Herpetological Conservation and Biology 6(2):277-286. Submitted: 27 January 2011; Accepted: 31 May 2011. 277 USING CAMERAS TO MONITOR TUNNEL USE BY LONG-TOED SALAMANDERS (AMBYSTOMA MACRODACTYLUM): AN INFORMATIVE, COST-EFFICIENT TECHNIQUE KATIE S. PAGNUCCO 1,3,4 , CYNTHIA A. PASZKOWSKI 1 , AND GARRY J. SCRIMGEOUR 2 1 Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9 2 Parks Canada, Western and Northern Service Centre, Calgary, Alberta, Canada T2P 3M3 3 Current address: Redpath Museum, McGill University, Montreal, Quebec, Canada H3A 2K6 4 Corresponding author e-mail: katie.pagnucco@mail.mcgill.ca Abstract.—Wildlife crossing structures must be monitored to assess their ability to restore animal movement patterns. Although cameras have been used effectively to record use of crossing structures by mammals, they have not been used to document amphibian movements. We installed four amphibian tunnels in Waterton Lakes National Park, Alberta, Canada to reduce road mortality in a declining population of Long-toed Salamanders (Ambystoma macrodactylum). Our goal was to determine if cameras offer an alternative to pitfall traps for monitoring tunnels for amphibians. We installed digital cameras on the ceilings of tunnel entrances to monitor tunnel floors. Cameras were set to take motion-triggered images and timed-interval images (1 photograph/minute from 2100–0600). We installed one pitfall trap at each tunnel exit to capture amphibians travelling through tunnels and assess camera performance. From May through August 2009, we captured 104 adult A. macrodactylum in traps, but only 58 crossings were documented by cameras, indicating that cameras missed at least 46 salamander crossings. Ambystoma macrodactylum failed to trigger motion detectors during 81.0% of the camera-documented tunnel crossings. Cameras revealed that salamanders moved slower at tunnel entrances than their average crossing speed, suggesting animals may have hesitated at entrances. Although cameras documented one case of snake predation, images indicated that tunnels were not significant predator traps for salamanders. Camera data revealed the same patterns of demographics and spatio-temporal variation in tunnel use for A. macrodactylum as did trapping; thus, cameras represent a novel, cost-efficient, noninvasive approach to monitoring amphibian tunnel use. However, we encourage managers to either augment motion-detection cameras or rely on images recorded at set time intervals to document tunnel use effectively by animals as small as, or smaller than, A. macrodactylum. Key Words.—Ambystoma macrodactylum; amphibian; camera; conservation; crossing structure; Long-toed Salamander; monitor; tunnel use. INTRODUCTION The negative effects of roads on wildlife, which include increased mortality and decreased habitat connectivity, have been well documented (Fahrig et al. 1995; Forman et al. 2003; Jaeger and Fahrig 2004). As a result, crossing structures, which include underpasses, overpasses, and under-road tunnels, are being designed and incorporated into road construction and improvement projects throughout North America and Europe (Spellerberg 2002; Cain et al. 2003; Forman et al. 2003). The success of such mitigation projects is typically assessed based on the extent that they reduce wildlife-vehicle collision rates or restore animal movement patterns (e.g., Ford et al. 2009). To document the re-establishment of travel corridors, crossing structures must be monitored to determine species use, especially when wildlife population persistence and connectivity are primary concerns (Clevenger 2005; Dodd et al. 2007). Despite the inherent need to measure the success of crossing structures, monitoring programs are rarely implemented. Given the increasing number of under-road tunnels being installed, evaluation of monitoring methods that document tunnel use is a conservation priority. We conducted a literature review of papers published between 1989–2009 using the BIOSIS Previews™ search engine and the following key terms: wildlife crossing structure, underpass, overpass, or culvert. We found 25 studies that described various methods employed to monitor use of crossing structures. We found an additional 19 studies that monitored crossing structures by conducting informal searches through a variety of databases and conference proceedings. Of the 44 studies that monitored crossing structures, we found that almost half used track plates, and 43% used remotely triggered cameras (see Pagnucco 2010 for full review). Although track plates and cameras are effective at documenting crossing structure use by mammals, they Copyright © 2011. Katie Pagnucco. All Rights Reserved.