中文摘要：

为了明确坡面水流中运移距离对不同层次红壤团聚体自身剥蚀程度的影响,选取3个地区（咸宁、进贤、长沙）典型红壤为研究对象,通过团聚体稳定性分析和室内冲刷水槽试验,就不同层次红壤团聚体稳定性与坡面水流中自身剥蚀程度间定量关系进行了初步探讨。研究结果表明,湿筛法及Le Bissonnais法中的快速湿润、预湿润振荡测得不同层次团聚体稳定性为表层红壤最弱,其次为网纹层,铁锰淀积层最强,而慢速湿润无明显规律,消散作用和机械破碎作用是不同层次红壤团聚体的主要破碎机制;不同层次红壤团聚体在5个运移距离后的剥蚀程度均表现为铁锰淀积层〈网纹层〈表层;描述团聚体相对机械破碎指数的RMI与5种运移距离后的团聚体剥蚀程度Wr/Wi、剥蚀系数α相关关系显著,利用RMI和运移距离x2个参数建立了坡面水流中不同层次红壤团聚体剥蚀程度预测方程,方程决定系数较高（R2=0.88）,显示了较好的预测性能。该结果为研究红壤侵蚀机理提供了参考,对完善土壤侵蚀模型具有一定的参考价值。

英文摘要：

In order to determine the effect of migration distance on the abrasion degree of red soil aggregate in different layer in overland flow, the relationship between aggregate stability of red soil in different layer and abrasion degree was studied based on the analysis of aggregate stability and the simulated scouring experiment in laboratory. Red soils from three areas （Xianning, Jinxian and Changsha city） were chosen for this study. The results showed that the aggregate stability of red soils from wetting sieving and LB methods （fast-wetting and stirring pre-wetted） was the greatest for iron-manganese illuvial layer, intermediate for reticulated mottling layer and least for the surface soil layer in the three areas. However, the slow-wetting treatment had no significant regularity between the different layers in this study. The main mechanisms of aggregate breakdown in different layers were slaking and mechanical breakdown. The abrasion degree of aggregate in different layers was iron-manganese illuvial layer 〈 reticulated mottling layer 〈 surface soil layer in five migration distances. The parameter RMI which describes mechanical breakdown showed a significant correlation with the abrasion degree and abrasion coefficient α. By introducing the parameter RMI and transport distance x, the abrasion degree prediction equation of red soil aggregate in different layer was established with a good correlation coefficient （R2= 0.88）. The equation showed a good prediction performance. The results can provide a reference for research on erosion mechanisms of red soil, which is of great significance for the development of adequate soil erosion models.