Ocean Color Research Articles

Using the automated HYPERNETS hyperspectral system for multi-mission satellite ocean colour validation in the Río de la Plata, accounting for different spatial resolutions

Validation of water reflectance using in situ data is essential to ensure the quality of ocean colour satellite-derived products useful for water quality monitoring, like turbidity and chlorophyll-a concentration. Since December 2021, the HYPERNETS automated hyperspectral system has been collecting data in the optically complex and highly turbid waters of the Río de la Plata, an ideal scenario for testing atmospheric correction algorithms’ performance. The site, located 60 km south of Buenos Aires (Argentina), is described in relation to the water reflectance spectral features and variability using high spatial resolution imagery and a methodology is proposed to objectively select a sensor-specific location of a reference pixel for satellite validation. Six months of data is used to evaluate surface water reflectance operational products from multi-spectral systems like Landsat 8&9/OLI (L89/OLI), Sentinel-2/MSI (S2/MSI) & Sentinel-3/OLCI (S3/OLCI), and PlanetScope SuperDoves (PS/SD), and also non standard products for Aqua/MODIS (Aqua/MODIS) and SNPP&JPSS1/VIIRS (SJ/VIIRS) missions. Moreover, the standard surface water reflectance product from the hyperspectral PRISMA mission could also be evaluated. The matchups show general good results when in situ measurements are compared to L2 standard products of high spatial resolution sensors that use land-based atmospheric correction approach, if sun glint contamination is avoided. Low mean relative percentage difference was found for S2/MSI (2.45%) and L89/OLI (−3.52%), but higher for PS/SD (30.7%). In turn, S3/OLCI medium resolution also showed low mean relative differences (2.31%), while SJ/VIIRS and Aqua/MODIS showed larger and negative differences (−16.35 for SJ/VIIRS and −35.6% for Aqua/MODIS) which showed a clear increase towards the shortest blue bands. The results show the great potential of the HYPERNETS automated system to provide high quality and quantity of data for validation of satellite data at all visible and near infrared (VNIR, 400–900 nm) wavelengths in a multi-mission perspective.

Evaluation and Refinement of Chlorophyll-a Algorithms for High-Biomass Blooms in San Francisco Bay (USA)

A massive bloom of the raphidophyte Heterosigma akashiwo occurred in summer 2022 in San Francisco Bay, causing widespread ecological impacts including events of low dissolved oxygen and mass fish kills. The rapidly evolving bloom required equally rapid management response, leading to the use of near-real-time image analysis of chlorophyll from the Ocean and Land Colour Instrument (OLCI) aboard Sentinel-3. Standard algorithms failed to adequately capture the bloom, signifying a need to refine a two-band algorithm developed for coastal and inland waters that relates the red-edge part of the remote sensing reflectance spectrum to chlorophyll. While the bloom was the initial motivation for optimizing this algorithm, an extensive dataset of in-water validation measurements from both bloom and non-bloom periods was used to evaluate performance over a range of concentrations and community composition. The modified red-edge algorithm with a simplified atmospheric correction scheme outperformed existing standard products across diverse conditions, and given the modest computational requirements, was found suitable for operational use and near-real-time product generation. The final version of the algorithm successfully minimizes error for non-bloom periods when chlorophyll a is typically <30 mg m−3, while also capturing bloom periods of >100 mg m−3 chlorophyll a.

Application of deep learning in predicting suspended sediment concentration: A case study in Jiaozhou Bay, China

Previous research methodologies for quantifying Suspended Sediment Concentration (SSC) have encompassed in-situ observations, numerical simulations, and analyses of remote sensing datasets, each with inherent constraints. In this study, we have harnessed Convolutional Neural Networks (CNNs) to create a deep learning model, which has been applied to the remote sensing data procured from the Geostationary Ocean Color Imager (GOCI) spanning April 2011 to March 2021. Our research indicates that on a small time scale, wind and hydrodynamic forces both have a significant impact on the prediction results of CNNs model. Considering both wind and hydrodynamic forces can effectively improve the model's prediction efficiency for SSC. Moreover, we have employed CNNs to interpolate absent values within the remote sensing datasets, yielding enhancements superior to those attained via linear or multivariate regression techniques. Finally, the correlation coefficient between CNN-derived SSC estimates for Jiaozhou Bay (JZB) and its corresponding remote sensing data is 0.72. Correlation coefficient and root mean square error differ in different regions. In the shallow water of JZB, due to water level changes, there is limited data, and the correlation coefficient in this area is about 0.5–0.6. In the central region of JZB with sufficient data, the correlation coefficient is generally higher than 0.75. Therefore, we believe that this CNNs model can be used to predict the hourly variation of SSC. When juxtaposed with alternative methodologies, the CNN approach is found to economize computational resources and enhance processing efficiency.

