Frogs on the hop: translocations of Green and Golden Bell Frogs Litoria aurea in Greater Sydney Arthur W. White and Graham H. Pyke Frog Ecology and Behaviour Group, Australian Museum, 6-8 College Street , Sydney, 2000. Email for corresponding author: [email protected] Translocations involving the endangered Green and Golden Bell Frog Litoria aurea have been under way in the Greater Sydney area since 1993. Case studies for four of these translocations are presented; the translocation sites being at Botany, Marrickville, Long Reef and Arncliffe. Bell frogs have persisted at only one of these sites since their introduction (Arncliffe). The success or failure of each translocation has provided insights into the habitat requirements and T management of bell frogs. In unsuccessful translocations, the reasons for the inability to establish a C permanent population became more apparent with monitoring; at Botany, young bell frogs failed to survive the winter because of inadequate or inappropriate over-winter habitat being available; at Long Reef, A foraging and breeding habitat were inadequate; at Marrickville, urban predators and disease eliminated R frogs. Other factors also appear to have a significant effect on the likely outcome of the translocation. These include: the proximity of a source population, the presence or absence of predatory fish, pond T water temperature and the timing of the release of tadpoles at translocation sites. S Despite the difficulties and uncertainties associated with habitat creation and the establishment B of translocated frogs, translocations remain as a last resort strategy for the conservation of frog A populations that may otherwise be lost. Key words: Green and Golden Bell Frog, Litoria aurea, frog translocations Introduction Translocations of frogs may be carried out both to In Australia, however, translocation has only recently been enhance the conservation status of a species or to adopted and so far has involved only a few species. Prior to protect a population from the threat of impending human 1995 there were no regulated programs for the relocation development (Seigal and Dodd 2002). Individuals of a of frogs. Since then translocation has been used as part of species may be introduced (or reintroduced) into an area programs to protect or conserve just three frog species in where the species does not presently occur in order to Australia; namely the Green and Golden Bell Frog Litoria increase its distribution (Griffiths et al. 1989). Individuals aurea (Pyke and White 2001), the Spotted Tree Frog Litoria may be moved from one population to another in order to spenceri (Gillespie and Hero 1999) and the Corroboree promote outbreeding and reduce the impacts of inbreeding Frog Pseudophryne corroboree (Hunter et al. 1999). (Griffiths et al. 1989). Individuals that are threatened by The Green and Golden Bell Frog has become a popular impending human development may be “rescued” and translocation species in New South Wales (NSW) moved to another location (Oldham et al. 1991). because of its conservation status, its occurrence at Translocation programs often include habitat sites where human development has been carried out, enhancement, threat reduction and captive breeding. It is and legislation regarding the assessment and mitigation generally assumed, for example, that there is little point in of likely impacts of proposed human developments on attempting to reintroduce a species into a location where the environment. Since 1992 it has been considered it once occurred, unless deleterious changes to the habitat to be under the highest risk of extinction in NSW are reversed and local threats to survival of the species are and is presently labelled ‘endangered’ under the NSW ameliorated (Seigal and Dodd 2002). Captive breeding Threatened Species Conservation Act 1995. In NSW, most is sometimes used to increase the numbers of animals of the known sites where this species occurs are on available for stocking or to maintain a population while privately owned land that have been greatly disturbed by suitable habitat is found or developed. past human activities and are likely to experience further For over 20 years translocation of frogs has been a widely human development (Pyke and White 2001). Since many and much used conservation strategy in North America of these sites are completely artificial or human-made, it and Europe. For example, during the 1980’s there were has been considered possible to create new habitat for over 100 translocations using crested newts (Triturus spp.) the species (e.g. White 1999). Under recent and current in the United Kingdom alone (Oldham et al. 1991). The environmental legislation in NSW, any adverse impacts success of many of these translocations was unknown as of proposed human development on ‘endangered’ species follow-up monitoring after release was rarely carried out. must be evaluated and mitigated if possible before the The failure to assess the effectiveness of these relocations proposed development receives necessary approval. In did not reduce the frequency of translocation of frogs and NSW, translocation of these frogs into a new site has salamanders in the northern hemisphere. been seen as a way of allowing development to proceed Australian October 2008 Zoologist volume 34 (3) 249 White and Pyke while protecting and conserving the frogs (e.g. Fanning the site consisted of a large, levelled rectangular block of and White 1994). In Victoria, on the other hand, the ground. The site was almost denuded of vegetation as a species is not considered at risk of extinction and there consequence of 30 years of sand mining. A drainage pond is less obvious overlap between the frogs and human was excavated in the south-western corner of the site and development. the land contoured so that stormwater runoff was diverted into the pond (Fig. 1). This pond was later colonised by The high public profile of the Green and Golden Bell bell frogs. The water level in the pond fluctuated markedly Frog (hereafter bell frog) as a consequence of its presence over time and would go dry for short periods each year. at Sydney Olympic Park led to interest in its fate at The eastern part of the Rosebery site contained mounds other locations where developments were proposed. One of bricks and concrete, which were being crushed for consequence of this notoriety is that bell frogs are often recycling. The remainder of the site was vacant. used to create media attention by opponents of particular developments where this species may be implicated (e.g. Although it would probably have originally been part of a Koch 1995). much larger and more extensive population of bell frogs, this population had become isolated from all others. Prior Research interest in this species has been prevalent since to 1980, the population would have been linked to the the species was listed as “endangered” in NSW. In 1995, a bell frog population in the Botany wetlands at Eastlakes conference on the biology