Inactivation of indicator organisms on different surfaces after urban floods

More Info


The high frequency and intensity of urban floods caused by climate change, urbanisation and infrastructure failures increase public health risks when the flood water contaminated from combined sewer overflows (CSOs) or other sources of faecal contamination remains on urban surfaces. This study contributes to a better understanding of the effects of urban and recreational surfaces on the occurrence of waterborne pathogens. The inactivation of selected indicator organisms was studied under controlled exposure to artificial sunlight for 6 h followed by 18 h in dark conditions. Concrete, asphalt, pavement blocks and glass as control were inoculated with artificial floodwater containing, as indicator organisms, Escherichia coli bacteria, which are common faecal indicator bacteria (FIB) for water quality assessment, Bacillus subtilis spores chosen as surrogates for Cryptosporidium parvum oocysts and Giardia cysts, and bacteriophages MS2 as indicators for viral contamination. On practically all the surfaces in this study, E. coli had the highest inactivation under light conditions followed by MS2 and B. subtilis, except asphalt where MS2 was inactivated faster. The highest inactivation under light conditions was seen with E. coli on a concrete surface (pH 9.6) with an inactivation rate of 1.85 h−1. However, the pH of the surfaces (varying between 7.0 and 9.6) did not have any influence on inactivation rates under dark conditions. MS2 bacteriophage had the highest inactivation under light conditions on asphalt with a rate of 1.29 h−1. No die-off of B. subtilis spores was observed on any of the surfaces during the experiment, neither in light nor in dark conditions. This study underpins the need to use different indicator organisms to test their inactivation after flooding. It also suggests that given the sunlight conditions, concentration of indicator organisms and type of surface, the fate of waterborne pathogens after a flood could be estimated.