Methods.

The study was conducted on the Wreck Rock beaches (22 km) adjacent to Deepwater National Park (lat 24°15′40″–24°21′39″S, long 151°53′28″–151°58′27″E) located in southeastern Queensland, Australia. The nesting beach was patrolled each night, and newly constructed turtle nests were marked with a stake. Early the following morning, these nests were found, and one of the treatments was applied or the nest left untouched as a “control.” For the aluminum cage treatment, the surface of the nest was dug down to a depth of 20 cm, and an aluminum mesh (mesh size 5 × 5 cm) cage 1 × 1 m with 10-cm sides was placed on top of the nest, and then sand was replaced on top of the cage to the original nest height. For the plastic mesh treatment, the surface of the nest was dug down to a depth of 10 cm, and a 1.2 × 1.5-m plastic mesh (mesh size 5 × 5 cm) was placed on top of the nest. The mesh was pegged down at the corners and the sides using 40-cm wooden stakes. Sand was then replaced on top of the nest to its original height. For the hot chili pepper treatment, the surface of the nest was dug down to a depth of 10 cm in a 0.5 × 0.5-m square, and 40 g of hot chili power (heat level 10 of 10; Herbie's Spices, Rozelle, New South Wales) were sprinkled evenly over the surface, and the sand was replaced on top of the nest to its original height. For the red flag treatment, a flag made from bright red canvas material (30 × 40 cm) mounted on a 1.2-m-high stake was inserted 50 cm into the sand 30 cm to the side of the nest. The control nests were marked with a wooden stake 30 cm to the side of the nest. The nests were visited every day from early December to the end of February, and predator visitation, as evidenced by the presence of tracks, and predation events, as evidenced by tunnels dug into the nest chamber, were recorded.

Results.

During the 2014–2015 sea turtle nesting season, from 12 May 2014 until 1 June 2015, 10 aluminum cages, 11 plastic meshes, 10 flagged nests, 10 chili powder–treated nests, and 11 control loggerhead turtle nests were set up. All treatments experienced some monitor lizard predation activities; however, even in predated nests, not all eggs were eaten, and some hatchlings escaped the nest successfully (Table 1). Almost all control nests experienced some monitor lizard predation, while nests protected with plastic mesh experienced the least predation (Table 1). Aluminum cages and plastic mesh were effective against monitor lizard predation if they were properly deployed, with predation rates significantly lower than control nests (Table 1). Successful attacks on aluminum cage–protected and mesh-protected nests occurred because wind erosion exposed an edge of the protective barrier and monitor lizards burrowed under the exposed edge. Both flags and hot chili powder–treated nests experienced a higher predation rate than aluminum cage and plastic mesh nests, and their predation rates were not significantly different from control nests (Table 1).

Table 1. Summary of nest protection treatment and emergence success results for monitored loggerhead turtle nests. 2 × 2 contingency tables and Fisher exact tests were used to compare nest protection treatments with the control group in each nesting season.

View Fullscreen

During the 2015–2016 sea turtle nesting season, from 12 July 2015 until 1 June 2015, 15 plastic mesh–protected nests, 15 chili powder–treated nests, and 16 control loggerhead turtle nests were set up. The overall predation rate of nests was lower in the 2015–2016 season compared with the 2014–2015 season, as judged by the predation rate of control nests (Table 1). As a consequence of the generally low nest predation rate in the 2015–2016 season, there was no difference in the predation rate of control nests, plastic mesh nests, and chili powder nests, although chili powder nests received the greatest amount of predation (Table 1).

Discussion.

During the 2014–2015 sea turtle nesting season, almost all control nests experienced some monitor lizard predation activities; however, only 12.5% of control nests were predated by monitor lizards in the 2015–2016 season. This indicated that monitor lizard predation rate can vary greatly between different seasons. Christian et al. (1995) reported that yellow-spotted goannas were active in the wet season in the wet–dry tropics of northern Australia but restricted their activity in the dry season to the wetlands, where there was still a high abundance of prey. This indicates that yellow-spotted goannas concentrated their foraging activities in habitats rich in prey. Ghost crabs were reported in yellow-spotted goanna scats all year-round (Blamires 2004), indicating that ghost crabs are a staple food source for this species. At our Wreck Rock study site, ghost crab burrows were numerous in the dune areas immediately behind the beach. We suspect that the abundance of ghost crabs in the dunes behind the beach may have been greater during the 2015–2016 season than in the 2014–2015 season, and this could explain the lower predation rate of sea turtle nests in the 2015–2016 season. Further studies of the connection between yellow-spotted goanna activity area and their food abundance are needed to test this hypothesis.

