The pool of soil organic carbon (SOC) in a forest forms an important component in the global carbon (C) cycle. SOC plays an important role in enhancing forest productivity and mitigating the net rate of global greenhouse gas emissions. The risk of global warming has caught the attention of the scientific community as it relates to SOC stocks in forest ecosystems. The precise measurement of SOC stocks and verification of the amount of C sequestered in the soil are critical factors for the implementation of C trading programs. SOC in mineral soils generally decreases with depth; however, this decrease is non-linear and has been frequently modeled as an exponential function. We selected four forest types (boreal forest, temperate deciduous forest, subtropical mixed forest, and tropical evergreen broadleaved forest) and analyzed the exponential function for SOC mass density. We established an SOC database for layers of the soil profile by measuring the SOC in typical areas in the four forest biomes. The depth distribution models for the mass density of SOC were established by a typical sampling method. The model was calibrated using 60% of the data of the profiles, and 40% of the data was used for validation purposes. The entire evaluation for the results of model simulation consisted of determining the coefficient of determination (R²), Nash-Sutcliffe coefficient of efficiency (E), and the percentage error (PE). Next, the depth distribution models evaluated here were used to simulate the distribution of SOC deeper into the soil profile. The results showed that the simulation values for the depth distribution models of the four forest biomes and the observed values were relatively consistent. The average values of R², E, and absolute PE were 0.88, 0.74, and 6.95%, respectively. The model simulations had a relatively high capacity (E > 0.6), and the PE of the model was simulated within a range with acceptable accuracy (PE < ±15%). The model could be used to simulate the depth distribution of forest soil organic carbon. Second, the boreal coniferous forest had a much higher density of SOC in the 0-20 cm layer than those of the tropical deciduous forest and the two other forest types. In contrast, the SOC densities in other layers of boreal coniferous forest were lower, while those of the tropical deciduous forest were higher. The regional SOC densities were lower when SOC densities in the 0-100 cm soil layers were used to characterize the regional SOC density. When compared with the SOC densities in the range of 0-200 cm in the soil profile, the SOC densities in the 0-100 cm soil layer were about 21.8% lower than the overall density. Any error in this calculation may be greater and more prominent in regions with high temperatures and precipitation rates. For rainfall events of a small magnitude, the model generally over-estimated mass density at the bottom of the soil profile, while the opposite was true; that is, for regions with large amounts of rainfall, the model generally under-estimated the surface SOC density. In general, the model performs well at simulating the depth distribution of SOC, and it can be used as a forest SOC management tool to simulate the depth distribution of SOC in some regions.