Exploring Domestic Discharge Patterns in Wastewater through LC-HRMS Screening and Temporal Clustering

Wastewater influent contains valuable epidemiological
information,
but the complexity of the wastewater matrix poses challenges for data
interpretation and linking signals to human exposure. This study aims
to analyze daily discharge patterns in influent wastewater to identify
recurring patterns for trace organic compounds, particularly those
from domestic sources, providing insights into discharge dynamics
originating from population chemical consumption and exposure. Over
three 24-h periods, hourly composite influent samples from a wastewater
treatment plant were analyzed using liquid chromatography coupled
to high-resolution mass spectrometry (LC-HRMS). Target and non-target
screening revealed over 72,000 features, with 402 target compounds
annotated. Temporal k-means clustering of target compounds identified
five distinct daily patterns, with two clusters linked to domestic
use: one correlated with wastewater flow, representing general daily
population activities, and another showing a morning peak, likely
associated with morning urine. Based on these patterns, cluster predictions
were applied to the non-targeted feature list, prioritizing features
with similar temporal trends. This led to 70 additional features associated
with the morning peak pattern, with four compounds exemplarily identified.
The findings highlight the value of combining targeted and non-targeted
analyzes with clustering methods to improve the interpretation of
complex wastewater data and unravel chemical discharge patterns linked
to population exposure.