Massive floating macroalgal blooms in the ocean result in many ecological consequences. Tracking their drifting pattern and predicting their biomass are essential for effective marine management. In this study, a physical–ecological model, the Floating Macroalgal Growth and Drift Model (FMGDM), was developed. Based on the tracking, replication, and extinction of Lagrangian particles, FMGDM is capable of determining the dynamic growth and drift pattern of floating macroalgal, with the position, velocity, quantity, and represented biomass of particles being updated synchronously between the tracking and the ecological modules. The particle tracking is driven by ocean flows and sea surface wind, and the ecological process is controlled by the temperature, irradiation, and nutrients. Based on the high-precision ocean numerical model, the FMGDM could simulate and predict the biomass and distribution of the floating macroalgal.  We focused on the large-scale green tide disasters bloomed repeatedly in the Yellow Sea of China. The flow and turbulence fields were provided by the unstructured grid Finite-Volume Community Ocean Model (FVCOM), and biological parameters were specified based on a culture experiment of Ulva prolifera, a phytoplankton species causing the largest worldwide bloom of green tide in the Yellow Sea, China. The FMGDM was applied to simulate the green tide around the Yellow Sea in 2014 and 2015. The model results, e.g., the distribution, and biomass of the green tide, were validated using the remote sensing observation data. Given the prescribed spatial initialization from remote sensing observations, the model was robust to reproduce the spatial and temporal developments of the green tide bloom and its extinction from early spring to late summer, with an accurate prediction for 7–8 days. With the support of the hydrodynamic model and biological macroalgae data, FMGDM can serve as a model tool to forecast floating macroalgae blooms in other regions.