VIIRS NASA Snow Cover Product Validation

(Reviewed February 2022)

Version 1 Validation Statement

Validation at stage 1 has been achieved for the NASA VIIRS snow cover (VNP10) product.

The VIIRS snow-cover product has been compared to snow-cover extent (SCE) and Normalized Difference Snow Index (NDSI) with both Terra and Aqua MODIS NDSI-derived SCE for various areas of North America. The accuracy of the MODIS snow-cover maps is well established (Hall et al., 2007). VIIRS NDSI snow maps accurately represent the extent of snow cover at regional scales (Riggs et al., 2017).

In a study area that includes the Wind River Range in Wyoming, the agreement between MODIS Collection 6 and VIIRS Collection 1 NDSI-derived SCE maps is 97 percent. The small differences in the location of snow cover mapped by MODIS and VIIRS are due to differences in the sensors, viewing geometry and acquisition times.

The images in the figure below show snow-cover maps where NDSI values over 0.1 signify snow cover; the deep purple represents the lowest NDSI values and white represents the highest NDSI and thus the most extensive snow cover. The left image is MODIS, the center image is VIIRS. The MODIS-VIIRS difference map is shown on the right. The small differences in the location of snow cover mapped by MODIS and VIIRS are due to differences in the sensors, viewing geometry and acquisition-times.


MODIS and VIIRS NDSI snow maps focusing on the Wind River Range, Wyoming on 2 December 2016. The light gray shows cloud masking in the difference map
(using the MODIS and VIIRS cloud masks combined) and snow-free land.

NDSI legend

In the Lower Peninsula of Michigan, comparisons between Aqua MODIS and VIIRS NDSI snow maps in the image below show an average accuracy of the VIIRS NDSI of ±0.05 - 0.10 NDSI for three different dates (see table below).


difference legend

Comparison of MODIS and VIIRS NDSI maps of the Lower Peninsula of Michigan for 11 Mar 2015. NDSI values from 0.1 - 1.0 represent snow cover. NDSI values of <0.1 are shown in green and are generally dense forests and/or cities. Aqua MODIS NDSI values are often slightly lower than VIIRS NDSI values. The left panel is an Aqua MODIS NDSI cloud-cleared map (MYD10A) representing daily snow cover; images were acquired between 17:35 and 21:00 UTC. The center panel is the VIIRS NDSI cloud-cleared map representing daily snow cover, acquired at 18:00 UTC. The right panel is the MODIS-VIIRS difference map.

There is very little difference in the Aqua and VIIRS NDSI snow maps. Mean NDSI values are provided below, from within a box located in west-central Michigan that is ~5000 km2 (location of the box is shown in center of the difference map above). A negative difference indicates the mean Aqua MODIS NDSI is lower than the mean VIIRS NDSI.

Date Aqua MODIS NDSI VIIRS NDSI Difference
11 Mar 2015 0.451 0.583 -0.132
27 Feb 2017 0.271 0.266 +0.005
04 Mar 2017 0.481 0.537 -0.056


Support Studies

Riggs, G.A. and D.K. Hall, 2020: Continuity of MODIS and VIIRS snow cover extent data products for development of an Earth
    Science Data Record, Remote Sensing, 12:3781.

Dorothy K. Hall, George A. Riggs, Nicolo E. DiGirolamo, and Miguel O. Román, 2019, Evaluation of MODIS and VIIRS cloud-gap-
   filled snow-cover products for production of an Earth science data record, Hydrol. Earth Syst. Sci., 23, 5227–52

Hall, D.K. and G.A. Riggs, 2007: Accuracy assessment of the MODIS snow-cover products, Hydrological Processes,
   21(12):1534-1547, https://doi:10.1002/hyp.6715.

Riggs, G.A., D.K. Hall and M.O. Román, 2017: Overview of NASA's MODIS and Visible Infrared Imaging Radiometer Suite (VIIRS)    snow-cover Earth System Data Records, Earth System Science Data, 9:1-13,