Geoscience frontiers最新文献

Landslide susceptibility assessment is crucial in predicting landslide occurrence and potential risks. However, traditional methods usually emphasize on larger regions of landsliding and rely on relatively static environmental conditions, which exposes the hysteresis of landslide susceptibility assessment in refined-scale and temporal dynamic changes. This study presents an improved landslide susceptibility assessment approach by integrating machine learning models based on random forest (RF), logical regression (LR), and gradient boosting decision tree (GBDT) with interferometric synthetic aperture radar (InSAR) technology and comparing them to their respective original models. The results demonstrated that the combined approach improves prediction accuracy and reduces the false negative and false positive errors. The LR-InSAR model showed the best performance in dynamic landslide susceptibility assessment at both regional and smaller scale, particularly when identifying areas of high and very high susceptibility. Modeling results were verified using data from field investigations including unmanned aerial vehicle (UAV) flights. This study is of great significance to accurately assess dynamic landslide susceptibility and to help reduce and prevent landslide risk.

High arsenic (As) groundwater is a global problem primarily originating from As-enriched sediments. The provenance (source) and release mechanisms (sinks) of high As sediment have been identified, but the source-sink transfer is poorly understood, especially the influence of geological and surface processes. In this study, we explore the roles of tectonic movement and Yellow River evolution in provenance formation processes and evaluate the combined effects of provenance and sediment age on the As content of aquifer sediments in the northern Hetao Basin of Inner Mongolia. Based on optically stimulated luminescence (OSL) and ¹⁴C dating and detrital zircon U-Pb, As content, and lithological analyses of a 400 m core, we reconstructed As changes over the last 160 ka. Our results show clay deposited in a paleo-lake during the Gonghe movement period in the late Pleistocene (∼100 ka B.P.) is enriched in As (31.8 μg/g) due to significant provenance contributions of the As-bearing Langshan Group under tectonic uplift and mountain erosion. In contrast, clay deposited in the middle Pleistocene (∼160 ka B.P.) has lower As content (7.3 μg/g) due to the Yellow River as the primary provenance. Accordingly, the provenance of basin As forced by tectonic uplift and Yellow River evolution determines the background As of aquifer sediments. After deposition, sediment As content decays over time, with higher decay rates in coarse-grained sands than fine-grained. Overall, both provenance formation and sediment age, representing initial and dynamic states of solid phase As, jointly determine the As content of aquifer sediments. More solid phase As provided by younger sediments from the proximal orogenic provenance and reducing conditions due to frequent river–lake transitions, jointly lead to higher As concentrations in shallow groundwater. The study highlights the potential for using a combined analysis of the tectonic movement-surface processes-environment system to improve understanding of geogenic high As groundwater over global large sedimentary basins in the proximity of young orogenic belts.

Landslide inventory is an indispensable output variable of landslide susceptibility prediction (LSP) modelling. However, the influence of landslide inventory incompleteness on LSP and the transfer rules of LSP resulting error in the model have not been explored. Adopting Xunwu County, China, as an example, the existing landslide inventory is first obtained and assumed to contain all landslide inventory samples under ideal conditions, after which different landslide inventory sample missing conditions are simulated by random sampling. It includes the condition that the landslide inventory samples in the whole study area are missing randomly at the proportions of 10%, 20%, 30%, 40% and 50%, as well as the condition that the landslide inventory samples in the south of Xunwu County are missing in aggregation. Then, five machine learning models, namely, Random Forest (RF), and Support Vector Machine (SVM), are used to perform LSP. Finally, the LSP results are evaluated to analyze the LSP uncertainties under various conditions. In addition, this study introduces various interpretability methods of machine learning model to explore the changes in the decision basis of the RF model under various conditions. Results show that (1) randomly missing landslide inventory samples at certain proportions (10%–50%) may affect the LSP results for local areas. (2) Aggregation of missing landslide inventory samples may cause significant biases in LSP, particularly in areas where samples are missing. (3) When 50% of landslide samples are missing (either randomly or aggregated), the changes in the decision basis of the RF model are mainly manifested in two aspects: first, the importance ranking of environmental factors slightly differs; second, in regard to LSP modelling in the same test grid unit, the weights of individual model factors may drastically vary.

