Water use practices

• Although limited, evidence suggests that when applying water to crops, ensuring the water does not contact the edible parts of the crop reduces microbial risk. The least risky method is subsurface irrigation, progressing to the most risky, overhead irrigation.
• Soil splash from rainwater can transfer soil and bacteria onto the undersides of crops grown outdoors.
• Pipes used to transport water can harbour microbes; Escherichia coli has been found capable of growing inside pipes between irrigation events.
• Crops can be contaminated with organisms pathogenic to humans either on the surface, or they can take up pathogens into their tissues - internalisation.
• The water interval (amount of time between last irrigation and harvest) can affect the presence of potentially harmful organisms on harvested produce.
• In addition to good irrigation water hygiene, the water interval has to be carefully chosen to avoid repeated reapplication of potentially harmful organisms.

Although the scientific evidence available is not extensive, it is sufficient to conclude that the method of applying irrigation water to fresh produce affects the microbiological risks associated with the crop. In general, and as might be expected, keeping water away from the edible parts of ready-to-eat crops (those consumed without cooking) results in a lowered risk of foodborne illness outbreaks. Thus, when choosing the most appropriate irrigation method for a particular crop, consideration should be given to the location of the edible part.

A growing body of scientific evidence shows that abstracting surface water during rainfall increases risk because surface water during rainfall contains elevated bacterial numbers 1-6. For crops that grow above ground with little clearance (e.g. salad onions, lettuce), figure 1 below illustrates the most and least risky irrigation methods.

In addition to keeping water away from edible parts, several publications have highlighted the importance of pathogen attachment in the contamination of fresh produce (8, 9). In simple terms, zoonotic agents (pathogens from animals with potential to infect humans) can recognise and adhere to the surface of some, but not all, fresh produce types. Different pathogens attach better to different crops. Research into attachment mechanisms and lists of produce susceptible to pathogens is still at an early stage. The results of the few published experiments regarding irrigation water application to various produce types are summarised below.

Song and colleagues compared flood furrow irrigation with a subsurface drip system for microbiological safety (8). Melons, lettuce, and bell peppers were grown, irrigated with water deliberately contaminated with E. coli, Clostridium, and a bacterial virus. Over 14 days, microorganism numbers were monitored on crops, soil surface, and subsoil (10 cm depth). Greater contamination was found in the furrow system compared to subsurface drip. The decline of microorganisms was fastest on leaf surfaces (phylloplanes) and soil surface, while microorganisms below the soil surface declined more slowly. Song et al. concluded that “subsurface drip had great potential to reduce health risks associated with contaminated irrigation water.” Implications for underground crops like carrots were not investigated in this study.

Barak et al. studied Salmonella fate in contaminated soils used to plant radish, turnip, and broccoli, finding higher contamination prevalence in the phyllosphere (i.e. the above-ground surfaces of plants) than in lettuce, tomatoes, and carrots grown in the same soils (12). Salmonella contamination of lettuce and tomatoes was exceptionally low, suggesting those strains do not readily attach, grow, or are chemotactically attracted to the root exudates of these crops.

Hutchison et al. studied contaminated irrigation water applied by sprinklers onto baby spinach and lettuce on two soil types in the UK (13). Pathogens (E. coli O157, Campylobacter, Listeria monocytogenes, Salmonella enteritidis) were tracked on produce and soil over time. Results showed overhead sprinklers can contaminate above-ground produce. Mootian et al. investigated low numbers of E. coli O157 transfer to crops from manure-amended soils and contaminated water, confirming transfer but without examining mechanisms (14). Fonseca et al. found bright sunshine rapidly reduces E. coli numbers (15). In Arizona, E. coli applied by overhead sprinklers was undetectable after seven days. Furrow irrigation increased E. coli soil survival (15).

Other researchers found E. coli persisted for nearly six months during winter (16). Oron estimated that drip irrigation with contaminated water reduces microbiological risks 100–1000 times compared to overhead sprinklers (17). Park et al. concluded spray irrigation using contaminated water was a major risk factor for contamination of leaf undersides (18).

