Search Results (164)

Karst aquifers present unique challenges for groundwater level prediction due to their dual-porosity structures and highly nonlinear hydrological responses. This study systematically evaluates nine machine learning and deep learning models (RF, XGBoost, LSTM, CNN, Transformer, N-BEATS, CNN-LSTM, Seq2Seq-LSTM, and Attention-Seq2Seq-LSTM) for rainfall-driven groundwater level forecasting in the Maocun subterranean river catchment, Guilin, Guangxi, China. Two years of hourly high-frequency data from three monitoring sites representing distinct hydrogeological zones (recharge, flow, and discharge) were employed within a multidimensional evaluation framework integrating single-step accuracy, multi-step stability, and computational efficiency. Results indicate that the Transformer achieved the highest single-step prediction accuracy, attaining the lowest RMSE (0.130–0.606 m) and highest R² (0.813–0.965) across all three sites. CNN-LSTM offered the best balance between predictive performance and computational cost, requiring an average training time of only 27.97 s and 28.0 convergence epochs. N-BEATS demonstrated superior long-term stability in 12-steps-ahead forecasting, achieving R² = 0.914 at ZK1, outperforming all other architectures. More fundamentally, hydrogeological complexity exerted a dominant control on predictive skill that systematically outweighed differences arising from model architecture. All models yielded R² below 0.813 at the geologically complex ZK2 site, whereas R² exceeded 0.950 across all models at ZK1, indicating that aquifer complexity, rather than algorithm selection, constitutes the primary constraint on prediction feasibility. This study presents the first application of N-BEATS to karst groundwater level forecasting and proposes a replicable multidimensional evaluation framework, providing a scientific reference for intelligent modelling of complex karst systems. Full article

Caves are fragile subterranean ecosystems whose conservation depends on accurate microclimatic zonation. Traditional fixed-distance sampling often overlooks non-linear thermodynamic transitions at geomorphological thresholds, hindering sustainable management of subterranean biodiversity. This study introduces the Adaptive Morpho-Statistical Protocol (AMSP), a novel, resource-efficient framework for functional cave profiling. The methodology integrates high-precision atmospheric monitoring with adaptive spatial positioning to identify three distinct sectors (S1–S3) based on thermodynamic homeostasis rather than linear distance. Validated across five diverse cave archetypes in the Vratsa Karst Region (Bulgaria), the AMSP demonstrated exceptional predictive power using second-order polynomial regressions (R² > 0.92). A key finding is the definition of a standardized reference threshold for deep-reach stability (Sector 3), consistently characterized by a Dew Point Standard Deviation (SD_DP < 0.40) and stabilized thermal coupling (∆T → 0). Furthermore, the adaptive strategy successfully captured extreme hygrometric jumps at morphological bottlenecks—critical inflection points for protecting sensitive biota. By providing a cost-effective and replicable standard, the AMSP bridges the gap between spatial resolution and logistical feasibility in challenging environments. These results confirm that morphological isolation is the primary driver of microclimatic inertia, offering a robust tool for sustainable subterranean heritage management and high-precision ecological monitoring in protected karst landscapes. Full article

In karst mountainous areas where high-dimensional features coexist with extremely limited sample sizes, accurate landslide susceptibility mapping remains challenging. To address this issue, we propose an ensemble framework termed the Triple-Source Probabilistic Fusion Bidirectional Long Short-Term Memory network (TSPF-BiLSTM). The approach was tested in Sangzhi County, Hunan Province, by integrating three base learners—Random Forest (RF), LightGBM, and AdaBoost. Their raw outputs were first calibrated using five-fold Platt scaling to generate posterior probabilities on a unified scale. A bidirectional LSTM was then employed to perform deep nonlinear fusion of these cross-model probability features. Using a total of 618 landslide and 618 non-landslide samples (split into training and testing sets), the TSPF-BiLSTM model achieved a mean AUC of 0.9525 (±0.0115) under ten-fold cross-validation, outperforming not only the individual base learners but also standalone deep learning models (CNN and Transformer). The frequency ratio in the very high susceptibility zone reached 3.97, significantly exceeding all benchmark models and confirming its superior capability in high-risk area identification. Multi-model importance analysis identified NDVI, elevation, and annual rainfall as the dominant regional landslide predisposing factors. Within the specific ranges of NDVI 0–0.686, elevation 155–462 m, and annual rainfall 1273.6–1301 mm, landslide frequency ratios consistently exceeded 1.96. The proposed framework, with its probability-level fusion and embedded regularization mechanisms, effectively mitigated overfitting despite the small sample size, providing a robust technical solution for geological hazard risk identification and prevention in the data-scarce karst terrain of the Wuling Mountains. Full article

