Real-Time Biological Early Warning System based on Freshwater Mussels’ Valvometry Data

Abstract. The aim of this study is to investigate the impact of natural river floods on biotic communities. To this purpose, we used freshwater mussels (FMs), recognized as one of the most reliable bioindicators in aquatic environments. A well-established valvometry technique was applied to measure the FMs valve gaping behaviour, considering both gaping amplitude and frequency. The mussels have been employed in two distinct configurations, either free to move or stuck on vertical bars.  We performed experiments in a laboratory flume and in the Paglia River (Italy). The FMs valve gaping movement was first recorded, then the continuous wavelet transform (CWT) analysis was applied to the signals to get the time-dependent frequency of the signals. Laboratory experiments allowed to assess to what extent stuck mussels react differently than free mussels to abrupt increases in flow conditions. Subsequently, we examined the response of thirteen stuck mussels installed in real riverine conditions during a moderate flood occurred on March 31, 2022, with a rapid increase of the water level. The experimental results demonstrate that stuck mussels produce signals that are more consistent and easier to interpret compared to free mussels, primarily due to the reduced number of features resulting from movement constraints. The stuck mussels in the field showed a sharp and timely change of valve gaping frequency as the flood ramped up, thus confirming the findings in the laboratory. The results highlight the effectiveness of using FMs as bioindicators for assessing the impact of floods on the aquatic ecosystem, and the utility of CWT as a suitable signal processing tool for analyzing valvometric time series. These findings provide a pathway towards the integration of FMs valvometry and CWT for the development of operational real-time Biological Early Warning Systems (BEWS) aimed at the monitoring and safeguarding of aquatic ecosystems.