An objective road risk assessment method for multiple species: ranking 166 reptiles and amphibians in California

An objective road risk assessment method for multiple species: ranking 166 reptiles and... Landscape Ecol (2018) 33:911–935 https://doi.org/10.1007/s10980-018-0640-1 RESEARCH AR TIC L E An objective road risk assessment method for multiple species: ranking 166 reptiles and amphibians in California . . Cheryl S. Brehme Stacie A. Hathaway Robert N. Fisher Received: 10 July 2017 / Accepted: 26 March 2018 / Published online: 9 May 2018 The Author(s) 2018 Abstract Results All chelonids, 72% of snakes, 50% of Context Transportation and wildlife agencies may anurans, 18% of lizards and 17% of salamander consider the need for barrier structures and safe species in California were ranked at high or very-high wildlife road-crossings to maintain the long-term risk from negative road impacts. Results were largely viability of wildlife populations. In order to prioritize consistent with local and global scientific literature in these efforts, it is important to identify species that are identifying high risk species and groups. most at risk of extirpation from road-related impacts. Conclusions This comparative risk assessment Purpose Our goal was to identify reptiles and method provides a science-based framework to iden- amphibians in California most susceptible to road tify species most susceptible to negative road impacts. mortality and fragmentation. With over 160 species The results can inform regional-scale road mitigation and a lack of species-specific research data, we planning and prioritization efforts and threat assess- developed an objective risk assessment method based ments for special-status species. We believe this upon road ecology science. approach is applicable to numerous landscapes and Methods Risk scoring was based upon a suite of life taxonomic groups. history and space-use characteristics associated with negative road effects applied in a hierarchical manner Keywords Reptile  Amphibian  Road mortality from individuals to species. We evaluated risk to both Habitat fragmentation  Road ecology  Risk aquatic and terrestrial connectivity and calculated assessment  Road buffer distances to encompass 95% of population- level movements. We ranked species into five relative categories of road-related risk (very-high to very-low) based upon 20% increments of all species scores. Introduction There have been many attempts to better characterize Electronic supplementary material The online version of and quantify threat criteria in order to classify species this article (https://doi.org/10.1007/s10980-018-0640-1) con- at higher risk of extinction at state, national, and global tains supplementary material, which is available to authorized users. levels (Congress 1973 (U.S. Endangered Species Act); Mace et al. 2008; Hobday et al. 2011; Thomson et al. C. S. Brehme (&)  S. A. Hathaway  R. N. Fisher 2016;IUCN 2017). Roads are a significant threat to U.S. Geological Survey, Western Ecological Research wildlife populations (e.g., Forman et al. 2003; Center, San Diego, CA, USA e-mail: cbrehme@usgs.gov 123 912 Landscape Ecol (2018) 33:911–935 Andrews et al. 2015a; van der Ree et al. 2015), causing systems, the use of barrier systems, underpasses, and both barrier (habitat fragmentation) and depletion overpasses can reduce road mortality and help to (road mortality) effects. Barrier effects occur when maintain connectivity and safe passage across roads animals avoid crossing roads, in which case roads for herpetofauna and other wildlife (Jochimsen et al. essentially fragment species habitat. Barrier effects 2004; Colino-Rabanal and Lizana 2012; Langton include reduced size and quality of available habitat, 2015; Langen et al. 2015b). Because it is currently reduced effective population size, reduced ability to unrealistic and cost prohibitive to mitigate all road- find mates and resources, increased genetic structur- ways for all species, it is vital to identify species most ing, and increased probability of local extirpation susceptible to road-related impacts. Within species (e.g., Forman et al. 2003; Fahrig and Rytwinski 2009; ranges, risks to populations and need for mitigation D’Amico et al. 2016). Depletion effects occur when can then be evaluated based upon local road densities animals attempt to cross roads and are killed by and matrix, road-types, traffic, and road locations in vehicles. Depletion effects include all of the risks from relation to species habitat and movement corridors barrier effects as well as reduced survivorship, making (e.g., Jaeger 2000; Litvaitis and Tash 2008; Langen high road mortality an even greater concern (Jackson et al. 2015b; Zimmermann Teixeira et al. 2017). and Fahrig 2011). Among other stressors, such as Here we describe a road risk assessment method- habitat loss and fragmentation, invasive species, ology applied to native amphibian and reptile species in California, a global biodiversity hotspot (Myers pesticide use, changing climate, and disease, the negative impacts from roads may independently or et al. 2000). We also included analysis of subspecies if cumulatively threaten the persistence of populations they had special federal or state protection status. This and even species. includes 166 species and subspecies of frogs, toads, Amphibians and reptiles have been identified as salamanders, snakes, lizards, turtles, and tortoise. being particularly susceptible to the negative effects of Rankings and prioritizations such as these can be very roads within their habitat (e.g., Klauber 1931; Forman subjective. In order to avoid including low risk species et al. 2003; Rytwinski and Fahrig 2012; Andrews et al. that may be favored by the assessors or to uninten- 2015a, b; D’Amico et al. 2015). Many are slow tionally overlook species that are at high risk, it was moving, do not avoid roads, and are simply too small important for this be done in an objective manner for drivers to see and avoid. During rains many informed by current road ecology literature. amphibians make long linear terrestrial movements Very few quantitative data are available on the regardless of the presence of intersecting roadways impact of roads on population persistence. Jaeger et al. (Glista et al. 2008), and because paved roads typically (2005) were the first to develop a relative ranking absorb and retain more heat than the surrounding system to compare the impact of roads on wildlife habitat, snakes and lizards are often attracted to roads populations. Their ranking system was largely based for thermoregulation (Case and Fisher 2001; Jochim- upon behavioral responses of animal species to the sen et al. 2004). In fact, road surveys are one of the road surface, road size, traffic noise, and vehicles with most common methods for surveying these reptiles varying road sizes and traffic volumes. However, (e.g., Sullivan 2012). Many herpetofauna species knowledge of these detailed behavioral responses to utilize both aquatic and terrestrial habitat for breeding, ranges in road and traffic characteristics is rarely found development, foraging, and overwintering and there- in literature and the link between individual behavior fore require connectivity within and between both and population-level effects has not been clearly aquatic and terrestrial habitats to support basic life established (Rytwinski and Fahrig 2012, 2013). history requirements. Rytwinski and Fahrig (2012) performed a meta- The primary goal of this study was to provide analysis of wildlife groups to test whether certain life information to transportation and other planning history characteristics were related to negative agencies in California to assist them in prioritizing responses to roads. High reproductive rate (fecundity) road mitigation efforts for amphibian and reptile was negatively associated with the magnitude of species. Although there is still a lot to learn about the population-level effects for amphibians. No associa- effectiveness of different designs of road mitigation tions were significant in reptiles, although there were 123 Landscape Ecol (2018) 33:911–935 913 few studies to inform this analysis. However, a strong foragers, seasonal migrants, sit-and-wait predators vs. link was shown between body size, greater mobility, sedentary species) (e.g., Bonnet et al. 1999; Carr and lower reproductive rates and the magnitude of nega- Fahrig 2001). Because many species are semi-aquatic, tive road effects in mammals, the most studied wildlife movement distance and frequency were scored sepa- group. Conversely, simulations predicted populations rately for both aquatic and terrestrial habitats. of species with small home ranges and high reproduc- There is a theorized higher risk associated with tive rates were the least likely to be affected by roads depletion effects (i.e., road mortality) in comparison to (Rytwinski and Fahrig 2013). barrier effects (Fahrig and Rytwinski 2009; Jackson We used these findings as a basis for creating a and Fahrig 2011). Therefore, we gave additional multi-tiered system to rank and identify reptile and weight to those species more likely to go out onto a amphibian species that may be most susceptible to road surface and be killed by vehicular traffic. For this road impacts. We based our ranking upon a suite of we considered factors of habitat preference (e.g., open species life history and space-use characteristics vs. closed), roads as potential attractants (e.g., for associated with negative road effects, as well as basking), and movement speed (e.g., slow vs. fast). including species distribution and conservation status. However, individuals within and among species may We evaluated risk to both aquatic and terrestrial respond differently to roads (attraction vs. avoidance) connectivity and include buffer distances that were based upon local landscape features, road width, traffic volume, and perceived danger (Forman et al. calculated to encompass 95% of population move- ments. Relative confidence in these distances is given 2003; Andrews 2005; Brehme et al. 2013; Jacobson for each species based upon the amount of support et al. 2016). Because a state-wide analysis encom- from scientific studies. We solely focused on the direct passes extreme variation in landscape and road effects of roads as barriers and sources of road characteristics, the extent to which roads act as mortality and not impacts from road construction barriers or sources of direct mortality within a species and maintenance or indirect effects from increased range is unknown. The risk disparity between deple- human use of the landscape once a road is in place (see tion and barrier effects could also be highly variable. review by Langen et al. 2015a). Therefore, we limited the additional weight for Because we based the risk assessment solely upon potential depletion effects to twenty percent of the space-use and life history characteristics, this repre- individual risk score. sents a species relative susceptibility to road impacts. We assessed population-level road risk by multi- It is understood that circumstances associated with plying individual risk with scores representing: (1) the particular populations (e.g., local road types, loca- relative proportion of the population at risk; and (2) the tions, densities) may elevate or reduce the risk for species ability to sustain higher rates of mortality. For certain populations and species. instance, the proportion of the population at risk was expected to be higher for migratory species than for territorial species. Highly fecund species were Methods expected to better withstand (or more quickly recover from) higher mortality in comparison to those with Road risk assessment (overview) few annual offspring. Finally, we assessed species-level road risk by multiplying population road risk with scores for range We assessed the relative risk of California herpeto- fauna species to negative road-related impacts at three size (both within and outside of California) and scales in a hierarchical fashion. We first assessed risk conservation status according to the U.S. Fish and at the scale of an individual animal and then expanded Wildlife Service (USFWS 2016) and the California the risk to the population and then to species (Fig. 1). Department of Fish and Wildlife (CDFW 2016a; At the individual-level, we based road risk primar- Thomson et al. 2016). Species with smaller ranges ily upon the likelihood that an individual would typically have fewer populations and are thus less encounter one or more roads. We considered this a resilient to population-level stressors. Endangered, product of movement distance (home range, seasonal threatened, and special concern species have already migrations) and movement frequency (e.g., active been designated at risk of extirpation, often due to 123 914 Landscape Ecol (2018) 33:911–935 Fig. 1 California reptile and amphibian road risk assessment conceptual model (ARSSC Amphibian and Reptile Species of Special Concern (Thomson et al. 2016)) multiple stressors, and are thus thought to be less 5. Conservation Status of Amphibians and Reptiles likely to be resilient to additional road impacts. on USDA National Forests, Pacific Southwest Region, 2012 (Evelyn and Sweet 2012). Although we present both aquatic and terrestrial risk scores for semi-aquatic species, we used the 6. Natureserve Explorer (natureserve.org): Species Accounts largely authored by G. Hammerson higher of the two scores for the overall risk ranking. (2003–2016). Literature review When these reviews were lacking life history information needed for the road risk assessment, we Species life history data were primarily taken from and then searched for supplementary peer-reviewed liter- cross-checked among the following species account ature using the Google Scholar search engine. Because review sources; movement distances (terrestrial, aquatic, home range, 1. U.S. Fish and Wildlife Service (USFWS) Recov- migratory) were so important for the risk assessment, ery Plans and 5-year Reviews https://www.fws. we acquired referenced articles from the species gov/endangered/. accounts and independently searched the literature to 2. California Amphibian and Reptile Species of acquire these data. Search terms included the species Special Concern (ARSSC; Thomson et al. 2016). common name, scientific name, or genus and terms 3. A Field Guide to Amphibians and Reptiles of such as ‘‘movement’’, ‘‘home-range’’, ‘‘spatial’’, and California (Stebbins and McGinnis 2012) ‘‘telemetry’’. We also reviewed articles for citations of 4. Amphibian declines: the conservation status of other studies to find more recent information on movement. This literature included published articles, United States species (Lannoo 2005). 123 Landscape Ecol (2018) 33:911–935 915 book chapters, M.S. Theses, Ph.D. dissertations, Table 1 Individual-level Road Risk (IRR): Score criteria for risk of individuals encountering a road agency reports, and consultant reports. In the case that specific life history or movement information was Risk of individuals encountering a road = Movement distance 9 not found for a species, we chose a surrogate species frequency based upon phylogeny, habitat, and body size. We first Movement Score Frequency Score looked for the closest related species within the genus distance (m) or family and chose a closely related surrogate based [ 1200 40 Active throughout home range 2 upon similar habitat and body size. If surrogates were 901–1200 32 Migratory (2–4 9 per year)/ 1.5 used, these are clearly reported. non-migratory sit and wait foragers Road risk metrics 601–900 24 Sedentary, confined to 1 specialized habitat The following section describes in detail the rank 451–600 16 scoring used for Individual-level Road Risk, Popula- 301–450 12 tion-level Road Risk, and Species-level Road Risk. 201–300 8 All rank values are meant to represent the relative 101–200 5 contribution of each attribute to either additive or 51–100 3 multiplicative road risk. 0–50 1 Individual-level risk (100 points possible) Out of a total of 100 points for individual road account for the majority of movement for these species. mortality risk, we attributed up to 80 points (80%) to the risk of encountering a road and up to 20 points The calculations and rankings for movement distances were well considered and deserve further (20%) for the risk of an individual moving onto a road and being killed by a motor vehicle. explanation. Our original thinking was that maximum distances should reflect relative movement distances The risk of encountering a road was based on a combination of movement distance and general across species and these data were commonly reported movement frequency. Movement distance was ranked in species accounts. However, it became increasingly 1–40 based upon home range movement distances difficult to determine whether maximum distances (diameter) for non-migrants or migration distances for reported were seasonal migration movements, home range movements or rarer dispersal events. We seasonal migrants that spanned from 0 to [ 1200 m (Table 1). The scores are linearly correlated with believed this assessment should reflect annual move- ment distances and not rare dispersal events. We increasing movement distance. For species that use both terrestrial and wetland/ considered using average/median movement dis- tances; however, these often underestimate the move- stream/riverine habitats, such as frogs, toads, aquatic snakes and turtles, we scored aquatic and terrestrial ment of seasonal migrants because in many cases a sizeable portion of the population may remain close to movement distances and frequencies separately. This was necessary as some species move much larger a breeding site, while another sizable portion make distances and at different frequencies in one habitat longer distance migrations causing an average or versus the other. This also informs the type(s) of median to be uninformative. Therefore, we decided to mitigation structures that may be warranted based use a buffer distance that incorporates the movement distances of 95% of the population studied. A 95% upon habitat type, buffer distances and risk scores for each species. Aquatic movement distances were not population movement distance is commonly accepted for the delineation of terrestrial buffer zones for calculated for pond-breeding amphibians. Ponds are typically small ephemeral bodies of water and terres- amphibians (i.e., Semlitsch 1998; Semlitsch and Bodie 2003) and we believe it was the most biologically trial movements of amphibians to and among ponds 123 916 Landscape Ecol (2018) 33:911–935 meaningful and useful measure for this study. This suitable overwintering sites are typically much farther measure, which we will refer to as Maximum Popu- away from breeding and summer activity areas than in lation Movement Distance (MPMD), should include milder California climates (e.g., Gregory 1984). We almost all population movements, such as seasonal did use study data from adjacent states or lower migration distances and annual home ranges (diame- estimates of migration distances from those reported ter), but not rare dispersal events. The MPMD should in Midwestern states. In some cases where little also be useful for local risk assessments as these information was available, we made an educated guess distances can be used to aide in mapping and based upon limited study data and/or closely related mitigation decisions. species and noted these in the tables. For all MPMDs, The calculation we used for MPMD is commonly we report a relative confidence level based upon the known as the 95% upper tolerance interval (Vangel number and quality of studies, sample sizes, and 2015). A tolerance interval is an interval that is meant locations in or adjacent to California. It is intended that to contain a specified percentage of individual popu- the scores be adjusted as new information becomes lation measurements. This should not be confused available. with a confidence interval, which is an interval that is To compute the risk of encountering a road, the meant to contain the population mean. We chose a MPMD was multiplied by a relative index of the 50% confidence level for the upper 95% confidence expected frequency of longer distance movements (1–2 points; Table 1). We defined three frequency limit of movement distances which is equal to the 95% prediction interval for future observations and is the categories largely based upon annual migratory mean ? 