Mangroves are the unique intertidal plant formations growing in sheltered tropical and subtropical coastal areas. The past decade, many measures were undertaken for mangrove forestation in the Aojiang Estuary, Zhejiang Province, with 47 hm² of existing Kandelia obovata forest. The present experiment was conducted to assess the population biomass of K. obovata in Aojiang Estuary. Mathematical methods that use easily measured variables to predict difficult-to-measure variables are important to mangrove managers. As a result, standard plant methods and allometric equations have been developed for several decades to estimate mangrove biomass. Single-stemed mangrove production was usually estimated by allometry between biomass and stem diameter at breast height. Because mangroves are usually dwarf forests in higher latitude sites, and moreover, the crown bases and multi-stems of dominant individuals may begin within a few decimeters of ground level, estimates of community production that depend on allometry based on single-stemed mangrove may not be accurate. Here, we develop allometric relations to predict total biomass and individual components of biomass (e.g., leaves, stemts, roots and butts) of K. obovata, a multi-stemmed mangrove, in the Aojiang Estuary, Zhejiang province. This procedure treated each stem as a discrete tree that shared a proportion of the butt and other elements common to all stems. Linear log-log relationships were obtained between biomass and stem diameter at one-tenth of the stem length nearly the ground. Population biomass of artificial K. obovata forest in Aojiang Estuary was calculated according to the function model. We compared the difference on population biomass of K. obovata in different regions of China. The results showed that K. obovata biomass (W) correlated to the stem diameter (D) at a significance level (P < 0.001). The function model between plant biomass (leaf, W_L; stem, W_S; Root and butt, W_B; and total, W_T) and stem diameter (D) was as follows: W_L=0.187D^1.855 (R² =0.612, P < 0.0001); W_S=0.267D^1.906 (R² =0.821, P < 0.0001); W_B=4.6D^1.136 (R² =0.644, P < 0.0001); W_T=3.614D^1.446 (R² =0.801, P < 0.0001). The regression relationship between K. obovata aboveground biomass and stand age and latitude was significant in different regions of China, that is, lg(aboveground biomass)=3.123 + 0.84lg(stand age)-2.019lg(latitude) (R² =0.431, F_{2, 11}=4.161, P =0.045). Population biomass of K. obovata increased with increased stand age, while trend to decrease with increased latitude. Population biomass of 3-, 5- and 10-year-old K. obovata forest in Aojiang Estuary was estimated at 7.13, 11.32 and 24.35 t/hm², respectively. The 5-year-old population biomass in this experiment was only of 18% compared with the same age population of K. obovata grown in natural wetlands in Zhanjiang, Guangdong province. However, the biomass of 3-year-old population grown in artificial wetlands in Shenzhen was only 9.3% of the same age population biomass in this experiment. In addition, the regression relationship between population density and mean individual biomass was estimated based on the data of ≤ 11-year-old artificial pure K. obovata forest: lg(mean individual aboveground biomass)=8.468-2.1 × lg(population density), (R² =0.961, F=99.764, P =0.001). This equation indicates that mean individual biomass increased with decreased population density and the self-thinning index was -2.1, which approximately accorded with the -3/2 power law. Therefore, not only stand age and latitude affect K. obovata population biomass, but habitat types and population density are crucial to K. obovata population biomass accumulation.