In karstic environments, water supply wells are vulnerable to rapid sediment transfer during intense rainfall events, often generating turbidity peaks that disrupt water-treatment operations. In Normandy (France), the high density of sinkholes and the complexity of transport processes in karsts complicate the identification and prioritization of sinkholes requiring mitigation to reduce sediment fluxes at water supply wells. This study aims to quantify the time-lagged impact of each sinkhole on turbidity peaks at a supply well using a cascade modeling approach that couples numerical surface erosion–runoff simulations with deep learning models representing hydrosedimentary responses through the karst network. Surface erosion–runoff was simulated using WaterSed. Hydroclimatic time series and WaterSed model outputs were used as inputs for our deep learning models. Several deep learning architectures were compared and optimized across multiple rounds to identify a best-performing model, which was then interpreted using interpretability methods. Interpretability analyses show that turbidity is primarily controlled by seasonal conditions and short-term rainfall accumulation, while multiple sinkholes contribute jointly to short time lags. Temporal attributions reveal rapid karst response followed by attenuation, consistent with reactive karst behavior. The contribution of each sinkhole to turbidity peaks allows us to identify the most important sinkholes requiring mitigation by stakeholders. Full article

Subtropical forest ecosystems, especially evergreen broad-leaved forests in the East Asian monsoon region, are a crucial component of the global terrestrial carbon cycle and make a key contribution to the “missing carbon sequestration” in the Northern Hemisphere. This review systematically integrates recent research progress on the carbon pool patterns, aboveground-subsurface biogeochemical processes, and global change responses of subtropical forests, summarizing the potential mechanisms of their sustainable carbon sequestration capacity and identifying current cognitive bottlenecks. Studies have shown that subtropical mature forests have carbon sequestration potential that exceeds traditional theoretical expectations, but there are still significant shortcomings in terms of carbon stability in deep soil (>1 m), quantitative constraints on rhizosphere activating effects, and assessment of ecosystem resilience under extreme climate events. Furthermore, the nonlinear interactions between factors such as climate warming, precipitation changes, and nitrogen deposition may trigger a critical turning point in carbon sink functions, and the water-carbon-geological coupling processes in special habitats such as karst and mangrove forests are often underestimated. We further propose that future research should focus on developing coupled models of “plant–soil–microbe hydrology”, combining molecular and isotopic techniques to elucidate microbial carbon pump mechanisms and strengthening long-term in situ experiments under combined extreme events to provide scientific support for subtropical forest carbon sink management and prediction. Full article

Regional curtain grouting for water interception serves as a critical technique for achieving safe and efficient mining under complex hydrogeological conditions in deep mine deposits. This study focuses on the Cuihongshan Iron-Polymetallic Mine, where repeated incidents of water inrush and sand outbursts have occurred due to complex hydrogeological conditions. By identifying the water-conducting pathways and characterizing the spatial distribution of relative aquitards within the mining area, a precise hydrogeological model was established. On this basis, the engineering application and performance evaluation of horizontal curtain grouting were systematically investigated. Through field monitoring and multi-method joint analysis, the water-blocking effectiveness of the grouting technique was comprehensively assessed. The results demonstrate a significant sequential reduction in grout take per meter for primary, secondary, and tertiary grouting holes, indicating a clear cumulative grouting effect. The grout effectively filled karst fractures, forming a continuous and stable water-resisting curtain. The project successfully severed the hydraulic connection between the highly water-rich Quaternary aquifer and the mine workings, substantially reducing mine water inflow. This study provides important theoretical support and practical reference for water hazard control in similar deep metal mines. Full article

