Gorges Reservoir Area Of China Research Articles

Risk assessment of exposure to 12 kinds ofmycotoxins throughconsumption of Pericarpium Citri Reticulataecollected fromThree Gorges Reservoir Area of China

Risk assessment of exposure to 12 kinds ofmycotoxins throughconsumption of Pericarpium Citri Reticulataecollected fromThree Gorges Reservoir Area of China

Housing-industry transformation development and rural revitalization path based on complex adaptive system in Three Gorges Reservoir Area of China

The core of dual circulation in the Three Gorges Reservoir Area (TGRA) and the greening Yangtze River is sustainable development. Reservoir housing-industry coordinate sustainable is extremely important to ensuring a more sustainable TGRA. Here, we investigate the overall characteristics of housing-industry transformation development and rural revitalization path in the TGRA at microscopic perspective. Taking Wutu village (WTV), a historical and cultural rural in the TGRA, as case study area, we also analyze the evolution, stage characteristics, and trends of housing-industry system (HIS) using theoretical analysis and empirical research from the perspective of spatio-temporal-development and complex adaptability system model. The findings reveal that: (1) In the traditional housing-industry separation stage(1978–1998), the development of villagers' self-organization and the positive and correct leadership of the village party branch are the dominant forces for the evolution of HIS in WTV. The adaptive behavior of the main body and the strength of the housing-industry effect are limited, and the evolution of the HIS is in the accumulation stage of quantitative change. (2) In the stage of cooperative transformation of housing-industry (1999–2018), the positive and correct leadership of the village party branch is the dominant force of the evolution of HIS in WTV, and the government's policy support is the main force of the evolution of HIS in WTV. (3) In the stage of modern housing-industry integration (2019 to present), the interaction of multiple subjects is the dominant force in the evolution of HIS in WTV. Multiple entities further adapt to various changes in the external environment, and the system structure and level have taken a qualitative leap. Overall, the evolution of HIS in WTV is a dynamic and continuous process. Based on the evolution stage and driving mechanism of HIS, it is necessary to improve the self-organization and adaptive response and ability of the complex adaptive system in WTV, adapt to the general law of system evolution, and effectively regulate the transformation and upgrading of HIS in WTV. Our research outcomes offer valuable insights for fostering rural transformation, enhancing the housing-industry environment, and ultimately propelling the TGRA towards a more sustainable and prosperous future.

Evaluating the long-term effects of best management practices on pollution reduction and soil quality improvement in sloping farmland of the Three Gorges Reservoir area

Reducing nonpoint-source pollution and improving soil quality by applying best management practices (BMPs) on sloping farmland are essential for ensuring water environmental safety and sustainable agricultural development. However, little is known about the accumulative effect of the long-term application of BMPs on reducing pollution and improving soil quality in major land use types of sloping farmland in watersheds. We conducted a nine-year field experiment for citrus orchard and cropland of sloping farmland in the Three Gorges Reservoir area of China. The treatment settings for the citrus experimental plots were as follows: (1) citrus intercropped with white clover (WC), (2) citrus orchard soil mulched with straw (SM), (3) citrus intercropped with Hemerocallis flava contour hedgerows (HF) and (4) conventional citrus planting pattern (C-CK). The treatment settings for the crop experimental plots were as follows: (1) wheat-peanut rotation intercropped with Toona sinensis contour hedgerows (TS), (2) wheat-peanut rotation intercropped with alfalfa contour hedgerows (AF), (3) ryegrass-sesame rotation (RS) and (4) conventional wheat-peanut rotation (W-CK). We found that compared with C-CK and W-CK, (1) runoff, sediment, total nitrogen and total phosphorus losses were reduced by 18.0 %, 66.1 %, 35.6 % and 40.2 %, respectively, in WC; by 22.4 %, 61.2 %, 43.9 % and 52.4 %, respectively, in SM; by 25.4%, 69.5 %, 52.0 % and 58.1 %, respectively, in HF; by 45.1 %, 73.7 %, 64.6 % and 66.0 %, respectively, in TS; by 41.8 %, 67.4 %, 59.3 % and 60.8 %, respectively, in AF; and by 24.1 %, 42.9 %, 39.3 % and 40.3 %, respectively, in RS; (2) the soil quality index values of the BMP treatments increased; and (3) the application of these BMPs did not significantly affect yield or product value. We suggest the application of these BMPs on sloping farmland to (1) efficiently reduce total runoff, total sediment, and soil total nitrogen and total phosphorus losses; (2) improve soil quality; and (3) have no significant effect on output, of which the BMP HF applied to citrus orchard and the BMP TS applied to cropland are the most effective.