Deconstructing co-occurring marine heatwave and phytoplankton bloom events in the Arkona Sea in 2018

Between May and August 2018, two separate marine heatwaves (MHWs) occurred in the Arkona Sea in the western Baltic Sea. These heatwaves bookended an extended period of phytoplankton growth in the region. Data from the Ocean and Land Colour Instrument (OLCI) on board the European Sentinel-3 satellite revealed an eddy-like structure containing high chlorophyll a (Chl-a) concentrations (ca. 25 mg.m-3) persisting for several days at the end of May in the Arkona Sea. Combining ocean colour observations, a coupled bio-optical ocean model and a particle tracking model, we examined the three dimensional relationship between these co-occurring MHW and phytoplankton bloom events. We find that the onset of the MHW in May provided the optimal conditions for phytoplankton growth, i.e. sufficient light and nutrients. Wind-driven surface eddy circulation, geostrophic eddy stirring and transient submesoscale dynamics along the edges of the eddy provided a transport path for nutrient fluxes and carbon export, and helped to sustain the phytoplankton bloom. The bloom may have indirectly had an enhancing effect on the MHW, through the impact of water constituent-induced heating rates on air-sea energy fluxes. The subsurface signature of the MHW plays a critical role in de-coupling surface and subsurface dynamics and terminating the phytoplankton bloom. Subsurface temperature anomalies of up to 8°C between 15 and 20 m depth are found to persist up to 15 days after the surface signature of the MHW has disappeared. The study reveals how surface and subsurface dynamics of MHWs and phytoplankton blooms are connected under different environmental conditions. It extends our knowledge on surface layer processes obtained from satellite data.

A Lagrangian coherent eddy atlas for biogeochemical applications in the North Pacific Subtropical Gyre

Abstract. Mesoscale eddies affect phytoplankton in several ways, including the horizontal dispersal and mixing of populations. Coherent eddies trap and contain fluid masses, whereas other eddies mix more freely with surrounding waters. To evaluate the role of lateral dispersal and trapping on the biogeochemical properties of eddies, we must accurately characterize their coherency. We employed a Lagrangian approach to identify materially coherent structures in remote sensing observations and developed a methodology to track them over their entire individual lifetimes. We provide an atlas of two decades of coherent eddies with an 8 d resolution in an intensely studied region of the North Pacific Subtropical Gyre (Jones-Kellett, 2023a, https://doi.org/10.5281/zenodo.8139149). The atlas was specifically designed to facilitate biogeochemical applications and was synchronized with available ocean color products. We identified coherent features using backward Lagrangian trajectories because the recent history of coherency (rather than the future coherency) will be most valuable for interpreting associated biogeochemical signatures. We compared the atlas of Lagrangian coherent eddies with an atlas of Eulerian eddies identified using the more conventionally used Sea Level Anomaly method. Although 65 % of tracked Sea Level Anomaly eddies are classified as coherent at some point in their lifetime, only 54 % contain a Lagrangian coherent structure at any given time. We found similar variations in the temporal and spatial distributions of coherent and Sea Level Anomaly eddies. However, strictly coherent eddies have a clearer relationship between size and longevity and form distinct regional regimes based on polarity. We illustrate the potential of the coherent eddy atlas for biogeochemical applications by examining the relationship between bloom development and eddy evolution in a case study of a Hawaiian Lee cyclone.

