Evaluating contaminated irrigation water in BC's Lower Mainland

Introduction

Foodborne pathogens can cause human illness and sometimes even result in loss of life. VTEC (Verotoxigenic Escherichia coli), Listeria (Listeria monocytogenes), and Salmonella are the major foodborne pathogens contributing to illnesses in people. 

Produce can become contaminated when surface water containing pathogens is used for irrigation. Foodborne pathogens are mostly associated with livestock, specifically their feces; livestock upstream can deposit contaminated fecal matter into surface water such as ditches which can then be washed downstream and contaminate other crops. Once pathogens have attached to crops through contact with contaminated water, the bacteria can survive for long periods of time and may not come off through washing. Water contamination has been shown to be linked to environmental factors such as landscape, weather, and season. 

Testing water for these disease-causing pathogens is too time consuming and expensive to do routinely, so water quality assessment and regulation is based on the use of microbial indicators. These indicators are bacteria that pose no risk to human health, but whose presence usually point to the presence of more dangerous disease-causing pathogens in a water sample. The most commonly used indicator organisms are fecal coliform bacteria, and generic E.coli. 

 

The objectives of this study are to: 

i) determine how much VTEC, Listeria, and Salmonella can be found in surface water used for irrigation in BC’s Lower Mainland;

ii) assess the effectiveness of fecal coliforms and generic E.coli as water quality indicators; and

iii) investigate how environmental factors increase or decrease the presence of VTEC, Listeria, and Salmonella in surface water used for irrigation

Research Process

Two watersheds in Lower Mainland BC were sampled over 18 months. Sampling sites were located in the Sumas watershed (Abbotsford, BC) and Serpentine watershed (Surrey, BC). Surface water samples from ditches, creeks, or streams were collected one to two times per month from each watershed. The samples were tested for VTEC, Listeria, and Salmonella. Weather measurements were taken from Environment Canada data, including air temperature and recent precipitation. The temperature and pH of the water was also measured, (pH assesses the acidity of water). All sampling sites were directly connected to surface water used for irrigation of the adjacent fields. Water flow direction was also noted. Upstream water sources were defined as any waterway directly connected to the sample, up to three kilometres from each site. Livestock (cows, poultry, goats, etc.) were considered to be upstream if the property where they were living was directly bordering the waterway. 

Results

How often were pathogens found in surface water? 

A total of 223 samples were collected from the Sumas and Serpentine watersheds over 18 months, with the overall pathogen occurrence in both watersheds being 15.7 per cent. Overall, Listeria was most commonly detected, followed by VTEC and Salmonella. Contamination was more likely to be found in the Serpentine watershed compared to Sumas.

 

Do microbial indicators successfully reveal the presence of pathogens?

 

Generic E.coli and fecal coliforms were assessed as microbial indicators to help identify presence of the three disease-causing pathogens. Generic E.coli was a better predictor for VTEC, while fecal coliforms was a better predictor for Listeria. Neither indicator was strongly associated with Salmonella. Overall, long-term trends in microbial indicator concentrations were more reliable than single measurements to predict the “risk” of finding foodborne pathogens.These findings suggest the importance of keeping a microbial water quality profile, which is a record of many water samples taken over time to provide a more reliable indication of water health. 

The impact of environmental factors

 

Temperature, precipitation, pH, season, and location were evaluated. Sampling revealed that contamination was greater when water was cooler, with a pH closer to neutral. The level of precipitation up to five days before sample collection was also observed to be correlated with pathogen presence. To examine seasonal difference, occurrence of pathogens was examined through winter (December to February), spring (March to May), summer (June to August), and fall (September to November). Contamination was more commonly found during fall and winter. Finally, upstream livestock (up to 2 km) adjacent to irrigation water sources may also increase contamination.

Since there is more rain observed in fall and winter months, contamination in some areas may be seasonally linked to cooler, rainier seasons. Rainfall is believed to increase the movement of pathogens into surface waters and carry them longer distances downstream, leading to an increase in surface water contamination three to five days after rainfall. The Serpentine watershed showed a stronger link between season and contamination than the Sumas watershed. Change in water flow may also affect seasonal changes; the region dams the irrigation ditches during the growing season to ensure sufficient water availability for irrigation in summer months, which may limit contamination by reducing the flow of water. 

 

Implications

VTEC, Listeria monocytogenes, and Salmonella are present in surface water used for irrigation in BC’s Lower Mainland. Listeria was the most commonly identified pathogen, followed by VTEC and Salmonella. The use of microbial indicators to identify pathogens from one water sample cannot reliably predict the true level of contamination; however, many samples taken over time can identify overall trends of water health. Other potential predictors of contamination include proximity of surface waters to livestock, cooler weather, and increased precipitation three to five days before irrigation. These findings speak to the importance of keeping a microbial water quality profile over time, being aware of livestock in upstream agriculture areas, and limiting irrigation three to five days after a rainfall.