Abstract:
Abstract: The soil layer structure of the slope would be disturbed during highway construction, leading to heterogeneity of soil properties. Water and solute are always moving in a non-equilibrium flow. Introducing the theory of preferential flow to analyze the migration mechanism of chloride pollutants in slope soil is of great significance to reveal the law of the migration of industrial salt in the protection of road environment. In order to observe the distribution of preferential flow paths in slope soil directly, dye infiltration experiment was used in soils of low subgrade slope of Heda highway. Image processing techniques and morphological computation procedures were used to determine the influence radius and coordinates of preferential flow paths. Spatial point pattern analysis method (O-ring statistic) of landscape ecology with 19 times simulation tests by using Monte Carlo stochastic was used to analyze the spatial distribution characteristics and the spatial correlation of the preferential flow paths with different influence radius. The results showed that the development depth of preferential flow paths in slope soil could reach to 40 cm by dye infiltration experiment. Plant roots were concentrated in topsoil (0-10 cm), and there was no trace of plant roots and biological activities in deeper soil. The number of preferential flow paths with same influence radius decreased as the soil depth increased, while the number of preferential flow paths decreased with the increasing of influence radius in the same soil layer. In the scale of this research, preferential flow paths showed a random distribution state at the range of 1.0≤R<5.0 mm and an aggregation distribution state when R≥5.0 mm. With the gradual increase of influence radius, aggregated distribution pattern was more evident. At the range of 5≤R<10.0 mm, preferential flow paths showed an uniform distribution state in small scale (0-20 mm). The peak of aggregation state usually appeared at a smaller scale (0-40 mm). A large number of gravels lead to a change in the direction of actual movement of water flow, and preferential flow paths will be separated and then merged. The difference of soil particle composition can also lead to lateral water flow, which shows poor correlation between adjacent soil layers. The spatial correlation analysis of preferential flow path shows that the correlation between different preferential flow paths is relatively independent which related to the structure of soil. Large pores in slope soil are mainly formed by the filling of the fine particles of gravel soil, which has strong randomness, good stability, and not easy to collapse. This kind of non-random distribution condition shows that it is very necessary to study the distribution of preferential flow paths in slope soil. O-ring statistic is much more simple and intuitive than the traditional water breakthrough curve method, and it can enrich soil preferential flow research methods. It can also quantitatively analyze the distribution characteristics and spatial correlation of soil preferential flow paths from a statistical point of view. In future, landscape ecology spatial point pattern analysis method will be used well in study on spatial distribution of soil preferential paths.