Particle sorting and transport dynamics of eroded sediment can be expected to clarify the erosion mechanism of spoil tips in soil and water conservation. This study aims to investigate the response of sediment particle sorting and transport to hydrodynamic parameters during erosion. Two spoil materials, namely the Lou soil and aeolian sandy soil, were selected as research objects. Runoff scouring experiments were conducted under two slopes (28°and 32°) and four inflow rates (8, 12, 16, and 20 L/min). A systematic analysis was also implemented on the particle-size distribution of eroded sediment, the sorting features of sediment particles, and the transport mechanisms under different soil types. In addition, the relationships between sediment particle characteristics and hydrodynamic parameters were examined to identify the key controlling factors of sediment sorting and transport during erosion. The results showed that the eroded sediments from both the Lou soil and aeolian sandy soil spoil tips were dominated by the 0.002~<0.050 mm size fraction, which accounted for more than 46% of the total. There was great variation in the response patterns of sediment particle composition to slope and inflow rate. In the aeolian sandy soil spoil tips, the particle-size composition of eroded sediment was more sensitive to slope, indicating that the slope played a more significant role in the detachment and transport of sediment particles. In the Lou soil, the transport of coarse particles in the 0.250~2.000 mm size fraction was controlled by the inflow rate. There was a significant correlation between slope and inflow rate. The influence of hydraulic conditions on sediment transport was strongly dependent on the soil type and particle composition. Compared with the Lou soil spoil tips, the eroded sediment from the aeolian sandy soil spoil tips shared a larger mean weight diameter, ranging from 0.061 to 0.085 mm, and a smaller fractal dimension, ranging from 2.494 to 2.561. The coarse particles in aeolian sandy soil were more easily detached and transported during runoff scouring, whereas the sediment sorting was weak. In contrast, the relatively stronger sorting was observed in the smaller mean weight diameter and larger fractal dimension of the eroded sediment from the Lou soil. Fine particles were also enriched in the transported sediment. Therefore, there were differences in soil texture and particle composition between the Lou soil and aeolian sandy soil, leading to sorting responses during erosion. Among the hydrodynamic parameters, stream power was identified as the optimal predictor for the sorting characteristics of eroded sediment from the aeolian sandy soil spoil tips. Specifically, stream power showed the strongest relationship with mean weight diameter and fractal dimension (R²= 0.92 and 0.55, respectively). In the Lou soil spoil tips, runoff shear stress was the optimal hydrodynamic parameter for sediment sorting (R²= 0.80 and 0.31, respectively). The contribution rates of suspension-saltation exceeded 61% for the aeolian sandy soil spoil tips and 83% for the Lou soil ones. Most eroded sediment was also transported in the form of suspended and saltating particles rather than as bed load. Moreover, the power and logarithmic functions were used to quantitatively describe the contribution rates of suspension-saltation and bed load in the eroded sediment from the Lou soil and aeolian sandy soil spoil tips under stream power. The stream power served as an effective hydraulic indicator to partition the different transport modes during erosion. Dynamic erosion mechanisms of spoil tips can provide a solid theoretical basis to optimize the differentiated soil and water conservation using specific regional soil types.

Soil salinization is a main environmental and ecological problem in irrigated agro-ecosystems especially for the area with shallow groundwater table. Most previous studies mainly focused on the surface and root zone soil salinity, however, the knowledge on spatio-temporal dynamics of soil salinity for both root zone and deep soil as well as driving factors, and the recommended seasonal and annual groundwater depth for salt control was rarely addressed. In this study, 68 cropland sampling experiments and investigations related to soil salinity from the top soil to the depth of 1.80 m, groundwater depth and salinity, and crop types were carried out before and after the crop growing season of 2017-2018 in the Hetao Irrigation District. Soil salinity statistical characteristics showed that the standard deviation of soil salinity in time and space became larger with increasing of temporal mean and spatial mean. The root zone (within depth of 0-0.60 m) soil salinity in the crop growing season were accumulated and desalinated in the fallow season, which was opposite for changes of the deep (within depth of 0.60-1.80 m) soil salinity. The multi-factor variance analysis of variance showed that the driving factors including crop type, groundwater depth at the beginning of crop growing season, averaged groundwater depth in the crop growing season, groundwater salinity and root zone soil salinity at the beginning of crop growing season had significant effects on root zone soil salinity changes in the crop growing season (P<0.05). The influence of groundwater condition and other factors on root zone soil salinity change during crop growing season was closely related to the hydrological year type. At the beginning of crop growing season, the soil salinity in the root zone decreased exponentially with increasing groundwater depth, and the sampling locations with soil salinity below average had a minimum groundwater depth of 1.60 m. Both salt value and salt reduction rate decreased with groundwater depth increased across different soil layers when the average annual groundwater depth was less than 2.50 m. Thus, the sampling locations were selected where the groundwater table depth was less than 2.50 m, and then the median of root zone soil salinity and groundwater depth was obtained by the frequency analysis method. The annual suitable critical control groundwater depth for soil salinity median was 1.77 m. The groundwater depth of sampling locations where soil salinity was significantly higher than the median or mean value were less than 1.77 m. Therefore, in order to control soil salinity, it is recommended that the groundwater depth should be no less than 1.60 m at the beginning of crop growing period, and to maintain an average annual groundwater depth of at least 1.77 m. The research results will provide theoretical support for the design of irrigation scheduling and the rational layout of drainage projects for similar irrigation area in the arid area.