Predicting culturable Escherichia coli and enterococci concentrations from reverse transcription quantitative PCR and hydro-meteorological data for bathing water management

Rapid decision-making for bathing-water management requires analytical methods faster than conventional culture-based assays, which typically need 18–72 h to deliver results. Reverse transcription quantitative PCR (RT-qPCR) enables same-day detection of Escherichia coli and enterococci but also quantifies viable but non-culturable cells, limiting direct comparison with culture-based regulatory thresholds. This study evaluates whether integrating RT-qPCR measurements with hydro-meteorological predictors through a modelling framework can generate culture-equivalent estimates suitable for operational use. Long-term monitoring data (2015–2024) from 50 coastal sites in France combined culture-based counts, RT-qPCR cycle threshold values, and environmental variables. Models achieved strong predictive performance on independent test datasets, with unbounded RMSE values of 0.34–0.41 log10 units. Within concentration ranges relevant for bathing-water management, RMSE decreased to 0.25 log10 units for E. coli and 0.19 log10 units for enterococci in seawater. Model-based predictions substantially improved agreement with culture-based regulatory classifications compared with raw RT-qPCR outputs, particularly by reducing false-positive classifications at low concentrations. Coupling RT-qPCR measurements with environmental predictors provides a robust and transferable framework for translating molecular signals into culture-equivalent concentrations, enabling faster and more reliable bathing-water quality assessments.