Under the Plant Protection Product (PPP) Authorisations Directive (91/414/EEC) the risk of a PPP to off-crop non-target aquatic organisms is assessed in a tiered approach. From the properties and use pattern of the product, the likely routes of entry into surface water are assessed for PPPs applied as sprays. These assessments are based upon a calculated percentage of the active substance being deposited on a static body of water, 30cm deep, related to the distance from the end of the spray boom to the edge of the water body (Ganzelmeier et al. 1995). For some PPPs, such as chlorpyrifos, a buffer (no spray zone) may be applied to "in use" situations to reduce drift off-crop. However, there is little data to demonstrate how well drift events with specific chemicals match Ganzelmeier data or the extent to which application factors such as wind speed and spray nozzle affect the degree and amount of drift. Using a large-scale wind tunnel, a series of controlled, replicated studies were carried out to measure the influence of two wind speeds in combination with a conventional and three star (UK) rated reduced drift nozzle on the spray drift of chlorpyrifos, applied as Dursban 4, and its deposition on to an artificial ditch, simulating a static edge of field water body, 30cm deep. Results showed a clear reduction in amounts of chlorpyrifos as distance from the nozzle increased. The combination of 3mph (low) wind speed and low drift nozzle had a significant influence in reducing drift by ca. ten-fold at 2m from the spray nozzle, and five-fold at the mid-ditch position (4.5 or 5m), as measured by polyethylene strings stretched horizontally across the path of the drift. Water concentrations were reduced by ca. half from an average of 1. 11 mug L-1 to 0.45mug L-1 The presence of a 50cm artificial bank had no significant influence on the concentration of chlorpyrifos in surface water. Results show that both low wind speed and low drift nozzle can contribute to risk reduction of certain PPPs in surface water.