Mapping Surface Water Bodies based Water Indices Extracted from Sentinel 2 Images
| Meeting | VIWC2026 |
| Topic | OS3: Water Environment and Mining |
| Author | Thi Ngoc TRAN | Thi Hoa TRAN |
| Organization | Lecturer Hanoi University of Mining and Geology (HUMG), Faculty of Environment (FoE), Department of Environment Engineering | Instructor Hanoi University of Mining and Geology (HUMG), IT Faculty, Department of Geoinformatics |
| DOI | 2022.1669935982 |
Abstract
Remote sensing technology has taken many advantages in water and hydrological systems management, such as: accurately determining water bodies; or providing up-to-date spatiotemporal changes of stream flows, water quantity, and water quality. Theoretically, based on the significant spectral signature of water in the visible to shortwave infrared wavelengths, it is feasible to distinguish water bodies from other phenomena on satellite images like vegetation, bare soils, clouds, or even shadows. However, due to restrictions of spatial resolutions and capabilities of satellite sensors, establishing threshold values to separate water bodies from the background may be challenging. To minimized the effects of environmental background noises, there are some indices investigated in extracting information about water resources such as the Normalized Difference Water Index (NDWI), the Normalized Difference Vegetation Index (NDVI), the Land Surface Water Index (LSWI), the Modified Normalized Difference Water Index (MNDWI), etc. This study examines performances of some widely used water indices derived from a freely remotely sensed data – the Sentinel 2 for mapping water bodies. Results are expected to support a process of automatic extraction water bodies from imagery scenes by comparing and recommending possible threshold values and indices in a particular case study.
Keywords: hydrological systems management, remote sensing, mapping, Sentinel 2.