Divergent behaviour of soil nutrients imprinted by different land management practices in the Three Gorges Reservoir Area, China

Soil nutrients are essentially regulated by land management practices via modulating biotic element input and metabolism. The Three Gorges Reservoir Area in China was dominated by a farming landscape, but land management has become diversified over recent decades. How these restorative management practices may have affected soil nutrients is not completely understood. In this study, a space-time substitution approach was applied to evaluate soil nutrients and their stoichiometric changes in response to post-farming land management practices. Soil samples (0–10 cm, 10–20 cm, and 20–40 cm) were collected from present-day croplands, cypress plantations, eucalyptus plantations, abandoned croplands, and citrus plantations. Soil organic matter, soil organic carbon, total nitrogen, alkaline hydrolyzed nitrogen, total phosphorus, and available phosphorus were determined. The results showed that soil organic matter and total nitrogen in abandoned croplands, cypress plantations, eucalyptus plantations and citrus plantations were increased by 186% and 190%, 184% and 107%, 45% and 33%, and 45% and 54%, respectively, in comparison with present-day croplands. Soil nutrients except for total phosphorus decreased more with soil depth by exclusion of tillage mixing. The analysis of the comprehensive soil nutrient index showed that abandoned croplands (0.90) and cypress plantations (0.72) exhibited better nutrient recovery capacity. Soil C:P and N:P ratios increase in abandoned croplands, cypress plantations, and eucalyptus plantations. Phosphorus may become a limiting factor for plant growth with prolonged recovery in abandoned croplands, cypress plantations, and eucalyptus plantations, while soil organic matter and total nitrogen deficiencies are exacerbated in citrus plantations and present-day croplands. Therefore, cropland abandonment and reforestation (particularly cypress trees plantation) are recommended practices for restoring soil nutrients in the Three Gorges Reservoir Area.

Nitrogen addition enhances nitrogen but not carbon mineralization in aggregate size fractions of soils in a Pinus massonia plantation

IntroductionAtmospheric nitrogen (N) deposition can impact the levels of soil organic carbon (SOC) and total nitrogen (total N) by altering the soil N availability. However, the effect of N input on the mineralization of SOC and total N in various soil aggregate size fractions requires further clarification.MethodsThe soil samples were collected from a Pinus massoniana plantation situated in the Three Gorges Reservoir Area of China. Over a period of three years, the soils from the plantation were subjected to four different levels of nitrogen addition (0 [N0], 30 [N30], 60 [N60], and 90 [N90] kg N ha−1 yr−1). The impact of N addition on the mineralization of SOC and total N in aggregates was evaluated through an incubation experiment, encompassing four aggregate sizes (2000 − 8000, 1000 − 2000, 250 − 1000, and < 250 μm).ResultsThe < 250 μm fraction showed the highest levels of cumulative C mineralization, while the lowest levels were observed in the 2000 − 8000 μm fraction. Compared to the < 250 um fraction, a drop of 9 − 21% in cumulative C mineralization was observed in the 2000 − 8000 μm fraction, indicating that soil aggregates enhance the stability of C in the soil. Cumulative N mineralization levels were consistently at their lowest in the 2000 − 8000 μm fraction, indicating aggregates reducing mineralization-related N loss. Adding N to forest soil samples led to a reduction in cumulative C mineralization. In contrast, an opposite trend was observed in the cumulative N mineralization after adding N in microaggregates. Nitrification was the main contributor to net N mineralization. SOC and total levels increased in response to N30 and N60. N addition leads to an increase in the weight ratio of the 1000 − 2000 μm fraction. Moreover, N90 was linked to decreases in microbial biomass C and N.DiscussionThese findings confirm that the structural characteristics of soil aggregates play a crucial role in sequestering organic carbon and total N sequestration in the presence of N deposition, while highlighting N loss from the soil caused by N input.

