AMPHIBIAN MOVEMENT ACROSS A NEW RESIDENTIAL ROAD IN WESTERN WASHINGTON STATE.
Breeding timing in western Washington State is influenced by a maritime climate characterized by mild, wet conditions during the autumn, winter, and spring, and a dry Mediterranean climate in the summer (Mass 2008). Most pond-breeding amphibians in western Washington have a bi-phasic lifestyle. Breeding and larval development occur in lentic habitats, but much of the year adults and juveniles occupy terrestrial habitats (Corkran and Thoms 1996; Jones and others 2005). Adults migrate to breeding habitats during periods of wet, mild winter, and early-spring weather (Corkran and Thoms 1996).
Thurston County, in western Washington, is among the most rapidly growing areas in Washington State. Between 2000 and 2010, the human population increased by 22% (Thurston County EDC 2017). Traffic-related amphibian mortality will undoubtedly increase with the growing human population and should be addressed if amphibian populations are to persist in the area. During 2006 to 2007, concentrated areas of amphibian mortality were identified through walking surveys on a 6-km length of county road (JSH, unpubl. data). Concurrently, a new development was initiated adjoining one of the high amphibian-mortality areas. This presented an opportunity to gather information on amphibian movement across new roads in the development, before traffic levels associated with occupancy occurred.
In the Pacific Northwest, limited research on this topic includes work in British Columbia (Beasley 2006) and Oregon (Beilke and Stratis 2018). Our study purpose was to document migration patterns for pond-breeding amphibians to support recommendations for safe-crossing measures. Our objectives were to: (1) share findings from our exploratory study of autumn-to-winter amphibian migrations on the new roads; and (2) interpret these findings in a framework that provides literature-based measures for those in Thurston County and elsewhere who are working to implement safe-crossing measures.
In 2008 and 2009, we surveyed newly constructed roads that provide access to 19 home sites in a 47-ha young Douglas-fir (Pseudotsuga menziesii) forest in Thurston County, Washington State (Fig. 1). The roads are dead-end residential roads accessed from a primary county road (Fig. 1). The 6-m wide main asphalt road (1260 m long) and similar spur road (133 m) totaled 1393 m. The main road follows a northwesterly path through forest that is adjacent to and includes parts of a palustrine wetland complex. A small stream originates in the wetland complex and flows under the main road through a 1.8-m diameter metal culvert.
We conducted linear walking transects of the roads over 24 h during each month from October to December and again in February. These months were chosen because amphibian movement to and from breeding ponds at and near the study area is high during this period (JSH, unpubl. data). Data we collected includes amphibian occurrences on the roads and in the culvert by species, sex, and lifestage. We also documented movement direction relative to nearby wetlands, and autumn-to-winter and 24-h movement timing. Eighteen volunteers, including the authors, participated in the surveys.
We divided the road system into 10-m segments marked with flagging and georeferenced using GPS (Trimble, Sunnyvale, California). We surveyed the roads and the small stream through the culvert during 24 h on: 16-17 October 2008; 11-12 November 2008; 30-31 December 2008; and 23-24 February 2009. Each 24-h sample period was composed of 6 sequential 4-h surveys, with the 1st period always starting at 1600 h. We attempted to conduct surveys when weather conditions were predicted to be wet and relatively warm, thus conducive to amphibian movement. However, the December survey date unexpectedly had snow on the ground.
During each 4-h interval observers walked together over the entire length of roads (main and spur roads), first to the ends of the roads, and then back, moving slowly to minimize the likelihood of missing amphibians crossing the road. The intent was to take a minimum of 1 h to walk in each direction. Therefore the full survey within each 4-h interval typically would require a minimum of 2 h and the fastest walking speed that would be undertaken during a survey would be approximately 1.4 km [h.sup.-1]. All surveys or portions of surveys taking place after dark or in low light levels were accomplished with spotlights.
