Project: Upgrade the KD(490) product to the normalized diffuse attenuation coefficient at 490 nm (NKD(490)) for SUOMI NPP

The diffuse attenuation coefficient at 490 nm, Kd(490), is one of the standard ocean data products of the Earth Observing System (EOS), which represents an important component of the Earth Data Records (EDR). Traditionally, and even today, its derivation from satellite ocean color measurements takes the same approach as that for producing chlorophyll-a concentration ([Chl]), i.e. using the blue-green ratios of remote- sensing reflectance (Rrs) as the sole input to empirically generate the desired product. Such an approach is generally appropriate for deriving [Chl] in oceanic (or Case-1) waters, but is subject to inherent ambiguity and errors for Kd(490) for both oceanic and coastal waters. This is because that the diffuse attenuation coefficient is an apparent optical property, i.e. it varies with sun angle, while [Chl] does not. We propose to upgrade this attenuation product from two aspects: 1) remove the sun-angle related ambiguity, and 2) improve data quality through a retrieval system based on the radiative transfer theory. To achieve this upgrade, we propose to generate the normalized diffuse attenuation coefficient at 490 nm, nKd(490), from the Rrs products provided by Suomi NPP VIIRS. nKd(490) is defined as the diffuse attenuation coefficient for the Sun at zenith with a black sky, which matches the definition of the normalized water-leaving radiance (nLw) for satellite ocean color missions. This nKd(490) product is thus not only clear in its representation, but also facilitates its evaluation with field measurements. In addition, because nKd(490) will be produced through a mechanistic system rooted in the radiative transfer theory, this product will have higher science quality for the global oceans, in particular for turbid coastal waters. Furthermore, because the semi-analytical system takes a step-wise approach for the calculation of nKd(490), an uncertainty map of the nKd(490) product will also be generated via the error-propagation theory. The results from this effort will provide data products important not only to support the decadal continuity of the EDR, but also to study Earth from space to advance scientific understanding and meet societal needs.