Herbivores play a key role in the turnover, gains and losses of nutrients in ecosystems. Because nutrients are often limiting, herbivores can influence plant growth and chemistry, and their own resource supply. Herbivores typically affect their environment in three ways: 1) they clip aboveground biomass, 2) they trample soil, and 3) they return nutrients to the system via waste deposition. The relative importance of these pathways is often unexplored because it requires conducting experiments that isolate these mechanisms. Millions of geese migrate in the spring to sub-arctic coastal wetlands where they play a key role in determining the amount and quality of forage in this habitat. We conducted two field experiments on Carex subspathacea grazing lawns in western Alaska to investigate how these individual processes (grazing, trampling, and fecal addition) influence foliage quality (C:N) and biogeochemical cycling, including greenhouse gas flux. In experiment 1, we isolated goose herbivore effects with five treatments: grazing only, trampling only, fecal addition only, all three treatments combined, and no herbivory. In experiment 2, we manipulated goose fecal material such that naturally grazed plots received no fecal matter, ambient amounts of fecal matter, or double ambient amounts of fecal matter.
None of the herbivore effects were individually responsible for maintaining high quality forage or affecting nutrient cycling; an interaction among the three was crucial in regulating plant nutrient availability and greenhouse gas exchange. In experiment 1, we found that C:N ratios of foliage in all-treatment plots were lower than in no herbivory plots, but the single-process treatments did not substantially change C:N. We also found soil respiration was higher in no herbivory plots compared to all other treatments. In experiment 2, we found that treatments did not affect C:N. However, plots with feces removal had a mean net ecosystem exchange of 0.9 µmol m-2 s-1 indicating net carbon loss from the system, while plots with ambient feces and feces addition were net carbon sinks of -0.35 and -0.32 µmol m-2 s-1, respectively. Similarly, plots with fecal removal had higher mean methane flux, 42.1 umol m-2 s-1, than plots with fecal addition, 15.7 umol m-2 s-1. We conclude that no single herbivore effect is responsible for increasing the quality of forage, but rather that the three processes interact to maintain high nutrient C. subspathacea lawn and sequester, rather than release, carbon as a greenhouse gas.