Abstract：The flow of surface water under vegetation cover has a significant impact on the hydraulic erosion of slope. Most previous studies focused on vegetation type, coverage, stiffness, diameter, submergence and height, particularly on the relationships between some botanical attributes and flow resistance. In recent years, the spatial location and distribution pattern of vegetation, such as sloping land, have become drawn much attention. However, few researches have been conducted on the sloping land under natural conditions, where less disturbed by human activities and covered by fragmented patchy vegetation. To clarify hydrodynamic characteristics of overland flow covered by patch vegetation, a fixed-bed resistance simulation scouring test was conducted at the State Key Laboratory of Soil Erosion and Dry－land Farming on the Loess Plateau, China. The test was set up with 6 slopes with a range of 0.034 9 to 0.207 9. According to the critical intensity of erosion rain on the Loess Plateau, seven single-wide flows were designed, ranging from 0.278×10-3 to 2.500×10-3 m2/s.To simulate the vegetation distribution in the natural state, the grass was randomly arranged in patches, where5 coverages were set up to explore the critical coverage of vegetation on the slope. A systematic analysis has been made to investigate the resistance mechanism of slope flow zone under patch-like vegetation. The results showed that: 1) Under the experimental conditions, the overland flow over vegetation coverage could be considered as the virtual laminar flow and transitional flow, and this evolution process of the water flow pattern was restricted by vegetation coverage and slopes. As the slope gradient increased, the flow pattern extended from the subcritical flow to supercritical flow, while it showed an adverse trend when the coverage degree increased. 2) The formation mechanism of slope resistance was closely related to the coverage condition, flow discharge and water depth. The proportion of particle resistance decreased when these three factors increased, but it had nothing to do with the slope gradient. 3) In the transitional flow zone, the comprehensive resistance coefficient was negatively correlated with Reynolds number, while the aforementioned relationship was mainly influenced by the varying vegetation coverage degree in the transitional zone. With the increase of coverage degree, it gradually changedfrom the negative correlation to positive correlation, and the influence of slopes gradually became dominate. 4) The comprehensive flow resistance coefficient showed a power function relationship with the coverage degree. When the coverage degree did not reach a critical value, the comprehensive flow resistance coefficient rose significantly as the coverage degree increased, but this trend tended to be stable as the coverage degree was relatively larger than the critical value. This relationship was alsorestricted by the flow discharge and slope gradient, that is why the f-Cr relationship curve would cross under the different flow discharges when the coverage degree increased. 5) The relationship between the comprehensive flow resistance coefficient and submergence degree was also related to the coverage degree. When it did not reach to the critical coverage degree, they were negatively related. But as it increased continuously, the comprehensive flow resistance coefficient first decreased, and then increased as the submergence degree increased. 6) In addition, usingthe stress analysis of water body covered by patch vegetation, a calculation model of overland flow resistance considering slope gradient, vegetation coverage, Reynolds number and submergence degree was established based on the equivalent principle. The determination coefficient was 0.831 and 0.806, respectively in the virtual laminar flow and transitional flow zone. It can be seen that the Reynolds number played a leading role in the virtual laminar flow zone, while the coverage degree was more important in the transitional zone.As the Nash coefficient was 0.836 and 0.784, it indicates that this model can simulate the flow resistance on slopes covered by patch vegetation. This finding provides a theoretical basis for adopting vegetation measures for soil and water conservation.