Risk assessment of birds foraging terrestrially at Marion and Gough Islands to primary and secondary poisoning by rodenticides

Abstract

Context. Aerial application of poison bait pellets is an established and widely used method for removing invasive rodents and restoring insular ecological processes. However, the non-target effects of saturation poisoning require very careful consideration and precautionary risk-avoidance strategies. Aims. We assessed the risk of primary and secondary poisoning by rodenticides to terrestrially foraging lesser sheathbills (Chionis minor marionensis), Gough moorhens (Gallinula comeri) and Gough buntings (Rowettia goughensis) at Marion and Gough Islands. Methods. Birds taken into temporary captivity were offered non-toxic bait pellets dyed different colours and the carcasses of house mice (Mus musculus). In addition, dead mice were offered to these three species in the field, as well as to sub-Antarctic skuas (Catharacta antarctica) at both islands. Response to captivity was assessed by daily weighings. Key results. Individual birds either gained or lost mass overall during their 4–7 days in captivity. Whereas all captive birds pecked at the pellets, minimal amounts were consumed. However, Gough moorhens offered pellets in the field did consume them. Sheathbills (in captivity and in the field) and moorhens (in the field) consumed mouse carcasses, whereas buntings in captivity ate little from them. Sub-Antarctic skuas offered mouse carcasses in the field at both islands readily consumed them. At Gough Island some, but not all, skuas consumed bait in the field. Conclusions. Although the levels of assessed risk to primary and secondary poisoning differed among the three main species studied, it is recommended that populations for subsequent reintroduction be taken into temporary captivity before and during a poison-bait exercise as a precautionary measure. It is not deemed necessary to take sub-Antarctic skuas into captivity because they will be largely absent during a poisoning exercise in winter (the most likely period). Implications. Captive studies to assess susceptibility to primary and secondary poisoning are useful for determining positive risk; however, cage effects can cause false negatives by altering behaviours, and should be conducted with complimentary field trials. Where endemic species show any degree of risk (e.g. are vulnerable to the poison, regardless of how it might be ingested), precaution dictates that the risk be mitigated.