A REANALYSIS OF FOOTPRINTS AND TRACKWAYS AT THE DINOSAUR STATE PARK MEGATRACKSITE USING BASIC STATISTICAL METHODS

Dinosaur footprints and trackways are routinely measured to infer trackmaker taxonomy, behavior, size, and speed. Utilizing published trackway orientations and new measurements of footprint length, width, interdigital angles, and stride for the twelve best-preserved trackways at the Dinosaur State Park megatracksite (DSP), we applied basic statistical methods to test some of the questions commonly addressed using trackway data. Previous authors concluded that there is no preferred orientation to the trackways at DSP. Trackway azimuths were tested for significant unidirectional and bidirectional orientation against the null hypothesis of random orientation using the circular normal (von Mises) distribution. The null hypothesis of random orientation could not be rejected, lending statistical support to the hypothesis that the trackmakers at DSP were moving in random directions. Comparing the mean footprint length and width of each the twelve measured trackways using analysis of variance (ANOVA) reveals that five of the twelve trackways are statistically indistinguishable and could have been made by the same individual. Two other trackways are marginally similar to these five, while the remaining five are significantly different in average footprint size. These results caution against assuming that each trackway at a megatracksite is the product of a unique individual trackmaker even when mean footprint sizes are different. Measurements of interdigital angles II-III and III-IV were found to be highly variable within individual trackways. ANOVA tests on interdigital angles failed to find any consistent significant differences between trackways, lending some support to the hypothesis that all trackways were made by the same species of dinosaur, but also casting doubt on the utility of such a variable trait for characterizing trackmakers. Estimates of trackmaker speed calculated using published methods based on average footprint length and stride length indicate all twelve trackmakers walking at speeds between 1.2 and 1.8 meters/sec (app. 3-4 MPH). Calculations using the upper and lower bounds on the 95% confidence intervals around mean footprint length for each trackway suggest that measurement error associated with footprint length affects estimates of trackmaker speed by less than +/- 10%.