When new individuals from a pest species are detected after an eradication program, it is important to determine if these individuals are survivors from the eradication attempt or reinvaders from another population, as this enables managers to adjust the methodologies for future eradications and for the quarantine procedures of these islands. Within the New Zealand environment, the primary threat to native fauna on coastal islands is the stoat (Mustela erminea); a voracious introduced predator that has invaded at least 90 islands. Several stoat eradication programs are currently underway on New Zealand's islands, however on many of these islands stoats are still being caught years into the control program. Using genetic assignment and kinship-based techniques we have assessed where these stoats came from - are they surviving residents or invading migrants? In each program, all stoats caught have been genotyped at 16 microsatellite markers along with stoats from nearby potential source populations. Along with this, the age of every individual caught was assessed using cementum analysis. A variety of assignment techniques were then applied to these data to determine the best method for genetic population assignment, and models were created to describe the invasion and survival rate on each island.
Results/Conclusions
The ability to accurately assign individuals to source populations varied across islands, with islands that are more distant more readily differentiable based on genetics. Results indicate that stoats can swim further, and do so more regularly than previously thought. Both invasion from the opposing mainland and in situ survival and breeding have been detected to various degrees. On one island previously thought to be beyond the swimming range of stoats (Rangitoto Island 3 km offshore), an individual trapped post-eradication was shown to be a reincursion from the mainland. On Secretary Island, low but significant levels of reinvasion were confirmed, possibly linked to pulsed food resources on the opposing mainland due to mast years of dominant tree species. Along with this, in-situ breeding was detected using kinship analysis. These results provide vital information to improve the management procedures and success of these eradication programs, and they highlight the efficacy of genetic techniques in assessing these questions.