Remote sensing of particulate absorption coefficients and their biogeochemical interpretation: A case study in the Irish Sea

Abstract

We propose a method for (i) obtaining particulate absorption coefficients from optical remote sensing reflectance signals and (ii) partitioning these coefficients into separate components for phytoplankton cells and mineral particles. This method makes the initial assumption that absorption by coloured dissolved organic matter (CDOM) in the region of interest is either constant, or so low that its variability can be neglected. It uses a slightly modified version of the Quasi-Analytical Algorithm (Lee et al., 2009) to recover the non-water absorption (anw) and particulate backscattering (bbp) coefficients, and then obtains an estimate of absorption by CDOM, and the backscattering to absorption ratios for minerals and phytoplankton, by fitting upper and lower boundaries to the resulting data cluster. These ratios are employed in linear un-mixing equations to resolve the contributions of phytoplankton and minerals to the particulate absorption coefficient. The performance of the method was evaluated using reflectance spectra generated by an optical model of the Irish Sea which incorporated realistic specific inherent optical properties and distributions of constituent concentrations. It was then used to recover absorption coefficients for phytoplankton and suspended minerals in the 488nm waveband from MODIS remote sensing data for the central Irish Sea for the period May 2005–November 2012. The partitioned coefficients showed strong seasonality, with suspended mineral absorption peaking in winter and phytoplankton absorption becoming dominant in spring and summer. By employing locally determined specific inherent optical properties, it was possible to estimate the average annual cycles in the concentrations of the two constituents. These cycles proved to be consistent with previously published models of particle dynamics in the region in which mineral concentrations are elevated by winter mixing and increasing water column stability in spring is followed by a period of phytoplankton growth. We conclude that it is possible to resolve the separate contributions of phytoplankton and minerals to particulate absorption in waters where these constituents are the dominant determinants of optical variability, and that the partitioned absorption coefficients can provide valuable insights into the dynamics of biogenic and minerogenic particles in shelf sea ecosystems.