Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-03-07T18:17:19Z No. of bitstreams: 1 LidiaGabrielaRodriguesDeSouza_DISSERT.pdf: 6421234 bytes, checksum: 89fa219b3a32406f43d2dd2fdeb0c1dd (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-03-07T18:34:41Z (GMT) No. of bitstreams: 1 LidiaGabrielaRodriguesDeSouza_DISSERT.pdf: 6421234 bytes, checksum: 89fa219b3a32406f43d2dd2fdeb0c1dd (MD5) / Made available in DSpace on 2017-03-07T18:34:41Z (GMT). No. of bitstreams: 1 LidiaGabrielaRodriguesDeSouza_DISSERT.pdf: 6421234 bytes, checksum: 89fa219b3a32406f43d2dd2fdeb0c1dd (MD5) Previous issue date: 2015-08-17 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) / O estudo sobre o clima vem crescendo constantemente no decorrer dos ?ltimos anos e a precipita??o ? uma das vari?veis em destaque para essa finalidade. Neste estudo levando em considera??o as dimens?es territoriais e as distin??es f?sico-clim?ticas existentes entre os estados do Nordeste Brasileiro (NEB), objetiva-se estudar as oscila??es da precipita??o no NEB, no per?odo de 1961 a 2010, tendo como refer?ncia as m?dias mensais das capitais da regi?o. Para tanto, inicialmente foi feita a descri??o dos principais sistemas meteorol?gicos geradores de chuva na regi?o e em seguida a an?lise dos dados com o aux?lio da estat?stica descritiva, bem como a divis?o dos nove estados em tr?s grupos homog?neos com o auxilio da an?lise de agrupamento (cluster), para assim poder observar as s?ries temporais. Partindo dessas quest?es, fez-se uso da Transformada Wavelet (TW), a qual se apresenta como ferramenta eficaz na obten??o da variabilidade peri?dica num dado recorte temporal dos elementos meteorol?gicos, visando detectar altera??es no padr?o da precipita??o. Os resultados s?o apresentados a partir das TW, a fim de visualizar as oscila??es da precipita??o, assim como a escala e o per?odo para as capitais do NEB, observando que ? poss?vel estudar a chuva da regi?o pela ferramenta. Nos resultados encontrados, pode-se observar que Macei?, Salvador e Aracaj? s?o cidades com padr?es de chuva semelhantes, assim como, Jo?o Pessoa, Natal e Recife, formando outro grupo e um terceiro grupo constitu?do por Fortaleza, Teresina e S?o Lu?s, com caracter?sticas semelhantes no padr?o de chuva, corroborando com trabalhos da literatura, que mostram grupos com alta similaridade. E tais resultados possuem import?ncia nos estudos clim?ticos, pois s?o obtidos por meio de um m?todo que se utiliza de representa??es de tempo e frequ?ncia de forma simult?nea e precisa, permitindo a an?lise de qualquer serie temporal, concluindo que ? poss?vel estudar a precipita??o de um recorte temporal por meio da TW. / The study about the climate change has been growing constantly over the past of the years, and the rainfall is one of the highlighted variables for this purpose. In this study the territorial dimension and existing physical-climatic distinctions between the states of Northeast Brazil (NEB) was taken into consideration. Indeed, the objective is to study the fluctuations in the rainfall of the NEB during the period of 1961-2010 taking as a reference the monthly average capital of the region. For this reason, initially the description of the meteorological rain generating systems in the region was made, and then the data was analyzed using descriptive statistics. Also, the nine states was divided in three homogeneous groups with the help of cluster analysis (cluster), so the time series could be observed. Based on these issues, the Wavelet Transform (WT) which presents itself as an effective tool in obtaining the periodic variability in a given time frame of meteorological elements was used to detect changes in the pattern of precipitation. The WT results are presented in order to verify the fluctuations of rainfall, the scale and the period for the capital of the NEB. Thus, it is possible to study the rainfall in the region using this tool. The WT results showed that Maceio, Salvador and Aracaju are cities with similar rainfall patterns as well as another group formed by Jo?o Pessoa, Natal and Recife. In addition, a third group consisting of Fortaleza, Teresina and S?o Luis have similar characteristics in rainfall patterns. In fact, these observations corroborate previous studies which show groups with high similarity. These results have importance in weather studies because they are obtained by a method that uses representations of time and frequency simultaneously and accurately. Furthermore, it allows the examination of any time series. Therefore, it is possible to study the precipitation of a time frame using WT.