Mapping water constituents in Lake Constance using CHRIS/Proba

Abstract

CHRIS-PROBA data were acquired at 3 days in 2003 and one day in 2004 at the Eastern part of Lake Constance. Field campaigns were organised at these days in order to measure optical parameters of the water and the atmosphere and concentrations of water constituents from ship for validation. Five deep water sampling stations were chosen. The modular inversion program (MIP) developed by Heege, Miksa and Kisselev was used for the coupled retrieval of water constituents and aerosol concentrations, for atmospheric and water surface corrections and for optical closure and sensor calibration tests. Different viewing angles of CHRIS images of 2003 were used to test the atmospheric correction algorithm. Spectra were calculated and adjusted to the measured ones by fitting the concentrations of 3 aerosol types (rural, maritime, urban) and 3 water constituents (suspended matter, chlorophyll, Gelbstoff). The resulting mean relative error for suspended matter is 17%, for chlorophyll it is 28% which is within the error margin of the measurements. Maps of chlorophyll, suspended matter and Gelbstoff were generated for all viewing angles of CHRIS/PROBA where sunglint contamination is absent. By comparing these maps the accuracies of the resulting water constituents are determined. Looking at one pixel from different angles, the water constituent concentration is supposed to be the same. The mean relative error for each sampling station is 37% for chlorophyll and 24% for suspended matter. A more detailed statistical analysis is in preparation.