Hot chili powder and red flags did not provide an effective deterrent against monitor lizard predation. Hot chili powder contains capsaicin, which is responsible for the immediate “burning” sensation on ingestion experienced by mammals and so actively deters its ingestion in mammals, but it apparently is not distasteful to birds and therefore has no repellant effect in birds (Jordt and Julius 2002; Levey et al. 2006; Baylis et al. 2012). Hence, it is probable that monitor lizards also do not find capsaicin distasteful and that this was the reason that hot chili powder–treated nests failed to deter monitor lizard predation. However, predation of hot chili powder–treated nests was lower than that of control nests during the first 2 wks after nest construction, so hot chili powder may have had a deterrent effect initially but became ineffective as time progressed possibly because rain diluted the effect. Further experimental work of adding chili powder at regular intervals throughout the incubation period is needed to test this hypothesis. Red flags did not appear to affect monitor lizard behavior; they continued to visit nests (as evidence by tracks) and frequently opened nests. Longo et al. (2009) reported that flags may be a less effective deterrent to mammalian predators in nonconstant coastal wind areas. We found when turtle nests were constructed behind the first dune, they were protected to a large degree from costal wind and the flags were limp for much of the time. However, even when nests were constructed on the front of the dune and flags consistently fluttered in the wind, monitor lizard predation of nests remained high. Flags might also serve as a visual cue to alert a predator to the presence of a nest by the predator learning to associate a visual cue with a food source, as has been observed with feral swine (Engeman et al. 2016), and thus potentially increase the predation rate of nests. For example, raccoons (Procyon lotor) were found to use nest cages as landmarks when searching for sea turtle nests (Mroziak et al. 2000).

Both plastic mesh sheets and aluminum cages, when protected against sand erosion around their edges, were highly effective in preventing monitor lizard predation of sea turtle nests. However, installing and removing aluminum cages requires much effort. Plastic mesh sheets could be deployed substantially faster than aluminum cages. In our procedures, 2 people installed nest protection, and on average it took 40 min to install an aluminum cage but just 20 min to install a plastic mesh sheet. Kurz et al. (2011) also reported that installing screens took 48% less time than installing cages under the same conditions. In addition, the costs of aluminum cages averaged AUD (Australian dollars) 200 per nest, but plastic mesh cost only AUD 20. Although cage protection is an established option for nest management at some sea turtle nesting beaches (Addison and Henricy 1994; Ratnaswamy et al. 1997; Kinsella et al. 1998), many programs probably lack the resources necessary to purchase and deploy hundreds of cages. Therefore, the high success rate and low cost for plastic mesh sheet–protected nests indicate that plastic mesh sheets are an effective way to reduce monitor lizard depredation of sea turtle nests. However, both aluminum cages and plastic mesh sheets failed if the edge of the barriers became exposed on the surface as a result of wind or rain erosion. Once an edge became exposed, monitor lizards were able to burrow under the barrier and access the egg chamber. As a consequence, we recommend that the upper surface of these physical barriers be buried 20 cm below the sand surface. Also, the larger the protected area immediately on top of the nest, the more effective it is as a barrier to predators, as previously shown with raccoon predation of sea turtle nests (Addison 1997). Our anecdotal observations indicated that both aluminum cages and plastic mesh sheets that had 5 × 5-cm mesh may impede some hatchlings during the nest escape process. For this reason, in the 2015–2016 sea turtle nesting season, we enlarged the mesh size to 10 × 5 cm in the region immediately above the egg chamber, allowing hatchlings to escape unimpeded but still providing effective protection against monitor lizard predation. Kurz et al. (2011) suggested that it may not be appropriate to use cage or mesh protection on high-density nesting beaches because the presence of cages or mesh may deter or impede nesting females. Therefore, management methods should also be beach specific, taking into account predation threats, resources for protection, and nesting turtle densities. In addition, the earth's magnetic field has been shown to be a vital navigational aid to sea turtle hatchlings (Lohmann 1991; Lohmann and Lohmann 1998; Irwin and Lohmann 2003); therefore, nest protection structures constructed with ferrous metals should not be used, but those constructed from plastic or aluminum should be safe.