The Huoshaoyun deposit in the Karakorum area of NW China is the world’s largest zinc-lead carbonate ore deposit. Here we investigate the genesis of the mineralization based on multiproxy investigations. The deposit contains zinc-lead carbonate and sulfide minerals, with smithsonite (Smt), cerussite (Cer), and sulfides accounting for 85%, 10%, and 5% of the total lead and zinc resources, respectively. Three ore-forming stages, involving Smt, Cer, and sulfides ores were summarized. The Smt mineralization is characterized by veined, massive, and stratified Smt forming horizontal sedimentary layered ore and vertical feeder veins similar to the SEDEX-type deposits. The sulfide and Cer veins display typical hydrothermal characteristics and are superimposed on the massive Smt ores. The Smt ores show high Li, Be, Cr, Y, Ba, Nd, Yb, and Zr contents, whereas the Cer veins have extremely high Sr contents (up to 3814–9174 ppm) and low Zr contents (less than 0.01 ppm). Galena and sphalerite show higher Cd concentrations compared to Smt and Cer ores.

The Smt ores differ with different spatial locations, with Smt ores formed at the vent have δ⁶⁶Zn values of +0.15‰ to +0.21‰, the massive Smt formed close to the vent show a value of +0.13‰, and those formed away from the vent show a value of 0.05‰, all values being close to 0. The sulfides have δ⁶⁶Zn values of −0.09‰ to +0.04‰. The C-O isotopes of Smt ores are similar to both altered and unaltered host limestone, suggesting that the limestone was a potential source for carbon and oxygen. Quartz with veined Smt shows magmatic signatures with δ¹⁸O_VSMOW of +1.14‰ to +2.23‰, high Pb (115–401 ppm) and Zn concentrations (390–997 ppm), whereas quartz associated with sulfide has meteoric fluid signature with the lowest δ¹⁸O_VSMOW (−14‰ to −10.7‰), low Pb (11.6–49.0 ppm) and Zn (18.1–72.8 ppm) concentrations. The temperature of equilibration computed based on oxygen isotope fractionation between Smt and coeval quartz indicate a dual source with that of quartz derived from an aqueous fluid, whereas the source for Smt might involve CO₂ or HCO₃⁻.

We trace multiple metallogenic stages for this deposit including exhalation, hydrothermal deposition, and fault-controlled sulfide vein formation. The largest orebody (III-1) preserves a 16 Mt reserve of Zn and was formed by crust-mantle interaction at ca. 195 Ma in the early development of the Linjitang post-arc rift system. Fluid convection, zinc enrichment driven by granitic magma, volcanic activity, and karst alteration induced by acid rain in a lagoonal environment promoted ore enrichment.

Joints shear strength is a critical parameter during the design and construction of geotechnical engineering structures. The prevailing models mostly adopt the form of empirical functions, employing mathematical regression techniques to represent experimental data. As an alternative approach, this paper proposes a new integrated intelligent computing paradigm that aims to predict joints shear strength. Five metaheuristic optimization algorithms, including the chameleon swarm algorithm (CSA), slime mold algorithm, transient search optimization algorithm, equilibrium optimizer and social network search algorithm, were employed to enhance the performance of the multilayered perception (MLP) model. Efficiency comparisons were conducted between the proposed CSA-MLP model and twelve classical models, employing statistical indicators such as root mean square error (RMSE), correlation coefficient (R²), mean absolute error (MAE), and variance accounted for (VAF) to evaluate the performance of each model. The sensitivity analysis of parameters that impact joints shear strength was conducted. Finally, the feasibility and limitations of this study were discussed. The results revealed that, in comparison to other models, the CSA-MLP model exhibited the most appropriate performance in terms of R² (0.88), RMSE (0.19), MAE (0.15), and VAF (90.32%) values. The result of sensitivity analysis showed that the normal stress and the joint roughness coefficient were the most critical factors influencing joints shear strength. This paper presented an efficacious attempt toward swift prediction of joints shear strength, thus avoiding the need for costly in-site and laboratory tests.