Guo et al. studied Salmonella colonisation of tomato plants in hydroponic systems (19). After nine days exposure to very high contamination levels (>10,000 cfu/ml), large Salmonella counts (>1,000 cfu/g) were isolated from roots and developing cotyledons (young leaves and shoots). Plants were too young for fruit development. While the study shows potential for outbreaks if severe contamination occurs, the bacteria concentrations used were unrealistically high. What the Gou study highlights clearly is that Salmonella appear to be readily internalised by tomato plants and can spread through the plant from the roots. Koseki et al. extended these findings, comparing E. coli O157:H7, Salmonella, and Listeria monocytogenes applied to roots or leaves of hydroponic spinach at low (10³) and high (10⁶) concentrations (20). Contamination occurred primarily via roots and only at high concentrations. Leaf contamination was infrequent even at high levels. Risk of contamination through roots was seven times higher than through leaves. L. monocytogenes contamination risk was 30% that of Salmonella and E. coli O157:H7. Internalisation appears to not have been a major factor in the spinach plant compared to the tomato plant. 

Hydroponic water contains nutrients that can support bacterial growth, so water management is critical. Although primary water supply is drinking quality, in-line treatment is necessary to maintain acceptable quality. Scientific information on this aspect is currently very limited.

Pachepsky et al. investigated irrigation water pipework’s influence on water quality, focusing on microbial growth inside pipes and residual water between irrigation events (21). Stainless steel squares were placed inside irrigation pipes for a week between irrigation events. Samples of creek water, sprinkler water, residual pipe water, and biofilms on steel squares were tested for E. coli. High E. coli levels were found in residual water, suggesting growth in pipes. Biofilm bacterial counts on pipe walls were higher than in water samples. The authors recommended flushing irrigation systems prior to irrigation to reduce microbial contamination risk. Dead legs and redundant spurs in pipework can also promote bacterial growth.

Antimicrobial resistance (AMR) is an increasing concern. Blaustein et al. found multi-drug resistance in bacteria isolated from biofilms inside irrigation pipes distributing surface water (22). AMR types varied weekly. AMR bacteria readily grow in and are isolated from surface water pipe systems.

Best practice advises keeping irrigation water away from edible crop parts. However, other factors matter. Pathogenic bacteria survive longer in and on soil than on crop leaf surfaces (phylloplanes). Heavy rain can cause soil splash, transferring pathogens from soil to crops (Figure 2).

AHDB Horticulture-funded research investigated how far pathogens in contaminated soil can be transferred by raindrops (23). Figure 3 shows an agar strip (used to culture bacteria) with purple E. coli colonies transferred by a single raindrop. A large raindrop (50 µl volume) can transfer E. coli up to 0.5 metres laterally from the impact point and reach a height of 0.3 metres. Smaller raindrops transfer pathogens shorter distances.

A single simulated raindrop fell from 6 metres onto contaminated soil; splashes landed on agar strips arranged to allow growth and counting of colonies to determine dispersal distances. No significant differences were found between two loamy soil types teste (23). However, Park et al. identified growing produce on clay soil as a contamination risk factor, though the precise nature of the risk was not established (18).

Habteselassie et al. used genetically modified E. coli to trace bacterial movement and growth in soil and crops (24). Introduction of E. coli via manure or irrigation water showed colonisation of the lettuce rhizosphere. Fifteen days after establishment, E. coli was detected on lettuce leaf surfaces at 2.5 log cfu/g. E. coli persisted in bulk and rhizosphere soil for 41 days but was not detected on leaf surfaces after 27 days. This study is important because it demonstrates E. coli can move chemotactically through soil towards nutrient sources. Lettuce roots leak carbon compounds that E. coli can use for growth and multiplication. Jacobson and Bech reviewed related work (2010–2012), confirming bacterial growth in rhizospheres of tomatoes and spinach (25).

Internalisation of pathogens harmful to humans into crops is an increasing concern. Deering et al. provide a clear literature summary (26).

