Aim: To assess the persistence and diversity of faecal bacterial populations (faecal coliforms and enterococci) that have recently been included in microbial source tracking (MST) predictive models. Methods and Results: The analysed bacterial populations included members of the enterococci group (ENT) [Enterococcus faecium (FM), Enterococcus faecalis (FS) and Enterococcus hirae (HIR)] and the faecal coliform group (FC) [diverse Escherichia coli phenotypes (ECP) and cellobiose-negative faecal coliforms (CNFC)]. The inactivation of these distinct groups was monitored over time on-site in river by biochemical fingerprinting, and diversity indices were calculated. Among the different analysed species belonging to the ENT group, HIR persisted longer and was able to replicate in the environment at a higher rate. On the other hand, ECP and NCFC showed a similar persistence throughout the different seasons. The diversity index (Di) for FC increased substantially in the summer after 96 h to a maximum value of 0.96. On the other hand, the Di for ENT diminished over the same period to a value of 0.86, suggesting a different persistence for the different species integrating this group. Conclusions: The persistence of ECP, CNFC, FM and FS in the aquatic environment is high, particularly for the members of the FC and in the summer season. On the contrary, HIR is able to replicate in the environment at a high rate even in winter, and therefore, its inclusion in MST predictive models is discouraged. Significance and Impact of the Study: ECP, CNFC, FMFS and HIR have been proposed as additional variables in MST predictive models. However, the different persistence of HIR compared with the other variables should be taken into account for the development of such models.