Standard Atmospheric Correction Research Articles

Enhanced POLYMER atmospheric correction algorithm for water-leaving radiance retrievals from hyperspectral/multispectral remote sensing data in inland and coastal waters.

Accurate retrieval of the water-leaving radiance from hyperspectral/multispectral remote sensing data in optically complex inland and coastal waters remains a challenge due to the excessive concentrations of phytoplankton and suspended sediments as well as the inaccurate estimation and extrapolation of aerosol radiance over the visible wavelengths. In recent years, reasonably accurate methods were established to estimate the enhanced contribution of suspended sediments in the near-infrared (NIR) and shortwave infrared (SWIR) bands to enable atmospheric correction in coastal waters, but solutions to derive the dominant phytoplankton contribution in the NIR and SWIR bands are less generalizable and subject to large uncertainties in the remotely-derived water color products. These issues are not only associated with the standard atmospheric correction algorithm in the SeaDAS processing system but with the non-traditional algorithms such as POLYMER (POLYnomial-based approach established for the atmospheric correction of MERIS data). This study attempts to enhance the POLYMER algorithm to enable atmospheric correction of hyperspectral and multispectral remote sensing data over a wide range of inland and ocean waters. The original POLYMER algorithm is less suitable owing to its complete reliance on a polynomial approach to model the atmospheric reflectance as a function of the wavelength and retrieve the water-leaving reflectance using two semi-analytical models (MM01 and PR05). The polynomial functions calculate the bulk atmospheric contribution instead of using an explicit method to estimate aerosol radiance separately, resulting the erroneous water color products in inland and coastal waters. The modified POLYMER algorithm (mPOLYMER) employs more realistic approaches to estimate aerosol contributions with a combination of UV and Visible-NIR bands and enables accurate retrievals of water-leaving radiance from both hyperspectral and multispectral remote sensing data. To assess the relative performance and wider applicability of mPOLYMER, the original and enhanced algorithms were tested on a variety of HICO, MSI and MODIS-Aqua data and the retrieved Lwn products were compared with AERONET-OC and OOIL-regional in-situ data. Expectedly, the mPOLYMER algorithm greatly improved the accuracy of Lwn (in terms of magnitude and spectral shape) when applied to MODIS-Aqua and HICO data in highly turbid productive waters (with higher concentrations of phytoplankton or with dense algal blooms) in Muttukadu Lagoon, Lake Erie, Yangtze River Estuary, Baltic Sea and Arabian Sea. In contrast, the original POLYMER algorithm overestimated Lwn in the visible and NIR bands and produced unphysical negative Lwn or distorted Lwn spectra in turbid productive waters. The mPOLYMER yielded a relative mean error reduction of more than 50% (i.e., from 79% to 34%) in Lwn for a large number of matchup data. The improved accuracy and data quality is because the mPOLYMER algorithm's funio and coefficients sufficiently accounted for the enhanced backscattering contribution of phytoplankton and suspended sediments in optically complex waters.

Uncertainty of atmospheric scattering functions relevant for Satellite Ocean Colour Radiometry in European Seas

ABSTRACT Uncertainty assessment of Ocean Colour Radiometry (OCR) usually relies on the comparison between satellite marine reflectance and concurrent in situ radiometric measurements. However, when dealing with the standard atmospheric correction (AC), uncertainty of marine reflectance originates fundamentally from two sources: (i) uncertainty of the Level-1 calibration at top of the atmosphere, and (ii) uncertainty of the atmospheric scattering functions computed by the AC (atmospheric path reflectance and transmittance). The aim of this alternative work is thus to assess the uncertainty of these scattering functions, thanks to inherent optical properties of aerosols extracted from the AErosol Robotic NETwork (AERONET) and ingested in a radiative transfer code. The study is conducted for the MEdium Resolution Imaging Spectrometer (MERIS), over three AERONET sites selected in European seas (Aqua Alta Oceanographic Tower, AAOT; Gustav Dalen Tower, GDT; Lampedusa) and Lanai in the Pacific Ocean as a reference. We develop an uncertainty budget based on the statistical analysis of the errors between MERIS and AERONET-derived data. At 443 nm, uncertainty of MERIS atmospheric path reflectance due to random effects is estimated around 2%, while that of total atmospheric transmittance is between 0.6% and 1%. Propagation of these uncertainties to OCR is achieved following metrological principles and varies between match-ups and sites due to the distinctive contribution of the marine signal to the total radiometry: uncertainty of MERIS marine reflectance at 443 nm is 10.8% (±1.1%) at Lanai, 18.3% (±4.6%) at Lampedusa, 29.4% (±12.2%) at AAOT and up to 68.7% (±127.4%) at GDT (i.e. with a significant intra-site variability). The methodology is a first step aimed at deriving a pixel-by-pixel uncertainty budget of OCR, instead of providing overall estimates as generally done in the OC community. Limitations due to uncertainties in the AERONET data are also discussed, with a tentative quantification.

