<p>Accurate rainfall forecasting is critical for agriculture-dependent nations and essential for watershed management, particularly for flood warnings. Rainfall prediction is a challenging task due to high variability in patterns, data inefficiency, and lack of comprehensive meteorological records. These issues complicate effective water resource management, crop productivity, and infrastructure planning. This work introduces a novel Bird Sea Lion Optimization (BSLnO) algorithm for rainfall forecasting, combining the strengths of the Bird Swarm Algorithm (BSA) and Sea Lion Optimization (SLnO). In this work, a Bird Sea Lion Optimization (BSLnO) algorithm is proposed for rainfall forecasting. Initially, the time series data is taken as the input, and then technical indicators, such as Simple Moving Average (SMA), Time Series Forecasting (TSF), Exponential Moving Average (EMA), momentum (MOM), On Balance Volume (OBV), Weighted Moving Average (WMA), Relative Strength Index (RSI), and Average Directional Index (ADX) are extracted. The fusion of features is done by RideNN and hamming distance followed by Data Augmentation for handling data dimensionality. Then, rainfall prediction is done using Non-linear autoregressive with external input (NARX), which is trained using BSLnO. The BSLnO is the combination of the Bird Swarm algorithm (BSA) and the Sea Lion optimization algorithm (SLnO). The BSLnO-based NARX outperformed the existing methods based on performance metrics, such as Mean Absolute Error (MAE) of 0.081, Mean Squared Error (MSE) of 0.147, Root Mean Squared Error (RMSE) of 0.384, Nash–Sutcliffe Efficiency (NSE) of 0.128, and R² of 0.058 using Rainfall in India dataset.</p>