1.645 9 standard deviation. In cases where a movements or foraging strategies for non-migratory standard deviation was not reported, we back calcu- species. The highest category included actively for- lated standard deviation from the standard error and aging predators which are characterized by frequent sample size, calculated it from the individual data, or wandering movements throughout their home range estimated it based on the methods recommended by (Pianka 1966). Less frequent movers included sea- Hozo et al. (2005). Although non-parametric tolerance sonal migrants traveling among breeding, summer intervals would be more appropriate for non-normally foraging, and/or overwintering sites and non-migra- distributed movement data, the data required to tory ‘sit-and-wait’ predators that remain still for long calculate these is rarely reported in the published periods of time to ambush prey (Pianka 1966). Finally, literature. In the case of non-normally distributed data low frequency included highly sedentary species with where medians, sample sizes and ranges are reported, high site fidelity, particularly specialized rock, cre- Hozo et al. (2005) methods allow for approximation of vice, soil, or tree dwellers that may rarely traverse means and standard deviations with no assumption of terrestrial or aquatic habitats. the underlying data distribution. We found the result- The risk of an individual moving onto a road and ing MPMDs to be reasonable in excluding large being killed by a moving vehicle was ranked by outliers but including multiple long distance move- attributes of habitat preference, road use, and move- ments below the maximum movement distance. ment speed (Table 2). Habitat preference represents We recognize that for any species there can be the degree to which an individual is expected to go out substantial variability in movement distances that onto or avoid an open road as predicted from their depend upon varying local, landscape, and climatic habitat and microhabitat preferences. Open habitat factors. This was often reflected in studies with specialists and generalists were expected to more sometimes widely varying estimates of home range readily move onto a road than species that prefer cover and migration distances. We attempted to be conser- (e.g., Forman et al. 2003; Brehme et al. 2013). vative by using the study data for calculation of Although many amphibians are closed habitat spe- MPMD in which the largest population movement cialists, most readily move through open habitats distances were observed. For studies where movement during rain events, when most overland migratory distance significantly varied between females and movements tend to occur (Glista et al. 2008). There- males, we used the information from the wider ranging fore, amphibians were considered open habitat spe- sex. For migratory distances, we did not use distances cialists for this ranking. An additional factor that may from extreme environments, such as Canada, where increase road use is for thermoregulation for lizards 123 Landscape Ecol (2018) 33:911–935 917 Table 2 Individual-level Road Risk (IRR): Score criteria for risk of road mortality Risk of road mortality = Habitat preference ? road use ? movement speed Habitat preference Score Road use Score Movement speed Score Open habitat specialist/amphibians 10 Thermoregulation (snakes/lizards) 4 Slow (\ 0.6 m/s) 6 Generalist 8 Other 0 Medium (0.6–2.0 m/s) 3 Edge specialist 4 Fast ([ 2.0 m/s) 0 Closed habitat or aquatic specialist 0 Table 3 Population-level Road Risk (PRR): Score criteria for population level road risk PRR = IRR 9 (Fecundity ? Proportion of population at risk) Fecundity Ave. potential offspring/year Score Proportion of population at risk Score Low 0–10 2 Seasonal migrants (Migratory) 2 Med 11–25 1.5 Wandering 1.5 High 26–100 1 Territorial 1 Very high [ 100 0 and snakes, as roads often retain more heat than the potential offspring was calculated by multiplying the surrounding environment (Colino-Rabanal and Lizana average clutch size by the average number of clutches 2012; Mccardle and Fontenot 2016). Finally, there is per year. an increased risk of road mortality for slow versus fast Individual mortality risk (1–100 points) was mul- moving species (see Andrews and Gibbons 2005; tiplied by the sum of these population-level factors Mazerolle et al. 2005; Andrews et al. 2015b). (1–4 points) to calculate population-level road risk. Population-level Road Risk (400 points possible) Species-level road risk (1200 points possible) To assess the risk of negative road impacts on the In comparison to population-level risk, we considered persistence of a population we incorporated scores for the overall risk of roads to species to be negatively population-level movement behavior and fecundity associated with species range and conservation status. (Table 3). For the proportion of a population expected Although some populations may be at high risk, to encounter a road, we scored the greatest risk to species with a wide distribution and many populations species that seasonally migrate to overwintering and should be more resilient to localized declines and breeding areas (Jackson et al. 2015). For those that do extirpations. Therefore, we assigned a range isolation not migrate, we expected higher proportions of non- score ranging from 0 to 1 that considered species territorial or loosely territorial species (‘‘wandering’’) distributions range-wide (North America) and within to encounter roads than species that defend distinct California (CA) (Table 4). Range-wide distribution territories. varied from ‘‘CA only’’ to ‘‘widespread’’ ([ 4 states). Species with low fecundity are less resilient to road If the species range extended into Mexico and/or mortality impacts than highly fecund species (Rytwin- Canada, these countries were counted as another state ski and Fahrig 2013). Relative fecundity was simply for calculation of the index. California-wide distribu- calculated from the average number of potential tion was calculated based upon the number of CA offspring per year whether the animals were oviparous geographic regions occupied out of twelve regions or live-bearing. For egg-laying species, the number of defined by Hickman (1993) and used in Stebbins and 123 918 Landscape Ecol (2018) 33:911–935 Table 4 Species-level Road Risk (SRR): Score criteria for species-level road risk SRR = PRR 9 ((Range isolation score ? Conservation status score)/2) (a) Range isolation score = (North America range ? CA range)/2 North America range Rank/score CA only 1.00 2 states (very restricted distribution) 1.00 2 states (restricted) 0.67 2–3 states 0.33 Widespread (4 ? states) 0.00 California range (No. of geographic regions occupied) Rank/score 1 0.92 2 0.83 3 0.75 4 0.67 5 0.58 6 0.50 7 0.42 8 0.33 9 0.25 10 0.17 11 0.08 12 0.00 (b) Conservation status score Conservation status Rank/score CA or federal threatened/endangered 1.00 SSC priority 1 0.75 SSC priority 2 0.50 SSC priority 3 0.25 None 0.00 Population-level risk [ 80 only McGinnis (2012). These two scores (Range-wide from 0.25 to 0.75 based upon their SSC ranking. isolation, CA isolation) were summed and divided Population-level Road Risk (score range 1–400) was by two in order to normalize the overall range isolation multiplied by (1 ? Range Isolation Score ? Conser- score to a 0 to 1 scale. vation Status Score; score range 1–3) to calculate the At the species-level, we also incorporated conser- final Species-level Road Risk. vation status (Table 4). Some species are declining Range and conservation status were only used as a and are at higher risk of extinction often due to multiplier for species-level road risk if the population- multiple stressors. Federal and State Threatened and level road risk was greater than 80 (20% of possible Endangered Species were given the highest score population score). This helped to prevent false infla- (1.0). In California, forty-five species are designated tion of the road risk metrics for low road susceptible ‘‘Species of Special Concern (SSC)’’ with a ranking of species. 1, 2, or 3 based upon severity and immediacy of threats Because all members of the genus Batrachoseps affecting each taxon (Thomson et al. 2016). SSC (slender salamanders) are similar in body size, range species were given a conservation status score ranging size and general life history characteristics, we scored 123 Landscape Ecol (2018) 33:911–935 919 Table 5 Species-level frequency distributions and road risk As a way to support the results of our ranking model rankings with species literature, we focused on special status species. We reviewed recovery plans and 5-year Percentile Scores Relative ranks reviews for federally listed species and state species 81–100 322–710 Very high accounts for California listed species and species of 61–80 213–321 High special concern (collectively referred to as special 41–60 63–212 Medium status species). For each rank group (i.e., ‘‘very low’’ 21–40 53–62 Low to ‘‘very high’’), we calculated the percentage of 1–20 0–52 Very Low special status species where roads were specifically listed as a threat. Similarly, we tallied the number of species identified in a recent California preliminary road risk assessment (Levine 2013, Amy Golden pers. comm.) and compared the number of species that fell the genus as whole with the most conservative within each of our road risk categories. estimates and conservation status but included all 20 species in the final count and calculations. Once all 166 species (including subspecies with Results conservation status) were scored for species-level road risk within both terrestrial and aquatic habitats, we All chelonids, 72% of snakes, 50% of anurans, 18% of took the maximum score for each species and sorted lizards and 17% of salamander species were ranked as them from the highest to lowest scores. We grouped high or very high risk from negative road impacts. species into categories of risk (Very high, high, (Table 6, Fig. 3). medium, low, and very low) based upon ranges of Review of species accounts, recovery plans, and values that represented frequency distributions in 20% 5-year reviews for all special status species showed increments of all species scores (Table 5, Fig. 2). Fig. 2 Histogram of species-level scores and approximate 20 percentile road risk categories 123 920 Landscape Ecol (2018) 33:911–935 Table 6 Numbers of species by taxa within each risk category Species group Species-level rankings Very high High Med Low Very low Salamander 4 4 3 26 9 Lizard 5 3 8 7 21 Anuran 5 6 6 4 1 Snake 15 21 13 0 1 Tortoise 1 0 0 0 0 Turtle 3 0 0 0 0 4/4 100% 80% 36/50 7/11 60% 4/11 40% 8/38 8/44 20% 0% Salamander Lizard Frog Toad Snake Chelonid Fig. 3 Percentages of species by taxa in high and very high road risk categories that 94% (17/18) of species accounts that referenced special status species accounts mentioned roads as a roads as a threat to the species were ranked as ‘‘high’’ potential threat. In addition, 79% (15/19) of species of or ‘‘very high’’ in our risk assessment (Table 7). Of the concern recommended in a recent Caltrans prelimi- special status species that ranked ‘high’ and ‘very nary road risk assessment scored as ‘high’ or ‘very high’, close to fifty percent (17/35) had road-related high’ risk in our analysis (Levine 2013, Amy Golden threats referenced in their listing literature. In com- pers. comm.). parison, only 4% (1/27) of ‘medium’ to ‘very low’ risk Table 7 Comparison of road risk results and number of special status species with roads listed as threat Road risk level Special status species Caltrans PI No. species in road risk level No. species with roads listed as threat % of Total No. Spp in road risk level Very high 25 14 56 11 High 11 3 27 4 Medium 5 1 20 3 Low 10 0 0 1 Very low 7 0 0 0 Caltrans PI are Caltrans identified sensitive species % High and Very High Risk Landscape Ecol (2018) 33:911–935 921 Table 8 Amphibian and reptile road risk assessment: very range), low risk species (20–40% percentile) and very low risk high risk species (80–100% percentile), high risk species species (0–20% percentile) (60–80% percentile), medium risk species (40–60% percentile Discussion Risk scores and relative rankings for California reptile and amphibian species in both terrestrial and To our knowledge, this is the first attempt to objec- aquatic habitats are presented in Tables 8. Terrestrial and Aquatic rankings are provided separately in tively assess the relative risk of roads at a species level Tables 9 and 10 and also include population-level using a logical and scientifically based framework and risk scores, 95% population buffer distances, confi- apply it across a large array of species and habitats. We dence levels, and identification of any surrogate believe this approach could be useful for assessing and species used for the distance calculations. Species comparing susceptibility of species to negative road scores for all ranking criteria and life history and impacts within and among all taxonomic groups. To date, such risk assessments have been based largely movement references are provided in Appendices 1 and 2. upon expert opinion, limited information available on 123 922 Landscape Ecol (2018) 33:911–935 Table 8 continued road mortality, and even less information available on aquatic habitats to further inform mitigation. Some population or species-level road effects (Levine 2013; aquatic species may greatly benefit from fish passages Rytwinski and Fahrig 2015). while others may better benefit from terrestrial barriers Overall, this is meant to be a first step in highlight- and wildlife crossings or both. ing reptile and amphibian species that may be at Although data are currently lacking to validate highest risk from roads transecting their habitat. These completely the scoring and results of the risk assess- species may deserve consideration for further study ment, our review of species accounts, recovery plans, and for implementing mitigation solutions to reduce 5-year reviews for federal and state-listed species and mortality and to maintain or enhance connectivity. California species of special concern show a strong The risk assessment was done for both terrestrial and association between elevated road risk from our 123 Landscape Ecol (2018) 33:911–935 923 Table 8 continued objective analysis and the probability that roads are resulting in less risk to aquatic movement of popula- listed as a potential threat to the species in the species tions that inhabit lake and river systems. However, listing literature. culverts that are more commonly constructed under Although more than 40% of special status species roads in streams and wetlands vary in passability are semi-aquatic, roads were rarely considered a threat depending on factors such as diameter, length, slope, to aquatic connectivity in the species literature. This outlet configuration, and other characteristics (Furniss may be accurate if bridges or large culverts currently et al. 1991; Clarkin et al. 2005; Kemp and O’Hanley exist for water flow that also provide permeability to 2010). In fact, Januchowski-Hartley et al. (2013) aquatic movement. Bridges are generally considered found that only 36% of road crossings were fully passable to fish in the Great Lakes basin. In addition, to be completely passable by all aquatic species. Bridges are more likely to be constructed adjacent to many low water crossings in arid regions of the state or over large water bodies and rivers, presumably are simply a dip in the road that allows water to flow 123 924 Landscape Ecol (2018) 33:911–935 Table 8 continued 123 Landscape Ecol (2018) 33:911–935 925 Table 9 Terrestrial risk ranking and population buffer distances 123 926 Landscape Ecol (2018) 33:911–935 Table 9 continued 123 Landscape Ecol (2018) 33:911–935 927 Table 9 continued over the surface during high flow events. These may be range and/or migratory), tend not to avoid roads (or are used as road crossings by species traveling along attracted to them for thermoregulation), are long lived, ephemeral stream corridors with or without water and have relatively low fecundity in comparison to flow. Given these potential vulnerabilities, we believe other herpetofaunal groups. Because of these traits, that road impacts to aquatic connectivity of herpeto- chelonids and snakes have been identified elsewhere as fauna deserve greater consideration. being particularly susceptible to negative population Across broad taxonomic groups, chelonids (tor- effects from roads (Gibbs and Shriver 2002;Andrews toises/turtles) and snakes had the greatest percentages et al. 2015b; Jackson et al. 2015). of species at ‘high’ or ‘very high’ risk from roads. They There are only four species of chelonids in are similar in that many move long distances (home California, (desert tortoise (Gopherus agazzii), 123 928 Landscape Ecol (2018) 33:911–935 Northwestern pond turtle (Actinemys marmorata), and underpasses (Dodd et al. 2004; Colley et al. 2017). Southwestern pond turtle (Actinemys pallida), and In our statewide risk analysis, coachwhips (genus the Sonoran mud turtle (Kinosternon sonoriense)). Masticophis/Coluber) were amongst the highest risk There has been a high level of attention to road groups at both the population and species-levels. impacts on the desert tortoise (Gopherus agazzii)as These are particularly wide-ranging and very active numerous studies have documented not only high road foragers in comparison to other snake genera (Stebbins mortality, but measurable road effect zones, and and McGinnis 2012). The coachwhip (Masticophis mostly positive responses to barriers and underpasses flagellum) was found to be ninefold more likely to be (e.g., Boarman and Sazaki 1996, 2006; Peaden et al. extirpated from habitats