Ecosystem engineering of harvester ants: Effects on vegetation across spatial scales in a sagebrush-steppe ecosystem

Background/Question/Methods

Owyhee harvest ants (Pogonomyrmex salinus) are influential in many arid ecosystems because they distribute and consume seeds, and create underground tunnels that redistribute soil particles and nutrients.  However, the effect of harvester ants on cover and composition of plants varies across spatial scale and ecosystem. The main objectives of our research were to assess (1) how density and size of harvester ant nests varied across three vegetation types (burned-untreated, burned-treated, and unburned) within a sagebrush-steppe community, and (2) how vegetation cover (by species and total) changed as a function of distance from harvester ant nests.  To examine how harvester ant nest size and density varied across the three sites with different vegetation types, we exhaustively surveyed for active harvester ant nests at a 1 ha site within each vegetation community, and then used nest counts and measurements to calculate nest size and density.  We examined how vegetation cover changed with distance from the edge of ant nests by taking photos of 0.25 x 0.5-m quadrats at 0 m, 1.5 m, and 3 m from the nest edge (0 m, 1.5 m, and 3.0 m) and measuring percent cover of each species using SamplePoint software.  

Results/Conclusions

Nest size was negatively correlated with harvester ant nest density across the three vegetation types.  The influence of Owyhee harvester ants on vegetation was larger at the edge of ant nests, but the relationship was not consistent among plant species.  Percent cover was positively correlated with distance to harvester ant nests for plant species that were considered undesirable food sources for the harvester ants and were densely distributed.  However, percent cover was negatively correlated with distance to harvester ant nests for patchily distributed and desirable plant species.  For some plant species, there was no change in cover associated with distance from harvester ant nests. Total vegetation cover (across all species) was associated with distance from ant nests at the unburned site but not the other two sites.  The dominant species at the unburned site was a densely distributed shrub (Krascheninnivokia lanata) that was actively defoliated by harvester ants.  Overall, these results suggest that Owyhee harvest ants facilitate vegetation heterogeneity through the mechanism of plant clearing, and that their influence may be associated with plant nutrition and spatial distribution.  This information may inform future management and plant restoration efforts by directly considering the “islands of influence” associated with harvester ant engineering.