Refined landslide susceptibility mapping in township area using ensemble machine learning method under dataset replenishment strategy

Landslide susceptibility mapping (LSM) warrants considerable attention as a prerequisite for risk assessment and prevention in response to the increasing global occurrences of landslides. However, refined LSM at the township scale still faces the problem of the limited number of landslides and the oversight of relevant factors associated with different landslide types. This work is dedicated to propose a novel dataset replenishment strategy designed to address landslide sample scarcity, and to identify dominant factors in high landslide susceptibility zones to refine LSM at township scale for decision makers. Herein, the Wuling County in the Three Gorges Reservoir area of China was taken as taken as the research case. Initially, field investigation and calibration of 21 landslides yielded accumulation and rock areas, and the corresponding impact factors were therefore proposed to construct input features for machine learning facilitated by factor diagnosis using Pearson correlation coefficient (PCC) and multicollinearity test. Subsequently, a training and validation dataset split of 8: 2 was randomly generated, enabling the application of ensemble machine learning model for LSM prediction. The study culminated in the feasibility of dataset replenishment strategy, which was employed to explore uncertainties and identify governing factors in LSM. The results indicate that the bagging-ANN model exhibits superior performance in both accumulation and rock areas, with Areas Under Curve (AUC) of 0.987 and 0.994. Furthermore, the dataset replenishment strategy emerges as instrumental in enhancing the reliability of LSM, as evidenced by an 8.97 % reduction in average value (AVG) and a 11.00 % increase in standard deviation (STD). The governing factors for landslides in accumulation and rock areas are designated as river erosion and height, respectively. Together with the excellent performance, this study is expected to provide a promising reference for LSM in other townships worldwide.

An integrated numerical model for landslide-generated waves: a case study of the Gongjiafang landslide, Three Gorges Reservoir, China

ABSTRACT Landslide is a frequent geological hazard in the Three Gorges reservoir area of China. Impulse waves could be created and cause substantial damage if the landslide collides with a water body. A nonlinear shallow water equation in combination with a progressive landslide model is proposed to simulate this catastrophic disaster. The proposed approach is distinguished by rapid calculation, less computational burden, and high practical value. By simulating the 2008 Three Gorges Reservoir Gongjiafang landslide and the subsequent impulse waves, we illustrated and validated this methodology. The initiation and motion processes of the landslide, as well as the generation, propagation, and run-up processes of impulse waves, were thoroughly performed. The simulation results were in good agreement with the field survey data, demonstrating the viability of the proposed approach for simulating the impulse waves triggered by landslide events. This numerical approach is beneficial for engineers to perform a preliminary estimation of landslide-generated waves and further facilitate the prevention and mitigation of geological hazards.

Monitoring the Area Change in the Three Gorges Reservoir Riparian Zone Based on Genetic Algorithm Optimized Machine Learning Algorithms and Sentinel-1 Data

Riparian zones play a critical role in ecosystems. Accurately extracting the area of a riparian zone in open water is challenging due to human activities and climate change. This study used Sentinel-1 satellite data to investigate the capabilities of the support vector machine, extreme gradient boosting, and random forest methods, which were optimized by genetic algorithms for the detection of area changes in the riparian zone in the heart region of the Three Gorges Reservoir area of China. A total of 29 images were collected in 2020, and three models were created for each image, which were then transferred to other phases. The models’ performance metrics were validated using all of the images. The results indicated that the SVM method achieved the best performance with an accuracy of 0.945, an F1_Score of 0.950, and a kappa coefficient of 0.889. The optimal model was then used to monitor the area changes in the riparian zone over the study area in 2020. It was calculated that the area of the riparian zones was the smallest on 26 December and the largest on 17 June, with a maximum riparian zone of 31.97 km2. Overall, this study demonstrates that an SVM is the most stable method for detecting area changes in a riparian zone when using Sentinel-1 data compared to the RF and XGB methods. The findings are anticipated to provide a feasible plan for detecting the area dynamics in open-water riparian zones and to provide valuable information for the rational utilization of land resources and the ecological safety of the riparian zone in the Three Gorges Reservoir.