Atmospheric correction under cloud edge effects for Geostationary Ocean Color Imager through deep learning

The Geostationary Ocean Color Imager (GOCI) is widely employed in tracking diurnal dynamics of oceanic conditions. However, the current atmospheric correction (AC) algorithms for GOCI often mask pixels around cloud edges to exclude pixels contaminated by cloud edge effects (CEEs; including stray light, cloud shadows, and cloud adjacent effects (AEs)), which results in massive data loss. In this paper, we propose a novel AC algorithm to correct these CEE-affected pixels based on deep learning (namely, DLACC) to achieve a comparable quality level to those pixels far away from cloud edges. We used the standard near-infrared iterative AC algorithm in SeaDAS (namely, NIR) to obtain Rayleigh-corrected reflectance (Rrc) and remote sensing reflectance (Rrs). Then, high-quality Rrs values were extracted using a quality assurance (QA) algorithm. Next, we obtained 2,821,668 matchups by matching CEE-free noontime Rrs values with CEE-affected Rrc values in a time window of 1 h. These matchups were used to develop DLACC. Validations using in situ data from Aerosol Robotic Network-Ocean Color (AERONET-OC) stations showed that more matchups were obtained with the DLACC model than with the NIR algorithm and Korea Ocean Satellite Center standard AC algorithm in GDPS 2.0 (KOSC), and the accuracies were similar. More importantly, the DLACC algorithm is more tolerant of AEs and reduces AEs in a 2-pixel distance from cloud edges by 10% in the blue bands and by 50% in the green band. As a result, more valid observations are obtained with the DLACC algorithm, with the daily percentage of valid observations (DPVOs) increasing by 71% and 62% over those with the NIR algorithm and the KOSC algorithm, respectively. These increases in valid observations could further result in more consistent patterns in terms of space and time. Our DLACC algorithm can not only be used to process GOCI images but also provide an open framework to develop corresponding deep-learning models for other geostationary satellites.

Aerosol Optical Properties above Productive Waters of Gorky Reservoir for Atmospheric Correction of Sentinel-3/OLCI Images

The main challenge that one has to face during the atmospheric correction (AC) of productive inland waters is the inability to correctly separate aerosol radiance from water-leaving radiance in the near-infrared range (NIR) bands. This leads both to incorrect estimates of the aerosol parameters and the remote-sensing reflectance (Rrs). For the Gorky Reservoir, where we are developing regional bio-optical models, the situation is complicated by the lack of field measurements of aerosol optical properties due to the significant remoteness of AERONET stations. The standard AC algorithms, as shown earlier, greatly overestimated the aerosol radiance in all spectral bands up to red bands during the period of intense cyanobacteria blooms, while the algorithm with a fixed aerosol optical depth (AOD) obtained in a clean water area gave encouraging results. Therefore, it was important to investigate the characteristics of the atmosphere above the reservoir and validate the proposed approach for regular use of Sentinel-3 imagery of the Gorky Reservoir. To solve these issues, regular in situ aerosol measurements using the handheld sun photometer SPM were performed. The measured AOD and the Angstrom exponent were compared with the estimates of these parameters from two Sentinel-3/OLCI Level-2 products, namely, Synergy (SYN) and Water Full Resolution products (OL_2_WFR). It was found that AOD and the Angstrom exponent from these standard products were overestimated by 2–3 times and almost 2 times in all cases. Atmospheric correction with fixed AOD, defined by measurements, allows us to completely get rid of negative Rrs, and its shapes and values became typical for the Gorky Reservoir. Despite the overestimation of AOD in traditional AC and its large variations in general, it was found that the minimum AOD spectrum is close to the measured spectrum. Therefore, the AOD spectra, which correspond to the two percentiles of the distribution, can be used for preliminary AC with a fixed AOD of the Sentinel-3/OLCI imaginary. The relative errors of the Rrs retrievals using the two percentile AOD compared to the measured AOD were 3–35% in the green and red bands of Sentinel-3/OLCI.

