Spectral modes of radiometric measurements in optically complex waters

Abstract

The case-1 (optically simple) oceanic environment is described by a smoothly transitioning gradient in optical properties arising from the concentration of the primary algal photopigment, chlorophyll a, plus its covarying and optically relevant constituents. As such, the spectral transition that defines case-1 water types is captured by a single primary spectral mode—the transition in peak wavelength between oligotrophic blue and mesotrophic green waters—which has substantiated blue-green algorithm designs and sensor capabilities for oceanographic observations that primarily focus on retrieving the visible (VIS) spectral domain. For case-2 water types, e.g., optically complex inland and coastal waters interacting with the continent and shelf, a similar unifying mode that captures common spectral changes is not defined. This study evaluates the potential to formulate optical complexity as a progression of discrete or continuous parameters in order to advance radiometry for a global range of water types. Within a gradient of increasing optical complexity, this study uses a comprehensive set of case-2 scenarios to compare spectral changes as a function of increasing water mass complexity, based on simple diagnostic tools and definitions. The spectral comparisons rely on instrument and data processing improvements—developed from an end-member analysis (EMA) perspective—that optimize a spectral domain spanning wavelengths shorter and longer than the primarily case-1 VIS domain, i.e., the ultraviolet (UV) and near-infrared (NIR) domains, and that enable sampling of shallow or non-navigable water bodies. The spectral comparisons in the case-2 scenarios indicate five primary modes capture the dominant spectral changes associated with increasing optical complexity. The primary modes and diagnostic tools presented herein enable the organization of global (i.e., oceanic, coastal, and inland) water bodies along a continuous complexity gradient defined by optical complexity instead of as a binary and discontinuous partition (i.e., case-1 and case-2). The expression and dynamic range of the modes establish the spectral regions most sensitive to water mass transitions in optically complex environments. These spectral expressions provide a basis for evaluating the spectral sensing requirements for next-generation remote sensing missions, which seek to characterize optically complex coastal (i.e., the continental shelf and its environs) and inland water bodies. In particular, three of five spectral modes indicate the spectral response to increasing optical complexity is greatest at the spectral end members. The results presented indicate that both discretized values and a continuum are useful for formulating a hierarchy in optical complexity and suggest next-generation activities should optimize the retrieval of the UV and NIR domains for improving the development of global algorithms.