Investigating the Ecology of Chelydra s. Serpentina, the Common Snapping Turtle, in a Highly Urban Setting
Tranquility was all around me. I couldn't even hear the sound of my own breath, for that was taken from me when I gazed out at the scene, simply awestruck by the amazing display of natural beauty set before me. Never before had I felt so at one with nature. The gleam of the sun on the crystal-clear water, the wind blowing through the trees, and the birds flocking and soaring above my head made it hard for me to remember where I was. I was not in the Amazon; I was in the wilds of Africa; I was not standing atop an isolated mountain range overlooking a desolate wilderness. No, I was right in the heart of the Bronx, doing what I was born to do.
From the time I was a small child, I was always finding myself in the outdoors, whether it be hiking, fishing, skiing, or kayaking. Thus, it was natural that I'd eventually bump into a reptile--and my lifelong passion would emerge. Only one word can describe the way I've felt ever since--fascinated. That word best describes why I do what I do. Every moment of free time I can get my hands on, I am in the field, searching for reptiles and loving every second of it.
During the warmer months of 2009, I was finally able to use this passion; I singlehandedly captured and collected data from 225 snapping turtles across the Bronx as a part of my research project. I look back on the experience and consider it one of the best times of my life. I spent almost every day in the field, whether it was after school, on weekends or during the summer break. Spending my time rowing in search of turtles, I was driven not only by passion but also by curiosity.
Large reptiles are not common in the midst of people, especially in northern regions. The common snapping turtle is certainly a large reptile, weighing up to 86 lbs (Hulse 2001). I found myself wondering how such a large and magnificent reptile manages to survive right here in New York City. It is one thing to see a snapping turtle in the country, but such a creature seems almost out of place in the city. I finally came up with a question. What are the characteristics of urban snapping turtle populations, and do they differ from more natural ones? My desire to answer this question made me ready to fully put my curiosity to the test.
The goals of my proposed research were simple. I wanted to investigate this species within an urban ecosystem. I wanted to know more about its life history by investigating its sex ratio, size and weight, population structure, and population density. By explaining these differences and attributing them to various environmental factors, I could have a better understanding of this species and its ecology in a highly urban setting.
Chelydra s. serpentina, the common snapping turtle, is one of the largest freshwater turtles in North America. This species has a large head, powerful jaws, a long tail, and big forelimbs with large claws. When facing a threat, snapping turtles face their attacker and use their long neck to aggressively lunge forward and bite. Females travel great distances over land to find nesting sites (Hulse 2001, Congdon et. al 1987). Males are highly territorial and fight aggressively (Hulse 2001). This species is known for its tolerance to pollution, and is able to survive and even thrive in habitats where other turtle species perish (Hulse 2001). It is one of a small group of species able to live among humans and tolerate highly unnatural settings. By continuing to learn more about this species, we can help insure that it will remain extant and add to the rich biodiversity of life on Earth.
The snapping turtle is a very common and frequently studied species. Studies have covered many aspects of its life history, including diet and food habits (Alexander 1943, Froese & Burghardt 1974), nesting habits (Congdon 1987, Petokas & Alexander 1980), population density (Froese & Burghardt 1975, Galbraith 1988) and movements (Obbard 1981). All have shown that the common snapping turtle is a highly adaptable and variable species. However, few studies have taken place in an urban setting such as New York City. I hypothesized that populations in urban settings would show the same variations seen in wild populations, but that living among people would make them different from wild populations. By comparing my results to previous studies, it would be possible to learn if urban populations of this species differ from those in more natural settings. This study could enhance our understanding of the effects of urbanization on wildlife. In a world where humans are swarming over the planet at a pace too fast for other species to catch up, it is important to have an idea of what is going to happen to the billions of other life forms with which we share our home.
