Wood ants (Formica spp.) as ecosystem engineers and their impact on the soil animal community

This thesis describes the impact of wood ants on the forest soil biology. The first question addressed is if wood ants affect the ecosystem by concentrating coniferous litter to their mounds. The results suggest that the chemical properties of Formica ant nests differ from the surrounding soil, and that the nests are likely to increase the spatial heterogeneity of the forest floor. Wood ants seem to maintain a specific environment in their nests by collecting conifer resin. Resin acts as a carbon source that increases C mineralisation and decreases net N mineralisation. The dry conditions in the ant nests and the addition of resin to the nest material might also determine the composition of soil fauna living in the nests. Collembola seemed to by negatively affected by resin. Oribatids seemed to be unaffected by resin and favoured by low moisture in the nests. The second question was if wood ants affect the soil food web structure by preying upon or interfering with the soil fauna. The results showed that ants can forage on the forest floor during a large part of their active season. This means that ants have a potential to affect the soil fauna in coniferous forests. However, I found little evidence supporting the hypothesis that wood ants have a large impact on the abundance or composition of the soil fauna. In one experiment, the foraging behaviour of wood ants was manipulated by excluding ants from their main protein resources in the tree canopy, with the intention to increase ant activity on the forest floor. When excluded from trees, ants changed their foraging behaviour by searching other trees further away from the nests rather than searching more intensively for prey on the forest floor. In the treated plots (ants excluded from the trees), only Linyphiidae spiders were negatively affected by the ants. This experiment as well as the results of two other studies, indicate that the hypothesis that wood ants are key-stone predators on soil fauna might be rejected. However, to test this hypothesis, more longterm experiments carried out at a large spatial scale are needed.