Reconstructing hyper-spectral downwelling irradiance from multi-spectral measurements

Checking the radiometric calibration of satellite hyper-spectral sensors such as the PACE Ocean Color Instrument (OCI) while they operate in orbit and evaluating remote sensing reflectance, the basic variable from which a variety of optical and biogeochemical ocean properties can be derived, requires measuring upwelling radiance just above the surface (Lw) and downwelling planar irradiance reaching the surface (Es). For this, the current HyperNav systems measure Lw at about 2 nm spectral resolution in the ultraviolet to near infrared, but Es in only four 10 nm wide spectral bands centered on 412, 489, 555, and 705 nm. In this study, the Es data acquired in these spectral bands in clear sky conditions are used to reconstruct via a multi-linear regression model the hyper-spectral Es signal at 0.5 nm resolution from 315 to 900 nm, the OCI spectral range, allowing an estimate of Es at the HyperNav, OCI, and other sensors’ resolutions. After correction of gaseous absorption and normalization by the top-of-atmosphere incident solar flux, the atmospheric diffuse transmittance is expressed as a linear combination of Es measured in those 4 spectral bands. Based on simulations for Sun zenith angles from 0 to 75° and a wide range of (i.e., expected) atmospheric, surface, and water conditions, the Es spectrum is reconstructed with a bias of less than 0.4% in magnitude and an RMS error (RMSE) ranging from 0% to 2.5%, depending on wavelength. The largest errors occur in spectral regions with strong gaseous absorption. In the presence of typical noise on Es measurements and uncertainties on the ancillary variables, the bias and RMSE become −2.5% and 7.0%, respectively. Using a General Additive Model with coefficients depending on Sun zenith angle and aerosol optical thickness at 550 nm improves statistical performance in the absence of noise, especially in the ultraviolet, but provides similar performance on noisy data, indicating more sensitivity to noise. Adding spectral bands in the ultraviolet, e.g., centered on 325, 340, and 380 nm, yields marginally more accurate results in the ultraviolet, due to uncertainties in the gaseous transmittance. Comparisons between the measured and reconstructed Es spectra acquired by the MOBY spectroradiometer show agreement within predicted uncertainties, i.e., biases less than 2% in magnitude and RMS differences less than 5%. Reconstruction can also be achieved accurately with other sets of spectral bands and extended to cloudy conditions since cloud optical properties, like aerosol properties, tend to vary regularly with wavelength. These results indicate that it is sufficient, for many scientific applications involving hyper-spectral Es, to measure Es in a few coarse spectral bands in the ultraviolet to near infrared and reconstruct the hyperspectral signal using the proposed multivariate linear modeling.

The phenology of the spring phytoplankton bloom in the North Atlantic does not trend with temperature

Climate change is anticipated to alter the phenology of phytoplankton blooms in the ocean, making their recent dynamics of interest to inform models of future ocean states. We characterized temperature change in the North Atlantic using metrics that track the patterns of sea surface water temperature (SST) defined by quantiles. To complement these thermal indicators, we estimated a thermal phenology index in the form of the date of the spring transition, taken as the date that temperature achieved the long-term mean at a specific location. We then used ocean color data (1998–2022) and characterized spring bloom phenology using change point methods to derive bloom initiation, duration, magnitude, and intensity. The North Atlantic has warmed over recent decades, averaging a rate of increase of 0.27°C decade−1, yet throughout most of the basin, spring transition timing has remained constant, with the exception of small areas with either delayed or advanced transitions. There were no clear trends in bloom start or duration in the North Atlantic, indicating that spring bloom phenology was independent of climate-driven temperature change. Bloom magnitude and intensity trended downward in some North Atlantic continental shelf seas, indicating that increased temperatures may have had negative effects on overall bloom productivity. However, exclusive of the areas where the bloom parameters were trending, there was a decrease in magnitude and intensity with warmer winter temperatures, suggesting that the inter-annual variability of these parameters may be affected by thermal conditions at the onset of the bloom. While temperature has increased in the North Atlantic, vernal light availability has remained unchanged, which may explain why spring bloom phenology has remained resistant to climate change. Consequently, it seems plausible that future climate change may have limited effects on spring bloom phenology, but could have substantial effects on overall phytoplankton production.

Variability of chlorophyll a concentration in surface waters of the open Baltic Sea

In situ, satellite and reanalysis data from numerical models were used to study the characteristic features of Chl variability in the Baltic Sea. The analysis is focused on the years 2003–2020 when regular observations of ocean color with the MODIS AQUA are available. In the Baltic Sea, there is a pronounced annual cycle in physical conditions in the water column, driven by seasonal cycles in atmospheric forcing. The seasonal cycle of Chl concentration does not conform to the picture known from classical models, with low phytoplankton concentration when nutrients are low. In contrast, in the Baltic Sea, the concentration of Chl is high even during the summer months when nutrients are depleted. This can be explained by a continuous supply of nutrients by runoff from land, as well as by a significant contribution to primary production by phytoplankton able to survive in environment poor in dissolved nutrients. There is also a considerable interannual variability in Chl. There are many possible cause/effect interactions involved, but the data series are still too short to make clear which of them are the most important. The most striking event was a spring bloom in 2008.

