View Report Details
A review of the use of water in UK agriculture and the potential risks to food safety
Project Code: B17001
ADAS Gleadthorpe Research Centre
Groves, S ;
SAC Heartland Environmental
There was a sixfold increase in the number of food-borne illnesses reported in the UK between 1982 and 1998. In the majority of cases these illnesses were caused by bacteria such as Salmonella, Campylobacter and Verocytotoxic Escherichia coli. Additionally, parasitic protozoa Cryptosporidium and Giardia, as well as viruses, were associated with human illness. It is likely that in a significant proportion of these cases, transmission to humans was via food or water contaminated with these organisms.
It is generally accepted that the goal of reducing the number of food borne illnesses is best served by the identification of potential points of pathogen entry into the food chain and implementation of effective controls.
The use of water in agriculture provides a possible point of entry into the food chain as many bacteria, viruses and protozoa of faecal origin can be found in waters which are used in the primary production of food crops. However, it should be noted that use of water containing pathogens does not necessarily result in an increased risk to food safety, as subsequent treatments such as produce washing or cooking may limited the potential for disease transmission.
This report summarises the agricultural practices involving the use of water, current knowledge of the levels of pathogens in agricultural waters and reviews published reports of disease transmission linked to agricultural water use. In particular, the management practices associated with crop irrigation are reviewed in conjunction with data describing the sources of irrigation water, and the geographical distribution of irrigated crops. Using this information together with our knowledge of likely levels of pathogens in irrigation water, the risks of transfer to the food chain are discussed. Relevant irrigation guideline documents for agricultural production in other countries are reviewed and their applicability to UK farming considered.
The report concludes with a tentative assessment of the relative risk to human health associated with crop irrigation. Recommendations are provided for future research designed to gather more information on the factors which affect the risks of pathogen transfer.
Disease outbreaks linked to water used in agriculture
Historically, the use of poor quality water in agriculture has frequently been linked to the outbreak of disease in under-developed countries. In developed nations, the incidence of pathogen transfer to humans from agricultural water is substantially lower, but still occurs. The literature indicates that the major routes of transmission are (i) irrigation water applied to ready to eat (RTE) crops; (ii) water used in pesticide sprays; and (iii) water used for post-harvest washing.
Pathogens in agricultural waters
The major uses of water in agriculture are crop irrigation, produce washing and livestock drinking. In 1995, approximately 191,900 hectares of land in the UK was irrigated with 177 million m3 of water to improve crop yields and quality. The sources of water used for irrigation included rivers and streams (51%), deep boreholes (32%), ponds and lakes (7%), springs and wells (4%), main supply (2%) and other sources (3%). In general, growers are not able to choose their water source as this is determined by availability within the local environment. A proportion of these waters will contain human pathogens which have the potential to enter the food chain.
In livestock systems, the risks are associated with the infection and cross-infection of livestock, and the subsequent excretion of pathogens in animal excreta. This can lead to the direct contamination of meat products during the slaughtering process or to the pollution of watercourses which may be used elsewhere as a source of irrigation water. Pathogens present in water used for crop irrigation may directly contaminate crop plants and thereby transmit disease.
River and stream water: The levels of pathogens, as assessed by faecal coliform numbers, have been shown to exceed 104 colony forming units (CFU)/100ml in some river systems used for crop irrigation. The pathogen loads in river water are influenced by the outfall of sewage treatment plants and by runoff from land occupied by livestock. The faecal coliform load of river water is further influenced by heavy rainfall events, season and abstraction location, so that the quality of water in a given river system will show substantial spatial and temporal variability.
There are currently no data on pathogen loads which relate specifically to water abstracted from rivers for use in irrigation. Such information is required before the risks to food safety associated within the use of river water can be accurately assessed.
Deep borehole water: There are few data which relate specifically to the levels of pathogens in borehole water used for irrigation. In general, it is likely that borehole water will contain fewer pathogens than river water as the potential for direct faecal contamination is limited. Boreholes sited where livestock occupy the land around the abstraction point, or where geological fissures allow rapid aquifer recharge present an increased risk of faecal contamination.
On-farm water storage: The majority of on-farm reservoirs are filled from river systems during the winter months. The pathogen load of reservoir water is therefore determined by the load of source water and the rate of pathogen decline which may occur between the time of abstraction and use (normally > 3 months). Laboratory studies with inoculated reservoir water have shown that where temperatures exceeds 15°C, a 3 log reduction of E. coli O157:H7 occured during a 12 week period. At 8°C pathogen decline was slow (1log over 14 weeks). Pathogen survival will be longer in 'clean' water than in water containing microbial organisms, which may be associated with pathogen predation.
As with river and borehole water, no data are available regarding the pathogen load of stored waters at the point of use.
Spring and well water: Spring and well waters are susceptible to faecal contamination by runoff from land occupied by livestock. In some cases, heavy rainfall and subsequent runoff have been associated with E. coli counts rising from near zero to 6x103 (CFU/100ml).
