Sensor-based Methods for Improved Understanding of Biological Water Treatment

Licentiate thesis, 2024

Access to clean and safe drinking water is essential for a functioning society. To produce drinking water in Sweden, raw water is taken from a natural resource and treated at a drinking water treatment plant, before being distributed to consumers via a pipe network. One important role of drinking water treatment is to remove natural organic matter (NOM). Natural organic matter is a collective term for a variety of organic compounds present in all natural water sources. The amounts as well as composition and characteristics of NOM vary between water resources as well as between seasons. In general NOM does not itself impose a health risk but it can alter the taste, colour, and odour of water, and may cause complications during drinking water treatment and distribution. One cause of concern for drinking water providers is microbial regrowth in treated drinking water while it is being transported in the distribution network. When drinking water is biologically stable, the microbial quality of water leaving the treatment plant will not change as it
travels to the consumer, when unstable there can be substantial bacterial regrowth in the distribution network, potentially including pathogens. Bacterial regrowth is prompted by the presence of biologically labile fractions of NOM that bacteria are able to utilise as nutrients.
There is a need for inexpensive tools that help to monitor biologically labile NOM throughout drinking water treatment and assess the effects of specific treatment steps on biostability. This thesis focuses on two new monitoring methods for measuring the availability of biologically labile NOM fractions, based upon (1) rates of oxygen consumption by bacteria, and (2) the composition of fluorescent NOM. The approach taken was to evaluate if these two methods could track labile organic carbon through dilution series, and secondly use each method to compare treatment processes at full-scale drinking water treatment plants. Treatment steps involving biological processes were studied particularly closely, because these are usually effective at removing labile organic carbon. Both new methods demonstrated potential for assessing the efficiencies of different types of treatment steps for removing labile NOM, for comparing performances of parallel filter beds, as well as for detecting changes in treatment efficiencies over time.