A novel water optical types framework for Chinese inland waters with the application of multitype satellite sensor

ABSTRACT Optical Water Type (OWT) analysis is crucial for comprehending water composition and quality, key factors in assessing water quality over extensive areas. However, China’s inland waters lack a standardized system for such analysis. To quantitatively analyze the classification results, our study compared three K-means clustering methods, for analyzing 1310 spectral data from various Chinese lakes and reservoirs, thereby addressing this gap. The innovative split-merge K-means method identified 13 distinct OWTs that more closely adhere to the principles of minimizing intra-class distance and maximizing inter-class distance. These were categorized into four groups: clear water, turbid water, eutrophic water, and special type water. Additionally, we developed a method based on Spectral Angle Distance (SAD) to evaluate the classification capabilities of 12 satellite sensors. The results show that Sentinel-3 OLCI (Ocean and Land Color Instrument), MERIS (Medium Resolution Imaging Spectrometer), and Sentinel-2 MSI (Multispectral Instrument) have the best water classification capabilities, making them well-suited for large-scale monitoring of OWT changes. Conversely, other sensors, such as the Sustainable Development Scientific Satellite-1 (SDGSAT-1), Landsat-8, GaoFen-6, GaoFen-1, GaoFen-2, Landsat-5, Landsat-7, Moderate Resolution Imaging Spectroradiometer (MODIS), and HuanJing-1, necessitate the consolidation of water types for effective categorization, indicative of their more limited classification capabilities.

Particle Size Distribution Slope Changes along the Yellow River Delta Observed from Sentinel 3A/B OLCI Images

Quantitative estimates of particle size in estuaries and shelf areas are important to understand ocean ecology and biogeochemistry. Particle size can be characterized qualitatively from satellite observations of ocean color. As a typical marginal sea, the Yellow River Delta (YRD) with the Bohai Sea experiences a complex hydrodynamic environment. Here, we attempt to quantify the particle size distribution (PSD) slope (ξ) based on its relationship with the particle backscattering exponent from Sentinel-3A/B OLCI. The PSD slope, ξ displays temporal and spatial variability in the YRD with the Bohai Sea. Its value varies between 3 and 4, and typically exceeds 5 in offshore areas. The lowest value of ξ occurs in the winter, indicating the presence of fine inorganic particles in the water, while high values are attained in the spring, when phytoplankton blooms increase the particle size. ξ decreases near the river mouth because of the large sediment-laden discharge debouching into the sea. We detected a slight increase in ξ when turbid waters were present in the period 2016–2022. Environmental factors, such as sea surface temperature, sea surface wave height, and wind, may control particle size and ξ in the long term. Inorganic suspended particle matter is derived along the YRD using the magnitude of ξ. The mean inorganic suspended particle matter area in winter approaches 23,900 km2 when ξ < 4.6. This study thoroughly characterizes variations in ξ in the YRD with the Bohai Sea and clarifies the contributions of driving factors from human activities and climate change.

Observational Evidence of Cold Filamentary Intensification in an Energetic Meander of the Antarctic Circumpolar Current

Abstract Eddy stirring at mesoscale oceanic fronts generates finescale filaments, visible in submesoscale-resolving model simulations and high-resolution satellite images of sea surface temperature, ocean color, and sea ice. Submesoscale filaments have widths of O(1–10) km and evolve on time scales of hours to days, making them extremely challenging to observe. Despite their relatively small scale, submesoscale processes play a key role in the climate system by providing a route to dissipation; altering the stratification of the ocean interior; and generating strong vertical velocities that exchange heat, carbon, nutrients, and oxygen between the mixed layer and the ocean interior. We present a unique set of in situ and satellite observations in a standing meander region of the Antarctic Circumpolar Current (ACC) that supports the theory of cold filamentary intensification—revealing enhanced vertical velocities and evidence of subduction and ventilation associated with finescale cold filaments. We show that these processes are not confined to the mixed layer; EM-APEX floats reveal enhanced downward velocities (>100 m day−1) and evidence of ageostrophic motion extending as deep as 1600 dbar, associated with a ∼20-km-wide cold filament. A finer-scale (∼5 km wide) cold filament crossed by a towed Triaxus is associated with anomalous chlorophyll and oxygen values extending at least 100–200 dbar below the base of the mixed layer, implying recent subduction and ventilation. Energetic standing meanders within the weakly stratified ACC provide an environment conductive to the generation of finescale filaments that can transport water mass properties across mesoscale fronts and deep into the ocean interior.

