Sentinel-3 OLCI Level-2 Ocean Colour operational processing has undergone a major update that is available from EUMETSAT from winter 2021 and constitutes OLCI Collection-3 products (OL_L2M.003.01, Collection 3 Report). The full OLCI-A and OLCI-B mission Level-2 reprocessed time-series is also available to users. Collection-3 ensures a consistent OLCI-A and -B Ocean Colour data record throughout the missions and into operations. Thanks to advances in System Vicarious Calibration, Ocean Colour products from both instruments are well harmonised, which allows a stable global merged coverage in just two days. Additional algorithm updates improve the quality of Ocean Colour products and pixel flagging. Collection-3 is a major advancement fulfilling the promise of Sentinel-3 OLCI Constellation.

Nevertheless, algorithm improvement activities are ongoing, as well as new product developments. These developments follow recommendations from the Copernicus Marine Environment Monitoring Service (CMEMS), the Sentinel-3 OLCI/SYN Quality Working Group (QWG) and Sentinel-3 Validation Team, Ocean Colour (S3VT-OC). The goal is to further improve the quality of OLCI Ocean Colour products for user applications, from monitoring of local ecosystems to climate change, and for CMEMS operational needs. Redevelopment of the standard atmospheric correction is the major algorithm evolution being currently investigated. Other algorithm improvements are also studied, including advancements in the correction for bidirectional distribution of water surface reflectance in clear and complex waters. New products have been developed and provided for user assessment and validations, such as Inherent Optical Properties of water constituents and Fluorescence.

The algorithm and product developments are planned for a future OLCI Level-2 Ocean Colour Collection-4 release. The Collection-4 planning is dependent on successful verifications and validations, where the user community feedback will be solicited, particularly from the S3VT-OC, QWG and CMEMS.

This presentation will summarize OLCI Collection-3 algorithm updates and it will present validation results and product status. The extensive use of Fiducial Reference Measurements for OLCI System Vicarious Calibration and for algorithm and product validations will be explained. The presentation will also outline the algorithm and product developments towards OLCI Collection-4.

Evolution of the CMEMS Ocean Colour global and regional products for Essential Ocean Variables during 2015-2021

Vittorio Ernesto Brando (1), Rosalia, Santoleri (1), Simone Colella(1), Gianluca Volpe(1), Javier Alonso Concha (1), Annalisa Di Cicco (1), Michela Sammartino (1), Marco Bracaglia(1), Mario Benincasa(1), Emanuele Böhm(1), Claudia Cesarini(1), Vega Forneris(1), Flavio La Padula(1)
Philippe Garnesson(2), Antoine Mangin(2), Odile Hembise Fanton d'Andon(2), Marine Bretagnon (2)
Silvia Pardo(3), Thomas Jackson(3), Ben Calton (3), Jane Netting, (3), Ben Howey (3)
Hajo Krasemann(4), Martin Hieronymi(4)
Davide D’ Alimonte (5), Tamito Kajiyama (5)
Jenni Attila (6), Seppo Kaitala (6), Sampsa Koponen (6)
Dimitry Van der Zande (7), João Felipe Cardoso Dos Santos (7), Quinten Vanhellemont (7)
Kerstin Stelzer (8), Martin Böttcher (8), Carole Lebreton (8)
Sindy Sterxck (9)

1: CNR ISMAR, 2: ACRI, 3: PML, 4: HEREON, 5: AEQUORA, 6 SYKE, 7 :RBINS, 8 : Brockmann Consult, 9 : VITO

The Ocean Colour Thematic Assembly Center (OCTAC) of the Copernicus Marine Environment Monitoring Service (CMEMS) provides high-quality core Ocean Colour products for the global ocean and the European seas based on multiple Ocean Colour missions.
The OCTAC serves users across the scientific and operational oceanography communities, commercial providers focused on the use of marine resources, and public agencies focused on environmental monitoring, with interests in data across oceanic, shelf and coastal waters. Depending on their applications, these users require different spatial resolutions (i.e., ~1km in ocean, 300m over the shelf, down to 10’s of metres in coastal waters).
To meet these needs, the global and regional higher-level combined OCTAC products provide added-value information not readily available from space agencies. From 2015 to 2021, the OCTAC has continued to improve the accuracy at the basin level of existing Essential Ocean Variables (EOVs), i.e., Chlorophyll-a concentration (CHL), Inherent Optical Properties (IOPs), with particular attention to the optically complex waters occurring in shelf the and coastal zones. Blended CHL datasets are produced for all basins applying the appropriate algorithms across the open ocean and coastal waters depending on the occurring water types, as well as the bio-optical characteristics of each regional Atlantic and Arctic Oceans, Baltic, Mediterranean and Black Seas). New EOVs related to Phytoplankton Functional Groups, community structure and Primary Production were introduced from 2019 onwards.
From 2015 to 2021, the upstream OC data shifted from science missions (i.e., SeaWiFS, MERIS and MODIS) towards operational missions (two OLCI and two VIIRS sensors). In 2015, all NRT regional products were based on single sensors (MODIS and VIIRS), these datasets were replaced by multi-sensor datasets products leading to a significant increase of the spatial coverage of daily observations. In May 2021, OLCI datasets at 300m resolution combining Sentinel-3 A and B, as well as the Sentinel-2/MSI datasets at 100m, were added to the catalogue.
This talk will provide an overview of the evolution of the CMEMS-OCTAC catalogue from 2015 to 2021, the accuracy of the global and regional products, and the plans for new products in coming years.

