da zona convectiva. Isto limita a criação de intensos campos toroidais não desejados nas altas latitudes. Verificamos também ser importante considerar o bombeamento magnético, pois ele provê advecção do fluxo magnético para o equador e para a base da camada convectiva, o que resulta em uma correta distribuição latitudinal e temporal dos campos toroidais e também permite certa penetração desses campos nas regiões mais estáveis onde podem adquirir maior amplificação. Esse mecanismo é ainda importante para produzir a paridade correta do campo (anti-simétrica) nos dois hemisférios do Sol. Também encontramos que o amortecimento da difusividade magnética é um mecanismo fundamental para a formação de pequenas estruturas de campo toroidal com maior tempo de vida, identificadas com os tubos de fluxo, que acredita-se existirem na base da zona de convecção. Além do mais, os campos magnéticos formados graças ao amortecimento de podem ser até ~2 vezes mais intensos que as estruturas magnéticas formadas sem o seu amortecimento. Por fim, nos últimos anos, alguns trabalhos teóricos vêm chamando a atenção para o papel da conservação da helicidade magnética no processo de dínamo, dando nova vida a modelos de dínamo turbulento, como originalmente proposto por Parker. Com o objetivo de investigar o papel da helicidade magnética e de buscar uma descrição dinâmica mais realista do mecanismo de dínamo, construímos recentemente um modelo numérico de convecção tridimensional (utilizando o código MHD, PLUTO) que tenta reproduzir o cenário natural do interior solar onde teria lugar o processo de dínamo. Exploramos a evolução de um campo magnético semente imposto sobre um estado convectivo estacionário. Os resultados preliminares indicam que a convecção pode facilmente excitar o efeito de dínamo, inclusive em casos sem rotação. Porém, nos casos com rotação, o dínamo parece produzir uma maior quantidade de campo magnético médio com relação aos casos sem a rotação nos quais o campo flutuante é dominante. Estes resultados suportam a existência de um dínamo turbulento y validam a teoria de campo médio, mas uma a análise mais detalhada ainda é necessária. / The solar cycle is one of the most interesting magnetic phenomenon in the Universe. Even though it was discovered more than 150 years ago, it remains until now as an open problem in Astrophysics. There are several observational evidences that suggest that the solar cycle corresponds to a dynamo process operating at some place of the solar interior. Parker, in 1955, was the first to try to explain the solar dynamo as hydromagnetic phenomena. Since then, although there has been important improvements in the observations, theory and numerical simulations, a definitive model for the solar dynamo is still missing. There is common agreement that in the solar case, at least two processes are necessary to close the dynamo loop: the transformation of an initial poloidal field into a toroidal field, the so called Omega effect, which is due to a large scale shear caused by the diferential rotation, and the transformation of the toroidal field into a new poloidal field of opposite polarity, which is still a poorly understood process that has been the subject of intense debate and research. Two main hypotheses have been formulated in order to explain the nature of this effect, usually denominated alpha effect: the first one is based on Parker\'s idea of a turbulent mechanism where the poloidal field results from cyclonic convective motions operating at small scales in the toroidal field ropes. These models, however face an important limitation: in the non-linear regime, i.e. when the back reaction of the toroidal field on the motions becomes important, the alpha effect can be catastrophically quenched leading to an ineffective dynamo. The second hypotheses is based on the formulation of Babcock (1961) and Leighton (1969) (BL), who proposed that the poloidal field is formed due to the emergence and decay of bipolar magnetic regions. In this model the meridional circulation plays an important role by acting as conveyor belt of the magnetic flux, so that the advection time must be dominant over the diffusion time. For this reason these models are often called flux-transport dynamo models. The flux-transport dynamo models has been relatively successful in reproducing the large scale features of the solar cycle, and have become popular between the solar community. However, they also present several problems that have been widely discussed in the literature. The main goal of this work was to identify the main problems concerning the flux-transport dynamo model and to explore possible solutions for them. For this aim, we have built a two-dimensional kinematic numerical model based on the mean-field theory in order to explore first the effects of the geometry and thickness of the solar tachocline on the dynamo amplification. Then, we considered the turbulent pumping as an alternative magnetic flux advection mechanism, and finally, we included the non-linear back-reaction of the magnetic field on the turbulent magnetic