Atmospheric correction based on the optimal solution of aerosol mix ratio: A case study for Pearl River Estuary

In ocean color remote sensing, the presence of absorbing aerosols and turbid waters will lead to increased errors in atmospheric correction (AC) for coastal regions. In order to reduce these errors, an alternative AC algorithm has been proposed based on the in situ measurements of the Pearl River Estuary (PRE) and the candidate aerosol models including continental, maritime, and absorbing urban aerosols. The new AC algorithm optimally solved the equation set which is constructed by the multi-spectral relationships of remote sensing reflectance (Rrs ) and the aerosol reflectance of each aerosol model. The performances of the new AC were evaluated using the simulated dataset and the in situ Rrs . The results indicated that this algorithm can effectively reduce the overestimation of aerosol reflectance caused by absorbing aerosols, and can retrieve reasonable Rrs values in PRE.

Fusing remote sensing data with spatiotemporal in situ samples for red tide (Karenia brevis) detection.

We present a novel method for detecting red tide (Karenia brevis) blooms off the west coast of Florida, driven by a neural network classifier that combines remote sensing data with spatiotemporally distributed in situ sample data. The network detects blooms over a 1-km grid, using seven ocean color features from the MODIS-Aqua satellite platform (2002-2021) and in situ sample data collected by the Florida Fish and Wildlife Conservation Commission and its partners. Model performance was demonstrably enhanced by two key innovations: depth normalization of satellite features and encoding of an in situ feature. The satellite features were normalized to adjust for depth-dependent bottom reflection effects in shallow coastal waters. The in situ data were used to engineer a feature that contextualizes recent nearby ground truth of K. brevis concentrations through a K-nearest neighbor spatiotemporal proximity weighting scheme. A rigorous experimental comparison revealed that our model outperforms existing remote detection methods presented in the literature and applied in practice. This classifier has strong potential to be operationalized to support more efficient monitoring and mitigation of future blooms, more accurate communication about their spatial extent and distribution, and a deeper scientific understanding of bloom dynamics, transport, drivers, and impacts in the region. This approach also has the potential to be adapted for the detection of other algal blooms in coastal waters. Integr Environ Assess Manag 2024;20:1432-1446. © 2024 SETAC.

Impacts of Hydrology and Extreme Events on Dissolved Organic Carbon Dynamics in a Heavily Urbanized Estuary and Its Major Tributaries: A View From Space

AbstractDissolved organic matter and its colored component, Colored Dissolved Organic Matter (CDOM), play a major role in global carbon budgets, and their fluxes provide an essential link between terrestrial and aquatic biogeochemical cycles. Satellite observations can uniquely capture the hydro‐biogeochemical connectivity of terrestrial and aquatic landscapes, across scales. Yet, accurate satellite retrievals of CDOM and dissolved organic carbon (DOC) dynamics remain challenging in urbanized estuaries and coasts. Here, we present an advanced unified algorithm for space‐based retrieval of coastal CDOM and DOC dynamics and its application in Long Island Sound—one of the world's most heavily urbanized estuaries that is becoming increasingly vulnerable to climate change stressors. A rich bio‐optical data set, encompassing a wide range of environmental conditions, was integrated into the algorithm training to retrieve DOC concentrations and CDOM spectral shape (i.e., spectral slope S275–295)—a proxy for DOC quality. The new algorithms were applied to full‐resolution satellite imagery from the Sentinel‐3 Ocean and Land Color Instrument (OLCI) after thoroughly evaluating the performance of six ocean color atmospheric correction approaches (ACOLITE, BAC, C2RCC, MUMM, l2gen, and Polymer). Evaluation of the algorithms yielded mean absolute percent differences of 28%, 12%, and 10% for aCDOM(300), S275–295, and DOC, respectively. Application of the algorithms to multi‐year satellite OLCI imagery captured, for the first time, the coupled impact of seasonal transitions, wind regimes, freshwater inputs, anthropogenic disturbances, and hydrological extremes (both intense precipitation and droughts) on DOC fluxes and CDOM quality at the ecosystem scale. Results have important implications for improved predictions of coastal biogeochemical fluxes in complex urban−estuary systems.