High-quality satellite-based ocean colour products can provide valuable support and insights in management and monitoring of coastal ecosystems. Today’s availability of Earth Observation (EO) data is unprecedented including traditional medium resolution ocean colour systems (e.g. SeaWiFS, MODIS-AQUA, MERIS, Sentinel-3/OLCI), high resolution land sensors (e.g. Sentinel-2/MSI, Landsat-8/OLI, Pleiades) and geostationary satellites (e.g. SEVIRI). Each of these sensors offers specific advantages in terms of spatial, temporal or radiometric characteristics.

With the High-Resolution Coastal Service (HROC), CMEMS provides high resolution ocean colour products based on Sentinel-2/MSI data for European coastal waters. It offers 12 different products which are categorized in three groups: 1) near real time (NRT) daily products, 2) aggregated monthly products and 3) gap-filled daily products. The products are generated for the coastal waters (20km stripe for the coastline) of all European Seas and are provided in a 100m spatial resolution. The primary variable from which it is virtually possible to derive all the geophysical and transparency products is the spectral Remote Sensing Reflectance (RRS). This, together with the Particulate Backscatter Coefficient (BBP), constitute the category of the optics products. The spectral BBP product is generated from the RRS products using a quasi-analytical algorithm. The transparency products include turbidity (TUR) and Suspended Particulate Matter (SPM) concentration. They are retrieved through the application of automated switching algorithms to the RRS spectra adapted to varying water conditions. With this approach we address the high variability of different water types with small scale changes. The geophysical product consists of the Chlorophyll-a concentration (CHL) retrieved via a multi-algorithm approach with optimized quality flagging. Validation of the variables has been performed by match-up analysis with in situ data as well as by inter-comparison of the high resolution products with the Low Resolution CMEMS Ocean Colour products. High-Resolution products are available from the 1st of January 2020 to current day and we will present our experiences after one year of operational processing together with examples of use cases with like eutrophication monitoring in the near shore WFD zones in the Southern North Sea.

Mapping water colour in different water body types using remote sensing techniques comes with several challenges, from choosing the most suitable atmospheric correction to applying the best retrieval algorithm that produces repeatable and consistent results. Often, the lack of suitable in-situ measurements prohibits researchers from validating their satellite-based results. On the other hand, the need for better and improved remote sensing techniques that complement in-situ campaigns is pivotal to holistically addressing UN Sustainable Development Goals and complying with European policy by reporting under the Water Framework Directive (WFD) and Marine Strategy Framework Directive (MSFD). This urgently calls for harmonised, transferable, robust and reliable approaches on mapping optical properties of different water bodies: from oceans to coasts, to inland waters.

The European Copernicus programme includes satellite sensors designed to map water colour and serves data and information to end-users in industry, policy, monitoring agencies and science. Three Copernicus services, namely Copernicus Marine, Copernicus Climate Change and Copernicus Land, provide satellite-based information on different water constituents such as phytoplankton and coloured dissolved organic matter in oceanic, shelf and lake waters. Though the transitional waters are partly covered by CMEMS coastal service, the approaches are distinct in the different services.

This presentation describes how the H2020 Copernicus Evolution: Research for harmonised Transitional water Observation (CERTO) project undertakes research and development necessary to address this gap by producing harmonised water colour datasets suitable for integration in the Copernicus services. With efforts focusing on improving methods for water optical classification, masking the land-sea interface and correcting atmospheric effects, CERTO builds on the most comprehensive existing in-situ database complemented by CERTO data collected with state-of-the-art biogeooptical instrumentation, while using Sentinel 2 MSI and 3 OLCI sensors. So far, local campaigns have been undertaken to characterise the bio-geochemical properties of the Curonian, Razelm-Sinoe and Venice Lagoon systems, and the Plymouth Sound. Dedicated campaigns to characterise the bio-geo-optical and above-water radiometry have been undertaken in the Curonian Lagoon, LI, Elbe Estuary, DE, Venice Lagoon and Adriatic Sea, IT, and the Tagus Estuary, PT, including the deployment of two new autonomous radiometer systems for extended data acquisition (Tagus Estuary and Razelm-Sinoe Lagoon System). Data collected are added to the LIMNADES and Ocean Colour CCI databases and used to attribute water types to optical-biogeochemical and environmental traits. The creation and implementation of optical water classification types is demonstrated using Sentinel 2 MSI and Sentinel 3 OLCI data, at the regional and European scale, across a range of optical environments.

