Search Results (63)

Land-based hyperspectral imaging provides high spatial and spectral resolution for detecting camouflaged targets, but practical deployment remains limited by strong target background spectral similarity, scarce annotated hyperspectral samples, and the computational cost of full-band processing. To address these issues, this paper proposes HCTDNet (Hyperspectral Camouflaged Target Detection Network), a land-based hyperspectral image analysis framework. The method first employs band extraction for data dimensionality reduction, compressing multi-channel hyperspectral images into 3-channel virtual RGB representations, which reduces spectral redundancy while preliminarily enhancing camouflaged target saliency. A pre-trained RGB camouflaged target detector is then adopted as the backbone model, with its parameters frozen to maintain stability, while trainable modality-specific prompts are learned to improve training efficiency. Finally, model fine-tuning is performed using a self-constructed camouflaged target dataset to enhance robustness in detecting camouflaged targets within virtual RGB images. During inference, preprocessed hyperspectral images are fed into the model to generate detection results for camouflaged target regions. The experiments performed on our self-collected land-based hyperspectral dataset with camouflaged targets reveal that HCTDNet achieves superior detection performance compared with seven classical hyperspectral target detection methods while maintaining an average inference speed of approximately 16 FPS. The proposed framework provides an efficient and near-real-time applicable solution for land-based hyperspectral camouflaged target detection, showing significant practical potential. Full article

Assessing climate performance through emission-centric metrics provides an incomplete picture of countries’ progress toward integrated climate objectives. This study proposes the Climate Integrated Performance Index (CIPI), a multidimensional and explainable composite indicator for 27 European countries during 2015–2023. CIPI integrates six thematic dimensions: emissions, energy systems, mitigation capacity, transport, agriculture, and waste–land-use interactions, using robust normalization, a policy-informed weighting framework, and formal statistical validation. Based on 243 country–year observations, the results indicate that CIPI is non-redundant. Pearson correlations reveal strong positive associations with the Energy Index (r = 0.899) and Mitigation Index (r = 0.894), alongside a significant negative association with the Agriculture Index (r = −0.659), highlighting sectoral trade-offs. Variance decomposition further shows that energy and mitigation dimensions jointly account for approximately 87% of explained variance, whereas agriculture exerts a systematic counterbalancing influence. To support structural interpretation, an explainable machine learning framework combining XGBoost and SHAP was implemented as a diagnostic layer. Renewable-energy capacity emerged as the dominant structural driver of integrated climate performance, and SHAP-based analyses revealed a nonlinear threshold effect, with positive contributions accelerating beyond a normalized renewable-capacity level of approximately 0.58 (95% bootstrap confidence interval: 0.54–0.62), particularly under low fossil-fuel dependency conditions. Because the machine learning models use indicators that also contribute to index construction, the results are interpreted as evidence of structural consistency and diagnostic interpretability rather than independent predictive discovery. To address this limitation, repeated cross-validation, subsample validation, benchmark comparisons, and weighting-sensitivity analyses were conducted. Ranking robustness remained high under alternative weighting schemes (Spearman ρ > 0.96), while comparison with an emission-centric benchmark demonstrated substantial rank reversals, indicating that broader sectoral and policy dimensions influence climate-performance assessment. Overall, CIPI functions not only as a benchmarking tool but also as a transparent diagnostic framework for identifying structural trade-offs, nonlinear relationships, and policy-relevant climate-transition dynamics. Full article

Hyperspectral remote sensing provides rich spectral information and has been widely used in fine-grained land-cover classification and forest monitoring. However, accurate tree species classification remains challenging due to subtle interspecific spectral differences, similar spatial structures among related species, redundant spectral bands, and the limited ability of existing methods to model long-range spatial–spectral dependencies efficiently. In addition, many existing hyperspectral image classification methods rely on patch-based inputs and sliding-window inference, which often lead to redundant computation and insufficient utilization of global image context. To address these issues, this paper proposes MambaHSINet, a dual-branch bidirectional state space network for full-image pixel-wise hyperspectral classification. Specifically, the proposed network employs a spectral branch and a spatial branch to explicitly extract complementary spectral responses and spatial structural features. Subsequently, a bidirectional Mamba global modeling module based on selective state space modeling is adopted to capture long-range contextual dependencies in both forward and backward directions with linear computational complexity. Unlike conventional patch-based methods, MambaHSINet takes the entire hyperspectral image as input and produces full-resolution pixel-wise classification maps, thereby avoiding repeated cropping and redundant sliding-window inference. We also construct two well-annotated subsets of a UAV-borne hyperspectral dataset dedicated to tree species classification. Experimental results on self-collected and public hyperspectral datasets demonstrate that the proposed method achieves excellent classification accuracy, inference efficiency, and generalization performance. It exhibits great potential for practical tree species classification and other general hyperspectral application scenarios. Full article

