Landcare Research - Manaaki Whenua

Landcare-Research -Manaaki Whenua

Assessing risk to non-target animals from brodifacoum bait in the Galápagos

In trials to assess non-target risk, captive Galapagos tortoises were presented with food containing the rodenticide brodifacoum. Image - Penny Fisher.

In trials to assess non-target risk, captive Galapagos tortoises were presented with food containing the rodenticide brodifacoum. Image - Penny Fisher.

The Galapágos Islands have a high international profile because of their unique biodiversity, including iconic species such as the giant tortoises. However, even such large, long-lived creatures are affected by invasive rodents, e.g. tortoise eggs and hatchlings are eaten by rats. As has been achieved in other islands around the world, significant benefits for endemic fauna and flora are expected if rats are permanently eradicated.

Rodent populations on many islands have been eradicated using cereal pellet bait containing the anticoagulant rodenticide brodifacoum, sown by plane or helicopter. The Galápagos National Park is using brodifacoum in a staged approach to eradicating the introduced rodents from selected islands, with the first two phases completed in 2007 and 2011. The next phase includes the island of Pinzón, which has a wild population of native tortoises.

Brodifacoum is a very effective rodenticide, with ‘single feed’ amounts of bait lethal to rodents. This high toxicity also poses an unwanted hazard to non-target mammals and birds that ingest bait (primary exposure) or live animals or carcasses that contain brodifacoum (secondary exposure). However, very little is known about the toxicity of brodifacoum to reptiles, including tortoises and turtles. As part of the planning for rodent eradication from tortoise habitat, Penny Fisher of Invasive Species International (Landcare Research) and colleagues at the Charles Darwin Research Station were commissioned to investigate the risk of brodifacoum to Galápagos tortoises on Santa Cruz Island, Galápagos, in March–April 2011.

In trials with captive Galápagos tortoises, relatively few showed interest in eating baits – the majority of the tortoises preferred their normal plant foods. This suggested that bait uptake by tortoises in field conditions is likely to be low when natural food is readily available. But because some tortoises did eat bait, blood samples were taken before and after the tortoises ate known amounts of brodifacoum relative to their bodyweight to indicate whether this was likely to affect them. An automated coagulometer was used to test the blood samples for clotting times, and to indicate any reduced ability of the blood to coagulate (a typical toxic effect of brodifacoum seen in mammals and birds). However, no significant changes to blood coagulation times in tortoises were measured over the 2 weeks following ingestion of brodifacoum. Overall, tortoises appeared to have higher normal blood coagulation times than those of mammals, supporting the idea that, in comparison to mammals, tortoises have a relatively low susceptibility to brodifacoum toxicity. In the context of the aerial bait application rates to be used in the upcoming rodent eradication programmes, tortoises weighing more than 20 kg would be unlikely to encounter and eat enough bait to harm themselves.

Bait trials with other Galápagos reptiles (lava lizards, geckos and snakes) were also undertaken, with small numbers of all three species brought into the laboratory and offered baits. Some of the lava lizards sampled very small quantities of bait but this did not produce any visible adverse effects on them, e.g. on bodyweight, food intake or behaviour. None of the geckoes or snakes showed any interest in eating bait over a 5-day period. To simulate secondary exposure to brodifacoum, lava lizards were also offered live cockroaches that had been feeding on bait and contained residual concentrations of brodifacoum in their gut. Most of the lizards ingested small amounts of brodifacoum in this way, but again no adverse effects were evident over the following 3 weeks. Because of the small sample sizes in this trial, limited conclusions can be drawn regarding non-target risk to reptiles, but Penny believes the results indicate that lava lizards are less susceptible than mammals to brodifacoum toxicity. However, it is possible that some lizards could eat sufficient bait or contaminated invertebrates to be lethally poisoned, and this requires further investigation. No conclusions about the toxicity of brodifacoum to the gecko or snake species tested can be drawn from the ‘nil exposure’ results, but it seems unlikely that either would be at risk of primary exposure through bait ingestion.

These results will be used to update non-target risk assessments for the planned rodent eradications, and to assist in the design of on-ground monitoring protocols for non-target species during and after the operations.

This work was supported by the Galápagos National Park, the Charles Darwin Research Station and Island Conservation.

Penny Fisher