We checked for amphibians from the upstream and downstream culvert ends twice during each 4-h interval. When there was no flowing water, or if the culvert was wadeable, an observer walked through the culvert to survey amphibians. In November, December, and February the stream was flowing through the culvert, making it difficult to observe amphibians. To improve observations under these conditions, we installed 2 minnow traps with 6.3-cm openings (Promar[TM] small collapsible minnow trap; 25 x 25 x 46 cm; mesh 1.6 mm) in the water at each end of the culvert. These traps were installed during the first pass on the survey start day, and thereafter routinely checked twice during each 4-h survey interval.
When we observed an amphibian, we recorded the following data: (1) sequential observation number; (2) time; (3) species; (4) age class; (5) sex; (6) snout-vent length (SVL) to nearest mm (to the anterior end of vent for salamanders); (7) tail length for salamanders (anterior vent to tail tip); (8) alive or dead; (9) 10-m segment of observation; and (10) observed movement direction. Following data collection we placed the animal off the road on the side of its direction of travel. We cleaned our hands after handling newts to prevent spreading tetrodotoxin to other animals.
Our analyses inform the following questions: (1) which species, sexes, and life stages are crossing the new roads or using the culvert; (2) when are they crossing; (3) are they heading toward the breeding wetland or away; and (4) where are they crossing? Commensurate with a small sample size, we provide summary numbers, proportions, and graphed analyses for non-spatial variables. For our 24-h analyses we excluded dead animals and culvert observations, and thus the December survey, as all December survey observations were inside the culvert. Additionally, we combined data from corresponding 4-h intervals across the October, November, and February surveys because light and dark conditions were similar: (1) 16:00 to 19:59 and 04:00 to 08:00 intervals always had both light and dark conditions; (2) 20:00 to 23:59 and 00:00 to 03:59 h intervals were always dark; and (3) 08:00 toll:59 and 12:00 to 15:59 intervals were always light. Traffic data for the road adjacent to the new development from 2011 to 2012 were provided by Thurston County.
For spatial variables we provide mean, standard deviation (SD), and range to describe breadth of road use. We identified relative densities of live amphibian observations along the study roads and in the culvert using the ArcGIS Spatial Analyst Tool, Kernal Density. To statistically test for observation hot spots we used ArcGIS Spatial Statistics Tool, Optimized Hot Spot Analysis (v.10.3; ESRI, Redlands, California), using the incident data aggregation method of count incidents within fishnet polygons. This analysis uses the Getis-Ord Gi* statistic (Getis and Ord 1992) to identify statistically significant hot and cold spots. Spatial analyses excluded 2 live animal observations with incomplete location data. Thus, for spatial analyses, the full set of live amphibian data is n = 180; for animals on the roads but not in the culvert, n = 155.
Across all surveys, we observed a total of 225 amphibians either on the roads or in the culvert. Of these, 157 were alive on the road, 25 were alive in the culvert, and 43 were dead on the road. Dead animals included 10 Pacific Treefrogs (Hyliola regilla), 31 Rough-skinned Newts (Tariclia granulosa), 1 Northwestern Salamander (Ambystoma gracilc), and 1 Long-toed Salamander (Ambystoma macrodactylum). Our study occurred before houses in the development were occupied. Nonetheless, construction and other vehicle traffic occurred on development roads, and all but one of the mortalities appeared to be traffic related. Dead animals are excluded from further results, as we cannot ascribe a time, date, or movement direction for most of the mortalities.
Of the total live-animal observations, we observed 7 species (Table 1). Pond-breeding amphibians were the dominant group, and the majority of the observations occurred in February for this group. Rough-skinned Newt and Pacific Treefrog were most frequently observed, comprising 50% and 22% of observations, respectively. We detected terrestrial breeding amphibians in low numbers and only in the October and November surveys (Table 1).
Species Richness and Abundance (Alive on the Roads)
Of the 7 amphibian species we observed alive on the roads, Rough-skinned Newt was most commonly recorded, representing slightly over half of our live-animal observations, followed by Pacific Treefrog (25%), and Northwestern Salamander (11%) (Table 1). Northern Red-legged Frog, Long-toed Salamander, Oregon Ensatina, and Western Red-backed Salamander were each encountered <10 times and collectively represent 13% of all live on-the-road amphibian observations. We observed the greatest number of live amphibians on the roads during the February survey (n = 99), and fewer animals during October and November (22 and 36 animals, respectively). No animals were recorded on the road during the December survey.