The prediction and assessment of water quality
The faecal contamination of surface waters is a continuous process which gives rise to temporal and spatial variations in water quality. As the factors that affect water quality are variable, the prediction of pathogen loads at a given abstraction point is problematic. Additionally, temporal changes present challenges for any monitoring schemes designed to characterise water quality at the point of use.
Water sources, such as on farm reservoirs and deep boreholes, are likely to exhibit less fluctuation in pathogen loads as they are less susceptible to significant contamination. In these cases, water quality monitoring schemes are likely to provide a good indication of quality at the point of use.
There are many different irrigation management practices used in the UK. Specific practices vary with crop type, market requirement and geographic location. In the context of this review, there are four main elements which may impact on food safety.
1. Water Source - The quality of water used for crop irrigation is governed by the quality of water available in the local environment and is therefore likely to vary from farm to farm. In most cases, growers cannot significantly improve the quality of water available to them through management practices, as the factors which determine quality are largely external to their farms. The on-farm treatment of water to improve quality is costly and rarely used. On-farm water storage, although costly, may allow pathogen numbers to decline during the storage period.
Although the widescale use of mains water would reduce the risk of pathogen transfer, the economic, social and environmental affects would be substantial.
2. Application Method - Application method is relevant in that, if contact between irrigation water and the edible parts of the crop can be avoided, the risks of pathogen entry into the food chain will be reduced. The overwhelming majority of irrigation water is applied as a spray from above the crop (95%) so that both the aerial and below ground parts of the crop are wetted. Soil surface drip irrigation is less common (<5% of irrigated crop area) and tends to be used on high value salad and fruit crops. This method avoids wetting the aerial parts of the crop, and therefore may reduce the risk of pathogen transfer in some crop species.
3. Irrigation Management - The application of irrigation water is planned to ensure optimum crop growth and produce quality. The appropriate schedule for irrigation varies within crop type, crop water demand, target market, soil texture, rainfall and other weather parameters. These factors determine the amount of irrigation water applied and the interval between the last application of water and harvest. The harvest interval may impact on food safety as the numbers of pathogens on crops surfaces are likely to decline over time due to the effects of UV radiation and desiccation. It is common for the harvest interval of many salad crops following the last application of irrigation water to be short (1 to 7 days).
4. Crop Type - Of all the crop species which receive irrigation, the risk of pathogen transfer into the human food chain is greatest for RTE crops. Risks are generally lower for other species which are cooked before consumption. Many salad and fruit crops rely heavily on crop irrigation to achieve the appropriate quality standards required by the retail trade.
Produce washing and cooling
Many vegetable and salad crops are washed and cooked prior to despatch to retail outlets. As this process often involves the re-circulation of wash water, there is potential for the cross contamination of produce during the early stages of the washing process. However, this risk is minimised where potable water is used in the final rinse or where chlorinated water is used during the hydro-cooling phase. Many RTE crops including lettuce, tomatoes, cucumber, soft and top fruit are commonly not washed prior to despatch to retail outlets.
Pesticide and fertiliser applications
The overwhelming majority of crops in the UK (excluding the organic sector) receive at least one pesticide or fertiliser application. In general, mains water is used for these applications as high quality water is required to ensure spray nozzles do not block. As a consequence, the risks of pathogen transfer via this route are limited.
Surface, spring, borehole and mains water are all used as sources for drinking water in livestock systems. However, reliable data regarding the amounts of water used in livestock production are difficult to obtain. Crude estimates suggest that between 210 and 360 million m³ of water per year is consumed by livestock, and a further 20-35million m³ is used during washdown procedures.
The now out of date (1981) MAFF (now DEFRA) guidance on Irrigation Water Quality makes no reference to the possibility of microbial contamination. However, some UK produced quality assurance documents discuss irrigation water quality issues but draw back from providing comprehensive guidelines or detailed procedures for the assessment of risk.
More comprehensive guidance is available for growers in other countries. Commonly, other countries use a threshold faecal coliform number as an indicator of irrigation water quality. This is often combined with recommendations for certain management practices such as irrigation application method, or harvest interval designed to reduce risk. Although the principals for risk reduction used in other warmer countries are applicable to the UK, the threshold values and time intervals suggested may be inappropriate for the cooler climate which prevails in the UK.
There is little consensus in the guidance documents regarding the frequency of water testing required to characterise the quality of waters used in agriculture. However, it is accepted that frequency should be related to potential variability within the water source. If farmers and growers are to be expected to decide on sampling patterns, more clear cut guidance would be required.
Recommendation to the FSA
• It is evident that there is little data available which quantifies the pathogen loads in water used in UK agriculture. A surveillance exercise is recommended to assess agricultural water quality at the point of use, so that the risks to food safety can be appropriately assessed.
• Appropriate on-farm guidelines are required to ensure that any risks to food safety from agricultural water usage is minimised. This guidance should accommodate factors such as water quality, irrigation application method, ‘harvest intervals’, crop type and post harvest treatment. Guidance is also required with regard to suitable procedures for the assessment of water quality prior to use.
• There is a need to investigate the rate of pathogen decline on RTE crops grown under UK climatic conditions and the effect leaf structure and orientation may have on pathogen survival.
Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: