The tropical cyclone is a majestic, yet violent atmospheric weather system occurring over tropical waters. Their majesty evolves from the significant range of spatial scales they operate over: from the mesoscale, to the larger synoptic-scale. Their associated violent winds and seas, however, are often the cause of damage and destruction for settlements in their path. Between 10/11/07 and 16/11/07, tropical cyclone Sidr formed and intensified into a category 5 hurricane over the southeast tropical waters of the northern Indian Ocean. Sidr tracked west, then north, during the course of its life, and eventually made landfall on 15/11/07, as a category 4 cyclone near the settlement of Barguna, Bangladesh. The storm affected approximately 2.7 million people in Bangladesh, and of that number 4234 were killed. In this study, the dynamics of tropical cyclone Sidr are simulated using version 2.2.1 of Advanced Weather Research and Forecasting — a non-hydrostatic, two-way interactive, triply-nested-grid mesoscale model. Three experiments were developed examining model sensitivity to ocean-atmosphere interaction; initialisation time; and choice of convective parameterisation scheme. All experiments were verified against analysed synoptic data. The ocean-atmosphere experiment involved one simulation of a cold sea surface temperature, fixed at 10 °C; and simulated using a 15 km grid resolution. The initialisation experiment involved three simulations of different model start time: 108-, 72-, and 48-hours before landfall respectively. These were simulated using a 15 km grid resolution. The convective experiment consisted of four simulations, with three of these using a different implicit convective scheme. The three schemes used were, the Kain-Fritsch, Betts-Miller-Janjic, and Grell-Devenyi ensemble. The fourth case simulated convection explicitly. A nested domain of 5km grid spacing was used in the convective experiment, for high resolution modelling. In all experiments, the Eta-Ferrier microphysics scheme, and the Mellor-Yamada-Janjic planetary boundary layer scheme were used. As verified against available observations, the model showed considerable sensitivity in each of the experiments. The model was found to be well suited for combining ocean-atmosphere interactions: a cool sea surface caused cyclone Sidr to dissipate within 24 hours. The initialisation simulations indicated moderate model sensitivity to initialisation time: variations were found for both cyclone track and intensity. Of the three simulations, an initialisation time 108 hours prior to landfall, was found to most accurately represent cyclone Sidr’s track and intensity. Finally, the convective simulations showed that considerable differences were found in cyclone track, intensity, and structure, when using different convective schemes. The Kain-Fritsch scheme produced the most accurate cyclone track and structure, but the rainfall rate was spurious on the sub-grid-scale. The Betts-Miller-Janjic scheme resolved realistic rainfall on both domains, but cyclone intensity was poor. Of particular significance, was that explicit convection produced a similar result to the Grell-Devenyi ensemble for both model domain resolutions. Overall, the results suggest that the modelled cyclone is highly sensitive to changes in initial conditions. In particular, in the context of other studies, it appears that the combination of convective scheme, microphysics scheme, and boundary layer scheme, are most significant for accurate track and intensity prediction.

Over the 1951-2009 time period, 47% of all tropical systems in the Atlantic Basin transitioned to post-tropical storms. These storms are capable of producing hurricaneforce winds, torrential, flooding rains and storm surge that floods coastal areas. This study adds to previous climatological work by completing a case-study of Hurricane Ike (2008) and examining how teleconnections such as the El Niño Southern Oscillation (ENSO), the Madden-Julian Oscillation (MJO), the Atlantic Multidecadal Oscillation (AMO) and the Pacific Decadal Oscillation (PDO) contribute to the strength of a transitioning post-tropical storm. T-tests performed show strong statistical relationships between an increase (decrease) in post-tropical storm frequency and warm PDO – La Niña (cold PDO – La Niña), cold PDO – ENSO neutral (warm PDO – ENSO neutral), and warm (cold) AMO conditions. Moreover, nearly significant results were found for the same increase (decrease) and La Niña seasons since (pre) 1980 and for cold (warm) PDO conditions. Modeling the MJO suggests that increased (decreased) relative humidity associated with the wet (dry) phase could increase (decrease) precipitation output from the storm and decrease (increase) forward speed of the storm, decreasing (increasing) wind speeds observed at the surface.

