Wetland connectivity: understanding the dispersal of organisms that occur in Victoria’s wetlands draft

Radio-tracking

5. Radar

5. Dispersal of wetland taxa

   1. Amphibians

Thirty-seven amphibian species occur in Victoria. Many amphibian species depend on permanent or ephemeral wetlands to complete their life cycle. Although reproductive strategies are varied, with some species laying their eggs on land or in water-filled burrows, the majority (86%) of species that occur in Victoria are pond breeders, indicating a strong dependency on wetland habitats (Appendix 1). A few species, however, show a greater affinity for flowing waters, but even these will utilise adjacent wetlands as refuges, or as stepping stones to other habitats (Appendix 1). Amphibian populations are in decline throughout the world, and are at greater risk of extinction than any other vertebrate group, with one third of species under threat (Beebee and Griffiths 2005). Seventeen Victorian frog species (one third of the total) are listed as threatened in the state, excluding species for which data is insufficient to assess the extinction risk (DSE 2007). Seven of the 17 species listed as threatened have affinities with wetland habitats:

• 
  Giant Bullfrog (Limnodynastes interioris)
  
• 
  Alpine Tree Frog (Litoria verreauxii alpina)
  
• 
  Baw Baw Frog (Philoria frosti)
  
• 
  Growling Grass Frog or Southern Bell Frog (Litoria raniformis)
  
• 
  Giant Burrowing Frog (Heleioporus australiacus)
  
• 
  Green and Golden Bell Frog (Litoria aurea)
  
• 
  Rugose Toadlet (Uperoleia rugosa).
  
  
  

1. Habitat preferences

Studies of amphibian habitat preferences conducted overseas and in Australia have shown that high vegetation cover and the absence of fish correlates with a higher species diversity of frogs (Pavignano et al. 1990, Ficetola and Bernardi 2004, Hazell et al. 2004). In Australia, for example, Growling Grass Frogs favour sites with a high proportion of emergent vegetation (Clemann and Gillespie 2010), and Green and Golden Bell Frogs prefer habitats with certain plant species such as Juncus kraussii that are used for basking and foraging, as well as habitats that are adjacent to other waterbodies (Goldingay 2008). The extent of native canopy cover in the landscape is positively related to frog species richness in ponds in south-eastern Australia (Hazell et al. 2001).

Fish predation of amphibian eggs and larvae can strongly influence amphibian community structure, although some species are more vulnerable to fish predation than others (Hecnar and M’Closkey 1997, Gillespie and Hero 1999, Gillespie 2001, Ficetola and Bernardi 2004). Frogs can resist predation by producing toxic or unpalatable substances at the tadpole stage, or through behaviours that limit predation (Ficetola and Bernardi 2004). The Australian riverine frogs Litoria spenceri, L. phyllochroa and L. lesueuri are susceptible to predation by introduced trout (Gillespie, 2001), and the introduced Mosquitofish (Gambusia holbrooki) is known to displace Green and Golden Bell Frogs (Hamer et al. 2002b). Native fish also prey on amphibians, but the susceptibility to predation may vary among amphibian species; lower rates of predation have been recorded for riverine frogs Litoria spenceri, L. phyllochroa and L. lesueuri compared with the marsh-dwelling frog Limnodynastes peroni (Gillespie 2001). Some amphibians, such as the Southern Bell Frog, have long larval stages and require more permanent water to complete their life cycle (Clemann and Gillespie 2010). As permanent water bodies favour the persistence of fish, amphibians with long larval stages that do not have adaptations to coexist with fish may be less common in permanent water bodies than amphibians with short larval stages. Ficetola and Bernardi (2004) found that amphibian diversity in a human-dominated landscape (Milan, northern Italy) was higher in temporary wetlands where fish were absent compared with permanent wetlands that were stocked with predatory fish for recreational fishing.

2. Movement

All the habitats that frogs utilise to complete their life cycle may be available in wetlands, necessitating only small migrations, or they may be available only at more distant sites and require larger migrations (Pyke and White 2001). Changes in habitat quality or water availability may also trigger frog movement. For example, the Southern Bell Frog (Litoria raniformis) abandons permanent waterbodies in favour of ephemeral waterbodies when they are available, returning again to permanent waterbodies as these sites dry (Wassens et al. 2008). Male Green and Golden Bell Frogs (Litoria aurea) make larger migrations to reach ephemeral waterbodies filled by heavy rains than they do to reach permanent waterbodies (Goldingay and Newell 2005, Hamer et al. 2008). The stimulus to utilise more distant ephemeral habitats may be that these sites have a lower risk of predation or more resources. Whatever the reason, occupying many habitat types serves to spread the risk of mortality, thus increasing the chance that some offspring will survive (Hamer et al. 2008). It is not known whether the preference to disperse to newly filled ephemeral waterbodies is a widespread phenomenon in frogs.

