New Zealand Journal of Ecology (2011) 35(2): 145- 152

Modelling detectability of kiore (Rattus exulans) on Aguiguan, Mariana Islands, to inform possible eradication and monitoring efforts

Review Article
Amy A. Yackel Adams 1*
James W. Stanford 2
Andrew S. Wiewel 3
Gordon H. Rodda 4
  1. ASRC Management Services under contract to USGS, Fort Collins Science Center, 2150 Centre Avenue, Building C, Fort Collins, Colorado 80526, USA
  2. USGS Brown Treesnake Laboratory, PO Box 8255 MOU-3, Dededo, Guam 96912, USA
  3. 26170 300th Street, Maywood, Missouri 63454, USA
  4. USGS Fort Collins Science Center, 2150 Centre Ave., Bldg. C, Fort Collins, Colorado, 80526, USA
*  Corresponding author

Estimating the detection probability of introduced organisms during the pre-monitoring phase of an eradication effort can be extremely helpful in informing eradication and post-eradication monitoring efforts, but this step is rarely taken. We used data collected during 11 nights of mark-recapture sampling on Aguiguan, Mariana Islands, to estimate introduced kiore (Rattus exulans Peale) density and detection probability, and evaluated factors affecting detectability to help inform possible eradication efforts. Modelling of 62 captures of 48 individuals resulted in a model-averaged density estimate of 55 kiore/ha. Kiore detection probability was best explained by a model allowing neophobia to diminish linearly (i.e. capture probability increased linearly) until occasion 7, with additive effects of sex and cumulative rainfall over the prior 48 hours. Detection probability increased with increasing rainfall and females were up to three times more likely than males to be trapped. In this paper, we illustrate the type of information that can be obtained by modelling mark-recapture data collected during pre-eradication monitoring and discuss the potential of using these data to inform eradication and post-eradication monitoring efforts.