A pesquisa estudou a saída de modelos de mudanças climáticas que melhor expressam a atuação dos Vórtices Ciclônicos em Altos Níveis (VCANs) no Nordeste Brasileiro (NEB). Os VCANs foram quantificados pela sua ocorrência diária durante 5 anos (1995-1999), no período de outubro a março. O objeto de estudo foram 13 modelos do CMIP5/IPCC/AR5 (Coupled Model Intercomparison Project Phase 5/Intergovernmental Panel on Climate Change/Fifth Assessment Report), comparados com os resultados do NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research), por meio de métodos estatísticos para escolha do modelo que melhor indica a presença dos VCANs no NEB. A primeira análise comparativa foi feita através das correlações de Pearson, Kendall e Spearman, Raiz quadrada do erro quadrático médio, Raiz quadrada do erro quadrático médio normalizada e os índices de Eficiência e desempenho, Nash-Sutcliffe (NSE), Kling-Gupta (KGE) e o Índice de Concordância de Willmott (d). Em seguida foram selecionados os modelos de melhor desempenho e com significância estatística para uma análise posterior de acertos e erros através dos índices: Índice de Proporção Correta (PC), Índice de Sucesso Crítico (ISC), Probabilidade de Detecção (POD), Taxa de alarme Falso (TAF) e Taxa de Tendência (VIÉS). Para os testes estatísticos aplicados na primeira avaliação realizada o modelo MIROC4h foi o que apresentou os melhores índices seguido pelo MIROC-ESM e inmCM4, respectivamente. Além destes, ainda apresentaram correlação estatística significante o MPI-ESM-LR,o MRI-CGCM3 e o CSIRO-MK3-6-0. A segunda análise também apresentou o MIROC4h com os melhores valores de PC, ISC e POD, excetuando-se o VIÉS que apresentou o segundo melhor resultado e o TAF com o pior resultado em relação aos outros 5 modelos. Dessa forma o MIROC4h apresentou-se como o mais indicado entre os modelos do CMIP5 para estudos de cenários presentes e futuros de VCANs no NEB. A B S T R A C T The research studied the output of climate change models that best express the actions of Upper Tropospheric Cyclonic Vortices (UTCV) in high levels in the Northeast Brazil (NEB). The UTCV were quantified by a daily occurrence for 5 years (1995-1999) in the period from October to March. The object of the study were 13 models from CMIP5/IPCC/AR5 (Coupled Model Intercomparison Project Phase 5 / Intergovernmental Panel on Climate Change / Fifth Assessment Report ), compared with results from the NCEP / NCAR (National Centers for Environmental Prediction / National Center for Atmospheric Research) by means of statistical methods for choosing the model which best indicates the presence of UTCV in the NEB. The first comparative analysis was performed using the Pearson, Spearman and Kendall correlations, mean square error, normalized mean square error and efficiency and performance indices, Nash-Sutcliff (NSE), Kling-Gupta (KGE) and Index of Agreement of the Willmott (d). Then models with better performance and statistical significance for further analysis of successes and mistakes through the indices were selected: Index Proportion Correct (PC), Critical Success Index (CSI), Probability of Detection (POD), False Alarm Rate (FAR) and Trend Rate (BIAS). For the statistical analyzes used in the first test performed MIROC4h model showed the best rates followed by MIROC-ESM and inmCM4 respectively. In addition, further significant statistical correlation MPI-ESM-LR, MRI-CGCM3 and CSIRO-MK3-6-0. The second analysis also showed the MIROC4h with the best values of PC, CSI and POD, except the BIAS that had the second best result and the FAR with the worst result in relation to the other five models considered in this phase. Thus the MIROC4h introduced himself as the most suitable model of the CMIP5 for studies of the present and future scenarios of UTCV in the NEB
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.