The extent and pattern of Eucalyptus regeneration in an agricultural landscape

Anthony Weinberg, Philip Gibbons, Sue Briggs, Stephen Bonser

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

The process of natural regeneration is critical for sustaining remnant native vegetation and the ecosystem services it supports. We quantified the extent and pattern of Eucalyptus regeneration within remnant veg- etation in a fragmented agricultural landscape in south-eastern Australia. Eucalyptus regeneration was absent at 42% of sites. Using an information-theoretic approach, we explored 13 possible models of Euca- lyptus regeneration across multiple scales. The explanatory variables in the four models with empirical support (and their summed Akaike weights) were: grazing intensity (1.0), native ground cover (0.99), remnant area (0.83), tenure (0.67), canopy cover (0.21) and vegetation type (0.11). Averaging across these four models we predicted that the probability of Eucalyptus regeneration was highest (0.95) in relatively unmodified remnant native vegetation, that is, remnant vegetation on public land where grazing was light and the understorey was dominated by native plants. In contrast, the predicted probability of Euca- lyptus regeneration was lowest (0.12) in small remnants on private land where grazing was heavy. Our results suggest that a large proportion of all remnant native vegetation in this landscape will disappear under existing land management and farming practices. Reducing grazing pressure within intensively grazed remnants appears to be the single most effective management intervention that will mitigate this threat. This will require a shift in conservation priorities away from large, intact remnants where regen- eration does not appear to be affected, to poorer quality remnantsâ¿¿often small remnants or scattered treesâ¿¿where regeneration is typically absent.
Original languageEnglish
Pages (from-to)227-233
Number of pages7
JournalBiological Conservation
Volume144
Issue number1
DOIs
Publication statusPublished - 2011
Externally publishedYes

    Fingerprint

Cite this