We examine the impact of atmospheric correction, specifically aerosol model selection, on retrieval of bio-optical properties from satellite ocean color imagery. Uncertainties in retrievals of bio-optical properties (such as chlorophyll, absorption, and backscattering coefficients) from satellite ocean color imagery are related to a variety of factors, including errors associated with sensor calibration, atmospheric correction, and the bio-optical inversion algorithms. In many cases, selection of an inappropriate or erroneous aerosol model during atmospheric correction can dominate the errors in the satellite estimation of the normalized water-leaving radiances (_nL_w), especially over turbid, coastal waters. These errors affect the downstream bio-optical properties. Here, we focus on the impact of aerosol model selection on the _nL_w radiance estimates by comparing Aerosol Robotic Network-Ocean Color (AERONET-OC) measurements of _nL_w and aerosol optical depth (AOD) to satellite-derived values from Moderate Resolution Imaging Spectroradiometer (MODIS) and Sea-viewing Wide Field-of-view Sensor (SeaWiFS). We also apply noise to the satellite top-of-atmosphere (TOA) radiance values in the two near-infrared (NIR) wavelengths used for atmospheric correction, to assess the effect on aerosol model selection and _nL_w retrievals. In general, for the data sets examined, we found that as little as 1% uncertainty (noise) in the NIR TOA radiances can lead to the selection of a different pair of bounding aerosol models, thus changing _nL_w retrievals. We also compare aerosol size fraction retrieved from AERONET and size fraction represented by aerosol models selected during atmospheric correction.

An increasing number of commercial nanosatellite-based Earth-observing sensors are providing high-resolution images for much of the coastal ocean region. Traditionally, to improve the accuracy of normalized water-leaving radiance ($n{L}_w$) estimates, sensor gains are computed using in-orbit vicarious calibration methods. The initial series of Planet nanosatellite sensors were primarily designed for land applications and are missing a second near-infrared band, which is typically used in selecting aerosol models for atmospheric correction over oceanographic regions. This study focuses on the vicarious calibration of Planet sensors and the duplication of its red band for use in both the aerosol model selection process and as input to bio-optical ocean product algorithms. Error measurements show the calibration performed well at the Marine Optical Buoy location near Lanai, Hawaii. Further validation was performed using in situ data from the Aerosol Robotic Network—Ocean Color platform in the northern Adriatic Sea. Bio-optical ocean color products were generated and compared with products from the Visual Infrared Imaging Radiometric Suite sensor. This approach for sensor gain generation and usage proved effective in increasing the accuracy of $n{L}_w$ measurements for bio-optical ocean product algorithms.

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 [L_w(l)], which is the basis for surface ocean optical products. Machine learning (ML) algorithms (KNN, decision tree regressors) were applied to determine relationships between L_w and top-of-atmosphere (L_t)/Rayleigh (L_r) 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.

The southward propagation of cold-air frontal boundaries into the Gulf of Mexico region initiates a cascade of coupled air–sea processes that manifests along the coastlines as an apparent brightness anomaly in the ocean color signals. Our hypothesis is that the color anomaly is largely due to the turbulent resuspension of sedimentary particles. Initially, there is significant wind-driven ocean turbulence as the frontal boundary passes, followed by the potential for sustained convective instability due to significant heat losses from the ocean surface. These cold front events occur during boreal autumn, winter, and into early spring, and the latter episodes occur in the context of the seasonally recurring thermal stratification of shelf waters. Here, we show that as stratification reasserts thermal stability in the waning days of a cold front episode, daily to hourly ocean color patterns are temporally coherent with the air–sea heat flux changes and the resulting impact on water column stability. Concomitant results from a nested, data-assimilative, and two-way coupled ocean-atmosphere numerical modeling system provides both corroboration and insight into how surface air–sea fluxes and in-water turbulent mixing manifest as hourly changes in apparent surface water turbidity due to the potential excitation and settling of reflective particles. A simple model of particle mixing and settling driven by the simulated turbulence mimics patterns seen in the satellite image sequences. This study offers a preview of potential application areas that may emerge following the launch of a dedicated ocean color geostationary sensor.

Atmospheric cold front propagation across the northern Gulf of Mexico is characterized by elevated surface wind velocities and a ~10–15 °C drop in surface air temperatures. These meteorological conditions result in significant heat energy losses from the surface ocean to the overlying atmosphere. These seasonally recurring cold-air outbreak events may penetrate the southern portion of the West Florida continental shelf and initiate turbulent and convective overturn of the water column. Examination of true color images derived from ocean-viewing, satellite-based radiometer data reveals coincident and substantial surface water discolorations that are optically similar to smaller-scale “whiting events,” despite the regional-scale extent of the observed phenomenon (>25,000 km²). Coupled air–sea numerical simulations suggest the surface water discoloration occurs and is sustained where the entire water column is dynamically unstable. The simulation results indicate significant density (σ_t) inversions between the surface and bottom waters. Thus, the combined numerical model and remote sensing analysis suggest that convective turbulence may be contributing to the sustained ventilation of bottom waters containing a high concentration of suspended particulates. High-temporal resolution true color images rendered from the GOES-R Advanced Baseline Imager (ABI) data appear to support the surface water discoloration’s turbulent-driven nature.