Encounter probability analysis of injured birds in Iowa: A spatial perspective
Topics: Geographic Information Science and Systems
, Animal Geographies
, Middle America
Keywords: GIS, birds, Iowa, analysis, rehabilitation, injury, roads, land cover, socioeconomic
Session Type: Virtual Poster
Day: Friday
Session Start / End Time: 4/9/2021 09:35 AM (Pacific Time (US & Canada)) - 4/9/2021 10:50 AM (Pacific Time (US & Canada))
Room: Virtual 52
Authors:
Grace Trenkamp, University of Wisconsin
L. Lynnette Dornak, University of Wisconsin-Platteville Geography Department
J. Huebschman, University of Wisconsin-Platteville
R.J. Haasl, University of Wisconsin-Platteville
,
,
,
,
,
,
Abstract
The Raptor Advocacy, Rehabilitation, and Education (R.A.R.E.) Group, located in Iowa City, Iowa, accepted 37 injured species (236 cases) to its facility in 2018, 13 of these species (97 cases) were raptors. Birds varied in injury severity and cause, age, rehabilitation time, and outcome (i.e., release or euthanasia). Although not every bird is rehabilitated, it is unlikely that an injured bird survives very long in the wild. Thus, understanding the factors associated with the discoverability of injured birds will better support avian conservation and education. The purpose of this study was to identify environmental (e.g., land cover type) and socioeconomic factors (e.g., average household size) associated with the probability of injured birds being encountered. We characterized land cover and socioeconomic landscape in 200-meter buffers surrounding each location where birds were found and for 145 randomly selected locations (i.e., for pseudoabsence comparisons) within our study area. We used three logistic regression techniques to predict found locations from 16 environmental and socioeconomic variables: basic logistic regression, penalized logistic (lasso) regression, and stepwise regression. Six variables had coefficients that occurred in two or more of the three regression models being either significant or non-zero (percent bare land cover (+/-), percent road land cover (+), percent grassland cover (+/-), percent agriculture land cover (-), percent tree land cover (+/-), and mean number of housing units(+). Currently, additional data for 2017 and 2019 are being added to this study to strengthen these relationships and determine if these variables are consistent through time.
