Is Water a Relevant Sampling Medium for Toxic Chemicals? An Alternative Environmental Sensing Strategy

Abstract

AbstractUsing an uncomplicated 960 km reach of the North Saskatchewan River, we examine pathways and distribution of anthropogenic contaminants in water and in suspended and bottom sediments. Synoptic sampling during stable low flow allows maximum inference of point sources and, therefore, of chemical complexity. After chemical extraction of contaminants into two fractions for aqueous samples, and five fractions for sediment samples, we are able to assess the toxicity of the fractions using two independent bioassays—the Panagrellus redivivus nematode assay and a nonstandard modification of the Ames bacterial test. Our observations include (i) little comparability between toxic response of whole water samples and suspended sediments contained in the whole water samples; (ii) a much greater toxic response and larger number of priority chemicals associated with suspended sediments than in water samples; (iii) observed priority chemicals can exist in water or on sediments, yet produce no toxic response in one or both of the bioassay procedures; and (iv) significant toxicity is observed in the absence of priority chemicals. Results lead to questioning of efficacy of a standard menu of priority chemicals applied to water samples for ambient (i.e., not end‐of‐pipe) aquatic quality sensing (monitoring) purposes. We conclude that water may be an inappropriate medium upon which to base toxic chemical criteria for toxic chemical sensing purposes in aquatic systems. The menu approach to toxic chemical sensing is neither cost nor information efficient. We propose an alternative sensing protocol using a coupled chemical‐bioassay protocol that reduces the number of chemicals analyzed, produces an objective biological index of relative toxic response, provides a backup for presence of toxic chemicals not included in the chemical screen, and conveys meaningful ecological information upon which agencies can make program decisions.