Hyperspectral and polarimetric imaging of the ocean for the characterization of the surface effects and measurement uncertainties

Author

Mateusz Malinowski, CUNY City CollegeFollow

Date of Award

2024

Document Type

Dissertation

Department

Engineering

First Advisor

Prof. Alexander Gilerson

Keywords

hyperspectral, polarimetric, imaging, ocean, surface, uncertainties

Abstract

Ocean and coastal waters are monitored by Ocean Color satellite sensors to determine concentrations of chlorophyll and water properties and identify areas of algal blooms and other events. The light radiance from the ocean is weak in comparison with the sky radiance, which requires very accurate atmospheric correction of the radiance measured at the top of the atmosphere (TOA) on the satellite and heavy validation of the derived water leaving radiance by field measurements from the ships and ocean platforms. For TOA and especially above water radiance the skylight reflected from the ocean surface represents one of the main sources of measurement uncertainties.

NASA’s Plankton, Aerosol, Cloud, and ocean Ecosystem (PACE) mission, successfully launched in February 2024, has the main hyperspectral imager and two polarimeters on board, which requires novel approaches in atmospheric correction, skylight correction, algorithms for retrieval of water optical properties as well as characterization of the surface and related uncertainties in the hyperspectral and polarimetric modes.

Polarized light provides important information in multiple environmental applications including the retrieval of parameters of atmospheric aerosols, ocean particulates, and ocean surfaces, information, which is not directly available from the traditional multi- and hyperspectral radiometric measurements. The polarization component is also critical for the accurate determination of the reflectance coefficient from the sea surface, which is widely used in the processing of above-water measurements and atmospheric correction models.

In this work, a unique optical system is used to make multi-and hyperspectral measurements above the ocean surface. The system includes a hyperspectral imager, in which each of 410x410 pixels contains hyperspectral information in the 350 -1000 nm range, and a polarimetric camera, which takes videos at 10 frames per second with 2464(H) x 2056(V) resolution, where each of 2x2 pixel area consists of 4 different subpixels with polarization orientations at 0, 45, 90, and 135 degrees. The images were acquired from the ships during the Visible Infrared Imaging Radiometer Suite (VIIRS) Calibration/Validation Cruises in the northern part of the Gulf of Mexico, around islands of Hawaii, at the US West Coast near Newport, OR as well as from the Coney Island Pier in Brooklyn, New York. Measurements were also conducted aerially from the helicopter during the Chesapeake Bay campaign, taken at 60, 150, 450, and 750 meters from the sea surface.

Based on the camera imaging, a modification of the previously available polarization sensing technique was developed, which reliably characterizes wave slopes of the ocean surfaces at different wind speeds. The estimated wave slopes agree well with the Cox and Munk statistics developed from measurements in the open ocean. This holds for different geographical regions and bodies of water from open ocean to coastal regions which previously has not been studied.

Data from the camera and imager measurements were also used for the estimation of uncertainties of above wind-roughened ocean surface radiometric measurements in polarized and unpolarized modes at various wind speeds and their impact on the atmospheric correction models. Propagation of uncertainties from the sea surface to higher helicopter altitudes was studied and showed uncertainties rising in the blue (400-450 nm) part of the spectrum as the result of the interaction of the surface-reflected light with Rayleigh scattering.

Recommended Citation

Malinowski, Mateusz, "Hyperspectral and polarimetric imaging of the ocean for the characterization of the surface effects and measurement uncertainties" (2024). CUNY Academic Works.
https://academicworks.cuny.edu/cc_etds_theses/1304