2008 SOUTHEASTERN NATURALIST 7(2):339–348 A Road Survey of Amphibians and Reptiles in a Bottomland Hardwood Forest Jessica L. Coleman1.*, Neil B. Ford1, and Kevin Herriman2 Abstract - The impact of roads on amphibian and reptile numbers at Texas Parks and Wildlife Department’s Old Sabine Bottom Wildlife Management Area (OSBWMA), a bottomland hardwood forest was studied from 1 June 2003 to 1 June 2004 by vehicular and pedestrian surveys. One weekly survey was conducted in the morning and the other in late afternoon. Vehicular surveys were conducted by driving 16.1 km/hr over the roads (9.5 km) through the bottomland twice per week. Pedestrian surveys were conducted on 6 sections of those roads (1.3, 1.3, 1.3, 1.1, 0.3, and 0.3 km), three on the larger main gravel road, and three sections on smaller dirt tributaries, twice weekly. One section of the main road (1.3 km) was located near three operating oil rigs. The other 5 sections of the survey route traversed forested regions of the management area. The 3 sections located along the main gravel road were subject to regular vehicular traffic by the public and OSBWMA staff. The three sections located along the dirt trail tributaries had primarily foot travel with much less and slower vehicular traffic. GPS coordinate, date, and condition (alive or dead) of the specimens were recorded. In the summer, 10.5% of the amphibians and reptiles were found dead on the road (DOR). The number of DORs in the fall was 47.1%, and 64.3% in the spring. For amphibians, 28.1% were DOR (summer), 20% in the fall, and 83.3% in the spring. In the winter months, no specimens were observed on the roads. Road traffic in this bottomland impacts reptiles extensively in the fall and amphibians in the summer, whereas both reptiles and amphibians are affected by road traffic in the spring. Smaller roads and the section of road with oil rigs had lower mortality rates, while areas with adjacent borrow pits had higher mortality rates. Introduction The impact of roads on animal populations is becoming an increasing concern throughout the developed world. Yet relatively little is actually known about the effects of roads on specific species (e.g., Oxley et al. 1974). Roads have multiple influences on animal populations by fragmenting habitats (Loughry and McDonough 1996, Scott 1938, Vos and Chardon 1998), by direct road mortalities (Groot Bruinderink and Hazebroek 1996, van Gelder 1973), and by creating barriers for some species (Noss 2004). Road surveys are often used to estimate road mortality (Haxton 2000, Lodé 2000, Scott 1938), species diversity (Enge and Wood 2002, Seigel et al. 2002), and abundances (Enge and Wood 2003, Fahrig et al. 1995). For reptiles and amphibians, most studies of the effects of roads have examined the number of animals dead on the road (DOR). Usually, these surveys 1Biology Department, University of Texas at Tyler, Tyler, TX 75799. 2Texas Parks and Wildlife Department, Old Sabine Bottom Wildlife Management Area, Lindale, TX 75771. *Corresponding author - j_l_coleman@hotmail.com. 340 Southeastern Naturalist Vol.7, No. 2 are conducted from vehicles (Carr and Fahrig 2001, Rosen and Lowe 1994), which may be biased against recording the young of smaller species because they are harder to view from a moving car. A recent pedestrian survey showed significant road mortalities among small species and neonatal snakes (Enge and Wood 2002). Although pedestrian surveys of amphibians and reptiles may find more small and young animals, they have the disadvantage of requiring more time. Combining these methods may produce a more accurate measure of the number of road mortalities and a more comprehensive species representation. Bottomland hardwood forests are some of the most productive habitats globally and are also among the most threatened (Molles et al. 1998). The periodic inundation of these fl oodplains deposits a tremendous nutrient load in pools and lowlands when the water recedes. Amphibians breed in these pools, but as the water recedes in the summer, these sites also become important prey sources for semi-aquatic snakes. The same nutrient input to terrestrial habitats translates to a large invertebrate detritivore fauna that serves as food for adult amphibians, lizards, and other terrestrial reptiles. Therefore, both diversity and abundance