that were fragmented by roads 2016; but see Peadon et al. 2017). Although not listed and urbanization, contributing to their decline as a primary threat to pond turtle populations in throughout California (Case and Fisher 2001; Mitro- California (Thomson et al. 2016), road mortality is a vich 2006). Similarly, habitat fragmentation from major concern for western pond turtle populations in roads and urbanization were identified as primary Oregon (Rosenberg et al. 2009). Pond turtles travel threats to the Alameda whipsnake (Masticophis later- kilometers within perennial waters and from pool to alis euryxanthus USFWS 2011). Although road use pool in intermittent aquatic habitats to forage and find and mortality have been documented for many other mates (Goodman and Stewart 2000). In addition, terrestrial California snake species on road-riding females nest and lay eggs in terrestrial habitats up to surveys (e.g., Klauber 1931; Jones et al. 2011; Shilling 0.5 km away from water which make roads that and Waetjen 2017), there is a paucity of studies parallel aquatic habitat a threat to both females and examining population-level effects of roads on Cali- hatchlings (Reese and Welsh 1997; Rathbun et al. fornia snake species. We could find only one such 2002; Pilliod et al. 2013). In fact, road mortality of study, where presence of a highway was shown to females has been identified as a cause for male-biased reduce gene flow in the Western diamond-backed sex ratios in some populations of pond turtles and rattlesnake (Crotalus atrox) in the Sonoran Desert, AZ other freshwater turtle species (Steen et al. 2006; (Hermann et al. 2017). Rosenberg et al. 2009; Reid and Peery 2014). There- Long foraging movements within aquatic habitats fore, this species requires consideration of both also contributed to the majority of garter snakes aquatic and terrestrial connectivity to satisfy their (genus: Thamnophis) falling within the highest road annual resource requirements. Sonoran mud turtles risk categories. Maintaining aquatic and wetland also travel long distances within intermittent streams connectivity is of primary concern for these species. and thus may be at risk of roads that transect their Garter snakes also use terrestrial habitats for overwin- aquatic habitat (Hensley et al. 2010). tering, reproduction, and for moving among wetland Larger colubrid snakes (Family Colubridae; many or aquatic patches. Some migrate long distances to genera) and rattlesnakes (genus Crotalus) were ranked winter hibernacula, making them also susceptible to among the highest risk from negative road effects. In roads within adjacent terrestrial habitats (Roe et al. addition to being attracted to paved road surfaces for 2006; Jackson et al. 2015). The highly aquatic giant thermoregulation, many large snakes have wide home- garter snake (Thamnophis gigas) had the highest ranges or may move large distances between winter aquatic road risk score. Because it moves only short hibernacula and summer foraging areas. In contrast to distances on land (Halstead et al. 2015), mitigation smaller species, larger snakes are also less likely to may best focus on functional aquatic passages with avoid roads (Rosen and Lowe 1994; Andrews and lengths of adjacent road barriers based upon their Gibbons 2005; Andrews et al. 2008; Siers et al. 2016). terrestrial movement distances. High road mortality (e.g., Klauber 1931; Rosen and Toads were the third highest ranking group with Lowe 1994; Jones et al. 2011), reduced abundance 64% ranked in the highest risk categories. In partic- near roads (Rudolph et al. 1999; Jones et al. 2011), ular, Bufonid toads (family Bufonidae) may move increased extinction risk (Row et al. 2007), and large distances ([ 1 km) in both aquatic and terrestrial decreased genetic diversity (Clark et al. 2010; Her- habitats to satisfy their annual resource requirements; mann et al. 2017) have been documented for numerous thus 5 of 7 bufonid species ranked high or very high snake species; as have positive responses to barriers risk from roads. Consistent with our risk assessment 123 Landscape Ecol (2018) 33:911–935 929 Table 10 Aquatic risk ranking and population buffer distances results, there is evidence that bufonid toads are fecund, with relatively limited movements and thus particularly susceptible to negative impacts from ranked low for road impacts. Four of 11 species ranked roads elsewhere (Trenham et al. 2003; Orłowski within the highest risk groupings; California red- 2007; Eigenbrod et al. 2008). legged frog (Rana draytonii), Oregon spotted frog (R. Roads and traffic have been associated with pretiosa), Northern red-legged frog (R. aurora), and reduced abundance and species richness of frog Cascades frog (R. cascadae). The Oregon spotted frog populations (e.g., Fahrig et al. 1995; Houlahan and (R. pretiosa) is known to move large distances within Findlay 2003). However, approximately half of Cal- aquatic habitats (Bourque 2008; USFWS 2009). ifornia species are small, primarily aquatic, highly Construction of a highway that bisected the 123 930 Landscape Ecol (2018) 33:911–935 Yellowstone population of Oregon spotted frogs was migratory salamander species were ranked within the one important factor that reduced the population highest risk categories from negative road effects. dramatically in the 1950s (see discussion in Watson There is substantial evidence that habitat fragmenta- et al. 2003). Although portions of the populations tion and mortality due to roads negatively affect many show high site fidelity, California red-legged frog and of these species. For instance, newts regularly migrate Northern red-legged frog migrants can move large long distances over land from and to breeding ponds, distances ([ 1 km) across both aquatic and terrestrial and to terrestrial foraging habitats ([ 2 km; Trenham habitats (Bulger et al. 2003; Fellers and Kleeman 1998). Large numbers are found dead on roads during 2007; Hayes et al. 2007). Road mortality or habitat dispersal periods and newt species are often the first to fragmentation from roads and urbanization were listed disappear in fragmented landscapes (Gibbs 1998; as primary threats to these species elsewhere (USFWS Trenham 1998, Shields pers. comm.). Similarly, road 2002; COSEWIC 2015). mortality and habitat fragmentation are primary Lizards had relatively low percentages of species in threats to the California tiger salamander and other the high risk groupings. Many lizard species are small, Ambystomid salamanders because terrestrial habitat is non-migratory, territorial, have small home ranges and used for interpond migration and overwintering are thus at low risk of negative road effects. Similar to (Semlitsch 1998; Trenham et al. 2001; Bolster 2010). snakes, lizards can also be attracted to road surfaces for Because this assessment covers a wide array of thermoregulation. A few wide ranging species scored in species and habitats, the risk to particular species the highest risk categories including the Gila monster populations must be re-assessed on a local level. This (Heloderma suspectum), leopard lizards (genus Gam- includes consideration of the locations, types, and belia) and two horned lizard species (genus Phryno- densities of roads in relation to population and species soma). The Gila monster has been negatively associated ranges along with goals for functional, meta-popula- with urbanization, where larger home ranges and tion, and genetic connectivity (e.g., Marsh and Jaeger greater movement rates result in higher mortality for 2015). Due to very low road densities in their limited males (Kwiatkowski et al. 2008). Sensitive to habitat ranges, some species and populations may be at lower fragmentation, the blunt-nosed leopard lizard (Gambe- risk. For instance the Gila monster, Oregon spotted lia sila) was found to be largely absent from habitat frog, Sonoran mud turtle, Sonoran desert toad (Incilius patches less than 250 ha (Bailey and Germano 2015). alvarius) and Yosemite toad (Anaxyrus canorus) Flat-tailed horned lizards (Phrynosoma mccallii)are scored high due to life history and space-use charac- also susceptible to habitat fragmentation with very large teristics, however their limited ranges are largely in home ranges for their size, particularly in wet years protected or low road density areas in the state. Thus (Young and Young 2000). In fact, road mortality is a roads may not be a significant threat to these species in well-known threat for this species (see review by California. In contrast, high road densities may CDFW 2016b). Horned lizards are also particularly increase the risk for species within coastal regions vulnerable to being killed on roads due to their tendency such as remaining populations of Santa Cruz long-toed to flatten and remain motionless while being salamander (Ambystoma macrodactylum croceum), approached (Young and Young 2000). Alameda striped racer (Masticophis lateralis euryx- Salamanders also had relatively low percentages of anthus), and San Francisco garter snake (Thamnophis species in the high risk grouping. Over 75% (35/46) of sirtalis tetrataenia). However, most species consist of the California salamanders are lungless salamanders numerous populations with a myriad of differing road- (Plethodontidae) and Torrent salamanders (Rhyacotri- related threat levels. Although detailed species ranges tonidae). These species are mostly small, sedentary, and occupancy within ranges are well known for some non-migratory, closed habitat specialists with limited species with very limited ranges, for most species movement distances and these traits have resulted in a range-wide surveys have not been conducted. There- high level of speciation. This is exemplified by there fore, only general range boundaries are available that being at least 20 species of slender salamanders (genus encompass large portions of the state and availability Batrachoseps) in California alone (Martinez-Solano of species distribution models of habitat suitability and et al. 2007; Vences and Wake 2007). However, within occupancy within their ranges is rare. This lack of the salamander group, newts and several other detailed spatial information on species distribution 123 Landscape Ecol (2018) 33:911–935 931 further limits the potential to incorporate road loca- The quantity and quality of life history information, tions, types, and densities in a state and species-wide particularly movement data, are highly variable assessment. among species (see confidence levels; Tables 9 and We also note that relative risk to negative road 10). Therefore it is important to re-assess risk as new impacts is provided for both populations and species. information becomes available. Finally, this is a Risk was elevated for species with small and isolated structured assessment of comparative risk across a ranges and that are facing a myriad of other threats. range of target species; therefore specific values for Because of this, a few common widespread species high risk have not been established. The ranking or scored high at the population-level but not at the assessment methodology should be adaptive and species-level. This included gopher snakes (Pituophis updated with advancements of road ecology science catenifer) and western toads (Anaxyrus boreas) where (e.g., Linkov et al. 2006). road mortality has been identified as a threat to the persistence of local populations (e.g., COSEWIC 2012; Jochimsen et al. 2014). Conclusion To potentially aid in local assessments, we have provided distance estimates or ‘‘buffer zones’’ that Although roads are a significant cause of mortality and contain estimates for 95% of population-level move- habitat fragmentation for many wildlife populations, ments for all species (e.g., Semlitsch and Bodie 2003). road-related risk rankings have been based largely on We provide all references evaluated for distance expert opinion due to a scarcity of literature on road estimates in Appendix 2. Meta-population movements effects for most species. Therefore, we developed an can be very important to the stability of pond-breeding objective and scientifically-based comparative risk amphibians (e.g., Semlitsch 2008; Jackson et al. 2015) approach to assess the potential threat from negative and are included in many of the buffer zone calcula- road impacts using species life history and movement tions. However, we note that buffer zones may not data. After applying it to over 160 herpetofaunal species include meta-population-level movements if the rate (and subspecies) in the state of California, the results of these dispersal movements was less than 5% in the are consistent with road ecology literature in identifying studies we used for our analyses. known high risk species, and call attention to some This should be considered an initial assessment of species not previously identified. Overall, we found that susceptibility to negative road impacts in a hierarchi- snakes and chelonids had the largest proportion of cal framework (e.g., see Level 2; Hobday et al. 2011). species at high risk for negative road impacts due to Therefore, as previously stated it will be important to longer movement distances (home range and/or migra- re-assess the risk of specific populations to roads tory), lack of road avoidance, and relatively low within their habitat and to evaluate and compare fecundity in comparison to other herpetofaunal groups. alternatives at the local scale (e.g., Suter 2016). This Results also indicated that consideration of aquatic may include more detailed information on specific connectivity appears to be under-represented for semi- road attributes (e.g., density, type, location), as well as aquatic herpetofauna that use both terrestrial and species behavior (Jaeger et al. 2005; Rouse et al. 2011; stream, riverine, or wetland habitats. Rytwinski and Fahrig 2013; Jacobson et al. 2016). Age In addition to informing transportation planning structured and spatially explicit population viability and mitigation considerations for California herpeto- models are valuable tools to predict long-term popu- fauna, we believe this approach may be useful for lation responses to roads and to compare outcomes of comparing the risk of road-related fragmentation and multiple mitigation scenarios (e.g., Gibbs and Shriver mortality for species elsewhere and for other taxo- 2005; Borda-de-Agua et al. 2014; Polak et al. 2014; nomic groups. The results can help to inform multi- Crawford 2015). Need and placement of mitigation criteria threat assessments for special status species or structures can be guided by local population or meta- those in consideration for listing. Finally, this serves to population dynamics, landscape attributes, movement highlight species that may deserve further study and routes, and road mortality hot spots (e.g., Bissonette consideration for aquatic and terrestrial road mitiga- and Adair 2008; Langen et al. 2009, 2015b; D’Amico tion to reduce mortality and to maintain population- et al. 2016; Loraamm and Downs 2016). level connectivity. 123 932 Landscape Ecol (2018) 33:911–935 sizes in the San Joaquin Desert of California. W Wild This risk assessment approach compares the sus- 2:23–28 ceptibility of species to negative road impacts. Com- Bissonette JA, Adair W (2008) Restoring habitat permeability to monly, there are numerous populations within a roaded landscapes with isometrically-scaled wildlife species range that occupy areas with greatly differing crossings. Biol Conserv 141(2):482–488 Boarman WI, Sazaki M (1996) Highway mortality in desert road pressures. Therefore, the actual risk to specific tortoises and small vertebrates: success of barrier fences species populations will depend upon local road and culverts (No. FHWA-PD-96-041) densities, road-types, traffic, and road locations in Boarman WI, Sazaki M (2006) A highway’s road-effect zone for relation to species habitat and movement corridors. desert tortoises (Gopherus agassizii). J Arid Environ 65(1):94–101 Bolster BC (2010) A status review of the California tiger sala- Acknowledgements We greatly appreciate the support and mander (Ambystoma californiense). Report to the Fish and feedback from Harold Hunt, Amy Golden, and James Henke Game Commission, State of California. Nongame Wildlife from the California Department of Transportation. Tony Program Report 2010–4 Clevenger, Tom Langton, Jeff Tracey, Amber Wright, Laura Bonnet X, Naulleau G, Shine R (1999) The dangers of leaving Patterson, Brian Halstead, Kari Gunson, Jon Richmond, and two home: dispersal and mortality in snakes. Biol Conserv anonymous reviewers gave thoughtful feedback that improved 89(1):39–50 this manuscript. We thank Tristan Edgarian for reviewing and Borda-de-Agua L, Grilo C, Pereira HM (2014) Modeling the cross-checking our life history data and movement data and impact of road mortality on barn owl (Tyto alba) popula- references. Finally, we appreciate all of the scientists that tions using age-structured models. Ecol Model 276:29–37 contributed life history information for California reptile and Bourque R (2008) [North Coast] spatial ecology of an inland amphibian species (Appendices). Funding came from Caltrans population of the foothill yellow-legged frog (Rana boylii) Department of Transportation Contract #65A0553 and the in Tehama County, California. Masters Thesis, California Ecosystems Mission Area in USGS. This is contribution number State University, Humboldt, pp 107 615 of the U.S. Geological Survey Amphibian Research and Brehme CS, Tracey JA, McClenaghan LR, Fisher RN (2013) Monitoring Initiative (ARMI). The use of trade, product or firm Permeability of roads to movement of scrubland lizards names in this publication does not imply endorsement by the and small mammals. Conserv Biol 27(4):710–720 U.S. Government. Bulger JB, Scott NJ, Seymour RB (2003) Terrestrial activity and conservation of adult California red-legged frogs Rana Open Access This article is distributed under the terms of the aurora draytonii in coastal forests and grasslands. Biol Creative Commons Attribution 4.0 International License (http:// Conserv 110(1):85–95 creativecommons.org/licenses/by/4.0/), which permits unre- California Department of Fish and Wildlife (CDFW) (2016a) stricted use, distribution, and reproduction in any medium, Threatened and endangered species. http://www.dfg.ca. provided you give appropriate credit to the original gov/wildlife/nongame/t_e_spp/ (accessed August 2016) author(s) and the source, provide a link to the Creative Com- California Department of Fish and Wildlife (CDFW) (2016b) A mons license, and indicate if changes were made. status review of the flat-tailed horned lizard (Phrynosoma mcallii) in California. Report to the Fish and Game Commission Carr LW, Fahrig L (2001) Effect of road traffic on two References amphibian species of differing vagility. Conserv Biol 15(4):1071–1078 Case TJ, Fisher RN (2001) Measuring and predicting species Andrews KM, Gibbons JW (2005) How do highways influence presence: coastal sage scrub case study. In: Hunsaker CT, snake movement? Behavioral responses to roads and Goodchild MF, Friedl MA, Case TJ (eds) Spatial uncer- vehicles. Copeia 4:772–782 tainty in ecology. Springer-Verlag, New York, pp 47–71 Andrews KM, Gibbons JW, Jochimsen D, Mitchell J (2008) Clark RW, Brown WS, Stechert R, Zamudio KR (2010) Roads, Ecological effects of roads on amphibians and reptiles: a interrupted dispersal, and genetic diversity in timber rat- literature review. In: Mitchell JC, Jung Brown RE, Bar- tlesnakes. Conserv Biol 24(4):1059–1069 tholomew B (eds) Urban Herpetology, Herpetological Clarkin K, Conner A, Furniss MJ, Gibernick B, Love M, Conservation 3. Society for the Study of Reptiles and Moynan K, Wilson S (2005) National inventory and Amphibians, Salt-Lake City, UT, pp 121–143 assessment procedure for identifying barriers to aquatic Andrews KM, Nanjappa P, Riley SP (eds) (2015a) Roads and organism passage at road-stream crossings. US Dep Agric ecological infrastructure: concepts and applications for for Serv. http://www.stream.fs.fed.us/publications/PDFs/ small animals. JHU Press, Baltimore, MD NIAP.pdf Andrews KM, Langen TA, Struijk RPJH (2015b) Reptiles: Colino-Rabanal VJ, Lizana M (2012) Herpetofauna and roads: a Overlooked but often at risk from roads. Handbook of Road review. Basic Appl Herpetol 26:5–31 Ecology. Wiley, New York, pp 271–280 Colley M, Lougheed SC, Otterbein K, Litzgus JD (2017) Miti- Bailey CV, Germano DJ (2015) Probability of occupancy of gation reduces road mortality of a threatened rattlesnake. blunt-nosed leopard lizards on habitat patches of various Wildlife Res. https://doi.org/10.1071/WR16130 123 Landscape Ecol (2018) 33:911–935 933 Congress US (1973) Endangered species act. US Code Glista DJ, DeVault TL, DeWoody JA (2008) Vertebrate road 16:1534–1544 mortality predominantly impacts amphibians. Herpetol COSEWIC (Committee on the Status of Endangered Wildlife in Conserv Biol 3(1):77–87 Canada) (2012) COSEWIC assessment and status report on Goodman RH, Stewart GR (2000) Aquatic home ranges of the Western Toad Anaxyrus boreas in Canada. Committee female Western Pond Turtles, Clemmys marmorata, at two on the Status of Endangered Wildlife in Canada. Ottawa. sites in Southern California. Chelonian Conserv Biol xiv ? 