Offshore high chlorophyll eddies: Separating upwelling from nearshore migration from ocean color remote sensing

Offshore eddies are often associated with high amounts of phytoplankton (represented by the chlorophyll-a concentration (Chla)), or, phytoplankton blooms, which can be detected from ocean color satellites. The phytoplankton “blooms” in these eddies are commonly explained as a result of enhanced nutrients - local growth - brought up from deeper waters by these eddies, although potentially they could simply be a migration of high Chla waters from nearshore regions. To better understand the interactions between physical forcing and phytoplankton dynamics, it is necessary to separate these “blooms” between local growth and migration. In this study, we first updated the multiple pigment inversion model using a synthetic dataset, for retrieving the absorption coefficient and absorption Gaussian peaks of phytoplankton from remote sensing reflectance in the broad aquatic environments. On this basis, a two-dimensional spatial model was developed to identify the sources of phytoplankton associated with offshore eddies. The model was based on the absorption coefficient of phytoplankton at 443 nm (aph(443)) and the ratio of two Gaussian peaks at 519 nm and 435 nm, where these two peaks represent different contributions of phytoplankton pigments to aph. This two-dimensional spatial model was applied to images collected by the Ocean and Land Color Instrument in the California offshore region to demonstrate that the scheme effectively separated offshore upwelling waters from those migrating from nearshore waters. Such separations provide independent sources for identifying offshore upwelling water that will be important for studying offshore circulation processes.

Toward a synthesis of phytoplankton community composition methods for global‐scale application

AbstractThe composition of the marine phytoplankton community has been shown to impact many biogeochemical processes and marine ecosystem services. A variety of methods exist to characterize phytoplankton community composition (PCC), with varying degrees of taxonomic resolution. Accordingly, the resulting PCC determinations are dependent on the method used. Here, we use surface ocean samples collected in the North Atlantic and North Pacific Oceans to compare high‐performance liquid chromatography pigment‐based PCC to four other methods: quantitative cell imaging, flow cytometry, and 16S and 18S rRNA amplicon sequencing. These methods allow characterization of both prokaryotic and eukaryotic PCC across a wide range of size classes. PCC estimates of many taxa resolved at the class level (e.g., diatoms) show strong positive correlations across methods, while other groups (e.g., dinoflagellates) are not well captured by one or more methods. Since variations in phytoplankton pigment concentrations are related to changes in optical properties, this combined dataset expands the potential scope of ocean color remote sensing by associating PCC at the genus‐ and species‐level with group‐ or class‐level PCC from pigments. Quantifying the strengths and limitations of pigment‐based PCC methods compared to PCC assessments from amplicon sequencing, imaging, and cytometry methods is the first step toward the robust validation of remote sensing approaches to quantify PCC from space.

Linking satellites to genes with machine learning to estimate phytoplankton community structure from space

Abstract. Ocean color remote sensing has been used for more than 2 decades to estimate primary productivity. Approaches have also been developed to disentangle phytoplankton community structure based on spectral data from space, in particular when combined with in situ measurements of photosynthetic pigments. Here, we propose a new ocean color algorithm to derive the relative cell abundance of seven phytoplankton groups, as well as their contribution to total chlorophyll a (Chl a) at the global scale. Our algorithm is based on machine learning and has been trained using remotely sensed parameters (reflectance, backscattering, and attenuation coefficients at different wavelengths, plus temperature and Chl a) combined with an omics-based biomarker developed using Tara Oceans data representing a single-copy gene encoding a component of the photosynthetic machinery that is present across all phytoplankton, including both prokaryotes and eukaryotes. It differs from previous methods which rely on diagnostic pigments to derive phytoplankton groups. Our methodology provides robust estimates of the phytoplankton community structure in terms of relative cell abundance and contribution to total Chl a concentration. The newly generated datasets yield complementary information about different aspects of phytoplankton that are valuable for assessing the contributions of different phytoplankton groups to primary productivity and inferring community assembly processes. This makes remote sensing observations excellent tools to collect essential biodiversity variables (EBVs) and provide a foundation for developing marine biodiversity forecasts.