diffusivity, a process known as eta-quenching. We have found that it is possible to build a flux-transport dynamo model able to reproduce the observations as long as a thin tachocline located below the convective zone is considered. This helps to prevent the amplification of undesirable strong toroidal fields at the high latitudes. We have also found that it is important to consider the turbulent magnetic pumping mechanism, because it provides magnetic field advection both equatorward and inwards, that results in a correct latitudinal and temporal distribution of the toroidal field and also allows the penetration of the toroidal fields down into the stable layers where they can acquire further amplification. Besides, this mechanism plays an important role in reproducing the correct field parity (anti-symmetric) on both solar hemispheres. We have also found that the eta-quenching may lead to the formation of long-lived small structures of toroidal field which resemble the flux-tubes that are believed to exist at the base of the convection zone. The magnetic fields that are formed thanks to the eta-quenching can be up to ~ twice as larger as the magnetic structures which are developed without this effect. Finally, a number of theoretical works in the last years have called the attention to the role of magnetic helicity conservation in the dynamo processes, giving a new life to the turbulent dynamo model as proposed by Parker. With the aim to study the role of magnetic helicity and explore a more realistic dynamical description of the dynamo mechanism, we have also recently built a 3D convective numerical model (based on the MHD-Goudunov type PLUTO code) where we try to reproduce the natural scenario of the solar interior where the dynamo might take place. We have studied the evolution of a seed field embedded in an initially steady state convection layer. Our preliminary results indicate that convection can easily drive the dynamo action, even in the case without rotation. However, in the rotating cases, the dynamo appears to produce a larger amount of large scale (coherent) magnetic field when compared to the case without rotation where small scale fluctuating fields are dominant. These results support the existence of a turbulent mean field dynamo, but furthermore detailed analysis is still required.

O objetivo deste trabalho foi a análise do movimento do fluido dentro de um segmento de vórtice em anel gerado através da ação magnetohidrodinâmica e o monitoramento do seu movimento resultante, de modo a oferecer condições de visualização usando velocimetria a laser com técnica de processamento de imagens de partículas envolvidas na circulação. O objetivo final desta linha de pesquisa é que através de análises de dados como as velocidades envolvidas na circulação obtida, possa modelar corpos em movimento em fluidos, de forma que o arrasto devido ao seu deslocamento possa ser reduzido. O movimento circulatório a que foi submetido o fluido sem meios mecânicos, ou seja, por meio de uma força resultante da ação de campos conjugados, teve a função de criar campos de baixa pressão tendo, para isso, acelerado o fluido em um circuito fechado restrito pela atuação da força e das paredes da célula de ensaio. O movimento do fluido, com a leitura da variação de velocidades em pontos diferentes do circuito teve o auxílio de um equipamento ótico que forneceu todas as variações, mesmo as mínimas, mostrando que tal aparato pode ser desenvolvido para a obtenção de melhores resultados. / The aim of this work involves the analysis of fluid flow, using sea water stimulated by localised magnetohydrodynamic induction within an experimental glass sided receptacle. The above mentioned apparatus simulated one segment of a three dimensional apparatus capable of creating a flow pattern very similar to that of a vortex ring. Particle Image Velocimetry techniques using a copper vapour laser with computerised image and data acquisition system were used to analyse the flow in the experimental segment. The final aim of this line of research is to be able to eventually model the shape of bodies in movement within a fluid when combined with the use of a magnetohydrodynamically induced flow field in such a way as to significantly reduce fluid drag forces on the body. Circulatory movement was induced in the fluid within the confines of the experimental tank using a locally applied Lorentz field force between electrodes to accelerate the fluid. Both the electrical and magnetic fields were controlled independently using separate external rectified a.c. circuits. The aim of the fluid movement described above was to induce low static pressure over a specific surface area. The analysis of the fluid movement, including the measurement of velocities in different points of the circuit was attained using specialised optical equipment linked to a computerised data acquisition and analysis system. The results, which included exceptionally low velocity readings, showed thatsuch equipment possesses the potential for developed for the attainment of better results in the future.

