Normalized Difference Water Index

Used to detect water bodies and monitor water content in vegetation. Positive values typically indicate water presence. Water indices exploit the strong absorption of shortwave infrared and near-infrared radiation by liquid water. They are critical for delineating water bodies, assessing moisture stress in vegetation, and monitoring hydrological changes over time. NDWI is particularly suited for water body mapping, flood monitoring, wetland assessment. The general formula is (Green - NIR) / (Green + NIR), which requires Green and NIR spectral bands.

NDWI is supported by 23 satellite sensors in our database, including BJ3A, BJ3N, Dragonette-1, Dragonette-2/3, Gaofen-1 and 18 more. Each sensor uses different band designations — for example, BJ3A uses the formula (Green - NIR) / (Green + NIR), while BJ3N uses (Green - NIR) / (Green + NIR). Select a sensor above to see its specific band mapping.

NDWI requires Green (510-580 nm), NIR (770-900 nm). Water absorbs strongly in the near-infrared and shortwave infrared portions of the spectrum, creating a measurable contrast with shorter wavelengths that penetrate the water surface.

Both Python and R code samples are provided above. In Python, use rasterio to load individual band GeoTIFF files and numpy for the arithmetic. In R, the terra package handles raster operations efficiently. The key is to load bands as floating-point arrays to avoid integer division, and to handle division-by-zero cases where the denominator equals zero. For production use, consider applying a valid data mask to exclude no-data pixels before calculation.

NDWI is designed to enhance water features, but dark surfaces like shadows, asphalt, and dark soils can produce similar values. For reliable water mapping, consider combining NDWI with a threshold analysis and, where possible, a secondary index to reduce false positives. Time-series analysis can also help distinguish permanent water bodies from temporary dark surfaces.