INTRODUCTION

The wetlands and mesic prairies of Illinois are among the few ecosystems of their kind left in the world, with national decline and habitat destruction largely contributing to agricultural development and suppression of fire that helps maintain these areas (Perry et al. 2009). Since 1818, Illinois has lost nearly 90% of its wetlands and over 95% of its prairies (Illinois DNR 2018). Destruction of ecosystems equates to a loss of habitats for the species that rely on it, and these spatial declines are coupled with plummeting populations of reptiles and amphibians worldwide (Durso et al. 2011). Herptiles native to wetland-prairie habitats are reliant on the natural hydrology and features of the landscape (Most 2013), making the degradation of the land even more destructive; populations are pushed out through lack of resources and overcompetition, and the habitats become fragmented, leaving them isolated from historically connected ecosystems (Baecher et al. 2018). Changing habitat structure due to human interference, specifically fire suppression, can cause prairies to give way to closed-canopy forests with shade-tolerant understories, which alters the original community of species able to survive there (Perry et al. 2009)

In the past several decades, however, there has been greater effort to restore these natural spaces to pre-industrial era quality, usually through eradication of non-native plant species and incorporation of prescribed burns (Illinois DNR 2018). This has aided in conserving prairies and wetlands, and in creating habitat corridors, which are a necessary feature that allows for interconnectedness between otherwise isolated fragments, increasing biodiversity (Baecher et al. 2018). With the return of historic vegetation structure and habitat heterogeneity (Litt et al. 2001), the goal is for native species’ populations to also rise and return to levels similar to those before destruction, or comparable to those of other restored or preserved habitats.

Despite comprising a substantial portion of vertebrate biomass and diversity in prairie and wetland habitats (Baecher et al. 2018), amphibians and reptiles are notoriously cryptic and therefore direct studies on the impact of habitat degradation and restoration on their populations are limited (Most, 2013). Survey efforts in fragmented or restored habitats are often focused on more conspicuous species - such as mammals, birds, or insects - and plants are typically the focal taxa for restoration in terms of reintroduction (Baecher et al. 2018). However, herptiles fulfill key trophic roles in their ecosystem and efforts to understand the impact of restoration on their populations is an important consideration for land management and effective restoration catering to the requirements of the species contained within (Durso et al. 2011), especially those that are rare or of distinct interest.

Melody Savanna is a recently restored portion of the wetland-prairie landscape of Illinois, and no previous observations or studies have been conducted at this location. Our objective through this herpetological survey is to provide insight into the quality of the habitat and types of populations present in the early years of restoration.

METHODS

The survey was conducted for seven weeks from September to October, 2018, at the Melody Savanna, Lake Forest, IL. Two professors and ten students from Lake Forest College participated in data collection and trappings. Five trapping methods were used in this survey: hoop net traps, minnow traps, pitfall traps, funnel traps, and coverboards. Students were broken into three groups responsible for either aquatic trappings (hoop net and minnow traps), trappings that utilized a drift fence (pitfall and funnel traps), or coverboards. Apart from traps, everyone also conducted time-constrained visual encounter surveys (TC-VES), also known as meandering surveys. This survey method collects species caught opportunistically while walking the field freely or through positive identifications of species by sight or sound (such as frog calls). On Mondays from 12-3pm, students checked their designated traps and recorded the species found in the trap and/or by TC-VES. On the remainingdays of the week from 12-3pm, a professor or volunteer students would check the traps and record meandering surveys. By the end of six weeks, every group shared their findings to combine a complete herpetological report of the property. 

Description of the study site

The Melody Savanna is privately owned and is approximately 20 acres in size. This is a newly restored habitat and little study has been done on it. The site is adjacent to busy roads, residential properties, and the Lake Forest Open Lands Association preserve. Although much of the Melody habitat is categorized as a savanna, other microhabitats were classified either as wet-mesic prairie, mesic prairie, shoreline, or wet savanna (Fig. 1). TC-VES surveys were conducted throughout the entire property. 

Trap Data and TC-VES survey

Each group designed their own trap and TC-VES data sheets. These sheets recorded measurements of every species caught, heard, or seen. When appropriate, the individual species were identified, sexed, weighed using a spring scale, and categorized as juvenile, sub-adult, or adult by body size. The total length (TL) and snout-vent length (SVL) were measured for snakes, the curved carapace length (CCL) was measured for turtles, and body length were measured for frogs. The weather, temperature, and location of the observed species were also noted. 

