New Zealand Journal of Ecology (2015) 39(1): 71- 78

Creek habitats as sources for the spread of an invasive herb in a New Zealand mountain landscape

Research Article
Alice L. Miller 1,2*
Susan K. Wiser 3
Jon J. Sullivan 1,4
Richard P. Duncan 1,5
  1. BioProtection Research Centre, PO Box 84, Lincoln University, Lincoln 7647, New Zealand
  2. Current address: Pyramid Botanical Consultants, 3001 Carriage Drive, Estes Park, CO 80517, USA
  3. Landcare Research, PO Box 69040, Lincoln 7640, New Zealand
  4. Department of Ecology, PO Box 84, Lincoln University, Lincoln 7647, New Zealand
  5. Institute for Applied Ecology, University of Canberra, ACT 2601, Australia
*  Corresponding author

Landscapes typically comprise various habitats that differ in their susceptibility to invasion by exotic species. Highly invasible habitats such as riparian corridors can act as a conduit for rapid movement across the landscape and as a propagule source to facilitate spread into adjacent, less invasible habitats. If this is so, there should be a decline in invader frequency and/or abundance with distance away from the source; the local abundance of invasive species in the source habitat should be positively correlated with local abundance in the adjacent habitat; and, having accounted for variation in local source population size, the slope of the decline in invader abundance with distance away from the source habitat should reflect the ease with which adjacent habitat can be invaded. Here, we test these predictions to assess whether creek habitats function as a source to facilitate the spread of the invasive plant Hieracium lepidulum Stenstr. (Asteraceae) in upland catchments of the South Island, New Zealand, by surveying H. lepidulum abundance in creek margin and adjacent beech forest and subalpine habitat in 17 creek catchments. Our results imply that propagule pressure from populations in creek margins and other disturbed areas is driving catchment-wide H. lepidulum invasion, but forest and subalpine habitats currently differ in the way H. lepidulum spreads from source populations. Our results suggest that H. lepidulum invasion is at an earlier stage in subalpine areas, that there are few barriers to invasion across this habitat, and that subalpine habitats will become more heavily invaded than forests. These findings can be used to underpin monitoring strategies and management prioritisation for this invader.