Data was collected from many parts of the Bronx. The four main study sites were located in Van Cortlandt Park, the Bronx Zoo, and the New York Botanical Gardens. In Van Cortlandt Park, most of the data was collected from snapping turtles living in Van Cortlandt Lake, a large, 16-acre lake lying in the middle of a heavily used public park. At the Bronx Zoo, turtles were captured in the Mitsubishi Riverwalk. A waterfall separates the Riverwalk section from the "Lower River," a section of the Bronx River stretching from the waterfall to the end of the Bronx Zoo property. In the Botanical Gardens, the Twin Lakes were studied. This is a body of water located in the northwestern section of the park. Some data was also collected from a tiny isolated pond in Pelham Bay Park.
Materials And Methods
Field research took place between April 2 and August 28 of 2009. Turtles were obtained by means of hand capture or trapping. Hoop traps manufactured by Dixie Gardens Nets were used. The traps were four feet long with a diameter of 2.5 feet. They were made of nylon netting with two-inch mesh, held together by three stainless-steel hoops. The nets were braced by two pipes made of polyvinyl chloride and held in place by a third. Bait was placed in a peanut butter jar with holes punched in it, or was hung from the netting by a zip tie. Whole bunker or chunks of carp were used as bait. Traps were checked at least once every 24 hours. At any given time, up to 12 traps were present in a given area. Observations of each habitat were made to assess the degree of human disturbance in the area.
The population structure for each location was analyzed. Each individual was placed in a general category based on its carapace size (Figure 1). The size of the carapace shows the general age class (juvenile, adult, old, etc.) of the individual turtle. Each age class made up a specific percentage of each population, and these percentages were analyzed.
To sample the populations, the mark-recapture sampling method was used. Every day was counted as a different sample, and the samples were averaged to obtain a proper estimate. On a given day, the number of previously marked turtles captured in the traps was counted along with the turtles that weren't marked. Then, based on the overall number of turtles that had been marked on the site, a proportion was set up to learn the size of the population. For example, if 3 marked turtles were captured along with 3 unmarked, and 50 turtles were previously marked, there would be 100 turtles in the population. Lagler (1943) also used the mark-recapture method to estimate population size in this species.
After learning the size of the population, the next goal was to discover the population density. This was obtained by dividing the total number of turtles in the population by the number of hectares of surface area.
Analyzing the data collected during this study revealed that, aside from several general similarities, each population exhibited significantly different parameters. Each population had a male-biased sex ratio, but to varying degrees. In the Bronx River, the sex ratio ranged from 1.24 to 1.37 males for every female (Table 2). In the other sites, there were three males for every female. This is a big difference, as some populations had more than twice as many females as others.
Population density at the three sites showed drastic differences as well, ranging from 24.11 to 125 turtles per hectare (Table 1). The highest density was more than five times greater than the lowest density. The structure of each population exhibited some similarity in that adult males always made up the largest percentage of the population. However, the structural makeup of the other age classes varied greatly between the populations (Figure 1). Average size and weight for females lay within a small range, but average size and weight for males was very different at each location (Table 1). In the Botanical Gardens at Twin Lakes, the males were the smallest, averaging 12.7 inches in length and weighing 18.2 pounds. In Van Cortlandt Park, the males were the largest, averaging 14.9 inches in length and weighing 28.5 pounds, twice as heavy as at Twin Lakes. Sex ratio, population structure, population density, and the size of the males showed a large degree of variation between the four populations studied.
Of the 225 Bronx specimens sexed, 147 were male and 78 were female. The lack of females in these populations could be related to the nesting habits of this species. Female snapping turtles come on land to lay their eggs. In some cases, females travel great distances over land to reach their nesting sites. Congdon (1987) studied the nesting habits of a Michigan population, discovering that females nested an average of about 120 feet from the water, and sometimes traveled distances of almost 600 feet. Snapping turtles are slow on land, making them vulnerable to motor vehicles and human harassment. Females searching for nest sites are commonly killed on roads. Thus, the sex ratio can be attributed to the high rate of mortality for adult females.
The overall sex ratio in the populations was male-biased, but examining the sex ratio of immature specimens revealed a 1:1 sex ratio. Petokas and Alexander (1980) studied maturity in a New York population of snapping turtles and concluded that specimens were mature once they reached a carapace length of about 8.75 inches, and only mature females lay eggs. Examining the sex ratio of all Bronx specimens smaller than 8.75 inches (immature specimens) showed a sex ratio of exactly 1:1. This supports the idea that females are dying once they are mature, and that the turtles aren't born with a male-biased sex ratio.