Accurate crop mapping is important for agricultural monitoring and land management; yet, identifying robust and compact feature subsets from high-dimensional multi-sensor remote sensing data remains challenging, particularly in heterogeneous agricultural landscapes affected by spatial autocorrelation. Although combining multi-sensor data provides complementary spectral and structural information, traditional workflows often neglect spatial dependence during feature evaluation, leading to over-optimistic validation metrics and spatially unstable feature subsets. To address this issue, this study proposes a hierarchical feature selection and subset optimization framework for crop mapping by integrating Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 optical imagery within the Google Earth Engine (GEE) platform. A total of 135 multi-sensor features were constructed, including spectral bands, vegetation indices, SAR metrics, texture descriptors, and phenological statistics. To improve feature compactness and spatial robustness, a multi-stage selection strategy combining correlation-based redundancy removal, spatial cross-validation (SCV) control, Boruta, recursive feature elimination (RFE), L1 regularization, SHapley Additive exPlanations (SHAP), and Non-dominated Sorting Genetic Algorithm II (NSGA-II) was developed. Results showed that temporal and phenological features contributed more strongly to crop discrimination than static spectral or SAR features, while multi-sensor integration further improved classification stability. Notably, the proposed framework reduced the feature space from 135 to 12 variables while slightly improving classification performance. The final optimized model achieved an overall accuracy (OA) of 96.98% under SCV and generated spatially consistent crop maps at 10 m resolution. The framework provides an efficient and scalable solution for fine-scale crop mapping in complex agricultural regions and demonstrates the practical potential of incorporating spatial dependence control into feature selection for large-scale agricultural monitoring applications. Full article

Rapid urbanization has intensified wetland fragmentation and ecological connectivity degradation, threatening the structural stability and functional sustainability of urban wetland ecosystems. Constructing resilient wetland ecological networks is therefore essential for maintaining regional ecological security and supporting sustainable urban development. Taking Wuhan as a case study, multi-temporal land-use data from 2004, 2014, and 2024, together with land-use transition matrices, were used to analyze urban expansion and wetland landscape transformation. Morphological Spatial Pattern Analysis (MSPA), the Minimum Cumulative Resistance (MCR) model, and circuit theory were integrated to identify ecological sources, construct ecological corridors, and evaluate the structural connectivity of the wetland ecological network. Ecological source importance was quantified using the Probability of Connectivity (PC) and dPC indices. In addition, robustness analysis based on the sequential removal of high-dPC ecological source patches was conducted to assess network stability under disturbance scenarios. The results identified 20 core ecological source areas and 45 ecological corridors, forming a relatively interconnected wetland ecological network centered around major lake clusters and key ecological hubs. High-current corridors and pinch points were mainly distributed in ecologically sensitive transition zones and urban expansion boundaries. Robustness analysis showed that sequential removal of high-dPC ecological hubs resulted in continuous declines in EC(PC) and corridor number, while corridor length increased substantially. Although overall connectivity was maintained through alternative ecological pathways, ecological movement efficiency decreased significantly under disturbance scenarios, indicating increasing dispersal costs and reduced structural stability. These findings suggest that the wetland ecological network possesses moderate structural connectivity through pathway redundancy but remains highly dependent on several dominant ecological hubs. This study extends traditional static connectivity assessment by incorporating robustness and disturbance response analysis into wetland ecological network evaluation. The proposed framework provides scientific support for resilient wetland conservation, ecological restoration, and sustainable spatial planning in rapidly urbanizing metropolitan regions. Full article

