Abstract: For a practical problem in the definite region, it is the most important work to identify parameter values efficiently after the structure of environmental model is defined.In fact, the model calibration is a function optimization problem, which searches the best or better parameter values in their acceptable ranges to attain a match between the observed and simulated distribution or distributions of a dependent variable or variables.By analyzing the features of parameter identification of environmental models, a survey on the identification approaches, improvements and applications and some further research contents and directions are presented.