Aims To better understand the persistence dynamics of enteropathogenic bacteria in freshwater wetland habitats, we constructed lab‐scale mesocosms planted with two different wetland plant species using a subsurface flow wetland design. Mesocosms were treated with either a high‐quality or a poor‐quality water source to examine the effects of water quality exposure and plant species on Escherichia coli, Salmonella spp. and Enterococcus spp. in the rhizoplane, rhizosphere and water of wetland habitats. Methods and Results Quantities of study micro‐organisms were detected using real‐time PCR in wetland mesocosms. A combination of molecular and culture‐based methods was also used to enumerate these organisms from surface water and plant material at high, medium and poor water quality sites in the field. We found that all three enteropathogenic micro‐organisms were influenced by microhabitat type and plant species. Organisms differed with respect to their predominant microhabitat and the extent of persistence associated with wetland plant species in the mesocosm study. Of the monitored pathogens, only E. coli was influenced by both water quality treatment and plant species. Salmonella spp. quantities in the rhizoplane consistently increased in all treatments over the course of the mesocosm experiment. Conclusions Plant species selection appears to be an overlooked aspect of constructed wetland design with respect to the removal of enteropathogenic micro‐organisms. Escherichia coli and Enterococcus concentrations in wetland outflow were significantly different between the two plant species tested, with Enterococcus concentrations being significantly higher in mesocosms planted with Phalaris arundinaceae and E. coli concentrations being higher in mesocosms planted with Veronica anagallis‐aquatica. Furthermore, there is evidence that the rhizoplane is a significant reservoir for Salmonella spp. within wetland habitats. Significance and Impact of the Study This is the first time that Salmonella spp. has been shown to proliferate under natural conditions within the rhizoplane. This will contribute to our understanding of wetland removal mechanisms for enteropathogenic bacteria. This study identifies the rhizoplane as a potentially important reservoir for human pathogenic micro‐organisms and warrants additional study to establish whether this finding is applicable in non‐wetland habitats.