The relative importance of aquatic and terrestrial variables for frogs in an urbanizing landscape: Key insights for sustainable urban development
Introduction
As the human population increases, a better understanding of how amphibian species respond to urbanization is needed worldwide. Urban human populations will increase by 2.7 billion from 2010 to 2050 (United Nations, 2012). Therefore, development for residential purposes will continue modifying landscapes and threatening biodiversity in many regions of the world (Beninde, Veith, & Hochkirch, 2015; McDonnell & Hahs, 2015; Villaseñor, Blanchard, & Lindenmayer, 2016). Among vertebrates, amphibians are the most threatened group: about 40% of amphibian species are threatened with extinction and the average size of populations has declined by up to 80% in the last four decades (based on 357 populations of 162 species from around the world; Baillie, Griffiths, Turvey, Loh, & Collen, 2010). Given that urban development threatens about 950 amphibian species with extinction (Baillie et al., 2010, Hamer and McDonnell, 2008), how can we conserve amphibians in an urbanizing world?
Amphibians rely on aquatic environments for breeding and larval development, and on terrestrial environments during juvenile and adult stages (Anstis, 2007; Semlitsch & Skelly, 2008). Thus, attributes of aquatic and terrestrial environments can affect amphibian distributions. Key variables from the aquatic environment influencing amphibian distributions include water body size, hydroperiod (i.e., period covered by water), the presence of fish and aquatic vegetation. Larger water bodies can support higher species richness (Parris, 2006). Hydroperiod can influence the species inhabiting a wetland (e.g., in ephemeral pools; Baldwin, Calhoun, & deMaynadier, 2006; Semlitsch, 2000). Predatory fish can have detrimental effects on amphibians (Shulse, Semlitsch, & Trauth, 2013), whereas aquatic vegetation may provide refuge against predation during amphibian larval stages (Hamer & Parris, 2011).
Modification of terrestrial environments, such as clearing of vegetation for urban development, can lead to habitat loss and fragmentation (Hamer and McDonnell, 2008, Semlitsch and Skelly, 2008), and influence amphibian populations from local to landscape scales (Hamer & Parris, 2011). Important variables from the terrestrial environment influencing amphibian distribution include fringing vegetation (i.e., vegetation adjacent to waterbodies), as well as forest cover and urban infrastructure at different landscape scales. At the local scale, fringing vegetation provides refuge for metamorphs and breeding adults (Hazell, Cunnningham, Lindenmayer, Mackey, & Osborne, 2001). Forest loss reduces terrestrial habitat for adults and leads to decreased richness, occurrence and abundance of several amphibian species (Ficetola, Marziali, Rossaro, De Bernardi, & Padoa-Schioppa, 2011; Hamer & McDonnell, 2008). Increased road cover, a surrogate variable for urbanization (Hamer & McDonnell, 2008), may reduce connectivity and isolate populations by limiting dispersal and migration between breeding habitats (Eigenbrod, Hecnar, & Fahrig, 2008; Hitchings & Beebee, 1997).
Urbanization may be an important anthropogenic driver of species loss in terrestrial landscapes, because only a few species able to adapt to rapid urbanization prevail (McDonnell & Hahs, 2015). Thus, management derived from ecological studies in urban landscapes may be biased towards a few common species that are recorded in sufficient numbers to perform statistical analysis. If management of landscapes under high human pressure is based on findings from these species, it may fail to cater for the needs of species at a higher risk of decline − such as previously common species that become locally-extinct or infrequently encountered as a result of habitat modification (Gaston & Fuller, 2007).
Increased urbanization of natural and rural lands can pose a serious threat to amphibians. However, our ability to guide conservation efforts remains limited because amphibians are among the least studied vertebrate groups in urbanizing landscapes (McDonnell & Hahs, 2008). Although the number of studies of amphibians in urban areas and in landscapes under urban development is growing, planners and managers still lack information to effectively guide amphibian conservation in most urbanizing landscapes worldwide (but see Calhoun, Jansujwicz, Bell, & Hunter, 2014).
