Landslides cause huge damage to social economy and human beings every year. Landslide susceptibility mapping (LSM) occupies an important position in land use and risk management. This study is to investigate a hybrid model which makes full use of the advantage of supervised learning model (SLM) and unsupervised learning model (ULM). Firstly, ten continuous variables were used to develop a ULM which consisted of factor analysis (FA) and k-means cluster for a preliminary landslide susceptibility map. Secondly, 351 landslides with “1” label were collected and the same number of non-landslide samples with “0” label were selected from the very low susceptibility area in the preliminary map, constituting a new priori condition for a SLM, and thirteen factors were used for the modeling of gradient boosting decision tree (GBDT) which represented for SLM. Finally, the performance of different models was verified using related indexes. The results showed that the performance of the pretreated GBDT model was improved with sensitivity, specificity, accuracy and the area under the curve (AUC) values of 88.60%, 92.59%, 90.60% and 0.976, respectively. It can be concluded that a pretreated model with strong robustness can be constructed by increasing the purity of samples.

The sedimentary structure is important for the engineering design, operation, and safety evaluation of tailing dams. For upstream-method tailing dams, tailing slurry flows and deposits in the pond and forms a complex dam structure. Ore drawing parameters (e.g., slurry concentration and flow rate) have significant influence on the sedimentary structure of tailing dams. However, there is a lack of unified and quantitative understanding of the complicated effects of ore drawing parameters on the deposition behaviour of tailings. In the present study, flume tests were applied to investigate the characteristics of the sedimentary structure of tailing dams. Seven ore drawing experiments were conducted to simulate different slurry concentrations and flow rates. The distribution of characteristic particle sizes d 50 and d 10 of sediment was obtained. Furthermore, considering two dominant features of particle size distribution, a mathematical model for the equation between characteristic particle size and deposition distance was established. The exponential part of this equation describes the decreasing trend of the characteristic particle size, and a smooth step function is introduced to characterize the abrupt decrease in particle size. The experimental data of d 50 and d 10 in all these test cases can be approximated by the equation with correlation coefficients R 2 greater than 0.861. As the slurry concentration of ore drawing increases, the hydraulic sorting gradually weakens. The characteristic particle size distribution curves corresponding to a larger flow rate are generally located above those corresponding to a small flow rate, indicating that the larger the flow rate is, the coarser the sediment. This study provided useful information for the determination of ore drawing parameters in actual tailing dams. The mathematical model of tailings’ particle size distribution can be further used for refined modelling of tailing dams, so as to analyse the safety and stability of the dams.

No abstract