The Effects of an Acute Insecticide Stress on a Semi‐Enclosed Grassland Ecosystem

Abstract

The effects of a carbamate insecticide, Sevin, on plant, arthropod, and mammal components within a grain crop grassland ecosystem were investigated. A single application of 2 lb. of the insecticide was applied in July to 1 of 2 comparable and adjacent once—acre fenced enclosures, each planted in a crop of millet (Panicum ramosum) and stocked with three species of small mammals, the cotton rat (Sigmodon hispidus), the house mouse (Mus musculus) and the old—field mouse (Peromyscus polionotus). Intensive and frequent sampling of plants, arthropods, and mammals on both control and treated enclosures provided the data for evaluating the effects of the insecticide on productivity, density, diversity, and equitability of the three components. Sevin residues on plants decreased rapidly from 35 ppm on the first day following spraying to 0.37 ppm on the 16th day. No effect of the insecticide could be detected on producer standing crop or net community primary production which averaged for the two areas 567 g dry wt/m2 for season or 3.96 g/day. A highly significant decrease in litter decomposition in the treated area was measured 3 weeks after spraying, presumed to be the result of a reduction in microarthropods and other decomposers. The total biomass and numbers of arthropods were reduced more than 95% in the treated area and remained well below the control area for 5 weeks; after 7 weeks total biomass but not total numbers returned to the control level. Photophagous insects (Homoptera and phytophagous Hemiptera were dominant at the time of spraying) were more severely affected than predaceous insects and spiders; density of the latter returned to control levels in 3 weeks. Species—numbers diversity in terms of S—1/1nN also was markedly reduced in all orders of insects immediately after treatment, but with the exception of the Hemiptera and Hymenoptera, returned to control levels within 1—2 weeks. Diversity in spiders was not affected by the treatment, even though numbers were reduced. In terms of trophic groupings, species diversity of phytophagous insects, although more severely reduced, "recovered" more quickly than did diversity in predaceous insects. The equitability index of Lloyd and Ghelardi (1964), in contrast to species/numbers diversity, tended to rise 1 to 3 weeks after treatment, especially in the phytophagous insect group. In terms of the seasonal succession, equitability index values for the predator populations increased to about 1.0 while that of herbivores remained at lower levels of 0.5 to 0.6 throughout the study on the control area, suggesting that predators tended to be food limited and to occupy non—overlapping niches. Cotton rat reproduction was delayed with ultimate density reduced in the treated grid; laboratory tests indicated this was a probable direct effect of the insecticide. However the total mammal population, which grew in sigmoid fashion to an asymptotic level of about 180/acre in both grids, was not affected by the treatment because there was a compensatory increase in the house mouse population in the treated grid. In summary, although the insecticide remained toxic in the environment for only a few days, long—term side effects on litter decomposition, arthropod density and diversity, and mammalian reproduction were demonstrated. It is hoped that this study will provide a model for an ecosystem approach to the testing of pesticides.