Ecological Systems and Devices Annotation << Back Biological Early Warning System for the Aquatic Environment Control Grekov A.N., Vyshkvarkova E.V., Ivakin Ya.A., Seleznev I.A., Grekov N.A., Kuzmin K.A. The growth of industry and agriculture, urbanization lead to an increase in the load on water bodies - more and more polluted substances enter them, worsening their ecological condition. Existing monitoring methods (mainly physicochemical) do not allow timely and complete assessment of the environmental threat. Only biological control methods, that is, the use of living organisms as biosensors, are capable of quickly and clearly responding to pollution and giving an integral assessment of the state of the aquatic environment in real time, regardless of the composition and origin of pollutants. For continuous monitoring of water bodies, biological early warning systems (BEWS) are used. The article presents the results of the development and research of a system for biological monitoring of the aquatic environment using machine learning algorithms and biosensors based on mussels “Ecobiocontrol”. The operation of the complex is based on recording and analyzing the activity of bivalve mollusks (valve opening). The work describes the scheme of the complex, its equipment, presents the results of laboratory and fi eld tests, describes the software, proposes a scheme for creating a monitoring system for water bodies and presents the results of using machine learning to detect anomalies measured in the activity data of mollusks to generate an alarm signal. The use of native species of mollusks makes it possible to scale the complex and the entire system to other water bodies. The most important advantages of the developed system are: a high degree of automation of the monitoring process, the ability to create large networks of automated control of aquatic environments, which allows real-time identifi cation of pollution sites and assessment of their severity for humans and living beings. Keywords: monitoring, ecology, biological early warning systems, biosensors, mollusks. DOI: 10.25791/esip.1.2024.1425 Pp. 38-48.