A Case Study: The deployment of a novel in situ fluorimeter for monitoring biological contamination within the urban surface waters of Kolkata, India.

With increasing pressures on water resources due to population, industrialization, agriculture, urbanization and climatic changes, improved temporal and spatial understanding of water quality is required. The development of new monitoring parameters, along with new monitoring technologies, are needed to provide real-time insight into the biogeochemical processes that underpin aquatic ecosystem health. Aquatic fluorescent organic matter (AFOM) has recently been explored for its potential to measure underpinning microbial activity within aquatic systems, which are essential in maintaining ecosystem health and function, with specific focus on the utilisation of tryptophan-like fluorescence (TLF or Peak T). In situ real-time portable fluorimeters have been extensively used for the identification and measurement of anthropogenic pollutants, such as polycyclic aromatic hydrocarbons (PAH) and optical brighteners. More recently, this portable fluorescence technology has been adapted for the monitoring and sensing of biological contamination, using microbially derived fluorescence signals (TLF). 

The principal aim of this research was to deploy, for the first time in the field, the VLux TPro sensor (Chelsea Technologies Ltd., UK) and to assess the ability of this novel fluorescence-based sensor to detect the presence of biological contamination and elevation of microbial activity. This sensor has been developed to correct in situ real-time sensing data for optical interferences (caused by high turbidity and absorbance), as well as to provide quantitative fluorescence data by reporting in standardised quinine sulphate units (QSU). The urban surface waters within the city of Kolkata provide an interesting challenge for water quality sensors, allowing exploration of sensor performance in a range of water bodies ranging from turbid river waters to open sewer canals. 

The sensor data collected demonstrates the ability of the VLux to identify waters with high bacterial loads using Peak T fluorescence. Moderate and weak positive correlations are seen for Peak T and E. coli or total coliform counts, R2 = 0.55 and 0.38 respectively. However, a strong significant correlation is identified between Peak T and the total bacterial cell counts (R2 = 0.75). This demonstrates that Peak T should not be used as a species-specific enumerator in complex surface water matrices. It does, however, demonstrate the ability of the VLux to successfully measure optically corrected and quantitative Peak T fluorescence in QSU. Therefore, data regarding the activity and fast-acting dynamics of freshwater bacterial communities, in response to pollution events, can now be reliably sensed and collected. This was demonstrated by the elevated Peak T fluorescence intensity observed when biologically contaminated water entered the main river channel, enabling identification of contamination hotspots. Sensing data has been further validated by laboratory analysis of spot samples confirming the significant correlations between Peak T and bacteria and nutrient concentrations. Further field-based research is required to determine the feasibility of long-term catchment scale sensor deployment as part of a sensing network, for the monitoring of biological activity and pollution events in freshwaters.  

 

Acknowledgements: This research was supported by the NERC-DST Indo-UK Water Quality Programme NE/R003106/1 and DST/TM/INDO-UK/2K17/30. We would also like to acknowledge Chelsea Technologies Ltd. for ongoing VLux TPro sensor support.