Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-10-11T22:53:53Z No. of bitstreams: 1 JoseUelitonPinheiro_TESE.pdf: 3704869 bytes, checksum: 22878ed4bfe495c030dc94c7fe144d8b (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-10-18T00:13:18Z (GMT) No. of bitstreams: 1 JoseUelitonPinheiro_TESE.pdf: 3704869 bytes, checksum: 22878ed4bfe495c030dc94c7fe144d8b (MD5) / Made available in DSpace on 2016-10-18T00:13:18Z (GMT). No. of bitstreams: 1 JoseUelitonPinheiro_TESE.pdf: 3704869 bytes, checksum: 22878ed4bfe495c030dc94c7fe144d8b (MD5) Previous issue date: 2015-03-18 / Neste trabalho foi estudado a sa?da de modelos de mudan?as clim?ticas do IPCC/AR5/CMIP5 que melhor expressam a atua??o dos V?rtices Cicl?nicos em Altos N?veis (VCANs) no Nordeste Brasileiro (NEB), no clima atual e cen?rios futuros, como tamb?m sua influ?ncia na precipita??o. Para alcan?ar os objetivos propostos na Tese foram feitos quatro ESTUDOS EXPERIMENTAIS. O primeiro, denominado de Piloto, onde estudou-se 13 modelos de mudan?as clim?ticas para avaliar e selecionar o modelo que melhor expressava a atua??o dos VCANs no NEB. O segundo onde, uma vez selecionado o melhor modelo, o MIROC4h, avaliou a efici?ncia desse modelo comparando com dados de rean?lise para um per?odo de 31 anos (1975-2005). O terceiro onde foram analisados os cen?rios clim?ticos futuros do MIROC4h para o per?odo de 21 anos (2015-2035). E o quarto onde foi analisado a contribui??o dos VCANs na precipita??o sobre o NEB atrav?s dos dados de rean?lise do NCEP/NCAR/DOE. Foram utilizadas duas abordagens estat?sticas comparativas nos Estudos Experimentais 1 e 2, a primeira utilizando o N?mero de Dias de Atua??o de VCANs (N.D.A.VCANs) mensais e a segunda usando o comparativo da atua??o de VCANs di?rio, com a utiliza??o dos ?ndices estat?sticos: correla??es de Pearson, Kendall e Spearman, Raiz quadrada do erro quadr?tico m?dio (RMSE), Raiz quadrada do erro quadr?tico m?dio normalizada (NRMSE), Nash-Sutcliffe (NSE), Kling-Gupta (KGE), ?ndice de Concord?ncia de Willmott (d), ?ndice de Propor??o Correta (PC), ?ndice de Sucesso Cr?tico (ISC), Probabilidade de Detec??o (POD), Taxa de alarme Falso (TAF) e Taxa de Tend?ncia (VI?S). E nos Experimentos 3 e 4 foram calculados desvios e m?dias. Os resultados mostraram a viabilidade na representa??o dos VCANs nos modelos de mudan?as clim?ticas do CMIP5, seja para o clima atual como nos cen?rios futuros. Com rela??o a contribui??o dos VCANs para a precipita??o do NEB estes apresentam percentuais que variam de 47,88% (LNE) a 49,89%(NNE) para o per?odo de outubro a mar?o. Sendo que Cear? (49,89%), Piau? (49,49%) e Maranh?o (47,88%) s?o os Estados onde os VCANs induzem mais precipita??o e Alagoas (41,93%) e Sergipe (38,03%) os Estados onde os VCANs induzem menos precipita??o. A proje??o de cen?rio futuro para os VCANs revelaram um desvio negativo 8,97% na ocorr?ncia deste fen?meno no NEB e ?reas adjacentes para o per?odo de 2015 a 2035. O que poder? impactar em -4,08% a precipita??o do NEB neste per?odo. / In this work, was analysed the climate change models's output from IPCC/AR5/CMIP5 that best express the performance of Upper Tropospheric Cyclonic Vortices (UTCV) in Northeast Brazil (NEB), its influence on rainfall and the possibility of projection of future climate scenarios. To achieve the proposed objectives in the thesis were made four EXPERIMENTAL STUDIES.The first, called Pilot, which were evaluated 13 models of climate change and selected the model that best expressed the actions of UTCV in the NEB. The second that evaluated the efficiency of the best model (MIROC4h), comparing it with reanalysis data for a period of 31 years (1975-2005). The third where the future climate scenarios from MIROC4h were analyzed for the period of 21 years (2015-2035). And the fourth where it was analyzed the contribution of UTCV in precipitation over the NEB through the reanalysis data from NCEP/NCAR/DOE. Two approaches comparative statistics were used in the Experimental Studies 1 and 2 with the use of statistical indices: Pearson , Kendall and Spearman's correlations, mean square error of the square root (RMSE), square of the normalized root mean square error (NRMSE), Nash-Sutcliffe (NSE), Kling-Gupta (KGE), Willmott Index (d) Proportion Correct Index (PCI), Critical Success Index (CSI), Probability of Detection (POD), False Alarm Rate (TAF ) and trend rate (BIAS). And in Experiments 3 and 4 were calculated deviations and averages. The results show that the representation of UTCV in climate change models of CMIP5 is feasible, either for the past climate and for the projection of future scenarios. Regarding the contribution of UTCV to precipitate the NEB these present percentages ranging from 47.88% (LNE) to 49.89% (NNE) for the period from October to March. Cear? (49.89%), Piau? (49.49%) and Maranh?o (47.88%) are the states where UTCV induce more rainfall and Alagoas (41.93%) and Sergipe (38.03%) are the states where UTCV induce less precipitation. The projection of future scenario for UTCV revealed a 8.97% negative difference in the occurrence of this phenomenon in the NEB and surrounding areas for the period 2015 to 2035. What may impact -4.08% the precipitation of the NEB.