Earth Observation for monitoring our aquatic environment

The SEAmBOTH project is coming to an end, and it is time to look back and see what we have accomplished in the field of Earth Observation (EO).  In a previous blog, the basics of aquatic Earth Observation were already explained. Here, we can concentrate on what were our main results and take-home messages.

EO can strongly support modelling efforts in Bothnian Bay

The SEAmBOTH project wanted to map the spatial and temporal characteristics of water quality in the Bothnian Bay area. For this, EO is a superior tool. In particular, the project found these spatially and temporally comprehensive observations are useful in modelling of the habitats. The project concluded, that monthly composites of turbidity observed in different times of the ice-free period are most beneficial support for the models, instead of trying to utilize all individual daily – and sometimes partially cloudy – satellite observations.  This compresses the necessary information and reduces artifacts present in daily images generated e.g. by clouds. An example of this is shown in Figure 1. Locations where elevated values of turbidity are consistently found are visible as yellow and red areas. These include e.g. river estuaries and areas where dredging is taking place, leading to a strong resuspension of sediments.

For more details on ecological modelling and its results, there is a specific blog available also.

Map showing that in Liminganlahti there is river input, around Hailuoto there is resuspension and around Pyhäjoki can be seen the nuclear power plant dredging area
Figure 1. Turbidity composite of summer 2017 (1.7-7.9) as visualized in the TARKKA map application. Note the increased values of turbidity (measured in FNU) in river estuaries (indicating run-off from rivers) and dredging areas (indicating resuspended sediments).

How can we know that the observations provided by the satellites are reliable?

The answer is validation. It is a process where EO estimates are compared to values measured in situ, that is in the water at specific station sites. An example of this is shown in Figure 2. These two types of turbidity observations correspond very well. The station in question is visited by boat about 10 times per year. EO can provide more estimates during the same time period. Furthermore, there are elevated values in the EO data during August 2018. These observations are from a time period affected by resuspension, presumably due to wind-wave stirring which causes bottom sediments to become resuspended in the water column. This effect is strongest in shallow waters, but the suspended material can drift over long distances along currents. Stationsamples were not taken during this resuspension event, and thus the traditional sampling method by boat completely missed the increased values in suspended particulate matter. A turbidity map of the event is shown in Figure 3 (left panel). For comparison, also the situation without the effects of resuspension is shown (right panel).

Figure showing that dot-like field samplings can't predict the whole truth of algorithms from earth measures
Figure 2. Turbidity time series at Hailuoto intensive station measured from stationsamples (MS) and with Sentine2-satellite observations (EO) using the SYKE algorithm (C2RCC processor and calibration based on samples collected at station sites.
Figure 3. Turbidity map in the Bay of Bothnia estimated from S2 data taken on Aug 19, 2018 (left) and July 10, 2018 (right) (the links lead to the images in the TARKKA map application). The Hailuoto station (Figure 2) is indicated with a triangle and a number 30372. The left panel shows the turbidity in the event of strong wind, the right panel shows the normal situation, indicating the run-off from land.

Importance of high-quality station observations

The reliability of the in situ data is essential for the validation and development of EO methods. It is therefore very important that these measurements are done with care and with correct protocols in all parts of the Bothnian Bay. Thehigh-qualityfield sample observations collected during SEAmBOTH have been a valuable resource for the algorithm testing and development. However, development of water quality algorithms over dark water types requires long time series before a sufficient level of confidence in the results can be reached. For future work, we recommend that water quality sampling is kept at high level in this region and that the sampling follows the optical protocols defined in SEAmBOTH.

What did we learn?

The turbidity and CDOM can be estimated well in the Bothnian Bay using Earth observations. In coastal regions, high resolution instruments are especially valuable, whereas the open sea areas can be well covered with moderate resolution instruments that provide more frequent observations. For Sentinel-3 OLCI data, there were many stations with good correspondence between field samples and EO values for chlorophyll-a – a quantity that is used to define the amount of algae in the water. One example of this is shown in Figure 4. The station in question was one of the stations sampled within the SEAmBOTH project. There were, however, also areas where the EO methods must still be improved, in particular the areas with high amounts of humic substances. We look forward to seeing how the ongoing development of EO algorithms will solve these remaining problems.  

Figure 4. Chlorophyll a time series at station RÅNEÅ 2R sampled by SU for the SEAmBOTH project in 2018 (MS) and observed by EO instrument OLCI using the currently best model, C2RCC.

SEAmBOTH put a lot of effort in solving problems related to sampling the water quality at field and ensuring that this important work will be done with as few errors as possible. This work involved writing dedicated optical protocols and training of monitoring groups. This will serve as basis for the future high-quality field work in the Bothnian Bay and was a huge joint effort of both participating countries.

About the EO team

This work has been based on collaboration between three partners:

  1. The Finnish Environment Institute (SYKE) was responsible mainly for the development and processing related to Sentinel-2 data and the publication of results in TARKKA. For any EO related questions you can contact us through
  2. Brockmann Geomatics Sweden AB (BG) was responsible for the development and processing of Sentinel-3 data.
  3. Stockholm University (SU) was responsible for providing in situ measurement protocols and performing measurements in the study area. Stockholm University also measured absorption and scattering properties of the Baltic Sea which were used by SYKE to develop an improved radiative transfer model for the Bothnian Sea.

This collaboration allowed each partner to concentrate on their strengths while also learning from others by sharing data and experience.

Further reading

  1. You can find more information about EO in here:
  2. SU have recently published a popular science article in Swedish on how to use Sentinel-3 data to derive inorganic matter in the coastal zone from scatter. The data from SEAmBOTH was included in the algorithm development for this article: Kratzer, S. Havet från rymden – satelliter berättar, Havsutsikt, Om Havsmiljö och Svensk havsforskning, 2/2019, s. 9-11.
  3. SEAmBOTH remote sensing report (pdf)
  4. A recent PhD thesis from SU gives more insight on how to use remote sensing for research and management: ‘Baltic Sea from Space – The use of ocean color data to improve our understanding of ecological drivers across the Baltic Sea basin – algorithm development, validation and ecological applications’. The thesis can be downloaded here:

Written by:

  1. Sampsa Koponen, Jenni Attila, Kari Kallio, Hanna Alasalmi, Mikko Kervinen, Vesa Keto, Eeva Bruun, Sakari Väkevä, Finnish Environment Institute (SYKE)
  2. Petra Philipson, Brockmann Geomatics Sweden AB (BG)
  3. Susanne Kratzer, Department of Ecology, Environment and Plant Sciences, Stockholm University

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s