71 pp. (www.registrelep-sararegistry.gc.ca/default_ 3(4):743–744 e.cfm) Gregory PT (1984) Communal denning in snakes. In: Seigel RA, COSEWIC (Committee on the Status of Endangered Wildlife in Hunt LE, Knight JL, Maralet LA, Zuschlag NL (eds) Canada) (2015) COSEWIC assessment and status report on Contributions to vertebrate ecology and systematics: a the Northern Red-legged Frog Rana aurora in Canada. tribute to Henry S. Fitch. University of Kansas Museum of Committee on the Status of Endangered Wildlife in Natural History Special Publication 10:57–75 Canada. Ottawa. xii ? 69 pp. (www.registrelep- Halstead BJ, Skalos SM, Wylie GD, Casazza ML (2015) Ter- sararegistry.gc.ca/default_e.cfm) restrial ecology of semi-aquatic giant gartersnakes Crawford BA (2015) Lessons from terrapin mortality and (Thamnophis gigas). Herpetol Conserv Bio 10(2):633–644 management on the Jekyll Island Causeway, Georgia, Hayes MP, Rombough CJ, Hayes CB (2007) Natural History: USA. In: Andrews KM, Nanjappa P, Riley SPD (eds) Rana aurora (Northern red-legged frog) movement. Her- Roads and ecological infrastructure. Concepts and appli- petol Rev 38:192–193 cations for small animals. Johns Hopkins University Press, Hensley FR, Jones TR, Maxwell MS, Adams LJ, Nedella NS Baltimore, pp 21–41 (2010) Demography, terrestrial behavior, and growth of D’Amico M, Roman J, De los Reyes L, Revilla E (2015) Ver- Sonora mud turtles (Kinosternon sonoriense) in an extreme tebrate road-kill patterns in Mediterranean habitats: who, habitat. Herpetol Monogr 24:174–193 when and where. Biol Conserv 191:234–242 Hermann HW, Pozarowski KM, Ochoa A, Schuett GW (2017) ´ ´ D’Amico M, Periquet S, Roman J, Revilla E (2016) Road An interstate highway affects gene flow in a top reptilian avoidance responses determine the impact of heteroge- predator (Crotalus atrox) of the Sonoran Desert. Conserv neous road networks at a regional scale. J Appl Ecol Genet 18:911–924 53(1):181–190 Hickman JC (ed) (1993) The Jepson manual of higher plants of Dodd CK, Barichivich WJ, Smith LL (2004) Effectiveness of a California. University of California Press, Berkeley barrier wall and culverts in reducing wildlife mortality on a Hobday AJ, Smith ADM, Stobutzki IC, Bulman C, Daley R, heavily traveled highway in Florida. Biol Conserv Dambacher JM, Deng RA, Dowdney J, Fuller M, Furlani 118(5):619–631 D, Griffiths SP (2011) Ecological risk assessment for the Eigenbrod F, Hecnar SJ, Fahrig L (2008) The relative effects of effects of fishing. Fish Res 108(2):372–384 road traffic and forest cover on anuran populations. Biol Houlahan JE, Findlay CS (2003) The effects of adjacent land use Conserv 141(1):35–46 on wetland amphibian species richness and community Evelyn CJ, Sweet SS (2012) Conservation status of amphibians composition. Can J Fish Aquat Sci 60:1078–1094 and reptiles on USDA National Forests, Pacific Southwest Hozo SP, Djulbegovic B, Hozo I (2005) Estimating the mean Region, 2012. Prepared for: USDA National Forest Ser- and variance from the median, range, and the size of a vice, Region 5, Pacific Southwest Region sample. BMC Med Res Methodol 5(1):1 Fahrig L, Rytwinski T (2009) Effects of roads on animal IUCN (2017) The IUCN Red List of Threatened Species. Ver- abundance: an empirical review and synthesis. Ecol Soc sion 2017-1. http://www.iucnredlist.org 14. http://www.ecologyandsociety.org/vol14/iss11/art21 Jackson ND, Fahrig L (2011) Relative effects of road mortality Fahrig L, Pedlar JH, Pope SE, Taylor PD, Wegner JF (1995) and decreased connectivity on population genetic diversity. Effect of road traffic on amphibian density. Biol Conserv Biol Conserv 144(12):3143–3148 73(3):177–182 Jackson SD, Langen TA, Marsh DM, Andrews KM (2015) Fellers GM, Kleeman PM (2007) California red-legged frog Natural history and physiological characteristics of small (Rana draytonii) movement and habitat use: implications animals in relation to roads. In: Andrews KM, Nanjappa P, for conservation. J Herpetol 41(2):276–286 Riley SPD (eds) Roads and ecological infrastructure. Forman RT, Sperling D, Bissonette JA, Clevenger AP, Cutshall Concepts and applications for small animals. Johns Hop- CD, Dale VH, Fahrig L, France R, Goldman CR, Heanue kins University Press, Baltimore, pp 21–41 K, Jones JA (2003) Road ecology. Science and Solutions. Jacobson SL, Bliss-Ketchum LL, Rivera CE, Smith WP (2016) Island Press, Washington A behavior-based framework for assessing barrier effects Furniss MJ, Roelofs TD, Yee CS (1991) Road construction and to wildlife from vehicle traffic volume. Ecosphere maintenance. Am Fish Soc Special Publ 19:297–323 7(4):e01345 Gibbs JP (1998) Amphibian movements in response to forest Jaeger JA (2000) Landscape division, splitting index, and edges, roads, and streambeds in southern New England. effective mesh size: new measures of landscape fragmen- J Wildl Manage 62:584–589 tation. Land Ecol 15(2):115–130 Gibbs JP, Shriver WG (2002) Estimating the effects of road mor- Jaeger JA, Bowman J, Brennan J, Fahrig L, Bert D, Bouchard J, tality on turtle populations. Conserv Biol 16(6):1647–1652 Charbonneau N, Frank K, Gruber B, von Toschanowitz KT Gibbs JP, Shriver WG (2005) Can road mortality limit popula- (2005) Predicting when animal populations are at risk from tions of pool-breeding amphibians? Wetl Ecol Manag roads: an interactive model of road avoidance behavior. 13:281–289 Ecol Model 185(2):329–348 123 934 Landscape Ecol (2018) 33:911–935 Januchowski-Hartley SR, McIntyre PB, Diebel M, Doran PJ, Mace GM, Collar NJ, Gaston KJ, Hilton-Taylor C, Akc ¸akaya Infante DM, Joseph C, Allan JD (2013) Restoring aquatic HR, Leader-Williams N, Milner-Gulland EJ, Stuart SN ecosystem connectivity requires expanding inventories of (2008) Quantification of extinction risk: IUCN’s system for both dams and road crossings. Front Ecol Environ classifying threatened species. Conserv Biol 11(4):211–217 22(6):1424–1442 Jochimsen DM, Peterson CR, Andrews KM, Gibbons JW, Marsh DM, Jaeger JAG (2015) Direct effects of roads on small Drawer E (2004) A literature review of the effects of roads animal populations. Roads and ecological infrastructure. on amphibians and reptiles and the measures used to Concepts and applications for small animals. Johns Hop- minimize those effects. USDA Forest Service, Idaho Fish kins University Press, Baltimore, pp 21–41 and Game Department, Washington, DC Martinez-Solano I, Jockusch EL, Wake DB (2007) Extreme Jochimsen DM, Peterson CR, Harmon LJ (2014) Influence of population subdivision throughout a continuous range: ecology and landscape on snake road mortality in a sage- phylogeography of Batrachoseps attenuatus (Caudata: brush-steppe ecosystem. Anim Conserv 17(6):583–592 Plethodontidae) in western North America. Mol Ecol Jones TR, Babb RD, Hensley FR, LiWanPo C, Sullivan BK 16(20):4335–4355 (2011) Sonoran Desert snake communities at two sites: Mazerolle MJ, Huot M, Gravel M (2005) Behavior of amphib- concordance and effects of increased road traffic. Herpetol ians on the road in response to car traffic. Herpetologica Conserv Biol 6(1):61–71 61(4):380–388 Kemp PS, O’hanley JR (2010) Procedures for evaluating and Mccardle LD, Fontenot CL (2016) The influence of thermal prioritising the removal of fish passage barriers: a synthe- biology on road mortality risk in snakes. J Thermal Biol sis. Fish Manag Ecol 17(4):297–322 56:39–49 Klauber LM (1931) A statistical survey of the snakes of the Mitrovich MJ (2006) A case study in conservation science: The southern border of California. Bull Zool Soc San Diego spatial ecology and evolutionary history of the coachwhip 8:1–93 snake (Masticophis flagellum) and striped racer (M. later- Kwiatkowski MA, Schuett GW, Repp RA, Nowak EM, Sullivan alis). Ph.D. dissertation. University of California, Davis, BK (2008) Does urbanization influence the spatial ecology of and San Diego State University. San Diego Gila monsters in the Sonoran Desert? J Zool 276(4):350–357 Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GA, Langen TA, Ogden KM, Schwarting LL (2009) Predicting hot Kent J (2000) Biodiversity hotspots for conservation pri- spots of herpetofauna road mortality along highway net- orities. Nature 403(6772):853–858 works. J Wildl Manag 73(1):104–114 NatureServe (2017) NatureServe explorer: An online encyclo- Langen TA, Andrews KM, Brady SP, Karraker NE, Smith DJ pedia of life [web application]. Version 7.1. NatureServe, (2015a) Road effects on habitat quality for small animals. Arlington, Virginia. http://explorer.natureserve.org. (Ac- In: Andrews KM, Nanjappa P, Riley SPD (eds) Roads and cessed: May 25, 2017) ecological infrastructure. Concepts and applications for Orłowski G (2007) Spatial distribution and seasonal pattern in small animals. Johns Hopkins University Press, Baltimore, road mortality of the common toad Bufo bufo in an agri- pp 21–41 cultural landscape of south-western Poland. Amphibia- Langen TA, Gunson KE, Jackson SD, Smith DJ, Ruediger W Reptilia 28(1):25–31 (2015b) Planning and designing mitigation of road effects Peaden JM, Tuberville TD, Buhlmann KA, Nafus MG, Todd BD on small animals. In: Andrews KM, Nanjappa P, Riley SPD (2016) Delimiting road-effect zones for threatened species: (eds) Roads and ecological infrastructure. Concepts and implications for mitigation fencing. Wildl Res applications for small animals. Johns Hopkins University 42(8):650–659 Press, Baltimore, pp 146–177 Peaden JM, Nowakowski AJ, Tuberville TD, Buhlmann KA, Langton TES (2015) A history of small animal road ecology. Todd BD (2017) Effects of roads and roadside fencing on Roads and ecological infrastructure. Concepts and appli- movements, space use, and carapace temperatures of a cations for small animals. Johns Hopkins University Press, threatened tortoise. Biol Conserv 214:13–22 Baltimore, pp 21–41 Pianka ER (1966) Convexity, desert lizards, and spatial Lannoo MJ (2005) Amphibian declines: the conservation status heterogeneity. Ecology 47(6):1055–1059 of United States species. University of California Press, Pilliod DS, Welty JL, Stafford R (2013) Terrestrial movement Berkeley patterns of western pond turtles (Actinemys marmorata)in Levine KK (2013) California amphibian and reptile crossing central California. Herpetol Conserv Bio 8(1):207–221 preliminary investigation. Report to Caltrans Division of Polak T, Rhodes JR, Jones D, Possingham HP (2014) Optimal Environmental Planning planning for mitigating the impacts of roads on wildlife. Linkov I, Satterstrom FK, Kiker G, Batchelor C, Bridges T, J Appl Ecol 51(3):726–734 Ferguson E (2006) From comparative risk assessment to Rathbun GB, Scott NJ Jr, Murphey TG (2002) Terrestrial habitat multi-criteria decision analysis and adaptive management: use by Pacific pond turtles in a Mediterranean climate. recent developments and applications. Environ Int Southwest Nat 47(2):225–235 32(8):1072–1093 Reese DA, Welsh HH (1997) Use of terrestrial habitat by Litvaitis JA, Tash JP (2008) An approach toward understanding western pond turtles, Clemmys marmorata: implications wildlife-vehicle collisions. Environ Manag 42(4):688–697 for management. In Proceedings: Conservation, restora- Loraamm RW, Downs JA (2016) A wildlife movement tion, and management of tortoises and turtles–an interna- approach to optimally locate wildlife crossing structures. tional conference. State University of New York, Purchase, Int J Geogr Inf Sci 30(1):74–88 New York, pp 352–357) 123 Landscape Ecol (2018) 33:911–935 935 Reid BN, Peery MZ (2014) Land use patterns skew sex ratios, Steen DA, Aresco MJ, Beilke SG, Compton BW, Condon EP, decrease genetic diversity and trump the effects of recent Dodd KC, Forrester H, Gibbons JW, Greene JL, Johnson G, climate change in an endangered turtle. Divers Distrib Langen TA (2006) Relative vulnerability of female turtles 20(12):1425–1437 to road mortality. Anim Conserv 9(3):269–273 Roe JH, Gibson J, Kingsbury BA (2006) Beyond the wetland Sullivan BK (2012) Road riding. In: McDiarmid RW (ed) border: estimating the impact of roads for two species of Reptile biodiversity: standard methods for inventory and water snakes. Biol Conserv 130(2):161–168 monitoring. University of California Press, Oakland, CA, Rosen PC, Lowe CH (1994) Highway mortality of snakes in the pp 215–218 Sonoran Desert of southern Arizona. Biol Conserv Suter GW II (2016) Ecological risk assessment. CRC Press, 68(2):143–148 Boca Raton, FL Rosenberg D, Gervais J, Vesely D, Barnes S, Holts L, Horn R, Thomson RC, Wright AN, Shaffer HB (2016) California Swift R, Todd L, Yee C (2009) Conservation assessment of amphibian and reptile species of special concern. Univer- the Western pond turtle in Oregon (Actinemys marmorata). sity of California Press, Oakland, CA USDI Bureau of Land Management and Fish and Wildlife Trenham PC (1998) Demography, migration, and metapopula- Service, USDA Forest Service Region 6, Oregon tion structure of pond-breeding salamanders. Ph.D. dis- Rouse JD, Willson RJ, Black R, Brooks RJ (2011) Movement sertation. University of California, Davis and spatial dispersion of Sistrurus catenatus and Hetero- Trenham PC, Koenig WD, Shaffer HB (2001) Spatially auto- don platirhinos: implications for interactions with roads. correlated demography and interpond dispersal in the Copeia 3:443–456 salamander Ambystoma californiense. Ecology Row JR, Blouin-Demers G, Weatherhead PJ (2007) Demo- 82(12):3519–3530 graphic effects of road mortality in black ratsnakes (Elaphe Trenham PC, Koenig WD, Mossman MJ, Stark SL, Jagger LA obsoleta). Biol Conserv 137(1):117–124 (2003) Regional dynamics of wetland-breeding frogs and Rudolph DC, Burgdorf SJ, Conner RN, Schaefer RR (1999) toads: turnover and synchrony. Ecol Appl 13(6):1522– Preliminary evaluation of the impact of roads and associ- 1532 ated vehicular traffic on snake populations in eastern US Fish and Wildlife (USFWS) (2016) United States species: Texas. In: Proceedings of the Third International Confer- Endangered and threatened wildlife (50 CFR 17.11). ence on Wildlife Ecology and Transportation. FL-ER-73- https://www.fws.gov/endangered/species/us-species.html 99. Evink GL, Garrett P, Zeigler D (eds) Florida Depart- (accessed August 2016) ment of Transportation, Tallahassee, Florida, pp 129–136 US Fish and Wildlife Service (USFWS) (2002) Recovery plan Rytwinski T, Fahrig L (2012) Do species life history traits for the California red-legged frog (Rana aurora draytonii). explain population responses to roads? A meta-analysis. US Fish and Wildlife Service, Portland, Oregon, viii ? Biol Conserv 147(1):87–98 173 pp Rytwinski T, Fahrig L (2013) Why are some animal populations US Fish and Wildlife Service (USFWS) (2009) Rana pretiosa. unaffected or positively affected by roads? Oecologia Species assessment and listing priority assignment form. 173(3):1143–1156 USFWS Region 1, Portland Rytwinski T, Fahrig L (2015) The impacts of roads and traffic on US Fish and Wildlife Service (USFWS) (2011) Alameda terrestrial animal populations. Handbook of Road Ecology. whipsnake (Masticophis lateralis euryxanthus) 5-year Wiley, New York, pp 237–246 review summary and evaluation. US Fish and Wildlife Semlitsch RD (1998) Biological delineation of terrestrial buffer Service, Sacramento, CA zones for pond-breeding salamanders. Conserv Biol van der Ree R, Smith DJ, Grilo C (2015) Handbook of road 12(5):1113–1119 ecology. Wiley, New York Semlitsch RD (2008) Differentiating migration and dispersal Vangel MG (2015) Tolerance interval. Statistics Reference processes for pond-breeding amphibians. J Wildl Manag Online. Wiley StatsRef, New York, pp 1–7 72(1):260–267 Vences M, Wake DB (2007) Speciation, species boundaries and Semlitsch RD, Bodie JR (2003) Biological criteria for buffer phylogeography of amphibians. Amphib Biol 7:2613–2671 zones around wetlands and riparian habitats for amphibians Watson JW, McAllister KR, Pierce DJ (2003) Home ranges, and reptiles. Conserv Biol 17(5):1219–1228 movements, and habitat selection of Oregon Spotted Frogs Shilling F, Waetjen D (2017) California Roadkill Observation (Rana pretiosa). J Herp 37(2):292–300 System. UC Davis Road Ecology Center. http:// Young KV, Young AT (2000) Final report: scientific study of wildlifecrossing.net/california(accessed April 2017) the flat-tailed horned lizard, Phrynosoma mcallii. US Dept. Siers SR, Reed RN, Savidge JA (2016) To cross or not to cross: of Navy Contracts N68711-95-LT-C0032, N68711-95-LT- modeling wildlife road crossings as a binary response C0035. 72 pp variable with contextual predictors. Ecosphere 7(5):19 Zimmermann Teixeira F, Kindel A, Hartz S, Mitchell S, Fahrig Stebbins RC, McGinnis SM (2012) Field guide to amphibians L (2017) When road-kill hotspots do not indicate the best and reptiles of California, vol 103. University of California sites for road-kill mitigation. J Appl Ecol 54(5):1544–1551 Press, Berkeley http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Landscape Ecology Springer Journals

An objective road risk assessment method for multiple species: ranking 166 reptiles and amphibians in California

Free
25 pages
Loading next page...