The 1st International Online Conference on Taxonomy (IOCTX2025) serves as a critical interdisciplinary nexus for addressing the contemporary “taxonomic impediment” through the lens of integrative systematics and computational innovation. By synthesizing research spanning from Paleozoic fossil records to extant microbial biodiversity, the conference illuminates the evolving methodology of species delimitation, moving beyond traditional morphometrics to incorporate multi-locus molecular phylogenetics, bioacoustics, and high-resolution 3D imaging. Key thematic clusters across the program examine the floristic complexity of Karst landscapes, the resolution of cryptic animal species complexes through genomic and proteomic data, and the role of machine learning in automating the identification of both fossil and living taxa. Furthermore, the proceedings underscore a paradigm shift toward “integrative taxonomy,” where the fusion of morphological rigor with eDNA metabarcoding and automated genomic scanning provides a more robust framework for understanding global biodiversity hotspots. Ultimately, IOCTX2025 reaffirms taxonomy as a high-technology discipline essential for conservation biology and evolutionary theory, providing a standardized scientific language to describe the complexities of the tree of life across deep time and modern ecosystems. Full article

Wind disturbance is the major driver of forest damage in Northern Europe, particularly during late autumn and winter when cyclonic activity might coincide with unfrozen soil conditions. We quantified the thermal regime of periodically waterlogged mineral soils in relation to snow cover thickness (SCT) in hemiboreal forests of Latvia. The study was conducted in 15 forest stands dominated by birch (Betula spp.), Scots pine (Pinus sylvestris L.), and Norway spruce (Picea abies (L.) H. Karst.) during two contrasting winters (2023/2024 and 2024/2025) across two regions differing in local climatic conditions. Soil temperature was monitored at 0, 10, and 20 cm depths, while SCT was measured at five points per plot. Linear mixed-effects models were used to assess the effects of air temperature, precipitation, region, season, and species composition to snow cover thickness (SCT) and effect of the other parameters to soil temperatures. SCT varied strongly between regions and seasons. Snow accumulation was lower in pine- and spruce-dominated stands compared to birch stands. Formation of snow layer increased soil temperatures at the surface, whereas SCT had a more pronounced insulating effect at depths of 10–20 cm, especially during prolonged snow cover (F = 15.43 − 54.25, p < 0.001). Heat transfer from deeper layers further enhanced thawing under waterlogged conditions. Snow cover significantly insulates soil in a depth-dependent manner, with its magnitude varying across regions and seasons. Promoting mixed-species stands and selecting deep-rooted species, such as birch, can enhance the formation of frozen soil, and thus soil–root anchorage, reducing wind damage risk on periodically waterlogged soils. Full article

A comprehensive study on the distribution characteristics and exploitation strategies of remaining oil was carried out in the Ordovician ultra-deep fault-controlled fractured–vuggy carbonate reservoir within the Shunbei No. 1 strike–slip fault zone. This research addresses challenges such as severe watered-out and gas channeling encountered during multi-stage development, marking a shift toward a development phase focused on residual oil recovery. By integrating seismic attributes, drilling, logging, and production performance data—and building upon previous methodologies of “hierarchical constraint and genetic modeling”—a three-dimensional geological model was constructed with a five-tiered architecture: strike–slip fault affected zone, fault-controlled unit, cave-like structure, cluster fillings, and fracture zone. Numerical simulations were subsequently performed based on this model. The results demonstrate that the distribution of remaining oil is dominantly controlled by the coupling between key geological factors—including fault kinematics, reservoir architecture formed by karst evolution, and fracture–vug connectivity—and the injection–production well pattern. Three major categories with five sub-types of residual oil distribution patterns were identified: (1) local low permeability, weak hydrodynamics; (2) shielded connectivity pathways; and (3) Well Pattern-Dependent. Accordingly, two types of potential-tapping measures are proposed: improve well control through optimized well placement and sidetrack drilling and reservoir flow field modification via adjusted injection–production parameters and sealing of high-permeability channels. Techniques such as gas (nitrogen) huff-and-puff, gravity-assisted segregation, and injection–production pattern restructuring are recommended to improve reserve control and sweep efficiency, thereby increasing ultimate recovery. This study provides valuable guidance for the efficient development of similar ultra-deep fractured–vuggy carbonate reservoirs. Full article

