Water Quality Monitoring for Lake Constance with a Physically Based Algorithm for MERIS Data

Sensors

Daniel Odermatt²  Thomas Heege¹  Jens Nieke²  Mathias Kneubühler² 

[1] EOMAP GmbH & Co. KG, Sonderflughafen Oberpfaffenhofen, D-82205 Gilching, Germany;Remote Sensing Laboratories (RSL), UZH, Winterthurerstr. 190, CH-8050 Zurich, Switzerland

关键词: remote sensing;    inland water;    chlorophyll;    monitoring;    operationalization;   

DOI  :  10.3390/s8084582

来源: mdpi

【 摘 要 】

A physically based algorithm is used for automatic processing of MERIS level 1B full resolution data. The algorithm is originally used with input variables for optimization with different sensors (i.e. channel recalibration and weighting), aquatic regions (i.e. specific inherent optical properties) or atmospheric conditions (i.e. aerosol models). For operational use, however, a lake-specific parameterization is required, representing an approximation of the spatio-temporal variation in atmospheric and hydrooptic conditions, and accounting for sensor properties. The algorithm performs atmospheric correction with a LUT for at-sensor radiance, and a downhill simplex inversion of chl-a, sm and y from subsurface irradiance reflectance. These outputs are enhanced by a selective filter, which makes use of the retrieval residuals. Regular chl-a sampling measurements by the Lake's protection authority coinciding with MERIS acquisitions were used for parameterization, training and validation.

【 授权许可】

CC BY   
© 2008 by the authors; licensee Molecular Diversity Preservation International, Basel, Switzerland.