 
/lp/springer_journal/an-objective-road-risk-assessment-method-for-multiple-species-ranking-30JJKNYzDj
Publisher
Springer Netherlands
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Landscape Ecology; Ecology; Nature Conservation; Landscape/Regional and Urban Planning; Sustainable Development; Environmental Management
ISSN
0921-2973
eISSN
1572-9761
D.O.I.
10.1007/s10980-018-0640-1
Publisher site
See Article on Publisher Site

Abstract

Landscape Ecol (2018) 33:911–935 https://doi.org/10.1007/s10980-018-0640-1 RESEARCH AR TIC L E An objective road risk assessment method for multiple species: ranking 166 reptiles and amphibians in California . . Cheryl S. Brehme Stacie A. Hathaway Robert N. Fisher Received: 10 July 2017 / Accepted: 26 March 2018 / Published online: 9 May 2018 The Author(s) 2018 Abstract Results All chelonids, 72% of snakes, 50% of Context Transportation and wildlife agencies may anurans, 18% of lizards and 17% of salamander consider the need for barrier structures and safe species in California were ranked at high or very-high wildlife road-crossings to maintain the long-term risk from negative road impacts. Results were largely viability of wildlife populations. In order to prioritize consistent with local and global scientific literature in these efforts, it is important to identify species that are identifying high risk species and groups. most at risk of extirpation from road-related impacts. Conclusions This comparative risk assessment Purpose Our goal was to identify reptiles and method provides a science-based framework to iden- amphibians in California most susceptible to road tify species most susceptible to negative road impacts. mortality and fragmentation. With over 160 species The results can inform regional-scale road mitigation and a lack of species-specific research data, we planning and prioritization efforts and threat assess- developed an objective risk assessment method based ments for special-status species. We believe this upon road ecology science. approach is applicable to numerous landscapes and Methods Risk scoring was based upon a suite of life taxonomic groups. history and space-use characteristics associated with negative road effects applied in a hierarchical manner Keywords Reptile  Amphibian  Road mortality from individuals to species. We evaluated risk to both Habitat fragmentation  Road ecology  Risk aquatic and terrestrial connectivity and calculated assessment  Road buffer distances to encompass 95% of population- level movements. We ranked species into five relative categories of road-related risk (very-high to very-low) based upon 20% increments of all species scores. Introduction There have been many attempts to better characterize Electronic supplementary material The online version of and quantify threat criteria in order to classify species this article (https://doi.org/10.1007/s10980-018-0640-1) con- at higher risk of extinction at state, national, and global tains supplementary material, which is available to authorized users. levels (Congress 1973 (U.S. Endangered Species Act); Mace et al. 2008; Hobday et al. 2011; Thomson et al. C. S. Brehme (&)  S. A. Hathaway  R. N. Fisher 2016;IUCN 2017). Roads are a significant threat to U.S. Geological Survey, Western Ecological Research wildlife populations (e.g., Forman et al. 2003; Center, San Diego, CA, USA e-mail: cbrehme@usgs.gov 123 912 Landscape Ecol (2018) 33:911–935 Andrews et al. 2015a; van der Ree et al. 2015), causing systems, the use of barrier systems, underpasses, and both barrier (habitat fragmentation) and depletion overpasses can reduce road mortality and help to (road mortality) effects. Barrier effects occur when maintain connectivity and safe passage across roads animals avoid crossing roads, in which case roads for herpetofauna and other wildlife (Jochimsen et al. essentially fragment species habitat. Barrier effects 2004; Colino-Rabanal and Lizana 2012; Langton include reduced size and quality of available habitat, 2015; Langen et al. 2015b). Because it is currently reduced effective population size, reduced ability to unrealistic and cost prohibitive to mitigate all road- find mates and resources, increased genetic structur- ways for all species, it is vital to identify species most ing, and increased probability of local extirpation susceptible to road-related impacts. Within species (e.g., Forman et al. 2003; Fahrig and Rytwinski 2009; ranges, risks to populations and need for mitigation D’Amico et al. 2016). Depletion effects occur when can then be evaluated based upon local road densities animals attempt to cross roads and are killed by and matrix, road-types, traffic, and road locations in vehicles. Depletion effects include all of the risks from relation to species habitat and movement corridors barrier effects as well as reduced survivorship, making (e.g., Jaeger 2000; Litvaitis and Tash 2008; Langen high road mortality an even greater concern (Jackson et al. 2015b; Zimmermann Teixeira et al. 2017). and Fahrig 2011). Among other stressors, such as Here we describe a road risk assessment method- habitat loss and fragmentation, invasive species, ology applied to native amphibian and reptile species in California, a global biodiversity hotspot (Myers pesticide use, changing climate, and disease, the negative impacts from roads may independently or et al. 2000). We also included analysis of subspecies if cumulatively threaten the persistence of populations they had special federal or state protection status. This and even species. includes 166 species and subspecies of frogs, toads, Amphibians and reptiles have been identified as salamanders, snakes, lizards, turtles, and tortoise. being particularly susceptible to the negative effects of Rankings and prioritizations such as these can be very roads within their habitat (e.g., Klauber 1931; Forman subjective. In order to avoid including low risk species et al. 2003; Rytwinski and Fahrig 2012; Andrews et al. that may be favored by the assessors or to uninten- 2015a, b; D’Amico et al. 2015). Many are slow tionally overlook species that are at high risk, it was moving, do not avoid roads, and are simply too small important for this be done in an objective manner for drivers to see and avoid. During rains many informed by current road ecology literature. amphibians make long linear terrestrial movements Very few quantitative data are available on the regardless of the presence of intersecting roadways impact of roads on population persistence. Jaeger et al. (Glista et al. 2008), and because paved roads typically (2005) were the first to develop a relative ranking absorb and retain more heat than the surrounding system to compare the impact of roads on wildlife habitat, snakes and lizards are often attracted to roads populations. Their ranking system was largely based for thermoregulation (Case and Fisher 2001; Jochim- upon behavioral responses of animal species to the sen et al. 2004). In fact, road surveys are one of the road surface, road size, traffic noise, and vehicles with most common methods for surveying these reptiles varying road sizes and traffic volumes. However, (e.g., Sullivan 2012). Many herpetofauna species knowledge of these detailed behavioral responses to utilize both aquatic and terrestrial habitat for breeding, ranges in road and traffic characteristics is rarely found development, foraging, and overwintering and there- in literature and the link between individual behavior fore require connectivity within and between both and population-level effects has not been clearly aquatic and terrestrial habitats to support basic life established (Rytwinski and Fahrig 2012, 2013). history requirements. Rytwinski and Fahrig (2012) performed a meta- The primary goal of this study was to provide analysis of wildlife groups to test whether certain life information to transportation and other planning history characteristics were related to negative agencies in California to assist them in prioritizing responses to roads. High reproductive rate (fecundity) road mitigation efforts for amphibian and reptile was negatively associated with the magnitude of species. Although there is still a lot to learn about the population-level effects for amphibians. No associa- effectiveness of different designs of road mitigation tions were significant in reptiles, although there were 123 Landscape Ecol (2018) 33:911–935 913 few studies to inform this analysis. However, a strong foragers, seasonal migrants, sit-and-wait predators vs. link was shown between body size, greater mobility, sedentary species) (e.g., Bonnet et al. 1999; Carr and lower reproductive rates and the magnitude of nega- Fahrig 2001). Because many species are semi-aquatic, tive road effects in mammals, the most studied wildlife movement distance and frequency were scored sepa- group. Conversely, simulations predicted populations rately for both aquatic and terrestrial habitats. of species with small home ranges and high reproduc- There is a theorized higher risk associated with tive rates were the least likely to be affected by roads depletion effects (i.e., road mortality) in comparison to (Rytwinski and Fahrig 2013). barrier effects (Fahrig and Rytwinski 2009; Jackson We used these findings as a basis for creating a and Fahrig 2011). Therefore, we gave additional multi-tiered system to rank and identify reptile and weight to those species more likely to go out onto a amphibian species that may be most susceptible to road surface and be killed by vehicular traffic. For this road impacts. We based our ranking upon a suite of we considered factors of habitat preference (e.g., open species life history and space-use characteristics vs. closed), roads as potential attractants (e.g., for associated with negative road effects, as well as basking), and movement speed (e.g., slow vs. fast). including species distribution and conservation status. However, individuals within and among species may We evaluated risk to both aquatic and terrestrial respond differently to roads (attraction vs. avoidance) connectivity and include buffer distances that were based upon local landscape features, road width, traffic volume, and perceived danger (Forman et al. calculated to encompass 95% of population move- ments. Relative confidence in these distances is given 2003; Andrews 2005; Brehme et al. 2013; Jacobson for each species based upon the amount of support et al. 2016). Because a state-wide analysis encom- from scientific studies. We solely focused on the direct passes extreme variation in landscape and road effects of roads as barriers and sources of road characteristics, the extent to which roads act as mortality and not impacts from road construction barriers or sources of direct mortality within a species and maintenance or indirect effects from increased range is unknown. The risk disparity between deple- human use of the landscape once a road is in place (see tion and barrier effects could also be highly variable. review by Langen et al. 2015a). Therefore, we limited the additional weight for Because we based the risk assessment solely upon potential depletion effects to twenty percent of the space-use and life history characteristics, this repre- individual risk score. sents a species relative susceptibility to road impacts. We assessed population-level road risk by multi- It is understood that circumstances associated with plying individual risk with scores representing: (1) the particular populations (e.g., local road types, loca- relative proportion of the population at risk; and (2) the tions, densities) may elevate or reduce the risk for species ability to sustain higher rates of mortality. For certain populations and species. instance, the proportion of the population at risk was expected to be higher for migratory species than for territorial species. Highly fecund species were Methods expected to better withstand (or more quickly recover from) higher mortality in comparison to those with Road risk assessment (overview) few annual offspring. Finally, we assessed species-level road risk by multiplying population road risk with scores for range We assessed the relative risk of California herpeto- fauna species to negative road-related impacts at three size (both within and outside of California) and scales in a hierarchical fashion. We first assessed risk conservation status according to the U.S. Fish and at the scale of an individual animal and then expanded Wildlife Service (USFWS 2016) and the California the risk to the population and then to species (Fig. 1). Department of Fish and Wildlife (CDFW 2016a; At the individual-level, we based road risk primar- Thomson et al. 2016). Species with smaller ranges ily upon the likelihood that an individual would typically have fewer populations and are thus less encounter one or more roads. We considered this a resilient to population-level stressors. Endangered, product of movement distance (home range, seasonal threatened, and special concern species have already migrations) and movement frequency (e.g., active been designated at risk of extirpation, often due to 123 914 Landscape Ecol (2018) 33:911–935 Fig. 1 California reptile and amphibian road risk assessment conceptual model (ARSSC Amphibian and Reptile Species of Special Concern (Thomson et al. 2016)) multiple stressors, and are thus thought to be less 5. Conservation Status of Amphibians and Reptiles likely to be resilient to additional road impacts. on USDA National Forests, Pacific Southwest Region, 2012 (Evelyn and Sweet 2012). Although we present both aquatic and terrestrial risk scores for semi-aquatic species, we used the 6. Natureserve Explorer (natureserve.org): Species Accounts largely authored by G. Hammerson higher of the two scores for the overall risk ranking. (2003–2016). Literature review When these reviews were lacking life history information needed for the road risk assessment, we Species life history data were primarily taken from and then searched for supplementary peer-reviewed liter- cross-checked among the following species account ature using the Google Scholar search engine. Because review sources; movement distances (terrestrial, aquatic, home range, 1. U.S. Fish and Wildlife Service (USFWS) Recov- migratory) were so important for the risk assessment, ery Plans and 5-year Reviews https://www.fws. we acquired referenced articles from the species gov/endangered/. accounts and independently searched the literature to 2. California Amphibian and Reptile Species of acquire these data. Search terms included the species Special Concern (ARSSC; Thomson et al. 2016). common name, scientific name, or genus and terms 3. A Field Guide to Amphibians and Reptiles of such as ‘‘movement’’, ‘‘home-range’’, ‘‘spatial’’, and California (Stebbins and McGinnis 2012) ‘‘telemetry’’. We also reviewed articles for citations of 4. Amphibian declines: the conservation status of other studies to find more recent information on movement. This literature included published articles, United States species (Lannoo 2005). 123 Landscape Ecol (2018) 33:911–935 915 book chapters, M.S. Theses, Ph.D. dissertations, Table 1 Individual-level Road Risk (IRR): Score criteria for risk of individuals encountering a road agency reports, and consultant reports. In the case that specific life history or movement information was Risk of individuals encountering a road = Movement distance 9 not found for a species, we chose a surrogate species frequency based upon phylogeny, habitat, and body size. We first Movement Score Frequency Score looked for the closest related species within the genus distance (m) or family and chose a closely related surrogate based [ 1200 40 Active throughout home range 2 upon similar habitat and body size. If surrogates were 901–1200 32 Migratory (2–4 9 per year)/ 1.5 used, these are clearly reported. non-migratory sit and wait foragers Road risk metrics 601–900 24 Sedentary, confined to 1 specialized habitat The following section describes in detail the rank 451–600 16 scoring used for Individual-level Road Risk, Popula- 301–450 12 tion-level Road Risk, and Species-level Road Risk. 201–300 8 All rank values are meant to represent the relative 101–200 5 contribution of each attribute to either additive or 51–100 3 multiplicative road risk. 0–50 1 Individual-level risk (100 points possible) Out of a total of 100 points for individual road account for the majority of movement for these species. mortality risk, we attributed up to 80 points (80%) to the risk of encountering a road and up to 20 points The calculations and rankings for movement distances were well considered and deserve further (20%) for the risk of an individual moving onto a road and being killed by a motor vehicle. explanation. Our original thinking was that maximum distances should reflect relative movement distances The risk of encountering a road was based on a combination of movement distance and general across species and these data were commonly reported movement frequency. Movement distance was ranked in species accounts. However, it became increasingly 1–40 based upon home range movement distances difficult to determine whether maximum distances (diameter) for non-migrants or migration distances for reported were seasonal migration movements, home range movements or rarer dispersal events. We seasonal migrants that spanned from 0 to [ 1200 m (Table 1). The scores are linearly correlated with believed this assessment should reflect annual move- ment distances and not rare dispersal events. We increasing movement distance. For species that use both terrestrial and wetland/ considered using average/median movement dis- tances; however, these often underestimate the move- stream/riverine habitats, such as frogs, toads, aquatic snakes and turtles, we scored aquatic and terrestrial ment of seasonal migrants because in many cases a sizeable portion of the population may remain close to movement distances and frequencies separately. This was necessary as some species move much larger a breeding site, while another sizable portion make distances and at different frequencies in one habitat longer distance migrations causing an average or versus the other. This also informs the type(s) of median to be uninformative. Therefore, we decided to mitigation structures that may be warranted based use a buffer distance that incorporates the movement distances of 95% of the population studied. A 95% upon habitat type, buffer distances and risk scores for each species. Aquatic movement distances were not population movement distance is commonly accepted for the delineation of terrestrial buffer zones for calculated for pond-breeding amphibians. Ponds are typically small ephemeral bodies of water and terres- amphibians (i.e., Semlitsch 1998; Semlitsch and Bodie 2003) and we believe it was the most biologically trial movements of amphibians to and among ponds 123 916 Landscape Ecol (2018) 33:911–935 meaningful and useful measure for this study. This suitable overwintering sites are typically much farther measure, which we will refer to as Maximum Popu- away from breeding and summer activity areas than in lation Movement Distance (MPMD), should include milder California climates (e.g., Gregory 1984). We almost all population movements, such as seasonal did use study data from adjacent states or lower migration distances and annual home ranges (diame- estimates of migration distances from those reported ter), but not rare dispersal events. The MPMD should in Midwestern states. In some cases where little also be useful for local risk assessments as these information was available, we made an educated guess distances can be used to aide in mapping and based upon limited study data and/or closely related mitigation decisions. species and noted these in the tables. For all MPMDs, The calculation we used for MPMD is commonly we report a relative confidence level based upon the known as the 95% upper tolerance interval (Vangel number and quality of studies, sample sizes, and 2015). A tolerance interval is an interval that is meant locations in or adjacent to California. It is intended that to contain a specified percentage of individual popu- the scores be adjusted as new information becomes lation measurements. This should not be confused available. with a confidence interval, which is an interval that is To compute the risk of encountering a road, the meant to contain the population mean. We chose a MPMD was multiplied by a relative index of the 50% confidence level for the upper 95% confidence expected frequency of longer distance movements (1–2 points; Table 1). We defined three frequency limit of movement distances which is equal to the 95% prediction interval for future observations and is the categories largely based upon annual migratory mean ? 