Aquatic – Hoop net and Minnow Traps

Our team was responsible for the aquatic traps. One large hoop net trap, two small hoop net traps, and a minnow trap were scattered throughout the shoreline with special preference to areas with previously observed turtle activity. The hoop net traps were placed in the water and staked to the ground while leaving approximately six inches above the water line for animals to surface and breathe. Each trap was baited with sardine cans attached to the net. If needed, the traps were replaced or relocated to a different area of the shoreline due to unsuccessful trappings or increased water levels submerging the traps completely. Aquatic traps must be checked every 24 hours. 

Drift fence array – Pitfall and Funnel Traps

The drift fence array was assembled in the savanna in quadrant 23 (Fig. 1). The drift fence stretched from North to South and consisted of 24 ft. burlap staked every four feet. Coffee cans were placed below ground level at each of the drift fence and one in the middle East side of the fence. A funnel trap was placed in various places along the drift fence. Like the aquatic traps, pitfall traps should be checked constantly. The traps may collect water, debris, or dirt because of strong weathers, which will lower the trapping efficiency. 

Coverboards 

A total of 9 coverboards were placed throughout quadrants 10, 11, 12 and 17, in the savanna or mesic prairie habitat (Fig. 1). Coverboards ranged from three feet to five feet in length. Some coverboards were already placed in the site before this study began and natural covers (e.g. logs, rocks, and tree barks) were also checked. To check coverboards, one person would slowly lift the coverboard using a snake hook while another member would capture the animal below in a snake bag. Since these coverboards serve as homes for some species, they should not be disturbed too much. Thus, unlike aquatic and drift fence traps, coverboards were left in place after this research project was complete. 

RESULTS

A total of seventy amphibians and reptiles were identified across eleven species during the seven-week survey period. Of the observed species, 66.2% were frogs, 21.1% were snakes, 11.3% were turtles, and 1.4% were salamanders (Fig. 2). The most observed species was the bullfrog (Rana catesbeianus), with a total of twenty-seven individuals, and the species with only one individual observed were the tiger salamander (Ambystoma tigrinum), the Eastern milksnake (Rana triangulum), and the North American racer (Coluber constrictor) (Fig. 3). 

The majority of the herpetofauna (72.9%) was either seen, captured, or both, utilizing the TC-VES method. The remaining 27.1% of herpetofauna were captured using other trapping methods. The aquatic traps accounted for 17.2% of captured species; the pitfalls, funnel traps, and drift fence captured 7.1%; and the coverboards captured the remaining 2.9% of the total herpetofauna during the survey period (Fig. 4). Species that were not physically captured during TC-VES surveys but were positively identified from a distance were frogs and painted turtles (Chrysemy picta) seen basking on rocks or logs. Also, Western chorus frogs (Pseudacris triseriata) were positively identified by mating calls in quadrant 5-7 on week 5, while the North American racer (Coluber constrictor) was identified by its fresh shedding on week 4 (Fig. 3).

The highest number of herpetofauna were observed or captured during week 2 through 4, and the fewest sightings and captures occurred during weeks 1 and 7 (Fig. 3). The temperature was highest in week 3 and lowest in week 7. Out of the five habitats in the Melody property, herptiles were found in only three of them: the shoreline, mesic-prairie, and savanna (Fig. 5). 67.6% of the animals were found by the shoreline while the savanna and mesic prairie contained 21.1% and 11.3% of the animals respectively (Fig. 5). 

Aquatic – Hoop Net and Minnow Traps

Aquatic trapping yielded a total of eleven individuals – seven R. catesbeianus (six adults, one tadpole) and four C. picta. (Fig. 4). The C. picta had a CCL range of 12.5 to 20.5 cm, classifying them as either sub-adult or adult. Only the large hoop trap (quadrant 17, Fig. 1) and minnow trap (quadrant 16, Fig. 1) captured any herptiles. All traps, however, did capture a native species of crayfish, Orconectes virilus, in various life stages from larvae to adult. Testing was also conducted on the pond water using strip tests and revealed the water contains little to no ammonia and nitrites levels, signifying healthy water chemistry (Fig. 6). However, the water was found to be slightly basic, with a pH of 8.53 (Fig. 6).

