Testing the Ideal Free Distribution on Turtles in the Field

Transcription

1 Testing the Ideal Free Distribution on Turtles in the Field Justin Carasa Nicole Cinquino Christopher Contreras Santiago Londoño Michelle Ortiz Andrea Remiro Alexander Rodriguez Research in Ecology University of Miami Summer 2012

2 Abstract We performed experiments to test the Ideal Free Distribution theory for how animals distribute themselves. We manipulated food quantities in a pond to see how resident turtles would distribute themselves in relation to amounts of food. This tested the hypothesis that the turtles would distribute themselves ideally. The data from four feeding experiments strongly support our hypothesis. We discuss some of the problems encountered in the experiments and make suggestions for future studies. Introduction The ideal free distribution (IFD) is a theory that describes how animals space themselves freely to obtain the maximum amount of food. According to the IFD, if animals have a free choice of where to feed, they will distribute themselves so that each gets as much food as possible. The IFD has been supported by numerous experiments with a variety of animals. Fretwell and Lucas (1970) were the first to introduce the theory of IFD. It is believed that many animals are intelligent enough to go to a place that has what they perceive to be the biggest source of food. The result of the choices of individual animals is that the largest source will have the most animals, the smallest source of food will have the least animals and intermediate sources of food will have intermediate numbers of animals. Even other simple animals like Goldfish have been tested (Sutherland, Townsend, and Patmore 1988), and the goldfish did as expected. Most of them went to the most amount of food. A little more than expected went to the least amount of food, and intermediate matched with the intermediate source. Black bears have been tested in Nevada (Beckmann and Berger 2003). The authors found that the black bears only partially supported the ideal free distribution concept. The larger, more dominant bears went to the bigger piles of food and scared off other smaller weaker bears suggesting that aggressive interaction might work against the IFD, which would instead lead to the ideal despotic distribution (IDD) theory. Ideal despotic distribution theory predicts that the quality of habitat controlled by territorial animals should vary depending on their competitive ability and the availability of resources. Other types of distribution are being tested such as the idea of both predators and prey distributing themselves freely (Kacelnik, Krebs, and Bernstein 1992) The purpose of our work was to further explore the IFD concept with turtles. We found it more testable in turtles than in other animals because some turtle species are less aggressive, so it would be easier to test the hypothesis. We were able to find a pond with a large number of turtles that were appropriate for testing by feeding experiments. The hypothesis was that the turtles would be able to tell the difference in the amounts of food at each feeding station and they would distribute themselves such that each would get the same amount of food. We tested it by having three feeding stations; from a large amount of food, to a medium amount of food, to a small

3 amount of food. The distribution of turtles was expected to be such that most turtles were at the feeding station with the largest amount of food, the fewest turtles would go to the feeding station with the least amount of food, and an intermediate number of turtles would go to the feeding station with the medium amount of food. This is to be expected because it is believed that animals are smart enough to distribute themselves accordingly. The null hypothesis is that the amount of food will not affect where they go to eat, and they won t be able to tell the difference. Materials & Methods This experiment was conducted mid morning June 15, 18, and 22, 2012, at Florida International University Modesto A. Maidique Campus in Miami, Florida. We selected the pond next to the library (N , W ) because it had a large number of turtles. There were four turtle species in the pond that participated in the experiment in some way. (See figure 1.) Red- Bellied turtle Spiny Softshell turtle Red- eared slider Peninsula cooter Pseudemys nelsoni Apalone spinifera Trachemys scripta elegans Pseudemys floridana peninsularis Figure 1. These are the turtles we tested. Three cups were filled with Pond Care Summer Staple Warm Water Koi and Goldfish Food. Each of the three groups of two students received one of three spoon sizes which were tablespoon, ½ tablespoon, and ½ teaspoon. Using a stopwatch the time keeper instructed the groups when to throw food and when to count turtles. Groups were instructed to throw food every minute, and all teams threw food directly in the pond in front of them using an underhand toss. After 20 seconds the time keeper said, Start count and each group counted the turtles that were in the area where they threw the food. Twenty seconds later the time keeper said, Stop count. The final count of the turtles was recorded in a field journal. This cycle was repeated 14 times concluding one 15-minute trial. For each trial we calculated the means and standard deviations for the number of turtles in each food level over time. Differences in the means were tested using the t-test. We composed line graphs in order to analyze if the distribution of turtles among the food levels changed over time. For the change in turtles over time at each food level we estimated a slope and an intercept, which were tested using t-test. All analyses were conducted using Microsoft Excel 2007and in the program R.