The maximum dispersal distance reported in European studies varies from 3 km in the toad Bufo bufo to 15 km in the frog Rana lessonae (Sinsch 1990). Species that migrate overland have been reported to utilise streams to disperse, with distances varying from 2.5 km to 10 km (Sinsch 2006). Research in America concluded that adult amphibians generally have a high site fidelity, rarely moving beyond a few hundred metres from breeding sites, while juveniles are capable of dispersing over larger distances. For example, adult American Wood Frogs (Rana sylvatica) have exceptional site fidelity; of 11 195 adults marked in one study, none were recaptured beyond their original pond over the six-year study (Berven and Grudzien 1990). In contrast, about 18% of 356 recaptured juveniles dispersed to other ponds, moving on average 1 126 m. Similar observations in other species have led to the conclusion that juveniles are largely responsible for maintaining amphibian metapopulations (Cushman 2006).

It is unclear to what extent this generalisation can be applied to Australian frogs, where extreme temporal variability in water availability has probably selected for greater mobility. For example, the Green and Golden Bell Frog (Litoria aurea) is less faithful to its breeding pond compared with the American Wood Frog, with only 53% of males and 65% of females remaining in the same permanent waterbody (Hamer et al. 2008). Mark–recapture studies demonstrate that this species may disperse up to 3 km (Pyke and White 2001), although sightings up to about 10 km from the nearest possible breeding pond have been made (White and Pyke 2008), and one individual was observed to travel 1.5 km in a single night while foraging (White, unpublished data cited in Pyke and White 2001). Similarly, the Growling Grass Frog (L. raniformis), considered among the more vagile frog species, has been reported to move up to 1 km in 24 hours (K. Jarvis, pers. comm. cited in Robertson et al. 2002). In contrast, in urban areas of Melbourne, Growling Grass Frogs that had been marked and recaptured over several seasons dispersed no more than 0.5 km from tagged sites (Heard 2010).

Although a wetland may lie in a frog’s dispersal range it may not be occupied because the habitat is unsuitable or the landscape between habitats is difficult to traverse. For example, Hamer et al. (2002a) reported that the probability of Green and Golden Bell Frogs occupying ponds in New South Wales decreased significantly when ponds were more than 50 m apart, even though radio-tracking and mark–recapture studies had shown that this species is capable of moving more than 500 m, sometimes in 24 hours (Hamer et al. 2008). This disparity may be because of habitat preferences, or because the landscape between habitats was inimical for dispersal.

Movements among habitats, however necessary for survival, also carry a high risk of mortality. When moving among habitats, frogs are vulnerable to predation and, because their skin is permeable to water, also desiccation (Wassens et al. 2008). In temperate landscapes the prevalence and diversity of amphibians declines where the distances among wetlands is large, where road densities are high, or where habitats are surrounded by agriculture or other intensive land uses (Rothermel and Semlitsch 2002, Parris 2006). All these factors decrease the capacity of amphibians to move between habitats and maintain metapopulations.

To minimise the risk of desiccation, movement often occurs during precipitation (Richter et al. 2001, Penchmann and Semlitsch 1986) or flooding (Wassens et al. 2008). To avoid predation and desiccation, some species preferentially move in closed-canopy forest habitats (Vasconcelos and Calhoun 2004). Open habitats such as agricultural fields increase the predation risk, and because temperatures are general higher there the risk of desiccation also increases (Rothermel and Semlitsch 2002). Decreasing forest cover within 2 km of wetlands has been associated with reductions in amphibian and reptile species richness (Findlay and Houlahan 1997). The American Toad (Bufo americanus) shows a strong preference for moving beneath forest canopy cover; a movement study by Rothermel and Semlitsch (2002) captured only 3 of 83 toads in open fields. This finding was surprising because adults are found in agricultural landscapes, suggesting that juveniles may be more prone to desiccation than adults.

Roads represent significant barriers to amphibian movement by increasing the risk of death and by exerting psychological or physical constraints on movement (Mader 1984, Gibbs 1998). Ehmann and Cogger (1985) estimated conservatively that 5.48 million reptiles and frogs were killed annually by traffic in Australia. Traffic intensity increases frog and toad kills and reduces population size, as measured by choruses (Fahrig et al. 1995). Increasing road density within 2 km of wetland boundaries has been correlated negatively with amphibian and reptile species richness in 30 wetlands in Ontario, Canada (Findlay and Houlahan 1997).

Because mobility and habitat requirements vary among species, the relative importance of local environmental variables, distances between habitats, and the traversability of the surrounding landscape will vary in shaping the spatial distribution of the frog species. For example, Ficetola and Bernardi (2004) assessed the relationships between wetland features and isolation on the presence or absence of amphibian species in 84 wetlands in a human-dominated landscape in northern Italy. The key finding of their study was that wetland features and wetland isolation both structure amphibian communities. Environmental features that were correlated with amphibian presence were water depth, water permanence, sun exposure and fish presence. Rare species were affected by both environmental features and isolation, whereas common species were less affected by wetland features and were more mobile. In environments where the landscape is inimical to movement, vagile species are probably at greater risk than sedentary species. In other systems the habitat itself may be under threat, and in this case sedentary species are at greater risk than vagile species (Rothermel and Semlitsch 2002).