of the species was hosted by the only 1.5 km away. In 1980, the Southern Cross Drive Royal Zoological Society of New South Wales to provide was created which excised the Rosebery site from the a forum for research and discussion about bell frogs. A Botany wetland corridor. The Botany wetlands’ bell frog recent review of the biology of the species has shown that population was already in decline at this time and the it has been the subject of more research studies and more population subsequently disappeared in 1993 (White and scientific publications than any other Australian frog Pyke 1996). It is assumed that the Rosebery bell frogs are (Pyke and White 2001). In respect of both translocation the last survivors of this much larger metapopulation. and research, the bell frog has become a ‘flagship’ species. Re-development of the site began in 1992 with further The bell frog has also been the subject of more translocation levelling of the site in preparation for the construction proposals and attempts than any other Australian frog. of a large high-rise residential complex. As part of the Within the greater Sydney metropolitan area, for example, translocations have been carried out at four locations (Arncliffe, Botany, Long Reef and Marrickville). The species is also the focus of other habitat modification programs that involve the colonisation of new areas from adjoining sites (e.g. Greenacre, White 1999; Port Kembla White 2001a; Woonona; White 2001b). Despite this high level of interest in bell frogs and their translocation there has been limited evaluation of how successful each translocation was in meeting conservation objectives and little assessment of the factors likely to have influenced this. The aim of this paper is: to provide an overview of four case histories (that we have been involved with), and to consider the possible factors that may explain the results. We consider attempted translocations that have taken place in relation to two source populations (i.e., Rosebery, Arncliffe) and the four recipient sites listed above. Source sites, history and reasons for translocations Rosebery population The Rosebery site is located in eastern Sydney on an urban block bounded by Southern Cross Drive to the east, Kimberley Grove to the south, Dalmeny Ave to the west and by residential and industrial units to the north. The frog pond that was the source of founder animals for a captive-breeding colony at Taronga Zoo was located in the south-western corner of the site (334380 E, 6245724 N). The site has been in private ownership since at least the 1950s when it was an active sand mine. Since 1993 it has been owned by Meriton Apartments Pty Ltd. In 1991, when Figure 1. Rosebery bell frog site, January 1991. bell frogs were first discovered at this site (Cogger 1992), Photo, A. White. Australian 250 Zoologist volume 34 (3) October 2008 Frogs on the hop re-development, the frog pond and surrounding habitat a known site for bell frogs (White 1998). The Arncliffe were to be destroyed. Evaluating the impacts of these bell frog population was one that was well documented development proposals became the first test of new and opportunistic survey records had been kept on this legislation in relation to impacts on species considered to population since 1980 (A.W. White unpubl. data). be under risk of extinction. The Endangered Fauna (Interim In 1996, a Supplementary Environmental Impact Protection) Act 1991 had come into force in NSW and the Statement (EIS) was prepared for the motorway. The EIS bell frog was listed under Schedule 12 as an “endangered” included a Species Impact Statement (Gunninah 1996) species. If the proposed development was deemed to that concluded that the motorway would have a significant pose a significant impact on the frogs then a Section 120 impact on the bell frogs at Arncliffe and that specific licence would be required by the developer. In order to conservation measures would be required to protect the gain the licence, a Fauna Impact Statement (FIS) would species. In 1998, a Plan of Management (White 1998) need to be undertaken to assess the viability of the frog for the Arncliffe bell frogs was accepted by the NPWS. population and to propose ameliorative measures that In brief, the plan detailed the creation of replacement could assist with the conservation of the Rosebery bell habitat (including two frog breeding ponds) on land frog population. next to the Kogarah Golf Course, the establishment of a This led to surveys being conducted by the Australian captive-breeding colony of Arncliffe bell frogs that would Museum from August of 1992 through to May 1993. The supplement the reduced number of bell frogs in the Marsh surveys located two adult bell frogs but found tadpoles Street area, and the establishment of a Management to be continually present in the drainage pond. On the Committee that would oversee the maintenance of the basis of the low number of adult frogs, it was suggested in wetlands and manage frog populations. the FIS (Cogger 1993) that the frog population may not be self-sustainable. In August 1993, the NSW National It was decided that the location of the recreated bell frog Parks and Wildlife Service (NPWS) convened a meeting habitat should be as close to the Marsh Street wetland and with the Museum staff, the developers, and invited the Kogarah Golf Course (known bell frog foraging area) frog biologists. The proposition that the population was as possible. A suitable site was found on the northern side unsustainable was the main topic for discussion. It was of the proposed M5 East Motorway corridor, next to the agreed that the best outcome for the frogs would be to Kogarah Golf Course. Frog underpasses were created to link try to relocate the population to a safer area where the the new frog habitat area with the Marsh Street wetland. population could be increased. A Section 120 licence was subsequently granted for the Captive Maintenance of Threatened proposed development subject to several preconditions, Populations the main one being that self-maintaining populations Captive populations based on the above two wild populations based on Rosebery stock be established both in captivity have been established at Taronga Zoo in the Sydney suburb and in the wild. Bell frogs were collected from the Rosebery of Mosman. Both of these programs are discussed separately Site and a captive-breeding colony was established at below. See also McFadden et al. (2008). Taronga Zoo in late 1993. The development consent also allowed for re-development of parts of the site away from Rosebery population the frog pond while the search for a suitable relocation site was under way. Searches for suitable relocation sites were Attempts to establish a captive breeding population of bell carried out within the eastern suburbs of Sydney, close to frogs based on Rosebery stock began in late 1994 with two previously known bell frog habitats. Several areas were translocations of adult frogs from the