Relationship of construction land expansion and ecological environment changes in the Three Gorges reservoir area of China

With the rapid development of urbanization, the disorderly construction land expansion takes up too much ecological space, leading to the exacerbation of the regional internal ecosystem. Therefore, exploring the coordinated relationship between construction land expansion and ecological environment change has become a key issue for regional sustainable development. Based on remote sensing technology, the evaluation system of construction land expansion and ecological environment change in the Three Gorges Reservoir area was constructed. The spatial–temporal evolution of construction land expansion and ecological environment change from 1995 to 2020 and their coordination were analyzed, and they were divided into coupling and coordinated development types. The research shows the following findings: Firstly, in the year 2020, the comprehensive value of construction land expansion reached 0.8635, an increase of more than 8 times compared to 1995, indicating a significant trend of rapid expansion. Secondly, the comprehensive value of ecological environment change from 1995 to 2020 remained between 0.4 and 0.6 and continued to change in a good direction. And improving the remote sensing ecological indicators has expanded the difference in ecological environment quality in the Three Gorges Reservoir Area. Thirdly, the coupled model increased from 0.0534 to 0.7951 from 1995 to 2020, which is in a low-level and slow-growing state, and social and economic development does not completely depend on destroying the ecological environment quality. Finally, three types of coordinated development are divided: ecological environment protection type, development and protection in tandem type, and ecological environment consumable type. Differentiated regulation strategies and suggestions are also proposed.

Spatiotemporal variations and its driving factors of soil conservation services in the Three Gorges Reservoir area in China

Soil conservation services play a vital role in regulating ecosystem services to prevent soil erosion and ensure regional ecological security. Therefore, effective evaluation and quantification of soil conservation services in the Three Gorges Reservoir Area (TGRA) are conducive to sustainable management under future global change. In this study, based on a basic database, including land use/cover data, soil data, topographic data, meteorological data, and NDVI (Normalized Difference Vegetation Index) data as the basic databases, to evaluate the temporal and spatial changes of soil conservation services in the TGRA from 1990 to 2015 at a regional-scale level using the general soil loss equation. The results showed that forest ecosystems (including coniferous and broad-leaved mixed forests, coniferous forests, shrub forests, and broad-leaved forests) made a greater contribution (69%) to regulating soil conservation in TGRA, followed by farmland ecosystems (29%). In total TGRA, large spatial variation in soil conservation, such as the highest appeared in the northern hinterland, whereas the lowest was mostly shown in the northwest with relatively frequent human activities and developed industry and agriculture. In general, soil conservation in the TGRA ecosystem gradually increased from 1990 to 2015, with a total increase of 6%. In this period, with the effective implementation of ecological projects, such as the conversion of farmland to forest and natural forest protection, the distributed proportion of forest land area in total TGRA showed a significant increase. In the meantime, the increase of vegetation coverage also helps the restoration of ecosystem structure and function and the improvement of soil conservation services. Our findings will aid our knowledge regarding the ecosystem services of the TGRA and provide implications for future sustainable land management and ecological protection.

Evaluation on processing suitability of 17 juice orange varieties planted in three gorges reservoir area of China

From 2014 to 2021, 17 sweet orange varieties and strains were continuously tested for their suitability for orange juice processing, focusing on the physical and chemical indicators of the fruit, namely juice yield, soluble solids, titratable acid, vitamin C, and sensory indicators including color value, defects, and flavor. We developed a numerical (percentage) method to evaluate the processing suitability for the juicing varieties, and all the tested data were processed using this method. A total of 12 sweet orange varieties with early, middle, and late maturity suitable for juicing were screened out, including six early matured varieties: Early Longleaf, Yuzaocheng, Westin, Early Gold, Yunguicheng, and Hamlin; two middle matured varieties: Middle Longleaf and Edangan (Jincheng); and four late matured varieties: Spring orange, Late Longleaf, Algerian Valencia, and Olinda Valencia. At the same time, five varieties, including the Early-maturings: Itaborai and Early Marrs; and the Middle-maturings: Tongshui 72-1, Trovita, and Midsweet, which can be juiced were also screened out. Among the 17 recommended varieties, the Early, Middle, and Late Longleaf are the most prominent.