In contrast, we observed only 4 amphibian species in the culvert, as no Pacific Treefrog, Oregon Ensatina, or Western Red-backed Salamander were observed in the culvert at any time (Table 1). The Rough-skinned Newt represented nearly half of all culvert observations. Long-toed Salamander and Northern Red-legged Frog represented 28% and 20%, respectively, of amphibians observed in the culvert, and we found only 1 Northwestern Salamander in the culvert. Amphibians used the culvert only when the ephemeral stream was flowing through and ponding in the culvert during November, December, and February surveys (Table 1).
Of the 182 live-amphibian observations on the road or in the culvert, most were adult, whereas lesser proportions were juvenile, or of unknown lifestage (Table 2). We identified adults among all 7 species of amphibians, and juvenile Pacific Treefrog, Rough-skinned Newt, and Northwestern Salamander. In October and February we encountered more females than males. In November and December we encountered more males than females. Juvenile amphibians provided a higher proportion of the total observations during October and November than in December or February.
We ascribed movement direction for 157 live amphibian observations. Of these, 61% were directed toward the wetlands, 26% were moving away, and 13% were moving in a direction indiscernible as toward or away from the wetland complex. Most (68%) adults were moving toward the wetlands, and most (76%) juveniles were moving away. Inter- and intraspecies movement patterns illustrate diverse migration timings (Fig. 2).
24-h Movement Timing
We found the largest proportion (31%) of observations during the dark interval of 00:00 to 03:59, whereas the 2 fully daylight intervals had the least (both 10%) (Fig. 3). Species richness was greatest during the 2 fully dark intervals (n = 7 and n = 6, respectively), and the interval that transitioned from dark to daylight (n = 7). We observed only 2 species (Rough-skinned Newt, Pacific Treefrog) in the interval that transitioned from daylight to dark, and just 1 (Rough-skinned
Newt) was observed during the 2 fully daylight intervals.
Movement timing across 24-h surveys was similar among 4 of the 5 pond-breeding amphibian species (Fig. 3). Movement numbers peaked during the fully dark 00:00 to 03:59 interval for the Northwestern Salamander, Long-toed Salamander, Pacific Treefrog, and Northern Red-legged Frog, and none were found during the 2 fully daylight intervals. In contrast, we observed Rough-skinned Newt during all time frames, and this species represented >90% of observations during both the fully daylight and light-to-dark transition intervals. Newts also represented 20 to nearly 40% of observations during the 2 fully dark intervals and during the interval of transition to daylight. Traffic levels for the county road adjoining our study roads demonstrate that most amphibian movement occurred during the lowest volumes of traffic. However, this was not the case for Rough-skinned Newts, which tended to have the highest numbers of animals moving during the highest traffic volumes (Fig. 3).
Live Amphibian Observation Locations, Densities, and Hot Spots along the Roads
We observed live amphibians on 66% of the 10-m segments on the study roads (Fig. 4), and in all 100-m segments. The mean number of amphibians per 100-m segment along the roads was 11 (SD = 4; range = 4 to 20; all surveys combined). Our analysis identified 2 locations with higher relative densities of amphibian observations along the road and in the culvert (Fig. 4). We identified a statistically significant hot spot using live amphibian on-the-road observations, in the vicinity of the junction of the main road and the spur road (z-score = 3.16; P = 0.01; Confidence Level = 99%) (Fig. 4). When we included culvert observations, the stream vicinity showed up as a significant hot spot (z-score = 3.91; P = 0.01; Confidence Level = 99%). We identified no significant cold spots.
Our 225 amphibian observations represent a subsample of the population of amphibians crossing the roads, and should not be construed as evidence of a small population magnitude or a low level of conservation concern (Hels and Buchwald 2001). A robust estimate of the numbers of animals crossing the roads might involve: (1) marking animals to ensure the same animals were not observed multiple times; (2) measuring the speed of different species as they crossed the road; and (3) accounting for surveyor visibility capabilities related to weather and light conditions. Our data confirm that all expected local species of pond-breeding amphibians were present, including both juvenile and adult life-stages. However, mortality from the new road system and loss of habitat owing to clearing and development of new home sites will put additional stress on amphibians breeding in the adjacent wetland.
