IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing最新文献

{"title":"Editorial Foreword to the Special Issue on Advances in Remote Sensing of Mountain Surface Processes","authors":"Li Peng;Wei Zhao;Gaofei Yin;Kai Yan;Tianjun Wu","doi":"10.1109/JSTARS.2025.3576138","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3576138","url":null,"abstract":"Mountains play a crucial role in providing essential ecosystem services, including biodiversity conservation, regional climate regulation, and water resource preservation. Accurate and continuous monitoring of mountain surface processes is therefore vital for assessing environmental change and mitigating natural hazards. Remote sensing has emerged as a powerful tool for mountain research, yet the inherent complexity of these processes demands increasingly higher spatio-temporal resolution in observations. Against this backdrop, we launched the special issue “Advances in Remote Sensing of Mountain Surface Processes”. This initiative aims to systematically showcase cuttingedge research findings and innovative technologies from the global remote sensing community, specifically tailored to the complex terrain of mountainous environments. The special issue collects 13 articles and highlights a multidimensional exploration of mountain surface processes. The featured manuscripts cover a wide range of topics, spanning multi-source data fusion, methodological innovations, and diverse applications across various fields. This Editorial Foreword provides a comprehensive summary of theoretical advancements, methodological innovations, validation techniques, and practical applications of remote sensing in studying mountain surface processes. It also offers a thorough discussion of the challenges and future opportunities within this dynamic field.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14495-14500"},"PeriodicalIF":4.7,"publicationDate":"2025-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11031222","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A General Model for Large-Scale Paddy Rice Mapping by Combining Biological Characteristics, Deep Learning, and Multisource Remote Sensing Data","authors":"Zhenjie Liu;Jialin Liu;Yingyue Su;Xiangming Xiao;Jingwei Dong;Luo Liu","doi":"10.1109/JSTARS.2025.3573750","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3573750","url":null,"abstract":"Due to the influences combined with global climate change and human activity, paddy rice area and distribution have undergone dramatic changes. Currently, many approaches for paddy rice mapping rely on the prior knowledge of paddy rice phenology or require widely distributed ground samples of paddy rice, which are limited for large-scale applications. In this work, we propose a general paddy rice mapping (GPRM) model by combining biological characteristics, deep learning, and multisource remote sensing data. The proposed GPRM first utilizes the normalized difference vegetation index and land surface water index to acquire large-scale remote sensing dataset in key phenology periods of paddy rice, such as the transplanting period and peak vegetative growth period. Then, a general model using object-based deep neural networks is developed and trained by the remote sensing dataset and the ground reference data collected in one region (e.g., Guangdong Province), which can be directly applied for generating 10-m paddy rice maps in other regions with different climate conditions and complex cropping systems (e.g., Jiangxi Province and Heilongjiang Province). The results demonstrate that the GPRM can realize remarkable performance of paddy rice mapping in China. The overall accuracies are over 99%, and the user accuracy, producer accuracy, and Kappa coefficient vary from 0.77 to 0.93, 0.94 to 0.97, 0.9 to 0.95, respectively. Overall, the GPRM is has significant promise for large-scale paddy rice mapping with complex cropping systems, thus supporting global agricultural development strategies and food security.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14705-14717"},"PeriodicalIF":4.7,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11031092","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144323232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CADFormer: Fine-Grained Cross-Modal Alignment and Decoding Transformer for Referring Remote Sensing Image Segmentation","authors":"Maofu Liu;Xin Jiang;Xiaokang Zhang","doi":"10.1109/JSTARS.2025.3576595","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3576595","url":null,"abstract":"Referring remote sensing image segmentation (RRSIS) is a challenging task, aiming to segment specific target objects in remote sensing images based on a given language expression. Existing RRSIS methods typically employ coarse-grained unidirectional alignment approaches to obtain multimodal features, and they often overlook the critical role of language features as contextual information during the decoding process. Consequently, these methods exhibit weak object-level correspondence between visual and language features, leading to incomplete or erroneous predicted masks, especially when handling complex expressions and intricate remote sensing image scenes. To address these challenges, we propose a fine-grained cross-modal alignment and decoding Transformer, CADFormer, for RRSIS. Specifically, we design a semantic mutual guidance alignment module (SMGAM) to achieve both vision-to-language and language-to-vision alignment, enabling comprehensive integration of visual and textual features for fine-grained cross-modal alignment. Furthermore, a textual-enhanced cross-modal decoder (TCMD) is introduced to incorporate language features during decoding, using refined textual information as context to enhance the relationship between cross-modal features. To thoroughly evaluate the performance of CADFormer, especially for inconspicuous targets in complex scenes, we constructed a new RRSIS dataset, called RRSIS-HR, which includes larger high-resolution remote sensing image patches and semantically richer language expressions. Extensive experiments on the RRSIS-HR dataset and the popular RRSIS-D dataset demonstrate the effectiveness and superiority of CADFormer.