Rosebery site to the assessed as potential habitat areas but finally (in 1995) Zoo. In October 1994, 5 pairs of bell frogs were collected one site, Sir Joseph Banks Park at Botany, was selected as and taken to the Zoo. In December 1994, an additional the best available. two adult frogs were taken to the Zoo. Only adult frogs were collected as these animals were to form the basis of a Ponds and other frog habitats were created at Botany (see breeding colony and tadpoles were required within the next below) and the translocation of captive-bred bell frogs year to stock the selected relocation sites. No tadpoles or from Taronga Zoo commenced in 1996. Subsequently, immature frogs were collected from Rosebery because of the the development condition that a self-maintaining heightened risk of undetected disease transmission wild population derived from Rosebery stock had to be established was abandoned. Despite the loss of Subsequently these frogs and their progeny have been support by the developer and the lack of success of the maintained at Taronga Zoo under a variety of conditions. initial translocation, subsequent translocations and site Incoming frogs were quarantined for a period of one modifications were carried out at Botany by the authors, month before being assessed by Zoo vets. A separate Taronga Zoo and Botany Council. holding room was established in the Serpentaria building at the Zoo. Frogs were maintained in two types of Arncliffe population enclosures; glass aquaria kept in a controlled climate room The Arncliffe translocation was carried out as part of the in the Serpentaria or in a wire-meshed outside enclosure development consent for the M5 East Motorway in southern that contained four small in-ground fibreglass ponds. At Sydney. The proposal to extend the motorway from Beverly least six pairs of frogs were retained in the frog room in Hills to Kyeemah meant that the transport corridor would the Serpentaria at all times. The remainder was kept in pass through part of the Marsh Street Wetland at Arncliffe, the outside enclosure. Australian October 2008 Zoologist volume 34 (3) 251 White and Pyke Rosebery bell frogs are still being maintained at Taronga Case histories for recipient sites Zoo (at the expense of the Zoo) as a source of animals for Captive-bred tadpoles from the Rosebery source population other translocation sites. Breeding of additional bell frogs maintained at Taronga Zoo have been released into sites at from the Rosebery founder frogs has been successful at Botany, Long Reef and Marrickville. Captive-bred tadpoles the Zoo with tadpoles being made available each year for from the Arncliffe (Marsh Street) source population have various projects, including Long Reef and Marrickville. been released into an adjacent site at Arncliffe. Each of Arncliffe population these recipient sites is discussed separately below. In November 1999, six bell frogs were collected from a. Arncliffe the Marsh Street wetlands (by A.W.White) and taken Two large ponds measuring 25 m long and 20 m wide to Taronga Zoo. Unlike the Rosebery bell frogs, all of were constructed in 1999 adjacent to the Kogarah Golf the Arncliffe frogs were maintained in glass aquaria in a Course (Fig. 2). The ponds and the surrounding area controlled environment room in Serpentaria. The frogs were of grassland were fenced off using cyclone wire fencing housed as pairs and when male frogs began calling, rain bars to exclude foxes and people from the site. The ponds were installed in the aquaria to help stimulate breeding. had a maximum depth of 1 m and had a shallow area Spawning first occurred in January 2000 and the first at one end of the pond only. The ponds were designed consignment of tadpoles for release in newly created in this fashion to make them less suitable for species frog ponds at Arncliffe was carried out in March 2000. such as Striped Marsh Frogs Limnodynastes peronii. Bell No additional frogs were collected from Arncliffe and frogs were still present in the Marsh Street wetlands and the breeder population was increased using the first and quickly colonised the new ponds. second-generation progeny. A second release of captive- bred tadpoles was carried out in February 2001. No further releases were carried out as monitoring indicated that the adult bell frog population had increased from about 10 individuals in 1999, to more than 70 adult frogs in 2002. Selection of recipient sites and individuals for release A site was considered suitable for the attempted introduction of bell frogs if known habitat requirements of the species were apparently (or potentially) satisfied in terms of breeding, foraging and sheltering (White 1998). Ponds were considered to provide suitable breeding habitat if they contained still, open and largely unshaded Figure 2. Bell frog ponds at Arncliffe and associated water that fluctuated significantly in level, some emergent boulder fields, May 2000. Photo, A. White. aquatic vegetation, and no Plague Minnow Gambusia The two frog ponds were completed in late November holbrooki or other predatory fish (Pyke and White 1996, 1999 and vegetation was well established in and around 2001). The Plague Minnow is known to feed on eggs and the ponds by the beginning of 2000. Extensive boulder tadpoles of bell frogs (Morgan and Buttemer 1996; Pyke fields were created around and between the ponds (Fig. and White 1999) and is considered a significant threat 2). The purpose of the boulder fields was to provide large to the survival of this and other frog species. Adjacent areas of over-winter habitat in protected areas close to the or nearby areas were considered to provide suitable ponds. In early 2000, construction work for the motorway shelter habitat if they contained thick, low terrestrial portals meant that parts of the Marsh Street Wetlands vegetation or rocks under which frogs could retreat. Areas would be lost. Frog searches were carried out in the of generally low vegetation were considered to provide affected areas and three bell frogs were captured, tagged potential foraging areas for these frogs. and translocated to the new ponds on the other side of the At all recipient sites tadpoles were the preferred life stage motorway, a distance of about 150 m. for release. Growth and development of tadpoles in the Night monitoring detected the presence of five non- ponds would verify their suitability for this stage in frog translocated bell frogs in the new ponds (as well as other development. Individuals that have grown up in a particular frog species) in January 2000. Bell frogs spawned in the pond were considered more likely to consider that pond as new ponds in late January. Fifty bell frog tadpoles were ‘home’, and hence to remain there, than individuals that had transferred from Taronga Zoo to the Arncliffe ponds in already metamorphosed into frogs when released (Cooke March 2000. By this time, eight adult bell frogs were and Oldham 1995). Furthermore, captive maintenance present in the ponds, previously-laid