The spatial and temporal patterns of movement we documented have important implications for vulnerability of the local amphibian populations to road mortality, and for the development of an effective mitigation strategy to minimize impacts. For example, the spatial pattern of movement was dispersed throughout the road system, because the new roads in this development run parallel to the wetland along the outside of protected wetland buffers. Consequently, pond-breeding amphibians cross the new roads when accessing the terrestrial habitats they use for much of the year, and again when returning to breed The segments of surveyed roads all lie within 400 m of the wetlands, creating a strong likelihood that amphibians would be moving across the entirety of the roads (Glista and others 2008; Langen and others 2009). Indeed, we observed amphibians along the full length of the roads (Fig. 4), and no statistical cold spots were identified. This can make considerations for safe crossing measures more complicated than when concise migratory pathways are identified (Patrick and others 2010).
We did, however, identify 2 statistically significant hot spots, one of which included the large culvert. Our spatial results merit follow-up to determine if the same densities and hot spots would be found across multiple years (Garrah and others 2015), and to more fully investigate culvert use. For example, the seasonal differences in migration timing among species and life stages we documented would be most useful in designing mitigation strategies if coupled with an analysis that identified populations at highest risk and the life stages most critical to the viability of those populations (Beaudry and others 2010; Sterrett and others 2019). In addition, specific information regarding weather conditions that may trigger movement could be used to identify critical times when risks to populations and benefits of mitigation measures were greatest.
Nearly one-third of our amphibian observations occurred during the 00:00 to 03:59 dark interval. Timing of this interval corresponded with lowest traffic volumes on an adjacent road (Fig. 3), which is an important finding for the conservation of animals migrating at this time (Hels and Buchwald 2001). Rough-skinned Newt activity in our study tended to mirror the daily traffic pattern, with greater numbers moving during highest levels of traffic (Fig. 3). Day-active, slow-moving species such as the newt can be most vulnerable to traffic (Hels and Buchwald 2001). We hypothesize that mortality will be greatest on study roads for Rough-skinned Newts because they are active during times when traffic is greatest, and lower for the remaining species that are most active at night when the probability of encountering vehicles is lower. Indeed, 72% of the dead animals we observed were newts.
A complex relationship between encounter rates of live and dead amphibians at varying levels of traffic volume is described in the literature (Fahrig and others 1995; Sutherland and others 2010). Whereas higher traffic volumes can correspond with higher amphibian mortality rates but few amphibians (Fahrig and others 1995; Sutherland and others 2010), locations where large numbers of amphibian mortalities are found may signal locations of the most intact populations, and can correspond with relatively low traffic volumes (Sutherland and others 2010). Additionally, higher volumes of traffic reported as daily average values include night traffic, and are correlated with population decline and loss (Fahrig and others 1995, Sutherland and others 2010). Thus, conservation benefits conveyed by night movement appear to be constrained by threshold traffic volumes.
A few studies have addressed specific traffic volumes and ensuing risk to amphibians crossing roads and to their populations (Fahrig and others 1995; Hels and Buchwald 2001; Mazerolle 2004; Sutherland and others 2010). Based on these studies, we have identified 3 traffic intensity thresholds for interpreting traffic intensity and amphibian road mortality risk (Table 3). A predicted traffic intensity for the study roads, after homes are occupied, is 8 daily vehicles [h.sup.-1] (assumptions: 10 vehicle trips [household.sup.-1] [d.sup.-2] [Kirby Nunn, Thurston County Public Works, pers. comm.]; 19 homes accessed by the study roads; thus study site homes provide 190 trips [d.sup.-1], or an average of 8 trips [h.sup.-1]). Using Table 3 thresholds, predicted traffic intensity for the new development falls within the <22 daily vehicles [h.sup.-1] category, where high levels of road mortality may occur but populations may be among the most intact.
