Foraging activity by an ecosystem engineer, the superb lyrebird, 'farms' its invertebrate prey.

Ecosystem engineers shape ecological communities worldwide by modifying the habitats of other taxa. Engineering activities may generate feedbacks that benefit the engineers themselves, though such evidence is sparse. The superb lyrebird (Menura novaehollandiae), a ground-dwelling species of moist eucalypt forests in south-eastern Australia, engineers habitats by extensively modifying litter and soil layers (~155 tonnes/ha/year in suitable habitat) whilst foraging for invertebrates. We examined whether this engineering activity by lyrebirds serves to promote a 'farming' effect on their prey, by increasing the biomass and taxonomic richness of the invertebrates on which they feed and altering the composition of invertebrate communities. We experimentally isolated the effects of engineering (soil/litter modification) from predation effects (on invertebrates) by establishing a manipulative experiment with three treatments: (1) lyrebird-free, fenced exclosures ('control'; no engineering or predation); (2) fenced exclosures with raking ('raked'; engineering effects only); and (3) unfenced plots accessible to lyrebirds ('lyrebird'; engineering and predation effects). Biomass and taxonomic richness of invertebrates were higher in 'raked' treatments compared with 'control' treatments, due to the engineering effects of raking. Biomass and richness were lower in 'lyrebird' treatments, due to predation effects by lyrebirds. Whilst the composition of invertebrate communities showed no significant change due to engineering, it was significantly influenced by predation. Lyrebird foraging actions that mix, turn over, and aerate litter and soil layers create a positive feedback loop that actively replenishes the biomass and diversity of their invertebrate prey, compensating for offtake by predation. These results provide evidence that lyrebirds 'farm' their prey resource. This process of farming by the superb lyrebird operates at a spatial scale that is unprecedented in non-human vertebrates, extending across millions of ha in moist forest ecosystems. The enhanced diversity and biomass of invertebrates generated via lyrebird foraging has potentially far-reaching implications for the structure and function of forest ecosystems; and for the temporal dynamics of forests and their response to disturbance processes, such as wildfire. Reported examples of such positive feedback loops arising from ecosystem engineering activities are scarce, but may be more common than thought and have extensive impacts on ecosystem function.