Deriving inherent optical properties from decomposition of hyperspectral non-water absorption

Semi-analytical algorithms (SAAs) developed for multispectral ocean color sensors have benefited from a variety of approaches for retrieving the magnitude and spectral shape of inherent optical properties (IOPs). SAAs generally follow two approaches: 1) simultaneous retrieval of all IOPs, resulting in pre-defined bio-optical models and spectral dependence between IOPs and 2) retrieval of bulk IOPs (absorption and backscattering) first followed by decomposition into separate components, allowing for independent retrievals of some components. Current algorithms used to decompose hyperspectral remotely-sensed reflectance into IOPs follow the first strategy. Here, a spectral deconvolution algorithm for incorporation into the second strategy is presented that decomposes a(t-w)(lambda) from in situ measurements and estimates absorption due to phytoplankton (a(ph)(lambda)) and colored detrital material (a(dg)(lambda)) free of explicit assumptions. The algorithm described here, Derivative Analysis and Iterative Spectral Evaluation of Absorption (DAISEA), provides estimates of a(ph)(lambda) and a(dg)(lambda) over a spectral range from 350 to 700 nm. Estimated a(ph)(lambda) and a(dg)(lambda) showed an average normalized root mean square difference of < 30% and < 20%, respectively, from 350 to 650 nm for the majority of optically distinct environments considered. Estimated S-dg median difference was < 20% for all environments considered, while distribution of S-dg uncertainty suggests that biogeochemical variability represented by S-dg can be estimated free of bias. DAISEA results suggest that hyperspectral satellite ocean color data will improve our ability to track biogeochemical processes affiliated with variability in a(dg)(lambda) and S-dg free of explicit assumptions.