VIIRS NASA Land Surface Reflectance Product Validation

(November 2019)

Version 1 Validation Statement

Validation at stage 3 has been achieved for the VIIRS land surface reflectance product.

The accuracy of the VIIRS surface reflectance product in all bands is at or better than 0.005±0.05SR (where SR is the surface reflectance) except for bands M1, M2 and M3 which are the blue bands used in the aerosol inversion. The accuracy of the derived vegetation indices [normalized difference vegetation index (NDVI) and enhanced VI (EVI)] is 0.02±0.02VI. Note that this statement only applies to normal observation conditions (no clouds, cloud shadows, or high aerosol) which can be determined by inspecting the product quality flags.

The SR product performance was determined over the AErosol RObotic NETwork (AERONET) sites during 2018 (see the Accuracy, Precision and Uncertainties, plots below in Figures 1a and 1b). The stability and consistency of the VIIRS reflectance and VI with MODIS Aqua is excellent (see Figure 2 for Red/NIR, and Skakun et al. 2018, below).


LSR bands1-7 APU graphic

Figure 1a: Performance of the VIIRS surface reflectance product for 2018 over the AERONET sites (bands M1-M7). The specification is met for band M4, M5 and M7, M1-3 (blue) are aerosol retrieval bands and are not expected to meet specification. Click on graphic to see larger version.


LSR M8-11 & VI APU graphic

Figure 1b: Same as figure 2a but for band M8, M10, M11 and EVI and NDVI.. Click on graphic to see larger version.

 

VIIRS comparison with MODIS Aqua and Terra

Figure 2. Automated monthly VIIRS surface reflectance cross-comparison with MODIS Aqua from 2012 to 2019. The stability of both VIIRS and MODIS Aqua is excellent in both red and NIR as shown (± 0.5%). An update in the VIIRS calibration methodology (~1%) is visible around 2018 that should disappear once collection 2 is produced.

 

Support Studies

Skakun, S., Justice, C.O., Vermote, E. and Roger, J.C., 2018. Transitioning from MODIS to VIIRS: an analysis of inter-consistency of    NDVI data sets for agricultural monitoring. International journal of remote sensing, 39(4), pp.971-992.

Vermote E., Justice C. O. and Csiszar I., Early evaluation of the VIIRS calibration, cloud mask and surface reflectance Earth data    records, Remote Sensing of Environment, Volume 148, 25 May 2014, Pages 134-145, ISSN 0034-4257,
   http://dx.doi.org/10.1016/j.rse.2014.03.028.