Integrated water resources management: a case study of on-farm water use for potato processing

Although Canada has large quantities of fresh water, its renewable supply dropped by 8.5% from 1971 to 2004 (Statistics Canada 2010b). Statistics Canada (2009) estimated that 4.8 billion m³ of water was used for irrigation, livestock, spraying, and cleaning in 2001, with most of that water used for irrigation.

Historically, little attention has been paid to studying and managing non-irrigation water uses in agriculture; however, recent increases in farm size and pressures on water availability and water quality have led to rapid changes in agricultural water-use regulations. Meanwhile, in the horticulture sector, farmers are increasingly being expected to wash produce fully before it leaves the farm. This expectation has created the situation in which our case study takes place: regulators had approached potato farmers to require them to comply with environmental standards for other industries, whereas none of the stakeholders had a firm scientific understanding of the quantity of water used or the quality of the effluent produced.

Integrated water resources management (IWRM) is described by the United Nations as a holistic approach to manage water efficiently, equitably, and sustainably. This approach has also been defined as the ‘coordinated development’ of strategies for managing water alongside other resources in an economically and environmentally sustainable manner (Roy et al. 2009). This case study documents how this approach, involving cooperation among researchers, farmers, and the provincial government, was applied to understand and manage water use on a large potato farm in Ontario, Canada.

Although the water requirements for irrigating potatoes are well known, the water requirements for storing, processing, and washing potatoes on the farm before processing are not well understood or reported (Statistics Canada 2010a). Because farmers use water sources that are not connected to water utilities, limited information has been gathered on actual water use (De Loë et al. 2001). Filling this knowledge gap is important for guiding science-based conservation targets and understanding water supply and demand to maintain future water availability (De Loë et al. 2001).

This research was performed as part of a larger study to understand the quantity and quality of potato wash water as well as develop a treatment system to improve the quality of the effluent that is produced. It was evident that focusing on treating the potato processing water alone was insufficient for effective management; rather, the essential first step was to understand the management factors that affect the quantity and timing of water use and wastewater production as well as the quality of the effluent. The project team was able to follow the IWRM approach thanks to the combined efforts (and overlapping interests) of three parties: university and government researchers, potato farmers, and provincial government regulators.

Farmers were informed by the Ontario Ministry of the Environment and Climate Change (i.e. the provincial regulatory body) that new water regulations were being implemented and would require a higher level of treatment for on-farm potato wash water. This news led the farmers to contact researchers, which led to a research project that included a water monitoring program. The project was led by a research team, and the farmers made a large contribution and were active participants throughout. The main function of the provincial regulatory body was enforcement and oversight, and the Ministry granted the farm a research permit to continue to improve its system without sanctions.

The potato processing took place at the storage facility of Sunrise Potato Storage Ltd. in Alliston, ON, Canada, which stores about 300,000 cwt of potatoes each year (3.0 × 10⁷ lbs; 1.4 × 10⁷ kg; hundredweight [cwt] is a standard unit in the potato industry; Bosak Zurowsky 2015). At harvest, the potatoes were graded by machines to remove soil and were piled into climate-controlled storage bins.

Three flow meters (Rain Bird, Azusa, CA, USA) were installed in August 2012 in the storage facility to monitor water use. The flow meters were all the same model, with a resolution of 0.378 L (0.1 USG), and were fitted to 5/8-in. piping. Each meter measured a different water source that served a specific function during potato handling, as follows:

• Flow meter 1 (FM_pump) measured the water used to lubricate the bearings on the flume water recirculation pump.

• Flow meter 2 (FM_flume) measured the water used to fill the flume before and during potato handling.

• Flow meter 3 (FM_tank) measured the water used to fill a large water tank, which in turn was used to wash the flume and remove sediment.

Another flow meter (Stenner Pump Co., Jacksonville, FL, USA) was installed in February 2013 on the main water line supplying well water to the entire storage facility (FM_tot; Figure 3(c)). This included water used for kitchens, bathrooms, storage humidity control, and shipping. The flow meter had a resolution of 0.378 L (0.1 USG), a manufacturer-specified accuracy of ±1.5%, and a maximum flow rate of 605.7 L min⁻¹ (160 USG min⁻¹).

Data was recorded manually by farm workers on data sheets, which were kept near the meters. Flow meter data record sheets were the responsibility of the potato storage secretary. Values for washing and processing were recorded after every shipping day, and values from the whole-storage flow meter were recorded monthly.