Validation of ocean color remote sensing reflectance data: Analysis of results at European coastal sites

From its initial measurements 20 years ago, the Ocean Color component of the Aerosol Robotic Network (AERONET-OC) has produced large validation data sets to assess the ocean color satellite data records. This study, applied to the standard atmospheric correction algorithm l2gen of the National Aeronautics and Space Administration, analyses the populations of residuals (differences between matching satellite and field data) of remote sensing reflectance RRS and, secondarily, aerosol optical thickness τa, and their validation statistics associated with data collected at seven AERONET-OC sites located in European coastal regions for six satellite missions. Validation statistics do not appear particularly sensitive to observation conditions, represented by viewing and solar geometry, aerosol optical thickness τa, Ångström exponent or water properties indexed by RRS, the most obvious exceptions being a systematic increase of the residuals for high solar zenith angles for all missions and sites. Residuals of RRS and τa tend to be inversely correlated as deviations in aerosol radiance and water-leaving radiance compensate each other. The spectra of a RRS uncertainty estimate σ decrease in a monotonous manner from 0.7 to 1.5 10−3 sr−1 for wavelengths around 412nm to 0.1–0.35 10−3 sr−1 around 667nm, with σ usually relatively high for waters associated with high amplitudes of RRS. At 412nm, σ for τa is mostly in the interval 0.03–0.04. For RRS, variations across missions of σ or of the relative differences between satellite and field data are overall smaller than variations across sites, which suggests that the products derived from various missions are rather consistent at a given location when processed with the considered algorithm. For a given mission, RRS residuals are well correlated between bands (inter-band correlation coefficient mostly larger than 0.5), which has implications on the propagation of uncertainties through bio-optical algorithms. For a given wavelength, residuals are correlated between missions to various degrees (most often with inter-mission correlation coefficients larger than 0.5). Besides implications as far as uncertainties of multi-mission merged products are concerned, this inter-mission correlation among residuals suggest that the residuals are not random and could be reduced.

Determining the Primary Sources of Uncertainty in Retrieval of Marine Remote Sensing Reflectance From Satellite Ocean Color Sensors

Uncertainties in the retrieval of the remote sensing reflectance, Rrs, from Ocean Color (OC) satellite sensors have a strong impact on the performance of algorithms for the estimation of chlorophyll-a, mineral concentrations, and inherent optical properties (IOPs). The uncertainties are highest in the blue bands. The total radiance measured at the top of the atmosphere captures the instantaneous state of the atmosphere-ocean system: the in-water conditions, sky and Sun glint reflected from the wind-roughened ocean surface, as well as light scattered from molecules and aerosols in the atmosphere. Each of these components has associated uncertainties, and when combined with the additional uncertainties from the instrument noise and the atmospheric correction process, they contribute to the total uncertainty budget for the retrieved Rrs. We analyzed the contribution of each component uncertainties to the total Rrs uncertainties in SNPP-VIIRS level 2 products, taking advantage of the spectral differences between the components. We examined multiple scenes in the open ocean and coastal waters at spatial resolutions ranging from 2250 to 5250 m by comparing the retrieved Rrs to in situ measurements made at several AERONET-OC sites and at the MOBY site. It was shown that uncertainties associated with the molecular (Rayleigh) scattering play the most significant role, while the contributions of other components are usually smaller. Uncertainties in Rayleigh scattering are primarily attributed to the variability of Rayleigh optical thickness (ROT) with a standard deviation of approximately 1.5% of ROT, which can largely explain the frequency of negative Rrs retrievals as observed using the current standard atmospheric correction process employed by NASA. Variability of the sky light reflected from the ocean surface in some conditions also contributed to uncertainties in the blue; water variability proportional to Rrs had a very pronounced peak in the green at coastal sites.

Development of a Deep Learning-Based Atmospheric Correction Algorithm for Oligotrophic Oceans

Although the 5% mission goal for NASA’s standard atmospheric correction (AC) algorithm (i.e., the near-infrared (NIR) algorithm) for oligotrophic oceans has been met, this algorithm applies only to blue bands and is highly sensitive to contamination from cloud straylight and sunglint. Here, we developed an AC algorithm for clear waters based on deep learning (namely, DLAC). The algorithm was trained using 3.6 million pairs of MODIS-Aqua high-quality Rrs from the NIR algorithm and Rayleigh-corrected reflectances selected across the global oceans and from all seasons. Validations using <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> data and a chlorophyll (Chl) constraint-based approach showed that the uncertainties in the R <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rs</sub> retrievals for DLAC are lower than those for the NIR algorithm, especially for the green and red bands. More importantly, the DLAC algorithm is more tolerant to cloud adjacency effects and moderate sunglint. As a result, the number of valid observations increased by ~50%, and the coverage of monthly global Level-3 R <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rs</sub> composites increased by up to 20%. More spatially and temporally consistent patterns were also found for the Level-3 R <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rs</sub> and Chl products, and large changes in their magnitudes (up to 20% for R <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rs</sub> and 30% for Chl) were detected in some oceanic regions. With these improvements in the quality and quantity of data, our DLAC algorithm may be valuable as another option for processing global data.

