Abstract: The 3-PG process-based forest growth model is increasingly widely used for both research and management of various forests in a number of countries in relation to climate,soil conditions,forest management and species.In order to reliably and rapidly predict plantation production of fast-growing Eucalyptus urophylla,the 3-PG model was applied to study the growth laws in the stands of E.urophylla on the Leizhou Peninsula of Guangdong Province.The model was calibrated on the basis of 4-year monitoring data at Jijia Forest Farm and was tested using growth data at Hetou Forest Farm.The calibration results for various stand variables were very satisfactory(overall mean r² > 0.93 and mean accuracy > 92% for stand growth;accuracy > 83% for stand biomass and LAI) except that root biomass was overestimated about 40%.In both cases of fine and coarse root measurements,root biomass may be underestimated because root system actually extended to greater depth than that can be assessed.In the test,all the mean accuracies were > 94%,with r² > 0.98.The results demonstrated that the 3-PG model was a convenient and efficient means of predicting the productivity of E.urophylla plantations,and can be used to predict the performance of that species and its response to conditions in new areas with the satisfactory accuracy and reliability when parameterized for a particular species.Therefore,it has considerable potential value as a research tool and practical means of providing a sound and credible basis on sustainable E.urophylla plantations for forest managers.