Validation of a new Water RS Algorithm with an Automatically Quality-Controlled Reflectance Data Set of Turbid Lakes
Ligi, Martin¹; Grötsch, P.M.M.¹; Reinart, Anu¹; Ansko, Ilmar¹; Stefan, S. G. H.^2
1Tartu Observatory, ESTONIA; ²Finnish Environment Institute, FINLAND

Water constituent concentrations can be obtained from optical satellite measurements after calibration and validation with in situ data. However, in situ measurements can be deteriorated by various issues - in the case of water leaving reflectances, e.g. due to illumination conditions, sensor tilting, adverse water surface conditions (whitecaps, waves, glint) and atmospheric effects (clouds, haze).

In this work, reflectance spectra from turbid Estonian lakes were collected with an above water TriOS Ramses set-up and a handheld WISP-3 system. We then used a novel approach to automatically reject unsuitable reflectance measurements. Additionally, water samples were collected and analysed in the lab for chlorophyll a, suspended matter and coloured dissolved organic matter concentrations.

This quality-controlled reflectance data set, as well as corresponding satellite data, was then processed with a new bio-optical algorithm, resulting in water constituent concentration. The algorithm performance was then evaluated with the lab retrieved concentrations.