The tropical cyclone is a majestic, yet violent atmospheric weather system occurring over tropical waters. Their majesty evolves from the significant range of spatial scales they operate over: from the mesoscale, to the larger synoptic-scale. Their associated violent winds and seas, however, are often the cause of damage and destruction for settlements in their path. Between 10/11/07 and 16/11/07, tropical cyclone Sidr formed and intensified into a category 5 hurricane over the southeast tropical waters of the northern Indian Ocean. Sidr tracked west, then north, during the course of its life, and eventually made landfall on 15/11/07, as a category 4 cyclone near the settlement of Barguna, Bangladesh. The storm affected approximately 2.7 million people in Bangladesh, and of that number 4234 were killed. In this study, the dynamics of tropical cyclone Sidr are simulated using version 2.2.1 of Advanced Weather Research and Forecasting — a non-hydrostatic, two-way interactive, triply-nested-grid mesoscale model. Three experiments were developed examining model sensitivity to ocean-atmosphere interaction; initialisation time; and choice of convective parameterisation scheme. All experiments were verified against analysed synoptic data. The ocean-atmosphere experiment involved one simulation of a cold sea surface temperature, fixed at 10 °C; and simulated using a 15 km grid resolution. The initialisation experiment involved three simulations of different model start time: 108-, 72-, and 48-hours before landfall respectively. These were simulated using a 15 km grid resolution. The convective experiment consisted of four simulations, with three of these using a different implicit convective scheme. The three schemes used were, the Kain-Fritsch, Betts-Miller-Janjic, and Grell-Devenyi ensemble. The fourth case simulated convection explicitly. A nested domain of 5km grid spacing was used in the convective experiment, for high resolution modelling. In all experiments, the Eta-Ferrier microphysics scheme, and the Mellor-Yamada-Janjic planetary boundary layer scheme were used. As verified against available observations, the model showed considerable sensitivity in each of the experiments. The model was found to be well suited for combining ocean-atmosphere interactions: a cool sea surface caused cyclone Sidr to dissipate within 24 hours. The initialisation simulations indicated moderate model sensitivity to initialisation time: variations were found for both cyclone track and intensity. Of the three simulations, an initialisation time 108 hours prior to landfall, was found to most accurately represent cyclone Sidr’s track and intensity. Finally, the convective simulations showed that considerable differences were found in cyclone track, intensity, and structure, when using different convective schemes. The Kain-Fritsch scheme produced the most accurate cyclone track and structure, but the rainfall rate was spurious on the sub-grid-scale. The Betts-Miller-Janjic scheme resolved realistic rainfall on both domains, but cyclone intensity was poor. Of particular significance, was that explicit convection produced a similar result to the Grell-Devenyi ensemble for both model domain resolutions. Overall, the results suggest that the modelled cyclone is highly sensitive to changes in initial conditions. In particular, in the context of other studies, it appears that the combination of convective scheme, microphysics scheme, and boundary layer scheme, are most significant for accurate track and intensity prediction.

Este estudo tem como objetivo a verificação das previsões diárias, das temperaturas máxima e mínima e precipitação acumulada, realizadas pelo modelo operacional de previsão numérica do tempo WRF (Weather Research Forecasting) para o estado de São Paulo. As condições iniciais e de fronteira fornecidas pela análise e previsão das 00UTC do modelo Global Forecast System (GFS), são usados no processamento do WRF, para previsões de 72 horas, em duas grades aninhadas (espaçamentos horizontais de grade de 50 km, D1, e 16,6 km, D2). O período avaliado foi de abril de 2010 a março de 2011. As comparações diárias das temperaturas máxima e mínima foram realizadas entre os valores preditos e observados nas estações de superfície de Registro, São Paulo, Paranapanema, Campinas, Presidente Prudente e Votuporanga (dados da CIIAGRO); através do erro médio (EM) e raiz do erro médio quadrático (REQM), para os prognósticos das 36, 60 e 72 horas. A precipitação acumulada diária