POSTER SESSION 4
10:30–12:30, Thursday, September 17
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
ABSTRACT 1175 | POSTER 236
PACE/OCI ATMOSPHERIC CORRECTION FOR COASTAL AND INLAND WATERS USING ACOLITE AND LIBRADTRAN
ACOLITE has been well established for processing of metre to kilometre scale multispectral satellite imagery, especially for coastal and inland waters. The default “dark spectrum fitting” algorithm (DSF) derives the aerosol concentration from the image with no prior assumptions on the surface reflectance. The DSF assumes there are targets that are dark at least in one of the sensor bands, so the atmospheric path reflectance can be estimated. ACOLITE/DSF processing for the first operational hyperspectral ocean colour mission, PACE/OCI, was made available a few days after the first data distribution.
In this study we demonstrate issues with DSF for hyperspectral processing that arise from the LUTs generated with 6SV that are hence limited in their spectral resolution. While suitable for multispectral sensors, they can produce or amplify spectral discontinuities in the hyperspectral context, i.e. increase the band to band variability in the second derivative. A hybrid DSF (HDSF) approach is introduced, which uses the DSF to estimate the aerosol optical depth from the image, and then performs hyperspectral radiative transfer simulations using libRadtran to perform the final atmospheric correction.
Results from the DSF, HDSF, and the standard processing are compared with in situ WATERHYPERNET measurements. The HDSF approach is shown to reduce spectral discontinuities, and proves to be suitable for processing over specific regions of interest, albeit at additional computational cost. The impact of the processing method on hyperspectral parameter retrieval algorithms is evaluated. The code for HDSF is available as part of the main ACOLITE distribution.
Quinten Vanhellemont, Royal Belgian Institute of Natural Sciences, [email protected], https://orcid.org/0000-0001-9195-6347
Kevin Ruddick, RBINS, [email protected]
Poster Session | 1 | 2 | 3 | 4 | Schedule at a Glance
