Weather forecasting is essential for various human activities. Recent data-driven models have outperformed numerical weather prediction in forecasting performance by utilizing deep learning. However, challenges remain in efficiently handling multiple meteorological variables. This study proposes a new variable aggregation scheme and an efficient learning framework for that challenge. Experiments show that VarteX outperforms conventional model in forecast performance while requiring significantly fewer parameters and resources. The effectiveness of learning through multiple aggregations and regional split training is demonstrated, enabling more efficient and accurate deep learning-based weather forecasting.