of amphibians and reptiles are high in such environments (Findlay and Houlahan 1997). Floodplain forests have declined in extent in the southeastern United States due to commercial logging and impoundments altering natural fl ood regimes (Clark and Benforado 1981). Most protected fl oodplains are designated and utilized primarily for hunting, and although vehicular traffic may be minimal, the impact of roads in these areas on amphibians and reptiles is important to document (Means 1999). In addition, roads in these habitats are typically unpaved, whereas most studies of the effect of roads and traffic on amphibians and reptiles have been with paved roads (Carr and Fahrig 2001, Rosen and Lowe 1994). The objective of this study was to examine the impact of roads on local amphibian and reptile populations in a bottomland floodplain. We explored differences in road mortality relative to seasons, road characteristics, presence of nearby borrow pits, and other impacts, such as oil rigs. Both vehicular and pedestrian surveys were employed at sites that differ in road width, presence of gravel, nearby water sources, and other environmental characteristics. Methods Study site Observations of amphibians and reptiles were made on the roads at Texas Parks and Wildlife Department’s Old Sabine Bottom Wildlife Management Area (OSBWMA). The OSBWMA is a 2318-ha bottomland hardwood forest along a section of the Sabine River in northeastern Smith County, TX. It consists of Quercus spp. (oak), Liquidambar styracifl ua L. (sweetgum), Ulmus spp. (elm), Fraxinus spp. (ash), and a diverse understory vegetation. The plant diversity is in part a response to the infl ux of nutrients that fl ood the bottomland and remain in borrow pits scattered throughout. Roads in 2008 J.L. Coleman, N.B. Ford, and K. Herriman 341 the OSBWMA are dirt, gravel, and some degree of vegetation, which varies along different sections. Data collection Vehicular and pedestrian surveys were conducted on 9.5 km of road twice a week from 1 June 2003 through 1 June 2004. The two weekly surveys, one in the morning (03:00–12:00) and the other in the evening (12:01–22:00), were conducted for a year at OSBWMA. Species observed on the road were identified and recorded as dead (DOR) or alive (AOR). For each observation we recorded date, GPS location, and species or the most specific level of identification possible for extensively damaged specimens. Vehicular surveys took place over all roads at 16.1 km/hr twice during each weekly survey. The pedestrian survey was conducted by walking a total of six sections with a combined distance of 5.3 km. Three sections on the larger main road of 1.3, 1.3, and 1.0 km and three smaller side roads, 1.1, 0.3, and 0.3 km, were walked. The main roads had widths of 491, 340, and 510 cm. These roads consisted of dirt and gravel, which were periodically graded to reduce vegetation. Along the main section of road, several borrow pits were located close to the road and across from each another. The side roads had borrow pits which were positioned further off the road and not directly across from one another (Fig. 1). Borrow Figure 1. Map of the roads and environmental characteristics throughout the study site (OSBWMA) in Northeast Texas between Smith and Wood counties. The brackets ([ ]) represent the areas of the road where pedestrian surveys were conducted, and A–F indicate the percent of reptiles and amphibians observed dead (DOR) in each section of road during June 2003 to June 2004. 342 Southeastern Naturalist Vol.7, No. 2 pits and oil rigs were adjacent to one section of the main road. The side roads had widths of 343, 345, and 336 cm. These smaller roads consisted of dirt and gravel with vegetation growing in the middle portion of the roads. Data analysis The abundance of AOR and DOR amphibians and reptiles were compared to investigate relative numbers of species affected during each season and the entire year by using a t-test. The seasons were categorized by the Table 1 Abundance of amphibians and reptiles observed seasonally on the roads at the OSBWMA from June 2003 to June 2004. The ratio represents number of AOR (alive on road)/DOR (dead on road). No species were recorded during the winter months (December, January, and February). ** = highly significant (P < 0.001). Species Summer Fall