INSECTICIDE EFFECTS ON SMALL MAMMALS: INFLUENCE OF VEGETATION STRUCTURE AND DIET

Abstract

We tested whether differences in the diets of nontarget organisms or in vegetation structure could result in adverse ecological effects not predicted by the Quotient Method, a laboratory-based risk assessment methodology used by the Environmental Protection Agency to evaluate pesticides for registration and use. We established populations of herbivorous gray-tailed voles (Microtus canicaudus) and omnivorous deer mice (Peromyscus maniculatus) in 24 0.2-ha enclosures planted with alfalfa (Medicago sativa) and monitored the populations by live trapping. Alfalfa in twelve enclosures was mowed on 22 June to reduce vegetation height, and 3 wk later, we applied azinphos-methyl at 0, 0.88, and 3.61 kg/ha. We compared predictions of risk with observed effects on the small mammals. Treatment with azinphos-methyl at 3.61 kg/ha reduced population density and growth, survival, recruitment, and body growth of voles in both mowed and unmowed enclosures. Survival of female voles exposed to 3.61 kg/ha was lower in mowed than in unmowed enclosures, and the 0.88 kg/ha treatment affected body growth of male voles only in mowed enclosures. Vole densities in 3.61 kg/ha enclosures remained depressed for ≥6 wk after spraying. Azinphos-methyl did not detectably affect reproductive activity of female voles. Precipitation after spraying may have increased exposure of the mammals to azinphos-methyl, resulting in greater effects on voles than reported in a previous, similar experiment. Deer mouse densities in mowed enclosures receiving 3.61 kg/ha decreased 47% within 5 d after spraying. No other adverse insecticide effects on deer mice were significant, but confidence intervals for such effects were wide. Analysis of deer mouse feces indicated that consumption of arthropods just after spraying was greater in insecticide-treated enclosures than in controls. In general, the Quotient Method adequately predicted effects on the small mammals, but its performance may be affected by vegetation structure and precipitation.