For nine weeks during the summer of 2002, the movement of stinkpot turtles (Sternotherus odoratus) in Lake Matoaka was monitored as part of a behavioral study. In this study, we hoped to determine whether or not stinkpot turtles, like other turtles, exhibit homing behavior. If the turtles did, in fact, display homing behavior, we wanted to test what factors influenced successful movements. For example, would depth, distance, or a combination of both affect the turtles' homing abilities?
To monitor the turtles' movement about Lake Matoaka, we placed three clusters of crabpots along the lake's shore. Two of these clusters (the "Ice House Cove" site and the "North Keck" site) were separated by a considerable distance (~500m) along the lake's eastern shore. The third cluster was situated on the western shore, separated from either site by water that reached a depth of ~4m. Each cluster contained three unbaited crabpots that would trap the turtle once it crawled through a one-way opening. Every day, my assistants and I checked each crabpot for stinkpots. Captured turtles were marked with a binary code on their carapace. This code was used to track the movement of individual turtles between traps throughout the study. For the first three weeks, turtles were displaced using a rotating release pattern. However, during the last six weeks of the study, turtles were simply released at the point of capture. As a result, we were able to monitor the homing behavior of displaced turtles relative to other movements in non-displaced turtles.
The results of this nine week study are as follows:
* 14 movements of displaced turtles over a considerable distance (~500m)
* 13 of these movements were homing - returned to site of original capture
* 9 displaced turtles displaying homing behavior over a considerable distance
*13 movements of displaced turtles across deep water (~4m)
*11 of these movements were homing - returned to site of original capture
*9 displaced turtles displaying homing behavior across deep water
* 12 movements of displaced turtles across both deep water and considerable distance
*10 of these movements were homing - returned to site of original capture
*9 displaced turtles displaying homing behavior across both deep water and considerable distance.
*4 random movements from non-displaced turtles
*316 total captures *197 different stinkpot turtles
*90 new stinkpots labeled during study
*average of 9.55 days between movements.
Based on the almost equal number of movements between the three clusters, neither distance, deep water, nor a combination of both elements deterred turtle movement. Twenty seven of 178 displaced turtles (15%) displayed homing behavior by returning to sites of original capture, and only five moved to non-home recapture sites, yielding a signicant difference between homing and non-homing movements by displaced turtles (chi-square = 15.1, p < 0.005). None of the 138 non-displaced turtles captured during the study, however, moved from one site to that of an earlier capture, and only three moved from any site of capture to another site. These results indicate that homing tendencies were responsible for the movement patterns observed in stinkpot turtles.
For additional documentation Christina Smar provided a Powerpoint Presentation entitled, "Homing Behavior in Stinkpot Turtles," provided here in PDF form.