| 摘要: |
| 基于遥感绿光波段和近红外波段对地表水分反射率的差异，探索使用归一化水体指数（NDWI）及扩展指数监测地表水分变化。利用2018至2022年的多时相Sentinel-2光学影像，比较NDWI阈值为0和OTSU自适应阈值在水体提取中的效果，提出了时序二值化NDWI差值法（NDWIb-del）用于识别水体转换，以及归一化NDWI差值法（NNDWIdel）用于监测水分变化，并以宿鸭湖大型水库为例进行分析。结果表明：NDWI阈值为0的提取效果优于OTSU阈值，后者有效期不足50%。水体分布与清淤扩容动态密切相关，2018年主要是西北部和西岸水体的自然变化，施工后，2019年库西水体减少，2020至2021年，深水带与一号人工岛附近及坝后排泥区水体面积增加；2022年，二号至三号人工岛附近水体持续增加。NDWIb-del方法显示，2019至2022年间，施工填筑区、深水带区与坝后堆泥区有明显的水体转换。NNDWIdel用于反映全局地表水分变化，值为0.5表明无变化，0或1表示极干或极湿变化。2019年至2022年，极端变化对应水体与非水体的转换，与NDWIb-del结果一致。适中变干和适中变湿对应极端变化的前序阶段，包括土方填筑导致的水体减少和淤泥堆放导致的水体聚积等不明显转换区。本研究扩展了遥感NDWI指数的应用，为地表水分和环境变化的自动高效监测提供了新方法，在水分含量、旱情、水质及浊度监测具有潜力。 |
| 关键词: 光学遥感 NDWI NDWIb-del NNDWIdel 地表水分 宿鸭湖水库 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学青年(52109026),国家重点研发计划课题(2021YFB3900603),国家联合(U2240217),水灾害防御全国重点实验室自主研究项目(5240152N2) |
|
| Surface Water Change Monitoring Based on NDWI Index and Its Extended Indices: A Case Study of Suyahu Reservoir |
|
LIU Suyi1, HUANG Dui2, Wang Jianjun3, YU Zhongbo1
|
|
1.:Hohai University,State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering;2.Nanjing Hydraulic Research Institute;3.Zhumadian City Suyahu Reservoir Operation Center
|
| Abstract: |
| This study investigates the use of the Normalized Difference Water Index (NDWI) and its extended methods for monitoring surface water changes, leveraging differences in surface water reflectance between green and near-infrared bands in remote sensing. Utilizing Sentinel-2 images from 2018 to 2022, the research examines NDWI with a threshold of 0 compared to the OTSU adaptive threshold for water body extraction. It proposes a time-series binary NDWI difference method (NDWIb-del) to identify conversions between water and non-water bodies and a normalized NDWI difference method (NNDWIdel) to track surface water changes, using the Suyahu Reservoir as a case study.The results show that NDWI with 0 threshold is more effective than the OTSU threshold, which is effective less than half of the time. Water body changes are linked to dredging and expansion projects, with notable shifts occurring around artificial islands and silt discharge areas. From 2019 to 2022, conversions mainly happened in construction fill areas and silt stacking zones. The NNDWIdel method indicates surface water changes, with index levels from 0 (extreme dryness) to 1 (extreme wetness), matching NDWIb-del findings. The study provides insights for enhancing NDWI use in environmental monitoring, including changes in water content, drought, water quality, and turbidity. |
| Key words: Optical Remote Sensing NDWI NDWIb-del NNDWIdel Surface Water Suyahu Reservoir |