Deep learning powered long-term warning systems for reservoir landslides

The reservoir landslides are characterized by repeated phases of acceleration and isokinetic deformation under long-term periodic external forces. The state-of-the-art research lacks reliable prediction methods and judgment of their evolution stages. This work promotes the application of the deep learning algorithm and landslide evolution model in long-term warning systems. The test site is the Sifangbei landslide in the Three Gorges reservoir area of China. The main innovative features are: (i) the displacement of the landslide is considered as the prediction target, and the optimal model (i.e., conditioning factors and hyper-parameters combination) driven by the deep learning framework is used for spatial prediction; (ii) different warning methods (from both literature and current practice) are compared to single out the one that can best describe the evolution stage of the reservoir landslide; and (iii) deep learning model and adaptive evolution model are combined to analyze the temporal-spatial kinematic characteristics and evolution trend of the landslide under extreme scenarios related to rainfall and reservoir water levels. The results show that the predicted displacements of the lower and central part of the landslide are respectively controlled by reservoir water level and rainfall; the five-stage evolution model can capture the long-term evolution trend of the Sifangbei landslide; under extreme scenarios, landslide deformation exhibits step-like characteristics and is more likely to start from the middle of the lower portion of the unstable area. These models represent the up-to-date steps of a long-term research plan. The gathered knowledge can be used to analyze the spatial evolution characteristics of landslides and promote the setup of long-term warning systems. Furthermore, the results show that combining the proposed deep learning and evolution methods provides forecast information that could help adjusting short-term warning strategies in such a complex risk area as the Three Gorges Reservoir.

Morphological evolution of desiccation cracking in purple soil under the influence of aggregate size distribution

Desiccation cracking in purple soil subjected to rainfall-evaporation cycles is a common natural phenomenon, in the Three Gorges Reservoir area of China, that detrimentally affects the hydraulic and mechanical properties of soils. The significant spatial variability in the aggregate size distribution (ASD) of purple soil in this area leads to variations in crack morphology. In this study, we experimentally investigated the effects of ASD on morphological evolution of cracks and conducted a regression analysis based on the ASD indices of purple soil. The dynamic geometrical, topological, and statistical characteristics of the crack morphology were analysed using digital image processing and morphological algorithms. The results indicated that the process of crack evolution in purple soil mainly contained three stages, two intersection shapes, and three network forms. The crack morphology in soil with a larger initial mean aggregate size was more irregular and discontinuous. We proposed a cracking evolution function that depicts the crack geometry and topology considering the gravimetric water content as a variable, which proved to be a good fit for modeling the dynamic morphological characteristics of desiccation cracks. The distribution of crack length, crack area, and clod area were well described by the ExpDec function, whereas the crack width revealed a bimodal Gaussian distribution, with primary and secondary peaks of 1–1.5 and 2–3.5 mm, respectively. Moreover, Multiple linear regression models between the crack geometric parameters and ASD indices were established, which indicated that the ASD in the micro-aggregate state determines the final morphological characteristics of cracks, with the content of clay-sized aggregates having the most significant effect. The tensile stress of aggregates in the unsaturated state decreases with increasing radius, based on the microscopic mechanical analysis. These results could provides a theoretical basis for predicting the macroscopic cracking behavior of purple soil under various ASD conditions.

Ensemble learning framework for landslide susceptibility mapping: Different basic classifier and ensemble strategy

