This research explores the modeling and simulation of groundwater dynamics applying the univariate and multivariate time series forecasting. The univariate analysis focuses on groundwater levels, whereas the multivariate analysis integrates related variable quantity such as rainfall, temperature, root and surface soil witness, and depth to groundwater level. The incorporation of progressive computational systems such as deep learning and machine learning offers significant enhancements in analytical exactness and model robustness compared to traditional numerical approaches. Key results of this study comprise the expansion of projecting models that can be used to estimate groundwater levels based on the existing and historic data of related variable quantity. The developed models can support policymakers and stakeholders in making informed results concerning groundwater usage and maintenance.