é avaliada com relação ao produto MERGE, pela aplicação da ferramenta MODE, na previsão das 36 horas, para um limiar de 0,3 mm, no domínio espacial abrangendo o Estado de São Paulo e vizinhanças. Primeiramente, fez-se uma análise, comparando os pares de grade dos campos previsto e observado, utilizando os índices estatísticos de verificação tradicional de probabilidade de acerto (PA); índice crítico de sucesso (ICS); viés (VIÉS); probabilidade de detecção (PD) e razão de falso alarme (RFA). Posteriormente, foram analisados os campos de precipitação com relação à razão de área (RA); distância dos centroides (DC); razões de percentil 50 (RP50) e 90 (RP90). Os resultados evidenciaram que as saídas numéricas do modelo WRF com D2 tiveram desempenho melhor comparado à grade de menor resolução (maior espaçamento de grade horizontal, D1), tanto no prognóstico diário das temperaturas (máxima e mínima) quanto da precipitação acumulada. A temperatura apresentou um padrão de amortecimento, com temperaturas diárias máxima subestimada e mínima superestimada. Com relação à precipitação, as saídas numéricas do modelo GFS e WRF com D2 mostraram desempenho semelhante, com o D2 apresentando índices ligeiramente melhores, enquanto que as saídas numéricas do modelo WRF com D1 exibiram pior desempenho. Verificou-se um padrão de superestimativa, tanto em termos de abrangência espacial quanto em intensidade, para o modelo GFS e WRF em ambos os domínios simulados, ao longo de todo o período analisado. O percentil 50 é, geralmente, maior que o observado; entretanto, o percentil 90 é mais próximo ao observado. Os resultados também indicam que o viés dos modelos varia ao longo do ano analisado. Os melhores índices tanto com relação à precipitação quanto à temperatura foram obtidos para a estação de verão, com o modelo WRF com D2 apresentando melhores prognósticos. Entretanto, os modelos apresentam os maiores erros no inverno e no outono. Estes erros foram decorrentes de subestimativas das temperaturas máximas e superestimativas de área e intensidade de precipitação. / Forecasts of daily maximum and minimum temperatures and rainfall performed by the operational numerical weather prediction WRF (Weather Research Forecasting) model in the São Paulo are evaluated. Initial and boundary conditions provided by the 00UTC Global Forecast System (GFS) Model and WRF run for 72 hours, with two nested grids (with horizontal grid spacing of 50 km, D1, and 16.6 km, D2). The study was made for April 2010 to March 2011 period. Daily maximum and minimum temperatures comparisons were made, between predicted and observed data of the surface weather stations of Registro, São Paulo, Paranapanema, Campinas, Presidente Prudente and Votuporanga (CIIAGRO Data), through the mean error (ME) and root mean square error(RMSE), for the 36, 60 and 72 hours forecasts. The daily accumulated rainfall is evaluated using MODE with respect to the MERGE product, for the 36 hours forecast, with threshold of 0.3 mm over the spatial domain covering the State of São Paulo and neighborhoods. First, an analysis was made comparing grid pairs of predicted and observed fields, through the traditional statistical verification indexes: accuracy (PA), critical success index (ICS), bias (VIES), probability of detection (PD) and false alarm ratio (RFA). Subsequently, we analyzed the precipitation field with respect to area ratio (AR), distance from the centroids (DC), ratio of the 50th percentile (RP50) and ratio of the 90th percentile (RP90). The WRF, with D2 nested grid, had better performance compared to the grid of lower space resolution (higher horizontal grid spacing, D1) for both, daily temperatures (maximum and minimum) and the accumulated rainfall forecasts. The temperature forecast presented a damped pattern, with underestimated maximum and overestimated minimum values. Rainfall was overall overestimated spatially and in intensity for the three models throughout the analized period. The forecasted 50th percentile is generally higher than that observed, however, the 90th percentile is closer to observations. The results also indicate that the bias of the models varies annually. The best performances for both rainfall and temperature were obtained for the summer season, with the D2 showing slightly better results. However, the models had the biggest errors during the winter and autumn seasons. These errors were due to underestimation of maximum temperatures and overestimation in area and intensity of precipitation.

