Remote sensing is becoming an increasingly reliable technique for the assessment of water environments, however, Latvia lacks a strong track record of using satellite data for continued water quality monitoring. In 2015, a year before the launch of Sentinel-3 satellite, the Institute for Environmental Solutions (IES) joined by the Estonian Marine Institute (EMI) started the SentiBalt project where a state-of-the-art remote sensing laboratory was used to simulate the data of future satellites. The aim of the project was to validate and improve remote sensing algorithms and methods for the assessment of water quality parameters in the coastal areas of the Baltic Sea.
!["Whirlpool]("/assets/upload/userfiles/images/62_2.png")
Baltic Sea has been identified as a case-2 water reservoir meaning that it has an optically complex water environment with high turbidity, mixture of fresh and salt water, shallowness and low biodiversity that creates a unique challenge for remote sensing. In such waterbodies, optical satellites can detect not only phytoplankton, but also other mineral and organic substances which disturbs the acquisition of accurate water surface parameters. Furthermore, the sea’s constantly changing conditions limit the upscaling of in situ collected data, thus failing to provide comprehensive and representative information at the necessary spatial scale. However, the launch of the second Sentinel-3 satellite will further increase revisit times of waterbodies, thus providing even more data for improved monitoring and decision-making.
Latvian Marine Environmental Monitoring Programme 2014-2020 is based on HELCOM’s Monitoring and Assessment Strategy. HELCOM has identified six marine monitoring sub-programmes where remote sensing is identified as a complementary monitoring technique to in situ observations. The detailed, regular and free-of-charge satellite data can be a cost-effective alternative to compulsory water quality assessment, however, to capitalise on the capabilities of the latest technologies competence and capacity building is necessary.
The simulation of Sentinel-3 data was planned through six simultaneous in situ and airborne data acquisition campaigns. Nevertheless, unfavourable weather conditions reduced the campaign count to four, with two campaigns in Gulf of Riga on 11 and 12 August 2015 and two in Estonian coastal waters on 31 May and 14 September 2016. The in situ and airborne data were further used to simulate the Sentinel-3 OLCI data.
!["Airborne]("/assets/upload/userfiles/images/62_4.png")
IES’s and EMI’s researchers have tested and developed different algorithms for the assessment of water quality (e.g. chlorophyll-a concentrations) in the Baltic Sea. Nonetheless, repeated field campaigns and tests of algorithms are required for increased uptake of remote sensing for marine monitoring. The main challenge is acquiring the best possible data from the rapidly changing and often cloudy Baltic Sea environment. The continued growth of the European Space Agency’s (ESA’s) Sentinel satellite fleet provides an increasingly frequent and wide array of data from different instruments covering all Earth’s ecosystems. The researchers recommend combining Sentinel-3 data with Sentinel-2 overpasses, spectral in situ campaigns and regular field campaigns to produce a variety of data of the same location from different instruments that could enable the validation and training of advanced machine learning algorithms for optical remote assessment of complex waterbodies.
The SentiBalt project (Simulating Performance of ESA Future Satellites for Water Quality Monitoring of the Baltic Sea) No LV1-27 is funded by the Government of Latvia and ESA Contract under the PECS (Plan for European Cooperation States). The view expressed herein can in no way to be taken to reflect the official opinion of ESA.
More about the SentiBalt project
