Background In semi-arid areas, soil erosion is a serious threat to land productivity and sustainability for both natural and human-managed ecosystems. Traditional vegetation techniques are recognized as effectively in reducing soil erosion, whereas the most evident vegetation source that protects soil against erosion is root wedging, which is an important mechanism where roots can bind soil together and tie weak surface soil layers into strong and stable subsurface layers. Plant roots significantly affect soil erosion process of overland flow by physical consolidation (root link and root-soil adhesive) and biochemical functions.

Methods This study introduced electrical impedence method, with the purpose to evaluate the relative contributions of root link, root-soil adhesive as well as root biochemical functions to soil erosion resistance non-directly. For this purpose, a simulated scouring experiment was conducted on a sandy soil with sand content 36.8%, silt content 51.2% and clay content 12.0%. Three treatments considered were:1) fallow (CK), 2) root-penetrated soil and 3) simulated-root-penetrated soil. Each treatment had 4 replicates. Rectangular, undisturbed soil samples (20 cm×10 cm×10 cm) were taken in the fallow and root pans and were conducted with a hydrological flume (2 m×0.10 m). The flume contained an opening at its lower base, equaling the size of metal sampling box, so that the surface of soil sample was at the same level of the flume surface. Space between the sample box and the flume edges was sealed with painter' mastic to prevent edge effects. The slope of the flume bottom could be varied and clear tap water flow was applied at 4.0 L/min rate discharge on a washing flume slope of 15° for 15 min. During the 15 minutes of each experiment, samples of runoff and detached soil were collected in every 1 min in the first 3 min and 2 min in the following time using 10 L buckets for determining sedimentation. Then, this paper analyzed the relative role in creating the soil configuration of soil resistance to erosion quantitatively, using no root-penetrated soil, root-penetrated soil erosion simulation test.

Results The linear equation can well fit the relationship between the soil resistance and the accumulated sediment loss, and the mean of the per 3 min electrical impedance of the flow can better reflect the soil physical consolidation effect of the root system. Physical consolidation effect is the key role in soil erosion resistance, with the contribution rate of 66.5%. The values obtained by electrical impedance method is 13.5% smaller than that of the collected sediment method, though a significant relationship is found between them.

Conclusions To some extent, the electrical impedance method is able to estimate soil erosion resistance of root non-directly. It is expected to provide new technical means for estimating soil consolidation effects of plant roots in the flow-induced erosion regions.