The application of ensemble learning models has been continuously improved in recent landslide susceptibility research, but most studies have no unified ensemble framework. Moreover, few papers have discussed the applicability of the ensemble learning model in landslide susceptibility mapping at the township level. This study aims at defining a robust ensemble framework that can become the benchmark method for future research dealing with the comparison of different ensemble models. For this purpose, the present work focuses on three different basic classifiers: decision tree (DT), support vector machine (SVM), and multi-layer perceptron neural network model (MLPNN) and two homogeneous ensemble models such as random forest (RF) and extreme gradient boosting (XGBoost). The hierarchical construction of deep ensemble relied on two leading ensemble technologies (i.e., homogeneous/heterogeneous model ensemble and bagging, boosting, stacking ensemble strategy) to provide a more accurate and effective spatial probability of landslide occurrence. The selected study area is Dazhou town, located in the Jurassic red-strata area in the Three Gorges Reservoir Area of China, which is a strategic economic area currently characterized by widespread landslide risk. Based on a long-term field investigation, the inventory counting thirty-three slow-moving landslide polygons was drawn. The results show that the ensemble models do not necessarily perform better; for instance, the Bagging based DT-SVM-MLPNN-XGBoost model performed worse than the single XGBoost model. Amongst the eleven tested models, the Stacking based RF-XGBoost model, which is a homogeneous model based on bagging, boosting, and stacking ensemble, showed the highest capability of predicting the landslide-affected areas. Besides, the factor behaviors of DT, SVM, MLPNN, RF and XGBoost models reflected the characteristics of slow-moving landslides in the Three Gorges reservoir area, wherein unfavorable lithological conditions and intense human engineering activities (i.e., reservoir water level fluctuation, residential area construction, and farmland development) are proven to be the key triggers. The presented approach could be used for landslide spatial occurrence prediction in similar regions and other fields.

Thermo-hydro-poro-mechanical responses of a reservoir-induced landslide tracked by high-resolution fiber optic sensing nerves

Thermo-poro-mechanical responses along sliding zone/surface have been extensively studied. However, it has not been recognized that the potential contribution of other crucial engineering geological interfaces beyond the slip surface to progressive failure. Here, we aim to investigate the subsurface multi-physics of reservoir landslides under two extreme hydrologic conditions (i.e. wet and dry), particularly within sliding masses. Based on ultra-weak fiber Bragg grating (UWFBG) technology, we employ special-purpose fiber optic sensing cables that can be implanted into boreholes as “nerves of the Earth” to collect data on soil temperature, water content, pore water pressure, and strain. The Xinpu landslide in the middle reach of the Three Gorges Reservoir Area in China was selected as a case study to establish a paradigm for in situ thermo-hydro-poro-mechanical monitoring. These UWFBG-based sensing cables were vertically buried in a 31 m-deep borehole at the foot of the landslide, with a resolution of 1 m except for the pressure sensor. We reported field measurements covering the period 2021 and 2022 and produced the spatiotemporal profiles throughout the borehole. Results show that wet years are more likely to motivate landslide motions than dry years. The annual thermally active layer of the landslide has a critical depth of roughly 9 m and might move downward in warmer years. The dynamic groundwater table is located at depths of 9–15 m, where the peaked strain undergoes a periodical response of leap and withdrawal to annual hydrometeorological cycles. These interface behaviors may support the interpretation of the contribution of reservoir regulation to slope stability, allowing us to correlate them to local damage events and potential global destabilization. This paper also offers a natural framework for interpreting thermo-hydro-poro-mechanical signatures from creeping reservoir bank slopes, which may form the basis for a landslide monitoring and early warning system.

Soil resistance to flowing water erosion as affected by tea planting age in Three Gorges Reservoir Area of China

Soil erosion resistance is influenced by intrinsic soil properties and multiple external factors. This study investigated the effect of tea planting age on soil resistance to flowing water erosion (reflected by rill erodibility (Kr) and critical shear stress (τc)) in Three Gorges Reservoir Area. One slope farmland (as the control) and five tea plantations cultivated for 3 to 34 years were selected for sampling sites. The results indicated that bulk density (BD), soil cohesion (Coh), water stable aggregate (WSA), mean weight diameter (MWD), soil organic carbon (SOC), litter density (LD), and root mass density (RMD) increased generally with tea planting age. Compared to the control, Kr of tea plantations reduced by 71.1%–85.3%. The temporal variation in soil erosion resistance was controlled greatly by the variations in most near-surface characteristics. Kr decreased with WSA, Coh, LD, RMD, and SOC following a power function (P < 0.01); τc increased with MWD, LD, RMD, and SOC as an exponential function, with BD a power function, and Coh a logarithmic function (P < 0.01). In this study, Kr could be simulated well by WSA and LD with a power function, and τc could be simulated well by MWD and RMD with an exponential function.

