Bison grazing intensity predicts the abundance of the dung fungus proxy Sporormiella at Konza Prairie: Implications for the Holocene paleoecology of the Great Plains

Background/Question/Methods

The role of mammalian herbivores as ecological keystones is typically quantified through experiments where herbivores are added, removed, or excluded, either due to experimental manipulations, local extirpations, or as the consequence of land-use or management decisions. Where records of herbivore biomass or presence is not available, however (i.e., in the paleoecological record), such study has been limited.  Spores from the coprophilous fungus Sporormiella are increasingly used as biomarkers for the presence of browsing and grazing herbivores in historic and prehistoric sediment records, but there remains a critical need to better understand the ecological factors governing the production, transportation, deposition, and taphonomy of spores. Konza Prairie LTER represents a unique opportunity to test the relationship between Sporormiella and megaherbivory, as it is essentially a single-megaherbivore system with annual records of bison biomass and landscape usage. Using modified Tauber traps, we collected pollen and spores from locations inside and outside the bison enclosure at Konza Prairie during the 2009 and 2010 flowering seasons. Bison grazing intensity (kg/m²/year) was calculated within concentric rings at increasing distances around each trap, using locations of radio-collared bison matriarchs as a representation of a fraction of the total herd biomass.

Results/Conclusions

Sporormiella showed a significant qualitative (presence-absence) and quantitative (grazing intensity) response to bison at Konza (for all results, p = <0.05). We found significantly higher amounts of Sporormiella in both relative (percent) and absolute (concentration) abundances from traps within the enclosure than in traps outside the enclosure, which supports previous modern process studies of Sporormiella but in a setting where megaherbivore densities and locations are known. Within the enclosure, grazing intensity was a significant predictor of Sporormiella abundance for both relative and absolute spore abundances. Grazing area is the strongest predictor of Sporormiella abundances at <100 m ring distance from the traps, suggesting that the overall source area of Sporormiella is very local. Additionally, grazed traps had significantly higher percentages of Ambrosia and lower percentages of Poaceae pollen than un-grazed traps, which is consistent with modern research on the effects of bison grazing on vegetation at Konza and elsewhere in the Great Plains. The results for each analysis were sometimes different for percent versus concentration data, which indicates the importance of both to interpreting Sporormiella records. We discuss the implications of these results to the Holocene paleoecology of the Great Plains, including testable hypotheses of the potential relationships between bison, drought, vegetation composition, and fire.