Measuring occupancy of breeding birds at a local scale in the Sacramento Valley using automated recorders

Link To Share This Poster: https://cdmcd.co/xvxB8Z
Live Discussion Link: https://cdmcd.co/93YGxz

Danielle T. Fradet and Andrew Engilis Jr., Wildlife, Fish, and Conservation Biology, University of California, Davis, CA, Brett Furnas, California Department of Fish and Wildlife

Background/Question/Methods
The Central Valley has undergone massive habitat change in the past century, creating a heterogeneous landscape. Many researchers lack the ability to undertake long-term biological monitoring on restoration projects and to measure long-term success, particularly on understudied private lands. Automated Recording Units (ARU) are a new monitoring methodology that have only been used in large-scale habitat studies. ARU can saturate a large landscape with time dependent surveys of birds; thus gathering more data on bird presence than traditional point counts. Other advantages include multiple reviewers of recordings to decrease bias and recorders collecting data simultaneously to increase standardization. This study tests if the ARU method can be applied at the site-specific scale and be used as a tool for long-term monitoring on restored sites.
To test site-level occupancy of breeding birds, 390, 5-minute recordings were taken in spring and summer of 2019 at Bird Haven Ranch (BHR) in Glenn County. BHR is a heterogeneous habitat of cultivated rice, managed wetlands, and riparian forest systems. Recordings were taken from 30 microsites, equally divided between the three habitat types. ARU were deployed for a minimum of 3 consecutive days per microsite, and 3 recordings were taken a day: 30 minutes before sunrise, at sunrise, and 30 minutes after sunrise. Three days of recordings were then interpreted from each site, totaling 270 interpretations.
Results/Conclusions
In total, 46 species were detected, including 29 passerines. Gamma diversity did not differ between marsh and riparian habitats but both were higher than rice; 33 species compared to 20 avian species. Alpha diversity had no statistically significant difference between the three habitat types. A detection history for use in occupancy estimation on BHR was constructed, which found no significant difference between habitat types, but did point to community differences. For example, Marsh Wren, Cistothorus palustris, had a higher occupancy in managed marsh habitat whereas Red-winged Blackbird, Agelaius phoeniceus, is least likely to be detected in riparian forest. The occupancy result comes from only 16 species meeting the threshold of detections necessary to fulfill the parameters of the model. Further refinement of the methodology through increasing microsite sampling could make ARU a viable option for long-term monitoring of biodiversity on restored lands, particularly because of this methodology’s advantages compared to traditional methods such as point counts. Together, these advantages and methodology refinement can help land managers better monitor avian diversity, aiding in making conservation decisions.