Skip to main content
ARS Home » Southeast Area » Tifton, Georgia » Southeast Watershed Research » Research » Publications at this Location » Publication #295892
Title: Landscape and seasonal factors influence salmonella and campylobacter prevalence in a rural mixed use watershed

Author
item VEREEN, ETHELL - Emory University
item Lowrance, Robert
item Jenkins, Michael
item ADAMS, PAIGE - University Of Georgia
item RAJEEV, SREEKUMARI - University Of Georgia
item LIPP, ERIN - University Of Georgia

Submitted to: Water Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/1/2013
Publication Date: 12/1/2013
Citation: Vereen, E., Lowrance, R.R., Jenkins, M., Adams, P., Rajeev, S., Lipp, E.K. 2013. Landscape and seasonal factors influence salmonella and campylobacter prevalence in a rural mixed use watershed. Water Research. 47:6075-6085.

Interpretive Summary: A byproduct of large-scale broiler chicken production is large quantities of broiler litter -- comprised of bedding material such as wood chips, fecal material, water, and chicken feed. This litter is also a source of the bacterial pathogens Salmonella and Campylobacter. Because most litter from broiler production facilities is applied to nearby agricultural fields as fertilizer, under conditions of rain and overland runoff, broiler litter has the potential to contaminate surface waters with these two pathogens. Scientists at the University of Georgia and the USDA-Agricultural Research Service sampled streams and rivers to determine if a connection exists between animal agriculture and the presence of Salmonella and Campylobacter in surface waters within the Satilla River watershed in the Coastal Plain of Georgia. Thirteen sampling sites were strategically placed at the mouths of sub-watershed within the Satilla River watershed. Each sub-watershed was characterized by having no agriculture (the control), those with poultry houses and fields that received broiler litter and those with no poultry houses but with fields that received broiler litter. One sampling site received effluent from a sewage treatment plant. Three sampling sites were at downstream locations on larger rivers. Water samples were taken monthly for two years. They were analyzed for the presence of Salmonella and Campylobacter.and for the standard bacterial indicator Escherichia coli and a proposed bacterial indicator Fecal enterococci. A sub-watershed without any agriculture was compared to sub-watersheds with broiler production facilities and/or fields amended with broiler litter, as well as the site with the sewage treatment plant. Both Salmonella and Campylobacter detection frequencies were positively associated with the number of poultry houses in the subwatersheds, but agricultural land use as a proportion of the watershed was not a significant predictor of either pathogen. Fecal indicator bacterial levels were assessed and evaluated for their ability to predict the presence of pathogens. Of those examined, enterococci were most predictive and E. coli were least predictive of the possible presence of the pathogens. Detection of the pathogens throughout the watershed indicated that there was potential for waterborne transmission especially in downstream areas that were more likely to have recreational users.

Technical Abstract: Salmonella and Campylobacter prevalence in stream networks of the Satilla River Basin (SRB) were monitored monthly from August 2007 to August 2009 to study relationships between these pathogens and land use, presence of poultry houses and wastewater treatment plant (WWTP) discharge. Salmonella and Campylobacter were detected at all 10 stream sites and the three sites at the sole wastewater treatment in the study area. In all, 43% (129/299) and 62% (96/156) of samples were positive for Salmonella and Campylobacter, respectively, with detection frequency increasing in downstream sites with more poultry production and influence of WWTP discharge. Both Salmonella and Campylobacter detection frequencies were positively associated with the number of poultry houses in the subwatersheds, but agricultural land use as a proportion of the watershed was not a significant predictor of either pathogen. Fecal indicator bacterial levels were assessed and evaluated for their ability to predict the presence of pathogens. Of those examined, enterococci was most predictive; of the 129 samples positive for Salmonella, 88% (113/129) were detected when enterococci were above EPA single sample threshold (61 CFU 100 ml-1); and of the 96 samples positive for Campylobacter, 90% (86/96) were detected when enterococci levels exceeded this level. Comparatively, E. coli concentrations were above EPA single sample thresholds in 38% (49/129) of the positive Salmonella samples. Detection of the pathogens throughout the watershed indicated that there was potential for waterborne transmission especially in downstream areas that were more likely to have recreational users.