Integrating spatiotemporal similarity for robust gap-filling in continuous surface water mapping with uncertainty quantification
文献类型：期刊论文
| 作者 | Xiao, Zhen2,4; Li, Runkui1,2; Song, Xianfeng1,2; Ding, Mingjun3; Liang, Shunlin4 |
| 刊名 | ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
 |
| 出版日期 | 2026-04-01 |
| 卷号 | 234页码:134-150 |
| 关键词 | Surface water Gap-filling framework Spatio-Temporal similarity Uncertainty quantification Landsat Remote sensing |
| ISSN号 | 0924-2716 |
| DOI | 10.1016/j.isprsjprs.2026.02.021 |
| 产权排序 | 3 |
| 文献子类 | Article |
| 英文摘要 | Continuous monitoring of surface water is critical for understanding Earth's hydrological cycle and managing water resources, yet it is severely hampered by data gaps in satellite imagery caused by cloud cover. Existing gapfilling methods are often limited by their reliance on single information sources and a lack of uncertainty assessment. To address these challenges, we introduce the Spatio-Temporal Gap-Filling (STGF) framework, which integrates inundation frequency, spatial similarity, and temporal similarity within a probabilistic structure. The framework makes three key contributions: (1) it quantifies the relative contributions of spatiotemporal drivers, (2) it introduces a pixel-wise confidence score to assess the reliability of filled data, and (3) it establishes a robust global performance benchmark. Through systematic validation across 838 diverse water bodies, the STGF framework achieves a relative mean absolute error (MAE) of 5.86%. In a direct comparison with established inundation-frequency-based methods, it demonstrates a 43.45% reduction in relative MAE. Critically, our analysis reveals that temporal similarity contributes more to accuracy than spatial similarity, especially under high data-loss conditions. The proposed confidence score shows a strong correlation with gap-filling accuracy, validating its utility for an operational quality control layer. The framework demonstrates robust performance across rivers, lakes, and reservoirs (relative MAE: 3.89%-7.22%), and reconstructed water area series show high correlations (Spearman's rho: 0.73-0.87) with independent water level and Sentinel-2 observations. This study presents an interpretable, open-source framework that advances our ability to produce reliable, high-frequency global surface water maps. The source code is openly available at https://code.earthengine.google.com/d2e4ae 14f76e46fd9d7e2db12f15b234. |
| URL标识 | 查看原文 |
| WOS关键词 | INLAND WATERS ; EMISSIONS ; RIVERS |
| WOS研究方向 | Physical Geography ; Geology ; Remote Sensing ; Imaging Science & Photographic Technology |
| 语种 | 英语 |
| WOS记录号 | WOS:001696750300001 |
| 出版者 | ELSEVIER |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/221336]  |
| 专题 | 资源与环境信息系统国家重点实验室_外文论文 |
| 通讯作者 | Li, Runkui |
| 作者单位 | 1.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Geog Informat Sci & Technol, Beijing 100101, Peoples R China; 2.Univ Chinese Acad Sci, Coll Resources & Environm, Beijing 100049, Peoples R China; 3.Jiangxi Normal Univ, Sch Geog & Environm, Nanchang 330022, Peoples R China 4.Univ Hong Kong, Dept Geog, Jockey Club STEM Lab Quantitat Remote Sensing, Hong Kong 999077, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Xiao, Zhen,Li, Runkui,Song, Xianfeng,et al. Integrating spatiotemporal similarity for robust gap-filling in continuous surface water mapping with uncertainty quantification[J]. ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING,2026,234:134-150. |
| APA | Xiao, Zhen,Li, Runkui,Song, Xianfeng,Ding, Mingjun,&Liang, Shunlin.(2026).Integrating spatiotemporal similarity for robust gap-filling in continuous surface water mapping with uncertainty quantification.ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING,234,134-150. |
| MLA | Xiao, Zhen,et al."Integrating spatiotemporal similarity for robust gap-filling in continuous surface water mapping with uncertainty quantification".ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING 234(2026):134-150. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。