Joint Characterization of the Vegetation by Satellite Observations From Visible to Microwave Wavelengths: A Sensitivity Analysis. Characterization of the Vegetation by Satellite Observations

摘要

This study presents an evaluation and comparison of visible, near-infrared, passive and active microwave observations for vegetation characterization, on a global basis, for a year, with spatial resolution compatible with climatological studies (0.25 deg x 0.25 deg). Visible and near-infrared observations along with the NDVI come from AVHRR. An atlas of monthly-mean microwave land surface emissivities between 19 and 85 GHz has been calculated from SSM/I observations for a year, suppressing the atmospheric contamination problems encountered with the use of simple channel combinations. The active microwave measurements are provided by the ERS-1 scatterometer backscattering coefficients at 5.25 GHz. The capacity to discriminate between vegetation types and to detect the vegetation phenology is assessed in the context of a vegetation classification obtained from in situ observations. A clustering technique derived from the Kohonen topological maps is used to merge the three data sets and interpret their relative variations with surface characteristics. NDVI varies with vegetation density, but saturates for and grassland and for forested areas. Spurious seasonal cycles and large spatial variability in several areas suggest that atmospheric contamination (cloud, water vapor, ozone, and aerosols) and/or solar zenith angle drift still modulate the NDVI signal. Passive and active microwave observations are primary sensitive to overall vegetation structure: They respond to absorption, emission and scattering by vegetation elements, including woody parts. Backscattering coefficients from ERS-1, measured with an accuracy of 5%, are not sensItive to atmospheric variations and exhibit good potential for vegetation discrimination with approximately 10 dB dynamic from rain forest to arid grassland. Passive microwave measurements also show some ability to characterize vegetation, but are less sensitive than active measurements. However, passive observations show sensitivity to the underlying surface wetness that enables detection of wetlands even in densely vegetated areas. This study illustrates overall features of vegetation cover characterized by the suite of data. Merging the data sets using clustering techniques capitalize on the complementary strengths of the instruments for vegetation discrimination and shows promising potential for land cover characterization on a global basis.