STAR / SMCD / VIIRS Aerosol Calibration and Validation

VIIRS Aerosol Algorithm

The VIIRS aerosol algorithm retrieves aerosol optical thickness (AOT) from reflected solar radiation observed in multiple moderate resolution (750 m at nadir) VIIRS bands (M bands). The M bands used by the algorithm are listed in the Table on the right.

Because of the difference in our ability to characterize the surface reflection, and because of the difference in the properties of aerosols, separate algorithms are used over land and ocean. Both algorithms have their heritages in algorithms that were developed for the Moderate Resolution Imaging Spectroradiometer(MODIS) flown aboard the NASA Earth Observing System (EOS) satellites. The VIIRS aerosol algorithms are similar but not identical to the MODIS algorithms.

• The VIIRS over-land algorithm is based on the MODIS atmospheric correction algorithm [Vermote and Kotchenova, 2008]. In this retrieval AOT and aerosol models are byproducts of surface reflectance retrievals. The basis of the retrieval is the fact that aerosols change the ratios of spectral reflectances (spectral contrast) from those of the surface (aerosol-free) values. The VIIRS algorithm retrieves surface reflectances in bands M1, M2, M3, M5 and M11 from VIIRS-observed top-of-atmosphere (TOA) reflectances assuming progressively increasing values of AOT for five candidate aerosol models (dust, low and high absorbing smoke, clean and polluted urban aerosol), and compares their ratios in channels to their expected, aerosol-free values. The retrieved AOT and aerosol model are the ones that provide the best match between the retrieved and expected surface reflectance ratios in multiple channels. The aerosol-free expected ratios were derived empirically by retrieving surface reflectances from VIIRS-observed top-of-atmosphere (TOA) reflectances using AOT measured over 99 globally distributed AERONET sites with vegetation (~60,000 matchups).
• The VIIRS over-ocean algorithm is based on the MODIS over-ocean aerosol algorithm [Tanré et al., 1997]. The VIIRS algorithm searches for AOT and aerosol model that most closely reproduces the VIIRS-measured TOA reflectances in bands M5, M6, M7, M8, M10, and M11. In this process the TOA reflectance is calculated for increasing amounts of AOT for each of 2020 candidate aerosol models obtained by mixing four fine mode and five coarse mode models with an increment of 0.01 in the fine mode weight. Contribution of the ocean surface reflectance is calculated analytically accounting for wind speed and direction in the bi-directional specular reflection, wind-speed dependent Lambertian whitecap reflection, and Lambertian water-leaving radiance due to assumed, fixed pigment concentration.

In calculating reflectances for comparison with observed/derived reflectances the VIIRS algorithm accounts for molecular scattering, aerosol scattering and absorption, gas absorption, and surface reflection using the approach in the vector radiative transfer code Second Simulation of the Satellite Signal in the Solar Spectrum (6S-V1.1) [Kotchenova and Vermote, 2007].

Details of the algorithms can be found in the Joint Polar Satellite System (JPSS) - VIIRS Aerosol Optical Thickness (AOT) and Particle Size Parameter Algorithm Theoretical Basis Document (ATBD), and in Jackson et al., 2013.

Aerosol Particle size Parameter (APSP, Ångström Exponent) is derived from AOTs at M2 (445 nm) and M5 (672 nm) over land, and M7 (865 nm) and M10 (1610 nm) over ocean. (Note: the over-land product is not recommended for use!)

Suspended Matter (SM) is reported as one of the aerosol types (dust, smoke, sea salt) identified by the AOT algorithm above some threshold AOT. Volcanic ash type is inherited from the VIIRS Cloud Mask. Smoke concentration is calculated from the smoke AOT.

Retrievals are performed only day time from cloud-free pixels over dark land and non-sunglint ocean.

As the Cal/Val team evaluates the VIIRS aerosol product changes (including bug fixes) are implemented. The major changes made since the October 2011 launch of VIIRS on the Suomi NPP satellite are listed in the Algorithm changes page.

References

Kotchenova, S. Y., and E. F. Vermote (2007), Validation of a vector version of the 6S radiative transfer code for atmospheric correction of satellite data. Part II. Homogeneous Lambertian and anisotropic surfaces, Appl. Opt., 46(20), 4455-4464, doi:10.1364/ao.46.004455.

Tanré, D., Y. J. Kaufman, M. Herman, and S. Mattoo (1997), Remote sensing of aerosol properties over oceans using the MODIS/EOS spectral radiances, J. Geophys. Res., 102(D14), 16971-16988, doi:Doi 10.1029/96jd03437.

Vermote, E. F., and S. Kotchenova (2008), Atmospheric correction for the monitoring of land surfaces, J. Geophys. Res., 113(D23), doi:D23s9010.1029/2007jd009662.