Abstract: Abstract: Soil organic carbon (SOC) is the key indicator in assessing soil quality, and it is also the important source and sink in global carbon cycle. Usually, the stratified summation method is used in estimation of SOC concentration at small scale, but it is costly and time-consuming process since it needs a large number of soil samples at regional scale. Recently, the vertical distribution models, such as the negative exponential, power and logarithmic functions are used to describe the changes of SOC content with the increasing of soil depth. The vertical distribution of soil texture in alluvial plain is very complicated. However, there are few reports on assessment of the suitability of the SOC vertical distribution model in an alluvial plain. The objectives of this study were to construct and assess a vertical distribution model to describe the changes of SOC content in an alluvial plain, and to determine the main variables that affected SOC concentration distribution. In this study, 605 soil samples were collected from 121 soil profiles in an alluvial plain area of Quzhou county, Hebei Province. SOC contents from topsoil to 1-m depth were determined. The vertical distribution model of SOC was constructed based on negative exponential function, and then regional spatial distribution of SOC concentration was obtained by geostatistical methods. The results indicated that SOC content showed a gradually decreasing trend with the increase of soil depth, and the mean SOC content in topsoil was the highest, reached to 8.25 g/kg soil. The coefficients of variation of SOC content for all layers ranged from 0.26 to 0.43, and all belonged to moderate degree of variation. The spatial continuity was better for SOC in 0-20 and >20-40 cm as compared to the rest soil depths, and their correlation distances were 14 and 3 km, respectively. However, SOC in subsoil (>40 cm) showed a pure nugget effect, which reflected the complex spatial distribution of soil textural layers in an alluvial plain. The negative exponential model can well describe the changes of SOC content with the increasing of depth in alluvial plain area, the root mean squared error was only 0.70 and the coefficient of determination of the predicted and measured SOC contents reached to 0.95. Spatial distribution of SOC density showed a decreasing trend from northwest to southeast across the county. Soil types and soil texture were the main influencing factors. The SOC concentration of fluvo-aquic soil and salinity fluvo-aquic soil were significantly higher than that of cinnamon fluvo-aquic soil. The SOC concentration of fine textural soil (light loam, medium loam and clay) was significantly higher than that of the coarse textural soil (sand and sandy loam). The constructed vertical distribution model can well describe the changes of SOC content in soil profile, which not only provides a new method to estimate SOC contentin alluvial plain area, but also can serve as guidance on evaluation and improvement of regional soil fertility.