The Prediction of Remote Sensing Reflectance at Leo-15

Abstract

The prediction of spectral water leaving radiance and remote sensing reflectance from a coupled physical/ecological/optical simulation is one of the goals of the ONR Hyperspectral Coastal Ocean Dynamics Experiment (HyCODE). This goal requires coupling a high-resolution circulation model (ROMS) to provide realistic physical forcing to a dynamic ecological model that produces depth-dependent distributions of inherent optical properties (EcoSim), which are subsequently fed to a robust radiative transfer model (Ecolight) to generate the water leaving radiance and remote sensing reflectance at the sea surface. The 2001 HyCODE experiment off the coast of New Jersey provided a data rich environment that included measurements of local currents from towed ADCPs and CODAR, in situ inherent and apparent optical properties (IOPs and AOPs), sea surface remote sensing reflectance, pigment distributions, fluorescence, and ocean color remote sensing (from three different aircraft sensors and four different satellites) with which to validate a coupled numerical simulation. The predictions over the New Jersey Bight have a 300 m resolution at the LEO-15 site, encompassing a total of 250,000 grid points. The ecological simulation is initialized from estimates of historical relationships between total phytoplankton, pigments, nutrients, Color Dissolved Organic Matter (CDOM), temperature, and salinity. The mass constituents of the ecological model include specific IOP functions, which are used in a numerically fast downwelling radiative transfer model to generate the time dependent spectral photon density for the calculation of primary production, light-dependent pigment adjustments, and CDOM cycling. The resultant IOPs are then used once per simulated day to generate the Ecolight solution for spectral upwelling radiance and remote sensing reflectance. The ability to directly compare physical measurements of radiance and remote sensing reflectance to simulated estimates provides an opportunity to validate numerical simulations via photon densities, rather than through derived products such as chlorophyll concentrations.

5 Figures and Tables

Cite this paper

@inproceedings{Bissett2002ThePO, title={The Prediction of Remote Sensing Reflectance at Leo-15}, author={W. Paul Bissett and Hernan G. Arango and Robert Arnone and Robert Chant and Scott Glenn and Curtis D. Mobley and Mark A. Moline and Oscar Schofield and Robert G Steward and John Wilkin}, year={2002} }