In most populations, old females make up only about 5% of the population, but in the Lower River, old females made up about 17% of the population (Figure 1). This further supports the conclusion that female mortality is related to human disturbance. The percentage of old females reflects the significantly higher survivorship of females in the Lower River population in relation to others. Females are able to survive to a greater age because they are less disturbed by humans.
The sex ratio varied throughout each population, so females faced more danger in some areas than others. The dangers females faced can be attributed to human disturbance. Potential exposure to human disturbance varied in each area. The main disturbing factors were roads and footpaths. The populations with the highest male-to-female ratio lived in habitats with the most heavily used paths and roads (Table 2). It can be inferred that females nesting in these areas were occasionally killed by automobiles or by cruel people. The Bronx River populations faced little exposure to people. In the Lower River, a habitat with no footpaths and only one small road, the male-to-female ratio was 1.24 to 1, which would be expected.
The populations analyzed had different population densities (Table 1). When the populations are placed in order from smallest to largest density, it is noted that they are also in order from largest to smallest habitat size (Table 1). Population density was negatively correlated with the size of the habitat. The smallest habitats had the highest densities.
This is consistent with a study published in the Canadian Journal of Zoology in 1988 (Galbraith 1988). This study examined the factors that affect the population density of the common snapping turtle through a study of two ponds in Ontario and a review of past literature. Galbraith determined that it was either surface area or latitude that affected population density. This Bronx study narrowed it down further, showing that it is surface area that has the biggest effect on density. The correlation between surface area and population density may be related to adult survivorship in this species.
It is reasonable to hypothesize that the high density of this species in smaller areas can also be directly attributed to the territoriality of males (Hulse 2001). In a larger area, a male can have a larger territory, which means that it has a larger area to defend. Since snapping turtle populations are less dense in larger areas, it is reasonable to assume that a male turtle can wander and occupy a larger home range with a smaller chance of encountering another male. Since it has a smaller chance of getting into fights, it can maintain a larger territory more easily. However, in a smaller environment that is much denser, it cannot maintain the same amount of territory because it would have to fight off too many other males. Therefore, turtles in smaller areas occupy smaller territories. It is known that snapping turtles tend to remain in the same home range for many years or even their entire lives (Obbard 1981), so once they establish their home range or territory, they will stay there. In a smaller area there are also fewer resources available for survival. Since resources are limited, there is a greater amount of competition, which may lead the turtles to become more territorial than they would be in an environment with less competition. Froese and Burghardt (1974, 1975) also hypothesized that the territoriality of this species may affect population density.
The highest population density was found in Pelham Bay Park. Galbraith (1988) reviewed most studies concerning population density in this species, and the highest density he came across was only 74 turtles per hectare (Major 1975). This is interesting, especially since the density of the Pelham Bay Park population is almost surely greater than I measured. Due to limited trapping success, an accurate sample was not taken from this area. However, a total of 10 individual turtles were captured, so the population consists of at least 10 individuals, most likely more. With 10 turtles living in the tiny pond, population density would be a staggering 125 turtles per hectare. Density in this pond is probably high due to the fact that the pond is incredibly small (only 0.08 hectares). The high density found by Major (1975) was also from a small pond, about 0.4 hectares in size. Of the other studies reviewed by Galbraith (1988), the highest densities were also found in small ponds (Major 1975, Froese and Burghardt 1975, Galbraith 1988). This shows that dense snapping turtle populations occur most commonly in small ponds.
This research shows that all four populations studied in the Bronx exhibited significantly different parameters. The study indicates that this species is very adaptable and can adjust to the limits of the habitat in which it lives. Studies have shown that descriptions of a "typical" snapping turtle population are invalid, as each population is different and highly variable in life history parameters such as population density (Froese and Burghardt 1974, Galbraith 1988), home range size (Obbard 1981), and even diet (Alexander 1943). This study shows that the Bronx is no exception. The populations here differ from those in other areas and from each other. My hypothesis was correct.
