BP artificial neural network technology has been increasingly used widely in the field of environmental assessment.In this paper,the method had been used to evaluate the ecological function of seabuckthom.A level model based on BP-artificial neural network was set up,which contained 4 input layer nodes,6 cryptic layer nodes and 1 output layer node with index of ecological functions as a sample input and comprehensive assessment level as the network output.The output value was 0.44,greater than grand II standard after actual data of target year(2008) as input.The results showed the seabuckthom significantly improve the ecological environment after seabuckthoms have been planted for ten years.The results was consistent with those obtained from the more mature AHP-fuzzy comprehensive evaluation method,indicating the BP artificial neural network model for evaluating the ecological functions of seabuckthom was feasible and evaluation result was objective.