Mourning doves (Zenaida macroura) are among the most abundant and harvested game birds in North America. As such, their population abundance and vital rates are annually monitored by the U.S. Fish and Wildlife Service in cooperation with state agencies. Current monitoring indicates a decline in absolute abundance across a large portion of its range in the United States, raising concerns. One theory for this apparent decline is problems with the data used to estimate these vital rates: specifically, biases in the data collection methods (including banding programs) not being representative of the overall population and harvest rates. Therefore, to assess one potential bias in our banding programs, we investigated whether band recoveries of mourning doves in North Carolina and South Carolina are affected by the proportion of urban landscape around banding sites. We hypothesized that there would be a negative linear association between proportions of urban area (i.e., developed areas as defined by the National Land Cover Database) around banding sites and proportion of dove band recoveries in the Carolinas. We used geographic information systems and land cover classifications from the National Land Cover Database to extract 5-km extents around banding sites and determined proportions of urban landscape within each 5-km extent. We used simple linear regression of the proportion of urban landscape associated with each site and the proportion of banded doves harvested per site. The regression lines did not differ significantly from horizontal and the relationships between variables were weak. From an urban vs. rural harvest perspective, banding as currently implemented in North Carolina and South Carolina likely provide an unbiased sample based on level of urbanization (on a limited continuum) at these sites.
