An investigation into the effects of floods of pesticide levels in the river tees

Abstract

Detection of significant concentrations of pesticide in drinking water (1993 to 1995) supplied from Broken Scar Treatment Works on the River Tees, County Durham prompted an investigation into the link between pesticides and river conditions. Northumbrian Water believed that pesticide contamination was related to river floods and needed to identify the conditions when the pollution risk was at its highest. Many previous studies have demonstrated pesticide in storm runoff from small catchments: however this study aimed to demonstrate the relationship between floods and pesticide concentrations in a much larger river system (Tees, 818.4 km(^2)).Three approaches were adopted to investigate the problem of pesticides in the Broken Scar catchment. Firstly, existing data of pesticide concentrations and river flows were examined. Secondly, in 1995 a sampling programme was undertaken examining pesticide concentrations in the River Tees and its major tributaries during floods of varying magnitude. Finally, ‘PESTVIEW', a GIS catchment investigation tool incorporating MAFF pesticide use data from 'FARMSTAT', was used to examine pesticide loading in the Tees catchment. The study found that the relationship between pesticide concentration in river water and floods exists, but that it is complex. Most existing pesticide data were of little value to the study as they were from customers' taps with no available accurate time-of-travel from the treatment works. However there were some data of pesticide contamination incidents, from the river, treatment works and the nearby town of Darlington, which indicated that 35% were from non-agricultural pesticides. 18% did not occur during floods and therefore were probably from point-source pollution: correspondingly 82% may be from diffuse sources. The sampling programme found simultaneous pesticide pollution in the tributaries and dilution in the flooded river. A pesticide 'spike' on the rising limb of the flood was also demonstrated. 'PESTVIEW' was found to be useful for investigation of agricultural pesticide loading and seasonal/annual trends. The catchment has two distinct areas: the upper catchment, which generates most of the river flow, has very little pesticide application and the middle catchment, which generates less flow in the river but higher loads of applied pesticide. This complexity within the catchment precluded the prediction of risk based on river flow alone. The conclusion was that more work is required and risk determination using the parameters of season, rainfall, antecedent conditions together with river flow is likely to be more successful.