TURTLES DEMONSTRATE THE IDEAL FREE DISTRIBUTION BY DISTRIBUTING TO MAXIMIZE FOOD CONSUMPTION By: Turtle-Tastic Task Force Jiyansh Agarwal Zahria Davis Sofia Diaz David Lopez Bianca Manzanares Gabriel Placido Varsha Udayakumar Nicole Verdecia ABSTRACT The ideal free distribution theory describes an optimal foraging strategy in which animals freely distribute themselves to obtain the most food. We fed turtles high, medium, and low amounts of food each minute and counted the number of turtles attracted to each food level. In six trials, a small number of turtles were attracted to the low food level, an intermediate number of turtles were attracted to the medium food level, and a large number of turtles were attracted to the high food level. Our results show that the turtles are capable of forming an ideal free distribution in response to food. INTRODUCTION All animals need food, but food is not abundant everywhere. Animals have to forage for food and try to get the largest amount of food. The foraging decisions that animals make affect their survival, reproduction, etc. The ideal free distribution (IFD) is one theory that describes how animals will distribute themselves according to the resources around them to maximize the amount of resources that they get. According to the theory most animals will go to the largest amount of a resource and fewer animals will go to a small amount of the same resource. IFD has
been validated using guppies (Abrahams 1989), cichlids (Godin and Keenleyside 1984), female black bears (Mitchell and Powell 2007), some insects (Kelly and Thompson 2000), female roe deer (Walhström and Kjellander 1995), and many other species. Our goal is to test the IFD on turtles. We used turtles because they are not dangerous, they can move freely between the different amounts of food, and because they are convenient to work with. We want to know if turtles can freely distribute themselves among various amounts of food to obtain the most food. To test this we fed turtles different amounts of food and observed the number of turtles at each feeding station. Our hypothesis was: if food is thrown to turtles in different positions and in different amounts then the turtles will distribute themselves to maximize their food intake. Our hypothesis predicts that the high food level would attract the greatest number of turtles, the medium food level will attract an intermidiate number of turtles, and the small food level would attract the least number of turtles. METHODS Our experiment was performed at Florida International University, University Park campus, during the morning hours on the 14 th, 19 th and 21 st of June. The weather was either sunny or partly cloudy. We selected the pond located near the Dorothea & Steven Green library because it had an abundance of turtles (see figure 1). The pond was a large, dark blue pond with a fountain in the center and a bridge dividing it into two parts. Among the turtles that participated, we identified four types of turtles: The red eared slider, the Florida redbelly slider, the Florida soft shelled turtle and the Florida cooter (see figure 2).
Figure 1. Map of FIU Campus. To establish three levels for the amounts of food we formed three teams. Each team had a different size spoon (tablespoon, half tablespoon and half teaspoon) to measure the amount of food (Hikari Gold Koi fish food). Food was delivered by one person in the group using these spoons every minute for 16 minutes. The other person in each group counted the turtles in front of each feeding station 10 seconds after food was thrown. A timekeeper who was not in any group would tell when to throw food and when to count turtles. Groups were spread out along the shoreline about 2 meters away from each other. Spoon size and position (right, middle and left) were chosen randomly for all six trials. The number of turtles counted each minute was entered in Excel, and line graphs were used to help visualize our data. The mean and standard deviation for number of turtles counted
every minute per trial were calculated, and t-tests were used to find differences between the respective means. Red-eared slider Trachemys scripta elegans Florida Redbelly slider Pseudemys nelsoni Florida Cooter Pseudemys floridana Florida Soft-shelled Apalone ferox Figure 2. These are the turtles we identified. RESULTS Our experiment had a good turn-out of turtles. On average, we had 32 turtles participating in our experiment at any time with a maximum of 49 turtles and a minimum of 10 turtles. We observed in all trials that the numbers of turtles that were attracted to each food level were about the same amount at the beginning of each trial, then as time passed the numbers of turtles that were attracted to each food level spread out until at the end of each trial there were larger
differences between the amounts of turtles that were attracted to each food level (see figure 3 and figure 4). Over all times and trials, about 6 turtles were attracted to the low food level per minute, 10 were attracted to the medium food level and 16 were attracted to the high food level (see figure 4). Due to the fact that the turtles were free to move between feeding stations, an increase in the number of turtles at one food level leads to a decrease in the number of turtles in another food level (see figure 5). The number of turtles in the high food level increase to a maximum at around the 8th minute in each trial (see figure 4 and 5). In two of the three trials when the low food level station was situated between the medium and high food level stations, we observed higher variance in the number of turtles each minute (see figure 5). In both trials 3 and 5, the number of turtles that were attracted to the low food level suddenly spiked above or almost equal to the amounts of turtles attracted to the medium and the high food levels (see figure 5). This is likely the result of the turtles moving between the medium and high food level stations, stopping for some food at the low food level, and being counted. Except for these trials, each trial ended with the high food level having the largest number of turtles attracted to it, the medium food level with a medium number of turtles, and the low food level with the smallest number of turtles (see figure 5). All trials started with the low food level having a higher number of turtles attracted to it than the medium food level except for the first trial (see figure 5). We did a t-test to compare the differences between the numbers of turtles attracted to each of the food levels. The t-test revealed that the high food level attracted more turtles than the medium food level (t-test: p<0.001), and the medium food level attracted more turtles than the low food level (t-test: p<0.001) (see figure 4). The high food level attracted the greatest number of turtles, and the low food level attracted the least number of turtles. These results validated the hypothesis of the experiment.