Methane measurements over lake Tämnaren show a pronounced diurnal variation with high values at night and low values during daytime. The atmosphere over the lake and its surroundings is simulated with two different settings and resolutions of the WRF model during a period of eight days in May 2011 to investigate if a lake/land breeze circulation could be the cause of the observed methane variation. A night time land breeze can give rise to convergence over Tämnaren of the natural methane emissions from the lake which possibly could explain the diurnal variation. Analysis show that although Tämnaren is large enough to initiate a fully closed circulation these events are likely going to be rare because of the strong dependence of the background wind speed and cannot therefore be the cause of the pronounced diurnal variation. A fairly moderate wind speed will dominate over the thermodynamical forcing necessary to create a lake breeze. Even so, it is possible that a closed or nearly closed circulation could enhance the diurnal pattern with an increase of methane concentration at night and a decrease during the day. The reason for the high night time methane concentration is more likely due to the accumulation in a shallow internal boundary layer that develops over the lake combined with high night time methane flux caused by waterside convection. / Mätningar av metankoncentrationen över Tämnaren visar en tydlig dygnsvariation med höga värden på natten och låga under dagtid. Atmosfären över sjön med omgivning modelleras med två olika inställningar och upplösningar av WRF modellen under en åttadagarsperiod i Maj 2011 för att undersöka om en sjö- och landbriscirkulation kan vara orsaken till den observerade metanvariationen. På natten kan en landbris ge upphov till konvergens över Tämnaren av de naturliga metanutsläppen vilket skulle kunna vara en möjlig förklaring till dygnsvariationen. Vidare analys visar att Tämnaren är tillräckligt stor för att initiera en sluten cirkulation men dessa händelser är troligtvis sällsynta på grund av det starka inflytandet av bakgrundsvinden och kan därför inte vara orsaken till den uttalade metanvariationen. En relativt måttlig vind kommer dominera över den termodynamiska effekt som är drivande för skapandet av sjö- och landbris. Trots detta är det möjligt att en sluten eller nästan sluten cirkulation kan förstärka metanhaltens dygnsvariation med en ökning på natten och minskning under dagen. Orsaken till den observerade höga metankoncentrationen på natten är troligare en ackumulering i ett grunt internt ytskikt som bildas över Tämnaren kombinerat med höga nattliga metanflöden till följd av konvektion i sjön.

Este estudo tem como objetivo a verificação das previsões diárias, das temperaturas máxima e mínima e precipitação acumulada, realizadas pelo modelo operacional de previsão numérica do tempo WRF (Weather Research Forecasting) para o estado de São Paulo. As condições iniciais e de fronteira fornecidas pela análise e previsão das 00UTC do modelo Global Forecast System (GFS), são usados no processamento do WRF, para previsões de 72 horas, em duas grades aninhadas (espaçamentos horizontais de grade de 50 km, D1, e 16,6 km, D2). O período avaliado foi de abril de 2010 a março de 2011. As comparações diárias das temperaturas máxima e mínima foram realizadas entre os valores preditos e observados nas estações de superfície de Registro, São Paulo, Paranapanema, Campinas, Presidente Prudente e Votuporanga (dados da CIIAGRO); através do erro médio (EM) e raiz do erro médio quadrático (REQM), para os prognósticos das 36, 60 e 72 horas. A precipitação acumulada diária é avaliada com relação ao produto MERGE, pela aplicação da ferramenta MODE, na previsão das 36 horas, para um limiar de 0,3 mm, no domínio espacial abrangendo o Estado de São Paulo e vizinhanças. Primeiramente, fez-se uma análise, comparando os pares de grade dos campos previsto e observado, utilizando os índices estatísticos de verificação tradicional de probabilidade de acerto (PA); índice crítico de sucesso (ICS); viés (VIÉS); probabilidade de detecção (PD) e razão de falso alarme (RFA). Posteriormente, foram analisados os campos de precipitação com relação à razão de área (RA); distância dos centroides (DC); razões de percentil 50 (RP50) e 90 (RP90). Os resultados evidenciaram que as saídas numéricas do modelo WRF com D2 tiveram desempenho melhor comparado à grade de menor resolução (maior espaçamento de grade horizontal, D1), tanto no prognóstico diário das temperaturas (máxima e mínima) quanto da precipitação acumulada. A temperatura