The bioturbation activity of ants in agricultural and naturally vegetated habitats in semiarid environments

Abstract

The effect of ant activity on bioturbation rates was assessed in the central wheatbelt of Western Australia in naturally vegetated and farmland sites, by bimonthly measurements of soil fauna activity and periodic measurements of ant nest dimensions. The percentage soil surface affected by soil fauna (%SSASF) was on average 0.5%, with the lowest activity in cropped areas and the highest %SSASF recorded in pasture sites (>2%). Overall %SSASF was greater on yellow sand in comparison with grey sandy loam. The estimated bioturbation rates of ant communities in naturally vegetated and farmland habitats on both soils ranged from 10 to 37 g m-2 yr-1. On average, the amount of soil excavated by ants to build a nest was 20 to 25 g. Aphaenogaster sp. D49 was identified as a significant contributor to soil turnover based on the frequency of excavation, nest density and mass of soil removed to build the nest (134 g). Aphaenogaster sp. D49 brought to the soil surface 465 g m-2 yr-1 on grey sandy loam and 223 g m-2 yr-1 on yellow sand. The activity period for the majority of ant nests was around 3 months, but some species such as Rhytidoponera spp. and Meranoplus spp. had nests active for up to 2 years. While other ant species such as Iridomyrmex sp. D13 have a high turnover of nests (<3 months). There was no clear seasonal trend in the number of active nest openings, although they appeared to be greatest in May, and lowest in January. Ant nest density was higher in yellow sand than in grey sandy loam, regardless of land use. Particle size distribution of ant spoils (soil brought to the surface during nest construction) and ant-affected soil at 26 cm was compared to the control soil at comparable depths, and found to be similar. Three instances where the ant spoil had a higher proportion of clay-sized particles in comparison with the control soil, and this was attributed to ants excavating into the clay subsoil or being size-selective in the material they excavated. The control soil had a texture-contrast between 2 and 26 cm, while all the ant nests examined had similiar texture between these two depths. This supports the hypothesis that ant activity homogenizes the soil texture profile.