However, the new roads are not the only source of road mortality faced by amphibians migrating across the study roads. Some, such as Northern Red-legged Frog females, migrate as much as 4.8 km between breeding and non-breeding habitats (Hayes and others 2007) likely crossing 2 or more additional roads in the vicinity as they make migrations between the wetland complex and summer habitat. The same migratory routes may be traveled many times over the lifespan of an individual animal, with each road crossing adding mortality risk (Gibbs and Shriver 2005). For example, migrating female Common Toads (Bufo bufo) incurred 18% mortality crossing a road enroute to breeding habitat, and an additional 14% mortality while leaving the breeding habitat (van Gelder 1973). Study area roads connect by T-junction to a primary road location previously identified as an amphibian mortality concentration area (JSH unpubl. data), with a traffic level of 25 vehicles [h.sup.-1] (2011 to 2012 data from Thurston County). The traffic intensity of this road falls in the 22 to 85 vehicles [h.sup.-1] category, in which populations may be at risk of loss (Table 3).
Developing effective mitigation strategies to minimize road-related mortality is challenging. Research in nearby King County, Washington highlights the importance of landscape modeling to benefit migratory amphibians (Grand and others 2017). Ideally, geographic information system (GIS) tools for landscape planning such as those used for wildlife habitat connectivity modeling in Washington State (WHCWG 2010) could be employed at local spatial scales to identify sensitive amphibian-migration zones. This would provide planners and developers with information to proactively integrate amphibian migrations into development planning. The information can be used with traffic-volume thresholds (Table 3) and traffic-calming analyses (Jaarsma 2004), to provide a systematic approach for maintaining low traffic levels in amphibian-migration zones.
Mitigation measures for existing roads can lessen numbers of amphibians killed on roads, and support amphibian population persistence; however, success is variable (Beebee 2013; Smith and Sutherland 2014). Measures fall into 2 categories: those that modify human behavior, and those that modify amphibian behavior (Beebee 2013) (Table 4). A spectrum of information is needed to inform implementation of measures, which can vary depending on site characteristics and perceived risk level to amphibian populations. For example, at the study roads we have strong lines of evidence, as documented in this study, that amphibians are actively using the breadth of the roads, and that 2 locations may have a higher density of use. As a starting point, these results support example measures such as road signs at the development entrance and higher-density crossing locations. Use of the culvert for migratory movement could possibly be encouraged by attaching directional fencing to guide migrating amphibians to the culvert entrances (Aresco 2005; Beebee 2013), and by ensuring that pathways inside the culvert remain above water during high-flow events (Patrick and others 2010). Over time, if additional information is obtained, additional measures could be identified, implemented, and monitored if certain species and life history elements are found to convey greatest risk to populations. This iterative approach can be integrated with monitoring to ensure effectiveness of the measures, and provide an adaptive-management feedback loop.
Amphibians are small, mobile animals, which must safely travel across roads to access breeding, summer, and winter habitats. Our results demonstrate that their autumn to winter migrations are complex, with seasonal and 24-h movement differences among species and life stages that have important implications for vulnerability to road mortality and development of successful mitigation strategies. Mortality resulting from presumed low traffic volumes on the new roads examined in this study might not be great enough to threaten the viability of local populations; however, the risk is additive for migrating animals that also move across other more heavily trafficked roads in the vicinity. To maintain native amphibian populations as rural areas continue to develop and traffic volumes increase, it is necessary to understand migration patterns and pro-actively plan and implement safe-crossing measures in amphibian-migration zones while using an adaptive-management framework to ensure that these measures are effective.
We thank M Hayes, T Hicks, A McIntyre, M Tirhi, and J Tyson for aspects of study design, field assistance, and analysis support, and G Stewart for G1S assistance. We are grateful to R Ariza, H Fuller, J Handfield, V Handfield, R Johnston, N Maggiulli, N Ricketts, L Salzer, and J Wallace for assistance with data collection. We thank J Kaufman for access to the pre-development roads and K Nunn, Thurston County Public Works for traffic data. Additional appreciation to B Blessing-Earle, M Stevie, W Peer, and J Terry for continued efforts on behalf of migrating amphibians. Handling protocols followed USGS ARMI SOP No. 100.