Decoration waste, because of its complex composition and the presence of volatile toxic and hazardous substances, has always been a difficult point in the management of urban construction waste. And with the continuous expansion of the town scale, the volume of decoration waste is gradually expanding, which constitutes a major challenge to the sustainable development of the construction industry. In order to solve this difficult problem, this paper took Henan Province as an example, and realized the accurate control of decoration waste based on GF-2 remote sensing images and a BP neural network model. The results of GF-2 remote sensing image interpretation and analysis showed that the spatial distribution of construction waste in the study area was extracted through a combination of manual visual interpretation and machine learning recognition, and as of 2021, the construction waste pile occupied a large proportion of the land area, of which the proportion of decoration waste was about 10%. Based on the trained BP neural network, the goodness-of-fit result was R = 0.95463. Selecting the research data from 2010 to 2021, the error of the predicted annual generation of decoration waste in Henan Province compared with the actual value was less than 15%, which had a high prediction accuracy. Based on the arithmetic sum of the projected figures for each year from 2022 to 2030, it is estimated that by 2030, the cumulative volume of construction and renovation waste generated in Henan Province will reach 49,827,200 tons. Visualization of spatial and temporal distribution characteristics was realized through ArcGIS, and the high production area of decoration waste was distributed from the beginning to the end of the distribution of multi-points to show the characteristics of a concentrated large area distribution, centrally located in southwestern and southeastern Henan Province, with the key cities of Zhumadian City, Luoyang City, Zhoukou City, and Xinyang City, which had obvious regional characteristics. At the same time, as the provincial capital, Zhengzhou has long ranked first in the province in terms of absolute case numbers and is therefore also a key focus of control measures. Uncertainty analysis indicates that the 95% confidence interval for the long-term forecast values is approximately ±12%. It is recommended to use the upper limit of this interval for the redundancy design of the absorption facilities to enhance the robustness of the decision. This study provides a theoretical basis and technical support for the governmental supervision of decoration waste during the development of national urban agglomerations, effectively solves regional urban planning and construction management problems, and promotes the sustainable development of the construction industry. Full article

Aiming to address the limitations in accuracy and reliability of land use classification models when encountering complex and variable land use features, a novel method for land use classification (4D-SOMKS) is developed, which constructs a low redundancy four-dimensional (4D) spectral feature system with biophysical meaning of distinct land surface properties and organizes pixel-level features in a topology-preserving space using the Self-Organizing Map (SOM), further groups the SOM neurons into spectrally coherent clusters through K-means, and uses a small number of labeled samples only for semantic assignment of the resulting clusters rather than for pixel-level supervised model training. Empirical research in the Bayannur and Hong Lake Basins (HLB) have revealed the driving role of key spectral indices in classification. High overall classification accuracies were achieved, reaching 98.45% and 98.55% respectively, with robust performance across evaluation metrics including precision, recall, F1-score, and the Kappa coefficient. The results show that 4D-SOMKS achieves high accuracy and robustness while significantly reducing reliance on large-scale labeled data, providing an effective avenue to improve the accuracy and reliability of land use classification under spatiotemporal dynamic changes. Full article

Accurate crop yield prediction increasingly relies on diverse data streams, including satellite observations, meteorological reanalysis, soil composition, and topographic information. However, despite advances in machine learning, many existing approaches remain crop- or region-specific and require substantial bespoke data engineering, limiting scalability and reproducibility. This study introduces UniCrop, a generalisable, configuration-driven data engineering pipeline that standardises the acquisition, harmonisation, and feature construction of multi-source agro-environmental data. Rather than proposing a new predictive model, UniCrop addresses a key bottleneck in agricultural machine learning: the lack of reproducible and scalable data preparation workflows. For any given location, crop type, and temporal window, the pipeline automatically retrieves, harmonises, and engineers over 160 environmental variables from heterogeneous sources (Sentinel-1/2, MODIS, ERA5-Land, NASA POWER, SoilGrids, and SRTM), reducing them to a compact, analysis-ready feature set using a structured feature selection process based on minimum redundancy maximum relevance (mRMR). The effectiveness of the pipeline is demonstrated through a case study, where the generated datasets enable robust baseline modelling across multiple machine-learning algorithms. Using a selected subset of 15 features, four baseline models (LightGBM, Random Forest, Support Vector Regression, and ElasticNet) were evaluated under rigorous cross-validation. LightGBM achieved the best single-model performance (RMSE = 465.1 kg/ha, $R^2=0.6576$), while a constrained ensemble provided a marginal improvement (RMSE = 463.2 kg/ha, $R^2=0.6604$). SHAP-based analysis further confirms that the selected features capture agronomically meaningful relationships across data modalities. UniCrop contributes a scalable and transparent data engineering pipeline that enables consistent, reproducible, and transferable dataset construction for crop yield prediction. By decoupling data specification from implementation and supporting flexible configuration across crops, regions, and temporal contexts, the framework provides a practical foundation for large-scale agricultural analytics. Full article

