This study presents an innovative approach employing an iterative Unscented Kalman Filter (IUKF) for the identification of amplitude-dependent nonlinear aerodynamic damping using wind tunnel free vibration data. The wind-structure interaction system is represented as a single-degree-of-freedom system, with the amplitude-dependent aerodynamic damping modeled as a polynomial function of structural displacement and velocity. The augmented state variables, encompassing structural vibration frequency and polynomial coefficients for aerodynamic damping, are concurrently estimated from free vibration data using the UKF technique. To enhance the robustness of the identification results against variations in initial conditions, the UKF is applied iteratively by assigning the estimated polynomial coefficients as new initial values for the state variables. Validation of the IUKF-based method is performed through a numerical example featuring a typical bridge deck sectional model, as well as experimental data from two spring-suspended sectional models experiencing vertical vortex-induced vibration (VIV) and torsional post-flutter limit cycle oscillation. The feasibility of identifying amplitude-dependent aerodynamic damping for amplitude range not covered by the displacement signal is examined.