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14557-14569"},"PeriodicalIF":4.7,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11023843","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144323152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Multicomponent Collaborative Fossil Fuel Power Plants Detection Framework Based on Geographic Analysis in Wide Areas","authors":"Ning Li;Min Jing;Wanxuan Geng;Shengkun Dongye;Hui Chen;Chen Ji;Liang Cheng","doi":"10.1109/JSTARS.2025.3573758","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3573758","url":null,"abstract":"Fossil fuel power plants (FFPPs) are major sources of carbon dioxide emissions in the power industry. Accurately locating these plants is essential for monitoring emissions, studying atmospheric pollution, and optimizing power supply structures. However, obtaining comprehensive geographic location data for FFPPs is challenging due to data availability and collection constraints. Therefore, we propose a wide-area FFPP detection framework that enhances detection efficiency through geographic constraints and improves detection accuracy using a multicomponent collaborative strategy. First, a geographic constraint method was developed, leveraging multisource geographic data to extract candidate FFPP regions based on their spatial characteristics. Next, we constructed a comprehensive FFPP dataset, including plants and their components, and trained two separate object detection models for FFPPs and their components. Subsequently, the FFPP model was used to perform coarse detection, followed by the refined detection of primary features (chimneys, square chimneys, and cooling towers) and auxiliary features (substations and storage tanks). After detecting these objects, the density-based spatial clustering of applications with noise clustering algorithm was applied to retain clusters with specific component combinations, yielding the final detection results. In the approximately 660 000-km² study area (Jiangsu Province, São Paulo, and Maharashtra), the proposed framework effectively minimized invalid regions by 94.8%, 91.12%, and 97.1%, respectively. Validation using high-resolution Google Earth images recalled 225 known FFPPs with a 91.46% recall rate and identified 167 previously unrecorded FFPPs. These results demonstrate the framework’s reliability for efficient and automated FFPP detection, representing a novel integration of multisource geographic analysis, deep-learning-based object detection, and wide-area FFPP recognition.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"13880-13894"},"PeriodicalIF":4.7,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11023143","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144264208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Method to Weaken Cloud Interference in Solar-Induced Chlorophyll Fluorescence (SIF) Reconstruction by Using Satellite VOD Observations","authors":"Jiajia Ding;Haiqiu Liu;Kai Zhang;Linyu Li","doi":"10.1109/JSTARS.2025.3576504","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3576504","url":null,"abstract":"Solar-induced chlorophyll fluorescence (SIF) satellite observations enable large-scale crop monitoring and yield assessment. Some optical vegetation indexes have been commonly used as predictors to reconstruct SIF. However, satellite optical vegetation indexes observations are highly susceptible to clouds, leading to degradations of EVI-based SIF reconstruction in cloud-covered situations. Unlike optical vegetation indexes, vegetation optical depth (VOD) can penetrate clouds and is highly sensitive to the changes in vegetation internal water. This study aims to investigate the potentials of VOD in reducing cloud-induced SIF reconstruction performance loss. First, a VOD-based model is established based on a dataset containing Global Ozone Monitoring Experiment-2 SIF, daily MODIS normalized bidirectional reflectance, land surface temperature, photosynthetically active radiation, and VOD data in 2015–2017. Second, comparisons between the VOD-based model and the non-VOD model are performed, and results suggest that as cloudage rises from 10% to 90%, the VOD-based SIF model reduces cloud-induced performance loss by 62% over the non-VOD model, proving that the introducing of VOD is effective in reducing cloud-induced SIF reconstruction performance loss, particularly under heavy cloudage. Finally, comparisons between the VOD-based model and the EVI-based model are performed, and results show that, in general, the VOD-based model mitigates the cloud-induced degradations in SIF reconstruction by 40% over the EVI-based model. But, under the cloudage less than 53.7%, the EVI-based model is recommended for easy access to higher-resolution optical vegetation indexes observations, and under the cloudage exceeding 53.7%, the VOD-based model is strongly recommended for its advantages in reducing cloud-induced degradation in SIF reconstruction.