International journal of applied earth observation and geoinformation : ITC journal最新文献

{"title":"Few-Shot multispectral segmentation of PV expansion and Land-Use dynamics in China","authors":"Xiaopu Zhang ,&nbsp;Huayi Wu ,&nbsp;Shuyang Hou ,&nbsp;Zhangyan Xu ,&nbsp;Yongxian Zhang ,&nbsp;Jianfang Ma ,&nbsp;Dan Liu ,&nbsp;Yuanyi Jiang ,&nbsp;Jianxun Wang","doi":"10.1016/j.jag.2026.105185","DOIUrl":"10.1016/j.jag.2026.105185","url":null,"abstract":"<div><div>Driven by global carbon neutrality initiatives, China’s rapid expansion of photovoltaic (PV) power generation necessitates large-scale and precise extraction of photovoltaic power stations (PPS) for effective resource management. However, in 10 m resolution remote sensing imagery, PPS targets frequently exhibit multi-scale and fragmented spatial distributions. Such characteristics often lead to limited model generalization, high omission rates, and elevated commission errors, particularly under sparse-sample conditions. To address these challenges, this study introduces PPS-SAM, a spectrum- and structure-aware extraction framework developed for sparse-sample scenarios based on the Segment Anything Model (SAM). PPS-SAM integrates a Spectral Enhancement Encoder to fuse near-infrared (NIR) and shortwave-infrared (SWIR) bands, thereby improving the spectral separability of PV targets from heterogeneous backgrounds. It also incorporates a High-Quality Mask Decoder to maintain edge integrity and delineate fragmented arrays more effectively. Evaluated on a newly developed 10 m multispectral PPS dataset (MSPV-Dataset), PPS-SAM demonstrated robust segmentation performance with only 11 training samples (F1: 91.47% ± 0.13%; mIoU: 91.40% ± 0.08%), notably surpassing baseline models trained on the complete 634-sample dataset. Ablation and generalization assessments indicate the effectiveness of each module in enhancing foreground detection and background suppression, with stable performance across diverse, unseen terrains and environmental disturbances (F1: 92.19%; mIoU: 92.19%). Applying this framework, nationwide PPS distribution maps for 2022 and 2024 were generated (excluding distributed rooftop systems). The results indicate that the total PPS area expanded from 3,486.41 km2 to 5,900.89 km2, representing an increase of approximately 70%. Regional analysis shows that Northwest China was characterized by predominantly centralized growth (78.77%), whereas eastern and southwestern regions exhibited significant distributed expansion. Although national spatial clustering weakened slightly (Global Moran’s I decreased from 0.4155 to 0.3112), it intensified within central and southwestern provinces. Land-use transition analysis suggests that new PV installations primarily originated from grasslands (59.04%) and barren lands (96.40%), highlighting substantial land-cover conversion. These spatio-temporal patterns underscore regional disparities in PV expansion and shifting spatial structures. This study offers a robust methodological framework for high-precision PPS identification at 10 m resolution under sparse-sample constraints, supporting efficient renewable energy management and evidence-based policy formulation. The dataset is publicly available via Figshare (<span><span>https://doi.org/10.6084/m9.figshare.29618429)</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105185"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A resource-efficient training framework for remote sensing text–image retrieval","authors":"Weihang Zhang ,&nbsp;Jihao Li ,&nbsp;Shuoke Li ,&nbsp;Ziqing Niu ,&nbsp;Jialiang Chen ,&nbsp;Wenkai Zhang ,&nbsp;Xin Gao ,&nbsp;Xian Sun","doi":"10.1016/j.jag.2026.105174","DOIUrl":"10.1016/j.jag.2026.105174","url":null,"abstract":"<div><div>Remote sensing text–image retrieval (RSTIR) aims to retrieve the matched remote sensing (RS) images from the database according to the descriptive text. Recently, the rapid development of large visual-language pre-training models provides new insights for RSTIR. Nevertheless, as the complexity of models grows in RSTIR, the previous studies suffer from suboptimal resource efficiency during transfer learning. To address this issue, we propose a computation and memory-efficient retrieval (CMER) framework for RSTIR. To reduce the training memory consumption, we propose the Focus-Adapter module, which adopts a side branch structure. Its focus layer suppresses the interference of background pixels for small targets. Simultaneously, to enhance data efficacy, we regard the RS scene category as the metadata and design a concise augmentation technique. The scene label augmentation leverages the prior knowledge from land cover categories