FRACTAL ANALYSES OF ANIMAL MOVEMENT - A CRITIQUE

Several recent papers developed and applied a novel approach for the a nalysis of animal movement paths, based on calculating the paths' frac tal dimensions. The estimated fractal dimension is used to describe th e pattern of the: interaction between animal movement and landscape he terogeneity, and possibly to extrapolate movement patterns of organism s across spatial scales. Here, I critically examine the key assumption of the fractal approach: that the estimated fractal dimension is cons tant over some biologically relevant range of spatial scales. Use of a correlated random walk as a null hypothesis for movement suggests tha r the fractal dimension should grade smoothly from near 1 at very smal l spatial scares to near 2 at very Large spatial scales. Several empir ical data sets exhibit a qualitative pattern in agreement with this pr ediction. I conclude that ecologists should avoid calculating and usin g the fractal dimension of movement paths, unless self-similarity (a c onstant fractal dimension) for some range of spatial scales is demonst rated. An alternative approach employing random-walk models provides a more powerful framework for translating individual movements in heter ogeneous space into spatial dynamics of populations.