O objetivo deste trabalho foi a análise do movimento do fluido dentro de um segmento de vórtice em anel gerado através da ação magnetohidrodinâmica e o monitoramento do seu movimento resultante, de modo a oferecer condições de visualização usando velocimetria a laser com técnica de processamento de imagens de partículas envolvidas na circulação. O objetivo final desta linha de pesquisa é que através de análises de dados como as velocidades envolvidas na circulação obtida, possa modelar corpos em movimento em fluidos, de forma que o arrasto devido ao seu deslocamento possa ser reduzido. O movimento circulatório a que foi submetido o fluido sem meios mecânicos, ou seja, por meio de uma força resultante da ação de campos conjugados, teve a função de criar campos de baixa pressão tendo, para isso, acelerado o fluido em um circuito fechado restrito pela atuação da força e das paredes da célula de ensaio. O movimento do fluido, com a leitura da variação de velocidades em pontos diferentes do circuito teve o auxílio de um equipamento ótico que forneceu todas as variações, mesmo as mínimas, mostrando que tal aparato pode ser desenvolvido para a obtenção de melhores resultados. / The aim of this work involves the analysis of fluid flow, using sea water stimulated by localised magnetohydrodynamic induction within an experimental glass sided receptacle. The above mentioned apparatus simulated one segment of a three dimensional apparatus capable of creating a flow pattern very similar to that of a vortex ring. Particle Image Velocimetry techniques using a copper vapour laser with computerised image and data acquisition system were used to analyse the flow in the experimental segment. The final aim of this line of research is to be able to eventually model the shape of bodies in movement within a fluid when combined with the use of a magnetohydrodynamically induced flow field in such a way as to significantly reduce fluid drag forces on the body. Circulatory movement was induced in the fluid within the confines of the experimental tank using a locally applied Lorentz field force between electrodes to accelerate the fluid. Both the electrical and magnetic fields were controlled independently using separate external rectified a.c. circuits. The aim of the fluid movement described above was to induce low static pressure over a specific surface area. The analysis of the fluid movement, including the measurement of velocities in different points of the circuit was attained using specialised optical equipment linked to a computerised data acquisition and analysis system. The results, which included exceptionally low velocity readings, showed thatsuch equipment possesses the potential for developed for the attainment of better results in the future.

Astrophysikalische Jets sind ausgedehnte, kollimierte Massenausflüsse von verschiedenen astronomischen Objekten. Zeitabhängige magnetohydrodynamische (MHD) Simulationen der Jet-Entwicklung müssen den Akrretionsprozess in der Scheibe berücksichtigen, da der Jet aus der Scheibenmaterie gespeist wird. Allerdings ist die simultane Berechnung der Entwicklung von Scheibe und Jet schwierig, da die charakteristischen Zeitskalen unterschiedlich sind. Selbstähnliche Modelle zeigten, daß eine Beschreibung der Jetentstehung aus einer Akkretionsscheibe durch rein magnetische Prozesse möglich ist. / In this thesis the magnetohydrodynamic jet formation and the effects of magnetic diffusion on the formation of axisymmetric protostellar jets have been investigated in three different simulation sets. The time-dependent numerical simulations have been performed, using the magnetohydrodynamic ZEUS-3D code.