and shrinks the search space. We propose the negative sample recycling strategy to make the negative sample pool decoupled from the mini-batch size. It improves the generalization performance without introducing additional encoders. We have conducted quantitative and qualitative experiments on public datasets and expanded the benchmark with some advanced approaches, which demonstrates the competitiveness of the proposed CMER. Compared with the recent advanced methods, the overall retrieval performance of CMER is 2%–5% higher on RSITMD. Moreover, our proposed method reduces memory consumption by 49% and has a 1.4x data throughput during training. The code of the CMER and the dataset will be released at <span><span>[Link]</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105174"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From canopy segmentation to accurate prediction: An UAV-based multi-feature fusion framework for plot-scale ratoon sugarcane seedling counting","authors":"Hongyan Zhu ,&nbsp;Zhihao Dong ,&nbsp;Litao Wei ,&nbsp;Shuai Qin ,&nbsp;Xiaoyan Qin ,&nbsp;Yong He","doi":"10.1016/j.jag.2026.105183","DOIUrl":"10.1016/j.jag.2026.105183","url":null,"abstract":"<div><div>Accurate and efficient monitoring of seedling emergence is critical for early-stage crop management and yield forecasting in sugarcane production. To meet this practical demand for precise field phenotyping, this study developed a high-throughput phenotyping framework leveraging unmanned aerial vehicle (UAV) remote sensing data and machine learning. This framework addresses the critical agricultural challenges of inefficient manual counting and the need for plot-scale monitoring in sugarcane production by enabling high-throughput sugarcane seedling number prediction through the integration of UAV-acquired RGB and multispectral imagery. Specifically, the sugarcane canopy was accurately segmented from the background using K-means clustering, a step that enabled the extraction of canopy area and the generation of a mask for obtaining canopy-level average features (including vegetation indices and texture features). These features together form a comprehensive feature set. Subsequently, six different feature selection methods were used to optimize the feature set, and eight machine learning models were combined for training and evaluation. The results showed that the combination of Gradient Boosting Regression (GBR) and KBest-F feature selection method yielded the optimal prediction performance, with a coefficient of determination (R²) of 0.7641, a root mean square error (RMSE) of 19.42, and a mean absolute error (MAE) of 15.93. Further analysis identified canopy area, the Normalized Difference Red Edge Index (NDRE), red edge contrast, and green entropy as core predictive features. They collectively contribute over 60% of total feature importance, and their synergistic effects support accurate seedling number estimation. This framework offers an efficient, scalable tool for plot-scale seedling monitoring, with substantial potential for precision field management of high-density crops.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105183"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the impact of road on sensitivity of large slow-moving landslides to precipitation by integrating multi-temporal InSAR and panel regression","authors":"Yi Zhang ,&nbsp;Wangcai Liu ,&nbsp;Guan Chen ,&nbsp;Tom Dijkstra ,&nbsp;Xingmin Meng ,&nbsp;Xiang Wu ,&nbsp;Jing Chang ,&nbsp;Yuanxi Li ,&nbsp;Yanzhong Yang","doi":"10.1016/j.jag.2026.105192","DOIUrl":"10.1016/j.jag.2026.105192","url":null,"abstract":"<div><div>Human-landscape interactions in mountainous terrains are complex and multi-faceted. Settlements tend to focus on relatively gently undulating terrains, which are often found in areas where ground conditions are weak and thus substantial ground movements prevail. Interventions in these precarious landscapes, such as progressive expansion of interconnecting transport infrastructure, affect the stress balance and hydrology of already critical slopes, potentially enhancing their sensitivity to changes. The mountainous Bailong River Corridor (BRC) in Northwest China is dominated by large slow-moving landslides, where any transport infrastructure expansion has to transit through. A complex interplay of human, precipitation, and seismic factors determines the triggering dynamics of these large movements. This study integrates displacement time series, precipitation, and road distribution to quantify the impact of road emplacement on the sensitivity of large slow-moving landslides to precipitation regionally using panel regression analysis. It is shown