Wind energy has been steadily growing in Brazil in recent years. To boost its growth, it is crucial to consider climate change, as wind energy generation is highly influenced by the weather. Therefore, it is essential to incorporate external climatic variables into the modeling of wind series, helping to reduce uncertainties. Periodic Autoregressive Models with Exogenous Variables (PARX) represent a viable approach to achieve this, including the ENSO exogenous variable. In the present study, wind speed series were modeled in the states of Rio Grande do Norte, Paraíba, Pernambuco, Alagoas, Sergipe, Rio Grande do Sul, and Santa Catarina. In this context, the covariance between these states in each Brazilian region was considered to assess the spatial correlation among them, creating the PARX-Cov modeling. Furthermore, the correlation between ENSO phenomenon indicators was also considered to enable out-of-sample forecasting of climatic variables, used for simulating wind speed scenarios. When comparing the PARX and PARX-Cov modeling with the current model in the Brazilian electric sector, the proposed models showed superior performance in simulating future wind speed series. The PARX-Cov model with the Accumulated ONI index is most suitable for Pernambuco, Rio Grande do Sul, and Santa Catarina. The PARX-Cov model with the SOI index is more appropriate for Rio Grande do Norte. For Alagoas and Sergipe, the PARX model with the Accumulated ONI index is the most recommended, while the PARX model with Accumulated Niño 4 is better for Paraíba.