Search

Here we present a machine-learning-based method for utilizing traditional ocean-viewing satellites to perform automated atmospheric correction of nanosatellite data. These sensor convolution techniques are required because nanosatellites do not usually possess the wavelength combinations required to atmospherically correct upwelling radiance data for oceanographic applications; however, nanosatellites do provide superior ground-viewing spatial resolution (~3 m). Coincident multispectral data from the Suomi National Polar-Orbiting Partnership Visible Infrared Imaging Radiometer Suite (Suomi NPP VIIRS; referred to herein as "VIIRS") were used to remove atmospheric contamination at each of the nanosatellite's visible wavelengths to yield an estimate of spectral water-leaving radiance [Lw(l)], which is the basis for surface ocean optical products. Machine learning (ML) algorithms (KNN, decision tree regressors) were applied to determine relationships between Lw and top-of-atmosphere (Lt)/Rayleigh (Lr) radiances within VIIRS training data, and then applied to test cases for (1) the Marine Optical Buoy (MOBY) in Hawaii and (2) the AErosol RObotic Network Ocean Color (AERONET-OC), Venice, Italy. For the test cases examined, ML-based methods appeared to improve statistical results when compared to alternative dark spectrum fitting (DSF) methods. The results suggest that ML-based sensor convolution techniques offer a viable path forward for the oceanographic application of nanosatellite data streams. [ABSTRACT FROM AUTHOR]

Results from first observational program of bioluminescence (BL) potential in the Delaware Bay area are presented. During the field program July 30-August 1, 2015, the satellite Visible Infrared Imaging Radiometer Suite (VIIRS) imagery shows the development of the submesoscale filament with elevated chlorophyll-a in the area of interaction of lighter water masses of the Bay outflow with denser upwelled water. We have shown that ageostrophic secondary circulation (ASC) cells contributed to the development of this filament. Analysis of BL potential observations have shown elevated values of BL potential in the area of the submesoscale filament. Analysis of observed temperature, salinity, sigma-t, and BL potential along the stations crossing the Bay mouth have shown a presence of a strong frontal structure separating colder, more saline, denser offshore water from the bay water masses. Over 3 days of sampling, this frontal structure moved onshore to the entrance of the bay, and brought offshore BL plankton communities with higher values of BL potential. We compared two surveys (at the end of July and in the middle of August) of water masses located in the area where the buoyant outflow of the Delaware Coastal Current is turned around and taken to the north up the shelf by upwelling favorable winds. Analysis of observations shows that the survey at the end of July has fresher water masses (due to higher river runoff before and during survey) and higher values of BL potential in comparison to the survey at the middle of August. [ABSTRACT FROM AUTHOR]

The Geostationary Ocean Color Imager (GOCI) is the first geostationary ocean color sensor in orbit that provides bio-optical properties from coastal and open waters around the Korean Peninsula at unprecedented temporal resolution. In this study, we compare the normalized water-leaving radiance (nLw) products generated by the Naval Research Laboratory Automated Processing System (APS) with those produced by the stand-alone software package, the GOCI Data Processing System (GDPS), developed by the Korean Ocean Research & Development Institute (KORDI). Both results are then compared to the nLw measured by the above water radiometer at the Ieodo site. This above-water radiometer is part of the Aerosol Robotic NETwork (AeroNET). The results indicate that the APS and GDPS processed nLw correlates well within the same image slot where the coefficient of determination (r2) is higher than 0.84 for all the bands from 412 nm to 745 nm. The agreement between APS and the AeroNET data is higher when compared to the GDPS results. The Root-Mean-Squared-Error (RMSE) between AeroNET and APS data ranges from 0.24 [mW/(cm2srμm)] at 555 nm to 0.52 [mW/(cm2srμm)] at 412 nm while RMSE between AeroNET and GDPS data ranges from 0.47 [mW/(cm2srμm)] at 443 nm to 0.69 [mW/(cm2srμm)] at 490 nm. [ABSTRACT FROM AUTHOR]

The spatial dynamics of coastal and inland regions are highly variable and monitoring these waters with ocean color remote sensors requires increased spatial resolution capabilities. A procedure for the spatial enhancement of ocean color products, including chlorophyll and inherent optical properties (IOPs), is developed using a sharpened visible water-leaving radiance spectrum for the visible infrared imaging radiometer suite (VIIRS). A new approach for spectral sharpening is developed by utilizing the spatial covariance of the spectral bands for sharpening the M bands (412, 443, 486, 551, 671 nm; 750-m resolution) with the I-1 band (645 nm; 375-m resolution). The spectral shape remains consistent by the use of a dynamic, wavelength-specific spatial resolution ratio that is weighted as a function of the relationship between proximate I- and M-band variance at each pixel. A comparison of bio-optical satellite products at 375-m and 750-m spatial resolution with in situ measurements of water leaving radiance and bio-optical properties show an improved capability of the VIIRS 375-m products in turbid and optically complex waters, such as the Chesapeake Bay and Mississippi River Plume. We demonstrate that the increased spatial resolution improves the ability for VIIRS to characterize bio-optical properties in coastal waters. [ABSTRACT FROM AUTHOR]