Aims Evaluate the T-90 and compare the decay of different faecal indicator bacteria (FIB) and molecular microbial source tracking (MST) markers of human and animal sources during summer and winter. Methods and Results Conclusions The persistence of Escherichia coli and enterococci and several MST molecular markers targeting host-specific Bifidobacterium and Bacteroidales species (BifHM, BifCW, BifPL, HF183/BFD, Rum2Bac and Pig2Bac) was assessed at the same time using mesocosms. Dialysis bags filled with diluted wastewater from different sources were kept in an outdoor water tank and monitored regularly to assess the inactivation rates. The T-90 values of E. coli by culture methods ranged from 1 center dot 52 to 5 center dot 69 days in summer and 2 center dot 06 to 6 center dot 19 days in winter, whereas with qPCR 2 center dot 29-4 center dot 23 days in summer and 4 center dot 17-8 center dot 09 days in winter. T-90 values for enterocci ranged from 1 center dot 15 to 3 center dot 10 days in summer and from 3 center dot 01 to 5 center dot 46 days in winter. Significant differences were observed between faecal sources for both markers. For the MST makers similar T-90 values were obtained in summer (1 center dot 05-1 center dot 91 days), whereas higher variability was observed in winter (2 center dot 90-6 center dot 12 days). Different decay rates were observed for the FIB from the different sources, especially for E. coli in ruminant samples. A higher variability among T-90 values of the different MST markers in winter was observed, whereas similar T-90 values were detected in summer highlighting the stronger effect of environmental parameters during this season. Significance and Impact of the Study The diverse inactivation rates observed in bacteria from different faecal sources have implications when these rates are used to model faecal pollution in water. The use of FIBT90 of different sources is essential to develop reliable predictive models. Since different inactivation of E. coli regarding the source of pollution has been observed, the source of the pollution has to be considered for modelling approaches.