Spring Anurans Acris crepitans Baird Northern Cricket Frog 1/0 0/0 1/0 Bufo velatus Braggs and East Texas Toad 1/0 8/1 0/2 Sanders Hyla cinerea (Schneider) Green Treefrog 1/0 0/0 0/0 Rana clamitans Latreille Bronze Frog 0/0 2/2 0/0 Rana sphenocephala L. Southern Leopard Frog 18/7 4/1 0/2 Unidentifiable anuran 2/2 2/0 0/1 Anuran total 23/9 16/4 1/5 Snakes Agkistrodon contortix (L.) Copperhead 7/0 0/0 2/1 Agkistrodon piscivorus Cottonmouth 4/0 0/0 1/0 (Lacépède) Crotalus horridus L. Timber Rattlesnake 1/0 0/0 1/2 Lampropeltis triangulum Louisiana Milk Snake 1/0 0/0 0/0 (Lacépède) Nerodia erythrogaster Yellow-Bellied Water Snake 5/0 0/0 0/0 (Forster in Bossu) Nerodia fasciata (L.) Broadband Water Snake 4/0 0/0 0/0 Opheodrys aestivus L. Rough Green Snake 1/1 0/0 0/0 Panterophis obsoleta Texas Rat Snake 4/0 0/0 0/0 (Baird and Girard) Storeia dekayi (Holbrook) Brown Snake 0/0 2/2 0/1 Thamnophis proximus Western Ribbon Snake 2/1 0/1 0/0 (Say in James) Unidentifiable snake 0/1 0/0 0/1 Snake total 29/3 2/3 4/5 Turtles Chelydra serpentina (L.) Common Snapping Turtle 0/1 0/0 0/0 Sternotherus carinatus (Gray) Razorback Musk Turtle 1/0 0/0 0/0 Terrapene carolina (L.) Three-Toed Box Turtle 1/0 0/0 0/0 Turtle total 2/1 0/0 0/0 Lizards Eumeces laticeps (Schneider) Broadhead Skink 3/0 1/6 3/0 Scincella lateralis Ground Skink 1/0 1/0 0/0 (Say in James) Eumeces sp. 0/0 0/0 1/0 Lizard total 4/0 2/6 4/0 Overall total 58/13** 20/13 9/10 2008 J.L. Coleman, N.B. Ford, and K. Herriman 343 following months: winter (November–January), spring (February–April), summer (May–July), and fall (August–October). A t-test was also used to compare the difference in the number of AOR and DOR found during morning (3:00–12:00) and evening (12:01–22:00) surveys. Results A total of 128 individuals representing 5 species of amphibians and 15 species of reptiles were recorded on the roads of the OSBWMA from June 2003 to June 2004 (Table 1). Anurans were the only amphibians seen during this study, although two species of salamanders (Ambystomidae: Ambystoma) were recorded on the roads subsequently. Three species of skinks were the only lizards observed, and 3 of roughly 10 possible turtle species were recorded. Ten of the approximately 17 possible snake species were recorded. Of the total amphibians and reptiles observed, 31.3% were found DOR (anurans 28.6%, lizards 41.2%, turtles 33.3%, and snakes 30.8%). DORs were concentrated in the area of the main road with borrow pits located across from each other (Fig. 1). The mortality rates for the size of the roads were 0. 031 DOR/km for the smaller side roads, and 0.075 DOR/km for the larger main roads. Figure 2. Comparison of reptiles and amphibians observed at the OSBWMA alive (AOR) and dead (DOR) (A) from June 2003 to June 2004, and (B) am and pm surveys. ** = highly significant (P < 0.001). 344 Southeastern Naturalist Vol.7, No. 2 Overall, there was a significant difference in the number of DORs and AORs during the entire study (P < 0.001; Fig. 2A). The survey period comparison indicated that there were significantly more AOR than DOR in the evening surveys (P < 0.001; Fig. 2B) than for the morning. Seasonal comparisons indicated that both amphibians and reptiles were more abundant on roads in the summer, but there was only a significant difference for reptiles (P < 0.001; Fig. 3A). Fewer reptiles were recorded in fall and spring, and no records for species in either group were recorded in the winter. Snakes and anurans were by far the most abundant animals recorded and so were examined separately. The number of snake mortalities mirrored the overall reptile trend, with more species observed on the roads in the summer. In the spring and fall, lower abundance and species diversity of snakes occurred, and a greater proportion were found dead. For anurans, some species were more abundant in particular seasons. Rana sphenocephala (Southern Leopard Frog) dominated records in the summer, whereas Bufo velatus (East Texas Toad) was the most abundant in the fall. R. sphenocephala and Rana clamitans (Bronze Frog) were also prevalent in that season. The few anurans recorded in the spring were found dead. Figure 3. S e a s o n a l com-parison of (A) reptiles and (B) amphibians observed at the OSBWMA alive (AOR) and dead (DOR) from June 2003 to June 2004. No reptiles or amphibians were seen d u r i n g the winter months. ** = highly significant (P < 0.001). 