Background information and definitions
Refuge habitats can provide amphibians with microclimates to keep them at the correct temperature and prevent them from dehydrating and can protect them from predation. Many amphibians seek shelter in rocks, logs or other refuges created by tree falls and other disturbances. Refuges can be created for amphibians where natural shelter habitat is limited, or to replace these habitats where they have been lost.
Other studies investigating the creation of shelter habitat are discussed in ‘Create artificial hibernacula or aestivation sites’ and ‘Biological resource use – Leave coarse woody debris in forests’.
Supporting evidence from individual studies
A before-and-after study in 1990–2000 of 12 endangered Hamilton's frog Leiopelma hamiltoni on Stephens Island, New Zealand (Brown 1994, Tocher & Brown 2004) found that three frogs survived in a created refuge within a predator-proof enclosure for at least eight years. There was no evidence of breeding by 1992 and only one juvenile was ever recorded, in 1996. Eight frogs survived the first year in the rock-filled pit and were recaptured 61 times by 2000. Two of three frogs that were not recorded at the release site after 1994, but were found back at their original habitat (76–89 m). After eight years, 42% of translocated frogs had been recaptured compared to 47% marked at the original site. In May 1992, frogs were translocated 40 m to a new rock-filled pit (72 m2) in a forest remnant. A predator-proof fence was built around the new habitat to exclude tuatara Sphenodon punctatus and the area was ‘seeded’ with invertebrate prey. Frogs were surveyed regularly from November 1990 to May 1992 (90 visits), intermittently in 1992–1996 and at least four times annually (over six days) in 1997–2000.Study and other actions tested
A randomized, replicated, controlled study in 1998–2005 of pine stands in South Carolina, USA (Owens et al. 2008) found that adding coarse woody debris to forest did not effect amphibian abundance, species richness or diversity. Plots with added downed woody debris did not differ significantly from controls in terms of amphibian abundance (1–2 vs 2), species richness (6–7 vs 7) or diversity (17 vs 19). One species, the southern leopard frog Rana sphenocephala, had lower capture rates with the addition compared to removal of woody debris (0.02 vs 0.11/night). Treatments were randomly assigned to 9 ha plots within three forest blocks. The first set of treatments was undertaken in 1996–2001 and a second set in 2002–2005. Woody debris was increased five-fold. Control plots had no manipulation of woody debris. Five drift-fence arrays with pitfall traps/plot were used for sampling in 1998–2005.Study and other actions tested
A before-and-after study in 1997–2004 of a golf course with degraded woodland and grassland in Sydney, Australia (Burgin & Wotherspoon 2009) found that restoration that included reintroducing coarse woody debris to the woodland floor increased frog species over two years. Frogs increased from seven to 10 species in the first year and then remained stable for the following six years. A total of 18 species of frogs were predicted in the area and so 56% were present following restoration. The golf course was developed in 1993 and restoration was undertaken in 1997–2001. Coarse woody debris was reintroduced onto the woodland floor, endemic shrubs and trees were planted and non-native weeds were removed. The mowing regime was changed to develop grasslands and a narrow band of herb vegetation retained around ponds as a buffer zone. Pond perimeters were walked to record frog calls in 1996–2004.Study and other actions tested
A replicated, controlled, before-and-after study in 2007–2008 of a cacao plantation in Sulawesi, Indonesia (Wanger et al. 2009) found that adding woody debris and/or leaf litter to plots had no effect on overall amphibian abundance or species richness. However, following addition of woody debris plus leaf litter, Hylarana celebensis abundance increased and Asian toad Duttaphrynus melanostictus decreased. Forty-two plots (40 x 40 m2) were divided into four treatments: addition of woody debris (trunks and branch piles), addition of leaf litter, addition of woody debris plus leaf litter and an unmanipulated control. Monitoring was undertaken twice 26 days before and twice 26 days after habitat manipulation. Visual surveys were undertaken along both plot diagonals (transects 113 x 3 m).Study and other actions tested
Referenced paperWanger T.C., Saro A., Iskandar D.T., Brook B.W., Sodhi N.S., Clough Y. & Tscharntke T. (2009) Conservation value of cacao agroforestry for amphibians and reptiles in South-East Asia: combining correlative models with follow-up field experiments. Journal of Applied Ecology, 46, 823-832.
A replicated study in 2007–2011 in Missouri, USA (Briggler & Ackerson 2012) found that artificial shelters were used by hellbenders Cryptobranchus alleganiensis in the wild and captivity, but breeding was limited. Six hellbenders used five of the seven shelters in the wild in 2010–2011. One clutch of 182 eggs was found being guarded within one shelter. In captivity, many shelters were used by hellbenders, but only one clutch of eggs was recorded. Artificial shelters were constructed from chicken wire covered with concrete (chamber: 41 x 37 cm). Six prototype shelters were installed in a riverbed in winter 2007–2008. A couple of these attracted females but no eggs were laid. Following modifications, seven L-shaped shelters were installed in a river in June 2010 and 20 in a captive enclosure in August 2011. ‘Wild’ shelters were checked in July and November 2010 and October 2011 and captive shelters were checked weekly.Study and other actions tested