Accurate simulation of compound flooding during tropical cyclones (TCs) requires the integration of rainfall-induced runoff with coastal storm surge. While high-fidelity hydrodynamic frameworks like the Discontinuous Galerkin Shallow Water Equation Model (DG-SWEM) provide the numerical fidelity to resolve these interactions, their predictive utility remains limited by the quality of the prescribed rainfall forcing. Existing parametric models offer operational efficiency but often fail to consistently reproduce the full set of observed spatiotemporal characteristics, including peak intensity, total volume, and radial distribution, simultaneously. This creates a critical need for a standardized modeling framework capable of bridging these discrepancies across diverse storm scenarios. To address this, we propose a novel TC rainfall model derived from a 22-year spatiotemporal analysis of North Atlantic storm data from NASA's IMERG precipitation data set. This framework defines precipitation as a function of key storm characteristics such as intensity, size, and translation speed. We evaluate both advanced parameterization and machine learning architectures to develop a model that balances predictive fidelity with computational efficiency. A central component of this work is the separate treatment of oceanic and terrestrial rainfall regimes. While oceanic precipitation is driven by relatively uniform air-sea interactions, the framework explicitly accounts for the increased structural complexity of landfalling systems by incorporating soil moisture influences and the "Brown Ocean Effect". Resolving these physical transitions allows for a more robust characterization of rainfall evolution and its subsequent conversion into terrestrial runoff. The proposed model is designed for direct integration as a source term within the DG-SWEM framework, avoiding the high computational cost of fully coupled atmospheric-oceanic models and making real-time ensemble forecasting feasible. By improving the representation of rainfall-induced forcing, this work better captures the non-linear compound flooding effects where simple linear superposition of surge and runoff is insufficient.