Open AccessInferring detailed space use from movement paths: A unifying, residence time‐based framework
Author(s) -
Kapota Dror,
Dolev Amit,
Saltz David
Publication year - 2017
Publication title -
ecology and evolution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.17
H-Index - 63
ISSN - 2045-7758
DOI - 10.1002/ece3.3321
Subject(s) - residence , resource (disambiguation) , computer science , scale (ratio) , movement (music) , path (computing) , variance (accounting) , residence time (fluid dynamics) , homogeneous , duration (music) , data mining , artificial intelligence , geography , cartography , mathematics , geology , computer network , philosophy , demography , accounting , geotechnical engineering , combinatorics , sociology , programming language , aesthetics , art , literature , business
The residence time is the amount of time spent within a predefined circle surrounding each point along the movement path of an animal, reflecting its response to resource availability/quality. Two main residence time‐based methods exist in the literature: (1) The variance of residence times along the path plotted against the radius of the circle was suggested to indicate the scale at which the animal perceives its resources; and (2) segments of the path with homogeneous residence times were suggested to indicate distinct behavioral modes, at a certain scale. Here, we modify and integrate these two methods to one framework with two steps of analysis: (1) identifying several distinct, nested scales of area‐restricted search ( ARS ), providing an indication of how animals view complex resource landscapes, and also the resolutions at which the analysis should proceed; and (2) identifying places which the animal revisits multiple times and performs ARS ; for these, we extract two scale‐dependent statistical measures—the mean visit duration and the number of revisits in each place. The association between these measures is suggested as a signature of how animals utilize different habitats or resource types. The framework is validated through computer simulations combining different movement strategies and resource maps. We suggest that the framework provides information that is especially relevant when interpreting movement data in light of optimal behavior models, and which would have remained uncovered by either coarser or finer analyses.