Proximate and long-term effects of agricultural pesticide runoff on native fish (California)

One of the primary challenges of ecotoxicology is to characterize, link, and predict effects of toxic stressors from molecular to organismal to population levels. The overarching goal of this dissertation was to evaluate consequences of dormant season pesticide runoff in the Central Valley of California on native fish at both the organismal and population levels. Toward this goal, the first major objective was to characterize proximate effects of pesticide runoff on a native fish (Sacramento sucker, Catostomus occidentalis ), by combined field and laboratory studies, and by examining multiple biomarker responses coupled with extensive environmental chemistry. Data indicated that dormant season runoff events were strongly associated with induction of DNA strand breakage in C. occidentalis exposed in both the field and laboratory. Genotoxicity of runoff was also supported by significant mutagenic responses of Ames Salmonella typhimurium tester strains. Given evidence of organism-level toxicity induced by pesticide runoff, the second major objective was to test whether long-term effects of such exposures were evident in genetic structures of C. occidentalis populations across Central Valley watersheds. Importantly, field sampling was designed to rigorously test for evidence of both contaminant stress and biogeography as forces for shaping genetic patterning among populations. A secondary goal within this objective was to evaluate the relative consistency and merits of two different marker systems (AFLPs and microsatellites) for the detection fine-scale structure, and the characterization of genetic patterning. Results indicated that both AFLPs and microsatellites were sensitive for detecting structure among C. occidentalis populations, and estimates of population parameters were highly concordant. Expectations derived from biogeographic hypotheses efficiently described patterns of variation among groups. In contrast, patterns of genetic variability indicating the influence of long-term pesticide contamination stress were weak or absent. Conclusions were that historical pesticide contamination stress was of insufficient duration or intensity, or that demographic forces such as migration were of such prevailing influence, that toxic stress evidenced at the organismal level for C. occidentalis had not propagated to the population level in habitats across the Central Valley. However, the observation of genotoxicity raises concern that new-generation pesticides, or other runoff-associated contaminants, may be of ecotoxicological consequence, and merits further investigation.