Predictive Mapping of Yellow Rail (Coturnicops noveboracensis) Density and Abundance in the Western Boreal Forest via Ground and Satellite Remote Sensors

• Author / Creator

  McLeod, Logan JT

• The Yellow Rail (Coturnicops noveboracensis) is a small, secretive, wetland bird, which is apparently rare throughout most of its range. Almost nothing is known about its abundance and density in the wetlands of the western boreal forest. Emerging technologies have enabled us to effectively survey for Yellow Rail in remote wetlands by using ground-based remote sensors (autonomous recording units; ARUs) to conduct passive acoustic monitoring. This technique was employed to survey Yellow Rail populations across two large study areas: one in the taiga plains of the Northwest Territories, and the other in the boreal plains of Alberta, Canada.

  For the Edéhzhíe Indigenous Protected Area (NWT), a predictive map of Yellow Rail density was developed based on data obtained from a systematic avian survey conducted in 2016, using 205 ARUs. Counts of Yellow Rail were converted to density estimates using habitat specific effective detection radii obtained via call-playback experiments. Generalized linear models and covariates from a detailed landcover classification effort were used to develop the spatial model. Yellow Rail appeared to breed at relatively high densities (0.07 males/ha compared to average densities of 0.04-0.05 males/ha) in Edéhzhíe and they were strongly associated with marsh wetlands. The Mills Lake wetland complex was identified as an important breeding area for Yellow Rail in the Northwest Territories based on a population estimate of ca. 560 breeding pairs.

  For the Alberta Oilsands Region, a predictive map of Yellow Rail breeding abundance was developed using acoustic data compiled from the first five years (2013-2017) of an ongoing bioacoustic monitoring program. Recent developments in open-access satellite data, cloud computing (Google Earth Engine), and data science were leveraged to secure large-scale, high-resolution (10 m) landcover data. Multiple satellite remote sensors were used to derive fifteen predictor variables: Sentinel-1 synthetic aperture radar, Sentinel-2 optical imagery, and Advanced Land Observation Satellite digital elevation maps. Gradient boosted regression was used to develop the spatial model. Six remote sensing predictors (DPOL, ΔVH, REIP, ARI, VH, and SWI), were identified as having strong predictive capacity. Several predictors had complex non-linear responses and multiple important interactions were identified. Approximately 1.5% of available wetland habitat in the region was predicted to be highly suitable for Yellow Rail.

• Subjects / Keywords

  • Remote Sensing
  • Predictive Models
  • Species at Risk
  • Yellow Rail
  • Bioacoustics
  • Wetlands

• Graduation date

  Fall 2019

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-f88r-dp27

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.