Hyperspectral Remote Sensing of the Coastal Ocean: Adaptive Sampling and Forecasting of Nearshore IN SITU Optical Properties.

January 1999 - January 2002

Project Summary:

Satellite ocean color data supplies a synoptic perspective on radiatively active constituents on space and time scales not possible using shipboard sampling. Remotely sensed data primarily provides data on the surface layers of the ocean. Consequently, remote sensing approaches must assume or rely upon ancillary data to estimate vertical changes in the optical properties. In the optically deep ocean (ODO), there is significant vertical variability in the concentrations of phytoplankton, colored dissolved organic matter (CDOM), suspended sediments, and detrital material which can change on the time scale of hours. This is especially true for most nearshore coastal ocean environments. This vertical heterogeneity in the IOPs thus limits the operational and predictive utility of algorithms based solely on satellite ocean color data.


	Time-series from 2001 of the cross-shore distribution of the percent composition of phytoplankton biomass for (A) diatoms, (B) dinoflagellates and (C) cyanobacteria. Horizontal bars are given along the top figure panels to indicate times of upwelling (blue) and downwelling (red).

Our research efforts within the New York Bight are focused on developing a relocatable data-assimilative model, which is coupled to an adaptive multi-platform sampling network. This program will integrate a bio-optical ecosystem model into our hydrodynamic model to provide a 3-dimensional forecasting capability for coastal optical properties. The development and validation of an integrated system to provide nowcasts and 2-3 day forecasts of the physical/optical state of the nearshore coastal ocean is an additional focus of this program. This project will also allow for the continued development of adaptive sampling approaches that can characterize episodic hydrographic events. Adaptive sampling approaches are critical for sampling episodic events, since standard techniques do not adequately resolve their frequency and their disproportionately large role on the optical, biological, and chemical processes in coastal waters.

Specific Objectives:

develop and deploy moored, shipboard, and autonomous bio-optical systems in the coastal ocean to ground-truth remote sensing imagery
quantify the physical, chemical and biological processes that define the spatial and temporal variability in the spectral IOPs for the nearshore coastal ocean
refine and calibrate existing hyperspectral optical models to derive IOPs from remotely sensed data using the above datasets
couple the radiative transfer module to the data-assimilative hydrodynamic model to 3-dimensionally forecast (horizon to horizon, surface to benthos) the impact of changing hydrography on the in situ IOPs
evaluate the coupled hydrodynamic-optical forecasting system by extending the Coastal Predictive Skill Experiments into the summers of 2000 and 2001 focusing on the adaptive sampling of recurrent coastal upwelling and the sub-optimal assimilation of data collected by the Long Term Ecosystem Observatory (LEO-15)
explore optimal assimilation schemes to generate 3-dimensional physical/biological/optical fields for dynamical analysis of key events

Field work