COMMUNITY ECOLOGY OF A PRE-RESTORATION BROWNFIELD AND VOCALIZATION ACTIVITY AND DETECTABILITY OF GRASSLAND WINTER BIRD COMMUNITIES

Author

• Isaac Grosner, Southern Illinois University CarbondaleFollow

Date of Award

8-1-2026

Degree Name

Master of Science

Department

Forestry

First Advisor

Pease, Brent

Second Advisor

Eichholz, Michael

Abstract

Grasslands and prairies are one of the most imperiled land covers in Illinois, with an estimated 0.01% remaining of the historic distribution. Because of this, grassland restoration efforts are widespread and opportunistic, frequently targeting land with presumed low wildlife and economic value, such as brownfield sites. While it is often assumed that brownfields have little wildlife value, and thus prime candidates for grassland restoration, recent research has shown that some brownfields may harbor surprisingly large amounts of biodiversity and rare species, and that restoration of these sites may have unintended negative consequences on biodiversity. Therefore, it is important that those involved in brownfield restoration assess the flora and fauna that actively exist on the site, and how these species may be impacted by restoration efforts. To illustrate this, I conducted surveys for arthropods, birds, snakes, salamanders, and small mammals on a brownfield site (17.8 ha) slated for grassland restoration and for birds, snakes, and small mammals on restored grasslands of various sizes (5.5 - 145 ha) which served as reference sites in comparison to the brownfield. Year-round Hill numbers richness (qD) of all bird species and species of conservation concern was statistically comparable between all grassland sites and the brownfield, but Hill numbers richness of grassland-obligate species during the breeding season was statistically higher on all grassland sites compared to the brownfield. The brownfield had lower small mammal richness, but similar abundance, to restored grasslands, but richness and abundance of snakes was lowest at the brownfield. While restoring brownfields to grasslands may result in some localized loss of generalist wildlife, it may be necessary for grassland-dependent wildlife, the more imperiled group, to persist and potentially recover in historic areas. As a result of the widespread loss of grasslands is the steep decline in populations of grassland-obligate bird species. While studies of grassland birds during the breeding season are common, studies during the non-breeding season are sparse. Additional bird surveys during the non-breeding season could further our ecological understanding of winter bird communities and improve study design during the nonbreeding season. However, current knowledge on the detectability of grassland species during the nonbreeding season is lacking. Here, I present results from non-breeding season surveys conducted during December 2024 - March 2025 using Passive Acoustic Monitoring at 15 sampling units across three grassland sites in Southern Illinois. My objectives were to assess how temporal and climatic factors such as day of season and temperature impact vocal activity and detectability. I used generalized additive models to estimate detection probabilities and AIC to rank detection models for each species. I documented species vocal activity and detectability to be highest for most species at the beginning of the season and decrease as the season progresses. Additionally, increasing temperature has a positive effect on detection probability across species. These results suggest that most species can be effectively surveyed during the non-breeding season using PAM.