L'origine de certains champs magnétiques stellaires observés et leur impact sur l'évolution des étoiles sont mal compris. C'est particulièrement vrai dans le cas des étoiles de masse intermédiaire de la séquence principale. Des relevés spectropolarimétriques récents ont en effet révélé l'existence d'une dichotomie magnétique inexpliquée, de 2 ordres de grandeur en terme de champ longitudinal, entre le fort champ des étoiles Ap/Bp et le faible champ des étoiles Vega-like. Le but de cette thèse est de tester la possibilité que cette dichotomie magnétique soit liée, comme proposé par Aurière el al. (2007), au développement d'instabilités magnétohydrodynamiques (MHD) dans la zone radiative des étoiles de masse intermédiaire. Pour cela, j'ai réalisé des simulations numériques MHD 2D et 3D qui permettent de suivre l'évolution d'un champ magnétique axisymétrique soumis initialement à une rotation différentielle dans une zone stratifiée de façon stable puis de considérer le développement d'instabilités MHD non-axisymétriques. L'influence de différents paramètres physiques des simulations, comme l'intensité initiale du champ magnétique poloïdal, le profil de rotation différentielle, la valeur des coefficients de diffusion ou encore l'importance de la stratification stable, a été testée. L'analyse des résultats des simulations montre que des instabilités MHD comme l'instabilité magnétorotationnelle et celle de Tayler peuvent se déclencher dans une zone radiative en rotation différentielle. En accord avec le scénario de Aurière et al. (2007), ces instabilités se développent assez pour modifier la structure spatiale à grande échelle d'un champ magnétique si l'intensité initiale du champ poloïdal est suffisamment faible par rapport à l'intensité initiale de la rotation différentielle. Le champ longitudinal calculé pour nos simulations les plus instables est diminué de 15% par rapport à un cas stable. Ce travail de thèse montre donc que les instabilités MHD sont des possibles candidats pour expliquer le désert magnétique des étoiles de masse intermédiaire de la séquence principale. / The origin of some of the observed stellar magnetic fields and their impact on stellar evolution are not well understood. This is particularly true for the main sequence intermediate-mass stars. Recent spectropolarimetric surveys have indeed exhibited an unexplained magnetic dichotomy, of 2 orders of magnitude in term of the longitudinal field, between the strong field of Ap/Bp stars and the weak field of Vega-like stars. This thesis aims to test the possibility that this magnetic dichotomy is linked to the development of magnetohydrodynamic (MHD) instabilities in the radiative zone of intermediate-mass stars, as proposed by Aurière et al. (2007). To do that, I have performed 2D and 3D MHD numerical simulations that allow to follow the evolution of an axisymetric magnetic field which is initially submitted to a differential rotation in a stably stratified zone and then to consider the development of non-axisymetric MHD instabilities. The influence of different physical parameters of the simulations, as the initial strength of the poloidal magnetic field, the differentially rotating profile, the diffusion coefficient values or the effect of the stable stratification, has been tested. The analysis of the simulation results show that MHD instabilities as the magneto-rotational instability or the Tayler instability can be triggered in a differentially rotating radiative zone. In agreement with the scenario of Aurière et al. (2007), these instabilities are enough developed to modify the large scale spatial structure of a magnetic field if the initial strength of the poloidal field is sufficiently weak with respect to the initial strength of the differentially rotation. The computed longitudinal field in our most unstable simulations is reduced by 15% with respect to a stable case. Therefore, this thesis work shows that the magnetic instabilities are possible candidates to explain the magnetic desert of the main sequence intermediate-mass stars.