ARESCO MJ. 2005. Mitigation measures to reduce highway mortality of turtles and other herpetofauna at a north Florida lake. Journal of Wildlife Management 69:549-560.
ASHLEY PE, ROBINSON JT. 1996. Road mortality of amphibians, reptiles and other wildlife on the Long Point causeway, Lake Erie, Ontario. Canadian Field-Naturalist 110:403-412.
BEASLEY BA. 2006. A study on the incidence of amphibian road mortality between Ucluelet and Tofino, British Columbia. Wildlife Afield 3:23-28.
BEAUDRY F, DEMAYNADIER P, HUNTER ML. 2010. Identifying hot moments in road-mortality risk for freshwater turtles. Journal of Wildlife Management 74:152-159. doi.org/10.2193/2008-370.
BEEBEE TJC. 2013. Effects of road mortality and mitigation measures on amphibian populations. Conservation Biology 4:657-668.
BEILKE G, STRATIS P. 2018. Saving an emperiled population of Northern Red-legged Frogs: How citizens came together to protect a local species declining due to habitat loss, climate change, and other human induced threats. In: Abstracts from the 2018 Annual meeting of the SNVB, held jointly with TWS--Washington Chapter, and Northwest PARC. Northwestern Naturalist 99:133.
CORKRAN CC, THOMS CR. 1996. Amphibians of Oregon, Washington, and British Columbia: A field identification guide. Renton, WA: Lone Pine Publishing. 175 p.
COSENTINO BJ, MARSH DM, JONES KS, APODACA JJ, BATES C, BEACH J, BEARD KH, BECKLIN K, BELL JM, CROCKE C, AND OTHERS. 2014. Citizen science reveals widespread negative effects of roads on amphibian distributions. Biological Conservation 180:331-340.
FAHRIG L, RYTWINSKI T. 2009. Effect of roads on animal abundance: an empirical review and synthesis. Ecology and Society 14:21. http://www.ecologyandsociety.org/vol14/iss1/art21.
FAHRIG L, PEDLAR JH, POPE SE, TAYLOR PD, WEGNER JF. 1995. Effect of road traffic on amphibian density. Biological Conservation 73:177-182.
GARRAH E, DANBY RK, EBERHARDT E, CUNNINGTON GM, MITCHELL S. 2015. Hot spots and hot times: Wildlife road mortality in a regional conservation corridor. Environmental Management 56:874-889. doi.org/10.1007/s00267-015-0566-1.
GETIS A, ORD JK. 1992. The analysis of spatial association by use of distance statistics. Geographical Analysis 24:189-206.
GIBBS JP, SCHRIVER WG. 2005. Can road mortality limit populations of pool-breeding amphibians? Wetlands Ecology and Management 13:281-289.
GLISTA DJ, DEVAULT TL, DEWOODY JA. 2008. Vertebrate road mortality predominantly impacts amphibians. Herpetological Conservation and Biology 3:77-87.
GRAND LA, HAYES MP, VOGT KA, VOGT DJ, YARNOLD PR, RICHTER KO, ANDERSON CD, OSTERGAARD EC, WILHELM JO. 2017. Identification of habitat controls on northern red-legged frog populations: implications for habitat conservation on an urbanizing landscape in the Pacific Northwest. Ecological Processes 6:44. doi.org/10.1186/s13717-017-0111-7.
HAYES MP, ROMBOUGH CJ, HAYES CB. 2007. Rana aurora (Northern Red-legged Frog): movement. Herpetological Review 38:192-193.
HAYES MP, QUINN T, RICHTER K, SCHUETT-HAMES JP, SERRA SHEAN JT. 2008. Maintaining lentic-breeding amphibians in urbanizing landscapes: The case study of the Northern Red-legged Frog (Rana aurora). In: Mitchell JC, Jung Brown RE, Bartholomew B, editors. Urban herpetology. Salt Lake City, UT: Society for the Study of Amphibians and Reptiles. p 445-461.
