El Niño and La Niña are critical phases of the El Niño-Southern Oscillation (ENSO), significantly impacting global climate patterns. This project aims to develop a machine learning-based predictive model and a deep learning forecasting model to understand and anticipate ENSO events. For the predictive model, Random Forest, Logistic Regression, and XGBoost classifiers are employed to classify climate conditions into El Niño or La Niña or Normal events using historical climate data. Additionally, Long Short-Term Memory (LSTM) and Recurrent Neural Network (RNN) models are used for time series forecasting of the Oceanic Niño Index (ONI) to predict future ENSO phases. The dataset used includes key oceanic and atmospheric variables, such as sea surface temperatures and pressure levels, obtained from historical ENSO records. The models are evaluated based on accuracy, precision, recall for classification, and RMSE and MAE for time series forecasting. The results demonstrate the effectiveness of machine learning and deep learning techniques in predicting and forecasting ENSO events, providing valuable insights for climate monitoring and forecasting.

Keywords: El Niño, La Niña, Oceanic Niño Index (ONI), El Niño-Southern Oscillation (ENSO).