Research on hub-and-spoke logistics networks can effectively advance the construction of the Northeast China Sea–Land Grand Corridor. In the context of regional synergistic development, this study investigates the optimization of the logistics network for the Northeast China Land–Sea Grand Corridor. Focusing on 43 prefecture-level cities across Liaoning, Jilin, Heilongjiang, and Inner Mongolia, a hub-and-spoke logistics network optimization model is developed. The model aims to minimize total network costs while satisfying specific network resilience thresholds. It integrates multi-modal transport and incorporates considerations such as economies of scale, node heterogeneity in resilience evaluation, and route redundancy. Based on this, the study employs the entropy weight method to establish a comprehensive evaluation system for regional logistics and economic development levels and applies an improved coupling coordination degree model to assess the synergistic relationship between these two systems. A modified gravity model, with the coupling coordination degree as a moderating coefficient, is constructed to quantify the strength of logistics–economic linkages between cities. Furthermore, social network analysis and a logistics affiliation model are used to identify key hub cities. The results demonstrate that the optimized network significantly enhances transport efficiency, achieves substantial economies of scale and strikes a balance between cost efficiency and system resilience. This research provides a quantitative foundation and practical reference for node layout planning and multi-modal transport organization along the Northeast China Sea–Land Grand Corridor, and its methodological framework can inform logistics network planning in similar regions. Full article

Soil salinization severely threatens agricultural sustainability in saline–alkali regions, and high-throughput, efficient screening of salt-tolerant rice varieties is critical to mitigating this threat. Traditional evaluation methods are constrained by low throughput, limited spatiotemporal resolution, and the lack of standardized indicators. To address these gaps, this study established a multi-scale spectral phenotyping framework integrating ground-based hyperspectral, UAV-borne multispectral, and Sentinel-2 satellite remote sensing data for high-throughput screening of salt-tolerant rice. Field experiments were conducted with 12 rice lines at five key growth stages in Ningxia, China, with synchronous ground spectral measurements and UAV image acquisition on the same day for each stage. Five feature selection methods were employed to screen salt stress-sensitive hyperspectral bands, with classification accuracy validated via a Support Vector Machine (SVM) model. The results showed that: (1) rice spectral characteristics varied dynamically across growth stages, and first-order differential transformation effectively amplified subtle spectral variations in stress-sensitive regions; (2) the Minimum Redundancy–Maximum Relevance (mRMR) method outperformed other methods, achieving 100% classification accuracy at key growth stages, with sensitive bands dominated by red edge bands (58.33%); (3) the constructed Salt Stress Index (SIR) showed strong correlations with classical vegetation indices and rice yield, and could clearly distinguish salt-tolerant and salt-sensitive rice varieties, with stable performance against field environmental noise; and (4) band matching between UAV and Sentinel-2 data enabled multi-scale data fusion and regional-scale salt stress monitoring. This framework realizes the transformation from qualitative spectral description to quantitative salt tolerance evaluation, providing standardized technical support for salt-tolerant rice breeding and precision management of saline–alkali lands. Full article

