Abstract

Animal movement tracks emerge from sequences of fundamental movement elements (FuMEs). The elements of a FuME time series are currently, exceptionally difficult to parse out: rather, we only observe a relocation points time series from which we can extract step length (SL) and turning angle (TA) time series. These time series can then be used to compute various derived quantities, such as radial and tangential velocities at each relocation point or auto correlations of variables along the  the movement track. The statistics of such variables, computed for fixed short segments of track (e.g., 10-30 points), can then be used to categorize such segments into statistical movement elements called StaMEs (previously called metaFuMEs).  These StaMEs then from the basis for constructing corresponding step-selection kernels. Such kernels can be used in simulation models to generate the tracks of canonical activity modes (CAMs: various length sequences of the same StaMEs), and even larger diel activity routines (DAR: a mixed string of CAMs over a 24 hour cycle). Here we present a multi-mode, a priori defined, step-selection kernel simulator (Numerus ANIMOVER 1 ), built using Numerus RAMP technology. We discuss methods for selecting kernel parameters that generate StaMEs comparable to those observed in empirical data, thereby providing a means for simulating the possible response of animal movement to landscape changes  We illustrate our methods for extracting StaMEs from both simulated and real data—-in our case the latter is movement data  of barn owls (Tyto alba) in the Hula Valley, Israel.  The methods described here provide a way to fit fine scale simulation data to movement models that can be used to predict changes in the movement behavior of animals under global change.