tadpoles had reached of frogs after metamorphosis is much more expensive and metamorphosis and 20 juvenile frogs were observed. time consuming than maintaining them as tadpoles (A.W. By late May 2000, no bell frog tadpoles remained in White & G.H. Pyke, pers. obs.). Nonetheless relatively the ponds; many had metamorphosed but others were small numbers of both immature and adult frogs have been observed to be eaten by adult bell frogs and Striped Marsh released at the various sites (see below). Frogs, as well as by Eastern Water Skinks Eulamprus quoyii. Australian 252 Zoologist volume 34 (3) October 2008 Frogs on the hop Juvenile bell frogs were found around the new ponds throughout winter and by spring 14 adult bell frogs were present at the site, of which 10 were first year adults. The resident bell frog population spawned again in November 2000 and late January 2001. An additional release of 150 bell frog tadpoles occurred in February 2001. No further releases of captive-bred tadpoles occurred because the resident bell frogs appeared to be generating sufficient offspring without external assistance. Population estimates based on mark-recapture data between 2004 and the present indicates that the adult bell frog population exceeds 50 animals. This population appears to be secure. b. Botany Figure 3. Ephemeral bell frog pond at Sir Joseph Banks The Botany Site was on reclaimed land along the northern Park, Botany, March 1997. Photo, A. White. foreshore of Botany Bay (part of Sir Joseph Banks Park; The two ponds proved to be suitable for tadpoles and 332830 E, 6241610 N). The park was created during the by April all of the tadpoles had metamorphosed. Large 1980’s when the Botany Bay Port facility and associated numbers (up to 40 in a single evening) of juvenile frogs access roads were being created. Dredge spoils excavated were detected around the ponds in May and early June. to make the channels and berth areas in the shipping During winter no frogs were found and none were observed container dock area were used to create a new shoreline in the following spring (A.W. White, pers. obs.). that extended from Matraville to Botany. At Botany, an The failure of this translocation was evaluated on the area 100-150 m wide was reclaimed and this became the basis of the limited monitoring data. The ponds and thoroughfare for Foreshore Drive as well as the western associated habitat were clearly suitable for tadpole growth extension of Sir Joseph Banks Park. Although several and provided foraging areas for young frogs, but not for ponds already existed within the park, none of them their survival beyond the winter. A possible reason for the were particularly suitable for bell frogs because Plague failure of the young frogs to survive the winter may have Minnows was widespread in the ponds. A fish-free pond been that they were too small and had insufficient body fat site was required that was distant from Plague Minnows- to last the non-feeding periods over winter. If the tadpoles containing water bodies. The best location for the bell frog had been introduced into the ponds before March, the ponds was the western end of the park where there were young frogs would have had a longer feeding period in few other ponds and public visitation was minimal. which to grow and may have been larger and more able In early 1996 excavators moved into the park to begin to withstand the rigours of winter The simplest method to construction of the first specifically-made bell frog test this notion was to carry out a second translocation of ponds in Australia. Two large and two small ponds were tadpoles at an earlier date, in late spring or early summer. excavated; the two larger ponds were excavated below the This was attempted the next season. water table and always contained water whereas the two smaller ponds were above the water table and only held 1997 – Second Botany Translocation water intermittently (Fig. 3). Bulrushes (cumbungi) Typha Despite intentions the second translocation could not be orientalis naturally established in the pond and by the end carried out under the same conditions as the first. During of the first year there were two large stands of bulrushes the summer of 1995/96 the new frog ponds at Sir Joseph at either end of the two large ponds. The area around the Banks Park became infested with Plague Minnows. These ponds was replanted; all bushes and trees were removed exotic fish are known to kill the eggs and tadpoles of Green within a 20 m radius of the ponds and native grasses and Golden Bell Frogs (Morgan and Buttemer 1996; Pyke (Danthonia sp. and Themeda australis) planted. and White 1999) and are listed as a Threatening Process under Schedule 3 of the Threatened Species Conservation 1996 – First Botany Translocation Act 1995. It is not known how the fish got into the pond The captive-breeding program at Taronga Zoo had proven although other ponds in the park contain the fish. Before very successful and in March 1996 the first translocation the second translocation could be carried out the fish had was carried out: 500 advanced bell frog tadpoles and 50 to be reduced or eliminated. One week before the second juvenile frogs were released into the two large ponds at translocation was planned to take place the pond was Botany. At about this time, events had taken place that hand-netted daily. Over 500 fish were removed but some allowed the developer to be released from some of the (less that 50) remained. conditions of the original licence; the translocation had In February 1997 a second translocation of tadpoles and become the determinant for the destruction of the frog young frogs was carried out. The result was an immediate pond at Rosebery. As no provision had been put in place disaster. The remaining Plague Minnows quickly killed the for monitoring the translocated population, the only tadpoles (A.W. White, pers. obs.) and within a month of follow-up monitoring was done on an opportunistic basis the release no tadpoles or juvenile frogs (if metamorphosis and was carried out by interested biologists. had occurred) could be found. Australian October 2008 Zoologist volume 34 (3) 253 White and Pyke 1999 – Third Botany translocation This site was chosen as a release site for bell frogs in 1997 when suitable breeding habitat for this species seemed to By the time it was apparent that the Botany pond had have been inadvertently created through the creation a failed, the Rosebery source pond had been destroyed and series of new ponds through the golf course. These ponds the population was now only represented by the captive were developed in such a way that they would receive colony at Taronga Zoo. The Sir Joseph Banks Site was runoff from the adjacent road and urban area with the no longer useable as a potential translocation site while goals of providing freshwater aquatic habitat for plants Plague Minnows were present. Meanwhile, pilot studies and animals, providing a source of irrigation water for (A.W. White, unpubl. data) had suggested that tadpoles use on the golf course and improving the quality of the of the Striped Marsh Frog could reduce the survival of water before it flowed out to the ocean. Originally there bell frog tadpoles in ponds that they cohabit. A system was a natural wetland in about the same area as where was needed that could exclude or reduce the number of these ponds were being developed but this was apparently Striped Marsh frogs from the bell Frog pond while bell destroyed when the golf course expanded from nine to 18 frogs were becoming established. holes in about 1930. In June 1999, Taronga Zoo undertook a series of habitat In 1997, two further ponds were developed specifically to alterations at Sir Joseph Banks Park. Rotenone and provide further habitat for the bell frog. The two ponds, chlorine, both potent fish poison were used to eradicate though smaller than those previously created, contained Plague Minnows from the ponds and frog barrier fences apparently suitable breeding habitat and were surrounded were constructed around the pond area. A monitoring by low vegetation. In 2003, another two ponds were program using local school children was devised and developed in response to the discovery that existing ponds in November the third translocation occurred with were too cold to support growth and development of bell 1,500 tadpoles and 50 juvenile frogs released. Tadpole frog tadpoles. These ponds were subsequently found to metamorphosis took place rapidly and the juvenile frogs contain suitably warm water. were well developed and approached adult proportions by The translocation program for bell frogs at Long Reef the onset of winter (T. Russell, unpubl. data). However, began in January 1998 with the first release of captive- the frog exclusion fencing was not successful in keeping bred tadpoles at the site and the commencement of out competing frogs and other species became established regular monitoring of the population and factors that alongside the bell frogs. No adult bell frogs survived might affect it. This first release consisted of about 2400 through the next spring. tadpoles which were introduced into each of seven ponds The loss of frogs over the winter period strongly suggested on the golf course, including the two newest frog ponds, that shelter habitat may have been inadequate around the one oldest golf pond and four of the ponds created the ponds. In late 1999, large sandstone boulders were during the wetland development. There have been further placed around the banks of the largest pond. In January releases since then. Between January 1998 and March 2000, 1,000 tadpoles were released into this pond and the 2004, total releases of bell frogs at Long Reef have amounted to about 9,300 tadpoles, 70 immatures and 5 other large pond. Sixty juvenile bell frogs were present adults (Table 1). Because of limited numbers of tadpoles around the two ponds in April of 2000. Unlike previous available for release and the effects of Plague Minnows in translocation attempts, several adult bell frogs survived many of the ponds, most of the releases following the first the winter (which was cold but wet) and emerged the have been restricted to the initial two frog ponds (see Pyke next spring. Adult frogs remained at the site until the next et al. 2008 for details). summer when they were no longer detected. A monitoring program was carried out every 1-2 weeks c. Long Reef during the warmer months of the year and about every 4 The site at Long Reef consisted of Long Reef Golf weeks during the remainder of the year. The monitoring Course and some adjacent areas. It is located on a initially included tadpoles and later life-stages of frogs, headland that lies between the Pacific Ocean and but subsequently also included fish when they were a residential area in the Sydney suburb of Collaroy discovered to be present soon after monitoring began. (343130 E, 6265240 N). The bell frog translocation Tadpoles and fish were monitored with daytime sweeps and monitoring program and release sites are discussed of a net at fixed points around each pond and through in more detail in Pyke et al. (2008). visual counts of animals visible at night from the pond Table 1. Summary of circumstance and results of translocation projects. Mean no. Total Mean no. Mean no. of Mean no. Bell frogs Rock Recipient tadpoles no. of On-site Chytrid Foxes metamorphs immatures of adults present shelter site released tadpoles breeding detected present observed observed observed nearby available per year released Arncliffe 100 230 >50 >25 >20 Yes Yes No No Yes Botany 500 2,500 >100 >25 <10 No No No Yes Yes Long Reef 1300 9,300 8 44 11 No No No Yes Yes Marrickville 22 162 14 13 12 Yes No Yes No No Australian 254 Zoologist volume 34 (3) October 2008 Frogs on the hop edges with a strong spotlight. Frogs were monitored through night-time searches within and adjacent to each pond and looking under artificial shelter boards positioned around each pond. Captured frogs were individually ‘marked’. Conditions in and around the ponds changed through the course of the project. As a consequence of growth of vegetation within the ponds, there was an increase in the amount of emergent aquatic vegetation in some of the ponds, especially in the two frog ponds. This necessitated some removal of aquatic vegetation from one of the frog ponds in order that some open water remain. Rock piles were added to the areas adjacent to the two frog ponds to provide more potential shelter for frogs. The two frog ponds, which have been stocked with bell frogs, were also Figure 4. Bell frog pond at Marrickville, March 1998. Photo, colonised by the Striped Marsh Frog. A. White. was lined with a waterproof mat. Sandstone blocks were Some habitat areas at Long Reef were initially adequate use to surround the pond. Pots containing the emergent for the development of bell frog tadpoles into young frogs, rushes Typha sp. and Eleocharis sp. were submerged in the although tadpole releases in autumn were unsuccessful. pond. Frog-exclusion fencing was set up around the outer Foraging and shelter habitat were sufficient to permit these perimeter of the nursery to exclude other frog species from frogs to develop to maturity. Male bell frogs were heard entering the site and to prevent Bell frogs from escaping calling in 1999 and 2000, but breeding did not taken place. the site; an area of 600 m2 was enclosed. Half of the With increases in aquatic vegetation, water temperature nursery site was allocated to seedling propagation, pots became too low to support tadpole growth and development. containing young plants were set out on prepared gravel Insufficient tadpoles were available from the Taronga Zoo beds and were regularly watered by timed sprinklers; the breeding program for release into additional warmer ponds remainder of the site was vegetated and consisted of a and the population has died out. mixture of native plants and exotic grasses. d. Marrickville Local residents