1.645 9 standard deviation. In cases where a movements or foraging strategies for non-migratory standard deviation was not reported, we back calcu- species. The highest category included actively for- lated standard deviation from the standard error and aging predators which are characterized by frequent sample size, calculated it from the individual data, or wandering movements throughout their home range estimated it based on the methods recommended by (Pianka 1966). Less frequent movers included sea- Hozo et al. (2005). Although non-parametric tolerance sonal migrants traveling among breeding, summer intervals would be more appropriate for non-normally foraging, and/or overwintering sites and non-migra- distributed movement data, the data required to tory ‘sit-and-wait’ predators that remain still for long calculate these is rarely reported in the published periods of time to ambush prey (Pianka 1966). Finally, literature. In the case of non-normally distributed data low frequency included highly sedentary species with where medians, sample sizes and ranges are reported, high site fidelity, particularly specialized rock, cre- Hozo et al. (2005) methods allow for approximation of vice, soil, or tree dwellers that may rarely traverse means and standard deviations with no assumption of terrestrial or aquatic habitats. the underlying data distribution. We found the result- The risk of an individual moving onto a road and ing MPMDs to be reasonable in excluding large being killed by a moving vehicle was ranked by outliers but including multiple long distance move- attributes of habitat preference, road use, and move- ments below the maximum movement distance. ment speed (Table 2). Habitat preference represents We recognize that for any species there can be the degree to which an individual is expected to go out substantial variability in movement distances that onto or avoid an open road as predicted from their depend upon varying local, landscape, and climatic habitat and microhabitat preferences. Open habitat factors. This was often reflected in studies with specialists and generalists were expected to more sometimes widely varying estimates of home range readily move onto a road than species that prefer cover and migration distances. We attempted to be conser- (e.g., Forman et al. 2003; Brehme et al. 2013). vative by using the study data for calculation of Although many amphibians are closed habitat spe- MPMD in which the largest population movement cialists, most readily move through open habitats distances were observed. For studies where movement during rain events, when most overland migratory distance significantly varied between females and movements tend to occur (Glista et al. 2008). There- males, we used the information from the wider ranging fore, amphibians were considered open habitat spe- sex. For migratory distances, we did not use distances cialists for this ranking. An additional factor that may from extreme environments, such as Canada, where increase road use is for thermoregulation for lizards 123 Landscape Ecol (2018) 33:911–935 917 Table 2 Individual-level Road Risk (IRR): Score criteria for risk of road mortality Risk of road mortality = Habitat preference ? road use ? movement speed Habitat preference Score Road use Score Movement speed Score Open habitat specialist/amphibians 10 Thermoregulation (snakes/lizards) 4 Slow (\ 0.6 m/s) 6 Generalist 8 Other 0 Medium (0.6–2.0 m/s) 3 Edge specialist 4 Fast ([ 2.0 m/s) 0 Closed habitat or aquatic specialist 0 Table 3 Population-level Road Risk (PRR): Score criteria for population level road risk PRR = IRR 9 (Fecundity ? Proportion of population at risk) Fecundity Ave. potential offspring/year Score Proportion of population at risk Score Low 0–10 2 Seasonal migrants (Migratory) 2 Med 11–25 1.5 Wandering 1.5 High 26–100 1 Territorial 1 Very high [ 100 0 and snakes, as roads often retain more heat than the potential offspring was calculated by multiplying the surrounding environment (Colino-Rabanal and Lizana average clutch size by the average number of clutches 2012; Mccardle and Fontenot 2016). Finally, there is per year. an increased risk of road mortality for slow versus fast Individual mortality risk (1–100 points) was mul- moving species (see Andrews and Gibbons 2005; tiplied by the sum of these population-level factors Mazerolle et al. 2005; Andrews et al. 2015b). (1–4 points) to calculate population-level road risk. Population-level Road Risk (400 points possible) Species-level road risk (1200 points possible) To assess the risk of negative road impacts on the In comparison to population-level risk, we considered persistence of a population we incorporated scores for the overall risk of roads to species to be negatively population-level movement behavior and fecundity associated with species range and conservation status. (Table 3). For the proportion of a population expected Although some populations may be at high risk, to encounter a road, we scored the greatest risk to species with a wide distribution and many populations species that seasonally migrate to overwintering and should be more resilient to localized declines and breeding areas (Jackson et al. 2015). For those that do extirpations. Therefore, we assigned a range isolation not migrate, we expected higher proportions of non- score ranging from 0 to 1 that considered species territorial or loosely territorial species (‘‘wandering’’) distributions range-wide (North America) and within to encounter roads than species that defend distinct California (CA) (Table 4). Range-wide distribution territories. varied from ‘‘CA only’’ to ‘‘widespread’’ ([ 4 states). Species with low fecundity are less resilient to road If the species range extended into Mexico and/or mortality impacts than highly fecund species (Rytwin- Canada, these countries were counted as another state ski and Fahrig 2013). Relative fecundity was simply for calculation of the index. California-wide distribu- calculated from the average number of potential tion was calculated based upon the number of CA offspring per year whether the animals were oviparous geographic regions occupied out of twelve regions or live-bearing. For egg-laying species, the number of defined by Hickman (1993) and used in Stebbins and 123 918 Landscape Ecol (2018) 33:911–935 Table 4 Species-level Road Risk (SRR): Score criteria for species-level road risk SRR = PRR 9 ((Range isolation score ? Conservation status score)/2) (a) Range isolation score = (North America range ? CA range)/2 North America range Rank/score CA only 1.00 2 states (very restricted distribution) 1.00 2 states (restricted) 0.67 2–3 states 0.33 Widespread (4 ? states) 0.00 California range (No. of geographic regions occupied) Rank/score 1 0.92 2 0.83 3 0.75 4 0.67 5 0.58 6 0.50 7 0.42 8 0.33 9 0.25 10 0.17 11 0.08 12 0.00 (b) Conservation status score Conservation status Rank/score CA or federal threatened/endangered 1.00 SSC priority 1 0.75 SSC priority 2 0.50 SSC priority 3 0.25 None 0.00 Population-level risk [ 80 only McGinnis (2012). These two scores (Range-wide from 0.25 to 0.75 based upon their SSC ranking. isolation, CA isolation) were summed and divided Population-level Road Risk (score range 1–400) was by two in order to normalize the overall range isolation multiplied by (1 ? Range Isolation Score ? Conser- score to a 0 to 1 scale. vation Status Score; score range 1–3) to calculate the At the species-level, we also incorporated conser- final Species-level Road Risk. vation status (Table 4). Some species are declining Range and conservation status were only used as a and are at higher risk of extinction often due to multiplier for species-level road risk if the population- multiple stressors. Federal and State Threatened and level road risk was greater than 80 (20% of possible Endangered Species were given the highest score population score). This helped to prevent false infla- (1.0). In California, forty-five species are designated tion of the road risk metrics for low road susceptible ‘‘Species of Special Concern (SSC)’’ with a ranking of species. 1, 2, or 3 based upon severity and immediacy of threats Because all members of the genus Batrachoseps affecting each taxon (Thomson et al. 2016). SSC (slender salamanders) are similar in body size, range species were given a conservation status score ranging size and general life history characteristics, we scored 123 Landscape Ecol (2018) 33:911–935 919 Table 5 Species-level frequency distributions and road risk As a way to support the results of our ranking model rankings with species literature, we focused on special status species. We reviewed recovery plans and 5-year Percentile Scores Relative ranks reviews for federally listed species and state species 81–100 322–710 Very high accounts for California listed species and species of 61–80 213–321 High special concern (collectively referred to as special 41–60 63–212 Medium status species). For each rank group (i.e., ‘‘very low’’ 21–40 53–62 Low to ‘‘very high’’), we calculated the percentage of 1–20 0–52 Very Low special status species where roads were specifically listed as a threat. Similarly, we tallied the number of species identified in a recent California preliminary road risk assessment (Levine 2013, Amy Golden pers. comm.) and compared the number of species that fell the genus as whole with the most conservative within each of our road risk categories. estimates and conservation status but included all 20 species in the final count and calculations. Once all 166 species (including subspecies with Results conservation status) were scored for species-level road risk within both terrestrial and aquatic habitats, we All chelonids, 72% of snakes, 50% of anurans, 18% of took the maximum score for each species and sorted lizards and 17% of salamander species were ranked as them from the highest to lowest scores. We grouped high or very high risk from negative road impacts. species into categories of risk (Very high, high, (Table 6, Fig. 3). medium, low, and very low) based upon ranges of Review of species accounts, recovery plans, and values that represented frequency distributions in 20% 5-year reviews for all special status species showed increments of all species scores (Table 5, Fig. 2). Fig. 2 Histogram of species-level scores and approximate 20 percentile road risk categories 123 920 Landscape Ecol (2018) 33:911–935 Table 6 Numbers of species by taxa within each risk category Species group Species-level rankings Very high High Med Low Very low Salamander 4 4 3 26 9 Lizard 5 3 8 7 21 Anuran 5 6 6 4 1 Snake 15 21 13 0 1 Tortoise 1 0 0 0 0 Turtle 3 0 0 0 0 4/4 100% 80% 36/50 7/11 60% 4/11 40% 8/38 8/44 20% 0% Salamander Lizard Frog Toad Snake Chelonid Fig. 3 Percentages of species by taxa in high and very high road risk categories that 94% (17/18) of species accounts that referenced special status species accounts mentioned roads as a roads as a threat to the species were ranked as ‘‘high’’ potential threat. In addition, 79% (15/19) of species of or ‘‘very high’’ in our risk assessment (Table 7). Of the concern recommended in a recent Caltrans prelimi- special status species that ranked ‘high’ and ‘very nary road risk assessment scored as ‘high’ or ‘very high’, close to fifty percent (17/35) had road-related high’ risk in our analysis (Levine 2013, Amy Golden threats referenced in their listing literature. In com- pers. comm.). parison, only 4% (1/27) of ‘medium’ to ‘very low’ risk Table 7 Comparison of road risk results and number of special status species with roads listed as threat Road risk level Special status species Caltrans PI No. species in road risk level No. species with roads listed as threat % of Total No. Spp in road risk level Very high 25 14 56 11 High 11 3 27 4 Medium 5 1 20 3 Low 10 0 0 1 Very low 7 0 0 0 Caltrans PI are Caltrans identified sensitive species % High and Very High Risk Landscape Ecol (2018) 33:911–935 921 Table 8 Amphibian and reptile road risk assessment: very range), low risk species (20–40% percentile) and very low risk high risk species (80–100% percentile), high risk species species (0–20% percentile) (60–80% percentile), medium risk species (40–60% percentile Discussion Risk scores and relative rankings for California reptile and amphibian species in both terrestrial and To our knowledge, this is the first attempt to objec- aquatic habitats are presented in Tables 8. Terrestrial and Aquatic rankings are provided separately in tively assess the relative risk of roads at a species level Tables 9 and 10 and also include population-level using a logical and scientifically based framework and risk scores, 95% population buffer distances, confi- apply it across a large array of species and habitats. We dence levels, and identification of any surrogate believe this approach could be useful for assessing and species used for the distance calculations. Species comparing susceptibility of species to negative road scores for all ranking criteria and life history and impacts within and among all taxonomic groups. To date, such risk assessments have been based largely movement references are provided in Appendices 1 and 2. upon expert opinion, limited information available on 123 922 Landscape Ecol (2018) 33:911–935 Table 8 continued road mortality, and even less information available on aquatic habitats to further inform mitigation. Some population or species-level road effects (Levine 2013; aquatic species may greatly benefit from fish passages Rytwinski and Fahrig 2015). while others may better benefit from terrestrial barriers Overall, this is meant to be a first step in highlight- and wildlife crossings or both. ing reptile and amphibian species that may be at Although data are currently lacking to validate highest risk from roads transecting their habitat. These completely the scoring and results of the risk assess- species may deserve consideration for further study ment, our review of species accounts, recovery plans, and for implementing mitigation solutions to reduce 5-year reviews for federal and state-listed species and mortality and to maintain or enhance connectivity. California species of special concern show a strong The risk assessment was done for both terrestrial and association between elevated road risk from our 123 Landscape Ecol (2018) 33:911–935 923 Table 8 continued objective analysis and the probability that roads are resulting in less risk to aquatic movement of popula- listed as a potential threat to the species in the species tions that inhabit lake and river systems. However, listing literature. culverts that are more commonly constructed under Although more than 40% of special status species roads in streams and wetlands vary in passability are semi-aquatic, roads were rarely considered a threat depending on factors such as diameter, length, slope, to aquatic connectivity in the species literature. This outlet configuration, and other characteristics (Furniss may be accurate if bridges or large culverts currently et al. 1991; Clarkin et al. 2005; Kemp and O’Hanley exist for water flow that also provide permeability to 2010). In fact, Januchowski-Hartley et al. (2013) aquatic movement. Bridges are generally considered found that only 36% of road crossings were fully passable to fish in the Great Lakes basin. In addition, to be completely passable by all aquatic species. Bridges are more likely to be constructed adjacent to many low water crossings in arid regions of the state or over large water bodies and rivers, presumably are simply a dip in the road that allows water to flow 123 924 Landscape Ecol (2018) 33:911–935 Table 8 continued 123 Landscape Ecol (2018) 33:911–935 925 Table 9 Terrestrial risk ranking and population buffer distances 123 926 Landscape Ecol (2018) 33:911–935 Table 9 continued 123 Landscape Ecol (2018) 33:911–935 927 Table 9 continued over the surface during high flow events. These may be range and/or migratory), tend not to avoid roads (or are used as road crossings by species traveling along attracted to them for thermoregulation), are long lived, ephemeral stream corridors with or without water and have relatively low fecundity in comparison to flow. Given these potential vulnerabilities, we believe other herpetofaunal groups. Because of these traits, that road impacts to aquatic connectivity of herpeto- chelonids and snakes have been identified elsewhere as fauna deserve greater consideration. being particularly susceptible to negative population Across broad taxonomic groups, chelonids (tor- effects from roads (Gibbs and Shriver 2002;Andrews toises/turtles) and snakes had the greatest percentages et al. 2015b; Jackson et al. 2015). of species at ‘high’ or ‘very high’ risk from roads. They There are only four species of chelonids in are similar in that many move long distances (home California, (desert tortoise (Gopherus agazzii), 123 928 Landscape Ecol (2018) 33:911–935 Northwestern pond turtle (Actinemys marmorata), and underpasses (Dodd et al. 2004; Colley et al. 2017). Southwestern pond turtle (Actinemys pallida), and In our statewide risk analysis, coachwhips (genus the Sonoran mud turtle (Kinosternon sonoriense)). Masticophis/Coluber) were amongst the highest risk There has been a high level of attention to road groups at both the population and species-levels. impacts on the desert tortoise (Gopherus agazzii)as These are particularly wide-ranging and very active numerous studies have documented not only high road foragers in comparison to other snake genera (Stebbins mortality, but measurable road effect zones, and and McGinnis 2012). The coachwhip (Masticophis