| The effects of vole (Microtus) herbivory on nitrogen dynamics were assessed using small-mammal exclosures in natural riparian meadows in Yellowstone National Park, USA. Voles are characterized by large fluctuations in population density and may have both short-term, pulsed effects and longer-term, chronic effects on nitrogen cycling. In the first year of the study, a peak in vole densities caused direct carbon and nitrogen inputs into the soil, quickly affecting nitrogen immobilization and mineralization rates. Immediately following this peak vole episode, more standing herbaceous litter accumulated in exclosures than in unexclosed plots. In addition, litter inside exclosures had a higher C:N ratio than litter from control plots. These direct effects of vole herbivory persisted for several years after the vole peak at the Crystal Bench site. But direct effects were eventually overshadowed by indirect effects caused by changes in the plant community at Blacktail where legumes were more common. At Blacktail, several years of selective foraging by voles on legumes and other high-nitrogen plants kept the litter quality outside exclosures low, while increases in legume biomass inside exclosures decreased the C:N ratio of exclosure plant litter. Because plant litter is a major source of soil organic matter, I expected the changes in plant litter quality to affect nitrogen mineralization rates. However, I did not detect any effects of three years of vole removal on nitrogen mineralization in the field, possibly because vole populations after the first year were relatively low and vole effects were more chronic than acute.; To assess vole effects on soil organic matter quality, I measured potentially mineralizable nitrogen (N{dollar}sb0{dollar}) by incubating soils in the laboratory. The pattern of nitrogen mineralized over time was best fit with a two-pool model, with a fast pool mineralizing at a decreasing (nonlinear) rate and a slow pool mineralizing at a low but constant rate. The proportion of the total soil nitrogen that was potentially mineralizable was similar (4-7%) for both sites, but total mineralization was greater in Crystal soils. At Crystal, N{dollar}sb0{dollar} was greater in soil from control plots than in soil from exclosures. At Blacktail, although the absolute pool of potentially mineralizable nitrogen did not differ between control and exclosure soils, the size of N{dollar}sb0{dollar} relative to the total pool of soil nitrogen was greater in exclosure than control soil. |
