Abstract:China's Natural Forest Conservation Program was launched in 1998. Since then, a harvesting model based on logging broadleaved species while retaining Pinus koraiensis has been widely applied in the Changbai Mountain region of Northeast China. The primary goal of this model is to promote the return of degraded secondary forests to primary broadleaved Korean pine mixed forests (BKFs). However, because of differences in tree species composition among secondary forests, this harvest model has led to a decrease in timber resources in some parts of the region. In order to explore a sustainable utilization program for BKFs, we simulated the dynamics of total timber stocks under different harvest scenarios. We selected five secondary 1-hm2 BKF plots, which had been established in 2007 in the area administered by the Lushuihe Forestry Bureau. In these plots, Pinus koraiensis accounted for approximately 20%, 30%, 40%, 50%, or 60% of the gross volume. We divided each plot into 25 subplots, each measuring 20 m×20 m. Within these plots, we identified and measured all free-standing trees with a diameter at breast height of ≥2 cm at 1.3 m above the ground. To simulate volume dynamics in the plots, we divided the tree species into three groups, namely, conifer trees, commercial broadleaved trees, and other trees. For each group, we applied volume growth equations and survival index equations to predict the volume dynamics of stands. In addition, we tested the validity of the harvesting model by using paired t-tests to analyze data obtained over a 10-year period from 10 permanent plots, ranging in area from 0.25 hm2 to 0.5 hm2. We calculated the gross volumes and the volumes available for harvesting of stands under a range of management programs, with harvest intensities of 20%, 30%, or 40% and cutting cycles of 10 years, 20 years, 30 years, or 40 years. In addition, we divided the programs into two groups: in the first group, harvesting of Pinus koraiensis was permitted, whereas in the second group, harvesting of Pinus koraiensis was prohibited. We verified the validity of the harvesting model (t=0.229, P=0.829). The logging schemes that prohibited harvesting of Pinus koraiensis not only restricted the volume of available timber but also made restoration more difficult. Even under the lowest harvest intensity of 20%, the available timber volumes did not return to original levels after 40 years of restoration in secondary forests, where Pinus koraiensis comprised >40% of the total volume. Hence, to achieve sustainable utilization in secondary forests, the characteristics of tree composition should be considered when developing harvest schemes. For secondary forests in which Pinus koraiensis comprises >40% of the total volume, the harvest scheme should permit harvesting of Pinus koraiensis with a cutting intensity of 20% at 30-year intervals. In contrast, for secondary forests in which Pinus koraiensis comprises < 40% of the total volume, the harvest scheme should prohibit harvesting of this species, with a cutting intensity of 20% at 40-year intervals. We further showed that the period required for restoring the volume available for harvesting was longer than the period required for restoring the gross volume. Hence, evaluation of the logging cycle according to the period required for restoring the volume available for harvest will more efficiently promote sustainable forest utilization.