Drift fence – Pitfalls and Funnel Traps

The pitfall traps caught four different species. Pitfall trap 1 recovered two juvenile Chelydra serpentina, likely less than two weeks old, based on size and presence of egg teeth. This trap also captured one juvenile Anaxyrus americanus (Fig. 3 &4). No captures were made in pitfall trap 2, and pitfall trap 3 captured one adult Ambystoma tigrinum (Fig. 3 & 4). The funnel trap situated along the drift fence captured a female T. sirtalis (Fig. 3 & 4). Additionally, a female T. sirtalis shed skin was found located just south of the drift fence (Fig. 3).

Coverboards

Two species were caught in the coverboards on week 4 (Fig. 3 and 4). The C. constrictor was not physically caught, but it had left a fresh skin shed under the coverboard (Fig. 3). Kim Wismann from Planet of Wonders positively identified the snake shed. The group also trapped a sub-adult T. sirtalis utilizing TC-VES method. 

DISCUSSION

Our survey of the Melody Savanna revealed high overall diversity of amphibians and reptiles with eleven total species identified, with two clades of amphibians (frogs and salamanders) and two clades of reptiles (snakes and turtles) present. Moreover, our predictions were on par with similar surveys of restored prairies in the area, such as the Middle Fork Savanna herpetofaunal survey (Mierzwa & Beltz 1994). Of the thirteen species found in the Middle Fork Savanna, this study site contains eight of the same species, including all the frog species they found, and we also recorded two species the original study did not find: Coluber constrictor and Lampropeltis triangulum. Thus, the Melody Savanna provides a habitat suited for these species while also housing more common species of the area. Despite the abbreviated length of the survey, a diverse number of species in various life stages were observed, indicating a functioning reproductive group. The property as a restored space is relatively young but already supporting healthy abundance and diversity of reptiles and amphibians.

Herptiles, especially frogs, were located mostly by the shoreline, as signified by aquatic traps capturing the most species. Amphibians constitute the majority of the total species captured, demonstrating that the Melody property provides a high-quality habitat for amphibians. However, reptiles like T. sirtalis and C. picta were also found by the shoreline. This is likely due to the pond being the only large, available water source in the Melody property. Many herptiles, particularly amphibians, rely on a steady water source for survival and reproduction. Another factor that could influence the high number of encountered species in the aquatic habitat is the presence and availability of food sources located in the shoreline, such as crayfish, minnows, insects, and tadpoles. Crayfish tunnels also provide optimal hibernacula for many amphibians and reptiles during the cold winter months as protection from low temperatures.  

There were a high number of observations between week 2 and week 4, but an overall downward trend in the number of species observed or trapped from week 1 to week 7. This is likely due to the cooling temperatures at the site during the survey period and conducting the survey late in the season. In the fall, many herptiles remain close to their hibernacula and spend less time above ground as they prepare for hibernation. Nevertheless, frogs were still observed and trapped at temperatures as low 51 F. This strongly indicates that during a more appropriate season, there could possibly be even greater diversity that went unseen at this time.

This was the first herptile survey report conducted at the Melody site, so there are a few important ideas to note for future surveying. For aquatic traps, only the large hoop net trap caught herptiles, so that trap size should continue to be used. Surveyors would also likely benefit from the use of other trapping methods. For example, floating turtle traps would allow for greater utilization of the open water habitat, as it is the most challenging to access. Other studies have also demonstrated how utilizing glowsticks as bait increases amphibian funnel trappings (Antonishak, 2015). 

Though the coverboard group found a wide variety of prey items, such as voles and mice – a prime food source for herpetofauna under the coverboards – the capture rate was still low. The varied success rate is likely due to the short length of time the coverboards were established in their respective locations. We recommend placing more coverboards throughout the property and leaving them in their locations for a longer period of time before checking or moving them; this way species will have time to adapt to using them without disturbance. The pitfalls and drift fence were more successful than coverboards but still showed a relatively low capture rate when compared to the aquatic hoop net traps. A higher capture success rate may be achieved if the drift fence is placed either directly in front of the wood pile that provides hiding and hibernation space for herptiles, or if it is placed closer to the water to intercept more species coming and going from the pond.

