Changes in overall and inter-variability of runoff and soil loss for a loess soil resulted from a freezing-thawing cycle.

The soil freeze-thaw process is a transition phase of soil water in cold areas that influences the soil's hydrological behavior. However, dynamic phenomena and corresponding consequences have yet to be studied adequately. Therefore, the present study was planned to comparatively analyze the effects of a freezing-thawing cycle on the hydrologic behaviors of loess soil from northeast Iran. Small-size (0.5 × 0.50 m) erosion plots were subjected to a freezing-thawing cycle under governing conditions of the region of the origin soil. The plots were subjected to a freezing-thawing treatment by inducing cold air until the temperature declined to below - 20 °C and lasted for 3 days using a cooling compartment system and then were kept in the laboratory with an ambient temperature of above 10 °C for 2 days. The treated plots and untreated plots were then exposed to a simulated rainfall with an intensity of 72 mm h-1 and 0.5 h duration while they were placed on a slope of 20%. The results indicated that the hybrid processes of freezing-thawing and splash and inter-rill erosions significantly increased runoff generation and soil loss. The time to runoff, runoff volume, and soil loss were 1.65 times less and 1.38 and 2.90 times more, respectively, compared to those reported for the control treatment with significant differences (p < 0.006). The performance of ice lenses, freezing fronts, and the creation of near-saturation moisture after completing the cycle were identified as the most critical factors affecting the different soil behaviors under the frozen-thawed cycle.

How can statistical and artificial intelligence approaches predict piping erosion susceptibility?

It is of fundamental importance to model the relationship between geo-environmental factors and piping erosion because of the environmental degradation attributed to soil loss. Methods that identify areas prone to piping erosion at the regional scale are limited. The main objective of this research is to develop a novel modeling approach by using three machine learning algorithms-mixture discriminant analysis (MDA), flexible discriminant analysis (FDA), and support vector machine (SVM) in addition to an unmanned aerial vehicle (UAV) images to map susceptibility to piping erosion in the loess-covered hilly region of Golestan Province, Northeast Iran. In this research, we have used 22 geo-environmental indices/factors and 345 identified pipes as predictors and dependent variables. The piping susceptibility maps were assessed by the area under the ROC curve (AUC). Validation of the results showed that the AUC for the three mentioned algorithms varied from 90.32% to 92.45%. We concluded that the proposed approach could efficiently produce a piping susceptibility map.