Evaluating bias in roadside surveys of secretive marsh birds in the Prairie Pothole Region, MN, ND, and SD 2008-2009

Publication Date

2021-08-11

Start Date

2008-05-06

End Date

2009-07-10

Buhl, T.K., Shaffer, T.L., Niemuth, N.D., and Granfors, D.A., 2021, Evaluating bias in roadside surveys of secretive marsh birds in the Prairie Pothole Region, MN, ND, and SD 2008-2009: U.S. Geological Survey data release, https://doi.org/10.5066/P9RCEIHF.

Population size and trends of most North American marsh bird species are poorly known, and continental marsh bird surveys were proposed to provide better insight into marsh bird populations. The Prairie Pothole Region (PPR) is a critical and expansive breeding area for several focal marsh bird species. Roadside point-count surveys were considered as a technique that would allow widespread areas within the PPR to be surveyed during the short crepuscular period during which these species can be detected, and would ease logistical constraints of gaining access to private lands and navigating potentially hazardous areas during periods of darkness. Counts from roadsides may differ from counts in non-roadside habitats either because densities differ between roadside and non-roadside habitats or because individual marsh birds in roadside habitats may be more (or less) detectable from the road than from an off-road location. We designed an experiment to investigate these questions. We followed a standardized protocol and conducted call-playback surveys on pairs of wetlands in 2008 and 2009 in Minnesota, North Dakota, and South Dakota. We fit occupancy models to estimate detection and occupancy probability in relation to whether the survey occurred in roadside versus non-roadside habitat, and whether detection probability was enhanced or depressed when surveying roadside habitat from the road itself. Models in which occupancy probability varied in proportion to distance between the survey point and the road were also considered.

• Northern Prairie Wildlife Research Center