Pyrethroids are the active ingredients in most insecticides available to consumers for residential use in the United States. Yet despite their dominance in the marketplace, there has been no attempt to analyze for most of these compounds in watercourses draining residential areas. Roseville, California was selected as a typical suburban development, and several creeks that drain subdivisions of single-family homes were examined. Nearly all creek sediments collected caused toxicity in laboratory exposures to an aquatic species, the amphipod Hyalella azteca, and about half the samples caused nearly complete mortality. This same species was also found as a resident in the system, but its presence was limited to areas where residential influence was least. The pyrethroid bifenthrin is implicated as the primary cause of the toxicity, with additional contributions to toxicity from the pyrethroids cyfluthrin and cypermethrin. The dominant sources of these pyrethroids are structural pest control by professional applicators and/ or homeowner use of insecticides, particularly lawn care products. The suburbs of Roseville are unlikely to be unique, and similar sediment quality degradation is likely in other suburban areas, particularly in dry regions where landscape irrigation can dominate seasonal flow in some water bodies.

The agricultural industry and urban pesticide users are increasingly relying upon pyrethroid insecticides and shifting to more potent members of the class, yet little information is available on residues of these substances in aquatic systems under conditions of actual use. Seventy sediment samples were collected over a 10-county area in the agriculture-dominated Central Valley of California, with most sites located in irrigation canals and small creeks dominated by agricultural effluent. The sediments were analyzed for 26 pesticides including five pyrethroids, 20 organochlorines, and one organophosphate. Ten-day sediment toxicity tests were conducted using the amphipod Hyalella azteca and, for some samples, the midge Chironomus tentans. Forty-two percent of the locations sampled caused significant mortality to one test species on at least one occasion. Fourteen percent of the sites (two creeks and four irrigation canals) showed extreme toxicity (>80% mortality) on at least one occasion. Pyrethroid pesticides were detected in 75% of the sediment samples, with permethrin detected most frequently, followed by esfenvalerate > bifenthrin >lambda-cyhalothrin. Based on a toxicity unit analysis, measured pyrethroid concentrations were sufficiently high to have contributed to the toxicity in 40% of samples toxic to C. tentans and nearly 70% of samples toxic to H. azteca. Organochlorine compounds (endrin, endosulfan) may have contributed to the toxicity at a few other sites. This study provides one of the first geographically broad assessments of pyrethroids in areas highly affected by agriculture, and it suggests there is a greater need to examine sediment-associated pesticide residues and their potential for uptake by and toxicity to benthic organisms.

-Toxicokinetic models are not constrained by assumptions of equilibrium as are thermodynamic (equilibrium-partitioning) models and are more accurate predictors of toxicant accumulation for non-steady-state exposures and multiple uptake routes. Toxicokinetic models -compartmentbased models, physiological-based models, and energetics-based models-are reviewed and the different mathematical formalisms compared. Additionally, the residue-based toxicity approach is reviewed. Coupling toxicokinetic models with tissue concentrations at which toxicity occurs offers a direct link between exposure and hazard. Basing hazard on tissue rather than environmental concentrations avoids the errors associated with accommodating multiple sources, pulsed exposures, and non-steady-state accumulation. Keywords-Kinetic models Bioaccumulation Tissue residue effects Sediment contaminationHazard assessment INTRODUCTlONAssessment and prediction of toxicant effects on aquatic organisms require evaluation of the extent of organism exposure. Exposure assessment establishes the relationship between environmental toxicant concentrations and organism accumulation while accounting for environmental and biological factors that modify exposure. If the relationships between the amount of toxicant accumulated and the resulting effects are known, then the hazard for a particular exposure regime can be established.Aquatic exposure assessments and predictions have employed mainly steady-state and equilibriumpartitioning models. Early efforts, using simple kinetic models, were designed to provide estimates of steady-state accumulation from water exposures [1,2]. These steady-state estimates were then utilized in hazard assessments based on thermodynamic limits (chemical equilibrium). Such models have been employed with good success for evaluation of general conditions, describing toxicant distribution among ecosystem components and *To whom correspondence may be addressed.identifying components dominating toxicant mass balance. This approach has been best refined using the fugacity concept and applied to describe the importance of sediment as a toxicant source [3] and toxicant distributions within ecosystems [4,5].Although there is a continued focus on equilibrium-partitioning models within regulatory agencies, it is clear that the environment is complex and variable. Therefore, to obtain more accurate predictions and assessments, kinetic models are needed to predict non-steady-state, nonequilibrium accumulation from temporally and spatially varying exposures when the simplifying assumptions of the equilibrium-partitioning models are inappropriate, for example, when multiple sources contribute significantly to accumulation.Kinetic models have been used successfully in pharmacology for decades. Such models permit prediction of the onset of drug action and allow the monitoring of drug clearance and termination of effects. Further, these kinetic models describe changes in tissue concentrations resulting from absorption, distribution, metabolism, a...

Pyrethroid pesticides have replaced organophosphates for many urban applications, including structural pest control, landscape maintenance, and residential home and garden use. This study was intended to determine if pyrethroids are detectable and widespread in diverse urban systems and if concentrations are high enough to cause associated aquatic toxicity. Urban creeks in California and Tennessee were tested on up to four occasions for pesticide residues in sediments, and aquatic toxicity was determined by acute toxicity tests using the amphipod, Hyalella azteca. In California, 12 of the 15 creeks tested were toxic on at least one sampling occasion, and sediment pyrethroid concentrations were sufficient to explain the observed toxicity in most cases. The pyrethroid bifenthrin, due to its high concentrations and relative toxicity as compared to other pyrethroids, was likely responsible for the majority of the toxicity at most sites. Cypermethrin, cyfluthrin, deltamethrin, and lambda-cyhalothrin also contributed to toxicity at some locations. The source of cypermethrin and deltamethrin was probably almost entirely structural pest control by professional applicators. Bifenthrin, cyfluthrin, and lambda-cyhalothrin may have originated either from professional structural pest control or from lawn and garden care by homeowners. None of the sediments collected from the 12 Tennessee creeks were toxic, and pyrethroids were rarely detectable. Regional differences between Tennessee and California are possibly attributable to climate, differences in types of residential development, and pesticide use practices.