Land use change is one of the major factors affecting soil organic carbon (SOC) and global carbon balance. The Grain for Green Project, implemented in the hilly Loess Plateau region, has remarkably changed the land use types in this region. However, it is not clear how this has affected SOC sequestration. Our objectives were (Ⅰ) to determine the temporal and spatial response of SOC sequestration to the land use changes, (Ⅱ) to understand the critical factors impacting the SOC sequestration. The research was conducted on the central of the hilly Loess Plateau region, which covered an area of 707km². Soil samples were taken in five catchments with different revegetation types and revegetation chronosequence. The results showed that at the regional scale, natural grasslands, planted shrub lands, planted wood lands and naturally revegetated grasslands showed remarkable SOC sequestration compared with slope croplands. However, no significant increase in SOC sequestration was found in planted grasslands and orchards compared with slope croplands. Setting the SOC density in natural grassland as the target, the naturally revegetated grassland had a SOC sequestration potential of 8.3t/hm² in top soil (0-20cm). We found no increase in SOC sequestration in the first 10 years after the Grain for Green Project implemented. However, after 10 years, we saw a significant increase in SOC sequestration in planted woodlands, planted shrub lands, revegetated grasslands and slope croplands. The effect of the Grain for Green project on SOC sequestration was mainly influenced by land use type and age, which explained 55.6% and 24.1% of the variation of SOC. Topographic factors, including slope aspect and slope position explained 8.5% of the variation in SOC. The SOC density in the deep soil layer (100-200 cm) in the planted woodland (dominated by Robinia pseudoacacia), planted shrub land (dominated by Caragana Korshinskii) and revegetated grassland was about 35%-40% of that in the 0-200 cm soil profile. The SOC sequestration in deep soil layer (100-200cm) was significantly increased after revegetation. Therefore, our study showed that evaluation of the Grain for Green project needs to account for 1) land use type and age, 2) SOC sequestration in the deep soil layer and 3) long-term effects (>10 years) rather than short-term effects. Otherwise, the effect of this program on SOC sequestration will be significantly under-estimated.