Water reuse is becoming a vital solution for water scarcity. Methods to make water reuse more efficient while protecting public health would increase its implementation. A groundbreaking study funded by The Water Research Foundation — project 5129: Demonstration of Innovation to Improve Pathogen Removal, Validation, and Monitoring in Carbon-Based Systems, led by Carollo Engineers and Virginia’s Hampton Roads Sanitation District (HRSD) — evaluated carbon-based advanced treatment (CBAT) processes for ozone disinfection and biologically activated carbon (BAC) filtration to validate and optimize their performance for potable reuse.
Traditional ozone validation relies on concentration×time (CT) values to quantify disinfection. CT requires measuring ozone residua, which often requires adding more ozone than needed for disinfection, resulting in excessive energy use and formation of disinfection byproducts (DBP). CT doesn’t account for rapid, variable ozone decay in real wastewater conditions, making it less reliable for predicting pathogen inactivation in CBAT systems. The study demonstrated that a more robust, energy-efficient surrogate, the ozone and nitrite-corrected total organic carbon ratio, or (O₃-NO₂):TOC, can be used to improve dose control.
Testing at six utilities revealed that this ratio reliably predicted virus log reduction values up to 7-log for MS2 bacteriophage, independent of variables like temperature, pH, nitrite, or peroxide addition. While UV absorbance showed potential as a secondary surrogate, its effectiveness varied by location, making site-specific testing essential.
The study integrates science, technology, and operations in advanced water treatment, providing an alternative to CT validation
The project proposed a two-tier ozone validation protocol: Tier 1 uses conservative 10th–25th percentile values from the dataset for initial crediting, whereas Tier 2 achieves higher credits using detailed, site-specific performance data when water quality is beyond the Tier 1 range. Using (O₃-NO₂):TOC lowers energy demand and reduces bromate and other DBPs typically associated with higher ozone dosages.
BAC filtration is valued for its ability to degrade organic and inorganic contaminants, but often not been credited for pathogen removal. Robust pilot-testing at HRSD’s full-scale indirect potable reuse facility and Polk County, Florida’s direct potable reuse pilot facility showed that BAC could achieve over 2-log removal of indicators like Clostridium perfringens and pepper mild mottle virus when preceded by coagulation, flocculation, and sedimentation. Without pretreatment, reductions were significantly lower, stressing the importance of upstream processes to maximise pathogen removal.
To allow more rapid (O₃-NO₂):TOC based ozone dosing, the study applied machine learning (ML) to predict Total Organic Carbon (TOC) levels in real-time. Using data from HRSD’s SWIFT Research Center, the best-performing ML model — boosted trees — achieved an impressive root mean square error (RMSE) of 0.35 mg/L. Impressively, this would be more accurate than assuming the same TOC from 2.5 hours prior, which could be the status quo with automated but not truly real-time instruments. These soft sensors allow utilities to refine ozone application more responsively, improving pathogen control and reducing chemical use.
The study provides a holistic approach to integrate science, technology, and operations in advanced water treatment, providing an alternative to CT validation that improves accuracy and efficiency, demonstrates pretreatment’s potential to boost BAC’s pathogen removal, and introduces ML tools that modernise process monitoring and control.
This equips facilities with scalable, site-adaptable tools to achieve regulatory compliance and safeguard public health. The study also complements related projects that address pathogen risk, water quality impacts, and validation protocols for other reuse technologies.