To provide conservationists, managers and urban planners with empirical insights to guide effective conservation of common and infrequently encountered amphibian species in urbanizing landscapes, we studied the distribution of pond-breeding frogs during the breeding season in a landscape comprised of forests, rural, and urban areas in southeastern Australia. Population growth and increased demand for holiday houses are triggering forest clearing for urban development in the region, but little is known about how this development affects the native fauna of the area. Currently, it is not known whether local environmental attributes or landscape context are the key factors that influence amphibian communities. The uncertain viability of approximately 96% of frog species in this region (State of the Environment 2011 Committee, 2011) demands urgent insights on the effects of urbanization on frogs to guide conservation actions. We asked: How is the distribution of pond-breeding frogs during the breeding season influenced by aquatic variables (water body size, aquatic vegetation, and presence of fish), and terrestrial variables (local habitat structure, and, at two spatial scales, forest and road cover)? In particular, we aimed to answer the following three key research questions:
Q1. Are common species (occurrence ≥36% of surveyed ponds) influenced by aquatic habitat, terrestrial habitat, or both? This knowledge will allow us to plan and manage urban areas to maintain common species.
Q2. What are the main factors (from the aquatic and terrestrial environments) influencing species richness of infrequently encountered pond-breeding frogs? If infrequently encountered frogs (occurrence <35% of surveyed ponds) are influenced by local-scale variables, conservation strategies can aim to improve local habitat condition within urban areas; whereas if they are sensitive to landscape variables, maintaining undisturbed habitat around breeding sites will be important to prevent species loss in urbanizing landscapes.
Q3. Does species richness reflect the variables influencing both common and infrequently encountered species? This is important because frog species richness has been proposed as a focus for conservation management in Australian urban environments (e.g., Hamer & Parris, 2011).
Knowledge of which variables from the aquatic and terrestrial environment influence amphibian distributions in urbanizing landscapes will help guide management and urban planning to conserve both common and infrequently encountered species. This knowledge is essential for amphibian conservation in forest ecosystems undergoing urban development worldwide.
Section snippets
Study area
This study was conducted between Nowra (34.86°S 150.60°E) and Booderee National Park (35.16°S 150.73°E), and covered approximately 600 km2 in New South Wales, southeastern Australia (Fig. 1). The study area is dominated by native eucalypt forests and wetlands, and rural and urban areas. Rural areas comprised cleared areas for livestock paddocks. Urban areas comprised several small towns of <10,000 inhabitants and the Nowra-Bomaderry urban center of ∼35,000 inhabitants (Australian Bureau of
Results
We recorded 14 frog species at 28 ponds (Table 2). All were native species and one (L. aurea) is endangered under state-level legislation (NSW Threatened Species Conservation Act 1995). Five species were recorded in a sufficient number of ponds to allow individual species analysis: Crinia signifera (71% of ponds), L. peronii (71%), Limnodynastes peronii (Lim. peronii from now on; 46%), Litoria fallax (43%) and L. tyleri (36%). Among these common species, the probability of detecting a species
Discussion
To help guide management and land use planning for amphibian conservation in urbanizing forested landscapes, we examined the relative importance of aquatic and terrestrial variables for pond-breeding frogs during a breeding season in a landscape of forests, rural and urban areas. In line of the three key questions posed at the Introduction, we found:
- (1)
The occurrence of common frogs was best explained by the local (aquatic) habitat.
- (2)
The richness of infrequently encountered frog species was best
Conclusions
What are the variables from aquatic and terrestrial environments that common and infrequently encountered species respond to? The key discovery in this study was that during the breeding season, common frogs responded to local aquatic variables (pond size and surface vegetation) whereas infrequently encountered frog species were highly sensitive to modification of the broader terrestrial environment. Managing ponds for appropriate water body size and a high percentage cover of surface
Acknowledgements
Martin Escobar and Kevin Mayes helped during fieldwork. Natalia Kandalaft helped with data entry. We thank Christopher MacGregor; Nicholas Dexter and Martin Fortescue [Booderee National Park]; Libby Shields and Alex Deura [NSW National Parks and Wildlife Service]; Kevin Petty [Forest Corporation NSW]; and the community of Shoalhaven for their support. Jeff Wood and Wade Blanchard made constructive comments on the experimental design and analysis. Ben Scheele, Claire Foster, Julie Watson and two
References (48)
- et al.
Assessing threats to pool-breeding amphibian habitat in an urbanizing landscape
Biological Conservation
(2009) - et al.