Sensitivity of Ocean Color Atmospheric Correction to Uncertainties in Ancillary Data: A Global Analysis With SeaWiFS Data

Atmospheric correction (AC) algorithms for ocean color (OC) data processing usually rely on ancillary data documenting the atmosphere and the sea state to help the calculation of the remote sensing reflectance <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\text {RS}}$ </tex-math></inline-formula> from the radiance measured by a space sensor. This study aims at assessing the impact that the uncertainties associated with these ancillary data have on the AC outputs. For this objective, a full year of global Sea-viewing Wide Field-of-view Sensor (SeaWiFS) imagery is processed with the standard AC algorithm <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l2gen</i> of the National Aeronautics and Space Administration with different sets of ancillary data, the reference case with National Centers for Environmental Prediction (NCEP) Reanalysis-2 meteorological data and satellite ozone products, as well as with ten ensemble members from the European Centre for Medium-Range Weather Forecast (ECMWF) CERA-20C data. The spread within the ensemble data and the differences with respect to the reference case are taken as a measure of the uncertainties associated with ancillary data. The impact on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\text {RS}}$ </tex-math></inline-formula> of perturbations in ancillary variables vary in space, the variables having the largest effects being wind speed and relative humidity, and ozone at bands where ozone absorption is largest, while sea-level pressure and precipitable water have the smallest effect. Sensitivity coefficients quantifying the relationship between perturbations in ancillary variables and effects on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\text {RS}}$ </tex-math></inline-formula> change with variable and wavelength. At the global scale, the variations found on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\text {RS}}$ </tex-math></inline-formula> when ancillary data are perturbed are usually small but not negligible and should be considered in the ocean color (OC) data uncertainty budget.

Assessment of Polymer Atmospheric Correction Algorithm for Hyperspectral Remote Sensing Imagery over Coastal Waters.

Spaceborne imaging spectroscopy, also called hyperspectral remote sensing, has shown huge potential to improve current water colour retrievals and, thereby, the monitoring of inland and coastal water ecosystems. However, the quality of water colour retrievals strongly depends on successful removal of the atmospheric/surface contributions to the radiance measured by satellite sensors. Atmospheric correction (AC) algorithms are specially designed to handle these effects, but are challenged by the hundreds of narrow spectral bands obtained by hyperspectral sensors. In this paper, we investigate the performance of Polymer AC for hyperspectral remote sensing over coastal waters. Polymer is, in nature, a hyperspectral algorithm that has been mostly applied to multispectral satellite data to date. Polymer was applied to data from the Hyperspectral Imager for the Coastal Ocean (HICO), validated against in situ multispectral (AERONET-OC) and hyperspectral radiometric measurements, and its performance was compared against that of the hyperspectral version of NASA’s standard AC algorithm, L2gen. The match-up analysis demonstrated very good performance of Polymer in the green spectral region. The mean absolute percentage difference across all the visible bands varied between 16% (green spectral region) and 66% (red spectral region). Compared with L2gen, Polymer remote sensing reflectances presented lower uncertainties, greater data coverage, and higher spectral similarity to in situ measurements. These results demonstrate the potential of Polymer to perform AC on hyperspectral satellite data over coastal waters, thus supporting its application in current and future hyperspectral satellite missions.

Atmospheric correction of geostationary satellite ocean color data under high solar zenith angles in open oceans

With a revisit time of 1 h, spatial resolution of 500 m, and high radiometric sensitivity, the Geostationary Ocean Color Imager (GOCI) is widely used to monitor diurnal dynamics of oceanic phenomena. However, atmospheric correction (AC) of GOCI data with high solar zenith angle (>70°) is still a challenge for traditional algorithms. Here, we propose a novel neural network (NN) AC algorithm for GOCI data under high solar zenith angles. Unlike traditional NN AC algorithms trained by radiative transfer-simulated dataset, our new AC algorithm was trained by a large number of matchups between GOCI-observed Rayleigh-corrected radiance in the morning and evening and GOCI-retrieved high-quality noontime remote-sensing reflectance (Rrs). When validated using hourly GOCI data, the new NN AC algorithm yielded diurnally stable Rrs in open ocean waters from the morning to evening. Furthermore, when validated by in-situ data from three Aerosol Robotic Network-Ocean Color (AERONET-OC) stations (Socheongcho, Gageocho and Ieodo), the GOCI-retrieved Rrs at visible bands obtained using the new AC algorithm agreed well with the in-situ values, even under high solar zenith angles. Practical application of the new algorithm was further examined using diurnal GOCI observation data acquired in clear open ocean waters. Results showed that the new algorithm successfully retrieved Rrs for the morning and evening GOCI data. Moreover, the amount of Rrs data retrieved by the new algorithm was much higher than that retrieved by the standard AC algorithm in SeaDAS. Our proposed NN AC algorithm can not only be applied to process GOCI data acquired in the morning and evening, but also has the potential to be applied to process polar-orbiting satellite ocean color data at high-latitude ocean that also include satellite observation with high solar zenith angles.

