The Response of Soil CO2 Fluxes to Progressively Excluding Vertebrate and Invertebrate Herbivores Depends on Ecosystem Type

Grasslands support large populations of herbivores and store up to 30% of the world’s soil carbon (C). Thus, herbivores likely play an important role in the global C cycle. However, most studies on how herbivory impacts the largest source of C released from grassland soils—soil carbon dioxide (CO₂) emissions—only considered the role of large ungulates. This ignores all other vertebrate and invertebrate herbivores and their collective effects on ecosystem properties. We progressively excluded large, medium, and small vertebrates and invertebrates from two subalpine grasslands (productive, heavily grazed short-grass; less productive, lightly grazed tall-grass) using size-selective fences, assessed the impact on soil CO₂ emissions and related biotic and abiotic variables. Exclusion resulted in significant changes in soil CO₂ emissions in both vegetation types. Short-grass soil CO₂ emissions progressively increased when large and medium mammals were excluded. However, no difference was detected among plots were all or no herbivores grazed. In contrast, tall-grass soil CO₂ emissions were not affected by mammal exclusion, but excluding all herbivores lead to reduced emissions. Soil micro-climatic parameters best predicted the patterns of soil CO₂ emissions in short-grass vegetation, whereas root biomass was the best predictor of CO₂ release in tall-grass vegetation. Our results showed that diverse herbivore communities affect soil respiration differently than assumed from previous studies that only excluded large ungulates. Such information is important if we are to understand how changes in herbivore species composition—as could happen through altered management practices, extinction or invasion—impact grassland C storage and release.