Agricultural impacts on water quality and implications for virtual water trading decisions

Research on the flow of virtual water associated with agricultural crop production and trade has focussed almost entirely on water quantity. It is pertinent to consider and quantify the opportunity costs in terms of reduced water quality associated with crop production. This paper investigates the impacts of water quality on virtual water trading by creating a proxy for water quality impacts by calculating the amount of water required to dilute nonpoint-source agrochemical inputs to relevant water quality guideline values. The quantity of water required for dilution of five agrochemicals (two nutrients; nitrogen and phosphorus and three insecticides; azinphos-methyl, chlorpyrifos and endosulfan) was estimated for five crops in South Africa (maize, wheat, sugar cane, citrus and cotton) and compared to consumption of irrigation water (blue water) and rainfall (green water) for the same crops. Results indicate that the volume of water required for dilution is similar to the total sum of green and blue water required for crop production, but significantly greater than blue water use (irrigation use). For all crops phosphorus losses require greater amounts of water for dilution than for nitrogen, while pesticides result in the greatest water quality use. Estimates of water quality use are based on assumptions for a number of input variables (i.e. fertilizer application rates, percentage loss of agrochemicals from cropped areas). A Monte Carlo analysis (5000 iterations) was run to randomly select input variables from within defined ranges. Water quality use was calculated and expressed as a factor of blue water use. For all crops the average factor indicated that the volume of water required for dilution of all agrochemicals was greater than that required for irrigation. The results of this study clearly indicate that the impacts of agriculture on water quality need to be considered in virtual water trading scenarios. The incorporation of a method to predict impacts on... [Copyright &y& Elsevier]