黄土丘陵沟壑区小流域土壤有机碳空间分布及其影响因素

研究局域尺度土壤有机碳空间分布特征,对准确估算大尺度土壤碳库储量和变化具有重要意义。以黄土丘陵沟壑区典型小流域为对象,采集0-10、10-20、20-40、40-60、60-80、80-100cm土层中(898个土壤样品),采用多元线性逐步回归和地理信息系统(GIS)相结合方法,分析了地形(峁顶、峁坡、沟底)、土地利用(农田、果园、川坝地、草地、灌木林、乔木林)等作用下,小流域不同深度土壤有机碳含量的空间分布特征。结果表明:地形因素不仅对表层(0-10cm)土壤有机碳含量空间分布差异影响显著,而且对深层(40-100cm)影响也显著,且空间格局图上40-100cm可以清晰地看地沟底与峁顶和峁坡显著差异。在0-10cm土层,峁顶以中值斑块(50%)和低值斑块(48%)为主;峁坡以中值斑块(62%)为主,其次是低值斑块(22%);沟底中值斑块占70%,其次为低值斑块(23%)。40-100cm均为低值斑块,沟底低值绿色斑块占34%,远高于峁坡(8%)和峁顶(13%)。土地利用对表层(0-40cm)有机碳含量影响显著,对40-100cm土层无影响。在0-10cm土层,乔木林、灌木林、草地上高值斑块分别占18%、47%、10%,川坝地、农田和果园没有高值斑块,中值斑块分别占80%、53%、85%、73%、39%、23%。10-40cm土层,乔木林、灌木林、草地、川坝地、农田和果园中值斑块分别占21%、46%、22%、19%、5%、4%。但在40-100cm土层,各土地利用下有机碳均处于低值斑块区。坡向上0-100cm各层土壤有机碳含量半阴坡(北部、东北、东部)最高,半阳坡(西部、西南、南部)含量较低。

The spatial distribution of soil organic carbon and it's influencing factors in hilly region of the Loess Plateau

Understanding of soil organic carbon (SOC) spatial distribution variability at a local scale are important when developing SOC budgets, explaining the role of SOC reservoir in regional, global climate and environmental issues. In this study, Yangou watershed in the hilly region of the Loess Plateau was selected to investigate the SOC distribution under different land uses and topographies. Topography at the watershed consists of tableland, slopeland, and gullyland. Land use type at the watershed includes farmland, orchard, plain land, grassland, shrubland, and woodland. A total of 53 soil cores was taken as a 100 m-100 m grid across the watershed. At each site, each soil cores with three replicates were segmented into 0-10, 10-20, 20-40, 40-60, 60-80 and 80-100cm sections. Spatial variability of SOC at different depths in relation to topography and land use was evaluated using multiple linear stepwise regression and geographic information system (GIS) analyses. A total of 898 SOC measurements was combined with the digital land use map and the digital topography map of Yangou watershed. The results showed as follows. Topography, land use and aspect significantly influence SOC spatial distribution at 1m soil layer at the watershed and formed spatially arborizations and strips of SOC at the different depths. Topography significantly affect spatial distribution of soil organic carbon in top soil (0-10 cm), but also significantly affect subsoil SOC spatial distribution (40-100 cm). For SOC at the top soil of the watershed, the topography of tableland was dominated by middle value polygon (50%) and low value polygon (48%); slopeland by middle value polygon (62%), followed by low value polygon (22%); gullyland by middle value polygon (70%), followed by low value polygon (23%). For SOC at the subsoil of the watershed, tableland, slopeland and gullyland were also dominated by low value polygon and low value green polygon of gullyland accounted for 34%, much higher than slopeland (8%) and tableland (13%). Land use significantly affect SOC spatial distribution at topsoil depth (0-40 cm), but little effect on subsoil SOC distribution(40-100 cm). Woodland, shrubland and grassland show a greater SOC accumulation than farmland and orchard. Based on SOC distribution across the three topographies, there appears significantly variability in the SOC within 1 m soil depth. For the topsoil depth (0-10 cm), the high value polygon of woodland, shrubland and grassland accounted for 18%, 47% and 10% respectively, while plain land, farmland and orchard were no high value polygon, the middle value plague of them accounted for 80%, 53%, 85%, 73%, 39% and 23% respectively. For the subsoil depth (40-100cm), the middle value polygons of woodland, shrubland, grassland, plain land, farmland and orchard accounted for 21%, 46%, 22%, 19%, 5% and 4 % respectively. But, SOC in 40 100cm across all land uses was at low value polygon. For aspect, SOC contents of the area of semi-shady (East+northeast+north) were high, while the area of semi-sunny (West + South West + South) were low. Soil organic carbon reserves at the watershed at 1 m soil layer was 217.6×10³ Mg, and 67.5% of SOC reserve was at subsoil (20-100cm). We concluded that topography, land use and depth control SOC spatial distribution at a watershed in the hilly region of the Loess Plateau.