5. Conclusions
The present study investigated the SDC of four treatments, namely a control, freeze-thaw, root system and freeze-thaw + root system group by a scouring test. The relationships of hydraulic parameters and root weight with SDC were explored. The present study made the following conclusions:
(1) The SDC of sand soil was higher than that of loessal soil, with both soils showing similar relative SDC responses to the four treatments. While SDC can be increased by freeze-thaw, the increase was not significant. In contrast, SDC was significantly reduced by the root system effect. Under conditions of a combined effect of freeze-thaw and the root system on SDC, the effect of the root system dominated. Among the three factors, the strength of relationships to SDC were: root system> freeze-thaw > soil type.
(2) Positive linear relationships existed for SDC with shear stress and stream power, whereas the relationships of SDC with unit energy of the water carrying section and unit stream power could be represented by a power function. In addition, stream power among the hydraulic parameters was found to have the strongest relationship with SDC under the different treatments.
(3) The SDC model prediction accuracy was improved by including root weight into the model based on stream power.
(4) The model for simulating SDC based on both root weight and stream power was found to be a stronger predictor of SDC when simultaneously considering the root system and freeze-thaw conditions. Furthermore, field observation experiments of SDC driven by stream scour should be increased in the future to verify the practicability of this model.