Key points include:

1. human pathogens such as Salmonella and E. coli O157 are present in water.
2. aeeds soaked in contaminated water, or irrigation/washing with contaminated water, can allow pathogens to invade plant tissues
3. invasion occurs through natural plant openings (stomata, lenticels, lateral roots) or damaged tissue (e.g. harvest cuts)
4. inside crops, pathogens are protected from harsh conditions and may survive extended periods but must compete with indigenous microbes
5. internalisation occurs in roots, lateral roots, root hairs, stems, leaves, and fruit

Riggio et al. reviewed six hydroponic culture types (nutrient film technique [NFT], deep water raft culture [DWC], flood and drain, continuous drip, wick, and aeroponics) (27). Each may confer different risks in leafy vegetable production. Variations in laboratory methods complicate conclusions. Internalisation has not yet been proven as a cause of human illness outbreaks. Pathogen numbers internalised are usually very low, often below infectious doses even if entire plants were consumed. It is unclear what role internalisation plays in commercial fresh produce safety. Nevertheless, compliance with relevant legislation keeps internalisation a concern for growers.

Water is also used to apply pesticides and chemicals such as nitrogen-, phosphate-, potassium (NPK) and liquid urea fertilisers. Stine et al. conducted a risk assessment using US source water contamination data broadly comparable to UK data (28). They found the annual risk of infection from Salmonella and enteric viruses could exceed 1:10,000 for some water sources. To reduce risks, surface water used for preparing pesticides or crop-applied chemicals in sprays should be microbiologically evaluated beforehand.

Irradiation, when not used to cover poor hygiene or manufacturing practices, is beneficial to consumers and poses no health hazards; it is legal to irradiate food in the UK. However, it is considered highly unlikely growers would irradiate fresh produce to ensure freedom from pathogens harmful to humans. The Food Irradiation Regulations 2009 (with near-identical versions across England, Scotland, Wales, and NI) set requirements for producing, importing, and selling irradiated food. A key issue is mandatory labelling of irradiated products, which may discourage consumers from purchasing them.

Puerta-Gomez et al. mathematically modelled irradiation as a critical control point (29). Their work showed spinach remains safe despite cross-contamination during post-harvest washing if harvested at 20°C, stored for at least 5 hours, washed with water containing chlorine at 220 mg/L, then irradiated at 1 kGy. However, 220 mg/L chlorine is unlikely to be permitted in the UK (due to legal limits of chlorate concentrations in foods), and irradiated spinach is not permitted for sale in Europe.

The water interval, the period between the final application of irrigation water and the harvest of fresh produce, plays a vital role in managing microbiological risks.

Fresh produce grown in Scotland, such as leafy greens, carrots, and soft fruits like raspberries and strawberries, is often irrigated using surface water, including rivers, lochs, or on-farm reservoirs. These sources are vulnerable to microbiological contamination, especially from faecal material, and present a recognised risk of introducing pathogens such as E. coli O157:H7, Salmonella, or Cryptosporidium to crops (30).

The water interval is a mitigation step that allows time for natural die-off of these pathogens on plant surfaces. Research has shown that E. coli O157:H7 can survive on crops for several days, depending on temperature, sunlight exposure, and humidity (16). Although Scotland’s cooler and wetter climate may slow pathogen die-off relative to warmer regions, UV radiation and dry periods still provide opportunities for risk reduction. Applying water intervals of several days (>48 hours) before harvest is recommended, particularly when using non-mains water.

Retail assurance schemes active in Scotland, such as Red Tractor, often require evidence that irrigation water quality is regularly monitored, and that irrigation water use is recorded (31). These schemes complement regulatory requirements under the Food Safety Act 1990 and Regulation (EC) No. 852/2004 on the hygiene of foodstuffs, both of which are enforced in Scotland.

As climate change increases pressure on water resources, and as the growing seasons extend, the role of the water interval will only become more critical. Used in combination with water source selection, testing, and hygienic harvest practices, the water interval provides a low-cost, effective barrier to reduce the microbial risk to consumers and protect Scotland’s reputation for high-quality fresh produce.