Pesticide levels in surface waters in an agricultural-forestry basin in Southern Chile.
Residues of five pesticides in surface water were surveyed during 2001 and 2003 in the Traiguen river basin in Southern Chile. Simazine, hexazinone, 2,4-D, picloram herbicides and carbendazim fungicide were selected through a pesticide risk classification index. Six sampling stations along the river were set up based on agricultural and forestry land use. The water sampling was carried out before and after the pesticide application periods and in correspondence to some rain events. Pesticides were analyzed by HPLC with DAD detection in a multiresidue analysis. During 2001, in the first sampling campaign (March), the highest concentrations of pesticides were 3.0 microg l(-1) for simazine and hexazinone and 1.8 microg l(-1) for carbendazim. In the second sampling (September), the highest concentration were 9.7 microg l(-1) for 2,4-D, 0.3 microg l(-1) for picloram and 0.4 microg l(-1) for carbendazim. In the last sampling period (December), samples indicated contamination with carbendazim fungicide at levels of up to 1.2 microg l(-1). In sampling carried out on May 2003, no pesticides were detected. In October 2003, the highest concentrations of pesticides were 4.5 microg l(-1) for carbendazim and 2.9 microg l(-1) for 2,4-D. Data are discussed in function of land use and application periods of the products, showing a clear seasonal pattern pollution in the Traiguen river. Risk assessment for these pesticides was calculated by using a risk quotient (RQ = PNEC/PEC). For picloram the calculated RQ < was 0, which indicates that no adverse effects may occur due to the exposure to this herbicide in the Traiguen river basin. For 2,4-D, simazine, hexazinone, carbendazim RQ > 1, meaning that adverse effects could occur and it is necessary to reduce pesticide exposure in surface waters. It is recommended to continue with a pesticide monitoring program and the implementation of ecotoxicological testing with local and standardized species in order to consider the probability of effects occurrence, with less uncertainty. Thus, it will be more feasible to make some recommendations to regulatory agencies regarding the pesticide use.