The residual effects of microtine herbivory and fire on an Illinois tall grass prairie

Includes bibliographical references.Includes illustrations.This study was undertaken to assess the effects of fire and residual effects of small mammal herbivory on an Illinois tall grass prairie during the 1979 growing season. Randomly located 2.5 m X 2.5 m exclosures and control sites were established on each of four previously investigated plots. Two of the plots studied during 1977-78 had been fenced and the principal small mammal herbivore, Mlcrotus pennsylvanicus, removed by trapping (previously ungrazed); the remaining two plots served as controls and were left open to voles (previously grazed). Prior to the initiation of this project in 1979* the study area was burned in April 1979 as part of the management program at Goose Lake Prairie State Park. Live trapping conducted from May to September revealed the voles did not establish residence on the burned area. This event prevented the testing of the possible'additive effects of herbivory but allowed the evaluation of the effects of small exclosures on tall grass prairie vegetation. Biweekly vegetation samples were taken using a point-frequency technique with a two meter point-frame having ten pins spaced 20 cm apart. Total number of contacts, total plant height, and average plant height were determined for each species from placements of the frame centered over the sampling point. Dry weight biomass of vegetation from 0.2-m by 2.0-m clip quadrats taken in July and September were regressed against these data. The number of contacts was found to be most closely correlated with the dry weight biomass. Equations derived from this relationship were used with the biweekly contact data to estimate the dry weight biomass trends. The results were analyzed with a two-way analysis of variance to examine the effects of residual herbivory and the exclosure effects. Nonparametric tests were utilized to determine the vegetation response to fire. The results indicated that the most abundant grass, Andropogon gerardl, did not show significant residual effects of herbivory, although greater biomass on previously ungrazed plots was found. A second grass, Sorghastrum nutans, demonstrated significant residual effects of herbivory with greater biomass on ungrazed plots. Andropogon biomass increased significantly on burned 1979 plots as did Sorghastrum biomass. However, increases in both fire and herbivory effects for Sorghastrum precluded a clear interpretation. Despite the greater biomass on the burned 1979 plots, greater peak biomass for both grass species was reached on the unburned plots in the previous year. Four Solldago spp., a forb, did not show a residual effect of herbivory, although there was a trend of greater biomass on previously grazed plots. A majority of the differences between grazed and ungrazed plots for another forb, Eupatorlum serotinum. were significant with greater biomass on previously grazed plots. The response of these forbs to fire appears to depend on previous exposure to herbivory, with previously grazed plots demonstrating more biomass. This differential response of the grasses and forbs to burning is mediated through differences in litter buildup and physiology of the plant species involved as well as climatic conditions. It was also concluded that the small exclosures of the size used in this study did not affect the vegetation significantly.M.S. (Master of Science