Through normal agricultural use, pesticides may reach environmental water bodies via several routes of entry. Various policies and initiatives exist to reduce the effects of pesticides in the environment. One such initiative in place in the UK is the Voluntary Initiative (VI). The VI is a voluntary scheme put forward by the Crop Protection Association with other crop protection and farming organisations to reduce the environmental impacts of pesticides. Mathematical models of pesticide fate can usefully be applied to examine the impact of factors influencing the contamination of water bodies by pesticides. The work reported here used water quality models to examine how changes in farmer behaviour could potentially impact pesticide contamination of environmental water bodies. As far as possible, uncalibrated, standard regulatory models were used. Where suitable models were not available, simple models were defined for the purposes of the study and calibrated using literature data. Scenarios were developed to represent different standards of practice with respect to pesticide user behaviour. The development of these scenarios was guided by the Crop Protection Management Plan (CPMP) aspect of the VI. A framework for the use of modelling in the evaluation of the VI is proposed. The results of the modelling study suggest that, in several areas, widespread adoption of the measures proposed in the VI could lead to reductions in pesticide contamination of environmental water bodies. These areas include pesticide contamination from farmyards, spray drift and field runoff. In other areas (including pesticide leaching to groundwater and contamination of surface water from field drains) the benefits that may potentially be gained from the VI are less clear. A framework to evaluate the VI should take into consideration the following aspects: (1) groundwater is more at risk when there is a combination of leachable compounds, vulnerable soils, shallow groundwater and high product usage; (2) surface water contamination from drains is most likely when heavy rain falls soon after application, the soils are vulnerable and product usage is high; (3) surface water contamination from drift is most likely when the distance between the spray boom and water body is small and product usage is high; (4) surface water contamination from farmyards is dependent on the nature of the farmyard surface, the competence of the spray operator and the level of product usage. Any policy or initiative to reduce pesticide contamination should be measured against farmer behaviour in these areas.