Level-2 Processing - Sentinel-3 OLCI Technical Guide - Sentinel Online
At Level-1B, the OLCI instrument provides radiance measurements of the Earth's surface in the visible and near infra-red spectral domain in 21 spectral channels. These measurements are accurately calibrated TOA radiances, annotated with geo-referencing data and observation geometry parameters. Level-2 data processing aims to derive geophysical products through three main steps:
- pre-processing, the main purpose of which is to decide whether a given pixel will be further analysed (ocean, land) or not (cloud)
- ocean processing
- land processing.
Ocean processing and land processing provide the expected geophysical properties corresponding to the surface types.
The pre-processing step aims to convert Level-1 TOA radiance into reflectance, which is a more relevant parameter for geophysical properties. This step is followed by a correction for gaseous absorption, pixel classification (in particular, water/land/cloud) and retrieval of the total column water vapour content. After having eliminated cloudy pixels, according to the nature of the pixel (water/land), the OLCI Level-2 is organised into two independent processes:
- water branch, consisting of a glint estimation and correction, a specific case two water reflectance estimation (in the NIR spectral range), aerosol identification and atmospheric correction allowing derivation of ocean colour and the final associated products
- land branch, using two independent modules (the Global Vegetation Index and the Terrestrial Chlorophyll Index).
Note: The above processing branches account for the spectral smile i.e. the in-FOV variation of the central wavelengths of OLCI channels, where relevant.