Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Brésil → APLICAÇÃO E AVALIAÇÃO DE METODOLOGIAS PARA PREVISÃO DE ÍNDICES REGIONAIS DE CHUVA NO PERÍODO DA QUADRA CHUVOSA NA REGIÃO SEMI-ÁRIDA DO NORDESTE DO BRASIL

Universidade Federal do Ceará (2007)

APLICAÇÃO E AVALIAÇÃO DE METODOLOGIAS PARA PREVISÃO DE ÍNDICES REGIONAIS DE CHUVA NO PERÍODO DA QUADRA CHUVOSA NA REGIÃO SEMI-ÁRIDA DO NORDESTE DO BRASIL

Silva Emerson Mariano da

Titre : APLICAÇÃO E AVALIAÇÃO DE METODOLOGIAS PARA PREVISÃO DE ÍNDICES REGIONAIS DE CHUVA NO PERÍODO DA QUADRA CHUVOSA NA REGIÃO SEMI-ÁRIDA DO NORDESTE DO BRASIL

Auteur : Silva Emerson Mariano da

Université de soutenance : Universidade Federal do Ceará

Grade : Doutor em Engenharia Civil 2007

Résumé
This study presents the application and evaluation of two distinct forecast systems Seasonal Regional Rainfall Indices on the Semi-Arid Northeast of Brazil and Basins of the Cearà State Regions. In the first case a set of numerical simulations (1971-2000) obtained from system forecast based on the theory of dynamic downscaling is used as data to input in the Regional Spectral Model (RSM97) atmospherics variables simulated by ECHAM4.5 global model. The second case is application of fuzzy sets theory as a tool for the period of 1985-1996. Thermodynamic parameters in Tropical Atlantic and Pacific Oceans were used as input variables for the model. The results from the application of each methodology had been evaluated for the climatological simulation and for composites of climatic years of contrast, categorized as dry, normal and rainy years, using statistical performance indices and correlation coefficients among the series simulated and observed on the regions. The results show that the use of the dynamic downscaling technique presents satisfactory performance, presenting ability in the task of simulation of the climatologic indices, capturing the trends series of the observed data, although to observe underestimated of the values observed in all regions analyzed. For the model application based on the theory of fuzzy sets, the results have shown that qualitatively, in an annual scale, this technique was able to forecast at least one of the categories of the output variable (total February/May rain in the of Semi-Arid Northeast Region of Brazil). Quantitatively, the smallest errors have been observed for the years classified according to the output variable in the categories of Normal, Rainy, and Very Rainy, with correlation coefficients ranging from 0.8 to 0.85. For regions of Basins in the State of CearÃ, the results showed that indexes simulated presents values close to that observed in the years ranked in the categories of Normal and Very Rainy, presenting larger differences for the years of Very Dry and Dry, with index correction varying between 0.66 and 0.84. Thus, concludes that the applications techniques can assist the forecast of the Seasonal Regional Rainfall Indices in the regions of the Basins in the State of CearÃ, and supply information of climatic forecasts that can assist the studies of quantification of the water resources and the agricultural sector of these regions.

Mots clés  : Modelagem ClimÃtica ; precipitaÃÃo pluviomÃtrica ; Bacias hidrogrÃficas ; Forecast. Seasonal Regional Rainfall. ; Recursos hÃdricos

Présentation (RCAAP)

Version intégrale

Page publiée le 6 décembre 2015, mise à jour le 10 juillet 2017