GRSG 34th Conference 2023

Title: A look into the influence of Sentinel-2 atmospheric correction and calibration on the repeatability of geological spectral indices

Author: Harald van der Werff

The weather and seasonal fluctuations impact the circumstances under which spectral measurements are acquired. To assess the influence of changing acquisition conditions on geological remote sensing products, we observed the repeatability of several geological spectral indices over a 5-year period and correlated them to atmospheric water vapour, aerosol content, and solar zenith position. The Sentinel-2 multispectral imager has a frequent 5-day overpass at the equator and its potential for geological applications and mineral mapping, particularly iron oxides, has been studied extensively. Sentinel-2 MSI data were therefore chosen to calculate spectral indices (band ratios) for green vegetation, ferric & ferrous iron oxide mineralogy and hydroxyl bearing alteration (clay) mineralogy in a semi-arid area in southern Spain.

In contrast to our expectations, it appears that top-of-atmosphere data has higher stability than bottom-of-atmosphere data. The Sen2Cor atmospheric processor seems to over-correct atmospheric influences. In line with our expectations, spectral index stability is highest during periods of heightened water vapour content, specifically under moist atmospheric conditions that mostly happen over summer. The same is observed for the solar zenith angle, during the astronomical summer, and periods of low aerosol concentrations also contribute to enhanced spectral index repeatability. Clearly, multiple environmental conditions exhibit confounding effects that can be compounded by seasonality cycles.

The weather and seasonal fluctuations impact the circumstances under which spectral measurements are acquired. To assess the influence of changing acquisition conditions on geological remote sensing products, we observed the repeatability of several geological spectral indices over a 5-year period and correlated them to atmospheric water vapour, aerosol content, and solar zenith position. The Sentinel-2 multispectral imager has a frequent 5-day overpass at the equator and its potential for geological applications and mineral mapping, particularly iron oxides, has been studied extensively. Sentinel-2 MSI data were therefore chosen to calculate spectral indices (band ratios) for green vegetation, ferric & ferrous iron oxide mineralogy and hydroxyl bearing alteration (clay) mineralogy in a semi-arid area in southern Spain.

In contrast to our expectations, it appears that top-of-atmosphere data has higher stability than bottom-of-atmosphere data. The Sen2Cor atmospheric processor seems to over-correct atmospheric influences. In line with our expectations, spectral index stability is highest during periods of heightened water vapour content, specifically under moist atmospheric conditions that mostly happen over summer. The same is observed for the solar zenith angle, during the astronomical summer, and periods of low aerosol concentrations also contribute to enhanced spectral index repeatability. Clearly, multiple environmental conditions exhibit confounding effects that can be compounded by seasonality cycles.