Figure 3. The total number of turtles summed for all trials at each time for each food level (Number of trials: 6).
Figure 4. The mean number of turtles per minute for each food level. Error bars represent ± one standard deviation. Results of t-tests indicate significant difference between the three food level treatments (p<0.001).
Figure 5. Number of turtles per minute for each food level at each trial.
DISCUSSION In ecology, an ideal free distribution (IFD) is a way in which animals distribute themselves among several patches of resources, in which they all get equal amounts of food. Are turtles capable of forming an IFD to spread themselves out among the patches of food? We carried out a series of trials, and the results indicate that the turtles are capable of forming an IFD. Our results showed that the different food levels attracted different numbers of turtles on average, supporting our alternative hypothesis. One of the many things we noticed was that there was a large amount of variability. The numbers of turtles at a given site were not constant, as displayed in the results above. There were some factors that could have disturbed the experiment such as fish interferences, intimidation by other turtles, or fighting. Size differences, weather, and water movement may have limited the ability of turtles to form an IFD though we have no proof of this. The importance of this experiment is that it helped us to determine whether turtles are capable of forming an IFD, and whether it could be used as a tactic that helps them get food. The IFD applies to the real world because turtles can be used as a study organism for other animals. This can be used to suggest that other animals have the possibility to distribute themselves among resources. The turtles seemed to acknowledge that more turtles are gathered in one area than another, so that may have attracted them toward that particular site. Future Reference: If we could alter this experiment, we propose trying different food brands to see which ones the turtles prefer. Another idea would be to have a controlled water area (like a pool with a specific number of turtles). One experiment that interested us involved guppies being fed
different amounts of food at different places (Abrahams 1989). It was very similar to our experiment, but different because they monitored the movement of each gender. Their results showed that female guppies remained in one place and the male guppies moved from place to place. We did not classify whether the turtles are male or female, but we did observe that some turtles moved more than others. Overall, our experiment was successful in demonstrating that turtles can form an IFD. ACKNOWLEDGMENTS We would like to thank Dr. Don DeAngelis, Mr. Robert McElderry, Andre Naranjo, FIU, Dr. Michael Gaines, Dr. Dana Krempels, Research in Ecology, and the Howard Hughes Medical Institute. LITERATURE CITED Abrahams, M.V., 1989, Foraging Guppies and the Ideal Free Distribution: The Influence of Information on Path Choice, Ethology, 82, 116-126 Godin, J.G.J.; Keenleyside, M.H.A., 1984, Foraging On Patchily Distributed Prey by a Cichlid Fish (Telostel, Cichlidae): A Test of the Ideal Free Distribution Theory, Animal Behavior, 32, 120-131 Kelly, D.W. ; Thompson, C.E., 2000, Parasitology, Epidemiology and Optimal Foraging Modeling the Ideal Free Distribution of Insects Vectors, Volume 120, 319-327 Mitchell, M.S.; Powell, R.A., 2007, Optimal Use of Resources Structures Home Ranges and Spatial Distribution, Science Direct, Animal Behavior, 219-230
Wahlström, L.K.; Kjellander, P., 1995, Ideal Free Distribution and Natal Dispersal in Female Roe Deer, Oecologia, 103, 302 308.