Escherichia coli Contamination of Vegetables Grown in Soils Fertilized with Noncomposted Bovine Manure: Garden-Scale Studies
Steven C. Ingham,1* Jill A. Losinski,1 Matthew P. Andrews,1 Jane E. Breuer,2 Jeffry R. Breuer,2 Timothy M. Wood,3 and Thomas H. Wright4
Department of Food Science,1 Hancock Agricultural Research Station,2 Lancaster Agricultural Research Station,3 West Madison Agricultural Research Station, University of Wisconsin—Madison, Madison, Wisconsin4
*Corresponding author. Mailing address: Department of Food Science, University of Wisconsin—Madison, 1605 Linden Dr., Madison, WI 53706-1565. Phone: (608) 265-4801. Fax: (608) 262-6872. E-mail: email@example.com .
Received March 5, 2004; Accepted July 1, 2004.
In this study we tested the validity of the National Organic Program (NOP) requirement for a ≥120-day interval between application of noncomposted manure and harvesting of vegetables grown in manure-fertilized soil. Noncomposted bovine manure was applied to 9.3-m2 plots at three Wisconsin sites (loamy sand, silt loam, and silty clay loam) prior to spring and summer planting of carrots, radishes, and lettuce. Soil and washed (30 s under running tap water) vegetables were analyzed for indigenous Escherichia coli. Within 90 days, the level of E. coli in manure-fertilized soil generally decreased by about 3 log CFU/g from initial levels of 4.2 to 4.4 log CFU/g. Low levels of E. coli generally persisted in manure-fertilized soil for more than 100 days and were detected in enriched soil from all three sites 132 to 168 days after manure application. For carrots and lettuce, at least one enrichment-negative sample was obtained ≤100 days after manure application for 63 and 88% of the treatments, respectively. The current ≥120-day limit provided an even greater likelihood of not detecting E. coli on carrots (≥1 enrichment-negative result for 100% of the treatments). The rapid maturation of radishes prevented conclusive evaluation of a 100- or 120-day application-to-harvest interval. The absolute absence of E. coli from vegetables harvested from manure-fertilized Wisconsin soils may not be ensured solely by adherence to the NOP ≥120-day limit. Unless pathogens are far better at colonizing vegetables than indigenous E. coli strains are, it appears that the risk of contamination for vegetables grown in Wisconsin soils would be elevated only slightly by reducing the NOP requirement to ≥100 days.
Appl Environ Microbiol. 2004 November; 70(11): 6420–6427.
Recent scrutiny of the role of agricultural practices in contamination of fresh vegetables with pathogenic microbes (47, 54) has led to concern about the safety of using animal manures as fertilizer in vegetable production. In the North Central region of the United States, an estimated 10.4 million dairy and feedlot cattle produce between 23.6 and 35.5 kg of manure (feces and urine) per 454 kg (live weight) per day (16). Bovine manure is a good source of macro- and micronutrients, so using it as fertilizer is an important disposal method (35), particularly for organic farmers. However, bovine manure is a well-known source of food-borne pathogenic bacteria (31, 33, 41, 64, 67), and using it without prior treatment to destroy pathogens increases the likelihood of contaminating vegetables grown in manure-fertilized soils. Composting is an accepted manure pathogen reduction treatment (58), and compost-generated heat is believed to eliminate pathogenic bacteria (27, 34). However, the heat-induced death of bacteria in composted materials is a complex phenomenon (15, 28). Tailing of pathogen inactivation curves has been reported (29), as have apparent regrowth or recontamination and growth of bacteria in cooled compost (7, 38). Composting of bovine manure is not yet widely practiced in Wisconsin (1). An alternative to composting may be to passively age or store and digest the manure before application so that pathogen populations either decrease or disappear (31, 32, 37). A variety of environmental and operational variables, including storage temperature (26) and the diet of the cattle (37), could affect the extent of pathogen decrease, and complete elimination of pathogens is not assured by this technique (20). Further death of manure-borne pathogens occurs once the manure is incorporated into the soil. Weather conditions, desiccation, soil type, predatory protozoan populations, and the degree of manure incorporation are all likely to have various effects on pathogen survival in manure-fertilized soil (5, 6, 8, 9, 18, 19, 21, 36, 40, 42, 51–53, 59, 66). Because of the wide range of variables associated with manure composition, preapplication storage, application, and incorporation, the United States Department of Agriculture National Organic Program (NOP) specifies a minimum manure application-to-harvest time interval necessary to provide adequate assurance of safety. To prevent manure-borne pathogen contamination of vegetables, the NOP regulations require at least 120 days between application of noncomposted manure and the harvesting of organic crops that have edible portions exposed to soil particles (58). Previous studies in which controlled-environment chambers and soil beds were used suggested that an application-to-harvest interval of less than 120 days might be safe in certain Wisconsin situations (39). These studies, along with the studies of other workers, validated the use of indigenous Escherichia coli as a surrogate for pathogenic E. coli O157:H7 (40) and Salmonella spp. (39). Subsequent field studies (unpublished data) suggested that indigenous E. coli may die faster in manure-fertilized soil under field conditions than in soil beds and resulted in the hypothesis that a 90-day application-to-harvest interval would adequately minimize the risk of contamination of leaf and root crops. The present field study was done to test this hypothesis in three different Wisconsin soil types by using garden-scale, low-chemical-input agricultural practices.