seaturtle.org : MTN : ARCHIVES : Sign In

During the 2003/2004 and 2004/2005 nesting seasons, an invasion of brown rats (Rattus novergicus) became a threat to nesting success, with an estimated loss of close to 3000 eggs and hatchlings (Figure 1). The brown rat, an aggressive species, is able to locate nests and kill hatchlings inside the egg chamber before emergence. To tackle the problem it was decided to eradicate the rodents along a 2.3km section of the nesting area that is limited north and south (7°02’11” S 23°50’09” W and 7°03’18” S 34°50’35” W) by two small intermittent streams. The objective was to eliminate brown rat depredation of eggs and hatchlings.

Figure 1. Hawksbill hatchlings, from nests laid in an urban area in northeastern Brazil, attacked by rats.

The eradication program was conducted in four steps: 1- survey of non-target fauna, 2- setup of the management area, 3- anticoagulant poisoning campaign, and 4- continued monitoring to prevent recolonization. The non-target species were determined by direct observation, bibliographic research and consulting the reference collection of the Mammalian Cytogenetic Laboratory of the Paraíba Federal University. The results indicated that the mastofauna is impoverished as a result of urbanization. Local populations of two alien species of rodents (R. novergicus and Mus musculus/) and one native species of marsupial (Didelphis aurita) were expected to be present, the latter population also expected to be in a state of disequibrium. However, there appeared to be a high diversity of predatory and scavenger birds, all of which could be affected through secondary poisoning (by feeding on dying or dead poisoned rats). Two species of owls were observed, the white owl (Tyto alba) and burrowing owl (Athene cunicularia), five species of hawks (Polyborus plancus, Rupornis magnirostris, Milvago chimachima, Elanus leucurus, and Falco sparverius) and three species of vultures (Cathartes aura, C. burrovianus, and Coragyps atratus). The herpetofauna and invertebrates were not considered to be threatened as the poison affects mammals, birds and fishes only. The fish were not surveyed because poisoning was conducted in the upper part of the beach and the bait was protected against rain and removal from bait stations, to avoid the risk of water contamination.

For poison, we used the anticoagulant brodifacoum, which is lethal to rats even in small concentrations. Brodifacoum acts by disrupting Vitamin K production needed for blood clotting. It has been successfully used in a similar project in Sangalaki, Indonesia, and evaluated by the Canadian Health Agency (Meier & Varnham 2004; HEALTH CANADA 2004). The poison was deployed in wheat grains embedded in 20g blocks of blue wax. To reduce the risk of poisoning non-target species, poisoned baits contained “bitrex,” which adds a bitter taste that is detectable by non-rodent species. The poisoned bait was placed in bait stations made of recycled plastic bottles with 15cm diameter end openings. These were anchored with wire to avoid the bait being removed from the station. To further reduce the risk of non-target species poisoning, the area was regularly monitored and exposed carcasses were removed to avoid consumption by scavengers. (Meier & Varnham 2004; Health Canada 2004).

A grid with 192 bait stations was established along the vegetation belt. Bait stations were distributed in three transects parallel to the sea shore. The first transect was placed seaside of a sidewalk along the edge of the vegetation belt. In this transect, stations were placed 100m apart. The second transect was placed along the middle axis of the vegetation belt, and the third transect was placed where the vegetation joins the sand beach. In the second and third transects, the stations were 25m apart. Two bait blocks were placed at each station and were replaced on a daily basis during the first two weeks. From the third week to the end of the program, the bait was replaced twice per week. The eradication program lasted 15 weeks, but all evidence of rats on the beach disappeared within six weeks. However, a small and somewhat constant bait consumption was observed throughout the program. After 15 weeks, the bait stations were removed and a new grid was established for the continued monitoring. Stations were placed every 100m along all three transects and poisoned baits were replaced monthly to detect new invasions.

The eradication process was considered complete when an evaluation of the direct and indirect evidences of the presence of rats on the beach met the following criteria: 1- the bait intake decreased to very low and constant levels, likely being due to non-target invertebrate consumption, such as cockroaches and other insects (Fig. 2); 2- no dead or alive rats or decaying odor was observed after 10 weeks; 3- no footprints, tracks on vegetation or evidence of food intake were observed; and 4- no further depredation of hawksbill turtle nests was observed.

A total of 461 poison blocks were consumed, or 9.5 kg (20.9 pounds). Despite the efforts to recover all the carcasses, only eight were actually found, although the odor of decay was present throughout the area. This is likely to be due to the features of the beach and the species behavior. Brown rats spend their inactive periods in burrows and were presumed to have died underground. Only one non-target marsupial (D. aurita) was found and was likely poisoned during the eradication program. Four poison blocks presented teeth marks compatible to those of a lizard (estimated body length 20 cm). During the final two weeks, several bait station were infested by dozens of cockroaches that completely covered the bait blocks.

Figure 2. Distribution of the bait intake. Y axis represents the number of 20 g bait blocks replaced each revisit to bait stations. Revisits axis represents each revisit of bait stations.

We successfully achieved rat eradication in the target area. No nests were attacked by rats in the 2005/2006 season or in the ongoing 2006/2007 season. A continued monitoring program to prevent new infestations is being conducted by the Urban Turtle Project staff, on a volunteer basis with support from the municipal administration for materials. The results show that a simple procedure with low financial and environmental costs can have significant conservation effects for sea turtles.

Acknowledgements: Albano Schulz Neto identified the birds; Valdi S. Moreira, Adailton Galdino, and the students and volunteers of the Urban Turtle Project conducted most of the field work; the Municipal Agencies of Environment and Health of Cabedelo (PB, Brazil) provided the materials for the program; the Laboratory of Cytogenetic of Mammalians, UFPB, opened the reference collection and library to the identification of the mammals, snakes and lizards; the anonymous referee revised the manuscript and deeply improved the language.

MASCARENHAS, R., D.F. ZEPPELINI & V.S. MOREIRA. 2003. Observations on sea turtles in the state of Paraíba, Brazil. Marine Turtle Newsletter 101:16-18.

MASCARENHAS, R., R.G. SANTOS, A.S. SANTOS & D. ZEPPELINI. 2004. Nesting of hawksbill turtles in Paraíba-Brazil: avoiding light pollution effects. Marine Turtle Newsletter 104:1-3.

MEIER, G. G. & K. VARNHAM. 2004. Rat eradication as part of a green turtle (Chelonia mydas) conservation programme in Indonesia. Marine Turtle Newsletter 106:11-12.

HEALTH CANADA. 2004. Re-evaluation of brodifacoum, bromadiolone, chlorophacinone, diphacinone and warfarin. Pest Management Regulatory Agency, Health Canada. Ottawa, Ontario. 36pp. Available: <http://www.pmra-arla.gc.ca/english/pdf/pacr/pacr2004-27-e.pdf>