2008 J.L. Coleman, N.B. Ford, and K. Herriman 345 Discussion This project used both vehicular and pedestrian surveys to examine the impact of roads on amphibians and reptiles in a bottomland. Previous studies with vertebrate road mortalities have suggested that the smaller species were difficult to detect from a moving vehicle (Taylor and Goldingay 2004). We found that larger mobile species were hard to identify during pedestrian surveys, since they were less likely to be encountered, or were only seen by a quick glimpse. By combining the two types of surveys, we believe that an accurate assessment of road mortality of all the amphibians and reptiles of the OSBWMA was made. Vehicular surveys provided information on the larger, quicker species, while the slower, smaller species were observed during the pedestrian surveys. Almost one third of the 128 records of all amphibians and reptiles on the roads were dead animals (Table 1). The actual number of DOR amphibians and reptiles is likely an underestimate, as scavengers probably consumed some of the road kills prior to detection. However, within our study, we assume this factor to be similar on all areas. Although salamanders and turtles are quite abundant at the OSBWMA, most of our observations were snakes and anurans. Terrestrial movements of salamanders tend to be restricted to rainy nights (Mazerolle 2004), whereas our surveys were conducted during daylight hours and clear nights, with little traffic occurring during nocturnal hours to cause mortality. Trachemys spp. (sliders), Pseudemys spp. (cooters), Apalone spp. (softshell turtles), and Graptemys spp. (map turtles) that occur at the OSBWMA primarily stay in the larger bodies of water, which are not near roads, and this may explain why only a few were seen. Similar studies have found that the size and other characteristics of the roads contribute to the number of reptile and amphibian mortalities (Clevenger et al. 2003). In our project, there was no apparent impact of size of the roads or presence of oil rigs on the number of records, but the mortality rate was lower on the secondary roads (0.031 DOR/km vs. 0.075 DOR/km for the main road). The smaller secondary roads were primarily subject to only foot travel and minimal vehicular traffic, while the main roads were subject to a higher volume of traffic, which may explain the difference in mortality rates. The numbers also indicate that mortality rates were much higher in the areas where borrow pits were located near the road (0.304 DOR/km; Fig. 1). Amphibians and reptiles cross the road due to the presence of adjacent ponds used for breeding and foraging. The presence of specific reptiles and amphibians on the roads varied during the seasons. While no data was collected on vehicular traffic during each season, increased road activity did occur during open hunting seasons at OSBWMA. Overall, more animals were seen during the summer, but with a larger percentage of the amphibians recorded dead. More amphibians were recorded during the summer, and lower numbers in the fall, yet both seasons had lower mortalities observed. In the spring, total observations were lower, and proportionately more amphibians and reptiles were DOR (Fig. 2). Mortality of snakes were affected the most in the spring and fall (61.5% and 60.0%), while 346 Southeastern Naturalist Vol.7, No. 2 anurans were affected more frequently in the spring (83.3%). For snakes, mortality appeared to be high for some species such as Crotalus horridus (Timber Rattlesnake), Storeria dekayi (Brown Snake), and Opheodrys aestivus (Rough Green Snake) (Fig. 3). These species may be slower to cross the road, making them more susceptible to traffic. Another potential cause of increased mortality is the more-open space of the roads creating basking sites for many species. Specific anuran observations varied more by season, which is probably related to breeding cycles. The higher mortality in the spring may refl ect that most reproduce in this season and migrate to standing water for breeding aggregation (Fig. 1). Winter had no road mortality, which may refl ect a cold, dry winter and hence the decrease in movement. The variation in reptile and amphibian mortalities on the roads in this bottomland