Biomass growth models for individual tree of main indigenous broadleaf tree species in Guangdong Province

Objective  To calculate quickly and precisely forest carbon sequestration in afforestation projects, we selected major broad-leaved tree species in Guangdong, including Cinnamomum camphora, Schima superba and Liquidambar formosana, and established biomass growth model of individual tree.

Method  All 270 sample trees with 90 sample trees for each tree species were obtained according to 10 diameter classes during the process of modeling. We established four types of biomass growth models for aboveground and underground biomass of three tree species from different origins (natural forest or planted forest) using age as the independent variable. The compatibility issue among growth models of different aboveground components (stem wood, bark, branch, leaf) was solved using optimized models with a set of simultaneous equations and controlled total biomass.

Result  Comparing trees from different origins including natural forest and planted forest, the biomass upper limits and ages of the maximum growth rate for the same species under the same biomass model were different. The biomass upper limits and ages of the maximum growth rate indicated by different equations for the same tree species under the same origin were largely different. When estimating the aboveground biomass, the optimal types of equations for different tree species were different. Logistic model was used to establish the compatibility model for the simultaneous equations of biomass for aboveground components of three tree species. The R_\rm adj^2 values from stem wood biomass equations of three species ranged from 0.560 to 0.768, and MPEs ranged from 3.05% to 6.73%. The R_\rm adj^2 values from bark biomass equations ranged from 0.552 to 0.866, and the MPEs ranged from 2.02% to 6.27%. The R_\rm adj^2 values from branch biomass equations ranged from 0.309 to 0.706, and the MPEs ranged from 3.01% to 14.33%. The R_\rm adj^2 values from leaf biomass equations ranged from 0.495 to 0.767, and the MPEs ranged from 4.16% to 7.14%.

Conclusion  Comparing the parameters and evaluation indexes of four models, the optimal model of aboveground biomass is the Logistic model and the optimal model of underground biomass is the Schumacher model. The proportion of each aboveground component in total aboveground biomass constantly changes with age during the growth process. The compatibility model for the simultaneous equations of biomass for aboveground components of three tree species is established using Logistic model, and the fitting effects of biomass models for stem wood and bark biomass are better than those for branch and leaf. These biomass models could estimate forest carbon combined with carbon coefficient in planted forest for known age in a certain period.