Identification of essential habitat is a fundamental component of amphibian conservation; however, species with complex life histories frequently move among habitats. To better understand dynamic habitat use, we evaluated Wood Frog (Lithobates sylvaticus (LeConte, 1825)) habitat selection and movement patterns during the spring migration and foraging periods and described the spatiotemporal variability of habitats used during all annual life-history periods. We radio-tracked 71 frogs in Maine during 2011–2013 and evaluated spring migration, foraging activity center (FAC), and within-FAC habitat selection. Telemetered frogs spent the greatest percentage of each field season in hibernacula (≥54.4%), followed by FACs (≥25.5%), migration habitat (≥16.9%), and breeding sites (≥4.5%). FACs ranged 49 – 1 335 m2 (568.0 ± 493.4 m2) and annual home ranges spanned 1 413 – 32 165 m2 (11 780.6 ± 12 506.1 m2). During spring migration, Wood Frogs exhibited different movement patterns (e.g., turn angles), selected different habitat features, and selected habitat features less consistently than while occupying FACs, indicating that the migration and foraging periods are ecologically distinct. Habitat-use studies that do not discriminate among annual life-history periods may obscure true ecological relationships and fail to identify essential habitat necessary for sustaining amphibian populations.

No abstract

Radiotelemetry provides fine-scale temporal and spatial information about an individual's movements and habitat use; however, its use for monitoring amphibians has been restricted by transmitter mass and lack of suitable attachment techniques. We describe a novel waistband for attaching external radiotransmitters to anurans and evaluate the percentages of resulting abrasions, lacerations, and shed transmitters. We used radiotelemetry to monitor movements and habitat use of wood frogs (Lithobates sylvaticus) in 2006 and 2011-2013 in Maine, USA; American toads (Anaxyrus americanus) in 2012 in North Carolina, USA; and, wood frogs, southern leopard frogs (L. sphenocephalus), and green frogs (L. clamitans) in 2012 in South Carolina, USA. We monitored 172 anurans for 1-365 days (56.4 AE 59.4) in a single year and 1-691 days (60.5 AE 94.1) across years. Our waistband resulted in an injury percentage comparable to 7 alternative anuran waistband attachment techniques; however, 12.5% fewer anurans shed their waistband when attached with our technique. Waistband retention facilitates longer monitoring periods and, thus, provides a greater quantity of data per radiotagged individual.

Ecological relationships and processes vary across species’ geographic distributions, life stages and spatial, and temporal scales. Montane landscapes are characterized by low wetland densities, rugged topographies, and cold climates. Consequently, aquatic‐dependent and low‐vagility ectothermic species (e.g., pool‐breeding amphibians) may exhibit unique ecological associations in montane landscapes. We evaluated the relative importance of breeding‐ and landscape‐scale features associated with spotted salamander (Ambystoma maculatum) and wood frog (Lithobates sylvaticus) wetland occupancy in Maine's Upper Montane‐Alpine Zone ecoregion, and we determined whether models performed better when the inclusive landscape‐scale covariates were estimated with topography‐weighted or circular buffers. We surveyed 135 potential breeding sites during May 2013–June 2014 and evaluated environmental relationships with multi‐season implicit dynamics occupancy models. Breeding site occupancy by both species was influenced solely by breeding‐scale habitat features. Spotted salamander occupancy probabilities increased with previous or current beaver (Castor canadensis) presence, and models generally were better supported when the inclusive landscape‐scale covariates were estimated with topography‐weighted rather than circular buffers. Wood frog occupancy probabilities increased with site area and percent shallows, but neither buffer type was better supported than the other. Model rank order and support varied between buffer types, but model inferences did not. Our results suggest pool‐breeding amphibian conservation in montane Maine include measures to maintain beaver populations and large wetlands with proportionally large areas of shallows ≤1‐m deep. Inconsistencies between our study and previous studies substantiate the value of region‐specific research for augmenting species’ conservation management plans and suggest the application of out‐of‐region inferences may promote ineffective conservation. © 2016 The Wildlife Society.

No abstract