RESUMEN
Traditionally, ecotoxicity quantitative structure-activity relationships (QSARs) for alcohol ethoxylate (AE) surfactants have been developed by assigning the measured ecotoxicity for commercial products to the average structures (alkyl chain length and ethoxylate chain length) of these materials. Acute Daphnia magna toxicity tests for binary mixtures indicate that mixtures are more toxic than the individual AE substances corresponding with their average structures (due to the nonlinear relation of toxicity with structure). Consequently, the ecotoxicity value (expressed as effects concentration) attributed to the average structures that are used to develop the existing QSARs is expected to be too low. A new QSAR technique for complex substances, which interprets the mixture toxicity with regard to the "ethoxymers" distribution (i.e., the individual AE components) rather than the average structure, was developed. This new technique was then applied to develop new AE ecotoxicity QSARs for invertebrates, fish, and mesocosms. Despite the higher complexity, the fit and accuracy of the new QSARs are at least as good as those for the existing QSARs based on the same data set. As expected from typical ethoxymer distributions of commercial AEs, the new QSAR generally predicts less toxicity than the QSARs based on average structure.
Asunto(s)
Alcoholes/toxicidad , Contaminación Ambiental/análisis , Tensoactivos/toxicidad , Contaminantes Químicos del Agua/toxicidad , Algoritmos , Animales , Cromatografía Líquida de Alta Presión , Daphnia , Peces , Modelos Estadísticos , Relación Estructura-Actividad CuantitativaRESUMEN
By means of GREAT-ER (Geo-Referenced Regional Exposure Assessment Tool for European Rivers) aquatic chemical fate simulations can be performed for river basins. To apply the resulting digital maps with local (river stretch specific) predicted concentrations in regional aquatic exposure and risk assessment, the output has to be aggregated to a (single) value representative of exposure in the catchment. Two spatially aggregated PEC definitions are proposed for this purpose: PECinitial (unweighted aggregation of concentrations just downstream of wastewater emissions) and PECcatchment (weighted aggregation of all average stretch concentrations). These PECs were tested using simulations for two pilot study catchments (Calder and Went, UK). This confirmed the theoretical considerations which led to the definitions, and it illustrated the need for weighting to resolve scale-dependencies.