that road disturbance significantly amplifies the sensitivity of landslide displacements to precipitation, and paved roads on the large slow-moving landslides increase their sensitivity to precipitation by 40%. Roads (both paved and unpaved) also reduce the threshold of antecedent cumulative precipitation required to trigger significant displacement, shortening the typical period from 132 days to only 120 days. The enhanced response frequency increases large reactivated landslide risk, and impacts road operation and management. This better understanding of the precipitation signature in the dynamics of large slow-moving landslides transited by roads contributes to improving future road planning, enhancing landslide risk mitigation, and strengthening urban resilience in vulnerable alpine environments.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105192"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Discriminating natural and planted forests in subtropical China using Sentinel-2 imagery and inventory data at 10 m resolution","authors":"Haiyang Guo ,&nbsp;Jia Sun ,&nbsp;Zenghui Fan ,&nbsp;Zhen Yu ,&nbsp;Feng Tian ,&nbsp;Xuemei Mao ,&nbsp;Lunche Wang ,&nbsp;Shaoqiang Wang ,&nbsp;Wei Gong ,&nbsp;Feiyue Mao ,&nbsp;Anders Ahlstrom","doi":"10.1016/j.jag.2026.105179","DOIUrl":"10.1016/j.jag.2026.105179","url":null,"abstract":"<div><div>Accurately distinguishing between natural forests and planted forests underpins effective biodiversity monitoring and climate policy, yet their spectral similarity and limited reference data challenge large-scale classification efforts. This study focuses on Guangxi Province, China, combining Sentinel-2 imagery, multi-source ecological indicators—including vegetation traits such as leaf chlorophyll content and LAI—and over 12 million visually interpreted forest stands to map the distribution of natural and planted forests. We assessed pixel- and object-based classification models under various sample scenarios, with accuracy independently validated using additional samples interpreted from high-resolution Google Earth imagery. Model evaluation relied primarily on random splits, supplemented with spatially stratified validation across ecological sub-regions to assess spatial generalization. Results show that the object-based random forest classifier achieved the highest accuracy (OA: 84.52%, F1: 84.04%), with topographic and vegetation functional traits as key predictors. Spatially, natural forests dominate mountainous zones, while planted forests prevail in flatter areas. This work delivers the first 10 m resolution forest type map for Guangxi based on a uniquely large training and validation dataset and demonstrates that combining functional traits with robust sampling improves classification performance. Our framework supports scalable forest monitoring and contributes to improved management and conservation strategies.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105179"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Field-scale water use assessment in the Lower Mississippi Alluvial Plain using OpenET","authors":"Weina Duan ,&nbsp;Yun Yang ,&nbsp;Martha C. Anderson ,&nbsp;Christopher R. Hain ,&nbsp;Forrest S. Melton ,&nbsp;John M. Volk ,&nbsp;Kyle R. Knipper","doi":"10.1016/j.jag.2026.105172","DOIUrl":"10.1016/j.jag.2026.105172","url":null,"abstract":"<div><div>Water scarcity and unprecedented rapid depletion of groundwater reserves are challenges for global sustainable agricultural development. Understanding water usage is especially critical in key agricultural production regions. In this study, we focus on the Lower Mississippi Alluvial Plain (LMAP) region, one of the most important intensive agricultural regions in the United States and facing severe groundwater depletion. However, the total consumptive water use and its relationships with drought events across different land cover types remain a key unknown in this region. Here we use recently developed field-scale monthly evapotranspiration (ET) data from OpenET to investigate water use dynamics over multiple years under various drought conditions. Water use patterns for soybeans, corn, cotton, rice, and surrounding forests are analyzed, with drought impacts assessed using the U.S. Drought Monitor (USDM). The results show that the water consumption for all four crops peaks in July, but early-season patterns differ by crop type and water use practices. Rice shows the highest average total ET for the growing season (688 mm), while cotton has the lowest (612 mm). In comparison, forest ET is higher than crop ET in the LMAP and also shows a seasonal peak in July. Crops show higher variability in ET than forests during