North American tallgrass prairie is characterized by high plant species diversity and substantial belowground productivity. Maintenance of these tallgrass prairie services is driven in part by the disturbances of fire and large herbivore grazing, and the interaction between the two. Less is known about prairie soil microbial characteristics and responses to fire and grazing, despite the importance of microbial activity in mediating carbon (C) and nitrogen (N) fluxes from the ecosystem. Burning removes N through volatilization, therefore we expected to measure lower microbial N cycling rates in burned watershed soils. We expected grazing to increase N availability thus enhance N cycling rates, and to decrease microbial C cycling rates due to lower belowground C allocation by plants in the less N-limited soil. To address these predictions, we quantified aspects of plant and soil microbial C and N cycling in watershed scale long-term burning (annual vs. 20 year) and bison grazing (grazed vs. ungrazed) experimental treatments in a 2-way factorial design. In each watershed, we installed resin bags along replicate linear transects to measure N availability, and collected corresponding samples to measure litter stock, belowground net primary production, soil available C, microbial respiration, nitrification and denitrification potentials, and soil edaphic factors.
Results/Conclusions
We performed 2-way ANOVA with blocking to evaluate treatment effects on each of the measured variables. Burning had a significant (P<0.05) direct or interactive effect on aboveground litter inputs, root production, soil microbial respiration, 30-day soil respiration, and total soil organic matter, with responses indicating higher C inputs and losses in the unburned treatment, and lowest C inputs and losses in the grazed and annually burned treatment. N availability was higher in grazed soils (P<0.005), as expected, and nitrification potential was higher in both grazed and unburned watersheds (graze: P=0.0086, burn: P=0.0028). Surprisingly, denitrification potential was higher in unburned soils and was not influenced by grazing (burn: P=0.004). Although soil N availability was higher in the presence of grazing animals, our data suggest that nitrification potential was influenced by both N availability and burning frequency, and that denitrification potential may be limited by low C availability even in relatively high-N soils. Therefore, the cessation of fire in tallgrass prairie could still result in significant N losses by denitrification rather than volatilization by fire. These results show that land management influences soil microbial activity and emphasizes that plant and microbially mediated C and N cycling are linked.