TY - JOUR
T1 - Relative importance of local and landscape variables on site occupancy by avian species in a pine forest, urban, and agriculture matrix
AU - Lee, Myung Bok
AU - Carroll, John P.
N1 - Funding Information:
The project was primarily funded by the Department of Defense, McIntire-Stennis Project GEOZ 136. We would like to thank the Georgia Ornithological Society and Warnell School of Forestry and Natural Resources for additional research grants. We thank Dr. Nate Nibbelink for providing valuable comments on analyzing spatial autocorrelation problems in our data and Dr. Shannon Albeke for developing R-code to calculate a spatial autocovariate. We also thank all staff, especially Robert Drumm and Paul Grimes, at the Fort Gordon Natural Resources Branch for their help and support, field technicians for their hard work, and land owners for allowing us to access to sites under their ownership or management for this study.
PY - 2014/5/15
Y1 - 2014/5/15
N2 - In the Southeastern USA, planted pine forests are increasingly embedded in a matrix of urban and/or agricultural land. Most avian conservation management in these forests has primarily focused on local (stand) characteristics and less attention has been given to landscape characteristics. We investigated the relative influences of local and landscape variables on site occupancy of 17 avian species in pine forests in urban and agricultural matrices. During May-August 2010 and 2011, we conducted bird surveys and vegetation surveys in stands dominated by loblolly pine (Pinus taeda) in the Sandhills Ecoregion of Georgia, USA. We developed 8 single-season occupancy models. An autocovariate was also incorporated to account for spatial autocorrelation. We used three principal components summarizing environmental characteristics at each of three spatial extents (one local and two landscape scales, 500. m-radius circle and 1. km-radius circle surrounding a point) as covariates. Of 17 species, occupancy by 5 species was associated with local vegetation variables and by 8 species with variables at 500. m and/or 1. km landscape scale. Occupancy by forest interior species had a greater association with landscape variables, whereas occupancy by pine-grassland species was related to local variables. Urban development and agricultural land use positively influenced the occupancy of species associated with landscape variables. To improve overall avian diversity, our results suggest that forest management needs to consider both local vegetation characteristics and landscape characteristics, and that the potential habitat value of pine forests in the urban/agricultural matrix should not be ignored.
AB - In the Southeastern USA, planted pine forests are increasingly embedded in a matrix of urban and/or agricultural land. Most avian conservation management in these forests has primarily focused on local (stand) characteristics and less attention has been given to landscape characteristics. We investigated the relative influences of local and landscape variables on site occupancy of 17 avian species in pine forests in urban and agricultural matrices. During May-August 2010 and 2011, we conducted bird surveys and vegetation surveys in stands dominated by loblolly pine (Pinus taeda) in the Sandhills Ecoregion of Georgia, USA. We developed 8 single-season occupancy models. An autocovariate was also incorporated to account for spatial autocorrelation. We used three principal components summarizing environmental characteristics at each of three spatial extents (one local and two landscape scales, 500. m-radius circle and 1. km-radius circle surrounding a point) as covariates. Of 17 species, occupancy by 5 species was associated with local vegetation variables and by 8 species with variables at 500. m and/or 1. km landscape scale. Occupancy by forest interior species had a greater association with landscape variables, whereas occupancy by pine-grassland species was related to local variables. Urban development and agricultural land use positively influenced the occupancy of species associated with landscape variables. To improve overall avian diversity, our results suggest that forest management needs to consider both local vegetation characteristics and landscape characteristics, and that the potential habitat value of pine forests in the urban/agricultural matrix should not be ignored.
KW - Occupancy model
KW - Pine-grassland species
KW - Spatial autocorrelation
KW - Spatial scale
KW - Urban-rural/agriculture-wildland
UR - http://www.scopus.com/inward/record.url?scp=84898771705&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84898771705&partnerID=8YFLogxK
U2 - 10.1016/j.foreco.2014.03.017
DO - 10.1016/j.foreco.2014.03.017
M3 - Article
AN - SCOPUS:84898771705
SN - 0378-1127
VL - 320
SP - 161
EP - 170
JO - Forest Ecology and Management
JF - Forest Ecology and Management
ER -