apresentou um padrão de amortecimento, com temperaturas diárias máxima subestimada e mínima superestimada. Com relação à precipitação, as saídas numéricas do modelo GFS e WRF com D2 mostraram desempenho semelhante, com o D2 apresentando índices ligeiramente melhores, enquanto que as saídas numéricas do modelo WRF com D1 exibiram pior desempenho. Verificou-se um padrão de superestimativa, tanto em termos de abrangência espacial quanto em intensidade, para o modelo GFS e WRF em ambos os domínios simulados, ao longo de todo o período analisado. O percentil 50 é, geralmente, maior que o observado; entretanto, o percentil 90 é mais próximo ao observado. Os resultados também indicam que o viés dos modelos varia ao longo do ano analisado. Os melhores índices tanto com relação à precipitação quanto à temperatura foram obtidos para a estação de verão, com o modelo WRF com D2 apresentando melhores prognósticos. Entretanto, os modelos apresentam os maiores erros no inverno e no outono. Estes erros foram decorrentes de subestimativas das temperaturas máximas e superestimativas de área e intensidade de precipitação. / Forecasts of daily maximum and minimum temperatures and rainfall performed by the operational numerical weather prediction WRF (Weather Research Forecasting) model in the São Paulo are evaluated. Initial and boundary conditions provided by the 00UTC Global Forecast System (GFS) Model and WRF run for 72 hours, with two nested grids (with horizontal grid spacing of 50 km, D1, and 16.6 km, D2). The study was made for April 2010 to March 2011 period. Daily maximum and minimum temperatures comparisons were made, between predicted and observed data of the surface weather stations of Registro, São Paulo, Paranapanema, Campinas, Presidente Prudente and Votuporanga (CIIAGRO Data), through the mean error (ME) and root mean square error(RMSE), for the 36, 60 and 72 hours forecasts. The daily accumulated rainfall is evaluated using MODE with respect to the MERGE product, for the 36 hours forecast, with threshold of 0.3 mm over the spatial domain covering the State of São Paulo and neighborhoods. First, an analysis was made comparing grid pairs of predicted and observed fields, through the traditional statistical verification indexes: accuracy (PA), critical success index (ICS), bias (VIES), probability of detection (PD) and false alarm ratio (RFA). Subsequently, we analyzed the precipitation field with respect to area ratio (AR), distance from the centroids (DC), ratio of the 50th percentile (RP50) and ratio of the 90th percentile (RP90). The WRF, with D2 nested grid, had better performance compared to the grid of lower space resolution (higher horizontal grid spacing, D1) for both, daily temperatures (maximum and minimum) and the accumulated rainfall forecasts. The temperature forecast presented a damped pattern, with underestimated maximum and overestimated minimum values. Rainfall was overall overestimated spatially and in intensity for the three models throughout the analized period. The forecasted 50th percentile is generally higher than that observed, however, the 90th percentile is closer to observations. The results also indicate that the bias of the models varies annually. The best performances for both rainfall and temperature were obtained for the summer season, with the D2 showing slightly better results. However, the models had the biggest errors during the winter and autumn seasons. These errors were due to underestimation of maximum temperatures and overestimation in area and intensity of precipitation.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The present study aims to evaluate the quality of nocturnal temperature forecast made by a mesoscale numerical model and to understand the reasons behind the difficulties found. To do that, the Weather Research and Forecast (WRF) model is used, with the same configuration employed for operational weather forecast. The model has been ran for the 31 nights of July 2012, and temperature outputs have been compared to hourly observations measured by 26 weather stations scattered over the entire state of Rio Grande do Sul. Four different schemes for turbulence have been considered. Three of them, Yonsei University (YSU), Mellor Yamada Janjic (MYJ) and Bougeault-Lacarrere (BOU) are formulations available from WRF code, while the fourth, Bougeault-Lacarrere Modified (BOU-Mod) is a change made to BOU, aiming at making it less turbulent. A general analysis shows that the different formulations present similar root mean squared errors (EQM), with YSU showing slightly smaller