In addition, a comparison of several atmospheric correction techniques is underway, showing the higher performance of the Polymer and C2RCC algorithms and their robustness to complex atmospheric effects (aerosols, thin clouds, sun glint); on the other hand, algorithms assuming spatial homogeneity of the atmospheric properties (iCOR, Acolite) can be more insensitive to complex variations of the water properties. The impact of adjacency and bathymetry effects (bottom visibility) on the observation of transitional waters are investigated and mitigation techniques are being developed.

All these CERTO outputs are brought together into a Software-as-a-Service (SaaS), the CERTO prototype, which is deployed in the CLMS Calimnos water quality processing system and used to generate an archive of data products from the Sentinel 2 MSI and 3 OLCI sensors focussed on the six case studies in transitional waters, namely: i) Danube Delta and Razelm-Sinoe Lagoon System; ii) Venice Lagoon and North Adriatic Sea; iii) Tagus Estuary; iv) Plymouth Sound; v) Elbe Estuary and German Bight; and vi) Curonian Lagoon. The prototype will also deliver the same suite of products for the CMEMS regions (Atlantic, Mediterranean, Black Sea, Baltic Sea and Arctic) and European CLMS regions. The CERTO prototype will make use of the various Data Access and Information Services (DIAS), which have evolved over the past several years. A comparison of the satellite data availability, range of services and compute resources, level of available support, and cost has been undertaken. The details of this comparison are published on a dedicated website (https://engage.certo-project.org/) and will inform CERTO on which DIAS is the most suitable host for the prototype, as well as others looking to make use of a DIAS but unsure which is the most suitable for their own needs.

By 2022, several satellite-borne sensors have continuously measured 24 years of global ocean colour data. Because sensors have a finite lifespan, combining the data without introducing artefacts in the process, is of utmost importance. Merged datasets protect the temporal continuity of the data and optimize the spatial coverage. However, all sensors have their own specific characteristics in terms of overpass and re-visit times, swaths, spatial and spectral resolutions. The multi-sensor merged ocean colour dataset of the European Space Agency’s Ocean Colour Climate Change Initiative (OC-CCI v5) is an error-characterized and intermission-corrected dataset, produced to provide continuous long-term observations of Essential Climate Variables like remote-sensing reflectance and chlorophyll-a concentration in the upper ocean. Despite the careful consideration of inter-mission biases, artefacts can still easily be transferred to multi-mission time series and create spurious signals in the derived trends. Temporally consistent Earth observations are necessary for accurate long-term statistical processing of ocean colour data. However, the consistency of time series produced with the CCI dataset variables suffers from some limitations. The most striking inconsistency is during the MERIS period, where the average timeline shows systematically increased values for chlorophyll-a (and other variables). This discrepancy is not caused by a true change in chlorophyll-a concentration, but rather a change in the spatial coverage of pixels used to compute statistics, caused by the introduction/discontinuation of a satellite sensor. In this presentation, the CCI dataset is used to demonstrate where the artefact is most prominent, and that differences in the spatiotemporal distribution of the merged satellite sensors cause this artefact. The objective of this study is to introduce a new method that achieves spatiotemporal consistency for improved statistical trend analysis of ocean colour variables. The running gap screening method provides a temporally equal daily timeline on every geographic pixel. Effectively, the method evades spatiotemporal differences between the satellite missions by masking records that are observed by one sensor, but not by another. It is particularly effective in coastal waters and scarcely observed areas (e.g. high latitude and cloudy areas), which are most affected by the inconsistencies between the satellite sensors. Moreover, this method is almost entirely independent of the reference parameter (e.g. chlorophyll-a), which is only required for optimizing. The running gap screening method is potentially of value for other merged satellite datasets, which may also suffer from spatiotemporal intermission biases.

The use of 'optical water type' classification schemes is becoming increasingly prevalent within limnological and oceanographic remote sensing research (Moore et al. 2001, 2014, Jackson et al. 2017, Spyrakos et al. 2018). Uses for optical water classes include, but are not limited to, algorithm blending (Moore et al. 2001), product uncertainty estimation (Jackson et al. 2017), data quality flagging (Wei et al. 2016) water quality monitoring (Uudeberg et al. 2020) and environmental phenology (Trochta et al. 2015) studies.