O ciclo solar é um dos fenômenos magnéticos mais interessantes do Universo. Embora ele tinha sido descoberto há mais de 150 anos, até agora permanece um problema em aberto para a Astrofísica. Há diferentes tipos de observações que sugerem que o ciclo solar corresponde a um processo de dínamo operando em algum lugar do interior solar. Parker foi o primeiro a tentar explicar o dínamo solar como um processo hidro-magnético acerca de 50 anos atrás. Desde então, embora tenha havido avanços significativos nas observações e investigações teóricas e numéricas, uma resposta definitiva para o dínamo solar ainda não existe. Acredita-se que no caso do Sol, pelo menos dois processos são necessários para completar o ciclo magnético observado: a transformação de um campo poloidal inicial em um campo toroidal, um processo conhecido como efeito , o qual se deve ao cisalhamento em grande escala ocasionado pela rotação diferencial; e a transformação do campo toroidal em um novo campo poloidal de polaridade oposta ao inicial. Esse segundo processo é menos conhecido e motivo de intensas discussões e pesquisas. Duas hipóteses principais foram formuladas para explicar a natureza deste processo, usualmente conhecido como efeito : a primeira, baseada na idéia de Parker de um mecanismo turbulento onde os campos poloidais resultam de movimentos convectivos ciclônicos operando em tubos de fluxo toroidais em pequena escala. Esses modelos se depararam, no entanto, com um serio inconveniente: na fase não-linear, i.e., quando a reação dinâmica do campo magnético ao fluido torna-se importante, o efeito pode ser amortecido de forma catastrófica, levando a um dínamo pouco efetivo. A segunda hipótese é baseada nas idéias de Babcock (1961) e Leighton (1969) (BL), que propuseram que o campo poloidal forma-se devido à emergência e decaimento posterior das regiões bipolares ativas. Neste modelo a circulação meridional tem um papel fundamental pois atua como mecanismo de transporte do fluxo magnético, de tal forma que a escala de tempo advectivo deve dominar sobre a escala de tempo difusiva. Por essa razão essa classe de modelos é comumente conhecida como modelo de dínamo dominado pelo transporte de fluxo, ou dínamo advectivo. Os modelos de dínamo dominados pelo transporte de fluxo são relativamente bem sucedidos em reproduzir as características em grande escala do ciclo solar, tornando-se populares entre a comunidade de Física solar, no entanto, também apresentam vários problemas amplamente discutidos na literatura. O objetivo principal deste trabalho foi identificar as principais limitações dessa classe de modelos e explorar as suas possíveis soluções. Para tal, construímos um modelo numérico bi-dimensional de dínamo cinemático baseado na teoria de campo médio e investigamos primeiro os efeitos da geometria e da espessura da tacoclina solar na amplificação do dínamo. Depois, consideramos o processo de bombeamento magnético turbulento como um mecanismo alternativo de transporte de fluxo magnético, e finalmente, incluímos a reação dinâmica do campo magnético sobre a difusividade magnética turbulenta, um processo conhecido como amortecimento de . Verificamos que é possível construir-se um modelo de dínamo dominado pelo transporte de fluxo capaz de reproduzir as observações ao considerar-se uma tacoclina de espessura fina localizada abaixo da zona convectiva. Isto limita a criação de intensos campos toroidais não desejados nas altas latitudes. Verificamos também ser importante considerar o bombeamento magnético, pois ele provê advecção do fluxo magnético para o equador e para a base da camada convectiva, o que resulta em uma correta distribuição latitudinal e temporal dos campos toroidais e também permite certa penetração desses campos nas regiões mais estáveis onde podem adquirir maior amplificação. Esse mecanismo é ainda importante para produzir a paridade correta do campo (anti-simétrica) nos dois hemisférios do Sol. Também encontramos que o amortecimento da difusividade magnética é um mecanismo fundamental para a formação de pequenas estruturas de campo toroidal com maior tempo de vida, identificadas com os tubos de fluxo, que acredita-se existirem na base da zona de convecção. Além do mais, os campos magnéticos formados graças ao amortecimento de podem ser até ~2 vezes mais intensos que as estruturas magnéticas formadas sem o seu amortecimento. Por fim, nos últimos anos, alguns trabalhos teóricos vêm chamando a atenção para o papel da conservação da helicidade magnética no processo de dínamo, dando nova vida a modelos de dínamo turbulento, como originalmente proposto por Parker. Com o objetivo de investigar o papel da helicidade magnética e de buscar uma descrição dinâmica mais realista do mecanismo de dínamo, construímos recentemente um modelo numérico de convecção tridimensional (utilizando o código MHD, PLUTO) que tenta reproduzir o cenário natural do interior solar onde teria lugar o processo de dínamo. Exploramos a evolução de um campo magnético semente imposto sobre um estado convectivo estacionário. Os resultados preliminares indicam que a convecção pode facilmente excitar o efeito de dínamo, inclusive em casos sem rotação. Porém, nos casos com rotação, o dínamo parece produzir uma maior quantidade de campo magnético médio com relação aos casos sem a rotação nos quais o campo flutuante é dominante. Estes resultados suportam a existência