mostly positive responses to barriers and underpasses flagellum) was found to be ninefold more likely to be (e.g., Boarman and Sazaki 1996, 2006; Peaden et al. extirpated from habitats that were fragmented by roads 2016; but see Peadon et al. 2017). Although not listed and urbanization, contributing to their decline as a primary threat to pond turtle populations in throughout California (Case and Fisher 2001; Mitro- California (Thomson et al. 2016), road mortality is a vich 2006). Similarly, habitat fragmentation from major concern for western pond turtle populations in roads and urbanization were identified as primary Oregon (Rosenberg et al. 2009). Pond turtles travel threats to the Alameda whipsnake (Masticophis later- kilometers within perennial waters and from pool to alis euryxanthus USFWS 2011). Although road use pool in intermittent aquatic habitats to forage and find and mortality have been documented for many other mates (Goodman and Stewart 2000). In addition, terrestrial California snake species on road-riding females nest and lay eggs in terrestrial habitats up to surveys (e.g., Klauber 1931; Jones et al. 2011; Shilling 0.5 km away from water which make roads that and Waetjen 2017), there is a paucity of studies parallel aquatic habitat a threat to both females and examining population-level effects of roads on Cali- hatchlings (Reese and Welsh 1997; Rathbun et al. fornia snake species. We could find only one such 2002; Pilliod et al. 2013). In fact, road mortality of study, where presence of a highway was shown to females has been identified as a cause for male-biased reduce gene flow in the Western diamond-backed sex ratios in some populations of pond turtles and rattlesnake (Crotalus atrox) in the Sonoran Desert, AZ other freshwater turtle species (Steen et al. 2006; (Hermann et al. 2017). Rosenberg et al. 2009; Reid and Peery 2014). There- Long foraging movements within aquatic habitats fore, this species requires consideration of both also contributed to the majority of garter snakes aquatic and terrestrial connectivity to satisfy their (genus: Thamnophis) falling within the highest road annual resource requirements. Sonoran mud turtles risk categories. Maintaining aquatic and wetland also travel long distances within intermittent streams connectivity is of primary concern for these species. and thus may be at risk of roads that transect their Garter snakes also use terrestrial habitats for overwin- aquatic habitat (Hensley et al. 2010). tering, reproduction, and for moving among wetland Larger colubrid snakes (Family Colubridae; many or aquatic patches. Some migrate long distances to genera) and rattlesnakes (genus Crotalus) were ranked winter hibernacula, making them also susceptible to among the highest risk from negative road effects. In roads within adjacent terrestrial habitats (Roe et al. addition to being attracted to paved road surfaces for 2006; Jackson et al. 2015). The highly aquatic giant thermoregulation, many large snakes have wide home- garter snake (Thamnophis gigas) had the highest ranges or may move large distances between winter aquatic road risk score. Because it moves only short hibernacula and summer foraging areas. In contrast to distances on land (Halstead et al. 2015), mitigation smaller species, larger snakes are also less likely to may best focus on functional aquatic passages with avoid roads (Rosen and Lowe 1994; Andrews and lengths of adjacent road barriers based upon their Gibbons 2005; Andrews et al. 2008; Siers et al. 2016). terrestrial movement distances. High road mortality (e.g., Klauber 1931; Rosen and Toads were the third highest ranking group with Lowe 1994; Jones et al. 2011), reduced abundance 64% ranked in the highest risk categories. In partic- near roads (Rudolph et al. 1999; Jones et al. 2011), ular, Bufonid toads (family Bufonidae) may move increased extinction risk (Row et al. 2007), and large distances ([ 1 km) in both aquatic and terrestrial decreased genetic diversity (Clark et al. 2010; Her- habitats to satisfy their annual resource requirements; mann et al. 2017) have been documented for numerous thus 5 of 7 bufonid species ranked high or very high snake species; as have positive responses to barriers risk from roads. Consistent with our risk assessment 123 Landscape Ecol (2018) 33:911–935 929 Table 10 Aquatic risk ranking and population buffer distances results, there is evidence that bufonid toads are fecund, with relatively limited movements and thus particularly susceptible to negative impacts from ranked low for road impacts. Four of 11 species ranked roads elsewhere (Trenham et al. 2003; Orłowski within the highest risk groupings; California red- 2007; Eigenbrod et al. 2008). legged frog (Rana draytonii), Oregon spotted frog (R. Roads and traffic have been associated with pretiosa), Northern red-legged frog (R. aurora), and reduced abundance and species richness of frog Cascades frog (R. cascadae). The Oregon spotted frog populations (e.g., Fahrig et al. 1995; Houlahan and (R. pretiosa) is known to move large distances within Findlay 2003). However, approximately half of Cal- aquatic habitats (Bourque 2008; USFWS 2009). ifornia species are small, primarily aquatic, highly Construction of a highway that bisected the 123 930 Landscape Ecol (2018) 33:911–935 Yellowstone population of Oregon spotted frogs was migratory salamander species were ranked within the one important factor that reduced the population highest risk categories from negative road effects. dramatically in the 1950s (see discussion in Watson There is substantial evidence that habitat fragmenta- et al. 2003). Although portions of the populations tion and mortality due to roads negatively affect many show high site fidelity, California red-legged frog and of these species. For instance, newts regularly migrate Northern red-legged frog migrants can move large long distances over land from and to breeding ponds, distances ([ 1 km) across both aquatic and terrestrial and to terrestrial foraging habitats ([ 2 km; Trenham habitats (Bulger et al. 2003; Fellers and Kleeman 1998). Large numbers are found dead on roads during 2007; Hayes et al. 2007). Road mortality or habitat dispersal periods and newt species are often the first to fragmentation from roads and urbanization were listed disappear in fragmented landscapes (Gibbs 1998; as primary threats to these species elsewhere (USFWS Trenham 1998, Shields pers. comm.). Similarly, road 2002; COSEWIC 2015). mortality and habitat fragmentation are primary Lizards had relatively low percentages of species in threats to the California tiger salamander and other the high risk groupings. Many lizard species are small, Ambystomid salamanders because terrestrial habitat is non-migratory, territorial, have small home ranges and used for interpond migration and overwintering are thus at low risk of negative road effects. Similar to (Semlitsch 1998; Trenham et al. 2001; Bolster 2010). snakes, lizards can also be attracted to road surfaces for Because this assessment covers a wide array of thermoregulation. A few wide ranging species scored in species and habitats, the risk to particular species the highest risk categories including the Gila monster populations must be re-assessed on a local level. This (Heloderma suspectum), leopard lizards (genus Gam- includes consideration of the locations, types, and belia) and two horned lizard species (genus Phryno- densities of roads in relation to population and species soma). The Gila monster has been negatively associated ranges along with goals for functional, meta-popula- with urbanization, where larger home ranges and tion, and genetic connectivity (e.g., Marsh and Jaeger greater movement rates result in higher mortality for 2015). Due to very low road densities in their limited males (Kwiatkowski et al. 2008). Sensitive to habitat ranges, some species and populations may be at lower fragmentation, the blunt-nosed leopard lizard (Gambe- risk. For instance the Gila monster, Oregon spotted lia sila) was found to be largely absent from habitat frog, Sonoran mud turtle, Sonoran desert toad (Incilius patches less than 250 ha (Bailey and Germano 2015). alvarius) and Yosemite toad (Anaxyrus canorus) Flat-tailed horned lizards (Phrynosoma mccallii)are scored high due to life history and space-use charac- also susceptible to habitat fragmentation with very large teristics, however their limited ranges are largely in home ranges for their size, particularly in wet years protected or low road density areas in the state. Thus (Young and Young 2000). In fact, road mortality is a roads may not be a significant threat to these species in well-known threat for this species (see review by California. In contrast, high road densities may CDFW 2016b). Horned lizards are also particularly increase the risk for species within coastal regions vulnerable to being killed on roads due to their tendency such as remaining populations of Santa Cruz long-toed to flatten and remain motionless while being salamander (Ambystoma macrodactylum croceum), approached (Young and Young 2000). Alameda striped racer (Masticophis lateralis euryx- Salamanders also had relatively low percentages of anthus), and San Francisco garter snake (Thamnophis species in the high risk grouping. Over 75% (35/46) of sirtalis tetrataenia). However, most species consist of the California salamanders are lungless salamanders numerous populations with a myriad of differing road- (Plethodontidae) and Torrent salamanders (Rhyacotri- related threat levels. Although detailed species ranges tonidae). These species are mostly small, sedentary, and occupancy within ranges are well known for some non-migratory, closed habitat specialists with limited species with very limited ranges, for most species movement distances and these traits have resulted in a range-wide surveys have not been conducted. There- high level of speciation. This is exemplified by there fore, only general range boundaries are available that being at least 20 species of slender salamanders (genus encompass large portions of the state and availability Batrachoseps) in California alone (Martinez-Solano of species distribution models of habitat suitability and et al. 2007; Vences and Wake 2007). However, within occupancy within their ranges is rare. This lack of the salamander group, newts and several other detailed spatial information on species distribution 123 Landscape Ecol (2018) 33:911–935 931 further limits the potential to incorporate road loca- The quantity and quality of life history information, tions, types, and densities in a state and species-wide particularly movement data, are highly variable assessment. among species (see confidence levels; Tables 9 and We also note that relative risk to negative road 10). Therefore it is important to re-assess risk as new impacts is provided for both populations and species. information becomes available. Finally, this is a Risk was elevated for species with small and isolated structured assessment of comparative risk across a ranges and that are facing a myriad of other threats. range of target species; therefore specific values for Because of this, a few common widespread species high risk have not been established. The ranking or scored high at the population-level but not at the assessment methodology should be adaptive and species-level. This included gopher snakes (Pituophis updated with advancements of road ecology science catenifer) and western toads (Anaxyrus boreas) where (e.g., Linkov et al. 2006). road mortality has been identified as a threat to the persistence of local populations (e.g., COSEWIC 2012; Jochimsen et al. 2014). Conclusion To potentially aid in local assessments, we have provided distance estimates or ‘‘buffer zones’’ that Although roads are a significant cause of mortality and contain estimates for 95% of population-level move- habitat fragmentation for many wildlife populations, ments for all species (e.g., Semlitsch and Bodie 2003). road-related risk rankings have been based largely on We provide all references evaluated for distance expert opinion due to a scarcity of literature on road estimates in Appendix 2. Meta-population movements effects for most species. Therefore, we developed an can be very important to the stability of pond-breeding objective and scientifically-based comparative risk amphibians (e.g., Semlitsch 2008; Jackson et al. 2015) approach to assess the potential threat from negative and are included in many of the buffer zone calcula- road impacts using species life history and movement tions. However, we note that buffer zones may not data. After applying it to over 160 herpetofaunal species include meta-population-level movements if the rate (and subspecies) in the state of California, the results of these dispersal movements was less than 5% in the are consistent with road ecology literature in identifying studies we used for our analyses. known high risk species, and call attention to some This should be considered an initial assessment of species not previously identified. Overall, we found that susceptibility to negative road impacts in a hierarchi- snakes and chelonids had the largest proportion of cal framework (e.g., see Level 2; Hobday et al. 2011). species at high risk for negative road impacts due to Therefore, as previously stated it will be important to longer movement distances (home range and/or migra- re-assess the risk of specific populations to roads tory), lack of road avoidance, and relatively low within their habitat and to evaluate and compare fecundity in comparison to other herpetofaunal groups. alternatives at the local scale (e.g., Suter 2016). This Results also indicated that consideration of aquatic may include more detailed information on specific connectivity appears to be under-represented for semi- road attributes (e.g., density, type, location), as well as aquatic herpetofauna that use both terrestrial and species behavior (Jaeger et al. 2005; Rouse et al. 2011; stream, riverine, or wetland habitats. Rytwinski and Fahrig 2013; Jacobson et al. 2016). Age In addition to informing transportation planning structured and spatially explicit population viability and mitigation considerations for California herpeto- models are valuable tools to predict long-term popu- fauna, we believe this approach may be useful for lation responses to roads and to compare outcomes of comparing the risk of road-related fragmentation and multiple mitigation scenarios (e.g., Gibbs and Shriver mortality for species elsewhere and for other taxo- 2005; Borda-de-Agua et al. 2014; Polak et al. 2014; nomic groups. The results can help to inform multi- Crawford 2015). Need and placement of mitigation criteria threat assessments for special status species or structures can be guided by local population or meta- those in consideration for listing. Finally, this serves to population dynamics, landscape attributes, movement highlight species that may deserve further study and routes, and road mortality hot spots (e.g., Bissonette consideration for aquatic and terrestrial road mitiga- and Adair 2008; Langen et al. 2009, 2015b; D’Amico tion to reduce mortality and to maintain population- et al. 2016; Loraamm and Downs 2016). level connectivity. 123 932 Landscape Ecol (2018) 33:911–935 sizes in the San Joaquin Desert of California. W Wild This risk assessment approach compares the sus- 2:23–28 ceptibility of species to negative road impacts. Com- Bissonette JA, Adair W (2008) Restoring habitat permeability to monly, there are numerous populations within a roaded landscapes with isometrically-scaled wildlife species range that occupy areas with greatly differing crossings. Biol Conserv 141(2):482–488 Boarman WI, Sazaki M (1996) Highway mortality in desert road pressures. Therefore, the actual risk to specific tortoises and small vertebrates: success of barrier fences species populations will depend upon local road and culverts (No. FHWA-PD-96-041) densities, road-types, traffic, and road locations in Boarman WI, Sazaki M (2006) A highway’s road-effect zone for relation to species habitat and movement corridors. desert tortoises (Gopherus agassizii). J Arid Environ 65(1):94–101 Bolster BC (2010) A status review of the California tiger sala- Acknowledgements We greatly appreciate the support and mander (Ambystoma californiense). Report to the Fish and feedback from Harold Hunt, Amy Golden, and James Henke Game Commission, State of California. Nongame Wildlife from the California Department of Transportation. Tony Program Report 2010–4 Clevenger, Tom Langton, Jeff Tracey, Amber Wright, Laura Bonnet X, Naulleau G, Shine R (1999) The dangers of leaving Patterson, Brian Halstead, Kari Gunson, Jon Richmond, and two home: dispersal and mortality in snakes. Biol Conserv anonymous reviewers gave thoughtful feedback that improved 89(1):39–50 this manuscript. We thank Tristan Edgarian for reviewing and Borda-de-Agua L, Grilo C, Pereira HM (2014) Modeling the cross-checking our life history data and movement data and impact of road mortality on barn owl (Tyto alba) popula- references. Finally, we appreciate all of the scientists that tions using age-structured models. Ecol Model 276:29–37 contributed life history information for California reptile and Bourque R (2008) [North Coast] spatial ecology of an inland amphibian species (Appendices). Funding came from Caltrans population of the foothill yellow-legged frog (Rana boylii) Department of Transportation Contract #65A0553 and the in Tehama County, California. Masters Thesis, California Ecosystems Mission Area in USGS. This is contribution number State University, Humboldt, pp 107 615 of the U.S. Geological Survey Amphibian Research and Brehme CS, Tracey JA, McClenaghan LR, Fisher RN (2013) Monitoring Initiative (ARMI). The use of trade, product or firm Permeability of roads to movement of scrubland lizards names in this publication does not imply endorsement by the and small mammals. Conserv Biol 27(4):710–720 U.S. Government. Bulger JB, Scott NJ, Seymour RB (2003) Terrestrial activity and conservation of adult California red-legged frogs Rana Open Access This article is distributed under the terms of the aurora draytonii in coastal forests and grasslands. Biol Creative Commons Attribution 4.0 International License (http:// Conserv 110(1):85–95 creativecommons.org/licenses/by/4.0/), which permits unre- California Department of Fish and Wildlife (CDFW) (2016a) stricted use, distribution, and reproduction in any medium, Threatened and endangered species. http://www.dfg.ca. provided you give appropriate credit to the original gov/wildlife/nongame/t_e_spp/ (accessed August 2016) author(s) and the source, provide a link to the Creative Com- California Department of Fish and Wildlife (CDFW) (2016b) A mons license, and indicate if changes were made. status review of the flat-tailed horned lizard (Phrynosoma mcallii) in California. Report to the Fish and Game Commission Carr LW, Fahrig L (2001) Effect of road traffic on two References amphibian species of differing vagility. Conserv Biol 15(4):1071–1078 Case TJ, Fisher RN (2001) Measuring and predicting species Andrews KM, Gibbons JW (2005) How do highways influence presence: coastal sage scrub case study. In: Hunsaker CT, snake movement? Behavioral responses to roads and Goodchild MF, Friedl MA, Case TJ (eds) Spatial uncer- vehicles. Copeia 4:772–782 tainty in ecology. Springer-Verlag, New York, pp 47–71 Andrews KM, Gibbons JW, Jochimsen D, Mitchell J (2008) Clark RW, Brown WS, Stechert R, Zamudio KR (2010) Roads, Ecological effects of roads on amphibians and reptiles: a interrupted dispersal, and genetic diversity in timber rat- literature review. In: Mitchell JC, Jung Brown RE, Bar- tlesnakes. Conserv Biol 24(4):1059–1069 tholomew B (eds) Urban Herpetology, Herpetological Clarkin K, Conner A, Furniss MJ, Gibernick B, Love M, Conservation 3. Society for the Study of Reptiles and Moynan K, Wilson S (2005) National inventory and Amphibians, Salt-Lake City, UT, pp 121–143 assessment procedure for identifying barriers to aquatic Andrews KM, Nanjappa P, Riley SP (eds) (2015a) Roads and organism passage at road-stream crossings. US Dep Agric ecological infrastructure: concepts and applications for for Serv. http://www.stream.fs.fed.us/publications/PDFs/ small animals. JHU Press, Baltimore, MD NIAP.pdf Andrews KM, Langen TA, Struijk RPJH (2015b) Reptiles: Colino-Rabanal VJ, Lizana M (2012) Herpetofauna and roads: a Overlooked but often at risk from roads. Handbook of Road review. Basic Appl Herpetol 26:5–31 Ecology. Wiley, New York, pp 271–280 Colley M, Lougheed SC, Otterbein K, Litzgus JD (2017) Miti- Bailey CV, Germano DJ (2015) Probability of occupancy of gation reduces road mortality of a threatened rattlesnake. blunt-nosed leopard lizards on habitat patches of various Wildlife Res. https://doi.org/10.1071/WR16130 123 Landscape Ecol (2018) 33:911–935 933 Congress US (1973) Endangered species act. US Code Glista DJ, DeVault TL, DeWoody JA (2008) Vertebrate road 16:1534–1544 mortality predominantly impacts amphibians. Herpetol COSEWIC (Committee on the Status of Endangered Wildlife in Conserv Biol 3(1):77–87 Canada) (2012) COSEWIC assessment and status report on Goodman RH, Stewart GR (2000) Aquatic home ranges of the Western Toad Anaxyrus boreas in Canada. Committee female Western Pond Turtles, Clemmys marmorata, at two on the Status of Endangered Wildlife in Canada. Ottawa. sites in Southern California. Chelonian Conserv Biol xiv ? 