the growing season, with larger differences in standard deviation in drought years. Across the three mid-drought years during the study period, groundwater consumption by the four major crops exceeded that of forests by an average of 10⁹ m³ per growing season. Anomalies in ET normalized by reference ET (fRET), a metric of evaporative stress, exhibited rapid response to drought as USDM drought severity intensified, demonstrating the potential of remote-sensing ET metrics for early drought detection. This study utilizes the OpenET dataset to analyze vegetation water use patterns at field scale (30 m), highlighting its value for detailed, spatially explicit monitoring of crop water dynamics and drought impacts and providing critical information for regional water accounting for the development of sustainable agriculture and effective water resources management in agriculture-intensive and drought impacted regions.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105172"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving fine-grained population distribution prediction by considering region-distinctive geographical factors-A case of Pearl River Delta, China","authors":"Ku Gao ,&nbsp;Xiaomei Yang ,&nbsp;Yueming Liu ,&nbsp;Qingyang Zhang ,&nbsp;Zhihua Wang","doi":"10.1016/j.jag.2026.105168","DOIUrl":"10.1016/j.jag.2026.105168","url":null,"abstract":"<div><div>Predicting fine-grained population distribution is crucial for effective urban planning. However, existing models widely ignore Region-Distinctive Geographical Factors (RDGF) in regional population modeling. This omission may compromise prediction accuracy, particularly in coastal zones where over 50% of the global population. To address this gap, we proposed an RDGF-incorporated approach for fine-grained population prediction, using the coastal Pearl River Delta as a case study. Leveraging multi-source geospatial data, based on generalized geographical factors (GGF) (e.g., topography, POI density, nighttime light intensity, etc.), we supplemented multi-dimensional RDGF including ecology, agriculture and transportation, etc. derived from unique regional environments (e.g., distance to shoreline, aquaculture, ports, etc.). We employed an interpretable machine learning framework (Random Forest + SHAP) to model and explain factor contribution. Results demonstrate: (1) incorporating RDGF substantially improves prediction accuracy in both model performance (with the average R² increasing by 6% under spatial cross-validation) and output (The relative error in densely populated areas can be reduced by up to 40%), thereby providing opportunity for more effective infrastructure planning and disaster risk management. (2) GGF still make the primary contribution to the model; however, RDGF are able to reveal local spatial heterogeneity and geographic decay patterns in population distribution, demonstrating greater potential for reducing prediction errors. This study provides region-specific insights for generating large-scale, fine-grained population map.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105168"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DeVCL: An end-to-end degradation-aware framework for vehicle counting and localization in satellite imagery","authors":"Ziqian Tan,&nbsp;Chen Wu","doi":"10.1016/j.jag.2026.105197","DOIUrl":"10.1016/j.jag.2026.105197","url":null,"abstract":"<div><div>Vehicle counting and localization using high-resolution satellite imagery have recently demonstrated substantial value in urban management and public services. However, satellite images routinely suffer from degradation issues, including blurring, insufficient resolution, noise, uneven illumination, and occlusion, due to sensor limitations, weather conditions, compression artifacts, and other environmental factors. These quality issues severely degrade the stability and accuracy of traditional vehicle counting and localization methods. To address this challenge, we propose Degradation-aware Vehicle Counting and Localization (DeVCL), a novel end-to-end point regression framework explicitly designed for degraded satellite imagery, which adaptively recognizes image degradation conditions and directly predicts vehicle positions. Specifically, DeVCL uses a self-supervised degradation representation jointly with image quality assessment to guide a degradation-aware feature modulation module, enhancing feature representations for low-quality inputs. We also introduce a feature-level adversarial mechanism without paired supervision to strengthen feature robustness. In addition, a density-sensitive feature refinement module is proposed to address matching ambiguities caused by densely packed and arranged vehicles, thus improving localization performance. We evaluated DeVCL using two synthetic degraded datasets built on FAIR1M_V and ITCVD, along with a newly collected dataset named SatPark that features high vehicle density and includes multiple naturally occurring degradations. Experimental results indicate that DeVCL consistently outperforms existing methods, particularly in SatPark, demonstrating strong generalization and practical adaptability.