Landslides cause a considerable amount of damage around the world every year. Landslide susceptibility assessments are useful for the mitigation of the associated potential risks to local economic development, land use planning, and decision makers. The main aim of this study was to present a novel hybrid approach of bagging (B)-based kernel logistic regression (KLR), named the BKLR model, for spatial prediction of landslides in the Shangnan County, China. We first selected 15 conditioning factors for landslide susceptibility modeling. Then, the prediction capability of all conditioning factors was evaluated using the least square support vector machine method. Model validation and comparison were performed based on the area under the receiver operating characteristic curve and several statistical-based indexes, including positive predictive rate, negative predictive rate, sensitivity, specificity, kappa index, and root mean square error. Results indicated that the BKLR ensemble model outperformed and outclassed the KLR and the benchmark support vector machine model. Our findings overall confirmed that a combination of the meta model with a decision tree classifier based on a functional algorithm can decrease the over-fitting and variance problems of data, which could enhance the prediction power of the landslide model. The resultant susceptibility maps could be useful for hazard mitigation in the study area and other similar landslide-prone areas.
The western Ordos Basin (OB) is situated at the junction of multiple tectonic units with distinct properties. The prolonged and complex tectonic interactions from adjacent tectonic units have resulted in diverse structural phenomena and intricate evolutionary history in this region. The late Paleozoic represents a critical period for the transition of the tectonic regime in this area. However, due to the effects of intense later-stage modification, the late Paleozoic provenance system and paleogeomorphology of this region remain poorly constrained. Against this background, systematic fieldwork and detrital zircon geochronological analyses of the Youjingshan and Quwushan Permian sections were conducted to determine sediment provenance, and spatial variations in detrital zircon geochronological characteristics across different parts of the OB are further discussed. The results indicate that the detrital zircon age spectra of the Permian Dahuangou and Yaogou formations in the Youjingshan and Quwushan sections are dominated by late Paleozoic (250–360 Ma), early Paleozoic (360–500 Ma), and Paleoproterozoic (1600–2500 Ma) age populations. However, significant differences in age composition are also observed among different samples. This study proposes that the detritus of the Dahuangou Formation in the Youjingshan area was mainly derived from the Alxa Block (AB), while that from the Yaogou Formation was sourced from the Yinshan-Daqingshan-Wulashan Orogenic Belt (YDWOB). In contrast, the West Qinling Orogenic Belt (WQOB) and North Qilian Orogenic Belt (NQOB) were identified as the source areas for the Dahuangou and Yaogou Formations in the Quwushan area. Based on a comprehensive comparison of detrital zircon geochronological data of the Permian strata in the OB, three major provenance systems can be identified: the southwestern source area (WQOB and NQOB); the northwestern source area (YDWOB and AB); and the interior source area (YDWOB). During the Permian, the tectonic-sedimentary evolution of the OB was primarily controlled by the combined effects of the Paleo-Asian Ocean (PAO) to the north and the Paleo-Tethys Ocean (PTO) to the south. Differences in the timing and intensity of subduction/collision between the PAO and the PTO resulted in a general paleogeographic pattern of “higher in the north and lower in the south” in the OB.
Late Permian–Early Triassic basic rocks, which are widespread in the western margin of the Yangtze block (SW China), provide critical information for regional tectonic evolution. The Qiaoqi intrusion, distributed in the western margin of the Yangtze block, is selected as a representative for discussion in this paper. LA-ICP-MS zircon U-Pb dating results show that the Qiaoqi intrusion was formed at 245 ± 1 Ma. It belongs to the medium-K calc-alkaline and tholeiitic basalt series. It is characterized by high concentrations of Fe2O3T (11.53 wt. % to 15.50 wt. %), TiO2 (1.81 wt. % to 3.20 wt. %), Al2O3 (11.80 wt. % to 15.60 wt. %), and low concentrations of MgO (4.51 wt. % to 8.93 wt. %). The LREE and LILE (such as Cs, Rb, Ba and Th) are enriched, with insignificant Eu anomalies (Eu/Eu* = 0.92 to 1.13). The chondrite-normalized REE distribution diagram shows a right-leaning pattern, similar to ocean island basalts (OIB), displaying the geochemical characteristics of enriched mantle sources. The zircon εHf(t) values are relatively high (+12.7 to +15.5) and the single-stage Hf model ages are relatively young (tDM = 272 to 386 Ma). Modeling further reveals that the parent magma was derived from 13% to 19% partial melting of garnet peridotite. Comprehensive analysis shows that the geochemical characteristics of the Qiaoqi intrusion bear resemblance to those of the Emeishan basalts, which are attributed to volumetrically minor melting of the fossil Emeishan plume head beneath the Yangtze crust following the eruption of the Emeishan Large Igneous Province (ELIP). This understanding further constrains the duration of the Emeishan Large Igneous Province and provides new support for understanding the formation, evolution and distribution of the Emeishan Large Igneous Province.
Karst collapse columns (KCCs) represent key concealed hazard-inducing factors that threaten the safety of coal mines in North China. To clarify their primary controlling geological factors and evolutionary processes, this study focuses on KCCs in the Hancheng Mining Area, situated on the southeastern margin of the Ordos Basin, China. A comprehensive methodological approach—integrating field surveys, petrographic and mineralogical identification, geochemical analysis, and structural interpretation—was employed to investigate the dominant factors controlling KCC development and their evolutionary mechanisms. The results indicate the following: (1) Thick-bedded dolomites of the 5th Member of the Majiagou Formation (Middle Ordovician Series) serve as the material foundation for karstification. These dolomites were deposited in an oxidized shallow-water tidal flat setting, which endowed them with favorable lithological properties for subsequent dissolution. (2) NE-SW trending erosional grooves within the paleogeomorphology of the Ordovician top surface functioned as preferential flow paths for karst water, channeling fluid movement and intensifying localized dissolution. (3) Multi-phase tectonic activities, particularly extensional deformation during the Himalayan orogeny, created the necessary stress conditions to trigger cave collapse. (4) KCCs undergo a multi-stage formation and evolution process: Starting with the Majiagou Formation’s 5th Member dolomites as the primary lithology, initial modification occurred via Caledonian weathering–crust karstification. Subsequently, compressional tectonism during the Yanshanian orogeny generated void spaces that facilitated deep-seated dissolution. Rapid uplift in the Paleogene exacerbated vertical dissolution, leading to extensive cavity development, which ultimately collapsed under the extensional tectonic regime of the Neogene. This study provides theoretical support for predicting and mitigating sudden water inrushes caused by KCCs in the Hancheng Mining Area. Furthermore, it offers novel insights and a scientific basis for advancing understanding of the developmental mechanisms of North China-type KCCs.