Author(s): M. Rice | W. Ballard | E. Fish | N. McIntyre | D. Holdermann
Journal: Wildlife Biology in Practice
ISSN 1646-1509
Volume: 4;
Issue: 2;
Start page: 48;
Date: 2008;
Original page
Keywords: black bear | GIS | Habitat selection | landcover/landuse | Ursus americanus
ABSTRACT
Landcover/landuse layers can be used to map species’ distributions as a potential indicator of habitat selection. We used this approach to assess habitat selection by a small population of black bears (Ursus americanus) in the Trans-Pecos of Texas. We compared the accuracy of the Gap Analysis Program (GAP) landcover/landuse layer and a second derived landcover/landuse layer (derived) from LANDSAT images. We then determined habitat selection using sighting data collected from 1901-2003 for both landcover/landuse layers. The derived landcover/landuse layer was 40% more accurate than the GAP layer and there was only 44% agreement between the classifications from the two layers. The discrepancy resulted in different landocover/landuse selections and different predictions of black bear occurrence depending on which layer was used. We recommend managers use derived landcover/landuse layers along with sighting databases to aid management of population when little data on species distribution and habitat use are available.
Journal: Wildlife Biology in Practice
ISSN 1646-1509
Volume: 4;
Issue: 2;
Start page: 48;
Date: 2008;
Original page
Keywords: black bear | GIS | Habitat selection | landcover/landuse | Ursus americanus
ABSTRACT
Landcover/landuse layers can be used to map species’ distributions as a potential indicator of habitat selection. We used this approach to assess habitat selection by a small population of black bears (Ursus americanus) in the Trans-Pecos of Texas. We compared the accuracy of the Gap Analysis Program (GAP) landcover/landuse layer and a second derived landcover/landuse layer (derived) from LANDSAT images. We then determined habitat selection using sighting data collected from 1901-2003 for both landcover/landuse layers. The derived landcover/landuse layer was 40% more accurate than the GAP layer and there was only 44% agreement between the classifications from the two layers. The discrepancy resulted in different landocover/landuse selections and different predictions of black bear occurrence depending on which layer was used. We recommend managers use derived landcover/landuse layers along with sighting databases to aid management of population when little data on species distribution and habitat use are available.
