Abstract: A method to identify modal parameters for large bridges is presented using empirical mode decomposition (EMD) and random decrement technique (RDT) simultaneously. Firstly, the non-stationary response of the bridge under ambient excitation is decomposed into a series of quasi-stationary intrinsic mode functions (IMFs) by carrying out EMD. Secondly, RDT is applied to certain IMFs to obtain the free responses. Finally, the free vibration signals are expressed in analytic forms in which modal parameters of the bridge are included. Parameter identification theories and optimum estimate theories are used in obtaining multiple modal parameters of the bridge. The modal parameters of the Nanjing Yangtze River Bridge identified from the measured response by the method presented in the paper are compared with those provided by finite element analysis and related measured values. The results demonstrate that the method is effective，robust，and promising to modal parameter identification for large bridges.