You are not currently logged in.
Access your personal account or get JSTOR access through your library or other institution:
Characterizing Independence of Observations in Movements of Columbian Black-Tailed Deer
R. Scott McNay, Jeffrey A. Morgan and Fred L. Bunnell
The Journal of Wildlife Management
Vol. 58, No. 3 (Jul., 1994), pp. 422-429
Stable URL: http://www.jstor.org/stable/3809312
Page Count: 8
Preview not available
Lack of independence in observations of animal locations and movements can cause underestimates of home-range sizes and may lead to inappropriate interpretations of temporal use of space. We used Schoener's Ratio (1981) to assess independence of observations in movements of black-tailed deer (Odocoileus hemionus columbianus). We examined the time interval between samples at which independence would occur with, and without, migratory movements, using 12,510 locations from 44 resident and 28 migratory deer. Most datasets contained dependent, or redundant, observations. Even with a 6-week interval between samples (i.e., 8 samples/yr), observations were still dependent for >50% of the deer tested. We found similar results when the data tested represented distance between consecutive locations rather than the locations themselves. In each dataset lacking independence, results were caused by migrations or by infrequent moves to unique places in the home range, both of which made the more common moves comparatively redundant. Such movements skewed data distributions and violated the assumption of normality in the independence test making it difficult to determine a time interval that would compensate for distribution problems. Because most animal location datasets are likely to have skewed data distributions, especially for those animals that migrate, we recommend placing emphasis on sampling animal locations systematically through time rather than trying to determine a time interval that will provide independent location samples. Time intervals between locations should be chosen with the understanding that potential gains in behavioral information are decreased with increasing time intervals between samples.
The Journal of Wildlife Management © 1994 Wiley