Vegetation-shadow indices based on differences in effect of atmospheric-path radiation between optical bands

Abstract

The fractional coverage of the illuminated vegetation (IV), illuminated soil (IS), shaded vegetation (SV), and shaded soil (SS) are four essential structural parameters (fIV, fIS, fSV, and fSS) describing the vegetation canopy. Their correct estimation can help improve the monitoring of vegetation growth. In this study, we propose and evaluate four coastal/aerosol-band-based vegetation-shadow indices (CVSIs) using the normalized difference of top-of-atmosphere (TOA) reflectance in the coastal/aerosol band and in the blue, green, red, and near-infrared (NIR) bands. We used the four CVSIs to evaluate the illumination conditions of the vegetation-soil canopy for two cases: shadows produced by vegetation and shadows produced by mountains. We also evaluated the combined use of linear spectral mixture analysis (LSMA), CVSIs, and the normalized difference vegetation index (NDVI) to estimate fIV, fIS, fSV, and fSS. The results led to several conclusions: (i) the effects of atmospheric path radiation mainly influence the optical bands, resulting in the coastal/aerosol-band TOA reflectance exceeding the TOA reflectance of the blue, green, red, and near-infrared bands; (ii) CVSIs can be used to evaluate the illumination conditions of the vegetation-soil canopy; (iii) the proposed strategy produces reasonably accurate estimates of fIV, fIS, fSV, and fSS (n = 191,268; R2 = 0.830–0.981; mean absolute error = 0.053–0.098; root mean square error = 0.063–0.126). Maps of fIV, fIS, fSV, and fSS are consistent with maps of the actual field illumination conditions.

Keywords

• Shadows
• TOA reflectance
• Surface reflectance
• Optical remote sensing
• Vegetation-shadow indices
• Illumination conditions