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14535-14544"},"PeriodicalIF":4.7,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11023846","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144308536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Elliptic Kernel Unsupervised Autoencoder—Graph Convolutional Network Ensemble Model for Hyperspectral Unmixing","authors":"Estefanía Alfaro-Mejía;Carlos J. Delgado;Vidya Manian","doi":"10.1109/JSTARS.2025.3576281","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3576281","url":null,"abstract":"Spectral unmixing is an important technique in remote sensing for analyzing hyperspectral images to identify endmembers and estimate fractional abundance maps. Over the past few decades, significant progress has been made in deep learning methods for endmember extraction and abundance estimation. This article introduces the autoencoder graph ensemble model (AEGEM), a novel ensemble-based framework designed to enhance performance in both endmember extraction and abundance estimation. In the initial stage, endmember extraction and abundance map estimation are carried out using a convolutional autoencoder. An elliptical kernel is then applied to compute spectral distances and generate an adjacency matrix based on elliptical neighborhoods. This information is used to construct an elliptical graph, where centroids serve as senders and surrounding pixels as receivers. A graph convolutional network (GCN) processes stacked input-abundance maps, senders, and receivers to refine the abundance estimations. Finally, an ensemble decision-making strategy selects the optimal abundance maps based on the root-mean-square error metric. The effectiveness of AEGEM is evaluated on benchmark datasets, including Samson, Jasper, and Urban, with additional performance validation on the Cuprite dataset. Experimental results demonstrate that AEGEM outperforms baseline algorithms in both endmember extraction and abundance estimation, particularly in complex and spectrally mixed scenarios.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14594-14614"},"PeriodicalIF":4.7,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11021649","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144323189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"WinMRSI: Feature Matching With Window Attention for Multimodal Remote Sensing Image","authors":"Yide Di;Yun Liao;Yunan Liu;Hao Zhou;Kaijun Zhu;Mingyu Lu;Qing Duan;Junhui Liu","doi":"10.1109/JSTARS.2025.3576233","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3576233","url":null,"abstract":"Multimodal remote sensing image matching is a crucial task with broad application potential. However, substantial nonlinear radiometric differences between multimodal images pose significant challenges, often leading to mismatches. To tackle these challenges, this article introduces WinMRSI, a window attention-based multimodal remote sensing image matching method designed to enhance cross-modal feature extraction and information interaction. For feature extraction, a siamese network with discrete cosine transform is employed to model inter-channel dependencies and extract multiscale features from cross-modal images. In addition, a dual-branch network is designed to capture contextual dependencies while refining local feature representations. For information interaction, WinMRSI integrates a window attention mechanism to strengthen fine-grained feature fusion within highly relevant windows, enabling the model to focus on discriminative regions. Furthermore, a multilevel matching module progressively refines matching accuracy in a coarse-to-fine manner across window, patch, and pixel levels. Extensive evaluations on benchmark datasets demonstrate that WinMRSI achieves state-of-the-art performance in multimodal remote sensing image matching. Ablation studies further validate the effectiveness of each component in WinMRSI.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14615-14629"},"PeriodicalIF":4.7,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11022727","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144323145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolution of GNSS-Derived Precipitable Water Vapor and Its Driving Factors During the “Dragon Boat Water” Rainfall Event in Guangxi, China","authors":"Ling Xiong;Liangke Huang;Zhixiang Mo;Xiangping Chen;Yifei Yang;Shaofeng Xie;Junyu Li;Lilong Liu","doi":"10.1109/JSTARS.2025.3575758","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3575758","url":null,"abstract":"Precipitable water vapor (PWV) is an important indicator for quantifying atmospheric water vapor, and its evolution is intrinsically linked to the formation and development of extreme weather. As a type of heavy rainfall occurring during the Dragon Boat Festival in southern China, the phenomenon known as “dragon boat water” (DBW) has caused a series of disasters in Guangxi Province. Therefore, it is crucial to investigate PWV evolution and contributing factors. The PWV retrieved by Global Navigation Satellite System (GNSS) technology offers a reliable method due to its advantages of high-temporal resolution, high-precision, and weather-independent. In this article, the GNSS-derived PWV at the stations was obtained based on the GNSS data from 2020 to 2022 at 121 GNSS stations as well as the European Centre for Medium-Range Weather Forecasts Reanalysis 5 (ERA5) atmospheric reanalysis data. To assess the performance of the GNSS-derived PWV, it was compared with Radiosondes, demonstrating its high accuracy. Additionally, combined with ERA5 reanalysis data, the evolution of GNSS-retrieval PWV during the DBW event in Guangxi was analyzed, and the results showed that the direction of the water vapor transport pathway was from the southeast to the northwest. The overall trend of PWV decreased from southeast to northwest, with higher values observed in coastal areas compared to inland areas, and greater concentrations in plains than in mountains. Further investigation revealed that the evolution of GNSS-derived PWV was governed by the synergistic effects of mean sea level pressure (MSLP), horizontal wind speed (u-wind), vertical wind speed (v-wind), and 2 m temperature (T2M). The direction of the wind field generally aligned with the direction of PWV movement, and the magnitude of PWV corresponded with wind field intensity. The distribution of PWV was found to be negatively correlated with MSLP and positively correlated with T2M. These findings could deepen the understanding of PWV dynamics and improve the prediction of extreme precipitation events.