Constraining the processes associated with the formation of new (juvenile) continental crust from mantle-derived (basaltic) sources is key to understanding the origin and evolution of Earth’s landmasses. Here we present high-precision measurements of stable isotopes of potassium (K) from Earth’s most voluminous plagiogranites, exposed near El-Shadli in the Eastern Desert of Egypt. These plagiogranites exhibit a wide range of δ⁴¹K values (–0.31‰ ± 0.06‰ to 0.36‰ ± 0.05‰; 2 SE, standard error) that are significantly higher (isotopically heavier) than mantle values (–0.42‰ ± 0.08‰). Isotopic (⁸⁷Sr/⁸⁶Sr and ¹⁴³Nd/¹⁴⁴Nd) and trace element data indicate that the large variation in δ⁴¹K was inherited from the basaltic source rocks of the El-Shadli plagiogranites, consistent with an origin through partial melting of hydrothermally-altered mid-to-lower oceanic crust. These data demonstrate that K isotopes have the potential to better constrain the source of granitoid rocks and thus the secular evolution of the continental crust.

Biogenic carbonate structures such as rhodoliths and foraminiferal-algal nodules are a significant part of marine carbonate production and are being increasingly used as paleoenvironmental indicators for predictive modeling of the global carbon cycle and ocean acidification research. However, traditional methods to characterize and quantify the carbonate production of biogenic nodules are typically limited to two-dimensional analysis using optical and electron microscopy. While micro-computed tomography (µCT) is an excellent tool for 3D analysis of inner structures of geomaterials, the trade-off between sample size and image resolution is often a limiting factor. In this study, we address these challenges by using a novel multi-scale µCT image analysis methodology combined with electron microscopy, to visualize and quantify the carbonate volumes in a biogenic calcareous nodule. We applied our methodology to a foraminiferal-algal nodule collected from the Red Sea along the coast of NEOM, Saudi Arabia. Integrated µCT and SEM image analyses revealed the main biogenic carbonate components of this nodule to be encrusting foraminifera (EF) and crustose coralline algae (CCA). We developed a multi-scale µCT analysis approach for this study, involving a hybrid thresholding and machine-learning based image segmentation. We utilized a high resolution µCT scan from the sample as a ground-truth to improve the segmentation of the lower resolution full volume µCT scan which provided reliable volumetric quantification of the EF and CCA layers. Together, the EF and CCA layers contribute to approximately 65.5 % of the studied FAN volume, corresponding to 69.01 cm³ and 73.32 cm³ respectively, and the rest is comprised of sediment infill, voids and other minor components. Moreover, volumetric quantification results in conjunction with CT density values, indicate that the CCA layers are associated with the highest amount of carbonate production within this foraminiferal-algal nodule. The methodology developed for this study is suitable for analyzing biogenic carbonate structures for a wide array of applications including quantification of carbonate production and studying the impact of ocean acidification on skeletal structures of marine calcifying organisms. In particular, the hybrid µCT image analysis we adopted in this study proved to be advantageous for the analysis of biogenic structures in which the textures and components of the internal layers are distinctly visible despite having an overlap in the range of CT density values.

This study aims to investigate the combined use of multi-sensor datasets (Landsat 4–5 & 8 OLI satellite imagery, spatial resolution = 30 m) coupled with field studies to evaluate spatio-temporal dynamics of soil salinization along the coastal belt in West Bengal, India. This study assesses soil salinization by mapping the salinity and electrical conductivity of saturation extract (EC_e) and utilizing spectral signatures for estimating soil salinity. The SI change (%) was analyzed (2021–1995), categorizing increases in salinity levels into 5%, 10%, and 50% changes possibly due to salt encrustation on the soil layers. The land use land cover (LULC) change map (2021–1995) demonstrates that the study area is continuously evolving in terms of urbanization. Moreover, in the study area, soil salinity ranges from 0.03 ppt to 3.87 ppt, and EC_e varies from 0.35 dSm⁻¹ to 52.85 dSm⁻¹. Additionally, vulnerable saline soil locations were further identified. Classification of soil salinity based on EC_e reveals that 26% of samples fall into the non-saline category, while the rest belong to the saline category. The Spectral signatures of the soil samples (n = 19) acquired from FieldSpec hand spectrometer show significant absorption features around 1400, 1900, and 2250 nm and indicate salt minerals. The results of reflectance spectroscopy were cross-validated using X-ray fluorescence and scanning electron microscopy. This study also employed partial least square regression (PLSR) approach to predict EC_e (r² = 0.79, RMSE = 3.29) and salinity parameters (r² = 0.75, RMSE = 0.51), suggesting PLSR applicability in monitoring salt-affected soils globally. This study’s conclusion emphasizes that remote sensing data and multivariate analysis can be crucial tools for mapping spatial variations and predicting soil salinity. It has also been concluded that saline groundwater used for irrigation and aqua-cultural activities exacerbates soil salinization. The study will help policymakers/farmers identify the salt degradation problem more effectively and adopt immediate mitigation measures.