were enlisted to help with the monitoring This site was located in the densely settled residential of the bell frog population that was to be established in suburb of Marrickville, in the grounds of the Marrickville the pond. Training and information days were held at the Community Nursery in the Addison Road Community site to prepare the monitoring team and to establish the Centre (329805 E, 6246950 N; Fig. 4). This site was monitoring routine. A “Friends of the Bell Frogs” group chosen as tightly controlled habitat could be created on (Marrickville Chapter) was formed. the site, whereby all other frogs and fish could be excluded In February 1998, 22 captive-bred bell frog tadpoles from the area. In addition, the site was amenable to regular (derived from Rosebery stock) were taken from Taronga monitoring as it was small in area and had no inaccessible Zoo and released in the new pond. At weekly intervals, areas where frogs could avoid detection. the tadpoles were caught and measured. Once the Unlike other translocation projects, the Marrickville tadpoles had metamorphosed, monitoring sessions were project was not aimed at creating a long-term habitat area reduced to fortnightly. When the juvenile frogs reached for the species. This project was a short-term experiment snout-vent lengths greater than 40 mm they were micro- designed to investigate factors that may impact on urban chipped with a passive induction transponder tag (Christy bell frog communities (White 2006). 1996). In this way growth records and movement around the site could be recorded for each frog. It appeared likely that the conservation of non-captive, translocated bell frogs would depend on the preservation Of the 22 tadpoles released 14 became adult frogs. Eight or creation of a range of secure habitats. One way of tadpoles were observed being attacked and eaten by achieving this outcome could be through “back-yard” Kookaburras Dacelo novaeguineae and Sacred Kingfishers conservation where frog ponds and frog habitat are created Todirhampus sancta. Once it was realised that these birds in urban residences. At the time of the development of were preying on the tadpoles, bird-scaring flutter tapes were this project, it was not known how much space would be strung across the pond and predation by birds ceased. required to create viable frog habitat and what additional The 14 young adult frogs survived the 1998 winter and by impacts may exist in urban settings. mid-August the first calling by male frogs was heard. During In 1997, Marrickville City Council made available a small the summer of 1998/99, calling occurred regularly but no portion of land in the community nursery in the Addison matings took place. An examination of the female frogs Road Community Centre for the creation of a “back-yard” revealed that none became gravid during this first year. frog pond (Caiwood 1997). The Georges River Catchment In early 1999, two adult bell frogs were killed and one Management Committee provided the funds to prepare was seriously wounded by Black Rats Rattus rattus (Fig. the site. A 4.5 m diameter pond was constructed which 5). Black Rats were observed attacking adult frogs and had a maximum depth of 50 cm. The base of the pond rat traps were set out and the rodents were quickly Australian October 2008 Zoologist volume 34 (3) 255 White and Pyke comm.). By the end of June only one bell frog remained alive. Other Striped Marsh Frogs invaded the pond and began breeding there. As exposure to diseased frogs was likely to be a problem for any “back-yard” frog population a simple test was carried out to try to protect it. Dr Michael Mahony at Newcastle University, in undertaking tadpole growth experiments with varying salinity levels, had a Chytrid outbreak that wiped out individuals except those exposed to higher NaCl concentrations (M. Mahony pers. comm.). Also a number of field-workers have noted that most remaining bell frog populations are close to the coast and perhaps within the reach of a saline influence. Consequently this has led to a perception that perhaps Chytrid spores are sensitive to certain levels of salinity. During the spring Figure 5. Injured bell frog rescued from Black Rat, of 2000, the frog-exclusion fence was reinstalled and Marrickville, January 1999. the frog pond was salted using commercial NaCl. The eliminated. No further deaths resulted from predation. salt concentration was brought to the equivalent of 3% One adult frog managed to escape the confines of the sea-water and this was tested regularly and adjusted as nursery and was found dead (run-over) the next day required (White 2006). Bell frog tadpoles are known to about 350 m from the frog pond be able to survive salinity levels of up to 10% and at least 5% sea-water without any obvious ill effect (Mahony In February 1999, 100 new tadpoles were added to the unpublished; Christy and Dickman 2002). Attempts to pond. Again, the tadpoles were caught and measured collect and remove all of the Striped Marsh Frogs from the weekly until metamorphosis, when monitoring sessions nursery area were unsuccessful. became less frequent. Eighty-four of these tadpoles reached metamorphosis but few survived more than a In December 2000, 40 bell frog tadpoles were released into few days on land. Cannabalism by the adult bell frogs the pond (which also contained tadpoles of the Striped was high and a number of juvenile frogs were observed Marsh Frog). By late January 2001, the first juvenile being eaten. Adult frogs were also observed to feed on bell frogs emerged. Only 7 bell frog tadpoles survived to large tadpoles. By May 1999, there were 24 adult bell metamorphosis (the others died in the pond). None of frogs present (11 from the initial release and 13 from the the frogs developed symptoms of chytridiomycosis. These second release). In August calling was heard from males young bell frogs were subsequently preyed upon by the from both age classes. No breeding had taken place by adult Striped Marsh Frogs and the population perished. In the end of September. March 2001, the Marrickville translocation experiment was terminated. Calling continued through spring and the first breeding event occurred in November 1999. Female bell frogs Discussion from the first tadpole release were gravid whereas female frogs from the second tadpole release were not. Criteria for success Over 3000 tadpoles were present in the pond by late November but by the end of December, this number Each of the translocation projects that we have described had fallen to about 200. Cannibalism of the tadpoles had distinct goals for habitat creation and population and emerging froglets was observed at each monitoring establishment: habitat components that were assessed night. Only two froglets survived to become adult included the establishment of foraging, shelter, over- frogs. These froglets did not remain around the central winter and breeding habitat; while the population goal frog pond but inhabited dense patches of grass and