errors than the others. An important discrepancy found refers to the fact that there is an appreciable difference between station real altitude and its altitude in the model, which is given by the height of the closest grid point. When such an altitude difference is corrected by a potential temperature, the errors are enhanced. In this case, the most turbulent formulations, YSU and BOU, tend to overestimate nocturnal temperatures, while the least turbulent ones, MYJ and BOU-Mod, tend to underestimate it. All schemes presented a tendency to underestimate the observed temporal variability. It means that they tend to overestimate the coldest observations and to underestimate the warmest ones. In the most stable nights, all parameterizations showed large EQM and overestimate the temperature. In the least stable nights there were some cases with reduced EQM, but all formulations tended to underestimate temperature, showing that it is necessary to increase the turbulent mixing in this cases. When the different stations are compared, it becomes evident that the height difference between station and model altitudes has a large influence in the nighttime temperature weather forecast. It happens mainly because stations lower than the nearest grid point the modeled winds tend to be larger than observed, causing more intense turbulent mixing and leading to warmer temperatures. The opposite happens in stations higher than the grid point. Such a situation occurs mainly in the more stable conditions, when the lower regions tend to have its surface decoupling from the higher atmospheric levels. The implications of these results and suggestions for improving nocturnal temperature forecasts are presented. / O objetivo do presente trabalho é avaliar a qualidade da previsão de temperaturas noturnas em um modelo numérico de mesoescala e compreender a razão das principais dificuldades encontradas. Para tanto, é utilizado o modelo Weather Research and Forecast (WRF), configurado da mesma maneira que é feito em previsões do tempo operacionais. O modelo foi rodado para as 31 noites do mês de Julho de 2012, e as saídas de temperatura foram comparadas com observações horárias feitas em 26 estações espalhadas por todo o estado do Rio Grande do Sul. Foram consideradas quatro representações diferentes para a turbulência no modelo, sendo que três delas, as de Yonsei University (YSU), Mellor-Yamada-Janjic (MYJ) e Bougeault-Lacarrere (BOU), são formulações disponibilizadas no próprio modelo e uma quarta, Bougeault-Lacarrere-Modificada (BOU-Mod), é uma alteração imposta à parametrização BOU com o propósito de torná-la menos turbulenta. Uma análise geral mostra que as formulações apresentam erros quadráticos médios (EQM) bastante próximos entre si, sendo que YSU tem erros levemente menores que as demais. Uma importante discrepância observada é que há diferença razoável entre a altura real da estação e a sua altitude no modelo, representada pelo valor do ponto de grade mais próximo. Quando essa diferença de altitude é corrigida pela temperatura potencial, os erros aumentam. Nesse caso, as duas parametrizações mais turbulentas, que são YSU e BOU, têm tendência geral de superestimar as temperaturas noturnas, enquanto as menos turbulentas, MYJ e BOU-Mod, tendem a subestimar essa grandeza. Todos os esquemas mostraram tendência de reduzir a variabilidade temporal observada, o que significa que elas tendem a superestimar as observações mais frias e subestimar as mais quentes. Nas noites mais estáveis, todas parametrizações apresentaram grande EQM e superestimam a temperatura. Nas noites menos estáveis houve casos com EQM reduzido, mas todas as parametrizações mostraram tendência de subestimar a temperatura observada, mostrando que é necessário que todas as formulações se tornem mais turbulentas nestes casos. Quando as diferentes estações são comparadas, a diferença de altura entre a estação e o ponto de grade mais próximo tem grande influência na previsão de temperatura noturna. Isso ocorre porque nas estações mais baixas que o ponto de grade, o vento do modelo tende a ser maior que o observado, causando maior mistura turbulenta, e levando a temperaturas maiores. O oposto ocorre nas estações mais altas que o ponto de grade. Esta situação ocorre principalmente nas noites mais estáveis, quando estações localizadas em regiões mais baixas tendem a experimentar o fenômeno do desacoplamento entre a superfície e os níveis mais altos da atmosfera. As implicações destes resultados são discutidas e propostas para melhorar as previsões noturnas de temperatura são apresentadas.

