Abstract: To clarify the effects of landuse types on the turnover of soil organic carbon in Karst mountain areas, soil organic carbon content and its mineralization rates were measured in three soil layers (0-10, 10-20, and 20-30 cm) under shrub (SR), paddy (PD), dry land (DL), 3-year-old grassland converted from cropland (GCC-3), and 15-year-old grassland converted from cropland (GCC-15) in a typical demonstration area of desertification control experiment in southwestern Guizhou. The results showed that the contents of soil organic carbon in SR, PD, DL, GCC-3 and GCC-15 were 30.37, 31.24, 21.86, 17.49 and 22.50 g·kg^-1, respectively. The contents of soil organic carbon in SR and PD were significantly higher than in DL, GCC-3, and GCC-15 (P<0.05). The mineralization rate of soil organic carbon was higher at the early stage, and lowered at the later stage. There were differences in the mineralization rate of soil organic carbon among the different land-use types. The mineralization rates of soil organic carbon in GCC-3 and GCC-15 were higher than under the other land-use types, and the mineralization rate of soil organic carbon in DL was higher than in PD. In the 0-10 and 10-20 cm layers, the longest halflife time of soil organic carbon was all found in SR, being 722 and 639 days in the corresponding layers, respectively. The content and half-life time of soil organic carbon in each of the three soil layers were greater in PD than in DL, GCC-3, and GCC-15, suggesting that PD could be one of the best land uses for long-term carbon sequestration in Karst mountain areas. In short, land use, especially returning cropland to grassland, was an important factor influencing soil organic carbon and its turnover rate in Karst mountain areas.

Key words: diameter class length distribution, soil warming, nitrogen addition, extreme value model.