A comparison of metrics predicting landscape connectivity for a highly interactive species along an urban gradient in Colorado, USA

摘要: Many organisms persist in fragmented habitat where movement between patches is essential for long-term demographic and genetic stability. In the absence of direct observation movement, connectivity or isolation metrics are useful to characterize potential patch-level connectivity. However, multiple exist at varying levels complexity, empirical data on species distribution rarely used compare performance metrics. We compared 12 degrees complexity determine which metric best predicts prairie dog colonies along an urban gradient 385 isolated Denver, Colorado, USA. found that a modified version incidence function model including area-weighting cost-weighted distance surface predicted occupancy, we assumed roads were fairly impermeable low-lying drainages provided dispersal corridors. also this result be robust range cost weight parameters. Our results suggest should incorporate both patch area composition surrounding matrix. These provide guidance improved landscape modeling landscapes can help identify target conservation management dogs systems.