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105197"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147360638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GAGNN: a geography-aware graph neural network for citywide commuting flows prediction","authors":"Youjun Tu ,&nbsp;Peixiao Wang ,&nbsp;Julie N.Y. Zhu ,&nbsp;Zhiyuan Zhao ,&nbsp;Junli Li ,&nbsp;Sheng Wu","doi":"10.1016/j.jag.2026.105175","DOIUrl":"10.1016/j.jag.2026.105175","url":null,"abstract":"<div><div>Urban commuting flow prediction is crucial for optimizing public transportation and improving efficiency, yet traditional models often focus on geographic adjacency, overlooking the complex cross-regional interactions within transportation networks. To address this, we propose a <strong>G</strong>eography-<strong>A</strong>ware <strong>G</strong>raph <strong>N</strong>eural <strong>N</strong>etwork (GAGNN) model for commuting flow prediction. The model first jointly encodes the geographic adjacency matrix and semantic adjacency from public transportation networks, developing a comprehensive attention mechanism to fuse regional proximity with cross-regional semantic connectivity. Subsequently, a Graph Attention Network (GAT) is employed to embed the multiple adjacency relations and multi-source geographic knowledge. Finally, graph embeddings are combined with spatial factors into multidimensional feature vectors, fed into an MLP for commuting flow prediction. The model was validated with Fuzhou workday mobile phone data from January to February 2023, assessing the impact of semantic adjacency from different transportation networks on performance. The results show that: (1) We proposed the GAGNN outperforms both traditional models and advanced graph neural network models (e.g., GSGNN), reducing MAE by 14.9% and improving CPC by 2.1%; (2) The type of semantic adjacency significantly impacts model prediction accuracy. Road-based semantic connections perform best, especially for long-distance commuting flows, followed by metro and bus semantic connections, while the absence of semantic connections yields the worst performance. (3) Spatial scale significantly affects model prediction performance. Under road-based semantic adjacency, accuracy slightly declines with increasing scale, whereas metro, bus, and non-semantic connections, prediction accuracy improves. These findings offer effective support for accurate regional commuting flow modeling and public transportation networks optimization.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105175"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Arctic navigation risk assessment using ICESat-2 and GIS-Based AHP","authors":"Sang-Hoon Lee ,&nbsp;Hong-Sik Yun ,&nbsp;Seung-Jun Lee","doi":"10.1016/j.jag.2026.105150","DOIUrl":"10.1016/j.jag.2026.105150","url":null,"abstract":"<div><div>The Arctic Ocean is among the world’s most hazardous maritime regions, where sea ice, uncharted bathymetry, and limited emergency infrastructure threaten vessel safety. Recent advances in remote sensing now provide critical spatial data that can mitigate these risks. This study proposes a GIS-based framework for monthly Arctic navigation risk assessment by integrating sea ice, bathymetry, satellite communication, and accessibility to rescue and evacuation infrastructure. Using Spatial Multi-Criteria Evaluation (SMCE) with Analytic Hierarchy Process (AHP), risks were evaluated from September 2022 to April 2023 with ICESat-2 observations. Five vessel classes were analyzed, from non-icebreaking ships to those with maximum icebreaking capacities of 1.0 m, 1.5 m, 2.0 m, and 2.8 m. The resulting maps delineate monthly risk zones and reveal spatial and temporal variability across the Arctic. These outputs provide decision support for safer routing, operational preparedness, and policy development. The framework also demonstrates practical relevance for e-Navigation systems and advances methodological approaches to maritime risk assessment.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"147 ","pages":"Article 105150"},"PeriodicalIF":8.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147278291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}