Significant conclusions were drawn from this study, but further research is needed to gain a better and more complete understanding of the life history of this species in an urban setting. Data was collected from only five locations. Collecting from many other sites within New York City or other urban areas will make the data analysis more accurate and give a broader understanding of snapping turtles. Also, collecting data from more natural locations would also be useful, and would make comparison between an urban and a natural environment easier.
Although it was seen that snapping turtles are able to adapt and survive in the Bronx, it was also seen that they are not shielded from the negative effects of human disturbance. There was a direct correlation between human activity in the area and the unbalanced sex ratio. The sex ratio was significantly more male-dominant in areas heavily exposed to humans. It was most male-dominated in Van Cortlandt Park and the Botanical Gardens, where human recreation is most common. This is similar to the results of a 20-year study of the relationship between human recreation and declining turtle populations (Garber 1995). A private piece of land was opened to the public, and turtle populations almost immediately declined as a result. Another long-term study documented the very heavy decline of reptile species due to development and human interference (Minton 1968). Snapping turtles are less vulnerable than other species due to their size and highly aquatic nature. However, nesting females, due to their smaller size and terrestrial habits, suffer greater negative effects from humans. Other studies have also pointed to nesting as the major issue facing turtles in urban areas (Conner 2005).
Overall, this study has improved the understanding of this species by fortifying and supporting the knowledge already obtained in past studies. The correlation between habitat surface area and population density is very important. It provides answers but also generates more questions about a species that has already been frequently studied. Since snapping turtle populations in urban areas face the most danger, it is important that we learn more about them to minimize this danger. Other studies have taken place in disturbed areas, but not in areas as heavily urbanized as the Bronx. This study is one of the few, if any, that have examined the life history of Chelydra s. serpentina in an environment so directly affected by humans.
Alexander, M.M. "Food habits of the snapping turtle in Connecticut." Journal of Wildlife Management 7 (1943): 278-282.
Conant, R., and J.T. Collins. A Field Guide to Reptiles and Amphibians: Eastern and Central North America. Boston: Houghton Mifflin, 2001.
Congdon, J.D., G.L. Breitenbach, R.C. van Loben Sels, and D.W. Tinkle. "Reproduction and nesting ecology of snapping turtle (Chelydra serpentina) in southeastern Michigan." Herpetologica 43 (1987): 39-54.
Conner, C.A. "Descriptive Ecology of a Turtle Assemblage in an Urban Landscape." American Midland Naturalist 153.2 (2005): 428-435.
Froese, A.D., and G.M. Burghardt. "A dense natural population of the common snapping turtle (Chelydra s. serpentina)." Herpetologica 31 (1975): 204-208.
Froese, A.D., and G.M. Burghardt. "Food competition in captive juvenile snapping turtles, Chelydra serpentina." Animal Behavior 22 (1974): 735-740.
Galbraith, D.A. "Factors affecting the density of populations of common snapping turtles." Canadian Journal of Zoology (1988): 1233-1240.
Garber, S.D. "A 20-year Study Documenting the Relationship Between Turtle Decline and Human Recreation." Ecological Applications 5.4 (1995): 1151-1162.
Hulse, Arthur C. Amphibians and Reptiles of Pennsylvania and the Northeast. Grand Rapids: Comstock, 2001.
Lagler, K.F. "Food habits and economic relations of the turtles of Michigan with special reference to fish management." American Midland Naturalist 29 (1943): 257-312.
Major, P.D. "Density of snapping turtles, Chelydra serpentina, in western West Virginia." Herpetologica 31 (1975): 332-335.
Minton, S.A. "The Fate of Amphibians and Reptiles in a Suburban Area." Journal of Herpetology 2.3/4 (1968): 113-116.
Obbard, M.E. "A radiotelemetry and mark-recapture study of activity in the common snapping turtle, Chelydra serpentina." Copeia 1981: 630-637.
Petokas, P.J., and M.M. Alexander. "The nesting of Chelydra serpentina in northern New York." Journal of Herpetology 14 (1980): 239-244.