Day 4

Detailed paper information

Back to list

Paper title Chlorophyll-a concentrations and water quality monitoring of French inland waters using satellite data from MSI Sentinel 2 and OLI Landsat 8. Recent improvements from optical water type classification in the scope of the Water Framework Directive
  1. Guillaume Morin INRAE - PACA Speaker
  2. Nathalie Reynaud RECOVER - Pôle ECLA, INRAE, Aix-en-Provence, France
  3. Tristan Harmel GET - CNRS - Magellium
  4. Arthur Coqué INRAE
  5. Pierre-Alain Danis Pôle R&D « ECLA », OFB
  6. Pierre Gernez Nantes Université, Institut des Substances et Organismes de la Mer, ISOMER, UR 2160 Nantes, France
  7. Thierry Tormos
Form of presentation Poster
  • A7. Hydrology and Water Cycle
    • A7.06 EO for monitoring water quality and ecological status in inland waters
Abstract text The Water Framework Directive (2000/60/EC) (WFD) states that all European Union (EU) members must implement the monitoring and estimation of the ecological status of their territorial inland water bodies. It demands to classify the status in 5 classes from “very bad” to “high” and aims to achieve at least “good” ecological status of inland waters by 2027 by all required means. However, monitoring of the several ecological parameters required by the WFD are based on in situ data sampling and further laboratory analysis, that are both time- and money-consuming. Hence, this cannot be achieved in timely and frequent manner at a country-scale. Therefore, satellite observation of water quality appears to be a promising and efficient tool to achieve the WFD requirements (Papathanasopoulou et al., 2019). Nonetheless, there is still a need to improve the accuracy of satellite-derived products used to classify the water bodies status, especially for each water quality parameters that can be remotely sensed: chlorophyll-a concentrations ([chlo-a]) values, Secchi-disk depth, turbidity, suspended matter concentrations. (Giardino et al., 2019).

In order to evaluate the relevancy of the current satellite products for ecological status monitoring, this study was based on numerous French lakes sites where field data were previously collected. A dataset was composed with in situ measurements from the french WFD regulatory monitoring network and the long-term Observatory on Lakes (OLA), as well as from other public institutes (research or territorial management). This dataset concerns ~325 sites from 2014 to 2017. Over the period covered by Lansat8 and Sentinel 2, this dataset includes ~1000 to 1900 chlorophyll-a concentrations ([chlo-a]) values, Secchi-disk depth, turbidity, suspended matter concentrations. Corresponding satellite data products were generated from Sentinel-2 and Landsat-8 imageries through our processing chain: first, level 2A water reflectances were produced with the atmospheric correction (AC) algorithm “Glint Removal for Sentinel-2 like data” (GRS) (Harmel et al., 2018), second, water reflectances images were masked based on watermask and cloudmask computed with sentinel-hub’s “s2clouless” for S2/MSI, and the original cloudmasks for Landsat 8. To evaluate the quality and representativeness of the satellite products, matchup comparisons were performed.

Our results demonstrate that, in certain environments and circumstances, sunglint signal can represent a major part of the water leaving reflectances. It can lead to a 10-fold bias on water quality estimations, and validate the importance of including a sunglint correction step, even though no consensus exists in the choice of a particular atmospheric correction algorithm (Pahlevan et al., 2021). Focusing on [chlo-a] retrieval, we implemented several widely used algorithms from literature, and adapted them to Landsat 8 and Sentinel 2 data. We calculated [chlo-a] following several modalities: (i) with the original paper’s calibration, (ii) after applying calibration to the region of interest, and (iii) with calibrations defined for each OWT by (Neil et al., 2019). We also implemented a spectral angle mapper method (SAM) to identify Optical Water Types (OWT hereafter) as defined by (Spyrakos et al., 2018), like it was recently implemented for MERIS by (Liu et al., 2021).
For altitude lakes in the Alps mountain chain, classified as clear oligotrophic waters, ocean-colour algorithm OC3 performs best, detecting low [chlo-a] in the range of 1 to 10 µg/l (MAE < 2.6 µg/L, RMSE < 4.8 µg/L, MAPE < 65 %, SSPB(signed bias) < 15%), which is comparable to recent neural-network algorithms. In meso- to eutrophic lakes, several algorithms performed satisfactorily such as red-fluorescence-, NDCI- or 2-3 bands- based algorithms, but with variable accuracies depending on sites. In a few lakes of Brittany and Aquitaine regions, optically classified as eutrophic to hypertrophic turbid lakes, performances are good enough to distinguish periods of blooms, and the shifts between ecological status from “moderate”, “poor” up to “bad”. Ranges of water quality parameters associated with the OWT classes as defined in (Spyrakos et al., 2018) are also in good agreement with the in situ data observed on our sites. Moreover, a matching score derived from SAM and OWT classes was implemented to measure similarity with OWT shapes. Matching scores will soon guide the choice of the best-suited algorithm. This matching score was also shown to be complementary to cloud and water masks and provide further ability for masking out pixels that are likely impacted by contributions from upwelling light of the bottom, adjacency effect from the shores, or badly-masked clouds.

