中文摘要：

土壤侵蚀是土壤有机碳（Soil organic carbon,SOC）动态过程的重要驱动因素,明确土壤侵蚀如何影响土壤微生物进而作用于SOC,有助于准确把握土壤侵蚀在全球碳循环中的作用。通过野外径流小区模拟降雨试验,结合定量聚合酶链式反应（quantitative Polymerase Chain Reaction,q PCR）技术,研究了水力侵蚀后短期内（10 d）坡耕地表层土壤微生物数量和SOC含量动态变化特征,并在此基础上探讨了微生物与SOC间的关系。结果表明：与雨前相比,降雨侵蚀后表层土壤SOC含量没有显著差异,而表层土壤细菌数量显著降低,为雨前细菌数量的58.76%（坡上）、55.22%（坡中）、55.82%（坡下）;降雨侵蚀同样显著改变了表层土壤真菌数量,雨后真菌数量为雨前真菌数量的105.51%（坡上）、2.29%（坡中）,12.20%（坡下）;降雨侵蚀后,SOC、细菌和真菌数量均在短时间内显著增加,达到峰值后下降;相关性分析表明,细菌和真菌数量与SOC之间的关系均未表现出显著正相关关系,仅有坡下细菌,坡中、坡下以及整个坡面真菌与SOC含量表现出显著正相关关系。

英文摘要：

Soil erosion is an important driving factor of soil organic carbon （SOC） dynamics and plays an extremely important role in the long-standing problem of ＂missing of carbon sink＂. Transport and deposition of soil particles on the earth＇ s surface triggered by soil erosion cause variation of soil microbes in distribution with the position of soil erosion. which in turn affects carbon sequestration and mineralization on the soil. Therefore, it is necessary to figure out how soil erosion affects soil microorganisms and then acts on soil organic carbon. The knowledge will help understand correctly the role of soil erosion in global carbon recycling A field simulated rainfall experiment was conducted on a runoff plot （2m x 5m ） in a red soil hilly region of South China, to study dynamic changes in SOC and number of microbes in the surface soil of the plot within a short time period （ 10 days） as affected by rainfall and the resultant soil erosion with the aid of quanti- tative Polymerase Chain Reaction （qPCR） technology. On such a basis, relationship between soil microbes and SOC was analyzed, so as to provide some fundamental theoretical basis for the exploration of the role of soil microbes in SOC dynam- ics as affected by soil erosion. The plot was evenly divided into five sections along the slope, namely, A, B, C, D, and E （each 1 m long）. Prior to rainfall simulation, soil samples were collected separately from the soil layers （0 - 10 cm depth） of Sections A （Up slope, US）, C （Middle slope, MS）, and E （Lower slope, LS） for analysis of basic soil prop- erties. Boreholes left by the samples were immediately filled up with soil from nearby and carefully leveled so as to mini- mize any possible impact of the sampling on effect of the later rainfall simulation. Soil samples were collected again in the same manner as soon as the simulated rainfall stopped. The second batch of samples were collected 35 h after the rainfall stopped, and then the third, forth batches were at 24 h in