Wastewater spiking in tap water CSTR

The data in this repository are from the published article: Sadler et al. 2020. Computational Surveillance of Microbial Water Quality With Online Flow Cytometry. Front. Water 2:586969. doi: 10.3389/frwa.2020.586969. A laboratory experiment was designed in which tap water was spiked with wastewater. The set-up consisted of a stirred tank through which tap water was continuously flowed and an automated flow cytometer connected to the inside of the vessel made measurements every 30 minutes. The purpose of the experiment was to assess whether the MCCD model (Microbial Community Change Detection) which is based on cytometric fingerprints and distance-based outlier calculation, better detected the spikes than the cell concentration. The experiment lasted 7 days, and 7 wastewater spikes as well as a blank spike (tap water spike) were performed.
Conclusion:
The MCCD was more performant in detecting the spiking times than the cell concentration. The concentration in the tap water underwent strong fluctuations which obscured the peaks observed during the spiking times, while a clear signal was obtained by the outlier score (output of the MCCD model) which increased during the spikes and remained flat otherwise.
Notes:
The .fcs files are organized as follows: - _ref_events.fcs: measurements corresponding to the defined reference period - _tapwater_events.fcs: measurements outside the reference period during which no external perturbations were applied - _Spike{num}_{1/2/3}_events.fcs: measurements corresponding to the spiking times. num is the number of the spike (1 to 7) and each spike lasted 90 minutes which resulted in 3 measurements per spike.