The combination of increased spectral resolution for in situ ocean optical instrumentation as well as future ocean remote sensing missions (e.g., PACE) provides an opportunity to examine new methods of analysis and ocean monitoring that were not feasible during the multispectral satellite era. For example, hyperspectral data enables a much more precise determination of the apparent true color for natural waters, one based on the full spectral shape of water-leaving radiance distributions. Herein we provide examples of how specific integrated biogeo-optical and physical processes in the northern Gulf of Mexico have characteristic hyperspectral signatures, and thusly, characteristic true color identifiers. Our emergent hypothesis is that once the characteristic hyperspectral color signature of a specific biophysical process is known, it can be detected and monitored even with multispectral or broad-band response digital imaging systems. To test this hypothesis, we examine archived imagery from MODIS and HICO to identify putative bottom boundary layer ventilation events along divergent shelf-frontal boundaries across the northern Gulf continental margin. Whereas on-demand in situ physical data that provide spatiotemporal correspondence with archived images are not available, we employ the data-assimilative Coupled Ocean-Atmosphere Mesoscale Prediction System (COAMPS) as a physical data surrogate. Preliminary results of this method appear to support the hypothesis, with the caveat that model results must be interpreted with due caution.
|