The occurrence of massif-type anorthosite intrusions is a widespread Proterozoic phenomenon. They are usually associated with gabbroic, charnockitic, and granitic rocks, comprising the so-called anorthosite-mangerite-charnockite-granite (AMCG) suite. Although these rocks have been extensively studied worldwide, several aspects concerning their formation remain unsettled. Among them, the magma source and the tectonic setting are the most important. To evaluate these issues, we first compiled geochemical and isotopic data of Proterozoic anorthosite massifs and AMCG suites worldwide and stored it in a database named datAMCG. This plethora of data allows us to make some important interpretations. We argue that the wide-ranging multi-isotopic composition of this group of rocks reflects varying proportions of juvenile mantle-derived melts and crustal components. We interpret that the precursor magmas of most massive anorthosite bodies and associated mafic rocks have a mantle-dominated origin. However, we highlight that a crustal component is indispensable to generate these lithologies. Adding variable amounts of this material during succeeding multi-stage assimilation-fractional crystallization (AFC) processes gives these intrusions their typical mantle-crustal hybrid isotopic traits. In contrast, a crustal-dominant origin with a complementary mantle component is interpreted for most MCG rocks. In summary, the isotopic information in datAMCG indicates that both sources are necessary to generate AMCG rocks. Therefore, we suggest that hybridized magmas with different mantle-crust proportions originate these rocks. This interpretation might offer a more nuanced and accurate depiction of this phenomenon in future work instead of choosing a single-sourced model as in the past decades. Finally, tectonomagmatic diagrams suggest that the rocks under study were likely generated in a tectonic environment that transitioned between collision and post-collisional extension, sometimes involving subduction-modified mantle sources. This interpretation is supported by geological and geochronological information from most complexes, thus challenging the Andean-type margins as an ideal tectonic setting.

The unique ore-forming processes and the key factors responsible for formation of skarn deposits are still obscure, and challenges exist in the determination of timing of Pb-Zn skarns owing to lacking suitable mineral chronometers. Here we present detailed paragenesis, bulk geochemistry, in situ U-Pb dating of zircon and garnet, and garnet oxygen isotopes together with in situ zircon Hf-O isotopes from the newly discovered Aqishan Pb-Zn deposit in the southern Central Asian Orogenic Belt (CAOB), northwest China. This comprehensive data set revealed a Late Carboniferous subduction-related distal Pb-Zn skarn system associated with the granitic magmatism. Pre-ore stage garnets are generally subhedral to euhedral with oscillatory zoning and show slightly fractionated rare earth element patterns with positive Eu anomalies that point to an infiltration metasomatism origin under high water/rock ratios. The syn-ore stage sphalerite is typically enriched in Mn and Cd and has moderate Zn/Cd ratios (337–482), with a formation temperature of 265 °C to 383 °C, which indicate magmatic-hydrothermal signatures. The isocons defined by P₂O₅ decipher that the principal factors for skarn formation were elevated activities of Fe, Ca, and Si species, where remobilization of Pb metals, meanwhile, contributed to ore-forming budgets to mineralizing fluids. SIMS U-Pb dating of zircons from granite porphyry that occurs distal to the skarns and Pb-Zn orebodies shows that these intrusions emplaced at ca. 311.3–310.6 Ma, recording the subduction of the Paleo-Tianshan oceanic plate. Hydrothermal garnets in close textural association with Pb-Zn sulfides yield indistinguishable in situ LA-ICP-MS U-Pb ages of 310.5 ± 4.1 Ma. Whole-rock geochemistry and in situ zircon Hf-O isotopes (δ¹⁸O = 4.6‰–6.0‰) indicate that the granite porphyry was derived from partial melting of juvenile crust and influenced by subducted oceanic crust. Oxygen isotope compositions of garnets (δ¹⁸O = 8.0‰–9.0‰) demonstrate that the equilibrated ore fluids were inherited from fluid-rock interactions between a primary magmatic water and host tuff rocks. Our study highlights the application of garnets as a potential robust U-Pb geochronometer and isotopic tracer of ore fluids in skarn mineralizing systems in subduction-related arc environments.