The objective of this investigation was to determine the simulation fidelity required to accurately model the aerodynamics of a horn-ice ice accretion.  Several simulations of differing fidelity, ranging from a low-fidelity, simple-geometric simulation to a high-fidelity, 3-D casting, were constructed to model a horn-ice accretion.  The aerodynamic performance parameters C_l, C_m, and C_d and flow visualization images were obtained for each simulation and compared to those of the 3-D casting.  Aerodynamic testing was performed in the University of Illinois 3 x 4 ft. subsonic wind tunnel at a Reynolds number of 1.8 million and a Mach number of 0.18.  The results of the study showed that the horn-ice casting caused a 55% reduction in maximum lift and a 400% increase in minimum drag at this Re and M.  The 2-D smooth simulation had a maximum lift coefficient within 0.2% of the casting, and was the simplest simulation to accurately model maximum lift.  It under-estimated drag slightly, but fell within the uncertainty caused by the spanwise variation present in the flowfield of the casting.  A higher fidelity simulation proved to have a slightly higher accuracy for modeling both lift and drag and the spanwise variation of the casting flowfield, but was substantially more difficult to construct.  Therefore, the 2-D smooth simulation is recommended for most applications.