2.2. Experimental design and methods
According to the change of riparian slope in topographic structure, 4 field natural slopes with few human disturbances (5°, 10°, 15°, 20°) were selected in the riparian zone (Fig. 1c). Meanwhile, the soil properties and vegetation types were similar among the 4 slopes (Table 1). Four experimental plots were set up on the slope of the riparian zone and each experimental plot was surrounded by steel flumes (300 cm length × 100 cm width × 30 cm height) so as to separate the plots from the surrounding environment to prevent runoff from overflowing.
[Insert Table 1]
Vegetation patches are severely fragmented according to field survey in the study area of riparian zone, therefore, 3 vegetation cover in this study were preliminarily designed as 0%, 15% and 30% respectively in the experimental plots. The extraction method of vegetation cover in the experimental plots is that: Firstly, simply weed the grass and repair the ground surface; Secondly, the vertical photography of the experimental plots is taken with a camera (Canon SX60 HS), Finally, Photoshop CS4 software (Adobe Systems Inc., San Jose, California, USA) was used to process the photos to obtain vegetation cover. In order to ensure the accuracy of vegetation cover, the process is repeated many times until the vegetation cover obtained is within 0.5% of the error compared with the actually designed vegetation cover.
The experimental system included a storage water tank, a water scouring supply system (composed of a constant head scouring setup, water pipes, water pump, and water valve), steel catchment collector, and plastic container (Fig. 1e). The water tank was filled with water to ensure consistent hydraulic pressure throughout the experiment. The constant head scouring setup was placed at the top of the steel flume to keep a uniform water flow into the experimental plots, and a steel catchment collector was placed at the foot of the steel flume to collect the runoff and sediments. Then, a plastic container (20 L) was used to catch the runoff and sediments at the outlet of the steel catchment collector. According to the discharge per unit width produced by the heavy rain in the study area, the scouring intensity was set to 15 L/min. In order to avoid any effects of river sediment on the experiment, the water in the tank was obtained from underground through a water pump.
The runoff scouring experiments were conducted with 4 slope gradients (5°, 10°, 15°, 20°) and 3 vegetation cover levels (0%, 15%, 30%), respectively. A total of 12 treatments were conducted, and each treatment lasted 30 min and repeated twice. Considering the heterogeneity in the initial soil moisture of the different slopes of the experimental plot, a sprinkle water process was applied to soil moisture uniform. The volume of the sprinkling water depended on the saturation of the soil surface, which should be fully saturated with no runoff. The water discharge was adjusted close to 15 L/min and repeated 3 times to ensure the error was within 2%. Runoff-initiating time was measured by a stopwatch at the beginning of each experiment. During each experiment, runoff and sediment samples were collected in a marked plastic container at 1 min (0–5 min) and then at 5 min (5–30 min) intervals for a total of 10 samples. The container was changed several times in each interval, and the total runoff volume of each interval was recorded.