Document: JUA-3-65-23

Scaling-up movements in heterogeneous landscapes: Importance of behavior

MANUEL MORALES, J.* ^{1, 2}

University of Connecticut, Storrs, CT 06269 USA ¹
LIEY ²

Abstract:
Movement of organisms within and across landscapes is a key process. Correlated random walks models (CRW) have been successful in modeling movement at the scale at which the data were collected. To examine whether extrapolation to bigger scales in heterogeneous landscapes is possible, I videotaped Tribolium confusum beetles moving in 'fast' and 'slow' experimental patches of 25 by 25 cm (small scale). These patches were constructed with paper and masking tape, and differed only by their physical structure. Observed turning angle and step length distributions were used in a spatially explicit, individual based model. The spatial position of an individual was represented by continuous x and y coordinates which were projected onto a landscape map. At every iteration of the model, individuals chose a step length and a turning angle according to the patch type in which they were located. Edge effects where incorporated as fixed probabilities of crossing a boundary between different patch types. Model iterations reflected observed time scale of beetle movement behavior. I compared simulated mean squared displacement over time to that observed for real beetles in an independent set of mosaic micro-landscape experiments of 4 by 4 meters (large scale). Only the most general prediction of slower spread in landscapes with slow patches and with edge effects held for the landscape experiments. Correlated random walk models parameterized at single patch scale under-predicted larger scale spread. These failures occurred because moving organisms violated the assumption of constant movement behavior. However, random walk simulations were flexible enough to accommodate changes in turning angle distributions with time and produce good fits to observed data. Beetle turning angle distributions depended not only on the patch type they occupied but also on how much time they spent in the experimental landscape. These experimental results illustrate the difficulties of extrapolating movement observations across temporal and spatial scales. Within the same type of patches, and depending on their physiological state and/or time since the beginning of dispersal, organisms could alternate between a thorough exploration (with high variance in turning angles) and escape from an area that is not suitable for them.

Keywords: movement, landscape experiment, scale-up, edge, Tribolium, heterogeneity, correlated random walk

This abstract is being presented at: 9:30 AM in session:
Oral Session #56: Metapopulation Analysis.