However, a harmonised approach to the creation and use of the classes has not yet emerged from the research community. Despite recent efforts to move to a unified fuzzy logic scheme (Jai et al. 2021), a diversity of distance metrics, data transformations and cluster optimisation schemes are applied at local scales (Bi et al. 2019, Botha et al. 2020, da Silva et al. 2020, Uudeberg et al. 2020). Though all these approaches provide interesting and useful results, the fragmented nature of the research makes the comparison of water types difficult, impeding collaboration and optimisation of methods.

As with most machine learning techniques, unsupervised clustering is susceptible to the problems of insufficient or biased training data, the ‘central tendency’ (Malik, 2020), and overtraining. Here we examine components of the clustering pipeline such as data transformation, data dimensionality, distance metric choice and output cluster variables to present a generalised approach that uses robust, data-driven principles to generate a cluster set with minimal 'human' decision making. The approach presented builds upon insight from two European Space Agency projects (Ocean Colour Climate Change Initiative, OC-CCI; Lakes-CCI). The approach is demonstrated using Sentinel 2 MSI and Sentinel 3 OLCI data, at the regional and pan-regional scale, across a range of optical environments. Matchups against in-situ measurements are used to validate the utility of the clusters generated.

This work was undertaken as part of the EC Horizon 2020 CERTO (Copernicus Evolution: Research for harmonized Transitional water Observation) project.

Bi, S, Li, Y., Xu, J., Liu, G., Song, K., Mu, M., Lyu, H., Miao, S., and Xu, J. "Optical classification of inland waters based on an improved Fuzzy C-Means method," Opt. Express 27, 34838-34856 (2019)
Botha, Elizabeth & Anstee, Janet & Sagar, Stephen & Lehmann, Eric & Galvao, Thais. (2020). Classification of Australian Waterbodies across a Wide Range of Optical Water Types. Remote Sensing. 12. 3018. 10.3390/rs12183018.
Jackson, T., Sathyendranath, S. and Mélin, F. “An improved optical classification scheme for the Ocean Colour Essential Climate Variable and its applications,” Remote Sens. Environ. 203, 152–161 (2017).
Jia T, Zhang Y, Dong R. A Universal Fuzzy Logic Optical Water Type Scheme for the Global Oceans. Remote Sensing. 2021; 13(19):4018. https://doi.org/10.3390/rs13194018
Malik, Momin M. "A hierarchy of limitations in machine learning." arXiv preprint arXiv:2002.05193 (2020).
Moore, Timothy & Campbell, Janet & Feng, Hui. (2001). A fuzzy logic classification scheme for selecting and blending satellite ocean color algorithms. IEEE Transactions on Geoscience and Remote Sensing, 39, 1764-1776. Geoscience and Remote Sensing, IEEE Transactions on. 39. 1764 - 1776. 10.1109/36.942555.
Moore, Timothy & Dowell, Mark & Bradt, Shane & Ruiz-Verdu, Antonio. (2014). An optical water type framework for selecting and blending retrievals from bio-optical algorithms in lakes and coastal waters. Remote sensing of environment. 143. 97-111. 10.1016/j.rse.2013.11.021.
Spyrakos, E., O’Donnell, R. , Hunter, P. D. , Miller, C. , Scott, M., Simis, S. G., Neil, C., Barbosa, C. C., Binding, C. E. and Bradt, S. “Optical types of inland and coastal waters,” Limnol. Oceanogr. 63(2), 846–870 (2018).
da Silva, E.F.F.; Novo, E.M.L.d.M.; Lobo, F.; Barbosa, C.; Noernberg, M.A.; Rotta, L.H.d.S.; Cairo, C.T.; Maciel, D.A.; Flores Júnior,R. Optical water types found in Brazilian waters. Limnology (2020).
Trochta, John & Mouw, Colleen & Moore, Timothy. (2015). Remote sensing of physical cycles in Lake Superior using a spatio-temporal analysis of optical water typologies. Remote Sensing of Environment. 171. 10.1016/j.rse.2015.10.008.
Uudeberg, K.; Aavaste, A.; Kõks, K.-L.; Ansper, A.; Uusõue, M.; Kangro, K.; Ansko, I.; Ligi, M.; Toming, K.; Reinart, A. Optical Water Type Guided Approach to Estimate Optical Water Quality Parameters. Remote Sens. 2020, 12, 931. https://doi.org/10.3390/rs12060931
Wei, Jianwei & Lee, Zhongping & Shang, Shaoling. (2016). A system to measure the data quality of spectral remote sensing reflectance of aquatic environments. Journal of Geophysical Research: Oceans. 10.1002/2016JC012126.