Snow Cover Detection {Methodology Using Landsat}
NASA MODIS Land February 2004
Calculated using a ratio of two bands captured and composed in a satellite image for a specific location and time. NDSI is a measure of the relative magnitude of the reflectance difference between visible (green) and short wave infrared (SWIR).
An important role of the NDSI is snow and cloud discrimination. For Landsat Thematic Mapper bandwidths snow is highly reflective in the visible band 2 but not reflective in the SWIR, therefore this ratio will allow snow to be detected apart from clouds at a given threshold. A physically based threshold of 0.4 as found by NASA for MODIS snow cover calculations is also used for the Landsat imagery analysis and is validated by visual analysis.
Normalized Difference Snow Index
Landsat
NDSI = (TM Band 2 - TM Band 5) / (TM Band 2 + TM Band 5)
Band 2: Visible Green, 0.53 - 0.61 micrometers
Band 5: Short Wave Infrared, 1.55 - 1.75 micrometers
Initial NDSI mask before masks are applied.
Aqua satellite
Source: NASA Earth Observatory
Landsat 4 Thematic Mapper, July 1982 - December 2001
Landsat 5 Thematic Mapper, March 1984 - January 2013
Orbit: 705 km altitude, sun-synchronous, near polar, 16 days for repeat cycle.
Bands: 7 bands across bandwidths from 0.45 to 2.35 micrometers.
Spatial Resolution: 30 meters, 120 meters for Thermal (Band 6)
Data Products: Level 1 processed, the map projection is UTM WGS 1984, 30 meter pixels, GeoTiff format
Landsat 4, 5
Aboard both Terra and Aqua satellites.
Orbit: 705 km altitude, sun-synchronous, near polar, 1 - 2 days for repeat cycle.
Bands: 36 bands across bandwidths from 620 nanometers to 14.385 micrometers.
Spatial Resolution: 250 meters (Bands 1-2), 500 meters (Bands 3-7), 1000 meters (Bands 8-36)
Snow cover data products: Spatial Resolution 500 meters
The overall absolute accuracy of the 500-m resolution products is ~93%