The researchers were in charge of checking the equipment, gathering the data, and synthesizing the data into a presentable format. Regression analysis was done using SigmaPlot software, version 12.0 (Systat Software Inc., San Jose, CA, USA). Paired t-tests were performed on total monthly water flows through each flow meter to find the statistical difference between year 1 (Y1) and year 2 (Y2) data.

After Y1 of water monitoring, the results were presented to farm management and the district environmental office. Because the water-use values for lubricating the pump bearings and for cleaning the flume with water from the tank were higher than anyone had expected, the focus was placed on reducing these uses where the farm managers thought conservation could be achieved. Together, all parties discussed and developed conservation strategies for Y2. The researchers presented their results to the farmers first, and together the researchers and farmers discussed in detail the patterns observed, the reasons for the patterns, and possible ways to improve the system to reduce water use. Following this discussion, a meeting was held with the district environmental officer to present and discuss the results and future strategies. This meeting led to the finalization of water-reducing strategies and a renewed research permit.

The first focus was the water used to prevent sand or debris in the wash water from damaging the pump bearings (FM_pump). Farm management described their practices in Y1: fully opening the bearing-lubrication valve at the start of each day and allowing it to flow all day. Although the farm managers knew from previous experience that the pump bearings required continuous water flow to prevent bearing failure, the actual flow required was unknown, and the flow supplied through the valve was also unknown. Before they saw the results, the farm managers had thought that the valve flow was negligible, and they had maintained maximum flow as a precaution against bearing failure. The reduction strategy agreed upon was to determine the minimum water required for proper bearing function and then install an appropriate flow-control device.

The second focus was the practice of washing the flume (FM_tank) after each day of potato handling. The farm managers explained that this was done to prevent the buildup of solids that could damage equipment, clog the wastewater outflow pipe, or be esthetically displeasing. Discussions led to two synergistic strategies for reducing this water use: one direct strategy and one indirect strategy. The direct strategy was that the farm would reduce the number of flume-washing events by modifying its procedure so that on days when fewer potatoes were handled, the flume would be washed down every other day or only when necessary, rather than daily. The indirect strategy was that the farm would decrease the need for flume washing by reducing the amount of solids entering the flume in the first place. The researchers highlighted that the wastewater contained 24.5 t of total suspended solids in Y1 and pointed out that the finger-roller system before the flume was removing solids and that enhancing that system would substantially reduce the amount of sediment in the flume. Putting the results into a unit that was meaningful to the farmers (t rather than mg L⁻¹) was valuable for providing perspective on the matter. The farmers were quick to suggest modifications that could be made to their equipment to remove soil and solids before they could enter the flume.

A flow restrictor was installed in January 2014, three months after the start of Y2 of monitoring. The flow restrictor was a simple device purchased from a local hardware store and attached to the hose delivering water to the pump. When the minimum flow volume needed was found, the knob was set at the desired level and locked in place to prevent accidental adjustment. Therefore, at the time of shipping, the main water valve was opened all the way, as it had been before, but now the new attachment automatically restricted to the volume to the desired level. This change resulted in consistent water conservation, regardless of which farm staff member was managing the water system on a particular day.

The new dirt-eliminating finger rollers were installed in February 2014 as part of a custom-built machine (Figure 2(c)) purchased by the farmers with partial cost recovery through a government environmental program. The machine was used in the same way as before (before the potatoes entered the flume), but the new machine (Figure 2(c)) had a significantly larger capacity and an improved design in comparison with the original machine (Figure 2(a)). A conveyor belt moved the potatoes from storage and through the series of finger rollers and brushes, removing dirt as well as potatoes that had been sliced during harvest or handling, were rotten, or were too small (e.g. Figure 3(b)). Then the potatoes were dropped into the flume, along which they were transported for the remainder of the washing and sorting process.

Reduced flume-washing events were implemented immediately at start of the shipping season (October 2013). Because the implementation of this change in practice was controlled by the storage workers, it was their conscientious efforts to conserve water every day that made it possible to reduce water use. Periodic updates were shared with the farm owners, and discussions were ongoing throughout Y2 of monitoring.

Final assessment began with the researchers giving a synthesis and presentation of the results obtained in Y2 and comparisons with Y1. In Y2, the average monthly water use was 234 ± 148 m³ and the average monthly potato shipment was 25,000 ± 19,000 cwt. The total water use for potato washing and fluming in Y2 was nearly half the total in Y1 (42% reduction from Y1 to Y2; p = 0.045), and the average water use per unit of potatoes shipped dropped by 31% (from 13.2 to 9.1 L cwt⁻¹). These results indicated that the conservation techniques were effective.

