Three-Dimensional Optical Fields Derived from Aircraft Remote Sensing Imagery:Results from the Hamlet's Cove Experiment, Eglin AFB, Florida

摘要

Three dimensional distributions of inherent optical properties (absorption, a;scattering, b; and single-scattering albedo, omega-o) at six SeaWiFS wavelengths were defined for a coastal region by coupling surface ocean color imagery and models of vertical optical profiles. Aircraft measurements of water leaving radiance collected over a nearshore environment in the northeastern Gulf of Mexico were incorporated into a modified SeaWiFS algorithm after atmospheric correction and removal of the bottom reflectance component, to provide surface estimates of a, b, and omega-o. To estimate subsurface distribution patterns, a priori knowledge of the depth distributions of the optical properties is required at several locations in the image. During the aircraft overflights, concurrent in situ depth profiles of absorption and scattering were collected at five stations along an offshore transect. The depth profiles for each station and wavelength were modeled using a sigmoidal or a bimodal Gaussian distribution. The individual modeled depth profile shapes were subsequently applied across the image by assigning the curve shapes to pixels with expected similar hydrographic properties. In this shallow coastal environment, the distribution patterns appeared related to coastal run-off and river discharge (and therefore bathymetry, indirectly), so profile shapes were assigned to pixels based on the water depth at each location. Thus, all pixels that fell within specified depth ranges were initially assigned the same profile shape. The modeled profiles of the optical properties were integrated over depth (weighted by the diffuse attenuation coefficient calculated from the corresponding a and b values) to provide a 'remote-sensing' estimate comparable to what the aircraft sensor would see.