HELS T, BUCHWALD E. 2001. The effect of road kills on amphibian populations. Biological Conservation 99:331-340.
JAARSMA CF. 2004. Ecological 'black spots' within the ecological network: An improved design for rural road network amelioration. In: Jongman RHG, Pungetti G, editors. Ecological networks and green-ways: Concept, design, implementation. Cambridge, UK: Cambridge University Press. p 171-187.
JONES LLC, LEONARD WP, OLSON DH, editors. 2005. Amphibians of the Pacific Northwest. Seattle, WA: Seattle Audubon Society. 227 p.
LANGEN TA, OGDEN KM, SCHWARTING LL. 2009. Predicting hot spots of herpetofauna road mortality along highway networks. Journal of Wildlife Management 73:104-114.
MASS C. 2008. The Weather of the Pacific Northwest. Seattle, WA: University of Washington Press. 280 p.
MAZEROLLE MJ. 2004. Amphibian road mortality in response to nightly variations in traffic intensity. Herpetologica 60:45-53.
ORLOWSKI G. 2007. Spatial distribution and seasonal pattern in road mortality of the Common Toad Bufo bufo in an agricultural landscape of southwestern Poland. Amphibia-Reptilia 28:25-31.
PATRICK DA, SCHALK CM, GIBBS JP, WOLTZ HW. 2010. Effective culvert placement and design to facilitate passage of amphibians across roads, journal of Herpetology 44:618-626.
SMITH RK, SUTHERLAND WJ. 2014. Amphibian conservation: Global evidence for the effects of interventions. Exeter, England: Pelagic Publishing. 281 p.
STERRETT SC, KATZ RA, FIELDS WR, GRANT EHC. 2019. The contribution of road-based citizen science to the conservation of pond-breeding amphibians. Journal of Applied Ecology 2019:1-8. doi.org/10.1111/1365-2664.13330.
SUTHERLAND RW, DUNNING PR, BAKER WM. 2010. Amphibian encounter rates on roads with different amounts of traffic and urbanization. Conservation Biology 24:1626-1635.
THURSTON COUNTY EDC. 2017. Population and Demographics. http://www.thurstonedc.com/why-thurston-county/thurshton-county-profile/population-demographics. Accessed 22 February 2017.
VAN GELDER JJ. 1973. A quantitative approach to the mortality resulting from traffic in a population of Bufo bufo. Oecologia 13:93-95.
[WHCWG] WASHINGTON WILDLIFE HABITAT CONNECTIVITY WORKING GROUP. 2010. Washington connected landscapes project: Statewide analysis. Olympia, WA: Washington Departments of Fish and Wildlife, and Transportation Technical Report. 209 p.
JOANNE P SCHUETT-HAMES, DAVE E SCHUETT-HAMES
5146 Blue Heron Lane NW, Olympia, WA 98502 USA; firstname.lastname@example.org
F TEAL WATERSTRAT
US Fish and Wildlife Service, 510 Desmond Dr. SE, Lacey, WA 98503 USA
ERIC M LUND
Minnesota Department of Natural Resources, 1801 S Oak St., Lake City, MN 55041 USA
Corresponding Editor: Christopher Pearl.
Submitted 16 March 2018, accepted 2 April 2019.