Reliable forecasting of karst spring discharge is critical for sustainable groundwater resource management under the dual pressures of climate change and intensified anthropogenic activities. This study proposes a Heterogeneous Spatiotemporal Graph Attention Network (H-STGAT) to predict spring discharge dynamics at Shentou Spring, Shanxi Province, China. Unlike conventional spatiotemporal networks that treat all relationships uniformly, our model derives its heterogeneity from a graph structure that explicitly categorizes spatial, temporal, and periodic dependencies as unique edge classes. Specifically, a dual-layer attention mechanism is designed to independently extract hydrological features within each relational channel while dynamically assigning importance weights to fuse these multi-source dependencies. This architecture enables the adaptive capture of spatial heterogeneity, temporal dependencies, and multi-year periodic patterns in karst hydrological processes. Results demonstrate that H-STGAT outperforms both traditional statistical and deep learning models in predictive accuracy, achieving an RMSE of 0.22 m³/s and an NSE of 0.77. The model reveals a long-distance recharge pattern dominated by high-altitude regions, a finding validated by independent isotopic evidence, and accurately identifies an approximately 4–6 month lag between precipitation and spring discharge, which is consistent with the characteristic hydrological lag identified through statistical cross-covariance analysis. This research enhances the understanding of complex mechanisms in karst hydrological systems and provides a robust predictive tool for sustainable groundwater management and ecological conservation, while offering a generalizable methodological framework for similar complex karst hydrological systems. Full article

The moldic pore-vuggy reservoirs of the Ma5₄-Ma5₁ sub-member in the Majiagou Formation, central Ordos Basin, are key targets for deep natural gas exploration, yet the alteration mechanisms and controlling factors of burial-stage pressure-released water karstification remain unclear. Herein, an integrated methodology encompassing core observation, thin-section analysis, and geochemical testing was adopted to systematically clarify the development characteristics and multi-factor coupling control mechanisms of this karst process. Results show that burial-stage pressure-released water karst is dominated by overprinting on pre-existing syndepositional and supergene pore networks, forming complex reservoir spaces via synergistic selective dissolution. The development of preferential dissolution zones is jointly controlled by differential compaction of the weathering crust, permeability heterogeneity of the overlying strata and weathered crust, and diagenetic fluid properties. After the supergene diagenetic stage, differential tectonic deformation and burial compaction induced overpressure in pore fluids, which drove acidic pressure-released water to migrate along high-permeability pathways such as the “sandstone windows” overlying the Ordovician weathering crust. These fluids preferentially dissolved high-permeability moldic pore-vuggy dolomites in paleo-karst platforms and steep slope zones, whereas tight micritic dolomites served as effective barriers. The acidic environment sustained by organic acids and H₂S in pressure-released water promoted carbonate dissolution, and carbon-oxygen isotopes as well as pyrite δ³⁴S values verify that the fluids were derived from mudstone compaction. This study reveals that the distribution of high-quality reservoirs is jointly determined by the synergistic preservation of moldic pore-vuggy systems in paleo-karst platforms and steep slopes and directional alteration of pressure-released water along preferential pathways, providing crucial geological guidance for the evaluation of deep carbonate reservoirs. Full article

Landslides are among the most prevalent geological hazards worldwide, posing severe threats to public safety due to their sudden onset and destructive potential. The rapid and accurate automated segmentation of landslide areas is a critical task for enhancing capabilities in disaster risk assessment, emergency response, and post-disaster management. However, existing deep learning models for landslide segmentation predominantly rely on unimodal remote sensing imagery. In complex Karst landscapes characterized by dense vegetation and severe shadow interference, the optical features of landslides are difficult to extract effectively, thereby significantly limiting recognition accuracy. Therefore, synergistically utilizing multimodal data while mitigating information redundancy and noise interference has emerged as a core challenge in this field. To address this challenge, this paper proposes a Triple-Stream Guided Enhancement and Fusion Network (TriGEFNet), designed to efficiently fuse three data sources: RGB imagery, Vegetation Indices (VI), and Slope. The model incorporates an adaptive guidance mechanism within the encoder. This mechanism leverages the terrain constraints provided by slope to compensate for the information loss within optical imagery under shadowing conditions. Simultaneously, it integrates the sensitivity of VIs to surface destruction to collectively calibrate and enhance RGB features, thereby extracting fused features that are highly responsive to landslides. Subsequently, gated skip connections in the decoder refine these features, ensuring the optimal combination of deep semantic information with critical boundary details, thus achieving deep synergy among multimodal features. A systematic performance evaluation of the proposed model was conducted on the self-constructed Zunyi dataset and two publicly available datasets. Experimental results demonstrate that TriGEFNet achieved mean Intersection over Union (mIoU) scores of 86.27% on the Zunyi dataset, 80.26% on the L4S dataset, and 89.53% on the Bijie dataset, respectively. Compared to the multimodal baseline model, TriGEFNet achieved significant improvements, with maximum gains of 7.68% in Recall and 4.37% in F1-score across the three datasets. This study not only presents a novel and effective paradigm for multimodal remote sensing data fusion but also provides a forward-looking solution for constructing more robust and precise intelligent systems for landslide monitoring and assessment. Full article

