Evaluating restoration potential of transferred topsoil

QUESTIONS: Translocation of topsoil and its seed bank for ecological restoration is increasingly popular. How representative is the soil seed bank of the extant vegetation at the source site? What influence does the transfer process have on germinant density, species and plant functional type composition? Does smoke and heat treatment of transferred topsoil enhance germination and potential for restoration success? LOCATION: Banksia woodland of the Swan Coastal Plain, Western Australia. METHODS: To assess the efficacy of topsoil transfer for return of native plant species, we measured topsoil seed bank characterisitcs in situ proir to vegetation clearing, and immediately after transfer of the topsoil to an adjacent, degraded mediterranean‐type woodland in southwest Australia. Glasshouse germination of topsoil samples from 24 pre‐ and 24 post‐transfer plots was used to quantify the effects of transfer and soil depth on germinant density, species richness, plant functional types and seed bank similarity to in situ vegetation. Application of germination cues (heat + smoke) was used to explore the impact of topsoil transfer on seed germination and emergence. RESULTS: Topsoil transfer significantly reduced germinant densities (pre‐transfer 1692–4239 germinants·m⁻², post‐transfer 795–1016 germinants·m⁻²; t = 6.7, P < 0.001) and shifted community structure (MRPP: A = 0.13, P < 0.001), including a reduction of woody species density by 81%. For the majority of functional types, heat and smoke failed to stimulate additional germination post‐transfer, suggesting soil transfer simulated the effect of fire‐related germination cues. CONCLUSION: Although topsoil transfer translocated many viable native seeds, potential restoration success was hindered by reduced germinant densities. This was mostly attributable to a dilution effect associated with mixing of transferred topsoil, so that many seeds were buried too deep to emerge. However, total reductions were greater than expected based on dilution alone, suggesting some seed mortality during the transfer process. Transfer shifted composition towards dominance by annual species, suggesting the need for topsoil transfer to be supplemented by other restoration techniques.