Tidal current turbines operate in a highly turbulent environment, which results in significant fluctuations in unsteady loads and consequent fatigue and dynamic failures in the rotor. Therefore, accurate calculation of these unsteady loads is crucial. However, the existing models do not take into account the influence of varying relative blade section thickness on dynamic stall when predicting the unsteady characteristics of horizontal axis tidal turbine (HATT) blades. To forecast the unsteady characteristics of the HATT blades, this study developed a method that considers the influence of relative blade thickness on dynamic stall. The modified Leishman-Beddoes (L-B) dynamic stall model, 3D rotational augmentation model, and steady-state BEM model were combined to predict the unsteady hydrodynamic characteristics of blades by synthesizing the transient inlet velocities induced by waves, shear flow and turbulence. In this method, the static parameters required for the modified L-B dynamic stall model were calculated by XFOIL within a range of small angle of attack (AoA) and then extrapolated to cover the entire range of AoA using the Viterna extrapolation method. The required dynamic parameters were obtained by CFD numerical simulation. This method, that we call HATT-UCPM, can consider both the effects of a single element on the unsteady hydrodynamic characteristics of the blades, as well as the combined effects of shear flow, waves, and turbulence on the blades in the actual marine environment. The accuracy of the method was validate through CFD numerical simulation. The results showed that the proposed method has a high consistency with the numerical simulation results, indicating that it has a good accuracy in the prediction of unsteady characteristics of HATT blades. Finally, the study examined several cases to investigate the impacts of shear flow, waves, and turbulence on the unsteady loads of blades, both individually and in combination.