Optical properties derived from ocean color imagery represent vertically-integrated values from roughly the first
attenuation length in the water column, thereby providing no information on the vertical structure. Robotic, in situ
gliders, on the other hand, are not as synoptic, but provide the vertical structure. By linking measurements from these
two platforms we can obtain a more complete environmental picture. We merged optical measurements derived from
gliders with ocean color satellite imagery to reconstruct vertical structure of particle size spectra (PSD) in Antarctic shelf
waters during January 2007. Satellite-derived PSD was estimated from reflectance ratios using the spectral slope of
particulate backscattering (γ<sub>bbp</sub>). Average surface values (0-20 m depth) of γ<sub>bbp</sub> were spatially coherent (1 to 50 km
resolution) between space and in-water remote sensing estimates. This agreement was confirmed with shipboard vertical
profiles of spectral backscattering (HydroScat-6). It is suggested the complimentary use of glider-satellite optical
relationships, ancillary data (e.g., wind speed) and ecological interpretation of spatial changes on particle dynamics (e.g.,
phytoplankton growth) to model underwater light fields based on cloud-free ocean color imagery.
The Autonomous Marine Optical System (AMOS) measures remote sensing reflectance (R<sub>rs</sub>) above the water surface and subsurface optical properties (irradiance at depth, beam attenuation, chlorophyll fluorescence, and light backscattering) at predetermined times throughout the day. Data are transmitted back by radio to a networked archival and processing station. AMOS was created to routinely monitor the optical properties of near-surface waters, and make those measurements available to researchers over an Ethernet connection with minimal delay. The Rrs measurements can be used not only to validate satellite and airborne remote sensing imagery, but also to be combined with the in situ measurements so that other water column properties can be estimated. The performance of visible and machine-aided hull inspection is strongly affected by the optical properties of the water. AMOS estimates of these optical properties can be used by optical models to predict both subsurface visibility and the amount of ambient light beneath ships at port inspection sites. An example of the application of an inverse hyperspectral Rrs model to AMOS data from the Port of St. Petersburg (FL) is shown to accurately estimate light absorption due to phytoplankton and colored dissolved organic matter (CDOM), and backscattering due to particles.