VU Scientists Find a New Method of Using Satellite Data to Determine “Bloom” in Water
We have definitely heard the phrase “water bloom” before and have seen this phenomenon in lakes that have turned a bright green colour. The question then arises as to what causes this phenomenon, and whether swimming in the water is dangerous?
Dalia Grendaitė, PhD student in the Department of Hydrology and Climatology, Faculty of Chemistry and Geosciences at Vilnius University (VU), together with her colleague from the Institute of Informatics in the Faculty of Mathematics and Informatics, Dr Linas Petkevičius, is implementing the VU Science Promotion Foundation’s Young Scientists Project titled “Automatic Detection of Water Bloom from Satellite Data”. The scientists are striving to develop artificial intelligence algorithms that can detect water bloom from Copernicus programme satellite data, and to use this data not only for research but also to address the public needs.
Water bloom requires both the right weather conditions and nutrients
The “blooming” of water is caused by favourable meteorological conditions: heat and light. This is because algae are photosynthetic organisms. The algae need adequate nutrients, so the strongest bloom is observed in water bodies that are rich in phosphorus and nitrogen.
“The amounts of phosphorus and nitrogen in a water body can depend on the erosion of the water body basin, when particles of these substances flow in with the affluent. The concentration of substances can also increase due to human activities: phosphorus and nitrogen are leached from fertilised fields, or can enter water bodies along with incompletely treated domestic wastewater,” said the researcher.
We have observed water bloom in the Baltic Sea, where it is caused by both cyanobacteria and macroalgae. The phenomenon is also observed in rivers with low water velocities. In rivers, the “bloom” in the water is more often caused by macroalgae – multicellular organisms that form aggregates – than by microalgae.
“These algae are clearly visible and are reminiscent of a green carpet. Our colleagues from the Nature Research Centre collect this algal biomass and look for areas where it can be used, because these organisms are rich in useful substances,” added Grendaitė.
Swimming in “blooming” water is not advisable
After seeing the green water of the lake, many ask: is it dangerous to swim in the water? According to Grendaitė, if the concentration of algae is low, we will not even notice the bloom – and then there is no risk when swimming. But intensive water blooms, which are usually caused by cyanobacteria, can be toxic and cause allergic reactions.
“Toxins in the algae can cause rashes, while neurotoxins cause tingling in the fingers. Another group of toxins affects the digestive system and the symptoms are like those of poisoning. Another important point is that these toxic substances can accumulate in the liver and cause chronic diseases. Thus, if a person ingests ‘blooming’ water while swimming, these risks increase,” explained the researcher.
Toxins can persist in the water for a couple of months and accumulate in the fish and other aquatic animals. In addition, strong water blooms significantly reduce the transparency of the water, leading to a lack of light and endangering aquatic flora and fauna.
“The algal biomass grows under favourable conditions, but when there is a lack of nutrients or changes in the weather, this mass begins to decay. Oxygen is consumed during the decomposition process, becoming scarce, which can be devastating to fish. Furthermore, toxins are released from the algal cells and enter the water during this process. Despite the fact that by that stage, the colour of the is no longer bright green, swimming can be even more dangerous because the toxins are present in the water itself, not just in the biomass. Consequently, you can also experience an unpleasant odour due to the decomposition of organic matter,” she explained.
Development of a model based on artificial intelligence
In the “Automatic Detection of Water Bloom from Satellite Data” project, Grendaitė and her colleague Dr Petkevičius are seeking to develop algorithms for the determination of water bloom from satellite data, as well as to determine water parameters such as the chlorophyll concentration.
“The aim is often to estimate the concentration of the main algal pigment, i.e. chlorophyll. Since this pigment is present in all algae and is a good indicator of a biomass, the chlorophyll content indicates the ecological status of a water body: the higher the chlorophyll concentration, the worse the water status. For this project, we collected satellite data from the Copernicus Programme Sentinel-2 satellites, as well as data from environmental agencies. We received this data from Lithuanian, Latvian and Estonian agencies. The project is aimed at creating a model that will be able to determine whether or not the water bloom occurs at any specific time,” said Grendaitė.
According to the researcher, the idea for the project came from the need to have more information about the “blooming” of water bodies, because although the observations made by the Environmental Protection Agency are accurate, they are infrequent and measurements are taken only four (in some cases six) times during the warm season.
“From the data that is currently available, it is difficult to determine when the bloom will begin and what its duration will be. It would also be useful for the public to receive more information about water bloom, in order to avoid allergic reactions and other negative effects on public health,” revealed the researcher.
Results of the project will benefit the public
The model being developed by the researchers will be based on artificial neural networks, allowing larger amounts of data to be processed and meaningful information to be extracted. The model is first planned to be applied to the lakes included in the state monitoring programme, and will be introduced to the scientific community.
“This data will be of great value to the Environmental Protection Agency, which monitors water bodies, as well as to the Center for Health Education and Disease Prevention, which monitors the quality of swimming water. We hope that our model will not only facilitate the information collection process, but also contribute to improving the quality of water bodies,” stated the researcher.
According to Grendaitė, water usually blooms most intensively in July, August and sometimes in September. Therefore, it would be valuable to make the Copernicus Programme Sentinel-2 satellite data available to the public around that time.
The satellites in the Copernicus Programme photograph the territory of Lithuania every 2-3 days, so we can definitely record changes in the “blooming” of water. Our goal for the future is to create an application or a website, where we could provide data as soon as the satellites pass over Lithuania. We believe that this tool would help residents plan their trips to water bodies in the summer,” concluded the researcher.