71 pp. (www.registrelep-sararegistry.gc.ca/default_ 3(4):743–744 e.cfm) Gregory PT (1984) Communal denning in snakes. In: Seigel RA, COSEWIC (Committee on the Status of Endangered Wildlife in Hunt LE, Knight JL, Maralet LA, Zuschlag NL (eds) Canada) (2015) COSEWIC assessment and status report on Contributions to vertebrate ecology and systematics: a the Northern Red-legged Frog Rana aurora in Canada. tribute to Henry S. Fitch. University of Kansas Museum of Committee on the Status of Endangered Wildlife in Natural History Special Publication 10:57–75 Canada. Ottawa. xii ? 69 pp. (www.registrelep- Halstead BJ, Skalos SM, Wylie GD, Casazza ML (2015) Ter- sararegistry.gc.ca/default_e.cfm) restrial ecology of semi-aquatic giant gartersnakes Crawford BA (2015) Lessons from terrapin mortality and (Thamnophis gigas). Herpetol Conserv Bio 10(2):633–644 management on the Jekyll Island Causeway, Georgia, Hayes MP, Rombough CJ, Hayes CB (2007) Natural History: USA. In: Andrews KM, Nanjappa P, Riley SPD (eds) Rana aurora (Northern red-legged frog) movement. Her- Roads and ecological infrastructure. Concepts and appli- petol Rev 38:192–193 cations for small animals. Johns Hopkins University Press, Hensley FR, Jones TR, Maxwell MS, Adams LJ, Nedella NS Baltimore, pp 21–41 (2010) Demography, terrestrial behavior, and growth of D’Amico M, Roman J, De los Reyes L, Revilla E (2015) Ver- Sonora mud turtles (Kinosternon sonoriense) in an extreme tebrate road-kill patterns in Mediterranean habitats: who, habitat. Herpetol Monogr 24:174–193 when and where. Biol Conserv 191:234–242 Hermann HW, Pozarowski KM, Ochoa A, Schuett GW (2017) ´ ´ D’Amico M, Periquet S, Roman J, Revilla E (2016) Road An interstate highway affects gene flow in a top reptilian avoidance responses determine the impact of heteroge- predator (Crotalus atrox) of the Sonoran Desert. Conserv neous road networks at a regional scale. J Appl Ecol Genet 18:911–924 53(1):181–190 Hickman JC (ed) (1993) The Jepson manual of higher plants of Dodd CK, Barichivich WJ, Smith LL (2004) Effectiveness of a California. University of California Press, Berkeley barrier wall and culverts in reducing wildlife mortality on a Hobday AJ, Smith ADM, Stobutzki IC, Bulman C, Daley R, heavily traveled highway in Florida. Biol Conserv Dambacher JM, Deng RA, Dowdney J, Fuller M, Furlani 118(5):619–631 D, Griffiths SP (2011) Ecological risk assessment for the Eigenbrod F, Hecnar SJ, Fahrig L (2008) The relative effects of effects of fishing. Fish Res 108(2):372–384 road traffic and forest cover on anuran populations. Biol Houlahan JE, Findlay CS (2003) The effects of adjacent land use Conserv 141(1):35–46 on wetland amphibian species richness and community Evelyn CJ, Sweet SS (2012) Conservation status of amphibians composition. Can J Fish Aquat Sci 60:1078–1094 and reptiles on USDA National Forests, Pacific Southwest Hozo SP, Djulbegovic B, Hozo I (2005) Estimating the mean Region, 2012. Prepared for: USDA National Forest Ser- and variance from the median, range, and the size of a vice, Region 5, Pacific Southwest Region sample. BMC Med Res Methodol 5(1):1 Fahrig L, Rytwinski T (2009) Effects of roads on animal IUCN (2017) The IUCN Red List of Threatened Species. Ver- abundance: an empirical review and synthesis. Ecol Soc sion 2017-1. http://www.iucnredlist.org 14. http://www.ecologyandsociety.org/vol14/iss11/art21 Jackson ND, Fahrig L (2011) Relative effects of road mortality Fahrig L, Pedlar JH, Pope SE, Taylor PD, Wegner JF (1995) and decreased connectivity on population genetic diversity. Effect of road traffic on amphibian density. Biol Conserv Biol Conserv 144(12):3143–3148 73(3):177–182 Jackson SD, Langen TA, Marsh DM, Andrews KM (2015) Fellers GM, Kleeman PM (2007) California red-legged frog Natural history and physiological characteristics of small (Rana draytonii) movement and habitat use: implications animals in relation to roads. In: Andrews KM, Nanjappa P, for conservation. J Herpetol 41(2):276–286 Riley SPD (eds) Roads and ecological infrastructure. Forman RT, Sperling D, Bissonette JA, Clevenger AP, Cutshall Concepts and applications for small animals. Johns Hop- CD, Dale VH, Fahrig L, France R, Goldman CR, Heanue kins University Press, Baltimore, pp 21–41 K, Jones JA (2003) Road ecology. Science and Solutions. Jacobson SL, Bliss-Ketchum LL, Rivera CE, Smith WP (2016) Island Press, Washington A behavior-based framework for assessing barrier effects Furniss MJ, Roelofs TD, Yee CS (1991) Road construction and to wildlife from vehicle traffic volume. Ecosphere maintenance. Am Fish Soc Special Publ 19:297–323 7(4):e01345 Gibbs JP (1998) Amphibian movements in response to forest Jaeger JA (2000) Landscape division, splitting index, and edges, roads, and streambeds in southern New England. effective mesh size: new measures of landscape fragmen- J Wildl Manage 62:584–589 tation. Land Ecol 15(2):115–130 Gibbs JP, Shriver WG (2002) Estimating the effects of road mor- Jaeger JA, Bowman J, Brennan J, Fahrig L, Bert D, Bouchard J, tality on turtle populations. Conserv Biol 16(6):1647–1652 Charbonneau N, Frank K, Gruber B, von Toschanowitz KT Gibbs JP, Shriver WG (2005) Can road mortality limit popula- (2005) Predicting when animal populations are at risk from tions of pool-breeding amphibians? Wetl Ecol Manag roads: an interactive model of road avoidance behavior. 13:281–289 Ecol Model 185(2):329–348 123 934 Landscape Ecol (2018) 33:911–935 Januchowski-Hartley SR, McIntyre PB, Diebel M, Doran PJ, Mace GM, Collar NJ, Gaston KJ, Hilton-Taylor C, Akc ¸akaya Infante DM, Joseph C, Allan JD (2013) Restoring aquatic HR, Leader-Williams N, Milner-Gulland EJ, Stuart SN ecosystem connectivity requires expanding inventories of (2008) Quantification of extinction risk: IUCN’s system for both dams and road crossings. Front Ecol Environ classifying threatened species. Conserv Biol 11(4):211–217 22(6):1424–1442 Jochimsen DM, Peterson CR, Andrews KM, Gibbons JW, Marsh DM, Jaeger JAG (2015) Direct effects of roads on small Drawer E (2004) A literature review of the effects of roads animal populations. Roads and ecological infrastructure. on amphibians and reptiles and the measures used to Concepts and applications for small animals. Johns Hop- minimize those effects. USDA Forest Service, Idaho Fish kins University Press, Baltimore, pp 21–41 and Game Department, Washington, DC Martinez-Solano I, Jockusch EL, Wake DB (2007) Extreme Jochimsen DM, Peterson CR, Harmon LJ (2014) Influence of population subdivision throughout a continuous range: ecology and landscape on snake road mortality in a sage- phylogeography of Batrachoseps attenuatus (Caudata: brush-steppe ecosystem. Anim Conserv 17(6):583–592 Plethodontidae) in western North America. Mol Ecol Jones TR, Babb RD, Hensley FR, LiWanPo C, Sullivan BK 16(20):4335–4355 (2011) Sonoran Desert snake communities at two sites: Mazerolle MJ, Huot M, Gravel M (2005) Behavior of amphib- concordance and effects of increased road traffic. Herpetol ians on the road in response to car traffic. Herpetologica Conserv Biol 6(1):61–71 61(4):380–388 Kemp PS, O’hanley JR (2010) Procedures for evaluating and Mccardle LD, Fontenot CL (2016) The influence of thermal prioritising the removal of fish passage barriers: a synthe- biology on road mortality risk in snakes. J Thermal Biol sis. Fish Manag Ecol 17(4):297–322 56:39–49 Klauber LM (1931) A statistical survey of the snakes of the Mitrovich MJ (2006) A case study in conservation science: The southern border of California. Bull Zool Soc San Diego spatial ecology and evolutionary history of the coachwhip 8:1–93 snake (Masticophis flagellum) and striped racer (M. later- Kwiatkowski MA, Schuett GW, Repp RA, Nowak EM, Sullivan alis). Ph.D. dissertation. University of California, Davis, BK (2008) Does urbanization influence the spatial ecology of and San Diego State University. San Diego Gila monsters in the Sonoran Desert? J Zool 276(4):350–357 Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GA, Langen TA, Ogden KM, Schwarting LL (2009) Predicting hot Kent J (2000) Biodiversity hotspots for conservation pri- spots of herpetofauna road mortality along highway net- orities. Nature 403(6772):853–858 works. J Wildl Manag 73(1):104–114 NatureServe (2017) NatureServe explorer: An online encyclo- Langen TA, Andrews KM, Brady SP, Karraker NE, Smith DJ pedia of life [web application]. Version 7.1. NatureServe, (2015a) Road effects on habitat quality for small animals. Arlington, Virginia. http://explorer.natureserve.org. (Ac- In: Andrews KM, Nanjappa P, Riley SPD (eds) Roads and cessed: May 25, 2017) ecological infrastructure. Concepts and applications for Orłowski G (2007) Spatial distribution and seasonal pattern in small animals. Johns Hopkins University Press, Baltimore, road mortality of the common toad Bufo bufo in an agri- pp 21–41 cultural landscape of south-western Poland. Amphibia- Langen TA, Gunson KE, Jackson SD, Smith DJ, Ruediger W Reptilia 28(1):25–31 (2015b) Planning and designing mitigation of road effects Peaden JM, Tuberville TD, Buhlmann KA, Nafus MG, Todd BD on small animals. In: Andrews KM, Nanjappa P, Riley SPD (2016) Delimiting road-effect zones for threatened species: (eds) Roads and ecological infrastructure. Concepts and implications for mitigation fencing. Wildl Res applications for small animals. Johns Hopkins University 42(8):650–659 Press, Baltimore, pp 146–177 Peaden JM, Nowakowski AJ, Tuberville TD, Buhlmann KA, Langton TES (2015) A history of small animal road ecology. Todd BD (2017) Effects of roads and roadside fencing on Roads and ecological infrastructure. Concepts and appli- movements, space use, and carapace temperatures of a cations for small animals. Johns Hopkins University Press, threatened tortoise. Biol Conserv 214:13–22 Baltimore, pp 21–41 Pianka ER (1966) Convexity, desert lizards, and spatial Lannoo MJ (2005) Amphibian declines: the conservation status heterogeneity. Ecology 47(6):1055–1059 of United States species. University of California Press, Pilliod DS, Welty JL, Stafford R (2013) Terrestrial movement Berkeley patterns of western pond turtles (Actinemys marmorata)in Levine KK (2013) California amphibian and reptile crossing central California. Herpetol Conserv Bio 8(1):207–221 preliminary investigation. Report to Caltrans Division of Polak T, Rhodes JR, Jones D, Possingham HP (2014) Optimal Environmental Planning planning for mitigating the impacts of roads on wildlife. Linkov I, Satterstrom FK, Kiker G, Batchelor C, Bridges T, J Appl Ecol 51(3):726–734 Ferguson E (2006) From comparative risk assessment to Rathbun GB, Scott NJ Jr, Murphey TG (2002) Terrestrial habitat multi-criteria decision analysis and adaptive management: use by Pacific pond turtles in a Mediterranean climate. recent developments and applications. Environ Int Southwest Nat 47(2):225–235 32(8):1072–1093 Reese DA, Welsh HH (1997) Use of terrestrial habitat by Litvaitis JA, Tash JP (2008) An approach toward understanding western pond turtles, Clemmys marmorata: implications wildlife-vehicle collisions. Environ Manag 42(4):688–697 for management. In Proceedings: Conservation, restora- Loraamm RW, Downs JA (2016) A wildlife movement tion, and management of tortoises and turtles–an interna- approach to optimally locate wildlife crossing structures. tional conference. State University of New York, Purchase, Int J Geogr Inf Sci 30(1):74–88 New York, pp 352–357) 123 Landscape Ecol (2018) 33:911–935 935 Reid BN, Peery MZ (2014) Land use patterns skew sex ratios, Steen DA, Aresco MJ, Beilke SG, Compton BW, Condon EP, decrease genetic diversity and trump the effects of recent Dodd KC, Forrester H, Gibbons JW, Greene JL, Johnson G, climate change in an endangered turtle. Divers Distrib Langen TA (2006) Relative vulnerability of female turtles 20(12):1425–1437 to road mortality. Anim Conserv 9(3):269–273 Roe JH, Gibson J, Kingsbury BA (2006) Beyond the wetland Sullivan BK (2012) Road riding. In: McDiarmid RW (ed) border: estimating the impact of roads for two species of Reptile biodiversity: standard methods for inventory and water snakes. Biol Conserv 130(2):161–168 monitoring. University of California Press, Oakland, CA, Rosen PC, Lowe CH (1994) Highway mortality of snakes in the pp 215–218 Sonoran Desert of southern Arizona. Biol Conserv Suter GW II (2016) Ecological risk assessment. CRC Press, 68(2):143–148 Boca Raton, FL Rosenberg D, Gervais J, Vesely D, Barnes S, Holts L, Horn R, Thomson RC, Wright AN, Shaffer HB (2016) California Swift R, Todd L, Yee C (2009) Conservation assessment of amphibian and reptile species of special concern. Univer- the Western pond turtle in Oregon (Actinemys marmorata). sity of California Press, Oakland, CA USDI Bureau of Land Management and Fish and Wildlife Trenham PC (1998) Demography, migration, and metapopula- Service, USDA Forest Service Region 6, Oregon tion structure of pond-breeding salamanders. Ph.D. dis- Rouse JD, Willson RJ, Black R, Brooks RJ (2011) Movement sertation. University of California, Davis and spatial dispersion of Sistrurus catenatus and Hetero- Trenham PC, Koenig WD, Shaffer HB (2001) Spatially auto- don platirhinos: implications for interactions with roads. correlated demography and interpond dispersal in the Copeia 3:443–456 salamander Ambystoma californiense. Ecology Row JR, Blouin-Demers G, Weatherhead PJ (2007) Demo- 82(12):3519–3530 graphic effects of road mortality in black ratsnakes (Elaphe Trenham PC, Koenig WD, Mossman MJ, Stark SL, Jagger LA obsoleta). Biol Conserv 137(1):117–124 (2003) Regional dynamics of wetland-breeding frogs and Rudolph DC, Burgdorf SJ, Conner RN, Schaefer RR (1999) toads: turnover and synchrony. Ecol Appl 13(6):1522– Preliminary evaluation of the impact of roads and associ- 1532 ated vehicular traffic on snake populations in eastern US Fish and Wildlife (USFWS) (2016) United States species: Texas. In: Proceedings of the Third International Confer- Endangered and threatened wildlife (50 CFR 17.11). ence on Wildlife Ecology and Transportation. FL-ER-73- https://www.fws.gov/endangered/species/us-species.html 99. Evink GL, Garrett P, Zeigler D (eds) Florida Depart- (accessed August 2016) ment of Transportation, Tallahassee, Florida, pp 129–136 US Fish and Wildlife Service (USFWS) (2002) Recovery plan Rytwinski T, Fahrig L (2012) Do species life history traits for the California red-legged frog (Rana aurora draytonii). explain population responses to roads? A meta-analysis. US Fish and Wildlife Service, Portland, Oregon, viii ? Biol Conserv 147(1):87–98 173 pp Rytwinski T, Fahrig L (2013) Why are some animal populations US Fish and Wildlife Service (USFWS) (2009) Rana pretiosa. unaffected or positively affected by roads? Oecologia Species assessment and listing priority assignment form. 173(3):1143–1156 USFWS Region 1, Portland Rytwinski T, Fahrig L (2015) The impacts of roads and traffic on US Fish and Wildlife Service (USFWS) (2011) Alameda terrestrial animal populations. Handbook of Road Ecology. whipsnake (Masticophis lateralis euryxanthus) 5-year Wiley, New York, pp 237–246 review summary and evaluation. US Fish and Wildlife Semlitsch RD (1998) Biological delineation of terrestrial buffer Service, Sacramento, CA zones for pond-breeding salamanders. Conserv Biol van der Ree R, Smith DJ, Grilo C (2015) Handbook of road 12(5):1113–1119 ecology. Wiley, New York Semlitsch RD (2008) Differentiating migration and dispersal Vangel MG (2015) Tolerance interval. Statistics Reference processes for pond-breeding amphibians. J Wildl Manag Online. Wiley StatsRef, New York, pp 1–7 72(1):260–267 Vences M, Wake DB (2007) Speciation, species boundaries and Semlitsch RD, Bodie JR (2003) Biological criteria for buffer phylogeography of amphibians. Amphib Biol 7:2613–2671 zones around wetlands and riparian habitats for amphibians Watson JW, McAllister KR, Pierce DJ (2003) Home ranges, and reptiles. Conserv Biol 17(5):1219–1228 movements, and habitat selection of Oregon Spotted Frogs Shilling F, Waetjen D (2017) California Roadkill Observation (Rana pretiosa). J Herp 37(2):292–300 System. UC Davis Road Ecology Center. http:// Young KV, Young AT (2000) Final report: scientific study of wildlifecrossing.net/california(accessed April 2017) the flat-tailed horned lizard, Phrynosoma mcallii. US Dept. Siers SR, Reed RN, Savidge JA (2016) To cross or not to cross: of Navy Contracts N68711-95-LT-C0032, N68711-95-LT- modeling wildlife road crossings as a binary response C0035. 72 pp variable with contextual predictors. Ecosphere 7(5):19 Zimmermann Teixeira F, Kindel A, Hartz S, Mitchell S, Fahrig Stebbins RC, McGinnis SM (2012) Field guide to amphibians L (2017) When road-kill hotspots do not indicate the best and reptiles of California, vol 103. University of California sites for road-kill mitigation. J Appl Ecol 54(5):1544–1551 Press, Berkeley

Journal

Landscape EcologySpringer Journals

Published: May 9, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off