suggests that there are several factors involved. Other studies have suggested that weather, amount of traffic, and seasonal activity of local species all contribute to the number of animals killed in particular areas (Dodd et al. 2004, Main and Allen 2002). Larger snakes are more susceptible to road mortalities, since they take up a larger amount of the road than smaller animals, which are easily avoided. The area around the oil rigs, although they had year-round activity, did not have higher mortalities, which is possibly due to a lower abundance of animals in those areas (Hampton 2004). This lower abundance may be attributed to more open areas, with fewer breeding and foraging areas. The secondary roads had lower mortality rates, which may refl ect fewer vehicles or those traveling at lower speeds. Seasonal comparisons can give additional insight into how these roads are impacting various species. For example, in summer as ephemeral pools begin to dry, amphibians become concentrated in and around them. Those areas become prime foraging sites for many semi-aquatic snakes (Bernardino and Dalrymple 1992). Frogs and snakes were frequently seen on roads. The use of ephemeral pools as foraging sites may explain why areas with borrow pits across from each other showed a higher mortality in our study. During the fall, several open hunting seasons are allowed on the OSBWMA, and the highest mortalities did occur in that season when traffic increased, and may pose the greatest threat to reptile and amphibian species. In the cooler months, they move more slowly, and may tend to bask on roads for longer periods of time, leaving them vulnerable to the increased traffic (Zug et al. 2001). The lack of records in the winter is obviously related to the weather, but as previously mentioned, the lack of records of winter breeding amphibians may also relate to the timing of traffic in the area. Although human hunting activity is reduced in spring and the activity of amphibians and reptiles is high, the number of DORs was just as high as in the fall. Possibly, the few animals occurring on the roads are remaining even longer, but this conclusion is premature. In fact, it may be in response to a variety of other possibilities including; seasonal activity periods, ambient temperatures, weather patterns, vagility, and behavioral ecology. In order to get a clear understanding of the number of animals found on the road, a more in-depth study needs to be conducted to address these issues. The OSBWMA, with relatively few roads, had mortality rates much lower than comparable areas with busy roads (Dodd et al. 1989, Pickles 2008 J.L. Coleman, N.B. Ford, and K. Herriman 347 1942). However, even with minimal roads in a given area, animals may still be affected because roadways interrupt the habitat and inhibit species movements from one area to the next (Forman and Alexander 1998). In particular, borrow pits create habitats that are ideal for the herpetofauna in bottomlands; however, the location of these pits must be considered given their role of increasing road mortalities as seen in this study. Obviously more information is needed to obtain an accurate picture of how reptiles and amphibians are being affected by roads in a floodplain, but one obvious management recommendation is to carefully consider the location of borrow pits during the construction of trails and roads. Based on this research, borrow pits should not be excavated across the road from one another and should be located far enough from the road to minimize mortality by species entering or leaving the pits. Acknowledgments We thank C. Whitfield, S. Wilson, J. Coleman, and D. Coleman for their assistance in the field, and Texas Parks and Wildlife Department for allowing us to conduct the research at OSBWMA. We thank P. Hampton, J. Hargrove, and R. Gutberlet, Jr. for their constructive criticism. Financial support was provided by the UT-System Louis Stokes Alliance for Minority Participation, with N. Ford serving as LSAMP faculty mentor. Literature Cited Bernardino, Jr., F.S., and G.H. Dalrymple 1992. Seasonal activity and road mortality of the snakes of the Pa-hay-okee wetlands of Everglades National Park, USA. Biological Conservation 62:71–75. Carr, L.W., and L. Fahrig. 