de um dínamo turbulento y validam a teoria de campo médio, mas uma a análise mais detalhada ainda é necessária. / The solar cycle is one of the most interesting magnetic phenomenon in the Universe. Even though it was discovered more than 150 years ago, it remains until now as an open problem in Astrophysics. There are several observational evidences that suggest that the solar cycle corresponds to a dynamo process operating at some place of the solar interior. Parker, in 1955, was the first to try to explain the solar dynamo as hydromagnetic phenomena. Since then, although there has been important improvements in the observations, theory and numerical simulations, a definitive model for the solar dynamo is still missing. There is common agreement that in the solar case, at least two processes are necessary to close the dynamo loop: the transformation of an initial poloidal field into a toroidal field, the so called Omega effect, which is due to a large scale shear caused by the diferential rotation, and the transformation of the toroidal field into a new poloidal field of opposite polarity, which is still a poorly understood process that has been the subject of intense debate and research. Two main hypotheses have been formulated in order to explain the nature of this effect, usually denominated alpha effect: the first one is based on Parker\'s idea of a turbulent mechanism where the poloidal field results from cyclonic convective motions operating at small scales in the toroidal field ropes. These models, however face an important limitation: in the non-linear regime, i.e. when the back reaction of the toroidal field on the motions becomes important, the alpha effect can be catastrophically quenched leading to an ineffective dynamo. The second hypotheses is based on the formulation of Babcock (1961) and Leighton (1969) (BL), who proposed that the poloidal field is formed due to the emergence and decay of bipolar magnetic regions. In this model the meridional circulation plays an important role by acting as conveyor belt of the magnetic flux, so that the advection time must be dominant over the diffusion time. For this reason these models are often called flux-transport dynamo models. The flux-transport dynamo models has been relatively successful in reproducing the large scale features of the solar cycle, and have become popular between the solar community. However, they also present several problems that have been widely discussed in the literature. The main goal of this work was to identify the main problems concerning the flux-transport dynamo model and to explore possible solutions for them. For this aim, we have built a two-dimensional kinematic numerical model based on the mean-field theory in order to explore first the effects of the geometry and thickness of the solar tachocline on the dynamo amplification. Then, we considered the turbulent pumping as an alternative magnetic flux advection mechanism, and finally, we included the non-linear back-reaction of the magnetic field on the turbulent magnetic diffusivity, a process known as eta-quenching. We have found that it is possible to build a flux-transport dynamo model able to reproduce the observations as long as a thin tachocline located below the convective zone is considered. This helps to prevent the amplification of undesirable strong toroidal fields at the high latitudes. We have also found that it is important to consider the turbulent magnetic pumping mechanism, because it provides magnetic field advection both equatorward and inwards, that results in a correct latitudinal and temporal distribution of the toroidal field and also allows the penetration of the toroidal fields down into the stable layers where they can acquire further amplification. Besides, this mechanism plays an important role in reproducing the correct field parity (anti-symmetric) on both solar hemispheres. We have also found that the eta-quenching may lead to the formation of long-lived small structures of toroidal field which resemble the flux-tubes that are believed to exist at the base of the convection zone. The magnetic fields that are formed thanks to the eta-quenching can be up to ~ twice as larger as the magnetic structures which are developed without this effect. Finally, a number of theoretical works in the last years have called the attention to the role of magnetic helicity conservation in the dynamo processes, giving a new life to the turbulent dynamo model as proposed by Parker. With the aim to study the role of magnetic helicity and explore a more realistic dynamical description of the dynamo mechanism, we have also recently built a 3D convective numerical model (based on the MHD-Goudunov type PLUTO code) where we try to reproduce the natural scenario of the solar interior where the dynamo might take place. We have studied the evolution of a seed field embedded in an initially steady state convection layer. Our preliminary results indicate that convection can easily drive the dynamo action, even in the case without rotation. However, in the rotating cases, the dynamo appears to produce a larger amount of large scale (coherent) magnetic field when compared to the case without rotation where small scale fluctuating fields are dominant. These results support the existence of a turbulent mean field dynamo, but furthermore detailed analysis is still required.