Performance of POLYMER Atmospheric Correction of Ocean Color Imagery in the Presence of Absorbing Aerosols

The atmospheric correction approach currently being used operationally by NASA [termed as NASA standard atmospheric correction (NSAC) approach] to process ocean color data relies on traditional “black pixel” approach, with additional modifications to account for nonnegligible water-leaving radiance in the near-infrared (NIR) bands. The NSAC approach underestimates remote-sensing reflectance ( $R_{\mathrm {rs}}$ , sr−1) in blue wavelengths in the presence of absorbing aerosols. Addressing this issue requires realistic absorbing-aerosol model and knowledge of the vertical distribution of aerosols, which are currently difficult to achieve. An alternative atmospheric correction approach has been evaluated in this paper for Moderate Resolution Imaging Spectroradiometer (MODIS) data. The approach is based on a previously developed spectra-matching optimization [POLYnomial-based approach established for the atmospheric correction of MERIS data (POLYMER)], where polynomial functions are used to express atmospheric contribution to the measured radiance and where a bio-optical model is used to estimate the water contribution. Evaluation against in situ data measured over the regions frequently affected by absorbing aerosols indicates that, compared with the NSAC approach, the POLYMER approach improves the $R_{\mathrm {rs}}$ retrievals in blue wavelengths while having a slightly worse performance in other wavelengths. Evaluation using NSAC-retrieved $R_{\mathrm {rs}}$ in adjacent days free of absorbing aerosols suggests that the POLYMER approach could improve the spectral shape and increase valid spatial coverage. When applied to time-series MODIS data, the POLYMER approach could generate more temporary coherent daily and monthly $R_{\mathrm {rs}}$ patterns than the NSAC approach. These results suggest that the POLYMER approach could be an alternative approach to partly correct for absorbing aerosols in the absence of explicit information on the aerosol type and the vertical distribution.

Multiband Atmospheric Correction Algorithm for Ocean Color Retrievals

NASA’s current Atmospheric Correction (AC) algorithm for ocean color utilizes two bands and their ratio in the Near Infrared (NIR) to estimate aerosol reflectance and aerosol type. The algorithm then extrapolates the spectral dependence of aerosol reflectance to the visible wavelengths based on modeled spectral dependence of the identified aerosol type. Future advanced ocean color sensors, such as the Ocean Color Instrument (OCI) that will be carried on the Plankton, Aerosol, Cloud, and ocean Ecosystem (PACE) satellite, will be capable of measuring the hyperspectral radiance from 340 to 890 nm at 5-nm spectral resolution and at 7 discrete Short-wave Infrared (SWIR) channels: 940, 1038, 1250, 1378, 1615, 2130, and 2260 nm. To optimally employ this unprecedented instrument capability, we propose an improved AC algorithm that utilizes all atmospheric-window channels in the NIR to SWIR spectral range to reduce the uncertainty in the AC process. A theoretical uncertainty analysis of this, namely Multi-Band AC (MBAC), indicates that the algorithm can reduce the uncertainty in remote sensing reflectance (Rrs) retrievals of the ocean caused by sensor random noise. Furthermore, in optically complex waters, where the NIR signal is affected by contributions from highly-reflective turbid waters, the MBAC algorithm can be adaptively weighted to the strongly-absorbing SWIR channels to enable improved ocean color retrievals in coastal waters. We provide here a description of the algorithm and demonstrate the improved performance in ocean color retrievals, relative to the current NASA standard AC algorithm, through comparison with field measurements and assessment of propagated uncertainties in applying the MBAC algorithm to MODIS and simulated PACE OCI data.

Atmospheric Correction of OLCI Imagery over Extremely Turbid Waters Based on the Red, NIR and 1016 nm Bands and a New Baseline Residual Technique

