Date Thesis Awarded


Access Type

Honors Thesis -- Access Restricted On-Campus Only

Degree Name

Bachelors of Arts (BA)


Environmental Science and Policy


Matthias Leu

Committee Members

Daniel Cristol

John Swaddle

Brent Kaup


Conservation efforts in the United States primarily focus on the protection of public lands, but despite the large quantity of protected area, species endangerment is rising at an alarming rate. Additional efforts need to address conservation on private lands to protect a greater variety of species and landscapes. Prior research has evaluated the consequences of land cover fragmentation on biodiversity, but the problem of parcelization of private land has been studied far less. I related avian guild richness with parcel area, parcel shape complexity, and distance to protected land to investigate the associations between species diversity and private property. I utilized parcel data from 19 randomly selected counties in Pennsylvania, which represent a broad range of forested and urban landscapes. I assessed the parcel variables at three home range sizes in univariate models for each guild to incorporate the biological response of the guilds to the landscape. I then evaluated all possible combinations of the parcel area and parcel shape complexity variables at the best home range size with distance to protected land. I identified the top models on the basis of the information-theoretic approach and used conditional model averaging to estimate the effect size of the important variables. Parcel area related positively with 10 of the 15 guilds, and distance to public land associated negatively with 11 guilds. Overall, parcel shape complexity was not an important variable for explaining the spatial variation of species richness. These results indicate that conservation of avian guild richness requires joint land management between owners of public and private property. My research is a preliminary step in understanding the impact that human delineation of the landscape has on species richness and should be evaluated with additional data.

On-Campus Access Only