The aquatic habitat located on the property is of vital importance to the overall health and biodiversity of the survey site. Many different species rely on this area for water access, hibernation sites, reproduction, development, and food sources. However, the aquatic zone is especially prone to degradation by other predators, particularly raccoons. The presence of raccoons in low numbers is not cause for alarm, but a large presence can be detrimental to the other species within the habitat. Raccoons can dramatically alter the biodiversity and health of a habitat by excessive predation of young, juvenile, and adult herptiles (Feinberg & Burke 2003). This habitat may also be vulnerable to invasive crayfish and turtle species. The Red Eared Slider (Trachemys scripta elegans) has become invasive to the Great Lakes region, and it is unknown how their presence can alter ecosystems. However, it is assumed they can alter native ecosystems by disturbing the habitat balance native species depend on (Thomson et al. 2010). It is of vital importance to protect species inhabiting the pond from outside influences and invasive species, which could dramatically alter this crucial ecosystem. Even though the property has only recently been restored, the abundance and diversity of its herpetofauna is promising for continued growth, and these methods will ensure accurate monitoring of the population.

ACKNOWLEDGEMENTS

We would like to thank our professors, Dr. Sean Menke and Prof. Rob Carmichael, for guiding us with the project and providing us with all the materials needed to conduct the traps. We also owe Robin Colburn a big thanks for allowing us to conduct our research project on her property. 

LITERATURE CITED 

Antonishak, M., Muñoz, D. J., & Miller, D. A. W. (2017). Using glow sticks to increase funnel trap capture rates for adult vernal pool amphibians. Herpetological Review, 43, 544-549.

Baecher, J.A., P.N. Vogrine, J.C. Guzy, C.S. Kross, & J.D. Wilson. 2017. Herpetofaunal communities in restored and unrestored remnant tallgrass prairie and associated wetlands in northwest Arkansas, USA. Wetlands 38: 157-168.

Durso, A.M., J.D. Wilson, C.T. Winne. 2011. Needles in haystacks: estimating detection probability and occupancy of rare and cryptic snakes. Biological Conservation 144: 1508-1515.

Feingberg, J.A. & R.L. Burke. 2003. Nesting ecology and predation of diamondback terrapins, Malaclemys terrapin, at Gateway National Recreation Area, New York. Journal of Herpetology 37(3): 517-526.

Illinois DNR. 2018. Conservation programs. www.dnr.illinois.gov.

Litt, A.R., L. Provencher, G.W. Tanner, & R. Franz. 2001. Herpetofaunal responses to restoration treatments of longleaf pine sandhills in Florida. Restoration Ecology 9(4): 462-474.

Mierzwa, K. S., & Beltz, E. (1994). Habitat Associations and Distribution of Amphibians and Reptiles at Middle Fork Savanna, Lake County, Illinois. A Report to the Lake Forest Open Lands Association, The Nature Conservancy, and the Lake County Forest Preserve District.

Most, M.G. 2013. Activity patterns and spatial resource selection of the Eastern garter snake (Thamnophis sirtalis sirtalis). Master’s Theses. Paper 1465. 

Perry, R.W., D.C. Rudolph, & R.E. Thill. 2009. Reptile and amphibian responses to restoration of fire-maintained pine woodlands. Restoration Ecology 17(6): 917-927.

Thomson, R.C., P.Q. Spinks, & H.B. Shaffer. 2010. Distribution and abundance of invasive red-eared sliders (Trachemys scripta elegans) in California’s Sacramento River Basin and possible impacts on native western pond turtles (Emys marmorata). Chelonian Conservation and Biology 9(2): 297-302.

Appendix: Figures

Figure 1: A map of the Melody Savanna and its different habitat types. Location of traps that only caught an amphibian or reptile

Figure 2: Percent Abundance of the herptile clades observed

Figure 3: Amphibians and reptiles collected at Melody Savanna. At least 1 species was identified by their * skin shed or ** mating callsFigure 4: Types and abundance of species caught in each trap. *At least 1 tadpole trapped.Figure 5: Abundance of species found in each habitatFigure 6: Water quality assessment of the pond using strip kits. This is not an exact measurement of the values but is within the range.