Estimating bird density and detection probability at five national park units in southern Oregon and northern California

Stephens, Jamie; Mohren, Sean; Barton, Daniel; Alexander, John; Sarr, Daniel

As part of the national Inventory and Monitoring Program of the National Park Service, the Klamath Inventory and Monitoring Network and Klamath Bird Observatory established a protocol to monitor long-term status and trends in landbird populations at National Park units in southern Oregon and northern California. This report examines detection probabilities and estimates densities using data from the first sequence of surveys in the five larger park units in the Network: Crater Lake National Park, Lassen Volcanic National Park, Lava Beds National Monument, Redwood National and State Parks, and Whiskeytown National Recreation Area. At each park, 25–35 sites were established using a stratified random approach. Breeding season point counts were conducted at each park from 2008–2010, along with the collection of vegetation data. For this analysis we examined factors related to detection probability using distance sampling. A number of factors were related to detection probability, including observer and type of detection. Effective distance radii varied among species suggesting that counts may not be proportionate based on a fixed radius. On the other hand, detection models fit the data poorly for some species, which showed evidence of observer avoidance or heaping, especially for song detections. Density estimates for all detections were highly correlated with song-based density estimates, and differences between the two estimates were related to the proportion of song detections. Model assumptions for distance sampling may not always be met, and our results suggest density estimates from this single year of surveys should be interpreted with caution. In the future, approaches of combining detections across years or park units may provide more robust estimates. Indices such as relative abundance remain useful in some contexts, and comparisons are needed with independently derived measures of density from spot-mapping. These results highlight the importance of considering differences among species in detection type and detection distances, as well as designing sampling to minimize observer effects.

Estimating bird density and detection probability at five national park units in southern Oregon and northern California

Series

Collections