In this study, using global liner shipping schedules, UNCTAD’s Port Liner Shipping Connectivity Index and Liner Shipping Bilateral Connectivity Index, together with bilateral trade-value data for 2022–2024, we construct a multilayer weighted port-to-port network that explicitly embeds port-level cargo-handling and service organisation capabilities, as well as demand-side routing pressure, into node and edge weights. Building on this network, we apply CONCOR-based structural-equivalence analysis to delineate functionally homogeneous port clusters, and adopt a structural role identification framework that combines multi-indicator connectivity metrics with Rank-Sum Ratio–entropy weighting and Probit-based binning to classify ports into high-efficiency core, bridge-control, and free-form bridge roles, thereby tracing the reconfiguration of cluster-level functional structures before and after the Red Sea crisis. Empirically, the clustering identifies four persistent communities—the Intertropical Maritime Hub Corridor (IMHC), Pacific Rim Mega-Port Agglomeration (PRMPA), Southern Commodity Export Gateway (SCEG), and Euro-Asian Intermodal Chokepoints (EAIC)—and reveals a marked spatial and functional reorganisation between 2022 and 2024. IMHC expands from 96 to 113 ports and SCEG from 33 to 56, whereas EAIC contracts from 27 to 10 nodes as gateway functions are reallocated across clusters, and the combined share of bridge-control and free-form bridge ports increases from 9.6% to 15.5% of all nodes, demonstrating a thicker functional backbone under rerouting pressures. Spatially, IMHC extends from a Mediterranean-centred configuration into tropical, trans-equatorial routes; PRMPA consolidates its role as the densest trans-Pacific belt; SCEG evolves from a commodity-based export gateway into a cross-regional Southern Hemisphere hub; and EAIC reorients from an Atlantic-dominated structure towards Eurasian corridors and emerging bypass routes. Functionally, Singapore, Rotterdam, and Shanghai remain dominant high-efficiency cores, while several Mediterranean and Red Sea ports (e.g., Jeddah, Alexandria) lose centrality as East and Southeast Asian nodes gain prominence; bridge-control functions are increasingly taken up by European and East Asian hubs (e.g., Antwerp, Hamburg, Busan, Kobe), acting as secondary transshipment buffers; and free-form bridge ports such as Manila, Haiphong, and Genoa strengthen their roles as elastic connectors that enhance intra-cluster cohesion and provide redundancy for inter-cluster rerouting. Overall, these patterns show that resilience under the Red Sea crisis is expressed through the cluster-level rebalancing of core–control–bridge roles, suggesting that port managers should prioritise parallel gateways, short-sea and coastal buffers, and sea–land intermodality within clusters when designing capacity expansion, hinterland access, and rerouting strategies. Full article

To investigate the influence of land use landscape patterns on lake water quality in the basin, the land use and water quality data of the Yilong Lake Basin from 1993 to 2023 were analyzed with a geographic information system, remote sensing, and landscape ecology methods in this research. The results show that (1) the land use landscape pattern and water quality of the Yilong Lake Basin had significant changes: the lake surface area, farmland, and shrubland declined, with grassland showing the sharpest decrease and serving as the main source of conversion to other land types, while forest land expanded and built-up land increased by five times. The landscape pattern analysis showed that the aggregation degree of the core habitat in the basin increased and the landscape had decreased patch density and increased heterogeneity. Regarding water quality, the concentrations of total nitrogen (TN), total phosphorus (TP), and ammonium nitrogen (NH₄⁺-N); permanganate index (I_Mn); and biochemical oxygen demand over 5 days (BOD₅) decreased. Furthermore, the concentration of dissolved oxygen (DO) increased and the concentration of chlorophyll-a (Chl-a) fluctuated for a long time but did not decrease dramatically at the end of the period compared with the beginning. In general, the eutrophication degree of Yilong Lake slightly decreased. (2) The landscape configuration strongly shaped the water quality: the redundancy analysis (RDA) revealed that the edge density (ED), landscape shape index (LSI), largest patch index (LPI), and patch density (PD) were negatively associated with the eutrophication of Yilong Lake (TN, TP, NH₄⁺-N, Chl-a), whereas the contagion index (CONTAG) was positively associated; the Shannon’s diversity index (SHDI) was closely linked with TN and I_Mn but negatively with DO; and the patch cohesion index (COHESION) had a low interpretation power for water quality changes. In particular, larger and more cohesive ecological patches supported a higher DO, while an increased patch density was linked to an elevated I_Mn and reduced DO. These results indicate that the restoration of key ecological patches and enhanced landscape cohesion helped to improve the water quality, whereas increased patch density and landscape heterogeneity negatively affected it. (3) In the past 30 years, the ecological management and protection work on Yilong Lake, such as returning farmland to forests and lakes, wetland restoration, and sewage pipe network construction, achieved remarkable results that were reflected in the change in the relationship between land use landscape pattern and water quality in the basin. However, human activities still affected the dynamic evolution of water quality: the expansion of built-up land increased the patch density, the reduction in shrubland and grassland weakened natural filtration, and the rapid urbanization process introduced more pollution sources. Although the increase in forest land helped to improve the water quality, the effect was not fully developed. These findings provide a scientific basis for the management and ecological restoration of plateau lakes. Strengthening land use planning, controlling urban expansion, and maintaining ecological patches are essential for sustaining water quality and promoting the coordinated development of the ecology and economy in the Yilong Lake Basin. Full article

