Limnetica 38

Albufera of València, a hypertrophic lake, has been studied extensively since the 1980s, but the efforts to revert the system to a clear water state have not yielded the expected results because pressure on this system is growing (increase in nutrient-rich water inputs, decrease in precipitation and increase in evaporation). The current state of the lake requires constant monitoring, and one of the main biological parameters used in ecology and water management to monitor and control the ecological status of aquatic ecosystems is chlorophyll a concentration [Chl-a]. In this sense, remote sensing is an optimal tool for continuous monitoring of the quality state of the water body through [Chl-a] and to obtain a better understanding of its spatial dynamics. This work aims to demonstrate the validity of an algorithm for [Chl-a] retrieval from Sentinel-2 (A and B), the new Earth observation satellites of the European Space Agency, with the sensor MSI, multispectral (13 bands) from 404 nm to 2200 nm, a spatial resolution of 10 m and a temporal frequency of 5 days - values unthinkable until now as regards to open access images. The study was carried out with images from 2016 and 2017, but only 40 images out of the 81 taken by the satellite could be used – such rate was mainly due to unfavourable weather conditions. Once images were downloaded, the SNAP 5 software was used for the processing. Using the Sen2cor tool, they were corrected atmospherically and, with the algorithm developed by Soria et al. 2017, lake [Chl-a] was estimated. Estimated data were validated against field samples: a total of 18 sampling campaigns were carried out and 92 samples were taken to measure the [Chl-a]. In addition, to better interpret results, data on conductivity and Secchi disk depths measurements were taken in the field and hydrological, precipitation and wind data were also collected. Results of the validation were deemed very good since an R = 0.8 was obtained when applying a linear correlation between field data and estimates, which shows the robustness of the algorithm used. From the interpretation of the thematic maps, it was possible to infer that the temporal evolution in [Chl-a] variations follows an annual bimodal pattern, where the decrease in [Chl-a] is determined either by a significant increase in water renewal of the lake or by the depletion of the available nutrients in the water due to a previous excessive growth of phytoplankton.