Effects of deer browsing on soil biogeochemical processes along a nitrogen gradient

Background/Question/Methods: Deer browsing impacts on temperate forest soils are mediated by herbivore selectivity and plant recovery after browsing, which is likely influenced by soil nutrient availability. The relative importance of these factors, however, remains unknown. When soil nutrients are abundant, plants respond to herbivore damage by fresh regrowth. When soil nutrients are scarce, plants increase their defense mechanisms. On the one hand, fresh regrowth can lead to more decomposable plant litter. Alternatively, increased defenses can slow litter decomposition and lower nutrient cycling rates. Although increased deer densities have driven changes in plant community composition and forest successional pathways, few studies address deer browsing effects on soil biogeochemistry. We examined how increased deer browsing in temperate forests affected soil nitrogen (N) mineralization and carbon (C) respiration across a soil N gradient. Using lab incubations, we measured N mineralization and CO₂respiration in soils collected from inside and outside 17-year-old deer exclosures at 19 sites in southeast Michigan. Differences in laboratory N mineralization (Δ N min) and turnover (ΔN turnover) rates between soils collected from inside and outside exclosures were analyzed as functions of the soil N gradient.

Results/Conclusions:  The magnitude and direction of deer browse effects on soil N mineralization and CO₂ respiration differed significantly across a soil N gradient. ΔN min and ΔN turnover values increased with soil N availability (p < 0.001, R²=0.80 and R²=0.70 respectively). This means that deer browsing decreased N cycling only at sites where N mineralization was high in the absence of deer. At sites with low N mineralization, however, no differences in N cycling were detected between soils from inside and outside deer exclosures. CO₂ respiration also decreased in the presence of deer (p=0.04, R²=0.074), again only at sites with high N mineralization. Importantly, these differences in impacts could also depend on browsing intensity. Preliminary surveys showed that N mineralization rates correlate negatively with browse intensity (p=0.029, R²=0.26). Together, these results suggest that forms of N become more resistant to microbial breakdown in the presence of deer, at sites with high N mineralization. We predict that this decrease in N turnover is mediated by a change in litter quality, not quantity. Therefore, an exploration of the chemical landscape of plant communities, inside and outside deer exclosures, is necessary to better predict deer browsing effects on temperate forest soils across nutrient gradients.