Ferrets are an invasive species especially common in New Zealand farmland. Ferrets become infected with bovine tuberculosis by feeding on infected carrion, and may also be a wildlife reservoir of the disease when their density, averaged through the year, exceeds a threshold of about 3 per km2. Reliable knowledge of ferret density is therefore useful for disease managers. Obtaining such knowledge is difficult because ferrets are highly mobile and estimates are required for large areas (>20 km2). We describe the results of 11 field trials of a new spatially explicit capture-recapture method implemented in program DENSITY. Ferrets were captured, marked and released over 6 days in 130-170 live traps per study area (16.8-42.1 km2). Estimated density ranged from 0.8 per km2 (SE 0.2) to 6.9 per km2 (SE 0.9). Relative precision (SE (Estimate) Estimate) ranged from 6% to 29%. Home-range size varied inversely with density; this effect could be expected to bias comparisons based on conventional estimators that use an effective trapping area. Some marked ferrets were caught repeatedly at a particular site, which suggests a lack of fit to the spatially explicit detection model. However, an improved method implemented in version 3.2 of the DENSITY software appears robust to this effect. Two problems require further consideration: extrapolation from the trapped sample to a management area requires the assumption that the sample is spatially representative. Ordinarily this can be ensured by random placement of lines, but topography and difficulties of access usually make random placement unacceptably expensive. We also encountered problems with occasional dispersal movements. We suggest that ferret monitoring be under taken at times of year when little dispersal is expected.
|Author||Efford, M. and Norbury, G.|
|Secondary title||13th Australasian Vertebrate Pest Conference|
|Place published||Wellington, NZ|