Sunset Frog

Population and DistributionSpicospina flammocaerulea was discovered in 1994 (Roberts et al. 1997). When first described in 1997, the species was only known from three well separated peat swamps in the s.w. corner of WA (Roberts et al. 1997). However, survey work undertaken from 1997 to 2000 raised the number of known populations to 27, all occurring near the WA s. coast, e. and n.e. of Walpole (Roberts et al. 1999; Burbidge & Roberts 2001). This species has a limited area of occurrence (approximately 305 km2), small area of occupancy (135 ha) and fragmented range (Roberts et al. 1999; Burbidge & Roberts 2001). No reliable data on population size is available, however, counts of males have been recorded at several sites from 1994-1997 (Roberts et al. 1999). Surveys of calling males usually report less then ten individuals, however 150 males were estimated to be present at Trent Road (Bow R.) in 1997 (Roberts et al. 1997). An apparent decline in the number of calling males has been recorded at Mountain Road (n. and s.) where 120 males were observed calling in 1994 and three years later only 2 males were recorded (Roberts et al. 1999). The actual population size at sites with few or no calling males is unknown (D. Roberts pers. comm.). Two sites with a long history of visitation and no calling activity contained individuals in 2000 (D. Roberts pers. comm.).Fourteen populations are on private property n., w. and e. of Bow Bridge, with the remainder in the Mt Franklin NP or on land designated to form part of the Mt Roe-Mt Lindesay NP but not yet declared (Roberts et al. 1999; D. Roberts pers. comm.).HabitatSpicospina flammocaerulea is a habitat specialist. The region from which S. flammocaerulea has been recorded is thought to have undergone a change from a subtropical wet to a seasonally arid climate about 5 to 6 million years ago and the peat swamps where the species occurs are considered to be relicts of an earlier environment (Wardell-Johnson et al. 1996). The persistence of the species in well separated swamps is no doubt attributable to this change in environment (Wardell-Johnson et al. 1996). The species is found in isolated and permanently moist peat based swamps with organically-rich soils (Roberts et al. 1997), in a high rainfall area of moderate relief with granite outcrops and associated ranges of hills rising to 300-400m (Roberts et al. 1999). These sites have a high moisture content in the soil and are protected from climatic extremes, often by local seepages that maintain water availability uncharacteristically into spring and summer (Roberts et al. 1997).

ReproductionMales call between Oct. and Dec. from shallow pools, water seepages, large hollows containing water or in open water along creek margins (Wardell-Johnson et al. 1996; Roberts et al. 1997). Less than 200 eggs are deposited singly and may be supported by algal mats just below the waters surface (Roberts et al. 1997). The tadpole stage is presumably free swimming (Roberts et al. 1997). Explosive breeding appears unlikely as numbers of calling males have been observed to remain relatively stable over extended periods throughout the breeding season at some sites (Roberts et al. 1999). Invasive speciesExcavation by feral pigs are common in swamps close to the type locality and pigs may have a direct impact on frog survival (Roberts et al. 1997). However, monitoring of known populations and adjacent control sites from 1997-1998 has shown little indication of pig damage (Roberts et al. 1999).

An extremely small geographic range makes this species particularly susceptible to local catastrophes. An apparent decline in frog numbers at one locality (Mountain Road, Mt Franklin NP) following wildfires in 1994 suggests a possible risk from fire (Roberts et al. 1999). Frequency of fire varies between localities but the majority of sites have experienced wildfires in the last 50 years (Roberts et al. 1999). This suggests some capacity to recover post-fire but the time and conditions required for full recovery, which could set an optimal fire interval and intensity regime, are unknown (Roberts et al. 1999). Fires which burn the substrate (peaty swamps) or fire regimes which lead to a greater propensity of substrate ignition, may well be detrimental to the persistence of the species (Roberts et al. 1997). Loss of vegetation through fire or disease (such as the fungus Phytophthora) may alter soil water tables affecting both availability of breeding sites and peat formation and maintenance (Wardell-Johnson et al. 1996; Roberts et al. 1997). Fourteen populations are on private property n., w. and e. of Bow Bridge, with the remainder in the Mount Franklin NP or on land designated to form part of the Mount Roe-Mount Lindesay NP but not yet declared (Roberts et al. 1999; D. Roberts pers. comm.). There are no threats to populations on publicly owned lands that cannot be controlled by appropriate management but there has been no analysis of threats to populations found on private property (Roberts et al. 1999). Fieldwork is currently being undertaken to evaluate declines and variation in population size by assessing population size more directly using mark-recapture techniques and surveys of tadpole populations (Roberts et al. 1999).

Relatively little is known about the ecology of this species. The breeding season occurs from October to January, when males call from water seepages or in large water-filled hollows and shallow pools (3). Females lay an average of 81 eggs singly; these are often supported on algal mats just below the surface of the water. The tadpoles are free-swimming but may be benthic specialists (7).

This frog is classified as Vulnerable by the IUCN Red List of Threatened Species because of its highly restricted range. With regards to the conservation of this species, perhaps the most pressing requirement at present is to increase knowledge of its distribution and the threats facing it, allowing accurate estimates of population size and in effective conservation measures to be devised. Research and monitoring to this end is on-going. In addition, many of the sites supporting this species are situated within national parks, or state forest so the species receives a high level of protection in over two thirds of its known range (8)

The sunset frog was first discovered in Western Australia as recently as 1994 and was scientifically described in 1997. It is so called because of its striking appearance; it is dark purplish-black above, with bright orange and blue on its belly, reminiscent of the colours at sunset (3). This small frog produces a deep 'dadukk dadukk' croak (3). This species is the only member of the genus Spicospina, and differs greatly from other Australian frogs in colour (orange and blue ventral surface) and appearance (massive glands behind the head) (4).

The sunset frog occupies a very specialised habitat. It is found in permanently moist peat swamps which may be relics of an earlier wet landscape with summer rainfall patterns that changed to a seasonally dry summer climate around 10-12 million years ago. The remaining peat bogs in which this frog lives are kept moist throughout the year, often as a result of seepages of water. Within the bogs, sunset frogs are typically found in the water seepages, in pools and along drainage lines. Modern peat swamps may represent the closest approximation available to Miocene summer wet climates (6).

When this species was first discovered, it was only known from three sites in south-west Western Australia. Since then, surveys have brought the total number of likely populations to 27, all of which are located near the southern coast of Western Australia. The range of this frog is highly fragmented and it occupies a very restricted area. There is at present a lack of data on the absolute size of known populations (5).

Classified as Vulnerable (VU D2) by the IUCN Red List 2004 (1) and listed on Schedule 1 of the Western Australia Wildlife Conservation Act (2).

The very restricted and highly fragmented geographic range of the sunset frog makes it extremely vulnerable to chance events, such as disease or freak weather. However, reports that fires caused a decline in the population size at one site (8) were not substantiated when this site was burnt again in the summer of 2002-2003 and frogs bred after the fire (6). Indeed, although wildfire can ignite peat in peat swamps there is also evidence of populations persisting for long periods post fire and that fire can induce breeding activity (9). Further threats may include introduced feral pigs, which may damage breeding habitat, and inappropriate land management, such as over-grazing, killing vegetation, or inappropriate fire regimes (3) (6). Outbreaks of dieback fungus, Phytophthora cinnamomi, could also pose a significant threat to the species (6). The fact that the populations are so fragmented could result in genetic problems due to inbreeding (3), but there are no reliable data on population size or connectedness (6).