2001. Effect of road traffic on two amphibian species of differing vagility. Conservation Biology 15:1017–1078. Clark, J.R., and J. Benforado (Eds.). 1981. Wetlands of Bottomland Hardwood Forests. Elsevier Scientific, Amsterdam, The Netherlands. Clevenger, A.P., B. Chruszcz, and K.E. Gunson. 2003. Spatial patterns and factors infl uencing small vertebrate fauna road-kill aggregations. Biological Conservation 109:15–26. Dodd, Jr., C.K., K.M. Enge, and J.N. Stuart. 1989. Reptiles on highways in North- Central Alabama, USA. Journal of Herpetology 23:197–200. Dodd, Jr., C.K., W.J. Barichvich, and L.L. Smith. 2004. Effectiveness of a barrier wall and culverts in reducing wildlife mortality on a heavily traveled highway in Florida. Biological Conservation 118:619–631. Enge, K.M., and K.N. Wood. 2002. A pedestrian road survey of an upland snake community in Florida. Southeastern Naturalist 1:365–380. Enge, K.M., and K.N. Wood. 2003. A pedestrian road survey of the Southern Hognose Snake (Heterodon simus) in Hernando County, Florida. Florida Scientist 66:189–203. Fahrig, L., J.H. Pedlar, S.E. Pope, P.D. Taylor, and J.F. Wegner. 1995. Effects of road traffic on amphibian density. Biological Conservation 73:177–182. Findlay, C.S., and J. Houlahan. 1997. Anthropogenic correlates of species richness in southeastern Ontario wetlands. Conservation Biology 11:1000–1009. Forman, R.T., and L.E. Alexander. 1998. Roads and their major ecological effects. Annual Review of Ecology and Systematics 29:207–231. 348 Southeastern Naturalist Vol.7, No. 2 Groot Bruinderink, G., and E. Hazebroek. 1996. Ungulate traffic collisions in Europe. Conservation Biology 10:1059–1067. Hampton, P. 2004. Effects of management techniques on amphibian and reptile populations in an East Texas fl oodplain. M.Sc. Thesis. University of Texas, Tyler, TX. Haxton, T. 2000. Road mortality of Snapping Turtles, Chelydra serpentina, in central Ontario during their nesting periods. Canadian Field Naturalist 114:106–110. Lodé, T. 2000. Effects of a motorway on mortality and isolation of wildlife populations. Ambio 29:163–166. Loughry, W.J., and C.M. McDonough. 1996. Are roadkills valid indicators of armadillo population structure? American Midland Naturalist 135:53–59. Main, M.B., and G.M. Allen. 2002. Landscape and seasonal infl uences on roadkill of wildlife in southwest Florida. Florida Scientist 65:149–158. Mazerolle, M.J. 2004. Amphibian road mortality in response to nightly variations in traffic intensity. Herpetologica 60:45–53. Means, D.B. 1999. The effects of highway mortality on four species of amphibians at a small, temporary pond in Northern Florida. In Proceedings of the Third International Conference on Wildlife Ecology and Transportation. September 13-16, Missoula, MT. Molles, Jr., M.C., C.S. Crawford, L.M. Ellis, H.M. Valett, and C.N. Dahm. 1998. Managed fl oods: Restoration of riparian forest ecosystems structure and function along the Rio Grande. Bioscience 48:749–756. Noss, R. 2004. The ecological effects of roads. Wildland CPR, Missoula, MT. Oxley, D.J., M.B. Fenton, and G.R. Carmody. 1974. The effect of roads on populations of small mammals. Journal of Applied Ecology 11:51–59. Pickles, W. 1942. Animal mortality on three miles of Yorkshire roads. Journal of Animal Ecology 11:37–43. Rosen, P.C., and C.H. Lowe. 1994. Highway mortality of snakes in the Sonoran Desert of southern Arizona. Biological Conservation 68:143–148. Scott, T.G. 1938. Wildlife mortality on Iowa highways. American Midland Naturalist 20:527–539. Seigel, R.A., R.B. Smith, J. Demuth, L.M. Ehrhart, and F.F. Snelson, Jr. 2002. Amphibians and reptiles of the John F. Kennedy Space Center, Florida: A long-term assessment of a large protected habitat (1975–2000). Florida Scientist 65:1–12. Taylor, B.D., and R.L. Goldingay. 2004. Wildlife road-kills on three major roads in north eastern New South Wales. Wildlife Research 31:83–91. van Gelder, J.J. 1973. A quantitative approach to the mortality resulting from traffic in a population of (Bufo bufo) L. Oecologia 13:93–95. Vos, C.C., and J.P. Chardon. 1998. Effects of habitat fragmentation and road density on the distribution pattern of the moor frog Rana arvalis. Journal of Applied Ecology 35:44–46. Zug, G.R., L.J. Vitt, and J.P. Caldwell. 2001. Herpetology: An Introductory Biology of Amphibians and Reptiles. 2nd Edition. Academic Press New York, NY. 630 pp.