The objective of this study was to determine how alternative on-site wastewater treatment systems (i.e. subsurface flow constructed wetlands, intermittent sand filters and intermittent peat filters) affect the viability and culturability of Salmonella choleraesuis (serotype typhimurium, ATCC 23567). Influent was a high strength septic tank effluent (BOD5 240-344 mgL(-1), TN approximately 100 mgL(-1), TP approximately 15 mgL(-1)) at the Natural Resources Research Institute's (NRRI) alternative treatment system test facility in northern Minnesota. Treatment systems were inoculated with cultures of S. choleraesuis for 5-7 consecutive days in summer and winter during 1998-99. After the seeding, outflow samples were taken until Salmonella counts were sustained at background levels. In addition to culture-based enumeration, S. choleraesuis abundances were also measured using fluorescent in situ hybridization (FISH) alone and in combination with the direct viable count method (DVC) to determine if plate counts underestimated total and viable Salmonella abundances and if the Salmonella cell viability changed after passing through the treatment systems. In most cases, total and viable cell abundances in treatment system effluents were several orders of magnitude higher than cultured cell abundances. Our results indicate that the culture-based method underestimated viable concentrations of the model pathogen, S. choleraesuis. Salmonella cell viability decreased in effluents during the summer but increased during the winter. Using a culture-based enumeration method alone to determine removal efficiencies of bacterial indicators and pathogens for wastewater treatment systems may result in artificially high estimates of effective treatment.