New Zealand Journal of Ecology (1999) 23(1): 45- 51

An evaluation of the efficiency of rodent trapping methods: The effect of trap arrangement, cover type, and bait

Research Article
Ji Weihong  
C. R. (Dick) Veitch  
John L. Craig  
  1. School of Environmental and Marine Sciences, University of Auckland, Private Bag 92019, Auckland 1, New Zealand
  2. Department of Conservation, Private Bag 68-908, Newton, Auckland, New Zealand

Eradication of rodent species from some offshore islands has proved to be an effective means of conserving native animal communities and restoring natural ecological processes on the islands. As methods of eradication differ fur different rodent species, a truthful monitoring method to detect species presence and relative density is essential for a successful eradication programme. This study compared two spatial arrangements (line vs, grid), 5 different baits (chocolate, cheese, soap, wax, oiled wood) and 3 cover types (transparent plastic, wire netting, galvanised iron) on the detection of 2 species of rodents on Browns (Motukorea) Island in June and August. The two species of rodents present on the island were Norway rats (Rattus norvegicus) and mice (Mus musculus). Trapping using conventional trapping lines and trapping grids was carried out in June and August, respectively. The traps were set for 8 nights for both lines and grids. Trap lines caught 12.40 rats per 100 corrected trap-nights (100 ctn(-1)) and no mice; trap grids caught 5.3 rats 100 ctn(-1) and 0.2 mice 100 ctn(-1). Trap grids appeared to be better than trap lines for detecting the presence/absence of rodent species when two species coexist and one appears subordinate to the other. On trap lines the trapping rate of rats was consistently high for five of the first six nights. On trap grids the trapping rate was variable on all nights with the first mice being caught on the third night. Three-night trapping sessions, conventionally used by most researchers, should be reliable for testing the relative densities of numerically dominant species but may not detect all rodent species present. Of five different bait types tested for monitoring rodent presence, the preferred order was chocolate, cheese, soap, wax and oiled wood. The efficiency of covers made of different materials, (galvanised iron, plastic, and wire netting) was also tested. Wire netting covers had the highest trapping rate and galvanised iron covers had the lowest. Three blackbirds (Turdus merula) were caught under wire netting covers, indicating a risk to non-target organisms.