With the growing energy demand and the opportunity for exploitation of renewable energy (wind energy), this paper performs the mapping of wind potential in Alagoas. This study is based on the use of anemometer measurements between August/2007 to July/2008, as also computational tools: mesoscale atmospheric model (WRF); numerical model of microscale (WAsP®); data from Geographic Information Systems (GIS). The stations with anemometers were geographically distributed in three mesoregions of the State: Backwoods Sertão (Água Branca), Rural Agreste (Girau do Ponciano and Palmeira dos Índios) and Coast Litoral (Feliz Deserto, Maragogi and Roteiro). WRF simulations have validated by time series (daily average, monthly average and diurnal cycles), statistical analysis and Weibull distributions. It was also evaluated using WAsP as a tool for calculating the spatial wind speed in microscale. The input data in WAsP, in addition to the WRF model simulations and the preparation of various maps, were all treated and developed in the ArcGIS® software. The results showed greater efficiency in the WRF simulation of wind speed for sites to the interior of Alagoas, with bias less than 1 m.s-1. These estimates were more accurate during the summer season (bias less than 0.5 m.s-1) and more disparate in winter, with average error of up to 3 m.s-1. For the Coast, these values were overestimated by more than 3 m.s-1, except in Maragogi. The wind direction from WRF was calculated with accuracy in all sites, being dominant in the east coast, east to northeast in Rural, and southeast in Água Branca. The annual wind map for WRF for Alagoas showed areas with good potential energy, the magnitude of the wind varies between 7-9 m.s-1 at 50 meters high. Application of the WAsP in the microscale showed good resemblance to the maps obtained in the Wind Atlas of the State of Alagoas - AEEA. The models achieved satisfactory results, but still need to be improved in some respects. / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Com a crescente demanda energética e a oportunidade de exploração de energia renovável (eólica), este trabalho realiza o mapeamento do potencial eólico em Alagoas. Este trabalho é baseado no uso de medições anemométricas entre agosto/2007 a julho/2008, além de ferramentas computacionais: modelo atmosférico de mesoescala (WRF); modelo numérico de microescala (WAsP®); dados de sistemas de informações geográficas (SIG). As estações anemométricas foram distribuídas geograficamente nas três mesorregiões do Estado: Sertão (Água Branca), Agreste (Girau do Ponciano e Palmeira dos Índios) e Litoral (Feliz Deserto, Maragogi e Roteiro). As simulações do WRF foram validadas através de séries temporais (médias diárias, mensais, ciclos diurnos), análises estatísticas e Distribuições de Weibull. Avaliaram-se ainda o uso do WAsP como ferramenta para cálculo espacial da velocidade do vento em microescala. Os dados de entrada no WAsP, além das simulações do modelo WRF e da elaboração de diversos mapas, foram todos tratados e desenvolvidos no software ArcGIS®. Os resultados mostraram que houve maior eficiência do WRF nas simulações da velocidade do vento para sítios do interior alagoano, com bias inferior a 1 m.s-1. Estas estimativas foram mais precisas durante a estação do verão (bias menor que 0,5 m.s-1) e mais discrepantes no inverno, com erro médio de até 3 m.s-1. Para o litoral, estes valores foram superestimados em mais de 3 m.s-1, exceto em Maragogi. A direção do vento do WRF foi calculada com acerto em todos os sítios, sendo dominante de leste no litoral, leste a nordeste no agreste, e sudeste em Água Branca. O mapa eólico anual do WRF para Alagoas apresentou áreas com bom potencial energético, cuja magnitude do vento varia entre 7 a 9 m.s-1 em 50 metros de altura. A aplicação do WAsP na microescala mostrou boa semelhança aos mapas obtidos no Atlas Eólico do Estado de Alagoas AEEA. Os modelos atingiram resultados satisfatórios, porém ainda precisam ser aperfeiçoados em alguns aspectos.

L'intégration de la production des énergies renouvelables intermittentes dans le mix énergétique est aujourd'hui limitée à un seuil de 30 % de la puissance totale produite. Cette mesure vise à assurer la sécurité de l'alimentation électrique des réseaux insulaires en France. La levée de ce verrou technique ne pourra se faire qu'en apportant des solutions au caractère intermittent des sources d'énergies éolienne et photovoltaïque. Les difficultés énergétiques auxquelles sont confrontés aujourd'hui les milieux insulaires préfigurent celles que rencontreront la planète à plus ou moins long terme. Ces territoires sont des laboratoires uniques pour éprouver les nouvelles technologies de stockage, de gestion et de prévision de l'énergie. La contribution de ce travail de thèse se focalise sur la prévision du rayonnement solaire global à différents horizons de temps car la puissance photovoltaïque produite découle directement de l'intensité du rayonnement solaire global. Dans un premier temps, l'étude bibliographique a permis de classer les modèles de prévision numériques et les modèles de prévision statistiques en fonction de la résolution spatiale et temporelle. Par ailleurs, elle montre que les meilleurs performances sont obtenues avec les modèles hybrides. Dans un deuxième temps, un