Validation of MERIS Products at Sagres, SW Portugal Between 2008-2012
Icely, John¹; Cristina, Sonia¹; Goela, Priscila¹; Moore, Gerald²; Danchenko, Sergei¹; Newton, Alice^3
1Sagremarisco Lda/ CIMA, U. of Algarve, PORTUGAL; ²Bio-Optika, UNITED KINGDOM; ³CIMA, U. of Algarve / CEE-NILU, Norway, PORTUGAL

Sagres, off the SW coast of Portugal, has been used for the validation of remote sensing products from the MERIS (MEdium Resolution Imaging Spectrometer) sensor on the ESA ENVISAT satellite between October 2008 and April 2012. This site benefits from many days with clear skies, thereby, increasing the number of potential days available for matching satellite images with in situ data. Furthermore, ready access to boats allows for the implementation of a sampling campaign at short notice. Three Stations at 2, 10 and 18 Kms from the shore have been selected to provide a contrast between coastal and more oceanic conditions. The validation campaigns have culminated in around 300 potential oematch ups¯ between MERIS and in situ data, where sampling has occurred within 0.5 h of the satellite overpass for the coastal station and within 1.5 h for the more offshore Stations.

Validation measurements were taken for hyperspectral water leaving reflectance and for atmospheric properties from the AERONET (AErosol RObotic NETwork) station at Sagres, coincident with field measurements from a hand-held sun photometer. The radiometric data that is presented here is FR (full resolution) Level 2 data from MERIS, with a spatial resolution of 290 ‚' 260 m. Based on the coordinates of the stations from each field campaign at Sagres, 3A-3 pixel matrices were extracted from the MERIS Level 2 products and filtered for contamination such as high glint, ice haze, and low sun. A matrix was only compared with in situ data, if five or more valid pixels remained after this filtration. The processing procedure involves taking MERIS Level 1b images that were processed with BEAM software (Basic ERS and Envisat (A)ATSR and MERIS Toolbox-version 4.9) up to Level 2 with the MEGS 8.1 processor (Meris Ground Segment data processing) using ODESA software (Optical Data Processor of the ESA-www.earth.eo.esa.int/odesa/). Using this procedure, it was possible to evaluate the performance of the sensor before or after vicarious adjustment by disabling or enabling the auxiliary data files that contained the procedure for vicarious adjustment before Level 1b images were processed to Level 2. The ICOL processor (Improve Contrast between Ocean and Land) was incorporated with the MEGS 8.1 processor and the retrieved Level 2 satellite data was evaluated against in situ data before and after ICOL processing.

Additional in situ validation measurements for water quality included pigment concentration by HPLC (High Performance Liquid Chromatography) for both MERIS Algal 1 and Algal 2 products, total suspended matter, coloured dissolved organic matter, and absorption coefficients for aquatic particles. As part of a longer time series at Sagres, measurements included temperature, salinity, Secchi depth, chlorophyll, and phytoplankton counts for species and community structure. Meteorological data was available, both on the sampling vessel and as a longer term time series at the AERONET station.

The outcome of these validation activities is summarised for this conference and all the data presented will be available from the MERis MAtchup In situ Datbase (MERMAID).