The accuracy of data-driven landslide susceptibility prediction depends heavily on the quality of non-landslide samples and the selection of machine-learning algorithms. Current methods rely on artificial prior knowledge to obtain negative samples from landslide-free regions or outside the landslide buffer zones randomly and quickly but often ignore the reliability of non-landslide samples, which will pose a serious risk of including potential landslides and lead to erroneous outcomes in training data. Furthermore, diverse machine-learning models exhibit distinct classification capabilities, and applying a single model can readily result in over-fitting of the dataset and introduce potential uncertainties in predictions. To address these problems, taking Chenxi County, a hilly and mountainous area in southern China, as an example, this research proposes a strategy-coupling optimised sampling with heterogeneous ensemble machine learning to enhance the accuracy of landslide susceptibility prediction. Initially, 21 landslide impact factors were derived from five aspects: geology, hydrology, topography, meteorology, human activities, and geographical environment. Then, these factors were screened through a correlation analysis and collinearity diagnosis. Afterwards, an optimised sampling (OS) method was utilised to select negative samples by fusing the reliability of non-landslide samples and certainty factor values on the basis of the environmental similarity and statistical model. Subsequently, the adopted non-landslide samples and historical landslides were combined to create machine-learning datasets. Finally, baseline models (support vector machine, random forest, and back propagation neural network) and the stacking ensemble model were employed to predict susceptibility. The findings indicated that the OS method, considering the reliability of non-landslide samples, achieved higher-quality negative samples than currently widely used sampling methods. The stacking ensemble machine-learning model outperformed those three baseline models. Notably, the accuracy of the hybrid OS–Stacking model is most promising, up to 97.1%. The integrated strategy significantly improves the prediction of landslide susceptibility and makes it reliable and effective for assessing regional geohazard risk.

Different land utilization types have unignorable impacts on adjacent aquifers, so studying the effects of varying land utilization types on groundwater balance and groundwater table in arid and semi-arid areas is crucial to facilitate the rational development of territorial space and groundwater resource management. This study investigated the relationship between land utilization type variations and groundwater dynamics from spatial and temporal scales in Huocheng plain, Xinjiang, China, via taking advantage of different land utilization maps provided by remote sensing techniques in 1990, 2000, and 2015. Based on the analysis of results, the findings can be summarized as follows. First, the cultivated and construction land has expanded significantly in the past 25 years. The migration of the regional center of cultivated land was insignificant, as it is still located almost in the center of the entire study area. However, the transfer span of the location center of construction land was more prominent, and its moving direction was mainly from south to northeast, directly affecting the groundwater balance state and groundwater table. Second, the total amount of groundwater storage exhibited a remarkable reduction, from a recharge in 1990 to a continuous discharge in 2000–2015, during which the total recharge and discharge had both continuously declined. Meanwhile, recharging and discharging elements were also changed profoundly as less groundwater recharges from ambient aqueous environments and more groundwater extraction. Thus, human activities were the fundamental driving terms causing the changes in groundwater sources. Third, the groundwater table gradually and continuously dropped from south to northeast in the study area over time. The drawdown of spatially averaged groundwater tables was positively correlated with construction and cultivated area extension. In addition, compared to the former literature, this study offers an alternative approach for groundwater protection and optimization of territorial space development and utilization in arid and semi-arid areas.