Comanaging fresh produce for nature conservation and food safety

Abstract

In 2006, a deadly Escherichia coli O157:H7 outbreak in bagged spinach was traced to California's Central Coast region, where >70% of the salad vegetables sold in the United States are produced. Although no definitive cause for the outbreak could be determined, wildlife was implicated as a disease vector. Growers were subsequently pressured to minimize the intrusion of wildlife onto their farm fields by removing surrounding noncrop vegetation. How vegetation removal actually affects foodborne pathogens remains unknown, however. We combined a fine-scale land use map with three datasets comprising ∼250,000 enterohemorrhagic E. coli (EHEC), generic E. coli, and Salmonella tests in produce, irrigation water, and rodents to quantify whether seminatural vegetation surrounding farmland is associated with foodborne pathogen prevalence in California's Central Coast region. We found that EHEC in fresh produce increased by more than an order of magnitude from 2007 to 2013, despite extensive vegetation clearing at farm field margins. Furthermore, although EHEC prevalence in produce was highest on farms near areas suitable for livestock grazing, we found no evidence of increased EHEC, generic E. coli, or Salmonella near nongrazed, seminatural areas. Rather, pathogen prevalence increased the most on farms where noncrop vegetation was removed, calling into question reforms that promote vegetation removal to improve food safety. These results suggest a path forward for comanaging fresh produce farms for food safety and environmental quality, as federal food safety reforms spread across ∼4.5 M acres of US farmland.

Keywords: E. coli; agriculture; biodiversity; disease ecology; foodborne pathogens.

Fig. 1.

(A) EHEC prevalence in leafy green vegetables increased significantly from 2007 to 2013 (likelihood ratio test: n = 21 region-years/482,208 tests, χ² = 16.8, P < 0.001). The rate of increase was marginally lower in California counties outside the Central Coast region and much lower in other states and countries (interaction: χ²= 10.7, P = 0.005). (B) Salmonella increased marginally over time (χ²= 2.75, P = 0.10) across all states and countries (interaction: χ²= 0.47, P = 0.79). Solid lines represent predicted pathogen prevalences from linear mixed models; dotted lines, prediction intervals. Points are observed percentages of infected samples in each year and region. (C) Percent change in Salinas Valley land cover from 2005 to 2012 within 50 m of cropland. Numbers represent absolute changes.

Fig. 2.

(A) Surrounding land cover did not significantly predict changes in the prevalence of generic E. coli in water (SI Appendix, Table S1). (B and C) Although EHEC was more prevalent in leafy greens at sites with more surrounding grazeable land (model averaging: n = 236,522 tests across 57 farms, Z = 3.8, P < 0.001), we found no evidence that riparian or other nongrazed natural vegetation increased EHEC prevalence (SI Appendix, Table S1). Points and lines in A and B are model-averaged estimates and confidence intervals of land cover effects on pathogen prevalence (filled circles; P < 0.05). The solid line in C relates surrounding grazeable land to pathogen prevalence; dotted lines represent prediction intervals.

Fig. 3.

(A) EHEC prevalence in leafy greens increased on farms that replaced nonriparian natural vegetation with crops between 2005 and 2012 (likelihood ratio test: n = 28 farms, χ² = 4.22, P = 0.04). (B and C) In contrast, EHEC did not change when riparian vegetation was removed (n = 28, χ²= 0.07, P = 0.79) (B) and increased when other natural vegetation was removed (note the negative scale of the x-axis; n = 28, χ² = 4.55, P = 0.03) (C). Solid lines depict predicted effects on EHEC from linear models, dotted lines are prediction intervals, and points are farms.

Fig. 4.

Collaborative action among growers, ranchers, and feedlot operators could reduce food-safety risk while maintaining the conservation value of agricultural landscapes. Promising practices include (1) planting low-risk crops between leafy green vegetables and pathogen sources (e.g., grazeable lands); (2) buffering farm fields with noncrop vegetation to filter pathogens from runoff (25, 26); (3) fencing upstream waterways from cattle and wildlife; (4) attracting livestock away from upstream waterways with water troughs, food supplements, and feed (33); (5) vaccinating cattle against foodborne pathogens (31); (6) creating secondary treatment wetlands near feedlots and high-intensity grazing operations (32); (7) reducing agri-chemical applications to bolster bacteria that depredate and compete with E. coli (34); (8) exposing compost heaps to high temperatures through regular turning to enhance soil fertility without compromising food safety (4); and (9) maintaining diverse wildlife communities with fewer competent disease hosts (21).