Quantifying the effects of precipitation, vegetation, and land preparation techniques on runoff and soil erosion in a Loess watershed of China.

The combination of land preparation techniques and vegetation could be an effective way to combat soil degradation on vulnerable, steep slopes. Because of the scale of re-vegetation and man-made micro-topographies in regions of hillies and gullies, quantifying the effects of land preparation techniques, precipitation, and vegetation on runoff and soil erosion remains challenging, particularly in semi-arid areas. This study investigated the runoff and erosion characteristics associated with different land preparation techniques (level bench, level ditch, fish-scale pits and adverse-grade tableland) in combination with different tree species (i.e., Caragana microphylla, Pinus tabulaeformis, Armeniaca sibirica, Platycladus orientalis) using network and redundancy analyses (RDA). Network analysis was used to identify the factors (rainfall features, vegetation types, and land preparation techniques) influential in surface runoff and erosion, while RDA was used to focus on the relations between surface runoff, soil erosion, and all of the influencing factors. Our results suggested that land preparation technique could reduce runoff generation on steep slopes, a key process that affects soil loss, although the extent was affected largely by the influence of the combination of the design of the engineering structure and the shape of the vegetation canopy. Our study indicated that Network analysis and RDA are practical methods to quantify the interactions and co-dependencies between rainfall features and other critical factors (vegetation type and ecological engineering) on runoff and soil loss that were difficult to assess previously using classical regression.