Occurrence of generic Escherichia coli, E. coli O157 and Salmonella spp. in water and sediment from leafy green produce farms and streams on the Central California coast

• Published in: International journal of food microbiology 2013-07, Vol.165 (1), p.65-76
• Main Authors: Benjamin, Lisa, Atwill, Edward R., Jay-Russell, Michele, Cooley, Michael, Carychao, Diana, Gorski, Lisa, Mandrell, Robert E.
• Format: Article
• Language: English
• Subjects: Agriculture; California; Climate; E. coli O157; Enterobacteriaceae - isolation & purification; Enterobacteriaceae - physiology; Escherichia coli; Escherichia coli - isolation & purification; Escherichia coli - physiology; Escherichia coli O157 - isolation & purification; Escherichia coli O157 - physiology; Generic E. coli; Geologic Sediments - microbiology; Produce; Risk Factors; Rivers - microbiology; Salmonella - isolation & purification; Salmonella - physiology; Salmonella spp; Water; Water Microbiology
• ISSN: 0168-1605; 1879-3460
• DOI: 10.1016/j.ijfoodmicro.2013.04.003

Irrigation with water of poor microbiological quality can elevate levels of bacteria on produce. This study aimed to identify climate and management variables associated with generic Escherichia coli in irrigation water on leafy green produce farms and to measure the prevalence of E. coli O157 and Salmonella spp. in irrigation and non-irrigation water sources on these farms. Water and sediment samples collected from various points along irrigation systems, as well as from streams and ponds on farms on the Central California coast between May 27th, 2008 and October 26th, 2010 were cultured for generic E. coli (MPN/100mL or cfu 100g) (n=436), E. coli O157 (n=437), and (n=163) Salmonella. Variables were based on grower's management practices, landscape features in proximity to samples (e.g., distance to roads and ranches/livestock), and climate data accessed from an online database. Negative binomial regression models were constructed to test associations between generic E. coli (MPN/100mL) in water from farms and variables. Arithmetic mean concentration of E. coli for water, not including those from Moore swabs, and sediment samples, was 7.1×102MPN/100mL and 1.0×104cfu/100g, respectively. Matched by collection day, E. coli concentration in sediment (cfu/100g) was typically 10- to 1000-fold higher than the overlying water (MPN/100mL) for these irrigation systems. Generic E. coli concentration (MPN/100mL) increased by 60.1% for each 1m/s increase in wind speed and decreased by 3% for each 10m increase in the distance between the sample location and rangeland. Moore swabs detected a higher proportion of E. coli O157 (13.8%) positive water samples compared to grab samples (1.8%); 1.7% of sediment samples had detectable levels of this pathogen. Interestingly, season was not significantly associated with E. coli O157 presence in water or sediments from produce farms or water sources with public access. Salmonella was detected in 6% (6/96) water and 4.3% (3/67) sediment samples. Generic E. coli concentration was not significantly associated with the presence of either E. coli O157 or Salmonella in water or sediment samples, suggesting that, for this 2.5-year period and geographical location, generic E. coli would likely be an unreliable indicator bacteria for predicting the presence of these food- and waterborne pathogens in a key produce production environment. •Generic E. coli in irrigation water associated with 24 hour precipitation.•Generic E. coli in irrigatio