In order to conserve important biodiversity values, eradication of feral cats (Felis catus) is planned on Auckland Island in the New Zealand subantarctic region. This eradication will require detailed knowledge of the abundance, distribution, movement behaviour and detection probability of cats on the island. We investigated these parameters on a peninsula at the northern end of the island using live trapping, camera trapping, and scat searches with and without detection dogs. Here, we compare the results of these methods, and discuss their utility for the planned eradication. Four cats were captured and fitted with GPS collars. Camera traps with 500 m spacing detected all these individuals on multiple occasions, and at multiple locations. At least 12 other individuals were also captured on camera. Excluding every second camera (to simulate 1000 m spacing) resulted in failure to detect 32% of known individuals. Population density estimates from camera trapping varied from 0.7–1.0 cats km-2. Humans found 29 cat scats, and dogs found 33. Genetic analysis estimated that these came from a minimum of ten individuals. Camera trapping should be repeated during the operational and confirmation phases of the eradication to monitor spatial and temporal variation in cat density, detect survivors, and help confirm eradication success. Scat collection, with and without dogs, can supplement data from camera trapping. With larger sample sizes of scats, DNA profiling may also allow cat abundance to be estimated.