modèle de prévisions à court terme (J+1) est proposé avec le modèle Weather Research and Forecasting (WRF) et un réseau de neurone bayésien. L'hybridation de ces deux méthodes améliore les performances de prévisions à J+1. Dans un troisième temps, un modèle de prévision à très court terme (t+h) est proposé avec le modèle hybride de Kalman. Cette méthode produit d'une part une prévision énergétique et d'autre part une prévision multi-horizon. La comparaison de la performance de ces modèles avec la méthode de référence dite de persistance montre une amélioration de la qualité de la prévision. Enfin, la combinaison du filtre de Kalman avec le modèle numérique WRF permet une mise en œuvre opérationnelle de la prévision. / The integration of intermittent renewable energy in the energy mix is currently limited to a threshold of 30% of the total power being produced. This restriction aims at ensuring the safety of the power input. The elimination of this technical obstacle will be possible with solutions to energy intermittence of wind and solar energy. The energy issues which islands are facing today prefigure global problems in a more or less long term. These territories constitute unique laboratories for testing new technologies of storage, management and forecasting of energy. The contribution of this thesis focuses on the forecasting of global horizontal irradiance at different time horizons. Indeed, the generated PV power stems directly from the intensity of the global horizontal irradiance. First, the review of solar irradiance forecasting methods allows to classify numerical weather models and statistical forecasting methods depending on spatial and temporal resolution. Moreover, it shows that best performance is obtained with hybrid models. Second, a short-term forecast model (day ahead forecast) is developed with the Weather Research and Forecasting model (WRF) and a Bayesian neural network. The hybridization of these methods improves the day ahead forecast performance. Third, a model for forecasting the very short term is developped with the Kalman hybrid model. This method offers on the one hand an energy forecasting and on the other hand a multi-horizon forecast. Comparing the performance of the aforesaid with the reference method, namely the persistence method, shows an improvement of the quality of the forecasts. Combining the Kalman filter with the WRF numerical model allows an operational implementation of the forecast.

Made available in DSpace on 2014-08-20T14:25:49Z (GMT). No. of bitstreams: 1 dissertacao_clovis_correa.pdf: 19288418 bytes, checksum: 98a82fd638a44d647e5b27b37e7a8288 (MD5) Previous issue date: 2010-02-26 / This work presents the study of atmospheric conditions associated the occurrence of Hurricane Catarina and two study cases. The Catarina occurred in March 2004 and reached the coast of southern Brazil.The other case studies correspond to cut-off lows that occurred in May 2005 and September 2008. It was used the WRF model, version 3.0, to obtain the meteorological fields during the period of occurrence of these events and images of the geostationary satellite GOES-12 and QuikSCAT. For Hurricane Catarina, data from surface meteorological stations were also analyzed. Initially, the study cases consisted of cut-off lows preceded by extratropical cyclones in the South Atlantic Ocean. When they shut the western flow, they occurred at the interface of different patterns of vertical variation of horizontal wind in the layer between 850-200hPa and showed closed cyclonic circulation at middle levels. At high levels, it was observed bifurcation of the jet with a trough quasi-stationnary near the southern coast of Brazil. The observed atmospheric blocking together with the thermodynamic and kinematic configurations of each case contributed to the occurrence of tropical transition and subsequent formation of Hurricane Catarina, in March 2004, while the other cut-off lows dissipated. / Neste trabalho é feito o estudo sobre as condições atmosféricas associadas à ocorrência do furacão Catarina e a dois casos de estudo. O Catarina ocorreu em março de 2004 e atingiu o litoral da região sul do Brasil. Os outros casos de estudo correspondem a baixas desprendidas que ocorreram em maio de 2005 e setembro de 2008. Foi utilizado o modelo WRF, versão 3.0, para a obtenção dos campos meteorológicos durante o período de ocorrência destes eventos e imagens do satélite geoestacionário GOES-12 e do QuikSCAT. Para o furacão Catarina, foram também analisados dados de estações meteorológicas de superfície. Os casos de estudo, inicialmente, corresponderam a baixas desprendidas precedidas por ciclones extratropicais no Oceano Atlântico Sul. Ao se isolarem do escoamento de oeste, ocorreram na interface de padrões distintos de variação vertical do vento horizontal na camada entre 850-200hPa e apresentaram circulação ciclônica fechada em níveis médios. Em altos níveis, foi observada bifurcação do jato com um cavado quase estacionário nas proximidades da costa sul do Brasil. Os bloqueios atmosféricos observados, juntamente com as configurações cinemáticas e termodinâmicas de cada caso, contribuíram para a ocorrência da transição tropical e posterior formação do furacão Catarina, em março de 2004, enquanto que as outras baixas desprendidas se dissiparam.