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14308-14323"},"PeriodicalIF":4.7,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11021307","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144299249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatiotemporal Change Detection of Ecological Quality and the Associated Influencing Factors in Yuzhou Mining Area With RSEIFE","authors":"Jiao Pan;Ziwei Wang;Tao Chen;Kun Jia;Antonio Plaza","doi":"10.1109/JSTARS.2025.3575810","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3575810","url":null,"abstract":"With the increasing attention to the construction of ecological civilization, sustainable development has been upgraded to green development, and the problem of ecological deterioration has been solved at the source of resource utilization. Ecological quality and its potential influencing factors are important elements of ecological security in Yuzhou mining area. In this work, we examined the geographical and temporal variations of ecological environmental quality within the Yuzhou Mining region using the remote sensing ecological index considering full elements (RSEIFEs). The method exploited Savitzky-Golay filtering (to smooth the four ecological index data), and the surface ecological effects of resource-based cities were evaluated using the entropy weight method and moving window technique. Our key findings include: 1) the filtered ecological index time series curves effectively reduce the data noise and make the data set smoother; 2) during mining time series monitoring, the mean RSEIFE values for five years were 0.56, 0.41, 0.44, 0.58, and 0.67, demonstrating an initial deterioration, subsequently followed by an enhancement in the ecological environmental condition; 3) the dominant index impacting the ecological environment of the open-pit mine are land surface temperature, and normalized difference built-up and bareness index; 4) while open-pit mining activities adversely affect the ecological environment, restoration efforts significantly improve conditions in both the open-pit and its surrounding areas. This study offers a quantitative method for monitoring the ecological environmental quality of resource-based cities and provides essential scientific data for the ecological environment restoration of mining areas and their surroundings.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14545-14556"},"PeriodicalIF":4.7,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11021284","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144323156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine Learning Model for Predicting Global Ionospheric TEC Maps Based on Constraint Conditions","authors":"Qingfeng Li;Hanxian Fang;Chao Xiao;Die Duan;Hongtao Huang;Ganming Ren","doi":"10.1109/JSTARS.2025.3575693","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3575693","url":null,"abstract":"The prediction and modeling of ionospheric total electron content (TEC) have consistently been a focal point for researchers, as it holds significant implications for satellite positioning, navigation, telemetry, control, and radio wave propagation. In this context, we propose a machine learning prediction model [predictive GAN variational autoencoder-label (PGVAE-label)] using a labeled graph of image segmentation as a constraint to predict the global ionospheric TEC. We use IGS TEC maps from 2003 to 2018 as training, test, and validation sets, respectively. Subsequently, we conducted comparative experiments using the unlabeled machine learning prediction model (PGVAE) and the one-day and two-day forecast maps published by the Center for Orbit Determination in Europe (CODE). In addition, the article analyzes the effect of predictions during the periods of geomagnetic quiet and disturbance, high solar activity years, and low solar activity years. The results show that the PGVAE-label model has superior TEC prediction capability, producing TEC prediction maps with the lowest average root-mean-square error values of 1.79, 1.80, and 1.83 TECU, and that the PGVAE-label model is also superior to the PGVAE and CODE models in the region of the peak ionospheric structure. The predictive ability of the PGVAE-label model is better in geomagnetically quiet periods than in geomagnetically disturbed periods, and better in solar low years than in solar high years. The work in this article provides new ideas and thoughts on the application of deep learning to the broader field of Earth sciences, particularly in the problem of prediction.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"14454-14466"},"PeriodicalIF":4.7,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11020806","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144299062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}