was to establish a self-sustaining, viable population of bell Lomandra, about 10 m away from the pond. frogs at each site. Using these criteria, only the Arncliffe site achieved all of the habitat requirements and also In April 2000, maintenance work at the nursery site achieved the population goal. The other three sites failed required that the frog-exclusion fence be removed for a in one or more of these criteria (Table 2). period of about 5 weeks. Within a week of the removal of the fence, two Striped Marsh Frogs were found in the Factors affecting the success of translocated pond. Shortly afterwards, more Striped Marsh Frogs were Green and Golden Bell Frogs found in the pond. In early May an adult bell frog was found dead. Over the next three weeks five other corpses The bell frog should be an ideal candidate species for were found and these were forwarded to Taronga Zoo for translocations because it appears to be “pre-adapted” autopsy. The autopsies revealed that the bell frogs were to colonising artificial habitats (Pyke and White 2001). infected with chytrid (Batrachochytrium dendrobatoides), a The failure of translocated frogs to become established at highly infectious micro-fungal disease of frogs (Berger et recipient sites is a strong indicator of the failure to provide al. 1998). A Striped Marsh Frog was caught and killed and suitable habitat. Factors such as predation, disease and forwarded to the Zoo for autopsy. It also contained the inadequate over-winter habitat had a major influence on skin vesicles that characterise this disease (K. Rose pers. the outcomes of the four projects described above. Australian 256 Zoologist volume 34 (3) October 2008 Frogs on the hop Table 2. Outcome of four translocations of Green and Golden Bell Frogs in Sydney Tadpoles reach On-site breeding Likelihood of population Recipient Site Duration of project (years) metamorphosis occurred survival Arncliffe 1999-2008 Yes Yes High Botany 1996-2001 Yes No Extinct Long Reef 1998-2004 Yes No Extinct Marrickville 1997-1999 Yes No Extinct The range of outcomes from the four translocations apparent inability of bell frogs to compete with some frog demonstrates quite dramatically how population survival species may account for the regular, long-range dispersal and occupation of habitat areas are influenced by a events observed for this species (Pyke and White 2001). variety of environmental and ontogenic factors. The Dispersal is particularly high in recently metamorphosed Marrickville translocation is an example of a heavily frogs and it appears that mortality rates are also high controlled translocation where potential predators and during dispersal. At Marrickville, where juvenile frogs competitors were excluded. Under these circumstances, were unable to disperse, juvenile survivorship was high very high survival rates were achieved for tadpoles when adult frogs were absent. Interactions between frogs reaching metamorphosis and for juvenile frogs reaching may also be determined by susceptibility to disease. adulthood. This is in stark contrast with most other sites The advent of new diseases is also a compounding where there were significant losses of bell frogs at the time factor in the success of translocated populations. The of metamorphosis and subsequent losses of adults. Marrickville population was the only population studied Predators of Green and Golden Bell Frogs in the wild have where exotic disease (chytrid) definitively affected the been described (Pyke and White 2001). These generally population, in this case it almost completely exterminated range from snakes, to wading birds and fish. However, for the population. Chytrid was not detected in the other, translocated populations in urban or near-urban areas less-confined populations and the chances of finding the predators were quite different. At Marrickville, most dead or dying frogs was lower. Unconfined populations predation was caused by adult bell frogs or Striped Marsh also have an advantage in that the spread of a disease Frogs eating juvenile frogs or tadpoles. Black Rats were through a population may be less comprehensive as some the main exotic predator of adult bell frogs while native individuals may not encounter infected individuals or birds were the main predator of tadpoles. At Arncliffe infected habitats. (A.W. White unpub. obs.) foxes and eels appear to be the The addition of a weak salt solution to a chytrid- main predators of adult bell frogs. infected pond suggested that this treatment is not harmful Plague Minnows were present at Botany and Long Reef. In to tadpoles or frogs and may confer some degree of both cases their presence had a marked influence on the protection against Chytridiomycosis (White 2006). The survival of bell frogs. At Botany, the second translocation use of weak salt as a preventative measure needs to be attempt failed completely due to the presence of these fish fully investigated but appears to have the potential to while at Long Reef, tadpoles metamorphosed to produce become a standard treatment for all bell frog ponds. frogs only at ponds where Plague Minnows were absent. Behavioral traits of bell frogs may also play a role in the Plague Minnow predation on bell frog eggs and tadpoles outcome of the translocations. The early translocations has been previously noted (Morgan and Buttemer 1996; of bell frogs at Botany were deemed unsuccessful because Pyke and White 2000). juvenile frogs were not found after winter. This was Fox predation stands out as a factor that could limit interpreted as a failure to provide sufficient over-winter the success of translocation projects involving bell frogs habitat for the young frogs. However, the construction of but further examination of this possibility is required. a frog-exclusion fence in 1998 resulted in the retention of If significant over-winter frog mortality arises from fox juvenile frogs at the release pond. In hindsight, it appears predation, then it should be possible to find some evidence that bell frog translocation areas require more than one of this through analyses of fox scats or stomach contents. functional pond and dispersal areas for juvenile frogs to Unfortunately, however, it is possible for such predation survive in until they are large enough to safely return to to be highly significant in terms of a relatively small frog the breeding ponds (where there is usually a high density population and yet represent only a very small component of attending male frogs). Goldingay and Lewis (1999) of fox diet in the area. recommend the establishment of multiple ponds as a safeguard against fish invasion and other adverse impacts. The impact of competition from other frog species has not been adequately demonstrated for Green and Golden The colonization of frogs, such as occurred at Arncliffe, Bell Frogs. The reasons for the inability of bell frogs to be suggests that translocation methods may be able to exploit successfully reintroduced into the Marrickville Site after