Soil salinization in arid regions threatens ecological security and sustainable agriculture. The Ebinur Lake wetland in Xinjiang, situated in an arid climate and subject to human disturbance, suffers from severe salt accumulation and ecological degradation. To overcome the lack of soil depth information and limited spatiotemporal monitoring, this study integrates multi-year field samples and Landsat imagery (1996–2024) to construct a six-layer (0–100 cm) soil salinity inversion framework. Multi-source spectral features were optimized using the Random Frog Leaping Algorithm (RFLA), and models based on Convolutional Neural Network (CNN), Long Short-Term Memory Network (LSTM), and Random Forest (RF) were compared. The results (1) demonstrated that RFLA effectively identified high-contribution features, enhancing efficiency and reducing redundancy; (2) showed that CNN outperformed LSTM and RF in capturing spatial salinity, with R² values of 0.75, 0.59, 0.63, 0.69, 0.57, and 0.56 for the six layers; and (3) revealed salinity migration: surface enrichment, mid-layer buffering, and deep-layer accumulation. In oases, surface salinity declined while deep layers accumulated; in deserts, surface salinity increased. The proposed framework enhances the accuracy of multi-depth salinity retrieval and provides technical support for salinization monitoring, irrigation management, ecological assessment, and control of land degradation in arid regions. Full article

In the rapid development process of cities, as important ecological corridors and landscape carriers, the water quality conditions of urban water bodies are not only related to the health of the ecological environment, but also closely linked to the quality of life of residents. The landscape pattern, as an important component of the urban ecosystem, has a potential impact on water quality. As a tropical coastal city, the unique water network pattern of Haikou City is facing the dual challenges of landscape fragmentation and water quality pollution in its rapid urban expansion. In order to study the impact of the landscape pattern of Haikou City on urban water bodies, this study takes the urban water bodies of Haikou City as the research object. By comprehensively applying landscape ecology methods and water quality monitoring techniques, and using landscape pattern indices (such as the number of patches, fragmentation degree, spread degree, etc.) and on-site investigation of water quality parameter data (such as chemical oxygen demand (COD), total nitrogen (TN), total phosphorus (TP), etc.), and by using correlation analysis and redundancy analysis, we explore the mechanism by which landscape patterns affect water quality. The results show that: (1) There are significant differences in water quality among water bodies. The concentrations of COD and TN in Hongcheng Lake are relatively high. The average values reached 86.603 mg/L and 13.368 mg/L, respectively, mainly affected by the high-intensity construction land around. Jinniu Lake has a high degree of landscape fragmentation and relatively high concentrations of NH3-N and TP. The average values are 2.086 mg/L and 0.154 mg/L, respectively. The Meishe River has a strong water purification capacity due to its good vegetation coverage. (2) The influence of landscape pattern on water quality has a scale effect. Hongcheng Lake, Jinniu Lake, and Meishe River all have the best interpretation rate of water quality in the 2000 m buffer zone landscape pattern. (3) The expansion of construction land has significantly exacerbated water pollution, while natural vegetation landscapes with high connectivity and low fragmentation can effectively improve water quality. The research reveals the correlation between urban landscape planning and water quality protection. It is suggested that by enhancing ecological connectivity, controlling non-point source pollution, and implementing differentiated seasonal management, the self-purification capacity of water bodies can be improved, providing a scientific basis for ecological restoration and sustainable development in Haikou City. Full article

Landslides represent a recurrent hazard in tropical mountain environments, where rapid urbanization and extreme rainfall amplify disaster risk. The Sentani region of Papua, Indonesia, is highly vulnerable, as demonstrated by the catastrophic debris flows of March 2019 that caused fatalities and widespread losses. This study developed high-resolution landslide susceptibility maps for Sentani using an ensemble machine learning framework. Three base learners—Random Forest, eXtreme Gradient Boosting (XGBoost), and CatBoost—were combined through a logistic regression meta-learner. Predictor redundancy was controlled using Pearson correlation and Variance Inflation Factor/Tolerance (VIF/TOL). The landslide inventory was constructed from multitemporal satellite imagery, integrating geological, topographic, hydrological, environmental, and seismic factors. Results showed that lithology, Slope Length and Steepness Factor (LS Factor), and earthquake density consistently dominated model predictions. The ensemble achieved the most balanced predictive performance, Area Under the Curve (AUC) > 0.96, and generated susceptibility maps that aligned closely with observed landslide occurrences. SHapley Additive Explanations (SHAP) analyses provided transparent, case-specific insights into the directional influence of key factors. Collectively, the findings highlight both the robustness and interpretability of ensemble learning for landslide susceptibility mapping, offering actionable evidence to support disaster preparedness, land-use planning, and sustainable development in Papua. Full article