Day 4

Detailed paper information

Back to list

Paper title Validation of Sentinel-3 Chlorophyll-a products in Mediterranean coastal lagoons
  1. Carolina Tenjo Universitat de València
  2. Antonio Ruiz-Verdú Image Processing Laboratory (IPL) - University of Valencia (Spain) Speaker
  3. Patricia Urrego University of Valencia
  4. Jesús Delegido University of Valencia
  5. Xavier Sòria-Perpinyà University of Valencia
  6. Eduardo Vicente University of Valencia
  7. Juan Miguel Soria Cavanilles Institute of Biodiversity and Evolutionary Biology (ICBiBE) - University of Valencia (Spain)
  8. Manuel Erena GIS and Remote Sensing - IMIDA
  9. Jose Moreno University of Valencia
Form of presentation Poster
  • A7. Hydrology and Water Cycle
    • A7.06 EO for monitoring water quality and ecological status in inland waters
Abstract text Chlorophyll-a concentration, as a proxy of phytoplankton biomass, is a key variable to monitor the highly dynamic transitional waters in coastal areas, often subjected to anthropogenic pressures that severely modify its ecological status. Sentinel-3 data is especially suited for this purpose in the large coastal lagoons characteristic of the Mediterranean floodplains. Its spatial, spectral and radiometric resolutions, together with a short revisit time, allow to monitor the spatiotemporal changes of the phytoplankton populations in these ecosystems in response to diffuse pollution, as well as the abrupt changes occurring after extreme meteorological events, such as floods, which alter its hydrodynamics and water composition.
We present the validation of the Sentinel-3 Chlorophyll-a concentration product ([Chl-a]), produced by the C2RCC processors available in the SNAP software, in two Mediterranean coastal lagoons in Eastern Spain: Albufera de Valencia, a shallow hypereutrophic brackish lagoon with an ongoing restoration plan to limit its nutrient content; and Mar Menor, a hypersaline mesotrophic lagoon which is undergoing an accelerated eutrophication, severely affecting its fisheries and its important recreational uses.
For this validation exercise, a set of 1413 and 185 in situ [Chl-a] samples were available in the Mar Menor and Albufera lakes, respectively. In the Mar Menor the in situ data was measured between August 2016 and October 2019, while for the Albufera the time span was between January 2016 and February 2018.
A total of 1142 Sentinel-3/OLCI images, from April 2016 to February 2020 were processed with C2RCC and filtered using the processor’s quality flags. The match-up points were statistically filtered and the correlation with the C2RCC [Chl-a] product was analyzed for the whole dataset, per lake and in the time series.
In the Mar Menor lagoon, with in situ [Chl-a] ranging from ~0.1 to ~25 mg·m-3, the C2RCC accuracy (R2= 0.69; RMSE= 4 mg·m-3) was acceptable for the spatial and temporal monitoring of this variable, closely following the time evolution of [Chl-a] in the studied period and identifying the bloom episodes and abrupt changes after flooding events. On the contrary, in the Albufera lagoon, C2RCC systematically underestimate [Chl-a] to an order of magnitude less than the in situ data, with large retrieval errors (R2=0.40 ; RMSE= 44 mg·m-3), precluding the spatio-temporal monitoring of [Chl-a] and suggesting that the C2RCC could not be appropriate for these eutrophic or hypereutrophic ecosystems.