Sources of uncertainty in satellite-derived chlorophyll-a concentration—An Adriatic Sea case study

This paper analyses a time series of chlorophyll-a profiles in the Adriatic from 1997 to 2019, and compares the data with satellite products with the view of analysing and reducing uncertainties in the corresponding satellite products. Three sources of uncertainties in satellite chlorophyll-a concentration are examined: (a) the algorithm itself; (b) the vertical structure of the water column; and (c) the phytoplankton community structure. Global and regional algorithms were examined, along with a local algorithm tuned using the time series data. The global algorithm produced the largest uncertainties when compared with the in situ data, followed by the regional and local algorithms. Correlation coefficient for the local algorithm was 0.690 - a significant increase from regional’s 0.420 and global’s 0.042. Both the global and the regional algorithms exhibited systemic errors that inversely were related to chlorophyll-a concentration, while the local algorithm displayed some reduction in the systematic errors, highlighting the value of local in situ observations, for improving sub-regional and local algorithms for retrieval of chlorophyll-a concentration from satellite ocean colour data. While the mixed layer has not shown any direct correlation with the uncertainties, it may facilitate exceptionally strong vertical gradients in chlorophyll-a profiles after summer blooms that take role as the main source of high differences between satellite observations and surface chlorophyll-a concentration. As such, it is important to supplement satellite measurements with vertical profiles to ensure valid readings and exercise caution when dealing with data post-blooms. These instances occurred in less than 3% of all cases. Differences in the phytoplankton community structures have shown direct correlation to estimation error - Miozoa is associated with low error, Bacillariophyta with high error, while Phytoflagellates abundance tips the error between underestimation and overestimation.

A gap-filling method for satellite-derived chlorophyll-a time series based on neighborhood spatiotemporal information

Satellite-derived Chlorophyll-a concentration (Chla) time series products are essential for large-scale marine environmental monitoring. However, the plenty of missing pixels in current satellite Chla products severely hinder their applications for marine research, due to cloud contamination, solar glint, and unfavorable observation conditions. This study proposed a Chla time series gap-filling method for MODIS 8-day composite Chla product by integrating spatiotemporal information (STGF). This method employed spatially neighboring pixels with similar temporal variation to fill the missing values in time series, without involving training or auxiliary data. The performance of the STGF is assessed quantitatively and qualitatively. The correlation coefficients (CC) between gap-filled data and actual observations for years of 2004, 2010, 2016, and 2022 across the entire study area are all greater than 0.97. The mean absolute percentage error (MAPE) and root mean square error (RMSE) are less than 16.1 % and 0.233 mg/m3, respectively. The proposed STGF outperformed linear interpolation and the DINEOF algorithm from both spatial and temporal perspectives, suggesting the effectiveness of STGF in handling continuous data gaps and capturing detailed Chla variation patterns, especially in regions with significant variability. The findings suggest that the proposed STGF method offers a viable alternative for filling missing values in Chla time series data. This supports the demand for long-term, large-scale, and high-coverage ocean color remote sensing data in marine environmental studies.

A new combined atmospheric correction algorithm for GOCI-2 Data over coastal waters assessed by long-term satellite ocean color platforms

ABSTRACT Accurate atmospheric correction (AC) is essential in retrieving ocean and coastal water ecosystem products from ocean colour remote sensing data. Generally, the atmospheric correction algorithms that are based on Near Infrared (NIR) bands are effective in clear water, whereas the algorithm based on ultraviolet (UV) bands deliver accurate results in turbid water. A practical approach is to employ a UV-NIR combined algorithm that uses both UV and NIR bands for clear and turbid water, respectively. However, currently, the UV-NIR algorithm generates a data discontinuity pattern when a single threshold value is used, as the coastal water exhibits a continuous variability between clear and turbid states. To address this issue, we propose a ‘clear and turbid water transition evenly combination’ algorithm that utilizes UV and NIR bands of the Second Geostationary Ocean Colour Imager (GOCI-2). We evaluated the new algorithm’s performance by comparing GOCI-2 images with in-situ data, and the results showed good agreement between the retrieved Remote Sensing Reflectance ( R rs ) values at visible bands and the in-situ observations. The new AC algorithm can achieve better precision, with Mean Absolute Percentage Difference (MAPD) reduced by 6.6 95.3% and 5.2 18.9% compared with UV AC and the standard AC algorithm employed by the Korea Ocean Satellite Center (KOSC).