This study explores the application of Long Short-Term Memory and AutoRegressive Integrated Moving Average models for forecasting climate variables, particularly temperature and the Standardized Precipitation Index. The research focuses on Vidarbha, Maharashtra, covering 11 districts, including Nagpur, Wardha, Bhandara, and Gondia. The region is known for its biodiversity, forests, and agricultural significance. Data collection includes daily temperature, precipitation, and Standardized Precipitation Index values. Preprocessing ensures data integrity by handling missing values, removing duplicates, and addressing outliers. The Standardized Precipitation Index calculation follows statistical transformations for standardized drought assessment. The statistical model captures linear trends, while the deep learning model models complex temporal dependencies. The integration of these models enhances climate forecasting, supporting agricultural decision-making, water management, and disaster preparedness. This study highlights the importance of hybrid statistical and deep learning approaches for accurate climate prediction, contributing to advanced climate science and real-world monitoring applications.