A common approach to the pixel-by-pixel atmospheric correction of satellite water colour imagery is to calculate aerosol and water reflectance at two spectral bands, typically in the near infra-red (NIR, 700–1000 nm) or the short-wave-infra-red (SWIR, 1000–3000 nm), and then extrapolate aerosol reflectance to shorter wavelengths. For clear waters, this can be achieved simply for NIR bands, where the water reflectance can be assumed negligible i.e., the “black water” assumption. For moderately turbid waters, either the NIR water reflectance, which is non-negligible, must be modelled or longer wavelength SWIR bands, with negligible water reflectance, must be used. For extremely turbid waters, modelling of non-zero NIR water reflectance becomes uncertain because the spectral slopes of water and aerosol reflectance in the NIR become similar, making it difficult to distinguish between them. In such waters the use of SWIR bands is definitely preferred and the use of the MODIS bands at 1240 nm and 2130 nm is clearly established although, on many sensors such as the Ocean and Land Colour Instrument (OLCI), such SWIR bands are not included. Instead, a new, cheaper SWIR band at 1016 nm is available on OLCI with potential for much better atmospheric correction over extremely turbid waters. That potential is tested here. In this work, we demonstrate that for spectrally-close band triplets (such as OLCI bands at 779–865–1016 nm), the Rayleigh-corrected reflectance of the triplet’s “middle” band after baseline subtraction (or baseline residual, BLR) is essentially independent of the atmospheric conditions. We use the three BLRs defined by three consecutive band triplets of the group of bands 620–709–779–865–1016 nm to calculate water reflectance and hence aerosol reflectance at these wavelengths. Comparison with standard atmospheric correction algorithms shows similar performance in moderately turbid and clear waters and a considerable improvement in extremely turbid waters.

Vicarious calibration of GOCI for the SeaDAS ocean color retrieval

ABSTRACTA key on-orbit calibration step for satellite remote sensing of ocean color is the vicarious calibration. This establishes the final gains for each spectral band on the sensor that minimize bias in the retrieved ocean color signal. The vicarious calibration is specific to the instrument and the atmospheric correction algorithm. The vicarious calibration gains for the Geostationary Ocean Color Imager (GOCI) are presented here, which were derived to optimize the performance of NASA’s standard atmospheric correction algorithm as implemented in the l2gen code and distributed through the SeaDAS open-source software package. Following NASA’s protocols, the near-infrared (NIR) bands were calibrated first, and the visible bands were then calibrated relative to this fixed NIR calibration. The gain for the 745-nm NIR band was derived using a fixed aerosol model, which was chosen based on the Angstrom Coefficients derived from MODIS on Aqua (MODISA). For the vicarious gains of the visible bands, two sources for the target water-leaving radiances were tested: matchups from MODISA and climatological data from SeaWiFS. A validation analysis using AERONET-OC data shows an improvement in sensor performance when compared with results using the current vicarious gains and results using no vicarious calibration. Good agreement was found in vicarious gains derived using both concurrent MODISA and climatological SeaWiFS as vicarious calibration data sources. These results support the use of a concurrent sensor for the vicarious calibration when in situ data are not available and demonstrate that using climatology from a well-calibrated sensor like SeaWiFS for the vicarious calibration is a valid alternative when it is not possible to use a concurrent sensor or in situ data. We recommend using the gains derived from concurrent GOCI matchups with MODISA for GOCI processing in SeaDAS/l2gen.

Atmospheric Correction Using the Information From the Short Blue Band

We describe our effort to use normalized water-leaving radiance at the short blue band 410 nm, nLw (410), derived from the Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the Suomi National Polar-orbiting Partnership to improve the VIIRS ocean color products over coastal and inland waters, in particular, over the regions contaminated by strongly absorbing aerosols. The current standard atmospheric correction algorithm has significant issues when dealing with cases of strongly absorbing aerosols, e.g., dust, smoke, and air pollution from nearby cities. For such cases, satellite-derived normalized water-leaving radiance spectra nLw ( $\lambda $ ) are usually biased low and may be negative, particularly for the short blue bands, e.g., for the VIIRS nLw (410). In addition, for extremely turbid waters and waters dominated by colored dissolved organic matter (CDOM) that is strongly absorbing at the short blue band, slightly negative nLw (410) data are also sometimes observed. Obviously, cases with nLw (410) nLw (410) nLw (410) information. Specifically, for cases with nLw (410) nLw (410) $\ge 0$ [e.g., assuming nLw (410) = 0], thereby removing unphysical retrievals for VIIRS-derived nLw ( $ \lambda $ ) spectra. Using this technique, VIIRS-derived nLw ( $ \lambda $ ) spectra are now all ≥ 0, showing considerable improvements for VIIRS ocean color products, particularly for biological and biogeochemical products [e.g., chlorophyll-a concentration and diffuse attenuation coefficient at 490 nm $K_{d}$ (490)] which are derived using VIIRS nLw ( $\lambda $ ) spectra. Several detailed examples from VIIRS measurements over various coastal and inland waters are provided and discussed. The new technique has been implemented in the Multi-Sensor Level-1 to Level-2 (MSL12) ocean color data processing system, which has been used for the routine production of VIIRS global ocean color products.