The breakdown of the water usage showed that the water used to fill the tank for cleaning the flume (FM_tank) had the largest reduction in usage (67%), from 1,481 m³ in Y1 to 483 m³ in Y2. The water used to lubricate the pump bearings (FM_pump) was reduced by half (53%), from 1,289 m³ in Y1 to 611 m³ in Y2 (Table 1; Figure 5). After the flow restrictor was installed in January 2014, midway through Y2, an even larger decrease in water use was observed (69%; comparing February–July). Lastly, the water used to fill the flume (FM_flume) was essentially unchanged from Y1 to Y2; that result was expected, given that conservation efforts were not focused on this function.

In just over a year (373 d), the entire storage facility used 5,533 m³ of water for all activities, including 3,793 m³ for shipping (23 February 2013 to 3 March 2014). The average water consumption per day was 14.7 m³; the highest rate of use for the entire storage facility was in July, averaging 27 m³ d⁻¹, whereas the lowest was in November, at 4.4 m³ d⁻¹. The total water used for washing and fluming potatoes during this time was a third of the whole-storage water use (31% or 3,793 m³). The remainder of the water was used for bathroom and kitchen facilities, equipment and storage cleaning, crop spraying, and humidity control for potato storage (in no particular order). During this time, 247,000 cwt of potatoes were flumed, washed, and shipped. Therefore, the rate of water use for the end goal of shipping those potatoes, from the entire storage facility, was 0.022 m³cwt⁻¹; of that, fluming and washing accounted for 0.015 m³ cwt⁻¹.

Through monitoring and simple changes, this farm reduced its water use by 31% per unit of potatoes sold. The changes required relatively little capital investment and resulted in no reduction in yield or disruption of operations. This demonstrates the value of IWRM and highlights that even modern farm businesses benefit from partnerships with other organizations to better understand farm operations. This is especially true when it comes to water management – although modern farms use sophisticated technologies to monitor and manage crop production and handling, very little attention is normally given to monitoring, understanding, and managing water use within the farm facility.

Global potato production in 2012 was estimated to be 365 Mt, with 4.6 Mt (1.014 × 10⁸ cwt) produced in Canada and 19 Mt produced in the United States (FAO 2015). Therefore, on-farm potato processing uses a substantial quantity of water that has not been studied carefully and is not well understood. That amount of water is even higher when other horticultural crops are taken into consideration. It is important that appropriate research is carried out to understand that water use before policies are implemented to regulate it. Water use in horticultural production and storage facilities is unlike water use on livestock or crop farms, and research is needed to properly understand and characterize these unique water uses.

Returning to the start of the IWRM cycle, all parties agreed on directions for future research:

• Reusing water to create a no-discharge system and reduce the reliance on groundwater.

• Carrying out more studies to gain a better representation of on-farm processing water use on potato and other crop farms in Ontario as well in as other areas.

This study used IWRM as an approach to understand water use and achieve water conservation with cooperative work by researchers, potato growers, and provincial governmental regulators. The cooperative strategy building and the farm's efforts to implement conservation strategies successfully reduced total water consumption for potato washing and handling by 42% (4,044 m³ in Y1 and 2,345 m³ in Y2). This reduction was achieved by reducing the water used to lubricate the pump bearings and clean the flume and decreasing the amount of solids on the potatoes, which in turn decreased the amount of water needed to clean the flume. The amount of water used for processing potatoes was found to average 13.2 L cwt⁻¹ in Y1 and 9.1 L cwt⁻¹ in Y2. Given that many of the changes were not implemented until January of the second shipping season, we expect that larger reductions are possible in future years. The ability to greatly reduce water use in Y2 highlights the importance of cooperation among multiple stakeholders to understand and manage water resources. In this case study, we were fortunate to identify relatively simple and inexpensive changes – both technology- and practice-based – that did not compromise yield or disrupt operations. This work draws attention to the water required for washing and handling processes on potato farms (and in the horticulture sector in general): the amount of water is significant but is not well researched or documented. Future research is needed to address these knowledge gaps and to help farmers understand and manage their water resources in Ontario and elsewhere.

This research was funded by the Ontario Ministry of Agriculture, Food, and Rural Affairs, with support from Agriculture and Agri-Food Canada and the University of Guelph. Special thanks go to the management and employees at Sunrise Potato Storage Ltd.