TABLE 1. Numbers of live amphibians observed on the roads and in the culvert by species and survey. Species and survey 16-17 Oct. 2008 11-12 Nov. 2008 30-31 Dec. 2008 type NORTHWESTERN SALAMANDER Road 3 4 0 Culvert 0 0 0 LONG-TOED SALAMANDER Road 0 0 0 Culvert 0 0 1 ROUGH-SKINNED NEWT Road 11 23 0 Culvert 0 4 8 OREGON ENSATINA Road 2 2 0 Culvert 0 0 0 WESTERN RED-BACKED SALAMANDER Road 2 0 0 Culvert 0 0 0 PACIFIC TREEFROG Road 4 3 0 Culvert 0 0 0 NORTHERN RED-LEGGED FROG Road 0 4 0 Culvert 0 0 2 TOTAL 22 40 11 Road 22 36 0 Culvert 0 4 11 Species and survey 23-24 Feb. 2009 Total type NORTHWESTERN 19 SALAMANDER Road 11 18 Culvert 1 1 LONG-TOED 13 SALAMANDER Road 6 6 Culvert 6 7 ROUGH-SKINNED 91 NEWT Road 45 79 Culvert 0 12 OREGON ENSATINA 4 Road 0 4 Culvert 0 0 WESTERN RED-BACKED 2 SALAMANDER Road 0 2 Culvert 0 0 PACIFIC 40 TREEFROG Road 33 40 Culvert 0 0 NORTHERN 13 RED-LEGGED FROG Road 4 8 Culvert 3 5 TOTAL 109 182 Road 99 157 Culvert 10 25 TABLE 2. Numbers of live amphibians on the roads and in the culvert by sex and lifestage. Survey Adult Adult Adult Adult Juvenile Lifestage male female sex unkn. total total unknown 16-17 Oct. 2008 2 4 5 11 7 4 11-12 Nov. 2008 13 6 14 33 6 1 30-31 Dec. 2008 5 0 3 8 0 3 23-24 Feb. 2009 33 62 3 98 8 3 Total 53 72 25 150 21 11 % of total 29.1 39.6 13.7 82.4 11.5 6.0 observations Survey Observation totals 16-17 Oct. 2008 22 11-12 Nov. 2008 40 30-31 Dec. 2008 11 23-24 Feb. 2009 109 Total 182 % of total 100 observations TABLE 3. Categories for interpreting traffic intensity and amphibian road-mortality risk. Supporting references: Fahrig and others 1995; Hels and Buchwald 2001; Mazerolle 2004; Sutherland and others 2010. Traffic Amphibian response - risk intensity (daily veh [h.sup.-1]) <22 High levels of road mortality may occur among populations that are the most intact, at these relatively low traffic volumes. Highest mortality numbers have been found where large amphibian populations exist at traffic rates as low as 4 veh [h.sup.-1]. 22-85 Road mortality in this traffic intensity range increasingly impacts amphibian populations. Populations may be at risk of loss. >85 Strong evidence exists for negative population impacts: fewer amphibians, and lesser densities are found. At increasing traffic intensities, approximately [greater than or equal to]200 veh [h.sup.-1], encounters of amphibians may be few, and many populations and species may have died out. TABLE 4. Example measures to support safe crossings for amphibians migrating across local roads, within an adaptive-management framework. Measures Information needed Determine effectiveness ADAPTING HUMAN BEHAVIORS: Migration locations: based Routine monitoring, Road signs: encourage on local knowledge, for example: drivers to watch for road migration surveys, index surveys animals, and if safe, or field verified of amphibian allow animals to pass connectivity modeling migrations Traffic calming: speed Traffic: volume, speeds, Traffic bumps, and lower routing options volume and speeds speed data trends Encourage drivers to Migration timing and Ensure take other routes, weather conditions measures or to make fewer facilitating movement adequately trips during migrations protect populations CHANCING AMPHIBIAN BEHAVIORS: Migration road crossing Crossing Road under-crossings hot-spot locations effectiveness with directional fencing supported by monitoring (include utility of baseline data existing culverts), and their maintenance Additional measures Demographic, life Ensure based on effectiveness history, and measures monitoring movement timing adequately protect populations
|Printer friendly Cite/link Email Feedback|
|Author:||Schuett-Hames, Joanne P.; Schuett-Hames, Dave E.; Waterstrat, F. Teal; Lund, Eric M.|
|Publication:||Northwestern Naturalist: A Journal of Vertebrate Biology|
|Date:||Dec 22, 2019|
|Previous Article:||DIURNAL ACTIVITY PATTERN OF A PAIR OF BREEDING NORTHERN PYGMY-OWLS.|
|Next Article:||EFFECTS OF SCAVENGING ON ASSUMPTIONS OF MORTALITY ANALYSES OF RADIO-MARKED GAMEBIRDS.|