Comparison of two atmospheric correction approaches applied to MODIS measurements over North American waters

Using in situ data of spectral remote sensing reflectance (Rrs, sr−1) collected over North American oceanic, coastal and estuarine waters between 2002 and 2016 (N = 942), we evaluate two atmospheric correction approaches applied to MODIS measurements. One is the POLYnomial based approach originally designed for MERIS (POLYMER) but adopted and implemented for MODIS, and the other is the traditional Gordon and Wang (1994b) near-infrared (NIR) approach with iteration to account for non-negligible NIR water-leaving radiance, which is currently embedded in the SeaWiFS Data Analysis System (SeaDAS) software package and used operationally by NASA for processing MODIS data (termed as NASA standard atmospheric correction or NSAC). The approaches are evaluated for both quality and quantity of their retrieved Rrs in the visible domain. The quality is gauged through three statistical measures between in situ and MODIS-retrieved Rrs: root mean square error (RMSE, sr−1), unbiased root mean square (uRMS), and mean bias (δ, sr−1). For common points where both approaches yield valid Rrs retrievals, POLYMER shows worse performance than NSAC for blue bands (<488 nm) and comparable performance for green and red bands. However, POLYMER shows the ability to retrieve more valid Rrs data points (2–3 folds) than NSAC for this evaluation dataset primarily because the latter fails over strong sun glint regions where the MODIS NIR bands saturate but the former is designed to work over sun glint regions using non-saturation MODIS bands. For those data points where only POLYMER yield valid Rrs retrievals, data quality is slightly worse than from the common data points. Although these results may vary slightly among individual subregions, it is generally true that POLYMER may be used as a surrogate of NSAC for atmospheric correction of MODIS when data quantity is significantly limited due to perturbations of sun glint and thin clouds that are typical for subtropical and tropical regions.

Simultaneous determination of aerosol optical thickness and water-leaving radiance from multispectral measurements in coastal waters

Abstract. Retrieval of aerosol optical properties and water-leaving radiance over ocean is challenging since the latter mostly accounts for ∼ 10 % of the satellite-observed signal and can be easily influenced by the atmospheric scattering. Such an effort would be more difficult in turbid coastal waters due to the existence of optically complex oceanic substances or high aerosol loading. In an effort to solve such problems, we present an optimization approach for the simultaneous determination of aerosol optical thickness (AOT) and normalized water-leaving radiance (nLw) from multispectral satellite measurements. In this algorithm, a coupled atmosphere–ocean radiative transfer model combined with a comprehensive bio-optical oceanic module is used to jointly simulate the satellite-observed reflectance at the top of atmosphere and water-leaving radiance just above the ocean surface. Then, an optimal estimation method is adopted to retrieve AOT and nLw iteratively. The algorithm is validated using Aerosol Robotic Network – Ocean Color (AERONET-OC) products selected from eight OC sites distributed over different waters, consisting of observations that covered glint and non-glint conditions from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument. Results show a good consistency between retrieved and in situ measurements at each site. It is demonstrated that more accurate AOTs are determined based on the simultaneous retrieval method, particularly in shorter wavelengths and sunglint conditions, where the averaged percentage difference (APD) of retrieved AOT is generally reduced by approximate 10 % in visible bands compared with those derived from the standard atmospheric correction (AC) scheme, since all the spectral measurements can be used jointly to increase the information content in the inversion of AOT, and the wind speed is also simultaneously retrieved to compensate the specular reflectance error estimated from the rough ocean surface model. For the retrieval of nLw, atmospheric overcorrection can be avoided in order to have a significant improvement of the inversion of nLw at 412 nm. Furthermore, generally better estimates of band ratios of nLw(443) / nLw(554) and nLw(488) / nLw(554) are obtained using the simultaneous retrieval approach with lower root mean square errors and relative differences than those derived from the standard AC approach in comparison to the AERONET-OC products, as well as the APD values of retrieved Chl which decreased by about 5 %. On the other hand, the standard AC scheme yields a more accurate retrieval of nLw at 488 nm, prompting a further optimization of the oceanic bio-optical module of the current model.

Sensor performance requirements for atmospheric correction of satellite ocean color remote sensing.

