The Landsat Surface Reflectance-derived normalized difference snow index (NDSI) are produced from Collection 2 (C2) Landsat 4-5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8-9 Operational Land Imagery (OLI)/Thermal Infrared Sensor (TIRS) scenes that successfully process to Landsat Collection 2 Level-2 Surface Reflectance products.
The C2 NDSI science product is the normalized difference between spectral bands green (G) and the shortwave infrared (SWIR). The NDSI is particularly useful for separating snow from vegetation, soils, and lithology endmembers. The NDSI uses the TM and ETM+ Bands 2 and 5, whereas for OLI the algorithm uses Bands 3 and 6:
(G – SWIR1) / (G + SWIR1)
In Landsat 4-7, NDSI = (Band 2 – Band 5) / (Band 2 + Band 5)
In Landsat 8-9, NDSI = (Band 3 – Band 6) / (Band 3 + Band 6)
The probability that snow is present is proportional to how close the NDSI pixel value is to 1 (range –1 to 1).
NDSI is delivered as a single band product, specified as shown in the table below.
Normalized Difference Snow Index
LC08SR_NDSI, LE07SR_NDSI, LT05SR_NDSI, or LT04SR_NDSI