4 Results We present here the results averaged over the four trials. Figure 2 shows the average number of turtles at each feeding station over a fifteen minute period. The number of turtles at each food level changed over time, as the turtles sorted themselves out according to the amount of food at each station. For the first seven minutes, it took the turtles a while to distribute themselves ideally, but in the last eight minutes they clearly show a consistent agreement with the ideal free distribution. There was a significant change in the number of turtles over time that depends on the food level. There were more turtles at the medium food level in the beginning, but an average of one turtle left the medium level every two minutes while the high food level gained one turtle every minute (see table 1). When we averaged over fifteen minutes we observed differences in the numbers of turtles at the three feeding stations. However these differences were not all statistically significant (see figure 3). This is due to the large variance that is shown in the overlapping error bars of the high and medium food levels. We also considered the average over the last eight minutes, because we noticed that the turtles needed about seven minutes to sort themselves out. This reduces the variance and shows a statistical difference between all three food levels (see figure 4). Table 1. Results for the linear model fit for all four trials. Estimates of the slope and intercept for the number of turtles at each food level overtime. Food level Estimate DF t-value p-value high <0.05 intercept medium <0.001 low <0.01 high <0.001 slope medium <0.01 low >0.05

5 20 Average number of turtles over 6me Number of turtles Food Level High Medium Low Time (min) Figure 2. Line graph showing the average count of turtles at the different feeding stations over time. The blue line represents the high food level (tablespoon), the red line represents the medium level (half tablespoon), and the green line represents the low food level (half teaspoon). Number of Turtles Average feeding habits High Medium Low Food Level Figure 3. Bar graph showing the average number of the turtles at the three feeding stations. Averages include all fifteen minutes of the four trials.

6 Number of Turtles Averages over last 8 minutes High Medium Low Food Level Figure 4. Bar graph showing the average number of the turtles at the three feeding stations. Averages include only the last eight minutes of the four trials. Discussion and Conclusion Ideal free distribution describes the situation in which animals distribute themselves according to the amount of food available. The histograms in Figure 3 and 4 show that, overall, the turtles in our experiment were able to distribute themselves in rough proportion to the different amounts of food. The highest food level attracted the most turtles and the lowest food level attracted the fewest turtles on average over time in the trials. This supports the hypothesis of the ideal free distribution. Even though the turtles were able to distribute themselves close to ideally, they seemed confused at times concerning the distribution of foods. It took the turtles at least five minutes to distribute themselves according to the three levels of food. Overall, the lowest level of food attracted the fewest turtles and the highest level of food attracted the most turtles. All of the turtles seemed to distribute themselves ideally except for the soft shell turtles. We observed the soft shell turtles biting each other, chasing each other away from the food, and biting other turtles like the red bellied turtles. We noticed that when we were feeding the turtles from three different feeding stations only one soft shell turtle was in each feeding station. It was possible that the distribution of soft shell turtles follows the ideal despotic distribution. It was shown in black bears that territorial behavior leads to unequal distribution of bears among food sources (Beckman and Berger 2003).

7 The conditions under which we did our experiment created some problems. For example, the wind created interference for our project because it caused a drift in the water, which would blow the food down along the bank of the pond distorting the intended distribution of food. We can conclude that three of the four turtle species were capable of sensing the amount of food available and distributed themselves accordingly, which supports the ideal free distribution. Aggressive interactions seemed to affect the distribution of soft shell turtles more than the amount of food, which supports the ideal despotic distribution. Future studies might focus on how the aggressive interactions of soft shell turtles create differences in their distribution relative to the other turtle species. Acknowledgments We would like to thank Dr. Michael Gaines, Dr. Dana Krempels, and Ms. Daritza Berio-Blanco for leading this program, and we would also like to thank Dr. Donald DeAngelis, Robert McElderry, Blake Simmons, Ms. April Danese, and Ms. Madeline Serrano for helping us on the project. We also thank Howard Hughes Medical Institute, UM, and FIU for their support. Literature Cited Beckmann, J. P. and J. Berger Using black bears to test ideal-free distribution. Journal of Mammalogy. Volume 93. Fretwell, S. D. and H. L. Lucas On territorial behavior and other factors influencing habitat distribution in birds. Theoretical Develoment. Volume 1 Kacelnik A., J.R. Krebs, and C. Bernstein Habitat Selection and Predator prey Dynamics. Trends in Ecology and Evolution. Volume 7. Sutherland W.J., C.R. Townsend, and J.M. Patmore A test of the IFD with unequal competitors. Behavioral Ecology and social biology. Volume 23 Photo Credits er.html Herpetology/index.php?option=com_content&task=view&id=76&Itemid=26