Citizen science reveals widespread negative effects of roads on amphibian distributions
Biological Conservation
(2014) - et al.
Amphibian ecology and conservation in the urbanising world: a review
Biological Conservation
(2008) Patch-occupancy dynamics in fragmented landscapes
Trends in Ecology & Evolution
(1994)- et al.
Use of farm dams as frog habitat in an Australian agricultural landscape: factors affecting species richness and distribution
Biological Conservation
(2001) - et al.
A simulation study of the number of events per variable in logistic regression analysis
Journal of Clinical Epidemiology
(1996) - et al.
Decline and re-expansion of an amphibian with high prevalence of chytrid fungus
Biological Conservation
(2014) - et al.
Anuran species in urban landscapes: relationships with biophysical, built environment and socio-economic factors
Landscape and Urban Planning
(2011) Frogs of Australia
(2012)Tadpoles of south-eastern Australia: a guide with keys
(2007)
Uninformative parameters and model selection using Akaike's Information Criterion
Journal of Wildlife Management
Census community profile
Evolution lost: status and trends of the world’s vertebrates
The significance of hydroperiod and stand maturity for pool-breeding amphibians in forested landscapes
Canadian Journal of Zoology-Revue Canadienne De Zoologie
Biodiversity in cities needs space: a meta-analysis of factors determining intra-urban biodiversity variation
Ecology Letters
Model selection and multimodel inference: a practical information-theoretical approach
Improving management of small natural features on private lands by negotiating the science-policy boundary for Maine vernal pools
Proceedings of the National Academy of Sciences of the United States of America
Conserving pool-breeding amphibians in human-dominated landscapes through local implementation of best development practices
Wetlands Ecology and Management
A resistant-kernel model of connectivity for amphibians that breed in vernal pools
Conservation Biology
Forest extent & change (version 9)
Accessible habitat: an improved measure of the effects of habitat loss and roads on wildlife populations
Landscape Ecology
Landscape-stream interactions and habitat conservation for amphibians
Ecological Applications
Unmarked: an R package for fitting hierarchical models of wildlife occurrence and abundance
Journal of Statistical Software
Cited by (35)
Mapping multigroup responses to land cover legacy for urban biodiversity conservation
2024, Biological ConservationSpatiotemporal dynamics in the roosting ecology of the green toad: Implications for urban planning and nature conservation
2024, Journal for Nature ConservationConserving nature's chorus: Local and landscape features promoting frog species richness in farm dams
2023, Biological ConservationEffects of multiple stressors on amphibian oviposition: Landscape and local determinants in central Japan
2021, Ecological IndicatorsCitation Excerpt :Such a phenomenon was also registered in other sites of central Japan (Osawa et al., 2013; Katayama et al., 2015). An increase in total road density, including roadside and urban construction areas, led to a decline of amphibian populations, which was also reported in Europe, Australia, and North America (Pillsbury and Miller, 2008; Marsh et al., 2017; Villasenor et al., 2017). Total road density within each circular area was used as an index of habitat fragmentation degree (Vos and Chardon, 1998).
Urbanization negatively impacts frog diversity at continental, regional, and local scales
2021, Basic and Applied EcologyAmphibian diversity in Polish cities: Taxonomic diversity, functional diversity and evolutionary distinctiveness
2020, Basic and Applied EcologyCitation Excerpt :Only some research has included surrounding areas in order to assess species composition along a urban gradient (Ficetola & De Bernardi 2004; Mollov 2011; MacGregor-Fors, Hernández Ordoñez, & Ortega-Álvarez 2013). However, several studies have focused on a wider geographical scale, including entire communities (Pillsbury & Miller 2008; Scheffers & Paszkowski 2012; Villaseñor, Driscoll, Gibbons, Calhoun, & Lindenmayer 2017; Ducatez, Sayol, Sol, & Lefebvre 2018). Currently, amphibian-based studies conducted in European urban areas appear limited (Ficetola & De Bernardi 2004; Löfvenhaft, Runborg, & Sjögren-Gulve 2004; Vignoli, Mocaer, Luiselli, & Bologna 2009; Mollov 2011; Herczeg et al. 2012; Pulev & Sakelarieva 2013) compared to those from North America (Scheffers & Paszkowski 2012; Barrett & Price 2014).