the juvenile dispersal trait to advantage. At Arncliffe, the arrival of Striped Marsh Frogs are unclear. In a study of juvenile frogs rarely stayed near the initial breeding ponds natural frog colonisation of ponds, Pyke and White (1999) but quickly colonised nearby sites. Regular recaptures found that bell frogs could be lost from stable ponds of juvenile frogs at these sites suggests that once the when other frog species had become established. The frogs have found a suitable new habitat they are likely to Australian October 2008 Zoologist volume 34 (3) 257 White and Pyke remain there as adults. On this basis, if a new bell frog and it has the extra advantage of simulating a major pond is created near the source population, it is likely to fluctuation in water level. If salt is to be added to ponds be passively colonized (Goldingay 1996) and the frogs that as a preventative measure against micro-fungal diseases, take up residence appear more likely to remain at this site the ability to drain the ponds is a great aid in regulating as adults. A frog pond constructed at Boilermakers Point, salt levels. Finally, when ponds become too overgrown near Port Kembla was 200 m from a known bell frog site. with emergent plants, ponds can be quickly and cheaply The pond was colonized by adult bell frogs within months drained, and the vegetation allowed to die back to and bred there over several years (Goldingay and Lewis preferred levels. 1999). Lack of maintenance of Typha subsequently let it Important issues in translocation become largely unsuitable (Goldingay and Newell 2005). Translocations should never be accepted lightly as a Further understanding of the relative roles of various means to achieve species conservation. Translocations are factors in determining the degree of success of bell risky and there is a high risk of failure. Even for species frog translocation projects may also arise through an like the Green and Golden Bell Frog whose gross habitat extension of the present review to include some of the requirements have been well studied and documented other similar translocation attempts that have been made. (Pyke and White 2001), success in translocation cannot Since these projects have all involved habitat creation or be guaranteed. enhancement, their consideration should include these processes as well as the translocation itself. In addition, there are conservation concerns about the possible impacts on species that are translocated. Greer Design of bell frog ponds (1996) lists several major concerns that are predicated The bell frog ponds used in the Marrickville and Arncliffe on the presumed knowledge of those undertaking the translocation were based on breeding habitat requirements translocation. The translocation of a species in space detailed in Pyke and White (1996). In particular, breeding may have serious implications in terms of genetic and ponds needed to have a deep and a shallow area, needed disease exchange with resident frog populations, and to have fluctuating water levels and the capacity to be because of this change of circumstance, the evolutionary drained, had surrounding rock piles, emergent water path of the population (species) may be diverted into a plants at one end and areas of open water. different direction. Such dramatic genetic impacts are only likely if bell frogs are relocated to areas distant from Several structural factors appear to be important in the source populations (Burns and Ferraria 2004). On determining the success of a bell frog pond. One structural a more pragmatic note, translocations are an expensive factor that appears to be important is the provision of undertaking. Significant costs are associated with the large rock fields (shelter and over-winter habitat) around creation of alternative habitat areas, ongoing monitoring the frog ponds. By providing these shelter areas close to and reporting, habitat management and maintenance. the ponds, there is less potential for predation of frogs In the case of translocations undertaken because of a moving between the ponds and shelter sites. Rock piles commercial development, there is often a considerable were created next to the ponds at Arncliffe and at Botany. time lag between the translocation of frogs and the In both cases, adult and juvenile bell frogs have been approval to destroy existing habitat. In NSW, the NPWS observed under the rocks. Rock fields may provide a high requires that recreated habitat must be demonstrated to density of refuge sites within a short distance of a pond. be successful before original habitat is destroyed. The If shelter sites are provided further from the pond, the ultimate proof of the suitability of the new habitat is the chances of predation (by foxes and rats) of moving bell successful breeding of the frogs in both the parental and frogs may increase. Dense vegetation cover can also be F1 generations. As female bell frogs take 18 months to used (such as at Marrickville) but this type of shelter has sexually mature, this requirement means that a time delay a drawback, the soil underneath can dry out much faster of between 3 and 4 years must elapse before it is possible than the soil under large rocks. Similarly, predators cannot to meet the breeding requirement. move rocks in search of frogs. One of the most important requirements for translocation Another structural feature is the provision of drains and is the need for regular long-term monitoring of the external water supplies (such as at Arncliffe). These translocated population. Only through long-term features allow the level of the pond to be raised and monitoring will it be possible to determine population lowered at will. The raising of the water level of a pond response to the new habitat and to assess population in early spring after a period of low water during winter fluctuations. Once these baseline data are established, was believed to trigger the early colonisation of a pond by any subsequent fluctuations can be reviewed in terms of male bell frogs. The pond could then be kept relatively full natural cycles. Changes in the frog population should be during spring and summer, making the pond available for used as the guide to direct modifications of frog habitat and opportunistic breeding over a long period. to ensure the effectiveness of breeding ponds. Monitoring The provision of drains in bell frog ponds is useful for other must be long-term as bell frog populations are not stable reasons. In the event of the ponds becoming infested with and population numbers vary greatly between and within unwanted fish or disease-carrying amphibians, the pond seasons. Ideally, monitoring should occur throughout the can be quickly drained and the contaminated water spring and summer months and should include measures discarded. Pond draining is a more acceptable method of reproduction as well as population estimation. of fish control than using fish poisons or electro-fishing Australian 258 Zoologist volume 34 (3) October 2008