This work, whilst ongoing, showed that spectral identification performs well with high resolution satellite data and is useful to optimize algorithm selection. Reasoning by analogy on optical types, we expect to successfully use OWT classification to retrieve [chlo-a] and other parameters on lakes where in situ data is not available. The perspective of this study is to proceed with a census of the ecological states of the French lakes. This can be foreseen as a crucial step to help respect the WFD engagements since field data is scarce or even absent for many sites included in the WFD.


Giardino, C., Brando, V.E., Gege, P., Pinnel, N., Hochberg, E., Knaeps, E., Reusen, I., Doerffer, R., Bresciani, M., Braga, F., Foerster, S., Champollion, N., Dekker, A., 2019. Imaging Spectrometry of Inland and Coastal Waters: State of the Art, Achievements and Perspectives. Surv Geophys 40, 401–429.
Harmel, T., Chami, M., Tormos, T., Reynaud, N., Danis, P.-A., 2018. Sunglint correction of the Multi-Spectral Instrument (MSI)-SENTINEL-2 imagery over inland and sea waters from SWIR bands. Remote Sensing of Environment 204, 308–321.
Liu, X., Steele, C., Simis, S., Warren, M., Tyler, A., Spyrakos, E., Selmes, N., Hunter, P., 2021. Retrieval of Chlorophyll-a concentration and associated product uncertainty in optically diverse lakes and reservoirs. Remote Sensing of Environment 267, 112710.
Neil, C., Spyrakos, E., Hunter, P.D., Tyler, A.N., 2019. A global approach for chlorophyll-a retrieval across optically complex inland waters based on optical water types. Remote Sensing of Environment 229, 159–178.
Pahlevan, N., Mangin, A., Balasubramanian, S.V., Smith, B., Alikas, K., Arai, K., Barbosa, C., Bélanger, S., Binding, C., Bresciani, M., Giardino, C., Gurlin, D., Fan, Y., Harmel, T., Hunter, P., Ishikaza, J., Kratzer, S., Lehmann, M.K., Ligi, M., Ma, R., Martin-Lauzer, F.-R., Olmanson, L., Oppelt, N., Pan, Y., Peters, S., Reynaud, N., Sander de Carvalho, L.A., Simis, S., Spyrakos, E., Steinmetz, F., Stelzer, K., Sterckx, S., Tormos, T., Tyler, A., Vanhellemont, Q., Warren, M., 2021. ACIX-Aqua: A global assessment of atmospheric correction methods for Landsat-8 and Sentinel-2 over lakes, rivers, and coastal waters. Remote Sensing of Environment 258, 112366.
Papathanasopoulou, E., Simis, S., Alikas, K., Ansper, A., Anttila, S., Attila, J., Barillé, A.-L., Barillé, L., Brando, V., Bresciani, M., Bučas, M., Gernez, P., Giardino, C., Harin, N., Hommersom, A., Kangro, K., Kauppila, P., Koponen, S., Laanen, M., Neil, C., Papadakis, D., Peters, S., Poikane, S., Poser, K., Pires, M.D., Riddick, C., Spyrakos, E., Tyler, A., Vaičiūtė, D., Warren, M., Zoffoli, M.L., 2019. Satellite-assisted monitoring of water quality to support the implementation of the Water Framework Directive, White paper. Eomores.
Spyrakos, E., O’Donnell, R., Hunter, P.D., Miller, C., Scott, M., Simis, S.G.H., Neil, C., Barbosa, C.C.F., Binding, C.E., Bradt, S., Bresciani, M., Dall’Olmo, G., Giardino, C., Gitelson, A.A., Kutser, T., Li, L., Matsushita, B., Martinez-Vicente, V., Matthews, M.W., Ogashawara, I., Ruiz-Verdú, A., Schalles, J.F., Tebbs, E., Zhang, Y., Tyler, A.N., 2018. Optical types of inland and coastal waters: Optical types of inland and coastal waters. Limnol. Oceanogr. 63, 846–870.