The retrieval of shrub fractional cover based on a geometric-optical model in combination with linear spectral mixture analysis

Vegetation fractional cover, which defines the amount of vegetation on the surface of the land, is a key parameter in land surface models. Based on a geometric-optical model in combination with a linear spectral mixture analysis, the retrieval of shrub fractional cover in Wushen Banner of Inner Mongolia in the Mu Us Sandland using HJ-1B multispectral images is discussed. We acquired the surface reflectance based on geometric correction and atmospheric correction of the HJ-1B image. Then we assumed that the reflectance of a mixed pixel is a simple linear combination of two components, namely illuminated background and illuminated canopy, and further calculated the areal proportion of the illuminated background within each pixel based on the linear spectral mixture analysis. Then, combined with the measured shrub structural parameters, the shrub fractional cover was estimated using an inverted geometric-optical model. Finally, the result was validated through the measured shrub cover of 13 sample plots and a comparison study was done with the NDVI regression method and simple linear spectral mixture analysis. The R-2 of the three methods are 0.898, 0.614, and 0.659, with corresponding root-mean-squared errors of 0.136, 0.154, and 0.175, which indicate the reliability of the combined method.