MERIS FUB-CSIRO Coastal Water Processor - Algorithm Specification

Algorithm Specification

TheMERIS FUB-CSIRO Coastal Water Processor makes use of MERIS Level-1b TOA radiances in the bands 1-7, 9-10 and 12-14 to retrieve the following case II water properties and atmospheric properties above case II waters :

• chlorophyll-a concentration (log scale, mg/m^3)
• yellow substance absorption @ 443 nm (log scale, 1/m)
• total suspended matter concentration (log scale, g/m^3)
• yellow substance absorption @ 443 nm (log scale, 1/m)
• aerosol optical depth @ 440 nm
• aerosol optical depth @ 550 nm
• aerosol optical depth @ 670 nm
• aerosol optical depth @ 870 nm
• water-leaving RS reflectance @ 412 nm (1/sr)
• water-leaving RS reflectance @ 442 nm (1/sr)
• water-leaving RS reflectance @ 490 nm (1/sr)
• water-leaving RS reflectance @ 510 nm (1/sr)
• water-leaving RS reflectance @ 560 nm (1/sr)
• water-leaving RS reflectance @ 620 nm (1/sr)
• water-leaving RS reflectance @ 665 nm (1/sr)
• water-leaving RS reflectance @ 708 nm (1/sr)

The retrieval is based on four separate artificial neural networks which were trained on the basis of the results of extensive radiative transfer simulations with the MOMO code by taking varying atmospheric and oceanic conditions into account. All networks were validated against in-situ measurements.

During the Plugin processing the MERIS Level-1b data are masked prior to the retrieval by applying the following combination mask :

• GLINT_RISK
• BRIGHT
• INVALID
• SUSPECT (only when not masking out almost all pixels)

The masked pixel's values are set to 5.

Non-masked pixels are then normalized for an atmosphere's ozone contents of 344 Dobson units by calculating transmission correction factors.

Notice that the AOT wavelengths 440, 670 and 870 nm correspond to the AERONET data wavelengths for a convenient direct comparison with in situ data.

Each pixel is checked against the input and output values margin of the trained networks. Additional flags are set in case of a neural network failure for input and output separately.

The radiative transfer simulation code and the retrieval algorithm are described in detail in the papers cited in the references section below. We would like to stress the fact that the operator is applicable over case II only, thus it is likely to fail over the open ocean by producing negative remote sensing (RS) water-leaving reflectances.

References

Schroeder Th., Schaale M., Fischer J., "Retrieval of atmospheric and oceanic properties from MERIS measurements: A new Case-2 water processor for BEAM", International Journal of Remote Sensing; Volume 28, 2007 - Issue 24

Fischer J., and Grassl H., "Radiative transfer in an atmosphere-ocean system: an azimuthally dependent matrixoperator approach", Applied Optics, 23, 1032-1039, 1984.

Fell F., and Fischer J., "Numerical simulation of the light field in the atmosphere-ocean system using the matrixoperator method", Journal of Quantitative Spectroscopy & Radiative Transfer, 69, 351-388, 2001.

Schroeder Th., Schaale M., Fell F. and Fischer J., "Atmospheric correction algorithm for MERIS data: A neural network approach", In: Proceedings of the Ocean Optics XVI Conference, Santa Fe, New Mexico, USA, published on CD ROM, 2002.

Schroeder Th., Fischer J., Schaale M. and Fell F., "Artificial neural network based atmospheric correction algorithm: Application to MERIS data", In: P roceedings of the International Society for Optical Engineering (SPIE), Vol. 4892, Hangzhou, China, 2002.

Schroeder Th., and Fischer J., "Atmospheric correction of MERIS imagery above case-2 waters", In: Proceedings of the 2003 MERIS User Workshop, ESA ESRIN, Frascati, Italy, 2003.

Schroeder, Th., "Fernerkundung von Wasserinhaltsstoffen in Kuestengeweassern mit MERIS unter Anwendung expliziter und impliziter Atmosphaerenkorrektur- verfahren", Ph.D. Dissertation, Freie Universitatet Berlin, Berlin (Germany), 2005, http:/www.diss.fu-berlin.de/2005/78