Identifying the Groundwater Sources of Huangtupo Landslide in the Three Gorges Reservoir Area of China

Groundwater plays a crucial role in triggering and reactivating deep-seated landslides. However, classical hydrogeological investigations have limitations in their applicability to deep-seated landslides due to anisotropic and heterogeneous media. The Huangtupo landslide in the Three Gorges Reservoir area has garnered significant attention due to its high hazard potential. Of particular interest is the NO.1 Riverside Sliding Mass (HTP-1), which has shown notable deformation and has become the primary focus of landslide research. The study aims to investigate the sources of water in the HTP-1 landslide through hydrochemical analysis. This was achieved by monitoring the major ion content in the groundwater within the landslide for one year. Furthermore, stable isotope investigations were conducted on the groundwater in and around the landslide area, and an analysis of the mineral composition of the landslide soil was also performed. The results indicate that the groundwater in the landslide area (LGW) is a mixture of karst groundwater (KGW) from the adjacent upslope and local precipitation (LP). The karst groundwater is a major contributor to the recharge of the landslide groundwater system, causing a high component of groundwater that can easily exceed the critical level that causes landside failure during heavy rainfall events. Furthermore, prior to the relocation of residents from the Huangtupo landslide, the landslide groundwater was also impacted by human sewage, which not only affected the chemical composition of groundwater, but also had potential implications for slope stability. These findings provide a more scientific basis for the design and implementation of interception and drainage measures for the Huangtupo landslide and other large-scale landslides with similar geological conditions in the Three Gorges Reservoir area.

Application of Bagging, Boosting and Stacking Ensemble and EasyEnsemble Methods for Landslide Susceptibility Mapping in the Three Gorges Reservoir Area of China

Since the impoundment of the Three Gorges Reservoir area in 2003, the potential risks of geological disasters in the reservoir area have increased significantly, among which the hidden dangers of landslides are particularly prominent. To reduce casualties and damage, efficient and precise landslide susceptibility evaluation methods are important. Multiple ensemble models have been used to evaluate the susceptibility of the upper part of Badong County to landslides. In this study, EasyEnsemble technology was used to solve the imbalance between landslide and nonlandslide sample data. The extracted evaluation factors were input into three bagging, boosting, and stacking ensemble models for training, and landslide susceptibility mapping (LSM) was drawn. According to the importance analysis, the important factors affecting the occurrence of landslides are altitude, terrain surface texture (TST), distance to residences, distance to rivers and land use. The influences of different grid sizes on the susceptibility results were compared, and a larger grid was found to lead to the overfitting of the prediction results. Therefore, a 30 m grid was selected as the evaluation unit. The accuracy, area under the curve (AUC), recall rate, test set precision, and kappa coefficient of a multi-grained cascade forest (gcForest) model with the stacking method were 0.958, 0.991, 0.965, 0.946, and 0.91, respectively, which a significantly better than the values produced by the other models.

A Novel Hybrid LMD–ETS–TCN Approach for Predicting Landslide Displacement Based on GPS Time Series Analysis

Landslide disasters cause serious property losses and casualties every year. Landslide displacement prediction is fundamental for mitigating landslide disasters. Several approaches have been used to predict landslide displacement, yet a more accurate and reliable displacement prediction still has a poor understanding of landslide early warning systems for landslide mitigation, due to limited data and mutational displacements. To boost the robustness and accuracy of landslide displacement prediction, this paper assembled a new hybrid model containing the local mean decomposition (LMD), innovations state space models for exponential smoothing (ETS), and the temporal convolutional network (TCN). The proposed model, which is based on over 10 years of long-term time series monitoring GPS data, was tested on the selected case—stepwise Baijiabao landslide in the Three Gorges Reservoir area of China (TGRA) was tested by the proposed model. The results presented that the LMD–ETS–TCN model has the best performance in comparison with other benchmark models. Compared with autoregressive integrated moving average (ARIMA), support vector regression (SVR), and long short-term memory neural network (LSTM), the accuracy was noticeably improved by an average of 40.9%, 46.2%, and 22.1%, respectively. The robustness and effectiveness of the presented approach are attested, and it has discernible improvements for landslide displacement prediction.

Landslide detection based on shipborne images and deep learning models: a case study in the Three Gorges Reservoir Area in China

Landslide detection based on shipborne images and deep learning models: a case study in the Three Gorges Reservoir Area in China