We analyze the effects of the sensor signal-to-noise ratio (SNR) requirements for atmospheric correction of satellite ocean color remote sensing using the near-infrared (NIR) and shortwave infrared (SWIR) bands. Using the Gaussian noise model for the sensor noise distribution in the NIR and SWIR bands, some extensive simulations have been carried out to evaluate and assess the effects of sensor NIR and SWIR SNR values on the retrieved normalized water-leaving reflectance spectra ρwN(λ), which are used to derive all ocean or inland water biological and biogeochemical property data. The standard atmospheric correction algorithm for global oceans and inland waters using the two NIR bands, i.e., Gordon and Wang (1994) [Appl. Opt.33, 443 (1994)Appl. Opt.46, 1535 (2007)], is assumed in the evaluation. Specifically, the minimum and goal SNR requirements for the NIR and SWIR bands for atmospheric correction are estimated. The minimum SNR values are those with which sufficiently accurate ρwN(λ) can be derived, while the goal SNR requirements are those with which the atmospheric correction algorithms reach to their corresponding inherent limitations (or inherent errors), i.e., no gains can be achieved with further increase of SNR values in the NIR and SWIR bands. Evaluation results show that the minimum SNR requirement for the two NIR bands is ~200-300, while the minimum SNR requirement for the three SWIR bands is ~100. For the goal SNR requirements, the recommendations are SNR's of ~600 and ~200 for the two NIR bands and three SWIR bands, respectively.

Using High-Resolution Airborne Data to Evaluate MERIS Atmospheric Correction and Intra-Pixel Variability in Nearshore Turbid Waters

The implementation of accurate atmospheric correction is a prerequisite for satellite observation and water quality monitoring in coastal areas. The potential of the fast-line-of-sight atmospheric analysis of spectral hypercubes (FLAASH) was investigated here for the medium resolution imaging spectrometer (MERIS). As the comparison between discrete field sampling points and macro-scale satellite pixels is subject to spatial biases associated with small-scale spatial patchiness in the turbid and highly dynamic nearshore zone, an alternative approach was proposed here using high spatial resolution (1 m) airborne hyperspectral images as radiometric truthing references. While FLAASH was not optimal for moderately turbid offshore waters (suspended particulate matter (SPM) concentration &lt; 50 g∙m−3), it yields satisfactory results in the 50–1500 g∙m−3 range, where MERIS standard atmospheric correction was subject to significant biases and failures. Due to the significant intra-pixel variability of SPM distribution in highly turbid areas, the acquisition of high resolution airborne images should be considered as a consistent strategy for the validation of medium resolution satellite remote sensing in the spatially heterogeneous and optically diverse nearshore waters.

Atmospheric correction over coastal waters using multilayer neural networks

Standard atmospheric correction (AC) algorithms work well in open ocean areas where the water inherent optical properties (IOPs) are correlated with pigmented particles. However, the IOPs of turbid coastal waters may independently vary with pigmented particles, suspended inorganic particles, and colored dissolved organic matter (CDOM). In turbid coastal waters standard AC algorithms often exhibit large inaccuracies that may lead to negative water-leaving radiances (Lw) or remote sensing reflectance (Rrs). We introduce a new atmospheric correction algorithm for coastal waters based on a multilayer neural network (MLNN) method. We use a coupled atmosphere-ocean radiative transfer model to simulate the Rayleigh-corrected radiance (Lrc) at the top of the atmosphere (TOA) and the Rrs just above the surface simultaneously, and train a MLNN to derive the aerosol optical depth (AOD) and Rrs directly from the TOA Lrc. The method is validated using both a synthetic dataset and Aerosol Robotic Network – Ocean Color (AERONET–OC) measurements. The SeaDAS NIR algorithm, the SeaDAS NIR/SWIR algorithm, and the MODIS version of the Case 2 regional water - CoastColour (C2RCC) algorithm are also included in the comparison with AERONET–OC measurements. The performance of the AC algorithms is evaluated with four statistical metrics: the Pearson correlation coefficient (R), the average percentage difference (APD), the mean percentage bias, and the root mean square difference (RMSD). The comparison with AERONET–OC measurements shows that the MLNN algorithm significantly improves retrieval of normalized Lw in blue bands (412nm and 443nm) and yields minor improvements in green and red bands compared with the other three algorithms. On a global scale, the MLNN algorithm reduces APD in normalized Lw by up to 13% in blue bands and by 2–7% in green and red bands when compared with the standard SeaDAS NIR algorithm. In highly absorbing coastal waters, such as the Baltic Sea, the MLNN algorithm reduces APD in normalized Lw by more than 60% in blue bands compared to the standard SeaDAS NIR algorithm, while in highly scattering coastal waters, such as the Black Sea, the MLNN algorithm reduces APD by more than 25%. These results indicate that the MLNN algorithm is suitable for application in turbid coastal waters. Application of the MLNN algorithm to MODIS Aqua images in several coastal areas also shows that it is robust and resilient to contamination due to sunglint or adjacency effects of land and cloud edges. The MLNN algorithm is very fast once the neural network has been properly trained and is therefore suitable for operational use. A significant advantage of the MLNN algorithm is that it does not need SWIR bands.