Le but de cette thèse est l'étude du développement de la turbulence dans le vent solaire entre 0.2 et 1 unité astronomique (UA) du soleil (i.e. l'orbite terrestre). L'étude est faite en résolvant numériquement les équations de la MHD après soustraction de l'écoulement moyen radial. Les deux aspects de la turbulence qui nous intéressent sont la structure 3D des spectres d’énergie et le chauffage du plasma qui résulte de la dissipation turbulente des tourbillons et couches de courant emportés par le vent. On cherche à déterminer quelles sont les conditions du plasma près du soleil qui permettent d’aboutir à ce qu'on observe à 1 UA. Un but important de mon travail est aussi de déterminer si la physique qui est présente dans les équations que j'intègre (la MHD) suffit pour arriver à reproduire ce qu'on a déjà observé dans cet intervalle de distance. Nous introduisons le contexte de notre travail dans la première partie. On y trouve les équations de base, une introduction à la turbulence, un résumé sur la physique du vent solaire et de la couronne solaire. La partie 2 sera consacrée à l'étude de l'anisotropie de la cascade turbulente, et plus précisément du spectre 3D. Dans la zone inertielle, les mesures in-situ vers 1 UA montrent des figures complexes pour ces spectres qu'on peut interpréter de plusieurs façons : nos simulations numériques permettent de lever toute ambiguïté. Plus précisément, la question est de savoir quand intervient l'axe soleil-terre, et quand intervient l'axe du champ magnétique moyen. La partie trois est centrée sur le chauffage turbulent dans les vents rapides et lents. Entre 0.3 et 1 UA, la température des protons diminue anormalement lentement, ce qui indique une source de chauffage, qu'on suppose ici être la dissipation des tourbillons et couches de courant emportés par le vent. Pour démontrer que cette hypothèse est raisonnable, nous considérons d’abord le modèle de Burgers qui est un modèle pour l'évolution d’ondes sonores. Ensuite, nous passons à l'étude du cas plus complexe d'un volume de plasma 3D. Nous examinerons les conditions initiales correspondant aux vents lents et rapides. Dans les deux cas, on adoptera des anisotropies spectrales différentes. Dans la dernière partie, nous exposerons les conclusions de notre travail et proposerons d'introduire l'anisotropie de la température dans un travail futur. / The aim of this thesis is the study of the development of turbulence in the solar wind between 0.2 and 1 astronomical unit (AU) from the Sun (i.e. Earth’s orbit). The study is done by solving the magnetohydrodynamics equations (MHD) after subtracting the mean radial flow. The two aspects of turbulence that interest us are the 3D structure of the energy spectra and the heating of plasma that results from the turbulent dissipation of eddies and current layers transported by the wind. We want to determine which conditions of the plasma close to the Sun can result into what we observe at 1 AU. We have relatively detailed measurements of what happens between 0.3 and 1 AU. One important goal of this work is to determine if the physics present in the equations that are integrated (MHD) is sufficient to reproduce what is observed in this interval of distances. We introduce the context of our work in the first part. We give a summary of the physics concerning the solar wind and the solar corona, and the basic equations used to describe the solar wind plasma and an introduction to turbulence. Part 2 is dedicated to the study of anisotropy in the turbulent cascade, which characterizes 3D spectra. In the inertial range, in-situ measurements at 1 AU show complex figures for these spectra that we can interpret in several ways : numerical simulations allow to clear ambiguities. An important question is to know whether the Earth-Sun symmetry axis or the mean magnetic field axis is dominant.The third part focuses on turbulent heating in fast and slow winds. Between 0.3 and 1 AU, proton temperature decreases more slowly than expected, which requires a heating source. This source is supposed to be the continuous dissipation of eddies and current layers transported by the wind. To start with, we consider the simple case of Burgers equation, which is a one-dimensional model for shock formation. Thereupon, we switch to the 3-dimensional case, where we consider initial conditions appropriate for slow and fast winds. In the last part we expose our conclusions and propose the implementation of temperature anisotropy as future work.