Chelmos Vouraikos UNESCO Global Geopark is located in North Peloponnesus, Greece. As a member of the Global Geoparks Network, it is valued for its rich geoheritage in combination with its natural and cultural wealth. Several different landforms of international value are located in the area. The scope of this work is to present an overview of its geomorphological features, link them with biodiversity and highlight their value for geotourism. Its geology is complicated due to intense tectonism. Three geotectonic units of the Alpine Orogeny can be found along with post-Alpine sediments related to the Corinth Gulf rifting. The area is highly covered by limestone creating important karst landforms. High peaks surround river valleys and deep gorges create breathtaking landscapes. Some of them cut through high and steep conglomerate slopes. Remnants of past glaciation have been preserved on Mt Chelmos. The exceptional geodiversity of the area is linked with rich vegetation and high endemism. The many identified geomorphological sites highlight the Geopark’s strong commitment to geomorphology and its importance as a key geomorphological destination. Highly visible geomorphological sites with ecological value can also promote environmental awareness and contribute to the protection of biodiversity. Full article

The dolomite of the fourth member of Dengying Formation in Gaoshiti-Moxi area of central Sichuan Basin is rich in hydrocarbon resources. It has experienced superimposition-reformation of multistage karstification, and is the key target for studying deep ancient carbonate reservoirs. Exploration and development practices show that there are great differences in the development of karst reservoirs of the fourth member of Dengying Formation between the platform margin and intraplatform in Gaoshiti-Moxi area. However, the differences in the genetic mechanism of karst reservoirs between these two zones are unclear. Therefore, based on an integrated analysis of core, thin section, drilling, logging, and geochemical test data, this study clarifies the differences in karstification between the platform margin and intraplatform and conducts a comparative analysis of the controlling factors for the differences in karst reservoirs. Results show that the fourth member of Dengying Formation experienced superimposition-reformation of four types of paleokarstification, including eogenetic meteoric water karst, supergene karst, coastal mixed water karst, and burial karst. Large-scale dissolved fractures and caves are mainly controlled by meteoric water karstification, primarily developing three types of reservoir space: vug type, fracture-vug type, and cave type. Dolomite and quartz fillings are mainly formed in the medium-deep burial period. Four types of paleokarstification are developed in the platform margin, while the coastal mixed water karst is not developed in the intraplatform. Eogenetic meteoric water karst and supergene karst in the platform margin are stronger than those in the intraplatform, while burial karst shows no notable difference between the two zones. The thickness of soluble rock (mound-shoal complex), karst paleogeomorphology, and different types of paleokarstification are the main controlling factors for the difference in karst reservoirs between the platform margin and the intraplatform. Full article

The Tarim Basin is currently the largest petroliferous basin in China, with hydrocarbons primarily hosted in Ordovician marine carbonate paleokarst fracture–vug reservoirs—a typical example being the Tahe Oilfield located in the northern structural uplift of the basin. The principle of “the present is the key to the past” serves as a core method for studying paleokarst fracture–vug reservoirs in the Tahe Oilfield. The deep and ultra-deep carbonate fracture–vug reservoirs in the Tahe Oilfield formed under humid tropical to subtropical paleoclimates during the Paleozoic Era, belonging to a humid tropical–subtropical paleoepikarst dynamic system. Modern karst types in China are diverse, providing abundant modern karst analogs for paleokarst research in the Tarim Basin. Carbonate regions in Eastern China can be divided into two major zones from north to south: the arid to semiarid north karst and the humid tropical–subtropical south karst. Karst in Northern China is characterized by large karst spring systems, with fissure–conduit networks as the primary aquifers; in contrast, karst in Southern China features underground river networks dominated by conduits and caves. From the perspective of karst hydrodynamic conditions, the paleokarst environment of deep fracture–vug reservoirs in the Tarim Basin exhibits high similarity to the modern karst environment in Southern China. The development patterns of karst underground rivers and caves in Southern China can be applied to comparative studies of carbonate fracture–vug reservoir structures in the Tarim Basin. Research on modern and paleokarst systems complements and advances each other, jointly promoting the development of karstology from different perspectives. Full article